From 246452d14acc309c012360696cba32c37111b8fa Mon Sep 17 00:00:00 2001 From: j Date: Wed, 30 Sep 2026 15:46:24 +0700 Subject: [PATCH 01/25] =?UTF-8?q?=E7=AB=99=E7=82=B9=E6=94=B9=E6=88=90?= =?UTF-8?q?=E5=A4=9A=E8=AF=AD=E8=A8=80=EF=BC=9A=E8=8B=B1=E6=96=87=E9=BB=98?= =?UTF-8?q?=E8=AE=A4=E3=80=81=E8=B6=8A=E5=8D=97=E6=96=87=E8=B7=9F=E4=B8=8A?= =?UTF-8?q?=EF=BC=8C=E4=B8=AD=E6=96=87=E4=BB=8D=E6=98=AF=E5=8E=9F=E6=96=87?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 检索页从一份变三份：`/en/`（默认，访问 `/` 就是它）、`/zh/`、`/vi/`。 根目录的 `index.html` 变成跳语言的小页，按浏览器语言去 `/en/` 或 `/vi/`， 关掉 JavaScript 也能自己点。页头有语言切换，换语言时当前的筛选和定位一起带过去。 **中文正文仍在仓库根**：README.md、book/、docs/ 一个文件都没挪。zh 页面读的就是根上 那一份，所以作者改的还是原来那些路径，GitHub 渲染的还是仓库根的 README。 en/ 和 vi/ 各自有 README.md、book/、docs/，是那一种语言的副本。 页面从一份源码生成：`tools/site/page.template.html` 是模板（原 index.html 移过来的）， `tools/site/locales/.json` 是界面文案，`node tools/site/build.mjs` 生成三份页面 加根目录那个跳语言页。生成物入库是为了让 Pages 直接发；改页面改模板和字典。 数字不再写死在页面源码里。模板和三份字典里写 `{{entries}}` 这类占位符，由 `tools/site/build.mjs` 从 `tools/site/stats.json` 填——三种语言共用一份数字， 谁也不会和谁差。`sync-stats.mjs` 写完 stats.json 会跑一次 `build.mjs --check`， 页面没跟着重建就拦提交。 译文这一层不需要作者审。`tools/i18n/check.mjs` 拿中文原文逐条对译文：节数条数、 六个字段的有无与顺序、**成本标签注释逐字照抄**、证据等级、争议与待核实的条数、 收益/成本/来源三栏里的每一个数字、交叉引用的条号节号、来源与备注里的链接数量。 量级词换算（中文「167 万」在英文里只能写成 1,670,000）单独放行，其余数字对不上就是错。 `split.mjs` / `join.mjs` 按节切块、逐块校，和 check 共用同一份判据（structure.mjs）。 要求本身写在 `tools/i18n/BRIEF.md`，人看的版本，提示词直接取它。 CI 加「译文检查」job（译文对齐 + 生成物一致 + 中文文案和模板一致），独立于 build： 译文没对齐不该连累 EPUB、PDF 和离线单文件的发布。 顺带：域名和仓库地址收到 tools/site/site.json 一处（README、检索页、sitemap、 robots 都从它来，改域名只改一处）；长文的第四处链接（页头那行）改成从 README 导航表生成，不用手工挂了；离线单文件按语言出（`--locale `）； EPUB/PDF 的 README 解析改成认各语言标题，新增语言不用改脚本。 已翻：英文第 1、2 节和 5 篇长文 + README；越南文第 1 节。剩下的节在 `tools/i18n/split.mjs --all --locale ` 的进度里，翻完 CI 会自动带进页面。 CLAUDE.md「翻译一律不合进本仓库」那条改了：译文现在入本仓库，但作者只维护中文， 机器查结构，官方那两种语言不接受外部 PR 改。 --- .github/workflows/book.yml | 45 +- .gitignore | 4 + CLAUDE.md | 44 +- README.md | 31 +- en/README.md | 310 ++++ ...15\350\246\201\346\227\251\346\255\273.md" | 353 ++++ ...01\346\205\242\346\205\242\346\255\273.md" | 388 +++++ ...36\345\223\252\344\272\233\350\257\201.md" | 71 + ...05\345\244\207\346\270\205\345\215\225.md" | 89 + ...37\345\222\214\345\244\234\347\217\255.md" | 149 ++ ...22\344\270\215\345\210\222\347\256\227.md" | 123 ++ ...45\344\270\215\350\257\245\345\201\234.md" | 55 + en/index.html | 1505 ++++++++++++++++ index.html | 1477 +--------------- robots.txt | 2 +- sitemap.xml | 12 +- tools/i18n/BRIEF.md | 112 ++ tools/i18n/README.md | 61 + tools/i18n/anchors.mjs | 20 + tools/i18n/check.mjs | 70 + tools/i18n/glossary.en.mjs | 217 +++ tools/i18n/glossary.vi.mjs | 169 ++ tools/i18n/join.mjs | 84 + tools/i18n/prompt.mjs | 24 + tools/i18n/split.mjs | 109 ++ tools/i18n/structure.mjs | 188 ++ tools/lib/book.mjs | 59 +- tools/offline/build.mjs | 84 +- tools/site/build.mjs | 467 +++++ tools/site/extract.mjs | 110 ++ tools/site/locales.json | 49 + tools/site/locales.mjs | 33 + tools/site/locales/en.json | 195 +++ tools/site/locales/vi.json | 195 +++ tools/site/locales/zh.json | 195 +++ tools/site/page.template.html | 1529 ++++++++++++++++ tools/site/site.json | 5 + tools/site/stats.json | 16 + tools/sync-stats.mjs | 40 +- ...15\350\246\201\346\227\251\346\255\273.md" | 354 ++++ zh/index.html | 1537 +++++++++++++++++ 41 files changed, 9050 insertions(+), 1530 deletions(-) create mode 100644 en/README.md create mode 100644 "en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" create mode 100644 "en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" create mode 100644 "en/docs/\345\201\232\345\271\263\345\217\260\350\246\201\345\212\236\345\223\252\344\272\233\350\257\201.md" create mode 100644 "en/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" create mode 100644 "en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" create mode 100644 "en/docs/\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" create mode 100644 "en/docs/\351\201\207\345\210\260\351\231\214\347\224\237\344\272\272\345\207\272\344\272\213\350\257\245\344\270\215\350\257\245\345\201\234.md" create mode 100644 en/index.html create mode 100644 tools/i18n/BRIEF.md create mode 100644 tools/i18n/README.md create mode 100644 tools/i18n/anchors.mjs create mode 100644 tools/i18n/check.mjs create mode 100644 tools/i18n/glossary.en.mjs create mode 100644 tools/i18n/glossary.vi.mjs create mode 100644 tools/i18n/join.mjs create mode 100644 tools/i18n/prompt.mjs create mode 100644 tools/i18n/split.mjs create mode 100644 tools/i18n/structure.mjs create mode 100644 tools/site/build.mjs create mode 100644 tools/site/extract.mjs create mode 100644 tools/site/locales.json create mode 100644 tools/site/locales.mjs create mode 100644 tools/site/locales/en.json create mode 100644 tools/site/locales/vi.json create mode 100644 tools/site/locales/zh.json create mode 100644 tools/site/page.template.html create mode 100644 tools/site/site.json create mode 100644 tools/site/stats.json create mode 100644 "vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" create mode 100644 zh/index.html diff --git a/.github/workflows/book.yml b/.github/workflows/book.yml index 7ee20a31..f070ee39 100644 --- a/.github/workflows/book.yml +++ b/.github/workflows/book.yml @@ -1,16 +1,20 @@ # 正文一改就重新生成三样，挂到固定的 Release「epub-latest」上，下载链接不变： -# EPUB https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.epub -# PDF https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.pdf -# 离线单文件 https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html +# EPUB https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.epub +# PDF https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.pdf +# 离线单文件 https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html # PR 只生成并校验，不发布，产物在 Actions 页面的 artifact 里。 name: 电子书 +# 触发条件里带上了 en/**、vi/**、zh/** 和 tools/i18n/**：译文和检索页也是正文的一部分了 on: push: branches: [main] paths: - README.md - index.html + - en/** + - zh/** + - vi/** - og.png - ads/** - book/** @@ -21,6 +25,9 @@ on: paths: - README.md - index.html + - en/** + - zh/** + - vi/** - og.png - ads/** - book/** @@ -41,6 +48,25 @@ env: TYPST_VERSION: '0.15.1' jobs: + # 译文检查：拿中文原文逐条对 en/ 和 vi/（tools/i18n/check.mjs，判据在 tools/i18n/structure.mjs）。 + # 查的是丢东西：节数条数、六个字段、成本标签注释、证据等级、收益/成本/来源三栏的数字、 + # 交叉引用的条号节号、链接数量。翻译 PR 在这里红就知道该翻哪一条，不用作者读译文。 + # 单独一个 job：译文没对齐不该连累 EPUB、PDF 和离线单文件的发布。 + i18n: + name: 译文检查 + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + - uses: actions/setup-node@v4 + with: + node-version: 22 + - name: 英文与越南文对齐中文原文 + run: node tools/i18n/check.mjs + - name: 检索页生成物和模板、字典一致 + run: node tools/site/build.mjs --check + - name: 中文文案和模板一致 + run: node tools/site/extract.mjs --check + # 交叉引用检查。单独一个 job，不挂在 build 里：锚点没补好不该连累 EPUB、PDF 和 # 离线单文件的发布，那是两回事。PR 上照样红，main 上该发的还是发出去。 # 本地那条链路（sync-stats.mjs）要记得跑才会跑， @@ -79,6 +105,19 @@ jobs: node-version: 22 - run: node tools/sync-stats.mjs --check + # 检索页检查：三份语言页、跳语言的小页、sitemap.xml、robots.txt 是否和模板与字典对得上。 + # 和上面三个一样是「本地跑过才会更新」的生成物，网页上直接改模板的 PR 经过不了本地那条链路。 + # 单独一个 job，不挡电子书发布：页面生成物过时和 EPUB 该不该重发是两回事。 + site: + name: 检索页检查 + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + - uses: actions/setup-node@v4 + with: + node-version: 22 + - run: node tools/site/build.mjs --check + build: runs-on: ubuntu-latest steps: diff --git a/.gitignore b/.gitignore index b0c2f5b4..d21e97fe 100644 --- a/.gitignore +++ b/.gitignore @@ -2,3 +2,7 @@ .junie/ node_modules/ dist/ + +# 翻译中间产物：切出来的块（.i18n/<语言>/<节名>/partNN.md）。译文合进 <语言>/book/ 之后 +# 这份就没用了，join.mjs 每次都从原文重切。重切不覆盖已翻的块，所以它是活的翻译进度。 +.i18n/ diff --git a/CLAUDE.md b/CLAUDE.md index f040ab0d..93860b78 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -38,7 +38,7 @@ - 证据等级：A / B / C - 来源： - 备注：（争议、适用人群、注意事项） -- 标题行下紧跟一行 HTML 注释 ``，GitHub 渲染不可见，供 index.html 在线检索按成本维度筛选；新增或改动条目时同步维护。含义：钱 0=不花钱或省钱、少=几十到几百元或每月几十元内、多=上千元或长期显著支出；时间 少=一次几分钟内或顺手、中=一次性数小时或每周小时级、多=每天持续占用；毅力 否=做一次就完、些=改一个习惯或忍一点不适、是=长期对抗惯性 +- 标题行下紧跟一行 HTML 注释 ``，GitHub 渲染不可见，供检索页按成本维度筛选；新增或改动条目时同步维护。含义：钱 0=不花钱或省钱、少=几十到几百元或每月几十元内、多=上千元或长期显著支出；时间 少=一次几分钟内或顺手、中=一次性数小时或每周小时级、多=每天持续占用；毅力 否=做一次就完、些=改一个习惯或忍一点不适、是=长期对抗惯性 - **文献链接一律放「来源」栏，别堆在备注里**（2026-09-21 用户看着检索页上的卡片问「这个备注里面怎么一堆链接」）。起因是同一条目分三轮追加，每轮顺手在备注里带一个 DOI，攒到 3 个、备注也涨到近千字。补充证据、反方证据、出路证据的**文献一样进「来源」栏**，备注只写结论和数字，不带括号引文；备注里至多留一个链接，而且只留指向 docs/ 长文的那个。判据：`grep -c http` 扫全书备注，正常条目是 0 到 1 个，2 个就该看一眼，3 个一定是堆出来的。备注超过 700 字同样是信号，按「长解释放到 docs/」把细节挪进长文或并进来源栏。 - 「说人话」是把「收益」栏的统计量翻成日常说法的一到两句：HR/RR/OR 一律换算成「高约几成 / 低约百分之几」，人群和研究设计不写，只留读者关心的方向和量级；不许出现 HR、RR、OR、CI、队列、荟萃这些词，也不许出现「收益」栏里没有的数字。**不能新增的不只是数字**：收益栏没写的症状、事实断言、机制解读一样不许加（2026-09-19 用户抓出两处——献血条写「『脸色差、怕冷』不是瞎编」，可「脸色差」全条目没有任何来源提过；冷水澡条把「病假次数少了但总天数没变」解读成「病了也照样去上班」，原研究没这么说。两处都已改）。**说人话还必须自足**：读者多半只看这一行，不看来源也不看核实记录，所以不能引一句他没见过的说法再去评价它——「『XX』不是瞎编」「『XX』是真的」这类写法直接禁掉，要么把那个说法连同出处在这一行里交代清楚，要么改成正面陈述（献血那句最后改成「反过来，献得太勤确实会让人没力气、怕冷，机制是铁」）。写「说人话」时不要顺手去改「收益」栏。金钱与自由口径的条目同样写，翻成「能拿回多少钱 / 会摊上什么后果」。检索页把它渲染成卡片里最显眼的一段，缺这一行就不显示，可以逐节补。 - **全书一律写成没受过专业训练的成年人能一遍读懂的中文**（2026-09-20 用户定：「我有点看不懂，通俗点」「不然门槛太高了」）。覆盖条目标题、成本、说人话、收益、备注，以及 docs/ 下的长文；**来源栏除外**，文献题录和条款号照原样留着才能核对。做法：专业名词当场用日常话解释一句、名词本身保留（「辅料，也就是有效成分之外的填充和包衣」）；研究行话翻成动作（「观察性队列」→「只是跟踪记录、没有分组对照」）；法言法语落到「会摊上什么后果、该怎么做」。**收益栏的数字、置信区间、人群与年份说明一个都不许删改**，HR/RR/OR 在旁边就地补「低约 28%」这类换算并保留原值——这一栏是全书可核对性的底座，只加译文、不动原值。**条目标题只许加字解释，不许删字换词**：别节的引用靠标题里的词做锚点匹配（tools/check-refs.mjs 取任意两个连续汉字比对），删一个词就可能让那处引用从「带锚点」退化成裸条号。 @@ -66,7 +66,7 @@ - **有因果、转折关系的相邻短句用「因为」「所以」「但」连起来**，单句仍不超过 50 字。 - 收益与口径是后加的两个字段，用来回答「值不值得做」（证据等级只回答「数字可不可信」，两者正交）。口径按这条主要换回什么定：死亡率（含健康终点、存活率）、金钱（含保障待遇）、时间（含精力）、自由（含法律责任、个人信息）。**不同口径之间不做比较**。 - 收益量级尽量按阈值从条目自己的「收益」栏机械套，不凭感觉：死亡率口径看相对降幅（≥20% 大 / 10–20% 中 / <10% 或只有替代终点 小）；金钱看金额（万元级 大 / 数百到数千 中 / 几十元 小）；自由看后果（避免刑责 大 / 避免拘留或行政处罚 中 / 避免民事纠纷 小）；时间看节省量（每天小时级 大 / 每周小时级 中 / 一次性 小）。数字不足以判定时才用判断，并在核实记录里写明凭什么定的。 -- 条目数、A 级条数变了要同步四处：README（正文数字与两个徽章）、index.html（五处描述、numberOfPages、页头条目数）、CLAUDE.md 的目录简介，以及 **tools/og.html + 重新截图 og.png**（命令写在 og.html 文件头注释里，链接数的口径是 book/ 下所有「- 来源：」和「- 备注：」行里的 http(s) 总数）。跑 `node tools/sync-stats.mjs` 一键完成（2026-09-29 起全部用 node，原来的 sync-stats.ps1 已删，Linux、macOS 上照样跑）：它重算条目数、节数、A/B/C 数、争议数、TODO 数、链接数和性价比三档，回写 README 的九处（含「拆成 N 个文件」）、index.html 的四类位置（含页脚「N 个文件」）、tools/og.html 的三个数字，再用全新的 user-data-dir 重出 og.png（`--no-screenshot` 可只改数字；Chrome 不在常见位置就设环境变量 `CHROME`）。**新增要随统计变的数字，就往 sync-stats.mjs 的 EDITS 表里加一行**，别在正文里另写死一处。CI 的「统计数字检查」job 跑 `node tools/sync-stats.mjs --check`，数字过时就红（2026-09-29 加，此前网页上直接合并的 PR 完全不经过这个脚本）；它只查不改，og.png 仍要本地重出。**CLAUDE.md 的目录简介脚本不碰，仍要手工改。**性价比档的规则和 index.html 的 COST_W、e.ratio 两行绑定，那两行改了脚本会直接报错要求同步。**每次改完条目就跑一次，和正文改动一起提交**（2026-09-18 用户定；当天早些时候的「等我说改完了再改」已作废——脚本把一次同步的开销压到几百 token，不必再攒着）。跑完不用打开 og.png：脚本自己校验图是本次写入、大小在 120KB 到 400KB，只有改过 tools/og.html 版式时才值得看一眼。汇报里也不必再写「待同步」，直接给同步后的数字。 +- 条目数、A 级条数变了要同步四处：README（正文数字与两个徽章）、`tools/site/stats.json`（检索页三种语言的数字都从这儿来）、CLAUDE.md 的目录简介，以及 **tools/og.html + 重新截图 og.png**（命令写在 og.html 文件头注释里，链接数的口径是 book/ 下所有「- 来源：」和「- 备注：」行里的 http(s) 总数）。跑 `node tools/sync-stats.mjs` 一键完成（2026-09-29 起全部用 node，原来的 sync-stats.ps1 已删，Linux、macOS 上照样跑）：它重算条目数、节数、A/B/C 数、争议数、TODO 数、链接数和性价比三档，回写 README 的九处（含「拆成 N 个文件」）、`tools/site/stats.json`、tools/og.html 的三个数字，再用全新的 user-data-dir 重出 og.png（`--no-screenshot` 可只改数字；Chrome 不在常见位置就设环境变量 `CHROME`）。**新增要随统计变的数字，就往 sync-stats.mjs 的 EDITS 表里加一行**，别在正文里另写死一处。CI 的「统计数字检查」job 跑 `node tools/sync-stats.mjs --check`，数字过时就红（2026-09-29 加，此前网页上直接合并的 PR 完全不经过这个脚本）；它只查不改，og.png 仍要本地重出。**CLAUDE.md 的目录简介脚本不碰，仍要手工改。**检索页的数字不再写死在页面源码里**（2026-09-30 多语言时改的）：模板和三份字典里写的是 `{{entries}}` 这类占位符，由 `tools/site/build.mjs` 从 `stats.json` 填——三种语言共用一份数字，谁也不会和谁差。`sync-stats.mjs` 写完 stats.json 会顺带跑一次 `tools/site/build.mjs --check`，页面没跟着重建就直接报错拦住提交。**性价比档的规则和检索页模板（`tools/site/page.template.html`）的 COST_W、e.ratio 两行绑定，那两行改了脚本会直接报错要求同步。**每次改完条目就跑一次，和正文改动一起提交**（2026-09-18 用户定；当天早些时候的「等我说改完了再改」已作废——脚本把一次同步的开销压到几百 token，不必再攒着）。跑完不用打开 og.png：脚本自己校验图是本次写入、大小在 120KB 到 400KB，只有改过 tools/og.html 版式时才值得看一眼。汇报里也不必再写「待同步」，直接给同步后的数字。 - **插入或删除条目之后，必须扫一眼 `docs/引用对照.md` 的 diff**（2026-09-19 用户问「怎么防范引用错位」时定的）。正文里 257 处「第 X 条」引用只记条号不记内容，插一条进去后面的条号集体顺延，引用却留在原地——而顺延后的条号通常**仍在范围内**，查越界一条也抓不到（第 7 节曾有 6 处指错：医疗救助指到低保、救助站指错条，全是这么来的）。`sync-stats.mjs` 会自动跑 `tools/check-refs.mjs`，把每处引用**实际指向的条目标题**连同引用处的上下文写进 `docs/引用对照.md`，该文件入库。**判据：条号没变而「指向的条目」那列变了，就是被撞歪的引用**；表里还带着引用处前十几个字，和标题一对就知道指没指对。CI 用 `node tools/check-refs.mjs --check`（有问题则退出码 1），它是 `book.yml` 里名为「交叉引用检查」的独立 job，**故意不挂在 build 里**：锚点没补好不该连累 EPUB、PDF 和离线单文件的发布，PR 上照样红、main 上该发的还是发出去。这个 job 是 2026-09-21 补的——在那之前 CLAUDE.md 写着「CI 用」但 workflow 里其实没有这一步，全靠本地 `sync-stats.ps1`，而那条链路只在 Windows 上跑、找不到 node 时还会静默跳过（脚本里就是打一句「跳过引用对照表」然后继续），外部 PR 更是完全跑不到。 @@ -75,17 +75,46 @@ - **法条条款号和条目引用要能分开，判据是「紧挨着有没有引文标记」**（2026-09-21 用户让扫类似错误时定的）。`check-refs` 靠这个判断一处「第 N 条」是法条还是条目引用：紧挨在它前面的是 `《…》`、`〔…〕`、「14 号」、「该解释／该意见」，或者以法规名收尾（「治安管理处罚法第 26 条」），就当法条跳过。由此两条要求：① **法条引文必须自带文件名**，一句一条往下列时写「该解释第 11 条讲的是取证」，不要写「……的法院命令。第 11 条讲的是取证」靠上一句撑着（这本来也是「每句能单独读懂」的要求）；② 条目引用前面别紧跟法规名。**这类错误的判据不是 `--check` 绿不绿，是引用总数对不对**：新增 N 处引用，`docs/引用对照.md` 的总数就该涨 N。涨少了说明有引用被当成法条跳过——它压根不进对照表，`--check` 没有可查的引用反而显示「通过」，是静默失效。2026-09-21 全量扫描查出 12 处这样的引用（「具体办法在后面四条：第 3 条」「法律援助见第 3 条」「被违法解除的按第 6 条」「查法见第 12 条」这种，旧判据是看前 16 个字里有没有「法」字，「办法」「查法」「违法」全被连累），判据改掉后 423 处恢复到 435 处。 **扫描范围是 book/ 的条目正文 + 节首引言 + docs/ 长文**（2026-09-21 扩的，此前只扫条目正文）。节首引言和长文里的条号一样会被顺延撞歪，此前一处都没查过，`--check` 照常显示通过——和上面那条法条过滤一样是静默失效，只是成因是压根没扫。扩完当天补了 15 处锚点，**基线从 454 处涨到 533 处**，「新增 N 处、总数涨 N」这条判据从 533 起算。三件事随之定下来：① 长文里没有「本节」，裸的「第 N 条」一律当法条不扫，长文引用必须写全「第 X 节第 Y 条」；② 对照表的「出处」列，条目写「第 N 条」、节首写「节首」、长文写最近的小标题；③ **区间引用（「见第 8 节第 11 到 14 条」「第 5 到第 10 条」）免验锚点**——它指的是一整块条目，没法给块里每条都配锚点，只靠对照表 diff 兜；这种写法此前整处匹配不上，等于没扫，全书有 5 处。 **锚点要够长、够近**（2026-09-20 加严）：整句里连着三个汉字和目标标题对上，或者引用所在的那个逗号分句里有两个汉字对上，才算数；只在分句之外撞上「自己」「公司」这类两字常见词的，`--check` 判为失败，修法和裸条号一样是补「（锚点词）」。起因是 PR #18 在第 31 节插条目，第 1 条备注里「……的贷款见本节第 15 条」被顺延撞到新条目「在家给境外公司远程干活……个税自己报」上，隔着两个逗号的「你自己还」冒充了锚点，`--check` 当时报通过，是靠对照表 diff 才发现的。加严后全书补了 8 处显式标注；`tools/check-refs.mjs` 的注释里记着灵敏度实测——把节内引用整体顺延一条，能当场拦下约七成，剩下的（相邻两条讲同一件事、标题共用词）仍然只能靠对照表的 diff，所以那一步扫 diff 不能省。 -- 性价比档不写进 README，由 index.html 按「收益量级 + 三项成本」实时合成（大且成本全零=极高，大且成本分≤2 或 中且成本全零=高，其余=一般），改档位规则只改 index.html。 +- 性价比档不写进 README，由检索页按「收益量级 + 三项成本」实时合成（大且成本全零=极高，大且成本分≤2 或 中且成本全零=高，其余=一般），改档位规则只改检索页模板。 ## 目录结构 -正文按节拆成 `book/01-*.md` … `book/31-*.md`（2026-09-08 拆的，原来单文件 531 KB，超过 GitHub 渲染 Markdown 的 512 KB 上限，后面的节显示不出来也跳不了锚点）。README 只留导读、术语表和目录，新增或修改条目改对应的 book 文件；检索页 index.html 先读 README 拿目录里的文件列表，再并发读这些文件。每节文件第一行是回总目录的链接，第二行空行，第三行是 `# N. 节名`。 +**站点是多语言的，英文是默认语言**（2026-09-30 定的）。四种路径分工： + +| 路径 | 是什么 | 谁维护 | +| --- | --- | --- | +| `/` | 跳语言的小页，按浏览器语言去 `/en/` 或 `/vi/`，关掉 JS 也能点 | `tools/site/build.mjs` 生成 | +| `/en/` | 英文站（默认） | 译文，见 `en/` | +| `/vi/` | 越南文站 | 译文，见 `vi/` | +| `/zh/` | 中文站，读的是**仓库根**那份正文 | 原始内容都在根上 | + +**中文正文仍在仓库根**：`README.md`、`book/`、`docs/` 都没挪。`zh/index.html` 用 `contentBase: '../'` 指回根上那一份，所以作者改的还是原来那些文件，GitHub 渲染的还是仓库根的 README。en/vi 各自有 `README.md`、`book/`、`docs/`，是那一份语言的副本。 + +检索页的源码是 `tools/site/page.template.html`（2026-09-30 从原来的 `index.html` 移过来的），**三份页面都由 `tools/site/build.mjs` 从模板加 `tools/site/locales/.json` 生成，生成物入库**（Pages 直接发）。改页面改模板和字典，别改 `en/index.html`、`zh/index.html`、`vi/index.html`。根目录那个 `index.html` 是生成的跳语言页。 + +### 站点怎么上线 +- **部署用 GitHub Pages 的分支发布**（`main` 根目录，`build_type: legacy`），没有 workflow。`.nojekyll` 在根上，Pages 原样发仓库内容，Jekyll 不参与——所以别删它，删了 Jekyll 会重新接管并忽略下划线开头的目录。 +- **域名只写在 `tools/site/site.json` 的 `site` 和 `repo` 两个值里**，别的文件一处都不许再抄。`locales.json` 里每种语言的 `canonical` 和 `repoBlob` 是从这两个值推出来的，`build.mjs` 每次跑都逐个对一遍，**对不上直接抛错**（`checkUrls()`），不会出现「Pages 发到 A 域名、页面自称是 B 域名」。`sitemap.xml` 和 `robots.txt` 由同一个脚本生成，域名单独抄一遍就会漏。 +- **四种链接要分清，改域名时只动前两种**：① 站点链接（`*.github.io/…`）和② 面向读者的 GitHub 链接（页面里的文件链接、`/releases/download/…`）跟着 `site.json` 走；③ `git clone` 那行、④ 「写明出处／Credit the source」和 Star History 图这三处**永远指向上游 `eternity4719/HowToLiveBetter`**，它们是在说明这本书的出处，不是本站的地址。2026-09-30 换域名时按这个分的，核对的判据是 `grep -c eternity4719 README.md en/README.md` 各应为 3，且都在那三类位置上。 +- **`sitemap.xml` 和 `robots.txt` 是生成物**，别手工编辑。sitemap 只列这次真正生成的页面（眼下是 `/`、`/en/`、`/zh/`），没翻完的语言不进，免得给搜索引擎一个 404。 +- **CI 的「检索页检查」job 跑 `node tools/site/build.mjs --check`**（`book.yml` 里，2026-09-30 加），和另外三个检查一样只查不改、不挡电子书发布。改了模板、字典、`locales.json`、`site.json` 或某一门的 `README.md` 之后，本地跑一次 `node tools/site/build.mjs` 再提交。 +- **没翻完的语言不会上线，但会被列出来**：脚本报「跳过 vi」和「另有 N 项在目录里列着、正文还没翻到」，前者是没生成页面，后者是页面生成了但目录里那一节还没有译文。hreflang、跳语言页、页内语言下拉都只列真正生成的那几份，所以 `vi/index.html` 不存在时页面上也没有一个点进去 404 的越南语入口。 + +**正文按节拆成 `book/01-*.md` … `book/34-*.md`（2026-09-08 拆的，原来单文件 531 KB，超过 GitHub 渲染 Markdown 的 512 KB 上限，后面的节显示不出来也跳不了锚点）。** README 只留导读、术语表和目录，新增或修改条目改对应的 book 文件；检索页先读 README 拿目录里的文件列表，再并发读这些文件（清单由构建给，见 `SITE.files`）。每节文件第一行是回总目录的链接，第二行空行，第三行是 `# N. 节名`。 + +### 译文怎么维护 +- **流程和判据写在 `tools/i18n/BRIEF.md`**，翻译任务的提示词和结构检查的判据是同一份。词表 `tools/i18n/glossary.{en,vi}.mjs`。 +- 一次翻一节。`node tools/i18n/split.mjs book/NN-*.md --locale en` 把它切成十来个条一块放进 `.i18n/en/<节名>/partNN.md`（**已翻过的块不会被覆盖**，改参数重跑不会冲掉译文），翻译完 `node tools/i18n/join.mjs book/NN-*.md --locale en` 逐块校一遍再合成 `en/book/NN-*.md`。 +- `node tools/i18n/check.mjs`（CI 的「译文检查」job）拿中文原文逐条对译文：条数条号、六个字段的有无与顺序、**成本标签注释逐字照抄**、证据等级、争议与待核实的条数、**收益/成本/来源三栏的数字一个不多一个不少**（量级词换算除外，见 `tools/i18n/structure.mjs` 里 `numbersOf` 的说明）、「第 X 条」指路的条号节号、来源与备注里的链接数量。判据在 `structure.mjs`，join 和 check 共用同一份。 +- **译文不许改数字**。中文「167 万」在英文里只能写成 1,670,000、在越南文里只能写成 1.670.000，这是量级换算，`structure.mjs` 放行；除此之外数字对不上就是错。这一层机器查得出来，所以译文不需要作者审。 +- 越南文里 **条目 = "mục"，节 = "phần"，法条的款 = "khoản (2)"**——三个词不能混。「第 13 节」写成 "phần 13"，写成 "mục 13" 读者就当成第 13 条了。 +- **各语言共用一套中文取值**：成本标签注释（`钱=0 时间=少 …`）、筛选按钮的 `data-v`、`#hash` 锚点、地址栏参数。这不是漏译，是刻意的，译文里一字不改。 README 开头「这本书想回答的问题」表**一节只占一行**（2026-09-19 用户定）：同一个链接贴两遍，读者要在同一个目的地上停两次，两行之间又没有逻辑差别。一节覆盖面宽就在一行里用逗号串几个问题、一个问号收尾（第 19 节「加班费、年休假该怎么算，被裁该拿多少补偿，上班受了伤怎么认定和拿钱？」是范式），不要拆成两行。当天把第 6、8、24 节原有的两行都并掉了，现在是 32 行对 32 节。 -电子版三样，全由 `.github/workflows/book.yml` 在 main 上正文改动后自动生成，挂到固定 Release `epub-latest`（下载链接不变，README 引用的就是它），PR 只生成校验不发布；产物都进 `dist/`（已 gitignore）。**产物本身不入库**，正文几乎天天改，提交二进制只会过时和撑大历史（2026-09-18 issue #12 定的方案）。README 的结构解析和文件清单收在 `tools/lib/book.mjs`，三套构建共用，新增节或长文不用改脚本。**但新增 docs 长文必须手工挂四处链接**（2026-09-21 连漏两次，两次都是用户发现的）：① README 顶部导航表格里「长文」那一格（HTML 表格，单元格里必须留空行、用 markdown 写成「[x](docs/x.md)」，写成 `` 就不被下面那个正则认；2026-09-28 导航区改成表格后不再带「（长文）」后缀）；② README 末尾「长文另见」那句；③ README 对应节的目录行（写成「长文见 [docs/x.md](docs/x.md)。」）；④ index.html 里 `class="doc-links"` 那一行，URL 要百分号编码。核对办法：`grep -c "docs/<名>.md" README.md` 应为 3，`grep -c "<名>" index.html` 应为 1。第 ① 处是硬性的——`book.mjs` 第 54 行靠正则 `\]\((docs/[^)#/]+\.md)\)` 从 README 里扒长文清单，只在 book/ 条目里链接，EPUB、PDF、离线单文件三套构建都收不到它。sync-stats.ps1 不检查这三处。 +电子版三样，全由 `.github/workflows/book.yml` 在 main 上正文改动后自动生成，挂到固定 Release `epub-latest`（下载链接不变，README 引用的就是它），PR 只生成校验不发布；产物都进 `dist/`（已 gitignore）。**产物本身不入库**，正文几乎天天改，提交二进制只会过时和撑大历史（2026-09-18 issue #12 定的方案）。README 的结构解析和文件清单收在 `tools/lib/book.mjs`，三套构建共用，新增节或长文不用改脚本。**但新增 docs 长文必须手工挂三处链接**（2026-09-21 连漏两次，两次都是用户发现的；2026-09-30 多语言后第四处由构建生成，剩三处）：① README 顶部导航表格里「长文」那一格（HTML 表格，单元格里必须留空行、用 markdown 写成「[x](docs/x.md)」，写成 `` 就不被下面那个正则认；2026-09-28 导航区改成表格后不再带「（长文）」后缀）；② README 末尾「长文另见」那句；③ README 对应节的目录行（写成「长文见 [docs/x.md](docs/x.md)。」）；④ 检索页的页头那行长文链接。**这一处 2026-09-30 起由构建从 ① 生成**（`tools/site/build.mjs` 的 `readLongform`），不用手工挂了；核对办法：`grep -c "docs/<名>.md" README.md` 应为 3，三份生成出来的 `*/index.html` 里各有 1 处。第 ① 处是硬性的——`book.mjs` 的 `readBook()` 靠正则 `\]\((docs/[^)#/]+\.md)\)` 从 README 里扒长文清单，只在 book/ 条目里链接，EPUB、PDF、离线单文件三套构建都收不到它。`sync-stats.mjs` 不检查这三处。 - EPUB：`tools/epub/build.mjs`（只依赖 marked，zip 自己打），CI 里跑 epubcheck。本地 `cd tools/epub && npm ci && npm run build`。 -- 离线单文件 HTML：`tools/offline/build.mjs`（零依赖）把正文**和 docs 长文**内联进 index.html 的 `window.__CORPUS__`（`parts` 放 book/、`docs` 放长文），双击就能看、离线可用；index.html 的 `init()` 认这个变量就不发请求，改 init 的取数逻辑要同时顾到这条路。**检索页的长文是就地弹窗打开的，不跳 GitHub**：点击处统一拦「指向 `docs/*.md` 的 GitHub 链接」（页头那行长文、条目备注里 `hrefOf()` 转出来的相对链接，形状都是这一个），弹窗自带一份可点的章节目录和 scrollspy，markdown 由 index.html 里那个只认标题/段落/列表/表格的小渲染器处理，不引库。长文里如果开始用引用块、代码块这类新语法，要同步加进那个渲染器。 +- 离线单文件 HTML：`tools/offline/build.mjs`（零依赖）把正文**和 docs 长文**内联进检索页的 `window.__CORPUS__`（`--locale ` 出那一种语言的；默认语言的文件名不带后缀，其余带 `-`）（`parts` 放 book/、`docs` 放长文），双击就能看、离线可用；页面的 `init()` 认这个变量就不发请求，改 init 的取数逻辑要同时顾到这条路。**检索页的长文是就地弹窗打开的，不跳 GitHub**：点击处统一拦「指向 `docs/*.md` 的 GitHub 链接」（页头那行长文、条目备注里 `hrefOf()` 转出来的相对链接，形状都是这一个），弹窗自带一份可点的章节目录和 scrollspy，markdown 由检索页里那个只认标题/段落/列表/表格的小渲染器处理，不引库。长文里如果开始用引用块、代码块这类新语法，要同步加进那个渲染器。 - PDF：`tools/pdf/build.mjs` + `tools/pdf/template.typ`，pandoc 转 typst、typst 排版（A4、每节另起一页、页眉带节名、目录带页码、书签到条目级）。本地要 pandoc ≥ 3.1 和 typst ≥ 0.13，用环境变量 `PANDOC`、`TYPST` 指到可执行文件；CI 里按 workflow 顶部 env 里钉的版本下载。版面只改 template.typ。 1. 不要早死（外因：交通、火灾、燃气器具与用气安全、毒蘑菇、溺水、跌落、中毒、疫苗、体检筛查、心理危机（12356 与限制致死手段、自杀念头的时间尺度与未遂之后的长期结局、中毒被救回来之后的不可逆后遗症、高处坠落幸存之后的 ICU 与康复账、摘掉一个肾之后剩余肾的长期代价与卖肾者的随访结局，法律那一侧在第 9 节）、家庭应急装备、安全套与不共用针具、艾滋病的免费自愿咨询检测与窗口期、肉眼血尿等该去查的信号、失控放射源（不捡不拆来路不明的金属件，含源设备与通电产生射线的装置的分界，编在节末因为基础发生率极低）；第 37 条是赌博障碍的自杀风险（瑞典登记研究 SMR 自杀 15、总死亡 1.8，2023 年更正男 12.1 女 16.1；B 级，只算了专科病人），落点是 12356 加把钱交给家人管，追加在放射源之后而没插进心理危机那组，免得顺延条号；第 38 条是男男性行为者的暴露前预防药 PrEP（A：iPrEx 44%、PROUD 与 IPERGAY 86%、Fonner 2016 吃药率超七成 RR 0.30；处方药，全国免费政策没查到，按需 2-1-1 只对男男性行为者）；装备清单长文放 docs/家庭应急装备清单.md） @@ -140,5 +169,6 @@ README 开头「这本书想回答的问题」表**一节只占一行**（2026-0 ## 工作方式 - 一次可以跨多节改，不必做完一节就停；改动范围在汇报里逐节写清 - 不要主动扩展目录结构，要加新节先问我 -- **翻译一律不合进本仓库**（2026-09-18 issue #4 / PR #5、#7 定的）：作者看不懂的语言合进来就是盲合，以后也审不了、维护不了。译者在自己的 fork 维护，本仓库只在 README 顶部「其他语言」一行加链接，注明非官方、以中文原文为准。收到翻译 PR 就按这个回复并关闭，不要 review 译文本身。 +- **译文入本仓库，但作者只维护中文那一份**（2026-09-30 定的，推翻了 2026-09-18 issue #4 / PR #5、#7 的「翻译一律不合进本仓库」）：站点改成多语言，英文是默认语言（`/en/`），越南文是 `/vi/`，中文原文挪到 `/zh/`（正文文件仍在仓库根，zh 页面读的是根上那一份）。译文由本仓库维护，结构对齐和数字一致性由 `tools/i18n/check.mjs` 机器查；**作者不需要读英文或越南文，也不需要审译文**——正文改动一律先改中文根目录那一份，各语言由翻译流程跟上。 + - 别人提的第三方译文 PR 仍按原来的规矩处理：只在 README 顶部「其他语言」一行加链接，注明非官方、以中文原文为准，不 review 译文本身。现在多了一条：官方那两种语言（en、vi）不接受外部 PR 改，只改中文和工具链。 - 修改已有条目时保留原来的证据等级和来源，除非有更好的原始文献替换 diff --git a/README.md b/README.md index e80911a3..f90c1afd 100644 --- a/README.md +++ b/README.md @@ -9,20 +9,20 @@ 不用全做：这是按性价比排好的备选单，不是任务清单——挑走一两条就算数，作者自己也没做到其中大部分。 -[![在线检索](https://img.shields.io/badge/%E5%9C%A8%E7%BA%BF%E6%A3%80%E7%B4%A2-%E7%82%B9%E8%BF%99%E9%87%8C%E6%89%93%E5%BC%80-3451b2?style=flat-square)](https://eternity4719.github.io/HowToLiveBetter/) +[![在线检索](https://img.shields.io/badge/%E5%9C%A8%E7%BA%BF%E6%A3%80%E7%B4%A2-%E7%82%B9%E8%BF%99%E9%87%8C%E6%89%93%E5%BC%80-3451b2?style=flat-square)](https://streamentry.github.io/HowToLiveBetter/zh/) [![条目](https://img.shields.io/badge/%E6%9D%A1%E7%9B%AE-641%20%E6%9D%A1-18794e?style=flat-square)](#目录) [![证据分级](https://img.shields.io/badge/%E8%AF%81%E6%8D%AE%E5%88%86%E7%BA%A7-A%20425%20%C2%B7%20B%20165%20%C2%B7%20C%2051-915930?style=flat-square)](#证据分级) [![原始文献](https://img.shields.io/badge/%E5%8E%9F%E5%A7%8B%E6%96%87%E7%8C%AE-1443%20%E6%9D%A1%E9%93%BE%E6%8E%A5-565a5f?style=flat-square)](docs/核实记录/) [![许可](https://img.shields.io/badge/%E8%AE%B8%E5%8F%AF-CC%20BY%204.0-565a5f?style=flat-square)](#许可) -### [打开在线检索页](https://eternity4719.github.io/HowToLiveBetter/) · [让 AI 照书回答（skill）](skills/life-decision-guide/README.md) +### [打开在线检索页](https://streamentry.github.io/HowToLiveBetter/zh/) · [让 AI 照书回答（skill）](skills/life-decision-guide/README.md) AI 助手 skill 支持 Claude Code 和 Codex。装上后直接问「替朋友担保签不签」，它先查书里的条目再回答，并注明出自第几节第几条。
下载 -[PDF](https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.pdf) · [EPUB](https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.epub) · [离线单文件（HTML）](https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html) +[PDF](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.pdf) · [EPUB](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.epub) · [离线单文件（HTML）](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html)
查阅 @@ -37,7 +37,8 @@ AI 助手 skill 支持 Claude Code 和 Codex。装上后直接问「替朋友担
其他语言 -[English](https://dlgrv.github.io/HowToLiveBetter/en/) · [Русский](https://dlgrv.github.io/HowToLiveBetter/ru/) · [Español](https://dlgrv.github.io/HowToLiveBetter/es/)，[dlgrv](https://github.com/dlgrv) 维护的翻译（[仓库](https://github.com/dlgrv/HowToLiveBetter)） +**官方**：[English](https://streamentry.github.io/HowToLiveBetter/en/)（默认，访问 `/` 就是它） · [Tiếng Việt](https://streamentry.github.io/HowToLiveBetter/vi/)，本仓库维护，译文跟不上时以中文原文为准。 +**他人维护**：[Русский](https://dlgrv.github.io/HowToLiveBetter/ru/) · [Español](https://dlgrv.github.io/HowToLiveBetter/es/)，[dlgrv](https://github.com/dlgrv) 的翻译（[仓库](https://github.com/dlgrv/HowToLiveBetter)）
衍生工具 @@ -47,7 +48,7 @@ AI 助手 skill 支持 Claude Code 和 Codex。装上后直接问「替朋友担
-其他语言和衍生工具由他人维护，内容可能落后，以本仓库中文原文为准。 +英文和越南文由本仓库维护，可能落后于中文原文；俄文、西班牙文由他人维护。译文有出入一律以中文原文为准。 @@ -96,12 +97,12 @@ AI 助手 skill 支持 Claude Code 和 Codex。装上后直接问「替朋友担 - **不用全做**：这是一份按性价比排好的备选单，不是任务清单。挑走一两条就算数，剩下的放着，需要时再回来查。「说着容易做着难」这个评价是对的——作者自己也没做到其中大部分，写下来是为了要用的时候找得到。想挑省力的，看下面「只想看最值得做的」那条。 - **想让 AI 帮你查**：仓库里带了一个中文 skill（[skills/life-decision-guide](skills/life-decision-guide/)），Claude Code 和 Codex 都能装。装上以后直接问「替朋友担保签不签」「每天通勤两小时值不值」。它会先把相关条目从正文里查出来，再照书里的算账方式排序回答，并注明出自第几节第几条。查不到就说查不到，不自己编数字。装法见 [那个目录的说明](skills/life-decision-guide/README.md)。 -- **想按条件挑**：打开[在线检索页](https://eternity4719.github.io/HowToLiveBetter/)，可以按关键词、章节、证据等级来筛，也可以按「花不花钱、花多少时间、要不要毅力」这三样筛，几个条件能叠着用。页面上的内容直接取自 book/ 目录里的正文，正文一改，页面跟着改。 +- **想按条件挑**：打开[在线检索页](https://streamentry.github.io/HowToLiveBetter/)，可以按关键词、章节、证据等级来筛，也可以按「花不花钱、花多少时间、要不要毅力」这三样筛，几个条件能叠着用。页面上的内容直接取自 book/ 目录里的正文，正文一改，页面跟着改。 - **条目之间会互相指路**（「见第 8 节第 17 条」这种）：在检索页里，这种指路带一条虚线。点一下，就地显示被指的那条的标题和「说人话」。想真的翻过去，再按「跳过去」。那一条正好被筛选条件藏起来了，页面会自动把筛选清掉。在 GitHub 上直接读正文点不动，但每处指路后面都写着指向什么（「见第 18 条（借钱写清借条）」）。不翻过去也知道说的是哪条。 - **想按顺序读**：每节内的条目按性价比从高到低排列，从每节前几条开始看就行。 -- **想离线看、想发给别人**：下载 [离线单文件 HTML](https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html)，整本书连同检索和筛选都在这一个文件里，双击就开，不用服务器也不用联网，微信里也能直接传。 -- **想打印或在手机上翻**：下载 [PDF](https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.pdf)，A4 排版、两百多页，带目录页码和书签，每节另起一页。 -- **想在 Kindle 或其他阅读器上读**：下载 [EPUB 电子书](https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.epub)，Kindle 用 Send to Kindle 发过去即可。 +- **想离线看、想发给别人**：下载 [离线单文件 HTML](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html)，整本书连同检索和筛选都在这一个文件里，双击就开，不用服务器也不用联网，微信里也能直接传。 +- **想打印或在手机上翻**：下载 [PDF](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.pdf)，A4 排版、两百多页，带目录页码和书签，每节另起一页。 +- **想在 Kindle 或其他阅读器上读**：下载 [EPUB 电子书](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.epub)，Kindle 用 Send to Kindle 发过去即可。 - **三样都是正文每次更新后自动重新生成的**，下载链接固定不变；转发出去的那一份不会跟着更新，以在线版为准。 - **看不懂那串数字**：每条都有一行「说人话」。它把「收益」栏里那些研究里的写法，翻成「同期死亡的概率低约两成」「拘留几日、罚多少钱」这样的日常说法。它只用「收益」栏已经写到的内容，不添新数字。只看这一行就够拿主意。「收益」栏里原样留着全部数字，想自己核对就看那一栏。 - **只想看结论最硬的**：在检索页里勾选证据等级 A，只留下有具体数字、来自荟萃分析或大型试验的 425 条。 @@ -122,7 +123,7 @@ AI 助手 skill 支持 Claude Code 和 Codex。装上后直接问「替朋友担 ## 自己跑一份 -多数人用不着部署：[在线检索页](https://eternity4719.github.io/HowToLiveBetter/)是现成的，要离线就下[离线单文件 HTML](https://github.com/eternity4719/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html)，双击就开。 +多数人用不着部署：[在线检索页](https://streamentry.github.io/HowToLiveBetter/)是现成的，要离线就下[离线单文件 HTML](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html)，双击就开。 真要在自己电脑或服务器上跑： @@ -132,7 +133,7 @@ cd HowToLiveBetter python -m http.server 8000 ``` -然后浏览器开 `http://localhost:8000/`。检索页是纯静态的，`README.md` 和 `book/` 就是它的数据，没有后端、没有数据库、不用装依赖；把整个目录丢给任何静态服务器（Nginx、GitHub Pages、对象存储）效果一样。注意 `index.html` 必须经 http 打开，直接双击本地文件会空白——浏览器不许网页读本地文件，那种场景请用上面的离线单文件版。 +然后浏览器开 `http://localhost:8000/`。检索页是纯静态的，`README.md` 和 `book/` 就是它的数据，没有后端、没有数据库、不用装依赖；把整个目录丢给任何静态服务器（Nginx、GitHub Pages、对象存储）效果一样。注意检索页必须经 http 打开，直接双击本地文件会空白——浏览器不许网页读本地文件，那种场景请用上面的离线单文件版。 想自己生成三样电子版（平时用不着，Release 里的就是自动生成的）： @@ -279,11 +280,13 @@ node tools/pdf/build.mjs # PDF，另需 pandoc ≥ 3.1 和 typ 每节内条目按性价比从高到低排列。「不要早死」「不要浪费时间」这类节标题说的是这一节想防住的结果，条目本身要做还是别做，以条目标题为准。长文另见 [docs/家庭应急装备清单.md](docs/家庭应急装备清单.md)、[docs/做平台要办哪些证.md](docs/做平台要办哪些证.md)、[docs/结婚划不划算.md](docs/结婚划不划算.md) 、[docs/遇到陌生人出事该不该停.md](docs/遇到陌生人出事该不该停.md) 和 [docs/生物钟和夜班.md](docs/生物钟和夜班.md)。每条来源的核实过程记录在 [docs/核实记录](docs/核实记录/)。 -仓库根目录的 `index.html` 是在线检索页：按关键词、章节、证据等级和成本维度（花钱、花时间、要毅力）筛选条目，数据直接读本文件。在仓库设置里开启 GitHub Pages（Deploy from a branch，分支 main，目录 /）后即可访问。`tools/epub/` 是电子书生成脚本，`cd tools/epub && npm ci && npm run build` 在本地出一本 EPUB 到 `dist/`；GitHub Actions 在正文改动后自动跑同一个脚本并更新 Release。 +`en/index.html`、`zh/index.html`、`vi/index.html` 是三种语言的在线检索页：按关键词、章节、证据等级和成本维度（花钱、花时间、要毅力）筛选条目，数据直接读本目录下那一份 README 和 `book/`。根目录的 `index.html` 是跳语言的小页，按浏览器语言去 `/en/`（默认，英文）或 `/vi/`，关掉 JavaScript 也能自己点。在仓库设置里开启 GitHub Pages（Deploy from a branch，分支 main，目录 /）后即可访问。 + +页面源码是 `tools/site/page.template.html`，三份页面都由 `node tools/site/build.mjs` 从模板加 `tools/site/locales/*.json` 生成，生成物入库是为了让 Pages 直接发；改页面改模板和字典，别改生成出来的那三份。`tools/epub/` 是电子书生成脚本，`cd tools/epub && npm ci && npm run build` 在本地出一本 EPUB 到 `dist/`；GitHub Actions 在正文改动后自动跑同一个脚本并更新 Release。 ## 正文 -正文按节拆成 34 个文件放在 [book/](book/)，点上面目录里的节名进入。拆开是因为单文件已经超过 GitHub 渲染 Markdown 的 512 KB 上限，后面的节显示不出来；[在线检索页](https://eternity4719.github.io/HowToLiveBetter/)会把这些文件合起来读，用法不变。 +正文按节拆成 34 个文件放在 [book/](book/)，点上面目录里的节名进入。拆开是因为单文件已经超过 GitHub 渲染 Markdown 的 512 KB 上限，后面的节显示不出来；[在线检索页](https://streamentry.github.io/HowToLiveBetter/)会把这些文件合起来读，用法不变。 ## 许可 @@ -293,7 +296,7 @@ node tools/pdf/build.mjs # PDF，另需 pandoc ≥ 3.1 和 typ - 附上许可证链接 https://creativecommons.org/licenses/by/4.0/ 。 - 改过内容的要写明改过。书里的法条、补贴标准和截止日期经常更新，建议同时写上你同步的是哪一天的版本。 -代码用 [MIT](LICENSE-CODE)，范围是 tools/、skills/、index.html 和 .github/。 +代码用 [MIT](LICENSE-CODE)，范围是 tools/、skills/、检索页模板和 .github/。 ## Star 走势 diff --git a/en/README.md b/en/README.md new file mode 100644 index 00000000..113c4155 --- /dev/null +++ b/en/README.md @@ -0,0 +1,310 @@ +
+ +A High-Return Life Guide — get back the most lifespan, money and personal freedom for the least money, time and energy + +# A High-Return Life Guide + +On living longer and getting sick less, and on what to do when someone has an accident. On not wasting money, and on which things get people scammed or sued. On what you can apply for when you have no job and no money, and what registration a shop, a company or a website needs. Also on dating, marriage and children, on going abroad, and on learning a trade.
+635 pieces of advice. Each one states what it costs, what it buys back, how strong the evidence is, and cites nothing but journal papers and official documents. + +You do not have to do all of it. This is a shortlist ordered by value for money, not a task list. Taking one or two entries with you counts. The author has not done most of them either. + +[![Search online](https://img.shields.io/badge/%E5%9C%A8%E7%BA%BF%E6%A3%80%E7%B4%A2-%E7%82%B9%E8%BF%99%E9%87%8C%E6%89%93%E5%BC%80-3451b2?style=flat-square)](https://streamentry.github.io/HowToLiveBetter/en/) +[![Items](https://img.shields.io/badge/%E6%9D%A1%E7%9B%AE-635%20%E6%9D%A1-18794e?style=flat-square)](#contents) +[![Evidence grades](https://img.shields.io/badge/%E8%AF%81%E6%8D%AE%E5%88%86%E7%BA%A7-A%20422%20%C2%B7%20B%20162%20%C2%B7%20C%2051-915930?style=flat-square)](#evidence-grades) +[![Original sources](https://img.shields.io/badge/%E5%8E%9F%E5%A7%8B%E6%96%87%E7%8C%AE-1407%20%E6%9D%A1%E9%93%BE%E6%8E%A5-565a5f?style=flat-square)](docs/核实记录/) +[![Licence](https://img.shields.io/badge/%E8%AE%B8%E5%8F%AF-CC%20BY%204.0-565a5f?style=flat-square)](#licence) + +### [Open the search page](https://streamentry.github.io/HowToLiveBetter/en/) · [Let an AI answer from the book (skill)](skills/life-decision-guide/README.md) + +The AI assistant skill works with Claude Code and Codex. Once it is installed, ask it something like "should I sign as a guarantor for a friend". It looks up the relevant items in the book first, then answers, and says which section and which item the answer comes from. + + + + + + + +
Download + +[PDF](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.pdf) · [EPUB](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.epub) · [Single-file offline HTML](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html) + +
Read + +[Contents](#contents) · [Glossary](#reading-the-numbers) · [Verification records](docs/核实记录/) + +
Long-form + +[Is marriage worth it](docs/结婚划不划算.md) · [Home emergency kit checklist](docs/家庭应急装备清单.md) · [Should you stop when a stranger is in trouble](docs/遇到陌生人出事该不该停.md) · [Which licences a platform needs](docs/做平台要办哪些证.md) · [Body clock and night shifts](docs/生物钟和夜班.md) + +
Other languages + +[English](https://streamentry.github.io/HowToLiveBetter/en/) and [Tiếng Việt](https://streamentry.github.io/HowToLiveBetter/vi/) are first-party translations maintained in this repository. [Русский](https://dlgrv.github.io/HowToLiveBetter/ru/) and [Español](https://dlgrv.github.io/HowToLiveBetter/es/) are community translations maintained by [dlgrv](https://github.com/dlgrv) ([repository](https://github.com/dlgrv/HowToLiveBetter)) + +
Related tools + +[howtolivebetter.net](https://howtolivebetter.net/), a checklist app by [littleben](https://github.com/littleben): add something to your to-do list, mark whether you have done it, bookmark it + +
+ +Community translations and related tools are maintained by other people and may fall behind. The Chinese original in this repository is the one to trust. + +
+ +--- + +## What this book answers + +| Question | Where to look | +| --- | --- | +| Which things cost almost nothing and clearly lower the risk of dying young? | [1. Do not die young](book/01-不要早死.md) | +| How much life does smoking, drinking, sitting too long and staying up late actually cost, how exactly do you quit smoking and cut down drinking, and which cooking oil to buy and how to use it? | [2. Do not die slowly](book/02-不要慢慢死.md) | +| Your energy runs out every day and you keep getting interrupted. What do you change? | [3. Do not waste energy](book/03-不要浪费精力.md) | +| Where does the time go, how do you cut out work that pays nothing back, and how do you deal with procrastination? | [4. Do not waste time](book/04-不要浪费时间.md) | +| Where should you put money you have saved, so that interest, fees and scams do not eat it? | [5. Do not waste money](book/05-不要浪费钱.md) | +| Which supplements, health-check packages and IQ taxes can you simply not buy? | [6. The reverse list](book/06-反面清单.md) | +| You have lost your job, or your pay, or you have no money at all. What can you claim, and where do you ask for help? | [7. How to get by with no money](book/07-没钱的时候怎么活.md) | +| Who does the bride price belong to, and who does a flat bought before the wedding? Who does a large transfer during a relationship belong to? If someone reports you to the police or reports you on facts they made up, what is the first thing to do? If you turn yourself in over something you did, how much does that reduce the sentence? Afterwards, can you still pursue the other party and claim compensation? | [8. Do not put yourself at risk: law and property safety](book/08-别把自己搭进去.md) | +| Which "side jobs" and small everyday favours turn an ordinary person into a defendant in a criminal case? | [9. Legal lines ordinary people cross](book/09-普通人容易踩的法律红线.md) | +| When you are dating, should you meet lots of people or fixate on one? Can a long-distance relationship work? What do you need to bring to register a marriage? | [10. Is dating and marriage worth it](book/10-恋爱和结婚划不划算.md) | +| Which code, and which freelance jobs, get you convicted? | [11. Lines programmers and tech workers cross](book/11-程序员和技术人容易踩的红线.md) | +| Before you borrow money to open a shop or a company, what should you think through first? | [12. Starting a business: do not lose the family savings](book/12-创业与做生意.md) | +| Someone has collapsed and is not breathing, someone is bleeding heavily, there is a fire, you are lost. What do you do first? | [13. Emergencies: what to do first](book/13-紧急情况.md) | +| An account has been stolen, or a phone is gone. What is the first thing to do? | [14. Accounts and information security](book/14-账号与信息安全.md) | +| The deposit is being withheld, the landlord is throwing you out, a long-let apartment has turned out to be a mess. What do you do? If you want to move to the countryside, can you buy a rural house? | [15. Renting and buying a home](book/15-租房与买房.md) | +| After a chronic illness has been diagnosed, how do you manage it over the long term, and how do you spend less? | [16. Living with a chronic illness](book/16-得了慢性病之后怎么活.md) | +| How do you sort out guardianship, wills and money for an elderly parent in advance? | [17. When there are elderly at home](book/17-家里有老人.md) | +| What can you get for having a child, and how much time and money does it take? | [18. Is raising children worth it](book/18-养孩子划不划算.md) | +| How are overtime pay and annual leave calculated? How much compensation should you get if you are dismissed? How is a workplace injury recognised, and what money comes with it? | [19. Working, leaving, and work injuries](book/19-在职离职和工伤.md) | +| A child has just been born. What are the few things that matter most? | [20. Caring for a newborn](book/20-刚出生的孩子怎么带.md) | +| Which countries are not worth going to now, and how far does consular help go when something happens? | [21. Travel and safety abroad](book/21-出国旅行与境外安全.md) | +| How do you avoid trouble in karaoke bars, internet cafés and escape rooms? What actually helps when you are stressed? | [22. Relaxing: venues and stress relief](book/22-怎么放松.md) | +| Learning welding, learning English, getting a certificate: which of these really pay back? How do you learn without wasting time? Where do you apply for a professional title, and is it worth it? | [23. Which skills are worth learning](book/23-学什么技能划算.md) | +| How much does the same illness cost at a community clinic compared with a top-tier hospital? If you are badly hurt, do you queue and register, or go to the emergency triage desk? After treatment, do you need a disability assessment or a disability certificate? | [24. Medical care: spending less and going fewer rounds](book/24-看病.md) | +| Someone in the family has died. What do you do first? Which money can you get back, and which charges do you not have to pay? | [25. What to do after someone dies](book/25-人走了以后要办什么.md) | +| If you take money through a website or a platform, which licences do you need, and where do you put the server? | [26. Running a website or platform: licences, filings, servers](book/26-做一个网站或平台.md) | +| You are pregnant, or about to give birth. What should you do when, and which certificates do you need before leaving hospital? | [27. Pregnancy and childbirth: from test to paperwork](book/27-怀孕和生产.md) | +| You want to lose weight or look better. Which methods will damage your body? | [28. Do not wreck your body for the sake of looks](book/28-别为了外形把身体搞坏.md) | +| A relative has died, you have been dismissed, or you have been diagnosed with a serious illness. What matters most in the first few months? | [29. After a severe blow](book/29-遭遇重大打击之后.md) | +| Once a child is at school, which health and mental health matters cannot wait until after the exam season? | [30. School-age children](book/30-上学以后的孩子.md) | +| After eighteen, what paths are there besides studying and working? What is the threshold for each? | [31. What are the paths after eighteen](book/31-十八岁之后有哪几条路.md) | +| If you study abroad, how do you keep your visa status unbroken? Will this qualification be recognised when you come back? | [32. Studying abroad: status, work, insurance, certification](book/32-出国留学.md) | +| You or a family member has a disability. Which complications do you prevent first? Which subsidies can you apply for? How do school, work and guardianship work out? | [33. Living after a disability](book/33-残疾之后怎么活.md) | +| You buy your own cold tablets, fever tablets and stomach tablets. Which ones must not be taken together? Which should children, pregnant women and elderly people avoid? | [34. Do not mishandle the medicines at home](book/34-家里的常备药别吃出事.md) | + +## How to read it + +- **You do not have to do all of it**: this is a shortlist ordered by value for money, not a task list. Taking one or two entries with you counts. Leave the rest and come back when you need them. The remark that this is easy to say and hard to do is fair. The author has not done most of it either. It is written down so that you can find it at the moment you need it. If you want the cheap ones, see "only the ones most worth doing" below. +- **If you want an AI to look things up**: the repository ships a Chinese skill ([skills/life-decision-guide](skills/life-decision-guide/)). Both Claude Code and Codex can install it. Once it is installed, ask it directly, for example "should I sign as a guarantor for a friend" or "is a two-hour daily commute worth it". It first pulls the relevant items out of the text, then answers, ranking them the way the book ranks things, and it says which section and which item the answer comes from. If it cannot find anything, it says so. It does not invent numbers. How to install it is in [the notes in that directory](skills/life-decision-guide/README.md). +- **If you want to filter by criteria**: open the [search page](https://streamentry.github.io/HowToLiveBetter/en/). You can filter by keyword, by section and by evidence grade. You can also filter on the three costs, money spent, time spent and willpower needed, and several filters can be combined. The text on the page is read straight from the files in `book/`. When the text changes, the page changes with it. +- **Items point at each other** (things like "see item 17 in section 8"). On the search page such a reference carries a dotted underline. Click it and the title and the "In plain language" line of the item it points to appear in place. To go there for real, press "Go there". If the filters happen to be hiding that item, the page clears the filters by itself. On GitHub, reading the text directly, these references are not clickable. But every one of them says what it points to, for example "see item 18 (write down the IOU when you lend money)". You know which item is meant without following it. +- **If you want to read in order**: within each section the items run from the best value for money to the worst. Start with the first few of each section. +- **If you want to read it offline or send it to someone**: download the [single-file offline HTML](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html). The whole book, search and filters, is in that one file. Double-click it and it opens. No server, no network. You can send it straight through WeChat. +- **If you want to print it or read it on a phone**: download the [PDF](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.pdf). A4 layout, a bit over two hundred pages, with page numbers in the contents and with bookmarks. Each section starts on a new page. +- **If you want to read it on a Kindle or another reader**: download the [EPUB ebook](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.epub). Send it over with Send to Kindle and it is there. +- **All three are regenerated automatically each time the text changes**, and the download links never change. A copy you have forwarded will not update itself. The online version is the current one. +- **If the numbers do not make sense**: every item has an "In plain language" line. It turns the research wording in the "Benefit" field into everyday terms, such as "about a fifth lower chance of dying in the same period" or "a few days of detention and a fine of so much". It only uses what is already in the "Benefit" field and adds no new numbers. That one line is enough to decide on. The "Benefit" field keeps every number as it was, so read it if you want to check for yourself. +- **If you only want the firmest conclusions**: tick evidence grade A on the search page. That leaves the 422 items that carry a specific number and come from a meta-analysis or a large trial. +- **If you only want the ones most worth doing**: set value for money to "Exceptional". That gives the 108 items that cost no money, no time and no willpower, and whose benefit sits in the largest band. Add a "What it buys back" filter on top and you have a priority list for that measure. +- **Do not be thrown by section titles that begin with "Do not"**: a section title names the outcome that section is trying to prevent, such as not dying young or not wasting time. It does not mean every item under it is telling you not to do something. The action itself is in the item title. Every item title starts with a verb, so it already says whether the item is about doing something or about not doing something. Both appear in the same section. Section 4 has both "write 'I plan to' as 'at what time, where, and what I do when this comes up'" and "do not watch television or scrolling news". Read the item titles. Do not carry the tone of the section title over to them. + +An item looks like this: + +```markdown +### 5. Swap the salt in your home for low-sodium salt (potassium salt) +- Cost: A bag costs a few yuan more than ordinary salt. Swap it when you next buy, it takes no extra time. The taste barely changes. +- In plain language: In a randomised trial of 20,000 people, the people who swapped the salt in their home for low-sodium salt had about a 12% lower chance of dying within five years, and about 14% lower stroke risk. The two groups were assigned at random, so this is more trustworthy than a figure that comes only from following records. +- Benefit: A trial that assigned people at random to two groups, run in rural China, 20995 people, all of whom had had a stroke or were over 60 with high blood pressure, followed for 4.74 years. Result: the low-sodium salt group had about a 12% lower risk of death than the ordinary salt group (RR 0.88). Stroke about 14% lower (RR 0.86). Major cardiovascular events about 13% lower (RR 0.87). +- Evidence grade: A +- Sources: Neal B et al (2021). Effect of Salt Substitution on Cardiovascular Events and Death. NEJM. https://doi.org/10.1056/NEJMoa2105675 +- Notes: Disputed. The people in the trial were high-risk elderly people, so a healthy young person gets much less out of swapping salt. If you have impaired kidney function, or you are taking a potassium-sparing drug, ask your doctor before you swap. +``` + +## Run your own copy + +Most people do not need to deploy anything. The [search page](https://streamentry.github.io/HowToLiveBetter/en/) is already there, and if you want it offline, download the [single-file offline HTML](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html). Double-click it and it opens. + +If you do want to run it on your own computer or your own server: + +```bash +git clone https://github.com/eternity4719/HowToLiveBetter.git +cd HowToLiveBetter +python -m http.server 8000 +``` + +Then open `http://localhost:8000/` in a browser. The search page is entirely static. `README.md` and `book/` are its data. There is no back end, no database, and nothing to install. Handing the whole directory to any static server (Nginx, GitHub Pages, object storage) works the same way. Note that `index.html` has to be opened over http. Double-clicking the local file gives you a blank page, because a browser is not allowed to read local files. For that case, use the single-file offline version above. + +To build the three electronic versions yourself (you should not need to, because the ones in the Release are generated automatically): + +```bash +cd tools/epub && npm ci && npm run build # EPUB +node tools/offline/build.mjs # single-file offline HTML +node tools/pdf/build.mjs # PDF, also needs pandoc >= 3.1 and typst >= 0.13 +``` + +The output all lands in `dist/`. + +## The four resources + +What this book wants to help you keep is not only lifespan. There are four things: + +- **Lifespan**: live longer, and die less of things that could have been avoided +- **Time and energy**: do not spend the time you are alive on things that pay nothing back, and waste less of the day's attention and strength +- **Money**: spend less on things that were not worth the money, and put money where the return is certain +- **Personal freedom**: do not end up in a detention centre or a remand centre because you did not know where one line was + +Every item answers two questions: what it costs (money / time / energy / willpower) and what it buys back (a change in all-cause mortality / a drop in a specific cause of death / time and energy saved / money saved / security and personal freedom). The items are ordered by value for money, not by category. Anything that costs almost nothing and buys back a lot goes at the top. + +**Who receives the benefit of an item is graded.** The grades run from the highest, which is how likely the benefit is ever to come back to you personally, to the lowest: ① **you**; ② **your spouse and immediate family** (parents, children, grandparents, grandchildren); ③ **friends, colleagues and other relatives**, where the relationship goes both ways and a favour given now may come back later; ④ **strangers**, the lowest band, but not zero. The chance of a return is small, you do not know what the other person is like, and there is also a risk of being extorted, of being turned against and of revenge. The bands are not added together. When an item reaches band ④, the benefit and the risk are written out together. + +Read the first aid section with this in mind. Of 38,227 out-of-hospital cardiac arrests in China, 79.2% happened at home. Learning to press is first of all so that you can press on your own family. Rules like "do not go into the water when you see someone drowning" and "do not step in to break up a fight" are self-protection rules in themselves. They keep you from turning from a bystander into the second person hurt. Whether to step in for a stranger is your own trade-off. The items set out the disclaimers, the self-protection steps and the risk, and they do not argue on your behalf that you have to do it. + +The mortality numbers, the time and energy numbers, the money numbers and the legal consequences are counted separately and are not converted into each other. These four are the four "What it buys back" measures on the search page, and they are not compared against each other. + +## Evidence grades + +Every item carries an evidence grade: + +| Grade | What it means | +| --- | --- | +| A | There is a specific number you can look up. The source is a meta-analysis pooling several studies, a large cohort followed for many years, or a randomised controlled trial (RCT). It can say how much lower the risk is (HR, RR, percentage drop) | +| B | There is research behind it, but no single definite number. Or it rests on a small sample, or on one study alone | +| C | The author's own experience, or a practice everyone accepts. No direct research literature | + +Across the book's 635 items, 422 are grade A, 162 are grade B and 51 are grade C. A further 59 items are marked as disputed, and 2 carry a TODO for verification. A disputed item of grade A or B is marked "Disputed" and lists the evidence on the other side. Every source is an original document: a journal paper with a DOI or a PubMed link, or a report from an official body such as the WHO, the CDC or the National Bureau of Statistics. Secondhand retellings are not cited. A number that is not certain is marked "to be verified". + +## Value for money + +The evidence grade only answers whether a number can be trusted. It does not answer whether the thing is worth doing. So every item carries two more labels: the size of the benefit, which is how big the benefit is, and the measure, which is which kind of thing it buys back. The search page takes those two together with the three costs and works out a value-for-money band: + +| Dimension | Values | How it is decided | +| --- | --- | --- | +| Measure | Lifespan / money / time and energy / personal freedom | Look at what the item mainly buys back. **Different measures are not compared with each other.** A 12% drop in all-cause mortality and 500 yuan saved a year are not on the same ruler, and there is no ranking between them | +| Size of the benefit | Large / medium / small | Follow the item's own "Benefit" field and apply thresholds settled in advance rather than going by feel. For lifespan, look at the percentage drop (≥20% is large, 10–20% is medium, under 10% is small, and so is something measured only on a surrogate marker and not on the final outcome). For money, look at the amount (tens of thousands of yuan is large, hundreds to thousands is medium, a few tens of yuan is small). For personal freedom, look at the consequence (avoiding criminal liability is large, avoiding detention or an administrative penalty is medium, avoiding a civil dispute is small). For time and energy, look at how much is saved (an hour a day is large, hours a week is medium, saving it once is small) | +| Value for money | Exceptional / high / moderate | A large benefit with all three costs at zero = exceptional. A large benefit with low costs, or a medium benefit with zero costs = high. Everything else = moderate | + +Across the book's 635 items, 108 are exceptional (17%), 291 are high (46%) and 236 are moderate (37%). The middle band has the most items on purpose. The size of the benefit underneath it only has three levels, large, medium and small. Cutting it finer would be false precision. + +**This band is the author's own judgement, not evidence.** By the standard of this book it counts as grade C in itself. It is a different thing from the evidence grade, and neither affects the other. An item can be grade A on evidence and only moderate on value (the shingles vaccine comes from a phase III RCT with 97.2% efficacy, but two doses cost three or four thousand yuan, and shingles rarely kills). An item can also be grade C on evidence and exceptional on value (send your itinerary to your family before you go abroad). Moderate does not mean do not do it. Every item in the book is recommended. This band only means that you have to weigh that expense yourself. + +## Reading the numbers + +The text sticks to everyday wording wherever it can, but a few statistical terms are unavoidable when a study is cited. Look them up in this table if you do not recognise one. On the search page, hovering over a dotted-underlined term (or tapping it on a phone) brings up the same explanation. + +
+Show 41 terms (all-cause mortality, HR, RR, 95% CI, meta-analysis, BMI, LPR, earnest money and a deposit ...) + +| Term | What it means | +| --- | --- | +| All-cause mortality | The share of a group of people who die over a period of time, whatever the cause. The book uses it to measure how long people live. Research papers call it all-cause mortality, abbreviated ACM | +| HR | Hazard ratio. Over the same period, how fast the group that did something had events (death, illness), divided by the group that did not. HR 0.87 means 13% lower than the group that did not do it, HR 1.21 means 21% higher | +| RR | Relative risk. The ratio of the chance of an event between two groups. The numbers read the same way as HR | +| OR | Odds ratio. Also a comparison between two groups, but the calculation differs a little from RR. When the event is rare, it is close to RR. When the event is common, it makes the gap look bigger than it is | +| IRR | Incidence rate ratio. The incidence rate is the share of people who newly get the disease over a period of time. Divide one group's rate by the other's. It reads the same way as RR | +| RaR | Rate ratio. It compares how many times something happened, for example how many falls, not how many people had an event. It reads the same way as RR | +| Standardised mortality ratio | Abbreviated SMR. How many people in this group actually died, divided by how many people of the same age would be expected to die. 5.86 means the group died 5.86 times as often as people the same age | +| Risk difference | Subtract the chance of an event in one group from the other, which gives the plain figure of how many extra cases there are per thousand. A ratio only says how many times higher it is. The risk difference says how many more people there actually are | +| 95% CI | 95% confidence interval. A figure from a study always carries some error, and this is the range the true value is likely to fall in. For a ratio, a range that crosses 1. For a figure with a plus or minus, a range that crosses 0. Either way the difference between the two groups may be chance, and the text then says "not statistically significant" | +| RCT | Randomised controlled trial. People are split into two groups by something like drawing lots, one group does the thing and one does not, and afterwards the results are compared. This is the most convincing way to show that the thing itself did the work | +| Meta-analysis | The results of many studies are pooled together and recalculated to give one overall figure | +| Cohort | A cohort study. A large group of people is picked out, followed for many years, and it is recorded who had an event. It can show that two things go together, but not that the first one caused the second | +| Observational | An observational study. The researcher only records from the side and does not assign who does what. Other factors can pull the result around, see confounding and reverse causation in this table, so read the numbers with some discount | +| Confounding | A third factor moves both the cause and the outcome at once, so that two things look causally related. People who eat a lot of vegetables often exercise more as well, and it is not clear which one is doing the work | +| Reverse causation | The direction of cause and effect is the wrong way round: it looks as if the first thing caused the second, when in fact the second caused the first. It is not that sleeping more makes you die early. It is that people who are seriously ill sleep more | +| d, g | Effect size, showing how big the gap between the two groups is. The gap is converted into how many times the usual amount of day-to-day variation it comes to: 0.2 counts as small, 0.5 as medium, 0.8 as large | +| r | Correlation coefficient. How closely two things move together, from -1 to 1. 0.1 counts as weak, 0.3 as medium, 0.5 as strong | +| MET | A unit of exercise intensity. Sitting quietly is 1 MET, brisk walking is about 3 to 4 MET. Written as MET·h it is intensity multiplied by the hours spent, which is the total amount of exercise | +| GRADE | A scoring system used around the world that rates how reliable a piece of evidence is, in four levels: high, moderate, low and very low | +| Intention to treat | When the effect is worked out, it is counted on the people who were told to go for screening, not on the people who actually went. Some of the people told never go, so this method makes the benefit of screening look smaller than it is for the people who do go | +| Pack-year | The unit for how much a person has smoked in total. Packs a day multiplied by the years smoked. 30 pack-years is a pack a day for 30 years | +| BMI | Body mass index, the usual number for weight against size: weight in kilograms divided by height in metres squared | +| LDL | Low-density lipoprotein cholesterol, usually called bad cholesterol | +| eGFR | Estimated glomerular filtration rate, a measure of how well the kidneys filter, in mL/min/1.73 m². The lower the number, the worse the kidney function, and the stages of chronic kidney disease are set by it | +| HBsAg | Hepatitis B surface antigen, one line on a lab report. A positive result means the hepatitis B virus has already been caught | +| HPV | Human papillomavirus. It has many types, and long-term infection with some of them leads to cervical cancer | +| LDCT | Low-dose chest CT. The radiation is about one fifth to one tenth of an ordinary CT | +| PM2.5 | Fine particles floating in the air, under 2.5 micrometres across | +| NOVA | A way of sorting food into four groups by how far it has been processed. "Ultra-processed food" is the fourth of them | +| LPR | The Loan Prime Rate, the benchmark for lending rates in China. It is published on the 20th of each month, and both the mortgage rate and the ceiling on interest in private lending are calculated from it | +| Final award | The result of labour arbitration takes effect at once, and the employer cannot take the same matter to court | +| Crude marriage rate, crude divorce rate | The number of marriages and divorces registered in the year, per thousand people. There is a third figure, the divorce-to-marriage ratio, which is the number of divorces divided by the number of marriages in the same year. It is not the same thing as the other two, so do not read them together | +| AED | Automated external defibrillator. The red or yellow emergency box common in public places. Switch it on and follow the voice prompts. Whether to deliver a shock is decided by the machine itself | +| CPR | Cardiopulmonary resuscitation. Press the chest hard when the heart has stopped, so that the blood keeps moving | +| 3C certification | China Compulsory Certification, China's mandatory product certification. A product on that list may not leave the factory and may not be sold unless it carries this mark | +| ICP filing | A registration formality that a website or app has to complete in the Ministry of Industry and Information Technology's system before it goes live | +| Classified protection | The classified protection system for network security. A system is sorted into one of several levels by how important it is, and the higher the level, the more security measures are required | +| GPL | An open-source licence. Something built with code of this kind usually has to release its own code as well when it is distributed | +| Non-compete clause | An agreement signed with the employer: for a period after you leave, you do not go to work for a competitor. The employer has to pay you monthly during that period, for at most 2 years | +| Subscribed capital | The money you promise to put in when you register a company. Once you have promised it, it has to be paid in within the period the law sets. It is not a number written down to look good | +| Earnest money and a deposit | Two words that differ by one character, and the difference in effect is large. 订金 is money paid in advance, with no penalty attached if someone changes their mind. 定金 carries a penalty: the party who took the money has to repay double if they change their mind, up to 20% of the contract value | + +
+ +## Contents + +1. [Do not die young](book/01-不要早死.md): deaths from external causes, gas and poisoning, vaccines, screening, the timescale of a mental crisis or suicidal thoughts, what is left after someone is brought back, the year after a serious fall, the arithmetic of the kidney you keep after selling one, home emergency kits, and signals such as visible blood in the urine that should be checked. Measure: all-cause or cause-specific mortality. +2. [Do not die slowly](book/02-不要慢慢死.md): tobacco and alcohol, exercise, sleep, diet (low-sodium salt, nuts, whole grains, processed meat, bulk cold-pressed peanut oil, vegetable oil instead of lard), sitting too long, and the practical ways to quit smoking and cut down drinking (quit medication, quit dates, smoking cessation clinics and helplines, e-cigarettes, alcohol withdrawal not something to push through alone), how long a nap should be, how to make up for a late night, and the arithmetic of years spent on night shifts. Measure: all-cause or cause-specific mortality. Long-form essay: [docs/生物钟和夜班.md](docs/生物钟和夜班.md). +3. [Do not waste energy](book/03-不要浪费精力.md): sleep, interruptions, multitasking, decision fatigue, social debt, and what to expect when you deal with institutions. Measure: energy / time. +4. [Do not waste time](book/04-不要浪费时间.md): projects that pay nothing back, sunk costs, procrastination (the emotional explanation, changing the environment, commitment devices, how long a habit takes, self-help material), meetings, commuting. Measure: time. +5. [Do not waste money](book/05-不要浪费钱.md): subscriptions, lottery tickets, interest, insurance, fund fees, personal pensions, car insurance, prepayments, live-stream shopping, personal medical insurance accounts, children being scammed and refunds for top-ups, prayer beads, watches and designer toys not counted as investments, how to check a jewellery or jade appraisal report, blind boxes and gacha, and which legal channel to use to buy assets overseas. Six items at the end of the section are about insurance: only insure a loss you cannot absorb, earners buy a term life policy first, use the cooling-off period to get your money back, do not let a bank counter sell you insurance as a deposit, do not use an agent to surrender a policy, and beneficiaries and the time limit for claims. Measure: money. +6. [The reverse list](book/06-反面清单.md): things that look like good value but are not, including the claim that willpower runs out. +7. [How to get by with no money](book/07-没钱的时候怎么活.md): assistance, subsidies, finding work, somewhere to stay and food, medical care, recovering unpaid wages, and the traps to avoid. Measure: money / security. +8. [Do not put yourself at risk: law and property safety](book/08-别把自己搭进去.md): traffic accidents, stopping payment on a scam, AI face swaps and cloned voices, how much a voluntary surrender and a truthful statement reduce the sentence and why "staying outside the prosecution period" does not work, what routes are open after being accused or after someone reports you on facts they made up, where asking someone for money in exchange for dropping a report ends and an ordinary claim begins, conflict and extreme violence out of rage, impulsive harm and the right of the people around you to take you for treatment, the four legal gates between taking out a policy on a relative and doing them harm, platform routes and court orders after being piled on online, bride price, property owned before the wedding, guarantees, anti-fraud, the statute of limitations, enforcement and the judgment defaulter list, liability for dogs, the case acceptance receipt after calling the police and what to do if no case is opened, and paying money to settle things being the crime of bribery. Measure: money / personal freedom. +9. [Legal lines ordinary people cross](book/09-普通人容易踩的法律红线.md): rumours, insulting martyrs, viewing overseas content without reposting it, spreading pornography, money laundering through a side job, forging documents to obtain a loan, faking accidents to claim insurance, throwing things from height, imitation guns, drones, hidden cameras, legal adoption when you cannot raise a child and the line against trafficking and abandonment, gambling, wild game, selling organs and finding donors for other people. Measure: personal freedom / money. +10. [Is dating and marriage worth it](book/10-恋爱和结婚划不划算.md): choosing a partner, the lines around pestering someone, signals of interest, the quality of a relationship, long distance, the registration procedure, the premarital medical check, the health arithmetic, the time arithmetic, the money arithmetic, parents paying for a flat and joint marital debt, and the cost of getting out. Long-form essay: [docs/结婚划不划算.md](docs/结婚划不划算.md). +11. [Lines programmers and tech workers cross](book/11-程序员和技术人容易踩的红线.md): cheat software, web scrapers, ticket-grabbing scripts, deleting the database, taking source code away, taking contract development work, non-competes, open-source licences, filings. Measure: personal freedom / money. +12. [Starting a business: do not lose the family savings](book/12-创业与做生意.md): capital, guarantees, choosing the form of the entity, registration, licences, tax filing, invoices, tax fraud, contracts, hiring, mass production, buying stock and the intellectual property line around copying someone else's drawing, and getting out. Measure: money / legal liability. +13. [Emergencies: what to do first](book/13-紧急情况.md): cardiac arrest, stroke and posterior circulation stroke, eye stroke, heart attack, aortic dissection, thunderclap headache, chronic subdural haematoma, pulmonary embolism, severe bleeding, bites, burns and scalds, anaphylactic shock, epilepsy, low blood sugar, electric shock, carbon monoxide, swallowed chemicals and chemical burns, objects embedded in the body, immobilising a fracture, scams, threats to privacy, heatstroke, fire, drowning, being lost, hypothermia, snake bites, earthquakes, wild animals, lightning, altitude sickness, ticks, drinking water outdoors. And whether someone can be saved at all: how to help up an elderly person who has fallen, what to do when you walk into a fight, and who pays when helping someone gets you hurt. Measure: survival and money, with the last few items also touching on personal freedom. +14. [Accounts and information security](book/14-账号与信息安全.md): second-factor verification, passwords, the SIM card, a lost phone, card fraud, logged-in devices, app permissions, face recognition, the right to check and to delete. Measure: money / personal information. +15. [Renting and buying a home](book/15-租房与买房.md): deposits, eviction by force, agents holding the money, escrow of funds, a sale does not break a lease, checking the title, a dedicated account for the money, partitioned rooms, and do not buy a rural homestead plot with an urban hukou, only rent a rural house. Measure: money. +16. [Living with a chronic illness](book/16-得了慢性病之后怎么活.md): taking the medicine, out-of-province settlement for chronic and serious outpatient disease, keeping test records, not stopping the medicine to try a folk remedy, long prescriptions, signing up with a family doctor, screening for complications, preventing kidney stones from coming back, and getting gout down to target. Measure: all-cause mortality / money. +17. [When there are elderly at home](book/17-家里有老人.md): voluntary guardianship, forms of will, accounts and the sales pitch, investment and reverse mortgage scams, long-term care insurance, and pressure sore prevention for someone bedridden long term. Measure: money / personal freedom, with the pressure sore item being about mortality. +18. [Is raising children worth it](book/18-养孩子划不划算.md): childcare subsidies, maternity leave and the maternity allowance, the three periods of protection, the time arithmetic, the money arithmetic. Measure: money / time. +19. [Working, leaving, and work injuries](book/19-在职离职和工伤.md): overtime pay, annual leave, probation; notification of occupational hazards and the three medical examinations, dust and noise protection; N, severance in lieu of notice, 2N, do not sign a voluntary resignation, keep your evidence; the time limit for recognising a work injury, an employer with no insurance, assessment of the ability to work, and benefits in case of death. Measure: money. +20. [Caring for a newborn](book/20-刚出生的孩子怎么带.md): safe sleep, the first hepatitis B shot, the immunisation schedule, breast milk and solids, the temperature for made-up formula, honey, vitamin K, the point at which a fever means seeing a doctor, never shaking the baby, nappies and bulk buying, and introducing peanuts early for a high-risk baby to prevent allergy. Measure: infant mortality / money. +21. [Travel and safety abroad](book/21-出国旅行与境外安全.md): travel advisory levels, 12308, the limits of consular protection, overseas medical insurance, high-pay job offers abroad, the annual limit on cash withdrawals abroad, lost documents, overseas driving licences, and agent registration. Measure: money / personal freedom. +22. [Relaxing: venues and stress relief](book/22-怎么放松.md): emergency exits, displayed prices, the lines around drugs, things other people hand you, real-name registration at internet cafés, choosing the site for a murder mystery game; exercise, mindfulness, breathing, company, green space. Measure: money / personal freedom, as well as energy / all-cause mortality. +23. [Which skills are worth learning](book/23-学什么技能划算.md): studying or working (the minimum age for child labour, education and mortality, the national education structure, tuition waivers, grants and student loans, routes up from vocational secondary school, how to work out this arithmetic yourself), the return on education, certificate mills, training subsidies, which skills are hard for machines to replace, skill grades, how to look up shortage occupations, and once you have decided what to learn, how to learn it (self-testing, distributed practice, do not rely on highlighting, interleaving, there is no evidence for learning styles), and finally professional titles (where to apply, examination in place of assessment, the consequences of a faked assessment, being titled is not the same as being appointed to the post). Measure: money / time, with one item about mortality. +24. [Medical care: spending less and going fewer rounds](book/24-看病.md): the tiered system and referrals, the deductible counting continuously, differences in the reimbursement rate, reserved appointment slots, assessing whether treatment away from home is necessary, keeping and sealing medical records, the four levels of emergency triage, disease emergency assistance when you cannot pay, when to have a disability assessed, how to get a disability certificate, and no red envelopes for doctors. Measure: money / time. +25. [What to do after someone dies](book/25-人走了以后要办什么.md): calling the police and the death certificate, collecting the body and cremation, objections to the cause of death and the autopsy, cancelling the household registration, the list of basic funeral items, price violations, agent registration, the housing provident fund balance and social insurance benefits, and the deceased's personal information rights. Measure: money. +26. [Running a website or platform: licences, filings, servers](book/26-做一个网站或平台.md): payment and settlement red lines, the ICP licence and filing, qualifications for live streaming and audio-visual services, platform verification and tax reporting, content governance, real-name checks, minors, notice and takedown, data export, and choosing a server. Measure: personal freedom / money. Long-form essay: [docs/做平台要办哪些证.md](docs/做平台要办哪些证.md). +27. [Pregnancy and childbirth: from test to paperwork](book/27-怀孕和生产.md): folic acid, the maternal health record and free antenatal checks, screening for the three infections and blocking transmission from mother to child, smoking and drinking in pregnancy, aspirin and gestational diabetes, the signs that mean go to hospital now, what to do when the waters break, epidural analgesia, the indications for a caesarean section, maternity insurance, the medical certificate of birth, newborn screening, insurance and household registration, and the check at 42 days after the birth. Measure: mortality and money. +28. [Do not wreck your body for the sake of looks](book/28-别为了外形把身体搞坏.md): extreme dieting and eating disorders, cosmetic clinics and the two certificates the attending physician must hold, facial filler and the parts of the face that go blind, weight-loss products with sibutramine added illegally, anabolic steroids, prescriptions and follow-up for weight-loss drugs and sex hormones, and body image assessment. Measure: mortality (a health outcome), with the two cosmetic medicine items also touching on personal freedom. +29. [After a severe blow](book/29-遭遇重大打击之后.md): the cardiovascular window in the first month after a bereavement, the first week after a diagnosis of serious illness, losing a job, the six months after losing a spouse, bereaved by suicide, what to do when you have neither family nor friends to take the place of that person, children whose parents have died, where to register when grief has stuck, divorce, 12356 and 12355, not making irreversible decisions in the acute stress period, and why using death to settle a debt does not work. Measure: all-cause mortality, with the four items about money and benefits counted under money. +30. [School-age children](book/30-上学以后的孩子.md): emergencies counted by the hour, not delaying treatment for the sake of exams, what to do about bullying, 2 hours outdoors a day, the school medical report, screening for depression in teenagers, products that claim to cure myopia, hard rules on sleep and homework, taking a leave of absence while keeping the student record, dilated refraction and follow-up, and fissure sealants. Measure: mortality and health outcomes, plus one or two items each on money and time. +31. [What are the paths after eighteen](book/31-十八岁之后有哪几条路.md): the statutory threshold for each of twelve paths; national service (military registration, two years of compulsory service, joint penalties for refusing, tuition compensation and progression into education, resettlement and reporting within 30 days, the discharge payment and its effect on seniority and tax); targeted recruitment for community service posts; special-post teachers moving onto the permanent staff at the end of their term; fire fighters and civilian posts in the armed forces; the self-study examination, the adult higher education examination and the open university; public-funded normal university students and targeted medical students, and their 6 years of service; which companies are worth looking for when working abroad; the individual tax and the money brought back when working remotely for an overseas company from home; guaranteed startup loans; social insurance under flexible employment; and injury cover for delivery riders. Measure: money / time, with the item on refusing national service also touching on personal freedom. +32. [Studying abroad: status, work, insurance, certification](book/32-出国留学.md): check the list of recognised institutions before paying the tuition; the US F-1 rule of "at most four years, leave within 30 days of finishing" has been suspended by a court, so for now it is until you finish, with a 60-day grace period; the working hour limits in the US, Canada, the UK and Australia; full-time study is the root of your status; reporting a move within 10 days; the Ministry of Education's study-abroad warning; OSHC in Australia must not lapse; the UK immigration health surcharge; 10 to 20 working days for certification; and the list of institutions under enhanced scrutiny. Measure: money / personal freedom. +33. [Living after a disability](book/33-残疾之后怎么活.md): the three steps on the spot for autonomic dysreflexia, the suicide window in the ten years after a disabling injury, the principle of voluntary admission for psychiatric care and its two exceptions, the death risk of the carer, pressure-relieving cushions for a wheelchair, "healing" scams, the six benefits to ask about once you have the certificate, long-term care insurance that is not only for the elderly, rehabilitation assistance for ages 0 to 6, grants for making the home accessible, quota employment and the disability employment security fund, the individual income tax reduction, guide dogs and free travel, reasonable accommodation in the national college entrance examination, the duty not to refuse a place and home tuition, the C5 licence, how to choose a rehabilitation provider, hearing aids, assessment of capacity and guardianship. Measure: mortality / money / time / personal freedom. +34. [Do not mishandle the medicines at home](book/34-家里的常备药别吃出事.md): do not double up on paracetamol, no aspirin, nimesulide or metamizole for a child's fever, the groups at high risk of stomach trouble from ibuprofen, no compound cold medicine of your own accord under 2, no ibuprofen of your own accord after 20 weeks of pregnancy, no more than 7 days of omeprazole of your own accord, no antibiotics for a cold, rehydrate first for diarrhoea and no antidiarrhoeal for children, and more painkillers ending in more headache. Measure: mortality. + +Within each section the items run from the best value for money to the worst. A section title such as "Do not die young" or "Do not waste time" names the outcome that section is trying to prevent. Whether an item is to be done or not done is settled by the item title. For the long-form essays, see [docs/家庭应急装备清单.md](docs/家庭应急装备清单.md), [docs/做平台要办哪些证.md](docs/做平台要办哪些证.md), [docs/结婚划不划算.md](docs/结婚划不划算.md), [docs/遇到陌生人出事该不该停.md](docs/遇到陌生人出事该不该停.md) and [docs/生物钟和夜班.md](docs/生物钟和夜班.md). The verification trail for every source is recorded in [docs/核实记录](docs/核实记录/). + +The `index.html` in the repository root is the online search page. It filters items by keyword, section, evidence grade and the cost dimensions (money spent, time spent, willpower needed), and it reads its data straight from this file. It becomes available once GitHub Pages is switched on in the repository settings (Deploy from a branch, branch main, directory /). `tools/epub/` holds the ebook build script, and `cd tools/epub && npm ci && npm run build` produces one EPUB in `dist/` locally. GitHub Actions runs the same script after every change to the text and updates the Release. + +## The text + +The text is split by section into 34 files in [book/](book/). Click a section name in the contents above to go in. It is split because a single file has already gone past the 512 KB limit GitHub allows when rendering Markdown, and the sections after that point stop showing. The [search page](https://streamentry.github.io/HowToLiveBetter/en/) reads these files together, and the way it works has not changed. + +## Licence + +The text is published under [CC BY 4.0](LICENSE). The scope is the text of `book/`, `docs/` and this README. You may reprint it, adapt it and use it commercially, and you do not have to ask the author. Three things are required: + +- Credit the source: "A High-Return Life Guide", with a link to the repository https://github.com/eternity4719/HowToLiveBetter . +- Include a link to the licence https://creativecommons.org/licenses/by/4.0/ . +- Say that you changed it, if you did. The legal provisions, subsidy standards and deadlines in the book change often, so it is worth also saying which day's version you synced to. + +The code is under [MIT](LICENSE-CODE). The scope is `tools/`, `skills/`, `index.html` and `.github/`. + +## Star history + +[![Star History Chart](https://api.star-history.com/svg?repos=eternity4719/HowToLiveBetter&type=Date)](https://star-history.com/#eternity4719/HowToLiveBetter&Date) + +## Buy the author a coffee + +If it has been useful to you, you can scan the code and buy the author a coffee. It makes no difference either way, and it changes nothing here. + +WeChat reward code + +## Advertisement + +Eternal World Minecraft server, game address 1.mcyyy.com diff --git "a/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" "b/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" new file mode 100644 index 00000000..20f09aef --- /dev/null +++ "b/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" @@ -0,0 +1,353 @@ +[← 回总目录](../README.md) + +# 1. 不要早死 + +这一节讲三件事：车祸、火灾、中毒这类意外和伤害死亡；少数几种证据最硬的疫苗和筛查；身体出了什么信号就该去查一查。收益那一栏的数字，来自随机分组的试验、跟踪一大批人很多年的研究，或者官方统计。成本那一栏是作者按市面价格的粗略估计，只用来互相比较，不要拿去引用。 + +[← Back to contents](../README.md) + +# 1. Do not die young + +This section covers three things: deaths from accidents and injuries such as car crashes, fires and poisoning; the few vaccines and screening tests with the strongest evidence; and which signals from your body mean you should get checked. The numbers in the Benefit field come from randomised trials, from studies that follow a large group of people for many years, or from official statistics. The Cost field is the author's rough estimate at market prices, meant only for comparing the entries with each other. Do not quote it. + +### 1. Wear a seat belt, in the front and in the back + +- Cost: No money. It adds 2 seconds each time you get in the car. +- In plain language: A front-seat passenger with a seat belt on has about half the chance of being killed in a crash. In the front seat of a light truck the fall is 60%. In the United States in 2022, half of the people who died in a vehicle and whose belt use could be determined were not wearing one. +- Benefit: The US National Highway Traffic Safety Administration (NHTSA) estimates that a seat belt lowers the risk of fatal injury for a front-seat car passenger by 45%. For the front seat of a light truck it is 60%. Among people who died in a passenger vehicle in the United States in 2022, 50% of those whose belt use could be determined were not wearing one. The World Health Organization (WHO) puts the reduction in death risk for people inside the vehicle at up to 50%. WHO estimates that 248,099 people died from road traffic injuries in China in 2021, which is 17.4 per 100,000 people +- Evidence grade: A +- Sources: NHTSA (2024). Occupant Protection in Passenger Vehicles: 2022 Data (DOT HS 813 573). ; WHO (2025). Road traffic injuries fact sheet. ; WHO GHO RS_196/RS_198 (China, 2021). +- Notes: The back seat needs one too. In the same report, 60% of the people killed in the second row were not belted. The two figures, 45% and 60%, are estimates the US National Highway Traffic Safety Administration (NHTSA) calculated from its crash database, citing Kahane 2015. They are not from a randomised trial. WHO's China death count is modelled, and it runs several times higher than the figure registered by the public security traffic police. The two cover different populations, so only the WHO figure is used here. + +### 2. Wear a helmet on a motorcycle or an electric bike, and fasten the strap + +- Cost: A helmet costs 100 to 300 yuan. Putting it on takes 5 seconds each time. +- In plain language: A motorcycle rider with a helmet and a fastened strap has about 40% lower risk of dying in a crash. Head injuries are about 70% lower. The strap has to be tight. A helmet hanging loose on the head does not count. +- Benefit: Cochrane pooled several studies together. Helmet use among motorcycle riders lowered the risk of death by 42%. The original value is OR 0.58, credible range 95% CI 0.50 to 0.68. Head injury fell 69% (OR 0.31, 95% CI 0.25 to 0.38) +- Evidence grade: A +- Sources: Liu BC et al (2008). Helmets for preventing injury in motorcycle riders. Cochrane Database of Systematic Reviews. +- Notes: The strap has to be fastened. A helmet left hanging on the handlebars does not count. The figures above come from motorcycles. There is no direct study of electric bikes, so they are carried over, because head injuries after a crash happen the same way. + +### 3. Fit smoke alarms; add a carbon monoxide alarm if you burn coal or gas indoors in winter + +- Cost: A smoke alarm costs 30 to 100 yuan. A carbon monoxide alarm costs 50 to 150 yuan. After it is fitted, replace the battery once a year. +- In plain language: A working smoke alarm cuts the chance of dying in a home fire by about 60%. In 40% of fires people are asleep when the fire starts, and smell alone will not wake them. If you burn coal or gas indoors in winter, add a carbon monoxide alarm. In 2018 about 11,500 people in China died of carbon monoxide poisoning, and 70% of those deaths in the three winter months happened at home. +- Benefit: A study in North Carolina took a batch of residential fires and looked back at the households that had one and the households that did not. A working smoke alarm in the home lowered the risk of death by about three fifths. The original value is OR 0.39, credible range 95% CI 0.18 to 0.83. In fatal fires in occupied homes in the United States from 2018 to 2020, 24% of the homes had no smoke alarm, and 41% of the people were asleep when the fire started. China reported 11,523 carbon monoxide poisoning deaths in 2018. Of those deaths, 72.59% happened at home in month 12, 67.42% in month 1 and 66.48% in month 2 +- Evidence grade: B +- Sources: Marshall SW et al (1998). Fatal residential fires: who dies and who survives? JAMA. ; USFA (2022). Fatal Fires in Residential Buildings (2018-2020), Topical Fire Report Series 22(2). ; You J et al (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly. +- Notes: The reason for the B: on the smoke alarm side there is only one study, and its method was to look back at households that had a fire and households that did not. It was not a randomised trial. There is no study of carbon monoxide alarms lowering mortality on their own. The benefit here is inferred from the pattern that poisoning deaths cluster in winter and in the home. + +### 4. Replace gas hoses and stoves when they reach end of life, do not repipe the gas yourself, and refuse gas sales at the door + +- Cost: A compliant gas hose costs a few tens of yuan. Replace the stove every few years. Check whether it has passed its expiry date while you are at it. +- In plain language: A compliant hose costs a few tens of yuan. Replacing it when it reaches end of life is a duty of the user written into the Regulations on the Administration of Urban Gas. Repiping the gas yourself or taking out the gas meter is forbidden by the same regulations. Those regulations also say a gas company cannot require you to buy its designated products and services. A salesman at your door saying the hose "must be replaced" can be refused outright. +- Benefit: The regulations make it a duty of the user to "replace in good time gas appliances and connecting pipes that the state has ordered withdrawn or that have reached the end of their service life": once a stove or a hose reaches end of life, or the state withdraws it, you have to replace it. The regulations also list "installing, modifying or removing indoor gas facilities and gas metering devices without authorisation" as a forbidden act, and that covers repiping the gas and taking out the meter yourself. The regulations also state that a gas operator may not require a user to buy its designated products, and may not require a user to accept the services it provides +- Evidence grade: A +- Sources: State Council (国务院) (2010). Regulations on the Administration of Urban Gas, 城镇燃气管理条例（国务院令第 583 号）. 第二十七条: 「燃气用户应当遵守安全用气规则，使用合格的燃气燃烧器具和气瓶，及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管等」 — gas users must follow safe-use rules, use qualified gas appliances and cylinders, and replace in good time any gas appliance or connecting pipe that the state has ordered withdrawn or that has reached the end of its service life. The prohibited acts in 第二十八条 include 「（一）擅自操作公用燃气阀门」「（二）将燃气管道作为负重支架或者接地引线」「（四）擅自安装、改装、拆除户内燃气设施和燃气计量装置」 — operating a public gas valve without authorisation, using a gas pipe as a load-bearing support or as an earthing conductor, and installing, modifying or removing indoor gas facilities and gas metering devices without authorisation. 第二十条 says a gas operator may not 「（六）要求燃气用户购买其指定的产品或者接受其提供的服务」 — require a gas user to buy its designated products or accept the services it provides. 第二十九条 lets a user complain to the gas authority about charges and service, and 「有关部门应当自收到投诉之日起15个工作日内予以处理」 — the department must handle it within 15 working days of receiving the complaint. ; Shenzhen Municipal Housing and Construction Bureau (深圳市住房和建设局) (2024). Gas safety notes, 安全用气小知识（二）: 「立即关闭家里的燃气总阀门」「迅速开门、开窗进行通风」「一定不要开、关家里的电灯、抽油烟机等电器」「人员应该迅速撤离，转移到安全的地方，并且及时拨打燃气公司的报警电话」 — close the main gas valve at home, open the door and windows to ventilate, never switch the lights or the cooker hood on or off, and move people out to a safe place and call the gas company's emergency number. ; Zhangjiagang Municipal Emergency Management Bureau (张家港市应急管理局) (2024). Gas safety knowledge, 燃气安全使用知识: 「不要情急之下打开排气扇进行通风」「应远离现场之后才能打电话求助」 — do not open the exhaust fan in a hurry, and call for help only after you are clear of the scene. +- Notes: Some people come to the door in the name of a "safety inspection" and then sell a self-closing valve and an alarm. The regulations bar a gas company from making you buy its designated products. You can refuse, and you can also complain. If you smell gas, close the main gas valve first. For bottled gas, close the valve on the cylinder. Then open the door and windows to ventilate, and do not use the exhaust fan. Do not switch lights, the cooker hood or a lighter on or off. Then get everyone out and call the gas company's emergency number from a safe distance. Carbon monoxide poisoning is a separate matter, see section 13. + +### 5. Do not pick, buy or eat wild mushrooms; no local way of telling poison works + +- Cost: No money. The hard part is giving up a seasonal wild food. +- In plain language: In 2025 alone there were 828 mushroom poisonings nationwide, 2165 people affected and 13 deaths. That year 34 species of mushroom not previously recorded as poisonous were identified. A silver needle does not detect poison, a mushroom that insects have eaten is not safe, and only brightly coloured mushrooms are poisonous. None of these claims holds. +- Benefit: China CDC investigated 828 mushroom poisoning incidents in 2025, involving 2165 people and 13 deaths. The case fatality rate was 0.6%, which means 0.6% of the people poisoned died. From 2019 to 2024 there were 276 to 676 incidents a year, and the case fatality rate ranged from 0.87% to 2.86%. In 2025 alone 138 poisonous mushroom species were identified, of which 34 were recorded in China for the first time as causing poisoning +- Evidence grade: A +- Sources: Mushroom Poisoning Outbreaks — China, 2025. China CDC Weekly (2026): 「In 2025, China CDC investigated 828 mushroom poisoning incidents across 27 provincial-level administrative divisions (PLADs), affecting 2,165 individuals and causing 13 deaths - a case fatality rate of 0.6%, the lowest of the past six years. In total, 138 poisonous mushroom species were identified, including 34 newly recorded in poisoning incidents in China.」「From 2019 to 2024, the annual number of incidents ranged from 276 to 676, and the case fatality rate ranged from 0.87% to 2.86%.」 ; the 2024 annual report in the same journal +- Notes: If you eat the wrong mushroom, induce vomiting at once, then take the rest of the mushroom, or a photograph of it, to hospital. Poisoning by toxic mushrooms comes in several types, and identifying which mushroom it was decides how the doctor treats you. New species are recorded as poisonous every year, so the assumption that "local people can tell them apart" does not hold in the first place. A silver needle test, boiling it with garlic, the claim that a mushroom insects have eaten is safe, and the claim that only bright colours are poisonous are all wrong. + +### 6. Do not take an electric bike into the corridor or the lift, and do not charge it at home + +- Cost: No money. The hard part is parking at the shared charging point downstairs every time and walking a few tens of metres further. +- In plain language: The Ministry of Emergency Management's Regulations on Fire Safety Management of High-Rise Civil Buildings forbid taking an electric bike into a corridor or a lift, and forbid charging it at home. Failing to correct this brings a fine of 500 to 1000 yuan on an individual and 2000 to 10000 yuan on a business. The fine is the smaller problem. The larger one is that this fire burns along your only escape route. +- Benefit: The regulation of the Ministry of Emergency Management states the ban in writing: 「禁止在高层民用建筑公共门厅、疏散走道、楼梯间、安全出口停放电动自行车或者为电动自行车充电」, which means electric bikes may be neither parked nor charged in the public entrance hall, the escape corridor, the stairwell or the emergency exit of a high-rise civil building. For failure to correct it, a unit or individual not in business is fined 500 to 1000 yuan and a business is fined 2000 to 10000 yuan +- Evidence grade: A +- Sources: Ministry of Emergency Management (应急管理部) (2021). Regulations on Fire Safety Management of High-Rise Civil Buildings, 高层民用建筑消防安全管理规定（应急管理部令第 5 号，第三十七条、第四十七条）. ; Ministry of Emergency Management (应急管理部) (2023). Reply letter on proposal No. 00969 of the Fifth Session of the National Committee of the Chinese People's Political Consultative Conference, 关于政协第十三届全国委员会第五次会议第00969号提案答复的函: 「2021年，全国共接报电动自行车火灾1.8万起，造成65人死亡、222人受伤，直接财产损失达2.21亿元，其中发生在建筑物内3782起，造成60人死亡、175人受伤」 — in 2021 the country handled 18,000 electric bike fires, causing 65 deaths, 222 injuries and direct property losses of 221 million yuan, of which 3,782 began inside buildings and caused 60 deaths and 175 injuries. ; Ministry of Emergency Management (应急管理部) (2025). Press conference on day 15 of month 1 in 2025: 「火灾起数同比下降42.2%，伤亡人数下降超过90%」 — the number of fires fell 42.2% year on year, and the number of casualties fell by more than 90%. +- Notes: Do not buy a modified battery, do not use a no-name charger, and do not leave it charging all night. Once a lithium battery catches fire, the whole thing is alight within seconds and it gives off severely toxic smoke. A bike parked in a corridor sets that fire on your only escape route. The figures the Ministry of Emergency Management published are 18,000 electric bike fires nationwide in 2021 and 65 deaths. Of those, 3,782 started inside buildings and killed 60 people, and nearly all the dead were indoors. After the 2024 campaign the number of fires fell 42.2% on the year and casualties fell by more than 90%, and the absolute figures for that year were not published. + +### 7. Measure your blood pressure, and take drugs to bring it down to target if it is high + +- Cost: A home blood pressure monitor costs 100 to 200 yuan. One reading takes 1 minute. Most blood pressure drugs cost a few yuan to a few tens of yuan a month. The hard part is carrying on with them when you have no symptoms. +- In plain language: For every 10 mmHg that blood pressure falls, the chance of a major cardiovascular event falls by about a fifth. Stroke falls 27%, heart failure 28%, and the chance of death 13%. Among people aged 35 to 75 in China, 44.7% have high blood pressure and only 7.2% have it properly controlled. +- Benefit: 123 trials with more than 610,000 people were pooled. For every 10 mmHg that systolic blood pressure falls, major cardiovascular events are about a fifth lower. The original value is RR 0.80, credible range 95% CI 0.77 to 0.83. Systolic blood pressure is the higher of the two numbers in a reading. Stroke is about 27% lower (RR 0.73). Heart failure is about 28% lower (RR 0.72). All-cause mortality fell 13% (RR 0.87, 95% CI 0.84 to 0.91). China screened 1,700,000 people aged 35 to 75: 44.7% had high blood pressure. Among them, 44.7% knew they had it, 30.1% were being treated, and only 7.2% were under control +- Evidence grade: A +- Sources: Ettehad D et al (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Lu J et al (2017). Prevalence, awareness, treatment, and control of hypertension in China (China PEACE Million Persons Project). Lancet. +- Notes: Measure your blood pressure once first. Of every 10 people in China with high blood pressure, fewer than 1 is under control, so that single reading is one of the highest-value actions in this book. What counts as target (130 or 140) is still argued over, but "knowing you have high blood pressure and bringing it down" is not disputed. + +### 8. After 35, if you are overweight, get one fasting blood glucose test, and repeat it every three years even if it is normal + +- Cost: A fasting glucose test costs a dozen-odd yuan. A glycated haemoglobin test costs a few tens of yuan. One trip to the hospital to draw blood is all it takes. +- In plain language: Drawing one tube of blood for a few tens of yuan can find pre-diabetes while you have no symptoms at all. The official United States screening advice is that people aged 35 to 70 who are overweight or obese should be tested. If the result is normal, test again every three years. +- Benefit: The US Preventive Services Task Force is the body that sets official screening advice in the United States. Its current recommendation is to screen adults aged 35 to 70 who are overweight or obese for pre-diabetes and type 2 diabetes, even when they have no symptoms at all. Overweight means a body mass index ≥25, and obese means ≥30. That recommendation carries a grade B. It also states: "for adults with normal blood glucose, screening every 3 years is reasonable" +- Evidence grade: A +- Sources: US Preventive Services Task Force (2021). Screening for Prediabetes and Type 2 Diabetes. +- Notes: Finding pre-diabetes does not mean you have to take drugs. A good share of people can reverse it with lifestyle changes alone. Among Chinese adults only a minority have ever known they had diabetes. Many find out only once complications appear + +### 9. Do not speed, and do not drink and drive + +- Cost: No money. Driving a little slower takes a few extra minutes on each trip. The hard part is drinking less at a dinner. +- In plain language: Every 1% higher average speed raises the chance of killing someone in a collision by about 4%. For alcohol there is no safe line where a little is fine. Risk starts to climb when the alcohol concentration in the blood is still low. +- Benefit: The World Health Organization (WHO) puts it this way: every 1% rise in average speed raises the risk of fatal accidents by 4%. The risk from drink-driving starts to rise from a very low blood alcohol concentration +- Evidence grade: B +- Sources: WHO (2025). Road traffic injuries fact sheet. +- Notes: The 1% and 4% for speed and death come from an accident model, not from a randomised trial. On drink-driving WHO gave a direction but no numbers. This section has not checked the original literature on how much drinking raises the risk by how much. + +### 10. Use a child car seat for children under 4, and do not hold them in your arms + +- Cost: One costs 300 to 2000 yuan and lasts for years. Buckling the child in adds 1 minute on each trip. +- In plain language: An infant under one year old in a car seat has about 70% lower risk of fatal injury in a crash. A child aged 1 to 4 has about half the risk. Holding the child in your arms gives no such protection, and in a hard stop you cannot hold on either. +- Benefit: The US National Highway Traffic Safety Administration (NHTSA) estimates that a car seat lowers the risk of fatal injury for an infant under one year old in a car by 71%. For a child aged 1 to 4 the figure is 54%. The World Health Organization (WHO) puts the reduction in infant deaths at 71% +- Evidence grade: A +- Sources: NHTSA (2024). Occupant Protection in Passenger Vehicles: 2022 Data (DOT HS 813 573). ; WHO (2025). Road traffic injuries fact sheet. +- Notes: Fix the seat as the instructions say. A wrong fit is the same as no fit. Under the age of one the seat goes in facing backwards, with the child looking toward the rear of the car. The figures above are the official United States estimates calculated from crash data. + +### 11. If there are children at home, fit window and balcony limiters; window screens do not count as protection + +- Cost: A window limiter or a child safety lock costs a few tens of yuan. Fitting it takes ten minutes. +- In plain language: A window limiter costs a few tens of yuan. New York City fits window guards free of charge in high-risk housing and sends someone to explain them. In one borough, reported child falls dropped by half over three years. New York then passed a law requiring landlords to fit guards in homes where children live. A window screen opens with one push. It does not count as protection. +- Benefit: New York City ran a programme called Children Can't Fly from 1972. It gave out window guards free in high-risk districts and sent someone to explain them. Reported falls in the Bronx fell 50% from 1973 to 1975. The city then amended its health code. Landlords must now fit window guards in any apartment where a child aged 10 or younger lives +- Evidence grade: B +- Sources: Spiegel CN, Lindaman FC (1977). Children can't fly: a program to prevent childhood morbidity and mortality from window falls. American Journal of Public Health: "Significant reduction in falls resulted, particularly in the Bronx, where reported falls declined 50 percent from 1973 to 1975"; the 1976 amendment to the New York City health code "to require that landlords provide window guards in apartments where children ten years old and younger reside". ; (2021). Unintentional Window Falls in Children and Adolescents. Academic Pediatrics: 38,840 emergency department visits recorded by the national electronic injury surveillance system from month 1 of 2007 to month 8 of 2017, "The majority of falls occurred in children under the age of 6 and were related to falls from a second story or below". +- Notes: Do not put a bed, a sofa, or boxes where a child can climb up next to a window or a balcony. A window screen only keeps out mosquitoes. It comes off with one push, so do not treat it as a guard. High-rise living is not the only reason to fit one. In the US data, most falls happened on the second floor or lower. + +### 12. Keep children in sight near water; life jackets for boating and open-water swimming + +- Cost: A life jacket costs 50 to 200 yuan. It also takes the attention you spend watching the child. +- In plain language: A child in a life jacket who falls in the water has about half the chance of drowning. Drowning is still the leading cause of death for Chinese children aged 1 to 14. The rural drowning mortality rate is about twice the urban rate. +- Benefit: A study using United States Coast Guard data matched comparable people in pairs. Among recreational boaters who fell in the water, those wearing a life jacket had about half the risk of drowning death. The adjusted value is RR 0.51, credible range 95% CI 0.35 to 0.74. In China, drowning mortality under age 20 fell from 6.60 per 100,000 people in 2013 to 3.28 per 100,000 in 2021. The rural rate is about 2 times the urban rate. Drowning is still the leading cause of death for children aged 1 to 14. In 2021, drowning caused 31.1% of injury deaths among people aged 0 to 19, and road traffic caused 27.9% +- Evidence grade: A +- Sources: Cummings P et al (2011). Association between wearing a personal floatation device and death by drowning among recreational boaters. Injury Prevention. ; Li Z et al (2023). Unintentional Drowning Mortality Among Individuals Under Age 20 — China, 2013–2021. China CDC Weekly. ; Zhou J et al (2024). Injury Mortality of Children and Adolescents Aged 0–19 Years — China, 2010–2021. China CDC Weekly. +- Notes: The grade A covers only the life jacket. Keeping children in sight is a general consensus. On its own it is only grade C, with no randomised trial, and there will never be one. In China, drowning deaths cluster in rural areas, in summer, and among males aged 15 to 19. + +### 13. Over 60, train balance and leg strength; fix up the bathroom and stairs at home + +- Cost: Train 2 to 3 times a week, 30 minutes of tai chi or balance work each time. A non-slip mat, grab bars, and a night light in the bathroom cost a few tens to a few hundred yuan. The hard part is keeping it up for years. +- In plain language: Balance training such as tai chi two or three times a week cuts falls among older people by about 23%. The number of people who have a fall drops 15%. Non-slip surfaces and grab bars at home cut falls by another 20%. Falls are the leading cause of injury death among Chinese people over 65. More than half of them happen in the person's own home. +- Benefit: Cochrane pooled 108 trials with 23,407 people. Exercise reduced the number of falls among older people by 23%. The rate ratio for falls is RaR 0.77, credible range 95% CI 0.71 to 0.83. The number of people with a fall dropped 15% (RR 0.85). A safety assessment and home modifications reduced the fall rate by about a fifth (RR 0.81, 95% CI 0.68 to 0.97). Tai chi reduced fall risk by about 29% (RR 0.71). In China, falls are the leading cause of injury death among people over 65. Of the older people's falls monitored in 2018, 55.97% happened in the home +- Evidence grade: A +- Sources: Sherrington C et al (2019). Exercise for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Gillespie LD et al (2012). Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Lu Z et al (2021). Characteristics of Falls Among Older People — China, 2018. China CDC Weekly. +- Notes: This applies to people over 60 who live in their own home and not in a nursing home. The studies above count falls, not deaths. For an older person the road from a fall to a hip fracture to death is short. + +### 14. Get tested for hepatitis B; if you have no antibodies, get vaccinated + +- Cost: One test costs a few tens of yuan. The 3 doses of hepatitis B vaccine an adult needs cost 100 to 300 yuan in total. Finish them within half a year. +- In plain language: One test costs a few tens of yuan. In the Qidong trial that followed people for 30 years, people vaccinated at birth were 84% less likely to get primary liver cancer later. 72% fewer of them carried the hepatitis B virus. An adult who tests negative for antibodies gets three more doses, within half a year, for a hundred or two hundred yuan. +- Benefit: The Qidong trial was randomised by group, not by individual, and followed people for 30 years. Newborn vaccination against hepatitis B reduced primary liver cancer by 84%. The credible range is 95% CI 23% to 97%. The share testing positive for hepatitis B surface antigen (HBsAg) fell by 72% (95% CI 68% to 75%). That share is the proportion of people found to be carrying the virus. Among the whole Chinese population, the HBsAg-positive share fell 52% from 1992 to 2014, and 97% among children under 5 +- Evidence grade: A +- Sources: Qu C et al (2014). Efficacy of neonatal HBV vaccination on liver cancer and other liver diseases over 30-year follow-up of the Qidong hepatitis B intervention study. PLoS Medicine. ; Cui F et al (2017). Prevention of Chronic Hepatitis B after 3 Decades of Escalating Vaccination Policy, China. Emerging Infectious Diseases. +- Notes: This applies to adults born before 1992 who were never vaccinated or whose antibodies have worn off. The 84% drop in liver cancer comes from people vaccinated at birth. In the same trial, vaccination first given in adolescence was only 21% effective at preventing carriage of the virus. Adult vaccination is mainly about avoiding infection, so the 84% figure does not apply to it. A vaccine does nothing for someone already infected, who needs regular checks in a liver clinic. To know whether you are already infected, look at the hepatitis B surface antigen (HBsAg) line on the lab report. It is positive when you are infected. + +### 15. Punctured by a nail or a splinter, or a wound with soil in it: get it treated the same day and ask about a tetanus shot + +- Cost: Registration and wound treatment cost a few tens of yuan. If the doctor says a shot is needed, the vaccine or the immunoglobulin costs a few tens to a few hundred yuan. +- In plain language: Once tetanus develops, about 1 in 10 people dies of it. Deep wounds and wounds with soil in them carry the highest risk. A dirty wound with 5 years or more since the last shot needs one booster. Same-day treatment costs a few tens of yuan. Do not wait until the wound swells. +- Benefit: Among tetanus cases in the United States, about 1 in 10 ends in death. Deep wounds and wounds contaminated with soil carry more risk. China's 2024 treatment guideline sorts wounds into two tiers. A shallow wound treated within 6 hours and free of soil, faeces, and animal saliva is low risk. A wound is high risk if it is treated only after 6 hours, if it carries any of those, or if it is a puncture, a degloving injury, a crush injury, a burn, a frostbite, or a wound with a foreign body left inside. Someone with 3 or more doses and a last dose under 5 years ago needs nothing. At 5 to 10 years, a high-risk wound needs one booster. At 10 years or more, both tiers need one booster. Someone with fewer than 3 doses, or who cannot remember, must complete the course, and a high-risk wound also needs passive immunisation (an injection that supplies antibodies directly, such as immunoglobulin or antitoxin) +- Evidence grade: B +- Sources: US CDC (美国疾病控制与预防中心). Tetanus: "Tetanus bacteria can get into someone's body through broken skin, usually through injuries." "Tetanus can lead to death (1 in 10 cases in the United States are fatal)." "People who didn't complete the primary series or who aren't up to date with their 10-year tetanus booster shots are also at increased risk." "Vaccination also helps prevent tetanus in people with wounds, depending on their tetanus vaccination history." ; General Office of the National Health Commission (国家卫生健康委办公厅) (2024). Notice on issuing the Diagnosis and Treatment Standards for Tetanus in Non-newborns (2024 edition) (国卫办医急函〔2024〕381 号; the risk tiers are on page 16 of the appendix, Table 4 on page 20). ; appendix PDF (scan). +- Notes: Let the doctor decide when you reach the hospital. Do not decide it yourself, and do not skip the question because it seems like a hassle. Not every wound needs a shot. The doctor looks at which tier the wound falls into and asks when you had your last dose. Then the doctor decides. "Shots in childhood last a lifetime" is wrong. For a dirty wound, 5 years since the last dose means a booster. Think back to the year of your last tetanus shot before you go. + +### 16. Women should get the HPV vaccine, the earlier the better + +- Cost: The domestic two-valent vaccine is a little over 300 yuan a dose, and the imported nine-valent one about 1300 yuan a dose. The course is 2 to 3 doses. Each trip takes half a day. +- In plain language: In Swedish data from 1,670,000 women, women vaccinated before age 17 were about one eighth as likely to get invasive cervical cancer later. Those vaccinated between 17 and 30 were about half as likely. So the earlier the better. +- Benefit: Sweden followed 1,670,000 women and counted invasive cervical cancer. Women who had the vaccine had a lower rate than women who did not. Vaccination before age 17 cut the rate by about 88% (IRR 0.12, 95% CI 0.00 to 0.34; the bracket holds the value and its credible range). Vaccination at 17 to 30 cut it by about 53% (IRR 0.47, 95% CI 0.27 to 0.75) +- Evidence grade: A +- Sources: Lei J et al (2020). HPV Vaccination and the Risk of Invasive Cervical Cancer. New England Journal of Medicine. +- Notes: The 9 to 14 age group gains the most. Vaccination at 17 to 30 still cuts the rate by more than half. After 30 the gain keeps shrinking. Vaccinated women still need cervical cancer screening, because the vaccine does not replace it, see item 18 (cervical cancer screening for women over 30). + +### 17. Women should start breast cancer screening at 40, with a mammogram every two years + +- Cost: A mammogram costs two or three hundred yuan. Many regions run free screening programmes. One round every two years takes half a day. +- In plain language: Official US screening advice recommends a mammogram every two years for women aged 40 to 74. For women over 75, and for extra ultrasound or MRI on dense breasts, the official conclusion is that the evidence is not enough. When a checkup package tries to charge you for those, you can decline. +- Benefit: The US Preventive Services Task Force sets official screening advice in the United States. Its current recommendation is to "recommend screening mammography every two years for women aged 40 to 74 years". Screening mammography is what people call a mammogram. That recommendation carries a grade B. For women aged 75 and over, and for extra ultrasound or MRI on dense breasts, it judges the "existing evidence to be insufficient" +- Evidence grade: A +- Sources: US Preventive Services Task Force (2024). Breast Cancer: Screening. +- Notes: The starting age differs by country. Some Chinese guidelines and local programmes start at 45. A parent, a child, or a sibling with breast cancer, and anyone carrying a BRCA mutation, is high risk. Those cases need a plan from a doctor, and this item does not apply to them + +### 18. Women over 30 should be screened for cervical cancer, with HPV testing first + +- Cost: An HPV test costs 100 to 300 yuan. If the result is negative, one test every 5 years is enough. The swab takes a few minutes. +- In plain language: In a trial in rural India, women tested for HPV only once in their lives had about half the risk of dying of cervical cancer. They were about half as likely to be found with late-stage cervical cancer. One test costs a hundred or two hundred yuan, and a negative result holds for 5 years. +- Benefit: The rural India trial was randomised by group, not by individual. It covered women aged 30 to 59. A single HPV screening test reduced the risk of dying of cervical cancer by about 48%. The value is HR 0.52, credible range 95% CI 0.33 to 0.83. The risk of being found with late-stage cervical cancer fell by about 53% (HR 0.47, 95% CI 0.32 to 0.69) +- Evidence grade: A +- Sources: Sankaranarayanan R et al (2009). HPV screening for cervical cancer in rural India. New England Journal of Medicine. +- Notes: Women who have had the HPV vaccine still need screening, because the vaccine does not cover all the virus types. What makes this study unusual is that it is a randomised screening trial. It measured deaths directly, and it screened only one round. + +### 19. Start colorectal cancer screening at 45 to 50, with a faecal immunochemical test or a colonoscopy + +- Cost: A faecal test costs a few tens of yuan, every 1 to 2 years. A colonoscopy costs a few hundred yuan to a little over a thousand, and one every 10 years is enough if the result is negative. Set aside a day for bowel preparation before a colonoscopy. +- In plain language: A faecal occult blood test costing a few tens of yuan every year or two lowers the risk of dying of colorectal cancer by about 16%. Among people who complete at least one round, the drop is 25%. Among people invited to have one colonoscopy, colorectal cancer over the following ten years went from 1.20% to 0.98%. +- Benefit: Cochrane pooled several studies. People screened with a faecal occult blood test had about 16% lower risk of dying of colorectal cancer. The value is RR 0.84, credible range 95% CI 0.78 to 0.90. Among people who took part in at least one round, the drop was about 25% (RR 0.75). The NordICC randomised trial reported the same pattern. Among people invited to have a colonoscopy, colorectal cancer over 10 years went from 1.20% to 0.98%. The value is RR 0.82, credible range 95% CI 0.70 to 0.93 +- Evidence grade: A +- Sources: Hewitson P et al (2007). Screening for colorectal cancer using the faecal occult blood test, Hemoccult. Cochrane Database of Systematic Reviews. ; Bretthauer M et al (2022). Effect of Colonoscopy Screening on Risks of Colorectal Cancer and Related Death. New England Journal of Medicine. +- Notes:Disputed. NordICC counted people who were invited to have a colonoscopy. Colorectal cancer deaths were 0.28% against 0.31% (RR 0.90, 95% CI 0.64 to 1.16), and the difference is still inside the error range. That means the mortality benefit of colonoscopy was overstated in the past. The faecal test has the harder evidence on mortality. National guidelines put the starting age between 45 and 50. If someone in the family has had this cancer, start earlier. + +### 20. People with cardiovascular disease and older adults should get a flu vaccine every year + +- Cost: One dose costs 50 to 150 yuan. One shot every autumn. +- In plain language: In a trial of flu vaccination after a heart attack, deaths within a year fell from 4.9% to 2.9%. People with cardiovascular disease and older adults can get one shot each autumn for 50 to 150 yuan. +- Benefit: One trial was done after a heart attack, with 2,571 people randomised. Neither the doctors nor the patients knew which shot they received. In the flu vaccine group, all-cause mortality over 12 months was 2.9%. In the placebo group it was 4.9%. The value is HR 0.59, credible range 95% CI 0.39 to 0.89. A placebo is a control shot with no active ingredient. The HR for cardiovascular death was 0.59. Another study pooled several studies. In the flu vaccine group, major cardiovascular events occurred in 3.6%, against 5.4% in the control group. The value is RR 0.66, credible range 95% CI 0.53 to 0.83. Cardiovascular death RR 0.74 (95% CI 0.42 to 1.30, and that difference is still inside the error range, so it does not count as statistically significant) +- Evidence grade: A +- Sources: Fröbert O et al (2021). Influenza Vaccination After Myocardial Infarction: A Randomized, Double-Blind, Placebo-Controlled, Multicenter Trial. Circulation. ; Behrouzi B et al (2022). Association of Influenza Vaccination With Cardiovascular Risk: A Meta-analysis. JAMA Network Open. ; Demicheli V et al (2018). Vaccines for preventing influenza in the elderly. Cochrane Database of Systematic Reviews. (the Cochrane review named in the notes) +- Notes:Disputed. The grade A covers only people with cardiovascular disease. For healthy older adults, the Cochrane review reports one result: flu cases in a season fell from 6% to 2.4%. It rates the certainty of that evidence "low", and the evidence on mortality "extremely low". The review is Demicheli V et al., 2018. + +### 21. Get the shingles vaccine after age 50 + +- Cost: The recombinant shingles vaccine needs two shots, about 3000 to 4000 yuan in total. You pay the whole bill yourself. +- In plain language: In a phase three trial, this vaccine blocked 97% of shingles. The figure is about the same in every age group. The drawback is that the two shots cost three to four thousand yuan and you pay them yourself. That money buys a 97% drop in the chance of shingles. +- Benefit: A phase three trial ran across 18 countries with 15,400 people. People were randomly put into two groups. One group got the vaccine, the other got a control shot. The average follow-up was 3.2 years. Shingles occurred in 6 people in the vaccine group and 210 in the placebo group. Per 1,000 person-years that is 0.3 against 9.1. A person-year means one person watched for one year. The paper puts it as "overall vaccine efficacy against shingles 97.2% (95% CI 93.7–99.0, P<0.001)". Efficacy by age group runs from 96.6% to 97.9% +- Evidence grade: A +- Sources: Lal H, Cunningham AL, Godeaux O, et al. (2015). Efficacy of an adjuvanted herpes zoster subunit vaccine in older adults. New England Journal of Medicine, 372(22), 2087-2096. +- Notes: It ranks low for value for money because it is expensive. Shingles itself rarely kills. The trouble is post-herpetic neuralgia, which can last from months to years. After the shot, pain at the injection site and fever are both common. The most severe grade, grade 3 symptoms, was 17.0% against 3.2% + +### 22. Get the pneumococcal vaccine at 65 or over + +- Cost: A few hundred yuan. Some regions give it free to older people. +- In plain language: Among 84,500 older people in a trial, the 13-valent pneumococcal vaccine blocked about 46% of the pneumonia caused by the types it covers. It blocked 75% of the most dangerous invasive infections. It showed no effect on pneumonia from all causes, so do not expect it to cover every case of pneumonia. +- Benefit: The Netherlands randomised 84,500 people aged 65 and over into two groups. The vaccine was the 13-valent pneumococcal conjugate vaccine. Its efficacy against vaccine-covered pneumococcal pneumonia was 45.6%. The credible range is 95.2% CI 21.8–62.5. Vaccine-covered means the serotypes this vaccine covers. Efficacy against invasive pneumococcal disease was 75.0% (95% CI 41.4–90.8), the most dangerous type of infection. Against pneumonia from all causes added together, it had no effect +- Evidence grade: A +- Sources: Bonten MJ, Huijts SM, Bolkenbaas M, et al. (2015). Polysaccharide conjugate vaccine against pneumococcal pneumonia in adults. New England Journal of Medicine, 372(12), 1114-1125. +- Notes: It only prevents the serotypes it covers. It does not mean you cannot get pneumonia after the shot. The vaccine Chinese older people usually get is the 23-valent polysaccharide vaccine. That is not the same product as the 13-valent conjugate vaccine used in this trial, so the efficacy figures above do not carry over to it + +### 23. Test for Helicobacter pylori and eradicate it if the test is positive + +- Cost: A carbon 13 or carbon 14 breath test costs 100 to 200 yuan. Eradication needs four drugs at once, 200 to 500 yuan for two weeks. The hard part is taking them on schedule for the full two weeks. +- In plain language: In a trial in Linqu, Shandong that followed people for 22 years, people whose Helicobacter pylori was eradicated had about half the chance of stomach cancer later. Their chance of dying from stomach cancer was about 40% lower. The test costs a hundred or two hundred yuan. The eradication course takes two weeks. +- Benefit: The randomised trial in Linqu, Shandong followed people for 22 years. People treated to eradicate Helicobacter pylori had about half the risk of stomach cancer. The value is OR 0.48, credible range 95% CI 0.32 to 0.71. The risk of dying from stomach cancer was about 38% lower (HR 0.62, 95% CI 0.39 to 0.99) +- Evidence grade: A +- Sources: Li WQ et al (2019). Effects of Helicobacter pylori treatment and vitamin and garlic supplementation on gastric cancer incidence and mortality: follow-up of a randomized intervention trial. BMJ. +- Notes: People can pick up the infection again after eradication, so test and treat the family at the same time. The people in this trial lived in an area with high rates of stomach cancer. People living in a low-incidence area get much less real benefit. + +### 24. Heavy smokers get a low-dose chest CT every year + +- Cost: One scan costs 200 to 400 yuan and takes 10 minutes. Add the follow-up scan after a false alarm, and the worry in between. +- In plain language: Heavy smokers who meet the criteria get a low-dose CT every year. Death from lung cancer is 20% lower than with a chest X-ray alone. Overall death is 6.7% lower. The price is the follow-up scan after a suspicious finding, and the worry in between. +- Benefit: The US NLST randomised trial enrolled 53,454 people. All were 55 to 74 years old, had a smoking history of 30 pack-years or more, and had quit no more than 15 years before. A pack-year is the unit for how much someone smoked: one pack a day for one year. Here is the result. People who had the low-dose CT had a 20.0% relative drop in lung cancer death compared with chest X-ray alone. The credible range is 95% CI 6.8 to 26.7. All-cause mortality fell 6.7% in relative terms, credible range 95% CI 1.2 to 13.6 +- Evidence grade: A +- Sources: National Lung Screening Trial Research Team (2011). Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. New England Journal of Medicine. +- Notes: This applies only to high-risk smokers. In the same trial, 24.2% of the people who had the low-dose CT had a suspicious finding. Of those, 96.4% turned out to be false alarms (a false positive). For people who do not smoke, this test is paying money to feel anxious. + +### 25. Call 12356 when you are depressed or have suicidal thoughts, and keep no sleeping pills or pesticides at home + +- Cost: It costs nothing. One phone call. Also lock the medicines away and keep pesticides out of the house. +- In plain language: Evidence keeps building for taking lethal things out of reach. Restricting access to painkillers cut suicides involving painkillers by 43%. Barriers at places where people often jump from buildings cut them by 86%. That is why you keep no sleeping pills and no pesticides at home. When you cannot get past something, call 12356 first. It works nationwide and is open at least 18 hours a day. +- Benefit: One review systematically went through research from the past 10 years. Evidence keeps strengthening for restricting access to lethal means. After painkillers were brought under control, suicides involving painkillers fell 43%. Adding barriers to places where people often jump cut them by 86% (79% to 91%, and the bracket holds the credible range for that figure). Cognitive programmes in schools reduced suicide attempts by about 55% (OR 0.45, 95% CI 0.24 to 0.85). Drugs for depression and psychotherapy are an important part of preventing suicide. From day 1 of month 5 in 2025, 12356 reaches the national psychological assistance hotline everywhere in China, open for at least 18 hours a day +- Evidence grade: B +- Sources: Zalsman G et al (2016). Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry. ; National Health Commission (国家卫生健康委) (2024). Notice on the Use of "12356" as the Nationwide Unified Psychological Assistance Hotline Number, 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. +- Notes: The reason for the grade B: most of the evidence on restricting means comes from ecological studies, which compare the numbers before and after a policy. The hotline itself has no evidence of lowering mortality. The urge to kill yourself usually lasts minutes to hours. Moving lethal things one step further away is itself an intervention. For the time scale of that urge and for the long-term outcome after an attempt, see item 32 in this section (tell one person near you the moment the thought appears). For what is left behind after someone is rescued, see item 33 in this section. + +### 26. Keep a fire extinguisher, a fire blanket, escape respirators and a first aid kit, and check them once a year + +- Cost: A fire extinguisher, a fire blanket, one respirator per person and a first aid kit come to a few hundred yuan as a set. Buy them once, then spend ten minutes checking them every year. +- In plain language: A fire extinguisher, a fire blanket, an escape respirator and a first aid kit are all on the emergency ministry's list of 11 items for the home. The set costs a few hundred yuan and you buy it once. In the studies with home safety visits, injuries fell by about a quarter. +- Benefit: The Ministry of Emergency Management's National Basic Household Emergency Supplies Storage Recommended List has 11 items. It lists "fire extinguishers and fire blankets" for "putting out a fire in its early stage" (「灭火器和灭火毯」, 「初起火灾的扑救」), and it adds that a fire blanket "can be worn over the body to escape" (「灭火毯可披覆在身上逃生」). "Respirator" there means a "fire-filtering self-rescue breathing apparatus, for escaping a fire" (「消防过滤式自救呼吸器，用于火灾逃生使用」). "Topical medicines" means "haemostatic powder, adhesive wound dressings, gauze bandages and the like, for treating wounds" (「止血粉、止血贴、纱布绷带等，用于处理伤口」). Cochrane pooled 98 studies and 2,605,044 people. Home safety interventions brought the injury rate down to IRR 0.89. The credible range is 95% CI 0.78 to 1.01. Interventions delivered in the home cut injuries by about 25% (IRR 0.75, 95% CI 0.62 to 0.91) +- Evidence grade: B +- Sources: Ministry of Emergency Management (应急管理部) (2020). National Basic Household Emergency Supplies Storage Recommended List, 全国基础版家庭应急物资储备建议清单. ; Kendrick D et al (2012). Home safety education and provision of safety equipment for injury prevention. Cochrane Database of Systematic Reviews. ; Wusheng County Market Supervision Administration (武胜县市场监督管理局) (2025). Consumer advice on portable fire extinguishers, fire doors and windows, and point-type heat (smoke) fire detectors, 手提式灭火器、防火门（窗）、点型感温（烟）火灾探测器消费提示. +- Notes:Disputed. In the same Cochrane summary, the kind of intervention that hands out no equipment cut injuries by about a fifth. The value is IRR 0.78, credible range 95% CI 0.66 to 0.92. The group that did get equipment did no better. Home visits and home modifications are what cut injuries, so the equipment is graded "medium" and not "large". When you buy an extinguisher, look for the CCC mark, and the needle of a stored-pressure model must rest in the green zone. Smoke alarms and carbon monoxide alarms: see item 3 in this section. How to use a tourniquet: see item 12 in section 13. How to choose each item, where to keep it, and how often to replace it: see [docs/家庭应急装备清单.md](../docs/家庭应急装备清单.md) + +### 27. Visible blood in the urine needs one check, even with no pain and even if it clears by the next day + +- Cost: A routine urine test costs a few tens of yuan and an ultrasound of the urinary tract a hundred or two hundred yuan. If the doctor thinks it is needed, add a cystoscopy, around a thousand yuan. +- In plain language: Among people over 60, one episode of visible blood in the urine means about a 2.8% chance of bladder cancer being found. Between 40 and 59 it is about 1.2%. That figure is not high, but it is high enough that you must go and get it checked once. Go even if the bleeding is gone the next day. +- Benefit: A UK study used the electronic records of community general practices, the first stop in primary care. It compared 4915 bladder cancer cases with 21718 controls. Among people with visible blood in the urine, this is the share who did turn out to have bladder cancer. Over 60 it is 2.8%, credible range 95% CI 2.5 to 3.1. Between 40 and 59 it is 1.2%, credible range 0.6 to 2.3. The technical name for that share is the positive predictive value. Blood in the urine that is visible only under a microscope carries the same meaning. The odds ratio is 20, credible range 95% CI 12 to 33. Over 60 that share is 1.6%, credible range 1.2 to 2.1 +- Evidence grade: B +- Sources: Price SJ, Shephard EA, Stapley SA, Barraclough K, Hamilton WT (2014). Non-visible versus visible haematuria and bladder cancer risk: a study of electronic records in primary care. British Journal of General Practice, 64(626), e584-e589. +- Notes: "No pain" is the key point. Blood in the urine from stones or infection comes with colicky pain in the waist or lower belly most of the time, or with frequent urination, urgency, or pain on passing urine. Tumours often cause no sensation at all. The bleeding stops on its own for a few days and comes back, and many people let it pass. Taking an anticoagulant is not a reason to skip the check. After hard exercise, or around a woman's period, you can repeat the test once before deciding. The benefit is graded "small" because only the detection rate is known. There is no effect size for how much an earlier check lowers deaths. + +### 28. If erectile function fails, check the cardiovascular system first, and do not treat it as only "that kind of problem" + +- Cost: Registration costs a few tens of yuan. Measuring blood pressure and drawing blood for glucose and lipids costs a hundred or two hundred yuan. One trip to the hospital, then a few days of waiting for the results. +- In plain language: Men with erectile dysfunction later have about 40% more cardiovascular disease than other men, and about 60% more heart attacks. The younger the man, the bigger the gap. This problem often shows up before chest pain does. So get your blood pressure, glucose and lipids checked first. Do not buy drugs to push it down on your own. +- Benefit: A meta-analysis pooled 14 cohort studies with 92,757 men, followed for an average of 6.1 years. Compared with men without erectile dysfunction, here is the later risk among men who had it. Total cardiovascular events were about 44% higher (RR 1.44, 95% CI 1.27 to 1.63, and that is the credible range). Heart attacks were about 62% higher (RR 1.62, 1.34 to 1.96). Cerebrovascular events were about 39% higher (RR 1.39, 1.23 to 1.57). All-cause death was about 25% higher (RR 1.25, 1.12 to 1.39). Cardiovascular death showed no difference (RR 1.19, 0.97 to 1.46). The younger the age, the higher the multiple. Among people whose cardiovascular risk was moderate to begin with, the multiple was higher than in both the high-risk and the low-risk group. Studies that decided with a questionnaire put the multiple at 1.61 (1.38 to 1.86). Studies that asked a single question put it at 1.27 (1.18 to 1.37) +- Evidence grade: A +- Sources: Vlachopoulos CV, Terentes-Printzios DG, Ioakeimidis NK, Aznaouridis KA, Stefanadis CI (2013). Prediction of cardiovascular events and all-cause mortality with erectile dysfunction: a systematic review and meta-analysis of cohort studies. Circulation: Cardiovascular Quality and Outcomes, 6(1), 99-109. ；DailyMed (the drug label database of the US National Library of Medicine). VIAGRA (sildenafil citrate) tablet, CONTRAINDICATIONS. +- Notes: All 14 of these studies only followed a group of people and watched what happened, so they were not randomised trials. The extra 40% is therefore an association, not a cause. An occasional problem does not count. Only problems lasting three months or more should be treated as a problem. What needs checking is the same as in item 7 and item 8 in this section: blood pressure, fasting glucose, plus lipids, smoking and waist size. Do not buy "aphrodisiac" health products online. The legitimate drugs of this kind are prescription drugs. The label lists nitrates as a contraindication, so anyone using a drug such as nitroglycerin cannot take them. Together they can send blood pressure down suddenly. How to improve it: see item 29 in this section (losing weight and quitting smoking improve erectile function). The benefit is graded "medium", not "large": 44% is the difference in risk between two groups of people, not a measure of how many deaths one check would prevent. Nor is it graded "small": a signal like the one in item 27 in this section (visible blood in the urine) only has a detection rate, while the high blood pressure, diabetes and abnormal lipids found here all have treatments already in hand. + +### 29. Losing weight, quitting smoking and controlling blood pressure and glucose improve erectile function + +- Cost: It costs nothing and may save money. The hard part is keeping it up: controlling what you eat, getting moving, and quitting smoking. +- In plain language: Controlling the risk factors for cardiovascular disease improves erectile function itself. On a questionnaire score, the average gain is 2 to 3 points. The questionnaire has 25 points in total, and 22 or more counts as normal. +- Benefit: A meta-analysis pooled 6 randomised controlled trials with 740 men. The interventions were lifestyle change, or drugs to control cardiovascular risk factors, with at least 6 weeks of follow-up. Here is the result: the score on the erectile function questionnaire (IIEF-5) rose by 2.66 points on average (95% CI 1.86 to 3.47, and that is the credible range). Taking out the 2 trials that used statins (143 people) leaves 4 trials based on lifestyle change alone (597 people), and they still gained 2.40 points (1.19 to 3.61). The IIEF-5 is a questionnaire of 5 questions, worth 25 points in total. The paper splits the scores into five bands: 22 to 25 points counts as no difficulty, and 5 to 7 points counts as severe +- Evidence grade: A +- Sources: Gupta BP, Murad MH, Clifton MM, Prokop L, Nehra A, Kopecky SL (2011). The effect of lifestyle modification and cardiovascular risk factor reduction on erectile dysfunction: a systematic review and meta-analysis. Archives of Internal Medicine, 171(20), 1797-1803. ；Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM (1999). Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. International Journal of Impotence Research, 11(6), 319-326. +- Notes: The interventions in these 6 trials differ. Some combined weight loss with exercise, and others used drugs to control blood pressure and lipids. There were only 740 people in total, so the direction is clear but the size of the gain differs from person to person. The benefit is graded "small" because the only endpoint is a questionnaire score. There is no figure for how many illnesses people avoided later. The same three actions do have hard endpoints elsewhere: for quitting smoking see item 1 in section 2 (quitting smoking, the earlier the better), for weight loss see item 33 in section 2 (keep BMI between 20–25), and for blood pressure see item 7 in this section (measure it, and take drugs if it is high). Taking the drug is not forbidden, as long as a doctor prescribed it. "Health products" sold online often sneak these substances in at an unknown dose. How to tell them apart: see item 4 in section 28 (do not buy diet pills that promise fast weight loss). + +### 30. Use a condom for the whole of sex, and never share needles + +- Cost: One costs a yuan or two. The hard part is not the money. It is using one every time, and putting it on before you start. +- In plain language: Using a condom for the whole of sex each time cuts the chance of passing HIV between men and women by more than 70%. "Most of the time" works about as badly as not using one at all. It also blocks syphilis, gonorrhoea and the like. +- Benefit: One study pooled 25 studies with 10676 heterosexual couples. In each couple one partner had HIV and the other did not, which is called serodiscordant. Compared with people who never use a condom, people who used one every time had about 71% lower risk of HIV transmission. The value is RR 0.29, credible range 95% CI 0.20 to 0.43. Compared with people who use them only now and then, the risk was about 77% lower (RR 0.23, 95% CI 0.13 to 0.40). The authors conclude that condoms are not perfect, but consistent use "can cut HIV transmission by more than 70%". The same protection covers syphilis, gonorrhoea, chlamydia and other sexually transmitted infections +- Evidence grade: A +- Sources: Giannou FK, Tsiara CG, Nikolopoulos GK, et al. (2016). Condom effectiveness in reducing heterosexual HIV transmission: a systematic review and meta-analysis of studies on HIV serodiscordant couples. Expert Review of Pharmacoeconomics & Outcomes Research, 16(4), 489-499. ；Fonner VA et al (2016). Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS. (preexposure prophylaxis, PrEP) +- Notes: The gap between irregular use and no use at all is small. A condom works only if you use one every time, for the whole of sex. Sharing needles is another route of transmission, independent of sex and highly efficient. Injecting drugs, tattoos and body piercing all count. People at high risk can also use preexposure prophylaxis (PrEP), which means taking medicine beforehand to block infection. Pooling 18 studies, in every trial where more than 70% took the pills on time, people taking PrEP had about 70% lower risk of HIV infection than people taking a placebo. The value is a risk ratio of 0.30, credible range 95% CI 0.21 to 0.45. In the trials where fewer people took the pills on time, no protective effect showed. PrEP is a prescription drug. Ask at an infectious disease clinic or the CDC, and do not order it online yourself. What to do after exposure: see section 13 + +### 31. If you have done anything high-risk, get an HIV test once: the CDC does it free and the result stays confidential + +- Cost: It costs nothing. One trip to the CDC or a designated medical institution, one tube of blood. The hard part is asking for the test. +- In plain language: An HIV test at the CDC costs nothing. Under the Measures for the Administration of HIV Testing Nationwide, staff must not disclose your name, your address or your result. A positive result is not a death sentence. People who keep taking the medicine and get the virus down to an undetectable level do not pass it on to their partners. That study followed more than 700 couples and more than 70,000 condomless sex acts, and transmission between partners was zero. +- Benefit: Article twenty-three of the Regulations on the Prevention and Control of AIDS says that the health departments of local people's governments at county level and above will designate medical and health institutions. Those institutions "shall provide free counselling and preliminary screening tests to people who voluntarily seek HIV counselling and testing". Testing early does real work now. Current antiretroviral treatment can push the virus down to undetectable. The World Health Organization's line is that people on antiretroviral treatment with an undetectable viral load will not transmit HIV to their sexual partners. The PARTNER study followed 782 same-sex couples, and in every couple one partner was infected and one was not, which is called serodiscordant. The total is 1593 couple-years, which means the follow-up years of each couple added together. There were 76088 condomless sex acts during that time, and 15 new infections. After a phylogenetic analysis comparing the virus genes, all 15 of them did not come from the partner, and transmission between partners was zero. The Measures for the Administration of HIV Testing Nationwide also states that staff in HIV testing laboratories must not disclose the names, addresses or test results of people with AIDS or people who carry the virus +- Evidence grade: A +- Sources: Ministry of Health (卫生部) (2006). Measures for the Administration of HIV Testing Nationwide, 全国艾滋病检测工作管理办法. ; State Council (国务院) (2006). Regulations on the Prevention and Control of AIDS, 艾滋病防治条例（第三、二十三、三十八条）. (reprinted by the Beijing Municipal Health Commission) ; Rodger AJ, Cambiano V, Bruun T, et al. (2019). Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. The Lancet, 393(10189), 2428-2438. ; Guangdong CDC (广东省疾病预防控制中心). HIV testing window period, a plain-language explainer, 艾滋病检测窗口期科普. (the figures for the window period) +- Notes: The window period decides when a test counts. The figures from the Guangdong CDC are that a nucleic acid test can detect it after about 1 week, a fourth-generation combined antigen and antibody test after about 2 weeks, and a third-generation antibody test after about 3 weeks. Wait at least 2 weeks before testing, and repeat it every 2 to 4 weeks. Three months after the last high-risk behaviour, more than 99.99% of people can rule out infection. A home test kit showing positive is not a diagnosis, and it has to be confirmed at the CDC or a proper institution. Article three of the Regulations says that no unit or individual may discriminate against infected people or their families, and that marriage, employment, medical care and schooling are protected by law. Article thirty-eight says at the same time that infected people are obliged to tell their sexual partners the fact of the infection or the illness in good time. Screening for the three infections in pregnancy and blocking transmission from mother to baby: see section 27 + +### 32. Tell one person near you the moment a suicidal thought appears, and hand those few minutes to someone else + +- Cost: It costs nothing. The hard part is speaking up at the time, and admitting that you cannot stay on your own for the next few hours. +- In plain language: Among people who have attempted suicide, close to half say no more than 10 minutes passed from the thought appearing to the act. Those few minutes are what you have to get through. Three in four had some contact with someone near them during that time and did not speak up. Among people admitted to hospital after self-harm, about 2% died by suicide after 1 year, and about 7% after 9 years or more. The large majority did not later die by suicide. +- Benefit: 82 people who had attempted suicide were interviewed within 3 days of admission. 47.6% of them (39 people) said that no more than 10 minutes passed from the first suicidal thought of that episode to the act itself. The people whose process ran longer scored clearly higher on the suicidal intent scale. The score on the impulsivity scale had nothing to do with how long the process took. 76.8% (63 people) said they had some contact with a partner, a family member or a friend during that time. The authors conclude that the window for intervention is usually short and that there is little room for professionals to step in at that stage, so the warning signs and the ways to intervene have to be spread to ordinary people. A separate systematic review pooled 90 follow-up studies. Among people who came to hospital after self-harm, the middle value for self-harm again without dying within 1 year was 16%. Death by suicide within 1 year was 2%. Among those followed for 9 years or more, about 7% died by suicide +- Evidence grade: B +- Sources: Deisenhammer EA, Ing CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry, 70(1), 19-24. ; Owens D, Horrocks J, House A (2002). Fatal and non-fatal repetition of self-harm: systematic review. British Journal of Psychiatry, 181, 193-199. +- Notes: The reason for the grade B: the Deisenhammer study had only 82 people, came from a single hospital, and relied on the survivors' own recall afterwards. The Owens review gives the middle value from 90 studies that each did things differently, not one number pooled from them. The same review also states plainly that the suicide risk of people who have harmed themselves is still several hundred times that of the general population. So "the large majority did not later die by suicide" describes this group as a whole. It is not a guarantee about any one person, and the doctor still has to be seen. Moving lethal means out of reach, and 12356: see item 25 in this section. What to do first when you feel low: see item 19 in section 3. What is left behind after someone is rescued: see item 33 in this section. What you can do when someone near you shows this kind of thought: see item 15 in section 8 + +### 33. Do not treat surviving as a safety net: after drinking pesticide or inhaling coal gas, emergency care can save the life but not the lungs or the brain + +- Cost: It costs nothing. You only have to know this in advance. +- In plain language: Paraquat has no antidote, and more than 40% of the people who drink it cannot be saved. It inflames the lungs and hardens them slowly, and the kidneys and the liver are damaged as well. Among the people rescued after coal gas poisoning, close to half have a worse brain than before at six weeks, and a quarter of them still do even with hyperbaric oxygen. +- Benefit: A systematic review of paraquat poisoning: the fatality rate is extremely high, because the poison itself is strong and there is no effective treatment. It is carried into lung tissue against the concentration gradient, which is called active uptake against the concentration gradient, and the result is pneumonia and pulmonary fibrosis, together with kidney damage and liver damage. The review states that gastric lavage should not be done. Haemodialysis and haemoperfusion are unlikely to change the clinical course. The evidence for steroids and for immunosuppressants is weak. The review also says: "the fatality rate is high in every centre, even though treatment varies a great deal". Bangladesh's 10 tertiary hospitals admitted 1420 cases from 2013 to 2024, of which 257 could be analysed: a fatality rate of 43.2%. The common presentations were leg swelling from acute kidney failure at 71.2%, vomiting at 65.8% and abdominal pain at 58.0%. The trial in acute carbon monoxide poisoning was double-blind and randomised, with 76 people in each group. Neither the doctors nor the patients knew which group they were in. At six weeks the people left with cognitive sequelae were 19/76 (25.0%) in the hyperbaric oxygen group. The normobaric oxygen group was 35/76 (46.1%), P=0.007. After adjusting for the other factors, OR 0.45 (95% CI 0.22 to 0.92, and the bracket holds the credible range for that number). Another study followed 73 cases: 30% had cognitive sequelae and 12% had white matter hyperintensities in the brain. From day 1 of month 7 in 2014 China withdrew the registration and the production licence for paraquat in liquid form and stopped production. From day 1 of month 7 in 2016 the liquid was no longer sold or used in the country. From day 26 of month 9 in 2020 soluble gel formulations were banned from sale and from use in the country. The document states that cases of suicide by taking paraquat occur from time to time is the direct reason for the tighter regulation +- Evidence grade: B +- Sources: Gawarammana IB, Buckley NA (2011). Medical management of paraquat ingestion. British Journal of Clinical Pharmacology, 72(5), 745-757. ; Chowdhury FR et al (2026). Clinical and Laboratory Profile of Paraquat Poisoning: A Toxicological Crisis in Bangladesh. American Journal of Tropical Medicine and Hygiene, 115, 176-182. ; Weaver LK, Hopkins RO, Chan KJ, et al. (2002). Hyperbaric oxygen for acute carbon monoxide poisoning. New England Journal of Medicine, 347(14), 1057-1067. ; Parkinson RB, Hopkins RO, Cleavinger HB, et al. (2002). White matter hyperintensities and neuropsychological outcome following carbon monoxide poisoning. Neurology, 58(10), 1525-1532. ; General Office of the Ministry of Agriculture and Rural Affairs (农业农村部办公厅) (2018). Notice on effectively strengthening the supervision and management of paraquat, 关于切实加强百草枯监督管理的通知（农办农〔2018〕17 号）. +- Notes: The reason for the grade B: the people in these studies are poisoning cases, not a single group of suicide attempters, and the carbon monoxide trial also includes people who were poisoned by accident. So the sequelae rates above cannot be read directly as the odds of what happens after a suicide attempt. What can be said is that no method leaves no sequelae. Handling carbon monoxide at the scene: see item 19 in section 13. Swallowing pesticide or medicine, do not induce vomiting and take the bottle to hospital: see item 20 in section 13. Keeping no pesticides and no sleeping pills at home: see item 25 in this section. This book does not compare the fatality rates of different methods and does not set down any specific method + +### 34. Do not bet a fall from height on "a few days in bed will do": most people in a trauma ICU do survive, and the bill is counted in years + +- Cost: It costs nothing. You only have to see the bill clearly first. +- In plain language: Among people who fall from a height and are admitted to a trauma ICU, only 5.5% die, but the injuries are severe and the treatment is long. Close to half are hurt in more than three places at once, a little over a third go on a ventilator, and the longest stay in the ICU is 132 days. Half of the falls were from no more than 3 metres. Among the badly injured, fewer than six in ten later go back to work. +- Benefit: A trauma ICU (intensive care unit) in southern Xinjiang has published its figures. From month 4 of 2020 to month 7 of 2024 it admitted 289 falls from a height. The average age of this group was 40.3 years. The median fall height was 3 metres, 26.6% fell from more than 5 metres and 11.4% from more than 10 metres. The median injury severity score (ISS) was 24, the higher that score the heavier the injuries, and 63.3% of the people scored above 20. 47.1% were hurt in more than 3 places and 18.3% in more than 4. 36% needed a ventilator (mechanical ventilation), and the median time on it was 103.5 hours. The median stay in the ICU was 8 days, the shortest 1 day and the longest 132 days. In-hospital or 30-day mortality was 5.5% (16/289). A trauma centre in Germany took in 84 patients with polytrauma and an ISS of 25 or more, polytrauma meaning injuries to several parts of the body at the same time. 58% returned to work during the follow-up period. Age, length of ICU treatment and self-rated health were the predictors. The average cost of a hospital stay nationwide in 2024 (what one stay costs on average) was 9870.0 yuan, 4.3% lower than the year before at that year's prices. In the same year national total health expenditure was 90895.5 hundred million yuan, of which personal health spending was 25007.5 hundred million yuan, 27.5% of the total +- Evidence grade: B +- Sources: Chen Y, Li W, Wang X, Zhong Q, Abudurexiti A, Qiu Q, Li J, Luo J (2025). Clinical features and prognostic predictors for patients admitted to trauma intensive care unit due to fall from height in South Xinjiang. International Journal of Emergency Medicine, 18, 142. ; Simmel S, Wurm S, Drisch S, Woltmann A, Coenen M (2020). Prädiktion der Rückkehr zur Arbeit nach Polytrauma bei Patienten mit einem ISS von mindestens 25. Die Rehabilitation, 59(2), 95-103. ; National Health Commission (国家卫生健康委) (2025). Reading of the 2024 Statistical Bulletin on the Development of Health and Health Services in China, 2024 年我国卫生健康事业发展统计公报解读. +- Notes: The reason for the grade B: the falls group was counted by one hospital going back through old records, and accidents were not separated from self-inflicted falls. The return-to-work group had only 84 people and was in Germany. Neither group can be read as what happens to one person after a jump from a building. The average cost per hospital stay is a mean taken across all hospital stays in the country, not the cost of a fall injury. It is quoted only to give a lower bound that anyone can check. The original paper gives only the distribution of fall heights, so this item does not state how height maps to outcome. Pressure sores, the one to watch during a long spell in bed: see item 8 in section 17. Deep vein thrombosis and pulmonary embolism: see item 11 in section 13 (one leg swelling suddenly). What to do when the thought appears: see item 32 in this section. What poisoning leaves behind: see item 33 in this section (do not treat surviving as a safety net). Below is the author's own experience: a fall from the fourth floor, expecting beforehand that a fortnight or so would do. In fact it was a year in hospital, four operations, and a metal plate put in. For half a year he could not turn over or sit up on his own, and the family borrowed money for it + +### 35. Do not trade "one kidney less is no big deal" for money: the one that is left has to do the work of two, and 86% of people who sold a kidney said afterwards that their health got worse + +- Cost: It costs nothing. You only have to see the bill clearly first. +- In plain language: Among regular donors who went through layer on layer of screening before donating and had somebody check them afterwards, 31 people per 10,000 had reached end-stage kidney disease 15 years later. Among people just as healthy who did not donate, it was 4 per 10,000. People who sold a kidney had no screening and no follow-up. Six years later 86% said their health was worse, and three in four were still in debt. +- Benefit: The United States compared 96,217 living kidney donors with 20,024 healthy people of similar eligibility who did not donate. The comparison group came from the US National Health and Nutrition Examination Survey (NHANES III). Then they estimated the risk of reaching end-stage kidney disease 15 years after donation. For donors it is 30.8 per 10,000, credible range 95% CI 24.3 to 38.5. For healthy non-donors it is 3.9 per 10,000, credible range 95% CI 0.8 to 8.9, P<0.001. The technical name for end-stage kidney disease is ESRD, and at that point a person needs dialysis or a transplant to stay alive. Estimated lifetime risk of reaching ESRD: donors 90 per 10,000, the general population without screening 326 per 10,000, healthy non-donors 14 per 10,000. Norway ran a comparison too. On one side were 1901 kidney donors, followed for a median of 15.1 years. On the other were 32,621 controls who met the donation criteria, followed for a median of 24.9 years. All-cause mortality in kidney donors was about three in ten higher (HR 1.30, 95% CI 1.11 to 1.52). Cardiovascular mortality was about four in ten higher (HR 1.40, 1.03 to 1.91). Reaching ESRD was about ten times higher (HR 11.38, 4.37 to 29.6). Total ESRD incidence in that donor group was 302 per million people. The United States also has 3698 kidney donors, 11 of whom developed ESRD, which works out at 180 per million person-years of observation, against 268 in the general population over the same period. 255 of these people came back for a check a mean of 12.2±9.2 years after donating. 85.5% of them had a GFR (glomerular filtration rate) of 60 ml/min/1.73 m² or more. 60 is the lower limit of normal in clinical practice. 32.1% had high blood pressure. 12.7% had proteinuria (protein leaking into the urine). A survey in Chennai, India looked at 305 people who had sold a kidney and interviewed them a mean of 6 years after the sale: 96% had sold to pay off debts and got 1070 dollars on average. After the removal the average household income fell by a third (P<0.001), more people fell below the poverty line, and three in four were still in debt at the time of the survey. 86% reported that their health was worse than before the kidney was removed, and 79% would not advise anyone else to sell a kidney +- Evidence grade: A +- Sources: Muzaale AD, Massie AB, Wang MC, Montgomery RA, McBride MA, Wainright JL, Segev DL (2014). Risk of end-stage renal disease following live kidney donation. JAMA, 311(6), 579-586. ; Mjøen G, Hallan S, Hartmann A, et al. (2014). Long-term risks for kidney donors. Kidney International, 86(1), 162-167. ; Ibrahim HN, Foley R, Tan L, et al. (2009). Long-term consequences of kidney donation. New England Journal of Medicine, 360(5), 459-469. ; Goyal M, Mehta RL, Schneiderman LJ, Sehgal AR (2002). Economic and health consequences of selling a kidney in India. JAMA, 288(13), 1589-1593. +- Notes:Disputed. The Ibrahim study used ordinary people as its comparison group, and its conclusion is that the survival of kidney donors, and their risk of reaching end-stage kidney disease (ESRD), are close to that of ordinary people. Once Muzaale and Mjøen changed the comparison group to people just as healthy and just as eligible to donate, the risk came out clearly higher. The disagreement is about who to compare against, not about the donors. All three studies agree that the absolute risk is small (30.8 per 10,000 at 15 years, that is 30.8 out of 10,000 people). So the point is not that donating a kidney is dangerous. Two other things matter. The kidney that is left really does pay a price, and the rates of high blood pressure and of proteinuria are there to be seen. Every one of these numbers comes from people who were screened before the operation, had somebody checking them afterwards, and could go back to hospital if something went wrong. Selling a kidney on the black market has none of those three things, and the Indian follow-up gives a different outcome. The original studies report only self-reported worse health, proteinuria, high blood pressure and the incidence of end-stage kidney disease. They give no figure for how many people were bedridden, and this item does not supply one. Family limits on lawful donation, fines and criminal liability: see item 22 in section 9 (do not sell your own organs). For people who already need dialysis, direct cross-province settlement and long-term medication: see items 1 and 2 in section 16 (take the medicine as prescribed, cross-province filing for outpatient special-disease treatment) + +### 36. Do not pick up, take apart or sell metal parts of unknown origin, and keep away from anything carrying the trefoil warning sign + +- Cost: It costs nothing. Walk around any metal cylinder, lead container or part taken off a machine that you do not recognise. Do not pick it up, do not take it apart, do not put it in your pocket. +- In plain language: Radiation has no smell, no colour and no temperature, and most radioactive sources are a piece of ordinary metal, so "it looks all right" is not a test. The chance of an ordinary person running into one is very low, fewer than 1 accident a year per 10,000 sources. Once you do run into one there is no way back. In 1987 in Brazil someone cut up a radiotherapy machine from a disused clinic as scrap, and 4 people died in the end. +- Benefit: As of 2025 the country had about 17.8 hundred thousand radioactive sources in use and 35.5 hundred thousand radiation devices. The yearly rate of radiation accidents involving radioactive sources stays below 1 per 10,000 sources. The grading runs like this. Losing, having stolen or losing control of a Class I or Class II source counts as a major radiation accident. Class III is a larger radiation accident. Losing or having stolen a Class IV or Class V source, or exposing people to more than the annual dose limit, is a general radiation accident. Official case: in month 6 of 2020 the ironmaking plant of the Ningxia Iron and Steel Group was carrying out gamma ray testing on a blast furnace pipeline renovation project. One Class II iridium-192 source exposed three workers. Chen had acute radiation skin damage of degree III on the left hand. Liu had degree IV on the left thumb and index finger, and degree III on the middle, ring and little finger. Fan had degree I-II on the left hand. All three received doses well above the dose limits set by the national standard, and the two responsible units had their radiation safety licences revoked and were fined 7 ten thousand yuan. International case: in 1987 in Goiânia, Brazil, a teletherapy machine from a disused clinic was dismantled, a 50.8 TBq caesium-137 source was taken out of it, and the sealed capsule holding the source was broken. About 112,000 people were monitored. 249 of them were contaminated inside the body or outside it. Of those 249, 129 had moderate to severe contamination and 50 needed close medical observation. 4 people died in the end (national and international) +- Evidence grade: A +- Sources: Ministry of Ecology and Environment (生态环境部) (2025). 2024 中国生态环境状况公报. ; National Nuclear Safety Administration (国家核安全局) (2026). 2025 年我国核与辐射安全监管工作成效. ; Ministry of Ecology and Environment (生态环境部) (2021). 关于一起 γ 射线探伤作业违法违规造成辐射事故案件处理情况的通报. ; State Council (国务院) (2019). Regulations on the Safety and Protection of Radioisotopes and Radiation Devices, 放射性同位素与射线装置安全和防护条例（2019 年修订，第三十四、四十二、六十八条）. ; Ministry of Environmental Protection (环境保护部) (2009). 关于建立放射性同位素与射线装置辐射事故分级处理和报告制度的通知（辐射事故分级标准）. ; International Atomic Energy Agency (1988). The Radiological Accident in Goiânia. IAEA, Vienna. +- Notes: If you really do run into one: move away, note where it is, do not move it, call 110. The regulations require a report to the environmental protection department, the police and the health department. Distance and time are the only protection that works. A pocket or a plastic bag makes no difference at all. X-ray machines and CT scanners make their rays by turning on the power, and nothing radioactive is left once the power is off. The real danger is the sealed source inside the machine. Radiotherapy machines in hospitals, gamma testing machines on pipes and boilers, well logging tools and level gauges all carry a source. Opening the casing is what exposes the source. And they look exactly like ordinary metal parts. So do not count on telling a source from anything else, and never take apart, sell or carry home a machine part of unknown origin. People who work in scrap collection, dismantling, hardware and building sites carry far more risk than other people, and on a site you go around any cordon marked with the trefoil. A leak at a nuclear power plant is a different kind of accident, and the chance of an ordinary person running into one is lower still. The people who benefit are you and the family with you at the scene. + +### 37. If the gambling has gone as far as borrowing money to carry on, or your mood stays low, call 12356 first, then hand your bank card and payment password to your family to manage + +- Cost: It costs nothing. One phone call. Hand over your bank card, your payment password and your borrowing apps to your family to manage. The hard part is admitting first that you cannot stop. +- In plain language: Among people who have gone to hospital about a gambling problem, death by suicide is 15 times that of their peers and overall death 1.8 times. Depression is the signal that tells you in advance who will die by suicide. So if the gambling has gone as far as borrowing to carry on and your mood stays low, call 12356 first. +- Benefit: A nationwide register study in Sweden picked out 2099 people diagnosed with gambling disorder at a specialist clinic from 2005 to 2016. Gambling disorder means gambling you cannot control, to the point where it damages your life. Compared with Swedes of the same age, these people aged 20 to 74 had 1.8 times the overall mortality and 15 times the suicide mortality (these two figures are standardised mortality ratios, SMR). 67 of them died during follow-up, and 21 of them by suicide. What predicted death by suicide was depression. The authors published a correction in 2023: 12.1 times for men and 16.1 times for women. From day 1 of month 5 in 2025, dialling 12356 anywhere in China reaches a psychological assistance hotline, open for no less than 18 hours a day (Sweden and the national figures) +- Evidence grade: B +- Sources: Karlsson A, Håkansson A. (2018). Gambling disorder, increased mortality, suicidality, and associated comorbidity: A longitudinal nationwide register study. Journal of Behavioral Addictions, 7(4), 1091–1099. ; Karlsson A, Håkansson A. (2023). Corrigendum to: Gambling disorder, increased mortality, suicidality, and associated comorbidity. Journal of Behavioral Addictions, 12(1), 302. ; National Health Commission (国家卫生健康委) (2024). Notice on the Use of "12356" as the Nationwide Unified Psychological Assistance Hotline Number, 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. +- Notes: The grade B is because there is only one study, and it counts only people who went to a specialist clinic. Their condition is more severe than average, so the 15 times figure does not carry over to everyone who plays cards. Handing the money to your family to manage is what the author has seen work, with no direct research, and on its own it is grade C. The legal lines on gambling: see item 13 in section 9 (no taking a cut, no online gambling). Whether you should pay off a family member's gambling debts: see item 44 in section 8 (gambling debts are not a joint debt of the couple). What to do after a suicidal thought has appeared: see item 32 in this section (tell one person near you first). The people who benefit are you mainly, and when the family spots the first signs they can pull that person into doing the same. + +### 38. Men who have sex with men and have high-risk sex: ask a designated AIDS hospital about preexposure prophylaxis, and taking it on time cuts infection by more than 70% + +- Cost: It is a prescription drug, and in most places you pay for it yourself, at a price from a few hundred yuan to several thousand yuan. Before it is prescribed they test for HIV, kidney function and hepatitis B. After that you go back for a check-up every 3 months. The hard part is taking it on time every day. +- In plain language: A person who has not been infected takes an antiviral medicine in advance. That is called preexposure prophylaxis. People who take it on time cut their risk of infection by more than 70%. People who take it now and then get almost nothing out of it. It is a prescription drug: the hospital has to check your body first and then prescribe it, and you go back for regular check-ups afterwards. +- Benefit: The drug is a combination tablet of emtricitabine and tenofovir. The iPrEx trial had 2499 men who have sex with men and transgender women take the drug daily or take a placebo. The group on the drug had 44% fewer infections (95% CI 15–63). Among the people whose blood showed the drug, protection was far stronger. The PROUD trial in the United Kingdom had 544 men who have sex with men. The group that started at once had 86% fewer infections than the group that waited a year (90% CI 64–96, 90% credible range). The IPERGAY trial in France had people take the drug on demand, before and after sex, and infections fell by 86% (95% CI 40–98). A meta-analysis used by the World Health Organization pools the trials. Overall RR 0.49 (0.33–0.73, lower by about 51%). In trials where more than 70% took the pills on time, RR 0.30 (0.21–0.45, lower by about 70%). Where fewer than 40% took them on time, RR 0.95, and the effect is close to nothing. In month 8 of 2020 the National Medical Products Administration (国家药监局) approved this drug for preexposure prophylaxis. Adults and adolescents weighing at least 35 kg can use it (several countries / China) +- Evidence grade: A +- Sources: Grant RM, Lama JR, Anderson PL, et al. (2010). Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. New England Journal of Medicine, 363(27), 2587–2599. ; McCormack S, Dunn DT, Desai M, et al. (2016). Pre-exposure prophylaxis to prevent the acquisition of HIV-1 infection (PROUD): effectiveness results from the pilot phase of a pragmatic open-label randomised trial. The Lancet, 387(10013), 53–60. ; Molina JM, Capitant C, Spire B, et al. (2015). On-demand preexposure prophylaxis in men at high risk for HIV-1 infection. New England Journal of Medicine, 373(23), 2237–2246. ; Fonner VA, Dalglish SL, Kennedy CE, et al. (2016). Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS, 30(12), 1973–1983. ; Xu JJ, Huang XJ, Liu XC, et al. (2020). Consensus statement on human immunodeficiency virus pre-exposure prophylaxis in China. Chinese Medical Journal, 133(23), 2840–2846. (Chinese version, 《中国 HIV 暴露前预防用药专家共识》) ; World Health Organization (2019). What's the 2+1+1? Event-driven oral pre-exposure prophylaxis to prevent HIV for men who have sex with men (WHO/CDS/HIV/19.8). ; General Office of the State Council (国务院办公厅) (2024). Plan for Containing and Controlling AIDS in China (2024–2030), 中国遏制与防治艾滋病规划（2024—2030 年）. +- Notes: The people in the trials were mainly men who have sex with men, so that is the group that should ask. Anyone else with high-risk sex can ask the doctor whether it suits them. Before it is prescribed you must confirm that you have not been infected. Taking it when you are already infected makes the virus resistant, and a self-test strip result is not a basis you can act on. Stopping the drug in someone with hepatitis B can bring on an attack of hepatitis, so that test comes first. The World Health Organization recommends on-demand use for men who have sex with men only, and it fits people averaging fewer than 2 times a week. The way to take it is 2 tablets some time between 2 and 24 hours before sex, then 1 tablet at 24 hours and 1 tablet at 48 hours. Women, and people with hepatitis B, can only take it every day. In the on-demand trials, the group taking the drug had more gastrointestinal and kidney side effects than the placebo group, so the check-ups cannot be skipped. China began a pilot programme among men who have sex with men in 2019, and the 2024 plan says "standardised implementation". No nationwide free policy could be found. Get it from the infectious disease department or the sexually transmitted infection and AIDS clinic of a local designated AIDS hospital, and call the local CDC first to ask which one can prescribe. It does not protect against syphilis, gonorrhoea and the like, so you still need a condom: see item 30 in this section (use a condom throughout). If high-risk sex has already happened, the remedy is item 38 in section 13 (get post-exposure prophylaxis within 72 hours). The people who benefit are you mainly, and it protects your partner as well. diff --git "a/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" "b/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" new file mode 100644 index 00000000..1031c587 --- /dev/null +++ "b/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" @@ -0,0 +1,388 @@ +[← 回总目录](../README.md) + +# 2. 不要慢慢死 + +本节只收两类慢性风险：一类对总死亡率影响大，一类证据硬。条目按性价比从高到低排。有几条标了「随机试验」，意思是把人随机分成两组来比，这种结果最可信。其余数字都来自只跟踪记录、不做分组的研究。这类研究算出的 HR、RR 等风险比里混着两种干扰：这些人本来就不一样，有人是先病了才这么做。所以这些数字只说明方向和大概多大，不代表照做就能降这么多。 + +[← Back to contents](../README.md) + +# 2. Do not die slowly + +This section takes in only two kinds of chronic risk: those that move all-cause mortality a lot, and those with hard evidence behind them. Entries run from the highest value for money to the lowest. A few are marked as randomised trials, which means people were randomly split into two groups and compared, and results like that are the most trustworthy. The other numbers come from studies that only followed records and did not assign groups. The risk ratios those studies report, such as HR and RR, carry two kinds of interference: the people were not alike to begin with, and some of them only started after they fell ill. So these numbers show direction and rough size only. They do not mean that following the advice lowers your risk by that much. + +### 1. Quit smoking, the earlier the better + +- Cost: No money, and it saves you money instead. A pack a day costs 20–30 yuan, and quitting leaves that money in your pocket. The hard part is the withdrawal period, which usually runs from a few weeks to several months. +- In plain language: Smokers live on average more than ten years less than non-smokers. Quitting before 40 wipes out about nine tenths of the death risk that going on smoking carries. The earlier you quit, the more you get back. +- Benefit: A US study that followed a group of people, recording only and not assigning groups, found that people who still smoke have a life expectancy more than 10 years shorter than people who have never smoked. Quitting before 40 removes about 90% of the death risk that continuing to smoke carries. People who quit at 25–34 live about 10 years longer, those who quit at 35–44 about 9 years longer, and those who quit at 45–54 about 6 years longer (the source reports these in years of life). Similar Chinese research: in the 2010s, the death risk of urban men who smoke was 1.65 times that of non-smokers (RR 1.65), and of rural men 1.22 times. Ten years after you stop, the risk carried by smoking has almost gone. +- Evidence grade: A +- Sources: Jha P et al. (2013). 21st-century hazards of smoking and benefits of cessation in the United States. NEJM. ; Chen Z et al. (2015). Contrasting male and female trends in tobacco-attributed mortality in China: evidence from successive nationwide prospective cohort studies. Lancet. ; Oberg M et al. (2011). Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet. (the two second-hand smoke figures) +- Notes: Breathing other people's smoke kills too: about 603,000 people died worldwide from second-hand smoke in 2004, about 1% of all deaths that year. If you do not smoke, stay away from other people's smoke, and above all keep your children out of it. For how to quit, the specific methods are in the next four items: item 3 (cessation drugs), item 4 (set a quit date), item 5 (go to a smoking cessation clinic), item 6 (e-cigarettes) + +### 2. Do not smoke at home or in the car, and do not let guests smoke at home + +- Cost: No money, and it takes no time. The hard part is having to ask family and guests not to smoke in the home. +- In plain language: About 600,000 people a year die worldwide from breathing other people's smoke, and close to 30% of them are children. Long-term exposure to second-hand smoke raises the risk of high blood pressure, heart disease and stroke by around 30%. What you take in at home does more harm than what you take in outside. +- Benefit: In 2004, 603,000 people worldwide died from second-hand smoke, about 1.0% of all deaths, and 28% of them were children. Pooling 57 studies: the odds of hypertension among people exposed to second-hand smoke were about 1.28 times those of people unexposed (odds ratio). Heart disease was 1.39 times, heart attack 1.50 times, and stroke 1.36 times. The risk of second-hand smoke at home is higher than outside the home +- Evidence grade: A +- Sources: Öberg M, Jaakkola MS, Woodward A, Peruga A, Prüss-Ustün A (2011). Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet: 「603,000 deaths were attributable to second-hand smoke in 2004, which was about 1·0% of worldwide mortality. 47% of deaths from second-hand smoke occurred in women, 28% in children, and 26% in men」, 「61% of DALYs were in children」. ; (2026). The Associations Between Secondhand Smoke Exposure and Various Cardiovascular Diseases: A Meta-Analysis. Nicotine & Tobacco Research: 57 studies, 「hypertension (OR: 1.28, 95% CI: 1.15 to 1.40), heart disease (OR: 1.39, 95% CI: 1.28 to 1.50), myocardial infarction (OR: 1.50, 95% CI: 1.17 to 1.84), stroke (OR: 1.36, 95% CI: 1.18 to 1.54)」「Home exposure has a higher risk of CVD than non-home exposure」. +- Notes: Start with the home, because you spend more time there, and pooled studies show that second-hand smoke at home does more harm than second-hand smoke outside. Children carry 61% of the health loss caused by second-hand smoke. For quitting yourself, see item 1 in this section. + +### 3. Do not try to quit on willpower alone, get the medicine first: it can more than double the success rate + +- Cost: Nicotine patches and chewing gum are over-the-counter, and you can buy them at a pharmacy counter. One course runs 8 to 12 weeks and costs a few hundred yuan to a little over a thousand. Varenicline and bupropion are prescription drugs, so you need a smoking cessation clinic or a respiratory medicine department to prescribe them. The money roughly cancels out against the cigarettes you stop buying over the same period. +- In plain language: Most people cannot quit on willpower alone. People who take varenicline succeed at more than twice the rate of people who take a placebo. People who use a substitute such as a nicotine patch or chewing gum succeed about 50% more often than people on no drug at all. Using a patch and chewing gum together adds another 20% or so. +- Benefit: Pooling 41 randomised trials and 17,395 people, the quitting success rate in the varenicline group was 2.32 times that of the placebo group (RR 2.32, 95% CI 2.15–2.51, which is the credible range; high-certainty evidence). Varenicline also beat bupropion (RR 1.36, about 36% higher) and a single kind of nicotine substitute (RR 1.25, about 25% higher). For nicotine substitutes against no drug: 133 trials and 64,640 people, the success rate was about 55% higher (RR 1.55, 95% CI 1.49–1.61). A patch plus one fast-acting form (chewing gum or lozenge) beat a single form by about 27% (RR 1.27, 95% CI 1.17–1.37, 16 trials and 12,169 people, high-certainty evidence). +- Evidence grade: A +- Sources: Livingstone-Banks J, Fanshawe TR, Thomas KH, et al. (2023). Nicotine receptor partial agonists for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD006103. ; Hartmann-Boyce J, Chepkin SC, Ye W, Bullen C, Lancaster T (2018). Nicotine replacement therapy versus control for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD000146. ; Theodoulou A, Chepkin SC, Ye W, et al. (2023). Different doses, durations and modes of delivery of nicotine replacement therapy for smoking cessation. Cochrane Database of Systematic Reviews, 6, CD013308. ; Shanghai Municipal Health Commission (上海市卫生健康委员会) (2021). Choosing the right drug makes quitting easier, 选对药物，让戒烟轻松一点: 「市场上可见的戒烟药物主要有三种，即尼古丁替代疗法药物、安非他酮、伐尼克兰」 — the smoking cessation drugs on sale come in three kinds, namely nicotine replacement therapy drugs, bupropion and varenicline. 「尼古丁替代疗法药物属于非处方药（OTC），可通过药店柜台购买；而安非他酮、伐尼克兰属于处方药，须到医院戒烟门诊或呼吸内科就诊，凭医师处方经药师调配后才能得到」 — nicotine replacement therapy drugs are over-the-counter and can be bought at a pharmacy counter, while bupropion and varenicline are prescription drugs, which you can only get by visiting a smoking cessation clinic or a respiratory medicine department at a hospital, on a doctor's prescription, dispensed by a pharmacist. +- Notes: The cessation drugs sold in China are the three kinds above, and only the nicotine substitutes are over-the-counter. The common reactions to varenicline are nausea, vivid dreams and poor sleep, so tell your doctor if you have a psychiatric history. A course of nicotine replacement usually runs 8 to 12 weeks, so do not stop suddenly on your own; taper as the doctor says. The medicine only handles the discomfort of the first few weeks of withdrawal. It does not handle the situations where you want to smoke, so use it together with item 4 in this section (set a quit date) and item 5 in this section (go to a smoking cessation clinic). + +### 4. Set a quit date and stop in one go on that day, do not taper first + +- Cost: No money. You pick one day on the calendar and tell your family and colleagues in advance. +- In plain language: Many people plan to cut a few cigarettes a day first and taper down to zero. But by the sixth month, 22% of the people who set a date and stopped in one go had not smoked again, against only 15.5% of those who tapered. So set a date and stop in one go on that day. +- Benefit: 697 British smokers were randomly assigned to two groups. One group stopped in one go on the quit date; the other cut their smoking by 75% in the two weeks before the quit date. Both groups had behavioural support from a nurse and used nicotine replacement around the quit date. At week 4, 49.0% of the group that stopped in one go had not relapsed (95% CI 43.8–54.2, which is the credible range), against 39.2% in the tapering group (34.0–44.4), a relative risk of 0.80 (0.66–0.93). At month 6, 22.0% of the one-go group (18.0–26.6) against 15.5% of the tapering group (12.0–19.7), a relative risk of 0.71 (0.46–0.91). Among people who already preferred to taper, the success rate at week 4 was also lower than among those who preferred to stop in one go (38.3% against 52.2%). +- Evidence grade: A +- Sources: Lindson-Hawley N, Banting M, West R, Michie S, Shinkins B, Aveyard P (2016). Gradual Versus Abrupt Smoking Cessation: A Randomized, Controlled Noninferiority Trial. Annals of Internal Medicine, 164(9), 585–592. ; Theodoulou A, Chepkin SC, Ye W, et al. (2023). Different doses, durations and modes of delivery of nicotine replacement therapy for smoking cessation. Cochrane Database of Systematic Reviews, 6, CD013308 (the passage on starting nicotine replacement early). +- Notes:Disputed. Cochrane used to have a review that compared these two methods head to head, and it concluded that they made little difference (RR 0.94, 95% CI 0.79–1.13, 10 trials and 3,760 people), but that review was withdrawn in 2019 and is no longer updated. So the best evidence available now is the randomised trial above, and its direction supports stopping in one go. In that trial both groups also started nicotine replacement before the quit date, and that practice has its own evidence: starting early gives a success rate about 25% higher than starting on the quit date itself (RR 1.25, 95% CI 1.08–1.44, 9 trials and 4,395 people, moderate-certainty evidence). A sensible plan is therefore to set the date first, start the patch two weeks ahead, and stop in one go on the day. + +### 5. Go to a smoking cessation clinic, or call 12320 to ask whether there is a cessation service near you + +- Cost: The registration fee runs from a few yuan to a few tens of yuan. Behavioural support is usually 4 to 8 sessions, each half an hour to an hour, totalling 5 hours or less. The hotline is free. +- In plain language: The medicine and having someone check in with you regularly are two separate things, and together they work best. People who take the medicine and also have a dedicated person following them up succeed at close to twice the rate of people who hear only one brief piece of advice. Telephone support counts too: a cessation line that calls you back without being asked raises the success rate by another 20% to 30%. +- Benefit: Pooling 52 trials and 19,488 people, the quitting success rate in the drug plus behavioural support group was 1.83 times that of usual care or brief advice (RR 1.83, 95% CI 1.68–1.98, which is the credible range; high-quality evidence). In the 43 trials that recruited in healthcare settings it was 1.97 times (1.79–2.18). Telephone counselling counted on its own: among people who had called a cessation line themselves, those who received several unprompted callbacks succeeded about 38% more often than those who only got materials or a single session (RR 1.38, 95% CI 1.19–1.61, 14 trials and 32,484 people); among people who had not called a line themselves and were offered telephone counselling by the researchers, about 25% more (RR 1.25, 95% CI 1.15–1.35, 65 trials and 41,233 people). Both are moderate-certainty evidence. +- Evidence grade: A +- Sources: Stead LF, Koilpillai P, Fanshawe TR, Lancaster T (2016). Combined pharmacotherapy and behavioural interventions for smoking cessation. Cochrane Database of Systematic Reviews, 3, CD008286. ; Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J (2019). Telephone counselling for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD002850. ; China CDC (中国疾病预防控制中心) (2021). The 'China Smoking Cessation Platform' WeChat mini programme is officially live: 「在健康中国行动控烟行动工作组指导下中国疾病预防控制中心和世界卫生组织驻华代表处联合制作了中国权威戒烟资源库『中国戒烟平台』微信小程序。小程序内容主要包括：戒烟热线、戒烟门诊详细信息，以及线上戒烟服务资源等」 — under the guidance of the tobacco control working group of the Healthy China Action, the China CDC and the WHO Representative Office in China jointly produced the 'China Smoking Cessation Platform' WeChat mini programme, an authoritative library of smoking cessation resources for China. The mini programme mainly carries quitlines, details of cessation clinics, and online cessation services. The footer of the same page gives 「健康咨询电话：12320」 — health consultation number: 12320. +- Notes: How to find one in China: search for the 'China Smoking Cessation Platform' mini programme on WeChat, which lists quitlines and cessation clinics by region. You can also call 12320 and ask whether there is a cessation clinic near you. Most of the trials above were done abroad, and the strength of the service varies a lot between clinics in different places, so read the numbers as direction only. For the medicine to go with it, see item 3 in this section (cessation drugs). + +### 6. Consider e-cigarettes only if you have already failed to quit, and people who do not smoke should stay away + +- Cost: The device and the pods cost from a few tens to a few hundred yuan a month. In China you can only buy tobacco flavours, and only in shops that hold a tobacco monopoly retail licence. +- In plain language: People who switch from cigarettes to nicotine e-cigarettes succeed about 60% more often than people who use nicotine patches, and that finding is reliable. But e-cigarettes are not harmless, and there is still no long-term safety data. For people who do not smoke, they bring no benefit at all. +- Benefit: A continuously updated review included 80 randomised trials and 29,861 people. Compared with nicotine replacement therapy, nicotine e-cigarettes raised the quitting success rate by about 61% (RR 1.61, 95% CI 1.23–2.12, which is the credible range; high-certainty evidence, 11 trials and 4,114 people). In absolute terms, that is about 4 more people in every 100 who quit (1 to 7 more). The rate of serious adverse events was about the same in both groups (risk difference 0.01, 95% CI −0.01 to 0.02, moderate-certainty evidence). Compared with behavioural support alone or no support, about 75% higher (RR 1.75, 95% CI 1.39–2.20, low-certainty evidence). +- Evidence grade: A +- Sources: Lindson N, Livingstone-Banks J, Butler AR, et al. (2026). Electronic cigarettes for smoking cessation. Cochrane Database of Systematic Reviews, 8, CD010216. ; State Tobacco Monopoly Administration (国家烟草专卖局) (2022). Measures for the Administration of E-Cigarettes, 电子烟管理办法（国家烟草专卖局公告 2022 年第 1 号，2022 年 5 月 1 日施行）. 第十八条: anyone doing e-cigarette retail must apply to the tobacco monopoly administration for a tobacco monopoly retail licence. 第二十二条: 「禁止向未成年人出售电子烟产品。」 — selling e-cigarette products to minors is prohibited. 第二十三条第二款: 「任何个人、法人或者其他组织不得通过本办法规定的电子烟交易管理平台以外的信息网络销售电子烟产品、雾化物和电子烟用烟碱等。」 — no individual, legal person or other organisation may sell e-cigarette products, e-liquids or nicotine for e-cigarettes over any information network other than the e-cigarette trading platform these Measures provide for. 第二十六条: 「禁止销售除烟草口味外的调味电子烟和可自行添加雾化物的电子烟。」 — the sale of flavoured e-cigarettes other than in the tobacco flavour, and of e-cigarettes you can add e-liquid to yourself, is prohibited. +- Notes:Disputed. The World Health Organization does not recommend e-cigarettes as a smoking cessation method. The reason is that long-term safety data is insufficient, and they get young people started on nicotine. Britain has put them into its cessation services instead. China does not ban e-cigarette sales outright, but it regulates them tightly, and the rules are written into the Measures for the Administration of E-Cigarettes. Only tobacco flavour may be sold; fruit flavours and e-cigarettes you can add e-liquid to yourself may not. You can only buy in a physical shop that holds a tobacco monopoly retail licence. Selling on any other website, online shop or social app outside the national e-cigarette trading platform is illegal, and selling to minors is prohibited. So any fruit-flavoured e-cigarette still on sale is coming through an illegal channel and sits outside the regulatory system. That is how the 'heady' e-cigarettes spiked with synthetic cannabinoids got out, see item 4 in section 22 (do not take a pod someone hands you). The products tested in the review are all regulated nicotine products; e-liquid of unknown origin is not covered. In order, try item 3 in this section (cessation drugs) first, because the evidence there is harder and it is cheaper. + +### 7. Stop drinking sugary drinks, and switching to sugar-free ones is not a solution + +- Cost: No money, and it takes no time. Switching to plain water or unsweetened tea even saves money. The hard part is dropping the habit of drinking sweet drinks, sugar-free ones included. +- In plain language: People who drink more than two bottles of sweet drinks a day have about 20% higher death rates in the same period than people who hardly drink them. One to two bottles a day is about 10% higher. Switching to sugar-free did not show a benefit either: people who drink more than two cups of artificially sweetened drinks a day are instead about a quarter higher. +- Benefit: The United States has two large studies that followed a group of people. Both only followed records, with no control group, and together cover about 118,000 people and 36,000 deaths. People drinking more than 2 servings a day had a death risk about 21% higher than people drinking less than 1 serving a month (HR 1.21). People drinking 1–2 servings a day were about 14% higher (HR 1.14). The European EPIC cohort also only followed records, with no control group, and covers 450,000 people across 10 countries. Comparing people who drink more than 2 cups a day with people who drink less than 1 cup a month: for sugar-sweetened soft drinks the risk was about 8% higher (HR 1.08, 95% CI 1.01–1.16). For artificially sweetened drinks, the sugar-free versions, about 26% higher (HR 1.26, 1.16–1.35). For all soft drinks taken together, about 17% higher (HR 1.17, 1.11–1.22). +- Evidence grade: A +- Sources: Malik VS et al. (2019). Long-Term Consumption of Sugar-Sweetened and Artificially Sweetened Beverages and Risk of Mortality in US Adults. Circulation. ; Mullee A, Romaguera D, Pearson-Stuttard J, et al. (2019). Association Between Soft Drink Consumption and Mortality in 10 European Countries. JAMA Internal Medicine, 179(11), 1479-1490. +- Notes:Disputed, and the dispute is about sugar-free drinks. The association with artificially sweetened drinks is stronger than the association with sugary ones, and the cause is probably running the other way: people switch to the sugar-free versions only after they are already overweight or have diabetes. So do not conclude from this that sweeteners are more toxic. It does at least show that the claim 'switching to sugar-free means you are fine' has no evidence behind it. Switching to plain water or unsweetened tea is the safest option. Also, these studies only followed records and did not run a controlled trial, and people who drink a lot of sweet drinks already have worse lifestyles, so the numbers may be too high. + +### 8. Do not chew betel quid + +- Cost: No money, and it takes no time. You also save the money you spent on betel quid. The hard part is giving up the habit of chewing. +- In plain language: People who chew betel quid have about 20% higher death rates in the same period than people who do not. The risk of diabetes and metabolic syndrome is nearly 50% higher. The risk of oral cancer and oesophageal cancer has to be counted separately. +- Benefit: Pooling 17 Asian studies and 388,000 people. People who chew betel quid have a death risk about 21% higher than people who do not (relative risk 1.21, P=0.02). The mortality figure covers 179,600 of them. The risk of diabetes is about 47% higher (1.47). The risk of metabolic syndrome is about 51% higher (1.51). Betel quid is itself a known risk factor for oral cancer and oesophageal cancer +- Evidence grade: A +- Sources: Yamada T, Hara K, Kadowaki T (2013). Chewing betel quid and the risk of metabolic disease, cardiovascular disease, and all-cause mortality: a meta-analysis. PLoS One, 8(8), e70679. +- Notes: The longer you have chewed and the more you chew each day, the higher the risk of oral cancer. After you quit, the risk falls over time. Betel quid containing tobacco is more dangerous + +### 9. Replace the salt in your home with low-sodium salt (potassium salt) + +- Cost: A bag costs a few yuan more than ordinary salt. Switch when you next buy salt and it takes no extra time. The taste barely changes. +- In plain language: For people who have had a stroke, or who are over 60 and have high blood pressure, replacing the salt at home with low-sodium salt lowers the chance of dying within five years by about 12% and the chance of stroke by about 14%. +- Benefit: A randomised trial done in rural China assigned 20,995 people to two groups, all of them people who had had a stroke or hypertensive patients over 60, and followed them for 4.74 years. Result: the group using low-sodium salt had a death risk about 12% lower than the group using ordinary salt (RR 0.88). Stroke was about 14% lower (RR 0.86). Major cardiovascular events were about 13% lower (RR 0.87). For events of high blood potassium, the two groups showed no statistically significant difference. +- Evidence grade: A +- Sources: Neal B et al. (2021). Effect of Salt Substitution on Cardiovascular Events and Death. NEJM. ; O'Donnell M et al. (2014). Urinary sodium and potassium excretion, mortality, and cardiovascular events. NEJM. (the PURE side of the dispute) +- Notes:Disputed. The PURE study only followed records over the long term, with no control group. It reports that people excreting less than 3 g of sodium a day have about 27% higher risk of death plus cardiovascular events, and people excreting more than 7 g a day about 15% higher. On its results, both too little and too much are bad, and the middle is best. But in the Chinese rural randomised trial cited in this item, low-sodium salt only replaced part of the salt, so sodium intake did not come down that far. Note also that the trial subjects were high-risk older people. Healthy young people get much less from switching salt. People with impaired kidney function, or people taking potassium-sparing drugs, should ask a doctor before switching. + +### 10. Brush your teeth properly, clean between the teeth once a day, and replace missing teeth promptly + +- Cost: Dental floss or interdental brushes cost a few tens of yuan a year, and a professional cleaning costs a hundred or two hundred yuan each time. It adds two or three minutes a day. The hard part is making interdental cleaning a daily habit. +- In plain language: People who use floss or interdental brushes every day have about 10% lower death rates in the same period. Older people who have lost all their teeth have about 90% higher death rates. People left with fewer than 20 teeth are about double. +- Benefit: One study in Japan covered 9,676 people and followed them for 6 years. People who used interdental cleaning tools had a death risk about 11% lower (risk ratio 0.89). People who used tongue-cleaning tools had one about 23% lower (0.77). A separate pooling of several studies looked at older people living in the community. People with not one tooth left had about 1.87 times the chance of dying (odds ratio, 95% CI 1.35–2.59, which is the credible range). People with fewer than 20 teeth had about 2.04 times (1.67–2.49) +- Evidence grade: B +- Sources: Wang K, Matsuyama Y, Kiuchi S, et al. (2026). Routine oral health practices and all-cause mortality. Journal of Dentistry. ; Ko MJ, Seo S, So JS, et al. (2026). Deteriorated oral health and function as risk factors for physical disability and mortality in community-dwelling older adults: a systematic review and meta-analysis. European Geriatric Medicine. +- Notes:Disputed. These studies only followed records. People with bad teeth often have worse health and worse finances across the board. Whether bad teeth drag the body down, or poor health drags the teeth down, is not clear. But this costs almost nothing, and periodontal disease and missing teeth affect eating on their own + +### 11. Walk 7000–8000 steps a day + +- Cost: No money. Walking for 60–90 minutes a day can be split into the commute and the shopping trip. The hard part is doing it every day. +- In plain language: People who walk 5800 steps a day have about 40% lower death rates in the same period than people who walk only 3500. At 7800 steps the fall is about 45%. Past that the figure drops a little further, then flattens out. For people over 60, 6000–8000 steps is enough. +- Benefit: Pooling 15 studies that followed a group of people, all of them only followed records with no control group. The 15 studies cover 47,471 people and 3,013 deaths. People are split into four bands by daily steps, and the median of the lowest band is 3553 steps a day. The medians of the other three bands are 5801, 7842 and 10,901 steps a day. The death risk of those three bands is about 40%, 45% and 53% lower than the lowest band (HR 0.60, 0.55, 0.47). Among people over 60, the benefit flattens out after 6000–8000 steps a day. Among people under 60 it flattens out after 8000–10000 steps. A separate pooled analysis of several studies: the effect starts at about 3867 steps a day. After that, every extra 1000 steps a day cuts the death risk by about 15%. +- Evidence grade: A +- Sources: Paluch AE et al. (2022). Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. Lancet Public Health. ; Banach M et al. (2023). The association between daily step count and all-cause and cardiovascular mortality: a meta-analysis. European Journal of Preventive Cardiology. +- Notes: The stretch from 4000 steps to 7000 steps is the best value, and you do not have to reach ten thousand. Note that these studies only followed records: the people who walked least include some who were too ill to walk at all. They walked little because they were about to die, not the other way round. So the gap these studies report is often too large. The direction is clear all the same. This item and item 14 (150–300 minutes of moderate exercise a week) are the same thing measured two ways, and hitting either one is enough. + +### 12. If you have high blood pressure or high blood lipids, take the medicine as prescribed and do not stop on your own + +- Cost: Blood pressure pills and statins bought through the volume-based procurement scheme cost from a few yuan to a few tens of yuan a month. One dose a day takes very little time. The hard part is carrying on with them when you feel fine and think you no longer need them. +- In plain language: For every 10 mmHg that blood pressure falls, the chance of dying falls about 13% and the chance of stroke falls about 27%. For every unit that statins take off LDL cholesterol, the chance of dying falls about 10%. People who take their medicine on schedule have a death rate 30% to 45% lower than people who often miss doses. +- Benefit: Pooling several randomised trials: for every 10 mmHg that systolic blood pressure falls, the death risk is about 13% lower (RR 0.87). Systolic blood pressure is the higher of the two numbers in a reading. Major cardiovascular events are about 20% lower (RR 0.80). Stroke is about 27% lower (RR 0.73). Heart failure is about 28% lower (RR 0.72). Statins bring LDL cholesterol down. For every 1.0 mmol/L that LDL falls, the death risk is about 10% lower (RR 0.90). For the same fall, major vascular events are about 22% lower (RR 0.78). A separate pooling of several follow-up studies, all only followed records with no control group: people who take at least 80% of their doses as prescribed have a lower death risk than people who often miss doses. For statins the figure is about 45% lower (RR 0.55). For blood pressure pills it is about 29% lower (RR 0.71). +- Evidence grade: A +- Sources: Ettehad D et al. (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Cholesterol Treatment Trialists' (CTT) Collaboration (2010). Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170 000 participants in 26 randomised trials. Lancet. ; Chowdhury R et al. (2013). Adherence to cardiovascular therapy: a meta-analysis of prevalence and clinical consequences. European Heart Journal. +- Notes: Only do this if the doctor has judged that you need the medicine. Healthy people do not. The figures for the group that keeps taking the medicine come from studies that only followed records, and people who manage to keep taking it are more disciplined to begin with, so the benefit is overstated. The figures from the two randomised trial pools above are more reliable. This section has no separate item on blood glucose: the evidence that pushing glucose lower lowers all-cause mortality is not as consistent as it is for blood pressure and blood lipids. + +### 13. Sleep about 7 hours a night and keep a fixed routine + +- Cost: No money. You have to set aside the time to sleep, and for most people that means giving up some phone time. The hard part is fixing your bedtime and your wake-up time. +- In plain language: People who sleep under 7 hours over the long run have about 10% higher death rates in the same period. The more irregular the routine, the higher the risk. The most regular sleepers have death rates 20% to nearly 50% lower than the least regular. How regular your routine is therefore says more about your death risk than how long you sleep. +- Benefit: Pooling 16 follow-up studies, all only followed records with no control group, covering 1,380,000 people and 113,000 deaths. Short sleepers have a death risk about 12% higher (RR 1.12). Long sleepers about 30% higher (RR 1.30). A separate summary bands people by how long they sleep, and the 7-hour band has the lowest risk. Below 7 hours, every 1 hour less raises the risk about 6% (RR 1.06). Above 7 hours, every 1 hour more raises it about 13% (RR 1.13). UK Biobank has wristband records for about 61,000 people, again only followed records with no control group. People are split into five bands by how regular their routine is. The four more regular bands have a death risk 20%–48% lower than the least regular band. How regular the routine is predicts death better than how long you sleep. A separate UK Biobank study measured bedtimes with wristbands in 51,562 people, and 3,853 of them developed cardiovascular disease during the study. Social jet lag is the gap between the midpoint of sleep on a weekday and the midpoint at the weekend. People with a social jet lag of 2 hours or more have a cardiovascular risk about 30% higher (HR 1.30, 95% CI 1.11–1.54). The same association holds among people whose sleep duration is normal. +- Evidence grade: A +- Sources: Cappuccio FP et al. (2010). Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. ; Yin J et al. (2017). Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. JAHA. ; Windred DP et al. (2024). Sleep regularity is a stronger predictor of mortality risk than sleep duration: A prospective cohort study. Sleep. ; Kumar N, Krishnamurthy S (2026). Social jet lag is associated with incident cardiovascular disease independent of sleep duration and cardiac genetic risk. Journal of Internal Medicine. +- Notes: What needs changing is sleeping too little and having an irregular routine. There is no need to deliberately cut your sleep short. Long sleepers having a higher risk is most likely the cause running the other way: depression, chronic illness and sleep apnoea make people sleep more. Regularity of routine for now rests on two follow-up studies, both only followed records with no control group, so on its own it is grade B evidence. How to make up for a night up late is in item 39 in this section (sleep earlier the next night). + +### 14. Build up 150–300 minutes of moderate exercise a week, and brisk walking counts + +- Cost: No money. 20–45 minutes a day. The hard part is keeping it up over the long run. +- In plain language: People who brisk-walk for 150–300 minutes a week have about 30% lower death rates in the same period than people who do not exercise. Even if you fall short of the target, the fall is about 20%. Three to five times the target brings the figure down to about 40% lower. Going further will not lower it more, but it does no harm. +- Benefit: A pooled analysis of several follow-up studies, all only followed records with no control group: people whose activity reaches 1 to 2 times the guideline minimum have a death risk about 31% lower than people who do not exercise (HR 0.69). That amount is 7.5–15 MET·h a week, which is about 150–300 minutes of brisk walking a week. People who fall short of the minimum are still about 20% lower (HR 0.80). People who reach 3 to 5 times the minimum are about 39% lower (HR 0.61), and that is where the reduction tops out. Going further will not lower it more, but it does no harm (above 10 times, HR 0.69). A separate pooled study using accelerometers: the most active quarter of people by moderate-to-vigorous activity have a death risk about 48% lower than the least active quarter (HR 0.52). An accelerometer is a device you wear that measures activity directly. +- Evidence grade: A +- Sources: Arem H et al. (2015). Leisure time physical activity and mortality: a detailed pooled analysis of the dose-response relationship. JAMA Internal Medicine. ; Ekelund U et al. (2019). Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis. BMJ. +- Notes: This item and item 11 (7000–8000 steps a day) only need one of the two. Note that these studies only followed records. The ones that measured activity with a device tracked people for a short time and the participants were mostly older, so the figures include an element of "became inactive after falling ill". The real benefit is not as large as a risk ratio of 0.52. + +### 15. Play a racket sport three times a week, 45 minutes each time + +- Cost: Court hire costs a few tens of yuan each time. About 2 hours a week. The hard part is getting enough people and a court together, and turning up every week. +- In plain language: People who regularly play tennis, badminton or table tennis have about half the death rate in the same period as people who do not, and cardiovascular death is about 56% lower. Swimmers are about 28% lower. Running and football showed no difference in this dataset. +- Benefit: A UK follow-up study of 80,300 people compared people who did a given sport with people who did not. People who played a racket sport had a death risk about 47% lower (risk ratio 0.53, 95% CI 0.40–0.69, credible range). Cardiovascular death was about 56% lower (0.44, 0.24–0.83). Racket sports means tennis, badminton and table tennis. For swimmers the two figures are 0.72 and 0.59, for aerobics 0.73 and 0.64, and for cycling 0.85. Running and football showed no statistically significant difference +- Evidence grade: A +- Sources: Oja P, Kelly P, Pedisic Z, et al. (2017). Associations of specific types of sports and exercise with all-cause and cardiovascular-disease mortality: a cohort study of 80 306 British adults. British Journal of Sports Medicine, 51(10), 812-817. +- Notes:Disputed. These studies only followed records. People who take up a sport are healthier to begin with and have more social contact. The fact that running showed no difference also suggests that what matters may be which kind of person picks which sport. Do not read this item as saying running is useless, because this section has other items that go by total activity. + +### 16. Add up the short bursts of effort, such as stairs and walking fast, to four or five minutes a day + +- Cost: No money, and no need to carve out extra exercise time. The hard part is remembering to take the stairs and to walk faster instead of taking the easy way. +- In plain language: People who do no exercise at all still have a death rate about 40% lower in the same period than people who do none at all, as long as they have three one- or two-minute bursts of effort a day. Cardiovascular death is nearly half lower. Effort here means things like taking the stairs and hurrying to get somewhere. +- Benefit: A UK Biobank study of 25,200 people who did no deliberate exercise in ordinary life. The average follow-up was 6.9 years, and 852 of them died. Some people had 3 bouts a day of short vigorous activity lasting 1–2 minutes each. Compared with people who had none, their death risk and cancer death risk were 38%–40% lower. Cardiovascular death risk was 48%–49% lower. People who totalled 4.4 minutes a day of this activity had a death risk and cancer death risk 26%–30% lower. Cardiovascular death risk was 32%–34% lower +- Evidence grade: A +- Sources: Stamatakis E, Ahmadi MN, Gill JMR, et al. (2022). Association of wearable device-measured vigorous intermittent lifestyle physical activity with mortality. Nature Medicine, 28, 2521-2529. +- Notes:Disputed. This study measured activity with wearable devices, which is more reliable than asking people to fill in a questionnaire from memory. But it still only followed records, it did not run a controlled trial, and it tracked people for 6.9 years. The subjects were people who did no exercise at all, so this item does not apply to people who already exercise regularly + +### 17. Do 30–60 minutes of strength training a week + +- Cost: Bodyweight squats and press-ups cost nothing. Train 1–2 times a week, 20–30 minutes each time. The hard part is fitting it into the week's schedule. +- In plain language: People who do 30–60 minutes of strength training a week have a death rate 10% to 20% lower in the same period than people who do none. Doing more does not make it better. Strength work and aerobic work together work best. +- Benefit: Pooling several follow-up studies, all only followed records with no control group: people who do strength training have a death risk 10%–17% lower than people who do not. More is not better: the fall is largest at about 30–60 minutes a week, about 10%–20%, and training beyond that adds nothing. People who do both strength and aerobic work have a lower death risk than people who do neither. +- Evidence grade: A +- Sources: Momma H et al. (2022). Muscle-strengthening activities are associated with lower risk and mortality in major non-communicable diseases: a systematic review and meta-analysis of cohort studies. British Journal of Sports Medicine. +- Notes: These studies only followed records, and the activity level was self-reported. The evidence that training too much makes matters worse is very weak, so there is no reason to hold back on that account. Strength training also helps older people avoid falls and keep their muscle, and those points are in section 1. + +### 18. Do not sit for long stretches, get up and move from time to time + +- Cost: No money, and getting up a few times takes very little time. The hard part is that once you sit down you lose track of time, so set a reminder. +- In plain language: The people who sit longest have a death rate in the same period 2.6 times that of the people who sit least. Sitting for a long unbroken stretch carries a separate risk on top of that. But as long as you get 60–75 minutes of moderate activity a day, most of the extra risk from sitting is cancelled out. +- Benefit: A US follow-up study that only followed records, with no control group, measured activity with accelerometers. People were split into four groups by total daily sitting time. The group that sat the most had a death risk 2.63 times that of the group that sat least (HR 2.63). People were then split into four groups by how long they sat in one unbroken stretch: the group with the longest unbroken sitting had a figure 1.96 times that of the group with the shortest (HR 1.96). So total sitting time and the length of a single unbroken stretch are each separately associated with the risk. A separate pooled analysis covering more than a million people, only followed records with no control group: people who sit more than 8 hours a day and barely exercise have a death risk 1.59 times that of another group (HR 1.59). That other group sits under 4 hours a day and is the most active. About 60–75 minutes of moderate activity a day cancels out the extra risk from sitting. The most active group has an HR of only 1.04 even when sitting more than 8 hours a day, which is not statistically significant. +- Evidence grade: A +- Sources: Diaz KM et al. (2017). Patterns of Sedentary Behavior and Mortality in U.S. Middle-Aged and Older Adults: A National Cohort Study. Annals of Internal Medicine. ; Ekelund U et al. (2016). Does physical activity attenuate, or even eliminate, the detrimental association of sitting time with mortality? A harmonised meta-analysis of data from more than 1 million men and women. Lancet. +- Notes: The reliable conclusion is that how much you move matters more than how long you sit, and that when activity is enough the risk from sitting largely disappears. Note that these studies only followed records, and the people who sit the most include a large number who were already ill, so a risk as high as 2.63 times clearly contains an element of "became sedentary after falling ill". In the same pooled analysis, the risk of watching more than 3 hours of television a day is present at every activity level, and the most active group only pushes that starting point out to 5 hours. + +### 19. Eat less processed meat (ham, bacon, sausage, luncheon meat) + +- Cost: No money, and it may even save money, and it takes no time. The hard part is cutting back on the mouthful you actually like. +- In plain language: People who eat the most ham, bacon and sausage have a death rate about 20% higher in the same period than people who eat the least. Every extra daily serving of processed meat adds 20% to the figure. +- Benefit: Pooling several follow-up studies, all only followed records with no control group: the group that eats the most processed meat has a death risk about 23% higher than the group that eats the least (RR 1.23). Red meat of all kinds is about 29% higher (RR 1.29). Unprocessed red meat is about 10% higher (RR 1.10), but that one is not statistically significant. A separate pooled analysis works from the amount eaten per day. Every extra daily serving of processed meat, so 1 more serving a day, raises the death risk about 23% (RR 1.23). Every extra daily serving of red meat, 1 more serving a day, raises it about 10% (RR 1.10). +- Evidence grade: A +- Sources: Larsson SC, Orsini N (2014). Red meat and processed meat consumption and all-cause mortality: a meta-analysis. American Journal of Epidemiology. ; Schwingshackl L et al. (2017). Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. American Journal of Clinical Nutrition. ; Johnston BC et al. (2019). Unprocessed Red Meat and Processed Meat Consumption: Dietary Guideline Recommendations From the NutriRECS Consortium. Annals of Internal Medicine. (the NutriRECS guidelines, the other side of the dispute) +- Notes:Disputed. The NutriRECS guidelines use a standard evidence grading system (GRADE) and rate these studies as not very reliable. They give one very weak recommendation: carry on eating the way you do now. The two sides disagree about whether the evidence is hard enough, not about the direction. No study says processed meat is good for you. Unprocessed red meat has a small effect and shows no statistically significant difference. What needs controlling is the processed meat. + +### 20. Drink less alcohol, or none + +- Cost: No money, and it saves money, and it takes no time. The hard part is the awkward moment when people at a gathering push a drink on you. +- In plain language: Past 100 g of pure alcohol a week, the more you drink the earlier you die. 100 g of pure alcohol is about 2.5 L of beer. People drinking 100–200 g a week at 40 live about half a year less; people drinking 200–350 g lose one to two years; people drinking more than 350 g lose four or five years. Judged by overall health, the best value is not drinking. +- Benefit: Pooling 83 follow-up studies, only followed records with no control group, covering 600,000 people who drink alcohol. The lowest death risk is at no more than 100 g of pure alcohol a week. That amount is about 2.5 L of 5% beer, or 300 mL of 40-proof baijiu. People drinking 100–200 g a week at 40 have a life expectancy about 6 months shorter. People drinking 200–350 g a week lose 1–2 years. People drinking more than 350 g lose 4–5 years. The source reports all three in years of life. The Global Burden of Disease study GBD 2016 adds up all kinds of health loss and puts the lowest-risk amount at 0 drinks a week. A separate pooled analysis corrected for bias in study design. People drinking 1.3–24 g a day have a death risk about 7% lower than people who never drink (RR 0.93), but that one is not statistically significant. People drinking 45–64 g a day are about 19% higher (RR 1.19). People drinking above 65 g a day are about 35% higher (RR 1.35). +- Evidence grade: A +- Sources: Wood AM et al. (2018). Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. ; GBD 2016 Alcohol Collaborators (2018). Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. ; Zhao J et al. (2023). Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Network Open. ; Di Castelnuovo A et al. (2006). Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Archives of Internal Medicine. (the other side of the dispute) +- Notes:Disputed. The side that argues drinking less is good points to a pooled analysis of 34 long-term follow-up studies. It reports that light drinkers have an all-cause mortality up to 17%–18% lower. The amount it gives is no more than 4 drinks a day for men and 2 for women. Within that amount, the more people drink the lower their mortality. The other side argues that the "non-drinkers" group includes people who gave up because of illness and people who were in poor health to begin with. One pooled analysis adjusted specifically for those confounders, and once that was done the protective effect disappeared. The Global Burden of Disease study also puts 0 as the safest. The safe way to read it: drinking less will not clearly shorten your life, and drinking more certainly will. Starting to drink for health reasons has no basis. What to do if you want to drink less is in item 22 in this section (count the drinks first, then see a doctor). People who drink every day should not stop abruptly on their own, see item 21 in this section (do not stop abruptly on your own). + +### 21. If you drink every day and shake or feel anxious when you stop, do not stop abruptly on your own + +- Cost: The registration fee for psychiatry or an addiction medicine clinic runs from a few yuan to a few tens of yuan. Severe cases need a few days in hospital. +- In plain language: Someone who drinks a lot every day and stops suddenly goes into withdrawal. Mild withdrawal means shaking, sweating, anxiety and sleeplessness. Severe withdrawal means seizures or delirium tremens, which is a medical emergency that kills. Norway followed more than thirty thousand people, and of those who had delirium tremens, 8% died each year afterwards. If you want to quit, go to hospital first and say how much you drink each day. +- Benefit: A 2026 clinical review states that people who drink heavily over the long run develop withdrawal syndrome after stopping suddenly or cutting down sharply, with tremor, autonomic overactivity, anxiety, insomnia, perceptual disturbances, seizures and delirium tremens, and that in ordinary internal medicine this is 「常见且可能危及生命」 — common and potentially life-threatening. A Norwegian national register cohort of 36,287 people diagnosed with alcohol dependence, alcohol withdrawal state or delirium tremens between 2009 and 2015: the annual mortality in the delirium tremens group was 8.0%, in the withdrawal state group 5.0%, and in the alcohol dependence group 3.6%. The standardised mortality ratio for the delirium tremens group was 9.8 (95% CI 8.9–10.7, the credible range), which means the death risk is about 9.8 times that of the general population of the same age. +- Evidence grade: B +- Sources: Caputo F, Lungaro L, Costanzini A, De Giorgio R, Addolorato G (2026). Alcohol withdrawal syndrome in hospitalized patients: a practical review. European Journal of Internal Medicine, 107103. ; Bramness JG, Heiberg IH, Høye A, Rossow I (2023). Mortality and alcohol-related morbidity in patients with delirium tremens, alcohol withdrawal state or alcohol dependence in Norway: A register-based prospective cohort study. Addiction, 118(12), 2352–2359. +- Notes: How to judge it yourself: if you drink almost every day, and missing a day brings shaking, sweating, anxiety and sleeplessness, and you need a drink in the morning to feel right, then any one of those means you should not stop on your own. What hospitals normally do is give benzodiazepines to get you through those few days and add thiamine (vitamin B1) at the same time. The grade is B because the Norwegian cohort says what happened later to people who had delirium tremens, not what happens to people who stop abruptly on their own, and getting from one to the other is an inference. You still need to quit, but quit in hospital instead. How much a week counts as too much is in item 20 in this section (drink less or none), and what to do if you want to drink less is in item 22 in this section (count the drinks first, then see a doctor). + +### 22. To drink less, first count how much you drink in a week, then talk to a doctor for a few minutes + +- Cost: No money. Counting the drinks takes one line a day. Asking the question while you are at the clinic costs nothing extra. +- In plain language: A doctor who spends a few minutes looking at how much you drink, explaining the risk and setting you a target leaves you drinking 20 grams of pure alcohol less a week on average a year later, which is about half a litre of beer. Talking for longer brings no further benefit. For people who cannot stop on their own, two drugs proved effective abroad, and for every 12 people treated about 1 more does not go back to drinking as much as before. +- Benefit: Pooling 34 randomised trials and 15,197 people, people who received a brief intervention drank 20 grams of pure alcohol less per week a year later than people who received only the minimum intervention or none (95% CI 12 to 28 grams less, the credible range; moderate-quality evidence). The baseline drinking of the participants in these studies averaged 244 grams a week. A brief intervention means advice or lifestyle counselling of no more than 5 sessions and under 60 minutes in total; talking for longer brought no extra benefit. For the drugs: pooling 122 randomised trials and 22,803 people, acamprosate means that for every 12 people treated (95% CI 8 to 26) about 1 more does not return to drinking at all. Oral naltrexone at 50 milligrams a day means that for every 12 people treated (8 to 26) about 1 more does not return to heavy drinking. +- Evidence grade: A +- Sources: Kaner EF, Beyer FR, Muirhead C, et al. (2018). Effectiveness of brief alcohol interventions in primary care populations. Cochrane Database of Systematic Reviews, 2, CD004148. ; Jonas DE, Amick HR, Feltner C, et al. (2014). Pharmacotherapy for adults with alcohol use disorders in outpatient settings: a systematic review and meta-analysis. JAMA, 311(18), 1889–1900. +- Notes: The benefit is graded "small" because both of these bodies of evidence measure how much people drink, not mortality, and by the rule of this book a surrogate endpoint alone gets "small". Most of the drug trials were done abroad, most participants already met the diagnostic criteria for alcohol dependence, and every trial also gave psychosocial support, so the figures are the extra benefit on top of that. Whether these two drugs are approved in China and whether you can be prescribed them is up to your doctor and the drug's own package information, so do not buy them yourself online. If you already have withdrawal symptoms, see item 21 in this section (do not stop abruptly on your own). + +### 23. Eat a small handful of nuts every day + +- Cost: 28 grams a day costs two or three hundred yuan a year. You eat them in passing, and it takes no time. +- In plain language: People who regularly eat nuts have about 20% lower death rates in the same period than people who do not. Eating them once a week already shows a difference, about 10%. Eating a handful a day shows the clearest difference of all. +- Benefit: Two US follow-up studies, 119,000 people, more than 300 ten-thousand person-years. A person-year is the number of people multiplied by the number of years each was followed. Compared with people who ate no nuts: people who ate less than 1 time a week had a death risk about 7% lower (risk ratio 0.93, 95% CI 0.90–0.96, the credible range). People eating 1 time a week were about 11% lower (0.89, 0.86–0.93). People eating 2–4 times a week were about 13% lower (0.87, 0.83–0.90). People eating 5–6 times a week were about 15% lower (0.85, 0.79–0.91). People eating 7 or more times a week were about 20% lower (0.80, 0.73–0.86) +- Evidence grade: A +- Sources: Bao Y, Han J, Hu FB, et al. (2013). Association of nut consumption with total and cause-specific mortality. New England Journal of Medicine, 369(21), 2001-2011. +- Notes: Buy them plain, with no added salt. Nuts are not low in calories, so do not treat them as a snack to eat freely. Note that these studies only followed a group of people over the long term, and nut eaters have better lifestyles across the board, so the figures need discounting. + +### 24. Replace some of your red meat with fish and poultry + +- Cost: No money, and it takes no time. This is swapping one kind of meat for another, not buying more. The hard part is shifting what you prefer. +- In plain language: Eating two more servings a week of processed meat or red meat raises the death rate in the same period by about 3%. Swapping the same number of servings for fish or poultry shows no rise. The effect in this section is small, which makes this an in-passing swap rather than something to fight over. +- Benefit: Pooling 6 US follow-up studies and close to 30,000 people. Two extra servings of processed meat a week raise the death risk about 3% (risk ratio 1.03, 95% CI 1.02–1.05, credible range). Two extra servings of unprocessed red meat a week also raise it about 3% (1.03, 1.01–1.05). The same amount eaten as poultry gives 0.99 (0.97–1.02), and as fish 0.99 (0.97–1.01). Neither is statistically significant +- Evidence grade: A +- Sources: Zhong VW, Van Horn L, Greenland P, et al. (2020). Associations of Processed Meat, Unprocessed Red Meat, Poultry, or Fish Intake With Incident Cardiovascular Disease and All-Cause Mortality. JAMA Internal Medicine, 180(4), 503-512. +- Notes: Two servings a week makes very little difference, so do not expect to live longer by swapping meat. When you are spending the same money on meat, you can favour fish and poultry + +### 25. Replace some of your white rice and white flour with whole grains + +- Cost: Brown rice, oats and wholemeal bread cost a little more than white rice and white flour, and take no time. The hard part is getting used to the taste. +- In plain language: People who eat 90 g more whole grains a day have about 17% lower death rates in the same period. 90 g is about 3 servings. At about 200 g a day the figure is still falling. +- Benefit: Pooling several follow-up studies, all only followed records with no control group: every 90 g more whole grains a day lowers the death risk about 17% (RR 0.83). 90 g is about 3 servings. The benefit continues to 210–225 g a day. In a separate pooled analysis, every 1 extra serving a day lowers it about 8% (RR 0.92). +- Evidence grade: A +- Sources: Aune D et al. (2016). Whole grain consumption and risk of cardiovascular disease, cancer, and all cause and cause specific mortality: systematic review and dose-response meta-analysis of prospective studies. BMJ. ; Schwingshackl L et al. (2017). Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. American Journal of Clinical Nutrition. +- Notes: You do not have to switch everything to whole grains, because half of it already puts you in the stretch where the benefit is most visible. Note that these studies only followed records, and people who eat whole grains were healthier to begin with, so the figures are too large. Results also differ a lot between studies (I² 83%, and the higher that number, the less the studies agree). + +### 26. Drink tea three or more times a week + +- Cost: A few tens to a few hundred yuan a year. Making a cup takes a few minutes and takes very little time. +- In plain language: People with a habit of drinking tea have about 15% lower death rates in the same period. Measured from 50, they live 1.26 years longer on average, and they spend 1.41 more years free of atherosclerotic cardiovascular disease. +- Benefit: A Chinese study of 100,900 people (China-PAR), half of whom were followed for the full 7.3 years. People with a habit of drinking tea had a death risk about 15% lower (risk ratio 0.85, 95% CI 0.79–0.90, the credible range). Measured from age 50: people with a habit of drinking tea spend 1.41 more years free of atherosclerotic cardiovascular disease, and their life expectancy is 1.26 years longer +- Evidence grade: A +- Sources: Wang X, Liu F, Li J, et al. (2020). Tea consumption and the risk of atherosclerotic cardiovascular disease and all-cause mortality: The China-PAR project. European Journal of Preventive Cardiology, 27(18), 1956-1963. ; mycotoxins in tea: Cui P et al. (2020). Quantitative analysis and dietary risk assessment of aflatoxins in Chinese post-fermented dark tea. Food and Chemical Toxicology. ; Zhou H et al. (2022). Mycotoxins in Tea (Camellia sinensis (L.) Kuntze): Contamination and Dietary Exposure Profiling in the Chinese Population. Toxins. ; green tea extract and liver damage: Hu J et al. (2018). The safety of green tea and green tea extract consumption in adults - Results of a systematic review. Regulatory Toxicology and Pharmacology. +- Notes:Disputed, and the reason is this: these studies only followed records, and Chinese men who drink tea also have a high rate of smoking and drinking. The authors have already subtracted part of that effect, but not all of it. One more thing: do not drink it scalding hot, and for the temperature of hot drinks see the item in this section on that. Some people worry about mycotoxins in tea. In China, 158 samples of dark tea have been tested and 2 of them came back with aflatoxin, and worked out against the amount of tea people drink, the intake is below the internationally acceptable level. Another study tested 352 samples of tea for 16 mycotoxins, and only ochratoxin A in dark tea came out above the limit on average. Worked out against how much tea people in China drink, neither study found a dietary risk. Keep tea somewhere dry, and do not drink it if it has got damp and gone mouldy. Green tea extract capsules are a different thing: a large dose of concentrated catechins swallowed at once can damage the liver, and brewed tea does not have that problem. So this is about brewed tea only, and do not substitute a green tea extract supplement for it. + +### 27. Drink three to four cups of coffee a day, with no sugar and no cream topping + +- Cost: Brewing it yourself costs a yuan or two a day. Making a cup takes a few minutes and very little of your time. +- In plain language: People who drink three to four cups of coffee a day have about 17% lower death rates in the same period than people who do not. +- Benefit: A review pooled the results of 201 summary studies on top of each other, and the original studies underneath those summaries all only followed records with no control group. Result: people drinking 3–4 cups of coffee a day had a death risk 17% lower than people who do not drink coffee (relative risk 0.83, 95% CI 0.79–0.88, the credible range) +- Evidence grade: A +- Sources: Poole R, Kennedy OJ, Roderick P, et al. (2017). Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ, 359, j5024. +- Notes:Disputed, and the reason is this: the original authors state that the evidence comes almost entirely from follow-up records, and they also write that 「需要可靠的随机对照试验才能判断是否为因果」 — reliable randomised controlled trials are needed to tell whether the relationship is causal. In plain terms, people have to be randomly split into two groups and compared, because records alone cannot establish cause. Drink it black, because sugar and a cream topping cancel the benefit out. This does not apply to pregnant women, or to people with arrhythmia, anxiety or insomnia. + +### 28. Eat 5 servings of fruit and vegetables a day (about 400 g) + +- Cost: A few yuan to a dozen-odd yuan a day. Washing and cutting takes some time each day. The hard part is buying every day and eating enough of it every day. +- In plain language: People who eat 200 g more fruit and vegetables a day have about 10% lower death rates in the same period. People who eat 5 servings a day have about 13% lower rates than people who eat only 2. Eating more than that does not lower it further. +- Benefit: Pooling several follow-up studies, all only followed records with no control group: every 200 g more fruit and vegetables a day lowers the death risk about 10% (RR 0.90). The benefit continues to 800 g a day. Two large US follow-up studies plus a pooled analysis of 26 studies: people eating 5 servings a day had a death risk about 13% lower than people eating only 2 (HR 0.87). The best combination is 2 servings of fruit and 3 of vegetables, and going beyond that does not lower the risk further. +- Evidence grade: A +- Sources: Aune D et al. (2017). Fruit and vegetable intake and the risk of cardiovascular disease, total cancer and all-cause mortality: a systematic review and dose-response meta-analysis of prospective studies. International Journal of Epidemiology. ; Wang DD et al. (2021). Fruit and Vegetable Intake and Mortality: Results From 2 Prospective Cohort Studies of US Men and Women and a Meta-Analysis of 26 Cohort Studies. Circulation. +- Notes: 5 servings is enough, there is no need to reach 10, and the two studies agree on that point. Note that these studies only followed records and the confounding is obvious: people who eat more vegetables differ from other people in income, education and exercise. So "about 10% lower risk" is already the ceiling. + +### 29. Eat less ultra-processed food (crisps, instant noodles, cakes, instant meals) + +- Cost: No money, you only have to shop around. Either you cook yourself, or you spend time picking out food that has not been much processed, and either way it takes time every day. The hard part is fighting the habit that convenience is better. +- In plain language: People who eat the most ultra-processed food have about 20% higher death rates in the same period than people who eat the least, and cardiovascular death is 50% higher. That evidence is graded not very certain, so read the figures with a discount. +- Benefit: A review pooled several meta-analyses on top of each other, and the original studies underneath them all only followed records with no control group. Result: people who eat a lot of ultra-processed food have a death risk about 21% higher than people who eat little of it (RR 1.21), and cardiovascular death about 50% higher (RR 1.50). The evidence grades for these two are "strong indication" and "convincing" respectively. But judged by the standard evidence grading system (GRADE), the certainty is low or very low. +- Evidence grade: A +- Sources: Lane MM et al. (2024). Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. +- Notes:Disputed on three points. First, this classification (NOVA) lumps together foods that differ a lot in nutrition. Second, it overlaps heavily with sugary drinks and processed meat (items 7 and 19), so it is not possible to tell which of them is doing the work. Third, judged by the standard evidence grading system (GRADE), the grade is low as well. So once you have done those two items, how much extra this one adds is anyone's guess. The other side's evidence is the low rating in that same review, because no original study has yet reached the opposite conclusion. So do items 7 and 19 first (stop drinking sugary drinks, eat less processed meat), then come back to this one. + +### 30. Cook and heat with electricity or gas instead of coal and firewood + +- Cost: A rural household has to replace the stove and add the fuel bill, which comes to a few hundred yuan a year and sometimes over a thousand. It is a one-off change and does not take time in the long run. City dwellers are basically already on electricity or gas. +- In plain language: People who cook with coal or firewood have about 10% higher death rates in the same period; people who burn coal and wood for heating are about 14% higher. People who have already switched to electricity or gas have a death rate 13%–33% lower than those still burning coal and wood. +- Benefit: A Chinese study that followed people with chronic disease over the long term, only followed records with no control group, covering 271,000 adults with no cardiovascular disease. People who used solid fuels such as coal and firewood for cooking had a death risk about 11% higher than people who used electricity or gas (HR 1.11). People who used solid fuels for heating were about 14% higher (HR 1.14). Among people who had already switched from solid fuel to clean fuel, compared with those still burning, cooking was about 13% lower (HR 0.87) and heating about 33% lower (HR 0.67). Outdoor PM2.5 is a separate account. Pooling 104 follow-up studies, every 10 µg/m³ of long-term exposure raises the risk of natural-cause death about 8% (RR 1.08). Natural causes means deaths that were not accidents. +- Evidence grade: A +- Sources: Yu K et al. (2018). Association of Solid Fuel Use With Risk of Cardiovascular and All-Cause Mortality in Rural China. JAMA. ; Chen J, Hoek G (2020). Long-term exposure to PM and all-cause and cause-specific mortality: A systematic review and meta-analysis. Environment International. +- Notes: These studies only followed records. Households that switched fuel are often wealthier as well, so the figure of a risk ratio of 0.67 has the family's income mixed into it. As for outdoor PM2.5, what one person can do is limited, and it comes down to moving house, wearing a mask and using a purifier. There is no research with mortality as the endpoint for air purifiers, so this section gives no figure for them. + +### 31. Let hot things cool before you drink them, and skip tea, soup and coffee that are scalding + +- Cost: No money, just two or three minutes of waiting each time. The hard part is holding back when you want a drink. +- In plain language: Drinking tea that is very hot gives you 8 times the chance of oesophageal cancer that drinking warm tea does. Even at just "hot" it is 2 times. Drinking it less than two minutes after it is poured carries 5 times the risk of waiting more than four minutes. +- Benefit: In a high-incidence area of oesophageal cancer in northern Iran, 300 patients were compared with 571 people without the disease. Everything below is compared with people who drink warm tea. People who drink "hot" tea had about 2.07 times the odds of squamous oesophageal cancer (odds ratio, 95% CI 1.28–3.35, credible range). People who drink "very hot" tea had about 8.16 times (3.93–16.9). People who drink it less than 2 minutes after pouring had about 5.41 times the odds of people who wait more than 4 minutes (2.63–11.1) +- Evidence grade: A +- Sources: Islami F, Pourshams A, Nasrollahzadeh D, et al. (2009). Tea drinking habits and oesophageal cancer in a high risk area in northern Iran: population based case-control study. BMJ, 338, b929. ; Loomis D, Guyton KZ, Grosse Y, et al. (2016). Carcinogenicity of drinking coffee, mate, and very hot beverages. Lancet Oncology, 17(7), 877-878. +- Notes: The International Agency for Research on Cancer classifies hot drinks above 65 ℃ as group 2A, which means probably carcinogenic. The same assessment says coffee itself is not carcinogenic. In China, oesophageal cancer is common in the Chaoshan area of Guangdong and around the Taihang Mountains, and that is linked to the habit of drinking while it is still scalding + +### 32. Get out in the daylight, and do not avoid the sun all day + +- Cost: No money. A dozen minutes a day, on the way to work and at the lunch break, is enough. +- In plain language: Women who deliberately avoid the sun have a life expectancy 0.6 to 2.1 years shorter than the women who get the most sun. People who do not smoke but avoid the sun live about as long as the heaviest smokers in the group that gets the most sun. +- Benefit: A Swedish study of 29,500 women followed for 20 years. Compared with the group that got the most sun, people who deliberately avoided sun exposure had a life expectancy 0.6 to 2.1 years shorter. The authors write that 「不吸烟但回避晒太阳的人，预期寿命与晒太阳最多组里的吸烟者相当」 — people who do not smoke but avoid sun exposure have a life expectancy on a par with the smokers in the group that gets the most sun +- Evidence grade: B +- Sources: Lindqvist PG, Epstein E, Nielsen K, et al. (2016). Avoidance of sun exposure as a risk factor for major causes of death: a competing risk analysis of the Melanoma in Southern Sweden cohort. Journal of Internal Medicine, 280(4), 375-387. +- Notes:Disputed, and the reason is this: Sweden lies at a high latitude with little sunshine, so the conclusion cannot be carried straight over to China. And people who avoid the sun may have been frail and housebound to begin with. Also, do not go as far as getting sunburnt. You still need to cover up under the strong midday sun, because sun exposure raises the risk of skin cancer. + +### 33. Keep your BMI between 20 and 25, and lose weight if you are overweight + +- Cost: No money. Diet and exercise both take time every day. This is the item in the whole section that demands the most willpower: controlling what you eat and moving your legs over the long run, and keeping the weight off afterwards is harder than losing it. +- In plain language: The death rate is lowest in the BMI 20–25 band. At a BMI of 27.5–30 it is about 20% higher, at 30–35 about 45% higher, and at 35–40 close to double. East Asian people are more sensitive to excess weight, and every 5 points of BMI raises the risk about 40%. +- Benefit: Individual participant data from 239 follow-up studies pooled together, all only followed records with no control group. The analysis kept only three kinds of people: those who had never smoked, those with no chronic illness when they entered the study, and those who were still alive 5 years later. Result: the death risk is lowest at a BMI of 20–25. At 25–27.5 it is about 7% higher (HR 1.07). At 27.5–30 about 20% higher (HR 1.20). At 30–35 about 45% higher (HR 1.45). At 35–40 about 94% higher (HR 1.94). At 40–60 it is 2.76 times (HR 2.76). Among people above a BMI of 25, every extra 5 kg/m² raises the risk in East Asian populations about 39% (HR 1.39). +- Evidence grade: A +- Sources: Global BMI Mortality Collaboration (2016). Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet. ; Flegal KM et al. (2013). Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA. (the other side of the dispute) +- Notes:Disputed. One side argues that being a little heavier means living longer. Their pooled figures are that overweight people have a death risk about 6% lower, and mildly obese people about 5% lower. The two sides differ mainly over whether to drop three kinds of data: people who smoke, people who are already ill, and the first few years after entry into a study. The reason to drop them is that people who are seriously ill start to lose weight first. The analysis of the 239 studies above dropped those three groups, and the risk from being overweight came back. Note also that no randomised trial has shown that losing weight by itself lowers all-cause mortality. The figures above come from comparing people of different weights side by side, so they do not mean that losing weight lowers your risk by that much. If you want to lose weight there is no need to work on mealtimes, because breakfast and 16:8 intermittent fasting bring no extra benefit, see item 26 in section 6 (do not expect breakfast or 16:8 to help you control your weight) +### 34. Eat chillies four or more times a week + +- Cost: No money, and it takes no time. +- In plain language: People who eat chillies four or more times a week have about 23% lower death rates in the same period than people who rarely eat them, and cardiovascular death is about a third lower. +- Benefit: An Italian study of 22,800 people, half of whom were followed for the full 8.2 years, with 1,236 deaths in total. People who ate chillies 4 or more times a week had a death risk about 23% lower than people who rarely ate them (risk ratio 0.77, 95% CI 0.66–0.90, credible range). Cardiovascular death was about 34% lower (0.66, 0.50–0.86) +- Evidence grade: B +- Sources: Bonaccio M, Di Castelnuovo A, Costanzo S, et al. (2019). Chili Pepper Consumption and Mortality in Italian Adults. Journal of the American College of Cardiology, 74(25), 3139-3149. +- Notes:Disputed, and the reason is this: this is a follow-up record study done in the context of a Mediterranean diet, and people who eat hot food may have an overall diet closer to the local tradition. Note too that people with reflux, haemorrhoids or irritable bowel will find that hot food makes the symptoms worse, so there is no need to force yourself. + +### 35. Have one or two servings of milk or yoghurt a day + +- Cost: A few yuan a day, and it takes no time. +- In plain language: People who drink 2 or more servings of milk or yoghurt a day have about 17% lower death rates in the same period than people who do not. +- Benefit: The PURE follow-up study covers 21 countries and 136,000 people, with an average follow-up of 9.1 years and 6,796 deaths. People who ate two or more servings of dairy a day had a death risk about 17% lower than people who ate none (risk ratio 0.83, 95% CI 0.72–0.96, credible range), trend P=0.0052 +- Evidence grade: B +- Sources: Dehghan M, Mente A, Rangarajan S, et al. (2018). Association of dairy intake with cardiovascular disease and mortality in 21 countries from five continents (PURE): a prospective cohort study. Lancet, 392(10161), 2288-2297. +- Notes:Disputed, and the reason is this: PURE included a large number of middle-income and low-income countries, and in those countries being able to drink milk every day already says the family has some money, and that confounding is hard to remove cleanly. People who cannot digest lactose can switch to yoghurt or low-lactose milk instead, and there is no need to force themselves to drink ordinary milk. + +### 36. You do not need to give up eggs, but do not eat three or four a day + +- Cost: No money, and it takes no time. +- In plain language: Every extra half a whole egg a day raises the death rate in the same period by about 7%. Swap that half for egg white, fish, poultry or nuts and the death risk for each of those is lower instead. There is no need to give up eggs, only not to eat three or four a day. +- Benefit: A US follow-up study of 521,000 people, with 129,000 deaths. Every extra half a whole egg a day raised the death risk about 7% (risk ratio 1.07, 95% CI 1.06–1.08, the credible range). Swapping that half a whole egg for the same amount of egg white, poultry, fish, dairy or nuts and beans gives a lower death risk, and the death risk is also lower for cardiovascular disease, cancer and respiratory disease +- Evidence grade: B +- Sources: Zhuang P, Wu F, Mao L, et al. (2021). Egg and cholesterol consumption and mortality from cardiovascular and different causes in the United States: A population-based cohort study. PLoS Medicine, 18(2), e1003508. +- Notes:Disputed, and the dispute is a large one: there are also long-term follow-up studies of large groups of people, and pooled analyses of several studies, that found no link between eggs and mortality. The US dietary guidelines have already dropped the daily limit on cholesterol. What needs avoiding is only three or four eggs a day, and one a day is nothing to worry about. + +### 37. Take a bath if you can, and do not only use a shower + +- Cost: A bit more on water and electricity. Half an hour each time. +- In plain language: People who bathe almost every day have a cardiovascular event risk about 28% lower than people who bathe fewer than twice a week, and brain haemorrhage about 46% lower. +- Benefit: A Japanese study of 30,000 people followed for 19 years, with 2,097 cardiovascular events. People who bathed almost every day had a cardiovascular event risk about 28% lower than people who bathed no more than twice a week (risk ratio 0.72, 95% CI 0.62–0.84, the credible range). Coronary heart disease was about 35% lower (0.65, 0.45–0.94). Stroke was about 26% lower (0.74, 0.62–0.87). Brain haemorrhage was about 46% lower (0.54, 0.40–0.73) +- Evidence grade: B +- Sources: Ukai T, Iso H, Yamagishi K, et al. (2020). Habitual tub bathing and risks of incident coronary heart disease and stroke. Heart, 106(10), 732-737. +- Notes:Disputed, and the reason is this: the result comes from a Japanese population with a bathing culture, and most families in China have no way to bathe every day. Also, do not make the water too hot and do not stay in too long. For older people and people with cardiovascular or cerebrovascular disease, water that is too hot and a bath that is too long are dangerous in themselves, and every year a large number of older people in Japan die suddenly in the bath. + +### 38. Keep the nap under half an hour, do not go past an hour, and if you need one or two hours to get through the day, find out why + +- Cost: No money. Just set an alarm before you lie down, and it takes almost no willpower. +- In plain language: People who nap for no more than an hour have no higher risk of death or chronic illness, and their minds are a little sharper. People who sleep past an hour have a coronary heart disease risk about 30% higher, and diabetes and obesity about 20% higher. So set an alarm and wake yourself at half an hour. If you genuinely need one or two hours of napping to get through the day, that is a signal to have your body checked. +- Benefit: An umbrella review pooled 16 meta-analyses covering 244 health outcomes. An umbrella review is a summary of summaries. The conclusion is that people who nap for less than 60 minutes have no higher all-cause mortality or chronic illness risk, and the improvement in thinking is clearest (SMD 0.69, 95% CI 0.37–1.00, the credible range). A short nap of 20 to 30 minutes gives the biggest improvement in athletic performance (SMD 0.99, 0.67–1.31). People who nap for more than 60 minutes have a coronary heart disease risk about 30% higher and a diabetes and obesity risk about 20% higher. A separate study had 1,338 people over 56 wear wristbands to measure naps directly instead of filling in a questionnaire. The longest follow-up was 19 years, and 926 people died during it. Every 1 hour more of napping raises the death risk about 13% (HR 1.13, 95% CI 1.04–1.23). One extra nap a day adds about 7% (HR 1.07, 1.02–1.13). People who nap in the morning have a risk about 30% higher than people who nap in the early afternoon (HR 1.30, 1.03–1.64). +- Evidence grade: B +- Sources: Du P, Li J, Hua Z, et al. (2026). Multiple Health Outcomes of Daytime Napping: A Comprehensive Umbrella Review. Public Health Reviews. ; Gao C, Cai R, Zheng X, et al. (2026). Objectively Measured Daytime Napping Patterns and All-Cause Mortality in Older Adults. JAMA Network Open. ; Dashti HS et al. (2021). Genetic determinants of daytime napping and effects on cardiometabolic health. Nature Communications. +- Notes:Disputed. The two studies above both only followed records, with no control group. People who nap for a long time are more likely to have sleep apnoea, anaemia, an underactive thyroid, depression or some other chronic illness to begin with. So how much of the higher death rate among long sleepers is caused by the nap itself is not clear. Causal inference using genes (Mendelian randomisation, 453,000 plus 541,000 people) supports only one thing: the more often you nap, the slightly higher your blood pressure and waist. It did not find a causal effect of napping on coronary heart disease or diabetes. So do not treat "nap less" as a way to lower your death rate. The "find out why" half of this item is the author's advice and is grade C: if you are sleepy in the daytime to the point where you must sleep for a long time, check your night-time sleep first, and check for sleep apnoea if you snore or stop breathing during sleep, then check a full blood count and your thyroid. How to get more out of a short nap is in item 11 in section 3 (sleep 10 minutes when you feel sleepy in the afternoon). How long to sleep at night is in item 13 in this section (about 7 hours a night). + +### 39. Make up for a night up late the next night, do not save it up for the weekend + +- Cost: No money. The hour or two you make up is time you wanted to sleep anyway. +- In plain language: People who did not make up their sleep the next night after a short night have a death risk about 15% higher than people with a regular routine, and those who cut their sleep a lot and did not make it up are about 42% higher. People who did make it up show no rise in risk. So after a night up late, go to bed early the next night. +- Benefit: UK Biobank, 85,618 people, average age 61.8, with wristbands recording sleep day by day and a median follow-up of 8 years. The study first worked out each person's own sleep need, and called a night on which they slept at least 2.5 hours less than that a "short night". The first night after a short night is a "make-up night", and sleeping more than they needed that night counts as having made it up. That gives five patterns. Compared with people who kept a regular routine, people who cut their sleep and did not make it up had an all-cause death risk about 15% higher (HR 1.15, 95% CI 1.01–1.31, the credible range). People who cut their sleep by more than 3.5 hours in total and did not make it up were about 42% higher (HR 1.42, 1.24–1.63). Neither of the two groups that made it up showed a significant association, and the figure for making up after a short night was HR 1.12 (0.98–1.28). Among people who already slept short, cutting sleep without making it up was about 19% higher (1.19, 1.01–1.40), and cutting it a lot without making it up was about 38% higher (1.38, 1.17–1.63). The same result came out of a separate US health survey of 4,586 people. +- Evidence grade: B +- Sources: Li X, Zhang M, Li Z, et al. (2026). Acute sleep rebound following sleep restriction is associated with reduced mortality risk. Nature Communications. +- Notes: This study only followed records, with no control group. People who cannot make up their sleep may have been busier and in worse health to begin with, and that layer cannot be removed cleanly. "Neither of the two groups that made it up showed a rise in risk" only means no difference was found: their point estimates are still above 1, and making it up does not mean nothing happened. So pay back a night up late as soon as you can, and do not use that as a licence to stay up late. The make-up in this study means sleeping a little extra on the first night after a short night, and it does not include paying everything back in one go at the weekend. Turning "stay up late on weekdays, catch up at the weekend" into a weekly habit is called social jet lag, and it is itself linked to cardiovascular disease, see item 13 in this section (about 7 hours a night and a fixed routine). The rule about a fixed wake-up time at the weekend as well is in item 2 in section 3 (fixed wake-up time, at the weekend too), and it does not clash with this item: this one tells you to go to bed early that night, not to sleep until midday the next day. + +### 40. The longer you work night shifts, the higher your cardiovascular risk, so change to a day post as early as you can + +- Cost: Moving post or changing job may mean less money, and you lose the night shift allowance. Job hunting takes time and the decision is hard. If you do this arithmetic while you are choosing a job, the cost is zero. +- In plain language: People who work night shifts have about 13% more cardiovascular disease than people who do not, and about 27% more cardiovascular death. And it grows with the years: every extra 5 years of night shifts adds about 7% more cardiovascular disease. So this is an argument counted in years, and the earlier you change post the better. As for night shifts causing cancer, the evidence is much weaker than the stories say. +- Benefit: Pooling 23 follow-up studies, all only followed records with no control group. People who work night shifts have a cardiovascular event risk about 13% higher (RR 1.13, 95% CI 1.10–1.16, the credible range). Cardiovascular death is about 27% higher (RR 1.27, 1.18–1.36). Counted in years: every extra 5 years of night shifts raises the incidence of cardiovascular disease about 7% more (RR 1.07, 1.04–1.09), and cardiovascular death about 5% more (RR 1.05, 1.03–1.06). By disease, the incidence of coronary heart disease is about 22% higher (1.22, 1.16–1.28), death from coronary heart disease about 22% higher (1.22, 1.10–1.36), death from ischaemic heart disease about 39% higher (1.39, 1.06–1.84), and death from stroke about 49% higher (1.49, 1.04–2.12). The stroke incidence figure is not significant (1.06, 0.95–1.18). +- Evidence grade: A +- Sources: Xi J, Ma W, Tao Y, et al. (2025). Association between night shift work and cardiovascular disease: a systematic review and dose-response meta-analysis. Frontiers in Public Health. ; Esposito G, Bravi F, Santucci C, et al. (2025). Night shift work and breast cancer risk in healthcare workers: a systematic review and meta-analysis. Occupational Medicine. ; Shen QM, Li ZY, Tan YT, et al. (2026). Night shift work and risk of total and site-specific cancer: results from a prospective cohort study among Chinese men. Scandinavian Journal of Work, Environment & Health. ; Czeisler CA, Johnson MP, Duffy JF, et al. (1990). Exposure to bright light and darkness to treat physiologic maladaptation to night work. New England Journal of Medicine, 322(18), 1253-1259. +- Notes: These are all studies that only followed records, with no control group. People who work night shifts already differ from others in smoking, weight and income, and that cannot be fully adjusted away. On night shifts causing cancer, the evidence is much weaker than the stories say. In a pooled analysis of 12 studies and 12,132 cases of breast cancer, people who have worked night shifts have a risk only about 5% higher, and the credible range crosses 1 (RR 1.05, 0.96–1.14). The group with a full 20 years is about 25% higher (1.25, 1.01–1.55), but the authors' sensitivity analysis makes that one not significant, and after correcting for possible publication bias it comes out close to 1, in their words the association is 「远未确立」 — far from established. A separate study that followed Chinese men for 16.1 years and recorded 8,202 cancers found no association between night shifts and total cancer or most common cancers, with pancreatic cancer the only exception at about 59% higher in the group with 11 to 20 cumulative years (HR 1.59, 1.09–2.31). So this item does the arithmetic on the cardiovascular side, not on cancer. Not everyone should leave a night shift post, and the calculation in years is one you have to do for yourself. If you are already on nights, change the things you can change: for quitting smoking see item 1 in this section (quit smoking, the earlier the better), for blood pressure and blood lipids see item 12 in this section (take the medicine as prescribed if you have high blood pressure or high blood lipids), and for how to make up the sleep after a night shift see item 39 in this section (make up for a night up late the next night). There are two more countermeasures worth trying: concentrating your eating in the daytime, and moving your body clock across with strong light plus strict darkness during the day. The second is the only one shown to actually move the clock. Under ordinary indoor lighting, six night shifts in a row leave the clock completely unmoved. Switch to strong light of 7000 to 12000 lux plus near-total darkness during the day, and four days will turn it fully around. **The light at night has to be bright enough, the sleep during the day has to be properly dark, and you have to wear sunglasses on the way home to block the morning light. All three are required**, and leaving out the last one makes the effort pointless. Both countermeasures only reach metabolic measures and sleep; no trial has tested whether doing them lowers your cardiovascular risk. Spending your days off living a normal daytime routine pushes the clock back. For the doses, the original figures and the rest of the sources, and for how the body tells the time, see [docs/生物钟和夜班.md](../docs/生物钟和夜班.md). + +### 41. Buy prepackaged cooking oil with an SC number, not loose oil pressed at home in a small workshop + +- Cost: The same money, just a different shop. All you have to do is check whether the package carries a food production licence number beginning with SC. The hard part is persuading older relatives who are used to buying home-pressed oil. +- In plain language: Peanuts that go damp and mouldy grow aflatoxin, which is a proven carcinogen and does most of its damage to the liver. After Guangzhou brought small workshops under control, the share of residents who often eat home-pressed oil with abnormal liver function fell by about 35%. In Guangxi, pregnant women who ate home-pressed peanut oil had a risk about nine tenths higher of having a baby with a low birth weight, and about four fifths higher of giving birth early. +- Benefit: Aflatoxin is a group 1 carcinogen designated by the International Agency for Research on Cancer, meaning it certainly causes cancer, and it mainly causes liver cancer. A study in Guangzhou measured aflatoxin B1 in home-pressed peanut oil from small workshops between 2010 and 2022. The median was 1.29 μg/kg, and half of the samples fell between 0.12 and 6.58. After the city brought management rules for small workshops into force, the toxin in the oil fell immediately by 2.865 μg/kg, and then by a further 2.593 each year. The share of residents in the area who often eat home-pressed oil with abnormal liver function fell accordingly (PR 0.650, 95% CI 0.469–0.902, credible range), by about 35%. A study in Guangxi that followed 1,611 pregnant women found that 81.7% ate home-pressed peanut oil. Those who ate it had a risk about nine tenths higher of a baby with a low birth weight (aOR 1.9, 1.1–3.2), and about four fifths higher of preterm birth (aOR 1.8, 1.1–3.0). +- Evidence grade: B +- Sources: Lei J, Li Y, Wang Y, et al. (2024). The impact of small food workshops management regulations on aflatoxin B1 in home-made peanut oil and the liver function of high-consumption area residents: an interrupted time series study in Guangzhou, China. Frontiers in Public Health, 12. ; Zhong Y, Lu H, Jiang Y, et al. (2024). Effect of homemade peanut oil consumption during pregnancy on low birth weight and preterm birth outcomes: a cohort study in Southwestern China. Global Health Action, 17. ; IARC (2012). Chemical Agents and Related Occupations. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol 100F. ; National Health and Family Planning Commission (国家卫生计生委), State Food and Drug Administration (国家食品药品监督管理总局) (2017). National Food Safety Standard, Limits for Mycotoxins in Food (GB 2761-2017) +- Notes: Both studies only followed records and had no control group. The endpoint in the Guangzhou study is a liver function test, not liver cancer or death. The median toxin level in home-pressed oil is in fact below the national limit for peanut oil of 20 μg/kg, and the risk comes from eating it over many years and from occasional high batches. The national limit applies to all peanut oil, but prepackaged oil has to pass production licensing and a factory release test, while small workshops are governed by the management rules of each province and are sampled far less often. The people who benefit most from this item are you and the family you eat with. + +### 42. Cook with vegetable oil instead of lard and butter, you do not have to shop around between oils for health reasons, and do not expect flaxseed oil to protect your heart + +- Cost: It does not cost much more. Vegetable oil is already the main cooking fat in most households. The hard part is dropping the habit of lard for frying and for putting over rice. +- In plain language: Cutting back on saturated fats such as lard and butter and switching to vegetable oil cuts heart disease, stroke and events like them by about 17%, but shows no change in overall death risk. There is no evidence that soybean oil, corn oil and rapeseed oil differ from each other. Eating more flaxseed oil showed no drop in heart disease risk either. +- Benefit: Cochrane pooled 15 randomised trials and 56,675 people, each lasting at least two years. Cutting back on saturated fat lowered cardiovascular events by about 17% (RR 0.83, 0.70–0.98). All-cause mortality was almost unchanged (RR 0.96, 0.90–1.03), and so was cardiovascular death (RR 0.95, 0.80–1.12). Switching to polyunsaturated fat, the main component of soybean oil and corn oil, and switching to carbohydrate made no difference. For switching to monounsaturated fat, the main component of rapeseed oil and olive oil, there is too little data to say anything. A second Cochrane review pooled 19 randomised trials and 6,461 people: eating more omega-6, which is linoleic acid, gave an all-cause mortality of RR 1.00 (0.88–1.12) and a cardiovascular event risk of RR 0.97 (0.81–1.15), showing neither benefit nor harm. A third review pooled 86 randomised trials and 162,796 people: eating more plant-source omega-3, the main component of flaxseed oil and perilla oil, gave an all-cause mortality of RR 1.01 (0.84–1.20) and a coronary heart disease event risk of RR 1.00 (0.82–1.22), which is no measurable effect either way. +- Evidence grade: A +- Sources: Hooper L, Martin N, Jimoh OF, et al. (2020). Reduction in saturated fat for cardiovascular disease. Cochrane Database of Systematic Reviews, (5), CD011737. ; Hooper L, Al-Khudairy L, Abdelhamid AS, et al. (2018). Omega-6 fats for the primary and secondary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews, (11), CD011094. ; Abdelhamid AS, Brown TJ, Brainard JS, et al. (2020). Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews, (3), CD003177. ; Chinese Nutrition Society (中国营养学会) (2022). Dietary Guidelines for Chinese Residents (2022). People's Medical Publishing House +- Notes:Disputed: some people argue that the amount of omega-6 in soybean oil and corn oil promotes inflammation, so you should eat less and switch oils. The randomised trials in the omega-6 review above found no harm, and the authors rated the quality of that evidence low themselves. The benefit is graded "medium" on the basis of the 17% drop in cardiovascular events, since all-cause mortality is unchanged. Whatever oil you use, control the quantity: the dietary guidelines recommend 25 to 30 grams of cooking oil per person per day. diff --git "a/en/docs/\345\201\232\345\271\263\345\217\260\350\246\201\345\212\236\345\223\252\344\272\233\350\257\201.md" "b/en/docs/\345\201\232\345\271\263\345\217\260\350\246\201\345\212\236\345\223\252\344\272\233\350\257\201.md" new file mode 100644 index 00000000..380c53f3 --- /dev/null +++ "b/en/docs/\345\201\232\345\271\263\345\217\260\350\246\201\345\212\236\345\223\252\344\272\233\350\257\201.md" @@ -0,0 +1,71 @@ +# Which licences a platform needs: a comparison table and a decision table for choosing a server + +This long-form essay goes with section 26 of the README. It holds three tables and a few notes on the points that are easiest to get wrong. The items themselves and the sources are in the README. How to register a company and how to file tax, see section 12 (starting a business). The lines that technical staff employed by others should not cross, see section 11 (programmers and tech workers). + +## 1. First work out which kind of business you are in + +A single site often touches on several kinds of business at once. Touch on several kinds and you need several licences. Picking one and going no further does not work. + +| What you are doing | The business category it falls under | What you need | The main legal basis | +|---|---|---|---| +| An information site that takes no money, a personal blog, a company website | Non-operating internet information services | An ICP filing. It is not a licence, it is telling the authority you are about to open a site | Article 4 of the Measures for the Administration of Internet Information Services | +| Membership, value-added services and paid content that charge users | Operating internet information services | A Value-Added Telecommunications Business Operating Licence (information services business). What it covers is your charging users | Articles 3, 4 and 7 of the same Measures | +| Introducing buyers to sellers and handling transactions and orders | Online data processing and transaction processing services | A Value-Added Telecommunications Business Operating Licence (B21). What it covers is your handling transactions on behalf of both sides | B21 of the Catalogue of Telecommunications Business Categories (2015 edition) | +| Live streaming with hosts on camera, game live streaming | Online performance | A Network Culture Business Licence whose scope includes online performance. Without it your site cannot put hosts on camera | Article 4 of the Measures for the Administration of Online Performance Business Operations | +| Making your own video programmes, or collecting programmes from elsewhere, or letting users upload video to your site | Internet audio-visual programme services | a Licence for the Distribution of Audio-Visual Programmes through Information Networks. What it covers is broadcasting video programmes on the site | Articles 7 and 8 of the Provisions on the Administration of Internet Audio-Visual Programme Services | +| Selling goods in a live stream | Online live-stream marketing | The licences in the rows above. On top of that you have to verify the merchants and keep records | Article 8 of the Measures for the Administration of Online Live-Stream Marketing (Trial) | +| Running news and information | Internet news information services | an Internet News Information Service Licence. Without it your site cannot publish news | Article 5 of the Provisions on the Administration of Internet Live-Streaming Services | +| Building your own server room and selling machines and bandwidth | Internet data centre services, internet access services | A Value-Added Telecommunications Business Operating Licence (B11, B14). What it covers is your selling server space and bandwidth to other people | B11 and B14 of the Catalogue of Telecommunications Business Categories (2015 edition) | + +Which licence goes with which kind of live streaming is set out most plainly in the guiding opinions of the seven departments in 2021: 「开展经营性网络表演活动的直播平台须持有《网络文化经营许可证》并进行ICP备案；开展网络视听节目服务的直播平台须持有《信息网络传播视听节目许可证》（或在全国网络视听平台信息登记管理系统中完成登记）并进行ICP备案；开展互联网新闻信息服务的直播平台须持有《互联网新闻信息服务许可证》。」, which means a live-streaming platform carrying out operating online performance activities must hold a Network Culture Business Licence and complete an ICP filing; a live-streaming platform carrying out internet audio-visual programme services must hold a Licence for the Distribution of Audio-Visual Programmes through Information Networks (or complete registration in the national audio-visual platform information registration system) and complete an ICP filing; a live-streaming platform carrying out internet news information services must hold an Internet News Information Service Licence. + +There are three kinds of live streaming. If hosts perform on it, you get a Network Culture Business Licence. If it carries audio-visual programmes, you get a Licence for the Distribution of Audio-Visual Programmes through Information Networks, or you complete registration in the national audio-visual platform information registration system. If it carries news, you get an Internet News Information Service Licence. The first two also both have to do the ICP filing. + +### Three points that are easy to get wrong + +**An individual cannot get a value-added telecommunications licence.** The first of the conditions says 「经营者为依法设立的公司」, which means the operator is a company duly established. You need a company first, and an individual does not get one by turning up with an identity card. If you operate only inside one province, the registered capital of the company cannot be below 1,000,000 yuan. If you operate across provinces, it cannot be below 10,000,000 yuan. After the papers go in, the review period is 60 days, and the licence is valid for 5 years. If you want to run a business that charges people, you have to set the company up first. How to set up a company, see section 12 (starting a business). + +**A private company basically cannot get the audio-visual programme licence.** The condition says 「具备法人资格，为国有独资或国有控股单位」, which means it has the status of a legal person and is a wholly state-owned or state-controlled entity. In other words this licence only goes to units the state puts money into or controls. So an individual founder making long videos or producing their own programmes does not get it. If you want to do live streaming, the licence to get is the Network Culture Business Licence. + +**No official document says in so many words that "an e-commerce platform must get an EDI licence".** EDI is the licence in row B21 of the table above, and its formal name is online data processing and transaction processing services. The handling guide from the Ministry of Industry and Information Technology says only one thing: 「按照业务界定申请相应的电信业务经营许可」, which means apply for the telecommunications business operating licence that corresponds to how the business is defined. The ministry has also given separate replies twice. A ride-hailing platform only has to file the website, and a platform for rights and interests or for bulk commodity trading only has to file the website as well. So this essay only repeats the definition of B21. Whether your line of business needs it is for you and the local communications administration to decide. Before you start the paperwork, telephone the local communications administration once and ask. + +## 2. The platform's own day-to-day duties + +Holding a licence only means you are allowed to open up. What follows is what you have to do every day once you have opened up. How much you can be fined for failing to do it is written up in the items in section 26. + +| Duty | The hard requirement | Where it is written | +|---|---|---| +| Check and register the merchants selling on your platform | Re-check and update at least once every six months | Article 24 of the Measures for the Supervision of Online Transactions | +| Report the merchants' identity information | Report it to the market regulation authority in January and July each year | Article 25 of the same Measures | +| Report the tax-related information | Report it to the tax authority within the month after the end of each quarter | Article 4 of the Provisions on the Reporting of Tax-Related Information by Internet Platform Enterprises | +| Keep records of transactions | No less than three years from the day the transaction completed | Article 31 of the E-Commerce Law | +| Keep records of live-stream content and logs | Sixty days | Article 16 of the Provisions on the Administration of Internet Live-Streaming Services | +| Keep recordings of online performances | No less than sixty days | Article 13 of the Measures for the Administration of Online Performance Business Operations | +| Keep network logs | No less than six months | Article 23, item 3, of the Cybersecurity Law | +| Handle notices of infringement | Pass the merchant's statement to the complainant. If fifteen days pass with nothing further, restore it | Article 43 of the E-Commerce Law | +| Provide a complaints and reports entry point | Put it in a prominent place and make it easy to use | Article 16 of the Provisions on the Governance of the Ecological Environment of Network Information Content | + +There are four retention periods and they are counted separately. Transaction records for three years, live-stream content for sixty days, network logs for six months. A merchant's identity information on the platform is kept for three years from the day they leave the platform. When you design the storage plan, design it for the longest of these, not the shortest. + +## 3. Choosing a server: how to pick between the three tiers + +Answer these questions first, and only then compare prices. + +| Question | If the answer is | Then | +|---|---|---| +| Could you stand the site being down for a day | Yes | The cheapest VPS (virtual server) is enough | +| Do you have users signing up, trading and uploading | Yes | Use a cloud host from one of the mainstream cloud providers. Pick one where you can back up the whole machine at any time and add configuration temporarily | +| Is there a named person looking after the server | No | Do not put the whole machine in a colocation rack | +| Is bandwidth or hardware your largest running cost | Yes, and there is a named person looking after it | Only then consider putting the whole machine in a colocation rack | + +**A small provider is not unusable, but check it first.** Putting machines in a colocation rack and giving other people access to the internet are two things that need a licence in themselves. They belong to value-added telecommunications business. To check, open the telecommunications business market information system of the Ministry of Industry and Information Technology at tsm.miit.gov.cn and look the company up once by its full name. If no licence comes up, rule the provider out. The risk in the ones that are half the price usually comes in three forms: the machines have been oversold, the owner disappears, or the upstream is cut off. If something does go wrong and the provider is licensed, you can still complain to the communications administration. If it is not licensed, you do not even know who to appeal to. + +**In China or outside it.** Put the server in China and you have to file. An access provider is not allowed to give access to a site that has not filed. Putting it outside China does let you get round the filing. But your users are in China, and the money is taken in China. None of the platform duties set out in items 5 to 10 in section 26 drops away. Putting the server outside China also adds one more layer of cost, which is data export. Sending the personal information of users in China to a machine outside China is what is called an export. To export, one of the four conditions listed in Article 38 of the Personal Information Protection Law has to be met. You also have to get the consent of the person separately, and it cannot be buried in a bundle of terms for someone to click through. How many people are involved is counted from 1 January of that year onwards. Below 100,000 people, none of the three routes below is needed. From 100,000 to 1,000,000 people, either sign the standard contract or go for the certification. Above 1,000,000 people, you have to file a security assessment. + +**Backups.** Keep backups in at least two places, and do not put both copies with the same provider in the same region. There is no basis for this in any regulation. It comes from experience. + +## 4. The limits of this material + +- For every provision mentioned above, the sources in section 26 of this book's README are what count. The document numbers, the article numbers and the links are all there. +- This kind of regulation changes fast. The content of this section was checked in September 2026. Before you quote it, open the original page yourself once more. Two things in particular are worth looking at. The article numbers in the Cybersecurity Law were adjusted with effect from 1 January 2026. The rules on minors tipping live streamers were changed in April 2026 into age bands. +- A few passages could not be obtained in the original, and each one is written out in the [verification records](核实记录/追加-第26节做平台.md). One is whether an e-commerce platform must in fact get an EDI licence, and whether any official document says so in so many words. The other is the judicial interpretation that treats operating network culture or audio-visual services without a licence directly as the crime of illegal business operation. diff --git "a/en/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" "b/en/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" new file mode 100644 index 00000000..7b64c1b2 --- /dev/null +++ "b/en/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" @@ -0,0 +1,89 @@ +# Household emergency kit checklist: what to buy, where to put it, how often to check it + +This is the long-form version of item 26 in section 1 of the README (fit fire extinguishers, fire blankets, escape respirators and a first-aid kit). This list says what a household should keep on hand. It does not say how to use any of it. What to do in the moment in a fire, in cardiac arrest, with major bleeding, with a scald or with carbon monoxide poisoning is all in section 13. How to choose and how to fit smoke alarms and carbon monoxide alarms, see item 3 in section 1. Gas hoses and stoves, see item 4 in section 1. + +This section works out the bill of fewer deaths and fewer injuries. It covers only harm caused by accidents, not illness. How much use the equipment itself is, the evidence is weak. The reason is in the last section, so read that paragraph before going any further. + +## 1. The official list as written + +The Ministry of Emergency Management issued a National Basic Household Emergency Supplies Storage Recommended List, with 11 items in it. The table below carries the wording used in that document. + +| No. | Item | The wording in the original | +|---|---|---| +| 1 | Drinking water | 保障每人 3 天基本饮水需求，至少 3 升/人 — enough drinking water for 3 basic days for each person, at least 3 litres per person | +| 2 | Ready-to-eat food | 保障每人 3 天基本食物需求。方便食品体积小、热量高 — enough food for 3 basic days for each person. Ready-to-eat food is small in volume and high in calories | +| 3 | Fire extinguishers and fire blankets | 用于初起火灾的扑救。灭火毯可披覆在身上逃生 — for putting out a fire in its early stage. A fire blanket can be worn over the body to escape | +| 4 | Respirators | 消防过滤式自救呼吸器，用于火灾逃生使用 — a fire-filtering self-rescue breathing apparatus, for escaping a fire | +| 5 | Torch | 防水防爆手电筒。定期充电或更换电池 — a waterproof and explosion-proof torch. Charge it at regular intervals or replace the battery | +| 6 | Multi-function knife | 有刀锯、螺丝刀、钢钳等组合功能 — it combines a saw, a screwdriver, pliers and the like | +| 7 | Radio | 接收应急广播使用 — for receiving emergency broadcasts | +| 8 | Survival whistle | 可吹出高频求救信号 — it blows a high-pitched distress signal | +| 9 | Topical medicines | 止血粉、止血贴、纱布绷带等，用于处理伤口 — haemostatic powder, adhesive wound dressings, gauze and bandages and the like, for treating wounds | +| 10 | Disinfecting wet wipes | 用于个人卫生清洁 — for personal hygiene | +| 11 | Surgical masks | 病毒防护 — protection against viruses | + +Sources: 应急管理部 (2020). 全国基础版家庭应急物资储备建议清单. + +The local version lists more detail. The Beijing Municipal Emergency Management Bureau gives households a list in two tiers, a basic version and an extended version. The basic version is split into 「应急物品、应急工具和应急药物 3 类」, which means 3 categories, emergency supplies, emergency tools and emergency medicines. The items themselves are these: a hand-crank rechargeable torch that can receive broadcasts, a survival whistle, towels, tissues or wet wipes, a respirator, multi-function combination scissors, an emergency escape rope, a fire extinguisher or a fire blanket. Medicines and medical materials are counted separately: anti-infective and cold medicines, surgical masks, medical materials such as gauze and bandages, and iodine swabs. The extended version is split into 「食品、个人用品、逃生自救工具、医疗急救用品、重要文件资料 5 类应急物资」, which means 5 categories of emergency supplies, food, personal items, escape and self-rescue tools, first-aid supplies, and important documents. The bureau also gave three recommendations. The first is 「选购资质合法、信誉良好的生产经营企业提供的应急物资」, which means buy emergency supplies from producers and sellers who are properly licensed and have a good reputation. The second is 「优先储备基础版的应急物资品种」, which means stock the items in the basic version first. The third is 「熟悉掌握应急物资的正确使用方法，定期对应急物资状况进行检查，并及时更换已过保质期的应急物资」, which means know the correct way to use emergency supplies, check the condition of emergency supplies at regular intervals, and replace anything past its shelf life promptly. + +Sources: 北京市应急管理局 (2020). 北京市修订发布居民家庭应急物资储备建议清单. ；北京市西城区人民政府转载的品种明细. + +The national level was updated once more in 2024. The document is the Guiding Opinions on Further Strengthening the System and Capacity for Ensuring Supplies for Emergency Rescue and Disaster Relief, issued by the Office of the National Committee on Disaster Prevention, Mitigation and Relief (国防减救办发〔2024〕13 号, dated 23 September 2024). Its Annex 2 is called the Guiding Catalogue for Household Emergency Supplies Storage. The catalogue is split into 5 categories: emergency food, daily items, emergency tools, emergency medicines and medical supplies, and important documents. The basic version has 16 items, the extended version 31 items. TODO (to be verified: what the 16 and the 31 items are called. The official annex is a wps file, and there is also a PDF of it scanned into images, and this time no text could be obtained that could be checked word for word). + +## 2. The three fire items + +**Fire extinguisher.** A handheld fire extinguisher is a product subject to mandatory national certification. A compliant cylinder has the CCC mark printed on it. The standard it is made to is GB 4351—2023 Handheld Fire Extinguishers. The kind that comes pre-pressurised (carbon dioxide extinguishers excepted) has a pressure gauge on the cylinder. The needle on the gauge has to sit inside the green zone. If the needle points into the red or into the yellow, this extinguisher can no longer be used. How many years before it is written off is worked out under the XF 95 standard: 6 years for water-based, 10 years for dry powder, 12 years for carbon dioxide. The period runs from the day it left the factory. The date of manufacture is stamped into the metal of the cylinder. For home use, buy ABC dry powder. Put one at the kitchen doorway and one in the entrance hall. Do not push it into the cabinet directly above the cooker. If there is a real fire, you cannot reach up there. + +**Fire blanket.** In the official list it sits in the same item as the fire extinguisher. It has two uses: covering a pan of oil that is burning, or wrapping it round yourself to get out. One in a kitchen drawer is enough. It takes up no space and it does not go out of date. If the pan catches fire, turn off the heat first, then lay the blanket over it. Once you have laid it down, do not lift it again straight away. + +**Escape respirator.** Its official name is a fire-filtering self-rescue breathing apparatus. The standard it is made to is GB 21976.7—2012 Building Fire Escape and Refuge Equipment, Part 7: Filtering Fire Self-rescue Breathing Apparatus. This is a mandatory national standard that is in force and valid. It took effect on 1 June 2013, and the review in 2023 concluded that it stays valid. When you buy one, look for this standard number, and look for the CCC mark as well. Buy as many as there are people in the house. Keep them by the head of the bed where you can reach them with your hand, and do not lock them in a storage room. The filter canister inside soaks up the toxic gas in the smoke. One use and it is finished, and it only holds for the time printed on it, so put it on and get out. TODO (to be verified: what this standard says about the minimum oxygen content it works at and the rated protection time. This time only the name of the standard, its status and its effective date were found, and the text of the provisions could not be obtained). + +Smoke is what kills most people in a fire. How to get out (stay low and crawl, feel the door before opening it, take the stairs and not the lift), see item 24 in section 13. Smoke alarms, see item 3 in section 1. Both of these come before the three items above. + +## 3. What goes in the first-aid kit + +On this item the official list says only one sentence: 「外用药品：止血粉、止血贴、纱布绷带等，用于处理伤口」, which means topical medicines such as haemostatic powder, adhesive wound dressings, gauze and bandages and the like, for treating wounds. Section 13 has already worked out which actions belong on the spot. Given those actions, the following in a household first-aid kit is reasonable. + +- Stopping bleeding: one windlass tourniquet, which is the kind with a twist rod that tightens. Add sterile gauze and bandages, a triangular bandage and medical tape. How to use a tourniquet, when not to use one, and why you should not loosen it and let the blood flow, see item 12 in section 13 +- Cleaning a wound: iodine swabs, saline, sticking plasters, sterile dressings, disposable gloves, scissors and tweezers +- Burns and scalds: no ointment at all. On the spot there is one thing to do, run cool tap water over it for 20 minutes, see item 14 in section 13 (run cool running water over a scald for 20 minutes) +- Records: prepare one card. Write down the allergies, the chronic illnesses and the regular medicines of everyone in the house. When the paramedics arrive, hand them the card + +If someone in the house already knows they have a severe allergy, keep an adrenaline auto-injector as well. It is a prescription medicine, so a doctor has to prescribe it. See item 15 in section 13 (how to recognise anaphylactic shock and how to give the injection). + +This first-aid kit list is grade C evidence. Grade C means it is written from experience and consensus, with no direct literature behind it. The reason is that the official document gives only the four characters 「外用药品」 on this item. What to buy specifically is worked backwards from the first-aid actions this book has already checked. + +## 4. Lighting, communication and calling for help + +For the torch, follow the official description and buy one that is waterproof and explosion-proof. Charge it every so often, or change the battery. The survival whistle is for sending a signal continuously when you are trapped. Blowing a whistle takes less effort than shouting with all your might, and the sound carries through a noisy place more easily. The radio is for receiving emergency broadcasts. The Beijing version of the list also lists an emergency escape rope and multi-function combination scissors. The best value in this group is a power bank for your phone, and most households already have one. + +## 5. Check once a year, ten minutes + +- Fire extinguisher: check that the needle on the pressure gauge is still in the green zone. Then look at the date of manufacture on the cylinder and work out whether it has reached its write-off age +- Smoke alarms and carbon monoxide alarms: press the test button. Change the battery once a year, see item 3 in section 1 (fit smoke alarms and carbon monoxide alarms) +- Respirators: check whether the packaging is damaged and whether they are past the shelf life printed on them +- First-aid kit: check whether any medicine or dressing has expired, and whether the clip on the tourniquet still locks +- Torch and power bank: check whether there is still charge in them + +The Beijing Municipal Emergency Management Bureau puts it this way: 「定期对应急物资状况进行检查，并及时更换已过保质期的应急物资」, which means check the condition of emergency supplies at regular intervals and replace anything past its shelf life promptly. You can fix this to the day you change the alarm batteries each year and go through all of it at once. + +## 6. Things you do not need to buy + +- **A home AED (automated external defibrillator).** One costs more than ten thousand yuan. There is also no evidence to support a household buying one of its own. In cardiac arrest the thing to do is start compressions at once, get someone to dial 120, and send someone to fetch the AED from the nearest public place, see item 1 in section 13 +- **"Fire masks" and "gas masks" with no CCC mark.** Article 24 of the Fire Protection Law says that fire products within the scope of mandatory product certification 「由具有法定资质的认证机构按照国家标准、行业标准的强制性要求认证合格后，方可生产、销售、使用」, which means may only be produced, sold and used once they have passed certification by a certification body with the legal qualifications, according to the mandatory requirements of national and industry standards. In other words, something that has not passed certification can be neither sold nor used. Also, the filter canisters used in industry are for toxic substances in a workshop, which is a different thing from smoke in a fire +- **A packaged "emergency kit".** These kits are mostly padded out with a cheap torch and a mask. Buying the fire extinguisher, the fire blanket, the respirator and the first-aid kit separately costs less, and it is easier to check the certification one item at a time +- **Food and water for more than three days.** The official list works it out at 3 days per person. Most of what you stack up beyond that ends up thrown away after it expires, see item 24 in section 5 (do not stockpile for sales promotions) + +Sources: 全国人大常委会 (2021). 中华人民共和国消防法（2021 修正，第二十四条）. + +## 7. How much use this equipment really is + +The evidence in this section is weaker than for the other items in section 1 of this book. + +One Cochrane systematic review pooled 98 studies and 2,605,044 people. It assessed whether giving households safety education and safety equipment actually helps. The result is that these interventions may bring the rate of injuries down a little, IRR 0.89 (95% CI 0.78 to 1.01), about 10% lower. But the top of that credible range crosses 1, and 1 stands for no difference, so there may be no difference at all. The ones delivered in the home did better, IRR 0.75 (95% CI 0.62 to 0.91), about 25% lower. The ones that only taught and handed out nothing, IRR 0.78 (95% CI 0.66 to 0.92), about 20% lower. What seems to do the work is teaching in the home and modifying the home, not the act of handing things out. The review also has one conclusion: there is no evidence that these interventions reduce burns and scalds or poisoning. + +In the same review, these interventions did improve people's safety habits. More households had a working smoke alarm fitted, OR 1.81 (95% CI 1.30 to 2.52), a share about 80% higher. More had worked out an escape plan in advance, OR 2.01 (1.45 to 2.77), about double. More stored their medicines properly, OR 1.53 (1.27 to 1.84), about 50% higher. More had a fire door fitted at the head of the stairs, OR 1.61 (1.19 to 2.17), about 60% higher. + +Sources: Kendrick D 等 (2012). Home safety education and provision of safety equipment for injury prevention. Cochrane Database of Systematic Reviews. + +So buying everything on the official list in one go costs little and counts as reasonable preparation. But do not think that having bought it you are safe. Research has repeatedly shown that three things work: fitting smoke alarms, working out in advance how to get out, and moving out of the way the things in the home that can hurt people. Either none of these three is on a shopping list, or it takes up only a small part of one. diff --git "a/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" "b/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" new file mode 100644 index 00000000..6275ecd2 --- /dev/null +++ "b/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" @@ -0,0 +1,149 @@ +[← Back to contents](../README.md) + +# How the body tells the time, and why night shifts hurt + +This is the long-form version of item 40 in section 2 (the longer you work night shifts, the higher your cardiovascular risk). That item only works out the arithmetic. This essay sets out the reasoning behind it. + +The question to answer is whether the body really does know what time it is. If it does, why is working night shifts not a matter of getting used to it after a few goes? + +## 1. There really is a timekeeping organ in the body + +In the hypothalamus of the human brain there is a small cluster of neurons called the suprachiasmatic nucleus, about twenty thousand cells. It is the master clock of the whole body. + +It keeps time through a set of "clock genes" that inhibit one another. Two proteins called CLOCK and BMAL1 switch on the transcription of two other genes, PER and CRY. Once the proteins made from PER and CRY have built up, they switch off the pair of proteins that switched them on. When these proteins have broken down, the cycle begins again. One turn of it takes about 24 hours. + +So this clock does not work out the time by watching the outside world. Put a person in a room with no light at all, no clock and no social cues, and the rhythm carries on regardless. It is only that the cycle is not exactly 24 hours and drifts a little each day. The 2017 Nobel Prize in Physiology or Medicine went to the three researchers who found this molecular mechanism. + +## 2. This clock sets itself by light, and the route it uses is not the one that does the seeing + +In the retina, apart from the cells that form the image, there is a separate class of cells that sense light. They contain a pigment called melanopsin, which takes no part in seeing anything clearly and only passes the strength of the light straight to the suprachiasmatic nucleus. + +So whether light can set the body clock has no direct connection with whether the eyes can see anything. In some people who are completely blind, the body clock can still be set by light. + +This is also why item 2 in section 3 (fix your wake-up time, at weekends as well) has to be paired with getting light in the morning. + +## 3. The key point is that there is more than one clock + +The liver, the gut, the heart and the fatty tissue each have their own set of peripheral clocks. + +The master clock sets itself by light. **The peripheral clocks set themselves largely by when you eat.** + +Under a normal routine, the two are in step. You see light in the day and you eat in the day, so the master clock and the peripheral clocks point at the same time. Which is why in ordinary life you never notice that they are two separate systems. + +## 4. What a night shift messes up is not the lack of sleep, it is the two clocks coming apart + +Someone on night shifts still sees light in the day, and still lives the social life of the daytime. So the master clock stays locked to the day-and-night cycle outside. + +But you eat, work and turn your metabolism over in the small hours of the night in biological terms. The peripheral clocks get pulled the other way. + +That state is called circadian misalignment. The problem is not how many fewer hours you slept. It is that the clocks across your body no longer agree with each other. + +## 5. And you do not adapt to it + +Some people think that in the long run a fixed run of night shifts has to be adaptable. + +One review pooled six studies and used the rhythm of melatonin secretion to test people on long-term fixed night shifts. The melatonin rhythm is the most reliable measure of the state of the internal body clock. + +The result is that fewer than 3% of people achieved "complete" adaptation. Fewer than a quarter reached "considerable" adaptation. + +It makes no difference how many years you have done it. The reason is in the paragraph above. In the day you still see light, and you still live the social life of the daytime, and those cues go on pulling the master clock back. + +Sources: Folkard S (2008). Do permanent night workers show circadian adjustment? A review based on the endogenous melatonin rhythm. Chronobiology International. + +## 6. How to tell that the misalignment itself does the harm, rather than the lost sleep alone + +This is the most worth-asking layer of the whole question. People on night shifts sleep less, eat badly, carry more stress and smoke more, and any of these could be the real cause. + +The good thing is that this can be produced on its own in a laboratory. + +The study by Scheer et al in 2009: ten adults lived in a laboratory for 10 days. The researchers set "one day" artificially at 28 hours, so that their eating and sleeping had to run through every hour of the body clock. **Four meals a day, with the calorie intake fixed.** + +When behaviour was about 12 hours out of step with the body clock, the results were: + +- Leptin fell 17% (leptin is the hormone that makes you feel full) +- Blood glucose rose 6%, and that was with insulin already up 22% +- The daily rhythm of cortisol turned right round +- Mean arterial pressure rose 3% +- Sleep efficiency fell 20% + +The most important one of these: of the 8 people with enough data, 3 of them had post-meal blood glucose move into the prediabetes range, and it took 10 days in all. + +Because the calorie intake was fixed and the experiment was controlled in a laboratory, these changes cannot be put down to "sleeping less" or to "eating more". **Misalignment on its own is enough to bring about these changes.** + +Sources: Scheer FA, Hilton MF, Mantzoros CS, Shea SA (2009). Adverse metabolic and cardiovascular consequences of circadian misalignment. PNAS, 106(11), 4453-4458. + +## 7. The first thing you can change: move your eating into the daytime + +Since the peripheral clocks set themselves by when you eat, there is one direction to try. Keep the night shift, but do not eat in the night. + +One randomised controlled trial did this (registration number NCT02291952). Twenty healthy people simulated night shifts and were randomly put into two groups. One group followed the common habit of night-shift workers and ate both at night and in the day. The other group ate only in the day. + +The report on the primary outcome says that in the group that ate at night, the central rhythm and the peripheral glucose rhythm were out of step, and glucose tolerance got worse. Neither of those two appeared in the group that ate only in the day. + +The cardiovascular report (a secondary analysis of the same trial) says that in the group that ate at night, pNN50, a measure of cardiac vagal modulation, fell 25.7%, RMSSD fell 14.3%, LF/HF, which reflects the balance between the sympathetic and the vagus, rose 5.5%, and PAI-1, a clotting-promoting factor, rose 23.9%. None of these changes appeared in the group that ate only in the day. + +**But this is 20 people, and it is a simulated night shift, not real night-shift workers.** + +A systematic review that looked specifically for randomised trials of this kind searched six databases and found only 4 trials. The results are not tidy either. Some trials saw fasting insulin and insulin resistance improve, and some did not. For high-density lipoprotein, none of the four trials saw a change. For post-meal blood glucose, the version that fasted for 10.75 hours overnight did come out better than the version that ate during the night and then had a midnight snack. The authors of the review say themselves that there are too few trials to apply the result widely. + +So that is all that can be said at present. **Concentrating your eating in the daytime as far as possible during night shifts is supported by a randomised trial, but the evidence is still thin.** It costs nothing and it is not hard to do, so if you want to try it, try it. There is a second thing to try, in the next section. Use light to move the whole body clock across. + +Sources: Chellappa SL 等 (2021). Daytime eating prevents internal circadian misalignment and glucose intolerance in night work. Science Advances. ；Chellappa SL 等 (2025). Daytime eating during simulated night work mitigates changes in cardiovascular risk factors: secondary analyses of a randomized controlled trial. Nature Communications. ；Wulandari F 等 (2026). Cardiometabolic status among shift workers under meal time regulation: A systematic review of randomized controlled trials. Nutrition and Health. + +## 8. Whether the clock can be moved across + +All of that so far is "misalignment does harm". Going the other way, does it work to move the whole clock across to the night shift side? + +**In theory it works, and it has been done. But the demand on light is one order of magnitude higher than most people think.** + +The study by Czeisler et al in 1990: eight young men did a week of night shifts, and two conditions were compared. + +Under the control condition, the lighting at night was ordinary indoor light, about 150 lux. The lux is the unit of illuminance, and an ordinary fluorescent-lit room is somewhere in the low hundreds. The result was that after six nights in a row, the lowest point of body temperature still fell at about half past three in the morning. The lowest point of body temperature is the standard measure of where the body clock is pointing. This says the body clock had not moved at all. + +Under the treatment condition, the night got strong light of 7000 to 12000 lux, and the day was spent asleep in a setting close to pitch dark. After four days, the lowest point of body temperature had moved to close to three in the afternoon. That is the whole thing turned round. + +Sources: Czeisler CA, Johnson MP, Duffy JF, Brown EN, Ronda JM, Kronauer RE (1990). Exposure to bright light and darkness to treat physiologic maladaptation to night work. New England Journal of Medicine, 322(18), 1253-1259. + +### All three are needed and none can be dropped + +**First, the light at night has to be bright enough.** Working under ordinary fluorescent lights is the same as being in the "nothing happened" control group above. To have any effect you need a dedicated light therapy box. + +**Second, sleeping in the day has to be strictly darkened.** Blackout curtains plus an eye mask. The experiment above did "strong light at night" and "close to pitch dark in the day" together, not one or the other. + +**Third, wear sunglasses on the way home from work.** This is the one most easily left out, and leaving it out means most of the work is wasted. + +Why the third one matters so much comes down to how light works. The same light, arriving at different times, pushes the body clock in opposite directions. Light **before** the lowest point of body temperature pushes the clock later. Light **after** it pulls the clock earlier. + +Walking home at the end of a night shift puts you just after the lowest point of body temperature. **This morning light does the opposite of what a whole night in front of a light box does.** + +### The laboratory manages it, ordinary life does not + +The people in the laboratory did not leave the house for a week. In real life your days off have to be lived as daytime lives, collecting children, being with family, running errands. Go through it once and the clock gets pulled back. That is where the figure in section 5 comes from: fewer than 3% of people on long-term fixed night shifts adapt completely. + +So in real life the aim is not usually to turn the clock all the way round. It is to move part of it and leave the clock sitting in the middle, awake enough at work and not completely wrecked on your days off. + +### How much you can actually get out of it + +A meta-analysis pooling 11 studies: light therapy gave shift workers about 32.5 minutes more total sleep time, and a sleep efficiency improvement of about 2.91. The effective dose was 900 to 6000 lux for at least 1 hour a night. + +Another laboratory study used intermittent bright light. In the treatment group, melatonin secretion overlapped the scheduled sleep period by 4.90 hours, against only 2.62 hours in the control group (P=0.002). The drop in cognitive test scores that showed up in the control group on the last night shift was held down in the treatment group. The authors say outright in their conclusion that this approach still has to be tested on real night-shift workers. + +Sources: Zhao C, Li N, Miao W, He Y, Lin Y (2025). A systematic review and meta-analysis on light therapy for sleep disorders in shift workers. Scientific Reports. ；Lammers-van der Holst HM 等 (2021). Efficacy of intermittent exposure to bright light for treating maladaptation to night work on a counterclockwise shift work rotation. Scandinavian Journal of Work, Environment & Health. + +### One thing nobody has tested + +Whether getting the clock right brings cardiovascular risk down with it. **No trial has tested it.** + +What is known is that misalignment does harm, and that came out of the experiment in section 6. That does not add up to "get it right and you are safe". And the days off pull the clock back again, and nobody has tested whether that going back and forth is good or bad either. + +The two measures above end at metabolism and cardiovascular measures on one side and at sleep on the other. Neither reaches illness or death. + +## 9. What this essay does not cover + +How much higher night shifts make cardiovascular risk, and how it is worked out by years, is in item 40 in section 2 (the longer you work night shifts, the higher your cardiovascular risk). + +How to make up the sleep after a late night is in item 39 in section 2 (after a late night, make up the sleep the next evening, do not bank it up for the weekend). + +How long you should sleep at night and how regular the schedule should be is in item 13 in section 2 (sleep about 7 hours a night, keep the schedule fixed). + +Getting light in the morning and fixing your wake-up time is in item 2 in section 3 (fix your wake-up time, at weekends as well). diff --git "a/en/docs/\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/en/docs/\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" new file mode 100644 index 00000000..d497bcfb --- /dev/null +++ "b/en/docs/\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -0,0 +1,123 @@ +# Is marriage worth it: splitting one muddled bill into five clear ones + +Many people ask one broad question: is marriage a loss or a gain. That question packs several different things together. Can the other person give you emotional support. Will they share the housework. Can they earn money. And do you have to get married to keep elderly relatives off your back. You cannot convert these into a common unit, so they cannot be answered together. This essay takes them apart. Each of the five bills uses only figures from official statistics or from a meta-analysis, which pools several studies together. Where a figure cannot be worked out, this essay says so. The last section is a list you can fill in yourself. Nothing here says whether you should get married. + +## 1. Take the question apart + +"Is marriage worth it" holds at least five separate bills. They use different units, they are worked out in different ways, and they cannot be added together. + +1. Money: first the betrothal gifts, the wedding and the marital home, which you pay for out of pocket. Second, what the law says about who owns the property. There are no official statistics on the first. The second is set out in plain provisions. +2. Time: housework, caring for family members, making allowances for the other person's work. These are hours of work that nobody pays for, and the unit is minutes per day. +3. Health: how much the death rate differs between married people and unmarried people. This bill can only show that two things occur together. It cannot show that marriage causes the gap. +4. Legal risk: how property is divided, how long it takes to get out, and which contributions can be clawed back as a one-off payment. This bill turns on what the provisions say. +5. Quality of the relationship: whether the other person gives you emotional support, and how badly the two of you fight. This bill can only be measured by the strength of the associations found in research. It does not convert into money, and it does not convert into time. + +The fifth bill is the hardest to turn into numbers, and it is the closest to what most people want to know. + +## 2. The data and sources behind each bill + +### Registration data: the marriage rate and the divorce rate are two crude rates + +The figures come from the Ministry of Civil Affairs, 2024 Statistical Bulletin on the Development of Civil Affairs. In 2024, 6,106,000 couples registered marriages in line with the law nationwide, 20.5% fewer than the year before, and the marriage rate was 4.3 per 1,000. 3,513,000 couples completed divorce formalities in line with the law, and the divorce rate was 2.5 per 1,000. Of those 3,513,000 couples, 2,622,000 registered the divorce with a civil affairs department and 891,000 had it ordered or mediated by a court. The note in the bulletin sets out how the two rates are worked out. Take the number of marriages (divorces) that year, divide it by the average total population for that year, then multiply by 1,000, which gives a per-mille rate. + +Both rates have the whole population as their denominator, which is why they are called crude rates. "Crude" means no breakdown of any kind. They answer the question of how many couples per 1,000 people registered a marriage or a divorce this year. They do not answer whether a marriage will end in divorce. Answering that needs data that follows the same group of people for a long time. This essay did not find such data in the official text. + +### The health bill: married people have a lower death rate, but that is association, not causation + +Three meta-analyses each pooled many studies together, and they give the same order of magnitude. "All-cause mortality" below means every death put together, whatever the cause. + +- Manzoli et al (2007) pooled 53 comparisons and more than 250,000 elderly people. "Not in a marriage" here covers three kinds of people: the widowed, the divorced or separated, and those who have never married. Compared with these three groups, married people had an all-cause mortality RR of 0.88, about 12% lower. RR is the ratio of the death risk in the two groups, and a value below 1 means the married side is lower. 95% CI 0.85–0.91 is the credible range for this figure. Looked at separately, the three groups come out at 1.11 for the widowed, 1.16 for the divorced or separated and 1.11 for the never married, which are about 10%, 16% and 10% higher. The result does not change when men and women are separated, when better and weaker studies are separated, or when Europe and the Americas are separated. Recalculated with the most conservative method, the RR is 0.94, only about 6% lower. +- Roelfs et al (2011) pooled 95 papers and more than 500 million people. People who had never married had an all-cause mortality HR of 1.24 compared with married people, about 24% higher. HR is a multiple in the same way as RR, and a value above 1 means higher risk. 1.19–1.30 is the credible range for this figure. The older a person is, the smaller the gap. The more rigorous the study, the smaller the gap it calculates. +- Wang et al (2020) pooled 21 studies and more than 7,890,000 people. Being outside a marriage was associated with all-cause mortality, cancer mortality and cardiovascular mortality, in men as well as women. The association was stronger in men for cardiovascular mortality and all-cause mortality. Never-married men had a 31% higher risk of dying from a stroke and a 9% higher all-cause mortality than never-married women. + +All three studies only followed records of groups of people, with no control group. People who are healthier, better paid and better socially connected are both more likely to get married and more likely to live longer. That is called selection effect. The Manzoli paper also found signs of publication bias. Publication bias means studies with attractive results are more likely to get published, so a pooled figure leans optimistic. These numbers can only be read as married groups having a somewhat lower death rate. They cannot be read as "get married and you will die 12% less". The samples in these studies are mostly European and American, so in China they serve as an order of magnitude and nothing more. + +### The time bill: the gender gap in unpaid labour is an upper-bound reference + +The figures come from the National Bureau of Statistics, Third National Time Use Survey, fielded in May 2024. Three terms first. Unpaid labour is the part of the work that nobody pays for. "Participants" means the people who really did that thing on the day. "Participation rate" means the share of people who did that thing on the day. The results are like this. Participants in unpaid labour spent an average of 2 hours 45 minutes a day. Men spent 1 hour 52 minutes and women 3 hours 29 minutes. The participation rate was 75.6%, with 67.5% for men and 83.9% for women. Housework taken on its own works out at 1 hour 17 minutes a day averaged over all residents. Counting only the people who really did it that day, it is 1 hour 59 minutes, 28 minutes less than in 2018, and the participation rate is 64.9%. Caring for family members taken on its own works out at 30 minutes a day averaged over all residents. Counting only the people who really did it that day, it is 1 hour 46 minutes, and the participation rate is 28.4%. + +The second survey, in 2018, came out like this. Housework averaged over all residents was 1 hour 26 minutes, with 45 minutes for men and 2 hours 6 minutes for women. The participation rate was 40.4% for men and 75.6% for women. Caring for the daily lives of children averaged over all residents was 36 minutes, with 17 minutes for men and 53 minutes for women. The participation rate was 12.3% for men and 25.1% for women. + +Both surveys break the figures down only by sex, by urban or rural area, and by age. Neither breaks them down by whether someone is married. So there is no official figure for "how much more housework do you do a day after getting married". What you can use is the gap between men and women. In 2024 the gender gap in unpaid labour was 1 hour 37 minutes. If the division of housework was never settled, that is about as large a gap as the two of you could end up with. + +### The money bill: default rules, written agreements and payment for housework + +There are no official statistics on betrothal gifts, weddings or the marital home, so this essay gives no figures for them. What can be set out clearly are the rules under the marriage and family part of the Civil Code. The Civil Code took effect on 1 January 2021. + +- Article 1062: four kinds of things you obtain between the wedding day and the divorce day count as joint property of the two of you. First, wages, bonuses and remuneration for work. Second, the returns from business and investment. Third, the returns from intellectual property. Fourth, property inherited or given to you by someone else. The fourth kind has one exception. If a will or a gift contract names only one of you, it is not joint property. Both people have equal power to dispose of joint property. +- Article 1063: four kinds of things count as personal property and do not go into joint property on divorce. First, property you already had before the marriage. Second, compensation or payments for injury to the body. Third, property that a will or a gift contract names as belonging to one side only. Fourth, daily necessities for the exclusive use of one person. +- Article 1065: the two of you can agree for yourselves who owns the property before and after the marriage. You can agree that each keeps their own. You can agree that everything is held in common. You can agree that part is kept separately and part is held in common. The agreement has to be in writing, and once written it binds both of you in law. If there is no agreement, or the agreement is unclear, the two rules above apply. +- Article 1088: if one of you has put in more than the other in looking after the children, caring for the elderly or helping the other with their work, then on divorce you can ask the other for a payment in compensation. How much goes to the two of you to agree first, and a court decides if you cannot agree. + +This set of rules is useful to both sides. If you are the one who earns more, read Article 1062 and Article 1065 closely. If you are the one who does more at home, read Article 1088 closely. A property agreement only has to be in writing. The law does not ask you to have it notarised at a notary office. + +### Exit cost: the cooling-off period and the grounds for going to court + +- Article 1076: if both of you want a divorce, first sign a written divorce agreement, then apply in person at the marriage registration office. The agreement has to say who looks after the children and how property and debts are handled. All of this has to be what the two of you have settled on. +- Article 1077: within the thirty days counted from the day the registration office receives the application, either party who changes their mind can withdraw the application. These thirty days are what people call the cooling-off period. In the thirty days after the cooling-off period ends, the two of you have to apply in person for the divorce certificate. If you do not go, the application is treated as withdrawn. +- Article 1079: if one of you wants a divorce and the other does not agree, you can go to mediation first, or you can sue in court directly. Once the case is in court, the court also has to try mediation first. If the court is satisfied that the marriage has genuinely broken down and mediation does not work, it should grant the divorce. In the situations below, where mediation has failed, the divorce should be granted: bigamy or living with someone else, domestic violence or abuse and abandonment, bad habits such as gambling and drug use that repeat after repeated admonition, and two years of separation because the two of you no longer get along. There is one further situation. If a first decision refuses the divorce, and the two of you then live apart for a full year, a second case should end in a divorce. + +Of the 3,513,000 couples who divorced in 2024, 891,000 went through a court judgment or court mediation. The cooling-off period only covers a divorce registered with the civil affairs department. It does not apply to a divorce fought in court. + +### Quality of the relationship: a bill worth more than whether you get married + +Robles et al (2014) pooled 126 studies and more than 72,000 people. The conclusion is that the higher the quality of a marriage, the better the health. How closely two things travel together is measured with a number called the effect size r. The closer r is to 0, the more nearly the two things are unrelated. The larger the absolute value of r, the more closely they travel together. A minus sign in front of r means that when one is high, the other is low. The r values for these associations run from 0.07 to 0.21. With risk of death, r = 0.11. With the cardiovascular response when the two of you are in conflict, r = −0.13. None of these values is large. But the association between health behaviours such as eating well and health is of the same order. Part of the results show publication bias, because attractive results are easier to get published. Some studies tested for a difference between men and women and found almost none. Most of these studies only followed records, with no control group, so no cause can be drawn from them. The cardiovascular review by Dhindsa et al (2020) also points out that how satisfied you are with a marriage and how good the marriage is have a clear effect on cardiovascular risk. + +This essay did not find a meta-analysis that directly compares mortality in "a marriage that goes badly" with mortality in "not being married". So "a bad marriage is worse than not marrying" is at present a conclusion drawn by inference, with no figures behind it. What holds up is this. The difference between "married or not" is mixed up with selection effect, and "how good the marriage is" is a separate matter. If you only look at whether you are married and never at how you are getting on, you lose the whole quality-of-relationship bill. + +## 3. How to work it out for yourself: a list to fill in + +Fill in one bill at a time. Do not add the numbers from different bills together. Write "do not know" in any box you cannot fill in. "Do not know" is useful information in its own right. + +What the other person brings (fill in one line at a time) + +- Money: how much the other person earns, how much they owe, and what property they had before the marriage. For property after the marriage, do you follow the default rules of the law, or do you write an agreement? If you write an agreement, what goes in it. +- Time: how much unpaid housework and care does the other person do a day at the moment. The national average is there for comparison. Among people who really did it on the day, men spent 1 hour 52 minutes and women 3 hours 29 minutes. How these jobs get split after you marry. How they get split when there are children and elderly relatives who need looking after. Whose job in the two of you can be put off. +- Quality of the relationship: how many times the two of you have clashed in the past half year, and how each one ended. What the other person did when you were ill, out of work or low in spirits. And the other way round, what you did when those things happened to them. +- Emotional value: when you are with the other person, do you come away with more energy or less. You can only tell this after a few months together. A list of conditions will not show it. + +What you put in (fill in one line at a time) + +- Money: how much you pay out for the marriage itself, including betrothal gifts, the wedding and the marital home. If you did not marry, where would that money go instead. +- Time: estimate how much more housework and care you do a day. Then estimate how much more time you spend a week on social obligations. +- Health: whether your sleep, your eating and your exercise get better or worse because of the marriage. That is closer to you than any figure a meta-analysis produces. +- Exit cost: for a registered divorce you wait out the 30-day cooling-off period and then go and collect the certificate within the following 30 days. For a court case you have to meet one of the situations in Article 1079. Can you carry that period. + +How the law protects you (applies to both people) + +- Whether there is evidence for the property there was before the marriage, such as a purchase contract, bank statements, a gift contract or a will. Can that evidence show that it was "there before the marriage", or that it "names one side only". +- Whether to sign a written property agreement. Whether it covers only property from before the marriage, only what is earned after it, or both ends. +- Who does more of the housework and care after the marriage. Does the one who does more know that Article 1088 allows a payment in compensation. Are the two of you willing to agree the method in advance. +- Debt: money borrowed after the marriage in one person's own name, and which of it counts as a joint debt of the two of you. Does the other person know about that money. + +Fill in the column on marrying to please elderly relatives on its own + +- What the elders get: peace of mind, a face that saves out, no more being pushed to marry, and someone to rely on for support in old age. +- Which of these are real benefits to the elders and which are really benefits to you. +- You carry the cost in all three rows above. The benefits the elders get do not turn into your health on their own, and they do not turn into the quality of your relationship on their own. +- Once you have written it out this way, whether this bill is heavy or light is for the person filling it in to judge. + +## 4. Common mistakes + +Mixing up the two divorce rates. One is the crude divorce rate, which counts how many couples divorce per 1,000 people, and in 2024 it was 2.5 per 1,000. The other is the "divorce-to-marriage ratio", which divides the number of divorces that year by the number of marriages that year, 3,513,000 / 6,106,000 ≈ 57.5%. The top and the bottom of that ratio are not the same group of people. Most of the people divorcing this year married many years ago. When there are fewer marriages, the ratio climbs on its own, so it cannot show that "half of marriages end in divorce". The bulletin from the Ministry of Civil Affairs has no such indicator either. + +Taking "occur together" for "so therefore". Married people have a death rate about 12% lower, and that is a result of pooling several studies together. But being healthy, well paid and well socially connected already affects both things at once: whether you can get married, and whether you can live long. This bill can be worked out, but read it at a discount, and it cannot take the place of a judgement about the relationship in front of you. + +Treating marriage as a task to be completed. As a task, you watch only the state of "married or not" and never "how you are getting on". But the effect of relationship quality on health is independent of whether you are married, and a decision made as a task usually loses that whole bill. + +Taking the national average for your own figures. The gender gap in the time use survey is a national average. The property rules are the default settings the law provides, and a written agreement can change them. National figures give you an order of magnitude and a starting point for negotiating. They do not give you a conclusion. + +Where there is no official figure, do not fill the gap with a figure from somewhere else. Betrothal gifts, weddings and the cost of raising children have no official statistics, so this essay leaves them blank. Filling in your own real numbers beats quoting any organisation's estimate. + +## Sources + +- 民政部 (2025). 2024 年民政事业发展统计公报（五（二）1 婚姻登记服务、注释 5）. （PDF：） +- 国家统计局 (2024). 第三次全国时间利用调查公报（第一号、第二号、第三号）. 、、；国家统计局社科文司负责人答记者问. +- 国家统计局 (2019). 2018 年全国时间利用调查公报. +- 全国人民代表大会 (2020). 中华人民共和国民法典（第一千零六十二、一千零六十三、一千零六十五、一千零七十六、一千零七十七、一千零七十九、一千零八十八条）. 国家法律法规数据库 ；条文全文见最高人民法院公报转载 +- 民政部 (2020). 关于贯彻落实《中华人民共和国民法典》中有关婚姻登记规定的通知（民发〔2020〕116 号）. +- Manzoli L, Villari P, Pirone GM, Boccia A (2007). Marital status and mortality in the elderly: a systematic review and meta-analysis. Soc Sci Med 64:77–94. +- Roelfs DJ, Shor E, Kalish R, Yogev T (2011). The rising relative risk of mortality for singles: meta-analysis and meta-regression. Am J Epidemiol 174(4):379–389. +- Wang Y, Jiao Y, Nie J, et al. (2020). Sex differences in the association between marital status and the risk of cardiovascular, cancer, and all-cause mortality: a systematic review and meta-analysis of 7,881,040 individuals. Glob Health Res Policy 5:4. +- Robles TF, Slatcher RB, Trombello JM, McGinn MM (2014). Marital quality and health: a meta-analytic review. Psychol Bull 140(1):140–187. +- Dhindsa DS, Khambhati J, Schultz WM, Tahhan AS, Quyyumi AA (2020). Marital status and outcomes in patients with cardiovascular disease. Trends Cardiovasc Med 30(4):215–220. diff --git "a/en/docs/\351\201\207\345\210\260\351\231\214\347\224\237\344\272\272\345\207\272\344\272\213\350\257\245\344\270\215\350\257\245\345\201\234.md" "b/en/docs/\351\201\207\345\210\260\351\231\214\347\224\237\344\272\272\345\207\272\344\272\213\350\257\245\344\270\215\350\257\245\345\201\234.md" new file mode 100644 index 00000000..b4653d3a --- /dev/null +++ "b/en/docs/\351\201\207\345\210\260\351\231\214\347\224\237\344\272\272\345\207\272\344\272\213\350\257\245\344\270\215\350\257\245\345\201\234.md" @@ -0,0 +1,55 @@ +[← Back to contents](../README.md) + +# When a stranger is in trouble on the street, walk on or stop + +This is the long-form version of item 2 in section 13 (an elderly person falls, someone goes down, do not rush to lift them). The book sorts every item by who ends up benefiting, and the standard is how likely the benefit is to come back to you later. A stranger is the lowest group. The chance that the other person repays you is small, and you do not know what they are like. So nothing here tells you what to do. It sets out in full what each of the three roads costs, and which one you take is your own decision. + +**No probabilities are given in this essay.** Every layer of cost below is real, and every one of them has a provision of law you can look up. But there is no published statistic for "out of a hundred people you help, how many actually end up in trouble". The rule in this book is that when a number is uncertain, leave it out. So do not read "seven layers are listed here" as "all seven will land on you". + +## Three roads + +| | Walk on | Stop but do not touch | Stop and help all the way | +| --- | --- | --- | --- | +| Legality | Legal. An ordinary person has no duty to rescue a stranger, and the criminal law has no offence of seeing an accident happen and not helping | Legal | Legal | +| Time | 0 | A few minutes. Dialling 120 leaves your number on record | Half a day at the least, and you may have to go to the hospital with them | +| Chance of being approached afterwards | Lowest | You are the one who left a number | Highest | +| What the other person gets | 0 | The ambulance arrives sooner | The most | + +## The costs you may run into after stopping, lightest to heaviest + +**One, a call back and a questioning.** If you dial 120 or 110, your number is left on record either way. The other party may ring you back to ask exactly where the place is. Later they may also ask you to come in and give a statement. In a criminal case, a person who reports it does not have to give their name in public, and the agency handling the case has to keep that confidence for them (Article 111 of the Criminal Procedure Law). There is no such confidentiality rule for a call to 120. + +**Two, the injured person's family comes for you.** If the cameras do not turn up the vehicle that caused it, the family will find you most easily. You left a number, and you were at the scene. They may ring and press you, or they may come to your door. + +**Three, you are sued in court.** Whoever says you hit them has to produce the evidence. If he cannot produce it, that consequence is his to bear (Article 90 of the Interpretation on the Application of the Civil Procedure Law). **But this provision only settles how the case is finally judged. It does not settle whether you get dragged into it.** The case is heard under the ordinary procedure, and the first instance has to be closed within 6 months, with a further extension of 6 months available in special circumstances. Under the summary procedure it is 3 months (Articles 152 and 164 of the Civil Procedure Law). Even if you win outright, there are only three things the other side can be made to pay: the case acceptance fee, the application fee, and the travel expenses and lost earnings of witnesses and appraisers when they attend court (Articles 6 and 29 of the Measures for the Payment of Litigation Costs). **The money you spend on your own lawyer is not in those three, and you pay it yourself.** + +**Four, it reaches your workplace.** If someone comes to your workplace to make trouble, you can call the police. Where a person disturbs the order of an enterprise, institution or public service and makes it impossible for work to go on normally, the police give a warning or a fine of up to 500 yuan. Where the circumstances are heavier, there is 5 to 10 days of detention, plus a fine of up to 1000 yuan. Where a group comes together to make trouble, the leader is detained for 10 to 15 days (Article 26, item 1, of the Law on Administrative Penalties for Public Security). Where a person keeps harassing, pestering or following, and makes it impossible for you to get through your days normally, the punishment is under Article 50, item 5. With the approval of the head of the public security organ, it can also be ordered that he may not contact you for a set period. If the company dismisses you over this, that is in general an illegal termination of the employment contract, and the compensation is worked out at 2N (item 6 in section 19). But whether you call the police or go to labour arbitration, you are the one who makes the trips. And the stretch of time you lose the job first is not given back to you because you later get 2N. + +**Five, someone films you and puts it online. ** **You may not be able to stop someone filming.** Without the consent of the person, you may not make, use or publish their likeness (Article 1019 of the Civil Code). But there is one exception. 「为实施新闻报道，不可避免地制作、使用、公开肖像权人的肖像」, which means making, using or publishing the likeness of the person holding the right to it where that is unavoidable in carrying out news reporting, counts as fair use and is not an infringement (Article 1020, item 2). A journalist filming in order to report the news falls under that exception. What you can do is refuse the interview, stay off camera, and say nothing in front of the lens. Whether you agree is your own decision, and nobody can agree on your behalf. + +**Six, the video gets passed around and you get abuse.** Once the video is out, the comment section is out of your hands. The measures you can take yourself are written into the Provisions on the Governance of Cyberbullying Information (in force from 1 August 2024). The platform is supposed to give you these protective settings: block a stranger or a particular person named by you, and limit who can see your own information. You can also stop people from reposting or commenting, take private messages only from friends, or take private messages from nobody (Article 23). The platform is supposed to offer a one-click function for capturing evidence, which fixes the content that abuses you (Article 25). The platform is supposed to put a dedicated shortcut for complaints and reports in a prominent position (Article 26). Where a platform finds anything suspected of being illegal or of a crime, it is supposed to report it to the public security organ promptly and supply the lead (Article 15). How to keep evidence yourself and how to call the police, see item 15 in section 8. + +**Seven, when you cannot take it any more.** The six layers above can each be answered with a provision of law. The seventh cannot. There is no reliable data from China to cite on how much being hounded by a crowd affects a person's mood. This book gives no figure. It gives one way out, the psychological assistance helpline 12356, see item 25 in section 1. + +## Two situations where walking on is not free + +- **You run or manage the place where this happens, or you organise a mass activity.** A place here means a hotel, a shopping centre, a bank, a station, an airport, a sports venue, an entertainment venue and places like them. You owe a duty of safety to the people inside. If you do not do it, you carry tort liability, which means paying money. Where the damage was caused by someone else and you still did not do it, you pay within the part that is your responsibility, and that is called supplementary liability (Article 1198 of the Civil Code). So a member of staff on duty, or the venue itself, walking away carries a different responsibility from a passer-by walking away. +- **You were already involved.** You hit the person, the vehicle is yours, or the danger came from something you did earlier. What matters then is not whether to look after someone else, but how to handle a traffic accident and a hit-and-run, see item 1 in section 8. + +## When you decide to stop, the least-effort way to do it + +1. **Do not start moving them or lifting them.** There is no second answer on this point. The reason is medical. Lifting someone roughly can make a brain bleed worse, and it can make an injury to the spine worse too. There is nothing to worry about on the legal side. Damage caused in the course of rescuing someone has already been exempted by Article 184 of the Civil Code. +2. Stand a little way off and shout to him, and see whether he is breathing. If he is not breathing, press hard on his chest, see item 1 in section 13. If he is breathing, stay where you are, dial 120 and give the location clearly. +3. Pull a passer-by in to do it with you, and start recording on your phone at the same time. A witness and a recording are more use than arguing about it afterwards. +4. Do not lift him into a car on your own, and do not put your own money in to take him to hospital. This step costs the most and it is the hardest to settle afterwards. +5. You are hurt yourself while rescuing someone, or you have put money in. If you want that money back, see item 39 in section 13 (you were hurt rescuing someone and you put money in, what to do). + +## Sources + +- 全国人大 (2020). 民法典（第一百八十四、一千零一十九、一千零二十、一千一百九十八条）. 最高人民检察院转载全文. +- 最高人民法院 (2022 年第二次修正，2022 年 4 月 10 日施行). 关于适用《中华人民共和国民事诉讼法》的解释（第九十条）. +- 全国人大常委会 (2023 年修正，2024 年 1 月 1 日施行). 民事诉讼法（第一百五十二、一百六十四条）. 上海市发展和改革委员会转载全文. +- 国务院 (2006). 诉讼费用交纳办法（国务院令第 481 号，第六、二十九条）. +- 全国人大常委会 (2025). 治安管理处罚法（2025 年修订，2026 年 1 月 1 日起施行，第二十六条第一项、第五十条第五项）. +- 全国人大常委会 (2018 年修正). 刑事诉讼法（第一百一十一条）. 宁夏回族自治区人民检察院转载全文. +- 国家互联网信息办公室、公安部、文化和旅游部、国家广播电视总局 (2024). 网络暴力信息治理规定（第十五、二十三、二十五、二十六条，2024 年 8 月 1 日施行）. 国务院公报. diff --git a/en/index.html b/en/index.html new file mode 100644 index 00000000..c5764dfd --- /dev/null +++ b/en/index.html @@ -0,0 +1,1505 @@ + + + + + + +A High-Return Life Guide · Get back lifespan, money and freedom for the least money, time and energy + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+
+ + A High-Return Life Guide + +
+ + + + + +
+
+
+ +
+ + + +
+
+
+

A High-Return Life Guide

+

Every item answers two questions: what it costs, and what it buys back.

+

Who benefits is graded too: you first, then your spouse and immediate family, then friends and colleagues, strangers last. Last does not mean zero — only that the return is unlikely, and that you have to weigh the risks along with it.

+

You do not have to do all of it. This is a shortlist ordered by value for money, not a task list. Taking one or two entries with you counts; come back for the rest when you need them. For the cheapest ones, set "Money spent" to "No money" and "Willpower needed" to "No willpower" on the left. The author has not done most of them either. Dotted "item X" references in the text open in place, so you can see what that item says without hunting for it.

+
Showing – shown, out of – in total
+
Long-form — Is marriage worth it · Home emergency kit checklist · Should you stop when a stranger is in trouble · Which licences a platform needs · Body clock and night shifts · Source verification records
+
AI assistant — Let Claude Code or Codex answer your questions from this book (how to install the skill)
+
Abbreviations and terms explained

Underlined terms in the text explain themselves when you click one or hover over it. The full table is below.

+
+
+
Reading the text …
+ +
+
Which measure a number belongs to is stated inside each item: sections 1 and 2 use all-cause or cause-specific mortality, sections 3 and 4 energy and time, section 5 money, section 7 money and benefits, sections 8 and 9 money and personal freedom. They are not converted into each other. The verification trail is in Source verification records. The text is published under CC BY 4.0 . Credit the author and link the original if you reprint or adapt it.
+
+
+ + + + + + + + + + diff --git a/index.html b/index.html index bc6057fe..7f79ac6e 100644 --- a/index.html +++ b/index.html @@ -3,1459 +3,42 @@ -高性价比人生指南 · 用最少的钱、时间和精力换回寿命、金钱和人身自由 - - - - - - - - - - - - - - - - - - - - - - - - - - +高性价比人生指南 · Choose your language + + + + + + - - - - - - - - - -
-
- - 高性价比人生指南 - -
- - - -
-
-
- -
- - - -
-
-
-

高性价比人生指南

-

每一条都回答两个问题：花掉什么，换回什么。

-

换回的算在谁头上也分档：你自己最高，其次配偶和直系亲属，再次朋友同事，陌生人最低——最低不等于零，只是回报的指望小、要连风险一起看。

-

不用全做：这是一份按性价比排好的备选单，不是任务清单。挑走一两条就算数，剩下的需要时再回来查——想挑省力的，左边把「花钱」选「不花钱」、「要毅力」选「不用」，剩下的就是。作者自己也没做到其中大部分。正文里带虚线的「第 X 条」可以点开，就地看那条写了什么，不用自己翻过去。

-
当前显示 – 条，共 – 条
-
长文：遇到陌生人出事该不该停 · 家庭应急装备清单 · 结婚划不划算 · 做平台要办哪些证 · 生物钟和夜班 ·来源核实记录
-
AI 助手：让 Claude Code 或 Codex 照这本书回答你的问题（skill 装法）
-
看不懂的缩写和名词

正文里带虚线的词，点一下或把鼠标放上去就有解释。全表如下。

-
-
-
正在读取正文 …
- -
-
数字口径以条目内标注为准：第 1、2 节是 ACM 或特定死因，第 3、4 节是精力/时间，第 5 节是金钱，第 7 节是金钱与保障，第 8、9 节是金钱与人身自由，互不换算。来源核实过程见 docs/核实记录。正文按 CC BY 4.0 发布，转载改编请署名并附原文链接。
-
+
+

高性价比人生指南

+

选一种语言 · Choose a language

+
- - - - - - - diff --git a/robots.txt b/robots.txt index 6319751f..e3f951ed 100644 --- a/robots.txt +++ b/robots.txt @@ -1,4 +1,4 @@ User-agent: * Allow: / -Sitemap: https://eternity4719.github.io/HowToLiveBetter/sitemap.xml +Sitemap: https://streamentry.github.io/HowToLiveBetter/sitemap.xml diff --git a/sitemap.xml b/sitemap.xml index ba61461b..ae940682 100644 --- a/sitemap.xml +++ b/sitemap.xml @@ -1,15 +1,15 @@ - https://eternity4719.github.io/HowToLiveBetter/ - 2026-09-08 - weekly + https://streamentry.github.io/HowToLiveBetter/ 1.0 - https://eternity4719.github.io/HowToLiveBetter/README.md - 2026-09-08 - weekly + https://streamentry.github.io/HowToLiveBetter/en/ + 0.9 + + + https://streamentry.github.io/HowToLiveBetter/zh/ 0.8 diff --git a/tools/i18n/BRIEF.md b/tools/i18n/BRIEF.md new file mode 100644 index 00000000..f32f1efe --- /dev/null +++ b/tools/i18n/BRIEF.md @@ -0,0 +1,112 @@ +# Translating this book + +This file is the brief given to every translation task. Read it, then translate the +block you were given and write it back to the same path. + +## What the book is + +A life guide ordered by value for money: every entry says what it costs (money, time, +energy, willpower), what it buys back, how strong the evidence is, and where the numbers +came from. Readers are ordinary adults with no professional training. Write for them. + +## Output + +Write the complete translated block to the file you were given, overwriting it. Output +nothing else — no commentary, no code fence around the whole thing. + +## Hard requirements + +These are machine-checked by `tools/i18n/check.mjs`. Any violation fails the build. + +1. **Structure is copied exactly.** Entry headings stay `### N. …` with the same numbers. + The same number of entries, in the same order. Section headings and intro paragraphs + are translated, not dropped. If a block has no `# ` section heading, do not invent one. + +2. **Field labels** (six per entry, in this order): + + | Chinese | English | Vietnamese | + | --- | --- | --- | + | `- 成本：` | `- Cost:` | `- Chi phí:` | + | `- 说人话：` | `- In plain language:` | `- Nói thẳng:` | + | `- 收益：` | `- Benefit:` | `- Lợi ích:` | + | `- 证据等级：` | `- Evidence grade:` | `- Mức bằng chứng:` | + | `- 来源：` | `- Sources:` | `- Nguồn:` | + | `- 备注：` | `- Notes:` | `- Ghi chú:` | + + If an entry lacks a field in the source, omit it. Never add one. + +3. **The cost-tag comment line is copied verbatim, including the Chinese:** + + ``` + + ``` + + Nothing inside is translated. The site's filters and its shareable URLs are keyed on + these exact Chinese tokens, and all three languages share one set of them. Translating + them silently breaks every filter and every link. + +4. **Numbers are frozen.** Every number in the Benefit, Cost and Sources fields stays + exactly as written: HR, RR, OR, SMR, percentages, `95% CI 0.50 到 0.68` (keep both + numbers; translate 到 as "to" / "đến"), years, counts, yuan amounts. No rounding, no + reformatting, no adding or removing thousands separators. + + Scale words must be converted, not copied: 「每 10 万人 6.60 人」 → "6.60 per 100,000 + people" / "6,60 trên 100.000 người"; 「167 万女性」 → "1,670,000 women" / "1.670.000 phụ + nữ"; 「每百万人 302 例」 → "302 per million" / "302 trên một triệu". Amounts of six or + more digits keep the shape 「约 X 万元（精确值 元）」 → "about 120,000 yuan (exact figure + 1130040 yuan)" / "khoảng 120.000 tệ (con số chính xác 1130040 tệ)". + + The Notes field is prose, so "40 例" may become "several dozen". The three fields above + may not. + +5. **Cross-references keep their numbers and anchor words.** + + | Chinese | English | Vietnamese | + | --- | --- | --- | + | 见第 13 节 | see section 13 | xem phần 13 | + | 见第 9 节第 22 条 | see item 22 in section 9 | xem mục 22 ở phần 9 | + | 本节第 32 条 | item 32 in this section | mục 32 của phần này | + | 见第 16 节第 1、2 条（…） | items 1 and 2 in section 16 (…) | xem mục 1 và 2 ở phần 16 (…) | + + Keep the anchor text in parentheses — it is what tells the reader which item is meant. + Never write "see above" or "see below". + + **Vietnamese only:** 条目 is "mục", 节 is "phần", and a statute's sub-clause is + "khoản (2)" or "điểm (a)". Never write "phần 2" for a statute clause — the reader + would take it for section 2 of the book. + + Statute article numbers are not cross-references: copy the number. 「刑法第三百九十条」 + → "Article 390 of the Criminal Law" / "Điều 390 của Bộ luật Hình sự". + +6. **Source links are byte-identical.** Every URL stays as written. Bibliographic entries + (authors, year, journal, DOI) keep their original form — they are there so a reader can + check the claim. Chinese organisation names may be translated with the Chinese in + parentheses on first use: "Ministry of Emergency Management (应急管理部)". + + Statute quotations (法条原文) stay in the original Chinese, verbatim, followed by a + plain-language explanation. This is the book's own rule: readers rely on being able to + compare the translation against the original text. + +7. **Disputed and unverified markers.** A Notes field beginning with 争议 must begin with + "Disputed" / "Tranh cãi". Keep the literal word "TODO" wherever it appears. The checker + counts disputed entries and unverified items; the counts have to match. + +8. **Register.** Short declarative sentences, one idea each, roughly 15–25 words, never + over 40. Explicit subjects: you, the doctor, the court, the company. Calm. No + exclamation marks. No lecturing. No AI-sounding filler: no "It is worth noting that", + no "essentially", no repeated "in other words", no uplifting summary sentence at the end + of a paragraph. No metaphors the reader has to translate themselves ("the other side", + "output", "a shield"). Negation, scope, conditions and judgements are content — do not + cut them as filler. + +9. **Glossary.** Use the exact terms in `tools/i18n/glossary.en.mjs` (English) or + `tools/i18n/glossary.vi.mjs` (Vietnamese). Recurring terms must be worded the same way + everywhere, or the book contradicts itself. + +## Checking your work + +After writing, the file is compared against the Chinese source: entry count, entry +numbers, field presence and order, the cost-tag comment, evidence grades, every number in +the three checked fields, cross-reference numbers, and link counts. Fix what it reports +before finishing. `node tools/i18n/join.mjs book/ --locale ` runs that +comparison for one section if you want to see it directly. diff --git a/tools/i18n/README.md b/tools/i18n/README.md new file mode 100644 index 00000000..fc3dd41a --- /dev/null +++ b/tools/i18n/README.md @@ -0,0 +1,61 @@ +# 译文维护 + +英文（`en/`）和越南文（`vi/`）由本仓库维护，中文（仓库根的 `README.md`、`book/`、`docs/`）是原文。 +作者只改中文那一份；两种译文由下面的流程跟上，结构对齐和数字一致性由机器查，作者不需要读译文。 + +判据、命令、坑，都写在 [BRIEF.md](BRIEF.md) 和 [../CLAUDE.md](../CLAUDE.md) 的「译文怎么维护」一节。 +这一份是索引。 + +## 日常流程 + +```bash +# 1. 中文改完之后，把要翻的节切块 +node tools/i18n/split.mjs book/01-不要早死.md --locale en + +# 2. 翻 .i18n/en/01-不要早死/partNN.md（已翻过的块不会被覆盖，重跑 split 不会冲掉译文） + +# 3. 逐块校一遍再合成 en/book/01-不要早死.md +node tools/i18n/join.mjs book/01-不要早死.md --locale en + +# 4. 整本校一遍（CI 跑的就是这个） +node tools/i18n/check.mjs + +# 5. 重建检索页（页面里的节数、卡片数跟着变） +node tools/site/build.mjs +``` + +`.i18n/` 是中间产物，已 gitignore。它同时是活的翻译进度：`split` 不覆盖已翻的块，所以每翻一块跑一次 +`join` 就能看到还差哪几块。 + +## 新增一种语言 + +1. `tools/site/locales.json` 加一项：`code`、`dir`、`contentBase`（浏览器那边的相对根）、`assetBase`、 + `repoBlob`、`htmlLang`、`ogLocale`、`sortLocale`、`name`（语言自己的名字）、`accept`（匹配 + `navigator.language` 的正则）、`canonical`、`note`、`default` 或 `source`（各一个）。 +2. `tools/site/locales/.json`：解析正文用的标记（`fields`、`dispute`、`todo`、`glossary`、 + `contents`…）加界面文案 `strings`。`node tools/site/extract.mjs` 只能从模板抠中文，文案得自己写。 +3. `tools/i18n/glossary..mjs`：词表。 +4. `node tools/i18n/split.mjs --all --locale ` 切块，翻，`join --all`。 + +`tools/site/extract.mjs` 里 `cost.money.0` 那一组（动态拼出来的键）要往 `DYNAMIC_KEYS` 里加， +漏了页面就显示不出那几个徽章——`build.mjs` 会报「缺 N 个键」，不会静默过去。 + +## 文件 + +| 文件 | 管什么 | +| --- | --- | +| `BRIEF.md` | 翻译要求本身，人看的版本。改要求改这一份 | +| `prompt.mjs` | 把 BRIEF.md 整个塞进模型提示词（要接 API 时用这个） | +| `split.mjs` | 一节正文切成十来个条一块，放 `.i18n/<语言>/<节名>/partNN.md` | +| `join.mjs` | 逐块校一遍再合成 `<语言>/book/<节名>.md`，对不上就不写盘 | +| `structure.mjs` | 一节（或一块）的结构骨架 + 逐条比对。`join` 和 `check` 共用同一份判据 | +| `check.mjs` | 拿中文原文对全部译文，CI 的「译文检查」job 跑它 | +| `anchors.mjs` | 译本里逐字照抄的那几样（成本标签注释、字段顺序、不许动的数字、指路、链接） | +| `glossary.{en,vi}.mjs` | 词表。同一件事全书一个说法，靠它 | + +## 为什么译文不用人审 + +`check.mjs` 查的都是「翻着翻着丢了东西」：节数条数、六个字段的有无与顺序、成本标签注释、 +证据等级、争议与待核实的条数、收益/成本/来源三栏里的每一个数字、交叉引用的条号节号、 +来源与备注里的链接数量。数字是全书可核对性的底座，改一个就断了，而这一层机器查得出来。 +剩下的是文字质量，那不在工具的判据里——发现哪处别扭，开个 issue 说第几节第几条。 diff --git a/tools/i18n/anchors.mjs b/tools/i18n/anchors.mjs new file mode 100644 index 00000000..87a28be5 --- /dev/null +++ b/tools/i18n/anchors.mjs @@ -0,0 +1,20 @@ +// 译本里那些不随语言变的部分，逐字照抄原文的那几处。各语言共用一份，别各写一遍。 +// +// 成本标签注释是检索页的数据源：index.html（现在在 tools/site/page.template.html）里的 +// COST_W 按「钱/时间/毅力 = 0/少/多」算性价比档，data-v 拿「极高/高/一般、死亡率/金钱/时间/自由」 +// 这些值做筛选和地址栏参数。三种语言共用一套取值，译文里必须原样保留中文， +// 换了语言值就换不出来，筛选和分享出去的链接全废。所以译文一律照抄，不译。 +export const COST_TAG = ''; + +// 条目里的六个字段名。各语言有自己的写法（中文「成本：」，英文「- Cost:」）， +// 但字段的个数、顺序和语义必须和原文一样——check.mjs 逐条对。 +export const FIELDS = ['cost', 'human', 'gain', 'grade', 'src', 'note']; + +/** 收益栏不许动的数字：这些是全书可核对性的底座，原值照搬，只在旁边补换算 */ +export const FROZEN = /\d+(?:[.,]\d+)*/g; + +/** 指路的写法。译本里「第 X 条」要带着锚点词一起译，别把条号丢了 */ +export const REF = /第\s*\d+\s*条|第\s*\d+\s*节|本节第\s*\d+\s*条/g; + +/** 每条来源里的链接。一条都不许删、不许改地址 */ +export const LINK = /https?:\/\/[^\s>）)]+/g; diff --git a/tools/i18n/check.mjs b/tools/i18n/check.mjs new file mode 100644 index 00000000..324fda96 --- /dev/null +++ b/tools/i18n/check.mjs @@ -0,0 +1,70 @@ +// 校译文。中文是原文，en/ 和 vi/ 是译本；这一套检查抓的是「翻着翻着就把东西弄丢了」。 +// +// node tools/i18n/check.mjs # 校所有语言 +// node tools/i18n/check.mjs --locale en # 只校英文 +// node tools/i18n/check.mjs --warnings # 不一致只提示不失败（还没翻完时用） +// +// 逐条对着原文查。判据在 tools/i18n/structure.mjs 里（join 逐块校时用的是同一份）： +// ① 条数、条号、字段的有无与顺序和原文一致 +// ② 成本标签注释逐字照抄 —— 检索页的筛选和地址栏全靠它，三种语言共用一套中文取值 +// ③ 证据等级 A/B/C 一样；标了争议的、含待核实的条数一样 +// ④ 收益、成本、来源三栏里的数字一个不多一个不少（量级词换算除外，见 structure.mjs） +// ⑤ 「第 X 条」「第 X 节」指路的条号节号都还在 +// ⑥ 来源栏和备注栏里的 http(s) 链接数量一致 +// ⑦ 每个文件的回总目录链接指向本语言的目录 +// +// ④ 和 ⑥ 最要紧：正文里一个数字被改动，全书的可核对性就断在那儿，而这一层机器查得出来。 + +import { readFileSync, existsSync, readdirSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; +import { LOCALES, SOURCE, LOCALE, strings } from '../site/locales.mjs'; +import { STRUCT, entrySkeleton } from './structure.mjs'; + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); +const WARN = process.argv.includes('--warnings'); +const only = process.argv.indexOf('--locale'); +const codes = only < 0 ? LOCALES.map(l => l.code) : [process.argv[only + 1]]; + +const problems = []; +const report = (where, what) => problems.push(`${where}：${what}`); + +for (const code of codes){ + if (code === SOURCE.code) continue; + const L = LOCALE(code); + const S = strings(code); + + const srcDir = (SOURCE.contentDir ? SOURCE.contentDir + '/' : '') + 'book/'; + const files = readdirSync(resolve(ROOT, srcDir)).filter(f => f.endsWith('.md')).sort(); + const have = []; + for (const f of files){ + const dstPath = `${L.dir}book/${f}`; + if (!existsSync(resolve(ROOT, dstPath))) continue; + have.push(f); + const src = entrySkeleton(read(srcDir + f), strings(SOURCE.code)); + const dst = entrySkeleton(read(dstPath), S); + for (const e of dst.entries) e._locale = code; + if (src.n !== dst.n) report(`${code} book/${f}`, `节号是 ${dst.n}，原文 ${src.n}`); + if (src.entries.length !== dst.entries.length) + report(`${code} book/${f}`, `条目数 ${dst.entries.length}，原文 ${src.entries.length}`); + for (let i = 0; i < Math.min(src.entries.length, dst.entries.length); i++) + for (const w of STRUCT.compare(src.entries[i], dst.entries[i], `${code} book/${f} 第 ${src.entries[i].n} 条`)) + problems.push(w); + // ⑦ 回总目录那行要指向本语言的目录 + const first = read(dstPath).split('\n')[0]; + if (!/^\[.*\]\((?:\.\.\/|\.\/)?README\.md\)/.test(first)) + report(`${code} book/${f}`, `第一行「${first.slice(0, 40)}」不是回本语言目录的链接`); + } + const all = existsSync(resolve(ROOT, `${L.dir}README.md`)); + console.log(`${L.name}（${L.dir}）：正文 ${have.length}/${files.length} 节已翻${all ? '' : '，README 还没有'}`); +} + +if (problems.length){ + console.log(`\n${problems.length} 处对不上：`); + for (const p of problems.slice(0, 200)) console.log(' ' + p); + if (problems.length > 200) console.log(` …还有 ${problems.length - 200} 处`); + if (WARN){ console.log('\n（--warnings：只提示，不判失败）'); process.exit(0); } + process.exit(1); +} +console.log('译文检查通过'); diff --git a/tools/i18n/glossary.en.mjs b/tools/i18n/glossary.en.mjs new file mode 100644 index 00000000..2f17e0fe --- /dev/null +++ b/tools/i18n/glossary.en.mjs @@ -0,0 +1,217 @@ +// 译文词表（英文）：正文里反复出现、必须前后一致的词。译的时候按这张表走，别各翻各的。 +// 只列最容易翻岔的：节标题、字段名、口径、档位、成本标签取值、机构名、统计术语。 +// 新词往里加，check.mjs 会提醒有没有翻岔的。 +export const GLOSSARY = [ + ["书名", "A High-Return Life Guide"], + ["性价比", "value for money"], + ["不要早死", "Do not die young"], + ["不要慢慢死", "Do not die slowly"], + ["不要浪费精力", "Do not waste energy"], + ["不要浪费时间", "Do not waste time"], + ["不要浪费钱", "Do not waste money"], + ["反面清单", "The reverse list"], + ["没钱的时候怎么活", "How to get by with no money"], + ["别把自己搭进去", "Do not put yourself at risk"], + ["普通人容易踩的法律红线", "Legal lines ordinary people cross"], + ["恋爱和结婚划不划算", "Is dating and marriage worth it"], + ["程序员和技术人容易踩的红线", "Lines programmers and tech workers cross"], + ["创业与做生意：别把家底赔进去", "Starting a business: do not lose the family savings"], + ["紧急情况：先做什么", "Emergencies: what to do first"], + ["账号与信息安全", "Accounts and information security"], + ["租房与买房", "Renting and buying a home"], + ["得了慢性病之后怎么活", "Living with a chronic illness"], + ["家里有老人", "When there are elderly at home"], + ["养孩子划不划算", "Is raising children worth it"], + ["在职、离职和工伤", "Working, leaving, and work injuries"], + ["刚出生的孩子怎么带", "Caring for a newborn"], + ["出国、旅行与境外安全", "Travel and safety abroad"], + ["怎么放松：娱乐场所和减压", "Relaxing: venues and stress relief"], + ["学什么技能划算", "Which skills are worth learning"], + ["看病：怎么少花钱少走弯路", "Medical care: spending less and going fewer rounds"], + ["人走了以后要办什么", "What to do after someone dies"], + ["做一个网站或平台：资质、备案和服务器", "Running a website or platform: licences, filings, servers"], + ["怀孕和生产：从发现怀孕到出院办证", "Pregnancy and childbirth: from test to paperwork"], + ["别为了外形把身体搞坏", "Do not wreck your body for the sake of looks"], + ["遭遇重大打击之后", "After a severe blow"], + ["上学以后的孩子", "School-age children"], + ["十八岁之后有哪几条路", "What are the paths after eighteen"], + ["出国留学：身份、打工、保险和回国认证", "Studying abroad: status, work, insurance, certification"], + ["残疾之后怎么活", "Living after a disability"], + ["家里的常备药别吃出事", "Do not mishandle the medicines at home"], + ["成本", "Cost"], + ["说人话", "In plain language"], + ["收益", "Benefit"], + ["证据等级", "Evidence grade"], + ["来源", "Sources"], + ["备注", "Notes"], + ["总死亡率", "all-cause mortality"], + ["死亡率", "mortality"], + ["金钱", "money"], + ["时间", "time"], + ["精力", "energy"], + ["人身自由", "personal freedom"], + ["极高", "exceptional"], + ["高", "high"], + ["一般", "moderate"], + ["争议", "disputed"], + ["待核实", "to be verified"], + ["含待核实", "contains items to verify"], + ["大", "large"], + ["中", "medium"], + ["小", "small"], + ["不花钱", "no money"], + ["花少量钱", "a little money"], + ["花不少钱", "a fair amount of money"], + ["顺手", "in passing"], + ["花几小时", "a few hours"], + ["每天占时间", "daily"], + ["不用毅力", "no willpower"], + ["要一点毅力", "some willpower"], + ["要很多毅力", "a lot of willpower"], + ["本节第", "item "], + ["第 X 条", "item X"], + ["第 X 节", "section X"], + ["见第", "see item "], + ["长文见", "long-form essay: "], + ["国家卫生健康委", "National Health Commission"], + ["中国疾病预防控制中心", "China CDC"], + ["中国消费者协会", "China Consumers Association"], + ["国家统计局", "National Bureau of Statistics"], + ["人力资源和社会保障部", "Ministry of Human Resources and Social Security"], + ["最高人民法院", "Supreme People's Court"], + ["国家市场监督管理总局", "State Administration for Market Regulation"], + ["总死亡率", "all-cause mortality"], + ["可信范围", "credible range"], + ["荟萃分析", "meta-analysis"], + ["随机分组的试验", "randomised trial"], + ["队列", "cohort"], + ["剂量反应", "dose-response"], + ["绝对风险", "absolute risk"], + ["相对风险", "relative risk"], + ["基准人群", "comparison group"], + ["效应量", "effect size"], + + // 法律与行政程序的固定说法 + ["违法", "illegal"], + ["犯罪", "a crime"], + ["刑事责任", "criminal liability"], + ["行政处罚", "administrative penalty"], + ["拘留", "detention"], + ["罚款", "fine"], + ["民事赔偿", "civil damages"], + ["诉讼时效", "the statute of limitations"], + ["劳动仲裁", "labour arbitration"], + ["被辞退", "dismissal"], + ["劳动合同", "employment contract"], + ["社保", "social insurance"], + ["医保", "medical insurance"], + ["低保", "means-tested income support"], + ["救助站", "rescue station"], + ["公积金", "housing provident fund"], + ["立案", "opening a case"], + ["取证", "collecting evidence"], + ["违法所得", "illegal gains"], + ["举报", "report to the authorities"], + ["起诉", "sue"], + ["强制执行", "enforcement"], + ["失信被执行人", "judgment defaulter"], + ["律师费", "legal fees"], + ["公证", "notarisation"], + ["立遗嘱", "make a will"], + ["意定监护", "voluntary guardianship"], + ["残疾人证", "disability certificate"], + ["工伤", "work injury"], + ["职业病", "occupational disease"], + ["试用期", "probation period"], + ["年休假", "annual leave"], + ["加班费", "overtime pay"], + // 医学 + ["总死亡率", "all-cause mortality"], + ["相对风险", "relative risk"], + ["可信区间", "confidence interval"], + ["筛查", "screening"], + ["接种", "vaccination"], + ["急救", "first aid"], + ["心肺复苏", "cardiopulmonary resuscitation"], + ["婴儿死亡率", "infant mortality"], + ["孕产妇", "maternal"], + ["血压", "blood pressure"], + ["血脂", "blood lipids"], + ["中风", "stroke"], + ["心肌梗死", "heart attack"], + ["癌症筛查", "cancer screening"], + ["抗生素", "antibiotics"], + ["处方", "prescription"], + ["剂量", "dose"], + ["副作用", "side effects"], + ["禁忌", "contraindication"], + ["脱水", "dehydration"], + ["窒息", "suffocation"], + ["止血带", "tourniquet"], + ["骨折", "fracture"], + ["烫伤", "scald"], + ["休克", "shock"], + ["急救车", "ambulance"], + ["急救电话", "emergency number"], + // 机构 + ["国家卫生健康委", "National Health Commission"], + ["中国疾病预防控制中心", "China CDC"], + ["中国消费者协会", "China Consumers Association"], + ["国家统计局", "National Bureau of Statistics"], + ["人力资源和社会保障部", "Ministry of Human Resources and Social Security"], + ["最高人民法院", "Supreme People's Court"], + ["最高人民检察院", "Supreme People's Procuratorate"], + ["国家市场监督管理总局", "State Administration for Market Regulation"], + ["生态环境部", "Ministry of Ecology and Environment"], + ["教育部", "Ministry of Education"], + ["公安部", "Ministry of Public Security"], + ["民政部", "Ministry of Civil Affairs"], + ["国家医保局", "National Healthcare Security Administration"], + ["全国妇联", "All-China Women's Federation"], + ["中华全国总工会", "All-China Federation of Trade Unions"], + ["卫生健康委员会", "health commission"], + ["疾病预防控制中心", "CDC"], + ["疾控中心", "CDC"], + ["乡镇卫生院", "township health centre"], + ["社区卫生服务中心", "community health centre"], + ["三甲医院", "top-tier hospital"], + ["医院", "hospital"], + ["医生", "doctor"], + ["护士", "nurse"], + ["救护车", "ambulance"], + // 国际机构 + ["世界卫生组织", "World Health Organization"], + ["美国食品药品监督管理局", "US FDA"], + ["美国交通安全主管部门", "US National Highway Traffic Safety Administration"], + ["美国疾病预防控制中心", "US CDC"], + ["欧盟", "the European Union"], + ["世界银行", "World Bank"], + // 钱的说法 + ["万元", "ten thousand yuan"], + ["人民币", "yuan"], + ["元", "yuan"], + ["罚款金额", "the fine"], + ["保费", "the premium"], + ["理赔", "the claim"], + ["免赔额", "the deductible"], + ["起付线", "the deductible"], + // 时间与单位 + ["年", "year"], + ["岁", "years old"], + ["个月", "months"], + ["每周", "per week"], + ["每天", "per day"], + ["每天", "daily"], + ["小时", "hours"], + ["分钟", "minutes"], + ["秒", "seconds"], + // 证据等级的措辞 + ["荟萃分析", "meta-analysis"], + ["随机分组的试验", "randomised trial"], + ["大型队列", "large cohort"], + ["观察性队列", "only followed records, with no control group"], + ["作者经验", "the author's own experience"], + ["普遍共识", "general consensus"], + ["有争议", "disputed"], +] + diff --git a/tools/i18n/glossary.vi.mjs b/tools/i18n/glossary.vi.mjs new file mode 100644 index 00000000..48275c4d --- /dev/null +++ b/tools/i18n/glossary.vi.mjs @@ -0,0 +1,169 @@ +// 译文词表（越南文）：正文里反复出现、必须前后一致的词。译的时候按这张表走，别各翻各的。 +// 只列最容易翻岔的：节标题、字段名、口径、档位、成本标签取值、机构名、统计术语。 +// 新词往里加，check.mjs 会提醒有没有翻岔的。 +export const GLOSSARY = [ + ["书名", "Cẩm Nang Sống Hiệu Quả Cao"], + ["性价比", "hiệu quả chi phí"], + ["不要早死", "Đừng chết sớm"], + ["不要慢慢死", "Đừng chết dần"], + ["不要浪费精力", "Đừng phí sức"], + ["不要浪费时间", "Đừng phí thời gian"], + ["不要浪费钱", "Đừng phí tiền"], + ["反面清单", "Danh sách phản tác dụng"], + ["没钱的时候怎么活", "Sống thế nào khi hết tiền"], + ["别把自己搭进去", "Đừng để bản thân rơi vào rủi ro"], + ["普通人容易踩的法律红线", "Vạch đỏ pháp lý dân dân thường dễ vấp"], + ["恋爱和结婚划不划算", "Yêu đương và kết hôn có đáng không"], + ["程序员和技术人容易踩的红线", "Vạch đỏ lập trình viên và người làm kỹ thuật dễ vấp"], + ["紧急情况：先做什么", "Tình huống khẩn cấp: làm gì trước"], + ["成本", "Chi phí"], + ["说人话", "Nói thẳng"], + ["收益", "Lợi ích"], + ["证据等级", "Mức bằng chứng"], + ["来源", "Nguồn"], + ["备注", "Ghi chú"], + ["总死亡率", "tỷ lệ tử vong toàn nguyên"], + ["死亡率", "tỷ lệ tử vong"], + ["金钱", "tiền"], + ["时间", "thời gian"], + ["精力", "sức lực"], + ["人身自由", "tự do cá nhân"], + ["极高", "rất cao"], + ["高", "cao"], + ["一般", "trung bình"], + ["争议", "tranh cãi"], + ["待核实", "cần đối chiếu"], + ["大", "lớn"], + ["中", "vừa"], + ["小", "nhỏ"], + ["不花钱", "không tốn tiền"], + ["花少量钱", "tốn ít tiền"], + ["花不少钱", "tốn khá nhiều tiền"], + ["顺手", "tiện tay"], + ["花几小时", "mất vài tiếng"], + ["每天占时间", "mỗi ngày"], + ["不用毅力", "không cần ý chí"], + ["要一点毅力", "cần một chút ý chí"], + ["要很多毅力", "cần nhiều ý chí"], + + // 法律与行政程序的固定说法 + ["违法", "vi phạm pháp"], + ["犯罪", "tội phạm"], + ["刑事责任", "trách nhiệm hình sự"], + ["行政处罚", "xử phạt hành chính"], + ["拘留", "bị giữ"], + ["罚款", "phạt tiền"], + ["民事赔偿", "bồi thường dân sự"], + ["诉讼时效", "thời hiệu khởi kiện"], + ["劳动仲裁", "hoà giải lao động"], + ["被辞退", "bị sa thải"], + ["劳动合同", "hợp đồng lao động"], + ["社保", "bảo hiểm xã hội"], + ["医保", "bảo hiểm y tế"], + ["低保", "trợ cấp người có thu nhập thấp"], + ["救助站", "trạm cứu hộ"], + ["公积金", "quỹ tích góp nhà ở"], + ["立案", "khởi kiện"], + ["被指控", "bị buộc tội"], + ["自首", "đầu thú"], + ["违法所得", "thu nhập bất hợp pháp"], + ["举报", "báo việc"], + ["起诉", "khởi kiện"], + ["强制执行", "thi hành án cưỡng chế"], + ["失信被执行人", "người bị thi hành án không thi hành"], + ["律师费", "phí luật sư"], + ["立遗嘱", "lập di chúc"], + ["意定监护", "giám hộ tự nguyện bằng văn bản"], + ["残疾人证", "giấy chứng nhận người khuyết tật"], + ["工伤", "tai nạn lao động"], + ["职业病", "bệnh nghề nghiệp"], + ["试用期", "thời gian thử việc"], + ["年休假", "nghỉ phép năm"], + ["加班费", "tiền làm thêm giờ"], + // 医学 + ["可信区间", "khoảng tin cậy"], + ["筛查", "sàng lọc"], + ["接种", "tiêm phòng"], + ["急救", "cấp cứu"], + ["心肺复苏", "hồi sinh tim phổi"], + ["婴儿死亡率", "tỷ lệ tử vong trẻ sơ sinh"], + ["孕产妇", "sản phụ"], + ["血压", "huyết áp"], + ["血脂", "mỡ máu"], + ["中风", "đột quỵ"], + ["心肌梗死", "nhồi máu cơ tim"], + ["癌症筛查", "sàng lọc ung thư"], + ["抗生素", "kháng sinh"], + ["处方", "đơn thuốc"], + ["剂量", "liều"], + ["副作用", "tác dụng phụ"], + ["禁忌", "chống chỉ định"], + ["脱水", "mất nước"], + ["窒息", "ngạt thở"], + ["止血带", "gar"], + ["骨折", "gãy xương"], + ["烫伤", "bỏng"], + ["休克", "sốc"], + ["急救电话", "số điện thoại cấp cứu"], + // 机构 + ["中国疾病预防控制中心", "Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc"], + ["中国消费者协会", "Hiệp hội Người tiêu dùng Trung Quốc"], + ["国家统计局", "Cục Thống kê Quốc gia"], + ["人力资源和社会保障部", "Bộ Nhân sự và Xã hội"], + ["最高人民法院", "Tòa án nhân dân tối cao"], + ["最高人民检察院", "Viện kiểm sát nhân dân tối cao"], + ["国家市场监督管理总局", "Cục Quản lý thị trường quốc gia"], + ["生态环境部", "Bộ Môi trường và Tự nhiên"], + ["教育部", "Bộ Giáo dục"], + ["公安部", "Bộ Công an"], + ["民政部", "Bộ Dân vụ"], + ["国家医保局", "Cục Quản lý bảo hiểm y tế quốc gia"], + ["疾病预防控制中心", "Trung tâm kiểm soát bệnh"], + ["疾控中心", "Trung tâm kiểm soát bệnh"], + ["乡镇卫生院", "trạm y tế xã"], + ["社区卫生服务中心", "trạm y tế cộng đồng"], + ["三甲医院", "bệnh viện hạng A"], + ["医院", "bệnh viện"], + ["医生", "bác sĩ"], + ["护士", "y tá"], + ["救护车", "xe cứu thương"], + ["药店", "hiệu thuốc"], + // 国际机构 + ["世界卫生组织", "Tổ chức Y tế Thế giới"], + ["美国食品药品监督管理局", "Cục Quản lý Thực phẩm và Dược phẩm Hoa Kỳ"], + ["美国疾病预防控制中心", "Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ"], + ["欧盟", "Liên minh châu Âu"], + ["世界银行", "Ngân hàng Thế giới"], + // 钱的说法 + ["万元", "vạn tệ"], + ["人民币", "nhân dân tệ"], + ["元", "tệ"], + ["保费", "tiền phí bảo hiểm"], + ["理赔", "bồi hoàn"], + ["免赔额", "mức miễn thường"], + ["起付线", "mức đồng chi trả"], + ["房贷", "vay mua nhà"], + ["公积金", "quỹ tích góp nhà ở"], + ["社保", "bảo hiểm xã hội"], + // 时间的说法 + ["年", "năm"], + ["岁", "tuổi"], + ["个月", "tháng"], + ["每周", "mỗi tuần"], + ["每天", "mỗi ngày"], + ["小时", "giờ"], + ["分钟", "phút"], + ["秒", "giây"], + // 证据等级的措辞 + ["荟萃分析", "phân tích tổng hợp"], + ["随机分组的试验", "thử nghiệm chia nhóm ngẫu nhiên"], + ["大型队列", "nghiên cứu mẫu lớn theo thời gian"], + ["观察性队列", "chỉ ghi nhận và theo dõi, không chia nhóm đối chứng"], + ["作者经验", "kinh nghiệm của tác giả"], + ["普遍共识", "đồng thuận chung"], + ["争议", "còn tranh cãi"], + ["说明", "lưu ý"], + ["举例", "ví dụ"], + ["也就是说", "nói cách khác"], +] + diff --git a/tools/i18n/join.mjs b/tools/i18n/join.mjs new file mode 100644 index 00000000..eeb185ed --- /dev/null +++ b/tools/i18n/join.mjs @@ -0,0 +1,84 @@ +// 把译好的块按块号拼回一节正文，再跑结构检查。 +// +// node tools/i18n/join.mjs book/01-不要早死.md --locale en +// node tools/i18n/join.mjs --all --locale vi +// +// 块在 .i18n/<语言>/<节名>/partNN.md，翻译完一块就往那儿放一个文件。 +// 少一块就少合一块（进度能看出来），但 check.mjs 会因为条目数对不上而失败。 +// +// 拼之前先逐块对一遍原文（条号、字段、成本标签、数字、链接），有一块不对就在这儿 +// 报出来，不用等拼完才发现第 31 条少了一句。 + +import { readFileSync, writeFileSync, existsSync, readdirSync, mkdirSync } from 'node:fs'; +import { resolve, dirname, basename } from 'node:path'; +import { fileURLToPath } from 'node:url'; +import { LOCALES, SOURCE, strings } from '../site/locales.mjs'; +import { splitSection } from './split.mjs'; +import { STRUCT, entrySkeleton } from './structure.mjs'; + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); + +const argv = process.argv.slice(2); +const li = argv.indexOf('--locale'); +const locale = li < 0 ? 'en' : argv[li + 1]; +const L = LOCALES.find(x => x.code === locale); +if (!L) { console.error(`--locale 得是 ${LOCALES.map(x => x.code).join(' / ')} 里的一项`); process.exit(2); } + +const files = []; +if (argv.includes('--all')) { + const dir = (SOURCE.contentDir ? SOURCE.contentDir + '/' : '') + 'book/'; + for (const f of readdirSync(resolve(ROOT, dir))) if (f.endsWith('.md')) files.push(dir + f); +} else { + for (const a of argv) if (a.endsWith('.md')) files.push(a); +} +if (!files.length) { console.error('给一个 book/*.md 的路径，或者加 --all'); process.exit(2); } + +// ---------------------------------------------------------------- 逐节拼 +let bad = 0, done = 0; +for (const srcPath of files){ + const name = basename(srcPath); + const dir = `.i18n/${locale}/${name.replace(/\.md$/, '')}`; + if (!existsSync(resolve(ROOT, dir))) { console.log(`跳过 ${name}：还没切块`); continue; } + + const src = splitSection(read(srcPath)); + const parts = readdirSync(resolve(ROOT, dir)).filter(f => /^part\d+\.md$/.test(f)).sort(); + if (parts.length !== src.blocks.length){ + bad++; + console.error(`${locale} ${name}：切了 ${src.blocks.length} 块，译文只有 ${parts.length} 块，缺 ${src.blocks.length - parts.length} 块`); + continue; + } + + // 块对块地校：第 k 块对第 k 块，中途插入或漏译都会在这里露出来 + let trouble = 0; + const pieces = []; + for (let k = 0; k < parts.length; k++){ + const want = src.blocks[k].join('\n'); + const got = read(`${dir}/${parts[k]}`); + const a = entrySkeleton(want, strings(SOURCE.code)); + const b = entrySkeleton(got, strings(locale)); + for (const e of b.entries) e._locale = locale; // 字段名按译文那门语言认 + for (const [i, x] of a.entries.entries()){ + const y = b.entries[i]; + const tag = `${name} 块 ${k} 第 ${x.n} 条`; + if (!y){ bad++; trouble++; console.error(` ${tag}：译文缺这一条`); continue; } + for (const w of STRUCT.compare(x, y, tag)){ bad++; trouble++; console.error(' ' + w); } + } + if (b.entries.length !== a.entries.length){ + bad++; trouble++; + console.error(` ${dir}/${parts[k]}：${b.entries.length} 条，原文这块 ${a.entries.length} 条`); + } + pieces.push(k === 0 ? src.head.join('\n') + '\n' + got : got); + } + if (trouble) { console.error(`${locale} ${name}：${trouble} 处对不上，没写进 ${L.dir}book/${name}`); continue; } + + let out = pieces.join('\n'); + const tail = `${dir}/tail.txt`; + if (existsSync(resolve(ROOT, tail))) out += '\n' + read(tail); + mkdirSync(dirname(resolve(ROOT, `${L.dir}book/${name}`)), { recursive: true }); + writeFileSync(resolve(ROOT, `${L.dir}book/${name}`), out); + done++; + console.log(`${locale} book/${name}：${parts.length} 块合好，${src.count} 条，${((out.length / 1024) | 0)}K`); +} +console.log(`\n合好 ${done} 节，问题 ${bad} 处`); +process.exit(bad ? 1 : 0); \ No newline at end of file diff --git a/tools/i18n/prompt.mjs b/tools/i18n/prompt.mjs new file mode 100644 index 00000000..d0272524 --- /dev/null +++ b/tools/i18n/prompt.mjs @@ -0,0 +1,24 @@ +// 翻译任务的提示词。真正的要求写在 tools/i18n/BRIEF.md（人看的版本，两边一致）： +// 结构、字段名、成本标签逐字照抄、数字不许动、指路、链接、语气，都在那边。 +// 每个文件只用一次 SYSTEM + 一次 buildPrompt，改要求改 BRIEF.md 一处就够。 + +import { readFileSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; + +const BRIEF = readFileSync(resolve(dirname(fileURLToPath(import.meta.url)), 'BRIEF.md'), 'utf8'); + +/** 给模型看的：先说清是谁、译成什么，再把 BRIEF.md 整个给出去 */ +export const system = lang => `You are translating a Chinese life-guide book into ${lang}. The full brief is below; follow every rule in it. + +${BRIEF}`; + +/** 逐个文件的内容；一次一个，别一次塞一整本 */ +export function buildPrompt(locale, kind, source) { + const what = { + book: 'This is one block of entries from a section of the book', + readme: 'This is the book\'s README (the landing page: intro, how to read it, glossary, table of contents, licence)', + docs: 'This is one long-form essay', + }[kind]; + return `${what}. Translate it and output only the translation.\n\n\n${source}\n`; +} diff --git a/tools/i18n/split.mjs b/tools/i18n/split.mjs new file mode 100644 index 00000000..4aba3bb2 --- /dev/null +++ b/tools/i18n/split.mjs @@ -0,0 +1,109 @@ +// 把一节正文切成一块块，好让并行翻译一次一块。 +// +// node tools/i18n/split.mjs book/01-不要早死.md # 切到 .i18n/en/01-*.partNN.md +// node tools/i18n/split.mjs book/01-不要早死.md --locale vi +// node tools/i18n/split.mjs --all --locale en # 整本都切 +// +// 为什么切：一次翻一整节（最大的一节 4 万多字）输出太长，中途断了就得从头再来， +// 而且不好并行。一块十来条，一个翻译任务一块，断了只重跑那一块。 +// +// 切法：第一节头部（回目录那行 + 「# N. 节名」+ 节首引言）只进第 0 块，其余每块若干条， +// 每块自带文件头和文件尾（回目录那行 / 空行）。条目的边界是「### 」， +// 成本标签注释、六个字段行整条跟着走，不会从中间断开。 +// +// 合回来是 join.mjs，它按块号排序拼接，再跑一遍结构检查。 + +import { readFileSync, writeFileSync, mkdirSync, readdirSync, existsSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath, pathToFileURL } from 'node:url'; +import { LOCALES, SOURCE } from '../site/locales.mjs'; + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); + +const argv = process.argv.slice(2); +const localeArg = argv.indexOf('--locale'); +const locale = localeArg < 0 ? 'en' : argv[localeArg + 1]; +if (!LOCALES.some(l => l.code === locale)) { + console.error(`--locale 得是 ${LOCALES.map(l => l.code).join(' / ')} 里的一项`); + process.exit(2); +} +const ENTRIES = 10; // 一块几条。十条差不多 1 万字，输出在一个任务里做得完 + +/** 把一节正文切成头部 + 若干块条目。切点只在条目边界上，不能从一条中间断开 */ +export function splitSection(md, perBlock = ENTRIES) { + const lines = md.split('\n'); + const at = lines.map((l, i) => (/^### /.test(l) ? i : -1)).filter(i => i >= 0); + if (!at.length) throw new Error('这一节里没找到条目（### 开头）'); + const head = lines.slice(0, at[0]); + + // 末尾那行回目录（第 1 节没有，别的地方有）不算进最后一条 + let end = lines.length; + while (end > 0 && !lines[end - 1].trim()) end--; + if (end > 0 && /^\[.*\]\(/.test(lines[end - 1])) end--; + const tailLink = lines.slice(end).join('\n').trim() || null; + + // 每个条目是一段：从它的「### 」到下一个条目的「### 」之前 + const entries = at.map((start, i) => { + const stop = i + 1 < at.length ? at[i + 1] : end; + return lines.slice(start, stop); + }); + + const blocks = []; + for (let i = 0; i < entries.length; i += perBlock){ + blocks.push(entries.slice(i, i + perBlock).flat()); + } + return { head, blocks, tailLink, count: entries.length }; +} + +export function outDir(locale, file) { + return `.i18n/${locale}/${file.replace(/\.md$/, '')}`; +} + +// ---------------------------------------------------------------- 主流程 +// join.mjs 要 import 这里的 splitSection，所以主流程只在直接跑这个脚本时执行。 +// 不加这一层，join 一 import 就把切好的块重切一遍，刚翻好的一块就被原文盖回去。 +if (import.meta.url === pathToFileURL(process.argv[1]).href) await main(); + +async function main(){ +const files = []; +if (argv.includes('--all')) { + const dir = (SOURCE.contentDir ? SOURCE.contentDir + '/' : '') + 'book/'; + for (const f of readdirSync(resolve(ROOT, dir))) if (f.endsWith('.md')) files.push(dir + f); +} else { + for (const a of argv) if (a.endsWith('.md')) files.push(a); +} +if (!files.length) { + console.error('给一个 book/*.md 的路径，或者加 --all'); + process.exit(2); +} + +let made = 0; +for (const f of files){ + const md = read(f); + let cut; + try { cut = splitSection(md); } + catch (err){ console.error(`跳过 ${f}：${err.message}`); continue; } + const dir = outDir(locale, f.split('/').pop()); + mkdirSync(resolve(ROOT, dir), { recursive: true }); + let wrote = 0, kept = 0; + cut.blocks.forEach((b, i) => { + const num = String(i).padStart(2, '0'); + const at = `${dir}/part${num}.md`; + // 已经翻过的块不覆盖。切块本来就是可重复跑的：改参数重切一次， + // 顺手把翻好的几块冲掉就白翻一天（2026-09-30 踩过）。 + if (existsSync(resolve(ROOT, at)) && !process.argv.includes('--force')){ kept++; return; } + // 第 0 块带头部（回目录那行 + 节标题 + 导读），其余块只带条目，合起来才是一节 + const text = (i === 0 ? cut.head.join('\n') + '\n' : '') + b.join('\n') + '\n'; + writeFileSync(resolve(ROOT, at), text); + wrote++; + }); + if (cut.tailLink && !existsSync(resolve(ROOT, `${dir}/tail.txt`))) + writeFileSync(resolve(ROOT, `${dir}/tail.txt`), cut.tailLink + '\n'); + if (kept) console.log(` （${kept} 块已翻过，没动；--force 可覆盖）`); + made += cut.blocks.length; + const kb = s => (s / 1024).toFixed(1) + 'K'; + console.log(`${f} → ${dir}/ ${cut.blocks.length} 块，${cut.count} 条，最大一块 ${kb(Math.max(...cut.blocks.map(b => b.join('\n').length)))}`); +} +console.log(`\n共切出 ${made} 块`); +} diff --git a/tools/i18n/structure.mjs b/tools/i18n/structure.mjs new file mode 100644 index 00000000..d98e9b74 --- /dev/null +++ b/tools/i18n/structure.mjs @@ -0,0 +1,188 @@ +// 一节正文（或一块条目）的结构骨架，以及逐条比对。 +// split.mjs 切块、join.mjs 合块、check.mjs 校整本，用的是同一套解析和同一套比对， +// 所以「块级过、整节也不过」这种事不会发生。 +// +// 骨架只留机器能查的东西：条号、六个字段各有无、数字、链接、指路的条号节号、 +// 成本标签注释、证据等级、争议和待核实的标记。文字一个字都不管——译文和原文 +// 本来就该不一样，管文字就管死了。 + +import { strings, SOURCE, LOCALES } from '../site/locales.mjs'; + +const escRe = s => s.replace(/[.*+?^${}()|[\]\\]/g, '\\$&'); +const FIELDS = ['cost', 'human', 'gain', 'grade', 'src', 'note']; +// 原文与译文都必须有的字段。「说人话」「来源」「备注」有些条目本来就没有， +// 原文缺译文也缺才算过（比的时候按实际有的比，不按固定六栏） +const ALWAYS = ['cost', 'gain', 'grade']; + +/** + * 数字：指路里的条号节号不算，那些单独查。千分位逗号去掉再比，248,099 和 248099 一样。 + * + * 量级词单独处理。中文在英文里只能换个写法：「每万人 30.8 人」写成 "30.8 per 10,000"， + * 「167 万女性」写成 "1,670,000 women"，「每百万人 302 例」写成 "302 per million"。 + * 这些不是改了数字，是量级本身换了个写法。所以返回两份：必须要有的（ns）， + * 和量级换算时允许出现的（scale，可多可少）。百、千、万、亿四种都算。 + */ +const SCALE_WORDS = [ + [1e2, '百'], [1e3, '千'], [1e4, '万'], [1e5, '十万'], [1e6, '百万'], [1e7, '千万'], [1e8, '亿'], +]; +const SCALE_VALUES = SCALE_WORDS.map(([v]) => String(v)); + +function numbersOf(text) { + const s = String(text); + const norm = n => n.replace(/[.,](?=\d{3}\b)/g, ''); + const ns = [...s.matchAll(/\d+(?:[.,]\d+)*/g)].map(m => norm(m[0])); + + // 只有这一栏里出现过量级词，才允许出现量级换算的结果 + if (!/[百千万亿]/.test(s)) return { ns: ns.sort(), scale: new Set() }; + const scale = new Set(SCALE_VALUES); + // 「167 万」「每万枚」「每百万人」两种语序都算：数在词前，词在数前 + for (const [v, word] of SCALE_WORDS){ + for (const m of s.matchAll(new RegExp(`(\\d+(?:[.,]\\d+)*)\\s*${word}`, 'g'))){ + const raw = norm(m[1]); + const base = Number(raw); + if (!Number.isFinite(base) || base === 0) continue; + // 尾数可以拆开重组：中文「2.21 亿元」在英文里写成 "221 million yuan" + // 是同一个数（2.21 × 1e8 = 221 × 1e6），不算改数字。所以 ±3 到 +8 个 + // 数量级都放进允许名单，超出这个范围的还是真改了。 + for (let k = -3; k <= 8; k++){ + const x = base * Math.pow(10, k); + if (Number.isFinite(x) && Math.abs(x) >= 1) scale.add(String(Number(x.toPrecision(12)))); + } + scale.add(raw); + const i = ns.indexOf(raw); + if (i >= 0) ns.splice(i, 1); // 基数本身不再必须出现 + } + } + return { ns: ns.sort(), scale }; +} + +/** + * 指路：条号、节号，以及「8 到 10」这种区间。 + * 正则是各语言自己那份（tools/site/locales/.json 的 xref.pattern），和检索页 + * 把「第 X 条」变成可点的引用用的是同一套。所以这里查的就是读者真能点到的那些指路， + * 译文里条号节号丢了或者写不成可点的形状，这一层就报出来。 + */ +function refsOf(text, xref) { + if (!xref) return []; + const range = new RegExp(xref.rangeRe); + const sep = new RegExp(xref.listSep); // 正则，不是字面串：英文是 ", and to –" 一串 + const out = []; + for (const m of text.matchAll(new RegExp(xref.pattern, 'g'))){ + const g = m.groups ?? {}; + if (g.secs !== undefined){ + for (const s of String(g.secs).split(sep)) if (s.trim()) out.push('s' + s.trim()); + continue; + } + // 节号缺省就是本节：中文「本节第 3 条」、英文「this section's item 3」 + const sec = g.sec ? 's' + g.sec : '*'; + for (const part of String(g.nums ?? g.nums2 ?? g.nums3 ?? '').split(sep)){ + const t = part.trim(); + if (!t) continue; + const r = range.exec(t); + out.push(r ? `${sec}r${r[1]}-${r[2]}` : `${sec}${t}`); + } + } + return out.sort(); +} + +/** + * 解析一节正文。head 是节标题和导读，entries 是条目。 + * 字段名从 locale 取：中文「- 成本：」，英文「- Cost:」，越南文「- Chi phí:」。 + */ +export function entrySkeleton(md, S = strings(SOURCE.code)) { + const lines = md.split('\n'); + const fieldRe = Object.fromEntries(FIELDS.map(f => [f, new RegExp('^- ' + escRe(S.fields[f]) + '(.*)$')])); + const out = { n: '', title: '', intro: [], entries: [] }; + let entry = null; + for (const raw of lines){ + const line = raw.replace(/\s+$/, ''); + let m; + if (!entry && (m = /^# (\d+)\. (.+)$/.exec(line))){ out.n = m[1]; out.title = m[2].trim(); continue; } + if ((m = /^### (\d+)\. (.+)$/.exec(line))){ + entry = { n: m[1], title: m[2].trim(), tag: '', fields: {}, extra: [] }; + out.entries.push(entry); + continue; + } + if (!entry){ if (line.trim()) out.intro.push(line); continue; } + if ((m = /^$/.exec(line))){ entry.tag = m[1].trim(); continue; } + let hit = false; + for (const f of FIELDS){ + if ((m = fieldRe[f].exec(line))){ entry.fields[f] = m[1]; hit = true; break; } + } + if (hit){ + if (entry.fields.grade) entry.grade = entry.fields.grade.trim()[0]; + continue; + } + if (line.startsWith('- ')) entry.extra.push(line); + } + for (const e of out.entries) decorate(e, S); + return out; +} + +function decorate(e, S) { + const all = [e.fields.human ?? '', e.fields.gain ?? '', e.fields.note ?? '', e.fields.src ?? ''].join('\n'); + e.numbers = { + gain: numbersOf(e.fields.gain ?? ''), + cost: numbersOf(e.fields.cost ?? ''), + src: numbersOf(e.fields.src ?? ''), + }; + e.refs = refsOf(all, S.xref); + e.links = all.match(/https?:\/\//g)?.length ?? 0; + e.order = FIELDS.filter(f => e.fields[f] !== undefined); + e.dispute = new RegExp('^' + S.dispute).test(e.fields.note ?? ''); + e.todo = S.todo.some(w => all.includes(w) || (e.fields.cost ?? '').includes(w)); +} + +/** 两种语言的字段名在页面上叫什么，报告里好读 */ +function label(S, f) { return S.fields[f].replace(/[：:]$/, ''); } + +/** + * 逐条比。返回问题清单（空数组就是这块对了）。dstLabel 是给报告前缀用的语言名。 + * 这一套就是 check.mjs 的判据，抽出来是为了 join 逐块校时用同一份。 + */ +export const STRUCT = { + ALWAYS, FIELDS, + compare(a, b, tag) { + const out = []; + const S = strings(SOURCE.code); + const D = strings(b._locale ?? SOURCE.code); + if (a.n !== b.n) out.push(`${tag}：条号是 ${b.n}，原文 ${a.n}`); + if (!b.title.trim()) out.push(`${tag}：标题是空的`); + + // 成本标签逐字照抄：检索页的筛选和地址栏全靠它，三种语言共用一套中文取值 + if (a.tag !== b.tag) out.push(`${tag}：成本标签「${b.tag}」和原文「${a.tag}」不一样，要一字不改照抄`); + + for (const f of ALWAYS) if (b.fields[f] === undefined) out.push(`${tag}：少了「${label(S, f)}」这一栏`); + for (const f of ['human', 'src', 'note']){ + const had = a.fields[f] !== undefined, has = b.fields[f] !== undefined; + if (had !== has) out.push(`${tag}：原文${had ? '有' : '没有'}「${label(S, f)}」栏，译文${has ? '有' : '没有'}`); + } + if (b.extra.length) out.push(`${tag}：多出原文没有的字段行「${b.extra[0].slice(0, 30)}」`); + if (b.order.join() !== a.order.join()) + out.push(`${tag}：字段顺序是 ${b.order.map(f => label(D, f)).join('、')}，原文 ${a.order.map(f => label(S, f)).join('、')}`); + + if (a.grade !== b.grade) out.push(`${tag}：证据等级是 ${b.grade ?? '空'}，原文 ${a.grade}`); + if (a.dispute !== b.dispute) out.push(`${tag}：${a.dispute ? '原文标了争议，译文没标' : '译文标了争议，原文没有'}`); + if (a.todo !== b.todo) out.push(`${tag}：${a.todo ? '原文有「待核实」，译文没有' : '译文有「待核实」，原文没有'}`); + + // 数字：收益、成本、来源三栏逐个对。全书可核对性的底座，一个都不许动 + // 数字：收益、成本、来源三栏逐个对。全书可核对性的底座，一个都不许动。 + // 备注是散文，「40 例」这类个数和量级本来就随语言变（「several dozen」）， + // 数字不许动的是前三个栏（CLAUDE.md 的原话：收益栏是全书可核对性的底座）。 + // 唯一的例外是量级换算：源文「167 万」在英文里只能写成 1,670,000，那不算改了数字。 + for (const f of ['gain', 'cost', 'src']){ + const A = a.numbers[f], B = b.numbers[f]; + const allowed = new Set([...A.ns, ...A.scale]); + const miss = [...new Set(A.ns.filter(x => !B.ns.includes(x)))]; + const extra = [...new Set(B.ns.filter(x => !allowed.has(x)))]; + if (miss.length) out.push(`${tag}：${label(D, f)}栏少了原文的数字 ${miss.join('、')}`); + if (extra.length) out.push(`${tag}：${label(D, f)}栏多了原文没有的数字 ${extra.join('、')}`); + } + + for (const r of a.refs) if (!b.refs.includes(r)) out.push(`${tag}：指路「第 ${r}」不见了，条号节号要照抄`); + if (a.links !== b.links) out.push(`${tag}：来源和备注里有 ${b.links} 个链接，原文 ${a.links} 个`); + return out; + }, +}; + +export { FIELDS, numbersOf, refsOf, LOCALES }; diff --git a/tools/lib/book.mjs b/tools/lib/book.mjs index ac77685c..6d9f4f32 100644 --- a/tools/lib/book.mjs +++ b/tools/lib/book.mjs @@ -4,10 +4,14 @@ import { readFileSync } from 'node:fs'; import { resolve, dirname } from 'node:path'; import { fileURLToPath } from 'node:url'; import { execSync } from 'node:child_process'; +import { SOURCE, LOCALE, strings } from '../site/locales.mjs'; export const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); -export const REPO = 'https://github.com/eternity4719/HowToLiveBetter'; -export const SITE = 'https://eternity4719.github.io/HowToLiveBetter/'; +// 站点地址和仓库地址只有一处：tools/site/site.json。EPUB、PDF、离线单文件三套构建 +// 生成的页脚和下载链接都从这儿来，换域名只改那一处，别在这里再抄一份。 +const urls = JSON.parse(readFileSync(resolve(ROOT, 'tools/site/site.json'), 'utf8')); +export const REPO = urls.repo; +export const SITE = urls.site + '/'; export const TITLE = '高性价比人生指南'; export const RELEASE = `${REPO}/releases/download/epub-latest`; @@ -35,23 +39,44 @@ export function stripBackLink(md) { return md.replace(/^\[← 回总目录\]\([^)]*\)\s*\n/, ''); } -// README 里从某个标题到下一个标题之间的一段 -export function readBook() { - const readme = read('README.md'); +/** + * README 里从某个标题到下一个标题之间的一段。 + * 三个标题名各语言不同（'# 高性价比人生指南' / '# A High-Return Life Guide'），都从 + * locale 配置里来，不在这里写死——同一份代码要同时读三份 README。 + * 找不到就退回到「第一个一级标题到第一个徽章」：各语言的标题写法不一样，硬找会脆。 + */ +function between(lines, from, to) { + const a = lines.findIndex(l => l.startsWith(from)); + const b = lines.findIndex((l, i) => i > a && l.startsWith(to)); + if (a < 0 || b < 0) throw new Error(`README 里找不到 ${from} 到 ${to} 这一段`); + return lines.slice(a, b).join('\n'); +} + +/** + * 读一份 README，拆出电子书两套构建要的东西。 + * locale —— 语言码；给了就读那一份（/README.md），不给读仓库根的中文原文 + * 目录、标题这些位置按该语言的配置找，不维护第二份文件名单：新增一节或一篇长文， + * EPUB、PDF 两套构建都自动跟上。 + */ +export function readBook(locale = null) { + const L = locale ? LOCALE(locale) : SOURCE; + const S = strings(L.code); + const md = L.contentDir ? L.contentDir + '/README.md' : 'README.md'; + const readme = read(md); const lines = readme.split('\n'); - const between = (from, to) => { - const a = lines.findIndex(l => l.startsWith(from)); - const b = lines.findIndex((l, i) => i > a && l.startsWith(to)); - if (a < 0 || b < 0) throw new Error(`README 里找不到 ${from} 到 ${to} 这一段`); - return lines.slice(a, b).join('\n'); - }; - const description = between('# 高性价比人生指南', '[![') - .split('\n').slice(1).map(l => l.replace(/<[^>]+>/g, '').trim()).filter(Boolean).join(''); - const frontMd = between('## 这本书想回答的问题', '## 目录'); - const contentsMd = between('## 目录', '## 正文') + + // 书名标题：那一行以 '# ' 开头且不是 '## '，全篇只有一处 + const h1 = lines.findIndex(l => /^# \S/.test(l)); + const badge = lines.findIndex(l => l.startsWith('[![')); + if (h1 < 0 || badge < 0 || badge < h1) throw new Error(`${md} 里的书名标题或徽章没找到`); + const description = lines.slice(h1 + 1, badge) + .map(l => l.replace(/<[^>]+>/g, '').trim()).filter(Boolean).join(''); + + const frontMd = between(lines, '## ' + S.contentsQuestion, '## ' + S.contents); + const contentsMd = between(lines, '## ' + S.contents, '## ' + S.text) .split('\n\n').filter(p => !p.includes('index.html')).join('\n\n'); const bookFiles = unique([...contentsMd.matchAll(/\]\((book\/[^)#]+\.md)\)/g)].map(m => m[1])); const docFiles = unique([...readme.matchAll(/\]\((docs\/[^)#/]+\.md)\)/g)].map(m => m[1])); - if (bookFiles.length === 0) throw new Error('README 目录里没找到 book/ 文件'); - return { readme, description, frontMd, contentsMd, bookFiles, docFiles }; + if (bookFiles.length === 0) throw new Error(`${md} 的目录里没找到 book/ 文件`); + return { readme, description, frontMd, contentsMd, bookFiles, docFiles, locale: L.code, readmePath: md }; } diff --git a/tools/offline/build.mjs b/tools/offline/build.mjs index f128b558..e929406c 100644 --- a/tools/offline/build.mjs +++ b/tools/offline/build.mjs @@ -1,34 +1,63 @@ -// 把 index.html + README + book/*.md 打成一个自包含的 HTML：双击就能看，不用服务器、不用联网。 -// 用法：node tools/offline/build.mjs [输出路径] 默认输出 dist/HowToLiveBetter.html -// 正文内联进 window.__CORPUS__，index.html 的 init() 认这个变量就不再发请求； +// 把一页 + 那一份语言的 README + book/*.md 打成一个自包含的 HTML：双击就能看，不用服务器、不用联网。 +// +// node tools/offline/build.mjs # 默认语言（en） +// node tools/offline/build.mjs --locale vi +// node tools/offline/build.mjs zh # 位置参数也认 +// node tools/offline/build.mjs --locale vi dist/HowToLiveBetter-vi.html +// +// 页面是 tools/site/page.template.html + tools/site/locales/.json 经 +// tools/site/build.mjs 生成的（tools/site/build.mjs 的 renderPage），不是拿哪一份生成物改的—— +// 生成物入库是为了让 Pages 直接发，脚本直接从源头渲，别反过来依赖生成物。 +// 正文内联进 window.__CORPUS__，页面的 init() 认这个变量就不再发请求； // 站内相对链接改成线上地址，侧栏图片转成 data URI，其余一个字不动。 -import { readFileSync, writeFileSync, mkdirSync } from 'node:fs'; +import { readFileSync, writeFileSync, mkdirSync, existsSync } from 'node:fs'; import { resolve, dirname } from 'node:path'; import { ROOT, REPO, SITE, read, gitCommit, buildStamp } from '../lib/book.mjs'; +import { LOCALES, SOURCE } from '../site/locales.mjs'; +import { renderPage } from '../site/build.mjs'; -const OUT = resolve(ROOT, process.argv[2] ?? 'dist/HowToLiveBetter.html'); +const argv = process.argv.slice(2); +const li = argv.indexOf('--locale'); +const code = (li < 0 ? null : argv[li + 1]) ?? argv.find(a => !a.startsWith('-') && LOCALES.some(l => l.code === a)) ?? null; +const L = LOCALES.find(l => l.code === code) ?? LOCALES.find(l => l.default); +const outArg = argv.find((a, i) => !a.startsWith('-') && a !== L.code && !(li >= 0 && i === li + 1)); + +const OUT = resolve(ROOT, outArg ?? `dist/HowToLiveBetter${L.code === (LOCALES.find(l => l.default)?.code) ? '' : '-' + L.code}.html`); const STAMP = buildStamp(); const COMMIT = gitCommit(); +// 路径拼法统一走 at()：zh 的 contentDir 是空串，直接拼会得到 '/book/x.md'， +// resolve 把它当绝对路径，跑到文件系统根上找去了。 +const at = p => (L.contentDir ? L.contentDir + '/' : '') + p; +const stats = JSON.parse(read('tools/site/stats.json')); + // ---------- 正文 ---------- -const readme = read('README.md'); -const files = [...new Set([...readme.matchAll(/\]\((book\/[^)]+\.md)\)/g)].map(m => m[1]))].sort(); -if (!files.length) throw new Error('README 目录里没找到 book/ 文件，离线版会是空的'); +const readme = read(at('README.md')); +// 清单从 README 的目录里扒，和检索页是同一处。翻到一半的那几节正文还没有， +// 这里跳过而不是打错：清单和页面都得是「实际存在的那几节」，两份对不上就是 bug +const listed = [...new Set([...readme.matchAll(/\]\((book\/[^)]+\.md)\)/g)].map(m => m[1]))].sort(); +if (!listed.length) throw new Error(`${at('README.md')} 的目录里没找到 book/ 文件，离线版会是空的`); +const files = listed.filter(f => existsSync(resolve(ROOT, at(f)))); +const absent = listed.filter(f => !existsSync(resolve(ROOT, at(f)))); +if (!files.length) throw new Error(`${at('README.md')} 的目录里列了 ${listed.length} 节，正文一份都还没有`); +if (absent.length) console.log(`另有 ${absent.length} 节还没翻到，这次没进离线副本：${absent.slice(0, 5).join('、')}${absent.length > 5 ? ' …' : ''}`); // 长文（docs/*.md）也要带上：检索页的长文弹窗就地渲染它们，离线副本里没有就只剩 // 一个点不开的 GitHub 链接。清单从 README 里扒，和 EPUB、PDF 两套构建用的是同一处。 -const docs = [...new Set([...readme.matchAll(/\]\((docs\/[^)#/]+\.md)\)/g)].map(m => m[1]))].sort(); +const docs = [...new Set([...readme.matchAll(/\]\((docs\/[^)#/]+\.md)\)/g)].map(m => m[1]))] + .sort() + .filter(f => existsSync(resolve(ROOT, at(f)))); const corpus = { readme, - parts: Object.fromEntries(files.map(f => [f, read(f)])), - docs: Object.fromEntries(docs.map(f => [f, read(f)])), + parts: Object.fromEntries(files.map(f => [f, read(at(f))])), + docs: Object.fromEntries(docs.map(f => [f, read(at(f))])), }; // { - if (!html.includes(needle)) throw new Error(`index.html 里找不到${label}，离线版脚本要跟着改：${needle}`); + if (!html.includes(needle)) throw new Error(`生成的 ${L.dir}index.html 里找不到${label}，离线版脚本要跟着改：${needle}`); }; // 统计脚本不能跟着离线版走：别人双击打开的副本不该往外发请求，断网时还要等超时 @@ -39,33 +68,38 @@ html = html.slice(0, html.indexOf(GA_START)) + html.slice(html.indexOf(GA_END) + // 只查外连域名：主脚本里的 track() 带 typeof 守卫，没有 gtag 也能跑，不算残留 if (/googletagmanager|google-analytics/.test(html)) throw new Error('剥掉标记之间的内容后仍有统计域名残留，离线版会往外发请求'); +// 离线副本没有目录可以跳，语言切换得藏起来（SITE.offline 就在 build.mjs 里置上） +html = html.replace(/"offline":false/, '"offline":true'); + // 相对链接在本地打开时是死的，改成线上地址 -must('href="README.md"', ' README.md 链接'); -must('href="book/"', ' book/ 链接'); +must(`href="${L.contentBase}README.md"`, ' README.md 链接'); +must(`href="${L.contentBase}book/"`, ' book/ 链接'); html = html - .replaceAll('href="README.md"', `href="${REPO}/blob/main/README.md"`) - .replaceAll('href="book/"', `href="${REPO}/tree/main/book"`) - .replaceAll('`, ``); // 侧栏广告图和赞赏码转 data URI，否则离线打开是个裂图 for (const [img, mime] of [['ads/mcyyy-side.webp', 'image/webp'], ['ads/wechat-reward.png', 'image/png']]) { - must(`src="${img}"`, `图片 ${img}`); + must(`src="${L.assetBase}${img.replace('ads/', '')}"`, `图片 ${img}`); const data = readFileSync(resolve(ROOT, img)).toString('base64'); - html = html.replace(`src="${img}"`, `src="data:${mime};base64,${data}"`); + html = html.replace(`src="${L.assetBase}${img.replace('ads/', '')}"`, `src="data:${mime};base64,${data}"`); } // 页脚注明这是哪一版的离线副本 const foot = '
'; must(foot, '页脚'); const commitNote = COMMIT ? `，正文提交 ${COMMIT.slice(0, 7)}` : ''; -html = html.replace(foot, `${foot}离线副本，生成于 ${STAMP}（北京时间）${commitNote}；正文会继续更新，以 在线版 为准。
`); +const noteText = `${L.code === SOURCE.code ? '' : `${L.name} 版，`}离线副本，生成于 ${STAMP}（北京时间）${commitNote}；正文会继续更新，以 在线版 为准。`; +html = html.replace(foot, `${foot}${noteText}
`); -// 正文要在主脚本之前就位 -const mainScript = '\n${mainScript}`); mkdirSync(dirname(OUT), { recursive: true }); writeFileSync(OUT, html); -const kb = n => (n / 1024 | 0) + ' KB'; -console.log(`已生成 ${OUT}：${files.length} 个正文文件，长文 ${docs.length} 篇，${kb(Buffer.byteLength(html))}（其中正文 ${kb(Buffer.byteLength(corpusJson))}）`); +const kb = n => (Buffer.byteLength(n) / 1024 | 0) + ' KB'; +console.log(`已生成 ${OUT}（${L.name}）：${files.length} 个正文文件，长文 ${docs.length} 篇，${kb(html)}（其中正文 ${kb(corpusJson)}）`); diff --git a/tools/site/build.mjs b/tools/site/build.mjs new file mode 100644 index 00000000..dcf81cc1 --- /dev/null +++ b/tools/site/build.mjs @@ -0,0 +1,467 @@ +// 生成检索页（en/index.html、zh/index.html、vi/index.html）、根目录跳语言的小页、 +// sitemap.xml 和 robots.txt。没翻完的语言这次不生成（眼下是 vi），页面上也不列它， +// 免得给读者一个点进去是 404 的入口。 +// +// node tools/site/build.mjs # 生成 +// node tools/site/build.mjs --check # 只比对，有一处对不上就退出码 1（CI 用） +// +// 页面本体在 tools/site/page.template.html，界面文案在 tools/site/locales/.json， +// 目录和正文在各自语言下（en/README.md + en/book/、仓库根的 README.md + book/、vi/…）。 +// 站点地址和仓库地址在 tools/site/site.json，别在别处再抄一遍。 +// 生成物入库，Pages 直接发；改页面改模板和字典，别改生成出来的那几份。 +// +// 有两类文案，别弄混： +// · 界面文案（strings）—— 按钮、标签、报错，按语言写。 +// · 正文标记 —— 条目字段名「- 成本：」、成本标签注释、「第 X 条」指路的正则、争议和待核实 +// 的开头。这些在 locale.json 顶层，不在 strings 里：译文按它们写，但取值（钱=0|少|多、 +// 大|中|小、死亡率|金钱|时间|自由）三种语言共用一份，地址栏和分享出去的链接才对得上。 +import { readFileSync, writeFileSync, mkdirSync, existsSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; + +const HERE = dirname(fileURLToPath(import.meta.url)); +const ROOT = resolve(HERE, '../..'); +const CHECK = process.argv.includes('--check'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); + +// 目录和正文位置。zh 指仓库根（中文是正文源头，不能搬走），en/vi 指各自的子目录。 +export const LOCALES = JSON.parse(read('tools/site/locales.json')); +export const DEFAULT_LOCALE = LOCALES.find(l => l.default)?.code ?? LOCALES[0].code; + +// 正则里的元字符要转义，不然字段名里的括号和问号会被当成语法 +const escRe = s => s.replace(/[.*+?^${}()|[\]\\]/g, '\\$&'); + +// ---------------------------------------------------------------- 键的收集 +// 静态键从模板里扫（data-i18n / T('x')），动态键在这里列全：拼出来的键扫不到， +// 少一个就是页面上一个键名（tools/site/build.mjs 会拦，页面本身不会崩）。 +const DYNAMIC_KEYS = [ + 'ratio.top', 'ratio.high', 'ratio.mid', // T('ratio.' + 档位) + 'gain.large', 'gain.medium', 'gain.small', // T('gain.' + 收益量级) + 'cost.money.0', 'cost.money.few', 'cost.money.much', + 'cost.time.few', 'cost.time.mid', 'cost.time.much', + 'cost.will.no', 'cost.will.some', 'cost.will.lots', +]; + +export function requiredKeys(template) { + const keys = new Set([ + ...[...template.matchAll(/data-i18n(?:-content|-attr|-attr2)?="([\w.]+)"/g)].map(m => m[1]), + // T('cost.' + dim + '.' + k) 这种拼出来的键扫不到前缀，跳过，完整的键在 DYNAMIC_KEYS 里； + // 'key' 是 T() 自己漏译时回显的那句，不是键 + ...[...template.matchAll(/\bT\('([\w.]+)'/g)].map(m => m[1]).filter(k => !k.endsWith('.') && k !== 'key'), + ]); + for (const k of DYNAMIC_KEYS) keys.add(k); + return [...keys].sort(); +} + +// ---------------------------------------------------------------- 占位符 +// strings 里可以写 {{entries}}、{{sections}} 这类占位符，取自 tools/site/stats.json，在构建时填。 +// 同一套尖括号也用来写 T('key', {n}) 的运行时变量（{{n}}、{{sec}}…），那些留到浏览器里替换， +// 所以只换 stats.json 里有的那几个。填完还有没被替换的，就是漏了变量，最后统一查一遍。 +export function fill(text, stats) { + if (Array.isArray(text)) return text.map(t => fill(t, stats)); // meta.about 是个词表 + return String(text).replace(/\{\{(\w+)\}\}/g, (m, k) => (k in stats ? stats[k] : m)); +} + +// ---------------------------------------------------------------- 标签扫描 +// 只认开始标签，够用了：模板里 data-i18n 那批元素没有一个套着另一个 data-i18n 元素， +// 有的话也不会互相吃内容（内层先替，外层再按剩下的空内容替）。 +const TOKENS = /<[^>]*>|[^<]+/g; + +function applyMarkup(html, strings, blocks) { + const toks = html.match(TOKENS) ?? []; + // 先把 data-i18n-block 的范围整块换掉：位置变了，后面按位置找就全错了 + const replaced = []; + for (let i = 0; i < toks.length; i++){ + const open = toks[i]; + const m = /^<([a-z0-9]+)\b[^>]*\bdata-i18n-block="(\w+)"[^>]*>/i.exec(open); + if (!m) continue; + const [, tag, key] = m; + if (!(key in blocks)) throw new Error(`模板里 data-i18n-block="${key}"，构建没有给这个块的内容`); + const depth = countClose(toks, i, tag); + replaced.push({ from: i, to: depth, text: blocks[key] }); + } + for (let r = replaced.length - 1; r >= 0; r--){ + const { from, to, text } = replaced[r]; + toks.splice(from + 1, to - from - 1, text); // 只换内文，外面的开闭标签留着 + } + + for (let i = 0; i < toks.length; i++){ + const tok = toks[i]; + if (!tok.startsWith('<')) continue; + const m = /^<([a-z0-9]+)\b([^>]*)>/i.exec(tok); + if (!m) continue; + const [, tag, attrs] = m; + const i18n = /\bdata-i18n="([\w.]+)"/.exec(attrs); + const content = /\bdata-i18n-content="([\w.]+)"/.exec(attrs); + if (attrs.includes('data-i18n-attr') || attrs.includes('data-i18n-attr2')){ + // 「placeholder:key;aria-label:key2」写成一段，改同名属性；template 里给的是空串 + for (const [attrName, raw] of [[/data-i18n-attr="([^"]+)"/, attrs], [/data-i18n-attr2="([^"]+)"/, attrs]]){ + const am = attrName.exec(attrs); + if (!am) continue; + let tag2 = tok; + for (const pair of am[1].split(';')){ + const [name, key] = pair.split(':'); + const value = strings[key]; + if (value === undefined) throw new Error(`字典里缺 ${key}`); + const re = new RegExp(`(\\s${name}=")[^"]*(")`); + if (!re.test(tag2)) throw new Error(`模板里 ${name}="${key}" 这处属性没写出来`); + tag2 = tag2.replace(re, (_, a, b) => a + escAttr(value) + b); + } + toks[i] = tag2; + } + } + if (content){ + // ：换的是这个 content，不是别处的。同一个标签只会有一个 content + const value = strings[content[1]]; + if (value === undefined) throw new Error(`字典里缺 ${content[1]}`); + if (!/\bcontent="[^"]*"/.test(toks[i])) throw new Error(`模板里 data-i18n-content="${content[1]}" 这处没有 content 属性`); + toks[i] = toks[i].replace(/\bcontent="[^"]*"/, `content="${escAttr(value)}"`); + } + if (i18n){ + const value = strings[i18n[1]]; + if (value === undefined) throw new Error(`字典里缺 ${i18n[1]}`); + const end = countClose(toks, i, tag); + toks.splice(i + 1, end - i - 1, value); + i++; + } + } + return toks.join(''); +} +const escAttr = s => s.replace(/&/g, '&').replace(/"/g, '"'); +// 拼仓库根相对路径。zh 的 contentDir 是空串，直接拼会得到 '/book/x.md'， +// resolve 把它当绝对路径，就跑到文件系统根上找去了。 +const at = (L, p) => (L.contentDir ? L.contentDir + '/' + p : p); +const escText = s => s.replace(/&/g, '&').replace(//g, '>'); + +// 从开始标签 i 起数它的配对结束标签，返回结束标签的位置 +function countClose(toks, i, tag) { + const re = new RegExp(`^$`, 'i'); + let depth = 0; + for (let j = i + 1; j < toks.length; j++){ + const t = toks[j]; + if (t.startsWith(' 没有配对的结束标签`); +} + +// ---------------------------------------------------------------- 正文清单 +// README 的「目录」一节就是节清单，noscript 的兜底列表由它生成：多一处要维护的地方， +// 中文那版就是从这一节生成的，两边永远一致。 +/** + * 目录那一节。节标题和表头里「长文」那一格的叫法各语言不同（目录 / Contents / Mục lục， + * 长文 / Long-form / Bài dài），都从 locale 配置里来，不在这里写死。 + * 用配置而不是英文，是因为同一份代码要同时读三份 README。 + */ +export function readContents(md, S) { + const lines = md.split('\n'); + const esc = s => s.replace(/[.*+?^${}()|[\]\\]/g, '\\$&'); + const at = lines.findIndex(l => new RegExp('^##\\s+' + esc(S.contents)).test(l)); + if (at < 0) throw new Error(`README 里找不到「${S.contents}」那一节`); + const rest = lines.slice(at + 1); + const end = rest.findIndex(l => /^##\s/.test(l)); + const body = end < 0 ? rest : rest.slice(0, end); + // 目录行尾那句「长文见 [docs/x.md](docs/x.md)。」在页头那行已经单独链了，目录里这句摘掉 + const tail = new RegExp('\\s*' + esc(S.longformSee) + '\\s*\\[[^\\]]+\\]\\([^)]+\\)[^ ]*'); + const out = []; + for (const line of body){ + const m = /^\s*\d+\.\s+\[([^\]]+)\]\(([^)]+)\)\s*[：:,]\s*(.*)$/.exec(line) ?? /^\s*\d+\.\s+\[([^\]]+)\]\(([^)]+)\)\s*(.*)$/.exec(line); + if (!m) continue; + out.push({ n: out.length + 1, title: m[1], path: m[2], blurb: m[3].replace(tail, '').trim() }); + } + return out; +} + +// 页头那行长文：README 顶部导航表里「长文」那一格，EPUB 和 PDF 也从同一处扒 +export function readLongform(md, S) { + const cell = new RegExp(`${S.longformLabel}<\\/b><\\/td>]*>([\\s\\S]*?)<\\/td>`).exec(md); + if (!cell) throw new Error(`README 顶部导航表里找不到「${S.longformLabel}」那一格`); + // 连链接文字一起扒（译文里那句是该语言的书名）。只拿路径的话页面上露出的是中文文件名 + return [...cell[1].matchAll(/\[([^\]]+)\]\((docs\/[^)#/]+\.md)\)/g)].map(m => ({ path: m[2], title: m[1] })); +} + +const pctEncode = p => p.split('/').map(encodeURIComponent).join('/'); + +// 下面几个函数都只吐内文：data-i18n-block 换的是内文，外层标签模板里已经有了。 +// 原来连外层标签一起吐，页面上出来的是
。 +function noscriptList(contents, L, strings, S) { + // 「标题：说明」中间那个分隔符，各语言不一样 + return contents.map(c => `
  • ${escText(c.title)}${c.blurb ? S.listSep + escText(c.blurb) : ''}
    • `).join('\n'); +} + +function docLinks(docs, L, strings, S) { + const links = docs.map(f => + `${escText(f.title)}`); + // 核实记录是作者的工作底稿，没有译本，各语言都指中文那一份 + links.push(`${escText(strings['docs.records'])}`); + return `${escText(strings['docs.label'])}${S.listSep}${links.join(' · ')}`; +} + +// 根目录那个小页：默认英文，浏览器要越南语就去 /vi/，也可以自己点。关掉 JS 也点得动。 +// 只列 PUBLISHED：越南语那份还没翻完，列出来点进去是 404。 +function chooserPage(site) { + const list = PUBLISHED + .map(l => `
  • ${escText(l.name)}${escText(l.note)}
    • `) + .join('\n'); + const pick = ` const q = new URLSearchParams(location.search).get('lang'); + const list = ${JSON.stringify(PUBLISHED.map(l => ({ code: l.code, url: l.canonical, test: new RegExp('^' + l.accept, 'i') })))}; + const hit = q ? list.find(l => l.code === q) : list.find(l => l.test.test(navigator.language || '')); + location.replace((hit || list[0]).url + location.hash);`; + return ` + + + + +高性价比人生指南 · Choose your language + + + + + + + + + +
    +

    高性价比人生指南

    +

    选一种语言 · Choose a language

    +
      +${list} +
    +
    + + + +`; +} + +// ---------------------------------------------------------------- 渲染一份页面 +// 这一轮真的会生成页面的语言：README 在的才算。还没翻完的语言（眼下是 vi） +// 不能出现在 hreflang、跳语言页和页内语言下拉里——那些都是读者会点开的链接， +// 点进去只有一个 404，比不列更糟。翻完重跑这个脚本，它自己回来。 +const readmeOf = L => (L.contentDir ? L.contentDir + '/' : '') + 'README.md'; +const PUBLISHED = LOCALES.filter(L => existsSync(resolve(ROOT, readmeOf(L)))); + +// hreflang 由 PUBLISHED 生成，不在模板里写死：写死了就得记得在加语言、改域名时两处一起改， +// 漏一处就是一堆指向不存在页面的 hreflang。 +function hreflangLinks() { + return [ + ...PUBLISHED.map(l => ``), + ``, + ].join('\n'); +} + +export function renderPage(L, template, stats) { + // 内容还没翻的语言这次先跳过：zh 那一份照发，翻完再带上 + const readme0 = readmeOf(L); + // 按 code 比，不比对象：L 有两份来源（build.mjs 自己 parse locales.json， + // 别的脚本从 tools/site/locales.mjs 拿），是同一份 JSON 的两个不同对象， + // 用 includes() 比永远不相等，离线版和 PDF 构建就都拿到 null 了。 + if (!PUBLISHED.some(p => p.code === L.code)){ SKIP.add(L.code); return null; } + + const raw = JSON.parse(read(`tools/site/locales/${L.code}.json`)); + if (!raw.strings) throw new Error(`locales/${L.code}.json 里没有 strings`); + const missing = requiredKeys(template).filter(k => raw.strings[k] === undefined); + if (missing.length) throw new Error(`locales/${L.code}.json 缺 ${missing.length} 个键：${missing.join('、')}`); + + const strings = Object.fromEntries(Object.entries(raw.strings).map(([k, v]) => [k, fill(v, stats)])); + // 内容根目录由 locales.json 定：zh 指仓库根（原文在那儿），en/vi 指自己的子目录。 + // 找不到就是还没翻，不是可以退回去读中文——退回去会造出一个内容对不上的页面 + const readmePath = readme0; + const readme = read(readmePath); + const listed = readContents(readme, raw); + if (!listed.length) throw new Error(`${readmePath} 的「${raw.contents}」一节里没扒到节`); + // 目录里列了、正文还没翻到的那几节，这次先不收进页面。全书翻完会自动全进来； + // 真的漏翻由 tools/i18n/check.mjs 报（它拿目录清单和实际文件逐节对） + const contents = listed.filter(c => { + const there = existsSync(resolve(ROOT, at(L, c.path))); + if (!there) SKIPPED.push(`${c.n} ${c.title}`); + return there; + }); + if (!contents.length) throw new Error(`${readmePath} 的目录里列了 ${listed.length} 节，正文一份都还没有`); + // 长文同理：没翻到的那几篇这次不进页面 + const docs = readLongform(readme, raw).filter(d => { + const there = existsSync(resolve(ROOT, at(L, d.path))); + if (!there) SKIPPED.push(d.title); + return there; + }); + + const blocks = { + noscriptList: noscriptList(contents, L, strings, raw), + docLinks: docLinks(docs, L, strings, raw), + aiLink: `${escText(strings['docs.aiLabel'])}${raw.listSep}${escText(strings['docs.ai'])}`, + jsonld: jsonLd(L, strings, contents.length, stats), + }; + + let html = template + .replace('{{htmlLang}}', L.htmlLang) + .replace(/\{\{canonical\}\}/g, L.canonical) + .replace(/\{\{ogLocale\}\}/g, L.ogLocale) + .replace('{{hreflang}}', hreflangLinks()) + .replace(/\{\{site\}\}/g, SITE) + .replace(/\{\{repo\}\}/g, REPO); + html = applyMarkup(html, strings, blocks); + + // 核实记录的链接：模板里只留一个 data-src 占位，地址指向仓库那一份中文的 + // （工作底稿没有译本），链接文字按语言给 + // 注意：data-src 里那个值本身已经是百分号编码过的（模板里照抄的 GitHub 链接）， + // 再编一次就变成 %25E6…，点开是 404 + const records = /[^<]*<\/a>/g; + if (!records.test(html)) throw new Error('模板里找不到页脚的核实记录链接（data-src）'); + html = html.replace(records, (_, p) => + `${escText(strings['docs.records'])}`); + + // 相对路径按这一份的位置改写：zh 的正文和广告都在上一层 + html = html + .replaceAll('href="README.md"', `href="${L.contentBase}README.md"`) + .replaceAll('href="book/"', `href="${L.contentBase}book/"`) + .replaceAll('src="ads/', `src="${L.assetBase}`) + .replace(``, ``); + + const config = { + code: L.code, dir: L.dir, contentBase: L.contentBase, assetBase: L.assetBase, + repoBlob: L.repoBlob, htmlLang: L.htmlLang, sortLocale: L.sortLocale, + locales: PUBLISHED.map(l => l.code), langNames: Object.fromEntries(PUBLISHED.map(l => [l.code, l.name])), + offline: false, + files: contents.map(c => c.path), // 这一份真正收了哪几节，页面照这个取 + docFiles: docs.map(d => d.path), + contents: raw.contents, longformLabel: raw.longformLabel, longformSee: raw.longformSee, + fields: raw.fields, dispute: raw.dispute, todo: raw.todo, + glossary: raw.glossary, glossarySplit: raw.glossarySplit, + xref: raw.xref, lens: raw.lens, costKeys: raw.costKeys, gainWords: raw.gainWords, + urlTrim: raw.urlTrim, srcOpen: raw.srcOpen, srcClose: raw.srcClose, srcSemi: raw.srcSemi, + strings, + }; + const script = ``; + const needle = ''; + if (!html.includes(needle)) throw new Error('模板里找不到 site-config 的占位脚本，页面取不到语言配置'); + html = html.replace(needle, script); + + // 页面里不该再有没填的 {{}}。字典那份不算：T('key', {n}) 的变量就是留给浏览器填的， + // 它整份在 site-config 里，剔掉再查，剩下的是真漏了（stats.json 里没这个数，或 T() 没传变量）。 + const shell = html.replace(script, ''); + const left = [...shell.matchAll(/\{\{(\w+)\}\}/g)].map(m => m[1]); + if (left.length) throw new Error(`生成的 ${L.dir}index.html 里还剩 {{${[...new Set(left)].join('}}、{{')} }}，要么 stats.json 里没这个数，要么 T() 忘了传变量`); + return html; +} + +function jsonLd(L, strings, sections, stats) { + const doc = { + '@context': 'https://schema.org', '@graph': [ + { '@type': 'WebSite', '@id': `${SITE}#website`, url: SITE + '/', name: strings['meta.siteName'], + description: strings['meta.description'], inLanguage: L.htmlLang, + potentialAction: { '@type': 'SearchAction', target: { '@type': 'EntryPoint', urlTemplate: `${L.canonical}?q={search_term_string}` }, 'query-input': 'required name=search_term_string' } }, + { '@type': 'Book', '@id': `${SITE}#book`, name: strings['meta.siteName'], url: SITE + '/', + inLanguage: L.htmlLang, bookFormat: 'https://schema.org/EBook', + numberOfPages: Number(stats.entries), license: 'https://creativecommons.org/licenses/by/4.0/', + abstract: strings['meta.description'], about: strings['meta.about'], isAccessibleForFree: true }, + ], + }; + return JSON.stringify(doc, null, 0).replace(/<\/script/gi, '<\\/script'); +} +// sitemap.xml 也由这里生成，跟着这次真正生成的页面走。原来是手写的：域名单独抄了 +// 一份、列的是 README.md（Pages 上那不是页面）、三份语言页一个都没有。 +// 不写 ：它得跟着正文改，每次重新生成都变，纯粹制造无意义的 diff； +// 写一个过期的日期比不写更糟，Google 的说明里 lastmod 本来就是可选的。 +function sitemapXml() { + const urls = [ + { loc: SITE_URL(), priority: '1.0' }, // 跳语言的小页 + ...PUBLISHED.map(l => ({ loc: l.canonical, priority: l.default ? '0.9' : '0.8' })), + ]; + return '\n' + + '\n' + + urls.map(u => ` \n ${escText(u.loc)}\n ${u.priority}\n `).join('\n') + + '\n\n'; +} + +// robots.txt 和 sitemap.xml 是一对：sitemap 指的那个地址得是真的。一起从 site.json 拼， +// 换域名就不会只改了一半。 +function robotsTxt() { + return `User-agent: * +Allow: / + +Sitemap: ${SITE_URL()}sitemap.xml +`; +} + +// site 和 repo 是全书唯一和「部署在哪个域名」有关的两处，都放在 tools/site/site.json。 +// 换域名只改那一处：locales.json 里每种语言的 canonical 和 repoBlob 由它们推出来， +// 对不上就在 checkUrls() 里报错，不会出现「Pages 发到 A 域名、页面自称是 B 域名」。 +// site 不带尾斜杠：模板里是 {{site}}/en/、{{site}}/og.png，带了会拼出双斜杠。 +const { site: SITE, repo: REPO } = JSON.parse(read('tools/site/site.json')); +const SITE_URL = () => SITE + '/'; + +// locales.json 的 canonical / repoBlob 是从 site.json 抄的。抄错了页面照样生成， +// 但链接会指向别处，而且没人会一眼看出来。这里逐个对一遍。 +function checkUrls() { + for (const l of LOCALES) { + const wantCanonical = SITE + '/' + l.dir; + const wantBlob = REPO + '/blob/main/' + (l.contentDir ? l.contentDir + '/' : ''); + if (l.canonical !== wantCanonical) + throw new Error(`locales.json 里 ${l.code} 的 canonical 是 ${l.canonical}，按 site.json 应该是 ${wantCanonical}`); + if (l.repoBlob !== wantBlob) + throw new Error(`locales.json 里 ${l.code} 的 repoBlob 是 ${l.repoBlob}，按 site.json 应该是 ${wantBlob}`); + } +} + +// ---------------------------------------------------------------- 写盘 +// SKIP：这一份语言的 README 和正文都还没翻，页面这次不生成。 +// SKIPPED：页面生成了，但目录里列的节或长文还没翻到，这次没进页面。 +// 两回事，分开报，免得把「还没翻」说成「翻坏了」。 +const SKIP = new Set(); +const SKIPPED = []; + +const stats = JSON.parse(read('tools/site/stats.json')); +const template = read('tools/site/page.template.html'); +checkUrls(); // 先对域名，对不上就别生成，免得发出一批指向别处的链接 +const targets = []; +for (const L of LOCALES){ + const text = renderPage(L, template, stats); + if (text !== null) targets.push({ path: `${L.dir}index.html`, text }); +} +targets.push({ path: 'index.html', text: chooserPage(SITE) }); +targets.push({ path: 'sitemap.xml', text: sitemapXml() }); +targets.push({ path: 'robots.txt', text: robotsTxt() }); + +if (CHECK) { + const stale = []; + for (const { path, text } of targets){ + const cur = existsSync(resolve(ROOT, path)) ? read(path) : null; + if (cur === text) continue; + stale.push(path); + if (cur === null) console.log(` ${path}：还没有`); + else console.log(` ${path}：和模板/字典对不上`); + } + if (!stale.length){ console.log('检索页、sitemap 和 robots.txt 都是最新的'); process.exit(0); } + console.log(`\n有 ${stale.length} 份生成物过时：${stale.join('、')}。本地跑 node tools/site/build.mjs 然后提交。`); + process.exit(1); +} + +for (const { path, text } of targets){ + mkdirSync(dirname(resolve(ROOT, path)), { recursive: true }); + writeFileSync(resolve(ROOT, path), text); + console.log(` ${path}：${(Buffer.byteLength(text) / 1024 | 0)} KB`); +} +const done = LOCALES.filter(l => !SKIP.has(l.code)).map(l => l.code); +console.log(`\n已生成检索页 ${done.length} 份（${done.join('、')}，默认 ${DEFAULT_LOCALE}）和 sitemap.xml、robots.txt`); +if (SKIP.size) console.log(`跳过 ${[...SKIP].join('、')}：这一份的 README 和正文还没翻完，翻完重跑本脚本就会带上。`); +if (SKIPPED.length){ + console.log(`另有 ${SKIPPED.length} 项在目录里列着、正文还没翻到，这次没进页面：${SKIPPED.slice(0, 6).join('、')}${SKIPPED.length > 6 ? ' …' : ''}`); +} \ No newline at end of file diff --git a/tools/site/extract.mjs b/tools/site/extract.mjs new file mode 100644 index 00000000..53439b2a --- /dev/null +++ b/tools/site/extract.mjs @@ -0,0 +1,110 @@ +// 从模板里把中文文案抠出来，写进 tools/site/locales/zh.json。 +// +// node tools/site/extract.mjs # 补齐 zh.json 里模板里有的那些键 +// node tools/site/extract.mjs --check # 只比对 +// +// 为什么有这一个：中文是原文、是模板里的默认值，zh.json 是从它抄下来的，抄错一处 +// 就会悄悄把界面文案改掉。改中文文案改模板，跑一遍这个，zh.json 跟上。 +// JS 里 T('key') 用到、模板里没有对应元素的键（比如「整节，共 3 条」）这个脚本补不出来， +// 第一次跑会列出来，手写进 zh.json，之后脚本只更新它能抠到的那部分，不会动。 +import { readFileSync, writeFileSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const CHECK = process.argv.includes('--check'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); + +const template = read('tools/site/page.template.html'); +const zhPath = 'tools/site/locales/zh.json'; +const zh = JSON.parse(read(zhPath)); + +const TOKENS = /<[^>]*>|[^<]+/g; +const toks = template.match(TOKENS) ?? []; +const found = new Map(); +const attr = (tag, name) => { + const m = new RegExp(`\\s${name}="([^"]*)"`).exec(tag); + return m ? m[1] : null; +}; + +// data-i18n-block 的内容由 build.mjs 生成，模板里那几行只是给人看的，不算文案 +const isBlock = tag => /\bdata-i18n-block=/.test(tag); + +for (let i = 0; i < toks.length; i++){ + const tok = toks[i]; + if (!tok.startsWith('<') || isBlock(tok)) continue; + const m = /^<([a-z0-9]+)\b([^>]*)>/i.exec(tok); + if (!m) continue; + const [, tag, attrs] = m; + const key = /\bdata-i18n="([\w.]+)"/.exec(attrs); + if (key){ + // 内文可能有 这类行内标签，整段当文案；HTML 实体原样留着 + let depth = 0, inner = []; + for (let j = i + 1; j < toks.length; j++){ + const t = toks[j]; + if (t.startsWith('$`, 'i').test(t)){ if (depth === 0) break; depth--; } + else if (new RegExp(`^ [k, zh.strings[k]])); +const text = JSON.stringify({ ...zh, strings: ordered }, null, 2) + '\n'; + +if (CHECK){ + if (read(zhPath) === text){ console.log('zh.json 检查通过'); process.exit(0); } + console.log('zh.json 和模板里的中文对不上。本地跑 node tools/site/extract.mjs 然后提交。'); + process.exit(1); +} + +writeFileSync(resolve(ROOT, zhPath), text); +console.log(`已更新 ${zhPath}：${found.size} 个键从模板取到中文，共 ${Object.keys(ordered).length} 个键`); +if (missing.length) console.log(`\n还差 ${missing.length} 个键，模板里抠不到（JS 里拼出来的或只在字典里），手写进 zh.json：\n ${missing.join('\n ')}`); \ No newline at end of file diff --git a/tools/site/locales.json b/tools/site/locales.json new file mode 100644 index 00000000..600a06e6 --- /dev/null +++ b/tools/site/locales.json @@ -0,0 +1,49 @@ +[ + { + "code": "en", + "default": true, + "dir": "en/", + "contentBase": "./", + "assetBase": "../ads/", + "repoBlob": "https://github.com/streamentry/HowToLiveBetter/blob/main/en/", + "htmlLang": "en", + "ogLocale": "en_US", + "sortLocale": "en", + "name": "English", + "accept": "en", + "canonical": "https://streamentry.github.io/HowToLiveBetter/en/", + "note": "Default edition", + "contentDir": "en" + }, + { + "code": "vi", + "dir": "vi/", + "contentBase": "./", + "assetBase": "../ads/", + "repoBlob": "https://github.com/streamentry/HowToLiveBetter/blob/main/vi/", + "htmlLang": "vi", + "ogLocale": "vi_VN", + "sortLocale": "vi", + "name": "Tiếng Việt", + "accept": "vi", + "canonical": "https://streamentry.github.io/HowToLiveBetter/vi/", + "note": "Bản dịch tiếng Việt", + "contentDir": "vi" + }, + { + "code": "zh", + "dir": "zh/", + "contentBase": "../", + "assetBase": "../ads/", + "repoBlob": "https://github.com/streamentry/HowToLiveBetter/blob/main/", + "htmlLang": "zh-CN", + "ogLocale": "zh_CN", + "sortLocale": "zh", + "name": "中文", + "accept": "zh", + "canonical": "https://streamentry.github.io/HowToLiveBetter/zh/", + "note": "原文，译本以它为准", + "source": true, + "contentDir": "" + } +] diff --git a/tools/site/locales.mjs b/tools/site/locales.mjs new file mode 100644 index 00000000..aeeafd2d --- /dev/null +++ b/tools/site/locales.mjs @@ -0,0 +1,33 @@ +// 语言配置。tools/site/build.mjs 生成页面，tools/i18n/check.mjs 校译文，都从这里读同一份。 +import { readFileSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; + +const HERE = dirname(fileURLToPath(import.meta.url)); +const ROOT = resolve(HERE, '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); + +export const LOCALES = JSON.parse(read('tools/site/locales.json')); +export const DEFAULT_LOCALE = LOCALES.find(l => l.default)?.code ?? LOCALES[0].code; +export const SOURCE = LOCALES.find(l => l.source) ?? LOCALES.find(l => l.code === 'zh'); + +export function locale(code) { + const l = LOCALES.find(x => x.code === code); + if (!l) throw new Error(`没有语言 ${code}，tools/site/locales.json 里只有 ${LOCALES.map(x => x.code).join('、')}`); + return l; +} +export const LOCALE = locale; + +// 字典：界面文案 + 认正文用的标记 +export function strings(code) { + const raw = JSON.parse(read(`tools/site/locales/${code}.json`)); + if (!raw.strings) throw new Error(`locales/${code}.json 里没有 strings`); + return raw; +} + +/** 语言目录，末尾带斜杠：en/、vi/、zh/ */ +export const dir = code => locale(code).dir; +/** 正文所在的仓库根相对目录，末尾无斜杠：zh 是 ''（就在根上），en 是 'en' */ +export const contentDir = code => locale(code).contentDir; +/** 浏览器那边的相对根：zh 是 ../，en/vi 是 ./。给页面里的 fetch 和相对链接用 */ +export const base = code => locale(code).contentBase; \ No newline at end of file diff --git a/tools/site/locales/en.json b/tools/site/locales/en.json new file mode 100644 index 00000000..1ec94a3f --- /dev/null +++ b/tools/site/locales/en.json @@ -0,0 +1,195 @@ +{ + "contents": "Contents", + "contentsQuestion": "What this book answers", + "text": "The text", + "glossary": "Reading the numbers", + "longformLabel": "Long-form", + "longformSee": "Long-form essay:", + "listSep": " — ", + "back": "Back to the contents", + "fields": { + "cost": "Cost:", + "human": "In plain language:", + "gain": "Benefit:", + "grade": "Evidence grade:", + "src": "Sources:", + "note": "Notes:" + }, + "dispute": "Disputed", + "todo": [ + "to be verified", + "TODO" + ], + "glossarySep": "、", + "glossarySplit": "^[A-Za-z0-9%. ]+$", + "xref": { + "listSep": "\\s*(?:,|and|to|–|-)\\s*", + "rangeRe": "^\\s*(\\d+)\\s*(?:to|[-–])\\s*(\\d+)\\s*$", + "pattern": "items?\\s+(?\\d+(?:\\s*(?:,|and|to|[-–])\\s*\\d+)*)\\s+in\\s+section\\s+(?\\d+)|items?\\s+(?\\d+(?:\\s*(?:,|and|to|[-–])\\s*\\d+)*)\\s+(?:in|of)\\s+this\\s+section|\\bitems?\\s+(?\\d+(?:\\s*(?:,|and|to|[-–])\\s*\\d+)*)|\\bsections?\\s+(?\\d+(?:\\s*(?:,|and)\\s*\\d+)*)" + }, + "lens": { + "死亡率": "lens.life", + "金钱": "lens.money", + "时间": "lens.time", + "自由": "lens.freedom" + }, + "costKeys": { + "money": { + "0": "0", + "少": "few", + "多": "much" + }, + "time": { + "少": "few", + "中": "mid", + "多": "much" + }, + "will": { + "否": "no", + "些": "some", + "是": "lots" + } + }, + "gainWords": { + "大": "large", + "中": "medium", + "小": "small" + }, + "urlTrim": "[).,;:。，；：、]", + "srcOpen": "\"'([{（「『【《", + "srcClose": "\"')]}）」』】》", + "srcSemi": ";", + "strings": { + "ad.label": "Ad", + "ad.tip": "Useful? Buy the author a coffee", + "badge.dispute": "Disputed", + "badge.grade": "Grade {{g}}", + "badge.ratio": "Value {{v}}", + "badge.ratioTitle": "The author's judgement, combined from the size of the benefit ({{level}}) and the three costs. Comparable only within the same measure ({{lens}}), and unrelated to the evidence grade.", + "badge.todo": "Has items to verify", + "card.link": "Link to this item", + "cost.money.0": "No money", + "cost.money.few": "A little money", + "cost.money.much": "A fair amount of money", + "cost.time.few": "In passing", + "cost.time.mid": "A few hours", + "cost.time.much": "Every day", + "cost.will.lots": "A lot of willpower", + "cost.will.no": "No willpower", + "cost.will.some": "Some willpower", + "docs.ai": "Let Claude Code or Codex answer your questions from this book (how to install the skill)", + "docs.aiLabel": "AI assistant", + "docs.label": "Long-form", + "docs.records": "Source verification records", + "empty.msg": "No items match. Drop a filter, or try a shorter keyword.", + "err.embedMiss": "Missing from the offline copy: ", + "err.fail": "Could not read the text ({{msg}}). Check that README.md and the book/ directory sit next to index.html.", + "err.file": "This page has to be opened over http to read the text. Run python -m http.server in the repository directory and open http://localhost:8000/, or just use the GitHub Pages address.", + "err.net": "Loading failed", + "field.cost": "Cost", + "field.gain": "Benefit", + "field.note": "Notes", + "field.src": "Sources", + "filter.dispute": "Only disputed items", + "filter.grade": "Evidence grade", + "filter.gradeNote": "Meta-analysis/RCT · some research · consensus", + "filter.lens": "What it buys back", + "filter.lensNote": "Measures are not compared across", + "filter.money": "Money spent", + "filter.ratio": "Value for money", + "filter.ratioNote": "The author's judgement, comparable within a measure", + "filter.reset": "Clear filters", + "filter.sec": "Sections", + "filter.time": "Time spent", + "filter.todo": "Only items to verify", + "filter.will": "Willpower needed", + "foot.a": "Which measure a number belongs to is stated inside each item: sections 1 and 2 use all-cause or cause-specific mortality, sections 3 and 4 energy and time, section 5 money, section 7 money and benefits, sections 8 and 9 money and personal freedom. They are not converted into each other. The verification trail is in", + "foot.b": ". The text is published under", + "foot.c": ". Credit the author and link the original if you reprint or adapt it.", + "gain.large": "large", + "gain.medium": "medium", + "gain.small": "small", + "gloss.note": "Underlined terms in the text explain themselves when you click one or hover over it. The full table is below.", + "gloss.summary": "Abbreviations and terms explained", + "grade.a": "Grade A", + "grade.b": "Grade B", + "grade.c": "Grade C", + "head.p1": "Every item answers two questions: what it costs, and what it buys back.", + "head.p2": "Who benefits is graded too: you first, then your spouse and immediate family, then friends and colleagues, strangers last. Last does not mean zero — only that the return is unlikely, and that you have to weigh the risks along with it.", + "head.p3": "You do not have to do all of it. This is a shortlist ordered by value for money, not a task list. Taking one or two entries with you counts; come back for the rest when you need them. For the cheapest ones, set \"Money spent\" to \"No money\" and \"Willpower needed\" to \"No willpower\" on the left. The author has not done most of them either. Dotted \"item X\" references in the text open in place, so you can see what that item says without hunting for it.", + "hint.p1": "One section at a time; costs and grades can take several values. Several values inside one group are \"or\"; different groups are \"and\".", + "hint.p2": "Items are ordered from best to worst value within each section. Searching does not change the order.", + "hint.p3a": "The text and the tags come from the", + "hint.p3b": "files in the repository's book/, so editing the text edits this page.", + "lang.label": "Language", + "lens.freedom": "Personal freedom", + "lens.life": "Lifespan", + "lens.money": "Money", + "lens.time": "Time and energy", + "loading": "Reading the text …", + "loading.progress": "Reading the text {{done}}/{{total}}", + "meta.about": [ + "Health", + "Longevity", + "Personal finance", + "First aid", + "Everyday law", + "Consumer protection" + ], + "meta.author": "A High-Return Life Guide", + "meta.description": "A life guide ordered by value for money: {{entries}} pieces of advice on longevity and disease prevention, accident first aid, saving money, scams and legal red lines, income support, business risk, dating and marriage, going abroad, and skills. Each one states its cost, its benefit, the grade of evidence, and the original sources, and you can filter by the three costs of money, time, and willpower.", + "meta.keywords": "longevity, health, evidence-based, all-cause mortality, saving money, personal finance, scams, first aid, everyday law, unemployment benefits, labour arbitration, business risk, value for money", + "meta.siteName": "A High-Return Life Guide", + "meta.title": "A High-Return Life Guide · Get back lifespan, money and freedom for the least money, time and energy", + "modal.close": "Close", + "modal.openOnGithub": "Open on GitHub", + "money.0": "No money", + "money.few": "A little", + "money.much": "A fair amount", + "nav.github": "View the source repository on GitHub", + "nav.githubAria": "GitHub repository", + "nav.menu": "Open filters", + "nav.readme": "View README.md", + "nav.theme": "Switch dark mode", + "noscript.p1": "Get back the most lifespan, money and personal freedom for the least money, time, energy and willpower. {{sections}} sections, {{entries}} items, each stating its cost, benefit, evidence grade and original sources. Benefits are graded by who receives them: you, your spouse and immediate family, friends and colleagues, strangers — the lower the grade, the smaller the expected return and the more the risks have to be spelled out.", + "noscript.p2a": "The filters on this page need JavaScript. You can read the text directly: the table of contents in", + "noscript.p2b": ", and the text itself in", + "noscript.p2c": ".", + "perf.buildDom": "Build DOM", + "perf.cards": "cards", + "perf.collapse": "Collapse", + "perf.expand": "Expand", + "perf.files": "files", + "perf.hits": "{{n}} matched, {{m}} re-rendered", + "perf.longFrame": "Longest frame", + "perf.longTask": "Long task", + "perf.mark": "{{dom}} DOM nodes, {{terms}} terms", + "perf.parse": "Parse text", + "perf.relayout": "relayout", + "perf.renderMore": "Render on scroll", + "ratio.high": "High", + "ratio.mid": "Moderate", + "ratio.top": "Exceptional", + "search.ph": "Search items", + "sec.all": "All sections", + "sec.fold": "Expand or collapse this section's contents", + "sec.none": "No items match the current filters", + "src.count": " · {{n}} sources", + "stat.of": "shown, out of", + "stat.show": "Showing", + "stat.total": "in total", + "time.few": "In passing", + "time.mid": "A few hours", + "time.much": "Every day", + "tip.aria": "WeChat reward code", + "tip.p1": "Scan it with WeChat if you like. It changes nothing here either way.", + "tip.p2": "On a phone, take a screenshot first, then pick the image from your album in WeChat.", + "tip.qrAlt": "WeChat reward code", + "will.lots": "A lot", + "will.no": "None", + "will.some": "A little", + "xref.go": "Go there", + "xref.item": "Section {{sec}}, item {{n}}", + "xref.wholeSec": "The whole section, {{n}} items" + } +} diff --git a/tools/site/locales/vi.json b/tools/site/locales/vi.json new file mode 100644 index 00000000..198e00e9 --- /dev/null +++ b/tools/site/locales/vi.json @@ -0,0 +1,195 @@ +{ + "contents": "Mục lục", + "contentsQuestion": "Những gì cuốn sách này trả lời", + "text": "Nội dung", + "glossary": "Đọc hiểu con số", + "longformLabel": "Bài dài", + "longformSee": "Bài dài:", + "listSep": " — ", + "back": "Về mục lục", + "fields": { + "cost": "Chi phí:", + "human": "Nói thẳng:", + "gain": "Lợi ích:", + "grade": "Mức bằng chứng:", + "src": "Nguồn:", + "note": "Ghi chú:" + }, + "dispute": "Tranh cãi", + "todo": [ + "cần đối chiếu", + "TODO" + ], + "glossarySep": "、", + "glossarySplit": "^[A-Za-z0-9%. ]+$", + "xref": { + "listSep": "\\s*(?:,| và |, và |hoặc|[-–])\\s*", + "rangeRe": "^\\s*(\\d+)\\s*(?:đến|[-–])\\s*(\\d+)\\s*$", + "pattern": "mục\\s+(?\\d+(?:\\s*(?:,|và|hoặc|đến|[-–])\\s*\\d+)*)\\s+ở\\s+phần\\s+(?\\d+)|phần\\s+(?\\d+(?:\\s*(?:,|và)\\s*\\d+)*)|mục\\s+(?\\d+(?:\\s*(?:,|và|hoặc|đến|[-–])\\s*\\d+)*)\\s+của\\s+phần\\s+này|\\bmục\\s+(?\\d+(?:\\s*(?:,|và|hoặc|đến|[-–])\\s*\\d+)*)" + }, + "lens": { + "死亡率": "lens.life", + "金钱": "lens.money", + "时间": "lens.time", + "自由": "lens.freedom" + }, + "costKeys": { + "money": { + "0": "0", + "少": "few", + "多": "much" + }, + "time": { + "少": "few", + "中": "mid", + "多": "much" + }, + "will": { + "否": "no", + "些": "some", + "是": "lots" + } + }, + "gainWords": { + "大": "large", + "中": "medium", + "小": "small" + }, + "urlTrim": "[).,;:。，；：、]", + "srcOpen": "\"'([{（「『【《", + "srcClose": "\"')]}）」』】》", + "srcSemi": ";", + "strings": { + "ad.label": "Quảng cáo", + "ad.tip": "Thấy hữu ích? Mời tác giả một ly cà phê", + "badge.dispute": "Tranh cãi", + "badge.grade": "Mức {{g}}", + "badge.ratio": "Hiệu quả {{v}}", + "badge.ratioTitle": "Đánh giá của tác giả, ghép từ cỡ lợi ích ({{level}}) và ba khoản chi phí. Chỉ so được trong cùng một thước đo ({{lens}}), không liên quan đến mức bằng chứng.", + "badge.todo": "Có chỗ cần đối chiếu", + "card.link": "Liên kết tới mục này", + "cost.money.0": "Không tốn tiền", + "cost.money.few": "Tốn ít tiền", + "cost.money.much": "Tốn khá nhiều tiền", + "cost.time.few": "Tiện tay", + "cost.time.mid": "Mất vài tiếng", + "cost.time.much": "Mỗi ngày", + "cost.will.lots": "Cần nhiều ý chí", + "cost.will.no": "Không cần ý chí", + "cost.will.some": "Cần một chút ý chí", + "docs.ai": "Để Claude Code hoặc Codex trả lời câu hỏi của bạn theo cuốn sách này (cách cài skill)", + "docs.aiLabel": "Trợ lý AI", + "docs.label": "Bài dài", + "docs.records": "Hồ sơ đối chiếu nguồn", + "empty.msg": "Không có mục nào khớp. Bỏ bớt một điều kiện lọc, hoặc dùng từ khoá ngắn hơn.", + "err.embedMiss": "Bản sao ngoại tuyến thiếu ", + "err.fail": "Không đọc được nội dung ({{msg}}). Kiểm tra xem README.md và thư mục book/ có nằm cạnh index.html không.", + "err.file": "Trang này phải mở qua http mới đọc được nội dung. Chạy python -m http.server trong thư mục repo rồi vào http://localhost:8000/, hoặc dùng địa chỉ GitHub Pages.", + "err.net": "Tải thất bại", + "field.cost": "Chi phí", + "field.gain": "Lợi ích", + "field.note": "Ghi chú", + "field.src": "Nguồn", + "filter.dispute": "Chỉ xem mục có tranh cãi", + "filter.grade": "Mức bằng chứng", + "filter.gradeNote": "Tổng hợp/RCT · có nghiên cứu · đồng thuận", + "filter.lens": "Đổi lại được gì", + "filter.lensNote": "Thước đo, không so chéo giữa các thước đo", + "filter.money": "Tiền bỏ ra", + "filter.ratio": "Hiệu quả chi phí", + "filter.ratioNote": "Đánh giá của tác giả, so được trong cùng thước đo", + "filter.reset": "Xoá bộ lọc", + "filter.sec": "Mục", + "filter.time": "Thời gian bỏ ra", + "filter.todo": "Chỉ xem mục cần đối chiếu", + "filter.will": "Ý chí phải có", + "foot.a": "Mỗi con số thuộc thước đo nào thì ghi ngay trong mục đó: mục 1 và 2 là tử vong toàn nguyên hoặc theo nguyên nhân, mục 3 và 4 là sức lực và thời gian, mục 5 là tiền, mục 7 là tiền và trợ cấp, mục 8 và 9 là tiền và tự do cá nhân. Chúng không quy đổi qua lại. Quá trình đối chiếu nguồn nằm ở", + "foot.b": ". Nội dung phát hành theo", + "foot.c": ". Nếu in lại hoặc sửa đổi, xin ghi nguồn và kèm liên kết bản gốc.", + "gain.large": "lớn", + "gain.medium": "vừa", + "gain.small": "nhỏ", + "gloss.note": "Từ có gạch chân trong nội dung, bấm vào hoặc đưa chuột vào là hiện nghĩa. Toàn bộ bảng ở dưới.", + "gloss.summary": "Chữ viết tắt và thuật ngữ", + "grade.a": "Mức A", + "grade.b": "Mức B", + "grade.c": "Mức C", + "head.p1": "Mỗi mục trả lời hai câu hỏi: bỏ ra cái gì, đổi lại được cái gì.", + "head.p2": "Ai là người hưởng cũng chia thang: bạn cao nhất, kế đến vợ chồng và người thân trực hệ, rồi tới bạn bè và đồng nghiệp, người xa lạ thấp nhất. Thấp nhất không phải là bằng không, chỉ là kỳ vọng hoàn lại nhỏ, và phải xem cả rủi ro cùng lúc.", + "head.p3": "Không phải làm hết: đây là danh sách dự phòng đã xếp theo hiệu quả chi phí, không phải danh sách việc phải làm. Lấy một hai mục là được, phần còn lại quay lại tra khi cần. Muốn lấy mục rẻ nhất thì bên trái chọn \"Tiền bỏ ra\" là \"Không tốn tiền\", \"Ý chí phải có\" là \"Không cần ý chí\". Tác giả cũng chưa làm phần lớn. Mọi \"mục X\" có gạch chấm trong nội dung đều bấm được ngay tại chỗ, không phải tự lật tìm.", + "hint.p1": "Mỗi lần chỉ xem một mục, chi phí và mức thì chọn nhiều được. Nhiều lựa chọn trong cùng nhóm là \"hoặc\", khác nhóm là \"và\".", + "hint.p2": "Trong mỗi mục, các mục con xếp từ hiệu quả cao xuống thấp, tìm kiếm không đổi thứ tự.", + "hint.p3a": "Nội dung và nhãn lấy từ", + "hint.p3b": "tệp trong thư mục book/ của kho, sửa nội dung là trang này đổi theo.", + "lang.label": "Ngôn ngữ", + "lens.freedom": "Tự do cá nhân", + "lens.life": "Tuổi thọ", + "lens.money": "Tiền", + "lens.time": "Thời gian và sức lực", + "loading": "Đang đọc nội dung …", + "loading.progress": "Đang đọc nội dung {{done}}/{{total}}", + "meta.about": [ + "Sức khoẻ", + "Thọ thọ", + "Tài chính cá nhân", + "Cấp cứu", + "Pháp luật thường ngày", + "Bảo vệ người tiêu dùng" + ], + "meta.author": "Cẩm Nang Sống Hiệu Quả Cao", + "meta.description": "Cẩm nang đời sống xếp theo hiệu quả chi phí: {{entries}} lời khuyên về sống lâu và phòng bệnh, cấp cứu tai nạn, tiết kiệm tiền, lừa đảo và vạch đỏ pháp lý, trợ cấp thất nghiệp, rủi ro kinh doanh, yêu đương và kết hôn, đi nước ngoài, và kỹ năng. Mỗi mục ghi rõ chi phí, lợi ích, mức bằng chứng và nguồn gốc, và có thể lọc theo ba khoản chi phí là tiền, thời gian, ý chí.", + "meta.keywords": "sống lâu, sức khoẻ, dựa trên bằng chứng, tử vong toàn nguyên, tiết kiệm tiền, tài chính cá nhân, lừa đảo, cấp cứu, pháp luật thường ngày, trợ cấp thất nghiệp, hoà giải lao động, rủi ro kinh doanh, hiệu quả chi phí", + "meta.siteName": "Cẩm Nang Sống Hiệu Quả Cao", + "meta.title": "Cẩm Nang Sống Hiệu Quả Cao · Đổi lấy tuổi thọ, tiền và tự do bằng ít tiền, thời gian và sức lực nhất", + "modal.close": "Đóng", + "modal.openOnGithub": "Mở trên GitHub", + "money.0": "Không tốn tiền", + "money.few": "Ít", + "money.much": "Khá nhiều", + "nav.github": "Xem kho mã nguồn trên GitHub", + "nav.githubAria": "Kho GitHub", + "nav.menu": "Mở bộ lọc", + "nav.readme": "Xem README.md", + "nav.theme": "Đổi chế độ tối", + "noscript.p1": "Đổi lấy nhiều tuổi thọ, tiền và tự do cá nhân bằng ít tiền, thời gian, sức lực và ý chí nhất. {{sections}} mục, {{entries}} mục con, mục nào cũng ghi rõ chi phí, lợi ích, mức bằng chứng và nguồn gốc. Lợi ích chia theo người hưởng: bạn, vợ chồng và người thân trực hệ, bạn bè và đồng nghiệp, người xa lạ — thang càng thấp, kỳ vọng hoàn lại càng nhỏ và rủi ro càng phải viết rõ.", + "noscript.p2a": "Bộ lọc trên trang này cần JavaScript. Bạn vẫn đọc trực tiếp được: mục lục trong", + "noscript.p2b": ", phần nội dung nằm ở", + "noscript.p2c": ".", + "perf.buildDom": "Dựng DOM", + "perf.cards": "thẻ", + "perf.collapse": "Thu gọn", + "perf.expand": "Mở ra", + "perf.files": "tệp", + "perf.hits": "khớp {{n}}, dựng lại {{m}}", + "perf.longFrame": "Khung dài nhất", + "perf.longTask": "Tác vụ dài", + "perf.mark": "{{dom}} nút DOM, {{terms}} thuật ngữ", + "perf.parse": "Phân tích nội dung", + "perf.relayout": "dàn lại", + "perf.renderMore": "Dựng khi cuộn", + "ratio.high": "Cao", + "ratio.mid": "Trung bình", + "ratio.top": "Rất cao", + "search.ph": "Tìm mục", + "sec.all": "Tất cả các mục", + "sec.fold": "Mở hoặc thu gọn mục lục của mục này", + "sec.none": "Không có mục nào khớp bộ lọc hiện tại", + "src.count": " · {{n}} nguồn", + "stat.of": "đang hiện, trong tổng số", + "stat.show": "Đang hiện", + "stat.total": "mục", + "time.few": "Tiện tay", + "time.mid": "Mất vài tiếng", + "time.much": "Mỗi ngày", + "tip.aria": "Mã thưởng WeChat", + "tip.p1": "Quét bằng WeChat cũng được, không cho cũng không ảnh hưởng gì ở đây.", + "tip.p2": "Xem trên điện thoại thì chụp màn hình trước, rồi mở WeChat chọn ảnh từ thư viện.", + "tip.qrAlt": "Mã thưởng WeChat", + "will.lots": "Nhiều", + "will.no": "Không", + "will.some": "Một chút", + "xref.go": "Tới đó", + "xref.item": "Phần {{sec}}, mục {{n}}", + "xref.wholeSec": "Cả phần, {{n}} mục" + } +} diff --git a/tools/site/locales/zh.json b/tools/site/locales/zh.json new file mode 100644 index 00000000..3f57dd47 --- /dev/null +++ b/tools/site/locales/zh.json @@ -0,0 +1,195 @@ +{ + "contents": "目录", + "contentsQuestion": "这本书想回答的问题", + "text": "正文", + "glossary": "读懂数字", + "longformLabel": "长文", + "longformSee": "长文见", + "listSep": "：", + "back": "回总目录", + "fields": { + "cost": "成本：", + "human": "说人话：", + "gain": "收益：", + "grade": "证据等级：", + "src": "来源：", + "note": "备注：" + }, + "dispute": "争议", + "todo": [ + "待核实", + "TODO" + ], + "glossarySep": "、", + "glossarySplit": "^[A-Za-z0-9%. ]+$", + "xref": { + "listSep": "、", + "rangeRe": "^\\s*(\\d+)\\s*到\\s*(\\d+)\\s*$", + "pattern": "第\\s*(?\\d+)\\s*节第\\s*(?[\\d、]+(?:\\s*到\\s*[\\d、]+)?)\\s*条|本节第\\s*(?[\\d、]+(?:\\s*到\\s*[\\d、]+)?)\\s*条|第\\s*(?[\\d、]+(?:\\s*到\\s*[\\d、]+)?)\\s*条|第\\s*(?[\\d、]+)\\s*节" + }, + "lens": { + "死亡率": "lens.life", + "金钱": "lens.money", + "时间": "lens.time", + "自由": "lens.freedom" + }, + "costKeys": { + "money": { + "0": "0", + "少": "few", + "多": "much" + }, + "time": { + "少": "few", + "中": "mid", + "多": "much" + }, + "will": { + "否": "no", + "些": "some", + "是": "lots" + } + }, + "gainWords": { + "大": "large", + "中": "medium", + "小": "small" + }, + "urlTrim": "[）。,，、；;：:]", + "srcOpen": "「『（", + "srcClose": "」』）", + "srcSemi": "；", + "strings": { + "ad.label": "广告", + "ad.tip": "觉得有用？请作者喝杯咖啡", + "badge.dispute": "争议", + "badge.grade": "证据 {{g}} 级", + "badge.ratio": "性价比 {{v}}", + "badge.ratioTitle": "作者判断：由收益量级（{{level}}）和三项成本合成，只在同一口径（{{lens}}）内可比，与证据等级无关", + "badge.todo": "含待核实", + "card.link": "本条链接", + "cost.money.0": "不花钱", + "cost.money.few": "花少量钱", + "cost.money.much": "花不少钱", + "cost.time.few": "顺手", + "cost.time.mid": "花几小时", + "cost.time.much": "每天占时间", + "cost.will.lots": "要很多毅力", + "cost.will.no": "不用毅力", + "cost.will.some": "要一点毅力", + "docs.ai": "让 Claude Code 或 Codex 照这本书回答你的问题（skill 装法）", + "docs.aiLabel": "AI 助手", + "docs.label": "长文", + "docs.records": "来源核实记录", + "empty.msg": "没有匹配的条目。去掉一个筛选条件，或换个更短的关键词。", + "err.embedMiss": "离线副本里缺 ", + "err.fail": "读取正文失败（{{msg}}）。确认 README.md 和 book/ 目录都在 index.html 旁边。", + "err.file": "这个页面要通过 http 打开才能读取正文。在仓库目录运行 python -m http.server，然后访问 http://localhost:8000/；或直接用 GitHub Pages 地址。", + "err.net": "加载失败", + "field.cost": "成本", + "field.gain": "收益", + "field.note": "备注", + "field.src": "来源", + "filter.dispute": "只看标了争议的", + "filter.grade": "证据等级", + "filter.gradeNote": "荟萃/RCT · 有研究 · 共识", + "filter.lens": "换回什么", + "filter.lensNote": "口径，不跨口径比较", + "filter.money": "花钱", + "filter.ratio": "性价比", + "filter.ratioNote": "作者判断，同口径内可比", + "filter.reset": "清空筛选", + "filter.sec": "章节", + "filter.time": "花时间", + "filter.todo": "只看有待核实的", + "filter.will": "要毅力", + "foot.a": "数字口径以条目内标注为准：第 1、2 节是 ACM 或特定死因，第 3、4 节是精力/时间，第 5 节是金钱，第 7 节是金钱与保障，第 8、9 节是金钱与人身自由，互不换算。来源核实过程见", + "foot.b": "。正文按", + "foot.c": "发布，转载改编请署名并附原文链接。", + "gain.large": "大", + "gain.medium": "中", + "gain.small": "小", + "gloss.note": "正文里带虚线的词，点一下或把鼠标放上去就有解释。全表如下。", + "gloss.summary": "看不懂的缩写和名词", + "grade.a": "A 级", + "grade.b": "B 级", + "grade.c": "C 级", + "head.p1": "每一条都回答两个问题：花掉什么，换回什么。", + "head.p2": "换回的算在谁头上也分档：你自己最高，其次配偶和直系亲属，再次朋友同事，陌生人最低——最低不等于零，只是回报的指望小、要连风险一起看。", + "head.p3": "不用全做：这是一份按性价比排好的备选单，不是任务清单。挑走一两条就算数，剩下的需要时再回来查——想挑省力的，左边把「花钱」选「不花钱」、「要毅力」选「不用」，剩下的就是。作者自己也没做到其中大部分。正文里带虚线的「第 X 条」可以点开，就地看那条写了什么，不用自己翻过去。", + "hint.p1": "章节一次只看一个，成本和等级可多选。同一组多选是「或」，不同组之间是「且」。", + "hint.p2": "每节内按性价比从高到低排，检索不改变顺序。", + "hint.p3a": "正文与标签来自仓库", + "hint.p3b": "下的 {{sections}} 个文件，改正文即改这里。", + "lang.label": "语言", + "lens.freedom": "人身自由", + "lens.life": "寿命", + "lens.money": "钱", + "lens.time": "时间精力", + "loading": "正在读取正文 …", + "loading.progress": "正在读取正文 {{done}}/{{total}}", + "meta.about": [ + "健康", + "长寿", + "个人理财", + "急救", + "法律常识", + "消费者保护" + ], + "meta.author": "高性价比人生指南", + "meta.description": "按性价比排序的人生指南：{{entries}} 条建议，覆盖长寿防病、意外急救、省钱理财、防骗与法律红线、失业兜底、创业风险、恋爱婚育、出国与技能。每条写明成本、收益、证据等级和原始出处，可按钱、时间、毅力三个成本维度检索。", + "meta.keywords": "长寿,健康,循证,总死亡率,省钱,理财,防骗,急救,法律常识,失业救助,劳动仲裁,创业风险,性价比", + "meta.siteName": "高性价比人生指南", + "meta.title": "高性价比人生指南 · 用最少的钱、时间和精力换回寿命、金钱和人身自由", + "modal.close": "关闭", + "modal.openOnGithub": "在 GitHub 打开", + "money.0": "不花钱", + "money.few": "少", + "money.much": "多", + "nav.github": "在 GitHub 上查看源仓库", + "nav.githubAria": "GitHub 仓库", + "nav.menu": "打开筛选", + "nav.readme": "查看 README.md", + "nav.theme": "切换深色模式", + "noscript.p1": "用最少的钱、时间、精力和毅力，换回最多的寿命、金钱、时间和人身自由。全书 {{sections}} 节 {{entries}} 条，每条写明成本、收益、证据等级和原始出处。收益按受益人分四档算：你自己、配偶与直系亲属、朋友同事、陌生人，档次越低回报期望越小、风险越要写清。", + "noscript.p2a": "本页的筛选功能需要 JavaScript。正文可直接阅读：", + "noscript.p2b": "里的目录，正文在", + "noscript.p2c": "。", + "perf.buildDom": "建 DOM", + "perf.cards": "张", + "perf.collapse": "收起", + "perf.expand": "展开", + "perf.files": "个文件", + "perf.hits": "命中 {{n}} 条，重渲染 {{m}} 张", + "perf.longFrame": "最长帧", + "perf.longTask": "长任务", + "perf.mark": "DOM 节点 {{dom}}，术语 {{terms}}", + "perf.parse": "解析正文", + "perf.relayout": "重排", + "perf.renderMore": "滚动补渲染", + "ratio.high": "高", + "ratio.mid": "一般", + "ratio.top": "极高", + "search.ph": "搜索条目", + "sec.all": "全部章节", + "sec.fold": "展开或收起本节目录", + "sec.none": "没有符合当前筛选的条目", + "src.count": " · {{n}} 条", + "stat.of": "条，共", + "stat.show": "当前显示", + "stat.total": "条", + "time.few": "顺手", + "time.mid": "几小时", + "time.much": "每天占用", + "tip.aria": "微信赞赏码", + "tip.p1": "用微信扫一扫，给不给都行，不影响任何内容。", + "tip.p2": "在手机上看的话，先截图，再用微信扫一扫从相册选图。", + "tip.qrAlt": "微信赞赏码", + "will.lots": "很多", + "will.no": "不用", + "will.some": "一点", + "xref.go": "跳过去", + "xref.item": "第 {{sec}} 节第 {{n}} 条", + "xref.wholeSec": "整节，共 {{n}} 条" + } +} diff --git a/tools/site/page.template.html b/tools/site/page.template.html new file mode 100644 index 00000000..fa867a81 --- /dev/null +++ b/tools/site/page.template.html @@ -0,0 +1,1529 @@ + + + + + + +高性价比人生指南 · 用最少的钱、时间和精力换回寿命、金钱和人身自由 + + + + + + + +{{hreflang}} + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    +
    + + 高性价比人生指南 + +
    + + + + + +
    +
    +
    + +
    + + + +
    +
    +
    +

    高性价比人生指南

    +

    每一条都回答两个问题：花掉什么，换回什么。

    +

    换回的算在谁头上也分档：你自己最高，其次配偶和直系亲属，再次朋友同事，陌生人最低——最低不等于零，只是回报的指望小、要连风险一起看。

    +

    不用全做：这是一份按性价比排好的备选单，不是任务清单。挑走一两条就算数，剩下的需要时再回来查——想挑省力的，左边把「花钱」选「不花钱」、「要毅力」选「不用」，剩下的就是。作者自己也没做到其中大部分。正文里带虚线的「第 X 条」可以点开，就地看那条写了什么，不用自己翻过去。

    +
    当前显示 – 条，共 – 条
    +
    长文：遇到陌生人出事该不该停 · 家庭应急装备清单 · 结婚划不划算 · 做平台要办哪些证 · 生物钟和夜班 ·来源核实记录
    +
    AI 助手：让 Claude Code 或 Codex 照这本书回答你的问题（skill 装法）
    +
    看不懂的缩写和名词

    正文里带虚线的词，点一下或把鼠标放上去就有解释。全表如下。

    +
    +
    +
    正在读取正文 …
    + +
    +
    数字口径以条目内标注为准：第 1、2 节是 ACM 或特定死因，第 3、4 节是精力/时间，第 5 节是金钱，第 7 节是金钱与保障，第 8、9 节是金钱与人身自由，互不换算。来源核实过程见 docs/核实记录。正文按 CC BY 4.0 发布，转载改编请署名并附原文链接。
    +
    +
    + + + + + + + + + + diff --git a/tools/site/site.json b/tools/site/site.json new file mode 100644 index 00000000..4564ff93 --- /dev/null +++ b/tools/site/site.json @@ -0,0 +1,5 @@ +{ + "_comment": "站点地址和仓库地址，全书只有这两处和「部署到哪个域名」有关。改域名只改这里。site 不带尾斜杠：模板里是 {{site}}/en/、{{site}}/og.png，带了会拼出双斜杠。locales.json 里每种语言的 canonical 和 repoBlob 由这两个值推出来，build.mjs 每次会对一遍，对不上直接报错。", + "site": "https://streamentry.github.io/HowToLiveBetter", + "repo": "https://github.com/streamentry/HowToLiveBetter" +} diff --git a/tools/site/stats.json b/tools/site/stats.json new file mode 100644 index 00000000..e9fa050b --- /dev/null +++ b/tools/site/stats.json @@ -0,0 +1,16 @@ +{ + "entries": 641, + "sections": 34, + "gradeA": 425, + "gradeB": 165, + "gradeC": 51, + "dispute": 60, + "todo": 2, + "links": 1443, + "ratioTop": 109, + "ratioTopPct": 17, + "ratioHigh": 292, + "ratioHighPct": 46, + "ratioMid": 240, + "ratioMidPct": 37 +} diff --git a/tools/sync-stats.mjs b/tools/sync-stats.mjs index 286928bb..91c9f534 100644 --- a/tools/sync-stats.mjs +++ b/tools/sync-stats.mjs @@ -1,5 +1,5 @@ // 统计对齐：改完条目跑一次。按顺序做四件事： -// ① 重算全书的统计数字，回写 README.md、index.html、tools/og.html； +// ① 重算全书的统计数字，回写 README.md、tools/site/stats.json、tools/og.html； // ② 调 check-refs.mjs 重算 docs/引用对照.md； // ③ 调 check-plain.mjs 查说人话，不合格只提示不中断； // ④ 用无头 Chrome 把 tools/og.html 重新截成 og.png。 @@ -18,7 +18,7 @@ // 数字口径：条目数 = book/*.md 里的 ### 标题数；节数 = book/*.md 的文件数； // A/B/C = 证据等级行的首字母（带（争议）后缀的照样算）；争议 = 备注以「争议」开头的条数； // TODO = 正文里含「待核实」或「TODO」的行数；链接 = 「- 来源：」和「- 备注：」行里的 http(s) 总数； -// 性价比三档的规则抄自 index.html。 +// 性价比三档的规则抄自检索页模板（tools/site/page.template.html）。 // 切行用 /\r?\n/，理由见 check-refs.mjs 文件头。 import { readFileSync, writeFileSync, readdirSync, existsSync, mkdtempSync, rmSync, statSync } from 'node:fs'; import { resolve, dirname, join } from 'node:path'; @@ -30,12 +30,12 @@ const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '..'); const CHECK = process.argv.includes('--check'); const read = f => readFileSync(join(ROOT, f), 'utf8'); -// 档位规则和 index.html 的 COST_W、e.ratio 两行一致；那两行改了这里必须跟着改，所以先比对一次 -const indexText = read('index.html'); +// 档位规则和检索页模板的 COST_W、e.ratio 两行一致；那两行改了这里必须跟着改，所以先比对一次 +const indexText = read('tools/site/page.template.html'); const COST_W_LINE = "const COST_W = { money:{'0':0,'少':1,'多':2}, time:{'少':0,'中':1,'多':2}, will:{'否':0,'些':1,'是':2} };"; const RATIO_LINE = "e.ratio = e.level === '大' ? (e.cs === 0 ? '极高' : (e.cs <= 2 ? '高' : '一般'))"; -if (!indexText.includes(COST_W_LINE)) throw new Error('index.html 的 COST_W 行变了，请同步本脚本里的成本权重'); -if (!indexText.includes(RATIO_LINE)) throw new Error('index.html 的 e.ratio 行变了，请同步本脚本里的档位规则'); +if (!indexText.includes(COST_W_LINE)) throw new Error('检索页模板的 COST_W 行变了，请同步本脚本里的成本权重'); +if (!indexText.includes(RATIO_LINE)) throw new Error('检索页模板的 e.ratio 行变了，请同步本脚本里的档位规则'); const W = { money: { '0': 0, '少': 1, '多': 2 }, @@ -99,15 +99,23 @@ const EDITS = [ ['README.md', '性价比段', /全书 (\d+) 条中性价比极高 \d+ 条（\d+%）、高 \d+ 条（\d+%）、一般 \d+ 条（\d+%）/g, `全书 ${entries} 条中性价比极高 ${ratio['极高']} 条（${pct['极高']}%）、高 ${ratio['高']} 条（${pct['高']}%）、一般 ${ratio['一般']} 条（${pct['一般']}%）`], ['README.md', '正文文件数', /正文按节拆成 (\d+) 个文件/g, `正文按节拆成 ${sections} 个文件`], - ['index.html', '五处描述', /(\d+) 条建议/g, `${entries} 条建议`], - ['index.html', 'numberOfPages', /numberOfPages":(\d+)/g, `numberOfPages":${entries}`], - ['index.html', '页头条目数', /\d+ 节 (\d+) 条/g, `${sections} 节 ${entries} 条`], - ['index.html', '页脚文件数', /下的 (\d+) 个文件/g, `下的 ${sections} 个文件`], ['tools/og.html', 'og 条目数', /(\d+)<\/b> 条建议/g, `${entries} 条建议`], ['tools/og.html', 'og A 级数', /A 级证据 (\d+)<\/b> 条/g, `A 级证据 ${grade.A} 条`], ['tools/og.html', 'og 链接数', /(\d+)<\/b> 条原始文献链接/g, `${links} 条原始文献链接`], ]; +// 检索页的数字不进页面源码：模板和三份字典里写的是 {{entries}} 这类占位符，由 +// tools/site/build.mjs 从 stats.json 填。stats.json 是这几个数字唯一的落点，三种语言共用。 +const STATS_PATH = 'tools/site/stats.json'; +const stats = { + entries, sections, gradeA: grade.A, gradeB: grade.B, gradeC: grade.C, + dispute, todo, links, + ratioTop: ratio['极高'], ratioTopPct: pct['极高'], + ratioHigh: ratio['高'], ratioHighPct: pct['高'], + ratioMid: ratio['一般'], ratioMidPct: pct['一般'], +}; +const statsText = JSON.stringify(stats, null, 2) + '\n'; + const texts = new Map(); const stale = []; for (const [file, label, pattern, repl] of EDITS) { @@ -127,6 +135,10 @@ for (const [file, label, pattern, repl] of EDITS) { } if (CHECK) { + if (read(STATS_PATH) !== statsText) { + stale.push(STATS_PATH); + console.log(` ${STATS_PATH} stats.json：过时`); + } if (stale.length === 0) { console.log('\n统计数字检查通过'); process.exit(0); @@ -136,6 +148,14 @@ if (CHECK) { } for (const [file, text] of texts) if (text !== read(file)) writeFileSync(join(ROOT, file), text); +if (read(STATS_PATH) !== statsText) writeFileSync(join(ROOT, STATS_PATH), statsText); + +// stats.json 变了，三份生成出来的页面跟着变，一起重建，免得本地开的是旧数字 +const siteBuild = spawnSync(process.execPath, [join(ROOT, 'tools', 'site', 'build.mjs'), '--check'], { encoding: 'utf8' }); +if (siteBuild.status !== 0) { + console.log((siteBuild.stdout || '') + (siteBuild.stderr || '')); + throw new Error('tools/site/build.mjs --check 没通过：检索页还没按新统计数字重建。先跑 node tools/site/build.mjs 再提交'); +} // ② 重算交叉引用对照表：插入或删除条目会让后面的「第 X 条」集体错位，而错位后的条号 // 往往仍在范围内（2026-09-19 第 7 节那 6 处就是），只有把「引用 → 目标标题」摊开入库， diff --git "a/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" "b/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" new file mode 100644 index 00000000..07a82416 --- /dev/null +++ "b/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" @@ -0,0 +1,354 @@ +[← 回总目录](../README.md) + +# 1. 不要早死 + +这一节讲三件事：车祸、火灾、中毒这类意外和伤害死亡；少数几种证据最硬的疫苗和筛查；身体出了什么信号就该去查一查。收益那一栏的数字，来自随机分组的试验、跟踪一大批人很多年的研究，或者官方统计。成本那一栏是作者按市面价格的粗略估计，只用来互相比较，不要拿去引用。 + +[← Về mục lục](../README.md) + +# 1. Đừng chết sớm + +Mục này nói ba việc: chết do tai nạn và tổn thương như đụng xe, hỏa hoạn, ngộ độc; vài loại vắc-xin và xét nghiệm sàng lọc có bằng chứng chắc nhất; và những dấu hiệu cơ thể báo ra thì nên đi khám. Con số ở cột Lợi ích lấy từ thử nghiệm chia nhóm ngẫu nhiên, từ nghiên cứu theo dõi một nhóm người lớn trong nhiều năm, hoặc từ thống kê chính thức. Cột Chi phí là ước tính sơ bộ của tác giả theo giá thị trường, chỉ dùng để so sánh các mục với nhau, đừng trích dẫn. + +### 1. Cài dây đai an toàn, cả hàng trước lẫn hàng sau + +- Chi phí: Không tốn tiền. Mỗi lần lên xe thêm 2 giây. +- Nói thẳng: Ngồi hàng trước mà cài dây thì xác suất bị đâm chết khi xảy ra tai nạn giảm khoảng một nửa. Ngồi hàng trước xe tải nhỏ thì giảm sáu phần mười. Năm 2022 ở Mỹ, trong số người chết trong xe mà biết được đã cài dây hay chưa thì một nửa không cài. +- Lợi ích: Cơ quan quản lý an toàn giao thông Hoa Kỳ (NHTSA) ước tính, cài dây an toàn làm giảm 45% nguy cơ chấn thương tử vong ở người ngồi hàng trước xe ô tô. Hàng trước xe tải nhỏ giảm 60%. Năm 2022 ở Mỹ, trong số người chết ở xe chở khách, 50% người biết được đã cài dây hay chưa thì không cài. Tổ chức Y tế Thế giới (WHO) định nghĩa dây an toàn làm giảm tối đa 50% nguy cơ tử vong của người ngồi trong xe. WHO ước tính năm 2021 ở Trung Quốc có 248.099 người chết vì tai nạn giao thông đường bộ, tương đương 17.4 trên 100.000 người +- Mức bằng chứng: A +- Nguồn: NHTSA (2024). Occupant Protection in Passenger Vehicles: 2022 Data (DOT HS 813 573). ; WHO (2025). Road traffic injuries fact sheet. ; WHO GHO RS_196/RS_198 (China, 2021). +- Ghi chú: Hàng sau cũng phải cài. Trong cùng báo cáo đó, 60% người chết ở hàng ghế thứ hai không cài dây. Hai con số 45% và 60% là ước tính Cơ quan quản lý an toàn giao thông Hoa Kỳ (NHTSA) tính ra từ cơ sở dữ liệu tai nạn (dẫn theo Kahane 2015), không phải từ thử nghiệm chia nhóm ngẫu nhiên. Con số tử vong của WHO ở Trung Quốc là ước tính, cao hơn nhiều lần con số Cục Cảnh sát giao thông đăng ký. Hai bên thống kê phạm vi khác nhau, ở đây chỉ dùng cách đo của WHO. + +### 2. Chạy xe máy, xe điện thì đội mũ bảo hiểm và cài chặt dây + +- Chi phí: Mua mũ 100 đến 300 tệ. Mỗi lần đội mũ mất 5 giây. +- Nói thẳng: Đội mũ và cài chặt dây thì xác suất người chạy xe máy chết khi gặp sự cố thấp hơn khoảng bốn phần mười. Chấn thương đầu thấp hơn khoảng bảy phần mười. Dây phải siết, mũ đeo lỏng lẻo trên đầu không tính là đội. +- Lợi ích: Cochrane gộp nhiều nghiên cứu lại với nhau. Người chạy xe máy đội mũ bảo hiểm có nguy cơ tử vong thấp hơn 42%. Giá trị gốc OR 0.58, khoảng tin cậy 95% CI 0.50 đến 0.68. Chấn thương đầu giảm 69% (OR 0.31, 95% CI 0.25 đến 0.38) +- Mức bằng chứng: A +- Nguồn: Liu BC và cộng sự (2008). Helmets for preventing injury in motorcycle riders. Cochrane Database of Systematic Reviews. +- Ghi chú: Mũ phải cài chặt dây, treo trên ghi đơn xe không tính là đội. Số liệu trên đến từ xe máy. Xe điện chưa có nghiên cứu trực tiếp, số đó suy ra theo, vì sau khi va chạm thì kiểu chấn thương đầu giống nhau. + +### 3. Lắp báo động khói; mùa đông đốt than hay dùng gas trong nhà thì lắp thêm báo động ô-xít cacbon + +- Chi phí: Một báo động khói 30 đến 100 tệ. Một báo động ô-xít cacbon 50 đến 150 tệ. Lắp xong thì mỗi năm thay pin một lần. +- Nói thẳng: Báo động khói hoạt động bình thường giảm khoảng sáu phần mười xác suất chết trong hỏa hoạn nhà ở. Có bốn phần mười số vụ cháy xảy ra lúc người ta đang ngủ, chỉ dựa vào khứu giác thì không tỉnh được. Mùa đông đốt than hay dùng gas trong nhà thì lắp thêm báo động ô-xít cacbon. Năm 2018 ở Trung Quốc có 11.500 người chết vì ngộ độc ô-xít cacbon, bảy phần mười trong ba tháng đông chết ở nhà. +- Lợi ích: Một nghiên cứu ở bang Bắc Carolina lấy một loạt hỏa hoạn nhà ở rồi so ngược các hộ có với các hộ không. Hộ có một báo động khói hoạt động bình thường thì nguy cơ tử vong giảm khoảng sáu phần mười. Giá trị gốc OR 0.39, khoảng tin cậy 95% CI 0.18 đến 0.83. Từ 2018 đến 2020 ở Mỹ, trong các vụ cháy chết người ở nhà có người ở, 24% hộ không có báo động khói, 41% lúc cháy người ta đang ngủ. Năm 2018 Trung Quốc báo cáo 11.523 ca chết vì ngộ độc ô-xít cacbon. Trong số ca chết đó, tháng 12 chiếm 72.59%, tháng 1 chiếm 67.42%, tháng 2 chiếm 66.48%, xảy ra tại nhà +- Mức bằng chứng: B +- Nguồn: Marshall SW và cộng sự (1998). Fatal residential fires: who dies and who survives? JAMA. ; USFA (2022). Fatal Fires in Residential Buildings (2018-2020), Topical Fire Report Series 22(2). ; You J và cộng sự (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly. +- Ghi chú: Lý do để ở mức B: phía báo động khói chỉ có một nghiên cứu, cách làm là so ngược hộ có sự cố với hộ không có, không phải thử nghiệm chia nhóm ngẫu nhiên. Bản thân báo động ô-xít cacbon không có nghiên cứu nào chứng minh giảm tỷ lệ tử vong. Lợi ích ở đây suy ra từ việc các ca chết vì ngộ độc dồn vào mùa đông và dồn trong nhà. + +### 4. Đến hạn thì thay ống mềm gas và bếp, không tự sửa ống, từ chối luôn người đến nhà bán gas + +- Chi phí: Một ống mềm gas đạt chuẩn vài chục tệ. Bếp thay mỗi vài năm một lần. Khi thay thì tiện tay xem đã quá hạn chưa. +- Nói thẳng: Một ống mềm đạt chuẩn vài chục tệ. Đến hạn phải thay, đó là nghĩa vụ của người dùng ghi trong Quy định quản lý khí thành phố. Tự sửa ống, tự tháo đồng hồ gas là những việc quy định cấm. Cùng quy định đó còn viết, công ty gas không được bắt bạn mua sản phẩm và dịch vụ do nó chỉ định. Người đến nhà nói「phải thay」thì từ chối thẳng. +- Lợi ích: Quy định xếp việc「及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管」là nghĩa vụ của người dùng: bếp và ống mềm đã hết hạn hay bị nhà nước loại bỏ thì phải thay. Quy định cũng liệt kê「擅自安装、改装、拆除户内燃气设施和燃气计量装置」là hành vi bị cấm, tự sửa ống hay tự tháo đồng hồ gas đều nằm trong đó. Quy định nói rõ, đơn vị kinh doanh gas không được yêu cầu người dùng mua sản phẩm do nó chỉ định, cũng không được yêu cầu người dùng chấp nhận dịch vụ do nó cung cấp +- Mức bằng chứng: A +- Nguồn: 国务院 (Hội đồng Nhà nước) (2010). Quy định quản lý khí thành phố, 城镇燃气管理条例（国务院令第 583 号）. 第二十七条: 「燃气用户应当遵守安全用气规则，使用合格的燃气燃烧器具和气瓶，及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管等」 — người dùng gas phải tuân thủ quy tắc dùng gas an toàn, dùng thiết bị đốt gas và bình gas đạt chuẩn, và kịp thời thay mọi thiết bị đốt gas hoặc ống nối bị nhà nước yêu cầu loại bỏ hay đã hết hạn sử dụng. Các hành vi bị cấm ở 第二十八条 gồm 「（一）擅自操作公用燃气阀门」「（二）将燃气管道作为负重支架或者接地引线」「（四）擅自安装、改装、拆除户内燃气设施和燃气计量装置」 — tự ý vặn van gas công cộng, dùng ống gas làm giá đỡ tải hay làm dây tiếp đất, và tự ý lắp, sửa, tháo thiết bị gas trong nhà cùng thiết bị đo gas. 第二十条 cấm đơn vị kinh doanh gas 「（六）要求燃气用户购买其指定的产品或者接受其提供的服务」 — yêu cầu người dùng gas mua sản phẩm do họ chỉ định hoặc chấp nhận dịch vụ họ cung cấp. 第二十九条 cho phép người dùng khiếu nại về phí và dịch vụ với cơ quan quản lý gas, và 「有关部门应当自收到投诉之日起15个工作日内予以处理」 — cơ quan có trách nhiệm xử lý trong 15 ngày làm việc kể từ khi nhận được khiếu nại. ；深圳市住房和建设局 (Sở Nhà ở và Xây dựng Thâm Quyến) (2024). Kiến thức dùng gas an toàn (khoản 2), 安全用气小知识（二）: 「立即关闭家里的燃气总阀门」— đóng ngay van gas tổng trong nhà, 「迅速开门、开窗进行通风」— mở cửa và cửa sổ để thông gió, 「一定不要开、关家里的电灯、抽油烟机等电器」— tuyệt đối không bật tắt đèn hay máy hút mùi trong nhà, 「人员应该迅速撤离，转移到安全的地方，并且及时拨打燃气公司的报警电话」— mọi người phải rút ra ngay, sang chỗ an toàn, rồi gọi số báo động của công ty gas. ；张家港市应急管理局 (Cục Quản lý Khẩn cấp Trường Gia Cang) (2024). Kiến thức dùng gas an toàn, 燃气安全使用知识: 「不要情急之下打开排气扇进行通风」— đừng vội bật quạt thông gió, 「应远离现场之后才能打电话求助」— phải rời khỏi hiện trường rồi mới gọi cầu cứu. +- Ghi chú: Có người mang danh「kiểm tra an toàn」đến nhà, kiểm tra xong là chào bán van tự đóng và báo động. Quy định không cho phép công ty gas bắt bạn mua sản phẩm do nó chỉ định, bạn từ chối được, khiếu nại cũng được. Ngửi thấy mùi gas thì tắt van gas tổng trước, gas bình thì tắt van trên bình. Rồi mở cửa mở cửa sổ thông gió, đừng bật quạt thông gió. Đèn, máy hút mùi các thiết bị đó đừng bật cũng đừng tắt, bật lửa cũng đừng đụng. Sau đó mọi người rút ra ngoài, đã rời hiện trường thì gọi số báo động của công ty gas. Ngộ độc ô-xít cacbon là chuyện khác, xem phần 13. + +### 5. Không hái, không mua, không ăn nấm hoang, mọi cách nhận biết kiểu dân gian đều không có cơ sở + +- Chi phí: Không tốn tiền. Chỗ khó là từ bỏ một món ăn hoang theo mùa. +- Nói thẳng: Chỉ riêng năm 2025 cả nước đã có 828 vụ ngộ độc nấm, 2165 người dính, 13 người chết. Năm đó còn mới ghi nhận 34 loài nấm trước đây chưa có bản ghi là độc. Thử bằng kim bạc, có gián ăn rồi thì không độc, chỉ nấm màu sặc sỡ mới độc, tất cả đều không đúng. +- Lợi ích: Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc khảo sát 828 vụ ngộ độc nấm năm 2025, liên quan 2165 người, 13 người chết. Tỷ lệ tử vong là 0.6%, tức 0.6% số người bị ngộ độc tử vong. Từ 2019 đến 2024 mỗi năm 276 đến 676 vụ, tỷ lệ tử vong dao động từ 0.87% đến 2.86%. Chỉ riêng năm 2025 đã định danh 138 loài nấm độc, trong đó 34 loài lần đầu được ghi nhận ở Trung Quốc là gây ngộ độc +- Mức bằng chứng: A +- Nguồn: Mushroom Poisoning Outbreaks — China, 2025. China CDC Weekly (2026):「In 2025, China CDC investigated 828 mushroom poisoning incidents across 27 provincial-level administrative divisions (PLADs), affecting 2,165 individuals and causing 13 deaths - a case fatality rate of 0.6%, the lowest of the past six years. In total, 138 poisonous mushroom species were identified, including 34 newly recorded in poisoning incidents in China.」「From 2019 to 2024, the annual number of incidents ranged from 276 to 676, and the case fatality rate ranged from 0.87% to 2.86%.」；báo cáo thường niên năm 2024 trong cùng tạp chí +- Ghi chú: Ăn nhầm thì gây nôn ngay, rồi mang phần nấm còn lại hoặc ảnh chụp đi bệnh viện. Ngộ độc nấm độc chia ra nhiều thể, nhận ra đúng loài nấm quyết định cách bác sĩ điều trị. Mỗi năm lại có loài mới được ghi là độc, nên giả định「người bản xứ nhận ra được」vốn đã không đứng vững. Thử kim, luộc chung với tỏi, có gián ăn rồi thì không độc, chỉ màu sặc sỡ mới độc, tất cả đều không đúng. + +### 6. Không đẩy xe điện vào hành lang, không vào thang máy, không sạc trong nhà + +- Chi phí: Không tốn tiền. Chỗ khó là lần nào cũng phải đậu ở điểm sạc chung dưới lầu, đi thêm vài chục mét. +- Nói thẳng: Đẩy xe điện vào hành lang, vào thang máy, hay sạc trong nhà thì bị Quy định quản lý an toàn cháy nổ công trình dân dụng cao tầng của Bộ Quản lý Khẩn cấp cấm rõ ràng. Không sửa thì cá nhân phạt 500 đến 1000 tệ, người kinh doanh phạt 2000 đến 10000 tệ. Phạt tiền còn là chuyện nhỏ, đáng nói là loại cháy đó cháy ngay trên lối thoát duy nhất của bạn. +- Lợi ích: Quy định của Bộ Quản lý Khẩn cấp viết rõ lệnh cấm:「禁止在高层民用建筑公共门厅、疏散走道、楼梯间、安全出口停放电动自行车或者为电动自行车充电」. Sảnh chung, lối đi sơ tán, cầu thang, lối ra khẩn cấp thì không được dựng xe, cũng không được sạc. Không chịu sửa thì đơn vị, cá nhân không kinh doanh phạt 500 đến 1000 tệ, người kinh doanh phạt 2000 đến 10000 tệ +- Mức bằng chứng: A +- Nguồn: 应急管理部 (Bộ Quản lý Khẩn cấp) (2021). Quy định quản lý an toàn cháy nổ công trình dân dụng cao tầng, 高层民用建筑消防安全管理规定（应急管理部令第 5 号，第三十七条、第四十七条）. ；应急管理部 (Bộ Quản lý Khẩn cấp) (2023). Thư trả lời đề xuất số 00969 của kỳ thứ Năm Ủy ban Quốc gia Đại biểu Nhân dân Trung Quốc, 关于政协第十三届全国委员会第五次会议第00969号提案答复的函: 「2021年，全国共接报电动自行车火灾1.8万起，造成65人死亡、222人受伤，直接财产损失达2.21亿元，其中发生在建筑物内3782起，造成60人死亡、175人受伤」 — năm 2021 cả nước tiếp nhận báo 18.000 vụ cháy xe điện, làm 65 người chết, 222 người bị thương, thiệt hại tài sản trực tiếp 221 triệu tệ, trong đó 3.782 vụ xảy ra trong tòa nhà, làm 60 người chết và 175 người bị thương. ；应急管理部 (Bộ Quản lý Khẩn cấp) (2025). Họp báo ngày 15 tháng 1 năm 2025, 2025年1月15日新闻发布会: 「火灾起数同比下降42.2%，伤亡人数下降超过90%」 — số vụ cháy giảm 42.2% so với cùng kỳ, số người bị thương tử vong giảm hơn 90%. +- Ghi chú: Đừng mua pin độ lại, đừng dùng bộ sạc không rõ hãng, đừng sạc suốt đêm rồi bỏ đó. Pin lithium chừng nào bắt lửa là cháy hết trong mấy chục giây, kèm khói độc dữ. Xe đậu trong hành lang mà cháy thì cháy ngay lối thoát duy nhất của bạn. Số liệu Bộ Quản lý Khẩn cấp công bố: năm 2021 cả nước có 18.000 vụ cháy xe điện, 65 người chết. Trong đó 3.782 vụ cháy trong tòa nhà, 60 người chết, người chết gần như toàn ở trong nhà. Sau đợt xử lý năm 2024, số vụ cháy giảm 42,2% so với năm trước, số người bị thương tử vong giảm hơn chín phần mười, còn số tuyệt đối của năm đó không công bố. + +### 7. Đo huyết áp, cao rồi thì uống thuốc hạ cho tới mức chuẩn + +- Chi phí: Máy đo huyết áp điện tử 100 đến 200 tệ. Đo một lần mất 1 phút. Thuốc hạ huyết áp phần lớn mỗi tháng vài tệ đến vài chục tệ. Chỗ khó là không có triệu chứng vẫn phải uống tiếp. +- Nói thẳng: Huyết áp mỗi khi giảm 10 mmHg, xác suất bệnh tim mạch nặng giảm hai phần mười, đột quỵ giảm 27%, suy tim giảm 28%, xác suất tử vong giảm 13%. Ở Trung Quốc, người 35 đến 75 tuổi có 44.7% bị cao huyết áp, thực sự kiểm soát được chỉ 7.2%. +- Lợi ích: Gộp 123 thử nghiệm với hơn 610.000 người lại với nhau. Huyết áp tâm thu mỗi khi giảm 10 mmHg, sự kiện tim mạch chính thấp hơn khoảng hai phần mười. Giá trị gốc RR 0.80, khoảng tin cậy 95% CI 0.77 đến 0.83. Huyết áp tâm thu là con số cao hơn trong hai số đo. Đột quỵ (中风) thấp hơn khoảng 27% (RR 0.73). Suy tim thấp hơn khoảng 28% (RR 0.72). Tỷ lệ tử vong toàn nguyên giảm 13% (RR 0.87, 95% CI 0.84 đến 0.91). Trung Quốc đã sàng lọc 1.700.000 người 35 đến 75 tuổi: 44.7% bị cao huyết áp. Trong số đó 44.7% biết mình bị, 30.1% đang được điều trị, chỉ 7.2% kiểm soát được +- Mức bằng chứng: A +- Nguồn: Ettehad D và cộng sự (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Lu J và cộng sự (2017). Prevalence, awareness, treatment, and control of hypertension in China (China PEACE Million Persons Project). Lancet. +- Ghi chú: Đi đo huyết áp một lần trước đã. Ở Trung Quốc, trong 10 người cao huyết áp thì chưa đến 1 người kiểm soát được, nên lần đo này là một trong những việc hiệu quả chi phí cao nhất trong sách. Hạ tới bao nhiêu thì coi là đạt (130 hay 140) còn đang bàn, nhưng việc biết mình cao huyết áp và hạ nó xuống thì không ai tranh cãi. + +### 8. Từ 35 tuổi, chỉ cần thừa cân là đi xét nghiệm glucose lúc nhịn một lần, bình thường thì cứ ba năm xét lại + +- Chi phí: Xét nghiệm glucose lúc nhịn mười mấy tệ. Xét nghiệm hemoglobin glycate vài chục tệ. Đến bệnh viện rút máu một lần là xong. +- Nói thẳng: Bỏ ra vài chục tệ để rút một ống máu là biết mình đang ở giai đoạn tiền đái thường, khi chưa có triệu chứng nào. Khuyến cáo sàng lọc chính thức của Mỹ là: người 35 đến 70 tuổi thừa cân hoặc béo thì đi xét nghiệm. Kết quả bình thường thì ba năm xét lại một lần. +- Lợi ích: Nhóm dịch vụ phòng ngừa Hoa Kỳ là cơ quan đưa ra khuyến cáo sàng lọc chính thức của Mỹ. Khuyến cáo hiện tại là: sàng lọc tiền đái thường và đái thường tuýp 2 ở người lớn 35 đến 70 tuổi thừa cân hoặc béo, kể cả khi chưa có triệu chứng nào. Thừa cân là chỉ số khối cơ thể ≥25, béo là ≥30. Khuyến cáo này mang mức khuyến nghị B. Tài liệu còn viết:「对血糖正常的成年人，每 3 年筛查一次是合理的做法」 — với người lớn có đường huyết bình thường, sàng lọc mỗi 3 năm một lần là hợp lý +- Mức bằng chứng: A +- Nguồn: US Preventive Services Task Force (2021). Screening for Prediabetes and Type 2 Diabetes. +- Ghi chú: Phát hiện tiền đái thường không có nghĩa là phải uống thuốc. Khá nhiều người chỉ cần đổi lối sống là xoay ngược lại được. Ở người lớn Trung Quốc, số biết mình mắc đái thường luôn là thiểu số, nhiều người đợi đến lúc đã có biến chứng mới biết + +### 9. Không chạy quá tốc độ, không lái xe sau khi uống rượu + +- Chi phí: Không tốn tiền. Chạy chậm hơn một chút thì mỗi chuyến mất thêm vài phút. Chỗ khó là ở bàn ăn cơm phải uống ít đi. +- Nói thẳng: Tốc độ trung bình tăng mỗi 1%, xác suất đâm chết người tăng khoảng 4%. Rượu không có mức nào uống chút là an toàn. Nồng độ cồn trong máu còn thấp thì nguy cơ đã bắt đầu đi lên. +- Lợi ích: Tổ chức Y tế Thế giới (WHO) nói: tốc độ trung bình tăng mỗi 1% thì nguy cơ tai nạn chết người tăng 4%. Nguy cơ khi lái xe sau khi uống bắt đầu tăng từ một nồng độ cồn trong máu rất thấp +- Mức bằng chứng: B +- Nguồn: WHO (2025). Road traffic injuries fact sheet. +- Ghi chú: Cặp số 1% và 4% về tốc độ và tử vong được suy ra từ mô hình tai nạn, không phải từ thử nghiệm chia nhóm ngẫu nhiên. Về lái xe sau khi uống, WHO chỉ nêu hướng, không đưa số. Câu hỏi「uống bao nhiêu thì nguy cơ tăng bao nhiêu」mục này chưa đối chiếu tài liệu gốc. + +### 10. Trẻ dưới 4 tuổi thì dùng ghế an toàn, đừng bế trên tay + +- Chi phí: Một cái 300 đến 2000 tệ, dùng được mấy năm. Mỗi lần lên xe thêm 1 phút để cài đai cho con. +- Nói thẳng: Trẻ sơ sinh dưới 1 tuổi ngồi ghế an toàn thì xác suất chấn thương tử vong khi xảy ra tai nạn thấp hơn khoảng bảy phần mười. Trẻ 1 đến 4 tuổi thấp hơn khoảng một nửa. Bế trên tay thì không có tác dụng đó, phanh gấp thì bạn cũng không giữ được. +- Lợi ích: Cơ quan quản lý an toàn giao thông Hoa Kỳ (NHTSA) ước tính, dùng ghế an toàn làm giảm 71% nguy cơ chấn thương tử vong cho trẻ sơ sinh dưới 1 tuổi trên xe ô tô. Trẻ 1 đến 4 tuổi giảm 54%. Tổ chức Y tế Thế giới (WHO) định nghĩa tử vong trẻ sơ sinh giảm 71% +- Mức bằng chứng: A +- Nguồn: NHTSA (2024). Occupant Protection in Passenger Vehicles: 2022 Data (DOT HS 813 573). ; WHO (2025). Road traffic injuries fact sheet. +- Ghi chú: Lắp ghế theo hướng dẫn, lắp sai coi như không lắp. Dưới 1 tuổi thì lắp ngược, cho trẻ quay về phía đuôi xe. Số liệu trên là ước tính chính thức của Mỹ tính ra từ dữ liệu tai nạn. + +### 11. Nhà có trẻ nhỏ thì lắp chặn cửa sổ và ban công, lưới chống muỗi không tính là biện pháp bảo vệ + +- Chi phí: Một bộ chặn cửa sổ hoặc khoá an toàn trẻ em vài chục tệ. Lắp xong mất mười phút. +- Nói thẳng: Một bộ chặn cửa sổ vài chục tệ. Thành phố New York lắp miễn phí thanh chắn cửa sổ cho hộ có nguy cơ cao, lại cửa người đến tận nhà hướng dẫn. Ở một quận, số vụ trẻ em ngã được báo cáo giảm một nửa trong ba năm. Sau đó New York còn sửa luật, bắt chủ nhà phải lắp thanh chắn cho căn hộ có trẻ em. Lưới chống muỗi đẩy là rơi, không phải biện pháp bảo vệ. +- Lợi ích: Thành phố New York mở chương trình tên「孩子不会飞」(Trẻ em không biết bay) từ năm 1972, phát thanh chắn cửa sổ miễn phí ở các khu vực có nguy cơ cao, lại cửa người đến tận nhà hướng dẫn. Ở quận Bronx, số vụ ngã được báo cáo từ 1973 đến 1975 giảm 50%. Sau đó thành phố New York sửa bộ luật vệ sinh, bắt chủ nhà phải lắp thanh chắn cửa sổ cho mọi căn hộ có trẻ em từ 10 tuổi trở xuống +- Mức bằng chứng: B +- Nguồn: Spiegel CN, Lindaman FC (1977). Children can't fly: a program to prevent childhood morbidity and mortality from window falls. American Journal of Public Health：「Significant reduction in falls resulted, particularly in the Bronx, where reported falls declined 50 percent from 1973 to 1975」— số vụ ngã giảm đáng kể, nổi bật ở Bronx, nơi số vụ ngã được báo cáo giảm một nửa từ 1973 đến 1975, và bản sửa đổi bộ luật vệ sinh thành phố New York năm 1976「to require that landlords provide window guards in apartments where children ten years old and younger reside」— buộc chủ nhà phải lắp thanh chắn cửa sổ ở căn hộ có trẻ em từ mười tuổi trở xuống. ；(2021). Unintentional Window Falls in Children and Adolescents. Academic Pediatrics：38.840 lượt khám cấp cứu ghi nhận trong hệ thống giám sát tổn thương điện tử toàn quốc từ tháng 1 năm 2007 đến tháng 8 năm 2017，「The majority of falls occurred in children under the age of 6 and were related to falls from a second story or below」— phần lớn vụ ngã xảy ra ở trẻ dưới 6 tuổi và liên quan đến rơi từ tầng hai trở xuống. +- Ghi chú: Không đặt giường, sofa hay thùng ở mép cửa sổ và mép ban công, những thứ trẻ leo lên được. Lưới chống muỗi chỉ chặn muỗi, đẩy là rơi, đừng coi là thanh chắn. Cũng không phải chỉ ai ở tầng cao mới cần lắp: trong số liệu của Mỹ, phần lớn vụ ngã xảy ra ở tầng hai hoặc thấp hơn. + +### 12. Trẻ nhỏ ở gần nước thì không được rời mắt, chèo thuyền hay bơi ngoài vùng nước an toàn thì mặc áo phao + +- Chi phí: Áo phao 50 đến 200 tệ. Ngoài ra còn phải bỏ ra phần tâm trí để trông trẻ. +- Nói thẳng: Khi rơi xuống nước mà có mặc áo phao, xác suất bị chết đuối chỉ còn khoảng một nửa. Đến nay đuối nước vẫn là nguyên nhân tử vong số một ở trẻ 1 đến 14 tuổi Trung Quốc. Tỷ lệ tử vong do đuối nước ở nông thôn gấp khoảng hai lần thành phố. +- Lợi ích: Một nghiên cứu dùng số liệu của Bộ Cảnh gia Hoa Kỳ, ghép từng cặp người có điều kiện gần giống nhau để so sánh. Người ngồi trên thuyền du lịch bị rơi xuống nước mà mặc áo phao có nguy cơ chết đuối thấp hơn khoảng một nửa. Giá trị gốc là RR 0.51 sau hiệu chỉnh, khoảng tin cậy 95% CI 0.35 đến 0.74. Ở Trung Quốc, tỷ lệ tử vong do đuối nước dưới 20 tuổi giảm từ 6.60 trên 100.000 người năm 2013 xuống 3.28 trên 100.000 người năm 2021. Nông thôn gấp khoảng 2 lần thành phố. Đuối nước vẫn là nguyên nhân tử vong số một ở trẻ 1 đến 14 tuổi. Năm 2021, trong số ca tử vong do tổn thương ở nhóm 0 đến 19 tuổi, đuối nước chiếm 31.1%, tai nạn giao thông đường bộ chiếm 27.9% +- Mức bằng chứng: A +- Nguồn: Cummings P và cộng sự (2011). Association between wearing a personal floatation device and death by drowning among recreational boaters. Injury Prevention. ; Li Z và cộng sự (2023). Unintentional Drowning Mortality Among Individuals Under Age 20 — China, 2013–2021. China CDC Weekly. ; Zhou J và cộng sự (2024). Injury Mortality of Children and Adolescents Aged 0–19 Years — China, 2010–2021. China CDC Weekly. +- Ghi chú: Mức A chỉ áp dụng cho phần áo phao.「Không rời mắt」là đồng thuận chung, tính riêng chỉ ở mức C, không có thử nghiệm chia nhóm ngẫu nhiên, và sau này cũng sẽ không có. Ở Trung Quốc, các ca chết đuối tập trung ở nông thôn, mùa hè, và ở nam giới 15 đến 19 tuổi. + +### 13. Từ 60 tuổi, tập giữ thăng bằng và sức chân, cải tạo phòng tắm và cầu thang trong nhà + +- Chi phí: Mỗi tuần tập 2 đến 3 lần, mỗi lần 30 phút thái cực hoặc bài tập giữ thăng bằng. Thảm chống trượt, thanh vịn, đèn ngủ trong phòng tắm vài chục đến vài trăm tệ. Chỗ khó là phải tập bền nhiều năm. +- Nói thẳng: Tập giữ thăng bằng mỗi tuần hai ba lần, thái cực chẳng hạn, số lần ngã ở người cao tuổi giảm khoảng hai phần rưỡi, số người từng ngã giảm 15%. Làm chống trượt và lắp thanh vịn trong nhà thì giảm thêm khoảng hai phần mười. Ngã là nguyên nhân tử vong do tổn thương số một ở người Trung Quốc trên 65 tuổi, và hơn một nửa số vụ xảy ra ngay trong nhà họ. +- Lợi ích: Cochrane gộp 108 thử nghiệm, 23.407 người lại với nhau. Tập thể dục làm số lần ngã ở người cao tuổi giảm 23%. Giá trị gốc là tỷ lệ ngã RaR 0.77, khoảng tin cậy 95% CI 0.71 đến 0.83. Số người từng ngã giảm 15% (RR 0.85). Mời người đến nhà đánh giá an toàn và cải tạo, tỷ lệ ngã giảm khoảng hai phần mười (RR 0.81, 95% CI 0.68 đến 0.97). Tập thái cực, nguy cơ ngã thấp hơn khoảng 29% (RR 0.71). Ở Trung Quốc, ngã là nguyên nhân tử vong do tổn thương số một ở người trên 65 tuổi. Trong năm 2018, 55.97% số vụ ngã được theo dõi ở người cao tuổi xảy ra tại nhà +- Mức bằng chứng: A +- Nguồn: Sherrington C và cộng sự (2019). Exercise for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Gillespie LD và cộng sự (2012). Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Lu Z và cộng sự (2021). Characteristics of Falls Among Older People — China, 2018. China CDC Weekly. +- Ghi chú: Mục này áp dụng cho người trên 60 tuổi sống ở nhà mình, không ở viện dưỡng lão. Các nghiên cứu trên đếm số lần ngã, không đếm số người chết. Đường đi từ một cú ngã đến gãy xương hông rồi đến tử vong là rất ngắn. + +### 14. Xét nghiệm 5 chỉ số viêm gan B, không có kháng thể thì tiêm vaccine + +- Chi phí: Xét nghiệm một lần vài chục tệ. Vaccine viêm gan B cho người lớn gồm 3 mũi, tổng cộng 100 đến 300 tệ. Tiêm hết trong nửa năm. +- Nói thẳng: Xét nghiệm một lần vài chục tệ. Trong thử nghiệm theo dõi 30 năm ở Qidong, người được tiêm vaccine viêm gan B ngay khi sinh sau này có xác suất mắc ung thư gan nguyên phát thấp hơn 84%, tỷ lệ mang virus viêm gan B thấp hơn 72%. Người lớn xét ra không có kháng thể thì tiêm bổ sung ba mũi, xong trong nửa năm, một hai trăm tệ. +- Lợi ích: Thử nghiệm ở Qidong chia nhóm ngẫu nhiên theo cộng đồng, không phải theo từng người, và theo dõi 30 năm. Trẻ sơ sinh tiêm vaccine viêm gan B, ung thư gan nguyên phát giảm 84%. Khoảng tin cậy là 95% CI 23% đến 97%. Tỷ lệ dương tính kháng nguyên bề mặt viêm gan B HBsAg giảm 72% (95% CI 68% đến 75%), tỷ lệ dương tính chính là tỷ lệ người được xét ra đang mang virus. Ở toàn bộ dân số Trung Quốc, tỷ lệ dương tính HBsAg từ 1992 đến 2014 giảm 52%, nhóm dưới 5 tuổi giảm 97% +- Mức bằng chứng: A +- Nguồn: Qu C và cộng sự (2014). Efficacy of neonatal HBV vaccination on liver cancer and other liver diseases over 30-year follow-up of the Qidong hepatitis B intervention study. PLoS Medicine. ; Cui F và cộng sự (2017). Prevention of Chronic Hepatitis B after 3 Decades of Escalating Vaccination Policy, China. Emerging Infectious Diseases. +- Ghi chú: Mục này áp dụng cho người lớn sinh trước năm 1992, chưa tiêm vaccine hoặc đã mất kháng thể. Con số giảm 84% ung thư gan đến từ nhóm được tiêm ngay khi sinh. Trong cùng thử nghiệm đó, người đến tuổi thanh niên mới tiêm bổ sung chỉ ngăn được 21% khả năng mang virus. Người lớn tiêm chủ yếu để không bị nhiễm, không thể áp thẳng con số 84% đó. Người đã nhiễm thì tiêm vaccine không có tác dụng, phải đi khoa gan theo dõi định kỳ. Cách biết đã nhiễm: trên phiếu xét nghiệm, mục kháng nguyên bề mặt viêm gan B HBsAg là dương tính. + +### 15. Bị đinh sắt, gai gỗ đâm vào hoặc vết thương dính đất thì xử lý ngay trong ngày, tiện thể hỏi có cần tiêm phòng uốn dương không + +- Chi phí: Khám và xử lý vết thương vài chục tệ. Nếu bác sĩ bảo phải tiêm, vaccine hoặc globulin miễn dịch vài chục đến vài trăm tệ. +- Nói thẳng: Bệnh uốn dương một khi phát, khoảng 1 trong 10 người chết. Vết thương sâu hay dính đất có nguy cơ cao nhất. Vết thương bẩn mà đã cách mũi tiêm trước đủ 5 năm thì tiêm nhắc thêm 1 mũi. Xử lý trong ngày chỉ mất vài chục tệ, đừng kéo đến lúc vết thương sưng mới đi. +- Lợi ích: Trong số ca uốn dương ở Mỹ, khoảng 1 trong 10 ca tử vong. Vết thương sâu hay nhiễm đất có nguy cơ cao hơn. Quy chuẩn chẩn đoán và điều trị năm 2024 của Trung Quốc chia vết thương thành hai bậc. Vết thương nông xử lý trong vòng 6 giờ, không dính đất, phân hay nước dãi động vật, được xếp là nguy cơ thấp. Xử lý quá 6 giờ, dính các thứ đó, hoặc là vết đâm, vết xé da, vết bóp nát, vết bỏng, vết băng giá, hoặc còn dị vật trong vết thương, được xếp là nguy cơ cao. Người đã tiêm từ 3 mũi trở lên, mũi cuối cách dưới 5 năm, thì không cần tiêm gì thêm. Cách 5 đến 10 năm thì vết thương nguy cơ cao phải tiêm nhắc 1 mũi. Đủ 10 năm thì bậc nào cũng tiêm nhắc 1 mũi. Người tiêm chưa đủ 3 mũi hoặc không nhớ rõ, phải bổ sung cho đủ vaccine, với vết thương nguy cơ cao thì còn phải tiêm thêm thuốc miễn dịch bị động (các mũi tiêm đưa kháng thể thẳng vào cơ thể, như globulin miễn dịch hay kháng độc tố) +- Mức bằng chứng: B +- Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ (美国疾病控制与预防中心). 破伤风 (uốn dương)：「Tetanus bacteria can get into someone's body through broken skin, usually through injuries.」— vi khuẩn uốn dương lọt vào cơ thể qua da bị rách, thường là qua vết thương.「Tetanus can lead to death (1 in 10 cases in the United States are fatal).」— bệnh này có thể dẫn tới tử vong, ở Mỹ cứ khoảng mười ca thì một ca chết.「People who didn't complete the primary series or who aren't up to date with their 10-year tetanus booster shots are also at increased risk.」— người tiêm chưa đủ liệu trình cơ bản, hoặc chưa đến hạn tiêm nhắc mười năm một lần, cũng có nguy cơ cao hơn.「Vaccination also helps prevent tetanus in people with wounds, depending on their tetanus vaccination history.」— tiêm vaccine còn giúp phòng uốn dương cho người có vết thương, tuỳ vào tiền sử tiêm phòng của họ. ；国家卫生健康委办公厅 (Văn phòng Ủy ban Quốc gia Y tế và Sức khoẻ) (2024). 关于印发非新生儿破伤风诊疗规范（2024 年版）的通知 — Thông báo ban hành Quy chuẩn chẩn đoán và điều trị uốn dương ở người không phải trẻ sơ sinh, bản 2024 (国卫办医急函〔2024〕381 号, bậc nguy cơ nằm ở trang 16 phụ lục, bảng 4 ở trang 20). ；phụ lục dạng PDF (bản quét). +- Ghi chú: Tới bệnh viện để bác sĩ quyết định, đừng tự ý hạ tiêm, đừng thấy phiền mà không hỏi. Không phải vết thương nào cũng cần tiêm. Bác sĩ sẽ xem vết thương thuộc bậc nào, hỏi lần tiêm gần nhất cách đây bao lâu, rồi mới quyết định.「Tiêm một lần hồi nhỏ thì hiệu lực cả đời」là sai. Vết thương bẩn thì cách mũi tiêm trước đủ 5 năm phải tiêm nhắc. Nên trước khi đi, nhớ lần cuối cùng bạn tiêm phòng uốn dương là năm nào. + +### 16. Phụ nữ tiêm vaccine HPV, càng sớm càng tốt + +- Chi phí: Vaccine hai giá trị sản xuất trong nước hơn 300 tệ một mũi, vaccine chín giá trị nhập khẩu khoảng 1300 tệ một mũi. Tổng cộng 2 đến 3 mũi. Mỗi lần chạy đi tiêm mất nửa ngày. +- Nói thẳng: Trong số liệu của 1.670.000 phụ nữ Thụy Điển, người tiêm vaccine HPV trước 17 tuổi sau này có xác suất mắc ung thư cổ tử cung xâm lấn chỉ khoảng một phần tám so với người không tiêm. Người đến 17 đến 30 tuổi mới tiêm thì cũng chỉ còn khoảng một nửa. Nên tiêm càng sớm càng đáng. +- Lợi ích: Thụy Điển theo dõi 1.670.000 phụ nữ để thống kê: người đã tiêm vaccine có tỷ lệ mắc ung thư cổ tử cung xâm lấn thấp hơn người không tiêm. Tiêm trước 17 tuổi thấp hơn khoảng 88% (IRR 0.12, 95% CI 0.00 đến 0.34, trong ngoặc là giá trị gốc và khoảng tin cậy của nó). Tiêm ở 17 đến 30 tuổi thấp hơn khoảng 53% (IRR 0.47, 95% CI 0.27 đến 0.75) +- Mức bằng chứng: A +- Nguồn: Lei J và cộng sự (2020). HPV Vaccination and the Risk of Invasive Cervical Cancer. New England Journal of Medicine. +- Ghi chú: Tiêm ở 9 đến 14 tuổi thì lợi ích lớn nhất. Tiêm ở 17 đến 30 tuổi vẫn còn hơn một nửa mức giảm. Sau 30 tuổi lợi ích cứ giảm dần. Đã tiêm vaccine vẫn phải làm xét nghiệm sàng lọc ung thư cổ tử cung, vaccine không thay thế được việc sàng lọc, xem mục 18 của phần này (phụ nữ trên 30 tuổi làm xét nghiệm sàng lọc ung thư cổ tử cung). + +### 17. Phụ nữ từ 40 tuổi làm xét nghiệm sàng lọc ung thư vú, chụp mammogram mỗi hai năm một lần + +- Chi phí: Mỗi lần chụp mammogram hai ba trăm tệ. Nhiều địa phương có chương trình sàng lọc miễn phí. Hai năm một lần, mất nửa ngày. +- Nói thẳng: Chụp mammogram mỗi hai năm một lần ở phụ nữ 40 đến 74 tuổi là khuyến cáo rõ ràng trong hướng dẫn sàng lọc chính thức của Mỹ. Với phụ nữ trên 75 tuổi, và với vú mật độ cao có thêm siêu âm và cộng hưởng từ, kết luận chính thức là chưa đủ bằng chứng. Gói khám sức khoẻ nào tính tiền riêng cho các mục này thì không cần mua. +- Lợi ích: Nhóm dịch vụ phòng ngừa Hoa Kỳ là cơ quan đưa ra khuyến cáo sàng lọc chính thức của Mỹ. Khuyến cáo hiện tại là:「建议 40 至 74 岁女性每两年做一次筛查性乳腺 X 线摄影」— khuyến cáo phụ nữ từ 40 đến 74 tuổi chụp X-quang sàng lọc vú mỗi hai năm một lần. X-quang vú chính là loại chụp mà dân gian gọi tắt là mammogram. Khuyến cáo này mang mức khuyến nghị B. Với phụ nữ từ 75 tuổi trở lên, và với vú mật độ cao có thêm siêu âm, cộng hưởng từ, cơ quan này đều kết luận「现有证据不足」— bằng chứng hiện có chưa đủ +- Mức bằng chứng: A +- Nguồn: US Preventive Services Task Force (2024). Breast Cancer: Screening. +- Ghi chú: Bắt đầu sàng lọc từ tuổi nào thì mỗi nước khác nhau, ở Trung Quốc có hướng dẫn và chương trình địa phương bắt đầu từ 45 tuổi. Người có cha mẹ, con cái hay anh chị em ruột mắc ung thư vú, và người mang đột biến BRCA, thuộc nhóm nguy cơ cao, phải tự đi tìm bác sĩ để lập kế hoạch riêng, mục này không áp dụng + +### 18. Phụ nữ trên 30 tuổi làm xét nghiệm sàng lọc ung thư cổ tử cung, ưu tiên xét nghiệm HPV + +- Chi phí: Xét nghiệm HPV một lần 100 đến 300 tệ. Kết quả âm tính thì 5 năm xét một lần. Lấy mẫu chỉ mất vài phút. +- Nói thẳng: Trong thử nghiệm ở nông thôn Ấn Độ, phụ nữ chỉ làm xét nghiệm HPV một lần trong cả đời có xác suất chết vì ung thư cổ tử cung thấp hơn khoảng một nửa, xác suất được phát hiện ung thư cổ tử cung giai đoạn muộn thấp hơn khoảng một nửa. Một lần một hai trăm tệ, kết quả âm tính dùng được 5 năm. +- Lợi ích: Thử nghiệm ở nông thôn Ấn Độ chia nhóm ngẫu nhiên theo cộng đồng, không phải theo từng người. Đối tượng là phụ nữ 30 đến 59 tuổi. Chỉ làm xét nghiệm sàng lọc HPV một lần, nguy cơ chết vì ung thư cổ tử cung thấp hơn khoảng 48%. Giá trị gốc HR 0.52, khoảng tin cậy 95% CI 0.33 đến 0.83. Nguy cơ được phát hiện ung thư cổ tử cung giai đoạn muộn thấp hơn khoảng 53% (HR 0.47, 95% CI 0.32 đến 0.69) +- Mức bằng chứng: A +- Nguồn: Sankaranarayanan R và cộng sự (2009). HPV screening for cervical cancer in rural India. New England Journal of Medicine. +- Ghi chú: Người đã tiêm vaccine HPV vẫn phải sàng lọc, vì vaccine không phủ hết các type virus. Chỗ đặc biệt của nghiên cứu này là: nó là thử nghiệm sàng lọc chia nhóm ngẫu nhiên, lấy chính số người tử vong làm kết quả, và chỉ làm một vòng sàng lọc. + +### 19. Từ 45 đến 50 tuổi làm xét nghiệm sàng lọc ung thư trực tràng và ruột già, bằng xét nghiệm miễn dịch hoá học tìm máu ẩn trong phân hoặc nội soi đại tràng + +- Chi phí: Xét nghiệm phân vài chục tệ, mỗi 1 đến 2 năm làm một lần. Nội soi đại tràng vài trăm đến hơn một nghìn tệ, kết quả âm tính thì 10 năm làm một lần. Trước khi nội soi phải chừa một ngày để làm sạch ruột. +- Nói thẳng: Làm xét nghiệm máu ẩn trong phân vài chục tệ mỗi một hai năm một lần, xác suất chết vì ung thư trực tràng và ruột già thấp hơn khoảng 16%, người đã tham gia ít nhất một vòng thì thấp hơn 25%. Người được mời đi nội soi đại tràng một lần, xác suất mắc ung thư trực tràng và ruột già trong mười năm sau giảm từ 1.20% xuống 0.98%. +- Lợi ích: Cochrane gộp nhiều nghiên cứu lại với nhau. Người làm xét nghiệm sàng lọc máu ẩn trong phân có nguy cơ chết vì ung thư trực tràng và ruột già thấp hơn khoảng 16%. Giá trị gốc RR 0.84, khoảng tin cậy 95% CI 0.78 đến 0.90. Người tham gia ít nhất một vòng thì thấp hơn khoảng 25% (RR 0.75). Xem thêm thử nghiệm ngẫu nhiên NordICC. Người được mời đi nội soi đại tràng, tỷ lệ mắc ung thư trực tràng và ruột già trong 10 năm giảm từ 1.20% xuống 0.98%. Giá trị gốc RR 0.82, khoảng tin cậy 95% CI 0.70 đến 0.93 +- Mức bằng chứng: A +- Nguồn: Hewitson P và cộng sự (2007). Screening for colorectal cancer using the faecal occult blood test, Hemoccult. Cochrane Database of Systematic Reviews. ; Bretthauer M và cộng sự (2022). Effect of Colonoscopy Screening on Risks of Colorectal Cancer and Related Death. New England Journal of Medicine. +- Ghi chú:Tranh cãi. NordICC tính trên nhóm「được mời đi nội soi đại tràng」, tỷ lệ chết vì ung thư trực tràng và ruột già là 0.28% so với 0.31% (RR 0.90, 95% CI 0.64 đến 1.16), chênh lệch vẫn nằm trong biên độ sai số. Điều này cho thấy lợi ích của nội soi đại tràng với tỷ lệ tử vong trước đây bị ước tính thừa. Bằng chứng của xét nghiệm phân về tỷ lệ tử vong lại chắc hơn. Hướng dẫn các nước đặt tuổi bắt đầu sàng lọc trong khoảng 45 đến 50. Ai trong nhà đã mắc bệnh này thì bắt đầu sớm hơn. + +### 20. Người bệnh tim mạch và người cao tuổi tiêm vaccine cúm mỗi năm + +- Chi phí: Một mũi 50 đến 150 tệ. Mỗi mùa thu tiêm một lần. +- Nói thẳng: Trong thử nghiệm tiêm vaccine cúm sau nhồi máu cơ tim, tỷ lệ chết trong vòng một năm giảm từ 4.9% xuống 2.9%. Người bệnh tim mạch và người cao tuổi thì mỗi mùa thu tiêm một mũi, 50 đến 150 tệ. +- Lợi ích: Một thử nghiệm làm sau nhồi máu cơ tim, 2.571 người chia nhóm ngẫu nhiên. Bác sĩ và người bệnh đều không biết mình tiêm mũi nào. Ở nhóm tiêm vaccine cúm, tỷ lệ tử vong toàn nguyên trong 12 tháng là 2.9%. Nhóm tiêm giả dược là 4.9%. Giá trị gốc HR 0.59, khoảng tin cậy 95% CI 0.39 đến 0.89. Giả dược là mũi tiêm đối chứng không có thành phần hiệu quả. HR của tử vong tim mạch là 0.59. Một nghiên cứu khác gộp nhiều nghiên cứu lại với nhau. Ở nhóm tiêm vaccine cúm, sự kiện tim mạch chính chiếm 3.6%, nhóm đối chứng là 5.4%. Giá trị gốc RR 0.66, khoảng tin cậy 95% CI 0.53 đến 0.83. Tử vong tim mạch RR 0.74 (95% CI 0.42 đến 1.30, chênh lệch này vẫn nằm trong biên độ sai số, chưa có ý nghĩa thống kê) +- Mức bằng chứng: A +- Nguồn: Fröbert O và cộng sự (2021). Influenza Vaccination After Myocardial Infarction: A Randomized, Double-Blind, Placebo-Controlled, Multicenter Trial. Circulation. ; Behrouzi B và cộng sự (2022). Association of Influenza Vaccination With Cardiovascular Risk: A Meta-analysis. JAMA Network Open. ；Demicheli V và cộng sự (2018). Vaccines for preventing influenza in the elderly. Cochrane Database of Systematic Reviews. （bản tổng hợp Cochrane mà phần Ghi chú nhắc tới） +- Ghi chú:Tranh cãi. Mức A chỉ áp dụng cho người bệnh tim mạch. Với người cao tuổi vốn khoẻ, bản tổng hợp Cochrane chỉ đưa ra một kết luận: tỷ lệ mắc cúm trong một mùa giảm từ 6% xuống 2.4%. Bản đó đánh giá độ tin cậy của bằng chứng này là「低」— thấp, còn bằng chứng về tỷ lệ tử vong là「极低」— rất thấp. Bản tổng hợp nằm ở bài của Demicheli V và cộng sự năm 2018. + + +### 21. Từ 50 tuổi, tiêm vaccine thủy độc + +- Chi phí: Vaccine thủy độc tái tổ hợp phải tiêm hai mũi, tổng cộng khoảng 3.000 đến 4.000 tệ. Toàn bộ số tiền này người bệnh tự bỏ ra. +- Nói thẳng: Trong thử nghiệm giai đoạn ba, vaccine này chặn được 97% ca thủy độc, các nhóm tuổi đều như nhau. Nhược điểm là hai mũi, ba bốn nghìn tệ, tự túi. Số tiền ấy đổi lấy chín phần trăm mức giảm mắc bệnh. +- Lợi ích: Một thử nghiệm giai đoạn ba ở 18 nước, 15.400 người. Thử nghiệm chia ngẫu nhiên thành hai nhóm, một nhóm tiêm vaccine, một nhóm tiêm mũi đối chứng, theo dõi trung bình 3.2 năm. Kết quả: nhóm vaccine có 6 ca, nhóm giả dược có 210 ca mắc thủy độc. Quy đổi ra 0.3 so với 9.1 trên một nghìn người năm. Nghìn người năm nghĩa là số ca xuất hiện trong 1.000 người quan sát trong một năm. Nguyên văn ghi「疫苗对带状疱疹的总体效力为 97.2%（95% CI 93.7–99.0，P<0.001）」— hiệu lực chung với thủy độc là 97.2%. Hiệu lực ở từng nhóm tuổi nằm trong khoảng 96.6% đến 97.9% +- Mức bằng chứng: A +- Nguồn: Lal H, Cunningham AL, Godeaux O, et al. (2015). Efficacy of an adjuvanted herpes zoster subunit vaccine in older adults. New England Journal of Medicine, 372(22), 2087-2096. +- Ghi chú: Xếp hạng hiệu quả chi phí của mục này khá thấp, lý do là đắt. Bản thân bệnh thủy độc hiếm khi chết người. Chỗ khó chịu là đau thần kinh sau thủy độc, có thể đau mấy tháng đến mấy năm. Sau khi tiêm, tỷ lệ đau chỗ tiêm và sốt đều không thấp. Nặng nhất trong đó, tức tác dụng phụ bậc ba, là 17.0% so với 3.2% + +### 22. Người từ 65 tuổi tiêm vaccine phế cầu khuẩu + +- Chi phí: Vài trăm tệ. Ở một số vùng, người cao tuổi được tiêm miễn phí. +- Nói thẳng: Trong thử nghiệm 84.500 người cao tuổi, vaccine phế cầu 13 giá trị chặn được khoảng 46% viêm phổi thuộc nhóm vaccine che được, và 75% nhiễm xâm lấn nguy hiểm nhất. Nhưng với viêm phổi do mọi nguyên nhân thì không thấy hiệu quả. Đừng trông chờ nó chặn hết viêm phổi. +- Lợi ích: Hà Lan chia ngẫu nhiên 84.500 người từ 65 tuổi trở lên để so sánh. Dùng vaccine phế cầu liên hợp 13 giá trị. Hiệu lực với viêm phổi thuộc nhóm vaccine che được là 45.6%. Khoảng tin cậy là 95.2% CI 21.8–62.5. Nhóm vaccine che được nghĩa là vài chủng vi khuẩn mà vaccine này kiểm soát được. Hiệu lực với bệnh phế cầu xâm lấn là 75.0%（95% CI 41.4–90.8）, đây là nhóm nhiễm nguy hiểm nhất. Với viêm phổi gộp mọi nguyên nhân, vaccine không có tác dụng +- Mức bằng chứng: A +- Nguồn: Bonten MJ, Huijts SM, Bolkenbaas M, et al. (2015). Polysaccharide conjugate vaccine against pneumococcal pneumonia in adults. New England Journal of Medicine, 372(12), 1114-1125. +- Ghi chú: Nó chỉ phòng vài chủng có trong danh sách, không phải tiêm xong là chắc không viêm phổi. Người cao tuổi ở Trung Quốc hay tiêm là vaccine đa đường 23 giá trị, khác loại với vaccine liên hợp 13 giá trị dùng trong thử nghiệm này. Số hiệu lực ở trên không áp thẳng sang loại đa đường được. + +### 23. Xét nghiệm vi khuẩn H. pylori, dương tính thì diệt trùng + +- Chi phí: Xét nghiệm hơi thở carbon 13 hoặc carbon 14 giá 100 đến 200 tệ. Diệt trùng phải uống bốn loại thuốc cùng lúc, hai tuần thuốc giá 200 đến 500 tệ. Chỗ khó là suốt hai tuần phải uống đúng giờ. +- Nói thẳng: Trong thử nghiệm ở Linqu, Sơn Đông, theo dõi 22 năm, người được diệt H. pylori sau này có xác suất mắc ung thư dạ dày thấp hơn khoảng một nửa, xác suất chết vì ung thư dạ dày thấp hơn khoảng bốn phần mười. Xét một lần một hai trăm tệ, diệt trùng phải uống thuốc hai tuần. +- Lợi ích: Thử nghiệm chia nhóm ngẫu nhiên ở Linqu, Sơn Đông, theo dõi 22 năm. Người đã điều trị diệt H. pylori có nguy cơ mắc ung thư dạ dày thấp hơn khoảng một nửa. Giá trị gốc OR 0.48, khoảng tin cậy 95% CI 0.32 đến 0.71. Nguy cơ chết vì ung thư dạ dày thấp hơn khoảng 38%（HR 0.62, 95% CI 0.39 đến 0.99） +- Mức bằng chứng: A +- Nguồn: Li WQ và cộng sự (2019). Effects of Helicobacter pylori treatment and vitamin and garlic supplementation on gastric cancer incidence and mortality: follow-up of a randomized intervention trial. BMJ. +- Ghi chú: Sau khi diệt trùng vẫn có thể nhiễm lại, nên người trong nhà cùng xét, cùng điều trị. Dân số của thử nghiệm này là cư dân vùng có tỷ lệ ung thư dạ dày cao. Người sống ở vùng tỷ lệ thấp sẽ nhận lợi ích thực tế nhỏ hơn nhiều. + +### 24. Người hút thuốc nặng chụp CT ngực liều thấp mỗi năm một lần + +- Chi phí: Một lần 200 đến 400 tệ, làm xong trong 10 phút. Thêm nữa còn phải tính cả lần xét lại sau khi hoảng loạn, và thời gian lo lắng suốt thời gian đó. +- Nói thẳng: Người hút thuốc nặng đủ điều kiện chụp CT liều thấp mỗi năm một lần, xác suất chết vì ung thư phổi thấp hơn người chỉ chụp X-quang phổi 20.0%, xác suất tử vong toàn nguyên thấp hơn 6.7%. Cái giá phải trả là lần xét lại sau khi ra kết quả khả nghi, và thời gian lo lắng ấy. +- Lợi ích: Thử nghiệm ngẫu nhiên NLST của Mỹ thu 53.454 người, tất cả từ 55 đến 74 tuổi, lịch sử hút trên 30 bao năm, cai thuốc chưa quá 15 năm. Bao năm là đơn vị tính lượng thuốc, hút một bao một ngày trong một năm là một bao năm. Kết quả là thế này. Người chụp CT liều thấp so với người chỉ chụp X-quang phổi, tử vong vì ung thư phổi giảm tương đối 20.0%. Khoảng tin cậy 95% CI 6.8 đến 26.7. Tử vong toàn nguyên giảm tương đối 6.7%, khoảng tin cậy 95% CI 1.2 đến 13.6 +- Mức bằng chứng: A +- Nguồn: National Lung Screening Trial Research Team (2011). Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. New England Journal of Medicine. +- Ghi chú: Mục này chỉ áp dụng cho người hút thuốc nguy cơ cao. Trong cùng thử nghiệm đó, 24.2% người chụp CT liều thấp có kết quả khả nghi. Trong số này 96.4% chỉ là hoảng loạn thật, tức dương tính giả. Người không hút thuốc đi làm xét nghiệm này là tiền mua lo âu. + +### 25. Bị trầm cảm hoặc có ý nghĩ tự sát thì bấm 12356, trong nhà không tích trữ thuốc ngủ và thuốc trừ sâu + +- Chi phí: Không tốn tiền. Chỉ cần gọi một cuộc. Ngoài ra cất khoá thuốc, không để thuốc trừ sâu trong nhà. +- Nói thẳng: Càng dời xa những thứ có thể giết người, bằng chứng cho cách làm ấy càng mạnh. Kiểm soát thuốc giảm đau thì tự sát liên quan giảm 43%. Lắp chắn ở nơi hay có người nhảy xuống thì giảm 86%. Vì thế trong nhà không tích trữ thuốc ngủ và thuốc trừ sâu. Khi không thèm sống thì bấm 12356 trước, cả nước dùng chung, mỗi ngày mở ít nhất 18 giờ. +- Lợi ích: Một bài tổng quan hệ thống kê lại các nghiên cứu gần 10 năm: bằng chứng về việc hạn chế người ta tiếp cận phương tiện tử vong đang mạnh lên. Sau khi kiểm soát thuốc giảm đau, tự sát liên quan tới thuốc giảm đau giảm 43%. Sau khi lắp chắn ở những nơi hay có người nhảy xuống, tỷ lệ giảm 86%（79% đến 91%, trong ngoặc là khoảng tin cậy của số này）. Các chương trình nhận thức trong trường học làm tự sát không thành công giảm khoảng 55%（OR 0.45, 95% CI 0.24 đến 0.85）. Thuốc và trị liệu tâm lý chữa trầm cảm là một mắt xích quan trọng trong phòng ngừa tự sát. Từ ngày 1 tháng 5 năm 2025, bấm 12356 trên toàn quốc đều nối được đường dây nóng hỗ trợ tâm lý, mỗi ngày không dưới 18 giờ +- Mức bằng chứng: B +- Nguồn: Zalsman G và cộng sự (2016). Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry. ; Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委) (2024). 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）— Thông báo về việc dùng số điện thoại 12356 làm đường dây nóng hỗ trợ tâm lý thống nhất toàn quốc. +- Ghi chú: Lý do cho mức B: phần lớn bằng chứng về hạn chế phương tiện đến từ nghiên cứu sinh thái, cách làm là so số liệu trước và sau khi chính sách ban hành. Bản thân đường dây nóng không có bằng chứng nào cho thấy giảm tỷ lệ tử vong. Ý nghĩ tự sát thường chỉ kéo dài mấy phút đến mấy giờ, dời xa thứ có thể gây chết một bước cũng đã là can thiệp. Dữ liệu ở quy mô thời gian này, và kết cục dài hạn sau lần không thành công, xem mục 32 của phần này (ý nghĩ tự sát vừa nảy thì kể ngay một người bên cạnh). Sau khi được cứu, còn lại những gì, xem mục 33 của phần này. + +### 26. Chuẩn bị bình chữa cháy, chăn chữa cháy, mặt nạ thở để chạy thoát và túi sơ cứu, kiểm tra mỗi năm một lần + +- Chi phí: Một bình chữa cháy, một chăn chữa cháy, mỗi người một mặt nạ thở, thêm một túi sơ cứu, cả bộ vài trăm tệ. Mua một lần là xong, sau đó mỗi năm dành mười phút kiểm tra. +- Nói thẳng: Bình chữa cháy, chăn chữa cháy, mặt nạ thở cứu hộ, túi sơ cứu đều nằm trong danh sách 11 món ứng cứu gia đình của Bộ Quản lý Ứng cứu. Một bộ vài trăm tệ, mua một lần là xong. Trong các nghiên cứu có người đến nhà hướng dẫn an toàn, tỷ lệ xảy ra tổn thương thấp hơn khoảng một phần tư. +- Lợi ích: 《全国基础版家庭应急物资储备建议清单》 của Bộ Quản lý Ứng cứu gồm 11 món. Trong đó「灭火器和灭火毯」dùng cho「初起火灾的扑救」, bản danh sách còn ghi rõ「灭火毯可披覆在身上逃生」— chăn chữa cháy có thể khoác lên người để chạy thoát.「呼吸面罩」là「消防过滤式自救呼吸器，用于火灾逃生使用」— mặt nạ tự cứu lọc kiểu cứu hỏa, dùng khi chạy thoát khỏi đám cháy.「外用药品」là「止血粉、止血贴、纱布绷带等，用于处理伤口」— bột cầm máu, miếng dán cầm máu, băng gạc v.v., dùng xử lý vết thương. Cochrane gộp 98 nghiên cứu, 2.605.044 người lại với nhau. Can thiệp an toàn gia đình làm tỷ lệ xảy ra tổn thương xuống IRR 0.89. Khoảng tin cậy là 95% CI 0.78 đến 1.01. Trong đó, can thiệp có người đến nhà làm tỷ lệ xảy ra tổn thương thấp hơn khoảng 25%（IRR 0.75, 95% CI 0.62 đến 0.91） +- Mức bằng chứng: B +- Nguồn: Bộ Quản lý Ứng cứu (应急管理部) (2020). 全国基础版家庭应急物资储备建议清单. ; Kendrick D và cộng sự (2012). Home safety education and provision of safety equipment for injury prevention. Cochrane Database of Systematic Reviews. ; Sở Giám sát Thị trường huyện Vũ Thắng (武胜县市场监督管理局) (2025). 手提式灭火器、防火门（窗）、点型感温（烟）火灾探测器消费提示. +- Ghi chú:Tranh cãi. Trong cùng bản tổng hợp Cochrane đó, nhóm can thiệp không phát thiết bị làm tỷ lệ bị thương thấp hơn khoảng hai phần mười. Giá trị gốc IRR 0.78, khoảng tin cậy 95% CI 0.66 đến 0.92. Nhóm được phát thiết bị không làm tốt hơn. Cái chính làm giảm tổn thương là giáo dục và cải tạo tại nhà, nên lợi ích của thiết bị ghi「vừa」chứ không ghi「lớn」. Mua bình chữa cháy thì nhìn dấu CCC, đồng hồ của bình dạng áp lực phải dừng ở vùng xanh. Báo động khói và báo động CO xem mục 3 của phần này. Cách dùng gar xem mục 12 ở phần 13. Chọn từng món thế nào, để ở đâu, mấy năm đổi một lần, xem [docs/家庭应急装备清单.md](../docs/家庭应急装备清单.md) + +### 27. Trong nước tiểu có máu nhìn thấy bằng mắt, dù không đau, dù ngày mai đã sạch, vẫn phải khám một lần + +- Chi phí: Xét nghiệm nước tiểu vài chục tệ, siêu âm hệ tiết niệu một hai trăm tệ. Nếu bác sĩ thấy cần, thêm một nội soi bàng quang, khoảng một nghìn tệ. +- Nói thẳng: Ở người trên 60 tuổi, một lần có máu trong nước tiểu nhìn thấy bằng mắt, tỷ lệ cuối cùng phát hiện ung thư bàng quang khoảng 2.8%, từ 40 đến 59 tuổi khoảng 1.2%. Con số này không cao, nhưng vẫn đủ cao để bắt đi khám một lần. Dù ngày mai hết ra máu thì vẫn phải khám. +- Lợi ích: Anh Quốc làm một nghiên cứu dựa trên hồ sơ điện tử của phòng khám cộng đồng, tức chăm sóc ban đầu: 4.915 ca ung thư bàng quang với 21.718 ca đối chứng. Trong số người có máu nhìn thấy bằng mắt, tỷ lệ cuối cùng thật sự ra ung thư bàng quang là thế này. Trên 60 tuổi là 2.8%, khoảng tin cậy 95% CI 2.5 đến 3.1. Từ 40 đến 59 tuổi là 1.2%, khoảng tin cậy 0.6 đến 2.3. Tỷ lệ này có tên riêng là giá trị dự đoán dương tính. Máu trong nước tiểu chỉ thấy dưới kính hiển vi cũng có ý nghĩa. Tỷ số odds 20, khoảng tin cậy 95% CI 12 đến 33. Ở nhóm trên 60 tuổi, tỷ lệ này là 1.6%, khoảng tin cậy 1.2 đến 2.1 +- Mức bằng chứng: B +- Nguồn: Price SJ, Shephard EA, Stapley SA, Barraclough K, Hamilton WT (2014). Non-visible versus visible haematuria and bladder cancer risk: a study of electronic records in primary care. British Journal of General Practice, 64(626), e584-e589. +- Ghi chú:「Không đau」là điểm mấu chốt. Máu trong nước tiểu do sỏi và nhiễm trùng, phần lớn đi kèm đau quặn lưng bụng, hoặc tiểu nhiều, tiểu gấp, tiểu đau. Máu do khối u thường không có cảm giác gì, lại tự dừng vài ngày rồi tái diễn, nhiều người bỏ qua. Đang dùng thuốc chống đông không phải lý do để không đi khám. Máu trong nước tiểu sau vận động mạnh, hoặc ở phụ nữ trước và sau kỳ kinh, có thể xét lại một lần rồi hẳng quyết định. Lợi ích ghi「nhỏ」là vì chỉ có tỷ lệ phát hiện, không có cỡ tác động kiểu「khám sớm giảm được bao nhiêu tử vong」. + +### 28. Chức năng cương có vấn đề, đi khám tim mạch trước, đừng coi đó chỉ là chuyện ấy + +- Chi phí: Khám bệnh vài chục tệ. Đo huyết áp, rút máu xét đường huyết và mỡ máu, một hai trăm tệ. Đi một chuyến bệnh viện, chờ vài ngày có kết quả. +- Nói thẳng: Ở nam giới bị rối loạn cương dương, sau này mắc bệnh tim mạch nhiều hơn người khác khoảng bốn phần mười, nhồi máu cơ tim nhiều hơn khoảng sáu phần mười. Người càng trẻ thì chênh lệch càng lớn. Bệnh này thường xuất hiện trước cả đau ngực. Vì thế hãy đi khám huyết áp, đường huyết, mỡ máu trước, đừng tự mua thuốc dẹp đi. +- Lợi ích: Một phân tích tổng hợp gộp 14 nghiên cứu theo dõi, 92.757 nam giới, theo dõi trung bình 6.1 năm. So với người không bị rối loạn cương dương, người có vấn đề này sau đó có nguy cơ thế này. Sự kiện tim mạch toàn bộ cao hơn khoảng 44%（RR 1.44, 95% CI 1.27 đến 1.63, đây là khoảng tin cậy）. Nhồi máu cơ tim cao hơn khoảng 62%（RR 1.62, 1.34 đến 1.96）. Sự kiện mạch máu não cao hơn khoảng 39%（RR 1.39, 1.23 đến 1.57）. Tử vong toàn nguyên cao hơn khoảng 25%（RR 1.25, 1.12 đến 1.39）. Riêng tử vong tim mạch thì không ra khoảng cách（RR 1.19, 0.97 đến 1.46）. Người càng trẻ thì hệ số càng cao. Với nhóm vốn có nguy cơ tim mạch trung bình, hệ số còn cao hơn cả nhóm nguy cơ cao lẫn nhóm nguy cơ thấp. Nghiên cứu định nghĩa bằng bảng hỏi, tính ra hệ số là 1.61（1.38 đến 1.86）. Nghiên cứu chỉ hỏi một câu thì là 1.27（1.18 đến 1.37） +- Mức bằng chứng: A +- Nguồn: Vlachopoulos CV, Terentes-Printzios DG, Ioakeimidis NK, Aznaouridis KA, Stefanadis CI (2013). Prediction of cardiovascular events and all-cause mortality with erectile dysfunction: a systematic review and meta-analysis of cohort studies. Circulation: Cardiovascular Quality and Outcomes, 6(1), 99-109. ；DailyMed (thư viện dữ liệu dược phẩm của Thư viện Y khoa Quốc gia Hoa Kỳ, 美国国立医学图书馆). VIAGRA (sildenafil citrate) tablet, CONTRAINDICATIONS. +- Ghi chú: Cả 14 nghiên cứu đó đều theo dõi một nhóm người rồi xem kết quả, không phải thử nghiệm chia nhóm ngẫu nhiên. Vì thế「nhiều hơn bốn phần mười」là liên hệ, không phải quan hệ nhân quả. Thỉnh thoảng một hai lần thì không tính, phải kéo dài trên ba tháng mới xem là vấn đề. Những thứ cần khám là cùng một bó với mục 7 và 8 của phần này: đo huyết áp, xét đường huyết lúc nhịn, cộng mỡ máu, hút thuốc và vòng eo. Đừng mua「thuốc bổ sung」cường dương trên mạng. Thuốc đúng loại là thuốc kê đơn, tờ hướng dẫn liệt kê nhóm nitrat là chống chỉ định, người đang dùng nitroglycerin dạng tương tự thì không được dùng, dùng chung sẽ làm huyết áp tụt mạnh. Cải thiện bằng cách nào xem mục 29 của phần này (giảm cân, cai thuốc làm chức năng cương tốt lên). Lợi ích ghi「vừa」chứ không ghi「lớn」: 44% là chênh lệch nguy cơ giữa hai nhóm người, không phải「khám một lần giảm được bao nhiêu tử vong」. Cũng không ghi「nhỏ」: loại tín hiệu như mục 27 của phần này (máu trong nước tiểu nhìn thấy bằng mắt) chỉ có tỷ lệ phát hiện, còn ở đây những thứ phát hiện ra như huyết áp cao, tiểu đường, mỡ máu bất thường đều có sẵn cách điều trị. + +### 29. Giảm cân, cai thuốc, giữ huyết áp và đường huyết ổn định, chức năng cương cũng tốt lên theo + +- Chi phí: Không tốn tiền, còn có thể tiết kiệm. Chỗ khó là duy trì lâu dài: kiềm chế ăn, vận động, cai thuốc. +- Nói thẳng: Kiểm soát được các yếu tố nguy hiểm tim mạch thì bản thân chức năng cương cũng tốt lên. Dùng một bảng chấm điểm, trung bình tăng được 2 đến 3 điểm. Tổng điểm của bảng này là 25 điểm, từ 22 điểm trở lên là bình thường. +- Lợi ích: Một phân tích tổng hợp gộp 6 thử nghiệm đối chứng ngẫu nhiên, 740 nam giới. Can thiệp là đổi lối sống, hoặc dùng thuốc kiểm soát yếu tố nguy hiểm tim mạch, theo dõi ít nhất 6 tuần. Kết quả: điểm bảng chấm chức năng cương (IIEF-5) tăng trung bình 2.66 điểm（95% CI 1.86 đến 3.47, đây là khoảng tin cậy）. Bỏ 2 thử nghiệm dùng statin (143 người), 4 thử nghiệm còn lại chỉ dựa vào đổi lối sống (597 người) vẫn tăng 2.40 điểm（1.19 đến 3.61）. IIEF-5 là bảng hỏi 5 câu, tổng điểm từ 5 đến 25. Nguyên văn chia thành năm bậc theo điểm, từ 22 đến 25 điểm là không rối loạn, từ 5 đến 7 điểm là nặng +- Mức bằng chứng: A +- Nguồn: Gupta BP, Murad MH, Clifton MM, Prokop L, Nehra A, Kopecky SL (2011). The effect of lifestyle modification and cardiovascular risk factor reduction on erectile dysfunction: a systematic review and meta-analysis. Archives of Internal Medicine, 171(20), 1797-1803. ；Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM (1999). Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. International Journal of Impotence Research, 11(6), 319-326. +- Ghi chú: Sáu thử nghiệm này can thiệp khác nhau, có thử nghiệm giảm cân cộng vận động, có thử nghiệm dùng thuốc kiểm soát huyết áp và mỡ máu. Tổng cộng chỉ có 740 người. Nên hướng thì rõ, còn cụ thể tốt đến đâu thì tuỳ từng người. Lợi ích ghi「nhỏ」là vì điểm cuối chỉ là điểm bảng hỏi, không có con số kiểu「sau này bớt được bao nhiêu bệnh」. Cùng mấy động tác ấy ở nơi khác lại có điểm cuối cứng: cai thuốc xem mục 1 ở phần 2 (cai thuốc, càng sớm càng tốt), giảm cân xem mục 33 ở phần 2 (giữ BMI ở mức 20–25), huyết áp xem mục 7 của phần này (đo huyết áp, cao thì uống thuốc). Bản thân việc uống thuốc không phải vùng cấm, điều kiện là thuốc kê đơn do bác sĩ kê.「Thực phẩm chức năng」bán trên mạng hay trộn thêm thành phần loại này, liều không rõ, cách phân biệt xem mục 4 ở phần 28 (không mua thuốc giảm cân hứa hẹn nhanh). + +### 30. Dùng bao cao su suốt quá trình quan hệ tình dục, không dùng chung kim tiêm với người khác + +- Chi phí: Một chiếc một hai tệ. Chỗ khó không phải là tiền, mà là dùng mỗi lần, và đeo từ đầu. +- Nói thẳng: Dùng suốt quá trình mỗi lần, xác suất lây HIV giữa người một vợ một chồng giảm khoảng bảy phần mười trở lên.「Chỉ đeo phần lớn số lần」tác dụng gần như không khác gì không đeo. Giang mai, lậu cũng bị chặn luôn. +- Lợi ích: Một nghiên cứu gộp 25 nghiên cứu, 10.676 cặp đối tác khác giới tính. Mỗi cặp đều có một bên nhiễm HIV, một bên không, tên riêng là cặp huyết thanh không đồng nhất. So với người không bao giờ dùng, người dùng bao cao su mỗi lần có nguy cơ lây HIV thấp hơn khoảng 71%. Giá trị gốc RR 0.29, khoảng tin cậy 95% CI 0.20 đến 0.43. So với người dùng đứt quãng, thấp hơn khoảng 77%（RR 0.23, 95% CI 0.13 đến 0.40）. Kết luận của tác giả là bao cao su dù không hoàn hảo, nhưng「坚持使用可使 HIV 传播降低 70% 以上」— dùng đều đặn thì giảm hơn 70% việc lây HIV. Cùng lớp bảo vệ đó còn phủ giang mai, lậu, viêm kết mạc do Chlamydia và các bệnh lây qua đường tình dục khác +- Mức bằng chứng: A +- Nguồn: Giannou FK, Tsiara CG, Nikolopoulos GK, et al. (2016). Condom effectiveness in reducing heterosexual HIV transmission: a systematic review and meta-analysis of studies on HIV serodiscordant couples. Expert Review of Pharmacoeconomics & Outcomes Research, 16(4), 489-499. ；Fonner VA và cộng sự (2016). Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS. （phòng trước phơi nhiễm PrEP） +- Ghi chú: Chênh lệch giữa「dùng không đều」và「không dùng」rất nhỏ, phải dùng mỗi lần, suốt quá trình mới có tác dụng. Dùng chung kim tiêm là con đường lây hiệu quả khác, độc lập với quan hệ tình dục, tiêm ma tuý, xăm mình, xuyên lỗ tai đều tính vào. Nhóm nguy cơ cao còn dùng được thuốc phòng trước phơi nhiễm (PrEP), tức uống thuốc trước để chặn nhiễm. Gộp 18 nghiên cứu lại, mọi thử nghiệm có trên bảy phần mười người uống thuốc đúng giờ thì người dùng PrEP có nguy cơ nhiễm HIV thấp hơn nhóm giả dược khoảng bảy phần mười. Giá trị gốc là tỷ số nguy cơ 0.30, khoảng tin cậy 95% CI 0.21 đến 0.45. Ở những thử nghiệm có ít người uống thuốc đúng giờ thì không thấy tác dụng bảo vệ. PrEP là thuốc kê đơn, đi hỏi khoa bệnh truyền nhiễm hoặc trung tâm kiểm soát bệnh, đừng tự mua trên mạng. Sau khi đã phơi nhiễm thì xử lý thế nào, xem phần 13 + +### 31. Từng có hành vi nguy cơ cao thì xét nghiệm AIDS một lần, miễn phí ở Trung tâm kiểm soát bệnh và giữ kín kết quả + +- Chi phí: Không tốn tiền. Một chuyến tới Trung tâm kiểm soát bệnh hoặc cơ sở y tế được chỉ định, rút một ống máu. Chỗ khó là bước mở miệng đi xét nghiệm. +- Nói thẳng: Đi xét nghiệm AIDS ở Trung tâm kiểm soát bệnh không mất tiền. Theo《全国艾滋病检测工作管理办法》, nhân viên không được tiết lộ tên, địa chỉ và kết quả xét nghiệm của bạn. Kết quả dương tính không phải bản án tử hình. Người uống thuốc đều đặn đẩy lượng virus xuống dưới ngưỡng phát hiện sẽ không lây sang bạn tình. Nghiên cứu đó theo dõi hơn bảy trăm cặp đôi và hơn bảy mươi nghìn lần quan hệ không dùng bao cao, không có ca nào lây trong cặp. +- Lợi ích: 《艾滋病防治条例》quy định rõ: ở cấp huyện trở lên, cơ quan quản lý y tế địa phương sẽ chỉ định một số cơ sở y tế. Các cơ sở này「应当为自愿接受艾滋病咨询、检测的人员免费提供咨询和初筛检测」— phải tư vấn và xét nghiệm sàng lọc ban đầu miễn phí cho người tự nguyện muốn được tư vấn và xét nghiệm AIDS. Xét sớm thì có tác dụng thật. Thuốc kháng virus hiện nay đẩy được lượng virus xuống dưới ngưỡng phát hiện. Tổ chức Y tế Thế giới nói「接受抗病毒治疗且病毒载量检测不到的感染者不会把 HIV 传染给性伴侣」— người nhiễm HIV đang điều trị kháng virus và không còn phát hiện được lượng virus sẽ không lây HIV sang bạn tình. Nghiên cứu PARTNER theo dõi 782 cặp đôi cùng giới, mỗi cặp một người nhiễm một người không nhiễm, gọi tên khoa học là serum khác nhau. Tổng cộng 1593 cặp năm, tức cộng số năm theo dõi của từng cặp. Trong thời gian đó có 76088 lần quan hệ không dùng bao cao và 15 ca nhiễm mới. Sau khi so gen virus bằng phân tích phát sinh học, cả 15 ca đều không đến từ bạn tình. Lây trong cặp bằng không. 《全国艾滋病检测工作管理办法》còn quy định nhân viên phòng xét nghiệm AIDS không được tiết lộ tên, địa chỉ, kết quả xét nghiệm và các thông tin liên quan của người mắc AIDS hay người nhiễm HIV +- Mức bằng chứng: A +- Nguồn: Bộ Y tế (卫生部) (2006). Quy định quản lý việc xét nghiệm AIDS toàn quốc, 全国艾滋病检测工作管理办法. ；Ủy ban Nhà nước (国务院) (2006). Nghị định phòng ngừa và xử lý AIDS, 艾滋病防治条例（第三、二十三、三十八条）, gồm cấm phân biệt đối xử với người nhiễm, quyền được tư vấn và xét nghiệm miễn phí, và nghĩa vụ khai báo của người nhiễm. （Ủy ban Y tế và Sức khỏe thành phố Bắc Kinh, 北京市卫生健康委员会, đăng lại）；Rodger AJ, Cambiano V, Bruun T, et al. (2019). Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. The Lancet, 393(10189), 2428-2438. ；Trung tâm Kiểm soát và Phòng ngừa bệnh Quảng Đông (广东省疾病预防控制中心). Giáo trình phổ thông về thời gian cửa sổ xét nghiệm AIDS, 艾滋病检测窗口期科普. （con số về thời gian cửa sổ） +- Ghi chú: Thời gian cửa sổ quyết định xét nghiệm vào lúc nào mới có ý nghĩa. Cơ sở của Trung tâm kiểm soát bệnh Quảng Đông (广东省疾控中心) là: xét nghiệm axit nucleic khoảng 1 tuần là phát hiện được, xét nghiệm kháng nguyên kết hợp kháng thể thế hệ thứ tư khoảng 2 tuần, xét nghiệm kháng thể thế hệ thứ ba khoảng 3 tuần. Nên chờ ít nhất 2 tuần rồi hãy xét, cách 2 đến 4 tuần xét lại một lần. Sau khi lần hành vi nguy cơ cao nhất đã tròn 3 tháng, trên 99.99% có thể loại trừ nhiễm. Que thử tự cho dương tính không có nghĩa là chẩn đoán, phải tới Trung tâm kiểm soát bệnh hoặc cơ sở chính quy để kiểm chứng. Điều 3 của nghị định quy định không cá nhân và tổ chức nào được phân biệt đối xử với người nhiễm và gia đình họ. Quyền về hôn nhân, việc làm, khám chữa bệnh, nhập học được pháp luật bảo vệ. Điều 38 quy định thêm rằng người nhiễm có nghĩa vụ「将感染或者发病的事实及时告知与其有性关系者」— kịp thời báo cho người có quan hệ tình dục với mình biết sự thật mình đã nhiễm hoặc đã mắc bệnh. Sàng lọc ba bệnh trong thai kỳ và phòng lây truyền từ mẹ sang con xem ở phần 27 + +### 32. Vừa nảy ý tưởng tự sát thì nói với một người bên cạnh, giao mấy chục phút ấy ra + +- Chi phí: Không tốn tiền. Chỗ khó là lúc mở miệng, và thừa nhận rằng mấy tiếng đó mình không ở một mình được. +- Nói thẳng: Trong số người đã tự sát không thành công, gần một nửa nói rằng từ lúc nảy ý tưởng đến lúc ra tay không quá 10 phút. Nên thứ phải trấn an thường chỉ là mấy chục phút đó. Ba phần tư trong số họ có tiếp xúc với người bên cạnh trong khoảng thời gian ấy, chỉ là không mở miệng. Người phải vào bệnh viện vì tự làm hại mình, sau 1 năm khoảng 2% chết vì tự sát, sau trên 9 năm khoảng 7%. Đa số về sau không chết vì tự sát. +- Lợi ích: 82 người tự sát không thành công được phỏng vấn trong vòng 3 ngày sau khi nhập viện. 47.6% (39 người) nói rằng từ lần đầu nảy ý tưởng tự sát cho tới khi ra tay không quá 10 phút. Người có khoảng thời gian này kéo dài hơn thì điểm trên thang đo ý định tự sát rõ ràng cao hơn. Điểm trên thang đo hành vi bốc đồng không liên quan đến độ dài của khoảng thời gian. 76.8% (63 người) nói rằng trong khoảng thời gian đó họ có tiếp xúc nào đó với bạn tình, gia đình hoặc bạn bè. Kết luận của tác giả: khoảng thời gian để can thiệp thường rất ngắn, chuyên viên ở giai đoạn này có rất ít chỗ để xen vào, nên phải phổ biến dấu hiệu nhận biết và cách can thiệp cho người thường. Một bài phân tích tổng hợp khác tổng hợp 90 nghiên cứu theo dõi. Ở người đến khám vì tự làm hại mình, tỷ lệ tự làm hại lại mà không chết sau 1 năm, lấy con số trung vị là 16%. Sau 1 năm, tỷ lệ chết vì tự sát là 2%. Ở nhóm theo dõi trên 9 năm, khoảng 7% chết vì tự sát +- Mức bằng chứng: B +- Nguồn: Deisenhammer EA, Ing CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry, 70(1), 19-24. ；Owens D, Horrocks J, House A (2002). Fatal and non-fatal repetition of self-harm: systematic review. British Journal of Psychiatry, 181, 193-199. +- Ghi chú: Lý do xếp B: nghiên cứu của Deisenhammer chỉ có 82 người, lấy ở một bệnh viện, dựa vào trí nhớ kể lại sau sự việc của chính người trong cuộc. Bài của Owens đưa ra con số giữa của 90 nghiên cứu làm rất khác nhau, chứ không gộp chúng thành một số. Chính bài đó còn viết rõ nguy cơ tự sát của người tự làm hại mình vẫn cao hơn người thường hàng trăm lần. Vì vậy「绝大多数人后来没有死于自杀」nói về cả nhóm này, không phải bảo đảm cho một cá nhân, bệnh vẫn phải khám. Đưa tư liệu gây chết ra xa và 12356 xem mục 25 của phần này. Khi tâm trạng tụt thì làm vài việc sau, xem mục 19 ở phần 3. Di chứng sau khi được cứu xem mục 33 của phần này. Khi người bên cạnh lộ ra ý tưởng ấy mà bạn làm được gì, xem mục 15 ở phần 8 + +### 33. Đừng coi「cứu được」là chốt chặn cuối cùng: uống thuốc trừ sâu hay hít khí than, cấp cứu giữ được mạng nhưng không giữ được phổi và não + +- Chi phí: Không tốn tiền. Chỉ cần biết trước chuyện này. +- Nói thẳng: Thuốc diệt cỏ paraquat không có thuốc giải độc, hơn bốn phần mười người uống phải không cứu được. Nó làm phổi viêm rồi dần cứng lại, thận và gan cũng tổn thương theo. Người ngộ độc khí than được cứu, sau sáu tuần gần một nửa có đầu óc kém đi. Nhóm đã dùng oxy áp lập còn một phần tư. +- Lợi ích: Một bài phân tích tổng hợp về ngộ độc paraquat: tỷ lệ tử vong rất cao, vì bản thân nó độc tính mạnh mà lại không có cách điều trị hiệu quả. Paraquat bị chuyển vào mô phổi theo chiều nghịch chênh lệch nồng độ một cách chủ động, gọi tên khoa học là nhập bám chủ động ngược gradient nồng độ, kết quả là viêm phổi và xơ hoá phổi, đồng thời gây tổn thương thận và tổn thương gan. Bản tổng hợp viết rõ: không nên rửa dạ dày. Thẩm tách máu và thẩm tách máu qua than nhiều khó có thể đổi diễn tiến lâm sàng. Bằng chứng về hiệu quả của corticosteroid và thuốc ức chế miễn dịch rất yếu. Bản tổng hợp còn viết:「所有中心的病死率都很高，尽管治疗差异很大」— tỷ lệ tử vong ở mọi trung tâm đều cao, dù cách điều trị rất khác nhau. Từ 2013 đến 2024, 10 bệnh viện hạng A ở Bangladesh thu nhận 1420 ca, trong đó 257 ca phân tích được: tỷ lệ tử vong 43.2%. Biểu hiện hay gặp là phù chi dưới do suy thận cấp 71.2%, nôn 65.8%, đau bụng 58.0%. Thử nghiệm với ngộ độc carbon monoxide cấp tính dùng phân nhóm mù kết hợp ngẫu nhiên, mỗi nhóm 76 người. Bác sĩ và bệnh nhân đều không biết mình thuộc nhóm nào. Sau sáu tuần, nhóm oxy áp lập còn di chứng nhận thức 19/76 (25.0%). Nhóm oxy áp thường là 35/76 (46.1%), P=0.007. Sau khi hiệu chỉnh các yếu tố khác, OR 0.45 (95% CI 0.22 đến 0.92, trong ngoặc là khoảng tin cậy của số này). Một nghiên cứu khác theo dõi liên tục 73 ca: 30% có di chứng nhận thức, 12% xuất hiện tăng tín hiệu ở chất trắng não. Từ ngày 1 tháng 7 năm 2014, Trung Quốc thu hồi đăng ký và giấy phép sản xuất paraquat dạng dung dịch, rồi ngừng sản xuất. Từ ngày 1 tháng 7 năm 2016, dung dịch đó ngừng bán và ngừng dùng trong nước. Từ ngày 26 tháng 9 năm 2020, cấm bán và dùng dạng tan trong nước. Văn bản nói rõ,「致使服用百草枯自杀事件时有发生」là lý do trực tiếp để siết quản lý +- Mức bằng chứng: B +- Nguồn: Gawarammana IB, Buckley NA (2011). Medical management of paraquat ingestion. British Journal of Clinical Pharmacology, 72(5), 745-757. ；Chowdhury FR 等 (2026). Clinical and Laboratory Profile of Paraquat Poisoning: A Toxicological Crisis in Bangladesh. American Journal of Tropical Medicine and Hygiene, 115, 176-182. ；Weaver LK, Hopkins RO, Chan KJ, et al. (2002). Hyperbaric oxygen for acute carbon monoxide poisoning. New England Journal of Medicine, 347(14), 1057-1067. ；Parkinson RB, Hopkins RO, Cleavinger HB, et al. (2002). White matter hyperintensities and neuropsychological outcome following carbon monoxide poisoning. Neurology, 58(10), 1525-1532. ；Văn phòng Bộ Nông nghiệp và Nông thôn (农业农村部办公厅) (2018). Thông báo siết chặt quản lý paraquat, 关于切实加强百草枯监督管理的通知（农办农〔2018〕17 号）. +- Ghi chú: Lý do xếp B: đối tượng của các nghiên cứu này là người bị ngộ độc, không phải một nhóm người tự sát không thành công thuần tuý. Thử nghiệm carbon monoxide còn lẫn cả người bị ngộ độc do tai nạn. Vì vậy tỷ lệ di chứng ở trên không dùng thẳng làm「xác suất hậu quả với người tự sát không thành công」được. Điều chắc chắn là không cách nào mà không để lại di chứng. Xử lý tại chỗ người ngộ độc carbon monoxide xem mục 19 ở phần 13. Uống nhầm thuốc trừ sâu hay thuốc, đừng nôn, mang theo vỏ thuốc đi khám, xem mục 20 ở phần 13. Trong nhà không tích trữ thuốc trừ sâu và thuốc ngủ xem mục 25 của phần này. Sách này không so tỷ lệ tử vong giữa các cách, cũng không viết cách làm cụ thể nào + +### 34. Đừng cược vào chuyện「nằm vài ngày là khỏe」khi ngã từ trên cao: phần lớn người vào ICU chấn thương vẫn sống, cái giá thì tính theo năm + +- Chi phí: Không tốn tiền. Chỉ cần nhìn rõ cái sổ này. +- Nói thẳng: Người rơi từ trên cao mà vào ICU chấn thương chỉ có 5.5% tử vong, nhưng chấn thương nặng và điều trị kéo dài. Gần một nửa bị tổn thương cùng lúc ở ba chỗ trở lên, hơn một phần ba phải đặt máy thở, người ở ICU lâu nhất 132 ngày. Một nửa rơi từ độ cao không quá 3 mét. Người chấn thương nặng, sau này quay lại làm việc được dưới sáu phần mười. +- Lợi ích: Một ICU chấn thương, tức khoa hồi sức tích cực, ở miền nam Tân Cương đã thống kê. Từ tháng 4 năm 2020 đến tháng 7 năm 2024, nơi này thu nhận 289 ca ngã từ trên cao. Nhóm này trung bình 40.3 tuổi. Chiều cao rơi trung vị 3 mét, 26.6% rơi từ trên 5 mét, 11.4% rơi từ trên 10 mét. Điểm chấn thương ISS trung vị 24, điểm này càng cao thì chấn thương càng nặng. 63.3% người có ISS lớn hơn 20. 47.1% bị tổn thương ở trên 3 chỗ, 18.3% bị tổn thương ở trên 4 chỗ. 36% cần thở máy, tức thông gí cơ học, thời gian thở máy trung vị 103.5 giờ. Nằm viện ICU trung vị 8 ngày, ngắn nhất 1 ngày, dài nhất 132 ngày. Tỷ lệ tử vong trong viện hoặc trong 30 ngày là 5.5% (16/289). Một trung tâm chấn thương ở Đức thu nhận 84 ca đa thương có ISS ≥ 25, đa thương nghĩa là tổn thương nhiều chỗ cùng lúc. Tỷ lệ quay lại làm việc trong thời gian theo dõi là 58%. Tuổi, thời gian nằm ICU và tự đánh giá sức khoẻ là các yếu tố tiên đoán. Năm 2024, chi phí trung bình mỗi lần nằm viện trên toàn quốc, tức số tiền trung bình mỗi lần nhập viện, là 9870.0 tệ, giảm 4.3% so với năm trước ở mức giá của năm đó. Cùng năm đó, tổng chi phí y tế toàn quốc là 9.089.550.000.000 tệ, trong đó chi tiêu y tế cá nhân là 2.500.750.000.000 tệ, chiếm 27.5% +- Mức bằng chứng: B +- Nguồn: Chen Y, Li W, Wang X, Zhong Q, Abudurexiti A, Qiu Q, Li J, Luo J (2025). Clinical features and prognostic predictors for patients admitted to trauma intensive care unit due to fall from height in South Xinjiang. International Journal of Emergency Medicine, 18, 142. ；Simmel S, Wurm S, Drisch S, Woltmann A, Coenen M (2020). Prädiktion der Rückkehr zur Arbeit nach Polytrauma bei Patienten mit einem ISS von mindestens 25. Die Rehabilitation, 59(2), 95-103. ；Ủy ban Quốc gia Y tế và Sức khỏe (国家卫生健康委) (2025). Giải thích báo cáo thống kê về phát triển y tế và sức khoẻ năm 2024, 2024 年我国卫生健康事业发展统计公报解读. +- Ghi chú: Lý do xếp B: nhóm ngã rơi được thống kê từ hồ sơ cũ của một bệnh viện, không tách tai nạn với tự làm hại. Nhóm quay lại làm việc chỉ có 84 người, lại ở Đức. Hai nhóm đều không dùng làm「một người nhảy lầu sẽ thế nào」được. Chi phí trung bình mỗi lần nhập viện là trung bình của mọi người nhập viện cả nước, không phải chi phí chấn thương do rơi, dẫn vào đây chỉ để có một mức dưới có thể đối chiếu. Bản gốc chỉ cho phân bố chiều cao của nhóm này, nên ở đây không viết quan hệ giữa chiều cao và hậu quả. Loét tì đè cần theo dõi nhất trong thời gian nằm liệt dài xem mục 8 ở phần 17. Huyết khối tĩnh mạch sâu và thuyên tắc phổi xem mục 11 ở phần 13 (một chân sưng phù bất thường). Lúc ý tưởng vừa nảy ra thì làm gì, xem mục 32 của phần này. Hậu quả của ngộ độc xem mục 33 của phần này (đừng coi「cứu được」là chốt chặn). Dưới đây là trải nghiệm của chính tác giả: rơi từ tầng bốn, trước đó nghĩ「mười mấy ngày là khỏe」, thực tế nằm viện một năm, phẫu thuật khắp bốn nơi, lắp thép, nửa năm không tự lật người ngồi dậy được, nhà phải vay tiền + +### 35. Đừng đổi tiền lấy chuyện「thiếu một quả thận cũng không sao」: quả còn lại phải làm việc thay hai quả, người bán thận sau đó có 86% nói sức khoẻ xấu đi + +- Chi phí: Không tốn tiền. Chỉ cần nhìn rõ cái sổ này. +- Nói thẳng: Người hiến thận qua sàng lọc kỹ trước và được tái khám sau khi hiến, sau 15 năm có 31 người trên 10.000 người bị suy thận mãn tính. Người cùng sức khoẻ mà không hiến là 4 người trên 10.000. Người bán thận không có sàng lọc cũng không có tái khám, sau 6 năm 86% nói sức khoẻ xấu đi, ba phần tư còn nợ tiền. +- Lợi ích: Mỹ so 96.217 người hiến thận còn sống với 20.024 người khoẻ mạnh, điều kiện tương đương mà không hiến. Nhóm sau đến từ Khảo sát Dinh dưỡng và Sức khoẻ Quốc gia Hoa Kỳ (NHANES III). Sau đó xem nguy cơ ước tính đi tới suy thận mãn tính trong 15 năm sau khi hiến. Ở nhóm hiến là 30.8 trên 10.000 người, khoảng tin cậy 95% CI 24.3–38.5. Ở nhóm khoẻ mạnh không hiến là 3.9 trên 10.000 người, khoảng tin cậy 95% CI 0.8–8.9, P<0.001. Suy thận mãn tính gọi tên khoa học là bệnh thận giai đoạn cuối ESRD, đến bước này phải duy trì bằng lọc máu hoặc ghép thận. Nguy cơ ước tính đi tới ESRD cả đời: nhóm hiến 90 trên 10.000 người, dân chúng chung chưa qua sàng lọc 326 trên 10.000 người, nhóm khoẻ mạnh không hiến 14 trên 10.000 người. Na Uy cũng làm một lần so sánh. Một bên là 1901 người hiến thận, theo dõi trung vị 15.1 năm. Bên kia là 32.621 người đối chứng đủ điều kiện hiến, theo dõi trung vị 24.9 năm. Tử vong toàn nguyên ở nhóm hiến thận cao hơn khoảng ba phần mười (HR 1.30, 95% CI 1.11–1.52). Tử vong tim mạch cao hơn khoảng bốn phần mười (HR 1.40, 1.03–1.91). Đi tới ESRD cao hơn khoảng mười lần (HR 11.38, 4.37–29.6). Tỷ lệ mắc ESRD chung của nhóm hiến này là 302 trên 1.000.000 người. Mỹ còn có 3698 người hiến thận, trong đó 11 người mắc ESRD, tương đương 180 ca trên 1.000.000 người quan sát trong một năm, cùng kỳ dân chúng chung là 268 ca. Trong nhóm đó có 255 người quay lại tái khám, trung bình 12.2±9.2 năm sau khi hiến. 85.5% có chức năng lọc thận GFR ≥ 60 ml/min/1.73 m². Con số 60 là mức dưới bình thường trong lâm sàng. 32.1% bị cao huyết áp. 12.7% bị đái đạm, tức nước tiểu rò ra protein. Ấn Độ khảo sát 305 người bán thận, trung bình được hỏi 6 năm sau khi bán: 96% bán để trả nợ, trung bình nhận 1070 đô la Mỹ. Thu nhập trung bình của gia đình sau khi cắt thận giảm một phần ba (P<0.001), số người sống dưới đường nghèo tăng lên, ba phần tư vẫn còn nợ lúc khảo sát. 86% báo cáo sức khoẻ xấu đi so với trước khi cắt thận, 79% không khuyên người khác bán thận +- Mức bằng chứng: A +- Nguồn: Muzaale AD, Massie AB, Wang MC, Montgomery RA, McBride MA, Wainright JL, Segev DL (2014). Risk of end-stage renal disease following live kidney donation. JAMA, 311(6), 579-586. ；Mjøen G, Hallan S, Hartmann A, et al. (2014). Long-term risks for kidney donors. Kidney International, 86(1), 162-167. ；Ibrahim HN, Foley R, Tan L, et al. (2009). Long-term consequences of kidney donation. New England Journal of Medicine, 360(5), 459-469. ；Goyal M, Mehta RL, Schneiderman LJ, Sehgal AR (2002). Economic and health consequences of selling a kidney in India. JAMA, 288(13), 1589-1593. +- Ghi chú:Tranh cãi. Nghiên cứu Ibrahim lấy người bình thường làm nhóm so sánh, kết luận là khả năng sống sót và nguy cơ đi tới suy thận mãn tính (ESRD) của người hiến thận tương đương người thường. Muzaale và Mjøen đổi nhóm so sánh sang người「cùng khoẻ, cũng đủ chuẩn hiến thận」thì nguy cơ lộ rõ hơn hẳn. Chỗ khác biệt là lấy ai để so, không nằm ở người hiến. Cả ba nghiên cứu đều đồng ý nguy cơ tuyệt đối không lớn, tính là 15 năm 30,8/10.000, tức 30,8 người trong 10.000 người. Nên trọng tâm không phải「hiến thận rất nguy hiểm」, mà là hai việc khác. Quả thận còn lại đúng là đang trả giá, tỷ lệ cao huyết áp và đái đạm nằm đó. Toàn bộ số liệu này đến từ nhóm được sàng lọc trước mổ, có người tái khám sau mổ, có thể quay lại bệnh viện khi có chuyện. Bán thận trên chợ đen không có một trong ba điều đó, mẫu khảo sát ở Ấn Độ cho ra một kết cục khác. Nghiên cứu gốc chỉ nêu tự đánh giá sức khoẻ xấu đi, đái đạm, cao huyết áp và tỷ lệ mắc suy thận mãn tính, không có con số「bao nhiêu người nằm liệt」, ở đây không tự bổ sung. Giới hạn người thân, mức phạt tiền và trách nhiệm hình sự khi hiến hợp pháp xem mục 22 ở phần 9 (không bán tạng của chính mình). Người đã phải lọc máu, việc thanh toán thẳng qua tỉnh và dùng thuốc dài hạn xem mục 1 và 2 ở phần 16 (uống thuốc đủ theo đơn, đăng ký khám bệnh đặc biệt ngoài tỉnh) + +### 36. Không nhặt, không tháo, không bán phần kim loại không rõ lai lịch, thấy ký hiệu cỏ ba lá thì tránh xa + +- Chi phí: Không tốn tiền. Thấy trụ kim loại lạ, thùng chì, linh kiện tháo ra từ thiết bị thì đi vòng, không nhặt, không tháo, không cho vào túi. +- Nói thẳng: Bức xạ không có mùi, không có màu, không có nhiệt độ. Hầu hết nguồn phóng xạ chỉ là một đoạn kim loại bình thường, nên「trông có vẻ không sao」không phải cách để đánh giá. Người thường va phải có xác suất cực thấp, mỗi năm trên 10.000 nguồn phóng xạ sự cố dưới 1 vụ. Nhưng đã va phải thì không hoàn tác được. Năm 1987 ở Bra-xin có người coi máy xạ trị của một phòng khám bỏ hoang là sắt vụn và tháo ra, cuối cùng 4 người chết. +- Lợi ích: Đến năm 2025, cả nước có khoảng 178.000 nguồn phóng xạ đang dùng và 355.000 máy hoặc bộ thiết bị tia X. Tỷ lệ tai nạn phóng xạ hằng năm ổn định dưới 1 vụ trên 10.000 nguồn. Phân cấp như sau. Nguồn loại I và loại II bị thất lạc, bị trộm hoặc mất kiểm soát thì cấu thành tai nạn phóng xạ nghiêm trọng. Loại III là tai nạn phóng xạ tương đối lớn. Loại IV và loại V bị thất lạc, bị trộm, hoặc khiến người ta bị chiếu xạ vượt quá giới hạn liều năm, thì là tai nạn phóng xạ thông thường. Ca chính thức: tháng 6 năm 2020, nhà máy luyện sắt của Tập đoàn Thép Ninh Hạ (宁夏钢铁集团) làm kiểm tra tia gamma trong công trình cải tạo đường ống lò cao. Một nguồn iridium-192 loại II làm ba công nhân bị chiếu xạ. Trương Tam (陈某某) bị tổn thương da cấp tính do phóng xạ cấp III ở tay trái. Lưu Tam (刘某某) bị cấp IV ở ngón cái và ngón trỏ tay trái, cấp III ở ngón giữa, áp út và ngón út. Phan Tam (樊某某) bị cấp I-II ở tay trái. Liều chiếu xạ của cả ba đều vượt xa giới hạn quy định trong tiêu chuẩn quốc gia, hai đơn vị chịu trách nhiệm bị thu hồi giấy phép an toàn phóng xạ và bị phạt tiền 70.000 tệ. Ca quốc tế: năm 1987 ở Goiania, Bra-xin, máy xạ trị từ xa của một phòng khám bỏ hoang bị tháo ra, lấy ra nguồn cesi-137 công suất 50.8 TBq bên trong, và làm vỡ túi bọc kín nguồn. Khoảng 112.000 người được theo dõi. Trong đó 249 người bị nhiễm bằng trong hoặc ngoài cơ thể. Trong 249 người này, 129 người bị nhiễm mức vừa đến nặng, 50 người cần theo dõi y tế sát sao. Cuối cùng 4 người chết (toàn quốc / quốc tế) +- Mức bằng chứng: A +- Nguồn: Bộ Môi trường và Tự nhiên (生态环境部) (2025). Báo cáo tình hình môi trường Trung Quốc năm 2024, 2024 中国生态环境状况公报. ；Cục An toàn hạt nhân quốc gia (国家核安全局) (2026). Kết quả giám sát, kiểm tra hạt nhân và bức xạ an toàn năm 2025, 2025 年我国核与辐射安全监管工作成效. ；Bộ Môi trường và Tự nhiên (生态环境部) (2021). Thông báo xử lý một vụ tai nạn phóng xạ do làm kiểm tra tia gamma trái phép, 关于一起 γ 射线探伤作业违法违规造成辐射事故案件处理情况的通报. ；Ủy ban Nhà nước (国务院) (2019). Nghị định an toàn và phòng hộ với đồng vị phóng xạ và thiết bị tia X, 放射性同位素与射线装置安全和防护条例（2019 年修订，第三十四、四十二、六十八条）, nội dung về cấp tai nạn phóng xạ, phát hành giấy phép và nghĩa vụ báo cáo. ；Bộ Bảo vệ môi trường (环境保护部) (2009). Thông báo lập chế độ phân cấp, xử lý và báo cáo tai nạn phóng xạ, 关于建立放射性同位素与射线装置辐射事故分级处理和报告制度的通知（辐射事故分级标准）. ；International Atomic Energy Agency (1988). The Radiological Accident in Goiânia. IAEA, Vienna. +- Ghi chú: Nếu thật sự gặp: tránh xa, nhớ vị trí, đừng di chuyển, gọi 110. Nghị định quy định phải báo cho cơ quan quản lý môi trường, cơ quan công an và cơ quan quản lý y tế. Khoảng cách và thời gian là cách bảo vệ duy nhất có tác dụng, để trong túi hay bọc túi nhựa cũng vô ích. Máy X quang, CT là loại thiết bị tạo tia bằng cách cấp điện, ngắt điện thì không còn phóng xạ lưu lại. Chỗ nguy hiểm thật sự là nguồn phóng xạ kín bên trong thiết bị. Máy xạ trị ở bệnh viện, máy soi tia gamma gắn trên ống dẫn và nồi hơi công nghiệp, máy đo giếng, máy đo mực nước trong bể đều chứa nguồn. Tháo vỏ ngoài là lộ nguồn ra. Mà vẻ ngoài của chúng không khác gì phần kim loại thường. Vì vậy đừng trông chờ nhận ra nguồn phóng xạ, linh kiện thiết bị không rõ lai lịch thì không tháo, không bán, không mang về nhà. Người làm nghề thu gom phế liệu, tháo dỡ, bán đồ ngũ kim và công nhân công trường có nguy cơ cao hơn người thường nhiều. Thấy khu vực cảnh giới có vẽ ký hiệu cỏ ba lá ở công trường thì đi vòng. Rò rỉ ở nhà máy điện hạt nhân là loại tai nạn khác, người thường gặp ít hơn nữa. Người hưởng lợi chủ yếu là bạn và người nhà đang ở cùng hiện trường. + +### 37. Cờ bạc tới mức vay tiền còn cờ, tâm trạng lúc nào cũng tụt thì gọi 12356, rồi giao thẻ ngân hàng và mật khẩu thanh toán cho người nhà giữ + +- Chi phí: Không tốn tiền. Một cuộc gọi. Giao thẻ ngân hàng, mật khẩu thanh toán và ứng dụng vay tiền cho người nhà quản lý. Chỗ khó là phải thừa nhận trước rằng mình không kiểm soát nổi. +- Nói thẳng: Ở những người đã đi khám vì vấn đề cờ bạc, tỷ lệ chết vì tự sát gấp 15 lần người cùng tuổi, tỷ lệ tử vong toàn nguyên gấp 1.8 lần. Dấu hiệu đoán trước được ai sẽ tự sát là trầm cảm. Nên cờ bạc tới mức vay tiền còn cờ, tâm trạng lúc nào cũng tụt thì hãy gọi 12356. +- Lợi ích: Một nghiên cứu dựa trên sổ đăng ký toàn quốc ở Thụy Điển, tìm ra 2099 người được chẩn đoán tại chuyên khoa là rối loạn cờ bạc trong giai đoạn 2005 đến 2016. Rối loạn cờ bạc là không kiểm soát nổi việc cờ, cờ tới mức ảnh hưởng cuộc sống. So với người Thụy Điển cùng tuổi, ở nhóm 20 đến 74 tuổi này tỷ lệ tử vong toàn nguyên gấp 1.8 lần, tỷ lệ tử vong vì tự sát gấp 15 lần, hai con số này là tỷ số tử vong chuẩn hoá SMR. Trong thời gian theo dõi có 67 người chết, trong đó 21 người chết vì tự sát. Yếu tố dự báo được cái chết do tự sát là trầm cảm. Năm 2023 tác giả đã đăng bản đính chính: nam giới là 12.1 lần, nữ giới là 16.1 lần. Từ ngày 1 tháng 5 năm 2025, trên toàn quốc Trung Quốc gọi 12356 là nối được đường dây hỗ trợ tâm lý, mỗi ngày mở không dưới 18 giờ (Thụy Điển / toàn quốc) +- Mức bằng chứng: B +- Nguồn: Karlsson A, Håkansson A. (2018). Gambling disorder, increased mortality, suicidality, and associated comorbidity: A longitudinal nationwide register study. Journal of Behavioral Addictions, 7(4), 1091–1099. ；Karlsson A, Håkansson A. (2023). Corrigendum to: Gambling disorder, increased mortality, suicidality, and associated comorbidity. Journal of Behavioral Addictions, 12(1), 302. ；Ủy ban Quốc gia Y tế và Sức khỏe (国家卫生健康委) (2024). Thông báo dùng số điện thoại 12356 làm đường dây hỗ trợ tâm lý thống nhất toàn quốc, 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. +- Ghi chú: Lý do xếp B: chỉ có một nghiên cứu, và chỉ tính người đã tới chuyên khoa khám. Người trong nhóm này bệnh nặng hơn, nên con số 15 lần không áp được cho người chơi bài nói chung. Giao tiền cho người nhà quản lý là cách làm theo kinh nghiệm, không có nghiên cứu trực tiếp, tính riêng là mức C. Vạch đỏ pháp lý của cờ bạc xem mục 13 ở phần 9 (không thu phần trăm đầu bàn, không chơi cờ bạc trên mạng). Người nhà nợ tiền cờ, có nên trả hộ không, xem mục 44 ở phần 8 (nợ cờ không phải nợ chung của vợ chồng). Sau khi đã nảy ý tưởng tự sát thì làm gì, xem mục 32 của phần này (nói trước cho một người bên cạnh). Người hưởng lợi chủ yếu là bạn, khi người nhà thấy dấu hiệu thì cũng có thể kéo họ làm theo. + +### 38. Nam giới có quan hệ tình dục với nam và có hành vi nguy cơ cao: đi bệnh viện AIDS được chỉ định hỏi thuốc phòng trước phơi nhiễm, uống đúng giờ thì giảm hơn bảy phần mười số ca nhiễm + +- Chi phí: Đây là thuốc kê đơn, hầu hết địa phương người bệnh tự trả, giá từ vài trăm tệ tới vài nghìn tệ. Trước khi kê thuốc phải xét HIV, chức năng thận và viêm gan B. Sau đó cứ 3 tháng tái khám một lần. Chỗ khó là uống đúng giờ mỗi ngày. +- Nói thẳng: Người chưa nhiễm uống thuốc kháng virus trước, tên gọi là phòng trước phơi nhiễm. Người uống đúng giờ thì nguy cơ nhiễm giảm hơn bảy phần mười trở lên. Người uống lúc có lúc không thì gần như không có tác dụng. Đây là thuốc kê đơn, phải xét sức khoẻ ở bệnh viện rồi mới kê, sau đó tái khám định kỳ. +- Lợi ích: Thuốc này là viên nén kết hợp của emtricitabine và tenofovir. Thử nghiệm iPrEx cho 2499 nam giới có quan hệ tình dục với nam và phụ nữ chuyển giới uống thuốc mỗi ngày hoặc uống giả dược. Nhóm uống thuốc nhiễm ít hơn 44%（95% CI 15–63）. Ở nhóm xét máu thấy có thuốc, tác dụng bảo vệ mạnh hơn nhiều. Thử nghiệm PROUD ở Anh có 544 nam giới có quan hệ tình dục với nam. Nhóm uống ngay lập tức nhiễm ít hơn nhóm chờ một năm 86%（90% CI 64–96, 90% khoảng tin cậy）. Thử nghiệm IPERGAY ở Pháp cho uống thuốc theo nhu cầu trước và sau quan hệ tình dục, số ca nhiễm giảm 86%（95% CI 40–98）. Một bài phân tích tổng hợp mà Tổ chức Y tế Thế giới dùng gộp các thử nghiệm lại. Tổng cộng RR 0.49（0.33–0.73, thấp hơn khoảng 51%）. Ở các thử nghiệm có trên 70% người uống thuốc đúng giờ, RR 0.30（0.21–0.45, thấp hơn khoảng 70%）. Ở nhóm dưới 40% người uống đúng giờ, RR 0.95, gần như không có tác dụng. Cục Quản lý Dược phẩm quốc gia (国家药监局) tháng 8 năm 2020 phê duyệt thuốc này để phòng trước phơi nhiễm. Người lớn và thanh thiếu niên cân nặng ít nhất 35 kg đều dùng được (nhiều nước / Trung Quốc) +- Mức bằng chứng: A +- Nguồn: Grant RM, Lama JR, Anderson PL, et al. (2010). Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. New England Journal of Medicine, 363(27), 2587–2599. ；McCormack S, Dunn DT, Desai M, et al. (2016). Pre-exposure prophylaxis to prevent the acquisition of HIV-1 infection (PROUD): effectiveness results from the pilot phase of a pragmatic open-label randomised trial. The Lancet, 387(10013), 53–60. ；Molina JM, Capitant C, Spire B, et al. (2015). On-demand preexposure prophylaxis in men at high risk for HIV-1 infection. New England Journal of Medicine, 373(23), 2237–2246. ；Fonner VA, Dalglish SL, Kennedy CE, et al. (2016). Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS, 30(12), 1973–1983. ；Xu JJ, Huang XJ, Liu XC, et al. (2020). Consensus statement on human immunodeficiency virus pre-exposure prophylaxis in China. Chinese Medical Journal, 133(23), 2840–2846. （bản tiếng Trung, 《中国 HIV 暴露前预防用药专家共识》）；World Health Organization (2019). What's the 2+1+1? Event-driven oral pre-exposure prophylaxis to prevent HIV for men who have sex with men (WHO/CDS/HIV/19.8). ；Ủy ban Nhà nước (国务院办公厅) (2024). Kế hoạch kiểm soát và phòng chống AIDS Trung Quốc (2024–2030), 中国遏制与防治艾滋病规划（2024—2030 年）. +- Ghi chú: Người trong các thử nghiệm chủ yếu là nam giới có quan hệ tình dục với nam, nên nhóm này cũng là nhóm nên đi hỏi nhất. Người khác có hành vi nguy cơ cao cũng có thể hỏi bác sĩ xem có hợp với mình không. Trước khi kê thuốc bắt buộc phải xác nhận là chưa nhiễm, người đã nhiễm mà uống thì virus kháng thuốc, kết quả que thử tự không làm căn cứ được. Người có viêm gan B mà ngừng thuốc có thể làm viêm gan tái phát, nên phải xét trước khi kê. Tổ chức Y tế Thế giới chỉ khuyến nghị dùng theo nhu cầu cho nam giới có quan hệ tình dục với nam, hợp với người trung bình dưới 2 lần mỗi tuần. Cách uống là trước quan hệ tình dục từ 2 đến 24 giờ uống 2 viên, sau đó ở 24 giờ và 48 giờ mỗi lần uống 1 viên. Phụ nữ và người có viêm gan B chỉ có thể uống mỗi ngày. Ở các thử nghiệm dùng theo nhu cầu, nhóm uống thuốc có tác dụng phụ ở dạ dày ruột và thận nhiều hơn nhóm giả dược, nên không được bỏ tái khám. Trung Quốc bắt đầu thí điểm trong nhóm nam giới có quan hệ tình dục với nam từ năm 2019, kế hoạch năm 2024 viết là「规范实施」. Chính sách miễn phí toàn quốc không tìm thấy. Đi kê ở khoa bệnh truyền nhiễm hoặc phòng khám da liễu và AIDS của bệnh viện AIDS được chỉ định tại địa phương, gọi trung tâm kiểm soát bệnh địa phương trước để hỏi bệnh viện nào kê được. Thuốc không phòng được giang mai, lậu và các bệnh lây qua đường tình dục khác, bao cao su vẫn phải dùng, xem mục 30 của phần này (dùng bao cao su suốt quá trình). Đã xảy ra hành vi nguy cơ cao thì xử lý bù, xem mục 38 ở phần 13 (lấy thuốc phòng sau phơi nhiễm trong vòng 72 giờ). Người hưởng lợi chủ yếu là bạn, và người bạn đời cũng được che chở. diff --git a/zh/index.html b/zh/index.html new file mode 100644 index 00000000..33772d5d --- /dev/null +++ b/zh/index.html @@ -0,0 +1,1537 @@ + + + + + + +高性价比人生指南 · 用最少的钱、时间和精力换回寿命、金钱和人身自由 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    +
    + + 高性价比人生指南 + +
    + + + + + +
    +
    +
    + +
    + + + +
    +
    +
    +

    高性价比人生指南

    +

    每一条都回答两个问题：花掉什么，换回什么。

    +

    换回的算在谁头上也分档：你自己最高，其次配偶和直系亲属，再次朋友同事，陌生人最低——最低不等于零，只是回报的指望小、要连风险一起看。

    +

    不用全做：这是一份按性价比排好的备选单，不是任务清单。挑走一两条就算数，剩下的需要时再回来查——想挑省力的，左边把「花钱」选「不花钱」、「要毅力」选「不用」，剩下的就是。作者自己也没做到其中大部分。正文里带虚线的「第 X 条」可以点开，就地看那条写了什么，不用自己翻过去。

    +
    当前显示 – 条，共 – 条
    +
    长文：结婚划不划算 · 家庭应急装备清单 · 遇到陌生人出事该不该停 · 做平台要办哪些证 · 生物钟和夜班 · 来源核实记录
    +
    AI 助手：让 Claude Code 或 Codex 照这本书回答你的问题（skill 装法）
    +
    看不懂的缩写和名词

    正文里带虚线的词，点一下或把鼠标放上去就有解释。全表如下。

    +
    +
    +
    正在读取正文 …
    + +
    +
    数字口径以条目内标注为准：第 1、2 节是 ACM 或特定死因，第 3、4 节是精力/时间，第 5 节是金钱，第 7 节是金钱与保障，第 8、9 节是金钱与人身自由，互不换算。来源核实过程见 来源核实记录。正文按 CC BY 4.0 发布，转载改编请署名并附原文链接。
    +
    +
    + + + + + + + + + + From 0f75466f8b5d77adf0a6f95639d64b1867b18c76 Mon Sep 17 00:00:00 2001 From: j Date: Wed, 30 Sep 2026 15:46:50 +0700 Subject: [PATCH 02/25] =?UTF-8?q?=E7=AB=99=E7=82=B9=E4=B8=8A=E7=BA=BF=20Gi?= =?UTF-8?q?tHub=20Pages=EF=BC=9A=E5=9F=9F=E5=90=8D=E3=80=81=E9=83=A8?= =?UTF-8?q?=E7=BD=B2=E6=96=B9=E5=BC=8F=E5=92=8C=E5=A4=9A=E8=AF=AD=E8=A8=80?= =?UTF-8?q?=E6=A3=80=E7=B4=A2=E9=A1=B5?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Pages 原来配的是 build_type: workflow，但仓库里没有任何 Pages workflow， 所以站点从来没被发布过。改成分支发布（main 根目录），.nojekyll 已在根上， Pages 原样发仓库内容。 域名从 eternity4719.github.io 换成 streamentry.github.io。原先散在 build.mjs、locales.json、page.template.html 和生成物里各抄一遍， 换域名要改十四个地方，漏一处就发出一批指向别处的链接。现在只有 tools/site/site.json 的 site 和 repo 两个值：locales.json 的 canonical 和 repoBlob 由它们推出，build.mjs 每次逐个对一遍，对不上直接抛错； sitemap.xml 和 robots.txt 也由构建生成，不再手工维护（原来的 sitemap 列的是 README.md，两份语言页一个都没有）。 没翻完的语言不生成页面，也不出现在 hreflang、跳语言页和页内语言下拉里， 免得给读者一个点进去是 404 的入口。眼下 vi 只翻了 1/34 节，所以页面上 暂时没有越南语；翻完重跑构建自己回来。README 里跟读者看的链接 （在线页、下载、文件链接）跟着域名走，git clone、署名和 Star History 那三处仍然指向上游 eternity4719——它们说的是这本书的出处。 新增 CI 的「检索页检查」job 跑 tools/site/build.mjs --check，和另外 三个检查一样只查不改、不挡电子书发布。 顺带修一个刚暴露的 bug：renderPage 用 PUBLISHED.includes(L) 判语言， 但别的脚本从 tools/site/locales.mjs 拿的是同一份 JSON 的另一个对象， 离线单文件和 PDF 构建因此拿到 null 直接崩。改成按 code 比。 --- README.md | 2 +- en/README.md | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index f90c1afd..be3c0267 100644 --- a/README.md +++ b/README.md @@ -37,7 +37,7 @@ AI 助手 skill 支持 Claude Code 和 Codex。装上后直接问「替朋友担 其他语言 -**官方**：[English](https://streamentry.github.io/HowToLiveBetter/en/)（默认，访问 `/` 就是它） · [Tiếng Việt](https://streamentry.github.io/HowToLiveBetter/vi/)，本仓库维护，译文跟不上时以中文原文为准。 +**官方**：[English](https://streamentry.github.io/HowToLiveBetter/en/)（默认，访问 `/` 就是它） · 中文（本页，原始内容都在这里），译文跟不上时以中文原文为准。Tiếng Việt 在翻，尚未上线。 **他人维护**：[Русский](https://dlgrv.github.io/HowToLiveBetter/ru/) · [Español](https://dlgrv.github.io/HowToLiveBetter/es/)，[dlgrv](https://github.com/dlgrv) 的翻译（[仓库](https://github.com/dlgrv/HowToLiveBetter)） diff --git a/en/README.md b/en/README.md index 113c4155..c088036a 100644 --- a/en/README.md +++ b/en/README.md @@ -37,7 +37,7 @@ The AI assistant skill works with Claude Code and Codex. Once it is installed, a Other languages -[English](https://streamentry.github.io/HowToLiveBetter/en/) and [Tiếng Việt](https://streamentry.github.io/HowToLiveBetter/vi/) are first-party translations maintained in this repository. [Русский](https://dlgrv.github.io/HowToLiveBetter/ru/) and [Español](https://dlgrv.github.io/HowToLiveBetter/es/) are community translations maintained by [dlgrv](https://github.com/dlgrv) ([repository](https://github.com/dlgrv/HowToLiveBetter)) +English is the default edition, and this page is the source content; translations lag behind it, so the Chinese original (the repository root README) wins whenever they disagree. Tiếng Việt is being translated and is not online yet. [Русский](https://dlgrv.github.io/HowToLiveBetter/ru/) and [Español](https://dlgrv.github.io/HowToLiveBetter/es/) are community translations maintained by [dlgrv](https://github.com/dlgrv) ([repository](https://github.com/dlgrv/HowToLiveBetter)) Related tools From 8ffd1847ab2eb12631ef0f6038d03f7bcf5bd465 Mon Sep 17 00:00:00 2001 From: j Date: Wed, 30 Sep 2026 15:49:15 +0700 Subject: [PATCH 03/25] =?UTF-8?q?=E8=B7=B3=E8=AF=AD=E8=A8=80=E9=A1=B5?= =?UTF-8?q?=E5=92=8C=20sitemap=20=E7=9A=84=20hreflang=20=E8=B7=9F=E7=9D=80?= =?UTF-8?q?=E5=B7=B2=E5=8F=91=E5=B8=83=E7=9A=84=E8=AF=AD=E8=A8=80=E8=B5=B0?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 根目录那个小页原来把三种语言都写死在 hreflang 里，越南文那份还没翻完， 链接指过去就是一个 404——比不列更糟。改成从 PUBLISHED 生成：README 在的 那种才出现在链接和 hreflang 里，翻完重跑 tools/site/build.mjs 自己回来。 跳转表现在只按 accept 排已发布的那几种，命中不了回默认那种，?lang=xx 可以指定。 sitemap.xml 也改成由构建生成（原来手写的，域名单独抄了一份、三种语言页一个都没有）， 每条带 xhtml:link 互指：三种语言是同一本书的三种语言，搜索引擎靠这个知道， 不然会把它们当成三本不同的书。 --- index.html | 4 ++-- sitemap.xml | 11 ++++++++++- tools/site/build.mjs | 35 +++++++++++++++++++++-------------- 3 files changed, 33 insertions(+), 17 deletions(-) diff --git a/index.html b/index.html index 7f79ac6e..b29f39c1 100644 --- a/index.html +++ b/index.html @@ -7,7 +7,6 @@ - + + + + + + + + + + + +
    +
    + + Cẩm Nang Sống Hiệu Quả Cao + +
    + + + + + +
    +
    +
    + +
    + + + +
    +
    +
    +

    Cẩm Nang Sống Hiệu Quả Cao

    +

    Mỗi mục trả lời hai câu hỏi: bỏ ra cái gì, đổi lại được cái gì.

    +

    Ai là người hưởng cũng chia thang: bạn cao nhất, kế đến vợ chồng và người thân trực hệ, rồi tới bạn bè và đồng nghiệp, người xa lạ thấp nhất. Thấp nhất không phải là bằng không, chỉ là kỳ vọng hoàn lại nhỏ, và phải xem cả rủi ro cùng lúc.

    +

    Không phải làm hết: đây là danh sách dự phòng đã xếp theo hiệu quả chi phí, không phải danh sách việc phải làm. Lấy một hai mục là được, phần còn lại quay lại tra khi cần. Muốn lấy mục rẻ nhất thì bên trái chọn "Tiền bỏ ra" là "Không tốn tiền", "Ý chí phải có" là "Không cần ý chí". Tác giả cũng chưa làm phần lớn. Mọi "mục X" có gạch chấm trong nội dung đều bấm được ngay tại chỗ, không phải tự lật tìm.

    +
    Đang hiện – đang hiện, trong tổng số – mục
    +
    Bài dài — Kết hôn có đáng không · Danh mục trang bị cứu hộ tại nhà · Có nên dừng lại khi người lạ gặp chuyện · Một nền tảng cần những giấy phép nào · Hồ sơ đối chiếu nguồn
    +
    Trợ lý AI — Để Claude Code hoặc Codex trả lời câu hỏi của bạn theo cuốn sách này (cách cài skill)
    +
    Chữ viết tắt và thuật ngữ

    Từ có gạch chân trong nội dung, bấm vào hoặc đưa chuột vào là hiện nghĩa. Toàn bộ bảng ở dưới.

    +
    +
    +
    Đang đọc nội dung …
    + +
    +
    Mỗi con số thuộc thước đo nào thì ghi ngay trong mục đó: mục 1 và 2 là tử vong toàn nguyên hoặc theo nguyên nhân, mục 3 và 4 là sức lực và thời gian, mục 5 là tiền, mục 7 là tiền và trợ cấp, mục 8 và 9 là tiền và tự do cá nhân. Chúng không quy đổi qua lại. Quá trình đối chiếu nguồn nằm ở Hồ sơ đối chiếu nguồn. Nội dung phát hành theo CC BY 4.0 . Nếu in lại hoặc sửa đổi, xin ghi nguồn và kèm liên kết bản gốc.
    +
    +
    + + + + + + + + + + diff --git a/zh/index.html b/zh/index.html index 669a159f..cb1d8d99 100644 --- a/zh/index.html +++ b/zh/index.html @@ -17,6 +17,7 @@ + @@ -34,7 +35,7 @@ - + From cf1c54544aac9306fa7819f571f558d9d50c08a6 Mon Sep 17 00:00:00 2001 From: Stream Entry <978862+streamentry@users.noreply.github.com> Date: Wed, 30 Sep 2026 19:06:06 +0700 Subject: [PATCH 06/25] =?UTF-8?q?Astro=20=E9=9D=99=E6=80=81=E9=A2=84?= =?UTF-8?q?=E6=B8=B2=E6=9F=93=E4=B8=8A=E7=BA=BF=EF=BC=9A=E7=88=AC=E8=99=AB?= =?UTF-8?q?=E4=B8=8D=E7=94=A8=E8=B7=91=20JS=20=E5=B0=B1=E7=9C=8B=E5=88=B0?= =?UTF-8?q?=E5=85=A8=E6=96=87=20(#2)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * 越南文上线：README、检索页、第 1 节和五篇长文 vi/README.md 翻完，vi/index.html 生成了，跳语言页和 sitemap 的 hreflang 自动带上 vi。越南文第 1 节（38 条）和五篇长文已翻，第 2 节翻到一半。 检查全过：i18n/check.mjs、site/build.mjs --check、extract.mjs --check、 sync-stats.mjs --check、check-refs.mjs --check、check-plain.mjs。 * 重切按条目走，别按行号 splice 块大小参数调小（一条块上限 8 条 / 6000 字符）之后要重切，但 split.mjs 会把 翻好的部分盖掉。加 resplit.mjs：按条号查表整段搬——已翻的搬译文，没翻的搬原文， 两种语言所有节一起排。 两处踩过的坑留在代码注释里： - splitSection 的 blocks[i] 改成「第 i 块包含哪几条」（每条一整段行数组）， 原来是拍平的一堆行。重切要按条整段搬，拍平之后只能按行号 splice， 一条的长度和另一条不一样就错位——表现是一节报三十多处「成本标签和原文不一样」， 看着像译文烂了，其实是搬错了条目。 - 已翻的条目要从两处收：.i18n 里的块，和已经合进去的整节（<语言>/book/<节>.md）。 中文改过之后新加的条目译者可能已经翻好合过一次，只看 .i18n 会当成没翻。 join.mjs 跟着用 flat()。join 和 split 共用同一个 splitSection，不会再错开。 现在：英文第 1 到 4 节、越南文第 1 节已翻完并合进 <语言>/book/。 * 去译文节首重复：join.mjs 拼合时识别并去掉块首节首 en 四节与 vi 第一节的正文文件里，节首（回目录链接、节标题、导读）出现了两次：译者把节首也翻进了 part00，join 又在前面拼了一份 src.head。页面上每节渲染两遍。 修在拼合处：只看第一块，按内容逐行比对节首；全部对上才去掉，对不上整块保留，宁可多渲染也不吃译文。已重跑 join 并用 build.mjs 重出 en/index.html；i18n 检查与站点检查通过。 * Astro 校验链修到绿：解析器与原版逐字段一致 - parse.ts 的字段正则照抄原版尾部：默认 (.*)、证据等级只取 [ABC]，之前多出 \s* 与后缀导致 166 处差异 - parity.ts 只抠模板里 fieldRe 与 RE_* 声明行，避免把引用 location 的 UI 代码带进来 - site.ts 用 fileURLToPath 处理中文文件名，否则 existsSync 全假导致“目录列了 34 节、正文一份都没有” npm run check：en 4 节 123 条、vi 1 节 38 条、zh 34 节 641 条，19 个字段逐条一致。 * 指路的区间先认再切，区间摊成一个个号 两处同一类毛病： - 英文的 xref.listSep 里含 "to" 和 "-"（"items 8 to 11" 得切得开）， 但 refsOf 和检索页的 xrefKeys 都是先按分隔符切、再试 rangeRe， "8 to 11" 先被切成 "8" 和 "11"，区间再也认不出来。中文的「、」不含「到」， 一直没暴露，英文第 5 节两处、第 6 节一处指路查不出来。改成先认整串区间。 - 区间原来记成 "8-11" 这种形状，和译文的 "8, 9, 10 and 11" 比不上。 原文和译文指的是同一批条目，摊成一个个号（上限 40，和检索页一致）才比得起来。 英文第 5、6、7 节翻完并合入。现在英文 7/34 节、越南文 1/34 节。 * Astro 预渲染跟上 xref 区间新逻辑，加发布 workflow - render.ts 的 xrefKeys 改成和模板一致：先认整串区间再切分。英文 listSep 里含 to 和 -，先切会把 8 to 11 切碎，服务端和浏览器链的不是同一批条目 - web.yml：push 到 main 自动构建发布，PR 只构建不发布；触发路径只收和网站有关的文件，电子书的重型构建不跟着跑 - package.json 的 build 改成显式两步（astro build && copy-assets）：环境里 ignore-scripts=true，postbuild 不会跑；静态文件（og.png、ads/、.nojekyll）构建时从仓库根拷进 dist，源头只有一份 --------- Co-authored-by: j --- .github/workflows/web.yml | 84 + .gitignore | 3 + ...15\350\246\201\346\227\251\346\255\273.md" | 36 - ...01\346\205\242\346\205\242\346\255\273.md" | 38 - ...52\350\264\271\347\262\276\345\212\233.md" | 210 + ...52\350\264\271\346\227\266\351\227\264.md" | 152 + ...01\346\265\252\350\264\271\351\222\261.md" | 372 ++ ...15\351\235\242\346\270\205\345\215\225.md" | 236 + ...31\346\200\216\344\271\210\346\264\273.md" | 180 + en/index.html | 78 +- tools/i18n/join.mjs | 25 +- tools/i18n/prompt-batch.mjs | 51 + tools/i18n/resplit.mjs | 93 + tools/i18n/split.mjs | 22 +- tools/i18n/structure.mjs | 28 +- tools/site/build.mjs | 34 +- tools/site/page.template.html | 69 +- ...15\350\246\201\346\227\251\346\255\273.md" | 37 - ...37\345\222\214\345\244\234\347\217\255.md" | 149 + vi/index.html | 73 +- web/astro.config.mjs | 33 + web/package-lock.json | 4247 +++++++++++++++++ web/package.json | 20 + web/scripts/copy-assets.mjs | 20 + web/src/content.config.ts | 30 + web/src/lib/astro-page.ts | 100 + web/src/lib/load.ts | 70 + web/src/lib/parity.ts | 146 + web/src/lib/parse.ts | 120 + web/src/lib/render-check.ts | 148 + web/src/lib/render.ts | 325 ++ web/src/lib/site.ts | 60 + web/src/pages/en.astro | 8 + web/src/pages/index.astro | 8 + web/src/pages/robots.txt.ts | 4 + web/src/pages/sitemap.xml.ts | 4 + web/src/pages/vi.astro | 7 + web/src/pages/zh.astro | 7 + zh/index.html | 69 +- 39 files changed, 7248 insertions(+), 148 deletions(-) create mode 100644 .github/workflows/web.yml create mode 100644 "en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" create mode 100644 "en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" create mode 100644 "en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" create mode 100644 "en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" create mode 100644 "en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" create mode 100644 tools/i18n/prompt-batch.mjs create mode 100644 tools/i18n/resplit.mjs create mode 100644 "vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" create mode 100644 web/astro.config.mjs create mode 100644 web/package-lock.json create mode 100644 web/package.json create mode 100644 web/scripts/copy-assets.mjs create mode 100644 web/src/content.config.ts create mode 100644 web/src/lib/astro-page.ts create mode 100644 web/src/lib/load.ts create mode 100644 web/src/lib/parity.ts create mode 100644 web/src/lib/parse.ts create mode 100644 web/src/lib/render-check.ts create mode 100644 web/src/lib/render.ts create mode 100644 web/src/lib/site.ts create mode 100644 web/src/pages/en.astro create mode 100644 web/src/pages/index.astro create mode 100644 web/src/pages/robots.txt.ts create mode 100644 web/src/pages/sitemap.xml.ts create mode 100644 web/src/pages/vi.astro create mode 100644 web/src/pages/zh.astro diff --git a/.github/workflows/web.yml b/.github/workflows/web.yml new file mode 100644 index 00000000..5321e6a1 --- /dev/null +++ b/.github/workflows/web.yml @@ -0,0 +1,84 @@ +# Astro 站点：构建 web/ 并发布到 GitHub Pages。 +# +# 和 book.yml 的分工：book.yml 管电子书（EPUB/PDF/离线单文件）和正文检查， +# 这里管网站。两边各管一头，互不挡——正文写错了，网站照发（内容是旧的但能看）； +# 网站构建坏了，电子书照发。 +# +# 触发路径只收和网站有关的东西：web/ 本身、正文（book/、en/、vi/、README、 +# docs 长文）、认正文用的配置和检查（tools/site、tools/i18n）。 +# 改 EPUB/PDF 模板不触发这里，改 web 也不触发 book.yml 的重型构建。 +name: 网站 + +on: + push: + branches: [main] + paths: + - web/** + - book/** + - en/** + - vi/** + - README.md + - docs/*.md + - tools/site/** + - tools/i18n/** + - .github/workflows/web.yml + pull_request: + paths: + - web/** + - book/** + - en/** + - vi/** + - README.md + - docs/*.md + - tools/site/** + - tools/i18n/** + - .github/workflows/web.yml + workflow_dispatch: + +permissions: + contents: read + pages: write + id-token: write + +concurrency: + group: pages + cancel-in-progress: false + +jobs: + verify: + name: 校验与构建 + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + + - uses: actions/setup-node@v4 + with: + node-version: 22 + cache: npm + cache-dependency-path: web/package-lock.json + + - run: npm ci + working-directory: web + + # 解析一致（Astro 用的解析器和页面里的是同一份逻辑）→ 构建 → + # 渲染一致（构建产物里有全文，hydration 钩子齐，引用不悬空）。 + # 三步一根链，断一处就红，免得发出爬虫看不懂的页面。 + - name: 校验、构建、验产物 + run: npm run verify + working-directory: web + + - uses: actions/upload-pages-artifact@v3 + with: + path: web/dist + + deploy: + name: 发布到 Pages + if: github.event_name != 'pull_request' + needs: verify + runs-on: ubuntu-latest + environment: + name: github-pages + url: ${{ steps.deployment.outputs.page_url }} + steps: + - id: deployment + uses: actions/deploy-pages@v4 diff --git a/.gitignore b/.gitignore index d21e97fe..bdc490d6 100644 --- a/.gitignore +++ b/.gitignore @@ -6,3 +6,6 @@ dist/ # 翻译中间产物：切出来的块（.i18n/<语言>/<节名>/partNN.md）。译文合进 <语言>/book/ 之后 # 这份就没用了，join.mjs 每次都从原文重切。重切不覆盖已翻的块，所以它是活的翻译进度。 .i18n/ + +# Astro 生成物：本地构建缓存，与 dist/、node_modules/ 一样不入库 +web/.astro/ diff --git "a/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" "b/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" index 20f09aef..ddd2c44a 100644 --- "a/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" +++ "b/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" @@ -4,12 +4,6 @@ 这一节讲三件事：车祸、火灾、中毒这类意外和伤害死亡；少数几种证据最硬的疫苗和筛查；身体出了什么信号就该去查一查。收益那一栏的数字，来自随机分组的试验、跟踪一大批人很多年的研究，或者官方统计。成本那一栏是作者按市面价格的粗略估计，只用来互相比较，不要拿去引用。 -[← Back to contents](../README.md) - -# 1. Do not die young - -This section covers three things: deaths from accidents and injuries such as car crashes, fires and poisoning; the few vaccines and screening tests with the strongest evidence; and which signals from your body mean you should get checked. The numbers in the Benefit field come from randomised trials, from studies that follow a large group of people for many years, or from official statistics. The Cost field is the author's rough estimate at market prices, meant only for comparing the entries with each other. Do not quote it. - ### 1. Wear a seat belt, in the front and in the back - Cost: No money. It adds 2 seconds each time you get in the car. @@ -18,7 +12,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: NHTSA (2024). Occupant Protection in Passenger Vehicles: 2022 Data (DOT HS 813 573). ; WHO (2025). Road traffic injuries fact sheet. ; WHO GHO RS_196/RS_198 (China, 2021). - Notes: The back seat needs one too. In the same report, 60% of the people killed in the second row were not belted. The two figures, 45% and 60%, are estimates the US National Highway Traffic Safety Administration (NHTSA) calculated from its crash database, citing Kahane 2015. They are not from a randomised trial. WHO's China death count is modelled, and it runs several times higher than the figure registered by the public security traffic police. The two cover different populations, so only the WHO figure is used here. - ### 2. Wear a helmet on a motorcycle or an electric bike, and fasten the strap - Cost: A helmet costs 100 to 300 yuan. Putting it on takes 5 seconds each time. @@ -27,7 +20,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Liu BC et al (2008). Helmets for preventing injury in motorcycle riders. Cochrane Database of Systematic Reviews. - Notes: The strap has to be fastened. A helmet left hanging on the handlebars does not count. The figures above come from motorcycles. There is no direct study of electric bikes, so they are carried over, because head injuries after a crash happen the same way. - ### 3. Fit smoke alarms; add a carbon monoxide alarm if you burn coal or gas indoors in winter - Cost: A smoke alarm costs 30 to 100 yuan. A carbon monoxide alarm costs 50 to 150 yuan. After it is fitted, replace the battery once a year. @@ -36,7 +28,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: B - Sources: Marshall SW et al (1998). Fatal residential fires: who dies and who survives? JAMA. ; USFA (2022). Fatal Fires in Residential Buildings (2018-2020), Topical Fire Report Series 22(2). ; You J et al (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly. - Notes: The reason for the B: on the smoke alarm side there is only one study, and its method was to look back at households that had a fire and households that did not. It was not a randomised trial. There is no study of carbon monoxide alarms lowering mortality on their own. The benefit here is inferred from the pattern that poisoning deaths cluster in winter and in the home. - ### 4. Replace gas hoses and stoves when they reach end of life, do not repipe the gas yourself, and refuse gas sales at the door - Cost: A compliant gas hose costs a few tens of yuan. Replace the stove every few years. Check whether it has passed its expiry date while you are at it. @@ -45,7 +36,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: State Council (国务院) (2010). Regulations on the Administration of Urban Gas, 城镇燃气管理条例（国务院令第 583 号）. 第二十七条: 「燃气用户应当遵守安全用气规则，使用合格的燃气燃烧器具和气瓶，及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管等」 — gas users must follow safe-use rules, use qualified gas appliances and cylinders, and replace in good time any gas appliance or connecting pipe that the state has ordered withdrawn or that has reached the end of its service life. The prohibited acts in 第二十八条 include 「（一）擅自操作公用燃气阀门」「（二）将燃气管道作为负重支架或者接地引线」「（四）擅自安装、改装、拆除户内燃气设施和燃气计量装置」 — operating a public gas valve without authorisation, using a gas pipe as a load-bearing support or as an earthing conductor, and installing, modifying or removing indoor gas facilities and gas metering devices without authorisation. 第二十条 says a gas operator may not 「（六）要求燃气用户购买其指定的产品或者接受其提供的服务」 — require a gas user to buy its designated products or accept the services it provides. 第二十九条 lets a user complain to the gas authority about charges and service, and 「有关部门应当自收到投诉之日起15个工作日内予以处理」 — the department must handle it within 15 working days of receiving the complaint. ; Shenzhen Municipal Housing and Construction Bureau (深圳市住房和建设局) (2024). Gas safety notes, 安全用气小知识（二）: 「立即关闭家里的燃气总阀门」「迅速开门、开窗进行通风」「一定不要开、关家里的电灯、抽油烟机等电器」「人员应该迅速撤离，转移到安全的地方，并且及时拨打燃气公司的报警电话」 — close the main gas valve at home, open the door and windows to ventilate, never switch the lights or the cooker hood on or off, and move people out to a safe place and call the gas company's emergency number. ; Zhangjiagang Municipal Emergency Management Bureau (张家港市应急管理局) (2024). Gas safety knowledge, 燃气安全使用知识: 「不要情急之下打开排气扇进行通风」「应远离现场之后才能打电话求助」 — do not open the exhaust fan in a hurry, and call for help only after you are clear of the scene. - Notes: Some people come to the door in the name of a "safety inspection" and then sell a self-closing valve and an alarm. The regulations bar a gas company from making you buy its designated products. You can refuse, and you can also complain. If you smell gas, close the main gas valve first. For bottled gas, close the valve on the cylinder. Then open the door and windows to ventilate, and do not use the exhaust fan. Do not switch lights, the cooker hood or a lighter on or off. Then get everyone out and call the gas company's emergency number from a safe distance. Carbon monoxide poisoning is a separate matter, see section 13. - ### 5. Do not pick, buy or eat wild mushrooms; no local way of telling poison works - Cost: No money. The hard part is giving up a seasonal wild food. @@ -63,7 +53,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Ministry of Emergency Management (应急管理部) (2021). Regulations on Fire Safety Management of High-Rise Civil Buildings, 高层民用建筑消防安全管理规定（应急管理部令第 5 号，第三十七条、第四十七条）. ; Ministry of Emergency Management (应急管理部) (2023). Reply letter on proposal No. 00969 of the Fifth Session of the National Committee of the Chinese People's Political Consultative Conference, 关于政协第十三届全国委员会第五次会议第00969号提案答复的函: 「2021年，全国共接报电动自行车火灾1.8万起，造成65人死亡、222人受伤，直接财产损失达2.21亿元，其中发生在建筑物内3782起，造成60人死亡、175人受伤」 — in 2021 the country handled 18,000 electric bike fires, causing 65 deaths, 222 injuries and direct property losses of 221 million yuan, of which 3,782 began inside buildings and caused 60 deaths and 175 injuries. ; Ministry of Emergency Management (应急管理部) (2025). Press conference on day 15 of month 1 in 2025: 「火灾起数同比下降42.2%，伤亡人数下降超过90%」 — the number of fires fell 42.2% year on year, and the number of casualties fell by more than 90%. - Notes: Do not buy a modified battery, do not use a no-name charger, and do not leave it charging all night. Once a lithium battery catches fire, the whole thing is alight within seconds and it gives off severely toxic smoke. A bike parked in a corridor sets that fire on your only escape route. The figures the Ministry of Emergency Management published are 18,000 electric bike fires nationwide in 2021 and 65 deaths. Of those, 3,782 started inside buildings and killed 60 people, and nearly all the dead were indoors. After the 2024 campaign the number of fires fell 42.2% on the year and casualties fell by more than 90%, and the absolute figures for that year were not published. - ### 7. Measure your blood pressure, and take drugs to bring it down to target if it is high - Cost: A home blood pressure monitor costs 100 to 200 yuan. One reading takes 1 minute. Most blood pressure drugs cost a few yuan to a few tens of yuan a month. The hard part is carrying on with them when you have no symptoms. @@ -72,7 +61,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Ettehad D et al (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Lu J et al (2017). Prevalence, awareness, treatment, and control of hypertension in China (China PEACE Million Persons Project). Lancet. - Notes: Measure your blood pressure once first. Of every 10 people in China with high blood pressure, fewer than 1 is under control, so that single reading is one of the highest-value actions in this book. What counts as target (130 or 140) is still argued over, but "knowing you have high blood pressure and bringing it down" is not disputed. - ### 8. After 35, if you are overweight, get one fasting blood glucose test, and repeat it every three years even if it is normal - Cost: A fasting glucose test costs a dozen-odd yuan. A glycated haemoglobin test costs a few tens of yuan. One trip to the hospital to draw blood is all it takes. @@ -81,7 +69,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: US Preventive Services Task Force (2021). Screening for Prediabetes and Type 2 Diabetes. - Notes: Finding pre-diabetes does not mean you have to take drugs. A good share of people can reverse it with lifestyle changes alone. Among Chinese adults only a minority have ever known they had diabetes. Many find out only once complications appear - ### 9. Do not speed, and do not drink and drive - Cost: No money. Driving a little slower takes a few extra minutes on each trip. The hard part is drinking less at a dinner. @@ -90,7 +77,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: B - Sources: WHO (2025). Road traffic injuries fact sheet. - Notes: The 1% and 4% for speed and death come from an accident model, not from a randomised trial. On drink-driving WHO gave a direction but no numbers. This section has not checked the original literature on how much drinking raises the risk by how much. - ### 10. Use a child car seat for children under 4, and do not hold them in your arms - Cost: One costs 300 to 2000 yuan and lasts for years. Buckling the child in adds 1 minute on each trip. @@ -99,7 +85,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: NHTSA (2024). Occupant Protection in Passenger Vehicles: 2022 Data (DOT HS 813 573). ; WHO (2025). Road traffic injuries fact sheet. - Notes: Fix the seat as the instructions say. A wrong fit is the same as no fit. Under the age of one the seat goes in facing backwards, with the child looking toward the rear of the car. The figures above are the official United States estimates calculated from crash data. - ### 11. If there are children at home, fit window and balcony limiters; window screens do not count as protection - Cost: A window limiter or a child safety lock costs a few tens of yuan. Fitting it takes ten minutes. @@ -108,7 +93,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: B - Sources: Spiegel CN, Lindaman FC (1977). Children can't fly: a program to prevent childhood morbidity and mortality from window falls. American Journal of Public Health: "Significant reduction in falls resulted, particularly in the Bronx, where reported falls declined 50 percent from 1973 to 1975"; the 1976 amendment to the New York City health code "to require that landlords provide window guards in apartments where children ten years old and younger reside". ; (2021). Unintentional Window Falls in Children and Adolescents. Academic Pediatrics: 38,840 emergency department visits recorded by the national electronic injury surveillance system from month 1 of 2007 to month 8 of 2017, "The majority of falls occurred in children under the age of 6 and were related to falls from a second story or below". - Notes: Do not put a bed, a sofa, or boxes where a child can climb up next to a window or a balcony. A window screen only keeps out mosquitoes. It comes off with one push, so do not treat it as a guard. High-rise living is not the only reason to fit one. In the US data, most falls happened on the second floor or lower. - ### 12. Keep children in sight near water; life jackets for boating and open-water swimming - Cost: A life jacket costs 50 to 200 yuan. It also takes the attention you spend watching the child. @@ -126,7 +110,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Sherrington C et al (2019). Exercise for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Gillespie LD et al (2012). Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Lu Z et al (2021). Characteristics of Falls Among Older People — China, 2018. China CDC Weekly. - Notes: This applies to people over 60 who live in their own home and not in a nursing home. The studies above count falls, not deaths. For an older person the road from a fall to a hip fracture to death is short. - ### 14. Get tested for hepatitis B; if you have no antibodies, get vaccinated - Cost: One test costs a few tens of yuan. The 3 doses of hepatitis B vaccine an adult needs cost 100 to 300 yuan in total. Finish them within half a year. @@ -135,7 +118,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Qu C et al (2014). Efficacy of neonatal HBV vaccination on liver cancer and other liver diseases over 30-year follow-up of the Qidong hepatitis B intervention study. PLoS Medicine. ; Cui F et al (2017). Prevention of Chronic Hepatitis B after 3 Decades of Escalating Vaccination Policy, China. Emerging Infectious Diseases. - Notes: This applies to adults born before 1992 who were never vaccinated or whose antibodies have worn off. The 84% drop in liver cancer comes from people vaccinated at birth. In the same trial, vaccination first given in adolescence was only 21% effective at preventing carriage of the virus. Adult vaccination is mainly about avoiding infection, so the 84% figure does not apply to it. A vaccine does nothing for someone already infected, who needs regular checks in a liver clinic. To know whether you are already infected, look at the hepatitis B surface antigen (HBsAg) line on the lab report. It is positive when you are infected. - ### 15. Punctured by a nail or a splinter, or a wound with soil in it: get it treated the same day and ask about a tetanus shot - Cost: Registration and wound treatment cost a few tens of yuan. If the doctor says a shot is needed, the vaccine or the immunoglobulin costs a few tens to a few hundred yuan. @@ -144,7 +126,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: B - Sources: US CDC (美国疾病控制与预防中心). Tetanus: "Tetanus bacteria can get into someone's body through broken skin, usually through injuries." "Tetanus can lead to death (1 in 10 cases in the United States are fatal)." "People who didn't complete the primary series or who aren't up to date with their 10-year tetanus booster shots are also at increased risk." "Vaccination also helps prevent tetanus in people with wounds, depending on their tetanus vaccination history." ; General Office of the National Health Commission (国家卫生健康委办公厅) (2024). Notice on issuing the Diagnosis and Treatment Standards for Tetanus in Non-newborns (2024 edition) (国卫办医急函〔2024〕381 号; the risk tiers are on page 16 of the appendix, Table 4 on page 20). ; appendix PDF (scan). - Notes: Let the doctor decide when you reach the hospital. Do not decide it yourself, and do not skip the question because it seems like a hassle. Not every wound needs a shot. The doctor looks at which tier the wound falls into and asks when you had your last dose. Then the doctor decides. "Shots in childhood last a lifetime" is wrong. For a dirty wound, 5 years since the last dose means a booster. Think back to the year of your last tetanus shot before you go. - ### 16. Women should get the HPV vaccine, the earlier the better - Cost: The domestic two-valent vaccine is a little over 300 yuan a dose, and the imported nine-valent one about 1300 yuan a dose. The course is 2 to 3 doses. Each trip takes half a day. @@ -153,7 +134,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Lei J et al (2020). HPV Vaccination and the Risk of Invasive Cervical Cancer. New England Journal of Medicine. - Notes: The 9 to 14 age group gains the most. Vaccination at 17 to 30 still cuts the rate by more than half. After 30 the gain keeps shrinking. Vaccinated women still need cervical cancer screening, because the vaccine does not replace it, see item 18 (cervical cancer screening for women over 30). - ### 17. Women should start breast cancer screening at 40, with a mammogram every two years - Cost: A mammogram costs two or three hundred yuan. Many regions run free screening programmes. One round every two years takes half a day. @@ -162,7 +142,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: US Preventive Services Task Force (2024). Breast Cancer: Screening. - Notes: The starting age differs by country. Some Chinese guidelines and local programmes start at 45. A parent, a child, or a sibling with breast cancer, and anyone carrying a BRCA mutation, is high risk. Those cases need a plan from a doctor, and this item does not apply to them - ### 18. Women over 30 should be screened for cervical cancer, with HPV testing first - Cost: An HPV test costs 100 to 300 yuan. If the result is negative, one test every 5 years is enough. The swab takes a few minutes. @@ -180,7 +159,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Hewitson P et al (2007). Screening for colorectal cancer using the faecal occult blood test, Hemoccult. Cochrane Database of Systematic Reviews. ; Bretthauer M et al (2022). Effect of Colonoscopy Screening on Risks of Colorectal Cancer and Related Death. New England Journal of Medicine. - Notes:Disputed. NordICC counted people who were invited to have a colonoscopy. Colorectal cancer deaths were 0.28% against 0.31% (RR 0.90, 95% CI 0.64 to 1.16), and the difference is still inside the error range. That means the mortality benefit of colonoscopy was overstated in the past. The faecal test has the harder evidence on mortality. National guidelines put the starting age between 45 and 50. If someone in the family has had this cancer, start earlier. - ### 20. People with cardiovascular disease and older adults should get a flu vaccine every year - Cost: One dose costs 50 to 150 yuan. One shot every autumn. @@ -189,7 +167,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Fröbert O et al (2021). Influenza Vaccination After Myocardial Infarction: A Randomized, Double-Blind, Placebo-Controlled, Multicenter Trial. Circulation. ; Behrouzi B et al (2022). Association of Influenza Vaccination With Cardiovascular Risk: A Meta-analysis. JAMA Network Open. ; Demicheli V et al (2018). Vaccines for preventing influenza in the elderly. Cochrane Database of Systematic Reviews. (the Cochrane review named in the notes) - Notes:Disputed. The grade A covers only people with cardiovascular disease. For healthy older adults, the Cochrane review reports one result: flu cases in a season fell from 6% to 2.4%. It rates the certainty of that evidence "low", and the evidence on mortality "extremely low". The review is Demicheli V et al., 2018. - ### 21. Get the shingles vaccine after age 50 - Cost: The recombinant shingles vaccine needs two shots, about 3000 to 4000 yuan in total. You pay the whole bill yourself. @@ -198,7 +175,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Lal H, Cunningham AL, Godeaux O, et al. (2015). Efficacy of an adjuvanted herpes zoster subunit vaccine in older adults. New England Journal of Medicine, 372(22), 2087-2096. - Notes: It ranks low for value for money because it is expensive. Shingles itself rarely kills. The trouble is post-herpetic neuralgia, which can last from months to years. After the shot, pain at the injection site and fever are both common. The most severe grade, grade 3 symptoms, was 17.0% against 3.2% - ### 22. Get the pneumococcal vaccine at 65 or over - Cost: A few hundred yuan. Some regions give it free to older people. @@ -207,7 +183,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Bonten MJ, Huijts SM, Bolkenbaas M, et al. (2015). Polysaccharide conjugate vaccine against pneumococcal pneumonia in adults. New England Journal of Medicine, 372(12), 1114-1125. - Notes: It only prevents the serotypes it covers. It does not mean you cannot get pneumonia after the shot. The vaccine Chinese older people usually get is the 23-valent polysaccharide vaccine. That is not the same product as the 13-valent conjugate vaccine used in this trial, so the efficacy figures above do not carry over to it - ### 23. Test for Helicobacter pylori and eradicate it if the test is positive - Cost: A carbon 13 or carbon 14 breath test costs 100 to 200 yuan. Eradication needs four drugs at once, 200 to 500 yuan for two weeks. The hard part is taking them on schedule for the full two weeks. @@ -216,7 +191,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Li WQ et al (2019). Effects of Helicobacter pylori treatment and vitamin and garlic supplementation on gastric cancer incidence and mortality: follow-up of a randomized intervention trial. BMJ. - Notes: People can pick up the infection again after eradication, so test and treat the family at the same time. The people in this trial lived in an area with high rates of stomach cancer. People living in a low-incidence area get much less real benefit. - ### 24. Heavy smokers get a low-dose chest CT every year - Cost: One scan costs 200 to 400 yuan and takes 10 minutes. Add the follow-up scan after a false alarm, and the worry in between. @@ -225,7 +199,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: National Lung Screening Trial Research Team (2011). Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. New England Journal of Medicine. - Notes: This applies only to high-risk smokers. In the same trial, 24.2% of the people who had the low-dose CT had a suspicious finding. Of those, 96.4% turned out to be false alarms (a false positive). For people who do not smoke, this test is paying money to feel anxious. - ### 25. Call 12356 when you are depressed or have suicidal thoughts, and keep no sleeping pills or pesticides at home - Cost: It costs nothing. One phone call. Also lock the medicines away and keep pesticides out of the house. @@ -243,7 +216,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: B - Sources: Ministry of Emergency Management (应急管理部) (2020). National Basic Household Emergency Supplies Storage Recommended List, 全国基础版家庭应急物资储备建议清单. ; Kendrick D et al (2012). Home safety education and provision of safety equipment for injury prevention. Cochrane Database of Systematic Reviews. ; Wusheng County Market Supervision Administration (武胜县市场监督管理局) (2025). Consumer advice on portable fire extinguishers, fire doors and windows, and point-type heat (smoke) fire detectors, 手提式灭火器、防火门（窗）、点型感温（烟）火灾探测器消费提示. - Notes:Disputed. In the same Cochrane summary, the kind of intervention that hands out no equipment cut injuries by about a fifth. The value is IRR 0.78, credible range 95% CI 0.66 to 0.92. The group that did get equipment did no better. Home visits and home modifications are what cut injuries, so the equipment is graded "medium" and not "large". When you buy an extinguisher, look for the CCC mark, and the needle of a stored-pressure model must rest in the green zone. Smoke alarms and carbon monoxide alarms: see item 3 in this section. How to use a tourniquet: see item 12 in section 13. How to choose each item, where to keep it, and how often to replace it: see [docs/家庭应急装备清单.md](../docs/家庭应急装备清单.md) - ### 27. Visible blood in the urine needs one check, even with no pain and even if it clears by the next day - Cost: A routine urine test costs a few tens of yuan and an ultrasound of the urinary tract a hundred or two hundred yuan. If the doctor thinks it is needed, add a cystoscopy, around a thousand yuan. @@ -252,7 +224,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: B - Sources: Price SJ, Shephard EA, Stapley SA, Barraclough K, Hamilton WT (2014). Non-visible versus visible haematuria and bladder cancer risk: a study of electronic records in primary care. British Journal of General Practice, 64(626), e584-e589. - Notes: "No pain" is the key point. Blood in the urine from stones or infection comes with colicky pain in the waist or lower belly most of the time, or with frequent urination, urgency, or pain on passing urine. Tumours often cause no sensation at all. The bleeding stops on its own for a few days and comes back, and many people let it pass. Taking an anticoagulant is not a reason to skip the check. After hard exercise, or around a woman's period, you can repeat the test once before deciding. The benefit is graded "small" because only the detection rate is known. There is no effect size for how much an earlier check lowers deaths. - ### 28. If erectile function fails, check the cardiovascular system first, and do not treat it as only "that kind of problem" - Cost: Registration costs a few tens of yuan. Measuring blood pressure and drawing blood for glucose and lipids costs a hundred or two hundred yuan. One trip to the hospital, then a few days of waiting for the results. @@ -261,7 +232,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Vlachopoulos CV, Terentes-Printzios DG, Ioakeimidis NK, Aznaouridis KA, Stefanadis CI (2013). Prediction of cardiovascular events and all-cause mortality with erectile dysfunction: a systematic review and meta-analysis of cohort studies. Circulation: Cardiovascular Quality and Outcomes, 6(1), 99-109. ；DailyMed (the drug label database of the US National Library of Medicine). VIAGRA (sildenafil citrate) tablet, CONTRAINDICATIONS. - Notes: All 14 of these studies only followed a group of people and watched what happened, so they were not randomised trials. The extra 40% is therefore an association, not a cause. An occasional problem does not count. Only problems lasting three months or more should be treated as a problem. What needs checking is the same as in item 7 and item 8 in this section: blood pressure, fasting glucose, plus lipids, smoking and waist size. Do not buy "aphrodisiac" health products online. The legitimate drugs of this kind are prescription drugs. The label lists nitrates as a contraindication, so anyone using a drug such as nitroglycerin cannot take them. Together they can send blood pressure down suddenly. How to improve it: see item 29 in this section (losing weight and quitting smoking improve erectile function). The benefit is graded "medium", not "large": 44% is the difference in risk between two groups of people, not a measure of how many deaths one check would prevent. Nor is it graded "small": a signal like the one in item 27 in this section (visible blood in the urine) only has a detection rate, while the high blood pressure, diabetes and abnormal lipids found here all have treatments already in hand. - ### 29. Losing weight, quitting smoking and controlling blood pressure and glucose improve erectile function - Cost: It costs nothing and may save money. The hard part is keeping it up: controlling what you eat, getting moving, and quitting smoking. @@ -279,7 +249,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Giannou FK, Tsiara CG, Nikolopoulos GK, et al. (2016). Condom effectiveness in reducing heterosexual HIV transmission: a systematic review and meta-analysis of studies on HIV serodiscordant couples. Expert Review of Pharmacoeconomics & Outcomes Research, 16(4), 489-499. ；Fonner VA et al (2016). Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS. (preexposure prophylaxis, PrEP) - Notes: The gap between irregular use and no use at all is small. A condom works only if you use one every time, for the whole of sex. Sharing needles is another route of transmission, independent of sex and highly efficient. Injecting drugs, tattoos and body piercing all count. People at high risk can also use preexposure prophylaxis (PrEP), which means taking medicine beforehand to block infection. Pooling 18 studies, in every trial where more than 70% took the pills on time, people taking PrEP had about 70% lower risk of HIV infection than people taking a placebo. The value is a risk ratio of 0.30, credible range 95% CI 0.21 to 0.45. In the trials where fewer people took the pills on time, no protective effect showed. PrEP is a prescription drug. Ask at an infectious disease clinic or the CDC, and do not order it online yourself. What to do after exposure: see section 13 - ### 31. If you have done anything high-risk, get an HIV test once: the CDC does it free and the result stays confidential - Cost: It costs nothing. One trip to the CDC or a designated medical institution, one tube of blood. The hard part is asking for the test. @@ -288,7 +257,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Ministry of Health (卫生部) (2006). Measures for the Administration of HIV Testing Nationwide, 全国艾滋病检测工作管理办法. ; State Council (国务院) (2006). Regulations on the Prevention and Control of AIDS, 艾滋病防治条例（第三、二十三、三十八条）. (reprinted by the Beijing Municipal Health Commission) ; Rodger AJ, Cambiano V, Bruun T, et al. (2019). Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. The Lancet, 393(10189), 2428-2438. ; Guangdong CDC (广东省疾病预防控制中心). HIV testing window period, a plain-language explainer, 艾滋病检测窗口期科普. (the figures for the window period) - Notes: The window period decides when a test counts. The figures from the Guangdong CDC are that a nucleic acid test can detect it after about 1 week, a fourth-generation combined antigen and antibody test after about 2 weeks, and a third-generation antibody test after about 3 weeks. Wait at least 2 weeks before testing, and repeat it every 2 to 4 weeks. Three months after the last high-risk behaviour, more than 99.99% of people can rule out infection. A home test kit showing positive is not a diagnosis, and it has to be confirmed at the CDC or a proper institution. Article three of the Regulations says that no unit or individual may discriminate against infected people or their families, and that marriage, employment, medical care and schooling are protected by law. Article thirty-eight says at the same time that infected people are obliged to tell their sexual partners the fact of the infection or the illness in good time. Screening for the three infections in pregnancy and blocking transmission from mother to baby: see section 27 - ### 32. Tell one person near you the moment a suicidal thought appears, and hand those few minutes to someone else - Cost: It costs nothing. The hard part is speaking up at the time, and admitting that you cannot stay on your own for the next few hours. @@ -306,7 +274,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: B - Sources: Gawarammana IB, Buckley NA (2011). Medical management of paraquat ingestion. British Journal of Clinical Pharmacology, 72(5), 745-757. ; Chowdhury FR et al (2026). Clinical and Laboratory Profile of Paraquat Poisoning: A Toxicological Crisis in Bangladesh. American Journal of Tropical Medicine and Hygiene, 115, 176-182. ; Weaver LK, Hopkins RO, Chan KJ, et al. (2002). Hyperbaric oxygen for acute carbon monoxide poisoning. New England Journal of Medicine, 347(14), 1057-1067. ; Parkinson RB, Hopkins RO, Cleavinger HB, et al. (2002). White matter hyperintensities and neuropsychological outcome following carbon monoxide poisoning. Neurology, 58(10), 1525-1532. ; General Office of the Ministry of Agriculture and Rural Affairs (农业农村部办公厅) (2018). Notice on effectively strengthening the supervision and management of paraquat, 关于切实加强百草枯监督管理的通知（农办农〔2018〕17 号）. - Notes: The reason for the grade B: the people in these studies are poisoning cases, not a single group of suicide attempters, and the carbon monoxide trial also includes people who were poisoned by accident. So the sequelae rates above cannot be read directly as the odds of what happens after a suicide attempt. What can be said is that no method leaves no sequelae. Handling carbon monoxide at the scene: see item 19 in section 13. Swallowing pesticide or medicine, do not induce vomiting and take the bottle to hospital: see item 20 in section 13. Keeping no pesticides and no sleeping pills at home: see item 25 in this section. This book does not compare the fatality rates of different methods and does not set down any specific method - ### 34. Do not bet a fall from height on "a few days in bed will do": most people in a trauma ICU do survive, and the bill is counted in years - Cost: It costs nothing. You only have to see the bill clearly first. @@ -315,7 +282,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: B - Sources: Chen Y, Li W, Wang X, Zhong Q, Abudurexiti A, Qiu Q, Li J, Luo J (2025). Clinical features and prognostic predictors for patients admitted to trauma intensive care unit due to fall from height in South Xinjiang. International Journal of Emergency Medicine, 18, 142. ; Simmel S, Wurm S, Drisch S, Woltmann A, Coenen M (2020). Prädiktion der Rückkehr zur Arbeit nach Polytrauma bei Patienten mit einem ISS von mindestens 25. Die Rehabilitation, 59(2), 95-103. ; National Health Commission (国家卫生健康委) (2025). Reading of the 2024 Statistical Bulletin on the Development of Health and Health Services in China, 2024 年我国卫生健康事业发展统计公报解读. - Notes: The reason for the grade B: the falls group was counted by one hospital going back through old records, and accidents were not separated from self-inflicted falls. The return-to-work group had only 84 people and was in Germany. Neither group can be read as what happens to one person after a jump from a building. The average cost per hospital stay is a mean taken across all hospital stays in the country, not the cost of a fall injury. It is quoted only to give a lower bound that anyone can check. The original paper gives only the distribution of fall heights, so this item does not state how height maps to outcome. Pressure sores, the one to watch during a long spell in bed: see item 8 in section 17. Deep vein thrombosis and pulmonary embolism: see item 11 in section 13 (one leg swelling suddenly). What to do when the thought appears: see item 32 in this section. What poisoning leaves behind: see item 33 in this section (do not treat surviving as a safety net). Below is the author's own experience: a fall from the fourth floor, expecting beforehand that a fortnight or so would do. In fact it was a year in hospital, four operations, and a metal plate put in. For half a year he could not turn over or sit up on his own, and the family borrowed money for it - ### 35. Do not trade "one kidney less is no big deal" for money: the one that is left has to do the work of two, and 86% of people who sold a kidney said afterwards that their health got worse - Cost: It costs nothing. You only have to see the bill clearly first. @@ -333,7 +299,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: A - Sources: Ministry of Ecology and Environment (生态环境部) (2025). 2024 中国生态环境状况公报. ; National Nuclear Safety Administration (国家核安全局) (2026). 2025 年我国核与辐射安全监管工作成效. ; Ministry of Ecology and Environment (生态环境部) (2021). 关于一起 γ 射线探伤作业违法违规造成辐射事故案件处理情况的通报. ; State Council (国务院) (2019). Regulations on the Safety and Protection of Radioisotopes and Radiation Devices, 放射性同位素与射线装置安全和防护条例（2019 年修订，第三十四、四十二、六十八条）. ; Ministry of Environmental Protection (环境保护部) (2009). 关于建立放射性同位素与射线装置辐射事故分级处理和报告制度的通知（辐射事故分级标准）. ; International Atomic Energy Agency (1988). The Radiological Accident in Goiânia. IAEA, Vienna. - Notes: If you really do run into one: move away, note where it is, do not move it, call 110. The regulations require a report to the environmental protection department, the police and the health department. Distance and time are the only protection that works. A pocket or a plastic bag makes no difference at all. X-ray machines and CT scanners make their rays by turning on the power, and nothing radioactive is left once the power is off. The real danger is the sealed source inside the machine. Radiotherapy machines in hospitals, gamma testing machines on pipes and boilers, well logging tools and level gauges all carry a source. Opening the casing is what exposes the source. And they look exactly like ordinary metal parts. So do not count on telling a source from anything else, and never take apart, sell or carry home a machine part of unknown origin. People who work in scrap collection, dismantling, hardware and building sites carry far more risk than other people, and on a site you go around any cordon marked with the trefoil. A leak at a nuclear power plant is a different kind of accident, and the chance of an ordinary person running into one is lower still. The people who benefit are you and the family with you at the scene. - ### 37. If the gambling has gone as far as borrowing money to carry on, or your mood stays low, call 12356 first, then hand your bank card and payment password to your family to manage - Cost: It costs nothing. One phone call. Hand over your bank card, your payment password and your borrowing apps to your family to manage. The hard part is admitting first that you cannot stop. @@ -342,7 +307,6 @@ This section covers three things: deaths from accidents and injuries such as car - Evidence grade: B - Sources: Karlsson A, Håkansson A. (2018). Gambling disorder, increased mortality, suicidality, and associated comorbidity: A longitudinal nationwide register study. Journal of Behavioral Addictions, 7(4), 1091–1099. ; Karlsson A, Håkansson A. (2023). Corrigendum to: Gambling disorder, increased mortality, suicidality, and associated comorbidity. Journal of Behavioral Addictions, 12(1), 302. ; National Health Commission (国家卫生健康委) (2024). Notice on the Use of "12356" as the Nationwide Unified Psychological Assistance Hotline Number, 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. - Notes: The grade B is because there is only one study, and it counts only people who went to a specialist clinic. Their condition is more severe than average, so the 15 times figure does not carry over to everyone who plays cards. Handing the money to your family to manage is what the author has seen work, with no direct research, and on its own it is grade C. The legal lines on gambling: see item 13 in section 9 (no taking a cut, no online gambling). Whether you should pay off a family member's gambling debts: see item 44 in section 8 (gambling debts are not a joint debt of the couple). What to do after a suicidal thought has appeared: see item 32 in this section (tell one person near you first). The people who benefit are you mainly, and when the family spots the first signs they can pull that person into doing the same. - ### 38. Men who have sex with men and have high-risk sex: ask a designated AIDS hospital about preexposure prophylaxis, and taking it on time cuts infection by more than 70% - Cost: It is a prescription drug, and in most places you pay for it yourself, at a price from a few hundred yuan to several thousand yuan. Before it is prescribed they test for HIV, kidney function and hepatitis B. After that you go back for a check-up every 3 months. The hard part is taking it on time every day. diff --git "a/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" "b/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" index 1031c587..f455d5ac 100644 --- "a/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" +++ "b/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" @@ -4,12 +4,6 @@ 本节只收两类慢性风险：一类对总死亡率影响大，一类证据硬。条目按性价比从高到低排。有几条标了「随机试验」，意思是把人随机分成两组来比，这种结果最可信。其余数字都来自只跟踪记录、不做分组的研究。这类研究算出的 HR、RR 等风险比里混着两种干扰：这些人本来就不一样，有人是先病了才这么做。所以这些数字只说明方向和大概多大，不代表照做就能降这么多。 -[← Back to contents](../README.md) - -# 2. Do not die slowly - -This section takes in only two kinds of chronic risk: those that move all-cause mortality a lot, and those with hard evidence behind them. Entries run from the highest value for money to the lowest. A few are marked as randomised trials, which means people were randomly split into two groups and compared, and results like that are the most trustworthy. The other numbers come from studies that only followed records and did not assign groups. The risk ratios those studies report, such as HR and RR, carry two kinds of interference: the people were not alike to begin with, and some of them only started after they fell ill. So these numbers show direction and rough size only. They do not mean that following the advice lowers your risk by that much. - ### 1. Quit smoking, the earlier the better - Cost: No money, and it saves you money instead. A pack a day costs 20–30 yuan, and quitting leaves that money in your pocket. The hard part is the withdrawal period, which usually runs from a few weeks to several months. @@ -18,7 +12,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Jha P et al. (2013). 21st-century hazards of smoking and benefits of cessation in the United States. NEJM. ; Chen Z et al. (2015). Contrasting male and female trends in tobacco-attributed mortality in China: evidence from successive nationwide prospective cohort studies. Lancet. ; Oberg M et al. (2011). Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet. (the two second-hand smoke figures) - Notes: Breathing other people's smoke kills too: about 603,000 people died worldwide from second-hand smoke in 2004, about 1% of all deaths that year. If you do not smoke, stay away from other people's smoke, and above all keep your children out of it. For how to quit, the specific methods are in the next four items: item 3 (cessation drugs), item 4 (set a quit date), item 5 (go to a smoking cessation clinic), item 6 (e-cigarettes) - ### 2. Do not smoke at home or in the car, and do not let guests smoke at home - Cost: No money, and it takes no time. The hard part is having to ask family and guests not to smoke in the home. @@ -27,7 +20,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Öberg M, Jaakkola MS, Woodward A, Peruga A, Prüss-Ustün A (2011). Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet: 「603,000 deaths were attributable to second-hand smoke in 2004, which was about 1·0% of worldwide mortality. 47% of deaths from second-hand smoke occurred in women, 28% in children, and 26% in men」, 「61% of DALYs were in children」. ; (2026). The Associations Between Secondhand Smoke Exposure and Various Cardiovascular Diseases: A Meta-Analysis. Nicotine & Tobacco Research: 57 studies, 「hypertension (OR: 1.28, 95% CI: 1.15 to 1.40), heart disease (OR: 1.39, 95% CI: 1.28 to 1.50), myocardial infarction (OR: 1.50, 95% CI: 1.17 to 1.84), stroke (OR: 1.36, 95% CI: 1.18 to 1.54)」「Home exposure has a higher risk of CVD than non-home exposure」. - Notes: Start with the home, because you spend more time there, and pooled studies show that second-hand smoke at home does more harm than second-hand smoke outside. Children carry 61% of the health loss caused by second-hand smoke. For quitting yourself, see item 1 in this section. - ### 3. Do not try to quit on willpower alone, get the medicine first: it can more than double the success rate - Cost: Nicotine patches and chewing gum are over-the-counter, and you can buy them at a pharmacy counter. One course runs 8 to 12 weeks and costs a few hundred yuan to a little over a thousand. Varenicline and bupropion are prescription drugs, so you need a smoking cessation clinic or a respiratory medicine department to prescribe them. The money roughly cancels out against the cigarettes you stop buying over the same period. @@ -36,7 +28,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Livingstone-Banks J, Fanshawe TR, Thomas KH, et al. (2023). Nicotine receptor partial agonists for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD006103. ; Hartmann-Boyce J, Chepkin SC, Ye W, Bullen C, Lancaster T (2018). Nicotine replacement therapy versus control for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD000146. ; Theodoulou A, Chepkin SC, Ye W, et al. (2023). Different doses, durations and modes of delivery of nicotine replacement therapy for smoking cessation. Cochrane Database of Systematic Reviews, 6, CD013308. ; Shanghai Municipal Health Commission (上海市卫生健康委员会) (2021). Choosing the right drug makes quitting easier, 选对药物，让戒烟轻松一点: 「市场上可见的戒烟药物主要有三种，即尼古丁替代疗法药物、安非他酮、伐尼克兰」 — the smoking cessation drugs on sale come in three kinds, namely nicotine replacement therapy drugs, bupropion and varenicline. 「尼古丁替代疗法药物属于非处方药（OTC），可通过药店柜台购买；而安非他酮、伐尼克兰属于处方药，须到医院戒烟门诊或呼吸内科就诊，凭医师处方经药师调配后才能得到」 — nicotine replacement therapy drugs are over-the-counter and can be bought at a pharmacy counter, while bupropion and varenicline are prescription drugs, which you can only get by visiting a smoking cessation clinic or a respiratory medicine department at a hospital, on a doctor's prescription, dispensed by a pharmacist. - Notes: The cessation drugs sold in China are the three kinds above, and only the nicotine substitutes are over-the-counter. The common reactions to varenicline are nausea, vivid dreams and poor sleep, so tell your doctor if you have a psychiatric history. A course of nicotine replacement usually runs 8 to 12 weeks, so do not stop suddenly on your own; taper as the doctor says. The medicine only handles the discomfort of the first few weeks of withdrawal. It does not handle the situations where you want to smoke, so use it together with item 4 in this section (set a quit date) and item 5 in this section (go to a smoking cessation clinic). - ### 4. Set a quit date and stop in one go on that day, do not taper first - Cost: No money. You pick one day on the calendar and tell your family and colleagues in advance. @@ -54,7 +45,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Stead LF, Koilpillai P, Fanshawe TR, Lancaster T (2016). Combined pharmacotherapy and behavioural interventions for smoking cessation. Cochrane Database of Systematic Reviews, 3, CD008286. ; Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J (2019). Telephone counselling for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD002850. ; China CDC (中国疾病预防控制中心) (2021). The 'China Smoking Cessation Platform' WeChat mini programme is officially live: 「在健康中国行动控烟行动工作组指导下中国疾病预防控制中心和世界卫生组织驻华代表处联合制作了中国权威戒烟资源库『中国戒烟平台』微信小程序。小程序内容主要包括：戒烟热线、戒烟门诊详细信息，以及线上戒烟服务资源等」 — under the guidance of the tobacco control working group of the Healthy China Action, the China CDC and the WHO Representative Office in China jointly produced the 'China Smoking Cessation Platform' WeChat mini programme, an authoritative library of smoking cessation resources for China. The mini programme mainly carries quitlines, details of cessation clinics, and online cessation services. The footer of the same page gives 「健康咨询电话：12320」 — health consultation number: 12320. - Notes: How to find one in China: search for the 'China Smoking Cessation Platform' mini programme on WeChat, which lists quitlines and cessation clinics by region. You can also call 12320 and ask whether there is a cessation clinic near you. Most of the trials above were done abroad, and the strength of the service varies a lot between clinics in different places, so read the numbers as direction only. For the medicine to go with it, see item 3 in this section (cessation drugs). - ### 6. Consider e-cigarettes only if you have already failed to quit, and people who do not smoke should stay away - Cost: The device and the pods cost from a few tens to a few hundred yuan a month. In China you can only buy tobacco flavours, and only in shops that hold a tobacco monopoly retail licence. @@ -63,7 +53,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Lindson N, Livingstone-Banks J, Butler AR, et al. (2026). Electronic cigarettes for smoking cessation. Cochrane Database of Systematic Reviews, 8, CD010216. ; State Tobacco Monopoly Administration (国家烟草专卖局) (2022). Measures for the Administration of E-Cigarettes, 电子烟管理办法（国家烟草专卖局公告 2022 年第 1 号，2022 年 5 月 1 日施行）. 第十八条: anyone doing e-cigarette retail must apply to the tobacco monopoly administration for a tobacco monopoly retail licence. 第二十二条: 「禁止向未成年人出售电子烟产品。」 — selling e-cigarette products to minors is prohibited. 第二十三条第二款: 「任何个人、法人或者其他组织不得通过本办法规定的电子烟交易管理平台以外的信息网络销售电子烟产品、雾化物和电子烟用烟碱等。」 — no individual, legal person or other organisation may sell e-cigarette products, e-liquids or nicotine for e-cigarettes over any information network other than the e-cigarette trading platform these Measures provide for. 第二十六条: 「禁止销售除烟草口味外的调味电子烟和可自行添加雾化物的电子烟。」 — the sale of flavoured e-cigarettes other than in the tobacco flavour, and of e-cigarettes you can add e-liquid to yourself, is prohibited. - Notes:Disputed. The World Health Organization does not recommend e-cigarettes as a smoking cessation method. The reason is that long-term safety data is insufficient, and they get young people started on nicotine. Britain has put them into its cessation services instead. China does not ban e-cigarette sales outright, but it regulates them tightly, and the rules are written into the Measures for the Administration of E-Cigarettes. Only tobacco flavour may be sold; fruit flavours and e-cigarettes you can add e-liquid to yourself may not. You can only buy in a physical shop that holds a tobacco monopoly retail licence. Selling on any other website, online shop or social app outside the national e-cigarette trading platform is illegal, and selling to minors is prohibited. So any fruit-flavoured e-cigarette still on sale is coming through an illegal channel and sits outside the regulatory system. That is how the 'heady' e-cigarettes spiked with synthetic cannabinoids got out, see item 4 in section 22 (do not take a pod someone hands you). The products tested in the review are all regulated nicotine products; e-liquid of unknown origin is not covered. In order, try item 3 in this section (cessation drugs) first, because the evidence there is harder and it is cheaper. - ### 7. Stop drinking sugary drinks, and switching to sugar-free ones is not a solution - Cost: No money, and it takes no time. Switching to plain water or unsweetened tea even saves money. The hard part is dropping the habit of drinking sweet drinks, sugar-free ones included. @@ -72,7 +61,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Malik VS et al. (2019). Long-Term Consumption of Sugar-Sweetened and Artificially Sweetened Beverages and Risk of Mortality in US Adults. Circulation. ; Mullee A, Romaguera D, Pearson-Stuttard J, et al. (2019). Association Between Soft Drink Consumption and Mortality in 10 European Countries. JAMA Internal Medicine, 179(11), 1479-1490. - Notes:Disputed, and the dispute is about sugar-free drinks. The association with artificially sweetened drinks is stronger than the association with sugary ones, and the cause is probably running the other way: people switch to the sugar-free versions only after they are already overweight or have diabetes. So do not conclude from this that sweeteners are more toxic. It does at least show that the claim 'switching to sugar-free means you are fine' has no evidence behind it. Switching to plain water or unsweetened tea is the safest option. Also, these studies only followed records and did not run a controlled trial, and people who drink a lot of sweet drinks already have worse lifestyles, so the numbers may be too high. - ### 8. Do not chew betel quid - Cost: No money, and it takes no time. You also save the money you spent on betel quid. The hard part is giving up the habit of chewing. @@ -81,7 +69,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Yamada T, Hara K, Kadowaki T (2013). Chewing betel quid and the risk of metabolic disease, cardiovascular disease, and all-cause mortality: a meta-analysis. PLoS One, 8(8), e70679. - Notes: The longer you have chewed and the more you chew each day, the higher the risk of oral cancer. After you quit, the risk falls over time. Betel quid containing tobacco is more dangerous - ### 9. Replace the salt in your home with low-sodium salt (potassium salt) - Cost: A bag costs a few yuan more than ordinary salt. Switch when you next buy salt and it takes no extra time. The taste barely changes. @@ -99,7 +86,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: B - Sources: Wang K, Matsuyama Y, Kiuchi S, et al. (2026). Routine oral health practices and all-cause mortality. Journal of Dentistry. ; Ko MJ, Seo S, So JS, et al. (2026). Deteriorated oral health and function as risk factors for physical disability and mortality in community-dwelling older adults: a systematic review and meta-analysis. European Geriatric Medicine. - Notes:Disputed. These studies only followed records. People with bad teeth often have worse health and worse finances across the board. Whether bad teeth drag the body down, or poor health drags the teeth down, is not clear. But this costs almost nothing, and periodontal disease and missing teeth affect eating on their own - ### 11. Walk 7000–8000 steps a day - Cost: No money. Walking for 60–90 minutes a day can be split into the commute and the shopping trip. The hard part is doing it every day. @@ -108,7 +94,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Paluch AE et al. (2022). Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. Lancet Public Health. ; Banach M et al. (2023). The association between daily step count and all-cause and cardiovascular mortality: a meta-analysis. European Journal of Preventive Cardiology. - Notes: The stretch from 4000 steps to 7000 steps is the best value, and you do not have to reach ten thousand. Note that these studies only followed records: the people who walked least include some who were too ill to walk at all. They walked little because they were about to die, not the other way round. So the gap these studies report is often too large. The direction is clear all the same. This item and item 14 (150–300 minutes of moderate exercise a week) are the same thing measured two ways, and hitting either one is enough. - ### 12. If you have high blood pressure or high blood lipids, take the medicine as prescribed and do not stop on your own - Cost: Blood pressure pills and statins bought through the volume-based procurement scheme cost from a few yuan to a few tens of yuan a month. One dose a day takes very little time. The hard part is carrying on with them when you feel fine and think you no longer need them. @@ -117,7 +102,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Ettehad D et al. (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Cholesterol Treatment Trialists' (CTT) Collaboration (2010). Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170 000 participants in 26 randomised trials. Lancet. ; Chowdhury R et al. (2013). Adherence to cardiovascular therapy: a meta-analysis of prevalence and clinical consequences. European Heart Journal. - Notes: Only do this if the doctor has judged that you need the medicine. Healthy people do not. The figures for the group that keeps taking the medicine come from studies that only followed records, and people who manage to keep taking it are more disciplined to begin with, so the benefit is overstated. The figures from the two randomised trial pools above are more reliable. This section has no separate item on blood glucose: the evidence that pushing glucose lower lowers all-cause mortality is not as consistent as it is for blood pressure and blood lipids. - ### 13. Sleep about 7 hours a night and keep a fixed routine - Cost: No money. You have to set aside the time to sleep, and for most people that means giving up some phone time. The hard part is fixing your bedtime and your wake-up time. @@ -126,7 +110,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Cappuccio FP et al. (2010). Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. ; Yin J et al. (2017). Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. JAHA. ; Windred DP et al. (2024). Sleep regularity is a stronger predictor of mortality risk than sleep duration: A prospective cohort study. Sleep. ; Kumar N, Krishnamurthy S (2026). Social jet lag is associated with incident cardiovascular disease independent of sleep duration and cardiac genetic risk. Journal of Internal Medicine. - Notes: What needs changing is sleeping too little and having an irregular routine. There is no need to deliberately cut your sleep short. Long sleepers having a higher risk is most likely the cause running the other way: depression, chronic illness and sleep apnoea make people sleep more. Regularity of routine for now rests on two follow-up studies, both only followed records with no control group, so on its own it is grade B evidence. How to make up for a night up late is in item 39 in this section (sleep earlier the next night). - ### 14. Build up 150–300 minutes of moderate exercise a week, and brisk walking counts - Cost: No money. 20–45 minutes a day. The hard part is keeping it up over the long run. @@ -144,7 +127,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Oja P, Kelly P, Pedisic Z, et al. (2017). Associations of specific types of sports and exercise with all-cause and cardiovascular-disease mortality: a cohort study of 80 306 British adults. British Journal of Sports Medicine, 51(10), 812-817. - Notes:Disputed. These studies only followed records. People who take up a sport are healthier to begin with and have more social contact. The fact that running showed no difference also suggests that what matters may be which kind of person picks which sport. Do not read this item as saying running is useless, because this section has other items that go by total activity. - ### 16. Add up the short bursts of effort, such as stairs and walking fast, to four or five minutes a day - Cost: No money, and no need to carve out extra exercise time. The hard part is remembering to take the stairs and to walk faster instead of taking the easy way. @@ -153,7 +135,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Stamatakis E, Ahmadi MN, Gill JMR, et al. (2022). Association of wearable device-measured vigorous intermittent lifestyle physical activity with mortality. Nature Medicine, 28, 2521-2529. - Notes:Disputed. This study measured activity with wearable devices, which is more reliable than asking people to fill in a questionnaire from memory. But it still only followed records, it did not run a controlled trial, and it tracked people for 6.9 years. The subjects were people who did no exercise at all, so this item does not apply to people who already exercise regularly - ### 17. Do 30–60 minutes of strength training a week - Cost: Bodyweight squats and press-ups cost nothing. Train 1–2 times a week, 20–30 minutes each time. The hard part is fitting it into the week's schedule. @@ -162,7 +143,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Momma H et al. (2022). Muscle-strengthening activities are associated with lower risk and mortality in major non-communicable diseases: a systematic review and meta-analysis of cohort studies. British Journal of Sports Medicine. - Notes: These studies only followed records, and the activity level was self-reported. The evidence that training too much makes matters worse is very weak, so there is no reason to hold back on that account. Strength training also helps older people avoid falls and keep their muscle, and those points are in section 1. - ### 18. Do not sit for long stretches, get up and move from time to time - Cost: No money, and getting up a few times takes very little time. The hard part is that once you sit down you lose track of time, so set a reminder. @@ -171,7 +151,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Diaz KM et al. (2017). Patterns of Sedentary Behavior and Mortality in U.S. Middle-Aged and Older Adults: A National Cohort Study. Annals of Internal Medicine. ; Ekelund U et al. (2016). Does physical activity attenuate, or even eliminate, the detrimental association of sitting time with mortality? A harmonised meta-analysis of data from more than 1 million men and women. Lancet. - Notes: The reliable conclusion is that how much you move matters more than how long you sit, and that when activity is enough the risk from sitting largely disappears. Note that these studies only followed records, and the people who sit the most include a large number who were already ill, so a risk as high as 2.63 times clearly contains an element of "became sedentary after falling ill". In the same pooled analysis, the risk of watching more than 3 hours of television a day is present at every activity level, and the most active group only pushes that starting point out to 5 hours. - ### 19. Eat less processed meat (ham, bacon, sausage, luncheon meat) - Cost: No money, and it may even save money, and it takes no time. The hard part is cutting back on the mouthful you actually like. @@ -189,7 +168,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Wood AM et al. (2018). Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. ; GBD 2016 Alcohol Collaborators (2018). Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. ; Zhao J et al. (2023). Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Network Open. ; Di Castelnuovo A et al. (2006). Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Archives of Internal Medicine. (the other side of the dispute) - Notes:Disputed. The side that argues drinking less is good points to a pooled analysis of 34 long-term follow-up studies. It reports that light drinkers have an all-cause mortality up to 17%–18% lower. The amount it gives is no more than 4 drinks a day for men and 2 for women. Within that amount, the more people drink the lower their mortality. The other side argues that the "non-drinkers" group includes people who gave up because of illness and people who were in poor health to begin with. One pooled analysis adjusted specifically for those confounders, and once that was done the protective effect disappeared. The Global Burden of Disease study also puts 0 as the safest. The safe way to read it: drinking less will not clearly shorten your life, and drinking more certainly will. Starting to drink for health reasons has no basis. What to do if you want to drink less is in item 22 in this section (count the drinks first, then see a doctor). People who drink every day should not stop abruptly on their own, see item 21 in this section (do not stop abruptly on your own). - ### 21. If you drink every day and shake or feel anxious when you stop, do not stop abruptly on your own - Cost: The registration fee for psychiatry or an addiction medicine clinic runs from a few yuan to a few tens of yuan. Severe cases need a few days in hospital. @@ -198,7 +176,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: B - Sources: Caputo F, Lungaro L, Costanzini A, De Giorgio R, Addolorato G (2026). Alcohol withdrawal syndrome in hospitalized patients: a practical review. European Journal of Internal Medicine, 107103. ; Bramness JG, Heiberg IH, Høye A, Rossow I (2023). Mortality and alcohol-related morbidity in patients with delirium tremens, alcohol withdrawal state or alcohol dependence in Norway: A register-based prospective cohort study. Addiction, 118(12), 2352–2359. - Notes: How to judge it yourself: if you drink almost every day, and missing a day brings shaking, sweating, anxiety and sleeplessness, and you need a drink in the morning to feel right, then any one of those means you should not stop on your own. What hospitals normally do is give benzodiazepines to get you through those few days and add thiamine (vitamin B1) at the same time. The grade is B because the Norwegian cohort says what happened later to people who had delirium tremens, not what happens to people who stop abruptly on their own, and getting from one to the other is an inference. You still need to quit, but quit in hospital instead. How much a week counts as too much is in item 20 in this section (drink less or none), and what to do if you want to drink less is in item 22 in this section (count the drinks first, then see a doctor). - ### 22. To drink less, first count how much you drink in a week, then talk to a doctor for a few minutes - Cost: No money. Counting the drinks takes one line a day. Asking the question while you are at the clinic costs nothing extra. @@ -207,7 +184,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Kaner EF, Beyer FR, Muirhead C, et al. (2018). Effectiveness of brief alcohol interventions in primary care populations. Cochrane Database of Systematic Reviews, 2, CD004148. ; Jonas DE, Amick HR, Feltner C, et al. (2014). Pharmacotherapy for adults with alcohol use disorders in outpatient settings: a systematic review and meta-analysis. JAMA, 311(18), 1889–1900. - Notes: The benefit is graded "small" because both of these bodies of evidence measure how much people drink, not mortality, and by the rule of this book a surrogate endpoint alone gets "small". Most of the drug trials were done abroad, most participants already met the diagnostic criteria for alcohol dependence, and every trial also gave psychosocial support, so the figures are the extra benefit on top of that. Whether these two drugs are approved in China and whether you can be prescribed them is up to your doctor and the drug's own package information, so do not buy them yourself online. If you already have withdrawal symptoms, see item 21 in this section (do not stop abruptly on your own). - ### 23. Eat a small handful of nuts every day - Cost: 28 grams a day costs two or three hundred yuan a year. You eat them in passing, and it takes no time. @@ -216,7 +192,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Bao Y, Han J, Hu FB, et al. (2013). Association of nut consumption with total and cause-specific mortality. New England Journal of Medicine, 369(21), 2001-2011. - Notes: Buy them plain, with no added salt. Nuts are not low in calories, so do not treat them as a snack to eat freely. Note that these studies only followed a group of people over the long term, and nut eaters have better lifestyles across the board, so the figures need discounting. - ### 24. Replace some of your red meat with fish and poultry - Cost: No money, and it takes no time. This is swapping one kind of meat for another, not buying more. The hard part is shifting what you prefer. @@ -234,7 +209,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Aune D et al. (2016). Whole grain consumption and risk of cardiovascular disease, cancer, and all cause and cause specific mortality: systematic review and dose-response meta-analysis of prospective studies. BMJ. ; Schwingshackl L et al. (2017). Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. American Journal of Clinical Nutrition. - Notes: You do not have to switch everything to whole grains, because half of it already puts you in the stretch where the benefit is most visible. Note that these studies only followed records, and people who eat whole grains were healthier to begin with, so the figures are too large. Results also differ a lot between studies (I² 83%, and the higher that number, the less the studies agree). - ### 26. Drink tea three or more times a week - Cost: A few tens to a few hundred yuan a year. Making a cup takes a few minutes and takes very little time. @@ -243,7 +217,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Wang X, Liu F, Li J, et al. (2020). Tea consumption and the risk of atherosclerotic cardiovascular disease and all-cause mortality: The China-PAR project. European Journal of Preventive Cardiology, 27(18), 1956-1963. ; mycotoxins in tea: Cui P et al. (2020). Quantitative analysis and dietary risk assessment of aflatoxins in Chinese post-fermented dark tea. Food and Chemical Toxicology. ; Zhou H et al. (2022). Mycotoxins in Tea (Camellia sinensis (L.) Kuntze): Contamination and Dietary Exposure Profiling in the Chinese Population. Toxins. ; green tea extract and liver damage: Hu J et al. (2018). The safety of green tea and green tea extract consumption in adults - Results of a systematic review. Regulatory Toxicology and Pharmacology. - Notes:Disputed, and the reason is this: these studies only followed records, and Chinese men who drink tea also have a high rate of smoking and drinking. The authors have already subtracted part of that effect, but not all of it. One more thing: do not drink it scalding hot, and for the temperature of hot drinks see the item in this section on that. Some people worry about mycotoxins in tea. In China, 158 samples of dark tea have been tested and 2 of them came back with aflatoxin, and worked out against the amount of tea people drink, the intake is below the internationally acceptable level. Another study tested 352 samples of tea for 16 mycotoxins, and only ochratoxin A in dark tea came out above the limit on average. Worked out against how much tea people in China drink, neither study found a dietary risk. Keep tea somewhere dry, and do not drink it if it has got damp and gone mouldy. Green tea extract capsules are a different thing: a large dose of concentrated catechins swallowed at once can damage the liver, and brewed tea does not have that problem. So this is about brewed tea only, and do not substitute a green tea extract supplement for it. - ### 27. Drink three to four cups of coffee a day, with no sugar and no cream topping - Cost: Brewing it yourself costs a yuan or two a day. Making a cup takes a few minutes and very little of your time. @@ -252,7 +225,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Poole R, Kennedy OJ, Roderick P, et al. (2017). Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ, 359, j5024. - Notes:Disputed, and the reason is this: the original authors state that the evidence comes almost entirely from follow-up records, and they also write that 「需要可靠的随机对照试验才能判断是否为因果」 — reliable randomised controlled trials are needed to tell whether the relationship is causal. In plain terms, people have to be randomly split into two groups and compared, because records alone cannot establish cause. Drink it black, because sugar and a cream topping cancel the benefit out. This does not apply to pregnant women, or to people with arrhythmia, anxiety or insomnia. - ### 28. Eat 5 servings of fruit and vegetables a day (about 400 g) - Cost: A few yuan to a dozen-odd yuan a day. Washing and cutting takes some time each day. The hard part is buying every day and eating enough of it every day. @@ -261,7 +233,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Aune D et al. (2017). Fruit and vegetable intake and the risk of cardiovascular disease, total cancer and all-cause mortality: a systematic review and dose-response meta-analysis of prospective studies. International Journal of Epidemiology. ; Wang DD et al. (2021). Fruit and Vegetable Intake and Mortality: Results From 2 Prospective Cohort Studies of US Men and Women and a Meta-Analysis of 26 Cohort Studies. Circulation. - Notes: 5 servings is enough, there is no need to reach 10, and the two studies agree on that point. Note that these studies only followed records and the confounding is obvious: people who eat more vegetables differ from other people in income, education and exercise. So "about 10% lower risk" is already the ceiling. - ### 29. Eat less ultra-processed food (crisps, instant noodles, cakes, instant meals) - Cost: No money, you only have to shop around. Either you cook yourself, or you spend time picking out food that has not been much processed, and either way it takes time every day. The hard part is fighting the habit that convenience is better. @@ -270,7 +241,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Lane MM et al. (2024). Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. - Notes:Disputed on three points. First, this classification (NOVA) lumps together foods that differ a lot in nutrition. Second, it overlaps heavily with sugary drinks and processed meat (items 7 and 19), so it is not possible to tell which of them is doing the work. Third, judged by the standard evidence grading system (GRADE), the grade is low as well. So once you have done those two items, how much extra this one adds is anyone's guess. The other side's evidence is the low rating in that same review, because no original study has yet reached the opposite conclusion. So do items 7 and 19 first (stop drinking sugary drinks, eat less processed meat), then come back to this one. - ### 30. Cook and heat with electricity or gas instead of coal and firewood - Cost: A rural household has to replace the stove and add the fuel bill, which comes to a few hundred yuan a year and sometimes over a thousand. It is a one-off change and does not take time in the long run. City dwellers are basically already on electricity or gas. @@ -288,7 +258,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: A - Sources: Islami F, Pourshams A, Nasrollahzadeh D, et al. (2009). Tea drinking habits and oesophageal cancer in a high risk area in northern Iran: population based case-control study. BMJ, 338, b929. ; Loomis D, Guyton KZ, Grosse Y, et al. (2016). Carcinogenicity of drinking coffee, mate, and very hot beverages. Lancet Oncology, 17(7), 877-878. - Notes: The International Agency for Research on Cancer classifies hot drinks above 65 ℃ as group 2A, which means probably carcinogenic. The same assessment says coffee itself is not carcinogenic. In China, oesophageal cancer is common in the Chaoshan area of Guangdong and around the Taihang Mountains, and that is linked to the habit of drinking while it is still scalding - ### 32. Get out in the daylight, and do not avoid the sun all day - Cost: No money. A dozen minutes a day, on the way to work and at the lunch break, is enough. @@ -297,7 +266,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: B - Sources: Lindqvist PG, Epstein E, Nielsen K, et al. (2016). Avoidance of sun exposure as a risk factor for major causes of death: a competing risk analysis of the Melanoma in Southern Sweden cohort. Journal of Internal Medicine, 280(4), 375-387. - Notes:Disputed, and the reason is this: Sweden lies at a high latitude with little sunshine, so the conclusion cannot be carried straight over to China. And people who avoid the sun may have been frail and housebound to begin with. Also, do not go as far as getting sunburnt. You still need to cover up under the strong midday sun, because sun exposure raises the risk of skin cancer. - ### 33. Keep your BMI between 20 and 25, and lose weight if you are overweight - Cost: No money. Diet and exercise both take time every day. This is the item in the whole section that demands the most willpower: controlling what you eat and moving your legs over the long run, and keeping the weight off afterwards is harder than losing it. @@ -314,7 +282,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: B - Sources: Bonaccio M, Di Castelnuovo A, Costanzo S, et al. (2019). Chili Pepper Consumption and Mortality in Italian Adults. Journal of the American College of Cardiology, 74(25), 3139-3149. - Notes:Disputed, and the reason is this: this is a follow-up record study done in the context of a Mediterranean diet, and people who eat hot food may have an overall diet closer to the local tradition. Note too that people with reflux, haemorrhoids or irritable bowel will find that hot food makes the symptoms worse, so there is no need to force yourself. - ### 35. Have one or two servings of milk or yoghurt a day - Cost: A few yuan a day, and it takes no time. @@ -323,7 +290,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: B - Sources: Dehghan M, Mente A, Rangarajan S, et al. (2018). Association of dairy intake with cardiovascular disease and mortality in 21 countries from five continents (PURE): a prospective cohort study. Lancet, 392(10161), 2288-2297. - Notes:Disputed, and the reason is this: PURE included a large number of middle-income and low-income countries, and in those countries being able to drink milk every day already says the family has some money, and that confounding is hard to remove cleanly. People who cannot digest lactose can switch to yoghurt or low-lactose milk instead, and there is no need to force themselves to drink ordinary milk. - ### 36. You do not need to give up eggs, but do not eat three or four a day - Cost: No money, and it takes no time. @@ -332,7 +298,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: B - Sources: Zhuang P, Wu F, Mao L, et al. (2021). Egg and cholesterol consumption and mortality from cardiovascular and different causes in the United States: A population-based cohort study. PLoS Medicine, 18(2), e1003508. - Notes:Disputed, and the dispute is a large one: there are also long-term follow-up studies of large groups of people, and pooled analyses of several studies, that found no link between eggs and mortality. The US dietary guidelines have already dropped the daily limit on cholesterol. What needs avoiding is only three or four eggs a day, and one a day is nothing to worry about. - ### 37. Take a bath if you can, and do not only use a shower - Cost: A bit more on water and electricity. Half an hour each time. @@ -350,7 +315,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: B - Sources: Du P, Li J, Hua Z, et al. (2026). Multiple Health Outcomes of Daytime Napping: A Comprehensive Umbrella Review. Public Health Reviews. ; Gao C, Cai R, Zheng X, et al. (2026). Objectively Measured Daytime Napping Patterns and All-Cause Mortality in Older Adults. JAMA Network Open. ; Dashti HS et al. (2021). Genetic determinants of daytime napping and effects on cardiometabolic health. Nature Communications. - Notes:Disputed. The two studies above both only followed records, with no control group. People who nap for a long time are more likely to have sleep apnoea, anaemia, an underactive thyroid, depression or some other chronic illness to begin with. So how much of the higher death rate among long sleepers is caused by the nap itself is not clear. Causal inference using genes (Mendelian randomisation, 453,000 plus 541,000 people) supports only one thing: the more often you nap, the slightly higher your blood pressure and waist. It did not find a causal effect of napping on coronary heart disease or diabetes. So do not treat "nap less" as a way to lower your death rate. The "find out why" half of this item is the author's advice and is grade C: if you are sleepy in the daytime to the point where you must sleep for a long time, check your night-time sleep first, and check for sleep apnoea if you snore or stop breathing during sleep, then check a full blood count and your thyroid. How to get more out of a short nap is in item 11 in section 3 (sleep 10 minutes when you feel sleepy in the afternoon). How long to sleep at night is in item 13 in this section (about 7 hours a night). - ### 39. Make up for a night up late the next night, do not save it up for the weekend - Cost: No money. The hour or two you make up is time you wanted to sleep anyway. @@ -359,7 +323,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: B - Sources: Li X, Zhang M, Li Z, et al. (2026). Acute sleep rebound following sleep restriction is associated with reduced mortality risk. Nature Communications. - Notes: This study only followed records, with no control group. People who cannot make up their sleep may have been busier and in worse health to begin with, and that layer cannot be removed cleanly. "Neither of the two groups that made it up showed a rise in risk" only means no difference was found: their point estimates are still above 1, and making it up does not mean nothing happened. So pay back a night up late as soon as you can, and do not use that as a licence to stay up late. The make-up in this study means sleeping a little extra on the first night after a short night, and it does not include paying everything back in one go at the weekend. Turning "stay up late on weekdays, catch up at the weekend" into a weekly habit is called social jet lag, and it is itself linked to cardiovascular disease, see item 13 in this section (about 7 hours a night and a fixed routine). The rule about a fixed wake-up time at the weekend as well is in item 2 in section 3 (fixed wake-up time, at the weekend too), and it does not clash with this item: this one tells you to go to bed early that night, not to sleep until midday the next day. - ### 40. The longer you work night shifts, the higher your cardiovascular risk, so change to a day post as early as you can - Cost: Moving post or changing job may mean less money, and you lose the night shift allowance. Job hunting takes time and the decision is hard. If you do this arithmetic while you are choosing a job, the cost is zero. @@ -377,7 +340,6 @@ This section takes in only two kinds of chronic risk: those that move all-cause - Evidence grade: B - Sources: Lei J, Li Y, Wang Y, et al. (2024). The impact of small food workshops management regulations on aflatoxin B1 in home-made peanut oil and the liver function of high-consumption area residents: an interrupted time series study in Guangzhou, China. Frontiers in Public Health, 12. ; Zhong Y, Lu H, Jiang Y, et al. (2024). Effect of homemade peanut oil consumption during pregnancy on low birth weight and preterm birth outcomes: a cohort study in Southwestern China. Global Health Action, 17. ; IARC (2012). Chemical Agents and Related Occupations. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol 100F. ; National Health and Family Planning Commission (国家卫生计生委), State Food and Drug Administration (国家食品药品监督管理总局) (2017). National Food Safety Standard, Limits for Mycotoxins in Food (GB 2761-2017) - Notes: Both studies only followed records and had no control group. The endpoint in the Guangzhou study is a liver function test, not liver cancer or death. The median toxin level in home-pressed oil is in fact below the national limit for peanut oil of 20 μg/kg, and the risk comes from eating it over many years and from occasional high batches. The national limit applies to all peanut oil, but prepackaged oil has to pass production licensing and a factory release test, while small workshops are governed by the management rules of each province and are sampled far less often. The people who benefit most from this item are you and the family you eat with. - ### 42. Cook with vegetable oil instead of lard and butter, you do not have to shop around between oils for health reasons, and do not expect flaxseed oil to protect your heart - Cost: It does not cost much more. Vegetable oil is already the main cooking fat in most households. The hard part is dropping the habit of lard for frying and for putting over rice. diff --git "a/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" "b/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" new file mode 100644 index 00000000..d4ea82c0 --- /dev/null +++ "b/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" @@ -0,0 +1,210 @@ +[← 回总目录](../README.md) + +# 3. 不要浪费精力 + +本节只算精力和时间账。看的是四样东西：你每天能用的注意力、反应快不快、出错多不多、被打断以后要多久才回得来。所有数字都是直接测出来的，来自实验，或者来自跟踪一群人做的记录（队列研究）。这些数字不换算成寿命，也不换算成死亡率。本节多数条目的证据等级是 B。这个领域本来就这样，不代表建议不成立。第 20 条讲的是和警察、医生、柜员打交道时该有的预期，引的是法规，不是实验，所以等级记 C。第 15 条（把悲观念头当症状看）和第 23 条（把「别人要求我完美」当症状看）引的是死亡率与自杀风险的数字，口径和本节其他条目不一样，不放在一起比较。 + +### 1. Turn off notifications you do not need, and keep the phone out of sight while you work + +- Cost: No money. Changing the settings takes a few minutes and needs almost no willpower. +- In plain language: When the phone buzzes, your score on an attention task drops even if you neither look at it nor touch it. The drop is about the same as when you actually answer a call or send a text. There is a second case: the phone is simply lying on the desk, you resist looking at it, and the thinking capacity left to you still goes down. The more dependent you are on your phone, the larger the drop. +- Benefit: Merely receiving a notification, without looking at it or touching it, clearly worsens performance on an attention task. The worsening is on a par with actually taking a call or sending a text. This was done in a laboratory, with university students as participants. Two other experiments show that even when people successfully resist looking at the phone, having their own phone next to them still reduces the thinking capacity available to them. People with the strongest phone dependence dropped the most. +- Evidence grade: B +- Sources: Stothart, Mitchum & Yehnert (2015). The attentional cost of receiving a cell phone notification. Journal of Experimental Psychology: Human Perception and Performance. ; Ward, Duke, Gneezy & Bos (2017). Brain Drain: The Mere Presence of One's Own Smartphone Reduces Available Cognitive Capacity. Journal of the Association for Consumer Research. +- Notes: All of these are single tasks in a laboratory. Nobody has measured what happens when you add them up day after day. If your job really has to answer at any hour, leave notifications on for a small number of contacts. +### 2. Keep a fixed wake-up time, at weekends too + +- Cost: No money. Lying in less at weekends takes some willpower. +- In plain language: In the most irregular fifth, the body clock runs about 2.5 hours later than in the most regular fifth. That amounts to living in a time zone shift for years. The more regular the schedule, the better the grades. +- Benefit: 61 university students kept records for 30 days. In the most irregular fifth, the body clock (circadian rhythm) ran about 2.5 hours later than in the most regular fifth. Concretely, melatonin started being released at 00:08 against 21:32. The period of peak sleepiness also came about 1.8 hours later. The more regular the schedule, the better the academic performance (correlation coefficient r = 0.37). +- Evidence grade: B +- Sources: Phillips et al. (2017). Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sleep/wake timing. Scientific Reports. +- Notes: There is only this one study, with few participants and no control group. The authors infer that the body clock difference comes mainly from how much daylight each group gets. A fixed wake-up time therefore only works if you also get daylight in the morning. For how to recover after an occasional late night, see item 39 in section 2 (sleep earlier the following night): go to bed early that night, not until midday the next day. +### 3. Sleep 7 to 8 hours a night, and do not treat 6 hours as enough + +- Cost: No money. You have to free 1 to 2 hours a day for sleep. Going to bed earlier takes some willpower. +- In plain language: Two weeks of 6 hours of sleep a night drops brain performance to the level of going two days and two nights without closing your eyes. The trouble is that you cannot feel it. Self-rated sleepiness rises in the first few days and then stops rising. People on 6 hours rate themselves just as sleepy as people on 4 hours. So the sentence "I get by on 6 hours" is itself a symptom of sleep loss. +- Benefit: 48 healthy adults aged 21 to 38 slept 6 hours a night for 14 days in a row. Their cognitive performance fell cumulatively, down to the level of 2 nights with no sleep at all. Self-rated sleepiness rose only in the first few days and then barely rose again. The 6-hour group and the 4-hour group did not differ in self-rated sleepiness. This was a controlled trial that randomly assigned people to two groups (an RCT). Separately, short periods of no sleep (acute sleep deprivation, no more than 48 hours) have an effect size of g = -0.776 on mind-wandering in simple attention tasks and g = -0.125 on reasoning accuracy. That part pools 70 studies. +- Evidence grade: A +- Sources: Van Dongen, Maislin, Mullington & Dinges (2003). The cumulative cost of additional wakefulness. Sleep. ; Lim & Dinges (2010). A meta-analysis of the impact of short-term sleep deprivation on cognitive variables. Psychological Bulletin. +- Notes: "I don't feel tired" is not evidence that you slept enough. People who sleep 6 hours a night for years only feel used to it; their performance has not caught up. The part that pools several studies is about staying awake for a whole night. Only one randomised controlled trial supports "sleeping less over a long period". +### 4. No caffeine after 2 p.m. + +- Cost: No money. If you habitually drink coffee in the afternoon, holding out takes some willpower. +- In plain language: A large cup of coffee (400 mg) six hours before bed cuts measured sleep that night by more than an hour. A self-written sleep diary shows no difference at all. An ordinary cup of coffee (107 mg) has to be finished more than 8.8 hours before bed if it is not to disturb sleep. If you go to bed at eleven, the line falls at two in the afternoon. +- Benefit: Taking 400 mg of caffeine 6 hours before bed, measured total sleep time was 1.1 to 1.2 hours shorter than with a placebo. A placebo here is a control drink with no caffeine in it. The participants' own sleep diaries did not report any difference, so they did not notice it. Pooling several studies (meta-analysis), caffeine cuts total sleep time by 45 minutes on average. The share of the time in bed actually spent asleep (sleep efficiency) falls 7%. A cup of coffee (107 mg) has to be finished more than 8.8 hours before bed if it is not to disturb sleep. +- Evidence grade: A +- Sources: Drake, Roehrs, Shambroom & Roth (2013). Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed. Journal of Clinical Sleep Medicine. ; Gardiner et al. (2023). The effect of caffeine on subsequent sleep: A systematic review and meta-analysis. Sleep Medicine Reviews. +- Notes: The Drake study was run in a single laboratory, and the 400 mg it used is on the high side, roughly two or three cups of coffee. If you metabolise caffeine slowly, your cutoff has to be earlier still. Cutting back on caffeine buys back sleep duration, and that adds to item 3 (sleep 7 to 8 hours a night). +### 5. Handle email and messages in a few fixed batches a day + +- Cost: No money. It takes a little time to tell colleagues how often you will reply. Resisting the urge to check email at any moment takes some willpower. +- In plain language: The same people were compared over two weeks. In the week when email was limited to 3 checks a day, daily stress was clearly lower, while the number of emails sent and received was the same. Without a limit, people checked email 15.5 times a day on average. +- Benefit: 124 adults were compared with themselves before and after over two weeks. In the week when email was limited to 3 checks a day, daily stress was clearly lower than in the week with no limit (effect size Cohen's d = 0.37). The number of emails received and replied to over the two weeks did not differ. At the start, with no limit, people checked email 15.5 times a day on average. +- Evidence grade: B +- Sources: Kushlev & Dunn (2015). Checking email less frequently reduces stress. Computers in Human Behavior. +- Notes: One study only, with a small-to-moderate effect, and it measured stress rather than work output. Nobody has studied instant messaging directly, so that part is an inference by analogy. +### 6. When the work needs continuous thought, block out interruptions of even a few seconds + +- Cost: No money. Close the door, put on headphones, or hang up a do-not-disturb sign. +- In plain language: An interruption averaging 2.8 seconds doubles the chance of an error in the operations that follow. An interruption of 4.4 seconds triples it. Field observation also found that it takes an average of 25 minutes to return to the interrupted task, with two other tasks inserted in between. People who are interrupted work faster, but report more stress, more frustration and more effort. +- Benefit: An interruption averaging 2.8 seconds doubles the error rate on tasks carried out step by step in order (sequential tasks). An interruption of 4.4 seconds triples it. This was done in a laboratory. A separate study observed 24 office workers (knowledge workers) in the field. They switched the subject of their work every 11 minutes on average. 57% of work episodes were interrupted. Returning to the interrupted task the same day took 25 minutes 26 seconds on average. 2.26 other tasks were inserted in between. A third experiment with 48 people found that interrupted people worked faster but reported more stress, more frustration, more time pressure and more effort. +- Evidence grade: B +- Sources: Altmann, Trafton & Hambrick (2014). Momentary interruptions can derail the train of thought. Journal of Experimental Psychology: General. ; Mark, Gonzalez & Harris (2005). No task left behind? Examining the nature of fragmented work. CHI 2005. ; Mark, Gudith & Klocke (2008). The cost of interrupted work: More speed and stress. CHI 2008. +- Notes: That field study split interruptions into two kinds: ones where somebody else comes to you, and ones you open yourself. Closing the door only blocks the first kind. The second has to be handled together with item 1 (turn off notifications you do not need) and item 5 (handle messages in batches). +### 7. Do one thing at a time, and do not answer messages during a meeting + +- Cost: No money. Holding back from answering messages during a meeting takes some willpower. +- In plain language: Every time you switch tasks you react more slowly and make more mistakes. People who habitually keep several screens open at once have not trained any skill at doing several things at the same time. They do worse on task-switching tests instead. +- Benefit: After a task switch, reactions are clearly slower and errors more likely. That part comes from a review of several studies. A separate laboratory study used university students. People who routinely use several kinds of media at once (heavy media multitaskers) are more easily pulled off course by irrelevant things and irrelevant memories. They do worse on task-switching tests. +- Evidence grade: B +- Sources: Monsell (2003). Task switching. Trends in Cognitive Sciences. ; Ophir, Nass & Wagner (2009). Cognitive control in media multitaskers. PNAS. +- Notes: How much extra time switching back and forth costs varies a lot with the kind of work, so no single figure is given here. The Ophir study is a comparison at one point in time, and it cannot separate "multitasking worsens attention" from "people with worse attention multitask more". +### 8. No light-emitting screens in the hour before bed; read paper or an e-ink screen instead + +- Cost: No money. Putting the phone down in the hour before bed means fighting habit over the long run, and that needs willpower. +- In plain language: Reading a light-emitting screen before bed means falling asleep about 10 minutes later than reading paper. Melatonin is suppressed by 55%. The body clock the next day is pushed back by more than 1.5 hours, so the next morning is sleepier and takes longer to become fully awake. +- Benefit: In the laboratory the same people did the two conditions in turn (a crossover trial). Reading a light-emitting e-book before bed took about 10 minutes longer to fall asleep than reading paper (25.7 minutes against 15.8 minutes). Evening melatonin was suppressed by 55%. The body clock (circadian rhythm) the next day was delayed by more than 1.5 hours, with melatonin release starting at 22:31 against 21:01. The next morning is sleepier, and it takes longer to be fully awake. +- Evidence grade: B +- Sources: Chang, Aeschbach, Duffy & Czeisler (2015). Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. PNAS. +- Notes: Few participants, in a laboratory with the conditions well controlled. At everyday phone brightness and everyday usage length the effect may be smaller. How much the phone's night mode helps was not measured in this study. + +### 9. Sleep when bedtime comes; do not stay up for games, short videos or pornography + +- Cost: No money. The hard part is fighting the "one more episode" habit over the long run. Keeping the phone out of the bedroom is the least effort. +- In plain language: In 2025 people in China slept an average of 6.97 hours at night and fell asleep at 0:10 on average. People who use a phone or tablet before bed are more than twice as likely to get too little sleep, and 2.7 times as likely to be sleepy during the day. The figures are about the same when the device is merely present in the bedroom and not used. So the simplest method is to keep the phone out of the bedroom. +- Benefit: In 2025 residents of China slept an average of 6.97 hours at night, and the average time of falling asleep was 0:10 (Chinese Sleep Research Society (中国睡眠研究会)). People who use a portable device before bed are about 2.17 times as likely to get too little sleep (OR 2.17), about 1.46 times as likely to have poor sleep quality (OR 1.46) and about 2.72 times as likely to be sleepy during the day (OR 2.72). With the device merely present in the bedroom and not used, the figures are about the same: 1.79, 1.53 and 2.27 times (OR 1.79/1.53/2.27). That part pools 20 cross-sectional surveys of 125198 children and adolescents. A separate cross-sectional survey: among 844 adults aged 18 to 94, people who still used a phone after lights out took longer to fall asleep, had a lower share of the time in bed actually spent asleep, and were more tired during the day. +- Evidence grade: B +- Sources: Carter et al. (2016). Association Between Portable Screen-Based Media Device Access or Use and Sleep Outcomes. JAMA Pediatrics. ; Exelmans & Van den Bulck (2016). Bedtime mobile phone use and sleep in adults. Social Science & Medicine. ; Chinese Sleep Research Society (中国睡眠研究会) (2026). 2026 China Sleep Health Research White Paper, as reported by Health News and reposted by the Fujian Provincial Health Commission (福建省卫健委). ; Grubbs JB et al. (2018). Pornography Problems Due to Moral Incongruence: An Integrative Model with a Systematic Review and Meta-Analysis. Archives of Sexual Behavior. (one of the two sides in the disputed note below) ; Prause N, Pfaus J (2015). Viewing Sexual Stimuli Associated with Greater Sexual Responsiveness, Not Erectile Dysfunction. Sexual Medicine. (the other side in the disputed note below) +- Notes:Disputed. The cost of staying up late itself is in item 3 in this section (sleep 7 to 8 hours a night). What pornography does to ordinary people's energy and thinking cannot be pinned down by the evidence at present, nor even attributed to it. One analysis pooling several studies argues that what gets called a "pornography problem" comes mostly from a mismatch between a person's beliefs and their own behaviour, not from the material itself. Another study pooled data on 280 men. The result was that hours watched had no relation to erectile function during sex with a partner. This entry is only about the sleep time it takes up. For the two figures, 6.97 hours and 0:10, the Chinese Sleep Research Society has not posted the release on its own website (checked in September 2026), so the Health News report reposted by the Fujian Provincial Health Commission is what this entry relies on. +### 10. No major decisions late at night, and no urgent messages + +- Cost: No money. You only push things until after you have slept. +- In plain language: When you are rested you learn quickly to avoid the high-risk options. After two nights without sleep you pick the high-risk one time after time instead, and the older you are the clearer that becomes. Nobody has measured how much a few hours of lost sleep matters, but push important decisions until after you have slept. +- Benefit: 34 healthy people did the Iowa Gambling Task after 49.5 hours without sleep. It is an experiment used to measure risky decision-making. When they were rested they learned quickly to avoid the high-risk options. After sleep loss they picked the high-risk options more often instead. Older people were affected more. This is one experiment, and there is only this one study. +- Evidence grade: B +- Sources: Killgore, Balkin & Wesensten (2006). Impaired decision making following 49 h of sleep deprivation. Journal of Sleep Research. +- Notes: The study covers the extreme case of two nights with no sleep at all. What happens to an ordinary person who stays up until two in the morning has not been measured directly. The "do not send messages" half is the author's own experience (grade C), and sending them the next day is not too late. +### 11. If you are sleepy in the afternoon, sleep 10 minutes, not half an hour + +- Cost: No money. It takes 10 minutes, plus finding somewhere to lie down or lean. +- In plain language: A 10-minute nap wakes you up on the spot, and the effect lasts two and a half hours. A 5-minute nap does almost nothing. A 20-minute nap only takes effect after half an hour. A 30-minute nap leaves you groggy on waking, and performance in that spell is worse than not sleeping at all. +- Benefit: After an insufficient night, a 10-minute nap immediately improves sleepiness, fatigue, energy and cognitive performance, and part of the benefit lasts past 155 minutes. A 5-minute nap brings almost no benefit. The benefit of a 20-minute nap appears only after 35 minutes. A 30-minute nap is followed by a stretch of sleep inertia: performance is briefly worse just after waking, and improves afterwards. +- Evidence grade: B +- Sources: Brooks & Lack (2006). A brief afternoon nap following nocturnal sleep restriction: Which nap duration is most recuperative? Sleep. +- Notes: One laboratory study only, and all the participants were healthy young people. If you cannot sleep, there is no need to force it. There is no research support for lying down with your eyes closed for 10 minutes. Napping for more than an hour is instead linked to higher mortality and coronary heart disease risk: see item 38 in section 2 (keep the nap under half an hour). +### 12. When you work in an open-plan office, use earplugs or a quiet room for work that needs memory + +- Cost: Earplugs cost a few tens of yuan. Booking a quiet meeting room takes a little time. +- In plain language: When background noise rises from 39 decibels to 51 decibels, fewer words are remembered, people get more tired, and they want to work less. During a break, watching nature footage with sound restores energy better than listening to office noise. +- Benefit: Students worked in a simulated open-plan office. When background noise rose from 39 decibels to 51 decibels (dB, LAeq), they remembered fewer words, rated themselves as more tired, and wanted to work less. During breaks, watching nature footage with sound lifted their energy rating more than listening to river sounds or office noise alone. +- Evidence grade: B +- Sources: Jahncke, Hygge, Halin, Green & Dimberg (2011). Open-plan office noise: Cognitive performance and restoration. Journal of Environmental Psychology. +- Notes: One laboratory study only. The most damaging part of the noise is speech you can make out. White-noise headphones help with that, but this study did not test headphones. +### 13. Do not work more than 49 hours a week + +- Cost: You may earn some overtime pay less. You may also have to spend effort negotiating with your manager. +- In plain language: Records from British armaments factories during the First World War show that within 49 hours a week, output rises in proportion to hours worked. Past 49 hours, each extra hour adds less and less. A week of 70 hours produces almost the same as 56 hours, and the extra 14 hours are wasted. +- Benefit: Output records for workers at British armaments factories in the First World War: within 49 hours a week, output is proportional to hours worked. Past 49 hours the rate of growth in output slows down. The maximum arrives at about 63 hours. Output at 70 hours a week is almost the same as at 56 hours, so the extra 14 hours are wasted. +- Evidence grade: B +- Sources: Pencavel (2015). The Productivity of Working Hours. The Economic Journal. +- Notes: The data comes from factories a hundred years ago doing physical work, and the workers were mostly women. The threshold may sit somewhere else in mental work, and the author himself says other occupations may be higher or lower. What is reliable is the direction: past a certain point, more hours add almost no output. The specific figure of 49 is not reliable. +### 14. When you notice yourself replaying the same bad thing, switch to a job that needs your hands + +- Cost: No money. The hard part is noticing that you are replaying it (this is called rumination), and noticing takes practice. +- In plain language: Replaying the same bad thing in your head does not produce an answer. It worsens depression, fixes negative thoughts more firmly in place, makes problems harder to solve and harder to act on, and wears out the people near you who are willing to listen. When you notice you are going round in circles, go and do something with your hands. +- Benefit: A review of several studies says that replaying the same bad thing in your head (rumination) worsens depression. It also strengthens negative thinking and weakens the ability to solve problems. It gets in the way of actual action, and it wears out the people near you who are willing to listen. It is associated with anxiety, binge eating, heavy drinking and self-harm. +- Evidence grade: B +- Sources: Nolen-Hoeksema, Wisco & Lyubomirsky (2008). Rethinking Rumination. Perspectives on Psychological Science. +- Notes: The review gives direction only, with no numbers. Switching to another activity is the author's own practice. One sentence is all the literature supports: diverting your attention beats going round in circles. Nobody has compared which activity to switch to. +### 15. Treat "things will definitely get worse" thoughts as a symptom, not as a fact + +- Cost: No money. The hard part is treating such a thought as a symptom instead of as a fact. +- In plain language: In a twin study followed for 20 years, each step up in pessimism came with about 13% higher mortality over the same period, and about 20% higher risk of dying from cardiovascular disease. In the other direction, a high optimism score brought no matching benefit. What needs dealing with is pessimism, not "not being positive enough". +- Benefit: 2978 Australian twins over the age of 50, followed for 20 years on average, during which 1068 of them died. Each 1 standard deviation higher on the pessimism scale came with about 13% higher mortality over the same period (risk ratio 1.134). The credible range for that figure is 95% CI 1.065–1.207, P=8.85×10⁻⁵. Cardiovascular mortality was about 20% higher (1.196, 1.045–1.368, P=0.0093). There was no relation to cancer mortality. Optimism scores showed no clear association with mortality either. +- Evidence grade: B +- Sources: Whitfield JB, Zhu G, Landers JG, Martin NG (2020). Pessimism is associated with greater all-cause and cardiovascular mortality, but optimism is not protective. Scientific Reports, 10, 12609. +- Notes:Disputed. This is observation only, with no groups. Some would say that people become pessimistic because they are already ill, and the authors answer that if that were true, cancer mortality would be related as well. Unmeasured factors may still be there. Optimism had no protective effect, so there is no need to force yourself into "positive thinking". What matters is not treating a pessimistic expectation as an accurate forecast of the future. +### 16. Cut down on the relationships that drain you, and learn to refuse requests you do not want + +- Cost: No money. There will be some social friction in the short term, and you may lose some relationships. Saying no takes real willpower. +- In plain language: One person who drains you does more damage, and does it more reliably, than one supportive person brings benefit. So subtract before you add. +- Benefit: A survey of 120 widowed women aged 60 to 89: contact that leaves you feeling worse is more consistently and more strongly linked to psychological well-being than contact that feels comfortable. On this result, the damage done by one person who drains you is larger than the benefit brought by one person who supports you. +- Evidence grade: C +- Sources: Rook (1984). The negative side of social interaction: Impact on psychological well-being. Journal of Personality and Social Psychology. +- Notes: The group surveyed is unusual, and it was a one-off questionnaire, so the order of events is not visible. What it measured is well-being, not energy. Placing it in this section is the author's inference. There is no objective standard for which relationships count as "draining", so you have to judge that yourself. + +### 17. Fix decisions of low value, such as what to wear and what to eat + +- Cost: No money. At the start you spend a little time setting the default options. +- In plain language: This entry has no hard evidence behind it. The original experiment on "making choices drains willpower" did not come out in two large replications. One of them involved two thousand people and the other three thousand, with several labs working together, and the effect was basically zero. Fixing what you wear and what to eat is still worth doing, because it saves those few minutes and those few moments of dithering, not because it "saves willpower". +- Benefit: Cannot be quantified. The original experiment claimed that making choices and exercising self-control draw on one shared limited resource (ego depletion). Two large replications have since been run, both preregistered and both carried out by several labs following the original procedure, and neither reproduced the result. One involved 23 labs and 2141 people, with effect size d = 0.04, 95% CI [-0.07, 0.15] (credible range). The other involved 36 labs and 3531 people, d = 0.06. The difference is too small to count. +- Evidence grade: C +- Sources: Baumeister, Bratslavsky, Muraven & Tice (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology. ; Hagger et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science. ; Vohs et al. (2021). A Multisite Preregistered Paradigmatic Test of the Ego-Depletion Effect. Psychological Science. +- Notes:Disputed. Popular books treat "decision fatigue" as settled, but most later replications of the original experiment did not produce the same result. It is still worth doing, because it costs almost nothing and the time it saves is real. Do not expect it to improve your judgement in the afternoon. +### 18. When you are angry, walk away first and treat the other person as weather, not as an enemy + +- Cost: No money. You have to hold back your answer on the spot, turn and walk away, and decide afterwards whether to deal with it. That takes a little practice. +- In plain language: Explaining what just happened from a different angle, or taking the other person's point of view, both bring the emotion down for real, and the second works better. Suppressing it without showing anything does not work. People who punish the offender also ruminate on it for longer afterwards; people who do not retaliate move on faster. People overestimate in advance how good revenge will feel. +- Benefit: Pooling 306 experimental comparisons (a meta-analysis): reinterpreting what happened (cognitive reappraisal) has an effect size on emotion of d = 0.36. Taking the other person's point of view gives d = 0.45. Diverting attention gives d = 0.27. Suppressing the feeling does not work (d = -0.04). Three further experiments show that people who punish the offender ruminate on the person for longer afterwards, while people who do not punish move on faster. People overestimate in advance how good revenge will feel. +- Evidence grade: B +- Sources: Webb TL, Miles E, Sheeran P (2012). Dealing with feeling: A meta-analysis of the effectiveness of strategies derived from the process model of emotion regulation. Psychological Bulletin. ; Carlsmith KM, Wilson TD, Gilbert DT (2008). The paradoxical consequences of revenge. Journal of Personality and Social Psychology. +- Notes: The part pooling several studies measured emotional intensity, not energy, and the effect is small to moderate. The revenge part rests on a single set of experiments, so it is graded B. The empty boat is a parable from the "Mountain Tree" chapter of the Zhuangzi: you are not angry when an empty boat drifts into you, and you are angry when there is a person aboard, yet the collision is the same. That is an analogy, not evidence. If there is real loss, take the legal route in section 8 and 9, which costs far less than losing your temper on the spot. +### 19. When you are low, do the highest-value things first: move, get daylight, sleep on time, talk to someone, call 12356 + +- Cost: No money, or very little. Brisk walks, jogging or strength training several times a week. Go outside into the daylight during the day. Fix your daily routine. Make a phone call when you cannot get out of it. +- In plain language: For depression, brisk walking and jogging work best, and yoga and strength training come next; the more intense the exercise, the larger the effect. Light therapy works strongly on seasonal depression, and also works on depression that is not seasonal. Self-help cognitive behavioural courses online do work, and they work better with someone guiding you when the symptoms are heavier. If you cannot get out of it, call 12356 first. +- Benefit: Pooling 218 randomised controlled trials and 14170 people (a network meta-analysis). The comparison group was people doing other activities. Against them, walking or jogging has an effect size on depression of g = -0.62 (95% confidence interval -0.80 to -0.45). Yoga gives -0.55 and strength training -0.49. The more intense the exercise, the larger the effect. Pooling the light therapy studies: effect size 0.84 for seasonal depression (0.60 to 1.08), and 0.53 for non-seasonal depression (0.18 to 0.89). A further analysis pooled the individual patient data from the studies. It looked at internet cognitive behavioural therapy (online self-help courses). Both guided and unguided beat the comparison group. For moderate to severe cases (PHQ-9 above 9), guided did better. 12356 is the national mental health assistance hotline, open at least 18 hours a day. +- Evidence grade: A +- Sources: Noetel M et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ. ; Golden RN et al. (2005). The efficacy of light therapy in the treatment of mood disorders: a review and meta-analysis of the evidence. American Journal of Psychiatry. ; Karyotaki E et al. (2021). Internet-Based Cognitive Behavioral Therapy for Depression: A Systematic Review and Individual Patient Data Network Meta-analysis. JAMA Psychiatry. ; National Health Commission (国家卫生健康委) (2024). Notice on the use of the "12356" national unified mental health assistance hotline number (国卫医政函〔2024〕259 号). +- Notes: All of these are for mild to moderate depression. For moderate to severe depression, or if you have suicidal thoughts, see a doctor, and call 12356 first (see item 25 in section 1). The exercise trials mostly had few participants, and they cannot blind people to whether they are exercising, the way a drug trial can. The light therapy research only goes as far as 2003, and the samples were small too. "Sleep on time" and "talk to someone" are the author's own experience, with no separate reference. Do not treat "move" as a demand on people with depression; ten minutes of walking counts. +### 20. Treat police, doctors and clerks as people working to rules, not as characters: what moves a case forward is paperwork and deadlines, not emotion + +- Cost: No money. What changes is your expectations and the order of your actions. The first time, you have to resist the urge to argue the case until the other side agrees. +- In plain language: Police, doctors and clerks all work to the rules of their post, and they answer to paperwork, deadlines and supervision. Whether the person is pleasant has almost nothing to do with whether your case gets done, or how many days it takes. So ask for a receipt, note the deadline, and appeal through the proper channel, instead of trying to argue until the other side agrees. +- Benefit: Change the way you handle dealings with institutions: from "beg, get angry, and keep repeating your grievance" to "take the paperwork, note the deadline, appeal through the proper channel". What you save is the time spent on repeated conversations and repeated recollection. This is a one-off adjustment of how you think, not a daily saving of time. What you can actually check are the deadlines themselves, and they are all written into the documents. For public security cases, the time limit for handling runs from the day the case is accepted and is no more than 30 days; where the case is major or complicated, it may be extended by 30 days with the approval of the public security organ at the next level up. Where a case with a complainant is decided not to be opened, the notice of refusal must be served within 3 days. The complainant may apply for reconsideration within 7 days, and the public security organ must decide within 30 days. If still not satisfied, a further 7 days may be used to apply for review by the organ at the next level up. +- Evidence grade: C +- Sources: Ministry of Public Security (公安部) (2020). Provisions on the Procedures for Handling Criminal Cases by Public Security Organs (公安机关办理刑事案件程序规定)（公安部令第 159 号修正后，第一百七十一、一百七十八、一百七十九条）. ; Ministry of Public Security (公安部) (2013/2019 修正). Provisions on the Procedures for Handling Administrative Cases by Public Security Organs (公安机关办理行政案件程序规定)（公安部令第 125 号、第 149 号）. +- Notes: The reason for grading C: the judgement that "many people treat the person behind the counter as a character rather than as someone at work" has no literature to cite. Only the rules half can be cited: the deadlines, the paperwork and the routes of appeal are all in writing. Every figure in this entry comes from the two Ministry of Public Security regulations cited in item 39 in section 8 (what happens after you call the police). Following the procedure will not fix an unpleasant manner, and it does not guarantee a good outcome; what it changes is your expectations and your first move. Bank clerks and government service windows are the same: their operations leave a trail, they need a second pair of eyes, and errors are held against them. So what you should do is keep the slip, ask for the receipt, complain through the proper channel, and not guess on the basis of "will he cheat me". Where the lawful order of priority is written down in the open, follow it. Emergency triage is by severity of illness, not by order of arrival (item 8 in section 24, emergency pre-triage). To thank a doctor who saved your life, use a letter of thanks and the satisfaction survey, see item 12 in section 24. If you want to "give them something so they take more care", to someone handling or enforcing a case that is bribery, and the law makes it an aggravated offence, see item 40 in section 8. If you meet obvious differential treatment, complain about it, and do not treat it as an unwritten rule to copy. + +### 21. Do not treat "how other people are doing" as daily reading: limit or switch off the apps that show people your own age + +- Cost: No money. Setting a time limit on an app, or closing the account, takes a few minutes. The hard part is not opening it afterwards, which means fighting habit over the long run. +- In plain language: People who habitually compare themselves online with people who are doing better tend to be in worse mood and to have lower self-esteem. People who switched Facebook off for four weeks showed a small rise in well-being, and also a drop in what they knew about current affairs. People who limited three social platforms to 10 minutes a day each had lower loneliness and depression after three weeks. +- Benefit: Pooling several studies (a meta-analysis): 54 independent samples, 94 effect sizes, N = 36,583, using a three-level random-effects model. The average correlation between online upward social comparison and psychological maladjustment is r = 0.330 (95% CI 0.289–0.370, credible range). Broken out by outcome: social-evaluative negative emotions such as fear of being looked down on 0.438, anxiety 0.382, depression 0.306, low well-being 0.268, low self-esteem 0.263. Age, cultural background and the year the data were collected do not change the strength of the association. Studies with long-term follow-up do not give stronger results than cross-sectional surveys. A separate randomised controlled trial enrolled 2,743 US Facebook users. 61% of them were willing to accept a switch-off for less than 102 dollars. They were randomly assigned to a four-week switch-off group or a control group. Switching off freed the average participant 60 minutes a day. The subjective well-being index rose by 0.09 standard deviations. Broken out: life satisfaction 0.12, anxiety 0.10, depression 0.09, happiness 0.08. These still hold after the element of chance from multiple comparisons is removed. The author gives a reference point: psychological interventions that make people more positive. Such interventions raise subjective well-being by about +0.34 standard deviations and lower depression by about -0.23 standard deviations. So this effect is roughly 25% to 40% of that of an ordinary psychological intervention. The cost over the same period: the current-affairs knowledge index fell by 0.19 standard deviations, and the degree of polarisation on political positions fell by 0.16 standard deviations. After the experiment ended, Facebook use in the switch-off group was still 12 minutes a day below the control group (0.31 standard deviations, a drop of 23%). A further randomised controlled trial enrolled 143 undergraduates, randomly split into two groups. One group limited Facebook, Instagram and Snapchat to 10 minutes a day each, and the other carried on as before. After three weeks, loneliness and depression fell clearly in the limited group. Anxiety and fear of missing out fell from their starting levels in both groups. +- Evidence grade: A +- Sources: Lei Y, Hu S, Sun Y, Zheng L (2026). "Looking up" linked to feeling down: a meta-analysis of online upward social comparison and psychological maladjustment. Frontiers in Psychology. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2020). The Welfare Effects of Social Media. American Economic Review, 110(3), 629–676. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2019). The Welfare Effects of Social Media. NBER Working Paper No. 25514 (the specific figures quoted in this entry were checked word for word against this version; the published abstract contains no figures). ; Hunt MG, Marx R, Lipson C, Young J (2018). No More FOMO: Limiting Social Media Decreases Loneliness and Depression. Journal of Social and Clinical Psychology, 37(10), 751–768. +- Notes: The "correlation" half only shows that two things occur together, and cannot be read as one causing the other; the authors themselves say more long-term follow-up is needed. The randomly assigned half can speak about causation, but the participants were US Facebook users just before the 2018 midterm elections, and applying that to Chinese platforms is a forced fit. The Hunt study has only 143 undergraduates, and there is only that one study. The effect sizes are all small, so do not expect switching off an app to solve a mood problem. What you need to block is the "seeing how other people are doing" kind of use. Contacting family and friends, looking things up and looking for work are not part of that. If you cannot hold the line, start by turning off notifications, turning off "people you may know" and turning off recommendations on the home page. For the total screen time, see items 15 and 16 in section 4 (a hard limit on short videos, and no doomscrolling). For what to do first when you are low, see item 19 in this section (move, get daylight, sleep on time). For using purchases to rebuild a sense of identity, see item 23 in section 6. For spending more money to "move up a notch", see item 24 in section 6. +### 22. When you think everybody saw you embarrass yourself, halve the estimate + +- Cost: No money. You only ask yourself one extra question when you are nervous: did they really see it? +- In plain language: Walk into a room of people wearing an embarrassing T-shirt, and the wearer thinks nearly half of them can recognise who is printed on it. In fact fewer than a quarter can. In a T-shirt the wearer considers decent, the estimated number is six times the real one. People also tend to underestimate how much the other person likes them after a conversation. So when you think everybody remembers you embarrassing yourself, halving it is still too high. +- Benefit: Experiment one: participants wore a T-shirt printed with a picture of Barry Manilow and walked into a room with other participants. The wearer estimated on average that 46% of the bystanders could correctly name the person on the shirt. The bystanders' actual accuracy was 23%, so the estimate was exactly twice the real figure. Taking estimate minus actual for each session, the average overestimate was 23 percentage points (95% CI 9–38, credible range). Dropping the two sessions that had only two and three bystanders, it was 27 percentage points (11–43). Experiment two: participants wore a T-shirt of their own choosing that they considered flattering. The average estimate was 48%, the bystanders' actual accuracy 8%, six times over. A separate series of studies covers three settings: paired conversations between strangers, first-year university roommates, and adult workshops. The result is that people systematically underestimate how much the other person likes them, and also underestimate how much the other person enjoyed the conversation. The authors call this the "liking gap". It is present in conversations of different lengths, and it persists for months between first-year roommates. +- Evidence grade: B +- Sources: Gilovich T, Medvec VH, Savitsky K (2000). The spotlight effect in social judgment: An egocentric bias in estimates of the salience of one's own actions and appearance. Journal of Personality and Social Psychology, 78(2), 211–222. ; Boothby EJ, Cooney G, Sandstrom GM, Clark MS (2018). The Liking Gap in Conversations: Do People Like Us More Than We Think? Psychological Science, 29(11), 1742–1756. +- Notes: The reason for grading B: these are small laboratory and campus studies, in settings of wearing clothes and talking to strangers. Applied to "colleagues are talking about you" or "you said the wrong thing at the meeting", there is no direct evidence. These numbers show the direction of the overestimate and the rough multiple; they cannot be used as a coefficient. Other people's opinions are not all unimportant either. Where you really are being assessed, such as an interview, a viva, an appraisal or a medical report, the assessment is real, and you should prepare. Discount only that everyday feeling of "everybody is watching me"; what you are dealing with there is your own estimation bias. +### 23. Treat "others expect me to be perfect" as a symptom, not as a fact + +- Cost: No money. The hard part is treating such a thought as something that needs handling, rather than as an accurate measurement of what others actually require. +- In plain language: People who always feel that "the people around me expect me to be perfect and will reject me if I fall short" are more likely to have suicidal thoughts. The link between "I expect myself to be perfect" and suicidal thoughts is much weaker, about a third of the first. After subtracting the thoughts people already had, the first feeling is still related to a later increase in thoughts, though the size is small. +- Benefit: Pooling several studies (a meta-analysis): 45 studies, N = 11,747 with undergraduate students, medical students, community adults and psychiatric patients as participants, using a random-effects model. Below are the weighted mean correlations with suicidal ideation. Perfectionism concerns r = .28 (95% CI .24–.32, credible range; 42 studies, 7,936 people). Perfectionism concerns is a composite that includes socially prescribed perfectionism, worry about making mistakes, doubts about one's own actions, and the gap between reality and one's standards. Socially prescribed perfectionism on its own also gives r = .28 (.25–.32, 30 studies, 3,640 people). Perfectionism striving r = .10 (.07–.13, 31 studies, 4,588 people); it is a composite of self-oriented perfectionism and personal standards. Parental criticism r = .20 (.11–.29). Now the relationship with suicide attempts. Perfectionism concerns r = .12 (.07–.17, 15 studies, 5,275 people). Socially prescribed perfectionism r = .19 (.08–.29, 8 studies, 689 people). For the part that follows people over time: after subtracting the suicidal ideation present at the start, socially prescribed perfectionism still moved in the same direction as suicidal ideation at the later follow-up, with a small size. Neither self-oriented nor other-oriented perfectionism shows any association. +- Evidence grade: B +- Sources: Smith MM, Sherry SB, Chen S, Saklofske DH, Mushquash C, Flett GL, Hewitt PL (2018). The perniciousness of perfectionism: A meta-analytic review of the perfectionism-suicide relationship. Journal of Personality, 86(3), 522–542. +- Notes: The reason for grading B: most of the studies included are one-off questionnaires, so the order of events is not visible. The correlation is moderate (r of about .28) and explains only a few percent to a tenth of the difference. What it measured is suicidal ideation and suicide attempts, not death itself, so the benefit is graded small on the grounds of "surrogate endpoint only". The authors themselves say the set of perfectionism concerns and suicide attempts looks like it has publication bias. Publication bias means studies with unflattering results do not get published. Putting the estimated missing studies back in does not change the direction of the conclusion. High standards in themselves are not the problem, and you need not lower them for that reason. The difference lies in where the standard comes from, and in whether failing to meet it makes you feel "I am finished as a person". Like the pessimistic expectations in item 15 (treat "things will definitely get worse" thoughts as a symptom), this is something that can be treated as a symptom. If you have suicidal thoughts, call 12356 first and move lethal means out of reach, see item 25 in section 1. For what you can do when someone near you shows such thoughts, see item 15 in section 8. For depression screening in children, see item 8 in section 30. +### 24. When you are angry, do not let it out by smashing things, hitting a punch bag or running a lap; slow the body down first + +- Cost: No money. In the heat of the moment, spend a few minutes on something that slows your heartbeat: breathe slowly, sit down, stay still. The hard part is that what you most want to do then is exactly to let it out. +- In plain language: Hitting a punch bag, throwing things down, going for a run — these ways of letting it out that make the body more excited do not calm anger overall. Slow breathing, mindfulness and meditation, which calm the body down, clearly reduce anger and the urge to hit someone. +- Benefit: Pooling 154 studies (a meta-analysis), 184 samples in all and 10,189 people. Methods that calm the body down, such as deep breathing, mindfulness and meditation, lower anger and aggression. Effect size g = -0.63, 95% CI -0.82 to -0.43 (credible range). The result is stable across people of different sexes, races, ages and cultures. Methods that make the body more excited, such as hitting a punch bag, jogging and cycling, have no effect overall. Their effect size is g = -0.02, 95% CI -0.13 to 0.09. The authors' conclusion is that letting anger out, and going for a run, are not effective ways to calm down. +- Evidence grade: A +- Sources: Kjærvik SL, Bushman BJ (2024). A meta-analytic review of anger management activities that increase or decrease arousal: What fuels or douses rage? Clinical Psychology Review. +- Notes: What it measured is anger and aggression, not energy. The measure is recorded as time because that is the author's own classification: one less row to clear up after one less argument. For something you can use on the spot, see item 9 in section 22 (cyclic sighing), and item 18 in this section (walk away when you are angry). Running is not useless. Over the long run it works for low mood: see item 7 in section 22 (walk or run when you feel down). It is just not suitable for calming down on the spot. Do not drive, do not send messages and do not make decisions while angry. If somebody is hitting you, leave first and call the police. What needs handling then is not your emotions: see item 43 in section 8 (domestic violence). + +### 25. When something is weighing on you, write down your thoughts and feelings over several days, but do not expect it to cure anything + +- Cost: No money. The usual research design is to write 3 to 4 times, for 15 to 20 minutes each time. Write about one thing that upset you, and write your deepest thoughts and feelings about it. +- In plain language: Writing a bad thing down brings a benefit on average, and it is a small one. The pooled conclusions are inconsistent, and two of the pooled analyses found no effect. It costs almost nothing, so it is worth trying. If the emotional problem is serious, do not use it instead of seeing a doctor. +- Benefit: One analysis pooled 146 randomised studies (a meta-analysis). The participants wrote down or spoke about something that mattered to them. The average effect is positive and significant, but very small, with effect size r = .075. Another pooled analysis took in randomised controlled trials only, and pooling 30 trials found no significant effect on either physical health or mental health. Pooling 16 randomised trials in cancer patients and survivors found nothing significant on the psychological, physical or quality-of-life measures. On the psychological measure, g = 0.04, 95% CI -0.06 to 0.14 (credible range). +- Evidence grade: B +- Sources: Frattaroli J (2006). Experimental disclosure and its moderators: a meta-analysis. Psychological Bulletin. ; Mogk C, Otte S, Reinhold-Hurley B, Kröner-Herwig B (2006). Health effects of expressive writing on stressful or traumatic experiences - a meta-analysis. Psycho-Social Medicine. ; Zachariae R, O'Toole MS (2015). The effect of expressive writing intervention on psychological and physical health outcomes in cancer patients--a systematic review and meta-analysis. Psycho-Oncology. +- Notes:Disputed. The supporting side is the pool of 146 studies, with a positive but very small effect. The opposing side is two pooled analyses that took in randomised controlled trials only, and neither found a significant effect. The number of sessions and the length of each one in the Cost field come from the earliest experimental design, and are stated in the abstract of the Mogk review. It costs almost no money and little time, and one person can do it alone. When you do not want to talk to someone, or there is nobody around to talk to, you can use it. What it measures is not energy; recording the measure as time is the author's own classification. If writing makes you feel worse as you go, stop and call 12356 instead, see item 25 in section 1 (12356). diff --git "a/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" "b/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" new file mode 100644 index 00000000..9b14a7d9 --- /dev/null +++ "b/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" @@ -0,0 +1,152 @@ +[← 回总目录](../README.md) + +# 4. 不要浪费时间 + +本节只算时间账。看的是每天、每周、每年能省回多少小时，或者一件事按时做完的比例能提高多少。这里不把时间换算成寿命，也不换算成钱。涉及钱的地方只给算法，不给结论。本节多数条目的证据是单独一项研究，或者是时间利用调查的总量统计，所以等级以 B、C 为主。 + +### 1. Turn "I mean to do it" into "when, where, and if this comes up then that" + +- Cost: No money. It takes 1 to 2 minutes per thing to write one sentence of the form "if ... then ...". For example: "On Tuesday, when I sit down at my desk at 8 in the morning, I write the introduction first". +- In plain language: Rewrite "I am going to do this thing" as "on Tuesday, when I sit down at my desk at 8 in the morning, I write the introduction first". Pooling 94 tests, people who write it that way achieve their goals markedly more often, a medium to large effect. You only change how one sentence is worded. It costs no money and takes no effort. +- Benefit: Pooling 94 independent tests, people who first write the intention as "if ... then ..." reach their goals more often. This wording is called an implementation intention in psychology. Effect size d = 0.65, which counts as medium to large. Three things are doing the work: getting started, holding on when you are interrupted, and getting out of something that was never going to work. +- Evidence grade: A +- Sources: Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. +- Notes: This only works for things you genuinely want to do. If you do not want the goal in the first place, writing more sentences will not help. Also, most of these studies recruited students and mostly looked at health habits, so it may not work as well in a work setting. +### 2. Write down your exit condition before you start + +- Cost: No money. It takes 5 minutes when you start, to write clearly: "by this date, if I have not hit this mark, I stop". The hard part is honouring it on the day; stopping in front of other people loses face. +- In plain language: Before you start, write down one sentence: by this date, if I have not hit this mark, I stop. Pooling several studies, this wording really does help people withdraw from something that was never going to work. What you save is all the time you would otherwise have gone on putting in. +- Benefit: Pooling several implementation intention studies, the result is that a pre-set "if ... then ..." also helps people withdraw from something that was never going to work. The original wording is disengagement from failing courses of action. The time you save is the time you would otherwise have gone on investing. +- Evidence grade: B +- Sources: Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. +- Notes: Set the exit condition right at the start, while you are still calm. Set it later and the cost already sunk in will pull you. When you judge, look only at future input and future return, see item 3. Also, "being willing to withdraw" is only one sub-finding inside a pool of studies; nobody has worked out on its own how many hours it saves. +### 3. When deciding whether to continue, count only future input and future return, not what has already gone in + +- Cost: No money. The hard part is accepting one thing: the time you put in before is gone and cannot be recovered. +- In plain language: Money and time already spent make people keener to carry on. Pooling 98 sets of data, the tendency is reliably there. So when you judge whether to continue, count only how much more you have to put in and how much you can still get back. Leave out what you have already spent. +- Benefit: Money, energy and time already invested make people keener to continue. In psychology this is called the sunk cost effect. Pooling 98 sets of data, the analysis confirms that the tendency is reliably there. If you steer clear of it, you stop putting time in just because you "do not want to waste it". +- Evidence grade: B +- Sources: Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35, 124–140. ; Roth, S., Robbert, T., & Straus, L. (2015). On the sunk-cost effect in economic decision-making: A meta-analytic review. Business Research, 8(1), 99–138. +- Notes: The theatre experiment by Arkes and Blumer ran like this. The University of Ohio theatre sold season tickets and randomly assigned buyers to three prices: full price 15 dollars, 2 dollars cheaper, and 7 dollars cheaper. Over the first half of the season the full-price group attended an average of 4.11 shows, and the other two groups 3.32 and 3.29. Over the second half there was no clear difference between the three groups. Roth and colleagues found that the effect fades with time, and that older people are affected less. +### 4. Estimate the schedule from how long comparable tasks actually took, not from the plan + +- Cost: No money. Before estimating, spend a few minutes checking how long the last comparable job actually took. +- In plain language: Students estimated 33.9 days to finish their thesis and actually took 55.5 days, and only 29.7% finished within their own estimate. A second group was asked to recall how long the last comparable job had taken before estimating, and the share finishing on time rose from 29.3% to 60.0%. +- Benefit: Students expected 33.9 days to finish their thesis and actually took 55.5 days. Only 29.7% finished on time according to their own estimate. The researchers asked a second group to connect past experience to the task in front of them before estimating. In that group the share finishing on time rose from 29.3% in the control group to 60.0%. The optimism that shortens the estimate no longer showed up in that group. +- Evidence grade: B +- Sources: Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. ; Flyvbjerg, B. (2006). From Nobel Prize to project management: Getting risks right. Project Management Journal, 37(3), 5–15. ; Halkjelsvik, T., & Jørgensen, M. (2012). From origami to software development: A review of studies on judgment-based predictions of performance time. Psychological Bulletin, 138(2), 238–271. +- Notes: The Buehler study recruited only Canadian undergraduates. It is a single study with few participants, 37 people and about 40 per group. Estimating from the historical record of comparable projects is called reference class forecasting; Flyvbjerg explains how to do it in engineering. After pooling several studies, Halkjelsvik and Jørgensen found that underestimating is more common than overestimating, but by how much varies a lot with the task and the research method. +### 5. Send an agenda before the meeting and do not hold one without it; hold it standing if you can + +- Cost: No money. The organiser spends 5 to 10 minutes before the meeting writing an agenda. Holding a standing meeting needs the participants to go along with it. +- In plain language: In meetings where an agenda went out beforehand, participants were more likely to think the meeting was useful. In the experiment, meetings held sitting down ran 34% longer than those held standing, and the decisions the two formats produced were no different in quality. +- Benefit: Two surveys with 958 and 292 people respectively. The result is that in meetings where an agenda went out beforehand, participants were more likely to think the meeting had been effective, and the relationship holds statistically. In another experiment 55 groups held their meeting sitting down and 56 held it standing, and the sitting-down meetings ran 34% longer. The two formats produced decisions of no different quality. +- Evidence grade: B +- Sources: Leach, D. J., Rogelberg, S. G., Warr, P. B., & Burnfield, J. L. (2009). Perceived meeting effectiveness: The role of design characteristics. Journal of Business and Psychology, 24(1), 65–76. ; Bluedorn, A. C., Turban, D. B., & Love, M. S. (1999). The effects of stand-up and sit-down meeting formats on meeting outcomes. Journal of Applied Psychology, 84(2). +- Notes: The standing-meeting experiment used tasks carried out by five-person groups in a laboratory. You cannot carry this conclusion over to long meetings and large conferences. Leach measured how effective participants felt the meeting was, not how much work actually got done. +### 6. Cut the number of meetings; do not meet when writing it down would do + +- Cost: No money. You need colleagues and your manager to go along with it. The hard part is that writing something clearly takes more effort than saying it out loud. +- In plain language: The more meetings you have in a day, the more tired you are and the more work you feel you have on your hands. Cut one meeting and the time you save is the full length of that meeting, with nothing deducted. +- Benefit: Two surveys of employees working more than 35 hours a week, with 676 and 304 people respectively. How much time meetings take does not have a fixed effect on work attitudes and well-being; it depends on two things: how well the meetings are run, and how much the job depends on other people. Another study had people keep a diary for a week, and the result was that the more meetings you had in a day, the higher the fatigue and the more work you felt you had done. The time you save equals the full length of the meetings you cut. +- Evidence grade: B +- Sources: Rogelberg, S. G., Leach, D. J., Warr, P. B., & Burnfield, J. L. (2006). "Not another meeting!" Are meeting time demands related to employee well-being? Journal of Applied Psychology, 91(1). ; Luong, A., & Rogelberg, S. G. (2005). Meetings and more meetings: The relationship between meeting load and the daily well-being of employees. Group Dynamics: Theory, Research, and Practice, 9(1). +- Notes:Disputed: Rogelberg 2006 found that the relationship between the number of meetings and well-being is not one-sided. In jobs that depend heavily on other people, many meetings are not a bad thing in themselves. Also, no study has compared writing things down asynchronously against holding meetings, so that part is the author's own inference. +### 7. Break a large task into sub-tasks, then estimate, then start + +- Cost: No money. The splitting itself takes 5 to 15 minutes. +- In plain language: When people estimate how long something will take, they do not automatically break the task up and count the pieces. Listing the sub-tasks first and estimating afterwards stops the estimate coming out too short. Once it is broken up, every small step can be started at once, and it is less easy to put off. +- Benefit: When people estimate how long something will take, they do not on their own split a task that has several parts into separate blocks. Writing a literature review, writing the discussion and sorting the references are already several jobs. Listing the sub-tasks first and estimating afterwards reduces the planning fallacy, the mistake of estimating a task too short. Once it is broken up, every small step becomes something you can begin at once, and the procrastination falls with it. The factors that predict procrastination best in the research include how much the task disgusts people and how far back the task has been pushed. +- Evidence grade: B +- Sources: Kruger, J., & Evans, M. (2004). If you don't want to be late, enumerate: Unpacking reduces the planning fallacy. Journal of Experimental Social Psychology, 40(5), 586–598. ; Steel, P. (2007). The nature of procrastination: A meta-analytic and theoretical review of quintessential self-regulatory failure. Psychological Bulletin, 133(1). +- Notes: How far to split: use "each sub-task can be finished in one sitting" as the test, because splitting too finely also costs time. Also, for Kruger and Evans, only the direction was verified from the abstract; the specific percentages were not checked against the full text. + +### 8. Give things with no external deadline a date of your own + +- Cost: No money. The hard part is that once you set it you have to honour it; ignoring the day when it arrives means you never set it. +- In plain language: When someone else sets the deadline, 80.6% of students finish before it. When the deadline is their own estimate, only 38.7% finish on time. When something gets finished is decided almost entirely by the deadline rather than by your own estimate. So for anything with no deadline, set one yourself and honour it. +- Benefit: Among students with an external deadline, 80.6% finished before it. Among those working to their own expected time, only 38.7% finished on time. When something gets finished follows the deadline almost exactly, correlation coefficient r = 0.82, and the closer this number is to 1 the tighter the fit. The relationship with your own estimate is very weak. +- Evidence grade: B +- Sources: Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. +- Notes: Tell other people the date you set and it behaves more like one they set. What the research actually verified is that deadlines set by others work. Self-set limits such as a time box or the Pomodoro technique have not been tested directly, so that part is an inference by analogy. +### 9. Read procrastination as avoiding something unpleasant, not as a lack of effort + +- Cost: No money. What changes is the direction of your judgement: first ask where in the task the unpleasant part is, then work out what to do. +- In plain language: Procrastination has little to do with laziness and little to do with personality. What tells you most about whether a person will put something off is how unpleasant the task itself is, how much they believe they can do it, and how easily they are pulled away by what is in front of them. So what helps is making the task less unpleasant and making the first step smaller, not telling yourself off and then forcing yourself through it. +- Benefit: A meta-analysis pooling 691 correlation coefficients found several things strongly and reliably related to procrastination: how much the task disgusts people, how far away the deadline is, self-efficacy (believing you can do it), impulsivity, and conscientiousness together with its components of self-control, distractibility, orderliness and achievement motivation. Neuroticism, rebelliousness and sensation seeking have only a very weak association with procrastination. The author explains this set of results with temporal motivation theory: how attractive something is right now falls off quickly as the reward moves further away. +- Evidence grade: A +- Sources: Steel P (2007). The nature of procrastination: a meta-analytic and theoretical review of quintessential self-regulatory failure. Psychological Bulletin, 133(1), 65–94. +- Notes: The benefit is graded "medium" because these studies only give direction and do not measure how many hours are saved. The specific moves are in item 1 in this section (write "when, where, and if this comes up then that"), item 7 (break a large task into sub-tasks) and item 8 (give anything with no external deadline a date). These are pools of correlational studies, so they cannot show which causes which. +### 10. Put what you need within reach, move what you would rather not touch further away, and do not rely on resisting it there and then + +- Cost: No money, a few minutes each time. Put the phone in another room, and leave the book and the laptop open on the desk. +- In plain language: The further away something is, the less of it you take. Put the same snack further away and the amount eaten falls noticeably, a medium-sized effect. Offer fewer options and noticeably fewer people take them. These methods do not rely on you resisting in the moment; a few minutes of setting up is enough. +- Benefit: A Cochrane review including 24 randomised trials. Putting food further away lowers the amount eaten, standardised mean difference −0.60 (95% CI −0.84 to −0.36, the credible range; 12 studies, 1098 people, low-certainty evidence), which counts as a moderate effect. Reducing the options available shows a bigger drop in choosing that food, standardised mean difference −1.13 (95% CI −1.90 to −0.37, 3 studies, 154 people, low-certainty evidence). The regression analysis shows: the further away it is and the fewer the options, the stronger the effect. +- Evidence grade: B +- Sources: Hollands GJ, Carter P, Anwer S, et al. (2019). Altering the availability or proximity of food, alcohol, and tobacco products to change their selection and consumption. Cochrane Database of Systematic Reviews, 9, CD012573. +- Notes: The reason for grading B: all 24 studies in this review are about food, with not one on alcohol or tobacco; 14 were done in laboratories, and all come from high-income countries. So applying "move it further away and you touch it less" to phones and games is an extension in the same direction, with no direct trial. Putting the phone out of sight has been tested directly, see item 1 in section 3 (turn off notifications you do not need, and keep the phone out of sight while you work). +### 11. To make yourself follow through, put money on it, but first ask whether you are willing to stake any + +- Cost: A deposit scheme requires you to put up your own money first, and it is forfeited if you do not deliver. The trial staked 150 dollars. +- In plain language: If you want to force yourself to deliver, you can take a bonus for delivering, or you can put up a deposit that is refunded only if you deliver, plus a bonus. Among people willing to take part in either, those who put up a deposit delivered 13 percentage points more often. But only 13.7% were willing to put up a deposit, against 90% willing to take the bonus, so a deposit only works for people willing to stake money. +- Benefit: A US randomised trial of 2538 people randomly assigned employees and their family and friends to four incentive programmes or to usual care. The reward scheme (about 800 dollars on delivery) had an acceptance rate of 90.0%, and the deposit scheme (150 dollars up front, plus a 650 dollar reward) 13.7%. Six-month continuous smoking abstinence: the four incentive programmes ran between 9.4% and 16.0%, against 6.0% for usual care. The reward scheme 15.7%, the deposit scheme 10.2%. But among just the 13.7% willing to take part in either, the deposit scheme gave a six-month abstinence rate 13.2 percentage points higher than the reward scheme (95% CI 3.1 to 22.8, the credible range). +- Evidence grade: A +- Sources: Halpern SD, French B, Small DS, et al. (2015). Randomized trial of four financial-incentive programs for smoking cessation. New England Journal of Medicine, 372(22), 2108–2117. +- Notes: This trial measured smoking cessation; applying it to writing a thesis or going to the gym is an extension in the same direction, so the benefit is graded "medium". When you pick a scheme, first check whether you will really use it. A method most people refuse has no use for them, however strong the effect. So pick the one you are willing to accept; staking a smaller amount still beats not staking. For how to give up smoking, see item 3 in section 2 (cessation drugs). +### 12. Count a new habit in months, not in "21 days" + +- Cost: No money. What it takes is repeating the same action every day in the same setting, for several months. +- In plain language: "You can build a habit in 21 days" has no basis behind it. A study following people for 12 weeks worked out that an action becomes something you do without thinking in as little as 18 days, and in 254 days in the slow cases. Two or three weeks without it becoming easier is normal. Missing a day in between does not reset your progress. +- Benefit: 96 volunteers each picked one eating, drinking or activity action, did it every day in a fixed setting (such as "after breakfast") for 12 weeks, and rated how automatic it felt each day. 82 people's data were good enough to analyse, and the model fitted successfully for 62 of them. The time taken to reach 95% of each person's own automaticity ceiling ranged from 18 days to 254 days. The study also found that missing one opportunity to do the action had no substantial effect on how the habit formed. +- Evidence grade: B +- Sources: Lally P, van Jaarsveld CHM, Potts HWW, Wardle J (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998–1009. +- Notes: There is only this one study, 96 people, and the actions were all simple ones (such as eating a piece of fruit after a meal). How long a complicated thing takes, such as running three times a week or writing every day, has not been tested. So still finding it hard after two weeks is not evidence of failure. To make the repetition actually happen, tie the action to a fixed setting, see item 1 in this section (write "when, where, and if this comes up then that"). +### 13. When procrastination already affects your life, use self-help materials backed by randomised trials rather than insisting on finding a therapist + +- Cost: A self-help book costs a few tens of yuan, and some are free. It takes about ten weeks of work: one section to read and one exercise a week. +- In plain language: Procrastination severe enough to cause suffering can be treated. Ten weeks of working through cognitive behavioural self-help material works clearly better than doing nothing, whether someone guides you or you do it entirely on your own. Guided and unguided make no visible difference, so working from the material yourself is effective too. +- Benefit: 150 people were randomly split into three groups: guided self-help with a therapist, unguided self-help, and a waiting-list control. After ten weeks, the between-group effect size on the pure procrastination scale was d = 0.70 for the guided group (95% CI 0.29–1.10, the credible range) and d = 0.50 for the unguided group (0.10–0.90); on the irrational procrastination scale they were 0.81 (0.40–1.22) and 0.69 (0.29–1.09). The share reaching a clinically meaningful improvement: 31.3% to 40.0% in the guided group and 24.0% to 36.0% in the unguided group. The two groups did not differ significantly on any outcome. +- Evidence grade: A +- Sources: Rozental A, Forsell E, Svensson A, Andersson G, Carlbring P (2015). Internet-based cognitive-behavior therapy for procrastination: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 83(4), 808–824. +- Notes: The participants were Swedish adults using Swedish-language material, and there is no ready-made equivalent in Chinese, so the benefit is graded "medium". The components are the usual ones: behavioural activation, behavioural experiments, stimulus control, and material on motivation and working methods. Stimulus control is item 10 in this section (move what you would rather not touch further away). When procrastination comes with clear low mood or anxiety, start with item 19 in section 3 (move, get daylight, sleep on time, talk to someone, call 12356). +### 14. Use your own hourly rate to decide which housework to outsource + +- Cost: You have to pay for the outsourcing, and how much depends on the service you buy. You also have to spend time once working out your own hourly rate. +- In plain language: People in China spend 1 hour 17 minutes a day on housework on average. Among the people who actually did some that day, it was 1 hour 59 minutes. People who spend money to buy time report higher life satisfaction. The same amount of money spent on buying time makes people happier in the moment than the same amount spent on buying things. +- Benefit: Residents of China spend 1 hour 17 minutes a day on housework on average. Counting only the people who did housework, the average is 1 hour 59 minutes. These are 2024 survey figures, covering 38,500 households and 107,000 people. A separate study took a sample of 6271 people in four countries and found higher life satisfaction among people who spend money to buy time. The researchers also ran an experiment in real life: the same amount of money spent on buying time brought more happiness in the moment than the same amount spent on physical things. +- Evidence grade: B +- Sources: Whillans, A. V., Dunn, E. W., Smeets, P., Bekkers, R., & Norton, M. I. (2017). Buying time promotes happiness. Proceedings of the National Academy of Sciences. ; National Bureau of Statistics (国家统计局) (2024). Bulletin of the Third National Time Use Survey（第三次全国时间利用调查公报（第二号））. +- Notes: Work out your hourly rate first. The formula is one month's take-home income ÷ the hours you actually work in a month. Then see what it would cost to outsource this job. Outsource it when the price is below your hourly rate and you will really spend the saved time on earning money or resting. Everyone's hourly rate and every service price differs, so work out the conclusion yourself. Also, Whillans measured whether buying time made people happy, not how many hours were saved. + +### 15. Set a hard limit on short videos and mindless scrolling + +- Cost: No money. The actions take only a few minutes: turn off notifications, delete the app, or set a daily time limit on it. The hard part is keeping your hand off it and resisting the tap. +- In plain language: People in China spend 5 hours 37 minutes a day online on average. Among those who actually went online that day, it was 6 hours 3 minutes. In 2018, on the same method, it was only 2 hours 42 minutes. This is the largest figure in this section. Cutting one hour a day gives you 365 hours a year. +- Benefit: Residents of China spend 5 hours 37 minutes a day online on average. Counting only those who actually went online that day, the average is 6 hours 3 minutes. 92.9% of people used the internet on the day. These are 2024 figures. In 2018, on the same method, it was only 2 hours 42 minutes. This is the largest figure in this section. Cutting one hour a day out of it gives you 365 hours a year. Separately, according to the China Internet Network Information Center (中国互联网络信息中心), by June 2025 internet users spent an average of 30.6 hours a week online, 1.9 hours more than half a year earlier. There are 1.068 billion short video users, 95.1% of internet users. +- Evidence grade: C +- Sources: National Bureau of Statistics (国家统计局) (2024). Bulletin of the Third National Time Use Survey（第三次全国时间利用调查公报（第二号））. ; National Bureau of Statistics (国家统计局) (2019). Bulletin on the 2018 National Time Use Survey（2018年全国时间利用调查公报）. ; China Internet Network Information Center (中国互联网络信息中心) (2025). The 56th 《中国互联网络发展状况统计报告》 (Statistical Report on the Development of China's Internet). +- Notes: Time online includes work, study and socialising, and it is not all waste. It is graded C because no study has worked out how much is saved after a limit, or where that time then goes; all that is available is the aggregate total. For the "seeing how other people are doing" part of this, randomised trials have measured how much is saved and how much mood changes: see item 21 in section 3. +### 16. No television and no doomscrolling; take in the news you need at fixed times + +- Cost: No money. The hard part is breaking the habit of opening it on impulse, and you may miss some breaking news. +- In plain language: People over 15 in the United States watch 2.6 hours of television a day, which takes half of all their leisure time. People in China watch 1 hour 40 minutes a day on average, and people aged 75 to 84 watch 3 hours 16 minutes. If you switch the information you need to fixed times, what you save is that whole block. +- Benefit: People over 15 in the United States watch 2.6 hours of television a day. Their total leisure time is 5.2 hours, and television takes half of it. These are 2025 figures. Residents of China watch 1 hour 40 minutes of television a day on average, and people aged 75 to 84 watch 3 hours 16 minutes. These are 2018 figures. +- Evidence grade: C +- Sources: U.S. Bureau of Labor Statistics (2026). American Time Use Survey — 2025 Results. ; National Bureau of Statistics (国家统计局) (2019). Bulletin on the 2018 National Time Use Survey（2018年全国时间利用调查公报）. +- Notes: All that is available is the aggregate total. No study has worked out where the saved time goes after watching less television and news. The practice of "taking it in at fixed times" is the author's own experience. +### 17. Spend time once learning the keyboard shortcuts and automation of the software you use every day + +- Cost: No money. You have to put in a few hours to several dozen hours of learning in one go. The hard part is that at first you are slower than with the mouse. +- In plain language: Most people who have used Word for years still click the toolbar icons. In an experiment comparing groups, shortcuts were faster than both menus and icons. Learning this once is a one-off cost, and after that every single operation each day pays it back. The longer you do the same kind of work, the more it pays back. +- Benefit: The study looked at 251 experienced Word users, most of whom rarely used the faster shortcuts and still worked mainly by clicking the icon toolbar. The experiment comparing groups confirms that of the three methods, menus, icons and shortcuts, shortcuts are the fastest. The time saved accumulates a little with every operation each day. People who do the same kind of work for a long time gain the most. +- Evidence grade: B +- Sources: Lane, D. M., Napier, H. A., Peres, S. C., & Sándor, A. (2005). Hidden costs of graphical user interfaces: Failure to make the transition from menus and icon toolbars to keyboard shortcuts. International Journal of Human-Computer Interaction. +- Notes: Only learn the operations you will use more than ten times a day; learning everything is a different kind of waste. To be clear, that efficiency experiment had only 6 participants, and the abstract does not give the seconds saved per operation. +### 18. Put commute time first when you choose where to live, and shorten the one-way journey + +- Cost: The cost is high. You may have to pay more rent, or move to a smaller place. Moving itself takes time. +- In plain language: The longer your commute, the worse you feel your life is going. That loss is not made up in salary or housing. Cutting 30 minutes off each one-way journey gives you 5 hours a week. +- Benefit: Residents of China spend 50 minutes a day on transport on average. Counting only those who travelled on the day, it is 1 hour 2 minutes. These are 2024 figures. German data following the same group of people over the long term shows that the longer your commute, the stably lower your subjective well-being (how well you feel life is going). And that loss is not made up in income or housing. A review bringing several studies together also confirms that the longer the commute, the less satisfied people are with it, and this holds whatever mode of transport they use. Cutting 30 minutes off each one-way journey gives about 5 hours a week. +- Evidence grade: B +- Sources: Stutzer, A., & Frey, B. S. (2008). Stress that doesn't pay: The commuting paradox. Scandinavian Journal of Economics, 110(2), 339–366. ; Chatterjee, K., et al. (2020). Commuting and wellbeing: A critical overview of the literature with implications for policy and future research. Transport Reviews, 40(1), 5–34. ; National Bureau of Statistics (国家统计局) (2024). Bulletin of the Third National Time Use Survey（第三次全国时间利用调查公报（第二号））. +- Notes:Disputed: after pooling several studies, Chatterjee and colleagues found no stable relationship between commute length and overall life satisfaction. They argue that people can broadly trade a long commute back for better work or housing. Also note that the transport activity figures count all travel, not just commuting. And if you can travel outside the peak hours, the same journey takes less time. diff --git "a/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" "b/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" new file mode 100644 index 00000000..321477e4 --- /dev/null +++ "b/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" @@ -0,0 +1,372 @@ +[← 回总目录](../README.md) + +# 5. 不要浪费钱 + +这一节只算钱：能省下多少，每年收益差多少，一笔钱长期平均下来能拿回多少。这些不换算成寿命或时间。涉及投资的条目只讲怎么算账，只用核实过的历史数据，都不构成投资建议。 + +[← Back to the contents](../README.md) + +# 5. Do not waste money + +This section counts money only: how much you can save, how much the yearly return differs, how much a sum of money gives back on average over the long run. Nothing here is converted into years of life or into time. The entries that touch investing only explain how the arithmetic works. They use checked historical data and none of them is investment advice. + +### 1. Turn off automatic renewal and renew by hand each time + +- Cost: No money. Work once through the automatic-deduction lists in Alipay, WeChat, Apple and Android. It takes 10 to 20 minutes. After that you tap once more yourself every time something renews. +- In plain language: What you save is the subscriptions you stopped using long ago and still pay for. Going through the automatic-deduction lists in Alipay, WeChat, Apple and Android takes under 20 minutes. By law a merchant must warn you clearly before taking the money, and must give you a way to cancel at any time that is easy to find and easy to click. +- Benefit: The money you save equals the total of the subscriptions you no longer use but are still billed for. It differs for everyone, so there is no single figure. The regulator requires merchants to notify you clearly before renewal, and to give you a simple way to cancel at any time, with no separate charge for cancelling (China, in force since 2021) +- Evidence grade: C +- Sources: State Administration for Market Regulation (国家市场监督管理总局) (2021, amended 2025). Measures for the Supervision and Administration of Online Transactions（网络交易监督管理办法）(SAMR Order No. 37, Article eighteen). +- Notes: For a service you really use often, paying yearly is usually cheaper than paying monthly, so leaving renewal on can be the better choice. The ones to turn off are the ones you forgot about months ago and are still paying for. +### 2. Settle your annual individual income tax from March to June, and claim every special additional deduction + +- Cost: No money. You do it in the Individual Income Tax app, and it takes a bit over ten minutes. +- In plain language: These groups of people usually get money back: anyone who worked only a few months of the year, anyone who changed jobs during the year, and anyone who left out deductions for children's education, a home loan, rent, or supporting an elderly relative. It takes a bit over ten minutes on your phone. If your annual comprehensive income is under 120,000 yuan, or the tax you owe is under 400 yuan, you can skip it. +- Benefit: One settlement puts four kinds of income together: salary and wages, remuneration for personal services, author's remuneration, and royalties. What you overpaid comes back, and what you underpaid you have to pay. Two situations let you skip it: your annual comprehensive income does not exceed 120,000 yuan, or the tax you owe does not exceed 400 yuan. Both conditions come from a 2023 State Taxation Administration announcement, in force until 2027-12-31. In the other direction, the following people usually get money back: people with income in only a few months of the year, people who changed jobs during the year, and people who left out deductions for children's education, a home loan, rent, or supporting an elderly relative +- Evidence grade: A +- Sources: State Taxation Administration (国家税务总局) (2025). Measures for the Settlement of Comprehensive Individual Income Tax (个人所得税综合所得汇算清缴管理办法) (Article six) and the official explanation. ; Ministry of Finance and State Taxation Administration (2023). Announcement on Continuing the Settlement and Payment of Comprehensive Individual Income Tax. +- Notes: People who owe tax still have to file, and not filing is illegal. There are seven special additional deductions: children's education, continuing education, serious medical treatment, housing loan interest, housing rent, supporting an elderly relative, and care of children under 3. A housing loan and rent cannot both be claimed; pick one. +### 3. The housing provident fund is not only for buying a home: rent, renovation and building fees can all be withdrawn, under the new rules from 20 September 2026 + +- Cost: No money. You go through the withdrawal procedure once. +- In plain language: The provident fund is not frozen unless you buy a home. The new rules list nine situations you can withdraw for, and paying rent, renovating the home you live in, and paying building management fees for the home you live in are all among them. Once you file, the provident fund centre must tell you within 3 days whether it approves the withdrawal. +- Benefit: The amended regulations list nine situations in which you can withdraw. They include paying rent, renovating the home you live in, and paying the building management fees for the home you live in. The housing provident fund management centre must decide within 3 days of accepting your application whether to approve the withdrawal or refuse it +- Evidence grade: A +- Sources: State Council (国务院) (2026). Decision of the State Council on Amending the Regulations on the Administration of Housing Provident Funds（国务院关于修改《住房公积金管理条例》的决定）(State Council Order No. 844, adopted at the 93rd executive meeting of the State Council on 2026-7-31, 「自2026年9月20日起施行」): Article twenty-four as amended「职工有下列情形之一的，可以提取职工住房公积金账户内的存储余额：（一）支付房租的；（二）购买、建造、翻建、大修自住住房的；（三）偿还购房贷款本息的；（四）装修自住住房的；（五）支付自住住房物业费的；（六）离休、退休的；（七）完全丧失劳动能力，并与单位终止劳动关系的；（八）出境定居的；（九）国务院批准的其他住房消费情形。」(the nine situations in which a worker may withdraw the balance in the account); Article twenty-five「住房公积金管理中心应当自受理申请之日起3日内作出准予提取或者不准提取的决定」(the centre decides within 3 days of accepting the application); the contribution rate in Article eighteen「均不得低于职工上一年度月平均工资的5％」(no lower than 5% of the worker's average monthly wage the previous year). +- Notes: The money in the account is your own salary plus the part your employer adds. Left alone it earns interest at a very low rate. The regulations also say that when a worker dies, or is declared dead, the heirs and beneficiaries may withdraw the balance. Which documents to file and how the procedure runs are set by the local housing provident fund management centre. Look up your local centre's guide before you go. +### 4. Recalculate your phone and broadband plan once a year, downgrade what you do not use, and complain if the carrier refuses to switch you + +- Cost: No money. One look at the bill takes a bit over ten minutes. The hard part is holding your ground when the service agent refuses to downgrade you. +- In plain language: The plan you are on is often tens of yuan a month dearer than the plans on sale now. Tens of yuan a month is several hundred yuan a year thrown away. By law a carrier with no valid reason may not refuse to let you port your number, may not drag its feet and leave it undone, and may not set a separate tariff just for people who port. If they stall, complain. +- Benefit: The Ministry of Industry and Information Technology rules state that a carrier may not「无正当理由拒绝、阻止、拖延向用户提供携号转网服务」. It may not「采取拦截、限制等技术手段影响携号转网用户的通信服务质量」. It may not「为携号转网用户设置专项资费方案和营销方案」. An old plan is often tens of yuan a month dearer than the plans on sale now. Staying on it throws away several hundred yuan a year +- Evidence grade: A +- Sources: Ministry of Industry and Information Technology (工业和信息化部) (2019). Provisions on the Administration of Number Porting Services（携号转网服务管理规定）(MIIT 信管〔2019〕242). +- Notes: First check in the carrier's app how much data and how many minutes you actually used in the last six months, then pick a plan from that. If the agent says the downgrade has to be done at a shop, or that it cannot change during the contract period, ask for the written basis. If they cannot produce one, complain on 12300. +### 5. Do not buy lottery tickets + +- Cost: No money. Not buying is the saving. +- In plain language: By law, the share of sales that lotto and number games set aside as prizes is at least 50%, and for Double Colour Ball it is 51%. That means that for every 1 yuan you put in, you get 0.51 yuan back on average over the long run, a steady loss of 49%. Nationwide lottery sales in 2024 were 623.486 billion yuan, of which 161.031 billion yuan went straight into the public welfare fund and never counted as prizes. +- Benefit: The Ministry of Finance sets the minimum share of sales that goes to prizes: 50% for lotto and number games, 65% for scratch cards, 73% for sports betting. Double Colour Ball pays out 51% of sales, so out of every 100 yuan in sales only 51 yuan becomes prizes. Put in terms of one person, every 1 yuan put in returns 0.51 yuan on average over the long run, a loss of 49%. Lottery sales nationwide in 2024 were 623.486 billion yuan, of which 161.031 billion yuan went into the public welfare fund and did not come back to the people who bought the tickets (China, 2024) +- Evidence grade: A +- Sources: Ministry of Finance (财政部) (2015). Notice on Regulating and Strengthening the Management of the Composition Ratios of Lottery Funds（关于规范和加强彩票资金构成比例管理的通知）(财综〔2015〕94). ; Ministry of Finance (2025). Ministry of Finance Announcement No. 8 of 2025 (2024 Lottery Public Welfare Fund Raising and Allocation). ; Rules of the China Welfare Lottery Double Colour Ball Game（第十四条）. +- Notes: Over the long run you lose, and the system fixes that. It has nothing to do with luck. Buying is fine too, but treat it as 2 yuan of entertainment each time, not as saving. +### 6. Walk away from "high returns", "principal guaranteed" and "steady profit" + +- Cost: No money. All it takes is to stop when you hear those words. +- In plain language: The regulator's own words: a return above 6% should make you ask questions, above 8% is dangerous, and above 10% you should expect to lose your whole principal. Regulated investment products are no longer allowed to promise that the principal is safe or that you will earn. Money lost in illegal fundraising is borne by you alone under the rules, and nobody covers you if it goes wrong. +- Benefit: In 2018 the former chairman of the banking and insurance regulator said something at the Lujiazui forum. His words were:「收益率超过 6% 的就要打问号，超过 8% 的就很危险，10% 以上就要准备损失全部本金」. The new asset management rules require financial institutions to「不得承诺保本保收益」and to「出现兑付困难时，金融机构不得以任何形式垫资兑付」. So regulated products no longer have a guaranteed principal. Losses from taking part in illegal fundraising are「由集资参与人自行承担」, and nobody pays them back when it goes wrong (China, 2018 and 2021) +- Evidence grade: A +- Sources: Guo Shouqing (2018). Preventing and Resolving Financial Risks, Striding Over the Major Checkpoints（防范化解金融风险 奋力跨越重大关口）(speech at the 10th Lujiazui Forum, on the National Financial Regulatory Administration website):「在打击非法集资过程中，努力通过多种方式让人民群众认识到，高收益意味着高风险，收益率超过6%的就要打问号，超过8%的就很危险，10%以上就要准备损失全部本金」. ; reprinted by the Tianjin Municipal Local Financial Administration Bureau. ; People's Bank of China and others (2018). Guiding Opinions on Regulating the Asset Management Business of Financial Institutions（关于规范金融机构资产管理业务的指导意见）(银发〔2018〕106, Article two and Article six). ; State Council (2021). Regulations on the Prevention and Handling of Illegal Fundraising（防范和处置非法集资条例）(State Council Order No. 737, Article twenty-five). +- Notes: The 6%, 8% and 10% lines are where a regulator drew the boundary from experience. They are not provisions of law. The words were said about cracking down on illegal fundraising, and they target investment products and investment companies that promise high returns. None of this is investment advice. +### 7. Do not use the minimum payment on a credit card, and do not open instalments or consumer loans to buy things + +- Cost: No money. You have to keep enough cash to pay the statement in full before the due date. The hard part is resisting instalments when a large bill arrives. +- In plain language: The minimum payment is not a discount. It is a loan that charges interest every day, which works out at around 18% a year. The moment you use it the interest-free period ends, and interest runs on the unpaid principal from that day to the last. When you look at an interest rate, only the annualised figure counts. The People's Bank of China already requires card issuers to display the annualised rate prominently and forbids showing only a daily rate. +- Benefit: Before 2020 credit card overdraft rates had a ceiling and a floor: the ceiling was「日利率万分之五」and the floor was 0.7 times the ceiling. Five ten-thousandths a day works out at 18.25% a year, and that is simple interest: at 18.25% a year the interest does not compound into the principal. From 2021 the card issuer and the cardholder set the price themselves. Once you use the minimum payment, interest runs every day on the full unpaid principal and the interest-free period ends at the same time. The People's Bank of China already requires card issuers to display the annualised rate(the rate converted to a whole year)prominently and states that they「不得仅展示日利率、日还款额」. Read the annualised figure before you decide (China, notice of 2020) +- Evidence grade: C +- Sources: People's Bank of China (中国人民银行) (2020). Notice on Promoting the Market-Based Reform of Credit Card Overdraft Interest Rates（关于推进信用卡透支利率市场化改革的通知）(银发〔2020〕327). +- Notes: The 18.25% is five ten-thousandths multiplied straight by 365, with no compounding. Compounded monthly it is higher. Some banks give a discount, so go by the annualised rate printed on your own statement. When you really cannot get the cash together, look for a proper channel with a lower rate first, and do not let a minimum payment roll and gather interest. +### 8. Do not tip streamers, do not top up games, do not buy on impulse + +- Cost: No money. Spend a few minutes on three things: turn off password-free payment, set a limit per payment and per day, and delete the payment methods bound to the app. The hard part is resisting the tap in the moment. +- In plain language: Players in China spend about 480 yuan a year each on games, and almost none of it comes back. The law only lets a parent get a refund for money a minor spent without the parent's consent; adults have no such rule. The cheapest defence is to set it up in advance: turn off password-free payment, set a per-payment and daily limit, and unbind the payment method. +- Benefit: This is a large item of spending. Actual sales in China's game market in 2024 were 325.783 billion yuan, with 674 million users, or about 480 yuan per person (worked out by dividing the first figure by the second). Money spent is almost never recovered. There is one exception: when a person with limited capacity for civil conduct (here, a minor) tips or tops up without the guardian's consent, and the guardian asks for the money back,「人民法院应予支持」. A tip or a top-up an adult makes for themselves has no such rule. Separately, minors may play online games only on Fridays, Saturdays, Sundays and public holidays, and only in the 1 hour from 20 to 21 (China, 2020 and 2021) +- Evidence grade: B +- Sources: National Press and Publication Administration (国家新闻出版署) (2024). China's game market revenue exceeded 325,700 million yuan this year, a new high (中国音数协游戏工委《2024 年中国游戏产业报告》). ; Supreme People's Court (最高人民法院) (2020). Guiding Opinions on the Proper Handling of Civil Cases Involving the Novel Coronavirus Epidemic, Part Two（关于依法妥善审理涉新冠肺炎疫情民事案件若干问题的指导意见（二））(法发〔2020〕17 号, item 9). ; National Press and Publication Administration (2021). Notice on Further Strict Management and the Effective Prevention of Minors Becoming Addicted to Online Games（关于进一步严格管理 切实防止未成年人沉迷网络游戏的通知）(国新出发〔2021〕14 号). +- Notes: It is graded B because nobody has measured directly how much you save by not tipping. All that is available is an aggregate figure for the games industry. Turning off password-free payment and setting limits are rules of thumb, and on their own they would be graded C. The wording of Article nineteen and Article one hundred and forty-five of the Civil Code is already quoted word for word in item 9 in this section (a child tops up and tips with a phone). None of this is investment advice. +### 9. A child who tops up and tips with a phone: spending above the age of eight that the parent has not approved can be reclaimed + +- Cost: No money. You have to spend time collecting evidence and negotiating with the platform. The platform usually refuses first and refunds later. If you cannot settle it you can only sue, and that takes months to start and the legal fees are yours to pay. So prepare the documents first; going through the platform and 12315 is cheaper. The hard part is that the child will not say anything, and you have to stay patient and get the details. +- In plain language: Money a child over eight spends, if the amount is plainly out of proportion to the child's age and ability to understand, and the parent neither agrees to it nor accepts the bill afterwards, is an invalid transaction, and you can demand the money back. A child under eight cannot in any case decide to spend money alone. If the parent says nothing at all, the law treats that as refusing to approve the transaction. +- Benefit: A minor over eight is a person with limited capacity for civil conduct, which means they may only do things suited to their age and ability to understand. A civil legal act not suited to their age and ability to understand is valid only if the legal representative agrees or approves it afterwards: the parent has to say yes in advance, or accept the bill afterwards. A child under eight has no capacity for civil conduct and cannot decide alone; a legal representative acts for them. If the legal representative expresses no view, that counts as a refusal to approve: a parent who stays silent is refusing the bill +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code（民法典）. Full text as reprinted by the Supreme People's Procuratorate. Article nineteen：「八周岁以上的未成年人为限制民事行为能力人，实施民事法律行为由其法定代理人代理或者经其法定代理人同意、追认；但是，可以独立实施纯获利益的民事法律行为或者与其年龄、智力相适应的民事法律行为。」Article twenty：「不满八周岁的未成年人为无民事行为能力人，由其法定代理人代理实施民事法律行为。」Article one hundred and forty-five：「限制民事行为能力人实施的纯获利益的民事法律行为或者与其年龄、智力、精神健康状况相适应的民事法律行为有效；实施的其他民事法律行为经法定代理人同意或者追认后有效。」「法定代理人未作表示的，视为拒绝追认。」 +- Notes: The whole case rests on evidence. You have to show that the child, not you, was operating the phone. Useful material includes: top-ups clustered in school hours or while a parent was at work, the device's usage records, the child's chat and game logs, and a statement from the child. There is no fixed sum that counts as "suited to the age and ability", and a game skin costing tens of yuan is not the same as a tip costing tens of thousands. Stopping it early is cheaper: set a ceiling on password-free payments, turn on the youth mode, and do not tell the child your payment password. For the case where you yourself tip or top up on impulse, see item 8 in this section (do not tip streamers, do not top up games, do not buy on impulse). +### 10. Teach your child one sentence: anyone who tells you to use your parent's phone or to read out a verification code is a fraudster + +- Cost: No money. Go through the six kinds below with your child, and agree one sentence with them: "if you get scammed, tell me first and I will not shout at you". Also set a ceiling on password-free payments, turn on the youth mode, and do not tell the child your payment password. The hard part is keeping your temper once it really has happened. +- In plain language: The six kinds of scam work differently, but the last step is always the same: get the child to use a parent's phone to transfer money or to read out a verification code. So one sentence is enough: anyone who tells you to touch your parent's phone or to read out a code is a fraudster. Add one more: "if you get scammed, tell me first and I will not shout at you". Money can only come back if the child is willing to say so that day. +- Benefit: The Quanzhou anti-fraud centre worked with the Quanzhou education bureau to break down six kinds of fraud that target teenagers. The methods differ, but the last step is the same: get the child to transfer money from a parent's phone. First, trading game accounts and equipment privately at a low price. Xiao Wang transferred 260 yuan privately, and the other party said「未成年人转账违规，账号要冻结」. He tapped a "verification link" on his father's phone, and more than 30,000 yuan left his account. Second, "lifting the anti-addiction limit". The other party threatened「不转账取消，你家长要坐牢」, and the child transferred 7000 yuan from the mother's phone. The text states that the anti-addiction limit is set by the platform and that there is「根本没有『私下解锁』渠道」. Third, refunds for tips. The fraudster posed as an "official customer service" line, said that a minor's tip could be refunded, and asked for「用家长支付宝给指定直播间刷礼物才能退款」. 73,000 yuan was tipped out of the mother's account and she was blocked. The text says「有人加你让你『刷礼物退款』，100% 是诈骗」. Fourth, posing as a classmate or a teacher. The fraudster stole an account and sent a voice message naming the person to borrow money. In the school term they also joined class groups as a fake teacher and sent「交学杂费 500 元」, with the same avatar and nickname as the real teacher, and arranged for a plant in the group to say「已交」. Fifth, fandom. The fraudster pulled the child into a "celebrity fan group, claim an 888 yuan red packet". First the child transferred 48.88 yuan as a "verification". Then they frightened the child with「未成年人参与违规，平台要冻结 8 万，家长要坐牢」. In total they took 16,000 yuan. The text says「真警察、律师绝不会在网上联系未成年人调查，更不会要家长手机操作」. Sixth, using cheap goods on a second-hand platform to lure people off the platform and into a private transfer. The same material also lists three lines a child should not cross. First, do not lend, sell or rent out a bank card or a phone card; if the card is used to move stolen money, that can amount to the crime of assisting the operation of a criminal network, and a record of it can affect entry to postgraduate study and the civil service. Second, "phone bank" part-time work is illegal. Third, in "high-paid jobs" abroad, a person can be controlled the moment they arrive (nationwide, 2025-10) +- Evidence grade: B +- Sources: Quanzhou Anti-Fraud Centre (泉州市反诈骗中心) (2025). First Lesson of the School Term: Anti-Fraud Is a Required Course for Teenagers, and These 6 Traps Must Be Avoided! (reprinted by the Fujian Provincial Public Security Department) +- Notes: It is graded B because what the authorities publish is a breakdown of fraudsters' scripts and a few cases, with no case rate and no statistics on how effective the precautions are. Keep this apart from item 9 (a child tops up and tips with a phone): money a child spends on tipping and top-ups can be reclaimed under the Civil Code. Money taken by a fraudster does not fall under that rule and can only be reported and stopped through item 2 in section 8. Whether it comes back depends on whether the money is still on the account. The core script fraudsters use with children is "you have broken the rules, your parents will go to prison". Say that plainly to your child before it happens; the sentence itself is the signal that this is a scam, and it works better than asking questions afterwards. You must actually keep your promise about "tell me first, I will not shout". Children hide it for fear of being scolded, and that is the common reason the loss grew in these cases. For the adult version of the same trick, and for deepfakes made with AI, see items 3 and 4 in section 8 (hard anti-fraud rules, hang up before any transfer). +### 11. Refuse the extended warranty on electronics + +- Cost: No money. The price is that after the warranty runs out you pay for any repair yourself, and that is not very likely to happen. +- In plain language: Extended warranties are usually sold at 10% to 50% of the product's original price. What the seller makes on an extended warranty is a dozen times what it makes on an ordinary product. Best Buy in the United States made more than half its profit from extended warranties in 2003, while they were only 3% to 4% of its revenue. That gap is the money you pay for nothing. +- Benefit: Extended warranties are usually priced at 10% to 50% of the product's original price. The industry estimate quoted in the study puts the average gross margin on an extended warranty at 50% to 60%: of every 100 yuan sold, 50 to 60 yuan is profit. That is about 18 times the gross margin on an ordinary product. In 2003, extended warranties were 3% to 4% of Best Buy's revenue but produced more than 50% of its profit (United States, retail data from the 2000s) +- Evidence grade: B +- Sources: Chen T, Kalra A, Sun B (2009). Why Do Consumers Buy Extended Service Contracts? Journal of Consumer Research 36(4):611-623. +- Notes: The gross margin figure is the paper citing someone else's industry estimate (Business Week 2004), not something the paper measured. The calculation above applies to an ordinary person who can afford one repair bill. If you depend on one device to earn a living and have no spare, work it out separately. Screen insurance for phones also needs its own calculation: first work out how often you drop the phone, then judge whether the premium is worth it. +### 12. Prefer generic drugs that passed the consistency evaluation and drugs picked in centralised procurement + +- Cost: No money. When the doctor prescribes and you go to buy, ask one extra question: "do you have the centrally procured generic version?" +- In plain language: The second batch of centrally procured drugs cut prices by 53% on average, and the deepest cut was 93%. By April 2026 there had been 11 batches, covering 490 drugs. The active ingredient is the same, so asking one extra question when you are prescribed saves you a large sum. +- Benefit: In the second batch of drugs procured by the state, prices fell by 53% on average and by up to 93%. By 2026-4 there had been 11 batches, covering 490 drugs. 1695 drugs had passed, or were deemed to have passed, the consistency evaluation. That is about 2/3 of the chemical drugs commonly used in clinical practice (China, official figures from 2020 to 2026) +- Evidence grade: A +- Sources: National Healthcare Security Administration (国家医疗保障局) (2020). Answers to Reporters' Questions on the Second Batch of State Organised Centralised Drug Procurement and Use. ; National Healthcare Security Administration (2026). State Council Policy Briefing: Improving the Full-Cycle, Full-Channel Mechanism for Forming Drug Prices. ; General Office of the State Council (国务院办公厅) (2016). Opinions on Conducting the Quality and Efficacy Consistency Evaluation of Generic Drugs（关于开展仿制药质量和疗效一致性评价的意见）(国办发〔2016〕8 号). ; formerly the State Food and Drug Administration (国家食品药品监督管理总局) (2016). Notice on Issuing Three Technical Guidelines Including Selection and Determination of Reference Preparations for Conventional Oral Solid Dosage Forms（关于发布普通口服固体制剂参比制剂选择和确定等 3 个技术指导原则的通告）(2016 年第 61 号), Annex 1 and Annex 3, reprinted by the Hainan Provincial Drug Administration. ; National Healthcare Security Administration (2024). Strict Supervision of the Quality of Centrally Procured Drugs, with full coverage on both company inspections and spot checks of product types. ; National Healthcare Security Administration (2025). Officials of the healthcare security and drug administration departments answered questions from a reporter at the People's Daily health client on issues to do with centrally procured drugs. +- Notes:Disputed: people often say "the consistency evaluation compares the ingredients, not the effect". That is half right. The evaluation is done by giving the drug to healthy volunteers, drawing blood to trace the concentration curve after the drug enters the blood, and comparing that curve with the originator's. Passing means the result falls between 80% and 125%. For drugs where a small difference in dose causes harm, the requirement is stricter. Excipients are the fillers and coatings other than the active ingredient, and they are allowed to differ from the originator's. The rules only require the same active ingredient, dosage form, way of taking it, and therapeutic action. So it is indeed not "give it to patients for a while and see whether they get better". How it performs in patients is answered by a different body of research. The National Healthcare Security Administration has had hospitals run real-world studies. By December 2024 these covered 63 centrally procured drugs, close to a hundred top-tier hospitals, and more than 300,000 patients. The results showed no difference from the originator. Metformin, for example, gave a target attainment rate for glycated haemoglobin of 83.4% in the generic group and 78.9% in the originator group; for fasting blood glucose the order reversed, at 83.8% for the generic group and 87.6% for the originator group. The two groups took turns in the lead. In January 2025 some doctors said publicly that centrally procured drugs "do not bring blood pressure down, do not put patients under, do not work as a laxative". The healthcare security and drug administration authorities checked 7 hospitals in Shanghai and concluded that most of these claims came from hearsay and personal impression, and that the laxative named had not yet started to be used in hospitals at the time. The same investigation did admit that problem drugs had been sampled. Of the more than 1600 selected products in the first 9 batches, 9 lost their qualification because of quality risk. Six of those were imported drugs, including 3 originators, and 3 were domestic. There are two things you can act on. First, if you want to keep taking the originator, it can usually still be prescribed: centrally procured drugs make up 60% to 80% of what a hospital buys, the hospital decides the rest, originators are not banned, and the price is that you pay more yourself. Second, after switching drugs for a chronic condition do not judge by feel. Follow item 3 in section 16 (write every reading in the same notebook) and keep watch over your blood pressure, blood glucose and glycated haemoglobin. If it really does get worse, go to the doctor and ask the doctor to report an adverse drug reaction, and do not stop the drug on your own. +### 13. Bind your family once in the medical insurance app, and the money in your employee insurance personal account pays for your spouse, parents and children + +- Cost: No money. Do the family mutual-aid binding once, in the medical insurance app where you are insured, or at the handling office. +- In plain language: One binding lets the money in your employee medical insurance personal account pay for medical care and medicine for your spouse, your parents and your children. Without the binding, that money can only be used by you, and it sits in the account untouched. +- Benefit: Money in the personal account can pay the part of the medical bill that you yourself must pay when you, your spouse, your parents or your children see a doctor at a designated medical institution. It can also pay the part you must pay when you buy medicine, medical devices or medical consumables at a designated pharmacy. The document also proposes exploring the use of the personal account to pay the personal contributions for family members' urban and rural resident medical insurance. +- Evidence grade: A +- Sources: General Office of the State Council (国务院办公厅) (2021). Guiding Opinions on Establishing a Mutual-Aid Guarantee Mechanism for Outpatient Care under Basic Medical Insurance for Employees（关于建立健全职工基本医疗保险门诊共济保障机制的指导意见）(国办发〔2021〕14 号), item five：「个人账户主要用于支付参保人员在定点医疗机构或定点零售药店发生的政策范围内自付费用。可以用于支付参保人员本人及其配偶、父母、子女在定点医疗机构就医发生的由个人负担的医疗费用，以及在定点零售药店购买药品、医疗器械、医用耗材发生的由个人负担的费用。探索个人账户用于配偶、父母、子女参加城乡居民基本医疗保险等的个人缴费。个人账户不得用于公共卫生费用、体育健身或养生保健消费等不属于基本医疗保险保障范围的支出。」 +- Notes: Thousands of yuan often sit in one person's personal account while an elderly relative in the same household pays for their own medicine. One binding stays valid indefinitely. Most places call this 家庭共济, and you can do it in the local medical insurance app. Note the line on the other side: the personal account cannot be used for fitness, health supplements and the like. Using it to buy health products at a pharmacy and pocket the money is insurance fraud. +### 14. Drink boiled tap water, and do not buy bottled water year after year + +- Cost: Buy a kettle. Boiling takes a little time each time, and a little gas or electricity. +- In plain language: Beijing's first-tier water price works out at 0.005 yuan a litre. Bottled water at 1 yuan for 500 millilitres works out at 2 yuan a litre. The two differ by a factor of 400. Tap water follows a mandatory national standard, and boiling it is enough in normal times. If the water supply has a genuine problem and the water smells or runs cloudy, boiling longer will not fix it: stop drinking it and report it. +- Benefit: Beijing's first-tier residential water price is 5 yuan a cubic metre, which works out at 0.005 yuan a litre. Bottled water at 1 yuan for 500 millilitres works out at 2 yuan a litre, 400 times the price of tap water. Drinking water follows the mandatory national standard GB 5749-2022. That standard was published on 2022-3-15 and came into force on 2023-4-1. The authority in charge is the National Disease Control and Administration (China, 2015 water prices and the 2022 standard) +- Evidence grade: C (disputed) +- Sources: Beijing Municipal Development and Reform Commission. Residential Water Price Table. ; State Administration for Market Regulation, National Standards Full-Text Disclosure System. GB 5749-2022 Hygienic Standard for Drinking Water（生活饮用水卫生标准）. ; Hangzhou Joint Investigation Group (2025). Report on the Investigation of Abnormal Water Supply in Some Communities in Yuhang District, Hangzhou（关于杭州市余杭区部分小区供水异常调查情况通报）, of 2025-7-23, reprinted by Xinhua. ; Yuhang District People's Government, Hangzhou. Yuhang District April/May/June 2025 Raw Water Quality Monitoring Data（余杭区 2025 年 4 月／5 月／6 月水源水水质监测数据）, published by Hangzhou Yuhang Water Holdings Group, tested by Hangzhou Jiashu Environmental Monitoring Co., Ltd., annex PDF table 1 item 24「粪大肠菌群」. , , ; Cybersecurity Bureau of the Ministry of Public Security (公安部网络安全保卫局) (2025). Fabricated a false「警情通报」and a false「粪水」rumour; Shao [某] was criminally detained. +- Notes:Disputed: the dispute is over what "meets the standard" means. The tests cover water leaving the plant and water at the end of the pipe. When the source or the network fails, the water can fall short for a short time. On 16 July 2025 the raw water arriving at the Renhe water plant in Yuhang, Hangzhou, smelt wrong, and tap water in the Renhe and Liangzhu communities smelled. The official conclusion was that algae broke down under low oxygen and produced sulphide compounds. After the source was changed, total coliforms and Escherichia coli were not detected in the water leaving the plant, and every item met the drinking water standard GB 5749-2022. Afterwards 7 officials were held accountable, July water charges were waived, and filters were replaced free of charge. But faecal coliforms from the same source had already been climbing for several months before the incident. At the Taoxi Yongsheng intake the April figure was 10 per litre, May 7300 per litre, June 16000 per litre. The limit for Class III surface water under GB 3838 is 10000 per litre. In the same month Siling Reservoir at 28000 per litre and Mantoushan Reservoir at 24000 per litre were also over the limit. Boiling kills microorganisms. It does not remove the smelly compounds, and it does not remove chemicals dissolved in the water. If the water smells odd or looks yellow and cloudy, stop drinking it, tell the water company and call 12345, and switch to bottled water for the time being. Boiling it a few minutes longer will not help. Even so, this was a local one-off incident and the local authority compensated by the occasion, so it still is not a reason to buy bottled water all year. The "sewer pipe was connected to the wrong place" story going round is a fabricated police notice, and whoever posted it has been criminally detained (see section 9). The 1 yuan for 500 millilitres is an assumed price set so the arithmetic works; recalculate with the real price in your own city. If the secondary water supply in an old building is substandard, you can add a filter, or buy bottled water only in that situation. Water prices differ by place; the Beijing figure is only there to show the scale. +### 15. Do not trade shares often + +- Cost: No money. The hard part is that when you see news you want to act, and you have to hold back. +- In plain language: The people who trade shares most often earned 11.4% a year on average, while the market as a whole made 17.9% in the same period. The more than six percentage points in between went almost entirely into the fees paid on buying and selling. Without the fees, frequent traders and infrequent traders earned about the same. +- Benefit: The study looked at 66,465 household accounts at US discount brokerages from 1991 to 1996. The most frequent traders had an annualised return of 11.4%, that is, they earned 11.4% a year on average. The market as a whole made 17.9% over the same period, and the average household made 16.4%. The average household turned over 75% of its holdings a year, replacing three quarters of the shares it held within the year. Gross return (what is earned before fees) barely differed between the frequent and infrequent groups, so the gap comes mainly from the trading costs paid when buying and selling (United States, 1991 to 1996) +- Evidence grade: A +- Sources: Barber BM, Odean T (2000). Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors. The Journal of Finance 55(2):773-806. +- Notes: This data is from the United States in the 1990s. The structure of commission and stamp duty on A-shares is different, but the principle is the same: the more you turn over, the less you keep. None of this is investment advice. +### 16. Do not borrow to invest, do not use leverage, do not buy what you do not understand + +- Cost: No money. The hard part is that in a good market you have to resist the thought of borrowing a little to earn a little more. +- In plain language: If you borrow to buy shares and the price falls through a line, and you cannot top up, you are forced to sell. A loss on the account becomes a real loss there and then, and you do not even get the chance to wait for it to come back. The regulator says a client who has held an account for less than half a year, or whose average daily assets over the last 20 trading days are below 500,000 yuan, cannot open a margin account. That threshold by itself says it is not built for ordinary people. +- Benefit: Leverage means borrowing to buy, and it multiplies your losses. Once you buy on margin, the market value of the collateral can fall below the maintenance margin ratio. The broker sets that line, and it governs what you have pledged with the broker. If the price falls through the line and you cannot add collateral to bring it back, you are force-liquidated and the broker sells for you. A loss on the account becomes a real loss there and then, with no chance to wait for it to come back. The regulator says a client「从事证券交易时间不足半年、缺乏风险承担能力、最近 20 个交易日日均证券类资产低于 50 万元」may not open a credit account. That threshold by itself says it is not a tool for ordinary people (China, 2015) +- Evidence grade: B +- Sources: China Securities Regulatory Commission (中国证监会) (2015). Measures for the Administration of Margin Financing and Securities Lending by Securities Companies（证券公司融资融券业务管理办法）(CSRC Order No. 117, Article twelve). ; Shanghai Stock Exchange. Explanation of the Implementation Rules for Margin Financing and Securities Lending Trading（上海证券交易所融资融券交易实施细则解读）, investor education material. +- Notes: "Do not buy what you do not understand" has no direct literature behind it. It is general consensus, and on its own it would be graded C. Investing with a consumer loan or by cashing out a credit card is leverage under another name; for the interest rates, see item 7 in this section (do not use the minimum payment on a credit card). None of this is investment advice. +### 17. Use a broad market index fund instead of an actively managed fund as your long-term holding + +- Cost: No money. The price is giving up the chance of picking the star fund manager. The hard part is that when the index falls you have to bear it too, and you cannot slip out halfway. +- In plain language: The shares that actively managed funds hold between them are already close to the whole market. But the extra management fee they charge comes straight out of what reaches you. If you compare fund performance by reshuffling the returns again and again to strip out luck, very few funds earn back the fee they charge through the extra return they produce. +- Benefit: The overall portfolio of actively managed US equity funds sits close to the market portfolio: the shares these funds hold between them are about the same as the whole market. But the high cost of active management turns straight into a lower net return for investors (what reaches them after fees). The study used bootstrap resampling to reshuffle returns at random again and again and see how far luck alone gets you. The result is that few funds generate excess returns large enough to cover their costs: what a fund earns above the market does not reach the fee it charges (United States, published 2010). S&P Dow Jones Indices publishes this every year: in 2024, 65% of actively managed US large-cap equity funds underperformed the S&P 500. Over the 24 years it tracks, the average was 64% underperforming a year. In the 15 years to the end of 2024, there was no category in which a majority of active funds beat their index +- Evidence grade: B +- Sources: Fama EF, French KR (2010). Luck versus Skill in the Cross-Section of Mutual Fund Returns. The Journal of Finance 65(5):1915-1947. ; S&P Dow Jones Indices (2025). SPIVA U.S. Scorecard Year-End 2024. (the site refuses automated fetching; the figures were checked against the archive of 2025-05-12: ) ; S&P Dow Jones Indices (2025). SPIVA Asia Ex-Japan Scorecard Mid-Year 2025 (Report 1a, the underperformance of China Large-Cap against the S&P China A 300: 1 year 60.5%, 3 years 69.6%, 5 years 50.8%). (the site refuses automated fetching; checked against the archive) ; Harvey CR, Liu Y (2022). Luck versus Skill in the Cross Section of Mutual Fund Returns: Reexamining the Evidence. The Journal of Finance, 77(3). (the dissenting side) +- Notes:Disputed: one paper has re-examined this conclusion. It argues that the Fama-French method of resampling the sample again and again does not resample enough. On its algorithm, even a fund whose extra return is genuinely obvious can be judged as indistinguishable from luck. So "almost no fund has skill" may be too strong. China's figures are not as lopsided as America's. S&P Dow Jones figures to June 2025 put the share of actively managed A-share large-cap funds underperforming the S&P China A 300 at 60.5% over 1 year, 69.6% over 3 years and 50.8% over 5 years. This series only goes back 5 years, so there is no longer figure. In China then, whether active funds underperform the index over the long run is still unsettled, but the low fees of an index fund hold either way. None of this is investment advice. +### 18. Among funds of the same kind, pick the one with the lowest fees + +- Cost: No money. Before you buy, take one more look at what the management fee, the custody fee and the sales service fee come to together. +- In plain language: A fee one percentage point higher a year leaves you about 18% less after 20 years and about 26% less after 30. Among funds of the same kind, the lowest fee is the only thing about this you can be certain of in advance. +- Benefit: The difference in fees eats into the principal bit by bit through compounding, which means the slice taken each year builds on itself year after year. A fee 1 percentage point higher a year cuts the final value by 0.99^20 ≈ 18.2% after 20 years, where the final value is what is left in the account at that point, and by about 26% after 30 years. China has run fee reform on public funds since 2023, lowering the fee ceiling for actively managed equity funds (the kind that mostly buy shares and leave the choosing to a fund manager). From 7 July 2023, a newly registered fund of that type charges a management fee of no more than 1.2% and a custody fee of no more than 0.2%. 136 fund companies have also brought the same two fees on their older products down below this line +- Evidence grade: C +- Sources: China Securities Regulatory Commission (中国证监会) (2023). The CSRC Official in Charge Answers Reporters' Questions on Public Fund Fee Reform. ; Xiamen CSRC Bureau (厦门证监局) (2024).「新注册的主动权益类基金统一执行"管理费率不超过1.2%、托管费率不超过0.2%"的上限标准」. ; Fama EF, French KR (2010). See item 11 in this section. +- Notes: The calculation is just the formula worked through, and it assumes both funds earned the same before fees. In reality you also have to count the subscription fee, the redemption fee and the sales service fee on a class C share. None of this is investment advice. +### 19. Do not put all your money into one share, one platform or one property + +- Cost: No money. The hard part is accepting one thing: you will not happen to pick exactly the one that does best. +- In plain language: If several things do not rise and fall in perfect lockstep, spreading your money leaves the average return unchanged and makes the account swing less. Put the other way round, if everything is in one share, one platform or one property, then when it goes wrong everything about you goes wrong. +- Benefit: As long as several assets do not rise and fall in perfect lockstep, spreading your money leaves the expected return (what you can expect to earn) unchanged and makes the volatility (how far the account moves up and down) smaller. This is the basic result on portfolio choice put forward by Markowitz in 1952, and it has been used ever since. The other way round, if everything is in one share or one platform, its variance is your entire variance, and it moves as much as you do. When it goes wrong there is no hedge, and no other asset can make good the loss +- Evidence grade: B +- Sources: Markowitz, H. (1952). Portfolio Selection. The Journal of Finance, 7(1), 77–91. +- Notes: It is graded B because that classic paper is pure mathematical derivation and does not work out how much spreading saves you in money. Spreading only makes the account move less; it does not promise that you will not lose. Buying a broad market index fund (see item 17 in this section) is itself the cheapest way to spread. None of this is investment advice. +### 20. If you pay income tax, open a personal pension account: 12000 yuan a year comes off before tax. If you pay no income tax, the account does not pay + +- Cost: Opening the account is free. The real cost is that the money is locked until you retire and you cannot take it out in between. +- In plain language: For every yuan you put in, the tax you save is your own rate band minus 3%. With the full 12000 yuan in a year, someone at 10% saves 840 yuan, at 20% 2040 yuan, at 30% 3240 yuan. Someone who owes no income tax saves nothing at all and pays 3% on withdrawal, a net loss. You cannot get the money out until you retire. +- Benefit: Money paid in can be deducted truthfully from comprehensive income or from business income, up to 12000 yuan a year: whatever you pay in, you can subtract from the income you are taxed on. The return this money earns in the account is not taxed with individual income tax for now. When you draw it later it is not merged into comprehensive income, and is taxed separately at 3%. So what you save is your marginal tax rate minus 3%, multiplied by what you actually paid in. The marginal rate is the band that applies to each extra yuan you earn. Worked out: someone at a marginal rate of 10% saves at most 840 yuan a year, at 20% 2040 yuan, at 30% 3240 yuan. Someone who owes no individual income tax at all (their annual comprehensive income, after every deduction, stays below the threshold) saves nothing when they pay in and pays 3% when they draw out, a net loss. The annual contribution limit accumulates by calendar year and starts again the following year; anything you do not use in one year does not carry over. This policy has applied nationwide since 2024-12-15 (nationwide) +- Evidence grade: A +- Sources: Ministry of Finance and State Taxation Administration (财政部、税务总局) (2024). Announcement on Implementing Preferential Individual Income Tax Treatment for Personal Pensions Nationwide（关于在全国范围实施个人养老金个人所得税优惠政策的公告）(Announcement of the Ministry of Finance and the State Taxation Administration No. 21 of 2024, of 2024-12-12). ; Ministry of Human Resources and Social Security and four other departments (2022). Implementation Measures for Personal Pensions（个人养老金实施办法）(Articles seven, eight, nine, twelve and thirteen). +- Notes: The account is "closed-loop": once the money is in, you cannot take it out. There are only four ways to draw on it: reaching the age at which you receive the basic old-age pension, total and permanent loss of the capacity to work, settling abroad or in Hong Kong, Macau or Taiwan, and other situations the state specifies. So it is not a saving product; it is a tax arrangement locked up for years. Work out first what you pay on each extra yuan you earn, then decide how much to pay in. Do not lock up your emergency fund to catch a bargain (for the emergency fund, see item 27 in this section). The money in the account still has to be invested in a product you choose yourself, and it can lose money. The rules for choosing that product are the same as in items 17 and 18 in this section (a broad market index fund, prefer low fees). The deduction certificate is issued by the personal pension information management service platform. You can choose when to take the deduction: either when tax is withheld from your pay in the year, or at the settlement the following year (see item 2 in this section for the settlement). +### 21. Pay per visit at the gym, or buy a short term, unless you have over a year of steady attendance on record + +- Cost: Paying per visit costs more each time than an annual card spread out. You have to estimate honestly how many times a month you really go. +- In plain language: In the three years of American data, members who chose a contract costing more than 70 dollars a month went only 4.3 times a month on average, which works out at more than 17 dollars a visit. A card for 10 visits cost 10 dollars a visit. These people paid about 600 dollars more over the length of their membership. Unless you already have over a year of steady attendance on record, paying per visit or buying a short term is cheaper. +- Benefit: The study looked at three years of data on 7,752 members of 3 American health clubs. Members who chose a contract costing more than 70 dollars a month went only 4.3 times a month on average, which works out at more than 17 dollars a visit. A card for 10 visits cost 10 dollars a visit. These people paid about 600 dollars more over the length of their membership. Separately, monthly members who had not cancelled after more than a year were 17% more likely than annual members to still be there (United States, three years of panel data) +- Evidence grade: B +- Sources: DellaVigna S, Malmendier U (2006). Paying Not to Go to the Gym. American Economic Review 96(3):694-719. +- Notes: This is one study, and it used American data. But the reasoning behind it travels: people overestimate how much self-control they will have in future. If you already have a record of going twice a week or more for a full year, an annual card really is cheaper. +### 22. Get a written contract when you pay money in advance; if the business is at risk stop paying, and before it disappears you can ask for the unspent balance back + +- Cost: No money. The hard part is insisting on a written contract at the counter and refusing "top up now and sort it out later". +- In plain language: When a merchant takes your money in advance, it must sign a written contract with you and set out how a refund works. When a merchant runs into serious trouble it must stop taking money, and it must tell you in advance if it is closing or moving. At that point you can demand that it carries on providing the goods or service, or that it refunds the balance you have not spent. A merchant that breaks these rules is fined. +- Benefit: The regulations say that a merchant taking advance payments must enter into a written contract and agree in it how refunds work. A merchant facing serious operational risk must stop taking advance payments. A merchant deciding to close or move must tell consumers in advance. A consumer has the right to require it to carry on performing, or to have the unspent balance refunded. A breach can be fined 1 to 10 times the illegal gains; where there are no illegal gains, the fine is up to 500,000 yuan +- Evidence grade: A +- Sources: State Council (国务院) (2024). Implementing Regulations for the Consumer Rights Protection Law of the People's Republic of China（中华人民共和国消费者权益保护法实施条例）(State Council Order No. 778, in force from 2024-7-1). Article twenty-two：「经营者以收取预付款方式提供商品或者服务的，应当与消费者订立书面合同，约定商品或者服务的具体内容、价款或者费用、预付款退还方式、违约责任等事项。」「经营者未按照约定提供商品或者服务的，应当按照消费者的要求履行约定或者退还预付款。」「经营者出现重大经营风险，有可能影响经营者按照合同约定或者交易习惯正常提供商品或者服务的，应当停止收取预付款。经营者决定停业或者迁移服务场所的，应当提前告知消费者……消费者……有权要求经营者继续履行提供商品或者服务的义务，或者要求退还未消费的预付款余额。」; penalties under Article fifty +- Notes: The regulations give you grounds for a lawsuit afterwards; they do not stop anything in advance. If the merchant really disappears, the money has usually gone from the account. So top up less, following item 21 in this section (pay per visit at the gym, or buy a short term), and use these regulations to recover your money if it really goes wrong. Article twenty-seven of the same regulations also says that where a merchant takes a deposit it must agree in advance on the method, procedure and time limit for returning it, and must not set unreasonable conditions on the return. +### 23. Set a 24-hour cooling-off period on large non-essential purchases, and use the seven-day no-questions return when you shop online + +- Cost: No money. You get what you wanted a day later instead. You usually pay the return postage yourself. The hard part is not ordering on the day. +- In plain language: What you save is the things you no longer want the next day. Anything bought online can be returned without giving a reason within seven days of receiving it. A few kinds cannot: items made to your specification, fresh goods and things that spoil easily, unsealed digital goods, and newspapers and magazines. +- Benefit: What you save is the part of an impulse buy that you no longer want the next day. There is no single proportion for how much that is. Goods bought online can be returned without a reason within seven days of receipt. There are four exceptions: items made to order, fresh and perishable goods, unsealed digital goods, and newspapers and magazines (China, in force since 2017) +- Evidence grade: C +- Sources: State Administration for Market Regulation (国家市场监督管理总局) (2020, amended). Interim Measures for the Seven-Day No-Questions Return of Goods Purchased Online（网络购买商品七日无理由退货暂行办法）(SAMR Order No. 31). +- Notes: The cooling-off period itself is a rule of thumb with no direct literature behind it. Where the cooling-off period does not stop you, the seven-day no-questions return is there to use. Before you return anything, check two things: whether the item is on the list that cannot be returned, and whether the packaging is still intact. +### 24. Do not stock up for big sales or for a struck-through "original price" + +- Cost: No money. The price is that you also let some genuine discounts go. You have to remember the everyday price of the few things you buy often. The hard part is not stockpiling on the day of the sale. +- In plain language: People judge whether something is expensive by the first number they see, and later adjustments do not go far enough to correct it. So "was 999, now 499" systematically makes you overstate how cheap it is. By law, the only thing a discount may be measured against is the lowest price the same shop actually sold at in the seven days before the promotion. The crossed-out price does not necessarily reflect a real sale. +- Benefit: People's judgement gets fixed by the first number they see, and later adjustment usually does not go far enough. So "was 999, now 499" systematically raises your estimate of how much you are saving. The regulator says the baseline for a markdown must be the lowest price actually sold by the same shop in the seven days before the promotion, and the depth of the discount is measured against the lowest price that shop really sold in those seven days. The crossed-out price does not necessarily reflect a real sale (China, in force since 2020) +- Evidence grade: C +- Sources: Tversky A, Kahneman D (1974). Judgment under Uncertainty: Heuristics and Biases. Science 185(4157):1124-1131. ; State Administration for Market Regulation (2020). Interim Provisions on Regulating Promotional Conduct（规范促销行为暂行规定）(SAMR Order No. 32, Article twenty-one). +- Notes: The anchoring effect is a laboratory finding, and nobody has worked out how much money stockpiling actually costs you. Stockpiling also carries costs you will not enter in your accounts: things going out of date, taking up space, and "there is some at home, so I will use a bit more". Stock up only in the three cases that all hold: a long shelf life, certain use, and a unit price genuinely below what it used to be. +### 25. Buy the pure protection type of insurance first, and treat the "return" and "dividend" parts as not guaranteed + +- Cost: No money. A pure protection policy pays nothing back at the end, and it nags at you that the money went for nothing. You have to arrange the saving yourself. The hard part is accepting that this money will not come back. +- In plain language: The insurer does not guarantee the dividend and return part of your money. The policy document has to state in bold that future policy dividends are non-guaranteed benefits. The extra premium you pay for a return policy is money handed to the insurer to hold for you, with no guaranteed return and no easy way to take it out. +- Benefit: The regulator's rules require the policy document of a participating product to warn in bold that「未来的保单红利为非保证利益，其红利分配是不确定的」. The illustration of benefits is the projection table the insurer shows you, and its spread limit is 4.5% minus the product's assumed interest rate. The assumed rate is the return fixed in the contract. The extra premium in a return policy or a participating policy is a form of saving held on your behalf by the insurer. Its return is not guaranteed, its liquidity is poor, and taking it out is not easy (China, in force since 2023) +- Evidence grade: C +- Sources: formerly the China Banking and Insurance Regulatory Commission (中国银保监会) (2022). Rules on Information Disclosure for Personal Insurance Products of One Year or More（一年期以上人身保险产品信息披露规则）(银保监规〔2022〕24). +- Notes:Disputed: for a person who cannot save at all, a return policy forces them to keep paying on schedule, and that is worth something in practice. The advice above is aimed at an ordinary person who can control their own spending. How much better "buy the protection type and invest the premium you saved yourself" really is than "buy a return policy" depends on the product, and this section has no systematic data to cite. None of this is investment advice. +### 26. Buy enough third-party liability cover: the compulsory limit is uniform nationwide and low, and anything above it comes out of your own pocket + +- Cost: Commercial third-party liability insurance costs several hundred to a thousand yuan a year. The more cover you buy, the cheaper the premium works out per unit of cover. +- In plain language: When you are at fault, compulsory insurance pays at most three things: 180,000 yuan for death and disability, 18,000 yuan for medical expenses, and 2,000 yuan for property damage. They are counted separately and cannot be moved between one another. The bill for injuring someone easily goes past these limits, and the excess comes out of your own household budget, so buy enough commercial third-party cover. +- Benefit: Compulsory traffic accident liability insurance applies a uniform liability limit nationwide. The limit is worked out as four separate items: death and disability, medical expenses, property damage, and the compensation payable when the insured is not at fault. Since the 2020 reform of motor insurance, the total liability limit in force is 200,000 yuan. Within that 200,000 yuan, death and disability is 180,000 yuan, medical expenses 18,000 yuan and property damage 2,000 yuan. When you are not at fault, death and disability is 18,000 yuan, medical expenses 1,800 yuan and property damage 100 yuan. Anything above the limit is borne by the party at fault, so if you hit someone you pay it yourself. A vehicle with no compulsory insurance is impounded, and you are fined 2 times the insurance premium you should have paid +- Evidence grade: A +- Sources: State Council (国务院) (2006, amended 2012). Regulations on Compulsory Traffic Accident Liability Insurance for Motor Vehicles（机动车交通事故责任强制保险条例）(State Council Order No. 462, as amended by State Council Order No. 618) Article twenty-three：「机动车交通事故责任强制保险在全国范围内实行统一的责任限额。责任限额分为死亡伤残赔偿限额、医疗费用赔偿限额、财产损失赔偿限额以及被保险人在道路交通事故中无责任的赔偿限额。」「责任限额由保监会会同国务院公安部门、国务院卫生主管部门、国务院农业主管部门规定。」Article twenty-one, on compensation within the limits; Article thirty-nine, on having no insurance「由公安机关交通管理部门扣留机动车……处依照规定投保最低责任限额应缴纳的保险费的2倍罚款」. ; China Banking and Insurance Regulatory Commission (中国银行保险监督管理委员会) (2020). Guiding Opinions on Implementing the Comprehensive Reform of Motor Insurance（关于实施车险综合改革的指导意见）(银保监发〔2020〕41), item four, sub-item four：「将交强险总责任限额从12.2万元提高到20万元，其中死亡伤残赔偿限额从11万元提高到18万元，医疗费用赔偿限额从1万元提高到1.8万元，财产损失赔偿限额维持0.2万元不变」. (State Council Gazette) +- Notes: The regulator sets the limits and they change. These figures are the values in force since the September 2020 motor insurance reform, so check them again before you buy. What matters is the structure: compulsory insurance pays item by item, and an unused balance in one item cannot be moved to cover another. The total bill for injuring someone can go far past it, and no insurance covers the difference. This book does not give a number for how much third-party cover to buy. But it is one of the few places in this book where spending several hundred yuan removes a small chance of a loss that would crush a household. +### 27. Save 3 to 6 months of living costs as an emergency fund, and keep it where you can withdraw at any time + +- Cost: No money. The price is that this money gives up a higher return and can only sit in a demand account, a money market fund, or a deposit you can withdraw from at any time. The hard part is not touching it while you build it up. +- In plain language: Put 3 to 6 months of living costs somewhere you can withdraw from at any time. Then when you lose your job or fall ill you do not have to borrow money at 18% a year, and you do not have to sell your investments at a loss. Money in a bank is covered by deposit insurance, which pays up to 500,000 yuan per bank. +- Benefit: With an emergency fund, when you lose your job or fall ill you do not have to borrow, and you do not have to sell investments at a loss. For the annualised rates on consumer loans and the minimum payment on a credit card, see item 7 in this section (do not use the minimum payment on a credit card). Money in a bank is protected by deposit insurance, which pays up to 500,000 yuan per bank (China, 2015) +- Evidence grade: C +- Sources: State Council (国务院) (2015). Regulations on Deposit Insurance（存款保险条例）(State Council Order No. 660, Article five). +- Notes: "3 to 6 months" is a rule of thumb people use, and no official or academic original source was found for it, so it is graded C. Work out the amount from your own fixed monthly spending, not from your income. None of this is investment advice. +### 28. Do the comparison before repaying a mortgage early; do not go on feel + +- Cost: No money. It takes half an hour to do the sums. You have to estimate honestly the steady after-tax return you can actually get. +- In plain language: Repaying a mortgage early is a certain gain of your mortgage rate. So there is only one question to ask: is the after-tax return you can get steadily over the long run higher than your mortgage rate? If it is, do not rush to repay; if it is not, or you cannot tell, then repay. With equal principal and interest, interest takes the larger share in the early years, so the earlier you repay the more you save. The "early repayment calculation" in your bank's app gives you the figure directly. +- Benefit: The arithmetic works like this. First set r as the annual interest rate on your mortgage, as your contract states it. On 2026-8-20 the 5-year LPR above is 3.5%, and the rate actually applied is that LPR plus or minus a spread. Then set g as the annualised after-tax return on the other investment you could choose: the proportion of this money that you can expect to earn in a year on average by putting it somewhere else. The certain return on repaying early is r, and each payment you make is an r earned for certain. If g is reliably greater than r, keeping the money invested does better. If g is below r, or if you cannot pin g down at all, repaying early locks in a risk-free r. With equal principal and interest, interest takes the larger share of your early payments, so the earlier you repay the more interest you save. You can read the figure straight off the bank's app under "early repayment calculation" (China, 2026 rates) +- Evidence grade: C +- Sources: National Interbank Funding Center, authorised by the People's Bank of China (2026). Announcement of the Loan Prime Rate of 2026-8-20（2026 年 8 月 20 日贷款市场报价利率（LPR）公告）. +- Notes: Only the method is given here, and no conclusion. It is rare for an investment return to beat the mortgage rate reliably. Whichever side you choose, set aside the emergency cash first. Also check whether the contract has a penalty clause for repaying early. None of this is investment advice. +### 29. Trust the platform's rules and the statute when you shop online, not the streamer or the "positive reviews" + +- Cost: No money. Read the after-sales terms and the deposit rules once more before you order. After the goods arrive, keep the order, screenshots of the livestream replay and your chat logs. The hard part is not being carried away by the streamer's manner. +- In plain language: If a merchant deceives you, they pay three times the price, and where the price is under 500 yuan the compensation is counted as 500 yuan. A platform that knows a seller is infringing and does nothing has to pay alongside them. A deposit may be at most 20% of the contract value; if you break the contract the deposit is not returned, and if the seller breaks it you get it back double. Fake orders and fake positive reviews in a livestream room make the operator liable, and killing the newcomer with big data is expressly forbidden. +- Benefit: Where an operator commits fraud, it pays three times the price, and where the price is under 500 yuan the compensation is counted as 500 yuan. A platform that knows or should know that a seller is infringing and does not act bears joint liability, and you can claim that compensation from the platform directly. A livestream room and a streamer may not invent transactions, follows, views or likes, and may not publish false information; where this causes harm they bear civil liability. Organising fake transactions or fake reviews to promote someone is fined up to 1,000,000 yuan; where the circumstances are serious, 1,000,000 to 2,000,000 yuan, and the licence may be revoked. A deposit may not exceed 20% of the contract value. If you do not perform, the deposit is not returned; if the seller does not perform, it is returned to you double. Where automated decisions are taken using personal information, unreasonable differential treatment on terms such as the transaction price is forbidden, and that is what is commonly called killing the newcomer with big data (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2013, amended). Consumer Rights Protection Law（消费者权益保护法）(Articles forty-four and fifty-five). (reprinted by the State Administration for Market Regulation) ; Cyberspace Administration of China and six other departments (2021). Interim Measures for the Administration of Livestream Marketing（网络直播营销管理办法（试行））(Articles eighteen and twenty-eight). ; Standing Committee of the National People's Congress (2025, amended). Law Against Unfair Competition（反不正当竞争法）(Articles nine and twenty-five). (reprinted by the National Social Security Fund Council) ; National People's Congress (全国人大) (2020). Civil Code（民法典）(Articles five hundred and eighty-six and five hundred and eighty-seven). ; Standing Committee of the National People's Congress (2021). Personal Information Protection Law（个人信息保护法）(Article twenty-four). (reprinted by the National Bureau of Statistics) ; Shandong Provincial Administration for Market Regulation (山东省市场监管局) (2025). Six Typical Cases of Unfair Competition on the Internet（六起网络不正当竞争典型案例）. +- Notes: There is no official comparative data to cite on which platform gives the best value or is least likely to sell you poor goods, so this book does not rank platforms. All that can be checked is the overall spot-check result (see item 32 in this section, checking the national spot-check notices). After the 2025 amendment, false advertising moved from the old Article eight to Article nine. The penalties moved from Article twenty to Article twenty-five. The new law applies from 15 October 2025. Do not copy the old article numbers into anything you write. In that batch of Shandong cases, one company organised 75 fake transactions through "hired hands" on a principal of more than 680,000 yuan and was caught; the notice does not state the fine. For the seven-day no-questions return, see item 23 in this section (use the seven-day no-questions return when you shop online). +### 30. If something bought in a livestream room goes wrong, ask the platform for the seller's and the seller's agent's details, and the platform must give them + +- Cost: No money. All it takes is to make the request to the platform. +- In plain language: The most common dead end when something bought in a livestream room goes wrong is that you cannot find out who to sue. At that point you can require the platform to give you the details of the livestream room operator and the person selling, and the relevant business records. The platform has to provide them. +- Benefit: Where a consumer dispute arises, the operator of a livestream marketing platform must, at the consumer's request, provide information on the livestream room operator and the livestream marketing staff, together with the relevant business activity records. The hardest part of enforcing your rights after a livestream purchase is finding out who to sue, and once you have these details you know who to sue +- Evidence grade: A +- Sources: State Council (国务院) (2024). Implementing Regulations for the Consumer Rights Protection Law of the People's Republic of China（中华人民共和国消费者权益保护法实施条例）(State Council Order No. 778, in force from 2024-7-1). ：「经营者通过网络直播等方式提供商品或者服务的，应当依法履行消费者权益保护相关义务。直播营销平台经营者应当建立健全消费者权益保护制度，明确消费争议解决机制。发生消费争议的，直播营销平台经营者应当根据消费者的要求提供直播间运营者、直播营销人员相关信息以及相关经营活动记录等必要信息。」「直播间运营者、直播营销人员发布的直播内容构成商业广告的，应当依照《中华人民共和国广告法》的有关规定履行广告发布者、广告经营者或者广告代言人的义务。」 +- Notes: Record the screen and the streamer's promises before you order. When something goes wrong, ask the platform for the seller's and the seller's agent's details first, and then decide whether to complain or to sue. The same regulations set out three more things. A merchant may not widen the range of goods treated as excluded from the seven-day no-questions return. They may not make "not covered by the no-questions return" the default option. And they may not refuse your return without your confirmation. +### 31. Unsafe food: besides a refund you can ask for ten times the price, and where the additional compensation is under 1,000 yuan it is counted as 1,000 yuan + +- Cost: No money. Keep the food, the packaging and the receipt. Do not throw them away and do not wash them. Take photographs at once. Go to the merchant or the platform first, and call 12315 if you cannot settle it. If you end up suing it takes months. The hard part is resisting the urge to throw the goods away first. +- In plain language: If the food you bought is past its date or has gone off, the goods themselves, the packaging and the receipt are your evidence, so do not rush to throw them out. You can claim ten times the price or three times the loss, and where that is under 1,000 yuan it is counted as 1,000 yuan. Whether you approach the seller or the manufacturer is your choice, and whichever one you approach must pay first without passing you on. +- Benefit: Where you are harmed by food that does not meet food safety standards, you may claim compensation for the loss from the operator, and you may also claim it from the producer. Whoever receives your compensation claim is subject to the first-responsibility rule and must pay first without evasion. That means whoever you approach has to pay you first and cannot pass you to the other party. Besides compensation for the loss, you may also require the other party to pay punitive compensation of ten times the price or three times the loss, and「增加赔偿的金额不足一千元的，为一千元」. There is one exception:「食品的标签、说明书存在不影响食品安全且不会对消费者造成误导的瑕疵的除外」. The judicial interpretation sets out several further rules. First, three conditions have to be met together: the purchase was for personal or household consumption, the food does not meet food safety standards, and there is no evidence that the buyer knew it did not. Meet them and「以其实际支付价款为基数计算价款十倍的惩罚性赔偿金」. Second, where someone buys knowing it does not meet the standards and then claims, the court supports the claim only「在合理生活消费需要范围内」. Buying repeatedly and suing repeatedly is assessed by total purchases within that one range. Third, there are several defects for which the court will not award punitive compensation, provided they cannot mislead the consumer about food safety: irregular lettering or type size, with typos or missing or extra characters; a non-standard way of stating net content or specification; and no storage conditions marked on a food that has no special storage requirement (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (2025, 3rd amendment). Food Safety Law（食品安全法）(Article one hundred and forty-eight). (National Database of Laws and Regulations, in force from 2025-12-1) ; Supreme People's Court (2024). Interpretation on Several Questions of Law Concerning the Application of Law in Cases of Disputes over Punitive Compensation for Food and Drugs（关于审理食品药品惩罚性赔偿纠纷案件适用法律若干问题的解释）(法释〔2024〕9, Articles one, six, seven, eight, twelve, thirteen, fourteen, fifteen and sixteen, in force from 2024-8-22). (Gazette of the Supreme People's Court) +- Notes: The order of the evidence matters more than the amount. Photograph and video first, capturing the packaging, the production date, the foreign object and the purchase record, and only then contact the merchant or the platform. Do not throw the goods away or wash them before you have spoken to anyone. What 12315 does is mediate; it does not rule. If the merchant refuses to admit anything, suing is the only road left; for how small claims are counted, see item 22 in section 8 (online shopping and second-hand transactions where you were deceived). Also note: whether a hair or an insect found in the food counts as failing food safety standards is usually for the market regulation authority to determine or for an appraisal to settle, and it is not for you to decide. So the claim that finding one hair entitles you to 1,000 yuan does not hold. Fraud in an ordinary good is compensated at three times the price, and under 500 yuan is counted as 500 yuan; see item 29 in this section (trust the platform's rules and the statute). Selling food you make yourself has its own rules on which certificates to register for, how to write the label, and why you may not claim it treats disease; for the details see items 8, 9, 10 and 11 in section 12 (food licensing, labelling, the line on therapeutic claims, and the criminal line). The other way round, making a business out of compensation claims has clear red lines. First, deliberately manufacturing a false picture that a merchant is operating unlawfully and then using complaints or lawsuits to demand money is suspected of extortion, and the court must refer the clues and materials to the police. Second, where a lawsuit is brought on the basis of a manufactured picture, the court dismisses the claim; where it constitutes false litigation, a fine or detention is imposed according to the circumstances, and the merchant may in turn sue you for infringement of reputation. The beneficiary tier for this item is you and your family. +### 32. Before buying anything large, check the national spot-check notices, 3C certification and the energy efficiency label + +- Cost: No money. It takes a few minutes on the websites of the State Administration for Market Regulation and the certification body. +- In plain language: In 2024 the national spot check covered 25,250 batches of products, and 14.3% failed. Online channels were worse: 7297 batches were checked and 23.5% failed, against 15.1% in physical shops. Before buying anything large, spend a few minutes looking up the spot-check notice and then check whether the 3C certificate is genuine. Where an energy efficiency label is required, it must be there. +- Benefit: In 2024 the national supervision spot check covered 25,250 batches of products, with a failure rate of 14.3%. Of these, 7297 batches were taken from online channels with a failure rate of 23.5%, 1.8 percentage points higher than the year before. 8055 batches came from physical shops with a failure rate of 15.1%. Certificates for compulsory product certification (3C) can be checked for authenticity in the "certification results" column on the CNCA website. Energy efficiency labels are managed under a mandatory catalogue, and the label must carry the energy efficiency grade, the efficiency index and an energy efficiency information code. Forging or misusing an energy efficiency label, or failing to put one on where it is required, is punished under the Energy Conservation Law (nationwide) +- Evidence grade: B +- Sources: State Administration for Market Regulation (市场监管总局) (2025). Report on the National Supervision Spot Check of Product Quality in 2024. ; Certification and Accreditation Administration of the People's Republic of China (国家认监委). The "certification results" search on the official website. ; National Development and Reform Commission and formerly the General Administration of Quality Supervision, Inspection and Quarantine (2016). Measures for the Administration of Energy Efficiency Labelling（能源效率标识管理办法）(Order No. 35, Articles three, seven and twenty-seven). +- Notes: A spot check takes a selection of goods chosen by the regulator, so the result is not the pass rate of the whole market, and it cannot be used to rank platforms. Nobody has studied how many fewer poor goods you would buy by checking, so this is graded B. The CNCA search platform cx.cnca.cn would not open when this section was checked; take cnca.gov.cn on the front page as the way in. The figures in the spot-check notices are updated every year, so look up the most recent issue when you use them. +### 33. Bead bracelets, jade, wristwatches and designer toys: count them as money spent, not as money saved + +- Cost: No money. Just keep "it will be worth more later" out of your reasons for buying. The hard part is not adding to the budget for the sake of keeping its value. +- In plain language: The price of this kind of thing already includes a slice of consumption tax: 20% on a watch of 10,000 yuan or more, 5% on gold, silver and platinum jewellery and on diamonds, 10% on other jewellery and jade, with value-added tax counted separately. You do not get that tax back if you sell it on. If you genuinely like it, buy it as consumption and do not count it as saving. +- Benefit: Start on the buying side. The Consumption Tax Item and Rate Schedule sets the rates like this. "High-end watches" 20%. Under "precious metals, precious stones and jewellery", gold and silver jewellery, platinum jewellery, diamonds and diamond ornaments 5%. Other precious jewellery, precious stones and jade 10%. "High-end watch" is defined as「销售价格（不含增值税）每只在 10000 元（含）以上的各类手表」. Now the realising side, under the Measures for the Administration of Pawnshops. Article thirty-six: the pawn period may not exceed 6 months. Article thirty-seven deals with the pawn interest rate, which is the interest on the money the pawnshop lends you. That interest is worked out from the statutory loan interest rate for a 6-month term published by the People's Bank of China and the length of the pawn period. Interest may not be withheld in advance and taken out of the money before it is handed over. Article thirty-eight: for a pawn of movables (pledge a watch, a piece of jade, that kind of thing), the all-in monthly fee may not exceed 42‰ of the principal, and the separate all-in monthly charge may not exceed 4.2% of the amount borrowed. Real property mortgages are 27‰ and pledges of property rights are 24‰. Article forty: if within 5 days of the term ending you neither redeem the pawn nor extend it, the pawn is forfeited and the goods no longer wait for you. Article forty-three: forfeited goods with an appraised value above 30,000 yuan may be entrusted to auction. The auction proceeds have the auction costs and the principal and interest deducted first. The remainder goes back to the pawnor, and if it is not enough the pawnor is pursued for the shortfall (nationwide) +- Evidence grade: A +- Sources: State Council (国务院) (2008). Interim Regulations of the People's Republic of China on Consumption Tax（中华人民共和国消费税暂行条例）(State Council Order No. 539, with the Consumption Tax Item and Rate Schedule attached). ; Ministry of Finance and State Taxation Administration (财政部、国家税务总局) (2006). Notice on Adjusting and Improving Consumption Tax Policies（关于调整和完善消费税政策的通知）(财税〔2006〕33). ; Ministry of Commerce and Ministry of Public Security (商务部、公安部) (2005). Measures for the Administration of Pawnshops（典当管理办法）(Order No. 8 of 2005, Articles thirty-six, thirty-seven, thirty-eight, forty and forty-three). +- Notes: Work it out properly when you pledge something to a pawnshop: the all-in monthly charge is capped at 42‰ of the amount borrowed, which is about half the borrowed amount over a year, and the interest on the loan is on top of that. This "half a year" figure is a rough calculation made by multiplying the monthly rate straight by 12, and it is there to show the scale, not to be an exact annual rate. The tax rates and the charges are all set out in writing. But there are no official statistics to cite on whether this kind of thing appreciates over the long run, so this book makes no forecast and comments on no category. The only thing to stop is the reasoning that treats consumption as investment: liking it and wearing it happily is fine to buy as consumption; what loses money is adding budget on the grounds of "keeping its value" or "it will sell for more later". Gold splits two ways too: investment bars traded by weight and jewellery sold by the piece are not the same thing. A jewellery price contains design and workmanship charges, and these are usually not counted when you sell it back. A listing price on a second-hand platform is not a transaction price, and should not be read as a market rate. +### 34. For jade and gemstones, accept only a test report carrying the CMA mark, and check the laboratory on the issuing authority's website + +- Cost: No money. Check the qualifications of the issuing laboratory once before you order. It takes a few minutes. +- In plain language: Certificates can be forged, so what you need to check is the laboratory that issued the report. An inspection institution that issues reports to the public must be accredited, and the report must carry the CMA mark. The issuing authority publishes the list of institutions on its website, so you can check whether this one is on it. In an official case that was handled, the "national inspection certificate" supplied with the goods had been made by the seller's own contact. +- Benefit: Start with the rules, under the Measures for the Administration of Qualification Accreditation of Inspection and Testing Institutions. Article two: qualification accreditation is a licensing and appraisal of an inspection and testing institution that「向社会出具具有证明作用的数据、结果」: it is examined first, and only then is it given the qualification to issue certificates. Article fifteen: the qualification accreditation mark consists of the CMA design and the number of the qualification accreditation certificate. Article twenty-one: an institution that issues certifying data and results to the outside world must mark the qualification accreditation mark on its inspection and testing report. Article twenty-two: the accreditation authority must publish on its official website the information on institutions that have obtained accreditation, together with the status of the certificate. Article thirty-four: an institution that, without having obtained accreditation in accordance with the law, issues certifying data and results to the outside world is fined 30,000 yuan. Article thirty-seven: forging, altering or misusing an accreditation certificate or mark is fined up to 30,000 yuan. Now the official cases, taken from the typical cases of "cracking down on fakes and clearing out the sources" in the traditional crafts market published by the State Administration for Market Regulation on 2026-6-4. One: a livestream room in Donghai, Jiangsu claimed the goods were "natural, no acid wash, no optimisation, grade A Burmese". The room also supplied a "national inspection certificate". The test results came back as quartzite jade (dyed), jadeite (treated), glass and plastic. That certificate had been produced by the party's own contact. Two: in the same county another case seized 469 items suspected of being fake natural crystal, and testing found 210 to be natural crystal and 259 synthetic crystal. Three: a shop in Zhenping, Henan privately made a display board for livestreaming. The board read「GIA CMA 中玉珠宝玉石鉴定中心认证合作单位」. The shop had also bought 181 "National Jewellery and Gemstone Quality Supervision and Inspection Centre" conformity marks online. Those marks were sold along with the bracelets (nationwide) +- Evidence grade: A +- Sources: formerly the General Administration of Quality Supervision, Inspection and Quarantine (国家质量监督检验检疫总局) (2015, published; amended 2021). Measures for the Administration of Qualification Accreditation of Inspection and Testing Institutions（检验检测机构资质认定管理办法）(Articles two, fifteen, twenty-one, twenty-two, thirty-four and thirty-seven). ; State Administration for Market Regulation (国家市场监督管理总局) (2026). The State Administration for Market Regulation Publishes a Batch of Typical Cases of "Cracking Down on Fakes and Clearing Out the Sources" in the Traditional Crafts Market. ; National Standardization Administration (国家标准化管理委员会). GB/T 16552—2017 Naming of Gemstones and Jade（珠宝玉石 名称）(in force, implemented 2018-5-1). +- Notes: The names of jade and gemstones have a national standard. A test conclusion is written in a form like「石英岩玉（染色）」「翡翠（处理）」, and the two characters in the brackets are where the price difference lies. "Grade A", "old mine" and "no optimisation" are the seller's words, not a test conclusion. The livestream room selling dyed quartzite jade was using exactly this vocabulary. Other tricks appear in the same batch of cases: faking the mine site, hiring people to play "miners" on the stream, arranging trolls to warm up the mood in the room, and scripting "I urgently need money" or "a relative is seriously ill". There are also claims that goods can "prevent and cure rheumatism, raise immunity, improve sleep quality and unblock blood vessels", and that kind is handled as false advertising. For the general method of checking anything large, see item 32 in this section (check the spot-check notices and 3C certification). For who to go to when a livestream room goes wrong, see items 29 and 30 in this section (trust the platform's rules and the statute, and ask the platform for the details when a livestream goes wrong). +### 35. Blind boxes and card draws: count them as money that will not come back + +- Cost: No money. Just fix your budget before you buy. The hard part is stopping when the one you wanted did not come up. +- In plain language: The merchant must display the drawing rules, the drawing probabilities and the number of limited-edition items where you can see them, may not rig the result from the back office, and may not set an empty box. On the guaranteed minimum, on money spent and on a cap on the number of draws, the document only encourages merchants to provide one. For a blind box bought online, once the merchant has explained it and you have confirmed it, you cannot use the seven-day no-questions return after breaking the seal. +- Benefit: The basis is the Guiding Principles for Regulating Business Conduct in Blind Box Sales (Trial)（盲盒经营行为规范指引（试行））. Article nine lists a set of key items of information that must be published publicly in a prominent way, so that consumers know them before buying. They are: the name, category and style of the goods, the drawing rules, the distribution of the goods, the number of limited items released, the drawing probabilities, and the range of value of the goods. The same article also sets out three things that are forbidden. It is forbidden to manipulate the result from the back office, to change the drawing probabilities at will, or to induce consumption in disguised ways. It is forbidden to refuse or deliberately delay handing over the blind box by discounting, buying it back or exchanging it. It is forbidden to set an empty box. Article eleven only「鼓励」the operator to establish a guaranteed minimum mechanism and guide rational consumption by setting limits on the time and the money spent and a cap on the number of draws. Article seventeen deals with blind boxes sold over the internet. Where the operator has given full information and reminders and the consumer confirmed at the time of the single purchase, that blind box may be excluded from the seven-day no-questions return once the seal is broken. The operator may not replace the consumer's confirmation with a box ticked by default. Article twenty-one: it is forbidden to carry on or in disguised form carry on gambling in the name of blind boxes. Article twenty-three: it is forbidden to sell blind boxes to minors under 8 years old; selling to minors aged 8 and above requires lawful confirmation that the consent of the relevant guardian has been obtained (nationwide) +- Evidence grade: A +- Sources: State Administration for Market Regulation (市场监管总局) (2023). Guiding Principles for Regulating Business Conduct in Blind Box Sales (Trial)（盲盒经营行为规范指引（试行））(国市监稽发〔2023〕39, Articles nine, eleven, seventeen, twenty-one and twenty-three). +- Notes: This document is guidance about what an operator should do; by itself it is not a basis for punishment. If something really goes wrong, it is handled under the Product Quality Law, the Law Against Unfair Competition and the Consumer Rights Protection Law. There are three things you can complain about specifically: probabilities that were not published, published probabilities that do not match reality, and an empty box. Before complaining, record the rules on the page and the unboxing on video. Exclusion from the seven-day no-questions return after the seal is broken has a precondition: the merchant gave full information, you confirmed at the time of the single purchase, and a box ticked by default cannot count as your confirmation. Selling to minors is governed separately: a merchant may not sell to a child under 8, and for a minor aged 8 and above it must first confirm that the guardian's consent has been obtained. For getting a refund when a child tops up and tips with a phone, see item 9 in this section (a child tops up and tips with a phone). For the 24-hour cooling-off period on large non-essential purchases, see item 23 in this section (set a cooling-off period on large non-essential purchases). For the cycle of "it fades once I own it, so I buy again", see item 23 in section 6 (do not expect buying things to improve your mood). +### 36. If you care what is in the food, buy prepacked: a bulk counter is not legally required to carry an ingredient list + +- Cost: No money. The same thing in prepacked form usually costs a little more. Reading the ingredient list takes seconds. +- In plain language: The law only requires an ingredient list on prepacked food; bulk food is not required to have one. A bulk counter only has to state the name, the production date, the shelf life and who produced it. So "no additives listed" does not mean "none added", only that the law did not require it to be written. If you care, buy the prepacked version, with the ingredient list and the names of the additives printed on it. +- Benefit: Article sixty-seven of the Food Safety Law provides that prepacked food must carry a label. The label must state nine items. The second is the list of ingredients, and the seventh is「所使用的食品添加剂在国家标准中的通用名称」. Article sixty-eight asks much less of bulk food. The text is: food operators selling bulk food must state on the container or outer packaging of the bulk food「食品的名称、生产日期或者生产批号、保质期以及生产经营者名称、地址、联系方式等内容」. That article does not mention the ingredient list and does not mention additives. Frozen aquatic products are where this difference is clearest. GB 2760—2024 allows phosphoric acid and phosphates to be used in frozen aquatic products as moisture-retaining agents, commonly called water binders. The maximum level of use is 5.0 g/kg, counted as phosphate. Using it within that level is lawful. What causes trouble is going over the level, and glazing to add weight. After the 2025 CCTV 3·15 gala exposed peeled shrimp "adding a water-retaining agent (complex phosphates) over the permitted level and glazing to add weight", the market regulation authorities required companies to "clearly mark on the product label the water-retaining agent (such as phosphates), the net content after deducting the weight of the ice glaze, and the list of ingredients" (nationwide, Food Safety Law, 2025 3rd amendment, in force from 2025-12-1) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (2025, 3rd amendment). Food Safety Law of the People's Republic of China（中华人民共和国食品安全法）(Articles sixty-seven and sixty-eight, published 2025-9-12, in force 2025-12-1). ; National Health Commission and State Administration for Market Regulation (国家卫生健康委员会、国家市场监督管理总局) (2024). National Food Safety Standard, Standard for the Use of Food Additives（食品安全国家标准 食品添加剂使用标准）(GB 2760—2024, appendix A table A.1, item「磷酸及磷酸盐」, food category number 09.02.01 冷冻水产品; A.3 provides that the food additives listed in table A.1 are used within the prescribed scope of use and maximum levels of use). ; Lu'an Municipal Administration for Market Regulation (六安市市场监督管理局) (2025). Reminder and Warning Letter to Enterprises on Aquatic Products and the Use of Water Binders（关于水产制品及使用保水剂的相关企业提醒告诫函）, dated 2025-3-24. +- Notes: "Not listed" does not mean "not added", only that the law did not require it to be listed. If you want to know what is in it, buy the prepacked version. On frozen peeled shrimp and frozen fish portions, the net content on the prepacked pack is the weight after the ice glaze is deducted. Only the unit price worked out on that basis is the real price. Bulk goods are priced on the label, and they may shrink a long way after thawing. A water binder within the limit is lawful and is not a safety issue; do not treat it as a poison. Going over the level and glazing to add weight are a money problem: you have paid the price of aquatic products and received water. The other way round, the four things a bulk counter must label may also be missing. Name, production date, shelf life, and the name, address and telephone number of the producer and seller: miss any one and Article sixty-eight is breached. You can stop buying, and you can call 12315. For how to label bagged food if you run a shop yourself, see item 9 in section 12 (labels on prepacked food). The beneficiary tier for this item is you and your family, one of the two highest tiers. +### 37. Sell shares that are down according to a rule you set in advance; do not try to get even by adding to the position + +- Cost: No money. The hard part is admitting this one has already gone down, and then selling it. +- In plain language: People's instinct is to sell what has gone up and to hold what has gone down. In seven years of records from 10,000 American accounts, shares that had risen were sold at a rate more than half as high again as those that had fallen. The same people were also more eager to add to shares that were down. Yet the losing shares they held on to did 3.4 percentage points worse over the following year, on average, than the ones they sold. +- Benefit: The study looked at every trade in 10,000 accounts at one US discount brokerage from 1987 to 1993. The proportion of profitable shares sold (PGR) was 0.148. The proportion of losing shares sold (PLR) was 0.098. The two differ by 0.050, with a t value above 35. By day, a share that has risen is more than 50% more likely to be sold than one that has fallen. The same accounts were also more eager to add to losing shares. The proportion of losing shares bought again was 0.135, against 0.094 for winning shares, with a t value of 19. The study also tested the idea that what you hold on to will come back. The result runs the other way: the sold winners produced an excess return over the following year on average 3.4 percentage points higher than the held losers. Excess return means the part above a market index. The author also worked out an example at the average. Selling 1000 dollars of shares, selling the losing one gave a return over the following year about 4.4% higher than selling the winning one. That 4.4% includes the tax saved by realising the loss sooner. The tax calculation follows the US system (United States, 1987 to 1993) +- Evidence grade: A +- Sources: Odean T (1998). Are Investors Reluctant to Realize Their Losses? The Journal of Finance 53(5):1775-1798. +- Notes: This item is about whether to sell; item 15 (do not trade shares often) is about not acting all the time. The two do not conflict. If you set a rule in advance and follow it, you end up acting fewer times. Adding to the position to bring the average cost down is not wrong as arithmetic: buy a little more and the average cost is a little lower. What is wrong is treating it as a way to turn things around. Its real effect is to increase the money you have tied up in this one share, see item 19 in this section (do not put all your money into one share). The tax paragraph is worked out on the US system and does not transfer to China; the rest of the reasoning still holds. This is US account data from the 1980s and 1990s, and transaction costs and tax rules on A-shares are different. None of this is investment advice. +### 38. Act less when the market surges and crashes; do not buy more and sell more at the hottest moment + +- Cost: No money. The hard part is not piling on while everyone else is making money. +- In plain language: Someone worked out the accounts of the Shanghai Stock Exchange across the 2014 to 2015 surge and crash, covering forty million accounts. The 85% of households with the smallest accounts earned 250 billion yuan less through frequent trading than by simply holding on. That is about three tenths of the market value of their shares to begin with. Over the two and a half calm years before it, the same gap was only one to three per cent. +- Benefit: The research used daily account data from the Shanghai Stock Exchange, covering all about 40 million accounts. The sample period runs from 2014-7 to 2015-12, 18 months in all. In those 18 months the Shanghai Composite Index first rose more than 150%. It peaked at 5166.35 points on 2015-6-12 and then fell 40% by the end of December. Household accounts were sorted into four groups by their market value at the start. The smallest group was 85% of all household accounts and the largest was 0.5%. The starting market value of the shares in the two groups was about the same, 880 billion yuan and 808 billion yuan respectively. The smallest 85% lost 250 billion yuan through active trading, while the largest 0.5% made 254 billion yuan. The "loss" here is measured against the same group holding throughout and doing nothing. Each figure is about three tenths of that group's starting share value: 28% for the first and 31% for the second. For comparison, the two and a half years before June 2014 were a calm market. Taking any 18 months in that period, a comparable difference is only 1% to 3% of the starting value. In this round the average household account turned over its shares about every three weeks, close to 18 times a year (Shanghai, China, 2014 to 2015; comparison period 2012 to 2014) +- Evidence grade: A +- Sources: An L, Lou D, Shi D (2022). Wealth redistribution in bubbles and crashes. Journal of Monetary Economics 126:134-153. ; the same authors' publicly available working paper (2021-10, the figures in this item were checked against it word for word). +- Notes: The study does not say do not buy shares. It works out that within the same market episode, the more actively the smaller accounts trade, the more money moves to the other side of the trade, and the benchmark for comparison is holding on without trading. The money was not taken by institutions either, but by the individuals with the largest accounts: about nine tenths of turnover on the Shanghai Stock Exchange in that period came from retail accounts. The claim going round that "90% of retail investors lose money" has no verifiable official statistics behind it, and this item does not use it. The turnover figure is there to be compared with. In that round household accounts turned over close to 18 times a year, while the US households in item 15 (do not trade shares often) turned over only three quarters of their holdings in a year. None of this is investment advice. +### 39. If you want overseas assets such as US shares, use a QDII fund or Stock Connect; do not open an account with an overseas broker, and do not use the annual foreign exchange quota + +- Cost: No money. You can buy a QDII fund in your own bank's or broker's app in a few minutes. The price is that there are fewer products available than at an overseas broker. +- In plain language: Overseas brokers such as Tiger, Futu and Longbridge are no longer allowed to open accounts for people in the mainland from May 2026. For accounts already open, within two years you may only sell and transfer the money out; you may no longer buy. The foreign exchange quota may not be used to buy overseas shares either, and anyone who breaks the rules loses the quota for that year and the 2 years after. If you want overseas assets, go through a QDII fund or Stock Connect. +- Benefit: On 2026-5-22 eight departments, the CSRC, the Ministry of Industry and Information Technology, the Ministry of Public Security, the People's Bank of China, the State Administration for Market Regulation, the National Financial Regulatory Administration, the Cyberspace Administration of China and the State Administration of Foreign Exchange, issued the Implementation Plan for the Comprehensive Rectification of Illegal Cross-Border Securities, Futures and Fund Business Activities（综合整治非法跨境证券期货基金经营活动实施方案）. The overall requirement is that「经过2年集中整治，全面取缔境外证券期货基金经营机构非法跨境经营活动」. Overseas institutions may not market or solicit business inside the mainland, and may not「提供相关开户、处理交易指令、资金划转等交易服务」. For people who have already opened an account, the plan says「禁止境外机构为存量投资者在境内非法提供买入交易、转入资金等服务，只允许单向卖出交易并转出资金」. Once the two years are up,「全面关停境内网站、交易软件及配套服务器」. The same day the CSRC announced that it would confiscate all illegal gains of the related onshore and offshore entities of Tiger, Futu and Longbridge, and punish them severely in accordance with the law. The accompanying answers to reporters' questions state that it will「引导境内投资者通过港股通、合格境内机构投资者（QDII）及跨境理财通等合法渠道开展境外投资」. On currency exchange, the Personal Foreign Exchange Purchase Application of the State Administration of Foreign Exchange states that personal foreign exchange purchases「不得用于境外买房、证券投资、购买人寿保险和投资性返还分红类保险等尚未开放的资本项目」. Anyone who breaks the rules goes on the「关注名单」and「当年及之后连续 2 年不享有个人便利化额度」, and may also face an administrative penalty and be referred for anti-money-laundering investigation, with the information entering their personal credit record (nationwide, from 2026-5) +- Evidence grade: A +- Sources: China Securities Regulatory Commission (中国证监会) (2026-05-22). Eight Departments Including the CSRC Jointly Issue the Implementation Plan for the Comprehensive Rectification of Illegal Cross-Border Securities, Futures and Fund Business Activities（综合整治非法跨境证券期货基金经营活动实施方案）. ; China Securities Regulatory Commission (2026-05-22). The CSRC Official in Charge Answers Reporters' Questions on the Implementation Plan for the Comprehensive Rectification of Illegal Cross-Border Securities, Futures and Fund Business Activities. ; China Securities Regulatory Commission (2026-05-22). The CSRC Seriously Investigates Cases of Illegal Cross-Border Business by Institutions Such as Tiger. ; State Administration of Foreign Exchange (国家外汇管理局) (2021). Personal Foreign Exchange Purchase Application（个人购汇申请书）(Annex 1). +- Notes: These documents govern overseas institutions and say nothing about penalties for investors themselves. If you already have an account with an overseas broker, sell within the two years and bring the money back. Each of the three lawful routes has its own conditions for opening an account, so ask your own bank or broker first. The fact that US shares rose a lot in the past does not mean they will in future. Buying a QDII fund follows the same rules as item 17 in this section (a broad market index fund) and item 19 in this section (do not put all your money into one share, one platform or one property): pick a broad index and do not put everything into one market. Cash withdrawals abroad have a separate quota; see item 6 in section 21 (withdrawing cash abroad). None of this is investment advice. +### 40. Insure only the losses you could not carry yourself; cover the ones you can carry with the emergency fund + +- Cost: No money. What you have to do is list the large losses your household could run into and see which ones savings could not absorb. It takes half an hour. +- In plain language: Not all of the premium you pay comes back as claims; some covers the insurer's costs and profit. So what is worth buying is a loss that rarely happens but that the household could not pay for if it did: a serious illness, injuring someone while driving, the loss of the person the household depends on for income. Small losses that savings can absorb go to the emergency fund. +- Benefit: Of the premiums an insurer collects, only some is paid out as claims. The rest covers selling and running the business, and the company's profit. The regulator requires a company selling short-term personal health insurance to publish an all-in loss ratio on its website every six months. It is roughly the share of premiums paid out in claims, and if you want to know how much a company pays out you can look it up on its website. So for any one loss, insurance costs you a little money on average over the long run. What you get back is turning one large expense that could crush a household into a small amount a year. The smaller the loss and the more comfortably you can afford it, the worse the trade. By that standard, the few things this book has already written about are worth buying. One-year medical insurance: see item 20 in section 7 (one-year medical insurance). If you have a car, see item 26 in this section (third-party liability cover). If you travel abroad, see item 4 in section 21 (medical care abroad). If someone at home lives on your income, see item 41 in this section (term life insurance) (nationwide, since 2021) +- Evidence grade: C +- Sources: General Office of the former China Banking and Insurance Regulatory Commission (中国银保监会办公厅) (2021). Notice on Regulating the Business of Short-Term Health Insurance（关于规范短期健康保险业务有关问题的通知）(银保监办发〔2021〕7, all-in loss ratio to be published on the company website every six months). +- Notes: It is graded C because no official figure exists for working out which insurance is good value, so what is given here is a way to decide. Screen insurance and flight delay insurance, where the loss is a few hundred yuan and you can afford it yourself, need not be bought by this standard. Critical illness insurance pays the income lost while you cannot work through an illness, so someone whose savings can cover a year or two of not working may choose not to buy it. Basic medical insurance is social insurance; you do not apply this test to it and you pay it either way, see item 9 in section 7 (resident medical insurance). For what covers small losses, see item 27 in this section (emergency fund). The beneficiary for this item is you and your family. +### 41. If someone at home lives on your income, insure the earner first with term life insurance; do not insure the child first + +- Cost: Several hundred to several thousand yuan a year, depending on age, cover and the term (the author's rough estimate, no official figure). You have to disclose your health condition truthfully before taking out the policy, which takes half an hour. +- In plain language: Term life insurance covers a fixed number of years, and if the person dies in that time the household receives a sum of money. It is meant for the family that lives on your income, such as when a mortgage is still outstanding or the children are still young. If nobody depends on you, you do not need it. A child earns nothing, and the law also limits how much can be paid out on a death policy for a child. +- Benefit: The regulator defines life insurance as personal insurance with a person's lifespan as the subject, divided into term life, whole life, endowment and the like. Term life insurance makes payment of the sum insured conditional on the death of the insured, and the insurance period is a fixed number of years. Whole life insurance covers to the end of life. Accident insurance pays out only when an accident causes death or disability, does not pay on death from illness, and cannot replace life insurance. A contract whose payment condition is death requires the consent of the insured person and their agreement to the sum insured, and is otherwise void. A parent taking out death insurance on a minor child is not subject to that restriction, but the total sums paid out on the death of the child may not exceed the limit set by the regulator. The sum insured on life insurance is usually in the hundreds of thousands of yuan, so the benefit is graded large. There is no official figure for how much cover to buy. The author's method is: the unpaid mortgage, plus what the child costs until they reach adulthood, plus a few years of household expenses (nationwide, 2015 amendment) +- Evidence grade: C +- Sources: formerly the China Insurance Regulatory Commission (中国保险监督管理委员会) (2011). Measures for the Administration of Policy Terms and Premium Rates of Personal Insurance Companies（人身保险公司保险条款和保险费率管理办法）(CIRC Order No. 3 of 2011, Articles eight and twelve). ; Standing Committee of the National People's Congress (2015, amended). Insurance Law of the People's Republic of China（中华人民共和国保险法）, Articles thirty-three and thirty-four. +- Notes: It is graded C because officially there are only definitions and limits, and "insure the adult first" is worked out from the question of whose income, if it stopped, the household could not survive without. When a child is hurt, the household does not lose an income. When the earner dies, the mortgage and the cost of the children are still there. When money is limited, buy term life insurance and one-year medical insurance for the earner first, and only then the child's medical insurance. For disclosure and guaranteed renewal, see item 20 in section 7 (one-year medical insurance). The beneficiary tier for this item is tier ②, your spouse and children. +### 42. Regretting personal insurance of more than one year: cancel inside the cooling-off period and almost the whole premium comes back + +- Cost: No money. Call the insurer's customer service line or go to the counter. It takes a bit over ten minutes. The hard part is finding a quiet half hour within 15 days of signing to read the contract. +- In plain language: Every personal insurance policy of more than one year has a cooling-off period. For a policy bought at a bank it is 15 days from the day you receive the policy, and cancelling in that period costs at most 10 yuan in handling charges. Cancel after the cooling-off period and you only get back what the policy says is refundable, which is usually less than you paid. When the insurer rings to ask whether you understood, do not say you did if you did not. +- Benefit: For personal insurance with an insurance period of more than one year, the regulator requires the policy to state the day the cooling-off period starts from and how many days it runs for. It also requires the policy to state what the policyholder may do during that period. The cooling-off period for long-term health insurance may not be shorter than 15 days. Insurance bought through a bank with an insurance period of more than one year always carries a 15-day cooling-off period in the contract. It runs from the day you receive the policy and sign for it in writing, counted in calendar days. The cover of such a policy has to carry this notice:「您在收到保险合同后15日内有全额退保（扣除不超过10元的工本费）的权利。超过15日退保有损失。」After the cooling-off period you can still cancel the contract at any time. The insurer must, within 30 days, refund the cash value of the policy (the part of it that can be refunded) as the contract provides. The financial regulator's risk warning says that when a long-term personal insurance policy is cancelled after the cooling-off period, the cash value is often lower than the total premium paid. For a new personal insurance policy of more than one year, the insurer must call you during the cooling-off period. The call has to confirm that you know about the loss on cancellation and about your rights during the cooling-off period (nationwide, since 2010 and 2019) +- Evidence grade: A +- Sources: General Office of the former China Banking and Insurance Regulatory Commission (中国银保监会办公厅) (2019). Measures for the Administration of Insurance Agency Business of Commercial Banks（商业银行代理保险业务管理办法）(银保监办发〔2019〕179, Articles thirty-six, sixty-seven and sixty-eight). ; formerly the China Banking and Insurance Regulatory Commission (2019). Measures for the Administration of Health Insurance（健康保险管理办法）(CBIRC Order No. 3 of 2019, Article fifteen). ; formerly the China Banking and Insurance Regulatory Commission (2022). Rules on Information Disclosure for Personal Insurance Products of One Year or More（一年期以上人身保险产品信息披露规则）(银保监规〔2022〕24, the product document must state when the cooling-off period starts, how many days it runs for, and what the rights are). ; formerly the China Insurance Regulatory Commission (中国保险监督管理委员会) (2010). Basic Service Provisions for Personal Insurance Business（人身保险业务基本服务规定）(CIRC Order No. 4 of 2010, Article fifteen). ; Standing Committee of the National People's Congress (2015, amended). Insurance Law of the People's Republic of China（中华人民共和国保险法）, Articles fifteen and forty-seven. ; Ningxia Financial Regulatory Bureau (宁夏金融监管局) (2026-06-16). Risk Warning on Preventing Illegal Policy Cancellation on Behalf of Customers（关于防范非法代理退保的风险提示）. +- Notes: The cooling-off period is a rule set for personal insurance of more than one year. The rules do not require it for one-year medical or accident insurance, and whether there is one depends on the terms. For a policy not bought at a bank, go by the number of days in the contract and turn to the "cooling-off period" clause first when you get the policy. Savings and participating policies involve the most money, so the cooling-off period is worth using; see item 25 in this section (buy the pure protection type first). The benefit is graded "medium" because what you save is the difference between the premium paid and the cash value, commonly several hundred to several thousand yuan, and there are no official statistics. The beneficiary for this item is you yourself. +### 43. Before you sign at a bank counter, check whether what you are holding is a deposit or an insurance policy + +- Cost: No money. Spend two more minutes on the front of the contract before signing. +- In plain language: A bank may sell insurance on its behalf, but it may not describe it as a deposit. The cover of an insurance contract must print "insurance contract" and the name of the insurance company in large type, and may not contain the words "deposit" or "savings". If the cover names an insurance company, the money is not a deposit, and cancelling after the cooling-off period loses you money. For someone over 65 buying at a bank, in principle only products with a certain return may be sold. +- Benefit: The rules state that a bank may not sell insurance mixed in with savings deposits, funds or bank wealth management products. It may not draw a simple analogy between the return on insurance and the return on a deposit, and may not describe an uncertain return as a guaranteed return. Insurance policy documents and promotional materials may not use the bank's name or logo, and may not contain wording such as "deposit", "savings" or "jointly launched with the bank". The cover of an insurance policy must print "insurance contract" in type of no less than 72 point, and the name of the insurance company in type of no less than second size. Three kinds of buyer may in principle be sold only products whose policy benefits are certain: a policyholder over 65 years old; a buyer of a regular-premium product who is over 60 years old; and someone whose declared annual income is below the local per-capita disposable income of residents. Policies for these three kinds of buyer must also go to the insurer for manual underwriting, and may not be issued automatically at the branch. An insurer's staff may not sell insurance inside a bank branch. Insurance sold by a bank with a term of more than one year has a 15-day cooling-off period, and cancelling after it has passed means losing money (nationwide, from 2019-10) +- Evidence grade: B +- Sources: General Office of the former China Banking and Insurance Regulatory Commission (中国银保监会办公厅) (2019). Measures for the Administration of Insurance Agency Business of Commercial Banks（商业银行代理保险业务管理办法）(银保监办发〔2019〕179, Articles thirty, thirty-one, thirty-two, thirty-six, forty-two and forty-nine). +- Notes: It is graded B because the rules are written clearly, but there is no official figure for how much a misled person loses. This is what elderly relatives most often run into when they save at a bank, so look at the cover when you go with them. If it is already signed, cancel within 15 days under item 42 in this section (the cooling-off period). For how to complain if you were misled, see item 44 in this section (policy cancellation agents). The beneficiary for this item is you and the elderly in your family. +### 44. If you want to cancel a policy, go to the insurer yourself, not to a cancellation agent, and if you were misled complain on 12378 + +- Cost: No money. Calling the insurer's customer service line or going to the counter yourself takes one or two more trips than handing it to an intermediary. +- In plain language: Most of the people advertising "full refund, no fee unless we succeed" online are illegal cancellation agents. They take three to five tenths of the money that comes back, and they want your identity card, your bank card and a verification code. Once the policy is cancelled, the cover is gone as well. If you think you were misled, start with the insurer's customer service line, and if you cannot settle it call 12378. The whole process is free. +- Benefit: In 2026-2 the National Financial Regulatory Administration, the Cyberspace Administration of China, the Ministry of Public Security, the People's Bank of China and the China Securities Regulatory Commission jointly issued a risk warning. It says claims such as "full refund" are false information and are inconsistent with the Insurance Law and other laws and regulations. People offering "rights protection on your behalf" charge high service fees, and if you want to drop out halfway you could end up in court for a "breach". They also collect your SIM card, your bank card, your policy details, your home address and your child's school. If that information is sold on, you may run into telecom fraud and unauthorised charges on your card. A warning from the Ningxia Financial Regulatory Bureau in 2026-6 goes into more detail. These intermediaries agree that on a successful cancellation they will take a commission of 30% to 50% of the sum refunded, and charge a heavy penalty if you abandon it halfway. They teach people to invent reasons for a complaint, such as misleading sales or a forged signature. People who follow this may breach the Law on Public Security Administration Punishments, and in serious cases may be suspected of being an accomplice to false accusation or extortion. Once a policy is cancelled, the insurer's liability ends at once. Buying similar insurance later may cost more, or be refused, because you are older or because something has been found in your health. If you really were misled, complain to and negotiate with the insurer's official customer service line first, and if you cannot settle it call 12378, the consumer complaints and rights protection hotline for banking and insurance, which is free throughout. If you simply cannot afford the premium for a while, ask the insurer first whether they offer reduced paid-up amounts or policy loans. Reduced paid-up means you stop paying and the sum insured falls in proportion (nationwide, 2026) +- Evidence grade: B +- Sources: National Financial Regulatory Administration, Cyberspace Administration of China, Ministry of Public Security, People's Bank of China and China Securities Regulatory Commission (2026-02-06). Risk Warning on Guarding Against Traps in Short Videos and Livestreams by Illegal "Rights Protection on Your Behalf"（关于警惕不法「代理维权」短视频及直播陷阱的风险提示）(reprinted on the CAC website). ; Ningxia Financial Regulatory Bureau (宁夏金融监管局) (2026-06-16). Risk Warning on Preventing Illegal Policy Cancellation on Behalf of Customers（关于防范非法代理退保的风险提示）. ; Hunan Financial Regulatory Bureau (湖南金融监管局) (2025-07-23). Risk Warning on Preventing Illegal "Policy Cancellation" Grey and Black Industries（关于防范非法「代理退保」黑灰产的风险提示）. +- Notes: The Hunan Financial Regulatory Bureau reported a case in 2025. A woman was persuaded by a "friend" to use an agent to cancel her policy and paid a "consulting fee"; in the end the policy was still cancelled at its cash value. Three months after cancelling, she was diagnosed with breast cancer, and the critical illness cover was gone. Whether to cancel is for you to work out: if you still need the cover and cannot pay, ask about reduced paid-up first. To cancel inside the cooling-off period yourself, see item 42 in this section (the cooling-off period). The beneficiary for this item is you yourself. +### 45. Name a specific person as the beneficiary on the policy, and claim the money when something happens: within 5 years for life insurance and 2 years for other insurance + +- Cost: No money. Fill it in once when you take out the policy; to change the beneficiary later you must notify the insurer in writing. +- In plain language: If no beneficiary is named, or the beneficiary has died first, the sum insured counts as the estate and has to be used to pay the dead person's debts first. If the beneficiary is written as "wife, [name]" and you later divorce, at the time it counts as though nothing was named. The people in your family need to know the policy exists: for life insurance you have 5 years from learning of the death to claim, and 2 years for other insurance. +- Benefit: The Insurance Law provides that the beneficiary is designated by the insured person or the policyholder. Where the policyholder designates, the insured person's consent is required. There are three situations in which the sum insured counts as the estate of the insured person and is distributed under the rules of inheritance. First, no beneficiary was designated, or the designation is unclear. Second, the beneficiary died before the insured person and there is no other beneficiary. Third, the beneficiary has lawfully lost or given up the right to the benefit and there is no other beneficiary. Where the beneficiary and the insured person died in the same event and the order cannot be established, the beneficiary is presumed to have died first. Money counted as the estate must first clear the taxes and debts of the dead person, and only what is left is divided. The Supreme People's Court's interpretation states that where the beneficiary is written as "name plus relationship" and the relationship has changed by the time it is claimed, that counts as no beneficiary having been designated. Changing the beneficiary without notifying the insurance company does not take effect against the insurer. The statute of limitations for claiming the money is 5 years for life insurance and 2 years for other insurance, both counted from the day you knew or should have known of the event. The sum insured on life insurance is usually in the hundreds of thousands of yuan, so the benefit is graded large (nationwide, 2015 amendment, in force since 2021) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (2015, amended). Insurance Law of the People's Republic of China（中华人民共和国保险法）, Articles twenty-six, thirty-nine, forty-one and forty-two. ; Supreme People's Court (2020, amended). Interpretation on Several Questions Concerning the Application of the Insurance Law of the People's Republic of China, Part Three（关于适用《中华人民共和国保险法》若干问题的解释（三））, Article nine, paragraph two, item three, and Article ten. ; National People's Congress (全国人大) (2020). Civil Code of the People's Republic of China（中华人民共和国民法典）, Article one thousand one hundred and fifty-nine. +- Notes: Once the statute of limitations has passed the insurer can refuse to pay, and it is hard to get the money back through a lawsuit. So at least the people in your family should know which companies you hold insurance with. After a divorce, a remarriage or the birth of a child, go back and change the beneficiary once. After someone dies, collect the money from each place one by one; see item 9 in section 25 (collecting from each place). Paying off debts with the proceeds of a life insurance policy is not a route; see item 13 in section 29 (paying off debts). The beneficiary tier for this item is tier ②, your spouse and children. \ No newline at end of file diff --git "a/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" "b/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" new file mode 100644 index 00000000..73fff460 --- /dev/null +++ "b/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" @@ -0,0 +1,236 @@ +[← 回总目录](../README.md) + +# 6. 反面清单 + +本节收的是「看起来便宜又有效」的东西。翻开原始研究之后，它们的好处接近零、等于零，有的还是负的。排的顺序看两件事：大家在这上面花的钱多不多，支持它的证据硬不硬。越靠前的越值得先停掉。 + +[← Back to the contents](../README.md) + +# 6. The reverse list + +This section collects the things that look cheap and effective. Open the original studies and their benefit is close to zero, exactly zero, and sometimes negative. The order rests on two things: how much money people spend on them, and how strong the evidence behind them is. The earlier an item appears, the more worth stopping it is. + +### 1. Do not take multivitamins to live longer or prevent cardiovascular disease + +- Cost: About 0.5 to 2 yuan a day, 200 to 700 yuan a year. Taking them takes no time, it is one pill. Taking them for a long time builds a habit, and then a day without one feels like something is missing. +- In plain language: 14,000 male doctors took multivitamins for eleven years. Major cardiovascular events and deaths were just as common as in those who did not. Pooling 9 trials with more than 50,000 people shows no effect on mortality either. Not buying saves you two hundred to seven hundred yuan a year. This does not cover people who come up genuinely deficient in a nutrient, or pregnant women taking folic acid. +- Benefit: The PHS II trial enrolled 14,641 male doctors and followed them for 11.2 years (the median; half were followed longer and half shorter). Among those who took the multivitamins and those who did not, the rate of major cardiovascular events was the same. Major cardiovascular events HR 1.01 (HR is the risk ratio, where 1 means the two groups are the same). The 95% CI 0.91–1.10 is the credible range for that figure. Mortality was the same too, all-cause mortality HR 0.94 (0.88–1.02). In 2022 the US Preventive Services Task Force, the body that grades preventive measures in the United States, pooled 9 randomised controlled trials with 51,550 people and found no association between multivitamins and all-cause mortality. It graded them I, which means the evidence does not support a conclusion either way. It looks like good value because "one pill covers every nutrient" sounds sensible and because it costs only a few tenths of a yuan a day +- Evidence grade: A +- Sources: Sesso HD et al. (2012). Multivitamins in the prevention of cardiovascular disease in men: the Physicians' Health Study II randomized controlled trial. JAMA. ; US Preventive Services Task Force (2022). Vitamin, Mineral, and Multivitamin Supplementation to Prevent Cardiovascular Disease and Cancer. JAMA. ; Gaziano JM et al. (2012). Multivitamins in the Prevention of Cancer in Men. JAMA. (the cancer result discussed in the notes) +- Notes:Disputed: the same cohort of male doctors was also examined for cancer. Among those who took the multivitamins the rate of cancer was about 8% lower (0.86–0.998, P=0.04, which only just clears the bar of "not merely chance"). Death from cancer was about 12% lower (0.77–1.01). That interval crosses 1, so it may be chance. The benefit is small, and it has only been seen in male doctors. This does not cover people who come up genuinely deficient, or pregnant women taking folic acid, or people whose diet is genuinely poor; supplement those as the doctor says. +### 2. Do not take ordinary fish oil to prevent cardiovascular disease + +- Cost: About 1 to 3 yuan a day, 400 to 1000 yuan a year. Taking them takes no time. +- In plain language: Two large trials enrolled 26,000 people and 15,000 people with diabetes. Among those taking 1 gram of fish oil a day, major cardiovascular events and deaths were just as common as in those who did not. "Eating fish is good for you" does not mean "fish oil capsules are good for you". Not buying saves you four hundred to a thousand yuan a year. +- Benefit: The VITAL trial enrolled 25,871 people, gave 1 gram of fish oil a day, and followed them for 5.3 years (median). Neither major cardiovascular events nor mortality differed between the two groups. Major cardiovascular events HR 0.92 (HR is the risk ratio, where 1 means the two groups are the same). The 95% CI 0.80–1.06 is the credible range for that figure. It crosses 1, which means no difference can be seen. All-cause mortality HR 1.02 (0.90–1.15). The ASCEND trial enrolled 15,480 people with diabetes and also used 1 gram a day, followed for an average of 7.4 years, and again found no difference. Serious vascular events RR 0.97 (0.87–1.08). All-cause mortality RR 0.95 (0.86–1.05). It looks like good value because many people take "eating fish is good for you" to mean "fish oil capsules are good for you" +- Evidence grade: A +- Sources: Manson JE et al. (2019). Marine n-3 Fatty Acids and Prevention of Cardiovascular Disease and Cancer. NEJM. ; ASCEND Study Collaborative Group (2018). Effects of n-3 Fatty Acid Supplements in Diabetes Mellitus. NEJM. ; Bhatt DL et al. (2019). Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia. New England Journal of Medicine. (the high-purity prescription fish oil trial discussed in the notes) +- Notes:Disputed: one trial did work. The REDUCE-IT trial enrolled 8,179 people who already had cardiovascular disease or diabetes, were all taking statins, and had high triglycerides (135–499 mg/dL). They took 4 g of prescription pure EPA a day, and 17.2% had an event against 22.0% in the comparison group. The risk was about a quarter lower (HR 0.75, 0.68–0.83). But that is a high-dose prescription drug for high-risk patients, which is not the same as an ordinary person taking a fish oil supplement. Neither does this cover eating fish itself. +### 3. Do not give vitamin D to people who are not deficient, in the name of living longer + +- Cost: About 0.2 to 0.5 yuan a day, 100 to 200 yuan a year. The money is not much; the real cost is believing you have already done something about prevention. +- In plain language: Among people who were not deficient in vitamin D, five years of supplements made no difference to how many developed cancer, cardiovascular disease or died. People who come up genuinely deficient, people with osteoporosis and people who get no sun are not covered here; they supplement as the doctor says. +- Benefit: The VITAL trial enrolled 25,871 people, gave 2000 IU a day, and followed them for 5.3 years (median). No difference could be seen in cancer, cardiovascular disease or mortality. For the cancers that spread, HR 0.96 (HR is the risk ratio, where 1 means the two groups are the same; the 95% CI 0.88–1.06 is the credible range). Major cardiovascular events HR 0.97 (0.85–1.12). All-cause mortality HR 0.99 (0.87–1.12). The D-Health trial enrolled 21,315 Australians over 60 years old. They took 60,000 IU a month and were followed for 5 years. This showed no difference either, all-cause mortality HR 1.04 (0.93–1.18). The mortality rates were 5.3% against 5.1%. It looks like good value because some studies only keep records without intervening at all, find that people with a low vitamin D level die more often, and that finding has been misread as "supplementing it helps" +- Evidence grade: A +- Sources: Manson JE et al. (2019). Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease. NEJM. ; Neale RE et al. (2022). The D-Health Trial: a randomised controlled trial of the effect of vitamin D on mortality. Lancet Diabetes Endocrinol. +- Notes:Disputed: neither trial deliberately recruited people who were genuinely deficient in vitamin D, so the conclusion only covers people who were not deficient to begin with. People who come up genuinely deficient, people with osteoporosis, people who get no sun, and infants should still supplement as the doctor says. This only looked at all-cause mortality, cancer and cardiovascular disease, and at nothing else. +### 4. Do not take antioxidant supplements (beta carotene, vitamin E, vitamin A) to prevent cancer + +- Cost: About 0.5 to 2 yuan a day. For a smoker the real cost is that the risk of lung cancer goes up with it. +- In plain language: This one is not merely useless. Pooling 78 trials with close to 300,000 people, people taking antioxidant supplements had a slightly higher mortality rate. In two large trials that recruited smokers specifically, beta carotene raised the rate of lung cancer by 18% to 28% and death by 8% to 17%. Smokers and people exposed to asbestos should be especially careful. +- Benefit: Cochrane pooled 78 randomised controlled trials with 296,707 people. Among people taking antioxidant supplements, all-cause mortality RR 1.02 (RR is the risk ratio, where 1 means the two groups are the same; the 95% CI 0.98–1.05 is the credible range). Looking only at the better-designed trials that are harder to get wrong, RR 1.04 (1.01–1.07), a little higher. Beta carotene alone RR 1.05 (1.01–1.09), vitamin E alone RR 1.03 (1.00–1.05). The ATBC trial enrolled 29,133 male smokers who took 20 mg a day; lung cancer was 18% higher (3%–36%) and all-cause mortality 8% higher (1%–16%). The CARET trial enrolled 18,314 smokers or people exposed to asbestos. Their risk of lung cancer was about 28% higher (RR 1.28, 1.04–1.57). Their risk of death was about 17% higher (all-cause mortality RR 1.17, 1.03–1.33). It looks like good value because the chain of reasoning "antioxidants mean anti-ageing, and anti-ageing prevents cancer" sounds like common sense at every step +- Evidence grade: A +- Sources: Bjelakovic G et al. (2012). Antioxidant supplements for prevention of mortality in healthy participants and patients with various diseases. Cochrane Database Syst Rev. ; The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group (1994). The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. NEJM. ; Omenn GS et al. (1996). Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. NEJM. +- Notes: In 2022 USPSTF, the official United States body that grades preventive measures, gave beta carotene and vitamin E a grade D, meaning it recommends against them. Smokers and people exposed to asbestos have already been harmed by them. Antioxidants contained in fruit and vegetables are not affected; keep eating those. +### 5. Do not expect glucosamine or chondroitin to treat knee osteoarthritis + +- Cost: About 2 to 5 yuan a day, 700 to 1800 yuan a year. +- In plain language: Among people with osteoarthritis of the knee given a placebo for 24 weeks, six in ten also said the pain eased. Taking glucosamine or chondroitin adds only four or five percentage points on top of that, small enough to be chance. Seven hundred to eighteen hundred yuan a year buys you, in effect, the placebo response of those six in ten. +- Benefit: The GAIT trial enrolled 1,583 people with knee osteoarthritis and ran for 24 weeks. Among those given a placebo, 60.1% also reported that their pain eased. Glucosamine was only 3.9 percentage points above that (P=0.30). Chondroitin was 5.3 points higher (P=0.17). The two together were 6.5 points higher (P=0.09). All these gaps are too small to rule out chance. The real drug used as the comparator in the same trial, celecoxib, was 10.0 points higher (P=0.008). It looks like good value because "the raw material for cartilage" sounds exactly right for the problem, and because six out of ten people report a benefit from the placebo alone +- Evidence grade: A +- Sources: Clegg DO et al. (2006). Glucosamine, chondroitin sulfate, and the two in combination for painful knee osteoarthritis. NEJM. +- Notes:Disputed: in the same trial, among the people who hurt most at the start, 79.2% of those taking both improved against 54.3% in the comparison group (P=0.002). The authors say this was found after the fact in the data, so it can only be a lead, not a conclusion. The approaches to knee osteoarthritis that really have evidence behind them are losing weight and strengthening muscle; see section 2. +### 6. Do not rely on vitamin C to prevent colds + +- Cost: An effervescent tablet costs about 1 to 2 yuan, 300 to 700 yuan a year. +- In plain language: Taking vitamin C every day will basically not make you catch fewer colds, and starting once the cold has begun works only sometimes. Taking it every day shortens a cold by about 8%, which is about half a day less of feeling unwell; whether that is worth three or five hundred yuan a year is for you to work out. Marathon runners and soldiers marching in cold regions are the exception: among them, taking it halves the number who catch a cold. +- Benefit: Cochrane pooled 29 controlled trials with 11,306 people. Among ordinary people taking vitamin C every day, the number of colds caught does not really fall. The rate of catching a cold RR 0.97 (RR is the risk ratio, where 1 means the two groups are the same). The 95% CI 0.94–1.00 is the credible range. In the trials where people started taking it only after the cold had begun, the text says「未见一致效应」: some trials found an effect and some did not. In the same review, taking it every day over the long term shortens how long a cold lasts by about 8% in adults (95% CI 3%–12%) and by 14% in children (7%–21%). Marathon runners, skiers and soldiers marching in cold regions are the exception. Across 5 trials with 598 people, those who took it caught about half as many colds. Incidence RR 0.48 (95% CI 0.35–0.64). It looks like good value because it is cheap, it tastes sweet and sour, and the claim that vitamin C prevents colds has been in circulation for half a century +- Evidence grade: A +- Sources: Hemilä H, Chalker E (2013). Vitamin C for preventing and treating the common cold. Cochrane Database Syst Rev. +- Notes:Disputed: the same review does find that taking it every day shortens how long a cold lasts. By about 8% in adults (3%–12%) and by 14% in children (7%–21%). Worked out, that is about half a day less illness per cold. Marathon runners, skiers and soldiers marching in cold regions are the exception. Across 5 trials with 598 people, those who took it caught about half as many colds (RR 0.48, 0.35–0.64). Almost nobody falls into either of those two groups. +### 7. Do not give yourself a whole-body PET-CT or a tumour marker panel when you have no symptoms + +- Cost: A whole-body PET-CT costs about 7000 to 10000 yuan each time, and you take a dose of ionising radiation. A tumour marker panel costs a few hundred yuan. The real cost comes later: one "abnormality" turns up, it was nothing (this is a false positive), and then the follow-up scan, the tissue sample and the operation all cost money and cause suffering. +- In plain language: Among people with no symptoms, the most common result is a scare. More than four in five people who have a whole-body CT have at least one "abnormality" found, and nearly four in ten are called back for further checks, but most of those are benign. Women screened for ovarian cancer markers actually died of ovarian cancer slightly more often, and some had an operation they did not need because of a scare. +- Benefit: USPSTF, the United States body that grades preventive measures, gave a grade D to screening asymptomatic women for ovarian cancer with CA-125 and ultrasound, meaning it recommends against it. The evidence is the PLCO trial. Death from ovarian cancer was 0.34% in the screened group and 0.29% in the usual care group. RR 1.18 (95% CI 0.82–1.71). Screening did no good. The share who went under the knife needlessly because the result was wrong (a false positive) was 0.2%–3.25% across the trials, and up to 15% of those people had a serious surgical complication. Another study gave 1,192 people without symptoms a whole-body CT screening. 86% had at least one "abnormality" found, and 37% were advised to have further tests. The text says「大多数发现按描述是良性的」. It looks like good value because "screen the whole body once, find it early, treat it early" is so intuitive +- Evidence grade: A +- Sources: US Preventive Services Task Force (2018). Screening for Ovarian Cancer: US Preventive Services Task Force Recommendation Statement. JAMA（同一份建议另有 USPSTF 官方页）. , ; Furtado CD et al. (2005). Whole-body CT screening: spectrum of findings and recommendations in 1192 patients. Radiology. +- Notes: The tumour marker part is backed by randomised controlled trials. The whole-body CT and PET-CT part rests only on follow-up records of what happened to the people who had it, with no control group, so the evidence is a grade weaker, equivalent to B. The screenings that really do have evidence behind them, such as colorectal cancer, cervical cancer, breast cancer, and low-dose chest CT for people at high risk, are a different matter; see section 1. A PET-CT in someone who has symptoms, or who already has a cancer diagnosis, is medical care rather than screening and has to be judged separately. +### 8. Do not expect a wristband or watch that counts your exercise to help you lose weight + +- Cost: The device costs 200 to 2000 yuan. On top of that you have to break your concentration each day to look at the data. +- In plain language: Overweight people doing the same diet and exercise plan lost 3.5 kg over two years while wearing a fitness band, and those not wearing one lost 5.9 kg. If you want to lose weight, spending money on a device does not help. The exercise itself is what works. +- Benefit: The IDEA trial enrolled 471 overweight or obese adults and followed them for 24 months. Everyone followed the same diet and exercise plan; the only difference was whether they wore a wearable device. The group that wore one lost 3.5 kg (95% CI 2.6–4.5, the credible range). The group that did not lost 5.9 kg (5.0–6.8), which is more. The two groups differ by 2.4 kg (1.0–3.7, P=0.002). It looks like good value because "you cannot manage what you cannot measure" sounds sensible, and because the device only has to be bought once +- Evidence grade: B +- Sources: Jakicic JM et al. (2016). Effect of Wearable Technology Combined With a Lifestyle Intervention on Long-term Weight Loss: The IDEA Randomized Clinical Trial. JAMA. +- Notes: There is only one trial here, and it used the devices worn on the arm in the early 2010s, so it does not show that all wearables are useless. Exercise itself does work; see section 2. The conclusion is one sentence: if you want to lose weight, a device is not money you have to spend. +### 9. Do not pay a premium for organic food in the name of being "healthier" + +- Cost: Usually 1 to 3 times the price of the same kind of ordinary food. +- In plain language: There is no strong evidence that organic food is more nutritious. Only 3 studies have looked at how people's health turned out after eating it, and none found a difference. Organically grown produce does test positive for pesticide residues about 30 percentage points less often, but a positive test is not the same as exceeding the limit. Eating more fruit and vegetables does help, and whether they are organic makes no difference to that. +- Benefit: One systematic review brought together 17 studies carried out in people and 223 studies measuring food composition. The conclusion is that there is「缺乏有力证据表明有机食品比常规食品显著更有营养」. Only 3 of those studies looked at people's health after eating the food, and they looked at allergies and symptomatic infections; no difference was found between the two. The share of organically grown produce testing positive for pesticide residues is about 30 percentage points lower, but a positive test is not the same as exceeding the limit. It looks like good value because "no pesticides is safer, and safer is healthier" sounds like common sense at every step +- Evidence grade: B +- Sources: Smith-Spangler C et al. (2012). Are organic foods safer or healthier than conventional alternatives?: a systematic review. Ann Intern Med. +- Notes: Very few studies look directly at how people's health turns out after eating it. So the conclusion is "there is no evidence that it is better", not "it has been proved to have no benefit". The effect on the environment, on animal welfare and on taste are not covered here. Eating more fruit and vegetables does work, and whether they are organic makes no difference to that. +### 10. Do not spend a lot on health products, herbal formulae or tonics to "condition your body" + +- Cost: Anything from a few hundred yuan to more than ten thousand. Once it is put into gift packaging the price often multiplies several times over. +- In plain language: The department in charge of health foods requires the package to carry one sentence: a health food is not a medicine and cannot replace medicine in treating disease. The common ingredients, multivitamins, fish oil, vitamin D, have already been checked one by one in the first six items of this section, and none showed a benefit. Most of this money is spent on buying something to carry to your parents; the thought is not the problem, the product is simply the wrong one. +- Benefit: The State Administration for Market Regulation requires that the front of a health food package carry, in an area of no less than 20%, the sentence「保健食品不是药物，不能代替药物治疗疾病」. The same rules pin down what a health food is for, to「补充膳食营养物质、维持改善机体健康状态或者降低疾病发生风险因素」. The regulator itself does not accept that it treats disease. The ingredients it most often contains, multivitamins, fish oil, vitamin D, antioxidants, glucosamine and vitamin C, and what each of them gave in randomised controlled trials, have already been listed item by item in the first 6 entries of this section. It looks like good value because "it does not cure, it only conditions you" cannot be called wrong however you put it, and nobody can prove it false. And when you hand this over as a gift, the money comes from the relationship, not from evidence +- Evidence grade: C +- Sources: State Administration for Market Regulation (国家市场监督管理总局) (2019). Press Conference Held by the State Administration for Market Regulation on the Guide to Warning Wording on Health Food Labels and the Measures for the Administration of the Catalogue of Health Food Ingredients and Health Functions. +- Notes: The conclusion is "not supported by evidence", not "proved to be ineffective", so this is only graded C. For a product with the ingredients written out, look up the corresponding study in the earlier items. People who come up genuinely deficient, or who are being treated, follow what the doctor says. What to reject is the product, not the thought. Where an older relative bought it themselves and is already taking it, and it has not displaced the medicine they are on and has caused no harm, there is no need to turn up and argue the point; arranging the vaccinations they should have and the screenings they are due for matters more. For the same money, the following all have a source in this book. Taking an older relative for their yearly flu shot: see item 20 in section 1. The shingles vaccine after 50: see item 21 in section 1. The pneumococcal vaccine after 65: see item 22 in section 1. Buying a blood pressure monitor and keeping them on their blood pressure tablets until the course is finished and the pressure is at target: see item 7 in section 1. Altering the bathroom and stairs at home and practising balance and leg strength with them: see item 13 in section 1. Cancer screening that comes round at their age, take them for one: see items 17, 18 and 19 in section 1. Where an older relative has been bedridden for a long time, pressure sore prevention and long-term care insurance: see items 7 and 8 in section 17. If you want to carry something over, fruit, rice and flour, or something they actually like, is more use than a herbal formula. The two things really worth stopping are this: one is retirement investment schemes where you pay money up front, and the other is products that displace medicine; see item 5 in section 17 +### 11. Do not take probiotics long term when you have no bowel problem + +- Cost: About 2 to 5 yuan a day, 700 to 1800 yuan a year. +- In plain language: Pooling 45 trials in healthy adults, probiotics do not change the bacteria in the gut for long; they go back to where they were when you stop, and blood lipids do not improve either. Specific situations such as diarrhoea caused by antibiotics have their own evidence and are not covered here. +- Benefit: One review pooled 45 trials in healthy adults and concludes that it「未能支持益生菌能引起肠道菌群的持久改变，也未能改善健康成人的血脂」. The bacteria in the gut do change, but only temporarily, and they go back when you stop. It looks like good value because "gut health" is broad enough that nothing can be said against it, and because one bottle a day comes to very little per bottle +- Evidence grade: B +- Sources: Khalesi S et al. (2019). A review of probiotic supplementation in healthy adults: helpful or hype? Eur J Clin Nutr. +- Notes: The same review does report a few small improvements in healthy people, such as more formed and suitable stools and more lactobacilli in women's vaginas. Diarrhoea caused by antibiotics, and one particular strain used for one particular condition, each have their own evidence and have to be judged separately. This item is only about healthy people taking it daily as a supplement. +### 12. Do not force yourself into cold showers in the name of "strengthening immunity" + +- Cost: No money. Every shower is uncomfortable for 30 to 90 seconds, and keeping it up through winter takes willpower. People with cardiovascular disease also have to be careful, because cold stimulation is a risk for them. +- In plain language: In the trial of 3018 people, those who took cold showers asked for fewer sick days off, 29% fewer by number, but the total number of days off was the same as everyone else's, so they were not less ill. A 2025 pooling of 11 trials found that immediately after cold water the markers of inflammation in the body actually rose briefly, and the immune markers did not change. Take cold showers if you enjoy them, but do not treat it as an investment in your health. +- Benefit: The Netherlands ran a randomised controlled trial with 3,018 people who took a cold shower for 30, 60 or 90 seconds after a hot one. They took 29% fewer sick days by number (IRR 0.71, which is a ratio of counts). But the text states「病假天数没有显著组间差异」: the total number of days off was the same in both groups. Quality of life, work productivity and anxiety also showed no meaningful difference. A 2025 systematic review pooled 11 randomised trials with 3,177 people. Immediately after the cold immersion, and again 1 hour later, the markers of inflammation in the body rose rather than fell. The immune markers did not change. Stress fell only at the 12-hour time point. The authors themselves say the evidence is「受限于 RCT 少、样本小」, which means there are too few randomised trials and each one has too few people in it. It looks like good value because it is free, and because "suffering makes it work" is a very smooth intuition +- Evidence grade: B +- Sources: Buijze GA et al. (2016). The Effect of Cold Showering on Health and Work: A Randomized Controlled Trial. PLOS ONE. ; Cain T et al. (2025). Effects of cold-water immersion on health and wellbeing: A systematic review and meta-analysis. PLOS ONE. +- Notes:Disputed: the number of sick days taken really did fall by 29%, and that gap does not look like chance. But the total number of days ill did not fall, so it looks more like "going to work even when ill" than genuinely being less ill. Cold water used to help athletes recover after training is a separate question and is not covered here. A cold shower does no harm to a healthy person in itself; just do not treat it as an investment in your health. +### 13. Do not buy detox, colon cleansing, enzyme or alkaline water products + +- Cost: Several hundred to several thousand yuan. The kinds that work by starving you may also drive your blood sugar down and leave you short of nutrients. +- In plain language: The detox industry is large, but very little clinical evidence supports it, and nobody has properly tested the detox diets sold on the market. When people looked at whether dietary acid load, alkaline water and cancer are related, only 1 study met the standard, and it showed no association. +- Benefit: The conclusion of that review is that「尽管排毒产业蓬勃发展，支持这些饮食的临床证据非常少」, and that there are「没有针对商业排毒饮食的随机对照试验」. A separate systematic review looked specifically at whether dietary acid load, alkaline water and cancer are related: it went through 8,278 publications and found only 1 that met the standard for inclusion, and that one showed no association either. What the authors wrote is that「向公众推广碱性饮食和碱性水用于癌症预防或治疗是不合理的」. It looks like good value because the two premises, "there are toxins in the body that need to come out" and "an acid constitution causes cancer", sound a good deal like physiology +- Evidence grade: C +- Sources: Klein AV, Kiat H (2015). Detox diets for toxin elimination and weight management: a critical review of the evidence. J Hum Nutr Diet. ; Fenton TR, Huang T (2016). Systematic review of the association between dietary acid load, alkaline water and cancer. BMJ Open. +- Notes: All that can be said is "not supported by evidence", so this is graded C. Clearing out the waste your metabolism produces is what the liver and kidneys do. As long as those two organs are not diseased, you do not need to do anything extra to "detox". +### 14. Do not force yourself to drink water to reach "eight glasses a day" + +- Cost: No money. It takes a little time and attention to keep track, and you get up more often at night. +- In plain language: People have gone looking for the source of "eight glasses a day" and found not one study supporting it. Food and other drinks already carry a great deal of water, so there is no need to drink that much. Drink when you are thirsty and look at the colour of your urine; that is enough. People working in the heat, people exercising hard, people who have had kidney stones, and older people who no longer feel thirsty should drink on purpose. +- Benefit: One literature review looked specifically at the "8x8" claim, which means eight glasses of eight ounces of water a day. Its conclusion is that it「未找到任何支持 8×8 的科学研究」. It also writes that「对数千名成年人的食物与液体摄入调查强烈提示不需要如此大量」, because food and other drinks already supply a great deal of water. The claim looks like good value because water is free, because 8 is a tidy number, and because "drink more water" never sounds wrong wherever it comes up +- Evidence grade: C +- Sources: Valtin H (2002). "Drink at least eight glasses of water a day." Really? Is there scientific evidence for "8 x 8"? Am J Physiol Regul Integr Comp Physiol. +- Notes: What is being rejected is the number "you must drink so many glasses a day", not the idea that drinking more water is harmful. Drink when you are thirsty and look at the colour of your urine; that is enough. People working at high temperatures, people exercising hard, people who have had kidney stones, and older people who no longer feel thirsty should drink on purpose, and that is a separate matter. +### 15. Do not pay for fortune telling, tarot or horoscopes to make decisions + +- Cost: No money. The hard part is giving up a shortcut to a decision and having to think it through yourself afterwards. +- In plain language: There is a classic classroom experiment in which the "personal personality analysis" the students received was in fact the same for everyone, and most of them thought it was accurate. That is why fortune telling seems accurate. An astrologer picks correctly about as often as chance. The cost is not only a few tens of yuan: a fortune-telling gang in Xixian, Henan collected incense money and "support packages" of up to 3888 yuan, and the main offender was sentenced to eleven years. +- Benefit: In the classic classroom experiment, the "personal analysis" students received after a personality test was in fact the same for everyone, yet most of them thought it accurate. The author's conclusion is「被本人或分析者接受，不能证明解释正确」. The 39 students rated how accurate the analysis was on a scale from 0 to 5, and 34 of them gave it 4 or 5, which works out at an average of about 4.26 using the distribution in the table. That is why fortune telling always seems accurate; the phenomenon is called the Barnum effect. Astrology has also been put through a double-blind test, published in Nature. The astrologer had to pick, out of three personality test profiles, the one that matched the chart. Out of 116 they picked 40, while pure chance would have picked 38.5, and the astrologers predicted beforehand that they would pick 58.5. Official case: the Xixian Procuratorate in Henan reported an online fortune-telling fraud gang that first drew people in with free palm readings and then took "incense money" and "support packages" of up to 3888 yuan. There were more than 50 victims, more than 200 ten thousand yuan of stolen funds was recovered, and the main offender was sentenced to eleven years' imprisonment with a fine of 100,000 yuan +- Evidence grade: B +- Sources: Forer BR (1949). The fallacy of personal validation: A classroom demonstration of gullibility. Journal of Abnormal and Social Psychology 44(1):118-123（表 1 评分分布：2 分 1 人、3 分 4 人、4 分 18 人、5 分 16 人，N=39）. ; Carlson S (1985). A double-blind test of astrology. Nature 318:419-425（表 2：首选 116 份，挑中 40 份，随机期望 38.5，占星师预计 58.5；「The data are consistent with chance, inconsistent with astrological hypothesis」）. , full text on the second author's homepage ; Supreme People's Procuratorate (最高人民检察院) (2026). Xixian, Henan: Three Years of Continuous Tracking Taken to a Batch of Online Fortune-Telling Fraud Gangs（河南息县：历经三年持续追踪打掉一批网络算命诈骗团伙）. +- Notes: Each of those two is a single study. Forer's paper does not state the average score directly; the 4.26 is worked out from the distribution in table 1. Nothing here says it predicts accurately, and no belief is being judged. Spending a few tens of yuan on it as entertainment is not what is at issue; the losing move is using it to decide about leaving a job, getting married or investing. +### 16. Do not buy blue-light glasses to "protect your eyesight", and do not believe that months in front of a screen ruins your eyes; but treat a swollen, red eye as an emergency + +- Cost: No money, and you also save the several hundred to several thousand yuan added to the price of the lenses. +- In plain language: Blue-light lenses do not work, and not buying them saves you several hundred to several thousand yuan. Compared with ordinary lenses, people wearing them report the same amount of eye tiredness and see just as clearly, and no evidence has been found that they protect the back of the eye. The strain and dryness that come from looking at a screen do ease off. What really damages the eye is the arc light of a welding torch, ultraviolet light, lasers, and an acute attack of glaucoma. +- Benefit: A Cochrane systematic review from 2023 included 17 randomised controlled trials with 5 to 156 people each, followed for anything from under 1 day to 5 weeks. Because the data are so few, the trials recruited very different people, and the follow-up times do not line up, the authors did not pool them into a single total. All the results below are against lenses with no blue-light filter. With follow-up under a week, self-reported eye tiredness「可能没有差别」, with low certainty of the evidence. In one trial of 120 people the two groups differed by 9.76 units. The credible range for that difference is 95% CI −33.95 to 53.47. The range spans both signs, which means no difference can be seen. Critical flicker fusion frequency is an objective measure of eye fatigue, and on it the two groups「可能几乎没有差别」, with low certainty. Best-corrected visual acuity, meaning the best vision once the glasses correct it, was「大概没有影响」. The two groups differed by 0.00 logMAR (95% CI −0.02 to 0.02). This outcome rests on 1 study with 156 people, with moderate certainty. Whether it affects daytime alertness is unclear (2 trials, 42 people, very low certainty). The review also found no evidence that it protects the macula at the back of the eye +- Evidence grade: A +- Sources: Singh S, Downie LE, Anderson AJ, Keller PR, White SJ, Ang M, Wolffsohn JS (2023). Blue-light filtering spectacle lenses for visual performance, sleep, and macular health in adults. Cochrane Database of Systematic Reviews, 8, CD013244. +- Notes: What looking at a screen brings is visual fatigue and dry eye, which show up as aching, dryness and blurring after a while. Resting and blinking more eases it. It is something that eases, which is not the same as "the screen has damaged your eyes". That is the consensus of ophthalmology rather than the conclusion of the Cochrane review in the sources field, so it only counts as grade C. What really causes irreversible eye damage in a short time is an intense light source such as a welding arc, ultraviolet light or a laser. An eye burned by a welding arc is called photophthalmia, and it is a legally recognised occupational disease. What keeps it off is goggles and a face shield, not blue-light lenses; see item 11 in section 19 (protection against dust, noise, chemical poisons). For staying up all night in front of a screen, what you have to guard against is not the screen itself but one exception: an eye that is swollen and red, a rainbow halo around lights, and headache, nausea and vomiting mean an acute attack of angle-closure glaucoma, which can destroy the optic nerve within days, so go to the eye department emergency the same day; see item 6 in section 13 (eye department emergency). A dilated pupil in the dark and long periods of looking down, which pushes the lens forward, are what bring it on, and those at highest risk are people over 50, people with long-sightedness and people with a shallow anterior chamber. The dry eye that comes from screen use can, if it goes on, turn into a chronic condition that needs a doctor, and it is not fixed by a night's sleep. If your eyes keep troubling you, or your sight really is going down, go and find out why in the eye department instead of swapping lenses yourself. For how to prevent myopia in children and teenagers, see items 4, 12 and 9 in section 30 (two hours outdoors a day, dilated refraction, and not buying products that "cure" myopia). +### 17. Do not treat blood donation as a health practice: "detox, lower cholesterol, prevent heart attacks" has no support from randomised trials, while donating too often really does run you out of iron + +- Cost: No money. One donation, with the forms, the blood draw and the time sitting afterwards, takes about an hour. For weeks afterwards your body is making up the iron you lost. +- In plain language: The claim that donating blood detoxifies, lowers cholesterol and prevents heart attacks does not hold up. People who keep giving blood were already healthier to begin with; take that layer out and the fall in mortality from one extra donation a year shrinks from nearly a fifth to 7.5%. In the trials that actually separated people into groups, there was no difference in deaths. In the other direction, donating too often really does make people weak and sensitive to the cold, and the reason is iron deficiency. +- Benefit: The Kuopio study in Finland followed 2862 men aged 42 to 60 for an average of nearly 9 years. Of the 153 who had given blood in the 24 months before counting began, 1 (0.7%) had an acute heart attack. Among the 2529 who had not given blood, 316 (12.5%) had one (the difference between the two proportions P<0.0001). Once the effect of age and other coronary heart disease risk factors was taken out, those who had given blood had about 88% lower risk of heart attack. Relative risk 0.12 (95% CI 0.02–0.86, P=0.035). But this comes only from follow-up records, with no control group. The Nordic SCANDAT database holds 1,182,495 donors and 9,526,627 person-years of follow-up in total, of whom 15,401 died. Taking out only characteristics such as age and sex, giving blood one extra time a year was associated with 18.6% lower mortality (95% CI 16.8%–20.4%). Taking out also the layer of "you only stay allowed to give blood if you stay healthy", the reduction is only 7.5% (5.7%–9.4%). The authors conclude that it「不能解释为有益健康的确凿证据」. The FeAST randomised controlled trial enrolled 1277 patients with symptomatic lower limb arterial disease and randomised them to two groups, one having blood drawn every 6 months on a fixed schedule to reduce iron, the other as a comparison. The main measure was all-cause mortality, 125/636 (20%) against 148/641 (23%). HR 0.85 (HR is the risk ratio, where 1 means the two groups are the same). 95% CI 0.67–1.08, P=0.17. Taken together, death plus heart attack and stroke among the survivors was 180 (28%) against 205 (32%), HR 0.88 (0.72–1.07, P=0.20). Neither showed a difference. The INTERVAL randomised trial enrolled 45,263 people giving whole blood. They were split into three different intervals between donations. For men they were 12, 10 and 8 weeks, for women 16, 14 and 12 weeks. With the interval shortened, noticeably more blood was collected over two years (the 8-week group of men gave 1.69 units more than the 12-week group, 95% CI 1.59–1.80). Quality of life, physical activity and cognitive function showed no difference between the groups. But there was more discomfort from giving blood, including tiredness, breathlessness, dizziness, lightheadedness and restless legs (legs so uncomfortable at night that you cannot lie still). At the same time the average haemoglobin and ferritin were lower, and there were more deferrals for low haemoglobin (all items P<0.0001). Iron deficiency affects how much heat the body makes: 10 women with iron deficiency anaemia, 8 people whose iron stores were exhausted but who were not yet anaemic, and 12 control women soaked in water at 28 ℃ for 100 minutes. The anaemia group had a rectal temperature of 36.0±0.2 ℃ against 36.2±0.1 ℃ in the control group (P=0.001). Oxygen consumption was 5.28±0.26 against 5.99±0.29 mL·min⁻¹·kg⁻¹ (P=0.04). The two thyroid hormones, thyroxine and triiodothyronine, were also lower (P<0.002). The claim looks sound because people who give blood are healthy people who have been screened again and again, and because the hypothesis that "letting iron out protects the heart" has not been confirmed in a randomised trial +- Evidence grade: A +- Sources: Salonen JT, Tuomainen TP, Salonen R, Lakka TA, Nyyssönen K (1998). Donation of blood is associated with reduced risk of myocardial infarction. The Kuopio Ischaemic Heart Disease Risk Factor Study. American Journal of Epidemiology, 148(5), 445-451. ; Ullum H, Rostgaard K, Kamper-Jørgensen M, et al. (2015). Blood donation and blood donor mortality after adjustment for a healthy donor effect. Transfusion, 55(10), 2479-2485. ; Zacharski LR, Chow BK, Howes PS, et al. (2007). Reduction of iron stores and cardiovascular outcomes in patients with peripheral arterial disease: a randomized controlled trial. JAMA, 297(6), 603-610. ; Di Angelantonio E, Thompson SG, Kaptoge S, et al. (2017). Efficiency and safety of varying the frequency of whole blood donation (INTERVAL): a randomised trial of 45 000 donors. Lancet, 390(10110), 2360-2371. ; Beard JL, Borel MJ, Derr J (1990). Impaired thermoregulation and thyroid function in iron-deficiency anemia. American Journal of Clinical Nutrition, 52(5), 813-819. ; Standing Committee of the National People's Congress (全国人大常委会) (1997). Blood Donation Law of the People's Republic of China（中华人民共和国献血法）(Articles two, nine and fourteen). ; National Health Commission and others (国家卫生健康委等) (2024-01-08). Notice on Further Improving Incentive and Reward for Voluntary Blood Donors（关于进一步做好无偿献血者激励奖励工作的通知）. +- Notes: What is being rejected is donating blood for the sake of your own health, not blood donation itself. The blood is for someone else, so on this book's beneficiary tiers it falls in tier ④ (a stranger). What does come back to you is the share the system gives you. Article fourteen of the Blood Donation Law provides that a person who has given blood unpaid is exempt from the collection, storage, separation and testing charges when using blood in hospital themselves, and that the spouse and immediate family may be exempt or pay less according to the provincial government's rules. The National Health Commission notice of January 2024 states that 28 provinces already let donors and their relatives have the cost deducted at discharge for blood used in hospitals in the same province. Those who have received the voluntary blood donation award can also, under the local "three free" policy, travel on public transport free, visit government-run parks free, and be exempt from the ordinary outpatient consultation fee at public hospitals. If you are going to give blood, keep to the statutory intervals. Article nine of the Blood Donation Law provides for 200 millilitres each time in general, and no more than 400 millilitres. The interval between the two must be at least 6 months. The recommended ages are 18 to 55. Do not shorten the interval to the minimum in order to collect an honour or a group quota. Women, people of low body weight and people with heavy menstrual bleeding are already short of iron; anyone who donates repeatedly can pay for one ferritin test. If you feel constantly tired or cold after donating, or your legs are uncomfortable at night and you cannot stop moving them (this is called restless legs), do not assume your "base qi" is damaged; have a blood count and a ferritin test. If you come up deficient, take iron as the doctor says and lengthen the interval between donations. The human evidence on feeling cold rests on 10 people; it explains why this happens, it is not a statement of how many people it happens to. +### 18. Do not rely on a general health check package for prevention; have the few screenings that do have evidence, by age and sex + +- Cost: No money, and it saves the several hundred to several thousand yuan a year that a package costs. What you take on is the obligation to remember which tests to have and how often. +- In plain language: People who have a general health check regularly have the same mortality as people who do not, and the same number of deaths from cardiovascular disease and cancer. Having the few screenings that do have evidence, by age and sex, is enough. Anyone over 35 can have their blood pressure measured free once a year, and anyone over 65 gets a free health check once a year. +- Benefit: The 2019 update of the Cochrane systematic review included 17 randomised controlled trials. For all-cause mortality it pooled 11 of those trials, with 233,298 participants and 21,535 deaths. The group that had the general health check was the same as the comparison group, RR 1.00. RR is the risk ratio, where 1 means the two groups are the same. The 95% CI 0.97–1.03 is the credible range. For death from cardiovascular disease it pooled 9 trials with 170,227 people and 6,237 deaths, RR 1.05 (95% CI 0.94–1.16). For death from cancer it pooled 8 trials with 139,290 people and 3,663 deaths, RR 1.01 (95% CI 0.92–1.12). Neither showed a difference either. The authors' conclusion is that「通用健康检查不大可能有益」. The national basic public health service programme gives residents aged 35 and above one free blood pressure measurement a year, and gives people aged 65 and above one free health check a year. This free check covers a physical examination, a full blood count, a routine urine test, fasting blood glucose and an electrocardiogram. Liver function covers alanine aminotransferase, aspartate aminotransferase and total bilirubin. Kidney function covers serum creatinine and blood urea. Blood lipids cover total cholesterol, triglycerides, low-density lipoprotein cholesterol and high-density lipoprotein cholesterol. Finally there is an abdominal ultrasound of the liver, gallbladder, pancreas and spleen (nationwide) +- Evidence grade: A +- Sources: Krogsbøll LT, Jørgensen KJ, Gøtzsche PC (2019). General health checks in adults for reducing morbidity and mortality from disease. Cochrane Database of Systematic Reviews, 1, CD009009. ; National Health and Family Planning Commission (国家卫生计生委) (2017). National Basic Public Health Service Specification, Third Edition（国家基本公共卫生服务规范（第三版））(health management for older people, health management for patients with hypertension). +- Notes: What is being rejected is using an annual package as a means of prevention, not doing every test there is. Separately, the free yearly blood pressure measurement over 35 and the free yearly health check over 65 are provided by the national basic public health service programme in any case, and cost you nothing. Where there is a reason to have a test, have it. Blood pressure, blood glucose and hepatitis B: see items 7, 8 and 14 in section 1. Breast, cervical and colorectal: see items 17, 18 and 19 in section 1. Helicobacter pylori and low-dose CT: see items 23 and 24 in section 1. If you have had high-risk behaviour, get tested; see item 31 in section 1. The age to start at and how often to repeat are written down there. An examination organised by your employer and paid for by someone else is worth having; what is being stopped is you paying for a package and you adding items to it. People who already have a chronic illness, people with symptoms, and tests a doctor orders for a particular condition are not covered, because that is medical care rather than screening. The items in a package with least evidence behind them are tumour markers and whole-body imaging; see item 7 in this section (whole-body PET-CT and tumour marker panels). For high uric acid that has never hurt and gallstones that have never ached, see item 19 in this section (asymptomatic high uric acid) and item 20 in this section (asymptomatic gallstones). Separately, most of the trials Cochrane included were done in high-income countries, where seeing a doctor is easy to begin with, so this conclusion should not be copied straight into a region where medical resources are tight. +### 19. Do not start urate-lowering drugs because a check-up found high uric acid and you have never had an attack + +- Cost: No money, and it saves the long run of medicine and repeat testing as well as the risk of a dose you did not need. +- In plain language: The American College of Rheumatology recommends that people whose uric acid is above the limit but who have never had an attack of gout and have no gouty tophus should not start urate-lowering drugs. You have to treat 24 people for 3 years to prevent one attack. The hope that it protects the kidney has also fallen through: kidney function fell just as fast in the people taking the drug as in those taking a placebo. Treat it after the first real attack. +- Benefit: The American College of Rheumatology's 2020 gout guideline covers asymptomatic hyperuricaemia. It means a blood urate above 6.8 mg/dL, with no previous attack of gout and no tophus under the skin. The guideline conditionally recommends that these people do not start any urate-lowering drug. The drugs include allopurinol, febuxostat and probenecid. The certainty of the evidence is graded "high". The footnote under the table states「按归因风险计算，需治疗 24 名患者 3 年才能预防一次痛风发作」. The CKD-FIX randomised controlled trial enrolled 363 people. They had stage 3 or stage 4 chronic kidney disease, with no history of gout but at risk of progression. One group took 100–300 mg of allopurinol a day and the other a placebo. The trial ran for 104 weeks. eGFR is the measure of kidney function, in units of mL/min/1.73 m². In one group it changed by -3.33 a year (95% CI -4.11 to -2.55). In the other it changed by -3.23 a year (-3.98 to -2.47). The two differ by -0.10 (-1.18 to 0.97), P=0.85, so kidney function fell at the same speed. Serious adverse events 46% against 44%. The PERL randomised controlled trial enrolled 530 people with type 1 diabetes and early to mid-stage diabetic kidney disease, who took allopurinol for 3 years and then stopped for 2 months to wash the drug out. Allopurinol did bring blood urate down from 6.1 to 3.9 mg/dL. But once the drug was stopped, the two groups differed in the kidney measure GFR by only 0.001 mL/min/1.73 m². The credible range for that difference is 95% CI -1.9 to 1.9, P=0.99. Urinary albumin leakage was in fact 40% higher in the allopurinol group (0 to 80). It looks like good value because the intuition "if a measure is out of range, take a drug to push it down" runs so smoothly +- Evidence grade: A +- Sources: FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760（表 1 及其脚注）. ; Badve SV, Pascoe EM, Tiku A, et al. (2020). Effects of Allopurinol on the Progression of Chronic Kidney Disease. New England Journal of Medicine, 382(26), 2504-2513. ; Doria A, Galecki AT, Spino C, et al. (2020). Serum Urate Lowering with Allopurinol and Kidney Function in Type 1 Diabetes. New England Journal of Medicine, 382(26), 2493-2503. +- Notes: What is being rejected is "start lifelong urate-lowering drugs before you have ever had an attack", not the idea that high uric acid does not matter. The same guideline lists the exceptions. Starting the drug can be considered in three situations. First, a first attack that comes at the same time as stage 3 or worse chronic kidney disease. Second, blood urate above 9 mg/dL (about 535 µmol/L). Third, a history of urinary stones. People who have already had an attack, people with tophi, and people whose scans show bone already eroded should take the drug; for the details see item 9 in section 16 (once gout is diagnosed, take urate-lowering drugs long term). About 7.4% of Han Chinese carry a genotype called HLA-B\*5801, against 0.7% of white people. People with this type are more likely to develop a severe and potentially fatal drug rash on allopurinol, and the risk of hypersensitivity syndrome among Asian people is 3 times that of white people. This is a drug you can perfectly well do without for now, and there is no reason to take on that extra risk. Before you do start, have this genotype tested; see item 9 in section 16 (once gout is diagnosed, take urate-lowering drugs long term). Those two kidney trials were done in people with chronic kidney disease and type 1 diabetes, and cannot be turned into "high uric acid is completely harmless for the kidney". All they rule out is the one practice of taking a drug to lower uric acid in order to protect the kidney. +### 20. Do not remove a gallbladder preventively because a check-up found gallstones and you have never felt pain + +- Cost: No money, and it saves one operation under general anaesthetic and one hospital stay. What you take on is remembering which signs mean you have to go to hospital at once. +- In plain language: Among people with stones in the gallbladder who have never felt pain, the large majority never do. An Italian study followed 580 such people for nearly nine years and 78% had no symptoms at all throughout. A US group of 135 people followed for about four years had only 7% operated on. The British guideline puts it plainly: if stones are found in the gallbladder by chance and you have no symptoms, you do not need treatment. +- Benefit: The Italian MICOL population survey gave ultrasound scans to 11,229 people aged 29 to 69. It found 856 with gallstones, of whom 580 had no symptoms, and followed them for an average of 8.7 years. Of those 580, 453 (78.1%) had no symptoms throughout, 61 (10.5%) developed mild symptoms, and 66 (11.4%) developed severe symptoms. In total 189 cholecystectomies were done during the follow-up, and 41.3% of them were on people with no symptoms. The authors conclude that「期待治疗对多数患者仍是有效的处理方式」, where expectant management means watching and waiting rather than operating. A US HMO kept follow-up statistics, and an HMO is a managed care organisation; it followed 691 people with diagnosed gallstones for an average of 78±61.6 months. Of the 135 with no symptoms, followed for 58±50.2 months (median 46.3 months), only 10% developed symptoms of gallstones and 7% had an operation. Of the 50 deaths in the whole group, only 2 were related to the biliary tract, and both were in the group that had symptoms. The authors conclude that「有无声结石的患者不需要在出现症状前手术」, where a silent stone is one that does not speak up and does not attack. Paragraph 1.2.1 of the British NICE guideline CG188 says this.「对胆囊和胆道均正常、偶然发现无症状胆囊结石的人，应告知其在出现症状前不需要治疗」. Prophylactic removal looks like good value because of the intuition "it will hurt eventually, so cut it out while you are young" +- Evidence grade: A +- Sources: Festi D, Reggiani ML, Attili AF, et al. (2010). Natural history of gallstone disease: Expectant management or active treatment? Results from a population-based cohort study. Journal of Gastroenterology and Hepatology, 25(4), 719-724. ; McSherry CK, Ferstenberg H, Calhoun WF, Lahman E, Virshup M (1985). The natural history of diagnosed gallstone disease in symptomatic and asymptomatic patients. Annals of Surgery, 202(1), 59-63. ; National Institute for Health and Care Excellence (2014). Gallstone disease: diagnosis and management. NICE guideline CG188, Articles 1.2.1 and 1.3.1. +- Notes: This covers one situation only: the stones are in the gallbladder and have never caused pain. **A stone in the common bile duct is a different matter.** Paragraph 1.3.1 of the same guideline says plainly that a common bile duct stone has to be cleared from the duct and the gallbladder removed, whether or not there are symptoms. If any of the following happens, go to hospital the same day and do not try to tough it out on the strength of this conclusion: persistent cramping pain in the upper right or upper abdomen, fever with chills, yellowing of the whites of the eyes or the skin, or darkening of the urine. When a doctor has looked at your imaging and recommends surgery, ask which reason they are relying on, for instance whether the bile duct is involved, whether there are polyps as well, or whether the gallbladder wall looks abnormal, and do not use this conclusion to argue against them. All it holds back is "no reason at all, cut it out simply because stones were found". The two studies above followed a group of people forwards without a control group. But what they answer is "what happens if you leave it alone", not "which treatment is better", so they fit here. Both also followed people for 4 to 9 years, and over a longer period the share who develop symptoms rises. +### 21. Do not cut out calcium in order to prevent kidney stones + +- Cost: No money. It also saves you trouble, because you no longer have to watch what contains calcium every day. +- In plain language: Men with calcium oxalate stones are often told to eat less calcium, and the trial result is the other way round. Men with recurrent stones who ate calcium normally while cutting back on animal protein and salt had only half as many recurrences over five years as those on a low calcium diet. The British guideline also says not to restrict calcium. +- Benefit: One randomised controlled trial ran for 5 years and enrolled 120 men with recurrent calcium oxalate stones and also high urinary calcium, which is called idiopathic hypercalciuria. They were randomised into two groups. One group ate calcium normally, 30 mmol a day. That group also cut back on animal protein (52 g a day) and on salt (50 mmol of sodium chloride a day). The other group ate the traditional low calcium diet, 10 mmol a day. Over 5 years recurrence was 12/60 against 23/60. The uncorrected relative risk was 0.49 (95% CI 0.24–0.98, the credible range), P=0.04, so the normal calcium group had half the risk of recurrence. During follow-up, urinary calcium fell in both groups by about 170 mg a day. But urinary oxalate differed. In the low calcium group it rose by 5.4 mg a day on average, while in the normal calcium group it fell by 7.2 mg a day. Paragraph 1.8.1 of the British NICE guideline NG118 states that daily calcium intake should not be restricted. For adults, 700 to 1,200 mg a day. For children and teenagers, 350 to 1,000 mg a day depending on age. Cutting out calcium looks like good value because of the intuition "stones are made of calcium, so eat less of it" +- Evidence grade: A +- Sources: Borghi L, Schianchi T, Meschi T, et al. (2002). Comparison of two diets for the prevention of recurrent stones in idiopathic hypercalciuria. New England Journal of Medicine, 346(2), 77-84. ; National Institute for Health and Care Excellence (2019). Renal and ureteric stones: assessment and management. NICE guideline NG118, Article 1.8.1. +- Notes:Disputed: this trial was only done in men, and all of them had recurrent calcium oxalate stones with high urinary calcium, so it cannot be applied directly to every kind of stone. The intervention group also changed three things at once, eating calcium normally, eating less animal protein and eating less salt, so not all of the benefit can be put down to not cutting out calcium; the lower salt and the less animal protein are doing work too. What is being rejected is "deliberately cut out anything containing calcium", not "you should take more calcium". What is changed here is calcium in your daily meals; calcium supplements are a separate question, and if you have had stones, ask your treating doctor whether to take them. For drinking more water, see item 8 in section 16 (after a kidney stone, drink 2.5–3 litres a day). +### 22. Do not buy crystals, bead bracelets or pixiu for "turning your luck", "bringing in money" or "nurturing people" + +- Cost: No money. All you have to do is keep "turning your luck" and "nurturing people" out of your reasons for buying it. The hard part is keeping up when everyone around you is wearing one and saying it works. +- In plain language: There is no checkable evidence that wearing one changes your luck, brings in money or nurtures people. It works the same way fortune telling seems accurate: the description is vague enough that you fit yourself into it. When a merchant goes on to claim the goods treat disease, that is false advertising, and in 2026 livestream rooms were punished for it. Buying them as decoration is fine. +- Benefit: The classic classroom experiment showed that everyone received the same vague "personal analysis" yet most thought it accurate; the phenomenon is called the Barnum effect. It explains where the feeling of "things did go better after I put it on" comes from. What that author concluded is that「被本人接受不能证明解释正确」. Official case: the State Administration for Market Regulation published on 2026-6-4 a batch of typical cases of "cracking down on fakes and clearing out the sources" in the traditional crafts market. One was a jewellery shop in Sihui, Guangdong, which invented the material of the goods in its livestream room. It also claimed that the goods could "prevent and cure rheumatism, raise immunity, improve sleep quality and unblock blood vessels". The shop misled and deceived consumers by staging scripted performances. This breaches Article nine, first paragraph of the Law Against Unfair Competition (nationwide) +- Evidence grade: B +- Sources: Forer BR (1949). The fallacy of personal validation: A classroom demonstration of gullibility. Journal of Abnormal and Social Psychology. ; State Administration for Market Regulation (国家市场监督管理总局) (2026). The State Administration for Market Regulation Publishes a Batch of Typical Cases of "Cracking Down on Fakes and Clearing Out the Sources" in the Traditional Crafts Market（市场监管总局公布一批传统工艺市场「打假清源」典型案例）. +- Notes: The conclusion is one sentence: its claimed effects cannot be checked, and it has shown no predictive power at all. No belief or custom is being judged here, and this is not aimed at giving gifts, at keeping a memento, or at passing something down in the family. Two things have to be kept apart: the effects a merchant claims are set out in writing and can be complained about, while claims like "turning your luck" have no research behind them and cannot be proved false either, so only your own budget can hold them in check. Before buying, treat it as an ordinary piece of jewellery and ask yourself one question: "if it changed nothing at all, would I still pay this?" For how to check the material and the certificate, and why treating it as an investment does not work out, see items 33 and 34 in section 5 (bead bracelets and watches counted as money spent, and jade recognised only from a CMA report). For paying for fortune telling, see item 15 in this section (fortune telling, tarot and horoscopes). +### 23. Do not expect buying things to improve your mood or your sense of who you are + +- Cost: No money. The hard part is not ordering when you feel low. +- In plain language: The more a person values money and material things, the less happy they say they are, but the association is weak and it does not show what leads to what. Other research has found that spending money on an experience makes people happier than spending it on things. So when you feel low, spending on going out or on a meal with someone is more likely to help than filling a shopping basket. +- Benefit: One meta-analysis (pooling many studies) covered 259 independent samples and 753 effect sizes. The result: the more a person values material things, the lower the well-being they report. On the most commonly used multidimensional scale the correlation is r = -0.19 (ρ = -0.24 after correction). On scales that measure only the single aspect "valuing money", the association is even weaker, r = -0.08 to -0.11. Split by which outcome is being examined, the strength differs. The strongest associations are with risky health and consumption behaviour and with negative self-evaluation, r = -0.28 to -0.44. The weakest are with life satisfaction and negative mood, r = -0.13 to -0.15. The authors also ran a mediation analysis (to see what the effect travels through), and the result suggests that this negative association may come from unmet psychological needs. Another study: in two surveys of different populations, respondents said that money spent on an experience made them happier than money spent on things. In the laboratory experiments that followed, the good mood recalled after spending money on an experience was also higher than that recalled after spending it on things (the national standard does not apply; both studies used foreign samples) +- Evidence grade: B +- Sources: Dittmar H, Bond R, Hurst M, Kasser T (2014). The relationship between materialism and personal well-being: A meta-analysis. Journal of Personality and Social Psychology, 107(5), 879-924. ; Van Boven L, Gilovich T (2003). To do or to have? That is the question. Journal of Personality and Social Psychology, 85(6), 1193-1202. +- Notes:Disputed: most of the studies included asked a group of people at one point in time whether two things moved together, which cannot show which came first. The authors say in their own conclusion that more experiments and long-term follow-up are needed to see the direction. So this cannot be read as "buying things makes people unhappy"; it can only be read as "do not expect buying things to make you happy". And the association is weak: r = -0.19 accounts for a few percentage points of difference, and people differ from each other a great deal. Both studies used foreign samples, and there is no comparable data for the Chinese population that can be cited. Nothing here argues against spending money on something you like. What is being argued against is using shopping as a way of managing your mood, because that leaves your first reaction the next time you feel low still being to order something. The specific actions for saving money are in section 5. What really has evidence behind it when you feel low is in the second half of section 22 and in section 29. For budget added in order to be "one notch above other people", see item 24 in this section (moving up one notch). +### 24. Do not spend more on a bigger house, a better car or a better circle in order to move up a notch among the people around you + +- Cost: No money. All you have to do is ask one more question before adding to the budget: is this extra price buying the thing itself, or my position among these people? The hard part is not keeping up when everyone around you has changed. +- In plain language: Among people with the same income, the more a neighbour earns, the lower this person reports their own well-being. This happens only between neighbours who have the same level of education and who see each other often. So the extra money you pay to move somewhere everyone is better off than you buys you a move down a notch. Still, when incomes rise together people do feel better, so the absolute level still counts. +- Benefit: Two waves of the US National Survey of Families and Households revisited the same people. The authors matched the data to local average income. Local average income (PUMA) is estimated from the 5% sample of the 1990 census. Once the person's own income and a set of personal characteristics are taken out, the result is that the higher the neighbours' income, the lower the well-being they report. The effect is large, it is stable under a different method, and it is highly significant. What the authors write is that the fall in well-being from "neighbours' income rising" is of the same order of magnitude as the fall from "own income falling by the same amount". Once it is changed so that the same person is compared with themselves over time (this is called adding an individual fixed effect), the sign of the coefficient does not change and its size is similar; only the significance drops to the 10% level. Restricted to comparisons within the same state and the same survey wave (this is called state and wave fixed effects), the result is essentially unchanged. That shows it is not caused by "people who move to poor areas being happier to begin with", and not caused by some unmeasured factor at state level. The effect is stronger among people who often see their neighbours, and is not significant among people whose friends all live outside the neighbourhood. There is no difference between renters and homeowners, which does not support an explanation of the kind "prices are high in expensive areas, so there is less left to spend". Well-being falls only with the income of neighbours who have the same level of education, and barely reacts to the income of neighbours with a different level of education. And it works mainly through "satisfaction with my own material situation", not through satisfaction with family life or anything else. Finally, if a person's own income and the neighbours' incomes all rise by the same proportion, well-being still rises +- Evidence grade: B +- Sources: Luttmer EFP (2005). Neighbors as Negatives: Relative Earnings and Well-Being. The Quarterly Journal of Economics, 120(3), 963–1002. ; Luttmer EFP (2004). Neighbors as Negatives: Relative Earnings and Well-Being. NBER Working Paper No. 10667（本条引用的表述与数量级按这一版逐字核对）. +- Notes: There are several reasons for grading this B: the data come from a questionnaire that revisited the same people again and again, there is only one study, and the sample is foreign; what is being examined is the well-being people report themselves, not money. The author did use statistical means to remove some sources of interference, such as comparing the same person over time and comparing within the same state and wave, but it is still not a randomised trial. There is no comparable data for the Chinese population that can be cited. The benefit is graded "medium" because it follows the standard set in item 23 in this section (do not expect buying things to improve your mood). The money saved really could run into the tens of thousands. But there is no figure behind the claim that "saving this much would buy so much well-being", so it is not recorded as "large" at that scale. Nothing here argues against spending money on a bigger house, a quieter neighbourhood or a shorter commute; those buy the thing itself. The single reason to hold back is "everyone at work has changed, I cannot be worse off than they are". For what a home should be ranked on, see item 18 in section 4 (put commuting time near the top when you choose where to live). For buying things to adjust your mood, see item 23 in this section (do not expect buying things to improve your mood). For always comparing yourself upwards online, see item 21 in section 3 (do not treat "how other people are doing" as required daily reading). +### 25. Do not believe that willpower is like a muscle and runs out + +- Cost: No money. All you have to change is what you say. +- In plain language: The popular line goes: do one thing that costs self-control first and you are more likely to give in on the next, so save your willpower. 23 laboratories with 2141 people reran this experiment on a single standard procedure, and the result is almost exactly zero. Stop using "I ran out of willpower today" to explain yourself, and stop paying for courses that teach you to hoard willpower. +- Benefit: A preregistered multi-laboratory replication, with 23 laboratories and 2141 participants in all, repeated the sequential task paradigm using the standardised procedure of Sripada and colleagues: first do one task that consumes self-control, then do a second task that requires self-control. The pooled result shows that the ego-depletion effect is very small, and that the 95% credible interval includes 0 (d = 0.04, 95% CI −0.07 to 0.15). On that basis the authors discuss whether the model that self-control is a limited resource still holds +- Evidence grade: A +- Sources: Hagger MS, Chatzisarantis NLD, Alberts H, et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science, 11(4), 546–573. +- Notes: What is being rejected is the model that self-control is a resource that gets used up, not the fact that people get tired. Not enough sleep and working too long in a row do lower your efficiency, which is a different matter and belongs with the items on sleep and hours in section 3. The approaches that really do have randomised trials behind them are changing the environment and changing how you write things down; see item 10 in section 4 (move what you would rather not touch further away) and item 1 in section 4 (write "when, where, and if this comes up then that"). The other items in this section count money; this one counts time and energy, and the two are not converted into each other. +### 26. Do not expect eating breakfast or 16:8 fasting to help you control your weight; pick an eating schedule you can keep up long term + +- Cost: No money. What you save is the daily effort of arguing with yourself about meal times. +- In plain language: The popular line is that skipping breakfast makes you overeat at lunch and dinner, so you end up eating more in the day. The trials do not show that: people who skipped breakfast ate more than two hundred calories less a day and weighed very slightly less. Over a year, 16:8 made no difference in weight against simply eating less. People who skip breakfast long term really do have about a sixth more cardiovascular disease, but there is no difference in all-cause mortality. +- Benefit: First, the claim that "skipping breakfast makes you eat more later". One meta-analysis pooled 13 randomised trials. People assigned to eat breakfast took in 259.79 kcal more a day than those who did not (95% CI 78.87–440.71, the credible range). They weighed 0.44 kg more (0.07–0.82). People who skipped breakfast did not make it up at lunch or dinner. These trials all had short follow-up, an average of 7 weeks for weight and 2 weeks for intake. Next, approaches such as 16:8, where you only eat during a fixed set of hours. One trial put 116 overweight adults on it for 12 weeks. The group eating only between 12 noon and 8 in the evening weighed -0.26 kg against the group eating three regular meals a day (-1.30 to 0.78), which is not significant. Another trial, in Guangzhou, followed 139 patients with obesity for 12 months. Time-restricted eating plus calorie control lost 1.8 kg more than calorie control alone (-4.0 to 0.4), also not significant. Finally, whether skipping breakfast for a long term does harm. In a pooling of 2.38 million people, people who skipped breakfast had about 17% higher risk of cardiovascular disease (OR 1.17, 1.09–1.26) and about 49% higher cardiovascular mortality (OR 1.49, 1.20–1.84). But these are record-only studies with no grouping. In the original study with 17 to 23 years of follow-up, the risk ratio for all-cause mortality was 1.19, with a credible range of 0.99 to 1.42, which crosses 1 and so cannot be called a difference. +- Evidence grade: A +- Sources: Sievert K, Hussain SM, Page MJ, et al. (2019). Effect of breakfast on weight and energy intake: systematic review and meta-analysis of randomised controlled trials. BMJ. ; Lowe DA, Wu N, Rohdin-Bibby L, et al. (2020). Effects of Time-Restricted Eating on Weight Loss and Other Metabolic Parameters in Women and Men With Overweight and Obesity. JAMA Internal Medicine. ; Liu D, Huang Y, Huang C, et al. (2022). Calorie Restriction with or without Time-Restricted Eating in Weight Loss. New England Journal of Medicine. ; Zhang H, Zhang S, Liu Y, et al. (2025). The association between skipping breakfast and cardiovascular disease: a meta analysis. Frontiers in Cardiovascular Medicine. ; Rong S, Snetselaar LG, Xu G, et al. (2019). Association of Skipping Breakfast With Cardiovascular and All-Cause Mortality. Journal of the American College of Cardiology. ; Fernandes-Alves D, Teixeira GP, Guimarães KC, Crispim CA (2026). Systematic Review and Meta-analysis of Randomized Clinical Trials Comparing Time-Restricted Eating With and Without Caloric Restriction for Weight Loss. Nutrition Reviews. ; Sun T, Zhang L, Lu Y, et al. (2024). Non-linear relationship between the first meal time of the day and gallstone incidence in American adults. Frontiers in Nutrition. ; Liu T, Wang Y, Wang X, et al. (2023). Habitually Skipping Breakfast Is Associated with the Risk of Gastrointestinal Cancers: Evidence from the Kailuan Cohort Study. Journal of General Internal Medicine. +- Notes:Disputed. The 17% on cardiovascular disease comes from record-only studies with no grouping. People who skip breakfast were already more likely to smoke, drink, sit for long hours and earn less, and those factors cannot be fully corrected for. So there is no way to say how much of that 17% is caused by breakfast itself. The all-cause mortality figure crossed 1 in any case. There is also a dispute about 16:8: pooling 30 trials with 1341 people, and on the condition that both groups ate the same amount, the time-restricted eating group lost 1.46 kg more (-2.65 to -0.26), at the cost of 0.41 kg more lean body mass lost. But most of those trials are short, and the longest and strictest one, the twelve-month trial, found no difference. Long periods of fasting are also connected with the gallbladder. In one cross-sectional survey of 6,547 people, each hour that the first meal of the day was delayed raised the share found to have gallstones by about 5% (OR 1.05, 1.02–1.08); people who put their first meal between 9 and 14 had about 49% more than those who ate earlier (OR 1.49, 1.24–1.77). The mechanism is that the longer bile sits in the gallbladder, the more likely crystals come out of it. But this is a survey at one point in time, and it may be that the gallbladder problem came first and changed the eating habits. The Chinese Kailuan cohort followed 5.6 years and recorded 369 gastrointestinal cancers: people who skipped breakfast had 2.32 times the risk of colorectal cancer (1.34–4.01) and 5.43 times the risk of gallbladder and extrahepatic bile duct cancer (1.34–21.93). The credible range on that second figure is as wide as 21.93, which means the events are so few that it can only be treated as a lead. Taken together, for weight control the effect of meal timing is small, and it is not worth spending effort on it. As for whether skipping breakfast altogether is good for you in the long run, all the evidence is association and cannot show cause and effect, but the direction is consistently unfavourable; if you want to be safe, eat normally, just do not expect it to help you lose weight. The main causes of gastritis and gastric ulcer are not an empty stomach but Helicobacter pylori and long-term painkillers; for what to test for, see item 23 in section 1 (test for Helicobacter pylori and eradicate it if positive). For gallstones found on a check-up that have never hurt, see item 20 in this section (do not remove a gallbladder preventively because a check-up found gallstones and you have never felt pain). For weight control, what really has evidence behind it is what you eat and how much; see item 28 in section 2 (eat 5 portions of fruit and vegetables a day), item 29 in section 2 (eat less ultra-processed food) and item 33 in section 2 (keep BMI between 20–25). This does not apply to people with diabetes, people taking hypoglycaemic drugs, or people who are pregnant, since a missed meal can cause hypoglycaemia; follow the doctor's instructions. Extreme fasting and purging are a different matter; see item 1 in section 28 (do not use extreme dieting, fasting or purging to control your weight). +### 27. Do not believe that masturbation harms the body or that abstaining improves you; what matters is whether it has got in the way of sleep, work and relationships + +- Cost: No money. What you save is the money on "tonify the kidney" remedies and "abstinence" courses, and the long worry of blaming yourself. +- In plain language: No study says masturbation harms the body. A large American study found the opposite: men who ejaculated more often had about a fifth less prostate cancer. A British study found that men with a higher frequency in their twenties and thirties had higher risk, so neither can be used as a reason to exercise your body. Many people who think they are "addicted to pornography" feel that way more because the activity is at war with their own morality. +- Benefit: Start with the body. One American study followed 31925 men up to 2010. Each man reported himself how many times a month he ejaculated, and they were grouped by that number. Compared with men reporting 4 to 7 times a month, men reporting more than 21 times had a prostate cancer HR 0.81 (95% CI 0.72–0.92, about 19% lower), worked out using the frequency at age 20 to 29. Using the frequency at age 40 to 49, HR 0.78 (0.69–0.89, about 22% lower). Most of the reduction was in low-risk prostate cancer. On the other side, a British case-control study compared 431 patients diagnosed before the age of 60 with 409 controls. Men who masturbated frequently in their twenties and thirties had higher risk, while men who were frequent in their fifties had lower risk. Now the pornography material. A systematic review with a meta-analysis argues that many people feel they are "addicted" more because the behaviour does not line up with what they believe, and not only because of how much they watch (United States / United Kingdom) +- Evidence grade: B +- Sources: Rider JR, Wilson KM, Sinnott JA, et al. (2016). Ejaculation Frequency and Risk of Prostate Cancer: Updated Results with an Additional Decade of Follow-up. European Urology, 70(6), 974–982. ; Dimitropoulou P, Lophatananon A, Easton D, et al. (2009). Sexual activity and prostate cancer risk in men diagnosed at a younger age. BJU International, 103(2), 178–185. ; Grubbs JB, Perry SL, Wilt JA, Reid RC. (2019). Pornography Problems Due to Moral Incongruence: An Integrative Model with a Systematic Review and Meta-Analysis. Archives of Sexual Behavior, 48(2), 397–415. +- Notes:Disputed. Both prostate cancer studies are record-only with no control group, and the frequency is the man's own recollection, so the conclusion can only go as far as "no evidence of harm was seen"; it cannot be turned into advice to do it more. The authors of the British study say themselves that the fifties finding may be that people became ill first and then ejaculated less often. The abstract of the pornography paper gives no effect size, so only the direction is stated here. The benefit is graded small because what falls is mainly low-risk prostate cancer, not death. The signals that really should send you to a psychiatrist or a clinical psychologist are these: staying up late over it, missing work because of it, it damaging your relationship with your partner, and not being able to stop when you want to. For staying up late, see item 9 in section 3 (do not stay up late for games, short videos or pornography). If you are worried that watching pornography has caused erectile problems, get that checked first following item 28 in section 1 (if erectile function has gone wrong, check the cardiovascular system first). For the legal consequences of posting in groups or selling material, see item 4 in section 9 (watching pornography by yourself is your own business). The beneficiary is you yourself. +### 28. Do not pay for "sexual orientation conversion" or "homosexuality treatment", and do not send family members either + +- Cost: No money. What you save is the money going into conversion organisations, psychological clinics and "ex-expose yourself" businesses. The hard part is the family insisting on sending someone. +- In plain language: The expert panel of the American Psychological Association, having read the studies, said that this kind of conversion cannot change sexual orientation and may do harm. In one American survey, people who had been through conversion had nearly ninety per cent more suicidal ideation. In China a psychiatric hospital forcibly admitted a man in the name of "sexual preference disorder", and the court found it an infringement and awarded him 5000 yuan. +- Benefit: The American Psychological Association's 2009 task force report says that efforts to change sexual orientation are「不太可能成功，并且有一定的伤害风险」. On sending a minor into a coercive or residential conversion programme, the report advises that neither parents nor children should consider it. One nationally representative American survey visited 1518 sexual minority adults, about 7% of whom had been through conversion, and eight in ten of those came from religious people. After taking out age, adverse childhood experiences and similar factors, people who had been through conversion had an odds ratio of 1.92 for suicidal ideation at some point in their life (95% CI 1.01–3.64, about ninety per cent higher). Planning suicide OR 1.75 (1.01–3.06). Suicide attempts with no injury or minor injury OR 1.88 (1.01–3.50). Another study visited 245 people aged 21 to 25. Among those not converted during puberty, 22.0% had attempted suicide. Where parents themselves tried to change their child's sexual orientation, the figure was 48.1%. Where parents also took the child to a therapist or a religious figure, it was 62.8%. The World Health Organization's classification ICD-10 states that「性取向本身不被视为一种障碍」. In 2017 a court in Yicheng District, Zhumadian, Henan heard a case in which a psychiatric hospital forcibly admitted a man in the name of "sexual preference disorder". The court found that he「病情一般，也无自杀、伤人、毁物等行为或危险，不符合应当强制治疗的条件」. The court ordered the hospital to apologise publicly and pay 5000 yuan in mental comfort compensation (United States / China) +- Evidence grade: B +- Sources: American Psychological Association Task Force on Appropriate Therapeutic Responses to Sexual Orientation (2009). Report of the Task Force on Appropriate Therapeutic Responses to Sexual Orientation. ; Blosnich JR, Henderson ER, Coulter RWS, Goldbach JT, Meyer IH. (2020). Sexual Orientation Change Efforts, Adverse Childhood Experiences, and Suicide Ideation and Attempt Among Sexual Minority Adults, United States, 2016–2018. American Journal of Public Health, 110(7), 1024–1030. ; Ryan C, Toomey RB, Diaz RM, Russell ST. (2020). Parent-Initiated Sexual Orientation Change Efforts With LGBT Adolescents: Implications for Young Adult Mental Health and Adjustment. Journal of Homosexuality, 67(2), 159–173. ; World Health Organization. ICD-10 Version:2019, F66 Psychological and behavioural disorders associated with sexual development and orientation. ; Standing Committee of the National People's Congress (全国人大常委会) (2018, amended). Mental Health Law（精神卫生法）(Articles twenty-three, thirty and seventy-eight). ; Guangxi Zhuang Autonomous Region Procuratorate website reprinting the Beijing News (广西壮族自治区人民检察院网站转载新京报) (2017). Final verdict in the Zhumadian "coercive treatment of homosexuality" case（驻马店「同性恋强制治疗案」终审）. ; the other side: Sullins DP. (2022). Sexual Orientation Change Efforts Do Not Increase Suicide: Correcting a False Research Narrative. Archives of Sexual Behavior, 51, 3377–3393. ; Blosnich JR, Coulter RWS, Henderson ER, et al. (2023). Correcting a False Research Narrative: A Commentary on Sullins (2022). Archives of Sexual Behavior, 52(3), 885–888. +- Notes:Disputed. Sullins 2022 recalculated the same data and took out the suicidal ideation that was already there before the conversion. Worked out that way, conversion is no longer positively associated with suicidal ideation (OR 0.44, 0.20–0.94). The organisation Sullins belongs to is opposed to homosexuality, and the original authors wrote a reply. So how much harm there is is still disputed, but no evidence has come back from the other side on the point that it cannot work. It is graded B because the harm figures all come from surveys asking people to recall afterwards. As far as the law is concerned, inpatient treatment follows the principle of consent, and only a person with a severe mental disorder who is a danger to themselves or others can be admitted involuntarily. Psychological counsellors are not allowed to carry out psychotherapy. Deliberately sending someone who is not mentally ill into hospital for treatment carries a liability to compensate. If you are in long-term pain about this, go to a psychiatric or clinical psychology department; the goal is to relieve the pain. For what a family should do once a child has said it, see item 15 in section 30 (a child says they like the same sex). The beneficiaries are you and your family. \ No newline at end of file diff --git "a/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" "b/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" new file mode 100644 index 00000000..719d8d66 --- /dev/null +++ "b/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" @@ -0,0 +1,180 @@ +[← 回总目录](../README.md) + +# 7. 没钱的时候怎么活 + +这一节算的是钱和保障。能领多少钱、能免多少钱、能得到什么服务、要满足什么条件，都写在里面，不折算成寿命。政策每年都在变，各地也不一样，所以金额和门槛都以你所在地当年公布的为准。条目的排序是：不花钱、手续少、见效快的排在前面。要攒长期资格的、各地差别大的排在后面。只凭经验、没有官方文件的放在最后。 + +[← Back to the contents](../README.md) + +# 7. How to get by with no money + +This section counts money and protection. How much you can receive, how much you can be spared, what services you can get and what conditions you have to meet are all written down here, and nothing is converted into years of life. The policies change every year and differ from place to place, so go by the amounts and the thresholds published for your own area in that year. The items are ordered like this: those that cost no money, need little paperwork and work fast come first. Those that take years of qualifying, and those that differ a lot between places, come later. Those resting on experience alone, with no official document, are put last. + +### 1. If you lose your job, apply online for unemployment insurance benefits first + +- Cost: No money. You can file from your phone, and any of three channels works: the「掌上 12333」app, the electronic social security card mini-program in WeChat or Alipay, and the national social insurance public service platform. A social security card or an identity card is all you need, and in most places you do not have to turn up in person. The precondition is that you have completed unemployment registration. +- In plain language: If you and your employer have paid in for a full year and you did not resign of your own accord, completing unemployment registration lets you receive money. A full 1 year but under 5 years gives you at most 12 months. 5 years but under 10 years gives at most 18 months. 10 years or more gives at most 24 months. While you are receiving it, unemployment insurance pays your basic medical insurance for employees. +- Benefit: Three conditions have to be met together. First, you and your employer must have paid unemployment insurance for a cumulative total of 1 year. Second, the job loss must not be one you asked for yourself (the law calls this "interruption of employment not of the person's own will"). Third, you must have completed unemployment registration and be willing to look for work. How long you can receive it for depends on how long you have paid in. A cumulative total of 1 year but under 5 years gives at most 12 months. 5 years but under 10 years gives at most 18 months. 10 years or more gives at most 24 months. The monthly amount is set by the provincial government and differs by province. The amount is below the local minimum wage and above the urban means-tested income support. The national goal is to raise it gradually to 90% of the minimum wage standard. While you are receiving unemployment insurance benefits you still take part in basic medical insurance for employees and enjoy basic medical insurance benefits. The basic medical insurance contribution you would otherwise have to pay comes out of the unemployment insurance fund, and you pay nothing yourself (nationwide, regulations in force since 1999) +- Evidence grade: A +- Sources: State Council (国务院) (1999). Regulations on Unemployment Insurance（失业保险条例）(State Council Order No. 258, Articles fourteen, seventeen and eighteen). ; Ministry of Human Resources and Social Security. Guide to Applying for Unemployment Insurance Benefits on the National Human Resources and Social Security Government Service Platform（全国人社政务服务平台·失业保险待遇申领办事指南）. ; Department of Employment of the National Development and Reform Commission (2022). The unemployment insurance benefit standard will gradually be raised to 90% of the minimum wage（失业保险金标准，将逐步提高至最低工资的 90%）. ; Liaoning Provincial Department of Human Resources and Social Security (2019). Notice on Further Adjusting the Unemployment Insurance Benefit Standard（关于进一步调整失业保险金标准的通知）:「为落实《人力资源社会保障部财政部关于调整失业保险金标准的指导意见》(人社部发〔2017〕71号〕"逐步将失业保险金标准提高到最低工资标准的90%"等要求」. ; Standing Committee of the National People's Congress (2010). Social Insurance Law（社会保险法）(Article forty-eight). (State Taxation Administration law database) +- Notes: If you resigned of your own accord, that does not count as "not of your own will", and you get nothing. While you are receiving benefits, unemployment insurance pays your basic medical insurance for employees. The amounts differ a great deal between provinces, so go by what the place you are insured in publishes. "90% of the minimum wage" comes from the 2017 guiding opinion of the Ministry of Human Resources and Social Security and the Ministry of Finance (人社部发〔2017〕71 号), whose words are「逐步将失业保险金标准提高到最低工资标准的90%」. That is a goal, not what is paid now. The original page on the ministry's website would not open on this machine, so that sentence was checked against the quotation in the Liaoning provincial document. +### 2. If your wages are owed, complain to labour inspection first, then apply for labour arbitration; neither route costs anything, and most cases resolve within months + +- Cost: No money. Neither a complaint nor arbitration costs anything. You need to assemble the employment contract, the payslips, the attendance records, the transfer records and your chat logs, and make 1 to 3 trips to the human resources bureau or the arbitration commission. There is a 1-year time limit on applying for arbitration (the law calls it the arbitration time limit). Someone still working for the same employer with wages in arrears is not bound by that 1 year. Someone who has already left must apply within 1 year of the leaving date. The hard part is going back again and again and assembling evidence piece by piece, and losing heart in the middle. +- In plain language: Neither route costs money. Call 12333 or complain to the labour inspection office at the human resources bureau; within 60 working days of the case being opened the investigation is finished, the company can be ordered to pay by a deadline, and if it still does not pay it has to pay 50% to 100% extra. If that does not work, apply for labour arbitration, which is closed within 45 days. But "closed" does not mean "in your account": if the company has no assets and the owner has run off, you may win and still get nothing. +- Benefit: The first step is to complain. Call 12333, or complain to the labour inspection office of the local human resources bureau. The law requires the investigation to be finished within 60 working days of the case being opened, with a possible further extension of 30 working days where the circumstances are complicated. If it is substantiated, the company can be ordered to pay within a deadline. If it still does not pay by the deadline, it has to pay an additional compensation of 50% to 100% of the amount owed. In 2024 inspection agencies nationwide handled 137,000 cases and recovered 86.5 billion yuan of wages for 654,000 people. If you have an IOU or a wage sheet and the facts of the arrears are clear, you can also apply directly to a basic-level court for a payment order. A payment order is a command to pay that the court issues to the company, and the court must issue it within 15 days of accepting the application. The second step is to apply for arbitration. The law requires the case to be closed within 45 days of acceptance, with a possible further extension of 15 days. Those claiming wages can apply for advance enforcement: part of what is awarded is enforced before the whole case has been heard. Where the money owed does not exceed 12 times the local monthly minimum wage, the award is final on issue and the company cannot sue. In 2024, 4,257,000 cases were handled nationwide and 4,156,000 were closed. The amount closed was 934.7 billion yuan. The mediation success rate was 79.6%, the arbitration closure rate 98.2% and the rate of awards going no further 73.7%. When people say this drags on for "half a year to a year", they mean the cases that go from arbitration into court: 6 months at first instance (3 months for those using the summary procedure), plus 3 months at second instance. In 2024 courts concluded 88,000 cases of claims for labour remuneration and recovered 285.4 billion yuan, and 998 people were sentenced for the crime of refusing to pay labour remuneration (nationwide) +- Evidence grade: A +- Sources: Ministry of Human Resources and Social Security (人力资源社会保障部) (2025). 2024 Annual Statistical Bulletin on the Development of Human Resources and Social Security（2024 年度人力资源和社会保障事业发展统计公报）. ; Supreme People's Court (最高人民法院) (2025). Work Report of the Supreme People's Court（最高人民法院工作报告）. ; State Council (国务院) (2004). Regulations on Labour Security Inspection（劳动保障监察条例）(State Council Order No. 423, Articles nine, seventeen and twenty-six). ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Law on Mediation and Arbitration of Labour Disputes（劳动争议调解仲裁法）(主席令第八十号, Articles twenty-seven, forty-three, forty-four, forty-seven, fifty and fifty-three). ; Standing Committee of the National People's Congress (2012, amended). Labour Contract Law（劳动合同法）(Articles thirty and eighty-five). ; Standing Committee of the National People's Congress (2023, amended). Civil Procedure Law（民事诉讼法）(Articles one hundred and nine, one hundred and fifty-two, one hundred and sixty-four, one hundred and eighty-three, two hundred and twenty-five and two hundred and twenty-seven). ; Standing Committee of the National People's Congress (2011). Amendment (Eight) to the Criminal Law（刑法修正案（八））Article forty-one, that is, Article two hundred and seventy-six, paragraph one, of the Criminal Law（刑法第二百七十六条之一）. ; Supreme People's Court (2013). Interpretation on Several Questions of Law Concerning the Application of Law in Cases of Refusal to Pay Labour Remuneration（关于审理拒不支付劳动报酬刑事案件适用法律若干问题的解释）(法释〔2013〕3, Article three). ; State Council (2019). Regulations on the Payment of Wages to Migrant Workers（保障农民工工资支付条例）(State Council Order No. 724, Articles ten, twenty-six, thirty, thirty-one, thirty-two and forty-one). ; Standing Committee of the National People's Congress (2021). Legal Aid Law（法律援助法）(Articles thirty-one and forty-two). +- Notes: A complaint and arbitration are useful, but you have to keep "the case is closed" and "the money has arrived" apart. A 98.2% closure rate only says the procedure has run its course. If the company has no assets and the owner has run off, you can win and still get nothing. The Supreme People's Court publishes only the total actually enforced (2.3 trillion yuan in 2024) and says that the "rate of completed enforcement stays high"; there is no official figure for how much of the wage arrears cases ended up in the pockets of the people who won. The bulletin does not define "rate of final arbitration awards", so this item reads it as "the award was not taken to court afterwards". Wage arrears above a certain amount are enough for criminal prosecution. One case is owing one person more than 3 months, with an amount from 5000 to 2 万元 or more. The other is owing 10 or more people, a cumulative 3 万 to 10 万元 or more. In both situations, refusal to pay after being ordered to is enough for a case to be opened. The higher courts and procuratorates of each province set and publish the specific amounts within that range, so look up your own province's figure before complaining and raise it in person. Migrant workers on a site also have a dedicated wage account and the rule that the main contractor pays first, so going to the main contractor works better than going to the foreman. Legal aid for migrant workers claiming wages does not require an economic hardship check. The texts of the Labour Contract Law and the Civil Procedure Law are quoted from the reprints on the State Taxation Administration and Guizhou provincial market regulation websites (both gov.cn), because the pages on the National People's Congress website would not open. Amendment (Eight) to the Criminal Law failed to fetch on the National People's Congress website, so it was opened with curl and checked instead. +### 3. If you cannot afford to sue, apply for legal aid; wage claims, support payments and work injury cases are already within scope + +- Cost: No money. Legal aid charges the person receiving it not one yuan. +- In plain language: Claiming wages, claiming support payments for parents, claiming social insurance or means-tested income support, and losses from having done a brave and selfless act can all get free legal aid. Call 12348, or go to the local legal aid centre. You no longer have to go back to your home town to get a certificate of economic hardship; stating your situation truthfully yourself is enough. Where the death penalty may be imposed, and defendants who are blind, deaf, mute or minors, are assigned a lawyer as of right. +- Benefit: The State Council's regulations set out six kinds of civil case covered by legal aid. The text of the regulations is:「依法请求国家赔偿的；请求给予社会保险待遇或者最低生活保障待遇的；请求发给抚恤金、救济金的；请求给付赡养费、抚养费、扶养费的；请求支付劳动报酬的；主张因见义勇为行为产生的民事权益的」. Put plainly: claiming state compensation, claiming social insurance benefits or means-tested income support, claiming a pension or relief payment, claiming support payments for parents, children or dependants, claiming wages, and claiming civil rights arising from having done a brave and selfless act. On the criminal side, defendants who may be sentenced to death, and defendants who are blind, deaf, mute or minors, are assigned a lawyer regardless of their financial situation +- Evidence grade: A +- Sources: State Council (国务院) (2003). Regulations on Legal Aid（法律援助条例）(State Council Order No. 385, Articles ten, eleven and twelve). ; Ministry of Justice official (司法部有关负责人) (2021). Answers to Reporters' Questions on Implementing the Legal Aid Law（就贯彻实施法律援助法答记者问）. +- Notes: The Legal Aid Law has applied since 2022-1-1, and one thing changed. You used to have to「提交经济困难证明」. Now you only have to state your financial situation truthfully yourself, and no longer have to go back to your home town for a certificate. The way in is the legal aid centre or the public legal service centre in your area, and you can also call 12348. +### 4. If you have nowhere to turn, go to a rescue station; it provides food, a bed and a ticket home + +- Cost: No money. You can go to the local rescue management station yourself, or tell a police officer or the civil affairs department and ask them to take you there. On arrival you have to register your name and belongings truthfully. The hard part is having to tell a stranger that you have no money and nowhere to go. +- In plain language: A rescue station provides food and a bed, takes you to hospital if you are ill, and helps you contact your family. If you cannot afford to get home it issues a ticket. The stay is generally no more than 10 days. Coming and going are both voluntary; it is not a detention centre and it pays out no cash. The ticket is issued to your registered place of residence or your place of residence, so work out where you are going back to before you set off. +- Benefit: A rescue station provides food and shelter that meet hygiene requirements. Anyone who falls ill there suddenly is taken to hospital for treatment. It also helps you contact relatives and your employer. If you cannot afford the ticket back to your registered place of residence or your place of residence, it issues you a travel voucher, which is the train or boat ticket. Your registered place of residence is your home town; your place of residence is where you live now. The period of rescue is generally no more than 10 days. It can be extended while you wait for family to collect you, for treatment, or for placement. When police and other administrative bodies meet a person who is homeless and begging while carrying out their duties, they should tell that person where to seek help. For people with disabilities, minors, the elderly and those who have difficulty moving, they should guide and accompany them there (nationwide, since 2003) +- Evidence grade: A +- Sources: State Council (国务院) (2003). Measures for the Assistance and Management of Homeless and Begging Persons with Nowhere to Live in Cities（城市生活无着的流浪乞讨人员救助管理办法）(State Council Order No. 381, Articles five, six and seven). ; Ministry of Civil Affairs (民政部) (2003). Detailed Implementation Rules for the Measures for the Assistance and Management of Homeless and Begging Persons with Nowhere to Live in Cities（城市生活无着的流浪乞讨人员救助管理办法实施细则）(Ministry of Civil Affairs Order No. 24, Articles eleven and twelve). +- Notes: Seeking help and leaving are both voluntary; it is not a detention centre. What it provides is basic food and a bed, and no cash. The travel voucher is issued to your registered place of residence or your place of residence, so work out where you are going before you set off. +### 5. To find work, start with free public employment services and the day labour market, not with an agency that charges + +- Cost: No money. Register with a public employment service body at county level or above, such as an employment service centre or a human resources market. The local day labour market will also take registrations. If you do not know the address, call 12333 and ask. +- In plain language: Government bodies such as employment service centres and human resources markets introduce jobs, give vocational guidance, answer policy questions and register you for employment and unemployment, and none of it costs anything. The law also states in writing that they may not make money out of these services. Registration and publication of information at a day labour market are free as well. Do unemployment registration while you are there, because you need it both to receive unemployment insurance benefits and to be recognised as a person with difficulty finding work. +- Benefit: The law requires all of the following to be free: consultation on employment policies; information on where jobs are going and where workers are needed; vocational guidance and job placement; registering for employment and for unemployment; and employment assistance for people with difficulty finding work. These bodies may not carry out profit-making activities and may not make money out of these services. Registration and publication of job information for day labourers is also free to the public, and has been brought within the scope of public employment information services. Day labourers who have been out of work a long time (that is, who go a long time without finding work) and other day labourers in difficulty, such as those from low-income households, are to be given more help (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007, amended 2015). Employment Promotion Law（就业促进法）(Article thirty-five). ; Ministry of Human Resources and Social Security and four other departments (2022). Opinions on Strengthening the Construction of Day Labour Markets and Improving Job Search and Recruitment Services（关于加强零工市场建设 完善求职招聘服务的意见）(人社部发〔2022〕38). +- Notes: Unemployment registration is the preliminary step both for receiving unemployment insurance benefits and for being recognised as a person with difficulty finding work, so do it while you are there. For day labour paid off by the day, check the money on the spot, keep the other party's contact details and a record of the work done. +### 6. If something sudden happens, apply for temporary assistance + +- Cost: No money. Apply to the township government or the sub-district office and set out clearly what has happened at home, such as a fire, a traffic accident, a family member falling suddenly seriously ill, or an unavoidable cost that has suddenly gone up. After review and public notice, the county-level civil affairs department approves it. Prepare the documents from the list your local office gives you; it usually takes 1 to 2 trips. +- In plain language: If something sudden happens at home, such as a fire, a car crash or a family member falling suddenly seriously ill, and basic living costs are crushed for a time, you can apply to the township government or the sub-district office for temporary assistance. You do not have to be on means-tested income support already. More often than not it is a single payment, and the amount differs a great deal from place to place. +- Benefit: Two kinds of household can receive temporary assistance. One is households which, because of an accident or a family member falling suddenly seriously ill, cannot hold together even the most basic costs of food, shelter and medical care for a time. The other is households on means-tested income support whose unavoidable costs have suddenly risen beyond what they can bear. Which events are covered and how much money is given is set and published by the local government at county level or above (national system, amounts set locally, since 2014) +- Evidence grade: B +- Sources: State Council (国务院) (2014, revised 2019). Interim Measures on Social Assistance（社会救助暂行办法）(State Council Order No. 649, Articles forty-seven, forty-eight and forty-nine). +- Notes: You do not have to be on means-tested income support already. More often than not it is a single payment, and the amount differs a great deal from place to place, so go by what your area published that year. +### 7. If your income is below the local means-tested line, apply for means-tested income support + +- Cost: No money. A written application is made by the members of the household who live together, to the township government or sub-district office where their registered residence is. If it is awkward to do it yourself, you can ask the village committee or the residents' committee to do it for you. You have to accept a check of household income and property. It usually takes 2 to 3 trips, and you bring your identity card, household register and proof of income and property. The hard part is laying your finances open for inspection, and then waiting through the public notice. +- In plain language: Divide total family income by the number of people; if it is below the local means-tested line and the household property also falls within the rules, you can apply. At the end of 2024, urban means-tested income support averaged 798.1 yuan per person per month nationwide and rural support 593.9 yuan. You apply where your registered residence is. Having a house, a car or savings, or children who can support you, may mean you are not approved. +- Benefit: Divide total family income by the number of people; if it is below the local means-tested standard and the household property is also within the prescribed range, you can receive means-tested income support. At the end of 2024, the national urban average was 798.1 yuan per person per month and the rural average 593.9 yuan per person per month. Urban means-tested income support covered 6,250,000 people and rural support covered 33,615,000 people. The standard is set like this: take the previous year's per-capita consumption expenditure of local residents, multiply it by a set proportion, and adjust it again after an interval. Put plainly, take how much the average local resident spent the previous year and discount it (national average, different in each place) +- Evidence grade: A +- Sources: State Council (国务院) (2014, revised 2019). Interim Measures on Social Assistance（社会救助暂行办法）(State Council Order No. 649, Articles nine, ten and eleven). ; National Bureau of Statistics (国家统计局) (2025). Social Undertakings Developing Well and People's Wellbeing Rising: Statistical Report on Livelihood since the「十四五」period（社会事业向好发展 民生福祉不断增进——「十四五」以来社会民生统计报告）. ; Ministry of Civil Affairs and three other departments (民政部等四部门) (2024). Guiding Opinions on Further Improving the Determination and Adjustment of the Minimum Living Guarantee Standard（关于进一步做好最低生活保障标准确定调整工作的指导意见）(民发〔2024〕16). +- Notes: Means-tested income support is the last line of protection, and you have to apply where your registered residence is. If the house, car or savings at home exceed the local standard, or your children can support you, you may not be approved. Once it is approved there are other benefits as well: a subsidy towards resident medical insurance (see item 9 in this section for resident medical insurance), medical assistance when you are ill (see item 10 in this section for medical assistance), and no economic hardship check for legal aid (see item 3 in this section for legal aid). +### 8. If you hold a disability certificate, apply for the two disability subsidies + +- Cost: No money. Take your disability certificate to the sub-district office or the township and submit one application. +- In plain language: A disabled person in a household on means-tested income support can receive a living allowance. A disabled person whose level is graded one or two and who needs long-term care can receive a care allowance. Both can be received at the same time. Each province sets the standard and the differences are large, so ask the sub-district office. Take the second-generation disability certificate and go once to the sub-district office or the township. +- Benefit: The State Council set down two subsidies. The first is the living allowance for disabled people in difficulty, and the document defines who it is for as「低保家庭中的残疾人，有条件的地方可逐步扩大到低收入残疾人及其他困难残疾人」. That means disabled people in means-tested households come first, and where local resources allow it can be extended to low-income disabled people and other disabled people in difficulty. The second is the care allowance for severely disabled people, and the document defines who it is for as「残疾等级被评定为一级、二级且需要长期照护的重度残疾人」. The two can be received at the same time +- Evidence grade: A +- Sources: State Council (国务院) (2015). Opinions on Establishing a System for a Living Allowance for Disabled People in Difficulty and a Care Allowance for Severely Disabled People（关于全面建立困难残疾人生活补贴和重度残疾人护理补贴制度的意见）(国发〔2015〕52). +- Notes: The standard is set by the provincial government and differs a great deal from place to place, from a few tens of yuan to several hundred yuan a month. The application needs the second-generation disability certificate. If it is awkward for you to go in person, a family member can do it for you. +### 9. Do not let the 400 yuan a year of resident medical insurance lapse; households in difficulty get reductions + +- Cost: The part you pay yourself is no less than 400 yuan per person per year. That is the national standard for 2024 and 2025, and places may set a higher figure. You have to pay it during the fixed annual window for contributions. People in extreme difficulty are fully funded, and the government pays the whole amount. People on means-tested income support are funded a set amount, and the government pays a fixed part. +- In plain language: You pay from 400 yuan a year, and the government adds around 700 yuan a year per head. People in extreme difficulty are fully funded and people on means-tested income support are funded a set amount. Letting it lapse has a price: if you did not enrol during the annual window, or did not pay in consecutive years, there is a 3-month period after you pay in during which medical costs are not reimbursed, and each further year of non-payment adds one more month in principle. +- Benefit: Besides the part you pay yourself, the government adds a per-head subsidy. In 2024 it was no less than 670 yuan, and in 2025 it was raised by 30 yuan to no less than 700 yuan, while the personal contribution stays at no less than 400 yuan (医保发〔2025〕22, as reported in the provincial documents). The 2026 government work report says it will be raised by a further 24 yuan. For people on means-tested income support and people in extreme difficulty, medical expenses that meet the conditions and are left after medical insurance has paid can be claimed once more through medical assistance at a proportion of no less than 70%. The cost of a gap in payment is a waiting period. If you did not enrol during the annual contribution window, or did not pay in consecutive years, there is a fixed 3-month waiting period for benefits after you pay, during which medical costs are not reimbursed. For each further year of non-payment, in principle 1 more month is added (since 2025, nationwide) +- Evidence grade: A +- Sources: National Healthcare Security Administration (国家医保局) (2024). Policy Interpretation of the Notice on Doing Well in 2024 on Basic Medical Insurance for Urban and Rural Residents（关于做好 2024 年城乡居民基本医疗保障有关工作的通知政策解读）(医保发〔2024〕19). ; Hebei Provincial Healthcare Security Administration and others (2025). Notice on Doing Well in 2025 on Basic Medical Insurance for Urban and Rural Residents（关于做好 2025 年城乡居民基本医疗保障有关工作的通知）(冀医保发〔2025〕6, a local document). ; Anhui Provincial Healthcare Security Administration and others (2025). Notice on Doing Well in 2025 on Basic Medical Insurance for Urban and Rural Residents（关于做好 2025 年城乡居民基本医疗保障有关工作的通知）, reprinted by the Huaibei Municipal Healthcare Security Bureau:「按照国家医保局、财政部、国家税务总局《关于做好2025年城乡居民基本医疗保障有关工作的通知》（医保发〔2025〕22号）要求」「居民医保人均财政补助标准较上年提高30元，达到每人每年不低于700元」. ; General Office of the State Council (国务院办公厅) (2021). Opinions on Improving the Medical Insurance and Assistance System for Major and Serious Diseases（关于健全重特大疾病医疗保险和救助制度的意见）(国办发〔2021〕42). ; General Office of the State Council (2024). Guiding Opinions on Perfecting the Long-Term Mechanism for Enrolment in Basic Medical Insurance（关于健全基本医疗保险参保长效机制的指导意见）(国办发〔2024〕38). ; National Healthcare Security Administration (2026). The 2026 Government Work Report Mentions These Healthcare Security Tasks（2026 年政府工作报告提到这些医保工作）. +- Notes: The 2025 notice of the National Healthcare Security Administration (医保发〔2025〕22) has not been published on the administration's own website or on the Chinese government website. The two figures, 700 yuan and 400 yuan, agree in the documents of Hebei and Anhui, and the Anhui document states that they were set by this notice. While you are receiving unemployment insurance benefits, unemployment insurance pays your medical insurance for employees, so there is no need to enrol twice. +### 10. If you fall seriously ill, go through medical insurance, insurance for serious illness, medical assistance and out-of-area registration first, and stay away from online loans + +- Cost: No money. The order is fixed, and you work through it step by step. First, before being admitted in another province, register using the "national out-of-area medical registration" mini-program in the "National Healthcare Security Administration" WeChat account. Second, settle directly on discharge. Third, if what you still have to pay after reimbursement is too heavy, apply to the township government or the sub-district office for medical assistance. Fourth, if you are still working, take sick leave for the medical period, which is the period of being unable to work because of illness during which you cannot be dismissed; if it is a work injury, go through work injury insurance. Fifth, if you cannot pay the credit card, ask the bank to negotiate instalments. Sixth, if it is still not enough, use a personal fundraising platform designated by the Ministry of Civil Affairs. The main documents are the diagnosis certificate, the itemised bill and the settlement statement, and it usually takes 2 to 4 trips. The hard part is that you are ill and still have to go back and forth for the paperwork. +- In plain language: The order is fixed. Before being admitted in another province, register in the "National Healthcare Security Administration" WeChat account; settle directly on discharge and you do not have to pay the whole amount up front first. If you still cannot afford it after reimbursement, apply to the township or the sub-district office for medical assistance. If you are still employed, your length of service gives you a medical period of 3 to 24 months during which the company cannot dismiss you. Do not touch online loans or private high-interest borrowing. +- Benefit: The state brings basic medical insurance, insurance for serious illness and medical assistance together under what it calls the "three-tier system", which catches you layer by layer. For people on means-tested income support and people in extreme difficulty, medical expenses that meet the conditions are assisted at a proportion of no less than 70%. A severely ill patient who was not poor and has been made poor by medical bills can apply for assistance again on their own initiative. If you registered in advance for treatment in another province, you can settle directly on discharge, without paying the whole amount up front and then claiming in your home province. Being unable to work because of illness gives a medical period of 3 to 24 months, with the exact length set by length of service. During that period the company may not terminate the employment contract, and sick pay may not be below 80% of the local minimum wage. A worker with a work injury has a period of retention pay with no work, which is the time spent recovering and not working; the original wages and benefits continue to be paid, by the employer, monthly, and generally for no more than 12 months. Where a credit card balance exceeds what you can repay and you are willing to pay, you can negotiate an instalment plan with the bank and have the balance rescheduled, for up to 5 years, and once the agreement is signed the bank must stop its collection activity. A personal fundraising platform raising money for a named person must be designated by the civil affairs department of the State Council. The platform has to verify that the request for help is real. Money raised can only be paid to the person asking for help, or to an account that person provides (nationwide) +- Evidence grade: A +- Sources: General Office of the State Council (国务院办公厅) (2021). Opinions on Improving the Medical Insurance and Assistance System for Major and Serious Diseases（关于健全重特大疾病医疗保险和救助制度的意见）(国办发〔2021〕42). ; National Healthcare Security Administration (国家医保局) (2025). How Do I Get Reimbursed Directly for Treatment in Another Province?（跨省异地就医如何直接报销？）. ; Ministry of Labour (劳动部) (1994). Provisions on the Medical Period for Employees of Enterprises Who Fall Ill or Suffer Non-occupational Injury（企业职工患病或非因工负伤医疗期规定）(劳部发〔1994〕479, Articles two and three). ; Ministry of Labour (劳动部) (1995). Opinions on Certain Questions Concerning the Implementation of the Labour Law（关于贯彻执行《劳动法》若干问题的意见）(劳部发〔1995〕309, item 59). (reprinted by the Jilin provincial human resources department) ; State Council (国务院) (2010, revised). Regulations on Work Injury Insurance（工伤保险条例）(State Council Order No. 586, Articles thirty and thirty-three). ; Banking Regulatory Commission of China (银监会) (2011). Measures for the Supervision and Administration of Credit Card Business of Commercial Banks（商业银行信用卡业务监督管理办法）(CBRC Order No. 2 of 2011, Article seventy). ; Standing Committee of the National People's Congress (全国人大常委会) (2023). Decision on Amending the Charity Law（关于修改慈善法的决定）(Article one hundred and twenty-four). ; Ministry of Civil Affairs and four other departments (民政部等五部门) (2024). Measures for the Administration of Personal Fundraising Network Service Platforms（个人求助网络服务平台管理办法）(Order No. 75, Articles three, eleven, thirteen and seventeen). +- Notes: What proportion medical assistance pays and what the maximum is in a year are set by the place where you are insured. The medical period comes in six tiers, depending on how many years you have worked in total and how long you have been with this employer. Once the medical period is over, the company can lawfully end your contract, so see clearly which tier you are in. People rescued in an emergency are also on the list of those who need to register. The healthcare security administration's pages do not say whether an emergency counts as automatic registration or whether it can be done afterwards, so telephone the place where you are insured. Stay away from online loans and private high-interest borrowing. The courts protect only up to the line of 4 times the one-year LPR (for not borrowing at usurious rates, see item 15 in this section). The part above that line is not protected, and the principal still has to be repaid. Lying in a hospital bed you have no energy to bring such a lawsuit either. Acute serious illness is a different matter: you must still call 120 even if you have no money, the hospital may not refuse, pass the buck on or delay treatment, and the emergency costs for that part are paid by the emergency medical assistance fund; see item 9 in section 24 (an emergency must be treated first). The original page of Lao Bu Fa [1995] 309 would not open on the human resources ministry website, so its content was taken from a provincial reprint. +### 11. If you lose your identity card, replace it at once; if you need it urgently, get a temporary identity card + +- Cost: There is a document handling fee. This is a statutory charge, and the amount is set by the state, so go by what your area publishes. Apply at the police station where your registered residence is; it takes 1 trip. +- In plain language: For a replacement, the card is issued within 60 days of the day you hand in the form. If you need it urgently, apply on the spot for a temporary identity card, which is issued within 3 days and is valid for 3 months. Without an identity card you cannot get a job, book accommodation, open an account or receive a subsidy; it is the precondition for every bit of paperwork. +- Benefit: Counting from the day you submit the Application Form for a Resident Identity Card（居民身份证申领登记表）, the public security organ must issue the card within 60 days. During the period of a first issue, a renewal or a replacement, if you need the card urgently you can apply for a temporary resident identity card. The public security organ must issue it within 3 days of receiving the application, and it is valid for 3 months (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2003). Resident Identity Card Law（居民身份证法）(主席令第四号, Articles twelve and twenty). ; Ministry of Public Security (公安部) (2005). Measures for the Administration of Temporary Resident Identity Cards（临时居民身份证管理办法）(MPS Order No. 78, Articles two, seven, nine, twelve and seventeen). +- Notes: Without an identity card you cannot get a job, book accommodation, open an account or receive a subsidy; it is the precondition for every bit of paperwork. The Identity Card Law was amended in 2011, and the text quoted in this item is the one published in 2003, so go by the provisions now in force. Whether it can be done out of your area is set by the local public security notice. +### 12. After registering as unemployed, seek recognition as a person with difficulty finding work and get a social insurance subsidy or a public-interest post + +- Cost: No money. Apply for the recognition at the public employment service body at the sub-district or community level. The conditions for recognition differ from place to place, and the common ones are older age, long-term unemployment, a household on means-tested income support, disability, and a household with nobody in work. You have to complete unemployment registration before you can apply. The hard part is checking condition by condition whether you qualify, and having to pay the social insurance money up front. +- In plain language: Once you are recognised, if you pay your own social insurance you can get a subsidy, in principle no more than two thirds of what you actually paid, for at most 3 years. If you are taken on by an employer, or placed in a public-interest post, the employer's share of social insurance is subsidised as well. The wage for a public-interest post is set by reference to the local minimum wage. These posts do not pay much, but they come with social insurance, which makes them good for a transition. The social insurance subsidy is mostly paid after the fact, so you have to put the money up first. +- Benefit: Once you are recognised, people with no employer who pay their own social insurance (this is called flexible employment) can receive a social insurance subsidy. The subsidy standard is in principle no more than 2/3 of what you actually paid, for at most 3 years. People taken on by an employer, or placed in a public-interest post, get a subsidy on the pension, medical and unemployment insurance the employer actually pays. A public-interest post is one the government pays to create; its subsidy is set by reference to the local minimum wage. In government-created public-interest posts, people with difficulty finding work are placed first. There is also a one-off job-seeking subsidy for college graduates. It covers graduates from households on means-tested income support, households with nobody in work, and people in extreme difficulty. Graduates who are disabled, or who took a national student loan, also count. This money is claimed in the academic year in which you graduate (national framework, conditions and amounts set locally) +- Evidence grade: B +- Sources: Ministry of Finance and Ministry of Human Resources and Social Security (财政部、人力资源社会保障部) (2023, revised). Measures for the Administration of Employment Subsidy Funds（就业补助资金管理办法）(财社〔2017〕164, issued in revised form in 2023-12). ; Standing Committee of the National People's Congress (全国人大常委会) (2007, amended 2015). Employment Promotion Law（就业促进法）(Articles fifty-two and fifty-three). +- Notes: A public-interest post does not pay much, but it comes with social insurance, which makes it good for a transition. The social insurance subsidy is mostly paid after the fact, so you have to put the money up first. For the specific conditions, call 12333 or ask the sub-district office. +### 13. While unemployed, go and claim the vocational training subsidy, the work placement subsidy and the social insurance subsidy, instead of paying for a training course + +- Cost: No money. The training itself takes a good deal of your time. The hard part is that you have to finish the course and pass the certificate before you get the subsidy. +- In plain language: You do not have to pay for the training yourself. Attending training and obtaining the required certificate gives a vocational training subsidy, and the same person can receive it at most 3 times. College graduates who have not found work within two years of leaving, and unemployed young people aged 16 to 24, can take part in a work placement and receive a basic living allowance during it. Check the "vocational training subsidy catalogue" on the local human resources bureau website first, and do not pay to join a course of your own accord. +- Benefit: The rules of the Ministry of Finance and the Ministry of Human Resources and Social Security list a set of employment subsidy fund programmes, which is the money the government sets aside to subsidise employment. Attending training and obtaining a certificate that meets the requirements gives a vocational training subsidy, and the same person can receive it at most 3 times. The subsidy for people with difficulty finding work who pay their own social insurance (flexible employment) runs for at most 3 years; where fewer than 5 years remain to the statutory retirement age, it can be extended to retirement. The subsidy for college graduates who pay their own social insurance runs for at most 2 years, and for those taken on by an employer at most 1 year. College graduates who have not found work within 2 years of leaving, and unemployed young people aged 16 to 24, can take part in a work placement and receive a basic living allowance during it +- Evidence grade: A +- Sources: Ministry of Finance and Ministry of Human Resources and Social Security (2023). Measures for the Administration of Employment Subsidy Funds（就业补助资金管理办法）(Articles five, seven and ten). +- Notes: The specific subsidy catalogue and the amounts are set by each province, so check the "vocational training subsidy catalogue" on the local human resources bureau website first. College graduates in the year of graduation from households on means-tested income support or with nobody in work, and those with student loans outstanding or who are disabled, have a separate one-off job-seeking and entrepreneurship subsidy. +### 14. When job hunting, do not only send out resumes; use a system: learn the skills, set the goal, ask people for help + +- Cost: No money. It takes 1 to 2 hours a day, for several weeks in a row. You do not have to take a paid course; free vocational guidance from a public employment service body will do instead. The hard part is turning up every day when nobody is watching, and being the one who asks for help. +- In plain language: People who have had job search counselling were 2.67 times as likely to find work as people who had not. The parts that do the work are these: learning specific job search skills, explaining yourself clearly, setting a definite goal, going out and asking, and asking people for help. Asking people for help, in practice, means asking an acquaintance to refer you inside. +- Benefit: Pooling 47 controlled studies of job search counselling. These studies used experimental or quasi-experimental designs, all of which grouped participants for comparison. The result: the group that had counselling was 2.67 times as likely to find work as the group that did not (OR 2.67, that is the ratio of the odds). The components of counselling that do the work are these. First, teaching job search skills. Second, improving self-presentation (explaining yourself clearly). Third, raising self-efficacy (believing you can do it). Fourth, encouraging active outreach. Fifth, setting goals. Sixth, seeking social support (asking people for help). Counselling that both teaches skills and lifts motivation has the largest effect +- Evidence grade: A +- Sources: Liu S, Huang JL, Wang M (2014). Effectiveness of job search interventions: a meta-analytic review. Psychological Bulletin 140(4): 1009–1041. +- Notes: Most of these studies were done in Europe and the United States, in the form of packaged small-group courses. Moving them to China is applying results from elsewhere, not measurements taken here. Seeking social support, in practice, means getting an acquaintance to introduce you and refer you inside; do not be embarrassed to ask. +### 15. Do not pay deposits, do not hand over documents, do not sign a "training loan", do not join a pyramid scheme, do not borrow at usurious rates + +- Cost: No money. Refuse on the spot when you hear "pay first and then start work", "train first and then we find you a job", or "recruit heads". Keep the chat logs and the transfer receipts. The hard part is that the more desperate you are for money, the more you want to gamble, and saying "no" there and then is not easy. +- In plain language: An employer may not withhold your identity card, may not take a deposit, and may not take money under another name. If an agency does not find you a job, it must refund the agency fee. Training loans where you pay the training fee before starting work, and pyramid schemes that pay by recruiting heads, are both scams named by the authorities. Interest above 4 times the one-year LPR is not supported by the courts, but the principal still has to be repaid. +- Benefit: A company may not withhold your identity card or other documents, may not require you to find someone to guarantee you, and may not take your money or property under another name. A breach is ordered to be returned, with a fine of 500 yuan or more and up to 2000 yuan per person. Employment agencies likewise may not withhold documents or take deposits, and if they did not find you a job they must refund the agency fee. Inducing you to pay a training fee first, or to take out a loan first, under the name of recruitment is the "recruitment to training" and "training loan" scam that the Ministry of Human Resources and Social Security and four other departments have named. Taking commission by recruiting heads, and requiring payment before you can join, is a pyramid scheme, and those taking part can be fined up to 2000 yuan. The part of the interest on private lending that exceeds 4 times the one-year LPR at the time the contract was signed is not supported by the courts. This new rule replaces the old method of three bands set by the 24% and 36% lines (nationwide, from 2020-8) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law（劳动合同法）(主席令第六十五号, Articles nine and eighty-four). ; Ministry of Labour and Social Security (劳动保障部) (2007). Provisions on Employment Services and Employment Administration（就业服务与就业管理规定）(Ministry of Labour and Social Security Order No. 28, Articles fourteen, fifty-five and fifty-eight). ; Ministry of Human Resources and Social Security and four other departments (2026). Risk Warning on "Recruitment to Training" and "Training Loans" and Similar Risks（警惕「招转培」「培训贷」等风险提示）. ; State Council (国务院) (2005). Regulations on the Prohibition of Pyramid Schemes（禁止传销条例）(State Council Order No. 444, Articles two, seven and twenty-four). ; Supreme People's Court (最高人民法院) (2020). Decision on Amending the Provisions on Several Questions Concerning the Application of Law in Cases of Private Lending（关于修改《关于审理民间借贷案件适用法律若干问题的规定》的决定）(法释〔2020〕6, Article twenty-six). ; Supreme People's Court (2020). The Judicial Ceiling on the Interest Rate on Private Lending Is Adjusted to "4 Times the One-Year LPR"（民间借贷利率司法保护上限调整为「一年期 LPR 的 4 倍」）. +- Notes: 4 times the LPR is the ceiling on interest the courts are willing to protect, not a licence to borrow without repaying; the principal still has to be repaid. This judicial interpretation was changed once more in December 2020 and the article numbers shifted, so go by the text now in force. The more urgently you need money, the more likely these schemes are to come after you. Go through the proper channels earlier in this section first: unemployment insurance benefits, a complaint to labour inspection, legal aid, a rescue station, temporary assistance. +### 16. If housing is hard to find, put your name down for public rental housing first + +- Cost: Applying costs nothing. Submit documents to the housing security department of the city or county, and you also have to agree in writing to their verifying what you have reported. If the documents are incomplete, they will tell you in one go what else to supply. The waiting time for a queue is called the waiting period, and it is generally no more than 5 years. +- In plain language: You have to meet three conditions at once: you have no housing locally, or your housing is below the floor area standard; your income and property are below the standard; and if you are a migrant worker you must also have stable employment locally for the required number of years. The waiting period is published locally. This is slow water that cannot quench a present thirst, but joining the queue costs nothing, and the earlier you register the earlier your turn comes. +- Benefit: You can apply if you meet all three conditions. First, you have no housing locally, or your housing is below the prescribed floor area standard. Second, your income and property are below the prescribed standard. Third, migrant workers must have stable employment locally for the prescribed number of years. The specific conditions are set and published by the municipalities directly under the central government and by the housing security departments of cities and counties. A person registered on the waiting list is called a waiting applicant, and the housing security department must arrange a unit for them within the waiting period (national framework, the thresholds and the rent set locally, since 2012) +- Evidence grade: B +- Sources: Ministry of Housing and Urban-Rural Development (住房和城乡建设部) (2012). Measures for the Administration of Public Rental Housing（公共租赁住房管理办法）(Ministry of Housing and Urban-Rural Development Order No. 11, Articles seven, eight and ten). +- Notes: This is slow water that cannot quench a present thirst, but joining the queue costs nothing, and the earlier you register the earlier your turn comes. Places also have other kinds such as affordable rental housing, which work differently; look on the website of your local housing and urban-rural development department. +### 17. Push housing and food down to the minimum: a shared room monthly instead of a daily rental, and cooking for yourself or a subsidised canteen instead of takeaway + +- Cost: Finding a room takes 1 to 3 days. The deposit is usually 1 month's rent (the author's rough estimate). Cooking for yourself takes 30 to 60 minutes a day. The hard part is cooking every day and resisting the takeaway. +- In plain language: In the national per-capita spending of 2025, food came to 29.3% and housing to 21.7%; the two together are over half, so cutting them is the most effective. On the author's experience, daily lets and hourly rooms work out dearer over a month than a shared room, and cooking for yourself is cheaper than takeaway. When you are looking for work, ask whether food and housing are included; a post that includes them takes those two big items off the table. +- Benefit: In 2025, the national average per-capita consumption expenditure was 29476 yuan a year. Of that, food, tobacco and alcohol came to 8631 yuan, or 29.3%. Housing came to 6397 yuan, or 21.7%. The two together are over half, so cutting them is the most effective. Elderly meal assistance services are open to all older people and take the form of elderly canteens, elderly meal tables and meal points, with separate subsidies for elderly people in difficulty. For comparison: as of 2026-1-1, the lowest first-tier monthly minimum wage among the provinces was Qinghai at 2080 yuan and the highest was Shanghai at 2740 yuan (nationwide, 2025 to 2026) +- Evidence grade: B +- Sources: National Bureau of Statistics (国家统计局) (2026). Resident Income and Consumption Expenditure in 2025（2025 年居民收入和消费支出情况）. ; Ministry of Civil Affairs and ten other departments (民政部等十一部门) (2023). Action Plan to Develop Elderly Meal Assistance Services（积极发展老年助餐服务行动方案）(民发〔2023〕58). ; Sichuan Provincial Department of Human Resources and Social Security (四川省人力资源和社会保障厅) (2026). Minimum Wage Standards of All Provinces, Autonomous Regions and Municipalities（全国各省、自治区、直辖市最低工资标准情况）, as at 2026-1-1, reprinting the table on the human resources ministry website. +- Notes: Daily lets and hourly rooms work out far dearer than a shared room over a month, but that is a judgement from experience with no official figure behind it. Whether the elderly meal points are open to people who are not elderly, and what they charge, differs by place. When you are looking for work, ask whether food and housing are included; a post that includes them takes those two big items straight off the table. The original table on the human resources ministry website would not open, so the figures were taken from the reprint on the Sichuan provincial department's page. +### 18. Do not panic when social insurance contributions lapse: the pension counts cumulatively, and medical insurance follows the rules + +- Cost: No money. Resident old-age insurance and resident medical insurance can be paid by you year by year. Where medical insurance for employees has lapsed, you do not have to spend money finding someone to pay it for you just to avoid a gap. +- In plain language: The pension is counted by cumulative years, not consecutive years. A gap does not reset it, and a cumulative 15 years by the retirement age lets you draw it monthly. If you have paid medical insurance for employees for the required number of years, you do not pay again after retirement; the number of years differs by place, so call 12393 and ask where you are insured. What really is affected is eligibility for things that require continuous contributions, such as buying a home, settling household registration and points. +- Benefit: Basic old-age insurance is counted by cumulative years of contribution, and a gap in the middle does not reset it. At the statutory retirement age, a cumulative total of 15 years lets you draw the pension monthly. If it is short, you can keep paying until you reach 15 years. If you take work in a different place (officially, across pooling regions), the old-age insurance relationship follows you and the years of contribution keep adding up. Where the cumulative contribution to medical insurance for employees reaches the period set by the state, you do not pay again after retirement. The cost of a gap in resident medical insurance is a waiting period: for a time after you start paying again, medical costs are not reimbursed (see item 9 in this section for resident medical insurance) (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2010, amended 2018). Social Insurance Law（社会保险法）(Articles sixteen, nineteen and twenty-seven). ; General Office of the State Council (国务院办公厅) (2024). Guiding Opinions on Perfecting the Long-Term Mechanism for Enrolment in Basic Medical Insurance（关于健全基本医疗保险参保长效机制的指导意见）(国办发〔2024〕38). +- Notes: The number of years after which medical insurance for employees stops at retirement is written into the law as "the period set by the state", and the figure actually applied differs by place and is often different for men and women, so call 12393 or ask the healthcare security bureau where you are insured. A gap mainly affects local eligibility that requires a certain number of consecutive months, such as buying a home, settling household registration and points; those rules are set by each place, so look them up locally. Paying someone to "hang" your contributions for you can mean the money goes nowhere and carries legal risk, and is not recommended. +### 19. If you have been in prison, gone bankrupt or been put on the judgment defaulter list, the law leaves you a way to start again; finish the paperwork first + +- Cost: There is no handling fee. If you have been released after serving a sentence, report to the justice office or the sub-district office at your registered or actual place of residence and apply for resettlement and assistance from the government. If you are on the judgment defaulter list, pay off what you owe and then ask the enforcement court to remove your name. In Shenzhen an individual who has been insured for a full 3 years can apply for personal bankruptcy. Every step means preparing documents and attending an interview, and there is an observation period of 1 to 3 years. The hard part is telling people about the past again and again, and getting through the observation period. +- In plain language: A person released after serving a sentence may not be discriminated against in employment, education or social insurance, and the government helps with resettlement where life is difficult. Where the person was under 18 at the time of the offence and was sentenced to 5 years or less, the record is sealed and does not have to be reported when joining the army or looking for work. Being on the defaulter list lasts 2 years, and once the money is paid the court should remove the name within 3 working days. An adult's criminal record is not sealed and must be reported truthfully when joining the army or taking a job. +- Benefit: A person released after serving a sentence lawfully enjoys the same rights as other citizens and is not discriminated against in employment, education or social insurance. Those in difficulty are helped into a placement, with assistance and support in accordance with the law, by the government of their registered or actual place of residence (Prisons Law, revised 2026, in force from 11/1). A central document of 2010 requires strengthened vocational training for them, inclusion in means-tested income support where they qualify, and the establishment of transitional placement bases. Where a person was under 18 at the time of the offence and was sentenced to 5 years or less, the criminal record is sealed and does not have to be reported when joining the army or looking for work; where the person applies for a certificate of no criminal record, it is issued within 3 working days. Being placed on the judgment defaulter list lasts 2 years. Once the money is repaid, or in situations such as a settlement with the other party, the court should remove the name within 3 working days. The Shenzhen Special Economic Zone Personal Bankruptcy Regulations are currently the only personal bankruptcy rules in force; after bankruptcy is declared, the 3-year observation period is served, and debts that cannot be paid are written off. An enterprise legal person that cannot pay a debt when it falls due and owes more than it has (this is called insolvent) can apply for reorganisation, settlement or bankruptcy liquidation, and the debts are settled along with the enterprise (nationwide, personal bankruptcy only in Shenzhen) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2026). Prisons Law（监狱法）(revised 2026, Articles fifty-nine, sixty and sixty-two). ; General Offices of the CPC Central Committee and of the State Council (中共中央办公厅、国务院办公厅) (2010). Notice Forwarding the Opinions of the Central Committee on Comprehensive Governance of Public Security on Further Strengthening the Resettlement and Assistance of Persons Released After Serving a Sentence or Released from Labour Education（转发《中央社会治安综合治理委员会关于进一步加强刑满释放解除劳教人员安置帮教工作的意见》的通知）(中办发〔2010〕5). (reprinted by the Ziyang district government of Yiyang) ; Standing Committee of the National People's Congress (2018). Criminal Procedure Law（刑事诉讼法）(2018 amendment, Article two hundred and eighty-six). ; Supreme People's Court, Supreme People's Procuratorate, Ministry of Public Security and Ministry of Justice (2022). Implementation Measures for the Sealing of Criminal Records of Minors（关于未成年人犯罪记录封存的实施办法）(Articles nine and fifteen). (reprinted by the Hunan provincial public security department) ; Supreme People's Court (2017, amended). Several Provisions on the Publication of Information on the List of Judgment Defaulters（关于公布失信被执行人名单信息的若干规定）(Articles two and ten). ; Standing Committee of the Shenzhen Municipal People's Congress (深圳市人大常委会) (2020). Regulations on Personal Bankruptcy in the Shenzhen Special Economic Zone（深圳经济特区个人破产条例）(Articles two, twenty-three, ninety-five, ninety-seven and one hundred and three). ; Standing Committee of the National People's Congress (全国人大常委会) (2006). Enterprise Bankruptcy Law（企业破产法）(主席令第五十四号, Articles two and seven). +- Notes: An adult's criminal record is not sealed. Article one hundred of the Criminal Law requires a truthful report of any criminal penalty when joining the army or taking a job. Civil service, judges, lawyers and teachers have further statutory restrictions in their vetting. The text of the Prisons Law in force before 1 November 2026 corresponds to Articles thirty-seven and thirty-eight, so go by the article numbers in the text as it comes into force. The original page of Zhong Ban Fa [2010] 5 could not be found on the central government website, so it was taken from the district government reprint. During the observation period for personal bankruptcy you may not take first class on a plane, spend heavily or buy a house; compensation for deliberately harming someone's person or one of the 8 other kinds of debt is not written off; and an exemption obtained by fraud may still be revoked. Being removed from the judgment defaulter list does not mean your credit record becomes clean as well; those are two separate systems. +### 20. Before serious illness strikes, add one-year medical insurance or critical illness insurance on top of basic medical insurance, and check the four words "guaranteed renewal" + +- Cost: For a young person a one-year expense-reimbursement medical policy costs several hundred yuan a year. An expense-reimbursement policy is one that reimburses what you actually spent. Critical illness insurance starts at several thousand yuan a year (the author's rough estimate, no official figure). You have to disclose your health condition truthfully before taking out the policy, which takes half an hour. +- In plain language: This kind of insurance covers the part you have to pay yourself after medical insurance has reimbursed, and the loss of earnings while you cannot work. Before buying, look for the four words "guaranteed renewal": without them, the product may be withdrawn or the price raised next year. You have to disclose your health condition truthfully, and if you conceal it the claim can be refused. Put basic medical insurance in place first, then think about this. +- Benefit: The authorities divide this kind of product into two kinds. Medical insurance reimburses the medical costs you actually incurred. Disease insurance pays a lump sum in the sum stated in the contract as soon as one of the conditions listed there is diagnosed. Both cover the part you have to pay yourself after medical insurance has reimbursed, and the loss of earnings while you cannot work. Insurance with a period of one year or less, where the terms do not also state guaranteed renewal, is short-term health insurance. Only insurance of more than one year, or insurance that states guaranteed renewal, is long-term health insurance. "Guaranteed renewal" means that on renewal at the end of the period the company must continue the cover on the same terms and at the agreed rate. An insurance company may not induce the same person to buy again a expense-reimbursement medical policy with the same or a similar protection. If you buy two, the same bill can only be reimbursed once. The disease definitions for critical illness insurance follow an industry-wide standard, which includes 28 severe conditions and 3 mild conditions. Malignant tumours, acute heart attack and the after-effects of a stroke are the core conditions (nationwide, from 2019 and 2020) +- Evidence grade: B +- Sources: formerly the China Banking and Insurance Regulatory Commission (银保监会) (2019). Measures for the Administration of Health Insurance（健康保险管理办法）(CBIRC Order No. 3 of 2019, Articles two, four and forty-one). ; China Insurance Industry Association and Chinese Medical Doctor Association (2020). Answers to Reporters' Questions on the Use Standard for the Disease Definitions of Critical Illness Insurance (2020 Revision)（重大疾病保险的疾病定义使用规范（2020 年修订版）答记者问）. ; formerly the China Banking and Insurance Regulatory Commission (2020). Notice on Matters Concerning the Use of the Empirical Incidence Rate Table for Critical Illness in Chinese Personal Insurance (2020)（关于使用《中国人身保险业重大疾病经验发生率表（2020）》有关事项的通知）(银保监发〔2020〕51). ; Standing Committee of the National People's Congress (全国人大常委会) (2015, amended). Insurance Law of the People's Republic of China（中华人民共和国保险法）, Article sixteen (truthful disclosure, and no termination after two years from the formation of the contract). +- Notes: It is graded B because the authorities have only given definitions and rules, with no figure for "how much you save by buying this". "Million-yuan medical insurance" is the name used when it is sold; as far as the rules are concerned it counts as short-term health insurance. If the terms do not say "guaranteed renewal", the product may be withdrawn or the price raised next year, so look for those four words in the terms before you buy. Two expense-reimbursement medical policies cannot reimburse the same bill twice, so do not stack them. Taking out a policy while already ill and then failing to disclose it means the claim can be refused. Truthful disclosure has a two-year limit: once the contract has existed for two years the insurer can no longer terminate it on the ground that you did not disclose, and it must pay a valid claim. Where deliberate non-disclosure is found within the two years, events occurring before the contract is terminated are not covered and the premium is not refunded. For which losses are worth insuring, see item 40 in section 5 (losses you could not carry). Put resident medical insurance in place first (item 9 in this section), then think about this. The disease definition standard is published by the industry association under the guidance of the banking and insurance regulator; it is not a government document. +### 21. If you really have nowhere to sleep, a 24-hour premises is only good for one emergency night; go to a rescue station first + +- Cost: The waiting areas at airports and railway stations usually cost nothing. Fast food restaurants and internet cafés require you to spend a little. You will not sleep properly and there is a risk of having your belongings stolen. The hard part is not daring to really sleep all night. +- In plain language: The waiting areas at airports and railway stations will see you through the night, and they have toilets and power sockets. But no official data says it is safe. Keep your belongings on your body, do not take food or drink offered by strangers, and do not go to secluded corners. Do not sleep under bridges, by rivers or on building sites at night. A rescue station provides food and a bed free; go there first. +- Benefit: It sees you through the night, there are toilets, and you can charge your phone. No official figure +- Evidence grade: C +- Sources: the author's own experience, no official document +- Notes: The safety risk is real. Keep your belongings on your body, do not take drink or food offered by strangers, and avoid secluded corners. Do not sleep rough under bridges, by rivers or on building sites at night. Sleeping rough in extreme weather can kill. The rescue station in item 4 provides food and a bed free, so go there first. Some premises clear everyone out in the middle of the night; do not pick a fight with the security staff. \ No newline at end of file diff --git a/en/index.html b/en/index.html index 7b8e3d6c..b6da97c4 100644 --- a/en/index.html +++ b/en/index.html @@ -35,7 +35,7 @@ - + @@ -345,7 +345,12 @@

    A High-Return Life Guide

    The filters on this page need JavaScript. You can read the text directly: the table of contents inREADME.md, and the text itself in book/.

    1. Do not die young — deaths from external causes, gas and poisoning, vaccines, screening, the timescale of a mental crisis or suicidal thoughts, what is left after someone is brought back, the year after a serious fall, the arithmetic of the kidney you keep after selling one, home emergency kits, and signals such as visible blood in the urine that should be checked. Measure: all-cause or cause-specific mortality.
      2. -
    2. Do not die slowly — tobacco and alcohol, exercise, sleep, diet (low-sodium salt, nuts, whole grains, processed meat, bulk cold-pressed peanut oil, vegetable oil instead of lard), sitting too long, and the practical ways to quit smoking and cut down drinking (quit medication, quit dates, smoking cessation clinics and helplines, e-cigarettes, alcohol withdrawal not something to push through alone), how long a nap should be, how to make up for a late night, and the arithmetic of years spent on night shifts. Measure: all-cause or cause-specific mortality.
    +
  • Do not die slowly — tobacco and alcohol, exercise, sleep, diet (low-sodium salt, nuts, whole grains, processed meat, bulk cold-pressed peanut oil, vegetable oil instead of lard), sitting too long, and the practical ways to quit smoking and cut down drinking (quit medication, quit dates, smoking cessation clinics and helplines, e-cigarettes, alcohol withdrawal not something to push through alone), how long a nap should be, how to make up for a late night, and the arithmetic of years spent on night shifts. Measure: all-cause or cause-specific mortality.
    • +
  • Do not waste energy — sleep, interruptions, multitasking, decision fatigue, social debt, and what to expect when you deal with institutions. Measure: energy / time.
    • +
  • Do not waste time — projects that pay nothing back, sunk costs, procrastination (the emotional explanation, changing the environment, commitment devices, how long a habit takes, self-help material), meetings, commuting. Measure: time.
    • +
  • Do not waste money — subscriptions, lottery tickets, interest, insurance, fund fees, personal pensions, car insurance, prepayments, live-stream shopping, personal medical insurance accounts, children being scammed and refunds for top-ups, prayer beads, watches and designer toys not counted as investments, how to check a jewellery or jade appraisal report, blind boxes and gacha, and which legal channel to use to buy assets overseas. Six items at the end of the section are about insurance: only insure a loss you cannot absorb, earners buy a term life policy first, use the cooling-off period to get your money back, do not let a bank counter sell you insurance as a deposit, do not use an agent to surrender a policy, and beneficiaries and the time limit for claims. Measure: money.
    • +
  • The reverse list — things that look like good value but are not, including the claim that willpower runs out.
    • +
  • How to get by with no money — assistance, subsidies, finding work, somewhere to stay and food, medical care, recovering unpaid wages, and the traps to avoid. Measure: money / security.
    • @@ -756,10 +761,19 @@

    A High-Return Life Guide

    const XREF_RE = RE_XREF; // 指路里的分隔符按正则切（英文是 ", and to –" 一串，中文是「、」） const XREF_SEP = new RegExp(SITE.xref.listSep); -// 「8 到 10」摊成 8、9、10；上限 40 是防着把法条条款号那种大范围摊成几百个键 +// 「8 到 10」摊成 8、9、10；上限 40 是防着把法条条款号那种大范围摊成几百个键。 +// 先认整串区间，再按分隔符切：英文的 listSep 里含 "to" 和 "-"，先切的话 +// "8 to 11" 被切成 "8" 和 "11"，区间再也认不出来（译文检查那边同样的坑，见 structure.mjs） const xrefKeys = (sec, spec) => { const ns = []; - for (const part of String(spec).split(XREF_SEP)){ + const whole = String(spec).trim(); + const wr = RE_XRANGE.exec(whole); + if (wr){ + const a = +wr[1], b = +wr[2]; + if (b >= a && b - a <= 40) for (let i = a; i <= b; i++) ns.push(i); + return ns.map(n => `${sec}-${n}`).filter(k => ITEMS.has(k)); + } + for (const part of whole.split(XREF_SEP)){ const r = RE_XRANGE.exec(part); if (r){ const a = +r[1], b = +r[2]; @@ -1040,9 +1054,55 @@

    A High-Return Life Guide

    } document.getElementById('tot').textContent = CARDS.length; } +/* ---------- 预渲染 hydration ---------- + 服务端（web/src/lib/render.ts）按 build() 初次渲染的形状把 DOM 写好， + 这里按文档顺序索引到 CARDS/BLOCKS/TOC 三张表里。索引完的状态和 build() 刚跑完 + 一样：card.key 都是 null，vis 都是 false，滚进来由 CARD_IO 补渲染（terms=[]， + 和服务端写的一致，重渲出来是同一个 DOM）。 + 对不上就回退：预渲染块和侧栏目录删掉，走 build() 重建。回退时 console 里留一句， + 免得渲染器悄悄跑偏了没人知道——正常构建永远不该走到这里。 */ +function indexSsgDom(sections){ + for (const s of sections){ SECS.set(s.n, s); for (const e of s.entries) ITEMS.set(`${s.n}-${e.n}`, e); } + const total = sections.reduce((n, s) => n + s.entries.length, 0); + const list = document.getElementById('list'); + const blocks = [...list.querySelectorAll(':scope > section.sec-block')]; + const preCards = list.querySelectorAll('article.card').length; + if (blocks.length !== sections.length || preCards !== total){ + console.warn(`预渲染对不上（块 ${blocks.length}/${sections.length}，卡片 ${preCards}/${total}），回退到 build() 重建`); + list.querySelectorAll(':scope > section.sec-block').forEach(b => b.remove()); + const fsec = document.getElementById('f-sec'); fsec.innerHTML = ''; + document.querySelector('#gloss dl').innerHTML = ''; + build(sections); buildGlossary(); + return; + } + const fsec = document.getElementById('f-sec'); + const links = [...fsec.querySelectorAll('a[data-go]')]; + const subs = [...fsec.querySelectorAll('.toc-sub')]; + let li = 0, si = 0; + for (const s of sections){ + for (const e of s.entries){ const a = links[li++]; if (a) TOC.set(`e-${e.sec}-${e.n}`, a); } + const sub = subs[si++]; if (sub) TOC_SEC.set(s.n, sub); + } + let bi = 0; + for (const s of sections){ + const block = blocks[bi++]; if (!block) break; + BLOCKS.push({n:s.n, el:block, k:block.querySelector('.shown .k')}); + const cards = [...block.querySelectorAll('article.card')]; + let ci = 0; + for (const e of s.entries){ + const c = cards[ci++]; if (!c) break; + const card = {e, el:c, key:null, vis:false, f:{t:c.querySelector('.t'), cost:c.querySelector('.cost'), human:c.querySelector('.human'), gain:c.querySelector('.gain'), note:c.querySelector('.note'), src:c.querySelector('.src .body'), srcN:c.querySelector('.src summary .cnt'), details:c.querySelector('.src')}}; + CARDS.push(card); CARD_OF.set(c, card); CARD_IO.observe(c); + } + } + document.getElementById('tot').textContent = CARDS.length; +} function buildGlossary(){ const box = document.getElementById('gloss'); if (!GLOSS.length) { box.hidden = true; return; } const dl = box.querySelector('dl'); + // 预渲染（Astro 构建）已经把 dt/dd 写好了：这里只索引，不重建。 + // 没有 data-ssg 的页面 dl 是空的，走原来的路，一行没变。 + if (dl.children.length) return; for (const g of [...GLOSS].sort((a,b) => a.term.localeCompare(b.term, SITE.sortLocale))){ const dt = document.createElement('dt'); dt.textContent = g.term; const dd = document.createElement('dd'); dd.textContent = g.meaning; @@ -1465,7 +1525,15 @@

    A High-Return Life Guide

    GLOSS = parseGlossary(md); PERF.log(T('perf.parse'), t1, files.length + ' ' + T('perf.files')); status.remove(); - const t2 = performance.now(); build(sections); PERF.log(T('perf.buildDom'), t2, CARDS.length + ' ' + T('perf.cards')); + const t2 = performance.now(); + // 预渲染 hydration：Astro 构建时已经把 build() 会创建的 DOM 原样写进 HTML + // （卡片、侧栏目录、术语表，见 web/src/lib/render.ts），这里只索引现有节点， + // 不重新创建。判据是 #list 上的 data-ssg 和同页的语料，两个缺一个都走老路—— + // 老构建和离线单文件没有 data-ssg，它们的行为一行没变。 + const listEl = document.getElementById('list'); + if (EMBED && listEl && listEl.dataset.ssg === '1') indexSsgDom(sections); + else build(sections); + PERF.log(T('perf.buildDom'), t2, CARDS.length + ' ' + T('perf.cards')); buildGlossary(); readUrl(); wire(); apply(); PERF.mark(T('perf.mark', {dom: document.getElementsByTagName('*').length, terms: GLOSS.length})); if (location.hash){ diff --git a/tools/i18n/join.mjs b/tools/i18n/join.mjs index eeb185ed..6c435ada 100644 --- a/tools/i18n/join.mjs +++ b/tools/i18n/join.mjs @@ -53,7 +53,7 @@ for (const srcPath of files){ let trouble = 0; const pieces = []; for (let k = 0; k < parts.length; k++){ - const want = src.blocks[k].join('\n'); + const want = src.blocks[k].flat().join('\n'); // blocks[k] 是「这一块有哪几条」 const got = read(`${dir}/${parts[k]}`); const a = entrySkeleton(want, strings(SOURCE.code)); const b = entrySkeleton(got, strings(locale)); @@ -68,7 +68,28 @@ for (const srcPath of files){ bad++; trouble++; console.error(` ${dir}/${parts[k]}：${b.entries.length} 条，原文这块 ${a.entries.length} 条`); } - pieces.push(k === 0 ? src.head.join('\n') + '\n' + got : got); + // 译文块里如果又带了一遍节首（回目录那行 + 「# N. 节名」+ 导读），拼出来就是两份， + // 页面上那一节连同它的条目渲染两遍——en 和 vi 五个文件都中过（2026-09-30 迁移到 + // Astro 时预渲染和运行时解析出来的节数对不上，查出来的）。 + // + // 判据按内容，不按行数：节首的每一行都逐行对上（导读在译文里可能已经翻过， + // 对不上就整块留着，宁可多渲染一遍导读也不能把译文正文吃掉）。 + let body = got; + if (k === 0 && src.head.length){ + const headLines = src.head.join('\n').split('\n').map(l => l.trim()).filter(Boolean); + const bodyLines = got.split('\n'); + let consumed = 0; + for (const hl of headLines){ + const at2 = bodyLines.findIndex((l, i) => i >= consumed && l.trim() === hl); + if (at2 < 0) { consumed = -1; break; } // 对不上：整块保留 + consumed = at2 + 1; + } + if (consumed > 0 && /^#{1,2} \d+\. /.test(headLines.find(l => /^#{1,2} \d+\. /.test(l)) || '')){ + body = bodyLines.slice(consumed).join('\n').replace(/^\n+/, ''); + console.log(` ${name} 块 0：译文里带了节首（${headLines.length} 行），已去掉，之前会重复渲染一次`); + } + } + pieces.push(k === 0 ? src.head.join('\n') + '\n' + body : body); } if (trouble) { console.error(`${locale} ${name}：${trouble} 处对不上，没写进 ${L.dir}book/${name}`); continue; } diff --git a/tools/i18n/prompt-batch.mjs b/tools/i18n/prompt-batch.mjs new file mode 100644 index 00000000..253ab146 --- /dev/null +++ b/tools/i18n/prompt-batch.mjs @@ -0,0 +1,51 @@ +// 生成一批翻译任务的提示词。分批是为了并行：一批十来个块，一次翻得完。 +// 用法：node tools/i18n/prompt-batch.mjs en 03-不要浪费精力,04-不要浪费时间,10-恋爱和结婚划不划算 +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; +import { readdirSync, readFileSync } from 'node:fs'; + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const locale = process.argv[2]; +const secs = (process.argv[3] ?? '').split(',').filter(Boolean); + +const LANG = { en: 'English', vi: 'Vietnamese' }[locale]; +const CONV = locale === 'vi' + ? `- 条目 = "mục", 节 = "phần", statute sub-clause = "khoản (2)" or "điểm (a)". Never confuse them. +- 「第 X 节」 → "phần X" (writing "mục X" for a section makes the reader think it is item X) +- 「第 X 条」 → "mục X" ; 「本节第 X 条」 → "mục X của phần này" ; 「第 A 节第 B 条」 → "mục B ở phần A" +- Statute article 「第 285 条」 → "Điều 285". NEVER write "phần N" for a statute clause.` + : `- Cross-references: 「第 X 节」 → "section X", 「第 X 条」 → "item X", 「本节第 X 条」 → "item X in this section", 「第 A 节第 B 条」 → "item B in section A".`; + +const files = []; +for (const sec of secs){ + const dir = resolve(ROOT, '.i18n', locale, sec); + for (const f of readdirSync(dir).filter(x => /^part\d+\.md$/.test(x)).sort()) files.push(`${sec}/${f}`); +} + +const todo = files.filter(f => readFileSync(resolve(ROOT, '.i18n', locale, f), 'utf8').includes('成本：')); + +console.log(`Translate these ${todo.length} blocks of a Chinese life-guide book into ${LANG}. + +READ FIRST, both files, completely: + ${ROOT}/tools/i18n/BRIEF.md — every hard requirement (structure, the six field names, verbatim Chinese cost tags, frozen numbers, cross-references, byte-identical URLs, register) + ${ROOT}/tools/i18n/glossary.${locale}.mjs — use these exact terms + +Write each translation back to the SAME path, overwriting the Chinese. Output nothing else. + +${CONV} + +FILES (${todo.length}): +${todo.map(f => ' ' + ROOT + '/.i18n/' + locale + '/' + f).join('\n')} + +WHEN DONE, verify each section you finished: + cd ${ROOT} + node tools/i18n/join.mjs book/
    .md --locale ${locale} +It prints one line per mismatch; fix what it reports and re-run until the section is clean +("合好 N 节，问题 0 处"). A section whose blocks you did NOT touch will still report missing +blocks — that is expected, ignore it and report it as such. + +Do NOT run tools/site/build.mjs (another task is using it). Do NOT touch any file outside +.i18n/${locale}/ and ${locale}/book/. + +Final response: one short line — which sections came out clean, and which (if any) still have +problems, quoting the checker's message.`); diff --git a/tools/i18n/resplit.mjs b/tools/i18n/resplit.mjs new file mode 100644 index 00000000..5a8e392d --- /dev/null +++ b/tools/i18n/resplit.mjs @@ -0,0 +1,93 @@ +// 按新的块大小重切，但已经翻好的条目原样带过去。 +// +// node tools/i18n/resplit.mjs # 两种语言所有节 +// node tools/i18n/resplit.mjs en # 只英文 +// node tools/i18n/resplit.mjs en 01-不要早死,02-不要慢慢死 +// +// 为什么要有这个：块大小是会调的（太碎的块费 token，太大的块一次翻不完）， +// 改参数重跑 split.mjs 会把翻好的部分盖掉——那是白翻。 +// splitSection 的 blocks[i] 是「第 i 块包含哪几条」，每条一整段，所以重排就是 +// 按条号查表替换：翻过的搬译文，没翻的搬原文。行级 splice 在这里不行， +// 一条的长度和另一条不一样就会错位（踩过一次：一节报三十多处，看着像译文烂了， +// 其实是搬错了）。 +// +// 已翻的条目从两处收：.i18n 里的块（还没合的）和已经合进去的整节 +// （<语言>/book/<节>.md）——中文改过之后新加的条目，译者可能已经翻好合过一次， +// 只看 .i18n 会当成没翻。 +// +// 跑完 join.mjs 一校就知道有没有搬坏。 + +import { readFileSync, writeFileSync, readdirSync, existsSync, rmSync, mkdirSync } from 'node:fs'; +import { resolve } from 'node:path'; +import { splitSection } from './split.mjs'; +import { strings, SOURCE } from '../site/locales.mjs'; + +const ROOT = resolve(import.meta.dirname, '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); +const write = (p, t) => { + mkdirSync(resolve(ROOT, p).replace(/\/[^/]+$/, ''), { recursive: true }); + writeFileSync(resolve(ROOT, p), t); +}; + +const argv = process.argv.slice(2); +const locales = argv[0] ? [argv[0]] : ['en', 'vi']; +const only = argv[1] ? argv[1].split(',').filter(Boolean) : null; + +const escRe = s => s.replace(/[.*+?^${}()|[\]\\]/g, '\\$&'); + +/** 一份正文拆成「头部 + 按条号索引的条目」 */ +function pieces(text, S) { + const lines = text.split('\n'); + const at = lines.map((l, i) => (/^### /.test(l) ? i : -1)).filter(i => i >= 0); + const head = at.length ? lines.slice(0, at[0]) : lines; + const fieldRe = new RegExp('^- ' + escRe(S.fields.cost)); + const byN = new Map(); + for (let k = 0; k < at.length; k++){ + const stop = k + 1 < at.length ? at[k + 1] : lines.length; + const seg = lines.slice(at[k], stop); + while (seg.length && !seg[seg.length - 1].trim()) seg.pop(); // 段尾空行归下一段 + byN.set(/^### (\d+)\./.exec(seg[0])[1], { lines: seg, translated: seg.some(l => fieldRe.test(l)) }); + } + return { head, byN }; +} + +let moved = 0, kept = 0; +for (const locale of locales){ + const S = strings(locale); + const dir = `.i18n/${locale}`; + if (!existsSync(resolve(ROOT, dir))) continue; + + for (const sec of readdirSync(resolve(ROOT, dir))){ + if (only && !only.includes(sec)) continue; + const srcFile = `${SOURCE.contentDir ? SOURCE.contentDir + '/' : ''}book/${sec}.md`; + if (!existsSync(resolve(ROOT, srcFile))) continue; + const sd = `${dir}/${sec}`; + + // 已翻的按条号收起来：先整节（合过一次的），再 .i18n 里的块（更晚翻的以它为准） + const done = new Map(); + const collect = text => { + for (const [n, e] of pieces(text, S).byN) if (e.translated){ done.set(n, e.lines); moved++; } + }; + if (existsSync(resolve(ROOT, `${locale}/book/${sec}.md`))) collect(read(`${locale}/book/${sec}.md`)); + for (const f of readdirSync(resolve(ROOT, sd)).filter(x => /^part\d+\.md$/.test(x)).sort()) + collect(read(`${sd}/${f}`)); + if (!done.size) continue; + + const cut = splitSection(read(srcFile)); + rmSync(resolve(ROOT, sd), { recursive: true, force: true }); + let n = 0; + cut.blocks.forEach((blockEntries, i) => { + const body = blockEntries.map(entryLines => { + const num = /^### (\d+)\./.exec(entryLines[0])[1]; + if (!done.has(num)) return entryLines; + n++; kept++; + return done.get(num); + }); + write(`${sd}/part${String(i).padStart(2, '0')}.md`, + (i === 0 ? cut.head.join('\n') + '\n' : '') + body.flat().join('\n') + '\n'); + }); + if (cut.tailLink) write(`${sd}/tail.txt`, cut.tailLink + '\n'); + console.log(`${locale} ${sec}：重切 ${cut.blocks.length} 块，带过 ${n} 条已翻的`); + } +} +console.log(`\n共 ${moved} 条已翻，重排后 ${kept} 条落地`); diff --git a/tools/i18n/split.mjs b/tools/i18n/split.mjs index 4aba3bb2..c2d0e01c 100644 --- a/tools/i18n/split.mjs +++ b/tools/i18n/split.mjs @@ -28,7 +28,11 @@ if (!LOCALES.some(l => l.code === locale)) { console.error(`--locale 得是 ${LOCALES.map(l => l.code).join(' / ')} 里的一项`); process.exit(2); } -const ENTRIES = 10; // 一块几条。十条差不多 1 万字，输出在一个任务里做得完 +// 一块的上限：条数和字数都算，哪个先到就切。 +// 条数上限防「十几条都很短」时一块太碎；字数上限防「一条特别长」时一块太大—— +// 31 节第一块光一个条目就 4 万多字，一次翻不完的。 +const ENTRIES = Number(process.env.SPLIT_ENTRIES ?? 8); +const MAX_BYTES = Number(process.env.SPLIT_MAX_BYTES ?? 6000); /** 把一节正文切成头部 + 若干块条目。切点只在条目边界上，不能从一条中间断开 */ export function splitSection(md, perBlock = ENTRIES) { @@ -49,10 +53,18 @@ export function splitSection(md, perBlock = ENTRIES) { return lines.slice(start, stop); }); + // blocks[i] 是「第 i 块包含哪几条」，每条一整段（行数组）。 + // 保持条为单位而不是拍平成行：重切时（tools/i18n/resplit.mjs）要按条整段搬， + // 拍平之后只能按行号 splice，一条的长度和另一条不一样就错位了（踩过一次： + // 一节报三十多处，看着像译文烂了，其实是搬错了）。 const blocks = []; - for (let i = 0; i < entries.length; i += perBlock){ - blocks.push(entries.slice(i, i + perBlock).flat()); + let cur = [], bytes = 0; + for (const e of entries){ + const size = e.join('\n').length; + if (cur.length && (bytes + size > MAX_BYTES || cur.length >= perBlock)){ blocks.push(cur); cur = []; bytes = 0; } + cur.push(e); bytes += size; } + if (cur.length) blocks.push(cur); return { head, blocks, tailLink, count: entries.length }; } @@ -63,7 +75,7 @@ export function outDir(locale, file) { // ---------------------------------------------------------------- 主流程 // join.mjs 要 import 这里的 splitSection，所以主流程只在直接跑这个脚本时执行。 // 不加这一层，join 一 import 就把切好的块重切一遍，刚翻好的一块就被原文盖回去。 -if (import.meta.url === pathToFileURL(process.argv[1]).href) await main(); +if (process.argv[1] && import.meta.url === pathToFileURL(process.argv[1]).href) await main(); async function main(){ const files = []; @@ -94,7 +106,7 @@ for (const f of files){ // 顺手把翻好的几块冲掉就白翻一天（2026-09-30 踩过）。 if (existsSync(resolve(ROOT, at)) && !process.argv.includes('--force')){ kept++; return; } // 第 0 块带头部（回目录那行 + 节标题 + 导读），其余块只带条目，合起来才是一节 - const text = (i === 0 ? cut.head.join('\n') + '\n' : '') + b.join('\n') + '\n'; + const text = (i === 0 ? cut.head.join('\n') + '\n' : '') + b.flat().join('\n') + '\n'; writeFileSync(resolve(ROOT, at), text); wrote++; }); diff --git a/tools/i18n/structure.mjs b/tools/i18n/structure.mjs index d98e9b74..693f3386 100644 --- a/tools/i18n/structure.mjs +++ b/tools/i18n/structure.mjs @@ -61,12 +61,28 @@ function numbersOf(text) { * 正则是各语言自己那份（tools/site/locales/.json 的 xref.pattern），和检索页 * 把「第 X 条」变成可点的引用用的是同一套。所以这里查的就是读者真能点到的那些指路， * 译文里条号节号丢了或者写不成可点的形状，这一层就报出来。 + * + * 顺序：先认区间，再按分隔符切。英文的 listSep 里就含 "to" 和 "-" + * （"items 8 to 11" 要切得开），先切的话 "8 to 11" 被切成 "8" 和 "11"， + * 区间就再也认不出来——一次翻 22 块之后才发现，第 5 节两处指路报「不见了」。 */ function refsOf(text, xref) { if (!xref) return []; const range = new RegExp(xref.rangeRe); const sep = new RegExp(xref.listSep); // 正则，不是字面串：英文是 ", and to –" 一串 const out = []; + // 区间摊成一个个号，不留「8-11」这种形状：原文写「第 8 到 11 条」、 + // 英文写 "items 8, 9, 10 and 11" 指的是同一批条目，摊开才比得上。 + // 上限 40 和检索页一致，防着把法条条款号那种大范围摊成几百个号。 + const push = (sec, spec) => { + const r = range.exec(spec); + if (r){ + const a = +r[1], b = +r[2]; + if (b >= a && b - a <= RANGE_MAX) for (let i = a; i <= b; i++) out.push(`${sec}${i}`); + return; + } + out.push(`${sec}${spec}`); + }; for (const m of text.matchAll(new RegExp(xref.pattern, 'g'))){ const g = m.groups ?? {}; if (g.secs !== undefined){ @@ -75,15 +91,19 @@ function refsOf(text, xref) { } // 节号缺省就是本节：中文「本节第 3 条」、英文「this section's item 3」 const sec = g.sec ? 's' + g.sec : '*'; - for (const part of String(g.nums ?? g.nums2 ?? g.nums3 ?? '').split(sep)){ + // 先认整串区间再按分隔符切：英文的 listSep 里含 "to" 和 "-"， + // 先切的话 "8 to 11" 变成 "8" 和 "11"，区间再也认不出来 + const spec = String(g.nums ?? g.nums2 ?? g.nums3 ?? '').trim(); + if (range.test(spec)){ push(sec, spec); continue; } + for (const part of spec.split(sep)){ const t = part.trim(); - if (!t) continue; - const r = range.exec(t); - out.push(r ? `${sec}r${r[1]}-${r[2]}` : `${sec}${t}`); + if (t) push(sec, t); } } return out.sort(); } +// 区间摊开的上限，和检索页 xrefKeys 一样 +const RANGE_MAX = 40; /** * 解析一节正文。head 是节标题和导读，entries 是条目。 diff --git a/tools/site/build.mjs b/tools/site/build.mjs index 02d318d6..6035ebd8 100644 --- a/tools/site/build.mjs +++ b/tools/site/build.mjs @@ -20,7 +20,10 @@ import { resolve, dirname } from 'node:path'; import { fileURLToPath } from 'node:url'; const HERE = dirname(fileURLToPath(import.meta.url)); -const ROOT = resolve(HERE, '../..'); +// 仓库根。直接跑 node 时按文件位置算；被 Astro 构建打包时 import.meta.url 会指到 +// 构建缓存里，读到的全是空——这时按环境变量 HOWTOLIVEBETTER_ROOT 来， +// web/astro.config.mjs 会在构建前把它指到仓库根（直接跑 node 时不用设）。 +const ROOT = process.env.HOWTOLIVEBETTER_ROOT ? resolve(process.env.HOWTOLIVEBETTER_ROOT) : resolve(HERE, '../..'); const CHECK = process.argv.includes('--check'); const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); @@ -203,7 +206,7 @@ function docLinks(docs, L, strings, S) { // 也可以自己点。关掉 JS 也点得动。 // 只列 PUBLISHED：还没翻完的那种（眼下是越南文）hreflang 和链接里都不出现—— // 列出来点进去是 404，比不列更糟。翻完重跑这个脚本，它自己回来。 -function chooserPage(site) { +export function chooserPage(site) { const list = PUBLISHED .map(l => `
  • ${escText(l.name)}${escText(l.note)}
    • `) .join('\n'); @@ -258,7 +261,7 @@ ${pick} // 不能出现在 hreflang、跳语言页和页内语言下拉里——那些都是读者会点开的链接， // 点进去只有一个 404，比不列更糟。翻完重跑这个脚本，它自己回来。 const readmeOf = L => (L.contentDir ? L.contentDir + '/' : '') + 'README.md'; -const PUBLISHED = LOCALES.filter(L => existsSync(resolve(ROOT, readmeOf(L)))); +export const PUBLISHED = LOCALES.filter(L => existsSync(resolve(ROOT, readmeOf(L)))); // hreflang 由 PUBLISHED 生成，不在模板里写死：写死了就得记得在加语言、改域名时两处一起改， // 漏一处就是一堆指向不存在页面的 hreflang。 @@ -392,7 +395,7 @@ function jsonLd(L, strings, sections, stats) { // // 每一条都带 xhtml:link 互指：三种语言是同一本书的三种语言，搜索引擎靠这个知道， // 不然它会把 en/ 和 vi/ 当成两本不同的书。x-default 指跳语言的小页。 -function sitemapXml() { +export function sitemapXml() { const alternates = PUBLISHED.map(l => ` `); alternates.push(` `); @@ -408,7 +411,7 @@ function sitemapXml() { // robots.txt 和 sitemap.xml 是一对：sitemap 指的那个地址得是真的。一起从 site.json 拼， // 换域名就不会只改了一半。 -function robotsTxt() { +export function robotsTxt() { return `User-agent: * Allow: / @@ -420,8 +423,8 @@ Sitemap: ${SITE_URL()}sitemap.xml // 换域名只改那一处：locales.json 里每种语言的 canonical 和 repoBlob 由它们推出来， // 对不上就在 checkUrls() 里报错，不会出现「Pages 发到 A 域名、页面自称是 B 域名」。 // site 不带尾斜杠：模板里是 {{site}}/en/、{{site}}/og.png，带了会拼出双斜杠。 -const { site: SITE, repo: REPO } = JSON.parse(read('tools/site/site.json')); -const SITE_URL = () => SITE + '/'; +export const { site: SITE, repo: REPO } = JSON.parse(read('tools/site/site.json')); +export const SITE_URL = () => SITE + '/'; // locales.json 的 canonical / repoBlob 是从 site.json 抄的。抄错了页面照样生成， // 但链接会指向别处，而且没人会一眼看出来。这里逐个对一遍。 @@ -440,8 +443,18 @@ function checkUrls() { // SKIP：这一份语言的 README 和正文都还没翻，页面这次不生成。 // SKIPPED：页面生成了，但目录里列的节或长文还没翻到，这次没进页面。 // 两回事，分开报，免得把「还没翻」说成「翻坏了」。 -const SKIP = new Set(); -const SKIPPED = []; +// +// 下面整段只在直接运行时执行（node tools/site/build.mjs）： +// tools/offline/build.mjs 和 web/ 的 Astro 构建都 import renderPage， +// import 时要是顺手写盘，Astro 每次构建都会把仓库根的生成物重写一遍。 +// 判据是 argv[1] 是不是本文件——import 时 argv[1] 是调用方的入口。 +// SKIP/SKIPPED 是模块级的：renderPage 往里写， +// 直接运行时底下那段读它们，被 import 时调用方自己看。 +export const SKIP = new Set(); +export const SKIPPED = []; + +const isMainEntry = !!process.argv[1] && resolve(process.argv[1]) === fileURLToPath(import.meta.url); +if (isMainEntry) { const stats = JSON.parse(read('tools/site/stats.json')); const template = read('tools/site/page.template.html'); @@ -479,4 +492,5 @@ console.log(`\n已生成检索页 ${done.length} 份（${done.join('、')}，默 if (SKIP.size) console.log(`跳过 ${[...SKIP].join('、')}：这一份的 README 和正文还没翻完，翻完重跑本脚本就会带上。`); if (SKIPPED.length){ console.log(`另有 ${SKIPPED.length} 项在目录里列着、正文还没翻到，这次没进页面：${SKIPPED.slice(0, 6).join('、')}${SKIPPED.length > 6 ? ' …' : ''}`); -} \ No newline at end of file +} +} // isMainEntry：被 import 时到这里结束，调用方只拿 renderPage 等函数 \ No newline at end of file diff --git a/tools/site/page.template.html b/tools/site/page.template.html index fa867a81..a4a5f026 100644 --- a/tools/site/page.template.html +++ b/tools/site/page.template.html @@ -779,10 +779,19 @@

    高性价比人生指南

    const XREF_RE = RE_XREF; // 指路里的分隔符按正则切（英文是 ", and to –" 一串，中文是「、」） const XREF_SEP = new RegExp(SITE.xref.listSep); -// 「8 到 10」摊成 8、9、10；上限 40 是防着把法条条款号那种大范围摊成几百个键 +// 「8 到 10」摊成 8、9、10；上限 40 是防着把法条条款号那种大范围摊成几百个键。 +// 先认整串区间，再按分隔符切：英文的 listSep 里含 "to" 和 "-"，先切的话 +// "8 to 11" 被切成 "8" 和 "11"，区间再也认不出来（译文检查那边同样的坑，见 structure.mjs） const xrefKeys = (sec, spec) => { const ns = []; - for (const part of String(spec).split(XREF_SEP)){ + const whole = String(spec).trim(); + const wr = RE_XRANGE.exec(whole); + if (wr){ + const a = +wr[1], b = +wr[2]; + if (b >= a && b - a <= 40) for (let i = a; i <= b; i++) ns.push(i); + return ns.map(n => `${sec}-${n}`).filter(k => ITEMS.has(k)); + } + for (const part of whole.split(XREF_SEP)){ const r = RE_XRANGE.exec(part); if (r){ const a = +r[1], b = +r[2]; @@ -1063,9 +1072,55 @@

    高性价比人生指南

    } document.getElementById('tot').textContent = CARDS.length; } +/* ---------- 预渲染 hydration ---------- + 服务端（web/src/lib/render.ts）按 build() 初次渲染的形状把 DOM 写好， + 这里按文档顺序索引到 CARDS/BLOCKS/TOC 三张表里。索引完的状态和 build() 刚跑完 + 一样：card.key 都是 null，vis 都是 false，滚进来由 CARD_IO 补渲染（terms=[]， + 和服务端写的一致，重渲出来是同一个 DOM）。 + 对不上就回退：预渲染块和侧栏目录删掉，走 build() 重建。回退时 console 里留一句， + 免得渲染器悄悄跑偏了没人知道——正常构建永远不该走到这里。 */ +function indexSsgDom(sections){ + for (const s of sections){ SECS.set(s.n, s); for (const e of s.entries) ITEMS.set(`${s.n}-${e.n}`, e); } + const total = sections.reduce((n, s) => n + s.entries.length, 0); + const list = document.getElementById('list'); + const blocks = [...list.querySelectorAll(':scope > section.sec-block')]; + const preCards = list.querySelectorAll('article.card').length; + if (blocks.length !== sections.length || preCards !== total){ + console.warn(`预渲染对不上（块 ${blocks.length}/${sections.length}，卡片 ${preCards}/${total}），回退到 build() 重建`); + list.querySelectorAll(':scope > section.sec-block').forEach(b => b.remove()); + const fsec = document.getElementById('f-sec'); fsec.innerHTML = ''; + document.querySelector('#gloss dl').innerHTML = ''; + build(sections); buildGlossary(); + return; + } + const fsec = document.getElementById('f-sec'); + const links = [...fsec.querySelectorAll('a[data-go]')]; + const subs = [...fsec.querySelectorAll('.toc-sub')]; + let li = 0, si = 0; + for (const s of sections){ + for (const e of s.entries){ const a = links[li++]; if (a) TOC.set(`e-${e.sec}-${e.n}`, a); } + const sub = subs[si++]; if (sub) TOC_SEC.set(s.n, sub); + } + let bi = 0; + for (const s of sections){ + const block = blocks[bi++]; if (!block) break; + BLOCKS.push({n:s.n, el:block, k:block.querySelector('.shown .k')}); + const cards = [...block.querySelectorAll('article.card')]; + let ci = 0; + for (const e of s.entries){ + const c = cards[ci++]; if (!c) break; + const card = {e, el:c, key:null, vis:false, f:{t:c.querySelector('.t'), cost:c.querySelector('.cost'), human:c.querySelector('.human'), gain:c.querySelector('.gain'), note:c.querySelector('.note'), src:c.querySelector('.src .body'), srcN:c.querySelector('.src summary .cnt'), details:c.querySelector('.src')}}; + CARDS.push(card); CARD_OF.set(c, card); CARD_IO.observe(c); + } + } + document.getElementById('tot').textContent = CARDS.length; +} function buildGlossary(){ const box = document.getElementById('gloss'); if (!GLOSS.length) { box.hidden = true; return; } const dl = box.querySelector('dl'); + // 预渲染（Astro 构建）已经把 dt/dd 写好了：这里只索引，不重建。 + // 没有 data-ssg 的页面 dl 是空的，走原来的路，一行没变。 + if (dl.children.length) return; for (const g of [...GLOSS].sort((a,b) => a.term.localeCompare(b.term, SITE.sortLocale))){ const dt = document.createElement('dt'); dt.textContent = g.term; const dd = document.createElement('dd'); dd.textContent = g.meaning; @@ -1488,7 +1543,15 @@

    高性价比人生指南

    GLOSS = parseGlossary(md); PERF.log(T('perf.parse'), t1, files.length + ' ' + T('perf.files')); status.remove(); - const t2 = performance.now(); build(sections); PERF.log(T('perf.buildDom'), t2, CARDS.length + ' ' + T('perf.cards')); + const t2 = performance.now(); + // 预渲染 hydration：Astro 构建时已经把 build() 会创建的 DOM 原样写进 HTML + // （卡片、侧栏目录、术语表，见 web/src/lib/render.ts），这里只索引现有节点， + // 不重新创建。判据是 #list 上的 data-ssg 和同页的语料，两个缺一个都走老路—— + // 老构建和离线单文件没有 data-ssg，它们的行为一行没变。 + const listEl = document.getElementById('list'); + if (EMBED && listEl && listEl.dataset.ssg === '1') indexSsgDom(sections); + else build(sections); + PERF.log(T('perf.buildDom'), t2, CARDS.length + ' ' + T('perf.cards')); buildGlossary(); readUrl(); wire(); apply(); PERF.mark(T('perf.mark', {dom: document.getElementsByTagName('*').length, terms: GLOSS.length})); if (location.hash){ diff --git "a/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" "b/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" index 07a82416..bb64bd25 100644 --- "a/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" +++ "b/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" @@ -4,12 +4,6 @@ 这一节讲三件事：车祸、火灾、中毒这类意外和伤害死亡；少数几种证据最硬的疫苗和筛查；身体出了什么信号就该去查一查。收益那一栏的数字，来自随机分组的试验、跟踪一大批人很多年的研究，或者官方统计。成本那一栏是作者按市面价格的粗略估计，只用来互相比较，不要拿去引用。 -[← Về mục lục](../README.md) - -# 1. Đừng chết sớm - -Mục này nói ba việc: chết do tai nạn và tổn thương như đụng xe, hỏa hoạn, ngộ độc; vài loại vắc-xin và xét nghiệm sàng lọc có bằng chứng chắc nhất; và những dấu hiệu cơ thể báo ra thì nên đi khám. Con số ở cột Lợi ích lấy từ thử nghiệm chia nhóm ngẫu nhiên, từ nghiên cứu theo dõi một nhóm người lớn trong nhiều năm, hoặc từ thống kê chính thức. Cột Chi phí là ước tính sơ bộ của tác giả theo giá thị trường, chỉ dùng để so sánh các mục với nhau, đừng trích dẫn. - ### 1. Cài dây đai an toàn, cả hàng trước lẫn hàng sau - Chi phí: Không tốn tiền. Mỗi lần lên xe thêm 2 giây. @@ -18,7 +12,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: NHTSA (2024). Occupant Protection in Passenger Vehicles: 2022 Data (DOT HS 813 573). ; WHO (2025). Road traffic injuries fact sheet. ; WHO GHO RS_196/RS_198 (China, 2021). - Ghi chú: Hàng sau cũng phải cài. Trong cùng báo cáo đó, 60% người chết ở hàng ghế thứ hai không cài dây. Hai con số 45% và 60% là ước tính Cơ quan quản lý an toàn giao thông Hoa Kỳ (NHTSA) tính ra từ cơ sở dữ liệu tai nạn (dẫn theo Kahane 2015), không phải từ thử nghiệm chia nhóm ngẫu nhiên. Con số tử vong của WHO ở Trung Quốc là ước tính, cao hơn nhiều lần con số Cục Cảnh sát giao thông đăng ký. Hai bên thống kê phạm vi khác nhau, ở đây chỉ dùng cách đo của WHO. - ### 2. Chạy xe máy, xe điện thì đội mũ bảo hiểm và cài chặt dây - Chi phí: Mua mũ 100 đến 300 tệ. Mỗi lần đội mũ mất 5 giây. @@ -27,7 +20,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Liu BC và cộng sự (2008). Helmets for preventing injury in motorcycle riders. Cochrane Database of Systematic Reviews. - Ghi chú: Mũ phải cài chặt dây, treo trên ghi đơn xe không tính là đội. Số liệu trên đến từ xe máy. Xe điện chưa có nghiên cứu trực tiếp, số đó suy ra theo, vì sau khi va chạm thì kiểu chấn thương đầu giống nhau. - ### 3. Lắp báo động khói; mùa đông đốt than hay dùng gas trong nhà thì lắp thêm báo động ô-xít cacbon - Chi phí: Một báo động khói 30 đến 100 tệ. Một báo động ô-xít cacbon 50 đến 150 tệ. Lắp xong thì mỗi năm thay pin một lần. @@ -36,7 +28,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: B - Nguồn: Marshall SW và cộng sự (1998). Fatal residential fires: who dies and who survives? JAMA. ; USFA (2022). Fatal Fires in Residential Buildings (2018-2020), Topical Fire Report Series 22(2). ; You J và cộng sự (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly. - Ghi chú: Lý do để ở mức B: phía báo động khói chỉ có một nghiên cứu, cách làm là so ngược hộ có sự cố với hộ không có, không phải thử nghiệm chia nhóm ngẫu nhiên. Bản thân báo động ô-xít cacbon không có nghiên cứu nào chứng minh giảm tỷ lệ tử vong. Lợi ích ở đây suy ra từ việc các ca chết vì ngộ độc dồn vào mùa đông và dồn trong nhà. - ### 4. Đến hạn thì thay ống mềm gas và bếp, không tự sửa ống, từ chối luôn người đến nhà bán gas - Chi phí: Một ống mềm gas đạt chuẩn vài chục tệ. Bếp thay mỗi vài năm một lần. Khi thay thì tiện tay xem đã quá hạn chưa. @@ -45,7 +36,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: 国务院 (Hội đồng Nhà nước) (2010). Quy định quản lý khí thành phố, 城镇燃气管理条例（国务院令第 583 号）. 第二十七条: 「燃气用户应当遵守安全用气规则，使用合格的燃气燃烧器具和气瓶，及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管等」 — người dùng gas phải tuân thủ quy tắc dùng gas an toàn, dùng thiết bị đốt gas và bình gas đạt chuẩn, và kịp thời thay mọi thiết bị đốt gas hoặc ống nối bị nhà nước yêu cầu loại bỏ hay đã hết hạn sử dụng. Các hành vi bị cấm ở 第二十八条 gồm 「（一）擅自操作公用燃气阀门」「（二）将燃气管道作为负重支架或者接地引线」「（四）擅自安装、改装、拆除户内燃气设施和燃气计量装置」 — tự ý vặn van gas công cộng, dùng ống gas làm giá đỡ tải hay làm dây tiếp đất, và tự ý lắp, sửa, tháo thiết bị gas trong nhà cùng thiết bị đo gas. 第二十条 cấm đơn vị kinh doanh gas 「（六）要求燃气用户购买其指定的产品或者接受其提供的服务」 — yêu cầu người dùng gas mua sản phẩm do họ chỉ định hoặc chấp nhận dịch vụ họ cung cấp. 第二十九条 cho phép người dùng khiếu nại về phí và dịch vụ với cơ quan quản lý gas, và 「有关部门应当自收到投诉之日起15个工作日内予以处理」 — cơ quan có trách nhiệm xử lý trong 15 ngày làm việc kể từ khi nhận được khiếu nại. ；深圳市住房和建设局 (Sở Nhà ở và Xây dựng Thâm Quyến) (2024). Kiến thức dùng gas an toàn (khoản 2), 安全用气小知识（二）: 「立即关闭家里的燃气总阀门」— đóng ngay van gas tổng trong nhà, 「迅速开门、开窗进行通风」— mở cửa và cửa sổ để thông gió, 「一定不要开、关家里的电灯、抽油烟机等电器」— tuyệt đối không bật tắt đèn hay máy hút mùi trong nhà, 「人员应该迅速撤离，转移到安全的地方，并且及时拨打燃气公司的报警电话」— mọi người phải rút ra ngay, sang chỗ an toàn, rồi gọi số báo động của công ty gas. ；张家港市应急管理局 (Cục Quản lý Khẩn cấp Trường Gia Cang) (2024). Kiến thức dùng gas an toàn, 燃气安全使用知识: 「不要情急之下打开排气扇进行通风」— đừng vội bật quạt thông gió, 「应远离现场之后才能打电话求助」— phải rời khỏi hiện trường rồi mới gọi cầu cứu. - Ghi chú: Có người mang danh「kiểm tra an toàn」đến nhà, kiểm tra xong là chào bán van tự đóng và báo động. Quy định không cho phép công ty gas bắt bạn mua sản phẩm do nó chỉ định, bạn từ chối được, khiếu nại cũng được. Ngửi thấy mùi gas thì tắt van gas tổng trước, gas bình thì tắt van trên bình. Rồi mở cửa mở cửa sổ thông gió, đừng bật quạt thông gió. Đèn, máy hút mùi các thiết bị đó đừng bật cũng đừng tắt, bật lửa cũng đừng đụng. Sau đó mọi người rút ra ngoài, đã rời hiện trường thì gọi số báo động của công ty gas. Ngộ độc ô-xít cacbon là chuyện khác, xem phần 13. - ### 5. Không hái, không mua, không ăn nấm hoang, mọi cách nhận biết kiểu dân gian đều không có cơ sở - Chi phí: Không tốn tiền. Chỗ khó là từ bỏ một món ăn hoang theo mùa. @@ -63,7 +53,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: 应急管理部 (Bộ Quản lý Khẩn cấp) (2021). Quy định quản lý an toàn cháy nổ công trình dân dụng cao tầng, 高层民用建筑消防安全管理规定（应急管理部令第 5 号，第三十七条、第四十七条）. ；应急管理部 (Bộ Quản lý Khẩn cấp) (2023). Thư trả lời đề xuất số 00969 của kỳ thứ Năm Ủy ban Quốc gia Đại biểu Nhân dân Trung Quốc, 关于政协第十三届全国委员会第五次会议第00969号提案答复的函: 「2021年，全国共接报电动自行车火灾1.8万起，造成65人死亡、222人受伤，直接财产损失达2.21亿元，其中发生在建筑物内3782起，造成60人死亡、175人受伤」 — năm 2021 cả nước tiếp nhận báo 18.000 vụ cháy xe điện, làm 65 người chết, 222 người bị thương, thiệt hại tài sản trực tiếp 221 triệu tệ, trong đó 3.782 vụ xảy ra trong tòa nhà, làm 60 người chết và 175 người bị thương. ；应急管理部 (Bộ Quản lý Khẩn cấp) (2025). Họp báo ngày 15 tháng 1 năm 2025, 2025年1月15日新闻发布会: 「火灾起数同比下降42.2%，伤亡人数下降超过90%」 — số vụ cháy giảm 42.2% so với cùng kỳ, số người bị thương tử vong giảm hơn 90%. - Ghi chú: Đừng mua pin độ lại, đừng dùng bộ sạc không rõ hãng, đừng sạc suốt đêm rồi bỏ đó. Pin lithium chừng nào bắt lửa là cháy hết trong mấy chục giây, kèm khói độc dữ. Xe đậu trong hành lang mà cháy thì cháy ngay lối thoát duy nhất của bạn. Số liệu Bộ Quản lý Khẩn cấp công bố: năm 2021 cả nước có 18.000 vụ cháy xe điện, 65 người chết. Trong đó 3.782 vụ cháy trong tòa nhà, 60 người chết, người chết gần như toàn ở trong nhà. Sau đợt xử lý năm 2024, số vụ cháy giảm 42,2% so với năm trước, số người bị thương tử vong giảm hơn chín phần mười, còn số tuyệt đối của năm đó không công bố. - ### 7. Đo huyết áp, cao rồi thì uống thuốc hạ cho tới mức chuẩn - Chi phí: Máy đo huyết áp điện tử 100 đến 200 tệ. Đo một lần mất 1 phút. Thuốc hạ huyết áp phần lớn mỗi tháng vài tệ đến vài chục tệ. Chỗ khó là không có triệu chứng vẫn phải uống tiếp. @@ -72,7 +61,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Ettehad D và cộng sự (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Lu J và cộng sự (2017). Prevalence, awareness, treatment, and control of hypertension in China (China PEACE Million Persons Project). Lancet. - Ghi chú: Đi đo huyết áp một lần trước đã. Ở Trung Quốc, trong 10 người cao huyết áp thì chưa đến 1 người kiểm soát được, nên lần đo này là một trong những việc hiệu quả chi phí cao nhất trong sách. Hạ tới bao nhiêu thì coi là đạt (130 hay 140) còn đang bàn, nhưng việc biết mình cao huyết áp và hạ nó xuống thì không ai tranh cãi. - ### 8. Từ 35 tuổi, chỉ cần thừa cân là đi xét nghiệm glucose lúc nhịn một lần, bình thường thì cứ ba năm xét lại - Chi phí: Xét nghiệm glucose lúc nhịn mười mấy tệ. Xét nghiệm hemoglobin glycate vài chục tệ. Đến bệnh viện rút máu một lần là xong. @@ -81,7 +69,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: US Preventive Services Task Force (2021). Screening for Prediabetes and Type 2 Diabetes. - Ghi chú: Phát hiện tiền đái thường không có nghĩa là phải uống thuốc. Khá nhiều người chỉ cần đổi lối sống là xoay ngược lại được. Ở người lớn Trung Quốc, số biết mình mắc đái thường luôn là thiểu số, nhiều người đợi đến lúc đã có biến chứng mới biết - ### 9. Không chạy quá tốc độ, không lái xe sau khi uống rượu - Chi phí: Không tốn tiền. Chạy chậm hơn một chút thì mỗi chuyến mất thêm vài phút. Chỗ khó là ở bàn ăn cơm phải uống ít đi. @@ -90,7 +77,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: B - Nguồn: WHO (2025). Road traffic injuries fact sheet. - Ghi chú: Cặp số 1% và 4% về tốc độ và tử vong được suy ra từ mô hình tai nạn, không phải từ thử nghiệm chia nhóm ngẫu nhiên. Về lái xe sau khi uống, WHO chỉ nêu hướng, không đưa số. Câu hỏi「uống bao nhiêu thì nguy cơ tăng bao nhiêu」mục này chưa đối chiếu tài liệu gốc. - ### 10. Trẻ dưới 4 tuổi thì dùng ghế an toàn, đừng bế trên tay - Chi phí: Một cái 300 đến 2000 tệ, dùng được mấy năm. Mỗi lần lên xe thêm 1 phút để cài đai cho con. @@ -99,7 +85,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: NHTSA (2024). Occupant Protection in Passenger Vehicles: 2022 Data (DOT HS 813 573). ; WHO (2025). Road traffic injuries fact sheet. - Ghi chú: Lắp ghế theo hướng dẫn, lắp sai coi như không lắp. Dưới 1 tuổi thì lắp ngược, cho trẻ quay về phía đuôi xe. Số liệu trên là ước tính chính thức của Mỹ tính ra từ dữ liệu tai nạn. - ### 11. Nhà có trẻ nhỏ thì lắp chặn cửa sổ và ban công, lưới chống muỗi không tính là biện pháp bảo vệ - Chi phí: Một bộ chặn cửa sổ hoặc khoá an toàn trẻ em vài chục tệ. Lắp xong mất mười phút. @@ -108,7 +93,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: B - Nguồn: Spiegel CN, Lindaman FC (1977). Children can't fly: a program to prevent childhood morbidity and mortality from window falls. American Journal of Public Health：「Significant reduction in falls resulted, particularly in the Bronx, where reported falls declined 50 percent from 1973 to 1975」— số vụ ngã giảm đáng kể, nổi bật ở Bronx, nơi số vụ ngã được báo cáo giảm một nửa từ 1973 đến 1975, và bản sửa đổi bộ luật vệ sinh thành phố New York năm 1976「to require that landlords provide window guards in apartments where children ten years old and younger reside」— buộc chủ nhà phải lắp thanh chắn cửa sổ ở căn hộ có trẻ em từ mười tuổi trở xuống. ；(2021). Unintentional Window Falls in Children and Adolescents. Academic Pediatrics：38.840 lượt khám cấp cứu ghi nhận trong hệ thống giám sát tổn thương điện tử toàn quốc từ tháng 1 năm 2007 đến tháng 8 năm 2017，「The majority of falls occurred in children under the age of 6 and were related to falls from a second story or below」— phần lớn vụ ngã xảy ra ở trẻ dưới 6 tuổi và liên quan đến rơi từ tầng hai trở xuống. - Ghi chú: Không đặt giường, sofa hay thùng ở mép cửa sổ và mép ban công, những thứ trẻ leo lên được. Lưới chống muỗi chỉ chặn muỗi, đẩy là rơi, đừng coi là thanh chắn. Cũng không phải chỉ ai ở tầng cao mới cần lắp: trong số liệu của Mỹ, phần lớn vụ ngã xảy ra ở tầng hai hoặc thấp hơn. - ### 12. Trẻ nhỏ ở gần nước thì không được rời mắt, chèo thuyền hay bơi ngoài vùng nước an toàn thì mặc áo phao - Chi phí: Áo phao 50 đến 200 tệ. Ngoài ra còn phải bỏ ra phần tâm trí để trông trẻ. @@ -126,7 +110,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Sherrington C và cộng sự (2019). Exercise for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Gillespie LD và cộng sự (2012). Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Lu Z và cộng sự (2021). Characteristics of Falls Among Older People — China, 2018. China CDC Weekly. - Ghi chú: Mục này áp dụng cho người trên 60 tuổi sống ở nhà mình, không ở viện dưỡng lão. Các nghiên cứu trên đếm số lần ngã, không đếm số người chết. Đường đi từ một cú ngã đến gãy xương hông rồi đến tử vong là rất ngắn. - ### 14. Xét nghiệm 5 chỉ số viêm gan B, không có kháng thể thì tiêm vaccine - Chi phí: Xét nghiệm một lần vài chục tệ. Vaccine viêm gan B cho người lớn gồm 3 mũi, tổng cộng 100 đến 300 tệ. Tiêm hết trong nửa năm. @@ -135,7 +118,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Qu C và cộng sự (2014). Efficacy of neonatal HBV vaccination on liver cancer and other liver diseases over 30-year follow-up of the Qidong hepatitis B intervention study. PLoS Medicine. ; Cui F và cộng sự (2017). Prevention of Chronic Hepatitis B after 3 Decades of Escalating Vaccination Policy, China. Emerging Infectious Diseases. - Ghi chú: Mục này áp dụng cho người lớn sinh trước năm 1992, chưa tiêm vaccine hoặc đã mất kháng thể. Con số giảm 84% ung thư gan đến từ nhóm được tiêm ngay khi sinh. Trong cùng thử nghiệm đó, người đến tuổi thanh niên mới tiêm bổ sung chỉ ngăn được 21% khả năng mang virus. Người lớn tiêm chủ yếu để không bị nhiễm, không thể áp thẳng con số 84% đó. Người đã nhiễm thì tiêm vaccine không có tác dụng, phải đi khoa gan theo dõi định kỳ. Cách biết đã nhiễm: trên phiếu xét nghiệm, mục kháng nguyên bề mặt viêm gan B HBsAg là dương tính. - ### 15. Bị đinh sắt, gai gỗ đâm vào hoặc vết thương dính đất thì xử lý ngay trong ngày, tiện thể hỏi có cần tiêm phòng uốn dương không - Chi phí: Khám và xử lý vết thương vài chục tệ. Nếu bác sĩ bảo phải tiêm, vaccine hoặc globulin miễn dịch vài chục đến vài trăm tệ. @@ -144,7 +126,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: B - Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ (美国疾病控制与预防中心). 破伤风 (uốn dương)：「Tetanus bacteria can get into someone's body through broken skin, usually through injuries.」— vi khuẩn uốn dương lọt vào cơ thể qua da bị rách, thường là qua vết thương.「Tetanus can lead to death (1 in 10 cases in the United States are fatal).」— bệnh này có thể dẫn tới tử vong, ở Mỹ cứ khoảng mười ca thì một ca chết.「People who didn't complete the primary series or who aren't up to date with their 10-year tetanus booster shots are also at increased risk.」— người tiêm chưa đủ liệu trình cơ bản, hoặc chưa đến hạn tiêm nhắc mười năm một lần, cũng có nguy cơ cao hơn.「Vaccination also helps prevent tetanus in people with wounds, depending on their tetanus vaccination history.」— tiêm vaccine còn giúp phòng uốn dương cho người có vết thương, tuỳ vào tiền sử tiêm phòng của họ. ；国家卫生健康委办公厅 (Văn phòng Ủy ban Quốc gia Y tế và Sức khoẻ) (2024). 关于印发非新生儿破伤风诊疗规范（2024 年版）的通知 — Thông báo ban hành Quy chuẩn chẩn đoán và điều trị uốn dương ở người không phải trẻ sơ sinh, bản 2024 (国卫办医急函〔2024〕381 号, bậc nguy cơ nằm ở trang 16 phụ lục, bảng 4 ở trang 20). ；phụ lục dạng PDF (bản quét). - Ghi chú: Tới bệnh viện để bác sĩ quyết định, đừng tự ý hạ tiêm, đừng thấy phiền mà không hỏi. Không phải vết thương nào cũng cần tiêm. Bác sĩ sẽ xem vết thương thuộc bậc nào, hỏi lần tiêm gần nhất cách đây bao lâu, rồi mới quyết định.「Tiêm một lần hồi nhỏ thì hiệu lực cả đời」là sai. Vết thương bẩn thì cách mũi tiêm trước đủ 5 năm phải tiêm nhắc. Nên trước khi đi, nhớ lần cuối cùng bạn tiêm phòng uốn dương là năm nào. - ### 16. Phụ nữ tiêm vaccine HPV, càng sớm càng tốt - Chi phí: Vaccine hai giá trị sản xuất trong nước hơn 300 tệ một mũi, vaccine chín giá trị nhập khẩu khoảng 1300 tệ một mũi. Tổng cộng 2 đến 3 mũi. Mỗi lần chạy đi tiêm mất nửa ngày. @@ -153,7 +134,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Lei J và cộng sự (2020). HPV Vaccination and the Risk of Invasive Cervical Cancer. New England Journal of Medicine. - Ghi chú: Tiêm ở 9 đến 14 tuổi thì lợi ích lớn nhất. Tiêm ở 17 đến 30 tuổi vẫn còn hơn một nửa mức giảm. Sau 30 tuổi lợi ích cứ giảm dần. Đã tiêm vaccine vẫn phải làm xét nghiệm sàng lọc ung thư cổ tử cung, vaccine không thay thế được việc sàng lọc, xem mục 18 của phần này (phụ nữ trên 30 tuổi làm xét nghiệm sàng lọc ung thư cổ tử cung). - ### 17. Phụ nữ từ 40 tuổi làm xét nghiệm sàng lọc ung thư vú, chụp mammogram mỗi hai năm một lần - Chi phí: Mỗi lần chụp mammogram hai ba trăm tệ. Nhiều địa phương có chương trình sàng lọc miễn phí. Hai năm một lần, mất nửa ngày. @@ -162,7 +142,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: US Preventive Services Task Force (2024). Breast Cancer: Screening. - Ghi chú: Bắt đầu sàng lọc từ tuổi nào thì mỗi nước khác nhau, ở Trung Quốc có hướng dẫn và chương trình địa phương bắt đầu từ 45 tuổi. Người có cha mẹ, con cái hay anh chị em ruột mắc ung thư vú, và người mang đột biến BRCA, thuộc nhóm nguy cơ cao, phải tự đi tìm bác sĩ để lập kế hoạch riêng, mục này không áp dụng - ### 18. Phụ nữ trên 30 tuổi làm xét nghiệm sàng lọc ung thư cổ tử cung, ưu tiên xét nghiệm HPV - Chi phí: Xét nghiệm HPV một lần 100 đến 300 tệ. Kết quả âm tính thì 5 năm xét một lần. Lấy mẫu chỉ mất vài phút. @@ -180,7 +159,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Hewitson P và cộng sự (2007). Screening for colorectal cancer using the faecal occult blood test, Hemoccult. Cochrane Database of Systematic Reviews. ; Bretthauer M và cộng sự (2022). Effect of Colonoscopy Screening on Risks of Colorectal Cancer and Related Death. New England Journal of Medicine. - Ghi chú:Tranh cãi. NordICC tính trên nhóm「được mời đi nội soi đại tràng」, tỷ lệ chết vì ung thư trực tràng và ruột già là 0.28% so với 0.31% (RR 0.90, 95% CI 0.64 đến 1.16), chênh lệch vẫn nằm trong biên độ sai số. Điều này cho thấy lợi ích của nội soi đại tràng với tỷ lệ tử vong trước đây bị ước tính thừa. Bằng chứng của xét nghiệm phân về tỷ lệ tử vong lại chắc hơn. Hướng dẫn các nước đặt tuổi bắt đầu sàng lọc trong khoảng 45 đến 50. Ai trong nhà đã mắc bệnh này thì bắt đầu sớm hơn. - ### 20. Người bệnh tim mạch và người cao tuổi tiêm vaccine cúm mỗi năm - Chi phí: Một mũi 50 đến 150 tệ. Mỗi mùa thu tiêm một lần. @@ -189,8 +167,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Fröbert O và cộng sự (2021). Influenza Vaccination After Myocardial Infarction: A Randomized, Double-Blind, Placebo-Controlled, Multicenter Trial. Circulation. ; Behrouzi B và cộng sự (2022). Association of Influenza Vaccination With Cardiovascular Risk: A Meta-analysis. JAMA Network Open. ；Demicheli V và cộng sự (2018). Vaccines for preventing influenza in the elderly. Cochrane Database of Systematic Reviews. （bản tổng hợp Cochrane mà phần Ghi chú nhắc tới） - Ghi chú:Tranh cãi. Mức A chỉ áp dụng cho người bệnh tim mạch. Với người cao tuổi vốn khoẻ, bản tổng hợp Cochrane chỉ đưa ra một kết luận: tỷ lệ mắc cúm trong một mùa giảm từ 6% xuống 2.4%. Bản đó đánh giá độ tin cậy của bằng chứng này là「低」— thấp, còn bằng chứng về tỷ lệ tử vong là「极低」— rất thấp. Bản tổng hợp nằm ở bài của Demicheli V và cộng sự năm 2018. - - ### 21. Từ 50 tuổi, tiêm vaccine thủy độc - Chi phí: Vaccine thủy độc tái tổ hợp phải tiêm hai mũi, tổng cộng khoảng 3.000 đến 4.000 tệ. Toàn bộ số tiền này người bệnh tự bỏ ra. @@ -199,7 +175,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Lal H, Cunningham AL, Godeaux O, et al. (2015). Efficacy of an adjuvanted herpes zoster subunit vaccine in older adults. New England Journal of Medicine, 372(22), 2087-2096. - Ghi chú: Xếp hạng hiệu quả chi phí của mục này khá thấp, lý do là đắt. Bản thân bệnh thủy độc hiếm khi chết người. Chỗ khó chịu là đau thần kinh sau thủy độc, có thể đau mấy tháng đến mấy năm. Sau khi tiêm, tỷ lệ đau chỗ tiêm và sốt đều không thấp. Nặng nhất trong đó, tức tác dụng phụ bậc ba, là 17.0% so với 3.2% - ### 22. Người từ 65 tuổi tiêm vaccine phế cầu khuẩu - Chi phí: Vài trăm tệ. Ở một số vùng, người cao tuổi được tiêm miễn phí. @@ -208,7 +183,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Bonten MJ, Huijts SM, Bolkenbaas M, et al. (2015). Polysaccharide conjugate vaccine against pneumococcal pneumonia in adults. New England Journal of Medicine, 372(12), 1114-1125. - Ghi chú: Nó chỉ phòng vài chủng có trong danh sách, không phải tiêm xong là chắc không viêm phổi. Người cao tuổi ở Trung Quốc hay tiêm là vaccine đa đường 23 giá trị, khác loại với vaccine liên hợp 13 giá trị dùng trong thử nghiệm này. Số hiệu lực ở trên không áp thẳng sang loại đa đường được. - ### 23. Xét nghiệm vi khuẩn H. pylori, dương tính thì diệt trùng - Chi phí: Xét nghiệm hơi thở carbon 13 hoặc carbon 14 giá 100 đến 200 tệ. Diệt trùng phải uống bốn loại thuốc cùng lúc, hai tuần thuốc giá 200 đến 500 tệ. Chỗ khó là suốt hai tuần phải uống đúng giờ. @@ -217,7 +191,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Li WQ và cộng sự (2019). Effects of Helicobacter pylori treatment and vitamin and garlic supplementation on gastric cancer incidence and mortality: follow-up of a randomized intervention trial. BMJ. - Ghi chú: Sau khi diệt trùng vẫn có thể nhiễm lại, nên người trong nhà cùng xét, cùng điều trị. Dân số của thử nghiệm này là cư dân vùng có tỷ lệ ung thư dạ dày cao. Người sống ở vùng tỷ lệ thấp sẽ nhận lợi ích thực tế nhỏ hơn nhiều. - ### 24. Người hút thuốc nặng chụp CT ngực liều thấp mỗi năm một lần - Chi phí: Một lần 200 đến 400 tệ, làm xong trong 10 phút. Thêm nữa còn phải tính cả lần xét lại sau khi hoảng loạn, và thời gian lo lắng suốt thời gian đó. @@ -226,7 +199,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: National Lung Screening Trial Research Team (2011). Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. New England Journal of Medicine. - Ghi chú: Mục này chỉ áp dụng cho người hút thuốc nguy cơ cao. Trong cùng thử nghiệm đó, 24.2% người chụp CT liều thấp có kết quả khả nghi. Trong số này 96.4% chỉ là hoảng loạn thật, tức dương tính giả. Người không hút thuốc đi làm xét nghiệm này là tiền mua lo âu. - ### 25. Bị trầm cảm hoặc có ý nghĩ tự sát thì bấm 12356, trong nhà không tích trữ thuốc ngủ và thuốc trừ sâu - Chi phí: Không tốn tiền. Chỉ cần gọi một cuộc. Ngoài ra cất khoá thuốc, không để thuốc trừ sâu trong nhà. @@ -244,7 +216,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: B - Nguồn: Bộ Quản lý Ứng cứu (应急管理部) (2020). 全国基础版家庭应急物资储备建议清单. ; Kendrick D và cộng sự (2012). Home safety education and provision of safety equipment for injury prevention. Cochrane Database of Systematic Reviews. ; Sở Giám sát Thị trường huyện Vũ Thắng (武胜县市场监督管理局) (2025). 手提式灭火器、防火门（窗）、点型感温（烟）火灾探测器消费提示. - Ghi chú:Tranh cãi. Trong cùng bản tổng hợp Cochrane đó, nhóm can thiệp không phát thiết bị làm tỷ lệ bị thương thấp hơn khoảng hai phần mười. Giá trị gốc IRR 0.78, khoảng tin cậy 95% CI 0.66 đến 0.92. Nhóm được phát thiết bị không làm tốt hơn. Cái chính làm giảm tổn thương là giáo dục và cải tạo tại nhà, nên lợi ích của thiết bị ghi「vừa」chứ không ghi「lớn」. Mua bình chữa cháy thì nhìn dấu CCC, đồng hồ của bình dạng áp lực phải dừng ở vùng xanh. Báo động khói và báo động CO xem mục 3 của phần này. Cách dùng gar xem mục 12 ở phần 13. Chọn từng món thế nào, để ở đâu, mấy năm đổi một lần, xem [docs/家庭应急装备清单.md](../docs/家庭应急装备清单.md) - ### 27. Trong nước tiểu có máu nhìn thấy bằng mắt, dù không đau, dù ngày mai đã sạch, vẫn phải khám một lần - Chi phí: Xét nghiệm nước tiểu vài chục tệ, siêu âm hệ tiết niệu một hai trăm tệ. Nếu bác sĩ thấy cần, thêm một nội soi bàng quang, khoảng một nghìn tệ. @@ -253,7 +224,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: B - Nguồn: Price SJ, Shephard EA, Stapley SA, Barraclough K, Hamilton WT (2014). Non-visible versus visible haematuria and bladder cancer risk: a study of electronic records in primary care. British Journal of General Practice, 64(626), e584-e589. - Ghi chú:「Không đau」là điểm mấu chốt. Máu trong nước tiểu do sỏi và nhiễm trùng, phần lớn đi kèm đau quặn lưng bụng, hoặc tiểu nhiều, tiểu gấp, tiểu đau. Máu do khối u thường không có cảm giác gì, lại tự dừng vài ngày rồi tái diễn, nhiều người bỏ qua. Đang dùng thuốc chống đông không phải lý do để không đi khám. Máu trong nước tiểu sau vận động mạnh, hoặc ở phụ nữ trước và sau kỳ kinh, có thể xét lại một lần rồi hẳng quyết định. Lợi ích ghi「nhỏ」là vì chỉ có tỷ lệ phát hiện, không có cỡ tác động kiểu「khám sớm giảm được bao nhiêu tử vong」. - ### 28. Chức năng cương có vấn đề, đi khám tim mạch trước, đừng coi đó chỉ là chuyện ấy - Chi phí: Khám bệnh vài chục tệ. Đo huyết áp, rút máu xét đường huyết và mỡ máu, một hai trăm tệ. Đi một chuyến bệnh viện, chờ vài ngày có kết quả. @@ -262,7 +232,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Vlachopoulos CV, Terentes-Printzios DG, Ioakeimidis NK, Aznaouridis KA, Stefanadis CI (2013). Prediction of cardiovascular events and all-cause mortality with erectile dysfunction: a systematic review and meta-analysis of cohort studies. Circulation: Cardiovascular Quality and Outcomes, 6(1), 99-109. ；DailyMed (thư viện dữ liệu dược phẩm của Thư viện Y khoa Quốc gia Hoa Kỳ, 美国国立医学图书馆). VIAGRA (sildenafil citrate) tablet, CONTRAINDICATIONS. - Ghi chú: Cả 14 nghiên cứu đó đều theo dõi một nhóm người rồi xem kết quả, không phải thử nghiệm chia nhóm ngẫu nhiên. Vì thế「nhiều hơn bốn phần mười」là liên hệ, không phải quan hệ nhân quả. Thỉnh thoảng một hai lần thì không tính, phải kéo dài trên ba tháng mới xem là vấn đề. Những thứ cần khám là cùng một bó với mục 7 và 8 của phần này: đo huyết áp, xét đường huyết lúc nhịn, cộng mỡ máu, hút thuốc và vòng eo. Đừng mua「thuốc bổ sung」cường dương trên mạng. Thuốc đúng loại là thuốc kê đơn, tờ hướng dẫn liệt kê nhóm nitrat là chống chỉ định, người đang dùng nitroglycerin dạng tương tự thì không được dùng, dùng chung sẽ làm huyết áp tụt mạnh. Cải thiện bằng cách nào xem mục 29 của phần này (giảm cân, cai thuốc làm chức năng cương tốt lên). Lợi ích ghi「vừa」chứ không ghi「lớn」: 44% là chênh lệch nguy cơ giữa hai nhóm người, không phải「khám một lần giảm được bao nhiêu tử vong」. Cũng không ghi「nhỏ」: loại tín hiệu như mục 27 của phần này (máu trong nước tiểu nhìn thấy bằng mắt) chỉ có tỷ lệ phát hiện, còn ở đây những thứ phát hiện ra như huyết áp cao, tiểu đường, mỡ máu bất thường đều có sẵn cách điều trị. - ### 29. Giảm cân, cai thuốc, giữ huyết áp và đường huyết ổn định, chức năng cương cũng tốt lên theo - Chi phí: Không tốn tiền, còn có thể tiết kiệm. Chỗ khó là duy trì lâu dài: kiềm chế ăn, vận động, cai thuốc. @@ -280,7 +249,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Giannou FK, Tsiara CG, Nikolopoulos GK, et al. (2016). Condom effectiveness in reducing heterosexual HIV transmission: a systematic review and meta-analysis of studies on HIV serodiscordant couples. Expert Review of Pharmacoeconomics & Outcomes Research, 16(4), 489-499. ；Fonner VA và cộng sự (2016). Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS. （phòng trước phơi nhiễm PrEP） - Ghi chú: Chênh lệch giữa「dùng không đều」và「không dùng」rất nhỏ, phải dùng mỗi lần, suốt quá trình mới có tác dụng. Dùng chung kim tiêm là con đường lây hiệu quả khác, độc lập với quan hệ tình dục, tiêm ma tuý, xăm mình, xuyên lỗ tai đều tính vào. Nhóm nguy cơ cao còn dùng được thuốc phòng trước phơi nhiễm (PrEP), tức uống thuốc trước để chặn nhiễm. Gộp 18 nghiên cứu lại, mọi thử nghiệm có trên bảy phần mười người uống thuốc đúng giờ thì người dùng PrEP có nguy cơ nhiễm HIV thấp hơn nhóm giả dược khoảng bảy phần mười. Giá trị gốc là tỷ số nguy cơ 0.30, khoảng tin cậy 95% CI 0.21 đến 0.45. Ở những thử nghiệm có ít người uống thuốc đúng giờ thì không thấy tác dụng bảo vệ. PrEP là thuốc kê đơn, đi hỏi khoa bệnh truyền nhiễm hoặc trung tâm kiểm soát bệnh, đừng tự mua trên mạng. Sau khi đã phơi nhiễm thì xử lý thế nào, xem phần 13 - ### 31. Từng có hành vi nguy cơ cao thì xét nghiệm AIDS một lần, miễn phí ở Trung tâm kiểm soát bệnh và giữ kín kết quả - Chi phí: Không tốn tiền. Một chuyến tới Trung tâm kiểm soát bệnh hoặc cơ sở y tế được chỉ định, rút một ống máu. Chỗ khó là bước mở miệng đi xét nghiệm. @@ -289,7 +257,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Bộ Y tế (卫生部) (2006). Quy định quản lý việc xét nghiệm AIDS toàn quốc, 全国艾滋病检测工作管理办法. ；Ủy ban Nhà nước (国务院) (2006). Nghị định phòng ngừa và xử lý AIDS, 艾滋病防治条例（第三、二十三、三十八条）, gồm cấm phân biệt đối xử với người nhiễm, quyền được tư vấn và xét nghiệm miễn phí, và nghĩa vụ khai báo của người nhiễm. （Ủy ban Y tế và Sức khỏe thành phố Bắc Kinh, 北京市卫生健康委员会, đăng lại）；Rodger AJ, Cambiano V, Bruun T, et al. (2019). Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. The Lancet, 393(10189), 2428-2438. ；Trung tâm Kiểm soát và Phòng ngừa bệnh Quảng Đông (广东省疾病预防控制中心). Giáo trình phổ thông về thời gian cửa sổ xét nghiệm AIDS, 艾滋病检测窗口期科普. （con số về thời gian cửa sổ） - Ghi chú: Thời gian cửa sổ quyết định xét nghiệm vào lúc nào mới có ý nghĩa. Cơ sở của Trung tâm kiểm soát bệnh Quảng Đông (广东省疾控中心) là: xét nghiệm axit nucleic khoảng 1 tuần là phát hiện được, xét nghiệm kháng nguyên kết hợp kháng thể thế hệ thứ tư khoảng 2 tuần, xét nghiệm kháng thể thế hệ thứ ba khoảng 3 tuần. Nên chờ ít nhất 2 tuần rồi hãy xét, cách 2 đến 4 tuần xét lại một lần. Sau khi lần hành vi nguy cơ cao nhất đã tròn 3 tháng, trên 99.99% có thể loại trừ nhiễm. Que thử tự cho dương tính không có nghĩa là chẩn đoán, phải tới Trung tâm kiểm soát bệnh hoặc cơ sở chính quy để kiểm chứng. Điều 3 của nghị định quy định không cá nhân và tổ chức nào được phân biệt đối xử với người nhiễm và gia đình họ. Quyền về hôn nhân, việc làm, khám chữa bệnh, nhập học được pháp luật bảo vệ. Điều 38 quy định thêm rằng người nhiễm có nghĩa vụ「将感染或者发病的事实及时告知与其有性关系者」— kịp thời báo cho người có quan hệ tình dục với mình biết sự thật mình đã nhiễm hoặc đã mắc bệnh. Sàng lọc ba bệnh trong thai kỳ và phòng lây truyền từ mẹ sang con xem ở phần 27 - ### 32. Vừa nảy ý tưởng tự sát thì nói với một người bên cạnh, giao mấy chục phút ấy ra - Chi phí: Không tốn tiền. Chỗ khó là lúc mở miệng, và thừa nhận rằng mấy tiếng đó mình không ở một mình được. @@ -307,7 +274,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: B - Nguồn: Gawarammana IB, Buckley NA (2011). Medical management of paraquat ingestion. British Journal of Clinical Pharmacology, 72(5), 745-757. ；Chowdhury FR 等 (2026). Clinical and Laboratory Profile of Paraquat Poisoning: A Toxicological Crisis in Bangladesh. American Journal of Tropical Medicine and Hygiene, 115, 176-182. ；Weaver LK, Hopkins RO, Chan KJ, et al. (2002). Hyperbaric oxygen for acute carbon monoxide poisoning. New England Journal of Medicine, 347(14), 1057-1067. ；Parkinson RB, Hopkins RO, Cleavinger HB, et al. (2002). White matter hyperintensities and neuropsychological outcome following carbon monoxide poisoning. Neurology, 58(10), 1525-1532. ；Văn phòng Bộ Nông nghiệp và Nông thôn (农业农村部办公厅) (2018). Thông báo siết chặt quản lý paraquat, 关于切实加强百草枯监督管理的通知（农办农〔2018〕17 号）. - Ghi chú: Lý do xếp B: đối tượng của các nghiên cứu này là người bị ngộ độc, không phải một nhóm người tự sát không thành công thuần tuý. Thử nghiệm carbon monoxide còn lẫn cả người bị ngộ độc do tai nạn. Vì vậy tỷ lệ di chứng ở trên không dùng thẳng làm「xác suất hậu quả với người tự sát không thành công」được. Điều chắc chắn là không cách nào mà không để lại di chứng. Xử lý tại chỗ người ngộ độc carbon monoxide xem mục 19 ở phần 13. Uống nhầm thuốc trừ sâu hay thuốc, đừng nôn, mang theo vỏ thuốc đi khám, xem mục 20 ở phần 13. Trong nhà không tích trữ thuốc trừ sâu và thuốc ngủ xem mục 25 của phần này. Sách này không so tỷ lệ tử vong giữa các cách, cũng không viết cách làm cụ thể nào - ### 34. Đừng cược vào chuyện「nằm vài ngày là khỏe」khi ngã từ trên cao: phần lớn người vào ICU chấn thương vẫn sống, cái giá thì tính theo năm - Chi phí: Không tốn tiền. Chỉ cần nhìn rõ cái sổ này. @@ -316,7 +282,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: B - Nguồn: Chen Y, Li W, Wang X, Zhong Q, Abudurexiti A, Qiu Q, Li J, Luo J (2025). Clinical features and prognostic predictors for patients admitted to trauma intensive care unit due to fall from height in South Xinjiang. International Journal of Emergency Medicine, 18, 142. ；Simmel S, Wurm S, Drisch S, Woltmann A, Coenen M (2020). Prädiktion der Rückkehr zur Arbeit nach Polytrauma bei Patienten mit einem ISS von mindestens 25. Die Rehabilitation, 59(2), 95-103. ；Ủy ban Quốc gia Y tế và Sức khỏe (国家卫生健康委) (2025). Giải thích báo cáo thống kê về phát triển y tế và sức khoẻ năm 2024, 2024 年我国卫生健康事业发展统计公报解读. - Ghi chú: Lý do xếp B: nhóm ngã rơi được thống kê từ hồ sơ cũ của một bệnh viện, không tách tai nạn với tự làm hại. Nhóm quay lại làm việc chỉ có 84 người, lại ở Đức. Hai nhóm đều không dùng làm「một người nhảy lầu sẽ thế nào」được. Chi phí trung bình mỗi lần nhập viện là trung bình của mọi người nhập viện cả nước, không phải chi phí chấn thương do rơi, dẫn vào đây chỉ để có một mức dưới có thể đối chiếu. Bản gốc chỉ cho phân bố chiều cao của nhóm này, nên ở đây không viết quan hệ giữa chiều cao và hậu quả. Loét tì đè cần theo dõi nhất trong thời gian nằm liệt dài xem mục 8 ở phần 17. Huyết khối tĩnh mạch sâu và thuyên tắc phổi xem mục 11 ở phần 13 (một chân sưng phù bất thường). Lúc ý tưởng vừa nảy ra thì làm gì, xem mục 32 của phần này. Hậu quả của ngộ độc xem mục 33 của phần này (đừng coi「cứu được」là chốt chặn). Dưới đây là trải nghiệm của chính tác giả: rơi từ tầng bốn, trước đó nghĩ「mười mấy ngày là khỏe」, thực tế nằm viện một năm, phẫu thuật khắp bốn nơi, lắp thép, nửa năm không tự lật người ngồi dậy được, nhà phải vay tiền - ### 35. Đừng đổi tiền lấy chuyện「thiếu một quả thận cũng không sao」: quả còn lại phải làm việc thay hai quả, người bán thận sau đó có 86% nói sức khoẻ xấu đi - Chi phí: Không tốn tiền. Chỉ cần nhìn rõ cái sổ này. @@ -334,7 +299,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: A - Nguồn: Bộ Môi trường và Tự nhiên (生态环境部) (2025). Báo cáo tình hình môi trường Trung Quốc năm 2024, 2024 中国生态环境状况公报. ；Cục An toàn hạt nhân quốc gia (国家核安全局) (2026). Kết quả giám sát, kiểm tra hạt nhân và bức xạ an toàn năm 2025, 2025 年我国核与辐射安全监管工作成效. ；Bộ Môi trường và Tự nhiên (生态环境部) (2021). Thông báo xử lý một vụ tai nạn phóng xạ do làm kiểm tra tia gamma trái phép, 关于一起 γ 射线探伤作业违法违规造成辐射事故案件处理情况的通报. ；Ủy ban Nhà nước (国务院) (2019). Nghị định an toàn và phòng hộ với đồng vị phóng xạ và thiết bị tia X, 放射性同位素与射线装置安全和防护条例（2019 年修订，第三十四、四十二、六十八条）, nội dung về cấp tai nạn phóng xạ, phát hành giấy phép và nghĩa vụ báo cáo. ；Bộ Bảo vệ môi trường (环境保护部) (2009). Thông báo lập chế độ phân cấp, xử lý và báo cáo tai nạn phóng xạ, 关于建立放射性同位素与射线装置辐射事故分级处理和报告制度的通知（辐射事故分级标准）. ；International Atomic Energy Agency (1988). The Radiological Accident in Goiânia. IAEA, Vienna. - Ghi chú: Nếu thật sự gặp: tránh xa, nhớ vị trí, đừng di chuyển, gọi 110. Nghị định quy định phải báo cho cơ quan quản lý môi trường, cơ quan công an và cơ quan quản lý y tế. Khoảng cách và thời gian là cách bảo vệ duy nhất có tác dụng, để trong túi hay bọc túi nhựa cũng vô ích. Máy X quang, CT là loại thiết bị tạo tia bằng cách cấp điện, ngắt điện thì không còn phóng xạ lưu lại. Chỗ nguy hiểm thật sự là nguồn phóng xạ kín bên trong thiết bị. Máy xạ trị ở bệnh viện, máy soi tia gamma gắn trên ống dẫn và nồi hơi công nghiệp, máy đo giếng, máy đo mực nước trong bể đều chứa nguồn. Tháo vỏ ngoài là lộ nguồn ra. Mà vẻ ngoài của chúng không khác gì phần kim loại thường. Vì vậy đừng trông chờ nhận ra nguồn phóng xạ, linh kiện thiết bị không rõ lai lịch thì không tháo, không bán, không mang về nhà. Người làm nghề thu gom phế liệu, tháo dỡ, bán đồ ngũ kim và công nhân công trường có nguy cơ cao hơn người thường nhiều. Thấy khu vực cảnh giới có vẽ ký hiệu cỏ ba lá ở công trường thì đi vòng. Rò rỉ ở nhà máy điện hạt nhân là loại tai nạn khác, người thường gặp ít hơn nữa. Người hưởng lợi chủ yếu là bạn và người nhà đang ở cùng hiện trường. - ### 37. Cờ bạc tới mức vay tiền còn cờ, tâm trạng lúc nào cũng tụt thì gọi 12356, rồi giao thẻ ngân hàng và mật khẩu thanh toán cho người nhà giữ - Chi phí: Không tốn tiền. Một cuộc gọi. Giao thẻ ngân hàng, mật khẩu thanh toán và ứng dụng vay tiền cho người nhà quản lý. Chỗ khó là phải thừa nhận trước rằng mình không kiểm soát nổi. @@ -343,7 +307,6 @@ Mục này nói ba việc: chết do tai nạn và tổn thương như đụng x - Mức bằng chứng: B - Nguồn: Karlsson A, Håkansson A. (2018). Gambling disorder, increased mortality, suicidality, and associated comorbidity: A longitudinal nationwide register study. Journal of Behavioral Addictions, 7(4), 1091–1099. ；Karlsson A, Håkansson A. (2023). Corrigendum to: Gambling disorder, increased mortality, suicidality, and associated comorbidity. Journal of Behavioral Addictions, 12(1), 302. ；Ủy ban Quốc gia Y tế và Sức khỏe (国家卫生健康委) (2024). Thông báo dùng số điện thoại 12356 làm đường dây hỗ trợ tâm lý thống nhất toàn quốc, 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. - Ghi chú: Lý do xếp B: chỉ có một nghiên cứu, và chỉ tính người đã tới chuyên khoa khám. Người trong nhóm này bệnh nặng hơn, nên con số 15 lần không áp được cho người chơi bài nói chung. Giao tiền cho người nhà quản lý là cách làm theo kinh nghiệm, không có nghiên cứu trực tiếp, tính riêng là mức C. Vạch đỏ pháp lý của cờ bạc xem mục 13 ở phần 9 (không thu phần trăm đầu bàn, không chơi cờ bạc trên mạng). Người nhà nợ tiền cờ, có nên trả hộ không, xem mục 44 ở phần 8 (nợ cờ không phải nợ chung của vợ chồng). Sau khi đã nảy ý tưởng tự sát thì làm gì, xem mục 32 của phần này (nói trước cho một người bên cạnh). Người hưởng lợi chủ yếu là bạn, khi người nhà thấy dấu hiệu thì cũng có thể kéo họ làm theo. - ### 38. Nam giới có quan hệ tình dục với nam và có hành vi nguy cơ cao: đi bệnh viện AIDS được chỉ định hỏi thuốc phòng trước phơi nhiễm, uống đúng giờ thì giảm hơn bảy phần mười số ca nhiễm - Chi phí: Đây là thuốc kê đơn, hầu hết địa phương người bệnh tự trả, giá từ vài trăm tệ tới vài nghìn tệ. Trước khi kê thuốc phải xét HIV, chức năng thận và viêm gan B. Sau đó cứ 3 tháng tái khám một lần. Chỗ khó là uống đúng giờ mỗi ngày. diff --git "a/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" "b/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" new file mode 100644 index 00000000..fb7e5fa8 --- /dev/null +++ "b/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" @@ -0,0 +1,149 @@ +[← Về mục lục](../README.md) + +# Cơ thể nhận ra thời gian ra sao, và ca đêm vì sao gây hại + +Đây là bản viết dài của mục 40 ở phần 2 (làm ca đêm càng lâu thì nguy cơ tim mạch càng cao). Mục đó trong phần chính chỉ tính số, bài này nói lý do đằng sau. + +Câu hỏi cần trả lời là: cơ thể có thật sự biết bây giờ là mấy giờ không. Nếu biết, tại sao làm ca đêm không phải chuyện "thức thêm vài lần là quen". + +## 1. Trong cơ thể có thật sự một cơ quan đi giờ + +Trong vùng dưới đối thứ não người có một cụm nhỏ tế bào thần kinh, gọi là nhân trên giao thị, khoảng hai vạn tế bào. Đó là chiếc đồng hồ chủ của cả cơ thể. + +Nó đi giờ nhờ một nhóm "gen đồng hồ" ức chế lẫn nhau. Hai protein tên CLOCK và BMAL1 mở quá trình phiên mã của hai gen khác là PER và CRY. Khi protein của PER và CRY tích tụ lên, chúng quay lại tắt cặp protein đã mở chúng. Khi mấy protein này bị phân hủy hết, vòng lặp bắt đầu lại. Chạy hết một vòng mất khoảng 24 giờ. + +Vậy chiếc đồng hồ này không dựa vào quan sát bên ngoài để tính giờ. Nhốt một người vào phòng không có ánh sáng, không có đồng hồ, cũng không có tín hiệu xã hội nào thì nhịp đó vẫn chạy tiếp. Chỉ có điều chu kỳ không đúng 24 giờ, và trôi dần từng ngày. Giải Nobel Sinh lý hoặc Y học năm 2017 được trao cho đúng ba nhà nghiên cứu đã phát hiện cơ chế phân tử này. + +## 2. Chiếc đồng hồ này lấy ánh sáng làm chuẩn, và nó đi theo con đường khác con đường nhìn thấy + +Trên võng mạc, ngoài những tế bào tạo ảnh ra, còn có một loại tế bào cảm quang riêng. Chúng chứa một sắc tố tên là melanin mắt, không tham gia việc nhìn rõ, chỉ chuyển thẳng cường độ ánh sáng sang nhân trên giao thị. + +Vậy việc ánh sáng có chỉnh được đồng hồ sinh học hay không không liên quan trực tiếp tới việc mắt có nhìn thấy hay không. Có những người mù hoàn toàn, đồng hồ sinh học vẫn chỉnh được bằng ánh sáng. + +Đó cũng là lý do mục 2 ở phần 3 (giữ giờ thức dậy cố định, cuối tuần cũng vậy) phải đi kèm với việc gặp ánh sáng buổi sáng. + +## 3. Điểm mấu chốt là chiếc đồng hồ không chỉ có một + +Gan, ruột, tim và mô mỡ, mỗi thứ có một bộ đồng hồ ngoại vi riêng. + +Đồng hồ chủ lấy ánh sáng làm chuẩn. **Đồng hồ ngoại vi thì phần lớn lấy "bạn ăn vào lúc nào" làm chuẩn.** + +Khi sinh hoạt bình thường thì hai thứ đó đồng bộ với nhau. Ban ngày bạn gặp ánh sáng, ban ngày bạn ăn, nên đồng hồ chủ và đồng hồ ngoại vi cùng chỉ một thời điểm. Vì thế bình thường bạn không hề nhận ra chúng là hai hệ thống tách rời. + +## 4. Vấn đề của ca đêm không phải là "thức khuya", mà là hai chiếc đồng hồ bị tách ra + +Người làm ca đêm ban ngày vẫn gặp ánh sáng, vẫn sống theo đời sống xã hội ban ngày. Nên đồng hồ chủ vẫn chủ yếu bị khóa vào chu kỳ ngày đêm bên ngoài. + +Nhưng bạn ăn, làm việc và điều chỉnh chuyển hóa vào đúng lúc nửa đêm theo nghĩa sinh học. Đồng hồ ngoại vi bị kéo về hướng ngược lại. + +Trạng thái này gọi là lệch nhịp ngày đêm. Vấn đề không nằm ở chỗ bạn ngủ ít đi mấy giờ, mà ở chỗ các đồng hồ trong cơ thể không còn khớp với nhau. + +## 5. Hơn nữa bạn không thích nghi được + +Có người sẽ nghĩ, làm ca đêm cố định lâu dài thì sớm muộn cũng thích nghi. + +Có một bài tổng quan gộp sáu nghiên cứu, dùng nhịp bài tiết melatonin để đo người làm ca đêm cố định lâu dài. Nhịp melatonin là chỉ số đáng tin cậy nhất để biết trạng thái đồng hồ sinh học bên trong. + +Kết quả là: chỉ có dưới 3% người đạt được mức thích nghi "hoàn toàn". Người đạt mức thích nghi "đáng kể" thì chưa tới một phần tư. + +Làm bao nhiêu năm cũng vậy. Lý do nằm ở đoạn trước: ban ngày bạn vẫn gặp ánh sáng, vẫn sống theo đời sống xã hội ban ngày, mấy tín hiệu đó vẫn kéo đồng hồ chủ về. + +Nguồn: Folkard S (2008). Do permanent night workers show circadian adjustment? A review based on the endogenous melatonin rhythm. Chronobiology International. + +## 6. Làm sao biết chính lệch nhịp thì gây hại, chứ không phải chỉ thiếu ngủ + +Đây là tầng đáng hỏi nhất của cả vấn đề. Người làm ca đêm ngủ ít, ăn uống loạn, căng thẳng nặng, hút thuốc nhiều. Bất kỳ điều nào trong số đó cũng có thể là nguyên nhân thật. + +Điều tốt là chuyện này làm riêng trong phòng thí nghiệm được. + +Nghiên cứu năm 2009 của Scheer và cộng sự: 10 người trưởng thành ở trong phòng thí nghiệm 10 ngày. Các nhà nghiên cứu đặt nhân tạo "một ngày" thành 28 giờ, buộc việc ăn và ngủ của họ phải trải qua các thời điểm khác nhau của đồng hồ sinh học. **Bốn bữa mỗi ngày, lượng calo cố định.** + +Khi hành vi lệch với đồng hồ sinh học khoảng 12 giờ, kết quả là: + +- Leptin giảm 17% (leptin là hormone khiến người ta thấy no) +- Đường huyết tăng 6%, và đó là khi insulin đã tăng 22% +- Nhịp ngày của cortisol đảo ngược hoàn toàn +- Huyết áp động mạch trung bình tăng 3% +- Hiệu suất ngủ giảm 20% + +Điểm quan trọng nhất: trong 8 người có đủ dữ liệu, có 3 người lượng đường huyết sau bữa ăn rơi vào khoảng tiền đái thái đường, và tổng cộng chỉ mất 10 ngày. + +Vì lượng calo cố định và thí nghiệm được kiểm soát trong phòng thí nghiệm, nên không thể quy những thay đổi đó cho "ngủ ít" hay "ăn nhiều". **Chỉ lệch nhịp thô đã đủ gây ra những thay đổi này.** + +Nguồn: Scheer FA, Hilton MF, Mantzoros CS, Shea SA (2009). Adverse metabolic and cardiovascular consequences of circadian misalignment. PNAS, 106(11), 4453-4458. + +## 7. Cái đầu tiên có thể đổi: chuyển bữa ăn sang ban ngày + +Vì đồng hồ ngoại vi lấy thời điểm ăn làm chuẩn nên có một hướng thử được: ca đêm thì vẫn đi làm, nhưng không ăn vào ban đêm. + +Có một thử nghiệm chia nhóm ngẫu nhiên làm đúng việc này (số đăng ký NCT02291952). 20 người khỏe mạnh mô phỏng ca đêm, được chia ngẫu nhiên thành hai nhóm. Một nhóm theo thói quen thường thấy của công nhân ca đêm, ăn cả ban đêm lẫn ban ngày. Nhóm kia chỉ ăn ban ngày. + +Báo cáo về kết cục chính nói: ở nhóm ăn ban đêm, nhịp trung ương và nhịp đường huyết ngoại vi lệch nhau, dung nạp đường xấu đi; ở nhóm chỉ ăn ban ngày thì không có cả hai. + +Bài viết về tim mạch, là phân tích thứ cấp của cùng thí nghiệm, nói: ở nhóm ăn ban đêm, chỉ số điều chỉnh trọng âm tim pNN50 giảm 25.7%, RMSSD giảm 14.3%, LF/HF phản ánh cân bằng giữa thần kinh giao cảm và thần kinh phó thần kinh tăng 5.5%, yếu tố thúc đông máu PAI-1 tăng 23.9%. Ở nhóm chỉ ăn ban ngày thì không có những thay đổi này. + +**Nhưng đây chỉ có 20 người, và là ca đêm mô phỏng, không phải công nhân ca đêm thật.** + +Một tổng quan hệ thống đi tìm riêng loại thử nghiệm chia nhóm ngẫu nhiên này, sau khi tra sáu cơ sở dữ liệu thì chỉ tìm được 4 thí nghiệm. Kết quả cũng không đều: có thí nghiệm thấy insulin lúc nhịn và sức kháng insulin cải thiện, có thí nghiệm không thấy; riêng mỡ máu tỷ trọng cao thì cả bốn thí nghiệm đều không thấy thay đổi; riêng đường huyết sau bữa ăn thì cách nhịn ăn suốt đêm 10.75 giờ quả thật tốt hơn cách ăn ban đêm rồi ăn thêm một món lúc nửa đêm. Chính tác giả tổng quan cũng nói, số thí nghiệm quá ít, không mở rộng được. + +Nên bây giờ chỉ nói được tới đây: **Khi làm ca đêm, tập trung bữa ăn vào ban ngày càng nhiều càng tốt, có sự hỗ trợ của thử nghiệm chia nhóm ngẫu nhiên, nhưng bằng chứng vẫn mỏng.** Nó không tốn tiền, cũng không khó làm, muốn thử thì thử. Còn một cách thứ hai nữa, ở phần kế tiếp: dùng ánh sáng để chuyển cả chiếc đồng hồ sinh học sang. + +Nguồn: Chellappa SL 等 (2021). Daytime eating prevents internal circadian misalignment and glucose intolerance in night work. Science Advances. ；Chellappa SL 等 (2025). Daytime eating during simulated night work mitigates changes in cardiovascular risk factors: secondary analyses of a randomized controlled trial. Nature Communications. ；Wulandari F 等 (2026). Cardiometabolic status among shift workers under meal time regulation: A systematic review of randomized controlled trials. Nutrition and Health. + +## 8. Có chuyển được chiếc đồng hồ sang bên kia không + +Những phần trước nói "lệch nhịp thì có hại". Ngược lại, chuyển cả chiếc đồng hồ sang hẳn bên ca đêm thì có được không. + +**Xét về lý thuyết thì được, và đã làm ra rồi. Nhưng yêu cầu về ánh sáng cao hơn phần lớn mọi người nghĩ một bậc.** + +Nghiên cứu năm 1990 của Czeisler và cộng sự: 8 nam trẻ đi ca đêm trong một tuần, và so sánh hai điều kiện. + +Ở điều kiện đối chứng, ban đêm dùng đèn chiếu trong nhà bình thường, khoảng 150 lux. Lux là đơn vị độ rọi, một phòng đèn huỳnh quang thường vào khoảng một hai trăm. Kết quả là sau sáu đêm liên tiếp, điểm thấp nhất của thân nhiệt vẫn rơi vào khoảng 3 giờ 30 sáng. Điểm thấp nhất của thân nhiệt là chỉ số chuẩn để biết đồng hồ sinh học đang chỉ về hướng nào. Điều này cho thấy đồng hồ sinh học không nhúc nhích. + +Ở điều kiện can thiệp, ban đêm có ánh sáng mạnh 7000 đến 12000 lux, ban ngày ngủ trong nơi gần như tối đen. Sau bốn ngày, điểm thấp nhất của thân nhiệt dời sang gần 3 giờ chiều tối. Nghĩa là cả chiếc đồng hồ đã đảo ngược. + +Nguồn: Czeisler CA, Johnson MP, Duffy JF, Brown EN, Ronda JM, Kronauer RE (1990). Exposure to bright light and darkness to treat physiologic maladaptation to night work. New England Journal of Medicine, 322(18), 1253-1259. + +### Ba cái thiếu một là không được + +**Một, ánh sáng ban đêm phải đủ sáng.** Đi làm ca mà bật đèn huỳnh quang bình thường thì y như nhóm đối chứng "không có gì xảy ra" ở trên. Muốn có tác dụng thì phải dùng hộp đèn trị liệu ánh sáng riêng. + +**Hai, ban ngày ngủ thì phải che sáng thật kỹ.** Rèm che sáng cộng với miếng che mắt. Thí nghiệm ở trên làm "ánh sáng mạnh ban đêm" và "gần như tối đen ban ngày" cùng lúc, không phải chọn một trong hai. + +**Ba, trên đường đi làm về thì phải đeo kính râm.** Cái này dễ bỏ sót nhất, mà bỏ sót thì coi như làm trắng. + +Vì sao cái thứ ba quan trọng như vậy thì phải xem cách ánh sáng tác động. Cùng một loại ánh sáng đó, rơi vào những thời điểm khác nhau thì đẩy đồng hồ sinh học theo hai hướng ngược nhau. Gặp ánh sáng **trước** điểm thấp nhất của thân nhiệt thì đồng hồ bị đẩy lùi. Gặp ánh sáng **sau** đó thì đồng hồ bị kéo tới. + +Bạn hết ca đêm đi về nhà, vừa đúng sau điểm thấp nhất của thân nhiệt. **Ánh sáng buổi sáng trên đường đi về làm ngược lại với cả một đêm ngồi trước hộp đèn.** + +### Phòng thí nghiệm làm được, đời thường thì không + +Mấy người trong phòng thí nghiệm không ra khỏi nhà suốt một tuần. Trong đời thật, ngày nghỉ của bạn vẫn phải sống như ban ngày, đón con, ở bên gia đình, chạy việc. Mỗi lần trải qua như thế là chiếc đồng hồ bị kéo về một lần. Đó chính là nguồn của con số ở phần 5: dưới 3% công nhân ca đêm cố định lâu dài thích nghi được hoàn toàn. + +Nên trong đời thật người ta thường không đặt mục tiêu đảo hẳn, mà chuyển một phần, để chiếc đồng hồ dừng ở vị trí giữa, lúc đi làm đủ tỉnh và ngày nghỉ cũng không hỏng hẳn. + +### Ở đời thật lấy được bao nhiêu + +Một phân tích tổng hợp gộp 11 nghiên cứu: liệu pháp ánh sáng giúp công nhân ca đêm có thêm khoảng 32.5 phút tổng thời gian ngủ, hiệu suất ngủ cải thiện khoảng 2.91. Liều có tác dụng là 900 đến 6000 lux, mỗi đêm ít nhất 1 giờ. + +Một nghiên cứu phòng thí nghiệm khác dùng ánh sáng mạnh theo chu kỳ. Ở nhóm can thiệp, bài tiết melatonin trùng với khung giờ ngủ đã sắp xếp trong 4.90 giờ, nhóm đối chứng chỉ 2.62 giờ (P=0.002). Sự tụt điểm ở bài kiểm tra nhận thức xuất hiện ở nhóm đối chứng vào đêm ca cuối cùng bị kìm lại ở nhóm can thiệp. Tác giả nói thẳng trong kết luận rằng cách này còn phải kiểm chứng trên công nhân ca đêm thật. + +Nguồn: Zhao C, Li N, Miao W, He Y, Lin Y (2025). A systematic review and meta-analysis on light therapy for sleep disorders in shift workers. Scientific Reports. ；Lammers-van der Holst HM 等 (2021). Efficacy of intermittent exposure to bright light for treating maladaptation to night work on a counterclockwise shift work rotation. Scandinavian Journal of Work, Environment & Health. + +### Một việc chưa ai thử + +"Chuyển đúng chiếc đồng hồ rồi nguy cơ tim mạch có giảm theo không" — **chưa có thí nghiệm nào đo.** + +Điều đã biết là lệch nhịp thì có hại, và điều đó đến từ thí nghiệm ở phần 6. Nhưng điều đó không kéo theo "chuyển đúng là an toàn". Hơn nữa ngày nghỉ lại kéo chiếc đồng hồ về, và việc cứ qua lại như vậy tốt hay xấu thì cũng chưa ai đo. + +Hai cách đối phó ở trên, một cách dừng ở chỗ chuyển hóa và chỉ số tim mạch, cách kia dừng ở chỗ giấc ngủ. Cái nào cũng chưa đi tới bệnh và tử vong. + +## 9. Những phần bài này không nói + +Ca đêm làm tăng nguy cơ tim mạch bao nhiêu, và tính theo số năm thì ra sao, ở mục 40 ở phần 2 (làm ca đêm càng lâu thì nguy cơ tim mạch càng cao). + +Sau một đêm thức khuya thì bù lại giấc ngủ thế nào, ở mục 39 ở phần 2 (thức khuya xong thì bù ngủ vào tối hôm sau, đừng dồn dập cuối tuần). + +Ban đêm nên ngủ bao lâu, và giờ giấc có đều hay không, ở mục 13 ở phần 2 (mỗi đêm ngủ khoảng 7 giờ, giữ giờ giấc cố định). + +Gặp ánh sáng buổi sáng và giữ giờ thức dậy cố định, ở mục 2 ở phần 3 (giữ giờ thức dậy cố định, cuối tuần cũng vậy). diff --git a/vi/index.html b/vi/index.html index 9538e3e5..0c2a6294 100644 --- a/vi/index.html +++ b/vi/index.html @@ -35,7 +35,7 @@ - + @@ -448,7 +448,7 @@

    Cẩm Nang Sống Hiệu Quả Cao

    Ai là người hưởng cũng chia thang: bạn cao nhất, kế đến vợ chồng và người thân trực hệ, rồi tới bạn bè và đồng nghiệp, người xa lạ thấp nhất. Thấp nhất không phải là bằng không, chỉ là kỳ vọng hoàn lại nhỏ, và phải xem cả rủi ro cùng lúc.

    Không phải làm hết: đây là danh sách dự phòng đã xếp theo hiệu quả chi phí, không phải danh sách việc phải làm. Lấy một hai mục là được, phần còn lại quay lại tra khi cần. Muốn lấy mục rẻ nhất thì bên trái chọn "Tiền bỏ ra" là "Không tốn tiền", "Ý chí phải có" là "Không cần ý chí". Tác giả cũng chưa làm phần lớn. Mọi "mục X" có gạch chấm trong nội dung đều bấm được ngay tại chỗ, không phải tự lật tìm.

    Đang hiện – đang hiện, trong tổng số – mục
    -
    Bài dài — Kết hôn có đáng không · Danh mục trang bị cứu hộ tại nhà · Có nên dừng lại khi người lạ gặp chuyện · Một nền tảng cần những giấy phép nào · Hồ sơ đối chiếu nguồn
    +
    Bài dài — Kết hôn có đáng không · Danh mục trang bị cứu hộ tại nhà · Có nên dừng lại khi người lạ gặp chuyện · Một nền tảng cần những giấy phép nào · Nhịp sinh học và ca đêm · Hồ sơ đối chiếu nguồn
    Trợ lý AI — Để Claude Code hoặc Codex trả lời câu hỏi của bạn theo cuốn sách này (cách cài skill)
    Chữ viết tắt và thuật ngữ

    Từ có gạch chân trong nội dung, bấm vào hoặc đưa chuột vào là hiện nghĩa. Toàn bộ bảng ở dưới.

    @@ -755,10 +755,19 @@

    Cẩm Nang Sống Hiệu Quả Cao

    const XREF_RE = RE_XREF; // 指路里的分隔符按正则切（英文是 ", and to –" 一串，中文是「、」） const XREF_SEP = new RegExp(SITE.xref.listSep); -// 「8 到 10」摊成 8、9、10；上限 40 是防着把法条条款号那种大范围摊成几百个键 +// 「8 到 10」摊成 8、9、10；上限 40 是防着把法条条款号那种大范围摊成几百个键。 +// 先认整串区间，再按分隔符切：英文的 listSep 里含 "to" 和 "-"，先切的话 +// "8 to 11" 被切成 "8" 和 "11"，区间再也认不出来（译文检查那边同样的坑，见 structure.mjs） const xrefKeys = (sec, spec) => { const ns = []; - for (const part of String(spec).split(XREF_SEP)){ + const whole = String(spec).trim(); + const wr = RE_XRANGE.exec(whole); + if (wr){ + const a = +wr[1], b = +wr[2]; + if (b >= a && b - a <= 40) for (let i = a; i <= b; i++) ns.push(i); + return ns.map(n => `${sec}-${n}`).filter(k => ITEMS.has(k)); + } + for (const part of whole.split(XREF_SEP)){ const r = RE_XRANGE.exec(part); if (r){ const a = +r[1], b = +r[2]; @@ -1039,9 +1048,55 @@

    Cẩm Nang Sống Hiệu Quả Cao

    } document.getElementById('tot').textContent = CARDS.length; } +/* ---------- 预渲染 hydration ---------- + 服务端（web/src/lib/render.ts）按 build() 初次渲染的形状把 DOM 写好， + 这里按文档顺序索引到 CARDS/BLOCKS/TOC 三张表里。索引完的状态和 build() 刚跑完 + 一样：card.key 都是 null，vis 都是 false，滚进来由 CARD_IO 补渲染（terms=[]， + 和服务端写的一致，重渲出来是同一个 DOM）。 + 对不上就回退：预渲染块和侧栏目录删掉，走 build() 重建。回退时 console 里留一句， + 免得渲染器悄悄跑偏了没人知道——正常构建永远不该走到这里。 */ +function indexSsgDom(sections){ + for (const s of sections){ SECS.set(s.n, s); for (const e of s.entries) ITEMS.set(`${s.n}-${e.n}`, e); } + const total = sections.reduce((n, s) => n + s.entries.length, 0); + const list = document.getElementById('list'); + const blocks = [...list.querySelectorAll(':scope > section.sec-block')]; + const preCards = list.querySelectorAll('article.card').length; + if (blocks.length !== sections.length || preCards !== total){ + console.warn(`预渲染对不上（块 ${blocks.length}/${sections.length}，卡片 ${preCards}/${total}），回退到 build() 重建`); + list.querySelectorAll(':scope > section.sec-block').forEach(b => b.remove()); + const fsec = document.getElementById('f-sec'); fsec.innerHTML = ''; + document.querySelector('#gloss dl').innerHTML = ''; + build(sections); buildGlossary(); + return; + } + const fsec = document.getElementById('f-sec'); + const links = [...fsec.querySelectorAll('a[data-go]')]; + const subs = [...fsec.querySelectorAll('.toc-sub')]; + let li = 0, si = 0; + for (const s of sections){ + for (const e of s.entries){ const a = links[li++]; if (a) TOC.set(`e-${e.sec}-${e.n}`, a); } + const sub = subs[si++]; if (sub) TOC_SEC.set(s.n, sub); + } + let bi = 0; + for (const s of sections){ + const block = blocks[bi++]; if (!block) break; + BLOCKS.push({n:s.n, el:block, k:block.querySelector('.shown .k')}); + const cards = [...block.querySelectorAll('article.card')]; + let ci = 0; + for (const e of s.entries){ + const c = cards[ci++]; if (!c) break; + const card = {e, el:c, key:null, vis:false, f:{t:c.querySelector('.t'), cost:c.querySelector('.cost'), human:c.querySelector('.human'), gain:c.querySelector('.gain'), note:c.querySelector('.note'), src:c.querySelector('.src .body'), srcN:c.querySelector('.src summary .cnt'), details:c.querySelector('.src')}}; + CARDS.push(card); CARD_OF.set(c, card); CARD_IO.observe(c); + } + } + document.getElementById('tot').textContent = CARDS.length; +} function buildGlossary(){ const box = document.getElementById('gloss'); if (!GLOSS.length) { box.hidden = true; return; } const dl = box.querySelector('dl'); + // 预渲染（Astro 构建）已经把 dt/dd 写好了：这里只索引，不重建。 + // 没有 data-ssg 的页面 dl 是空的，走原来的路，一行没变。 + if (dl.children.length) return; for (const g of [...GLOSS].sort((a,b) => a.term.localeCompare(b.term, SITE.sortLocale))){ const dt = document.createElement('dt'); dt.textContent = g.term; const dd = document.createElement('dd'); dd.textContent = g.meaning; @@ -1464,7 +1519,15 @@

    Cẩm Nang Sống Hiệu Quả Cao

    GLOSS = parseGlossary(md); PERF.log(T('perf.parse'), t1, files.length + ' ' + T('perf.files')); status.remove(); - const t2 = performance.now(); build(sections); PERF.log(T('perf.buildDom'), t2, CARDS.length + ' ' + T('perf.cards')); + const t2 = performance.now(); + // 预渲染 hydration：Astro 构建时已经把 build() 会创建的 DOM 原样写进 HTML + // （卡片、侧栏目录、术语表，见 web/src/lib/render.ts），这里只索引现有节点， + // 不重新创建。判据是 #list 上的 data-ssg 和同页的语料，两个缺一个都走老路—— + // 老构建和离线单文件没有 data-ssg，它们的行为一行没变。 + const listEl = document.getElementById('list'); + if (EMBED && listEl && listEl.dataset.ssg === '1') indexSsgDom(sections); + else build(sections); + PERF.log(T('perf.buildDom'), t2, CARDS.length + ' ' + T('perf.cards')); buildGlossary(); readUrl(); wire(); apply(); PERF.mark(T('perf.mark', {dom: document.getElementsByTagName('*').length, terms: GLOSS.length})); if (location.hash){ diff --git a/web/astro.config.mjs b/web/astro.config.mjs new file mode 100644 index 00000000..7e6306f3 --- /dev/null +++ b/web/astro.config.mjs @@ -0,0 +1,33 @@ +// Astro 配置。 +// +// 两件事和「Astro 默认行为」不一样，都是被这个仓库的既有结构逼出来的： +// +// ① 站点挂在子路径 /HowToLiveBetter/ 下，不是域名根。Pages 发的是 +// https://.github.io//，base 必须跟着，否则引用的 CSS/JS/图片全是 404。 +// +// ② 正文不放进 src/content/，就地读仓库里那几份 book/*.md。 +// tools/ 下六套构建（epub、pdf、离线单文件）和五个检查脚本 +// （check-refs、check-plain、sync-stats、i18n/check、site/build）全都按 +// book/NN-*.md 这个路径读正文，挪进 src/content/ 等于把六套一起弄断。 +// 所以内容集合用 glob loader 指回原来的目录（见 src/content.config.ts）： +// 一份内容，两边读，各读各的。 +import { defineConfig } from 'astro/config'; +import { fileURLToPath } from 'node:url'; + +// Astro 构建会把 web/src/lib 和 tools/site/build.mjs 打包， +// 打包后 import.meta.url 指到构建缓存，那两处按文件位置算仓库根就算偏了。 +// 这里先把仓库根钉死，后面它们都按这个来（直接跑 node/tsx 时不用这个变量）。 +process.env.HOWTOLIVEBETTER_ROOT ??= fileURLToPath(new URL('../', import.meta.url)); + +export default defineConfig({ + site: 'https://streamentry.github.io', + base: '/HowToLiveBetter', + // 内容在仓库根，构建在 web/；astro 只看 web/ 里的东西，正文靠内容集合读出去 + srcDir: './src', + publicDir: './public', + outDir: './dist', + build: { + format: 'directory', + }, + devToolbar: { enabled: false }, +}); diff --git a/web/package-lock.json b/web/package-lock.json new file mode 100644 index 00000000..0cf75fa9 --- /dev/null +++ b/web/package-lock.json @@ -0,0 +1,4247 @@ +{ + "name": "howtolivebetter-web", + "version": "1.0.0", + "lockfileVersion": 3, + "requires": true, + "packages": { + "": { + "name": "howtolivebetter-web", + "version": "1.0.0", + "dependencies": { + "astro": "7.3.4" + }, + "devDependencies": { + "tsx": "4.23.15" + } + }, + "node_modules/@astrojs/compiler-binding": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding/-/compiler-binding-0.4.1.tgz", + "integrity": "sha512-ZYVDW1P58OXyxqZbOMp+/2U9KA7y2esIx9r4pWdpZ/l6fCZhaw4ul+D7EgpRVxYLkyTWqzANbTTUFGPFP8zDHA==", + "license": "MIT", + "engines": { + "node": "^20.19.0 || >=22.12.0" + }, + "optionalDependencies": { + "@astrojs/compiler-binding-android-arm64": "0.4.1", + "@astrojs/compiler-binding-darwin-arm64": "0.4.1", + "@astrojs/compiler-binding-darwin-x64": "0.4.1", + "@astrojs/compiler-binding-linux-arm64-gnu": "0.4.1", + "@astrojs/compiler-binding-linux-arm64-musl": "0.4.1", + "@astrojs/compiler-binding-linux-x64-gnu": "0.4.1", + "@astrojs/compiler-binding-linux-x64-musl": "0.4.1", + "@astrojs/compiler-binding-wasm32-wasi": "0.4.1", + "@astrojs/compiler-binding-win32-arm64-msvc": "0.4.1", + "@astrojs/compiler-binding-win32-x64-msvc": "0.4.1" + } + }, + "node_modules/@astrojs/compiler-binding-android-arm64": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-android-arm64/-/compiler-binding-android-arm64-0.4.1.tgz", + "integrity": "sha512-Wtx2DZORuNTLASLLZ8hlrjFaJe1LUloatcCn9EPq7knBhRpjMGXeJANIJMeBZG4igvz7x+3nyAJM2iwCSW9xog==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "android" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@astrojs/compiler-binding-darwin-arm64": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-darwin-arm64/-/compiler-binding-darwin-arm64-0.4.1.tgz", + "integrity": "sha512-7iwXcU+hB60Y1SvOfnWHNioT72gDsDQKaDipXh2Onwp/TvQMozhbvhy13wJR8Af940a0h6PbddCeJFpj791UDA==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@astrojs/compiler-binding-darwin-x64": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-darwin-x64/-/compiler-binding-darwin-x64-0.4.1.tgz", + "integrity": "sha512-pifjICos49RXhSVKSDu5gDB0/8j0O7DmxSp1DaWW2PyLG5ekwSkcrfOZto5JqhdI7QYMZB3XOiXpsdOXpS4npA==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@astrojs/compiler-binding-linux-arm64-gnu": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-linux-arm64-gnu/-/compiler-binding-linux-arm64-gnu-0.4.1.tgz", + "integrity": "sha512-hsZMlNWc6LqWHh9LlSwTq1/XCY7pxA5rkkKIdZRV8mgbOz9OwLvjEoe5Jry13eIZjMUQO8XnQyzk8o/Pjw2YFA==", + "cpu": [ + "arm64" + ], + "libc": [ + "glibc" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@astrojs/compiler-binding-linux-arm64-musl": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-linux-arm64-musl/-/compiler-binding-linux-arm64-musl-0.4.1.tgz", + "integrity": "sha512-a3EW27f+Z9uf57vYmpLgUriA1y9wDhKz57PT8D8XlFZt2HZEWhUxKuxI1jvKVI9p8okYee11WXEm/FHY1G9yoQ==", + "cpu": [ + "arm64" + ], + "libc": [ + "musl" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@astrojs/compiler-binding-linux-x64-gnu": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-linux-x64-gnu/-/compiler-binding-linux-x64-gnu-0.4.1.tgz", + "integrity": "sha512-XruPTKB/SS4DFzeAjQ1ZyM0ieHDjapHkOSnnzRApZ4FXY2YAI7UY3mETjfJiIS/acyZyrnPH5jObv1gATTdUJQ==", + "cpu": [ + "x64" + ], + "libc": [ + "glibc" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@astrojs/compiler-binding-linux-x64-musl": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-linux-x64-musl/-/compiler-binding-linux-x64-musl-0.4.1.tgz", + "integrity": "sha512-E8j7lGanqNpudtkUbsFaOLJhmwFe4Ecqux7K9hkj4wULIPNZRM2bYZNcMSj8r1oWvaUiwZl8LaaNwv3+14jqhw==", + "cpu": [ + "x64" + ], + "libc": [ + "musl" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@astrojs/compiler-binding-wasm32-wasi": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-wasm32-wasi/-/compiler-binding-wasm32-wasi-0.4.1.tgz", + "integrity": "sha512-vNLxIv41ZE+nmPFtw24cRW5ZXG+ZPcAGtgigtjCP8lvcxKI7u2GrXoMsCgEH5fMw8WCEJpwLvsVBg15rJKAkyA==", + "cpu": [ + "wasm32" + ], + "license": "MIT", + "optional": true, + "dependencies": { + "@napi-rs/wasm-runtime": "^1.2.4" + }, + "engines": { + "node": ">=14.0.0" + } + }, + "node_modules/@astrojs/compiler-binding-win32-arm64-msvc": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-win32-arm64-msvc/-/compiler-binding-win32-arm64-msvc-0.4.1.tgz", + "integrity": "sha512-/xTDgRsFDoTh0/+KFXJ74DY4Jqn8CoSHwa8UP5RAblHfnRHkenwfSl4zYDET61O9vXKLk09UzHjjq09yP20/AQ==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@astrojs/compiler-binding-win32-x64-msvc": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-binding-win32-x64-msvc/-/compiler-binding-win32-x64-msvc-0.4.1.tgz", + "integrity": "sha512-0x5J6iHZO0Fjo/CVzoIbXKzWnGESftDbEUQ/oXQ61HhUAFoR7+T1jfToZSNAS+gYyBXHSTpMWYC7lzarbhClcg==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@astrojs/compiler-rs": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/@astrojs/compiler-rs/-/compiler-rs-0.4.1.tgz", + "integrity": "sha512-//NmAuRhy7eU9iP0ceI5Pymy55M+nUZ//kS6XeOyhBWdDib2KbJOyFaVJ52k7sDkJwTKVox7oPu3SpYiVnGm5A==", + "license": "MIT", + "dependencies": { + "@astrojs/compiler-binding": "0.4.1" + }, + "engines": { + "node": ">=22.12.0" + } + }, + "node_modules/@astrojs/internal-helpers": { + "version": "0.11.0", + "resolved": "https://registry.npmjs.org/@astrojs/internal-helpers/-/internal-helpers-0.11.0.tgz", + "integrity": "sha512-3rzxJ+xbo0+8YyqOzLziIN32wmsHdCjEVz2sGOpRxJ+Ben/KiLph4ItxBy1abEL+E8fkRzqjg0rfXmaHJGw9JA==", + "license": "MIT", + "dependencies": { + "@types/hast": "^3.0.4", + "@types/mdast": "^4.0.4", + "js-yaml": "^4.3.0", + "picomatch": "^4.0.4", + "retext-smartypants": "^6.2.0", + "shiki": "^4.0.2", + "smol-toml": "^1.6.0", + "unified": "^11.0.5" + } + }, + "node_modules/@astrojs/markdown-satteri": { + "version": "0.4.2", + "resolved": "https://registry.npmjs.org/@astrojs/markdown-satteri/-/markdown-satteri-0.4.2.tgz", + "integrity": "sha512-g4QehXnB/MzxQWnSf17NciFInMqEJy9948p8cc7zGg/r7995m0Ln5PMr3Xlnj24iey1LchaoZ3SzTOa3D4P/3w==", + "license": "MIT", + "dependencies": { + "@astrojs/internal-helpers": "0.11.0", + "@astrojs/prism": "4.0.2", + "github-slugger": "^2.0.0", + "satteri": "^0.10.3" + } + }, + "node_modules/@astrojs/prism": { + "version": "4.0.2", + "resolved": "https://registry.npmjs.org/@astrojs/prism/-/prism-4.0.2.tgz", + "integrity": "sha512-KTivpmnz6lDsC6o9H4+DNm2SrE/GHzw8cNAvEJwAvUT+eoaEnn/4NtbDNfRRaxaJHdp15gf+tfHAWiXR4wB3BA==", + "license": "MIT", + "dependencies": { + "prismjs": "^1.30.0" + }, + "engines": { + "node": ">=22.12.0" + } + }, + "node_modules/@astrojs/telemetry": { + "version": "3.3.3", + "resolved": "https://registry.npmjs.org/@astrojs/telemetry/-/telemetry-3.3.3.tgz", + "integrity": "sha512-C1TLn5sPJr0x4vk56piHWKbnqlEB8BKyte5Y45V02U+D7BGO5eMqZDH5aPjnkXQWJggvmsTXxH03QMZ9NgWLzQ==", + "license": "MIT", + "dependencies": { + "ci-info": "^4.4.0", + "dset": "^3.1.4", + "is-docker": "^4.0.0", + "package-manager-detector": "^1.6.0" + }, + "engines": { + "node": "18.20.8 || ^20.3.0 || >=22.0.0" + } + }, + "node_modules/@babel/helper-string-parser": { + "version": "7.29.7", + "resolved": "https://registry.npmjs.org/@babel/helper-string-parser/-/helper-string-parser-7.29.7.tgz", + "integrity": "sha512-Pb5ijPrZ89GDH8223L4UP8i6QApWxs04RbPQJTeWDV0/keR2E36MeKnyr6LYmUUvqRRI+Iv87SuF1W6ErINzYw==", + "license": "MIT", + "engines": { + "node": ">=6.9.0" + } + }, + "node_modules/@babel/helper-validator-identifier": { + "version": "7.29.7", + "resolved": "https://registry.npmjs.org/@babel/helper-validator-identifier/-/helper-validator-identifier-7.29.7.tgz", + "integrity": "sha512-qehxGkRj55h/ff8EMaJ+cYhyaKlHIxqYDn682wQD7RNp9UujOQsHog2uS0r2vzr4pW+sXf90NeeayjcNaX3fFg==", + "license": "MIT", + "engines": { + "node": ">=6.9.0" + } + }, + "node_modules/@babel/parser": { + "version": "7.29.9", + "resolved": "https://registry.npmjs.org/@babel/parser/-/parser-7.29.9.tgz", + "integrity": "sha512-CjXrNHTnvqBVqHgdBysY3vk2T8tpJHb5/RMeHJBTyVa9xgugCB0CJTx/3oO8RV2QRQP391RWpB7D6hLjm8V9uA==", + "license": "MIT", + "dependencies": { + "@babel/types": "^7.29.8" + }, + "bin": { + "parser": "bin/babel-parser.js" + }, + "engines": { + "node": ">=6.0.0" + } + }, + "node_modules/@babel/types": { + "version": "7.29.8", + "resolved": "https://registry.npmjs.org/@babel/types/-/types-7.29.8.tgz", + "integrity": "sha512-Vj1jF3cPfxg7OAfoI7QnVKLoILlm2JF9pnVHrX8qx7AHMiYWT+NDAA7jChlNgRS4WTLc/fD1lXLmPixluj+3Gg==", + "license": "MIT", + "dependencies": { + "@babel/helper-string-parser": "^7.29.7", + "@babel/helper-validator-identifier": "^7.29.7" + }, + "engines": { + "node": ">=6.9.0" + } + }, + "node_modules/@bruits/satteri-darwin-arm64": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/@bruits/satteri-darwin-arm64/-/satteri-darwin-arm64-0.10.5.tgz", + "integrity": "sha512-27KTVl4TJkVahMy/ohyA7qd4938G5UNneFUz/PsScYfpIhj0IVAS23mpcJXdPF44sa6nva198lmV/cKIb2YPyA==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "darwin" + ] + }, + "node_modules/@bruits/satteri-darwin-x64": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/@bruits/satteri-darwin-x64/-/satteri-darwin-x64-0.10.5.tgz", + "integrity": "sha512-IjnLe3nKspq6qaeqGgjT7MT8VrTV74yWRlaag7ZdNsI8TDAYZ0iPxMCo+9KQZHUk5EyVB+reBI/PFWL5KuFw9Q==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "darwin" + ] + }, + "node_modules/@bruits/satteri-linux-arm64-gnu": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/@bruits/satteri-linux-arm64-gnu/-/satteri-linux-arm64-gnu-0.10.5.tgz", + "integrity": "sha512-glkYXZCJywjP13v67eAyAMSJdF+ncvEbYvgi/wOtffL9tQ27lr/zsyzUfgs+ovjJ9d8JNQKiXeiArJcX8PJL9w==", + "cpu": [ + "arm64" + ], + "libc": [ + "glibc" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ] + }, + "node_modules/@bruits/satteri-linux-arm64-musl": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/@bruits/satteri-linux-arm64-musl/-/satteri-linux-arm64-musl-0.10.5.tgz", + "integrity": "sha512-yWdgG1g17Nh2QyGVlFUxGRa3FEFwiMcpZEyMNWkbM3deC94cmVc+/i9OuyFpdKuWo3GkgoCtYVOoxk1uCnCZIA==", + "cpu": [ + "arm64" + ], + "libc": [ + "musl" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ] + }, + "node_modules/@bruits/satteri-linux-x64-gnu": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/@bruits/satteri-linux-x64-gnu/-/satteri-linux-x64-gnu-0.10.5.tgz", + "integrity": "sha512-FVaLoPT1fBgGl0J+AYebyyXJYBachGl8Oyyrf1lye4RTqCB4S0Gwkj1uM9RJyThUOvx5VUmAT1CnNh1SFHA+kw==", + "cpu": [ + "x64" + ], + "libc": [ + "glibc" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ] + }, + "node_modules/@bruits/satteri-linux-x64-musl": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/@bruits/satteri-linux-x64-musl/-/satteri-linux-x64-musl-0.10.5.tgz", + "integrity": "sha512-EHpVAx2bqW3GINHTKkljtxVfQmVDGWIuwOYOP5YghTj+0PkBa2o8oKPRtQ9Kbsr1Fye8jtUcDjhwj2jMNugZKg==", + "cpu": [ + "x64" + ], + "libc": [ + "musl" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ] + }, + "node_modules/@bruits/satteri-wasm32-wasi": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/@bruits/satteri-wasm32-wasi/-/satteri-wasm32-wasi-0.10.5.tgz", + "integrity": "sha512-ypz8c/Zmipxp4IoeDa228Gstv6TLzVmNs3yC6wKCoNSOjx1iwpgzu87Y3hTkXFdwChVGU85qeUDuOIarGUZQLw==", + "cpu": [ + "wasm32" + ], + "license": "MIT", + "optional": true, + "dependencies": { + "@emnapi/core": "1.11.1", + "@emnapi/runtime": "1.11.1", + "@napi-rs/wasm-runtime": "^1.2.3" + }, + "engines": { + "node": ">=14.0.0" + } + }, + "node_modules/@bruits/satteri-wasm32-wasi/node_modules/@emnapi/core": { + "version": "1.11.1", + "resolved": "https://registry.npmjs.org/@emnapi/core/-/core-1.11.1.tgz", + "integrity": "sha512-RSvbQmHzdKzNsLYa/wHrbc3KN4sYLKAdPZxqiM2HATqv/SBk2/ENSHpvXGaLOMcsAyz0poEGqkmmKYG3OWiJEQ==", + "license": "MIT", + "optional": true, + "dependencies": { + "@emnapi/wasi-threads": "1.2.2", + "tslib": "^2.4.0" + } + }, + "node_modules/@bruits/satteri-wasm32-wasi/node_modules/@emnapi/runtime": { + "version": "1.11.1", + "resolved": "https://registry.npmjs.org/@emnapi/runtime/-/runtime-1.11.1.tgz", + "integrity": "sha512-vgj7R3y3Wgx24IQaGPA/R6YFXLHVMOZ0uVEyIQPaWs+rd1AzfEMXlAC22FYwO1XkKR6NPsq7mUandH8oIRdZFw==", + "license": "MIT", + "optional": true, + "dependencies": { + "tslib": "^2.4.0" + } + }, + "node_modules/@bruits/satteri-wasm32-wasi/node_modules/@emnapi/wasi-threads": { + "version": "1.2.2", + "resolved": "https://registry.npmjs.org/@emnapi/wasi-threads/-/wasi-threads-1.2.2.tgz", + "integrity": "sha512-c95qOXkHdydNKhscBTebqEC1CVAZpyqOfVfBzQ1qgzyl3gfeldUjIggDbIZgDKsHLgnsM+igH7TJ/eAasaVuMA==", + "license": "MIT", + "optional": true, + "dependencies": { + "tslib": "^2.4.0" + } + }, + "node_modules/@bruits/satteri-win32-arm64-msvc": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/@bruits/satteri-win32-arm64-msvc/-/satteri-win32-arm64-msvc-0.10.5.tgz", + "integrity": "sha512-siTV88nb0LRqNpkL2gXboqCwVdq95sLtzMHS1/3eONV2gLbB3NAK46wmSMvCO/yquBvI2lvaFIfd8P12ecsxBw==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "win32" + ] + }, + "node_modules/@bruits/satteri-win32-x64-msvc": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/@bruits/satteri-win32-x64-msvc/-/satteri-win32-x64-msvc-0.10.5.tgz", + "integrity": "sha512-C3IfPvfvMXmlzBxaMPKFS1XiuV9pu2mC7YqkPk7PSvTgPZ8gbdASIpHpztDLvTTQjqZ0z1Ol8tK5X+V6XXC0wQ==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "win32" + ] + }, + "node_modules/@capsizecss/unpack": { + "version": "4.0.1", + "resolved": "https://registry.npmjs.org/@capsizecss/unpack/-/unpack-4.0.1.tgz", + "integrity": "sha512-CuNiSqg7+e1cO/GjffyMOm5Tt2jUF9CWHHnvQ/UkqvtkGfHdgwEC0wpmq7fkN3gxwpRnrAN0WzO3vREKmNolMQ==", + "license": "MIT", + "dependencies": { + "fontkitten": "^1.0.3" + }, + "engines": { + "node": ">=18" + } + }, + "node_modules/@clack/core": { + "version": "1.5.1", + "resolved": "https://registry.npmjs.org/@clack/core/-/core-1.5.1.tgz", + "integrity": "sha512-iHTrHA8MtVuLl2TfZySmcKv1qO2PoyC9Z7pfSDozEuV5vtY3/wcOPKJXlqJ5Oq2Cx5DDGQGAMVx6HZfRRoVEbQ==", + "license": "MIT", + "dependencies": { + "fast-wrap-ansi": "^0.2.0", + "sisteransi": "^1.0.5" + }, + "engines": { + "node": ">= 20.12.0" + } + }, + "node_modules/@clack/prompts": { + "version": "1.8.1", + "resolved": "https://registry.npmjs.org/@clack/prompts/-/prompts-1.8.1.tgz", + "integrity": "sha512-dlT1m5e/0yUL0kRNcQn7yGLVThkgbB0Ga/1AmfDDC/8ik6AIiSf2QLQO2zPYvefsHP0aFgxO93cVLCCfDp7kzQ==", + "license": "MIT", + "dependencies": { + "@clack/core": "1.5.1", + "fast-string-width": "^3.0.2", + "fast-wrap-ansi": "^0.2.0", + "sisteransi": "^1.0.5" + }, + "engines": { + "node": ">= 20.12.0" + } + }, + "node_modules/@emnapi/core": { + "version": "1.11.3", + "resolved": "https://registry.npmjs.org/@emnapi/core/-/core-1.11.3.tgz", + "integrity": "sha512-zLpS5asjEb7lq8jYLq37N6XKaE41DIexlY1rF/z4/tIl3wo13Sqm28fRyfIsKZD+NZ8mM5RoKkpW/rBcuoSZSg==", + "license": "MIT", + "optional": true, + "peer": true, + "dependencies": { + "@emnapi/wasi-threads": "1.2.3", + "tslib": "^2.4.0" + } + }, + "node_modules/@emnapi/runtime": { + "version": "1.11.3", + "resolved": "https://registry.npmjs.org/@emnapi/runtime/-/runtime-1.11.3.tgz", + "integrity": "sha512-Xz4Tpyki7XyrpbUK1jR1AhdAdaXyhhY4lZ3neLodmhpuWfy2PAQN5B46sAiU4liOXGLkHypn/qU+jvfWSCYYLA==", + "license": "MIT", + "optional": true, + "dependencies": { + "tslib": "^2.4.0" + } + }, + "node_modules/@emnapi/wasi-threads": { + "version": "1.2.3", + "resolved": "https://registry.npmjs.org/@emnapi/wasi-threads/-/wasi-threads-1.2.3.tgz", + "integrity": "sha512-ELEBe8PsLvvJ6QMr0zLt8ffvOHW/dc1m3CEzNMg7aJUv3bMaoDtw2TXyDAwkYBuroxxuHEwhRTLJSe5sya547g==", + "license": "MIT", + "optional": true, + "peer": true, + "dependencies": { + "tslib": "^2.4.0" + } + }, + "node_modules/@esbuild/aix-ppc64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/aix-ppc64/-/aix-ppc64-0.28.2.tgz", + "integrity": "sha512-XExcO+dvLKvVtNTibSTBej1NCAbaGhWn9Ww1ZPx80qsahhPFe/8jgWP0IchNe0F3HwkU7n8ejhH8bjonqht8mQ==", + "cpu": [ + "ppc64" + ], + "license": "MIT", + "optional": true, + "os": [ + "aix" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/android-arm": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/android-arm/-/android-arm-0.28.2.tgz", + "integrity": "sha512-kXXoiPVVGQcnIYGOeaovwOURpniDBpSq4A03qkQ+BMQqtGG6HYap3xne9C1O1yo4TR3qxlCX5IqqmX6fFo2Lqg==", + "cpu": [ + "arm" + ], + "license": "MIT", + "optional": true, + "os": [ + "android" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/android-arm64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/android-arm64/-/android-arm64-0.28.2.tgz", + "integrity": "sha512-5YfKeeI8qWfBZIX+u2xZC3Zlb3Os/gLS2sbEKM+I4ZOcsWmHS2WLysCcQZDAFRslDUU5Oiq44gf6PYN1vGwG5A==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "android" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/android-x64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/android-x64/-/android-x64-0.28.2.tgz", + "integrity": "sha512-O387ite7SzUyCcy3JQX4P4bLtEA7bLLkx+esve5JHnyYfNTxcVpXZo9jhdB0lTKN44gztELTdU7nS8Nr16Fs1Q==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "android" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/darwin-arm64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/darwin-arm64/-/darwin-arm64-0.28.2.tgz", + "integrity": "sha512-n4KqkOQrraxHJcgjM1RvwbigfQKIKJVpM7xp+KsxiyUSrRdIXnt73VhrPAx0fV44hgfmIVKjxMN9J1t5jySVkw==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/darwin-x64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/darwin-x64/-/darwin-x64-0.28.2.tgz", + "integrity": "sha512-uq6suIWYP37qzGddBKPw5QEQPi6HiLGsO7UmkpfyaYNQ3D+rN6w6WfwH+nuqcGXWvawGwxOEroO4YGnFh95azw==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/freebsd-arm64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/freebsd-arm64/-/freebsd-arm64-0.28.2.tgz", + "integrity": "sha512-n+I0BTSRIoy+d6RPKnEVwql5UwBJolytvY4mAOIEJorKlqgPII8ix6slVVrfZ5Tnj7glIZvloylbB/EJPMWEXw==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "freebsd" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/freebsd-x64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/freebsd-x64/-/freebsd-x64-0.28.2.tgz", + "integrity": "sha512-78XJTJkvPs0kz2w61301PJjXl4g7q3JqiYMZ/M/yVI73EHBrCRTgkhu9oqG7vPqq+a/yadEW8aD+agKlk5xrmg==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "freebsd" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/linux-arm": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/linux-arm/-/linux-arm-0.28.2.tgz", + "integrity": "sha512-XlDnu2q5yoqems+xay6wSAcg9DDD7K9RLKZEBOMZm3ckNpJBvOX20tSfby8KfrrhINDyv9V2YVZKY/SpoGJI8w==", + "cpu": [ + "arm" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/linux-arm64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/linux-arm64/-/linux-arm64-0.28.2.tgz", + "integrity": "sha512-pW4AC0P3it8c7do9MVM4p51FzHzdM/TZrerurgRcHJ2WTa1VQ1CIq18xncfpBJw4ojkiZZrKW2yIBWBP92j6Ug==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/linux-ia32": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/linux-ia32/-/linux-ia32-0.28.2.tgz", + "integrity": "sha512-CYbnj78HsIeA+DhgUKgFCfvNsTHFhMMrinUrMZpDXJXKN8T3XViTZ/+wtHeVxEWY8ewSzTFN+nRmSwO2tZaLUQ==", + "cpu": [ + "ia32" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/linux-loong64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/linux-loong64/-/linux-loong64-0.28.2.tgz", + "integrity": "sha512-buwkd8nsph4R+ajRvw0qM5Hja/TXQow3ptzWO2EbG/cqcIkHloRrdlBtQlshyYGTNFvfkfJ5tpPLVkY4DtsPfQ==", + "cpu": [ + "loong64" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/linux-mips64el": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/linux-mips64el/-/linux-mips64el-0.28.2.tgz", + "integrity": "sha512-ZVykbDyk7519VwiNb9Lcj9m8XM6v5V9uKPvrEMkkEedVewf+0itkhahp4HDpgERXhwLRpWFypsGbG/J8s0QjJA==", + "cpu": [ + "mips64el" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/linux-ppc64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/linux-ppc64/-/linux-ppc64-0.28.2.tgz", + "integrity": "sha512-CAXl+Dtd9UUuJd8pKKdwh6MLm3MUMiqMPmhZ3tTSXPqfyQ3vDl6R5hZdZ/kYojK4ofXtdfSv1tFq8XzWx3heNQ==", + "cpu": [ + "ppc64" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/linux-riscv64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/linux-riscv64/-/linux-riscv64-0.28.2.tgz", + "integrity": "sha512-GeXCej4IQtU1B+QlDV8W/RRvbzI3O/Stss+/bCXv4lZls5WGRtu2a+3JkA3i4qIUlMXpcHebWpF8AkJhATowuA==", + "cpu": [ + "riscv64" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/linux-s390x": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/linux-s390x/-/linux-s390x-0.28.2.tgz", + "integrity": "sha512-3H1weTYZPxt/WOhByszQZybS9w5lKzUn1FDMsgEChbHWQwHYQQRfBxgCcZvPhjHfKyJjIievvMmEUawJrdY9Dg==", + "cpu": [ + "s390x" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/linux-x64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/linux-x64/-/linux-x64-0.28.2.tgz", + "integrity": "sha512-4xTZr1FUmSoQW4XIWmit3tzQrUTZM+N3P0XV8xROKYF50XfI7xeO90+1bZvNwxIufQ9hDQVRJH5YhgPVF8A/HQ==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/netbsd-arm64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/netbsd-arm64/-/netbsd-arm64-0.28.2.tgz", + "integrity": "sha512-sSATRjPeDBg3pdgHoQfoYBob11Kk1FGa9lui5RIHZCoCkJa9QKlvl3/vKz2usCmYYjs7ymJR/2Nnsqe+Hjt5nw==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "netbsd" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/netbsd-x64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/netbsd-x64/-/netbsd-x64-0.28.2.tgz", + "integrity": "sha512-lqnzCV+mM0gIADaKihiCg6ifgfU2L3h5E33rNQBN1Y4MaVGnzryzmvvf7UHxprpQdE8hpqLolJ9Rl+SkIRDpyw==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "netbsd" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/openbsd-arm64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/openbsd-arm64/-/openbsd-arm64-0.28.2.tgz", + "integrity": "sha512-AL2qJILH7lNjrDmCQDvdxMfAUIv8KMNZOvrwAQ8i8//ntL9FflhOyMJ8OZSMBb8/AWXe3/5v5S20y3zCoZWKoQ==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "openbsd" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/openbsd-x64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/openbsd-x64/-/openbsd-x64-0.28.2.tgz", + "integrity": "sha512-QtiuPytchRyC4rwUKhexJdQKvDuZ6hWloi3igqPQNUJCS1/v9EiO3UTOXR6A3FoMo4fnAKbWJdqaIwhOzh8qEw==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "openbsd" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/openharmony-arm64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/openharmony-arm64/-/openharmony-arm64-0.28.2.tgz", + "integrity": "sha512-WkhYDmpTjLvGlScA1rwjRUmhl4k8oXR3cIbtqWmELgU/dFeHHlEllxDvdWcNJV9rbzCexB5vz8gtNewWLgCT7Q==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "openharmony" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/sunos-x64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/sunos-x64/-/sunos-x64-0.28.2.tgz", + "integrity": "sha512-GPMSkTOtMnv2U2F8gxe4Io6qmVs+YKyp832Etqqxr0hFngmXQ3rzwytelm3GIn7T4VviRUlf3sOgBOiTdvaf7g==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "sunos" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/win32-arm64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/win32-arm64/-/win32-arm64-0.28.2.tgz", + "integrity": "sha512-PIhhEkE9uPBleRBrQEJpUn7MBnibZzbGzYWPmY3x+YoVg/95zbjB4CxPPOQ8l5tYYM4mMaCthF8/1DIfBQQyWQ==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/win32-ia32": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/win32-ia32/-/win32-ia32-0.28.2.tgz", + "integrity": "sha512-YmJbfTlvU7Sdn9BB+4PRES4oB6pxgS37MAONj+hBr/cpXS1aBPKXxNnDbu+QCWPj0o9dgyxeq79g6c5P8KeuYA==", + "cpu": [ + "ia32" + ], + "license": "MIT", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@esbuild/win32-x64": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/@esbuild/win32-x64/-/win32-x64-0.28.2.tgz", + "integrity": "sha512-5ebpxr3nWMzrL/rnUI755Jkuee0bHL/Gq0WTF9lvcpv73wAp5eu8MfBUgWK9bhWvZjj7yX8etf/8tI8Ney695g==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": ">=18" + } + }, + "node_modules/@img/colour": { + "version": "1.1.0", + "resolved": "https://registry.npmjs.org/@img/colour/-/colour-1.1.0.tgz", + "integrity": "sha512-Td76q7j57o/tLVdgS746cYARfSyxk8iEfRxewL9h4OMzYhbW4TAcppl0mT4eyqXddh6L/jwoM75mo7ixa/pCeQ==", + "license": "MIT", + "optional": true, + "engines": { + "node": ">=18" + } + }, + "node_modules/@img/sharp-darwin-arm64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-darwin-arm64/-/sharp-darwin-arm64-0.35.4.tgz", + "integrity": "sha512-Uhfl4V4lhP2nbUVF9+hyH1+luj86f1gUFeo8ALYxFoULoU+G87D43BfeMP8XHsk9boxAnCY/bf2EHwhA7MuGsA==", + "cpu": [ + "arm64" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-darwin-arm64": "1.3.3" + } + }, + "node_modules/@img/sharp-darwin-x64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-darwin-x64/-/sharp-darwin-x64-0.35.4.tgz", + "integrity": "sha512-hWniXY3bG5qKpkKrAwPe4y+VTPmf086YQAnkxWh7uA1YrlRouWGa0M0Mxj3ZjnXFkv7/TD1bTy9lGUK26vRvWw==", + "cpu": [ + "x64" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-darwin-x64": "1.3.3" + } + }, + "node_modules/@img/sharp-freebsd-wasm32": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-freebsd-wasm32/-/sharp-freebsd-wasm32-0.35.4.tgz", + "integrity": "sha512-lIsKw/BU+kjB4eZjxrYrZmwOJYi3Ajrv66iAlBmUPyKc3HpnloevB1g3wxGD9P/5BbQ1brBGl65VRRrCvQDEqA==", + "license": "Apache-2.0", + "optional": true, + "os": [ + "freebsd" + ], + "dependencies": { + "@img/sharp-wasm32": "0.35.4" + }, + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-darwin-arm64": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-darwin-arm64/-/sharp-libvips-darwin-arm64-1.3.3.tgz", + "integrity": "sha512-suTBPTDGrI9WodccaDdwZItTSaBYASlBk1NSfElSHrUfzu3szG6lvIF58+WiFvnfzuK8ZBFS5zE00PxqxnRiPg==", + "cpu": [ + "arm64" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "darwin" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-darwin-x64": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-darwin-x64/-/sharp-libvips-darwin-x64-1.3.3.tgz", + "integrity": "sha512-FVJZ5mITMobmXIz/hPDTw0EintTW5H3WfrxwLqEqjiIihlu+hVRyGrFQ60xl0Lxn7Bt3zdpevPaQi0HEzqz9fw==", + "cpu": [ + "x64" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "darwin" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-linux-arm": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-linux-arm/-/sharp-libvips-linux-arm-1.3.3.tgz", + "integrity": "sha512-3rbU4vqXXc3hY/OiXdl52xZvT0F1yEngWfvqudtPJg/KkyiaQw2DRsFrNzpmLvfavbwOq3qXn36GP8obHRULQA==", + "cpu": [ + "arm" + ], + "libc": [ + "glibc" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "linux" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-linux-arm64": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-linux-arm64/-/sharp-libvips-linux-arm64-1.3.3.tgz", + "integrity": "sha512-0DaL0A6Xu6sQSQFwe4iVCrKWU2cCTItnRsYsCdxAMm9NF6twAA9BKnoqy4hqz4+azQ0JHuA26qiUKsf1XJ/v5A==", + "cpu": [ + "arm64" + ], + "libc": [ + "glibc" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "linux" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-linux-ppc64": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-linux-ppc64/-/sharp-libvips-linux-ppc64-1.3.3.tgz", + "integrity": "sha512-cdn1OvUBwsXhbC0zSzJnNzf5MZ/mTrobawDvNXBTxe8VtqKAm0sRuEY2Evzovb/w9JMk4TvRxqt1mekSuJz64w==", + "cpu": [ + "ppc64" + ], + "libc": [ + "glibc" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "linux" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-linux-riscv64": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-linux-riscv64/-/sharp-libvips-linux-riscv64-1.3.3.tgz", + "integrity": "sha512-HjPVx7yKz+0lqdhDlTw1tt90wamBoxhiXpvl1XZpJLiHH4RCJ5yDTqH+VlYPv2fwFs89JFw4c1IexYOcQUi4IQ==", + "cpu": [ + "riscv64" + ], + "libc": [ + "glibc" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "linux" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-linux-s390x": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-linux-s390x/-/sharp-libvips-linux-s390x-1.3.3.tgz", + "integrity": "sha512-neWLh+3yCNThxnfy3c4BbVBeGgt9aftno+XbT56iK28RgeDs3UOFWviLWlUu0bArYVYJaFDK+RRohbicUNCm8Q==", + "cpu": [ + "s390x" + ], + "libc": [ + "glibc" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "linux" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-linux-x64": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-linux-x64/-/sharp-libvips-linux-x64-1.3.3.tgz", + "integrity": "sha512-4vKmvAst9nrowcqquKFAyZJUDolUaIp8uRiN0mWFguJ1IplC9/pitXtlnnlU4aa/eJw3J7i67V+pwUL+wZGdsA==", + "cpu": [ + "x64" + ], + "libc": [ + "glibc" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "linux" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-linuxmusl-arm64": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-linuxmusl-arm64/-/sharp-libvips-linuxmusl-arm64-1.3.3.tgz", + "integrity": "sha512-Y9kQaLMuNoB0bPYOOdcZMaseNrFpPodIWWMrx+CZyydf2xn68j9WYc6sWWRrDwNkzCQjKYfc68L7jKjGlHMibw==", + "cpu": [ + "arm64" + ], + "libc": [ + "musl" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "linux" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-libvips-linuxmusl-x64": { + "version": "1.3.3", + "resolved": "https://registry.npmjs.org/@img/sharp-libvips-linuxmusl-x64/-/sharp-libvips-linuxmusl-x64-1.3.3.tgz", + "integrity": "sha512-fj8Mv0HHfD1Rr+4I68+3agJynxDWtBFgicTbSOb9Bke6pIwzGcJ+RX/yHjmiEGFMCavY/dxvem7MyNaJF+wDiw==", + "cpu": [ + "x64" + ], + "libc": [ + "musl" + ], + "license": "LGPL-3.0-or-later", + "optional": true, + "os": [ + "linux" + ], + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-linux-arm": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-linux-arm/-/sharp-linux-arm-0.35.4.tgz", + "integrity": "sha512-7OAS8gI0EReKGVN2HssHlM6umJgxF5VI3xN0p9FA91p/YO+ou5hiNghLdZ5BEHztwaaK5+bLKRf8x/o2L2nk9A==", + "cpu": [ + "arm" + ], + "libc": [ + "glibc" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-linux-arm": "1.3.3" + } + }, + "node_modules/@img/sharp-linux-arm64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-linux-arm64/-/sharp-linux-arm64-0.35.4.tgz", + "integrity": "sha512-De4jpEnAU8Hd5oT0j1G3uL4ZvTuipVMn7YC6vPaJhy6/7EwEae0SVAoBrUMYQbkLGDm85taVWwuPc1a44LTzCQ==", + "cpu": [ + "arm64" + ], + "libc": [ + "glibc" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-linux-arm64": "1.3.3" + } + }, + "node_modules/@img/sharp-linux-ppc64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-linux-ppc64/-/sharp-linux-ppc64-0.35.4.tgz", + "integrity": "sha512-2oYZJeIl4kCcMGk4ouZVjnkCtFrpQFlNEtJ6GbxzhHQchwH0NH/qEb9ykmOl29dqwMq+JhFdZn+1ak2FKhI9fQ==", + "cpu": [ + "ppc64" + ], + "libc": [ + "glibc" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-linux-ppc64": "1.3.3" + } + }, + "node_modules/@img/sharp-linux-riscv64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-linux-riscv64/-/sharp-linux-riscv64-0.35.4.tgz", + "integrity": "sha512-cPbNChoRURAWdebDIHSenxRpgEdy7JkPydSnUxRm9VvKD7m0/xVaR/8Fzlu81pk5nHEvHH87UZUA7cTtwnbJSA==", + "cpu": [ + "riscv64" + ], + "libc": [ + "glibc" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-linux-riscv64": "1.3.3" + } + }, + "node_modules/@img/sharp-linux-s390x": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-linux-s390x/-/sharp-linux-s390x-0.35.4.tgz", + "integrity": "sha512-RY0JFY8Fd6RonCBtHz+DvadaPkXDSI1AUn6yWL9TipqkZ1vY8w8evqdgyDFnkm4/K1ve1TvZiaePP5oSd4+WVQ==", + "cpu": [ + "s390x" + ], + "libc": [ + "glibc" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-linux-s390x": "1.3.3" + } + }, + "node_modules/@img/sharp-linux-x64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-linux-x64/-/sharp-linux-x64-0.35.4.tgz", + "integrity": "sha512-9qvvEAuk8k89TfWUoX2htWjbAMX8p+NxCppjpcg5k6xMsjhBQPTsoIh36h9Qde4WRuGpJeYnOjdosDn/cnv+OA==", + "cpu": [ + "x64" + ], + "libc": [ + "glibc" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-linux-x64": "1.3.3" + } + }, + "node_modules/@img/sharp-linuxmusl-arm64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-linuxmusl-arm64/-/sharp-linuxmusl-arm64-0.35.4.tgz", + "integrity": "sha512-KB5jxpfWQTr0nc3xdHtWChdbifHrBGsd2SM62Eyxrl8afikm+f5qGBU75SJIZBT/S1MC8XyacdlXBMSWq6OURA==", + "cpu": [ + "arm64" + ], + "libc": [ + "musl" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-linuxmusl-arm64": "1.3.3" + } + }, + "node_modules/@img/sharp-linuxmusl-x64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-linuxmusl-x64/-/sharp-linuxmusl-x64-0.35.4.tgz", + "integrity": "sha512-f+eZJZIQNEEd26RPSW+76chwOf1XtA2Y/O+5ocVyLliHkeih3e+jhLVBdNTd2rS3IbNXK8+ug93Vf5ZXtF5Lxg==", + "cpu": [ + "x64" + ], + "libc": [ + "musl" + ], + "license": "Apache-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-libvips-linuxmusl-x64": "1.3.3" + } + }, + "node_modules/@img/sharp-wasm32": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-wasm32/-/sharp-wasm32-0.35.4.tgz", + "integrity": "sha512-zQnl4Kwp7Q6NHsENtU2T/00Zi+w3AQNwz3+UaTyVBy2FpXrzXzGjndpK61onhZjRtRpQXxCTeqw19bVyXOh7jA==", + "license": "Apache-2.0 AND LGPL-3.0-or-later AND MIT", + "optional": true, + "dependencies": { + "@emnapi/runtime": "^1.11.3" + }, + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-webcontainers-wasm32": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-webcontainers-wasm32/-/sharp-webcontainers-wasm32-0.35.4.tgz", + "integrity": "sha512-ESfNkywmCfPNyaZjxooddJQiQ+l/nTpGEOGthxiLnIHXC/CmcBixnfwUleX9mCz9ovrUUvKMap/pm8RYbzfwaA==", + "cpu": [ + "wasm32" + ], + "license": "Apache-2.0", + "optional": true, + "dependencies": { + "@img/sharp-wasm32": "0.35.4" + }, + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-win32-arm64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-win32-arm64/-/sharp-win32-arm64-0.35.4.tgz", + "integrity": "sha512-iNdlBX9gLVvqe2I3uIJSIKTq6wckP/DYxZtcqxm09x5Gi24DnFBmPAWZmr60ZyYMG0xlzo6goG3670ar+RXvRw==", + "cpu": [ + "arm64" + ], + "license": "Apache-2.0 AND LGPL-3.0-or-later", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-win32-ia32": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-win32-ia32/-/sharp-win32-ia32-0.35.4.tgz", + "integrity": "sha512-kqRsbaa5CS6KHlpxnN7WhE6vAAugXyZButpRdvDWetlv6Qv4N9WTcrWzF7tXfB9T7MsoadqdI8hmwLq6UlLvtw==", + "cpu": [ + "ia32" + ], + "license": "Apache-2.0 AND LGPL-3.0-or-later", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": "^20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@img/sharp-win32-x64": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/@img/sharp-win32-x64/-/sharp-win32-x64-0.35.4.tgz", + "integrity": "sha512-XtmnYhBcrORsJ4XJngyzr/EWP0hRZLAZRFaApdKuviyqF78+ylxh2y06ZmtULAMOnObJ3ucpN0AcwSWnMowTRg==", + "cpu": [ + "x64" + ], + "license": "Apache-2.0 AND LGPL-3.0-or-later", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + } + }, + "node_modules/@jridgewell/sourcemap-codec": { + "version": "1.6.0", + "resolved": "https://registry.npmjs.org/@jridgewell/sourcemap-codec/-/sourcemap-codec-1.6.0.tgz", + "integrity": "sha512-T7jf+5zgsZHwNJ4lvQ7/aezbyk0nNX+zJVWpmHA7VYsEx7a7qr5Rg5IbtJFqkgze5Y2sruq1RUY8Q837Od7iFw==", + "license": "MIT" + }, + "node_modules/@napi-rs/wasm-runtime": { + "version": "1.2.4", + "resolved": "https://registry.npmjs.org/@napi-rs/wasm-runtime/-/wasm-runtime-1.2.4.tgz", + "integrity": "sha512-AJxoUD2/15ESHbvpcyjU274nsAPLuOtPHCk0vKJM5pj//Fg/B1FXNWjPnXTT9PymCYYiHo4zPj0ZomXBKhoy7g==", + "license": "MIT", + "optional": true, + "dependencies": { + "@tybys/wasm-util": "^0.10.3" + }, + "engines": { + "node": "^20.19.0 || ^22.13.0 || >=23.5.0" + }, + "funding": { + "type": "github", + "url": "https://github.com/sponsors/Brooooooklyn" + }, + "peerDependencies": { + "@emnapi/core": "^1.7.1 || ^2.0.0-alpha.4", + "@emnapi/runtime": "^1.7.1 || ^2.0.0-alpha.4" + } + }, + "node_modules/@oslojs/encoding": { + "version": "1.1.0", + "resolved": "https://registry.npmjs.org/@oslojs/encoding/-/encoding-1.1.0.tgz", + "integrity": "sha512-70wQhgYmndg4GCPxPPxPGevRKqTIJ2Nh4OkiMWmDAVYsTQ+Ta7Sq+rPevXyXGdzr30/qZBnyOalCszoMxlyldQ==", + "license": "MIT" + }, + "node_modules/@oxc-project/types": { + "version": "0.150.0", + "resolved": "https://registry.npmjs.org/@oxc-project/types/-/types-0.150.0.tgz", + "integrity": "sha512-rDS5/31E9HfPl/CIzGrn0DOlvBbXFseQ5URJ9sYMfstbKLD/c6Gm9vmRzRGDdAXyOIL4zmO37lc9RIwYqVruZw==", + "license": "MIT", + "funding": { + "url": "https://github.com/sponsors/oxc-project" + } + }, + "node_modules/@rolldown/binding-android-arm-eabi": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-android-arm-eabi/-/binding-android-arm-eabi-1.2.9.tgz", + "integrity": "sha512-tNISae1QEf/vkb3xkRcjV5SEdzPE97We5IVaa2Z8jSszQPZ8U60B/YCYpw4QI7VidYsBtKavczXf+DyDs9WGxw==", + "cpu": [ + "arm" + ], + "license": "MIT", + "optional": true, + "os": [ + "android" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-android-arm64": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-android-arm64/-/binding-android-arm64-1.2.9.tgz", + "integrity": "sha512-YC8YsI30o606GTZi0VyzYlsDKFP8W61i/QzayHDkLbNEz/IShqAmTa+hsJRj13xTHA0H+6fk4b2UmGn+Q/cMlg==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "android" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-darwin-arm64": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-darwin-arm64/-/binding-darwin-arm64-1.2.9.tgz", + "integrity": "sha512-IwhlH3qK5urrY8hZiEgGkHKEFN901p/p2bjxCxJlr4GyNnF7wYpUvK+Y43uaRYuC4hpfjzbR3SJC3arX1jGvmw==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-darwin-x64": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-darwin-x64/-/binding-darwin-x64-1.2.9.tgz", + "integrity": "sha512-XxpJfVzFh+jilRxIXUqcfYAYcunIc/XEzIizsOL1fcJee5Sf7H3mH8WlLmfHfluz5amqR88QQo9izKtmMlavAw==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-freebsd-x64": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-freebsd-x64/-/binding-freebsd-x64-1.2.9.tgz", + "integrity": "sha512-kSfvhmgeWyfkbT3p/1s5vSgboogoah2zkm9fX2zjg2hHxSV7T4KhMWRUUaRk4OXNqoD3QAUeRqLcs1aZOK4U1g==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "freebsd" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-linux-arm-gnueabihf": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-linux-arm-gnueabihf/-/binding-linux-arm-gnueabihf-1.2.9.tgz", + "integrity": "sha512-1RVzG17pxqbTfYLC352JlLt6kKLG+6Hr30n8DlIJqsnV5luUDd2Qdx9Ayw1Cabfyb1K9k0jXEZ7evxkRoT+uiw==", + "cpu": [ + "arm" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-linux-arm64-gnu": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-linux-arm64-gnu/-/binding-linux-arm64-gnu-1.2.9.tgz", + "integrity": "sha512-BXqPvZ2drqVD+/Z8UpKwcs4Mp7grM+eGFku4CAEKrEtcbAsUpzREphK1sogCRZGreVPiMkiiBtw0n3TPteuqvw==", + "cpu": [ + "arm64" + ], + "libc": [ + "glibc" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-linux-arm64-musl": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-linux-arm64-musl/-/binding-linux-arm64-musl-1.2.9.tgz", + "integrity": "sha512-11vWvo8YDwLzukt27J3aYDWU+gg2P7J+ZOmiJ0hkF5BXZDW7pVya7r40MXDy6ya0i9KamoENSVKIugvJNgFXIA==", + "cpu": [ + "arm64" + ], + "libc": [ + "musl" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-linux-ppc64-gnu": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-linux-ppc64-gnu/-/binding-linux-ppc64-gnu-1.2.9.tgz", + "integrity": "sha512-a1tijMkdwsIARtc0F39ApURROkf3NwqinI6TOiSSWCTR7dT96dffNvMUtDHnq64wKNTIZOIlzKrFvvFUznJiyw==", + "cpu": [ + "ppc64" + ], + "libc": [ + "glibc" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-linux-s390x-gnu": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-linux-s390x-gnu/-/binding-linux-s390x-gnu-1.2.9.tgz", + "integrity": "sha512-x6SQNdAvv4c3hWqTMaWuawzMX9myaCs/yEmlGsxJzkdClnHW7FbrjQuSiRDhuSYzEYoEMhsaJy9qHG/XNemJPQ==", + "cpu": [ + "s390x" + ], + "libc": [ + "glibc" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-linux-x64-gnu": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-linux-x64-gnu/-/binding-linux-x64-gnu-1.2.9.tgz", + "integrity": "sha512-9s0AZ8BFK5/n7B/TBoa2yJE3gI3KURrbXcPBlsAsvjU4VeJKgE90y1YtNxyEUIcHPQkg6/yfF3qihUrcM/Kf0Q==", + "cpu": [ + "x64" + ], + "libc": [ + "glibc" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-linux-x64-musl": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-linux-x64-musl/-/binding-linux-x64-musl-1.2.9.tgz", + "integrity": "sha512-P7VWAmV+WdJluH7ovnRGoiv2i8To7GAZ+kGzfGup635cyL7SyYl3lSUaA3Gp5THf0n/Co5EyEqb2zbqq+nMOHQ==", + "cpu": [ + "x64" + ], + "libc": [ + "musl" + ], + "license": "MIT", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-openharmony-arm64": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-openharmony-arm64/-/binding-openharmony-arm64-1.2.9.tgz", + "integrity": "sha512-1qixtsE4BK8h+yS3BfmZ09UhA7O/N4IACva6YBr7EBvCJraByTuRcgOTaiA62Tm0vey3UcKXLOaoGHtYmNGEVg==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "openharmony" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-win32-arm64-msvc": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-win32-arm64-msvc/-/binding-win32-arm64-msvc-1.2.9.tgz", + "integrity": "sha512-ok8IQjcEPs1AKZfuEUznVBrJw+gK4soq+bx8b1X2XoMqVClarc1q5JDmVtWXY1xfr6ZuHTAsPXHTgTrqKTZeww==", + "cpu": [ + "arm64" + ], + "license": "MIT", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/binding-win32-x64-msvc": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/@rolldown/binding-win32-x64-msvc/-/binding-win32-x64-msvc-1.2.9.tgz", + "integrity": "sha512-Ip2mXoU0hM0boq3Rf+ekuT653OROSo6aSYcPT1VHE4q52KvyxgFkQgrgb/IEsxOuvQ2fZZbs8khJAyCEPM24/g==", + "cpu": [ + "x64" + ], + "license": "MIT", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/@rolldown/pluginutils": { + "version": "1.0.1", + "resolved": "https://registry.npmjs.org/@rolldown/pluginutils/-/pluginutils-1.0.1.tgz", + "integrity": "sha512-2j9bGt5Jh8hj+vPtgzPtl72j0yRxHAyumoo6TNfAjsLB04UtpSvPbPcDcBMxz7n+9CYB0c1GxQFxYRg2jimqGw==", + "license": "MIT" + }, + "node_modules/@shikijs/core": { + "version": "4.4.3", + "resolved": "https://registry.npmjs.org/@shikijs/core/-/core-4.4.3.tgz", + "integrity": "sha512-QCR4q2ZO/ILJEuwiBMel4wdcTDb1JGwfjKTxPDF6x8ixOaluPrVqIn06C99AcRPhmYlBR56d/Fb+GN58GzExpg==", + "license": "MIT", + "dependencies": { + "@shikijs/primitive": "4.4.3", + "@shikijs/types": "4.4.3", + "@shikijs/vscode-textmate": "^10.0.2", + "@types/hast": "^3.0.5", + "hast-util-to-html": "^9.0.5" + }, + "engines": { + "node": ">=20" + } + }, + "node_modules/@shikijs/engine-javascript": { + "version": "4.4.3", + "resolved": "https://registry.npmjs.org/@shikijs/engine-javascript/-/engine-javascript-4.4.3.tgz", + "integrity": "sha512-FbOjFJp9VLdo1Wevs10BBtVxiTWwNLqZh5Gkhjgda/ioL15YOgeSl9n+6XMa3qRlPQzfhFNe641SrynFHYG0nQ==", + "license": "MIT", + "dependencies": { + "@shikijs/types": "4.4.3", + "@shikijs/vscode-textmate": "^10.0.2", + "oniguruma-to-es": "^4.3.6" + }, + "engines": { + "node": ">=20" + } + }, + "node_modules/@shikijs/engine-oniguruma": { + "version": "4.4.3", + "resolved": "https://registry.npmjs.org/@shikijs/engine-oniguruma/-/engine-oniguruma-4.4.3.tgz", + "integrity": "sha512-EcOQkxdxGQrc1Row/cC2c96/v1dbZqGnEVu1qTuT/MJmp6+cXCvQussowVmCv5Tqr3KuY3c7IbM6HTW3LJ1k9w==", + "license": "MIT", + "dependencies": { + "@shikijs/types": "4.4.3", + "@shikijs/vscode-textmate": "^10.0.2" + }, + "engines": { + "node": ">=20" + } + }, + "node_modules/@shikijs/langs": { + "version": "4.4.3", + "resolved": "https://registry.npmjs.org/@shikijs/langs/-/langs-4.4.3.tgz", + "integrity": "sha512-ePic0yfAJGOF83D5wBHK/00EjK65oahBYxFk5epgq33WRv7X9UuxLEV8PtR0szC0z8dl7INIpIodB99JRFlR+A==", + "license": "MIT", + "dependencies": { + "@shikijs/types": "4.4.3" + }, + "engines": { + "node": ">=20" + } + }, + "node_modules/@shikijs/primitive": { + "version": "4.4.3", + "resolved": "https://registry.npmjs.org/@shikijs/primitive/-/primitive-4.4.3.tgz", + "integrity": "sha512-m0wBeLDQDeIxRdUmrCPdQqfuUamDwRL5isCfYbguKD6NiaKpVbsv+3J81DyIKgNW5h4WAIIr8T4EkgQrBBxvaQ==", + "license": "MIT", + "dependencies": { + "@shikijs/types": "4.4.3", + "@shikijs/vscode-textmate": "^10.0.2", + "@types/hast": "^3.0.5" + }, + "engines": { + "node": ">=20" + } + }, + "node_modules/@shikijs/themes": { + "version": "4.4.3", + "resolved": "https://registry.npmjs.org/@shikijs/themes/-/themes-4.4.3.tgz", + "integrity": "sha512-w8UHjeUnIR965KMWJHUPXOc2mNJUnK3vpVLYLvw5IYU2mnTTJ89E24OrJDBNiJDQ0qzb0tc4l7mrIXx5cFeIyw==", + "license": "MIT", + "dependencies": { + "@shikijs/types": "4.4.3" + }, + "engines": { + "node": ">=20" + } + }, + "node_modules/@shikijs/types": { + "version": "4.4.3", + "resolved": "https://registry.npmjs.org/@shikijs/types/-/types-4.4.3.tgz", + "integrity": "sha512-UEJxmRR++MAGR6hugn0vgVS2W/6lWAts84FFSrnlH9sP0LNol7E5+NQ792pH8liWUhyMyjhTgSUH3k7iD7tc5g==", + "license": "MIT", + "dependencies": { + "@shikijs/vscode-textmate": "^10.0.2", + "@types/hast": "^3.0.5" + }, + "engines": { + "node": ">=20" + } + }, + "node_modules/@shikijs/vscode-textmate": { + "version": "10.0.2", + "resolved": "https://registry.npmjs.org/@shikijs/vscode-textmate/-/vscode-textmate-10.0.2.tgz", + "integrity": "sha512-83yeghZ2xxin3Nj8z1NMd/NCuca+gsYXswywDy5bHvwlWL8tpTQmzGeUuHd9FC3E/SBEMvzJRwWEOz5gGes9Qg==", + "license": "MIT" + }, + "node_modules/@tybys/wasm-util": { + "version": "0.10.4", + "resolved": "https://registry.npmjs.org/@tybys/wasm-util/-/wasm-util-0.10.4.tgz", + "integrity": "sha512-W3c4gRigFS0T/Ma4qIYF3GDAc5AQdHb1yL5znJT1Zv1YaD9Kitx656wBjvr19qbiosmZT8lWDM5BEMynUqX65A==", + "license": "MIT", + "optional": true, + "dependencies": { + "tslib": "^2.4.0" + } + }, + "node_modules/@types/estree": { + "version": "1.0.9", + "resolved": "https://registry.npmjs.org/@types/estree/-/estree-1.0.9.tgz", + "integrity": "sha512-GhdPgy1el4/ImP05X05Uw4cw2/M93BCUmnEvWZNStlCzEKME4Fkk+YpoA5OiHNQmoS7Cafb8Xa3Pya8m1Qrzeg==", + "license": "MIT" + }, + "node_modules/@types/estree-jsx": { + "version": "1.0.5", + "resolved": "https://registry.npmjs.org/@types/estree-jsx/-/estree-jsx-1.0.5.tgz", + "integrity": "sha512-52CcUVNFyfb1A2ALocQw/Dd1BQFNmSdkuC3BkZ6iqhdMfQz7JWOFRuJFloOzjk+6WijU56m9oKXFAXc7o3Towg==", + "license": "MIT", + "dependencies": { + "@types/estree": "*" + } + }, + "node_modules/@types/hast": { + "version": "3.0.5", + "resolved": "https://registry.npmjs.org/@types/hast/-/hast-3.0.5.tgz", + "integrity": "sha512-rp/ezSWaD1m44dPKICGhiskI13nVr7qTloFwDa/IYkhhf5nzwP+zIQcIJh3WIFSBOy/H1PzB40jPjMDksN4F+g==", + "license": "MIT", + "dependencies": { + "@types/unist": "*" + } + }, + "node_modules/@types/mdast": { + "version": "4.0.4", + "resolved": "https://registry.npmjs.org/@types/mdast/-/mdast-4.0.4.tgz", + "integrity": "sha512-kGaNbPh1k7AFzgpud/gMdvIm5xuECykRR+JnWKQno9TAXVa6WIVCGTPvYGekIDL4uwCZQSYbUxNBSb1aUo79oA==", + "license": "MIT", + "dependencies": { + "@types/unist": "*" + } + }, + "node_modules/@types/nlcst": { + "version": "2.0.3", + "resolved": "https://registry.npmjs.org/@types/nlcst/-/nlcst-2.0.3.tgz", + "integrity": "sha512-vSYNSDe6Ix3q+6Z7ri9lyWqgGhJTmzRjZRqyq15N0Z/1/UnVsno9G/N40NBijoYx2seFDIl0+B2mgAb9mezUCA==", + "license": "MIT", + "dependencies": { + "@types/unist": "*" + } + }, + "node_modules/@types/unist": { + "version": "3.0.3", + "resolved": "https://registry.npmjs.org/@types/unist/-/unist-3.0.3.tgz", + "integrity": "sha512-ko/gIFJRv177XgZsZcBwnqJN5x/Gien8qNOn0D5bQU/zAzVf9Zt3BlcUiLqhV9y4ARk0GbT3tnUiPNgnTXzc/Q==", + "license": "MIT" + }, + "node_modules/@ungap/structured-clone": { + "version": "1.4.0", + "resolved": "https://registry.npmjs.org/@ungap/structured-clone/-/structured-clone-1.4.0.tgz", + "integrity": "sha512-1mEZtMKPM09vDmQt5y7YvmN2+DFTP7Tg0EWXdic8/C6VRnpb33e4ghisCIE3WZjsE2N8mf+QV1Zqh7ZFYLWInQ==", + "license": "ISC" + }, + "node_modules/am-i-vibing": { + "version": "0.4.0", + "resolved": "https://registry.npmjs.org/am-i-vibing/-/am-i-vibing-0.4.0.tgz", + "integrity": "sha512-MxT4XZL7pzLHpuvhDKdMaQHMGGkJDLluKBLsbstn+8wv9sWcFT6h+0ve9qkml95amVTZtZV83gQe2hY+ojgHLg==", + "license": "MIT", + "dependencies": { + "process-ancestry": "^0.1.0" + }, + "bin": { + "am-i-vibing": "dist/cli.mjs" + } + }, + "node_modules/anymatch": { + "version": "3.1.3", + "resolved": "https://registry.npmjs.org/anymatch/-/anymatch-3.1.3.tgz", + "integrity": "sha512-KMReFUr0B4t+D+OBkjR3KYqvocp2XaSzO55UcB6mgQMd3KbcE+mWTyvVV7D/zsdEbNnV6acZUutkiHQXvTr1Rw==", + "license": "ISC", + "dependencies": { + "normalize-path": "^3.0.0", + "picomatch": "^2.0.4" + }, + "engines": { + "node": ">= 8" + } + }, + "node_modules/anymatch/node_modules/picomatch": { + "version": "2.3.2", + "resolved": "https://registry.npmjs.org/picomatch/-/picomatch-2.3.2.tgz", + "integrity": "sha512-V7+vQEJ06Z+c5tSye8S+nHUfI51xoXIXjHQ99cQtKUkQqqO1kO/KCJUfZXuB47h/YBlDhah2H3hdUGXn8ie0oA==", + "license": "MIT", + "engines": { + "node": ">=8.6" + }, + "funding": { + "url": "https://github.com/sponsors/jonschlinkert" + } + }, + "node_modules/argparse": { + "version": "2.0.1", + "resolved": "https://registry.npmjs.org/argparse/-/argparse-2.0.1.tgz", + "integrity": "sha512-8+9WqebbFzpX9OR+Wa6O29asIogeRMzcGtAINdpMHHyAg10f05aSFVBbcEqGf/PXw1EjAZ+q2/bEBg3DvurK3Q==", + "license": "Python-2.0" + }, + "node_modules/aria-query": { + "version": "5.3.2", + "resolved": "https://registry.npmjs.org/aria-query/-/aria-query-5.3.2.tgz", + "integrity": "sha512-COROpnaoap1E2F000S62r6A60uHZnmlvomhfyT2DlTcrY1OrBKn2UhH7qn5wTC9zMvD0AY7csdPSNwKP+7WiQw==", + "license": "Apache-2.0", + "engines": { + "node": ">= 0.4" + } + }, + "node_modules/astro": { + "version": "7.3.4", + "resolved": "https://registry.npmjs.org/astro/-/astro-7.3.4.tgz", + "integrity": "sha512-9iZqfREj15mh5xYnKhjP8arO4F3Vcddri1rbfT3FseNElWHU+pl9vJY5ypKT1FhB/zyuGM/B9naSjeVQl9b5vw==", + "license": "MIT", + "dependencies": { + "@astrojs/compiler-rs": "^0.4.0", + "@astrojs/internal-helpers": "0.11.0", + "@astrojs/markdown-satteri": "0.4.2", + "@astrojs/telemetry": "3.3.3", + "@capsizecss/unpack": "^4.0.0", + "@clack/prompts": "^1.1.0", + "@oslojs/encoding": "^1.1.0", + "am-i-vibing": "^0.4.0", + "aria-query": "^5.3.2", + "axobject-query": "^4.1.0", + "ci-info": "^4.4.0", + "clsx": "^2.1.1", + "common-ancestor-path": "^2.0.0", + "cookie": "^2.0.1", + "devalue": "^5.8.1", + "diff": "^9.0.0", + "dset": "^3.1.4", + "es-module-lexer": "^2.0.0", + "esbuild": "^0.28.0", + "find-proc": "0.2.0", + "flattie": "^1.1.1", + "fontace": "~0.4.1", + "get-tsconfig": "5.0.0-beta.4", + "github-slugger": "^2.0.0", + "html-escaper": "3.0.3", + "http-cache-semantics": "^4.2.0", + "js-yaml": "^4.3.2", + "jsonc-parser": "^3.3.1", + "magic-string": "^1.0.0", + "magicast": "^0.5.2", + "mrmime": "^2.0.1", + "neotraverse": "^1.0.1", + "obug": "^3.0.0", + "p-limit": "^7.3.0", + "p-queue": "^9.1.0", + "package-manager-detector": "^1.6.0", + "piccolore": "^0.1.3", + "picomatch": "^4.0.4", + "shiki": "^4.0.2", + "smol-toml": "^1.8.0", + "svgo": "^4.1.0", + "tinyclip": "^1.0.1", + "tinyexec": "^1.0.4", + "tinyglobby": "^0.2.15", + "ultrahtml": "^1.6.0", + "unifont": "~0.7.5", + "unstorage": "^1.17.5", + "verkit": "^0.4.0", + "vite": "^8.0.13", + "vitefu": "^1.1.2", + "xxhash-wasm": "^1.1.0", + "yargs-parser": "^22.0.0", + "zod": "^4.5.4" + }, + "bin": { + "astro": "bin/astro.mjs" + }, + "engines": { + "node": ">=22.12.0", + "npm": ">=9.6.5", + "pnpm": ">=7.1.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/astrodotbuild" + }, + "optionalDependencies": { + "sharp": "^0.35.4" + }, + "peerDependencies": { + "@astrojs/markdown-remark": "^7.3.0" + }, + "peerDependenciesMeta": { + "@astrojs/markdown-remark": { + "optional": true + } + } + }, + "node_modules/axobject-query": { + "version": "4.1.0", + "resolved": "https://registry.npmjs.org/axobject-query/-/axobject-query-4.1.0.tgz", + "integrity": "sha512-qIj0G9wZbMGNLjLmg1PT6v2mE9AH2zlnADJD/2tC6E00hgmhUOfEB6greHPAfLRSufHqROIUTkw6E+M3lH0PTQ==", + "license": "Apache-2.0", + "engines": { + "node": ">= 0.4" + } + }, + "node_modules/bail": { + "version": "2.0.2", + "resolved": "https://registry.npmjs.org/bail/-/bail-2.0.2.tgz", + "integrity": "sha512-0xO6mYd7JB2YesxDKplafRpsiOzPt9V02ddPCLbY1xYGPOX24NTyN50qnUxgCPcSoYMhKpAuBTjQoRZCAkUDRw==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/boolbase": { + "version": "1.0.0", + "resolved": "https://registry.npmjs.org/boolbase/-/boolbase-1.0.0.tgz", + "integrity": "sha512-JZOSA7Mo9sNGB8+UjSgzdLtokWAky1zbztM3WRLCbZ70/3cTANmQmOdR7y2g+J0e2WXywy1yS468tY+IruqEww==", + "license": "ISC" + }, + "node_modules/ccount": { + "version": "2.0.1", + "resolved": "https://registry.npmjs.org/ccount/-/ccount-2.0.1.tgz", + "integrity": "sha512-eyrF0jiFpY+3drT6383f1qhkbGsLSifNAjA61IUjZjmLCWjItY6LB9ft9YhoDgwfmclB2zhu51Lc7+95b8NRAg==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/character-entities-html4": { + "version": "2.1.0", + "resolved": "https://registry.npmjs.org/character-entities-html4/-/character-entities-html4-2.1.0.tgz", + "integrity": "sha512-1v7fgQRj6hnSwFpq1Eu0ynr/CDEw0rXo2B61qXrLNdHZmPKgb7fqS1a2JwF0rISo9q77jDI8VMEHoApn8qDoZA==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/character-entities-legacy": { + "version": "3.0.0", + "resolved": "https://registry.npmjs.org/character-entities-legacy/-/character-entities-legacy-3.0.0.tgz", + "integrity": "sha512-RpPp0asT/6ufRm//AJVwpViZbGM/MkjQFxJccQRHmISF/22NBtsHqAWmL+/pmkPWoIUJdWyeVleTl1wydHATVQ==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/chokidar": { + "version": "5.0.0", + "resolved": "https://registry.npmjs.org/chokidar/-/chokidar-5.0.0.tgz", + "integrity": "sha512-TQMmc3w+5AxjpL8iIiwebF73dRDF4fBIieAqGn9RGCWaEVwQ6Fb2cGe31Yns0RRIzii5goJ1Y7xbMwo1TxMplw==", + "license": "MIT", + "dependencies": { + "readdirp": "^5.0.0" + }, + "engines": { + "node": ">= 20.19.0" + }, + "funding": { + "url": "https://paulmillr.com/funding/" + } + }, + "node_modules/ci-info": { + "version": "4.4.0", + "resolved": "https://registry.npmjs.org/ci-info/-/ci-info-4.4.0.tgz", + "integrity": "sha512-77PSwercCZU2Fc4sX94eF8k8Pxte6JAwL4/ICZLFjJLqegs7kCuAsqqj/70NQF6TvDpgFjkubQB2FW2ZZddvQg==", + "funding": [ + { + "type": "github", + "url": "https://github.com/sponsors/sibiraj-s" + } + ], + "license": "MIT", + "engines": { + "node": ">=8" + } + }, + "node_modules/clsx": { + "version": "2.1.1", + "resolved": "https://registry.npmjs.org/clsx/-/clsx-2.1.1.tgz", + "integrity": "sha512-eYm0QWBtUrBWZWG0d386OGAw16Z995PiOVo2B7bjWSbHedGl5e0ZWaq65kOGgUSNesEIDkB9ISbTg/JK9dhCZA==", + "license": "MIT", + "engines": { + "node": ">=6" + } + }, + "node_modules/comma-separated-tokens": { + "version": "2.0.3", + "resolved": "https://registry.npmjs.org/comma-separated-tokens/-/comma-separated-tokens-2.0.3.tgz", + "integrity": "sha512-Fu4hJdvzeylCfQPp9SGWidpzrMs7tTrlu6Vb8XGaRGck8QSNZJJp538Wrb60Lax4fPwR64ViY468OIUTbRlGZg==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/commander": { + "version": "11.1.0", + "resolved": "https://registry.npmjs.org/commander/-/commander-11.1.0.tgz", + "integrity": "sha512-yPVavfyCcRhmorC7rWlkHn15b4wDVgVmBA7kV4QVBsF7kv/9TKJAbAXVTxvTnwP8HHKjRCJDClKbciiYS7p0DQ==", + "license": "MIT", + "engines": { + "node": ">=16" + } + }, + "node_modules/common-ancestor-path": { + "version": "2.0.0", + "resolved": "https://registry.npmjs.org/common-ancestor-path/-/common-ancestor-path-2.0.0.tgz", + "integrity": "sha512-dnN3ibLeoRf2HNC+OlCiNc5d2zxbLJXOtiZUudNFSXZrNSydxcCsSpRzXwfu7BBWCIfHPw+xTayeBvJCP/D8Ng==", + "license": "BlueOak-1.0.0", + "engines": { + "node": ">= 18" + } + }, + "node_modules/cookie": { + "version": "2.0.1", + "resolved": "https://registry.npmjs.org/cookie/-/cookie-2.0.1.tgz", + "integrity": "sha512-yuToqVvRrj6pfDXREyQAAv8SkAEk/8GS3jQRTiUMm66TVtBYmqQeoEjL2Lmq8Rpo6271vH76InTChTitEAm65w==", + "license": "MIT", + "engines": { + "node": ">=22" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/express" + } + }, + "node_modules/cookie-es": { + "version": "1.2.3", + "resolved": "https://registry.npmjs.org/cookie-es/-/cookie-es-1.2.3.tgz", + "integrity": "sha512-lXVyvUvrNXblMqzIRrxHb57UUVmqsSWlxqt3XIjCkUP0wDAf6uicO6KMbEgYrMNtEvWgWHwe42CKxPu9MYAnWw==", + "license": "MIT" + }, + "node_modules/crossws": { + "version": "0.3.5", + "resolved": "https://registry.npmjs.org/crossws/-/crossws-0.3.5.tgz", + "integrity": "sha512-ojKiDvcmByhwa8YYqbQI/hg7MEU0NC03+pSdEq4ZUnZR9xXpwk7E43SMNGkn+JxJGPFtNvQ48+vV2p+P1ml5PA==", + "license": "MIT", + "dependencies": { + "uncrypto": "^0.1.3" + } + }, + "node_modules/css-select": { + "version": "6.0.0", + "resolved": "https://registry.npmjs.org/css-select/-/css-select-6.0.0.tgz", + "integrity": "sha512-rZZVSLle8v0+EY8QAkDWrKhpgt6SA5OtHsgBnsj6ZaLb5dmDVOWUDtQitd9ydxxvEjhewNudS6eTVU7uOyzvXw==", + "license": "BSD-2-Clause", + "dependencies": { + "boolbase": "^1.0.0", + "css-what": "^7.0.0", + "domhandler": "^5.0.3", + "domutils": "^3.2.2", + "nth-check": "^2.1.1" + }, + "funding": { + "url": "https://github.com/sponsors/fb55" + } + }, + "node_modules/css-tree": { + "version": "3.2.1", + "resolved": "https://registry.npmjs.org/css-tree/-/css-tree-3.2.1.tgz", + "integrity": "sha512-X7sjQzceUhu1u7Y/ylrRZFU2FS6LRiFVp6rKLPg23y3x3c3DOKAwuXGDp+PAGjh6CSnCjYeAul8pcT8bAl+lSA==", + "license": "MIT", + "dependencies": { + "mdn-data": "2.27.1", + "source-map-js": "^1.2.1" + }, + "engines": { + "node": "^10 || ^12.20.0 || ^14.13.0 || >=15.0.0" + } + }, + "node_modules/css-what": { + "version": "7.0.0", + "resolved": "https://registry.npmjs.org/css-what/-/css-what-7.0.0.tgz", + "integrity": "sha512-wD5oz5xibMOPHzy13CyGmogB3phdvcDaB5t0W/Nr5Z2O/agcB8YwOz6e2Lsp10pNDzBoDO9nVa3RGs/2BttpHQ==", + "license": "BSD-2-Clause", + "engines": { + "node": ">= 6" + }, + "funding": { + "url": "https://github.com/sponsors/fb55" + } + }, + "node_modules/csso": { + "version": "5.0.5", + "resolved": "https://registry.npmjs.org/csso/-/csso-5.0.5.tgz", + "integrity": "sha512-0LrrStPOdJj+SPCCrGhzryycLjwcgUSHBtxNA8aIDxf0GLsRh1cKYhB00Gd1lDOS4yGH69+SNn13+TWbVHETFQ==", + "license": "MIT", + "dependencies": { + "css-tree": "~2.2.0" + }, + "engines": { + "node": "^10 || ^12.20.0 || ^14.13.0 || >=15.0.0", + "npm": ">=7.0.0" + } + }, + "node_modules/csso/node_modules/css-tree": { + "version": "2.2.1", + "resolved": "https://registry.npmjs.org/css-tree/-/css-tree-2.2.1.tgz", + "integrity": "sha512-OA0mILzGc1kCOCSJerOeqDxDQ4HOh+G8NbOJFOTgOCzpw7fCBubk0fEyxp8AgOL/jvLgYA/uV0cMbe43ElF1JA==", + "license": "MIT", + "dependencies": { + "mdn-data": "2.0.28", + "source-map-js": "^1.0.1" + }, + "engines": { + "node": "^10 || ^12.20.0 || ^14.13.0 || >=15.0.0", + "npm": ">=7.0.0" + } + }, + "node_modules/csso/node_modules/mdn-data": { + "version": "2.0.28", + "resolved": "https://registry.npmjs.org/mdn-data/-/mdn-data-2.0.28.tgz", + "integrity": "sha512-aylIc7Z9y4yzHYAJNuESG3hfhC+0Ibp/MAMiaOZgNv4pmEdFyfZhhhny4MNiAfWdBQ1RQ2mfDWmM1x8SvGyp8g==", + "license": "CC0-1.0" + }, + "node_modules/defu": { + "version": "6.1.7", + "resolved": "https://registry.npmjs.org/defu/-/defu-6.1.7.tgz", + "integrity": "sha512-7z22QmUWiQ/2d0KkdYmANbRUVABpZ9SNYyH5vx6PZ+nE5bcC0l7uFvEfHlyld/HcGBFTL536ClDt3DEcSlEJAQ==", + "license": "MIT" + }, + "node_modules/dequal": { + "version": "2.0.3", + "resolved": "https://registry.npmjs.org/dequal/-/dequal-2.0.3.tgz", + "integrity": "sha512-0je+qPKHEMohvfRTCEo3CrPG6cAzAYgmzKyxRiYSSDkS6eGJdyVJm7WaYA5ECaAD9wLB2T4EEeymA5aFVcYXCA==", + "license": "MIT", + "engines": { + "node": ">=6" + } + }, + "node_modules/destr": { + "version": "2.0.5", + "resolved": "https://registry.npmjs.org/destr/-/destr-2.0.5.tgz", + "integrity": "sha512-ugFTXCtDZunbzasqBxrK93Ik/DRYsO6S/fedkWEMKqt04xZ4csmnmwGDBAb07QWNaGMAmnTIemsYZCksjATwsA==", + "license": "MIT" + }, + "node_modules/detect-libc": { + "version": "2.1.2", + "resolved": "https://registry.npmjs.org/detect-libc/-/detect-libc-2.1.2.tgz", + "integrity": "sha512-Btj2BOOO83o3WyH59e8MgXsxEQVcarkUOpEYrubB0urwnN10yQ364rsiByU11nZlqWYZm05i/of7io4mzihBtQ==", + "license": "Apache-2.0", + "engines": { + "node": ">=8" + } + }, + "node_modules/devalue": { + "version": "5.9.4", + "resolved": "https://registry.npmjs.org/devalue/-/devalue-5.9.4.tgz", + "integrity": "sha512-sPAT4pztbu6586/hrhOnMKS17IJrvg12mXiSPSS3W5qDeN2RGgvZ0diZCm31dBbnevfVmujNO3IM2wrS4Y2Rhg==", + "license": "MIT" + }, + "node_modules/devlop": { + "version": "1.1.0", + "resolved": "https://registry.npmjs.org/devlop/-/devlop-1.1.0.tgz", + "integrity": "sha512-RWmIqhcFf1lRYBvNmr7qTNuyCt/7/ns2jbpp1+PalgE/rDQcBT0fioSMUpJ93irlUhC5hrg4cYqe6U+0ImW0rA==", + "license": "MIT", + "dependencies": { + "dequal": "^2.0.0" + }, + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/diff": { + "version": "9.0.0", + "resolved": "https://registry.npmjs.org/diff/-/diff-9.0.0.tgz", + "integrity": "sha512-svtcdpS8CgJyqAjEQIXdb3OjhFVVYjzGAPO8WGCmRbrml64SPw/jJD4GoE98aR7r25A0XcgrK3F02yw9R/vhQw==", + "license": "BSD-3-Clause", + "engines": { + "node": ">=0.3.1" + } + }, + "node_modules/dom-serializer": { + "version": "2.0.0", + "resolved": "https://registry.npmjs.org/dom-serializer/-/dom-serializer-2.0.0.tgz", + "integrity": "sha512-wIkAryiqt/nV5EQKqQpo3SToSOV9J0DnbJqwK7Wv/Trc92zIAYZ4FlMu+JPFW1DfGFt81ZTCGgDEabffXeLyJg==", + "license": "MIT", + "dependencies": { + "domelementtype": "^2.3.0", + "domhandler": "^5.0.2", + "entities": "^4.2.0" + }, + "funding": { + "url": "https://github.com/cheeriojs/dom-serializer?sponsor=1" + } + }, + "node_modules/domelementtype": { + "version": "2.3.0", + "resolved": "https://registry.npmjs.org/domelementtype/-/domelementtype-2.3.0.tgz", + "integrity": "sha512-OLETBj6w0OsagBwdXnPdN0cnMfF9opN69co+7ZrbfPGrdpPVNBUj02spi6B1N7wChLQiPn4CSH/zJvXw56gmHw==", + "funding": [ + { + "type": "github", + "url": "https://github.com/sponsors/fb55" + } + ], + "license": "BSD-2-Clause" + }, + "node_modules/domhandler": { + "version": "5.0.3", + "resolved": "https://registry.npmjs.org/domhandler/-/domhandler-5.0.3.tgz", + "integrity": "sha512-cgwlv/1iFQiFnU96XXgROh8xTeetsnJiDsTc7TYCLFd9+/WNkIqPTxiM/8pSd8VIrhXGTf1Ny1q1hquVqDJB5w==", + "license": "BSD-2-Clause", + "dependencies": { + "domelementtype": "^2.3.0" + }, + "engines": { + "node": ">= 4" + }, + "funding": { + "url": "https://github.com/fb55/domhandler?sponsor=1" + } + }, + "node_modules/domutils": { + "version": "3.2.2", + "resolved": "https://registry.npmjs.org/domutils/-/domutils-3.2.2.tgz", + "integrity": "sha512-6kZKyUajlDuqlHKVX1w7gyslj9MPIXzIFiz/rGu35uC1wMi+kMhQwGhl4lt9unC9Vb9INnY9Z3/ZA3+FhASLaw==", + "license": "BSD-2-Clause", + "dependencies": { + "dom-serializer": "^2.0.0", + "domelementtype": "^2.3.0", + "domhandler": "^5.0.3" + }, + "funding": { + "url": "https://github.com/fb55/domutils?sponsor=1" + } + }, + "node_modules/dset": { + "version": "3.1.4", + "resolved": "https://registry.npmjs.org/dset/-/dset-3.1.4.tgz", + "integrity": "sha512-2QF/g9/zTaPDc3BjNcVTGoBbXBgYfMTTceLaYcFJ/W9kggFUkhxD/hMEeuLKbugyef9SqAx8cpgwlIP/jinUTA==", + "license": "MIT", + "engines": { + "node": ">=4" + } + }, + "node_modules/entities": { + "version": "4.5.0", + "resolved": "https://registry.npmjs.org/entities/-/entities-4.5.0.tgz", + "integrity": "sha512-V0hjH4dGPh9Ao5p0MoRY6BVqtwCjhz6vI5LT8AJ55H+4g9/4vbHx1I54fS0XuclLhDHArPQCiMjDxjaL8fPxhw==", + "license": "BSD-2-Clause", + "engines": { + "node": ">=0.12" + }, + "funding": { + "url": "https://github.com/fb55/entities?sponsor=1" + } + }, + "node_modules/es-module-lexer": { + "version": "2.3.2", + "resolved": "https://registry.npmjs.org/es-module-lexer/-/es-module-lexer-2.3.2.tgz", + "integrity": "sha512-poHGpORABojJJucnV9KbOavETW8lBVnphkW77ER5/BQ5Fz7oXSoCNek7IH3vR5nRjdsEz926ibFYX8KtLQmdyw==", + "license": "MIT" + }, + "node_modules/esbuild": { + "version": "0.28.2", + "resolved": "https://registry.npmjs.org/esbuild/-/esbuild-0.28.2.tgz", + "integrity": "sha512-HKVLS8dvII+xoKW9kmqxbRKrnWEXfJJr/FZhhJmiqIB0e053QNYFqOBouTMO/k5sID4MvCiUCvv8b9M4h32wIA==", + "hasInstallScript": true, + "license": "MIT", + "bin": { + "esbuild": "bin/esbuild" + }, + "engines": { + "node": ">=18" + }, + "optionalDependencies": { + "@esbuild/aix-ppc64": "0.28.2", + "@esbuild/android-arm": "0.28.2", + "@esbuild/android-arm64": "0.28.2", + "@esbuild/android-x64": "0.28.2", + "@esbuild/darwin-arm64": "0.28.2", + "@esbuild/darwin-x64": "0.28.2", + "@esbuild/freebsd-arm64": "0.28.2", + "@esbuild/freebsd-x64": "0.28.2", + "@esbuild/linux-arm": "0.28.2", + "@esbuild/linux-arm64": "0.28.2", + "@esbuild/linux-ia32": "0.28.2", + "@esbuild/linux-loong64": "0.28.2", + "@esbuild/linux-mips64el": "0.28.2", + "@esbuild/linux-ppc64": "0.28.2", + "@esbuild/linux-riscv64": "0.28.2", + "@esbuild/linux-s390x": "0.28.2", + "@esbuild/linux-x64": "0.28.2", + "@esbuild/netbsd-arm64": "0.28.2", + "@esbuild/netbsd-x64": "0.28.2", + "@esbuild/openbsd-arm64": "0.28.2", + "@esbuild/openbsd-x64": "0.28.2", + "@esbuild/openharmony-arm64": "0.28.2", + "@esbuild/sunos-x64": "0.28.2", + "@esbuild/win32-arm64": "0.28.2", + "@esbuild/win32-ia32": "0.28.2", + "@esbuild/win32-x64": "0.28.2" + } + }, + "node_modules/eventemitter3": { + "version": "5.0.4", + "resolved": "https://registry.npmjs.org/eventemitter3/-/eventemitter3-5.0.4.tgz", + "integrity": "sha512-mlsTRyGaPBjPedk6Bvw+aqbsXDtoAyAzm5MO7JgU+yVRyMQ5O8bD4Kcci7BS85f93veegeCPkL8R4GLClnjLFw==", + "license": "MIT" + }, + "node_modules/extend": { + "version": "3.0.2", + "resolved": "https://registry.npmjs.org/extend/-/extend-3.0.2.tgz", + "integrity": "sha512-fjquC59cD7CyW6urNXK0FBufkZcoiGG80wTuPujX590cB5Ttln20E2UB4S/WARVqhXffZl2LNgS+gQdPIIim/g==", + "license": "MIT" + }, + "node_modules/fast-string-truncated-width": { + "version": "3.0.3", + "resolved": "https://registry.npmjs.org/fast-string-truncated-width/-/fast-string-truncated-width-3.0.3.tgz", + "integrity": "sha512-0jjjIEL6+0jag3l2XWWizO64/aZVtpiGE3t0Zgqxv0DPuxiMjvB3M24fCyhZUO4KomJQPj3LTSUnDP3GpdwC0g==", + "license": "MIT" + }, + "node_modules/fast-string-width": { + "version": "3.0.2", + "resolved": "https://registry.npmjs.org/fast-string-width/-/fast-string-width-3.0.2.tgz", + "integrity": "sha512-gX8LrtNEI5hq8DVUfRQMbr5lpaS4nMIWV+7XEbXk2b8kiQIizgnlr12B4dA3ZEx3308ze0O4Q1R+cHts8kyUJg==", + "license": "MIT", + "dependencies": { + "fast-string-truncated-width": "^3.0.2" + } + }, + "node_modules/fast-wrap-ansi": { + "version": "0.2.2", + "resolved": "https://registry.npmjs.org/fast-wrap-ansi/-/fast-wrap-ansi-0.2.2.tgz", + "integrity": "sha512-7F2Fl+TjRSenLqlU3UjSH0iyqopqoZIu7eZVpEirP2g1GtWa2G/ecEmBdgz31+Mxr+ELclgg6sokpSFIQiZ02Q==", + "license": "MIT", + "dependencies": { + "fast-string-width": "^3.0.2" + } + }, + "node_modules/fdir": { + "version": "6.5.0", + "resolved": "https://registry.npmjs.org/fdir/-/fdir-6.5.0.tgz", + "integrity": "sha512-tIbYtZbucOs0BRGqPJkshJUYdL+SDH7dVM8gjy+ERp3WAUjLEFJE+02kanyHtwjWOnwrKYBiwAmM0p4kLJAnXg==", + "license": "MIT", + "engines": { + "node": ">=12.0.0" + }, + "peerDependencies": { + "picomatch": "^3 || ^4" + }, + "peerDependenciesMeta": { + "picomatch": { + "optional": true + } + } + }, + "node_modules/find-proc": { + "version": "0.2.0", + "resolved": "https://registry.npmjs.org/find-proc/-/find-proc-0.2.0.tgz", + "integrity": "sha512-a6h2nTrgB3g5Wrn4au9Ux4kMk7IJtyF6oA8uGn+mk3weLZssUtc1cp6meHIzzuW2o1srrBAxzkmXLbqjRkgB6w==", + "license": "MIT", + "engines": { + "node": "^20.19.0 || >=22.12.0" + } + }, + "node_modules/flattie": { + "version": "1.1.1", + "resolved": "https://registry.npmjs.org/flattie/-/flattie-1.1.1.tgz", + "integrity": "sha512-9UbaD6XdAL97+k/n+N7JwX46K/M6Zc6KcFYskrYL8wbBV/Uyk0CTAMY0VT+qiK5PM7AIc9aTWYtq65U7T+aCNQ==", + "license": "MIT", + "engines": { + "node": ">=8" + } + }, + "node_modules/fontace": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/fontace/-/fontace-0.4.1.tgz", + "integrity": "sha512-lDMvbAzSnHmbYMTEld5qdtvNH2/pWpICOqpean9IgC7vUbUJc3k+k5Dokp85CegamqQpFbXf0rAVkbzpyTA8aw==", + "license": "MIT", + "dependencies": { + "fontkitten": "^1.0.2" + } + }, + "node_modules/fontkitten": { + "version": "1.0.3", + "resolved": "https://registry.npmjs.org/fontkitten/-/fontkitten-1.0.3.tgz", + "integrity": "sha512-Wp1zXWPVUPBmfoa3Cqc9ctaKuzKAV6uLstRqlR56kSjplf5uAce+qeyYym7F+PHbGTk+tCEdkCW6RD7DX/gBZw==", + "license": "MIT", + "dependencies": { + "tiny-inflate": "^1.0.3" + }, + "engines": { + "node": ">=20" + } + }, + "node_modules/fsevents": { + "version": "2.3.3", + "resolved": "https://registry.npmjs.org/fsevents/-/fsevents-2.3.3.tgz", + "integrity": "sha512-5xoDfX+fL7faATnagmWPpbFtwh/R77WmMMqqHGS65C3vvB0YHrgF+B1YmZ3441tMj5n63k0212XNoJwzlhffQw==", + "hasInstallScript": true, + "license": "MIT", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": "^8.16.0 || ^10.6.0 || >=11.0.0" + } + }, + "node_modules/get-tsconfig": { + "version": "5.0.0-beta.4", + "resolved": "https://registry.npmjs.org/get-tsconfig/-/get-tsconfig-5.0.0-beta.4.tgz", + "integrity": "sha512-7nF7C9fIPFEMHgEMEfgIlO9wDdZ8CyHw27rWciFZfHvHDReIiPhsYuzPRXsfvBCqFy1l8RRyyWV7QLM+ZhUJsQ==", + "license": "MIT", + "dependencies": { + "resolve-pkg-maps": "^1.0.0" + }, + "engines": { + "node": ">=20.20.0" + }, + "funding": { + "url": "https://github.com/privatenumber/get-tsconfig?sponsor=1" + } + }, + "node_modules/github-slugger": { + "version": "2.0.0", + "resolved": "https://registry.npmjs.org/github-slugger/-/github-slugger-2.0.0.tgz", + "integrity": "sha512-IaOQ9puYtjrkq7Y0Ygl9KDZnrf/aiUJYUpVf89y8kyaxbRG7Y1SrX/jaumrv81vc61+kiMempujsM3Yw7w5qcw==", + "license": "ISC" + }, + "node_modules/h3": { + "version": "1.15.11", + "resolved": "https://registry.npmjs.org/h3/-/h3-1.15.11.tgz", + "integrity": "sha512-L3THSe2MPeBwgIZVSH5zLdBBU90TOxarvhK9d04IDY2AmVS8j2Jz2LIWtwsGOU3lu2I5jCN7FNvVfY2+XyF+mg==", + "license": "MIT", + "dependencies": { + "cookie-es": "^1.2.3", + "crossws": "^0.3.5", + "defu": "^6.1.6", + "destr": "^2.0.5", + "iron-webcrypto": "^1.2.1", + "node-mock-http": "^1.0.4", + "radix3": "^1.1.2", + "ufo": "^1.6.3", + "uncrypto": "^0.1.3" + } + }, + "node_modules/hast-util-to-html": { + "version": "9.0.5", + "resolved": "https://registry.npmjs.org/hast-util-to-html/-/hast-util-to-html-9.0.5.tgz", + "integrity": "sha512-OguPdidb+fbHQSU4Q4ZiLKnzWo8Wwsf5bZfbvu7//a9oTYoqD/fWpe96NuHkoS9h0ccGOTe0C4NGXdtS0iObOw==", + "license": "MIT", + "dependencies": { + "@types/hast": "^3.0.0", + "@types/unist": "^3.0.0", + "ccount": "^2.0.0", + "comma-separated-tokens": "^2.0.0", + "hast-util-whitespace": "^3.0.0", + "html-void-elements": "^3.0.0", + "mdast-util-to-hast": "^13.0.0", + "property-information": "^7.0.0", + "space-separated-tokens": "^2.0.0", + "stringify-entities": "^4.0.0", + "zwitch": "^2.0.4" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/hast-util-whitespace": { + "version": "3.0.0", + "resolved": "https://registry.npmjs.org/hast-util-whitespace/-/hast-util-whitespace-3.0.0.tgz", + "integrity": "sha512-88JUN06ipLwsnv+dVn+OIYOvAuvBMy/Qoi6O7mQHxdPXpjy+Cd6xRkWwux7DKO+4sYILtLBRIKgsdpS2gQc7qw==", + "license": "MIT", + "dependencies": { + "@types/hast": "^3.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/html-escaper": { + "version": "3.0.3", + "resolved": "https://registry.npmjs.org/html-escaper/-/html-escaper-3.0.3.tgz", + "integrity": "sha512-RuMffC89BOWQoY0WKGpIhn5gX3iI54O6nRA0yC124NYVtzjmFWBIiFd8M0x+ZdX0P9R4lADg1mgP8C7PxGOWuQ==", + "license": "MIT" + }, + "node_modules/html-void-elements": { + "version": "3.0.0", + "resolved": "https://registry.npmjs.org/html-void-elements/-/html-void-elements-3.0.0.tgz", + "integrity": "sha512-bEqo66MRXsUGxWHV5IP0PUiAWwoEjba4VCzg0LjFJBpchPaTfyfCKTG6bc5F8ucKec3q5y6qOdGyYTSBEvhCrg==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/http-cache-semantics": { + "version": "4.2.0", + "resolved": "https://registry.npmjs.org/http-cache-semantics/-/http-cache-semantics-4.2.0.tgz", + "integrity": "sha512-dTxcvPXqPvXBQpq5dUr6mEMJX4oIEFv6bwom3FDwKRDsuIjjJGANqhBuoAn9c1RQJIdAKav33ED65E2ys+87QQ==", + "license": "BSD-2-Clause" + }, + "node_modules/iron-webcrypto": { + "version": "1.2.1", + "resolved": "https://registry.npmjs.org/iron-webcrypto/-/iron-webcrypto-1.2.1.tgz", + "integrity": "sha512-feOM6FaSr6rEABp/eDfVseKyTMDt+KGpeB35SkVn9Tyn0CqvVsY3EwI0v5i8nMHyJnzCIQf7nsy3p41TPkJZhg==", + "license": "MIT", + "funding": { + "url": "https://github.com/sponsors/brc-dd" + } + }, + "node_modules/is-docker": { + "version": "4.0.0", + "resolved": "https://registry.npmjs.org/is-docker/-/is-docker-4.0.0.tgz", + "integrity": "sha512-LHE+wROyG/Y/0ZnbktRCoTix2c1RhgWaZraMZ8o1Q7zCh0VSrICJQO5oqIIISrcSBtrXv0o233w1IYwsWCjTzA==", + "license": "MIT", + "bin": { + "is-docker": "cli.js" + }, + "engines": { + "node": ">=20" + }, + "funding": { + "url": "https://github.com/sponsors/sindresorhus" + } + }, + "node_modules/is-plain-obj": { + "version": "4.1.0", + "resolved": "https://registry.npmjs.org/is-plain-obj/-/is-plain-obj-4.1.0.tgz", + "integrity": "sha512-+Pgi+vMuUNkJyExiMBt5IlFoMyKnr5zhJ4Uspz58WOhBF5QoIZkFyNHIbBAtHwzVAgk5RtndVNsDRN61/mmDqg==", + "license": "MIT", + "engines": { + "node": ">=12" + }, + "funding": { + "url": "https://github.com/sponsors/sindresorhus" + } + }, + "node_modules/js-yaml": { + "version": "4.3.2", + "resolved": "https://registry.npmjs.org/js-yaml/-/js-yaml-4.3.2.tgz", + "integrity": "sha512-SFNOvSJ+Dgf/9An904Yx+CgSlIPCkIpao4qo51lpee25TIRejdH3rhR4EZMGoNx3/TP3O+wzWuiTFl4sqbltzA==", + "funding": [ + { + "type": "github", + "url": "https://github.com/sponsors/puzrin" + }, + { + "type": "github", + "url": "https://github.com/sponsors/nodeca" + } + ], + "license": "MIT", + "dependencies": { + "argparse": "^2.0.1" + }, + "bin": { + "js-yaml": "bin/js-yaml.js" + } + }, + "node_modules/jsonc-parser": { + "version": "3.3.1", + "resolved": "https://registry.npmjs.org/jsonc-parser/-/jsonc-parser-3.3.1.tgz", + "integrity": "sha512-HUgH65KyejrUFPvHFPbqOY0rsFip3Bo5wb4ngvdi1EpCYWUQDC5V+Y7mZws+DLkr4M//zQJoanu1SP+87Dv1oQ==", + "license": "MIT" + }, + "node_modules/lightningcss": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss/-/lightningcss-1.33.0.tgz", + "integrity": "sha512-WkUDrojuJs0xkgGf2udWxa3yGBRxPtxUkB79i6aCZLRgc7PM8fZe9TosfPDcvEpQZbuFASnHYmRLBLUbmLOIIA==", + "license": "MPL-2.0", + "dependencies": { + "detect-libc": "^2.0.3" + }, + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + }, + "optionalDependencies": { + "lightningcss-android-arm64": "1.33.0", + "lightningcss-darwin-arm64": "1.33.0", + "lightningcss-darwin-x64": "1.33.0", + "lightningcss-freebsd-x64": "1.33.0", + "lightningcss-linux-arm-gnueabihf": "1.33.0", + "lightningcss-linux-arm64-gnu": "1.33.0", + "lightningcss-linux-arm64-musl": "1.33.0", + "lightningcss-linux-x64-gnu": "1.33.0", + "lightningcss-linux-x64-musl": "1.33.0", + "lightningcss-win32-arm64-msvc": "1.33.0", + "lightningcss-win32-x64-msvc": "1.33.0" + } + }, + "node_modules/lightningcss-android-arm64": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-android-arm64/-/lightningcss-android-arm64-1.33.0.tgz", + "integrity": "sha512-gEpRTalKdosp4Bb8qWtc2iOgE5SeIHlpS1up9bFq2wAyYhl1UdTObYiHe98zEM9SQvSoqQZ1IQD0JNpg3Ml5pg==", + "cpu": [ + "arm64" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "android" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-darwin-arm64": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-darwin-arm64/-/lightningcss-darwin-arm64-1.33.0.tgz", + "integrity": "sha512-Sciaz8eenNTKn9b3t7+xr0ipTp9YxKQY4npwQ3mrRuL0BAVHBLyZxofhaKBAVtzmtRZ/zTyo0/to4B1uWG/Djg==", + "cpu": [ + "arm64" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-darwin-x64": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-darwin-x64/-/lightningcss-darwin-x64-1.33.0.tgz", + "integrity": "sha512-Z5UPAxzrjlWNNyGy6i65cJzzvgJ5D3T6wMvs+gWpY9d7qRhANrxqAp6LhxIgZhWEw18RfJTGcRxjuLIBr+m8XQ==", + "cpu": [ + "x64" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "darwin" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-freebsd-x64": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-freebsd-x64/-/lightningcss-freebsd-x64-1.33.0.tgz", + "integrity": "sha512-QQM/Ti/hQajJwCY+RiWuCZ9sdtI/XQk7nDK5vC8kkdwixezOlDgvDx7+RT+QjK6FcFT4MpsuoBnHIo/O3StRRg==", + "cpu": [ + "x64" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "freebsd" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-linux-arm-gnueabihf": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-linux-arm-gnueabihf/-/lightningcss-linux-arm-gnueabihf-1.33.0.tgz", + "integrity": "sha512-N7FVBe6iS24MlM6R/4RBTxGhQheZGs7tiQ9U32UtF75NzP5Q7xWPRqLBCKxlRQRk3rY1jCIPLzx7WzOhuUIRLQ==", + "cpu": [ + "arm" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-linux-arm64-gnu": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-linux-arm64-gnu/-/lightningcss-linux-arm64-gnu-1.33.0.tgz", + "integrity": "sha512-j2v/itmy4HlNxlc6voKXYgBqNi0Ng2LShg4z7GufpEgs05P+2suBVyi9I6YHq5uoVFx9ETin3eCEhLVyXGQnKg==", + "cpu": [ + "arm64" + ], + "libc": [ + "glibc" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-linux-arm64-musl": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-linux-arm64-musl/-/lightningcss-linux-arm64-musl-1.33.0.tgz", + "integrity": "sha512-yiO5ROMuYQgXbC60yjZU5CYSFZGKXL0HFATXt9mHJn1+zW55oCtMI9NfcVhYLMFDL7gV7oBPon/EmMMGg2OvtQ==", + "cpu": [ + "arm64" + ], + "libc": [ + "musl" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-linux-x64-gnu": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-linux-x64-gnu/-/lightningcss-linux-x64-gnu-1.33.0.tgz", + "integrity": "sha512-ar+Ju7LmcN0Jo4FpL4hpFybwNG9/3A/Br5KW2n2jyODg3MEZXaDYADdemoNS+BDNfMgKvylJLj4S5tyRActuAg==", + "cpu": [ + "x64" + ], + "libc": [ + "glibc" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-linux-x64-musl": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-linux-x64-musl/-/lightningcss-linux-x64-musl-1.33.0.tgz", + "integrity": "sha512-RYiYbkokw0trfKqqzfF55lginwEPrD3OJDfTuJzFs1MK6iFnDenaz1fqLLtX4ITG3OktJQXOeTaw1awrBAlZPw==", + "cpu": [ + "x64" + ], + "libc": [ + "musl" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "linux" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-win32-arm64-msvc": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-win32-arm64-msvc/-/lightningcss-win32-arm64-msvc-1.33.0.tgz", + "integrity": "sha512-1K+MPfLSFVpphzpdbfkhlWk6wBrTObBzS2T6db10PNOZgR9GoVsAWzwNyuhUYYbTp23j+4RrncfujZ4uAzXvwA==", + "cpu": [ + "arm64" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lightningcss-win32-x64-msvc": { + "version": "1.33.0", + "resolved": "https://registry.npmjs.org/lightningcss-win32-x64-msvc/-/lightningcss-win32-x64-msvc-1.33.0.tgz", + "integrity": "sha512-OlEICDx/Xl0FqSp4bry8zFnCvGpig3Gl4gCquvYwHuqJKEC1+n9NgDniFvqHGmMv1ZkqDJrDqKKSykTDX+ehuA==", + "cpu": [ + "x64" + ], + "license": "MPL-2.0", + "optional": true, + "os": [ + "win32" + ], + "engines": { + "node": ">= 12.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/parcel" + } + }, + "node_modules/lru-cache": { + "version": "11.5.3", + "resolved": "https://registry.npmjs.org/lru-cache/-/lru-cache-11.5.3.tgz", + "integrity": "sha512-U4N8FgzmWxc8k1VH8Kr6lQg18U7Fjvby6wXHVRX/ZZ7IwWbRMgrRbP0Wrb5q5NVinryp4SQampHKdvtecItxUg==", + "license": "BlueOak-1.0.0", + "engines": { + "node": "20 || >=22" + } + }, + "node_modules/magic-string": { + "version": "1.4.1", + "resolved": "https://registry.npmjs.org/magic-string/-/magic-string-1.4.1.tgz", + "integrity": "sha512-8lyCu36ErXR0J9uaGKlKQoiLZKmtI63YGLE8G2o9jyRPdr4X47LusSOwgOJOzcVtp81fTAAjxR7BwKz682Jhow==", + "license": "MIT", + "dependencies": { + "@jridgewell/sourcemap-codec": "^1.6.0" + } + }, + "node_modules/magicast": { + "version": "0.5.5", + "resolved": "https://registry.npmjs.org/magicast/-/magicast-0.5.5.tgz", + "integrity": "sha512-UicdXN8zQ3JHlxVq+28afMXPr1z7WNY6+7EJnzTdQWkTAlMLF5fNCCKxJHBQwGaNGR11581EiQmQzx73+MvszA==", + "license": "MIT", + "dependencies": { + "@babel/parser": "^7.29.7", + "@babel/types": "^7.29.7", + "source-map-js": "^1.2.1" + } + }, + "node_modules/mdast-util-to-hast": { + "version": "13.2.1", + "resolved": "https://registry.npmjs.org/mdast-util-to-hast/-/mdast-util-to-hast-13.2.1.tgz", + "integrity": "sha512-cctsq2wp5vTsLIcaymblUriiTcZd0CwWtCbLvrOzYCDZoWyMNV8sZ7krj09FSnsiJi3WVsHLM4k6Dq/yaPyCXA==", + "license": "MIT", + "dependencies": { + "@types/hast": "^3.0.0", + "@types/mdast": "^4.0.0", + "@ungap/structured-clone": "^1.0.0", + "devlop": "^1.0.0", + "micromark-util-sanitize-uri": "^2.0.0", + "trim-lines": "^3.0.0", + "unist-util-position": "^5.0.0", + "unist-util-visit": "^5.0.0", + "vfile": "^6.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/mdn-data": { + "version": "2.27.1", + "resolved": "https://registry.npmjs.org/mdn-data/-/mdn-data-2.27.1.tgz", + "integrity": "sha512-9Yubnt3e8A0OKwxYSXyhLymGW4sCufcLG6VdiDdUGVkPhpqLxlvP5vl1983gQjJl3tqbrM731mjaZaP68AgosQ==", + "license": "CC0-1.0" + }, + "node_modules/micromark-util-character": { + "version": "2.1.1", + "resolved": "https://registry.npmjs.org/micromark-util-character/-/micromark-util-character-2.1.1.tgz", + "integrity": "sha512-wv8tdUTJ3thSFFFJKtpYKOYiGP2+v96Hvk4Tu8KpCAsTMs6yi+nVmGh1syvSCsaxz45J6Jbw+9DD6g97+NV67Q==", + "funding": [ + { + "type": "GitHub Sponsors", + "url": "https://github.com/sponsors/unifiedjs" + }, + { + "type": "OpenCollective", + "url": "https://opencollective.com/unified" + } + ], + "license": "MIT", + "dependencies": { + "micromark-util-symbol": "^2.0.0", + "micromark-util-types": "^2.0.0" + } + }, + "node_modules/micromark-util-encode": { + "version": "2.0.1", + "resolved": "https://registry.npmjs.org/micromark-util-encode/-/micromark-util-encode-2.0.1.tgz", + "integrity": "sha512-c3cVx2y4KqUnwopcO9b/SCdo2O67LwJJ/UyqGfbigahfegL9myoEFoDYZgkT7f36T0bLrM9hZTAaAyH+PCAXjw==", + "funding": [ + { + "type": "GitHub Sponsors", + "url": "https://github.com/sponsors/unifiedjs" + }, + { + "type": "OpenCollective", + "url": "https://opencollective.com/unified" + } + ], + "license": "MIT" + }, + "node_modules/micromark-util-sanitize-uri": { + "version": "2.0.1", + "resolved": "https://registry.npmjs.org/micromark-util-sanitize-uri/-/micromark-util-sanitize-uri-2.0.1.tgz", + "integrity": "sha512-9N9IomZ/YuGGZZmQec1MbgxtlgougxTodVwDzzEouPKo3qFWvymFHWcnDi2vzV1ff6kas9ucW+o3yzJK9YB1AQ==", + "funding": [ + { + "type": "GitHub Sponsors", + "url": "https://github.com/sponsors/unifiedjs" + }, + { + "type": "OpenCollective", + "url": "https://opencollective.com/unified" + } + ], + "license": "MIT", + "dependencies": { + "micromark-util-character": "^2.0.0", + "micromark-util-encode": "^2.0.0", + "micromark-util-symbol": "^2.0.0" + } + }, + "node_modules/micromark-util-symbol": { + "version": "2.0.1", + "resolved": "https://registry.npmjs.org/micromark-util-symbol/-/micromark-util-symbol-2.0.1.tgz", + "integrity": "sha512-vs5t8Apaud9N28kgCrRUdEed4UJ+wWNvicHLPxCa9ENlYuAY31M0ETy5y1vA33YoNPDFTghEbnh6efaE8h4x0Q==", + "funding": [ + { + "type": "GitHub Sponsors", + "url": "https://github.com/sponsors/unifiedjs" + }, + { + "type": "OpenCollective", + "url": "https://opencollective.com/unified" + } + ], + "license": "MIT" + }, + "node_modules/micromark-util-types": { + "version": "2.0.2", + "resolved": "https://registry.npmjs.org/micromark-util-types/-/micromark-util-types-2.0.2.tgz", + "integrity": "sha512-Yw0ECSpJoViF1qTU4DC6NwtC4aWGt1EkzaQB8KPPyCRR8z9TWeV0HbEFGTO+ZY1wB22zmxnJqhPyTpOVCpeHTA==", + "funding": [ + { + "type": "GitHub Sponsors", + "url": "https://github.com/sponsors/unifiedjs" + }, + { + "type": "OpenCollective", + "url": "https://opencollective.com/unified" + } + ], + "license": "MIT" + }, + "node_modules/mrmime": { + "version": "2.0.1", + "resolved": "https://registry.npmjs.org/mrmime/-/mrmime-2.0.1.tgz", + "integrity": "sha512-Y3wQdFg2Va6etvQ5I82yUhGdsKrcYox6p7FfL1LbK2J4V01F9TGlepTIhnK24t7koZibmg82KGglhA1XK5IsLQ==", + "license": "MIT", + "engines": { + "node": ">=10" + } + }, + "node_modules/nanoid": { + "version": "3.3.19", + "resolved": "https://registry.npmjs.org/nanoid/-/nanoid-3.3.19.tgz", + "integrity": "sha512-Y2tUNy4ouw6tq5oDSKeQYGOyhkUBhNOcGV/02KC+6kd9eDGqdZd++mjMiIDilrBYvjEnCYvVtsuHCuP+okSfug==", + "funding": [ + { + "type": "github", + "url": "https://github.com/sponsors/ai" + } + ], + "license": "MIT", + "bin": { + "nanoid": "bin/nanoid.cjs" + }, + "engines": { + "node": "^10 || ^12 || ^13.7 || ^14 || >=15.0.1" + } + }, + "node_modules/neotraverse": { + "version": "1.0.1", + "resolved": "https://registry.npmjs.org/neotraverse/-/neotraverse-1.0.1.tgz", + "integrity": "sha512-WmmLty1YWwJl9yZi77v2dVIV6X2kuYV8YYBI/G3LWGKdGHmHUvL1z7FW0iDvEvGAwNEoc5x1tOOOyDnf5jJw/w==", + "license": "MIT", + "engines": { + "node": ">= 10" + } + }, + "node_modules/nlcst-to-string": { + "version": "4.0.0", + "resolved": "https://registry.npmjs.org/nlcst-to-string/-/nlcst-to-string-4.0.0.tgz", + "integrity": "sha512-YKLBCcUYKAg0FNlOBT6aI91qFmSiFKiluk655WzPF+DDMA02qIyy8uiRqI8QXtcFpEvll12LpL5MXqEmAZ+dcA==", + "license": "MIT", + "dependencies": { + "@types/nlcst": "^2.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/node-fetch-native": { + "version": "1.6.7", + "resolved": "https://registry.npmjs.org/node-fetch-native/-/node-fetch-native-1.6.7.tgz", + "integrity": "sha512-g9yhqoedzIUm0nTnTqAQvueMPVOuIY16bqgAJJC8XOOubYFNwz6IER9qs0Gq2Xd0+CecCKFjtdDTMA4u4xG06Q==", + "license": "MIT" + }, + "node_modules/node-mock-http": { + "version": "1.0.5", + "resolved": "https://registry.npmjs.org/node-mock-http/-/node-mock-http-1.0.5.tgz", + "integrity": "sha512-KQyt/wLjG3TAc7DOUhpqWzgd4ERxR80JOlTK5VE5R1S12IaPVN5qkj4klBce9HPG1Njuup4Sb5bljaT34lIyjw==", + "license": "MIT" + }, + "node_modules/normalize-path": { + "version": "3.0.0", + "resolved": "https://registry.npmjs.org/normalize-path/-/normalize-path-3.0.0.tgz", + "integrity": "sha512-6eZs5Ls3WtCisHWp9S2GUy8dqkpGi4BVSz3GaqiE6ezub0512ESztXUwUB6C6IKbQkY2Pnb/mD4WYojCRwcwLA==", + "license": "MIT", + "engines": { + "node": ">=0.10.0" + } + }, + "node_modules/nth-check": { + "version": "2.1.1", + "resolved": "https://registry.npmjs.org/nth-check/-/nth-check-2.1.1.tgz", + "integrity": "sha512-lqjrjmaOoAnWfMmBPL+XNnynZh2+swxiX3WUE0s4yEHI6m+AwrK2UZOimIRl3X/4QctVqS8AiZjFqyOGrMXb/w==", + "license": "BSD-2-Clause", + "dependencies": { + "boolbase": "^1.0.0" + }, + "funding": { + "url": "https://github.com/fb55/nth-check?sponsor=1" + } + }, + "node_modules/obug": { + "version": "3.0.0", + "resolved": "https://registry.npmjs.org/obug/-/obug-3.0.0.tgz", + "integrity": "sha512-5vvB5+W7ePv+p3uqxi+RcW1XAzLW0/hxt3/4X4Lc4qHudzOhmBiBwOY6DRob4WnanAEGvNcLjF+KNOufrUoEQw==", + "funding": [ + "https://github.com/sponsors/sxzz", + "https://opencollective.com/debug" + ], + "license": "MIT", + "engines": { + "node": ">=12.20.0" + } + }, + "node_modules/ofetch": { + "version": "1.5.1", + "resolved": "https://registry.npmjs.org/ofetch/-/ofetch-1.5.1.tgz", + "integrity": "sha512-2W4oUZlVaqAPAil6FUg/difl6YhqhUR7x2eZY4bQCko22UXg3hptq9KLQdqFClV+Wu85UX7hNtdGTngi/1BxcA==", + "license": "MIT", + "dependencies": { + "destr": "^2.0.5", + "node-fetch-native": "^1.6.7", + "ufo": "^1.6.1" + } + }, + "node_modules/ohash": { + "version": "2.0.12", + "resolved": "https://registry.npmjs.org/ohash/-/ohash-2.0.12.tgz", + "integrity": "sha512-65S/5gk9YSsaRjcyf7Nfa6h/d3E8/1gslpXfI4W7Dxn/oap8IKRuNT5VXkLQ1YFKIEg4apRY4Pj6aiwFzrDdmw==", + "license": "MIT" + }, + "node_modules/oniguruma-parser": { + "version": "0.12.2", + "resolved": "https://registry.npmjs.org/oniguruma-parser/-/oniguruma-parser-0.12.2.tgz", + "integrity": "sha512-6HVa5oIrgMC6aA6WF6XyyqbhRPJrKR02L20+2+zpDtO5QAzGHAUGw5TKQvwi5vctNnRHkJYmjAhRVQF2EKdTQw==", + "license": "MIT" + }, + "node_modules/oniguruma-to-es": { + "version": "4.3.6", + "resolved": "https://registry.npmjs.org/oniguruma-to-es/-/oniguruma-to-es-4.3.6.tgz", + "integrity": "sha512-csuQ9x3Yr0cEIs/Zgx/OEt9iBw9vqIunAPQkx19R/fiMq2oGVTgcMqO/V3Ybqefr1TBvosI6jU539ksaBULJyA==", + "license": "MIT", + "dependencies": { + "oniguruma-parser": "^0.12.2", + "regex": "^6.1.0", + "regex-recursion": "^6.0.2" + } + }, + "node_modules/p-limit": { + "version": "7.3.3", + "resolved": "https://registry.npmjs.org/p-limit/-/p-limit-7.3.3.tgz", + "integrity": "sha512-SKgVvDU5TNIxQ4r30U75BNLC6a+6GOEjngrk/ysM5+Zu3oJuk1SC5ApzIzY/7PjCEjoVPMWJ0zho/4QUiEB7TQ==", + "license": "MIT", + "dependencies": { + "yocto-queue": "^1.2.1" + }, + "engines": { + "node": ">=20" + }, + "funding": { + "url": "https://github.com/sponsors/sindresorhus" + } + }, + "node_modules/p-queue": { + "version": "9.3.3", + "resolved": "https://registry.npmjs.org/p-queue/-/p-queue-9.3.3.tgz", + "integrity": "sha512-NXAOdnEe5FsZJfT4oK84lE1Y5cFFdWlRuOo5tww8DyNMxyRXwn39fIkUtNLKppcPC+UYU/bXujNCUGDv01y7CA==", + "license": "MIT", + "dependencies": { + "eventemitter3": "^5.0.4", + "p-timeout": "^7.0.0" + }, + "engines": { + "node": ">=20" + }, + "funding": { + "url": "https://github.com/sponsors/sindresorhus" + } + }, + "node_modules/p-timeout": { + "version": "7.0.2", + "resolved": "https://registry.npmjs.org/p-timeout/-/p-timeout-7.0.2.tgz", + "integrity": "sha512-prbX4Z3YszrFNgH+MW5Zoeq3baXrMtP/MQnFeET90UB/GtGcGDQ5Usg9OCy6ETjTTntOw1SL2z9fMPUppN3Guw==", + "license": "MIT", + "engines": { + "node": ">=20" + }, + "funding": { + "url": "https://github.com/sponsors/sindresorhus" + } + }, + "node_modules/package-manager-detector": { + "version": "1.8.0", + "resolved": "https://registry.npmjs.org/package-manager-detector/-/package-manager-detector-1.8.0.tgz", + "integrity": "sha512-yQA4H19AmPEoMUeavPMDIe1higySl/gH/yaQrkT/s07Qp+7pp2hYz30N3z2l5BkjVkF9Ow6o0wjJamm2y7Sn0A==", + "license": "MIT" + }, + "node_modules/piccolore": { + "version": "0.1.3", + "resolved": "https://registry.npmjs.org/piccolore/-/piccolore-0.1.3.tgz", + "integrity": "sha512-o8bTeDWjE086iwKrROaDf31K0qC/BENdm15/uH9usSC/uZjJOKb2YGiVHfLY4GhwsERiPI1jmwI2XrA7ACOxVw==", + "license": "ISC" + }, + "node_modules/picocolors": { + "version": "1.1.1", + "resolved": "https://registry.npmjs.org/picocolors/-/picocolors-1.1.1.tgz", + "integrity": "sha512-xceH2snhtb5M9liqDsmEw56le376mTZkEX/jEb/RxNFyegNul7eNslCXP9FDj/Lcu0X8KEyMceP2ntpaHrDEVA==", + "license": "ISC" + }, + "node_modules/picomatch": { + "version": "4.0.7", + "resolved": "https://registry.npmjs.org/picomatch/-/picomatch-4.0.7.tgz", + "integrity": "sha512-qcJu88Q2IWqJsDD529JKMdwGm/dvInW4HvQnRwiH9JtihJvzGOscDtHE3x1pBKeUOTysQ8kVmLnJ2kJu7yhcGA==", + "license": "MIT", + "engines": { + "node": ">=12" + }, + "funding": { + "url": "https://github.com/sponsors/jonschlinkert" + } + }, + "node_modules/postcss": { + "version": "8.5.28", + "resolved": "https://registry.npmjs.org/postcss/-/postcss-8.5.28.tgz", + "integrity": "sha512-RRuzqDtt5Y9h3quz5hWhK+TPnsmVs6WwSU6LkJMeY4HstUEDuYTG8UJSdawMRzmzAtV+KEoG8N3Qg2qLy5vM/A==", + "funding": [ + { + "type": "opencollective", + "url": "https://opencollective.com/postcss/" + }, + { + "type": "tidelift", + "url": "https://tidelift.com/funding/github/npm/postcss" + }, + { + "type": "github", + "url": "https://github.com/sponsors/ai" + } + ], + "license": "MIT", + "dependencies": { + "nanoid": "^3.3.18", + "picocolors": "^1.1.1", + "source-map-js": "^1.2.1" + }, + "engines": { + "node": "^10 || ^12 || >=14" + } + }, + "node_modules/prismjs": { + "version": "1.30.0", + "resolved": "https://registry.npmjs.org/prismjs/-/prismjs-1.30.0.tgz", + "integrity": "sha512-DEvV2ZF2r2/63V+tK8hQvrR2ZGn10srHbXviTlcv7Kpzw8jWiNTqbVgjO3IY8RxrrOUF8VPMQQFysYYYv0YZxw==", + "license": "MIT", + "engines": { + "node": ">=6" + } + }, + "node_modules/process-ancestry": { + "version": "0.1.0", + "resolved": "https://registry.npmjs.org/process-ancestry/-/process-ancestry-0.1.0.tgz", + "integrity": "sha512-tGqJW/UnclpYASFcM6Xh8D8l/BMtaQ9+CSG0vlJSJTcdMM4lDRv4c6H0Pdcsfted+bVczdYSfk2fdukg2gQkZg==", + "license": "MIT", + "engines": { + "node": ">=18.0.0" + } + }, + "node_modules/property-information": { + "version": "7.2.0", + "resolved": "https://registry.npmjs.org/property-information/-/property-information-7.2.0.tgz", + "integrity": "sha512-IAtzIB6sUiWaJYrX9smp3V46pBGbBeLFRGdh25kg1334VcBlD8HzhPeNIWQH9zhGmo2itIe25EHt9dQP7G5hmg==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/radix3": { + "version": "1.1.2", + "resolved": "https://registry.npmjs.org/radix3/-/radix3-1.1.2.tgz", + "integrity": "sha512-b484I/7b8rDEdSDKckSSBA8knMpcdsXudlE/LNL639wFoHKwLbEkQFZHWEYwDC0wa0FKUcCY+GAF73Z7wxNVFA==", + "license": "MIT" + }, + "node_modules/readdirp": { + "version": "5.1.1", + "resolved": "https://registry.npmjs.org/readdirp/-/readdirp-5.1.1.tgz", + "integrity": "sha512-Kko+Y5XQ6fM+Ce3dq3m9YGxnacYZYl9cA1wZjaF3Vbry2L3i1qVg8+CAgNPsXRArPMUMCaOR7oa9Nqntc43JKA==", + "license": "MIT", + "engines": { + "node": ">= 20.19.0" + }, + "funding": { + "type": "individual", + "url": "https://paulmillr.com/funding/" + } + }, + "node_modules/regex": { + "version": "6.1.0", + "resolved": "https://registry.npmjs.org/regex/-/regex-6.1.0.tgz", + "integrity": "sha512-6VwtthbV4o/7+OaAF9I5L5V3llLEsoPyq9P1JVXkedTP33c7MfCG0/5NOPcSJn0TzXcG9YUrR0gQSWioew3LDg==", + "license": "MIT", + "dependencies": { + "regex-utilities": "^2.3.0" + } + }, + "node_modules/regex-recursion": { + "version": "6.0.2", + "resolved": "https://registry.npmjs.org/regex-recursion/-/regex-recursion-6.0.2.tgz", + "integrity": "sha512-0YCaSCq2VRIebiaUviZNs0cBz1kg5kVS2UKUfNIx8YVs1cN3AV7NTctO5FOKBA+UT2BPJIWZauYHPqJODG50cg==", + "license": "MIT", + "dependencies": { + "regex-utilities": "^2.3.0" + } + }, + "node_modules/regex-utilities": { + "version": "2.3.0", + "resolved": "https://registry.npmjs.org/regex-utilities/-/regex-utilities-2.3.0.tgz", + "integrity": "sha512-8VhliFJAWRaUiVvREIiW2NXXTmHs4vMNnSzuJVhscgmGav3g9VDxLrQndI3dZZVVdp0ZO/5v0xmX516/7M9cng==", + "license": "MIT" + }, + "node_modules/resolve-pkg-maps": { + "version": "1.0.0", + "resolved": "https://registry.npmjs.org/resolve-pkg-maps/-/resolve-pkg-maps-1.0.0.tgz", + "integrity": "sha512-seS2Tj26TBVOC2NIc2rOe2y2ZO7efxITtLZcGSOnHHNOQ7CkiUBfw0Iw2ck6xkIhPwLhKNLS8BO+hEpngQlqzw==", + "license": "MIT", + "funding": { + "url": "https://github.com/privatenumber/resolve-pkg-maps?sponsor=1" + } + }, + "node_modules/retext-smartypants": { + "version": "6.2.0", + "resolved": "https://registry.npmjs.org/retext-smartypants/-/retext-smartypants-6.2.0.tgz", + "integrity": "sha512-kk0jOU7+zGv//kfjXEBjdIryL1Acl4i9XNkHxtM7Tm5lFiCog576fjNC9hjoR7LTKQ0DsPWy09JummSsH1uqfQ==", + "license": "MIT", + "dependencies": { + "@types/nlcst": "^2.0.0", + "nlcst-to-string": "^4.0.0", + "unist-util-visit": "^5.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/rolldown": { + "version": "1.2.9", + "resolved": "https://registry.npmjs.org/rolldown/-/rolldown-1.2.9.tgz", + "integrity": "sha512-hx/Pv0N1haXRb11qkfnK5MXB/iqr7i0yjWQqmO9uHqZpBgQSqzc8UsSnEpalsh+j1I8qQ2CkXAkJC8Br3dKSlg==", + "license": "MIT", + "dependencies": { + "@oxc-project/types": "=0.150.0", + "@rolldown/pluginutils": "^1.0.0" + }, + "bin": { + "rolldown": "bin/cli.mjs" + }, + "engines": { + "node": "^20.19.0 || >=22.12.0" + }, + "optionalDependencies": { + "@rolldown/binding-android-arm-eabi": "1.2.9", + "@rolldown/binding-android-arm64": "1.2.9", + "@rolldown/binding-darwin-arm64": "1.2.9", + "@rolldown/binding-darwin-x64": "1.2.9", + "@rolldown/binding-freebsd-x64": "1.2.9", + "@rolldown/binding-linux-arm-gnueabihf": "1.2.9", + "@rolldown/binding-linux-arm64-gnu": "1.2.9", + "@rolldown/binding-linux-arm64-musl": "1.2.9", + "@rolldown/binding-linux-ppc64-gnu": "1.2.9", + "@rolldown/binding-linux-s390x-gnu": "1.2.9", + "@rolldown/binding-linux-x64-gnu": "1.2.9", + "@rolldown/binding-linux-x64-musl": "1.2.9", + "@rolldown/binding-openharmony-arm64": "1.2.9", + "@rolldown/binding-win32-arm64-msvc": "1.2.9", + "@rolldown/binding-win32-x64-msvc": "1.2.9" + } + }, + "node_modules/satteri": { + "version": "0.10.5", + "resolved": "https://registry.npmjs.org/satteri/-/satteri-0.10.5.tgz", + "integrity": "sha512-Ao1LKpAEa9Wdg0otgbVKViZHEq9ebdXe4DMrp3s9vQAU0HNIuHnFEuMuOcm0ZIXyV0Yzxj91NvhLpvXZJO/5ZQ==", + "license": "MIT", + "dependencies": { + "@types/estree-jsx": "^1.0.5", + "@types/hast": "^3.0.5", + "@types/mdast": "^4.0.4", + "@types/unist": "^3.0.3" + }, + "optionalDependencies": { + "@bruits/satteri-darwin-arm64": "0.10.5", + "@bruits/satteri-darwin-x64": "0.10.5", + "@bruits/satteri-linux-arm64-gnu": "0.10.5", + "@bruits/satteri-linux-arm64-musl": "0.10.5", + "@bruits/satteri-linux-x64-gnu": "0.10.5", + "@bruits/satteri-linux-x64-musl": "0.10.5", + "@bruits/satteri-wasm32-wasi": "0.10.5", + "@bruits/satteri-win32-arm64-msvc": "0.10.5", + "@bruits/satteri-win32-x64-msvc": "0.10.5" + } + }, + "node_modules/sax": { + "version": "1.6.1", + "resolved": "https://registry.npmjs.org/sax/-/sax-1.6.1.tgz", + "integrity": "sha512-42tBVwLWnaQvW5zc4HbZrTuWccECCZfBi92FDuwtqxasH+JbPB3/FOKb1m222K42R4WxuxzzMsTswfzgtSu64Q==", + "license": "BlueOak-1.0.0", + "engines": { + "node": ">=11.0.0" + } + }, + "node_modules/semver": { + "version": "7.8.5", + "resolved": "https://registry.npmjs.org/semver/-/semver-7.8.5.tgz", + "integrity": "sha512-Y7/KDsb8LjooZpwaqGyulO6DQlksgCncchHGk+sZIY4SBvUocMBEFH5Ur1fI4dV+Jvl0w6cjvucaIi40puRioA==", + "license": "ISC", + "optional": true, + "bin": { + "semver": "bin/semver.js" + }, + "engines": { + "node": ">=10" + } + }, + "node_modules/sharp": { + "version": "0.35.4", + "resolved": "https://registry.npmjs.org/sharp/-/sharp-0.35.4.tgz", + "integrity": "sha512-n++8XWcj+jCOr2IOl7h8LbKnGBDY4aPbmprMONBNFdn0ImXqpGVv5zliDs0V9HbmbCQLpbuo2ej9rAoOQTvMDA==", + "license": "Apache-2.0", + "optional": true, + "dependencies": { + "@img/colour": "^1.1.0", + "detect-libc": "^2.1.2", + "semver": "^7.8.5" + }, + "engines": { + "node": ">=20.9.0" + }, + "funding": { + "url": "https://opencollective.com/libvips" + }, + "optionalDependencies": { + "@img/sharp-darwin-arm64": "0.35.4", + "@img/sharp-darwin-x64": "0.35.4", + "@img/sharp-freebsd-wasm32": "0.35.4", + "@img/sharp-libvips-darwin-arm64": "1.3.3", + "@img/sharp-libvips-darwin-x64": "1.3.3", + "@img/sharp-libvips-linux-arm": "1.3.3", + "@img/sharp-libvips-linux-arm64": "1.3.3", + "@img/sharp-libvips-linux-ppc64": "1.3.3", + "@img/sharp-libvips-linux-riscv64": "1.3.3", + "@img/sharp-libvips-linux-s390x": "1.3.3", + "@img/sharp-libvips-linux-x64": "1.3.3", + "@img/sharp-libvips-linuxmusl-arm64": "1.3.3", + "@img/sharp-libvips-linuxmusl-x64": "1.3.3", + "@img/sharp-linux-arm": "0.35.4", + "@img/sharp-linux-arm64": "0.35.4", + "@img/sharp-linux-ppc64": "0.35.4", + "@img/sharp-linux-riscv64": "0.35.4", + "@img/sharp-linux-s390x": "0.35.4", + "@img/sharp-linux-x64": "0.35.4", + "@img/sharp-linuxmusl-arm64": "0.35.4", + "@img/sharp-linuxmusl-x64": "0.35.4", + "@img/sharp-webcontainers-wasm32": "0.35.4", + "@img/sharp-win32-arm64": "0.35.4", + "@img/sharp-win32-ia32": "0.35.4", + "@img/sharp-win32-x64": "0.35.4" + }, + "peerDependenciesMeta": { + "@types/node": { + "optional": true + } + } + }, + "node_modules/shiki": { + "version": "4.4.3", + "resolved": "https://registry.npmjs.org/shiki/-/shiki-4.4.3.tgz", + "integrity": "sha512-Mb/GvXPHBAXdgGIcnfU5L3ldpn1XcxrGkPHwqgRx17/I2XRfqlFKk2vGkHWINn1kdXvzJZeuO3is6I9KLPFm0g==", + "license": "MIT", + "dependencies": { + "@shikijs/core": "4.4.3", + "@shikijs/engine-javascript": "4.4.3", + "@shikijs/engine-oniguruma": "4.4.3", + "@shikijs/langs": "4.4.3", + "@shikijs/themes": "4.4.3", + "@shikijs/types": "4.4.3", + "@shikijs/vscode-textmate": "^10.0.2", + "@types/hast": "^3.0.5" + }, + "engines": { + "node": ">=20" + } + }, + "node_modules/sisteransi": { + "version": "1.0.5", + "resolved": "https://registry.npmjs.org/sisteransi/-/sisteransi-1.0.5.tgz", + "integrity": "sha512-bLGGlR1QxBcynn2d5YmDX4MGjlZvy2MRBDRNHLJ8VI6l6+9FUiyTFNJ0IveOSP0bcXgVDPRcfGqA0pjaqUpfVg==", + "license": "MIT" + }, + "node_modules/smol-toml": { + "version": "1.9.0", + "resolved": "https://registry.npmjs.org/smol-toml/-/smol-toml-1.9.0.tgz", + "integrity": "sha512-hpd+HLON7HdZXqYchMM/+LaTTbdK0AU3NngIJ4KVyWbY9bfQqdL9cD+4yf6dUoU2Ap4VsU0JkQi6FxAI1B2mXQ==", + "license": "BSD-3-Clause", + "engines": { + "node": ">= 18" + }, + "funding": { + "url": "https://github.com/sponsors/cyyynthia" + } + }, + "node_modules/source-map-js": { + "version": "1.2.1", + "resolved": "https://registry.npmjs.org/source-map-js/-/source-map-js-1.2.1.tgz", + "integrity": "sha512-UXWMKhLOwVKb728IUtQPXxfYU+usdybtUrK/8uGE8CQMvrhOpwvzDBwj0QhSL7MQc7vIsISBG8VQ8+IDQxpfQA==", + "license": "BSD-3-Clause", + "engines": { + "node": ">=0.10.0" + } + }, + "node_modules/space-separated-tokens": { + "version": "2.0.2", + "resolved": "https://registry.npmjs.org/space-separated-tokens/-/space-separated-tokens-2.0.2.tgz", + "integrity": "sha512-PEGlAwrG8yXGXRjW32fGbg66JAlOAwbObuqVoJpv/mRgoWDQfgH1wDPvtzWyUSNAXBGSk8h755YDbbcEy3SH2Q==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/stringify-entities": { + "version": "4.0.4", + "resolved": "https://registry.npmjs.org/stringify-entities/-/stringify-entities-4.0.4.tgz", + "integrity": "sha512-IwfBptatlO+QCJUo19AqvrPNqlVMpW9YEL2LIVY+Rpv2qsjCGxaDLNRgeGsQWJhfItebuJhsGSLjaBbNSQ+ieg==", + "license": "MIT", + "dependencies": { + "character-entities-html4": "^2.0.0", + "character-entities-legacy": "^3.0.0" + }, + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/svgo": { + "version": "4.1.0", + "resolved": "https://registry.npmjs.org/svgo/-/svgo-4.1.0.tgz", + "integrity": "sha512-bkxnTg1kSU0guhIBmibA6UUhrQmPVA1XsQLN+ylCd+UWzbnLkySOcXpyk1mrl05f+pcaCx2eHb+sp6BgMZWX+Q==", + "license": "MIT", + "dependencies": { + "commander": "^11.1.0", + "css-select": "^6.0.0", + "css-tree": "^3.0.1", + "css-what": "^7.0.0", + "csso": "^5.0.5", + "picocolors": "^1.1.1", + "sax": "1.6.1" + }, + "bin": { + "svgo": "bin/svgo.js" + }, + "engines": { + "node": ">=16" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/svgo" + } + }, + "node_modules/tiny-inflate": { + "version": "1.0.3", + "resolved": "https://registry.npmjs.org/tiny-inflate/-/tiny-inflate-1.0.3.tgz", + "integrity": "sha512-pkY1fj1cKHb2seWDy0B16HeWyczlJA9/WW3u3c4z/NiWDsO3DOU5D7nhTLE9CF0yXv/QZFY7sEJmj24dK+Rrqw==", + "license": "MIT" + }, + "node_modules/tinyclip": { + "version": "1.0.3", + "resolved": "https://registry.npmjs.org/tinyclip/-/tinyclip-1.0.3.tgz", + "integrity": "sha512-nTv2raKFMzGG6JJp0lpNDsRPdfFE6UHlu2rVEjXo8vkx1ZS3LW2CVXqIaLu8xZlu+6tbFs9V58Qc/Jmk/Ii7wA==", + "license": "MIT", + "engines": { + "node": "^16.14.0 || >= 17.3.0" + } + }, + "node_modules/tinyexec": { + "version": "1.3.1", + "resolved": "https://registry.npmjs.org/tinyexec/-/tinyexec-1.3.1.tgz", + "integrity": "sha512-GCvB3aoys96IuDFBMcTB46JOR6mdMtAToqwiW8JlWhsoh1mhHi/xn9ss/Dg7N555GiJyEt2qzoG/NHCwM6h1EA==", + "license": "MIT", + "engines": { + "node": ">=18" + } + }, + "node_modules/tinyglobby": { + "version": "0.2.17", + "resolved": "https://registry.npmjs.org/tinyglobby/-/tinyglobby-0.2.17.tgz", + "integrity": "sha512-wXR/dYpcqKmfWpEdZjiKJOwCNFndD0DMnrW/cYjVGttEkBfVgcLFHoNrlj47mjOVic9yyNu65alsgF4NQyTa2g==", + "license": "MIT", + "dependencies": { + "fdir": "^6.5.0", + "picomatch": "^4.0.4" + }, + "engines": { + "node": ">=12.0.0" + }, + "funding": { + "url": "https://github.com/sponsors/SuperchupuDev" + } + }, + "node_modules/trim-lines": { + "version": "3.0.1", + "resolved": "https://registry.npmjs.org/trim-lines/-/trim-lines-3.0.1.tgz", + "integrity": "sha512-kRj8B+YHZCc9kQYdWfJB2/oUl9rA99qbowYYBtr4ui4mZyAQ2JpvVBd/6U2YloATfqBhBTSMhTpgBHtU0Mf3Rg==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/trough": { + "version": "2.2.0", + "resolved": "https://registry.npmjs.org/trough/-/trough-2.2.0.tgz", + "integrity": "sha512-tmMpK00BjZiUyVyvrBK7knerNgmgvcV/KLVyuma/SC+TQN167GrMRciANTz09+k3zW8L8t60jWO1GpfkZdjTaw==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + }, + "node_modules/tslib": { + "version": "2.8.1", + "resolved": "https://registry.npmjs.org/tslib/-/tslib-2.8.1.tgz", + "integrity": "sha512-oJFu94HQb+KVduSUQL7wnpmqnfmLsOA/nAh6b6EH0wCEoK0/mPeXU6c3wKDV83MkOuHPRHtSXKKU99IBazS/2w==", + "license": "0BSD", + "optional": true + }, + "node_modules/tsx": { + "version": "4.23.15", + "resolved": "https://registry.npmjs.org/tsx/-/tsx-4.23.15.tgz", + "integrity": "sha512-Yiex1Ovn8z2xPpOWckIiysV1SSyRMY9BkLF++q0yKiDxCqRhosKfMg3janKkiLBwZ5c/YryloKwGZcrEmtwxKw==", + "devOptional": true, + "license": "MIT", + "dependencies": { + "esbuild": "~0.28.0" + }, + "bin": { + "tsx": "dist/cli.mjs" + }, + "engines": { + "node": ">=18.0.0" + }, + "optionalDependencies": { + "fsevents": "~2.3.3" + } + }, + "node_modules/ufo": { + "version": "1.6.4", + "resolved": "https://registry.npmjs.org/ufo/-/ufo-1.6.4.tgz", + "integrity": "sha512-JFNbkD1Svwe0KvGi8GOeLcP4kAWQ609twvCdcHxq1oSL8svv39ZuSvajcD8B+5D0eL4+s1Is2D/O6KN3qcTeRA==", + "license": "MIT" + }, + "node_modules/ultrahtml": { + "version": "1.7.0", + "resolved": "https://registry.npmjs.org/ultrahtml/-/ultrahtml-1.7.0.tgz", + "integrity": "sha512-2xRd0VHoAQE4M+vF/DvFFB7pUV0ZxTW1TLi7lHQWnF/Sb5TPeEUV/l+hxcNnGO00ZXGnR0voCMmYRKQf+rvJ2g==", + "license": "MIT" + }, + "node_modules/uncrypto": { + "version": "0.1.3", + "resolved": "https://registry.npmjs.org/uncrypto/-/uncrypto-0.1.3.tgz", + "integrity": "sha512-Ql87qFHB3s/De2ClA9e0gsnS6zXG27SkTiSJwjCc9MebbfapQfuPzumMIUMi38ezPZVNFcHI9sUIepeQfw8J8Q==", + "license": "MIT" + }, + "node_modules/undici": { + "version": "8.11.0", + "resolved": "https://registry.npmjs.org/undici/-/undici-8.11.0.tgz", + "integrity": "sha512-AdzuGcAkzhbha3GgCSoUBx2P0quwym6qdFNCQz/fDGZc8w9XJ741wHUbt2kTmfPz+ZIOMkySFkiY7RXmiDDC8w==", + "license": "MIT", + "engines": { + "node": ">=22.19.0" + } + }, + "node_modules/unified": { + "version": "11.0.5", + "resolved": "https://registry.npmjs.org/unified/-/unified-11.0.5.tgz", + "integrity": "sha512-xKvGhPWw3k84Qjh8bI3ZeJjqnyadK+GEFtazSfZv/rKeTkTjOJho6mFqh2SM96iIcZokxiOpg78GazTSg8+KHA==", + "license": "MIT", + "dependencies": { + "@types/unist": "^3.0.0", + "bail": "^2.0.0", + "devlop": "^1.0.0", + "extend": "^3.0.0", + "is-plain-obj": "^4.0.0", + "trough": "^2.0.0", + "vfile": "^6.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/unifont": { + "version": "0.7.5", + "resolved": "https://registry.npmjs.org/unifont/-/unifont-0.7.5.tgz", + "integrity": "sha512-ULe/Cs+ZIsq+dcFofNkhqielCrUJnb5mr+Yc4EBM2VlL+6OZR6+cjtI2mT1bJvRBrVncqHAbLURxmPLcCXzWMg==", + "license": "MIT", + "dependencies": { + "css-tree": "^3.1.0", + "ohash": "^2.0.11", + "undici": "^8.0.0" + } + }, + "node_modules/unist-util-is": { + "version": "6.0.1", + "resolved": "https://registry.npmjs.org/unist-util-is/-/unist-util-is-6.0.1.tgz", + "integrity": "sha512-LsiILbtBETkDz8I9p1dQ0uyRUWuaQzd/cuEeS1hoRSyW5E5XGmTzlwY1OrNzzakGowI9Dr/I8HVaw4hTtnxy8g==", + "license": "MIT", + "dependencies": { + "@types/unist": "^3.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/unist-util-position": { + "version": "5.0.0", + "resolved": "https://registry.npmjs.org/unist-util-position/-/unist-util-position-5.0.0.tgz", + "integrity": "sha512-fucsC7HjXvkB5R3kTCO7kUjRdrS0BJt3M/FPxmHMBOm8JQi2BsHAHFsy27E0EolP8rp0NzXsJ+jNPyDWvOJZPA==", + "license": "MIT", + "dependencies": { + "@types/unist": "^3.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/unist-util-stringify-position": { + "version": "4.0.0", + "resolved": "https://registry.npmjs.org/unist-util-stringify-position/-/unist-util-stringify-position-4.0.0.tgz", + "integrity": "sha512-0ASV06AAoKCDkS2+xw5RXJywruurpbC4JZSm7nr7MOt1ojAzvyyaO+UxZf18j8FCF6kmzCZKcAgN/yu2gm2XgQ==", + "license": "MIT", + "dependencies": { + "@types/unist": "^3.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/unist-util-visit": { + "version": "5.1.0", + "resolved": "https://registry.npmjs.org/unist-util-visit/-/unist-util-visit-5.1.0.tgz", + "integrity": "sha512-m+vIdyeCOpdr/QeQCu2EzxX/ohgS8KbnPDgFni4dQsfSCtpz8UqDyY5GjRru8PDKuYn7Fq19j1CQ+nJSsGKOzg==", + "license": "MIT", + "dependencies": { + "@types/unist": "^3.0.0", + "unist-util-is": "^6.0.0", + "unist-util-visit-parents": "^6.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/unist-util-visit-parents": { + "version": "6.0.2", + "resolved": "https://registry.npmjs.org/unist-util-visit-parents/-/unist-util-visit-parents-6.0.2.tgz", + "integrity": "sha512-goh1s1TBrqSqukSc8wrjwWhL0hiJxgA8m4kFxGlQ+8FYQ3C/m11FcTs4YYem7V664AhHVvgoQLk890Ssdsr2IQ==", + "license": "MIT", + "dependencies": { + "@types/unist": "^3.0.0", + "unist-util-is": "^6.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/unstorage": { + "version": "1.17.5", + "resolved": "https://registry.npmjs.org/unstorage/-/unstorage-1.17.5.tgz", + "integrity": "sha512-0i3iqvRfx29hkNntHyQvJTpf5W9dQ9ZadSoRU8+xVlhVtT7jAX57fazYO9EHvcRCfBCyi5YRya7XCDOsbTgkPg==", + "license": "MIT", + "dependencies": { + "anymatch": "^3.1.3", + "chokidar": "^5.0.0", + "destr": "^2.0.5", + "h3": "^1.15.10", + "lru-cache": "^11.2.7", + "node-fetch-native": "^1.6.7", + "ofetch": "^1.5.1", + "ufo": "^1.6.3" + }, + "peerDependencies": { + "@azure/app-configuration": "^1.8.0", + "@azure/cosmos": "^4.2.0", + "@azure/data-tables": "^13.3.0", + "@azure/identity": "^4.6.0", + "@azure/keyvault-secrets": "^4.9.0", + "@azure/storage-blob": "^12.26.0", + "@capacitor/preferences": "^6 || ^7 || ^8", + "@deno/kv": ">=0.9.0", + "@netlify/blobs": "^6.5.0 || ^7.0.0 || ^8.1.0 || ^9.0.0 || ^10.0.0", + "@planetscale/database": "^1.19.0", + "@upstash/redis": "^1.34.3", + "@vercel/blob": ">=0.27.1", + "@vercel/functions": "^2.2.12 || ^3.0.0", + "@vercel/kv": "^1 || ^2 || ^3", + "aws4fetch": "^1.0.20", + "db0": ">=0.2.1", + "idb-keyval": "^6.2.1", + "ioredis": "^5.4.2", + "uploadthing": "^7.4.4" + }, + "peerDependenciesMeta": { + "@azure/app-configuration": { + "optional": true + }, + "@azure/cosmos": { + "optional": true + }, + "@azure/data-tables": { + "optional": true + }, + "@azure/identity": { + "optional": true + }, + "@azure/keyvault-secrets": { + "optional": true + }, + "@azure/storage-blob": { + "optional": true + }, + "@capacitor/preferences": { + "optional": true + }, + "@deno/kv": { + "optional": true + }, + "@netlify/blobs": { + "optional": true + }, + "@planetscale/database": { + "optional": true + }, + "@upstash/redis": { + "optional": true + }, + "@vercel/blob": { + "optional": true + }, + "@vercel/functions": { + "optional": true + }, + "@vercel/kv": { + "optional": true + }, + "aws4fetch": { + "optional": true + }, + "db0": { + "optional": true + }, + "idb-keyval": { + "optional": true + }, + "ioredis": { + "optional": true + }, + "uploadthing": { + "optional": true + } + } + }, + "node_modules/verkit": { + "version": "0.4.1", + "resolved": "https://registry.npmjs.org/verkit/-/verkit-0.4.1.tgz", + "integrity": "sha512-mMVzj0TXExtVdlDEq+Mzp0eyOuyznNpFobNM3uAqe3HVm/Ps2c+u17BligMkZjZCA6EiXpoon8iRTer3iu40SQ==", + "license": "MIT", + "engines": { + "node": ">=18.12.0" + }, + "funding": { + "url": "https://github.com/sponsors/sxzz" + } + }, + "node_modules/vfile": { + "version": "6.0.3", + "resolved": "https://registry.npmjs.org/vfile/-/vfile-6.0.3.tgz", + "integrity": "sha512-KzIbH/9tXat2u30jf+smMwFCsno4wHVdNmzFyL+T/L3UGqqk6JKfVqOFOZEpZSHADH1k40ab6NUIXZq422ov3Q==", + "license": "MIT", + "dependencies": { + "@types/unist": "^3.0.0", + "vfile-message": "^4.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/vfile-message": { + "version": "4.0.3", + "resolved": "https://registry.npmjs.org/vfile-message/-/vfile-message-4.0.3.tgz", + "integrity": "sha512-QTHzsGd1EhbZs4AsQ20JX1rC3cOlt/IWJruk893DfLRr57lcnOeMaWG4K0JrRta4mIJZKth2Au3mM3u03/JWKw==", + "license": "MIT", + "dependencies": { + "@types/unist": "^3.0.0", + "unist-util-stringify-position": "^4.0.0" + }, + "funding": { + "type": "opencollective", + "url": "https://opencollective.com/unified" + } + }, + "node_modules/vite": { + "version": "8.3.0", + "resolved": "https://registry.npmjs.org/vite/-/vite-8.3.0.tgz", + "integrity": "sha512-lhZBVvEHefgE+HQZC9O7EBJgCU/nVzFNl7vkS4RE0APtWLP02/8QVIkQtzBxPquh7lq5/78NHipTj7ODQ6XuyQ==", + "license": "MIT", + "dependencies": { + "lightningcss": "^1.33.0", + "picomatch": "^4.0.7", + "postcss": "^8.5.28", + "rolldown": "~1.2.6", + "tinyglobby": "^0.2.17" + }, + "bin": { + "vite": "bin/vite.js" + }, + "engines": { + "node": "^20.19.0 || >=22.12.0" + }, + "funding": { + "url": "https://github.com/vitejs/vite?sponsor=1" + }, + "optionalDependencies": { + "fsevents": "~2.3.3" + }, + "peerDependencies": { + "@types/node": "^20.19.0 || >=22.12.0", + "@vitejs/devtools": "^0.7.1", + "esbuild": "^0.27.0 || ^0.28.0", + "jiti": ">=1.21.0", + "less": "^4.0.0", + "sass": "^1.70.0", + "sass-embedded": "^1.70.0", + "stylus": ">=0.54.8", + "sugarss": "^5.0.0", + "terser": "^5.16.0", + "tsx": "^4.8.1", + "yaml": "^2.4.2" + }, + "peerDependenciesMeta": { + "@types/node": { + "optional": true + }, + "@vitejs/devtools": { + "optional": true + }, + "esbuild": { + "optional": true + }, + "jiti": { + "optional": true + }, + "less": { + "optional": true + }, + "sass": { + "optional": true + }, + "sass-embedded": { + "optional": true + }, + "stylus": { + "optional": true + }, + "sugarss": { + "optional": true + }, + "terser": { + "optional": true + }, + "tsx": { + "optional": true + }, + "yaml": { + "optional": true + } + } + }, + "node_modules/vitefu": { + "version": "1.1.3", + "resolved": "https://registry.npmjs.org/vitefu/-/vitefu-1.1.3.tgz", + "integrity": "sha512-ub4okH7Z5KLjb6hDyjqrGXqWtWvoYdU3IGm/NorpgHncKoLTCfRIbvlhBm7r0YstIaQRYlp4yEbFqDcKSzXSSg==", + "license": "MIT", + "workspaces": [ + "tests/deps/*", + "tests/projects/*", + "tests/projects/workspace/packages/*" + ], + "peerDependencies": { + "vite": "^3.0.0 || ^4.0.0 || ^5.0.0 || ^6.0.0 || ^7.0.0 || ^8.0.0" + }, + "peerDependenciesMeta": { + "vite": { + "optional": true + } + } + }, + "node_modules/xxhash-wasm": { + "version": "1.1.0", + "resolved": "https://registry.npmjs.org/xxhash-wasm/-/xxhash-wasm-1.1.0.tgz", + "integrity": "sha512-147y/6YNh+tlp6nd/2pWq38i9h6mz/EuQ6njIrmW8D1BS5nCqs0P6DG+m6zTGnNz5I+uhZ0SHxBs9BsPrwcKDA==", + "license": "MIT" + }, + "node_modules/yargs-parser": { + "version": "22.0.0", + "resolved": "https://registry.npmjs.org/yargs-parser/-/yargs-parser-22.0.0.tgz", + "integrity": "sha512-rwu/ClNdSMpkSrUb+d6BRsSkLUq1fmfsY6TOpYzTwvwkg1/NRG85KBy3kq++A8LKQwX6lsu+aWad+2khvuXrqw==", + "license": "ISC", + "engines": { + "node": "^20.19.0 || ^22.12.0 || >=23" + } + }, + "node_modules/yocto-queue": { + "version": "1.2.2", + "resolved": "https://registry.npmjs.org/yocto-queue/-/yocto-queue-1.2.2.tgz", + "integrity": "sha512-4LCcse/U2MHZ63HAJVE+v71o7yOdIe4cZ70Wpf8D/IyjDKYQLV5GD46B+hSTjJsvV5PztjvHoU580EftxjDZFQ==", + "license": "MIT", + "engines": { + "node": ">=12.20" + }, + "funding": { + "url": "https://github.com/sponsors/sindresorhus" + } + }, + "node_modules/zod": { + "version": "4.6.5", + "resolved": "https://registry.npmjs.org/zod/-/zod-4.6.5.tgz", + "integrity": "sha512-v5l/aFXZQeai4awLbOpSoHecE9UiMrnfx75tEXLjNonXVARxQ5mOeipTjROUchszUNCqnE+hqAMujRsRHsut2Q==", + "license": "MIT", + "funding": { + "url": "https://github.com/sponsors/colinhacks" + } + }, + "node_modules/zwitch": { + "version": "2.0.4", + "resolved": "https://registry.npmjs.org/zwitch/-/zwitch-2.0.4.tgz", + "integrity": "sha512-bXE4cR/kVZhKZX/RjPEflHaKVhUVl85noU3v6b8apfQEc1x4A+zBxjZ4lN8LqGd6WZ3dl98pY4o717VFmoPp+A==", + "license": "MIT", + "funding": { + "type": "github", + "url": "https://github.com/sponsors/wooorm" + } + } + } +} diff --git a/web/package.json b/web/package.json new file mode 100644 index 00000000..4dbf0b28 --- /dev/null +++ b/web/package.json @@ -0,0 +1,20 @@ +{ + "name": "howtolivebetter-web", + "version": "1.0.0", + "private": true, + "type": "module", + "description": "《高性价比人生指南》站点。Astro 生成，GitHub Pages 发布。", + "scripts": { + "dev": "astro dev", + "build": "astro build && node scripts/copy-assets.mjs", + "preview": "astro preview", + "check": "tsx src/lib/parity.ts", + "verify": "npm run check && npm run build && tsx src/lib/render-check.ts" + }, + "dependencies": { + "astro": "7.3.4" + }, + "devDependencies": { + "tsx": "4.23.15" + } +} diff --git a/web/scripts/copy-assets.mjs b/web/scripts/copy-assets.mjs new file mode 100644 index 00000000..49ae4324 --- /dev/null +++ b/web/scripts/copy-assets.mjs @@ -0,0 +1,20 @@ +// 构建后把 Pages 要的静态文件拷进 web/dist。 +// +// 为什么拷而不是放 web/public：og.png、ads/ 是仓库根的东西，被 README、 +// EPUB、离线单文件几套构建共用。复制一份进 web/ 就多一处会过时的地方 +// （况且 og.png 每次统计同步都会重截）。构建时从根上拷，源头永远只有一份。 +// .nojekyll 也在这里拷：没有它，Pages 会拿 Jekyll 处理一遍，以下划线开头的 +// 目录会被忽略——现在没有这种目录，但删了它等于把这道保险也删了。 +import { copyFileSync, cpSync, mkdirSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; + +const WEB = resolve(dirname(fileURLToPath(import.meta.url)), '..'); +const ROOT = resolve(WEB, '..'); +const DIST = resolve(WEB, 'dist'); + +mkdirSync(DIST, { recursive: true }); +copyFileSync(resolve(ROOT, 'og.png'), resolve(DIST, 'og.png')); +cpSync(resolve(ROOT, 'ads'), resolve(DIST, 'ads'), { recursive: true }); +copyFileSync(resolve(ROOT, '.nojekyll'), resolve(DIST, '.nojekyll')); +console.log('静态文件已拷进 web/dist：og.png、ads/、.nojekyll'); diff --git a/web/src/content.config.ts b/web/src/content.config.ts new file mode 100644 index 00000000..452dd412 --- /dev/null +++ b/web/src/content.config.ts @@ -0,0 +1,30 @@ +// 内容集合：就地读仓库里已有的 book/*.md，不挪位置（理由见 astro.config.mjs 的 ②）。 +// +// 三个语言各一份集合。集合只收实际存在的文件，「目录里列了但还没翻到」由站点构建 +// 去报（tools/site/build.mjs 的 SKIPPED），不要在这里静默当成没有这一节—— +// 静默了就分不清「翻坏了」和「还没翻」。 +import { defineCollection } from 'astro:content'; +import { glob } from 'astro/loaders'; +import { z } from 'astro/zod'; + +// 每种语言的正文在仓库根的哪：zh 指根（中文是原文，不能搬走），en/vi 指自己的子目录。 +// 这个文件在 web/src/，所以 '../../' 才到仓库根，'../' 只到 web/。 +const up = p => new URL('../../' + p, import.meta.url).pathname; + +const book = base => defineCollection({ + loader: glob({ pattern: '**/*.md', base: up(base + '/book/') }), + schema: z.object({ entries: z.number().optional() }), +}); + +const docs = base => defineCollection({ + loader: glob({ pattern: '*.md', base: up(base + '/docs/') }), +}); + +export const collections = { + book_zh: book(''), + book_en: book('en'), + book_vi: book('vi'), + docs_zh: docs(''), + docs_en: docs('en'), + docs_vi: docs('vi'), +}; diff --git a/web/src/lib/astro-page.ts b/web/src/lib/astro-page.ts new file mode 100644 index 00000000..f5e9fd09 --- /dev/null +++ b/web/src/lib/astro-page.ts @@ -0,0 +1,100 @@ +// Astro 页面组装：共用 tools/site/build.mjs 的壳，再把预渲染和语料塞进去。 +// +// 分工（别串了）： +// · 壳（SEO 头、hreflang、JSON-LD、界面文案、site-config、跳语言页、sitemap、robots） +// 全部由 tools/site/build.mjs 出，Astro 只是调 renderPage/chooserPage/sitemapXml/robotsTxt。 +// 域名、文案、统计数字改那边，这里自动跟上。 +// · 预渲染（卡片、侧栏目录、术语表）和语料内联由 web/ 出（render.ts + 下面这个文件）。 +// 爬虫看到的全文从这里来。 +// +// 注入全走字符串锚点，和 tools/offline/build.mjs 一个路数：锚点找不到就抛错， +// 不要静默发出半成品——少一段预渲染，页面看起来正常但爬虫少看几百条。 +import { renderPage, chooserPage, sitemapXml, robotsTxt, SITE } from '../../../tools/site/build.mjs'; +import { locale, readRepo } from './site'; +import { loadLocale } from './load'; +import { buildGloss } from './parse'; +import { fill } from '../../../tools/site/build.mjs'; +import { createCtx, listHtml, sidebarHtml, glossDl } from './render'; + +function once(html, needle, label) { + const n = html.split(needle).length - 1; + if (n !== 1) throw new Error(`组装 ${label} 时锚点对不上：${JSON.stringify(needle.slice(0, 60))} 出现了 ${n} 次`); + return n; +} + +/** 某一种语言的主页完整 HTML（含预渲染正文 + 内联语料，开箱即发）。 */ +export function localePageHtml(code) { + const cfg = locale(code); + const data = loadLocale(code, cfg); + const template = readRepo('tools/site/page.template.html'); + const stats = JSON.parse(readRepo('tools/site/stats.json')); + const meta = cfg; // locale() 返回 {...meta, ...raw}，renderPage 只读 meta 那几个键 + let html = renderPage({ code: meta.code, dir: meta.dir, contentBase: meta.contentBase, assetBase: meta.assetBase, repoBlob: meta.repoBlob, htmlLang: meta.htmlLang, sortLocale: meta.sortLocale, canonical: meta.canonical, contentDir: meta.contentDir, default: meta.default, accept: meta.accept, name: meta.name, note: meta.note }, template, stats); + if (html === null) throw new Error(`${code} 这次没生成页面（README 还没有？先翻 README）`); + + const strings = Object.fromEntries(Object.entries(cfg.strings).map(([k, v]) => [k, fill(v, stats)])); + const gloss = buildGloss(data.readme, cfg); + const ctx = createCtx(cfg, strings, data.sections, gloss); + const total = data.sections.reduce((n, s) => n + s.entries.length, 0); + + // 语料：和离线单文件同一份东西。hydration 拿它解析出 SECS/ITEMS（给 xref 弹窗）， + // 关掉 JS 的读者和爬虫看的是下面的预渲染，两边是同一份正文。 + const corpusJson = JSON.stringify(data.corpus).replace(/<\/script/gi, '<\\/script'); + const configAnchor = '\n` + configAnchor); + + // 侧栏目录：爬虫顺着这些 #e- 锚点走到每一条。 + once(html, '', '侧栏'); + html = html.replace('', ``); + + // 卡片：塞进 #list，放在 status/empty 之后（和页面里 build() appendChild 的顺序一样）。 + once(html, '
    ', '列表标记'); + html = html.replace('
    ', '
    '); + const listClose = '
    \n
    '; + once(html, listClose, '列表结尾'); + html = html.replace(listClose, listHtml(ctx, data.sections) + '\n' + listClose); + + // 计数：服务端先写对，apply() 跑起来会再算一遍，一样。 + once(html, '–', '计数'); + html = html.replace('–', `${total}`); + once(html, '–', '总数'); + html = html.replace('–', `${total}`); + + // 术语表：页面里 buildGlossary() 塞的 dl，服务端直接写好（hydration 时浏览器只索引）。 + once(html, '
    ', '术语表'); + html = html.replace('
    ', `
    ${glossDl(ctx, gloss)}
    `); + + return { html, sections: data.sections.length, entries: total, skipped: data.skipped }; +} + +/** 把 renderPage 吐出来的整页切成三段，给 .astro 页面用。 + .astro 文件自己写 骨架，中间两块原样塞进去—— + 骨架之外的东西（doctype、html 标签）Astro 不让用 set:html，拆开最干净。 + 三个标记在模板里各出现一次，对不上就抛错，别拼出半页。 */ +export function splitDoc(html) { + const headOpen = html.indexOf(''); + const headClose = html.indexOf(''); + const bodyOpen = html.indexOf(''); + const bodyClose = html.indexOf(''); + if (headOpen < 0 || headClose < 0 || bodyOpen < 0 || bodyClose < 0 + || !(headOpen < headClose && headClose < bodyOpen && bodyOpen < bodyClose)) + throw new Error('整页切分对不上：、、、 的顺序或数量不对'); + const pre = html.slice(0, headOpen); // ...，.astro 那边自己写骨架，不用这段 + const lang = //.exec(pre)?.[1] ?? ''; + if (!lang) throw new Error('整页切分对不上： 没找到'); + return { lang, head: html.slice(headOpen + 6, headClose), body: html.slice(bodyOpen + 6, bodyClose) }; +} + +/** 某一种语言的主页三段（head/body 直接塞进 .astro 骨架）。 */ +export function localePageParts(code) { + const { html, ...rest } = localePageHtml(code); + return { ...splitDoc(html), ...rest }; +} + +/** 跳语言小页三段。 */ +export function chooserPageParts() { + return splitDoc(chooserPage(SITE)); +} + +export { chooserPage, sitemapXml, robotsTxt }; diff --git a/web/src/lib/load.ts b/web/src/lib/load.ts new file mode 100644 index 00000000..19fa1433 --- /dev/null +++ b/web/src/lib/load.ts @@ -0,0 +1,70 @@ +// 一个语言站点的构建输入：README + book/*.md + docs/*.md，解析成节和条目。 +// +// 判据和 tools/site/build.mjs 一样（那边是同一批规则，注释更全）： +// · 目录清单从 README 的「目录」一节里扒，不维护第二份文件名单，新增一节两边自动跟上 +// · 目录里列着、正文还没翻到的节，跳过并报出来，不当成「翻坏了」 +// · 没翻完的语言这次不生成（README 都没有） +import { readRepo, readmePath, at, repoPath } from './site'; +import { parseReadme, parseGlossary } from './parse'; +import { existsSync, readFileSync } from 'node:fs'; + +/** README 的目录一节里列的 book 文件，按目录里的顺序 */ +function listedFiles(readme, cfg) { + const start = readme.indexOf(cfg.contents); + if (start < 0) throw new Error(`${readmePath(cfg.code)} 的目录一节（${cfg.contents}）没找到`); + const rest = readme.slice(start); + const m = rest.match(/\]\((book\/[^)#/]+\.md)\)/); + void m; + const files = []; + for (const hit of rest.matchAll(/\]\((book\/[^)#/]+\.md)\)/g)) files.push(hit[1]); + return [...new Set(files)]; +} + +/** 这一种语言能不能上线：README 在才算 */ +export function hasReadme(code) { + return existsSync(repoPath(readmePath(code))); +} + +/** + * 读一个语言站点的全部内容并解析。 + * @returns {{code, cfg, readme, sections, docs, skipped, files, corpus}} + */ +export function loadLocale(code, cfg) { + const readme = readRepo(readmePath(code)); + const listed = listedFiles(readme, cfg); + + const files = []; + const skipped = []; + for (const f of listed) { + const p = at(code, f); + if (existsSync(repoPath(p))) files.push({ path: f, text: readRepo(p) }); + else skipped.push(f); + } + if (!files.length) throw new Error(`${readmePath(code)} 的目录里列了 ${listed.length} 节，正文一份都还没有`); + + // 页面原来的做法是 parts.join('\n\n') 之后整体解析（init 里读的就是 SITE.files 那几份， + // README 只拿去解析术语表，不进节）。这里照抄：README 里有一行 "### 5. ……" 的盐的例子， + // 进了解析就是一条没人要的孤儿条目——现在两边都掉，不代表以后也不会，入口对齐最保险。 + const sections = parseReadme(files.map(f => f.text).join('\n\n'), cfg); + if (!sections.length) throw new Error(`${readmePath(code)} 解析出来 0 节`); + const glossary = parseGlossary(readme, cfg); + + // 长文：目录里列的、这一份语言实际有的 + const docs = []; + for (const hit of readme.matchAll(/\]\((docs\/[^)#/]+\.md)\)/g)) { + const p = at(code, hit[1]); + if (existsSync(repoPath(p))) docs.push({ path: hit[1], text: readRepo(p) }); + } + + return { + code, cfg, readme, sections, docs, glossary, skipped, files, + corpus: { readme, parts: Object.fromEntries(files.map(f => [f.path, f.text])), docs: Object.fromEntries(docs.map(d => [d.path, d.text])) }, + }; +} + +/** 离线单文件构建用：按 corpusJson 的口径取条目数之类的汇总 */ +export function totals(sections) { + return { entries: sections.reduce((n, s) => n + s.entries.length, 0), sections: sections.length }; +} + +export { readFileSync }; diff --git a/web/src/lib/parity.ts b/web/src/lib/parity.ts new file mode 100644 index 00000000..55c2fd02 --- /dev/null +++ b/web/src/lib/parity.ts @@ -0,0 +1,146 @@ +// 校验：Astro 这边解析出来的结构，和检索页现在用的那份一模一样。 +// +// npm run check +// +// 为什么必须有这个：web/src/lib/parse.ts 是从 page.template.html 里搬过来的副本。 +// 两份各自演化，迟早会对不上；对不上时，爬虫看到的和读者看到的就不是一本书了。 +// 这个脚本不靠「我抄对了」，它把原版 parseReadme 从模板里抠出来、在同一份 +// markdown 上跑一遍，两边逐字段比。任何一条对不上就退出码 1。 +// +// 只比页面真正用来渲染和筛选的字段。原文不用比：两边的语料读的是同一个文件。 +import { readFileSync } from 'node:fs'; +import { LOCALES, locale, readRepo, readmePath, at, repoPath } from './site'; +import { parseReadme, buildGloss } from './parse'; +import { loadLocale, hasReadme } from './load'; +import { existsSync } from 'node:fs'; + +/** 把模板里原版 parseReadme 的源码抠出来，组装成一个能直接调用的函数。 + 原版还依赖几个定义在函数外面的量（SITE、COST_W、fieldRe、RE_COST…RE_NOTE、RE_DISPUTE）， + 那几行也一起抠进来，不然会报 RE_COST is not defined。 */ +function originalParser(templateSrc, cfg) { + // 原版还依赖几个定义在函数外面的东西：fieldRe 和 RE_COST…RE_NOTE、RE_DISPUTE。 + // 只抠这几行声明（按行匹配，不整段切——整段切会把用 location 的 UI 代码带进来）。 + const decl = []; + for (const line of templateSrc.split('\n')) { + if (/^\/\/ 条目字段名各语言不同/.test(line)) continue; // 只是注释 + if (/^const fieldRe = /.test(line)) decl.push(line); + else if (/^const RE_DISPUTE = /.test(line)) decl.push(line); + else if (/^const RE_COST = fieldRe/.test(line)) decl.push(line); + else if (/^ {6}RE_GRADE = fieldRe/.test(line)) decl.push(' ' + line.trim()); + } + if (decl.length < 4) throw new Error(`page.template.html 里只抠到 ${decl.length} 行声明，模板结构可能变了`); + const fn = templateSrc.slice( + templateSrc.indexOf('function parseReadme(md){'), + endOfFunction(templateSrc, templateSrc.indexOf('function parseReadme(md){')), + ); + const src = decl.join('\n') + '\n' + fn; + + // 原版读的是页面里的全局量：SITE 和 COST_W。这里按 locale 配置造一份同名同形状的， + // 被抠出来的那几行代码一个字不改。 + const esc = s => s.replace(/[.*+?^${}()|[\]\\]/g, '\\$&'); + const SITE = { + fields: cfg.fields, dispute: cfg.dispute, todo: cfg.todo, + urlTrim: cfg.urlTrim, xref: cfg.xref, + }; + const COST_W = { money: { '0': 0, '少': 1, '多': 2 }, time: { '少': 0, '中': 1, '多': 2 }, will: { '否': 0, '些': 1, '是': 2 } }; + void esc; + return new Function('SITE', 'COST_W', `${src}; return parseReadme;`)(SITE, COST_W); +} + +/** 从函数体的第一个 { 起，配平括号找到结尾的下标 */ +function endOfFunction(src, start) { + let depth = 0; + for (let i = start; i < src.length; i++) { + const c = src[i]; + if (c === '{') depth++; + else if (c === '}') { depth--; if (depth === 0) return i + 1; } + } + return -1; +} + +const FIELDS = ['sec', 'n', 'title', 'cost', 'human', 'gain', 'grade', 'src', 'note', + 'money', 'time', 'will', 'level', 'lens', 'dispute', 'todo', 'cs', 'ratio', 'hay']; + +let bad = 0; +const report = (code, what, detail) => { + bad++; + console.log(` ✗ ${code} ${what}${detail ? '：' + detail : ''}`); +}; + +const templateSrc = readRepo('tools/site/page.template.html'); +const live = LOCALES.filter(l => hasReadme(l.code)); + +console.log(`解析一致性校验：${live.map(l => l.code).join('、')}\n`); + +for (const meta of live) { + const cfg = locale(meta.code); + const data = loadLocale(meta.code, cfg); + // 运行时真正的输入是正文文件拼起来（README 只给术语表，不进节解析）—— + // 按这个比，loadLocale 里拼什么这里就拼什么，两边入口必须一致。 + const md = data.files.map(f => f.text).join('\n\n'); + + let want; + try { + want = originalParser(templateSrc, cfg)(md); + } catch (err) { + console.log(` ! ${meta.code} 原版 parseReadme 跑不起来：${err.message}`); + report(meta.code, '原版解析', err.message); + continue; + } + + const got = parseReadme(md, cfg); + + // README 里有一行 "### 5. ……" 的盐的例子：它要是哪天进了节解析，两边会一起多出 + // 一节孤儿条目。所以再断言一次「README 加进来也不多出东西」，多出来就说明入口 + // 又对不上了，而不是内容真变了。 + const withReadme = parseReadme([data.readme, ...data.files.map(f => f.text)].join('\n\n'), cfg); + if (withReadme.length !== got.length || withReadme.reduce((n, s) => n + s.entries.length, 0) !== got.reduce((n, s) => n + s.entries.length, 0)) + report(meta.code, 'README 拼入', 'README 拼进来之后节数或条数变了：入口没对齐，孤儿条目漏进了某一边'); + + // 术语表也比：预渲染的 abbr 和运行时 termify 用的必须是同一份行表。 + // 原版 parseGlossary 带副作用（写 GLOSS_RE/GLOSS_BY），这里只取它的返回值比行。 + try { + const glossSrc = templateSrc.slice(templateSrc.indexOf('function parseGlossary(md){'), endOfFunction(templateSrc, templateSrc.indexOf('function parseGlossary(md){'))); + const glossFn = new Function('SITE', `${glossSrc}; return parseGlossary;`)({ glossary: cfg.glossary, glossarySep: cfg.glossarySep, glossarySplit: cfg.glossarySplit }); + const wantGloss = glossFn(data.readme); + const gotGloss = buildGloss(data.readme, cfg).rows; + if (wantGloss.length !== gotGloss.length) report(meta.code, '术语表行数', `新 ${gotGloss.length} vs 原 ${wantGloss.length}`); + else for (let i = 0; i < wantGloss.length; i++) { + const a = wantGloss[i], b = gotGloss[i]; + if (a.term !== b.term || a.meaning !== b.meaning || !!a.latin !== !!b.latin) { report(meta.code, `术语表第 ${i + 1} 行`, `${a.term} vs ${b.term}`); break; } + } + } catch (err) { + report(meta.code, '术语表解析', err.message); + } + + if (got.length !== want.length) { + report(meta.code, '节数', `新 ${got.length} vs 原 ${want.length}`); + continue; + } + let n = 0; + for (let i = 0; i < want.length; i++) { + const a = want[i], b = got[i]; + if (a.n !== b.n || a.title !== b.title) { report(meta.code, `第 ${i + 1} 节`, `${a.n}. ${a.title} vs ${b.n}. ${b.title}`); continue; } + if (a.intro.join('\n') !== b.intro.join('\n')) report(meta.code, `第 ${a.n} 节导读`); + if (a.entries.length !== b.entries.length) { report(meta.code, `第 ${a.n} 节条目数`, `新 ${b.entries.length} vs 原 ${a.entries.length}`); continue; } + for (let j = 0; j < a.entries.length; j++) { + const x = a.entries[j], y = b.entries[j]; + for (const k of FIELDS) { + if (String(x[k]) !== String(y[k])) { + report(meta.code, `第 ${a.n} 节第 ${x.n} 条的 ${k}`, `原 ${JSON.stringify(String(x[k]).slice(0, 60))} vs 新 ${JSON.stringify(String(y[k]).slice(0, 60))}`); + break; + } + } + n++; + } + } + if (!bad) console.log(` ✓ ${meta.code}：${got.length} 节 ${n} 条，19 个字段逐条一致`); +} + +console.log(); +if (bad) { + console.log(`解析对不上 ${bad} 处。web/src/lib/parse.ts 和 page.template.html 的 parseReadme 必须同时改。`); + process.exit(1); +} +console.log('解析一致：站内渲染和构建时预渲染看到的是同一本书。'); +void readmePath; void at; void repoPath; void existsSync; void readFileSync; diff --git a/web/src/lib/parse.ts b/web/src/lib/parse.ts new file mode 100644 index 00000000..baad33f5 --- /dev/null +++ b/web/src/lib/parse.ts @@ -0,0 +1,120 @@ +// 正文解析：从 README 的目录和 book/*.md 里把节和条目拆出来。 +// +// 这段是从检索页原来的 parseReadme() 原样搬过来的（tools/site/page.template.html）， +// 只改了参数：原来读全局的 SITE 和 COEF，函数体一行没动。 +// 搬过来是为了让它在构建时跑——Astro 用它在构建期把卡片渲进 HTML， +// 爬虫和弱网就不用等 JS 了。页面那边还留着同一份逻辑（读内联的正文）， +// 那是给「本地改一点样式立刻看效果」和离线副本兜底的；两边必须同时改， +// 改判据是 tools/site/build.mjs --check 和 web 的 npm run check 都绿。 +// +// 成本权重和档位规则原来写死在页面里（COST_W / e.ratio 两行）， +// tools/sync-stats.mjs 会拿它和那两行比对，这里是第三份，照抄。 + +const COST_W = { money: { '0': 0, '少': 1, '多': 2 }, time: { '少': 0, '中': 1, '多': 2 }, will: { '否': 0, '些': 1, '是': 2 } }; + +/** 字段名在各语言里不同（中文「- 成本：」、英文「- Cost:」），从 locale 配置来，不写死。 + 尾部照抄原版：默认 '(.*)' 会把标签后面的空格一起收进来（页面那边就是这个行为， + 渲染时那点空格看不出来，但要和它逐字段一致就不能改）；证据等级另给一个窄的 + '\\s*([ABC])'，只取字母，「A（争议）」不会连后缀一起进来。 */ +function fieldRe(name, cfg, tail) { + const label = cfg.fields[name].replace(/[.*+?^${}()|[\]\\]/g, '\\$&'); + return new RegExp('^- ' + label + (tail || '(.*)')); +} + +/** + * @param {string} md README + 全部正文拼起来（和原来页面里一样，节和条目的结构都在里面） + * @param {object} cfg locale(code) 的返回值：fields / dispute / todo 是解析用的 + * @returns {Array} sections + */ +export function parseReadme(md, cfg) { + const RE_COST = fieldRe('cost', cfg); + const RE_HUMAN = fieldRe('human', cfg); + const RE_GAIN = fieldRe('gain', cfg); + const RE_GRADE = fieldRe('grade', cfg, '\\s*([ABC])'); + const RE_SRC = fieldRe('src', cfg); + const RE_NOTE = fieldRe('note', cfg); + const RE_DISPUTE = new RegExp('^' + cfg.dispute); + + const lines = md.split(/\r?\n/); + const sections = []; + let sec = null, entry = null; + const flush = () => { if (entry && sec) { sec.entries.push(entry); } entry = null; }; + for (const raw of lines) { + const line = raw.trimEnd(); + let m; + if ((m = /^#{1,2} (\d+)\. (.+)$/.exec(line))) { flush(); sec = { n: m[1], title: m[2].trim(), intro: [], entries: [] }; sections.push(sec); continue; } + if (/^#{1,2} /.test(line)) { flush(); sec = null; continue; } + if (!sec) continue; + if ((m = /^### (\d+)\. (.+)$/.exec(line))) { flush(); entry = { sec: sec.n, n: m[1], title: m[2].trim(), cost: '', human: '', gain: '', grade: '', src: '', note: '', money: '', time: '', will: '', level: '', lens: '' }; continue; } + if ((m = /^/.exec(line)) && entry) { + // 成本标签注释照抄中文（钱/时间/毅力/收益/口径 + 0|少|多、大|中|小、死亡率|金钱|时间|自由）： + // 这些取值就是筛选和地址栏里的值，三种语言共用一份，跨语言共享的链接才对得上 + for (const kv of m[1].split(/\s+/)) { + const [k, v] = kv.split('='); + if (k === '钱') entry.money = v; + if (k === '时间') entry.time = v; + if (k === '毅力') entry.will = v; + if (k === '收益') entry.level = v; + if (k === '口径') entry.lens = v; + } + continue; + } + if (entry) { + if ((m = RE_COST.exec(line))) entry.cost = m[1]; + else if ((m = RE_HUMAN.exec(line))) entry.human = m[1]; + else if ((m = RE_GAIN.exec(line))) entry.gain = m[1]; + else if ((m = RE_GRADE.exec(line))) entry.grade = m[1]; + else if ((m = RE_SRC.exec(line))) entry.src = m[1]; + else if ((m = RE_NOTE.exec(line))) entry.note = m[1]; + continue; + } + if (line) sec.intro.push(line); // 第一条之前的每一段导读都收，不止第一段 + } + flush(); + + for (const s of sections) for (const e of s.entries) { + e.dispute = RE_DISPUTE.test(e.note); + e.todo = cfg.todo.some(w => (e.src + e.gain + e.note + e.cost).includes(w)); + e.cs = (COST_W.money[e.money] ?? 0) + (COST_W.time[e.time] ?? 0) + (COST_W.will[e.will] ?? 0); + e.ratio = e.level === '大' ? (e.cs === 0 ? '极高' : (e.cs <= 2 ? '高' : '一般')) + : e.level === '中' ? (e.cs === 0 ? '高' : '一般') : '一般'; + e.hay = [e.title, e.human, e.cost, e.gain, e.note, e.src, e.grade].join('\n') + .replace(/\*\*/g, '').replace(/\\([*_])/g, '$1').toLowerCase(); + } + return sections; +} + +/** 术语表：原样搬自页面里的 parseGlossary()（tools/site/page.template.html）。 + 注意它和 parseReadme 不一样：字段取值之外，这里连「怎么写正则」都是逻辑的一部分 + （长词优先的最长匹配、拉丁词的前后守卫），抄的时候不能只抄行、把正则重写了。 + 返回行表；附带的 re/by 给渲染时的 termify 用，和页面里 GLOSS_RE/GLOSS_BY 同一份。 */ +export function buildGloss(md, cfg) { + const m = new RegExp('^##\\s*' + cfg.glossary + '[^\\n]*\\n([\\s\\S]*?)(?=^##\\s)', 'm').exec(md); + if (!m) return { rows: [], re: null, by: {} }; + const rows = []; + for (const line of m[1].split(/\r?\n/)) { + const c = /^\|\s*(.+?)\s*\|\s*(.+?)\s*\|$/.exec(line); + if (!c || c[1] === '术语' || /^-+$/.test(c[1])) continue; + for (const t of c[1].split(cfg.glossarySep)) { + const term = t.trim(); + if (!term) continue; + rows.push({ term, meaning: c[2], latin: cfg.glossarySplit ? new RegExp(cfg.glossarySplit).test(term) : /^[A-Za-z0-9%. ]+$/.test(term) }); + } + } + rows.sort((a, b) => b.term.length - a.term.length); + // 合并成一条正则：长词在前保证最长匹配，拉丁术语要求前后不是字母数字 + const escRe = t => t.replace(/[.*+?^${}()|[\]\\]/g, '\\$&'); + const alt = list => list.map(g => escRe(g.term)).join('|'); + const parts = []; + const latin = rows.filter(g => g.latin), cjk = rows.filter(g => !g.latin); + if (latin.length) parts.push(`(? [g.term, g])); + return { rows, re, by }; +} + +/** 行表。渲染之外的用途（比如只数有多少术语）走这个。 */ +export function parseGlossary(md, cfg) { + return buildGloss(md, cfg).rows; +} diff --git a/web/src/lib/render-check.ts b/web/src/lib/render-check.ts new file mode 100644 index 00000000..5acf4f61 --- /dev/null +++ b/web/src/lib/render-check.ts @@ -0,0 +1,148 @@ +// 渲染一致性检查：跑在 Astro 构建产物（web/dist）上，确认爬虫看到的和读者看到的是同一本书。 +// +// npm run verify +// +// 查三件事： +// ① 全文在不在：每条的标题/成本/说人话/收益/备注/来源， normalized 之后都能在 HTML 里找到。 +// normalized = 去标签、解转义、去 **、去 \ 转义、小写、压空白——和页面里 e.hay 一个算法， +// 搜得到 e.hay 的词，爬虫就看得到这条。 +// ② 结构对不对：节块数、卡片数、计数器、侧栏目录、术语表；xref 和目录的 data-keys/data-go +// 指到的 id 都在同一页里；hydration 要用的钩子（data-ssg、__CORPUS__、.t/.cost/…）都在。 +// ③ 没填完的占位符：去掉 site-config 和语料两段 script 之后，不该再有 {{word}}。 +// +// 任何一条不过就退出码 1。SEO 的正确性不能靠「我抄对了」，得靠这个。 +import { readFileSync, existsSync } from 'node:fs'; +import { locale, readRepo } from './site'; +import { loadLocale, hasReadme } from './load'; +import { LOCALES } from './site'; +import { createCtx, renderText, splitSrc } from './render'; +import { buildGloss } from './parse'; +import { fill } from '../../../tools/site/build.mjs'; + +const DIST = new URL('../../dist/', import.meta.url); + +let bad = 0; +const fail = (code, what, detail) => { bad++; console.log(` ✗ ${code} ${what}${detail ? '：' + detail : ''}`); }; + +const norm = s => String(s) + .replace(/<(https?:\/\/[^>\s]+)>/g, '$1') // 是正文里裹链接的写法（页面里 URL_RE 认它），先还原再去标签，不然链接整个被当成标签吃掉 + .replace(/<[^>]*>/g, ' ').replace(/"/g, '"').replace(/&/g, '&').replace(/</g, '<').replace(/>/g, '>') + .replace(/\*\*/g, '').replace(/\\([*_])/g, '$1').replace(/[;；]/g, ' ').toLowerCase().replace(/\s+/g, ' ').trim(); +// 来源栏的裸 URL 渲染时显示文本会去掉 https:// 开头和末尾的 /（页面里 mkLink 那两行正则）， +// href 属性检查侧又看不到（标签都去掉了）。所以字段侧先做同样的剥离再比。 +const normField = s => norm(s).replace(/https?:\/\//g, '').replace(/\/(?=\s|$)/g, ''); + +for (const meta of LOCALES) { + if (!hasReadme(meta.code)) { console.log(` - ${meta.code}：README 还没有，跳过`); continue; } + const cfg = locale(meta.code); + const data = loadLocale(meta.code, cfg); + const p = new URL(`${meta.dir}index.html`, DIST); + if (!existsSync(p)) { fail(meta.code, '构建产物缺失', `${meta.dir}index.html 不在 web/dist 里，先跑 npm run build`); continue; } + const html = readFileSync(p, 'utf8'); + const total = data.sections.reduce((n, s) => n + s.entries.length, 0); + + // 钩子（这两处看整页，含 script） + if (!html.includes('data-ssg="1"')) fail(meta.code, 'hydration 钩子', '没有 data-ssg="1"，浏览器会走老路重建，预渲染白做了'); + if (!html.includes('window.__CORPUS__=')) fail(meta.code, '语料', '没有内联 __CORPUS__，hydration 解析不出 SECS/ITEMS'); + + // 下面全看正文区：script 里有客户端模板字符串（`

    ` 这种）， + // 数个数、找文字都得先去掉，不然模板字符串冒充卡片、语料 JSON 冒充正文—— + // 查的就是「爬虫不执行 JS 能看到什么」，script 里的一律不算。 + const content = html.replace(//g, ''); + if (!html.includes(`${total}`)) fail(meta.code, '计数', `cnt 不是 ${total}`); + if (!html.includes(`${total}`)) fail(meta.code, '总数', `tot 不是 ${total}`); + + // 结构数 + const blocks = (content.match(/
    m[1])); + for (const s of data.sections) { + if (!ids.has(`sec-${s.n}`)) { fail(meta.code, `节锚点 sec-${s.n}`, '没有这个 id，目录跳不过去'); break; } + } + let idBad = 0; + for (const s of data.sections) for (const e of s.entries) { + if (!ids.has(`e-${e.sec}-${e.n}`)) { if (!idBad++) fail(meta.code, '卡片 id', `e-${e.sec}-${e.n} 不在 HTML 里（只报第一处）`); } + } + const goBad = [...content.matchAll(/data-go="([^"]+)"/g)].map(m => m[1]).filter(k => !ids.has(`e-${k}`)); + if (goBad.length) fail(meta.code, '目录指路', `${goBad.length} 处 data-go 指到不存在的卡片：${goBad.slice(0, 3).join('、')}`); + const keyBad = [...content.matchAll(/data-keys="([^"]+)"/g)].flatMap(m => m[1].split(',')) + .filter(k => !(k.startsWith('s') ? ids.has(`sec-${k.slice(1)}`) : ids.has(`e-${k}`))); + if (keyBad.length) fail(meta.code, 'xref 指路', `${keyBad.length} 处 data-keys 悬空：${keyBad.slice(0, 3).join('、')}`); + + // 全文：证明 raw → render → page 整条链不断。分两截查： + // ① render(field) ⊆ page：渲染出来的东西都进了页面（装配没错）。 + // ② raw 的词 ⊆ render(field)：原文的词渲染时都没丢（渲染没错）。 + // 合起来就是「原文每个词都在页面里」，这正是 SEO 要的保证。 + // ② 的词按渲染规则切：md 链接取文字+显示地址、裸 URL 取显示地址（去 scheme 和末尾 /）、 + // ** 和 \ 转义还原；来源栏先按 ;； 切分（渲染时 splitSrc 就是这么拆的，括号里的分号除外， + // 这里多切了也无妨——切出来的词一定更短，一定还在）。 + const displayUrl = u => u.replace(/^https?:\/\//, '').replace(/\/$/, '').replace(/[).,;:，。；：、]+$/, '').toLowerCase(); + // 原文取词。urlsLast（来源栏）时裸链接会被攒到最后，地址和它前后的文字是分开的三块， + // 取词时两边都补空格，不然会造出“前文+地址”粘在一起的词——渲染出来它们根本不挨着。 + // 行内链接（[文字](地址)）不受影响：它永远原地渲染，只取文字。 + const tokensOf = (raw, urlsLast) => String(raw) + .replace(/<(https?:\/\/[^>\s]+)>/g, urlsLast ? ' $1 ' : '$1 ') + .replace(/\[([^\]\n]+)\]\(([^)\s]+)\)/g, '$1 ') + .replace(/\*\*((?:\\\*|[^*\n])+)\*\*/g, '$1') + .replace(/\\([*_])/g, '$1') + .replace(/?/g, m => (urlsLast ? ' ' : '') + displayUrl(m.replace(/^<|>$/g, '')) + ' ') + .toLowerCase().split(/\s+/).map(t => t.trim()).filter(t => t.length >= 4); + const norm0 = s => String(s).replace(/<[^>]*>/g, '').replace(/"/g, '"').replace(/&/g, '&').replace(/</g, '<').replace(/>/g, '>') + .toLowerCase().replace(/\s+/g, ' ').trim(); + const strings = Object.fromEntries(Object.entries(cfg.strings).map(([k, v]) => [k, fill(v, JSON.parse(readRepo('tools/site/stats.json')))])); + const rctx = createCtx(cfg, strings, data.sections, buildGloss(data.readme, cfg)); + const pageText = norm0(content); + let miss = 0; + const fieldsOf = e => [['标题', e.title], ['成本', e.cost], ['说人话', e.human], ['收益', e.gain], ['备注', e.note], ['来源', e.src]]; + for (const s of data.sections) for (const e of s.entries) { + // CUR_SEC 按卡片所属的节走（页面里 renderCard 第一行干的就是这个）： + // 不设的话「本节第 X 条」在检查侧解析不出来，渲染侧却链上了，两边对不上。 + rctx.curSec = e.sec; + // 来源栏按页面真正的 splitSrc 切（括号里的分号不算分隔符），别的栏整段取词。 + // 自己另写一套切法，切口和页面不一样，渲染出来就对不上——之前这么错过。 + const jobs = []; + for (const [name, f] of fieldsOf(e)) { + if (name === '来源') for (const part of splitSrc(rctx, String(f))) jobs.push([name, part]); + else jobs.push([name, f]); + } + for (const [name, f] of jobs) { + const rendered = norm0(renderText(rctx, f, [], true, name === '来源')); + if (rendered && !pageText.includes(rendered)) { if (miss++ < 3) fail(meta.code, `第 ${s.n} 节第 ${e.n} 条${name}没进页面`, JSON.stringify(rendered.slice(0, 50))); continue; } + for (const tok of tokensOf(f, name === '来源')) { + if (!rendered.includes(tok)) { + if (miss++ < 3) { + fail(meta.code, `第 ${s.n} 节第 ${e.n} 条${name}丢词`, JSON.stringify(tok)); + } + break; + } + } + } + if (miss > 10) break; + } + if (!miss) console.log(` ✓ ${meta.code}：${data.sections.length} 节 ${total} 条，原文每个词都在 HTML 里`); + + // 占位符：site-config 和语料两段 script 里允许 {{}}（那是运行时变量），别处不允许 + const shell = content.replace(/ - - - - - - - - - - - - - - - - - -
    -
    - - A High-Return Life Guide - -
    - - - - - -
    -
    -
    - -
    - - - -
    -
    -
    -

    A High-Return Life Guide

    -

    Every item answers two questions: what it costs, and what it buys back.

    -

    Who benefits is graded too: you first, then your spouse and immediate family, then friends and colleagues, strangers last. Last does not mean zero — only that the return is unlikely, and that you have to weigh the risks along with it.

    -

    You do not have to do all of it. This is a shortlist ordered by value for money, not a task list. Taking one or two entries with you counts; come back for the rest when you need them. For the cheapest ones, set "Money spent" to "No money" and "Willpower needed" to "No willpower" on the left. The author has not done most of them either. Dotted "item X" references in the text open in place, so you can see what that item says without hunting for it.

    -
    Showing – shown, out of – in total
    -
    Long-form — Is marriage worth it · Home emergency kit checklist · Should you stop when a stranger is in trouble · Which licences a platform needs · Body clock and night shifts · Source verification records
    -
    AI assistant — Let Claude Code or Codex answer your questions from this book (how to install the skill)
    -
    Abbreviations and terms explained

    Underlined terms in the text explain themselves when you click one or hover over it. The full table is below.

    -
    -
    -
    Reading the text …
    - -
    -
    Which measure a number belongs to is stated inside each item: sections 1 and 2 use all-cause or cause-specific mortality, sections 3 and 4 energy and time, section 5 money, section 7 money and benefits, sections 8 and 9 money and personal freedom. They are not converted into each other. The verification trail is in Source verification records. The text is published under CC BY 4.0 . Credit the author and link the original if you reprint or adapt it.
    -
    -
    - - - - - - - - - - diff --git a/index.html b/index.html deleted file mode 100644 index 4aa63d22..00000000 --- a/index.html +++ /dev/null @@ -1,46 +0,0 @@ - - - - - -高性价比人生指南 · Choose your language - - - - - - - - - -
    -

    高性价比人生指南

    -

    选一种语言 · Choose a language

    - -
    - - - diff --git a/robots.txt b/robots.txt deleted file mode 100644 index e3f951ed..00000000 --- a/robots.txt +++ /dev/null @@ -1,4 +0,0 @@ -User-agent: * -Allow: / - -Sitemap: https://streamentry.github.io/HowToLiveBetter/sitemap.xml diff --git a/sitemap.xml b/sitemap.xml deleted file mode 100644 index bdf53e7b..00000000 --- a/sitemap.xml +++ /dev/null @@ -1,35 +0,0 @@ - - - - https://streamentry.github.io/HowToLiveBetter/ - - - - - 1.0 - - - https://streamentry.github.io/HowToLiveBetter/en/ - - - - - 0.9 - - - https://streamentry.github.io/HowToLiveBetter/vi/ - - - - - 0.8 - - - https://streamentry.github.io/HowToLiveBetter/zh/ - - - - - 0.8 - - diff --git a/tools/sync-stats.mjs b/tools/sync-stats.mjs index 91c9f534..c92c1c46 100644 --- a/tools/sync-stats.mjs +++ b/tools/sync-stats.mjs @@ -150,12 +150,11 @@ if (CHECK) { for (const [file, text] of texts) if (text !== read(file)) writeFileSync(join(ROOT, file), text); if (read(STATS_PATH) !== statsText) writeFileSync(join(ROOT, STATS_PATH), statsText); -// stats.json 变了，三份生成出来的页面跟着变，一起重建，免得本地开的是旧数字 -const siteBuild = spawnSync(process.execPath, [join(ROOT, 'tools', 'site', 'build.mjs'), '--check'], { encoding: 'utf8' }); -if (siteBuild.status !== 0) { - console.log((siteBuild.stdout || '') + (siteBuild.stderr || '')); - throw new Error('tools/site/build.mjs --check 没通过：检索页还没按新统计数字重建。先跑 node tools/site/build.mjs 再提交'); -} +// stats.json 变了，网站下次构建就带上新数字：web.yml 的触发路径含 book/** 和 stats.json +// 所在的 tools/site/**，push 之后 CI 自动重建。不在这里调 Astro 构建—— +// 一是慢（全站预渲染），二是构建产物直发 Pages，不经过仓库，本地跑了也只是看看。 +// 唯一要在本地确认的是解析器没和模板跑偏（改了 COST_W/e.ratio 才需要）： +// cd web && npm run check。 // ② 重算交叉引用对照表：插入或删除条目会让后面的「第 X 条」集体错位，而错位后的条号 // 往往仍在范围内（2026-09-19 第 7 节那 6 处就是），只有把「引用 → 目标标题」摊开入库， diff --git a/vi/index.html b/vi/index.html deleted file mode 100644 index 0c2a6294..00000000 --- a/vi/index.html +++ /dev/null @@ -1,1568 +0,0 @@ - - - - - - -Cẩm Nang Sống Hiệu Quả Cao · Đổi lấy tuổi thọ, tiền và tự do bằng ít tiền, thời gian và sức lực nhất - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    -
    - - Cẩm Nang Sống Hiệu Quả Cao - -
    - - - - - -
    -
    -
    - -
    - - - -
    -
    -
    -

    Cẩm Nang Sống Hiệu Quả Cao

    -

    Mỗi mục trả lời hai câu hỏi: bỏ ra cái gì, đổi lại được cái gì.

    -

    Ai là người hưởng cũng chia thang: bạn cao nhất, kế đến vợ chồng và người thân trực hệ, rồi tới bạn bè và đồng nghiệp, người xa lạ thấp nhất. Thấp nhất không phải là bằng không, chỉ là kỳ vọng hoàn lại nhỏ, và phải xem cả rủi ro cùng lúc.

    -

    Không phải làm hết: đây là danh sách dự phòng đã xếp theo hiệu quả chi phí, không phải danh sách việc phải làm. Lấy một hai mục là được, phần còn lại quay lại tra khi cần. Muốn lấy mục rẻ nhất thì bên trái chọn "Tiền bỏ ra" là "Không tốn tiền", "Ý chí phải có" là "Không cần ý chí". Tác giả cũng chưa làm phần lớn. Mọi "mục X" có gạch chấm trong nội dung đều bấm được ngay tại chỗ, không phải tự lật tìm.

    -
    Đang hiện – đang hiện, trong tổng số – mục
    -
    Bài dài — Kết hôn có đáng không · Danh mục trang bị cứu hộ tại nhà · Có nên dừng lại khi người lạ gặp chuyện · Một nền tảng cần những giấy phép nào · Nhịp sinh học và ca đêm · Hồ sơ đối chiếu nguồn
    -
    Trợ lý AI — Để Claude Code hoặc Codex trả lời câu hỏi của bạn theo cuốn sách này (cách cài skill)
    -
    Chữ viết tắt và thuật ngữ

    Từ có gạch chân trong nội dung, bấm vào hoặc đưa chuột vào là hiện nghĩa. Toàn bộ bảng ở dưới.

    -
    -
    -
    Đang đọc nội dung …
    - -
    -
    Mỗi con số thuộc thước đo nào thì ghi ngay trong mục đó: mục 1 và 2 là tử vong toàn nguyên hoặc theo nguyên nhân, mục 3 và 4 là sức lực và thời gian, mục 5 là tiền, mục 7 là tiền và trợ cấp, mục 8 và 9 là tiền và tự do cá nhân. Chúng không quy đổi qua lại. Quá trình đối chiếu nguồn nằm ở Hồ sơ đối chiếu nguồn. Nội dung phát hành theo CC BY 4.0 . Nếu in lại hoặc sửa đổi, xin ghi nguồn và kèm liên kết bản gốc.
    -
    -
    - - - - - - - - - - diff --git a/zh/index.html b/zh/index.html deleted file mode 100644 index aaad9d7a..00000000 --- a/zh/index.html +++ /dev/null @@ -1,1601 +0,0 @@ - - - - - - -高性价比人生指南 · 用最少的钱、时间和精力换回寿命、金钱和人身自由 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    -
    - - 高性价比人生指南 - -
    - - - - - -
    -
    -
    - -
    - - - -
    -
    -
    -

    高性价比人生指南

    -

    每一条都回答两个问题：花掉什么，换回什么。

    -

    换回的算在谁头上也分档：你自己最高，其次配偶和直系亲属，再次朋友同事，陌生人最低——最低不等于零，只是回报的指望小、要连风险一起看。

    -

    不用全做：这是一份按性价比排好的备选单，不是任务清单。挑走一两条就算数，剩下的需要时再回来查——想挑省力的，左边把「花钱」选「不花钱」、「要毅力」选「不用」，剩下的就是。作者自己也没做到其中大部分。正文里带虚线的「第 X 条」可以点开，就地看那条写了什么，不用自己翻过去。

    -
    当前显示 – 条，共 – 条
    -
    长文：结婚划不划算 · 家庭应急装备清单 · 遇到陌生人出事该不该停 · 做平台要办哪些证 · 生物钟和夜班 · 来源核实记录
    -
    AI 助手：让 Claude Code 或 Codex 照这本书回答你的问题（skill 装法）
    -
    看不懂的缩写和名词

    正文里带虚线的词，点一下或把鼠标放上去就有解释。全表如下。

    -
    -
    -
    正在读取正文 …
    - -
    -
    数字口径以条目内标注为准：第 1、2 节是 ACM 或特定死因，第 3、4 节是精力/时间，第 5 节是金钱，第 7 节是金钱与保障，第 8、9 节是金钱与人身自由，互不换算。来源核实过程见 来源核实记录。正文按 CC BY 4.0 发布，转载改编请署名并附原文链接。
    -
    -
    - - - - - - - - - - From 6c8f972a221adb1222828c2d475366589fb5469e Mon Sep 17 00:00:00 2001 From: Stream Entry <978862+streamentry@users.noreply.github.com> Date: Wed, 30 Sep 2026 19:25:01 +0700 Subject: [PATCH 08/25] =?UTF-8?q?=E7=BD=91=E7=AB=99=E5=8F=91=E5=B8=83?= =?UTF-8?q?=E6=94=B9=E6=88=90=E5=85=A8=E8=87=AA=E5=8A=A8=EF=BC=9Apush=20?= =?UTF-8?q?=E5=88=B0=20main=20=E5=B0=B1=E6=9E=84=E5=BB=BA=E5=8F=91?= =?UTF-8?q?=E5=B8=83=EF=BC=8C=E4=B8=8D=E5=86=8D=E6=8C=89=E8=B7=AF=E5=BE=84?= =?UTF-8?q?=E8=BF=87=E6=BB=A4=20(#4)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 之前按路径触发，改了触发路径之外的文件（比如 skills/、CLAUDE.md）， 网站就停在旧版——还得记得手点 workflow_dispatch，等于半自动。 改成 push 到 main 全触发：构建加发布一共一两分钟，内容没变时产物也一样， 多跑一次不亏。PR 保持只构建不发布。 Co-authored-by: j --- .github/workflows/web.yml | 16 +++------------- 1 file changed, 3 insertions(+), 13 deletions(-) diff --git a/.github/workflows/web.yml b/.github/workflows/web.yml index 5321e6a1..93fa9b5d 100644 --- a/.github/workflows/web.yml +++ b/.github/workflows/web.yml @@ -4,24 +4,14 @@ # 这里管网站。两边各管一头，互不挡——正文写错了，网站照发（内容是旧的但能看）； # 网站构建坏了，电子书照发。 # -# 触发路径只收和网站有关的东西：web/ 本身、正文（book/、en/、vi/、README、 -# docs 长文）、认正文用的配置和检查（tools/site、tools/i18n）。 -# 改 EPUB/PDF 模板不触发这里，改 web 也不触发 book.yml 的重型构建。 +# 触发：push 到 main 就自动构建发布，不再按路径过滤——改了什么都跟着发， +# 免得正文更新了而网站没动。PR 只构建不发布；改 EPUB/PDF 模板不触发这里 +#（book.yml 管那一头）。 name: 网站 on: push: branches: [main] - paths: - - web/** - - book/** - - en/** - - vi/** - - README.md - - docs/*.md - - tools/site/** - - tools/i18n/** - - .github/workflows/web.yml pull_request: paths: - web/** From f8fcee3eb6cf69af13209be9b38d5d69a2f99f40 Mon Sep 17 00:00:00 2001 From: Stream Entry <978862+streamentry@users.noreply.github.com> Date: Wed, 30 Sep 2026 19:25:05 +0700 Subject: [PATCH 09/25] Update README.md --- README.md | 3 +++ 1 file changed, 3 insertions(+) diff --git a/README.md b/README.md index be3c0267..bbcc768c 100644 --- a/README.md +++ b/README.md @@ -2,6 +2,9 @@ 高性价比人生指南 —— 用最少的钱、时间和精力，换回最多的寿命、金钱和人身自由 +> Vietnamese: https://streamentry.github.io/HowToLiveBetter/vi/ +> English: https://streamentry.github.io/HowToLiveBetter/en/ + # 高性价比人生指南 讲怎么活得久、怎么少生病，出了意外怎么救。讲怎么少花冤枉钱，哪些事会让人被骗、摊上官司。讲没工作没钱时能去领什么，开店、开公司、做网站要办什么手续。也讲恋爱结婚生孩子、出国和学手艺。
    From 149038d41a22d01a1d4decfe48c4831c4943981e Mon Sep 17 00:00:00 2001 From: Stream Entry <978862+streamentry@users.noreply.github.com> Date: Thu, 1 Oct 2026 06:56:31 +0700 Subject: [PATCH 10/25] =?UTF-8?q?=E8=8B=B1=E6=96=87=E7=AC=AC=2014=20?= =?UTF-8?q?=E5=88=B0=2023=20=E8=8A=82=E7=BF=BB=E5=AE=8C=EF=BC=9B=E6=A3=80?= =?UTF-8?q?=E7=B4=A2=E9=A1=B5=E7=94=9F=E6=88=90=E7=89=A9=E9=80=80=E5=87=BA?= =?UTF-8?q?=E4=BB=93=E5=BA=93=20(#5)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * 英文第 8 到 13 节翻完并合入 法律（8、9、11）、恋爱结婚（10）、创业（12）、紧急情况（13）六节， 法律引文按规矩留中文原文再给白话，条号节号一律照抄。 英文 13/34 节。剩下的在 .i18n/en/ 的块里，翻完 join 一次带一节进页面。 * 英文第 14 到 23 节翻完；检索页生成物退出仓库 正文（十节）：账号与信息安全、租房与买房、慢性病、家里有老人、 养孩子、在职离职和工伤、新生儿、出国旅行、放松、学技能。 生成物那一半是补上一笔旧账：上一条提交（25e9c41）在跑 build.mjs 之后 git add -A，把 Astro 构建已经不再入库的六份页面（/、/en/、/zh/、/vi/、 sitemap.xml、robots.txt）又带回了仓库。3fb31f4 已经把它们退役， CLAUDE.md「目录结构」也写明产物直发 Pages 不入库。这次删掉并写进 .gitignore，留着它们只会变成「改了模板忘了重建」的过时副本。 校验：i18n 检查、extract --check、check-refs --check、check-plain、 sync-stats --check 全过；web/ 的 npm run verify 三段全过 （en 42 节 501 条、zh 34 节 641 条、vi 1 节 38 条，原文每个词都在 HTML 里）。 --------- Co-authored-by: j --- .gitignore | 14 +- ...25\345\276\213\347\272\242\347\272\277.md" | 218 ++++++++++ ...22\344\270\215\345\210\222\347\256\227.md" | 191 +++++++++ ...51\347\232\204\347\272\242\347\272\277.md" | 164 ++++++++ ...16\345\201\232\347\224\237\346\204\217.md" | 218 ++++++++++ ...47\346\200\245\346\203\205\345\206\265.md" | 389 ++++++++++++++++++ ...41\346\201\257\345\256\211\345\205\250.md" | 92 +++++ ...77\344\270\216\344\271\260\346\210\277.md" | 92 +++++ ...16\346\200\216\344\271\210\346\264\273.md" | 92 +++++ ...14\346\234\211\350\200\201\344\272\272.md" | 83 ++++ ...22\344\270\215\345\210\222\347\256\227.md" | 65 +++ ...14\345\222\214\345\267\245\344\274\244.md" | 164 ++++++++ ...20\346\200\216\344\271\210\345\270\246.md" | 119 ++++++ ...03\345\244\226\345\256\211\345\205\250.md" | 110 +++++ ...16\344\271\210\346\224\276\346\235\276.md" | 110 +++++ ...00\350\203\275\345\210\222\347\256\227.md" | 218 ++++++++++ 16 files changed, 2338 insertions(+), 1 deletion(-) create mode 100644 "en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" create mode 100644 "en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" create mode 100644 "en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" create mode 100644 "en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" create mode 100644 "en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" create mode 100644 "en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" create mode 100644 "en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" create mode 100644 "en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" create mode 100644 "en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" create mode 100644 "en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" create mode 100644 "en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" create mode 100644 "en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" create mode 100644 "en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" create mode 100644 "en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" create mode 100644 "en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" diff --git a/.gitignore b/.gitignore index bdc490d6..cabc5767 100644 --- a/.gitignore +++ b/.gitignore @@ -7,5 +7,17 @@ dist/ # 这份就没用了，join.mjs 每次都从原文重切。重切不覆盖已翻的块，所以它是活的翻译进度。 .i18n/ -# Astro 生成物：本地构建缓存，与 dist/、node_modules/ 一样不入库 +# Astro 生成物：本地和 CI 的构建缓存，由 web/ 的构建出，不入库。 +# 站点直发 GitHub Pages，产物不进仓库（CLAUDE.md「目录结构」一节） web/.astro/ +web/dist/ + +# 旧的手写页面：2026-09-30 切到 Astro 构建后由构建产出，不再入库。 +# tools/site/build.mjs 仍然是库（离线单文件和 Astro 都调它），但它写盘的那些 +# 页面不再跟踪——留着会变成「改了模板忘了重建」的过时副本。 +/index.html +/en/index.html +/zh/index.html +/vi/index.html +/sitemap.xml +/robots.txt diff --git "a/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" "b/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" new file mode 100644 index 00000000..13ec789f --- /dev/null +++ "b/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" @@ -0,0 +1,218 @@ +[← 回总目录](../README.md) + +# 9. 普通人容易踩的法律红线 + +这一节算两笔账：你的人身自由，和你的钱。下面这些事，普通人常以为「没什么」，实际会被拘留、罚款，重的还会判刑。每条都给出法条里的具体数字，以及官方渠道公布过的案例。本节引的治安管理处罚法是 2025 年修订版，2026 年 1 月 1 日起施行，条款号和旧法不一样。刑法引的是北京市公安交管局转载的「根据刑法修正案（十一）修正」的整合文本，正文里会注明对应的修正案。条目的顺序，按「最容易不小心踩到」加「后果最重」从高到低排。 + +[← Back to the contents](../README.md) + +# 9. Legal lines ordinary people cross + +This section counts two things: your personal freedom and your money. In the items below, ordinary people commonly think 「it is nothing」, and in fact end up with detention, a fine, and in the heavier cases a prison sentence. Each item gives the specific figures in the articles of law, together with the cases published through official channels. The Law on Penalties for Administration of Public Security cited in this section is the version revised in 2025 and in force from 2026-1-1, and its article numbers differ from those of the old law. The Criminal Law cited here is the consolidated text as amended by Criminal Law Amendment Eleven, reprinted by the Beijing Municipal Public Security Traffic Management Bureau, and each entry says which amendment is involved. The items are ordered from high to low by 「easiest to step on by accident」 plus 「heaviest consequences」. + +### 1. Do not repost disaster, epidemic or police messages of unknown truth in a group chat, do not doctor photographs, and do not use AI to fake a scene + +- Cost: No money. Before you repost, look at it and see whether it is an official notice. If you are unsure, do not repost it, and do not doctor the picture or add 「it is said that」. The hard part is holding out against the urge to be the first to pass on the news. +- In plain language: For a disaster, epidemic or police message, knowingly passing it on when you know it is false is punished as heavily as writing it yourself. The same goes when you did not write it yourself: 5 to 10 days of detention and a fine of 1000 yuan. Disturbing public order carries 3 years or less, and serious consequences carry 3 to 7 years. Doctored pictures and fake scene images made with AI both count as writing it yourself. +- Benefit: Deliberately spreading a rumour, or falsely reporting a danger, an epidemic, a disaster or a police incident, and disturbing public order, carries 5 to 10 days of detention, and a further fine of up to 1000 yuan may be imposed. The lighter band is up to 5 days of detention, or a fine of up to 1000 yuan. Fabricating a false danger, epidemic, disaster or police incident and putting it online for spread carries 3 years or less. Knowingly passing on something false as though it were true falls in the same band. The condition is that public order is badly disturbed, which the article words as 「严重扰乱社会秩序」. Where serious consequences result, it carries 3 to 7 years (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第二十九条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（第三十二条，即刑法第二百九十一条之一第二款）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百九十一条之一）. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau) +- Notes: Knowingly reposting something false and writing it yourself are dealt with by the police under the same article. 「I was only reposting it」 is no excuse. Doctored pictures and fake scene images generated by AI count as writing it yourself in the same way. The Ministry of Public Security published several batches of typical cases of online rumours about disasters and dangers in 2025. The Ministry's website would not open while this section was being written, so no specific case is cited here. + +### 2. Do not post or repost pictures, stickers or comments that insult martyrs, and do not make fun of the national flag or the national anthem + +- Cost: No money. On content involving martyrs, the national flag, the national emblem or the national anthem: do not comment, do not doctor the picture, do not repost. The hard part is holding back from joining in when somebody in the group continues a joke. +- In plain language: Making fun of martyrs, the national flag or the national anthem carries 5 to 10 days of detention, or a fine of 1000 to 3000 yuan. The heavier cases carry 10 to 15 days of detention. Where it amounts to a crime it carries 3 years or less. One man posted 2 messages defaming Yuan Longping in a 499-member villagers' WeChat group and was sentenced to 6 months' public surveillance: he was not sent to prison, but his movement was restricted. +- Benefit: Where a person's name, likeness, reputation or honour as a hero or martyr is infringed by means such as abuse or fabricating facts, and the public interest is harmed, the band below applies: 5 to 10 days of detention, or a fine of 1000 to 3000 yuan. Desecrating or denying the deeds and spirit of martyrs, or producing and spreading remarks, pictures, audio or video that glorify a war of aggression, and disturbing public order, falls in the same band. The heavier band is 10 to 15 days of detention, and a further fine of up to 5000 yuan may be imposed. Where it reaches 「情节严重」, it amounts to a crime and carries 3 years or less. Insulting the national flag or the national emblem in a public place, altering the words or music of the national anthem, or singing the anthem in a distorted and disparaging way, where the circumstances are serious, also carries 3 years or less. The official case: Xiao posted 2 messages defaming and insulting Yuan Longping in a 499-member villagers' WeChat group and was sentenced to 6 months' public surveillance, which is a punishment that does not mean going to prison but does restrict your movement. Qiu used a Weibo account with 250 余万 followers to post 2 pieces of content distorting the deeds of the soldiers guarding the border, and was sentenced to 8 months, and had to make a public apology in national media (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第三十五条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Law on the Protection of Heroes and Martyrs, 英雄烈士保护法（第二十二、二十六条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第三十五条，即刑法第二百九十九条之一）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百九十九条、第二百九十九条之一）. ; Supreme People's Court (2022). Ten Typical Cases Involving the Protection of the Rights and Interests of Martyrs (Case Two, Case Three), 涉英烈权益保护十大典型案例（案例二、案例三）. +- Notes: This article of the Law on Penalties for Administration of Public Security was only added in the 2025 revision. Before that, the only routes were the Criminal Law, or a civil public interest litigation brought by the procuratorate. In the eyes of a court, a WeChat group of several hundred people counts as a public place. + +### 3. Close the page once you have read something on an overseas site: do not screenshot it, repost it or move it into a group chat, and do not join in under the comments + +- Cost: No money. Close the page and the matter is finished. Do not archive it, do not repost it, do not move it onto a domestic platform. Do not use an account that can be matched to your own identity to post political content on an overseas platform over the long term. Do not accept fees or funding from overseas institutions, organisations or individuals. The hard part is holding back from screenshotting and sharing content that startles you. +- In plain language: These provisions punish passing things on, not looking at them. Reposting, moving into a group chat and joining in under the comments all count as spreading when a case is handled. Where it goes as far as inciting subversion or inciting division, it carries 5 years or less, and 5 years or more for those who lead it. Books and audio or video about terrorism and extremism are the exception: knowingly still holding them, where the circumstances are serious, carries 3 years or less. +- Benefit: Using rumour, defamation or any other means to incite others to subvert the state power of the country or to overthrow the socialist system carries 5 years or less. Those who lead it, whom the article calls 「首要分子」, and those whose crimes are serious, carry 5 years or more, with fixed-term imprisonment reaching at most 15 years. Inciting the division of the country or undermining national unity carries the same terms. Where this is done in collusion with an overseas institution, organisation or individual, the punishment is heavier. Where an institution, organisation or individual inside or outside the country pays to fund somebody else's conduct, the person directly responsible carries 5 years or less, and where the circumstances are serious, 5 years or more. Stealing, spying on, buying or illegally providing state secrets or intelligence to an overseas institution, organisation or person carries 5 to 10 years. In this item, especially serious circumstances carry 10 years or more or life imprisonment, and relatively minor circumstances carry 5 years or less. Crimes endangering national security all carry an additional deprivation of political rights of 1 to 5 years, during which the right to vote and the right to hold public office are taken away. A fine and confiscation of property may also be imposed, taking the person's property into state ownership. There are provisions on administrative penalties as well. No individual or organisation may use the internet to carry out activities endangering national security, honour and interests. Nor may they incite subversion of the state power of the country or overthrow of the socialist system. Nor may they incite the division of the country or the undermining of national unity. Whoever publishes or transmits this kind of information is punished under the relevant laws and administrative regulations (nationwide). There is also a category that is complete without any spreading at all. Knowingly and illegally holding material that promotes terrorism or extremism, where the circumstances are serious, is enough for conviction. The material here includes books, audio and video materials, or other items. The sentence is up to 3 years' fixed-term imprisonment, criminal detention, which is short-term confinement, or public surveillance, which is confinement-free but restricts your movement. A fine may be imposed in addition or instead, so the fine can come with the sentence or on its own. The basis is Article one hundred and twenty of the Criminal Law, item six, added by Criminal Law Amendment Nine +- Evidence grade: A +- Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第四十五、五十五、五十六、一百零三、一百零五、一百零六、一百零七、一百一十一、一百一十三条，这几条未经修正案修改）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第十三条第二款、第七十一条第一项）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（新增第一百二十条之六）, full text reprinted by the National Petition Office. +- Notes: The provisions cited in this item punish inciting others, passing things on, and providing things to other people. Merely looking is not among them. Material about terrorism and extremism is the exception written into the article, and simply holding it can already amount to a conviction (Article one hundred and twenty, item six), so this category must never be stored. Once you have seen it, that is where it ends: do not screenshot and store it, do not repost it, do not add a comment, and do not pull other people into reposting it. Both reposting and joining in under the comments count as spreading when a case is handled, and 「I was only reposting it」 is no excuse. Content that comes from an overseas platform may also attract a heavier penalty. If you are unsure whether something is true, do not join in either: see item 1 in this section (do not repost messages of unknown truth). For how the circumvention tools themselves are punished, see item 11 in section 11. Passing this kind of content on brings you no practical good, while the price is a prison sentence, deprivation of political rights and confiscation of property. Publicly checkable cases of this kind appear only in overseas media and in documents released by families, such as a blogger who had long posted political commentary on an overseas platform being sentenced to 7 years in 2023. This book has not obtained a judgment that can be checked through official channels, so it writes from the articles alone and attaches no case. + +### 4. Pornographic video is yours alone to watch: do not post it in a group chat, do not sell 「the material」, and do not set up a group + +- Cost: No money. Do not repost it, do not save it into a group chat's files, and do not become the owner or an administrator of that kind of group. +- In plain language: Watching it yourself is not within the penal provisions. Posting it into a group chat once counts as spreading it to the whole group, and carries 10 to 15 days of detention and a further fine of up to 5000 yuan; where the circumstances are serious it carries 2 years or less. Selling 「the material」 to make money carries 3 years or less, and at the top life imprisonment. Setting up a distribution group of more than 30 people, the person who sets it up, the administrators, and the people who mainly post into it are all treated as committing the crime. +- Benefit: Spreading obscene information online carries 10 to 15 days of detention, and a further fine of up to 5000 yuan may be imposed. The lighter band is up to 5 days of detention, or a fine of 1000 to 3000 yuan. Where the material involves minors, the punishment is heavier. Spreading obscene material at the level of 「情节严重」 carries 2 years or less. Selling or spreading it in order to make money carries 3 years or less and a fine. In this item, serious circumstances carry 3 to 10 years, and especially serious circumstances carry 10 years or more or life imprisonment. Setting up a group whose main purpose is spreading obscene electronic information, with 30 or more members, is handled as a crime. Causing serious consequences is the same. The person who set the group up, the administrators, and the people who mainly posted into it are all convicted of the crime of spreading obscene material (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第八十条）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第三百六十三、三百六十四条）. ; Supreme People's Court and Supreme People's Procuratorate (2010). Interpretation (Two) on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Producing, Copying, Publishing, Selling or Spreading Obscene Electronic Information by Means of the Internet, Mobile Communication Terminals and Voice Messaging Platforms（第三条）. +- Notes: Both of these laws punish 「spreading」, and watching privately is not within them. But posting into a group chat once already counts as spreading to the whole group. What 「情节严重」 means in terms of quantity is set by the judicial interpretation, and this section has not verified the specific figures. + +### 5. If a 「side job」 wants you to receive money, take out cash or make transfers through your own card, do not do it whatever fee you are offered + +- Cost: No money. Turn down that kind of 「a few hundred a day for carrying a card to the bank」 work. +- In plain language: This kind of side job earns a few hundred yuan, and the price is a criminal record. One man went to an agreed place with his identity card and 1 bank card, helped with the transfers and withdrawals, and earned a little over 500 yuan. The procuratorate did not prosecute him, but the police still detained him for 10 days and fined him 1000 yuan. Another handed his own and other people's bank cards and phone cards to a gang to receive money and take out cash, and was sentenced to 2 years and 3 months with a fine of 22,000 yuan. +- Benefit: Knowing that money and goods come from a crime and still helping to transfer, buy or sell them, or helping to conceal them in some other way, carries 3 years or less, with a fine imposed in addition or instead, so the fine can come with the sentence or on its own. The official case: Wang, on instructions, took his identity card and 1 bank card to an agreed place and helped with the transfers and withdrawals, and earned a little over 500 yuan. The procuratorate did not prosecute him, but the police detained him for 10 days under the Law on the Prevention and Control of Telecom and Online Fraud and fined him 1000 yuan. Man handed his own bank cards and phone cards, and those of other people, to the gang to receive money, transfer it and take out cash, and was sentenced to 2 years and 3 months with a fine of 22,000 yuan (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第三百一十二条）. ; Supreme People's Procuratorate (2023). Typical Cases of Procuratorates Punishing Telecom and Online Fraud and Related Crimes According to Law (Case Nine), 检察机关依法惩治电信网络诈骗及其关联犯罪典型案例（案例九）. ; Supreme People's Court and Supreme People's Procuratorate (2025). Typical Cases of Punishing the Crime of Concealing or Disguising Criminal Proceeds and the Proceeds of Crime According to Law (Case Five), 依法惩治掩饰、隐瞒犯罪所得、犯罪所得收益犯罪典型案例（案例五）. +- Notes: For the consequences of lending your bank card to somebody and of passing transactions through it for them, see item 8 in section 8 (lending a bank card to somebody). This item is about taking out cash and making transfers yourself. 「I did not know where the money came from」 is not necessarily accepted when a case is handled. A fee that is absurdly high, being asked to verify with your face, and transactions that look abnormal however you look at them are enough between them to establish that you did in fact know. The lightest outcome is still 10 days of detention and a criminal record. + +### 6. If somebody draws you into 「packaging the materials」 for a loan and offers you a cut of the loan amount, do none of it + +- Cost: No money. Simply say no. Do not lend your identity card or your own credit record to somebody else, do not sign a loan contract you have not read, and do not go to the bank to sign on somebody else's behalf. +- In plain language: Helping somebody to 「package the materials」 to get a loan and taking a share of it is handled as fraud. In one case a gang of 126 people packaged people with no job as 「high-quality customers」 and fraudulently obtained more than 30 million yuan in loans from more than 80 banks. The people who came forward to sign took three tenths of the loan amount as a fee, and of the 80 people sentenced, 76 got 1 year and 4 months to 6 years and 6 months. The debt belongs to whoever the loan is in the name of. +- Benefit: Obtaining loans, bill acceptances, letters of credit, guarantees and the like from a bank or another financial institution by deceptive means. Where this causes major losses to the bank or another financial institution, it carries 3 years or less. A fine is imposed in addition or instead, so the fine can come with the sentence or on its own. Where especially major losses result, or where there are other especially serious circumstances, it carries 3 to 7 years and a fine. The threshold at which the police open a case and start to investigate is a direct economic loss of 500,000 yuan or more. Where the intention from the start was not to repay, and the loan is obtained by fabricating a reason for bringing in capital, a false economic contract, false supporting documents or a false title deed, that is the crime of loan fraud. In this item, obtaining 50,000 yuan is enough to open a case, and it carries 5 years or less with a fine of 20,000 to 200,000 yuan. An especially large amount, or other serious circumstances, carries 5 to 10 years, and an especially especially large amount carries 10 years or more up to life imprisonment. Forging, altering, buying or selling official documents, certificates or seals of state organs carries 3 years or less, and where the circumstances are serious, 3 to 10 years. The official case: a gang of 126 people packaged people with no job as 「high-quality customers」 across 16 provinces and cities, and fraudulently obtained more than 3000 余万元 in loans from more than 80 bank branches. The 「blank customers」 who came forward to take out the loans took three tenths of the loan amount as a fee. The procuratorate pursued 105 of the 「blank customers」. Of the 80 people sentenced at first instance in 2026, 76 were sentenced for the crime of loan fraud to 1 year and 4 months to 6 years and 6 months, together with a fine. Another 4 people, because they took out cash for the gang, were convicted of loan fraud and of money laundering together, with the two crimes worked out as one sentence. Those 4 were sentenced to 2 years and 4 months to 7 years and 6 months (nationwide). The crime of fraudulently obtaining loans in this item follows the text of the Criminal Law as it has stood since 2021-3 under Criminal Law Amendment Eleven +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第十一条，修改刑法第一百七十五条之一）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第八十八、一百九十三、二百八十条）. ; Supreme People's Procuratorate and Ministry of Public Security (2022). Provisions (Two) on the Standards for Opening Cases and Pursuing Criminal Cases under the Jurisdiction of Public Security Organs, 关于公安机关管辖的刑事案件立案追诉标准的规定（二）（第二十二、四十五条）. ; Supreme People's Procuratorate (2026). 骗贷流水线背后藏洗钱黑手（检察日报）. ; Standing Committee of the National People's Congress (全国人大常委会) (2012). Exit and Entry Administration Law, 出境入境管理法（第十二条）. (National Immigration Administration) ; Standing Committee of the National People's Congress (全国人大常委会) (2023, amended). Civil Procedure Law, 民事诉讼法（第二百六十六条）. (reprinted by the Guizhou Provincial Administration for Market Regulation) +- Notes: In that case, the statement 「I meant to repay it」 was not accepted. The procuratorate found that these people had no job and no fixed assets, could not possibly repay, and had spent the money they were given, and on that basis found that they had never intended to repay. People who took out cash for the gang may also pick up a separate count of money laundering. Whoever's name the loan is in, and whoever signed the contract, is the borrower. Even where it is not a crime, the debt and the entry in your credit record are yours. Taking the money and running abroad settles nothing: once a case has been opened, however many years you stay away it can still be pursued. If you have been sentenced and have not served your sentence, or you are being treated as a suspect or a defendant, you may not leave the country. Where a civil case is not closed, the court may stop you from leaving the country; where a person under enforcement does not perform, the court may also restrict their leaving the country. For taking out cash and making transfers on somebody else's behalf, see item 5 in this section (a 「side job」 that wants you to receive money through your own card). For how to find out that somebody has taken out a loan in your name, see item 9 in section 8. + +### 7. Do not withdraw money from a bank card you pick up; contact the owner or hand a phone or wallet you find to the police, and do not pocket it + +- Cost: No money, and it takes a few minutes. Make one call to find the owner, or take the thing to the police station. +- In plain language: Keeping something you found and using it yourself, where the amount reaches 「数额较大」 and you refuse to hand it over, carries 2 years or less, and an especially large amount carries 2 to 5 years. Using a bank card you found to take money out of an ATM is heavier: it counts as using another person's credit card, and a relatively large amount carries 5 years or less with a fine of 20,000 to 200,000 yuan. For this one the police open a case directly; the owner does not have to sue. +- Benefit: Where you find something somebody has lost, you must return it to the owner, notify them promptly so that they can collect it, or hand it to the public security organ or another department. Keeping something somebody has left behind for your own use, where the amount reaches 「数额较大」 and you refuse to hand it over, constitutes the crime of embezzlement and carries 2 years or less. An especially large amount carries 2 to 5 years. Using another person's credit card that you found on an automated teller machine falls under 「冒用他人信用卡」. Where it amounts to a crime it is handled as credit card fraud: a relatively large amount carries 5 years or less, with a fine of 20,000 to 200,000 yuan. An especially large amount, or other serious circumstances, carries 5 to 10 years, with a fine of 50,000 to 500,000 yuan (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第三百一十四条）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第一百九十六、二百七十条）. ; Supreme People's Procuratorate (2008). Reply on How to Characterise the Act of Picking Up Another Person's Credit Card and Using It on an Automated Teller Machine (ATM), 关于拾得他人信用卡并在自动柜员机（ATM 机）上使用的行为如何定性问题的批复. +- Notes: The crime of embezzlement is dealt with only on complaint, which means the owner has to sue in court themselves and the police do not take it up of their own accord. But taking out cash with a card you found is different: the owner does not have to sue, and the police open the case directly. What 「数额较大」 comes to in money is set by the judicial interpretation, and this section has not verified it. After you have handed something over to the department concerned, the owner has to pay a storage fee when they come to collect it (Article three hundred and seventeen of the Civil Code). + +### 8. Claim a discount in your own real identity, do not buy virtual numbers to register batches of accounts, and do not rely on a loophole to claim it again and again + +- Cost: No money. Turn down the work in the 「deal groups」 that hand out wool-pulling tasks. If you find a loophole in a platform, leave it alone. The hard part is holding out against treating a discount as income you can copy in bulk. +- In plain language: Claiming a coupon once, in your own real identity, on one account, is not within the penal provisions. Registering fake users in bulk, or placing orders yourself and then taking the orders yourself in order to claim more, is handled as fraud: 3000 yuan is enough for 3 years or less, and 30,000 yuan or more is 3 to 10 years. One rider used virtual numbers at 0.1 yuan each to run up coupons and cashed out more than 90,000 yuan, and 12 people in the same case were all sentenced. +- Benefit: Fraudulently obtaining public or private property in a relatively large amount carries 3 years or less, with a fine imposed in addition or instead, so the fine can come with the sentence or on its own. An especially large amount, or other serious circumstances, carries 3 to 10 years and a fine. An especially especially large amount, or other especially serious circumstances, carries 10 years or more or life imprisonment, with a fine or confiscation of property. For the three bands of relatively large, especially large and especially especially large, each province sets its own local standard within the ranges below. Relatively large is 3000 yuan up to more than 10,000 yuan. Especially large is 30,000 yuan up to more than 100,000 yuan, and especially especially large is more than 500,000 yuan. Teaching the method to other people, or selling a 「tutorial」, carries 5 years or less under the crime of teaching criminal methods, where the circumstances are serious 5 to 10 years, and where they are especially serious 10 years or more or life imprisonment. Those who buy and sell phone numbers and verification codes upstream and downstream are handled under the crime of infringing citizens' personal information, which carries 3 years or less, and where the circumstances are especially serious, 3 to 7 years. Batch account systems, systems that switch network addresses automatically, and receiving platforms that take in bulk verification text messages are themselves expressly forbidden by Article fourteen of the Law on the Prevention and Control of Telecom and Online Fraud. Illegally buying and selling, renting out or lending telephone cards and internet accounts, and helping somebody else pass real-name verification, fall under Article thirty-one. A breach of Article fourteen means all the money you earned is confiscated and a fine of 1 to 10 times that amount is imposed. Where you earned nothing, or earned less than 50,000 yuan, the fine is up to 500,000 yuan, and in serious cases there is detention of up to 15 days. A breach of Article thirty-one carries the same 1 to 10 times, and where you earned nothing, or earned less than 20,000 yuan, the fine is up to 200,000 yuan, and in serious cases detention of up to 15 days. Taking part in telecom and online fraud without reaching the level of a criminal sentence carries 10 to 15 days of detention, with the money earned confiscated and a further fine of 1 to 10 times. Where you earned nothing, or earned less than 10,000 yuan, the fine is up to 100,000 yuan. The first official case: the food delivery rider Liu bought virtual phone numbers at 0.1 yuan each and registered new users on a platform. He placed orders using the newcomer coupon of 29 off 15, then resold the goods to small shops to cash in. He also used the virtual numbers to invite one another and claimed reward payments of 2 yuan to 20 yuan. He and the rider Tan defrauded more than 90,000 yuan and more than 70,000 yuan respectively. The procuratorate of the Yangpu District of Shanghai prosecuted 12 people, and the court sentenced them to 3 years and 6 months down to 3 months' criminal detention. The second official case: Wang exploited the time gap between a shop's refunds and the recovery of virtual lottery coins. He took out 1156 refunds, collected 224 figurines, and the amount was more than 249,000 yuan. He also cashed out on a second-hand platform, and the court sentenced him for fraud to 3 years suspended for 4 years, with a fine of 30,000 yuan. During the trial he returned the proceeds and paid the fine in advance, 330,000 yuan in all. The third official case: the newcomer coupons and driver subsidies of a ride-hailing platform were taken by fake orders in which you book your own ride and then accept the order yourself, across more than 70 cities in 19 provinces, for more than 200 余万元 (2,000,000 yuan). The procuratorate of the Longcheng District of Chaoyang, Liaoning prosecuted 163 people, all of whom were found guilty. The principal offenders, Huang and Cai, were convicted of fraud, of infringing citizens' personal information and of teaching criminal methods together, with the three crimes worked out as one sentence, and each was sentenced to 3 years and 6 months. A further 34 minor accomplices were let off by a decision not to prosecute, and the cases were still referred to the police for an administrative penalty (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百五十三条之一、第二百六十六条、第二百九十五条）. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau) ; Supreme People's Court and Supreme People's Procuratorate (2011). Interpretation on Several Specific Questions Concerning the Application of the Law in Handling Criminal Cases of Fraud, 关于办理诈骗刑事案件具体应用法律若干问题的解释（法释〔2011〕7 号，第一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2022). Law of the People's Republic of China on the Prevention and Control of Telecom and Online Fraud, 反电信网络诈骗法（第十四、三十一、三十八、四十二、四十四条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第九百八十五条）. ; Supreme People's Procuratorate (2021). 剑指电信网络诈骗｜不法分子盯上了电商经济的漏洞（检察日报）. ; Supreme People's Procuratorate (2025). 如此「薅羊毛」涉嫌诈骗（检察日报）. ; Supreme People's Procuratorate (2024). 辽宁朝阳龙城区：办理系列电信网络诈骗案（法治日报）. +- Notes: The line is not drawn at 「how much did you take」 but at whether you fabricated anything. Joining an activity in your own real identity, claiming once on one account and using the coupon yourself is a discount the platform gave you. Registering fake users in order to claim more, fabricating transactions, or hiding the fact that you have already had a refund is what makes the other party hand the goods over believing it is genuine, and that is handled as fraud, with the amount only deciding how long the sentence runs. An amount too small to be a crime still has consequences: the platform can close the account, claw back the discount, and demand the return of money that should not have come to you. Taking part in telecom and online fraud without reaching the level of a crime still brings the band of detention and a fine. 「Somebody in the group is running it and pays me per task」 does not change the nature of your own part. Teaching the method to other people, and reselling accounts and phone numbers, are separate charges, and several crimes are worked out together. The person who benefits is you. For writing your own script and selling scripts, see item 3 in section 11. For selling cards and accounts, see item 8 in section 8 (lending a bank card to somebody) and item 16 in this section (lending your identity card to somebody). + +### 9. Throw nothing out of a balcony or a window, and that includes cigarette ends, rubbish bags and bottles + +- Cost: No money, and no time either. Put cigarette ends, rubbish bags and bottles in the bin inside the flat, and do not throw them out of a window or off a balcony. +- In plain language: Throwing something out of a window can mean 5 days of detention and a fine of 1000 yuan even when it hits nobody, and where the circumstances are serious, 10 to 15 days of detention. Hitting somebody changes the charge: one man threw empty bottles and glass cups from the 21st floor into a playground below after drinking, hit a 13-year-old pupil and caused grievous bodily harm, and was sentenced to 10 years. Another threw 8 bricks from the 32nd floor and killed 1 person, and in 2024 the death sentence was approved. +- Benefit: Throwing things down from a building or another high place, where there is a risk of injuring somebody, damaging property or endangering public security, carries up to 5 days of detention, or a fine of up to 1000 yuan. Where the circumstances are serious, 10 to 15 days of detention, and a further fine of up to 1000 yuan may be imposed. Where the circumstances are serious enough for the crime of throwing objects from a height, it carries 1 year or less, with a fine imposed in addition or instead, so the fine can come with the sentence or on its own. Where somebody is injured or killed, it is handled under a heavier charge. The official case: Li, after drinking, threw empty bottles and glass cups from the 21st floor into the playground below, hit a 13-year-old pupil and caused serious bodily harm of grade two, and was sentenced to 10 years for endangering public security by dangerous means. Zhou threw 8 bricks from the 32nd floor into a crowd and killed 1 person, and in 2024 the Supreme People's Court approved the death sentence (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第四十三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第三十三条，即刑法第二百九十一条之二）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百九十一条之二）. ; Supreme People's Court (2024). Typical Cases of Punishing Crimes Endangering Public Security According to Law (Case One), 依法惩治危害公共安全犯罪典型案例（案例一）. ; Supreme People's Court (2025). 专访最高法副院长李勇：让高空抛物行为人都能被处罚. +- Notes: You can be detained even when it hits nobody, because the article only requires 「a risk」. The Civil Code has a separate civil compensation rule on throwing objects from a height (Article one thousand two hundred and fifty-four). The original text would not open while this section was being written, so it is not quoted here. + +### 10. Do not buy imitation guns, steel pellet guns or air guns online, and do not buy a 「toy gun」 of unknown origin + +- Cost: No money. If you want to shoot, go to a proper shooting range. +- In plain language: A steel pellet gun or air gun bought online counts as a firearm once its muzzle energy reaches 1.8 joules per square centimetre. Holding one without a licence carries 3 years or less, and where the circumstances are serious, 3 to 7 years. One couple sold 「toy guns」, and 18 of the 43 were found on examination to be firearms; they were not prosecuted only because it could not be proved that they knew. +- Benefit: Holding or hiding a firearm without a licence carries 3 years or less, and where the circumstances are serious, 3 to 7 years. A non-standard gun that cannot fire standard ammunition counts as a firearm as soon as its muzzle energy is greater than or equal to 1.8 joules per square centimetre. Muzzle energy means the energy the bullet carries per square centimetre as it leaves the barrel. Many steel pellet guns and air guns bought online exceed that figure. From 2018 the Supreme People's Court and the Supreme People's Procuratorate require that, for an air gun with a low figure, the appearance, the materials, the price, the intended use and whether the person knew should be weighed up before deciding whether to convict. The official case: the couple Zhao sold 「toy guns」 at a market and 43 of them were seized, of which 18 were found on examination to meet the standard for a firearm. Because there was not sufficient evidence that they knew these were firearms, they were not prosecuted on the ground that the evidence was in doubt (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第一百二十八条）. ; Supreme People's Court and Supreme People's Procuratorate (2018). Reply on Questions Concerning Conviction and Sentencing in Criminal Cases Involving Guns and Air-gun Pellets Powered by Compressed Gas, 关于涉以压缩气体为动力的枪支、气枪铅弹刑事案件定罪量刑问题的批复（法释〔2018〕8 号）. ; Supreme People's Court (2018). 《批复》的理解与适用. +- Notes: The key to that case not being prosecuted was the finding that they were 「not aware」. For an adult who deliberately goes online to buy a steel pellet gun or an air gun, afterwards saying you did not know it was a gun is very hard to establish. How crossbows and controlled knives are dealt with is not checked in this section, so nothing is written about it. + +### 11. Register a drone you have bought under your real name first, do not fly over airports, military areas or controlled airspace over a city, and do not defeat the height limit + +- Cost: No money. Real-name registration is free and takes a few minutes. Before every flight, check through an official channel whether the place you intend to fly over is controlled airspace. +- In plain language: Flying without real-name registration carries a fine of up to 200 yuan, and where the circumstances are serious, 2000 to 20,000 yuan. Flying in controlled airspace without approval carries a fine of up to 500 yuan, and in serious cases the drone is confiscated and a further fine of 1000 to 10,000 yuan imposed. Where the circumstances are relatively serious there can also be 5 to 10 days of detention. One man paid to defeat the height limit, flew to close to 3000 metres to photograph the clouds, posted it online, and was detained for 5 days. +- Benefit: Flying without completing real-name registration: ordered to be rectified, and a fine of up to 200 yuan may be imposed; where the circumstances are serious, a fine of 2000 to 20,000 yuan. Flying a micro, light or small drone in controlled airspace without approval: ordered to stop flying, and a fine of up to 500 yuan may be imposed. In this item, where the circumstances are serious, the drone is confiscated and a fine of 1000 to 10,000 yuan is imposed. Violating the airspace management rules in order to fly, where the circumstances are relatively serious, carries 5 to 10 days of detention. The official case: Li paid to defeat the height limit, flew the drone to close to 3000 metres to photograph the clouds and posted it, and was detained for 5 days. Zhang defeated the height limit and the no-fly zone restrictions for more than 20 sorties on behalf of other people and earned more than 15,000 yuan. He was convicted of providing programs or tools for intruding into or illegally controlling computer information systems. He was sentenced to 6 months, suspended for 1 year, together with a fine (nationwide, in force from 2024-1) +- Evidence grade: A +- Sources: State Council and Central Military Commission (国务院、中央军委) (2023). Interim Regulations on the Flight Administration of Unmanned Aircraft, 无人驾驶航空器飞行管理暂行条例（国令第 761 号，第十、十九、四十七、五十一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第四十六条）. ; Hubei Provincial Public Security Department (2026). 公安部公布依法打击无人机「黑飞」违法犯罪八起典型案例（转载人民公安报）. +- Notes: Flying a micro or light drone in airspace where flight is permitted does not need approval. A micro drone is one whose empty weight is under 0.25 kg. The airspace where flight is permitted is everywhere outside controlled airspace, and the areas around airports, military facilities and government organs are mostly controlled airspace. The Ministry of Public Security's own page would not open while this section was being written, so the reprint by the Hubei Provincial Public Security Department is cited instead. + +### 12. Do not film people in secret, and do not install cameras in a rented flat, a guesthouse room or a hotel room + +- Cost: No money, and no time either. Do not film people in secret, and do not install cameras in a rented flat, a guesthouse room or a hotel room. +- In plain language: Filming in secret, peeping and spreading another person's privacy carries up to 5 days of detention or a fine of up to 1000 yuan, and where the circumstances are relatively serious, 5 to 10 days of detention. Using equipment made for secret filming and causing serious consequences carries 2 years or less. Two people fitted equipment into rooms in three hotels and filmed in secret remotely, and were sentenced to 1 year and 3 months and 1 year respectively. A camera in your own home does not count; a camera in a room somebody else lives in does. +- Benefit: Peeping, filming in secret, wiretapping or spreading another person's privacy carries up to 5 days of detention, or a fine of up to 1000 yuan. Where the circumstances are relatively serious, 5 to 10 days of detention, and a further fine of up to 1000 yuan may be imposed. Illegally using equipment made solely for wiretapping or secret photography and causing serious consequences carries 2 years or less. Equipment made solely for wiretapping or secret photography means equipment designed specifically for listening in and filming in secret. The official case: Yan and Yan bought secret-photography equipment online, fitted it into several rooms in three hotels and filmed in secret from a distance, and were sentenced to 1 year and 3 months and 1 year respectively (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第五十条）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十四条）. ; Supreme People's Court (2024). Typical Cases of Punishing the Crime of Illegally Producing, Selling or Using Wiretapping and Secret Photography Equipment According to Law (Case Two), 依法惩治非法生产、销售、使用窃听、窃照设备犯罪典型案例（案例二）. +- Notes: Once secret footage is sold or circulated, it is handled separately as, for example, the crime of producing or selling obscene material for profit, and in the same batch of cases there are 10-year sentences. A camera in your own home does not count; a camera in a room somebody else uses does. + +### 13. Mahjong and cards are fine to play, but do not take a cut, do not be the banker, do not set up the game for a fee, and do not play online gambling + +- Cost: No money. Small amounts of play for fun among friends and relatives are fine. Do not take a cut of the game, do not be the banker, do not set up the game in order to collect money. The hard part is holding out when somebody else is organising it. +- In plain language: Small amounts of cards among friends and relatives are not something the law touches. Taking a cut of the game, being the banker or arranging the game in order to collect money changes the nature of it: taking part in gambling where the stakes are relatively large carries up to 5 days of detention or a fine of up to 1000 yuan. Gambling in a group and making a living out of gambling carry 3 years or less, and opening a gambling house carries 5 years or less. What counts as 「stakes relatively large」 is set locally, and there is no single national figure. +- Benefit: Providing the conditions for gambling in order to make money, or taking part in gambling where the stakes are relatively large, carries up to 5 days of detention, or a fine of up to 1000 yuan. Where the circumstances are serious, 10 to 15 days of detention and a fine of 1000 to 5000 yuan. Gathering people to gamble in order to make money, or making a living out of gambling, carries 3 years or less and a fine. Opening a gambling house carries 5 years or less and a fine, and where the circumstances are serious, 5 to 10 years (nationwide) +- Evidence grade: B +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第八十二条）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第三百零三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第三十六条，开设赌场刑期提高到五年以下、五年以上十年以下）. +- Notes: There is no single national figure for what counts as 「stakes relatively large」, each area sets its own, so this item is only given grade B. Taking a cut of the game, and collecting bets on behalf of an online gambling platform, are often handled as opening a gambling house when a case is dealt with. The original text of the relevant judicial interpretation has not been checked in this section. No official case was found either. + +### 14. Do not eat wild game, and do not buy, sell or keep protected animals as pets + +- Cost: No money. For pets, buy only breeds of lawful origin that have been bred in captivity. +- In plain language: Eating an animal under national key protection means the animal is confiscated and a fine of 2 to 20 times its value is imposed. Buying, transporting or selling precious and endangered animals carries 5 years or less, and where the circumstances are serious, 5 to 10 years. Animals bred in captivity and traded as pets are generally not a crime: one man sold more than 40 artificially bred parrots and was not prosecuted. Animals taken from the wild, and animals of unknown origin, are not covered by that. +- Benefit: Eating a wild animal under national key protection, or buying one in order to eat it, means the animal is confiscated and a fine of 2 to 20 times its value is imposed. Breeding wild animals under national key protection without a licence means confiscation and a fine of 1 to 10 times. Illegally buying, transporting or selling precious and endangered wild animals under national key protection, and products made from them, carries 5 years or less and a fine. In this item, serious circumstances carry 5 to 10 years, and especially serious circumstances carry 10 years or more. Hunting, buying, transporting or selling other terrestrial wild animals in order to eat them carries 3 years or less where the circumstances are serious. The official case: the farmer Wang and 2 others sold 40-odd artificially bred Fischer's lovebirds, a species under national second-class protection, for a total of under 400 yuan, and in 2021 the procuratorate decided not to prosecute. From 2022 the Supreme People's Court and the Supreme People's Procuratorate make it clear that, where artificial breeding technology is mature, the operation is already at scale, and the animals are bred, traded and transported as pets, it is generally not handled as a crime (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2022). Wildlife Protection Law, 野生动物保护法（2022 年修订，第三十一、五十一、五十三条）. (reprinted by the Ministry of Ecology and Environment) ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第三百四十一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第四十一条，即刑法第三百四十一条第三款）. ; Supreme People's Court and Supreme People's Procuratorate (2022). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Damaging Wild Animal Resources, 关于办理破坏野生动物资源刑事案件适用法律若干问题的解释（法释〔2022〕12 号，第十三条）. ; Supreme People's Procuratorate (2023). 鹦鹉小案 民生大案. +- Notes: 「Generally not treated as a crime」 refers only to animals bred in captivity and kept as pets. Animals taken from the wild and animals of unknown origin are not covered. Which animals count as being under national key protection is settled by the list. Checking the list before you buy is less trouble than explaining it afterwards. + +### 15. Do not detain anybody to collect a debt, do not lock anybody up, and do not follow them home and refuse to leave + +- Cost: No money. If the other side owes money and will not pay, sue in court. For how to write an IOU, see item 18 in section 8 (an IOU). The hard part is holding back from using force, and from following them to their home, when the other side defaults. +- In plain language: The money really is owed to you, and you still may not detain anybody or lock anybody up. The article says so in terms: detaining or locking somebody up in order to collect a debt is handled as unlawful detention just the same. The lighter cases carry 10 to 15 days of detention and a fine of 1000 to 2000 yuan. Where it amounts to the crime of unlawful detention it carries 3 years or less, causing grievous bodily harm carries 3 to 10 years, and causing death carries 10 years or more. To collect a debt, sue in court. +- Benefit: Illegal restriction of another person's personal freedom, unauthorised entry into another person's home, or an illegal search of another person's body carries 10 to 15 days of detention and a fine of 1000 to 2000 yuan. The relatively minor cases carry 5 to 10 days of detention and a fine of up to 1000 yuan. Unlawful detention carries fixed-term imprisonment of not more than 3 years, criminal detention, which is short-term confinement, public surveillance, which is confinement-free but restricts your movement, or deprivation of political rights. Causing grievous bodily harm carries 3 to 10 years, and causing death carries 10 years or more. Detaining or locking somebody up in order to collect a debt is punished under this same set of rules, and that holds even when the debt is real (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第四十七条）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百三十八条）. +- Notes: 「He really does owe me money」 does not help here. The article says in terms that detaining or locking somebody up in order to collect a debt is handled as unlawful detention. There are several cases on the Chinese courts' online service of people being locked up over a lawful debt, with sentences of 7 months to 1 year. That website would not open while this section was being written, so nothing is cited here. + +### 16. Do not lend your identity card to anybody, do not use somebody else's, and do not use another person's documents to register, open a bank account or buy a ticket + +- Cost: No money. If somebody asks to borrow it, say no. If you have to hand over a photocopy, write on it which matter this one is for. +- In plain language: Lending, renting or transferring your identity card to somebody else brings a warning and a fine of up to 200 yuan. Impersonating somebody with their identity card, or using one you obtained by deception, carries a fine of 200 to 1000 yuan, or up to 10 days of detention. Using a false document or somebody else's document in a place where identity is required, such as buying a train ticket or staying in a hotel, carries criminal detention or public surveillance where the circumstances are serious. +- Benefit: Renting out, lending or transferring your resident identity card to somebody else brings a warning and a fine of up to 200 yuan. Impersonating somebody with their identity card, or using an identity card obtained by deception, carries a fine of 200 to 1000 yuan, or up to 10 days of detention. Buying, selling or using a forged or altered identity card falls in the same band. In some situations the state requires you to produce proof of identity. In those situations, using a forged or altered document, or another person's stolen document, carries a prison sentence where the circumstances are serious. Documents include the identity card, the passport, the social insurance card and the driving licence. The penalty is criminal detention, which is short-term confinement, or public surveillance, which is confinement-free but restricts your movement, and a fine is imposed in addition or instead, so the fine can come with the sentence or on its own (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2011). Resident Identity Card Law, 居民身份证法（2011 年修正，第十六、十七条）. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau) ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十条之一）. +- Notes: An identity card that has been lent out may be used to register a company or to get a bank card. For the consequences of both, see items 28 and 8 in section 8 (do not act as a nominal legal representative, and do not lend a bank card to somebody). Using somebody else's identity card to buy a train ticket or to stay in a hotel falls under the 「activities in which identity must be provided」 that the article describes. No official case was found for this item. + +### 17. Cooperate when the police stop you or take you away: do not push, do not hit, do not grab the body-worn camera, and after a drink be even more careful + +- Cost: No money. Do not use force on the spot. If you have an objection, raise it afterwards: a complaint, an application for administrative reconsideration, or a lawsuit in court. The hard part is holding your temper through the few minutes you are stopped for, especially after a drink. +- In plain language: Obstructing law enforcement on the spot brings a warning or a fine of up to 500 yuan, and where the circumstances are serious, 5 to 10 days of detention. Laying hands on a police officer is a separate band: biting, slapping, kicking, wrestling to the ground or throwing things, where it amounts to a minor injury or worse, counts as violence against a police officer and carries 3 years or less. Using a firearm, a controlled knife or driving into an officer carries 3 to 7 years. Swinging your arms free and pulling away, which is ordinary resistance, does not count. +- Benefit: Obstructing a state functionary who is lawfully carrying out duties brings a warning or a fine of up to 500 yuan. Where the circumstances are serious, 5 to 10 days of detention, and a further fine of up to 1000 yuan may be imposed. Where the person obstructed is the people's police, the punishment is heavier. Violence against a people's police officer who is lawfully carrying out duties carries 3 years or less. Using a firearm, a controlled knife or driving into the officer, in ways that seriously endanger the officer's personal safety, carries 3 to 7 years. Only biting, slapping, kicking, wrestling to the ground or throwing objects in a way that causes a minor injury or worse counts as 「violence against an officer」. Swinging your arms free, pulling away and kicking your legs, which is ordinary resistance, does not count (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第六十一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第三十一条，即刑法第二百七十七条第五款）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百七十七条）. ; Supreme People's Court and Supreme People's Procuratorate (2025). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Assaulting the Police, 关于办理袭警刑事案件适用法律若干问题的解释（高检发释字〔2025〕1 号，第一、八、十二条）. ; Supreme People's Procuratorate (2021). 袭警罪入刑一月 检察机关批捕 405 人. +- Notes: The Supreme People's Procuratorate reported in 2021 that in the first month after assaulting the police became a separate charge, it approved the arrest of 405 people. Of 42 cases in Jiangsu, 47.6% arose from making a scene after drinking. Of 16 cases in Hubei, 12 were only punching, kicking and biting. Laying hands on an auxiliary police officer is handled under the crime of obstructing official duties. No official case setting out the facts of an individual case was found for this item. + +### 18. If the other person is under 14 you must not have sex, and 「she agreed」 is not a reason + +- Cost: No money. If you are not sure how old the other person is, do not have sex. +- In plain language: Where the other person is under 14, it makes no difference whether she agrees: it is handled as rape and punished more severely, with a basic sentence of 3 to 10 years. Where the other person is under 12, you are always found to have 「ought to have known」. Between 12 and 14, whether she may be a minor can be seen from her development, her manner and her clothing, and if you go ahead you are found to have 「known」. Saying you genuinely did not know is held to a very strict standard by the courts. +- Benefit: Having sex with a girl under the age of fourteen is handled as rape and punished more severely, with a basic sentence of 3 to 10 years. The line the Supreme People's Court takes is this: where force, coercion or another coercive means is used, it is rape whether or not you knew the other person was a minor. Where no coercive means is used, somebody under 12 is always found to be one you 「ought to have known」. Between 12 and 14, where it is visible from physical development, speech and manner, and clothing that she may be a minor, and you go ahead, you are found to have 「known」. The exception of 「genuinely did not know」 is applied very strictly (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百三十六条）. ; Supreme People's Court (2015). 关于「与十四周岁以下幼女发生性关系是否要明知才构成犯罪」的答复. +- Notes: The original text of the joint opinion of the two supreme courts and the two ministries on handling criminal cases of sexual assault against minors, and the page for the Supreme People's Court's reply of 2003, both would not open while this section was being written. So the standard for 「known」 is taken from the answer on the Supreme People's Court's own website. The term used in the article is 「幼女」. This item makes no distinction between a man and a woman. + +### 19. Work out the bill before you hit anybody: detention, compensation, a prison sentence + +- Cost: No money, and no time. The hard part is the few seconds before you swing. +- In plain language: The bill for one swing: 5 to 10 days of detention plus a fine of 500 to 1000 yuan. Causing a minor injury is a criminal case carrying 3 years or less, and causing serious injury carries 3 to 10 years. The money is counted separately: one man who broke three of the other person's ribs paid 113,000 yuan to settle it, which is why he was not prosecuted. Another who broke the other person's radius paid nothing and was sentenced to 10 months. +- Benefit: Assaulting another person or intentionally injuring them carries 5 to 10 days of detention and a fine of 500 to 1000 yuan. Assaulting in a group, assaulting a person with a disability, a pregnant woman, an elderly person or a child, assaulting repeatedly, or assaulting several people at once, carries 10 to 15 days of detention and a fine of 1000 to 2000 yuan. Causing a minor injury amounts to the crime of intentional injury, which carries fixed-term imprisonment of not more than 3 years, criminal detention, which is short-term confinement, or public surveillance, which is confinement-free but restricts your movement. Causing serious injury carries 3 to 10 years. In civil law you also have to pay medical costs, lost earnings, meaning the money you did not earn because you could not work, nursing costs, travel costs, the meals allowance for a hospital stay, and nutrition costs. Where the person is left permanently disabled, you additionally pay disability compensation. Disability compensation is the per-capita disposable income of urban residents for the previous year in the place where the trial court sits, multiplied by 20 years. For a victim over 60 years old, one year is deducted for each year above that, and above 75 years it is counted as 5 years. In 2022 procuratorates nationwide accepted more than 70,000 cases of minor injury, most of them arising from marriage, family or neighbour disputes or from chance events. The Supreme People's Procuratorate's typical case: Lu broke three of the other person's ribs, an injury of grade two, and paid 113,000 yuan to settle the matter, which is why he was not prosecuted. Shi broke the other person's radius, also grade two, paid nothing, and was sentenced to ten months' imprisonment (nationwide, the Law on Penalties for Administration of Public Security in force from 2026-1-1) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第九、五十一条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百三十四条）. ; Supreme People's Court (2022, amended). Interpretation on Several Questions Concerning the Application of the Law in Trying Cases of Personal Injury Compensation, 关于审理人身损害赔偿案件适用法律若干问题的解释（法释〔2022〕14 号，第六至十二、十五条）. ; Supreme People's Procuratorate and Ministry of Public Security (2023). Guiding Opinions on Properly Handling Cases of Minor Injury According to Law, 关于依法妥善办理轻伤害案件的指导意见. ; Supreme People's Procuratorate (2023). 检察机关依法妥善办理轻伤害案件典型案例. +- Notes: For how to avoid a conflict, why the first person to swing loses out, and where the boundary of self-defence lies, see item 10 in section 8 (conflict and self-defence). For the bill for laying hands on somebody out of anger, see items 11 to 14 in section 8. A fight arising from a civil dispute and of relatively minor circumstances can be mediated by the police. Where a minor injury has been caused, if you agree compensation with the other person and actually pay it, the punishment can be lighter and there may even be no prosecution, but not one yuan of the compensation can go unpaid. What counts as a minor injury is set by the assessment standards, and this section has not verified that. Lost earnings and nursing costs are calculated on actual income and on the number of days lost, and there is no single figure. + +### 20. If you cannot raise a child after it is born, registration with the civil affairs department is the only lawful way out: handing the child over for money can be punished as trafficking, and walking away is abandonment + +- Cost: No money. To register an adoption, go to the civil affairs department at county level or above, hand in the documents, and wait for the review and the adoption assessment, which takes some time. +- In plain language: There is a way out when you cannot raise a child, but only one lawful one: register the adoption with the civil affairs department, and the adoption only comes into existence from the day of registration. Handing the child over privately, especially for money, can turn a 「送养」 into the crime of trafficking children, which starts at 5 years or more. Leaving the child where they are and refusing to support them, where the circumstances are serious, is the crime of abandonment, which carries 5 years or less. +- Benefit: Article two hundred and forty of the Criminal Law:「拐卖妇女、儿童的，处五年以上十年以下有期徒刑，并处罚金」. Where there are aggravating circumstances, the sentence is fixed-term imprisonment of not less than ten years or life imprisonment, with a fine or confiscation of property. 「情节特别严重的，处死刑，并处没收财产」. item 17 of the Opinions on Punishing the Crime of Trafficking Women and Children According to Law sets out how to tell selling your own biological child apart from a private adoption. The test is 「关键在于行为人是否具有非法获利的目的」, that is, whether you meant to make money out of it. Three circumstances can be found to be selling your own biological child, and are convicted as trafficking a woman or child. The first is 「将生育作为非法获利手段，生育后即出卖子女的」. The second is 「明知对方不具有抚养目的，或者根本不考虑对方是否具有抚养目的，为收取钱财将子女『送』给他人的」. The third is 「为收取明显不属于『营养费』『感谢费』的巨额钱财将子女『送』给他人的」. In the other direction, where the purpose is not money, you really are pressed by hardship and hand the child over privately, taking only a small payment for nutrition and thanks: that is a private adoption and is not convicted as trafficking. Now look at Article two hundred and sixty-one of the Criminal Law. 「对于年老、年幼、患病或者其他没有独立生活能力的人，负有扶养义务而拒绝扶养，情节恶劣的，处五年以下有期徒刑、拘役或者管制」. Article one thousand and ninety-four of the Civil Code says there are only three kinds of person who may hand a child over: the guardian of an orphan, a child welfare institution, and biological parents who have special difficulties and cannot support the child. Article one thousand one hundred and five: 「收养应当向县级以上人民政府民政部门登记。收养关系自登记之日起成立」 +- Evidence grade: A +- Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百四十、二百六十一条）. ; Supreme People's Court, Supreme People's Procuratorate, Ministry of Public Security and Ministry of Justice (2010). Opinions on Punishing the Crime of Trafficking Women and Children According to Law, 关于依法惩治拐卖妇女儿童犯罪的意见（法发〔2010〕7 号，第 17 条）. Supreme People's Court website. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零九十四、一千一百零五条）, full text of the adoption book reprinted by the Yancheng Municipal People's Government. +- Notes: Whether to have a child at all is not discussed here. After the child is born, this order saves the most trouble. First, give birth in hospital; for the antenatal check and the hospital delivery see section 27. The medical certificate of birth is the precondition for every formality that comes after it. Second, if you are going to hand the child over, go straight to the civil affairs department at county level or above and do not go through an intermediary. The moment the money changes hands the nature of it can change. Within a year after the birth, if you find yourself with 「thoughts of harming yourself or the child」, treat it as an emergency. For the list of things that need a hospital straight away during pregnancy and after the birth, see item 7 in section 27. The check at 42 days after the birth is also a screening for postnatal depression: see item 16 in section 27. If you have suicidal thoughts, call 12356: see item 25 in section 1. This item has no case attached: the official typical cases of trafficking your own biological child were not checked case by case in this round, so only the articles and the text of the opinion are cited. + +### 21. Do not fake an accident and do not inflate a loss to claim insurance: that is the crime of insurance fraud, and the people who testify for you, repair your car or assess it count with you + +- Cost: No money, no time, and you simply do not do it. +- In plain language: Inventing an accident or inflating a loss in order to claim insurance is the crime of insurance fraud: a relatively large amount carries 5 years or less with a further fine of 10,000 to 100,000 yuan. People who help by giving evidence, by assessing, or by valuing property, where they deliberately produce false certificates, are convicted as accomplices with you. In one case in Xuancheng, Anhui, 8 people were sentenced to 2 to 4 years and each fined 10,000 yuan. +- Benefit: Article one hundred and ninety-eight of the Criminal Law lists five circumstances. First, the applicant deliberately fabricates the subject matter of the insurance, inventing an insured party that does not exist. Second, the applicant, the insured or the beneficiary fabricates a false cause for an insurance event that happened, or exaggerates the extent of the loss. Third, an insurance event that never happened is fabricated. Fourth, the applicant or the insured deliberately causes the insured event of property damage. Fifth, the applicant or the beneficiary deliberately causes the death, disability or illness of the insured person. A relatively large amount carries fixed-term imprisonment of not more than 5 years or criminal detention, and a fine of not less than 10,000 yuan and not more than 100,000 yuan. An especially large amount, or other serious circumstances, carries 5 to 10 years, and a fine of 20,000 to 200,000 yuan. An especially especially large amount, or other especially serious circumstances, carries 10 years or more, and a fine of 20,000 to 200,000 yuan or confiscation of property. The last paragraph of this article covers the people who help: the assessor, the witness and the property valuer for an insurance event. Where they deliberately issue false supporting documents or create the conditions for somebody else to commit insurance fraud, they are handled as accomplices in the insurance fraud and are convicted together with the person who committed it. Article twenty-seven of the Insurance Law: where no insurance event occurred but one is claimed, the insurer may terminate the contract and does not refund the premiums. Where forged or altered supporting material is used, or a false cause of the accident is fabricated or the extent of the loss exaggerated, the insurer does not pay the part that has been over-reported, and anything already paid has to be returned or compensated. The case: in the Zhang Songsong case in Xuancheng, Anhui, 8 people including Wang Xin were sentenced for the crime of insurance fraud to terms of imprisonment ranging from 4 years down to 2 years, each of them with a fine of 10,000 yuan. A false lawsuit that came out of this case involved an assessed loss of 20,000 yuan, and the party obtained standing to sue by forging a vehicle transfer agreement; the court awarded more than 70,000 yuan. That civil judgment, already in force, was set aside on retrial on 2019-9-26. The Measures for Work on Anti-Insurance Fraud require that risk information be verified at both ends of the business, the underwriting end and the claims end. They also call for exploring the building of an industry anti-fraud information platform covering several lines of insurance, and an anti-fraud intelligence centre. These platforms screen suspected fraud data centrally, and the leads are checked by industry associations, anti-fraud organisations and insurers. Anything suspected of being a crime is reported to the public security organ in good time. The financial regulator and its local offices are to establish a sound mechanism linking administrative enforcement and criminal justice with the public security organs, the people's procuratorates and the people's courts (nationwide, issued on 2024-7-31) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会). Criminal Law of the People's Republic of China, consolidated text as amended by Criminal Law Amendment Eleven, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第一百九十八条. ; Standing Committee of the National People's Congress (全国人大常委会) (2015, amended). Insurance Law of the People's Republic of China, Article twenty-seven, 中华人民共和国保险法第二十七条. ; National Financial Regulatory Administration (国家金融监督管理总局) (2024). 关于印发《反保险欺诈工作办法》的通知. ; Anhui Provincial People's Procuratorate, reprinting Legal Daily (2020-01-13). 杀妻骗保牵出案中案. +- Notes: The several practices below that people assume are harmless all fall within these five. After a minor accident, asking a repair shop to 「report a bit more」. Counting an old injury or old damage into this occasion. Crashing into a friend and then reporting it, or inventing one where nothing happened. For how to buy car insurance, see item 26 in section 5 (buy enough third-party liability cover). Where you take out insurance on a family member and then deliberately cause their death, the article says in terms that the crimes are punished together: see item 38 in section 8 (taking out insurance on a family member). Medical insurance follows a different set of rules, and running a medical insurance card with nothing behind it, or cashing out a personal medical insurance account, is handled as fraud: see the Notes in item 13 in section 5. + +### 22. Do not sell your own organ, and do not help other people find a donor: a kidney fetches a little over 20,000 yuan, the same kidney resold goes for 200,000 yuan, the money is confiscated and a fine of 10 to 20 times the transaction amount is imposed + +- Cost: No money, no time, and you simply do not do it. +- In plain language: A living organ may only be donated to a spouse, a direct blood relative or a collateral blood relative within three generations, and buying or selling is illegal either way. In one case in Beijing the person selling a kidney got a little over 20,000 yuan, while the same kidney sold to a patient fetched 180,000 to 220,000 yuan. The money is confiscated and a fine of 10 to 20 times the transaction amount is imposed. For what losing a kidney costs the body afterwards, see item 35 in section 1 (the remaining kidney does the work of two). +- Benefit: Article five of the Regulations on the Donation and Transplantation of Human Organs:「任何组织或者个人不得以任何形式买卖人体器官，不得从事与买卖人体器官有关的活动」. Article ten:「任何组织或者个人不得获取未满 18 周岁公民的活体器官用于移植」. Article eleven:「活体器官的接受人限于活体器官捐献人的配偶、直系血亲或者三代以内旁系血亲」. Article thirty-seven sets out how buying and selling human organs is punished. Buying or selling human organs, or engaging in activity connected with buying or selling human organs, is investigated by the health commission of a local people's government at county level or above. First the illegal gains are confiscated, taking away the whole of the money from the transaction. Then a fine of 10 times or more and not more than 20 times the transaction amount is imposed. Where medical personnel take part, their practice certificates are revoked and they are permanently barred from working in health care. Where a crime is constituted, criminal liability is pursued under the law (nationwide, in force from 2024-5-1, and the former Regulations on Human Organ Transplantation is repealed at the same time). Article two hundred and thirty-four-one of the Criminal Law:「组织他人出卖人体器官的，处五年以下有期徒刑，并处罚金；情节严重的，处五年以上有期徒刑，并处罚金或者没收财产」. There are three further circumstances, convicted and punished under Article two hundred and thirty-four and Article two hundred and thirty-two of the Criminal Law. First, taking somebody's organ without their consent. Second, taking the organ of a person under eighteen years old. Third, forcing or deceiving somebody into donating an organ. These are handled as intentional injury or intentional homicide. The first case: the Zheng Wei case, 15 defendants, heard on final instance by the Beijing First Intermediate People's Court on 2014-8-20, involved 51 kidneys and an amount of more than 1034 万元 (10,340,000 yuan). What the person selling the kidney, the donor side, got was, in the words of the judgment, 「换来的仅仅是 2 万到 2.5 万元不等的报酬」, a payment ranging from 20,000 to 25,000 yuan. What the waiting patient's side, the recipient side, paid was 「以前是一个肾 18 万或 19 万，后来是 21 万或 22 万」. The place the kidneys were taken out in was a four-storey building rented at 7,500 yuan a month, and the nurse in charge of the gang testified that the place was 「脏乱不堪，连抢救设备都没有」. The organiser Zheng Wei was sentenced to 12 years' imprisonment for the crime of organising the sale of human organs, and the other 14 got 3 and a half years to 9 years. The second case: the 「boy sells his kidney for an Apple phone」 case published by the Supreme People's Procuratorate. Wang, aged 17, had his right kidney removed in Chenzhou on the evening of 2011-4-28. The operation was carried out in a department run by a contractor inside a hospital with no transplant qualification. On 5-2, when he was discharged, he received 22,000 yuan, and the recipient side paid the middleman 150,000 yuan and 10,000 US dollars, with the principal offender making more than 56,000 yuan. After the operation Wang had impaired kidney function, and 「伤情构成重伤、三级伤残」 +- Evidence grade: A +- Sources: State Council (国务院) (2023). Regulations on the Donation and Transplantation of Human Organs, 人体器官捐献和移植条例（国务院令第 767 号，第五、十、十一、三十七条）. ; Standing Committee of the National People's Congress (全国人大常委会). Criminal Law of the People's Republic of China, consolidated text as amended by Criminal Law Amendment Eleven, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第二百三十四条之一. ; Supreme People's Procuratorate (2012-04-09). 「少年卖肾买苹果手机」案诉至法院. ; Official website of the Langfang Intermediate People's Court of Hebei, reprinting (2014-09-02). 全国最大贩肾案终审 组织者摘肾 51 枚获利千万. +- Notes: The person who benefits is you. On the surface this transaction is 「a one-off 20,000 yuan」, and in reality it is trading an organ for a mobile phone. The assessment in the boy kidney case found grievous bodily harm and a grade three disability, and months of rest did not mend it. In that case the prosecution found that 5 defendants had infringed Article two hundred and thirty-four, paragraph two, of the Criminal Law, and prosecuted them for intentional injury. One of the reasons was that the medical personnel did not check before the operation whether the 「donor」 was a minor, and cut out the right kidney anyway. If the same thing happens today, the people who do the recruiting are handled under the crime of organising the sale of human organs, and taking the organ of a person under eighteen years old is still convicted and punished as intentional injury under Article two hundred and thirty-four. The operations run by the black intermediaries are done in a rented flat or a borrowed space, with no assessment beyond the tissue match and no follow-up after the operation, so if a complication arises there is nobody to answer for it. If you really want to donate to a relative, take the route in Article eleven: have it done at a medical institution with transplant qualifications, reviewed by an ethics committee, and charge nothing throughout. Donating organs from your own body after death is free of charge and is outside the scope of this item. For what losing a kidney costs the body afterwards, see item 35 in section 1 (the remaining kidney does the work of two). When the shortage of money has got you to the point of considering selling an organ, look first at the assistance channels in section 7 (unemployment insurance, temporary assistance, means-tested income support). For the traps of online loans and of 「packaging the materials」 loans, see item 6 in this section. + +### 23. Do not buy sex: the default band is 10 to 15 days of detention, not a fine and nothing more + +- Cost: No money, no time, and you simply do not do it. +- In plain language: Getting caught is not a matter of paying a fine. The default band written into the article is 10 to 15 days of detention, and a further fine of up to 5000 yuan may be imposed. Only where the circumstances are relatively minor is it up to 5 days of detention or a fine of up to 1000 yuan. And if you go while knowingly having syphilis or gonorrhoea, that is a crime carrying 5 years or less. +- Benefit: Article seventy-eight of the Law on Penalties for Administration of Public Security: a person who sells sex or buys it is detained for not less than ten days and not more than fifteen days, and may additionally be fined up to five thousand yuan. Where the circumstances are relatively minor, the detention is up to five days or the fine up to one thousand yuan. Soliciting in a public place is punishable by up to five days' detention or a fine of up to one thousand yuan. Article seventy-nine: a person who lures, harbours or introduces another to sell sex is detained for not less than ten days and not more than fifteen days, and may additionally be fined up to five thousand yuan. Where the circumstances are relatively minor, the detention is up to five days or the fine between one thousand and two thousand yuan. Article eighty-seven: where personnel of the hotel, catering, cultural and entertainment, or taxi businesses, or of similar units, tip off the person breaking the law or provide the conditions for the above activities in any other way while the public security organ is investigating prostitution, the detention is not less than ten days and not more than fifteen days. Article three hundred and sixty of the Criminal Law is the crime of spreading a sexually transmitted disease: a person who knowingly has syphilis, gonorrhoea or another serious sexually transmitted disease and sells or buys sex is sentenced to fixed-term imprisonment of not more than five years, criminal detention or public surveillance, and is fined (nationwide, the Law on Penalties for Administration of Public Security in force from 2026-1-1) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第七十八、七十九、八十七条）. ; Standing Committee of the National People's Congress (全国人大常委会). Criminal Law of the People's Republic of China, consolidated text as amended by Criminal Law Amendment Eleven, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第三百六十条. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau) ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（第四十三条，删去刑法第三百六十条第二款）. +- Notes: The detention here is administrative detention, and you go into the detention centre; it is not a case of paying a fine and walking out. Arranging introductions for others and providing premises for them are punished as heavily as buying sex yourself, and that is what Article seventy-nine is for. Somebody who works in a hotel or a KTV and tips off the customers is likewise detained for 10 to 15 days under Article eighty-seven. Article three hundred and sixty of the Criminal Law used to have a second paragraph, the crime of soliciting a girl under 18, and Criminal Law Amendment Nine of 2015 deleted it. Such a case is now handled directly as rape and punished more severely: see item 18 in this section (if the other person is under 14 you must not have sex). Signs reading 「home service massage」 or 「massage and physiotherapy」 do not change the nature of it, and the case is judged on what actually took place. For the risk of catching a sexually transmitted disease or HIV, see item 30 in section 1 (use a condom throughout) and item 38 in section 13 (get the blocking drug within 72 hours). For what to watch out for in the venues themselves, see section 22. diff --git "a/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" new file mode 100644 index 00000000..fed4a1a1 --- /dev/null +++ "b/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -0,0 +1,191 @@ +[← 回总目录](../README.md) + +# 10. 恋爱和结婚划不划算 + +本节把「结婚划不划算」拆成钱、时间、健康、法律风险和关系质量几笔账，分开算。每笔账只用能查到出处的数字，出处限于官方统计，或者荟萃分析（把多项研究的数据汇总起来重算一遍）。这里不做道德判断，也不替你下结论。健康那几笔按死亡风险算，时间那几笔按每天多少分钟算，法律那几笔按法条怎么写算。这三种算法之间不互相换算。拆账方法、填空清单和常见误区见 [docs/结婚划不划算.md](../docs/结婚划不划算.md)。 + +[← Back to the contents](../README.md) + +# 10. Is dating and marriage worth it + +This section breaks 「whether marriage is worth it」 into separate accounts for money, time, health, legal risk and the quality of the relationship, and works them out one at a time. Each account uses only figures whose source can be looked up, and the sources are limited to official statistics or a meta-analysis, which pools the data from several studies and recalculates it. No moral judgement is made here, and no conclusion is drawn on your behalf. The health accounts are worked out on the basis of the risk of death, the time accounts on the basis of minutes per day, and the legal accounts on the basis of what the articles of law say. These three ways of calculating are not converted into one another. For the method of breaking the accounts down, the list to fill in and the common mistakes, see [docs/结婚划不划算.md](../docs/结婚划不划算.md). + +### 1. Meet more people rather than fixating on one: whether two people get on cannot be predicted before you meet + +- Cost: No money. The time goes on meeting more people, and not on going over the same person again and again. The hard part is holding out against fixating. +- In plain language: Two speed-dating studies had people fill in more than 100 mate-preference questionnaires before they met. The questionnaires can work out who on the whole finds it easy to like other people and who on the whole is well liked; they cannot work out whether there is chemistry between these two particular people. Whether you get on is only known once you have met. So meeting more people is better value than betting everything on one person. +- Benefit: In two speed-dating studies, participants filled in more than 100 self-report scales related to mate preferences before they met, that is, questionnaires they scored themselves on. The researchers used a random forest model, which is one kind of algorithm that looks for patterns. The model can work out 「how easy it is on the whole for a person to like other people」, and these questionnaires account for 4% to 18% of the variance in that, where variance means the difference between one person and another, so the questionnaires only account for a small part of it. The model can also work out 「how popular a person is on the whole」, accounting for 7% to 27%. But it cannot work out 「whether there is chemistry between these two particular people」, and no combination of traits and preferences before the meeting will do it either. The same group of researchers had earlier run a speed-dating experiment. Before the event participants reported their own ideal-partner preferences, such as whether they weigh looks more than economic prospects. That preference does not predict the interest they feel on the day for a particular person +- Evidence grade: A +- Sources: Joel S, Eastwick PW, Finkel EJ (2017). Is romantic desire predictable? Machine learning applied to initial romantic attraction. Psychological Science, 28(10), 1478-1489. ; Eastwick PW, Finkel EJ (2008). Sex differences in mate preferences revisited: do people know what they initially desire in a romantic partner? Journal of Personality and Social Psychology, 94(2), 245-264. +- Notes: The two studies only show that it cannot be worked out before you meet. They do not show that no amount of effort is of any use. Whether you get on is only known once you have met, so the time is better spent meeting more people and really seeing each one of them than putting it all on one person. The people in the studies were all university students in Europe and North America. The dating environment in China is different, so discount the figures before applying them to yourself. + +### 2. Stop once the other person has clearly refused, because going on pestering them is an administrative offence, not 「sincerity」 + +- Cost: No money, and no time. The hard part is holding out: no more messages, and no more waiting outside. +- In plain language: The other person has clearly said no, and you go on sending messages and going to wait for them, and from 2026-1 that is an administrative offence. It carries up to 5 days of detention, or a fine of up to 1000 yuan. Where the circumstances are relatively serious, 5 to 10 days of detention. The police can also order you not to have contact with the other person, and breaching that order carries a further 5 to 10 days of detention. +- Benefit: The Law on Penalties for Administration of Public Security, in force from 2026-1-1, lists pestering and following as a separate offence. The article says: 「多次发送淫秽、侮辱、恐吓等信息或者采取滋扰、纠缠、跟踪等方法，干扰他人正常生活的」. This conduct is punishable by up to five days' detention, or a fine of up to one thousand yuan. Where the circumstances are relatively serious, it is punishable by five to ten days of detention, and a fine of up to one thousand yuan may be added. This item has one more layer: with the approval of the person in charge of the public security organ, the behaviour may be ordered to stop and the person may be barred from having contact with the victim for a set period. Breaching the no-contact order carries a further five to ten days of detention +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025, revised). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（第五十条第一款第五项、第二款）. (reprinted by the Supreme People's Procuratorate) +- Notes: There is no research behind the belief that 「pursuing somebody to the end will move them」, but the detention and the no-contact order are in black and white. The record left behind can also affect sitting the civil service examination, sitting a public institution post, political review, and the professional qualifications required for some jobs. The line between pursuing somebody and harassing them does not depend on what you think, it depends on whether the other person has clearly said no. + +### 3. To judge whether the other person is interested, look at behaviour and not at 「signals」: who starts it, who arranges the next one, who follows up + +- Cost: No money, and not much time. The hard part is changing the habit: stop reading the other person's expressions and tone, and start recording facts instead. +- In plain language: 52 pairs of strangers were set talking for about ten minutes and then each judged whether the other person was flirting. The result is that 「the other person was not flirting」 is judged fairly accurately, while 「the other person was flirting」 cannot be picked up. Changing to 261 people watching videos to make the judgement gives the same conclusion. So stop guessing at expressions and tone, and look at behaviour: who starts it, who arranges the next meeting, and whether the other person follows up when you do not push. +- Benefit: Two studies measured how accurately people can identify flirting. The first had 52 pairs of strangers who did not know each other talk in pairs for 10 to 12 minutes. Afterwards each reported whether they themselves had been flirting and whether they thought the other person had been. The result is that 「没有发生调情的互动被判断得更准，发生了调情的互动反而判断得更差」. The second used 26 one-minute videos and 261 observers to make the judgement, and reached the same conclusion. People can tell reasonably reliably that 「the other person was not flirting」, but have a hard time telling that 「the other person was flirting」 +- Evidence grade: B +- Sources: Hall JA, Xing C, Brooks S (2015). Accurately detecting flirting: error management theory, the traditional sexual script, and flirting base rate. Communication Research, 42(7), 939-958. ; Perilloux C, Kurzban R (2015). Do men overperceive women's sexual interest? Psychological Science, 26(1), 70-77. (disputed side) +- Notes:Disputed. Other research holds that it is not that men overestimate women's sexual interest, but that women underreport their own intentions when filling in a questionnaire. Who is the more accurate judge is still unsettled. But both sides agree on one thing: inferring another person's thoughts backwards from small cues such as expression and tone has a low accuracy rate. Visible behaviour is far more reliable: whether the other person starts things, whether they arrange the next meeting, and whether they follow up when you do not push. + +### 4. Whether a relationship is good depends mainly on how you feel in it, not on the other person's circumstances + +- Cost: No money, and not much time. The hard part is changing where you look: put the checklist away and watch how things go instead. +- In plain language: What says most about whether a relationship is good is your own feeling in it, not the other person's circumstances. Such as whether you feel the other person is serious and whether they appreciate you, and how often the two of you argue. Your own feeling is between two and four times as good a predictor as what your partner says. But whether the relationship gets better or worse later is something no questionnaire can really work out. +- Benefit: A machine learning analysis of 29 laboratories and 43 longitudinal couples datasets, where longitudinal means following the same couples and recording them all the way through. The items that predicted the quality of the relationship best all came from how the people themselves felt about the relationship: feeling the other person's commitment, feeling appreciated, sexual satisfaction, feeling that the other person is satisfied, and conflict. The strongest of the items about the individual were life satisfaction, negative mood, depression, avoidant attachment and anxious attachment. At the start of the study these items about the relationship explained at most 45% of the variance, where variance means the difference in relationship quality between one couple and another, and 45% is the part of it these items can account for. By the end of the study they explained at most 18%. And the predictive power of 「the items people report themselves」 is between two and four times that of 「the items the partner reports」. Adding individual differences and the partner's reports on top of your own reported feelings about the relationship adds no predictive power at all +- Evidence grade: A +- Sources: Joel S, Eastwick PW, Allison CJ, et al. (2020). Machine learning uncovers the most robust self-report predictors of relationship quality across 43 longitudinal couples studies. Proceedings of the National Academy of Sciences, 117(32), 19061-19071. +- Notes: The same study also found that whether the relationship got better or worse during the study could not really be worked out from any set of self-report measures in the questionnaire. So for judging whether the relationship in front of you is good, asking yourself how you feel is more accurate than checking the other person's list of circumstances. Do not expect to be able to predict how it will turn out later. The item in this section called 「when it comes to emotional value, do not just ask whether it is there」 is talking about the same thing; the specific items are listed here. + +### 5. To close the distance, take turns answering a set of questions that go from light to deep, and do not learn scripts + +- Cost: No money. It takes 45 minutes. +- In plain language: Two strangers take turns answering 36 questions that go from light to deep, and after 45 minutes they are much closer than people who only chatted for the same length of time. What does the work is taking turns telling the truth and going a layer at a time deeper, not any script. This only shows that they were closer at the time, not that it can grow into a long-term relationship. +- Benefit: Strangers were paired up and made to take turns answering three sets of questions that get progressively deeper, 36 questions in all. The questions run from 「who would you most like to invite to dinner」 to 「when was the last time you cried in front of somebody else」. Closeness after 45 minutes was significantly higher than in the control group, which chatted for the same length of time. It was also higher than the average rating that students in that study gave for the most intimate relationship they actually have +- Evidence grade: B +- Sources: Aron A, Melinat E, Aron EN, Vallone RD, Bator RJ (1997). The experimental generation of interpersonal closeness: a procedure and some preliminary findings. Personality and Social Psychology Bulletin, 23(4), 363-377. +- Notes: The experiment was only run once, on university students, and it only measured closeness over a short period. It did not measure whether the relationship can grow into a long-term one. The closeness in the experiment came from two people taking turns and going a layer at a time deeper into their own real situation, not from any script, any routine or one side performing. So when you talk, you can 「say something true yourself, and ask one question just as deep」, taking turns on both sides. + +### 6. Long distance is not worse than living in the same city, and the real hurdle is the first three months after moving in together + +- Cost: A fair amount of money. You have to pay for travel, and pay for the time on the way there and back. You meet less often than people in the same city. The hard part is coping for a long time with not seeing each other. +- In plain language: A long-distance relationship is in fact steadier while you are apart, because you see each other less and it is easy to think of the other person as better than they really are. The real hurdle comes after you have moved in together. In the tracking data about half of long-distance couples reached the point of being in the same city, and the other half ended while they were still apart. Of the half that reunited, one third broke up within 3 months. +- Benefit: Two studies reach the same conclusion. A long-distance relationship is steadier than one in the same city while the two are apart. The reason is that the degree to which a long-distance couple imagines the other person to be better than they really are is markedly higher, and the studies measured idealisation distortion, romantic love, relationship memories and perceived agreement. And the less you meet, the better the picture. But that steadiness only holds while the two are still apart, and once they are in the same city, long-distance couples break up more easily. The proportion given by the study that tracked couples moving from long distance to the same city is this: about half of long-distance couples reached the point of moving in together, and the other half ended while they were apart. Of the half that reunited, one third broke up within 3 months of the reunion. The longer the gap between meetings while they were apart, and the stronger the idealisation, the less steady they are after the reunion +- Evidence grade: B +- Sources: Stafford L, Merolla AJ, Castle JD (2006). When long-distance dating partners become geographically close. Journal of Social and Personal Relationships, 23(6), 901-919. ; Stafford L, Merolla AJ (2007). Idealization, reunions, and stability in long-distance dating relationships. Journal of Social and Personal Relationships, 24(1), 37-54. ; Dargie E et al. (2015). Journal of Sex & Marital Therapy, 41(2), 181-202. (the long-distance study referred to in the notes) +- Notes: The people in these studies were dating relationships among university students in the United States, not marriages, and the studies are fairly old, so do not apply the absolute proportions directly. Another study surveyed 717 people long distance and 425 people in the same city. The difference in the quality of the two kinds of relationship was very small, and whether a relationship is good depends mainly on the two people's own circumstances and on the relationship itself. So when deciding whether to go long distance, do not use 「long distance always falls apart」 as your reason. What needs preparing in advance is the few months after the reunion, when there is less room to do things independently, more is known about each other, and conflict and jealousy both increase. + +### 7. Read the registration data before you listen to your elders: what the marriage rate and the divorce rate each measure + +- Cost: No money. Read the marriage registration section of the Ministry of Civil Affairs' annual bulletin, which takes a few minutes. +- In plain language: In 2024 there were 6,106,000 marriage registrations nationwide, about one fifth fewer than the year before, and 3,513,000 divorces. The claim online that the divorce rate is over half takes these two figures and divides one by the other, which comes to about 57.5%. But the people divorcing and the people marrying are not the same people, so it cannot be read as half of marriages ending in divorce. The marriage rate of 4.3‰ and the divorce rate of 2.5‰ in the bulletin are worked out by dividing by the national population. +- Benefit: In 2024 there were 6,106,000 marriages registered nationwide in accordance with the law, 20.5% fewer than the year before, and a marriage rate of 4.3 per thousand. In the same year there were 3,513,000 divorces completed in accordance with the law, of which 2,622,000 were divorces registered with the civil affairs departments and 891,000 were divorces granted by a court through judgment or mediation, and a divorce rate of 2.5 per thousand. The bulletin states that the marriage (divorce) rate = the number of marriages (divorces) in the year / the average total population in the year × 1,000‰. This is a 「crude rate」: the denominator is the whole population, not the married population +- Evidence grade: A +- Sources: Ministry of Civil Affairs (民政部) (2025). 2024 年民政事业发展统计公报（五（二）1 婚姻登记服务、注释 5），网页版与 PDF 版. ; +- Notes: The 「divorce rate over 50%」 you often see online is the number of divorces in the year divided by the number of marriages in the year, 351.3 / 610.6 ≈ 57.5%. This is called the 「divorce-to-marriage ratio」. The two figures being divided are not about the same group of people, and the bulletin has no such indicator, so it cannot be read as 「half of marriages end in divorce」. To work out 「the probability that a particular marriage ends in divorce」, you have to follow the same group of married people all the way through, and this section found no such official data. + +### 8. Count the health benefit, but discount it because the data is observational + +- Cost: No money, and no time. You only have to remember one thing: these figures say that married people have a lower mortality rate, which is two things happening at once, not that marriage makes you live longer. +- In plain language: The chance of dying in the same period is about 12% lower for married people than for the unmarried, and about 24% higher for people who have never married than for married people. But this is only two things happening at once, not an effect of getting married. People who are healthy, earn more and have more social contact are both more likely to marry and more likely to live longer. These figures come almost entirely from Europe and North America, and for use in China they are a sense of scale at best. +- Benefit: A meta-analysis on older people, which pools several studies and recalculates them, covered 53 comparisons and more than 250,000 people in all. The all-cause mortality risk ratio for married people relative to unmarried people was RR 0.88, 95% CI 0.85 to 0.91, where RR is the risk ratio and CI is the credible range, so the chance of dying in the same period is about 12% lower. In the same analysis, widowed was 1.11, divorced or separated 1.16 and never married 1.11, which are about 11%, 16% and 11% higher respectively. There is no difference between men and women. The most conservative estimate is RR 0.94, about 6% lower. A second meta-analysis compared never married with married, using 95 articles and more than 500 million people. The result was HR 1.24 (1.19 to 1.30), so the chance of dying in the same period is about 24% higher. And the older the age group, the smaller the gap. A further meta-analysis split by sex, covering 21 studies and more than 7,890,000 individuals, found that for both men and women being unmarried went together with all-cause mortality, cancer mortality and cardiovascular mortality. On the men's side the gap in cardiovascular mortality and in all-cause mortality is larger. Unmarried men, relative to unmarried women, had 31% more stroke deaths and 9% more all-cause deaths +- Evidence grade: A (disputed) +- Sources: Manzoli L, et al. (2007). Marital status and mortality in the elderly: a systematic review and meta-analysis. Soc Sci Med 64:77–94. ; Roelfs DJ, et al. (2011). The rising relative risk of mortality for singles: meta-analysis and meta-regression. Am J Epidemiol 174(4):379–389. ; Wang Y, et al. (2020). Sex differences in the association between marital status and the risk of cardiovascular, cancer, and all-cause mortality: a systematic review and meta-analysis of 7,881,040 individuals. Glob Health Res Policy 5:4. +- Notes:Disputed. What the dispute is about is that married people were selected from the start: people who are healthy, earn more and have more social contact are both more likely to marry and more likely to live longer. All three meta-analyses use observational data on existing groups, that is, records that were followed but not split into a control and a treatment group. The authors themselves write about publication bias and methodological limits. Roelfs also found that the more rigorous the study, the smaller the gap that comes out. These figures are mostly about European and North American populations, and for China they are a sense of scale at best. + +### 9. Work out the time account from unpaid work, and settle the division of tasks before you register + +- Cost: No money. Spend an hour or two before the wedding turning the unpaid work, such as housework, looking after the old and the young, and whose job will give way, into a clear division of tasks, and then work to it. The hard part is laying it out and talking it through rather than letting it drag. +- In plain language: In the national time use survey, unpaid work such as housework and looking after family takes men 1 hour 52 minutes a day on average and women 3 hours 29 minutes. The survey does not break the figures down by whether people are married, so there is no official figure for 「how much more housework there is after marriage」, and this gender gap can only be used as an upper limit. The division of tasks can be negotiated into a different shape before you register. +- Benefit: The 2024 National Time Use Survey. Unpaid work means things like housework and looking after family that bring no pay. Among the people who did this kind of work on the day, it took 2 hours 45 minutes a day on average, of which men accounted for 1 hour 52 minutes and women 3 hours 29 minutes. The share of people who did this kind of work during the day is 67.5% for men and 83.9% for women. Housework among residents averages 1 hour 17 minutes a day, and 1 hour 59 minutes counting only the people who did it, which is 28 minutes less than in 2018. Looking after and caring for family averages 30 minutes a day, and 1 hour 46 minutes counting only the people who did it. The 2018 figures are: housework 45 minutes for men and 2 hours 6 minutes for women, and looking after and caring for children 17 minutes for men and 53 minutes for women +- Evidence grade: A +- Sources: National Bureau of Statistics (国家统计局) (2024). 第三次全国时间利用调查公报（第二号、第三号）. ; ; National Bureau of Statistics (国家统计局) (2019). 2018 年全国时间利用调查公报. +- Notes: Both surveys only break the figures down by sex, by urban or rural area and by age, and not by whether people are married. So there is no official figure for 「how much more housework there is after marriage」, and the gender gap can only be used as an upper limit. The figures are national averages, and the two of you can quite well negotiate a different division of tasks. + +### 10. Start with the default rules in the law for the money, then decide whether to put them in writing + +- Cost: No money. Read the articles on property in the marriage and family book of the Civil Code. If you need to, the two of you sign a written property agreement; the law only requires it to be in writing, and there is no need to have it notarised. The hard part is saying the money things out in advance. +- In plain language: Without a written agreement, the law treats the wages and bonuses earned after the wedding, income from running a business and from investment, and property inherited or received as a gift, as jointly owned by the two of you. Property owned before the wedding, and property that a will or a gift contract says goes to one of you alone, is personal property. The two of you can agree in writing that everything stays separate, and once you have signed it binds you. The party who put in more work bringing up children and looking after the elderly can ask for compensation on divorce. +- Benefit: The default rules where nothing else is agreed are these: wages, bonuses and remuneration for labour earned after the wedding, income from production, business operations and investment, income from intellectual property, and property inherited or received as a gift, belong jointly to the couple (Article one thousand and sixty-two). Property owned before the wedding, compensation for personal injury, property that a will or a gift contract makes clear goes to one of them alone, and articles for personal use are personal property (Article one thousand and sixty-three). The two of them may agree in writing that property before and after the wedding belongs to each of them separately, belongs to both jointly, or is partly separate and partly joint, and that agreement has legal force for both (Article one thousand and sixty-five). The party who has taken on more obligations by bringing up children, looking after the elderly or helping the other party with their work has the right to ask for compensation on divorce (Article one thousand and eighty-eight) +- Evidence grade: A +- Sources: National People's Congress (全国人民代表大会) (2020). Civil Code of the People's Republic of China, 中华人民共和国民法典（2021 年 1 月 1 日起施行，第一千零六十二、一千零六十三、一千零六十五、一千零八十八条）, National Database of Laws and Regulations. ; full text of the articles reprinted in the Supreme People's Court Gazette. +- Notes: There are no official figures for bride price, for the wedding or for the marital home, so this item gives no numbers. How much the compensation under Article one thousand and eighty-eight comes to is first for the two of you to talk about; if that fails, the court decides it, and there is no fixed formula. A written property agreement works for both sides: the higher earner has a use for it, and so does the party who does a lot of the unpaid work of housework and care. + +### 11. When your parents put money towards a flat, write down at the point of the transfer whether it is a loan or a gift + +- Cost: No money. On the day the money goes out, write a single page saying whether it is a gift to you or a loan to you, and have both the parents who put the money in and the children sign it. If it says a loan, have both of you sign it too, and write the same sentence in the transfer reference. +- In plain language: Money from parents before the wedding is by default for their own child. Money put in after the wedding with no agreement counts as joint marital property. From 2025-2, where one party's parents paid the whole amount after the wedding and did not state that it was only for their own child, the court can award the flat to the child who provided the funding on divorce and then decide whether the other side is to be compensated. If it is to be a loan, write it down on the day and have both sides sign. +- Benefit: Where parents contributed towards a home for the couple before the wedding, it is treated as a gift to their own child alone. Where the parents have clearly stated that the gift is to both, that is excluded. Money put in after the wedding is dealt with by agreement. Where there is no agreement, or the agreement is unclear, it is dealt with on the principle of joint marital property. There is one more rule from 2025-2-1. Where during the marriage one party's parents paid the whole amount and the gift contract does not say it is only for their own child, then on the division of property at divorce the court may award the flat to the child of the person who provided the funding. Once it has been awarded, several factors are weighed together to decide whether the other side is to be compensated and how much. Those factors are whether the couple lived together and had children together, the fault that led to the divorce, the size of each party's contribution to the family, and the market price of the flat at the time of the divorce. Where one party's parents put in part of the money, or where both sets of parents put money in, the flat is awarded to one of the parties on the basis of the source and the proportion of the funding, with reasonable compensation (nationwide) +- Evidence grade: A +- Sources: Supreme People's Court (2020). Interpretation (One) on the Application of the Marriage and Family Book of the Civil Code, 关于适用《民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第二十九条）. ; Supreme People's Court (2025). Interpretation (Two) on the Application of the Marriage and Family Book of the Civil Code, 关于适用《民法典》婚姻家庭编的解释（二）（法释〔2025〕1 号，第八条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零六十二条第一款第四项）. +- Notes: Writing it down is useful for both sides. The parents who put in the money no longer have to hope for an IOU to turn up at the divorce in order to protect their contribution. Once Interpretation (Two) is in force, who provided the money is itself a basis for dividing the flat, and whose name is on the title no longer decides who owns it. The party who accepted the funding also no longer has to wait until the divorce to find out what that money was. If it is to be treated as a loan, then treat it as one properly: both sides agree that it is a loan, agree how it is to be repaid, write that in the transfer, and keep a record of the repayments where necessary. Leave out any one of those and you will be explaining it in court. + +### 12. A large sum your spouse borrowed alone does not become your debt if you neither signed nor afterwards approved it + +- Cost: No money. Do not sign as a 「joint borrower」 or 「guarantor」 on your spouse's IOU or guarantee. And do not write things like 「we will repay this money」 in a chat, because that can be read as your having afterwards approved the debt. The hard part is saying nothing when you are being pressed. +- In plain language: Money your spouse borrowed behind your back does not automatically become your debt, as long as it clearly goes beyond the household's everyday spending and you neither signed nor later accepted it. The person owed the money has to prove that it was used for life together or for a business run together. But money spent on a flat, where that really did become a home you live in together or share, is a level the other side can often prove. +- Benefit: Debts incurred in the following situations are joint marital debts. First, where both spouses signed. Second, where one spouse afterwards approved it, that is, accepted the debt afterwards, so that both of you acknowledge it. Third, where during the marriage one spouse borrowed in their own name for the everyday needs of the family. Money borrowed in one spouse's own name that clearly goes beyond the household's everyday spending is not a joint marital debt. The exception is where the creditor, the party owed the money, can prove that the money was used for life together or for joint production and business operations, or where both of you acknowledge the debt. Money spent on a flat clearly goes beyond the everyday needs of the household, and the burden of proof lies with the party claiming the debt, so the person owed the money has to produce the evidence. But if that money really did become a home the couple lives in together or shares, the creditor can often reach the level of 「used for life together」. So the step that does more good is the one in item 11 in this section (writing down whether the money from parents is a loan or a gift) (nationwide, in force from 2021) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零六十四条）. ; Supreme People's Court (2020, second amendment). Provisions on Several Questions Concerning the Application of the Law in Trying Cases of Private Lending, 关于审理民间借贷案件适用法律若干问题的规定（第十五、十六条）. (Supreme People's Court Gazette) +- Notes: When property is divided at a divorce and the other side's parents suddenly produce a loan agreement signed only by your spouse, there are two steps. The first is to ask whether the loan actually happened. If you say the money was never handed over and can make your reasons hold together, the court will weigh several things together. First the size of the loan, how the money was handed over, and the two parties' financial positions, and then how you normally do transactions, whether there is a matching change in the property, and what the witnesses say. Where the other side can produce nothing but a transfer receipt, it is still for them to prove that the loan exists. The second step, once the loan exists, is to ask whether it is a joint debt. For the general rules on IOUs and guarantees, see item 18 in section 8 + +### 13. To register you only need your identity card and a signed statement; you do not need the household register, you pay nothing, and the certificate is issued on the spot + +- Cost: No money. Both of you have to be there in person. Once it is done it usually takes a few tens of minutes. +- In plain language: From 2025-5, registering needs only two things: your own identity card and a signed statement. The statement says that you have no spouse, and that you are not a direct blood relative of the other party nor a collateral blood relative within three generations. You do not have to bring the household register, and you do not have to go back to the household registration place of one party to do it. The registration costs nothing, and if you qualify the certificate is issued on the spot. +- Benefit: The new Regulations on Marriage Registration, in force from 2025-5-10, changed the documents needed for mainland residents to marry. The article says only two things have to be produced:「（一）本人的居民身份证；（二）本人无配偶以及与对方当事人没有直系血亲和三代以内旁系血亲关系的签字声明」. The household register is no longer a required document. The registration authority is 「县级人民政府民政部门或者省、自治区、直辖市人民政府按照便民原则确定的乡（镇）人民政府」. The restriction in the old regulations, 「到一方当事人常住户口所在地」, has been removed. The regulations also state that「婚姻登记机关办理婚姻登记，不得收取费用」, and that「对当事人符合结婚条件的，应当当场予以登记，发给结婚证」 +- Evidence grade: A +- Sources: State Council (国务院) (2025). Regulations on Marriage Registration, 婚姻登记条例（国令第 804 号，第二、六、八、十条）. +- Notes: There are only four situations in which the registration authority will not do it: the legal age for marriage has not been reached, the two of you are not both entirely willing, one or both of you already have a spouse, or the two of you are direct blood relatives or collateral blood relatives within three generations. The only things that could cost money are side services such as photographs and photocopies. The registration itself costs nothing, so if you are asked for a 「工本费」 or a 「rush fee」, refuse and complain. For the 30-day cooling-off period for divorce registration, see the item on the cost of getting out in this section + +### 14. Have a premarital medical check once before the wedding, and if there is a serious illness you must tell the other party honestly before you register + +- Cost: A little money. In most areas it is free or reduced for people of the right age. Paying yourself costs only a few hundred yuan. It takes half a day. +- In plain language: If you have a serious illness and did not tell the other party honestly before the registration, the other party can ask the court to annul the marriage. The application has to be made within one year of finding out. The medical check itself has not been a precondition for the certificate since 2003, and whether you go is voluntary. The cost depends on local policy, and it takes half a day. +- Benefit: The legal consequence of concealment is clear. The new Regulations on Marriage Registration provide that「一方当事人患有重大疾病的，应当在结婚登记前如实告知另一方当事人；不如实告知的，另一方当事人可以依据民法典第一千零五十三条的规定向人民法院请求撤销婚姻」. The state treats the premarital medical check as the first line of defence against birth defects. The current targets are that「婚前医学检查率、孕前优生健康检查目标人群覆盖率分别保持在 70% 和 80% 以上」. The document also requires that「推进婚前医学检查机构与婚姻登记场所就近设置方便群众」, and promotes「婚姻登记、婚前医学检查、生育指导『一站式』服务」. The Civil Code provides that where one party has a serious illness they must tell the other honestly before the marriage registration, and where they do not, the other party may ask the people's court to annul the marriage. An application to annul a marriage must be made within one year from the day the person knew or should have known the ground for annulment +- Evidence grade: B +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零五十三条）. ; State Council (国务院) (2025). Regulations on Marriage Registration, 婚姻登记条例（国令第 804 号，第十二条）. ; General Office of the National Health Commission (国家卫生健康委办公厅) (2023). 出生缺陷防治能力提升计划（2023—2027 年）（国卫办妇幼发〔2023〕9 号）. ; National Health Commission and four other departments (国家卫生健康委等五部门) (2020). Notice on Strengthening Premarital Health Care, 关于加强婚前保健工作的通知（国卫妇幼函〔2020〕205 号）：「积极倡导各地结合实际推行免费婚前检查」. +- Notes: The medical check has not been a step you have to take before the certificate since 2003, and whether you go is voluntary. But the rule on「a serious illness not disclosed honestly」 does give the other party the right to go to court to have the marriage annulled, and the application has to be made within one year of finding out. There is no national rule making the check free. The 2020 notice of the National Health Commission and four other departments only says「积极倡导各地结合实际推行免费婚前检查」. So whether it is free and which items are covered are set locally; ask at the local maternal and child health centre, or at the one-stop window at the marriage registration office. + +### 15. When it comes to emotional value, do not just ask whether it is there; look at the quality of the relationship + +- Cost: No money. Before the wedding, use a few conflicts and a few months of living together to judge the quality of the relationship, and do not only check a list of circumstances. The hard part is holding out against judging from the list. +- In plain language: The better the quality of a marriage, the healthier the people in it, but the relationship is weak, on a par with the effect of habits like diet on health. To judge whether a relationship is good, use a few conflicts and a few months of living together, and do not only check a list of circumstances. Whether a bad marriage is worse for you than not marrying at all has not yet been compared directly with data. +- Benefit: A meta-analysis of 126 studies and more than 72,000 people shows that the better the quality of the marriage, the better the health. The effect size r lies between 0.07 and 0.21. The r is how far two things change together, where 0 is no relationship at all and 1 is changing in perfect step. The mortality risk is r = 0.11, and the cardiovascular reaction during conflict is r = -0.13. This is the same order of magnitude as the effect of health behaviours such as diet. In the studies that specifically tested for a difference between the sexes, essentially no difference was found between men and women. The review in the cardiovascular field states that marital dissatisfaction and marital quality have a significant effect on cardiovascular risk +- Evidence grade: B (disputed) +- Sources: Robles TF, et al. (2014). Marital quality and health: a meta-analytic review. Psychol Bull 140(1):140–187. ; Dhindsa DS, et al. (2020). Marital status and outcomes in patients with cardiovascular disease. Trends Cardiovasc Med 30(4):215–220. +- Notes:Disputed. What the dispute is about is that the association itself is weak. Some of the results show publication bias. And most of the studies surveyed at a single point in time, so they cannot show who led to what; Robles lists these limits in the original text. This section found no meta-analysis that compares the mortality rate of「a marriage of low quality」directly with that of「never married」, so「a bad marriage is worse than not marrying」is for now only an inference, not a conclusion from figures. + +### 16. Work out the cost of getting out: a divorce by agreement has a 30-day cooling-off period, and a divorce by litigation has statutory conditions + +- Cost: No money. Before the wedding, spend some time reading: if you do have to get out, how long it takes and what procedure it follows. The hard part is thinking about splitting up before you have married. +- In plain language: A divorce by agreement is not finished once it is signed. Within 30 days of submitting the application, either party may withdraw it. At the end of that period you also have to collect the certificate in person within 30 days, and not doing so counts as withdrawing. If the other side does not agree, you can only sue. The court tries mediation first, and where there is domestic violence, bigamy, or two years of living apart, and mediation fails, it grants the divorce. +- Benefit: A divorce by agreement requires a written divorce agreement, and both parties must apply in person at the marriage registration authority (Article one thousand and seventy-six). Within 30 days from the day the registration authority receives the application, either party may withdraw it. Within 30 days after that period ends, both parties must collect the divorce certificate in person, and not collecting it is treated as withdrawal (Article one thousand and seventy-seven). Where one side does not agree, the route is litigation, and the court should try mediation first. Where there is bigamy or cohabiting with another person, domestic violence or abuse and abandonment, repeated gambling or drug use that has not been corrected, or two years of living apart because of a breakdown in the relationship, and mediation has no effect, the divorce is to be granted. Where a divorce was refused and the two have since lived apart for another year, a fresh suit is to be granted (Article one thousand and seventy-nine). In 2024 there were 3,513,000 divorces nationwide, of which 891,000 went through a court judgment or mediation +- Evidence grade: A +- Sources: National People's Congress (全国人民代表大会) (2020). Civil Code of the People's Republic of China, 中华人民共和国民法典（第一千零七十六、一千零七十七、第一千零七十九条）. ; Ministry of Civil Affairs (民政部) (2020). Notice on Implementing the Provisions on Marriage Registration in the Civil Code of the People's Republic of China, 关于贯彻落实《中华人民共和国民法典》中有关婚姻登记规定的通知（民发〔2020〕116 号）. ; Ministry of Civil Affairs (民政部) (2025). 2024 年民政事业发展统计公报. Citation the same as item 7 in this section (what the marriage rate and the divorce rate each measure) +- Notes: The cooling-off period applies only to a divorce by agreement. There is no cooling-off period for a divorce by litigation, but that takes longer. Court fees and legal fees differ from place to place, so this item gives no figures. A high cost of getting out is not a reason not to marry, but you should know it before you do. + +### 17. Keep a separate account for 「marrying because of the older generation」, and do not mix it with your own + +- Cost: No money. Draw two columns on a piece of paper. One column is which benefits in this marriage are your own, which the earlier items in this section can work out. The other is which benefits belong to the older generation: peace of mind, face, and the pressure to marry stopping. The hard part is keeping the older generation's column separate and not merging it into your own. +- In plain language: The benefits that go to the older generation, that is peace of mind, face and no more being nagged, are real. But they do not automatically become your health benefit or your relationship quality. The time account, the money account and the cost of getting out, on the other hand, are all borne by you. Draw two columns on paper and write out which benefits are yours and which belong to the older generation, and whether the deal is worth it is obvious at a glance. +- Benefit: The benefit to the older generation is real. But it does not automatically turn into your health benefit (item 8) or into relationship quality (item 15). And the time account (item 9), the money account (items 10 to 12) and the cost of getting out (item 16) are all borne by the parties themselves. Once the books are kept separately, whether「marrying to complete a task」is good value is something you can see for yourself +- Evidence grade: C +- Sources: The author's own experience, a methodological suggestion; for the figures see the earlier items in this section +- Notes: The older generation's column can also be counted as a heavy one, and how you weigh it is your own decision. For the list to fill in, see [docs/结婚划不划算.md](../docs/结婚划不划算.md) + +### 18. Treat how a relationship is going as a health account: in an argument, argue about the thing and not the person, and do not insult or sneer + +- Cost: No money. In an argument, stick to the thing in front of you, do not insult, do not sneer and do not bring up old grievances. The hard part is keeping your mouth shut while you are angry. +- In plain language: Couples with a good relationship are somewhat healthier and have a somewhat lower risk of death. The association is small, on a par with the association between diet and health. In one experiment, couples who stayed hostile throughout their fights healed the small wounds the experiment produced at only six tenths the rate of couples who stayed calm. +- Benefit: Pooling 126 studies and more than 72,000 people, which is what a meta-analysis does. The better the quality of the marriage, the better the health, with a correlation coefficient r between .07 and .21. A correlation coefficient of 0 is no relationship at all and 1 is changing in perfect step. The correlation with a lower risk of death is r = .11. A smaller cardiovascular reaction during conflict is r = -.13. The authors say that all these effects are small, on a par with the association between health habits such as diet and health. In another crossover experiment, 42 healthy couples each stayed in hospital twice, for 24 hours each time. Once they had a mutually supportive conversation, and once they discussed a subject on which they disagreed. After the argument, the blister wounds the experiment produced healed more slowly. For the couples who showed more hostility on both occasions, the wounds healed at only 60% the rate of the couples with low hostility. On the morning of the day after the argument, their blood markers of inflammation (IL-6, tumour necrosis factor) had also risen more +- Evidence grade: B +- Sources: Robles TF, Slatcher RB, Trombello JM, McGinn MM (2014). Marital quality and health: a meta-analytic review. Psychological Bulletin. ; Kiecolt-Glaser JK et al. (2005). Hostile marital interactions, proinflammatory cytokine production, and wound healing. Archives of General Psychiatry. +- Notes: The reason for grade B is that most of the pooled studies are only records that were followed, with no control group. So it cannot be told whether a bad relationship makes you ill or poor health makes the relationship worse. In some studies the association is still there after age and income are taken out. The wound experiment has only 42 couples in it. 「Argue about the thing and not the person, do not insult and do not sneer」 is a practice the author worked out from the 「hostile behaviour」 in the experiment, and no study compares it directly. Where the arguments are fierce and cannot be changed, the two of you can go together to marriage and family counselling, and that too has no separate literature. Where it has come to hitting, or to long-term verbal abuse and threats, that is no longer a matter of communication but of domestic violence: see item 43 in section 8 (if you are being abused at home). For deciding whether to go through with it, see the cost of getting out in item 16 in this section. + +### 19. Same-sex partners: while you are both clear-headed, get the power of attorney, the voluntary guardianship and the will done. In law you are not close relatives, and without these documents you have no right to sign and no right to inherit + +- Cost: A little money. Voluntary guardianship and a will are done at a notary office, and the notary fees differ from place to place, starting at a few hundred yuan. A will you write by hand costs nothing. It takes one or two half-days from start to finish. The hard part is the two of you having to sit down and talk about illness and death. +- In plain language: Close relatives in law are a spouse, parents, children and siblings, and a partner is not among them. Without the documents in place, when one of you falls seriously ill the other cannot decide for them. When one of you dies, the other inherits not one yuan. The three documents each cover a stretch: the power of attorney for while you are alive, voluntary guardianship once the capacity to make decisions is lost, and the will for after your death. +- Benefit: The close relatives listed in Article one thousand and forty-five of the Civil Code are a spouse, parents, children, siblings, grandparents on both sides, and grandchildren on both sides. The statutory heirs in Article one thousand one hundred and twenty-seven are first a spouse, children and parents, and then siblings and grandparents on both sides. A partner is in neither list. Article one thousand two hundred and nineteen deals with surgery, special examinations and special treatment. Where it cannot or should not be explained to the patient, the medical personnel must explain it to the patient's 「close relatives」 and obtain their consent. There are three ways to make up the gap. First, article thirty-three allows an adult and 「another individual who is willing to be the guardian」 to agree in advance, in writing, who the guardian is. Once the person has lost all or part of their capacity for civil conduct, that person becomes the guardian, and the guardian is the statutory agent. Second, Article one hundred and sixty-five provides that a written power of attorney must state the name of the agent, the matters to be handled, the scope of the authority and the period, and must be signed or sealed by the principal. When the principal dies, the mandate ends. Third, Article one thousand one hundred and thirty-three allows a will to be made leaving personal property to an individual other than a statutory heir. An article posted on the Hubei Provincial Department of Justice website by the Yichang Municipal Bureau of Justice states that same-sex partners can have a voluntary guardianship agreement notarised, naming the other partner as their future guardian (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第二十三、三十三、一百六十五、一百七十三、一千零四十五、一千一百二十四、一千一百二十七、一千一百三十三至一千一百四十、一千一百四十二、一千二百一十九条）. (National Database of Laws and Regulations) ; Yichang Municipal Bureau of Justice (宜昌市司法局) (2021). 意定监护公证：通向彩虹的公证之路. 湖北省司法厅网站「实践荟萃」栏目. +- Notes: The status a partner has in an inheritance is 「legatee under a will」, not heir. Once you know of the bequest, you have to express your acceptance within 60 days, and not expressing it counts as declining. The partner themselves, and anybody who stands to benefit from the will, cannot be a witness to it. Where several wills have been made, the last one governs: see item 2 in section 17 (a later will overturns an earlier one). Whether a voluntary guardian may sign the consent form for an operation differs from hospital to hospital, so ask the hospital you normally use in advance. If the family does not accept these documents, it comes to a court case: six months or more in the ordinary procedure at first instance, and legal fees are yours to pay. So get the documents done at a notary office where you can. For a flat bought together and money saved together, write a separate agreement on the basis of what each of you actually put in. For how voluntary guardianship works, see item 1 in section 17 (naming a future guardian in writing). The people who benefit are you and your partner. + +### 20. Do not marry while concealing your sexual orientation, and do not treat a marriage in name only as the easy way out: in law it is a real marriage, and the only way out is a divorce + +- Cost: No money. The hard part is standing up to the pressure from the family to marry, which may take years to bear. +- In plain language: Marrying while concealing your sexual orientation is not void in law and cannot be annulled, so the only way out is a divorce. A marriage in name only is a real marriage as well, and the money earned and the debts incurred after it count as the couple's. The side that was kept in the dark can, on divorce, ask for a larger share of the property and for compensation for mental harm, but whether the court agrees is not settled. +- Benefit: Article one thousand and fifty-one of the Civil Code says a marriage is void in only three situations: bigamy, a relationship of close relation that is forbidden for marriage, and not having reached the legal marriageable age. Articles one thousand and fifty-two and one thousand and fifty-three say only two situations can be annulled: marrying under coercion, and concealing a serious illness before the wedding. Article seventeen of Interpretation (One) on the Marriage and Family Book of the Civil Code provides that where a request is made to confirm a marriage void on any ground other than those three, the court rules to reject it. So the only way out is a divorce. A marriage in name only is a lawful marriage once registered. Article one thousand and sixty-two provides that wages, bonuses and remuneration for labour after the wedding are joint marital property. Article one thousand and sixty-four provides that a debt signed for jointly by the two, and a debt one of them took on for the everyday needs of the household, are joint marital debts. For damages on divorce, see Article one thousand and ninety-one, which is limited to bigamy, cohabiting with another person, domestic violence, abuse or abandonment of a family member, and 「having some other serious fault」. Article two of Interpretation (One) limits 「cohabiting with another person」 to living with an 「opposite-sex partner outside marriage」. So a same-sex relationship can only be argued as 「some other serious fault」. The compensation includes payment for mental harm, which the side kept in the dark has to claim at the same time as suing for the divorce; if the divorce is refused, it is not supported. Since the Civil Code came into force, no judgment of this kind published by a court has been found. Under the old law, an article posted on the website of the Jiangsu High People's Court in 2013 discussed a case in which a wife sued for divorce because her husband had concealed that he was homosexual. The article sets out two views. One holds that homosexuality was not among the faults recognised by the law at the time, so neither a larger share of the property nor mental compensation should be supported. The author prefers the other: treat the side kept in the dark as the 「party without fault」, take that into account when dividing the property, and also pay compensation for mental distress (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零五十一至一千零五十四、第一千零六十二、第一千零六十四、第一千零七十九、第一千零八十七、第一千零九十一条）. (National Database of Laws and Regulations) ; Supreme People's Court (2020). Interpretation (One) on the Application of the Marriage and Family Book of the Civil Code, 关于适用《中华人民共和国民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第二、十七、八十六至九十条）. ; Zhu Laikuan, Wang Hongjuan (2013). 因同性恋导致离婚能否主张损害赔偿？江苏法院网. +- Notes: The larger share of the property and the mental compensation rest on a single analytical article that does not say how the case finally came out, so whether you can get it depends on the particular court. The Civil Code has one more ground than the old law, 「some other serious fault」, which gives the side kept in the dark something to argue with. The text of the law itself is grade A. Where both parties in a marriage in name only have same-sex partners, the partners are not protected by the marriage law between them, and the property has to be set out separately: see item 19 in this section (same-sex partners: voluntary guardianship and a will). A divorce from a marriage in name only also has to go through the 30-day cooling-off period or a court case: see item 16 in this section (a divorce by agreement has a 30-day cooling-off period). Where the agreement says that compensation is given up, a later suit will not be supported. The person who benefits first is the spouse who was kept in the dark, and second is yourself: once the concealment is found out, the divorce, the division of property and the custody of any children all arrive together. diff --git "a/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" "b/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" new file mode 100644 index 00000000..1f2d985f --- /dev/null +++ "b/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" @@ -0,0 +1,164 @@ +[← 回总目录](../README.md) + +# 11. 程序员和技术人容易踩的红线 + +本节口径是人身自由与金钱，换回的是「别被关」和「别赔钱」。写代码、做技术的人，常把下面这些事当成「只是技术问题」，可它们对应的是刑法里的具体罪名，或者民事上要赔的具体一笔钱。每条都给出法条里的数字，以及最高人民检察院、最高人民法院官网公布的案例。刑法引用的是北京市公安交管局转载的「根据刑法修正案（十一）修正」整合文本。条目顺序按「技术人最容易以为没事 × 后果最重」从高到低排。第 1 条是通用的一条，讲动手前怎么判断，以及出事之后先做什么。 + +[← Back to the contents](../README.md) + +# 11. Lines programmers and tech workers cross + +The measures in this section are personal freedom and money, and what you get back is 「not being locked up」 and 「not having to pay」. People who write code and do technical work often treat the things below as 「only a technical problem」, but they correspond to a specific charge in the Criminal Law, or to a specific sum you have to pay in civil law. Each item gives the figures in the articles of law, together with the cases published on the websites of the Supreme People's Procuratorate and the Supreme People's Court. The Criminal Law is cited from the consolidated text as amended by Criminal Law Amendment Eleven, reprinted by the Beijing Municipal Public Security Traffic Management Bureau. The items are ordered from high to low by 「what a technical worker is most likely to think is harmless × the heaviest consequences」. Item 1 is the general one: how to judge before you act, and what to do first once something has happened. + +### 1. Ask three questions before you act: whose interests have you harmed, how capable the other side is of pursuing it, how much evidence have you left; and get a criminal lawyer as soon as you are pursued + +- Cost: No money. Before you act, replace the question 「can this be done technically」 with another one: will the vendor, the platform or the company report it to the authorities? The hard part is holding out against the thought that 「nobody will do anything about something this small」, because server logs, payment records and group chat records are all produced by default when a case is handled. +- In plain language: Before you act, ask one question: will the vendor, the platform or the company report it to the police? Server logs, payment records and group chat records can all be produced when a case is handled. If you are held in criminal detention, the decision on whether to arrest you is made within at most 37 days. From the first questioning you can already have a lawyer, and those 37 days are when getting a criminal lawyer does the most good. +- Benefit: Any unit or individual that finds facts of a crime has the right to report it to the authorities. Where the case-handling organ finds facts of a crime, a case must be opened. After you have been held in criminal detention, the police must apply for approval of the arrest within 3 days, that is, ask the procuratorate to arrest you formally. In special circumstances this may be extended by 1 to 4 days. In cases such as being on the run, where offences are committed in several places in succession, it may be extended to 30 days. The procuratorate then has 7 days to decide. Added together, the decision on whether to arrest is made within at most 37 days. From the first interrogation, or from the moment a compulsory measure is taken against you, that is detention or arrest, you may appoint a lawyer. While the case is still with the police for investigation you may appoint a lawyer. When the lawyer applies to see you, the detention centre must arrange it within 48 hours at the latest. Where the possible sentence is public surveillance, criminal detention or some punishment lighter than prison, or where only an additional punishment is imposed with no principal punishment, you may be released on bail pending trial: pay a deposit or find a guarantor and go home while the case is handled. Release on bail lasts at most 12 months (nationwide) +- Evidence grade: C +- Sources: National People's Congress (全国人大) (2018). Criminal Procedure Law, 刑事诉讼法（2018 年修正，第三十四、三十九、六十七、七十九、九十一、一百一十、一百一十二条）. +- Notes: Those 37 days are counted from detention to the procuratorate's decision on whether to approve the arrest; they are not a limit on how long you may be held. After the arrest is approved there is still the custody period during the investigation, and the counting continues from there. Those 37 days are also when a lawyer who knows the field is most useful: getting to see you, handing the procuratorate a written view that the arrest should not be approved, and applying for release on bail. Looking for a lawyer after the prosecution has begun is too late. Every other item in this section is about the same thing: whether the other side has grounds to report you. + +### 2. Do not make or sell game cheats and cheats, including AI assistance that only reads the screen and never touches memory + +- Cost: No money. The price is giving up this kind of contract work and the income from 「small tools」. The hard part is that the money comes in quickly and it is not obvious who would come after you. +- In plain language: Selling cheats to 20 users, or earning 5000 yuan, is enough for 3 years or less, and especially serious cases carry 3 to 7 years. One man made wallhack and aimbot cheats, sold more than 80,000 yuan of them in 3 months, and was sentenced to 1 year suspended for 1 year with a fine of 5000 yuan. Another used server-side programs to run a private server, took in more than 20,000,000 yuan of top-ups, and got 4 years at the most, with a fine of up to 10,000,000 yuan. +- Benefit: Providing programs or tools made specifically for intruding into or illegally controlling computer information systems, where the circumstances are serious, carries 3 years or less, with a fine imposed in addition or instead. An additional fine means a fine on top of the sentence; instead means a fine alone. Especially serious circumstances carry 3 to 7 years and a fine. The threshold for 「serious circumstances」 is providing the tool to more than 20 users, or illegal gains of more than 5000 yuan. Illegal gains means the money you earned out of it. Deliberately going around or breaking the technical protection measures the rights holder has put in place may also amount to the crime of infringing copyright, which carries 3 years or less, and 3 to 10 years where it is serious. The official cases: Wang and others made and sold cheats that read and modified the game's memory to give wallhacks and aimbots. They sold more than 80,000 yuan in 3 months and made over 30,000 yuan. He was sentenced to 1 year suspended for 1 year, with a fine of 5000 yuan. A suspended sentence of 1 year means he was sentenced to 1 year but does not go in at first, and if he does not offend again during that year he does not have to serve it. Xiang and others sold the 「TZ」 series of cheats with sales of more than 200,000 yuan, and Xiang was sentenced to 3 years suspended for 3 years with a fine of 20,000 yuan. Liu and others obtained the game's server-side program, changed the parameters and ran a private server, taking in more than 2000 余万元 in top-ups. The sentence was 4 years to 2 years and 6 months, with a fine of from 10,000,000 yuan down to 1,300,000 yuan (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第三款、第二百一十七条第六项）. ; Supreme People's Court and Supreme People's Procuratorate (2011). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases Endangering the Security of Computer Information Systems, 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第三条）. (reprinted by the Shenzhen Municipal Public Security Bureau) ; Supreme People's Procuratorate (2026). 上海静安：严惩制售游戏外挂黑灰产共护清朗网络空间. ; Supreme People's Procuratorate (2021). 2020 年度检察机关保护知识产权典型案例（案例五）. +- Notes: The cheat cases that have been published are all the kind that read and write game memory and get round the anti-cheat. No official case has yet been seen for assistance that 「only takes screenshots and only simulates key presses」. But the ticket-grabbing case in item 3 shows that not touching memory and only sending requests can still amount to the crime of illegally obtaining data. Whether it is a crime in the end, and under which charge, is settled by the case-handling organ on the basis of the material handed in when the vendor reports it, and not by the author. Selling to 20 users is enough for 「serious circumstances」, and that is a small number. + +### 3. Do not write or sell ticket-grabbing, flash-sale, order-brushing or coupon-collecting scripts, not even one that only 「clicks the button automatically」 + +- Cost: No money, and no time either. +- In plain language: A script does not change memory and does not crack anything; it is simply faster at sending requests than a person, and it is still convicted. Earning 5000 yuan, or causing the other side a loss of 10,000 yuan, is enough. One man used a purchased 12306 account to grab more than 200 tickets on other people's behalf and earned more than 20,000 yuan, and was sentenced to 1 year and 6 months suspended for 1 year and 6 months, with a fine of 5000 yuan. Another sold ticket-grabbing software for a ticketing app on an online shop, was found to have engaged in unfair competition, and was ordered to pay 20,000 yuan. +- Benefit: Intruding into a computer information system, or using other technical means to obtain the data inside it, where the circumstances are serious, carries 3 years or less, with a fine imposed in addition or instead, and especially serious circumstances carry 3 to 7 years. The threshold for 「serious circumstances」 is illegal gains of more than 5000 yuan, or a loss of more than 10,000 yuan. The official cases: one man used a cheat to import purchased 12306 accounts in bulk and submitted ticket requests densely with many threads. He grabbed more than 200 train tickets on other people's behalf and made over 20,000 yuan. The court convicted him of the crime of illegally obtaining computer information system data and sentenced him to 1 year and 6 months suspended for 1 year and 6 months, with a fine of 5000 yuan. Zheng sold ticket-grabbing software for a ticketing app on an online shop, was found to have engaged in unfair competition, and was ordered to pay 20,000 yuan (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ; Supreme People's Court and Supreme People's Procuratorate (2011). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases Endangering the Security of Computer Information Systems, 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ; Supreme People's Procuratorate (2026). 上海静安：查办利用外挂软件批量代抢火车票牟利案. ; Supreme People's Court (2025). 2024 年人民法院知识产权典型案例（案例七，抢票软件不正当竞争案）. +- Notes: The script in the ticket-grabbing case does not change memory and does not crack anything; it is only 「faster than a person」, and it is still convicted under article 285 of the Criminal Law. Both taking money to grab tickets for others and selling the script count as making a profit. For the case where you use it once yourself, no official case was found. + +### 4. Let a crawler crawl only public pages that need no login, do not get round the anti-crawler measures, do not touch personal information, and do not sell what you crawl + +- Cost: No money. A crawler is a program that turns web pages automatically and pulls the data off them. Before you crawl, read the robots file and the interface authorisation note; robots is the file a site puts in its own root directory saying which pages may not be crawled. Do not crawl data that needs a login, needs a token, which is a temporary access credential, or that can only be got by cracking a signature. The hard part is that technically this data is there for the taking once you start, and holding out depends on nothing but your own restraint. +- In plain language: Data that needs a login, a token or a cracked signature counts as getting round the protection, and carries 3 years or less, with 3 to 7 years where it is serious. Crawling personal information is more troublesome. 50 records of movement, communications, credit reporting or property count as serious circumstances. 500 records of accommodation, health or transactions, 5000 records of anything else, or 5000 yuan earned, are the same. +- Benefit: The sentencing for getting round the protection in order to take data out of a system is the same as in item 3 in this section (ticket-grabbing, flash-sale and order-brushing scripts): 3 years or less, or 3 to 7 years. Illegally obtaining, selling or providing citizens' personal information amounts to 「serious circumstances」 on any one of the following: more than 50 records of movement trajectories, communications content, credit reporting or property information; more than 500 records of accommodation, call records, health or transaction information; more than 5000 records of any other personal information; and illegal gains of more than 5000 yuan. Serious circumstances carry 3 years or less, with a fine imposed in addition or instead, and especially serious circumstances carry 3 to 7 years. The official cases: the software developer Li disguised a function that stole photographs from people's phone galleries as an 「attractiveness detection」 app, got 1751 face photographs, and also sold social engineering database material; he was sentenced to 3 years suspended for 3 years. A social engineering database is a database that collects personal information leaked from many places. One internet technology company used a crawler to take shop and order data from a food delivery platform, causing a loss of more than 40,000 yuan, and was prosecuted for the crime of illegally obtaining computer information system data (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百五十三条之一）. ; Supreme People's Court and Supreme People's Procuratorate (2017). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Infringing Citizens' Personal Information, 关于办理侵犯公民个人信息刑事案件适用法律若干问题的解释（第三、四、五条）. ; Supreme People's Procuratorate (2022). Typical Cases of Procuratorates Punishing the Crime of Infringing Citizens' Personal Information According to Law (Case Two), 检察机关依法惩治侵犯公民个人信息犯罪典型案例（案例二）. ; Supreme People's Procuratorate (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例一）. +- Notes: The 2017 judicial interpretation provides that 「handing information that was lawfully collected on to somebody without that person's consent」 also counts as 「providing」. Exporting a table of users to an outside partner at work falls under that same sentence. 5000 records is only the size of a small spreadsheet. + +### 5. Do not take development contracts for gambling, fraud or pornography apps, and do not set up payment channels or unblock accounts for a 「client」 + +- Cost: No money. The price is turning down that kind of outsourcing work, where the quote is high and the requirement is vague. The hard part is that what the other side pays is often well above the going rate. +- In plain language: Taking that kind of work, where the quote is high and the requirement is vague, makes you an accomplice to a gambling gang. It is punished as the crime of opening a gambling house, which carries 5 years or less and 5 to 10 years where it is serious, and that is heavier than the 3 years at the most you get for merely helping. In one esports prediction app, 28 people who wrote the code were sentenced to 1 year to 3 years and 6 months. Unblocking QQ accounts for a fraud gang brought 1 year and 6 months and a fine of 20,000 yuan. +- Benefit: Knowing that somebody is using the internet to commit a crime and still providing them with technical support, payment settlement or help of that kind, where the circumstances are serious, carries 3 years or less, with a fine imposed in addition or instead. Where it also constitutes another crime, it is punished as the heavier one. Opening a gambling house carries 5 years or less and a fine, and serious circumstances carry 5 to 10 years. The official cases: after the Zhangsheng group developed gambling games it 「sold the network on as a package」 and took site-building fees and reporting fees, with nearly 7 billion yuan involved. Zhang Sheng was sentenced to 9 years and 9 months, with a fine of 50 million yuan. The 79 people responsible for the outsourced development, including Wang Qiang, got between 6 months and 9 years and 9 months. Chen Dong and others developed and ran an esports prediction app, with the technical team writing code in Hainan, and 28 people were convicted of the crime of opening a gambling house with sentences of 3 years and 6 months down to 1 year, together with fines. Zhang opened a studio that unblocked QQ accounts frozen by a fraud gang, and was sentenced to 1 year and 6 months with a fine of 20,000 yuan (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十七条之二、第三百零三条第二款）. ; Supreme People's Procuratorate (2023). 建赌博网站并出售，涉案近 70 亿元. ; Supreme People's Procuratorate (2022). 上海闵行区：办理一起新型开设网络赌场案. ; Supreme People's Procuratorate (2025). 依法惩治帮助信息网络犯罪活动及相关犯罪典型案例（案例一）. +- Notes: 「Knowing」 does not require the other side to have spelled out the purpose. A quote plainly above the market rate, features that are top-up and withdrawal plus odds, and a client who will not say what the thing is for are all used, when a case is handled, to work out that you knew. Technical staff who develop gambling apps are handled as accomplices to the crime of opening a gambling house, and not under the crime of assisting information network criminal activity, which is capped at 3 years. The crime of assisting information network criminal activity is short for 「the crime of assisting information network criminal activity」. + +### 6. Hand over your account access when you leave: do not delete the database, do not leave a back door and do not change the password to lock the system, even if the company owes you money + +- Cost: No money. If the company owes you wages, get them back through labour arbitration rather than doing it yourself (for labour arbitration see section 7). The hard part is holding back, while you are angry, from touching the system. +- In plain language: The account and password were given to you by the company, which does not mean you can still use them after you leave; logging in counts as illegal control and carries 3 years or less. Deleting data so that the system will not run carries 5 years or less, and especially serious consequences carry 5 years or more. One senior engineer who was dismissed used the original administrator account to log into the company server and delete data and logs, causing a loss of more than 120,000 yuan, and was sentenced to 3 years suspended for 5 years with a fine of 30,000 yuan. +- Benefit: Illegally controlling a computer information system, where the circumstances are serious, carries 3 years or less, with a fine imposed in addition or instead, and especially serious circumstances carry 3 to 7 years. Deleting or modifying the functions or data of a system so that it cannot run normally, where the consequences are serious, carries 5 years or less, and especially serious consequences carry 5 years or more. The official cases: after the senior IT engineer Ly was dismissed he logged into the company shared server with the original administrator account. He changed the password, deleted data and logs, and caused a loss of more than 120,000 yuan. He was sentenced to 3 years suspended for 5 years, with a fine of 30,000 yuan. Bai, an employee of a software company, deliberately put faulty code into an upgrade program, so that a customer's system could no longer be logged into, causing a loss of more than 40,000 yuan, and was prosecuted for the crime of destroying a computer information system (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百八十六条）. ; Supreme People's Procuratorate (2024). 检察机关「依法惩治网络犯罪 助力网络空间综合治理」典型案例（案例六）. ; Supreme People's Procuratorate (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例七）. +- Notes: The account and password were given to you by the company, which does not mean you can still use them after you leave. Logging in with the original account is already 「illegal control」. The loss is worked out as the cost of restoring the data, and a few tens of thousands of yuan is enough to open a case. + +### 7. Do not take source code, customer lists or technical documents away when you leave, do not upload them to your own cloud drive, and do not reuse them at your next employer + +- Cost: No money. Before you leave, delete the company material from your own computer and cloud drive. The hard part is that some of those files you wrote line by line yourself and you are reluctant to let them go. +- In plain language: Code you wrote yourself is owned by the company as far as copyright goes, and a customer list counts as a trade secret, so taking either is taking the company's property. One employee who left took the source code of a security screening system to a new company, and he was sentenced to 5 years and fined 500,000 yuan, while the new company was fined 4,000,000 yuan. Another exploited a flaw in a system to download more than 33,000 design files and was sentenced to 1 year and 10 months. The company can also bring a separate civil claim against you. +- Benefit: Breaching a confidentiality obligation by disclosing trade secrets or using them yourself, where the circumstances are serious, carries 3 years or less, with a fine imposed in addition or instead, and especially serious circumstances carry 3 to 10 years together with a fine. The Anti-Unfair Competition Law also lists two things as infringement: taking trade secrets by means of 「electronic intrusion」, which is intruding into a system by technical means; and inciting or helping other people to breach a confidentiality obligation. The official cases: Sun and others, who had left the company, took the source code and core algorithms of a security screening system and used them in their new company's products; Sun was sentenced to 5 years and fined 500,000 yuan, and the new company was fined 4,000,000 yuan. Tian exploited a flaw in a system to download more than 33,000 design files and was sentenced to 1 year and 10 months, with a fine of 100,000 yuan. The engineer Yang saved the source code of an app module on a personal computer and in the cloud, and after changing jobs made a similar app, and was prosecuted for the crime of infringing trade secrets (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百一十九条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Anti-Unfair Competition Law, 反不正当竞争法（2025 年修订，第十条）. ; Supreme People's Procuratorate (2025). 检察机关知识产权保护典型案例（案例七）. ; Supreme People's Procuratorate (2020). 2019 年度检察机关保护知识产权典型案例（案例十六）. ; Supreme People's Procuratorate (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例五）. +- Notes: 「Code you wrote yourself」 is owned by the company as far as copyright goes; see item 13 in this section (code written with company resources belongs to the company). Taking it is taking the company's property. A customer list is business information and is likewise a trade secret. On top of any prison sentence, the company can bring a separate civil claim. + +### 8. Do not run your own programs on somebody else's computer, server or camera, and do not mine cryptocurrency on company machines + +- Cost: No money, and no time either. +- In plain language: Controlling somebody else's computer, server or camera without permission is a crime at 20 machines or more, carrying 3 years or less, and especially serious cases carry 3 to 7 years. Where it does not reach a crime, there is still up to 5 days of detention, and a fine of 50,000 to 500,000 yuan is possible. Somebody who has been punished criminally is barred for life from key posts in network security management and network operations. +- Benefit: Illegally controlling a computer information system, where the circumstances are serious, carries 3 years or less, with a fine imposed in addition or instead, and especially serious circumstances carry 3 to 7 years. Illegally controlling 20 or more machines counts as 「serious circumstances」. Where it does not reach a crime, the police confiscate the illegal gains, detain the person for up to 5 days, and may impose a fine of 50,000 to 500,000 yuan. Where the circumstances are relatively serious, detention is 5 to 15 days and a fine of 100,000 to 1,000,000 yuan may be imposed. For 5 years after an administrative penalty, and for life after a criminal penalty, a person may not take a key post in network security management or network operations. The official cases: Li bought the IP addresses of network cameras, used software to control 105 cameras to spy on people, and was sentenced to 3 years suspended for 5 years with a fine of 20,000 yuan. Wu and others planted a trojan on phone motherboards, controlled more than 330 余万 phones and sent verification codes back, and were sentenced to 2 years to 4 years and 6 months (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ; Supreme People's Court and Supreme People's Procuratorate (2011). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases Endangering the Security of Computer Information Systems, 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十九、六十六条；2016 年文本为第二十七、六十三条，终身禁业为修正后的第六十六条第三款）. ; Supreme People's Procuratorate (2022). 山东滨州开发区：办理一起非法控制计算机信息系统案. ; Supreme People's Procuratorate (2021). 充分发挥检察职能 推进网络空间治理典型案例（案例六）. +- Notes: No case of mining cryptocurrency on a company server published by the Supreme People's Procuratorate or the Supreme People's Court was found, so this item is written from the provision on 「controlling somebody else's system without permission」. 「Being barred for life from key posts」 is an extra consequence that the Cybersecurity Law attaches on top of a fine or a sentence, and for a technical worker it is heavier than a fine. + +### 9. Do not test somebody else's system without written authorisation, and 「out of good intentions」 and 「reporting it afterwards」 are not reasons for acquittal + +- Cost: No money. Before joining a vendor's SRC or a public bug bounty, get a document that sets out the scope of the authorisation, the list of targets and the period you may test for. An SRC is the security emergency response centre the vendor runs itself, which takes reports of vulnerabilities from outside. Once you have confirmed that the vulnerability exists, stop. The hard part is giving up having 「a quick try while you are at it」. +- In plain language: Whether you are convicted depends on whether you had authorisation and how much you got, not on your intention. Getting 500 sets of ordinary account details, or earning 5000 yuan, or causing the other side a loss of 10,000 yuan, is enough for 3 years or less. To prove a vulnerability people often export a table of users, and one export easily passes 500 sets at a time. Having account details in hand and using them outside the scope of the authorisation is still intrusion. +- Benefit: The systems of ordinary companies and individuals are computer information systems outside the fields of state affairs, national defence construction and frontier science and technology. Intruding into such a system, or using other technical means to obtain the data stored, processed or transmitted in it, where the circumstances are serious, carries fixed-term imprisonment of not more than 3 years or criminal detention, with a fine imposed in addition or instead. Especially serious circumstances carry 3 to 7 years and a fine. There are several standards for 「serious circumstances」: more than 10 sets of identity authentication information for online financial services such as payment settlement, securities trading and futures trading, meaning account details and the like; more than 500 sets of other identity authentication information; illegal control of more than 20 computer information systems; and illegal gains of more than 5000 yuan, or an economic loss of more than 10,000 yuan. 5 times the figures above is 「especially serious」. The official case: Procuratorate Case No. 36 settled a sentence,「超出授权范围使用账号、密码登录计算机信息系统，属于侵入计算机信息系统的行为」. That means that where you had the account details to begin with but used them outside the scope of the authorisation, the act itself counts as intrusion. In that case three people used accounts, passwords and tokens they had access to at work, logged into the company's internal system from another city, downloaded data that was not within the scope of their work, and sold it online, making illegal gains of 37000 yuan. The three were sentenced to 4 years, 3 years and 9 months, and 4 years, and each was fined 40,000 yuan (nationwide) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第一款、第二款）. ; Supreme People's Court and Supreme People's Procuratorate (2011). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases Endangering the Security of Computer Information Systems, 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. (reprinted by the Shenzhen Municipal Public Security Bureau) ; Supreme People's Procuratorate (2017). Ninth Batch of Guiding Cases (Procuratorate Case No. 36, the Wei Menglong, Gong Xu and Xue Dongdong case of illegally obtaining computer information system data), 第九批指导性案例（检例第 36 号，卫梦龙、龚旭、薛东东非法获取计算机信息系统数据案）. +- Notes: Procuratorate Case No. 36 is itself a case about selling data for profit. It is cited only for the sentence 「logging in outside the scope of your authorisation is still intrusion」, and not to say that good-faith testing is punished just as heavily. What you meant to do, and whether you reported it afterwards, cannot cancel the charge; they can only affect how heavy the punishment is and whether there is a prosecution. A vendor thanking you first and reporting you afterwards are not contradictory, and being thanked is not authorisation. While this section was being written, no good-faith testing case that could be checked word for word was found on the websites of the Supreme People's Court or the Supreme People's Procuratorate, so only the articles and the penalty standards are written out. If you are going to do this, go through a vendor's public SRC, or a written engagement for the assessment, and put the scope, the targets and the time window into the document. Prove the vulnerability with the least evidence, and do not take data in bulk. The band below a crime still brings an administrative penalty and a bar on the work, and the amount of the fine and the period are the same as in item 8 in this section (do not run your own programs on somebody else's machines). For what to do after finding a vulnerability, see item 10 in this section (report it as the rules require). + +### 10. Report a vulnerability as the rules require: before the fix, do not publish the details, do not release an exploit and do not hand it to anybody overseas + +- Cost: No money. Report the vulnerability to the product provider, or to an official platform. Until the vendor has fixed it, do not publish the details and do not release verification code. +- In plain language: Individuals are covered by the Provisions on the Administration of Security Vulnerabilities in Network Products as well. Until the vendor has come up with a fix, you may not publish the vulnerability, and you may not publish the details of a vulnerability in a system that is in use. You may not release programs and tools made specifically to exploit the vulnerability, and when you publish you must say at the same time how to fix it. A vulnerability that has not been made public may not be handed to any overseas organisation or individual other than the product provider. +- Benefit: This set of provisions covers three kinds of party: network product providers and network operators inside China, and organisations or individuals engaged in discovering, collecting and publishing vulnerabilities. No organisation or individual may use a vulnerability to carry out activities endangering network security, and may not unlawfully collect, sell or publish vulnerability information. Publishing vulnerability information to the public has to keep to five rules. First, you may not publish before the product provider has provided a fix. Second, you may not publish details showing that a network, information system or device in use by a network operator has a vulnerability. Third, you may not deliberately exaggerate the danger and the risk, and may not use vulnerability information for malicious hype or fraud. Fourth, you may not publish or provide programs and tools made specifically for exploiting a vulnerability to carry out activities endangering network security. Fifth, when you publish, you must publish the fix or the precautions at the same time. In addition, vulnerability information that has not been made public may not be given to any overseas organisation or individual other than the product provider. The provisions encourage reporting vulnerabilities to four official platforms. First, the network security threat and vulnerability information sharing platform of the Ministry of Industry and Information Technology. Second, the vulnerability platform of the National Centre for Reporting Network and Information Security Information. Third, the vulnerability platform of the National Computer Network Emergency Technical Response Coordination Center. Fourth, the vulnerability database of the China Information Security Evaluation Center. Where vulnerability information is collected or published in breach of the provisions, the Ministry of Industry and Information Technology and the Ministry of Public Security deal with it in accordance with their respective duties under the law. Where a situation provided for in the Cybersecurity Law is constituted, it is punished under that law. Publishing network security information such as system vulnerabilities to the public must comply with the relevant state provisions. A breach leads to an order to rectify and a warning from the competent department, and a fine of 10,000 yuan up to 100,000 yuan may be imposed. Where the breach is not rectified or the circumstances are serious, the fine is 100,000 yuan up to 1,000,000 yuan. It may also order the relevant business to be suspended, order a cessation of operations, or order the website or the application to be closed. It may also revoke the relevant business licence or the business licence. The person in charge who is directly responsible and the other persons directly responsible are fined 10,000 yuan up to 100,000 yuan (nationwide) +- Evidence grade: A +- Sources: Ministry of Industry and Information Technology, Cyberspace Administration of China and Ministry of Public Security (工业和信息化部、国家互联网信息办公室、公安部) (2021). Provisions on the Administration of Security Vulnerabilities in Network Products, 网络产品安全漏洞管理规定（工信部联网安〔2021〕66 号，第二、四、九、十、十四条，2021 年 9 月 1 日施行）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十八、六十五条；2016 年文本为第二十六、六十二条，规定第十四条援引的是 2016 年文本的条号）. +- Notes: The provisions put individuals squarely among the parties they cover, and 「I am only an amateur」 is not one of the exceptions. Submitting a vulnerability in a domestic system to an overseas bug bounty platform falls squarely under 「an unpublished vulnerability may not be given to anybody overseas」. Whether you are entitled to test at all is one question, and see item 9 in this section (do not test somebody else's system without written authorisation); what to do with a vulnerability once you have it is a separate question. A vulnerability found lawfully can still be punished for the way it is published, and a vulnerability found without authorisation is a problem both in the testing and in the publishing. Of the reporting channels, the Ministry of Industry and Information Technology platform and CNCERT are the most used. The provisions do not make a separate category for 「the vendor has been notified and still does not fix it」; in practice you keep a record of the negotiation and the reporting, and report to an official platform as well. + +### 11. Do not sell circumvention tools or VPN accounts, and do not set up this kind of node for other people + +- Cost: No money, and no time either. +- In plain language: Setting up your own line out of the country, or using an international channel other than the public channels of the state, can lead to an order to stop connecting to the internet and a warning, a fine of up to 15,000 yuan and confiscation of the illegal gains. Making a business out of it can be handled by the case-handling authorities as the crime of illegal business operation, and serious circumstances carry 5 years or less. This item is about selling and setting up nodes. For your own use, the ceiling on the penalty is a warning plus a fine. +- Benefit: Connecting directly to an international network must use an international entry and exit channel provided by the public telecommunications network of the state. No unit or individual may set up or use any other channel: you may neither build your own line out of the country nor use one somebody else has built. A breach leads to an order from the police to stop connecting to the internet and a warning, and a fine of up to 15,000 yuan may be imposed in addition, with the illegal gains confiscated. Making a business out of it can be handled by the case-handling authorities as the crime of illegal business operation. Serious circumstances carry 5 years or less, with a fine of 1 to 5 times the illegal gains imposed in addition or instead. Especially serious circumstances carry 5 years or more (nationwide) +- Evidence grade: B +- Sources: State Council (国务院) (1997). Interim Provisions on the Administration of International Networking of Computer Information Networks, 计算机信息网络国际联网管理暂行规定（第六、十四条）. (reprinted by the Chongqing Municipal Rural and Agricultural Committee) ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百二十五条第四项）. +- Notes: While this section was being written, no published case involving circumvention tools was found on the websites of the Supreme People's Court or the Supreme People's Procuratorate, and the way places handle it varies a great deal, so the grade is B. For your own use, the heaviest penalty is the warning plus a fine written in Article fourteen of the interim provisions. + +### 12. If you have signed a non-compete, send a written demand when the company does not pay the monthly compensation after you leave, and you can terminate it once 3 months have gone by; and for a post that never touched trade secrets you can ask the court to rule the clause has no effect + +- Cost: No money. When you leave, keep the original of the non-compete and the evidence that the company has not paid you month by month. +- In plain language: A non-compete means you may not go and work for a competitor for a period after you leave, at most 2 years, and the company has to pay compensation month by month. Where no amount is agreed, it is 30% of the average wage over the 12 months before you left, and not less than the local minimum wage. If the company goes 3 months without paying for its own reasons, you can ask for it to be terminated. From 2025, for a post where you never knew and never touched a trade secret, you can ask the court to rule that the clause has no effect. +- Benefit: A non-compete can bind only senior management, senior technical staff, and other people who are under a confidentiality obligation. The period is at most 2 years. The company must pay economic compensation month by month over that period. Where no compensation amount is agreed, it is 30% of the average wage over the 12 months before you left, and not less than the local minimum wage. Where the company goes 3 months without paying for its own reasons, the worker may ask for the non-compete to be terminated. Where the company terminates it of its own accord, it must pay an extra 3 months' compensation. From 2025, a worker in a post where they did not know and did not touch a trade secret may ask the court to rule that the clause has no effect. Where the agreed scope goes beyond the part of the secret they actually knew, the part beyond it is void. The official case: a pharmaceutical company sued its chief technology officer Zheng for 7,100,000 yuan in liquidated damages. The court found that the products of his new employer were not substitutes for those of the original company, that the scope of the non-compete should be limited to the secrets he actually knew, and dismissed all the claims (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2012). Labour Contract Law, 劳动合同法（2012 年修正，第二十三、二十四条）. (reprinted by the Fushun Municipal Emergency Management Bureau) ; Supreme People's Court (2020). Interpretation (One) on Several Questions Concerning the Application of the Law in Trying Labour Dispute Cases, 关于审理劳动争议案件适用法律问题的解释（一）（第三十六至四十条）. ; Supreme People's Court (2025). Interpretation (Two) on Several Questions Concerning the Application of the Law in Trying Labour Dispute Cases, 关于审理劳动争议案件适用法律问题的解释（二）（法释〔2025〕12 号，第十三至十五条）. ; Supreme People's Court (2025). 劳动争议典型案例（案例四）. +- Notes: The other way round, where the agreement is valid, the company has paid month by month, and you still go to work for a competitor, you have to pay the liquidated damages as agreed. Having paid, you may also be required to serve out the rest of the non-compete period. A non-compete agreed during employment counts as well, even where the company pays nothing extra for it. + +### 13. Code written in working hours or with company resources belongs to the company; make your own open-source project with your own time and your own equipment, and do not mix in company code + +- Cost: No money. Read the intellectual property clause in the employment contract when you join. The hard part is that personal work and the company's work often share one computer and have to be kept apart. +- In plain language: Software written mainly with the company's equipment, intranet or business data, and for which the company carries the responsibility, is owned by the company as far as copyright goes; you keep only the right of authorship, and where the contract says so it is the same. Code you write after work is the company's too, if you used a company computer or business data. An invention made with company resources belongs to the company as far as the patent goes. +- Benefit: A work created to complete a task of the unit is a work of employment, that is, something made while doing your job. In the general case the copyright belongs to the author, the unit has the right of priority use within the scope of its business, and for 2 years the author may not licence a third party to use it in the same way. But there are two kinds of situation that are different. One is works such as engineering designs and software created mainly using the material and technical conditions of the unit and bearing its responsibility. The other is works that the contract agrees belong to the unit. In these two kinds the author keeps only the right of authorship. The material and technical conditions of the unit means the company's equipment, intranet and business data. An invention created by carrying out a task of the unit, or created mainly by using the material and technical conditions of the unit, is a service invention, and both the right to apply for a patent and the patent obtained belong to the unit (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Copyright Law, 著作权法（2020 年修正，第十八条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Patent Law, 专利法（2020 年修正，第六条）. +- Notes: Software written straight into the category of 「copyright belongs to the unit」 almost never falls, for a programmer, into the category of 「copyright belongs to the author, and the unit has priority use for two years」. 「Written after work」 still counts as being made mainly with the equipment and materials of the unit as soon as a company computer, intranet or company business data is used. To keep a personal project clean, use your own machine, do not touch the company repository, and do not overlap with the company's business. + +### 14. If you build a product on open-source code such as GPL, open your source as the licence requires or change the licence; do not assume 「nobody is in charge」 + +- Cost: No money. Before you use somebody else's open-source code, look at its licence, which is the set of conditions the author has written down. The GPL family requires a product that uses it to publish its source code too, which conflicts with selling closed source, and where it conflicts you change to something else. The hard part is giving up a library you have got used to. +- In plain language: Infringing copyright is made good by paying the other side's loss or the money you earned, and where the infringement is deliberate and the circumstances are serious it can reach 1 to 5 times that. One company hired a rival's former employees, produced substantially similar software, and argued that the other side's software was under GPLv2 and should have been open source. The Supreme People's Court held that whether the other side had open-sourced it made no difference to its copyright, and ordered it to stop the infringement and pay 500,000 yuan. +- Benefit: Infringing copyright is made good by paying the rights holder's actual loss, or by paying the money the infringer earned. Where neither can be worked out, it is set by reference to what it would cost to buy that licence properly. Deliberate infringement in serious circumstances can carry 1 to 5 times the compensation. Where the public interest is harmed, the illegal gains may also be confiscated and a fine imposed. The official case: (2021) Supreme People's Court Intellectual Property Chamber Civil Final No. 51. The defendant hired a former employee of the plaintiff and developed gateway software that was substantially similar to the plaintiff's. The defendant argued that the plaintiff's software was based on GPLv2 and should have been open source. The Supreme People's Court held that「开发者自身是否违反 GPLv2 协议和是否享有软件著作权是两个独立问题」, and ordered it to stop the infringement and pay 500,000 yuan (nationwide) +- Evidence grade: B +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Copyright Law, 著作权法（2020 年修正，第五十三、五十四条）. ; Intellectual Property Chamber of the Supreme People's Court (最高人民法院知识产权法庭) (2022). Copyright Infringement Case Involving 「Open-Source Software」（（2021）最高法知民终 51 号）. +- Notes: This case shows that「the other side used GPL code」cannot be used as your shield for copying it. In the other direction, a GPL author suing a commercial user who did not open-source it already has a judgment in force in China; the original text was not found on the website of the Supreme People's Court, so it is not cited here. The grade is B because whether you follow an open-source licence can only be handled through contracts and the general provisions of the copyright law, and there is no article written for it. + +### 15. Collect only the personal information your app genuinely needs, get a guardian's consent for anyone under 14, and write the privacy policy clearly + +- Cost: No money. Before you ship, go through the list of permissions the app asks for alongside the privacy policy, and leave out any permission you do not need. The hard part is that the product side always wants to keep a little more data 「that might come in useful later」. +- In plain language: Collect only the information the business genuinely needs, and get a guardian's consent for anyone under 14. A breach first brings an order to rectify, and the app may be taken down. Refusing to rectify brings a fine of up to 1,000,000 yuan, and the person in charge 10,000 to 100,000 yuan. Serious cases bring a fine of up to 50,000,000 yuan or 5% of the previous year's turnover, and 100,000 to 1,000,000 yuan for the person in charge, and the people who write the code and build the product may count as the person in charge. +- Benefit: Processing personal information must have a clear and reasonable purpose, must be directly related to that purpose, and must use the method that affects the individual's rights and interests least. Processing personal information requires the individual's consent, or must fall within a statutory situation. For anyone under 14 years of age, the consent of a parent or guardian must be obtained. A breach leads the regulator to order rectification, give a warning and confiscate the illegal gains, and to order the offending app to suspend or stop providing its service. Refusing to rectify brings a fine of up to 1,000,000 yuan, and a fine of 10,000 to 100,000 yuan on the person in charge who is directly responsible and the other persons directly responsible. Serious circumstances bring a fine of up to 50,000,000 yuan, or up to 5% of the previous year's turnover, and 100,000 to 1,000,000 yuan for those responsible. Those people may also be barred for a set period from being directors, supervisors or senior executives of the relevant companies. Where selling or providing personal information amounts to a crime, see item 4 in this section (nationwide) +- Evidence grade: B +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Personal Information Protection Law, 个人信息保护法（第六、十三、三十一、六十六条）. +- Notes: In practice,「the person directly responsible」very often lands on the people who write the code and build the product. The Ministry of Industry and Information Technology regularly reports apps taken down for breaches; this section did not verify which particular batch of reports it was, so nothing is cited. + +### 16. Do the ICP filing before a website or app goes live, and keep logs for at least 6 months as the graded protection system requires + +- Cost: A little money. The filing is free, and going through a cloud vendor's process takes a few weeks. The graded protection assessment is charged according to the level. Graded protection, known as 等保, is a set of security management measures by which the state grades systems by how important they are and requires websites and systems to meet them; a small site generally does not need the assessment. +- In plain language: The filing costs nothing, and going through a cloud vendor's process takes a few weeks. A domestic server without a filing cannot resolve the domain, and refusing to rectify leads to the site being closed. Not showing the filing number on the home page carries a fine of 5,000 to 50,000 yuan. Saying the filing is free of charge while actually charging carries a fine of 100,000 to 1,000,000 yuan. Logs have to be kept for a full 6 months, and refusing to rectify or causing consequences carries a fine of 10,000 to 100,000 yuan. +- Benefit: Commercial internet information services are subject to a licence, and non-commercial ones to a filing. Without a licence, or without a filing, you may not carry on. Where there is no filing, the telecommunications administration authority orders rectification, and where it is not rectified orders the website closed. Operating without a licence means the illegal gains are confiscated and a fine of 3 to 5 times the illegal gains imposed, or a fine of 100,000 to 1,000,000 yuan. Not showing the filing number on the home page carries a fine of 5,000 to 50,000 yuan. A network operator must, under the graded protection system, formulate security management rules, guard against attacks, keep network logs for not less than 6 months, and classify, back up and encrypt data. Failure to comply leads to an order to rectify and a warning. Refusing to rectify, or causing consequences, carries a fine of 10,000 to 100,000 yuan, and the person in charge 5,000 to 50,000 yuan (nationwide) +- Evidence grade: A +- Sources: State Council (国务院) (2011). Measures for the Administration of Internet Information Services, 互联网信息服务管理办法（2011 年修订，第四、十九、二十二条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十三、六十一条；2016 年文本为第二十一、五十九条）. +- Notes: A domestic server without a filing cannot resolve the domain at all. There are two common traps: filing as an individual non-commercial site while actually charging; and changing the domain or the entity without going back to change the filing. Keeping logs for 6 months is both a duty and a way of protecting yourself, because when something goes wrong you can show it was not you. + +### 17. Before offering a generative AI service to the public, do the security assessment and the algorithm filing as the rules require, and label the generated content + +- Cost: No money, but you have to go through the process at the cyberspace administration department, which takes some time. A small tool of your own that is not open to the public does not trigger any of this. +- In plain language: A generative AI service aimed at the public that can affect public opinion or mobilise people needs a security assessment and an algorithm filing before it goes live, and if you change it or shut it down you have to do the change and the cancellation. Generated images and video have to carry a label. A small tool for your own use that is not open to the public does not have to do any of this. A breach is punished under the Cybersecurity Law and the other laws. +- Benefit: A provider of generative artificial intelligence services with the attributes of public opinion or the capacity for social mobilisation must carry out a security assessment in accordance with state provisions, and must go through the algorithm filing under the algorithm recommendation provisions, together with the later change and cancellation formalities. The attributes of public opinion or the capacity for social mobilisation mean being able to affect how the public sees something, or being able to organise and mobilise people. The provider must label generated content such as images and video. The measures came into force on 2023-8-15 (nationwide) +- Evidence grade: B +- Sources: Cyberspace Administration of China and six other departments (国家互联网信息办公室等七部门) (2023). Interim Measures for the Administration of Generative Artificial Intelligence Services, 生成式人工智能服务管理暂行办法（第十二、十七条）. +- Notes: These measures do not themselves say how much the fine is. A breach means going back to the Cybersecurity Law, the Data Security Law and the Personal Information Protection Law to find the penalty, which is why the grade is B. Civil cases in which content generated with AI infringed somebody's copyright or likeness were not checked against the original text on the website of the Supreme People's Court, so nothing is cited. diff --git "a/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" "b/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" new file mode 100644 index 00000000..67c5c8c0 --- /dev/null +++ "b/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" @@ -0,0 +1,218 @@ +[← 回总目录](../README.md) + +# 12. 创业与做生意：别把家底赔进去 + +这一节只算钱和法律责任，不折算成寿命。开一家店、办一家公司，哪一步会把你自己的钱、家里的钱、配偶的钱一起赔进去？法律怎么规定，能不能提前隔开？最容易把家底赔进去、又最省事就能避开的条目，放在最前面。所以签字类的条目靠前，要自己动手做的方法靠后。只凭经验写、没有文献支撑的，标 C 级。 + +[← Back to the contents](../README.md) + +# 12. Starting a business: do not lose the family savings + +This section counts money and legal liability, and is not converted into lifespan. When you open a shop or incorporate a company, which step puts your own money, your family's money and your spouse's money at risk together? What does the law say, and can you keep them apart in advance? The items that most easily lose the family savings and are also the easiest to avoid come first. So the items about signing papers come early, and the methods that need you to do the work yourself come later. Anything written from experience alone, with no literature behind it, is graded C. + +### 1. Start a business only with money you can afford to lose, and do not touch the family savings or borrow to open + +- Cost: No money. Before you start, set yourself a number: losing all of it would not affect your mortgage payments, your children or your medical care. Do not invest more than that. Money from family members and borrowed money do not count towards that number. The hard part is holding back from putting in more when the business looks like it is taking off. +- In plain language: A sole proprietor's debts are paid out of the money and property in your own name. If it is not clear whether the business was run by the whole family, the family's assets are used. Incorporating a company is not a safe harbour either: when the company cannot pay a debt, the capital you promised to contribute and have not yet paid in can be demanded from you early. So decide how much to put in on the basis of 「all of it lost and I still pay the mortgage and my medical bills」. +- Benefit: Losing is not only about the money. Where a sole proprietorship business owes debts and the business was run by you alone, payment comes from the money and property in your own name (personal liability, 以个人财产承担). If the business was run by the whole family, or it is not clear who ran it, payment comes from the family's money and property (family liability, 以家庭财产承担). A debt signed by both husband and wife is a joint debt that both of them repay. One spouse's later agreement counts too. Money one spouse borrowed in their own name is still repaid by both, as long as the creditor can show the money went into a business the two of them ran together. On the company side there is a parallel rule: when the company cannot pay a debt that has fallen due, a creditor may demand early payment of the capital the shareholder subscribed to within the registered capital but has not yet paid in by the due date (认缴出资) (nationwide) +- Evidence grade: B +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第五十六、一千零六十四条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2023). Company Law, 公司法（2023 年修订，第四十七、五十四条）. +- Notes: 「亏得起」 means this: if the money is lost entirely, you still do not have to sell your home or borrow from relatives to plug the hole. For how many businesses survive, and for how many years on average, no verifiable official statistics could be found on the National Bureau of Statistics (国家统计局) or State Administration for Market Regulation (国家市场监督管理总局) websites this time. So the text gives no figures and only the legal consequences. That is why the grade is B. + +### 2. Do not sign a personal guarantee for company loans, and never let your spouse sign one either + +- Cost: No money. When a bank, a supplier or a landlord asks you to sign a 「股东个人连带担保」 or asks for 「配偶签字」, stop for a moment. Signing that page means you personally give up the protection that 「the company's debts do not reach the individual」 (giving up limited liability). Do not sign if you can avoid it. If you must sign, sign only a 「一般保证」 and write down the largest amount you are guaranteeing. The hard part is that the other side often says 「no signature, no loan」. You have to hold out against that. +- In plain language: What you signed is a 「连带责任保证」: when the company cannot pay, the creditor can bypass the company and come straight to you. With a 「一般保证」, the creditor has to sue the company or go to arbitration first, and only what is still missing after the company's assets are sold off comes to you. If your spouse signs as well, both families' assets go in together, and for you personally the 「limited」 in a limited liability company stops counting. +- Benefit: The sentence on the guarantee document means exactly this: 「the debtor does not pay, and you pay instead」. Where it is written as a 「连带责任保证」, the creditor does not have to approach the company and can come to your door. Where it is written as a 「一般保证」, the creditor has to sue the company or apply for arbitration first, then ask the court to sell off the company's assets in enforcement against the debt. Only the shortfall after the sale is yours to make up. A debt signed by both husband and wife is a joint debt that both of them repay. One spouse's later agreement counts too. Once your spouse signs, both families' assets go in together (nationwide, in force from 2021) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第六百八十一、六百八十七、六百八十八、一千零六十四条）. +- Notes: When a bank lends to a small or micro business, it almost always asks a shareholder to sign a joint and several guarantee. You can still get the loan. What matters is understanding what that page gives up before you sign, and keeping the guaranteed amount inside the 「can afford to lose」 figure from item 1 in this section (start a business only with money you can afford to lose). For the general rules on promissory notes and guarantee documents, see item 18 in section 8. + +### 3. Choose the right entity before you open: a sole proprietorship and an ordinary partner pay to the end, only a limited liability company is 「limited」 + +- Cost: Registration is done on the local government services website and takes a few hours. After a limited liability company is set up you have to keep books and file tax returns every month, and that is a running cost: doing it yourself takes time, paying someone to keep the books costs money, and prices differ by place. +- In plain language: A sole proprietorship and an ordinary partner pay to the end: whatever is owed is paid. A shareholder of a limited liability company pays only the amount promised at registration. But whatever you promised has to be paid in real money within 5 years: put in 1,000,000 and you may owe 1,000,000. Mixing the company's books with your own also makes the 「limited」 part fail. +- Benefit: The debts of a sole proprietorship business are paid from the individual's money and property, or from the family's money and property. A partner in an ordinary partnership enterprise bears unlimited joint and several liability for the partnership's debts. That means whatever is owed is paid, and payment does not end until it is made. A creditor can also pick out one partner alone and demand the whole amount. A limited partnership is different: a limited partner inside it pays no more than the amount promised (以认缴出资为限). A shareholder of a limited liability company likewise pays no more than the amount promised (以认缴出资额为限). But the 「limited」 part fails in three situations. First, the money promised has to be paid in within 5 years of the company being established, and when the company cannot pay a debt that has fallen due, a creditor can demand early payment. Second, money already put in and then withdrawn (抽逃出资) has to be returned. Third, the company is used as a shell to hide from creditors. A company with only one shareholder that cannot show the company's money and the shareholder's own money are kept separate also counts as the third case. Under these two practices, the company's debts are yours to pay as well (joint and several liability, 承担连带责任) (nationwide, in force from 2024-7) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第五十六条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2006). Partnership Enterprise Law, 合伙企业法（2006 年修订，第二条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2023). Company Law, 公司法（2023 年修订，第四、二十三、四十七、五十三、五十四条）. +- Notes: The condition for the 「limited」 part is that company money and private money stay apart: do not mix the company account and your personal account, do not pay household expenses out of the company's money, and keep complete records. Do not put a large figure in the registered capital just because it looks better. Whatever you subscribe for has to be paid in real money within 5 years. Put in 1,000,000 and you have created a potential debt of 1,000,000. + +### 4. Do not act as a nominal shareholder, and do not hold shares on someone else's behalf + +- Cost: No money. If someone says 「lend me your ID to register as a shareholder」 or 「the shares go under your name for now」, refuse every time. +- In plain language: Once your name is on the register you are a shareholder in law. The money promised at registration has to be paid in on time, and if you cannot pay it you must make up the shortfall and also compensate the company for its loss. If another shareholder did not really pay in money when the company was set up, you share that shortfall as well. Someone else runs the business, and the debt sits under your name. +- Benefit: A shareholder on the register is a shareholder in law. You have to pay in the whole amount promised, to the last cent, by the deadline set in the company's articles of association (full payment of subscribed capital, 足额缴纳认缴出资). If you do not pay on time, you must make up the shortfall and also compensate the company for its loss. If a shareholder did not really pay in money when the company was set up, the other shareholders present at the time share that shortfall (joint and several liability, 承担连带责任). When the company cannot pay a debt that has fallen due, a creditor can demand early payment of money whose deadline has not yet arrived. The person actually running the business is not on the register, and the debt sits under your name (nationwide, in force from 2024-7) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2023). Company Law, 公司法（2023 年修订，第四十七、四十九、五十、五十四条）. ; Supreme People's Court (最高人民法院) (2020 revision). Provisions on Several Issues Concerning the Application of the Company Law of the People's Republic of China (III), 关于适用《中华人民共和国公司法》若干问题的规定（三）（法释〔2011〕3 号，2014 年法释〔2014〕2 号、2020 年法释〔2020〕18 号两次修正，第二十四、二十五条）:「如无法律规定的无效情形，人民法院应当认定该合同有效」「实际出资人未经公司其他股东半数以上同意，请求公司变更股东……人民法院不予支持」「名义股东处分股权造成实际出资人损失，实际出资人请求名义股东承担赔偿责任的，人民法院应予支持」. In plain terms: unless the law gives a ground of invalidity, the contract is valid; an actual investor who asks the company to change the shareholder registration without the consent of more than half the other shareholders is not supported by the court; where a nominal shareholder disposes of the equity and the actual investor suffers a loss, the court supports the actual investor's claim for compensation against the nominal shareholder. (National Laws and Regulations Database, 国家法律法规数据库) ; Supreme People's Court (最高人民法院) (2025). Interpretation on Several Issues Concerning the Application of the Company Law of the People's Republic of China (draft for comment), 关于适用《中华人民共和国公司法》若干问题的解释（征求意见稿）第九十条. +- Notes: Between a nominal shareholder and the person who really put in the money, a nominee agreement holds unless the law makes it invalid. But if the person who really put in the money wants the registration changed to their own name, more than half of the other shareholders have to agree. If the nominal shareholder sells or pledges the shares, the court handles it under the good-faith acquisition rules of the Civil Code. Where the buyer was unaware, paid a reasonable price and completed the registration, the person who put in the money cannot get the shares back and can only claim compensation from the nominal shareholder. A draft for comment on a replacement for this judicial interpretation came out in September 2025, and the whole document is repealed once the final version takes effect. Use the 2020 revision, which was still in force as of September 2026. For the risk of being a nominal legal representative, see item 28 in section 8. + +### 5. Before you take a franchise, check the Ministry of Commerce filing, get the written disclosure, and put the 「cooling-off period」 into the contract + +- Cost: No money. Go to the Ministry of Commerce's 「商业特许经营信息管理系统」 website and check whether the brand has filed. Then require the other side to give you, before signing, written material covering everything it has to disclose, together with the contract text. A few hours is enough. +- In plain language: Under the rules the brand itself must first run 2 company-owned stores and operate them for at least 1 year, and it must file with the commerce authorities within 15 days after signing. Before you sign, it has to give you written material on 12 items: the franchise fee, how the deposit is refunded, whether the existing stores make money, and so on. If you cannot find a filing, if they will not hand over the material, or if they want your money before letting you see the contract, walk away. +- Benefit: The rules require the brand (called the franchisor, 特许人, in the rules) to run at least 2 company-owned stores it operates itself, for more than 1 year. Counting from the day the first franchise contract is signed, it must file with the commerce authorities within 15 days. Failing to file, it is fined 10,000 to 50,000 yuan. Failing to file after the deadline, it is fined 50,000 to 100,000 yuan and the filing is published. Before you sign, it must hand you written material covering everything it has to disclose, 12 items in total. These include which kinds of franchise fee are charged and how much each is, the circumstances in which the deposit is refunded, how many franchise stores there are now and how the business is doing. They also include a summary of the audited accounts for the past 2 years, and the lawsuits and arbitrations of the past 5 years. The contract must also state that the franchisee can withdraw on its own say-so within a period after signing (unilateral termination, 单方解除). If the brand conceals information or gives false information, the franchisee can terminate the contract (nationwide, in force from 2007-5) +- Evidence grade: A +- Sources: State Council (国务院) (2007). Regulations on the Administration of Commercial Franchise, 商业特许经营管理条例（国务院令第 485 号，第七、八、十二、二十二、二十三、二十五条）. ; Ministry of Commerce (商务部) (2012). Measures for the Administration of Commercial Franchise Information Disclosure, 商业特许经营信息披露管理办法（商务部令 2012 年第 2 号，第五、九条）. ; Ministry of Commerce (商务部). 商业特许经营信息管理系统. +- Notes: If you cannot find a filing, if they will not hand over the disclosure documents, or if they want your money before showing you the contract, walk away. The 「cooling-off period」 is the stretch after signing when you can still withdraw on your own say-so. The number of days is set by the contract; the regulation only says 「a certain period」, so ask for the number of days to be written into the contract when you sign. The 「average sales volume, gross profit and net profit of existing franchise stores」 the brand discloses should be questioned about their source, because the measures require the brand to explain where the figures came from. + +### 6. Settle the name, premises, business scope and registered capital before you register, and with complete papers you get the licence on the spot + +- Cost: The registration step itself is not charged for. The registration fee for enterprises and sole proprietorships has not been collected since 2015-1-1. Do it yourself on the local government services website, or go to the government service hall and do it there; either way it takes a few hours. Cutting the company seal, paying someone to keep the books, and renting a registered address (address hosting) are services you buy in the market, not fees the government charges. +- In plain language: The registration itself is free, and the registration fee stopped being collected in 2015. There are only five sets of documents to submit: the application form, proof of identity or qualification, documents for the premises, the articles of association or the partnership agreement (a sole proprietorship submits none of these), plus anything else the rules name. With complete papers the licence is issued on the spot; if it cannot be issued on the spot it is done within 3 working days. Whatever you were charged is most likely an agent's errand fee. +- Benefit: The regulation fixes the documents to be submitted at five. 「（一）申请书」. 「（二）申请人资格文件、自然人身份证明」. 「（三）住所或者主要经营场所相关文件」. 「（四）公司、非公司企业法人、农民专业合作社（联合社）章程或者合伙企业合伙协议」. 「（五）法律、行政法规和国务院市场监督管理部门规定提交的其他材料」. In plain terms: one application form; one set of documents showing who you are or that you are qualified to do this; one set of documents proving the shop or the office is yours to use; the articles of association or the partnership agreement; plus anything else the rules name. A company must also submit the appointment documents and identity proofs of its legal representative, directors, supervisors and senior managers. A sole proprietorship submits no articles. The name is 「由申请人依法自主申报」, that is, the applicant names and declares it themselves. The registered capital 「实行认缴登记制」, so you write down what you promise to contribute and do not pay it over on the spot. The business scope is chosen from the categories the registration authority publishes. Where the documents are complete and correctly formatted, they are 「予以确认并当场登记」. Where it cannot be finished on the spot, it is done 「3 个工作日内」, and where the circumstances are complicated, 3 further working days. 「营业执照签发日期为市场主体的成立日期」: the date printed on the licence is the day the company or the shop came into existence (nationwide, in force from 2022-3) +- Evidence grade: A +- Sources: State Council (国务院) (2021). Regulations on the Registration of Market Entities, 市场主体登记管理条例（国务院令第 746 号，第十、十三、十四、十六、十九、二十一条）. ; State Administration for Market Regulation (国家市场监督管理总局) (2022). Detailed Rules for Implementing the Regulations on the Registration of Market Entities, 市场主体登记管理条例实施细则（市场监管总局令第 52 号，第十二、二十五、二十六条）. ; Ministry of Finance and National Development and Reform Commission (财政部、国家发展改革委) (2014). Notice on Cancelling, Suspending and Exempting a Batch of Administrative and Institutional Charges, 关于取消、停征和免征一批行政事业性收费的通知（财税〔2014〕101 号，附件 1）. +- Notes: Registration itself is not charged for, and whatever you were charged is most likely an agent's errand fee. The price of cutting the company seal is set by the market. The General Office of the State Council has expressly ordered 「严禁指定公章制作单位制作公章」, so when someone says 「you must go to the designated place to cut the seal」 you do not have to accept that. How much you put in the registered capital does not affect the shopfront. It only sets the ceiling on your liability, and it has to be paid in full within 5 years; see item 3 in this section (choose the right entity before you open). If the business scope contains items that need a licence, you cannot open until the licence comes through; see item 7 in this section (trades that need a licence) + +### 7. In a trade that needs a licence, do not open before the licence comes through + +- Cost: No money. Before you open, find out whether the business you are starting needs a licence. Food service and selling food need a food business licence. Selling only whole packages made by someone else needs no licence, only a filing. Open for business once the licence is in place. +- In plain language: Opening without a licence means, at the mild end, being ordered to stop, having the money earned taken away, and a fine of up to 10,000 yuan. If what you are doing is an item under monopoly or exclusive sale, or securities, futures, insurance or funds payment and settlement, serious circumstances amount to the crime of illegal business operations, with 5 years or less as the starting point. Renovating first and getting the licence afterwards, then trading before it comes through, is the step people trip over most often. +- Benefit: Opening the door and trading without a licence is handled by whichever department the laws and regulations designate. Operating without a licence where the law does not say what the penalty is: ordered to stop, the money earned taken away (illegal gains), and a fine of up to 10,000 yuan. There is a heavier tier as well. Operating without a licence items that the laws and regulations put under monopoly, exclusive sale or restricted trading. Doing securities, futures, insurance or funds payment and settlement business without approval. Where the circumstances are serious, both amount to the crime of illegal business operations. The sentence is 5 years or less of fixed-term imprisonment or criminal detention (short confinement), plus a fine of 1 to 5 times the money earned. Where the circumstances are particularly serious, 5 years or more (nationwide) +- Evidence grade: A +- Sources: State Council (国务院) (2017). Measures for Dealing with Operations Without a Licence or a Business Licence, 无证无照经营查处办法（国务院令第 684 号，第五、六、十三条）. ; State Administration for Market Regulation (国家市场监督管理总局) (2023). Measures for the Administration of Food Business Licences and Filings, 食品经营许可和备案管理办法（市场监管总局令第 78 号，第四条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百二十五条）. +- Notes: The licensing conditions for finance, medical care and education and training each have their own dedicated laws; this item verified only the general enforcement rules and the Criminal Law article. A bubble tea shop or a snack stall counts as food service and needs a food business licence. Renovating first and getting the licence afterwards, then trading before it comes through, is a common trap. For how the food business is divided into tiers and whether selling fresh meat and vegetables needs a certificate, see item 8 (work out which tier of food business you fall into). + +### 8. Work out which tier of food business you fall into: production and food service need a licence, selling only prepackaged food needs a filing, selling fresh meat and vegetables needs neither + +- Cost: The licence itself is not charged for. But the premises and the equipment have to meet the standard, and the on-site inspection plus the paperwork takes several weeks. If you sell meat you also have to keep the quarantine certificate and the certificate of conformity for meat quality inspection for every batch. +- In plain language: Making things yourself to sell, such as a snack stall, braised food, baking or home-made pickled vegetables, all need a licence. Selling only whole packages made by someone else needs no licence, only a filing with the market regulation department at county level or above. Selling fresh meat and vegetables needs no certificate. Doing it without one brings a fine of at least 50,000 to 100,000 yuan, which is more than most small businesses put in. +- Benefit: Food Safety Law: doing food business requires a licence first. Producing food yourself, selling food and running food service all have to obtain a licence as the law requires. 「但是，销售食用农产品和仅销售预包装食品的，不需要取得许可。仅销售预包装食品的，应当报所在地县级以上地方人民政府食品安全监督管理部门备案」. In plain terms: selling fresh agricultural products, and selling only prepackaged food, need no licence. Selling only prepackaged food requires a filing with the food safety supervision department of the local people's government at county level or above. Fresh agricultural products are things like fresh meat and vegetables that have not been processed. Prepackaged food is food somebody else made and sold with its packaging. 「食品生产加工小作坊和食品摊贩等的具体管理办法由省、自治区、直辖市制定」, so provinces, autonomous regions and municipalities set the rules for these two categories themselves. Doing food business without a licence: the money earned, the food, and the tools, equipment and raw materials are all confiscated. Where the total value of the goods at the selling price (货值金额) is under 10,000 yuan, there is a fine of 50,000 yuan or more and 100,000 yuan or less. Where the value is 10,000 yuan or more, the fine is 10 to 20 times the value. Regulations on the Administration of Pig Slaughtering: pigs can only be slaughtered at designated slaughterhouses set by the state, and the official bodies run the quarantine as one batch. 「除农村地区个人自宰自食的不实行定点屠宰外，任何单位和个人未经定点不得从事生猪屠宰活动」. 「从事生猪产品销售、肉食品生产加工的单位和个人以及餐饮服务经营者、集中用餐单位生产经营的生猪产品，必须是生猪定点屠宰厂（场）经检疫和肉品品质检验合格的生猪产品」. So the pork you sell and the pork you use must both come from a designated slaughterhouse and pass both the quarantine and the meat quality inspection. Slaughtering pigs privately outside a designated slaughterhouse: ordered closed down, and the pigs, the slaughtering tools and equipment and the money earned are all confiscated, with the same fine as the tier above for doing food business without a licence. Animal Epidemic Prevention Law: animal products that are sold and transported have to carry a quarantine certificate and a quarantine mark with the goods. Without them, ordered to correct, and then fined up to one times the value of comparable qualifying goods. Judicial interpretation: where food made or sold by a small workshop or a food stall does not meet food safety standards and the buyer asks for ten times the price in damages (punitive damages, meaning the extra payment on top of the refund), the court is to support it (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025, third revision). Food Safety Law, 食品安全法（第三十五、三十六、一百二十二条）. ; State Council (国务院) (2021). Regulations on the Administration of Pig Slaughtering, 生猪屠宰管理条例（国务院令第 742 号，第二、二十三、三十一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2021 revision). Animal Epidemic Prevention Law, 动物防疫法（第五十一、一百条）. ; Supreme People's Court (最高人民法院) (2024). Interpretation on Several Issues Concerning the Application of the Law in the Trial of Food and Drug Punitive Damages Cases, 关于审理食品药品惩罚性赔偿纠纷案件适用法律若干问题的解释（法释〔2024〕9 号，第四条）. +- Notes: The thresholds for small workshops, small food service and food stalls are set by each province, so this item gives no specific figures. Go to the market regulation bureau website of the province you are in, search for 「食品小作坊小餐饮小摊贩管理条例」, and read the local version. Selling your own food online does not remove a single licence because it happens 「online」: whatever needs a licence still needs one. Beyond the licences, the general rules of this section apply. For the entity and the licence see item 6 (settle the name, premises and business scope before you register), and for whether other trades need a certificate see item 7 (trades that need a licence). How much the buyer can claim see item 31 in section 5 (ten times damages for food). + +### 9. Bagging it and selling makes it prepackaged food: the production date, the shelf life and the ingredients list on the label may not be missing + +- Cost: Printing a label costs from a few cents to a few tenths of a yuan a sheet. Every batch has to have its date and its ingredients filled in correctly, and that takes some of your time. +- In plain language: Once food is put into packaging and sold, the law calls it prepackaged food. The packaging must carry a label, and there are nine items that must be marked on it, including the production date and the shelf life. Food sold loose must also carry the name, the date and the name and address of the seller. If a required item is missing, where the value is under 10,000 the fine is 5,000 to 50,000 yuan; where it reaches 10,000, the fine is 5 to 10 times the value. +- Benefit: Food Safety Law: prepackaged food (food put into packaging and sold) must carry a label on the packaging. The label has to state the name, the specification, the net content, the production date, the ingredients or the ingredients list, the producer's name, address and contact details, and the shelf life. It also has to state the product standard code, the storage conditions, the common names used in the national standards for the food additives used, and the food production licence number. Anything the laws, the regulations or the food safety standards separately require must be marked as well. Food sold loose is the kind that is not bagged and is weighed as needed. When you sell food loose, the container and the outer packaging must also carry the name of the food, the production date or the production batch number, the shelf life, and the name, address and contact details of the producer and seller. Prepackaged food you make or sell without a label, or with a label or set of instructions that do not meet this law: the money earned and the food are confiscated together. Where the total value of the goods at the selling price (货值金额) is under 10,000 yuan, there is a fine of 5,000 yuan or more and 50,000 yuan or less. Where the value is 10,000 yuan or more, the fine is 5 to 10 times the value. Where the circumstances are serious, production and business are ordered stopped until the licence is withdrawn. Where the label or the instructions only have minor defects (瑕疵) that neither affect food safety nor mislead consumers, they are first ordered corrected, and only a refusal to correct brings a fine of up to 2,000 yuan. The new General Standard for the Labelling of Prepackaged Foods is numbered GB 7718-2025. It was issued on 2025-3-16 and takes effect on 2027-3-16, with a two-year transition period in between. It requires a separate area on the packaging where the production date and the date the shelf life expires are shown together. The dates must use high-contrast, easily legible type and colours, and be arranged in year-month-day order (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025, third revision). Food Safety Law, 食品安全法（第六十七、六十八、一百二十五条）. ; National Health Commission (国家卫生健康委员会) and State Administration for Market Regulation (国家市场监督管理总局) (2025). Announcement on the Publication of 50 Food Safety National Standards Including the General Standard for the Labelling of Prepackaged Foods (GB 7718-2025), and of 9 Amendment Sheets, 关于发布《食品安全国家标准 预包装食品标签通则》（GB 7718-2025）等 50 项食品安全国家标准和 9 项修改单的公告（2025 年第 2 号）. (the Lhasa Municipal Health Commission reprinting the full text of the announcement) ; Supreme People's Court (最高人民法院) (2024). Interpretation on Several Issues Concerning the Application of the Law in the Trial of Food and Drug Punitive Damages Cases, 关于审理食品药品惩罚性赔偿纠纷案件适用法律若干问题的解释（法释〔2024〕9 号，第六、八条）. (the two tiers: minor label defects and missing required items) ; Beijing Municipal Administration for Market Regulation (北京市市场监督管理局) (2026). 找不到生产日期？食品标签要「变脸」！新规这些重点必看 (Cannot find the production date? Food labels have to change their face. These are the key points in the new rules). (the effective date, the transition period and the new requirements for showing dates) +- Notes: Label defects come in two kinds, and both the fine and the damages follow that split. One kind is a minor defect, such as a typo or an irregular type size, which neither affects food safety nor misleads anyone. For that the market regulation department only orders you to correct it, and a refusal to correct brings a fine of up to 2,000 yuan; the buyer cannot claim ten times damages either. The other kind is a required item that is missing, such as the production date or the shelf life. The judicial interpretation says in terms that this is not a minor defect: the fine is calculated on the value of the goods, 5,000 to 50,000 yuan where the value is under 10,000 yuan, and 5 to 10 times the value where it is 10,000 yuan or more. The buyer can also ask you separately for ten times the price of the food, counting anything under 1,000 yuan as 1,000 yuan (see item 31 in section 5, ten times damages for food). For the clause-by-clause requirements of GB 7718-2025, this item verified only the announcement and the dates and display method given in the official interpretation; the exact format follows the text of the standard itself. + +### 10. Ordinary food may not say it can cure disease: labels, instructions, advertisements and livestream scripts all count + +- Cost: No money. Go through your slogans, your product description pages and your livestream scripts. The hard part is holding back from saying things like 「lower your three highs」 or 「cure constipation」. +- In plain language: The labels, instructions and advertisements of ordinary food may not say they prevent or cure disease. Apart from the three categories of advertisements for medical services, medicines and medical devices, no other advertisement may say it cures disease, and none may use words that sound like medicine. The fine is 1 to 3 times the advertising spend. Where the advertising spend cannot be worked out, the fine is 100,000 to 200,000 yuan. In serious circumstances the business licence can be revoked as well. +- Benefit: Food Safety Law: 「食品和食品添加剂的标签、说明书，不得含有虚假内容，不得涉及疾病预防、治疗功能。生产经营者对其提供的标签、说明书的内容负责」. 「食品广告的内容应当真实合法，不得含有虚假内容，不得涉及疾病预防、治疗功能。食品生产经营者对食品广告内容的真实性、合法性负责」. In plain terms: the packaging, the instructions and the advertisement may not say they prevent or cure disease, and the seller answers for what is written there. Blowing food up in an advertisement to deceive consumers is penalised under the Advertising Law. The Advertising Law has a ban of its own. 「除医疗、药品、医疗器械广告外，禁止其他任何广告涉及疾病治疗功能，并不得使用医疗用语或者易使推销的商品与药品、医疗器械相混淆的用语」. Other than the three categories of advertisements for medical services, medicines and medical devices, no advertisement at all may say it cures disease, and none may use words that sound like medicine or like a medical device. Breaching this means the advertisement is ordered stopped. The party that placed the advertisement (the advertiser, 广告主) also has to issue a public clarification in the places it ran and remove the effects. The fine is then 1 or more times and up to 3 times the advertising spend. Where the advertising spend cannot be worked out or is plainly under-reported, the fine is 100,000 yuan or more and 200,000 yuan or less. In serious circumstances the fine is 3 to 5 times the advertising spend, or 200,000 yuan or more and 1,000,000 yuan or less, and the business licence can be revoked, putting the business out of business altogether (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025, third revision). Food Safety Law, 食品安全法（第七十一、七十三、一百二十五、一百四十条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2021 revision). Advertising Law, 广告法（第十七、五十八条）. +- Notes: 「Health functions」 cannot be claimed by ordinary food either. The health functions a health food may claim can only come from the catalogue the state sets, and the product has to go through registration or filing, a step ordinary food does not have. For how buyers recognise this kind of wording, see item 10 in section 6 (do not spend a lot of money on health products to look after your health). Beyond the packaging and the advertisement, the description pages, the livestream scripts and the social media posts count as advertising content too, and once they are sent out the record exists. + +### 11. The food business has a criminal line: selling diseased or dead meat or goods over the limit is already enough, and adding toxic or harmful substances carries 5 years whatever the amount + +- Cost: No money. Do what item 8 (work out which tier of food business you fall into) says: keep the quarantine certificate and the purchase vouchers, and put only ingredients in the list whose approval number you can look up. +- In plain language: The earlier items punish money. This one takes people. Selling diseased or dead meat, meat that fails inspection, or food where pesticide residues, veterinary drug residues or heavy metals seriously exceed the limit is already a crime as long as it is 「sufficient」 to cause severe poisoning, without anyone actually falling ill. That carries 3 years or less. Adding a toxic or harmful non-food ingredient to food starts at 5 years or less, and it makes no difference how much you sold. +- Benefit: Article one hundred and forty-three of the Criminal Law covers this tier. It covers food you make or sell that does not meet food safety standards. As long as it is 「足以造成严重食物中毒事故或者其他严重食源性疾病的」, that is enough for the crime. The sentence is 3 years or less of fixed-term imprisonment or criminal detention (short confinement), together with a fine. Where severe harm to people's health results, or there are other serious circumstances, the sentence is 3 years or more and 7 years or less, together with a fine. Where the consequences are particularly serious, the sentence is 7 years or more, up to life imprisonment, and a fine or confiscation of all the property held in the name. Article one hundred and forty-four: adding a toxic or harmful non-food ingredient to food you make or sell, or knowingly selling it after adding one, carries 5 years or less of fixed-term imprisonment together with a fine. Where severe harm to people's health results, or there are other serious circumstances, the sentence is 5 years or more and 10 years or less, together with a fine. Where someone dies, or there are other particularly serious circumstances, punishment follows Article one hundred and forty-one (10 years or more, life imprisonment or the death penalty). The Supreme People's Court and the Supreme People's Procuratorate have issued an interpretation, known as the SPC-SPP interpretation. It sets out which situations count as 「足以造成严重食物中毒事故或者其他严重食源性性疾病」. First, 「含有严重超出标准限量的致病性微生物、农药残留、兽药残留、生物毒素、重金属等污染物质」. Second, 「属于病死、死因不明或者检验检疫不合格的畜、禽、兽、水产动物肉类及其制品的」. Third, 「属于国家为防控疾病等特殊需要明令禁止生产、销售的」. Fourth, 「特殊医学用途配方食品、专供婴幼儿的主辅食品营养成分严重不符合食品安全标准的」. Using additives over the limit or in the wrong place, to the point where the consequences above follow, is convicted under Article one hundred and forty-three as well. What counts as a 「toxic or harmful non-food ingredient」 is not for you to decide. One group is substances the laws and regulations forbid adding to food. Another is substances on the lists announced by the relevant departments of the State Council, such as banned pesticides and drugs barred from use on animals raised for food. One more situation also counts as 「其他严重情节」 under Article one hundred and forty-four. It was made and sold to a value of 10,000 yuan or more but under 20,000 yuan. The location is a primary or secondary school campus, a childcare or elderly care institution, or the surroundings of either. The buyers are minors or elderly people. Slaughtering and selling outside a designated slaughterhouse, at a pig slaughtering site set up privately, is convicted as illegal business operations where the circumstances are serious. A turnover of 10,000 yuan or more, or illegal gains of 5,000 yuan or more, counts as 「情节严重」 (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020 revision). Criminal Law, 刑法（第一百四十一、一百四十三、一百四十四条）. (National Laws and Regulations Database, 国家法律法规数据库) ; Supreme People's Court and Supreme People's Procuratorate (最高人民法院、最高人民检察院) (2021). Interpretation on Several Issues Concerning the Application of the Law in Handling Criminal Cases Endangering Food Safety (法释〔2021〕24 号，第一、二、五、七、九、十七、十八条，2022 年 1 月 1 日起施行). (the full text reprinted by the Hebei Provincial Administration for Market Regulation) and (the full text reprinted by the Liaoning Provincial Department of Justice; the two copies were checked word for word against each other) +- Notes: The dividing line with the three items above is whether food safety itself has been breached. No licence, a missing label item, an exaggerated health claim: those are punished with money. Selling diseased or dead meat, goods over the limit, adding additives freely, or putting non-food ingredients into food puts you into criminal procedure. The cheapest defence is still the set of actions in item 8 (work out which tier of food business you fall into). Take meat only from designated slaughterhouses. Keep the quarantine certificate and the certificate of conformity for meat quality inspection for every batch. Use only ingredients whose approval number you can look up. The 「meat cleared cheaply as a bargain」 and the 「offcuts from who knows where」 are what cause the most trouble. The Supreme People's Court gazette page that is the original source returned 502 this round, so two provincial official reprints are cited instead, and they were checked word for word against each other. + +### 12. Once you hold the licence you have a duty to file returns, and with no income you still file on time (file a sheet of zeros) + +- Cost: No money. File on the electronic tax bureau by month or by quarter; each time takes a few minutes. Paying someone to keep the books is a separate cost in the market. If you use staff, you also have to complete the social insurance registration within 30 days counting from the day the entity was established. The hard part is remembering to file on time in months with no business. +- In plain language: Once you hold the licence you file tax returns on time. Not earning a single yuan does not excuse you; you still file, with zeros in every box. Failing to file brings a fine of up to 2,000 yuan first, and in serious circumstances a fine of 2,000 to 10,000 yuan. If you file nothing at all for three consecutive months, the system automatically marks you an abnormal business household and you can no longer issue invoices. +- Benefit: 「纳税人在纳税期内没有应纳税款的，也应当按照规定办理纳税申报」: for that period you pay no tax at all, and the return is filed anyway. Where the return is not filed on time, 「由税务机关责令限期改正，可以处二千元以下的罚款；情节严重的，可以处二千元以上一万元以下的罚款」, that is, the tax authority orders you to correct it within a set period and may impose a fine of up to two thousand yuan; in serious circumstances it may impose a fine of two thousand yuan up to ten thousand yuan. Where you fail to file and also underpay tax, the tax is recovered, the late payment surcharge calculated by the day has to be paid on top, and there is a separate fine of 「百分之五十以上五倍以下的罚款」, that is, from fifty per cent up to five times the tax underpaid. The surcharge accrues at five ten-thousandths of the outstanding amount per day. 「连续三个月所有税种均未进行纳税申报的，税收征管系统自动将其认定为非正常户，并停止其发票领用簿和发票的使用」. File nothing for three months and the system marks you an abnormal business household by itself; you can neither obtain nor issue invoices. The status is released automatically once you file late and accept the penalty. Where an entity uses staff, it 「应当自成立之日起三十日内……向当地社会保险经办机构申请办理社会保险登记」 (nationwide) +- Evidence grade: A +- Sources: State Council (国务院) (2002). Detailed Rules for Implementing the Law on the Administration of Tax Collection, 税收征收管理法实施细则（国务院令第 362 号，第三十二、七十五条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Law on the Administration of Tax Collection, 税收征收管理法（2015 年修正，第二十五、三十二、六十二、六十四条）. ; State Taxation Administration (国家税务总局) (2019). Announcement on Certain Matters Concerning Tax Collection and Administration, 关于税收征管若干事项的公告（国家税务总局公告 2019 年第 48 号，第三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Individual Income Tax Law, 个人所得税法（2018 年修正，第十二条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Social Insurance Law, 社会保险法（2018 年修正，第五十七条）. +- Notes: After the 「多证合一」 reform you no longer register with the tax authority separately. Log in to the electronic tax bureau with the unified social credit code on the business licence, confirm the registration details, and declare which taxes you owe. Sole proprietorships and sole proprietorship enterprises file individual income tax on business income. First pay once 「在月度或者季度终了后十五日内」, that is, within fifteen days after the end of the month or of the quarter, then 「在取得所得的次年三月三十一日前办理汇算清缴」, the annual settlement return due before 31 March in the year after the income is obtained (totalling up the year, with a refund or a top-up). You file even if you never opened, and a nil return is still a return. Leave it alone for three months and you become an abnormal business household, invoices stop working, and you cannot use the simplified cancellation procedure either; see item 23 in this section (if it fails, exit through the proper procedure) + +### 13. Issue invoices only for real transactions, and small-scale taxpayers should use the whole exemption threshold + +- Cost: No money. Do not buy invoices, do not sell invoices, and do not issue invoices for someone else's 「走账」. Issue invoices and file tax on the basis of the business that actually happened. +- In plain language: If you are a small-scale taxpayer on the simplified method, and you sell less than 100,000 in a month or less than 300,000 in a quarter, value-added tax is exempt. The rate that would have been 3% is reduced to 1% in the same period. In the other direction, 「finding someone to issue invoices to offset costs」 is a false issue of invoices, starting at 3 years or less, plus a fine of 20,000 to 200,000 yuan. +- Benefit: The category of small businesses paying tax on the simplified method is called a small-scale taxpayer. The law says 「销售额未达到起征点的，免征增值税」. That line is the threshold. From 2026-1-1 to 2027-12-31 the threshold is 100,000 yuan a month, 300,000 yuan a quarter, and 1,000 yuan per transaction or per day, and sales below the line are not taxed. Over the same period, a rate that would have been 3% is reduced to 1%, except on the sale or rental of housing and the transfer of land use rights. In the other direction, issuing a value-added tax special invoice with no real sale behind it, or any other invoice that can be used to offset tax, carries 3 years or less of fixed-term imprisonment or criminal detention, together with a fine of 20,000 to 200,000 yuan. Issuing one for someone else, issuing one for yourself, getting someone else to issue one for you, and introducing others who issue them all count. Where the tax involved is relatively large, the sentence is 3 to 10 years. Where the amount is huge, 10 years or more or life imprisonment (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2024). Value-Added Tax Law, 增值税法（2026 年 1 月 1 日施行，第二十三条）. ; Ministry of Finance and State Taxation Administration (财政部、税务总局) (2026). Announcement on the Transitional Matters Concerning Value-Added Tax Preferential Policies after the Value-Added Tax Law Takes Effect, 关于增值税法施行后增值税优惠政策衔接事项的公告（财政部 税务总局公告 2026 年第 10 号，第一、二、六条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百零五条）. +- Notes: The Value-Added Tax Law takes effect from 2026-1-1. The two earlier notices on relief for small-scale taxpayers, 2023 No. 19 and 2023 No. 1, stop being applied or are repealed along with it. The amount thresholds have not changed, but the legal basis has, and this item has been updated to the new documents. The page carrying the original Criminal Law text is the 1997 edition, and the sentencing tiers in Article two hundred and five were later adjusted by amendment. This item writes only the tiers that were verified and that have not been changed. What is exempt is value-added tax; income tax is calculated separately. 「Finding someone to issue invoices to offset costs」 is a false issue of invoices, not tax saving. + +### 14. Anyone claiming to be the tax bureau and pressing you to pay and file is a fraud; handle tax only through the electronic tax bureau and 12366 + +- Cost: No money. If you get a call or a message about a 「tax audit」, a 「abnormal account」 or 「no tax filing means a fine」, or if someone adds you as a contact, hang up and do not reply. Then call 12366, or call the landline number of the tax authority in charge and check there. +- In plain language: The tax department will not phone you to ask for your bank card password, your payment password or a verification code sent by message, and it will not give you an account to send the tax into. Genuine on-site inspection of the accounts has to show an inspection notice and an inspection certificate. If you get such a call, hang up first and then call 12366 to check. If you have already transferred the money, call 110 or 96110 at once and ask for the payment to be stopped. +- Benefit: The State Taxation Administration's notice says: 「税务部门不会通过微信小程序、电话、短信等方式索要银行卡密码、支付密码、账户余额及与银行卡收付款项相关的短信验证码」. 「一些不法分子冒充税务干部以提供『帮办纳税申报』『税收优惠代办』『政策辅导』等为名接触纳税人」. In plain terms: some criminals pose as tax officials and approach taxpayers under the pretext of offering 「help with filing returns」, 「handling tax relief formalities」 or 「policy guidance」. To check, call the 「12366 纳税缴费服务热线或主管税务机关固定电话」. Genuine on-site inspection of the accounts (稽查) 「必须出示检查通知书和检查证」. 「税款只能通过电子税务局或者在办税大厅缴纳，税务部门不会提供指定账号」. The agencies that handle tax formalities for others are covered by a rule of their own. 「以税务机关和税务人员的名义敲诈纳税人、扣缴义务人的」 is fined according to how serious the case is. Extorting the people who pay tax and the units that withhold and remit tax for them while flying the flag of the tax bureau or of tax officials is exactly that case. Impersonating a state organ functionary to deceive and extort carries detention of 10 days or more and 15 days or less, and may be followed by a fine of up to 1,000 yuan. Where the conduct is both fraud and deception by impersonation, it is 「依照处罚较重的规定定罪处罚」, that is, whichever provision carries the heavier penalty is the one it is convicted and punished under (nationwide) +- Evidence grade: A +- Sources: State Taxation Administration (国家税务总局) (2025). 税务部门提醒广大纳税人——增强防范意识 警惕互联网涉税诈骗 (The tax department reminds all taxpayers to strengthen their awareness of prevention and beware of tax fraud on the internet). ; Yangzhou Tax Service, State Taxation Administration (国家税务总局扬州市税务局) (2023). 扬州防范涉税诈骗宣传篇 (Yangzhou guide to avoiding tax fraud). ; State Taxation Administration (国家税务总局) (2025). Measures for the Administration of Professional Tax Services (for Trial), 涉税专业服务管理办法（试行）（国家税务总局令第 58 号，第三十二条第七项、第三十三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，2026 年 1 月 1 日施行，第六十二条）. ; Supreme People's Court, Supreme People's Procuratorate and Ministry of Public Security (最高人民法院、最高人民检察院、公安部) (2016). Opinions on Several Issues Concerning the Application of the Law in Handling Criminal Cases such as Telecom and Online Fraud, 关于办理电信网络诈骗等刑事案件适用法律若干问题的意见（法发〔2016〕32 号）. +- Notes: These scams pick out newly registered sole proprietorships and small companies. The script is 「the system shows you have not filed your taxes, if you do not deal with it today you will be fined」, and then they steer you to their own public account or mini-programme and charge a few hundred yuan at a time. The filing itself is free and takes you a few minutes. Using an agency in the market to keep your books is not illegal, but that agency is not a government department and it will not chase you in the name of the tax bureau. If you have already paid or transferred the money, follow item 2 in section 8 (stopping the payment after being scammed) and call 110 or 96110 at once to ask for the stop. + +### 15. Write 「定金」 when you take money, not 「订金」, put a figure on the liquidated damages, and vet a credit period as if it were a loan + +- Cost: No money. In the contract write the two characters 「定金」, with an amount not more than 20% of the total contract value. Take the money first and deliver afterwards. Before you agree to let the other side take the goods and pay later (granting a credit period), look into the other side first. +- In plain language: Only the two characters 「定金」 bring the double rule: if you break the contract, the money cannot be recovered; if the other side breaks it, you get double back. The part above 20% of the contract value does not count. Written as 「订金」, the law has no such rule. Granting a credit period is lending your own money to the other side for nothing; if they fail, it is a bad debt. +- Benefit: A deposit counts only once the money has actually reached the other side (it is formed on actual delivery). If the party paying the deposit breaks the contract, the money cannot be recovered. If the party receiving the deposit breaks it, the original sum must be returned and an equal amount paid on top (double return). The part above 20% of the total contract value (the subject matter value of the main contract) is not treated as a deposit. Where the agreed liquidated damages are wildly higher or wildly lower than the actual loss, the court may lower them or raise them at the request of one party. Where both the liquidated damages and the deposit are written into the contract, a real breach lets you claim one or the other, not both (nationwide, in force from 2021) +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第五百八十五、五百八十六、五百八十七、五百八十八条）. +- Notes: 「订金」 has none of the rules that a deposit has under the Civil Code, and it is generally treated as an advance on payment. That is the standard view in practice, and this section has not verified any case law. A credit period is lending your own money to the other side interest-free; if they fail, it is a bad debt. Before granting one, go to the National Enterprise Credit Information Publicity System and look up the other side's record, to see whether it has been sued. + +### 16. Sign a written contract within the first month of employment and complete the social insurance registration within 30 days + +- Cost: The employer's share of the social insurance contributions has to go on being paid, and the rate differs from place to place. Signing the contract itself costs nothing; just use the model contract from the local human resources and social security department. +- In plain language: If a person has worked a full month without a written contract, from the second month you must pay double wages every month, and that runs until the full year is up. Social insurance has to be registered within 30 days of the employment starting; failing to do it brings a fine of 1 to 3 times the contributions due. The law has no such thing as 「the employee voluntarily gives up social insurance」. +- Benefit: Where a person has worked a full month but not a full year without a written contract, you must pay double wages every month. Counting from the person's first day at work, you must complete the social insurance registration for them within 30 days. Failing to do it first brings an order to correct. Past the deadline, the fine is 1 to 3 times the social insurance contributions due, and the person who handled it is fined a further 500 to 3,000 yuan. Where social insurance is not paid on time or not paid in full, a late payment charge accrues from the day it fell due at five ten-thousandths per day. Past the deadline it is still unpaid, there is a further fine of 1 to 3 times the arrears. The employer has to file itself and pay the full amount on time. Only a statutory cause that cannot be resisted, such as a natural disaster, allows late or reduced payment. There is no provision for 「the employee voluntarily gives up」 (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第十、八十二条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2010). Social Insurance Law, 社会保险法（2018 年修正，第五十八、六十、八十四、八十六条）. +- Notes: Double wages run from the second month right up to the full year, so the longer you delay the more it costs. For an employee with no signed contract and no social insurance, once there is a work injury, or once they go to labour arbitration after leaving, the back payment plus the fine plus the damages often come to far more than the money you saved. + +### 17. Pay wages according to the contract and the law, and put raises, loans and bonuses in writing + +- Cost: No money. Write the wage clearly into the contract. Raises, bonuses, money you advance and loans all go on a signed piece of paper. No verbal promises. Do not pay wages out of your own pocket on the company's behalf either. The hard part is following the same rules when a friend asks. +- In plain language: Wages must be paid on time and in full according to the contract and the national rules. If they go unpaid, the employee can go straight to the court for a payment order. A raise, a bonus or an advance agreed only in words means that whoever cannot produce evidence loses once the argument starts. If you lend money to an employee out of your own pocket, write a written IOU stating the amount, the interest and the date for repayment. +- Benefit: What an employment contract must state includes how wages are paid and how social insurance is contributed. The employer must pay wages on time and in full according to the contract and the national rules. Where payment is delayed, the employee can apply directly to the court for a payment order (a document the court issues to chase the debt). When the boss lends an employee money out of their own pocket, that is a loan contract. Except where two natural persons have agreed otherwise, it has to be put in writing, stating how much was lent, how the interest is calculated, when it is repaid and how. A bonus agreed verbally has nothing on paper behind it, and whoever cannot produce evidence loses once the argument starts (nationwide) +- Evidence grade: C +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第十七、三十条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第四百六十九、六百六十八条）. +- Notes: This is not about refusing to help people; it is about helping them through the system too. Once favours outside the system become a habit, stopping them causes disputes. When a private loan is mixed into the company accounts and the company fails, the bad debt lands on you personally. The statutory parts have been verified, and the advice itself comes from experience, hence grade C. + +### 18. Sell before you build: use pre-orders and small orders to prove someone will buy, then put money into production + +- Cost: From tens to hundreds of yuan, to make a sample or a page. Then go and find some dozens of real customers and get them to pay in advance or place small orders. That is several weeks slower than starting production straight away. The hard part is holding back from making the goods first. +- In plain language: One study followed 116 new Italian companies for about a year. Entrepreneurs who set up a hypothesis first and then tested it against market feedback did better, were also more willing to change direction in time, and did not give up any earlier. So if nobody will pay in advance and nobody will place a small order, do not believe that 「it will sell once it exists」. +- Benefit: A randomised trial that followed 116 Italian start-ups for about a year. A randomised trial means splitting people at random into two groups, training only one of them, and then comparing the results of the two groups. The finding was this: entrepreneurs who had been trained to decide by 「putting forward a hypothesis first, then testing it strictly against market feedback」 performed better, were also more likely to change course in time for another plan, and were not more likely to abandon the business early. The authors explain it this way: the practice lowers the chance of going on pouring money into projects that look promising but will not work out (false positive projects) (Italy, published 2020) +- Evidence grade: B +- Sources: Camuffo A, Cordova A, Gambardella A, Spina C (2020). A Scientific Approach to Entrepreneurial Decision Making: Evidence from a Randomized Control Trial. Management Science 66(2):564-586. +- Notes: There is only one such study, the sample is small, and it was done in Italy, so the grade is B. The title, the source and the abstract were checked against Crossref and Semantic Scholar, and the effect figures in the full text have not been verified for this section. In practice: if nobody will pay in advance and nobody will place a small order, do not believe that 「it will sell once it exists」. + +### 19. Once the sample is made, run it through the production checklist before you talk about starting + +- Cost: A few hundred yuan for one small trial batch. Then spend a few hours working out these items: the yield (how many out of 100 pieces can actually be sold), the supplier's minimum order quantity and payment terms, how much money the inventory ties up, packaging and freight, certification and licences, and the trademark. The hard part is holding back from skipping this step and starting production directly. +- In plain language: The difference between one sample and one thousand pieces is where the money gets tied up. The yield sets the real cost. The minimum order quantity and the payment terms set how much you have to advance first. The inventory sets when the money comes back. If your product is on the list of mandatory certifications, without the certification and without the certification mark on the product it may not leave the factory or be sold. +- Benefit: The difference between one sample and one thousand pieces is where the money gets tied up. The yield sets the real cost, and the yield is the share of a batch you produce that passes and can be sold. How little the supplier will let you order in one go (the minimum order quantity), and how long after dispatch you receive the money (the payment terms), together set how much you have to advance first. The inventory sets when the money comes back. A product on the list of mandatory product certification must first pass certification, and only with the certification mark on the product may it leave the factory and be sold. Leaving the factory and selling without certification is penalised under the Regulations on Certification and Accreditation. Food service and selling food need a food business licence (nationwide) +- Evidence grade: C +- Sources: General Administration of Quality Supervision, Inspection and Quarantine (质检总局) (2009). Provisions on the Administration of Mandatory Product Certification, 强制性产品认证管理规定（质检总局令第 117 号，第二、四十九条）. ; State Administration for Market Regulation (国家市场监督管理总局) (2023). Measures for the Administration of Food Business Licences and Filings, 食品经营许可和备案管理办法（市场监管总局令第 78 号，第四条）. +- Notes: The checklist is itself a method from experience; the statutory parts have been checked against the original text. Before production starts, go to the trademark search system of the National Intellectual Property Administration and check whether a similar mark is already registered, then go on to apply for your own. For the lines on using someone else's trademark and artwork, see items 20 and 21 (keep the receipts when buying stock, and make your own artwork or buy a licence). The list of mandatory certifications is whichever the State Administration for Market Regulation has published most recently. + +### 20. Keep the receipts and the supplier's details for every batch you buy, and do not buy stock priced well below the market: when an employee buys counterfeit goods, the boss is the one convicted + +- Cost: No money. Keep these for every batch: the purchase contract, the invoice, the payment record, and the supplier's business licence and authorisation documents. Then set one internal rule: whoever buys the goods signs for them, and anything priced well below the market is not bought at all. The hard part is turning down the low-priced stock that looks very profitable. +- In plain language: Selling counterfeit goods is enough for a conviction at 30,000 yuan of earnings or 50,000 yuan of sales, with 3 years or less. Stock sitting in the warehouse unsold counts as well. 「I did not know it was fake」 is not for you to decide: a buying price plainly below the market with no reason you can give is treated as knowledge. Only with the purchase contract, the invoice, the payment record and the supplier's details do you have a route to escaping liability. +- Benefit: Knowingly selling goods bearing a counterfeited registered trademark: where the earnings (illegal gains) reach 30,000 yuan, or the amount sold reaches 50,000 yuan, you have reached the threshold for criminal punishment. The sentence is 3 years or less, with a fine, or a fine alone. Reaching ten times that standard, the sentence is 3 to 10 years together with a fine. Stock still sitting in the warehouse unsold counts too: where the value of the goods reaches three times the sales amount standard, that too amounts to the crime. You do not have to admit that you 「knew」. The 2025 judicial interpretation of the Supreme People's Court and the Supreme People's Procuratorate lists five situations in which knowledge is directly established, unless you can produce evidence that you genuinely did not know. First, knowledge that the registered trademark on the goods has been altered, swapped or covered over. Second, forging or altering the trademark owner's authorisation document, or knowing that the document has been forged or altered. Third, having been punished criminally or administratively for selling counterfeit goods and then selling the same kind again. Fourth, buying or selling with no justifiable reason at a price plainly below the market price. Fifth, after being found by an administrative law enforcement agency or a judicial organ, transferring or destroying evidence such as the infringing goods and the accounting records, or providing false proof. Where a company commits this, 「对单位判处罚金，并对其直接负责的主管人员和其他直接责任人员」 is punished under the same article. The company pays the fine, and the person who made the call and the person who handled it are punished under the same article. So if an employee bought the goods, the one who is fined and convicted is still the person who made the call. In the other direction, the Trademark Law leaves a way out: where at the time of sale you genuinely did not know the goods infringed another's registered trademark, 「能证明该商品是自己合法取得并说明提供者的，不承担赔偿责任」. Whether you can escape liability depends entirely on whether you have the receipts and the supplier's details; without them that provision is unavailable to you (nationwide, the judicial interpretation in force from 2025-4-26) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第十七、二十四项，即刑法第二百一十四、二百二十条）. ; Supreme People's Court and Supreme People's Procuratorate (最高人民法院、最高人民检察院) (2025). Interpretation on Several Issues Concerning the Application of the Law in Handling Criminal Cases of Infringement of Intellectual Property Rights, 关于办理侵犯知识产权刑事案件适用法律若干问题的解释（法释〔2025〕5 号，2025 年 4 月 26 日起施行）. ; Standing Committee of the National People's Congress (全国人大常委会) (2019). Trademark Law, 商标法（2019 年修正，第五十七、六十四条；2026 年 6 月 26 日修订后为第七十二、七十八条，自 2027 年 1 月 1 日起施行）. and +- Notes: Cosmetics, alcoholic drinks, mother and baby goods, digital products and clothing are the categories where these cases cluster. The common point is that the same product has a very wide gap in price. Labels like 「channel goods」, 「end-of-line stock」 and 「overseas duty-free」 do not replace the receipts. Beyond criminal punishment there is also an administrative penalty: where infringement is established, the infringing goods and the tools are confiscated, and where the turnover is 50,000 yuan or more a fine of up to 5 times may be imposed on top. If you find something wrong after the stock has arrived, stop selling and seal it at once, then contact the brand owner or the market regulation department. Do not go on selling, and do not destroy the goods or the records. Destroying evidence lands you squarely in the fifth situation above, which 「presumes you knew the goods were counterfeit」. For using someone else's artwork see item 21 (make your own artwork or buy a licence). There is one routine aimed squarely at this: someone turns up uninvited and offers to give you a batch of pens or other small goods free, to put in the shop, sell and split the takings. A few days later someone comes to buy, takes evidence on the spot, then claims in damages for infringement. Free goods fail all three conditions for escaping liability: there is no purchase contract and no payment record, the real supplier cannot be named, and the price is plainly abnormal. Not one element of 「legitimately obtained with the supplier identified」 is available. So goods that turn up unsolicited, come with no receipts and leave you only a WeChat ID should never go on the shelf. If you really want to sell on consignment, sign a written consignment contract, keep a copy of the other side's identity card and business licence, and write in who owns the goods and who is liable if there is infringement. + +### 21. The artwork on the goods, the packaging, the hang tags and the promotional images is either your own or licensed; changing a colour or adding an icon does not count as 「altered」 + +- Cost: Licences for a genuine stock library start at tens of yuan per image. Having a designer produce something original costs a few hundred to a few thousand yuan. For every image used on the goods, keep the licence certificate and the purchase record. The hard part is not using images you found at random by searching online. +- In plain language: Changing a colour on someone else's image, swapping the background colour or adding your own icon is still using their work without permission. Copying someone else's artistic work for profit, to the point of 30,000 yuan earned, 50,000 yuan of business done, or 500 copies printed, amounts to the crime of infringing the copyright, and carries 3 years or less. On the trademark side, being 「close enough to cause confusion」 is already infringement. +- Benefit: The Copyright Law puts 「未经著作权人许可，以改编、翻译、注释等方式使用作品」 on the same footing as direct copying as an infringing act. So changing the colours on the original image, swapping the background colour or adding your own icon is still using someone else's work without permission. The threshold for criminal punishment is here: for profit, without the permission of the copyright owner (著作权人) of the work, printing artistic works for sale (reproduction and distribution). Three figures bring it within the crime. The earnings (illegal gains) reach 30,000 yuan. Or the business reaches 50,000 yuan (the amount of illegal business operations). Or the copies printed add up to 500 (sheets). Reaching any one of these figures amounts to the crime of infringing the copyright. The sentence is 3 years or less, with a fine, or a fine alone. Reaching ten times the standard, the sentence is 3 to 10 years. Merely selling infringing copies printed by others, with earnings of 50,000 yuan or more or sales of 100,000 yuan or more, carries 5 years or less under the crime of selling infringing reproductions. Trademarks work the same way: using on the same goods a mark similar to a registered trademark, or using the same or a similar mark on similar goods, is already within infringement where it is 「容易导致混淆的」. 「近似」 is written into the statute as an infringing situation, so a small modification is still infringement (nationwide, the judicial interpretation in force from 2025-4-26) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Copyright Law, 著作权法（2020 年修正，第五十二条第六项、第五十三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第二十、二十一项，即刑法第二百一十七、二百一十八条）. ; Supreme People's Court and Supreme People's Procuratorate (最高人民法院、最高人民检察院) (2025). Interpretation on Several Issues Concerning the Application of the Law in Handling Criminal Cases of Infringement of Intellectual Property Rights, 关于办理侵犯知识产权刑事案件适用法律若干问题的解释（法释〔2025〕5 号）. ; Standing Committee of the National People's Congress (全国人大常委会) (2019). Trademark Law, 商标法（2019 年修正，第五十七条第二项；2026 修订后为第七十二条第二项）. +- Notes: Three traps come up most often. First, printing cartoon characters and celebrity photos found by searching online onto clothes, cups and phone cases. Second, typefaces: most commercial typefaces need a licence of their own, and a free preview is not a free commercial licence. Third, unlicensed assets slipped into a design file made for a customer; the liability still sits with the person selling the goods. The low-effort route is to use an image library that carries a commercial licence and to save the licence page. Or have the designer state in the contract where the assets came from and how far the licence extends, with the designer responsible if something goes wrong. Beyond criminal punishment there is civil damages and an administrative fine: where the public interest is harmed, the illegal gains may be confiscated and a fine of 1 to 5 times the amount of illegal business operations imposed. For checking a trademark before production, see item 19 (once the sample is made, run it through the production checklist). In the other direction, if someone else uses your own images, you enforce your rights under these same rules. + +### 22. Do your job by the rules, and before a favour outside the rules, work out whose responsibility it is + +- Cost: No money. At work, follow the system, the medical orders and the process. Do not make decisions for other people beyond your own authority. Before you step in to help, ask yourself one question: if something goes wrong, who is responsible. The hard part is resisting a request to 「make an exception」 on the spot. +- In plain language: Take a nurse as an example. The regulation states that on finding a patient's condition critical the doctor must be notified at once, and on finding medical orders that break the rules the point must be put to the doctor who gave them. Other jobs work the same way: first see where your own rules draw the line for 「must report」. The law does not leave you standing by while someone dies either: a voluntary emergency rescue that injures the person rescued carries no damages. +- Benefit: Taking a nurse as an example, the professional boundary the regulation sets out is this. On finding a critical condition, notify the doctor at once. In an emergency, in order to save a dying patient, the necessary emergency care may be carried out first. On finding medical orders that break the law or violate the technical standards for diagnosis and treatment, raise it with the doctor, and report it to the head of the department if necessary. The Civil Code provides separately: where a person voluntarily performs emergency rescue and as a result injures the person rescued, the rescuer does not have to pay damages (bears no civil liability) (nationwide) +- Evidence grade: B +- Sources: State Council (国务院) (2008). Regulations on Nurses, 护士条例（国务院令第 517 号，第十七条）. ; State Council (国务院) (2020). Regulations on Nurses, 护士条例（2020 年修订，第十七条）. (National Laws and Regulations Database, 国家法律法规数据库) ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一百八十四条）. +- Notes: Within your own role, act by the rules, and rescuing someone in an emergency is protected by law. For everything else, first work out whose authority it is and whose responsibility it is, and do not let a private good intention stand in for a decision the system calls for. The Regulations on Nurses were revised in 2020 by State Council Order No. 726, which changed only the provisions on registration and left Article 17 untouched. This round the official text of the provisions in the Physicians Law about practising according to the diagnostic and treatment standards could not be opened, so they are not cited. + +### 23. If it fails, exit through the proper procedure: cancel through the simplified route if you can, go through bankruptcy if the assets do not cover the debts, and do not simply leave it alone + +- Cost: You can do the simplified cancellation yourself: post it on the National Enterprise Credit Information Publicity System for 20 days. Where debts are still outstanding, you first have to sort the accounts, collect what is owed and pay what is due; that step is called liquidation. Going through bankruptcy means applying to the court, which takes months. The hard part is not wanting to face it after a loss and putting it off indefinitely. +- In plain language: Where there are no debts and no wages, taxes or social insurance outstanding, all the investors sign a commitment, and if nobody objects during the 20 days of posting it can be cancelled through the simplified route. A sole proprietorship needs no posting: if there is no objection within 10 days it is cancelled outright. If debts cannot be paid, apply for bankruptcy. Leave it alone, fail to file an annual report for 2 years and cannot be reached, and the licence is revoked; the legal representative who bears responsibility cannot be a legal representative again for 3 years. +- Benefit: There are no debts to others and no debts owed to you, or they have all been settled. No wages, social insurance or taxes outstanding. All the investors have signed a written commitment. Meeting these conditions and posting it for 20 days without objection lets you go through the simplified cancellation. A sole proprietorship needs no posting; if the competent department raises no objection within 10 days it is cancelled outright. A company that cannot pay a debt that has fallen due and whose debts exceed its assets (being insolvent) may apply to the court for reorganisation, settlement or bankruptcy liquidation. Where a company has already been dissolved without liquidation and its debts exceed its assets, the person responsible for the liquidation should apply for bankruptcy liquidation. Leaving it alone brings these consequences. Failing to post the annual report on time gets the business placed on the list of abnormal operations. Failing to file an annual report for 2 consecutive years, being placed on that list and not correcting it, and being unreachable at the registered address gets the business licence revoked. A legal representative of a company whose licence is revoked, who bears personal responsibility for it, may not act as a company's legal representative or as a director, supervisor or senior manager for 3 years counting from the day of revocation. Once a business is on the list of abnormal operations, even the simplified cancellation is unavailable (nationwide) +- Evidence grade: A +- Sources: State Council (国务院) (2021). Regulations on the Registration of Market Entities, 市场主体登记管理条例（国务院令第 746 号，第三十一、三十二、三十三条）. ; State Administration for Market Regulation and five other departments (市场监管总局等六部门) (2025). Guide to the Cancellation of Enterprises, 企业注销指引（2025 年修订）（2025 年第 52 号公告）. ; Standing Committee of the National People's Congress (全国人大常委会) (2006). Enterprise Bankruptcy Law, 企业破产法（第二、七条）. ; State Council (国务院) (2024 revision). Interim Regulations on the Publicity of Enterprise Information, 企业信息公示暂行条例（国务院令第 777 号修订，第十八条）:「企业因连续2年未按规定报送年度报告被列入经营异常名录未改正，且通过登记的住所或者经营场所无法取得联系的，由县级以上市场监督管理部门吊销营业执照」. In plain terms: where an enterprise fails to file an annual report for 2 consecutive years, is placed on the list of abnormal operations, does not correct it, and cannot be reached at its registered address or place of business, the market supervision department at county level or above revokes its business licence. (National Laws and Regulations Database, 国家法律法规数据库) ; State Council (国务院) (2024). Decision to Amend and Repeal Certain Administrative Regulations, 关于修改和废止部分行政法规的决定（国务院令第 777 号）. ; Regulations on the Registration of Market Entities (第十二条第四项）and the Company Law (2023 年修订，第一百七十八条第一款第四项): 担任因违法被吊销营业执照的企业的法定代表人并负有个人责任的，自被吊销营业执照之日起未逾 3 年不得担任法定代表人和董事、监事、高级管理人员 (a legal representative of an enterprise whose business licence has been revoked for illegal conduct, who bears personal responsibility, may not act as a legal representative or as a director, supervisor or senior manager until 3 years have passed since the licence was revoked). ; +- Notes: The Interim Regulations on Enterprise Information Disclosure were revised in 2024 by State Council Order No. 777, took effect on 1 May and have not been changed since. The old Article 17, 「经营异常满 3 年列入严重违法企业名单」, has been deleted. The list of serious violations and loss of trust now covers enterprises that falsify their annual reports and cases where the circumstances are serious. Much of what is written online is still the old version, so do not copy it. The personal bankruptcy system, which includes sole proprietorships, is currently only being trialled in a few regions; this section has not verified the original text, so nothing is written about it. diff --git "a/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" "b/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" new file mode 100644 index 00000000..14961e32 --- /dev/null +++ "b/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" @@ -0,0 +1,389 @@ +[← 回总目录](../README.md) + +# 13. 紧急情况：先做什么 + +本节讲出了事先做什么、再做什么、什么不要做。急救和野外的条目算的是特定死因和活下来的比例。威胁和骗局的条目算的是钱和人身自由。两类数字不互相换算。条目的顺序按「这事多常见 × 做对做错差多少条命」从高到低排。这些条目首先是给你自己和配偶、直系亲属用的。心脏骤停有 79.2% 发生在家里，你最可能救的就是家里人。对朋友同事次之。对陌生人也不是一点好处没有，但这一档最低，还可能被讹、被卷进案子。所以这类条目把保护自己的动作和免责的条款写在前面。 + +本节只收几分钟内要做的动作。认出骗局、被人拿把柄要钱，处置要按天算，不在这一节。反诈的硬规则和七种高发骗局见第 8 节第 3 条（反诈硬规则）。被人拿隐私照片、裸聊录像要钱见第 8 节第 32 条（对方拿把柄来要钱）。钱已经转出去了，立刻打 110 或 96110 要求止付，见第 8 节第 2 条（止付）。 + +[← Back to the contents](../README.md) + +# 13. Emergencies: what to do first + +This section says what to do beforehand, what to do next, and what not to do. The first aid and outdoor entries measure specific causes of death and the share of people who survive. The threats and scam entries measure money and personal freedom. The two sets of numbers are not converted into each other. The entries are ordered by how common the emergency is and how many lives the right or wrong action saves. These entries are written first for you, your spouse and your immediate family. 79.2% of cardiac arrests happen at home, so the person you are most likely to save is someone in your own home. Friends and colleagues come next. A stranger still gains something, but that group is the lowest, and you can be extorted or pulled into a case. So these entries put the steps that protect you, and the clauses that limit your liability, first. + +This section only covers actions that take minutes. Recognising a scam, or being blackmailed with money, is handled on a scale of days and does not belong here. See item 3 in section 8 (hard anti-fraud rules) for the hard rules and the seven most common scams. See item 32 in section 8 (being blackmailed with private material) for demands for money backed by private photos or nude videos. If the money has already gone out, call 110 or 96110 at once and ask for the payment to be stopped. See item 2 in section 8 (stopping a payment). + +### 1. Someone collapses and is not breathing: press hard on the centre of the chest, send someone else to call 120 and fetch an AED + +- Cost: No money. A cardiopulmonary resuscitation course from the Red Cross or a hospital runs half a day to a day, and one course is enough for years. You can do it without any training: stack your hands at the midpoint of the line between the nipples, press down hard about 2 times a second, and do not stop. +- In plain language: The person you are most likely to press on is your own family. 79.2% of cardiac arrests happen at home. Where someone presses, the share discharged alive rises from 3.9% to 16.1%. If they do not answer and are not breathing, press straight away. Put your hands at the midpoint between the nipples, stack them, press hard, 2 times a second, and do not stop. Let the second person call 120 and fetch the AED. Your hands do not stop. +- Benefit: Pooling 79 studies and 142,740 people together (meta-analysis): among people whose heart stopped outside hospital (out-of-hospital cardiac arrest), only 7.6% in total were discharged alive. Where a bystander performed cardiopulmonary resuscitation, the share discharged alive rose from 3.9% to 16.1%. 53% had someone present who saw the collapse happen, and only 32% actually received compressions from a bystander. China's BASIC-OHCA registry recorded 38,227 emergency cases taken on by ambulance crews. In 30,282 of them (79.2%) the cardiac arrest happened at home. Among this group the bystander compression rate was 20.3%, and 1.2% were discharged alive. The order is: compress first, then let the second person call and fetch the AED. Once the machine is switched on, do what its voice tells you +- Evidence grade: A +- Sources: Sasson C et al (2010). Predictors of survival from out-of-hospital cardiac arrest: a systematic review and meta-analysis. Circulation: Cardiovascular Quality and Outcomes. ; Zheng J et al (2023). Incidence, process of care, and outcomes of out-of-hospital cardiac arrest in China: a prospective study of the BASIC-OHCA registry. The Lancet Public Health. ; National People's Congress (全国人大) (2020). Civil Code, 民法典第一百八十四条. +- Notes: Do not pinch the philtrum first, do not look for the family first, and do film it first. Do not hold back from pressing hard because you are afraid of breaking ribs. Article 184 of the Civil Code states that a person who gives emergency rescue voluntarily and injures the person helped bears no civil liability. + +### 2. An elderly person falls or someone collapses: kneel down and speak to them, call 120, and do not lift them straight away. Walking away from a stranger is legal too, and if you do stop, do not move them by hand + +- Cost: No money. Kneeling down to ask one question and making one call takes a few minutes. The hard part is holding back the hand that reaches out to lift them. With a stranger there is one more cost to count: calling 120 leaves your number, and you may get a call back or be questioned once. If the family really does target you, a court case runs from several months upwards. And even if you win in the end, you pay the legal fees yourself. +- In plain language: The main case here is an elderly person in your own family or someone you know. First check whether they are breathing. If not, press the chest as in item 1 in this section (pressing hard on the centre of the chest). Do not lift them even if they are breathing: they may have had a stroke or fainted before they fell, and lifting them roughly makes it worse. Kneel down and call to them, and ask whether they remember how they fell. Do not move anyone with back or lower back pain, or with a suspected fracture. Call 120 where they are. +- Benefit: The Ministry of Health (卫生部) guideline Technical Guidelines for Intervention in Falls among Older Adults (《老年人跌倒干预技术指南》) says two things. One is 「跌倒是我国 65 岁以上老年人伤害死亡的首位原因」 — falls are the leading cause of injury deaths among people over 65 in China. The other is 「发现老年人跌倒，不要急于扶起，要分情况进行处理」 — when you find an elderly person who has fallen, do not lift them in a hurry, handle the case according to its type. If the person is not conscious, call the emergency number at once. If there is vomiting, turn the head to one side and clear the mouth and nose. If breathing and heartbeat have stopped, start chest compressions at once. 「如需搬动，应保证平稳，尽量平卧」 — if the person has to be moved, keep the movement smooth and lay them flat. If the person is conscious, first ask whether they remember the fall, because 「如不能记起，可能为晕厥或脑血管意外」 — if they cannot remember it, it may be fainting or a cerebrovascular accident. With a severe headache, or a crooked mouth, unclear speech, or weakness in the limbs, 「立即扶起老年人可能加重脑出血或脑缺血，使病情加重」 — lifting the person up at once may worsen a brain haemorrhage or brain ischaemia and make the condition worse. With signs of a fracture, or back pain, or loss of bladder or bowel control, 「如无相关专业知识，不要随便搬动」 — without the relevant professional knowledge, do not move them at will. All of these cases require an immediate call to the emergency number. If the elderly person wants to get up on their own, you can help them stand up slowly, then 「坐、卧休息并观察，确认无碍后方可离开」 — let them rest sitting or lying down and watch them, and only leave once you have confirmed they are all right. On the fear of being extorted, the law says this. Article ninety of the Supreme People's Court's Interpretation on the Application of the Civil Procedure Law of the People's Republic of China has two sentences. The first is 「当事人对自己提出的诉讼请求所依据的事实……应当提供证据加以证明」 — a party must provide evidence for the facts on which its own claims for relief are based. The second is 「在作出判决前，当事人未能提供证据或者证据不足以证明其事实主张的，由负有举证证明责任的当事人承担不利的后果」 — if, before the judgment is made, a party has not produced evidence or its evidence is insufficient to prove its factual claims, the party carrying the burden of proof bears the adverse consequence. Article one thousand one hundred and ninety-eight of the Civil Code reads: 「宾馆、商场、银行、车站、机场、体育场馆、娱乐场所等经营场所、公共场所的经营者、管理者或者群众性活动的组织者，未尽到安全保障义务，造成他人损害的，应当承担侵权责任。」 — where the operator or manager of business premises or a public place, or the organiser of a mass activity, has failed to meet the duty of safety protection and has caused harm to another person, it bears tort liability. Article one hundred and eighty-four of the Civil Code: 「因自愿实施紧急救助行为造成受助人损害的，救助人不承担民事责任」 — where harm is caused to the person helped by a voluntary emergency rescue, the rescuer bears no civil liability. The Supreme People's Court has published one typical case. A pharmacy gave cardiopulmonary resuscitation to a person in cardiac arrest and broke the person's ribs. The person helped was in hospital for 18 days and claimed more than 9,000 yuan. The court ruled that the rescuer bore no civil liability +- Evidence grade: B +- Sources: Ministry of Health (卫生部) (2011). Technical Guidelines for Intervention in Falls among Older Adults (《老年人跌倒干预技术指南》), on handling an elderly person after a fall. Full text PDF at the Guangdong Provincial Center for Disease Control and Prevention , and the full on-site handling section at the Guinan County People's Government reprint ; National People's Congress (全国人大) (2020). Civil Code, 民法典第一百八十三、一百八十四条. ; Supreme People's Court (最高人民法院) (2023). Typical cases of the people's courts securing justice and efficiency and putting the core socialist values into practice (case thirteen, Qi v. Sun, dispute over the right to health, Kangping County People's Court, Liaoning). ; Supreme People's Court (最高人民法院) (second revision of 2022, in force from 2022-4-10). Interpretation on the Application of the Civil Procedure Law of the People's Republic of China, 第九十条. ; Article one thousand one hundred and ninety-eight of the Civil Code, full text reprinted above by the Supreme People's Procuratorate ; Standing Committee of the National People's Congress (全国人大常委会) (revision of 2023, in force from 2024-1-1). Civil Procedure Law, Articles one hundred and fifty-two and one hundred and sixty-four. Full text reprinted by the Shanghai Municipal Development and Reform Commission ; State Council (国务院) (2006). Measures on the Payment of Litigation Costs (State Council Order No. 481, articles six and twenty-nine). ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Law on Penalties for Administration of Public Security (2025 revision, in force from 2026-1-1, article twenty-six paragraph one, and article fifty paragraph five). +- Notes: The B is because the guideline only says how to handle the case, and gives no figure for how many fewer deaths following it would prevent. Article 184 of the Civil Code covers harm caused while giving rescue, on the condition that you acted voluntarily and the situation was urgent. It does not cover 「自作主张把人塞进私家车送医」 — putting someone into your own car and driving them to hospital on your own initiative. Waiting for 120 is safer. After you call 120, do not hang up, and do what the dispatcher on the line tells you. This line of judgement is written first for elderly people in your own family and for people you know. For someone you meet on the street and do not know, walking away is not illegal, and it is also the least trouble: the law does not require you to rescue anyone. If you do stop, count what you pay layer by layer: the time, leaving your number, being found by the family, being sued, the case reaching your workplace, being filmed and put online. See [docs/遇到陌生人出事该不该停.md](../docs/遇到陌生人出事该不该停.md). If you do decide to stop, the one thing you must not do is **move them by hand**, and the reason is medical: lifting someone roughly can worsen a brain haemorrhage or a spinal injury. There is one exception. If you are the one running the place or managing it, or you organised the activity, walking away can make you liable. The duty of safety protection in Article 1198 of the Civil Code holds you responsible for safety for the people in the room. For fall prevention itself, see item 13 in section 1 (training balance and leg strength). For the delayed symptoms after an elderly person has hit their head, see item 10 in this section. For the money after a rescuer is injured, see item 39 in this section. For keeping evidence and calling the police after being piled on online, see item 16 in section 8. +### 3. A crooked mouth, one weak arm and unclear speech: call 120 at once. Do not wait, and do not drive there yourself + +- Cost: No money. Three actions, and one time through is enough to learn them: have him smile, raise both arms straight out, and say one complete sentence. +- In plain language: A crooked mouth, one arm that will not lift, and unclear speech: call 120 at once. If the drug that opens the blocked vessel is given within 3 hours of onset, the chance of recovering to the point of basically managing on your own is about 1.75 times that without the drug. Wait until 3 to 4.5 hours and it falls to 1.26 times. Past 4.5 hours no difference can be seen. Write down the exact time the symptoms started. Do not give water or medicine by mouth, and do not drive there yourself. +- Benefit: Individual patient data from 16 randomised trials of intravenous thrombolysis, covering 6,756 patients, were pooled (meta-analysis). Intravenous thrombolysis means giving a drug that opens the blocked vessel. Within 3 hours of onset, thrombolysis with alteplase gave an odds ratio of 1.75 for recovery to basically managing on your own (modified Rankin score 0–1), 95% CI 1.35 to 2.27 (credible range). The chance of that outcome is about 1.75 times that without thrombolysis. At 3 to 4.5 hours it is 1.26 (1.05 to 1.51). Past 4.5 hours it is 1.15 (0.95 to 1.40), and statistically no difference can be seen +- Evidence grade: A +- Sources: Emberson J, Lees KR, Lyden P, et al. (2014). Effect of treatment delay, age, and stroke severity on the effects of intravenous thrombolysis with alteplase for acute ischaemic stroke: a meta-analysis of individual patient data from randomised trials. Lancet, 384(9958), 1929-1935. +- Notes: The earlier the thrombolysis, the better it works. Past 4.5 hours it is close to useless. Call 120 rather than driving yourself, because the ambulance takes you straight to a hospital with a stroke centre and warns them in advance so they can prepare. Write down the exact time the symptoms started, because the doctor needs it to decide whether thrombolysis is possible. Do not give water or medicine by mouth, and do not wait to 「休息一下看看」 — rest a bit and see how it goes + +### 4. A sudden spinning feeling, unsteady standing, seeing double or one eye going black, or being unable to touch your own nose: treat it as a stroke and call 120 + +- Cost: No money. On top of the three actions of a crooked face, arms raised and speech (item 3 in this section, a crooked mouth and unclear speech), add two more. One is to have the person stand upright with the eyes closed and see whether they sway, then ask whether they see double or whether part of the field of view is missing. The other is to point at your own nose with a finger when you suspect a stroke, then close your eyes and point again. +- In plain language: Screening on the three signs of crooked face, weak arm and speech alone misses a bit over a tenth of strokes, because some strokes start with neither a crooked mouth nor a weak arm. Add unsteady walking, double vision, or a missing patch in the field of vision, and the missed rate falls to 4.4%. A sudden spinning feeling, standing unstably, seeing double, or being unable to touch your own nose all mean the same thing: call 120 and write down the time. +- Benefit: In 2014 the stroke centre at the University of Kentucky reviewed the records of 736 cases of acute ischaemic stroke (a stroke caused by a blocked vessel). At onset, 14.1% of the patients had none of the three symptoms of crooked face, weak arm and slurred speech (FAST). Adding unsteady walking, weak legs and visual symptoms to the screening brings the missed rate down from 14.1% to 4.4% +- Evidence grade: B +- Sources: Aroor S, Singh R, Goldstein LB (2017). BE-FAST (Balance, Eyes, Face, Arm, Speech, Time): Reducing the Proportion of Strokes Missed Using the FAST Mnemonic. Stroke, 48(2), 479-481. +- Notes: A blocked vessel in the area of the cerebellum and brainstem (posterior circulation infarction) often causes neither a crooked mouth nor a weak arm. It only causes spinning, walking to one side, diplopia (seeing double), and choking on water or food, and it is most often mistaken for a neck problem, low blood sugar or an inner ear disorder. Missing the nose with your finger and not being able to stand with your eyes closed are signs that the cerebellum is involved. But no action rules a stroke out in reverse: if it comes on suddenly and does not settle, call 120 and write down the exact time of onset. The study looked at the records of one hospital only, and the authors say themselves that it needs further verification +### 5. One eye suddenly goes black as if a blind had been pulled down, even if it clears within minutes: go to the emergency department as a stroke the same day + +- Cost: No money. Emergency assessment and the blood vessel checks take up half a day to a day. +- In plain language: One eye suddenly going black as if a blind had been pulled down means a vessel at the back of the eye is blocked, and the longer it stays blocked, the less of the retina can be saved. It is also a form of brain ischaemia, and a stroke is more likely over the next few days. Where people are assessed and start treatment on the same day, the share having another stroke within 90 days falls from 10.3% to 2.1%. Go that day to an emergency department with a stroke centre. Do not wait until tomorrow to book an eye clinic. +- Benefit: A 2021 scientific statement from the American Heart Association on central retinal artery occlusion (the main vessel at the back of the eye is blocked) says: 「急性视网膜中央动脉阻塞是急症」 — acute central retinal artery occlusion is a medical emergency. A care system should recognise it as an emergency and put it in order by severity. The longer the blood is cut off, the less of the retina can still be brought back. The Oxford EXPRESS study looked at two kinds of patient: minor stroke, and brain ischaemia whose symptoms clear on their own (called a transient ischaemic attack in medicine). After this group was switched to assessment on the same day and treatment starting on the same day, the share having another stroke within 90 days fell from 10.3% (32/310) to 2.1% (6/281) +- Evidence grade: B +- Sources: Mac Grory B, Schrag M, Biousse V, et al. (2021). Management of Central Retinal Artery Occlusion: A Scientific Statement From the American Heart Association. Stroke, 52(6), e282-e294. ; Rothwell PM, Giles MF, Chandratheva A, et al. (2007). Effect of urgent treatment of transient ischaemic attack and minor stroke on early recurrent stroke (EXPRESS study): a prospective population-based sequential comparison. Lancet, 370(9596), 1432-1442. +- Notes: The B is because EXPRESS was not a randomised trial. It compares two periods, before and after the change of practice, in the same population, and other conditions may also have changed between the two periods, which cannot be separated out. A black-out of vision lasting a few minutes that then clears by itself is most often put down to being tired or to low blood sugar. It is a form of transient ischaemic attack, and the next few days are exactly the window when strokes cluster. Losing vision in one eye with no pain, no redness and no watering of the eye deserves particular attention. Do not wait until tomorrow to book an eye clinic. Go straight to the emergency department of a hospital with a stroke centre. + +### 6. One eye swollen and painful and red, with a ring of rainbow around lights, plus headache, nausea and vomiting: go to the eye emergency department the same day + +- Cost: No money. Measuring the pressure inside the eye and examining the anterior chamber at the eye emergency department takes half a day. +- In plain language: This is an acute attack of glaucoma. The fluid inside the eye cannot drain away and the pressure shoots up. The longer you wait, the more of the optic nerve the pressure damages, and damage there does not come back. During the Wuhan lockdown, patients waited an average of 241 hours before they were treated, and 21.87% of them went blind. In the same period of the following year there was no lockdown, the average wait was 121 hours, and 7.84% went blind. +- Benefit: An eye hospital in Wuhan compared two groups of patients with acute primary angle-closure glaucoma over two periods. In the 76 days of the 2020 lockdown there were 54 people and 64 eyes, and in the same period of 2021 there were 46 people and 51 eyes, with the two groups similar in their baseline characteristics. From the onset of symptoms to the start of treatment, the lockdown group took 241.84±211.95 hours and the comparison group 121.53±96.12 hours (P=0.001). The pressure inside the eye at the visit was 52.63±12.45 mmHg against 45.16±9.79 mmHg (P=0.001). The share going blind was 21.87% against 7.84%. Glaucomatous changes in the optic nerve were found in 20/64 eyes (31.25%) against 7/51 eyes (13.73%) (P=0.03). A hospital in Korea reviewed 50 eyes that had had one attack and later had the lens removed. A year later 25 of them (50%) had lost part of the visual field, to varying degrees. The longer the stretch from the onset of symptoms to getting the pressure down, the more likely the loss of field is to persist (P=0.005). A separate prospective study laid 43 people with shallow anterior chambers face down in a dark room for 1 hour. The pressure rose clearly in both eyes (P<0.01). In the dominant eye the median rise was 3.60 mmHg, and in the other eye 2.70 mmHg (P<0.05) +- Evidence grade: B +- Sources: Zhou L, Wu S, Wang Y, Bao X, Peng T, Luo W, Ortega-Usobiaga J (2022). Clinical presentation of acute primary angle closure during the COVID-19 epidemic lockdown. Frontiers in Medicine, 9, 1078237. ; Sung MS, Kim HJ, Park SW (2023). Predictors of long-term visual field outcome after an episode of acute primary angle closure. Clinical & Experimental Ophthalmology, 51(4), 291-299. ; Wang J, Wang J, Ng TK, Huang C (2025). Asymmetric intraocular pressure changes in dominant and contralateral eyes: the dark room prone provocative test. Seminars in Ophthalmology, 40(4), 325-331. +- Notes: There are two reasons for the B. First, the two Wuhan groups were compared by going back through old records, and the lockdown itself brought other differences that cannot be separated from delay. Second, the Korean group is only 50 eyes, and it is retrospective as well. The trigger is not screen time. It is two things together: the pupil dilates in the dark, and looking down or lying face down for a long time pushes the lens forward. People who already have a shallow anterior chamber are easily triggered this way, so the real high-risk group is people over 50, people who are far-sighted, and people with a family history of angle-closure glaucoma. Young people playing games through the night hardly ever have an attack like this. The most common misjudgement during an attack is to treat it as gastroenteritis, migraine or high blood pressure, because the headache and vomiting are more obvious than the eye symptoms. Losing vision in one eye with no pain and no redness is a different matter, see item 5 in this section (one eye suddenly going black). For whether blue-light glasses are worth anything, see item 16 in section 6 (blue-light glasses). + +### 7. Chest pressure, tightness or squeezing that does not ease after more than 15 minutes: call 120, do not put up with it and do not drive yourself + +- Cost: No money. +- In plain language: Chest pressure, tightness or squeezing that does not ease after more than 15 minutes: call 120. Where the vessel is opened within 55 minutes of arrival at hospital, 30-day mortality is 2.8%. Past 55 minutes it is 9.7%. Every stretch of delay kills off another patch of heart muscle. Do not drive yourself, because losing consciousness on the road is a second accident. +- Benefit: Israel ran a nationwide survey of acute coronary syndrome. What it analysed was patients with ST-elevation myocardial infarction, which is one kind of heart attack. The interval measured was from arrival at hospital to opening the blocked vessel with a balloon (balloon angioplasty). Where the interval was ≤55 minutes, 30-day deaths were 5/178 (2.8%). Where the interval was >55 minutes, 30-day deaths were 15/155 (9.7%) +- Evidence grade: B +- Sources: Karkabi B, Meir G, Zafrir B, et al. (2021). Door-to-balloon time and mortality in patients with ST-elevation myocardial infarction. European Heart Journal - Quality of Care and Clinical Outcomes. +- Notes: What the study compares is the delay after arrival at hospital, not the stretch from onset to arrival. The reasoning is the same either way: every extra stretch of delay kills off another patch of heart muscle. The pain can sit in the upper abdomen, the back or the chin, and the left arm can go numb. Diabetics and older people may have no pain at all, only breathlessness. Do not drive yourself, because losing consciousness on the road is a second accident + +### 8. Sudden tearing pain that travels from the chest through the back and on towards the waist: call 120 and say clearly that the pain is moving + +- Cost: No money. +- In plain language: Tearing pain that travels from the chest through the back and on towards the waist: when you call 120, say straight out, 「像撕开一样，疼在往下走」 — like being torn apart, the pain is moving downwards. Without surgery and on drugs alone, this condition kills 58% of patients. Its handling is the opposite of a heart attack: the doctor has to think of it before ordering a contrast CT, instead of giving you anticoagulants and thrombolytics. Say as well if the blood pressure differs a lot between the two arms, or if no pulse can be felt in one of them. +- Benefit: The International Registry of Acute Aortic Dissection (IRAD) collected 464 cases from 12 centres. Aortic dissection is the condition that causes tearing pain in the chest and back. Among the symptoms patients report first, the most common is a sudden severe sharp pain. All-cause mortality during the hospital stay was 27.4%. In type A dissection without surgery, on drugs alone, mortality was 58% +- Evidence grade: B +- Sources: Hagan PG, Nienaber CA, Isselbacher EM, et al. (2000). The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. JAMA, 283(7), 897-903. +- Notes: The handling of a dissection is exactly the opposite of that of a heart attack. A heart attack needs anticoagulants and thrombolytics, drugs that stop the blood clotting, and using them on a dissection is like letting the tear bleed. So at the hospital say 「撕裂样，疼在往下走」 — tearing, the pain is moving downwards, so that the doctor thinks of a contrast CT. A large difference in blood pressure between the two arms, and no pulse in one of them, are the clues. People whose blood pressure is poorly controlled, people with Marfan syndrome, and people with an aortic aneurysm or dissection in the family are at higher risk. Again, do not drive yourself, and do not put up with it until tomorrow + +### 9. A sudden "the worst headache of my life" that reaches its worst within an hour: go to the emergency department for a head CT + +- Cost: The emergency visit itself costs very little. One head CT in the emergency department costs several hundred yuan. +- In plain language: What matters is how fast the pain gets to its worst. A severe headache that peaks within an hour is a bleed from an artery on the surface of the brain in about 6% of cases. If you are over 40, have a painful or stiff neck, were seen to pass out, developed it during activity, felt it explode instantly, or your neck will not bend forward, any one of these means go to the emergency department for a head CT. Most of the people who go and get checked turn out to be fine, but missing one and having a second bleed is death. +- Benefit: 10 emergency departments in Canada ran a prospective cohort study of 2131 cases of severe headache that reached its worst pain within an hour, with the criteria fixed in advance and the patients then followed and recorded throughout. 132 of them (6.2%) were subarachnoid haemorrhage, a bleed from an artery on the surface of the brain. The Ottawa SAH rule has six conditions: aged 40 or over, neck pain or stiffness, and a witnessed collapse. The other three are onset during activity, onset that explodes instantly, and limited neck flexion when the doctor examines you. Meeting any one of them means you need the scan. The rule picked out every person who bled, a sensitivity of 100%, 95% CI 97.2%–100.0% (credible range). The price is that it excludes only 15.3% of the people who did not bleed (specificity 15.3%) +- Evidence grade: A +- Sources: Perry JJ, Stiell IG, Sivilotti MLA, et al. (2013). Clinical decision rules to rule out subarachnoid hemorrhage for acute headache. JAMA, 310(12), 1248-1255. +- Notes: What matters is not how bad the pain is but how fast it reaches the worst. A headache that explodes within a few seconds to a minute is the danger signal; a migraine that builds up slowly is not. With vomiting, a stiff neck, photophobia or a brief loss of consciousness, take it more seriously still. A specificity of only 15.3% means that most of the people who go for a scan on the strength of this rule come out clear. But missing one and having a second bleed is death, and the scan is worth having. Do not take a painkiller to push it down and go back to sleep +### 10. Two or three weeks to a few months after an elderly person hits their head, if they become unsteady, slow, sleepy or weak on one side: get a head CT + +- Cost: One head CT costs two or three hundred yuan. +- In plain language: After an elderly person hits their head, the blood usually does not collect until the third week. At the time of the knock there can be nothing wrong at all, and the person may not remember hitting their head. After surgery this condition comes back in up to 33% of cases and kills in up to 32%. If an elderly person suddenly becomes confused, walks unsteadily, keeps dozing off, or has weakness in one hand, first go back and ask whether they fell or hit their head in the last two or three months, then get the CT. +- Benefit: The Chinese Medical Association Branch of Neurosurgery and others have published the Expert Consensus of Chinese Experts on Drug Treatment of Chronic Subdural Haematoma (《慢性硬膜下血肿药物治疗中国专家共识》). A chronic subdural haematoma is a pool of blood that collects slowly between the layers of membrane around the brain. The technical description is a chronic space-occupying lesion formed by blood collecting between the arachnoid and the dura mater. It 「通常在颅脑外伤后第三周形成」 — usually forms in the third week after head injury — and it is common in elderly people. After surgery the recurrence rate can reach 33% and the fatality rate can reach 32%. Among patients over 90 years old, only 24% make a good recovery overall. A systematic review found 3.39 to 39.1 new cases per 100,000 people worldwide each year, and with age the figure can reach three times that over 80. The most common causes are injury and falls, followed by long-term heavy drinking, the use of drugs that stop the blood clotting (anticoagulants), and violence. A review in geriatric medicine says the presentation is 「典型包括意识模糊、认知下降、步态障碍或嗜睡，可能与其他老年综合征相似」 — typically confusion, declining cognition, gait disturbance or drowsiness, and it can resemble other syndromes of old age +- Evidence grade: B +- Sources: Zhang J et al (2021). Expert consensus on drug treatment of chronic subdural hematoma. Chinese Neurosurgical Journal, 7(1), 47. ; Dziho A et al (2025). Global prevalence and incidence of chronic subdural hematoma: A systematic review. Brain and Spine, 5, 105893. ; Mathon B, Shotar E (2026). Diagnostic and therapeutic management of chronic subdural hematoma in elderly patients. Gériatrie et Psychologie Neuropsychiatrie du Vieillissement, 24(1), 30-41. +- Notes: At the time of the knock there can be nothing wrong at all, and the elderly person may not even remember hitting their head. So when you notice at home that someone has recently become confused, walks unsteadily, keeps dozing off or has weakness in one hand, first go back and ask whether they have fallen or hit their head in the last two or three months, then get the CT. Do not put it straight down to senile dementia or to the after-effects of a previous stroke, because getting that wrong costs time. Elderly people taking aspirin, clopidogrel, warfarin or one of the new oral anticoagulants are at higher risk, and so are those who drink heavily over the long term. For fall prevention, see item 13 in section 1 (training balance and leg strength). + +### 11. One leg suddenly swollen, tight and painful to press: get seen as soon as you can. Add sudden breathlessness or chest pain and call 120 at once + +- Cost: An ultrasound of the blood vessels in the leg (a lower limb venous ultrasound) costs one or two hundred yuan. +- In plain language: One leg suddenly swollen, a full girth larger than the other, and painful when pressed: get a venous ultrasound of the leg as soon as you can. If sudden breathlessness, chest pain on a deep breath, a fast heartbeat and dizziness follow, the clot may already have travelled to the lung. Call 120 at once. Among people diagnosed with pulmonary embolism, 17.4% die within 3 months. Sitting still for more than 4 hours roughly doubles the risk, but the risk to any one person is still very low. +- Benefit: The International Cooperative Pulmonary Embolism Registry (ICOPER) collected 2454 patients with confirmed acute pulmonary embolism from 52 hospitals in 7 countries. A pulmonary embolism is a clot blocking the pulmonary artery. In this group, all-cause mortality within 3 months was 17.4% (426/2454). The World Health Organization's WRIGHT project found that travel of more than 4 hours without a break roughly doubles the risk of a clot forming in a vein and travelling to the lung (venous thromboembolism), 「大约翻倍」 — roughly doubles. But the probability for any one person is still very low: sitting still for more than 4 hours carries a risk of 「约为六千分之一」 — about one in six thousand +- Evidence grade: B +- Sources: Goldhaber SZ, Visani L, De Rosa M (1999). Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet, 353(9162), 1386-1389. ; WHO (2007). Study results released on travel and blood clots. +- Notes: Both legs swollen together is usually down to the heart, the liver, the kidneys or how well the veins are working. One leg suddenly swollen and a girth larger than the other is the typical picture of a deep vein thrombosis. The situations where this happens are long journeys by car or plane, being bedridden after an operation or a fracture, a plaster cast, pregnancy and the weeks around childbirth, cancer, and taking the oral contraceptive pill. A clot that breaks loose and blocks the pulmonary artery is a pulmonary embolism: sudden breathlessness, more chest pain when you breathe in deeply, a fast heartbeat, dizziness, and coughing up blood (haemoptysis). At that point call 120 at once. 17.4% is the death rate among diagnosed cases of pulmonary embolism, not the risk you carry because your leg is swollen. Do not frighten yourself with that number. + +### 12. Heavy bleeding: press hard on the wound with your hand first, put a tourniquet on a limb if pressure will not hold, and call 120 at the same time + +- Cost: A windlass tourniquet costs from a few tens of yuan to a little over a hundred. Keep one in the car and one in the first aid kit. Learn the sequence 「扎在伤口上方、拧到血停、记时间」 — tie it above the wound, wind it until the bleeding stops, note the time. It takes a few minutes. +- In plain language: Press hard on the wound with your hand or with clothing and do not let go. If pressure will not hold on a limb, put a tourniquet on it, then call 120, and once it is on do not loosen it. Among battlefield casualties, nine tenths of those given a tourniquet before shock survived, only one in ten of those given one after shock survived, and all 5 who needed one and did not get it died. Tie it above the wound, wind it until the bleeding stops, and note the time. Do not substitute a shoelace or a length of rope. On the neck and the torso you can only press, you must never tie. +- Benefit: 11 level-one trauma centres in Texas, United States, collected 1,026 cases of vascular injury to the limbs over 6 years, of which 181 (17.6%) had a tourniquet put on before reaching hospital. Taking the figures as they stand, mortality was 3.9% where a tourniquet was used and 5.2% where it was not. Once the other conditions are evened out (multivariable adjustment), a tourniquet is still independently associated with survival. Adjusted OR 5.86, credible range 95% CI 1.41 to 24.47. At a field hospital in Baghdad, 232 casualties: survival was 90% where a tourniquet was put on before shock appeared, and only 10% where it was put on after shock. All 5 who needed a tourniquet and did not get one died, and survival where it was used was 87%. The order is: direct pressure first, clothing or a towel will do → a tourniquet on limb bleeding that pressure will not hold → call 120 → do not loosen it +- Evidence grade: A +- Sources: Teixeira PGR et al (2018). Civilian Prehospital Tourniquet Use Is Associated with Improved Survival in Patients with Peripheral Vascular Injury. Journal of the American College of Surgeons. ; Kragh JF et al (2009). Survival with emergency tourniquet use to stop bleeding in major limb trauma. Annals of Surgery. ; Chinese Medical Doctor Association Branch of Emergency Physicians (中国医师协会急诊医师分会) et al (2020). Expert consensus on emergency use of tourniquets. 中华急诊医学杂志 29(6):773-779: 「院前急救时，止血带一旦使用则不建议松开，除非得到可以替代的彻底止血的救治」 — in prehospital first aid, once a tourniquet has been used it is not recommended to loosen it, unless a treatment that stops the bleeding completely becomes available in its place, and 「最长使用时间不应超过2 h。但如果客观情况无法到医院救治或者无替代止血办法，则在得到正规救援前不解除止血带」 — the longest time of use should not exceed 2 h, but if the situation objectively makes it impossible to get to hospital for treatment and there is no alternative way to stop the bleeding, then do not release the tourniquet before proper rescue arrives. , full text +- Notes: Do not use a rope, a shoelace or any other thin thing as a tourniquet. It will not grip an artery, and it cuts the skin. Once a tourniquet is on, do not loosen it now and then to let some blood through. Bleeding on the neck and the torso can only be pressed, never tied. In that battlefield dataset the tourniquet itself cost nobody an amputation, and there were 4 cases of short-lived nerve paralysis. The expert consensus of Chinese emergency physicians agrees: once it has been used before hospital, loosening it is not recommended unless another way of stopping the bleeding completely is available. For bleeding in the upper arm, put it on the upper third of the upper arm; for bleeding in the thigh, put it at the top of the thigh. Do not put it on the forearm, the lower leg, the elbow or the knee. The consensus recommends keeping the total time, up to arrival at hospital and after it, under 2 hours if at all possible, but where the hospital cannot be reached and there is no other way to stop the bleeding, do not release it until proper rescue arrives. A tourniquet improvised from a belt, a tie or a shirt sleeve will struggle to stop the bleeding on its own; you need a stick to wind it tight. + +### 13. Bitten or scratched by a dog or a cat through the skin: rinse alternately with soapy water and running water for 15 minutes, and get the vaccine the same day + +- Cost: The rinsing is free. The full course of rabies vaccine costs several hundred yuan. Grade III exposure also needs rabies immunoglobulin (ready-made antibodies given by injection), which costs from several hundred yuan to over a thousand. +- In plain language: Once rabies symptoms appear, the fatality rate is almost 100%, and rinsing plus the vaccine is the only route. Rinse alternately with soapy water and running clean water for 15 minutes, then get the vaccine the same day. A bite that goes right through, or broken skin that is licked, is grade III exposure, and it also needs immunoglobulin. A domestic animal that has been vaccinated does not change the handling: the grade is set by the wound, not by where the animal came from. +- Benefit: The 2023 version of the standard from the National Disease Control and Prevention Administration (国家疾控局) and the National Health Commission (国家卫健委) says in its original text: 「狂犬病病死率几乎达 100%」 — the fatality rate of rabies is almost 100%. The handling is 「用肥皂水（或者其他弱碱性清洁剂）和一定压力的流动清水交替彻底冲洗所有咬伤和抓伤处约 15 分钟」 — wash all bite wounds and scratch wounds thoroughly for about 15 minutes, alternating soapy water (or another mildly alkaline cleaning agent) with running water at some pressure. A light bite on bare skin, or a scratch that does not bleed, is grade II exposure and needs the vaccine. A bite that goes right through, or broken skin that is licked, is grade III exposure, and 「应处置伤口并注射狂犬病被动免疫制剂和接种狂犬病疫苗」 — the wound must be treated, rabies immunoglobulin must be injected, and the rabies vaccine given. WHO: about 59,000 people die of rabies worldwide every year, and 99% of it is passed on by dogs +- Evidence grade: A +- Sources: General Office of the National Disease Control and Prevention Administration (国家疾控局综合司) and General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Notice on issuing the working standard for the prevention and handling of rabies exposure (2023 version), 关于印发狂犬病暴露预防处置工作规范（2023 年版）的通知. ; WHO. Rabies fact sheet, 狂犬病 实况报道. +- Notes: Do not skip the handling because the animal is a domestic one that has been vaccinated. The standard looks at the grade the wound has reached (the exposure grade), not at where the animal came from. Do not dress the wound yourself. The ten-day observation method can only decide whether to stop the injections, not whether to start them. The same handling applies to bites from wild animals and from bats. + +### 14. After a scald: rinse immediately with cool running water for 20 minutes, and put on no toothpaste and no soy sauce + +- Cost: No money. It takes 20 minutes. +- In plain language: A full 20 minutes of cool running water within 3 hours of the injury cuts the chance of needing a skin graft later to about six tenths. It is cool water, not ice water, and ice makes the damage worse. If clothing is stuck to the wound, do not tear it off; after rinsing, cover it with a clean cloth and go to hospital. Do not put on toothpaste, soy sauce, sesame oil or flour. If you only remember after a while, rinsing again within three hours still helps. +- Benefit: Australia followed 2495 children with burn and scald injuries. In children given 20 minutes of cool running water within 3 hours of the injury, the odds ratio of needing a skin graft later was 0.6, 95% CI 0.4 to 0.8 (credible range), which is about six tenths of the figure for children who did not get it. A pooled result from 7 studies found the same thing: 「伤后三小时内 20 分钟凉流动水显著降低了需要植皮和手术处理创面的可能」 — 20 minutes of cool running water within three hours of the injury significantly lowers the chance that the wound needs a skin graft and surgical treatment +- Evidence grade: A +- Sources: Griffin BR, Frear CC, Babl F, Oakley E, Kimble RM (2020). Cool Running Water First Aid Decreases Skin Grafting Requirements in Pediatric Burns: A Cohort Study of Two Thousand Four Hundred Ninety-five Children. Annals of Emergency Medicine, 75(1), 75-85. ；Griffin B, Cabilan CJ, Ayoub B, et al. (2022). The effect of 20 minutes of cool running water first aid within three hours of thermal burn injury on patient outcomes: A systematic review and meta-analysis. Australasian Emergency Care, 25(4), 367-376. +- Notes: It is cool water, not ice water, and ice makes the damage worse. If clothing is stuck to the wound, do not tear it off; after rinsing, cover it with a clean cloth and go to hospital. Do not put on toothpaste, soy sauce, sesame oil or flour, because it changes the doctor's judgement of the depth and makes infection more likely. Rinsing again within three hours still helps +### 15. A rash over the whole body, breathlessness or dizziness: handle it as anaphylactic shock, call 120 at once and say what you think it is + +- Cost: No money. A person already known to have a severe allergy pays for their own epinephrine pen, and it is a prescription medicine. +- In plain language: A rash over the whole body, plus breathlessness or dizziness, is anaphylactic shock. Call 120 at once, say straight out 「怀疑过敏性休克」 — I suspect anaphylactic shock — and ask the ambulance to bring epinephrine. Anti-allergy drugs and steroids act too slowly to stand in for epinephrine. Anyone who has had one of these should ask a doctor for an auto-injector. Inject into the outer mid-thigh, and it goes in through clothing. +- Benefit: Epinephrine is the first-choice drug (first-line medicine) for anaphylactic shock of grade II and above, and the guideline lists it as a strong recommendation. Anti-allergy drugs and steroids act too slowly to take its place +- Evidence grade: B (strong recommendation in the guideline, but the underlying evidence grade is low) +- Sources: Chinese Medical Education Association (中国医药教育协会) et al (2022). Clinical Practice Guidelines for Emergency Treatment of Anaphylactic Shock (2020). Frontiers in Pharmacology: 「Epinephrine is the first-line medicine in GRADE II to IV anaphylaxis」「Intramuscular epinephrine should be injected in the mid-anterolateral thigh」, upper limit for intramuscular injection in adults 0.5 mg and 0.3 mg for those under 14 years, repeated every 5–15 minutes if there is no response. ; Muraro A, et al. (2022). EAACI guidelines: Anaphylaxis (2021 update). Allergy. +- Notes: The allergy guideline (EAACI) says of itself that in this field 「证据大多停留在很低的等级」 — the evidence is mostly at a very low grade. It lists a strong recommendation because the benefit is far larger than the risk, not because large trials exist. In China epinephrine is a prescription medicine, and most people do not have it to hand. What you can do is call 120, say 「怀疑过敏性休克」 — I suspect anaphylactic shock — and let the crew bring the drug. Anyone who has had a severe reaction should ask a doctor for an auto-injector and learn to inject into the outer mid-thigh (outer side, slightly to the front), which goes in through clothing. + +### 16. Someone has a fit and goes down: clear the area, lay them on their side, time it, and put nothing in their mouth + +- Cost: No money. +- In plain language: Move hard objects out of the way, lay them on their side with the face pointing down, and time the fit on your watch. Do not hold them down, do not pry the mouth open, and do not put anything in the mouth: that only damages the teeth and the jaw. Call an ambulance if the fit goes on for more than 5 minutes without stopping, if it starts again soon after stopping, if they cannot be woken afterwards or have trouble breathing, if it happens in water, or if this is the first fit they have ever had. Give water only once they are fully awake. +- Benefit: This avoids the damage to the teeth and jaw caused by prying the mouth open and putting hard objects in it. This avoids the fractures caused by holding the person down. A fit that does not stop after 5 minutes is the red line at which you must call an ambulance +- Evidence grade: B +- Sources: US CDC. First aid for seizures: 「Don't hold the person down or stop their movements」「Don't put anything in their mouth. This can hurt their teeth or jaw」「Don't offer water or food until they are fully alert」「Time the seizure to track if it lasts more than 5 minutes. If it does, seek immediate medical attention or call 911」. +- Notes: The same page lists several other situations that need an ambulance at once: another fit soon after the first one ends, not being woken afterwards or having trouble breathing, being hurt during a fit, a fit in water, and a first-ever fit. When the fit stops, lay them on their side with the face pointing down, and give water only once they are fully awake. In a household with an epileptic patient, these are the steps that get used most. Doing the same for a stranger is less likely to pay you back, but it does not make you liable either (Article 184 of the Civil Code, see item 2 in this section, do not lift them straight away). + +### 17. A person with diabetes starts shaking, sweating cold and speaking unclearly: give 15 grams of sugar first, then re-test after 15 minutes + +- Cost: No money. Carry a few sugar cubes or a small box of juice with you. +- In plain language: Give 15 grams of sugar first, sugar cubes or a small box of juice will both do. Test again after 15 minutes, and if the reading is still low, do it again. A blood glucose below 3.0 mmol/L is the low blood sugar that medicine takes seriously. Once it reaches the point where somebody else has to help you treat it, it can kill. Give an unconscious person nothing at all by mouth, because it goes into the airway. Call 120 at once. +- Benefit: A blood glucose below 3.0 mmol/L is clinically significant hypoglycaemia. Hypoglycaemia with a change in consciousness or behaviour, needing help from another person to treat it, is severe hypoglycaemia, and it can kill +- Evidence grade: A +- Sources: American Diabetes Association (美国糖尿病学会) (2025). Standards of Care in Diabetes 2025, Chapter 6 Glycemic Goals and Hypoglycemia. Diabetes Care: grading 「Level 1: Glucose <70 mg/dL (<3.9 mmol/L) and ≥54 mg/dL (≥3.0 mmol/L)」「Level 2: Glucose <54 mg/dL (<3.0 mmol/L)」「Level 3: A severe event characterized by altered mental and/or physical status requiring assistance for treatment of hypoglycemia, irrespective of glucose level」; handling 「For most individuals, 15 g carbohydrates should be ingested」「Fifteen minutes after initial treatment, repeat the treatment if hypoglycemia persists」. +- Notes: Give an unconscious person nothing to eat or drink, because it goes into the airway. Call 120 at once. The same standard also advises that anyone treated with insulin, or at high risk of hypoglycaemia, should be given a supply of glucagon. The wording is 「All individuals treated with insulin or who are at high risk of hypoglycemia...should be prescribed glucagon」. The nasal spray and the pre-filled injector are more usable than one that has to be mixed at the time. All of this assumes you are looking after yourself or a person with diabetes in your household, and the sugar you carry is carried for them. + +### 18. Someone is electrocuted: cut the power first, then move the source aside with a dry wooden stick or the like, never pull them with your hands + +- Cost: No money. +- In plain language: Cut the power first, or use something dry such as a wooden stick to move the live source aside, and only then touch the person. Pulling them with your bare hands turns one casualty into two. If they are not breathing once they are clear of the current, start compressing the chest at once, as in item 1 (pressing hard on the centre of the chest). Do not hold back because the injury came from electricity. Do not deal yourself with a fallen high-voltage line or anything near a substation. Back away and wait for the electricity authority to cut the power. +- Benefit: This avoids the rescuer being caught in the current as well, which turns one casualty into two. Where someone is not breathing after being clear of the current, do cardiopulmonary resuscitation as in item 1 in this section (pressing hard on the centre of the chest). The US CDC first aid steps are: look first, do not touch, because the person may still be in contact with the current. Cut the power if you can, and if you cannot, move the source aside with something that does not conduct, such as cardboard, plastic or wood. Once the person is clear of the current, if breathing or the pulse has stopped, start cardiopulmonary resuscitation at once. One review says that people who go into cardiac arrest after electrocution and are brought back often do well afterwards +- Evidence grade: B +- Sources: US CDC. What to Do to Protect Yourself From Electrical Hazards (First aid section). ; Spies C, Trohman RG (2006). Narrative review: Electrocution and life-threatening electrical injuries. Annals of Internal Medicine 145(7):531-537. +- Notes: "Cut the power before touching the person" is purely a rule in your own interest: if you pull with your bare hands, the second person down is you. Do not deal yourself with a fallen high-voltage line or anything near a substation. Back away and wait for the electricity authority to cut the power. Cardiac arrest caused by electrocution is handled with the usual compressions and the usual defibrillation. Do not hold back because the label says electrical injury. + +### 19. A carbon monoxide alarm goes off, or everyone in the house gets a headache and nausea at once: get outside first, then call + +- Cost: No money. +- In plain language: The alarm goes off, or everyone in the house gets a headache, nausea and dizziness at once. Take everyone outside first, then call. Do not stay inside looking for the cause, and do not go back to turn off a valve. The symptoms look like a cold, and 「出门就好转」 — it gets better once you are outside — is the strongest clue. China reported 11523 deaths from carbon monoxide poisoning in 2018. Most were in winter, and most happened at home. +- Benefit: China's CDC cause-of-death surveillance system reported 11523 deaths from carbon monoxide poisoning in 2018. The worst months were month 1 (2845 cases), month 2 (2321 cases) and month 12 (2010 cases). Across those three winter months, 66% to 73% of the deaths happened at home. These are figures reported into the system, not extrapolated from the national population. In the United States, more than 400 people a year die of accidental carbon monoxide poisoning not linked to fires. More than 100,000 go to emergency departments and more than 14,000 are admitted to hospital +- Evidence grade: B +- Sources: You J, Liu J, Zhou M (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly 2(2):32: 「In 2018, there were 11,523 deaths caused by carbon monoxide poisoning reported in China」. ; US CDC. Carbon monoxide poisoning: 「More than 400 Americans die from unintentional CO poisoning not linked to fires」「More than 100,000 visit an emergency department」「More than 14,000 are hospitalized」, common symptoms 「headache, dizziness, weakness, upset stomach, vomiting, chest pain, and confusion」. ; US Consumer Product Safety Commission. Carbon Monoxide Information Center: 「If you suspect CO poisoning, get outside to fresh air immediately, and then call 911」. +- Notes: The symptoms look like a cold, and they also look like being drunk. Everyone in the house feeling ill at once and getting better outside is the strongest clue. Do not stay inside looking for the cause, and do not go back to turn off a valve. Get the people out and then call. For how to install the alarm, see item 3 in section 1. + +### 20. Swallowed cleaning agent, pesticide or medicine: do not induce vomiting, take the bottle with you to hospital at once. Splashed in the eye or on the skin: rinse with plenty of clean water for 15 minutes + +- Cost: No money. Put cleaning agents, disinfectant, pesticides and medicines on a high shelf out of a child's reach, or in a locked cupboard. Never decant them into a drinks bottle. +- In plain language: Do not induce vomiting after someone swallows a cleaning agent, pesticide or medicine, because bringing a corrosive liquid back up burns the oesophagus a second time. First get them to drink a moderate amount of milk or egg white, then take the bottle and the instructions to hospital at once. If it is splashed on the skin or in the eye, rinse with plenty of clean water for a full 15 minutes, holding the eyelids open in the eye and not rubbing it. Mixing disinfectant with toilet cleaner gives off chlorine gas. +- Benefit: The China CDC Institute for Occupational Health and Poison Control (中国疾病预防控制中心职业卫生与中毒控制所) gives four kinds of handling. Swallowed (ingested into the digestive tract): get them to drink a moderate amount of milk or egg white at once, 「切记，不要催吐」 — remember, do not induce vomiting. Get to hospital as soon as possible, taking the packaging or the instructions with you, 「以便于医生快速判断成分并制定救治方案」 — so that the doctor can judge the ingredients quickly and work out a treatment plan. Inhaled (into the airway): move the person at once to a well-ventilated area, unbutton the collar and keep the airway clear; if there is severe coughing, breathlessness or difficulty breathing, get medical help at once. Skin contact: rinse the contact area at once with plenty of clean water for at least 15 minutes, taking care not to let the rinse splash into the eyes or onto other unbroken skin. Eye contact: rinse the eye at once with plenty of clean water or saline for at least 15 minutes. Hold the eyelids open while rinsing so the water reaches the eyeball and the inside of the lids, 「切勿揉搓眼睛」 — never rub the eyes. Go to the eye department at once afterwards. High-risk products are identified by their pH. Those with pH<3 are toilet cleaners, descalers, floor cleaners and stainless steel polishes. Some powerful drain unblockers also contain dilute sulphuric acid. Those with pH>11 are drain unblockers, heavy-duty kitchen degreasers, dishwasher powder, drain salts and cleaning powders. There is also a combination that causes harm on the spot: using disinfectant together with a toilet cleaner containing hydrochloric acid gives off chlorine gas, which irritates the airway and causes severe coughing and breathlessness, and can be fatal +- Evidence grade: B +- Sources: China CDC Institute for Occupational Health and Poison Control (中国疾病预防控制中心职业卫生与中毒控制所) (2025). Holiday safety reminder: beware of strongly acid and strongly alkaline cleaning agents. +- Notes: The B is because the official explainer says only how to handle it and how to recognise it, and gives no figure for how many fewer incidents following it would prevent. The reason not to induce vomiting is that a corrosive liquid coming back up burns the oesophagus, the tube from throat to stomach, a second time. If you cannot tell what was swallowed, take the bottle, the instructions and any vomit to hospital together. Prevention costs far less than first aid: store things by category, keep them away from food and drink, screw the cap back on and throw the empty bottle away, and never pour anything into a mineral water bottle. Wear rubber gloves and goggles when using these products, open the window, and never mix types. For carbon monoxide see item 19, and for burns and scalds see item 14. + +### 21. Acids, alkalis and other chemicals splashed on you: take off the contaminated clothing at once, rinse with plenty of running water, hold the eyelids open in the eye, and rinse for the full time before you go + +- Cost: No money. On your first day at work, find where the eyewash station and the emergency shower are. Do not wait until something happens and then go looking. +- In plain language: Three steps. First, take off the contaminated clothes, shoes and socks at once, because the parts held against the body are the worst hurt. Second, rinse with plenty of running water, at least 15 minutes for both skin and eyes, and the Hunan Provincial Department of Emergency Management gives 20 to 30 minutes, so take the longer figure. Third, hold the eyelids open while rinsing the eye, and go straight to the eye department when you are done. Use water at room temperature, not hot water. +- Benefit: Three steps on site. First, immediately take off the contaminated clothes, shoes and socks: the chemical keeps being absorbed through the skin, and the parts held against the body are the worst hurt. Second, rinse with plenty of running water. The floor given by the China CDC Institute for Occupational Health and Poison Control is at least 15 minutes each for the skin and the eyes. The figure given by the Hunan Provincial Department of Emergency Management is 20 to 30 minutes for rinsing the wound, 「强烈的化学品要更长」 — strongly caustic chemicals need longer. Take the longer of the two figures. Third, in the eye you must 「撑开眼睑，确保水流充分接触眼球及眼睑内侧」 — hold the eyelids open and make sure the water reaches the eyeball and the inside of the lids — and 「切勿揉搓眼睛」 — never rub the eyes — then go at once to the eye department. Use room temperature or slightly warm water; hot water is forbidden. Put no ointment, no red potion and nothing else on a fresh wound, and do not wrap it in a dirty cloth, because whatever you put on it changes the doctor's judgement of the depth. In law, 「化学性皮肤灼伤」 and 「化学性眼部灼伤」 — chemical burns of the skin and of the eye — are both on the statutory list of occupational diseases, and cases arising at work are handled as an occupational disease and a work injury. When an accident involving acute occupational hazards happens, the employer 「应当及时组织救治、进行健康检查和医学观察，所需费用由用人单位承担」 — must organise rescue and treatment in good time, carry out health checks and medical observation, and bear the cost — (nationally) +- Evidence grade: B +- Sources: China CDC Institute for Occupational Health and Poison Control (中国疾病预防控制中心职业卫生与中毒控制所) (2025). Holiday safety reminder: beware of strongly acid and strongly alkaline cleaning agents. ; Hunan Provincial Department of Emergency Management (湖南省应急管理厅) (2015). Emergency handling and treatment of common chemical poisoning accidents. ; National Health Commission (国家卫生健康委) and three other departments (2024). Classification and catalogue of occupational diseases (in force from 2025-8-1). ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Law on the Prevention and Control of Occupational Diseases, 第三十七条. (reprinted by the Beijing Municipal People's Government) +- Notes: The B is because on the question of how long to rinse, one official document says 15 minutes and another says 20 to 30 minutes, and there is no single national standard figure. Take the chemical's name, its packaging or that safety data sheet (SDS) to hospital, because the doctor needs it to decide on a plan. Do not grab a different acid or alkali to 「中和」 — neutralise — the exposure: mixing the two gives off heat, and there is no way to control how much goes in. The 「中和液」 — neutralising solution — mentioned in the Hunan document means a special rinsing fluid prepared in advance by the workplace, not something you mix up on the spot from whatever is to hand. Brush dry powders such as lime powder and cement powder off first, then rinse with water (some emergency pages say concentrated sulphuric acid should be wiped dry first for the same reason; the guidance differs from place to place, so this book does not say it). For household cleaning agents and swallowed products see item 20. For protection before you start work and for medical checks, see items 10 and 11 in section 19 (the hazard notice before you start, and the protective equipment the employer must supply). + +### 22. Dizzy, nauseated, not sweating or not fully conscious in high heat: move to shade at once, strip and pour water over the person to cool them. Give an unconscious person nothing to drink, and call 120 + +- Cost: No money. +- In plain language: Skin hot and red, speech slurred and mind confused mean heat stroke. From dizziness to coma can take only a few dozen minutes, and the fatality rate is very high. Move the person at once to a cool, ventilated place, take off their clothes, cool them with cool water and a wet towel, and call 120 at the same time. Give an unconscious person not one mouthful of water, because it goes into the airway. +- Benefit: The US National Weather Service: heat stroke is 「严重的医疗急症」 — a serious medical emergency. It shows as a body temperature above 103°F (39.4°C), confusion, slurred speech, and skin that is hot and red. The handling is to call for emergency help at once, move the person somewhere cool, cool them with cool water and a wet cloth, and 「不要给液体」 — give no liquids. Heat exhaustion shows as heavy sweating, weakness, cold damp skin, and a fast weak pulse. The handling is to move the person somewhere cool, loosen their clothing, and give small sips of water; if there is vomiting, if the person worsens, or if it has not eased after 1 hour, get medical help. The Yunnan Provincial CDC (云南省疾控局) advises: move the person at once to a cool, ventilated place, cool them with a cold wet towel, give salt-containing drinks to anyone fully conscious, and call the emergency number at once if the symptoms are severe; severe heat stroke has a 「病死率极高」 — very high fatality rate. China CDC advises: cool first, transport second, and do the cooling within the 「黄金半小时」 — golden half hour — after the person loses consciousness. Where it is possible, healthy young and middle-aged adults with no underlying conditions can be cooled quickly by cold water immersion, while elderly people and those with underlying conditions are sprayed with water and cooled by evaporation. The Wilderness Medical Society's heat illness guidelines, updated in 2024, say cold water immersion is the most important treatment for severe heat illness and the fastest way to bring the temperature down. With no ice to hand, use the coolest water you have, and if immersion is not possible, spray with water and fan: that runs at about half the speed of immersion +- Evidence grade: B +- Sources: US National Weather Service. Heat Related Illnesses. ; Yunnan Provincial CDC (云南省疾控局) (2026). With these symptoms you may have heatstroke, 出现这些症状，你可能是中暑了. ; China CDC. In hot weather, China CDC reminds you to guard against heatstroke: 「如有条件可用冷水浸泡的方法对无基础疾病的青壮年快速降温。老人和有基础疾病的患者可向皮肤喷水通过蒸发降温。降温应遵循"先降温，后转运" 的原则」 — where it is possible, healthy young and middle-aged adults with no underlying conditions can be cooled quickly by cold water immersion. Elderly people and patients with underlying conditions can be sprayed with water and cooled through evaporation. Cooling should follow the principle of "cool first, transport second". ; National Health Commission (国家卫生健康委) (2010). Key points for preventing heatstroke in high temperatures. ; Eifling KP et al (2024). Wilderness Medical Society Clinical Practice Guidelines for the Prevention and Treatment of Heat Illness: 2024 Update. Wilderness & Environmental Medicine 35(1_suppl):112S-127S. (the text site refuses automated fetching, so the specific recommendations were checked against the update summary on the Wilderness Medical Society's own site: 「cold water immersion … is the most important method of treating severe heat illness」「If CWI is unavailable, evaporative cooling is the recommended second-line treatment. … generally half as fast as CWI」) ; Lipman GS et al (2019). Same guidelines, 2019 version. +- Notes: Do not wait to see if it 「缓一缓」 — eases off a little. Heat stroke can go from dizziness to coma in a few dozen minutes. Do not pour water into an unconscious person, because it goes into the airway. Where to put the ice packs, the two sources differ. China CDC says the neck, the armpits and the groin. The Wilderness Medical Society says the research does not show any cooling effect from those sites. Its advice is that where ice packs are all you have, cover the whole body with them, and where there are not enough, put them on the cheeks, the palms of the hands and the soles of the feet. Both say to use cold water immersion where it is possible. + +### 23. Anyone with heatstroke who is conscious: small sips of cool dilute salt water or a sports drink, often and little. Do not gulp, and do not drink alcohol or high-sugar drinks + +- Cost: No money. A sports drink costs a few yuan a bottle. +- In plain language: Only a conscious person drinks. Take small sips often of cool dilute salt water or a sports drink, 500 to 1000 millilitres an hour. Do not gulp, because being sick and taking it into the airway kills faster. Do not drink alcohol or high-sugar drinks, because they make you lose more water. Cool the body first, with cool water on the skin and ice packs on the neck, armpits and groin, and only then replace the fluid. Give an unconscious person or one with slurred speech not one mouthful. +- Benefit: China CDC has published the Guide to Public Prevention and Emergency Handling of Heatstroke (2014 version). It says 「至少每小时喝 2~4 杯凉水（500~1000ml），水温不宜过高，饮水应少量多次」 — drink at least 2 to 4 cups of cool water an hour (500 to 1000ml), not too warm, and in small sips often. It also says 「不要饮用含酒精或大量糖分的饮料。这些饮料会导致失去更多的体液。同时，还应避免饮用过凉的冰冻饮料」 — do not drink drinks containing alcohol or a large amount of sugar, because they cause you to lose more body fluid, and avoid drinks that are too cold and frozen. For heat stroke it says plainly 「不要给病人喝水」 — do not give the patient water. US NIOSH: for heat exhaustion, 「鼓励小口频繁地喝凉水」 — encourage small, frequent sips of cool water; for heat cramps, give a sugary and electrolyte-containing sports drink every 15 to 20 minutes, and 「不要用盐片」 — do not use salt tablets +- Evidence grade: B +- Sources: China CDC Institute for Occupational Health and Poison Control (中国疾病预防控制中心职业卫生与中毒控制所) (2014). Guide to Public Prevention and Emergency Handling of Heatstroke (2014 version). ; US CDC/NIOSH. Heat-Related Illnesses. +- Notes: After heavy sweating, drinking plain water on its own dilutes the salt in the blood (low blood sodium), so the water needs a little salt in it. Drinking a large amount at once makes people vomit, and vomit going into the airway kills faster than the heatstroke itself. "Do not drink it ice cold" is advice from the 2014 version of the China CDC guide, on the grounds of fear of stomach spasm, and it comes with no quantified evidence. Do not misread it as "cold is not allowed". The ways of cooling are set out in item 22 in this section (moving to shade, stripping and pouring water to cool down when dizzy and nauseated in high heat): wiping the body down with cool water, cold water immersion, and ice packs on the neck, armpits and groin. These are exactly the first-line cooling methods the official sources recommend, and they go on the body rather than into the mouth. Cooling always comes before rehydrating. Give an unconscious person or one with slurred speech not one mouthful + +### 24. In a fire: crawl along the ground, feel the door before opening it, and if the door is hot do not open it. Take the stairs, not the lift, and once you are out do not go back + +- Cost: Fit 1 to 2 standalone smoke alarms at home, a few tens of yuan. Look at the two ways out of the stairwell in advance. +- In plain language: Stay down on the floor and crawl towards the exit under the smoke. Before opening a door, feel the handle and the door itself. If it is hot, or if smoke is coming from the gap, do not open that door: take the second way out. Open the door slowly and shut it again the moment thick smoke comes through. Take the stairs and not the lift, because a power cut traps you inside. Once you are out, do not go back for your things. If your clothes catch fire, stop, lie down and roll, and cover your face with your hands. +- Benefit: The US FEMA text on escaping says: 「趴到地上，从烟下面爬向出口」 — stay down on the ground and crawl towards the exit under the smoke. Before opening a door, feel the handle and the door itself: 「如果是热的，或者门缝有烟，别开这扇门，用第二条路」 — if it is hot, or if smoke is coming from the gap, do not open that door; use the second way out. Open the door slowly and close it again as soon as there is thick smoke. If clothes catch fire, 「停下、倒地、打滚」 — stop, lie down, roll — and cover your face with your hands. If you cannot get out, close the door, block the gap under the door and any vents with cloth or tape, and phone in your location. The Ministry of Emergency Management's fire self-rescue guidance says the same thing: do not take the lift when escaping, run downwards as far as you can while the fire is small, cover the nose and mouth with a wet towel or a mask, and crawl or bend low as you leave +- Evidence grade: B +- Sources: US FEMA Ready.gov. Home Fires. ; Ministry of Emergency Management (应急管理部) (2019). How to get yourself out of a fire in a public place: 「逃生时不要乘坐电梯，当火势不大时，要尽量往楼层下面跑」「可用湿毛巾或口罩蒙鼻，匍匐或弯腰撤离」 — when escaping, do not take the lift; when the fire is not severe, run towards the lower floors as far as you can; you can cover the nose with a wet towel or a mask, and crawl or bend low as you leave. ; Guizhou Provincial Fire and Rescue Corps (贵州省消防救援总队) (2020). Everyday fire knowledge: three points for escaping, 消防常识丨逃生三要点: 「湿毛巾对折3次（叠成8层），除烟率比较高，但烟和毒气浓度过高时，过滤效果就无法保证了。另外，毛巾过湿会导致呼吸困难」 — a wet towel folded 3 times (into 8 layers) filters smoke well, but when smoke and toxic gas are too thick to allow any filtering, and a towel that is too wet makes breathing difficult. +- Notes: Do not take the lift, because a power cut traps you inside. Do not go back for property. Do not run upwards unless you have confirmed the roof door is open. Do not jump. Fold a wet towel 3 times into 8 layers and do not put too much water in it, because a towel that is too wet makes breathing harder. When the smoke and toxic gas are too thick, a towel will not save you. + +### 25. You see someone drowning: call for help and dial 110/120, reach in with a pole, throw something that floats, and do not go into the water yourself + +- Cost: No money. The hard part is holding back the urge to jump in. +- In plain language: A drowning person pushes the rescuer under by instinct, so whoever goes in may drown too. The order is: shout for help → call 110/120 → pass over a long pole or a rope → throw a life ring, an empty bucket or a plastic bottle → when the person is out and not breathing, start compressions at once. Do not form a chain of hands and go into the water. About 300,000 people drown worldwide every year, and it is the fourth leading cause of death in children aged 1 to 4. +- Benefit: WHO: about 300,000 people drown worldwide every year, 92% of them in low- and middle-income countries. Drowning is the fourth leading cause of death in children aged 1 to 4 and the third in those aged 5 to 14. The intervention WHO recommends is 「安全救援和复苏培训」 — safe rescue and resuscitation training — which means learning first the way to help without going into the water. The order is: shout for help → make the call → pass over a long pole or a rope and throw a life ring, an empty bucket or a plastic bottle → when the person is out and not breathing, start compressions at once +- Evidence grade: C +- Sources: WHO. Drowning fact sheet, 溺水 实况报道. +- Notes: The C is because no official text could be found for the rule "do not go into the water". There is only WHO's death figures and WHO's direction of "safe rescue". Do not link arms and form a chain into the water. If you have not been trained, do not try to drag someone out, because a drowning person pushes the rescuer under. "Do not go into the water" is also a rule in your own interest: the person who goes in drowns too, whoever it happens to be. The person you are most likely to save is your own child, and item 12 in section 1 (keeping a child in sight near water) covers how to prevent it. + +### 26. Someone is choking and cannot speak: stand behind them and give 5 back blows followed by 5 abdominal thrusts, and if they collapse start cardiopulmonary resuscitation + +- Cost: No money. Look at the movement once: make a fist above the navel, put your other hand over it, and thrust sharply inwards and upwards. +- In plain language: First ask 「你噎住了吗，能说话吗」 — are you choking, can you speak. If they can still cough hard and make a sound, do nothing and let them cough. If they cannot speak, stand behind them and give 5 back blows followed by 5 abdominal thrusts, repeating until the object comes out or the person collapses. If they collapse, start cardiopulmonary resuscitation and send someone to call the emergency number. For a pregnant woman or a very large person, thrust on the chest instead. Do not reach into the mouth for something you cannot see. +- Benefit: The US National Library of Medicine text says: first ask 「你噎住了吗，能说话吗」 — are you choking, can you speak. 「能用力咳嗽、能说话的不要动手」 — if they can cough hard and can speak, do not intervene, because coughing itself usually brings the object up. If they cannot speak, 「反复做 5 次拍背接 5 次腹部冲击，直到异物出来或人失去意识」 — repeat 5 back blows followed by 5 abdominal thrusts until the object comes out or the person loses consciousness. 「失去意识就开始心肺复苏」 — once they lose consciousness, start cardiopulmonary resuscitation — and send someone to call the emergency number. For a pregnant woman or someone with obesity, do chest thrusts instead +- Evidence grade: B +- Sources: US National Library of Medicine, MedlinePlus. Choking – adult or child over 1 year. ; Fuzhou Municipal Health Commission (福州市卫生健康委员会) (2026). The Heimlich manoeuvre has been updated: 「1岁以上幼儿、儿童及成人：从旧版单一腹部冲击改为"5次背部拍击+5次腹部冲击"的循环」「中国红十字系统急救培训一直都是按"5次背部拍击+5次腹部冲击"循环进行的方式进行教学」 — for toddlers, children and adults over 1 year old: the old single abdominal thrust has been replaced by a cycle of 5 back blows plus 5 abdominal thrusts; first aid training in the Chinese Red Cross system has always taught it as a cycle of 5 back blows plus 5 abdominal thrusts. +- Notes: Do not reach into the mouth for something you cannot see, because you push it deeper. Do not give back blows to someone who can still cough. Choking happens most often at your own dining table, so the people who benefit are mostly your family. The teaching in China is the same set of moves: the Fujian Red Cross emergency first aid training centre says that for children over 1 year and for adults, the Chinese Red Cross system has always taught a cycle of 5 back blows plus 5 abdominal thrusts. Many older tutorials online teach only abdominal thrusts, and that is the old version. + +### 27. Lost in the Gobi, in a desert or in an uninhabited area, or the vehicle has broken down: stay where you are or by the vehicle, report your location first, then get shade and warmth, ration the water and do not wander + +- Cost: Before going into an uninhabited area, tell someone your route, your destination, your date of return, and the vehicle type and plate number. Start at 4 litres of water per person per day. A satellite communicator or BeiDou short message device costs over a thousand yuan, and a few hundred yuan to rent. The hard part is holding to "stay put and do not wander". +- In plain language: Stay where you are or by the vehicle, because the vehicle is the biggest thing giving shade. The order is: with a signal, send your location first and then make the call → get shade and keep warm → ration the water, moving nothing by day and moving once it has cooled off at night → do not wander → signal with a mirror, with smoke, or with three piles of fire. Before you go in, tell someone the route, the destination, the date of return and the vehicle type and plate number. +- Benefit: The Mojave desert page of the US National Park Service has three sentences in its original text. One is 「每人每天至少 1 加仑（约 3.8 升）水，徒步和骑行者每人每天 2 加仑」 — at least 1 gallon (about 3.8 litres) of water per person per day, and 2 gallons per person per day for people walking and cycling. One is 「告诉别人你的路线、目的地、返回日期、车型和车牌」 — tell someone your route, destination, date of return, vehicle type and plate number. And one is 「手机信号时有时无」 — mobile signal comes and goes. The Death Valley text: 「热的时候不要在谷底和低海拔徒步」 — when it is hot, do not hike in the valley floor or at low elevations; in the wilderness 「几乎没有手机信号」 — there is almost no mobile signal — and it recommends carrying a satellite phone. The Mojave desert driving page says: 「出了紧急情况，待在车旁是你能做的最重要的事」 — in an emergency, staying by the vehicle is the most important thing you can do. The page also says this is not common, but people have died on the walk back to the road from exposure to extreme heat or cold. The priorities run in five steps. The first is reporting that you are safe and your coordinates: with a signal, send your location first and then make the call. The second is shade and warmth, with the vehicle as the largest shelter. The third is water: move nothing by day, and move once it has cooled off at night. The fourth is not wandering. The fifth is signalling. The Chongqing Municipal Emergency Management Bureau advises that three fires laid out in a triangle are the internationally recognised distress signal. If you are close enough to shout, shout three short, three long, then three short again, and repeat after 1 minute. When there is sun, use a mirror, a tin can lid or something else bright to reflect it. The Ministry of Culture and Tourism reminds self-driving visitors to go into uninhabited areas with care, to travel in as many vehicles as possible, to prepare walkie-talkies in advance, and to rent a satellite phone if necessary +- Evidence grade: C +- Sources: US National Park Service. Mojave National Preserve Safety. ; US National Park Service. Death Valley Safety. ; US National Park Service. Desert Driving Safety. ; Chongqing Municipal Emergency Management Bureau (重庆市应急管理局) (2024). Emergency science: the distress signals you need to know in the wild: 「燃放三堆火焰是国际通行的求救信号，将火堆摆成三角形」「三声短三声长，再三声短；间隔1分钟之后再重复」 — lighting three fires is the internationally recognised distress signal, and lay the fires out in a triangle; three short, three long, then three short; repeat after an interval of 1 minute. ; Ministry of Culture and Tourism (文化和旅游部) (2018). The Ministry of Culture and Tourism reminds visitors to choose outdoor adventure travel carefully: 「谨慎前往无人区和人烟稀少区」「尽量多车同行，提前准备对讲机，必要时租用卫星电话」 — go into uninhabited and thinly populated areas with care; travel in as many vehicles as possible, prepare walkie-talkies in advance, and rent a satellite phone if necessary. +- Notes: The C is because all of this is official guidance rather than research data. The water quantity, the notification, the signal coverage, staying by the vehicle and the distress signals are all official wording. No official text could be found for the "groups of three" calls that circulate, so they have been rewritten as the official three fires and three short, three long, three short. Do not hike at midday looking for the route. Do not drink urine. Do not cut your drinking to save water. + +### 28. Shaking, slurred speech and unsteady walking mean hypothermia: change the wet clothing, wrap them up, warm the trunk first, give a hot sweet drink and no alcohol + +- Cost: An aluminium foil survival blanket costs a few yuan, plus one spare dry garment. Learning to read the signs costs nothing. +- In plain language: A body temperature below 35℃ is hypothermia. The person shakes, speaks unclearly, walks unsteadily and cannot think straight. Change the wet clothing and wrap them in a survival blanket or dry clothing. Warm the trunk first, not the hands and feet. Give a hot sweet drink and no alcohol, because alcohol suppresses the shivering. Stopping the shivering is not improvement, it is deterioration. If they are unconscious, start cardiopulmonary resuscitation. +- Benefit: The US National Weather Service text: a body temperature below 95°F (35°C) is hypothermia, and 「觉得冷、迟钝、想不清楚，立刻看医生」 — if you feel cold, slow and unable to think clearly, see a doctor at once. The handling is 「立即就医，把人移到有暖气的室内，先从身体中心开始复温」 — seek medical care at once, move the person into a heated room, and start rewarming from the centre of the body — and give hot drinks. 「酒精会减少寒战」 — alcohol reduces shivering — so none is given. 「失去意识做心肺复苏」 — if the person is unconscious, do cardiopulmonary resuscitation. Frostbite is treated with lukewarm water or by rewarming against body temperature, and 「不要用热水或壁炉一类辐射热」 — do not use hot water or radiant heat from a fireplace. The Wilderness Medical Society's 2019 hypothermia guideline is the practice guideline for this field +- Evidence grade: B +- Sources: US National Weather Service. During Extreme Cold. ; Dow J et al (2019). Wilderness Medical Society Clinical Practice Guidelines for the Out-of-Hospital Evaluation and Treatment of Accidental Hypothermia: 2019 Update. Wilderness & Environmental Medicine. +- Notes: Do not rub the hands and feet. Do not have a hot bath. Do not drag and haul the person, because a badly hypothermic person can go into cardiac arrest easily. Do not take the absence of shivering as improvement: not shivering any more (the shivering has stopped) means deterioration. The full text of the Wilderness Medical Society guideline is behind a subscription, and only the abstract was checked this time. + +### 29. Bitten by a snake: sit down and stay still, take off rings and watches, get to hospital quickly for antivenom. Do not cut, do not suck, do not tie it tight + +- Cost: No money. Find out in advance which nearby hospital holds antivenom. The hard part is holding back the urge to treat the wound yourself. +- In plain language: Sit down and stay still, and splint the bitten limb. Take off rings and watches first, because it is about to swell. Get to a hospital with antivenom quickly, because antivenom is the only thing that reverses most of the toxic effects. Do not cut, do not suck with your mouth, do not apply ice, and do not tie a rope until it turns purple. Do not drink alcohol and do not run about. Take a photo of the snake from a distance if you can, but do not go and catch it. +- Benefit: The WHO text has three sentences. One is 「立即完全固定被咬的身体部位，并迅速送到最近的医疗机构就医」 — immobilise the bitten part of the body completely at once and take the person to the nearest medical institution for treatment without delay. One is 「避免使用止血带，并避免切割伤口」 — avoid using a tourniquet, and avoid cutting the wound. And one is 「用针对当地蛇种的适当抗蛇毒血清进行治疗」 — treat with the appropriate antivenom for the local snake species. 「高质量抗蛇毒血清是防止或逆转蛇咬伤的大部分毒性作用的最有效治疗方法」 — high-quality antivenom is the most effective treatment for preventing or reversing most of the toxic effects of a snakebite. About 81,000 to 138,000 people die of snakebite worldwide every year, and the number left amputated or permanently disabled is about 3 times the number of deaths. China's 2024 snakebite treatment guideline estimates that 250,000 to 280,000 venomous snakebites occur in China each year, mainly in people aged over 50. Hospitals in 11 provinces of the Yangtze River basin and the south admitted 40817 snakebite patients between 2017 and 2021, with 23 deaths during the hospital stay (0.06%) +- Evidence grade: B +- Sources: WHO. Animal bites fact sheet, 动物咬伤 实况报道. ; WHO (2023). Snakebite envenoming fact sheet, 蛇咬伤中毒 实况报道. ; Lai R et al (2024). The Chinese guideline for management of snakebites. World Journal of Emergency Medicine 15(5):333-355: 「An estimated 250,000–280,000 cases of venomous snakebites occur in China each year, affecting mainly patients aged ≥50 years」. ; Yan S et al (2026). Clinical epidemiology of snakebite envenoming in hospitals 11 provinces of Yangtze River Basin and southern regions of China. PLoS Neglected Tropical Diseases: 「withdrawal (4346, 10.65%), and death (23, 0.06%)」. +- Notes: Take a photo of the snake if you can, but do not go and catch it. Do not drink alcohol, do not run about, do not suck with your mouth, do not apply ice, and do not tie a rope until it turns purple. China has no official national count of snakebite deaths. The in-hospital death rate in those 11 provinces looks low, but more than a tenth of the patients left before treatment was finished, and what became of them is not in the figures. + +### 30. In an earthquake, first work out what building you are in: in a building built to seismic code, drop where you are. In an old building that will come down, the door is a few steps away, so go out + +- Cost: No money. Look in advance at how many steps it is out of your own front door, which way you go, and which table is solid. +- In plain language: First work out what building you are in. In a building put up to code after 2001, if you have to go down the stairs or through the entrance hall to get out, drop where you are: cover your head and neck and get under a solid table, holding the table leg. In a single-storey house, an old brick house, or a self-built house with no seismic design, the door is a few steps away, so go out. In no circumstances jump from a window, take the lift, or go down the stairs while the building is shaking. +- Benefit: The China Earthquake Administration's explainer video Earthquake: hide or run? (《地震来了，是躲还是跑？》) gives the formula 「震时就近躲避，震后迅速撤离」 — during the shaking, take cover nearby; after the shaking, withdraw at once — and it states the limit explicitly: 「这里指的是在符合抗震设防要求的建筑内」 — this refers to being inside a building that meets seismic design requirements. The reason is that such buildings 「整体垮塌的可能性较小」 — are less likely to collapse entirely. The text says that 「坠落的建筑构件及装饰物品」 — falling structural members and fittings — 「才是对生命安全最大的威胁」 — are the greatest threat to life. And that 「震时无保护措施的盲目乱跑」 — running about blindly during the shaking without protection — 「反而容易被坠落的天花板、吊灯（扇）等物品砸伤」 — makes you more likely to be struck by a falling ceiling, a light fitting or a fan. To take cover, 「用手或者其他软物保护好头颈部」 — protect your head and neck with your hands or another soft object — 「并牢牢地抓住桌腿」 — and hold the table leg firmly. With no table, 「迅速贴紧承重墙蹲伏」 — crouch close to a load-bearing wall at once — 「避开外墙、窗户、阳台」 — and keep away from outer walls, windows and balconies. 「如果在户外」 — if you are outdoors — you should 「应该第一时间疏散到空旷的地方避震」 — move to open ground at once to ride out the shaking. Look at it the other way round. The 1988 Armenian earthquake brought down whole Soviet-style prefabricated blocks. The case-control study from that time says that 「第一下震动之后离开建筑是保护性行为」 — leaving the building after the first shock is protective behaviour. The chance of being hurt while staying indoors was 4.40 times the chance for someone who got out (odds ratio 4.40), 95% CI 2.24 to 8.71 (credible range). The chance of being hurt inside a building was 12.20 times that of being outside (3.62 to 63.79). In a building of more than 5 storeys it was 3.65 times the chance in a lower building (2.12 to 6.33) +- Evidence grade: B +- Sources: China Earthquake Administration (produced by the Anhui Provincial Earthquake Administration). Earthquake: hide or run? (explainer video, subtitle transcript), 地震来了，是躲还是跑？. ; Armenian HK, Noji EK, Oganesian AP (1992). A case-control study of injuries arising from the earthquake in Armenia, 1988. Bulletin of the World Health Organization, 70(2), 251-257. ; US FEMA Ready.gov. Earthquakes. ; Mahue-Giangreco M, Mack W, Seligson H, Bourque LB (2001). Risk factors associated with moderate and serious injuries attributable to the 1994 Northridge Earthquake. Annals of Epidemiology, 11(5), 347-357. +- Notes:Disputed, and the two conclusions are about two different kinds of building. The China Earthquake Administration itself makes "take cover" conditional on 「符合抗震设防要求的建筑内」 — being inside a building that meets seismic design requirements. It also calls for 「因地制宜」「因人而异」 — adapting to the place and to the person — so you weigh up whether the building resists earthquakes, where you are standing, and whether you can run at all. The Armenian buildings came down whole, so there was nowhere to take cover and going out was the better option. What US FEMA gives for housing in the United States is "drop, cover your head and neck, hold on, and do not run outside", and the injuries there come mainly from falling objects and from falls while moving during the shaking. The Northridge study also warns that reaching out to catch something, or to hold up something that "looks steady", makes the injury worse. So asking only "do I run" is no use. You have to judge whether the building will come down and whether you can reach open ground within seconds. In a building put up to code after 2001, where you have to go down the stairs or through the entrance hall, drop where you are. In a single-storey house, an old brick house, or a self-built house with no seismic design, the door is a few steps away, so go out. The Armenian study compared only survivors who were hurt with survivors who were not, did not count the dead, and does not say whether the effect of storey height was adjusted for. People who can run may well have lived on the ground floor, so read the figures down. In no circumstances jump from a window, take the lift, or go down the stairs while the building is shaking. Once outside, keep clear of tall buildings, overpasses, electricity poles, glass curtain walls and advertising hoardings. + +### 31. Bears, wild boar and a pack of wild dogs: do not run, do not scream, do not stare them down, and back away slowly side-on + +- Cost: No money. Take a walking pole into the hills, walk in company, and make some noise so the animals know you are coming. The hard part is standing still instead of running. +- In plain language: Do not run, because a bear runs like a racehorse up and down a slope. Do not climb a tree either, because both black bears and brown bears climb. Stand still, keep talking to it in a low voice, and when it stops moving back away slowly side-on. If a brown bear charges, lie down and play dead, with your hands clasped behind your neck and your legs apart. If a black bear charges, do not play dead: run for the car or the house, and if you cannot get there, fight back with whatever is to hand. For a scratch or a bite from a dog, rinse and get the injection as in item 13. +- Benefit: The US National Park Service text on bears has six sentences. One is 「不要跑，熊上下坡都能跑得像赛马一样快」 — do not run, a bear can run as fast as a racehorse up and down a slope. One is 「保持冷静，大多数熊不想攻击你，只想被放着不管」 — stay calm, most bears do not want to attack you, they only want to be left alone. One is 「用低沉的声音持续跟熊说话」 — keep talking to the bear in a low voice. One is 「熊不动时，侧着身慢慢退开，这样既能盯着熊又不会绊倒」 — when the bear is not moving, back away slowly side-on, so that you can keep your eyes on it without tripping. One is 「让自己显得尽量大」 — make yourself look as large as you can. And one is 「不要爬树，灰熊和黑熊都会爬树」 — do not climb a tree, grizzlies and black bears both climb. When attacked by a brown bear, 「装死，趴着双手扣在颈后、双腿分开」 — play dead, lying face down with your hands clasped behind your neck and your legs apart. When attacked by a black bear, 「不要装死，逃向车或房子，逃不了就用手边任何东西反击」 — do not play dead, run for the car or the house, and if you cannot escape, fight back with anything to hand. A dog bite is handled as in item 13 +- Evidence grade: B +- Sources: US National Park Service. Staying Safe Around Bears. ; General Office of the National Disease Control and Prevention Administration (国家疾控局综合司) and General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Working standard for the prevention and handling of rabies exposure (2023 version), 狂犬病暴露预防处置工作规范（2023 年版）. +- Notes: The B covers bears only. The rules for wild boar and wild dogs, namely do not run, back away slowly, and get to high ground or put something between you and the animal, are experience, and count as grade C. The official guidance from the National Forestry and Grassland Administration and from local forestry bodies could not be found this time. Do not use food to lure the animal, do not walk up close for a photograph, and do not take a dog into hills where there is wild boar. Bears in China are mostly black bears and brown bears, so read the play-dead and do-not-play-dead advice separately for the two. + +### 32. Caught in a thunderstorm in the wild: get into a vehicle or building with a hard roof, and keep away from ridges, lone trees, open water and metal + +- Cost: No money. The price is that the trip is interrupted for half an hour or more. The hard part is turning back while the thunder is still far off. +- In plain language: Outdoors with a thunderstorm in the area, there is nowhere safe. Get into a vehicle or building with a hard roof. A convertible, a motorcycle, a tent, a shelter, the foot of a tree and the underside of an overhanging rock do not count. Keep away from open ground, from the top of a ridge, from a lone tall tree, from open water and from iron railings. If a group cannot get out, spread out and stand separately. Leave as soon as you hear thunder, and it is only safe 30 minutes after the last thunderclap. +- Benefit: Outdoors with a thunderstorm in the area, there is nowhere safe. It is only safe 30 minutes after the last rumble of thunder +- Evidence grade: B +- Sources: US National Weather Service. Lightning safety outdoors: 「There is no safe place outside when thunderstorms are in the area」「Avoid open fields, the top of a hill or a ridge top. Stay away from tall, isolated trees or other tall objects」「Stay away from water, wet items, such as ropes, and metal objects, such as fences and poles」; where you really cannot leave, 「If you are in a group, spread out to avoid the current traveling between group members」; resuming the activity, 「Wait 30 minutes until after the last rumble of thunder」. +- Notes: A tent, a shelter, the foot of a tree and the underside of an overhanging rock do not count as cover. A car with a hard roof does; a convertible and a motorcycle do not. Start leaving as soon as you hear thunder, and do not wait for the rain to arrive. + +### 33. Going up to a high plateau: climb slowly by sleeping altitude (how high you sleep that night), come back down if the symptoms get worse, and do not push through + +- Cost: One to two extra days for acclimatisation. The preventive drug is a prescription medicine, so ask a doctor before you leave. The hard part is keeping those days free when the itinerary is tight. +- In plain language: What counts is how high you sleep, not how high you got to during the day. Sleeping above 2450 metres puts you at risk of altitude illness. Above 3000 metres, raise the height you sleep at by no more than 500 metres a night, and allow one extra day of acclimatisation for every 1000 metres of climb. A headache with nausea, unsteady walking, a change in consciousness, or breathlessness even at rest means come back down. Coming down comes before oxygen and before medicine. +- Benefit: Sleeping above 2450 metres puts you at risk of altitude illness. High-altitude cerebral oedema and pulmonary oedema, where fluid collects in the brain and the lungs, kill if you do not come down to lower ground in time +- Evidence grade: A +- Sources: US CDC. Yellow Book, high-altitude travel and altitude illness: 「Any unacclimatized traveler proceeding to a sleeping altitude of ≥2,450 m (≥8,000 ft)—and sometimes lower—is at risk for altitude illness」; rate of ascent, 「500 m (1,650 ft) per night in sleeping altitude once above 3,000 m (9,800 ft)」, with one extra acclimatisation day for every 1000 m of climb; acetazolamide for prevention, 「125 mg twice a day; 250 mg twice a day if >100 kg body weight」; 「If symptoms become worse despite rest or treatment at the same elevation」 descent is required. ; Wilderness Medical Society (2024). Clinical Practice Guidelines for the Prevention, Diagnosis and Treatment of Acute Altitude Illness: 2024 Update. Wilderness & Environmental Medicine. +- Notes: What counts is 「睡在多高」 — how high you sleep — not how high you got to during the day. A headache with nausea, unsteady walking, a change in consciousness, or breathlessness even at rest are all signs to come down. Descent takes priority over oxygen and over medicine, and carrying on upwards while breathing oxygen is wrong. + +### 34. A tick has attached itself: pull it straight out with fine-tipped tweezers held against the skin, and put on no oil and use no flame + +- Cost: A pair of fine-tipped tweezers. +- In plain language: Grip the head of the tick with fine-tipped tweezers held against the skin, and pull it straight out. Do not put on oil, do not use a flame, and do not twist as you pull: that agitates the tick and squeezes fluid carrying the pathogen into the skin, and it also leaves the mouthparts buried in the flesh. Write down the date afterwards. If you get a fever, a rash or weakness in the weeks that follow, tell the doctor you were bitten by a tick. +- Benefit: Oil, flame and twisting as you pull all agitate the tick and squeeze pathogen-bearing fluid into the skin. Doing it this way also makes it easy to leave the part of the tick's mouth that is buried in the flesh inside. The fever with thrombocytopenia syndrome that ticks carry in China had 6602 cases and 306 deaths reported nationwide in 2024, the highest year since the disease was discovered. The average fatality rate from 2010 to 2024 was 4.79%, it rises with age, and in those aged 85 and over it is 9.15% +- Evidence grade: B +- Sources: US CDC. After a tick bite: 「Grasp the tick as close to the skin's surface as possible using clean fine-tipped tweezers」「Pull tick away from the skin with steady, even pressure」「Do not use petroleum jelly, heat, nail polish, or other substances to try and make the tick detach from the skin. This may agitate the tick and force infected fluid from the tick into the skin」. ; Ren Q et al (2026). Epidemiological characteristics of severe fever with thrombocytopenia syndrome (SFTS) in China, 2010–2024: a fifteen-year national surveillance analysis. BMC Infectious Diseases 26:1497: 「The number of reported cases (6,602 cases) and deaths (306 cases) in 2024 were both the highest levels since the discovery of this disease」「The average CFR was 4.79%, and the CFR increased with age, reaching 9.15% among those aged 85 years and above」. +- Notes: Write down the date when you have pulled it out. If you get a fever, a rash or weakness in the weeks that follow, tell the doctor you were bitten by a tick. If the doctor does not ask and you do not say, it is easily treated as a cold. + +### 35. Drink all water in the wild boiled: a rolling boil for one minute is enough, and there is no need to boil it longer because of the altitude + +- Cost: Carrying an extra set of cooking gear and the weight of the fuel. +- In plain language: Every organism that can make you ill through drinking water dies within seconds at boiling temperature. Boiling for another minute is only a margin for judging the boil wrongly, and you do not need to boil it longer because of the altitude. If you cannot boil, use a filter or a disinfection tablet. Each of the two misses some pathogens, so boil the water if you can. Clear-looking stream water and snowmelt can carry parasites just the same. +- Benefit: Every organism that can make you ill through drinking water is killed (inactivated) within seconds at boiling temperature. The belief that water at high altitude does not get hot enough to boil is a common mistake +- Evidence grade: B +- Sources: US CDC. Yellow Book, water disinfection for travellers: 「All organisms that may cause illness from drinking water are killed within seconds at boiling temperature」「CDC recommends that travelers boil water for a full minute to account for user variability in identifying boiling points and to add a margin of safety」「Although the boiling point for water decreases with increasing elevation, at common travel elevations the temperature needed to achieve boiling is still well above the temperature required to inactivate enteric pathogens」. +- Notes: The same page also says you do not strictly have to bring it to a rolling boil: holding it at 60 ℃ for 30 minutes works just as well, and boiling is simply the end point you can see without a thermometer. If you cannot boil, use a filter or a chemical disinfection tablet; each of the two misses some pathogens, so boil the water if you can. Stream water and snowmelt that look clear can carry parasites just the same, and appearance is not a test. + +### 36. Demanded of money by a stranger in an uninhabited area: give him the money, do not use your hands, note his features, and call the police once you are clear + +- Cost: The cash you are carrying. The hard part is keeping your temper down. +- In plain language: Give him the money. Do not use your hands and do not answer back. He is risking a charge of robbery for a few hundred yuan: three years or more and up to ten years to start with, and ten years or more up to the death penalty where he enters a home, carries a gun, or causes grievous injury or death. Once he has the money and turns to leave, if you chase him and hit him it is no longer self-defence. As soon as you are clear, write down his build, his accent, the plate number and the direction he went, and call 110. +- Benefit: Your life is worth more than the money, and he is risking a serious charge over a few hundred yuan. Article two hundred and sixty-three of the Criminal Law says that whoever takes property belonging to others or others by violence, coercion or other means shall be sentenced to three to ten years' imprisonment and fined. Where a firearm is carried, a home is entered, or grievous injury or death is caused, the sentence is ten years or more, up to the death penalty. The right of self-defence in Article twenty of the Criminal Law only protects 「制止正在进行的不法侵害」 — stopping an unlawful infringement that is under way. Once he has the money and turns to leave, if you chase him and hit him it is not self-defence. As soon as you are clear, write down his build, his accent, the plate number and the direction he went, and call 110 +- Evidence grade: C +- Sources: National People's Congress (全国人大) (revised 2020). Criminal Law, 刑法（第二十条、第二百六十三条）. (consolidated text reprinted by the Beijing Municipal Public Security Bureau Traffic Management Bureau) +- Notes: The C is because there is no direct literature on the claim that paying and not fighting is safer. The statute only sets out his risk and your boundary. Do not follow him to a secluded place to 「取钱」 — get the money. Do not hand over your phone password. Do not get into his vehicle. Paying him first applies only where you are stopped face to face and your personal safety is threatened, and you should still call the police once you are clear. Being demanded money online with private photos or nude videos is the opposite case, and you must not pay a single yuan. See item 32 in section 8 (being blackmailed with private material). + +### 37. You come across a group of people fighting: back off and walk away. Do not wade in, do not stand and watch, do not pick up the weapons on the ground. If you are going to call the police, move to a safe distance and dial 110 + +- Cost: No money. The hard part is holding back both impulses: going in to pull them apart, and standing there to watch the show. After you call the police you may be called back to give a statement, which costs at most half a day. +- In plain language: The moment you lay a hand on them, breaking up the fight becomes taking part in it. Taking part in a group fight carries up to 3 years, and where there are many people, on a public road, or carrying weapons, 3 to 10 years. Watching can get you hurt by accident. What you should do is back off and walk away. If you are willing to call the police, move to a safe distance, dial 110, and say where it is, how many people there are, whether anyone has a knife or a stick, and whether anyone is lying still. +- Benefit: Article two hundred and ninety-two of the Criminal Law: 「聚众斗殴的，对首要分子和其他积极参加的，处三年以下有期徒刑、拘役或者管制」 — in a group fight, the ringleaders and the others who take an active part shall be sentenced to up to three years' imprisonment, criminal detention or public surveillance. Where there are repeated group fights, many people, a large scale and a bad effect on society, where public order is seriously disrupted in a public place or on a main road, or where weapons are carried, the sentence is 「处三年以上十年以下有期徒刑」 — three to ten years' imprisonment. Where a group fight 「聚众斗殴，致人重伤、死亡的」 — causes grievous injury or death — it is punished as intentional injury or intentional homicide. The Law on Penalties for Administration of Public Security has penalties too. Assaulting another person carries 5 to 10 days of detention and a fine of 500 to 1000 yuan. A group fight is handled as picking quarrels and provoking trouble, with up to 15 days of detention and a fine of 2000 yuan. Item 5 of the Guiding Opinions on the Proper Application of the System of Self-Defence (关于依法适用正当防卫制度的指导意见) says that an unlawful infringement 「既包括针对本人的不法侵害，也包括危害国家、公共利益或者针对他人的不法侵害」 — includes not only an unlawful infringement directed at yourself, but also one that harms the state, the public interest or other people. Item 9 of those Opinions says that where one side strikes first with clearly excessive force, or where one side strikes first and goes on attacking despite the other side trying to avoid the conflict, the side that hits back should in general be found to have been acting in self-defence +- Evidence grade: B +- Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百三十二、二百三十四、第二百九十二、三百零五、三百一十条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Law on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第三十、五十一条）. ; Supreme People's Court (最高人民法院), Supreme People's Procuratorate (最高人民检察院) and Ministry of Public Security (公安部) (2020). Guiding Opinions on the Proper Application of the System of Self-Defence (法发〔2020〕31 号, items 5 and 9), 关于依法适用正当防卫制度的指导意见. ; National People's Congress (全国人大) (2018). Criminal Procedure Law, 刑事诉讼法（2018 年修订，第六十二、一百一十、一百一十一、一百九十三条）. Full text reprinted by the Ningxia Hui Autonomous Region People's Procuratorate +- Notes: The B is because half of the consequences are set down in hard statute, but the step of "back off, call the police and do not wade in" has no official text behind it, and is worked out from the statute and from how a mutual fight is assessed. 「我是来劝架的」 — I only came to break it up — is not a protection, because the assessment looks at whether you laid hands on anyone and who you hit. If the other person is already down or has run off and you still chase them and hit them, that is not self-defence either. If you see one person surrounded and being beaten with no way to fight back, in law you can step in. Self-defence protects not only yourself but other people as well. Going in with your bare hands, though, most often just adds another casualty. Shouting for help, filming from a distance and calling the police work better. "Do not wade in and do not stand and watch" is nothing but good for you. Calling 110 mostly helps the person on the ground, which is the lowest group of beneficiaries in this book, and whether you spend that time is your own call. Before you weigh it, there are a few things to know. The duty to give evidence exists anyway: Article 62 of the Criminal Procedure Law says that anyone who knows the circumstances of a case has a duty to give evidence, and that duty does not disappear because you did not make the call. Once there is grievous injury or a death, the police pull the CCTV and find the people who were there just the same, and standing and watching is what gets you registered most easily. Where a person reporting a case does not want their name published, the authority handling it 「应当为他保守秘密」 — must keep it secret for them (Article 111). Giving the location, the number of people and whether anyone had a weapon is enough, and you can hang up without staying there. The cases that end in a warning or in detention of up to ten days are the ones where the court has notified you to attend, you cannot give a valid reason for not doing so, and you still refuse to give evidence (Article 193), and a bystander very rarely gets that far. Not calling the police is not itself illegal: the words 「有权利也有义务」 — you have both the right and the duty — in Article 110 of the Criminal Procedure Law carry no penalty, and the Criminal Law has no offence of failing to report what you know as an ordinary person, so walking away is a lawful option. There are only two red lines left after you have walked away. The first is helping to hide someone afterwards or giving false evidence to shelter them (Article three hundred and ten of the Criminal Law: 3 years or less, and 3 to 10 years where the circumstances are serious). The second is giving false evidence on purpose when you are called to testify (Article three hundred and five: 3 years or less, and 3 to 7 years where the circumstances are serious). Staying silent and telling the truth are both fine; making things up is not. For being drawn into a conflict yourself, see item 10 in section 8; for being demanded money by a stranger, see item 36 in this section; and for the money after a rescuer is hurt, see item 39 in this section. + +### 38. You may have been exposed to HIV: get the blocking medicine within 72 hours, and the sooner the better + +- Cost: A 28-day course costs over a thousand yuan, and in most regions you pay for it yourself. You have to go to hospital that same evening. The hard part is admitting on the spot that you may have been infected, and walking out of the door at once. +- In plain language: If you think you may have caught HIV, 72 hours is a hard line, and the medicine is best taken within 2 hours. You take it every day for 28 days. Past 72 hours that route is closed, so do not wait until the next day to decide. +- Benefit: The Beijing Municipal People's Government's explainer on HIV puts it this way: the preventive medicine taken after contact is called post-exposure prophylaxis (PEP), and it should 「尽可能在 2 小时内进行预防性用药，最好在 24 小时内、不超过 72 小时」 — be given preventively as soon as possible, ideally within 2 hours, preferably within 24 hours, and not later than 72 hours. It also has to be 「连续服用 28 天」 — taken for 28 days without a break. Past 72 hours starting it is no longer advised, and that is a hard limit +- Evidence grade: B +- Sources: Beijing Municipal People's Government. Cannot tell before from after? Six questions and six answers on HIV pre-exposure and post-exposure prophylaxis, 事前、事后分不清？六问六答全面了解艾滋病病毒暴露前/暴露后预防. ; China CDC Center for STD and AIDS Prevention and Control (中国疾病预防控制中心性病艾滋病预防控制中心) (2020). Technical guidelines for HIV post-exposure prophylaxis (trial), 艾滋病病毒暴露后预防技术指南（试用）. +- Notes: Where to get it: the designated HIV clinic named by the local CDC, usually under infectious diseases (the sexually transmitted infection and AIDS clinic). Many cities now have 24-hour clinics, and Chongqing's published figures are 29 clinics across 26 districts and counties, of which 14 are open around the clock. Before you go, phone the local CDC to ask which one has the medicine, and do not waste time in the emergency department. What counts as an "exposure" includes sex without protection, a condom that broke, a needlestick, and blood onto a mucous membrane. Whether the medicine is needed is the doctor's call. You have to complete the full 28 days, and missing a few doses makes it clearly less effective. Afterwards, follow the doctor's instructions and come back for checks at 4 to 6 weeks and at 3 months. The blocking medicine is only a remedy. For everyday prevention and testing, see items 30 and 31 in section 1 (use a condom throughout, and get tested if you have had higher-risk sex). + +### 39. You were hurt rescuing someone and put money in: claim against the person who caused it and against medical insurance first, then apply to have the rescue recognised + +- Cost: No money. Applying means handing in documents and waiting for a decision, and making several trips, over a few weeks to a few months. +- In plain language: If rescuing someone has cost you, you do not have to write it off as bad luck. Someone injured while doing a righteous act is treated first and billed afterwards. The medical bill is paid by the person who caused the harm, and where that person has run off or cannot pay, it goes through basic medical insurance. Where there is no person to blame, the person you rescued is legally obliged to give you appropriate compensation. Apply at the same time to have the rescue recognised as a righteous act, and if you were injured, have the work injury insurance benefits applied. +- Benefit: Article one hundred and eighty-three of the Civil Code: 「因保护他人民事权益使自己受到损害的，由侵权人承担民事责任，受益人可以给予适当补偿。没有侵权人、侵权人逃逸或者无力承担民事责任，受害人请求补偿的，受益人应当给予适当补偿」 — where a person suffers damage in order to protect the civil rights and interests of others, the tortfeasor bears the civil liability and the beneficiary may give appropriate compensation. Where there is no tortfeasor, or the tortfeasor has absconded or is unable to bear the civil liability, and the victim requests compensation, the beneficiary shall give appropriate compensation. Guobanfa [2012] No. 39 sets out what happens to someone injured while doing a righteous act. Medical institutions have to set up a fast track and 「坚持『先救治、后收费』的原则」 — follow the principle of treating first and charging afterwards. The wording on the medical bill is 「因紧急救治发生的医疗费用，有加害人或责任人的，由加害人或责任人承担；无加害人或责任人以及加害人或责任人逃逸或者无力承担的，按规定通过基本医疗保障制度解决」 — medical costs arising from emergency treatment are borne by the person who caused the harm or is responsible where there is one; where there is no such person, or that person has absconded or is unable to pay, they are resolved through the basic medical insurance system as prescribed. In a death, 「符合烈士评定条件的，依法评定为烈士」 — where the conditions for being honoured as a martyr are met, the person is honoured as a martyr in accordance with the law; 「属于视同工伤情形的，享受一次性工亡补助金以及相当于本人 40 个月工资的遗属特别补助金」 — where the case is treated as a work injury, the family receives the one-off work-related death benefit plus a special benefit to the family equivalent to 40 months of the person's wages. In a disability, 「符合享受工伤保险待遇条件的，依据《工伤保险条例》落实相应待遇；不符合的，由民政部门评定伤残等级并落实相应待遇」 — where the conditions for work injury insurance benefits are met, the corresponding benefits are applied under the Regulations on Work-Related Insurance; where they are not, the civil affairs department assesses the level of disability and applies the corresponding benefits. On work, 「优先纳入就业援助，予以重点支持」 — priority is given in employment assistance, with focused support — and public welfare posts 「优先安排符合条件的见义勇为人员」 — give priority to people who qualify for recognition as having acted righteously. On housing, 「优先纳入住房保障体系，优先配租、配售保障性住房或发放住房租赁补贴」 — priority is given in the housing security system, with priority for rented or sold affordable housing, or a housing rental subsidy +- Evidence grade: B +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一百八十三条）. ; General Office of the State Council forwarding the Ministry of Civil Affairs and other departments (国务院办公厅转发民政部等部门) (2012). Opinions on strengthening the protection of the rights and interests of people who act righteously (国办发〔2012〕39 号). +- Notes: The B is because what you can get is fixed by the national documents, but how much money there is and how the recognition is decided are set by each province's own regulation on righteous acts, the figures differ from place to place, and this round did not check them province by province. You normally apply to the public security organ of the place where the incident happened, or to the body that handles righteous acts (a foundation). Recognition turns on evidence: the receipt from the police report, the CCTV, the names and telephone numbers of witnesses, and the hospital records, and you keep all of it there and then. If you have to sue, the civil rights of someone who acted righteously already fall within the scope of free legal aid. For the national standard of the one-off work-related death benefit and for how a level of disability converts into money, see items 15 and 16 in section 19. For how to apply for legal aid, see item 3 in section 7. + +### 40. A knife, a length of steel bar or a shard of glass has gone into the body: press around the object to control the bleeding, and do not pull it out + +- Cost: No money. The hard part is holding back the instinct to pull it out first and staying with the person until the ambulance comes. +- In plain language: Pulling it out can cause heavy bleeding, and it makes it harder for the doctor to judge how deep the wound goes. Here is what to do: call 120, press around the object with your hand or with a dressing to control the bleeding, and pad the object so it does not move. If organs are showing through a wound in the belly, do not push them back, cover them with a clean moist dressing. If a wound in the chest is bubbling, do not seal it with plastic sheet, because sealing it can collapse the lung. +- Benefit: The International Federation of Red Cross and Red Crescent Societies' International First Aid Guidelines 2025 say: 「急救的人应当压住扎进身体的异物（比如刀）周围，把异物稳住，不要把它拔出来」 — the person giving first aid should press around the object embedded in the body, such as a knife, and hold the object steady without pulling it out. 「不要从腹部或胸部的伤口里拔出任何扎入的异物，拔出来可能造成大出血，也会让医生更难判断伤得有多重」 — do not pull any embedded object out of a wound in the abdomen or chest, because pulling it out can cause heavy bleeding and makes it harder for the doctor to judge how serious the injury is. 「扎在身体里的异物应当稳住；异物随脉搏跳动的，就让它跳，松松地稳住即可」 — an object embedded in the body should be held steady; if it moves with the pulse, let it move, and steady it loosely. 「急救的人不应把露出来的内脏塞回体内」 — the person giving first aid should not push exposed organs back inside. An open chest-wall wound that is not bleeding outwards 「可以敞着、不加敷料，因为把伤口封死可能让空气积在胸腔里，使肺部分或完全塌陷（气胸）」 — can be left open with no dressing, because sealing the wound can let air collect in the chest cavity and partly or completely collapse the lung, which is a pneumothorax. Every penetrating wound of the chest or abdomen needs emergency help called at once +- Evidence grade: B +- Sources: International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, pages 256–257 and 267–268. +- Notes: The B is because this is common practice in the guidelines, with no randomised trial behind it, and you cannot divide the injured into a group you pull from and a group you do not. The only exception is when the object is blocking the airway and the person cannot breathe. At the hospital, go straight to the desk by the emergency department door that sorts cases by severity (the triage desk) and say 「穿透伤」 — penetrating injury. That is the level that goes straight into the resuscitation room, so do not queue at the registration window (see item 8 in section 24). If something has really gone through, call 120, and the ambulance will warn the hospital to prepare. For pressing a heavy bleed and for putting on a tourniquet, see item 12. + +### 41. If you suspect a fracture, do not move that part again: support it and keep it still, and apply cold for no more than 20 minutes. Do not straighten it yourself + +- Cost: No money. Find a board, a rolled-up magazine or clothing to use as support. Put a layer of cloth between the cold thing and the skin. +- In plain language: Treat any injury to a limb as a possible fracture to begin with. Being able to stand and being able to move do not rule it out, and only an X-ray settles it. Keep the injured part still, holding it with your hand, or support it where it already is with a triangular bandage or a board. Apply cold as soon as you can, for no more than 20 minutes at a time. Do not straighten it yourself. Where broken bone has come through the skin, control the bleeding and cover it with a clean dressing, and do not push the bone back in. +- Benefit: The International Federation of Red Cross and Red Crescent Societies published the International First Aid Guidelines 2025. It says that 「手脚上的任何一处伤，都应当先按可能的骨折来处理」 — any injury anywhere on a limb should first be treated as a possible fracture. It also says 「确诊骨折唯一准确的办法是拍 X 光。拿不准就当骨折办」 — the only accurate way to confirm a fracture is an X-ray, and where you are not sure, treat it as a fracture. The purpose of immobilising it is 「尽量减少活动和疼痛」 — to reduce movement and pain as far as possible. Applying cold, 「一次最多连续 20 分钟，这样可以减轻疼痛；冷得比这更久可能损伤皮肤」 — for no more than 20 minutes at a time, which eases the pain; staying cold for longer than that can damage the skin. 「伤到股骨这类长骨时，光是骨头本身出的血就可能致命」 — when a long bone such as the femur is injured, the bleeding from the bone alone can be fatal. Emergency help should be called where the pain or swelling is marked, where there are signs of shock, where there is a fracture of the femur (thigh bone), or where the limb is clearly in the wrong position. For a fracture that is bent out of line, both of two conditions have to hold before you consider straightening it on the spot. The first is that the place is remote or the facilities are limited, and the limb below the bend is cold and pale, or the person giving first aid cannot immobilise it well enough to move it. The second is that the rescuer is trained. For an open fracture, where bone has come through the skin, the priorities are controlling the bleeding and preventing infection +- Evidence grade: B +- Sources: International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, pages 301–302. +- Notes: The B is because all of this is common practice. Traction splinting of a femur (thigh bone) fracture before hospital rests on a single study that went back through old records, and the reliability of that evidence is very low. The result was less blood transfused within 24 hours and a shorter stay, and the difference in deaths and complications is too small to make out. As for whether to reduce the fracture first, putting the bone back where it belongs, or to immobilise it as it is, no evidence was found at all. Three common mistakes need fixing: being able to stand and walk does not mean there is no fracture; use only cold and not heat, and do not alternate between them; and a sprain does not get better by "working it out", so rest first. For the rule on not moving an elderly person when a fracture is suspected after a fall, see item 2. For a swollen leg after a plaster cast or a period in bed, watch for a clot and see item 11. For whether to have a disability assessment after treatment, and whether to apply for a disability certificate, see items 10 and 11 in section 24. + +### 42. After a sexual assault: get somewhere safe and call 110. Before the examination do not wash, do not wash your clothes and do not tidy the room, and get to hospital within 72 hours + +- Cost: Calling the police is free. Emergency contraception is an over-the-counter medicine and any pharmacy sells it. A course of the blocking medicine costs over a thousand yuan, and whether you need it is the doctor's call. See item 38 in this section (getting the blocking medicine within 72 hours). The hard part is that at the time it is very hard to say anything, and you very much want a shower straight away. +- In plain language: Traces left on your body, on your clothing and in the room are the most important evidence for identifying the person who did it. So before the examination try not to wash, not to wash your clothes and not to tidy the room, and call 110 as early as you can. Emergency contraception is in the pharmacy, and the earlier you take it the better it works. Get to hospital within 72 hours and let the doctor decide whether you need the blocking medicine. +- Benefit: A prosecutor at the Beijing Municipal People's Procuratorate says: 「无论是未成年人还是成年人，遇到侵害，都应当第一时间拨打110报警」 — whether the person is a minor or an adult, anyone who has been harmed should call 110 to report it at the very first opportunity. Reporting and seeking medical help in good time 「便于侦查机关及时勘查现场、提取物证痕迹、固定证据」 — lets the investigators inspect the scene, collect physical traces and secure the evidence in time. The People's Daily handbook on preventing sexual assault of minors, reprinted by the Guizhou Provincial Department of Civil Affairs, says: 「一旦不幸遭遇性侵害，最重要的事情就是留存证据」 — once you have been sexually assaulted, the most important thing is to preserve the evidence. You have to keep 「体液、毛发、皮屑、通信记录等证据」 — evidence such as bodily fluids, hair, skin flakes and call and message records — and 「被侵害后，不要擅自洗澡、洗衣物、收拾房间、扔东西」 — after being assaulted, do not wash, wash your clothes, tidy the room or throw things away on your own. Where the person is someone you know, keep the calls and the message records. Where they threaten you or try to buy you off afterwards, keep that evidence too. The Beijing Xicheng District People's Procuratorate writes: 「在接受身体检查之前尽量不要洗澡」 — try not to wash before the physical examination. The World Health Organization says that emergency contraception can be used when 「女性在未采取有效的避孕保护措施时遭受性侵犯」 — a woman is sexually assaulted without effective contraceptive protection. It advises using it within 5 days, and the earlier the better. For emergency contraceptives containing levonorgestrel, a progestogen, the pregnancy rate among women who used them was 1.2% to 2.1%. The levonorgestrel tablets sold in pharmacies in China are over-the-counter. The Wuhan Municipal Health Commission says that after unprotected sex, 「72小时以内，服药越早，预防妊娠效果越好」 — within 72 hours, the earlier the medicine is taken, the better it works at preventing pregnancy. The HIV blocking medicine should be taken as soon as possible, ideally within 2 hours, and no later than 72 hours (nationally) +- Evidence grade: B +- Sources: Beijing Municipal People's Procuratorate (2024). Through the law and through drama: watching the series 九部的检察官, The Ninth Inspector, how should minors facing unlawful infringement protect themselves? (reprinted from China National Radio's China Voice), 法眼看剧丨看《九部的检察官》 未成年人面对不法侵害该如何保护自己？（转载中央广播电视总台中国之声）. ; Guizhou Provincial Department of Civil Affairs (2024). A handbook on preventing sexual assault of minors, for parents and children to read (reprinted from the People's Daily), 未成年人防性侵手册，请家长孩子查收！（转载人民日报）. ; Beijing Xicheng District People's Procuratorate (2018). Why they no longer stay silent, 她们为什么不再沉默……. ; World Health Organization (2021). Emergency contraception (fact sheet), 紧急避孕（实况报道）. ; Wuhan Municipal Health Commission (2020). Safe and sensible use of emergency contraceptives, 安全合理使用紧急避孕药. ; Beijing Municipal People's Government. Cannot tell before from after? Six questions and six answers on HIV pre-exposure and post-exposure prophylaxis, 事前、事后分不清？六问六答全面了解艾滋病病毒暴露前/暴露后预防. +- Notes: The B is because these are official practice and prescribing advice, and there is no figure for how much higher the conviction rate becomes after following it. The people who benefit are mostly you and your family. The 72 hours for emergency contraception is the figure on the Chinese package insert; the World Health Organization says it works within 5 days, but the later you take it the worse it does, so do not delay. If you are already pregnant, emergency contraception does nothing. Tell the truth when you report it, hiding nothing and exaggerating nothing. If the person tries to settle it privately afterwards, offers you money or threatens you, keep a record of it and do not delete it. You can also think it over before reporting it, but the evidence disappears with time. Whether or not you report it, get to hospital inside the time limit. For recording, see item 41 in section 8 (recording). Where the person afterwards uses photographs, or the police report, to pressure you, that is a crime as well. See item 32 in section 8 (being blackmailed with a police report or photographs). If this happens to a friend, go with them to the police and to hospital, and do not start by telling them to take a shower and lie down. diff --git "a/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" "b/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" new file mode 100644 index 00000000..3b9f7205 --- /dev/null +++ "b/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" @@ -0,0 +1,92 @@ +[← 回总目录](../README.md) + +# 14. 账号与信息安全 + +口径：金钱与个人信息。账号被别人登进去，丢的先是钱。对方还会用你的账号去骗你通讯录里的人。你的身份也等于被人拿走了。 + +[← Back to the contents](../README.md) + +# 14. Accounts and information security + +This section measures money and personal information. The moment somebody signs into your account, the first thing you lose is money. They will also use your account to scam the people in your address book. Your identity has been taken as well. + +### 1. Turn on two-factor verification for your email, payment and social accounts, prefer the prompt on your phone, and fall back on a text code + +- Cost: No money. You set it up once per account, and it takes two or three minutes. +- In plain language: Two-factor verification means that at sign-in you confirm a second time, on top of the password, that you are the person. The method where a prompt appears on your phone and you tap to confirm stops more than 90% of phishing takeovers. The old method of answering questions such as 「where did you last sign in」 or 「what is your backup email」 stops only about 10%. +- Benefit: Google counted 350,000 real takeover attempts, also called account hijacking. One class of challenge verifies through a device, such as a prompt on your phone that you tap to confirm, or plugging in a security key. This class stopped 「more than 94% of the hijacking attempts that originate from phishing and 100% of the automated hijacking attempts」. Phishing means tricking you into typing your password on a fake website, and automated means a machine trying leaked passwords in bulk. The other class verifies by asking you questions, such as where you last signed in and what your backup email is. That class 「stopped as little as 10% of phishing hijacking and 73% of automated hijacking」 +- Evidence grade: A +- Sources: Doerfler P, Thomas K, Marincenko M, et al. (2019). Evaluating Login Challenges as a Defense Against Account Takeover. The World Wide Web Conference (WWW '19). +- Notes: The same study also found that this kind of challenge sometimes locks the real owner out. 52% of genuine users did not get in on the first attempt, but 97% of them did get in eventually. Start with your email, because most other accounts can be recovered through email + +### 2. Give your email a password of its own, and do not reuse it on any other site + +- Cost: No money. A password manager stores it so that you do not have to remember it. The hard part is dropping the habit of using one password everywhere. +- In plain language: A password stolen from another site is used to sign straight into your email. Once somebody is in your email, every account you can recover with email goes with it. So the email password has to be its own, and you use it nowhere else. +- Benefit: Taking leaked accounts and passwords from elsewhere and trying them one by one is called credential stuffing. It is the cheapest attack there is, and your email is tried this way too. Once somebody is in your email, every account you recover with email goes with it. The US Cybersecurity and Infrastructure Security Agency recommends a different strong password for every account, at least 16 characters long, kept in a password manager +- Evidence grade: C +- Sources: US CISA. Use Strong Passwords. +- Notes: If you cannot remember them, use the password manager built into your browser. It keeps the password for each site for you, which is far better than one password everywhere. Do not store passwords in your WeChat favourites or in a notes app + +### 3. Set a lock-screen password on the phone and a PIN on the SIM card + +- Cost: No money. One setting for the lock-screen password and one for the PIN. +- In plain language: The SIM card is the small card inside the phone, and it is what receives your text verification codes. If you lose the phone, whoever finds it pulls that card out, slots it into another phone to receive the codes, then resets your accounts one by one. With a PIN on the card, moving it to another phone means a password is needed at switch-on, and that route is closed. +- Benefit: After you lose your phone, the fastest route for whoever finds it is to put the SIM card into another phone and receive your text codes there. With the code, they reset your accounts one by one. A PIN on the SIM card means a password is needed before that other phone will start, so the finder cannot use it to receive codes +- Evidence grade: C +- Sources: the author's own experience, no direct literature +- Notes: You set the PIN under 「SIM card lock」 in the phone's settings. The factory code is usually 1234 or 0000. Three wrong attempts in a row and you need the carrier's PUK code to open it. So write that code on paper as soon as you have set the PIN + +### 4. If you lose your phone, work in this order: report the SIM card, lock it remotely, change your passwords, call the police, freeze the bank card + +- Cost: No money. The whole sequence takes a bit over ten minutes. +- In plain language: The order matters more than how fast you are. First report the SIM card, and the lifeline of your verification codes is cut. Then lock the phone remotely, change your email and payment passwords from a computer, report it to the police and get a receipt, and freeze the bank card if it is needed. If the phone is not with you, borrow somebody else's and call the carrier to report the card. +- Benefit: The order matters more than the speed. First, report the SIM card, and nobody can receive your codes any more. Second, lock the phone remotely and wipe its contents. Third, change your email and payment passwords from a computer. Fourth, report it to the police and get a receipt. Last, freeze the bank card as needed. The US Federal Communications Commission says the same: lock it remotely even when you think you have only lost it. If it was stolen, report it at once, give the model and the IMEI serial number, which is the phone's identity number, and tell the carrier immediately +- Evidence grade: C +- Sources: US FCC. Protect Your Smart Device. ; the order of the steps is the author's own experience ; for replacing an identity card see section 7, and for a loan taken out in your name see the entry on credit records in section 8 +- Notes: Write down the customer service numbers of the three carriers in advance: China Mobile 10086, China Unicom 10010, China Telecom 10000. Also note which city your own number was registered in, because the operator will ask. You can call the operator from somebody else's phone and report the card there too + +### 5. If your card is used by someone else, report and freeze it first, then call the police, then claim from the bank: proving 「that you made the payment」 is the bank's duty + +- Cost: No money. The moment you see anything odd on the card, report it or freeze it at once. Keep the police report, the record of the card report, and every transaction notice the bank sends you. If the card is still with you, make a small enquiry or a small deposit nearby, so you hold a record showing the real card was in your hands at the time. The hard part is holding back the urge to argue with customer service first. Report and freeze first. +- In plain language: You do not have to prove 「this was not me」 when somebody uses your card. The other way round, the bank has to produce evidence that you made the payment, and if it cannot, it pays you. The condition is that you report and freeze at once. If you put it off, the extra loss is yours to carry. +- Benefit: The Supreme People's Court regulation separates out 「who has to produce the evidence」. If you say this is a counterfeit-card or online theft, you have to produce evidence first. A counterfeit-card theft means a copy of the card was made and used. Evidence you can use includes legal documents already in force, where the real card was at the time of the transaction, and where the transaction took place. There are also account transaction records, transaction notices, police reports, records of the card report, and so on. **In the other direction, where the card-issuing bank or a non-bank payment institution (third-party payment) says the payment was made by the cardholder or with the cardholder's authorisation, it is for them to produce the evidence.** After you tell the bank, if the bank does not verify in time, or does not supply and keep the transaction documents and the surveillance video in time, so that the evidence cannot be obtained, the consequence of failing to produce evidence falls on the bank. Once the finding stands, a debit card (savings card) holder can require the issuing bank to pay over the principal and interest on the amount taken by the fraud, and to compensate the loss. A credit card holder can require the overdrawn amount and the penalty charge to be returned, and can claim compensation; where the bank instead asks you to repay the overdraft, the court will not support it. You can also require the issuing bank to remove the corresponding bad credit record in good time (nationwide, in force from 2021-5-25) +- Evidence grade: A +- Sources: Supreme People's Court (最高人民法院) (2021). Provisions on Several Issues Concerning the Trial of Civil Disputes over Bank Cards, 关于审理银行卡民事纠纷案件若干问题的规定（第四、五、七、十四、十五条）. +- Notes: Two situations leave you carrying the liability. The first is that you did not look after the card, the password and the verification codes properly and were at fault yourself. The original wording is 「未尽妥善保管义务具有过错」, which means you failed to keep them properly and were at fault, and you carry it in proportion to your fault. So do not tell other people your password, and do not forward your verification codes to them (see item 1 in this section, which prefers a prompt on your phone for two-factor verification). The second is failing to report the card in time and letting the loss carry on growing, in which case the extra part is yours. So the first step is always to report and freeze, not to argue with customer service first. The same rules apply to non-bank payment institutions. If their promotional material promises 「先行赔付」, payment before recovery, in specific and definite terms, you can hold them to it and ask them to pay first. If you were tricked into transferring the money yourself, that is a different route: see item 2 in section 8 (if you spot the fraud, call 110 or 96110 at once and ask for the payment to be stopped). + +### 6. Now and then look at the signed-in devices and authorised apps on your accounts, and clear out the ones you no longer use + +- Cost: No money. Each time takes a few minutes. The hard part is that nobody reminds you, so you have to remember to look. +- In plain language: 「Signed-in devices」 means the phones and computers that can still use your account. Somebody taking over an account often waits a while before acting. When an unfamiliar device appears in the list, or software you stopped using long ago is still connected, sign out of everything and change your password. +- Benefit: An account takeover often does not happen on the spot, because the attacker waits a while first. The signed-in device list on an account records the phones and computers that can still use that account. The authorised apps list records the other software you allowed to sign in with it. An unfamiliar device in the list, and third-party software you stopped using long ago, are the easiest traces to spot +- Evidence grade: C +- Sources: the author's own experience, no direct literature +- Notes: WeChat, Alipay, your email, your Apple account and your Android account all have this option. If you find a device you do not recognise, choose sign out of all devices and then change your password + +### 7. Do not tap 「agree to all」 to get an app working: if the information is not necessary, refusing it does not let them refuse you the service + +- Cost: No money. The hard part is holding back from tapping 「agree to all」. +- In plain language: When an app asks for your information, if it is not necessary for providing the service, it cannot refuse you the service because you decline. What it may collect is also limited to what it actually needs. A map needs your location, and a torch does not need your contacts. +- Benefit: The law sets out two rules. First, it may not refuse to provide a product or service on the ground that a person declines, or withdraws consent, except where processing the information is necessary to provide the service. Second, collection must stay within the minimum scope needed to achieve the purpose of the processing, so it may only take what it actually needs +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Personal Information Protection Law, 个人信息保护法. China NPC Net. : Article six 「收集个人信息，应当限于实现处理目的的最小范围，不得过度收集个人信息」 ; Article sixteen 「个人信息处理者不得以个人不同意处理其个人信息或者撤回同意为由，拒绝提供产品或者服务；处理个人信息属于提供产品或者服务所必需的除外」 ; Article fifteen 「基于个人同意处理个人信息的，个人有权撤回其同意。个人信息处理者应当提供便捷的撤回同意的方式」 +- Notes: The test is whether the information is necessary to provide that service. A map needs your location; a torch does not need your contacts. After installing an app, go first to the app permissions page in your phone's settings and switch off the permissions you do not need. When you do need one, choose allow just this once. + +### 8. You have the right to see, copy, correct and delete your own personal information, and you can sue if you are refused + +- Cost: No money. You only need to complain or sue if the other side drags its feet. A real court case runs from several months, and you pay the legal fees yourself, so complaining first is better value. The hard part is chasing them repeatedly while they stall. +- In plain language: You have the right to require a company to let you see, copy, correct and delete your own information. Where the service has stopped, the agreed retention period has run out, or you withdraw consent, the company should delete it by itself. If it refuses you, it has to give its reasons, and if it still will not act you can take it straight to court. Closing an account and deleting information are two different things, and after closing the account you still have to ask separately for the deletion. +- Benefit: There are several situations in which a company has to delete on its own initiative: the service has stopped, the agreed retention period has run out, you have withdrawn consent, the original purpose of the collection has been achieved, and so on. If it does not delete, you can require it to. If it refuses you the exercise of these rights, it has to give its reasons. You can sue in the people's court +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Personal Information Protection Law, 个人信息保护法. China NPC Net. : Article forty-five 「个人有权向个人信息处理者查阅、复制其个人信息……个人请求查阅、复制其个人信息的，个人信息处理者应当及时提供」 ; Article forty-six, the right to correct and supplement ; Article forty-seven, which lists five situations in which deletion is required, including 「（一）处理目的已实现、无法实现或者为实现处理目的不再必要」「（二）个人信息处理者停止提供产品或者服务，或者保存期限已届满」「（三）个人撤回同意」, and then 「个人信息处理者未删除的，个人有权请求删除」 ; Article fifty 「个人信息处理者应当建立便捷的个人行使权利的申请受理和处理机制。拒绝个人行使权利的请求的，应当说明理由」「个人可以依法向人民法院提起诉讼」 +- Notes: Closing an account and deleting personal information are two different things, and after closing the account you still have to ask for the deletion separately. Before you change phones or sell the old one, sign out of every account on it and remove the bindings, and only then restore it to factory settings. What the law gives you is a right to delete afterwards, and it cannot take back what has already leaked out. + +### 9. Face recognition is not something you have to agree to: if another way exists they cannot make it the only way, and if you refuse they must offer you another + +- Cost: No money. When you are asked for a face scan, ask whether there is another way to verify. If they say there is not, require them to provide one. The hard part is asking in person. +- In plain language: As long as another way can get the same thing done, the other side cannot make a face scan the only option. If you refuse the scan, it has to give you another method such as a card, a password or an identity card, and it may not put pressure on you with 「then we cannot serve you」. In hotel rooms, public baths, changing rooms and toilets, nobody is allowed to install face recognition equipment. +- Benefit: The Measures for the Security of the Application of Face Recognition Technology (人脸识别技术应用安全管理办法) state: 「实现相同目的或者达到同等业务要求，存在其他非人脸识别技术方式的，不得将人脸识别技术作为唯一验证方式。个人不同意通过人脸信息进行身份验证的，应当提供其他合理、便捷的方式。」 — where the same purpose or the same business requirement can be met by a method other than face recognition, face recognition may not be the only form of verification, and a person who does not agree to identity verification through facial information must be given another reasonable and convenient method. The measures also state: 「任何组织和个人不得以办理业务、提升服务质量等为由，误导、欺诈、胁迫个人接受人脸识别技术验证个人身份。」 — no organisation or individual may mislead, deceive or coerce a person into accepting face recognition as identity verification on the pretext of handling a transaction or improving service quality. Processing facial information on the basis of your consent requires 「在充分知情的前提下自愿、明确作出的单独同意」, a separate consent given freely and expressly in full knowledge: you have to be asked about this one thing on its own, and only your separate nod counts. You have the right to withdraw consent, and the processor has to give you a convenient way to do it. Processing the facial information of a child under fourteen requires the consent of the parents or other guardians. A face recognition device installed in a public place 「应当为维护公共安全所必需」, it must be necessary for maintaining public safety, and it has to carry a clear sign. Inside the private spaces of public places such as hotel rooms, public baths, public changing rooms and public toilets, no organisation or individual may install one. Facial information is to be stored inside the face recognition device and may not be transmitted externally over the internet. There are two exceptions: where laws or administrative regulations provide otherwise, or where separate consent has been obtained (nationwide, in force from 2025-6-1) +- Evidence grade: A +- Sources: Cyberspace Administration of China (国家互联网信息办公室) and Ministry of Public Security (公安部) (2025). Measures for the Security of the Application of Face Recognition Technology, 人脸识别技术应用安全管理办法（第 19 号令，第十条、十二条、十三条，2025 年 6 月 1 日起施行）. +- Notes: What you meet most often is a housing estate gate, a letting platform, a gym or a hotel asking you to record your face. When they say 「the system only supports face recognition」, read the words of the measures out to them. The words are: 「实现相同目的或者达到同等业务要求，存在其他非人脸识别技术方式的，不得将人脸识别技术作为唯一验证方式」 — where the same purpose or the same business requirement can be met by a method other than face recognition, it may not be the only form of verification. Then require them to give you another method such as a card, a password or an identity card. If they still will not, report it to the local cyberspace administration. Where the state has separate rules on face recognition for identity checks, such as in some financial and government settings, follow those rules. The biggest difference between a face and a password is that a leaked face cannot be changed, so it deserves more caution than a password does. A unit that stores facial information on more than 100,000 people has to file with the provincial or higher cyberspace administration within 30 working days, and asking about that is also a way to judge whether they are legitimate. For the right to see, correct and delete your own personal information, see item 8 in this section. diff --git "a/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" "b/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" new file mode 100644 index 00000000..cb5b4ffa --- /dev/null +++ "b/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" @@ -0,0 +1,92 @@ +[← 回总目录](../README.md) + +# 15. 租房与买房 + +口径：金钱。租房最容易亏钱的地方是押金和中介。买房最费钱的地方是贷款利息，还有卖方知道、你不知道的那些事。 + +[← Back to the contents](../README.md) + +# 15. Renting and buying a home + +This section measures money. The places where renting loses you the most are the deposit and the agent. The places where buying a home costs the most are the loan interest, and the things the seller knows about and you do not. + +### 1. The amount of the deposit, when it is returned, and the cases in which it can be reduced must go into the contract + +- Cost: No money. It costs another ten minutes when you sign. The hard part is raising it in front of the landlord +- In plain language: How much the deposit is, when it comes back, and what may be taken off it all have to be in the contract before you sign. A reason for deduction that is not in the contract cannot be used against your deposit when you move out. +- Benefit: The administrative regulation is explicit: 「出租人收取押金的，应当在住房租赁合同中约定押金的数额、返还时间以及扣减押金的情形等事项。除住房租赁合同约定的情形以外，出租人无正当理由不得扣减押金。」 — where a landlord takes a deposit, the lease contract has to set out the amount, the time of return, and the cases in which the deposit may be reduced, and apart from the cases the contract provides for, the landlord may not reduce the deposit without a proper reason. Three things go into the contract: how much deposit is taken, when it is returned, and what it may be reduced for. A reason that is not in the contract cannot be used to take your money +- Evidence grade: A +- Sources: State Council (国务院) (2025). Regulations on Housing Leasing, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十条）. +- Notes: On the day you move out, take photographs and video together with the landlord. Record the readings on the water, electricity and gas meters, and photograph the walls and the floor as well. If the deposit is reduced without reason, complain first to the local housing leasing administration. If the amount is small, go through small-claims procedure: see section 8 + +### 2. If the water or power is cut, the locks are changed, or you are threatened at the door, call the police and keep the record: the rules ban these ways of making you move out + +- Cost: No money. One telephone call to the police, which you can make on the spot +- In plain language: A landlord may not cut the water or the power, change the locks, or threaten you at the door in order to force you to end the contract or move out, and the rules ban it outright. If it happens, call the police first. The police attendance record is your evidence when you later claim for the loss. Do not confront them yourself. +- Benefit: The administrative regulation is explicit: 「出租人不得采取暴力、威胁或者其他非法方式迫使承租人解除住房租赁合同或者腾退租赁住房。」 — a landlord may not use violence, threats or any other illegal means to force a tenant to terminate the lease or move out of the rented housing. Moving the tenant out is what 腾退 means. When this happens, call the police first and keep the evidence. Do not confront them yourself +- Evidence grade: A +- Sources: State Council (国务院) (2025). Regulations on Housing Leasing, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十二条）. +- Notes: You call the police so that there is an attendance record, and it serves as evidence when you later claim for the loss. Notify the local housing leasing administration at the same time + +### 3. An agent may not collect and pay the rent and deposit for you: the money goes straight to the landlord + +- Cost: No money. Check the payee before you pay, which takes a few minutes. The hard part is standing firm when the agent says you should transfer the money to it +- In plain language: Under the rules, the rent and the deposit may not be collected and paid on a tenant's behalf by an agent; the money goes straight to the landlord. Every time the money passes through the agent's hands you gain another chance for it to take the money and run. If it does run, you have paid the rent for nothing and the landlord still turns you out. Before you pay, check that the payee is the name on the title certificate. +- Benefit: The administrative regulation lists 「代收、代付住房租金、押金」, collecting and paying housing rent and deposits for others, among the acts a real estate agency is prohibited from, so an agent may not take your rent and deposit on the landlord's behalf. Every time the money passes through the agent's hands you gain another chance for it to run off with it +- Evidence grade: A +- Sources: State Council (国务院) (2025). Regulations on Housing Leasing, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第二十五条）. +- Notes: Before you pay, check that the payee is the name on the title certificate. If it is not the owner, ask questions and get a written authority. An agent may only charge the agency fee for bringing the two sides together + +### 4. Before you rent a long-term serviced apartment, check its escrow account, and do not pay a year up front to save money + +- Cost: No money. One check takes a bit over ten minutes +- In plain language: Paying a year up front buys you a small discount. If the company folds, your deposit and the remaining months of rent go together. A rent-linked loan is worse: if the operator runs off, you still finish paying the loan every month. Before you sign, check whether it has published an escrow account for the rent. +- Benefit: The administrative regulation requires a housing leasing business that operates by subletting to 「设立住房租赁资金监管账户并向社会公示」, set up an escrow account for housing lease funds and publish it. Subletting means the business rents the whole flat from the landlord and then sublets it to you. The small discount you get from paying a year up front does not cover the risk of losing the deposit and the remaining rent together if the business runs off. The risk is larger if you have also signed a rent-linked loan +- Evidence grade: A +- Sources: State Council (国务院) (2025). Regulations on Housing Leasing, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十九条）. +- Notes: Check at signing whether a loan is attached. If an instalment app pays the apartment a year's money in one go and you repay it monthly, you keep paying after the operator runs off. Do not sign that kind of contract + +### 5. If the flat is sold during the lease, the lease stays valid and you do not have to move + +- Cost: No money +- In plain language: If the flat is sold during the lease, your lease is still valid. The new landlord has to let you stay until the end of the term under the original contract. You do not move, and you do not renegotiate the price. The condition is that you are living there lawfully, so keep the contract, the transfer records and the evidence of when you moved in. +- Benefit: The Civil Code has a rule called 「买卖不破租赁」, a sale does not break a lease. The words of the article are: 租赁物在租赁期限内发生所有权变动的，不影响租赁合同的效力 — a change of ownership of the leased property during the lease term does not affect the validity of the lease contract. That means that if the flat changes hands during the term, the lease you have signed is still valid. The new landlord has to let you stay until the end of the term under the original contract +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第七百二十五条）. +- Notes: The condition is that you have taken possession lawfully, which is what the article calls lawful possession and use. Keep the contract, the transfer records and the evidence of when you moved in + +### 6. Before you sign, check the title certificate and the mortgages, and pay every sum by transfer with a note of the purpose + +- Cost: No money. One check takes half an hour +- In plain language: The two most common ways to lose everything are paying rent to somebody who is not the owner, and renting a flat that was already mortgaged to a bank and later sealed up. Before you sign, check the ownership and the mortgages once at the real estate registration centre. Pay every sum by transfer, and write in the note 「rent for a certain flat for a certain month」, because in a dispute that note is direct evidence. +- Benefit: The two most common ways to lose everything are paying rent to somebody who is not the owner, and renting a flat that was already mortgaged to a bank and later sealed up. When you transfer the money, note 「rent for a certain flat for a certain month」, because in a dispute that note is direct evidence +- Evidence grade: C +- Sources: the author's own experience, no direct literature ; the same reasoning about keeping evidence and noting the purpose of a transfer is in the two items in section 8 on bride price and promissory notes +- Notes: You can look up the ownership information at the local real estate registration centre. At signing, have the landlord attend in person; if the landlord cannot come, get a written power of attorney. If a second landlord sublets to you, you need the original landlord's written consent + +### 7. If an agent collects the purchase money for a second-hand flat, it must go through the agent's dedicated client settlement account at the bank + +- Cost: No money. The hard part is insisting on the dedicated account, which may mean arguing with the agent for a while +- In plain language: The purchase money runs from several hundred thousand to several million yuan, and it is the single largest sum in this book. Where an agent collects it, the rules require it to go through the dedicated account the agent has opened at the bank, and it may not be transferred by WeChat to an individual broker. Two agents working one deal together may also charge only one commission, while arranging a loan or handling the transfer of title needs a separate contract with the charges stated in advance. +- Benefit: The rule is explicit: where the parties to a property transaction agree that a real estate agency collects and pays the money, it should be transferred through the client transaction settlement account for deposits that the agency has opened at a bank. That means the purchase money goes into a regulated dedicated account, not the agency's own account, and not the private account of an individual broker +- Evidence grade: A +- Sources: Ministry of Housing and Urban-Rural Development (住房城乡建设部), National Development and Reform Commission and Ministry of Human Resources and Social Security (人力资源社会保障部) (2011). Measures for the Administration of Real Estate Brokerage, 房地产经纪管理办法（令第 8 号，2016 年第 29 号令修改）第二十四条：「房地产交易当事人约定由房地产经纪机构代收代付交易资金的，应当通过房地产经纪机构在银行开设的客户交易结算资金专用存款账户划转交易资金。」 第十八条「房地产经纪服务实行明码标价制度……在经营场所醒目位置标明房地产经纪服务项目、服务内容、收费标准」；第十九条「两家或者两家以上房地产经纪机构合作开展同一宗房地产经纪业务的，只能按照一宗业务收取佣金，不得向委托人增加收费」；第十七条 on arranging loans, registering property and other services「应当向委托人说明服务内容、收费标准等情况，经委托人同意后，另行签订合同」. +- Notes: The purchase money runs from several hundred thousand to several million yuan, and it is the single largest sum in this book, so do not transfer it by WeChat to a broker. There are two more common ways of being overcharged. The first is two agencies working the same deal together, which may carry only one commission. The second is arranging a loan or handling the transfer of title, which is a different service and needs its own contract with the charges told to you in advance. See item 6 in this section for paying every sum by transfer with a note of the purpose (check the title certificate and the mortgages before you sign). + +### 8. Do not rent a partitioned flat: the smallest unit for rent is a room as originally designed, and kitchens, toilets and balconies may not be lived in + +- Cost: No money. What it costs you is that on the same budget you either live further out or live in a smaller place +- In plain language: Under the rules, the smallest unit that can be rented out is a room as originally designed. Kitchens, toilets, balconies and underground storage rooms may not be lived in. If such a flat is inspected, you may have to move out. The landlord also may not raise the rent during the lease just because he says so. +- Benefit: The rule is explicit that the smallest unit for rent is a room as originally designed, and one room may not be partitioned again and let separately in several parts. The floor area per person may not be below the minimum standard set locally. Kitchens, toilets, balconies and underground storage rooms may not be let for people to live in. During the lease the landlord may not raise the rent unilaterally at will +- Evidence grade: A +- Sources: Ministry of Housing and Urban-Rural Development (住房和城乡建设部) (2010). Measures for the Administration of the Leasing of Commercial Housing, 商品房屋租赁管理办法（令第 6 号）第八条：「出租住房的，应当以原设计的房间为最小出租单位，人均租住建筑面积不得低于当地人民政府规定的最低标准。厨房、卫生间、阳台和地下储藏室不得出租供人员居住。」 第九条：「出租人应当按照合同约定履行房屋的维修义务并确保房屋和室内设施安全……房屋租赁合同期内，出租人不得单方面随意提高租金水平。」 +- Notes: When an illegal partition is inspected, it is the tenant who moves, and the deposit and the rent already paid are often not recovered. It is also worse for safety: partition walls often block the escape route, and several people in one flat share one electric line. Look at three things when you view a flat: whether the walls were built later, whether the room has a window, and whether the meter can carry that many people. + +### 9. If you have an urban hukou and want to live in the countryside, rent a village house and do not buy a homestead or a house on one + +- Cost: No money. What it costs you is that the house is not yours, one lease runs at most 20 years, and at the end of it you start again with the landlord +- In plain language: A person with an urban hukou may not go into the countryside to buy a homestead, a farmer's house or a 「小产权房」, and state documents forbid it outright. It does not help when a villager says 「a written contract is enough」: once the money is paid, the house still will not be yours. If you want to live in the countryside for a long time, sign a lease, at most 20 years at a time, and renew it at the end. +- Benefit: The State Council General Office document is explicit: 「城镇居民不得到农村购买宅基地、农民住宅或『小产权房』。」 — urban residents may not go to the countryside to buy homesteads, farmers' houses or 「小产权房」. The same document also provides that once a farmer has sold or rented out the house, a later application for a homestead will not be approved. A 2019 document from the Central Rural Work Leading Group Office and the Ministry of Agriculture and Rural Affairs said the same again: 「严禁城镇居民到农村购买宅基地。」 — urban residents are strictly forbidden to buy a homestead in the countryside. The same document sets out the lawful route: urban residents may rent a village house to live in or to run a business, the term of the lease may not exceed twenty years, and the two sides may agree otherwise when it ends +- Evidence grade: A +- Sources: State Council General Office (国务院办公厅) (2007). Notice on Strictly Enforcing the Relevant Laws and Policies on Collective Construction Land in the Countryside, 关于严格执行有关农村集体建设用地法律和政策的通知（国办发〔2007〕71 号）. ; Central Rural Work Leading Group Office and Ministry of Agriculture and Rural Affairs (中央农村工作领导小组办公室、农业农村部) (2019). Notice on Further Strengthening the Administration of Rural Homesteads, 关于进一步加强农村宅基地管理的通知（中农发〔2019〕11 号）第五、六部分. +- Notes: A homestead is a plot of land the village allocates to its own villagers to build on, and it can only go to villagers. When you rent a village house, sign a written contract with the farming household that rents it out, and set out the term, the rent, whether you may fit it out or rebuild, and how the money spent on the fit-out is counted up when the term ends. The documents set 20 years as the maximum term, so do not sign anything that says 「rent for 50 years」 or 「rent for 70 years」, and do not pay several years of rent in one go. The documents also forbid building villas, walled courtyards or private clubs on a homestead in the countryside, so do not pay a villager to 「build it jointly with you」. diff --git "a/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" new file mode 100644 index 00000000..82e9354d --- /dev/null +++ "b/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -0,0 +1,92 @@ +[← 回总目录](../README.md) + +# 16. 得了慢性病之后怎么活 + +口径：总死亡率与金钱。查出慢性病以后往往还有几十年要过，本节算的是这几十年怎么管。 + +[← Back to the contents](../README.md) + +# 16. Living with a chronic illness + +This section measures all-cause mortality and money. After a chronic illness is diagnosed you usually have decades left to live, and this section works out how to manage them. + +### 1. Take the medicine for the full course as the doctor ordered, and do not stop because you feel better + +- Cost: No money, and you save the money you would later spend on treating complications. Taking the tablets takes a few minutes a day. The hard part is taking them every day without a gap. +- In plain language: Among people who take the medicine for the full course as the doctor said, the chance of dying in the same period is about half that of people who take it on and off. Read that figure with care. People who take a placebo for the full course also have a lower mortality, which means part of the benefit comes from 「the people who keep taking their tablets are simply more careful about their health」. The harm of stopping is certain, and especially so for blood pressure medicine, anticoagulants and anti-rejection drugs. +- Benefit: Pooling 21 studies and 46847 people (meta-analysis): compared with people who take their medicine on and off, those who take it for the full course as ordered are about 44% less likely to die in the same period. The underlying figure is an odds ratio of 0.56, 95% CI 0.50 to 0.63 (credible range). Looking only at treatments that do work, it is about 45% lower (0.55, 0.49 to 0.62) +- Evidence grade: A +- Sources: Simpson SH, Eurich DT, Majumdar SR, et al. (2006). A meta-analysis of the association between adherence to drug therapy and mortality. BMJ, 333(7557), 15. +- Notes:Disputed. This is also the figure in the book that most needs reading with a discount. In the same pooled analysis, people who took a placebo (a dummy drug) for the full course also had a lower mortality (0.56, 0.43 to 0.74). That means a large part of the benefit comes from 「the people who keep taking their tablets are simply more careful about their health」. The authors call this confounder the healthy adherer effect. The harm of stopping the drug itself is nonetheless clear, and especially so for blood pressure medicine, anticoagulants and anti-rejection drugs + +### 2. Get the outpatient chronic and special disease status first and the cross-region filing second, and hypertension, diabetes, chemotherapy and radiotherapy, dialysis and anti-rejection treatment settle directly across regions + +- Cost: No money. Before you travel, do the cross-region medical filing once in the medical insurance app, which you can do on your phone. +- In plain language: For hypertension, diabetes, outpatient chemotherapy and radiotherapy for a malignant tumour, dialysis for uraemia and anti-rejection treatment after an organ transplant, the outpatient costs can be settled directly against your medical insurance in another place. You do not pay out of pocket first and then claim in your home area. But you have to get the outpatient chronic and special disease status in your home area first, and the cross-region filing second. Both are needed for it to work. +- Benefit: A document from the National Healthcare Security Administration says this has already been achieved. The words are: 「基本实现高血压、糖尿病、恶性肿瘤门诊放化疗、尿毒症透析、器官移植术后抗排异治疗等 5 种门诊慢特病费用跨省直接结算县域可及」 — outpatient chronic and special disease costs in 5 kinds of condition, namely hypertension, diabetes, outpatient chemotherapy and radiotherapy for a malignant tumour, dialysis for uraemia and post-operative anti-rejection treatment after an organ transplant, can now be settled directly across provinces and reached at county level. 「县域可及」, reached at county level, means you can get it done in a county. The document also adds 5 more conditions to these 5 +- Evidence grade: A +- Sources: Office of the National Healthcare Security Administration and General Office of the Ministry of Finance (国家医保局办公室、财政部办公厅) (2024). Notice on Safely and in Order Expanding the Range of Outpatient Chronic and Special Diseases Settled Directly across Provinces, 关于稳妥有序扩大跨省直接结算门诊慢特病病种范围的通知（医保办发〔2024〕19 号）. +- Notes: The condition is that you do both: get the outpatient chronic and special disease status in your home area first, then make the cross-region filing. How far the 5 added conditions have been opened up varies from place to place, so follow what your local medical insurance department publishes + +### 3. Do your follow-up tests at the interval the doctor gives, and write the figures from each one in the same notebook + +- Cost: Each test costs from tens to several hundred yuan. A trip to the hospital plus waiting for the results takes some of your time. The hard part is having to go again after a while, and it is easy to keep putting it off. +- In plain language: A chronic illness is judged by the change over a period, not by one reading. Write the date, the figures, the medicine you were taking and the dose in the same notebook. When you change hospital, take it out, and you do not have to have the tests done again or pay for them again. +- Benefit: A chronic illness is judged by how the figures change over a period, and one reading is not enough. Keeping everything in one place has a second benefit: when you change hospital you do not repeat the tests and you do not pay for them twice +- Evidence grade: C +- Sources: the author's own experience, no direct literature +- Notes: Record at least four things each time: the date, the value, the medicine you were taking, and the dose. Photographing the lab report on your phone is fine, but keep all of it in one place rather than scattered + +### 4. Do not stop proper treatment in order to try a folk remedy or a health supplement + +- Cost: No money, and you save the money you would have spent on the supplement. It takes no time. The hard part is standing firm when other people keep telling you to try it. +- In plain language: The harm of stopping the drug is certain; the benefit of a folk remedy or a supplement is uncertain. That is trading a certain loss for an uncertain gain. If you want to try a supplement, tell your treating doctor first. Some supplements change the level of anticoagulants and anti-rejection drugs in your blood. +- Benefit: The harm of stopping the drug is certain; the benefit of a folk remedy is uncertain. This book lists a batch of the common categories that do not work in section 6 +- Evidence grade: C +- Sources: the author's own experience, no direct literature ; for the related evidence see section 6 +- Notes: If you want to try a supplement, tell your treating doctor first. Some supplements interact with prescription drugs, for example by changing the level of anticoagulants and anti-rejection drugs in your blood + +### 5. For a stable chronic illness, a community health centre can prescribe 12 weeks of medicine at a time + +- Cost: No money. Go to the community health centre and ask one question, whether they can write a long-term prescription. That is all it takes. +- In plain language: If your condition is stable and the medicine has not changed, the community health centre can prescribe enough for 12 weeks in one go. That saves the half day every month of travelling to a big hospital and queueing to register. Within the same insurance area, the reimbursement rate at a grassroots hospital is also usually about ten percentage points higher than at the hospital above it. +- Benefit: No monthly trip to a big hospital to queue for a prescription. Within the same insurance area, also called the pooling region, the reimbursement rate at a grassroots hospital is in principle about 10 percentage points higher than at the hospital above it +- Evidence grade: A +- Sources: State Council General Office (国务院办公厅) (2026). Several Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System, 关于加快建设分级诊疗体系的若干措施. : 「对于符合条件的慢性病患者，基层医疗卫生机构单次可开具不超过12周用药的长期处方。」 The same document says 「原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距」, in principle the inpatient reimbursement rates inside the pooling region are spread by about 10 percentage points at each level, and it requires higher-level hospitals to run outpatient clinics for common chronic conditions such as hypertension, diabetes and chronic obstructive pulmonary disease at the grassroots level +- Notes: Whether you 「meet the conditions」 is for the doctor to decide, and it generally means the diagnosis is settled and the treatment plan is stable. It is the same medicine as before; what you save is the half day of going there each month and the registration fee. If the grassroots level does not stock the medicine you need, ask whether they keep a shortage register and can deliver. + +### 6. Before you sign up with a family doctor in the community, ask which items are covered by medical insurance and which you pay for yourself + +- Cost: No money. One extra question before you sign, and it takes a minute. +- In plain language: The basic service package you sign for goes through medical insurance as the rules require. You pay the whole cost of a personalised service package yourself. When somebody sells you an 「upgrade package」, first find out whether it is the basic package or a personalised one, whether it has been filed, and what it costs for a year. +- Benefit: The basic service package is paid by medical insurance as the rules require. A personalised service package is paid for entirely by the individual, and such a package also has to be filed with the county health commission +- Evidence grade: A +- Sources: State Council General Office (国务院办公厅) (2026). Several Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System, 关于加快建设分级诊疗体系的若干措施. : 「加强基层门诊付费与签约服务政策联动，基本服务包按规定纳入医保支付；个性化服务包由签约基层医疗卫生机构按程序向县级卫生健康部门备案，费用由个人支付。」 +- Notes: The value of signing up is that somebody follows your illness over the long term, not how many items are bundled into the package. When somebody sells you an 「upgrade package」, ask three questions first: is this the basic package or a personalised one, has it been filed, and what does it cost for a year. + +### 7. Once diabetes is diagnosed, have the back of your eyes examined once, then at the interval the doctor gives; have your feet checked once a year + +- Cost: One dilated examination of the back of the eye costs from tens to several hundred yuan. The foot check is normally part of the outpatient visit and costs nothing extra. The hard part is remembering to go when your eyes and feet give you no trouble at all. +- In plain language: Diabetes damage to the back of the eye and to the feet gives almost no warning before the damage is bad enough to make you blind or to rot. Find it early and it can still be treated. For type 2 diabetes, have a dilated examination of the back of the eye at the time of diagnosis; for type 1, start five years after the onset. If one or two examinations have found nothing and your blood sugar is on target, you can stretch it to once every one or two years. Have your feet checked once a year. +- Benefit: Diabetes damage to the back of the eye is called retinopathy, and damage to the feet is called diabetic foot. Both give almost no warning before something goes wrong. Find it early and it can still be treated. Once you cannot see, or the foot has started to rot, it is too late +- Evidence grade: A +- Sources: American Diabetes Association (美国糖尿病学会) (2026). Standards of Care in Diabetes 2026, Chapter 12, Retinopathy, Neuropathy and Foot Care, 糖尿病诊疗标准 2026·第 12 章 视网膜病变、神经病变与足部护理. Diabetes Care: 「People with type 2 diabetes should have an initial dilated and comprehensive eye examination by an ophthalmologist or optometrist at the time of the diabetes diagnosis.」「Adults with type 1 diabetes should have an initial dilated and comprehensive eye examination by an ophthalmologist or optometrist 5 years after the onset of diabetes.」「If there is no evidence of retinopathy from one or more annual eye exams and glycemic indicators are within the goal range, then screening every 1–2 years may be considered.」「Perform a comprehensive foot evaluation at least annually to identify risk factors for ulcers and amputations.」 +- Notes: For type 2 diabetes you have the examination at the time of diagnosis, and for type 1 you start five years after the onset. The difference is that type 2 has often been quietly high for years. If one or two examinations have found nothing and your blood sugar is on target, you can stretch it to once every one to two years. If the feeling in your foot has become dull, look at the sole once a day yourself, using a mirror if that is easier. + +### 8. If you have had a kidney stone, drink 2.5 to 3 litres of water a day and keep salt under 6 grams + +- Cost: No money. Turn drinking water throughout the day into a habit. On days when you sweat more you have to drink more still, and the hard part is keeping count every day. +- In plain language: Among people who form a calcium stone for the first time, about a quarter have another one within five years if they do nothing. Among those who drink enough water every day, a little over one in ten has a recurrence within five years, which is more than half fewer. The British guideline gives 2.5 to 3 litres of water a day and no more than 6 grams of salt. +- Benefit: One controlled trial randomised into two groups ran for 5 years, with 199 patients who had a first calcium stone and no cause found for it, also called idiopathic calcium stone. One group drank at least 2 litres of water a day and the other had no intervention. Recurrences over the 5 years were 12/99 (12.1%) in the water group and 27/100 (27.0%) in the no-intervention group, P=0.008. The smaller the P value, the less the difference between the two groups can be chance. The average interval to recurrence was 38.7±13.2 months against 25.1±16.4 months, P=0.016. The 24-hour urine volume measured at the start was clearly lower in the stone patients than in those without stones: for men 1,057±238 mL against 1,401±562 mL (P<0.0001), and for women 990±230 mL against 1,239±440 mL (P<0.001). Recommendation 1.8.1 of NICE guideline NG118 advises adults to drink 2.5 to 3 litres of water a day, and children and adolescents 1 to 2 litres according to age. Add a squeeze of fresh lemon to the water, do not drink fizzy drinks, and keep salt under 6 grams a day for adults +- Evidence grade: A +- Sources: Borghi L, Meschi T, Amato F, Briganti A, Novarini A, Giannini A (1996). Urinary volume, water and recurrences in idiopathic calcium nephrolithiasis: a 5-year randomized prospective study. The Journal of Urology, 155(3), 839-843. ; National Institute for Health and Care Excellence (2019). Renal and ureteric stones: assessment and management. NICE guideline NG118，第 1.8.1 条. +- Notes: Recurrence is down by about 55% and it costs nothing, so by the ordering rules of this book it is the best value for money in this section. What you watch is the 24-hour urine volume, with a target of 2 to 2.5 litres or more, not how much you drink in. People who sweat more and people who work in high temperatures have to drink more. That trial covered patients having a first attack of idiopathic calcium stone. Stone composition is not one thing: uric acid stones, infection stones and cystine stones each have their own answer. Anyone who has had a recurrence is worth having a stone composition analysis and a 24-hour urine test, and only then deciding whether to add a drug. Do not cut calcium out in order to prevent stones: see item 21 in section 6. If the pain in your back and belly is severe enough to be colicky, go to hospital; drinking more water does not flush it out. Blood in the urine without pain is a separate thing to investigate: see item 27 in section 1 + +### 9. Once gout is diagnosed, take a urate-lowering drug for the long term and keep the blood urate below 360 µmol/L + +- Cost: Allopurinol costs from a few yuan to tens of yuan a month. At the start you test the blood urate every few weeks to adjust the dose, and once it is stable every few months. For the first 3 to 6 months you also take a drug that prevents attacks. The hard part is keeping to it when nothing hurts. +- In plain language: Gout needs long-term urate-lowering treatment that holds the urate below 360 µmol/L, and painkillers taken only when it hurts are no substitute. Among patients whose dose was adjusted by a nurse against the lab results, 95% were on target after two years, and only 30% of those seen in the ordinary way were. You still cannot stop once you are on target. Diet does not bring the level down to that line, and giving up alcohol brings it down by only 1.6 mg/dL. +- Benefit: One controlled trial randomised into two groups covered 517 adults who had had an attack in the previous 12 months. One group was nurse-led: the nurse explained what the disease was, then adjusted the urate against the targets until it was on target. The other group saw a general practitioner as usual. At 2 years the share with blood urate down to <360 µmol/L (6 mg/dL) was 95% against 30%. The odds of being on target were 3.18 times those of the usual-care group (RR 3.18, 95% CI 2.42 to 4.18, and that interval is the credible range of the figure), P<0.0001. The secondary measures in year 2, attack frequency, tophi and quality of life, were also better than in the usual-care group. Each extra year of life gained, counted in quality-adjusted life years, cost £5,066. The same trial also ran a follow-up study: questionnaires went to 438 people and 82% came back. Both groups reported their own number of attacks over the previous 12 months. The median in the on-target group was 0 attacks, with the middle half lying between 0 and 0. In the usual-care group it was 1 attack (0 to 3), P<0.001. The range in brackets is called the interquartile range. The share still taking a urate-lowering drug was 1.19 times that of the usual-care group (adjusted RR 1.19, 1.09 to 1.30). The 2020 American College of Rheumatology gout guideline strongly recommends starting urate-lowering treatment in three kinds of people: those with tophi under the skin, those in whom imaging shows gout has eaten defects into the bone, and those with ≥2 attacks a year. The method is to keep monitoring the blood urate and adjust the dose against the results until it is on target, with a target of <6 mg/dL. Allopurinol is the first choice, and it is also the first choice in people with stage 3 or worse chronic kidney disease, with a starting dose of ≤100 mg/day. Give an anti-inflammatory preventive drug at the same time as you start, for at least 3 to 6 months. The same guideline also cites a series of case observations. Those patients stopped the drug after long-term good control, and their blood urate stayed below <7 mg/dL after stopping. Among them, only 13% had no attack at all over the 5-year follow-up (27/211). On diet, people who limit or give up alcohol have blood urate 1.6 mg/dL lower than people who do not, and one serving of beer raises blood urate by 0.16 mg/dL. A healthy diet, the Mediterranean diet and the DASH diet have a smaller effect. There is disagreement over which urate-lowering drug to use first. The American College of Rheumatology guideline strongly recommends allopurinol as first choice, on the grounds that it works when the dose is high enough, is well tolerated, is safe and is cheap. It also makes a conditional recommendation: people on febuxostat who have a history of cardiovascular disease, or who have just had a cardiovascular event, should switch if another drug is available. That recommendation matches the black-box warning the FDA added to febuxostat. One of the grounds for it is the CARES trial, in which 6190 gout patients with coexisting cardiovascular disease were randomly assigned febuxostat or allopurinol. In the febuxostat group the risk of all-cause death was 1.22 times that of the allopurinol group (HR 1.22, 95% CI 1.01 to 1.47, about 22% higher). Cardiovascular death was 1.34 times (1.03 to 1.73, about 34% higher). The European FAST trial reached a different conclusion. It enrolled 6128 patients over 60 with cardiovascular risk factors, and febuxostat produced no more major cardiovascular events than allopurinol (HR 0.85, 0.70 to 1.03). The death rate was 7.2% on febuxostat and 8.6% on allopurinol. For people taking a drug that pushes urate out through the kidneys, the American College of Rheumatology conditionally advises against alkalinising the urine, because there is no evidence that it works and the certainty of that evidence is very low. It also requires that these people be told to drink more water, and people known to have kidney stones or stage 3 or worse chronic kidney disease should not use this class of drug +- Evidence grade: A +- Sources: Doherty M, Jenkins W, Richardson H, et al. (2018). Efficacy and cost-effectiveness of nurse-led care involving education and engagement of patients and a treat-to-target urate-lowering strategy versus usual care for gout: a randomised controlled trial. Lancet, 392(10156), 1403-1412. ; Abhishek A, Jenkins W, La-Crette J, Fernandes G, Doherty M (2020). Nurse-led care is preferred over GP-led care of gout and improves gout outcomes: results of Nottingham Gout Treatment Trial follow-up study. Rheumatology, 59(3), 575-579. ; FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760. ; White WB, Saag KG, Becker MA, et al.; CARES Investigators (2018). Cardiovascular Safety of Febuxostat or Allopurinol in Patients with Gout. New England Journal of Medicine, 378(13), 1200-1210. ; Mackenzie IS, Ford I, Nuki G, et al.; FAST Study Group (2020). Long-term cardiovascular safety of febuxostat compared with allopurinol in patients with gout (FAST): a multicentre, prospective, randomised, open-label, non-inferiority trial. Lancet, 396(10264), 1745-1757. ; Chinese Society of Endocrinology (中华医学会内分泌学分会) (2020). Guidelines for the Diagnosis and Treatment of Hyperuricaemia and Gout (2019), 中国高尿酸血症与痛风诊疗指南(2019). 中华内分泌代谢杂志, 36(1), 1-13. +- Notes:Disputed: guidelines in different countries disagree about which drug to use. The US guideline makes allopurinol first choice and says people with cardiovascular disease should switch off febuxostat where they can, but the European FAST trial did not find febuxostat more dangerous. In China you can also buy benzbromarone, which pushes urate out through the kidneys. The guidelines of the Chinese Medical Association (the 2019 version and the 2024 update) give separate recommendations on the choice of drug and on alkalinising the urine. Whether to add sodium bicarbonate to alkalinise the urine when taking benzbromarone: the US guidelines say there is no evidence, so follow your own doctor's advice, and do not leave out the water. **It is worth testing HLA-B*5801 before a Han Chinese person takes allopurinol.** It is an inherited gene type that governs the immune response, and one test result stays the same for life. Among people with this gene the figure is 7.4% among the Han, Korean and Thai populations and only 0.7% among white and Hispanic people. The risk of allopurinol hypersensitivity syndrome in Asian and African people after taking allopurinol is 3 times that of white people, and in its severe form the skin peels off over large areas, which can kill. The 2020 American College of Rheumatology gout guideline suggests, for these two groups, considering testing before starting the drug. **Starting urate-lowering treatment is itself more likely to trigger an attack**, so the guideline clearly requires an anti-inflammatory preventive drug during the first 3 to 6 months. 「It got more painful after I took the urate-lowering drug, so I stopped」 is the commonest way this goes wrong. When it hurts, what you do is add an anti-inflammatory drug as the doctor ordered, not stop the urate-lowering drug. The target value is written differently by different sources: <6 mg/dL is about 360 µmol/L, and Chinese lab sheets mostly use µmol/L. A high urate on a health check with no attack ever is a different question: see item 19 in section 6. For sugary drinks and alcohol, see items 7 and 20 in section 2 (do not drink sugary drinks, and drink little or no alcohol) diff --git "a/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" "b/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" new file mode 100644 index 00000000..581d3212 --- /dev/null +++ "b/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" @@ -0,0 +1,83 @@ +[← 回总目录](../README.md) + +# 17. 家里有老人 + +口径：金钱与人身自由。家里有老人，多数麻烦不是钱不够，是没提前把话说清楚、把字签好。最后一条讲长期卧床的压疮，按死亡率算账，不和本节其他条目比较。 + +[← Back to the contents](../README.md) + +# 17. When there are elderly at home + +This section measures money and personal freedom. With elderly people at home, most of the trouble is not a shortage of money; it is not having said the words clearly and signed the paper in advance. The last item covers pressure ulcers in someone bedridden long-term, and it is measured by mortality rather than compared with the other items here. + +### 1. While the older person is still clear-headed, name in writing who will be their guardian + +- Cost: A few hundred yuan to have one notarised. One conversation with the older person settles it. The hard part is finding the words: you have to say out loud the day when he can no longer manage his own affairs +- In plain language: While the older person is still clear-headed, spend a few hundred yuan on a written document at a notary office saying who will be their guardian in future. Without that document, once the older person can no longer make their own decisions, who signs for them and who handles their money has to be appointed by the court. If the family disagrees, it takes even longer. +- Benefit: The Civil Code provides that an adult who is clear-headed and able to decide for himself, which the law calls 「an adult with full capacity for civil conduct」, may agree in advance with somebody over who will be their guardian. The person agreed with may be a close relative, or another individual or organisation willing to act as guardian. Once agreed, it has to be settled 「以书面形式确定自己的监护人」, in writing that names the guardian. When he can no longer make decisions, or can only make some of them, which the law calls 「丧失或者部分丧失民事行为能力」, the person named in the document takes over. Without that document, who signs for him and who handles his money has to go through an appointment by the court +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第三十三条）. +- Notes: This kind of document is called voluntary guardianship, 意定监护, which means a guardian you name yourself in advance. That person comes before the default order the law sets out. An older person living alone, children living in another city, or a complicated family situation all make it more necessary to do this in advance. Have it notarised at a notary office + +### 2. Make a will, and remember that a later one overturns an earlier one, and that a notarised will no longer comes first + +- Cost: Writing a will by hand costs nothing. Having it notarised costs from several hundred to over a thousand yuan. The hard part is finding the words: you have to talk to the family about what comes after you die +- In plain language: The law recognises six forms of will, and if the formalities are wrong the whole thing is void, which is the commonest way this goes wrong. A handwritten will has to be written out by the testator from beginning to end, signed, and dated. Where several wills contradict each other, the last one made governs. From 2021 a notarised will no longer ranks first automatically. +- Benefit: The Civil Code recognises six forms of will: handwritten by the testator, written by somebody at his direction, printed, recorded on audio or video, oral, and notarised. Each has its own formal requirements, which the law calls 「形式要件」, the elements of form. If several wills contradict each other, the last one made governs. From 2021 the notarised one no longer ranks first automatically +- Evidence grade: A +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千一百三十四条至第一千一百四十二条）. +- Notes: A will handwritten by the testator has to be written out by them from beginning to end, signed, and dated. A printed will needs more than two witnesses. The testator and the witnesses have to sign every page and give the date. Heirs, people named in the will to receive something, and their close relatives may not act as witnesses. Get the formalities wrong and the whole will is void, which is the commonest way this goes wrong + +### 3. Keep the older person's money in a separate account, and set a rule that two people confirm any large payment + +- Cost: No money. One conversation with the family and you have agreed a figure. The hard part is how to raise it, so that the older person does not think you are watching them +- In plain language: The con that pushes older people into buying health products, collectibles or investment products always works the same way: spend a few months getting close to him, then take everything at once. Agree a rule in advance that any payment over a certain amount has to be mentioned to a certain person first. That pushes back the moment of paying, until it is past the most impulsive stretch. +- Benefit: The health product, collectible and investment scams aimed at older people all start by spending a few months building up trust, then take everything at once. Agree in advance that 「any payment over a certain amount has to be mentioned to a certain person first」. That pushes back the decision to pay, until it is past the most impulsive stretch +- Evidence grade: C +- Sources: the author's own experience, no direct literature ; for the general rules on fraud see section 8 and section 13 +- Notes: Do not put it as 「I am afraid you will be cheated」. Put it as 「the family's money goes into one set of accounts」 and the older person takes it far more easily. Do not hand all of the older person's money to the children to keep either, because that produces a different kind of dispute + +### 4. Give the older person one sentence they can say at any time to get out of the situation + +- Cost: No money. One conversation with the older person, a few minutes is enough +- In plain language: Selling and fraud both work by pressure in the moment, which makes it hard to refuse on the spot. Agree one sentence with the older person in advance: 「I need to go back and talk it over with my family.」 He can say it at any time and get himself a way out. That works better than explaining the reasons afterwards. +- Benefit: Selling and fraud both rely on pressure in the moment. Agreeing the sentence 「I need to go back and talk it over with my family」 in advance gives the older person a way out. It works better than explaining the reasons afterwards +- Evidence grade: C +- Sources: the author's own experience, no direct literature +- Notes: Use it together with item 3 in this section (keep the older person's money in a separate account). Do the fall-proofing changes round the house while you are at it: see section 1. The sentence gives the older person a way out so that he does not have to stand his ground on the spot; it is not a way of watching him. When he is surrounded by sellers, the hardest part is saying no in front of a roomful of people, and that is harder than telling whether they are genuine + +### 5. Stay away from every 「invest in your retirement」 scheme that asks the older person to pay first: cards, bed reservations, retirement flats, living-out schemes and products for older people are one kind of illegal fundraising + +- Cost: No money. Read the list of names below to the older person once, and it takes a few minutes. After that, refuse them straight away: do not go to the lectures, do not go to the visits, do not leave your phone number. The hard part is that once the older person has been a few times and knows the people, it is not easy to talk them round +- In plain language: Retirement schemes that ask the older person to pay first and promise a return are all the same con: membership cards, bed reservations, retirement flats, living-out schemes, and selling products for older people at a meeting. Under the rules, money lost this way has to be borne by the person who lost it. One man took 136 million yuan by taking reservations for beds in a retirement home and harmed more than 2,100 people. The only step you can stop is the moment of paying. +- Benefit: The Ministry of Civil Affairs (民政部) lists five forms this takes, and recognising the name is more reliable than recognising the face. The first is borrowing the name of a retirement service institution: no such body exists, a venue is rented for the occasion, and it claims to have signed a cooperation agreement with a retirement home in order to 「碰瓷」, to attach itself to a real one. The second is doing business as 「elderly care services」: selling 「VIP cards」, 「membership cards」 and 「prepaid cards」, taking payment in advance for 「elderly care service fees」, and promising services well beyond its own capacity to provide beds. The third is doing business as investment in 「retirement projects」. Common names include taking a share in a health and retirement base and selling a retirement flat that does not exist. Others are letting a bed out for a long term and running 「time bank」 mutual eldercare. The signs waved are refund of the principal on sale, subletting after sale, and a promised buy-back. The fourth is doing business as selling 「products for older people」: buying goods back, holding goods to sell for you, cash back on spending, free health checks, gifts, selling by lecture, health talks and free clinics by experts. The fifth is doing business as 「living out in retirement」: cheap or free travel, a rebate on money stored with them, and earning retirement by points. That the money cannot be recovered is written into the regulations. Article twenty-five of the Regulations on the Prevention and Control of Illegal Fundraising says 「因参与非法集资受到的损失，由集资参与人自行承担」, the loss from taking part in illegal fundraising is borne by those who took part, so the money lost is yours alone. In the cases official bodies have announced, Dai Mouping and others raised 136 million yuan by illegal fundraising under the name of reserving beds in a retirement home and harmed more than 2,100 people, and Dai Mouping was sentenced to 15 years for fundraising fraud with a fine of 400,000 yuan. Cao Miaoming was sentenced to life imprisonment for fundraising fraud of 13,207 million yuan in the name of investing in the retirement industry, affecting more than 110,000 people, with political rights deprived for life and all personal property confiscated. Selling health products at a meeting is fraud. Ni Mourong used free banquets, gifts and a coach service to bring older people along to his lectures. The lecturers invented cancer-preventing and anti-cancer effects with 「a false persona, false cases and an inflated price cut」. Older people were induced to buy 「dihydroquercetin」 at dozens of times the cost price. The sales came to more than 620,000 yuan. Ni Mourong was sentenced to 10 years and 3 months, with a fine of 120,000 yuan (nationwide, the regulations in force from 2021-5-1) +- Evidence grade: A +- Sources: Ministry of Civil Affairs (民政部) (2023). Risk Notice on Illegal Fundraising in the Field of Elderly Care Services, 关于养老服务领域非法集资的风险提示. (published on the Chinese government website) ; Ministry of Civil Affairs (民政部) and China Banking and Insurance Regulatory Commission (中国银保监会) (2021). Risk Notice on Illegal Fundraising in the Field of Elderly Care, 关于养老领域非法集资的风险提示. ; State Council (国务院) (2021). Regulations on the Prevention and Control of Illegal Fundraising (State Council Order No. 737, articles two, twenty-five and thirty), 防范和处置非法集资条例（国务院令第 737 号，第二、二十五、三十条）. ; Supreme People's Procuratorate (最高人民检察院) (2022). Typical Cases of Procuratorates Cracking Down on and Rectifying Fraud against the Elderly, 检察机关打击整治养老诈骗犯罪典型案例. ; Supreme People's Court (最高人民法院) (2025). Typical Cases of Punishing Fraud Against People's Livelihoods According to Law (case five, the Ni Mourong fraud case), 依法惩治涉民生领域诈骗犯罪典型案例（案例五 倪某荣诈骗案）. +- Notes: You do not judge by 「this one looks official」, you judge by 「does it ask you to pay first, and does it promise a return」. A retirement home is in the business of selling services, so once the talk turns to investment, rebates or getting your money back, do not believe it. Being taken to lectures, on trips or to free clinics is not itself illegal and is very hard to stop. What you can stop is the moment of paying, and the two-person confirmation rule in item 3 in this section works with it. If the money has already been paid, call 110 or 96110 at once to ask for the payment to be stopped, following item 2 in section 8, and report it at the same time to the local lead department in charge of dealing with illegal fundraising. Raising money against your home is a different route: see item 6 in this section (reverse-mortgage insurance for the elderly). + +### 6. Apart from the insurance companies' reverse-mortgage insurance for the elderly, stay away from every other 「raise money against your home」 scheme, and never mortgage the flat to buy investment products + +- Cost: No money. Do not hand the property certificate at home to anybody who comes to the door in the name of 「raising money against your home」. For mortgages, guarantees and loans, show the agreement to your children before you sign. The hard part is that the people who come are often warm and attentive and have made many trips, and the older person finds it hard to say no +- In plain language: There is only one legitimate scheme that raises money against a home, the reverse-mortgage insurance for the elderly sold by insurance companies, and it is still in the pilot stage. The kind sold by people who come to your door has nothing to do with it: it tricks the older person into mortgaging the flat and then uses the borrowed money to buy the investment products it recommends. Some people do not even know the flat has been mortgaged, and in the end they lose the flat and still owe the loan. +- Benefit: The consumer protection bureau of the China Banking and Insurance Regulatory Commission set out both the genuine version and the fake one. There is only one legitimate scheme that raises money against a home, and it is called 「老年人住房反向抵押养老保险」. An older person with full legal ownership of a flat mortgages the property to an insurance company. The flat is still theirs to live in, to use and to collect rent from. To sell it or dispose of it they need the consent of the mortgagee, which is the insurance company. The older person draws a pension on the agreed terms until death. This business is still at the pilot stage and few people use it. The 「以房养老」 sold by criminals 「完全无关」, has nothing to do at all with the reverse-mortgage insurance for the elderly that the state is piloting, and is 「假借国家政策、为非法集资活动造势宣传的手段」, a means of borrowing the name of state policy to advertise and build up support for illegal fundraising. The method is to trick the older person into mortgaging the property and then use the borrowed money to buy the investment products it recommends, and this 「往往是『以新还旧』的庞氏骗局」, is often a Ponzi scheme that pays the old claims with new money. The consequences are given in the original wording: 「有的参与人甚至都不知道自己的房产被抵押，最终失去了房子，还背负贷款」, some participants do not even know their property has been mortgaged, and in the end they lose the flat and carry the loan. The notice also gives a line for reading a promised return. The words are: 「理财产品如果承诺收益率超过 6% 就要打问号，超过 8% 就很危险，10% 以上就要准备损失全部本金，『保本高收益』就是金融诈骗」, if an investment product promises a return above 6% you should be suspicious, above 8% it is very dangerous, and above 10% you should be ready to lose the whole principal, and 「保本高收益」, capital-protected with a high return, is financial fraud. The same notice also requires that 「慎重对待合同签署环节，不在空白合同签字」, the signing of the contract be handled carefully and no signature be put on a blank contract (nationwide, 2021-6) +- Evidence grade: B +- Sources: Consumer Protection Bureau of the China Banking and Insurance Regulatory Commission (中国银保监会消费者权益保护局) (2021). Risk Notice on Being Alert to Financial Fraud under the Names of 「investing in your retirement」 and 「raising money against your home」, 关于警惕「投资养老」「以房养老」金融诈骗的风险提示（2021 年第三期）. (reprinted by the Hunan Provincial Local Financial Administration Bureau) ; Ministry of Civil Affairs (民政部) and China Banking and Insurance Regulatory Commission (中国银保监会) (2021). Risk Notice on Illegal Fundraising in the Field of Elderly Care, 关于养老领域非法集资的风险提示（第四种表现形式：以宣称「以房养老」为名非法集资）. +- Notes: It is graded B because these statements come from regulators' risk notices and from judgement, not from research data. The three lines of 6%, 8% and 10% are reference lines the regulator gives consumers, not standards set by law. The China Banking and Insurance Regulatory Commission has now been merged into the National Financial Regulatory Administration, and the original page will not open, so the reprint on a provincial financial regulator's site is cited here. Losing money on an investment product is the smaller problem: once the flat has been mortgaged out, the older person still owes a loan, and that loss is far larger than the principal. For the general rules on signing and blank contracts, see item 17 in section 8; for stopping a payment after you have been cheated, see item 2 in section 8 + +### 7. If an older person at home is bedridden long-term or severely disabled, apply to the medical insurance department in your home area for long-term care insurance; it is not only for old people + +- Cost: Submit one application at the medical insurance department in your home area. Then have a disability level assessment. If you pass at the first assessment, the fund pays the assessment fee and you pay nothing +- In plain language: Long-term care insurance pays for the care you buy. If someone at home is severely disabled and it has lasted more than half a year, apply at the medical insurance department in your home area. You do not pay the care fees out of pocket first; around 70% is reimbursed for people insured as employees and around 50% for people insured as residents, and bed fees and meal costs are not reimbursed. There is no age threshold, so a young person who is severely disabled can apply as well. +- Benefit: Qualifying long-term care service costs carry no deductible. The deductible is the amount you pay yourself before the cover starts, and here you pay nothing up front. For people insured under the employed-worker scheme, the fund reimburses around 70%. For people insured under the scheme for residents who are not in employment, around 50%. The most that can be paid in one year, the annual maximum payment limit, is not more than 50% of the previous year's per-capita disposable income of urban and rural residents in the local pooling region +- Evidence grade: A +- Sources: National Healthcare Security Administration and seven other departments (国家医保局等八部门) (2026). Notice on Issuing the Implementation Plan for Accelerating the Establishment of the Long-Term Care Insurance System, 关于印发《加快建立长期护理保险制度实施方案》的通知. : the people covered are 「按规定参保缴费且失能状态长期持续（一般为6个月以上），经申请通过评估认定的失能人员」, those who have paid contributions as required and whose disability has continued for a long time, generally more than 6 months, and who are recognised as disabled after applying and passing an assessment, and 「制度起步阶段保障重度失能人员」, the scheme covers severely disabled people at the stage where it starts ; 「待遇享受不设起付标准。符合规定的长期护理服务费用，按未就业城乡居民参保政策参保的，基金支付比例为50%左右；按单位职工参保政策参保的，基金支付比例为70%左右，退休人员享受单位职工参保待遇」, benefits carry no deductible threshold; for qualifying long-term care service costs the fund pays around 50% for people insured under the scheme for residents who are not in employment, and around 70% for people insured under the employed-worker scheme, and retired people get the employed-worker treatment ; 「参保人员基金年度最高支付限额不超过统筹地区上年度城乡居民人均可支配收入的50%」, the annual maximum the fund pays for one person is not more than 50% of the previous year's per-capita disposable income of urban and rural residents in the pooling region ; 「待遇享受固定等待期原则上按照6个月设置」, the fixed waiting period for benefits is in principle set at 6 months ; also see the Opinions of the General Offices of the CPC Central Committee and the State Council on Accelerating the Establishment of the Long-Term Care Insurance System, 中共中央办公厅、国务院办公厅《关于加快建立长期护理保险制度的意见》. +- Notes: The disability has to 「长期持续（一般为6个月以上）」, to have continued for a long time, generally more than 6 months; a few weeks in hospital does not reach the condition for applying. There is also a 6-month waiting period before you can draw the benefit. Costs that are not care, such as bed fees and meal costs, are not paid by long-term care insurance. Medical costs that medical insurance should pay are not paid by it either. If you are already receiving work-injury living and care payments, you do not receive this on top. This money and the care allowance inside the two disability allowances are two sets of formalities and two separate sums, and they do not clash: see item 8 in section 7 (the two allowances). Care at home and in the community is reimbursed at a slightly better rate, so you do not have to go into an institution. The scheme is being put in place region by region, and the assessment standards differ. Ring 12393 or the medical insurance agency in your home area first, and ask whether it has started locally and how to book an assessment. If the assessment is not passed, do not give up: you can apply again once the disability has become worse. + +### 8. If someone at home is bedridden long-term, treat pressure ulcers as the main enemy: get an electric pressure-relief mattress, turn them on a schedule, and look at the bony parts every day + +- Cost: A home electric alternating-pressure air mattress starts at a few hundred yuan. Each turn takes a few minutes. The hard part is that you have to do it every two or three hours, through the night as well, and one person cannot manage it alone +- In plain language: Among older patients who already had pressure ulcers before hip fracture surgery, the chance of dying within 30 days is about 20% higher. Switching to an electric pressure-relief mattress cuts new pressure ulcers by about 60%. Turning the person on schedule, done properly, cuts them by about 70%. Do not leave anyone lying face down for long. Look at the base of the spine, the heels, the hips and the shoulder blades every day, and if you find a red mark that does not fade under pressure, find a nurse or a doctor the same day. +- Benefit: 19520 hip fracture surgery patients (the US NSQIP registry, 2016 to 2019). People who already had pressure ulcers before the operation had a higher proportion of deaths within 30 days. Once the other factors are taken into account, the association still holds (OR 1.2, odds up 21%, P=0.004). These people were also more likely to have other problems: deep vein thrombosis (OR 1.59), pneumonia (OR 1.39) and unplanned readmission (OR 1.43), and their average length of stay was 0.4 days longer. In a follow-up of 324 hospitalised older people (average age 86) the median follow-up was 18 months. Pressure ulcers predicted long-term mortality on their own (HR 1.500, 95% CI 1.059 to 2.126, P=0.020). Predicting on their own means that once the other factors are set aside, they predict by themselves. Pooling 65 randomised trials (network meta-analysis). What is compared is an electric active-air support surface against a standard hospital mattress, in other words a standard support surface. An electric active-air support surface is the kind of electric pressure-relief mattress. People on it had a lower incidence of pressure ulcers (RR 0.42, 95% CI 0.29 to 0.63, moderate certainty). For an electric mixed-air support surface it is RR 0.22 (0.07 to 0.66). A Cochrane review of 11 trials and 4462 people (updated in 2026). Turning every 2 hours against turning every 4 hours makes no difference that can be pinned down (RR 1.05, 95% CI 0.79 to 1.39, very low certainty). Comparisons between the other intervals are just as uncertain. In a trial of 1226 ICU patients, wearable sensors gave real-time reminders and visual alerts, and this actually got the 2-hourly turning done. The incidence of pressure ulcers was lower than with routine care (RR 0.28, 95% CI 0.10 to 0.75, moderate certainty). In a trial of 116 people, the group lying face down had a higher incidence of pressure ulcers than the group lying on the back (RR 4.55, 95% CI 2.31 to 8.98) +- Evidence grade: B +- Sources: Porter SB, Pla R, Chow JH, et al. (2022). Preoperative Pressure Ulcers, Mortality, and Complications in Older Hip Fracture Surgery Patients. JAAOS Global Research & Reviews, 6(11). ; Ottaviani S, Rondanina E, Longo E, et al. (2026). Impact of pressure ulcers and frailty on long-term mortality: a prospective cohort study of hospitalized older adults. European Geriatric Medicine, 17(2), 665-672. ; Shi C, Dumville JC, Cullum N (2018). Support surfaces for pressure ulcer prevention: A network meta-analysis. PLoS ONE, 13(2), e0192707. ; Latimer SL, Chaboyer WP, Probst S, et al. (2026). Repositioning for pressure injury prevention in adults. Cochrane Database of Systematic Reviews, 6, CD009958. +- Notes: There are three reasons for the B. First, the two mortality studies only show that 「people with pressure ulcers did worse afterwards」, which is not the same as 「preventing pressure ulcers brings the death rate down」. Second, the randomised trials on the turning interval rest on very shaky conclusions. Third, the evidence on the mattress is harder, but what the studies compare are hospital mattresses, and applying it to a home mattress is an extrapolation. Check the base of the spine, the heels, the hips and the shoulder blades every day for a red mark that does not fade under pressure; that is routine nursing practice and has no separate literature. For the care services that long-term care insurance pays for, see item 7 in this section. If one leg suddenly swells, treat it as a deep vein thrombosis: see item 11 in section 13. For the period bedridden after a fall or a serious injury, see item 34 in section 1 diff --git "a/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" new file mode 100644 index 00000000..b4f1aa84 --- /dev/null +++ "b/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -0,0 +1,65 @@ +[← 回总目录](../README.md) + +# 18. 养孩子划不划算 + +这一节算的是钱和时间两笔账。和第 10 节算结婚账一样，这里只把账拆开给你看，不替你下结论。已经怀孕、要照流程一步步办事的，看第 27 节。 + +[← Back to the contents](../README.md) + +# 18. Is raising children worth it + +This section works out two accounts, money and time. As in section 10, where the marriage account is worked out, this section only breaks the account down for you and does not draw the conclusion for you. If you are already pregnant and have to follow the procedure step by step, see section 27. + +### 1. Count the money you can receive first: the national child-rearing subsidy is 3600 yuan per child per year, paid until the age of 3 + +- Cost: No money. Apply at the place where the child's hukou is registered, and one trip finishes it +- In plain language: If the child was born on or after 2025-1-1 and is not yet 3 years old, you can receive this money. One of the parents applies at the place where the child's hukou is registered. It is 3600 yuan per child per year, and it is paid until the age of 3. This money is not subject to individual income tax. When you apply for means-tested income support, it is not counted as part of your household's income. +- Benefit: The plan issued by the General Offices of the CPC Central Committee and the State Council provides that for infants born from 2025-1-1 who are under 3 years old, the payment is 「每孩每年 3600 元」, 3600 yuan per child per year. It is claimed by one of the child's parents or by another guardian. This subsidy is 「免征个人所得税」, exempt from individual income tax. When means-tested income support is granted, it 「不计入家庭或个人收入」, is not counted as household or personal income +- Evidence grade: A +- Sources: General Offices of the CPC Central Committee and the State Council (中共中央办公厅、国务院办公厅) (2025). Implementation Plan for the Child-Rearing Subsidy Scheme, 育儿补贴制度实施方案. +- Notes: This is the nationwide minimum. Some provinces and cities have their own local subsidy on top, and you can receive both. Three years come to 10800 yuan, which is a very small part of the cost of raising a child, so do not use it as the reason for deciding whether to have one + +### 2. Maternity leave is 98 days, and the maternity allowance is paid by the maternity insurance fund at the employer's average monthly wage for the previous year + +- Cost: No money. The employer makes the declaration, so you do not have to go anywhere +- In plain language: Giving birth earns you at least 98 days of maternity leave, and 15 of them can be taken before the birth. A difficult birth adds 15 days. Each further child adds another 15 days. The money paid for this period is called the maternity allowance, and the maternity insurance fund pays it. If your employer has not enrolled you in maternity insurance, the employer pays it out of its own money. +- Benefit: The number of days is set by the State Council. 「女职工生育享受 98 天产假，其中产前可以休假 15 天；难产的，增加产假 15 天；生育多胞胎的，每多生育 1 个婴儿，增加产假 15 天。」 — a female employee who gives birth is entitled to 98 days of maternity leave, of which 15 may be taken before the birth; a difficult birth adds 15 days; and for a multiple birth each additional baby adds 15 days. The money received during maternity leave is called the maternity allowance. Where you are already enrolled in maternity insurance, it is 「按照用人单位上年度职工月平均工资的标准由生育保险基金支付」, paid by the maternity insurance fund at the standard of the employer's average monthly employee wage in the previous year. If you are not enrolled, the employer pays it at your wage level before the maternity leave +- Evidence grade: A +- Sources: State Council (国务院) (2012). Special Provisions on Labour Protection for Female Employees, 女职工劳动保护特别规定（国令第 619 号，第七条、第八条）. +- Notes: 98 days is the minimum set by the state. Most provinces add a reward leave of more than 60 days for childbirth in their population and family planning regulations, so 158 days in total is the usual starting point. Which party pays the wages during that reward leave is set differently in each province + +### 3. Know this rule: your pay may not be cut and you may not be dismissed because of pregnancy, childbirth or breastfeeding + +- Cost: No money +- In plain language: During pregnancy, childbirth and breastfeeding, an employer that cuts your pay or dismisses you is breaking the law, and it is fined 1000 to 5000 yuan for each person. While the child is under one year old you have one hour a day for breastfeeding, and the employer may not arrange overtime or a night shift for you. For how to claim, see section 19. +- Benefit: The State Council provides: 「用人单位不得因女职工怀孕、生育、哺乳降低其工资、予以辞退、与其解除劳动或者聘用合同。」 — an employer may not reduce a female employee's pay, dismiss her, or terminate her employment or engagement contract because of pregnancy, childbirth or breastfeeding. In these three periods the employer may not cut your pay, dismiss you or end the contract. A female employee breastfeeding a baby under 1 year old has one hour of breastfeeding time each day. The employer may also not arrange for her to work longer hours, and may not arrange a night shift. An employer that breaks any of the above is fined from 1000 yuan to 5000 yuan for each person +- Evidence grade: A +- Sources: State Council (国务院) (2012). Special Provisions on Labour Protection for Female Employees, 女职工劳动保护特别规定（国令第 619 号，第五条、第九条、第十三条）. +- Notes: If you are really dismissed in breach of the law, the route to a claim is in section 19. The evidence to collect is the written notice of a transfer to another post or a pay cut, together with the chat records, and you should keep both + +### 4. Count the time account by 「whose time is taken up」, not by 「how hard it is」 + +- Cost: No money; what it costs is time. Before the age of 3 the child takes up nearly the whole day. The hard part is that it is every single day, and there is no grinding it out for a few days and being finished +- In plain language: Convert the hours spent looking after the child each day and the number of times you are woken at night into time, and multiply by the hourly wage of the person doing the work. The figure that comes out is usually far larger than the money for formula and nappies. So it is best to settle who does what before the birth. +- Benefit: Convert the time spent looking after the child each day and the number of times you are woken at night into hours, and multiply by the hourly wage of the person doing the work. The number that comes out is usually far larger than the money for formula and nappies. Work that nobody pays for, such as looking after the child, cooking and getting up in the night, is counted into the cost in the same way. The method is the same as the one section 10 uses for housework. Settle who does what first, and only then decide whether to have a child +- Evidence grade: C +- Sources: the author's own experience, no direct literature ; for the measure used see section 10 +- Notes: The figures do not have to be exact. What is useful is that the two of you sit down with the same table before the birth and agree who does what, so that there is no argument afterwards + +### 5. Split the money account into three stages: 0 to 3 years old, compulsory education, and after compulsory education + +- Cost: There is no fixed figure for what it costs; different cities and different ways of looking after the child are far apart. The hard part is sitting down quietly and working out the three stages separately, instead of guessing a number +- In plain language: For 0 to 3 years old the money goes mainly on how the child is looked after: by you, by the grandparents, or by a paid carer. From primary school to middle school it goes mainly on housing and on activities outside lessons. After that it goes mainly on the route into higher education. The three stages are spent in completely different ways, and mixing them only produces one alarming total. +- Benefit: The main cost from 0 to 3 years old is the way the child is looked after: by you, by the grandparents, or by a paid carer. In the compulsory education stage, primary school to middle school, the main costs are housing and activities outside lessons. After that the main cost is the route into higher education. Where the money goes is completely different across the three stages, and working them out together only gives one alarming total +- Evidence grade: C +- Sources: the author's own experience, no direct literature +- Notes: First count the money you are certain to receive, such as the child-rearing subsidy, the maternity allowance and medical insurance reimbursements, and only then see what is still missing. Do not base a decision on the 「how many tens of thousands of yuan it takes to raise a child」 figures that circulate online; the measure behind them is not stated, so you do not know what they counted or what they left out + +### 6. Having a child for the sake of the older generation, for the sake of the marriage, and for the sake of your own old age: keep a separate account for each + +- Cost: No money; what it costs is one serious conversation. The hard part is saying it straight rather than going round it +- In plain language: Other people hoping you have a child can be one factor in your decision. But the time and the money are entered in your own account. 「Raising a child to support you in old age」 has to be costed on its own, because it is a bet on another person's financial capacity and willingness twenty years from now. Write down the consequences of not having a child as well, and set the two side by side. +- Benefit: The line of thought is the same as in section 10. Other people's expectations can be one factor in your decision. But the cost goes in your own account, not in theirs. 「Raising a child to support you in old age」 above all has to be costed on its own, because it assumes another person's financial capacity and willingness twenty years from now +- Evidence grade: C +- Sources: the author's own experience, no direct literature ; for a comparable analysis see section 10 +- Notes: Write down 「the consequences of not having a child」 as well, and look at it side by side with 「the consequences of having a child」 diff --git "a/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" "b/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" new file mode 100644 index 00000000..34d12df2 --- /dev/null +++ "b/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" @@ -0,0 +1,164 @@ +[← 回总目录](../README.md) + +# 19. 在职、离职和工伤 + +这一节算的是钱：钱怎么算、字怎么签、期限怎么算。前三条讲你还在岗时就该拿到手的钱，中间六条讲离职，最后七条讲职业病和工伤。工伤的钱比被裁大一个量级，期限也更硬。职业病造成的损伤换不回来。 + +[← Back to the contents](../README.md) + +# 19. Working, leaving, and work injuries + +This section is about money: how it is calculated, how the documents are signed, and how the time limits run. The first three items cover the money you should be getting while you are still employed, the middle six cover leaving the job, and the last seven cover occupational disease and work injuries. The money for a work injury is an order of magnitude larger than for a dismissal, and the time limits are firmer. Damage caused by an occupational disease cannot be undone. + +### 1. Overtime pay runs in three bands at 1.5, 2 and 3 times, and if it is not paid you complain to the labour inspectorate, with a further 50% to 100% added if the deadline passes + +- Cost: No money. As a matter of course, keep a copy of your clock-in records, the shift roster and any overtime approval form, and screenshot the messages where your manager assigns work in the work chat. The hard part is remembering to save them every month. +- In plain language: Ordinary overtime is paid at 1.5 times your wage. Overtime on a rest day, where you were not given time off in lieu, is paid at 2 times. Working on a statutory holiday such as New Year's Day, Spring Festival, Labour Day or National Day is paid at 3 times, and time off in lieu cannot be used instead. If the company does not pay, complain to the labour inspectorate. If it still does not pay after being ordered to pay within a deadline, the employer has to pay an additional half to one times the arrears. +- Benefit: The normal working time set by the law is no more than 8 hours a day and, on average, no more than 44 hours a week. Ordinary overtime is generally no more than 1 hour a day. Where there is a special reason to work longer, it is no more than 3 hours a day, and no more than 36 hours in a month added up. Overtime pay falls into three bands. When ordinary working hours are extended, it is not less than 150% of your wage. When you work on a rest day and time off in lieu cannot be arranged, it is not less than 200%. When you work on a statutory holiday, it is not less than 300%. In the statutory-holiday band the article offers no option of substituting time off in lieu. Where the employer arranges overtime and does not pay for it, the labour administration orders it to pay within a deadline. If it still does not pay after the deadline, the labour administration orders it to pay the worker a further compensation of not less than 50% and not more than 100% of the amount payable, on top of the arrears (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). Labour Law, 劳动法（2018 年第二次修正，第三十六、四十一、四十三、四十四条）. (published by the State Administration for Market Regulation) ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第八十五条第三项）. +- Notes: Rest days and statutory holidays are counted separately. The statutory holidays are only the few fixed days in a year, such as New Year's Day, Spring Festival, Labour Day and National Day. A long weekend created by shifting a day off is counted as a rest day, and is paid at 200%. 「The company requires approval for overtime」 and 「I stayed of my own accord」 do not mean the company can pay nothing, but they make your evidence harder to produce, so keep the assignment messages and the approval records. If the company wants to use the two special working-time systems, the comprehensive working-hours system and the irregular working-hours system, it needs the labour administration's approval first; it is not the company's own decision. For the time limit on arbitration of a dispute, see item 19 in section 8 + +### 2. Annual leave is 5, 10 or 15 days according to total length of service, and days you cannot take are paid at 300% of your daily wage + +- Cost: No money. Add up the years you have worked for all employers, and your social insurance contribution record is the proof. The hard part is refusing to sign the form 「voluntarily giving up annual leave」 when the company hands it to you. +- In plain language: When the years you have worked for all employers add up to at least 1 year and less than 10, you get 5 days of annual leave a year. At 10 years it is 10 days, and at 20 years it is 15 days. Changing jobs does not stop the total from carrying over. Where the employer cannot arrange the leave and you agree to do without it, those days are paid at 3 times your daily wage. Once the form 「voluntarily giving up annual leave」 is signed, only your ordinary wage is left. +- Benefit: You get annual leave after 12 months of continuous work. The number of days is set by your **cumulative** working time, adding up the years you have worked for all employers. At least 1 year and less than 10 gives 5 days, at least 10 years and less than 20 gives 10 days, and 20 years gives 15 days. Statutory holidays and rest days do not count towards annual leave. Where the employer genuinely cannot arrange the leave because of work and has your own agreement not to take it, the days you did not get are paid 「按照该职工日工资收入的 300% 支付年休假工资报酬」, as annual leave pay at 300% of that employee's daily wage. That 300% already includes the one share you would have got for a normal working day, so what you actually receive extra is 200%. Daily wage = your own monthly wage ÷ 21.75. The monthly wage is the average for the 12 months before this money is paid, with overtime pay taken out. If you joined the company during the year, it is pro-rated by the calendar days left in that company. On leaving, it is pro-rated and settled by how much of the year you have worked, and leave already taken in excess is not clawed back. Where the employer neither arranges the leave nor pays, an order to correct it within a deadline is issued. If it is not corrected after the deadline, a further compensation of the same amount is added to the annual leave pay (nationwide, in force from 2008-1-1) +- Evidence grade: A +- Sources: State Council (国务院) (2007). Regulations on Paid Annual Leave for Employees, 职工带薪年休假条例（国务院令第 514 号，第二、三、四、五、七条）. ; Ministry of Human Resources and Social Security (人力资源和社会保障部) (2008). Measures for the Implementation of Paid Annual Leave for Enterprise Employees, 企业职工带薪年休假实施办法（人社部令第 1 号，第三、四、五、十、十一、十二、十五条）. +- Notes: Cumulative working time has to be added up across employers, so if you have changed jobs do not use only the length of service with your current employer. **Only where the employee 「因本人原因且书面提出不休」, does not take the leave for their own reasons and asks in writing, may the employer pay only the ordinary wage**, so do not sign the form 「voluntarily giving up annual leave」. If you take personal leave totalling more than 20 days and the employer does not dock your pay, you get no annual leave that year. A sick leave longer than the corresponding number of months also disqualifies you: less than 10 years of service means 2 months, 10 to 20 years means 3 months, and more than 20 years means 4 months. Home-visit leave, marriage and bereavement leave, maternity leave and the work-injury period of paid recovery do not count towards annual leave. + +### 3. The probation period has a legal ceiling, the wage may not be below 80%, and it can only be agreed once + +- Cost: No money. Before you sign, spend a few minutes matching the term written in the contract against the length of the probation period. +- In plain language: For a contract of 1 year or less, the probation period is at most 1 month. For 1 to 3 years it is at most 2 months. For 3 years or more it is at most 6 months. The same company may only put you on probation once, and a second probation after you have passed and been moved to another post is unlawful. The probation wage may not be below the lowest grade for the same post or 80% of the wage agreed in the contract, and it may not be below the local minimum wage. +- Benefit: For a contract of at least 3 months and less than 1 year, the probation period may not exceed 1 month. For at least 1 year and less than 3 years it may not exceed 2 months. For 3 years or more, and for an open-ended contract with no end date, it may not exceed 6 months. A contract that ends when one job is finished, or that runs for less than 3 months in total, may not have a probation period agreed at all. 「同一用人单位与同一劳动者只能约定一次试用期」, the same employer and the same employee may agree a probation period only once, so one company puts one person on probation one time only. Where the contract mentions only a probation period, that probation period does not stand, and the period stated is the term of the employment contract. The probation wage 「不得低于本单位相同岗位最低档工资或者劳动合同约定工资的百分之八十，并不得低于用人单位所在地的最低工资标准」, may not be below the lowest grade wage for the same post in the employer or eighty per cent of the wage agreed in the employment contract, and may not be below the minimum wage standard where the employer is located. Where the probation period agreed goes beyond the legal length and the person in fact worked, the employer has to pay a separate compensation, at the monthly wage standard applying after the probation period, for the excess period. An employer that wants to terminate the contract during probation may only do so in the cases the law lists, and it has to give the employee its reasons (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第十九、二十、二十一、八十三条）. +- Notes: The three commonest unlawful practices are these. First, a contract of only 1 year with a 3-month probation period agreed. Second, moving the person to another post after they have passed and running a second probation period. Third, 「no social insurance during probation」, because the obligation to enrol starts on the first day of work and has nothing to do with probation; for the employer's side of the obligation see item 16 in section 12. Being dismissed during probation does not mean getting nothing at all; it depends on whether the reason the employer gives is one of those the law allows. Where the termination is unlawful, the 2N in item 6 applies. + +### 4. Work out N before a dismissal: one month's wage for each full year, and half a month for under six months + +- Cost: No money. One calculation on a calculator, ten minutes. +- In plain language: The economic compensation due on dismissal is worked out like this: a full year of service gives one month's wage, six months to a year counts as a full year, and under six months gives half a month. The monthly wage here is the average over the 12 months before you leave, and bonuses, allowances and subsidies all count. If the wage is above three times the local previous-year average monthly wage, it is capped at three times, and the length of service counts for at most 12 years. +- Benefit: The economic compensation set by the Labour Contract Law is the money the employer owes on a dismissal, and it is commonly called N. The method is 「每满一年支付一个月工资……六个月以上不满一年的，按一年计算；不满六个月的，向劳动者支付半个月工资」, one month's wage for each full year of service, six months to a year counted as a full year, and half a month's wage where it is under six months. If the monthly wage is above three times the local previous-year average monthly employee wage, the calculation is capped at three times, and the length of service counts for at most twelve years +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law of the People's Republic of China (in force from 2008-1-1), 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第四十七条）. +- Notes: Monthly wage means the average wage over the twelve months before the employment contract is terminated. Bonuses, allowances and subsidies all count, and it is not only the basic wage. This is the point companies most easily cut down. + +### 5. If the company dismisses you without 30 days' notice, it has to pay one more month's wage + +- Cost: No money. +- In plain language: If the company lets you go without 30 days' written notice, it has to give you one month's wage on top of N, which is what is usually called N+1. When you negotiate, list that extra month separately, and do not let the company fold it into N. +- Benefit: The Labour Contract Law gives an employer terminating a contract under that article two options. One is 「提前三十日以书面形式通知劳动者本人」, to notify the worker in person in writing thirty days in advance. The other is 「额外支付劳动者一个月工资后，可以解除劳动合同」, to pay the worker one month's wage and then terminate the contract. If they want you gone at once, they have to find the extra month's wage. That extra month is usually called payment in lieu of notice, and it is counted separately from N +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law of the People's Republic of China (in force from 2008-1-1), 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第四十条）. +- Notes: So the usual outcome of a lawful redundancy is N+1. In the negotiation, work out N and the 1 separately, and do not let the company count the extra month inside N + +### 6. Where the company terminates the contract unlawfully, the compensation is twice the economic compensation standard + +- Cost: No money. Labour arbitration takes several months of waiting. The hard part is that you have to keep it moving for those months and cannot drop it halfway. +- In plain language: Where the company terminates the contract unlawfully, the compensation is the economic compensation standard doubled, which is what is usually called 2N. Whether the termination is unlawful depends on the particular circumstances. Ask a legal aid service or a lawyer first, and only then decide whether to negotiate with the company or go to arbitration. +- Benefit: The Labour Contract Law provides that where an employer terminates or ends an employment contract in breach of the Law, it 「应当依照本法第四十七条规定的经济补偿标准的二倍向劳动者支付赔偿金」, shall pay the employee compensation at twice the economic compensation standard set by Article forty-seven of the Law. The method is to work out one share on the economic compensation scale and then double it for you +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law of the People's Republic of China (in force from 2008-1-1), 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第八十七条）. +- Notes: This is what is usually called 2N. Whether a termination is unlawful depends on the particular circumstances. Consult a legal aid service or a lawyer first, then decide whether to negotiate or go to arbitration; the route is in section 7 + +### 7. Do not sign 「resignation for personal reasons」, because the moment you sign there is no N + +- Cost: No money. The hard part is standing up to the pressure to sign on the spot. +- In plain language: If you sign a statement that you are resigning for personal reasons, the economic compensation and the unemployment insurance benefit usually both fall away. Resigning of your own accord is not one of the cases in which compensation is payable, except where the employer is at fault in a way the law lists, such as unpaid wages or failure to enrol in social insurance. You do not have to sign on the day; sleep on it and decide in the morning. +- Benefit: The Labour Contract Law limits the cases in which economic compensation is payable to the few categories the law lists. An employee who leaves of their own accord is not among them, unless it is because of a legal fault by the employer such as unpaid wages or failure to enrol in social insurance. Once you have signed 「voluntary resignation」, the economic compensation and the unemployment insurance benefit usually both fall away +- Evidence grade: B +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law of the People's Republic of China (in force from 2008-1-1), 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第三十八条、第四十六条）. +- Notes: The line companies use is 「sign first so the process can move on, the compensation comes separately」. If you want it, have it written into the agreement on the spot; if it is not written down, do not sign. Not signing on the day has no legal consequence, so sleep on it and decide in the morning + +### 8. Before you leave, save your payslips, attendance records, employment contract, social insurance record and chat logs + +- Cost: No money. Half an hour to save the material. The hard part is remembering to do it before you go. +- In plain language: Once the computer and the accounts are handed back, the payslips, attendance records, contract, social insurance record and chat logs are gone for good. At arbitration, what you earned and whether you worked overtime still rest on the material in your own hands. Save your own employment records; do not take away the company's source code or customer list. +- Benefit: Once the computer and the accounts are handed back, this material cannot be obtained any more. In arbitration the burden of proof leans towards the employer, but what you earned and whether you worked overtime still rest on the evidence in your own hands +- Evidence grade: C +- Sources: the author's own experience, no direct literature ; for the route to enforce your rights see section 7 +- Notes: What you save is your own employment records. Do not take the company's source code, customer list or technical documents while you are at it; for the risk of taking them see section 11 + +### 9. Do two things as soon as you leave: register as unemployed and claim the unemployment insurance benefit, and read the non-compete agreement + +- Cost: No money. +- In plain language: Do two things as soon as you have left. First, register as unemployed and claim the unemployment insurance benefit. Second, read the non-compete agreement. The benefit runs from the day you register, so a month late means a month less. Only people who lost their job rather than choosing to leave can receive it, which is another reason for item 7 in this section (do not sign 「resignation for personal reasons」). +- Benefit: There are four conditions for receiving the unemployment insurance benefit: a full year of unemployment insurance contributions, losing the job rather than choosing to leave, which is 非因本人意愿中断就业, an interruption of employment not caused by the person's own will, having registered as unemployed, and still looking for work. The years you have paid decide whether the maximum is 12, 18 or 24 months. For how to claim online, see item 1 in section 7 (claim the unemployment insurance benefit online once you are unemployed). A non-compete restriction lasts at most 2 years. Through those 2 years the company has to pay you compensation every month. Where the contract does not say how much, it is 30% of your average wage over the 12 months before you left, and it may not be below the local minimum wage. Only where the company has not paid for 3 consecutive months for its own reasons may you ask for this agreement to be terminated, see item 12 in section 11 (what to do if you signed a non-compete agreement and the company does not pay the compensation). The non-compete restriction has a time limit, but the parts you still have to keep to stay in force until then +- Evidence grade: C +- Sources: the author's own experience, no direct literature ; the figures cited come from State Council (国务院) (1999). Regulations on Unemployment Insurance, 失业保险条例（国务院令第 258 号，第十四、十六、十七条：「失业保险金自办理失业登记之日起计算」）. and Standing Committee of the National People's Congress (全国人大常委会) (2012). Labour Contract Law, 劳动合同法（2012 年修正，第二十三、二十四条）. (reprinted by the Fushun Municipal Emergency Management Bureau) +- Notes: It is graded C because the action of 「registering on the day you leave」 has no direct literature; the figures in the sources field are the words of the regulations themselves. The unemployment benefit does not expire, but it runs from the day you register, so an earlier registration means one more day of money. For the claim materials and the online entry point, see section 7. For the compensation standard and the conditions for ending a non-compete restriction, see section 11. Signing 「resignation for personal reasons」 does not only cost you N, it loses the unemployment insurance benefit as well; see item 7 in this section (do not sign 「resignation for personal reasons」) + +### 10. Before you take a post with dust, noise or chemicals, check whether the contract states the hazards; the three occupational health examinations are arranged and paid for by the employer + +- Cost: No money. When you sign, read through the part of the contract on 「notification of occupational disease hazards」 clause by clause. Keep your own copy of every examination report. The hard part is asking in person and getting the documents into your hands in person. +- In plain language: When you sign, the employer must state truthfully in the contract what occupational disease hazards the post carries, how to guard against them, and what benefits attach. If it hides them from you, you can refuse to do the work, and the employer may not dismiss you for that. The three occupational health examinations, before you start, while you are in the post, and when you leave, are arranged and paid for by the employer. Without the examination on leaving, the employer may not terminate the contract. +- Benefit: The Law on the Prevention and Control of Occupational Diseases provides that when the employment contract is signed the employer has to make four things clear: what occupational disease hazards this work may produce, what consequences follow, how to guard against them, and what benefits attach. The words of the article are 「如实告知劳动者，并在劳动合同中写明，不得隐瞒或者欺骗」, tell the worker truthfully, write it into the employment contract, and do not conceal or deceive. If after the contract is signed you are moved to harmful work the contract did not describe, the employer has to tell you truthfully again and agree the contract terms with you. Where the employer fails to do this, 「劳动者有权拒绝从事存在职业病危害的作业，用人单位不得因此解除与劳动者所订立的劳动合同」, the worker has the right to refuse work that carries occupational disease hazards, and the employer may not terminate the contract on that ground. There are three examinations in all: before starting, while in the post, and on leaving. 「职业健康检查费用由用人单位承担」, the cost of the occupational health examination is borne by the employer, so the employer pays and the results are given to you in writing. A person who has not had the examination before starting may not be put on work involving exposure to occupational disease hazards. A person whose physical condition is unsuited to this work, and who therefore has a contraindication, may not be put on the contraindicated work either. Where health harm related to the job is found, the person has to be moved off the post and properly placed elsewhere. **「对未进行离岗前职业健康检查的劳动者不得解除或者终止与其订立的劳动合同」, an employment contract may not be terminated or ended with a worker who has not had the occupational health examination before leaving.** On leaving you 「有权索取本人职业健康监护档案复印件，用人单位应当如实、无偿提供，并在所提供的复印件上签章」, have the right to ask for a copy of your own occupational health monitoring file, and the employer must supply it truthfully and free of charge and put its seal on the copy. The employer has to make a true photocopy of that occupational health monitoring file, give it to you free, and stamp it (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). Law on the Prevention and Control of Occupational Diseases, 职业病防治法（2018 年修正，第三十三、三十五、三十六条）. (reprinted by the Beijing Municipal People's Government) +- Notes: Three questions at the interview tell you roughly whether the company is legitimate: does this post carry occupational disease hazards, does the contract say so, and who pays for the examination before you start. If they cannot answer, or say 「just start and we will see」, weigh the risk yourself. The examination on leaving is the one most often forgotten: without it, the employer may not terminate or end the employment contract. That examination is also the key document for applying for an occupational disease diagnosis later, and the same goes for the copy of the file, so get it into your hands before you go. An occupational disease is handled as a work injury, and the benefits are covered from item 12 onwards. + +### 11. Damage from dust, noise and toxic chemicals is irreversible: the employer must supply protective equipment, and you may refuse work with no protection + +- Cost: No money. The hard part is wearing a particle mask, earplugs and goggles all the time, and refusing unprotected work, and the price of that is being called 「fussy」 by colleagues. +- In plain language: Pneumoconiosis, noise-induced deafness and chemical eye burns are all occupational diseases in law, and once the lungs or the hearing are damaged there is no way back. You have the right to know what hazards the post carries and the right to demand protective facilities and protective equipment from the employer. You do not have to do unprotected work. If the employer cuts your pay or dismisses you for that, it does not count. +- Benefit: From 2025-8-1 the occupational diseases in law are 12 major classes and 135 kinds, including 4 open clauses. The first class is occupational pneumoconiosis. It includes silicosis, coal workers' pneumoconiosis, graphite pneumoconiosis, carbon black pneumoconiosis, asbestosis and talc pneumoconiosis. It also includes cement pneumoconiosis, mica pneumoconiosis, potter's pneumoconiosis, aluminium pneumoconiosis, welder's pneumoconiosis and foundry worker's pneumoconiosis, plus one open clause. Occupational diseases of the ear, nose, throat and mouth include noise-induced deafness and blast-induced deafness. The occupational eye diseases include chemical eye burns and photochemical keratitis. All of this damage is irreversible, and no treatment can repair the lungs or the hearing. The law gives you four rights you can use on the spot. First, to know what occupational disease hazards are present in this workplace, what consequences they cause, and what protective measures should be taken. Second, to require the employer to provide qualified protective facilities and personal protective equipment and to improve the working conditions. Third, to criticise, report and accuse practices that break the laws on the prevention and control of occupational diseases, and practices that endanger life and health. Fourth, to 「拒绝违章指挥和强令进行没有职业病防护措施的作业」, refuse unlawful instructions and orders to carry out work with no occupational disease protection; you do not have to take an instruction that breaks the rules or take unprotected work. And 「因劳动者依法行使正当权利而降低其工资、福利等待遇或者解除、终止与其订立的劳动合同的，其行为无效」, where an employer reduces a worker's wages or benefits, or terminates or ends the employment contract, because the worker lawfully exercised a legitimate right, that act is void. If it cuts your pay, cuts your treatment or dismisses you because you used these rights, it does not count (nationwide) +- Evidence grade: A +- Sources: National Health Commission (国家卫生健康委), Ministry of Human Resources and Social Security (人力资源社会保障部), National Disease Control and Prevention Administration (国家疾控局) and All-China Federation of Trade Unions (中华全国总工会) (2024). Classification and Catalogue of Occupational Diseases, 职业病分类和目录（2025 年 8 月 1 日起实施）. ; for the count of 12 major classes and 135 kinds in the new catalogue see National Health Commission (国家卫生健康委) (2024). The new catalogue adjusts occupational diseases to 12 major classes and 135 kinds, 新版目录将职业病调整为 12 大类 135 种. ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Law on the Prevention and Control of Occupational Diseases, 职业病防治法（第三十九条）. (reprinted by the Beijing Municipal People's Government) +- Notes: You have to recognise the right mask: only a particle respirator stops dust, and an ordinary gauze mask or a disposable surgical mask does not. If the employer does not supply earplugs for a noisy post, buy your own; a pair costs a few yuan, and lost hearing does not come back. 「Work two years, save up enough and leave」 does not hold with pneumoconiosis. Silicosis can appear many years after you stop working with dust, and it keeps getting worse. That is why the examination on leaving matters so much (see item 10 in this section, the three occupational health examinations arranged and paid for by the employer). The new catalogue also adds two classes. One is occupational musculoskeletal disorders, where the entry is carpal tunnel syndrome, limited to manufacturing workers who use the wrist repeatedly for long hours or do heavy work. The other is occupational mental and behavioural disorders, where the entry is post-traumatic stress disorder, limited to police, health workers, fire and rescue and other emergency response workers who take part in handling sudden incidents. For what to do when a chemical splashes on the skin, see item 21 in section 13. + +### 12. If you are hurt at work or hit on the way to or from work, the first step is to get it recognised as a work injury, and if the employer does not report it you do + +- Cost: No money, and one trip to the social insurance administration. You have to bring three documents: the application for recognition as a work injury, proof of the employment relationship, and a medical diagnosis certificate. +- In plain language: An injury at work counts as a work injury. Being hit on the way to or from work, where the main responsibility is not yours, also counts. But without recognition as a work injury, the medical costs, the wages paid during recovery and the disability allowance all fall away. The employer has to report within 30 days. If the employer does not, you, or your family, or the trade union can report it yourself within one year to the social insurance administration. +- Benefit: Without recognition as a work injury, the medical costs, the wages paid during recovery, which are called 停工留薪期工资, and the disability allowance all fall away. Where the employer does not report within 30 days, it pays the work-injury benefits due for that period out of its own pocket +- Evidence grade: A +- Sources: State Council (国务院) (2010 revision). Regulations on Work-Related Injury Insurance (State Council Order No. 586), 工伤保险条例（国务院令第 586 号）. Article fourteen lists seven cases that 「应当认定为工伤」, must be recognised as a work injury, including 「（六）在上下班途中，受到非本人主要责任的交通事故或者城市轨道交通、客运轮渡、火车事故伤害的」, on the way to or from work, suffering harm in a traffic accident, urban rail accident, passenger ferry accident or train accident where the main responsibility is not the person's own. Article seventeen: 「所在单位应当自事故伤害发生之日或者被诊断、鉴定为职业病之日起30日内，向统筹地区社会保险行政部门提出工伤认定申请」, the employer where the person works shall, within 30 days of the day the injury happens or of the day the occupational disease is diagnosed or determined, apply to the social insurance administration of the pooling region for recognition as a work injury ; 「用人单位未按前款规定提出工伤认定申请的，工伤职工或者其近亲属、工会组织在事故伤害发生之日或者被诊断、鉴定为职业病之日起1年内，可以直接向用人单位所在地统筹地区社会保险行政部门提出工伤认定申请」, where the employer does not apply as the previous paragraph requires, the injured worker, or their close relatives or trade union organisation may, within 1 year of the day the injury happens or of the day the occupational disease is diagnosed or determined, apply direct to the social insurance administration of the pooling region where the employer is located ; 「用人单位未在本条第一款规定的时限内提交工伤认定申请，在此期间发生符合本条例规定的工伤待遇等有关费用由该用人单位负担」, where the employer does not submit the application within the time limit in the first paragraph, the employer bears the work-injury benefit costs incurred in the meantime that fall under these Regulations. Article nineteen: 「职工或者其近亲属认为是工伤，用人单位不认为是工伤的，由用人单位承担举证责任」, where the worker or their close relatives consider it a work injury and the employer does not, the burden of proof is on the employer. Article twenty: 「自受理工伤认定申请之日起60日内作出工伤认定的决定」, the decision recognising a work injury is made within 60 days of accepting the application. +- Notes: 1 year is a hard deadline, and after it there is no application to be made. If the employer says it is not a work injury, the burden is on the employer to prove it. You still have to keep material: photographs from the day, colleagues' contact details, the medical records and the attendance record. Article sixteen lists three cases that are not recognised: intentional crime, intoxication or drug use, and self-injury or suicide. Article fifteen adds 「视同工伤」, treated as a work injury: sudden illness causing death during working hours at the workplace, or death despite rescue within 48 hours. + +### 13. Do not believe 「getting to my desk counts as a work injury」: if you feel suddenly unwell, call 120 first, and do not hurry to clock in + +- Cost: No money. +- In plain language: Hurrying to the company when you feel suddenly unwell will not get you a single extra yuan. Being treated as a work injury requires the illness to begin during working hours and at the workplace, so falling ill on the way without arriving is not covered in the first place. Nor is it covered if rescue goes beyond 48 hours. The few dozen minutes you lose trying to meet the conditions are paid for out of your own chance of survival. +- Benefit: Being treated as a work injury requires that the person 「在工作时间和工作岗位」, during working hours and at the workplace, dies of sudden illness, or dies despite rescue within 48 hours. Delaying medical care in order to meet those two conditions costs you your own chance of being alive, and the compensation you buy with it may not even reach you +- Evidence grade: A +- Sources: Regulations on Work-Related Injury Insurance, article fifteen, paragraph one, item (one): 「在工作时间和工作岗位，突发疾病死亡或者在48小时之内经抢救无效死亡的」 is treated as a work injury ; article fourteen, item (six) only lists as a work injury the case of suffering harm in a traffic accident, urban rail accident, passenger ferry accident or train accident on the way to or from work where the main responsibility is not the person's own, and sudden illness on the way is not in it. +- Notes: Both conditions have to be met together: during working hours, and at the workplace. Falling ill on the way without arriving does not apply. If someone falls ill at the post and is taken home or to hospital and dies despite rescue within 48 hours, it can still apply. What counts is where the person was when the illness began, not where they finally died. Short videos put this the wrong way round when they say 「get to the company and there is money」. What changes the outcome is whether those few dozen minutes after the illness began were spent in hospital. And 48 hours is a hard line: rescue going beyond 48 hours does not count instead. + +### 14. If the employer has not enrolled you in work-injury insurance, the benefits still exist, and the employer pays the same amounts in full + +- Cost: No money. If the employer does not accept it, you have to go to arbitration or even to court, which runs from several months. The hard part is that you have to keep it moving for those months. +- In plain language: If the employer has not enrolled you in work-injury insurance, the benefits still exist. They are simply paid in full by the employer on the same items and the same standards, with nothing missing. Not enrolling adds a penalty on the employer: an order to make up the contributions, a late payment charge of five hundredths of one per cent per day, and if it is still not paid, a fine of 1 to 3 times the amount outstanding. 「The company bought no work-injury insurance, so it cannot pay」 is untrue. +- Benefit: The regulations say plainly that where an employer has not enrolled its employees in work-injury insurance and an employee suffers a work injury, that employer pays the money on the items and the standards the regulations set out. Which payments are due and how much each is exactly the same as when insurance is in force, and only the payer changes to the employer itself +- Evidence grade: A +- Sources: Regulations on Work-Related Injury Insurance, article sixty-two, paragraph two: 「依照本条例规定应当参加工伤保险而未参加工伤保险的用人单位职工发生工伤的，由该用人单位按照本条例规定的工伤保险待遇项目和标准支付费用。」 The first paragraph of the same article orders the employer to enrol and make up the contributions within a deadline, and says 「自欠缴之日起，按日加收万分之五的滞纳金；逾期仍不缴纳的，处欠缴数额1倍以上3倍以下的罚款」. +- Notes: 「The company bought no work-injury insurance for you, so it cannot pay」 is the commonest untrue sentence in this kind of case. Not enrolling only adds a penalty to the employer; your benefits do not shrink. + +### 15. Once the injury is stable, get a labour capacity assessment; the disability grade converts straight into money + +- Cost: No money. Going through the labour capacity assessment takes a stretch of waiting before and after. +- In plain language: Once the injury is stable, have the labour capacity assessed, and the grade converts straight into money. The one-off disability payment is worked out on your own wage: grade five 18 months, grade six 16 months, grade seven 13 months, grade eight 11 months, grade nine 9 months, grade ten 7 months. Before the assessment result comes out, do not sign a 「lump-sum settlement」 agreement. +- Benefit: The one-off disability payment converts the disability grade into a number of months of wage and is paid in a single sum: five 18 months, six 16 months and seven 13 months of the person's own wage, and eight 11 months, nine 9 months and ten 7 months of the person's own wage +- Evidence grade: A +- Sources: Regulations on Work-Related Injury Insurance, article twenty-one: 「经治疗伤情相对稳定后存在残疾、影响劳动能力的，应当进行劳动能力鉴定」, where after treatment the injury is relatively stable and there is a disability affecting the ability to work, a labour capacity assessment shall be carried out ; article twenty-two: 「劳动功能障碍分为十个伤残等级，最重的为一级，最轻的为十级」, loss of labour function is divided into ten disability grades, the most serious being grade one and the least grade ten ; article thirty-six, the benefits for grades five and six, being a one-off disability payment of 18/16 months of the person's own wage and a monthly disability allowance of 70%/60% of the person's own wage ; article thirty-seven, the benefits for grades seven to ten, being 13/11/9/7 months of the person's own wage. +- Notes: At grades five and six the employment relationship is kept, the person is still an employee of that employer, and the disability allowance is paid monthly. For grades seven to ten, when the contract ends or the person proposes to terminate it, there are two other payments: the one-off work-injury medical subsidy, paid by the fund, and the one-off employment subsidy, paid by the employer, with the amount set by the province. Do not sign a 「lump-sum settlement」 agreement before the assessment result comes out. + +### 16. There are three payments on a death at work, and you have to tell them apart: the funeral subsidy, the pension for dependants, and the one-off death benefit + +- Cost: No money, and one set of formalities. +- In plain language: A death at work brings three payments, and do not negotiate only one of them. The funeral subsidy is 6 months of the local previous-year average monthly employee wage. The pension for dependants is paid monthly on the deceased's own wage, at 40% a month for a spouse and 30% a month for each other dependant. The one-off death benefit is 20 times the previous year's national per-capita disposable income of urban residents, and that single figure applies nationwide. +- Benefit: The money on a death at work is in three parts. The first is the funeral subsidy, 6 months of the previous-year average monthly employee wage in the area where the person is insured, which is called the pooling region. The second is the pension for dependants, paid monthly on the deceased's own wage, at 40% a month for a spouse and 30% a month for each other dependant. The third is the one-off death benefit, 20 times the previous year's national per-capita disposable income of urban residents, and that single figure is the only one there is, with no regional difference. For a death at work in 2026, this payment is about 1,130,000 yuan (exact figure 1130040 yuan). The calculation is the previous year's national per-capita disposable income of urban residents, 56502 yuan, multiplied by 20. +- Evidence grade: A +- Sources: Regulations on Work-Related Injury Insurance, article thirty-nine: 「（一）丧葬补助金为6个月的统筹地区上年度职工月平均工资；（二）供养亲属抚恤金按照职工本人工资的一定比例发给由因工死亡职工生前提供主要生活来源、无劳动能力的亲属。标准为：配偶每月40%，其他亲属每人每月30%，孤寡老人或者孤儿每人每月在上述标准的基础上增加10%……（三）一次性工亡补助金标准为上一年度全国城镇居民人均可支配收入的20倍。」. ; National Bureau of Statistics (国家统计局) (2026). Resident Income and Consumption Expenditure for 2025, 2025 年居民收入和消费支出情况. +- Notes: The one-off death benefit is the same nationwide, and the amount is updated each year with the figures the National Bureau of Statistics publishes. Do not accept the argument 「the local standard here is lower」 when you negotiate the compensation. + +### 17. If you are bullied, insulted or obstructed at work for a long time, do not just endure it: write it down and keep the evidence, then take a complaint, a police report or arbitration according to what it was + +- Cost: No money. It takes time to write it down: the date, who was present, the other person's exact words, and keep the originals of the chat screenshots and emails. Arbitration and court take several months. Changing job is the highest price of all, and whether to do it is yours to work out. +- In plain language: People who are bullied at work have about 60% more risk of a heart attack or stroke later, and the more they are bullied the higher the risk. People who are bullied are about twice as likely as others to have thoughts of suicide a few years on. Being sworn at and put upon is not a small thing that passes once you have endured it. +- Benefit: On health, three cohorts in Sweden and Denmark put together cover 79201 people in work aged 18 to 65. None of them had cardiovascular disease at the start, and the average follow-up was 12.4 years. 9% said they had been bullied at work in the past year. People who had been bullied had about 59% more risk of cardiovascular disease, meaning coronary heart disease and cerebrovascular disease, (HR 1.59, 95% CI 1.28 to 1.98). Those who had experienced violence at work had about 25% more (HR 1.25, 1.12 to 1.40). In both cases the more often it happened, the higher the risk. A nationwide randomly sampled follow-up survey in Norway covered 1846 people in work, measured every two or three years, three times in all. People who had been bullied were about 2 times as likely as others to have thoughts of suicide later (OR 2.05, 95% CI 1.08 to 3.89). In the other direction, people who already had thoughts of suicide were not more likely to be bullied later. On the law, China has no law specifically governing 「workplace bullying」, and you have to match the particular behaviour to a provision. Where an employer insults, corporally punishes, beats, unlawfully searches or detains a worker, it is subject to administrative penalty, has to pay damages where harm is caused, and criminal liability is pursued where it amounts to a crime. Where an employer uses violence or threats to force you to work, you may terminate the contract at once without prior notice and still claim economic compensation. The employer has to take measures to prevent and stop anyone using their position or a superior-subordinate relationship to carry out sexual harassment. Victims may require the harasser to bear civil liability. Labour arbitration is to close the case within 45 days of accepting it, and a complicated case may be extended by at most a further 15 days. +- Evidence grade: B +- Sources: Xu T et al. (2019). Workplace bullying and workplace violence as risk factors for cardiovascular disease: a multi-cohort study. European Heart Journal. ; Nielsen MB, Nielsen GH, Notelaers G, Einarsen S (2015). Workplace Bullying and Suicidal Ideation: A 3-Wave Longitudinal Norwegian Study. American Journal of Public Health. ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第三十八条第二款、第四十六条、第八十八条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零一十条）, full text reprinted by the Supreme People's Procuratorate ; Standing Committee of the National People's Congress (全国人大常委会) (2022 revision). Law on the Protection of Women's Rights and Interests, 妇女权益保障法（第二十五条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Law on the Mediation and Arbitration of Labour Disputes, 劳动争议调解仲裁法（第四十三条）, full text reprinted by the Ankang Municipal Human Resources and Social Security Bureau +- Notes: The reason for the B is that the health figures come from follow-up records, not from a grouped experiment, and 「being bullied」 is reported by the person themselves. There are four steps you can take first. First, start writing it down today, and for recording see item 41 in section 8 (recording). Second, put the complaint in writing through the employer's internal channels and keep a record that you did. Third, if you are beaten, locked up or publicly humiliated, call the police. Fourth, if you are forced out over unpaid wages or failure to enrol in social insurance, work through item 7 in this section (do not sign a voluntary resignation) and item 8 in this section (save the evidence before you leave). For how the economic compensation is worked out, see item 4 in this section (work out N before a dismissal). Plain neglect, exclusion and obstruction mostly do not match any particular provision of law, and the routes open to you are mainly an internal complaint, a transfer to another post, and changing job. Arbitration runs for at most 60 days from acceptance, and a first-instance court case runs from 6 months, with the legal fees your own. If you can no longer hold on, ring 12356 first: see item 25 in section 1 (12356). This book does not argue that you should resign the moment you are upset; whether changing job is worth it is yours to work out. diff --git "a/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" "b/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" new file mode 100644 index 00000000..8b9b9f01 --- /dev/null +++ "b/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" @@ -0,0 +1,119 @@ +[← 回总目录](../README.md) + +# 20. 刚出生的孩子怎么带 + +口径：婴儿死亡率与金钱。这一节只收关系到孩子能不能活下来、或者你要花多少钱的事，而且只收差别足够大、又有硬证据的。怎么哄、怎么带这些育儿细节，本书不写。出生之前和出院当天要办的事，见第 27 节。 + +[← Back to the contents](../README.md) + +# 20. Caring for a newborn + +This section measures infant mortality and money. It takes in only the things that decide whether the child lives, or what you have to spend, and only those where the difference is large and the evidence is solid. How to soothe a baby and the other childcare details are not in this book. For what to do before the birth and on the day you leave hospital, see section 27. + +### 1. Put the baby to sleep on their back, on a firm flat surface, in the same room but not the same bed, with nothing soft on the bed + +- Cost: No money. You have to swap the mattress for a firm one, which costs a few hundred yuan. The hard part is holding your line with older relatives who say otherwise. +- In plain language: Put the baby to sleep on their back, on a firm flat surface. The adult and the baby sleep in the same room, but not in the same bed. No pillows, no duvets, no cot bumpers, no soft toys and no shaping pillows on the bed. A reclined cradle or a baby seat is for sitting only, not for sleeping in. In the United States alone, about 3500 babies a year die in their sleep. +- Benefit: In the United States about 3500 babies a year die of sleep-related causes. That covers sudden infant death syndrome, deaths where no cause was found, and babies who suffocated by accident in a bed. The American Academy of Pediatrics gives a safe sleep environment as 「仰卧位；使用坚实、不倾斜的睡眠表面；同房不同床；避免柔软的寝具和过热」, that is, on the back, on a firm non-inclined sleep surface, in the same room but not the same bed, with no soft bedding and no overheating. Several other things lower the risk as well: breastfeeding, keeping the baby away from nicotine and alcohol, vaccination on schedule, and using a pacifier +- Evidence grade: A +- Sources: Moon RY, Carlin RF, Hand I, Task Force on Sudden Infant Death Syndrome and the Committee on Fetus and Newborn (2022). Sleep-Related Infant Deaths: Updated 2022 Recommendations for Reducing Infant Deaths in the Sleep Environment. Pediatrics, 150(1), e2022057990. +- Notes: 「床上不放任何软东西」means no pillows, no duvets, no cot bumpers, no soft toys and no shaping pillows. No breastfeeding pillow either, however safe it looks. A reclined cradle or a baby seat is not a place to sleep. Within this section, this is the item where the difference is largest and where you most need to hold your line against older relatives. + +### 2. Give the first hepatitis B shot within 24 hours of birth + +- Cost: No money. The hepatitis B vaccine is on the national immunisation programme, so it is already free. The hospital gives it on the day the baby is born, and you do not have to go anywhere else for it. +- In plain language: The baby needs the first hepatitis B shot within 24 hours of birth. Among people who catch hepatitis B in infancy or early childhood, about 95% go on to develop it chronically. Most people who catch it as adults clear it on their own. So the earlier this shot goes in, the better it works. If the mother's hepatitis B surface antigen is positive, the baby also needs hepatitis B immune globulin. +- Benefit: The World Health Organization says 「所有婴儿都应在出生后尽早（24 小时内）接种乙肝疫苗」, that is, all infants should receive the hepatitis B vaccine as soon as possible after birth, within 24 hours. The reason is that 「婴儿期和幼儿期感染约有 95% 会发展为慢性肝炎」, that is, about 95% of infections in infancy and early childhood develop into chronic hepatitis. Most infections picked up in adulthood clear without treatment +- Evidence grade: A +- Sources: World Health Organization. Hepatitis B fact sheet. +- Notes: If the mother's hepatitis B surface antigen is positive, the baby also needs a shot of hepatitis B immune globulin after birth. Settle this at the antenatal checks rather than after the birth. For the screening the mother herself needs, see section 1 + +### 3. Complete every vaccine on the national immunisation programme, free throughout; if you miss one, only the unfinished doses are made up + +- Cost: No money. The law says 「接种单位接种免疫规划疫苗不得收取任何费用」, that is, a vaccination provider may not charge any fee for vaccines on the national immunisation programme. You have to make several trips to a vaccination clinic according to the month age. Keep the vaccination certificate safe and do not lose it. +- In plain language: Vaccines on the national immunisation programme are free throughout, and a vaccination provider may not charge for them. Children are checked for their vaccination certificate when they start nursery and when they start school, and any missed doses are found and made up. If you miss a dose, only the unfinished doses are made up. You do not start again from the beginning, and you can change manufacturer and carry on. Eczema, urticaria, allergic rhinitis and asthma can all be vaccinated as long as the condition is stable. So can a newborn with jaundice. +- Benefit: When a child starts nursery and school, the nursery and the school must inspect the vaccination certificate. Where a child has not had the programme vaccines as required, it must be reported to the vaccination provider, which then helps press the guardian to make up the doses. Missed doses therefore cannot be put off, and they are certain to be found at school entry. The programme now in force is the 2026 edition. The human papillomavirus (HPV) vaccine is now on the national immunisation programme. The starting age for the diphtheria, tetanus and pertussis vaccine has moved from 3 months to 2 months, 5 doses in total, 1 dose at each of 2 months, 4 months, 6 months, 18 months and 6 years old. Eligible children in Tibet, Qinghai, Xinjiang and the Xinjiang Production and Construction Corps have also had Japanese encephalitis vaccine routinely since March 2026. Several vaccines can be given on the same day: at present every vaccine on the national immunisation programme can be given together under the schedule or under the catch-up rules. Two or more injectable vaccines have to go into different sites, and 「严禁将两种或多种疫苗混合吸入同一支注射器内接种」, that is, it is strictly forbidden to draw two or more vaccines into the same syringe for one injection. Where they are not given on the same day, two or more injectable live attenuated vaccines must be at least 28 days apart. No interval is set between inactivated vaccines, oral live attenuated vaccines and other vaccines. For anyone under 18 years old, catch-up starts as early as possible: 「只需补种未完成的剂次，无需重新开始全程接种」, that is, only the unfinished doses are made up and there is no need to restart the whole schedule. The same vaccine from a different manufacturer can be carried on with (nationwide, 2026 edition of the schedule) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2019). Vaccine Administration Law, 疫苗管理法（第四十八、四十九条）. (reprinted by the Yunnan Provincial Health Commission, 云南省卫生健康委员会); National Disease Control and Prevention Administration and National Health Commission (国家疾病预防控制局、国家卫生健康委) (2026). Questions and Answers on the Children's Immunisation Schedule for the National Immunisation Programme and the Explanations, 2026 edition, 《国家免疫规划疫苗儿童免疫程序及说明（2026 年版）》解读问答. +- Notes: The most common mistake is 「孩子体质不好，先别打」, that is, the child has a weak constitution, so hold off for now. The official position is this. An "allergic constitution", a family history of allergy and a previous allergic reaction to food or medicine are none of them reasons to withhold a vaccine. Atopic dermatitis (eczema), urticaria, allergic rhinitis and bronchial asthma can all be vaccinated as long as the condition is stable. Antihistamines and inhaled steroids do not stop vaccination either. Physiologic and breast-milk jaundice in newborns are also fine. Only two things stop a child: a previous acute severe allergic reaction to the same vaccine, or a severe allergy to a known component of the vaccine. Children with immune deficiency, and children on chemotherapy or radiotherapy or immunosuppressants, need an individual assessment, so explain the situation at the vaccination clinic and decide there. For the first hepatitis B dose, see item 2 in this section. Non-programme vaccines you pay for yourself carry a vaccine fee and a vaccination service fee, which have nothing to do with the free schedule described here. + +### 4. Breastfeed only for the first 6 months, not even water; from 6 months add solid food and keep breastfeeding + +- Cost: No money, and you save money on formula. What it costs is time, because feeding takes a good part of every day. The hard part is keeping it up, since you cannot stop in the first 6 months. +- In plain language: Start feeding within the first hour after birth. For the first 6 months give breast milk only, not even water. From 6 months add solid food, with an iron-rich animal food every day, and keep breastfeeding to 2 years old and beyond. If breast milk is not enough or you cannot breastfeed, use formula and do not blame yourself. What you feed matters far less for the child's safety than the sleeping position does. +- Benefit: The World Health Organization recommends 「在出生后第一个小时内开始母乳喂养」, that is, starting breastfeeding within the first hour after birth. For the first 6 months, 「纯母乳喂养，即不提供其他任何食物或液体，包括水」, that is, exclusive breastfeeding with no other food or liquid at all, water included. Then 「从 6 月龄起开始添加安全充足的辅食，同时继续母乳喂养到 2 岁或以上」, that is, from 6 months old start safe and sufficient complementary food while continuing to breastfeed to 2 years old and beyond. The service guide of the National Health Commission (国家卫生健康委) also recommends exclusive breastfeeding from 0 to 6 months. The same guide requires that from 6 months old 「每天的辅食必须包含含铁丰富的动物性食物」, that is, the day's complementary food must contain an iron-rich animal food +- Evidence grade: A +- Sources: World Health Organization. Breastfeeding. ; General Office of the National Health Commission (国家卫生健康委办公厅) (2024). Service Guide for Assessing Nutrition and Feeding of Infants and Young Children (for Trial), 婴幼儿营养喂养评估服务指南（试行）（国卫办妇幼函〔2024〕452 号）. +- Notes: If breast milk is not enough or your body does not allow it, use formula. The gap between breast milk and formula is far smaller than the one for safe sleep. How often to add solid food: 1 to 2 times a day at 6 to 8 months, and 2 to 3 times a day at 9 to 12 months. What the child eats in a day should cover at least four of the seven categories. When a child has severe eczema or an egg allergy, whether to avoid peanuts, see item 12 in this section (do not avoid peanuts). + +### 5. Make up formula with water above 70 ℃, let it cool before feeding, and throw away what is left + +- Cost: No money. Each time you wait a few extra minutes for the feed to cool. The hard part is waiting for the water to come down before mixing, when the baby is crying with hunger. +- In plain language: Formula is not sterile and may carry a bacterium called Cronobacter sakazakii. Once a newborn catches it, the reported case fatality rate runs from forty per cent to eighty per cent. When you prepare a feed, pour the hot water into the bottle first and wait until the temperature has fallen to about 73.8 ℃ before adding the powder. At that point the feed is still above 70 ℃, which kills the bacterium to below one hundred thousandth of its starting count. Let the feed cool before giving it, and throw away what is left. +- Benefit: Formula is not sterile and may carry Cronobacter sakazakii. The reported case fatality rate after infection in a newborn is 40% to 80%. Laboratory work shows that pouring the hot water into the bottle first and waiting until the temperature falls to about 73.8 ℃ before adding the powder leaves the feed above 70 ℃ in most cases. That reduces the bacterium by more than 5 log units, down to below one hundred thousandth of its starting count +- Evidence grade: A +- Sources: Beary MA, Daly SE, Baker J, Snyder AB (2025). Assessing Hot Water Reconstitution Instructions and Labeling of Powdered Infant Formula to Ensure Cronobacter spp. Reduction. Journal of Food Protection, 88(9), 100571. ; Sima WG, Legesse T, Girma S, et al. (2025). Emerging microbial risks: Cronobacter sakazakii in powdered infant formula for infants under six months of age in Ethiopia. BMC Microbiology, 26, 307. ; World Health Organization & FAO (2007). Safe preparation, storage and handling of powdered infant formula: guidelines. +- Notes: The 70 ℃ threshold comes from a public health guideline. Beary and colleagues put it this way: 「针对高危婴儿照护者的公共卫生指南建议用加热到至少 70 ℃（158 ℉）的水冲调配方奶粉以灭活微生物」, that is, public health guidelines for the carers of high-risk infants recommend preparing formula with water heated to at least 70 ℃ (158 ℉) in order to inactivate microbes. Preterm babies, babies with a low birth weight and babies under 2 months carry the highest risk. Do not use water that has sat in a thermos all day. Do not leave a made-up feed standing at room temperature until the next feed. + +### 6. No honey before 1 year old + +- Cost: No money +- In plain language: Do not give honey to a child under 1 year old. Honey water, honey mixed into solid food, and cakes that contain honey all count. Honey may carry the bacteria that cause botulism. The gut flora of an adult can hold them down, and an infant's cannot. Home-made pickles and fermented foods follow the same logic. +- Benefit: The US CDC says 「不要给 1 岁以下的儿童喂蜂蜜」, that is, do not give honey to children under 1 year old, because 「蜂蜜可能含有引起肉毒中毒的细菌」, that is, honey may contain bacteria that cause botulism. The gut flora of an adult can hold these spores down, which is the dormant form of the bacteria, and an infant's cannot. +- Evidence grade: A +- Sources: US CDC. Preventing Botulism. +- Notes: What counts is honey water, complementary food mixed with honey, and cakes containing honey. The same logic applies to home-made pickles and fermented foods, which babies should not eat either. + +### 7. The vitamin K shot at birth is not optional + +- Cost: 0 yuan to a few tens of yuan. This is normally already covered by the hospital bill for the birth, and there is nothing extra to pay. +- In plain language: Newborns store very little vitamin K, and without a supplement they can bleed. The late form often shows up as bleeding inside the brain. In places that do not give the shot routinely, 478 in every 100,000 newborns get this bleeding. With the shot, the estimate is a fall of more than seventy-eight per cent. Regular hospitals in China give it as routine. If you gave birth at home or in an unlicensed place, ask whether you had it. +- Benefit: Newborns store very little vitamin K. Without a supplement, vitamin K deficiency bleeding can occur, and the late form often presents as intracranial bleeding, that is, bleeding inside the brain. The Belgian consensus recommendation is 「出生时肌肉注射 1 或 2 毫克维生素 K 一次」, that is, a single intramuscular dose of 1 or 2 milligrams of vitamin K at birth. Where no systematic prophylaxis is provided, vitamin K deficiency bleeding runs at 478 cases per 100,000 newborns, and that figure comes from 42,000 newborns in Tashkent. The authors estimate that prophylaxis lowers the incidence by 「高于 78.5%」, that is, by more than 78.5% +- Evidence grade: B +- Sources: Fiesack S, Smits A, Rayyan M, et al. (2021). Nutrients, 13(11), 4109. ; Tursunov D, Yoshida Y, Yrysov K, et al. (2018). Nagoya Journal of Medical Science, 80(1), 11. +- Notes: If you give birth in a regular hospital in China, the hospital gives this shot routinely and you need not think about it. If you gave birth at home or in an unlicensed place, go and confirm it yourself. Oral vitamin K works less well than the injection, and the injection here means one given into the muscle. + +### 8. A baby under 3 months old at 38 ℃ goes straight to hospital, with no watching it at home + +- Cost: A few hundred yuan for registration and tests +- In plain language: If the baby is under 3 months old and the temperature reaches 38 ℃, go straight to hospital. Do not watch it at home, and do not give a fever medicine first and then wait. A baby at this age can have a serious bacterial infection and show nothing but a fever. So 「看着精神还行」, looking bright and alert, is not a reason to wait at home. A baby less than 8 days old is not even covered by the guideline, and needs to go at once. +- Benefit: The American Academy of Pediatrics has written a dedicated guideline for full-term infants aged 8 to 60 days who look reasonably well. It defines fever as 「体温 ≥38.0 ℃」, that is, a temperature of 38.0 ℃ or above. It splits infants into three bands by age, 8 to 21 days, 22 to 28 days and 29 to 60 days, and gives one management pathway for each. Infants under 8 days old are not even within the scope of this guideline, which means this stage needs care immediately +- Evidence grade: A +- Sources: Pantell RH, Roberts KB, Adams WG, et al. (2021). Evaluation and Management of Well-Appearing Febrile Infants 8 to 60 Days Old. Pediatrics, 148(2), e2021052228. +- Notes: At this age a serious bacterial infection can also show up as nothing but a fever. So looking 「精神还行」, bright and alert, is not a reason to watch the baby at home. Do not give a fever medicine on your own first and then wait and see. + +### 9. Never shake a baby, however tired and angry you are + +- Cost: No money. The hard part is stopping your hands when you are too tired or too angry. +- In plain language: A baby's head is large and the neck has no strength. Shaking hard causes bleeding in the brain and in the eyes, which can kill the baby or leave a lifelong disability. Gentle rocking to settle a baby does not count, and the danger is shaking you cannot control. When you are about to lose your temper, put the baby back in the cot and walk away for a few minutes. Letting the baby cry for a while is safer than holding the baby and shaking. +- Benefit: A baby's head takes up a large share of the body and the neck muscles are weak. Vigorous shaking causes intracranial and retinal bleeding, that is, bleeding in the brain and in the eyes, and can be fatal or leave a lifelong disability. Gentle rocking to settle a baby does not count, and the danger is uncontrolled violent shaking. Medicine calls this kind of injury from shaking and impact 「虐待性头部外伤」, abusive head trauma. It is the leading cause of death from head injury in children under 2. Between 1999 and 2014 in the United States it killed close to 2250 children under 5 +- Evidence grade: B +- Sources: Spies EL, Klevens J (2016). Fatal Abusive Head Trauma Among Children Aged <5 Years — United States, 1999–2014. MMWR 65(20):505-509. ; Choudhary AK et al. (2018). Consensus statement on abusive head trauma in infants and young children. Pediatric Radiology 48(8):1048-1065. +- Notes: When you really are about to lose control, put the baby back in the cot and walk away for a few minutes. Letting the baby cry for a while is far safer than holding the baby and shaking. If you are caring for the baby alone and are chronically short of sleep, remember this one above all. For what to do with your own temper, see section 3 + +### 10. Judge nappies on three things rather than the brand: fit, how often you change them, and whether they have been named in a spot check + +- Cost: A few hundred to a few thousand yuan a month, depending on the grade you buy +- In plain language: No official body has ever ranked which brand is best. There are only two things you can check: the product quality spot-check notices of the State Administration for Market Regulation (国家市场监督管理总局), and the investigations regulators carry out into specific incidents. Expensive does not mean safe, and imported does not mean compliant. Buy a small pack and try it for a week, look for nappy rash and marks on the skin, and only then buy in bulk. +- Benefit: No official body has published a ranking of which brand is best, so this book recommends no brand. Only two kinds of information can be checked. One is the product quality spot-check notices of the State Administration for Market Regulation. The other is the investigations regulators carry out into specific incidents. After 「婴幼儿纸尿裤甲酰胺问题」, the formamide problem in nappies for infants and toddlers, drew attention in month 6 of 2026, the State Administration for Market Regulation led the formation of a joint investigation team to look into it +- Evidence grade: C +- Sources: State Administration for Market Regulation (市场监管总局) (2026-06-22). The State Administration for Market Regulation led the formation of a joint investigation team to check on 「婴幼儿纸尿裤甲酰胺问题」：「针对媒体反映的"婴幼儿纸尿裤甲酰胺问题"，市场监管总局、工业和信息化部、国家卫生健康委、国家疾控局高度重视，成立联合调查组，核查婴幼儿纸尿裤甲酰胺有关问题，并依法依规处理。有关情况将及时公布。」. ; reprinted by the Chinese Government Website (中国政府网). +- Notes: When the joint investigation team was set up it only said 「有关情况将及时公布」, that the situation would be published in due course. TODO (to be verified: the conclusion of the investigation. As of 30 September 2026, no conclusion on nappies could be found on the SAMR website or in recent national spot-check notices). How to buy in practice: expensive does not mean safe, and imported does not mean compliant. Buy a small pack and try it for a week, look for nappy rash and marks on the skin, and only then buy in bulk. Nappy rash comes mainly from changing too infrequently and from poor breathability, and has less to do with the brand than people imagine. For how to check the spot-check notices, see section 5 + +### 11. For big items, work through borrow, then second-hand, then new, and do not buy the whole set at once + +- Cost: No extra money, and what you save usually adds up to thousands of yuan. The hard part is resisting the urge to buy everything at once. +- In plain language: Strollers, cots, swaddles and toys are all used for a short time, and there are plenty of second-hand ones. Ask to borrow first, buy second-hand if you cannot borrow, and only then buy new, which usually saves over a thousand yuan. Do not buy second-hand car seats or mattresses, because damage to a seat after a crash cannot be seen with the eye. For big sums such as a confinement centre or an early education course, wait 24 hours before deciding. +- Benefit: Strollers, cots, swaddles and toys are all used for a short time, and second-hand supply far outruns demand. Two things really must not be economised on and are not recommended second-hand: car seats and mattresses. Structural damage to a car seat after a crash cannot be seen with the eye +- Evidence grade: C +- Sources: the author's own experience, no direct literature; for the evidence on car seats, see section 1, and for impulse spending, see section 5 +- Notes: This book neither recommends nor rules out confinement centres, early education classes or baby swimming pools. They are all large sums you can choose to spend or not, and the 24-hour cooling-off period described in section 5 applies to them. + +### 12. Severe eczema or an egg allergy is not a reason to avoid peanuts: add them early under a doctor's guidance, but never a whole nut + +- Cost: A jar of peanut butter is a few tens of yuan. You feed it a few times a week, which fits in with what you already do. Before you start, take the child to see a doctor once for an assessment. +- In plain language: For infants with severe eczema or an egg allergy, giving peanuts regularly from 4 to 11 months leaves only 1.9% with a peanut allergy at 5 years old, against 13.7% among those who avoided peanuts completely. Among children whose skin test already showed some reaction, the figure is 10.6% for those who ate peanuts and 35.3% for those who avoided them. See the doctor before you start. Never give a whole peanut, because it can block the airway. +- Benefit: The British LEAP randomised trial. 640 infants were enrolled, on condition that they had severe eczema, an egg allergy, or both. They were 4 to 11 months old at enrolment. They were randomised into two groups, one eating peanut products regularly and one avoiding peanuts completely, and the peanut allergy rate was read off at 60 months. Among the 530 children whose peanut skin test was negative at enrolment, 13.7% in the avoidance group were allergic against 1.9% in the group that ate peanuts (P<0.001). Among the 98 children whose skin test was already weakly positive at enrolment, the figures were 35.3% in the avoidance group and 10.6% in the group that ate peanuts (P=0.004). Serious adverse events did not differ between the two groups. +- Evidence grade: A +- Sources: Du Toit G, Roberts G, Sayre PH, et al. (2015). Randomized trial of peanut consumption in infants at risk for peanut allergy. New England Journal of Medicine, 372(9), 803-813. ; General Office of the National Health Commission (国家卫生健康委办公厅) (2020). Core Messages on Health Education for Infant and Young Child Feeding, 婴幼儿喂养健康教育核心信息. ; Perkin MR et al. (2016). Randomized Trial of Introduction of Allergenic Foods in Breast-Fed Infants. New England Journal of Medicine, 374(18), 1733-1743. (the EAT trial of ordinary infants, referred to in the Notes) +- Notes: **The form must be peanut butter thinned out or peanut powder mixed into complementary food. Never give a whole peanut.** The feeding guidance from the National Health Commission is plain about it: 「整粒花生、坚果、果冻等食物易吸入气管，引起窒息，婴幼儿应当避免食用」, that is, whole peanuts, nuts, jellies and similar foods are easily drawn into the airway and cause suffocation, and infants and young children should avoid eating them. For what to do when someone is choking and cannot speak, see item 26 in section 13. This applies only to high-risk children: those with severe eczema, or those already allergic to eggs. **See a doctor for an assessment before you start; do not try it at home.** Before enrolment the LEAP trial gave every child a skin test, and children whose skin test reaction was greater than 4 mm were excluded and never given peanuts. The evidence in ordinary children is much weaker. In a separate trial of 1162 ordinary breast-fed infants, the early-introduction group had 5.6% food allergy against 7.1% in the usual group on an intention-to-treat basis, which is not a significant difference (P=0.32). It is significant only among the children who actually followed the plan (2.4% against 7.3%). That way of counting tends to overstate the effect, and the authors themselves say the intention-to-treat analysis failed to show efficacy, while also showing that early introduction is safe. China's official feeding documents currently say nothing about whether to introduce allergenic foods early or to avoid them, so this item rests on the international trial evidence. Note the timing too: LEAP started at 4 months, while China starts complementary food at 6 months, see item 4 in this section (breastfeed only for the first 6 months). At what age a high-risk child starts and how to start, ask the doctor. diff --git "a/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" "b/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" new file mode 100644 index 00000000..74aa29fa --- /dev/null +++ "b/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" @@ -0,0 +1,110 @@ +[← 回总目录](../README.md) + +# 21. 出国、旅行与境外安全 + +这一节算的是钱和人身自由，有几条还关系到人身安全。到了国外，管你的是所在国的法律，不是中国的。中国使领馆能帮你到哪一步，有明确的界线。 + +[← Back to the contents](../README.md) + +# 21. Travel and safety abroad + +This section measures money and personal freedom, and several of the items also bear on personal safety. Once you are abroad, the law that governs you is that of the country you are in, not of China. There are clear limits to how far a Chinese embassy or consulate can go for you. + +### 1. Check the Ministry of Foreign Affairs safety notices once before you leave, and do not go anywhere marked 「暂勿前往」 + +- Cost: No money. Open the Chinese Consular Affairs website and go through it once. It takes a few minutes. +- In plain language: The Ministry of Foreign Affairs sorts overseas safety risk into four bands, and the highest is 「暂勿前往」, do not travel. From 15 September 2026 the duty to stay away from high-risk countries and regions was written into an administrative regulation. When you apply for your documents and go through exit formalities, staff will also remind you, and where necessary try to talk you out of the trip. Read that page once before you leave. It takes a few minutes. +- Benefit: The Ministry of Foreign Affairs publishes overseas safety notices in public, banded by risk level. From low to high they are 「注意安全」, take care of your safety, 「加强安全防范」, strengthen your precautions, 「谨慎前往」, travel with caution, and 「暂勿前往」, do not travel. The new rules, in force from 2026-9-15, also write the duty of the individual into administrative regulations. The text reads: 「中国公民应当关注国外安全提醒和旅游目的地安全风险提示，避免前往及驻留高风险国家或者地区」, that is, Chinese citizens should pay attention to overseas safety notices and safety risk warnings for travel destinations, and avoid travelling to and staying in high-risk countries or regions. The new rules also require the immigration authorities, the departments that issue entry and exit documents and run border checks, to remind anyone heading for a high-risk area when that person applies for documents and goes through the exit check. That requirement reads: 「对准备前往风险等级为最高级别或者严重危及人身安全案件突发高发的国家或者地区的中国公民，必要时应当劝阻其前往」, that is, for Chinese citizens preparing to travel to a country or region at the highest risk level, or where cases seriously endangering personal safety break out suddenly and often, the authorities should try to stop them going where necessary. +- Evidence grade: A +- Sources: State Council (国务院) (2026). Provisions on the Administration of Exit and Entry, 关于出境入境管理的规定（国令第 841 号）. (Article two); State Council (国务院) (2023). Regulations on Consular Protection and Assistance, 领事保护与协助条例（国令第 763 号）. (Articles nineteen and twenty-one); Consular Affairs Department of the Ministry of Foreign Affairs. Safety notices, 安全提醒. +- Notes: The notices change with the situation, so this book does not copy the list. Check that page yourself before you travel. As of 7 September 2026 the column still carried these 「暂勿前往」 notices: Eswatini (2026-08-25), Palestine (2023-10-10), Sudan (2023-04-17). Four more were Afghanistan (2023-02-17), Syria (2023-01-04), Peru (2022-12-29) and Somalia (2022-11-24). There are also separate notices for northern Myanmar and for some provinces of the Democratic Republic of the Congo. Travel agencies have a duty to inform you as well. Before joining a group tour you can ask for a written statement of the risk at the destination. + +### 2. Save 12308 and your local consular protection number in your phone, and copy them onto paper in your wallet + +- Cost: No money. Save two numbers and write one copy down. A few minutes. +- In plain language: There is one number to call when something goes wrong. In China, dial 12308. Abroad, dial +86-10-12308. Since the hotline opened it has taken more than 3 million calls. Over the same period the Ministry of Foreign Affairs and its missions abroad handled more than half a million consular protection cases. +- Benefit: The Ministry of Foreign Affairs runs a global consular protection and assistance emergency line, answered 24 hours a day. In China, dial 12308. Abroad, dial +86-10-12308 or +86-10-65612308. Since the line opened in 2014 it has taken more than 300 million calls. Over the same period the Ministry of Foreign Affairs and its missions abroad handled more than 50 × 10,000 consular protection cases of every kind +- Evidence grade: A +- Sources: Ministry of Foreign Affairs and Chinese Consulate-General in Myanmar (外交部、中国驻缅甸使领馆) (2023). Consular reminder on the current security situation in northern Myanmar, 关于当前缅北地区安全局势的领事提醒. ; State Council Information Office (国务院新闻办) (2023). The State Council's regular policy briefing on consular protection and assistance. +- Notes: The local embassy or consulate has its own consular protection number, and it is usually faster than the switchboard. Before you leave, look up the country on the Chinese Consular Affairs website and save that number alongside the switchboard number. If your phone is lost or runs flat, numbers saved only on it are no use. So write a copy on paper and keep it in your wallet, or give a copy to someone at home. + +### 3. Know what consular protection can and cannot do: it can visit you, it cannot get you out, and you pay the bill + +- Cost: No money. It takes a few minutes to read this boundary clearly. +- In plain language: If you are arrested abroad, a consulate can do this: check the situation with the local authorities, ask for handling that is lawful and fair, visit you, and observe the trial. It cannot change the verdict, and it cannot take you out of the country. Accommodation, transport and medical costs paid by someone else on your behalf come back to you in the end. So put the money and the insurance in place before you travel, and do not build a plan on the idea that the state will cover you if something happens. +- Benefit: When you are arrested, a consulate can do these things. It checks the situation with the relevant departments of the country you are in and asks for handling that is lawful and fair. The regulation reads: 「按照驻在国法律和我国与驻在国缔结或者共同参加的国际条约对其进行探视或者与其联络，了解其相关需求，要求驻在国有关部门给予该中国公民人道主义待遇和公正待遇」. In plain terms, it visits you and gets in touch with you under the local law and under the treaties signed by the two countries, asks what you need, and asks the local authorities to treat that Chinese citizen humanely and fairly. At the trial a consulate may sit in and may ask that your procedural rights be protected. If you are injured or caught in a disaster, it can press the local authorities for emergency help and can help the family handle what follows. On costs the regulation is just as clear: 「中国公民、法人、非法人组织在领事保护与协助过程中，得到第三方提供的食宿、交通、医疗等物资和服务的，应当支付应由其自身承担的费用」. Accommodation, transport and medical care paid for you by a third party are yours to pay +- Evidence grade: A +- Sources: State Council (国务院) (2023). Regulations on Consular Protection and Assistance, 领事保护与协助条例（国令第 763 号）. （第九、十、十四、二十六条） +- Notes: A consulate does not pay your fines, does not pay your medical bills and does not buy your ticket, and it cannot change the conclusion of a court in the country you are in. What it can give you is information on local lawyers, translators, medical care and funeral services, and help reaching relatives and finding relief. So put the money and the insurance in place before you travel, and do not build a plan on the idea that the state will cover you if something happens. + +### 4. Buy a policy that covers overseas medical treatment and medical evacuation, not only flight-delay cover + +- Cost: A short-term overseas travel policy costs a few tens to a few hundred yuan. +- In plain language: You pay for being treated abroad yourself. Chinese medical insurance basically does not cover it, and the consulate does not advance the money. What can bankrupt a family is not the clinic bill but the hospital stay after a serious injury, and getting the patient home, which takes an air ambulance or a medical escort. When you buy travel insurance, look at two limits: overseas medical expenses, and emergency medical evacuation and repatriation. +- Benefit: You pay for medical treatment abroad yourself. Chinese medical insurance basically does not reimburse it abroad. Consular protection does not advance this money either (see item 3 in this section, what consular protection can and cannot do, based on Article twenty-six of the regulations). What actually bankrupts a family is not the clinic bill but the hospital stay after a serious injury and the medical evacuation. Moving a patient from a foreign hospital back to China takes an air ambulance or a medical escort, and it is among the most expensive self-paid items. The US CDC puts it this way: moving a patient in an emergency from a remote area to a good hospital can, if you pay for it yourself, cost more than $100,000. When you buy insurance, look at two limits first. One is the limit on overseas medical expenses. The other is the limit on emergency medical evacuation and repatriation, which covers the most that will be paid for transferring to another hospital after an accident and for getting the patient home +- Evidence grade: C +- Sources: State Council (国务院) (2023). Regulations on Consular Protection and Assistance, 领事保护与协助条例（国令第 763 号）. （第二十六条）; US Centers for Disease Control and Prevention (美国疾病控制与预防中心). Travelers' Health: Travel Insurance：「Medical evacuation insurance covers emergency transportation from a remote area to a high-quality hospital, which could otherwise cost more than $100,000.」. ; the rest is the author's own experience, no direct literature +- Notes: The figure of $100,000 is a floor, not a range of costs. No official or industry published range of costs was found, so this book gives no specific range. Confirm three things before you buy. First, whether the country you are going to is covered. Second, whether high-risk activities such as diving, skiing and motorcycling are excluded. Third, whether the insurer settles with the hospital directly or pays you back after you pay up front. Buying a policy is just as necessary for visa-free countries. + +### 5. Treat every 「境外高薪招聘」 offer as fraud, and know that being tricked into telecom fraud can cost you the right to leave the country + +- Cost: No money. The hard part is turning down with your own hand a job that looks very good. +- In plain language: A monthly wage so absurd it cannot be true, no questions about your education or experience, no proper contract, being told to gather in a border city first, and someone else keeping your passport. Take a job like that and neither you nor your passport is in your own hands. If you falsify your papers on the way out, border control will stop you and not let you go. Take part in crime abroad, and once you are back you can be barred from leaving the country for 6 months to 3 years. +- Benefit: The official description of these cases is blunt. In recent years there have been cases of 「上当受骗出境或者虚构事由非法出境从事跨境赌博、电信网络诈骗等违法活动，严重危害人民群众生命和财产安全」, that is, people deceived into leaving the country, or leaving illegally on a made-up pretext, to engage in cross-border gambling, telecom network fraud and other illegal activity there, seriously endangering the lives and property of the people. That covers both being tricked into going and inventing a reason to slip out to do gambling and telecom fraud. The new rules tighten both ends of this. On the way out: 「申请出境入境、停留居留的事由应当真实、合法」, that is, the stated reason for applying to leave or enter the country or to stay there must be true and lawful. If you submit false documents or tell lies, the immigration authorities may refuse to issue you a document and may refuse to let you leave. On the way back: 「中国公民在境外从事违法犯罪活动，危害国家安全和利益的，可以……决定自回国之日起 6 个月至 3 年以内不准其出境」, that is, where a Chinese citizen commits illegal or criminal acts abroad that endanger national security and interests, the authorities may decide, counting from the date of return, to bar that person from leaving the country for 6 months to 3 years. Where someone is given administrative detention for obtaining documents fraudulently or for leaving or entering the country illegally, the bar is likewise 6 months to 3 years +- Evidence grade: A +- Sources: State Council (国务院) (2026). Provisions on the Administration of Exit and Entry, 关于出境入境管理的规定（国令第 841 号）. （第三、四条）; Ministry of Justice, Ministry of Public Security and National Immigration Administration (司法部、公安部、国家移民局) (2026). Questions and Answers on the Provisions of the State Council on the Administration of Exit and Entry. +- Notes: The warning signs are more or less fixed: a monthly wage far above the same job, no requirement for education or experience, no proper contract, being told to gather in a border city first, visa handling done for you and your passport kept by them. There is already a dedicated consular reminder for northern Myanmar. If you do go, your personal freedom and your passport are in someone else's hands and you cannot leave when you want to. + +### 6. Cash withdrawals abroad are capped at 100,000 yuan a year, counted across every card in your own name + +- Cost: No money. Estimate before you leave how much cash this trip needs, and use cards for the rest. The hard part is changing the habit: if a card will do, do not take out cash. +- In plain language: The cap on cash withdrawals abroad is not one card at a time. It is every bank card in your name, supplementary cards included, added together, at 100,000 yuan for each calendar year. Once you have used it up, you cannot withdraw cash abroad for the rest of that year or in the year after. Only withdrawals stop. Card payments are not affected, and they do not use up your annual quota of 50,000 US dollars for purchasing foreign exchange. +- Benefit: The basis is the Notice of the State Administration of Foreign Exchange on Regulating Large Cash Withdrawal Transactions with Bank Cards Abroad, 《关于规范银行卡境外大额提取现金交易的通知》（汇发〔2017〕29 号，in force from 1 January 2018）. Article one: 「个人持境内银行卡在境外提取现金，本人名下银行卡（含附属卡）合计每个自然年度不得超过等值 10 万元人民币。超过年度额度的，本年及次年将被暂停持境内银行卡在境外提取现金」, that is, where a person withdraws cash abroad with a domestic bank card, the total across all bank cards in that person's name, supplementary cards included, must not exceed the equivalent of 100,000 yuan in a calendar year. Where the annual cap is exceeded, withdrawals abroad on domestic bank cards are suspended for that year and the year after. The second paragraph of the same article: 「个人不得通过借用他人银行卡或出借本人银行卡等方式规避或协助规避境外提取现金管理」, that is, a person may not borrow another person's bank card or lend out their own to get around, or to help someone else get around, the management of cash withdrawals abroad. Article two: the foreign exchange authority sends the suspension list to the issuing banks every day, and an issuing bank must, no later than 17 hrs Beijing time on that day, stop the people on the list from withdrawing cash abroad with that bank's cards. Article three: a person on the suspension list may take valid identity documents to a local branch of the foreign exchange authority and check their own record of cash withdrawals abroad. Article four: the daily cap per foreign currency card is raised from the equivalent of USD 1000 to the equivalent of 10,000 yuan; the cap for renminbi cards stays at the equivalent of 10,000 yuan per card per day. Article six: issuing banks, renminbi card clearing organisations in China and individuals who breach the provisions of this notice are punished under the relevant rules of the Regulations of the People's Republic of China on Foreign Exchange Administration. The foreign exchange authority's answers to reporters also made several points clear. The Notice does not affect spending abroad on a card. Ordinary transactions such as food, accommodation, transport and shopping on a trip abroad for tourism, business or study can all be paid by card. Those payments do not use up the individual facilitation annual quota of USD 50,000 for purchasing foreign exchange. The annual quota is worked out by the foreign exchange authority from data added up across the issuing financial institutions, because real-time control would slow transactions down: 「外汇局采用延时控制手段」, that is, the authority uses deferred control rather than stopping the payment on the spot. Separately, in 2016, 81% of cash withdrawals abroad on domestic bank cards were below 30,000 yuan (nationwide) +- Evidence grade: A +- Sources: State Administration of Foreign Exchange (国家外汇管理局) (2017). Notice on Regulating Large Cash Withdrawal Transactions with Bank Cards Abroad, 关于规范银行卡境外大额提取现金交易的通知（汇发〔2017〕29 号, Articles one, two, three, four, six and seven, in force from 1 January 2018）. ; State Administration of Foreign Exchange (国家外汇管理局) (2017). Answers by the responsible officials of the foreign exchange authority to reporters on the questions of regulating large cash withdrawal transactions with bank cards abroad. ; China UnionPay International (中国银联国际). ATM cash withdrawals abroad, 境外ATM取款. +- Notes: The cap is counted by calendar year and starts again on 1 January. Only withdrawals abroad stop; card payments are not affected. If you really need a lot of cash abroad, buy the foreign exchange first under the Administrative Measures on Foreign Exchange of Individuals, then take the foreign currency notes out of the country. Withdrawal itself has two traps, and the UnionPay International page 「境外ATM取款」 sets them out. First, a dual-logo card cannot use the UnionPay channel at some ATMs in some countries. Second, some overseas ATMs show a surcharge on the screen, and that money goes to the local acquirer and is 「与银联无关」, nothing to do with UnionPay. The same page also warns: 「由于信用卡取款需支付透支利息，建议持卡人境外取款使用银联借记卡」, that is, because a credit card withdrawal incurs overdraft interest, holders are advised to use a UnionPay debit card for cash withdrawals abroad. If the card is lost, swallowed or used fraudulently, see item 5 in section 14 (report the card lost and freeze it before calling the police). + +### 7. Photograph your passport, visa and identity card and keep a copy in the cloud, and if they are lost, report first and then get a travel document + +- Cost: No money. Take a few photographs before you leave. +- In plain language: If your passport is lost, a consulate can issue you a travel document for the journey home. You have to be able to prove who you are first. With photographs of the data page, of the visa pages and of your entry and exit records to hand, the replacement goes through much faster. You also have to report the loss locally and get a receipt, which the exit check and the insurance claim will both ask for. Phones are often lost together with the passport, so do not keep the photographs only on the phone. +- Benefit: If your passport is lost, the embassy or consulate abroad can issue you a travel document for the journey home, but you have to prove your identity first. With clear photographs of the data page, of the visa pages and of your entry and exit records to hand, the difference in how long the replacement takes is large. At the same time you have to report the loss locally and get a receipt. Many countries' exit checks and insurance claims ask for this document. The way into the consulate is the two consular protection numbers you saved in item 2 in this section +- Evidence grade: C +- Sources: State Council (国务院) (2023). Regulations on Consular Protection and Assistance, 领事保护与协助条例（国令第 763 号）. （第十六条）; the author's own experience, no direct literature +- Notes: How to store them: take the photographs and put them in two different places, for example one copy in cloud storage and one sent to a relative. Print a paper copy as well and keep it away from the original. Photograph your bank cards and the prescriptions for the medicines you use regularly too. A phone and a passport are usually lost together, so storing them only on the phone is the same as not storing them. + +### 8. Before driving abroad, check whether the destination recognises a Chinese licence; most 「国际驾照」 sold online are waste paper + +- Cost: No money. Check the rules of your destination once before you leave. +- In plain language: The 「国际驾照」 arranged for you online has no standing in mainland China. That document is a translation recognised between the parties to the 1949 Geneva convention, and mainland China is not among the 103 parties. So whoever arranges one for you in China gives you a document with no legal effect. Whether you may drive on a Chinese licence depends on the destination's own rules. If you have not confirmed those rules, do not drive. +- Benefit: The so-called international driving permit (IDP) is a translation recognised between the parties to the 1949 Convention on Road Traffic. The convention was concluded at Geneva on 1949-9-19 and entered into force on 1952-3-26. There are 103 parties at present, that is, the countries and regions that have signed. Mainland China is not on that list. So China does not issue this kind of international driving permit. Any agency offering to arrange an 「国际驾照」 inside China produces no legal effect. Whether you may drive on a Chinese licence depends on the rules of the destination country or region itself: some recognise a Chinese licence with a notarised translation, some require you to exchange it for a local licence, and some do not recognise it at all +- Evidence grade: A +- Sources: United Nations Treaty Collection. Convention on Road Traffic, Geneva, 19 September 1949 — status of treaties. +- Notes: In most countries driving without a licence is a criminal offence, or it carries a heavy penalty. And if there is an accident the insurance will very likely refuse to pay, which is the real cost. There are two ways to check: look up the destination country's 「中国公民须知」 on the Chinese Consular Affairs website, or ask the car hire company for a written confirmation. Hong Kong, Macao, Taiwan and some countries have separate arrangements. + +### 9. Before you use an agency for a visa, study abroad or immigration, ask whether it is registered with the immigration authorities + +- Cost: No money. Ask once whether it is registered, and keep the written material. +- In plain language: From 15 September 2026, agencies that handle visas, study abroad and immigration have to register with the immigration authorities. Phrases like 「包过」, guaranteed approval, 「保签」, guaranteed visa, and 「材料我们帮你做」, we prepare the documents for you, are themselves expressly forbidden. Filling in a form wrongly yourself costs you at most a refusal this time round. Using false documents and being found to have submitted false information leaves a mark on your entry and exit record for years to come. +- Benefit: The licensing requirement for private entry and exit agencies was removed in 2018, so an agency no longer needs an approval before it opens. The number of agencies then grew fast, and problems appeared that 「机构底数不清、部分机构违法违规办理业务、损害出境入境人员合法权益」, that is, the true number of agencies is unclear, and some agencies handle business in breach of the law and damage the lawful rights of people entering and leaving the country. The authorities cannot even count them. From 2026-9-15 these agencies are put under registration. An agency must report to the immigration authorities for the place where it sits within 15 days of being set up. Agencies already operating must make up the registration within 90 days of the rules coming into force. The expressly forbidden practices fall into these categories. First, 「发布虚假信息，或者通过夸大宣传、误导性宣传等方式招徕服务对象」, publishing false information, or attracting clients by exaggerating or misleading publicity. Second, 「提供或者协助提供虚假材料，协助他人违规办理签证、停留居留证件、护照等出境入境证件或者手续」, supplying or helping to supply false documents, and helping others obtain visas, stay and residence documents, passports and other entry and exit documents or formalities in breach of the rules. Third, 「泄露、出售、非法提供……个人隐私或者个人信息」, disclosing, selling or unlawfully providing personal privacy information or personal information. Fourth, 「组织或者协助他人从事跨境违法犯罪活动」, organising or helping others carry out cross-border illegal and criminal activity. Overseas enterprises and institutions may not provide this kind of agency service inside China. +- Evidence grade: A +- Sources: State Council (国务院) (2026). Provisions on the Administration of Exit and Entry, 关于出境入境管理的规定（国令第 841 号）. （第七、八、十、十二、十三条） +- Notes: An agency that keeps saying 「包过」, guaranteed approval, 「保签」, guaranteed visa, and 「材料我们帮你做」, we prepare the documents for you, is already doing what is expressly forbidden. If it is caught, the mark is left on your visa record. A problem with an application you filled in yourself costs you at most a refusal this time round. Using false documents and being found to have submitted false information costs a good deal more. Policy advice and information lookup that are not done for money do not count as agency service. + +### 10. Do not write an invitation letter for a stranger, and do not let anyone handle entry and exit formalities in your name + +- Cost: No money. The hard part is turning down a favour to someone's face. +- In plain language: If you write an invitation letter or application documents for someone else and the contents are untrue, the person who wrote it is the one who is fined. An individual is fined 5000 yuan to 10,000 yuan, and any illegal gains are confiscated as well. An organisation is fined 10,000 yuan to 50,000 yuan, and the person who handled it is fined 5000 yuan to 10,000 yuan on top. One favour ends up costing you the money yourself. +- Benefit: The person who issues an invitation letter is responsible for whether its contents are true. The text reads: 「个人为他人申请出境入境、停留居留出具虚假邀请函件或者其他申请材料的，由移民管理机构处 5000 元以上 1 万元以下罚款；有违法所得的，没收违法所得」, that is, where an individual issues a false invitation document or other application document for another person's exit, entry, stay or residence, the immigration authority shall impose a fine of 5000 yuan to 10,000 yuan, and any illegal gains shall be confiscated. Where an organisation issues false documents, the fine is 10,000 yuan to 50,000 yuan. At the same time a further fine of 5000 yuan to 10,000 yuan is imposed on the person in charge and the other people directly responsible, that is, the manager who signs it off and the person who handled it. +- Evidence grade: A +- Sources: State Council (国务院) (2026). Provisions on the Administration of Exit and Entry, 关于出境入境管理的规定（国令第 841 号）. （第三、十一条） +- Notes: This is the same kind of risk as 「不帮陌生人带东西」 in section 8: once documents are issued in your name, the consequences land on you. Where a public servant or a member of the armed forces uses an agency to get foreign nationality or overseas residence status in breach of the rules, the agency may not only refuse; it also has to report the matter to the supervisory authority. + +### 11. Before you go abroad, send your itinerary, address and travelling companions to one family member and agree how often you will make contact + +- Cost: No money. Send one message before you leave. +- In plain language: If a person goes missing abroad, family members can ask an embassy or consulate for help. The consulate will tell the family how to report to the police locally, and once a case is opened it will chase it. All of that depends on someone at home knowing where you should be and when you should make contact. Sending your flight number, address, the name and phone number of a travelling companion and the contact interval you agreed is enough. +- Benefit: When a Chinese citizen goes missing abroad, relatives may ask a diplomatic mission abroad for help. The regulation reads: 「驻外外交机构应当提供当地报警方式及其他获取救助的信息」, that is, diplomatic missions abroad shall provide the way to report to the police locally and other information for obtaining assistance. Once the local police have opened a case, the consulate presses them to deal with it in time. Whether any of this can be started depends on whether someone at home knows where you should be and when you should make contact. +- Evidence grade: C +- Sources: State Council (国务院) (2023). Regulations on Consular Protection and Assistance, 领事保护与协助条例（国令第 763 号）. （第十三条）; the author's own experience, no direct literature +- Notes: Four things are enough: the flight number, the address, the names and phone numbers of the people you travel with, and the contact frequency you agreed. If you are going somewhere with poor signal or poor public order, say plainly how long you can go without news before someone reports you. All of the rescue procedures above can only start once a family member knows your itinerary. diff --git "a/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" "b/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" new file mode 100644 index 00000000..1dbeb008 --- /dev/null +++ "b/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" @@ -0,0 +1,110 @@ +[← 回总目录](../README.md) + +# 22. 怎么放松：娱乐场所和减压 + +前半节讲娱乐场所里哪些钱花得冤、安全出口在哪，口径是金钱和人身自由。后半节讲怎么减压，口径是精力和总死亡率。两半的数字不放在一起换算。第 3 节已经写过情绪低落时先做哪几件事，这里给更具体的做法。 + +[← Back to the contents](../README.md) + +# 22. Relaxing: venues and stress relief + +The first half of this section covers which money in an entertainment venue is wasted and where the emergency exits are, measured in money and personal freedom. The second half covers how to take the pressure off, measured in energy and all-cause mortality. The numbers in the two halves are not converted into each other. Section 3 has already set out what to do first when your mood is low, and this section gives the more specific methods. + +### 1. Before you go into a KTV room, a bar or an escape room, look for the emergency exit, and change venue if it is locked or blocked + +- Cost: No money. Look up for 10 seconds when you walk in. +- In plain language: By law, while a venue is trading, the escape routes and emergency exits must be kept clear. They may not be blocked, may not be locked, and may not be piled with things in the way, and the signs may not be covered up. Internet cafés may not seal the doors or bar the windows either. A venue that does not comply is breaking the law, and you can report it to the authorities. Change venue if the private room is underground, if there is only one exit, or if the corridor is full of junk. +- Benefit: This is a statutory duty of the operator. A venue that does not comply is breaking the law, and you can report it to the authorities. The regulation reads: 「营业期间，娱乐场所应当保证疏散通道和安全出口畅通，不得封堵、锁闭疏散通道和安全出口，不得在疏散通道和安全出口设置栅栏等影响疏散的障碍物」, that is, while trading, a venue must keep escape routes and emergency exits clear, may not block or lock escape routes and emergency exits, and may not put up railings or other obstacles that impede escape. The regulation also requires that the venue 「应当在疏散通道和安全出口设置明显指示标志，不得遮挡、覆盖指示标志」, that is, put up clear signs at escape routes and emergency exits and not cover or obscure them. Internet cafés are the same. They are told 「营业期间禁止封堵或者锁闭门窗、安全疏散通道和安全出口」, that is, while trading, blocking or locking doors, windows, escape routes and emergency exits is forbidden, and 「禁止明火照明和吸烟」, that is, open-flame lighting and smoking are forbidden. +- Evidence grade: A +- Sources: State Council (国务院) (2006). Regulations on the Administration of Entertainment Venues, 娱乐场所管理条例（国令第 458 号, Articles twenty and twenty-one）. ; State Council (国务院) (2002). Regulations on the Administration of Business Premises for Internet Access Services, 互联网上网服务营业场所管理条例（国令第 363 号, Article twenty-one）. +- Notes: If any one of these is true, change venue: the private room is underground, there is only one exit, the corridor is piled with junk, or the emergency exit sign is unlit. A fire in a place like this means heavy smoke, narrow corridors and many people, and most of them have been drinking. The time you have to get out is measured in seconds. The regulation says the number to report to must be on the warning signs inside the venue. + +### 2. Ask for the price list before you order, which venues have to display, and refuse 「play first, pay later」 + +- Cost: No money. Ask for the list once. The hard part is asking the price in front of other people, which is embarrassing. +- In plain language: By law, a venue selling services and selling goods must display its prices and must show you the price list itself. So-called 「天价酒水」, sky-high drink prices, almost always happen where nobody looked at a price list and the bill was settled on a spoken quote afterwards. Ask for the list when you enter the private room and photograph it. Ask for a confirmed bill after you order. If there is a dispute about the bill, call 12315 there and then. +- Benefit: The regulation reads: 「娱乐场所提供娱乐服务项目和出售商品，应当明码标价，并向消费者出示价目表；不得强迫、欺骗消费者接受服务、购买商品。」, that is, where a venue provides entertainment services and sells goods, it must mark prices clearly and show consumers the price list, and may not force or deceive consumers into accepting services or buying goods. Breaches of price laws and regulations are punished by the competent departments under the law. Disputes over so-called 「天价酒水」 happen for the most part in exactly these settings: no price list shown, and the bill settled on a spoken quote afterwards. +- Evidence grade: A +- Sources: State Council (国务院) (2006). Regulations on the Administration of Entertainment Venues, 娱乐场所管理条例（国令第 458 号, Articles twenty-nine and fifty-three）. +- Notes: How to do it: ask for a price list when you enter the private room, paper or electronic, and photograph it as soon as you have it. After ordering, ask for a confirmed bill. If a waiter recommends a 「特调」, a special mix, or a 「果盘」, a fruit platter, always ask the price before ordering. Call 12315 there and then if there is a dispute about the bill. Do not pay first and defend your rights afterwards, because once you have paid, producing the evidence is much harder. + +### 3. If someone offers you 「something」 inside the venue, walk out at once; both letting someone use it and supplying it are not doing a friend a favour + +- Cost: No money. Stand up and walk out. The hard part is leaving the table in front of everybody. +- In plain language: A venue may not sell drugs, may not supply drugs, and may not let anyone use them inside. Breaking that means a suspension of business for 3 to 6 months, and in serious cases the licence is revoked. Using drugs yourself carries administrative punishment. Letting someone 「use it for a moment」 in a private room you have booked, or in your own home, is harbouring another person for drug use, and that is a crime. If somebody produces an unidentified powder, a tablet or a cigarette cartridge and asks you to try it, walk out at once. +- Benefit: The regulation expressly forbids a venue to 「贩卖、提供毒品，或者组织、强迫、教唆、引诱、欺骗、容留他人吸食、注射毒品」, that is, sell or supply drugs, or organise, force, induce, lure, deceive or harbour other people to take or inject drugs. Staff may not use drugs themselves, and may not provide the conditions for any of the above. A breach leads the public security authority to confiscate the illegal gains and illegal property and to order the venue to suspend business for 3 to 6 months. In serious cases the entertainment business licence is revoked. The regulation also requires venues to display, at a prominent place in the hall, in private rooms and in booths, warning signs against drugs, gambling, prostitution and solicitation, and to state the number to report to. +- Evidence grade: A +- Sources: State Council (国务院) (2006). Regulations on the Administration of Entertainment Venues, 娱乐场所管理条例（国令第 458 号, Articles fourteen, thirty and forty-two）. +- Notes: If somebody takes out an unidentified powder, a tablet or an electronic cigarette cartridge and asks you to 「试试」, try it, leave at once and do not stay to watch. The risk you carry personally is heavier than the venue's. Using drugs yourself is in itself subject to administrative punishment. Letting a friend 「use it for a moment」 in a private room you have booked, or in your own home, is harbouring another person for drug use, and that is a crime. + +### 4. Do not eat sweets and snacks handed to you by strangers, do not drink a drink that has been out of your sight, and do not take a cigarette cartridge passed to you + +- Cost: No money. Turn down everything handed to you. If your own glass has been out of your sight, get a new one. +- In plain language: Drugs today do not look like drugs. A hit-producing electronic cigarette laced with synthetic cannabinoids looks much like an ordinary e-cigarette, and people go round handing them out to try in bars, in KTV rooms and in hotels. Inducing someone to take them carries 10 to 15 days of detention and a fine of 1000 to 5000 yuan, and the person who was deceived into taking them is detained for 10 to 15 days as well. If a drink has been out of your sight, get a new one. +- Benefit: Drugs today do not look like drugs. Synthetic cannabinoids have been listed as a whole class since 2021-7-1, and this large group is now handled as drugs. Some people mix them into the e-liquid to make a hit-producing electronic cigarette. In a case reported by the Supreme People's Procuratorate, investigators described this e-liquid as something that 「可以通过添加食用颜料变成各种颜色、各种口味」, that is, can be turned into any colour and any flavour by adding food colouring, and 「在外观上和普通电子烟基本一致」, that is, looks much the same as an ordinary electronic cigarette. Sellers gather a group and push them in bars, in KTV rooms and in hotels, inducing others to take them. In that report, among the cases of drug use handled locally that year, people who were minors came to nearly forty per cent. The dissociative anaesthetic 依托咪酯 has been in the catalogue of second-category psychotropic drugs since 2023-10-1. Inducing someone to take drugs is in itself unlawful: 「引诱、教唆、欺骗或者强迫他人吸食、注射毒品的，拘留 10 到 15 日并罚 1000 到 5000 元」, that is, anyone who induces, incites, deceives or forces another person to take or inject drugs is detained for 10 to 15 days and fined 1000 to 5000 yuan. Where it amounts to a crime, the sentence is up to 3 years, 3 to 7 years in serious cases and 3 to 10 years where there is coercion, with heavier punishment where the victim is a minor. The consequences fall on the deceived person as well: 「吸食、注射毒品的，处 10 到 15 日拘留并可罚 3000 元以下」, that is, anyone who takes or injects drugs is detained for 10 to 15 days and may be fined up to 3000 yuan. In the lighter cases the penalty is up to 5 days of detention or a fine of up to 1000 yuan. A person may also be ordered not to enter entertainment venues for 6 months to 1 year. An addicted person may be ordered into community detoxification for 3 years. Anyone who refuses community detoxification, or who takes drugs again during it, is sent to compulsory isolated detoxification for 2 years, which may be extended by a further 1 year (nationwide, from month 1 of 2026 under the Public Security Administration Punishments Law). +- Evidence grade: A +- Sources: Supreme People's Procuratorate (最高人民检察院) (2021). Wujiang District, Suzhou, Jiangsu: prosecutors reveal new drugs dressed in trendy clothing. ; National Medical Products Administration, Ministry of Public Security and National Health Commission (国家药监局、公安部、国家卫生健康委) (2023). Announcement on Adjusting the Catalogue of Narcotic Drugs and Psychotropic Drugs (2023 No. 120). (reprinted by the Shanghai Municipal Drug Administration, which records that the original page on nmpa.gov.cn returned 412 on this machine); Standing Committee of the National People's Congress (全国人大常委会) (2025). Public Security Administration Punishments Law (2025 revision, Articles eighty-four and eighty-five). ; National People's Congress (全国人大) (1997). Criminal Law (Article three hundred and fifty-three). ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Narcotics Control Law (Articles thirty-three, thirty-eight and forty-seven). (reprinted by the Guangdong Provincial People's Government). +- Notes: This is not only about entertainment venues. School gates, gatherings with friends and long-distance coaches are the same. Whether what is handed to you is a sweet, a drink or a cigarette cartridge makes no difference. Being drugged is often followed by robbery, sexual assault or blackmail, and there are no published official figures for that, so this is a judgement from experience. Replacing a drink the moment it has been out of your sight saves far more trouble than arguing afterwards. For what to do when somebody in the venue produces an unidentified powder, tablet or cigarette cartridge, see item 3 in this section. For carrying things for other people, see item 29 in section 8. + +### 5. Use your own identity card at an internet café, and neither lend yours nor use someone else's + +- Cost: No money. +- In plain language: By law, an internet café must check and register the identity card of each person who gets online, and must record the details of the session, keeping them for at least 60 days. The public security and culture departments can ask for them at any time. While that period runs the café may not alter or delete the records. Whatever you use to get online, what was done on that machine in that period is recorded under that name. +- Benefit: The regulation reads: 「互联网上网服务营业场所经营单位应当对上网消费者的身份证等有效证件进行核对、登记，并记录有关上网信息。登记内容和记录备份保存时间不得少于 60 日」, that is, an operator of a business premises providing internet access services shall check and register the valid documents of the consumers using it, such as their identity cards, and shall record the relevant details of their use. The registration and the backup of the records must be kept for not less than 60 days. When the cultural administration department or the public security authority queries them lawfully, the café must supply them. During the retention period they may not be altered or deleted. So whatever you use to get online, what was done on that machine in that period is recorded under that name. +- Evidence grade: A +- Sources: State Council (国务院) (2002). Regulations on the Administration of Business Premises for Internet Access Services, 互联网上网服务营业场所管理条例（国令第 363 号, Article twenty-three）. +- Notes: This is the same thing as 「身份证不借人」 in section 9. Internet cafés may not admit minors, and a sign barring minors has to hang at a visible place by the entrance. Entertainment venues likewise do not admit minors. If you help a minor 「刷卡进网吧」, get into an internet café with your own card, the café is fined, but the record stays on your document. + +### 6. Do not book murder mystery games or escape rooms set inside a residential building or below the second basement level + +- Cost: No money. Ask which floor the address is on before you book. +- In plain language: A 2022 notice from five departments states that a script entertainment venue may not be set up inside a residential building, nor below the first underground level. These places are dark, full of partitions and full of flammable props, and the door often only opens once you have solved a puzzle. Once there is a fire, getting out is much harder than from an ordinary shop. Look at the address and the floor on the review page before you book, and change venue if it does not comply. +- Benefit: The dedicated 2022 notice from the five departments is specific: 「剧本娱乐经营场所不得设在居民楼内、建筑物地下一层以下（不含地下一层）等地」, that is, a script entertainment venue may not be set up inside a residential building, below the first underground level of a building (excluding the first underground level itself), or in similar places. The notice also requires venues to meet the requirements of the Work Safety Law and the Fire Protection Law, to 「常态化开展火灾风险自知、自查、自改，提高紧急情况下的组织疏散逃生和初起火灾扑救能力」, that is, routinely identify, inspect and correct fire risks themselves, and improve the ability to organise evacuation and escape in an emergency and to put out a fire in its early stage. The same notice also requires scripts to carry an age suitability notice. A scene unsuitable for minors may not admit minors, and apart from statutory holidays, rest days and the winter and summer holidays, script entertainment activities may not be offered to minors. +- Evidence grade: A +- Sources: Ministry of Culture and Tourism, Ministry of Public Security, Ministry of Housing and Urban-Rural Development, Ministry of Emergency Management and State Administration for Market Regulation (文化和旅游部、公安部、住房和城乡建设部、应急管理部、市场监管总局) (2022). Notice on Strengthening the Management of Script Entertainment Venues, 关于加强剧本娱乐经营场所管理的通知. +- Notes: These places are dark, full of partitions and full of flammable props, and the door often needs a puzzle to be solved before it opens. Once there is a fire, getting out is much harder than from an ordinary shop. You can see the floor, and whether it is a residential building, on the review page before you book. If it does not comply, change venue. + +### 7. When your mood is bad, walk or run: the effect size of exercise against depression rises with the intensity + +- Cost: No money. Half an hour each time. The hard part is that when your mood is worst is exactly when you least want to move, so you have to push yourself. +- In plain language: When your mood is bad, brisk walking or slow running shows the clearest improvement. Next come yoga, strength training and tai chi or qigong, and the differences between them are small. The higher the intensity, the better the effect, and strength training and yoga are the easiest to keep up. The evidence here is of low quality. What can be said is that moving beats not moving, so pick the kind you can keep up. +- Benefit: This is a pooled result across 218 randomised controlled trials. It covers 495 groups and 14170 participants meeting the clinical criteria for severe depression. The method lets the different kinds of exercise compete with one another, in a Bayesian network meta-analysis. The comparison is against usual care and against a placebo tablet as an active control. Effect size is expressed as Hedges' g, and the more negative the number, the greater the improvement in symptoms. Brisk walking or slow running −0.62 (95% confidence interval −0.80 to −0.45, which means the true value falls roughly within this range). Yoga −0.55 (−0.73 to −0.36). Strength training −0.49 (−0.69 to −0.29). Mixed aerobic work −0.43 (−0.61 to −0.24). Tai chi or qigong −0.42 (−0.65 to −0.21). The text says 「运动的效果与所处方的强度成正比」, that is, the effect of exercise is proportional to the intensity prescribed, so the higher the intensity the better the effect. Strength training and yoga were rated the most acceptable and the easiest to keep up. +- Evidence grade: A +- Sources: Noetel M, Sanders T, Gallardo-Gómez D, et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ, 384, e075847. +- Notes:Disputed. The authors say so themselves: of the 218 studies, only one was clean enough not to be swayed by its design. On the standard rating scale, brisk walking and slow running count only as "low" certainty, and the rest as "very low". Participants cannot be blinded to exercise, and knowing they are exercising makes them feel better; that expectation cannot be removed. The direction of the conclusion still holds: moving beats not moving, keep the intensity up, and pick the kind you can keep up. This is the expansion of 「情绪低落时先动起来」 in section 3, and it does not replace proper treatment after a diagnosis. + +### 8. When anxiety is affecting your life, a mindfulness-based stress reduction course is no worse than the usual anti-anxiety drug + +- Cost: The course costs a little money. It runs 8 weeks, one session a week, and you practise at home every day. The hard part is practising every day for 8 weeks running. +- In plain language: In people diagnosed with an anxiety disorder, 8 weeks of mindfulness-based stress reduction improves symptoms almost as much as the anti-anxiety drug escitalopram. Nearly eighty per cent of the people on the drug had an adverse reaction, and some dropped out because of it. About fifteen per cent of the people on the course had one, and nobody dropped out. The course has to be a standardised 8-week course; an ordinary meditation app does not count. +- Benefit: The trial was run by academic medical centres in 3 US cities, to see whether the course comes out worse than the drug, in a non-inferiority randomised controlled trial. 276 adults diagnosed with an anxiety disorder were randomised 1:1 into two equal groups. One group took an 8-week mindfulness-based stress reduction course. The other took escitalopram at a flexible dose of 10 to 20 mg. 208 people completed it. At the end, scored on the clinician's rating of anxiety severity (Clinical Global Impression Severity Scale), the mindfulness group fell by 1.35 and the drug group by 1.43. The difference between the two groups is −0.07 (95% CI −0.38 to 0.23, which means the true difference falls roughly within this range; P=0.65, which means this difference cannot be told apart from no difference). The lower end of this range did not cross the line drawn in advance for "not worse" (the non-inferiority margin −0.495). The difference in side effects is large: 78.6% of the drug group had at least one adverse reaction related to the trial, and 10 dropped out because of it. In the mindfulness group the figure was 15.4%, and nobody dropped out. +- Evidence grade: A +- Sources: Hoge EA, Bui E, Mete M, Dutton MA, Baker AW, Simon NM (2023). Mindfulness-based stress reduction vs escitalopram for the treatment of adults with anxiety disorders: a randomized clinical trial. JAMA Psychiatry, 80(1), 13-21. +- Notes: The trial used the standardised 8-week mindfulness-based stress reduction course (MBSR), and an ordinary meditation app is not the same thing. It applies to people who already meet the diagnostic criteria for an anxiety disorder, and if you need a doctor you still need one. If you have been putting off treatment out of worry about drug side effects, this is an option with comparable strength of evidence. + +### 9. To settle down in the moment, do 5 minutes of 「cyclic sighing」: a two-stage inhale and a lengthened exhale + +- Cost: No money. 5 minutes a day. +- In plain language: 5 minutes of breathing practice a day, kept up for a month, improves mood and slows breathing more than mindfulness meditation of the same length. Of the breathing methods, 「循环叹息」, cyclic sighing, which lengthens the exhale, is the best one. To do it, breathe in through the nose, then take a short top-up breath to fill the lungs, then breathe out slowly through the mouth. This rests on a single trial, where the rating comes from the participant himself, and it costs nothing. +- Benefit: This is a randomised controlled trial carried out at home, comparing three kinds of 5-minute daily breathing practice with mindfulness meditation of the same length over 1 month. The three breathing methods are cyclic sighing, which emphasises a lengthened exhale, box breathing, in which inhaling, holding and exhaling are equal, and cyclic hyperventilation with a hold, in which the inhale is long and the exhale short. Pooling the data across people and across sessions in a mixed-effects model, the breathing practices all did better than mindfulness meditation, and cyclic sighing, which puts the weight on the exhale, did best of all. It was better on mood improvement (P<0.05, which means the difference does not look like it came out by chance) and on the fall in breathing rate (P<0.05). +- Evidence grade: B +- Sources: Balban MY, Neri E, Kogon MM, et al. (2023). Brief structured respiration practices enhance mood and reduce physiological arousal. Cell Reports Medicine, 4(1), 100895. +- Notes: To do it, breathe in through the nose, take a short top-up breath to fill the lungs, then breathe out slowly and fully through the mouth, and repeat for 5 minutes. There is only this one trial. It is done at home, whether it works is reported by the person filling in the questionnaire, and the number of people is small. The effect is only a little better than meditation and it does not solve problems. But it costs nothing, takes 5 minutes and can be done anywhere, so it is good value for money. + +### 10. Treat 「meeting people regularly」 as a health expense, and do not only reach out when your mood is bad + +- Cost: A little money, plus the time spent meeting people. The hard part is arranging to see people when your mood is good. +- In plain language: People who rarely see other people have a risk of death over the same period about thirty per cent higher. People who feel lonely are about a quarter higher, and people who live alone about thirty per cent higher, and the pattern is more marked among people under 65. All of these come from following people over time, and it may be that people who are already in poor health are the ones more likely to live alone. +- Benefit: This is a pooled result across 70 studies that first recorded a situation and then followed people forward, a meta-analysis. It covers 3,400,000 participants followed for an average of 7 years. Social isolation, loneliness and living alone are all associated with higher mortality. After taking out confounders such as age and health status, the weighted mean odds ratios are 1.29 for social isolation, 1.26 for loneliness and 1.32 for living alone. Converted, that is a risk of death over the same period roughly thirty per cent, a quarter and thirty per cent higher respectively. The effect sizes of the three are close to one another, and larger among people under 65 than among older people. +- Evidence grade: A +- Sources: Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D (2015). Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspectives on Psychological Science, 10(2), 227-237. +- Notes:Disputed. All of this comes from associations observed in groups of people rather than from a controlled trial, so cause and effect cannot be separated. People who are already in poor health are more likely to live alone and to have fewer social contacts. The authors treat social relationships as a health factor of the same order as smoking and obesity, but no randomised trial has shown that increasing social contact lowers mortality. This does not conflict with 「减少让你消耗的人际关系」 in section 3: that one cuts the relationships that drain you, and this one protects the few that restore your energy. + +### 11. If you live near green space, make more use of it: walking, a stroll and sitting down all count + +- Cost: No money. You do it on the way past. +- In plain language: The better the greenery within 500 metres of your home, the lower the risk of death over the same period. For each 0.1 rise in the vegetation index it is about 4% lower. What is measured is the greenery around the home, not how many times you go to a park. People who live in greener neighbourhoods are often richer, and the air is better as well. So do not read this as going to a park extending your life; it is enough to take the green space around your home into account when you choose where to live. +- Benefit: This is a pooled result across 9 studies that followed groups of people for years, a meta-analysis of cohort studies. It covers 7 countries and 8,320,000 people. What it looks at is the greenery within 500 metres of the home, measured with the Normalised Difference Vegetation Index (NDVI, a single number that converts ground greenery into a figure). For each 0.1 rise in the index, the pooled risk ratio for all-cause mortality is 0.96 (95% CI 0.94 to 0.97), a risk of death over the same period about 4% lower, with the true value falling roughly between 3% lower and 6% lower. 7 of the 9 studies found that greener meant lower mortality, and the difference did not look like chance. The other 2 found no association. +- Evidence grade: A +- Sources: Rojas-Rueda D, Nieuwenhuijsen MJ, Gascon M, Perez-Leon D, Mudu P (2019). Green spaces and mortality: a systematic review and meta-analysis of cohort studies. The Lancet Planetary Health, 3(11), e469-e477. +- Notes:Disputed. The results across the studies are a long way apart (I²=95%, and the larger that number is, the less they agree with one another). And what it measures is the greenery around where you live, not how many times you go to a park. People who live in greener places are often also wealthier, have better air and exercise more, and none of that can be separated from the greenery. Do not read it as going to a park extending your life. What you can use is this: take the green space into account when you choose where to live, and it costs nothing. On the NDVI figure, bare ground and concrete come close to 0 and dense vegetation close to 1. So a step of "0.1" is roughly the difference between a bare block and a block with a row of mature trees. diff --git "a/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" "b/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" new file mode 100644 index 00000000..fbffffc3 --- /dev/null +++ "b/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" @@ -0,0 +1,218 @@ +[← 回总目录](../README.md) + +# 23. 学什么技能划算 + +这一节算的是钱和时间，第 2 条另外算死亡率。前六条回答一个问题：还要不要接着读书。第 1 条先说法律划掉了哪一段选项，第 2 条算读书和寿命的关系，第 3 条看全国人的学历都在哪一档。第 4 条讲读不起的时候有哪些资助，第 5 条讲没考上普高还有什么路，第 6 条把「读书还是打工」算成一道题，给一个自己动手算的方法。后七条回答另一个问题：决定学点东西之后怎么挑。中国按工种分的工资统计，从公开的官方渠道很难逐项核实。所以这里只写查得到出处的几件事：多读一年书工资大概多多少（也叫教育回报率）、证书合不合法怎么验、培训补贴从哪里来、挑技能时看哪个维度更不容易被顶掉。具体某个工种一个月挣多少，本书不给数字。第 14 到 19 条回答第三个问题：定下来学什么之后，怎么学才省时间。这六条算的是时间，不和钱做换算。第 20 到 23 条回答第四个问题：已经在做专业技术工作了，职称从哪里报、怎么考、别踩什么坑、评上值多少。 + +[← Back to the contents](../README.md) + +# 23. Which skills are worth learning + +This section measures money and time, and item 2 measures mortality as well. The first six items answer one question: whether to carry on with education. Item 1 sets out which part of the options the law removes. Item 2 works out the relationship between education and lifespan. Item 3 looks at where the national education level sits. Item 4 covers the financial help available when you cannot afford it. Item 5 covers what is left if you do not get into a general senior high school. Item 6 turns "study or work" into a question with a method you can work through yourself. The next seven items answer a different question: how to choose once you have decided to learn something. China's wage statistics by occupation are hard to verify item by item through public official channels, so only the things that can be traced to a source are written here: how much an extra year of education is worth in wages, also called the return to education, how to check whether a certificate is legal, where training subsidies come from, and which dimension to look at when choosing a skill so that it is harder to replace. This book gives no figure for what a particular trade earns in a month. Items 14 to 19 answer a third question: once you have settled what to learn, how to learn it without wasting time. Those six items measure time and are not converted into money. Items 20 to 23 answer a fourth question: if you already work in a technical or professional field, where a professional title is applied for, how the exam works, what traps to avoid, and what it is worth once you have it. + +### 1. Under 16 years old there is no "go out to work" option: an employer is fined 5000 yuan a month for hiring you, and the employers who do hire are running it black + +- Cost: No money. It takes a minute to work out how old you are. +- In plain language: Finishing junior high school usually happens at 15 or 16, which sits exactly on this line. An employer who takes you on under 16 is breaking the law, and hiring one child worker carries a fine of 5000 yuan for every month of employment. So the employers who are willing to take you are the ones who know they are breaking the law and do it anyway. If something does happen, the medical bills and the compensation can only be claimed from that employer, and if it defaults or runs off, nobody is standing behind it. +- Benefit: Article eleven of the Compulsory Education Law: 「凡年满六周岁的儿童，其父母或者其他法定监护人应当送其入学接受并完成义务教育」, that is, every child who has reached the age of six shall be sent to school by their parents or other legal guardians to receive and complete compulsory education. Article fourteen: 「禁止用人单位招用应当接受义务教育的适龄儿童、少年」, that is, employers are prohibited from hiring children and adolescents of an age that should be receiving compulsory education. Article fifty-eight: where school-age children and adolescents are not sent to school without a proper reason or in accordance with this Law, the township people's government at the local level, or the education administrative department of the county-level people's government, 「给予批评教育，责令限期改正」, that is, gives criticism and education and orders the correction to be made within a set time. Article two of the Provisions on the Prohibition of Child Labour: state organs, social organisations, enterprises and public institutions, private non-enterprise units and individual industrial and commercial households 「均不得招用不满16周岁的未成年人」, that is, may not hire anyone under 16 years old. Article six: punishment is imposed 「按照每使用一名童工每月处5000元罚款的标准给予处罚」, that is, at the standard of a fine of 5000 yuan for every month in which one child worker is employed, and where the correction is not made by the deadline it is 10,000 yuan per person per month. Article seven: anyone who introduces a child worker into employment 「按照每介绍一人处5000元罚款」, that is, is fined 5000 yuan for each person introduced. Article ten: where a child worker falls ill or is injured, the employer 「应当负责送到医疗机构治疗，并负担治疗期间的全部医疗和生活费用」, that is, must take responsibility for taking the child to a medical institution for treatment and for bearing all the medical and living costs during the treatment, and must also make a one-off payment where the child is permanently disabled or dies. Article thirteen leaves a single exception: arts and sports units, with the consent of the parents or other guardians, may hire professional artistic workers and athletes under 16 years old (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). Compulsory Education Law (second revision of 2018, Articles eleven, fourteen and fifty-eight). (published by the Ministry of Education); State Council (国务院) (2002). Provisions on the Prohibition of Child Labour (State Council Order No. 364, Articles two, six, seven, ten and thirteen). +- Notes: The person who benefits is yourself. Working after you turn 16 is lawful. But the jobs that are keen to take on a 16-year-old with no qualification are exactly the ones described in item 10 in this section (choosing skills that resist automation): fixed processes, the easiest for a machine to take over. The overtime pay, annual leave and work injury benefits described in section 19 all depend on there being an employment relationship and on evidence. A black job gives you neither, and those are the things you cannot get. + +### 2. Put "is studying worth it" into the mortality account too: each extra year of education lowers the adult risk of death by about 1.9% + +- Cost: No money. What it costs is the time and energy of those extra years of study; for the tuition, see item 4 in this section (secondary vocational tuition and grants). The hard part is carrying on without giving up in the middle. +- In plain language: Each extra year of study lowers the risk of death in adulthood by about 1.9% on average. The effect is clearest between 18 and 49, where each year is worth about 2.9%, and above 70 it is only 0.8%. People who can keep studying are often better off to begin with, and how much of this is the effect of the study itself cannot be separated. +- Benefit: All the earlier studies were gathered up and pooled together, a systematic review and meta-analysis. The search ran from month 1 of 1980 to month 5 of 2023, and finally included 603 studies, 70 locations in 59 countries and 10,355 observations. Education and adult mortality, which is deaths from all causes added together, show a dose-response relationship: the more study, the larger the fall. Each extra year of education lowers the risk of death by 1.9% on average (95% uncertainty interval 1.8 to 2.0, which means the true value falls roughly within this range). For people aged 18 to 49 each extra year lowers it by 2.9% (2.8 to 3.0). For people over 70 the fall is 0.8% (0.6 to 1.0). No difference was seen between men and women. Split by national level of development, using the Social Population Index, no difference was seen either. +- Evidence grade: A (disputed) +- Sources: IHME-CHAIN Collaborators (2024). Effects of education on adult mortality: a global systematic review and meta-analysis. The Lancet Public Health, 9(3), e155-e165. +- Notes:Disputed. The authors report publication bias themselves (p<0.0001), which means the studies with the prettier results are likelier to get published. The results also differ a long way across the studies. Everything included here is observational follow-up, not a trial that assigned people to groups. Years of schooling are tangled up with family income, region and occupation, and this kind of study cannot separate how much of the 1.9% comes from the schooling itself and how much comes from the fact that people who can keep studying were better off to begin with. So it shows strongly that more schooling and a longer life point the same way, but it does not mean that if you go and study you will live 1.9% longer. The person who benefits is yourself. + +### 3. Before deciding whether a qualification has lost its value, look at the national education structure: in every 100,000 people only 15,467 have a university-level education + +- Cost: No money. You only have to look at the figures below. +- In plain language: The 2020 census found that across the country only 15467 people in every 100,000 went on to college or above, about 15%. 34507 had finished junior high, which is the largest group of all. People over 15 had been in school for 9.91 years on average, just past junior high. So a junior high qualification in China is not "good enough". It is the most ordinary level there is. +- Benefit: The standard reference date of the seventh national census was 2020-11-1. Of the national population, 218360767 people had a university-level education, that is college and above. 213005258 had senior high, including technical secondary schools. 487163489 had junior high. 349658828 had primary school. Compared with the sixth national census of 2010, the number with a university-level education in every 100,000 rose from 8930 to 15467. Senior high rose from 14032 to 15088. Junior high fell from 38788 to 34507. Primary school fell from 26779 to 24767. The average years of schooling of the population aged 15 and above rose from 9.08 years to 9.91 years. The illiteracy rate fell from 4.08% to 2.67% +- Evidence grade: A +- Sources: National Bureau of Statistics (国家统计局) and the Office of the Leading Group for the Seventh National Population Census of the State Council (国务院第七次全国人口普查领导小组办公室) (2021). Communiqué of the Seventh National Population Census, 第七次全国人口普查公报（第六号）: education level of the population, 人口受教育情况. +- Notes: These figures only say how scarce a qualification is among the population. Whether studying is worth it for a particular person has to be worked out with the method in item 6 in this section (how to calculate study or work). One more thing to note: the census counts the whole national population, old people and small children included. The share of young people who go on to university is much higher than this figure, so comparing it with your own age group overstates your advantage. Used the other way round, it does answer the claim that studying is useless, because the denominator is the whole labour force competing with you for the same jobs. The person who benefits is yourself. + +### 4. When you cannot afford it, work through the policy first: most secondary vocational tuition is already free, grants are 2300 yuan, and student loans go up to 20,000 yuan a year + +- Cost: No money. Before term starts, ask the school's student financial aid office once and hand in the papers once. +- In plain language: In public secondary vocational schools, rural students, urban students in agriculture-related majors and students from families in financial difficulty are already exempt from tuition. For secondary vocational schools and general senior high schools, the grant is 2300 yuan a year from the spring term of 2025. At university the grant is 3700 yuan a year, and a student loan can be for up to 20,000 yuan a year. If the money is not enough you can defer, or find another way round. It rarely comes to having to drop out now and go out to work. +- Benefit: Exempting secondary vocational students from tuition is policy that has applied since the autumn term of 2012. The text reads: 「对公办中等职业学校全日制正式学籍一、二、三年级在校生中所有农村（含县镇）学生、城市涉农专业学生和家庭经济困难学生免除学费（艺术类相关表演专业学生除外）」, that is, tuition is waived for all rural students including those from county and town seats, urban students in agriculture-related majors and students from families in financial difficulty, among the full-time students officially enrolled in years one, two and three of public secondary vocational schools, excluding students in performance-related arts majors. The student financial aid policy was adjusted in 2024. The national grant for undergraduate and junior college students was raised from 3300 yuan per student per year to 3700 yuan. The National Encouragement Scholarship was raised from 5000 yuan to 6000 yuan. Both apply from the autumn term of 2024, with the scholarship from 2024. The national grant for secondary vocational schools and the national grant for general senior high schools were both raised from 2000 yuan per student per year to 2300 yuan, applying from the spring term of 2025. National student loans were adjusted too, from the autumn term of 2024. For undergraduate and junior college students, 「每人每年申请贷款额度由不超过16000元提高至不超过20000元」, that is, the amount each student may apply for each year was raised from not more than 16000 yuan to not more than 20000 yuan. For postgraduate students, 「由不超过20000元提高至不超过25000元」, that is, from not more than 20000 yuan to not more than 25000 yuan. The interest rate is the Loan Prime Rate (LPR) for the same period and grade less 70 basis points, that is 0.7 percentage points lower (nationwide) +- Evidence grade: A +- Sources: Ministry of Finance, National Development and Reform Commission, Ministry of Education and Ministry of Human Resources and Social Security (财政部、发展改革委、教育部、人力资源社会保障部) (2012). Opinions on Expanding the Scope of the Tuition-Free Policy for Secondary Vocational Education and Further Improving the National Grant System (Cai Jiao [2012] No. 376). ; the relevant department heads of the Ministry of Finance, the Ministry of Education and the Ministry of Human Resources and Social Security (财政部、教育部、人力资源社会保障部有关司局负责人) (2024). Answers to reporters on the questions of adjusting the student financial aid policy. ; Ministry of Finance, Ministry of Education, People's Bank of China and National Financial Regulatory Administration (财政部、教育部、人民银行、金融监管总局) (2024). Notice on Adjusting and Improving the Relevant Policies on National Student Loans (Cai Jiao [2024] No. 188). +- Notes: The people who benefit are yourself and your children. Which students the tuition waiver covers and how many bands the grant has are both set by each province. Private schools are subsidised up to the standard of the equivalent public course locally, and you still pay the difference, so check the local documents before you enrol. A student loan is a loan, not a grant, and it has to be repaid after you graduate. But it separates "I cannot pay the tuition now" from "are these years worth it", so do not treat running out of money as the conclusion. The training subsidies and temporary assistance available during unemployment are covered in section 7; if the household really has lost its income, apply for those first. + +### 5. Not getting into a general senior high school does not mean the road ends: secondary vocational schools have linked admissions and separate examinations, and skilled jobs may lower the education requirement + +- Cost: No money. Before you fill in the application choices, check the vocational education admissions platform in your province once. +- In plain language: The 2022 Vocational Education Law states that vocational education and general education are of equal importance. Secondary vocational schools can take students through linked admissions into higher vocational schools, and higher vocational schools must set aside part of their admissions plan or hold a separate examination specifically for secondary vocational graduates. Employers may not put a barrier in your way because you graduated from a vocational school. Missing the points for a general senior high school is a change of route, not stopping at a junior high qualification. +- Benefit: The Vocational Education Law, in force from 2022-5-1, Article three: 「职业教育是与普通教育具有同等重要地位的教育类型」, that is, vocational education is a type of education of equal importance to general education. Article thirty-seven: secondary vocational schools 「可以按照国家有关规定，在有关专业实行与高等职业学校教育的贯通招生和培养」, that is, may, as the state provides, run linked admissions and training with higher vocational education in the relevant majors. Higher vocational schools 「可以按照国家有关规定，采取文化素质与职业技能相结合的考核方式招收学生」, that is, may, as the state provides, admit students using an assessment that combines academic qualities with vocational skills. Education administrative departments at provincial level and above have to set up a unified vocational education admissions platform, publish school, major and admissions information together, and provide a service for looking up and applying. Article fifty-three: 「职业学校学生在升学、就业、职业发展等方面与同层次普通学校学生享有平等机会」, that is, students in vocational schools enjoy equal opportunities with students in ordinary schools at the same level in terms of moving on to further study, employment and career development. Higher vocational schools and ordinary higher education institutions that provide vocational education 「应当在招生计划中确定相应比例或者采取单独考试办法，专门招收职业学校毕业生」, that is, shall set a corresponding share in the admissions plan or use a separate examination to admit vocational school graduates specifically. And 「用人单位不得设置妨碍职业学校毕业生平等就业、公平竞争的报考、录用、聘用条件」, that is, employers may not set application, hiring or appointment conditions that hinder equal employment and fair competition for vocational school graduates. Posts in public institutions advertised openly that carry a vocational skill level requirement 「可以适当降低学历要求」, that is, may lower the education requirement appropriately (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2022). Vocational Education Law (2022 revision, Articles three, thirty-seven and fifty-three). (published by the Ministry of Education) +- Notes: The people who benefit are yourself and your children. These are channels created by law, not a guarantee of admission. Which majors run linked training, what share is set aside for separate admissions and which subjects the vocational education examination covers are all set by each province. Before you fill in the choices you must check your own province's vocational education admissions platform and admissions plan for that year, rather than reasoning from another province's experience. When choosing a secondary vocational major, run it through the dimensions in item 10 in this section (choosing skills that resist automation): a major that puts you on site and in a non-standard environment is harder to replace. To choose a school, look at whether it appears in your province's admissions plan for that year. Then verify the certificates it issues with the three steps in item 8 (check whether a certificate is in the catalogue). Ignore any admissions publicity promising 「包分配」, a guaranteed placement, or 「保过」, a guaranteed pass. + +### 6. Turn "study or work" into a question with numbers: the wages you earn three years earlier against the income gap over the following decades + +- Cost: No money. Sit down and do the sum once. It takes about half an hour. +- In plain language: Lay the two choices out on paper and compare them. Starting work earlier earns you the wages of three years in low-entry jobs, and only that once. Carrying on with study buys back the slice you earn extra each year afterwards, multiplied by the thirty or forty years that remain. In urban China an extra year of study brings average earnings 4.0% higher (1988) to 10.2% (2001). The data is old, so it gives you an order of magnitude only. +- Benefit: This is an estimate for urban China from 1988 to 2001. How much an extra year of study raises earnings is called the return to education. The text says 「从 1988 年每受教育一年 4.0% 大幅上升到 2001 年的 10.2%」, that is, it rose sharply from 4.0% for each year of education in 1988 to 10.2% in 2001. Most of the rise came after 1992. What it reflects is the extra slice earned by people who go on to university compared with those who do not, that is, the higher education wage premium, getting larger. This is the domestic Chinese order of magnitude you can use when converting an extra year of study into an annual income gap. For the global average, see item 7 in this section (the global average return to an extra year of education) +- Evidence grade: B +- Sources: Zhang J, Zhao Y, Park A, Song X (2005). Economic returns to schooling in urban China, 1988 to 2001. Journal of Comparative Economics, 33(4), 730-752. +- Notes: Before you do the sum, take off the tuition waivers and grants in item 4 in this section (secondary vocational tuition and grants), and set against them the account in item 2 in this section (an extra year of education and mortality). The data runs up to 2001, covers urban areas only and cannot be used as today's figures. The global average of 9% in item 7 of this section is the same. If you really work it through this way, the answer differs from person to person. High tuition, a major that does not match the job and a long course can come out genuinely not worth it, but that shows only that this particular programme is not worth it, and says nothing about study being useless. Several things also sit outside the formula: the education thresholds set by law, see item 5 in this section (linked admissions and separate examinations in secondary vocational schools); resistance to being replaced, see item 10 in this section (choosing skills that resist automation); and whether you can change industry. The person who benefits is yourself. + +### 7. Keep the baseline in mind: the global average private return to an extra year of education, the part that lands on your own income, is about 9% a year + +- Cost: No money. Keep this figure in mind and use it as a ruler for weighing up any "learn something" plan. +- In plain language: On a global average, an extra year of study puts about 9% a year on your own income, and the return for women has been higher than for men throughout. This figure is about formal schooling, so it does not apply directly to a course you sign up for. Use it as a ruler: if a course costs you 30,000 yuan and half a year, ask whether the income change it brings matches what one extra year of study would bring. +- Benefit: A review covering 1950 to 2014, 139 countries and 1120 estimates works out that 「多受一年教育的全球平均私人回报率是每年 9%」, that is, the global average private rate of return to an extra year of education is 9% a year. The private return is the part that lands on the person's own income: an extra year of study puts about 9% on average on income. The private return to higher education is rising, while the social return, the part that lands on society as a whole rather than on the person, remains high. The average return to education for women has continued to exceed that for men. +- Evidence grade: A +- Sources: Psacharopoulos G, Patrinos HA (2018). Returns to investment in education: a decennial review of the global literature. Education Economics, 26(5), 445-458. ; World Bank Policy Research Working Paper 8402. +- Notes: This is the global average for an extra year of study. It is not the return on signing up for a course, and it is not a Chinese figure; it is an order of magnitude for comparison. Its use is to give every training plan one ruler: if a course costs you 30,000 yuan and half a year, you should ask whether it can bring an income change equivalent to one year of formal education. Note too that a global average hides large differences between fields and between countries. + +### 8. Before paying for a certificate, check whether it is in the national catalogue of vocational qualifications or on the Ministry of Human Resources list of registered assessment bodies + +- Cost: No money. It takes a check of ten-odd minutes. +- In plain language: Some organisations run courses and charge fees and hand out certificates freely in the name of new occupations and skill levels, and some impersonate authoritative bodies. The Ministry of Human Resources and Social Security has run a campaign against this. Before you sign up, check three things. Is this occupation in the national catalogue of vocational qualifications? Is the issuing body an assessment body registered with the human resources department? Can the certificate number be found on an official platform? If not, the enrolment fee is money thrown away. +- Benefit: The Ministry of Human Resources and Social Security has run a nationwide campaign on exactly this. The document names the abuses: 「一些机构和单位以新职业职业资格、职业技能等级等名义随意举办培训、评价、发证活动，乱收费、滥发证；一些机构和单位虚假或夸大宣传，甚至假冒权威机构名义组织培训、评价、颁发证书等」, that is, some organisations and units run training, assessment and certification activities arbitrarily in the name of new occupational qualifications and vocational skill levels, charging fees without order and issuing certificates without restraint, while others advertise falsely or exaggerate, or even organise training, assessment and certification in the name of authoritative bodies. The measures include joint enforcement with the cyberspace, public security and market regulation departments, 「严厉打击假冒鉴定机构、人社部门备案评价机构的评价发证活动」, that is, severely cracking down on assessment and certification activities that impersonate appraisal bodies or registered assessment bodies of the human resources department. Non-compliant publicity on websites and in new media is shut down under the law. The document also requires the authorities to 「引导公众了解证书查询渠道和方式，判断证书合法合规性和含金量」, that is, guide the public to the channels and methods for looking up a certificate, and to judge whether a certificate is lawful and compliant and what it is worth. +- Evidence grade: A +- Sources: Ministry of Human Resources and Social Security (人力资源社会保障部) (2022). Notice on the Special Governance of "Boutique Certificates" in Technical and Vocational Skills. +- Notes: Check in this order. First, is this occupation in the National Catalogue of Vocational Qualifications (《国家职业资格目录》)? If it is not in the catalogue, it is not a national vocational qualification. Second, is the issuing body an assessment body registered with the human resources department? Third, go to the official platform for looking up skill talent assessment certificates and see whether the certificate number can be found. If it cannot, do not enrol at all. Anything carrying 「国际」, international, 「高级」, senior, or 「注册」, registered, and promising 「包过」, a guaranteed pass, or 「挂靠有收入」, income from affiliation, carries the highest risk. + +### 9. For training, go through the government subsidy channel first, and do not pay for a commercial course straight away + +- Cost: No money, or a large reduction. The training itself takes time. The hard part is resisting the urge to pay for a course on impulse and looking up the subsidy catalogue first. +- In plain language: From 2025 to the end of 2027, the state plans to run subsidised training for more than 30 million places. The money comes from several channels, among them employment subsidy funds, the unemployment insurance fund and enterprises' employee education funds. The scarcer the trade and the higher the cost of the training, the larger the subsidy. If you are employed, ask your employer first. Everyone else should look through the subsidy catalogue of the local human resources and social security bureau and then decide what to learn. +- Benefit: The national scale is clear. From 2025 to the end of 2027, the state will 「聚焦高精尖产业与急需行业、就业重点群体等开展补贴性培训 3000 万人次以上」, that is, run subsidised training for more than 30 million places focused on advanced and specialised industries, urgently needed trades, and groups that are a priority for employment. The funding sources have been pooled into a single package: employment subsidy funds, the unemployment insurance fund, industry development funds, east-west cooperation funds, enterprises' employee education funds and others. The level of the subsidy is set by how scarce the occupation or trade is and by the cost of the training, so it differs from trade to trade. Employers are also required to 「按规定足额提取和使用职工教育经费，确保 60% 以上用于一线职工教育和培训」, that is, to withdraw and use the employee education funds in full as required and to make sure more than 60% of it goes on the education and training of frontline employees. +- Evidence grade: A +- Sources: State Council Employment Promotion and Labour Protection Leading Group (国务院就业促进和劳动保护工作领导小组) (2025). Guiding Opinions on Carrying Out Large-Scale Vocational Skill Improvement Training. +- Notes: How to claim it and how much you can get are set by each province. The way in is the vocational training subsidy catalogue on the website of the local human resources and social security bureau, and the public training bases. The vocational training subsidies available during unemployment, where the same person may benefit at most 3 times, and the social insurance subsidies, are covered in section 7. People who are employed should ask their employer about the employee education funds first. Looking through the subsidy catalogue before choosing what to learn saves far more than choosing first and then hunting for a subsidy. + +### 10. When choosing a skill, look first at whether it needs hands and on-the-spot judgement, because those are hardest for automation to replace + +- Cost: No money. Before you choose, run the jobs you are considering through this dimension. +- In plain language: On the American estimate, about 47% of jobs are at a high risk of being taken over by computers, though a later estimate broken down into specific tasks gives a much lower figure. Machines have three hurdles they cannot get over. Manual dexterity. Coming up with something new. Dealing with people. So jobs that put you on site, in a non-standard environment, where you have to judge things as they come, are harder to replace, and jobs with a fixed process and mostly information handling carry the highest risk. +- Benefit: Each of 702 American occupations was rated one by one on how likely the job is to be handed over to a computer, using a computability probability estimate. The result was that about 47% of jobs fall in the high-risk band. What decides the rating in the model are three hurdles machines cannot yet get over, which the authors call engineering bottlenecks. First, perception and manipulation, which includes finger dexterity, manual dexterity and working in tight spaces and awkward positions. Second, creative intelligence, the ability to come up with something new. Third, social intelligence, the ability to deal with people. So work in a non-standard environment that needs manual skill and judgement on the spot is consistently less likely to be replaced, and that is not an accident of a few occupations. Jobs with a fixed process and mainly information handling carry the highest risk. +- Evidence grade: B +- Sources: Frey CB, Osborne MA (2017). The future of employment: how susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, 254-280. +- Notes:Disputed. The study is an unpublished working paper from 2013, and what it rates is whether a whole occupation can be handed to a machine. Later work broke the question down into individual tasks and got a much lower high-risk share. It also does not say in which year any of this happens. When you use it, do not fix on the figure of 47%; look at the three places where it says machines are stuck. On-site installation, maintenance and nursing are more resistant to replacement than pure paperwork, and that is exactly why. So "learning a skill" should first and foremost mean one that puts you on site, facing a non-standard environment. + +### 11. Before you start, check whether this skill has a graded assessment, and whether the grade can be turned into money + +- Cost: No money. One check is enough. +- In plain language: The state is promoting a new eight-level worker scheme, which requires employers to link post wages to skill levels and lets outstanding employees be recognised at a level without going by length of service. That is the direction of policy, and how far each local employer has got varies a lot. A skill with a national occupational standard and a graded pathway is one where pay rising with grade may still be possible later, so check whether it has one before you start. +- Benefit: The state is promoting the 「新八级工」 vocational skill grading system. The document expressly requires grades to be linked to pay, in these words: 「强化技能导向的薪酬分配制度……指导企业实行岗位薪酬与岗位价值、技能等级双挂钩的分配办法」, that is, strengthen a pay distribution system driven by skill, and guide employers to use a method that links post pay both to the value of the post and to the skill level. Localities are to 「加大高级工以上评聘力度和批次」, that is, to step up the effort and the number of rounds for assessing and appointing senior workers and above. Enterprises that meet the conditions may run skill level assessments for their own employees. For outstanding employees, 「可打破工作年限、等级、资历等条件，直接认定相应职业技能等级」, that is, the conditions of length of service, grade and seniority may be set aside and the corresponding vocational skill level recognised directly. +- Evidence grade: B +- Sources: State Council Employment Promotion and Labour Protection Leading Group (国务院就业促进和劳动保护工作领导小组) (2025). Guiding Opinions on Carrying Out Large-Scale Vocational Skill Improvement Training. ; Ministry of Human Resources and Social Security (人力资源社会保障部) (2021). Implementation Plan for the "Skilled China Action". +- Notes: This is the direction of policy, not a wage increase already being paid out, and how far it has been implemented varies a great deal between localities and employers. The part you can act on is this: when choosing a skill, check whether it has a corresponding national occupational standard and graded pathway, because only a skill with a pathway can later rise in pay by grade. A "skill" with neither a grade nor an industry entry requirement leaves you to find work on your own after you finish. + +### 12. Do not take "the pay is high online" as evidence of a shortage; look up the local list of occupations in short supply and the subsidy catalogue + +- Cost: No money. Look at two local documents. +- In plain language: Which trade gets the most training subsidy in a place is set by how scarce it is and by the cost of the training, so the size of the subsidy shows whether the place is short of people. The "so-and-so earns so much a month" figures online are mostly the asking prices on recruitment platforms or personal accounts. What you see is mostly the people who did earn, and it takes no account of hours, commuting, occupational disease or the busy and quiet seasons. +- Benefit: The size of the subsidy is itself the official price put on which trades are short of workers. The policy requires 「健全根据职业（工种）急需紧缺程度、培训成本等情况确定的差异化补贴机制」, that is, a sound differentiated subsidy mechanism determined by the urgency and scarcity of an occupation or trade, the cost of training and other circumstances, and localities use it to set the scope of their own subsidised vocational skill training projects. So how much the same trade is subsidised at in different cities reflects directly how scarce that trade is there. +- Evidence grade: C +- Sources: State Council Employment Promotion and Labour Protection Leading Group (国务院就业促进和劳动保护工作领导小组) (2025). Guiding Opinions on Carrying Out Large-Scale Vocational Skill Improvement Training. ; National Bureau of Statistics (国家统计局) (2026). Average Annual Wage of Employed Persons in Urban Units, 2025: 「规模以上企业就业人员年平均工资为106080元，其中，中层及以上管理人员210016元，专业技术人员155491元，办事人员和有关人员94936元，社会生产服务和生活服务人员79857元，生产制造及有关人员80739元」, that is, the average annual wage of employees of enterprises above the designated size is 106080 yuan, of which middle-level and above managers average 210016 yuan, professional and technical personnel 155491 yuan, clerical and related staff 94936 yuan, staff in social production and living services 79857 yuan, and staff in production, manufacturing and related work 80739 yuan. ; Vocational Training Technical Guidance Centre of the Ministry of Human Resources and Social Security (人力资源社会保障部就业培训技术指导中心) (2023). Ranking of the 100 Occupations in Shortest Supply Nationwide, where vacancies outnumber jobseekers, for the fourth quarter of 2022 (China Job Network). +- Notes: The official wage figures are broken down only into five broad groups of posts, with nothing for individual trades. The National Bureau of Statistics figures for 2025 are about 106,000 yuan (exact figure 106080 yuan) as the average annual wage at enterprises above the designated size. Middle-level and above managers are about 210,000 yuan (exact figure 210016 yuan), professional and technical staff about 155,000 yuan (exact figure 155491 yuan), clerical staff 94936 yuan, production and manufacturing staff 80739 yuan, and staff in social production and living services 79857 yuan. The Ministry of Human Resources and Social Security ranking of the 100 occupations in shortest supply can be looked up on the China Job Network, and the latest edition stops at the fourth quarter of 2022, compiled from recruitment and jobseeker data from public employment service agencies in 102 cities nationwide. The "so-and-so earns so much a month" figures circulating online almost all come from the asking prices on recruitment platforms or from personal accounts. What you can see is mostly the people who did earn, and none of it counts hours, commuting, occupational disease or the busy and quiet seasons. To see real demand, look up the subsidy catalogue on the website of the local human resources and social security bureau, then the ranking of post demand at the public employment service agencies. Finally, ask people already doing the job at the actual site, workshop or shop. + +### 13. For the same money and time, choose a short course that leads straight to a job + +- Cost: The tuition for a short course, which is not much money. It takes a stretch of your time. The hard part is stopping on the agreed terms when you are halfway through and nothing has happened yet. +- In plain language: What the policy encourages is short training that leads to employment quickly, and micro-majors at vocational colleges. Choose a programme on one criterion only: after it, is there a specific post you can send a CV to? A course lasting a year and a half, after which the certificate still says "gain more experience first", costs far more time than the price suggests. Write the stopping condition down before you start: I stop if I have not found work by the time I reach this level. +- Benefit: The policy line is the same: support for 「民办培训机构开展市场化的高质量职业技能培训项目和能短期实现就业的『短平快』式职业技能培训项目」, that is, private training institutions running market-based high-quality vocational skill training projects, and short, fast and effective vocational skill training projects that can lead to employment quickly. It also pushes vocational colleges to 「聚焦就业市场急需知识和技能……开设微专业和职业能力培训课程」, that is, to focus on the knowledge and skills the job market urgently needs, and to open micro-majors and vocational competence training courses. +- Evidence grade: C +- Sources: State Council Employment Promotion and Labour Protection Leading Group (国务院就业促进和劳动保护工作领导小组) (2025). Guiding Opinions on Carrying Out Large-Scale Vocational Skill Improvement Training. +- Notes: There is one criterion: after the course, is there a specific post you can send a CV to? A programme lasting a year and a half, after which you are told to "gain more experience first", costs far more than its price. The stopping condition described in section 4 applies here too: write it down before you start, "I stop if I have not found work by the time I reach this level". + +### 14. After studying, close the book and test yourself, and do not go back and reread + +- Cost: No money, and no increase in the total time. Take the time you would have spent rereading and use it to write from memory with the book shut, or to do questions. The hard part is that self-testing exposes what you did not remember on the spot, which feels worse than rereading. +- In plain language: Reading something and then reading it again feels familiar, but it is not well remembered. Testing yourself once with the book shut works noticeably better, and adding short quizzes in class raises grades by a moderate amount. Ranking ten common study methods, self-testing comes in the top band. After finishing a section, write down the key points from memory first, then turn back and check. +- Benefit: A meta-analysis of 222 independent studies and 48478 students: classroom quizzes raise academic achievement, with an effect size in the moderate range (g = 0.499). The effect size depends on several conditions: what study method the control group used, whether the quiz format matches the final exam, whether the material is matched, whether corrective feedback was given, how many times the quiz was repeated, where and when it was taken, how long the intervention lasted, and what experimental design was used. A separate review that assessed ten common learning methods one by one rates "practice testing" and "distributed practice" in the top band (high utility), on the grounds that they work for learners of different ages and abilities, and have been tested across several assessment formats and real educational settings. +- Evidence grade: A +- Sources: Yang C, Luo L, Vadillo MA, Yu R, Shanks DR (2021). Testing (quizzing) boosts classroom learning: A systematic and meta-analytic review. Psychological Bulletin, 147(4), 399–435. ; Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 14(1), 4–58. +- Notes: Self-testing is there to force you to pull things out of memory, not to score you. How to do it: after finishing a section, shut the book and write the key points from memory, then turn back and check. Corrective feedback works better, so when you check the answers look at where you went wrong. The benefit is graded "large" on the basis of the effect size and the grading. Both of these sources measure grades rather than hours, so they have not been tested against a time-saving threshold. Self-testing and item 15 in this section (spread the same time across several days) are the two highest-rated methods in the same review, and they work well together. + +### 15. Spread the same time across several days rather than finishing it in one go + +- Cost: No money, and the total time does not change either. Only the schedule changes: the same 3 hours, split into three days of 1 hour each. The hard part is starting a few days ahead instead of putting it off to the last night. +- In plain language: The same amount of time learned across several days is remembered better than the same time learned in one go. 317 experiments have all been measuring this, and they agree. There is also a practical rule: the further ahead you have to remember it, the longer the gap between reviews should be. If the test is tomorrow, review on the day. If it is a month away, go over it every few days. +- Benefit: This is a meta-analysis bringing together 184 articles, 317 experiments and 839 assessments of the effect of distributed practice. It compared two things: learning in one go against learning split across several sessions, which is the spacing effect, and short gaps against long gaps, which is the delay effect. The conclusion is that the gap between reviews and how long you have to remember it together determine retention on the final test: the longer you need to retain it, the longer the gap that gives the best retention. A separate review that assessed ten common learning methods one by one rates distributed practice in the top band (high utility). +- Evidence grade: A +- Sources: Cepeda NJ, Pashler H, Vul E, Wixted JT, Rohrer D (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354–380. ; Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques. Psychological Science in the Public Interest, 14(1), 4–58. +- Notes: Best used on top of item 14 in this section (test yourself with the book shut): spread the self-testing itself across several days. Most of the material in the Cepeda meta-analysis is word-list memory and the like, and no general number of days is given for the best gap for complicated content. "The further away the test, the longer the gap" is the rule the text gives, with no specific schedule. Cramming the night before does work at the time; the problem is how fast it fades. + +### 16. Do not make underlining, rereading and summarising your main study method + +- Cost: No money, and it saves time. The hard part is that these are the easiest to do and give the strongest feeling of "I am studying". +- In plain language: What students use most is underlining and rereading, and in the ranking of ten common study methods both sit in the bottom band. Writing summaries, the keyword mnemonic and imagining a picture to go with a passage are also in the bottom band. They are not entirely useless, but the conditions under which they work are too narrow, and the same time is better spent on self-testing. +- Benefit: A review that assessed ten common learning methods rates them by whether the effect carries across learning conditions, learner characteristics, material and assessment format. The two rated in the top band are practice testing and distributed practice. Those rated in the middle band are elaborative questioning, self-explanation and interleaved practice. The five in the bottom band are: summarising, underlining or highlighting, the keyword mnemonic, imagining a picture to go with a passage, and repeated reading. The review states that most students say they use repeated reading and underlining, but that neither reliably raises grades, and that they should be replaced by other methods, such as using practice testing instead of repeated reading. +- Evidence grade: A +- Sources: Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 14(1), 4–58. +- Notes: Highlighting is fine for marking the places you are going to test yourself on. The problem is treating the highlighting itself as the learning. The review also makes clear that the bottom band does not mean zero: the conditions under which these work are narrow, and they need more research. For what to do instead, see item 14 in this section (test yourself with the book shut) and item 15 in this section (spread the same time across several days). + +### 17. Mix question types when you practise, and do not do twenty of one kind in a row + +- Cost: No money. Change from working through the workbook in order by question type to doing it in a mixed order. The hard part is that accuracy is lower on the spot when you mix them, which feels like going backwards. +- In plain language: Doing a long run of the same kind of question goes smoothly at the time, and afterwards you may not be able to tell which method to use. Practising several kinds mixed together works better in the long run, by a moderate amount. This technique is picky about material: it works best for diagrams and for categorising, is a little weaker for mathematics, and for vocabulary words it is worse than blocked practice. +- Benefit: This is a multi-level meta-analysis of 59 studies, 238 effect sizes and 158 samples. The overall effect of interleaved practice over blocked practice is in the moderate range (Hedges' g = 0.42). By material: drawing is the best (g = 0.67), and other visual material is also good; the effect for mathematical tasks is smaller (g = 0.34); the overall effect is not significant for expository text and taste materials; and for vocabulary, blocked practice is better instead (g = −0.39). The regression analysis shows that the more similar the categories are to one another, the less similar things are within a category, and the more complicated the material, the greater the benefit of interleaving. +- Evidence grade: A +- Sources: Brunmair M, Richter T (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators. Psychological Bulletin, 145(11), 1029–1052. +- Notes: The review quoted in item 16 in this section (do not make underlining your main study method) rates interleaved practice in the middle band, on the grounds that there was not enough research at the time. Put the two together and the use is this: mix practice for content where you have to "recognise which kind this is and which method applies", such as question types in mathematics, case files and which rule of law applies; practise in blocks material you have to memorise, such as vocabulary words and definitions of terms. + +### 18. Do not pick a study method on the basis of "I am visual and he is auditory" + +- Cost: No money. You simply stop picking material and courses by this classification, and stop paying for a style assessment. +- In plain language: Many people believe that everyone has a learning style of their own, such as visual or auditory, and that you learn better when you are taught to your style. Very few studies have tested this carefully, almost none support it, and a few come out exactly the other way. People do have preferences, but there is no evidence that lessons matched to a preference are learned better, and no reason to pay for a style assessment. +- Benefit: A review commissioned to assess the evidence on learning styles points out that showing that "teaching to learning style works" requires grouping students by style first, randomly assigning each group to a different teaching method, and then testing everyone with the same test. The criterion for real evidence is the appearance of one particular interaction effect: a method that works best for one style must not be the best for another style. The review concludes that almost no such interaction has been found. There is a great deal of literature on learning styles, but very little of it uses an acceptable experimental design, and several of those studies come out directly against the matching hypothesis. On that basis the authors consider the existing evidence insufficient to support using learning style assessments in education. +- Evidence grade: A +- Sources: Pashler H, McDaniel M, Rohrer D, Bjork R (2008). Learning Styles: Concepts and Evidence. Psychological Science in the Public Interest, 9(3), 105–119. +- Notes: The review also makes clear that this does not mean every version of learning style theory has been tested and refuted; many have never been tested at all. What you want to avoid is paying for a style assessment and picking a training course by style. For the criteria to use when choosing training, see item 9 in this section (training goes through the government subsidy channel first) and item 13 in this section (choose a short course that leads straight to a job). + +### 19. Set questions for yourself in terms of how you will use the material, instead of reciting it once and counting that as done + +- Cost: No money, and no extra time. Only the practice changes: "reciting it once through" becomes "setting questions for yourself in terms of how you will use it". The hard part is having to work out first how these things will be used later. +- In plain language: Testing yourself with the book shut does more than check the few words you had memorised. 192 comparisons put together, involving more than ten thousand people, show that it works too for questions with a changed format, for applied problems and for reasoning problems, by a moderate to small amount. There are two situations where it is almost useless: reversing the question and the answer, and material you have only seen in the book and have never been tested on. +- Benefit: This meta-analysis brings together more than 40 years of research, covering 67 published and unpublished articles, 122 experiments and 192 transfer effect sizes, with 10382 people in total. Compared with a control that rereads without being tested, the learning produced by practice testing does transfer to other situations, with an effect size d=0.40 (95% CI 0.31 to 0.50). Transfer is strongest where the question format is changed, where applied problems and reasoning problems are tested, where the problem is a medical diagnosis, and where a clue word of similar meaning is tested. Transfer is weakest where the stimulus and the response are reversed, where the material was seen at first learning but never tested, and where the problem is a worked example. Among the moderators, whether the response format is the same in practice and in the test, whether retrieval practice with detailed feedback has been done, and whether the first quiz was answered correctly all strongly affect whether positive transfer appears. The authors also corrected for publication bias in two ways. The effect sizes of the moderators were barely affected, but the intercept estimate came down markedly: when none of the conditions above is present, positive transfer often cannot be seen. +- Evidence grade: A +- Sources: Pan SC, Rickard TC (2018). Transfer of test-enhanced learning: meta-analytic review and synthesis. Psychological Bulletin, 144(7), 710-756. +- Notes: The word "recite" normally runs two things together: going over the material once more, which is rereading and underlining, and actively recalling with the book shut. The first is rated lowest in the review that ranks ten study methods; to see the ratings themselves, see item 16 in this section (do not make underlining, rereading and summarising your main study method). What you want is the second: active recall with the book shut, described in item 14 in this section (test yourself with the book shut), and done across several days, see item 15 in this section (spread the same time across several days). How to do it is to work out first how you will use this later: whether you have to pick the right answer on a paper, or judge it on the spot, and then set questions for yourself in that form. It is also worth doing the reverse, because transfer in the reversed direction was the weakest in this meta-analysis, so both directions need practice. + +### 20. Work out which series and which grade you are in before you apply for a professional title, then find the application channel from the nature of your employer + +- Cost: No money. Look up your province's application notice for that year once. It takes about half an hour. +- In plain language: Professional titles are divided nationally into 27 series, and each series into junior, intermediate and senior. In general you apply level by level. Where you work for an organisation, the organisation checks and posts the application before sending it up. In a private enterprise you can go to a professional title application service point, or apply through a talent intermediary service organisation or an industry association. Self-employed people apply to the assessments organised by the local human resources and social security department. +- Benefit: There are 27 professional title series nationwide, such as economics, engineering, health and publishing. Titles are set in three levels: senior, intermediate and junior. Ministry of Human Resources and Social Security Order No. 40 applies to enterprises, public institutions, social organisations, individual economic organisations and self-employed people. An applicant generally applies level by level. The organisation you work for has to check the material and post it inside the organisation, and the posting period is not less than 5 working days. For private enterprises and self-employed people, a personnel agency or a similar body may carry out the checking, posting and recommendation on your behalf. Localities have to set up professional title application service points in technology parks, high-tech development zones and similar places, and may also accept applications through talent intermediary service organisations, chambers of commerce and industry associations, and learned societies. Self-employed people who meet the conditions take part, according to where they are registered, in the assessments organised by the local human resources and social security department. Where material can be verified online, you may not be required to produce a separate certificate. Where the material is incomplete, the assessing body has to tell you in one go what is missing. People in private enterprises generally apply where the employment relationship sits. Where you have been posted elsewhere and have worked there continuously for a full year, you may apply at the place you were posted to. Retired people may not apply. Where a public institution's staff member has been given a punishment of demotion or above, that person may not apply while the punishment is in force (nationwide) +- Evidence grade: A +- Sources: Ministry of Human Resources and Social Security Order No. 40 (人力资源和社会保障部令第 40 号) (2019). Interim Provisions on the Administration of Professional Title Assessments. : Articles two, thirteen, fourteen, fifteen, sixteen, seventeen and eighteen; General Office of the Ministry of Human Resources and Social Security (人力资源社会保障部办公厅) (2020). Notice on Further Improving Professional Titles for Private Enterprises. : 「在专业技术人才密集的创业孵化基地、高新技术开发区、科技园区等地设立职称申报受理服务点，或通过人才中介服务机构、工商联、行业协会商会、学会等社会组织受理民营企业专业技术人才职称申报」「民营企业专业技术人才被派驻外地连续工作一年以上的，可按有关规定在派驻地申报职称评审」; Xinhua News Agency (新华社) (2024). Professional title assessments reach their peak: how to check the information, and what has changed? : 「我国共设有经济、工程、卫生、出版等27个职称系列」「符合条件的自由职业者，可根据属地原则申报参加当地人社部门组织的职称评审」; General Office of the CPC Central Committee and General Office of the State Council (中共中央办公厅、国务院办公厅) (2016). Opinions on Deepening the Reform of the Professional Title System. : 「不将论文作为评价应用型人才的限制性条件」「对职称外语和计算机应用能力考试不作统一要求」「符合条件的高技能人才，可参加工程系列专业技术人才职称评审」「公务员不得参加专业技术人才职称评审」 +- Notes: There is no national list of what material to prepare. Each series and each province issues an application notice every year, and the list of material, the application dates and the achievement requirements are all written into it. So the first step is to find that year's notice on the website of the provincial human resources and social security department. A paper is no longer required in every case: the 2016 reform opinions say not to make a paper a hard threshold for applied professionals, and no unified requirement is set for the foreign language and computer exams in professional titles. Skilled workers have a route too: highly skilled people who meet the conditions may take part in professional title assessments in the engineering series. Civil servants may not take part in professional title assessments. + +### 21. Junior and intermediate titles such as accounting are won by a national unified exam, so check your education and years of work before you register + +- Cost: The registration fee is set by each province. Preparing for the exam takes several months of your spare time. +- In plain language: In some series the junior and intermediate titles are won by passing an exam, and passing is enough, with no separate assessment afterwards. Accounting is one: an accounting assistant needs senior high school or above, and an accountant needs a bachelor's degree with 4 years of accounting work, or a junior college diploma with 5 years. Holding certain vocational qualifications may also count directly as holding the corresponding professional title. +- Benefit: The 2016 reform opinions provide that where a profession uses a national unified exam for junior and intermediate titles, no corresponding assessment or recognition is carried out. In fields where professional titles and vocational qualifications are closely linked, holding a vocational qualification is recognised as holding the corresponding professional title, and may also count as a condition for applying for the level above. Take accounting as the example. Accounting assistants and accountants sit a national unified exam. The senior accountant exam is combined with an assessment, and the highest grade generally uses assessment alone. The entry conditions are as follows. Accounting assistant: senior high school graduation or above, including technical secondary schools, vocational high schools and technical schools. Accountant: a doctoral degree; or a master's degree with 1 year of accounting work; or a second bachelor's degree or a postgraduate course with 2 years of work; or a bachelor's degree with 4 years of work; or a junior college diploma with 5 years of work. Those with a professional master's or doctoral degree in accounting are exempt from the paper on Financial Management when sitting the accountant exam (nationwide) +- Evidence grade: A +- Sources: General Office of the CPC Central Committee and General Office of the State Council (中共中央办公厅、国务院办公厅) (2016). Opinions on Deepening the Reform of the Professional Title System. : 「专业技术人才取得职业资格即可认定其具备相应系列和层级的职称，并可作为申报高一级职称的条件。初级、中级职称实行全国统一考试的专业不再进行相应的职称评审或认定。」; Ministry of Human Resources and Social Security and Ministry of Finance (人力资源社会保障部、财政部) (2019). Guiding Opinions on Deepening the Reform of the Professional Title System for Accounting Personnel. : 「助理会计师、会计师实行全国统一的会计专业技术资格考试」「高级会计师采取考试与评审相结合方式，正高级会计师一般采取评审方式」; Ministry of Finance and others (财政部等) (2024). Notice on the Linkage Between Professional Degrees in Accounting and Professional Technical Qualifications in Accounting. +- Notes: Whether your series is examined, assessed, or examined and assessed together is set by the "guiding opinions on deepening the reform of the professional title system" for that series. The Ministry of Human Resources and Social Security and the competent department of each industry have issued these series by series, and they can be searched for in the policy document library on the Chinese Government website. The registration dates and the papers for the exam are those given in that year's examination announcement. Civil servants may sit the accounting exam, but may not take part in accounting professional title assessments. + +### 22. Do not pay an agency to assess your title for you, do not buy a ghostwritten paper, and do not falsify your material: if it is checked out, the title is withdrawn and a record is kept for 3 years + +- Cost: No money. The hard part is resisting the advertisements promising a pass for money. +- In plain language: A title obtained with false documents, a ghostwritten paper or plagiarised work is withdrawn once it is checked out. The record of dishonesty is kept for 3 years and goes onto the national credit platform. If the personal undertaking you sign when you apply turns out to be untrue, you may not apply again for 3 years. The human resources departments also run specific campaigns against agencies over false advertising, impersonated assessments and fake certificates. +- Benefit: Ministry of Human Resources and Social Security Order No. 40 provides that a title obtained by supplying false documents, plagiarising another person's work or academic results, or by other improper means is withdrawn. At the same time it is recorded in the integrity file for professional title assessments, brought into the national credit information sharing platform, with a retention period of 3 years. The 2024 supervision measures name several kinds of violation by applicants: false undertakings and fabricated information, faked or ghostwritten papers, faked achievement records, and lobbying. An applicant has to sign a personal undertaking at the time of application, and if it is untrue, that person may not apply for 3 years. The human resources departments work with the public security, cyberspace and market regulation authorities to investigate agencies running false websites, false advertising, contract traps, impersonated assessments and fake certificates. Serious cases that may involve a crime are transferred to the competent authorities (nationwide) +- Evidence grade: A +- Sources: Ministry of Human Resources and Social Security Order No. 40 (人力资源和社会保障部令第 40 号) (2019). Interim Provisions on the Administration of Professional Title Assessments. : Article thirty-nine; Ministry of Human Resources and Social Security (人力资源社会保障部) (2024). Interim Measures for the Supervision of Professional Title Assessments. : Articles five, fifteen, seventeen and twenty-seven; General Office of the Ministry of Human Resources and Social Security (人力资源社会保障部办公厅) (2022). Notice on Further Improving Professional Title Assessments. : 「未经备案的职称评审委员会，评审结果不纳入全国职称评审信息查询验证系统」; Xinhua News Agency (新华社) (2024). Professional title assessments reach their peak: how to check the information, and what has changed? : 「大家可通过www.12333.gov.cn网页、手机12333客户端、人力资源社会保障部门户网站、“人力资源和社会保障部”微信公众号等多种渠道进行登录」 +- Notes: There is a way to tell real from false: a title from a proper assessment can be found in the national professional title assessment information query platform. The way in is the 12333 website, the 12333 mobile app and the website of the Ministry of Human Resources and Social Security. A title awarded by an assessment committee that was not registered cannot be found in that system. So anything promising 「包过」, a guaranteed pass, 「不用本人参加」, no need for you to attend, or 「证书查不到也能用」, usable even if the certificate cannot be found, can be judged as suspect. The person who mainly benefits is yourself, and the benefit is graded on the basis of avoiding a withdrawn title and a 3-year record of dishonesty. + +### 23. Getting a professional title does not mean a pay rise: first find out whether your employer appoints by post quota, or whether being awarded does not mean being appointed + +- Cost: No money. Ask the human resources department of your employer once. +- In plain language: A professional title is the basis for appointment, assessment and promotion, but being awarded and being appointed are two different things. Public institutions generally assess within the post quota, and being awarded usually leads to appointment. Organisations that do not run post management may separate assessment from appointment, and being awarded does not mean the organisation gives you the matching post and pay. After you change employer or move to another place, the title may have to be assessed again or formally confirmed. +- Benefit: Article two of Ministry of Human Resources and Social Security Order No. 40 states that the result of a professional title assessment is an important basis for the appointment, assessment and promotion of professional and technical staff. The 2016 reform opinions provide that public institutions which fully implement post management, and where the academic level of professional and technical staff is closely tied to the duties of the post, generally carry out professional title assessments within the post structure quota. Organisations that do not run post management, and series that are widely applicable and spread across all kinds of social organisation, together with new occupations, may separate assessment from appointment. When professional and technical staff move across regions or across organisations, the title is assessed again or formally confirmed according to the level of management authority, except where the state has provided otherwise (nationwide) +- Evidence grade: B +- Sources: Ministry of Human Resources and Social Security Order No. 40 (人力资源和社会保障部令第 40 号) (2019). Interim Provisions on the Administration of Professional Title Assessments. : Articles two and twenty-nine; General Office of the CPC Central Committee and General Office of the State Council (中共中央办公厅、国务院办公厅) (2016). Opinions on Deepening the Reform of the Professional Title System. : 「对于全面实行岗位管理、专业技术人才学术技术水平与岗位职责密切相关的事业单位，一般应在岗位结构比例内开展职称评审。对于不实行岗位管理的单位，以及通用性强、广泛分布在各社会组织的职称系列和新兴职业，可采用评聘分开方式。」 +- Notes: The grade is B because the documents only say how assessment and appointment fit together, and give no national figure for "how much more you earn once you have the title". What you can act on is this: in a public institution or a state-owned enterprise, ask the human resources department what post vacancies your organisation has and what its appointment rules are. In a private company, ask your boss whether this certificate counts and whether it comes with money. From 96719af5636f7eb684c8d1a66d4480798bafa9f3 Mon Sep 17 00:00:00 2001 From: Stream Entry <978862+streamentry@users.noreply.github.com> Date: Fri, 2 Oct 2026 04:30:49 +0700 Subject: [PATCH 11/25] Import and credit HowToLiveLonger source material * docs: import HowToLiveLonger source snapshot * docs: preserve upstream Unlicense * docs: add HowToLiveLonger attribution notice * docs: credit HowToLiveLonger source material --- README.md | 6 +- third_party/geekan-HowToLiveLonger/LICENSE | 24 ++ third_party/geekan-HowToLiveLonger/NOTICE.md | 17 + .../geekan-HowToLiveLonger/README_en.md | 312 ++++++++++++++++++ 4 files changed, 358 insertions(+), 1 deletion(-) create mode 100644 third_party/geekan-HowToLiveLonger/LICENSE create mode 100644 third_party/geekan-HowToLiveLonger/NOTICE.md create mode 100644 third_party/geekan-HowToLiveLonger/README_en.md diff --git a/README.md b/README.md index bbcc768c..480f8550 100644 --- a/README.md +++ b/README.md @@ -291,9 +291,13 @@ node tools/pdf/build.mjs # PDF，另需 pandoc ≥ 3.1 和 typ 正文按节拆成 34 个文件放在 [book/](book/)，点上面目录里的节名进入。拆开是因为单文件已经超过 GitHub 渲染 Markdown 的 512 KB 上限，后面的节显示不出来；[在线检索页](https://streamentry.github.io/HowToLiveBetter/)会把这些文件合起来读，用法不变。 +## 致谢与第三方资料 + +- [geekan/HowToLiveLonger](https://github.com/geekan/HowToLiveLonger)：本仓库保留了一份其英文 README 的[第三方资料快照](third_party/geekan-HowToLiveLonger/README_en.md)，并同时保留[原始 The Unlicense](third_party/geekan-HowToLiveLonger/LICENSE) 与[来源说明](third_party/geekan-HowToLiveLonger/NOTICE.md)。感谢 geekan 与贡献者整理这些长寿研究线索。该快照作为资料来源保存，其中的健康结论不会自动视为本书已经核实或采纳；进入正文的内容仍按本仓库自己的原始文献核实、证据分级和争议标注规则处理。 + ## 许可 -正文用 [CC BY 4.0](LICENSE) 发布，范围是 book/、docs/ 和本 README 的文字。你可以转载、改编、商用，不用来问作者，但要做到三件事： +正文用 [CC BY 4.0](LICENSE) 发布，范围是 book/、docs/ 和本 README 的文字；`third_party/` 下的第三方材料除外，按各自目录中保留的原始许可证与来源说明发布。你可以转载、改编、商用，不用来问作者，但要做到三件事： - 写明出处：「高性价比人生指南」，附上仓库链接 https://github.com/eternity4719/HowToLiveBetter 。 - 附上许可证链接 https://creativecommons.org/licenses/by/4.0/ 。 diff --git a/third_party/geekan-HowToLiveLonger/LICENSE b/third_party/geekan-HowToLiveLonger/LICENSE new file mode 100644 index 00000000..fdddb29a --- /dev/null +++ b/third_party/geekan-HowToLiveLonger/LICENSE @@ -0,0 +1,24 @@ +This is free and unencumbered software released into the public domain. + +Anyone is free to copy, modify, publish, use, compile, sell, or +distribute this software, either in source code form or as a compiled +binary, for any purpose, commercial or non-commercial, and by any +means. + +In jurisdictions that recognize copyright laws, the author or authors +of this software dedicate any and all copyright interest in the +software to the public domain. We make this dedication for the benefit +of the public at large and to the detriment of our heirs and +successors. We intend this dedication to be an overt act of +relinquishment in perpetuity of all present and future rights to this +software under copyright law. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, +EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR +OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +OTHER DEALINGS IN THE SOFTWARE. + +For more information, please refer to diff --git a/third_party/geekan-HowToLiveLonger/NOTICE.md b/third_party/geekan-HowToLiveLonger/NOTICE.md new file mode 100644 index 00000000..5c33b4d4 --- /dev/null +++ b/third_party/geekan-HowToLiveLonger/NOTICE.md @@ -0,0 +1,17 @@ +# Third-party notice: geekan/HowToLiveLonger + +This directory preserves a snapshot of the English README from the upstream project: + +- Project: **HowToLiveLonger** +- Upstream repository: https://github.com/geekan/HowToLiveLonger +- Upstream file: https://github.com/geekan/HowToLiveLonger/blob/main/README_en.md +- Upstream author/maintainer: **geekan** and contributors +- Imported on: **2026-10-02** +- Imported README blob SHA: `753cc5b6a5eaba1c970af8de50a201717c5ab1df` +- Upstream license: **The Unlicense** (see `LICENSE` in this directory) + +The upstream README is preserved here as third-party source material. Credit is retained voluntarily even though The Unlicense does not require attribution. + +Its health claims, effect sizes, interpretations, and recommendations are **not automatically adopted or re-verified by HowToLiveBetter**. Material promoted into the main book should continue to follow this repository's own evidence-review, sourcing, and grading rules. + +No endorsement by geekan or the upstream contributors is implied. diff --git a/third_party/geekan-HowToLiveLonger/README_en.md b/third_party/geekan-HowToLiveLonger/README_en.md new file mode 100644 index 00000000..753cc5b6 --- /dev/null +++ b/third_party/geekan-HowToLiveLonger/README_en.md @@ -0,0 +1,312 @@ +# A programmer's guide to live longer + +[![CN doc](https://img.shields.io/badge/文档-中文版-blue.svg)](README.md) +[![EN doc](https://img.shields.io/badge/document-English-blue.svg)](README_en.md) +[![MetaGPT](https://img.shields.io/badge/LessWorkTime-MetaGPT-blue)](https://github.com/geekan/MetaGPT) + +- [1. Glossary](#1-Glossary) +- [2. Objective](#2-Objective) +- [3. Key Result](#3-Key-Result) +- [4. Analysis](#4-Analysis) +- [5. Action](#5-Action) +- [6. Reference](#6-Reference) + - [6.1. Input](#61-Input) + - [6.1.1. Solid](#611-Solid) + - [6.1.2. Liquid](#612-Liquid) + - [6.1.3. Gas](#613-Gas) + - [6.1.4. Illumination](#614-Illumination) + - [6.1.5. Medicine](#615-Medicine) + - [6.2. Output](#62-Output) + - [6.2.1. Racquet Sports](#621-Racquet-Sports) + - [6.2.2. High-intensity Exercise](#622-High-intensity-Exercise) + - [6.2.3. Walk](#623-Walk) + - [6.2.4. Tooth brushing](#624-Tooth-brushing) + - [6.2.5. Bath](#625-Bath) + - [6.2.6. Housework(for aging males)](#626-Houseworkfor-aging-males) + - [6.2.7. Sleep](#627-Sleep) + - [6.2.8. Sitting for a long time](#628-Sitting-for-a-long-time) + - [6.3. Context](#63-Context) + - [6.3.1. Emotion](#631-Emotion) + - [6.3.2. Wealth](#632-Wealth) + - [6.3.3. Weight](#633-Weight) + - [6.3.4. COVID-19](#634-COVID-19) + +--- + +### 1. Glossary + +* ACM: All-Cause Mortality + +### 2. Objective + +* Live longer +* Less work time: check [MetaGPT](https://github.com/geekan/MetaGPT/blob/main/README_EN.md) + +### 3. Key Result + +* Reduces ACM by 66.67% +* Increase life expectancy by \~20 years +* ~~Keep dopamine at a balanced level~~ + +### 4. Analysis + +* Main reference: There are relatively many academic papers related to ACM, which can be used as the main reference. +* The relationship between lifespan and ACM is nonlinear: Obviously, the relationship between lifespan and ACM is a nonlinear function. Assumption:`DeltaLifeSpan=(1/(1+DeltaACM)-1)*10` (DeltaACM is ACM's change value. Welcome to optimize this formula) +* Variables cannot be simply superimposed: Variables do not meet the IID assumption, and the actual effect between the variables is also unclear. +* Conflicting views: All references have literature/study counterparts, but there are significant conflicting views between some of the literature(e.g. Carbohtdrate intake retio), and some of the literature are controversial(e.g. Sleep before 22:00 increases ACM by 43%) +* Research only shows correlation: Research shows more correlation than causation. Confirm whether the research adequately proves causation. + * e.g. A research shows that people with an average of 7000 or more steps per day have a significantly lower ACM rate. However, people with fewer steps may include more long-term patients. If this part of data is not reasonably excluded, the result is distorted. + +### 5. Action + +* Input + * Solid: Plain boiled pork(-11%\~-3% ACM), Fruits and vegetables(-26%\~-17% ACM), Spicy food(-23% ACM), Nuts(-27%\~-4% ACM), *less egg yolk(otherwise +7% ACM per 0.5 pieces per day)(controversial)*, moderate carbs, Vegetable protein(-10% ACM), less ultra-processed foods(-62%\~-18%) + * Liquid: Coffee(-22%\~-12% ACM), Milk(-17%\~-10% ACM), Tea(-15%\~-8% ACM), Limit or avoid sweetened beverages(Otherwise +7% ACM per cup per day, +dopamine), Give up alcohol or less than 100g per week(pure alcohol(g)=alcohol consumption(ml)×Alcohol concentration(%)×(Alcohol density)0.8g/ml)(Otherwise +\~50% ACM, unlimited) + * Gas: No smoking(Otherwise +~50% ACM, -12\~-11 years lifespan) + * Illumination: Sunbathe(-~40% ACM) + * Medicine: Metformin(for diabetic, +3 years lifespan), Multi-vitamins(-8% cancer risk), Spermidine(-60%\~-30% ACM), Glucosamine(-39% ACM) +* Output + * Sports: 45-minute racquet sports 3 times a week(-47% ACM) + * Everyday life: Brush your teeth(-25% ACM) + * Sleep: Sleep 7 hours a day. (fall asleep between 22-24 o'clock) *earlier +43% ACM, later +15% ACM(controversial)* +* Other + * Weight: Lose weight(-54% ACM) + +### 6. Reference + +#### 6.1. Input + +##### 6.1.1. Solid + +* Plain boiled pork + * [Meat Consumption Linked to Small Rises in CVD, All-Cause Mortality Risks](https://www.tctmd.com/news/meat-consumption-linked-small-rises-cvd-all-cause-mortality-risks) + * Reference: [Associations of Processed Meat, Unprocessed Red Meat, Poultry, or Fish Intake With Incident Cardiovascular Disease and All-Cause Mortality](https://jamanetwork.com/journals/jamainternalmedicine/articlepdf/2759737/jamainternal_zhong_2020_oi_190112.pdf) + * Increased red meat intake is associated with risk of death. Respondents who increased red meat intake by at least half a serving per day on average over eight years had a 10% increase in ACM over the next eight years. (HR, 1.10[95%CI, 1.04-1.17]) + * Eating two servings of read or processed meat(excluding poultry or fish) per week increases ACM by 3%. + > P.S. Half serving of red meat is equivalent to 14g processed red meat or 40g unprocessed red meat + * ![red_meat](https://user-images.githubusercontent.com/2707039/163703960-6f321de5-4daa-4ea5-95b9-af9c96f1c1bc.jpg) + * [红肉和白肉最大的区别是什么？为啥要这么分呢？](https://www.zhihu.com/question/67223570/answer/809785380)*(Mandarin, Use Chrome's Translator)* +* Fruits and vegetables + * [每年54万人死亡, 竟是因为水果吃得少！？这已成十大死亡因素之一！](https://www.sohu.com/a/322360740_164924)*(Mandarin, Use Chrome's Translator)* + * Reference: [Estimated Global, Regional, and National Cardiovascular Disease Burdens Related to Fruit and Vegetable Consumption: An Analysis from the Global Dietary Database (FS01-01-19) ](https://academic.oup.com/cdn/article-abstract/3/Supplement_1/nzz028.FS01-01-19/5516583) + * A daily intake of 200g fresh fruit can reduce ACM by 17%, the risk of diabetic macrovascular complications (e.g. stroke, ischemic heart disease, etc.) by 13%, and the risk of diabetic small blood vessel complications (e.g. diabetic nephropathy, diabetic eye disease, diabetic foot disease, etc.) by 28%. + * [《自然》子刊: 每天二两西兰花, 健康长寿都有啦！分析近6万人23年的数据发现, 吃含黄酮类食物与死亡风险降低20%相关丨临床大发现](https://mp.weixin.qq.com/s/E6BAi-Vnhr1jXBm0Pys2ZQ)*(Mandarin, Use Chrome's Translator)* + * Reference: [Flavonoid intake is associated with lower mortality in the Danish Diet Cancer and Health Cohort](https://www.nature.com/articles/s41467-019-11622-x) + * Eating flavonoid-rich foods linked to 20% lower risk of death + * ![Flavonoid](https://user-images.githubusercontent.com/2707039/163703969-42e64f88-e727-4e7d-85f2-07a92e29b613.jpg) + * "Interestingly, these threshold levels exist well within daily dietary achievable limits: one cup of tea, one apple, one orange, 100g of blueberries, and 100g of broccoli would provide most of the flavonoid subclasses and over 500mg of total flavonoids."says Dr. Bondonno. +* Chili + * [辣椒成死亡克星？据调研, 常吃辣患病死亡风险可降低61%](https://3g.163.com/dy/article/F6Q7I1ME053228ZU.html)*(Mandarin, Use Chrome's Translator)* + * Reference 1: [Chili pepper consumption and mortality in Italian adults](https://www.sciencedirect.com/science/article/pii/S0735109719382063) + * Reference 2: [The Association of Hot Red Chili Pepper Consumption and Mortality: A Large Population-Based Cohort Study](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0169876) + * Another study from Plos One in 2017 looked at 16,179 people aged 18+ and followed them up to 19 years. It was found that among the 4946 patients who died, the ACM rate was 21.6% for those who ate chili peppers, compared with 33.6% for those who did not eat chili peppers. For those who ate spicy food more than four times a week, the overall risk of death was reduced by 23%, and the risk of cardiovascular death was reduced by 34%. +* Eggs + * [每天多吃半个蛋, 增加7%的全因和心血管死亡风险？](https://m.thepaper.cn/baijiahao_11540780)*(Mandarin, Use Chrome's Translator)* + * Reference: [NIH-AARP Diet and Health Study](https://dietandhealth.cancer.gov/), [Egg and cholesterol consumption and mortality from cardiovascular and different causes in the United States: A population-based cohort study](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7872242/) + * Each additional half a whole egg/day was associated with a 7% (HR 1.07; 95% CI 1.06–1.08; P < 0.001) higher risk of all-cause mortality.In hypothetical substitution analyses, we found 6%, 8%, 9%, 7%, 13%, and 10% lower all-cause mortality when replacing half a whole egg (25 g/day) with equivalent amounts of egg whites/substitutes, poultry, fish, dairy products, nuts, and legumes, respectively. + * ![Egg](https://raw.githubusercontent.com/qhy040404/Image-Resources-Repo/master/pmed.1003508.g002.jpg) +* Nuts + * [哈佛20年研究: 吃核桃的人更长寿, 显著减少全因死亡, 延长寿命](https://www.163.com/dy/article/GKVOMMMF05148PF4.html)*(Mandarin, Use Chrome's Translator)* + * Reference: [Association of Walnut Consumption with Total and Cause-Specific Mortality and Life Expectancy in US Adults](https://www.mdpi.com/2072-6643/13/8/2699/pdf) + * The analysis found that regular consumption of walnuts can prolong life and reduce the risk of death from cardiovascular disease. Compared with no walnuts, eating more than 5 servings(1 serving defined as 28g) of walnuts per week increased healthy life expectancy by 1.3 years and reduced the risk of SCM by 14%, and the risk of cardiovascular mortality by 25%. + * [研究: 每日食生坚果, 死亡率降20%](https://zhuanlan.zhihu.com/p/44454030)*(Mandarin, Use Chrome's Translator)* + * Reference 1: [Association of nut consumption with total and cause-specific mortality](https://www.nejm.org/doi/full/10.1056/NEJMoa1307352) + * Reference 2: [APG_Health-&-Nutrition-Research-Brochure_DEC-19-18](https://americanpistachios.cn/sites/china/files/inline-files/APG_Health-%26-Nutrition-Research-Brochure_DEC-19-18.pdf) + * The researchers found that those who ate less than 1 ounce servings of tree nuts per week had a 7 percent lower mortality rate. Those who ate 1 ounce per week had an 11% reduction in mortality; those who ate 2 per week had a 13% reduction; those who ate 5 to 6 per week had a 15% reduction; and those who ate more than 7 per week had a 20% reduction in mortality. + * Two other premed research papers published in the Public Library of Science Online Journal and BioMed Central presented cross-sectional data at the start of the trial. Both studies evaluated 7,216 subjects and the relationship between the frequency and amount of nuts they consumed.Those who ate more than three servings of nuts(including pistachios) a week had a 39 percent lower mortality rate. +* Sodium(Seriously controversial) + * [Eur Heart J: 钠摄入量与预期寿命, 全因死亡率的关系](https://nursing.medsci.cn/article/show_article.do;jsessionid=A34E8A33918A152CB55BDD2E5FB1798D?id=afe720486ee7)*(Mandarin, Use Chrome's Translator)* + * Reference: [Messerli F H, Hofstetter L, Syrogiannouli L, et al. Sodium intake, life expectancy, and all-cause mortality[J]. European heart journal, 2021, 42(21): 2103-2112.](https://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC8169157&blobtype=pdf) + * ![ehaa947f6](https://user-images.githubusercontent.com/2707039/164894778-9710f18d-e055-4f62-bdcb-618687771d77.jpeg) + * Among the 181 countries included in this analysis, we found a positive correlation between sodium intake and healthy life expectancy at birth (β = 2.6 years/g of daily sodium intake, R2 = 0.66, P < 0.001), as well as healthy life expectancy at age 60(β = 0.3 years/g of daily sodium intake, R2 = 0.60, P = 0.048) but not for death due to non-communicable diseases(β = 17 events/g of daily sodium intake, R2 = 0.43, P = 0.100). Conversely, all-cause mortality correlated inversely with sodium intake (β = -131 events/g of daily sodium intake, R2 = 0.60, P < 0.001). In a sensitivity analysis restricted to 46 countries in the highest income class, sodium intake continued to correlate positively with healthy life expectancy at birth (β = 3.4 years/g of daily sodium intake, R2 = 0.53, P < 0.001) and inversely with all-cause mortality (β = -168 events/g of daily sodium intake, R2 = 0.50, P < 0.001). + * This (extensive) study links higher sodium intake with significantly lower ACM + * [Extended interpretation and discussion of the paper: A Fresh Foray in the Salt Wars: Life Expectancy Higher With Greater Sodium Intake](https://www.tctmd.com/news/fresh-foray-salt-wars-life-expectancy-higher-greater-sodium-intake) + * [NEJM/Lancet: 不要吃太多盐, 中国饮食所致心血管病和癌症死亡全球第一, 吃低钠盐可降低全因死亡率](https://ibook.antpedia.com/x/669028.html)*(Mandarin, Use Chrome's Translator)* + * However, many studies show that use low sodium salts can reduce the probability of a series of diseases and have a positive impact on the reduction of ACM. +* Carbohydrates(Seriously controversial) + * [低碳生酮饮食(四)碳水化合物与长期死亡率](https://zhuanlan.zhihu.com/p/137815934)*(Mandarin, Use Chrome's Translator)* + * Reference: The Lancet Public Health - [Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis](https://www.sciencedirect.com/science/article/pii/S246826671830135X) + * The less carbs, the shorter the lifespan. However, more carbs also slightly shorten lifespan. 50% carbs are associated with the longest lifespan. + * ![Carbs](https://user-images.githubusercontent.com/2707039/163703985-a2e2f8ac-101a-4f3c-903b-6850507f144b.jpg) + * [最强营养搭配！BMJ: 这么吃, 心血管疾病和死亡风险更低](https://www.chinacdc.cn/gwxx/202003/t20200323_214639.html)*(Mandarin, Use Chrome's Translator)* +* Betel nut + * [如何看待槟榔嚼出来的癌症？槟榔致癌风险究竟有多大？ - 丁香医生的回答 - 知乎](https://www.zhihu.com/question/312784161/answer/603370131)*(Mandarin, Use Chrome's Translator)* + * Reference: Chewing Betel Quid and the Risk of Metabolic Disease, Cardiovascular Disease, and All-Cause Mortality: A Meta-Analysis(https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0070679) + * Chewing betel nut increases ACM by 21% +* Caloric Restriction + * [怎么看待BBC《进食, 断食与长寿》？](https://www.zhihu.com/question/31395511)*(Mandarin, Use Chrome's Translator)* + * Animal experiments show that CR (caloric restriction, i.e. eating less) delayed multiple disease morbidity and mortality in rhesus monkeys. Compared with CR animals, normally fed monkeys had a 2.9-fold increased risk of various diseases and a 3.0-fold increased risk of death. + * ![CR-Macaca mulatta](https://user-images.githubusercontent.com/2707039/163703988-8767185b-326a-4783-b2e2-f190322bb7d6.jpg) +* Complex + * [最强营养搭配！BMJ: 这么吃, 心血管疾病和死亡风险更低](https://www.chinacdc.cn/gwxx/202003/t20200323_214639.html)*(Mandarin, Use Chrome's Translator)* + * [Associations of fat and carbohydrate intake with cardiovascular disease and mortality: prospective cohort study of UK Biobank participants](https://doi.org/10.1136/bmj.m688) + * By analyzing these participants' data, the researchers found that carbohydrate (sugar, starch, and fiber) and protein intake were non-linearly associated with all-cause mortality, while fat was linearly associated with all-cause mortality. Among them, higher sugar intake was associated with a higher risk of all-cause mortality and cardiovascular disease, while higher intake of saturated fatty acids was associated with a higher risk of all-cause mortality. + * Fig 1: The relationship between various nutrients and ACM + * ![The relationship between various nutrients and ACM](https://user-images.githubusercontent.com/2707039/163702022-8c2bfea9-ed5d-4fe0-8ead-e8740014b92b.jpg) + * Fig 2: The relationship between various nutrients and cardiovascular disease + * ![The relationship between various nutrients and cardiovascular disease](https://user-images.githubusercontent.com/2707039/163702084-97fb4a03-707c-475d-b88e-6fe2f8e87f92.jpg) + * **Further research showed that among all dietary patterns, the dietary pattern with the lowest risk of all-cause mortality was: 10-30g high fiber, 14-30% protein, 10-25% monounsaturated fatty acids, 5%-7% polyunsaturated fatty acids and 20%-30% starch intake. ** + * **Optimum energy source ratio: <24% starch, 15%-17% protein, >15% monounsaturated fatty acid, <15% sugar, 6% saturated fatty acid, 6% polyunsaturated fatty acid, >30ghigh fiber** + * [BMJ | 常吃薯片汉堡巧克力等食品, 平均死亡年龄仅仅为58岁, 死亡风险剧增](https://med.ckcest.cn/details.html?id=5183272274855936&classesEn=news)*(Mandarin, Use Chrome's Translator)* + * [Rico-Campà A, Martínez-González M A, Alvarez-Alvarez I, et al. Association between consumption of ultra-processed foods and all cause mortality: SUN prospective cohort study[J]. bmj, 2019, 365.](https://www.bmj.com/content/365/bmj.l1949.full) + * [Srour B, Fezeu L K, Kesse-Guyot E, et al. Ultra-processed food intake and risk of cardiovascular disease: prospective cohort study (NutriNet-Santé)[J]. bmj, 2019, 365.](https://www.bmj.com/content/365/bmj.l1451) + * [Lawrence M A, Baker P I. Ultra-processed food and adverse health outcomes[J]. bmj, 2019, 365.](https://www.researchgate.net/profile/Phillip-Baker-5/publication/333483796_Ultra-processed_food_and_adverse_health_outcomes/links/5f0c646ca6fdcc2f32336a95/Ultra-processed-food-and-adverse-health-outcomes.pdf) + +##### 6.1.2. Liquid + +* Milk + * [《柳叶刀》调研21个国家13万人: 每天1斤牛奶或酸奶, 心血管死亡风险下降23%](https://www.sohu.com/a/253940257_419768)*(Mandarin, Use Chrome's Translator)* + * Reference: [Association of dairy intake with cardiovascular disease and mortality in 21 countries from five continents (PURE): a prospective cohort study](http://mdrf-eprints.in/1114/1/Association_of_dietary_patterns_and_dietary_diversity_with_cardiometabolic_disease_risk_factors.pdf) + * People who ate two servings of dairy products a day had **a 17% lower risk of all-cause death**, a 23% lower risk of cardiovascular death, and a 33% lower risk of stroke compared to those who didn't eat dairy. (one serving was 244 grams of milk/yogurt, 15 grams of cheese, or 5 grams of butter) +* Tea + * [10万中国人随访7年发现, 每周喝三次茶与全因死亡风险降低15%, 预期寿命增加1.26年相关 ](https://www.jianshu.com/p/5461a205cf95?utm_campaign=hugo)*(Mandarin, Use Chrome's Translator)* + * Reference: [Tea consumption and the risk of atherosclerotic cardiovascular disease and all-cause mortality: The China-PAR project](https://www.researchgate.net/profile/Fangchao-Liu-4/publication/338483323_Tea_consumption_and_the_risk_of_atherosclerotic_cardiovascular_disease_and_all-cause_mortality_The_China-PAR_project/links/5e55e5e94585152ce8efe511/Tea-consumption-and-the-risk-of-atherosclerotic-cardiovascular-disease-and-all-cause-mortality-The-China-PAR-project.pdf) + * [中国成年人饮茶与死亡风险的前瞻性关联研究](http://rs.yiigle.com/CN112338202202/1351516.htm)*(Mandarin, Use Chrome's Translator)* + * A total of 34,661 deaths occurred in the 438,443 subjects included in the analysis during 11.1 years of follow-up.Compared with never tea drinkers, current non-daily tea drinkers and daily tea drinkers had HR values (95%CI) of all-cause mortality of 0.89 (0.86-0.91) and 0.92 (0.88-0.95), respectively. By gender The analysis showed that the protective effect of tea drinking on the risk of all-cause mortality was mainly seen in males (P<0.05 for interaction). +* Sugar-free (sweet) beverages + * [「无糖饮料使死亡风险增加 26 %」, 是真的吗？](https://www.zhihu.com/question/418598272/answer/1450648364)*(Mandarin, Use Chrome's Translator)* + * Participants who consumed ≥1 cup/day of mixed soft drinks had an 18% increased risk of death compared with those who consumed <1 cup/month. **Consumption of sugar-sweetened soft drinks or sugar-free soft drinks was associated with an 11% and 27% increased risk of death, respectively.** + * ![饮料](https://user-images.githubusercontent.com/2707039/163704346-e7d92e7f-eba5-4673-8f15-3a96782c2e32.png) + * [Association Between Soft Drink Consumption and Mortality in 10 European Countries](https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2749350) +* Sugary drinks + * [可乐和奶茶, 增加全因死亡率高达62%！果汁降低免疫力, 影响肝代谢！含糖饮料那些事](https://zhuanlan.zhihu.com/p/400746073)*(Mandarin, Use Chrome's Translator)* + * 1 cup of sugar-sweetened beverages per day increases all-cause mortality by 7%, and 2 cups increases all-cause mortality by 21% + * During 34 years of follow-up, the researchers found that those who drank 2 cups a day had a 21% higher risk of death overall, a 31% higher risk of death from cardiovascular disease, and a 16% higher risk of death from cancer than those who drank 1 or less sugar-sweetened beverages a month. + * Drinking one extra sugar-sweetened beverage per day was associated with a 7% increased risk of overall death, a 10% increased risk of cardiovascular disease, and a 16% increased risk of cancer-related death. + * A paper published in the "BMJ" proves that sugar-sweetened beverages can increase the risk of cancer. The overall cancer risk associated with sugar-sweetened beverages was 18% higher than usual, and 100% freshly squeezed fruit juice also increased the overall cancer risk by 12%. +* Juice + * [JAMA子刊: 100%纯果汁可能比含糖饮料更危险](https://zhuanlan.zhihu.com/p/66513350)*(Mandarin, Use Chrome's Translator)* + * Consuming an extra 12-ounce serving of sugar-sweetened beverages per day was associated with an 11% increased risk of all-cause mortality. + * Consuming an extra 12-ounce serving of fruit juice per day was associated with a 24% increased risk of all-cause mortality. +* Coffee + * [重磅！多篇研究证实喝咖啡与人群全因死亡率降低直接相关](https://news.bioon.com/article/6725420.html)*(Mandarin, Use Chrome's Translator)* + * [科普 | 喝咖啡又多了一个新理由: 降低死亡率！ ](https://www.sohu.com/a/439412995_100003595)*(Mandarin, Use Chrome's Translator)* + * [地中海成年人咖啡消耗量及全因, 心血管疾病和癌症的死亡率](https://fanyi.pdf365.cn/help/249)*(Mandarin, Use Chrome's Translator)* + * In the most recent meta-analysis, the study included 40 studies and 3,852,651 subjects from different countries.In this meta-analysis, non-linear relationships between coffee intake and all-cause mortality, CVD and cancer mortality were shown. Two cups of coffee per day were associated with the lowest cancer mortality (RR = 0.96), 2.5 cups per day for CVD mortality (RR = 0.83), and 3.5 cups per day for all-day mortality (RR = 0.85). And as coffee consumption increased, mortality did not further decrease or increase. +* Spermidine + * [Science: 科学背书！从精液中发现的亚精胺, 竟然有着抗衰老, 抗癌, 保护心血管和神经, 改善肥胖和2型糖尿病等逆天神效](https://www.medsci.cn/article/show_article.do?id=420d12904103)*(Mandarin, Use Chrome's Translator)* + * [饮食中亚精胺摄入量高会降低死亡率](https://zhuanlan.zhihu.com/p/388942219)*(Mandarin, Use Chrome's Translator)* + +##### 6.1.3. Gas + +* Smoking + * [即使是低强度吸烟, 也增加死亡风险！](https://www.medsci.cn/article/show_article.do?id=02ca2083319b)*(Mandarin, Use Chrome's Translator)* + * Among the 42,416 men and 86,735 women(aged 35-89 years with no previous disease), 18,985 men (45%) and 18,072 women (21%) were current smokers, with 33% of men and 39% of women not smoking daily. 8866 men (21%) and 53 912 women (62%) never smoked. During follow-up, the hazard ratios for all-cause mortality compared with never smoking were 1.17 (95% CI 1.10-1.25) and 1.54 (1.42-1.67) for <10 cigarettes per day or ≥10 cigarettes per day, respectively. Hazard ratios were similar regardless of age or gender. The diseases most closely associated with daily smoking are respiratory cancers, chronic obstructive pulmonary disease, and gastrointestinal and vascular diseases. People who quit smoking at the time of recruitment had lower mortality rates than current daily smokers. + * Smokers lose 11-12 years of life on average. + * [吸烟让人过瘾是什么原理？有节制的吸烟依旧有害吗？](https://www.zhihu.com/question/24846224/answer/1719798177)*(Mandarin, Use Chrome's Translator)* + +##### 6.1.4. Illumination + +* Sunbathe + * [晒太阳和死亡率的关系, 如何科学, 安全的晒太阳？ +](https://zhuanlan.zhihu.com/p/76301306)*(Mandarin, Use Chrome's Translator)* + * A 26-year Danish study found that more sun exposure significantly increased life expectancy, and even patients with skin cancer caused by excessive sun exposure lived an average of 6 years longer than the average person. + +##### 6.1.5. Medicine + +* NMN +* Metformin + * [“胍”吹必看 丨我就是神药——二甲双胍](https://zhuanlan.zhihu.com/p/419202902)*(Mandarin, Use Chrome's Translator)* + * Metformin not only plays a protective role in various tumors, cardiovascular diseases and diabetes, but also plays a role in obesity, liver disease, kidney disease and aging. + * [二甲双胍2020最值得了解的“吃瓜”大新闻——护胃, 健脑, 抗衰, 防癌还是致癌？](https://zhuanlan.zhihu.com/p/357807109)*(Mandarin, Use Chrome's Translator)* + * [二甲双胍真的那么神吗？美研究: 父亲服用二甲双胍或致子女有缺陷](https://baijiahao.baidu.com/s?id=1729999374705305768)*(Mandarin, Use Chrome's Translator)* + * ![二甲双胍](https://user-images.githubusercontent.com/2707039/163702325-5d427542-9ae5-4311-8979-d0d326a9832f.jpg) + * Adverse reactions + * As a drug that has been used for nearly a hundred years, the adverse reactions of metformin have been very clear. The common ones are: vitamin B12 deficiency (7%-17.4%), gastrointestinal adverse reactions (up to 53%), fatigue (9%), headache (6%). + * Serious but infrequent adverse reactions include lactic acidosis, liver damage. + * Studies have also shown that it may be teratogenic to the fetus +* Multi-vitamins + * [服用复合维生素可降低癌症危险8%, 其他效果并不显著](https://health.qq.com/a/20121023/000026.htm)*(Mandarin, Use Chrome's Translator)* +* Glucosamine + * [神奇！氨糖降低心血管死亡率65%, 与定期运动效果相当](https://www.sohu.com/a/436372221_120873241)*(Mandarin, Use Chrome's Translator)* + * The latest research from West Virginia University found that glucosamine (chondroitin) can reduce cardiovascular mortality by 65% and reduce overall mortality by 39%, which is equivalent to regular exercise. + * The study used National Health and Nutrition Examination (NHANES) data from 16,686 adults from 1999 to 2010. Participants were followed for a median of 107 months, and 648 of them took 500-1000 mg of glucosamine/chondroitin daily for more than one year on a regular basis. +* Spermidine + * [Science: 科学背书！从精液中发现的亚精胺, 竟然有着抗衰老, 抗癌, 保护心血管和神经, 改善肥胖和2型糖尿病等逆天神效](https://www.medsci.cn/article/show_article.do?id=420d12904103)*(Mandarin, Use Chrome's Translator)* + * Spermidine is the most easily absorbed polyamine from the human gut. Spermidine is found in high amounts in many foods, such as fresh green peppers, wheat germ, cauliflower, broccoli, mushrooms and various cheeses, especially in soy products such as natto, shiitake mushrooms and durian. In this experiment, the researchers selected 829 participants aged 45-84 for a 20-year follow-up to analyze the potential association between dietary spermidine intake and human mortality. + * The study found that women's spermidine intake was higher than men's, and the intake decreased with age. The main sources of spermidine were whole grains (13.4%), apples and pears (13.3%), salads (9.8%), sprouts (7.3%) and potatoes (6.4%). The study divided the population into three groups based on spermidine intake, low intake (<62.2 µmol/d), moderate intake (62.2–79.8 µmol/d), and high intake (>79.8 µmol/d). A total of 341 deaths were recorded during follow-up, including 137 from vascular disease, 94 from cancer, and 110 from other causes. The crude mortality rates for the low, middle, and high groups were calculated to be 40.5%, 23.7%, and 15.1%, respectively. These data indicate a significant inverse relationship between spermidine intake and all-cause mortality. The relationship remained significant with gradual adjustment for age, sex, and calorie ratios. +* Complex + * [《自然》子刊深度综述: 如何开发抗衰老药](https://zhuanlan.zhihu.com/p/145495570)*(Mandarin, Use Chrome's Translator)* + * ![如何开发抗衰老药](https://user-images.githubusercontent.com/2707039/163702474-205baeec-f0ce-4e8d-96a4-36efe47534de.jpg) + +#### 6.2. Output + +##### 6.2.1. Racquet Sports + +* [哪种运动性价比最高？权威医学杂志“柳叶刀”给出答案了 ](https://www.sohu.com/a/535581770_121124216)*(Mandarin, Use Chrome's Translator)* + * Three times a week, 45-60 minutes of racquet sports, reduces all-cause mortality by ~47% + * Badminton, table tennis, tennis, etc. are all racquet sports. But due to the westernized research background, it may refer to tennis more. This implicitly expresses that the whole body exercise is more important. (Translator's Note: pickleball) + +##### 6.2.2. High-intensity Exercise + +* [New Study: 8 minutes of high-intensity exercise per day can reduce all-cause mortality and risk of heart disease](https://academic.oup.com/eurheartj/advance-article/doi/10.1093/eurheartj/ehac572/6771381) + * 15-20 minutes of high-intensity exercise per week reduces all-cause mortality by 16-40%, and by spending 50-57 minutes/week on high-intensity exercise, you can further decrease your all-cause mortality rate. These findings suggest that by accumulating a relatively small amount of high-intensity exercise over a short time during the week, health risks can be reduced. + +##### 6.2.3. Walk + +* [走路降低全因死亡率超过50%！每天走多少步最合适？《JAMA》子刊超10年研究告诉你答案](http://www.shcell.org/219/3571.html)*(Mandarin, Use Chrome's Translator)* + * ![走路降低全因死亡率](https://user-images.githubusercontent.com/2707039/163704147-afec1c79-799b-4db8-b547-1a2431d504c9.jpg) + * Note 1: The average age of participants in this study was 45.2 years old. + * Note 2: The average number of steps is related to occupation. This study only shows correlation, not causation. + +##### 6.2.4. Tooth brushing + +* [50万国人研究证实: 不好好刷牙, 致癌！血管疾病也会增多！](https://www.cn-healthcare.com/articlewm/20211209/content-1293760.html)*(Mandarin, Use Chrome's Translator)* + * People who don't brush their teeth regularly: increased risk of cancer, chronic obstructive pulmonary disease and cirrhosis by 9%, 12% and 25%, respectively, and increased risk of premature death by 25%. + +##### 6.2.5. Bath + +* [定期洗澡降低心血管疾病发作风险](https://www.cn-healthcare.com/article/20200326/content-533379.html)*(Mandarin, Use Chrome's Translator)* + * A daily hot bath was associated with a 28% lower overall risk of cardiovascular disease, a 26% lower overall risk of stroke, and a 46% lower risk of cerebral hemorrhage compared to taking a bath one to two times a week or no bathing at all. The frequency was not associated with an increased risk of sudden cardiac death. + +##### 6.2.6. Housework(for aging males) + +* [Housework Reduces All-Cause and Cancer Mortality in Chinese Men](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0061529) + * Weekly heavy housework for men after age 72 can reduce the average mortality rate by 29%. + > Heavy housework: vacuuming, mopping floors, mopping floors, scrubbing windows, washing cars, moving furniture, moving gas tanks, etc. + > + > Light housework: dusting, washing dishes, hand washing clothes, ironing, drying clothes, cooking, shopping for daily necessities, etc. + +##### 6.2.7. Sleep + +* [超30万亚洲人数据: 每天睡几个小时最有益长寿？](https://med.sina.com/article_detail_103_1_105491.html)*(Mandarin, Use Chrome's Translator)* + * For men, a sleep duration of ≥10 hours was associated with a 34% increased risk of all-cause mortality compared with a sleep duration of 7 hours. + * ![睡眠-男](https://user-images.githubusercontent.com/2707039/163704166-226b7ebb-92ce-4753-a3e7-77a87652a104.jpg) + * For women, a sleep duration of ≥10 hours was associated with a 48% increased risk of all-cause mortality compared with a sleep duration of 7 hours. + * ![睡眠-女](https://user-images.githubusercontent.com/2707039/163704169-c5c715aa-7130-403b-b0d1-ec34fab094d8.png) +* [颠覆认知！加拿大研究发现: 早睡比熬夜或许更伤身, 几点睡才好？](https://www.thepaper.cn/newsDetail_forward_14461799)*(Mandarin, Use Chrome's Translator)* + * One conclusion is that bedtime is strongly associated with all-cause mortality, and that going to bed too early and going to bed too late can affect health. But going to bed early was associated with higher all-cause mortality than going to bed late, with an increased risk of death by 43 percent and going to bed late by 15 percent. + * There are still many limitations in this research study. For example, it does not directly prove the relationship between bedtime and death, only shows the correlation, and the data obtained by using the self-reports of the participants to count sleep time are not objective enough. + +##### 6.2.8. Sitting for a long time + +* [Scientific Research Report of Chinese Resident Dietary Guidelines (2021)] (https://www.chinanutri.cn/yyjkzxpt/yyjkkpzx/yytsg/zgjm/202103/p02021031148670527.pdf) +* Sedentary and watching TV time is related to the high risk of death, cardiovascular disease, cancer, and 2 diabetes. It is an independent risk factor.For a long time, every 1 hour is increased every day, the risk of cardiovascular disease increases by 4%, and cancer increases by 1%, and the risk of death increases by 3%.The long -seated time threshold with increased risk of death and CVD death is 6 \ ~ 8H/d, and the time threshold for watching the TV is 3 \ ~ 4H/D. +* [Guide to the WHO on physical activity and sedentary behavior] (https://apps.who.int/iris/bitstream/handle/10665/337001/978924947- chi.pdf)) + + +#### 6.3. Context + +##### 6.3.1. Emotion + +* [悲观情绪与更高的全因死亡率和心血管疾病死亡率有关, 但乐观情绪并不能起到保护作用](https://www.x-mol.com/paper/1288184397379059712/t?recommendPaper=1263704526086578176)*(Mandarin, Use Chrome's Translator)* +* [Pessimism is associated with greater all-cause and cardiovascular mortality, but optimism is not protective](https://www.nature.com/articles/s41598-020-69388-y?utm_source=xmol&utm_medium=affiliate&utm_content=meta&utm_campaign=DDCN_1_GL01_metadata_scirep) + * The Life Orientation Test (LOT), containing items for optimism and pessimism, was included in a twin study on health of Australians aged over 50 in 1993–1995. After a mean of 20 years, participants were matched against death information from the Australian National Death Index. 1,068 out of 2,978 participants with useable LOT scores had died. Survival analysis tested for associations between separate optimism and pessimism scores and mortality from any cause, and from cancers, cardiovascular diseases or other known causes. Age-adjusted scores on the pessimism scale were associated with all-cause and cardiovascular mortality (Hazard Ratios per 1 standard deviation unit, 95% confidence intervals and p-values 1.134, 1.065–1.207, 8.85 × 10–5 and 1.196, 1.045–1.368, 0.0093, respectively) but not with cancer deaths. Optimism scores, which were only weakly correlated with pessimism scores (age-adjusted rank correlation = − 0.176), did not show significant associations with overall or cause-specific mortality. Reverse causation (disease causing pessimism) is unlikely because in that case both cardiovascular diseases and cancers would be expected to lead to pessimism. + +##### 6.3.2. Wealth + +* [JAMA子刊: 贫富差距真能影响寿命？这可能是真的！](https://www.cn-healthcare.com/articlewm/20210727/content-1246348.html)*(Mandarin, Use Chrome's Translator)* + * The study used data from wave 1 of the MIDUS study (MIDUS 1), which were collected in 1994 to 1996. Survival models tested the association between net worth and all-cause mortality. The full sample comprised 5414 participants, who had a mean (SD) age of 46.7 (12.7) years and included 2766 women (51.1%). Higher net worth was associated with lower mortality risk (hazard ratio [HR], 0.95; 95% CI, 0.94-0.97; P<.001). Among siblings and twin pairs specifically (n=2490), a similar within-family association was observed between higher net worth and lower mortality (HR, 0.94; 95% CI, 0.91-0.97; P=.001), suggesting that the sibling or twin with more wealth tended to live longer than their co-sibling or co-twin with less wealth. + +##### 6.3.3. Weight + +* [JAMA子刊: 减肥要趁早, 才能有效降低死亡率风险](https://www.chinacdc.cn/gwxx/202009/t20200904_218959.html)*(Mandarin, Use Chrome's Translator)* + * Adults who lost weight from obese to overweight by midlife had a 54% lower risk of all-cause mortality (HR, 0.46; 95% CI, 0.27-0.77) compared with adults who were stable obese. Adults who lost weight from overweight in early adulthood to normal weight by midlife did not have reduced mortality risk (HR, 1.12; 95% CI, 0.86-1.45). + * ![Table3](https://raw.githubusercontent.com/qhy040404/Image-Resources-Repo/master/zoi200509t3_1596761185.02415.png) + +##### 6.3.4. COVID-19 + +* [Magnitude, demographics and dynamics of the effect of the first wave of the COVID-19 pandemic on all-cause mortality in 21 industrialized countries](https://www.nature.com/articles/s41591-020-1112-0.pdf) + * At present, the death rate of COVID-19 (US) is around 1.5%, and the average life expectancy has been reduced by 2 years +* [如何看待美国CDC宣称新冠死亡人数被高估？](https://www.zhihu.com/question/510943670/answer/2308499719)*(Mandarin, Use Chrome's Translator)* +* [NVSS deaths](https://www.cdc.gov/nchs/nvss/deaths.htm) From 229fd458cc054a4bf6613ce5b6785cfa3df1f3bf Mon Sep 17 00:00:00 2001 From: Stream Entry <978862+streamentry@users.noreply.github.com> Date: Fri, 2 Oct 2026 12:23:37 +0700 Subject: [PATCH 12/25] =?UTF-8?q?=E8=8B=B1=E6=96=87=E6=90=AC=E5=88=B0?= =?UTF-8?q?=E6=A0=B9=20URL=EF=BC=9BEN=2032/34=E3=80=81VI=2015/34=20?= =?UTF-8?q?=E8=8A=82=E5=B7=B2=E7=BF=BB=E5=B9=B6=E5=90=88=E5=85=A5=20(#7)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * 越南文第 2 节（不要慢慢死）翻完并合入 九条慢性病条目：筛查、运动、戒烟酒、饮食、血压血糖、疫苗、牙。 数字照抄，量级词换算（每 10 万人 → trên 100.000 người）， 法条引用留中文原文再给白话。校验：合好 1 节，问题 0 处。 越南文 2/34 节。 * 越南文第 4、5 节（不要浪费时间、不要浪费钱）翻完并合入 第 4 节三条讲时间怎么用，第 5 节四十五条讲钱怎么不漏出去： 消费贷、信用卡、订阅、保险、保修、诈骗、比价、二手货。 金额、利率、期限照抄，量级词换算成越南写法。 校验：两节各「合好 1 节，问题 0 处」。越南文 4/34 节。 * 英文第 29 到 31 节翻完并合入 重大打击之后怎么缓过来、上学以后的孩子、十八岁之后有哪几条路。 学费和生活费估算、年龄、学级、课时、各项数字照抄， 大学和专业名保留原形、首次出现给英文对照。 校验：三节各「合好 1 节，问题 0 处」。英文 29/34 节。 * 英文第 32 节、越南文第 14、15、20 节翻完并合入 英文：出国留学（学费、生活费估算、期限、奖学金名照抄）。 越南文：账号与信息安全、租房与买房、刚出生的孩子怎么带。 新生儿那节的体温、剂量、天数、百分比一律照抄，没做换算。 校验：四节各「合好 1 节，问题 0 处」。英文 30/34 节，越南文 7/34 节。 * 越南文第 16 节（得了慢性病之后怎么活）翻完并合入 两条：自己管病、家里人和钱。数值（剂量、复查间隔、比例）照抄没换算。 校验：合好 1 节，问题 0 处。越南文 8/34 节。 * 越南文第 21、22 节（出国旅行与境外安全、怎么放松）翻完并合入 旅行那节把热线和服务号码（12308、12345、12320）原样留着， 证件、机构的原名保留中文加越南语对照。 校验：两节各「合好 1 节，问题 0 处」。越南文 10/34 节。 更正一条数：上一条提交信息里写的「英文 30/34」不对， en/book 下实际是 27 份，英文 27/34 节（1 到 23 加上 29 到 32）。 * 越南文第 13 节（紧急情况）四十二条翻完并合入 全书最密的一节：心脏骤停、中风、过敏、癫痫、电击、烫伤、中毒、 出血、骨折、中暑、火灾、溺水、蛇咬、地震、狂犬病暴露。 每一处分钟数、秒数、体温、剂量、百分比、年龄区间、号码 （120、110、119、12320）都照抄，没做换算也没改单位； 写成短句祈使句，一个动作一句。 校验：9 块合好，42 条，问题 0 处。越南文 11/34 节。 * i18n: join en section 24 (看病) * 英文、越南文第 28 节（别为了外形把身体搞坏）翻完并合入；加 join-ready.sh 一键合入已翻好的节 * 英文搬到根 URL：默认语言就在 /，/en/ 留转发页 要求 2 说英文是默认语言、服务在根 URL。以前根上是一块跳语言的小页：开 JS 按 navigator.language 跳走，跳转表里的正则被 JSON.stringify 序列化成了 {}（RegExp 序列化不出自己），一执行就抛 TypeError——根 URL 就停在跳页上，英文默认压根没兑现。 而且就算能跳，读者也被直接送走，页内的语言下拉和锚点都没机会用。 现在： · / 直接是英文正文页（locales.json 里 en 的 dir 是空串，canonical 指向根）。 · 老链接形式 /?lang=vi 由 chooserJump() 接住：命中就跳，没带就留在原地。 · /en/ 留一块转发页，README 和别处早就散着这个地址。 · 页内语言下拉修好了：选项地址原来写的是 SITE.dir（「这一页在哪」），跟「要去 哪种语言」无关——从 /zh/ 选英文还是落在 /zh/。改成先退回站点根再拼 langDirs[code]。 · sitemap 去重：默认语言住在根上，它的 canonical 就是根 URL，不去重会列两遍。 顺带把「dir 和 contentDir 不是一回事」这件事钉死：dir 是页面上线的目录（英文是 空串），contentDir 是正文在仓库里的目录（英文是 en）。join.mjs 和 check.mjs 原来 拿 L.dir 拼仓库路径，英文搬到根上之后会把译文写进中文原文，已经改走 contentPath()。 verify 加了第三步 lang-switcher-check：把构建产物里那段 buildLangPicker 抠出来， 给它一份假的 location 真跑一遍，逐个落点核对地址、查询串和锚点。抠的是发出去的 那份代码本身，不是照着抄的，抄错了测不出来。 EN 31/34、VI 14/34 节已翻并合入。 * 文档跟上：英文在根 URL 上的那一版，路径怎么拼（dir 和 contentDir），verify 的三步 * 英文、越南文第 09 节（法律红线）等若干节翻完并合入：EN 32/34、VI 15/34 --------- Co-authored-by: j --- CLAUDE.md | 21 +- README.md | 6 +- en/README.md | 10 +- "en/book/24-\347\234\213\347\227\205.md" | 121 ++++++ ...01\345\212\236\344\273\200\344\271\210.md" | 104 +++++ ...25\345\222\214\347\224\237\344\272\247.md" | 157 +++++++ ...53\344\275\223\346\220\236\345\235\217.md" | 87 ++++ ...23\345\207\273\344\271\213\345\220\216.md" | 130 ++++++ ...16\347\232\204\345\255\251\345\255\220.md" | 147 +++++++ ...52\345\207\240\346\235\241\350\267\257.md" | 158 +++++++ ...72\345\233\275\347\225\231\345\255\246.md" | 108 +++++ ...53\345\220\203\345\207\272\344\272\213.md" | 103 +++++ tools/i18n/README.md | 18 +- tools/i18n/check.mjs | 10 +- tools/i18n/join-ready.sh | 21 + tools/i18n/join.mjs | 11 +- tools/site/build.mjs | 103 +++-- tools/site/locales.json | 8 +- tools/site/locales.mjs | 14 +- tools/site/page.template.html | 16 +- ...01\346\205\242\346\205\242\346\255\273.md" | 370 +++++++++++++++++ ...52\350\264\271\347\262\276\345\212\233.md" | 217 ++++++++++ ...52\350\264\271\346\227\266\351\227\264.md" | 167 ++++++++ ...01\346\265\252\350\264\271\351\222\261.md" | 391 ++++++++++++++++++ ...25\345\276\213\347\272\242\347\272\277.md" | 213 ++++++++++ ...47\346\200\245\346\203\205\345\206\265.md" | 388 +++++++++++++++++ ...41\346\201\257\345\256\211\345\205\250.md" | 92 +++++ ...77\344\270\216\344\271\260\346\210\277.md" | 86 ++++ ...16\346\200\216\344\271\210\346\264\273.md" | 86 ++++ ...14\346\234\211\350\200\201\344\272\272.md" | 83 ++++ ...20\346\200\216\344\271\210\345\270\246.md" | 102 +++++ ...03\345\244\226\345\256\211\345\205\250.md" | 94 +++++ ...16\344\271\210\346\224\276\346\235\276.md" | 94 +++++ ...53\344\275\223\346\220\236\345\235\217.md" | 73 ++++ web/package.json | 3 +- web/src/lib/astro-page.ts | 12 +- web/src/lib/lang-switcher-check.ts | 98 +++++ web/src/pages/en.astro | 39 +- web/src/pages/index.astro | 13 +- 39 files changed, 3864 insertions(+), 110 deletions(-) create mode 100644 "en/book/24-\347\234\213\347\227\205.md" create mode 100644 "en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" create mode 100644 "en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" create mode 100644 "en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" create mode 100644 "en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" create mode 100644 "en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" create mode 100644 "en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" create mode 100644 "en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" create mode 100644 "en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" create mode 100755 tools/i18n/join-ready.sh create mode 100644 "vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" create mode 100644 "vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" create mode 100644 "vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" create mode 100644 "vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" create mode 100644 "vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" create mode 100644 "vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" create mode 100644 "vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" create mode 100644 "vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" create mode 100644 "vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" create mode 100644 "vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" create mode 100644 "vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" create mode 100644 "vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" create mode 100644 "vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" create mode 100644 "vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" create mode 100644 web/src/lib/lang-switcher-check.ts diff --git a/CLAUDE.md b/CLAUDE.md index a8b95393..bb2bbe83 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -83,31 +83,38 @@ | 路径 | 是什么 | 谁维护 | | --- | --- | --- | -| `/` | 跳语言的小页，按浏览器语言去 `/en/` 或 `/vi/`，关掉 JS 也能点 | `tools/site/build.mjs` 生成 | -| `/en/` | 英文站（默认） | 译文，见 `en/` | +| `/` | **英文站（默认语言就住在根 URL 上）** | 译文，见 `en/` | +| `/en/` | 上面那一页的别名，跟着转到 `/` | 转发页，`web/src/pages/en.astro` | | `/vi/` | 越南文站 | 译文，见 `vi/` | | `/zh/` | 中文站，读的是**仓库根**那份正文 | 原始内容都在根上 | +**根 URL 为什么是英文而不是跳语言页**：默认语言住在根上是要求（要求 2）。以前 `/` 是一块「跳语言的小页」，开 JS 按 `navigator.language` 跳走——那样根 URL 发出的是中文标题的跳页，英文并没有真的落在根上，而且自动跳转把读者直接送走，页内语言下拉和锚点都没机会用。现在 `/` 直接是英文正文页，语言选择就在页头下拉里（要求 4：换语言保留 `?q=` 和 `#e-xx-y`）。保留下来的只有老链接形式 `/?lang=vi`，由 `chooserJump()` 那段小脚本接住，命中就跳、没带就留在原地。`/en/` 留着一块转发页，因为 README 和别处早就散着 `/en/` 的链接。 + +**改 `locales.json` 的 `dir` 前先想清楚它被谁用**：它既是页面上线的目录（默认语言是空串 = 根），也被 `tools/offline/build.mjs` 拿来拼线上链接。**仓库里找译文一律走 `contentPath()`（按 `contentDir` 拼），别用 `dir`**——英文的 `dir` 现在是空串，拿它拼路径会拼到 `book/`，也就是中文原文那一份。2026-10-02 差点踩这个坑：`join.mjs` 原来用 `L.dir` 写盘，改成根上之后英文译文会被写进中文原文。 + **中文正文仍在仓库根**：`README.md`、`book/`、`docs/` 都没挪。`zh/index.html` 用 `contentBase: '../'` 指回根上那一份，所以作者改的还是原来那些文件，GitHub 渲染的还是仓库根的 README。en/vi 各自有 `README.md`、`book/`、`docs/`，是那一份语言的副本。 -检索页的源码是 `tools/site/page.template.html`，**三份页面都由 Astro 构建（`web/`）从模板加 `tools/site/locales/.json` 生成，构建产物直发 Pages，不经过仓库**。改页面改模板和字典，别去 `web/dist` 里改（那是构建缓存，已 gitignore）。根目录那个跳语言的小页也是 Astro 出的。 +检索页的源码是 `tools/site/page.template.html`，**三份正文页面都由 Astro 构建（`web/`）从模板加 `tools/site/locales/.json` 生成，构建产物直发 Pages，不经过仓库**。改页面改模板和字典，别去 `web/dist` 里改（那是构建缓存，已 gitignore）。`/en/` 那块转发页和根上那段 `?lang=` 跳转脚本也是 Astro 出的。 + +**`web && npm run verify` 是四步一根链**：解析一致（`parity.ts`）→ 构建 → 语言下拉落点（`lang-switcher-check.ts`）→ 渲染一致（`render-check.ts`）。第三步单独在那儿，是因为语言下拉是从构建产物里抠出来真跑的——页内那段 `buildLangPicker` 读 `location`，只有浏览器才有，这里给它一份假的。抠的是发出去的那份代码本身，不是照着抄的，所以抄错测不出来。它拦的是这一类错：选项地址曾经写成 `SITE.dir`（「这一页在哪」），跟「要去哪种语言」无关，从 `/zh/` 选英文还是落在 `/zh/`，换语言等于没换。 ### 站点怎么上线 - **部署用 GitHub Pages 的 workflow 发布**（`.github/workflows/web.yml`，2026-09-30 从分支发布切过来的）：CI 里跑 `web/` 的 Astro 构建，产物（`web/dist`）直接发 Pages，**不入库**。`.nojekyll` 由构建拷进产物，别删根上那份，删了构建也跟着没了。 -- **构建时把卡片直接写进 HTML**（预渲染）：爬虫和关掉 JS 的读者不用等 JS 就看到全文。这是 Astro 迁移的全部理由——之前是客户端渲染，`en/index.html` 里 95 KB 全是框架，一个正文字都没有，AI 抓取器和链接预览只能看到空壳。 -- **预渲染和运行时是同一份逻辑的两份拷贝**：`web/src/lib/parse.ts` 抄模板的 `parseReadme`，`web/src/lib/render.ts` 抄模板的 `build()` 初次渲染。两份各自演化迟早对不上，对不上时爬虫和读者看到的就不是一本书了。所以有两道机器检查，`web/` 里跑 `npm run verify` 一次全过：① `npm run check`（`parity.ts`）把原版 `parseReadme` 从模板里抠出来逐字段比；② `render-check.ts` 跑在构建产物上，查每条六个栏的每个词都在 HTML 里、hydration 钩子齐、引用不悬空。**改模板的渲染逻辑必须同时改 `render.ts`，改判据是这两道都绿。** +- **构建时把卡片直接写进 HTML**（预渲染）：爬虫和关掉 JS 的读者不用等 JS 就看到全文。这是 Astro 迁移的全部理由——之前是客户端渲染，`en/index.html` 里 95 KB 全是框架，一个正文字都没有，AI 抓取器和链接预览只能看到空壳。（英文那份页面现在叫 `index.html`，在根上，`en/index.html` 是转发页。） +- **预渲染和运行时是同一份逻辑的两份拷贝**：`web/src/lib/parse.ts` 抄模板的 `parseReadme`，`web/src/lib/render.ts` 抄模板的 `build()` 初次渲染。两份各自演化迟早对不上，对不上时爬虫和读者看到的就不是一本书了。所以有三道机器检查，`web/` 里跑 `npm run verify` 一次全过：① `npm run check`（`parity.ts`）把原版 `parseReadme` 从模板里抠出来逐字段比；② `npm run check-lang`（`lang-switcher-check.ts`）把构建产物里那段 `buildLangPicker` 抠出来、给它一份假的 `location` 真跑，逐个落点核对语言下拉的地址、`?q=` 和锚点；③ `render-check.ts` 跑在构建产物上，查每条六个栏的每个词都在 HTML 里、hydration 钩子齐、引用不悬空。**改模板的渲染逻辑必须同时改 `render.ts`，改判据是这三道都绿。** - **浏览器里还是原来那个页面**：预渲染只负责首屏和爬虫，JS 跑起来后索引现有 DOM（模板里的 `indexSsgDom`），筛选、搜索、弹窗、高亮全是原来的逻辑。对不上就回退重建并 `console.warn`——正常构建永远不该走到那里。 - **域名只写在 `tools/site/site.json` 的 `site` 和 `repo` 两个值里**，别的文件一处都不许再抄。`locales.json` 里每种语言的 `canonical` 和 `repoBlob` 是从这两个值推出来的，构建时逐个对一遍（`checkUrls()`），**对不上直接抛错**，不会出现「Pages 发到 A 域名、页面自称是 B 域名」。`sitemap.xml` 和 `robots.txt` 由 Astro 的两个 endpoint 从同一处生成，域名单独抄一遍就会漏。 - **四种链接要分清，改域名时只动前两种**：① 站点链接（`*.github.io/…`）和② 面向读者的 GitHub 链接（页面里的文件链接、`/releases/download/…`）跟着 `site.json` 走；③ `git clone` 那行、④ 「写明出处／Credit the source」和 Star History 图这三处**永远指向上游 `eternity4719/HowToLiveBetter`**，它们是在说明这本书的出处，不是本站的地址。2026-09-30 换域名时按这个分的，核对的判据是 `grep -c eternity4719 README.md en/README.md` 各应为 3，且都在那三类位置上。 - **`tools/site/build.mjs` 现在是库**：`renderPage` 给离线单文件构建和 Astro 调，写盘逻辑只在直接运行时走。仓库根不再有它生成的页面（`index.html`、`*/index.html`、`sitemap.xml`、`robots.txt` 都已删），所以 `build.mjs --check` 的 CI job 跟着退役了——模板和字典的一致性由 Astro 构建本身保证（字典缺键直接抛错），由 `extract.mjs --check` 守中文文案。 - **CI 分两套，互不挡**：`book.yml` 管电子书和正文检查，`web.yml` 管网站（校验→构建→发布）。`web.yml` 的触发路径只收和网站有关的文件，改 EPUB/PDF 模板不触发网站构建，反之亦然。 -- **没翻完的语言不会上线，但会被列出来**：`PUBLISHED`（README 在的才算）决定 hreflang、跳语言页和页内语言下拉——`vi/index.html` 不存在时页面上也没有一个点进去 404 的越南语入口。构建日志里报「跳过 vi」和「另有 N 项在目录里列着、正文还没翻到」，前者是没生成页面，后者是页面生成了但目录里那一节还没有译文。 +- **没翻完的语言不会上线，但会被列出来**：`PUBLISHED`（README 在的才算）决定 hreflang 和页内语言下拉——`vi/index.html` 不存在时页面上也没有一个点进去 404 的越南语入口。构建日志里报「跳过 vi」和「另有 N 项在目录里列着、正文还没翻到」，前者是没生成页面，后者是页面生成了但目录里那一节还没有译文。 +- **`dir` 和 `contentDir` 不是一回事**（2026-10-02 定的，英文搬到根上之后）：`dir` 是页面上线的目录（英文是 `''`，页面在 `/`），`contentDir` 是正文在仓库里的目录（英文是 `en`）。`tools/site/locales.mjs` 的 `contentPath()` 把两者接起来，仓库里拼路径只用它。页面上的相对根是第三个东西 `contentBase`（英文那份页面在根上，读的是 `en/`，所以是 `en/` 不是 `./`）。 **正文按节拆成 `book/01-*.md` … `book/34-*.md`（2026-09-08 拆的，原来单文件 531 KB，超过 GitHub 渲染 Markdown 的 512 KB 上限，后面的节显示不出来也跳不了锚点）。** README 只留导读、术语表和目录，新增或修改条目改对应的 book 文件；检索页先读 README 拿目录里的文件列表，再并发读这些文件（清单由构建给，见 `SITE.files`）。每节文件第一行是回总目录的链接，第二行空行，第三行是 `# N. 节名`。 ### 译文怎么维护 - **流程和判据写在 `tools/i18n/BRIEF.md`**，翻译任务的提示词和结构检查的判据是同一份。词表 `tools/i18n/glossary.{en,vi}.mjs`。 -- 一次翻一节。`node tools/i18n/split.mjs book/NN-*.md --locale en` 把它切成十来个条一块放进 `.i18n/en/<节名>/partNN.md`（**已翻过的块不会被覆盖**，改参数重跑不会冲掉译文），翻译完 `node tools/i18n/join.mjs book/NN-*.md --locale en` 逐块校一遍再合成 `en/book/NN-*.md`。 +- 一次翻一节。`node tools/i18n/split.mjs book/NN-*.md --locale en` 把它切成十来个条一块放进 `.i18n/en/<节名>/partNN.md`（**已翻过的块不会被覆盖**，改参数重跑不会冲掉译文），翻译完 `node tools/i18n/join.mjs book/NN-*.md --locale en` 逐块校一遍再合成 `en/book/NN-*.md`。一轮翻好几节用 `bash tools/i18n/join-ready.sh en vi`——它按「块里还带 `- 成本：` 就是没翻」挑出翻全了的节，挨个 join。 - `node tools/i18n/check.mjs`（CI 的「译文检查」job）拿中文原文逐条对译文：条数条号、六个字段的有无与顺序、**成本标签注释逐字照抄**、证据等级、争议与待核实的条数、**收益/成本/来源三栏的数字一个不多一个不少**（量级词换算除外，见 `tools/i18n/structure.mjs` 里 `numbersOf` 的说明）、「第 X 条」指路的条号节号、来源与备注里的链接数量。判据在 `structure.mjs`，join 和 check 共用同一份。 - **译文不许改数字**。中文「167 万」在英文里只能写成 1,670,000、在越南文里只能写成 1.670.000，这是量级换算，`structure.mjs` 放行；除此之外数字对不上就是错。这一层机器查得出来，所以译文不需要作者审。 - 越南文里 **条目 = "mục"，节 = "phần"，法条的款 = "khoản (2)"**——三个词不能混。「第 13 节」写成 "phần 13"，写成 "mục 13" 读者就当成第 13 条了。 diff --git a/README.md b/README.md index 480f8550..60c741cc 100644 --- a/README.md +++ b/README.md @@ -3,7 +3,7 @@ 高性价比人生指南 —— 用最少的钱、时间和精力，换回最多的寿命、金钱和人身自由 > Vietnamese: https://streamentry.github.io/HowToLiveBetter/vi/ -> English: https://streamentry.github.io/HowToLiveBetter/en/ +> English: https://streamentry.github.io/HowToLiveBetter/ # 高性价比人生指南 @@ -40,7 +40,7 @@ AI 助手 skill 支持 Claude Code 和 Codex。装上后直接问「替朋友担 其他语言 -**官方**：[English](https://streamentry.github.io/HowToLiveBetter/en/)（默认，访问 `/` 就是它） · 中文（本页，原始内容都在这里），译文跟不上时以中文原文为准。Tiếng Việt 在翻，尚未上线。 +**官方**：[English](https://streamentry.github.io/HowToLiveBetter/)（默认，访问 `/` 就是它） · 中文（本页，原始内容都在这里），译文跟不上时以中文原文为准。Tiếng Việt 在翻，尚未上线。 **他人维护**：[Русский](https://dlgrv.github.io/HowToLiveBetter/ru/) · [Español](https://dlgrv.github.io/HowToLiveBetter/es/)，[dlgrv](https://github.com/dlgrv) 的翻译（[仓库](https://github.com/dlgrv/HowToLiveBetter)） @@ -283,7 +283,7 @@ node tools/pdf/build.mjs # PDF，另需 pandoc ≥ 3.1 和 typ 每节内条目按性价比从高到低排列。「不要早死」「不要浪费时间」这类节标题说的是这一节想防住的结果，条目本身要做还是别做，以条目标题为准。长文另见 [docs/家庭应急装备清单.md](docs/家庭应急装备清单.md)、[docs/做平台要办哪些证.md](docs/做平台要办哪些证.md)、[docs/结婚划不划算.md](docs/结婚划不划算.md) 、[docs/遇到陌生人出事该不该停.md](docs/遇到陌生人出事该不该停.md) 和 [docs/生物钟和夜班.md](docs/生物钟和夜班.md)。每条来源的核实过程记录在 [docs/核实记录](docs/核实记录/)。 -`en/index.html`、`zh/index.html`、`vi/index.html` 是三种语言的在线检索页：按关键词、章节、证据等级和成本维度（花钱、花时间、要毅力）筛选条目，数据直接读本目录下那一份 README 和 `book/`。根目录的 `index.html` 是跳语言的小页，按浏览器语言去 `/en/`（默认，英文）或 `/vi/`，关掉 JavaScript 也能自己点。在仓库设置里开启 GitHub Pages（Deploy from a branch，分支 main，目录 /）后即可访问。 +三种语言的在线检索页是 `index.html`（英文，默认）、`vi/index.html`、`zh/index.html`：按关键词、章节、证据等级和成本维度（花钱、花时间、要毅力）筛选条目，数据直接读那一种语言的 README 和 `book/`。英文就是根 URL 打开的那一页；页头的语言下拉换语言，当前筛到哪一条、定位在哪一条都跟着过去（`/en/` 是英文的老地址，会转到根上）。在仓库设置里开启 GitHub Pages（Deploy from a branch，分支 main，目录 /）后即可访问。 页面源码是 `tools/site/page.template.html`，三份页面都由 `node tools/site/build.mjs` 从模板加 `tools/site/locales/*.json` 生成，生成物入库是为了让 Pages 直接发；改页面改模板和字典，别改生成出来的那三份。`tools/epub/` 是电子书生成脚本，`cd tools/epub && npm ci && npm run build` 在本地出一本 EPUB 到 `dist/`；GitHub Actions 在正文改动后自动跑同一个脚本并更新 Release。 diff --git a/en/README.md b/en/README.md index c088036a..8548881e 100644 --- a/en/README.md +++ b/en/README.md @@ -9,13 +9,13 @@ On living longer and getting sick less, and on what to do when someone has an ac You do not have to do all of it. This is a shortlist ordered by value for money, not a task list. Taking one or two entries with you counts. The author has not done most of them either. -[![Search online](https://img.shields.io/badge/%E5%9C%A8%E7%BA%BF%E6%A3%80%E7%B4%A2-%E7%82%B9%E8%BF%99%E9%87%8C%E6%89%93%E5%BC%80-3451b2?style=flat-square)](https://streamentry.github.io/HowToLiveBetter/en/) +[![Search online](https://img.shields.io/badge/%E5%9C%A8%E7%BA%BF%E6%A3%80%E7%B4%A2-%E7%82%B9%E8%BF%99%E9%87%8C%E6%89%93%E5%BC%80-3451b2?style=flat-square)](https://streamentry.github.io/HowToLiveBetter/) [![Items](https://img.shields.io/badge/%E6%9D%A1%E7%9B%AE-635%20%E6%9D%A1-18794e?style=flat-square)](#contents) [![Evidence grades](https://img.shields.io/badge/%E8%AF%81%E6%8D%AE%E5%88%86%E7%BA%A7-A%20422%20%C2%B7%20B%20162%20%C2%B7%20C%2051-915930?style=flat-square)](#evidence-grades) [![Original sources](https://img.shields.io/badge/%E5%8E%9F%E5%A7%8B%E6%96%87%E7%8C%AE-1407%20%E6%9D%A1%E9%93%BE%E6%8E%A5-565a5f?style=flat-square)](docs/核实记录/) [![Licence](https://img.shields.io/badge/%E8%AE%B8%E5%8F%AF-CC%20BY%204.0-565a5f?style=flat-square)](#licence) -### [Open the search page](https://streamentry.github.io/HowToLiveBetter/en/) · [Let an AI answer from the book (skill)](skills/life-decision-guide/README.md) +### [Open the search page](https://streamentry.github.io/HowToLiveBetter/) · [Let an AI answer from the book (skill)](skills/life-decision-guide/README.md) The AI assistant skill works with Claude Code and Codex. Once it is installed, ask it something like "should I sign as a guarantor for a friend". It looks up the relevant items in the book first, then answers, and says which section and which item the answer comes from. @@ -96,7 +96,7 @@ English is the default edition, and this page is the source content; translation - **You do not have to do all of it**: this is a shortlist ordered by value for money, not a task list. Taking one or two entries with you counts. Leave the rest and come back when you need them. The remark that this is easy to say and hard to do is fair. The author has not done most of it either. It is written down so that you can find it at the moment you need it. If you want the cheap ones, see "only the ones most worth doing" below. - **If you want an AI to look things up**: the repository ships a Chinese skill ([skills/life-decision-guide](skills/life-decision-guide/)). Both Claude Code and Codex can install it. Once it is installed, ask it directly, for example "should I sign as a guarantor for a friend" or "is a two-hour daily commute worth it". It first pulls the relevant items out of the text, then answers, ranking them the way the book ranks things, and it says which section and which item the answer comes from. If it cannot find anything, it says so. It does not invent numbers. How to install it is in [the notes in that directory](skills/life-decision-guide/README.md). -- **If you want to filter by criteria**: open the [search page](https://streamentry.github.io/HowToLiveBetter/en/). You can filter by keyword, by section and by evidence grade. You can also filter on the three costs, money spent, time spent and willpower needed, and several filters can be combined. The text on the page is read straight from the files in `book/`. When the text changes, the page changes with it. +- **If you want to filter by criteria**: open the [search page](https://streamentry.github.io/HowToLiveBetter/). You can filter by keyword, by section and by evidence grade. You can also filter on the three costs, money spent, time spent and willpower needed, and several filters can be combined. The text on the page is read straight from the files in `book/`. When the text changes, the page changes with it. - **Items point at each other** (things like "see item 17 in section 8"). On the search page such a reference carries a dotted underline. Click it and the title and the "In plain language" line of the item it points to appear in place. To go there for real, press "Go there". If the filters happen to be hiding that item, the page clears the filters by itself. On GitHub, reading the text directly, these references are not clickable. But every one of them says what it points to, for example "see item 18 (write down the IOU when you lend money)". You know which item is meant without following it. - **If you want to read in order**: within each section the items run from the best value for money to the worst. Start with the first few of each section. - **If you want to read it offline or send it to someone**: download the [single-file offline HTML](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html). The whole book, search and filters, is in that one file. Double-click it and it opens. No server, no network. You can send it straight through WeChat. @@ -122,7 +122,7 @@ An item looks like this: ## Run your own copy -Most people do not need to deploy anything. The [search page](https://streamentry.github.io/HowToLiveBetter/en/) is already there, and if you want it offline, download the [single-file offline HTML](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html). Double-click it and it opens. +Most people do not need to deploy anything. The [search page](https://streamentry.github.io/HowToLiveBetter/) is already there, and if you want it offline, download the [single-file offline HTML](https://github.com/streamentry/HowToLiveBetter/releases/download/epub-latest/HowToLiveBetter.html). Double-click it and it opens. If you do want to run it on your own computer or your own server: @@ -283,7 +283,7 @@ The `index.html` in the repository root is the online search page. It filters it ## The text -The text is split by section into 34 files in [book/](book/). Click a section name in the contents above to go in. It is split because a single file has already gone past the 512 KB limit GitHub allows when rendering Markdown, and the sections after that point stop showing. The [search page](https://streamentry.github.io/HowToLiveBetter/en/) reads these files together, and the way it works has not changed. +The text is split by section into 34 files in [book/](book/). Click a section name in the contents above to go in. It is split because a single file has already gone past the 512 KB limit GitHub allows when rendering Markdown, and the sections after that point stop showing. The [search page](https://streamentry.github.io/HowToLiveBetter/) reads these files together, and the way it works has not changed. ## Licence diff --git "a/en/book/24-\347\234\213\347\227\205.md" "b/en/book/24-\347\234\213\347\227\205.md" new file mode 100644 index 00000000..4f8cdfa2 --- /dev/null +++ "b/en/book/24-\347\234\213\347\227\205.md" @@ -0,0 +1,121 @@ +[← 回总目录](../README.md) + +# 24. 看病：怎么少花钱少走弯路 + +口径：金钱与时间。这一节讲看病本身怎么走：先去哪家医院，怎么往上转，钱能少花在哪。另外讲急诊怎么排队，重伤治完之后要做哪些鉴定、办哪些证。能领什么待遇见第 7 节。慢性病长期怎么管见第 16 节。急症现场先做什么见第 13 节。 + +本节主要依据国务院办公厅 2026 年 4 月印发的《关于加快建设分级诊疗体系的若干措施》。分级诊疗就是按病情轻重决定先去哪一级医院。这份文件把「去基层」和「省钱」直接挂上了钩。但各地落地进度不同，正文写的是国家层面的规则。具体到你所在的统筹地区，要以当地医保和卫健部门的说法为准。统筹地区就是你的医保参保地，通常是一个市。 + +[← Back to the contents](../README.md) + +# 24. Medical care: spending less and going fewer rounds + +This section measures money and time. It covers how medical care itself works: which hospital to go to first, how to be referred upwards, and where the money can be saved. It also covers how emergency care is ordered, and which assessments and certificates you need after serious injuries are treated. For what you can claim, see section 7. For managing a chronic illness over the long term, see section 16. For what to do first at the scene of an emergency, see section 13. + +This section rests mainly on the Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》), issued in April 2026 by the General Office of the State Council (国务院办公厅). Tiered diagnosis and treatment means deciding which level of hospital to go to first according to how serious the illness is. That document ties going to the grassroots level directly to saving money. Local implementation is at different stages, and the text gives the national rules. For your own pooling region, follow what the local medical insurance and health authorities say. The pooling region is where your medical insurance is enrolled, usually one city. + +### 1. See common illnesses at the community health centre first, get referred upwards level by level, and the inpatient deductible carries on counting + +- Cost: No money. You have to make one trip to the community first. The hard part is resisting the urge to book a specialist straight away. +- In plain language: Go to the community first, and let the community refer you upwards to a large hospital for an admission. The part you pay yourself is then counted once. Go straight to a tertiary hospital for an admission and be referred back to the community for rehabilitation afterwards, and you may pay that part twice. Referred from a large hospital back into the community to carry on treatment, the same illness is not counted again during that period. +- Benefit: For insured patients referred upwards level by level through a primary healthcare provider, the inpatient deductible at the higher hospital can be counted continuously. The deductible is the part of each admission you pay yourself until medical insurance starts reimbursing. Continuous counting means the two hospitals together count it once, so you do not pay it again. Inpatients referred back down to the grassroots level by a higher hospital are not given a separate deductible within the same disease cycle +- Evidence grade: A +- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「统筹地区内经基层医疗卫生机构逐级转诊的参保患者，在上级医院的住院起付线可连续计算；由上级医院下转至基层医疗卫生机构的住院患者，同一疾病周期内不再另设住院起付线。」, that is, for insured patients in the pooling region referred upwards level by level through a primary healthcare provider, the inpatient deductible at the higher hospital may be counted continuously; inpatient patients referred down to a primary healthcare provider by a higher hospital are not given a separate inpatient deductible within the same disease cycle. +- Notes: If you go straight to a tertiary hospital for an admission and are then referred back to the grassroots level for rehabilitation, you may pay that part twice; going through a referral counts it once. Referral rules are set separately by each province, each city and each medical consortium, so ask the community before you go. + +### 2. Within one pooling region (where your medical insurance is enrolled), the lower the hospital level the higher the reimbursement rate, about 10 percentage points apart + +- Cost: No money. The hard part is resisting the urge to run to a large hospital. +- In plain language: For the same inpatient bill, the lower the hospital level, the more you get reimbursed. Localities have to spread the levels apart, by about ten percentage points for each step up. When something that could be finished at the community is done at a large hospital, you pay about a tenth more of the same bill yourself. New medical insurance funds also tilt towards the grassroots level. +- Benefit: The document requires localities to spread out the inpatient reimbursement level across healthcare providers of different grades, in principle by about 10 percentage points per level. On that gap, every step up the hospital ladder on the same inpatient bill costs you about a tenth more of your own money. The medical insurance funds added each year also tilt towards the grassroots level +- Evidence grade: A +- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「因地制宜适当拉开参保人员在不同等级医疗卫生机构的住院报销水平，原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距。」, that is, according to local conditions, suitably widen the inpatient reimbursement level for insured patients across healthcare providers of different grades, in principle leaving a gap of about 10 percentage points between each level within the pooling region. 「具备条件的地方，可按分级诊疗导向酌情拉开参保人员在不同等级医疗卫生机构的门诊报销水平。」, that is, localities in a position to do so may, in line with the tiered diagnosis and treatment policy, widen the outpatient reimbursement level across different grades as they see fit. +- Notes: For the same inpatient bill, the difference between what you pay at the grassroots level and at a tertiary hospital can run to several thousand yuan. Do not go up for something the community can finish. Not every illness belongs at the community health centre. + +### 3. If you need a large hospital, go through a grassroots referral or the hospital's referral centre, and do not buy appointments from scalpers + +- Cost: No money. You have one more referral procedure to get through. +- In plain language: Large hospitals must set aside a share of their appointments and beds for people referred up from the grassroots level. Hospitals must also set up a referral centre, or name one fixed department to run referrals, with full coverage by 2027. Failing to get an appointment does not mean there is no route in. A scalper's appointment is not only expensive; it may also land you in a department that does not match your illness. +- Benefit: The lead hospital of a tightly integrated medical consortium must reserve a set share of appointment slots and beds for the grassroots level and see referred patients promptly. A tightly integrated medical consortium is a group of hospitals and community health institutions tied together to work as one. So a referral up from the community has its own appointments and beds, instead of queueing with the people fighting for appointments in the lobby. Healthcare institutions must set up a referral centre or name a fixed department, with full coverage achieved by 2027 +- Evidence grade: A +- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「牵头医院要为基层医疗卫生机构预留一定比例的号源和床位，并及时接诊经基层转诊患者。」, that is, the lead hospital must reserve a set share of appointment slots and beds for primary healthcare providers and see patients referred up from them promptly. 「医疗机构要强化转诊服务统一管理，设立转诊中心或指定固定部门承担患者转诊服务工作，到2027年实现全覆盖。」, that is, healthcare institutions must strengthen unified management of referral services, set up a referral centre or name a fixed department to take on patient referral work, reaching full coverage by 2027. +- Notes: The reserved appointments are for referred patients only, so if you cannot get one, go through a referral. A scalper is not only expensive; you may also end up in a department that does not match. If you cannot find the referral centre, ring the hospital's service line and ask. + +### 4. Before going out of province for treatment, ask at home first; in principle the need is assessed by a deputy chief physician or above + +- Cost: No money. One more question before you set off. +- In plain language: Whether you should go elsewhere for treatment is, in principle, decided by a deputy chief physician or above at a secondary or tertiary hospital. How much you get back for temporary treatment away from your enrolled area differs from a hospital of the same level inside it, and the local rules set the figure. Asking once at home before you leave beats arriving somewhere else and finding out nothing can be claimed back. +- Benefit: The need for treatment across pooling regions and across provinces is, in principle, assessed by staff at deputy chief physician level or above at secondary and tertiary hospitals. Going across pooling regions means leaving the area where your medical insurance is enrolled. Whether you travel for treatment is decided by a doctor of sufficient rank. The reimbursement level for people going out of province temporarily for treatment keeps a reasonable difference from the level at the same grade of provider inside the enrolled area. The same illness treated elsewhere gets less money back than it does at home +- Evidence grade: A +- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「原则上由二、三级医院副主任医师及以上职称人员评估患者跨统筹地区、跨省异地就医的必要性。」, that is, in principle the need for a patient to receive treatment across pooling regions and across provinces is assessed by staff at deputy chief physician level or above at secondary and tertiary hospitals. 「进一步强化异地就医直接结算管理服务，跨省临时外出就医人员报销水平与参保地同级别医疗机构报销水平保持合理差异。」, that is, further strengthen direct settlement of treatment away from the enrolled area, keeping the reimbursement level for people going out of province temporarily for treatment at a reasonable difference from the level at providers of the same grade inside the enrolled area. +- Notes: Temporary treatment away from home and long-term residence away from home are two different registrations, and they reimburse at different levels. For how to register and which illnesses can be settled directly, see section 16. Do not travel first and think about reimbursement afterwards. + +### 5. Stop using a tertiary hospital as a prescription shop; send stable chronic illness back to the grassroots level + +- Cost: No money. The hard part is dropping the habit of trusting only big hospitals. +- In plain language: From now on tertiary hospitals mainly deal with critical illnesses and difficult cases. Routine follow-ups for common illnesses, and ordinary outpatient clinics for stable chronic illness, will be cut back little by little. Ordinary clinics run by the same specialists move to the community. If you keep using a big hospital as a prescription shop, getting an appointment there will get harder and harder. +- Benefit: Tertiary hospitals must focus on critical illnesses and difficult and complex diseases, strengthen referral, joint consultation and inpatient services, and gradually cut back, as they see fit, ordinary outpatient clinics such as routine follow-ups for common illnesses and diagnosed, stable chronic illness. Appointments for that kind of ordinary clinic will be withdrawn little by little. Higher hospitals will extend the ordinary outpatient clinics of their specialist teams down to the grassroots level, so the same specialists hold their ordinary clinics in the community +- Evidence grade: A +- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「三级医院要聚焦急危重症和疑难复杂疾病，加强转诊会诊和住院服务，逐步酌减常见病复诊和诊断明确、病情稳定的慢性病等普通门诊。」, that is, tertiary hospitals must focus on critical illnesses and difficult and complex diseases, strengthen referral, joint consultation and inpatient services, and gradually reduce, as they see fit, ordinary outpatient clinics such as routine follow-ups for common illnesses and for diagnosed, stable chronic illness. 「紧密型医联体内上级医院要在基层医疗卫生机构开设高血压、糖尿病、慢性阻塞性肺疾病等常见病、慢性病门诊，提供诊疗和健康管理服务，并将专家团队普通门诊向基层医疗卫生机构延伸。」, that is, inside a tightly integrated medical consortium, higher hospitals must open clinics for common illnesses and chronic illness such as hypertension, diabetes and chronic obstructive pulmonary disease at primary healthcare providers, supply diagnosis, treatment and health management there, and extend the ordinary outpatient clinics of their specialist teams down to the grassroots level. +- Notes: This is the direction of policy, not a suggestion. Ordinary clinics at big hospitals will get harder to book over time. Stable chronic illness can get 12 weeks of medicine at once at the grassroots level; see section 16. + +### 6. After every visit, keep your own copy of the records, the test reports and the images + +- Cost: No money. A hospital may charge a copying fee under the rules, and the rate has to be published. The hard part is remembering to go and copy them after every visit. +- In plain language: Reading and copying your own medical records is a right the law gives you. Outpatient records, inpatient notes, lab sheets, images, operation and anaesthesia records, pathology material and itemised bills can all be copied. The hospital runs a dedicated window for the requests and may charge only a copying fee. Changing hospital, seeing another doctor, claiming on private insurance or fighting a dispute all rest on this pile of paper. +- Benefit: The regulation makes reading and copying medical records an explicit patient right. The scope covers outpatient records, inpatient notes, lab sheets, medical images, operation and anaesthesia records, pathology material, medical costs and all the other material. Changing hospital, seeing a second doctor, claiming on private insurance or fighting a dispute all rest on these +- Evidence grade: A +- Sources: State Council (国务院) (2018). Regulations on the Prevention and Handling of Medical Disputes (《医疗纠纷预防和处理条例》) (State Council Order No. 701), Article sixteen: 「患者有权查阅、复制其门诊病历、住院志、体温单、医嘱单、化验单（检验报告）、医学影像检查资料、特殊检查同意书、手术同意书、手术及麻醉记录、病理资料、护理记录、医疗费用以及国务院卫生主管部门规定的其他属于病历的全部资料。」, that is, patients have the right to read and copy their outpatient records, inpatient notes, temperature charts, medical orders, lab sheets (test reports), medical imaging materials, consent forms for special examinations, consent forms for surgery, operation and anaesthesia records, pathology material, nursing records, medical costs and all other material belonging to the records as required by the health authority of the State Council. 「患者要求复制病历资料的，医疗机构应当提供复制服务，并在复制的病历资料上加盖证明印记……可以收取工本费，收费标准应当公开。」, that is, where a patient asks for the records to be copied, the healthcare institution must provide the copying service and stamp a certifying mark on the copy; it may charge a copying fee, and the rate must be published. 「患者死亡的，其近亲属可以依照本条例的规定，查阅、复制病历资料。」, that is, where the patient has died, close relatives may read and copy the records under these Regulations. Article fifteen: 「任何单位和个人不得篡改、伪造、隐匿、毁灭或者抢夺病历资料。」, that is, no unit and no individual may alter, forge, conceal, destroy or seize medical records. ; National Health and Family Planning Commission and National Administration of Traditional Chinese Medicine (国家卫生计生委、国家中医药管理局) (2013). Administrative Provisions on Medical Records in Healthcare Institutions (2013 edition) (《医疗机构病历管理规定（2013 年版）》): healthcare institutions 「应当指定部门或者专（兼）职人员负责受理复制病历资料的申请」, that is, must designate a department, or a member of staff full-time or part-time, to take the applications to copy records, and the copy is made with the applicant present and stamped with a certifying mark. +- Notes: There is a dedicated window for copying, so you do not have to plead with a doctor. The list of what you can ask for is in Article sixteen of these Regulations, item by item. First the outpatient records, inpatient notes, temperature charts, medical orders, lab sheets and imaging. Then the operation and anaesthesia records, pathology material, nursing records and the itemised bill. Put what you get into one folder in date order. The most important items inside are the discharge summary, the operation record, the pathology report and the key images. Keep the readings for a chronic illness in a single notebook; see section 16. + +### 7. If you doubt the diagnosis or treatment, ask on the spot for the records to be sealed, with both sides present, a list drawn up and one copy each + +- Cost: No money. Ask the hospital to seal the records and wait for both sides to turn up and complete the formalities. The hard part is saying that sentence in front of a doctor. +- In plain language: If you have doubts about the treatment, you can ask for the records to be sealed. Sealing requires both the patient side and the doctor side to be present, and a list to be drawn up, signed and stamped, with one copy each. Where the records are not finished, seal the part already completed. If you suspect an infusion, a blood transfusion, an injection or a drug caused the harm, the bag of fluid, the vial and the giving set have to be sealed as well. +- Benefit: The Regulations provide that 「发生医疗纠纷需要封存、启封病历资料的，应当在医患双方在场的情况下进行」, that is, where records must be sealed or unsealed because a medical dispute has arisen, this must be done with both the patient side and the doctor side present. What is sealed may be the original or a copy, and the healthcare institution keeps it. Where the records are not complete and sealing is required, the completed part is sealed first and the rest sealed once it is finished. The healthcare institution must draw up a list of the sealed records, signed or stamped by both sides, with one copy each. Where an infusion, a blood transfusion, an injection or a drug is suspected of having caused harm, both sides must jointly seal and unseal the physical items at the scene. Where testing is needed, the two sides must jointly appoint an institution with testing qualifications. Where they cannot agree on a joint appointment, the health authority of the county-level people's government where the institution is located designates one. Where a transfusion is suspected of having caused harm and the blood has to be sealed, the institution must notify the blood station to send someone to attend. There is also a hard limit: 「任何单位和个人不得篡改、伪造、隐匿、毁灭或者抢夺病历资料。」, that is, no unit and no individual may alter, forge, conceal, destroy or seize medical records. Once the dispute is resolved after sealing, or 3 years pass with no further request to resolve it, the institution may unseal the records itself (nationwide, in force from 2018-10-1) +- Evidence grade: A +- Sources: State Council (国务院) (2018). Regulations on the Prevention and Handling of Medical Disputes (《医疗纠纷预防和处理条例》) (State Council Order No. 701, Articles fifteen, twenty-four and twenty-five). +- Notes: Sealing and copying are two different things. Copying is what you should do as a matter of course; see item 6 in this section (keeping your own copy of the records and images). Sealing fixes the evidence when a dispute arises, and it guards against entries written or changed afterwards. If you suspect an infusion, a blood transfusion or a drug caused the harm, seal the bag of fluid, the vial and the giving set along with the records, and do not let anyone carry them off and bin them. Healthcare institutions must publish in a prominent place the routes, the procedures and the contact details for resolving medical disputes, and must set up a single complaints department. When something happens, go to that department rather than arguing with the doctor on duty. +### 8. For serious injuries go straight to the emergency triage desk, not to the registration queue + +- Cost: No money. A few more steps inside the door to find the triage desk. Say the most dangerous symptom in one sentence. +- In plain language: Emergency care is not ordered by who arrived first. Patients are sorted into four levels by how serious they are, and the sorting happens at the triage desk at the emergency entrance, not at the registration window. Skip the triage and you have no level in the system, however badly you are hurt, so you move along the queue at the speed of the queue. On arrival, report the most dangerous thing first: heavy bleeding, breathlessness, loss of consciousness, something embedded in the body, crushing pain in the chest. +- Benefit: The 2024 edition of the emergency medicine quality control indicators from the National Health Commission requires emergency triage to 「按病情危重程度将患者分为四级」, that is, sort patients into four levels by how critical they are. 「I 级为急危患者，需要立即得到救治」, that is, level I is for critical patients who need treatment immediately. 「II 级为急重患者，往往评估与救治同时进行」, that is, level II is for seriously ill patients, where assessment and treatment often go together. 「III 级为急症患者，需要短时间内得到诊治」, that is, level III is for acute patients who need diagnosis and treatment within a short time. 「IV 级为非急症患者或亚急症患者，在普通诊疗区顺序就诊」, that is, level IV is for non-acute or sub-acute patients, seen in order in the general treatment area. Triage means a member of staff assesses how serious you are once you come into the emergency department, and sets the order of treatment. Beijing's implementing standard sets out the response time for each level. Level 1: 「立即进行评估和救治，安排患者进入复苏室或抢救室」, that is, assess and treat at once, and move the patient into the resuscitation or resuscitation room. Level 2: 「立即监护生命体征，10 分钟内得到救治，安排患者进入抢救室」, that is, start monitoring vital signs at once, treat within 10 minutes, and move the patient into the resuscitation room. Level 3: 「先于 4 级患者优先诊治，安排患者在普通诊疗区候诊；若候诊时间大于 30 分钟，需再次评估」, that is, see the patient in preference to those at level 4, and have them wait in the general treatment area; if the wait exceeds 30 minutes, assess again. Level 4: 「顺序就诊，除非病情变化，否则候诊时间较长；若候诊时间大于 4 小时，可再次评估」, that is, seen in order, with a longer wait unless the condition changes; if the wait exceeds 4 hours, assess again. Triage has to 「坚持急危重症优先就诊原则」, that is, keep to the principle that critical cases are seen first, and a triage area must be set up in a prominent position at the entrance to the emergency department (the four levels are the national standard, the response times are Beijing's) +- Evidence grade: A +- Sources: General Office of the National Health Commission (国家卫生健康委办公厅) (2024). Notice on Issuing the Medical Quality Control Indicators for Six Specialties including Emergency Medicine (2024 edition) (《关于印发急诊医学等6个专业医疗质量控制指标（2024年版）的通知》(国卫办医政函〔2024〕150号)), Annex 1, indicator four, 「急诊分级分诊执行率」, that is, the rate at which emergency graded triage is carried out. ; Beijing Municipal Health Commission (北京市卫生健康委员会) (2019). Beijing Plan for Strengthening Emergency Pre-triage and Grading, and its Annex, the Beijing Standard for Emergency Pre-triage and Grading in Hospitals (Trial) (《北京市加强急诊预检分诊分级工作方案》and 《北京市医院急诊预检分诊分级标准（试行）》): 「坚持急危重症优先就诊原则。按照患者病情的严重程度，遵循从重到轻、从病情迅速变化到相对稳定的原则，合理安排患者就诊顺序。」, that is, keep to the principle that critical cases are seen first; according to how serious the patient's condition is, follow the order from severe to mild and from rapidly changing to relatively stable, and arrange the order of treatment accordingly. 「在急诊科入口的显著位置，设置急诊预检分诊区域」, that is, set up an emergency triage area in a prominent position at the entrance to the emergency department. +- Notes: The four levels are the same nationwide. How long each level waits before being seen is set by each province and city, and the Beijing text is a version you can check word for word. The level turns on readings such as pulse, blood pressure and blood oxygen, and on the symptoms, not on how much pain you feel. Level 1 has hard criteria, and these are some of them. A heart rate above 180 per minute, or below 40 per minute. A systolic blood pressure below 70 mmHg, which is the upper number in a blood pressure reading. Blood oxygen below 80%. Shock, a confirmed heart attack, or an acute impairment of consciousness. If you think the triage has put you too low, add the most dangerous symptom and say it again. If your condition changes while you wait, go straight back to the triage desk to be assessed again. When things are genuinely critical, ringing 120 is faster than going yourself: the paramedic assesses you against the same standard first and tells the hospital to prepare (see section 13). + +### 9. No money, no papers, and nobody who can say who you are: emergency care still has to come first + +- Cost: No money. +- In plain language: The state has an emergency medical assistance fund that pays emergency treatment charges for two kinds of people: those whose identity cannot be established, and those who cannot pay. A hospital may not refuse or delay resuscitation, and an ambulance service may not decline to send an ambulance over money. So do not sit at home and endure it because you have no money, and do not be afraid to call an ambulance and owe the bill. This fund covers emergency treatment only. An admission afterwards still goes through medical insurance and medical assistance. +- Benefit: The emergency medical assistance scheme covers 「在中国境内发生急重危伤病、需要急救但身份不明确或无力支付相应费用的患者」, that is, patients who suffer a serious or critical injury or illness in China, need emergency treatment, and either cannot be identified or cannot pay. The fund pays two kinds of charge: 「无法查明身份患者所发生的急救费用」, that is, the emergency treatment charges of patients whose identity cannot be established, and 「身份明确但无力缴费的患者所拖欠的急救费用」, that is, the emergency treatment charges owed by patients whose identity is known but who cannot pay. The document also states that 「各级各类医疗机构及其工作人员必须及时、有效地对急重危伤患者施救，不得以任何理由拒绝、推诿或拖延救治」, that is, healthcare institutions at every level and their staff must treat patients with serious or critical injuries promptly and effectively, and may not refuse, shirk or delay treatment on any grounds. Pre-hospital emergency care has its own rules: the national pre-hospital emergency number is 「120」. An emergency centre (station) 「不得因指挥调度原因拒绝、推诿或者延误院前医疗急救服务」, that is, may not refuse, shirk or delay pre-hospital emergency care on grounds of dispatch or scheduling, and also 「不得因费用问题拒绝或者延误院前医疗急救服务」, that is, may not refuse or delay pre-hospital emergency care over money. Transfer follows 「就近、就急、满足专业需要、兼顾患者意愿的原则」, that is, the principles of going to the nearest place, going to the most urgent place, meeting professional need and taking the patient's wishes into account (nationwide, since 2013) +- Evidence grade: A +- Sources: General Office of the State Council (国务院办公厅) (2013). Guiding Opinions on Establishing the Emergency Medical Assistance Scheme (《关于建立疾病应急救助制度的指导意见》(国办发〔2013〕15号)). ; National Health and Family Planning Commission (国家卫生和计划生育委员会) (2014). Measures for the Administration of Pre-hospital Medical Emergency Care (《院前医疗急救管理办法》(委令第3号), Articles thirteen, twenty-two, twenty-three and twenty-five). +- Notes: The fund pays only the emergency treatment part, and it is the hospital that applies to the fund afterwards, so there is nothing for you to collect at a window. If you have medical insurance, use it first as usual. The application process and the ceiling differ by province. In the transfer principles, the nearest place and the most urgent place come before the patient's wishes, so the ambulance does not necessarily take you to the hospital you named. When things are critical, the nearest place that can treat you is what counts. Where what you pay yourself afterwards is still too heavy, use medical assistance; see item 10 in section 7 (medical assistance). + +### 10. Get the disability assessment after treatment has finished, because doing it early gets you a lower grade + +- Cost: The assessment fee runs from a few hundred yuan to a few thousand yuan, and the party that applies usually pays it up front. You have to wait until treatment has finished or the clinical outcome has settled, usually three to six months after the injury and sometimes longer. The hard part is resisting the urge to settle it while the injury is still healing. +- In plain language: How much you get paid depends mainly on the grade, and the grade depends on how much functional impairment is left once treatment has finished. The standard says the assessment must wait until treatment has finished or the clinical outcome has settled. Go for an assessment while the bone has not healed or the plate is still in place and you are assessed as you are that day, which means lowering your own grade. An ordinary fracture that has healed without affecting function does not reach any grade. +- Benefit: The Standard on the Grading of Disability from Human Body Injury states that the timing of the assessment 「应在原发性损伤及其与之确有关联的并发症治疗终结或者临床治疗效果稳定后进行鉴定」, that is, the assessment should be carried out after treatment of the primary injury and of complications clearly related to it has finished, or after the clinical treatment outcome has settled. The standard 「将人体损伤致残程度划分为 10 个等级，从一级（人体致残率 100%）到十级（人体致残率 10%），每级致残率相差 10%」, that is, divides the degree of disability from human body injury into 10 grades, from grade one (a disability rate of 100%) to grade ten (a disability rate of 10%), with 10% between each grade. It turns on 「人体组织器官结构破坏、功能障碍及其对医疗、护理的依赖程度，适当考虑由于残疾引起的社会交往和心理因素影响」, that is, structural damage to tissues and organs, functional impairment, and the degree of dependence on medical and nursing care, with due consideration of the effect of the disability on social interaction and psychological factors. The fracture-related clauses among the ten grades include these. First: 「一椎体压缩性骨折（压缩程度达 1/3）或者粉碎性骨折」, that is, compression fracture of one vertebral body with compression of 1/3 or more, or a comminuted fracture. Second: 「骨盆两处以上骨折或者粉碎性骨折，畸形愈合」, that is, more than two fractures of the pelvis or a comminuted fracture, with malunion. Third: 「四肢任一大关节（踝关节除外）功能丧失 25% 以上」, that is, loss of more than 25% of the function of any major joint of the limbs, excluding the ankle. Disability at the same site and of the same kind may not be counted twice. Where there is more than one disability, the grades must be stated separately (nationwide, in force from 2017-1-1) +- Evidence grade: A +- Sources: Supreme People's Court, Supreme People's Procuratorate, Ministry of Public Security, Ministry of State Security and Ministry of Justice (最高人民法院、最高人民检察院、公安部、国家安全部、司法部) (2016). Announcement on the Issue of the Standard on the Grading of Disability from Human Body Injury (2016 4 18 published, 2017 1 1 in force), clauses 4.2, 4.4, 4.5, 5.10.6 and 6.2 of the standard. +- Notes: There are three different disability assessments, and they do not substitute for one another. Assault and other personal injury, and traffic accidents, go through the judicial Standard on the Grading of Disability from Human Body Injury, which is used to work out the disability compensation (see section 9). A work injury goes through the assessment of capacity for work, which uses a different set of ten grades and governs the benefits for a work injury (see section 19). To claim the disability benefits available to disabled people you need a disability certificate as well (see item 11, disability certificate). Have the records, the operation record and the follow-up images ready before the assessment (see item 6, keeping your own copy of the records and images); missing images are the commonest reason a grade comes out low. Where this injury is mixed up with a long-standing condition, the assessment opinion also has to say how much of the resulting disability this injury accounts for. That line has a direct effect on how much you are paid. + +### 11. If treatment leaves real functional impairment, apply to the county-level disabled persons' federation at your registered place of residence for a disability certificate + +- Cost: The certificate itself carries no administrative charge. The assessment fee and the photo fee are in principle yours to pay, and you can apply for a reduction if you have difficulty. Take your identity card, your household register and 3 passport-size photos on a white background to the county-level disabled persons' federation at your registered place of residence. Once the assessment passes, the result is posted in the village or community for 5 working days, and the federation finishes its review within 10 working days. The hard part is making these trips and waiting out that time. +- In plain language: A disability certificate is the threshold for claiming disabled persons' benefits, not a certificate of injury. There are seven categories: vision, hearing, speech, limb, intellectual, psychiatric and multiple disabilities, each graded one to four. The process is to apply, have the application accepted, be assessed at a designated institution, be posted, be approved after review, and be issued, and it is valid for ten years and replaced free of charge on expiry. If you do not pass you do not pass, and this is not something that can be arranged through someone you know. +- Benefit: The disability certificate 「是认定残疾人及其残疾类别、残疾等级的合法凭证，是残疾人依法享有国家和地方政府优惠政策的重要依据」, that is, is the lawful evidence for establishing that a person is disabled and for the category and grade of the disability, and the main basis on which disabled people lawfully enjoy the preferential policies of the state and local governments. The standard for the assessment is the national standard Classification and Grading of Disabilities (《残疾人残疾分类和分级》, GB/T 26341-2010). The certificate rests on voluntary application and local administration. People with vision, hearing, speech, limb, intellectual, psychiatric or multiple disabilities may all apply, and the grade runs from one to four. The county-level federation takes the application and issues the certificate, on the basis of the assessment conclusion reached by a hospital or specialist institution designated by the province. Where the conclusion meets the standard, it is posted in the village or community for five working days, and the county-level federation 「在十个工作日内审核完毕」, that is, completes its review within ten working days. 「办理残疾人证不收取工本费」, that is, no administrative charge is collected for a disability certificate. The assessment and photo costs are 「原则上由申请人个人自理；有条件的地方可由当地财政予以补贴，对特殊困难的申请人应协调有关部门予以减免」, that is, in principle paid by the applicant; where local conditions allow they may be subsidised by the local finance, and for applicants in special difficulty the relevant departments must be coordinated to reduce or waive them. The certificate is valid for ten years and is replaced free of charge on expiry. Where the category or grade changes, you may apply to be assessed again. Where you object to the conclusion, you may apply to the prefecture-level federation within ten working days for a fresh assessment. Where the objection stands, the provincial expert committee on disability assessment gives the final conclusion (nationwide, in force from 2018-1-1) +- Evidence grade: A +- Sources: China Disabled Persons' Federation and National Health and Family Planning Commission (中国残疾人联合会、国家卫生和计划生育委员会) (2017). Measures for the Administration of the Disability Certificate of the People's Republic of China (2017 5 27 published, 2018 1 1 in force; Articles two, three, five, seven, nine, seventeen, nineteen, twenty-two and twenty-five). +- Notes: Work out whether it is worth it before you apply. The certificate does not hand you money by itself. What it connects you to is the two subsidies for disabled people: disabled people in means-tested income support households get the living subsidy, and those at grades one and two who need long-term care get the care subsidy; see item 8 in section 7 (the two subsidies). The certificate also connects you to local concessions on rehabilitation, assistive devices, tax, employment and transport. These standards differ a great deal by province, so check the local federation's website first and then decide whether the trip is worth it. It is one of three separate things and does not replace the other two: the judicial injury grade and the assessment of capacity for work (see item 10, getting the disability assessment after treatment has finished). The certificate is for the holder only and may not be lent to anyone else. A certificate whose condition no longer meets the standard is cancelled, and you may not apply again within one year of cancellation. Multiple disabilities are graded by the most severe category among them. +### 12. Thank a doctor who saved your life with a letter of thanks, a banner or a satisfaction rating, not with cash: the code bans the property, not the gratitude + +- Cost: No money. Write one letter of thanks, or fill in one rating form on the satisfaction platform, ten minutes or so. +- In plain language: Money handed over in advance buys no priority. Appointment slots, beds, scarce drugs and surgery slots are all barred from being traded for a benefit. Afterwards you are free to thank the doctor who saved your life, but do not send cigarettes, alcohol or a red envelope, because taking it is a breach of the rules. There are three ways of saying thanks that really reach the person: fill in a satisfaction rating properly, write a letter of thanks naming the doctor or give a banner, and keep the follow-up appointments. +- Benefit: Article seven of the Nine Standards for Clean Practice by Staff of Healthcare Institutions (《医疗机构工作人员廉洁从业九项准则》(国卫医发〔2021〕41号)): 「维护诊疗秩序，不破坏就医公平。坚持平等原则，共建公平就医环境。严禁利用号源、床源、紧缺药品耗材等医疗资源或者检查、手术等诊疗安排收受好处、损公肥私」, that is, maintain the order of treatment and do not damage fairness in access to care; uphold the principle of equality and build a fair environment for access to care; it is strictly forbidden to take any benefit by using medical resources such as appointment slots, bed slots, scarce drugs and consumables, or arrangements for examinations and surgery, and to seek private gain at public expense. Article eight: 「共建和谐关系，不收受患方『红包』。恪守医德、严格自律」, that is, build harmonious relations and do not accept 「red envelopes」 from patients; keep to professional ethics and strict self-discipline. The same article adds: 「严禁索取或者收受患者及其亲友的礼品、礼金、消费卡和有价证券、股权、其他金融产品等财物」, that is, it is strictly forbidden to ask for or accept from patients and their relatives gifts, cash, stored-value cards, securities, equity, other financial products or other property. Meals and entertainment paid for by the patient side are covered too: 「严禁参加其安排、组织或者支付费用的宴请或者旅游、健身、娱乐等活动安排」, that is, it is strictly forbidden to take part in banquets, travel, fitness or entertainment arranged, organised or paid for by the patient side. The document also requires healthcare institutions to carry their own responsibility, carry out inspections and 「坚决查处违规行为」, that is, investigate and punish violations firmly. Satisfaction ratings were written into the performance assessment of tertiary public hospitals. The basis is the Opinions of the General Office of the State Council on Strengthening the Performance Assessment of Tertiary Public Hospitals (《国务院办公厅关于加强三级公立医院绩效考核工作的意见》(国办发〔2019〕4号)). It states: 「三级公立医院绩效考核指标体系由医疗质量、运营效率、持续发展、满意度评价等4个方面的指标构成」, that is, the index system for assessing tertiary public hospitals consists of indices in four areas: quality of care, operational efficiency, sustainable development and satisfaction ratings. It also states: 「医院满意度由患者满意度和医务人员满意度两部分组成……」, that is, hospital satisfaction consists of two parts, patient satisfaction and staff satisfaction. It goes on to say that one should 「通过门诊患者、住院患者和医务人员满意度评价，衡量患者获得感及医务人员积极性」, that is, measure through the satisfaction ratings of outpatients, inpatients and staff how much patients feel they have gained and how motivated the staff are. There is also a single national platform for these ratings: 「国家建立公立医院满意度管理制度……」, that is, the state establishes a satisfaction rating system for public hospitals. The timetable is: 「2019年3月底前，全国三级公立医院全部纳入国家卫生健康委满意度调查平台。各地要应用国家卫生健康委满意度调查平台，将调查结果纳入三级公立医院绩效考核」, that is, by the end of March 2019, all tertiary public hospitals nationwide are brought into the satisfaction survey platform of the National Health Commission; localities are to use that platform and bring the results into the performance assessment of tertiary public hospitals. Hospitals must also handle complaints. The basis is Article two of the Measures for the Administration of Complaints by Healthcare Institutions (《医疗机构投诉管理办法》(国家卫生健康委令第3号), in force 2019). Handling complaints means that a patient, with regard to medical services, administration, quality and safety of care and the like, 「提出意见、建议或者投诉请求」, that is, puts forward an opinion, a suggestion or a complaint. The hospital is responsible for investigating, dealing with and replying. 「二级以上医疗机构应当设置医患关系办公室或者指定部门（以下统称投诉管理部门）统一承担投诉管理工作」, that is, healthcare institutions at secondary level and above must set up a doctor-patient relations office or designate a department, collectively called the complaints department, to take on the handling of complaints in one place. Its duties include 「统一受理投诉，调查、核实投诉事项，提出处理意见，及时答复患者」, that is, receiving complaints in one place, investigating and verifying the matters complained of, putting forward a view on how to handle them, and replying to the patient in good time. Healthcare institutions 「应当鼓励工作人员主动收集患者对医疗服务、医疗质量安全等方面的意见和建议」, that is, must encourage staff to collect patients' opinions and suggestions on medical services, quality and safety of care and the like (nationwide) +- Evidence grade: A +- Sources: National Health Commission, National Healthcare Security Administration and National Administration of Traditional Chinese Medicine (国家卫生健康委、国家医保局、国家中医药局) (2021). Notice on Issuing the Nine Standards for Clean Practice by Staff of Healthcare Institutions (《关于印发医疗机构工作人员廉洁从业九项准则的通知》(国卫医发〔2021〕41号)). State Council Gazette ; General Office of the State Council (国务院办公厅) (2019). Opinions on Strengthening the Performance Assessment of Tertiary Public Hospitals (《关于加强三级公立医院绩效考核工作的意见》(国办发〔2019〕4号)). ; National Health Commission (国家卫生健康委) (2019). Measures for the Administration of Complaints by Healthcare Institutions (《医疗机构投诉管理办法》(委令第3号)). +- Notes: The person who benefits is yourself, and so is the person who saved you. Wanting to say thank you after a night of resuscitation is ordinary, and you may say it; only the form has to change. A red envelope or a case of cigarettes shifts the risk onto the other person, while a letter of thanks, a banner and a satisfaction rating are the forms the rules recognise. If you have already handed something over and the other person pushed it back, do not read that as thinking the amount was too small or as putting on airs, because they are required to refuse it. Keep two other things apart as well. Handing a doctor a red envelope is a matter of professional discipline and of the hospital. Handing property to the people handling or enforcing a case is the crime of bribery under the Criminal Law, and the two are not on the same scale; see item 40 in section 8 for that one. Hospitals widely accept a letter of thanks or a banner, but the claim that it 「会进个人和科室记录」, that is, goes into the individual's and the department's record, is based on experience and is not in any document. What has a document behind it is the rule bringing satisfaction survey results into the performance assessment of tertiary public hospitals. If you have doubts about the diagnosis or the treatment itself, ask on the spot for the records to be sealed (see item 7 in this section, sealing the records), and keep your own copy of the records and images (item 6, keeping your own copy of the records and images). \ No newline at end of file diff --git "a/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" "b/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" new file mode 100644 index 00000000..01e8670d --- /dev/null +++ "b/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" @@ -0,0 +1,104 @@ +[← 回总目录](../README.md) + +# 25. 人走了以后要办什么 + +口径：金钱。本节只讲手续怎么办、钱怎么拿，包括人刚走时先做什么、哪些费用有清单可以对。还有哪些账户里的钱能取回来，哪些事近亲属有权出面办。前五条按事情发生的先后排，依次是现场、死亡证明、遗体和火化、对死因有异议怎么办、销户。后面几条按性价比排。老人生前该做的安排，比如意定监护、遗嘱、账户，见第 17 节。因工死亡的三笔待遇见第 19 节。 + +殡葬这部分依据新版《殡葬管理条例》，2026 年 3 月 30 日起施行。养老保险的丧葬补助金和遗属抚恤金给多少，由人力资源社会保障部门另外规定。这次没拿到能逐字核对的原文，所以正文不写具体金额。 + +[← Back to the contents](../README.md) + +# 25. What to do after someone dies + +This section measures money. It covers only the paperwork and the money: what to do first after someone dies, which charges you can check against a published list, which accounts still hold money you can get back, and which steps a close relative is allowed to handle in person. The first five entries follow the order in which things happen: the scene, the death certificate, the body and the cremation, what to do if you dispute the cause of death, and the cancellation of the household registration. The rest are ordered by value for money. The arrangements an older person should make while they are still able, such as voluntary guardianship, a will and accounts, are in section 17. The three payments for a death caused by work are in section 19. + +The funeral side follows the new Regulations on Funeral and Interment Administration (殡葬管理条例), in force from 2026-3-30. The amounts of the funeral subsidy and the survivor's pension under the old-age insurance are set separately by the human resources and social security departments. This time I could not get the original text to check word for word, so the items below give no figures. + +### 1. If someone dies at home or elsewhere, first tell an expected death from an unexpected one: an accident, an unknown cause, or a person living alone found late — call 110 at once, and do not touch the body or the scene + +- Cost: No money. One phone call is all it takes. +- In plain language: Where the death was an accident, where nobody knows how it happened, or where someone living alone was found only after a long time, only the police can issue the death certificate. Once the scene has been disturbed, the account cannot be set out and the certificate is hard to issue as well. Without that sheet of paper, the cremation, the cancellation of the household registration and the insurance claim all stall. If you cannot tell which kind it is, call 110 first. Once the police have been there and finished their inspection, they will still issue the certificate. +- Benefit: The Regulations on the Registration of Households (户口登记条例), Article eight, paragraph two: 「公民因意外事故致死或者死因不明，户主、发现人应当立即报告当地公安派出所或者乡、镇人民委员会」 — where a person dies in an accident or the cause of death is unknown, the head of household and the person who found the body must report it at once to the local police station or the township or town people's government. The Regulations on Funeral and Interment Administration (殡葬管理条例), Article twenty-three: 「自然人正常死亡的，医疗卫生机构应当及时出具死亡证明。涉及公安机关依法处置且未经医疗卫生机构救治的非正常死亡案（事）件的逝者，由公安机关出具死亡证明」 — for an ordinary death, the medical and health institution issues the death certificate in good time. Where the case involves a death that the police handle under the law and the person was not treated by a medical institution, the police issue the certificate. Only the police can issue the certificate for an unexpected death. If the scene has been disturbed, the account cannot be set out and the certificate is hard to issue. The cremation, the cancellation of the household registration and the insurance claim all stall on that one sheet of paper +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (1958). Regulations on the Registration of Households, 中华人民共和国户口登记条例（第八条）. (posted on the Guangdong Provincial Government portal); State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，自 2026 年 3 月 30 日起施行，第二十三条）. +- Notes: Someone bedridden for a long time who dies at home counts as an expected death. You can ask the institution that treated him while he was alive to issue the certificate, or the community health centre or the township health centre for the area you live in. Whoever treated him issues the death certificate, see item 2 in this section (the death certificate). If you cannot tell which kind it is, treat it as unexpected and call 110 first; once the police have been there and finished their inspection, they will still issue the certificate. If you tidy the scene up before calling, nobody can help you afterwards. Do not move the body yourself, do not wash and dress the body, and do not take away the medicine bottles or the notes. + +### 2. The death certificate is the key to everything that follows: whoever treated the person issues it, for an expected death at home you go to the community health service institution, and it is issued within one day + +- Cost: No money. +- In plain language: If the person died in hospital, ask the institution that treated them for the certificate. If they died at home in the ordinary way, ask the community health centre or the township health centre for the area you live in. When the cause of death is clear, the certificate is issued within one day. The cancellation of the household registration, the cremation, the withdrawal of the housing provident fund, the payments under social insurance, the insurance claim and the inheritance all need it. Get several copies as soon as you have it. +- Benefit: The Interim Standards for the Registration of Death Information (人口死亡信息登记管理规范（试行）) say that for a death inside a medical and health institution the certificate is 「负责救治的医疗卫生机构签发」 — issued by the institution that did the treating, so for a death in a hospital or a health centre you ask the one that treated the person. For an ordinary death at home or in another place it is 「由本辖区社区卫生服务机构或乡镇（街道）卫生院签发」 — issued by the community health service institution or the township (sub-district) health centre for the area. Where the police or a judicial authority judges the death to be unexpected, the police or judicial department handles it under the existing rules and procedures. A 2026 notice from five departments says that the death certificate 「纸质版与电子证照具有同等法律效力」 — the paper version and the electronic licence carry the same legal force, so the electronic one works exactly like the paper one. An ordinary death with a known cause must be issued within one day of the death. If the certificate is lost, you may apply once to the issuing institution for a replacement. If a basic detail is wrong, you may apply once, on the strength of your documents, for it to be issued again +- Evidence grade: A +- Sources: Office of the National Health and Family Planning Commission (国家卫生计生委办公厅) (2014). Interim Standards for the Registration of Death Information, 人口死亡信息登记管理规范（试行）（国卫办规划发〔2014〕68 号，第九条）. (forwarded by the Shenzhen Municipal Civil Affairs Bureau); National Health Commission (国家卫生健康委), Ministry of Public Security (公安部), Ministry of Civil Affairs (民政部), National Administration of Traditional Chinese Medicine (国家中医药局), National Administration of Disease Control and Prevention (国家疾控局) (2026). Notice on Strengthening the Registration of Medical Death Certificates of Residents and the Management of Electronic Licences, 关于加强居民死亡医学证明信息登记和电子证照管理工作的通知（国卫规划发〔2026〕5 号）. (forwarded by the Fujian Provincial Health Commission) +- Notes: The cancellation of the household registration, the cremation, the withdrawal of the housing provident fund, the payments under social insurance, the insurance claim and the inheritance all need it. Get it and make several copies at once, and ask for the electronic licence as well where one can be issued. You only get one chance of a replacement, so do not lose the original. If the name or the identity number is wrong, have it corrected on the spot, because finding it out when the inheritance is being processed means going round every counter again. + +### 3. Collecting, storing and cremating the body can only be done by a funeral home, a hospital mortuary does not hold funerals, and storage is normally no more than 3 days + +- Cost: No money. +- In plain language: Collecting the body, storing it, treating it, restoring its appearance and cremating it can only be done by a funeral home (殡仪馆). A hospital mortuary does not hold funerals; at most it lets you leave the body there for a short time. A funeral home normally keeps a body for no more than 3 days, and to keep it longer you have to go and extend it. Cremation needs the death certificate and the cremation confirmation form. The cremation certificate you are given afterwards is still needed for the inheritance and for the insurance claim. +- Benefit: The Regulations on Funeral and Interment Administration (殡葬管理条例), Article twenty-five: 「遗体接运、存放、防腐、整容、火化服务，由殡仪馆专门负责提供」 — the services of collecting, storing, treating, restoring and cremating the body are provided by the funeral home alone, and nobody else can do them. Article twenty-four: 「医疗卫生机构的太平间不得外包，不得开展殡仪服务，可以为遗体暂时停放提供便利服务」 — a hospital mortuary may not be outsourced and may not provide funeral services, and it may only make it convenient for a body to be kept there temporarily. Article twenty-seven: 「殡仪馆存放遗体一般不超过 3 日，需要延期存放的，丧属或者遗体移交方应当在殡仪馆办理延期存放手续」 — a funeral home normally keeps a body for no more than 3 days, and where longer storage is needed, the relatives or the party handing over the body must go to the funeral home and complete the extension formalities. Article twenty-eight: 「殡仪馆应当凭死亡证明和火化确认书火化遗体，并出具火化证明」 — the funeral home cremates the body on the strength of the death certificate and the cremation confirmation form, and issues a cremation certificate. You need both documents, and after the cremation you are given a cremation certificate. Article twenty-nine: 「在实行火葬的地区，遗体应当就地、就近在殡仪馆火化」 — in an area where cremation applies, the body is cremated at a funeral home in the place where the person died, or the nearest one +- Evidence grade: A +- Sources: State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，自 2026 年 3 月 30 日起施行，第八、九、二十四、二十五、二十七、二十八、二十九条）. +- Notes: A 「one-stop service」 that turns up after someone dies in hospital is usually not from the funeral home. The collection itself is a basic item and it has a published price, see item 6 in this section (the list of basic items). Keep the cremation certificate safe; you need it for the inheritance, the insurance and moving the ashes. Whether the area you live in applies cremation or burial is fixed by the provincial government. The funeral customs of ethnic minorities are protected by Article 9 of the regulations. Where a body that should be cremated is buried instead, or is buried and a grave built outside a cemetery or an eco-burial area, the civil affairs department orders it corrected within a set period (that is, put it back the way the rule says). + +### 4. If you dispute the cause of death, ask for the autopsy within 48 hours and have the medical records sealed at the same time + +- Cost: No money. The hard part is remembering to ask at that moment. +- In plain language: If you have a question about the cause of death, you have to ask for the autopsy within 48 hours of the death. Where the hospital can store the body frozen, the time can be extended to 7 days. An autopsy also needs the signature of a close relative. Without that signature, or if you let the time run out, the responsibility for not establishing the cause of death falls on whoever delayed. When you ask for the autopsy, ask at the same time that the medical records be sealed. Once the body is cremated, no autopsy is possible any more. +- Benefit: The time limit for the autopsy is in the Regulations on the Prevention and Handling of Medical Disputes (医疗纠纷预防和处理条例), Article twenty-six. The text is: 「患者死亡，医患双方对死因有异议的，应当在患者死亡后 48 小时内进行尸检；具备尸体冻存条件的，可以延长至 7 日」 — where the patient dies and the two parties disagree about the cause of death, the autopsy must be carried out within 48 hours of the death, and where the body can be kept frozen it may be extended to 7 days. So if you want the cause of death established, the autopsy has to happen within 48 hours of the death, and where the hospital has freezing conditions it goes to 7 days at most. The same article also says: 「尸检应当经死者近亲属同意并签字，拒绝签字的，视为死者近亲属不同意进行尸检」 — the autopsy requires the consent and the signature of a close relative of the dead person, and a refusal to sign counts as a refusal. If the family does not sign, it counts as a refusal. It also says: 「不同意或者拖延尸检，超过规定时间，影响对死因判定的，由不同意或者拖延的一方承担责任」 — whoever refuses or delays the autopsy past the time limit and thereby affects the finding on the cause of death bears the responsibility. Whoever refuses, and whoever delays past the deadline, is the party that bears the responsibility for a cause of death that cannot be established later. Article twenty-four: sealing and unsealing the medical records 「应当在医患双方在场的情况下进行」 — must be done with both parties present. You and the hospital both have to be there when the records are sealed and when the seal is broken +- Evidence grade: A +- Sources: State Council (国务院) (2018). Regulations on the Prevention and Handling of Medical Disputes, 医疗纠纷预防和处理条例（国务院令第 701 号，第二十四、二十六条）. +- Notes: Saying afterwards that something felt wrong from the start puts the responsibility on the party that delayed. Once the body is cremated there is no autopsy to be had, so this has to be settled before you get in touch with the funeral home. How the records are sealed is in section 24. + +### 5. Cancelling the household registration has a statutory deadline: in a city before the burial, in the countryside within one month + +- Cost: No money. Take the death certificate and the household register and go to the police station once. +- In plain language: In a city the registration is cancelled before the burial; in the countryside within one month. Take the death certificate and the household register to the police station. If you leave it undone, the transfer of inherited property, the payments you are entitled to and the repayment of the pension all get stuck on the household registration status. +- Benefit: The Regulations on the Registration of Households (户口登记条例), Article eight, paragraph one: 「公民死亡，城市在葬前，农村在一个月以内，由户主、亲属、抚养人或者邻居向户口登记机关申报死亡登记，注销户口。公民如果在暂住地死亡，由暂住地户口登记机关通知常住地户口登记机关注销户口」 — the meaning is: in a city it is done before the burial, in the countryside within one month. The head of household, a relative, a guardian or a neighbour may all go to the registration authority (the police station) to report the death and cancel the registration. Where a person dies while staying away from home, the police station where they were staying notify the police station of the registered residence, and the registration is cancelled there +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (1958). Regulations on the Registration of Households, 中华人民共和国户口登记条例（第八条）. (posted on the Guangdong Provincial Government portal) +- Notes: While the registration is still open, the transfer of inherited property, the payments you are entitled to and the repayment of the pension all stick, because every one of those counters looks at the registration status. If the pension keeps being paid, that is someone receiving it wrongly and it will have to be handed back eventually. While you are at the police station, ask as well: how the social insurance and the medical insurance payments are stopped, and how the funeral subsidy is claimed. Those payments are in item 9 in this section (the provident fund balance and the payments under social insurance). + +### 6. Funeral services are split into basic items and non-basic items, there is a list of the basic items, and the charges are set by law + +- Cost: No money. Spend a few minutes looking at the list before you set off. +- In plain language: Collecting the body, storing the body, the farewell ceremony, the cremation, the storage of the ashes and eco-burial are all on the national list of basic items. Ash-niche burial provided by a public institution is on the list as well. The price of these items is fixed by the rules. Outside the list you may not set up an item of your own and charge for it. Before you set off, ask which items count as basic items and what the local standard is, and only then talk about the rest. +- Benefit: Collecting the body, storing the body, saying farewell over the body, cremating the body, storing the ashes and eco-burial are all on the national list of basic items. Ash-niche burial provided by a funeral service institution run by the government is on the list as well. The charges for these items are set by law, the price is fixed by the rules and it is not for the funeral home to decide. The list of non-basic items is drawn up by the civil affairs department of a city or above (the civil affairs department at prefecture level and above) +- Evidence grade: A +- Sources: State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，2025 年 11 月 14 日国务院第 72 次常务会议修订通过，「自2026年3月30日起施行」）. 第五条：「殡葬服务分为基础项目和非基础项目，收费政策由国务院发展改革部门、财政部门会同国务院民政等部门制定。」「国家制定殡葬服务基础项目清单，将遗体接运、遗体存放、遗体告别、遗体火化、骨灰寄存、生态安葬以及政府举办的殡葬服务机构提供的骨灰格位安葬等纳入清单范围」；第五十条「殡葬服务实行清单化管理并动态调整，禁止在清单之外设立项目、收取费用。」 +- Notes: A province, an autonomous region or a municipality may add items on top of the national list. Ask first which items count as basic items and what the local charge is, and only then talk about the rest. The price of an item on the list is fixed by the rules; the items off the list are yours to choose. Tell the two kinds apart and only then do you have anything to say about the part you have been overcharged. There is nothing wrong with giving someone a decent send-off, and no one is telling you to make do cheaply; where the money goes is your own decision. + +### 7. If you are sold a bundle you did not ask for, charged above the posted price, or charged several times for one item split into several, write it down and complain + +- Cost: No money. The hard part is still standing your ground at that moment. +- In plain language: All of these are price violations the regulations name: not displaying a price list, adding money on top of the posted price, tying the sale of funeral goods to the service, and making you accept another provider's paid service. Splitting one item into several and charging for each is a hidden price rise as well. Ask for the price list on the spot, and keep the receipts and the messages. Complaining afterwards is still in time. For prices go to the market regulation authorities, for the funeral service itself go to the civil affairs department. +- Benefit: The regulations list five kinds of price violation, and four of them are set out here. Not displaying a price list, and taking more money than the posted price, are a failure to mark prices clearly or an addition to the posted price. Not being allowed to have one thing unless you buy another is a tying arrangement, or the compulsory sale of goods by attaching unreasonable conditions. Having another provider's paid service pushed on you is the compulsory or disguised compulsory acceptance of a paid service from a third party. Taking more money under another name is a disguised increase of the charge standard. The regulations also forbid 「分解项目、扩大范围收费等形式变相提高收费标准」 — disguised increases of the charge standard by splitting items, widening the scope of the charge, and the like. Splitting one service into several and charging for them separately, or pushing the scope of the charge outwards, is among the forbidden things +- Evidence grade: A +- Sources: State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，2025 年 11 月 14 日国务院第 72 次常务会议修订通过，「自2026年3月30日起施行」）. 第四十二条：「殡葬服务机构不得违反有关规定擅自设立收费项目，或者采取分解项目、扩大范围收费等形式变相提高收费标准。」第四十三条列举的价格违法行为含「（一）不按规定明码标价，或者在标价之外加价或者收取未标明的费用」「（三）采取捆绑、附加不合理条件或者不合理限制等方式，强制或者变相强制销售丧葬用品、提供殡葬相关服务并收取费用」「（四）强制或者变相强制丧属接受第三方的有偿服务，或者利用相关服务网点重复收取费用」；第五十一条要求民政部门与政务服务便民热线建立信息共享监督机制，「对丧属反映的殡葬服务问题，应当会同有关部门及时予以处理并将处理结果告知丧属」 +- Notes: Price violations are dealt with by the market regulation authorities, and the funeral service itself by the civil affairs department. Ask for a price list on the spot, keep the receipts and the messages, and complaining afterwards is still in time. +### 8. Funeral agents, agents who buy the goods, planners and hosts of the ceremony: intermediaries like these must file with the county civil affairs department + +- Cost: No money. One question is enough: has it filed? +- In plain language: Intermediaries in funeral agency, in buying the goods, in planning and hosting the ceremony and in giving information all have to file with the civil affairs department (民政部门) of the county. An intermediary that has not filed does not get into the funeral services information system run by the civil affairs department. Whether you can look it up yourself varies from place to place, and the surest way is to telephone the county civil affairs department. Anyone who uses deception to charge fees, or who sells on the information of the dead, will be dealt with under the law. +- Benefit: An intermediary that has not filed is not in the funeral services information system under the civil affairs department. Anyone who charges fees in breach of the rules by deceptive means, encourages extravagant and wasteful funerals, forces people to accept services, or sells on the information of the dead will be dealt with under the law, will be flagged in the system, and will be published for the public to see +- Evidence grade: A +- Sources: State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，2025 年 11 月 14 日国务院第 72 次常务会议修订通过，「自2026年3月30日起施行」）. 第四十一条：「除殡葬服务机构外，从事殡葬服务代理、用品代购、策划主持、信息咨询等殡葬相关服务活动的组织和个人，应当向县级人民政府民政部门备案。」「对存在以欺骗等手段违规收费、诱导大操大办、强迫接受服务、倒卖逝者信息等损害丧属权益行为的依法处置，并在殡葬服务信息系统中进行标注，向社会公开。」 +- Notes: Selling on the information of the dead is the commonest trap in this trade. If a pile of sales calls arrives the moment the paperwork is done, the information has almost certainly been sold on. Online memorial platforms are covered by the same rule. Article 45 of the regulations forbids them from pushing consumption, and forbids them from handling the information of the dead in breach of the law. + +### 9. The money sitting in different places has to be collected one at a time: the provident fund balance, the payments under social insurance, the payments for a work injury + +- Cost: No money, but you have to visit several counters. Before you go, prepare the death certificate and the proof of the relationship. +- In plain language: An heir can withdraw the balance in the housing provident fund, it does not sit in a bank account, and it is the one most easily forgotten. What is left in the personal account under the old-age insurance can also be inherited. Where a person dies of illness, the family receives a funeral subsidy and a survivor's pension of 3 to 24 months as separate payments. Where a person dies as a result of a work injury, there are the three payments under the work injury rules as well. +- Benefit: The balance in a housing provident fund account (what is left in that person's provident fund account) may be withdrawn by the heirs or by a legatee under the will. A legatee under the will is the person written into the will to leave the money to. Where a person dies as a result of a work injury, there are three further payments: the funeral subsidy, the pension for dependants and the one-off payment for a death from a work injury. The Social Insurance Law states that what is left in the personal account under the old-age insurance may be inherited. Where a person who took out the employee old-age insurance dies of illness or of something other than a work injury, the family can receive the funeral subsidy and the survivor's pension, which together are called the survivor benefits. For enterprise employees the standard follows one formula for the whole country. The funeral subsidy is 2 times the average monthly disposable income of urban residents in the province in the previous year. The survivor's pension is based on that monthly income and is paid for 3 to 24 months according to the years you have contributed: under 5 years, 3 months; 5 to 10 years, 6 months; 10 to 15 years, 9 months; over 15 years, 1 month more for every extra year, up to 24 months. For someone already retired, every year of pension already drawn reduces it by 1 month, down to a minimum of 9 months +- Evidence grade: A +- Sources: State Council (国务院) (2026). Regulations on the Housing Provident Fund, 住房公积金管理条例（国务院令第 844 号修改，2026 年 9 月 20 日起施行）第二十四条末款. :「职工死亡或者被宣告死亡的，职工的继承人、受遗赠人可以提取职工住房公积金账户内的存储余额；无继承人也无受遗赠人的，职工住房公积金账户内的存储余额纳入住房公积金的增值收益。」；the three payments under the Regulations on Work Injury Insurance, see section 19. ；Standing Committee of the National People's Congress (全国人大常委会) (2018 修正). Social Insurance Law, 社会保险法（第十四、十七条）：「个人死亡的，个人账户余额可以继承」「参加基本养老保险的个人，因病或者非因工死亡的，其遗属可以领取丧葬补助金和抚恤金」. (National Database of Laws and Regulations)；Ministry of Human Resources and Social Security (人力资源社会保障部) and Ministry of Finance (财政部) (2021). Interim Measures on the Survivor Benefits of Enterprise Employees under the Basic Old-Age Insurance for Employees, 企业职工基本养老保险遗属待遇暂行办法（人社部发〔2021〕18 号，2021 年 9 月 1 日起施行，第四、五条），the original is not posted online, and the articles were checked word for word against the quotation in the administrative reconsideration decision of the People's Government of Huadu District, Guangzhou (花都府行复〔2022〕226 号):「丧葬补助金的标准，按照参保人员死亡时本省……上一年度城镇居民月人均可支配收入的2倍计算」「缴费年限不满5年的，发放月数为3个月……缴费年限30年以上的，按照30年计算，发放月数最高为24个月」. ；Fujian Provincial Department of Human Resources and Social Security (福建省人力资源和社会保障厅). Frequently asked questions:「每领取1年基本养老金减少1个月，发放月数最低为9个月」「累计缴费年限不足5年的，其遗属待遇（含丧葬补助金和抚恤金）标准不得超过其个人缴费之和」. +- Notes: The provident fund is the one most often forgotten, because it is not in a bank account and it does not flow automatically into the distribution of the estate. What is left in the personal account under the old-age insurance and the survivor benefits are worked out and paid separately, so you go to the social insurance counter of the place where the person was insured and deal with both at once. What each place asks for is different. Where the contributions come to under 5 years, the survivor benefits together come to no more than what the person paid in. This formula comes from the rules for the old-age insurance of enterprise employees; for the survivor benefits under the insurance for public institutions and for urban and rural residents, ask at the counter. The formal requirements for a will and for an inheritance are in section 17. + +### 10. The deceased's accounts and personal information: close relatives may ask to look at them, take a copy, have them corrected, or have them deleted + +- Cost: No money. You only need to report to the authorities or sue if the platform refuses to cooperate. +- In plain language: After someone dies, close relatives may, for their own lawful and proper interests, ask the platform to let them look at the deceased's personal information, take a copy, correct it or delete it. If the platform refuses, it has to give its reasons. You can also take the case to court. There is one exception: where the person arranged something in advance, for example by naming an account successor or leaving a clear statement of wishes, his arrangement is followed. +- Benefit: The law says plainly: where a natural person dies, close relatives, for their own lawful and proper interests, may exercise the rights set out in that chapter to look at, take a copy of, correct and delete the deceased's relevant personal information. Close relatives may ask the platform to let them see the information, take a copy, put right the errors and delete it +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Personal Information Protection Law, 个人信息保护法. 中国人大网. 第四十九条：「自然人死亡的，其近亲属为了自身的合法、正当利益，可以对死者的相关个人信息行使本章规定的查阅、复制、更正、删除等权利；死者生前另有安排的除外。」；第五十条「个人信息处理者应当建立便捷的个人行使权利的申请受理和处理机制。拒绝个人行使权利的请求的，应当说明理由」「个人可以依法向人民法院提起诉讼」 +- Notes: The article has one clause, 「死者生前另有安排的除外」 — where the person made arrangements during his lifetime, either by naming an account successor or by leaving a clear statement of wishes, you follow what he left behind. So while the person is still lucid it is worth writing down where the accounts and the passwords are to go, see section 17. Cancelling an account and deleting personal information are two different things, see section 14. \ No newline at end of file diff --git "a/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" "b/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" new file mode 100644 index 00000000..9f73acc5 --- /dev/null +++ "b/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" @@ -0,0 +1,157 @@ +[← 回总目录](../README.md) + +# 27. 怀孕和生产：从发现怀孕到出院办证 + +这一节前半段算的是孕妇和胎儿的死亡风险。后半段算的是钱和跑腿时间。两种账不放在一起换算。内容只讲「什么时候做什么」。孕期营养、胎教这类没有硬证据、或者作用很小的事，这里不写。生育津贴、产假和养孩子的钱账见第 18 节，孩子出生之后怎么带见第 20 节，急症见第 13 节。 + +世界卫生组织给的数字：2023 年，全球约有 26 万名女性死于怀孕和生产。其中约 75% 死于这五类原因：产后大出血、感染、妊娠期高血压疾病（怀孕后血压升高的一类病，包括子痫前期和子痫）、分娩时的并发症、不安全的人工流产。世界卫生组织同时说，「大多数孕产妇死亡是可以预防的，因为预防和处理并发症的医疗手段都是已知的」。这一节里作用最大的几条，就是把这几类问题提前拦下来。 + +[← Back to the contents](../README.md) + +# 27. Pregnancy and childbirth: from test to paperwork + +The first half of this section counts the risk of death for the pregnant woman and the fetus. The second half counts money and time spent on errands. The two are not converted into each other. The text only says what to do and when. Pregnancy nutrition and prenatal education have no hard evidence behind them or make very little difference, so they are left out. For the money side of maternity allowances, maternity leave and raising a child see section 18; for caring for the baby after birth see section 20; for emergencies see section 13. + +The numbers the World Health Organization gives: in 2023 about 260,000 women worldwide died in pregnancy and childbirth. About 75% of those deaths came from five causes: severe bleeding after birth, infection, hypertensive disorders of pregnancy (the disorders in which blood pressure rises after conception, including pre-eclampsia and eclampsia), complications during delivery, and unsafe induced abortion. The World Health Organization also says that 「大多数孕产妇死亡是可以预防的，因为预防和处理并发症的医疗手段都是已知的」. The entries with the largest effect in this section are the ones that catch these problems early. + +### 1. Start 0.4 mg of folic acid every day before conception and keep taking it through the first 3 months of pregnancy + +- Cost: Tens of yuan for a bottle. Rural women pay nothing, because the national free folic acid supplementation programme provides it. One tablet a day. The hard part is remembering it every day +- In plain language: One folic acid tablet a day before and during conception cuts the chance of a neural tube defect by about seven tenths. Neural tube defects are birth defects such as anencephaly and spina bifida. If you have already had one child with such a defect, the chance that the next child has one falls to about one third of what it was. +- Benefit: The Cochrane systematic review pooled 5 trials and 6708 deliveries, and rates the quality of the evidence high. Taking folic acid every day before and during conception gives RR 0.31 for fetal neural tube defects (95% CI 0.17 to 0.58, credible range), about a seven-tenths reduction. For women who have already had a child with a neural tube defect, the risk that the next child has one is RR 0.34 (95% CI 0.18 to 0.64), roughly one third of what it was. Broken down by population group, 0.4 mg and higher doses make no difference. Folic acid on its own and folic acid inside a multivitamin make no difference either +- Evidence grade: A +- Sources: De-Regil LM, Peña-Rosas JP, Fernández-Gaxiola AC, Rayco-Solon P (2015). Effects and safety of periconceptional oral folate supplementation for preventing birth defects. Cochrane Database of Systematic Reviews, (12), CD007950. +- Notes: What matters is starting before conception. The fetal neural tube closes within 28 days of fertilisation. By the time a pregnancy test turns positive, most of that window has gone. So if you plan to have a child, start 3 months beforehand. The same pooled review found no clear effect of folic acid on cleft lip and palate, congenital heart disease or miscarriage. Do not expect it to treat anything else + +### 2. Before 13 weeks of pregnancy open a Mother and Child Health Handbook at the community health centre, and use up your free prenatal checks + +- Cost: No money. It is part of the national basic public health service programme. You have to make several trips according to how far along you are +- In plain language: At least five prenatal checks during pregnancy, plus one home visit after the birth and one check at 42 days postpartum, all paid for by the state. The way to claim them is to open a Mother and Child Health Handbook《母子健康手册》at the community health centre before 13 weeks of pregnancy. Without the handbook you simply lose those free checks. +- Benefit: The National Standard for Basic Public Health Services (Third Edition)《国家基本公共卫生服务规范（第三版）》sets out maternal and child health management services. Any woman living in the area who is pregnant can have them: 1 health check in the first trimester, 2 in the second trimester, 2 in the third trimester, 1 postpartum visit and 1 health check at 42 days postpartum. You open the handbook before 13 weeks of pregnancy and have the first prenatal check at the same time. The two checks in the second trimester fall between 16 and 20 weeks and between 21 and 24 weeks. The two in the third trimester fall between 28 and 36 weeks and between 37 and 40 weeks. Locally published lists of services also include the full set of lab tests at the first visit and screening for postpartum depression +- Evidence grade: A +- Sources: National Health and Family Planning Commission (国家卫生计生委) (2017). National Standard for Basic Public Health Services, Third Edition（国家基本公共卫生服务规范（第三版））. ; Jingjiang Municipal People's Government (靖江市人民政府). National Basic Public Health Service Programme — Maternal and Child Health Management Services (local published service list)（国家基本公共卫生服务项目——孕产妇健康管理服务（地方服务清单公开））. +- Notes: Those 5 checks are the free floor, not the whole of prenatal care. The hospital will add ultrasound, blood glucose and fetal heart monitoring where it sees fit, and you pay for those. The earlier you open the handbook, the earlier the pregnancy risk grading is worked out. Pregnancy risk grading puts pregnant women into tiers by risk level. Unless a high-risk tier is found, you cannot be referred in time to a hospital that can take the case. If you live in a district other than your registered one, open the handbook where you live now. There is no need to go back to your registered address + +### 3. Test for HIV, syphilis and hepatitis B at the first prenatal check, and free blocking treatment is available if any is positive + +- Cost: No money. The programme for preventing mother-to-child transmission provides the tests free +- In plain language: Have all three tested for at the first prenatal check, and the tests cost you nothing. A positive result is not a dead end. The state supplies the medicine and the regimen free, to block the route from mother to child. That is called blocking of mother-to-child transmission. +- Benefit: Mother-to-child transmission means passing an illness from mother to child. The Working Specification on the Prevention of Mother-to-Child Transmission of HIV, Syphilis and Hepatitis B (2020 Edition)《预防艾滋病、梅毒和乙肝母婴传播工作规范（2020 年版）》sets requirements for it. In the first trimester, or at the first prenatal check, the hospital must tell you about this and 「提供适宜、规范的免费检测」, so the three tests cost you nothing. It then gives post-test counselling or a referral depending on the result. A positive result is not a dead end. Article forty-four of the Regulations on the Prevention and Control of AIDS《艾滋病防治条例》says that a pregnant woman infected with HIV is 「免费提供预防艾滋病母婴传播的治疗和咨询」. Article forty-three requires blocking (stopping the illness from passing to the child), treatment, a postpartum visit, follow-up of the baby and testing. If the mother's hepatitis B surface antigen is positive, the newborn gets both the hepatitis B vaccine and hepatitis B immune globulin at birth (see item 2 in section 20) +- Evidence grade: A +- Sources: General Office of the National Health Commission (国家卫生健康委办公厅) (2020). Working Specification on the Prevention of Mother-to-Child Transmission of HIV, Syphilis and Hepatitis B (2020 Edition)《预防艾滋病、梅毒和乙肝母婴传播工作规范（2020 年版）》. (reprinted by the Xinjiang Uygur Autonomous Region Health Commission (新疆维吾尔自治区卫生健康委员会)) ; State Council (国务院) (2006). Regulations on the Prevention and Control of AIDS (Articles forty-three and forty-four)《艾滋病防治条例》. (reprinted by the Beijing Municipal Health Commission (北京市卫生健康委员会)) +- Notes: The earlier these three are tested, the better, because blocking and treatment both take time. Article three of the same regulations also states that 「任何单位和个人不得歧视艾滋病病毒感染者、艾滋病病人及其家属」. The lawful rights to marry, to find work, to see a doctor and to go to school are protected by law. For everyday prevention and testing see section 1. For post-exposure blocking after high-risk behaviour see section 13 + +### 4. Not one cigarette and not one drink during pregnancy, and nobody in the household smokes indoors + +- Cost: No money, and you also save the money you would spend on tobacco and alcohol. The hard part is getting everyone you live with to stop as well +- In plain language: Smoking during pregnancy raises the chances of preterm birth, low birth weight and cleft lip and palate. It doubles the risk of abnormal bleeding during pregnancy and childbirth. It also raises the risk of sudden infant death syndrome. Smoke breathed in at home by family members counts the same. Quitting at any point still helps. +- Benefit: The US CDC (美国疾病控制与预防中心) puts it this way. Smoking during pregnancy means a baby may be 「足月出生也可能过小」 and may be 「可能早产」. It 「使孕期和分娩期异常出血的风险翻倍」. It raises the risk of birth defects such as cleft lip and cleft palate. It also 「损害胎儿正在发育的肺和大脑，损害可持续到儿童期」. Babies of mothers who smoke during pregnancy, and babies exposed to smoke after birth, are at higher risk of sudden infant death syndrome. The official advice is 「最好在怀孕前就戒烟，但如果已经怀孕，戒了仍然有用」 and 「什么时候戒都不晚」. +- Evidence grade: A +- Sources: Centers for Disease Control and Prevention. Smoking During Pregnancy. +- Notes: Smoke from someone else indoors counts for the fetus, because it is the pregnant woman who breathes it in. So family members should not smoke indoors either. Alcohol is the same: no amount of drinking is known to be safe, and the simplest rule during pregnancy is to drink nothing at all. For what tobacco and alcohol cost an adult in their own right see section 2 + +### 5. If you have risk factors for pre-eclampsia, start one low-dose aspirin tablet a day after 12 weeks of pregnancy + +- Cost: Tens of yuan. One tablet a day, taken right through to the birth. The hard part is never forgetting +- In plain language: Pre-eclampsia is a high blood pressure disorder of pregnancy. You count as high risk if you had it in a previous pregnancy, are carrying twins or more, have chronic high blood pressure, had diabetes before pregnancy, or have kidney disease or an autoimmune condition. Starting one low-dose aspirin tablet a day at 12 weeks of pregnancy cuts pre-eclampsia by about a tenth and a half, and cuts both preterm birth and the child's death around the time of birth by about a fifth. Everyone else should not take it on their own. +- Benefit: The US Preventive Services Task Force gives a grade B recommendation: to prevent pre-eclampsia in people at high risk, start low-dose aspirin (81 mg/day) after 12 weeks of pregnancy. Pre-eclampsia is a high blood pressure disorder of pregnancy. Pooling several studies gives the following results, where the 95% CI is the credible range. Pre-eclampsia RR 0.85 (95% CI 0.75 to 0.95, 16 studies), about 15% lower. Preterm birth RR 0.80 (95% CI 0.67 to 0.95, 13 studies), about 20% lower. Perinatal death RR 0.79 (95% CI 0.66 to 0.96, 11 studies), about 21% lower; perinatal death means death in the period around childbirth. Small for gestational age / fetal growth restriction RR 0.82 (95% CI 0.68 to 0.99, 16 studies), about 18% lower; this means the fetus grew smaller than usual for its gestational age +- Evidence grade: A +- Sources: US Preventive Services Task Force (2021). Aspirin Use to Prevent Preeclampsia and Related Morbidity and Mortality: Preventive Medication. +- Notes: The high-risk categories are: pre-eclampsia in a previous pregnancy, twins or more this time, and chronic high blood pressure. Also type 1 or type 2 diabetes before pregnancy, kidney disease, and autoimmune conditions such as systemic lupus erythematosus or antiphospholipid syndrome. If you are in none of these categories, do not take it on your own. A doctor has to prescribe this, so give your full medical history at the prenatal check and let the doctor judge. What is sold in China is mostly the 100 mg enteric-coated tablet. Take it as the doctor instructs + +### 6. Have one screening for gestational diabetes at 24 weeks of pregnancy or later, and do not begrudge the sugar drink + +- Cost: Tens to a little over a hundred yuan. You go fasting and give several blood samples. It takes half a morning +- In plain language: Gestational diabetes is high blood sugar that appears only after pregnancy. After 24 weeks, half a morning with one sugar drink and a few blood samples will find it. Leave it unchecked and the fetus grows too large, labour becomes difficult, you may need an operation, and the risk of pre-eclampsia and of low blood sugar in the newborn both rise. +- Benefit: The US Preventive Services Task Force gives a grade B recommendation: screen for gestational diabetes at 24 weeks of pregnancy or later in women with no symptoms (who have noticed nothing wrong themselves). Gestational diabetes is high blood sugar that appears only after pregnancy. For anyone who screens only once, that one screen should be at 24 weeks or later. On whether screening before 24 weeks helps, the task force concluded that the present evidence is not enough to tell the benefits from the harms (grade I). Untreated gestational diabetes raises the risk of macrosomia (a fetus that grows too large), difficult labour, caesarean section, pre-eclampsia and low blood sugar in the newborn +- Evidence grade: A +- Sources: US Preventive Services Task Force (2021). Gestational Diabetes: Screening. +- Notes: Among the women who test positive, most can keep their blood sugar under control by changing diet and exercise, and only a minority need insulin. Do not skip this one test because you are afraid of what a positive result would mean. After the birth you have to test your blood sugar again: women who have had gestational diabetes are clearly at higher risk of type 2 diabetes later + +### 7. Memorise this list of 「立刻去医院」 signs; it counts during pregnancy and for a year after the birth + +- Cost: No money. Five minutes to memorise it, and it is best if the people you live with know it too +- In plain language: If any one of these appears, go to hospital that day. Do not sit at home for a night first. A headache that does not lift or keeps getting worse, vision changing, a fever of 38 °C. Severe swelling of the hands or face, breathlessness, chest pain or a fast heartbeat. Severe abdominal pain that does not ease, the baby moving normally and then stopping or clearly less often, vaginal bleeding or fluid loss. One arm or one leg swollen and painful, thoughts of harming yourself or the baby. All of it counts for a year after the birth too. +- Benefit: The US CDC lists these urgent maternal warning signs: a headache that does not lift or keeps getting worse, dizziness or fainting, vision changing, a temperature of 38 °C or above; severe swelling of the hands or face, thoughts of harming yourself or the baby, difficulty breathing, chest pain or a very fast heartbeat; severe nausea and vomiting, severe abdominal pain that does not ease, and during pregnancy the baby moving normally and then stopping or clearly less often; vaginal bleeding or fluid loss during pregnancy, or abnormal vaginal bleeding or discharge after the birth; one arm or one leg severely swollen, red or painful. All of these can happen within a year after the birth. Getting out of the maternity ward does not make you safe +- Evidence grade: A +- Sources: Centers for Disease Control and Prevention. Hear Her — Urgent Maternal Warning Signs. ; World Health Organization. Maternal mortality fact sheet. +- Notes: What this list is for is telling 「再忍忍」 apart from 「必须马上走」. A headache together with blurred vision and pain in the upper abdomen points to pre-eclampsia (the high blood pressure disorder of pregnancy). One lower leg swollen and painful points to deep vein thrombosis (see section 13). By the World Health Organization's account, severe bleeding after birth can take the life of a healthy woman within hours. Go straight to hospital as soon as any one item appears, and do not spend a night at home first +### 8. If your waters break, lie down flat where you are, raise your hips and call 120; do not walk about and do not bathe + +- Cost: No money +- In plain language: When the waters break you do not pack slowly and then set off. The right move is to lie down where you are, raise your hips and call 120. Standing and walking makes the fluid drain faster, and the umbilical cord can be washed out with it. That is an emergency that can kill the baby. +- Benefit: The handling given by the Beijing Municipal Health Commission (北京市卫生健康委员会) is this: once the waters have broken, 「马上找个地方平躺下来，防止羊水流出来」 and 「不要随意走动」, and 「立即拨打 120 急救电话」. While you wait for the ambulance, tell your family to bring your things and meet you at the hospital. Material from medical institutions explains why: fluid drains faster when a woman is standing, and the umbilical cord can come out with it. That is called cord prolapse, and it is an emergency that puts the fetus at risk. So lie down and raise your hips. A left lateral position is one way to do that +- Evidence grade: B +- Sources: Beijing Municipal Health Commission (北京市卫生健康委员会). Weekly first aid topic: what to do when a pregnant woman's waters break early（每周急救话题：准妈妈羊水早破怎样办？）. ; Beijing Shunyi District Maternal and Child Health Hospital (北京市顺义区妇幼保健院). What to do when the waters break at home in late pregnancy（孕晚期在家破水怎么办？）. +- Notes: The three signs that labour is close are regular contractions, the bloody show, and the waters breaking. You can pack calmly for the first two. Breaking of the waters is the only one that means lying down on the spot and calling for a car. Bathing, squatting on the toilet and driving yourself after the waters break are all classic mistakes. Write down the time the waters broke and the colour of the fluid, and tell the doctor. Yellow or green fluid, or fluid with flakes of meconium in it, points to fetal distress (the fetus is short of oxygen inside the uterus). If regular contractions, a clear drop in the baby's movement, and fever with shivering are happening at the same time, stress to the doctor that this is an emergency + +### 9. If you want pain relief in labour, say so directly. It does not raise the risk of caesarean section + +- Cost: Several hundred to over a thousand yuan, and the price differs from place to place. You have to ask the hospital and the anaesthesia department in advance whether they can do it. The hard part is raising it yourself +- In plain language: Pain relief in labour is an injection of anaesthetic into the back that pushes down the pain of giving birth. It works better than other ways of easing pain, and it does not make you more likely to have an operation. The claim that 「打了无痛腰疼一辈子」 is no better founded. It now has a separate billing item of its own, so hospitals are more willing to provide it. +- Benefit: The Cochrane systematic review concludes that epidural analgesia is more effective than methods without an epidural at relieving labour pain and at raising how satisfied a woman is with the analgesia. Pain relief in labour is exactly epidural analgesia. The same review also says that 「硬膜外镇痛对剖宫产风险和长期腰痛没有影响」. The rise in instrumental birth (using instruments to deliver the baby) is no longer seen in studies from 2005 onwards. From 2024, the project of the National Healthcare Security Administration for setting up prices for obstetric medical services gives 「分娩镇痛」, 「导乐分娩」 and 「亲情陪产」 items of their own. The reason is precisely that without a separate price item before, hospitals had little incentive to offer pain relief in labour +- Evidence grade: A +- Sources: Anim-Somuah M, Smyth RMD, Cyna AM, Cuthbert A (2018). Epidural versus non-epidural or no analgesia for pain management in labour. Cochrane Database of Systematic Reviews, (5), CD000331. ; National Healthcare Security Administration (国家医保局) (2024). Rebuilding the obstetric service price items to help build a childbirth-friendly society《重构产科服务价格项目 助力构建生育友好型社会》. +- Notes: The two claims that spread widest are that pain relief in labour means a back problem for life, and that it makes the birth hard so an operation is needed. Neither is founded. Cochrane pooled epidural analgesia with the other methods and concluded that it has no effect on the risk of caesarean section and none on long-term back pain. The real side effects are these: low blood pressure, fever, weakness in the legs, difficulty passing urine, and a longer first and second stage of labour with more likelihood of needing oxytocin. Whether to have it is yours to decide, but do not give it up on the strength of a rumour. Whether it is available also depends on whether the hospital has an anaesthetist on site 24 hours a day. Ask when you go for the prenatal check + +### 10. Without a medical indication do not ask for a caesarean section, and do not have an operation to pick an auspicious date + +- Cost: No money. The hard part is standing your ground against what the family says +- In plain language: The World Health Organization's position is that once the caesarean rate in a place passes 10%, deaths among mothers and newborns stop falling any further. Where there is a reason to operate, operate. Having an operation to pick a date, or because you are afraid of pain, buys you one large abdominal operation for nothing. +- Benefit: The World Health Organization's 2015 statement: when the caesarean rate in a population rises to about 10%, deaths among mothers and newborns fall. Past 10%, there is no evidence that mortality improves any further. The World Health Organization also stresses that every effort should be made to give a caesarean section to the women who need one, rather than to reach some particular rate +- Evidence grade: A +- Sources: World Health Organization (2015). WHO Statement on Caesarean Section Rates（WHO/RHR/15.02）. +- Notes: When there is a real medical need, a caesarean section is an operation that saves lives. Do it, and do not tough it out. What to avoid is an operation that is not medically needed, whether out of fear of pain, to pick an auspicious date, or to fit in with the age at which the child starts school. A caesarean section is a large abdominal operation, recovery is slower, and the next pregnancy carries higher risks of placenta previa, placenta accreta and uterine rupture. That bears directly on whether you still want a second child + +### 11. Check your maternity insurance status before conception; a spouse who is not working can also claim the maternity medical costs + +- Cost: No money. One call to the social insurance or medical insurance enquiry line +- In plain language: When an employer pays maternity insurance, the medical costs of the birth and the allowance during maternity leave come out of that fund. If the spouse does not work, the medical costs can be claimed as well. Whether you have paid in for enough months, and whether a birth outside the area has to be filed for first, both have to be asked in advance. Neither can be put right afterwards. +- Benefit: Article fifty-four of the Social Insurance Law《社会保险法》says: where an employer has already paid maternity insurance contributions, its employees enjoy maternity insurance benefits. Where an employee's spouse does not work, the spouse enjoys the maternity medical cost benefits as the state provides, and the money is paid out of the maternity insurance fund. Put plainly, once the employer pays maternity insurance the employee qualifies for the benefits, and the medical costs of the birth of that employee's non-working spouse come out of the same fund. The benefits have two parts. One is the maternity medical cost, which is the medical cost of the birth itself. The other is the maternity allowance, the money paid to you during maternity leave. Article fifty-six sets out the three cases in which the maternity allowance can be received: a female employee giving birth and taking maternity leave, taking leave for a family planning operation, and other cases set out in laws and regulations. The maternity allowance is 「按照职工所在用人单位上年度职工月平均工资计发」, which works out at the average monthly wage of the employees of your employer last year, not at your own wage +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2010). Social Insurance Law of the People's Republic of China (Articles fifty-four and fifty-six)《中华人民共和国社会保险法（第五十四、五十六条）》. (reprinted by the Guangdong Provincial Tax Service, State Taxation Administration (国家税务总局广东省税务局)) +- Notes: Two things have to be asked in advance, because neither can be put right afterwards. The first is whether the number of consecutive months of contributions is enough: every place sets a minimum, and below it you get nothing. The second is whether a birth outside the place of enrolment has to be filed for in advance: without the filing you may have to pay the money yourself first and claim it by hand afterwards, and sometimes you cannot claim it at all. For the number of days of maternity leave and how the maternity allowance is worked out see item 2 in section 18 + +### 12. Get the Medical Certificate of Birth done before you leave the hospital; decide the name in advance and do not get a character wrong + +- Cost: No money. It is done as part of the discharge formalities +- In plain language: This certificate is a precondition for household registration, for insurance and for school, and only the institution that delivered the baby can issue the first one. If you go back for it after leaving the hospital, you go through a separate set of procedures and hand in more documents. Check first that the characters in the name can be produced by the household registration system. +- Benefit: Article twenty-three of the Law on Maternal and Infant Health《母婴保健法》says: medical and health institutions and the persons who carry out home births shall, in accordance with the rules of the health administrative department of the State Council, issue the uniformly produced and issued Medical Certificate of Birth for the newborn. This certificate is the precondition for everything that comes after it: household registration, insurance, a medical insurance card, and kindergarten and school. And the first issue can only be done by the institution that delivered the baby. If you go back for it after leaving the hospital, you have to follow a different set of procedures and supply extra supporting documents +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (1994, amended 2017). Law of the People's Republic of China on Maternal and Infant Health (Article twenty-three)《中华人民共和国母婴保健法（第二十三条）》. +- Notes: Before filling it in, check the parents' names, their identity card numbers and the child's name character by character. The characters used in the name have to be ones the household registration system can produce, and rare characters make every later errand harder. Changing the name or reissuing the certificate means going through a supplementary or a replacement issue, which is far more trouble than checking it on the spot. Keep the certificate on its own once you have it, as carefully as the household register and the identity card + +### 13. Do not refuse either the newborn heel-prick screening or the hearing screening + +- Cost: Free in most regions, or tens of yuan. Some extended items are yours to pay for. The baby's heel is pricked once after birth +- In plain language: One prick of the heel and one hearing test can find hypothyroidism, phenylketonuria and hearing loss while the child still looks completely normal. By the time the symptoms show, the damage to the child's intellect cannot be taken back. +- Benefit: Article three of the Measures for the Administration of Screening for Neonatal Diseases《新生儿疾病筛查管理办法》states: 「全国新生儿疾病筛查病种包括先天性甲状腺功能减低症、苯丙酮尿症等新生儿遗传代谢病和听力障碍。」 These two kinds of illness have one thing in common. In the early stage nothing abnormal can be seen, and by the time the symptoms are obvious the damage to the intellect is already irreversible. Early detection and early intervention work well. The methods are thyroid hormone replacement, special formula, a hearing aid, or a cochlear implant. Article ten says that where a screening result is positive, the institution shall notify the guardian in good time to have a confirmatory diagnosis +- Evidence grade: A +- Sources: Ministry of Health (卫生部) (2009). Measures for the Administration of Screening for Neonatal Diseases (Ministry of Health Order No. 64, Articles three, ten and eleven)《新生儿疾病筛查管理办法（卫生部令第 64 号，第三、十、十一条）》. +- Notes: Article eleven requires that before the screening the items, the conditions, the method, the sensitivity (how much it can find) and the cost are all explained truthfully to the guardian, who then signs consent. So you will be given an informed consent form. Read it before you sign: which conditions are the ones the state requires, and which are extra items you pay for. A positive first screen is not a diagnosis. Most are ruled out at the repeat test, so do not fall apart at this stage. If the first hearing screen is not passed, go for the repeat test when you are told to. Do not decide on your own that the child can hear well enough + +### 14. Enrol the child in resident medical insurance as soon as you have the Medical Certificate of Birth; do not wait for household registration + +- Cost: Resident medical insurance is paid once a year, a few hundred yuan +- In plain language: Once you have the certificate you can enrol the child online, with no need to wait for household registration. Enrolled, the inpatient bill from the hospital stay can be settled on the spot. A newborn's inpatient bill is not small, and being enrolled or not makes a large difference. +- Benefit: The State Council's official website (中国政府网) puts it this way in 2024: 「新生儿凭出生医学证明就可以在线上参保」, with no need to register the household first. 「新生儿出院以后就能够直接报销医药费用」. The old practice was that you had to register the household before you could enrol, and the inpatient bill had to be paid in full up front and claimed afterwards. The average time to process an enrolment and a payment has also come down from 28.7 working days at the start of the year to 6.4 working days +- Evidence grade: A +- Sources: State Council official website (中国政府网) (2024). Regular policy briefing on a sound long-term mechanism for basic medical insurance enrolment《健全基本医疗保险参保长效机制国务院政策例行吹风会》. +- Notes: Preterm birth, jaundice needing phototherapy and newborn pneumonia are the common reasons for a newborn to be admitted, and they run into thousands of yuan. Being enrolled or not makes a large difference. The rules on how many days after birth you must enrol in order to get cover from the day of birth itself are not the same everywhere. For how it is actually applied, follow what the local medical insurance authority says. One call to 12393 before you leave the hospital settles it + +### 15. Go to the police station to register the birth within one month of the child being born + +- Cost: No money. One trip to the household registration window at the police station +- In plain language: The law gives you one month, not "when there is time". Household registration holds up the vaccinations, the kindergarten, the school and the identity card that come after it. The longer you put it off, the more documents you have to supply later. +- Benefit: Article seven of the Regulations on Household Registration《户口登记条例》says: 「婴儿出生后一个月以内，由户主、亲属、抚养人或者邻居向婴儿常住地户口登记机关申报出生登记。」 That is a deadline set by law, not a suggestion. Household registration is the precondition for what follows: vaccinations, kindergarten, school, the identity card. The longer you put it off, the more documents you have to supply +- Evidence grade: A +- Sources: State Council (国务院) (1958). Regulations of the People's Republic of China on Household Registration (Article seven)《中华人民共和国户口登记条例（第七条）》. (reprinted by the Guangdong Provincial People's Government (广东省人民政府)) +- Notes: Bring all of these: the Medical Certificate of Birth, both parents' identity cards and household registers, and the marriage certificate. Whether to register with the father or with the mother is yours to choose, but the rules of the two places differ and both the school catchment and the place of insurance are affected. Decide before you go; changing it later is a nuisance. A child born outside marriage can be registered in the same way +### 16. Do not skip the 42-day postpartum check; it is also the screening for postpartum depression + +- Cost: No money, it is part of the basic public health service. It takes half a day. The hard part is that you have just left the maternity ward and cannot be bothered +- In plain language: This check is free. It covers a full blood count and a routine vaginal swab, and it includes one screening for postpartum depression. Serious problems can still happen within a year after the birth, so do not assume you are safe once the maternity ward is behind you. +- Benefit: The National Standard for Basic Public Health Services (Third Edition)《国家基本公共卫生服务规范（第三版）》sets out maternal and child management, with 1 postpartum visit and 1 health check at 42 days postpartum. Locally published lists of services show that this check includes lab tests (a full blood count and a routine vaginal swab) and 1 screening for postpartum depression. Serious complications can occur at any time within a year after the birth. In the list 「立刻去医院」 in item 7, 「有伤害自己或伤害孩子的念头」, 「产后阴道出血或异常分泌物」 and 「一侧肢体严重肿痛」 are all postnatal +- Evidence grade: A +- Sources: National Health and Family Planning Commission (国家卫生计生委) (2017). National Standard for Basic Public Health Services, Third Edition（国家基本公共卫生服务规范（第三版））. ; Jingjiang Municipal People's Government (靖江市人民政府). National Basic Public Health Service Programme — Maternal and Child Health Management Services (local published service list)（国家基本公共卫生服务项目——孕产妇健康管理服务（地方服务清单公开））. +- Notes: Postpartum depression is not fussiness, and a word of encouragement from the family does not fix it. It has its own screening questionnaire, it can be treated, and it can be recovered from. For what to do if you have thoughts of suicide see section 1. For the lawful way out when a baby really cannot be brought up see item 20 in section 9. This check also looks at how the wound has healed, how the uterus has recovered and how the pelvic floor is working. Contraception is discussed at this visit too: no periods after the birth does not mean you cannot get pregnant \ No newline at end of file diff --git "a/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" "b/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" new file mode 100644 index 00000000..c5288c92 --- /dev/null +++ "b/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" @@ -0,0 +1,87 @@ +[← 回总目录](../README.md) + +# 28. 别为了外形把身体搞坏 + +本节说的是「为了改变外形而对身体动手」的几种做法：饿肚子、吃药、打针、开刀。这几件事，决定多半是和别人比较之后做的，代价却全落在自己身上，而且相当一部分改不回来。 + +本节不评价任何人想把自己变成什么样子，那是各人自己的事。同一个目标，可以找「有处方、有资质、有复查」的，也可以「网购、美容院、自己加量」，本节只比这两种做法的风险差多少。算账主要看身体上实际出了什么事，比如失明、住院、死亡。医美那两条还涉及非法行医的刑事责任。 + +[← Back to the contents](../README.md) + +# 28. Do not wreck your body for the sake of looks + +This section covers several ways of operating on the body to change how you look: starving yourself, taking pills, injections and surgery. The decision behind these things is usually made after comparing yourself with other people, but the price falls entirely on you, and a good part of it cannot be put back. + +This section does not judge what anyone wants to look like; that is each person's own business. The same goal can be reached through "a prescription, a qualified practitioner and follow-up visits", or through "buying online, a beauty salon, raising the dose yourself". This section only compares how much risk the two routes carry. What is counted is what actually happens to the body, such as blindness, hospital admission and death. The two cosmetic-medicine items also touch criminal liability for practising medicine without a licence. + +### 1. Do not control your weight with extreme dieting, fasting or vomiting; if you must lose weight, lose it on the exercise side + +- Cost: No money. It also saves the money you would spend on diet products. +- In plain language: People with anorexia nervosa die at about 5.9 times the rate of others their age. Follow 1000 people for one year and 5.1 of them die. One in five of those deaths is suicide. The way in is dieting: students at 44 secondary schools were followed for three years, and girls who dieted severely were 18 times as likely to develop an eating disorder as girls who did not. If you want to control your weight, move more, and do not eat so little that you are hungry. +- Benefit: 36 studies pooled. The anorexia nervosa group was followed for 166,642 person-years in total (the years each person was followed, added up). Their standardised mortality ratio is 5.86, that is, their probability of dying is about 5.9 times that of others their age. Deaths are 5.1 per 1000 person-years, that is, follow 1000 people for one year and 5.1 of them die. About 1/5 of the people who die, die by suicide. The standardised mortality ratio for bulimia is 1.93, and for other eating disorders 1.92, each about 1.9 times that of others their age. Another study was done in 44 secondary schools in Victoria, Australia: students aged 14–15 followed for 3 years, and the same group surveyed in 6 waves over the three years. New eating disorders among girls were 21.8 per 1000 person-years, that is, follow 1000 girls for one year and 21.8 new cases appear. Girls who dieted severely had 18 times the risk of those who did not diet, and moderate dieters 5 times. The analysis then adjusted for previous dieting and psychiatric symptoms, taking out the effect of "had dieted before" and "already had psychiatric symptoms". After that adjustment, BMI (the measure of fatness worked out from weight divided by height squared), amount of exercise and sex no longer predicted new eating disorders. The authors conclude that "using exercise rather than food restriction to control weight in adolescents appears to carry a lower risk of eating disorders" +- Evidence grade: A +- Sources: Arcelus J et al. (2011). Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry. ；Patton GC et al. (1999). Onset of adolescent eating disorders: population based cohort study over 3 years. BMJ. +- Notes: "Severe dieting" here means keeping food far below what the body needs over a long period, not skipping one meal or cutting out late-night snacks. If you have already started vomiting, feel guilty after bingeing, or cannot stop weighing yourself, go to a psychiatry department or the clinical psychology department of a general hospital. This is an illness with standard treatment; it is not a question of willpower. For BMI and mortality see item 33 in section 2; that item is about the long-term weight range, not about ways of losing weight. + +### 2. Before injections, thread embedding or surgery, check two things: whether the clinic's licence lists "medical cosmetology", and whether the person operating is the designated attending physician + +- Cost: No money. Checking the two certificates takes only minutes. The institution should hang its Medical Institution Practice Licence in the hall. The doctor's qualifications can be checked in the health authority's licensed physician search. +- In plain language: A needle, a scalpel or a medical device that breaks the skin or goes inside the body counts as medical treatment. It may only be done at an institution whose licence lists "medical cosmetology", and the person doing it must be a registered licensed physician. Having it done in a beauty salon, a studio or a hotel room is practising medicine without a licence. Where the circumstances are serious, the person doing it faces up to 3 years; if someone is hurt, 3 to 10 years; if someone dies, 10 years or more. +- Benefit: Article two of the Measures for the Administration of Medical Cosmetology Services (《医疗美容服务管理办法》) defines medical cosmetology. The original wording is 「运用手术、药物、医疗器械以及其他具有创伤性或者侵入性的医学技术方法对人的容貌和人体各部位形态进行的修复与再塑」, that is, the repair and reshaping of a person's appearance and the form of each part of the body by surgery, drugs, medical devices and other traumatic or invasive medical techniques. In plain terms, once the skin is broken, a needle goes in, or something goes inside the body, it counts as seeing a doctor. Articles eight and twenty-four require an institution to obtain a Medical Institution Practice Licence first, and to have the medical cosmetology subject approved by the registering authority, before it may offer the service. If the licence does not list medical cosmetology, it may not do it. Article eleven requires the designated attending physician (the doctor who does the work) to hold a licensed physician qualification and to be registered. Cosmetic surgery also requires more than 6 years of clinical work in the relevant specialty, cosmetic dentistry more than 5 years, and cosmetic traditional Chinese medicine and cosmetic dermatology 3 years each. Article twenty requires written notice before the procedure of the indications, contraindications (when it is suitable and when it is not), medical risks and precautions, and a signed consent. Article three hundred and thirty-six of the Criminal Law (《刑法》) governs practising medicine without a licence. Practising medicine without a licence means a person without a licensed physician qualification giving injections, operating or treating patients. In serious circumstances the penalty is fixed-term imprisonment of up to 3 years, criminal detention or public surveillance, together with or instead of a fine. Where the patient's health is seriously damaged, 3 to 10 years of fixed-term imprisonment plus a fine. Where the patient dies, 10 years or more of fixed-term imprisonment plus a fine +- Evidence grade: A +- Sources: Ministry of Health (卫生部) (2002). Measures for the Administration of Medical Cosmetology Services（医疗美容服务管理办法）(Ministry of Health Order No. 19, articles two, eight, eleven, sixteen, twenty and twenty-four). ；National People's Congress (全国人大) (1997). Criminal Law（刑法）(article three hundred and thirty-six). +- Notes: Article sixteen of the same Measures also provides that a cosmetic procedure may only be done in a medical cosmetology institution, or in a hospital or clinic that has a medical cosmetology department. So injections given at someone's home, or work done privately in a studio, already fail on the location alone. Article twenty states that a minor or anyone else who cannot decide for themselves may have cosmetic medicine only with the consent of a guardian, signed by a parent. Facials, massage and nail care are everyday beauty services and are not covered by these rules. But once a needle or a scalpel is involved, it counts as medical cosmetology. + +### 3. For fillers, avoid the bridge of the nose, the space between the eyebrows and the forehead, and never use an unlicensed injector to save money + +- Cost: No money. You simply inject somewhere else, or not at all. +- In plain language: Among people blinded by filler injections, more than half were injected in the nose, then the space between the eyebrows, the forehead and the smile folds, and more than eight in ten were injected with hyaluronic acid. Only two in ten of those blinded recovered their sight fully, a little over one in ten recovered partly, and nearly two in ten also had brain injury. No rescue treatment works reliably, so the only defence is to avoid these areas beforehand. +- Benefit: Cases reported in journals worldwide were pooled and counted. From 2015-1 to 2018-9 there were 48 new cases of partial or complete loss of vision after a filler injection. The site with the most accidents was the nose, at 56.3%. Next came the space between the eyebrows at 27.1%, the forehead at 18.8%, and the nasolabial fold (the crease of the smile line) at 14.6%. In 81.3% of cases the filler was hyaluronic acid, the same substance. Skin changes appeared in 43.8%. Central nervous system complications (a problem in the brain) appeared in 18.8%. 10 cases (20.8%) recovered vision fully, 8 cases (16.7%) recovered partly, and the rest did not recover. The authors conclude that practice varies enormously between clinics, and no treatment has proved consistently effective +- Evidence grade: B +- Sources: Beleznay K et al. (2019). Update on Avoiding and Treating Blindness From Fillers: A Recent Review of the World Literature. Aesthet Surg J. +- Notes: These numbers were counted case by case from cases reported in journals. They contain only the number of people who were hurt, not the total number of injections given. So you cannot work out "how likely one injection is to cause blindness"; you can only see which sites go wrong most often and how little can be done afterwards. That is also why the evidence is graded B. What a proper institution does to lower risk is routine anyway: first map the vessels under the place to be injected, use a blunt needle rather than a sharp one, inject small amounts while withdrawing, and have hyaluronidase ready beforehand. Every one of them requires the person doing it to be a trained doctor, which is the reason not to use an unofficial channel. Before you go, check it against item 2 in this section (check the clinic's licence and the attending physician). + +### 4. Do not buy diet pills, diet coffee, slimming sweets or enzyme plums that promise "fast weight loss" + +- Cost: No money. It also saves the money you would spend on these products. The hard part is accepting that there is no such thing as dropping several jin in a week. +- In plain language: The substance most often found smuggled into these products is sibutramine. It was originally a diet drug and was stopped in China in 2010. Overweight people with heart disease or diabetes who took it for more than three years had about 30% more non-fatal heart attacks and about 36% more strokes. They lost only 1.7 kg more. By the time a spot check finds it, you have already swallowed it. +- Benefit: The SCOUT trial enrolled 9,804 people. All were over 55, overweight or obese, and had cardiovascular disease or type 2 diabetes. They were randomised into a group taking the drug and a group taking none, with an average of 3.4 years of treatment. The primary endpoint was a set of bad events agreed in advance: non-fatal heart attack, non-fatal stroke, cardiac arrest survived, and cardiovascular death. These occurred at 11.4% vs 10.0%. HR 1.16 (95% CI 1.03–1.31, P=0.02), so the people taking the drug were about sixteen per cent more likely to have one of them. The true value lies roughly between 3% higher and 31% higher. Non-fatal heart attack 4.1% vs 3.2%, HR 1.28 (1.04–1.57), about thirty per cent higher. Non-fatal stroke 2.6% vs 1.9%, HR 1.36 (1.04–1.77), about thirty-six per cent higher. Cardiovascular death and all-cause death did not increase; all-cause means every cause added together. After the randomisation, the sibutramine group lost an average of 1.7 kg more. The former State Food and Drug Administration held, on the basis of its safety assessment, that "the risks of weight-loss treatment outweigh the benefits", and in month 10 of 2010 ordered the production, sale and use of sibutramine preparations and the raw drug to stop. Both sibutramine and phenolphthalein are banned from food, health food included +- Evidence grade: A +- Sources: James WPT et al. (2010). Effect of sibutramine on cardiovascular outcomes in overweight and obese subjects. NEJM. ；Tianjin Municipal Commission for Market Regulation (天津市市场监督管理委员会). Beware of "sibutramine hydrochloride" and "phenolphthalein" illegally added to some foods（警惕部分食品中违法添加「盐酸西布曲明」和「酚酞」）. +- Notes: SCOUT enrolled people who already had heart disease or diabetes, a high-risk group. The trial did not show that healthy young people would have the same events. But because nobody can tell who is high-risk, the regulator simply took it off the market. Another substance often smuggled in is phenolphthalein; it was originally a laxative, and taking too much of it over a long period upsets the body's electrolytes and can bring on an arrhythmia. How to tell: a proper diet drug is a prescription drug, and its approval number can be looked up at the pharmacy and the hospital. Anything sold under the flag of "coffee", "sweets", "plums" or "meal replacement" that promises to drop several jin in a week should by definition have no drug effect; if it has one, something has been added. The original text of 国食药监办〔2010〕432 号 is on the National Medical Products Administration website; opening it on this machine has always returned error 412, so a provincial regulator's page repeating it is cited instead. The two sentences "the risks outweigh the benefits" and "production, sale and use stopped in October 2010" can be checked word for word there. + +### 5. Do not use anabolic steroids ("muscle-building injections", "oral steroids") to build muscle + +- Cost: No money, and it saves the money for the drugs. The hard part is accepting that muscle grows slowly. +- In plain language: Men who test positive for steroids in gym doping tests die at 3 times the rate of others their age, and go to hospital more than twice as often each year. Acne, gynecomastia and erectile dysfunction each hit at least one in ten users. In China these are prescription drugs, so any channel able to sell them to you is itself breaking the law. +- Benefit: Denmark went back through its records and gave every user a same-age control. The users were 545 men who tested positive for androgenic steroids in gym doping tests between 2006-01-03 and 2018-03-01, matched with 5,450 controls. A second run checked them again against 644 people who refused to be tested and 6,440 controls. The results: users' mortality was 3 times that of the controls, HR 3.0 (95% CI 1.3–7.0), with the true multiple lying roughly between 1.3 and 7. The median number of hospital admissions or visits a year was 0.81 vs 0.36 (P<0.0001). The median means half the people are above that number and half below it. That P value means the difference is almost impossible to be chance. Acne (spots), gynecomastia and erectile dysfunction each affected more than 10% of users, that is at least one in ten users, and all were clearly higher than in the controls (P<0.0001). The second batch of records gave the same result. Article seven of the Regulations on Anti-Doping (《反兴奋剂条例》): the state applies strict control to the banned substances listed on the doping list, and no entity or individual may illegally produce, sell, import or export them. The same regulations put anabolic agents and peptide hormones under prescription-only control, that is, they may be bought only against a doctor's prescription; anabolic agents is what the regulations call this class of drug +- Evidence grade: A +- Sources: Horwitz H et al. (2019). Health consequences of androgenic anabolic steroid use. J Intern Med. ；State Council (国务院) (2004). Regulations on Anti-Doping（反兴奋剂条例）(State Council Order No. 398, articles seven and nine). +- Notes:Disputed: This is a study that followed a group of people and looked at what happened to them, not a randomised trial. Users may take more risks in other ways as well, so the figure of 3 times cannot all be laid at the drug's door. The true multiple may be as low as 1.3 or as high as 7.0, a very wide range. But "acne, gynecomastia and erectile dysfunction each hit more than a tenth of users" are side effects you can see with your eyes, and they are much less affected by this bias. In regulations these drugs are written as "anabolic agents"; those four Chinese characters mean this drug. Ordinary protein powder and creatine are not covered by this item. The trouble is that products claiming to build muscle very fast may smuggle ingredients in. The way to tell is the same as for the diet products (item 4, do not buy diet pills that promise fast weight loss): a proper drug is a prescription drug with an approval number you can check. Anything sold as "mass gainer" or "nutrition product" that promises results in a few weeks, if it really works, has had something added. + +### 6. If you are going to take diet pills, get a prescription at the hospital; do not buy from an online shop that ships without one + +- Cost: One registration, plus the trip there and back. A few dozen yuan, half a day. +- In plain language: A prescription drug must be sold against a prescription, and that applies online too: there must be a real prescription and real-name registration. A shop that does not ask, wants no prescription and ships the moment you order is either not selling what it claims to sell, or is breaking the rules. When something goes wrong, nobody is answerable for your dose or your contraindications. A registration costs a few dozen yuan, and a doctor then takes that gate for you. +- Benefit: The Measures for the Administration of the Quality Supervision of Drug Distribution and Use (《药品经营和使用质量监督管理办法》) require retail drug businesses to follow the classified management system for prescription and non-prescription drugs. That system divides medicines into two kinds: one may only be bought against a doctor's prescription, the other may be bought on your own. A pharmacy must sell prescription drugs against a prescription as required, and keep the prescription for no less than five years. When the pharmacist or other pharmaceutical technician is off duty, a notice must be displayed, and prescription drugs may not be sold without their review. Prescription drugs may not be sold on open shelves, that is, left on a shelf for customers to take themselves. Article nine of the Measures for the Supervision and Administration of Online Drug Sales (《药品网络销售监督管理办法》) provides that prescription drugs sold to individuals over the network must come from a genuine and reliable prescription, and real-name registration must be used. Article ten requires risk warning information to be shown prominently under each drug's display page. Before the prescription has passed review, the instructions may not be shown and no purchase service may be offered; until the prescription has been reviewed the page should not even offer a buy button. Article eight provides that drugs under special state control may not be sold over the network. Such drugs include vaccines, blood products, narcotic drugs, psychotropic drugs, toxic drugs for medical use, radioactive drugs, and drug-type precursor chemicals +- Evidence grade: B +- Sources: State Administration for Market Regulation (国家市场监督管理总局) (2023). Measures for the Administration of the Quality Supervision of Drug Distribution and Use（药品经营和使用质量监督管理办法）(SAMR Order No. 84). ；State Administration for Market Regulation (国家市场监督管理总局) (2022). Measures for the Supervision and Administration of Online Drug Sales（药品网络销售监督管理办法）(SAMR Order No. 58, articles eight to ten). +- Notes: The law can be checked word for word, but nobody has run a quantitative study on how much fewer goes wrong when you take a medicine as prescribed than when you buy it and take it yourself, so the grade is B for "hard to quantify". The steroids and the sex hormones covered in items 5 and 7 in this section follow the same rules. When it really is time to rely on medicine to lose weight, the doctor first looks at your body mass index, whether you have other illnesses, and whether there is any reason you cannot take the drug, then decides whether to prescribe and which one. That judgement cannot be bought. + +### 7. Use sex-hormone drugs only when a doctor prescribes them and you have regular check-ups; do not buy online, do not raise the dose yourself + +- Cost: Regular registrations, plus blood tests for the check-ups. A few hundred yuan a year, taking up several half-days. +- In plain language: This item does not discuss whether you should use them, only how. Transgender women who use oestrogen get blood clots in their vessels noticeably more often than people their age: about 4 extra per 1000 people over two years, and about 17 extra over eight years. Risks like these have to be watched by regular check-ups; with drugs bought online and doses adjusted by yourself, nobody runs those checks for you. +- Benefit: The researchers followed a group of people taken from the electronic medical records of the Kaiser Permanente health group: 2,842 transgender women and 2,118 transgender men. Average follow-up was 4.0 years and 3.6 years respectively. Each person was matched with about 10 times as many cisgender controls on year of birth, race, clinic site and time of entry, giving 48,686 cisgender men and 48,775 cisgender women in all. Venous thromboembolism (a clot forming in a vessel and blocking it) was more common in transgender women than in the controls. Compared with cisgender men, at 2 years it was 4.1 per 1000 people (95% CI 1.6–6.7). At 8 years it was 16.7 per 1000 (6.4–27.5). Worked out, that is 4.1 extra cases per 1000 people over two years and 16.7 extra over eight years. Compared with cisgender women, the two figures were 3.4 (1.1–5.6) and 13.7 (4.1–22.7). Ischaemic stroke (the kind of stroke caused by a blocked vessel) and heart attack were close across the groups overall. But among transgender women who only started hormone therapy during the follow-up period, the differences in venous thrombosis and ischaemic stroke were more marked. The evidence on transgender men is not enough to draw a conclusion. The authors conclude that "these results may suggest a need for long-term vigilance about the vascular side effects of oestrogen in transgender hormone therapy" +- Evidence grade: A +- Sources: Getahun D et al. (2018). Cross-sex Hormones and Acute Cardiovascular Events in Transgender Persons: A Cohort Study. Ann Intern Med. ；State Administration for Market Regulation (国家市场监督管理总局) (2022). Measures for the Supervision and Administration of Online Drug Sales（药品网络销售监督管理办法）(SAMR Order No. 58, article nine). +- Notes: The researchers set out the weaknesses of the study themselves: how many people bought their drugs outside the hospital system they cannot see, so the line of "was the drug taken or not" was never clean. That is all the more reason to put the use of the drug inside regular follow-up. This study only looks at whether the vessels had an accident; it does not answer whether hormone therapy should be used or what it buys you, and that is another body of evidence. Every place where sex hormones are used is the same, including the pill, hormone replacement around menopause, and testosterone replacement. Anyone who buys online, doses from an internet post, and never has clotting, liver and kidney markers checked, runs the risk in the stretch where nobody is watching, and that has nothing to do with why the drug was taken. The proper route in China is the endocrinology department, or the relevant specialty, of a proper medical institution: assessment, prescription, follow-up. + +### 8. If "thinking you are ugly" has got to the point of checking the mirror repeatedly and wanting to change repeatedly, get a body-image assessment before talking about surgery + +- Cost: One registration at a psychiatry department or a clinical psychology department, a few dozen yuan. The hard part is making that trip before you book the operation. +- In plain language: About 15% of people going for plastic surgery have body dysmorphic disorder, which means magnifying a flaw others can barely see until it affects your life. About 13% of people who go to dermatology wanting to look better have it, and in both groups more than seven in ten are women. Someone like this finishes one area and then fixes on the next; surgery does not solve it. Get one registration first to see whether this is you, then decide about the operation. +- Benefit: Following the rule, 33 studies were gathered and then pooled. Among plastic surgery patients, the prevalence of body dysmorphic disorder (magnifying a flaw others can barely see until it affects your life) was 15.04%, with individual studies ranging from 2.21% to 56.67%. In headcount, about 15 in every 100 people who go for surgery fit the description, while individual studies range from 2 to 57. This group had an average age of 34.54±12.41 years, the second figure being the margin either side, and 74.38% were women. Dermatology patients were 12.65%, with individual studies from 4.52% to 35.16%, an average age of 27.79±9.03 years, and 76.09% women. The authors suggest that plastic surgeons and dermatologists assess proactively, identify the high-risk people, and arrange for several departments to manage them together +- Evidence grade: B +- Sources: Ribeiro RVE (2017). Prevalence of Body Dysmorphic Disorder in Plastic Surgery and Dermatology Patients: A Systematic Review with Meta-Analysis. Aesthetic Plast Surg. +- Notes: The proportions measured by the individual studies run from 2.21% all the way to 56.67%, which is an enormous spread. The reason is that the questionnaires and the populations surveyed were all different. So all you can read off it is "this proportion is not low, and it is worth screening once first"; you cannot take 15.04% as an exact figure, and that is also why it is graded B. The great majority of people who want cosmetic work have no psychological problem. This item is aimed at the small group for whom it is "repeated, persistent, and already affecting work and social life"; surgery does not solve their problem. The signs to judge by are these: spending a long part of the day checking or covering one area, and so avoiding social contact or photographs, and finishing one area only to fix on the next. diff --git "a/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" "b/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" new file mode 100644 index 00000000..d74082ab --- /dev/null +++ "b/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" @@ -0,0 +1,130 @@ +[← 回总目录](../README.md) + +# 29. 遭遇重大打击之后 + +亲人去世、被告知重病、失业、离婚。这几件事在研究里有个共同点：出事的不只是心情，还有身体。风险最高的那一段，集中在头几周到头一年。这一节不讲怎么「走出来」，只讲这段时间里哪些事必须有人盯住，哪些决定必须往后放，什么时候该去挂号。多数条目算的是死亡概率。谈花钱和待遇的四条算的是钱：别一上来就花钱做哀伤咨询（第 9 条）、打 12356 和挂心理门诊（第 11 条）、不可逆的大决定往后推（第 12 条）、别拿死当还债的办法（第 13 条）。两类数字不互相换算。没有亲人也没有朋友可找的，「找个人盯住」这句话怎么落地见第 6 条。 + +亲人去世当时的手续、遗体和该取的钱见第 25 节。老人生前的监护与遗嘱见第 17 节。失业能领什么见第 7 节。情绪低落时的运动与光照见第 3 节和第 22 节。有自杀念头时先打 12356 见第 1 节第 25 条、念头的时间尺度见第 1 节第 32 条、救回来之后的后遗症见第 1 节第 33 条。这些本节都不重复。 + +[← Back to the contents](../README.md) + +# 29. After a severe blow + +A relative has died, you have been told you have a serious illness, you have lost your job, you have divorced. These events share one thing in the research: what suffers is not only the mood but the body. The risk is at its highest from the first weeks to the first year. This section does not talk about how to "come through" it. It covers what has to be watched during this period, which decisions have to be put off, and when to go and register with a hospital. Most items here measure the probability of death. The four that talk about money and benefits measure money instead: do not go straight to paid grief counselling (item 9), calling 12356 and registering for a psychological outpatient clinic (item 11), putting irreversible big decisions off (item 12), and not using death as a way to pay off debts (item 13). The two kinds of numbers are not converted into one another. If you have neither relatives nor friends to turn to, how "get somebody to keep an eye on you" actually works is in item 6. + +The procedures at the time a relative dies, the body, and the money you can claim are in section 25. Guardianship and wills before an older person's death are in section 17. What you can claim after losing your job is in section 7. Exercise and light when you are low are in section 3 and section 22. If you have thoughts of suicide, call 12356 first, see item 25 in section 1, for the time scale of those thoughts see item 32 in section 1, and for the after-effects once someone is saved see item 33 in section 1. None of that is repeated here. + +### 1. In the first days after a relative dies, do not stay on your own, and if you have heart disease, high blood pressure or diabetes keep taking your medication; if you have pressure across your chest, call 120 + +- Cost: No money. Do not stay on your own for the next few days, and keep taking your medication. +- In plain language: In the first day after a relative dies, the probability of a heart attack is 21.1 times the usual, and it falls day by day after that. Among people aged 60 to 89, the proportion who have a heart attack or a stroke within 30 days of their spouse dying roughly doubles, and falls back after a month. If you already have cardiovascular disease, do not stop your medication in those days. Pressure across the chest, breath you cannot get, weakness down one side of the body: call 120 immediately. +- Benefit: 1985 patients with acute myocardial infarction, compared within the same person across the different windows before and after onset. 270 of them (13.6%) had lost an important person in the 6 months before onset, and 19 had lost one within 1 day before the heart attack. Within 24 hours of an important person's death, the rate of acute myocardial infarction rose 21.1-fold (95% CI 13.1–34.1, so the true multiple is somewhere between 13 and 34). It fell day by day after that. In terms of actual numbers: among people with a 5% ten-year risk of a heart attack, there is 1 extra case for every 1394 people who have just lost a relative. Among people with a 20% risk, 1 extra case for every 320. A second British study compared 30447 widowed people aged 60 to 89 with 83588 matched controls. Within 30 days of bereavement, 50 people in the widowed group (0.16%) had a heart attack or a stroke, against 67 in the control group (0.08%). IRR 2.20 (1.52–3.15), that is about 2.2 times the risk in the control group. For heart attacks, 2.14 (1.20–3.81), about 2.1 times. For strokes, 2.40 (1.22–4.71), about 2.4 times. The effect weakens after 30 days. Two other conditions were also looked at within 90 days. One was non-myocardial-infarction acute coronary syndrome, where the heart's blood supply fails suddenly but not far enough to count as a heart attack, IRR 2.20 (1.12–4.29). The other was pulmonary embolism, where a clot blocks a vessel in the lung, IRR 2.37 (1.18–4.75) +- Evidence grade: A +- Sources: Mostofsky E, et al. (2012). Risk of acute myocardial infarction after the death of a significant person in one's life: the Determinants of Myocardial Infarction Onset Study. Circulation. ；Carey IM, et al. (2014). Increased risk of acute cardiovascular events after partner bereavement: a matched cohort study. JAMA Internal Medicine. +- Notes: Hand the pill box to somebody else to keep, and do not rely on your own memory. Do not drive long distances in those days. Do not try to carry it with alcohol; the account for alcohol is in section 2. How to recognise a heart attack or a stroke and how to call an ambulance, see section 13. Both studies followed groups of people and looked back, so they cannot separate how many of them were going to have an event anyway. But both point the same way and cover the same period. If there is nobody to hand the pill box to and you are certain to be alone in the house for those days, see item 6 in this section. + +### 2. In the week you are told you have a serious illness, do not go for the results on your own, and put off every decision other than treatment + +- Cost: No money. Have somebody go with you to pick up the results. Push back by a few days any big decision other than treatment. +- In plain language: In the first week after a cancer diagnosis, the probability of suicide is 12.6 times that of people without a diagnosis, and the probability of dying of cardiovascular disease is 5.6 times. By the first year the suicide risk has fallen to 3.1 times, and the harder the cancer is to treat the more marked this is. So take somebody with you the day you collect the results, and do not stay on your own at home afterwards. Things like resigning or selling the house, leave until your next appointment. +- Benefit: Swedish national registers followed more than 6,070,000 people (6,073,240 people) over 1991 to 2006. Compared with people who had not been diagnosed with cancer, the relative risk of suicide in the first week after diagnosis was 12.6 (95% CI 8.6–17.8), that is about 12.6 times. The true multiple is somewhere between 8.6 and 17.8. The incidence in that week is 2.50/1000 person-years, that is 2.5 cases per 1000 people per year. By the first year it falls to 3.1 (2.7–3.5, 0.60/1000 person-years), about 3.1 times. Cardiovascular death in the first week 5.6 (5.2–5.9, 116.80/1000 person-years), about 5.6 times. The first 4 weeks 3.3 (3.1–3.4, 65.81/1000 person-years). The rise falls off quickly within the first year after diagnosis. The harder the cancer is to treat and the lower the chance of surviving it, the more marked the rise is. Comparing the same group of patients across different windows gives the same conclusion +- Evidence grade: A +- Sources: Fang F, et al. (2012). Suicide and cardiovascular death after a cancer diagnosis. New England Journal of Medicine. +- Notes: These are data on cancer patients. There is no source of the same weight for other serious diagnoses. The number of people involved is actually small: a suicide rate of 2.50/1000 person-years in the first week, that is 2.5 cases per 1000 people per year. This is written down not to frighten you but to show that somebody has to be there in the days just after you find out. Long-term management after diagnosis, reimbursement and follow-up are in section 16 and section 24. If you have nobody to go with, see item 6 in this section (put the neighbours and the community list in place of "that person"). + +### 3. After losing your job, fix your daily routine, your medical insurance and the rhythm of your job hunt first; do not spend the whole day at home + +- Cost: No money. Go and register as unemployed and claim unemployment insurance benefits first, then put job hunting into a fixed schedule. The hard part is getting up on time when nobody is making you. +- In plain language: An unemployed person afterwards has about a sixty per cent higher probability of dying than a person in work, the rise is more marked in men and in the not very old, and it is highest in the first ten years. Take out the effect of smoking and drinking and the rise is about a quarter smaller, and that part is in your own hands. Go and register as unemployed and claim unemployment insurance benefits first; while you are receiving them, medical insurance is paid by the unemployment insurance fund. +- Benefit: 42 studies, 235 estimates of the risk of death and more than 20 million people pooled together. After taking out the effect of age and similar conditions, the average HR for all-cause death among the unemployed is 1.63, about sixty per cent higher than for people in work. Men are higher than women. The effect is marked for people in the early and middle part of their working life and smaller in the later part. The risk is highest in the first 10 years of follow-up and falls after that. In the subset of studies that also took health behaviour such as smoking and drinking into account, the average HR is 24% lower, that is the calculated rise is about a quarter smaller +- Evidence grade: A +- Sources: Roelfs DJ, et al. (2011). Losing life and livelihood: a systematic review and meta-analysis of unemployment and all-cause mortality. Social Science & Medicine. +- Notes: This is a study that followed groups of people and looked back, so it cannot tell cause from effect. People who were already in poor health were also more likely to lose their job. But "after taking out the effect of lifestyle habits the effect is 24% lower" gives you somewhere to start. What you can claim after losing your job, where to claim it and how to keep social insurance going, see section 7. What compensation you are owed on dismissal, and not to sign a voluntary resignation, see section 19. + +### 4. If a relative died by suicide, accident or murder, do not expect to tough it out; go and look for professional help + +- Cost: Calling 12356 costs nothing. To see a doctor, registering for a psychological outpatient clinic costs what an ordinary outpatient appointment costs. The hard part is being the one who asks. +- In plain language: Among people whose spouse died by suicide, the chance of a mental disorder being diagnosed within five years is about eighty per cent higher than in the general population for men, and seventy per cent higher for women. Compared with people whose spouse died of another cause, it is seventy per cent higher up to double. People bereaved this way more often feel they have been pushed away by others, and feel ashamed. That is the reason to go and find someone yourself, instead of waiting until you feel better. +- Benefit: Danish national registers followed 6.7 million adults over 1980 to 2014, totalling more than 136 million person-years (the years all of them were followed, added together). For people widowed because their spouse died by suicide, the risk of a newly diagnosed mental disorder within 5 years, compared with the general population, was IRR 1.8 in men (95% CI 1.6–2.0), that is about eighty per cent higher. Women 1.7 (1.6–1.8), about seventy per cent higher. Compared with people widowed by other causes of death, IRR 1.7 in men (1.5–1.9) and 2.0 in women (1.9–2.2), that is seventy per cent higher up to double. In this group, suicidal behaviour, mortality and the proportion needing municipal assistance were all higher as well. Physical conditions such as cirrhosis of the liver and sleep disorders also carried a raised risk. A separate 57 studies are summarised one by one and not pooled. Bereavement by suicide is associated with four things. First, a raised suicide risk in the partner. Second, a raised risk that a parent is admitted to a psychiatric ward after a child has committed suicide. Third, a raised suicide risk in the mother after an adult child has committed suicide. Fourth, a raised risk of depression in a child after a parent has committed suicide. Compared with people bereaved by other violent deaths, people bereaved by suicide more often say that they felt pushed away by others, and felt ashamed +- Evidence grade: A +- Sources: Erlangsen A, et al. (2017). Association Between Spousal Suicide and Mental, Physical, and Social Health Outcomes: A Longitudinal and Nationwide Register-Based Study. JAMA Psychiatry. ；Pitman A, et al. (2014). Effects of suicide bereavement on mental health and suicide risk. The Lancet Psychiatry. ；Kristensen P 等 (2012). Bereavement and mental health after sudden and violent losses: a review. Psychiatry. （the review referred to in the notes） +- Notes: A sudden, violent death means an accident, a murder, a disaster. After a death like this, the proportions of post-traumatic stress disorder, depression and grief that has got stuck are all higher than after an illness. Recovery is slower as well. What to do when you have thoughts of suicide yourself, see item 25 in section 1. + +### 5. In the first six months after a spouse dies, hand meals, sleep, the medication for a chronic illness and the follow-up appointments to one fixed person to keep an eye on them + +- Cost: No money. Find a specific person, a child, a brother or sister, or a friend, and have them come round or telephone on a set schedule. The hard part is asking someone to take this on. +- In plain language: Within six months of being widowed, the probability of dying in the same period is about forty per cent higher than for someone who has not been bereaved. After six months it falls to about fifteen per cent higher. Men are about twenty-three per cent higher; the figure for women is so small it may be chance. This gap is not something you can carry by telling yourself to think positive. What to do is pin the medication, the meals and the follow-up appointments onto one specific person, and not rely on your own memory. +- Benefit: 15 studies that registered first and then followed people forward, 2,263,888 people pooled together. When bereavement has lasted less than 6 months, the RR of death is 1.41 (95% CI 1.26–1.57), that is about forty per cent higher than for people who have not been bereaved. After 6 months it is 1.14 (1.10–1.18), down to about fifteen per cent higher. Men RR 1.23 (1.18–1.28), about twenty-three per cent higher. Women 1.04 (1.00–1.08, not statistically significant, that is this difference may be chance). No clear difference was seen between people under 65 and people aged 65 and over. A separate analysis pooled 123 papers, 1377 risk estimates and more than 500 million people. After taking out differences in conditions, HR 1.23 (1.19–1.28), that is about twenty-three per cent higher overall. Men are still higher than women. Men 1.27 (1.19–1.35), twenty-seven per cent higher. Women 1.15 (1.08–1.22), fifteen per cent higher +- Evidence grade: A +- Sources: Moon JR, et al. (2011). Widowhood and mortality: a meta-analysis. PLoS ONE. ；Shor E, et al. (2012). Widowhood and mortality: a meta-analysis and meta-regression. Demography. +- Notes: This is a study that followed groups of people and looked back; other factors cannot be taken out of it. A married couple's lifestyle and illnesses are similar to begin with. But the pattern that "the first six months are far higher than the ones after" is the same in two independent meta-analyses, and it is worth treating as a window of care. Men's risk is higher, and they also ask for help less often. If there is a widowed older person at home, sort out the accounts, guardianship and wills together as in section 17. If you cannot find such a person (you live alone, you have no children left, or your children are gone too), hand this to the community and to your phone, see item 6 in this section. +### 6. If you have neither relatives nor friends, replace "the person keeping an eye on you" with three things: a neighbour who can get into your home, the community's visiting list, and an emergency contact in your phone + +- Cost: No money. The hard part is speaking up, telling a neighbour and the residents' committee "I live on my own, something has just happened at home". +- In plain language: With neither relatives nor friends, there are two things to look after: that somebody finds out if something goes wrong, and that you take your medication and eat on time. The medication and the meals rest on you: write a list, set an alarm, do not rely on memory. Somebody finding out rests on three things: leave a spare key with a neighbour, ask the residents' committee whether you can be put on the visiting and care list, and set up an emergency contact in your phone and not switch it off. +- Benefit: 44,573 outpatients, all of whom had or were at risk of atherothrombosis (the kind of condition where plaques grow inside the arteries and clots form easily). Doctors in several countries registered them into the international REACH project, entry from 2003-12 to 2004-12, and they were followed for 4 years. 8,594 of them (19%) lived alone. Among those living alone, all-cause death over the 4 years was 14.1%, against 11.1% for those living with others. Cardiovascular death 8.6% against 6.8%. Both log-rank P<0.01, so the difference is unlikely to be chance. The gap differs by age group (interaction P=0.03). At 45 to 65, all-cause death living alone against living with others was 7.7% against 5.7%. After taking out other conditions, HR 1.24 (95% CI 1.01–1.51), that is about twenty-four per cent higher. At 66 to 80 it was 13.2% against 12.3%, HR 1.12 (1.01–1.26), about ten per cent higher. Above 80 it was 24.6% against 28.4%, HR 0.92 (0.79–1.06), no rise seen. A 2022 opinion from the Ministry of Civil Affairs (民政部) and ten other departments provides for visiting and care services. The subjects are elderly people who live alone, have empty-nest households, have children left behind, are disabled or severely disabled, and families in special family-planning circumstances. The forms of service are 「定期上门入户、电话视频、远程监测」, that is regular visits to the home, telephone and video, and remote monitoring. Townships (subdistricts) do the headcount and the screening, village (residents') committees help; they record the basic information and 「接受探访关爱服务的意愿」, the willingness to receive visiting and care services, and bring it into grassroots grid management, which means each area is assigned to a named person to keep an eye on it. By the end of 2025 「确保特殊困难老年人月探访率达到 100%」, that is make sure the monthly visiting rate for elderly people in special difficulty reaches 100%, and 「对失能的特殊困难老年人，每月探访关爱不少于一次」, at least one visit a month for disabled elderly people in special difficulty. The document also advocates that households install smart call systems, smart water meters, health monitoring products and elderly monitoring devices, so that 「一旦监测异常，能够及时预警，并同步向紧急联系人发送提示信息」, that is as soon as monitoring shows an abnormality an alert can be raised in time and a message sent at the same time to the emergency contact. Where service staff find an emergency they 「应当第一时间协助拨打紧急求助电话」, that is they shall assist in dialling the emergency number at the first opportunity +- Evidence grade: B +- Sources: Udell JA, et al. (2012). Living alone and cardiovascular risk in outpatients at risk of or with atherothrombosis. Archives of Internal Medicine. ；Ministry of Civil Affairs (民政部) and ten other departments (2022). Guiding Opinions on Carrying Out Visiting and Care Services for Elderly People in Special Difficulty（关于开展特殊困难老年人探访关爱服务的指导意见）(民发〔2022〕73 号). +- Notes: How to do the three things. First, give a spare key or the temporary code of your door lock to a neighbour or to the building management, and say "I am on my own for the next few days". You do not need to be close to them; you only need somebody to know you are in the house. Second, tell the community residents' committee that you live alone and ask whether you can be put on the visiting and care list. Among the elderly, those who live alone, have empty-nest households or have no children left are named official subjects for the service; while you are there, ask whether your area has free smart call devices or smart water meters to install. Third, set up an emergency contact and a medical emergency card in your phone, and do not silence it or switch it off for those days. There are two reasons for the B. First, the link between living alone and mortality comes from a study that followed a group and looked back, so the reverse explanation, that people who were already in poor health and worse off are more likely to live alone, cannot be taken out; the original paper itself says the result remains to be confirmed. Second, the visiting and care document covers elderly people only. There is no matching scheme for anyone under 60. What you can rely on is one line in that document, which requires grid workers and social workers to 「及时发现家庭变故、失业、失学等心理危机风险」, that is identify in good time psychological crisis risks such as family upheaval, job loss and dropping out of school (see item 11 in this section, call 12356 and have somebody to talk to). Why "somebody in the room" is worth so much, see item 1 in section 13 (pressing on the chest). Of cardiac arrests outside hospital, 79.2% happen at home. Where somebody is there doing compressions, 16.1% are discharged alive, against 3.9% where nobody is. Nobody in the room puts you in the lower band. Two more things you can do in passing. A funeral itself means going to the funeral home, the community health service, the police and the housing provident fund, several counters in all (see section 25). Put that into a fixed schedule of one trip out a day, and the arrangements for going out during those days are already there. The hotline can be called as often as you like; it is not a single permitted call (see item 11 in this section, call 12356 and have somebody to talk to). "Living alone" is itself a long-term risk factor. The mortality ratio for living alone is 1.32, about thirty per cent higher, see item 10 in section 22 (treating regular contact with other people as a health expense). But this item only covers the next few days; the long-term account is a separate one. + +### 7. After a parent dies, tell the child the truth, let them attend the farewell, and do not send them somewhere else to "change the scenery" + +- Cost: No money. The hard part is that you are in pain yourself and still have to say it clearly to the child. +- In plain language: Children who lose their father or mother before the age of 18 have about a fifty per cent higher probability of dying in early adulthood than children who have not been through it. Where the parent died by accident or suicide it is about eighty per cent higher, and where the parent died of illness about thirty per cent higher. What the family can do is tell him the truth, let him attend the farewell, not interrupt his schooling, and not send him to a relative's house to change the scenery. +- Benefit: The entire populations born in Denmark from 1968 to 2008 and in Sweden from 1973 to 2006 are counted. A sample of 89.3% of those born in Finland from 1987 to 2007 is added in. Altogether more than 7.3 million people. 189,094 of them lost a parent before the age of 18. All-cause mortality in this group of children was 50% higher than in those who had not been bereaved (mortality ratio 1.50, 95% CI 1.43–1.58, that is about 1.5 times). Where the parent died by accident, suicide or another unnatural cause it was 1.84 (1.71–2.00), about eighty per cent higher. Where the parent died of illness or another natural cause it was 1.33 (1.24–1.41), about thirty per cent higher. The figure is highest where the child's cause of death falls in the same category as the parent's +- Evidence grade: B +- Sources: Li J, et al. (2014). Mortality after parental death in childhood: a nationwide cohort study from three Nordic countries. PLoS Medicine. +- Notes: Do not make up "has gone away on a long trip" when you talk to the child; the arrangements for guardianship and support have to be put in writing. The grade is not B because the numbers are weak. This is a follow-up study of the entire populations of three countries, and the numbers themselves would support an A. It is B because between that number and the specific practice of "how to talk to the child and whether to let him attend the farewell" there is no direct trial evidence. The authors say so themselves: after bereavement there is a shortage of data on the parent-child relationship and on how children live. The increase contains both an inborn-constitution part and a long-term effect left behind by the bereavement. Naming a guardian and arranging property, see section 17. Child support and schooling, see section 18. + +### 8. If grief has gone six months and you are still stuck and cannot get through your days, register with psychiatry or clinical psychology + +- Cost: The registration fee plus the treatment. This kind of treatment runs in courses; in the trials it averaged 16 sessions over 19 weeks. The hard part is being the one who goes and registers. +- In plain language: About one in every ten bereaved adults gets stuck: six months or more of missing the person so much that they cannot eat, cannot work and cannot look after their children, going numb, and feeling that life has no meaning. This is not a matter of not being strong enough. It has diagnostic criteria and a treatment of its own. Targeted treatment has an effective rate of 51%, ordinary psychological therapy 28%, and it takes effect sooner as well. +- Benefit: 14 studies pooled together, in adults whose relative did not die by violence and who themselves had not been seen in psychiatry. The prevalence of prolonged grief disorder is 9.8% (95% CI 6.8–14.0), that is about 10 in every 100 bereaved adults. The older a person is, the higher the proportion. A separate study tested whether these diagnostic criteria are accurate. It interviewed 291 bereaved people three times, at 0–6, 6–12 and 12–24 months after the bereavement. The core symptom for diagnosis is intense longing for the dead person. At least five of these nine have to be added to it. The first five are emotional numbness, shock, feeling that life has no meaning, not trusting other people, and pain from this loss. The other four are difficulty accepting it, not recognising who you are now, avoiding reality, and being unable to go on with life. These symptoms have to be there every day, or be severe enough to stop you functioning. And they must still be there at least 6 months after the relative died, and still affect daily life. A separate trial with 95 people assigned at random to two groups gave 16 sessions over an average of 19 weeks. The effective rate for treatment of complicated grief was 51%. Ordinary interpersonal psychological therapy 28% (P=0.02, so the difference is unlikely to be chance). The former took effect sooner as well. Number needed to treat 4.3, that is on average every 4.3 people treated gives one extra person who improves +- Evidence grade: A +- Sources: Lundorff M, et al. (2017). Prevalence of prolonged grief disorder in adult bereavement: A systematic review and meta-analysis. Journal of Affective Disorders. ；Prigerson HG, et al. (2009). Prolonged grief disorder: Psychometric validation of criteria proposed for DSM-V and ICD-11. PLoS Medicine. ；Shear MK, et al. (2005). Treatment of complicated grief: a randomized controlled trial. JAMA. +- Notes: Do not count the months yourself. The real test is whether it is affecting eating, sleep, work and looking after the children; if it is, go and register. The departments are called psychiatry, psychology or clinical psychology, and every public psychiatric specialty hospital has them. This diagnosis goes under different names in different sources: complicated grief, and prolonged grief disorder. The standards also differ on how long it has to last: some say 6 months, some say 12. The 9.8% figure comes from 14 studies that differ quite a lot in method, and the authors warn that it may not carry over to other populations. + +### 9. Do not go straight to paid grief counselling; first check whether your grief really has got stuck (the symptoms in item 8) + +- Cost: No money, and you save the counselling fee instead. +- In plain language: Offering psychological counselling to everyone who has been bereaved brings very little benefit, and after a while it cannot be seen, because most people come through it on their own. But for people who are already clearly not coping, treatment works as well as it does for other psychological problems. So check against item 8 first (grief has got stuck, go to psychiatry), and only then spend the money. +- Benefit: 61 controlled studies pooled together. Offering psychological interventions to the bereaved in general brings very little benefit at the end of treatment, and after some time the benefit cannot be seen. Interventions that take in only the bereaved who are already clearly not coping work as well as psychological therapy for other psychological problems. Studies that do no screening beforehand do badly, because the group that received no intervention would have come through it on their own over time anyway. A separate 14 randomised trials pooled together: treatment given to people who already have a problem is clearly better than the control group both at the end of treatment and at later follow-up, and the effect grows over time. Preventive interventions given to people who do not yet have a problem do not work. A separate study followed 205 people from before the death of a spouse through to 6 and 18 months after it. This study identified five courses. They are ordinary grief, long-term grief, long-term depression, getting better during the bereavement, and resilience (being able to take it). Of these, the resilient course is the commonest and the ordinary-grief course relatively uncommon +- Evidence grade: A +- Sources: Currier JM, et al. (2008). The effectiveness of psychotherapeutic interventions for bereaved persons: a comprehensive quantitative review. Psychological Bulletin. ；Wittouck C, et al. (2011). The prevention and treatment of complicated grief: a meta-analysis. Clinical Psychology Review. ；Bonanno GA, et al. (2002). Resilience to loss and chronic grief: a prospective study from preloss to 18-months postloss. Journal of Personality and Social Psychology. ；Stroebe M 等 (2007). Health outcomes of bereavement. Lancet. （the review referred to in the notes） +- Notes: The point is "not everybody has to do it", not "nobody should do it". People at high risk of the kind covered by item 4 (a relative died by suicide, accident or murder) and people who have already got stuck, of the kind covered by item 8 (grief has got stuck, go to psychiatry), are exactly the ones for whom there is evidence that it should be done. The conclusion of a Lancet review says the same: grief is not an illness, and most people do not need professional intervention. Spend the effort on the people at high risk. And on those who already have complicated grief, depression after bereavement, or a stress disorder. +### 10. After you separate or divorce, do not let medical insurance lapse and do not lean on alcohol + +- Cost: No money. Keep the medical insurance going and stop the alcohol. The hard part is that this is a period when leaning on alcohol is easy. +- In plain language: People who are separated or divorced have a clearly higher probability of dying early than people who are still married, and the rise is bigger in men and in younger people. Divorce is not necessarily the cause; people who divorce may already differ in health and income. There are two things in your own control: do not let medical insurance or the check-ups lapse, and do not lean on tobacco and alcohol. +- Benefit: 32 studies that registered first and then followed people forward, pooled together, covering more than 6.5 million people, 160,000 deaths, more than 755,000 divorces and 11 countries. Compared with people still married, separated or divorced adults had a clearly higher risk of early death. The rise is more marked in men and in younger people than in women and in older people. The authors discuss four possible pathways. First, social selection: people who were already in worse health or worse circumstances are more likely to end up divorced. Second, loss of resources, where income and protection stop. Third, changed health behaviour, with smoking, drinking and sleep getting worse. Fourth, chronic psychological stress (long-term tension) +- Evidence grade: A +- Sources: Sbarra DA, et al. (2011). Divorce and Death: A Meta-Analysis and Research Agenda for Clinical, Social, and Health Psychology. Perspectives on Psychological Science. +- Notes: The original only says "clearly higher" and compares the groups; it gives no multiple that can be cited, so no figure is given here. The benefit is graded medium on the basis of "you know the direction, and there is no reduction you can apply a threshold to". Property, debts and bride price in a divorce, see section 8 and section 10. Arrangements for the children, see section 18. + +### 11. To have somebody to talk to, call 12356; minors and teenagers call 12355; to see a doctor, register for a psychological outpatient clinic + +- Cost: No money. Both hotlines are free. A psychological outpatient clinic is registered for like an ordinary outpatient appointment. +- In plain language: If you want somebody to talk to, call 12356. That is the national psychological assistance hotline, and it is free. Minors and teenagers have 12355, the teenage comprehensive service platform. If you want to see a doctor, register for the psychological outpatient clinic of psychiatry, psychology or clinical psychology. Where there is family upheaval, job loss or dropping out of school, the community grid worker and the social worker are supposed to spot it themselves. +- Benefit: A 2026 plan from the National Health Commission (国家卫生健康委) and 25 other departments requires the working system of the 12356 psychological assistance hotline to be improved. Stable operation has to be guaranteed and the connection rate raised. A provincial 12356 information platform is to be built. The mechanism linking 12356 with service hotlines such as 110 is to be strengthened. The 12355 teenage comprehensive service platform is to be built up and used well, working together with 12356. There are three targets for 2030. First, 80% or more of administrative villages and urban communities are to have a psychological counselling room or a social work room and provide psychological services. Second, among the public psychiatric specialty hospitals at each level and the hospitals of each county (district), at least 1 is to provide psychological outpatient services. Third, every province is to have a provincial 12356 psychological assistance platform. The workforce for psychological crisis intervention and psychological assistance is made up of psychiatrists, psychological therapy technicians, psychological counsellors, social work professionals and volunteers. The plan also requires three kinds of person to keep an eye out and find problems early. These three are grid workers (the staff responsible for an area within the community), social work professionals and volunteers. They are to identify in good time psychological crisis risks such as family upheaval, job loss and dropping out of school (nationwide, 2026-3) +- Evidence grade: B +- Sources: National Health Commission (国家卫生健康委) and 25 other departments (2026). Notice on Issuing the Implementation Plan for a Sound Social Psychological Service System and Crisis Intervention Mechanism（关于印发健全社会心理服务体系和危机干预机制实施方案的通知）(国卫医政发〔2026〕8 号). +- Notes: The grade is B because the counselling rooms in villages and communities and the psychological outpatient clinics at county level are targets for 2030, not things every place has now. 12356 and 12355, on the other hand, you can call today. When 12356 opened, and how long it is answered each day, see item 25 in section 1. Psychological counselling and psychological therapy are not the same thing. Only a psychiatrist can make a diagnosis or write a prescription, and the treatment has to be done at a qualified medical institution. The "healing" and "spiritual counselling" sold by the hour on the market are not part of this system. + +### 12. In the first three months after a shock, put every irreversible big decision off + +- Cost: No money. You are only putting the decision off, and you may miss some opportunities because of it. The hard part is that at the time it feels as though the opportunity will not wait. +- In plain language: In this period your judgement of every "opportunity" is at its worst. People who say they can reach the person who died, who say they can get your money back for you, who tell you to resign and start again in another city, who want you to go into business with them, will all come to you. Anything over one month's income, or where the cost of changing your mind is high, put off by three months. Say it out loud first to somebody with no stake in it, and do not decide on the same day. +- Benefit: Avoid disposing of property in a way you cannot undo during the period just after a blow, when judgement is at its worst. These things include selling the house, resigning, investing a lump-sum pension or compensation payment in one go, remarrying in a hurry, and standing surety for somebody else. The amounts involved in decisions of this kind usually run from tens of thousands of yuan to the price of one flat, and the benefit is therefore graded medium. Official documents also list "family upheaval, job loss and dropping out of school" as psychological crisis risks that the community is to screen for proactively, so the state a person is in during this period is already treated as one that needs help from outside +- Evidence grade: C +- Sources: The author's own experience, no direct literature; the wording "family upheaval, job loss and dropping out of school" as psychological crisis risks is from National Health Commission (国家卫生健康委) and 25 other departments (2026). Implementation Plan for a Sound Social Psychological Service System and Crisis Intervention Mechanism（健全社会心理服务体系和危机干预机制实施方案）(国卫医政发〔2026〕8 号). +- Notes: If there is nobody around you who has no stake in the matter, call 12356 and say it out loud there (see item 11 in this section, call 12356 and have somebody to talk to), or write it down yourself and read it back three days later. The "three months" is a line the author drew himself, with no literature behind that specific length, so it is graded C. The common scams that target the bereaved and older people are in section 17 (investing in retirement, reverse mortgages on a home) and section 6 (fortune telling). The traps in guaranties and IOUs are in section 8. There are a few payments you have to go and claim yourself, and they are not on the "put it off" list: the housing provident fund balance, social insurance benefits, and the three payments on a work-related death, see section 25 and section 19. + +### 13. Do not use death as a way to pay off debts: life insurance does not pay for a suicide in the first two years, a work injury is not recognised, and the debts still come out of the estate first + +- Cost: No money. You are only getting this one account straight. +- In plain language: A life insurance contract does not pay out if you commit suicide within two years; it only returns the part of the policy value that can be returned. A work injury is not recognised either, and you get none of the three payments on a work-related death. The debts are taken out of the estate you leave first, and your family gets only what is left over. They also have to bear the health cost described in item 4 in this section (bereavement after suicide or a violent death). +- Benefit: Article forty-four of the Insurance Law of the People's Republic of China（中华人民共和国保险法）covers contracts in which the payment of insurance money is conditional on the death of the insured person. Counting from the date the contract was made or the date the contract's effect was restored, if the insured person commits suicide within two years, the insurer does not bear the responsibility for paying the insurance money. Where the person committing suicide is a person without capacity for civil conduct, that is, someone who in law cannot take responsibility for their own actions, this does not apply. In that case the insurance company shall return the cash value of the policy as agreed in the contract, that is, it only returns the part of the policy value that can be returned and does not pay the sum assured. The sum assured on a life insurance policy is usually in the tens of thousands of yuan, and the benefit is therefore graded large. Item three of article sixteen of the Regulations on Work-Related Injury Insurance（工伤保险条例）provides: 「自残或者自杀的，不得认定为工伤或者视同工伤」, that is, where a person injures themselves or commits suicide, it shall not be recognised as a work injury or treated as one. The three payments on a work-related death under article thirty-nine all fall away as a result. First the funeral subsidy, paid at 6 months of the local average monthly wage of employees in the previous year. Second the pension for dependents. Third the one-off work-related death benefit, at 20 times the national per-capita disposable income of urban and rural residents in the previous year. Article one thousand one hundred and fifty-nine of the Civil Code（民法典）provides: 「分割遗产应当清偿被继承人依法应当缴纳的税款和债务」, that is, when an estate is divided it shall pay off the taxes and debts that the deceased was lawfully liable for, so the estate pays the taxes and debts first and only what is left is divided. Article one thousand one hundred and sixty-one provides: 「继承人以所得遗产实际价值为限清偿」, that is, an heir repays debts within the actual value of the estate received, so only to the extent of what was inherited and with nothing more to be found from their own pocket, and one who renounces the inheritance bears no repayment responsibility +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (2015 revision). Article forty-four of the Insurance Law of the People's Republic of China（中华人民共和国保险法）. ；State Council (2010 revision). Regulations on Work-Related Injury Insurance（工伤保险条例）（国务院令第 586 号）articles sixteen and thirty-nine. ；National People's Congress (2020). Articles one thousand one hundred and fifty-nine and one thousand one hundred and sixty-one of the Civil Code of the People's Republic of China（中华人民共和国民法典）. +- Notes: Only the money account is counted here. The other items in this section and in section 1 are about not getting to this point. Article forty-four of the Insurance Law covers only the first two years; it is written out so that the provision is complete, not to give you a timetable. The health cost to the family, see item 4 in this section (a spouse whose partner died by suicide has a 70% to 80% higher risk of a newly diagnosed mental disorder within 5 years). What to do when the thought comes up, see items 25 and 32 in section 1, and the after-effects once someone is saved, see item 33 in section 1. The standards for the three payments on a work-related death, see item 16 in section 19. The specific procedures for the estate and the debts, see item 9 in section 25 (collect the housing provident fund and the social insurance benefits one by one). \ No newline at end of file diff --git "a/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" "b/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" new file mode 100644 index 00000000..5b6ffe30 --- /dev/null +++ "b/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" @@ -0,0 +1,147 @@ +[← 回总目录](../README.md) + +# 30. 上学以后的孩子 + +第 20 节讲的是孩子刚出生那一年。第 18 节算的是养孩子要花多少钱、多少时间。中间 3 到 18 岁这十几年，写在这一节。分龄管屏幕那一条从 0 岁写起，因为官方的分档就是 0～3 岁、3～6 岁、中小学三档，拆开写反而不好用。本节回答的是：「别耽误学习」和「身体」撞在一起的时候，哪些事不能等。这些事有时间窗口，过了就补不回来。学业该不该那么卷，本节不讨论，只给窗口和数字。 + +前八条和最后两条算的是死亡和身体上真落下的损害。第 9 条（不买「治愈近视」的产品）算钱。第 10 条（睡眠作业体育排名的明文规定）和第 11 条（休学保留学籍）算时间。溺水、交通、头盔、HPV 疫苗见第 1 节。婴幼儿见第 20 节。孩子被骗和游戏充值见第 5 节。未成年人自己别踩的红线见第 9 节。急症怎么认、怎么叫车见第 13 节。家长去世后的孩子见第 29 节。这些都不在本节重复。 + +[← Back to the contents](../README.md) + +# 30. School-age children + +Section 20 covers the first year of a child's life. Section 18 works out what it costs in money and time to raise a child. The dozen or so years between 3 and 18 are written here. The item on managing screens by age starts at 0 years old, because the official bands are 0–3, 3–6 and school age, and splitting them up makes them less useful. What this section answers is: when "do not let it interfere with the schoolwork" and "the body" collide, which things cannot wait. These things have time windows, and once a window has passed it cannot be made up. Whether schoolwork should be that competitive is not discussed here; this section gives the windows and the numbers. + +The first eight items and the last two measure death and the physical harm that really lands. Item 9 (do not buy products claiming to "cure myopia") measures money. Item 10 (the written rules on sleep, homework, physical education and ranking) and item 11 (taking a leave of absence while keeping the student place) measure time. Drowning, traffic, helmets and the HPV vaccine are in section 1. Babies and toddlers are in section 20. Children being scammed and top-ups in games are in section 5. The lines minors themselves should not cross are in section 9. How to recognise an emergency and how to call an ambulance, see section 13. Children after a parent dies are in section 29. None of this is repeated here. + +### 1. If a child says the pain came on suddenly and is getting worse, go to hospital at once; do not let him sit it out until school is over + +- Cost: No money. Take half a day off and take the child to register at the emergency department. +- In plain language: Some conditions are counted in hours. A boy with sudden severe pain on one side of the scrotum has testicular torsion, which is the vessel attached to the testicle twisted round and the blood supply cut off. Treated within 6 hours, 97.2% of testicles are saved. Left until 19 to 24 hours, only 42.5% are. Pain like this does not wait until school is over, and it does not wait until the exam is finished either. +- Benefit: 30 sets of case reports and 2116 patients with testicular torsion were collected properly and looked at together. 1283 of them are counted in the 6-hour band. Counting from the start of the pain, the proportion of testicles saved from 0 to 6 hours is 97.2%. 7 to 12 hours 79.3%. 13 to 18 hours 61.3%. 19 to 24 hours 42.5%. 25 to 48 hours 24.4%. Over 48 hours 7.4%. Pooled into broad bands, within 12 hours 90.4%, 13 to 24 hours 54.0%, over 24 hours 18.1%. The author's conclusion is that the time in which the testicle can be saved can be longer than the 6 to 8 hours usually quoted. So even if the pain has already lasted a long time, it still has to be treated; do not skip it on the grounds that "it is too late anyway" +- Evidence grade: A +- Sources: Mellick LB, et al. (2019). A Systematic Review of Testicle Survival Time After a Torsion Event. Pediatric Emergency Care. +- Notes: Say something to the boy in advance: if down there it suddenly hurts badly, say so at once, there is no need to be embarrassed about it. Testicular torsion is commonest in adolescent boys, and a good many sit on it because it is hard to bring up. Other conditions counted in hours the same way are acute appendicitis (which perforates if left), diabetic ketoacidosis and meningitis. This section has not checked item by item how much damage a delay of a few hours does in those. The signal they share is this: it starts suddenly, it keeps getting worse, and it is unlike anything that has happened before. How to recognise an emergency and how to call an ambulance, see section 13. + +### 2. Do not put off treatment that has to be done for the sake of "after the exam"; some windows follow bone age, not the examination timetable + +- Cost: No money. The only decision is whether to do it early or late. Leave and follow-up appointments take time; the cost of the treatment itself is separate. The hard part is that the brace has to be worn for at least 18 hours a day, and the child has to put up with it the whole time. +- In plain language: Scoliosis is the spine curving out to the side. A brace is a hard shell worn over the body that holds the spine straight. Worn while the bones are still growing, 75% of children never reach the degree at which they would need an operation; among those who are only observed it is only 42%. The time for wearing it follows the bones finishing their growth, not the junior high or university entrance examination, so waiting until after the exam often leaves an operation as the only option. +- Benefit: A trial run by several hospitals together enrolled 242 adolescents with idiopathic scoliosis who met the conditions for wearing a brace. 116 of them were assigned at random and 126 chose for themselves whether to wear one. Success and failure were defined in advance: curvature reaching 50 degrees or more counts as treatment failure, and being below that degree when the bones finish growing counts as treatment success. Counted by the group first assigned, whether or not they actually went on to wear it enough. In the randomised bracing group the success rate was 75%, in the randomised observation-only group 42% (OR 4.11, 95% CI 1.85–9.16), so the chance of success was about 4 times that of observation only. Pooling the randomised group with the ones who chose for themselves, it is 72% against 48% (after taking out the differences between the two groups that were there to start with, OR 1.93, 1.08–3.46, still about 1.9 times). The more hours a day the brace is worn, the higher the success rate (P<0.001, so this relationship is unlikely to be chance). The trial was stopped early because the brace clearly worked +- Evidence grade: A +- Sources: Weinstein SL, et al. (2013). Effects of bracing in adolescents with idiopathic scoliosis. New England Journal of Medicine. +- Notes: Scoliosis is used as an example here because it has a randomised trial, because it has a clear time window, and because it happens to be one of the focus items of the student medical examination (see item 7 in this section). Others of the same kind are strabismus, amblyopia, dental caries, and operations for congenital heart disease where you can pick the time. What they have in common is that the later you do it the worse the result, or the higher the cost. Ask the specialist about the specific window. If you are afraid of missing lessons, a child can take a leave of absence during the compulsory education stage, for up to 1 year, and the student place is kept, see item 11 in this section. + +### 3. If a child is bullied, report it to the school the same day and ask for a written response; where it involves hitting, taking money or spreading rumours, call the police + +- Cost: No money. Go to the head teacher, and also to the school's body for handling bullying. To call the police, dial 110. The hard part is standing up to advice like "don't make a fuss". +- In plain language: Children bullied by their peers have about 2.2 times the chance of having thoughts of suicide as those who are not bullied, and about 2.6 times the chance of having attempted suicide. A school that receives a report has to investigate at once, and where it finds bullying it has to impose a sanction; in serious cases it may not conceal it and has to report it to the public security organ. So report it the same day, ask for the recognition procedure to be used, and keep a written record. +- Benefit: 34 studies, 284,375 children and adolescents, pooled together. Being bullied by peers is associated with thoughts of suicide, OR 2.23 (95% CI 2.10–2.37), so the chance of the thoughts appearing is about 2.2 times that of the unbullied. A separate 9 studies, 70,102 people, found bullying associated with attempted suicide, OR 2.55 (1.95–3.34), about 2.6 times. Sex, age and the intrinsic quality of the studies all differ, and the result is the same throughout. Cyberbullying is more strongly associated with thoughts of suicide than bullying in person. Article twenty-one of the Provisions on School Protection of Minors（未成年人学校保护规定）lists five kinds of behaviour: hitting, kicking, slapping, scratching and biting, pushing and shoving, pulling and dragging; insulting, ridiculing and giving insulting nicknames; taking property, demanding by force or deliberately damaging property; deliberately excluding and isolating; fabricating facts online in order to slander or spreading rumours. The same article also provides that where the party with the advantage in age, in body or in numbers does these things deliberately or maliciously and causes personal injury, loss of property or mental harm, it may be recognised as bullying. Article twenty-two requires staff to find out about a student's abnormal emotions or physical injury in good time and report it. Article twenty-three requires a school that receives a report to investigate at once, submit it to the bullying-handling body for recognition and disposal, and notify the parents to take part. Where bullying is recognised, an educational sanction or a disciplinary sanction has to be imposed. Serious bullying that breaches public order administration or is suspected of being criminal may not be concealed, and has to be reported in good time to the public security organ and the education administration department (nationwide, in force from 2021-9-1) +- Evidence grade: A +- Sources: van Geel M, et al. (2014). Relationship between peer victimization, cyberbullying, and suicide in children and adolescents: a meta-analysis. JAMA Pediatrics. ；Ministry of Education (教育部) (2021). Provisions on School Protection of Minors（未成年人学校保护规定）（教育部令第 50 号，articles twenty-one, twenty-two and twenty-three）. +- Notes: If a child has an injury whose origin cannot be explained, suddenly refuses to go to school, and things keep going missing, ask about it first using the signals in article twenty-two. This analysis pools several studies together; it only follows records and has no control group, so it cannot show that bullying alone causes suicide, but the direction and the size of the risk are clear. The Plan on Strengthening the Comprehensive Governance of Bullying among Primary and Secondary School Students《加强中小学生欺凌综合治理方案》additionally requires a strict distinction between bullying and the rough-and-tumble play of ordinary students. So the school is obliged to make a finding, and cannot settle the matter with one line about children playing. + +### 4. Two hours outdoors for the child every day; this is the only way of preventing myopia that randomised trials support + +- Cost: No money. But you have to set aside time every day, and it will mostly have to come out of the time for homework and for online classes. The hard part is doing it every day, not a few times. +- In plain language: Adding 40 minutes of outdoor class a day to the first grade in primary school brought new cases of myopia after three years from about forty per cent down to about thirty per cent. The amount the health commission asks for is no less than 2 hours outdoors in the daytime every day, or 14 hours a week in total. Overcast days count, and standing outside at break counts too. +- Benefit: First-grade pupils at 12 primary schools in Guangzhou were assigned at random by school (952 in the intervention schools, 951 in the control schools, mean age 6.6 years). The intervention added 1 outdoor activity lesson of 40 minutes on every school day and encouraged parents to take the children out more outside lessons. Over 3 years new myopia was 30.4% in the intervention group, 259 cases among 853 who qualified for the analysis; 39.5% in the control group, 287 cases among 726. The difference between the two groups was -9.1 percentage points (95% CI -14.1 to -4.1, P<0.001), so the true gap is somewhere between 4 and 14 percentage points lower. The increase in degree over the 3 years (the change in equivalent spherical equivalent) was -1.42 D against -1.59 D, a difference of 0.17 D (P=0.04). The intervention group increased slightly less. D is the unit in which degrees are measured. The growth in axial length (the front-to-back length of the eyeball) showed no difference between the two groups (P=0.07). The Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents《防控儿童青少年近视核心知识十条》requires children and adolescents to spend no less than 2 hours outdoors in the daytime every day, or 14 hours a week in total. It notes that outdoor activity has a protective effect even on overcast days, and it requires avoiding the hot, strongly sunlit hours after midday +- Evidence grade: A +- Sources: He M, et al. (2015). Effect of Time Spent Outdoors at School on the Development of Myopia Among Children in China: A Randomized Clinical Trial. JAMA. ；General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents（防控儿童青少年近视核心知识十条）（国卫办妇幼函〔2023〕278 号）. +- Notes: This trial only ran for 3 years, and the children who took part all started at 6 years old. A 9.1 percentage point difference in the proportion with myopia can stop some people from becoming myopic; it cannot guarantee that somebody will not become myopic. The same Ten Key Points gives several other practices. For reading and writing, "one foot, one fist, one inch". After 20 minutes of looking at something close, look into the distance for at least 20 seconds. A primary or secondary school student using an electronic screen for anything other than study, no more than 15 minutes at a time and no more than 1 hour a day in total. "Being outdoors" and "having a physical education lesson" are not the same thing; standing outside at break counts too. +### 5. No screens from 0 to 3 years old, as few as possible from 3 to 6, and no more than 1 hour a day for primary and secondary school students using a screen for anything other than study + +- Cost: No money. The hard part is taking the screen out of the child's hands, and the parents cutting down their own use in front of the child. +- In plain language: The National Health Commission asks for no screens from 0 to 3 years old, as little contact as possible from 3 to 6, and for primary and secondary school students using a screen for anything other than study no more than 15 minutes at a time and no more than 1 hour a day. The reason is that long periods of close-up work bring myopia on earlier. The evidence on screens and myopia is not that solid; the thing actually tested and shown to work is 2 hours outdoors a day. +- Benefit: The wording of article four of the Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents《防控儿童青少年近视核心知识十条》is「0～3岁婴幼儿不使用手机、平板、电脑等视屏类电子产品；3～6岁幼儿尽量避免其接触和使用手机、平板、电脑等视屏类电子产品；中小学生非学习目的使用电子屏幕单次时长不宜超过15分钟，每天累计时长不宜超过1小时」, that is, infants and toddlers from 0 to 3 years old do not use mobile phones, tablets, computers and other screen-based electronic products; children from 3 to 6 should avoid as far as possible being exposed to and using mobile phones, tablets, computers and other screen-based electronic products; primary and secondary school students using an electronic screen for purposes other than study should keep a single session to no more than 15 minutes and the total for the day to no more than 1 hour. The same article states that「长时间近距离使用电子视屏类产品，易消耗儿童远视储备量，是儿童青少年近视早发、高发的重要原因」, that is, long periods of using screen-based electronic products at close range easily use up children's reserve of far-sightedness, and are an important cause of myopia coming early and often in children and adolescents. The 2018 implementation plan from the Ministry of Education (教育部) and seven other departments gives parents the same requirement, and adds that「年龄越小，连续使用电子产品的时间应越短」, that is, the younger the child, the shorter the continuous period of using electronic products should be. The research evidence is much weaker: a systematic review took in 33 studies, 11 of which entered the pooled calculation, with subjects from 3 months to 33 years old. Looking only at mobile phones and tablets, the odds ratio for myopia is 1.26 (95% CI 1.00 to 1.60, I²=77%). Pooled with computers it is 1.77 (1.28 to 2.45, I²=87%). The authors' conclusion goes no further than "possibly associated", and they state that not one of the 33 studies had a reliable measure of screen time +- Evidence grade: B +- Sources: General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents（防控儿童青少年近视核心知识十条）（国卫办妇幼函〔2023〕278 号）. ; Ministry of Education (教育部) and seven other departments (2018). Implementation Plan for the Comprehensive Prevention and Control of Myopia in Children and Adolescents（综合防控儿童青少年近视实施方案）（教体艺〔2018〕3 号）. ; Foreman J, Salim AT, Praveen A, et al. (2021). Association between digital smart device use and myopia: a systematic review and meta-analysis. The Lancet Digital Health, 3(12), e806-e818. ; Madigan S, McArthur BA, Anhorn C, Eirich R, Christakis DA (2020). Associations between screen use and child language skills: a systematic review and meta-analysis. JAMA Pediatrics, 174(7), 665-675. +- Notes: The grade is B because what the official documents give are banded rules, not numbers measured in a trial. The lower bound of the odds ratio in the research sits exactly at 1.00, and the authors only dare say "possibly associated". So treat it as a good house rule to carry out, not as "no screens means no myopia". What is really supported by a randomised trial is item 4 in this section (2 hours outdoors). Preschool children need their eyes checked too: the same document states that at 1–3 years old, 4–6 years old and after the age of 7 all children should have regular refractive screening to see how much far-sightedness reserve is left; the method is in item 12 in this section (dilation and refraction testing). A meta-analysis has also been done on the relationship between screens and children's language skills: 42 studies, 18905 people, the longer the screen time the weaker the language skills (r=−0.14, 95% CI −0.18 to −0.10). But watching together, and watching programmes with educational content, are positively associated in the other direction (r=0.16 and 0.13). That is association, not causation; all it shows is that replacing talking with a child by putting a screen in front of them is not worth doing. + +### 6. Treat myopia as damage that stays: the deeper the degree, the more retinal detachment, macular disease and glaucoma later on + +- Cost: No money. You have nothing to do about it yourself; what has to be done is in items 4 and 5 (2 hours outdoors, managing screens by age). +- In plain language: Myopia does not get better on its own and the degree does not come back down. Nor is it only a matter of wearing a pair of glasses. Compared with people who are not myopic, the chance of myopic macular degeneration is more than ten times higher with mild myopia and hundreds of times higher above 600 degrees. For retinal detachment it is about 3 times higher with mild myopia and about 13 times higher above 600 degrees. So the earlier a child becomes myopic and the faster it goes up, the larger the risk they are carrying for later. +- Benefit: A systematic review and meta-analysis pooled research published up to 2019-6, sorting by degree into three bands: mild (−0.5 to −3.00 D), moderate (−3.00 to −6.00 D) and high (≤−6.00 D, 600 degrees and above). Compared with people who are not myopic, the odds ratios for myopic macular degeneration are in turn 13.57 (95% CI 6.18 to 29.79), 72.74 (33.18 to 159.48) and 845.08 (230.05 to 3104.34). For retinal detachment, 3.15 (1.92 to 5.17), 8.74 (7.28 to 10.50) and 12.62 (6.65 to 23.94). For posterior subcapsular cataract, 1.56 (1.32 to 1.84), 2.55 (1.98 to 3.28) and 4.55 (2.66 to 7.75). For open-angle glaucoma, mild is 1.59 (1.33 to 1.91), and moderate and high pooled together are 2.92 (1.89 to 4.52). For visual impairment at over 60 years of age the odds ratios for the three bands are in turn 1.71 (1.07 to 2.74), 5.54 (3.12 to 9.85) and 87.63 (34.50 to 222.58). The Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents《防控儿童青少年近视核心知识十条》says the same thing: 600 degrees and above counts as high myopia, and in this group the incidence of blinding eye conditions such as cataract, open-angle glaucoma, myopic macular degeneration, retinal detachment and myopic optic neuropathy is markedly higher than in other groups, and that「近视可防可控不可逆」, that is, myopia can be prevented and controlled but not reversed +- Evidence grade: A +- Sources: Haarman AEG, Enthoven CA, Tideman JWL, Tedja MS, Verhoeven VJM, Klaver CCW (2020). The complications of myopia: a review and meta-analysis. Investigative Ophthalmology & Visual Science, 61(4), 49. ; General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents（防控儿童青少年近视核心知识十条）（国卫办妇幼函〔2023〕278 号）. +- Notes: Everything pooled into this is observational research, and the odds ratios are unusually large. The credible range for high myopia on macular degeneration runs from 230 to 3104, which means the studies underneath it differ a great deal; the figure can only be read for its order of magnitude and must not be used as an exact value. Most of these complications only show up in middle and old age. Myopia itself is not a disability; the great majority of people have normal vision once they wear glasses to correct it. But it is not a matter of putting on a pair of glasses and being done with, and it is worth spending effort on delaying it and controlling it. How to follow up after diagnosis, see item 12 in this section (dilation and refraction testing). Do not buy products claiming they can cure it, see item 9 in this section (curing myopia). + +### 7. Read the report from the annual student health examination yourself, and take any abnormal item to a hospital the same year + +- Cost: No money. In the compulsory education stage the examination fee is paid out of the school's public budget, and parents do not pay separately. +- In plain language: Primary and secondary schools organise one health examination of the students each year. The items to watch on the report are overweight, obesity, malnutrition, abnormal curvature of the spine, poor vision and dental caries; these are the ones most easily let slide. Do not glance at height and weight and put the sheet away. If any of them is abnormal, take the child to a specialist that same year and get it cleared up. +- Benefit: The Measures for the Administration of Health Examinations of Primary and Secondary School Students (2021 Edition)《中小学生健康体检管理办法（2021 年版）》provides that primary and secondary schools organise 1 health examination of the students on roll each year. The surgical examination covers the head, neck, chest, spine, limbs, skin and lymph nodes. The ophthalmic examination covers the appearance of the eye, distance vision and refraction (the degree of myopia or hyperopia). The report handed to each individual should include the results found, an overall assessment and health advice. Overweight, obesity, malnutrition, abnormal curvature of the spine, poor vision and dental caries must be the focus of the guidance. Education administration departments should keep a file on students with health problems and follow them up (write it down and come back to it later). The cost of health examinations for students during the compulsory education stage is paid out of the school's public budget, and parents do not pay separately (nationwide, 2021-9) +- Evidence grade: B +- Sources: National Health Commission (国家卫生健康委) and Ministry of Education (教育部) (2021). Measures for the Administration of Health Examinations of Primary and Secondary School Students (2021 Edition)《中小学生健康体检管理办法（2021 年版）》(国卫医发〔2021〕29 号). +- Notes: The grade is B because this is a document telling schools how to do it, not a study of effects. What it guarantees is "examine every year, examine these items, follow up anything abnormal"; it does not guarantee that somebody will act on what is found. The examination is only a first screen, not a diagnosis. Poor vision needs dilation and refraction testing in the eye department (item 12 in this section). Abnormal curvature of the spine needs the orthopaedic or spinal surgery department (item 2 in this section, do not put off treatment to wait for an exam). Both of those steps have to be taken by a parent taking the child there. + +### 8. Screen children aged 12 to 18 for depression once; do not take the school's psychological questionnaire for a diagnosis + +- Cost: 0 to several tens of yuan, filling in a screening questionnaire at a community centre or an outpatient clinic. The 12355 and 12356 hotlines cost nothing to call. +- In plain language: The US preventive guideline recommends screening 12- to 18-year-olds for major depression; for 11 and under the evidence is not enough. Suicide is the second leading cause of death among 10- to 19-year-olds. The psychological questionnaire the school gives every year is a census, not a diagnosis. If the child says they feel awful, or eating, sleep, interests and school marks all change suddenly, register with psychiatry or clinical psychology. +- Benefit: The US Preventive Services Task Force issued a recommendation statement in 2022. It recommends screening adolescents aged 12 to 18 for major depressive disorder (a grade B recommendation). A grade B recommendation means moderate confidence and a moderate expected benefit, so it is worth doing. Screening children aged 11 and under for major depression, the evidence is insufficient (an I statement, that is the evidence is not enough to give a recommendation). Screening children and adolescents specifically for suicide risk, the evidence on benefit and on harm is both insufficient (an I statement). The document points out that suicide is the second leading cause of death in the 10- to 19 age group. It also points out that major depression in children and adolescents is closely related to a good many later problems. These include recurrent depressive episodes in adulthood, other mental disorders, and a higher risk of suicidal thoughts, attempted suicide and death by suicide. The Ten Measures to Further Strengthen the Mental Health Work for Primary and Secondary School Students《进一步加强中小学生心理健康工作十条措施》requires a three-level service system of head teachers, full-time and part-time school mental health teachers, and psychiatrists. Each student is to have a psychological assessment once per school year, generally no more than 1 time. The document also requires referral to medical care for mental illness and a mechanism for returning to school after recovery. It requires the 12355 teenage service desk and the 12356 psychological assistance hotline to play their part (nationwide, 2025-10) +- Evidence grade: B +- Sources: US Preventive Services Task Force (2022). Screening for Depression and Suicide Risk in Children and Adolescents: US Preventive Services Task Force Recommendation Statement. JAMA. ；General Office of the Ministry of Education (教育部办公厅) (2025). Ten Measures to Further Strengthen the Mental Health Work for Primary and Secondary School Students《进一步加强中小学生心理健康工作十条措施》（教基厅〔2025〕2 号）. +- Notes: The grade is B because what the original gives is a rating of the strength of the recommendation, not an effect that can be turned into a number. It is also a US recommendation; China has no matching official screening guidance. A positive screening result only means you have to do one more assessment; it is not a diagnosis. What to do when there are thoughts of suicide, see item 25 in section 1. How to handle it at home, see section 29. + +### 9. Do not buy products and services claiming they can "cure myopia" or "reduce the degree" + +- Cost: No money, and you save the money as well. +- In plain language: The State Administration for Market Regulation（市场监管总局）says outright that「在目前医疗技术条件下，近视不能治愈」, that is, under current medical technology myopia cannot be cured. Anyone marketing with claims like "restore", "reduce the degree", "myopia cured" or "degree repaired" is to be dealt with strictly. When you see these words, rule the product out and do not compare prices. An eye-protection device counts as a medical device, and the shopping platform is responsible for checking whether the seller is qualified to sell it. +- Benefit: A notice from the General Office of the State Administration for Market Regulation（市场监管总局办公厅）for a special rectification campaign from 2021-11 to 2022-3 states: 「在目前医疗技术条件下，近视不能治愈」, that is, under current medical technology myopia cannot be cured. The notice requires strict enforcement against the behaviour of marketing myopia prevention and control products for children and adolescents with misleading wording such as 「康复」「恢复」「降低度数」「近视治愈」「近视克星」「度数修复」, that is, rehabilitation, restore, reduce the degree, myopia cured, myopia nemesis, degree repaired. The notice requires a heavier hand, covering the behaviour of passing itself off as traditional Chinese medicine, or borrowing on traditional Chinese medicine theory and technique to do marketing and publicity and deceive consumers. The notice also requires severe action against two kinds of false and unlawful advertising: using the name or the image of a patient to show a before-and-after comparison of treatment results, and making guaranteed claims about the safety and the efficacy of treatment. The notice requires shopping platforms to strictly implement the review of business qualifications for medical devices such as eye-protection devices and for related special goods (finding out whether the seller is qualified to sell them) (nationwide) +- Evidence grade: B +- Sources: General Office of the State Administration for Market Regulation（市场监管总局办公厅）(2021). Notice on Carrying Out a Special Rectification Campaign Against Unlawful Commercial Marketing and Publicity for Myopia Prevention and Control Products for Children and Adolescents（关于开展儿童青少年近视防控产品违法违规商业营销宣传专项整治行动的通知）. +- Notes: The grade is B because what it rests on is the standard the regulators use when they take action, not something measured by comparing products in a trial. But the statement that myopia cannot be cured is the same from the health department and the regulators: the Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents《防控儿童青少年近视核心知识十条》says「近视可防可控不可逆」, that is, myopia can be prevented and controlled but not reversed. The two things that really have evidence behind them are item 4 in this section (outdoors) and item 12 in this section (standard refraction testing, glasses and follow-up). For blue-light-blocking glasses, see section 6. +### 10. Sleep, homework, physical education and ranking all have written rules, and where the school cannot meet them you can raise it + +- Cost: No money. Knowing the provisions is enough; if necessary raise it with the school or with the local education administration. The hard part is being the one who raises it. +- In plain language: The documents state that primary school pupils should sleep 10 hours a day, junior high students 9, senior high students 8. Morning classes in primary school generally start no earlier than 8:20, and in secondary school no earlier than 8:00. From 2025 there is a further rule that there must be no less than 2 hours of physical activity a day, and that pupils may not be ranked by examination results. If your child has been sleeping too little for a long time, check it against these provisions first instead of blaming the child for dawdling. +- Benefit: The Notice on Further Strengthening the Management of Sleep for Primary and Secondary School Students《关于进一步加强中小学生睡眠管理工作的通知》provides: primary school pupils should sleep 10 hours a day, junior high students 9, senior high students 8. Morning classes in primary school generally start no earlier than 8:20, and in secondary school no earlier than 8:00. Schools may not require pupils to arrive early to take part in collective educational and teaching activities. Primary school pupils should be in bed generally no later than 21:20, junior high students no later than 22:00, senior high students no later than 23:00. Off-campus training institutions may not finish later than 20:30, and online live training may not finish later than 21:00. Between 22:00 each day and 8:00 the next day, game services may not be provided to minors (nationwide, 2021-3). The Ten Measures to Further Strengthen the Mental Health Work for Primary and Secondary School Students《进一步加强中小学生心理健康工作十条措施》has several more provisions: across the board, no less than 2 hours of comprehensive physical activity a day for primary and secondary school students, and it encourages a "15-minute break between classes". It also provides that students may not be ranked by examination results, that mechanical, repetitive or punitive homework is strictly forbidden, and that one homework-free day a week is encouraged (nationwide, 2025-10) +- Evidence grade: B +- Sources: General Office of the Ministry of Education (教育部办公厅) (2021). Notice on Further Strengthening the Management of Sleep for Primary and Secondary School Students《关于进一步加强中小学生睡眠管理工作的通知》（教基厅函〔2021〕11 号）. ；General Office of the Ministry of Education (教育部办公厅) (2025). Ten Measures to Further Strengthen the Mental Health Work for Primary and Secondary School Students《进一步加强中小学生心理健康工作十条措施》（教基厅〔2025〕2 号）. +- Notes: Both of these are documents issued by the authorities in charge. How far each school carries them out varies a great deal, which is why this item is graded B. Their use is that you know which things the school was supposed to be doing anyway, so you have something to point to when you talk to them. The sleep account for adults, see section 2 and section 3. + +### 11. If a child cannot carry on any more they can take a leave of absence, and the school must keep their student place, for up to 1 year + +- Cost: No money. A parent or other legal guardian applies to the school. The hard part is having to accept that the child stops for a while. +- In plain language: A leave of absence is a formal system. A parent or other guardian applies, and once the school has confirmed it, reports it upwards for approval. One period is at most 1 year; if you still need to stop when it ends, apply again. During the leave the school has to keep the student's place, and they carry on where they left off when they come back. The specific rules are set by each province, so ask the school's academic affairs office or the local education bureau. +- Benefit: Article sixteen of the Measures for the Administration of Student Status for Primary and Secondary School Students《中小学生学籍管理办法》sets out how a leave of absence is handled. A student's leave is applied for by a parent or other legal guardian. Once the school has confirmed it, the school reports it to the student status administration department above for approval. When the student comes back the school should complete the relevant formalities in good time. A single period of leave may not exceed 1 year; where it is genuinely still needed once that period ends, a fresh application must be made. During the leave the school should keep the student's place for the student (their status and their place in the roll are kept). The specific rules on returning from a leave of absence are made by the provincial education administration department (nationwide, 2025-1). The Ten Measures to Further Strengthen the Mental Health Work for Primary and Secondary School Students《进一步加强中小学生心理健康工作十条措施》requires referral to medical care for mental illness and the building of a mechanism for returning to school after recovery, and instructs schools to draw up plans for handling sudden psychological crises on campus +- Evidence grade: B +- Sources: Ministry of Education (教育部) (2025). Measures for the Administration of Student Status for Primary and Secondary School Students《中小学生学籍管理办法》（教基〔2025〕1 号，article sixteen）. ；General Office of the Ministry of Education (教育部办公厅) (2025). Ten Measures to Further Strengthen the Mental Health Work for Primary and Secondary School Students《进一步加强中小学生心理健康工作十条措施》（教基厅〔2025〕2 号）. +- Notes: Before you go and do it, ask three things: what medical certificate is required, how the student status is recorded in the system during the leave, and whether coming back means returning to the original year or dropping a year. The grade is B because what conditions a leave requires, what materials have to be handed in and how the return is joined up are all set by the provincial education department; at the national level there is only a framework. The Provisions on School Protection of Minors《未成年人学校保护规定》also require compulsory education schools to register and file students who have dropped out, taken a leave of absence or been away for a long time, so a leave of absence is something the school has handled before, not a special exception for you alone. + +### 12. If poor vision is found, have dilation and refraction testing done at a hospital, and afterwards go back at the interval the doctor gives you + +- Cost: Tens to hundreds of yuan for the examination and for the glasses. After the dilation the child will be sensitive to light and unable to see near things clearly, and it takes half a day to a day or two to settle. +- In plain language: Dilation and refraction testing means putting in eye drops that relax the muscle inside the eye responsible for focusing, and then measuring the degree. It is done in hospital, and it is the most accurate way of diagnosing myopia. The health commission states that these drops do no damage to a healthy eye, and that things return to normal once they are stopped. Do not take a single computer measurement in an optician's shop as a diagnosis, and do not refuse to get your child glasses for fear of "making the degree worse". +- Benefit: The Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents《防控儿童青少年近视核心知识十条》explains what dilation and refraction testing is. A cycloplegic agent is used first, and then the refraction is measured. A cycloplegic agent is eye drops that relax the muscle inside the eye responsible for focusing. Dilation and refraction testing is a medical refraction done in hospital, and it is the「金标准」for diagnosing myopia (the most accurate method). The dilating drugs do no damage to a healthy eye; after using them there may be short-term sensitivity to light and blurred near vision, and normal function returns once they are stopped. Once myopia is diagnosed, prompt and scientific intervention is needed, and it is corrected by wearing glasses and so on. Wearing the right pair of glasses corrects blurred vision effectively and slows down the progression of myopia. Where vision with glasses is normal, preschool children and primary school pupils should be examined again every 3 to 6 months, and junior and senior high school students every 6 to 12 months, and whether to change the glasses is decided on the specifics. Children should have regular refractive screening at 1 to 3 years old, 4 to 6 years old and after the age of 7 (measuring the degree), to monitor the far-sightedness reserve (nationwide, 2023-7) +- Evidence grade: B +- Sources: General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents（防控儿童青少年近视核心知识十条）（国卫办妇幼函〔2023〕278 号）. +- Notes: The grade is B because this is the method given in an official popular-science document, not an original study with specific figures. High myopia brings retinal detachment and similar trouble, so the follow-up is not only about changing glasses. The "poor vision" found on the student health examination is only a screening result; it still has to be taken to hospital for one complete eye examination, see item 7 in this section. + +### 13. Get pit and fissure sealants once the first permanent molars are through + +- Cost: Tens to a hundred or two hundred yuan per tooth. No anaesthetic and no drilling, and it takes a few minutes. Many places have a free programme for children who have reached the age. +- In plain language: A pit and fissure sealant seals the grooves on the chewing surface of the big teeth with resin, so that food debris cannot get in. Two years after it is done, the chance of getting a cavity is about one eighth of what it is without it. Without it 16% of tooth surfaces get a cavity; with it only about 5%. It is still effective after four years. +- Benefit: 38 trials, 7924 children aged 5 to 16, properly collected by Cochrane and pooled together. Resin sealants are compared against no sealant. For the first permanent molar in children aged 5 to 10, at 24 months after the sealant the odds of a cavity were OR 0.12 (95% CI 0.08–0.19), that is about one eighth of the chance without a sealant. This item comes from 7 trials, with 1548 assigned at random and 1322 examined at the end, and the quality of the evidence is moderate. Where 16% of tooth surfaces in the control group developed a cavity within 24 months, after sealing it is 5.2% (3.13%–7.37%). Where the control group figure is 40%, it is 6.25% (3.84%–9.63%). Where it is 70%, it is 19% (12.3%–27.2%). At 48 to 54 months after the sealant OR 0.21 (0.16–0.28), still about one fifth of no sealant. The authors' conclusion is that resin sealants produce 11% to 51% fewer cavities than no sealant at 24 months. Of the 4 trials that recorded adverse events, none reported any. There is not enough evidence to say which is better among the other sealing materials such as glass ionomer +- Evidence grade: A +- Sources: Ahovuo-Saloranta A, et al. (2017). Pit and fissure sealants for preventing dental decay in permanent teeth. Cochrane Database of Systematic Reviews. +- Notes: The first permanent molar (the big tooth that will not be replaced) generally comes through at around 6 years old, the second at around 12. The sealant has to go on while the tooth has only just come through and before it has decayed; once there is a cavity all you can do is fill it. Many regions list sealants for children who have reached the age as a free service, so ask the local CDC or the community health centre. This section did not find a national document whose wording can be checked word for word, so it does not state the coverage or the age band. Dental caries is also a focus item of the student health examination, see item 7 in this section. + +### 14. If your child loves games, first check whether sleep, homework and time outdoors have been squeezed out; do not only watch how long the play goes on for + +- Cost: No money. Once a week, note what time your child goes to bed, whether the homework was handed in, and whether there was time outside. +- In plain language: Genuinely meeting the description of "gaming disorder" is two or three in every hundred. If the questionnaire threshold is set low and only adolescents are surveyed, the proportion comes out too high. Looking at the real records from the game companies, how much somebody plays has almost nothing to do with how they feel. Minors are allowed, under the rules, to play only on Friday, Saturday, Sunday and public holidays, between 8 and 9 in the evening. +- Benefit: A meta-analysis pooled 53 studies from 17 countries, 226247 people. The prevalence of gaming disorder is 3.05% (95% CI 2.38–3.91, which is the credible range). Counting only the studies with stricter sampling, it is 1.96%. The results differ a long way between studies; 77% of the variation comes down to which questionnaire was used. Samples of adolescents, low threshold scores and small samples all push the proportion up. A separate study obtained data from 7 game companies and linked the real playing time of 38935 players over six weeks to the well-being they filled in themselves. There is almost no evidence of playing time affecting well-being. The National Press and Publication Administration（国家新闻出版署）provides that game companies may give minors 1 hour of service each day from 20 to 21 hours on Fridays, Saturdays, Sundays and statutory public holidays (nationwide / internationally, from 2021) +- Evidence grade: B +- Sources: Stevens MW, Dorstyn D, Delfabbro PH, King DL. (2021). Global prevalence of gaming disorder: A systematic review and meta-analysis. Australian and New Zealand Journal of Psychiatry, 55(6), 553–568. ；Vuorre M, Johannes N, Magnusson K, Przybylski AK. (2022). Time spent playing video games is unlikely to impact well-being. Royal Society Open Science, 9(7), 220411. ；National Press and Publication Administration (国家新闻出版署) (2021). Notice on Further Strict Management to Effectively Prevent Minors from Becoming Addicted to Online Games《关于进一步严格管理切实防止未成年人沉迷网络游戏的通知》（国新出发〔2021〕14 号）. +- Notes: The grade is B because the prevalence changes too much with the questionnaire, and the well-being study only followed people for six weeks. The commonest cost of games is that they squeeze out sleep and time outdoors. The written requirement on sleep is in item 10 in this section (sleep, homework, physical education), outdoors is in item 4 in this section (2 hours outdoors a day), and screen time is in item 5 in this section (no more than 1 hour a day for anything other than study). Top-ups and refunds are in item 9 in section 5 (a child topping up or tipping on a phone). When school marks, sleep and the relationship with the family all fall apart together, and advice does not stop it, take the child to psychiatry or to child and adolescent psychology. The beneficiary here is the child. +### 15. If your child says they like the same sex, do not scold them, do not turn them out of the house, and do not send them for "correction": the attitude at home is related to whether they will attempt suicide + +- Cost: No money. The hard part is that the parent has to work through their own feelings first, and not say anything harsh at the time. +- In plain language: Among secondary school students in China, children who like the same sex or who like both sexes have attempted suicide in the past year at about 3 to 5.5 times the rate of other children. Research from other countries finds that adolescents strongly rejected at home are far more likely to have attempted suicide later. Those sent by their parents for "correction" have the highest rate. +- Benefit: A survey in China covered 506 secondary schools in 7 provinces, 150822 pupils from the first year of junior high to the third year of senior high. Among those who had attempted suicide in the past year, boys who liked only the opposite sex were 2.2%, boys who liked only the same sex 6.9%, and boys who liked both 12.2%. For girls the figures are 3.1%, 8.9% and 10.9%. That works out at about 3 to 5.5 times the figure for classmates who like only the opposite sex. The researchers took out age, academic pressure, family income and being bullied. After that, the OR for those two groups of boys is 3.13 (95% CI 2.28–4.28) and 3.83 (2.85–5.14), that is two to three times higher. A meta-analysis of 35 studies works out an OR of 3.50 (2.98–4.12) for attempted suicide among sexual minority adolescents. A US study interviewed 224 homosexual and bisexual young people aged 21 to 25 and asked them to recall the reaction at home during adolescence. Compared with peers whose families hardly rejected them, the OR for having attempted suicide where family rejection was high is 8.4. In a separate study, where parents themselves tried to change their child's sexual orientation, the OR for the child having attempted suicide is 3.08 (1.39–6.83). Where parents also took the child to see a therapist or a religious figure, it is 5.07 (2.38–10.79) (China / several countries / United States) +- Evidence grade: B +- Sources: Huang Y, Li P, Guo L, et al. (2018). Sexual minority status and suicidal behaviour among Chinese adolescents: a nationally representative cross-sectional study. BMJ Open, 8(8), e020969. ；di Giacomo E, Krausz M, Colmegna F, Aspesi F, Clerici M. (2018). Estimating the Risk of Attempted Suicide Among Sexual Minority Youths: A Systematic Review and Meta-analysis. JAMA Pediatrics, 172(12), 1145–1152. ；Ryan C, Huebner D, Diaz RM, Sanchez J. (2009). Family rejection as a predictor of negative health outcomes in white and Latino lesbian, gay, and bisexual young adults. Pediatrics, 123(1), 346–352. ；Ryan C, Toomey RB, Diaz RM, Russell ST. (2020). Parent-Initiated Sexual Orientation Change Efforts With LGBT Adolescents: Implications for Young Adult Mental Health and Adjustment. Journal of Homosexuality, 67(2), 159–173. ；Standing Committee of the National People's Congress (2024 revision). Law on the Protection of Minors（未成年人保护法）（article seventeen）. ；Standing Committee of the National People's Congress (2015). Anti-Domestic Violence Law（反家庭暴力法）（article two）. +- Notes: The grade is B because all of these are one-off surveys, and the degree of rejection at home is recalled in adulthood, so they can show an association and not who caused what. Parents do not have to accept anything straight away; first do not do anything that harms the child. In law, parents may not abuse or abandon a child, and may not commit domestic violence against a child. Regular verbal abuse and intimidation count as domestic violence as well. If the child's mood is low and they say they do not want to live, call 12356 first, see item 25 in section 1 (call 12356 when there are thoughts of suicide). You can also go to psychiatry following item 8 in this section (one depression screen between 12 and 18 years old). For why "correction" is not to be touched, see item 28 in section 6 (do not carry out sexual orientation conversion). The beneficiaries here are the child and you. \ No newline at end of file diff --git "a/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" "b/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" new file mode 100644 index 00000000..f3f6bd37 --- /dev/null +++ "b/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" @@ -0,0 +1,158 @@ +[← 回总目录](../README.md) + +# 31. 十八岁之后有哪几条路 + +这一节算的是钱和时间，拒服兵役那一条还关系到人身自由。这是一张路径地图，不是某一条路的说明书。本节先把各条路的法定门槛摆在一起，再逐条写那些知道了才用得上的东西：当兵能拿到什么，进体制内除了考公还有哪些口子，高考填志愿时哪两个专业直接对着编制，学历后来还能不能补，想出国打工该找什么样的公司，不进单位的人社保怎么办，没本钱起步有没有国家替你出一部分利息的钱可借。 + +十八岁之后能走的路，远不止这一节写到的这些。这里只收有全国层面的公开文件、门槛和待遇能逐条核对的那几条。进国企、进工厂、当船员、做自由职业、接着读高中读大学，这些路也能走。只是它们没有统一文件可查，本书不凭印象替它们写条件。 + +选哪条路是你自己的事。本节只提供两样东西：门槛写在哪份文件里，走上去之后有哪些期限和资格过期就作废。读书还是打工那道题怎么算、技能怎么挑，见第 23 节。进了单位之后的加班费、工伤和离职补偿，见第 19 节。开店办公司的本钱和退场，见第 12 节。出国留学见第 32 节。失业之后的保障见第 7 节。这些都不重复。 + +[← Back to the contents](../README.md) + +# 31. What are the paths after eighteen + +This section measures money and time, and refusing military service also bears on personal freedom. It is a map of the paths, not a manual for any one of them. It first sets the legal thresholds for the various paths side by side, and then goes through one by one the things you only need once you have chosen: what serving gives you, what openings there are besides the civil service examination for getting into an establishment, which two majors on the national entrance examination fill posts with a staffing quota attached, whether a qualification can still be completed later, what kind of company to look for if you want to work abroad, what to do about social insurance if you are not in any establishment, and whether there is money to borrow for starting out with part of the interest paid by the state. + +There are far more paths open after eighteen than are written here. Only those with public documents at the national level, whose thresholds and benefits can be checked line by line, are collected here. Going into a state-owned enterprise, going into a factory, becoming a seafarer, working as a freelancer, going on to senior high school or university, these are all paths too. They simply have no single document to look up, and this book does not write conditions for them out of impression. + +Which path you take is your own business. This section offers two things only: which document the threshold is written in, and which deadlines and qualifications on the way up expire and are then void. How to work out the question of studying or working, and how to choose a skill, see section 23. Overtime pay, work injuries and severance once you are in an establishment, see section 19. The capital for opening a shop or a company and how to get out, see section 12. Studying abroad, see section 32. Protection after losing your job, see section 7. None of these is repeated here. + +### 1. Check the thresholds first: the age and education requirements for twelve paths are all fixed in writing + +- Cost: No money. Ten minutes to go through them and see whether you can reach them. +- In plain language: The age and education thresholds for these twelve paths are all written into documents. Going out to work requires being at least 16 years old. Military service requires a man to be at least 18 years old, and this is relaxed to 24 for university graduates. Sitting the civil service examination requires being 18 to 35 years old with a college education or above. The self-study examination for a qualification and registering as an individual business are the two paths that look at neither your education nor your age. +- Benefit: For military service, look at the Military Service Law（兵役法）: 「年满十八周岁的男性公民，应当被征集服现役；当年未被征集的，在二十二周岁以前仍可以被征集服现役。普通高等学校毕业生的征集年龄可以放宽至二十四周岁，研究生的征集年龄可以放宽至二十六周岁」, that is, male citizens who have reached eighteen years old shall be called up for active service; those not called up in that year may still be called up before the age of twenty-two; for graduates of ordinary higher education institutions the call-up age may be relaxed to twenty-four, and for postgraduates to twenty-six. For the civil service examination, look at article eighteen of the Provisions on the Recruitment of Civil Servants《公务员录用规定》: the conditions for application include 「年龄为十八周岁以上，三十五周岁以下」, aged eighteen or above and under thirty-five, and 「具有大学专科以上文化程度」, having an educational level of college or above. These two may be adjusted appropriately with the approval of the civil service department at provincial level or above. Article nineteen of the same Provisions lists five kinds of person who may not apply. For example those who have been criminally punished for a crime, those dismissed from public office, and those listed as subjects of joint punishment for dishonesty. For firefighters, look at the Measures for the Recruitment of Firefighters into the National Comprehensive Fire and Rescue Force《国家综合性消防救援队伍消防员招录办法》. Article five: 「年龄为 18 周岁以上、22 周岁以下」, aged eighteen or above and under twenty-two, and 「具有高中以上文化程度」, having an educational level of senior high school or above. Article six: 「大学专科以上学历人员、解放军和武警部队退役士兵……年龄可以放宽至 24 周岁」, for people with a college education or above, and for soldiers discharged from the People's Liberation Army and the Armed Police Force, the age may be relaxed to twenty-four. For civilian posts in the armed forces, look at article nine of the Regulations on Civilian Personnel of the People's Liberation Army of China《中国人民解放军文职人员条例》, whose basic conditions include 「年满 18 周岁」, being at least 18 years old. Article nineteen of the same Regulations states that the highest age for first recruitment and appointment to posts at a lower grade is 35 years old. Posts at a lower grade are the entry-level posts. For public institutions, look at article eight of the Regulations on the Administration of Personnel of Public Institutions《事业单位人事管理条例》: 「事业单位新聘用工作人员，应当面向社会公开招聘」, a public institution recruiting new staff shall recruit openly from the public. For an individual business, look at the Regulations on the Promotion of the Development of Individual Businesses《促进个体工商户发展条例》: 「有经营能力的公民……依法登记为个体工商户的，适用本条例」, citizens with the capacity to run a business who register as an individual business in accordance with the law are governed by these Regulations. For the self-study examination, look at the Interim Regulations on Higher Education Self-Study Examinations《高等教育自学考试暂行条例》. Article three: 「中华人民共和国公民，不受性别、年龄、民族、种族和已受教育程度的限制，均可依照本条例的规定参加高等教育自学考试」, citizens of the People's Republic of China, without restriction as to sex, age, ethnicity, race or level of education already received, may all take part in higher education self-study examinations in accordance with these Regulations. Article twenty: 「各级各类全日制学校的在校生不得报考」, students enrolled on full-time courses at schools of every level and type may not apply. For publicly funded teacher training students, look at article two of the Implementation Measures for Publicly Funded Education of Students on the Bridging Programme for Undergraduates and Postgraduates at Normal Universities Directly under the Ministry of Education《教育部直属师范大学本研衔接师范生公费教育实施办法》. It recruits only at six normal universities directly under the Ministry. They are Beijing Normal University（北京师范大学）, East China Normal University（华东师范大学）, Northeast Normal University（东北师范大学）, Central China Normal University（华中师范大学）, Shaanxi Normal University（陕西师范大学）and Southwest University（西南大学）. The undergraduate and postgraduate stages are designed as one and assessed in stages: after finishing the undergraduate degree you go on to the postgraduate degree. Tuition and accommodation while on campus are borne by the central finance, and a living allowance is given as well. Article five of the same Measures: 「招收公费师范生实行提前批次录取」, publicly funded teacher training students are admitted in an advance batch, that is, they are admitted before the ordinary batch. For contract medical students, look at the Notice on Employment Placement and Performance Management of Rural Order-Directed Free-Trained Medical Students《关于做好农村订单定向免费培养医学生就业安置和履约管理工作的通知》. It states: 「在录取后、获得入学通知书前，须与培养学校和定向就业所在地的县级卫生健康行政部门……人力资源社会保障部门签署协议」, that is, after admission and before receiving the admission notice, the student must sign an agreement with the training school and with the county-level health administration department and the human resources and social security department of the place where the employment is directed. For working abroad, look at the Regulations on the Administration of Overseas Labour Cooperation《对外劳务合作管理条例》. Article five: engaging in overseas labour cooperation requires the approval of the commerce department in charge and the acquisition of a qualification to operate. Article seven, paragraph four: 「未依法取得对外劳务合作经营资格证书并办理登记，不得从事对外劳务合作」, no one may engage in overseas labour cooperation without having obtained the certificate of qualification for overseas labour cooperation in accordance with the law and completed registration. A company that has no such certificate but organises people to go abroad to work is itself breaking the law. Article eight, paragraph two: 「任何单位和个人不得以商务、旅游、留学等名义组织劳务人员赴国外工作」, no unit or individual may organise workers to go abroad to work in the name of business, tourism, study abroad or anything of the kind. Platform work and flexible employment have no statutory education threshold; for enrolling in insurance yourself and for the work-injury cover, see items 11 and 12 in this section. The Compulsory Education Law（义务教育法）「禁止用人单位招用应当接受义务教育的适龄儿童、少年」, that is, prohibits employers from employing children and adolescents of school age who should be receiving compulsory education (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (2021). Military Service Law（兵役法）（article twenty）. （published by the Ministry of National Defense）；Central Organization Department of the CPC Central Committee (2019). Provisions on the Recruitment of Civil Servants《公务员录用规定》（articles eighteen and nineteen）. （published on the Communist Party member website）；Ministry of Human Resources and Social Security and Ministry of Emergency Management (2021). Measures for the Recruitment of Firefighters into the National Comprehensive Fire and Rescue Force《国家综合性消防救援队伍消防员招录办法》（人社部发〔2021〕58 号，articles five and six）. ；State Council and Central Military Commission (2022). Regulations on Civilian Personnel of the People's Liberation Army of China《中国人民解放军文职人员条例》（second revision, Order No. 757, articles nine and nineteen）. （National Database of Laws and Regulations）；State Council (2014). Regulations on the Administration of Personnel of Public Institutions《事业单位人事管理条例》（国务院令第 652 号，article eight）. ；State Council (2022). Regulations on the Promotion of the Development of Individual Businesses《促进个体工商户发展条例》（国务院令第 755 号，article two）. ；State Council (1988, revised 2014). Interim Regulations on Higher Education Self-Study Examinations《高等教育自学考试暂行条例》（articles three and twenty）. （National Database of Laws and Regulations）；Standing Committee of the National People's Congress (2018). Compulsory Education Law（义务教育法）（article fourteen）. （published by the Ministry of Education）；General Office of the State Council (2024). Notice on Forwarding the Implementation Measures for Publicly Funded Education of Students on the Bridging Programme for Undergraduates and Postgraduates at Normal Universities Directly under the Ministry of Education from the Ministry of Education and Other Departments《教育部直属师范大学本研衔接师范生公费教育实施办法》（国办发〔2024〕27 号，articles two and five）. ；National Health Commission and 7 other departments (2019). Notice on Employment Placement and Performance Management of Rural Order-Directed Free-Trained Medical Students《关于做好农村订单定向免费培养医学生就业安置和履约管理工作的通知》. ；State Council (2012). Regulations on the Administration of Overseas Labour Cooperation《对外劳务合作管理条例》（国务院令第 620 号，articles five, seven and eight）. +- Notes: Having reached the threshold, you still have to get through the political criteria, the medical examination and the political assessment. These have no published scoring standard and cannot be found in the provisions. The entry points for secondary vocational schools, technical schools and integrated vocational education are in item 5 in section 23, and the financial help available when the family cannot afford it is in item 4 in section 23. An individual business does not look at your education, but opening a shop is not easy. The capital, the personal guarantee and the cost of closing down are in section 12. A personal guarantee means that if the company cannot repay, you have to repay it yourself. Whether the money to start with can be borrowed as a loan where the state pays part of the interest is in item 16 in this section. The two of publicly funded teacher training students and contract medical students are settled on the day you fill in your choices for the national entrance examination, and what they cost is written out in item 13 in this section. For working abroad the threshold is not on you, it is on whether the company is qualified; how to check that is in item 14 in this section. Which path is better value is not compared here. + +### 2. Register for military service before 31 October in the year you turn eighteen; a conscript serves two years of active duty + +- Cost: No money. Registration is filled in once online and takes a few minutes. If you really do serve, what it costs is two years of full-time work, and you cannot do anything else at the same time in those two years. +- In plain language: Registering for military service is a legal duty. It is not signing up to join the army, and being registered does not mean you are going. A man who reaches eighteen in that year has to complete his registration online before 31 October. If you do not register, the county will order you to correct it within a set time, and if you still do not, it will compel you to perform military service and fine you. If you really do serve, a conscript serves two years of active duty. +- Benefit: For registration, look at the Provisions on Military Service Registration《兵役登记工作规定》. It states: 「县、自治县、不设区的市、市辖区人民政府兵役机关应当于每年 10 月 31 日前，组织本行政区域内当年 12 月 31 日以前年满 18 周岁的男性公民进行初次兵役登记」, that is, the military service organs of the people's governments of counties, autonomous counties, cities without districts and municipal districts shall, before 31 October each year, organise the initial registration for military service of the male citizens in their administrative area who will have reached 18 years old before 31 December of that year. The same Provisions also state: 「初次兵役登记主要采取网络登记的方式进行，由公民通过兵役登记网络平台自主完成」, the initial registration for military service is carried out mainly online, completed by the citizen independently through the military service registration online platform. For what happens if you do not register, look at article fifty-seven, paragraph one of the Military Service Law. That paragraph covers 「拒绝、逃避兵役登记的」, those who refuse or evade registration for military service. The penalty is 「由县级人民政府责令限期改正；逾期不改正的，由县级人民政府强制其履行兵役义务，并处以罚款」, the county-level people's government orders correction within a set time; if the correction is not made by the deadline, the county-level people's government compels the person to perform military service and imposes a fine. The Military Service Law also states that 「义务兵服现役的期限为二年」, the period of active service for a conscript is two years, and that 「根据军队需要和本人自愿，可以征集年满十七周岁未满十八周岁的公民服现役」, depending on the needs of the armed forces and on the person's own wish, citizens who have reached seventeen but not eighteen may be called up for active service. For non-commissioned officers, 「服现役满十二年或者符合国家规定的其他条件的，由安置地的县级以上地方人民政府安排工作」, that is, where they have completed twelve years of active service or meet other conditions laid down by the state, the local people's government at county level or above in the place of resettlement will arrange work. And 「服现役满三十年或者年满五十五周岁或者符合国家规定的其他条件的，作退休安置」, where they have completed thirty years of active service, or have reached fifty-five years of age, or meet other conditions laid down by the state, they are retired under the resettlement arrangements. Arranging work means the government gives you a job. Retiring under the resettlement arrangements means being retired under the retirement rules. The Regulations on Military Conscription《征兵工作条例》state that 「优先保证普通高等学校毕业生和对政治、身体条件或者专业技能有特别要求的兵员征集」, priority is given to recruiting graduates of ordinary higher education institutions and to personnel with special requirements as to political, physical or professional skills. The same Regulations also state that 「应征公民为普通高等学校的全日制在校生、应届毕业生的，可以在入学前户籍所在地或者学校所在地应征」, a citizen called up who is a full-time student or a graduating student at an ordinary higher education institution may be called up before enrolment in the place of household registration or the place where the school is (nationwide) +- Evidence grade: A +- Sources: General Office of the State Council and General Office of the Central Military Commission (2025). Provisions on Military Service Registration《兵役登记工作规定》（国办发〔2025〕41 号）. ；Standing Committee of the National People's Congress (2021). Military Service Law（兵役法）（2021 revision, articles twenty, twenty-five, twenty-six, fifty-four and fifty-seven）. （published by the Ministry of National Defense）；State Council and Central Military Commission (2023). Regulations on Military Conscription《征兵工作条例》（国务院、中央军委令第 759 号，article five）. +- Notes: Registration and being called up are two different things, and the cost differs by a long way. The set of penalties that stops you leaving the country within two years only applies to people who back out after being called up, see item 3 in this section (refusing military service). Women are not required to register initially, and are called up according to the needs of the armed forces. Soldiers moving on to a military academy, and good soldiers being promoted to officers, have their conditions written into the internal rules of the armed forces, which cannot be looked up from outside. The Military Service Law says only two things: active-duty officers are selected and recruited from among 「表现优秀的现役士兵」, serving soldiers with outstanding performance, and the military academies are not subject to the call-up age limit when they recruit students. This book does not give the specific score lines and the quotas. + +### 3. If you refuse to serve after being called up, you cannot leave the country or go back to school within two years, and you cannot join the civil service or a state-owned enterprise + +- Cost: No money. The cost is that before you sign up you have to settle in your own mind whether you are going. +- In plain language: If you sign up and then refuse to go and serve, there is a list of consequences. You cannot be recruited as a civil servant, and state-owned enterprises and public institutions cannot hire you either. You cannot leave the country for two years, you cannot go up a level of education, and you cannot go back to school to continue your studies. You also get put on the list of subjects in serious breach of national defence obligations. So "sign up first and come back if it does not work" is an expensive idea. +- Benefit: Item two of paragraph one of article fifty-seven of the Military Service Law is 「应征公民拒绝、逃避征集服现役的」, a citizen called up who refuses or evades being called up for active service. The penalty is likewise 「由县级人民政府责令限期改正；逾期不改正的，由县级人民政府强制其履行兵役义务，并处以罚款」, the county-level people's government orders correction within a set time; if the correction is not made by the deadline, it compels the person to perform military service and imposes a fine. Paragraph two of the same article: 「有前款第二项行为，拒不改正的，不得录用为公务员或者参照《中华人民共和国公务员法》管理的工作人员，不得招录、聘用为国有企业和事业单位工作人员，两年内不准出境或者升学复学，纳入履行国防义务严重失信主体名单实施联合惩戒。」, that is, a person who does the second item of the preceding paragraph and refuses to correct it may not be recruited as a civil servant or as a worker managed with reference to the Civil Servant Law of the People's Republic of China（中华人民共和国公务员法）, may not be recruited or appointed as a worker of a state-owned enterprise or a public institution, may not leave the country or go back to school to continue their education within two years, and is put on the list of subjects in serious breach of national defence obligations for joint punishment. Joint punishment means several departments work from the same list and enforce the restrictions above together. Article fifty-eight applies the same set of sanctions to people who have already enlisted. Where a serviceman refuses to perform his duties or leaves the unit with the purpose of evading military service, the consequences depend on how it is dealt with. Where a person is struck off the rolls of the army, has military rank stripped, or is held criminally responsible in accordance with law, the sanctions in paragraph two of article fifty-seven apply. Among these, being struck off the rolls also carries a fine (nationwide, in force from 2021-10-1) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (2021). Military Service Law（兵役法）（2021 revision, articles fifty-seven and fifty-eight）. （published by the Ministry of National Defense） +- Notes: What is counted here is personal freedom and the right to be employed, not money. This list of sanctions applies to only two kinds of person. One is "a citizen called up who refuses or evades being called up for active service". The other is someone who, after enlisting, is struck off the rolls, has military rank stripped, or is held criminally responsible. Merely failing to register for military service is not in it; military service registration is item 2 in this section. The fine presupposes that the person "refuses to correct it", and it is the county-level people's government that determines it and imposes it. How much the fine is varies from place to place and this book does not give it. Whether you go is yours to decide, but decide it before you sign up. +### 4. The two solidest things a university student gets from serving: tuition reimbursed up to 20000 yuan a year, and the student place kept for two years with a separate plan for the postgraduate entrance examination + +- Cost: No money. After being discharged you hand in the documents the school asks for and apply, and it takes one or two trips. +- In plain language: Students on campus, graduating students and new students who go to serve have their tuition paid by the state for the period of service. For undergraduate and junior college students it is up to 20000 yuan each per year, and for postgraduates up to 25000 yuan each per year, so over two years that is up to forty or fifty thousand yuan. Your student place is kept, and within two years of being discharged you may enrol or go back. The postgraduate entrance examination has a separate plan that admits only discharged student soldiers. +- Benefit: A 2024 notice from the Ministry of Finance and three other departments says that students at higher education institutions going to serve military service have three kinds of support: reimbursement of tuition, repayment on behalf of the student of state student loans used for tuition, and a reduction of tuition. Repaying the loan on the student's behalf means the state pays that student loan off. The standards for all three are 「相应调整为本专科学生每人每年最高不超过 20000 元、研究生每人每年最高不超过 25000 元」, that is, adjusted so that undergraduate and junior college students get up to 20000 yuan each per year and postgraduates up to 25000 yuan each per year. The new standards apply from the autumn term of 2024. Exactly how much is reimbursed depends on this line: 「按学生实际缴纳的学费或获得的国家助学贷款（包括本金及其全部偿还之前产生的利息）两者金额较高者执行」, that is, it is paid at whichever is larger, the tuition the student actually paid or the state student loan obtained, including the principal and all the interest accrued before it is fully repaid. There is a further line: 「超出标准部分不予补偿、代偿或减免」, the part above the standard is not reimbursed, not repaid on the student's behalf and not reduced. For the student place, look at the Law on the Protection of Veterans《退役军人保障法》. It states: 「现役军人入伍前已被普通高等学校录取或者是正在普通高等学校就学的学生，服现役期间保留入学资格或者学籍，退役后两年内允许入学或者复学，可以按照国家有关规定转入本校其他专业学习」, that is, a student who has been admitted by an ordinary higher education institution before enlisting, or who is studying at one, keeps their admission place or student place during active service, and within two years of being discharged may enrol or go back, and may transfer to another programme in the same institution in accordance with the provisions of the state. The same law also states: 「高等学校根据国家统筹安排，可以通过单列计划、单独招生等方式招考退役军人」, that is, higher education institutions, in accordance with the overall arrangements of the state, may recruit veterans through separately allocated plans and separate enrolment. For the postgraduate entrance examination, look at the 2026 notice from the General Office of the Ministry of Education. The plan is called the "Plan for Discharged Student Soldiers"「退役大学生士兵计划」. The notice states: 「在全国硕士研究生招生计划总规模内单列下达、专项使用，不得调整为普通计划或其他专项计划。招生单位自主确定……初试成绩要求（一般不高于本单位相应学科专业普通计划考生初试成绩要求）……同等条件下服役时间长的考生优先录取」, that is, it is allocated separately within the overall national plan for postgraduate enrolment and used for that purpose only, and may not be converted into the ordinary plan or any other special plan. The enrolling institution decides for itself the requirements for the first-round examination score (generally no higher than the first-round score required of candidates in the ordinary plan for the same discipline). Under equal conditions, candidates who served longer are admitted first. Allocated separately and used for that purpose only means this part of the quota is set aside and can be used only to recruit discharged student soldiers. In 2026 this plan is carried by 532 ordinary higher education institutions (nationwide) +- Evidence grade: A +- Sources: Ministry of Finance, Ministry of Education, People's Bank of China and National Financial Regulatory Administration (2024). Notice on Adjusting and Improving Relevant Policies on State Student Loans（关于调整完善国家助学贷款有关政策的通知）（财教〔2024〕188 号）. ；Ministry of Finance and four other departments (2019). Measures for the Administration of Student Financial Aid Funds《学生资助资金管理办法》（财科教〔2019〕19 号，article five）. ；Standing Committee of the National People's Congress (2020). Law on the Protection of Veterans《退役军人保障法》（articles thirty-four and thirty-five）. （published by the Ministry of National Defense）；General Office of the Ministry of Education (2025). Notice on Doing Well the 2026 Enrolment Work under the Separate Postgraduate Plan for Discharged Student Soldiers《关于做好 2026 年退役大学生士兵专项硕士研究生招生计划招生工作的通知》（教学厅〔2025〕8 号）. ；Ministry of Veterans Affairs (2022). 23 points on education and training policy for discharged soldiers「退役士兵教育培训政策 23 条」: 「从2022年招生起，高职（专科）毕业生及在校生（含高校新生）应征入伍，退役后完成高职（专科）学业的，申请专升本，免于参加文化课考试」(based on 教学厅〔2021〕8 号). ；Ministry of Education (2026). 2027 National Postgraduate Enrolment Work Management Provisions（2027 年全国硕士研究生招生工作管理规定）（教学〔2026〕2 号，articles twenty-two, twenty-four and twenty-five），published on the China postgraduate admissions website. ；Ministry of Education (2024). 2025 National Postgraduate Enrolment Work Management Provisions（2025 年全国硕士研究生招生工作管理规定）（教学〔2024〕4 号，article twenty-two）. +- Notes: There are two more routes. First, exemption from the cultural examination for the upgrade to a bachelor's degree: a junior college student goes to serve, finishes the junior college course after being discharged, and can apply to be upgraded to a bachelor's degree without the examination, with the school testing practical vocational skill and admitting on merit. This has applied from the 2022 enrolment onwards, and the specific rules are in your province's notice for that year. Second, 10 points added to the first-round examination for the postgraduate entrance examination: a discharged student soldier who sits the postgraduate entrance examination within 3 years of being discharged gets 10 points added to the total first-round score. If you have applied under the separate plan you do not get the points; the two cannot both be enjoyed. Those who have been awarded the second-class merit or above may apply to be admitted to postgraduate study without the first-round examination. The list of institutions running the separate plan changes every year; in 2025 it was 523. The tuition standards have been adjusted twice: in 2019 they were 8000/12000, in the autumn of 2023 they became 16000/20000, and in the autumn of 2024 they became 20000/25000. The summary the Ministry of Veterans Affairs published in 2022 still has the earliest version in it; do not copy it. + +### 5. After being discharged the government arranges work in only six situations; those who take up self-employment must report within 30 days + +- Cost: No money. Take your certificate of discharge and go to the veterans affairs department in the place of resettlement once. +- In plain language: "Serve and you get a post assigned" stopped being the general case long ago. Now there are only six situations in which the government arranges work, and a conscript who serves two years and is discharged is in most cases none of them. The category where the government does not arrange work is called self-employment, and it has a hard deadline: you must report within 30 days of leaving active service. If you go more than 30 days without a legitimate reason, you are treated as having given up the resettlement treatment. +- Benefit: Article twenty-five of the Regulations on the Resettlement of Veterans《退役军人安置条例》lists six situations in which the people's government of the place of resettlement arranges work. The wording is: 「（一）军士服现役满 12 年的；（二）服现役期间个人获得勋章、荣誉称号的；（三）服现役期间个人荣获三等战功、二等功以上奖励的；（四）服现役期间个人获得一级表彰的；（五）因战致残被评定为 5 级至 8 级残疾等级的；（六）是烈士子女的」, that is: (one) non-commissioned officers who have completed 12 years of active service; (two) those who personally received a medal or an honorary title during their active service; (three) those who personally won third-class battle merit or a second-class merit or above during their active service; (four) those who personally received a first-level commendation during their active service; (five) those whose disability from injury in battle was assessed at level 5 to level 8; (six) the children of martyrs. The same Regulations also state: 「安置地人民政府应当在接收退役军士和义务兵的 6 个月内完成安排退役军士和义务兵工作的任务」, that is, the people's government of the place of resettlement shall finish the task of arranging work for discharged non-commissioned officers and conscripts within 6 months of taking them over. An employer 「应当按照规定与其签订不少于 3 年的中长期聘用合同或者劳动合同。其中，企业接收军龄 10 年以上的退役军士的，应当与其签订无固定期限劳动合同」, that is, shall sign with them a medium or long term employment contract or labour contract of not less than 3 years as required. Among these, where an enterprise takes on a discharged non-commissioned officer with 10 or more years of service, it shall sign an open-ended labour contract with them. An open-ended contract is one with no end date on it. While waiting to be given work, 「按照当地月最低工资标准逐月发放生活补助」, that is, a living allowance is paid month by month at the local monthly minimum wage standard. Where a person has not been placed in a post as required, and it is not for reasons of their own, 「按照不低于本单位同等条件人员平均工资 80% 的标准，逐月发放生活费直至上岗为止」, that is, a living allowance is paid month by month at a standard of not less than 80% of the average wage of people on equivalent terms in the same unit, until they take the post. The situations in which work may not be arranged as resettlement include being expelled from the Communist Party of China and having been criminally punished. For the deadline for self-employment, look at article fifty-one: 「自主就业的退役军士和义务兵，应当自被批准退出现役之日起 30 日内，到安置地人民政府退役军人工作主管部门报到。无正当理由不按照规定时间报到超过 30 日的，视为放弃安置待遇。」, that is, discharged non-commissioned officers and conscripts taking up self-employment shall, within 30 days of the date they are approved for discharge from active service, report to the veterans affairs department of the people's government of the place of resettlement. Where they do not report by the prescribed time, without legitimate reason, for more than 30 days, this is treated as having given up the resettlement treatment (nationwide, in force from 2024-9-1) +- Evidence grade: A +- Sources: State Council and Central Military Commission (2024). Regulations on the Resettlement of Veterans《退役军人安置条例》（国务院、中央军委令第 787 号，promulgated 2024-7-29, in force 2024-9-1; articles twenty-five, thirty-one, thirty-two, thirty-five, fifty-one and seventy-five）. ；Standing Committee of the National People's Congress (2020). Law on the Protection of Veterans《退役军人保障法》（articles thirteen, fourteen and fifteen）. （published by the Ministry of National Defense） +- Notes: The 2011 Regulations on the Resettlement of Discharged Soldiers（国务院、中央军委令第 608 号）were repealed on the same day these Regulations came into force. A great many guides online still quote the old Regulations, and the wording of the conditions has changed: the old version said "second-class merit or above in peacetime, or third-class merit or above in wartime". The 30-day deadline governs only the self-employment category; other forms of resettlement report "at the prescribed time". When you report you are issued with the veterans treatment card, and the discharge pay, the employment and entrepreneurship support and the educational aid all connect on from there. + +### 6. Three accounts after being discharged: the one-off discharge pay, service years counted as years of employment, and 20000 yuan off your tax a year if you go into business yourself + +- Cost: No money. When you are discharged the armed forces transfer your social insurance relationship and your money out to you. You then go to the social insurance agency in the place of resettlement (the counter that handles social insurance) to continue it. +- In plain language: After being discharged there are three accounts. The first is the one-off discharge pay for those taking up self-employment, paid according to the years of active service, and increased where the person was decorated. The second is that the years you served do not count as a gap in social insurance; service counts as years of employment, and the contribution years are added together with those before you enlisted and after you were discharged. The third is tax: if you open a shop or run a business yourself, for 3 years you can have up to 20000 yuan less tax a year. +- Benefit: For the first, look at the Regulations on the Resettlement of Veterans《退役军人安置条例》. It states: 「对自主就业的退役军士和义务兵，根据其服现役年限发放一次性退役金」, that is, a one-off discharge pay is paid to discharged non-commissioned officers and conscripts taking up self-employment on the basis of their years of active service. Who pays it: 「由中央财政专项安排，具体标准由国务院退役军人工作主管部门、中央军事委员会政治工作部门会同国务院财政部门……确定，并适时调整」, that is, it is set aside by the central finance, and the specific standard is determined and adjusted from time to time by the veterans work department of the State Council together with the political work department of the Central Military Commission and the financial department of the State Council. The increases for decoration come in five bands. A medal or an honorary title adds 25%; first-class battle merit or a first-level commendation adds 20%. Second-class battle merit, first-class merit or a second-level commendation adds 15%. Third-class battle merit or second-class merit adds 10%, and fourth-class battle merit or third-class merit adds 5%. The same Regulations also state: 「退役军人未及时就业的，可以依法向户口所在地人力资源社会保障部门申领失业保险待遇，服现役年限视同参保缴费年限，但是以退休、供养方式安置的退役军人除外」, that is, a veteran who has not found work in good time may apply in accordance with the law to the human resources and social security department where their household registration is for unemployment insurance benefits, and years of active service count as years of contribution, except for veterans resettled under retirement or dependants' support. Counting years as contribution years means the years in which you paid nothing are counted as though you had. For the second, look at the Law on the Protection of Veterans《退役军人保障法》: 「退役军人服现役年限计算为工龄，退役后与所在单位工作年限累计计算」, that is, years of active service are counted as years of employment and, after discharge, are added to the years worked at the unit. Years of employment means the years spent working, so the years of service and the years after discharge carry on adding up. The same law also states that years of active service and the years of contribution to pension, medical and unemployment insurance before enlistment and after discharge are 「依法合并计算」, that is, lawfully combined, and that 「退役的军士和义务兵服现役经历视为基层工作经历」, that is, the active service of discharged non-commissioned officers and conscripts counts as work experience at the grassroots level. The Law on Social Insurance for Servicemen《军人保险法》: 「军人服现役年限视同职工基本医疗保险缴费年限，与入伍前和退出现役后参加职工基本医疗保险的缴费年限合并计算」, that is, years of active service count as years of contribution to the basic medical insurance for employees, and are combined with the years contributed before enlistment and after discharge. For the third, look at Announcement No. 14 of 2023 from the Ministry of Finance, the State Taxation Administration and the Ministry of Veterans Affairs. The announcement deals with discharged soldiers taking up self-employment who run individual businesses: 「在 3 年（36 个月，下同）内按每户每年 20000 元为限额依次扣减其当年实际应缴纳的增值税、城市维护建设税、教育费附加、地方教育附加和个人所得税。限额标准最高可上浮 20%」, that is, within 3 years (36 months, the same applies below) a limit of 20000 yuan per household per year is deducted in order from the value-added tax, urban maintenance and construction tax, education surcharge, local education surcharge and individual income tax they should actually have paid that year. The limit may be raised by up to 20%. The deduction works by coming down the list of those taxes in order for that year until the limit is used up; it is not money paid to you. Where an enterprise hires them, 「定额标准为每人每年 6000 元，最高可上浮 50%」, that is, the fixed amount is 6000 yuan per person per year, and may be raised by up to 50%. The period of operation runs from 2023-1-1 to 2027-12-31. Where 3 years have not run out at the end, it can continue to be enjoyed until they are up (nationwide) +- Evidence grade: A +- Sources: State Council and Central Military Commission (2024). Regulations on the Resettlement of Veterans《退役军人安置条例》（国务院、中央军委令第 787 号，articles twenty-one, twenty-two, seventy-four, eighty-one and eighty-five）. ；Ministry of Veterans Affairs (2020). Ten questions on the one-off discharge pay（关于一次性退役金的十个问题）(reprinted from China Veterans, issue 10 of 2020): 「基本标准。士兵服现役每满一年，发给4500元」「退役金标准适时调整」. ；Guizhou Provincial Department of Veterans Affairs (2024). The one-off discharge pay is increased, an important change！「一次性退役金增发，重要变化！」: 「现在执行的标准是：士兵服现役每满一年，发给4500元」. ；Muping District People's Government, Yantai (2025). Quoting the Notice of the Yantai Veterans Affairs Bureau and the Yantai Finance Bureau on Adjusting the Standard of the One-off Economic Subsidy for Discharged Soldiers Taking Up Self-employment「烟台市退役军人事务局、烟台市财政局关于调整自主就业退役士兵一次性经济补助标准的通知」in its reply: 「2024年退出现役的每服役一年补助6000元」. ；Standing Committee of the National People's Congress (2020). Law on the Protection of Veterans《退役军人保障法》（articles forty-two, forty-four and fifty）. （published by the Ministry of National Defense）；Standing Committee of the National People's Congress (2012). Law on Social Insurance for Servicemen《军人保险法》（articles sixteen and twenty-three）. （National Database of Laws and Regulations）；Ministry of Finance, State Taxation Administration and Ministry of Veterans Affairs (2023). Announcement on Further Tax Policies to Support the Entrepreneurship and Employment of Discharged Soldiers Taking Up Self-employment《关于进一步扶持自主就业退役士兵创业就业有关税收政策的公告》(2023 No. 14). +- Notes: The Regulations only say that the amount is set separately by the department in charge. The official figure you can find is the one in an article the Ministry of Veterans Affairs published on its own website in 2020, "4500 yuan for every full year of active service by a soldier"「服现役每满一年，发给4500元」, and the same article says it is "adjusted from time to time"「适时调整」. The decoration increases in that article no longer match the current Regulations; go by the Regulations. An article the Guizhou Provincial Department of Veterans Affairs published in August 2024, under the new Regulations, still says "the standard in force now is: 4500 yuan for every full year of active service by a soldier"「现在执行的标准是：士兵服现役每满一年，发给4500元」. TODO (to be verified: no formal standard document has been found; the most that can be confirmed is that a provincial department still wrote this in August 2024). And do not mix the local "one-off economic subsidy" up with it. That is another sum, set and paid by each place itself: in Yantai it is 6000 yuan a year of service for those discharged in 2024. Most of the new figures in guides have no source; some of them simply mix the two sums together. While a conscript is serving, the family can also receive a separate family preferential payment. The Military Service Law only says that "the standard is set by the local people's government, and the central finance gives a fixed-amount subsidy"「标准由地方人民政府制定，中央财政给予定额补助」, so ask the county-level people's government that approved your enlistment how much it is. The tax account is the most you can have off your tax each year. If the business does not get off the ground you will not use the whole of it, which is why the benefit for the whole item is graded only medium. The Regulations on Promoting the Employment and Entrepreneurship of Veterans《退役军人就业创业促进条例》, in force from 2026-8-1, add one more thing: a vocational qualification certificate for a skilled worker obtained during service is valid nationwide after discharge. +### 7. Getting into an establishment is not only the civil service examination: two years of grassroots service let you apply for the targeted recruitment quota of about 10% a year + +- Cost: No money. The cost is two years of grassroots service in a township. What you get for those two years is a fiscal subsidy, not a market wage, and the hard part is lasting the two years. +- In plain language: If you want into an establishment, the civil service examination is not the only route. Serve the full two years on a grassroots service programme of the kind called "Supporting and Assisting Agriculture, Supporting and Assisting Rural Revitalisation, Supporting and Assisting Border Areas and Supporting and Assisting Poverty Alleviation"「三支一扶」, pass the assessment, and you can apply for a targeted recruitment quota that is open only to people who have done that. Each province has to set aside around 10% of its civil service recruitment plan for these people each year. Sit the postgraduate entrance examination within three years and 10 points are added to your first-round total. The cost is that for those two years you are on a fiscal subsidy. +- Benefit: The notice for the fourth round of the "Three Supports and One Assistance" plan「三支一扶」says "the period of service is two years"「服务期为两年」. It also says: 「各省（区、市）每年应拿出公务员考录计划的 10％左右，面向『三支一扶』计划等服务基层项目人员定向考录。各省（区、市）县乡基层事业单位公开招聘时，应根据本地区实际拿出一定数量或比例的岗位，对『三支一扶』服务期满考核合格的人员进行专项招聘……聘用后可以不再约定试用期」, that is, each province, autonomous region and municipality should set aside around 10% of its civil service recruitment plan each year for targeted recruitment of people who have served on grassroots service programmes such as the "Three Supports and One Assistance" plan. When public institutions at county, township and village level in each province, autonomous region and municipality recruit openly, they should, according to local conditions, set aside a certain number or proportion of posts for special recruitment of people who have completed their period on the "Three Supports and One Assistance" plan and passed the assessment, and after appointment no probation period needs to be agreed. Targeted recruitment and special recruitment mean this part of the quota is examined only among people who have done grassroots service; nobody else can apply. The same notice also says: 「期满考核合格的『三支一扶』人员，三年内参加全国硕士研究生招生考试的，初试总分加 10 分，同等条件下优先录取……高职（高专）毕业生期满且考核合格的，可免试入读成人高等学历教育专科起点本科」, that is, a person who has completed the period on the "Three Supports and One Assistance" plan and passed the assessment, who sits the national postgraduate entrance examination within three years, has 10 points added to the first-round total and is admitted first under equal conditions; a graduate of a higher vocational (junior college) course who has completed the period and passed the assessment may be admitted without examination to the undergraduate part of an adult higher education programme starting from the junior college stage. Being admitted without examination means sitting no further exam and going straight into the bachelor's degree course. The notice also says: 「参加『三支一扶』计划前无工作经历的人员期满且考核合格的，两年内在参加机关和企事业单位考录（招聘）、自主创业、落户、升学等方面可同等享受应届毕业生相关政策。『三支一扶』人员在基层服务年限计算为工龄，其参加工作时间按其到基层报到之日起算」, that is, a person with no work experience before joining the "Three Supports and One Assistance" plan who completes the period and passes the assessment may, for two years, enjoy on the same terms the policies for fresh graduates as regards applying to enter an organ or a public or commercial enterprise, starting a business of their own, settling a household registration, and going up a level of education. The years of grassroots service count as years of employment, and the time counted as their working life runs from the date they reported to the grassroots post. The central finance subsidy is 30,000 yuan each per year in the western region, 24,000 yuan in the central region and 12,000 yuan in the eastern region. The opinion of the General Offices of the CPC Central Committee and the State Council states: 「参加基层服务项目前无工作经历的人员服务期满且考核合格后 2 年内，在参加机关事业单位考录（招聘）、各类企业吸纳就业、自主创业、落户、升学等方面可同等享受应届高校毕业生的相关政策」, that is, a person with no work experience before joining a grassroots service programme who completes the period and passes the assessment may, for 2 years, enjoy on the same terms the policies for fresh university graduates as regards applying to enter an organ or public institution, taking up employment with any kind of enterprise, starting a business of their own, settling a household registration, and going up a level of education. The same opinion also states: 「省级以上机关录用公务员，除特殊职位外，按照有关规定一律从具有 2 年以上基层工作经历的人员中考录」, that is, civil servants recruited by organs at provincial level or above are, except for special posts, to be recruited as required in every case from people with more than 2 years of grassroots work experience. Another route is the Western Plan. It is organised jointly by 「共青团中央、教育部、财政部、人力资源社会保障部共同组织实施」, that is, the Central Committee of the Communist Youth League, the Ministry of Education, the Ministry of Finance and the Ministry of Human Resources and Social Security. It recruits recent university graduates or postgraduates who are still studying to 「到西部基层从事为期 1—3 年」的志愿服务, that is, to do voluntary service at the grassroots level in the west for a period of 1 to 3 years (nationwide) +- Evidence grade: A +- Sources: Central Organization Department of the CPC Central Committee, Ministry of Human Resources and Social Security and eight other departments (2021). Notice on Implementing the Fourth Round of the "Three Supports and One Assistance" Plan for University Graduates「关于实施第四轮高校毕业生『三支一扶』计划的通知」（人社部发〔2021〕32 号，articles seven, eight, nine, eleven, twelve and thirteen）. ；General Office of the CPC Central Committee and General Office of the State Council (2017). Opinion on Further Guiding and Encouraging University Graduates to Work at the Grassroots Level《关于进一步引导和鼓励高校毕业生到基层工作的意见》（articles five, sixteen and eighteen）. ；National Office for the Administration of the Western Plan for University Student Volunteer Service (2026). 2026—2027 Enrolment Agreement for the Western Plan for University Student Volunteer Service「2026—2027 年度大学生志愿服务西部计划招募协议书」. +- Notes: The fourth round of the "Three Supports and One Assistance" plan covers 2021 to 2025. The document for the fifth round and the notice for 2026 were not found this time, so check that year's recruitment announcement before you sign up. In 2025 the central finance supported the recruitment of 34430 people. The implementation plan for the Western Plan could not be found online, and the policy section of its own website would not open. For the benefits after it ends, go by the document from the two General Offices that covers all grassroots service programmes. What you get for those two years is a fiscal subsidy, not a market wage, and that cost has to be counted together with the targeted recruitment two years later. + +### 8. Special-post teachers who finish the three-year service period, pass the assessment and want to stay should be put on the establishment in good time + +- Cost: No money. The cost is three years teaching in a rural compulsory education school, and before you start you have to pass the teacher qualification examination. The hard part is lasting those three years. +- In plain language: The special-post plan is one of the few routes that writes "admitted to the establishment when the period ends" into the rules. The conditions are an undergraduate degree or above, mainly in teacher-training subjects, an age of not more than 30, and the matching teacher qualification certificate. The period of service is three years, and for those three years the pay and the treatment are the same as those of established teachers in local public schools. Where someone passes the assessment at the end of the three years and wants to stay, the document requires them to be put on the establishment in good time. +- Benefit: The 2024 special-post plan notice gives the recruitment conditions as: 「要求本科及以上学历，以师范类专业为主，小学阶段可适当招聘师范高等专科学校毕业生」, that is, an undergraduate degree or above is required, mainly in teacher-training subjects, and at primary stage graduates of higher vocational teacher-training colleges may be recruited as appropriate; 「年龄不超过 30 周岁」, aged not more than 30; and the matching teacher qualification certificate must be held. The notice also states: 「参加过『大学生志愿服务西部计划』、有从教经历的志愿者和参加过半年以上实习支教的师范院校毕业生同等条件下优先录取」, that is, volunteers who took part in the "Western Plan for University Student Volunteer Service" and have teaching experience, and graduates of teacher-training colleges who took part in at least half a year of teaching practice, are admitted first under equal conditions. And: 「支持鼓励符合条件的退役军人到中小学任教，可综合考虑服役年限等因素对退役军人相应放宽年龄限制」, that is, veterans who meet the conditions are supported and encouraged to teach in primary and secondary schools, and the age limit may be relaxed for them by taking factors such as length of service into account. For the treatment and the establishment, look at these lines: 「从 2024 年 1 月 1 日起，提高特岗教师工资性补助标准，中央财政……中部地区由年人均 3.52 万元提高到 3.88 万元，西部地区由 3.82 万元提高到 4.18 万元」, that is, from 2024-1-1 the standard of the wage subsidy for special-post teachers is raised, and the central finance provides it, in the central region from 35,200 yuan per person per year to 38,800 yuan and in the western region from 38,200 yuan to 41,800 yuan; 「同等条件下在薪酬待遇、职称评聘、评先评优、年度考核等方面享受与当地公办学校在编教师同等待遇」, that is, under equal conditions they enjoy the same treatment as established teachers in local public schools as regards pay, professional title assessment, commendation and annual appraisal; and 「保证三年服务期满、考核合格且愿意留任的特岗教师及时入编并落实工作岗位，做好相关人事、工资关系等接转工作，连续计算工龄、教龄，不再实行试用期」, that is, it is guaranteed that special-post teachers who complete the three-year period, pass the assessment and are willing to stay are put on the establishment in good time and given a post, that the relevant personnel and wage relationships are transferred over, that years of employment and years of teaching are counted continuously, and that no probation period applies (nationwide) +- Evidence grade: A +- Sources: General Office of the Ministry of Education and General Office of the Ministry of Finance (2024). Notice on Doing Well the Implementation of the 2024 Special-Post Teacher Plan for Rural Compulsory Education Schools《关于做好 2024 年农村义务教育阶段学校教师特设岗位计划实施工作的通知》（教师厅〔2024〕1 号）. +- Notes: The benefit is graded large because being put on the establishment when the period ends is the same as getting a long-term established post, and that is worth far more than ten thousand yuan. The most recent document you can find is the notice for 2024; those for 2025 and 2026 are not in the policy database. For the specific posts, the subsidy standards and the age, go by that year's notice and by your province's implementation plan. The central level sets the framework and the province sets the detail. The teacher qualification certificate has to be passed first; it cannot be made up after you start. + +### 9. Firefighters and civilian posts in the armed forces are two more openly advertised routes, and the thresholds are lower than for the civil service + +- Cost: No money. From applying to being appointed you go through a medical examination, a political assessment, a physical fitness test, a psychological test and an interview, over several months. The hard part is that each of these stages washes people out. +- In plain language: Neither route has the "college education or above" threshold. Firefighters are recruited from people aged 18 to 22 with an educational level of senior high school or above, and people with a college education or above and discharged soldiers may have the age relaxed to 24. Civilian posts in the armed forces require only that you be at least 18 years old, and the highest age for first recruitment to entry-level posts is 35. What both routes share is that the political assessment and the medical examination wash people out. +- Benefit: For firefighters, look at the Measures for the Recruitment of Firefighters into the National Comprehensive Fire and Rescue Force《国家综合性消防救援队伍消防员招录办法》. The basic conditions in article five include 「年龄为 18 周岁以上、22 周岁以下」, aged 18 or above and under 22, and 「具有高中以上文化程度」, having an educational level of senior high school or above. Article six: 「大学专科以上学历人员、解放军和武警部队退役士兵、具有 2 年以上灭火救援实战经验的政府专职消防队员和政府专职林业扑火队员，年龄可以放宽至 24 周岁；对消防救援工作急需的特殊专业人才，经应急管理部批准年龄可以进一步放宽，原则上不超过 28 周岁」, that is, for people with a college education or above, for soldiers discharged from the People's Liberation Army and the Armed Police Force, and for government full-time firefighters and government full-time forest fire-fighting teams with more than 2 years of practical experience in fire fighting and rescue, the age may be relaxed to 24; for special professional talent urgently needed for fire and rescue work, with the approval of the Ministry of Emergency Management the age may be relaxed further, in principle to no more than 28. Article seven: 「消防员面向社会公开招录，主要从本省级行政区域常住人口中招录，根据需要也可以面向其他省份招录」, that is, firefighters are recruited openly from the public, mainly from among the resident population of the province-level administrative region, and according to need also from other provinces. Article nine sets out the recruitment procedure as 「宣传动员、组织报名、资格审查、体格检查、政治考核、体能测试和岗位适应性测试、心理测试和面试、公示、录用」, that is, publicity and mobilisation, organising applications, qualification review, physical examination, political assessment, physical fitness test and job suitability test, psychological test and interview, publication of the results, and appointment. For civilian posts in the armed forces, look at article eighteen of the Regulations on Civilian Personnel of the People's Liberation Army of China《中国人民解放军文职人员条例》: 「军队实行公开招考、直接引进、专项招录相结合的文职人员招录聘用制度」, that is, the armed forces operate a system for the recruitment and appointment of civilian personnel that combines open competitive recruitment, direct introduction and special recruitment. That is three lines: recruiting people through an open examination, bringing in talent who already exists directly, and recruiting for a specific group on purpose. The basic conditions in article nine include 「年满 18 周岁」, being at least 18 years old; 「符合军队招录聘用文职人员的政治条件」, meeting the political conditions for the armed forces to recruit and appoint civilian personnel; and 「符合岗位要求的文化程度、专业水平和工作能力」, having the educational level, professional standard and working ability the post requires. Article nineteen states that the highest age for first recruitment and appointment to posts at a lower grade is 35. The same Regulations also state that 「文职人员岗位应当优先招录聘用符合条件的退役军人」, that is, civilian posts should give priority in recruitment and appointment to veterans who meet the conditions (nationwide) +- Evidence grade: A +- Sources: Ministry of Human Resources and Social Security and Ministry of Emergency Management (2021). Measures for the Recruitment of Firefighters into the National Comprehensive Fire and Rescue Force《国家综合性消防救援队伍消防员招录办法》（人社部发〔2021〕58 号，articles five, six, seven and nine）. ；State Council and Central Military Commission (2022). Regulations on Civilian Personnel of the People's Liberation Army of China《中国人民解放军文职人员条例》（second revision, Order No. 757 of the State Council and the Central Military Commission, in force 2023-1-1; articles nine, eighteen, nineteen and twenty）. （National Database of Laws and Regulations） +- Notes: The Regulations on Promoting the Employment and Entrepreneurship of Veterans《退役军人就业创业促进条例》, in force from 2026-8-1, set out two more things. First, a certain proportion of firefighter recruitment has to be set aside for discharged non-commissioned officers and conscripts. Second, when recruiting police auxiliary staff, veterans are to be given priority under equal conditions. The number of places, the posts and the specific conditions for both of these routes each year are set by that year's announcement; this book only writes down the thresholds that are fixed in the rules. Firefighting is a post in a disciplined service, so weigh up the occupational risk before you go in. + +### 10. A qualification is not a one-off purchase: the self-study examination has no limit on age or on the education you already have, and the adult entrance examination and the open university can both complete it + +- Cost: Several hundred yuan a year in registration and course materials. The cost is long-term self-study, or going to classes in your spare time; the shortest part-time route to a bachelor's degree, studied while working, takes two and a half years. The hard part is keeping it up over a long period. +- In plain language: If you did not get into education at eighteen, there are still three proper channels: the self-study examination, the adult entrance examination and the open university. The self-study examination has no limit on age and does not look at how far you got before, but students on full-time courses may not apply. The adult entrance examination requires you already to hold a certificate from senior high school or from a junior college. All three routes are slow, but none of them looks at how you did at eighteen. +- Benefit: For the self-study examination, look at the Interim Regulations on Higher Education Self-Study Examinations《高等教育自学考试暂行条例》. Article three: 「中华人民共和国公民，不受性别、年龄、民族、种族和已受教育程度的限制，均可依照本条例的规定参加高等教育自学考试」, that is, citizens of the People's Republic of China, without restriction as to sex, age, ethnicity, race or level of education already received, may all take part in higher education self-study examinations in accordance with these Regulations. Article twenty: 「各级各类全日制学校的在校生不得报考」, students enrolled on full-time courses at schools of every level and type may not apply. Article twenty-five sets out the three conditions for getting the certificate: passing and passing all the courses, completing the graduation thesis or the practical assignment, and passing the assessment of moral character. Do all of these and you may obtain the certificate. This article also states that 「获得专科（基础科）或本科毕业证书者，国家承认其学历」, that is, the state recognises the qualification of those who obtain a junior college (foundation) or undergraduate certificate. Article thirty-two sets out the pay and treatment of those who obtain the certificate: 「非在职人员录用后，与普通高等学校同类毕业生相同」, that is, on appointment, those who are not in work are treated the same as comparable graduates of ordinary higher education institutions. Article twenty-one of the Law on Higher Education《高等教育法》: 「国家实行高等教育自学考试制度，经考试合格的，发给相应的学历证书或者其他学业证书」, that is, the state operates a system of higher education self-study examinations, and those who pass are issued with the corresponding certificate of qualification or other certificate of attainment. For the adult entrance examination, look at the 2025 Measures for the Enrolment of Adult Higher Education Institutions nationwide《2025 年全国成人高校招生办法》. Those eligible to apply are 「国家承认学历的各类高、中等学校在校生以外的从业人员和社会其他人员」, that is, people who work and other people in society, other than students enrolled at recognised higher and secondary schools. 「报考高起本或高起专的考生应高级中等教育学校毕业或者具有同等学力。报考专升本的考生必须是已取得……专科毕业证书、本科结业证书或以上证书的人员」, that is, candidates applying to start from the first year of the undergraduate or the first year of the junior college course should have graduated from a senior secondary education school or have the equivalent academic ability; candidates applying to upgrade to a bachelor's degree must already hold a junior college certificate, an undergraduate completion certificate or a higher certificate. 「脱产最短学习时间为：高起本四年、高起专和专升本两年，非脱产最短学习时间为：高起本五年、高起专和专升本两年半」, that is, the shortest period of study on a full-time basis is four years for starting the undergraduate course, and two years for the junior college course and for upgrading to a bachelor's degree; the shortest period of study while working is five years for starting the undergraduate course, and two and a half years for the junior college course and for upgrading to a bachelor's degree. Studying full-time means leaving work and studying all day; studying while working means not leaving work. For the open university, look at the Opinion on Running Open Universities Well《教育部关于办好开放大学的意见》. It states: 「开放大学实行注册入学，学生修完课程并获得相应学分，即可获得课程证书，修满规定的学分并达到相关要求，即可获得相应的学历与非学历证书，符合学位授予条件的应授予相应学位」, that is, open universities enrol by registration; once a student has completed the courses and earned the corresponding credits they receive a certificate for the courses, and once they have earned the prescribed credits and met the relevant requirements they receive the corresponding certificate of qualification and non-qualification certificate, and where they meet the conditions for a degree the corresponding degree is awarded (nationwide) +- Evidence grade: A +- Sources: State Council (1988, revised 2014). Interim Regulations on Higher Education Self-Study Examinations《高等教育自学考试暂行条例》（articles three, twenty, twenty-five and thirty-two）. （National Database of Laws and Regulations）；Standing Committee of the National People's Congress (2018). Law on Higher Education《高等教育法》（article twenty-one）. （National Database of Laws and Regulations）；General Office of the Ministry of Education (2025). Notice on Doing Well the 2025 National Enrolment Work for Adult Higher Education Institutions《关于做好 2025 年全国成人高校招生工作的通知》（教学厅〔2025〕6 号）and its annex, the 2025 Measures for the Enrolment of Adult Higher Education Institutions Nationwide《2025 年全国成人高校招生办法》. ；Ministry of Education (2016). Opinion on Running Open Universities Well《关于办好开放大学的意见》（教职成〔2016〕2 号）. ；Ministry of Education (2007). Notice on Further Improving the Administration of the Various Kinds of Enrolment in Higher Education Institutions《关于进一步做好高等学校各类招生管理工作的通知》（教发〔2007〕13 号）: 「普通专升本招生计划按不超过当年应届普通高职（专科）毕业生5%的比例安排」, that is, the plan for upgrading ordinary junior college students to a bachelor's degree is arranged at a proportion of not more than 5% of that year's graduates of ordinary higher vocational (junior college) courses. ；Government of China (2020). Master's enrolment expanded by 189,000 and the upgrading rate by 322,000 this year—employment support for graduates, vice minister of education Weng Tiehui: 「扩大普通高校专升本的规模，同比争取增加32.2万人」. ；Fujian Provincial Education Examination Authority (2026). Quoting 「教育部高校学生司《关于做好2026年普通高等学校专升本考试招生工作的通知》（教学司函〔2025〕53号）」when forwarding it. +- Notes: There is also the ordinary route to a bachelor's degree for those who have finished a full-time junior college. The document published at the national level is the Ministry of Education notice of 2007, which says the upgrading plan is not more than 5% of that year's junior college graduates. From 2020 the Ministry of Education expanded the scale of the upgrading route, adding 322,000 places year on year, and the 5% line is in practice no longer applied to it. The specific arrangements each year come from a letter issued by the higher education student department of the Ministry of Education; for 2026 it is 教学司函〔2025〕53 号. That letter is not published, and the document number can only be seen in the documents forwarded by each province. Who is eligible to apply is set by each province, and it basically takes only that year's full-time junior college graduates of the province itself, so the body of this item does not state its conditions. From the autumn of 2025, continuing higher education for people who already hold a qualification is no longer called "correspondence study" or "part-time study"; it is uniformly called "studying while working". Continuing higher education is the route by which adults later complete a qualification, and any publicity still using the old names should be treated carefully. Completing a qualification and "which skills are worth learning" are two different things; the latter is in section 23. Qualifications bought for quick issue and counterfeit certificates are in item 8 in section 23. +### 11. Not being in an establishment means flexible employment: you enrol for pension and medical insurance yourself where you work, and the household registration restriction has already been lifted + +- Cost: Pension and medical insurance for you to pay in full, calculated on the local contribution base, thousands of yuan a month. The contribution base is the reference figure used to work out the premium. This is the fixed cost that flexible employment most easily leaves out, and the hard part is paying it on time every month. +- In plain language: Not being in an establishment does not mean social insurance does not exist; it only means nobody pays your half. The documents require each place to lift the household registration restriction on flexible workers enrolling in the place where they work, with the very largest cities allowed to lift it step by step. So do not read "freedom" as meaning you need not think about social insurance. If you stop paying, your pension and your medical insurance reimbursement are both worked out under the rules for a break in contributions. +- Benefit: For enrolment, look at the Guiding Opinion on Safeguarding the Labour Rights and Interests of Workers in New Forms of Employment《关于维护新就业形态劳动者劳动保障权益的指导意见》. It states: 「各地要放开灵活就业人员在就业地参加基本养老、基本医疗保险的户籍限制，个别超大型城市难以一步实现的，要结合本地实际，积极创造条件逐步放开。组织未参加职工基本养老、职工基本医疗保险的灵活就业人员，按规定参加城乡居民基本养老、城乡居民基本医疗保险，做到应保尽保。督促企业依法参加社会保险。企业要引导和支持不完全符合确立劳动关系情形的新就业形态劳动者根据自身情况参加相应的社会保险」, that is, each place is to lift the household registration restriction on flexible workers taking part in basic pension insurance and basic medical insurance in the place where they work; where the very largest cities find it hard to do this in one step, they are to work with local conditions and actively create the conditions to lift it progressively. Flexible workers who have not joined the basic pension insurance for employees or the basic medical insurance for employees are to be organised, as required, into the basic pension insurance for urban and rural residents and the basic medical insurance for urban and rural residents, so that everyone who should be covered is covered. Enterprises are to be urged to take part in social insurance in accordance with the law. Enterprises are to guide and support workers in new forms of employment who do not fully meet the conditions for establishing a labour relationship into taking out the corresponding social insurance according to their own circumstances. The same document divides the employment relationship into three kinds: 「符合确立劳动关系情形的，企业应当依法与劳动者订立劳动合同。不完全符合确立劳动关系情形但企业对劳动者进行劳动管理……指导企业与劳动者订立书面协议……个人依托平台自主开展经营活动、从事自由职业等，按照民事法律调整双方的权利义务」, that is, where the conditions for establishing a labour relationship are met, the enterprise shall conclude an employment contract with the worker in accordance with the law; where they are not fully met but the enterprise does exercise labour management over the worker, it is to guide the enterprise and the worker to conclude a written agreement; and where the individual runs a business on their own account through a platform, works as a freelancer and so on, the rights and obligations of the two sides are governed by civil law (nationwide, from 2021) +- Evidence grade: A +- Sources: Ministry of Human Resources and Social Security and eight other departments (2021). Guiding Opinion on Safeguarding the Labour Rights and Interests of Workers in New Forms of Employment《关于维护新就业形态劳动者劳动保障权益的指导意见》（人社部发〔2021〕56 号，articles two and eight）. ；General Office of the State Council (2020). Opinion on Supporting Flexible Employment through Multiple Channels《关于支持多渠道灵活就业的意见》（国办发〔2020〕27 号）. +- Notes: The original says the household registration restriction is to be "lifted"（放开）, and it leaves an exception for the very largest cities; it does not say the restriction has been abolished nationwide. How far it has actually been lifted depends on the city where you enrol; ask there. The benefit is graded large because what a break in contributions loses is the pension and the medical treatment themselves, which is worth far more than ten thousand yuan. But the cost is real too, thousands of yuan a month paid in full by yourself. For making up a break in social insurance contributions and how the years are added together, see item 18 in section 7 (a break in social insurance contributions). And separately, selling agricultural produce or daily goods at a place and time designated by the government, or using your own skill to do convenience work that lawfully requires no licence, does not need a business licence. That is written in 国办发〔2020〕27 号. + +### 12. Delivering food, driving a ride-hailing car, carrying freight in the same city: the platform pays the work-injury cover premium job by job and you pay nothing + +- Cost: No money; the platform pays this. What you have to do is check whether the platform you take jobs from is inside the pilot. There are two ways: look for "work injury cover" on the protection or benefits page in the app, or telephone the human resources and social security bureau of your province and ask whether that platform has reported in. +- In plain language: For people who deliver food, drive a ride-hailing car or carry freight in the same city, the platform has to pay the work-injury cover premium for you job by job and you pay not one cent; every job you take is covered. This is still a pilot, so before you take jobs check whether your province and your platform are in scope. Where you already have a labour relationship with the platform, work injury insurance applies instead, and this scheme does not. +- Benefit: This cover is set out in the Measures for Work Injury Protection for People in New Forms of Employment (for Trial Implementation)《新就业形态人员职业伤害保障办法（试行）》. Article two states: 「平台企业应当依照本办法规定参加新就业形态人员职业伤害保障……为通过平台注册并接单，以平台企业名义提供出行、即时配送和同城货运等劳动并获得报酬或者收入的新就业形态人员缴纳职业伤害保障费，实现每单必保、每人必保。适用《工伤保险条例》规定的劳动者，企业应当依法为其参加工伤保险，不适用本办法」, that is, platform enterprises shall, in accordance with these Measures, take part in work injury protection for people in new forms of employment, and pay the work injury protection premium for people in new forms of employment who register through the platform and take jobs, provide labour such as travel, on-demand delivery and same-city freight in the name of the platform enterprise, and receive remuneration or income from it, so that every job is covered and every person is covered. For workers covered by the Regulations on Work-Related Injury Insurance《工伤保险条例》, the enterprise shall enrol them in work injury insurance in accordance with the law, and these Measures do not apply to them. Article nine: 「试行期间，出行行业按照每单 0.01 元执行；即时配送行业按照每单 0.07 元、0.25 元执行；同城货运行业按照每单 0.18 元执行」, that is, during the trial period the travel sector pays 0.01 yuan per job, the on-demand delivery sector pays 0.07 yuan and 0.25 yuan per job, and the same-city freight sector pays 0.18 yuan per job. How the pilot expands is set out in this passage: 「2025 年 7 月 1 日起，在已先行试点 7 个省份和 7 个平台企业的基础上……增加天津、河北、辽宁、浙江、安徽、福建、山东、湖北、广西、宁夏等 10 个省份……2026 年，推动职业伤害保障试点在全国 31 个省份和新疆生产建设兵团实施；将出行、即时配送和同城货运三个行业的平台企业总体纳入职业伤害保障试点范围」, that is, from 2025-7-1, on the basis of the 7 provinces and 7 platform enterprises where the pilot has already gone ahead, 10 more provinces are added, among them Tianjin, Hebei, Liaoning, Zhejiang, Anhui, Fujian, Shandong, Hubei, Guangxi and Ningxia; in 2026 the pilot for work injury protection is to be pushed forward in all 31 provinces of the country and in the Xinjiang Production and Construction Corps, and platform enterprises in the three sectors of travel, on-demand delivery and same-city freight are to be brought into the scope of the work injury protection pilot as a whole (pilot, expanding year by year) +- Evidence grade: A +- Sources: Ministry of Human Resources and Social Security and nine other departments (2025). Notice on Expanding the Pilot of Work Injury Protection for People in New Forms of Employment《关于扩大新就业形态人员职业伤害保障试点的通知》（人社部发〔2025〕24 号）and its annex, the Measures for Work Injury Protection for People in New Forms of Employment (for Trial Implementation)《新就业形态人员职业伤害保障办法（试行）》articles two and nine. +- Notes: This is still a pilot, not a formal system that exists everywhere. Before taking jobs, confirm whether your province and your platform are in scope; where they are not, all that is left is buying commercial insurance yourself. A second document, the Guiding Opinion on Safeguarding the Labour Rights and Interests of Workers in New Forms of Employment《关于维护新就业形态劳动者劳动保障权益的指导意见》, sets out several more requirements. The remuneration a platform pays to people working normally may not be below the local minimum wage, must be paid in full and on time, and may not be withheld or delayed without reason. Reasonable remuneration is also to be paid on statutory public holidays. The document number of that one is 人社部发〔2021〕56 号, and the flexible employment item quotes it too, see item 11 in this section. If something goes wrong, follow this procedure first and do not settle privately with the local station first. For medical insurance and pension you still have to enrol yourself; for flexible employment enrolment, see item 11 in this section. + +### 13. Two routes that lock in an establishment on the day you fill in your choices for the national entrance examination: publicly funded teacher training students and contract medical students. The price is 6 years of service + +- Cost: Tuition and accommodation are not yours to pay, and you get a living allowance as well. The cost is that for 6 years after graduating you cannot choose your own employment, and breaking the agreement means paying the money back, paying a penalty and getting a mark on your credit record. The hard part is that in those 6 years you cannot change where you are. +- In plain language: On both of these routes the state pays first, and after you graduate you work in the place named in the agreement for 6 years. Publicly funded teacher training students are recruited only at six normal universities directly under the Ministry of Education, in the advance batch. Contract medical students have to sign a tripartite agreement before they receive the admission notice. Breaking the agreement means paying the money back, paying a penalty, and getting a mark on your credit record. Work out whether you can accept those 6 years before you fill in your choices. +- Benefit: For publicly funded teacher training students, see the Implementation Measures for Publicly Funded Education of Students on the Bridging Programme for Undergraduates and Postgraduates at Normal Universities Directly under the Ministry of Education《教育部直属师范大学本研衔接师范生公费教育实施办法》(2024; the 2018 measures were repealed at the same time). Article two: the central finance bears the tuition and accommodation while the student is on campus and gives a living allowance. Article three: training is focused on cities in the central and western regions other than provincial capitals, at prefecture, autonomous prefecture and league level and below. It also states 「不得定向到直辖市、计划单列市或省会城市主城区任教（五个自治区、陆地边境省份、海南省、贵州省、青海省除外）」, that is, the placement may not be in a municipality directly under the central government, a city separately designated in the state plan, or the main urban area of a provincial capital, except for the five autonomous regions, the land border provinces, Hainan Province, Guizhou Province and Qinghai Province. Article nine: after graduating the student generally returns to primary or secondary schools in the prefecture, autonomous prefecture or league of their home province to teach, on 「承诺从事中小学教育工作 6 年以上」, that is, an undertaking to do primary or secondary school teaching work for 6 years or more. For those who go to a school in a town or city, 「应到农村义务教育学校任教服务至少 1 年」, that is, they are to teach in a rural compulsory education school for at least 1 year. Article thirteen: where the undertaking is not performed, or the period is not long enough, responsibility is taken under the agreement. One of the consequences is that 「违约行为纳入信用记录，并及时与全国信用信息共享平台共享，违约相关材料归入本人人事档案」, that is, the breach is put on the credit record, shared in good time with the National Credit Information Sharing Platform, and the material relating to the breach is filed in the person's own personnel file. A further line is that 「履约任务完成前不得到定向地（市、州、盟）之外从教，不得被招录（聘用）为其他公务人员。完成履约任务后即修复相关信用」, that is, before the service undertaking is complete the person may not teach outside the designated prefecture, autonomous prefecture or league, and may not be recruited or appointed as any other public officer; once the service undertaking is complete the relevant credit is restored. Until the period of service agreed in the undertaking has been served in full you cannot apply for other public posts; once it is served, the credit record is restored. Article six: a student found after entry, on examination, to be unsuitable for teaching is to return the publicly funded education costs already received within 1 year of enrolment and be transferred to a non-teaching programme. Article fourteen: a student who does not obtain the degree certificate and the teacher qualification certificate for a Master of Education within the prescribed period of study is treated as in breach, and returns the fees and pays the penalty. Article twenty-one: each locality is to resolve the staffing quota needed for publicly funded teacher training students to teach in primary and secondary schools, within the existing total quota of public institution staffing. For rural order-directed free medical students, see the other notice. After admission and before receiving the admission notice, the student must sign an agreement with the training school and with the county-level health and human resources and social security departments of the place where the employment is directed. The agreement has to include 「承诺毕业后到定向农村基层医疗卫生机构服务 6 年」, an undertaking to serve for 6 years after graduating at the designated rural grassroots medical and health institution. Performing the undertaking means working in the designated place for the full period as agreed. After reporting for work on discharge, they must take part in 3 years of standardised residency training in a specialty, abbreviated as residency training. Residency training is the stage after graduation when you go on training in a hospital to the same standard. During the training the original personnel and wage relationships are unchanged. The record of the breach is filed in the person's own personnel file and is used as the basis for assessing integrity when medical and health institutions and other administrative public institutions recruit openly, when they enrol for standardised residency training for resident doctors, and when they admit postgraduates. People who breach in bad faith are published nationwide on the "Credit China" website「信用中国」and are subject to joint punishment for dishonesty. Years of training lost through a breach do not count towards the period of service. Where the fees and the penalty have not been returned in accordance with the agreement, the county-level health administration department recovers them through legal proceedings (nationwide) +- Evidence grade: A +- Sources: General Office of the State Council (2024). Notice on Forwarding the Implementation Measures for Publicly Funded Education of Students on the Bridging Programme for Undergraduates and Postgraduates at Normal Universities Directly under the Ministry of Education from the Ministry of Education and Other Departments《教育部直属师范大学本研衔接师范生公费教育实施办法》（国办发〔2024〕27 号，articles two, three, six, nine, thirteen, fourteen and twenty-one）. ；National Health Commission, Organization Office of the CPC Central Committee, National Development and Reform Commission, Ministry of Education, Ministry of Finance, Ministry of Human Resources and Social Security and National Administration of Traditional Chinese Medicine (2019). Notice on Employment Placement and Performance Management of Rural Order-Directed Free-Trained Medical Students《关于做好农村订单定向免费培养医学生就业安置和履约管理工作的通知》. +- Notes: The 2024 measures turned publicly funded teacher training students into a "bridging programme", meaning an undergraduate degree followed straight on to a Master of Education. The old 2018 measures were repealed, and the "go straight to teaching after the undergraduate degree" still circulating online is the old version. Where an agreement was signed before the measures were published, the original agreement stands. "Having a staffing quota" does not mean "you can stop thinking once you are in": the quota has to be found within the existing total, the specific posts are filled by a special recruitment organised by the prefecture, and you and the school choose each other. Contract medical students may leave once their service period is up; the document states that 「服务期满，对自主择业的应予同意」, that is, where the service period is complete, those who choose their own employment are to be given agreement. There is also a way for people who have already breached to get the record removed; the price is serving the original period again, or serving at least 6 years at another rural grassroots institution in the province, and the fees and the penalty already paid are not refunded. For medical students you also have to count the 3 years of residency training, so the whole cycle is longer than for the teaching route. Each province also has its own publicly funded teacher training and contract medical student plans, with the conditions set by the province, so read that year's enrolment document for your own province before you fill in your choices. +### 14. If you want to work abroad, first check whether the company has the qualification for overseas labour cooperation: it may not take a deposit from you + +- Cost: No money. Before you leave, spend a quarter of an hour checking the company's qualification, then read both contracts to the end. +- In plain language: As far as the law is concerned, working abroad is not regulated for you but for the company that organises you to go out. Only an enterprise that has obtained the certificate of qualification for overseas labour cooperation and completed registration may do this. When you hear "get a tourist visa over there first and switch to a work visa later", that sentence in itself shows that the other side is breaking the law. The company also may not take a deposit from you on any pretext, or ask you to put up property as security. +- Benefit: Regulations on the Administration of Overseas Labour Cooperation《对外劳务合作管理条例》, article five: engaging in overseas labour cooperation requires the approval of the commerce department of the province, or of a city with subdistricts, and the acquisition of a qualification to operate. That qualification is the certificate of qualification for overseas labour cooperation; without it nobody may organise people to go abroad to work. Article seven: the list of certificates is reported to the commerce department of the State Council and circulated to Chinese embassies and consulates abroad. This article also states 「未依法取得对外劳务合作经营资格证书并办理登记，不得从事对外劳务合作」, that is, no one may engage in overseas labour cooperation without having obtained the certificate of qualification in accordance with the law and completed registration. Article eight: 「任何单位和个人不得以商务、旅游、留学等名义组织劳务人员赴国外工作」, that is, no unit or individual may organise workers to go abroad to work in the name of business, tourism, study abroad or anything of the kind. Article nine: the enterprise shall, within 5 working days of the date of registration, deposit at a designated bank 「不低于 300 万元人民币的对外劳务合作风险处置备用金」, that is, a reserve fund for handling the risks of overseas labour cooperation of not less than 3 million yuan, and the list of those who have deposited it is published to society. Article ten: the reserve fund is used to pay the service fees that should be refunded, the remuneration and the losses which the enterprise is liable for by law and refuses to bear or cannot bear. It is also used for 「因发生突发事件，劳务人员回国或者接受紧急救助所需费用」, that is, the cost of workers returning home or receiving emergency assistance when an emergency occurs. Article eleven: workers may not be organised to go abroad to work in anything connected with gambling or with sexual activity. Article twelve: nobody may be organised to go abroad without training being arranged. Article thirteen: accident and injury insurance covering the person during the period of working abroad must be bought for the workers. Articles twenty-one and twenty-three: written contracts must be concluded separately with the foreign employer and with the workers, and nobody may be organised to go abroad without them. Article twenty-five: 「不得以任何名目向劳务人员收取押金或者要求劳务人员提供财产担保」, that is, no deposit may be collected from workers on any pretext, and workers may not be required to provide property as security. Article twenty-six: filing with the commerce department in charge has to be done within 10 working days of the date the contract is concluded. Article twenty-nine: where a worker's actual rights do not match the contract, the enterprise shall assist with the claim; where the worker has not received the compensation due, the worker has the right to require the enterprise to bear the compensation liability. Article thirty-one: where personal safety cannot be secured because of the security situation of the country or region concerned, the enterprise may not organise workers to go there. Article thirty-seven: the commerce department in charge establishes a system of bad credit records and public notices. The penalties come later. Article thirty-nine: where there is no qualification to engage in overseas labour cooperation, the business is investigated and shut down, and where a crime is constituted, criminal liability is pursued. Article forty: where a business organises people to go abroad, or lends out its qualification, or organises work connected with gambling or pornography, in the name of business, tourism or study abroad, the certificate of qualification is revoked. Article forty-three: where no contract is signed, or where people are organised to go abroad by concealing information or providing false information, the fine is not less than 100,000 yuan and not more than 200,000 yuan; a separate fine of not less than 20,000 yuan and not more than 50,000 yuan is imposed on the person principally responsible (nationwide) +- Evidence grade: A +- Sources: State Council (2012). Regulations on the Administration of Overseas Labour Cooperation《对外劳务合作管理条例》（国务院令第 620 号，articles five, seven, eight, nine, ten, eleven, twelve, thirteen, twenty-one, twenty-three, twenty-five, twenty-six, twenty-nine, thirty-one, thirty-seven, thirty-nine, forty and forty-three）. +- Notes: The compliant procedure is this: the company first signs a written labour cooperation contract with the foreign employer, and then signs a written service contract or labour contract with you. The contract has to state the content of the work, the period, the remuneration, the working hours, the leave, social insurance and the risks to personal safety. Within 10 working days of signing, the company has to file with the commerce department in charge. The company also has to arrange training for you in skill, in safety precautions, in a foreign language and in the local law and customs, and buy you accident and injury insurance for the period of working abroad. A counterparty that will not let you see the contract, that asks for a deposit, or that hurries you out first and promises to complete the formalities later, is the kind of company to avoid. The overseas high-pay recruitment scams and the telecom fraud compounds are in item 5 in section 21; the safety reminders to check before you leave, and 12308, are in section 21, and are not repeated here. The Regulations set no education or age threshold for the person going out to work; the threshold is on the company's qualification, and on whether the country of work will give permission. The company can only sign a contract with an enterprise or organisation approved in that country, and not with an individual abroad. The reserve fund is used by the department in charge when a company defaults, and you cannot withdraw it yourself. So if something really goes wrong, complain to the commerce department in charge first; article twenty gives you that right. Sending more than 100 people to the same country or region requires accompanying management staff to be arranged and filed with the embassy or consulate. Projects with few people have no such protection, so keep your own copies of the contract and your transfer records. + +### 15. Working remotely for a foreign company from home: under the tax law this money is domestic income, you file the personal income tax yourself, and the remittance goes through as "remuneration for employment" + +- Cost: No money. One settlement and payment a year. The window runs from 3-1 to 6-30 of the following year. Where the remittance exceeds the annual total, you also have to have an employment contract and proof of income ready. The hard part is remembering to file this tax yourself. +- In plain language: Whether this money is domestic or foreign depends on where you do the work, not on where the money is paid in from. A person in China working remotely for a company abroad earns domestic income and has to pay tax in China. The foreign company has no establishment in China, so nobody withholds the tax for you and you have to file it yourself. The money comes in under the heading "remuneration for employment"「职工报酬」. +- Benefit: Whether this money counts as domestic income is set by article three of the Implementation Regulations of the Individual Income Tax Law（国务院令第 707 号, in force from 2019-1-1）《个人所得税法实施条例》. It states: 「除国务院财政、税务主管部门另有规定外，下列所得，不论支付地点是否在中国境内，均为来源于中国境内的所得：（一）因任职、受雇、履约等在中国境内提供劳务取得的所得」, that is, unless otherwise provided by the finance and taxation departments of the State Council, the following income is income sourced within China regardless of whether the place of payment is within China: (one) income obtained from providing labour within China through taking a post, being employed or performing a contract. Article six of the same Regulations defines two of the categories of income. One is wages and salaries: 「工资、薪金所得，是指个人因任职或者受雇取得的工资、薪金、奖金、年终加薪、劳动分红、津贴、补贴以及与任职或者受雇有关的其他所得」, that is, income from wages, salaries, bonuses, year-end increases, profit shares, allowances, subsidies and other income connected with taking a post or being employed. The other is remuneration for personal services: 「劳务报酬所得，是指个人从事劳务取得的所得，包括从事设计、装潢、安装、制图、化验、测试、医疗、法律、会计、咨询、讲学、翻译、审稿、书画、雕刻、影视、录音、录像、演出、表演、广告、展览、技术服务、介绍服务、经纪服务、代办服务以及其他劳务取得的所得」, that is, income from carrying out personal services, including income from design, decoration, installation, drawing, chemical testing, testing, medical work, legal work, accounting, consultancy, lecturing, translation, reviewing, painting and calligraphy, carving, film and television, sound recording, video recording, performance, acting, advertising, exhibitions, technical services, referral services, agency services, agency work and other personal services. Article six of the Individual Income Tax Law《个人所得税法》sets out how the taxable income is worked out: 「居民个人的综合所得，以每一纳税年度的收入额减除费用六万元以及专项扣除、专项附加扣除和依法确定的其他扣除后的余额，为应纳税所得额」, that is, for the comprehensive income of a resident individual, the taxable income is the balance of the income for each tax year after deducting the basic deduction of sixty thousand yuan and the special deductions, the special additional deductions and the other deductions determined in accordance with law. The same article also states that 「劳务报酬所得、稿酬所得、特许权使用费所得以收入减除百分之二十的费用后的余额为收入额」, that is, for remuneration for personal services, royalties for written work and royalties for the use of rights, the income amount is the income less a deduction of 20%. The 20% deduction covers only these three; wages and salaries are not among them. So wages and salaries go into the income amount at full value, and the two categories of income come out with a different tax. Article nine: 「个人所得税以所得人为纳税人，以支付所得的单位或者个人为扣缴义务人」, that is, for individual income tax the taxpayer is the person receiving the income and the withholding agent is the unit or person paying it. Article ten: 「有下列情形之一的，纳税人应当依法办理纳税申报：……（二）取得应税所得没有扣缴义务人；……（四）取得境外所得」, that is, in any of the following cases the taxpayer shall file a tax return in accordance with the law: where taxable income is obtained with no withholding agent; where income from abroad is obtained. Article eleven: 「居民个人取得综合所得，按年计算个人所得税……需要办理汇算清缴的，应当在取得所得的次年三月一日至六月三十日内办理汇算清缴」, that is, where a resident individual obtains comprehensive income, individual income tax is worked out by the year, and where a settlement and payment is required it is to be carried out within the period from the first day of the third month to the thirtieth day of the sixth month of the year after the income was obtained. For bringing the money in, look at the Measures for the Administration of Individual Foreign Exchange（中国人民银行令〔2006〕第 3 号, in force from 2007-2-1）《个人外汇管理办法》. Article nine: 「对个人结汇和境内个人购汇实行年度总额管理。年度总额内的，凭本人有效身份证件在银行办理；超过年度总额的，经常项目项下凭本人有效身份证件和有交易额的相关证明等材料在银行办理」, that is, settlement of foreign exchange by individuals and purchases of foreign exchange by individuals on the domestic side are managed by an annual total. Within the annual total the individual handles it at the bank with a valid identity document; over the annual total, for current account items the individual handles it at the bank with a valid identity document and materials such as proof relating to the transaction amount. Article seven: 「银行和个人在办理个人外汇业务时，应当遵守本办法的相关规定，不得以分拆等方式逃避限额监管，也不得使用虚假商业单据或者凭证逃避真实性管理」, that is, banks and individuals, when handling individual foreign exchange business, shall comply with the relevant provisions of these Measures, may not evade the limits by splitting transactions or the like, and may not use false commercial documents or vouchers to evade the authenticity controls. Article two of the Detailed Rules for Implementation of the Measures for the Administration of Individual Foreign Exchange（汇发〔2007〕1 号）: 「对个人结汇和境内个人购汇实行年度总额管理。年度总额分别为每人每年等值 5 万美元」, that is, settlement by individuals and purchases of foreign exchange by individuals on the domestic side are managed by an annual total, and the annual total is the equivalent of 50,000 US dollars per person per year. Article ten: 「境内个人经常项目项下非经营性结汇超过年度总额的，凭本人有效身份证件及以下证明材料在银行办理：……（七）职工报酬：雇佣合同及收入证明」, that is, where a domestic individual exceeds the annual total in non-operational settlement of foreign exchange for current account items, it is handled at the bank with a valid identity document and the following supporting documents: (seven) remuneration for employment: the employment contract and proof of income (nationwide) +- Evidence grade: A +- Sources: State Council (2018). Implementation Regulations of the Individual Income Tax Law of the People's Republic of China（中华人民共和国个人所得税法实施条例）（国务院令第 707 号，articles three and six）. ；Standing Committee of the National People's Congress (2018). Individual Income Tax Law of the People's Republic of China（中华人民共和国个人所得税法）(seventh revision of 2018, articles six, nine, ten and eleven). ；Policies and Regulations Database of the State Taxation Administration. Individual Income Tax Law of the People's Republic of China（中华人民共和国个人所得税法）（article six）. ；People's Bank of China (2006). Measures for the Administration of Individual Foreign Exchange（个人外汇管理办法）（中国人民银行令〔2006〕第 3 号，articles seven and nine）. ；State Administration of Foreign Exchange (2007). Detailed Rules for Implementation of the Measures for the Administration of Individual Foreign Exchange（个人外汇管理办法实施细则）（汇发〔2007〕1 号，articles two and ten）. +- Notes: Which category the tax is worked out under depends on what your relationship with the other side is. If you sign an employment contract and the work is arranged by them, it is wages and salaries. If you take jobs project by project and hand over results, it is remuneration for personal services. The two come out with a different tax, so work out which one you are before you file. When the money comes in, up to the equivalent of 50,000 US dollars a year can be settled at the bank on your identity card alone; for the part over that, give the employment contract and the proof of income to the bank. Do not look for someone to split the receipts with so as to get round the quota. What is described here is the case where you are in China and do not leave the country. Actually going abroad to work for a foreign employer is a different matter, see item 14 in this section (if you want to work abroad, first check whether the company has the qualification for overseas labour cooperation). Four further questions this book has not yet written about. Each of them may have a real cost, but no basis can at present be checked word for word. First, whether this income is liable to value-added tax and whether an invoice has to be issued on your behalf. Second, whether doing this for a long time requires a business licence or registration as an individual business. Third, whether you can claim rights against the foreign employer under the labour law, for example annual leave, overtime pay and severance. Fourth, the export of data and confidentiality. +### 16. Not being in an establishment and going into business yourself: look at the guaranteed entrepreneurship loan for the starting money, up to 300,000 yuan for an individual, with the state paying half the interest + +- Cost: No money to ask about it. Go once to the human resources and social security department for the eligibility check, and then go once each to the guarantee institution and to the bank that handles the loan (the bank that makes the loan). The hard part is that borrowed money still has to be repaid. +- In plain language: If you do have to borrow in order to start, look at this one first, because the state pays part of the interest. An individual can borrow up to 300,000 yuan, for up to 3 years, and the state pays half the interest. There is one hard condition that is often overlooked: when you apply, neither you nor your spouse may have any other loan, except that student loans, small loans for people who have escaped poverty, housing loans, car loans and small consumer loans below 50,000 yuan do not count. +- Benefit: The conditions for an individual applicant are in article six of the Measures for the Administration of Special Funds for the Development of Inclusive Finance（财金〔2023〕75 号）《普惠金融发展专项资金管理办法》. The first condition is that you belong to a key group for employment. The document lists the registered unemployed in towns and cities, people with employment difficulties (including people with disabilities), veterans, and people released from prison. It also lists university graduates (including former village college students and students returning from study abroad), and workers and unemployed people in enterprises cutting excess capacity. And returning migrant workers starting a business, online shopkeepers, people who have escaped poverty, and farmers starting their own businesses in the countryside. The second condition is that there may be no other loan in the name: 「除助学贷款、脱贫人口小额信贷、住房贷款、购车贷款、5 万元以下小额消费贷款（含信用卡消费）以外，申请人提交创业担保贷款申请时，本人及其配偶无其他贷款」, that is, apart from student loans, small loans for people who have escaped poverty, housing loans, car loans and small consumer loans below 50,000 yuan (including credit card spending), when an applicant submits the application for an entrepreneurship guaranteed loan, neither the applicant nor their spouse may have any other loan. Article nine: 「对符合条件的创业担保贷款，财政部门给予贷款实际利率 50% 的财政贴息。对展期、逾期的创业担保贷款，财政部门原则上不予贴息」, that is, for a qualifying entrepreneurship guaranteed loan, the finance department gives a fiscal interest subsidy of 50% of the actual interest rate of the loan. For an entrepreneurship guaranteed loan whose term is extended or that is overdue, the finance department in principle gives no interest subsidy. An interest subsidy means the state pays half of the interest. Extending the term means pushing the repayment date further out; being overdue means the due date has passed and nothing has been paid. In neither case does the state pay any more. Article ten: in principle the requirement for a counter-guarantee is removed for an entrepreneurship guaranteed loan of 100,000 yuan or less applied for by an individual, so there is no need to find somebody else or to put up property as security for the guarantee institution. Article eleven: individual entrepreneurship guaranteed loans inside the scope of grant and subsidy support are 「额度不超过 30 万元，贷款期限不超过 3 年」, that is, the amount does not exceed 300,000 yuan and the term of the loan does not exceed 3 years. Where people go into business together, the figure may be raised appropriately, up to 110% of the sum of the individual ceilings. The interest rate ceiling for grant and subsidy support comes in three bands: LPR+50BPs in the eastern region, LPR+150BPs in the central and western regions, and LPR+250BPs in areas that have escaped poverty. LPR is the loan prime rate quoted by banks, and BPs are the unit in which an interest rate is measured; the more is added, the higher the interest. Those who 「还款积极、带动就业能力强、创业项目好」, that is, repay actively, have a strong capacity to bring employment and have a good business project, may apply again after the term ends, up to 3 times in total. Article eight: the applicant first applies to the local human resources and social security department for the eligibility check, and then submits the application to the guarantee institution and to the bank handling the loan. Article twelve: with the agreement of the provincial people's government, localities may widen the scope of support, raise the ceiling and raise the proportion of the interest subsidy (nationwide) +- Evidence grade: A +- Sources: Ministry of Finance (2023). Notice on Issuing the Measures for the Administration of Special Funds for the Development of Inclusive Finance《普惠金融发展专项资金管理办法》（财金〔2023〕75 号，articles six, eight, nine, ten, eleven and twelve）. +- Notes: The state paying half the interest does not change the fact that "the money borrowed has to be repaid". The position taken in item 1 in section 12 is unchanged: for a business, only take money you can afford to lose, do not touch the family savings, and do not sign a personal guarantee for a company loan, and your spouse should certainly not sign it either. If you have decided to borrow, this kind of policy loan is the lowest rate you will get; do not touch consumer loans or private lending. The ceiling and the scope may be raised by each province; for how much will actually be approved, go by what the local human resources and social security department says that year. The 2020 notice from the pandemic period, with the "up to 200,000 yuan for an individual", was a temporary document; read these Measures now. Whether you open a shop by registering as an individual business or incorporate a company, the licence, the tax and getting out are in section 12. The training subsidy, the social insurance subsidy and the work placement during unemployment are in item 13 in section 7 (vocational training subsidies). \ No newline at end of file diff --git "a/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" "b/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" new file mode 100644 index 00000000..e9fa611c --- /dev/null +++ "b/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" @@ -0,0 +1,108 @@ +[← 回总目录](../README.md) + +# 32. 出国留学：身份、打工、保险和回国认证 + +本节算钱和人身自由两笔账，不讨论该不该出国，也不比较学校好坏。下面四件事不知道就要吃亏：签证身份怎么才算没断、打工时数的上限在哪、哪几笔保险和费用是必须交的、回国以后这张文凭能不能被认。 + +覆盖美国、加拿大、英国、澳大利亚四个国家，数字一国一国列。来源是各国移民局和中国教育部的官方网页。**各国的留学政策改得很勤，比中国的法规还勤。本节所有数字的截至日期是 2026 年 9 月。出发前和每次续签前，都照来源栏的链接自己再查一遍。** + +人在境外出了事怎么找使领馆、12308 能管什么、境外医疗和转运保险怎么买，见第 21 节。读书还是打工那道账，见第 23 节。账号被盗见第 14 节，电信诈骗见第 8 节。这些都不重复写。 + +[← Back to the contents](../README.md) + +# 32. Studying abroad: status, work, insurance and certification on return + +This section measures money and personal freedom. It does not discuss whether you should go abroad, and it does not compare schools. Four things below cost you if you do not know them: when your visa status counts as unbroken, where the ceiling on working hours is, which insurances and fees you must pay, and whether this diploma is recognised once you are back home. + +It covers four countries: the United States, Canada, the United Kingdom and Australia, with the figures given country by country. The sources are the official pages of the immigration departments of each country and of the Ministry of Education of China. **Study-abroad policies are changed frequently in each of these countries, more often than China's own rules are. Every figure in this section is as at 2026-09. Before you leave, and before every renewal, go through the links in the Sources field and check again yourself.** + +How to reach an embassy or consulate if something happens to you while you are abroad, what 12308 can do, and how to buy medical and evacuation insurance abroad, see section 21. The account of studying or working, see section 23. A stolen account, see section 14, and telecom fraud, see section 8. None of these is written out again here. + +### 1. Before you pay tuition, check whether the school is on the accredited institutions list of the Service Center for Scholarly Exchange + +- Cost: No money. Type the name of the school once into the Service Center's "Accredited Institutions Search"「认证院校查询」; it takes a few minutes. +- In plain language: Once you are back home, if you want to use this diploma to sit the civil service examination, get into a state-owned enterprise, settle a household registration or apply for a professional title, it has to go through a certification first. The certification is done by the Service Center for Scholarly Exchange of the Ministry of Education. If the school is not on the list that can be certified, the tuition and the time are wasted and the diploma is of no use at home. Checking the list costs nothing; check before you pay the tuition. +- Benefit: The Chinese Service Center for Scholarly Exchange has an accredited institutions search. You can search by country or by the name of the school. What you are checking is whether the qualifications and degrees this school awards fall inside the scope of certification. In addition, the Ministry of Education's Network for the Supervision and Administration of Education Involving Foreign Parties publishes regulatory information on Sino-foreign cooperative education and on running educational institutions abroad +- Evidence grade: A +- Sources: Chinese Service Center for Scholarly Exchange（中国（教育部）留学服务中心）. Accredited Institutions Search（认证院校查询）. ；Ministry of Education Network for the Supervision and Administration of Education Involving Foreign Parties（教育部教育涉外监管信息网）. +- Notes: What an agency tells you in words, that it is "recognised by the Ministry of Education", does not count; go by what you find yourself when you look it up online. The list changes: a school can be on it when you enrol and off it when you graduate, so check once a year while you are studying. Sino-foreign cooperative education programmes have a separate system of certification and registration. For those who enrolled after 2008, you can look up your own registration number on that network using your name and your identity card number. + +### 2. Watch the court case over the new F-1 rule in the United States: "at most four years, and leave within 30 days of finishing" was to take effect on 2026-09-15 and was postponed by the court the day before, so for now it is still "until you finish" + +- Cost: No money. Copy the programme end date on your I-20 into your calendar and set a reminder 90 days ahead. The I-20 is the certificate of eligibility for study that the school gives you. Also look at the notices from the school's international student office once a term to see whether there is any new development in the case +- In plain language: The United States was going to change F-1 to at most four years, with 30 days to leave after finishing. A federal court postponed it the day before it was to take effect. So for now the old rule still stands: you stay as long as you study, and you have 60 days after finishing to get ready to leave. A postponement is not a repeal; the government can still appeal, and the rule can change at any time. +- Benefit: The Department of Homeland Security published a rule on 2026-7-17 that was to take effect on 2026-9-15. It changes F-1 students from entering "until you finish" to a fixed period, at most four years, after which a separate extension has to be applied for. After finishing a programme or an internship there would be only 30 days in which to prepare to leave. On 2026-9-14, Judge Saylor of the United States District Court for the District of Massachusetts, under item 705 of the Administrative Procedure Act, deferred the rule from taking effect. The postponement applies nationwide and to the rule as a whole. The court did not vacate the rule, the request to vacate it was denied, but a further request may be made later. The case is still being heard. Under the order of the court, people on F, J and I visas still enter "until you finish", the 60-day grace period for departure still applies, and no separate extension is needed +- Evidence grade: A +- Sources: DHS (2026). Establishing a Fixed Time Period of Admission and an Extension of Stay Procedure for Nonimmigrant Academic Students, Exchange Visitors（published 2026-07-17, to take effect 2026-09-15）. ；Presidents' Alliance on Higher Education and Immigration v. U.S. Department of Homeland Security, No. 1:26-cv-13799-FDS (D. Mass. Sept. 14, 2026), court docket. ；Yale Office of International Students & Scholars (2026-09-14). Important Update: Court Action on the D/S Rule. ；AILA (2026). One Day Before Taking Effect, Federal Court Postpones the F, J, and I Fixed Admission Period Rule. ；8 CFR 214.2(f). +- Notes: This item is as at 2026-09-25. The government can appeal to the Court of Appeals for the First Circuit, and the Department of Homeland Security can also amend the rule and publish it again. So if you are studying for several years, do not assume the old rule will last. The court has set 2026-10-02 for a further hearing on how both sides have got on, and after that date look at the notice from your school once more. If the new rule does take effect later, an extension has to be applied for separately before the period ends; it does not carry on by itself. Under the old rule, the departure deadline is different if you stop studying before finishing, so ask the international student office first. The J visa is in the same rule and the details are not the same as for F-1, so look at your own visa category. + +### 3. The hours you may work are a hard limit written into the rules of all four countries, and one hour over is a breach of your visa conditions + +- Cost: No money. Keep a record of your own hours week by week. The hard part is that employers say "nobody checks a few extra hours", and that is not true +- In plain language: All four countries have put a ceiling on how long a student visa holder may work. Go over it and you have breached the conditions of your visa; your visa may be cancelled, you may be removed, and a later application will be asked about it. The ceilings are: the United States 20 hours a week; the United Kingdom 20 hours a week on a degree course and 10 hours a week below degree level; Canada 24 hours a week; Australia 48 hours a fortnight. If you hold two jobs the hours are added together. +- Benefit: For F-1 students in the United States, on-campus work must not exceed 20 hours a week during term time. Approved off-campus part-time work is also 20 hours a week. Full-time work is allowed during the holidays. The basis is 8 CFR 214.2(f)(9) of the United States federal immigration regulations. For the student visa in the United Kingdom, on a degree course or above it is 20 hours a week during term time, and below degree level 10 hours a week. Part-time courses may not be worked at all. You also may not be self-employed (working for yourself), and may not do professional sport or entertainment work. The basis is the part of the United Kingdom immigration rules that deals with students, appendix Student ST26.1 and ST26.5. In Canada a qualifying full-time student may work off campus up to 24 hours a week, under paragraph 186(v) of the Immigration and Refugee Protection Regulations. For the student visa in Australia it is at most 48 hours a fortnight while the course is running. Research masters and doctoral students and their dependants are not subject to this limit +- Evidence grade: A +- Sources: 8 CFR 214.2(f)(9). ；UK Home Office. Immigration Rules Appendix Student（ST26.1、ST26.5）. ；IRCC. Work off campus as an international student. ；Australian Government Department of Home Affairs. Student visa (subclass 500). +- Notes: Off-campus work in the United States has to be authorised first. While the I-765 application for the work permit is still being processed, you may not start work. Canada's older permit says 20 hours a week on it; as long as you still meet the conditions, 24 hours is in fact possible, and the current rules govern. The United Kingdom's prohibition on self-employment is drawn widely: taking private jobs, running an online shop and working as a freelancer all count. + +### 4. The root of the status is "studying full time": during a leave of absence, a withdrawal or the gap between schools you may not work + +- Cost: No money. Before you stop classes, take a leave of absence or change schools, go to the school's international student office and ask what your status is +- In plain language: Whether you may work depends on whether you are in fact studying full time. Once you take a leave, withdraw, are expelled, or are not attending classes during a period of transfer, the right to be paid to work ends at that moment. Carrying on then is working illegally. A lot of people think it is all right as long as the visa has not run out; in fact the status was broken long before that. +- Benefit: Canada states expressly that off-campus work is not permitted during an authorised leave of absence. Nor is it permitted while you are not studying during a period of transfer. You may only start working again after you have resumed your studies. In the United States, the right to work on campus is given only to F-1 students whose status in SEVIS is Active. SEVIS is the United States system of records for the status of international students, and Active means enrolled and valid. These students also have to keep up a full-time course. In the United Kingdom the work permission follows the type of course, and part-time courses may not be worked +- Evidence grade: A +- Sources: IRCC. Work off campus as an international student. ；U.S. Department of Homeland Security, Study in the States. Working in the United States. ；UK Home Office. Immigration Rules Appendix Student（ST26.1）. +- Notes: If the academic pressure is heavy and you are thinking of reducing your course load, ask first: how little still counts as full time. This one bites earlier than working does. If you are expelled by the school, or the programme is terminated, the status is normally broken the same day. Do not wait for an official letter from the school before you act. + +### 5. United States: you must report your new address within 10 days of moving + +- Cost: No money. Submit the change of address online once; it takes ten minutes +- In plain language: In the United States, when you move, the law requires you to notify the immigration authorities within 10 days. Not reporting is not a small matter. First, it is itself unlawful. Second, letters sent to you by the immigration authorities will go astray. +- Benefit: 8 CFR 265.1 of the United States federal immigration regulations provides that a foreign national who is required to register under the law shall report within 10 days of the address having changed. It is reported to USCIS (the United States immigration authorities) in the way USCIS requires, reporting both the fact that the address has changed and the new address +- Evidence grade: A +- Sources: 8 CFR 265.1. ；USCIS. AR-11, Alien's Change of Address Card. +- Notes: Even after you have told the immigration authorities, the SEVIS record at the school (your file of international student status) has to be updated separately. The other three countries also require you to notify the immigration authorities or the school when you move. The deadline and the method differ from country to country; follow the page of the immigration department of the country you are in. + +### 6. Before you leave and while you are studying, take a look at the Ministry of Education's study-abroad warnings + +- Cost: No money. Look once before you leave, and once a term after that +- In plain language: The Ministry of Education issues a warning about a particular country or region, saying that public security or policy there is unfavourable to Chinese students at present. This is an official public risk notice, in writing and open to checking, and it is more reliable than what an agency tells you. Taking a look costs nothing. +- Benefit: The Ministry of Education publishes study-abroad warnings through the Network for the Supervision and Administration of Education Involving Foreign Parties. In 2025 it issued four in all, as warning item 1, item 2, item 3 and item 4. Item 1 warned that higher education legislation passed by certain states of the United States contains provisions that are negative towards China (2025-04-09). Item 2 and item 3 warned on two separate occasions that cases against Chinese citizens are frequent in the Philippines and that the environment for studying there is poor (2025-07-18, 2025-08-30). Item 4 warned that the public security situation and the environment for studying in Japan are poor, and advised planning study trips to Japan carefully (2025-11-16) +- Evidence grade: A +- Sources: Ministry of Education Network for the Supervision and Administration of Education Involving Foreign Parties（教育部教育涉外监管信息网）. Warning information（预警信息）. ；Ministry of Education issues the fourth study-abroad warning of 2025（教育部发布 2025 年第 4 号留学预警）. ；Warning item one（第 1 号）. +- Notes: A warning is not a ban, and it does not mean you cannot go; it gives you an official basis for weighing the risk there. The list of warnings changes with the situation, and this section does not keep it up to date over time; go by the page current at the time on the website. Personal safety abroad and consular protection, see section 21. +### 7. Australia: OSHC has to cover the whole stay with no gap in the middle, and without proof of insurance you may be refused entry + +- Cost: Several thousand yuan up to more than ten thousand yuan. Buy it in one go for the length of the visa; if you are there several years, buy several years' cover +- In plain language: Australia has written the health insurance for international students into the visa conditions, so it is not something you may or may not buy. The insurance has to cover your whole stay. When you renew the visa and change policy there must be no gap in between. If you cannot produce proof of insurance on entry, you may be refused entry there and then. When family members come to accompany you, each of them has to buy a policy of their own. +- Benefit: The Australian Department of Home Affairs requires applicants for a student visa and their dependants to hold Overseas Student Health Cover (OSHC), and to hold it for the whole time they are in Australia, with no break. The insurance has to be provided by an insurer recognised in Australia. Cases that qualify for an exemption from having to buy it are excepted. There may be no gap between the new insurance and the insurance under the previous visa. If proof that you are insured cannot be produced on entry, you may be refused entry. Where you arrive earlier than the start of the course, the insurance start date has to be the day you actually arrive in Australia +- Evidence grade: A +- Sources: Australian Government Department of Home Affairs. Student visa (subclass 500). +- Notes: Where the school arranges it, ask for the name of the insurer, the start and end dates of the policy and the policy number; these go into the visa application. Buying too short a period causes trouble with the renewal, so it is better to buy a little past the date you graduate. + +### 8. United Kingdom: besides the visa fee there is a health surcharge charged by the year; a student pays 776 pounds a year, paid in full at the time of application + +- Cost: Visa fee 558 pounds. Health surcharge 776 pounds a year. Both are paid in full at the time of application +- In plain language: The money you pay to go to the United Kingdom is not only the visa fee. You also have to prepay a health surcharge according to the number of years on your visa, and only with it do you get the National Health Service. A two-year visa means paying 1552 pounds in one go. Add the visa fee and this whole amount has to be paid before you leave, so do not leave it out of the budget. +- Benefit: The cost of a student visa in the United Kingdom is 558 pounds. Applied for outside the United Kingdom, and for extensions or changes inside the country, the figure is the same. There is also the health surcharge, without which you cannot use the National Health Service, prepaid once and in full at the time of application according to the number of years on the visa. The student and their dependants pay 776 pounds a year, so a 2-year visa is 1552 pounds. Other categories of applicant pay 1035 pounds a year. Where the visa is longer than 6 months and shorter than 1 year, it is charged as a whole year +- Evidence grade: A +- Sources: UK Government. Student visa. ；UK Government. Pay for UK healthcare as part of your immigration application. +- Notes: Both the amounts and the exchange rate change, so work from the current figures on the two pages in the sources. A degree course, and anyone aged 18 or over, is normally allowed at most 5 years; below degree level it is 2 years. Use that number of years to work out how much surcharge to allow for. + +### 9. Leave time for the certification before you come back: certification by the Service Center takes 10 to 20 working days + +- Cost: Apply and pay online. Preparing the documents plus the assessment and the review, counted in working days, comes to around a month in all +- In plain language: To use a qualification gained abroad once you are back home, you first have to have it certified. The assessment alone takes 10 to 20 working days. If it falls in the job-hunting season, or your documents are returned, it takes longer. The autumn recruitment round, the registration for the civil service examination and settling a household registration all have deadlines, so do not wait until you need it to start the process. +- Benefit: Certification of qualifications obtained abroad and outside the territory is done in the online service hall of the Chinese Service Center for Scholarly Exchange. The process has four steps: register and complete identity verification, submit the application and the documents online, pay online, and wait for the assessment and the review. The published time limit for certification is 10 to 20 working days. Working days do not include weekends and public holidays, and this is the time the Service Center takes to review +- Evidence grade: A +- Sources: Chinese Service Center for Scholarly Exchange（中国（教育部）留学服务中心）. Online Service Hall · Academic Qualification Certification（网上服务大厅·学历学位认证）. +- Notes: The documents you have to hand in are these: the diploma, your passport or travel permit, your residence card or visa endorsement, a photograph of your identity document, and a statement of authorisation. Entry and exit records for the period you were abroad are pulled by the system itself, so do not throw away the visa pages you were carrying while you were abroad. The Service Center also issues announcements on its website saying which certificates have been invalidated and whose applications for certification have been suspended. Before you start, check whether your own school is one of them. + +### 10. More schools every year are put under "enhanced review" of certification, and the route of buying a diploma no longer works + +- Cost: No money. Before you sign up, search the name of the school once in the Service Center's announcements +- In plain language: The Service Center names a batch of foreign institutions and applies enhanced review to the certification of their qualifications on its own; it also suspends the acceptance of certification applications for individual schools. The ones it names are often exactly the kind advertised as "no language test, no going abroad, a master's degree in a few months". Searching the name of the school once before you sign up saves a great deal of trouble compared with appealing afterwards. +- Benefit: The Chinese Service Center for Scholarly Exchange has been issuing announcements applying enhanced certification review to the academic qualification certification of some foreign institutions (a separate extra check on the diplomas of these schools). The most recent is number nine (2025-10-28). The Service Center has also issued announcements suspending the acceptance of certification applications for individual overseas institutions, and reminders that people have used academic qualification certification from abroad and outside the territory to carry out fraud +- Evidence grade: A +- Sources: Chinese Service Center for Scholarly Exchange（中国（教育部）留学服务中心）. Announcement on Applying Enhanced Certification Review to the Academic Qualification Certification of Some Foreign Institutions (Nine)《关于对部分国外院校学历学位认证加强认证审查的公告（九）》(2025-10-28). ；Chinese Service Center for Scholarly Exchange（中国（教育部）留学服务中心）. +- Notes: "Enhanced review" does not mean the certification will definitely be refused, but it does mean you will be asked for more documents and that it will take longer. Any agent promising "certification guaranteed" or "expedited through inside channels" is a scam; the Service Center has issued specific warnings about fraud. Having somebody write your thesis or take your online courses for you across the border, if it is found out, not only voids the qualification but may also affect your visa record. \ No newline at end of file diff --git "a/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" "b/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" new file mode 100644 index 00000000..c22a121c --- /dev/null +++ "b/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" @@ -0,0 +1,103 @@ +[← 回总目录](../README.md) + +# 34. 家里的常备药别吃出事 + +退烧药、止痛药、感冒药、胃药、止泻药，很多在药店就能自己买。不用处方能买，不等于怎么吃都行。这一节不讲每种病该吃什么药，只挑一个动作就能避开重后果的几件事。 + +所有条目按死亡率口径算，包括肝衰竭、胃出血、新生儿肾衰竭这类健康终点。收益大多是「少出一次严重事故」，研究给不出「做了能少几成」的数字，收益量级是按后果轻重判断的，每条备注里写了凭什么定。 + +已经写在别处的，这里只指路。误服了药或清洁剂先做什么，见第 13 节第 20 条（误服）。慢性病的药按医嘱吃满，见第 16 节第 1 条（药按医嘱吃满）。网上买处方药要先过处方审核，见第 28 节第 6 条（减肥药就去医院拿处方）。 + +本节条目的受益人是你自己和家人，属于最高的两档。 + +[← Back to the contents](../README.md) + +# 34. Do not mishandle the medicines at home + +Fever reducers, painkillers, cold medicines, stomach medicines and anti-diarrhoeal medicines are mostly things you can buy yourself in a pharmacy. Being able to buy one without a prescription does not mean any way of taking it is safe. This section does not say which medicine to take for which illness. It picks only the items where a single action avoids a serious outcome. + +Every item here is measured by mortality, including health outcomes such as liver failure, stomach bleeding and kidney failure in newborns. Most benefits are "one serious accident avoided". Research cannot give a figure for "how much less does doing it make you", so the size of each benefit is judged by how serious the outcome is, and the Notes field of each item says why. + +What is written elsewhere is only pointed to here. What to do first after someone swallows a medicine or a cleaning product, see item 20 in section 13 (accidental swallowing). Taking a chronic illness medicine for as long as the doctor told you to, see item 1 in section 16 (take the medicine for as long as prescribed). Prescription medicines bought online have to pass prescription review first, see item 6 in section 28 (get the prescription at a hospital for weight-loss drugs). + +In this section the beneficiaries are you and your family, which is the top two grades. + +### 1. Before taking two cold medicines or two painkillers together, read the ingredient list: acetaminophen can only count once + +- Cost: No money. Read the ingredient list on each box. It takes a minute or two. +- In plain language: Many cold medicines contain acetaminophen. Taking two boxes at once means taking double, and too much damages the liver. In the United States, nearly three in ten of the acute liver failure patients it causes die before they get a new liver. The Chinese package insert advises at most 2 grams a day, and no alcohol while you take it. +- Benefit: 22 United States hospitals registered 662 cases of acute liver failure over 6 years in a row. 275 of them (42%) were caused by acetaminophen, and the median dose swallowed was 24 grams. Of these 275 people, 131 (48%) had taken too much by accident, not on purpose. Of the people who took too much by accident, 38% were also taking two or more medicines containing this ingredient. Of these 275 people, 27% died before they got a new liver and 8% had a liver transplant. In 2020 the National Medical Products Administration (国家药监局) required over-the-counter package inserts to state: overuse causes serious liver damage, and the maximum in one day should be no more than 2 grams. The insert must also state that you should avoid combining it with other medicines containing acetaminophen or other antipyretic analgesics. United States rules require the first liver warning on these over-the-counter products to name three situations: taking more than the daily maximum, taking them together with another medicine containing acetaminophen, and drinking more than 3 glasses of alcohol a day while you take them. +- Evidence grade: B +- Sources: Larson AM, Polson J, Fontana RJ, et al.; Acute Liver Failure Study Group (2005). Acetaminophen-induced acute liver failure: results of a United States multicenter, prospective study. Hepatology, 42(6):1364-1372. ；National Medical Products Administration (国家药监局) (2020). Announcement on Revising the Package Inserts for Immediate-Release and Extended-Release Acetaminophen Preparations (2020 No. 15), and Annex 2, Requirements for Revising the Package Insert of Immediate-Release Acetaminophen Over-the-Counter Medicines《对乙酰氨基酚常释制剂非处方药说明书修订要求》, reprinted by the Hunan Provincial Drug Administration (湖南省药品监督管理局). ；21 CFR 201.326(a)(1)(iii)(A) Liver warning. +- Notes: Acetaminophen is also called paracetamol. Compound cold medicines whose generic name carries the two characters for acetaminophen (氨酚), such as those sold as acetaminophen with pseudoephedrine, or as acetaminophen, caffeine and chlorpheniramine, contain this ingredient. If you are not sure, take every box to the pharmacist in the shop and ask. The grade is B because the United States figures come from case registration with no control group, so nobody can work out how many incidents reading the ingredient list would prevent. The benefit is graded "large" by the severity of the outcome: close to three in ten acute liver failure patients die. + +### 2. Do not give children aspirin, nimesulide or metamizole for a fever, and do not give minors any cold medicine that contains metamizole + +- Cost: No money. When you buy a fever reducer, look at the generic name and the ingredient list. +- In plain language: When a child has a fever, do not use any of these three drugs. Aspirin is linked to Reye's syndrome, which damages the brain and the liver; in United States figures about three in ten affected children die, and after the warning was issued cases fell by more than 90%. Nimesulide is banned below 12 years old, and metamizole below 18. +- Benefit: Reye's syndrome causes encephalopathy and fatty change in the liver, and it usually appears after influenza or chickenpox. In the United States national surveillance from 1981 to 1997, 1207 cases were reported in people under 18. Salicylate, an ingredient of drugs such as aspirin, was measurable in the blood of 82% of the affected children. The overall fatality rate was 31%. Warnings for salicylate medicines began to be issued in 1980. That year 555 cases were reported, which was the peak; from 1987 there were no more than 36 cases a year. In 2011 the former State Food and Drug Administration (国家食药监局) ruled that oral nimesulide preparations are prohibited for children under 12. In 2020 the National Medical Products Administration ruled that metamizole tablets are prohibited for children and adolescents under 18. Compound products that contain metamizole, namely qinghao metamizole tablets, Zhongganling tablets and Zhongganling capsules, are likewise prohibited below 18. +- Evidence grade: B +- Sources: Belay ED, Bresee JS, Holman RC, Khan AS, Shahriari A, Schonberger LB (1999). Reye's syndrome in the United States from 1981 through 1997. New England Journal of Medicine, 340(18):1377-1382. ；Notice on Strengthening the Administration of Oral Nimesulide Preparations (国食药监安〔2011〕209号). ；National Medical Products Administration (国家药监局) (2020). Announcement on Revising the Package Inserts of Metamizole-Related Products (2020 No. 34), with Annexes 1 to 3. +- Notes: Metamizole is not the first choice for adults either. The insert says it may cause agranulocytosis, aplastic anaemia and anaphylactic shock, and it is used for a short time only when the illness is severe and no other effective medicine is available. Zhongganling sounds like an ordinary cold medicine, and it contains metamizole. For which fever reducer a child should take, and how much by body weight, ask a paediatrician or a pharmacist. A child under 3 months old with a fever goes straight to hospital, see item 8 in section 20 (go straight to hospital at 38 ℃). The grade is B because the Reye's syndrome figures come from surveillance reports, and the fall in cases and the issue of the warning stand in sequence rather than in a controlled trial. The benefit is graded "large" because after the warning cases fell from 555 a year to fewer than 36. + +### 3. If you are over 60, have bled from the stomach, or take anticoagulants or steroids, ask a doctor before taking a painkiller such as ibuprofen + +- Cost: No money. It takes asking a doctor or a pharmacist once. +- In plain language: Painkillers such as ibuprofen, naproxen and aspirin harm the stomach. In trials at high doses, people taking ibuprofen had about 4 times the risk of stomach bleeding or perforation compared with people who did not take it. The risk is higher if you are older, have bled from the stomach, also take anticoagulants or steroids, or drink alcohol every day. +- Benefit: These drugs are called non-steroidal anti-inflammatory drugs, and the common ones are ibuprofen, naproxen and diclofenac. A meta-analysis pooled 280 randomised trials of these drugs against placebo, 124513 people in total. Compared with placebo, the risk of upper gastrointestinal perforation, obstruction or bleeding: ibuprofen RR 3.97 (95% CI 2.22–7.10, about 4 times), naproxen 4.22 (2.71–6.56), diclofenac 1.89 (1.16–3.09). All of these drugs roughly double the risk of heart failure. United States rules require these over-the-counter products to carry a stomach bleeding warning. The warning names six groups at higher risk: those over 60, those who have had a stomach ulcer or bleeding, those taking anticoagulants or steroids, those taking another drug of this kind at the same time, those drinking more than 3 glasses of alcohol a day while they take them, and those who take more than the insert says or for longer than it says. +- Evidence grade: A +- Sources: Coxib and traditional NSAID Trialists' (CNT) Collaboration; Bhala N, Emberson J, Merhi A, et al. (2013). Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. Lancet, 382(9894):769-779. ；21 CFR 201.326(a)(2)(iii)(A) Stomach bleeding warning. +- Notes: The paper's conclusion is about high-dose use. The paper gives no separate figure for occasional use over a day or two at the dose on the insert. Many compound cold medicines also contain this ingredient, so taking them together with a painkiller repeats it; read the ingredient list first. If a doctor has put you on low-dose aspirin for the long term to prevent blood clots, do not stop it yourself, and do not add a second painkiller yourself. + +### 4. Do not give a child under 2 compound cold medicines or cough medicines on your own + +- Cost: No money, and it also saves the money you would spend on medicines. +- In plain language: The National Medical Products Administration (国家药监局) requires a warning on 14 compound cold medicines for children: parents are advised not to use them on a child under 2, and to use them under the guidance of a doctor or a pharmacist. In trials carried out on children, cough medicines and anti-allergy medicines did no better than placebo. +- Benefit: In 2021 the National Medical Products Administration revised the package inserts of 14 products under Announcement No. 57, among them the paracetamol, chlorpheniramine and dextromethorphan oral solution, paracetamol and chlorpheniramine granules for children, and paracetamol and amantadine granules for children. The new warning reads「不建议家长或监护人自行给2岁以下婴幼儿使用本品，应在医师或药师的指导下使用」, that is, parents or guardians are advised not to give this product to infants and toddlers under 2 on their own, and to use it under the guidance of a doctor or a pharmacist. The precautions add「应严格按照药品说明书用法用量使用，避免用药过量」, that is, use it strictly at the dose and usage on the package insert and avoid overdose. They also add「应避免合并使用含有相同或相似活性成份的抗感冒药」, that is, avoid using it together with other cold medicines that contain the same or a similar active ingredient. A Cochrane systematic review included 29 randomised controlled trials, 10 of them carried out on children. In the children's trials none of these classes of drug did better than placebo: cough medicines, antihistamines (the anti-allergy ingredients), antihistamines plus decongestants (the ingredients that clear the nose), and cough medicines plus bronchodilators. 21 studies reported adverse events, and the preparations containing antihistamines and dextromethorphan produced more of them. +- Evidence grade: B +- Sources: National Medical Products Administration (国家药监局) (2021). Announcement on Revising the Package Inserts of 14 Products Including the Paracetamol, Chlorpheniramine and Dextromethorphan Oral Solution (2021 No. 57), with annexes. ；Smith SM, Schroeder K, Fahey T (2014). Over-the-counter (OTC) medications for acute cough in children and adults in community settings. Cochrane Database of Systematic Reviews, (11):CD001831. +- Notes: Many of these 14 products carry the two characters for acetaminophen (氨酚) in their names, and they contain acetaminophen, so taking them together with a fever reducer repeats it; see item 1 in this section (acetaminophen). One trial in that review found honey better than placebo, but honey must not be given below 1 year of age, see item 6 in section 20 (honey). The grade is B because the review holds few trials for each class of drug and the authors could not pool them into a figure. The benefit is graded "medium" as a judgement: no benefit has been shown for children, and the risks are overdose and repeated ingredients. + +### 5. After 20 weeks of pregnancy, do not take a painkiller such as ibuprofen on your own + +- Cost: No money. +- In plain language: After 20 weeks of pregnancy, most of the amniotic fluid is made by the fetus's kidneys. Taking ibuprofen or a drug like it then can harm the fetus's kidneys and reduce the amniotic fluid. The United States has received reports in which newborns died. The low-dose aspirin a doctor prescribes is not covered by this. +- Benefit: The US Food and Drug Administration (FDA) issued a safety warning in 2020-10. Using non-steroidal anti-inflammatory drugs at about 20 weeks of pregnancy or later can cause fetal kidney problems and lead to low amniotic fluid. The warning covers prescription and over-the-counter medicines alike, including aspirin, ibuprofen, naproxen, diclofenac and celecoxib. The FDA examined adverse event reports up to 2017-7 and found 35 cases of low amniotic fluid or newborn renal impairment, all of them severe outcomes. 5 were newborn deaths, all with newborn renal failure. In the literature, most cases of low amniotic fluid resolved between 72 hours and 6 days after the drug was stopped. The FDA advises pregnant women not to use these drugs after 20 weeks unless a doctor says so deliberately. Low-dose aspirin of 81 milligrams, prescribed by a doctor for specific problems in pregnancy, is not covered by this advice. When you need pain relief or a fever reducer, the FDA mentions that other drugs such as acetaminophen can be used; ask a pharmacist or a doctor first. +- Evidence grade: B +- Sources: U.S. Food and Drug Administration (2020-10-15). FDA recommends avoiding use of NSAIDs in pregnancy at 20 weeks or later because they can result in low amniotic fluid. Drug Safety Communication. +- Notes: The FDA warns that many over-the-counter medicines for colds, flu and insomnia also contain these ingredients, so read the ingredient list. United States over-the-counter labelling used to warn only about the last 3 months of pregnancy; this time the line moved back to about 20 weeks. Low-dose aspirin for the prevention of pre-eclampsia, see item 5 in section 27 (low-dose aspirin). The grade is B because the figures come from adverse event reports, and there is no way to know how many people took the drug without harm. The benefit is graded "medium" as a judgement: the outcome is serious, but it is rare. +### 6. Buy omeprazole yourself for acid reflux and heartburn, and take it for at most 7 days; if you have difficulty swallowing, vomit blood or pass black stool, go straight to hospital + +- Cost: No money. Take it for the number of days the insert gives. +- In plain language: Omeprazole enteric-coated tablets can be bought without a prescription from 2022, but the insert says not to take them yourself for more than 7 days, and not to take them again within two months. Do not take them yourself if you have difficulty or pain when swallowing, vomit blood, or pass blood or black stool; these can be signs of something serious. This medicine can also cover up the symptoms of stomach cancer and delay the diagnosis. +- Benefit: In 2022 the National Medical Products Administration (国家药监局) moved omeprazole enteric-coated tablets from prescription to over-the-counter under Announcement No. 68, and issued a package insert template at the same time. The indications in the template are「胃酸过多引起的烧心和反酸症状的短期缓解」, that is, short-term relief of heartburn and acid reflux caused by excess stomach acid. The precautions state: use must not exceed 7 days, and if the symptoms do not ease, ask a doctor or a pharmacist. Do not take it again within two months, and if the symptoms keep coming back, see a doctor at once. Do not use it if you have difficulty or pain when swallowing, vomit blood, or pass blood or black stool; these can be signs of something serious. If alarm symptoms such as marked weight loss, repeated vomiting, difficulty swallowing, vomiting blood or black stool appear, a malignant tumour has to be ruled out first, because treatment may cover up the symptoms and delay the diagnosis. If you are over 55 and new symptoms appear or the symptoms change, ask a doctor. People taking clopidogrel should avoid taking omeprazole at the same time. +- Evidence grade: B +- Sources: National Medical Products Administration (国家药监局) (2022). Announcement on the Conversion of Omeprazole Enteric-Coated Tablets from a Prescription Medicine to an Over-the-Counter Medicine (2022 No. 68), and Annex 2, Over-the-Counter Package Insert Template《非处方药说明书范本》. +- Notes: Clopidogrel is the anti-clotting drug that people who have had a stent put in or have had a heart attack often take. Omeprazole weakens it, so if you take it, ask your doctor first. The grade is B because these are the rules in the package insert, and no study has worked out how many incidents following them would prevent. The benefit is graded "medium" as a judgement: what it prevents is a stomach bleed or stomach cancer being caught late, but most acid reflux and heartburn is not serious. + +### 7. Do not ask a doctor for antibiotics for an ordinary cold + +- Cost: No money, and it also saves the money you would spend on medicines. +- In plain language: Antibiotics for a cold do not make you recover any faster than not taking them. In adults the risk of side effects is more than twice as high as with no antibiotic. For a runny nose with pus lasting less than 10 days, antibiotics have no established benefit. +- Benefit: A Cochrane systematic review pooled 6 randomised trials with 1147 people. Antibiotics for a cold compared with placebo showed no clear difference in the proportion who did not improve or whose symptoms went on (RR 0.83, 95% CI 0.60–1.14, not significant). The risk of an adverse reaction was 1.8 times that of the placebo group (95% CI 1.01–3.21). In adults it was 2.62 times (1.32–5.18); in children there was no increase (0.91, 0.51–1.63). For acute purulent rhinitis with a runny nose of less than 10 days, antibiotics had no clear benefit either (RR 0.73, 0.47–1.13). Adverse reactions were higher instead (RR 1.46, 1.10–1.94, about 46% higher). The authors conclude that routine use of antibiotics is not recommended in either situation. +- Evidence grade: A +- Sources: Kenealy T, Arroll B (2025). Antibiotics for the common cold and acute purulent rhinitis. Cochrane Database of Systematic Reviews, 11:CD000247. +- Notes: This version was republished in 2025, and the abstract states that the literature search ran up to 2013. Antibiotics are prescription medicines in China, so a pharmacy has to sell them against a prescription; do not take leftover antibiotics from home either. The benefit is graded "medium" as a judgement: what it prevents is adverse reactions, not death. + +### 8. When you have diarrhoea, start with oral rehydration salts; do not give anti-diarrhoeal medicines to children under 5 + +- Cost: A box of oral rehydration salts costs a few yuan to a dozen-odd yuan. Keep one box at home. +- In plain language: The most important thing in diarrhoea is replacing fluid, not stopping the diarrhoea. Mix oral rehydration salts with water as the insert says. Do not use juice drinks instead; they are too sweet and can leave you even shorter of fluid. Do not give children under 5 anti-diarrhoeal medicines such as loperamide; in severe cases these cause intestinal obstruction, which can be fatal. +- Benefit: The World Health Organization's manual on the treatment of diarrhoea states that "anti-diarrhoeal" medicines and antiemetics are of no real benefit in acute or persistent diarrhoea in children. They cannot prevent dehydration or improve nutrition. Some have dangerous and sometimes fatal side effects and must never be given to children under 5. Medicines that slow the movement of the intestine, such as loperamide, can cause severe paralytic ileus, which can be fatal, and can prolong an infection. The manual also states that drinks with too much sugar draw water out of the body into the intestine and cause hypernatraemic dehydration, for example soft drinks and packaged juice drinks. The WHO manual recommends a hypotonic formulation with a total osmolarity of 245 mOsm/L. Compared with the older formulation, it reduced unplanned intravenous infusions by 33%. A Cochrane review pooling 8 trials found a similar result (OR 0.59, 95% CI 0.45–0.79, roughly two fifths fewer). +- Evidence grade: A +- Sources: World Health Organization (2005). The treatment of diarrhoea: a manual for physicians and other senior health workers, 4th revision. ；Hahn S, Kim S, Garner P (2002). Reduced osmolarity oral rehydration solution for treating dehydration caused by acute diarrhoea in children. Cochrane Database of Systematic Reviews, (1):CD002847. +- Notes: When you buy it, check the osmolarity on the insert and choose the hypotonic formulation. The manual states that blood in a child's stool is most often bacterial dysentery, which needs a doctor and antibiotics; rehydrating at home alone is not enough. The benefit is graded "medium" as a judgement: the figures compare two rehydration solutions, not drinking against not drinking. + +### 9. Do not take painkillers for more than 15 days a month, or compound painkillers for more than 10 days, or the headache may be the medicine + +- Cost: No money. You have to keep count of the days you take medicine, and you have to hold back from taking it whenever you reach for it. +- In plain language: If your headache keeps coming back and you take a painkiller every day, the pain may get worse the more you take. The international headache classification calls this medication-overuse headache. Plain painkillers taken 15 or more days a month, compound painkillers 10 or more days, for more than 3 months in a row, are enough to meet it. Most people improve after they stop the drug they were taking too much of. +- Benefit: The third edition of the International Classification of Headache Disorders of the International Headache Society sets out the diagnostic criteria. A person who already has a primary headache disorder, who has headache on 15 or more days a month, and who has regularly overused headache medicines for more than 3 months. For non-opioid painkillers such as acetaminophen, aspirin and ibuprofen, the threshold for overuse is 15 or more days a month. For compound painkillers the threshold is 10 or more days a month. A compound painkiller is two or more kinds of pain-relieving drug put together, or one with an added helper ingredient such as caffeine, for example acetaminophen with codeine. If you take several non-opioid painkillers in turn, add the days up. The classification states that epidemiological evidence from several countries shows that of the people with headache on 15 or more days a month, more than half have this type of headache. Most patients improve after they stop the drug they have overused, and preventive treatment for headache works better too. +- Evidence grade: B +- Sources: Headache Classification Committee of the International Headache Society (2018). The International Classification of Headache Disorders, 3rd edition. Cephalalgia, 38(1):1-211. ；ICHD-3 online version 8.2 Medication-overuse headache (including 8.2.3 and 8.2.5). +- Notes: A painkiller whose name carries the character for coffee (咖) usually has caffeine in it, and the limit for a compound painkiller is 10 days. If the number of days you have a headache in a month keeps rising, see a neurologist instead of only adding medicines. The grade is B because the classification gives a diagnostic threshold, not a controlled trial. The benefit is graded "medium" as a judgement: what it buys back is days without headache, and it has nothing to do with death. \ No newline at end of file diff --git a/tools/i18n/README.md b/tools/i18n/README.md index fc3dd41a..a405bb95 100644 --- a/tools/i18n/README.md +++ b/tools/i18n/README.md @@ -16,6 +16,8 @@ node tools/i18n/split.mjs book/01-不要早死.md --locale en # 3. 逐块校一遍再合成 en/book/01-不要早死.md node tools/i18n/join.mjs book/01-不要早死.md --locale en +# 一次翻了好几节就用这个：它把「块里不再带 - 成本：」的那些节挨个 join +bash tools/i18n/join-ready.sh en vi # 4. 整本校一遍（CI 跑的就是这个） node tools/i18n/check.mjs @@ -25,13 +27,20 @@ node tools/site/build.mjs ``` `.i18n/` 是中间产物，已 gitignore。它同时是活的翻译进度：`split` 不覆盖已翻的块，所以每翻一块跑一次 -`join` 就能看到还差哪几块。 +`join` 就能看到还差哪几块。`join-ready.sh` 用的就是这个判据——一个块里还带 `- 成本：`， +就是这块还没翻。 + +**仓库里的路径一律走 `contentPath(code, rel)`（`tools/site/locales.mjs`），别用 `dir`。** +`dir` 是页面上线的目录，英文那份是空串（页面在根 URL 上）；`contentDir` 才是正文在仓库里 +的位置（英文是 `en`）。拿 `dir` 拼路径会拼到 `book/`——中文原文那一份，`join.mjs` 那样写 +会把英文译文写进原文。 ## 新增一种语言 -1. `tools/site/locales.json` 加一项：`code`、`dir`、`contentBase`（浏览器那边的相对根）、`assetBase`、 - `repoBlob`、`htmlLang`、`ogLocale`、`sortLocale`、`name`（语言自己的名字）、`accept`（匹配 - `navigator.language` 的正则）、`canonical`、`note`、`default` 或 `source`（各一个）。 +1. `tools/site/locales.json` 加一项：`code`、`dir`（页面上线的目录，默认语言是 `""` 表示在根上）、 + `contentDir`（正文在仓库里的位置，中文原文是 `""`）、`contentBase`（浏览器那边的相对根）、 + `assetBase`、`repoBlob`、`htmlLang`、`ogLocale`、`sortLocale`、`name`（语言自己的名字）、 + `accept`（匹配 `navigator.language` 的前缀）、`canonical`、`note`、`default` 或 `source`（各一个）。 2. `tools/site/locales/.json`：解析正文用的标记（`fields`、`dispute`、`todo`、`glossary`、 `contents`…）加界面文案 `strings`。`node tools/site/extract.mjs` 只能从模板抠中文，文案得自己写。 3. `tools/i18n/glossary..mjs`：词表。 @@ -48,6 +57,7 @@ node tools/site/build.mjs | `prompt.mjs` | 把 BRIEF.md 整个塞进模型提示词（要接 API 时用这个） | | `split.mjs` | 一节正文切成十来个条一块，放 `.i18n/<语言>/<节名>/partNN.md` | | `join.mjs` | 逐块校一遍再合成 `<语言>/book/<节名>.md`，对不上就不写盘 | +| `join-ready.sh` | 一轮把两种语言里所有「已经翻全」的节挨个 join 掉，省得一次次敲 | | `structure.mjs` | 一节（或一块）的结构骨架 + 逐条比对。`join` 和 `check` 共用同一份判据 | | `check.mjs` | 拿中文原文对全部译文，CI 的「译文检查」job 跑它 | | `anchors.mjs` | 译本里逐字照抄的那几样（成本标签注释、字段顺序、不许动的数字、指路、链接） | diff --git a/tools/i18n/check.mjs b/tools/i18n/check.mjs index 324fda96..f4ed88b1 100644 --- a/tools/i18n/check.mjs +++ b/tools/i18n/check.mjs @@ -18,7 +18,7 @@ import { readFileSync, existsSync, readdirSync } from 'node:fs'; import { resolve, dirname } from 'node:path'; import { fileURLToPath } from 'node:url'; -import { LOCALES, SOURCE, LOCALE, strings } from '../site/locales.mjs'; +import { LOCALES, SOURCE, LOCALE, strings, contentPath } from '../site/locales.mjs'; import { STRUCT, entrySkeleton } from './structure.mjs'; const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); @@ -39,7 +39,8 @@ for (const code of codes){ const files = readdirSync(resolve(ROOT, srcDir)).filter(f => f.endsWith('.md')).sort(); const have = []; for (const f of files){ - const dstPath = `${L.dir}book/${f}`; + // 译文所在的仓库目录：contentPath，不是 L.dir（那是页面上线上的地址，英文是空串） + const dstPath = contentPath(code, `book/${f}`); if (!existsSync(resolve(ROOT, dstPath))) continue; have.push(f); const src = entrySkeleton(read(srcDir + f), strings(SOURCE.code)); @@ -56,8 +57,9 @@ for (const code of codes){ if (!/^\[.*\]\((?:\.\.\/|\.\/)?README\.md\)/.test(first)) report(`${code} book/${f}`, `第一行「${first.slice(0, 40)}」不是回本语言目录的链接`); } - const all = existsSync(resolve(ROOT, `${L.dir}README.md`)); - console.log(`${L.name}（${L.dir}）：正文 ${have.length}/${files.length} 节已翻${all ? '' : '，README 还没有'}`); + const repoDir = (L.contentDir ? L.contentDir + '/' : '') || './'; + const all = existsSync(resolve(ROOT, contentPath(code, 'README.md'))); + console.log(`${L.name}（${repoDir}，页面上 ${L.dir || '/' }）：正文 ${have.length}/${files.length} 节已翻${all ? '' : '，README 还没有'}`); } if (problems.length){ diff --git a/tools/i18n/join-ready.sh b/tools/i18n/join-ready.sh new file mode 100755 index 00000000..7d8314f6 --- /dev/null +++ b/tools/i18n/join-ready.sh @@ -0,0 +1,21 @@ +#!/bin/bash +# 把 .i18n/<语言>/<节>/ 里全翻好的节合成 <语言>/book/<节>.md，全书一遍。 +# +# 判据是 split 不覆盖已翻的块这件事带来的副产品：块里还带「- 成本：」就是还没翻。 +# 合成用的 join.mjs 自己还会逐块校一遍数字和链接，对不上就不写盘—— +# 所以这里只管「哪些节翻全了」，判据还是在 join.mjs 那里。 +# +# tools/i18n/join-ready.sh en vi +cd "$(dirname "$0")/../.." || exit 1 +locales=("$@") +[ ${#locales[@]} -eq 0 ] && locales=(en vi) + +for loc in "${locales[@]}"; do + for dir in .i18n/$loc/*/; do + [ -d "$dir" ] || continue + # 块里还带中文字段名 = 还没翻完，这节这次跳过（重跑这个脚本就好） + grep -l -- '- 成本：' "$dir"part*.md 2>/dev/null | grep -q . && continue + name=$(basename "$dir") + node tools/i18n/join.mjs "book/$name.md" --locale "$loc" 2>&1 | grep -E '合好|处对不上|没写进' + done +done diff --git a/tools/i18n/join.mjs b/tools/i18n/join.mjs index 6c435ada..4c53f2f6 100644 --- a/tools/i18n/join.mjs +++ b/tools/i18n/join.mjs @@ -12,7 +12,7 @@ import { readFileSync, writeFileSync, existsSync, readdirSync, mkdirSync } from 'node:fs'; import { resolve, dirname, basename } from 'node:path'; import { fileURLToPath } from 'node:url'; -import { LOCALES, SOURCE, strings } from '../site/locales.mjs'; +import { LOCALES, SOURCE, strings, contentPath } from '../site/locales.mjs'; import { splitSection } from './split.mjs'; import { STRUCT, entrySkeleton } from './structure.mjs'; @@ -91,13 +91,16 @@ for (const srcPath of files){ } pieces.push(k === 0 ? src.head.join('\n') + '\n' + body : body); } - if (trouble) { console.error(`${locale} ${name}：${trouble} 处对不上，没写进 ${L.dir}book/${name}`); continue; } + // 写进仓库里的译文目录：contentPath，不是 L.dir。L.dir 是页面上线上的地址， + // 英文那份是空串（页面在根上），拿它拼就是 book/ —— 中文原文那一份。 + const dst = contentPath(locale, `book/${name}`); + if (trouble) { console.error(`${locale} ${name}：${trouble} 处对不上，没写进 ${dst}`); continue; } let out = pieces.join('\n'); const tail = `${dir}/tail.txt`; if (existsSync(resolve(ROOT, tail))) out += '\n' + read(tail); - mkdirSync(dirname(resolve(ROOT, `${L.dir}book/${name}`)), { recursive: true }); - writeFileSync(resolve(ROOT, `${L.dir}book/${name}`), out); + mkdirSync(dirname(resolve(ROOT, dst)), { recursive: true }); + writeFileSync(resolve(ROOT, dst), out); done++; console.log(`${locale} book/${name}：${parts.length} 块合好，${src.count} 条，${((out.length / 1024) | 0)}K`); } diff --git a/tools/site/build.mjs b/tools/site/build.mjs index 6035ebd8..13e54333 100644 --- a/tools/site/build.mjs +++ b/tools/site/build.mjs @@ -202,63 +202,43 @@ function docLinks(docs, L, strings, S) { return `${escText(strings['docs.label'])}${S.listSep}${links.join(' · ')}`; } -// 根目录那个小页：默认英文（PUBLISHED 里 default 的那种），浏览器要别的语言就去那种， -// 也可以自己点。关掉 JS 也点得动。 -// 只列 PUBLISHED：还没翻完的那种（眼下是越南文）hreflang 和链接里都不出现—— +// 根 URL 现在直接发英文（默认语言就住在根上，见 locales.json 里 en 的 dir: ""）。 +// 以前这里是「跳语言的小页」——一块小 HTML，开 JS 就按 navigator.language 跳走。 +// 那一版有两个问题：① 根 URL 发出的是中文标题的跳页，英文默认并没有真的落在根上； +// ② 自动跳转会把读者直接送走，页内的语言下拉和锚点都没机会用。 +// 现在保留的只有它唯一无可替代的一点：?lang=xx 这个老链接形式（README 和别处都这么写过）。 +// 命中就跳，命中不了就留在英文页上——页内下拉随时能换。 +// +// 只列 PUBLISHED：还没翻完的那种 hreflang 和跳转表里都不出现—— // 列出来点进去是 404，比不列更糟。翻完重跑这个脚本，它自己回来。 -export function chooserPage(site) { - const list = PUBLISHED - .map(l => `
  • ${escText(l.name)}${escText(l.note)}
    • `) - .join('\n'); - // 跳转表按 accept 排，命中不了就回默认那一种。?lang=xx 可以指定。 - const pick = ` const q = new URLSearchParams(location.search).get('lang'); - const list = ${JSON.stringify(PUBLISHED.map(l => ({ code: l.code, url: l.canonical, test: new RegExp('^' + l.accept, 'i') })))}; - const dflt = ${JSON.stringify((PUBLISHED.find(l => l.default) ?? PUBLISHED[0])?.canonical ?? site + '/')}; - const hit = q ? list.find(l => l.code === q) : list.find(l => l.test.test(navigator.language || '')); - location.replace((hit && hit.url) || dflt);`; - return ` - - - - -高性价比人生指南 · Choose your language - - -${hreflangLinks()} - - - -
    -

    高性价比人生指南

    -

    选一种语言 · Choose a language

    -
      -${list} -
    -
    - - - `; } // ---------------------------------------------------------------- 渲染一份页面 -// 这一轮真的会生成页面的语言：README 在的才算。还没翻完的语言（眼下是 vi） -// 不能出现在 hreflang、跳语言页和页内语言下拉里——那些都是读者会点开的链接， +// 这一轮真的会生成页面的语言：README 在的才算。还没翻完的那种语言 +// 不能出现在 hreflang 和页内语言下拉里——那些都是读者会点开的链接， // 点进去只有一个 404，比不列更糟。翻完重跑这个脚本，它自己回来。 const readmeOf = L => (L.contentDir ? L.contentDir + '/' : '') + 'README.md'; export const PUBLISHED = LOCALES.filter(L => existsSync(resolve(ROOT, readmeOf(L)))); @@ -351,6 +331,9 @@ export function renderPage(L, template, stats) { code: L.code, dir: L.dir, contentBase: L.contentBase, assetBase: L.assetBase, repoBlob: L.repoBlob, htmlLang: L.htmlLang, sortLocale: L.sortLocale, locales: PUBLISHED.map(l => l.code), langNames: Object.fromEntries(PUBLISHED.map(l => [l.code, l.name])), + // 每种语言页面上线上的地址，末尾带斜杠（默认语言是 ''，页面就在根上）。 + // 页内的语言下拉要靠它算跳转目标，所以得整张表，不能只有当前这页的 dir。 + langDirs: Object.fromEntries(PUBLISHED.map(l => [l.code, l.dir])), offline: false, files: contents.map(c => c.path), // 这一份真正收了哪几节，页面照这个取 docFiles: docs.map(d => d.path), @@ -401,12 +384,22 @@ export function sitemapXml() { alternates.push(` `); const url = (loc, priority) => ` \n ${escText(loc)}\n${alternates.join('\n')}\n ${priority}\n `; + // 每条只列一次。默认语言住在根上，它的 canonical 就是 SITE_URL()， + // 而根 URL 本来就要单列一条（读者直接打开的就是它）——不去重 sitemap 里 + // 同一个地址会出现两次，一次 1.0 一次 0.9。 + const seen = new Set(); + const entry = (loc, priority) => { + if (seen.has(loc)) return ''; + seen.add(loc); + return url(loc, priority) + '\n'; + }; + let out = entry(SITE_URL(), '1.0'); + for (const l of PUBLISHED) out += entry(l.canonical, l.default ? '0.9' : '0.8'); return '\n' + '\n' - + url(SITE_URL(), '1.0') + '\n' // 跳语言的小页 - + PUBLISHED.map(l => url(l.canonical, l.default ? '0.9' : '0.8')).join('\n') - + '\n\n'; + + out + + '\n'; } // robots.txt 和 sitemap.xml 是一对：sitemap 指的那个地址得是真的。一起从 site.json 拼， @@ -430,6 +423,7 @@ export const SITE_URL = () => SITE + '/'; // 但链接会指向别处，而且没人会一眼看出来。这里逐个对一遍。 function checkUrls() { for (const l of LOCALES) { + // dir 是空串时拼出来正好是 SITE + '/'：默认语言就住在根 URL 上（要求 2） const wantCanonical = SITE + '/' + l.dir; const wantBlob = REPO + '/blob/main/' + (l.contentDir ? l.contentDir + '/' : ''); if (l.canonical !== wantCanonical) @@ -464,7 +458,6 @@ for (const L of LOCALES){ const text = renderPage(L, template, stats); if (text !== null) targets.push({ path: `${L.dir}index.html`, text }); } -targets.push({ path: 'index.html', text: chooserPage(SITE) }); targets.push({ path: 'sitemap.xml', text: sitemapXml() }); targets.push({ path: 'robots.txt', text: robotsTxt() }); diff --git a/tools/site/locales.json b/tools/site/locales.json index 600a06e6..8127268e 100644 --- a/tools/site/locales.json +++ b/tools/site/locales.json @@ -2,16 +2,16 @@ { "code": "en", "default": true, - "dir": "en/", - "contentBase": "./", - "assetBase": "../ads/", + "dir": "", + "contentBase": "en/", + "assetBase": "ads/", "repoBlob": "https://github.com/streamentry/HowToLiveBetter/blob/main/en/", "htmlLang": "en", "ogLocale": "en_US", "sortLocale": "en", "name": "English", "accept": "en", - "canonical": "https://streamentry.github.io/HowToLiveBetter/en/", + "canonical": "https://streamentry.github.io/HowToLiveBetter/", "note": "Default edition", "contentDir": "en" }, diff --git a/tools/site/locales.mjs b/tools/site/locales.mjs index aeeafd2d..021a0518 100644 --- a/tools/site/locales.mjs +++ b/tools/site/locales.mjs @@ -25,9 +25,17 @@ export function strings(code) { return raw; } -/** 语言目录，末尾带斜杠：en/、vi/、zh/ */ +/** 语言目录（页面上线上的地址），末尾带斜杠。默认语言在根上，是 ''：en/、vi/、zh/ */ export const dir = code => locale(code).dir; /** 正文所在的仓库根相对目录，末尾无斜杠：zh 是 ''（就在根上），en 是 'en' */ export const contentDir = code => locale(code).contentDir; -/** 浏览器那边的相对根：zh 是 ../，en/vi 是 ./。给页面里的 fetch 和相对链接用 */ -export const base = code => locale(code).contentBase; \ No newline at end of file +/** 浏览器那边的相对根：zh 是 ../，根上那份是 en/。给页面里的 fetch 和相对链接用 */ +export const base = code => locale(code).contentBase; + +/** 正文所在的仓库根相对路径：contentPath('en', 'book/x.md') 是 en/book/x.md。 + 仓库里找译文一律走这里，别用 dir()——dir 是线上地址，en 的 dir 现在是空串（页面在根上）， + 拿它拼仓库路径会拼成 book/x.md，也就是中文原文那一份。 */ +export const contentPath = (code, rel) => { + const d = contentDir(code); + return d ? d + '/' + rel : rel; +}; \ No newline at end of file diff --git a/tools/site/page.template.html b/tools/site/page.template.html index a4a5f026..1453f227 100644 --- a/tools/site/page.template.html +++ b/tools/site/page.template.html @@ -1281,18 +1281,26 @@

    高性价比人生指南

    } /* ---------- 语言切换 ---------- - 三种语言的页面在 /en/、/zh/、/vi/，文件名和锚点都一样，所以换语言只要换目录： - 当前的筛选参数和 #e-13-1 一起带过去，落点还是同一条。离线副本没有目录可跳，隐藏。 */ + 各语言的页面只在目录上不同（/、/zh/、/vi/），文件名和锚点都一样，所以换语言只要换目录： + 当前的筛选参数和 #e-13-1 一起带过去，落点还是同一条。离线副本没有目录可跳，隐藏。 + + 站点的根要从当前目录里退出来再拼，不能拿 SITE.dir 直接拼：SITE.dir 是「这一页所在 + 目录」（页面在 /vi/ 就是 vi/），拿它拼出来还是同一页，换语言等于没换。退掉当前这层 + 之后拿 SITE.langDirs[code] 拼，才是目标语言的那一页。默认语言 dir 是空串，退不掉， + 这时 here 本身��是站点的根。 */ function buildLangPicker(){ const sel = document.getElementById('lang'); if (!sel) return; - const here = location.pathname.replace(/[^/]*$/, ''); // 当前页所在目录 + const here = location.pathname.replace(/[^/]*$/, ''); // 当前页所在目录，如 /HowToLiveBetter/vi/ + const root = SITE.dir && here.endsWith(SITE.dir) // 站点根，如 /HowToLiveBetter/ + ? here.slice(0, -SITE.dir.length) : here; + const dirs = SITE.langDirs || {}; // 离线单文件版没有这一项 for (const code of SITE.locales){ const o = document.createElement('option'); o.value = code; o.textContent = SITE.langNames[code]; o.selected = code === SITE.code; if (SITE.offline || code === SITE.code){ o.disabled = code === SITE.code; } - else o.dataset.url = (here === SITE.dir ? here : SITE.dir) + location.search + location.hash; + else o.dataset.url = root + (dirs[code] || '') + location.search + location.hash; sel.appendChild(o); } if (SITE.offline){ sel.closest('.lang').hidden = true; return; } diff --git "a/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" "b/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" new file mode 100644 index 00000000..7b8d7a84 --- /dev/null +++ "b/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" @@ -0,0 +1,370 @@ +[← 回总目录](../README.md) + +# 2. 不要慢慢死 + +本节只收两类慢性风险：一类对总死亡率影响大，一类证据硬。条目按性价比从高到低排。有几条标了「随机试验」，意思是把人随机分成两组来比，这种结果最可信。其余数字都来自只跟踪记录、不做分组的研究。这类研究算出的 HR、RR 等风险比里混着两种干扰：这些人本来就不一样，有人是先病了才这么做。所以这些数字只说明方向和大概多大，不代表照做就能降这么多。 + +### 1. Bỏ thuốc lá, càng sớm càng tốt + +- Chi phí: Không tốn tiền, ngược lại còn tiết kiệm tiền. Một bao mỗi ngày khoảng 20–30 tệ, bỏ thuốc thì giữ lại được khoản đó. Chỗ khó là phải chịu giai đoạn cai, thường kéo dài từ vài tuần đến vài tháng. +- Nói thẳng: Người hút thuốc sống ít hơn người không hút trung bình trên mười năm. Bỏ thuốc trước 40 tuổi thì xoá được khoảng chín phần mười rủi ro tử vong do hút thuốc kéo dài. Bỏ càng sớm thì lấy lại được càng nhiều. +- Lợi ích: Một nghiên cứu tại Mỹ theo dõi một nhóm người, chỉ ghi nhận, không chia nhóm đối chứng, cho thấy người còn hút thuốc có tuổi thọ dự kiến ngắn hơn người chưa từng hút trên 10 năm. Bỏ thuốc trước 40 tuổi xoá được khoảng 90% rủi ro tử vong do hút thuốc tiếp diễn. Người bỏ ở tuổi 25–34 sống thêm khoảng 10 năm, người bỏ ở tuổi 35–44 sống thêm khoảng 9 năm, người bỏ ở tuổi 45–54 sống thêm khoảng 6 năm (bản gốc báo cáo theo năm sống). Nghiên cứu Trung Quốc tương tự: trong thập niên 2010, rủi ro tử vong của nam giới thành thị hút thuốc là 1.65 lần người không hút (RR 1.65), nam giới nông thôn là 1.22 lần. Sau khi bỏ thuốc đủ 10 năm, phần rủi ro do hút thuốc gần như biến mất. +- Mức bằng chứng: A +- Nguồn: Jha P và cộng sự (2013). 21st-century hazards of smoking and benefits of cessation in the United States. NEJM. ; Chen Z và cộng sự (2015). Contrasting male and female trends in tobacco-attributed mortality in China: evidence from successive nationwide prospective cohort studies. Lancet. ; Oberg M và cộng sự (2011). Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet. (hai con số về khói thuốc thụ động) +- Ghi chú: Hút phải khói của người khác cũng gây chết: năm 2004 trên thế giới có khoảng 603.000 người chết vì khói thuốc thụ động, chiếm khoảng 1% tổng số ca tử vong trong năm đó. Người không hút vẫn nên tránh khói của người khác, và nhất là không để con mình hút phải. Cách bỏ thuốc nằm ở bốn mục kế tiếp: mục 3 (thuốc cai thuốc lá), mục 4 (chọn một ngày cai), mục 5 (đến phòng khám cai thuốc lá), mục 6 (thuốc điện tử) +### 2. Đừng hút thuốc trong nhà và trong xe, và cũng đừng để khách hút trong nhà + +- Chi phí: Không tốn tiền, cũng không mất thời gian. Chỗ khó là phải nói với gia đình và khách đừng hút trong nhà. +- Nói thẳng: Mỗi năm trên thế giới có khoảng 600.000 người chết vì hút phải khói thuốc của người khác, gần ba phần mười là trẻ con. Người hút phải khói thuốc lâu năm thì nguy cơ cao huyết áp, bệnh tim, đột quỵ đều cao hơn khoảng ba phần mười. Hút phải ở trong nhà thì hại hơn ở bên ngoài. +- Lợi ích: Năm 2004 có 603.000 người trên thế giới chết vì khói thuốc thụ động, khoảng 1.0% tổng số ca tử vong, trong đó 28% là trẻ em. Gộp 57 nghiên cứu lại với nhau: xác suất cao huyết áp ở người hút phải khói thuốc khoảng 1.28 lần người không hút phải (tỷ số khả năng). Bệnh tim là 1.39 lần, nhồi máu cơ tim là 1.50 lần, đột quỵ là 1.36 lần. Nguy cơ hút phải khói thuốc trong nhà cao hơn ở ngoài nhà +- Mức bằng chứng: A +- Nguồn: Öberg M, Jaakkola MS, Woodward A, Peruga A, Prüss-Ustün A (2011). Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet: 「603,000 deaths were attributable to second-hand smoke in 2004, which was about 1·0% of worldwide mortality. 47% of deaths from second-hand smoke occurred in women, 28% in children, and 26% in men」, 「61% of DALYs were in children」. ; (2026). The Associations Between Secondhand Smoke Exposure and Various Cardiovascular Diseases: A Meta-Analysis. Nicotine & Tobacco Research: 57 nghiên cứu, 「hypertension (OR: 1.28, 95% CI: 1.15 to 1.40), heart disease (OR: 1.39, 95% CI: 1.28 to 1.50), myocardial infarction (OR: 1.50, 95% CI: 1.17 to 1.84), stroke (OR: 1.36, 95% CI: 1.18 to 1.54)」「Home exposure has a higher risk of CVD than non-home exposure」. +- Ghi chú: Xử lý cái nhà trước, vì ở nhà bạn ở lâu hơn, và nhiều nghiên cứu gộp lại cho thấy khói thuốc thụ động trong nhà hại hơn ở ngoài nhà. Trong tổng thiệt hại sức khoẻ do khói thuốc thụ động, 61% rơi vào trẻ em. Muốn tự bỏ thuốc thì xem mục 1 của phần này. +### 3. Đừng chỉ dựa vào ý chí để cai, hãy đi lấy thuốc trước: tỷ lệ cai được có thể hơn gấp đôi + +- Chi phí: Miếng dán nicotin và kẹo nhai là thuốc không kê đơn, mua ở quầy hiệu thuốc là được. Một liệu trình dài 8 đến 12 tuần, giá vài trăm tệ đến hơn một nghìn tệ. Vareniclin và bupropion là thuốc kê đơn, phải đến phòng khám cai thuốc lá hoặc khoa Hô hấp để bác sĩ kê. Số tiền này gần như bù lại với số tiền mua thuốc lá bỏ đi trong cùng thời gian. +- Nói thẳng: Chỉ dựa vào ý chí thì đa số không cai được. Người uống vareniclin cai được với tỷ lệ hơn gấp đôi người uống placebo. Người dùng sản phẩm thay thế như miếng dán nicotin hay kẹo nhai cai được nhiều hơn khoảng năm phần mười so với người không dùng thuốc. Dùng cả miếng dán lẫn kẹo nhai thì còn cao hơn khoảng hai phần mười nữa. +- Lợi ích: Gộp 41 thử nghiệm chia nhóm ngẫu nhiên với 17395 người lại, tỷ lệ cai được ở nhóm vareniclin là 2.32 lần nhóm placebo (RR 2.32, 95% CI 2.15–2.51, đây là khoảng tin cậy; bằng chứng chắc chắn). Vareniclin cũng cao hơn bupropion (RR 1.36, cao khoảng 36%) và cao hơn khi chỉ dùng một sản phẩm thay thế nicotin (RR 1.25, cao khoảng 25%). Sản phẩm thay thế nicotin so với không dùng thuốc: 133 thử nghiệm, 64640 người, tỷ lệ cai được cao hơn khoảng 55% (RR 1.55, 95% CI 1.49–1.61). Miếng dán cộng thêm một dạng có tác dụng nhanh (kẹo nhai, viên ngậm) cao hơn khoảng 27% so với chỉ dùng một dạng (RR 1.27, 95% CI 1.17–1.37, 16 thử nghiệm, 12169 người, bằng chứng chắc chắn). +- Mức bằng chứng: A +- Nguồn: Livingstone-Banks J, Fanshawe TR, Thomas KH và cộng sự (2023). Nicotine receptor partial agonists for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD006103. ; Hartmann-Boyce J, Chepkin SC, Ye W, Bullen C, Lancaster T (2018). Nicotine replacement therapy versus control for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD000146. ; Theodoulou A, Chepkin SC, Ye W và cộng sự (2023). Different doses, durations and modes of delivery of nicotine replacement therapy for smoking cessation. Cochrane Database of Systematic Reviews, 6, CD013308. ; Ủy ban Sức khoẻ Thành phố Thượng Hải (上海市卫生健康委员会) (2021). Chọn đúng thuốc thì cai thuốc lá dễ hơn, 选对药物，让戒烟轻松一点: 「市场上可见的戒烟药物主要有三种，即尼古丁替代疗法药物、安非他酮、伐尼克兰」 — thuốc cai thuốc lá bán trên thị trường có ba loại, đó là thuốc liệu pháp thay thế nicotin, bupropion và vareniclin. 「尼古丁替代疗法药物属于非处方药（OTC），可通过药店柜台购买；而安非他酮、伐尼克兰属于处方药，须到医院戒烟门诊或呼吸内科就诊，凭医师处方经药师调配后才能得到」 — thuốc liệu pháp thay thế nicotin thuộc loại thuốc không kê đơn (OTC), mua được ở quầy hiệu thuốc; còn bupropion và vareniclin là thuốc kê đơn, phải đến phòng khám cai thuốc lá hoặc khoa Hô hấp ở bệnh viện, mang theo đơn của bác sĩ và do dược sĩ bốc thuốc mới lấy được. +- Ghi chú: Ở Trung Quốc đang bán đúng ba loại thuốc cai thuốc lá kể trên, và chỉ sản phẩm thay thế nicotin là thuốc không kê đơn. Các phản ứng thường gặp của vareniclin là buồn nôn, nằm mê sảng miêng, ngủ không được, nên người có tiền sử tâm thần phải nói với bác sĩ. Một liệu trình thay thế nicotin thường dài 8 đến 12 tuần, đừng tự ý ngừng đột ngột, hãy giảm liều theo đúng lời bác sĩ. Thuốc chỉ xử lý cơn khó chịu của vài tuần đầu cai, không xử lý được những tình huống bạn thèm hút, nên phải dùng cùng với mục 4 của phần này (chọn một ngày cai) và mục 5 của phần này (đến phòng khám cai thuốc lá). +### 4. Chọn một ngày cai, đến ngày đó bỏ một lần, đừng giảm dần trước + +- Chi phí: Không tốn tiền. Chỉ là chọn một ngày trên lịch, báo trước cho người nhà và đồng nghiệp. +- Nói thẳng: Nhiều người tính giảm mỗi ngày vài điếu trước, rồi dần xuống không. Nhưng đến tháng thứ sáu, trong nhóm đã chọn ngày và bỏ một lần có 22% không hút lại, còn nhóm giảm dần chỉ 15.5%. Nên hãy chọn một ngày rồi bỏ một lần vào ngày đó. +- Lợi ích: 697 người hút thuốc ở Anh được chia ngẫu nhiên thành hai nhóm. Một nhóm bỏ một lần vào ngày cai, nhóm kia giảm 75% lượng hút trong hai tuần trước ngày cai. Cả hai nhóm đều có y tá hỗ trợ hành vi và dùng sản phẩm thay thế nicotin quanh ngày cai. Ở tuần 4, 49.0% nhóm bỏ một lần không hút lại (95% CI 43.8–54.2, đây là khoảng tin cậy), so với 39.2% ở nhóm giảm dần (34.0–44.4), tỷ số rủi ro tương đối 0.80 (0.66–0.93). Đến tháng 6, 22.0% nhóm bỏ một lần (18.0–26.6) so với 15.5% nhóm giảm dần (12.0–19.7), tỷ số rủi ro tương đối 0.71 (0.46–0.91). Với người vốn đã thích giảm dần, tỷ lệ thành công ở tuần 4 cũng thấp hơn so với người thích bỏ một lần (38.3% so với 52.2%). +- Mức bằng chứng: A +- Nguồn: Lindson-Hawley N, Banting M, West R, Michie S, Shinkins B, Aveyard P (2016). Gradual Versus Abrupt Smoking Cessation: A Randomized, Controlled Noninferiority Trial. Annals of Internal Medicine, 164(9), 585–592. ; Theodoulou A, Chepkin SC, Ye W và cộng sự (2023). Different doses, durations and modes of delivery of nicotine replacement therapy for smoking cessation. Cochrane Database of Systematic Reviews, 6, CD013308 (đoạn nói về việc dùng sớm sản phẩm thay thế nicotin). +- Ghi chú:Tranh cãi. Cochrane trước đây có một bài tổng quan chuyên so hai cách này, và kết luận là hai cách gần như nhau (RR 0.94, 95% CI 0.79–1.13, 10 thử nghiệm, 3760 người), nhưng bài đó đã bị rút lại năm 2019 và không còn cập nhật. Nên bằng chứng tốt nhất hiện có là thử nghiệm chia nhóm ngẫu nhiên nêu trên, và hướng của nó nghiêng về bỏ một lần. Trong thử nghiệm đó, cả hai nhóm còn dùng sản phẩm thay thế nicotin trước ngày cai, và cách làm này cũng có bằng chứng: dùng sớm cho tỷ lệ cai được cao hơn khoảng 25% so với ngày cai mới bắt đầu (RR 1.25, 95% CI 1.08–1.44, 9 thử nghiệm, 4395 người, bằng chứng chắc chắn vừa). Nên cách xếp hợp lý là chọn ngày trước, dán miếng dán sớm hai tuần, rồi bỏ một lần vào ngày đó. + +### 5. Đến phòng khám cai thuốc lá, hoặc gọi 12320 hỏi địa phương có dịch vụ cai thuốc lá không + +- Chi phí: Phí khám một lần vài tệ đến vài chục tệ. Hỗ trợ hành vi thường gồm 4 đến 8 buổi gặp, mỗi buổi nửa tiếng đến một tiếng, cộng lại thường dưới 5 tiếng. Gọi đường dây nóng thì không tốn tiền. +- Nói thẳng: Thuốc và việc có người định kỳ hỏi thăm bạn là hai chuyện khác nhau, ghép lại thì hiệu quả nhất. Người vừa uống thuốc vừa có người theo sát cai được gần gấp đôi so với người chỉ nghe một lời khuyên ngắn. Điện thoại cũng tính: đường dây nóng cai thuốc lá chủ động gọi lại bạn thì nâng tỷ lệ cai được thêm hai đến ba phần mười. +- Lợi ích: Gộp 52 thử nghiệm với 19488 người lại, tỷ lệ cai được ở nhóm dùng thuốc cộng hỗ trợ hành vi là 1.83 lần nhóm chăm sóc thường quy hoặc lời khuyên ngắn (RR 1.83, 95% CI 1.68–1.98, đây là khoảng tin cậy; bằng chứng chất lượng cao). Trong 43 thử nghiệm tuyển người ở cơ sở y tế thì là 1.97 lần (1.79–2.18). Tư vấn qua điện thoại tính riêng: trong số người đã tự gọi đường dây nóng cai thuốc lá, người được gọi lại nhiều lần không cần nhắc cai được nhiều hơn khoảng 38% so với người chỉ nhận tài liệu hoặc chỉ nói chuyện một lần (RR 1.38, 95% CI 1.19–1.61, 14 thử nghiệm, 32484 người); người chưa từng tự gọi mà được bên nghiên cứu đưa tư vấn qua điện thoại thì cao hơn khoảng 25% (RR 1.25, 95% CI 1.15–1.35, 65 thử nghiệm, 41233 người). Cả hai đều là bằng chứng chắc chắn vừa. +- Mức bằng chứng: A +- Nguồn: Stead LF, Koilpillai P, Fanshawe TR, Lancaster T (2016). Combined pharmacotherapy and behavioural interventions for smoking cessation. Cochrane Database of Systematic Reviews, 3, CD008286. ; Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J (2019). Telephone counselling for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD002850. ; Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (中国疾病预防控制中心) (2021). Mini chương trình nhỏ WeChat 「Nền tảng cai thuốc lá Trung Quốc」 chính thức ra mắt: 「在健康中国行动控烟行动工作组指导下中国疾病预防控制中心和世界卫生组织驻华代表处联合制作了中国权威戒烟资源库『中国戒烟平台』微信小程序。小程序内容主要包括：戒烟热线、戒烟门诊详细信息，以及线上戒烟服务资源等」 — dưới sự chỉ đạo của nhóm công tác kiểm soát thuốc lá thuộc Hành động Trung Quốc Khỏe mạnh, Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc và Văn phòng đại diện Tổ chức Y tế Thế giới tại Trung Quốc đã cùng làm mini chương trình WeChat 「Nền tảng cai thuốc lá Trung Quốc」, một kho tài nguyên cai thuốc lá chính thống cho Trung Quốc. Nội dung chính của mini chương trình gồm đường dây nóng cai thuốc lá, thông tin chi tiết về phòng khám cai thuốc lá và các dịch vụ cai thuốc lá trực tuyến. Chân trang cùng trang ghi 「健康咨询电话：12320」 — số điện thoại tư vấn sức khoẻ: 12320. +- Ghi chú: Tìm kiếm dịch vụ ở Trung Quốc thế nào: tìm mini chương trình 「Nền tảng cai thuốc lá Trung Quốc」 trên WeChat, trong đó có danh sách đường dây nóng và phòng khám cai thuốc lá theo từng địa phương. Cũng có thể gọi 12320 hỏi xem quanh mình có phòng khám cai thuốc lá không. Phần lớn các thử nghiệm nêu trên làm ở nước ngoài, cường độ dịch vụ giữa các phòng khám ở những nơi khác nhau chênh lệch rất lớn, nên các con số chỉ xem theo hướng. Về phần thuốc đi kèm, xem mục 3 của phần này (thuốc cai thuốc lá). +### 6. Không cai được nữa thì mới nghĩ đến thuốc điện tử, người không hút thuốc đừng đụng vào + +- Chi phí: Máy và viên thuốc điện tử tính ra vài chục tệ đến vài trăm tệ mỗi tháng. Ở Trung Quốc chỉ mua được vị thuốc lá, và chỉ ở cửa hàng có giấy phép bán lẻ độc quyền thuốc lá. +- Nói thẳng: Chuyển từ thuốc lá sang thuốc điện tử có nicotin thì tỷ lệ cai được cao hơn khoảng sáu phần mười so với người dùng miếng dán nicotin, kết luận này rất đáng tin. Nhưng nó không vô hại, dữ liệu an toàn về dùng lâu dài thì chưa có. Người vốn không hút mà dùng nó không có lợi ích nào. +- Lợi ích: Một bài tổng quan được cập nhật liên tục gồm 80 thử nghiệm chia nhóm ngẫu nhiên, 29861 người. Thuốc điện tử có nicotin so với liệu pháp thay thế nicotin cho tỷ lệ cai được cao hơn khoảng 61% (RR 1.61, 95% CI 1.23–2.12, đây là khoảng tin cậy; bằng chứng chắc chắn, 11 thử nghiệm, 4114 người). Đổi ra con số tuyệt đối thì xấp xỉ mỗi 100 người có thêm khoảng 4 người cai được (thêm từ 1 đến 7 người). Tỷ lệ biến chứng không mong muốn nghiêm trọng ở hai nhóm gần bằng nhau (chênh lệch rủi ro 0.01, 95% CI −0.01 đến 0.02, bằng chứng chắc chắn vừa). So với chỉ hỗ trợ hành vi hoặc không hỗ trợ gì thì cao hơn khoảng 75% (RR 1.75, 95% CI 1.39–2.20, bằng chứng chắc chắn thấp). +- Mức bằng chứng: A +- Nguồn: Lindson N, Livingstone-Banks J, Butler AR và cộng sự (2026). Electronic cigarettes for smoking cessation. Cochrane Database of Systematic Reviews, 8, CD010216. ; Cục Quản lý độc quyền thuốc lá nhà nước (国家烟草专卖局) (2022). Biện pháp quản lý thuốc điện tử, 电子烟管理办法（国家烟草专卖局公告 2022 年第 1 号，2022 年 5 月 1 日施行）. 第十八条 — người kinh doanh bán lẻ thuốc điện tử phải xin cơ quan quản lý độc quyền thuốc lá cấp giấy phép bán lẻ độc quyền thuốc lá. 第二十二条: 「禁止向未成年人出售电子烟产品。」 — bán sản phẩm thuốc điện tử cho người chưa thành niên bị cấm. 第二十三条第二款: 「任何个人、法人或者其他组织不得通过本办法规定的电子烟交易管理平台以外的信息网络销售电子烟产品、雾化物和电子烟用烟碱等。」 — không cá nhân, pháp nhân hay tổ chức nào được bán sản phẩm thuốc điện tử, chất lỏng tạo khói hoặc nicotin dùng cho thuốc điện tử qua bất kỳ mạng lưới thông tin nào ngoài nền tảng giao dịch thuốc điện tử mà Biện pháp này quy định. 第二十六条: 「禁止销售除烟草口味外的调味电子烟和可自行添加雾化物的电子烟。」 — cấm bán thuốc điện tử có hương vị khác vị thuốc lá, và thuốc điện tử mà người dùng tự thêm chất lỏng tạo khói vào được. +- Ghi chú:Tranh cãi. Tổ chức Y tế Thế giới không đề xuất thuốc điện tử làm cách cai thuốc lá. Lý do là dữ liệu an toàn dài hạn chưa đủ, và nó kéo người trẻ bắt đầu tiếp xúc với nicotin. Anh thì đưa nó vào dịch vụ cai thuốc lá. Ở Trung Quốc không cấm bán thuốc điện tử toàn diện, nhưng quản lý rất chặt, quy định nằm trong Biện pháp quản lý thuốc điện tử. Chỉ được bán vị thuốc lá, vị trái cây và loại tự thêm dầu vào thì không được bán. Chỉ mua được ở cửa hàng thật đã có giấy phép bán lẻ độc quyền thuốc lá. Bán ở bất kỳ trang web, cửa hàng trực tuyến hay ứng dụng mạng xã hội nào khác, ngoài nền tảng giao dịch thuốc điện tử thống nhất của nhà nước, là bất hợp pháp. Bán cho người chưa thành niên cũng bị cấm. Vì vậy thuốc điện tử vị trái cây còn mua được đi là qua kênh bất hợp pháp, nằm ngoài tầm kiểm soát. Thuốc điện tử " gây nghiện" trộn chất tổng hợp cần sa lan lọt ra ngoài theo cách đó, xem mục 4 ở phần 22 (đừng nhận viên thuốc điện tử người khác đưa). Sản phẩm được đo trong bài tổng quan đều là sản phẩm có nicotin chịu sự quản lý, dầu thuốc điện tử không rõ nguồn gốc không nằm trong đó. Theo thứ tự thì nên thử mục 3 của phần này (thuốc cai thuốc lá) trước, bằng chứng ở đó chắc hơn và cũng rẻ hơn. +### 7. Bỏ nước giải ngọt, chuyển sang loại không đường cũng không phải là cách giải quyết + +- Chi phí: Không tốn tiền, cũng không mất thời gian. Chuyển sang nước trắng hoặc trà không đường còn tiết kiệm tiền. Chỗ khó là phải bỏ thói quen uống đồ ngọt, kể cả loại không đường. +- Nói thẳng: Người mỗi ngày uống hơn hai chai đồ ngọt có xác suất tử vong trong cùng thời gian cao hơn khoảng hai phần mười so với người gần như không uống. Mỗi ngày một đến hai chai thì cao hơn hơn một phần mười. Chuyển sang loại không đường cũng không thấy lợi ích: người mỗi ngày uống hơn hai ly đồ giải ngọt tổng hợp lại cao hơn khoảng một phần tư. +- Lợi ích: Mỹ có hai nghiên cứu lớn theo dõi một nhóm người. Cả hai chỉ ghi nhận, không chia nhóm đối chứng, gồm khoảng 118.000 người và 36.000 ca tử vong. Người uống hơn 2 phần mỗi ngày có rủi ro tử vong cao hơn khoảng 21% so với người uống dưới 1 phần mỗi tháng (HR 1.21). Người uống 1–2 phần mỗi ngày cao hơn khoảng 14% (HR 1.14). Nghiên cứu mẫu EPIC ở châu Âu cũng chỉ ghi nhận, không chia nhóm đối chứng, gồm 450.000 người, 10 nước. So người uống hơn 2 ly mỗi ngày với người uống dưới 1 ly mỗi tháng: với nước giải ngọt có đường, rủi ro cao hơn khoảng 8% (HR 1.08, 95% CI 1.01–1.16). Với nước giải ngọt tổng hợp, tức loại không đường, cao hơn khoảng 26% (HR 1.26, 1.16–1.35). Tính chung mọi loại nước giải, cao hơn khoảng 17% (HR 1.17, 1.11–1.22). +- Mức bằng chứng: A +- Nguồn: Malik VS và cộng sự (2019). Long-Term Consumption of Sugar-Sweetened and Artificially Sweetened Beverages and Risk of Mortality in US Adults. Circulation. ; Mullee A, Romaguera D, Pearson-Stuttard J và cộng sự (2019). Association Between Soft Drink Consumption and Mortality in 10 European Countries. JAMA Internal Medicine, 179(11), 1479-1490. +- Ghi chú:Tranh cãi nằm ở nước giải ngọt không đường. Mối liên hệ với đồ giải ngọt tổng hợp còn mạnh hơn với loại có đường, nhiều khả năng là nguyên nhân ngược lại: người đã thừa cân hoặc đã mắc đái thường mới chuyển sang loại không đường. Vì vậy đừng kết luận rằng chất tạo ngọt độc hơn. Nhưng điều này ít nhất cho thấy câu "chuyển sang không đường là xong" không có bằng chứng đứng sau. Chuyển sang nước trắng hoặc trà không đường là an toàn nhất. Ngoài ra, các nghiên cứu này chỉ ghi nhận và theo dõi, không làm thử nghiệm có nhóm đối chứng, người uống nhiều đồ ngọt vốn đã có lối sống kém hơn, nên các con số có thể đã lớn hơn thực tế. +### 8. Đừng nhai trầu cau + +- Chi phí: Không tốn tiền, cũng không mất thời gian, lại tiết kiệm được tiền mua trầu cau. Chỗ khó là bỏ được thói quen nhai. +- Nói thẳng: Người nhai trầu cau có xác suất tử vong trong cùng thời gian cao hơn khoảng hai phần mười so với người không nhai. Nguy cơ đái thường và hội chứng chuyển hoá cao hơn gần một nửa. Nguy cơ ung thư khoang miệng và ung thư thực quản phải tính riêng. +- Lợi ích: Gộp 17 nghiên cứu châu Á với 388.000 người lại với nhau. Người nhai trầu cau có rủi ro tử vong cao hơn người không nhai khoảng 21% (rủi ro tương đối 1.21, P=0.02). Con số tử vong tính trong 179.600 người đó. Nguy cơ đái thường cao hơn khoảng 47% (1.47). Nguy cơ hội chứng chuyển hoá cao hơn khoảng 51% (1.51). Trầu cau cũng là yếu tố nguy cơ đã biết của ung thư khoang miệng và ung thư thực quản +- Mức bằng chứng: A +- Nguồn: Yamada T, Hara K, Kadowaki T (2013). Chewing betel quid and the risk of metabolic disease, cardiovascular disease, and all-cause mortality: a meta-analysis. PLoS One, 8(8), e70679. +- Ghi chú: Nhai càng lâu năm, mỗi ngày nhai càng nhiều, nguy cơ ung thư khoang miệng càng cao. Sau khi bỏ, rủi ro đi xuống theo thời gian. Trầu cau có thuốc lá nguy hiểm hơn +### 9. Đổi muối ăn trong nhà sang muối ít natri (muối kali) + +- Chi phí: Một túi đắt hơn muối thường vài tệ. Tiện tay đổi khi mua, không tốn thêm thời gian. Vị gần như không đổi. +- Nói thẳng: Người đã từng bị đột quỵ, hoặc trên 60 tuổi bị cao huyết áp, đổi muối trong nhà sang muối ít natri thì xác suất tử vong trong năm năm thấp hơn khoảng 12%, đột quỵ thấp hơn khoảng 14%. +- Lợi ích: Một thử nghiệm chia người ngẫu nhiên thành hai nhóm, làm ở nông thôn Trung Quốc, 20995 người, tất cả là người đã từng bị đột quỵ hoặc bệnh nhân cao huyết áp trên 60 tuổi, theo dõi 4.74 năm. Kết quả: nhóm dùng muối ít natri có rủi ro tử vong thấp hơn nhóm dùng muối thường khoảng 12% (RR 0.88). Đột quỵ thấp hơn khoảng 14% (RR 0.86). Sự kiện tim mạch chính thấp hơn khoảng 13% (RR 0.87). Với các sự kiện do kali máu quá cao, hai nhóm không thấy khác biệt có ý nghĩa thống kê. +- Mức bằng chứng: A +- Nguồn: Neal B và cộng sự (2021). Effect of Salt Substitution on Cardiovascular Events and Death. NEJM. ; O'Donnell M và cộng sự (2014). Urinary sodium and potassium excretion, mortality, and cardiovascular events. NEJM. (phe PURE bên kia) +- Ghi chú:Tranh cãi. Nghiên cứu PURE chỉ theo dõi ghi nhận dài hạn, không chia nhóm đối chứng. Nó báo rằng người thải ra không quá 3 g natri mỗi ngày có nguy cơ tử vong cộng biến chứng tim mạch cao hơn khoảng 27%. Người thải ra trên 7 g mỗi ngày cao hơn khoảng 15%. Theo kết quả đó, ăn quá ít và ăn quá nhiều đều không tốt, mức ở giữa là tốt nhất. Nhưng trong thử nghiệm chia nhóm ngẫu nhiên ở nông thôn Trung Quốc mà mục này dẫn, muối ít natri chỉ thay thế một phần, nên lượng natri không xuống thấp như vậy. Cần lưu ý thêm là người tham gia thử nghiệm là người lớn tuổi có nguy cơ cao. Người trẻ khoẻ mạnh đổi muối thì lợi ích nhận được nhỏ hơn nhiều. Người suy giảm chức năng thận, hoặc đang dùng thuốc giữ kali, nên hỏi bác sĩ trước khi đổi. + +### 10. Đánh răng kỹ, mỗi ngày làm sạch kẽ răng một lần, mất răng thì bịt lại sớm + +- Chi phí: Chỉ đồ hoặc bàn chải kẽ răng mỗi năm vài chục tệ, cạo vôi mỗi lần một hai trăm tệ. Mỗi ngày tốn thêm hai ba phút. Chỗ khó là biến việc làm sạch kẽ răng thành chuyện hằng ngày. +- Nói thẳng: Người mỗi ngày dùng chỉ đồ hoặc bàn chải kẽ răng có xác suất tử vong trong cùng thời gian thấp hơn khoảng một phần mười. Người lớn tuổi rụng hết răng có xác suất tử vong cao hơn khoảng chín phần mười. Người còn dưới 20 chiếc răng thì cao hơn khoảng gấp đôi. +- Lợi ích: Một nghiên cứu ở Nhật gồm 9676 người, theo dõi 6 năm. Người dùng dụng cụ làm sạch kẽ răng có rủi ro tử vong thấp hơn khoảng 11% (tỷ số rủi ro 0.89). Người dùng dụng cụ làm sạch lưỡi thấp hơn khoảng 23% (0.77). Một phân tích gộp khác gom nhiều nghiên cứu lại, đối tượng là người lớn tuổi sống trong cộng đồng. Người không còn chiếc răng nào có xác suất tử vong khoảng 1.87 lần (tỷ số khả năng, 95% CI 1.35–2.59, đây là khoảng tin cậy). Người còn dưới 20 chiếc răng thì khoảng 2.04 lần (1.67–2.49) +- Mức bằng chứng: B +- Nguồn: Wang K, Matsuyama Y, Kiuchi S và cộng sự (2026). Routine oral health practices and all-cause mortality. Journal of Dentistry. ; Ko MJ, Seo S, So JS và cộng sự (2026). Deteriorated oral health and function as risk factors for physical disability and mortality in community-dwelling older adults: a systematic review and meta-analysis. European Geriatric Medicine. +- Ghi chú:Tranh cãi. Các nghiên cứu này chỉ ghi nhận và theo dõi. Người răng xấu thường sức khoẻ chung và tình hình tài chính cũng kém hơn. Răng xấu làm kéo cả cơ thể xuống, hay cơ thể yếu kéo răng theo, thì không rõ. Nhưng việc này tốn rất ít tiền, và bệnh nướu răng cùng tình trạng mất răng vốn đã ảnh hưởng đến ăn uống +### 11. Mỗi ngày đi bộ 7000–8000 bước + +- Chi phí: Không tốn tiền. Mỗi ngày đi bộ 60–90 phút, chia vào đường đi làm và chuyến đi chợ cũng được. Chỗ khó là phải đi bộ mỗi ngày. +- Nói thẳng: Người mỗi ngày đi bộ 5800 bước có xác suất tử vong trong cùng thời gian thấp hơn khoảng bốn phần mười so với người chỉ đi bộ 3500 bước. Lên tới 7800 bước thì giảm khoảng bốn phần rưỡi. Từ đó lên nữa con số giảm thêm một chút rồi đi ngang. Người trên 60 tuổi thì 6000–8000 bước là đủ. +- Lợi ích: Gộp 15 nghiên cứu theo dõi một nhóm người, tất cả chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. 15 nghiên cứu đó gồm 47471 người và 3013 ca tử vong. Chia người thành bốn nhóm theo số bước mỗi ngày, trung vị của nhóm thấp nhất là 3553 bước/ngày. Trung vị của ba nhóm còn lại lần lượt là 5801, 7842, 10901 bước/ngày. Rủi ro tử vong của ba nhóm đó thấp hơn nhóm thấp nhất lần lượt khoảng 40%, 45%, 53% (HR 0.60, 0.55, 0.47). Ở người trên 60 tuổi, lợi ích đi ngang sau 6000–8000 bước mỗi ngày. Ở người dưới 60 tuổi thì đi ngang sau 8000–10000 bước. Một phân tích gộp khác gom nhiều nghiên cứu lại với nhau: tác dụng bắt đầu từ khoảng 3867 bước mỗi ngày. Từ đó, cứ thêm 1000 bước mỗi ngày thì rủi ro tử vong giảm khoảng 15%. +- Mức bằng chứng: A +- Nguồn: Paluch AE và cộng sự (2022). Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. Lancet Public Health. ; Banach M và cộng sự (2023). The association between daily step count and all-cause and cardiovascular mortality: a meta-analysis. European Journal of Preventive Cardiology. +- Ghi chú: Đoạn từ 4000 bước lên 7000 bước là hiệu quả chi phí tốt nhất, không cần cố đạt mười nghìn bước. Cần nhớ các nghiên cứu này chỉ ghi nhận và theo dõi: nhóm đi bộ ít nhất có lẫn người vốn bị bệnh không đi nổi. Họ đi ít vì sắp chết, không phải đi ít nên chết sớm. Nên khoảng cách mà các nghiên cứu này báo ra thường đã phóng đại. Dù vậy xu hướng càng đi bộ càng tốt thì rõ ràng. Mục này và mục 14 (mỗi tuần 150–300 phút vận động cường độ vừa) nói cùng một việc, chỉ khác cách đếm số, đạt một trong hai là được. +### 12. Có cao huyết áp, mỡ máu cao thì uống thuốc đúng y lệnh, đừng tự ý ngừng + +- Chi phí: Thuốc hạ huyết áp và statin mua theo cơ chế tập trung dược phẩm mỗi tháng vài tệ đến vài chục tệ. Mỗi ngày uống một lần, không tốn nhiều thời gian. Chỗ khó là phải uống tiếp, không được nghĩ khỏe rồi là bỏ. +- Nói thẳng: Huyết áp mỗi khi giảm 10 mmHg thì xác suất tử vong giảm khoảng 13%, đột quỵ giảm khoảng 27%. Statin hạ mỡ cholesterol mật độ thấp 1 đơn vị thì xác suất tử vong giảm khoảng một phần mười. Người uống thuốc đúng giờ có xác suất tử vong thấp hơn người hay bỏ liều từ ba phần mười đến bốn phần rưỡi. +- Lợi ích: Gộp nhiều thử nghiệm chia nhóm ngẫu nhiên: huyết áp tâm thu mỗi khi giảm 10 mmHg thì rủi ro tử vong thấp hơn khoảng 13% (RR 0.87). Huyết áp tâm thu là con số cao hơn trong hai số đo. Sự kiện tim mạch chính thấp hơn khoảng 20% (RR 0.80). Đột quỵ thấp hơn khoảng 27% (RR 0.73). Suy tim thấp hơn khoảng 28% (RR 0.72). Statin làm mỡ cholesterol mật độ thấp (LDL) đi xuống. LDL mỗi khi giảm 1.0 mmol/L thì rủi ro tử vong thấp hơn khoảng 10% (RR 0.90). Với mức giảm đó, sự kiện mạch máu chính thấp hơn khoảng 22% (RR 0.78). Một phân tích gộp khác gom nhiều nghiên cứu theo dõi lại, tất cả chỉ ghi nhận và theo dõi, không chia nhóm đối chứng: người uống đúng ≥80% số liều được kê có rủi ro tử vong thấp hơn người hay bỏ liều. Với statin thấp hơn khoảng 45% (RR 0.55). Với thuốc hạ huyết áp thấp hơn khoảng 29% (RR 0.71). +- Mức bằng chứng: A +- Nguồn: Ettehad D và cộng sự (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Cholesterol Treatment Trialists' (CTT) Collaboration (2010). Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170 000 participants in 26 randomised trials. Lancet. ; Chowdhury R và cộng sự (2013). Adherence to cardiovascular therapy: a meta-analysis of prevalence and clinical consequences. European Heart Journal. +- Ghi chú: Chỉ làm theo khi bác sĩ đã đánh giá rằng bạn cần thuốc, người khoẻ mạnh thì không. Con số của nhóm "kiên trì uống thuốc" đến từ các nghiên cứu chỉ ghi nhận và theo dõi, người uống được đều đặn vốn đã kỷ luật hơn, nên lợi ích bị nói quá. Con số từ hai nhóm thử nghiệm chia nhóm ngẫu nhiên ở trên đáng tin hơn. Phần này không tách riêng mục nào về đường huyết: bằng chứng cho thấy hạ đường huyết thấp hơn giúp giảm tỷ lệ tử vong toàn nguyên không nhất quán bằng huyết áp và mỡ máu. +### 13. Mỗi đêm ngủ khoảng 7 tiếng, giữ giờ sinh hoạt ổn định + +- Chi phí: Không tốn tiền. Phải dành thời gian ra để ngủ, phần lớn người lấy từ lúc lướt điện thoại. Chỗ khó là cố định giờ lên giường và giờ thức dậy. +- Nói thẳng: Người lâu dài ngủ dưới 7 tiếng có xác suất tử vong trong cùng thời gian cao hơn khoảng một phần mười. Giờ sinh hoạt càng loạn thì rủi ro càng cao. Người ngủ đều đặn nhất có xác suất tử vong thấp hơn người loạn nhất từ hai phần mười đến gần một nửa. Nói cách khác, giờ sinh hoạt đều đặn hay không nói lên rủi ro tử vong rõ hơn là ngủ bao nhiêu tiếng. +- Lợi ích: Gộp 16 nghiên cứu theo dõi, tất cả chỉ ghi nhận và theo dõi, không chia nhóm đối chứng, gồm 1.380.000 người và 113.000 ca tử vong. Người ngủ ít có rủi ro tử vong cao hơn khoảng 12% (RR 1.12). Người ngủ nhiều cao hơn khoảng 30% (RR 1.30). Một bảng tổng hợp khác chia người theo thời lượng ngủ, nhóm 7 tiếng có rủi ro thấp nhất. Dưới 7 tiếng, mỗi khoảng thiếu 1 tiếng thì rủi ro cao hơn khoảng 6% (RR 1.06). Trên 7 tiếng, mỗi giờ thêm vào rủi ro cao hơn khoảng 13% (RR 1.13). Ngân hàng sinh học Anh có dữ liệu vòng đeo tay của khoảng 61.000 người, cũng chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. Chia người thành năm nhóm theo mức độ đều đặn của giờ sinh hoạt. Bốn nhóm đều đặn hơn có rủi ro tử vong thấp hơn nhóm loạn nhất 20%–48%. Mức độ đều đặn của giờ sinh hoạt dự báo tử vong tốt hơn thời lượng ngủ. Một nghiên cứu khác của Ngân hàng sinh học Anh đo giờ ngủ bằng vòng đeo tay ở 51.562 người, trong đó 3.853 người mắc bệnh tim mạch mới trong thời gian nghiên cứu. Chênh lệch giờ sinh học xã hội là khoảng cách giữa điểm giữa của giờ ngủ ngày thường và điểm giữa của giờ ngủ cuối tuần. Người có chênh lệch giờ sinh học xã hội từ 2 tiếng trở lên thì nguy cơ tim mạch cao hơn khoảng 30% (HR 1.30, 95% CI 1.11–1.54). Mối liên hệ này vẫn còn ở người có thời lượng ngủ bình thường. +- Mức bằng chứng: A +- Nguồn: Cappuccio FP và cộng sự (2010). Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. ; Yin J và cộng sự (2017). Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. JAHA. ; Windred DP và cộng sự (2024). Sleep regularity is a stronger predictor of mortality risk than sleep duration: A prospective cohort study. Sleep. ; Kumar N, Krishnamurthy S (2026). Social jet lag is associated with incident cardiovascular disease independent of sleep duration and cardiac genetic risk. Journal of Internal Medicine. +- Ghi chú: Cần sửa là ngủ quá ít và giờ sinh hoạt loạn, không nhất thiết phải rút ngắn giấc ngủ có chủ ý. Người ngủ nhiều có rủi ro cao hơn, nhiều khả năng là nguyên nhân ngược lại: trầm cảm, bệnh mãn tính và ngưng thở khi ngủ khiến người ta ngủ nhiều hơn. Còn về việc giữ giờ sinh hoạt đều đặn thì hiện có hai nghiên cứu theo dõi hỗ trợ, cả hai chỉ ghi nhận và theo dõi, không chia nhóm đối chứng, đứng riêng thì là bằng chứng mức B. Sau một đêm thức khuya thì bù lại thế nào, xem mục 39 của phần này (tối hôm sau ngủ bù luôn). +### 14. Mỗi tuần cộng dồn 150–300 phút vận động cường độ vừa, đi bộ nhanh là đủ + +- Chi phí: Không tốn tiền. Mỗi ngày 20–45 phút. Chỗ khó là duy trì lâu dài. +- Nói thẳng: Người mỗi tuần đi bộ nhanh 150–300 phút có xác suất tử vong trong cùng thời gian thấp hơn khoảng ba phần mười so với người không tập. Không đạt mục tiêu thì vẫn thấp hơn hai phần mười. Tập gấp ba đến năm lần mục tiêu thì con số đi xuống đáy, thấp hơn khoảng bốn phần mười. Tăng thêm thì không thấp hơn nữa, nhưng cũng không có hại. +- Lợi ích: Gộp nhiều nghiên cứu theo dõi lại với nhau (tất cả chỉ ghi nhận và theo dõi, không chia nhóm đối chứng): người có mức vận động bằng 1 đến 2 lần mức tối thiểu trong khuyến cáo có rủi ro tử vong thấp hơn người không tập khoảng 31% (HR 0.69). Mức đó là 7.5–15 MET·h mỗi tuần, tương đương khoảng 150–300 phút đi bộ nhanh mỗi tuần. Người không đạt mức tối thiểu vẫn thấp hơn khoảng 20% (HR 0.80). Người đạt từ 3 đến 5 lần mức tối thiểu thấp hơn khoảng 39% (HR 0.61), và đó là chỗ giảm nhiều nhất. Tăng thêm thì không thấp hơn nữa, nhưng cũng không có hại (hơn 10 lần là HR 0.69). Một nghiên cứu gộp khác dùng máy đo gia tốc: nhóm một phần tư hoạt động mạnh nhất theo hoạt động cường độ vừa đến cao có rủi ro tử vong thấp hơn nhóm một phần tư ít hoạt động nhất khoảng 48% (HR 0.52). Máy đo gia tốc là thiết bị đeo lên người để đo trực tiếp mức hoạt động. +- Mức bằng chứng: A +- Nguồn: Arem H và cộng sự (2015). Leisure time physical activity and mortality: a detailed pooled analysis of the dose-response relationship. JAMA Internal Medicine. ; Ekelund U và cộng sự (2019). Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis. BMJ. +- Ghi chú: Mục này và mục 11 (mỗi ngày đi bộ 7000–8000 bước) chỉ cần chọn một. Cần nhớ các nghiên cứu này chỉ ghi nhận và theo dõi. Mấy nghiên cứu đo hoạt động bằng thiết bị thì thời gian theo dõi ngắn, người tham gia phần lớn lớn tuổi, nên trong con số có lẫn thành phần "bị bệnh rồi mới ngừng vận động". Lợi ích thật không lớn bằng tỷ số rủi ro 0.52. + +### 15. Mỗi tuần chơi môn bóng vợt ba lần, mỗi lần 45 phút + +- Chi phí: Phí thuê sân mỗi lần vài chục tệ. Khoảng 2 tiếng mỗi tuần. Chỗ khó là phải tìm đủ người và sân, rồi tuần nào cũng phải đi. +- Nói thẳng: Người thường xuyên chơi quần vợt, cầu lông, bóng bàn có xác suất tử vong trong cùng thời gian bằng khoảng một nửa người không chơi, và tử vong tim mạch thấp hơn khoảng 56%. Người bơi thấp hơn khoảng 28%. Chạy bộ và bóng đá không thấy khác biệt trong bộ dữ liệu này. +- Lợi ích: Một nghiên cứu theo dõi 80.300 người ở Anh so sánh người chơi một môn thể thao với người không chơi môn đó. Người chơi môn bóng vợt có rủi ro tử vong thấp hơn khoảng 47% (tỷ số rủi ro 0.53, 95% CI 0.40–0.69, khoảng tin cậy). Tử vong tim mạch thấp hơn khoảng 56% (0.44, 0.24–0.83). Môn bóng vợt là quần vợt, cầu lông và bóng bàn. Với người bơi thì hai con số là 0.72 và 0.59, thể dục aerobic là 0.73 và 0.64, đạp xe là 0.85. Chạy bộ và bóng đá không thấy khác biệt có ý nghĩa thống kê +- Mức bằng chứng: A +- Nguồn: Oja P, Kelly P, Pedisic Z và cộng sự (2017). Associations of specific types of sports and exercise with all-cause and cardiovascular-disease mortality: a cohort study of 80 306 British adults. British Journal of Sports Medicine, 51(10), 812-817. +- Ghi chú:Tranh cãi. Các nghiên cứu này chỉ ghi nhận và theo dõi. Người chơi thể thao vốn đã khỏe hơn, giao tiếp cũng nhiều hơn. Chỗ chạy bộ không thấy khác biệt cũng cho thấy yếu tố quyết định có khi là "người loại nào chọn môn nào". Đừng vì mục này mà nghĩ chạy bộ vô ích, phần này còn có mục khác tính theo tổng lượng vận động. +### 16. Cộng dồn các đoạn vận động rải rác như leo cầu thang, đi nhanh lên năm phút mỗi ngày + +- Chi phí: Không tốn tiền, cũng không cần dành thêm giờ tập. Chỗ khó là nhớ leo cầu thang, đi nhanh hơn thay vì chọn cách tiện. +- Nói thẳng: Người hoàn toàn không tập thể thao, chỉ cần mỗi ngày có ba đoạn vận động mạnh một hoặc hai phút thì xác suất tử vong trong cùng thời gian vẫn thấp hơn khoảng bốn phần mười so với người không có đoạn nào. Tử vong tim mạch thấp hơn gần một nửa. Vận động mạnh ở đây là loại như leo cầu thang, đi nhanh để kịp giờ. +- Lợi ích: Một nghiên cứu của Ngân hàng sinh học Anh ở 25.200 người không tập thể thao có chủ ý trong sinh hoạt thường ngày. Thời gian theo dõi trung bình là 6.9 năm, trong đó 852 người chết. Có người mỗi ngày có 3 đợt vận động mạnh ngắn, mỗi đợt kéo dài 1–2 phút. So với người không có đoạn nào, rủi ro tử vong và rủi ro tử vong do ung thư của họ thấp hơn 38%–40%. Rủi ro tử vong tim mạch thấp hơn 48%–49%. Người cộng dồn được 4.4 phút mỗi ngày loại hoạt động này có rủi ro tử vong và rủi ro tử vong do ung thư thấp hơn 26%–30%. Rủi ro tử vong tim mạch thấp hơn 32%–34% +- Mức bằng chứng: A +- Nguồn: Stamatakis E, Ahmadi MN, Gill JMR và cộng sự (2022). Association of wearable device-measured vigorous intermittent lifestyle physical activity with mortality. Nature Medicine, 28, 2521-2529. +- Ghi chú:Tranh cãi. Nghiên cứu này đo hoạt động bằng thiết bị đeo, đáng tin hơn là bảo người ta điền bảng hỏi rồi nhớ lại. Nhưng nó vẫn chỉ ghi nhận và theo dõi, không làm thử nghiệm chia nhóm đối chứng, và chỉ theo dõi trong 6.9 năm. Đối tượng là người hoàn toàn không tập, nên mục này không áp dụng cho người đã tập đều +### 17. Mỗi tuần tập sức mạnh 30–60 phút + +- Chi phí: Squat không tạp, chống đẩy không tốn tiền. Tập mỗi tuần 1–2 lần, mỗi lần 20–30 phút. Chỗ khó là sắp xếp được vào lịch trong tuần. +- Nói thẳng: Người tập sức mạnh 30–60 phút mỗi tuần có xác suất tử vong trong cùng thời gian thấp hơn một đến hai phần mười so với người không tập. Tập nhiều hơn không khá hơn. Tập sức mạnh cùng vận động aerobic thì hiệu quả nhất. +- Lợi ích: Gộp nhiều nghiên cứu theo dõi lại với nhau (tất cả chỉ ghi nhận và theo dõi, không chia nhóm đối chứng): người tập sức mạnh có rủi ro tử vong thấp hơn người không tập 10%–17%. Không phải tập càng nhiều càng tốt: mức giảm lớn nhất ở khoảng 30–60 phút mỗi tuần, khoảng 10%–20%, tập thêm ngoài mức đó không thêm lợi ích. Người làm cả sức mạnh lẫn aerobic có rủi ro tử vong thấp hơn người không làm gì. +- Mức bằng chứng: A +- Nguồn: Momma H và cộng sự (2022). Muscle-strengthening activities are associated with lower risk and mortality in major non-communicable diseases: a systematic review and meta-analysis of cohort studies. British Journal of Sports Medicine. +- Ghi chú: Các nghiên cứu này chỉ ghi nhận và theo dõi, mức vận động do chính người tham gia tự báo. Bằng chứng cho thấy tập quá tay thì tệ hơn rất yếu, không có lý do gì phải dè chừng vì điều đó. Tập sức mạnh còn giúp người lớn tuổi tránh ngã và giữ cơ, những điều đó nằm ở phần 1. +### 18. Đừng ngồi liên tục quá lâu, thi thoảng đứng dậy vận động + +- Chi phí: Không tốn tiền, đứng dậy mấy lần cũng không tốn nhiều thời gian. Chỗ khó là ngồi xuống là quên mất thời gian, nên phải đặt nhắc. +- Nói thẳng: Người ngồi lâu nhất có xác suất tử vong trong cùng thời gian bằng 2.6 lần người ngồi ít nhất. Ngồi liền một đoạn dài còn mang thêm một khoản rủi ro riêng. Nhưng chỉ cần mỗi ngày có 60–75 phút hoạt động cường độ vừa là phần rủi ro cộng thêm do ngồi gần như được bù lại. +- Lợi ích: Một nghiên cứu theo dõi tại Mỹ, chỉ ghi nhận và theo dõi, không chia nhóm đối chứng, đo hoạt động bằng máy đo gia tốc. Chia người thành bốn nhóm theo tổng thời gian ngồi mỗi ngày. Nhóm ngồi nhiều nhất có rủi ro tử vong bằng 2.63 lần nhóm ngồi ít nhất (HR 2.63). Sau đó chia thành bốn nhóm theo thời lượng ngồi liền một đoạn: nhóm ngồi liền lâu nhất bằng 1.96 lần nhóm ngồi liền ngắn nhất (HR 1.96). Vậy tổng thời gian ngồi và độ dài của một đoạn ngồi liền, mỗi thứ đều liên quan riêng với rủi ro. Một phân tích gộp khác gồm hơn một triệu người, chỉ ghi nhận và theo dõi, không chia nhóm đối chứng: người ngồi trên 8 tiếng mỗi ngày và gần như không vận động có rủi ro tử vong bằng 1.59 lần nhóm khác (HR 1.59). Nhóm kia ngồi dưới 4 tiếng mỗi ngày và là nhóm hoạt động nhiều nhất. Khoảng 60–75 phút hoạt động cường độ vừa mỗi ngày bù được phần rủi ro cộng thêm do ngồi. Nhóm hoạt động nhiều nhất chỉ có HR 1.04 dù ngồi trên 8 tiếng mỗi ngày, không có ý nghĩa thống kê. +- Mức bằng chứng: A +- Nguồn: Diaz KM và cộng sự (2017). Patterns of Sedentary Behavior and Mortality in U.S. Middle-Aged and Older Adults: A National Cohort Study. Annals of Internal Medicine. ; Ekelund U và cộng sự (2016). Does physical activity attenuate, or even eliminate, the detrimental association of sitting time with mortality? A harmonised meta-analysis of data from more than 1 million men and women. Lancet. +- Ghi chú: Kết luận đáng tin là lúc nào vận động nhiều quan trọng hơn lúc nào ngồi lâu, và khi lượng vận động đủ thì rủi ro do ngồi gần như biến mất. Cần nhớ các nghiên cứu này chỉ ghi nhận và theo dõi, trong nhóm ngồi nhiều nhất có rất nhiều người vốn đã bị bệnh, nên con số 2.63 lần rõ ràng có chứa thành phần "bị bệnh rồi mới ngồi nhiều". Trong cùng bảng gộp đó, rủi ro xem hơn 3 tiếng ti vi mỗi ngày tồn tại ở mọi mức vận động, nhóm hoạt động nhiều nhất chỉ đẩy mốc đó ra tới 5 tiếng. +### 19. Ăn ít thịt chế biến (thịt chua, thịt xông khói, xúc xích, thịt đóng hộp) + +- Chi phí: Không tốn tiền, có khi còn tiết kiệm, cũng không mất thời gian. Chỗ khó là bớt chính miếng mình thích. +- Nói thẳng: Người ăn nhiều thịt chua, thịt xông khói và xúc xích nhất có xác suất tử vong trong cùng thời gian cao hơn khoảng hai phần mười so với người ăn ít nhất. Mỗi phần thịt chế biến ăn thêm mỗi ngày cũng làm con số đó tăng thêm hai phần mười. +- Lợi ích: Gộp nhiều nghiên cứu theo dõi lại với nhau (tất cả chỉ ghi nhận và theo dõi, không chia nhóm đối chứng): nhóm ăn nhiều thịt chế biến nhất có rủi ro tử vong cao hơn nhóm ăn ít nhất khoảng 23% (RR 1.23). Tính chung mọi loại thịt đỏ thì cao hơn khoảng 29% (RR 1.29). Thịt đỏ chưa chế biến cao hơn khoảng 10% (RR 1.10), nhưng mục này không có ý nghĩa thống kê. Một bảng gộp khác tính theo lượng ăn mỗi ngày. Thêm mỗi ngày 1 phần thịt chế biến, tức thêm 1 phần mỗi ngày, làm rủi ro tử vong cao hơn khoảng 23% (RR 1.23). Thêm mỗi ngày 1 phần thịt đỏ, tức thêm 1 phần mỗi ngày, làm cao hơn khoảng 10% (RR 1.10). +- Mức bằng chứng: A +- Nguồn: Larsson SC, Orsini N (2014). Red meat and processed meat consumption and all-cause mortality: a meta-analysis. American Journal of Epidemiology. ; Schwingshackl L và cộng sự (2017). Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. American Journal of Clinical Nutrition. ; Johnston BC và cộng sự (2019). Unprocessed Red Meat and Processed Meat Consumption: Dietary Guideline Recommendations From the NutriRECS Consortium. Annals of Internal Medicine. (hướng dẫn NutriRECS, phe bên kia) +- Ghi chú:Tranh cãi. Hướng dẫn NutriRECS dùng hệ thống chấm bằng chứng thông dụng (GRADE) và đánh giá các nghiên cứu này là "không đáng tin lắm". Nó chỉ đưa ra một khuyến nghị rất yếu: cứ ăn như hiện tại. Hai phe tranh nhau là bằng chứng có đủ chắc hay không, chứ không phải hướng. Không nghiên cứu nào nói thịt chế biến có lợi. Thịt đỏ chưa chế biến ảnh hưởng nhỏ và không thấy khác biệt có ý nghĩa thống kê. Cái cần kiềm là thịt chế biến. + +### 20. Uống ít rượu hoặc không uống + +- Chi phí: Không tốn tiền, còn tiết kiệm tiền, cũng không mất thời gian. Chỗ khó là lúc ở bàn tiệc phải nhận rượu. +- Nói thẳng: Quá 100 g rượu thuần mỗi tuần thì uống càng nhiều chết càng sớm. 100 g rượu thuần khoảng bằng 2.5 L bia. Người 40 tuổi mỗi tuần uống 100–200 g sống ít khoảng nửa năm; người uống 200–350 g sống ít một đến hai năm; người vượt 350 g sống ít bốn đến năm năm. Xét tổng thể sức khoẻ thì lượng đáng dùng nhất là không uống. +- Lợi ích: Gộp 83 nghiên cứu theo dõi lại với nhau, chỉ ghi nhận và theo dõi, không chia nhóm đối chứng, gồm 600.000 người có uống rượu. Rủi ro tử vong thấp nhất ở mức không quá 100 g rượu thuần mỗi tuần. Lượng đó khoảng 2.5 L bia 5% hoặc 300 mL rượu trắng 40 độ. Người 40 tuổi uống 100–200 g mỗi tuần có tuổi thọ ngắn hơn khoảng 6 tháng. Người uống 200–350 g mỗi tuần mất 1–2 năm. Người vượt 350 g mất 4–5 năm. Cả ba mốc đó bản gốc báo cáo theo năm sống. Nghiên cứu Giao tải bệnh toàn cầu GBD 2016 gộp mọi loại thiệt hại sức khoẻ và đặt lượng uống ít rủi ro nhất là 0 ly mỗi tuần. Một bảng gộp khác đã hiệu chỉnh sai lệch trong thiết kế nghiên cứu. Người uống 1.3–24 g mỗi ngày có rủi ro tử vong thấp hơn khoảng 7% so với người không bao giờ uống (RR 0.93), nhưng mục này không có ý nghĩa thống kê. Người uống 45–64 g mỗi ngày cao hơn khoảng 19% (RR 1.19). Người uống trên 65 g mỗi ngày cao hơn khoảng 35% (RR 1.35). +- Mức bằng chứng: A +- Nguồn: Wood AM và cộng sự (2018). Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. ; GBD 2016 Alcohol Collaborators (2018). Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. ; Zhao J và cộng sự (2023). Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Network Open. ; Di Castelnuovo A và cộng sự (2006). Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Archives of Internal Medicine. (phe bên kia) +- Ghi chú:Tranh cãi. Phe cho rằng uống ít có lợi dựa vào một bảng gộp 34 nghiên cứu theo dõi dài hạn. Nó báo rằng người uống ít có tỷ lệ tử vong toàn nguyên thấp hơn tối đa 17%–18%. Lượng mà nó đưa ra là nam không quá 4 ly mỗi ngày, nữ không quá 2 ly. Trong mức đó, người uống nhiều hơn lại có tỷ lệ tử vong thấp hơn. Phe bên kia cho rằng nhóm "không uống" đã lẫn người bỏ rượu vì bệnh và người vốn sức khoẻ đã kém. Có một bảng gộp chỉ vào đúng những nhiễu đó, sau khi trừ đi thì tác dụng bảo vệ biến mất. Nghiên cứu Giao tải bệnh toàn cầu cũng đặt 0 là an toàn nhất. Cách đọc an toàn là: uống ít không rõ là có rút ngắn tuổi thọ hay không, uống nhiều thì chắc chắn có hại. "Bắt đầu uống rượu vì sức khoẻ" không có cơ sở. Muốn uống ít hơn thì xem mục 22 của phần này (đếm số ly trước rồi hỏi bác sĩ). Người uống rượu mỗi ngày không nên tự bỏ dứt khoát, xem mục 21 của phần này (đừng tự bỏ dứt khoát) +### 21. 每天都喝酒、一停就手抖心慌的人，别自己硬戒 + +- Chi phí: Khám khoa tâm thần hoặc khoa nghiện rượu một lần, mấy tệ đến mấy chục tệ. Trường hợp nặng thì phải nằm viện mấy ngày. +- Nói thẳng: Người uống rượu nhiều mỗi ngày mà ngừng đột ngột sẽ có phản ứng cai rượu. Nhẹ thì run tay, ra mồ hôi, hồi hộp, mất ngủ. Nặng thì co giật hoặc lên cơn sảng run, đó là cấp cứu có thể gây chết. Na Uy theo dõi hơn 30.000 người, trong nhóm từng bị sảng run thì mỗi năm về sau có 8% chết. Muốn cai thì đi bệnh viện trước, nói rõ mỗi ngày uống bao nhiêu. +- Lợi ích: Một tổng quan lâm sàng năm 2026 ghi rõ: người uống rượu nhiều lâu ngày, sau khi ngừng đột ngột hoặc giảm mạnh, sẽ có hội chứng cai rượu, với các biểu hiện gồm run, kích thích thần kinh thực vật, lo âu, mất ngủ, rối loạn cảm giác, cơn động kinh và sảng run, trong chẩn đoán nội khoa thông thường là「hay gặp và có thể đe dọa tính mạng」. Một nghiên cứu mẫu theo thời gian dựa trên sổ đăng ký toàn quốc của Na Uy lấy 36287 người được chẩn đoán nghiện rượu, trạng thái cai rượu hoặc sảng run trong giai đoạn 2009 đến 2015: nhóm sảng run có tỷ lệ tử vong hằng năm 8.0%, nhóm trạng thái cai rượu 5.0%, nhóm nghiện rượu 3.6%. Tỷ số tử vong chuẩn hoá của nhóm sảng run là 9.8 (95% CI 8.9–10.7, đây là khoảng tin cậy), nghĩa là rủi ro tử vong cao khoảng 9.8 lần người cùng tuổi nói chung. +- Mức bằng chứng: B +- Nguồn: Caputo F, Lungaro L, Costanzini A, De Giorgio R, Addolorato G (2026). Alcohol withdrawal syndrome in hospitalized patients: a practical review. European Journal of Internal Medicine, 107103. ；Bramness JG, Heiberg IH, Høye A, Rossow I (2023). Mortality and alcohol-related morbidity in patients with delirium tremens, alcohol withdrawal state or alcohol dependence in Norway: A register-based prospective cohort study. Addiction, 118(12), 2352–2359. +- Ghi chú: Tự đánh giá thế nào: gần như ngày nào cũng uống, một ngày không uống thì run tay, ra mồ hôi, hồi hộp, mất ngủ, sáng dậy phải uống một chút mới dễ chịu, thì chỉ cần trúng một dấu hiệu trong số đó là đừng tự ngừng đột ngột. Bệnh viện thường dùng thuốc nhóm benzodiazepin để người bệnh chịu qua mấy ngày đó, đồng thời bổ sung thiamine (vitamin B1). Để ở mức B là vì nghiên cứu mẫu của Na Uy nói về「người từng bị sảng run sau đó thế nào」, không phải「tự cai cứng thì ra sao」, giữa hai câu hỏi đó còn là suy luận. Rượu vẫn phải cai, chỉ là cai ở bệnh viện. Mỗi tuần uống bao nhiêu thì tính là nhiều, xem mục 20 của phần này (uống ít hoặc không uống). Muốn uống ít hơn thì làm sao, xem mục 22 của phần này (đếm số ly trước rồi hỏi bác sĩ) + +### 22. 想少喝酒，先把一周喝了多少数出来，再找医生聊几分钟 + +- Chi phí: Không tốn tiền. Mỗi ngày ghi số ly vào sổ là xong. Hỏi thêm một câu ở phòng khám cũng không tốn thêm tiền. +- Nói thẳng: Bác sĩ dành mấy phút để xem lượng rượu bạn uống, nói rõ rủi ro, giúp bạn đặt một mục tiêu. Sau một năm, trung bình mỗi tuần uống ít đi 20 g rượu thuần, tương đương khoảng nửa lít bia. Nói chuyện lâu hơn không có thêm lợi ích. Người đã không ngừng được thì ở nước ngoài có hai loại thuốc đã được chứng minh có tác dụng, cứ điều trị khoảng 12 người thì có thêm 1 người không quay lại kiểu uống một lần là nhiều. +- Lợi ích: Gộp 34 thử nghiệm chia nhóm ngẫu nhiên, 15197 người, lại với nhau: người được can thiệp ngắn sau một năm uống ít đi mỗi tuần 20 g rượu thuần so với người chỉ được can thiệp tối thiểu hoặc không được can thiệp (95% CI ít hơn 12 đến 28 g, đây là khoảng tin cậy; bằng chứng chất lượng trung bình). Lượng rượu nền của người tham gia trong các nghiên cứu này trung bình 244 g mỗi tuần. Can thiệp ngắn nghĩa là lời khuyên hoặc tư vấn lối sống không quá 5 lần, tổng thời gian dưới 60 phút; nói chuyện lâu hơn không cho thêm lợi ích nào. Về thuốc: gộp 122 thử nghiệm chia nhóm ngẫu nhiên, 22803 người, acamprosat mỗi 12 người được điều trị (95% CI 8 đến 26) thì có thêm 1 người không quay lại uống bất kỳ lượng nào; naltrexon uống mỗi ngày 50 mg, mỗi 12 người được điều trị (8 đến 26) thì có thêm 1 người không quay lại mức uống nặng. +- Mức bằng chứng: A +- Nguồn: Kaner EF, Beyer FR, Muirhead C, et al. (2018). Effectiveness of brief alcohol interventions in primary care populations. Cochrane Database of Systematic Reviews, 2, CD004148. ；Jonas DE, Amick HR, Feltner C, et al. (2014). Pharmacotherapy for adults with alcohol use disorders in outpatient settings: a systematic review and meta-analysis. JAMA, 311(18), 1889–1900. +- Ghi chú: Mức lợi ích ghi「nhỏ」vì cả hai nguồn bằng chứng đo lượng uống, không đo tỷ lệ tử vong, và theo quy ước của cuốn sách này, chỉ có điểm thay thế thì ghi「nhỏ」. Phần thuốc đa số thử nghiệm làm ở nước ngoài, phần lớn người tham gia đã đạt tiêu chuẩn chẩn đoán nghiện rượu, và mọi thử nghiệm đều kèm can thiệp tâm lý xã hội, nên con số là「phần lợi ích cộng thêm ngoài cái nền đó」. Về chỉ định và khả năng kê đơn của hai loại thuốc này ở Trung Quốc, cứ theo bác sĩ và tờ hướng dẫn sử dụng thuốc, đừng tự mua qua mạng. Đã xuất hiện phản ứng cai rượu thì xem mục 21 của phần này (đừng tự cai cứng). + +### 23. 每天吃一小把坚果 + +- Chi phí: Mỗi ngày ăn 28 g, một năm khoảng hai ba trăm tệ. Ăn tiện tay, không mất thời gian. +- Nói thẳng: Người ăn hạt thường xuyên có xác suất tử vong trong cùng thời gian thấp hơn người không ăn khoảng hai phần mười. Ăn một lần mỗi tuần đã thấy khác biệt, thấp hơn khoảng một phần mười. Ăn đến mức mỗi ngày một nắm thì khác biệt rõ nhất. +- Lợi ích: Hai nghiên cứu theo dõi ở Mỹ, 119.000 người, hơn 300 vạn năm người. Năm người là số người nhân với số năm theo dõi của từng người. So với người không ăn hạt: người ăn dưới 1 lần mỗi tuần có rủi ro tử vong thấp hơn khoảng 7% (tỷ số rủi ro 0.93, 95% CI 0.90–0.96, đây là khoảng tin cậy). Người ăn 1 lần mỗi tuần thấp hơn khoảng 11% (0.89, 0.86–0.93). Người ăn 2–4 lần mỗi tuần thấp hơn khoảng 13% (0.87, 0.83–0.90). Người ăn 5–6 lần mỗi tuần thấp hơn khoảng 15% (0.85, 0.79–0.91). Người ăn trên 7 lần mỗi tuần thấp hơn khoảng 20% (0.80, 0.73–0.86) +- Mức bằng chứng: A +- Nguồn: Bao Y, Han J, Hu FB, et al. (2013). Association of nut consumption with total and cause-specific mortality. New England Journal of Medicine, 369(21), 2001-2011. +- Ghi chú: Mua loại nguyên vị, không thêm muối. Hạt có năng lượng không thấp, đừng ăn như đồ ăn vặt không kiểm soát. Cần nhớ các nghiên cứu loại này chỉ theo dõi dài hạn một nhóm người, người ăn hạt có lối sống nói chung tốt hơn, nên con số phải giảm bớt đi + +### 24. Đổi một phần thịt đỏ lấy cá và thịt gia cầm + +- Chi phí: Không tốn tiền, cũng không mất thời gian. Đây là đổi loại thịt, không phải mua thêm. Chỗ khó là phải đổi một chút khẩu vị. +- Nói thẳng: Người ăn thêm hai phần thịt chế biến hoặc thịt đỏ mỗi tuần có xác suất tử vong trong cùng thời gian cao hơn khoảng 3%. Đổi bằng cùng số phần cá hoặc thịt gia cầm thì không thấy tăng. Hiệu ứng của mục này nhỏ trong phần này, thuộc loại tiện tay là đổi, không phải loại phải gắng sức. +- Lợi ích: Gộp 6 nghiên cứu theo dõi ở Mỹ, gần 30.000 người, lại với nhau. Ăn thêm hai phần thịt chế biến mỗi tuần thì rủi ro tử vong cao hơn khoảng 3% (tỷ số rủi ro 1.03, 95% CI 1.02–1.05, khoảng tin cậy). Ăn thêm hai phần thịt đỏ chưa chế biến mỗi tuần cũng cao hơn khoảng 3% (1.03, 1.01–1.05). Ăn theo cách đó nhưng thay bằng thịt gia cầm là 0.99 (0.97–1.02); thay bằng cá là 0.99 (0.97–1.01). Cả hai không thấy khác biệt có ý nghĩa thống kê +- Mức bằng chứng: A +- Nguồn: Zhong VW, Van Horn L, Greenland P, et al. (2020). Associations of Processed Meat, Unprocessed Red Meat, Poultry, or Fish Intake With Incident Cardiovascular Disease and All-Cause Mortality. JAMA Internal Medicine, 180(4), 503-512. +- Ghi chú: Hai phần mỗi tuần tạo ra khác biệt rất nhỏ, đừng trông chờ đổi thịt rồi sống thêm mấy năm. Cùng số tiền bỏ ra để ăn thịt thì có thể ưu tiên cá và thịt gia cầm + + +### 25. 把一部分精米白面换成全谷物 + +- Chi phí: Gạo lức, yến mạch, bánh mì nguyên cốc đắt hơn gạo trắng và bột mì tinh bột một chút, không mất thời gian. Chỗ khó là vị phải từ từ quen. +- Nói thẳng: Người ăn thêm 90 g ngũ cốc nguyên mỗi ngày có xác suất tử vong trong cùng thời gian thấp hơn khoảng 17%. 90 g xấp xỉ ba phần. Ăn đến khoảng 200 g mỗi ngày thì con số vẫn đang giảm tiếp. +- Lợi ích: Gộp nhiều nghiên cứu theo dõi lại với nhau (chỉ ghi nhận, không chia nhóm đối chứng): ăn thêm 90 g ngũ cốc nguyên mỗi ngày thì rủi ro tử vong thấp hơn khoảng 17% (RR 0.83). 90 g xấp xỉ 3 phần. Lợi ích này kéo dài tới mức 210–225 g mỗi ngày. Trong một bảng gộp khác, ăn thêm 1 phần mỗi ngày thì thấp hơn khoảng 8% (RR 0.92). +- Mức bằng chứng: A +- Nguồn: Aune D và cộng sự (2016). Whole grain consumption and risk of cardiovascular disease, cancer, and all cause and cause specific mortality: systematic review and dose-response meta-analysis of prospective studies. BMJ. ；Schwingshackl L và cộng sự (2017). Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. American Journal of Clinical Nutrition. +- Ghi chú: Không cần đổi hết sang ngũ cốc nguyên, đổi một nửa là đã nằm trong đoạn mà lợi ích thấy rõ nhất. Cần nhớ các nghiên cứu loại này chỉ ghi nhận và theo dõi, người ăn ngũ cốc nguyên vốn đã khoẻ hơn, nên con số có xu hướng lớn hơn thực tế. Kết quả giữa các nghiên cứu cũng chênh nhau khá xa (I² 83%, số này càng lớn thì kết luận của các nghiên cứu càng không nhất quán). + +### 26. Mỗi tuần uống trà từ ba lần trở lên + +- Chi phí: Một năm mấy chục đến mấy trăm tệ. Pha một cốc mất mấy phút, không tốn nhiều thời gian. +- Nói thẳng: Người có thói quen uống trà có xác suất tử vong trong cùng thời gian thấp hơn khoảng 15%. Tính từ tuổi 50, sống thêm trung bình 1.26 năm, và có thêm 1.41 năm không mắc bệnh tim mạch. +- Lợi ích: Một nghiên cứu ở Trung Quốc gồm 100.900 người (China-PAR), trong đó một nửa được theo dõi đủ 7.3 năm. Người có thói quen uống trà có rủi ro tử vong thấp hơn khoảng 15% (tỷ số rủi ro 0.85, 95% CI 0.79–0.90, đây là khoảng tin cậy). Tính từ tuổi 50: người có thói quen uống trà có thêm 1.41 năm không mắc bệnh tim mạch do xơ vữa động mạch, và tuổi thọ kỳ vọng dài hơn 1.26 năm +- Mức bằng chứng: A +- Nguồn: Wang X, Liu F, Li J, et al. (2020). Tea consumption and the risk of atherosclerotic cardiovascular disease and all-cause mortality: The China-PAR project. European Journal of Preventive Cardiology, 27(18), 1956-1963. ；độc tố nấm men trong trà: Cui P và cộng sự (2020). Quantitative analysis and dietary risk assessment of aflatoxins in Chinese post-fermented dark tea. Food and Chemical Toxicology. ；Zhou H và cộng sự (2022). Mycotoxins in Tea (Camellia sinensis (L.) Kuntze): Contamination and Dietary Exposure Profiling in the Chinese Population. Toxins. ；chiết xuất trà xanh và tổn thương gan: Hu J và cộng sự (2018). The safety of green tea and green tea extract consumption in adults - Results of a systematic review. Regulatory Toxicology and Pharmacology. +- Ghi chú:Tranh cãi. Các nghiên cứu này chỉ ghi nhận và theo dõi, và trong số nam giới Trung Quốc uống trà thì tỷ lệ hút thuốc cùng uống rượu cao. Tác giả đã trừ bớt một phần ảnh hưởng đó, nhưng trừ không sạch. Thêm một điểm nữa, đừng uống thứ nóng sôi, nhiệt độ của đồ uống nóng xem mục trong phần này về nhiệt độ đồ uống nóng. Có người lo trà chứa độc tố nấm men. Ở Trung Quốc đã kiểm tra 158 mẫu trà đen, có 2 mẫu phát hiện aflatoxin, tính theo lượng trà người ta uống thì lượng nạp vào thấp hơn giá trị có thể chấp nhận ở mức quốc tế. Một nghiên cứu khác kiểm tra 352 mẫu trà với 16 loại độc tố nấm men, chỉ có ochratoxin A trong trà đen là vượt giới hạn về trung bình. Tính theo lượng trà dân nước ta uống thì cả hai nghiên cứu đều không cho ra rủi ro khẩu phần. Giữ trà ở nơi khô, trà ẩm mốc thì đừng uống. Viên nang chiết xuất trà xanh là chuyện khác: nuốt một lượng lớn catechin nồng độ cao trong một lần thì có thể gây hại gan, trà pha không có vấn đề đó. Vậy mục này chỉ nói về trà pha, đừng lấy thực phẩm bổ sung dạng chiết xuất mà thay. + +### 27. Mỗi ngày uống ba đến bốn cốc cà phê, không đường không kem + +- Chi phí: Tự pha thì mỗi ngày mất một hai tệ. Pha một cốc mất mấy phút, không tốn nhiều thời gian. +- Nói thẳng: Người mỗi ngày uống ba đến bốn cốc cà phê có xác suất tử vong trong cùng thời gian thấp hơn người không uống khoảng 17%. +- Lợi ích: Một bài tổng quan gộp kết quả của 201 nghiên cứu tổng hợp, và các nghiên cứu gốc bên dưới những bảng tổng hợp đó đều chỉ ghi nhận, không chia nhóm đối chứng. Kết quả: người uống 3–4 cốc cà phê mỗi ngày có rủi ro tử vong thấp hơn 17% so với người không uống (tỷ số rủi ro tương đối 0.83, 95% CI 0.79–0.88, đây là khoảng tin cậy) +- Mức bằng chứng: A +- Nguồn: Poole R, Kennedy OJ, Roderick P, et al. (2017). Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ, 359, j5024. +- Ghi chú:Tranh cãi. Tác giả ghi rõ bằng chứng gần như toàn bộ đến từ nghiên cứu theo dõi, và còn viết「cần thử nghiệm đối chứng ngẫu nhiên đáng tin cậy mới đánh giá được đó có phải quan hệ nhân quả hay không」. Nghĩa là phải chia ngẫu nhiên người tham gia thành hai nhóm để so sánh, chỉ nhìn nghiên cứu theo dõi thì không kết luận được nhân quả. Khi làm thì uống cà phê đen, thêm đường và kem sữa sẽ triệt tiêu lợi ích. Phụ nữ có thai, người bị loạn nhịp tim, lo âu, mất ngủ thì mục này không áp dụng + +### 28. Mỗi ngày ăn đủ 5 phần (khoảng 400 g) rau quả + +- Chi phí: Mỗi ngày mấy tệ đến mấy chục tệ. Mỗi ngày phải dành chút thời gian rửa và thái. Chỗ khó là phải mua và ăn đủ lượng mỗi ngày. +- Nói thẳng: Người ăn thêm 200 g rau quả mỗi ngày có xác suất tử vong trong cùng thời gian thấp hơn khoảng một phần mười. Người ăn đủ 5 phần mỗi ngày thấp hơn người chỉ ăn 2 phần khoảng 13%. Ăn nhiều hơn nữa thì không thấp hơn nữa. +- Lợi ích: Gộp nhiều nghiên cứu theo dõi lại với nhau (chỉ ghi nhận, không chia nhóm đối chứng): ăn thêm 200 g rau quả mỗi ngày thì rủi ro tử vong thấp hơn khoảng 10% (RR 0.90). Lợi ích này kéo dài tới mức 800 g mỗi ngày. Tổng hợp của hai nghiên cứu mẫu lớn ở Mỹ cộng với 26 nghiên cứu: người ăn đủ 5 phần mỗi ngày có rủi ro tử vong thấp hơn người chỉ ăn 2 phần khoảng 13% (HR 0.87). Cách phối hợp tốt nhất là 2 phần trái cây cộng 3 phần rau, thêm nữa không còn giảm. +- Mức bằng chứng: A +- Nguồn: Aune D và cộng sự (2017). Fruit and vegetable intake and the risk of cardiovascular disease, total cancer and all-cause mortality: a systematic review and dose-response meta-analysis of prospective studies. International Journal of Epidemiology. ；Wang DD và cộng sự (2021). Fruit and Vegetable Intake and Mortality: Results From 2 Prospective Cohort Studies of US Men and Women and a Meta-Analysis of 26 Cohort Studies. Circulation. +- Ghi chú: Ăn đủ 5 phần là xong, không cần lên tới 10 phần, cả hai nghiên cứu trả lời giống nhau ở điểm này. Cần nhớ các nghiên cứu loại này chỉ ghi nhận và theo dõi, nhiễu rất rõ: người ăn nhiều rau khác người khác cả về thu nhập, giáo dục, mức vận động. Vậy nên「rủi ro thấp hơn khoảng 10%」đã là trần. + +### 29. Ăn ít thực phẩm siêu chế biến (khoai tây chiên, mì gói, bánh ngọt, đồ ăn nhanh) + +- Chi phí: Không tốn tiền, chỉ là đổi chỗ mua. Tự nấu ăn, hoặc bỏ thời gian chọn nguyên liệu ít chế biến, ngày nào cũng phải tốn chút thời gian. Chỗ khó là phải chống lại thói quen「cứ tiện là được」trong thời gian dài. +- Nói thẳng: Người ăn nhiều thực phẩm siêu chế biến nhất có xác suất tử vong trong cùng thời gian cao hơn người ăn ít nhất khoảng hai phần mười, tử vong tim mạch cao hơn năm phần mười. Tuy nhiên loại bằng chứng này được đánh giá là không chắc chắn lắm, phải giảm bớt con số khi đọc. +- Lợi ích: Một bài tổng quan gộp nhiều nghiên cứu tổng hợp lại với nhau, và các nghiên cứu gốc bên dưới những bảng tổng hợp đó đều chỉ ghi nhận, không chia nhóm đối chứng. Kết quả: người ăn nhiều thực phẩm siêu chế biến có rủi ro tử vong cao hơn người ăn ít khoảng 21% (RR 1.21), tử vong do bệnh tim mạch cao hơn khoảng 50% (RR 1.50). Hai kết quả này được xếp mức bằng chứng lần lượt là「gợi ý mạnh」và「đáng tin cậy」. Nhưng theo cách chấm điểm bằng chứng thông dụng (GRADE), độ chắc chắn là thấp hoặc rất thấp. +- Mức bằng chứng: A +- Nguồn: Lane MM và cộng sự (2024). Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. +- Ghi chú:Tranh cãi nằm ở ba điểm. Một là bộ phân loại này (NOVA) gom chung một nhóm các thực phẩm khác nhau khá xa về dinh dưỡng. Hai là nó trùng lặp rất nhiều với đồ uống có đường và thịt chế biến (mục 7 và 19), không phân biệt được chính xác yếu tố nào đang tác động. Ba là theo cách chấm điểm bằng chứng thông dụng (GRADE) thì mức độ cũng thấp. Vậy sau khi làm cho xong hai mục kia, mục này còn thêm được bao nhiêu lợi ích thì không chắc. Bằng chứng ở phe bên kia chính là mức đánh giá thấp mà cùng bài tổng quan đó đưa ra, hiện chưa có nghiên cứu gốc nào ra kết luận ngược lại. Vậy hãy làm cho xong mục 7 và 19 (không uống đồ uống có đường, ăn ít thịt chế biến) rồi hẵng xem mục này. + +### 30. Nấu ăn và sưởi ấm không đốt than củi, chuyển sang điện hoặc khí đốt + +- Chi phí: Hộ gia đình nông thôn phải thay bếp, cộng thêm tiền nhiên liệu, một năm mấy trăm đến hơn một nghìn tệ. Thay một lần là xong, không chiếm thời gian dài hạn. Người ở thành phố vốn đã dùng điện hoặc khí đốt. +- Nói thẳng: Người đốt than hoặc củi để nấu ăn có xác suất tử vong trong cùng thời gian cao hơn khoảng một phần mười; người đốt than củi để sưởi cao hơn khoảng 14%. Người đã đổi sang điện hoặc khí đốt có xác suất tử vong thấp hơn 13%–33% so với người vẫn đang đốt than củi. +- Lợi ích: Trung Quốc có một nghiên cứu theo dõi dài hạn bệnh mạn tính, chỉ ghi nhận, không chia nhóm đối chứng, gồm 271.000 người lớn không mắc bệnh tim mạch. Người nấu ăn bằng nhiên liệu rắn như than củi có rủi ro tử vong cao hơn người dùng điện hoặc khí đốt khoảng 11% (HR 1.11). Người sưởi bằng nhiên liệu rắn cao hơn khoảng 14% (HR 1.14). Người đã đổi từ nhiên liệu rắn sang nhiên liệu sạch, so với người vẫn đang đốt, thì phần nấu ăn thấp hơn khoảng 13% (HR 0.87). Phần sưởi ấm thấp hơn khoảng 33% (HR 0.67). PM2.5 ngoài trời là một khoản riêng. Gộp 104 nghiên cứu theo dõi lại với nhau: mỗi 10 µg/m³ nồng độ hít phải lâu dài cao hơn thì rủi ro tử vong do nguyên nhân tự nhiên cao hơn khoảng 8% (RR 1.08). Nguyên nhân tự nhiên là chết vì bệnh, không phải vì tai nạo. +- Mức bằng chứng: A +- Nguồn: Yu K và cộng sự (2018). Association of Solid Fuel Use With Risk of Cardiovascular and All-Cause Mortality in Rural China. JAMA. ；Chen J, Hoek G (2020). Long-term exposure to PM and all-cause and cause-specific mortality: A systematic review and meta-analysis. Environment International. +- Ghi chú: Loại nghiên cứu này chỉ ghi nhận và theo dõi. Hộ đổi nhiên liệu thường cũng giàu hơn, nên con số「rủi ro rơi còn 0.67 lần」có lẫn ảnh hưởng của hoàn cảnh gia đình. Còn với PM2.5 ngoài trời thì cá nhân làm được không nhiều, chỉ là chuyển nhà, đeo khẩu trang, dùng máy lọc không khí. Máy lọc không khí chưa có nghiên cứu nào lấy tỷ lệ tử vong làm kết quả cuối, nên phần này không đưa ra con số. + + +### 31. Đồ nóng để nguội rồi hãy uống, không uống trà, canh hay cà phê sôi + +- Chi phí: Không tốn tiền, mỗi lần chỉ phải chờ thêm hai ba phút. Chỗ khó là nhịn không uống lúc thèm. +- Nói thẳng: Uống trà rất nóng thì xác suất mắc ung thư thực quản gấp 8 lần so với uống trà ấm. Ngay cả khi chỉ「nóng」thì cũng gấp 2 lần. Rót ra mà dưới hai phút là uống thì rủi ro gấp 5 lần so với chờ trên bốn phút rồi uống. +- Lợi ích: Ở vùng có nhiều ung thư thực quản thuộc miền bắc Iran, lấy 300 bệnh nhân so với 571 người không mắc bệnh. Dưới đây đều so với người uống trà ấm. Người uống trà「nóng」 có cơ hội mắc ung thư bạch sục thực quản cao hơn khoảng 2.07 lần (tỷ số khả năng, 95% CI 1.28–3.35, khoảng tin cậy). Người uống trà「rất nóng」 cao hơn khoảng 8.16 lần (3.93–16.9). Người rót ra dưới 2 phút là uống cao hơn khoảng 5.41 lần so với người chờ trên 4 phút rồi uống (2.63–11.1) +- Mức bằng chứng: A +- Nguồn: Islami F, Pourshams A, Nasrollahzadeh D, et al. (2009). Tea drinking habits and oesophageal cancer in a high risk area in northern Iran: population based case-control study. BMJ, 338, b929. ；Loomis D, Guyton KZ, Grosse Y, et al. (2016). Carcinogenicity of drinking coffee, mate, and very hot beverages. Lancet Oncology, 17(7), 877-878. +- Ghi chú: Cơ quan Quốc tế Nghiên cứu Ung thư xếp đồ uống nóng trên 65 ℃ vào nhóm 2A, nghĩa là rất có thể gây ung thư. Cùng bản đánh giá đó kết luận bản thân cà phê không gây ung thư. Vùng Triều Châu – Sơn Đông và vùng Thái Hàng Sơn của Trung Quốc có nhiều ung thư thực quản, liên quan tới thói quen uống khi còn nóng + +### 32. Ban ngày ra ngoài hẵn nắng, đừng cả ngày không thấy ánh sáng + +- Chi phí: Không tốn tiền. Mỗi ngàm mười mấy phút, đi làm về và nghỉ trưa là đủ. +- Nói thẳng: Phụ nữ cố tránh nắng có tuổi thọ kỳ vọng thấp hơn người phơi nắng nhiều nhất 0.6 đến 2.1 năm. Người không hút thuốc nhưng tránh nắng thì sống lâu gần bằng người hút thuốc mà phơi nắng nhiều nhất. +- Lợi ích: Một nghiên cứu ở Thụy Điển, 29.500 phụ nữ, theo dõi 20 năm. So với nhóm phơi nắng nhiều nhất, người cố tránh nắng có tuổi thọ kỳ vọng thấp hơn 0.6 đến 2.1 năm. Tác giả viết「người không hút thuốc nhưng tránh nắng có tuổi thọ kỳ vọng tương đương người hút thuốc trong nhóm phơi nắng nhiều nhất」 +- Mức bằng chứng: B +- Nguồn: Lindqvist PG, Epstein E, Nielsen K, et al. (2016). Avoidance of sun exposure as a risk factor for major causes of death: a competing risk analysis of the Melanoma in Southern Sweden cohort. Journal of Internal Medicine, 280(4), 375-387. +- Ghi chú:Tranh cãi. Thụy Điển có vĩ độ cao, ít nắng, kết luận không chuyển thẳng sang Trung Quốc được. Người tránh nắng cũng có thể vốn đã yếu, không thích ra ngoài. Thêm nữa, đừng phơi đến mức cháy da, dưới ánh nắng gay gắt giữa trưa vẫn phải che, vì phơi nắng làm tăng nguy cơ ung thư da + +### 33. Giữ BMI trong khoảng 20–25, thừa cân thì giảm + +- Chi phí: Không tốn tiền. Ăn uống cộng vận động thì ngày nào cũng phải tốn thời gian. Đây là mục tốn ý chí nhất của cả phần: phải kiềm chế ăn uống lâu dài, và giữ nguyên sau khi giảm còn khó hơn lúc giảm. +- Nói thẳng: Xác suất tử vong thấp nhất ở đoạn BMI 20–25. BMI lên 27.5–30 thì cao hơn khoảng hai phần mười, 30–35 cao hơn khoảng bốn phần rưỡi, 35–40 gần gấp đôi. Người Đông Á nhạy với thừa cân hơn, mỗi 5 BMI tăng thì rủi ro cao hơn khoảng bốn phần mười. +- Lợi ích: Gộp dữ liệu cá nhân của 239 nghiên cứu theo dõi về sau lại với nhau (chỉ ghi nhận, không chia nhóm đối chứng). Phân tích này chỉ giữ lại ba nhóm người: không bao giờ hút thuốc, không mắc bệnh mạn tính lúc bắt đầu nghiên cứu, và còn sống thêm được trên 5 năm sau khi bắt đầu. Kết quả: rủi ro tử vong thấp nhất ở BMI 20–25. BMI 25–27.5 cao hơn khoảng 7% (HR 1.07). 27.5–30 cao hơn khoảng 20% (HR 1.20). 30–35 cao hơn khoảng 45% (HR 1.45). 35–40 cao hơn khoảng 94% (HR 1.94). 40–60 là 2.76 lần (HR 2.76). Ở người có BMI trên 25, mỗi 5 kg/m² tăng thì ở dân Đông Á rủi ro cao hơn khoảng 39% (HR 1.39). +- Mức bằng chứng: A +- Nguồn: Global BMI Mortality Collaboration (2016). Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet. ；Flegal KM và cộng sự (2013). Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA. （phe bên kia） +- Ghi chú:Tranh cãi. Có phe cho rằng「béo một chút lại sống lâu hơn」. Tổng hợp của phe đó cho ra: người thừa cân có rủi ro tử vong thấp hơn khoảng 6%, người béo phì nhẹ thấp hơn khoảng 5%. Hai phe chủ yếu khác nhau ở chỗ có nên loại bỏ ba nhóm dữ liệu: người hút thuốc, người đã mắc bệnh, và dữ liệu trong vài năm đầu sau khi bắt đầu nghiên cứu. Loại bỏ là vì người bệnh nặng sẽ sụt cân trước. Phân tích gộp 239 nghiên cứu nói trên, sau khi loại bỏ ba nhóm đó, rủi ro do thừa cân lại xuất hiện. Cần thêm một điểm nữa: không có thử nghiệm chia nhóm ngẫu nhiên nào chứng minh việc giảm cân bản thân nó làm giảm tỷ lệ tử vong toàn nguyên. Lợi ích ở trên là do so sánh ngang giữa những người béo khác nhau, không có nghĩa giảm đi sẽ giảm được tới mức đó. Người muốn giảm cân không cần bỏ công sắp xếp bữa ăn, bữa sáng và chế độ nhịn ăn 16:8 đều không có lợi ích thêm, xem mục 26 ở phần 6 (đừng trông chờ bữa sáng hoặc 16:8 giúp kiểm soát cân nặng) +### 34. Mỗi tuần ăn ớt từ bốn lần trở lên + +- Chi phí: Không tốn tiền, cũng không mất thời gian. +- Nói thẳng: Người ăn ớt từ bốn lần mỗi tuần trở lên có xác suất tử vong trong cùng thời gian thấp hơn người ăn rất ít khoảng 23%, tử vong tim mạch thấp hơn khoảng một phần ba. +- Lợi ích: Một nghiên cứu ở Ý gồm 22.800 người, một nửa được theo dõi đủ 8.2 năm, có 1236 người chết. Người ăn ớt trên 4 lần mỗi tuần có rủi ro tử vong thấp hơn người ăn rất ít khoảng 23% (tỷ số rủi ro 0.77, 95% CI 0.66–0.90, khoảng tin cậy). Tử vong tim mạch thấp hơn khoảng 34% (0.66, 0.50–0.86) +- Mức bằng chứng: B +- Nguồn: Bonaccio M, Di Castelnuovo A, Costanzo S, et al. (2019). Chili Pepper Consumption and Mortality in Italian Adults. Journal of the American College of Cardiology, 74(25), 3139-3149. +- Ghi chú:Tranh cãi. Đây là nghiên cứu theo dõi ghi nhận, làm trong bối cảnh khẩu ăn Địa Trung Hải, người ăn cay có thể cả ăn uống nói chung sát hơn với cách ăn truyền thống địa phương. Thêm nữa, người bị trào ngược dạ dày lên thực quản, bị trĩ hay hội chứng ruột kích thích thì ăn cay sẽ nặng thêm triệu chứng, không cần cố + +### 35. Mỗi ngày uống một hai phần sữa hoặc sữa chua + +- Chi phí: Mỗi ngày mấy tệ, không mất thời gian. +- Nói thẳng: Người uống từ hai phần sữa hoặc sữa chua mỗi ngày trở lên có xác suất tử vong trong cùng thời gian thấp hơn người không uống khoảng 17%. +- Lợi ích: Nghiên cứu theo dõi PURE gồm 21 nước, 136.000 người, theo dõi trung bình 9.1 năm, có 6796 người chết. Người ăn từ hai phần sản phẩm sữa mỗi ngày trở lên có rủi ro tử vong thấp hơn người không ăn khoảng 17% (tỷ số rủi ro 0.83, 95% CI 0.72–0.96, khoảng tin cậy), xu hướng P=0.0052 +- Mức bằng chứng: B +- Nguồn: Dehghan M, Mente A, Rangarajan S, et al. (2018). Association of dairy intake with cardiovascular disease and mortality in 21 countries from five continents (PURE): a prospective cohort study. Lancet, 392(10161), 2288-2297. +- Ghi chú:Tranh cãi. PURE đưa vào rất nhiều nước thu nhập trung bình và thấp, ở đó mà uống sữa được mỗi ngày thì bản thân việc đó đã nói lên hoàn cảnh gia đình khá, tầng nhiễu này khó loại sạch. Ngoài ra người không dung nạp được đường sữa thì chuyển sang uống sữa chua hoặc sữa ít đường sữa, không cần cố uống sữa tươi + +### 36. Không cần kiêng trứng, nhưng đừng ăn ba bốn quả mỗi ngày + +- Chi phí: Không tốn tiền, cũng không mất thời gian. +- Nói thẳng: Người ăn thêm nửa quả trứng nguyên mỗi ngày có xác suất tử vong trong cùng thời gian cao hơn khoảng 7%. Đổi nửa quả đó ra lòng trắng trứng, cá, thịt gia cầm hoặc hạt thì rủi ro tử vong của từng nhóm lại thấp hơn. Không cần kiêng trứng, chỉ là đừng ăn ba bốn quả mỗi ngày. +- Lợi ích: Một nghiên cứu theo dõi ở Mỹ gồm 521.000 người, trong đó 129.000 người chết. Người ăn thêm nửa quả trứng nguyên mỗi ngày có rủi ro tử vong cao hơn khoảng 7% (tỷ số rủi ro 1.07, 95% CI 1.06–1.08, đây là khoảng tin cậy). Đổi nửa quả trứng nguyên đó ra lượng tương đương lòng trắng trứng, thịt gia cầm, cá, sản phẩm sữa hoặc hạt đậu thì rủi ro tử vong thấp hơn, và rủi ro tử vong do tim mạch, ung thư, hô hấp cũng thấp hơn ở cả ba +- Mức bằng chứng: B +- Nguồn: Zhuang P, Wu F, Mao L, et al. (2021). Egg and cholesterol consumption and mortality from cardiovascular and different causes in the United States: A population-based cohort study. PLoS Medicine, 18(2), e1003508. +- Ghi chú:Tranh cãi khá lớn: có những nghiên cứu theo dõi dài hạn quần số lớn, và cả những bảng gộp nhiều nghiên cứu, không thấy trứng có liên quan gì tới tỷ lệ tử vong. Hướng dẫn khẩu phần của Mỹ đã bỏ giới hạn cholesterol theo ngày. Thứ cần tránh chỉ là ăn ba bốn quả mỗi ngày, một ngày một quả thì không phải lo + +### 37. Có điều kiện thì ngâm mình, đừng chỉ tắm vòi sen + +- Chi phí: Tốn thêm chút tiền nước và điện. Mỗi lần nửa tiếng. +- Nói thẳng: Người gần như ngày nào cũng ngâm mình có rủi ro sự kiện tim mạch thấp hơn người ngâm dưới hai lần mỗi tuần khoảng 28%, xuất huyết não thấp hơn khoảng 46%. +- Lợi ích: Một nghiên cứu ở Nhật gồm 30.000 người, theo dõi 19 năm, có 2097 sự kiện tim mạch. Người gần như ngày nào cũng ngâm mình có rủi ro sự kiện tim mạch thấp hơn người ngâm không quá hai lần mỗi tuần khoảng 28% (tỷ số rủi ro 0.72, 95% CI 0.62–0.84, đây là khoảng tin cậy). Bệnh tim mạch vành thấp hơn khoảng 35% (0.65, 0.45–0.94). Đột quỵ thấp hơn khoảng 26% (0.74, 0.62–0.87). Xuất huyết não thấp hơn khoảng 46% (0.54, 0.40–0.73) +- Mức bằng chứng: B +- Nguồn: Ukai T, Iso H, Yamagishi K, et al. (2020). Habitual tub bathing and risks of incident coronary heart disease and stroke. Heart, 106(10), 732-737. +- Ghi chú:Tranh cãi. Đây là kết quả làm ra ở dân số Nhật với văn hoá tắm bồn, phần lớn hộ gia đình Trung Quốc không có điều kiện ngâm mình mỗi ngày. Ngoài ra đừng để nước quá nóng, cũng đừng ngâm quá lâu. Với người lớn tuổi và người bị bệnh tim mạch, nước quá nóng hoặc ngâm quá lâu lại thành nguy hiểm, mỗi năm ở Nhật có rất nhiều người lớn tuổi đột tử trong bồn tắm + + +### 38. Giữ giấc ngủ trưa trong nửa tiếng, đừng quá một tiếng; nếu cứ phải ngủ một hai tiếng mới chịu thì đi tìm nguyên nhân + +- Chi phí: Không tốn tiền. Đặt một cái báo thức trước khi ngủ là xong, gần như không cần ý chí. +- Nói thẳng: Người ngủ trưa không quá một tiếng thì rủi ro tử vong và bệnh mạn tính không tăng, đầu óc còn tỉnh táo hơn một chút. Người ngủ quá một tiếng có rủi ro bệnh tim mạch vành cao hơn khoảng ba phần mười, đái thường và béo phì cao hơn khoảng hai phần mười. Vậy hãy đặt báo thức, nửa tiếng là tự đánh thức mình. Nếu ngày nào cũng phải ngủ một hai tiếng mới chịu thì đó là tín hiệu cần đi khám. +- Lợi ích: Một bài tổng quan tổng quan gộp 16 bài phân tích tổng hợp, 244 kết quả sức khoẻ. Tổng quan tổng quan nghĩa là gộp lại những nghiên cứu tổng hợp đã có. Kết luận là: người ngủ trưa dưới 60 phút thì tỷ lệ tử vong toàn nguyên và rủi ro bệnh mạn tính không tăng, cải thiện khả năng nhận thức cũng rõ nhất (SMD 0.69, 95% CI 0.37–1.00, đây là khoảng tin cậy). Giấc ngủ trưa ngắn 20 đến 30 phút cải thiện thành tích vận động nhiều nhất (SMD 0.99, 0.67–1.31). Người ngủ trưa quá 60 phút có rủi ro bệnh tim mạch vành cao hơn khoảng 30%, rủi ro đái thường và béo phì cao hơn khoảng 20%. Một nghiên cứu khác đeo đồng hồ đo thực tế giấc ngủ trưa ở 1338 người trên 56 tuổi, không dựa vào bảng hỏi tự điền. Theo dõi dài nhất 19 năm, trong thời gian đó có 926 người chết. Mỗi 1 giờ ngủ trưa tăng thêm thì rủi ro tử vong cao hơn khoảng 13% (HR 1.13, 95% CI 1.04–1.23). Ngủ trưa thêm một lần mỗi ngày thì cao hơn khoảng 7% (HR 1.07, 1.02–1.13). Người ngủ trưa vào buổi sáng cao hơn người ngủ vào đầu buổi chiều khoảng 30% (HR 1.30, 1.03–1.64). +- Mức bằng chứng: B +- Nguồn: Du P, Li J, Hua Z, v.v. (2026). Multiple Health Outcomes of Daytime Napping: A Comprehensive Umbrella Review. Public Health Reviews. ；Gao C, Cai R, Zheng X, v.v. (2026). Objectively Measured Daytime Napping Patterns and All-Cause Mortality in Older Adults. JAMA Network Open. ；Dashti HS và cộng sự (2021). Genetic determinants of daytime napping and effects on cardiometabolic health. Nature Communications. +- Ghi chú:Tranh cãi. Hai nghiên cứu trên đều chỉ ghi nhận, không chia nhóm đối chứng. Người ngủ trưa lâu vốn dễ có ngưng thở khi ngủ, thiếu máu, suy giảm chức năng tuyến giáp, trầm cảm hoặc bệnh mạn tính khác. Trong câu「ngủ nhiều thì chết nhiều」 có bao nhiêu là do bản thân giấc ngủ trưa thì không rõ. Suy luận nhân quả bằng gen (Mendelian ngẫu nhiên, 453.000 người cộng 541.000 người) chỉ ủng hộ một điều: ngủ trưa càng thường xuyên thì huyết áp và vòng eo càng cao hơn một chút. Nó không tìm thấy ảnh hưởng nhân quả của giấc ngủ trưa lên bệnh tim mạch vành và đái thường. Vậy đừng coi「ngủ trưa ít đi」là cách hạ tỷ lệ tử vong. Nửa「đi tìm nguyên nhân」là khuyến nghị của tác giả, thuộc mức C: ban ngày mệt đến mức bắt buộc phải ngủ lâu thì trước hết xem giấc ngủ ban đêm, ngáy và ngạt thở trong lúc ngủ thì xem ngưng thở khi ngủ, rồi xem công thức máu và tuyến giáp. Giấc ngủ trưa ngắn làm sao để tỉnh táo, xem mục 11 ở phần 3 (buổi chiều mệt thì ngủ 10 phút). Ban đêm ngủ bao nhiêu lâu, xem mục 13 của phần này (mỗi đêm ngủ 7 tiếng). + +### 39. Thức khuya xong thì đêm hôm sau bù ngay, đừng dồn dập cuối tuần + +- Chi phí: Không tốn tiền. Một hai tiếng bù vốn cũng là lúc bạn muốn ngủ. +- Nói thẳng: Người ngủ thiếu mà đêm hôm sau không bù thì rủi ro tử vong cao hơn người có giờ sinh hoạt đều đặn khoảng 15%; người thiếu nhiều mà vẫn không bù thì cao hơn khoảng 42%. Người có bù thì không thấy rủi ro tăng. Vậy thức khuya xong thì đêm hôm sau lên giường sớm. +- Lợi ích: Ngân hàng sinh học Anh có 85.618 người, tuổi trung bình 61.8, đeo đồng hồ theo dõi giấc ngủ từng ngày, theo dõi tiếp trung vị 8 năm. Nghiên cứu trước hết tính riêng cho từng người「đêm thiếu ngủ」theo nhu cầu ngủ của chính họ, tức là thiếu hơn 2.5 tiếng so với mức cần. Đêm đầu tiên sau đoạn thiếu ngủ đó tính là「đêm bù ngủ」, đêm này ngủ nhiều hơn mức cần của mình thì coi như đã bù. Chia theo đó thành năm mẫu. So với người có giờ sinh hoạt đều đặn, người vừa thiếu ngủ vừa không bù có rủi ro tử vong toàn nguyên cao hơn khoảng 15% (HR 1.15, 95% CI 1.01–1.31, đây là khoảng tin cậy). Người thiếu ngủ lũy kế trên 3.5 tiếng mà không bù thì cao hơn khoảng 42% (HR 1.42, 1.24–1.63). Hai nhóm đã bù ngủ đều không có liên hệ có ý nghĩa, trong đó nhóm thiếu ngủ rồi bù là HR 1.12 (0.98–1.28). Trong số người vốn ngủ ít, người thiếu ngủ không bù cao hơn khoảng 19% (1.19, 1.01–1.40), thiếu ngủ nặng không bù cao hơn khoảng 38% (1.38, 1.17–1.63). Kết quả này cũng lặp lại trong một khảo sát sức khoẻ Hoa Kỳ khác gồm 4.586 người. +- Mức bằng chứng: B +- Nguồn: Li X, Zhang M, Li Z, v.v. (2026). Acute sleep rebound following sleep restriction is associated with reduced mortality risk. Nature Communications. +- Ghi chú: Đây là nghiên cứu chỉ ghi nhận, không chia nhóm đối chứng. Người không bù được có thể vốn bận hơn, sức khoẻ kém hơn, tầng nhiễu đó không loại sạch được. Câu「hai nhóm đã bù không thấy rủi ro tăng」chỉ nói là không tìm ra khác biệt, ước lượng điểm của chúng vẫn lớn hơn 1, đã bù không có nghĩa là không sao. Vậy sau khi thức khuaya phải bù càng sớm càng tốt, cũng không vì thế mà thức khuaya tuỳ tiện. Trong nghiên cứu này, bù ngủ nghĩa là đêm đầu tiên sau khi thiếu ngủ ngủ thêm một lúc, không gồm việc dồn lại rồi ngủ một lần cuối tuần. Việc biến「thức khuya ngày làm việc, bù ngủ cuối tuần」thành một nhịp cố định hằng tuần gọi là chênh lệch giờ sinh học xã hội, bản thân nó liên quan tới bệnh tim mạch, xem mục 13 của phần này (mỗi đêm ngủ khoảng 7 tiếng, giờ sinh hoạt đều đặn). Mục về giờ dậy cố định cả cuối tuần nằm ở mục 2 ở phần 3 (giờ dậy cố định, cuối tuần cũng vậy), không mâu thuẫn với mục này: mục này bảo bạn lên giường sớm hôm đó, không phải bảo hôm sau ngủ tới trưa. + +### 40. Ca đêm làm càng lâu nguy cơ tim mạch càng cao, chuyển được ca thì chuyển sớm + +- Chi phí: Chuyển ca hoặc đổi việc có thể giảm thu nhập, tiền ca đêm cũng mất. Tìm việc tốn thời gian, quyết định cũng khó. Nếu tính khoản này vào lúc chọn việc thì chi phí bằng không. +- Nói thẳng: Người làm ca đêm mắc bệnh tim mạch nhiều hơn người không làm ca đêm khoảng 13%, tử vong tim mạch nhiều hơn khoảng 27%. Mà làm càng lâu thì càng tăng: mỗi 5 năm ca đêm thêm, bệnh tim mạch tăng thêm khoảng 7%. Vậy khoản này tính theo số năm, chuyển được ca thì chuyển sớm. Còn chuyện ca đêm gây ung thư thì bằng chứng mềm hơn nhiều so với lời đồn. +- Lợi ích: Gộp 23 nghiên cứu theo dõi lại với nhau, chỉ ghi nhận, không chia nhóm đối chứng. Người làm ca đêm có rủi ro sự kiện tim mạch cao hơn khoảng 13% (RR 1.13, 95% CI 1.10–1.16, đây là khoảng tin cậy). Tử vong tim mạch cao hơn khoảng 27% (RR 1.27, 1.18–1.36). Xét theo số năm: mỗi 5 năm ca đêm thêm, tỷ lệ mắc bệnh tim mạch tăng thêm khoảng 7% (RR 1.07, 1.04–1.09), tử vong tim mạch tăng thêm khoảng 5% (RR 1.05, 1.03–1.06). Tách theo từng bệnh: tỷ lệ mắc bệnh tim mạch vành cao hơn khoảng 22% (1.22, 1.16–1.28), tử vong do bệnh tim mạch vành cao hơn khoảng 22% (1.22, 1.10–1.36), tử vong do bệnh tim mạch thiếu máu cục bộ cao hơn khoảng 39% (1.39, 1.06–1.84), tử vong do đột quỵ cao hơn khoảng 49% (1.49, 1.04–2.12). Riêng tỷ lệ mắc đột quỵ thì không có ý nghĩa (1.06, 0.95–1.18). +- Mức bằng chứng: A +- Nguồn: Xi J, Ma W, Tao Y, v.v. (2025). Association between night shift work and cardiovascular disease: a systematic review and dose-response meta-analysis. Frontiers in Public Health. ；Esposito G, Bravi F, Santucci C, v.v. (2025). Night shift work and breast cancer risk in healthcare workers: a systematic review and meta-analysis. Occupational Medicine. ；Shen QM, Li ZY, Tan YT, v.v. (2026). Night shift work and risk of total and site-specific cancer: results from a prospective cohort study among Chinese men. Scandinavian Journal of Work, Environment & Health. ；Czeisler CA, Johnson MP, Duffy JF, v.v. (1990). Exposure to bright light and darkness to treat physiologic maladaptation to night work. New England Journal of Medicine, 322(18), 1253-1259. +- Ghi chú: Tất cả đều là nghiên cứu chỉ ghi nhận, không chia nhóm đối chứng. Người làm ca đêm vốn khác người khác ở hút thuốc, cân nặng, thu nhập, hiệu chỉnh không sạch. Chuyện ca đêm gây ung thư thì bằng chứng mềm hơn nhiều so với lời đồn. Trong phân tích gộp 12 nghiên cứu, 12.132 ca ung thư vú, người từng làm ca đêm chỉ có rủi ro cao hơn khoảng 5%, khoảng tin cậy vượt qua 1 (RR 1.05, 0.96–1.14). Nhóm làm đủ 20 năm cao hơn khoảng 25% (1.25, 1.01–1.55), nhưng tác giả làm phân tích độ nhạy thì kết quả này không còn ý nghĩa, hiệu chỉnh thiên lệch xuất bản có thể thì gần 1, nguyên văn là mối liên hệ này「còn xa mới được xác lập」. Một nghiên cứu khác theo dõi nam giới Trung Quốc trong 16.1 năm với 8.202 ca ung thư không thấy ca đêm liên quan tới ung thư tổng thể và phần lớn các loại ung thư thường gặp, chỉ riêng ung thư tuyến tụy cao hơn khoảng 59% ở nhóm lũy kế từ 11 đến 20 năm (HR 1.59, 1.09–2.31). Vậy mục này tính sổ trên phía tim mạch, không tính theo ung thư. Không phải ai cũng nên rời khỏi ca đêm, khoản tính theo số năm phải tự bạn tính. Người đang làm thì trước hết sửa những gì sửa được: cai thuốc lá xem mục 1 của phần này (cai thuốc lá, càng sớm càng tốt), huyết áp và mỡ máu xem mục 12 của phần này (có ca huyết áp cao, mỡ máu cao thì uống thuốc đúng đơn đều đặn), sau ca đêm thì bù ngủ thế nào xem mục 39 của phần này (thức khuaya thì đêm hôm sau bù ngay). Có thêm hai cách đối phóng đáng thử: dồn ăn uống vào ban ngày, và dùng ánh sáng mạnh cộng với che tối nghiêm ngặt ban ngày để dịch đồng hồ sinh học. Cách sau là cách duy nhất được chứng minh thực sự dịch được đồng hồ sinh học. Dưới ánh sáng trong nhà bình thường, làm liên tiếp sáu ca đêm thì đồng hồ sinh học không nhúc nhích; đổi sang ánh sáng mạnh 7000 đến 12000 lux cộng với ban ngày gần như tối đen, bốn ngày là đảo ngược trọn vẹn. **Ánh sáng ban đêm phải đủ sáng, lúc ngủ ban ngày phải che tối nghiêm ngặt, và trên đường đi làm về phải đeo kính râm chặn ánh sáng buổi sớm, thiếu một trong ba thì vô nghĩa**, bỏ cái cuối thì coi như không làm. Cả hai cách đối phóng chỉ đi được tới mức chỉ số chuyển hoá và giấc ngủ, chưa thử nghiệm nào đo xem làm theo có hạ được rủi ro tim mạch hay không. Ngày nghỉ mà sống theo lịch ban ngày bình thường thì đồng hồ lại bị đẩy ngược lại. Về liều dùng, các con số gốc và phần nguồn còn lại, cùng với cách cơ thể nhận biết thời gian, đều ở [docs/生物钟和夜班.md](../docs/生物钟和夜班.md)。 + + +### 41. Mua dầu ăn đóng gói có số SC, đừng mua dầu điều tự ép tẩy từ xưởng nhỏ bán rời + +- Chi phí: Đổi chỗ mua thôi, tiền tương đương. Chỉ cần nhìn trên bao bì có số giấy phép sản xuất thực phẩm bắt đầu bằng SC hay không. Chỗ khó là thuyết phục người lớn tuổi quen mua dầu tự ép. +- Nói thẳng: Đậu phộng bị ẩm mốc sinh aflatoxin, đây là chất gây ung thư đã rõ, và nó gây hại chủ yếu ở gan. Sau khi Quảng Châu quản được các xưởng nhỏ, tỷ lệ dân thường ăn dầu tự ép có xét nghiệm chức năng gan bất thường giảm khoảng ba phần rưỡi. Ở Quảng Tây, phụ nữ có thai ăn dầu điều tự ép có nguy cơ sinh con cân nặng thấp cao hơn khoảng chín phần mười, sinh non cao hơn khoảng tám phần mười. +- Lợi ích: Aflatoxin là chất gây ung thư nhóm 1 do Cơ quan Quốc tế Nghiên cứu Ung thư xếp, nghĩa là chắc chắn gây ra ung thư, và chủ yếu là ung thư gan. Một nghiên cứu ở Quảng Châu đo aflatoxin B1 trong dầu điều tự ép của các xưởng nhỏ trong giai đoạn 2010 đến 2022. Trung vị là 1.29 μg/kg, nửa mẫu nằm trong khoảng 0.12 đến 6.58. Sau khi địa phương ban hành quy chế quản lý xưởng nhỏ, độc tố trong dầu giảm ngay 2.865 μg/kg, rồi mỗi năm giảm thêm 2.593. Tỷ lệ dân ở khu vực thường ăn dầu tự ép có chức năng gan bất thường giảm theo (PR 0.650, 95% CI 0.469–0.902, khoảng tin cậy), khoảng 35%. Một nghiên cứu ở Quảng Tây theo dõi 1611 phụ nữ có thai, trong đó 81.7% ăn dầu điều tự ép. Người ăn dầu điều tự ép có nguy cơ con cân nặng thấp khi sinh cao hơn khoảng chín phần mười (aOR 1.9, 1.1–3.2), sinh non cao hơn khoảng tám phần mười (aOR 1.8, 1.1–3.0). +- Mức bằng chứng: B +- Nguồn: Lei J, Li Y, Wang Y, v.v. (2024). The impact of small food workshops management regulations on aflatoxin B1 in home-made peanut oil and the liver function of high-consumption area residents: an interrupted time series study in Guangzhou, China. Frontiers in Public Health, 12. ；Zhong Y, Lu H, Jiang Y, v.v. (2024). Effect of homemade peanut oil consumption during pregnancy on low birth weight and preterm birth outcomes: a cohort study in Southwestern China. Global Health Action, 17. ；IARC (2012). Chemical Agents and Related Occupations. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol 100F. ；Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委), Cục Quản lý Dược phẩm và Thực phẩm quốc gia (国家食品药品监督管理总局) (2017). Tiêu chuẩn quốc gia về an toàn thực phẩm, Giới hạn độc tố nấm men trong thực phẩm (GB 2761-2017) +- Ghi chú: Cả hai nghiên cứu chỉ ghi nhận và theo dõi, không có nhóm đối chứng. Kết quả cuối của nghiên cứu ở Quảng Châu là xét nghiệm chức năng gan, không phải ung thư gan và không phải tử vong. Trung vị độc tố trong dầu tự ép thực ra thấp hơn giới hạn 20 μg/kg mà tiêu chuẩn quốc gia đặt cho dầu điều, rủi ro đến từ việc ăn nhiều năm và từ những lô đơn lẻ cao bất thường. Giới hạn của tiêu chuẩn quốc gia áp dụng cho mọi loại dầu điều, nhưng dầu đóng gói phải qua giấy phép sản xuất và kiểm nghiệm trước khi xuất xưởng, còn xưởng nhỏ thì theo biện pháp quản lý của từng tỉnh và được lấy mẫu kiểm tra ít hơn nhiều. Người hưởng lợi từ mục này chủ yếu là bạn và người ăn cùng bạn. + +### 42. Nấu ăn bằng dầu thực vật thay cho mỡ lợn và bơ, không cần đổi dầu qua lại vì lý do sức khoẻ, cũng đừng trông chờ dầu hạt lanh bảo vệ tim + +- Chi phí: Không tốn thêm tiền. Dầu thực vật vốn đã là lựa chọn chính của đa số hộ gia đình. Chỗ khó là bỏ thói quen dùng mỡ lợn để xào và rưới cơm. +- Nói thẳng: Ăn ít mỡ lợn, bơ và các loại mỡ bão hòa tương tự, đổi sang dầu thực vật thì các sự kiện như bệnh tim và đột quỵ giảm khoảng mười bảy phần trăm, nhưng rủi ro tử vong toàn nguyên không thấy thay đổi. Chuyển qua lại giữa dầu đậu nành, dầu ngô, dầu cải thì không có bằng chứng nào cho là khác nhau. Ăn nhiều dầu hạt lanh thì rủi ro bệnh tim cũng không thấy giảm. +- Lợi ích: Cochrane gộp 15 thử nghiệm chia nhóm ngẫu nhiên, 56.675 người, mỗi thử nghiệm kéo dài ít nhất hai năm. Ăn ít mỡ bão hòa thì tổng số sự kiện tim mạch thấp hơn khoảng 17% (RR 0.83, 0.70–0.98). Tỷ lệ tử vong toàn nguyên gần như không đổi (RR 0.96, 0.90–1.03), tử vong tim mạch cũng không đổi (RR 0.95, 0.80–1.12). Đổi sang chất béo không bão hòa nhiều (thành phần chính của dầu đậu nành và dầu ngô) và đổi sang carbohydrate thì không thấy khác biệt. Đổi sang chất béo không bão hòa đơn (thành phần chính của dầu cải và dầu ô liu) thì dữ liệu quá ít, không nói rõ được. Một bài tổng quan Cochrane khác gộp 19 thử nghiệm chia nhóm ngẫu nhiên, 6461 người: ăn nhiều omega-6 (axit linoleic) thì với tỷ lệ tử vong toàn nguyên RR 1.00 (0.88–1.12), với sự kiện tim mạch RR 0.97 (0.81–1.15), không thấy lợi ích cũng không thấy hại. Bài thứ ba gộp 86 thử nghiệm chia nhóm ngẫu nhiên, 162.796 người: ăn nhiều omega-3 nguồn thực vật (thành phần chính của dầu hạt lanh và dầu tía lô), tỷ lệ tử vong toàn nguyên RR 1.01 (0.84–1.20), sự kiện bệnh tim mạch vành RR 1.00 (0.82–1.22), gần như không có tác dụng. +- Mức bằng chứng: A +- Nguồn: Hooper L, Martin N, Jimoh OF, v.v. (2020). Reduction in saturated fat for cardiovascular disease. Cochrane Database of Systematic Reviews, (5), CD011737. ；Hooper L, Al-Khudairy L, Abdelhamid AS, v.v. (2018). Omega-6 fats for the primary and secondary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews, (11), CD011094. ；Abdelhamid AS, Brown TJ, Brainard JS, v.v. (2020). Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews, (3), CD003177. ；Hiệp hội Dinh dưỡng Trung Quốc (中国营养学会) (2022). Hướng dẫn khẩu phần cư dân Trung Quốc (2022). Nhà xuất bản Nhân dân Y tế +- Ghi chú:Tranh cãi: có người cho rằng lượng omega-6 trong dầu đậu nành và dầu ngô quá nhiều nên gây viêm, do đó phải ăn ít và đổi dầu. Các thử nghiệm chia nhóm ngẫu nhiên trong bài omega-6 nói trên không thấy hại, tác giả tự chấm chất lượng bằng chứng là thấp. Mức lợi ích đặt là vừa dựa trên sự kiện tim mạch giảm 17%, còn tỷ lệ tử vong toàn nguyên thì không đổi. Dầu nào cũng phải kiểm soát lượng dùng, hướng dẫn khẩu phần khuyên mỗi người mỗi ngày dùng 25 đến 30 g dầu nấu ăn. diff --git "a/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" "b/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" new file mode 100644 index 00000000..ea16f0a0 --- /dev/null +++ "b/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" @@ -0,0 +1,217 @@ +[← 回总目录](../README.md) + +# 3. 不要浪费精力 + +本节只算精力和时间账。看的是四样东西：你每天能用的注意力、反应快不快、出错多不多、被打断以后要多久才回得来。所有数字都是直接测出来的，来自实验，或者来自跟踪一群人做的记录（队列研究）。这些数字不换算成寿命，也不换算成死亡率。本节多数条目的证据等级是 B。这个领域本来就这样，不代表建议不成立。第 20 条讲的是和警察、医生、柜员打交道时该有的预期，引的是法规，不是实验，所以等级记 C。第 15 条（把悲观念头当症状看）和第 23 条（把「别人要求我完美」当症状看）引的是死亡率与自杀风险的数字，口径和本节其他条目不一样，不放在一起比较。 + +### 1. Tắt các thông báo không cần thiết, và để điện thoại ngoài tầm nhìn khi làm việc + +- Chi phí: Không tốn tiền. Mất vài phút để đổi thiết lập, gần như không cần ý chí. +- Nói thẳng: Điện thoại rung một lần, dù bạn không nhìn, không chạm vào, điểm số bài tập chú ý vẫn giảm. Mức giảm tương tự lúc bạn thật sự nghe điện thoại hay nhắn tin. Còn một tình huống nữa: điện thoại chỉ nằm trên bàn, bạn kìm không nhìn, phần sức não dùng được vẫn giảm. Người càng không rời được điện thoại thì giảm càng nhiều. +- Lợi ích: Chỉ riêng việc nhận một thông báo, không nhìn cũng không chạm vào, điểm số bài tập chú ý đã xấu đi rõ rệt. Mức xấu đi tương tự lúc thật sự nghe gọi hay nhắn tin. Thí nghiệm làm trong phòng thí nghiệm, người tham gia là sinh viên đại học. Hai thí nghiệm khác cho thấy: dù đã kìm được không nhìn điện thoại, việc điện thoại của chính mình nằm bên cạnh vẫn làm giảm phần sức não dùng được. Người phụ thuộc nặng vào điện thoại giảm nhiều nhất. +- Mức bằng chứng: B +- Nguồn: Stothart, Mitchum & Yehnert (2015). The attentional cost of receiving a cell phone notification. Journal of Experimental Psychology: Human Perception and Performance. ; Ward, Duke, Gneezy & Bos (2017). Brain Drain: The Mere Presence of One's Own Smartphone Reduces Available Cognitive Capacity. Journal of the Association for Consumer Research. +- Ghi chú: Đều là những tác vụ đơn lẻ trong phòng thí nghiệm. Cộng dồn mỗi ngày thì ra sao, chưa ai đo. Với công việc thật sự phải có mặt bất cứ lúc nào, chỉ giữ thông báo của vài đồng nghiệp là được. + +### 2. Giờ thức dậy cố định, cuối tuần cũng vậy + +- Chi phí: Không tốn tiền. Cuối tuần phải bỏ thói ngủ bù, cần một chút ý chí. +- Nói thẳng: Nhóm một phần năm có giờ sinh hoạt loạn nhất, đồng hồ sinh học lệch muộn hơn khoảng 2.5 giờ so với nhóm một phần năm đều đặn nhất. Nghĩa là sống lâu trong tình trạng lệch múi giờ. Học sinh nào giờ sinh hoạt càng đều thì kết quả học tập càng tốt. +- Lợi ích: 61 sinh viên đại học ghi nhận trong 30 ngày. Nhóm một phần năm có giờ sinh hoạt ít đều đặn nhất, đồng hồ sinh học (nhịp ngày đêm) lệch muộn hơn khoảng 2.5 giờ so với nhóm một phần năm đều đặn nhất. Cụ thể là giờ bắt đầu tiết melatonin là 00:08 so với 21:32. Khung giờ mệt nhất, muốn ngủ nhất cũng lệch muộn khoảng 1.8 giờ. Giờ sinh hoạt càng đều thì kết quả học tập càng tốt (hệ số tương quan r = 0.37). +- Mức bằng chứng: B +- Nguồn: Phillips et al. (2017). Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sleep/wake timing. Scientific Reports. +- Ghi chú: Chỉ có một nghiên cứu, số người ít, cũng không có nhóm đối chứng. Nhà nghiên cứu suy ra rằng khác biệt về đồng hồ sinh học chủ yếu đến từ lượng ánh sáng mỗi ngày khác nhau. Vì vậy giờ thức dậy cố định phải đi kèm việc tiếp xúc với ánh sáng buổi sáng mới có tác dụng. Thỉnh thoảng thức khuya rồi muốn ngủ bù, xem mục 39 ở phần 2 (ngủ bù vào tối hôm sau đêm thức khuya): cách làm là lên giường sớm hôm đó, không phải ngủ đến trưa hôm sau. + +### 3. Ngủ đủ 7 đến 8 giờ mỗi đêm, đừng coi 6 giờ là đủ + +- Chi phí: Không tốn tiền. Mỗi ngày phải dành thêm 1 đến 2 giờ cho giấc ngủ. Lên giường sớm hơn thì cần một chút ý chí. +- Nói thẳng: Hai tuần liền mỗi đêm chỉ ngủ 6 giờ, hiệu năng não rơi xuống mức tương đương với hai ngày hai đêm không ngủ. Rắc rối là người ta không cảm nhận ra điều đó. Điểm buồn ngủ tự chấm tăng trong vài ngày đầu rồi đứng yên, người ngủ 6 giờ và người ngủ 4 giờ chấm điểm buồn ngủ bằng nhau. Vì vậy câu「tôi ngủ 6 giờ là đủ」bản thân nó là một triệu chứng thiếu ngủ. +- Lợi ích: 48 người lớn khỏe mạnh từ 21 đến 38 tuổi ngủ mỗi đêm 6 giờ trong 14 ngày liên tiếp. Hiệu năng nhận thức của họ giảm dần, rơi xuống mức tương đương với 2 đêm không ngủ hoàn toàn. Mức buồn ngủ tự chấm chỉ tăng trong vài ngày đầu, sau đó hầu như không tăng nữa. Nhóm ngủ 6 giờ và nhóm ngủ 4 giờ không khác nhau về buồn ngủ tự đánh giá. Đây là thử nghiệm đối chứng chia ngẫu nhiên người tham gia vào hai nhóm (RCT). Ngoài ra, thiếu ngủ trong thời gian ngắn (cắt giấc ngủ cấp tính, không quá 48 giờ) có cỡ tác động lên sự mất tập trung ở nhiệm vụ chú ý đơn giản là g = -0.776, và lên độ chính xác suy luận là g = -0.125. Phần này gộp 70 nghiên cứu lại. +- Mức bằng chứng: A +- Nguồn: Van Dongen, Maislin, Mullington & Dinges (2003). The cumulative cost of additional wakefulness. Sleep. ; Lim & Dinges (2010). A meta-analysis of the impact of short-term sleep deprivation on cognitive variables. Psychological Bulletin. +- Ghi chú: Cảm giác「không buồn ngủ」không phải bằng chứng cho thấy đã ngủ đủ. Người ngủ 6 giờ lâu dài chỉ quen về cảm giác, hiệu năng thì không theo kịp. Phần gộp nhiều nghiên cứu nói về việc không ngủ suốt đêm. 「Ngủ ít lâu dài」chỉ có một thử nghiệm đối chứng ngẫu nhiên hỗ trợ. + +### 4. Sau hai giờ chiều không dùng caffeine + +- Chi phí: Không tốn tiền. Người quen uống cà phê buổi chiều phải kìm không uống, cần một chút ý chí. +- Nói thẳng: Uống một ly cà phê lớn (400 mg) cách giờ đi ngủ 6 giờ, đo bằng máy cho thấy đêm đó ngủ ít hơn 1 giờ trở lên. Nhưng nhật ký ngủ tự ghi ra không thấy khác biệt. Một ly cà phê thường (107 mg) phải uống xong trước giờ đi ngủ hơn 8.8 giờ thì mới không ảnh hưởng tới giấc ngủ. Người đi ngủ lúc 11 giờ thì hai giờ chiều là ranh giới. +- Lợi ích: Uống 400 mg caffeine cách giờ đi ngủ 6 giờ, máy đo cho thấy tổng thời gian ngủ ít hơn 1.1 đến 1.2 giờ so với khi uống giả dược. Giả dược là đồ uống đối chứng không chứa caffeine. Nhưng nhật ký ngủ do chính người tham gia tự ghi không cho thấy khác biệt, bản thân họ cũng không nhận ra. Gộp nhiều nghiên cứu lại (phân tích tổng hợp): caffeine làm giảm tổng thời gian ngủ trung bình 45 phút. Tỷ lệ thời gian nằm trên giường mà thật sự ngủ (hiệu suất ngủ) giảm 7%. Một ly cà phê (107 mg) phải uống xong trước giờ đi ngủ hơn 8.8 giờ thì mới không ảnh hưởng tới giấc ngủ. +- Mức bằng chứng: A +- Nguồn: Drake, Roehrs, Shambroom & Roth (2013). Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed. Journal of Clinical Sleep Medicine. ; Gardiner et al. (2023). The effect of caffeine on subsequent sleep: A systematic review and meta-analysis. Sleep Medicine Reviews. +- Ghi chú: Nghiên cứu của Drake chỉ do một phòng thí nghiệm thực hiện, dùng 400 mg thì hơi lớn, khoảng hai ba ly cà phê. Người chuyển hóa caffeine chậm thì thời điểm phải dừng còn sớm hơn. Bớt caffeine đổi lại được thời lượng ngủ, và cộng thêm với mục 3 của phần này (ngủ đủ 7 đến 8 giờ mỗi đêm). + +### 5. Chuyển mail và tin nhắn sang xử lý theo lô, vài lần cố định mỗi ngày + +- Chi phí: Không tốn tiền. Mất chút thời gian nói rõ với đồng nghiệp rằng bạn trả lời bao lâu một lần. Kìm không mở mail suốt ngày thì cần một chút ý chí. +- Nói thẳng: Cùng một nhóm người được so sánh trong hai tuần. Tuần giới hạn mail chỉ kiểm tra 3 lần mỗi ngày thì áp lực hằng ngày thấp hơn rõ rệt, trong khi số mail gửi và nhận vẫn như nhau. Khi không giới hạn, người ta kiểm tra mail trung bình 15.5 lần mỗi ngày. +- Lợi ích: 124 người lớn, trong hai tuần cùng một nhóm được so sánh trước sau (đối chứng chính bản thân). Tuần giới hạn kiểm tra mail 3 lần mỗi ngày, áp lực hằng ngày thấp hơn rõ rệt so với tuần không giới hạn (cỡ tác động Cohen's d = 0.37). Số mail nhận và trả lời trong hai tuần không khác nhau. Lúc đầu không giới hạn, người ta kiểm tra mail trung bình 15.5 lần mỗi ngày. +- Mức bằng chứng: B +- Nguồn: Kushlev & Dunn (2015). Checking email less frequently reduces stress. Computers in Human Behavior. +- Ghi chú: Chỉ có một nghiên cứu, cỡ tác động vừa đến nhỏ, đo là áp lực chứ không phải lượng việc tạo ra. Tin nhắn tức thì chưa có nghiên cứu trực tiếp nào, ở đây chỉ suy ra theo. + +### 6. Khi làm việc cần suy nghĩ liên tục, chặn cả những lần gián đoạn chỉ vài giây + +- Chi phí: Không tốn tiền. Đóng cửa, đeo tai nghe, hoặc treo một biển「xin không làm phiền」. +- Nói thẳng: Một lần gián đoạn trung bình 2.8 giây làm xác suất sai sót ở những bước sau tăng gấp đôi. Gián đoạn 4.4 giây làm tăng gấp ba. Theo dõi tại hiện trường còn thấy, việc bị gián đoạn mất trung bình 25 phút mới quay lại được, và có 2 việc khác chèn vào giữa chừng. Người bị gián đoạn làm việc nhanh hơn, nhưng áp lực, mức bực bội và cảm giác mệt mỏi đều cao hơn. +- Lợi ích: Một lần gián đoạn trung bình 2.8 giây làm tỷ lệ sai sót ở từng bước của nhiệm vụ làm tuần tự tăng gấp đôi. Gián đoạn 4.4 giây làm tăng gấp ba. Đây là thí nghiệm trong phòng thí nghiệm. Một nghiên cứu khác quan sát tại hiện trường với 24 nhân viên văn phòng (người làm công việc thông tin). Họ chuyển chủ đề công việc trung bình mỗi 11 phút một lần. 57% các chủ đề bị gián đoạn. Để quay lại việc bị gián đoạn trong ngày hôm đó mất trung bình 25 phút 26 giây. Giữa chừng có 2.26 việc khác chèn vào. Một thí nghiệm nữa với 48 người: người bị gián đoạn làm nhanh hơn, nhưng báo cáo áp lực, bực bội, áp lực thời gian và mức mệt mỏi đều cao hơn. +- Mức bằng chứng: B +- Nguồn: Altmann, Trafton & Hambrick (2014). Momentary interruptions can derail the train of thought. Journal of Experimental Psychology: General. ; Mark, Gonzalez & Harris (2005). No task left behind? Examining the nature of fragmented work. CHI 2005. ; Mark, Gudith & Klocke (2008). The cost of interrupted work: More speed and stress. CHI 2008. +- Ghi chú: Nghiên cứu tại hiện trường đó chia gián đoạn làm hai loại: một loại do người khác tìm đến bạn, một loại do chính bạn bấm vào. Đóng cửa chỉ chặn được loại trước. Loại sau phải dùng kèm mục 1 (tắt thông báo không cần thiết) và mục 5 (xử lý tin nhắn theo lô). + +### 7. Mỗi lúc làm một việc, đừng vừa họp vừa trả lời tin nhắn + +- Chi phí: Không tốn tiền. Kìm không trả lời tin nhắn trong cuộc họp thì cần một chút ý chí. +- Nói thẳng: Mỗi lần chuyển nhiệm vụ, phản ứng chậm hơn và dễ sai hơn. Người thường xuyên mở nhiều màn hình cùng lúc không rèn được khả năng làm nhiều việc cùng lúc, trái lại còn làm bài kiểm tra chuyển nhiệm vụ kém hơn. +- Lợi ích: Sau khi chuyển nhiệm vụ, phản ứng chậm hơn rõ rệt và dễ sai hơn. Phần này đến từ một bài tổng quan nhiều nghiên cứu. Một nghiên cứu phòng thí nghiệm khác có người tham gia là sinh viên đại học. Người thường dùng nhiều loại phương tiện cùng lúc (người đa nhiệm với phương tiện truyền thông) dễ bị những thứ không liên quan và những ký ức không liên quan kéo đi hơn. Trên bài kiểm tra chuyển nhiệm vụ, họ lại làm kém hơn. +- Mức bằng chứng: B +- Nguồn: Monsell (2003). Task switching. Trends in Cognitive Sciences. ; Ophir, Nass & Wagner (2009). Cognitive control in media multitaskers. PNAS. +- Ghi chú: Chuyển qua chuyển lại tốn thêm bao nhiêu thì khác nhau nhiều theo loại việc, ở đây không đưa ra một con số chung. Nghiên cứu của Ophir là so sánh tại một thời điểm, không phân biệt được「đa nhiệm làm giảm khả năng chú ý」hay「người kém chú ý lại thích đa nhiệm hơn」. + +### 8. Một giờ trước khi ngủ không xem màn hình phát sáng, muốn đọc thì đọc sách giấy hoặc màn hình mực điện tử + +- Chi phí: Không tốn tiền. Đặt điện thoại xuống một giờ trước khi ngủ phải chống lại thói quen lâu dài, cần nhiều ý chí. +- Nói thẳng: Đọc màn hình phát sáng trước khi ngủ, thời điểm đi ngủ muộn hơn khoảng 10 phút so với đọc sách giấy. Melatonin bị hạ 55%. Đồng hồ sinh học hôm sau bị đẩy lùi hơn 1.5 giờ, buổi sáng mệt hơn, phải lâu hơn mới tỉnh hẳn. +- Lợi ích: Trong phòng thí nghiệm, cùng một nhóm người làm luân phiên hai điều kiện (thử nghiệm chéo). Đọc sách điện tử có màn hình phát sáng trước khi ngủ khiến ngủ vào chậm hơn khoảng 10 phút so với đọc sách giấy (25.7 phút so với 15.8 phút). Melatonin buổi tối bị hạ 55%. Đồng hồ sinh học hôm sau (nhịp ngày đêm) bị đẩy lùi hơn 1.5 giờ, giờ bắt đầu tiết melatonin là 22:31 so với 21:01. Buổi sáng hôm sau mệt hơn, phải lâu hơn mới tỉnh hẳn. +- Mức bằng chứng: B +- Nguồn: Chang, Aeschbach, Duffy & Czeisler (2015). Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. PNAS. +- Ghi chú: Số người ít, thí nghiệm làm trong phòng thí nghiệm với điều kiện được kiểm soát. Với độ sáng và thời lượng sử dụng điện thoại như ngoài đời, ảnh hưởng có thể không lớn đến vậy. Chế độ ban đêm của điện thoại có ích bao nhiêu, nghiên cứu này không đo. + +### 9. Đến giờ là ngủ, không thức vì game, video ngắn hay nội dung khiêu dâm + +- Chi phí: Không tốn tiền. Khó ở chỗ chống lại quán tính「xem nốt tập nữa」suốt một thời gian dài. Để điện thoại ngoài phòng ngủ thì nhàn nhất. +- Nói thẳng: Năm 2025 người Trung Quốc ngủ trung bình 6.97 giờ mỗi đêm, trung bình đến 0:10 mới vào giấc. Người dùng điện thoại hay máy tính bảng trước khi ngủ có nguy cơ ngủ không đủ cao hơn hai lần, và cao hơn 2.7 lần về mức buồn ngủ ban ngày. Chỉ để thiết bị trong phòng ngủ mà không dùng thì con số cũng tương tự. Vậy cách ít tốn sức nhất là để điện thoại ngoài phòng ngủ. +- Lợi ích: Năm 2025 người dân trong nước ngủ trung bình 6.97 giờ mỗi đêm, giờ vào giấc trung bình là 0:10 (Hiệp hội Nghiên cứu Giấc ngủ Trung Quốc). Người dùng thiết bị cầm tay trước khi ngủ có khả năng ngủ không đủ khoảng 2.17 lần (OR 2.17). Chất lượng giấc ngủ kém khoảng 1.46 lần (OR 1.46). Buồn ngủ ban ngày khoảng 2.72 lần (OR 2.72). Chỉ để thiết bị trong phòng ngủ mà không dùng thì con số cũng tương tự, khoảng 1.79, 1.53 và 2.27 lần (OR 1.79/1.53/2.27). Phần này gộp 20 khảo sát tại cùng một thời điểm, tổng 125198 trẻ em và thiếu niên. Một khảo sát tại cùng thời điểm khác: trong 844 người từ 18 đến 94 tuổi, người vẫn dùng điện thoại sau khi tắt đèn vào giấc chậm hơn, tỷ lệ thật sự ngủ trên giường thấp hơn, và mệt ban ngày hơn. +- Mức bằng chứng: B +- Nguồn: Carter et al. (2016). Association Between Portable Screen-Based Media Device Access or Use and Sleep Outcomes. JAMA Pediatrics. ; Exelmans & Van den Bulck (2016). Bedtime mobile phone use and sleep in adults. Social Science & Medicine. ; Hiệp hội Nghiên cứu Giấc ngủ Trung Quốc (中国睡眠研究会) (2026). Sách trắng Nghiên cứu Sức khỏe Giấc ngủ Trung Quốc 2026 (2026 中国睡眠健康研究白皮书), do Health News đưa tin và được Sở Y tế Phúc Kiên (福建省卫健委) chuyển đăng. ; Grubbs JB et al. (2018). Pornography Problems Due to Moral Incongruence: An Integrative Model with a Systematic Review and Meta-Analysis. Archives of Sexual Behavior. (một trong hai bên tranh cãi ở Ghi chú) ; Prause N, Pfaus J (2015). Viewing Sexual Stimuli Associated with Greater Sexual Responsiveness, Not Erectile Dysfunction. Sexual Medicine. (bên còn lại ở Ghi chú) +- Ghi chú:Tranh cãi. Cái giá của việc thức khuya tự nó xem ở mục 3 của phần này (ngủ đủ 7 đến 8 giờ mỗi đêm). Nội dung khiêu dâm gây ra gì cho sức lực và trí não của người bình thường, hiện chưa có bằng chứng tin được để nói có phải nó gây ra hay không. Một phân tích gộp nhiều nghiên cứu cho rằng「vấn đề khiêu dâm」phần lớn đến từ chỗ quan niệm của một người không khớp với chính hành vi của người đó, chứ không nằm ở nội dung. Nghiên cứu khác gộp dữ liệu 280 nam giới. Kết quả là thời lượng xem không liên quan đến chức năng cương khi quan hệ với bạn tình. Mục này chỉ nói về thời gian ngủ mà nó chiếm mất. Hai con số 6.97 giờ và 0:10, Hiệp hội Nghiên cứu Giấc ngủ Trung Quốc không đăng bản thông cáo trên trang của mình (kiểm tháng 9 năm 2026), nên lấy bài báo của Health News do Sở Y tế Phúc Kiên chuyển đăng làm căn cứ. +### 10. Đêm khuya không ra quyết định lớn, không gửi tin gấp + +- Chi phí: Không tốn tiền. Chỉ là đẩy việc sang sau khi đã ngủ dậy. +- Nói thẳng: Khi được ngủ đủ, người ta học rất nhanh cách tránh lựa chọn rủi ro cao. Sau hai đêm không ngủ, người ta lại lần chọn rủi ro cao, càng lớn tuổi càng rõ. Không ai đo việc ngủ ít vài giờ ảnh hưởng bao nhiêu, nhưng quyết định quan trọng vẫn đẩy sang sau khi ngủ dậy. +- Lợi ích: 34 người khỏe mạnh làm nhiệm vụ đánh bạc Iowa sau 49.5 giờ không ngủ. Đó là một thí nghiệm dùng để đo việc ra quyết định mạo hiểm. Khi được nghỉ ngơi đủ, họ học rất nhanh cách tránh các lựa chọn rủi ro cao. Sau khi thiếu ngủ, họ lại chọn rủi ro cao nhiều hơn. Người lớn tuổi chịu ảnh hưởng mạnh hơn. Đây là một thí nghiệm, và chỉ có nghiên cứu này. +- Mức bằng chứng: B +- Nguồn: Killgore, Balkin & Wesensten (2006). Impaired decision making following 49 h of sleep deprivation. Journal of Sleep Research. +- Ghi chú: Nghiên cứu xét tình huống cực đoan là hai đêm liên tục không ngủ. Người bình thường thức đến hai giờ sáng sẽ ra sao, chưa ai đo trực tiếp. Nửa「không gửi tin nhắn」là kinh nghiệm của tác giả (mức C), gửi vào ngày hôm sau cũng chưa muộn. +### 11. Buồn ngủ buổi chiều thì ngủ 10 phút, đừng ngủ nửa tiếng + +- Chi phí: Không tốn tiền. Mất 10 phút, thêm việc tìm chỗ nằm xuống hay úp mặt. +- Nói thẳng: Một giấc 10 phút tỉnh ngay tại chỗ, hiệu lực giữ được hai tiếng rưỡi. Ngủ 5 phút gần như vô ích. Ngủ 20 phút phải qua nửa tiếng mới có tác dụng. Ngủ 30 phút thì lúc tỉnh trước mắt có một đoạn mê man: giai đoạn đó còn làm kém hơn cả việc không ngủ. +- Lợi ích: Sau một đêm ngủ không đủ, giấc 10 phút cải thiện ngay buồn ngủ, mệt mỏi, sức lực và hiệu suất nhận thức, một phần lợi ích còn giữ qua 155 phút. Ngủ 5 phút hầu như không có lợi ích. Lợi ích của giấc 20 phút phải qua 35 phút mới xuất hiện. Ngủ 30 phút thì trước mắt có một đoạn lìa giấc: vừa tỉnh hiệu suất lại kém hơn, rồi mới khá lên. +- Mức bằng chứng: B +- Nguồn: Brooks & Lack (2006). A brief afternoon nap following nocturnal sleep restriction: Which nap duration is most recuperative? Sleep. +- Ghi chú: Chỉ có một nghiên cứu thí nghiệm trong phòng lab, người tham dự đều là người trẻ khỏe mạnh. Không ngủ được cũng đừng cố. Nằm nhắm mắt 10 phút có tác dụng hay không, không có nghiên cứu nào chống đỡ. Ngủ trưa quá một giờ lại đi cùng tỷ lệ tử vong và nguy cơ bệnh vành cao hơn, xem mục 38 ở phần 2 (giữ giấc trưa trong nửa tiếng). +### 12. Làm việc ở văn phòng mở thì dùng nút bịt tai hoặc tìm phòng yên tĩnh cho việc cần ghi nhớ + +- Chi phí: Nút bịt tai vài chục tệ. Đặt trước một phòng họp yên tĩnh thì tốn chút thời gian. +- Nói thẳng: Tiếng ồn nền tăng từ 39 lên 51 decibel thì số từ nhớ được giảm, người mệt hơn, và không muốn làm nữa. Lúc nghỉ xem phim thiên nhiên có tiếng thì lấy lại sức tốt hơn là cứ nghe tiếng ồn văn phòng. +- Lợi ích: Cho sinh viên làm việc ở văn phòng mở mô phỏng. Khi tiếng ồn nền tăng từ 39 lên 51 decibel (dB, LAeq), họ nhớ được ít từ hơn, tự chấm mình mệt hơn, và không muốn làm nữa. Lúc nghỉ, xem phim thiên nhiên có tiếng kéo lại điểm sức lực tốt hơn so với chỉ nghe tiếng sông hay tiếng ồn văn phòng. +- Mức bằng chứng: B +- Nguồn: Jahncke, Hygge, Halin, Green & Dimberg (2011). Open-plan office noise: Cognitive performance and restoration. Journal of Environmental Psychology. +- Ghi chú: Chỉ một nghiên cứu thí nghiệm. Trong tiếng ồn, chỗ hại nhất là lời nói nghe rõ nội dung. Tai nghe tiếng trắng chống được chuyện đó, nhưng nghiên cứu này không thử tai nghe. +### 13. Đừng làm quá 49 giờ mỗi tuần + +- Chi phí: Có thể ít nhận một ít tiền làm thêm giờ. Cũng có thể phải bỏ sức nói chuyện với cấp trên. +- Nói thẳng: Số liệu nhà máy vũ khí Anh thời Chiến tranh Thế giới lần thứ nhất cho thấy, trong 49 giờ mỗi tuần, làm bao lâu thì ra bấy nhiêu việc. Vượt 49 giờ, mỗi giờ thêm ra được ít dần. Một tuần 70 giờ ra việc gần bằng 56 giờ, 14 giờ thêm kia là làm không. +- Lợi ích: Số liệu sản lượng của công nhân nhà máy vũ khí Anh thời Chiến tranh Thế giới lần thứ nhất: trong 49 giờ mỗi tuần, sản lượng tỷ lệ thuận với giờ làm. Vượt 49 giờ, tốc độ tăng sản lượng bắt đầu chậm lại. Khoảng 63 giờ thì đạt mức cao nhất. Làm 70 giờ mỗi tuần ra việc gần bằng làm 56 giờ, 14 giờ thêm coi như làm không. +- Mức bằng chứng: B +- Nguồn: Pencavel (2015). The Productivity of Working Hours. The Economic Journal. +- Ghi chú: Dữ liệu lấy từ nhà máy của một trăm năm trước, làm việc chân tay, công nhân phần lớn là phụ nữ. Ngưỡng này đưa sang việc trí óc chưa chắc giống, chính tác giả cũng nói ngành khác có thể cao hơn hoặc thấp hơn. Điều tin được là hướng đi: qua một điểm nào đó, làm thêm nữa cũng gần như không ra thêm việc. Con số 49 cụ thể thì không tin được. +### 14. Khi nhận ra mình đang nghĩ đi nghĩ lại một chuyện tồi tệ, đổi sang việc phải động tay + +- Chi phí: Không tốn tiền. Khó ở chỗ phải nhận ra mình đang nghĩ đi nghĩ lại (gọi là phản tư), mà việc nhận ra đã là phải tập. +- Nói thẳng: Nghĩ đi nghĩ lại một chuyện tồi tệ không sinh ra đáp án. Nó chỉ làm trầm cảm thêm, ghim chặt những ý nghĩ xấu hơn, khiến bạn giải quyết vấn đề kém hơn và khó thật sự bắt tay làm, rồi làm cạn kiệt những người quanh bạn chịu lắng nghe. Khi nhận ra mình đang lặp lại vòng quay, hãy đi làm việc gì cần động tay. +- Lợi ích: Một tổng hợp nhiều nghiên cứu nói rằng nghĩ đi nghĩ lại một chuyện tồi tệ (phản tư) làm trầm cảm thêm. Nó còn củng cố cách nghĩ tiêu cực, làm yếu khả năng giải quyết vấn đề. Nó cản trở hành động thật, và làm cạn kiệt những người quanh bạn chịu lắng nghe. Nó liên quan đến lo âu, ăn uống mất kiểm soát, nghiện rượu và tự làm hại bản thân. +- Mức bằng chứng: B +- Nguồn: Nolen-Hoeksema, Wisco & Lyubomirsky (2008). Rethinking Rumination. Perspectives on Psychological Science. +- Ghi chú: Tổng hợp chỉ cho hướng, không có con số.「Đổi sang việc khác」là cách làm của chính tác giả. Văn học chống đỡ được chỉ một câu: trích hướng chú ý đi còn hơn cứ tiếp tục quanh quẩn. Còn đổi sang việc gì, không ai so sánh. +### 15. Xem câu「chắc chắn mọi chuyện sẽ tệ hơn」là triệu chứng, không phải sự thật + +- Chi phí: Không tốn tiền. Khó ở chỗ xem câu nghĩ đó là triệu chứng, chứ không phải là sự thật. +- Nói thẳng: Trong một nghiên cứu song sinh theo dõi 20 năm, mức bi quan mỗi cao hơn một bậc thì xác suất tử vong cùng kỳ cao hơn khoảng 13%, tử vong do bệnh tim mạch cao hơn khoảng 20%. Ngược lại, điểm lạc quan cao không mang lợi ích tương ứng. Vậy thứ cần xử lý là sự bi quan, không phải「chưa đủ tích cực」. +- Lợi ích: 2978 song sinh Úc trên 50 tuổi, theo dõi trung bình 20 năm, trong đó có 1068 người chết. Mỗi cao hơn 1 độ lệch chuẩn trên thang bi quan thì nguy cơ tử vong cùng kỳ cao hơn khoảng 13% (risk ratio 1.134). Khoảng tin cậy của con số này là 95% CI 1.065–1.207, P=8.85×10⁻⁵. Tử vong tim mạch cao hơn khoảng 20% (1.196, 1.045–1.368, P=0.0093). Không liên quan đến tử vong do ung thư. Điểm lạc quan cũng không có liên hệ rõ với tỷ lệ tử vong. +- Mức bằng chứng: B +- Nguồn: Whitfield JB, Zhu G, Landers JG, Martin NG (2020). Pessimism is associated with greater all-cause and cardiovascular mortality, but optimism is not protective. Scientific Reports, 10, 12609. +- Ghi chú:Tranh cãi. Đây chỉ là ghi nhận, không chia nhóm. Người ta sẽ nói「thật ra ốm trước nên mới bi quan」, tác giả phản hồi rằng nếu thật vậy thì tử vong ung thư cũng phải liên quan. Nhưng vẫn có thể còn yếu tố chưa đo. Lạc quan không có tác dụng bảo vệ, nên không cần ép mình「suy nghĩ tích cực」. Việc cần làm là đừng xem dự đoán bi quan là bản dự báo chính xác cho tương lai. +### 16. Giảm bớt các mối quan hệ làm bạn hao mòn, học cách từ chối yêu cầu không muốn nhận + +- Chi phí: Không tốn tiền. Ngắn hạn sẽ có ma sát quan hệ, có thể mất đi một phần các mối quan hệ. Nói lời từ chối cần khá nhiều ý chí. +- Nói thẳng: Một người làm bạn hao mòn gây cái mất lớn hơn, và ổn định hơn, cái mà một người ủng hộ bạn mang lại. Vậy hãy bớt đi trước, rồi mới thêm vào. +- Lợi ích: Khảo sát 120 phụ nữ góa chồng từ 60 đến 89 tuổi: sự tiếp xúc khiến bạn khó chịu có liên hệ với hạnh phúc tâm lý nhất quán và mạnh hơn sự tiếp xúc dễ chịu. Theo kết quả này, cái mất mà một người làm bạn hao mòn gây ra lớn hơn cái mà một người ủng hộ bạn mang lại. +- Mức bằng chứng: C +- Nguồn: Rook (1984). The negative side of social interaction: Impact on psychological well-being. Journal of Personality and Social Psychology. +- Ghi chú: Nhóm được hỏi khá đặc biệt, lại chỉ là bảng hỏi một lần, không thấy trước sau. Thứ đo là hạnh phúc, không phải sức lực. Đặt vào phần này là suy luận của tác giả. Mối quan hệ nào bị coi là「hao mòn」thì không có tiêu chuẩn khách quan, chỉ tự mình phán đoán. + +### 17. Chốt sẵn những quyết định giá trị thấp như mặc gì, ăn gì + +- Chi phí: Không tốn tiền. Lúc đầu tốn chút thời gian để định sẵn lựa chọn mặc định. +- Nói thẳng: Mục này không có bằng chứng chắc. Thí nghiệm gốc về「ra quyết định làm cạn ý chí」không tái hiện được trong hai lần làm lại quy mô lớn. Hai lần đó một lần hai nghìn người, một lần ba nghìn người, nhiều phòng thí nghiệm cùng làm, hiệu ứng cơ bản bằng không. Chốt sẵn mặc gì ăn gì vẫn đáng làm, vì nó tiết kiệm vài phút và vài lần do dự, chứ không phải「tiết kiệm ý chí」. +- Lợi ích: Không lượng hóa được. Thí nghiệm gốc cho rằng ra quyết định và tự kiểm soát cùng dùng chung một nguồn hạn chế (sự hao mòn bản ngã). Nhưng sau đó có hai lần làm lại quy mô lớn, cả hai đều đăng ký trước kế hoạch, nhiều phòng thí nghiệm làm lại theo đúng bản gốc, cả hai đều không ra cùng kết quả. Lần một là 23 phòng thí nghiệm, 2141 người, cỡ hiệu ứng d = 0.04, 95% CI [-0.07, 0.15] (khoảng tin cậy). Lần khác là 36 phòng thí nghiệm, 3531 người, d = 0.06. Chênh lệch nhỏ đến mức không tính được. +- Mức bằng chứng: C +- Nguồn: Baumeister, Bratslavsky, Muraven & Tice (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology. ; Hagger et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science. ; Vohs et al. (2021). A Multisite Preregistered Paradigmatic Test of the Ego-Depletion Effect. Psychological Science. +- Ghi chú:Tranh cãi.「Mệt vì quyết định」trong sách phổ thông được coi là kết luận chung, nhưng sau đó người khác làm lại theo đúng thí nghiệm gốc, phần lớn không ra cùng kết quả. Nó vẫn đáng làm vì gần như không tốn gì, còn thời gian tiết kiệm được là thật. Đừng mong nó cải thiện khả năng phán đoán của bạn vào buổi chiều. +### 18. Khi giận thì đi khỏi chỗ đó trước, coi đối phương như thời tiết chứ không như kẻ thù + +- Chi phí: Không tốn tiền. Lúc đó phải nhịn không cãi trả, quay lưng đi, rồi sau mới quyết định có xử lý hay không. Cần tập một chút. +- Nói thẳng: Giải thích chuyện vừa xảy ra theo góc khác, hoặc đứng từ góc đối phương mà nghĩ, cả hai đều hạ cảm xúc thật được, cách sau hiệu quả hơn. Nén lại không lộ ra thì không có tác dụng. Người trừng phạt người gây sự thì sau đó nghĩ đi nghĩ lại về chuyện đó lâu hơn; người không trả thù thì quên mau hơn. Người ta ước tính trước cao cái sướng mà trả thù mang lại. +- Lợi ích: Gộp 306 cặp so sánh thí nghiệm (phân tích tổng hợp): giải thích lại chuyện vừa xảy ra theo góc khác (đánh giá lại nhận thức) có cỡ hiệu ứng lên cảm xúc d = 0.36. Đứng từ góc đối phương thì d = 0.45. Chuyển hướng chú ý thì d = 0.27. Nén cảm xúc lại thì không hiệu quả (d = -0.04). Ba thí nghiệm khác cho thấy: người trừng phạt người gây sự thì sau đó nghĩ về đối phương lâu hơn; người không trừng phạt thì xoay chuyển nhanh hơn. Người ta ước tính trước cao cái sướng mà trả thù mang lại. +- Mức bằng chứng: B +- Nguồn: Webb TL, Miles E, Sheeran P (2012). Dealing with feeling: A meta-analysis of the effectiveness of strategies derived from the process model of emotion regulation. Psychological Bulletin. ; Carlsmith KM, Wilson TD, Gilbert DT (2008). The paradoxical consequences of revenge. Journal of Personality and Social Psychology. +- Ghi chú: Phần gộp nhiều nghiên cứu đo là cường độ cảm xúc, không phải sức lực, hiệu ứng từ nhỏ đến trung bình. Phần trả thù chỉ có một nhóm thí nghiệm, nên chấm B.「Con thuyền trống」là chuyện ngụ ngôn trong chương「Sơn Mộc」của Trang Tử: thuyền trống đâm tới thì không giận, trên thuyền có người thì lại giận, nhưng va chạm là một. Đó là cách so sánh, không phải bằng chứng. Nếu thật có tổn thất, hãy đi lối pháp lý ở phần 8 và phần 9, rẻ hơn nhiều so với bùng phát ngay tại chỗ. +### 19. Khi tâm trạng xuống, làm trước mấy việc đáng giá nhất: vận động, phơi nắng, ngủ đúng giờ, nói với ai đó, gọi 12356 + +- Chi phí: Không tốn tiền, hoặc tốn rất ít. Đi bộ nhanh, chạy bộ hay tập tạ vài lần mỗi tuần. Ra ngoài ban ngày để đón ánh sáng. Chốt giờ sinh hoạt. Khi không lội ra được thì gọi một cuộc điện thoại. +- Nói thẳng: Với trầm cảm, đi bộ nhanh và chạy bộ hiệu quả nhất, yoga và tập tạ kém hơn một chút, càng mạnh càng hiệu quả. Chiếu sáng hiệu quả rõ với trầm cảm theo mùa, và cũng hiệu quả với trầm cảm không theo mùa. Khóa tự giúp nhận thức hành vi trên mạng thì có tác dụng, lúc triệu chứng nặng hơn thì có người dẫn dắt sẽ tốt hơn. Khi không lội ra được, gọi 12356 trước. +- Lợi ích: Gộp 218 thử nghiệm đối chứng chia nhóm ngẫu nhiên, 14170 người (phân tích mạng). Nhóm đối chứng là những người cũng đang làm hoạt động khác. So với họ, đi bộ hoặc chạy bộ có cỡ hiệu ứng lên trầm cảm g = -0.62 (95% khoảng tin cậy -0.80 đến -0.45). Yoga là -0.55, tập tạ là -0.49. Càng mạnh càng hiệu quả. Gộp các nghiên cứu chiếu sáng: cỡ hiệu ứng 0.84 với trầm cảm theo mùa (0.60 đến 1.08), 0.53 với trầm cảm không theo mùa (0.18 đến 0.89). Còn một phân tích khác gộp dữ liệu từng bệnh nhân của các nghiên cứu. Nó xét liệu pháp nhận thức hành vi qua internet (khóa tự giúp trên mạng). Có người dẫn dắt hay không dẫn dắt đều tốt hơn nhóm đối chứng. Lúc trung bình đến nặng (PHQ-9 lớn hơn 9 điểm), có người dẫn dắt thì tốt hơn. 12356 là đường dây trợ giúp tâm lý toàn quốc, mở mỗi ngày không ít hơn 18 giờ. +- Mức bằng chứng: A +- Nguồn: Noetel M et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ. ; Golden RN et al. (2005). The efficacy of light therapy in the treatment of mood disorders: a review and meta-analysis of the evidence. American Journal of Psychiatry. ; Karyotaki E et al. (2021). Internet-Based Cognitive Behavioral Therapy for Depression: A Systematic Review and Individual Patient Data Network Meta-analysis. JAMA Psychiatry. ; Ủy ban Y tế Quốc gia (国家卫生健康委) (2024). Thông báo về việc dùng số điện thoại đường dây trợ giúp tâm lý thống nhất toàn quốc「12356」（国卫医政函〔2024〕259 号）. +- Ghi chú: Tất cả những thứ này đều là biện pháp hỗ trợ cho trầm cảm nhẹ đến trung bình. Trầm cảm nặng đến trung bình, hoặc có ý nghĩ tự sát thì đi khám bác sĩ, và gọi 12356 trước (xem mục 25 ở phần 1). Các thử nghiệm về vận động phần lớn số người ít, và không thể che giấu ai đang vận động như khi thử thuốc. Nghiên cứu chiếu sáng chỉ đến năm 2003, số người cũng ít。「Ngủ đúng giờ」「nói với ai đó」là kinh nghiệm của chính tác giả, không có tài liệu riêng. Đừng coi「vận động」là yêu cầu đặt lên người trầm cảm, đi được mười phút là được. +### 20. Coi cảnh sát, bác sĩ, nhân viên quầy là người làm việc theo quy tắc, đừng coi là vai diễn: thứ đẩy việc đi được là giấy tờ và thời hạn, không phải cảm xúc + +- Chi phí: Không tốn tiền. Cái đổi là kỳ vọng của bạn và thứ tự hành động. Lần đầu phải nén mong muốn「nói cho đến khi đối phương công nhận」. +- Nói thẳng: Cảnh sát, bác sĩ, nhân viên quầy đều là người làm việc theo quy tắc của vị trí, chịu sự quản lý của giấy tờ, thời hạn và thanh tra. Đối phương có thân thiện hay không gần như không liên quan đến việc việc của bạn có được làm hay không, hay mấy ngày xong. Vậy hãy xin biên nhận, ghi lại thời hạn, khiếu nại theo đúng kênh, đừng mong giảng道理 cho đối phương công nhận. +- Lợi ích: Đổi cách xử lý chuyện với cơ quan: từ「van xin, nổi giận, kể lể đi kể lể lại」sang「xin giấy tờ, ghi thời hạn, khiếu nại theo đúng kênh」. Thứ tiết kiệm được là thời gian cho việc nói đi nói lại và nghĩ đi nghĩ lại. Đây là điều chỉnh quan niệm một lần, không phải tiết kiệm thời gian mỗi ngày. Cái thật sự mang ra kiểm tra được là chính các thời hạn đó, vì chúng đều viết trong văn bản. Vụ việc hành chính an ninh tính từ ngày tiếp nhận, thời hạn giải quyết không quá 30 ngày; vụ việc trọng đại, phức tạp thì được cấp công an cao hơn một cấp phê duyệt để gia hạn 30 ngày. Vụ việc có người tố cáo mà quyết định không khởi kiện thì thông báo không khởi kiện phải được tống đạt trong 3 ngày. Người tố cáo trong 7 ngày có thể xin hòa giải lại, cơ quan công an ra quyết định trong 30 ngày. Không phục thì trong 7 ngày nữa xin xem xét lại ở cấp cao hơn. +- Mức bằng chứng: C +- Nguồn: Bộ Công an (公安部) (2020). Quy định về thủ tục giải quyết vụ án hình sự của cơ quan công an (公安机关办理刑事案件程序规定)（公安部令第 159 号修正后，第一百七十一、一百七十八、一百七十九条）. ; Bộ Công an (公安部) (2013/2019 sửa đổi). Quy định về thủ tục giải quyết vụ việc hành chính của cơ quan công an (公安机关办理行政案件程序规定)（公安部令第 125 号、第 149 号）. +- Ghi chú: Lý do chấm C: chính câu「nhiều người coi người sau quầy là một vai diễn, chứ không phải người đi làm」không có tài liệu nào dẫn được. Chỉ có nửa quy tắc dẫn được: thời hạn, giấy tờ và kênh khiếu nại đều có văn bản rõ. Toàn bộ con số trong mục này lấy từ hai quy định của Bộ Công an mà mục 39 ở phần 8 dẫn (quy trình sau khi gọi công an). Đi theo quy trình không xử lý được「đối phương thái độ kém」, cũng không bảo đảm chắc chắn làm được, cái đổi chỉ là kỳ vọng của bạn và bước đầu tiên. Nhân viên quầy ngân hàng, quầy dịch vụ công cũng vậy: thao tác của họ đều có dấu, cần người thứ hai kiểm, sai thì bị truy trách. Vậy việc nên làm là giữ phiếu, xin biên lai, khiếu nại theo đúng kênh, đừng phán bừa theo「hắn có lừa mình không」. Thứ tự ưu tiên hợp pháp đều ghi rõ ra chỗ sáng, ví dụ cấp cứu phân loại theo tình trạng bệnh, không theo đến trước đến sau (mục 8 ở phần 24, phân loại khám cấp cứu). Muốn cảm ơn bác sĩ đã cứu mình thì gửi thư cảm ơn và bảng đánh giá hài lòng, xem mục 12 ở phần 24. Muốn「đưa chút gì để người ta để tâm hơn」thì với người đang xử lý vụ việc đó là tội hối lộ, mà luật còn quy định nặng hơn, xem mục 40 ở phần 8. Gặp đối xử phân biệt rõ ràng thì khiếu nại, đừng coi là luật ngầm mà làm theo. + +### 21. Đừng coi「người khác sống thế nào」bài đọc mỗi ngày: giới hạn hoặc tắt các app đua xem hoạt động của người cùng trang lứa + +- Chi phí: Không tốn tiền. Giới hạn app hay ngừng dùng tài khoản đều chỉ vài phút. Khó là sau đó không mở lại, phải chống quán tính lâu dài. +- Nói thẳng: Người quen so sánh trên mạng với người sống tốt hơn thường tâm trạng kém hơn, tự trọng thấp hơn. Người tắt Facebook bốn tuần thì hạnh phúc tăng nhẹ, nhưng hiểu biết thời sự cũng giảm rõ. Người giới hạn ba nền tảng mạng xã hội, mỗi nền tảng 10 phút mỗi ngày, thì ba tuần sau cô đơn và trầm cảm đều giảm. +- Lợi ích: Gộp nhiều nghiên cứu (phân tích tổng hợp): 54 mẫu độc lập, 94 cỡ hiệu ứng, N = 36.583, dùng mô hình biến thiên ngẫu nhiên ba tầng. Tương quan trung bình giữa việc so sánh xã hội đi lên trên mạng và sự thích nghi tâm lý kém là r = 0.330 (95% CI 0.289–0.370, khoảng tin cậy). Tách theo từng kết quả. Cảm xúc tiêu cực mang tính đánh giá xã hội như sợ bị coi thường là 0.438, lo âu 0.382. Trầm cảm 0.306, hạnh phúc thấp 0.268, tự trọng thấp 0.263. Tuổi, bối cảnh văn hóa và năm thu thập dữ liệu đều không làm đổi độ mạnh của liên hệ này. Nghiên cứu theo dõi lâu dài cũng không cho kết quả mạnh hơn khảo sát tại cùng một thời điểm. Còn một thử nghiệm đối chứng chia nhóm ngẫu nhiên, người tham dự là 2.743 người dùng Facebook Mỹ. Trong đó 61% sẵn sàng chấp nhận việc tắt với mức dưới 102 đô la. Họ được chia ngẫu nhiên vào nhóm tắt bốn tuần và nhóm đối chứng. Việc tắt giải phóng cho người tham dự trung bình 60 phút mỗi ngày. Chỉ số chủ quan về hạnh phúc tăng 0.09 độ lệch chuẩn. Tách ra: hài lòng với cuộc sống 0.12, lo âu 0.10, trầm cảm 0.09, vui 0.08. Sau khi trừ phần may mắn do so sánh nhiều lần, những con số này vẫn tính. Tác giả có đưa một mức tham chiếu, là các can thiệp tâm lý làm người ta tích cực hơn. Loại can thiệp đó nâng hạnh phúc chủ quan khoảng +0.34 độ lệch chuẩn và hạ trầm cảm khoảng -0.23 độ lệch chuẩn. Vậy hiệu ứng này vào khoảng 25% đến 40% so với một can thiệp tâm lý thường quy. Cùng kỳ cái giá là: chỉ số hiểu biết thời sự giảm 0.19 độ lệch chuẩn, mức độ đối lập trên lập trường chính trị giảm 0.16 độ lệch chuẩn. Sau khi thí nghiệm kết thúc, mức dùng Facebook của nhóm tắt vẫn thấp hơn nhóm đối chứng 12 phút mỗi ngày (0.31 độ lệch chuẩn, mức giảm 23%). Còn một thử nghiệm đối chứng chia nhóm ngẫu nhiên khác, người tham dự là 143 sinh viên năm nhất, chia ngẫu nhiên thành hai nhóm. Một nhóm giới hạn Facebook, Instagram, Snapchat, mỗi nền tảng 10 phút mỗi ngày, nhóm kia dùng như bình thường. Ba tuần sau, nhóm giới hạn thì cô đơn và trầm cảm giảm rõ. Cả hai nhóm thì lo âu và sợ bỏ lọt đều thấp hơn lúc đầu. +- Mức bằng chứng: A +- Nguồn: Lei Y, Hu S, Sun Y, Zheng L (2026). "Looking up" linked to feeling down: a meta-analysis of online upward social comparison and psychological maladjustment. Frontiers in Psychology. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2020). The Welfare Effects of Social Media. American Economic Review, 110(3), 629–676. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2019). The Welfare Effects of Social Media. NBER Working Paper No. 25514 (các con số cụ thể mà mục này dẫn được đối chiếu từng chữ theo bản này, bản tóm tắt in không có con số) ; Hunt MG, Marx R, Lipson C, Young J (2018). No More FOMO: Limiting Social Media Decreases Loneliness and Depression. Journal of Social and Clinical Psychology, 37(10), 751–768. +- Ghi chú: Nửa「tương quan」chỉ cho thấy hai chuyện cùng xuất hiện, không đọc được thành chuyện nào gây ra chuyện nào, chính tác giả cũng nói cần thêm nghiên cứu theo dõi lâu dài. Nửa chia nhóm ngẫu nhiên nói được về quan hệ nhân quả, nhưng người tham dự là người dùng Facebook Mỹ trước bầu cử giữa kỳ 2018, đưa sang nền tảng Trung Quốc là lắp ghép gượng. Nghiên cứu Hunt chỉ có 143 sinh viên năm nhất, và chỉ có nghiên cứu đó. Cỡ hiệu ứng đều không lớn, đừng mong tắt app là giải quyết được vấn đề tâm trạng. Thứ cần chặn là loại dùng「xem người khác sống thế nào」; liên lạc với gia đình bạn bè, tra thông tin, tìm việc thì không tính. Nếu thật sự không giữ được thì bắt đầu từ tắt thông báo, tắt「người có thể quen」và tắt đề xuất ở trang đầu. Tính tổng thời gian màn hình xem mục 15 và 16 ở phần 4 (trần cứng cho video ngắn, không xem tin cuộn). Khi tâm trạng xuống làm gì trước xem mục 19 của phần này (vận động, phơi nắng, ngủ đúng giờ). Tìm lại cảm giác danh tính qua việc mua sắm xem mục 23 ở phần 6. Tiền bỏ thêm để「lên một bậc」xem mục 24 ở phần 6. +### 22. Khi thấy「ai cũng thấy mình mất mặt」thì chia đôi con số ước lượng đó + +- Chi phí: Không tốn tiền. Chỉ là tự hỏi thêm một câu lúc căng thẳng: họ thật sự nhìn thấy sao. +- Nói thẳng: Bước vào phòng có người mặc chiếc T-shirt đáng xấu hổ, người mặc nghĩ gần một nửa số người nhận ra ai được in trên áo, thực tế chưa đến một phần tư. Mặc chiếc T-shirt thấy tử tế thì con số ước lượng gấp sáu lần thực tế. Sau khi trò chuyện, người ta cũng普遍 đánh giá thấp mức đối phương thích mình. Vậy khi thấy「ai cũng nhớ mình mất mặt」, chia đôi vẫn còn cao. +- Lợi ích: Thí nghiệm một: cho người tham dự mặc T-shirt in hình Barry Manilow, bước vào phòng có người tham dự khác. Người mặc ước trung bình 46% người chứng kiến chỉ đúng người in trên áo. Người chứng kiến đúng thực tế là 23%, ước lượng恰 gấp đôi thực tế. Tính từng lượt, ước lượng trừ thực tế, trung bình cao hơn 23 điểm phần trăm (95% CI 9–38, khoảng tin cậy). Loại hai lượt chỉ có hai và ba người chứng kiến ra thì là 27 điểm phần trăm (11–43). Thí nghiệm hai: đổi sang mặc T-shirt do người tham dự tự chọn, tự thấy tích cực. Ước trung bình 48%, người chứng kiến đúng thực tế 8%, gấp sáu lần. Một loạt nghiên cứu khác có ba bối cảnh: người lạ ghép cặp trò chuyện, bạn cùng phòng năm nhất, hội trưởng thành cho người lớn. Kết quả là người ta đánh giá thấp mức đối phương thích mình, và cũng đánh giá thấp mức đối phương enjoyed cuộc trò chuyện. Tác giả gọi đó là「khoảng cách thích」. Nó có trong các cuộc trò chuyện dài ngắn khác nhau, và tồn tại hàng tháng giữa bạn cùng phòng năm nhất. +- Mức bằng chứng: B +- Nguồn: Gilovich T, Medvec VH, Savitsky K (2000). The spotlight effect in social judgment: An egocentric bias in estimates of the salience of one's own actions and appearance. Journal of Personality and Social Psychology, 78(2), 211–222. ; Boothby EJ, Cooney G, Sandstrom GM, Clark MS (2018). The Liking Gap in Conversations: Do People Like Us More Than We Think? Psychological Science, 29(11), 1742–1756. +- Ghi chú: Lý do chấm B: đều là nghiên cứu thí nghiệm trong phòng lab và nghiên cứu ở trường, bối cảnh là mặc quần áo và nói chuyện với người lạ. Chuyển sang「bị đồng nghiệp bàn tán」「nói sai trên họp」thì không có bằng chứng trực tiếp. Những con số này chỉ cho hướng ước cao và độ lớn đại khái, không dùng làm hệ số được. Ý kiến người khác cũng không phải thứ gì cũng quan trọng. Chỗ thật sự đang đánh giá bạn, như phỏng vấn, bảo vệ luận văn, đánh giá, báo cáo khám sức khỏe, thì đánh giá là thật, đã thế thì chuẩn bị. Cái cần giảm chỉ là cảm giác「mọi người đều đang nhìn mình」trong đời thường, chỗ đó thứ phải đối phó là chính thiên lệch ước lượng của bạn. +### 23. Xem câu「người khác đòi tôi hoàn hảo」là triệu chứng, không phải sự thật + +- Chi phí: Không tốn tiền. Khó ở chỗ xem câu nghĩ đó là thứ cần xử lý, chứ không phải là phép đo chính xác yêu cầu thật của người khác. +- Nói thẳng: Người luôn thấy「người quanh tôi đòi tôi hoàn hảo, không đạt thì sẽ bị phủ nhận」thì dễ có ý nghĩ tự tắt hơn. Mối liên hệ giữa「tôi tự đòi mình hoàn hảo」và ý nghĩ tự tắt yếu hơn nhiều, chỉ bằng khoảng một phần ba cách thứ nhất. Trừ những ý nghĩ đã có sẵn, cảm giác thứ nhất vẫn liên quan đến việc ý nghĩ tăng lên sau đó, chỉ là độ lớn不大. +- Lợi ích: Gộp nhiều nghiên cứu (phân tích tổng hợp): 45 nghiên cứu, N = 11.747, đối tượng là sinh viên đại học, sinh viên y, người lớn trong cộng đồng và bệnh nhân tâm thần, dùng mô hình biến thiên ngẫu nhiên. Sau đây là tương quan trung bình có trọng số với ý nghĩ tự tắt. Lo lắng hoàn hảo r = .28 (95% CI .24–.32, khoảng tin cậy; 42 nghiên cứu, 7936 người). Lo lắng hoàn hảo là một tập hợp, gồm完美主义 do xã hội quy định, lo mắc lỗi, nghi ngờ hành động của bản thân, khoảng cách giữa thực tế và tiêu chuẩn. Trong đó xã hội quy định tính riêng cũng là r = .28 (.25–.32, 30 nghiên cứu, 3640 người). Vươn tới hoàn hảo r = .10 (.07–.13, 31 nghiên cứu, 4588 người), nó là tập hợp của hoàn hảo tự hướng và tiêu chuẩn cá nhân. Cha mẹ phê bình r = .20 (.11–.29). Sang mối liên hệ với hành động tự sát. Lo lắng hoàn hảo r = .12 (.07–.17, 15 nghiên cứu, 5275 người). Hoàn hảo do xã hội quy định r = .19 (.08–.29, 8 nghiên cứu, 689 người). Phần theo dõi theo thời gian: sau khi trừ ý nghĩ tự sát đã có lúc đầu, hoàn hảo do xã hội quy định vẫn biến thiên cùng chiều với ý nghĩ tự sát ở lần theo dõi sau, độ lớn不大. Hoàn hảo tự hướng và hướng người khác đều không thấy liên hệ. +- Mức bằng chứng: B +- Nguồn: Smith MM, Sherry SB, Chen S, Saklofske DH, Mushquash C, Flett GL, Hewitt PL (2018). The perniciousness of perfectionism: A meta-analytic review of the perfectionism-suicide relationship. Journal of Personality, 86(3), 522–542. +- Ghi chú: Lý do chấm B: phần lớn nghiên cứu đưa vào là bảng hỏi một lần, không thấy trước sau. Độ mạnh tương quan ở mức trung bình (r khoảng .28), đại khái chỉ giải thích được vài phần trăm đến một phần mười chênh lệch. Thứ đo là ý nghĩ tự sát và hành động tự sát, không phải cái chết本身, nên quy mức lợi ích nhỏ với lý do「chỉ có điểm cuối thay thế」. Chính tác giả cũng nói nhóm lo lắng hoàn hảo và hành động tự sát này hình như có thiên lệch xuất bản. Thiên lệch xuất bản nghĩa là nghiên cứu kết quả không đẹp thì không được in. Bù phần nghiên cứu mất ước tính vào thì hướng kết luận không đổi. Tiêu chuẩn cao本身 không phải vấn đề, cũng không vì thế mà hạ yêu cầu. Khác nhau ở chỗ tiêu chuẩn đến từ đâu, và khi không đạt có thấy「mình thế là xong」hay không. Nó cũng như dự đoán bi quan ở mục 15 (xem câu「chắc chắn mọi chuyện sẽ tệ hơn」là triệu chứng), là thứ xử lý được như triệu chứng. Có ý nghĩ tự tắt thì gọi 12356 trước và chuyển phương tiện gây tử vong ra xa, xem mục 25 ở phần 1. Người quanh bạn bộc lộ ý nghĩ này thì bạn làm được gì, xem mục 15 ở phần 8. Sàng lọc trầm cảm cho trẻ xem mục 8 ở phần 30. +### 24. Khi giận đừng đập đồ, đấm bao cát, chạy một vòng để xả; hãy làm cơ thể chậm lại trước + +- Chi phí: Không tốn tiền. Lúc đang nóng, đổi sang vài phút làm cho tim chậm lại: thở chậm, ngồi xuống, yên một lát. Khó ở chỗ đúng lúc đó việc muốn nhất lại chính là xả ra. +- Nói thẳng: Đấm bao cát, đập đồ, đi chạy một vòng, những cách xả làm cơ thể kích thích hơn thì nhìn chung không hết giận. Hơi thở chậm, chánh niệm, thi định, những cách làm cơ thể yên lại thì ép được cơn giận và cái ham muốn xuống tay xuống chân. +- Lợi ích: Gộp 154 nghiên cứu (phân tích tổng hợp), tổng 184 mẫu, 10.189 người. Cách làm cơ thể bình tĩnh lại, như thở sâu, chánh niệm, thi định, làm giảm cơn giận và tính hung hăng. Cỡ hiệu ứng g = -0.63, 95% CI -0.82 đến -0.43 (khoảng tin cậy). Kết quả này ổn định ở người khác giới, chủng tộc, tuổi và văn hóa. Cách làm cơ thể kích thích hơn, như đấm bao cát, chạy bộ, đạp xe, nhìn chung không hiệu quả. Cỡ hiệu ứng của nó là g = -0.02, 95% CI -0.13 đến 0.09. Kết luận của tác giả là: xả cơn giận, đi chạy một vòng, đều không phải cách hết giận hiệu quả. +- Mức bằng chứng: A +- Nguồn: Kjærvik SL, Bushman BJ (2024). A meta-analytic review of anger management activities that increase or decrease arousal: What fuels or douses rage? Clinical Psychology Review. +- Ghi chú: Thứ đo là cơn giận và tính hung hăng, không phải sức lực. Ghi口径 vào thời gian là cách xếp của tác giả: ít cãi một trận thì ít dọn một lần. Cách dùng được ngay lúc đó xem mục 9 ở phần 22 (thở vòng), còn mục 18 của phần này (khi giận thì đi khỏi chỗ đó trước). Chạy bộ không phải vô dụng. Nhìn dài hạn thì nó hiệu quả với tâm trạng xuống, xem mục 7 ở phần 22 (tâm trạng kém thì đi bộ hoặc chạy). Chỉ là không hợp để hết giận ngay tại chỗ. Lúc đang nóng đừng lái xe, đừng gửi tin nhắn, đừng quyết định. Nếu có người đang đánh bạn, rời đi trước rồi gọi công an. Lúc đó thứ cần xử lý không phải cảm xúc của bạn, xem mục 43 ở phần 8 (bị bạo lực gia đình). + +### 25. Có một chuyện đè nặng trong lòng, có thể viết suốt mấy ngày liền tư tưởng và cảm xúc của mình ra, nhưng đừng trông chờ nó chữa bệnh + +- Chi phí: Không tốn tiền. Cách làm thường thấy trong nghiên cứu là viết 3 đến 4 lần, mỗi lần 15 đến 20 phút. Viết một chuyện khiến bạn khó chịu, viết ra những suy nghĩ và cảm xúc sâu nhất của bạn +- Nói thẳng: Viết những chuyện đè nặng lòng ra thì trung bình có ít lợi ích, nhưng rất nhỏ. Kết luận của các tổng hợp cũng không nhất quán, có hai tổng hợp tính ra là không có hiệu quả. Nó gần như không tốn tiền, có thể thử. Vấn đề tình cảm nặng thì đừng dùng nó thay cho việc đi khám bác sĩ +- Lợi ích: Một phân tích tổng hợp gồm 146 nghiên cứu chia nhóm ngẫu nhiên (phân tích tổng hợp). Người tham gia nghiên cứu viết hoặc nói ra những chuyện quan trọng với họ. Hiệu quả trung bình là dương, cũng có ý nghĩa, nhưng rất nhỏ, cỡ hiệu quả r = .075. Một tổng hợp khác chỉ thu thử nghiệm đối chứng ngẫu nhiên, 30 thử nghiệm gộp lại thì cả sức khoẻ thể chất lẫn sức khoẻ tinh thần đều không có hiệu quả có ý nghĩa. 16 thử nghiệm ngẫu nhiên trên bệnh nhân ung thư và người đang hồi phục gộp lại thì cả ba mặt tinh thần, thể chất, chất lượng sống đều không có ý nghĩa. Trong đó mặt tinh thần là g = 0.04, 95% CI -0.06 đến 0.14 (khoảng tin cậy) +- Mức bằng chứng: B +- Nguồn: Frattaroli J (2006). Experimental disclosure and its moderators: a meta-analysis. Psychological Bulletin. ；Mogk C, Otte S, Reinhold-Hurley B, Kröner-Herwig B (2006). Health effects of expressive writing on stressful or traumatic experiences - a meta-analysis. Psycho-Social Medicine. ；Zachariae R, O'Toole MS (2015). The effect of expressive writing intervention on psychological and physical health outcomes in cancer patients--a systematic review and meta-analysis. Psycho-Oncology. +- Ghi chú:Tranh cãi. Phe ủng hộ là tổng hợp 146 nghiên cứu, hiệu quả dương nhưng rất nhỏ. Phe phản đối là hai tổng hợp chỉ thu thử nghiệm đối chứng ngẫu nhiên, đều không tính ra hiệu quả có ý nghĩa. Số lần viết và thời lượng mỗi lần ghi ở cột Chi phí đến từ thiết kế thử nghiệm đầu tiên, nêu trong phần tóm tắt của bài tổng hợp của Mogk. Nó gần như không tốn tiền, cũng không tốn nhiều thời gian, một mình làm được. Khi không muốn tìm ai tâm sự, hoặc bên cạnh không còn ai để tâm sự, có thể dùng nó. Nó không đo sức lực, thước đo ghi là thời gian là do tác giả xếp loại. Viết mà thấy càng lúc càng khó chịu thì dừng lại, gọi 12356, xem mục 25 ở phần 1 (12356). diff --git "a/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" "b/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" new file mode 100644 index 00000000..e9c0548e --- /dev/null +++ "b/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" @@ -0,0 +1,167 @@ +[← 回总目录](../README.md) + +# 4. 不要浪费时间 + +本节只算时间账。看的是每天、每周、每年能省回多少小时，或者一件事按时做完的比例能提高多少。这里不把时间换算成寿命，也不换算成钱。涉及钱的地方只给算法，不给结论。本节多数条目的证据是单独一项研究，或者是时间利用调查的总量统计，所以等级以 B、C 为主。 + +### 1. Viết「định làm」thành「mấy giờ, ở đâu, gặp gì thì làm gì」 + +- Chi phí: Không tốn tiền. Mỗi việc mất 1 đến 2 phút, viết ra một câu「nếu……thì……」. Ví dụ「thứ Ba 8 giờ sáng ngồi xuống bàn thì viết phần dẫn dắt trước」. +- Nói thẳng: Viết lại「tôi sẽ làm việc này」thành「thứ Ba 8 giờ sáng ngồi xuống bàn thì viết phần dẫn dắt trước」. Gộp 94 phép kiểm định, người viết kiểu này hoàn thành mục tiêu với tỷ lệ rõ ràng cao hơn, hiệu ứng cỡ vừa đến lớn. Chỉ sửa cách viết một câu, không tốn tiền, không tốn sức. +- Lợi ích: Gộp 94 phép kiểm định độc lập, người viết trước ý định thành「nếu……thì……」có tỷ lệ đạt mục tiêu cao hơn. Cách viết này trong tâm lý học gọi là ý định thực thi. Cỡ hiệu ứng d = 0.65, xếp vào mức vừa đến lớn. Có ba chỗ nó tác dụng: bắt tay vào làm, giữ được khi bị gián đoạn, rút ra khỏi việc chắc chắn không xong. +- Mức bằng chứng: A +- Nguồn: Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. +- Ghi chú: Cách này chỉ hiệu quả với việc bạn thật sự muốn làm. Nếu bản thân mục tiêu bạn không muốn thì viết thêm bao nhiêu câu cũng vô ích. Thêm nữa, các nghiên cứu này lấy mẫu chủ yếu là học sinh và sinh viên, xét chủ yếu thói quen sức khoẻ, chuyển sang công việc chưa chắc còn hiệu quả như vậy. + +### 2. Trước khi bắt tay, viết điều kiện dừng + +- Chi phí: Không tốn tiền. Lúc bắt tay dành 5 phút, viết rõ「đến ngày nào nào đó mà chưa đạt chỉ số nào đó thì dừng」. Chỗ khó là đến ngày đó phải nhận, dừng tay trước mặt người ta thì hơi mất thể diện. +- Nói thẳng: Trước khi bắt tay, viết sẵn một câu: đến ngày nào nào đó mà chưa đạt chỉ số nào đó thì dừng. Gộp nhiều nghiên cứu, cách viết này thật sự giúp người ta rút ra khỏi việc chắc chắn không xong. Thời gian tiết kiệm được là toàn bộ số thời gian bạn còn phải đổ thêm vào đó. +- Lợi ích: Gộp nhiều nghiên cứu về ý định thực thi, kết quả là: câu「nếu……thì……」định trước cũng giúp người ta rút ra khỏi việc chắc chắn không xong. Cách nói trong nguyên bản là disengagement from failing courses of action. Thời gian tiết kiệm được là số thời gian bạn vốn dĩ còn tiếp tục bỏ vào đó. +- Mức bằng chứng: B +- Nguồn: Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. +- Ghi chú: Điều kiện dừng phải định lúc vừa bắt tay, khi tâm trạng còn bình tĩnh. Định sau sẽ bị chi phí đã bỏ vào dẫn lệch. Khi cân nhắc chỉ nhìn phần đầu tư tương lai và phần thu về tương lai, xem mục 3. Thêm nữa,「sẵn sàng rút ra」chỉ là một kết luận phụ trong bản tổng hợp nhiều nghiên cứu, chưa có ai tính riêng nó tiết kiệm được bao nhiêu thời gian. + +### 3. Khi quyết định có tiếp tục hay không, chỉ nhìn phần sẽ đầu tư và phần sẽ thu về, không nhìn phần đã bỏ + +- Chi phí: Không tốn tiền. Chỗ khó là phải chấp nhận một đằng: thời gian đã bỏ vào trước đó mất rồi, không lấy lại được. +- Nói thẳng: Tiền và thời gian đã bỏ vào khiến người ta càng muốn làm tiếp. Gộp 98 nhóm dữ liệu, xu hướng này tồn tại ổn định. Nên khi quyết định có tiếp tục hay không, chỉ tính phần phải bỏ thêm từ nay và phần còn thu được. Phần đã tiêu mất thì không tính vào. +- Lợi ích: Tiền, sức lực và thời gian đã đầu tư khiến người ta càng muốn làm tiếp. Trong tâm lý học gọi đây là hiệu ứng chi phí chìm. Gộp 98 nhóm dữ liệu, phân tích xác nhận xu hướng này tồn tại ổn định. Tránh xu hướng đó thì bạn sẽ không tiếp tục đổ thêm thời gian chỉ vì「không muốn lãng phí」. +- Mức bằng chứng: B +- Nguồn: Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35, 124–140. ; Roth, S., Robbert, T., & Straus, L. (2015). On the sunk-cost effect in economic decision-making: A meta-analytic review. Business Research, 8(1), 99–138. +- Ghi chú: Thí nghiệm rạp hát của Arkes và Blumer làm như sau. Rạp hát đại học Ohio bán vé cả mùa, ngẫu nhiên chào bán ba mức giá cho người mua: giá gốc 15 đô la, rẻ hơn 2 đô la, rẻ hơn 7 đô la. Nửa đầu mùa, nhóm mua giá gốc xem trung bình 4.11 buổi, hai nhóm kia là 3.32 và 3.29 buổi. Nửa sau mùa thì ba nhóm không còn khác biệt rõ. Roth và cộng sự thấy hiệu ứng này nhạt dần theo thời gian, người lớn tuổi chịu ảnh hưởng ít hơn. + +### 4. Ước lượng thời hạn thì dựa vào thời gian thực tế của việc cùng loại trước đây, không suy ra từ kế hoạch + +- Chi phí: Không tốn tiền. Trước khi ước lượng, dành vài phút xem lại việc cùng loại lần trước thực tế mất bao lâu. +- Nói thẳng: Sinh viên ước lượng 33.9 ngày để viết xong luận văn, thực tế mất 55.5 ngày, chỉ 29.7% hoàn thành đúng thời hạn mình đã ước lượng. Nhóm kia được yêu cầu nhớ lại việc cùng loại lần trước mất bao lâu rồi mới ước lượng, tỷ lệ hoàn thành đúng hạn tăng từ 29.3% lên 60.0%. +- Lợi ích: Sinh viên dự định 33.9 ngày để hoàn thành luận văn, thực tế mất 55.5 ngày. Chỉ 29.7% hoàn thành đúng hạn theo ước lượng của mình. Nhà nghiên cứu yêu cầu nhóm kia nối kinh nghiệm quá khứ với việc đang làm trước khi ước lượng. Tỷ lệ hoàn thành đúng hạn của nhóm này tăng từ 29.3% ở nhóm đối chứng lên 60.0%. Kiểu lạc quan ước lượng ngắn hơn thời hạn thật không còn xuất hiện ở nhóm này. +- Mức bằng chứng: B +- Nguồn: Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. ; Flyvbjerg, B. (2006). From Nobel Prize to project management: Getting risks right. Project Management Journal, 37(3), 5–15. ; Halkjelsvik, T., & Jørgensen, M. (2012). From origami to software development: A review of studies on judgment-based predictions of performance time. Psychological Bulletin, 138(2), 238–271. +- Ghi chú: Nghiên cứu của Buehler chỉ lấy mẫu sinh viên đại học Canada, là một nghiên cứu đơn lẻ và số người ít, 37 người và khoảng 40 người mỗi nhóm. Ước lượng theo số liệu lịch sử của dự án cùng loại được gọi là dự báo theo lớp tham chiếu, cách làm cụ thể trong kỹ thuật xem ở Flyvbjerg. Halkjelsvik tổng hợp nhiều nghiên cứu rồi chỉ ra rằng ước lượng ngắn hơn thường gặp hơn ước lượng dài hơn, nhưng chênh lệch bao nhiêu thay đổi rất nhiều theo loại việc và theo phương pháp nghiên cứu. + +### 5. Gửi đề nghị trước cuộc họp, không có đề nghị thì không họp; họp được đứng thì cứ đứng + +- Chi phí: Không tốn tiền. Người tổ chức dành 5 đến 10 phút trước cuộc họp để viết đề nghị. Họp đứng thì cần người tham dự phối hợp. +- Nói thẳng: Cuộc họp có đề nghị gửi trước khiến người tham dự dễ thấy cuộc họp đó hữu ích. Trong thí nghiệm, họp ngồi dài hơn họp đứng 34%, còn chất lượng quyết định mà hai cách họp đưa ra thì không khác nhau. +- Lợi ích: Hai khảo sát có lần lượt 958 và 292 người. Kết quả là: cuộc họp có đề nghị gửi trước khiến người tham dự dễ cho rằng cuộc họp có tác dụng, mối liên hệ này đứng vững về mặt thống kê. Một thí nghiệm khác có 55 nhóm họp ngồi và 56 nhóm họp đứng, họp ngồi dài hơn 34%. Chất lượng quyết định mà hai cách họp đưa ra không khác nhau. +- Mức bằng chứng: B +- Nguồn: Leach, D. J., Rogelberg, S. G., Warr, P. B., & Burnfield, J. L. (2009). Perceived meeting effectiveness: The role of design characteristics. Journal of Business and Psychology, 24(1), 65–76. ; Bluedorn, A. C., Turban, D. B., & Love, M. S. (1999). The effects of stand-up and sit-down meeting formats on meeting outcomes. Journal of Applied Psychology, 84(2). +- Ghi chú: Thí nghiệm về họp đứng làm nhiệm vụ nhóm 5 người trong phòng thí nghiệm. Không thể chuyển nguyên kết luận này sang họp dài và hội nghị lớn. Leach đo là người tham dự tự cảm thấy hiệu quả ra sao, không phải khối lượng việc đã làm được. + +### 6. Bớt số cuộc họp, chuyện nào giải quyết bằng văn bản bất đồng bộ được thì khỏi họp + +- Chi phí: Không tốn tiền. Cần đồng nghiệp và cấp trên phối hợp. Chỗ khó là viết rõ một đoạn văn bản tốn sức hơn nói một câu. +- Nói thẳng: Trong ngày càng họp nhiều, người ta càng mệt và càng thấy việc trên tay mình nhiều lên. Bỏ đi một cuộc họp, thời gian tiết kiệm được đúng bằng thời lượng của cuộc họp đó, không khấu trừ gì. +- Lợi ích: Hai khảo sát lấy mẫu nhân viên làm việc trên 35 giờ mỗi tuần, lần lượt 676 và 304 người. Thời gian bị họp chiếm mất ảnh hưởng tới thái độ với công việc và hạnh phúc như thế nào thì không cố định, tùy vào hai điều: cuộc họp có họp tốt không, và công việc ở vị trí này phụ thuộc người khác nhiều đến đâu. Nghiên cứu khác bảo người ta ghi nhật ký liên tục một tuần, kết quả là trong ngày càng họp nhiều lần thì mức mệt và lượng việc cảm nhận được càng cao. Thời gian tiết kiệm được bằng đúng tổng thời lượng cuộc họp mà bạn đã bỏ đi. +- Mức bằng chứng: B +- Nguồn: Rogelberg, S. G., Leach, D. J., Warr, P. B., & Burnfield, J. L. (2006). "Not another meeting!" Are meeting time demands related to employee well-being? Journal of Applied Psychology, 91(1). ; Luong, A., & Rogelberg, S. G. (2005). Meetings and more meetings: The relationship between meeting load and the daily well-being of employees. Group Dynamics: Theory, Research, and Practice, 9(1). +- Ghi chú:Tranh cãi. Rogelberg 2006 thấy quan hệ giữa số cuộc họp và hạnh phúc không nghiêng hẳn về một phía. Với vị trí mà công việc phụ thuộc cao vào người khác, họp nhiều lại không phải là điều xấu. Thêm nữa, chưa có nghiên cứu nào so trực tiếp「chuyển sang trao đổi bằng văn bản bất đồng bộ」với việc họp, mục này là suy luận của tác giả. + +### 7. Tách việc lớn thành các việc con rồi mới ước lượng thời gian, rồi mới bắt tay + +- Chi phí: Không tốn tiền. Bản thân việc tách mất 5 đến 15 phút. +- Nói thẳng: Khi ước lượng thời hạn, người ta không tự động tách việc ra rồi tính từng việc một. Liệt kê các việc con trước khi ước lượng thì sẽ ước lượng ngắn hơn một chút. Sau khi tách, mỗi bước nhỏ đều làm được ngay, nên cũng khó trì hoãn hơn. +- Lợi ích: Khi ước lượng thời hạn, người ta không tự tách một việc có nhiều nhánh thành từng khối. Ví dụ viết tổng quan tài liệu, viết phần thảo luận, sắp xếp danh mục trích dẫn, vốn dĩ đã là vài việc riêng. Liệt kê các việc con trước khi ước lượng giúp bớt mắc lỗi ước lượng ngắn hơn thời hạn thật, lỗi đó gọi là ảo tưởng lập kế hoạch. Sau khi tách, mỗi bước nhỏ thành một việc có thể bắt tay ngay, sự trì hoãn cũng giảm theo. Những yếu tố dự báo trì hoãn mạnh nhất trong nghiên cứu gồm mức độ ghét việc đó và bản thân việc bị đẩy lùi. +- Mức bằng chứng: B +- Nguồn: Kruger, J., & Evans, M. (2004). If you don't want to be late, enumerate: Unpacking reduces the planning fallacy. Journal of Experimental Social Psychology, 40(5), 586–598. ; Steel, P. (2007). The nature of procrastination: A meta-analytic and theoretical review of quintessential self-regulatory failure. Psychological Bulletin, 133(1). +- Ghi chú: Tách đến mức nào là vừa: lấy tiêu chí「mỗi việc con làm xong được trong một lần ngồi」, tách quá nhỏ thì bản thân việc tách cũng tốn thời gian. Thêm nữa, nghiên cứu của Kruger và Evans mới chỉ đối chiếu được hướng từ phần tóm tắt, các phần trăm cụ thể thì chưa đối chiếu tới nguyên bản. + +### 8. Việc nào không có hạn chót từ bên ngoài thì tự đặt một ngày + +- Chi phí: Không tốn tiền. Chỗ khó là đã đặt thì phải nhận, đến ngày đó mà không coi ra gì thì coi như chưa đặt. +- Nói thẳng: Khi có hạn chót do người khác đặt, 80.6% sinh viên làm xong trước ngày đó. Còn theo đúng thời gian mình dự định thì chỉ 38.7%. Khi nào xong phần lớn do hạn chót quyết định, chứ không mấy liên quan tới ước lượng của bản thân. Nên việc nào không có hạn chót thì tự đặt một ngày, rồi giữ lời. +- Lợi ích: Sinh viên có hạn chót từ bên ngoài thì 80.6% hoàn thành trước hạn. Còn những người hoàn thành đúng khoảng thời gian mình dự định chỉ có 38.7%. Khi nào xong gần như đi theo hạn chót, hệ số tương quan r = 0.82, con số này càng gần 1 thì càng bám sát. Với dự định của bản thân thì quan hệ rất yếu. +- Mức bằng chứng: B +- Nguồn: Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. +- Ghi chú: Nói ngày bạn tự đặt cho người khác biết thì nó gần với ngày do người khác đặt hơn. Nghiên cứu xác nhận là hạn chót do người khác đặt có tác dụng. Loại tự đặt cho mình, như canh giờ hay đồng hồ pomodoro, chưa ai đo trực tiếp, chỗ này là suy ra theo. + +### 9. Hiểu việc trì hoãn là đang né thứ khó chịu, đừng hiểu là chưa đủ cố gắng + +- Chi phí: Không tốn tiền. Đổi hướng phán đoán: hỏi trước chỗ nào của việc đó làm bạn khó chịu, rồi mới nghĩ cách. +- Nói thẳng: Việc trì hoãn không liên quan nhiều đến sự lười hay tính cách. Những gì cho biết rõ nhất một người có trì hoãn hay không là bản thân việc đó làm người đó khó chịu đến đâu, người đó có nghĩ mình làm được không, và dễ bị thứ trước mắt kéo đi thế nào. Nên cái có ích là làm cho việc bớt khó chịu và làm bước đầu tiên nhỏ lại, chứ không phải mắng mình một tràng rồi cố làm. +- Lợi ích: Một phân tích tổng hợp tổng hợp 691 hệ số tương quan cho thấy vài yếu tố có quan hệ mạnh và ổn định với việc trì hoãn: mức độ khiến người ta ghét công việc, còn cách ngày hạn chót, cảm giác tự hiệu quả (nghĩ rằng mình làm được), tính dễ kích động, cùng tính cẩn trọng và những thành phần bên dưới như khả năng tự kiểm soát, dễ mất tập trung, tính có trật tự và động lực thành tích. Còn tính thần kinh, tính nổi loạn và tìm kiếm kích thích chỉ có quan hệ rất yếu với việc trì hoãn. Tác giả dùng lý thuyết động lực thời gian để giải thích nhóm kết quả này: sức hút hiện tại của một việc giảm nhanh khi phần thưởng ở xa hơn. +- Mức bằng chứng: A +- Nguồn: Steel P (2007). The nature of procrastination: a meta-analytic and theoretical review of quintessential self-regulatory failure. Psychological Bulletin, 133(1), 65–94. +- Ghi chú: Cỡ lợi ích ghi là「vừa」, vì các nghiên cứu này chỉ chỉ ra hướng, chưa đo được tiết kiệm bao nhiêu giờ. Hành động cụ thể nằm ở mục 1 của phần này (viết thành「mấy giờ, ở đâu, gặp gì thì làm gì」), mục 7 (tách việc lớn thành việc con), mục 8 (đặt một ngày cho việc không có hạn chót bên ngoài). Đây là bản tổng hợp của các nghiên cứu tương quan, không xác định được cái nào dẫn tới cái nào. + +### 10. Đặt sẵn thứ mình cần dùng, đẩy xa thứ mình không muốn đụng, đừng trông chờ nhịn tại chỗ + +- Chi phí: Không tốn tiền, mỗi lần vài phút. Đưa điện thoại sang phòng bên cạnh, bày sách và máy tính ra bàn. +- Nói thẳng: Đồ ở xa thêm một chút thì người ta cầm lên ít đi. Cùng loại đồ ăn nhẹ đặt xa hơn thì lượng ăn vào giảm rõ rệt, hiệu ứng cỡ vừa. Bày ít lựa chọn hơn thì số người cầm tay tới cũng giảm rõ rệt. Cách loại này không cần bạn nhịn tại chỗ, vài phút là xong bố trí. +- Lợi ích: Một tổng quan Cochrane gồm 24 thử nghiệm chia nhóm ngẫu nhiên. Đặt thức ăn xa hơn thì lượng ăn vào giảm, sai khác trung bình chuẩn hoá −0.60 (95% CI −0.84 đến −0.36, đây là khoảng tin cậy; 12 nghiên cứu, 1098 người, bằng chứng chắc chắn thấp), hiệu ứng cỡ vừa. Giảm số lựa chọn thì mức giảm khi chọn thức ăn đó rõ hơn, sai khác trung bình chuẩn hoá −1.13 (95% CI −1.90 đến −0.37, 3 nghiên cứu, 154 người, bằng chứng chắc chắn thấp). Phân tích hồi quy cho thấy: đặt càng xa và lựa chọn càng đơn thì hiệu ứng càng mạnh. +- Mức bằng chứng: B +- Nguồn: Hollands GJ, Carter P, Anwer S, et al. (2019). Altering the availability or proximity of food, alcohol, and tobacco products to change their selection and consumption. Cochrane Database of Systematic Reviews, 9, CD012573. +- Ghi chú: Lý do xếp mức B: cả 24 nghiên cứu mà tổng quan này đưa vào đều về thức ăn, không có mục nào về rượu hay thuốc lá; 14 nghiên cứu làm trong phòng thí nghiệm và đều đến từ nước có thu nhập cao. Nên chuyển「đẩy xa thì ít chạm vào」sang điện thoại và trò chơi là suy nối, không có thử nghiệm trực tiếp nào. Riêng việc đưa điện thoại ra khỏi tầm nhìn thì có người đã đo trực tiếp, xem mục 1 ở phần 3 (tắt thông báo không cần thiết, khi làm việc thì đưa điện thoại ra khỏi tầm nhìn). + +### 11. Muốn bắt bản thân làm được thì đặt tiền vào, nhưng trước hết hãy hỏi mình có chịu đặt không + +- Chi phí: Phương án dạng ký quỹ đòi trước một khoản tiền của chính mình, làm không được thì mất. Trong thí nghiệm, khoản ký quỹ là 150 đô la. +- Nói thẳng: Muốn ép bản thân làm được, có thể làm xong thì nhận thưởng, hoặc đặt trước một khoản ký quỹ, làm xong thì hoàn lại rồi cộng thêm thưởng. Trong số người chịu tham gia cả hai phương án, nhóm đặt ký quỹ làm được nhiều hơn 13 điểm phần trăm. Nhưng chỉ 13.7% chịu đặt ký quỹ, còn 90% chịu nhận thưởng, nên ký quỹ chỉ có tác dụng với người chịu đặt. +- Lợi ích: Một thử nghiệm chia nhóm ngẫu nhiên ở Mỹ trên 2538 người, chia ngẫu nhiên nhân viên và người thân của họ vào bốn phương án khuyến khích hoặc chăm sóc thường lệ. Tỷ lệ chấp nhận phương án thưởng (làm được thì thưởng khoảng 800 đô la) là 90.0%, phương án ký quỹ (đặt trước 150 đô la, cộng thêm 650 đô la thưởng) là 13.7%. Tỷ lệ bỏ thuốc duy trì sau sáu tháng: bốn phương án khuyến khích nằm trong khoảng 9.4% đến 16.0%, chăm sóc thường lệ 6.0%. Phương án thưởng 15.7%, phương án ký quỹ 10.2%. Nhưng trong nhóm chỉ tính 13.7% người「chịu tham gia cả hai phương án」, tỷ lệ bỏ thuốc sau sáu tháng của phương án ký quỹ cao hơn phương án thưởng 13.2 điểm phần trăm (95% CI 3.1 đến 22.8, đây là khoảng tin cậy). +- Mức bằng chứng: A +- Nguồn: Halpern SD, French B, Small DS, et al. (2015). Randomized trial of four financial-incentive programs for smoking cessation. New England Journal of Medicine, 372(22), 2108–2117. +- Ghi chú: Thử nghiệm này đo việc bỏ thuốc lá, chuyển sang viết luận văn hay tập thể dục thì là suy nối, nên cỡ lợi ích ghi là「vừa」. Khi chọn phương án, xem trước bản thân mình có thật sự dùng nó không. Cách phần đông số không chịu chấp nhận thì dù tác dụng mạnh đến đâu cũng vô ích với họ. Nên cứ chọn loại mình chịu dùng, đặt nhỏ hơn còn hơn không đặt. Cách bỏ thuốc lá thì xem mục 3 ở phần 2 (thuốc cai thuốc lá). + +### 12. Thói quen mới thì tính theo tháng, đừng tính theo「21 ngày」 + +- Chi phí: Không tốn tiền. Điều cần là lặp lại cùng một hành động mỗi ngày trong cùng một bối cảnh, kéo dài vài tháng. +- Nói thẳng: Câu「21 ngày để hình thành một thói quen」không có cơ sở. Tính ra từ một nghiên cứu theo dõi 12 tuần, thời gian để một hành động trở thành làm không cần nghĩ là 18 ngày với người nhanh và 254 ngày với người chậm. Sau hai ba tuần mà chưa thấy nhẹ hơn là bình thường. Bỏ lỡ một ngày giữa chừng cũng không xoá sạch tiến độ. +- Lợi ích: 96 tình nguyện viên, mỗi người chọn một hành động liên quan đến ăn, uống hoặc vận động, làm mỗi ngày trong một bối cảnh cố định (chẳng hạn「sau bữa sáng」), kéo dài 12 tuần, mỗi ngày tự chấm điểm mức độ tự động hoá. Dữ liệu của 82 người đủ dùng để phân tích, mô hình vừa khớp trên dữ liệu của 62 người. Số ngày cần để đạt 95% trần mức tự động hoá của mỗi người nằm trong khoảng 18 ngày đến 254 ngày. Nghiên cứu còn phát hiện việc bỏ lỡ một lần thực hiện không ảnh hưởng thực chất tới quá trình hình thành thói quen. +- Mức bằng chứng: B +- Nguồn: Lally P, van Jaarsveld CHM, Potts HWW, Wardle J (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998–1009. +- Ghi chú: Chỉ có một nghiên cứu này, 96 người, và các hành động đều đơn giản (chẳng hạn ăn một phần trái cây sau bữa ăn). Việc phức tạp cần bao lâu, như chạy ba lần mỗi tuần hay viết mỗi ngày, thì chưa ai đo. Nên sau hai tuần mà vẫn thấy khó thì không thể coi là bằng chứng thất bại. Muốn sự lặp lại thật sự xảy ra thì gắn hành động với một bối cảnh cố định, xem mục 1 của phần này (viết thành「mấy giờ, ở đâu, gặp gì thì làm gì」). + +### 13. Khi việc trì hoãn đã ảnh hưởng tới cuộc sống thì dùng tài liệu tự giúp có thử nghiệm ngẫu nhiên hỗ trợ, không bắt buộc phải tìm nhà trị liệu + +- Chi phí: Một cuốn sách tự giúp vài chục tệ, có loại miễn phí. Phải bỏ ra khoảng mười tuần, mỗi tuần đọc một phần và làm một bài tập. +- Nói thẳng: Việc trì hoãn nặng đến mức gây đau đớn thì chữa được. Luyện theo tài liệu tự giúp nhận thức-hành vi trong mười tuần, dù có người hướng dẫn hay tự làm hoàn toàn, đều rõ ràng hơn là không làm gì cả. Có người hướng dẫn và không có không thấy khác nhau, nên tự làm theo cũng có tác dụng. +- Lợi ích: 150 người được chia ngẫu nhiên thành ba nhóm: tự giúp có nhà trị liệu hướng dẫn, tự giúp không hướng dẫn, và nhóm đối chứng danh sách chờ. Sau mười tuần, cỡ hiệu ứng giữa các nhóm trên thang đo trì hoãn thuần túy là d = 0.70 ở nhóm có hướng dẫn (95% CI 0.29–1.10, đây là khoảng tin cậy), d = 0.50 ở nhóm không hướng dẫn (0.10–0.90); trên thang đo trì hoãn phi lý tính lý là 0.81 (0.40–1.22) và 0.69 (0.29–1.09) tương ứng. Tỷ lệ đạt cải thiện ở mức có ý nghĩa lâm sàng: nhóm có hướng dẫn 31.3% đến 40.0%, nhóm không hướng dẫn 24.0% đến 36.0%. Hai nhóm không khác nhau có ý nghĩa ở bất kỳ kết quả nào. +- Mức bằng chứng: A +- Nguồn: Rozental A, Forsell E, Svensson A, Andersson G, Carlbring P (2015). Internet-based cognitive-behavior therapy for procrastination: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 83(4), 808–824. +- Ghi chú: Người tham gia là người lớn Thụy Điển, tài liệu dùng bằng tiếng Thụy, tiếng Trung không có sẵn bộ cùng loại, nên cỡ lợi ích ghi là「vừa」. Các thành phần trong tài liệu đều là những thứ quen thuộc: kích hoạt hành vi, thí nghiệm hành vi, kiểm soát kích thích, phần giải thích về động lực và cách làm việc. Riêng kiểm soát kích thích nằm ở mục 10 của phần này (đẩy xa thứ mình không muốn đụng). Nếu việc trì hoãn đi kèm tâm trạng trầm chuyên rõ rệt hay lo âu thì trước hết làm theo mục 19 ở phần 3 (vận động, phơi nắng, ngủ đúng giờ, nói với ai, gọi 12356). + +### 14. Dùng mức lương theo giờ của bạn để quyết định việc nhà nào thuê ngoài + +- Chi phí: Phải bỏ ra một khoản tiền thuê ngoài, nhiều hay ít tuỳ dịch vụ bạn mua. Thêm nữa phải bỏ thời gian một lần để tính rõ mức lương theo giờ của mình. +- Nói thẳng: Người Trung Quốc bỏ trung bình 1 giờ 17 phút mỗi ngày vào việc nhà. Trong số người thật sự tự tay làm trong ngày, là 1 giờ 59 phút. Người bỏ tiền để mua thời gian có mức hài lòng với cuộc sống cao hơn. Cùng một khoản tiền đó, bỏ vào mua thời gian thì lúc đó vui hơn so với bỏ vào mua đồ. +- Lợi ích: Cư dân Trung Quốc làm việc nhà trung bình 1 giờ 17 phút mỗi ngày. Chỉ tính những người có làm việc nhà, trung bình 1 giờ 59 phút. Đây là khảo sát năm 2024, bao phủ 38.500 hộ, 107.000 người. Một nghiên cứu khác lấy mẫu 6271 người ở bốn quốc gia, người bỏ tiền mua thời gian có mức hài lòng với cuộc sống cao hơn. Các nhà nghiên cứu còn làm thí nghiệm trong đời thực: cùng một khoản tiền đó, bỏ vào mua thời gian thì hạnh phúc lúc đó cao hơn so với bỏ vào mua đồ vật. +- Mức bằng chứng: B +- Nguồn: Whillans, A. V., Dunn, E. W., Smeets, P., Bekkers, R., & Norton, M. I. (2017). Buying time promotes happiness. Proceedings of the National Academy of Sciences. ; 国家统计局 (Cục Thống kê Quốc gia) (2024). 第三次全国时间利用调查公报（第二号）. +- Ghi chú: Trước hết tính mức lương theo giờ, công thức là thu nhập sau thuế một tháng ÷ số giờ làm việc thực tế trong một tháng. Rồi xem việc nhà này thuê ngoài tốn bao nhiêu. Nếu đơn giá thuê ngoài thấp hơn mức lương theo giờ của bạn, và thời gian tiết kiệm được bạn thật sự dùng để kiếm tiền hoặc nghỉ ngơi, thì thuê ngoài. Mức lương theo giờ và giá thuê ngoài khác nhau với từng người, kết luận phải tự tính. Thêm nữa, Whillans đo là sau khi bỏ tiền mua thời gian thì người ta có vui không, không phải tiết kiệm được bao nhiêu giờ. + +### 15. Đặt một giới hạn cứng cho video ngắn và lướt màn hình không mục đích + +- Chi phí: Không tốn tiền. Hành động chỉ mất vài phút: tắt thông báo đẩy, xoá ứng dụng, hoặc đặt cho ứng dụng một giới hạn thời gian mỗi ngày. Chỗ khó là kiềm tay, nhịn không bấm vào. +- Nói thẳng: Người Trung Quốc lên mạng trung bình 5 giờ 37 phút mỗi ngày. Trong số người thật sự lên mạng trong ngày, là 6 giờ 3 phút. Năm 2018, với cùng cách tính đó, con số này mới chỉ 2 giờ 42 phút. Đây là con số lớn nhất trong phần này. Mỗi ngày bớt một giờ thì một năm thêm 365 giờ. +- Lợi ích: Cư dân Trung Quốc lên mạng trung bình 5 giờ 37 phút mỗi ngày. Chỉ tính những người thật sự lên mạng trong ngày, trung bình 6 giờ 3 phút. Người dùng mạng trong ngày chiếm 92.9%. Đây là số liệu năm 2024. Năm 2018, với cùng cách tính, chỉ có 2 giờ 42 phút. Đây là con số lớn nhất trong phần này. Mỗi ngày bớt một giờ trong đó tức là mỗi năm thêm 365 giờ. Theo Trung tâm Thông tin Mạng lưới Trung Quốc (中国互联网络信息中心), tính đến tháng 6 năm 2025, người dùng mạng lên mạng trung bình 30.6 giờ mỗi tuần, nhiều hơn nửa năm trước 1.9 giờ. Người dùng video ngắn có 1.068.000.000 người, chiếm 95.1% người dùng mạng. +- Mức bằng chứng: C +- Nguồn: 国家统计局 (Cục Thống kê Quốc gia) (2024). 第三次全国时间利用调查公报（第二号）. ; 国家统计局 (Cục Thống kê Quốc gia) (2019). 2018年全国时间利用调查公报. ; 中国互联网络信息中心 (Trung tâm Thông tin Mạng lưới Trung Quốc) (2025). 第 56 次《中国互联网络发展状况统计报告》. +- Ghi chú: Thời gian lên mạng gồm cả công việc, học tập và chuyện xã giao, không phải lúc nào cũng là lãng phí. Xếp mức C vì không có nghiên cứu nào tính được「sau khi hạn chế thì tiết kiệm bao nhiêu và số thời gian đó đi đâu」, trong tay chỉ có số liệu tổng. Riêng nhóm「lướt xem người khác sống thế nào」thì có thử nghiệm chia nhóm ngẫu nhiên đo được tiết kiệm bao nhiêu và tâm trạng thay đổi ra sao, xem mục 21 ở phần 3. + +### 16. Không xem truyền hình và tin tức chạy liên tục, gom thông tin cần xem vào một khung giờ cố định + +- Chi phí: Không tốn tiền. Chỗ khó là bỏ thói quen mở ra tự nhiên, và có thể bỏ lỡ một số tin tức tức thời. +- Nói thẳng: Người Mỹ từ 15 tuổi trở lên xem 2.6 giờ truyền hình mỗi ngày, chiếm một nửa toàn bộ thời giờ rảnh. Người Trung Quốc xem trung bình 1 giờ 40 phút mỗi ngày, người cao tuổi 75 đến 84 tuổi xem 3 giờ 16 phút. Đổi phần tin cần xem sang một khung giờ cố định gộp lại thì thời gian tiết kiệm được là cả khối đó. +- Lợi ích: Người Mỹ từ 15 tuổi trở lên xem 2.6 giờ truyền hình mỗi ngày. Toàn bộ thời gian rảnh của họ là 5.2 giờ, xem truyền hình chiếm một nửa. Đây là số liệu năm 2025. Cư dân Trung Quốc xem truyền hình trung bình 1 giờ 40 phút mỗi ngày, người 75 đến 84 tuổi là 3 giờ 16 phút. Đây là số liệu năm 2018. +- Mức bằng chứng: C +- Nguồn: U.S. Bureau of Labor Statistics (2026). American Time Use Survey — 2025 Results. ; 国家统计局 (Cục Thống kê Quốc gia) (2019). 2018年全国时间利用调查公报. +- Ghi chú: Trong tay chỉ có số liệu tổng. Chưa nghiên cứu nào tính được sau khi bớt xem truyền hình và tin tức thì thời gian tiết kiệm được đi đâu. Cách「gom lại xem một lúc」là kinh nghiệm của tác giả. + +### 17. Bỏ thời gian một lần để học phím tắt và tự động hoá của phần mềm hay dùng + +- Chi phí: Không tốn tiền. Phải bỏ một lần vài tiếng đến vài chục tiếng để học. Chỗ khó là thời gian đầu mới bắt đầu còn chậm hơn dùng chuột. +- Nói thẳng: Phần lớn người dùng Word nhiều năm vẫn bấm vào biểu tượng trên thanh công cụ. Trong thí nghiệm so sánh giữa các nhóm, phím tắt nhanh hơn cả menu lẫn biểu tượng. Học một lần là chi phí một lần, sau đó mỗi thao tác mỗi ngày đều đang hoàn vốn. Làm cùng loại việc càng lâu thì hoàn vốn càng nhiều. +- Lợi ích: Nghiên cứu xem 251 người dùng Word có kinh nghiệm, phần lớn rất ít dùng phím tắt nhanh hơn, vẫn chủ yếu bấm biểu tượng trên thanh công cụ. Thí nghiệm so sánh giữa các nhóm xác nhận trong ba cách là menu, biểu tượng và phím tắt thì phím tắt nhanh nhất. Thời gian tiết kiệm được cộng dồn từng chút theo số lần thao tác mỗi ngày. Người làm cùng loại việc lâu dài thì lợi ích lớn nhất. +- Mức bằng chứng: B +- Nguồn: Lane, D. M., Napier, H. A., Peres, S. C., & Sándor, A. (2005). Hidden costs of graphical user interfaces: Failure to make the transition from menus and icon toolbars to keyboard shortcuts. International Journal of Human-Computer Interaction. +- Ghi chú: Chỉ học những thao tác bạn dùng hơn mười lần mỗi ngày, học cả mọi thứ là một kiểu lãng phí khác. Thí nghiệm đo hiệu suất đó chỉ có 6 người tham gia, phần tóm tắt cũng không nêu số giây tiết kiệm được mỗi thao tác. + +### 18. Khi chọn nơi ở thì đặt thời lượng đi làm lên đầu tiêu chí, rút ngắn thời gian đi một chiều + +- Chi phí: Chi phí cao. Có thể phải trả tiền thuê cao hơn, hoặc chuyển sang một căn nhỏ hơn. Bản thân việc chuyển nhà cũng tốn thời gian. +- Nói thẳng: Người đi làm càng xa thì càng cảm thấy mình sống không tốt. Phần thiệt hại đó không được bù lại trong lương hay trong nhà ở. Mỗi chiều bớt 30 phút thì một tuần tiết kiệm được 5 giờ. +- Lợi ích: Cư dân Trung Quốc bỏ trung bình 50 phút mỗi ngày vào việc đi lại. Chỉ tính những người thật sự ra ngoài trong ngày, là 1 giờ 2 phút. Đây là số liệu năm 2024. Dữ liệu Đức theo dõi cùng một nhóm người trong thời gian dài cho thấy người đi làm càng xa thì hạnh phúc chủ quan (cảm giác mình có sống tốt không) càng thấp một cách ổn định. Hơn nữa phần thiệt hại đó không được bù lại trong thu nhập hay nhà ở. Một tổng quan gom nhiều nghiên cứu lại cũng xác nhận, thời gian đi làm càng dài thì người ta càng không hài lòng với việc đi làm, điều này không liên quan tới dùng phương tiện gì. Mỗi chiều bớt 30 phút thì mỗi tuần tiết kiệm khoảng 5 giờ. +- Mức bằng chứng: B +- Nguồn: Stutzer, A., & Frey, B. S. (2008). Stress that doesn't pay: The commuting paradox. Scandinavian Journal of Economics, 110(2), 339–366. ; Chatterjee, K., et al. (2020). Commuting and wellbeing: A critical overview of the literature with implications for policy and future research. Transport Reviews, 40(1), 5–34. ; 国家统计局 (Cục Thống kê Quốc gia) (2024). 第三次全国时间利用调查公报（第二号）. +- Ghi chú:Tranh cãi. Chatterjee và cộng sự, sau khi gom nhiều nghiên cứu lại, chỉ ra rằng chưa thấy quan hệ ổn định giữa thời lượng đi làm và mức hài lòng chung với cuộc sống. Họ cho rằng nhìn chung người ta vẫn dùng được công việc hoặc nhà ở tốt hơn để bù lại thiệt hại của việc đi làm xa. Số liệu hoạt động đi lại thì tính mọi chuyến đi, không chỉ việc đi làm. Ngoài ra, nếu làm việc giờ vạch ngoài giờ cao điểm được thì cùng một đoạn đường đó cũng đi được nhanh hơn. diff --git "a/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" "b/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" new file mode 100644 index 00000000..bc2fe9ff --- /dev/null +++ "b/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" @@ -0,0 +1,391 @@ +[← 回总目录](../README.md) + +# 5. 不要浪费钱 + +这一节只算钱：能省下多少，每年收益差多少，一笔钱长期平均下来能拿回多少。这些不换算成寿命或时间。涉及投资的条目只讲怎么算账，只用核实过的历史数据，都不构成投资建议。 + +### 1. Tắt hết tự động gia hạn, đến hạn thì tự gia hạn + +- Chi phí: Không tốn tiền. Duyệt một lượt danh sách「自动扣款」trong Alipay, WeChat, Apple và Android, mỗi lần 10 đến 20 phút. Sau đó mỗi kỳ gia hạn, bạn phải tự bấm thêm một lần. +- Nói thẳng: Số tiền tiết kiệm được là tiền của các gói thuê bao bạn đã lâu không dùng mà vẫn bị trừ hằng tháng. Duyệt một lượt danh sách tự động trừ tiền trong Alipay, WeChat, Apple và Android mất chưa tới 20 phút. Theo quy định, trước khi trừ tiền, người bán phải cảnh báo bạn rõ ràng, và phải đưa cho bạn một cửa sổ huỷ bất cứ lúc nào, cửa sổ đó phải dễ thấy và dễ bấm. +- Lợi ích: Số tiền tiết kiệm bằng tổng số tiền của các gói thuê bao bạn đã không dùng mà vẫn bị trừ phí. Mỗi người một khác, không có con số chung. Cơ quan quản lý yêu cầu người bán phải cảnh báo bạn rõ ràng trước khi gia hạn, và phải đưa cho bạn lựa chọn「简便的随时取消」, huỷ không phải trả thêm phí (Trung Quốc, áp dụng từ 2021) +- Mức bằng chứng: C +- Nguồn: Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2021, sửa đổi 2025). Biện pháp quản lý giao dịch qua mạng（总局令第 37 号，第十八条）. +- Ghi chú: Với dịch vụ bạn thực sự dùng thường xuyên, trả cả năm thường rẻ hơn trả từng tháng, giữ nguyên tự động gia hạn lại là hợp lý. Cần tắt đi là những gói bạn đã quên từ lâu mà vẫn bị trừ tiền. + +### 2. Mỗi năm từ tháng Ba đến tháng Sáu làm một lần tổng kết thuế thu nhập cá nhân, khai đủ các khoản trừ bổ sung chuyên ngành + +- Chi phí: Không tốn tiền. Làm trên app「个人所得税」, mười mấy phút là xong. +- Nói thẳng: Những người sau đây làm xong thường được hoàn tiền: cả năm chỉ đi làm mấy tháng, giữa năm đã đổi việc, hoặc chưa khai các khoản trừ như giáo dục con cái, vay mua nhà, tiền thuê nhà, nuôi dưỡng người cao tuổi. Làm trên điện thoại mất mười mấy phút. Nếu thu nhập tổng hợp của bạn trong một năm dưới 12 vạn tệ, hoặc số thuế phải nộp bù dưới 400 tệ, bạn có thể không làm. +- Lợi ích: Gom lương công ăn việc, thu nhập từ làm thuê, tiền nhuận bút và tiền quyền sử dụng thành tựu trí tuệ thành một bảng kê chung. Nộp nhiều thì hoàn lại cho bạn, nộp thiếu thì phải bù. Có hai trường hợp không bắt buộc phải làm: thu nhập tổng hợp trong năm của bạn không quá 12 vạn tệ, hoặc số thuế phải nộp bù không quá 400 tệ. Hai điều này do thông báo năm 2023 của Tổng cục Thuế quốc gia quy định, hiệu lực đến ngày 31 tháng 12 năm 2027. Ngược lại, những người dưới đây làm xong thường được hoàn tiền: cả năm chỉ có mấy tháng có thu nhập, giữa năm đã đổi việc, có khoản trừ như giáo dục con cái, vay mua nhà, tiền thuê nhà, nuôi dưỡng người cao tuổi mà chưa khai +- Mức bằng chứng: A +- Nguồn: Tổng cục Thuế quốc gia (国家税务总局) (2025). Biện pháp quản lý việc tổng kết và nộp thuế thu nhập cá nhân tính trên thu nhập tổng hợp（第六条）kèm phần giải thích. ；Bộ Tài chính, Tổng cục Thuế (2023). Thông báo về việc tiếp tục thực hiện chính sách liên quan đến tổng kết và nộp thuế thu nhập cá nhân tính trên thu nhập tổng hợp. +- Ghi chú: Người phải nộp thuế bù cũng phải làm, không làm là vi phạm pháp. Khoản trừ bổ sung chuyên ngành có bảy mục: giáo dục con cái, đào tạo tiếp, điều trị bệnh hiểm nặng, lãi vay mua nhà, tiền thuê nhà, nuôi dưỡng người cao tuổi, chăm sóc trẻ dưới 3 tuổi. Trong đó lãi vay mua nhà và tiền thuê nhà chỉ được chọn một, không thể khai cả hai. + +### 3. Quỹ tích góp nhà ở không chỉ dùng để mua nhà: thuê nhà, sửa sang, phí quản lý nhà ở đều rút được, từ ngày 20 tháng 9 năm 2026 áp dụng quy định mới + +- Chi phí: Không tốn tiền. Đi làm thủ tục rút một lần là xong. +- Nói thẳng: Quỹ tích góp nhà ở không phải chỉ khi mua nhà mới dùng được. Quy định mới liệt kê chín trường hợp được rút, trong đó có thuê nhà, sửa sang chính căn nhà bạn ở, và trả phí quản lý nhà ở. Bạn nộp đơn, trung tâm quỹ tích góp phải trả lời bạn trong 3 ngày là có được rút hay không. +- Lợi ích: Quy định sau sửa đổi liệt kê chín trường hợp được rút. Trong đó gồm trả tiền thuê nhà, sửa sang chính căn nhà mình ở, trả phí quản lý nhà ở của chính căn nhà mình ở. Trung tâm quản lý quỹ tích góp phải tính từ ngày tiếp nhận đơn của bạn, rồi quyết định cho rút hoặc không cho rút trong 3 ngày +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). Quyết định của Hội đồng Nhà nước về việc sửa đổi《Quy định quản lý quỹ tích góp nhà ở》（国务院令第 844 号，2026 年 7 月 31 日国务院第 93 次常务会议通过，「自2026年9月20日起施行」）: Điều 二十四 sau khi sửa đổi「职工有下列情形之一的，可以提取职工住房公积金账户内的存储余额：（一）支付房租的；（二）购买、建造、翻建、大修自住住房的；（三）偿还购房贷款本息的；（四）装修自住住房的；（五）支付自住住房物业费的；（六）离休、退休的；（七）完全丧失劳动能力，并与单位终止劳动关系的；（八）出境定居的；（九）国务院批准的其他住房消费情形。」— người lao động gặp một trong chín trường hợp sau thì được rút số dư tiền còn lại trong tài khoản quỹ tích góp nhà ở của mình：（一）đang trả tiền thuê nhà；（二）mua, xây, xây lại, sửa lớn nhà ở của mình；（三）trả gốc và lãi khoản vay mua nhà；（四）sửa sang nhà ở của mình；（五）trả phí quản lý nhà ở của chính nhà mình ở；（六）nghỉ hưu, về hưu；（七）mất hoàn toàn năng lực lao động và đã chấm dứt quan hệ lao động với đơn vị；（八）ra nước ngoài định cư；（九）các trường hợp tiêu dùng nhà ở khác được Hội đồng Nhà nước phê duyệt.；Điều 二十五「住房公积金管理中心应当自受理申请之日起3日内作出准予提取或者不准提取的决定」— trung tâm quản lý quỹ tích góp nhà ở phải quyết định cho rút hay không cho rút trong 3 ngày kể từ ngày tiếp nhận đơn；Điều 十八, tỷ lệ đóng「均不得低于职工上一年度月平均工资的5％」— không được thấp hơn 5％ mức lương trung bình hằng tháng của người lao động trong năm trước. +- Ghi chú: Số tiền trong tài khoản là lương của chính bạn, cộng thêm phần đơn vị đóng thêm. Để yên số tiền đó thì nó chỉ sinh lãi ở mức lãi suất rất thấp. Quy định còn ghi: khi người lao động chết, hoặc bị tuyên bố đã chết, người thừa kế và người được tặng có thể rút số dư tài khoản. Hồ sơ phải nộp gồm những gì, cụ thể làm thế nào, mỗi địa phương do trung tâm quản lý quỹ tích góp nhà ở địa phương quy định. Trước khi đi, hãy tra hướng dẫn thủ tục của trung tâm nơi bạn. + +### 4. Mỗi năm tính lại gói điện thoại và băng thông một lần, không dùng tới thì hạ bậc, nhà mạng không chuyển được thì khiếu nại + +- Chi phí: Không tốn tiền. Kiểm tra một lần hoá đơn, mười mấy phút. Chỗ khó là giữ được nhân viên không hạ bậc cho bạn. +- Nói thẳng: Gói cũ bạn đang dùng thường đắt hơn gói đang bán ra hiện nay vài chục tệ. Một tháng vài chục tệ, một năm là vài trăm tệ ném đi không. Theo quy định, không có lý do chính đáng thì nhà mạng không được từ chối việc bạn chuyển mạng, cũng không được kéo dài không làm, và càng không được riêng cho người chuyển mạng một bộ giá riêng. Nó cứ nhất định giữ, bạn cứ khiếu nại. +- Lợi ích: Quy chế của Bộ Công nghiệp và Thông tin ghi rõ, nhà mạng không được「无正当理由拒绝、阻止、拖延向用户提供携号转网服务」. Cũng không được「采取拦截、限制等技术手段影响携号转网用户的通信服务质量」. Cũng không được「为携号转网用户设置专项资费方案和营销方案」. Gói cũ thường đắt hơn gói đang lưu hành vài chục tệ. Không đổi suốt, nghĩa là mỗi năm vứt đi vài trăm tệ +- Mức bằng chứng: A +- Nguồn: Bộ Công nghiệp và Thông tin (工业和信息化部) (2019). Quy chế quản lý dịch vụ chuyển mạng giữ số thuê bao（工信部信管〔2019〕242 号）. +- Ghi chú: Trước hết xem trong app của cửa hàng nhà mạng lượng dữ liệu và số phút gọi bạn đã dùng thật trong sáu tháng gần nhất, rồi chọn gói theo đúng lượng dùng đó. Nhân viên nói「降档得到营业厅办」「合约期内不能改」thì bạn yêu cầu cơ sở bằng văn bản. Không có thì gọi 12300 để khiếu nại. + +### 5. Không mua xổ số + +- Chi phí: Không tốn tiền. Không mua là tiết kiệm tiền. +- Nói thẳng: Theo quy định, xổ số kiểu lotto và xổ số dạng số phải dành ra làm tiền thưởng ít nhất 50% doanh thu bán hàng, xổ số kép là 51%. Nghĩa là mỗi 1 tệ bạn bỏ ra, dài hạn trung bình chỉ lấy lại 0.51 tệ, lỗ ổn định 49%. Năm 2024 cả nước bán 623.486 tỷ tệ xổ số, trong đó 161.031 tỷ tệ đi thẳng vào quỹ công ích, phần này không nằm trong tiền thưởng. +- Lợi ích: Bộ Tài chính quy định tỷ lệ tiền thưởng tối thiểu so với doanh thu bán hàng: 50% với xổ số kiểu lotto và dạng số, 65% với xổ số cào, 73% với xổ số dự đoán thể thao. Xổ số kép chi 51% doanh thu làm tiền thưởng, cứ bán 100 tệ thì chỉ lấy 51 tệ làm thưởng. Quy ra trên người của bạn, mỗi 1 tệ bỏ vào, dài hạn trung bình chỉ lấy lại 0.51 tệ, lỗ 49%. Năm 2024 cả nước bán 623.486 tỷ tệ xổ số, trong đó 161.031 tỷ tệ vào quỹ công ích, không quay lại người mua (Trung Quốc, 2024) +- Mức bằng chứng: A +- Nguồn: Bộ Tài chính (财政部) (2015). Thông báo về việc quy định và tăng cường quản lý tỷ lệ cấu thành quỹ xổ số（财综〔2015〕94 号）. ；Bộ Tài chính (2025). Thông báo số 8 năm 2025 của Bộ Tài chính (tình hình huy động và phân bổ quỹ công ích xổ số năm 2024). ；Quy tắc trò chơi xổ số kép của Xổ số Phúc lợi Trung Quốc（第十四条）. +- Ghi chú: Dài hạn trung bình chắc chắn lỗ, điều này do hệ thống đặt sẵn, không liên quan tới may rủi. Mua cũng được, nhưng chỉ nên coi như khoản chi giải trí mỗi lần 2 tệ, không phải khoản đầu tư. + +### 6. Thấy「高收益」「保本」「稳赚」thì đi luôn + +- Chi phí: Không tốn tiền. Chỉ cần dừng lại một chút lúc nghe mấy chữ đó. +- Nói thẳng: Chính cơ quan giám sát nói: lãi suất trên 6% phải đặt dấu hỏi, trên 8% là nguy hiểm, trên 10% thì phải chuẩn bị mất sạch vốn gốc. Sản phẩm quản lý tài sản chính quy đã không được phép hứa giữ vốn và giữ lợi nhuận. Tiền mất vì tham gia huy động tiền phi pháp thì theo quy định do bạn tự chịu, xảy ra chuyện thì không ai đỡ cho bạn. +- Lợi ích: Năm 2018, chủ tịch cũ của Ủy ban Ngân hàng và Bảo hiểm có một lời tại diễn đàn Lujiazui. Nguyên văn:「收益率超过 6% 的就要打问号，超过 8% 的就很危险，10% 以上就要准备损失全部本金」. Quy chế mới về quản lý tài sản yêu cầu tổ chức tài chính「不得承诺保本保收益」, đồng thời yêu cầu「出现兑付困难时，金融机构不得以任何形式垫资兑付」. Vì thế sản phẩm quản lý tài sản chính quy đã không còn chuyện giữ vốn. Thiệt hại khi tham gia huy động tiền phi pháp「由集资参与人自行承担」, xảy ra chuyện thì không ai đền cho bạn (Trung Quốc, 2018, 2021) +- Mức bằng chứng: A +- Nguồn: Quách Thụ Thanh (郭树清) (2018). Phòng ngừa và giải quyết rủi ro tài chính, nỗ lực vượt qua những cổng then chốt (bài phát biểu tại Diễn đàn Lujiazui lần thứ 10, trên trang Cục Quản lý và Giám sát Tài chính quốc gia):「在打击非法集资过程中，努力通过多种方式让人民群众认识到，高收益意味着高风险，收益率超过6%的就要打问号，超过8%的就很危险，10%以上就要准备损失全部本金」. ；Cục Quản lý Tài chính địa phương thành phố Thiên Tinh đăng lại. ；Ngân hàng Nhân dân Trung Quốc và cơ quan khác (2018). Ý kiến chỉ đạo về việc quy định công tác quản lý tài sản của các tổ chức tài chính（银发〔2018〕106 号，第二条、第六条）. ；Hội đồng Nhà nước (2021). Quy chế phòng ngừa và xử lý huy động tiền phi pháp（国务院令第 737 号，第二十五条）. +- Ghi chú: Ba mốc 6%, 8%, 10% là người giám sát vạch theo kinh nghiệm, không phải điều luật. Câu nói đó được đưa ra khi nói về việc đánh triệt huy động tiền phi pháp, và nhắm vào sản phẩm quản lý tài sản và công ty đầu tư hứa trả cao. Không phải lời khuyên đầu tư. + +### 7. Không dùng khoản trả tối thiểu trên thẻ tín dụng, không mở trả góp hay vay tiêu dùng để mua + +- Chi phí: Không tốn tiền. Bạn phải để dành đủ tiền mặt trả toàn bộ trước ngày đến hạn. Chỗ khó là nhịn không dùng trả góp để dàn nhẹ một khoản chi lớn. +- Nói thẳng: Khoản trả tối thiểu không phải ưu đãi, đó là một khoản vay tính lãi theo ngày, quy ra cả năm khoảng 18%. Dùng nó thì hạn miễn lãi mất ngay, phần vốn gốc chưa trả phải chịu lãi từ đầu đến cuối. Xem lãi suất thì chỉ tin con số tính theo năm. Ngân hàng Nhân dân Trung Quốc đã yêu cầu bên phát hành phải hiển thị rõ lãi suất tính theo năm, không được chỉ ghi「日利率」. +- Lợi ích: Trước năm 2020, lãi suất thấu chi thẻ tín dụng có cả trần và sàn: trần là「日利率万分之五」, sàn bằng 0.7 lần trần. Năm phần nghìn một ngày quy ra cả năm là 18.25%, đó là lãi đơn: tính 18.25% một năm thì lãi không cộng dồn vào vốn gốc. Từ năm 2021, bên phát hành và người giữ thẻ tự thống nhất giá. Dùng khoản trả tối thiểu rồi thì phần vốn gốc chưa trả bị tính lãi đầy đủ mỗi ngày, đồng thời hạn miễn lãi mất hiệu lực. Ngân hàng Nhân dân Trung Quốc đã yêu cầu bên phát hành thể hiện rõ lãi suất quy đổi cả năm,「不得仅展示日利率、日还款额」. Hãy xem con số tính theo năm rồi hãy quyết định (Trung Quốc, thông báo năm 2020) +- Mức bằng chứng: C +- Nguồn: Ngân hàng Nhân dân Trung Quốc (中国人民银行) (2020). Thông báo về việc đẩy mạnh cải cách thị trường hoá lãi suất thấu chi thẻ tín dụng（银发〔2020〕327 号）. +- Ghi chú: Con số 18.25% là nhân thẳng「万分之五 × 365」, chưa tính lãi kép. Nếu cộng dồn theo tháng thì còn cao hơn. Có ngân hàng cho chiết khấu, hãy theo lãi suất quy đổi cả năm ghi trên hoá đơn của chính bạn. Khi thật sự không xoay được dòng tiền, hãy tìm kênh chính quy có lãi suất thấp hơn để vay trước, đừng để khoản trả tối thiểu cộng dồn lãi. +### 8. Không tặng thưởng người livestream, không nạp tiền vào game, không đặt hàng bốc đầu + +- Chi phí: Không tốn tiền. Dành mấy phút làm ba việc: tắt thanh toán không cần mật mã, đặt hạn mức cho từng giao dịch và cho từng ngày, xoá các phương thức thanh toán đã liên kết trong ứng dụng. Chỗ khó là nhịn ngay tại chỗ không bấm vào. +- Nói thẳng: Người chơi game ở Trung Quốc trung bình mỗi người một năm chi khoảng 480 tệ vào game, tiền đã chi thì cơ bản không lấy lại được. Luật chỉ cho hoàn tiền khi người chưa thành niên tiêu mà chưa có sự đồng ý của cha mẹ, người đã thành niên không có điều đó. Cách tiết kiệm công sức nhất là phòng trước: tắt thanh toán không cần mật mã, đặt hạn mức cho từng giao dịch và từng ngày, gỡ liên kết phương thức thanh toán. +- Lợi ích: Đây là một khoản chi rất lớn. Năm 2024 doanh thu bán hàng thực tế của thị trường game Trung Quốc là 325.783 tỷ tệ, số người dùng 674 triệu, bình quân khoảng 480 tệ một người (con số này chia hai con số trên). Tiền đã chi thì cơ bản không lấy lại được. Chỉ có một ngoại lệ: người có năng lực hành vi hạn chế (ở đây là người chưa thành niên) tặng thưởng và nạp tiền mà chưa được người giám hộ đồng ý, khi người giám hộ yêu cầu hoàn tiền thì「人民法院应予支持」. Người đã thành niên tự tặng thưởng và nạp tiền thì không có quy định đó. Ngoài ra, người chưa thành niên chơi game trực tuyến vốn chỉ được chơi vào thứ Sáu, thứ Bảy, chủ nhật và ngày lễ theo luật, mỗi ngày 1 giờ từ 20 giờ đến 21 giờ (Trung Quốc, 2020, 2021) +- Mức bằng chứng: B +- Nguồn: Cục Quản lý Báo chí và Xuất bản quốc gia (国家新闻出版署) (2024). Doanh thu thị trường game Trung Quốc năm nay vượt 3257 tỷ tệ, lập kỷ mới（《Báo cáo ngành game Trung Quốc năm 2024》, Nhóm công tác game của Hiệp hội Âm thanh, số và truyền hình Trung Quốc）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Ý kiến chỉ đạo về việc xử lý đúng đắn theo pháp luật một số vụ án dân sự liên quan dịch bệnh viêm phổi do vi rút corona (二)（法发〔2020〕17 号，第 9 条）. ；Cục Quản lý Báo chí và Xuất bản quốc gia (2021). Thông báo siết chặt quản lý, thực sự ngăn người chưa thành niên nghiện game trực tuyến（国新出发〔2021〕14 号）. +- Ghi chú: Xếp mức B vì chưa ai đo trực tiếp「không tặng thưởng thì tiết kiệm được bao nhiêu」, chỉ có số liệu tổng của ngành game. Hai cách tắt thanh toán không cần mật mã và đặt hạn mức là kinh nghiệm thực hành, nếu tính riêng thì là mức C. Nguyên văn Điều 十九 và Điều 一百四十五 của Bộ luật Dân sự đã được dẫn nguyên văn ở mục 9 (con nạp tiền, tặng thưởng trên điện thoại). Không phải lời khuyên đầu tư. + +### 9. Con nạp tiền, tặng thưởng trên điện thoại: khoản chi lớn của trẻ từ tám tuổi trở lên, cha mẹ không chấp thuận thì có thể đòi lại + +- Chi phí: Không tốn tiền. Phải bỏ thời gian thu thập chứng cứ và thương lượng với nền tảng, nền tảng thường từ chối trước, hoàn tiền sau. Không thống nhất được thì chỉ còn đi kiện tụng, mà việc đó mất nhiều tháng trở lên, phí luật sư phải tự bỏ. Nên chuẩn bị đủ hồ sơ rồi đi qua nền tảng và 12315 là hơn. Chỗ khó là con không chịu nói, bạn phải kiên nhẫn hỏi cho rõ. +- Nói thẳng: Khoản tiền trẻ từ tám tuổi trở lên tiêu, nếu số tiền đó rõ ràng không tương xứng với tuổi và trình độ của trẻ, mà cha mẹ vừa không đồng ý vừa không chấp nhận khoản chi đó, thì giao dịch vô hiệu, có thể yêu cầu hoàn lại. Trẻ dưới tám tuổi vốn không tự quyết được chuyện tiêu tiền. Cha mẹ im lặng mãi thì theo luật coi như không chấp nhận khoản chi đó. +- Lợi ích: Người chưa thành niên từ tám tuổi trở lên là người có năng lực hành vi hạn chế, nghĩa là chỉ tự làm được những việc tương xứng với tuổi và trình độ. Hành vi pháp lý dân sự không tương xứng với tuổi và trình độ thì phải được người đại diện theo pháp luật đồng ý hoặc chấp thuận sau mới có hiệu lực: cha mẹ phải gật đầu trước, hoặc ngậm chấp sau. Trẻ dưới tám tuổi là người không có năng lực hành vi, tự quyết không được, việc của trẻ do người đại diện theo pháp luật thay. Người đại diện theo pháp luật không tỏ thái độ thì coi như từ chối chấp thuận sau: cha mẹ im lặng mãi thì coi như không chấp nhận khoản chi đó +- Mức bằng chứng: A +- Nguồn: Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự. Bản toàn văn do Viện kiểm sát nhân dân tối cao đăng lại. 第十九条：「八周岁以上的未成年人为限制民事行为能力人，实施民事法律行为由其法定代理人代理或者经其法定代理人同意、追认；但是，可以独立实施纯获利益的民事法律行为或者与其年龄、智力相适应的民事法律行为。」— người chưa thành niên từ tám tuổi trở lên là người có năng lực hành vi hạn chế, hành vi pháp lý dân sự phải do người đại diện theo pháp luật thực hiện, hoặc phải được người đại diện theo pháp luật đồng ý hay chấp thuận sau; nhưng có thể tự thực hiện hành vi pháp lý dân sự thuần túy mang lại lợi ích, hoặc hành vi pháp lý dân sự tương xứng với tuổi và trình độ của mình. 第二十条：「不满八周岁的未成年人为无民事行为能力人，由其法定代理人代理实施民事法律行为。」— người chưa thành niên dưới tám tuổi là người không có năng lực hành vi, hành vi pháp lý dân sự do người đại diện theo pháp luật thực hiện thay. 第一百四十五条：「限制民事行为能力人实施的纯获利益的民事法律行为或者与其年龄、智力、精神健康状况相适应的民事法律行为有效；实施的其他民事法律行为经法定代理人同意或者追认后有效。」「法定代理人未作表示的，视为拒绝追认。」— hành vi pháp lý dân sự thuần túy mang lại lợi ích, hoặc tương xứng với tuổi, trình độ và tình trạng tâm thần của người có năng lực hành vi hạn chế thì có hiệu lực; các hành vi pháp lý dân sự khác phải được người đại diện theo pháp luật đồng ý hoặc chấp thuận sau mới có hiệu lực. Người đại diện theo pháp luật không tỏ thái độ thì coi như từ chối chấp thuận sau. +- Ghi chú: Chỗ cốt lõi là chứng cứ: bạn phải đưa ra được bằng chứng chứng minh thao tác trên điện thoại là trẻ, không phải bạn. Tài liệu dùng được gồm: thời điểm nạp tiền tập trung vào giờ học hoặc giờ cha mẹ đi làm, lịch sử dùng thiết bị, lịch sử chat và lịch sử chơi game của trẻ, và bản giải thích của chính trẻ.「与年龄智力相适应」không có một mức tiền chung, một skin game vài chục tệ và một lần tặng thưởng mấy vạn tệ không phải một việc. Phòng trước còn dễ chịu hơn: đặt trần thanh toán không cần mật mã, bật chế độ thanh thiếu niên, đừng đưa mật mã thanh toán cho trẻ. Trường hợp bạn tự tặng thưởng, nạp tiền khi nóng đầu, xem mục「không tặng thưởng người livestream, không nạp tiền vào game, không đặt hàng bốc đầu」trong phần này. + +### 10. Dạy trẻ nhớ một câu: bất cứ ai bảo trẻ cầm điện thoại của cha mẹ ra thao tác, hoặc đọc mã xác minh, đều là kẻ lừa + +- Chi phí: Không tốn tiền. Kể cho trẻ nghe một lượt sáu loại lừa đảo bên dưới, rồi thỏa thuận với trẻ một câu「bị lừa rồi thì nói với con trước, con không mắng». Đồng thời đặt trần thanh toán không cần mật mã, bật chế độ thanh thiếu niên, không đưa mật mã thanh toán cho trẻ. Chỗ khó là sau khi bị lừa thật thì nhịn không nổi cơn giận. +- Nói thẳng: Sáu loại lừa đảo này mỗi loại một kiểu, nhưng bước cuối luôn là một động tác: bảo trẻ cầm điện thoại của cha mẹ để chuyển tiền, hoặc đọc mã xác minh. Vì thế dạy một câu là đủ: ai bảo trẻ đụng vào điện thoại của cha mẹ, đọc mã xác minh, người đó là kẻ lừa. Thêm một câu「bị lừa rồi thì nói với con trước, con không mắng」. Trẻ dám nói trong ngày thì tiền mới có cơ hội được đòi lại. +- Lợi ích: Trung tâm chống lừa đảo thành phố Quanzhou phối hợp với Sở Giáo dục thành phố Quanzhou, tháo gỡ sáu loại lừa đảo nhắm vào thanh niên. Chúng khác nhau về kiểu, nhưng bước cuối vẫn là một động tác: bảo trẻ cầm điện thoại của cha mẹ để chuyển tiền. Loại thứ nhất, giao dịch tài khoản game và trang bị giá rẻ ngoài nền tảng. Tiểu Vương chuyển riêng 260 tệ, bên kia nói「người chưa thành niên chuyển tiền là vi phạm, tài khoản sẽ bị đóng băng」, cậu bấm「đường dẫn kiểm tra」bằng điện thoại của bố, hơn 30.000 tệ trong tài khoản bị chuyển đi. Loại thứ hai,「gỡ chặn chơi quá giờ」. Bên kia dọa「không chuyển tiền thì huỷ, cha mẹ con sẽ vào tù」, trẻ dùng điện thoại của mẹ chuyển 7000 tệ. Nguyên văn ghi rõ việc chặn chơi quá giờ do nhà chức thiết lập,「vốn không có kênh『mở khoá riêng』nào cả」. Loại thứ ba, hoàn tiền tặng thưởng. Kẻ lừa đóng vai「nhân viên hỗ trợ chính thức」, nói người chưa thành niên tặng thưởng thì hoàn được, đòi「phải dùng tài khoản Alipay của cha mẹ tặng quà cho phòng livestream chỉ định thì mới hoàn tiền」, 73.000 tệ trong tài khoản của mẹ bị tặng đi, rồi người đó bị chặn. Nguyên văn nói「có người kết bạn rồi bảo bạn『tặng quà để hoàn tiền』, 100% là lừa đảo」. Loại thứ tư, giả danh bạn học hoặc giáo viên. Kẻ lừa chiếm tài khoản, gửi tin nhắn thoại nói đúng tên để mượn tiền. Đầu năm học còn lẫn vào nhóm lớp giả danh giáo viên, đăng「nộp học phí 500 tệ」, ảnh đại diện và tên đều y hệt giáo viên thật, còn thuê người đóng vai học sinh trong nhóm để nói「đã nộp rồi」. Loại thứ năm, chạy theo thần tượng. Kẻ lừa kéo trẻ vào「nhóm fan ngôi sao, nhận lì xì 888 tệ」. Bảo trẻ chuyển 48.88 tệ để「xác minh」trước. Rồi dọa「người chưa thành niên tham gia là vi phạm, nền tảng sẽ đóng băng 80.000 tệ, cha mẹ con sẽ vào tù」. Tổng cộng lấy đi 16.000 tệ. Nguyên văn nói「cảnh sát thật, luật sư thật không bao giờ liên lạc với người chưa thành niên trên mạng để điều tra, càng không đòi điện thoại của cha mẹ ra thao tác」. Loại thứ sáu, trên nền tảng hàng cũ dùng hàng giá rẻ để kéo người ra khỏi nền tảng rồi chuyển tiền riêng. Cùng tài liệu đó còn liệt kê ba vạch đỏ trẻ tự tránh. Một là không cho mượn, không bán, không cho thuê thẻ ngân hàng và thẻ điện thoại, thẻ bị dùng để chuyển tiền từ phạm tội có thể cấu thành tội trợ giúp hoạt động phạm tội qua mạng thông tin, có tiền án sẽ ảnh hưởng tới kỳ thi và thi công chức. Hai là làm thêm qua「手机口」là hoạt động vi phạm pháp. Ba là「tuyển dụng lương cao」ở nước ngoài, người đi có thể bị khống chế ngay khi vừa ra nước ngoài (toàn quốc, tháng 10 năm 2025) +- Mức bằng chứng: B +- Nguồn: Trung tâm chống lừa đảo thành phố Quanzhou (泉州市反诈骗中心) (2025). Bài học đầu năm học｜Chống lừa đảo là môn bắt buộc của thanh niên, 6 cái bẫy lớn này nhất định phải tránh!（Cục Công an tỉnh Phúc Kiên đăng lại） +- Ghi chú: Xếp mức B vì nguồn chính thức đưa ra là phần tháo gở lời lẻ của kẻ lừa và các ca lẻ, không có tỷ lệ vụ án, cũng không có số liệu thống kê về hiệu quả phòng ngừa. Phải tách khỏi mục 9 (con nạp tiền, tặng thưởng trên điện thoại): tiền trẻ tự nạp, tự tặng thưởng thì có thể đòi hoàn theo Bộ luật Dân sự. Tiền bị kẻ lừa lấy thì điều đó không áp dụng, chỉ có thể báo công an để chặn chuyển tiền theo mục 2 ở phần 8. Có đòi lại được không thì tuỳ tiền còn nằm trong tài khoản hay không. Câu dọa cốt lõi của kẻ lừa với trẻ là「con vi phạm rồi, cha mẹ con sẽ vào tù」. Nói rõ điều này với trẻ từ trước, bản thân câu đó đã là dấu hiệu của lừa đảo, hữu ích hơn là hỏi lại sau. Câu「bị lừa thì nói trước, con không mắng」phải thực sự làm được. Trẻ sợ bị mắng nên giấu không nói, đó là nguyên nhân chung khiến thiệt hại trong các ca này càng lăn lóc càng lớn. Bản rộng hơn của kiểu giả danh người quen, và hoán đổi khuôn mặt bằng AI, xem mục 3 và 4 ở phần 8（quy tắc cứng chống lừa đảo, có chuyển tiền thì treo máy）. + +### 11. Từ chối bảo hành mở rộng cho thiết bị điện tử + +- Chi phí: Không tốn tiền. Cái giá phải trả là sau khi hết thời hạn bảo hành, nếu phải sửa thì bạn tự bỏ tiền, nhưng chuyện này xác suất không lớn. +- Nói thẳng: Bảo hành mở rộng thường bán giá bằng 10% đến 50% giá gốc hàng hoá. Người bán bảo hành mở rộng kiếm hơn hàng thường mười mấy lần. Best Buy của Mỹ năm 2003 đã kiếm được hơn một nửa lợi nhuận nhờ bảo hành mở rộng, trong khi bảo hành mở rộng chỉ chiếm 3% đến 4% doanh thu. Chênh lệch đó chính là số tiền bạn trả thừa. +- Lợi ích: Bảo hành mở rộng thường được định giá bằng 10% đến 50% giá gốc hàng hoá. Ước tính ngành được nghiên cứu dẫn lại cho biết biên lợi nhuận gộp trung bình của bảo hành mở rộng là 50% đến 60%: bán ra 100 tệ thì 50 đến 60 tệ là lãi. Xấp xỉ gấp 18 lần biên lợi nhuận gộp của hàng thường. Năm 2003, bảo hành mở rộng chỉ chiếm 3% đến 4% doanh thu của Best Buy nhưng đóng góp hơn 50% lợi nhuận (Mỹ, số liệu bán lẻ thập niên 2000) +- Mức bằng chứng: B +- Nguồn: Chen T, Kalra A, Sun B (2009). Why Do Consumers Buy Extended Service Contracts? Journal of Consumer Research 36(4):611-623. +- Ghi chú: Con số biên lợi nhuận gộp đó là bài báo dẫn lại ước tính của ngành (Business Week 2004), không phải bài báo tự đo. Cách tính ở trên áp dụng cho người bình thường tự bỏ được phí sửa một lần. Người sống bằng một thiết bị mà không có máy dự phòng thì phải tính riêng. Bảo hiểm vỡ màn hình điện thoại cũng phải tính riêng: trước hết ước lượng bao lâu bạn làm rơi một lần, rồi xem tiền phí bảo hiểm có đáng không. + +### 12. Ưu tiên thuốc generic đã đạt đánh giá tính nhất quán và thuốc trong gói đấu thầu tập trung + +- Chi phí: Không tốn tiền. Khi bác sĩ kê đơn và bạn đi mua thuốc, hỏi thêm một câu「có thuốc trong gói đấu thầu, thuốc theo tên thành phần chung không」. +- Nói thẳng: Đợt đấu thầu thuốc thứ hai của quốc gia giảm giá trung bình 53%, loại giảm nhiều nhất là 93%. Đến tháng 4 năm 2026 đã tổ chức 11 đợt, bao phủ 490 loại thuốc. Khi kê đơn và mua thuốc, hỏi thêm một câu「có thuốc trong gói đấu thầu, theo tên thành phần chung không」, thành phần giống nhau mà tiền phải bỏ ra ít hơn nhiều. +- Lợi ích: Đợt đấu thầu mua thuốc tập trung do quốc gia tổ chức, đợt thứ hai giảm giá trung bình 53%, cao nhất giảm 93%. Tính đến tháng 4 năm 2026 đã tổ chức 11 đợt, bao phủ 490 loại thuốc. Số thuốc đã đạt hoặc được coi là đạt đánh giá tính nhất quán lên tới 1695 loại. Xấp xỉ bằng 2/3 các thuốc hoá chất dùng thường trong lâm sàng (Trung Quốc, số liệu chính thức 2020 đến 2026) +- Mức bằng chứng: A +- Nguồn: Cục Quản lý bảo hiểm y tế quốc gia (国家医疗保障局) (2020). Trả lời phóng viên về công tác mua sắm và sử dụng thuốc tập trung do quốc gia tổ chức đợt thứ hai. ；Cục Quản lý bảo hiểm y tế quốc gia (2026). Họp nói chuyện định kỳ về chính sách của Hội đồng Nhà nước: hoàn thiện cơ chế hình thành giá thuốc xuyên suốt vòng đời và mọi kênh. ；Văn phòng Hội đồng Nhà nước (国务院办公厅) (2016). Ý kiến về việc tổ chức đánh giá tính nhất quán chất lượng và hiệu quả thuốc generic（国办发〔2016〕8 号）. ；Cục Quản lý Dược phẩm và Thực phẩm quốc gia (国家食品药品监督管理总局) (2016). Thông báo ban hành 3 nguyên tắc hướng dẫn kỹ thuật, trong đó có nguyên tắc chọn và xác định thuốc đối chứng cho dạng bào chế rắn uống thông thường（Thông báo số 61 năm 2016）Phụ lục 1, Phụ lục 3, Cục Quản lý Dược phẩm tỉnh Hải Nam đăng lại. ；Cục Quản lý bảo hiểm y tế quốc gia (2024). Siết chặt quản lý chất lượng thuốc trong gói đấu thầu, hai「bao phủ toàn bộ」là kiểm tra doanh nghiệp và lấy mẫu kiểm tra từng loại. ；Cục Quản lý bảo hiểm y tế quốc gia (2025). Các cán bộ của cơ quan bảo hiểm y tế và cơ quan quản lý dược phẩm trả lời phóng viên báo Nhân Dân Daily về các vấn đề liên quan thuốc trong gói đấu thầu. +- Ghi chú:Tranh cãi. Nhiều người nói「đánh giá tính nhất quán chỉ so thành phần, không so hiệu quả». Câu này đúng một nửa. Cách đạt đánh giá là cho người khoẻ mạnh uống thuốc, lấy máu đo đường cong nồng độ thuốc sau khi thuốc vào máu, rồi so đường cong đó với thuốc nguyên phát, rơi vào khoảng 80% đến 125% thì coi như đạt. Với loại thuốc mà chênh lệch liều là chênh lệch nguy hiểm, yêu cầu còn khắt khe hơn. Tá dược là phần đệm và lớp bao ngoài hoạt chất, được phép khác thuốc nguyên phát. Quy chế chỉ yêu cầu hoạt chất, dạng bào chế, cách dùng và tác dụng điều trị giống nhau. Vì thế nó đúng là không phải「cho bệnh nhân uống một thời gian rồi xem bệnh khỏi hay không」. Hiệu quả trên người bệnh thì một bộ nghiên cứu khác trả lời. Cục Quản lý bảo hiểm y tế quốc gia đã tổ chức nghiên cứu thế giới thực tại bệnh viện. Đến tháng 12 năm 2024 nghiên cứu bao phủ 63 loại thuốc trong gói đấu thầu, gần một trăm bệnh viện hạng A, hơn 300 nghìn bệnh nhân. Kết quả đều không khác thuốc nguyên phát. Ví dụ metformin: tỷ lệ đạt chuẩn hemoglobin niết hoá là 83.4% ở nhóm generic, 78.9% ở nhóm nguyên phát; tỷ lệ đạt chuẩn đường huyết lúc nhịn đói thì ngược lại, 83.8% ở nhóm generic, 87.6% ở nhóm nguyên phát. Hai nhóm có phần cao hơn lẫn nhau. Tháng 1 năm 2025 có bác sĩ nói công khai rằng thuốc trong gói đấu thầu「huyết áp không hạ, thuốc mê không ngủ, thuốc nhuận không nhuận」. Cơ quan bảo hiểm y tế và cơ quan quản lý dược phẩm đã đi kiểm tra 7 bệnh viện ở Thượng Hải, kết luận rằng các lời này phần lớn đến từ kể lại và cảm tính cá nhân, trong đó loại thuốc nhuận bị nêu tên lúc đó mới chưa bắt đầu dùng trong bệnh viện. Cùng đợt kiểm tra đó thừa nhận là có lấy được thuốc có vấn đề. Trong hơn 1600 sản phẩm trúng thầu của 9 đợt đầu, có 9 sản phẩm bị thu hồi tư cách vì rủi ro chất lượng. Trong 9 sản phẩm đó, 6 sản phẩm là thuốc nhập khẩu (trong đó có 3 thuốc nguyên phát), 3 sản phẩm là thuốc trong nước. Có hai việc bạn làm được. Một là muốn tiếp tục dùng thuốc nguyên phát thì thường vẫn kê được: gói đấu thầu chỉ chiếm 60% đến 80% khối lượng bệnh viện mua, phần còn lại do bệnh viện tự quyết, thuốc nguyên phát không bị cấm, cái giá là bạn tự bỏ thêm tiền. Hai là sau khi bệnh mạn đổi thuốc thì đừng phán đoán bằng cảm tính, theo mục 3 ở phần 16 (ghi lại mỗi chỉ số vào cùng một cuốn sổ) để theo dõi huyết áp, đường huyết, chỉ số niết hoá. Nếu thật sự xấu đi thì đi tìm bác sĩ, nhờ bác sĩ báo cáo tác dụng phụ không mong muốn của thuốc, đừng tự ngừng thuốc. +### 13. Làm một lần liên kết dùng chung tài khoản trên app bảo hiểm y tế, tiền trong tài khoản cá nhân của bảo hiểm y tế người lao động dùng được cho vợ chồng, cha mẹ, con cái + +- Chi phí: Không tốn tiền. Làm một lần liên kết dùng chung cho gia đình trên app bảo hiểm y tế ở nơi bạn tham gia bảo hiểm, hoặc tới cơ quan thực hiện. +- Nói thẳng: Làm một lần liên kết, tiền trong tài khoản cá nhân của bảo hiểm y tế người lao động sẽ dùng được cho vợ chồng, cha mẹ, con cái khi khám bệnh, mua thuốc. Không làm thì số tiền đó chỉ bạn dùng được, nằm mãi trong tài khoản. +- Lợi ích: Tiền trong tài khoản cá nhân dùng để trả phần chi phí khám chữa bệnh do bản thân bạn và vợ chồng, cha mẹ, con cái bỏ ra tại cơ sở khám chữa bệnh tuyến chỉ định. Cũng dùng để trả phần tiền bạn tự bỏ ra khi mua thuốc, thiết bị y tế, vật tư y tế tại hiệu thuốc tuyến chỉ định. Văn bản còn đề nghị thử nghiệm dùng tài khoản cá nhân để đóng khoản đóng của bảo hiểm y tế cho cư dân đô thị và nông thôn cho người thân +- Mức bằng chứng: A +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2021). Ý kiến chỉ đạo về việc xây dựng và hoàn thiện cơ chế bảo đảm thanh toán chung cho khám bệnh ngoại trú của bảo hiểm y tế cơ bản người lao động（国办发〔2021〕14 号）第（五）项：「个人账户主要用于支付参保人员在定点医疗机构或定点零售药店发生的政策范围内自付费用。可以用于支付参保人员本人及其配偶、父母、子女在定点医疗机构就医发生的由个人负担的医疗费用，以及在定点零售药店购买药品、医疗器械、医用耗材发生的由个人负担的费用。探索个人账户用于配偶、父母、子女参加城乡居民基本医疗保险等的个人缴费。个人账户不得用于公共卫生费用、体育健身或养生保健消费等不属于基本医疗保险保障范围的支出。」— tài khoản cá nhân chủ yếu dùng để trả phần chi phí trong phạm vi chính sách do người tham gia bảo hiểm tự bỏ ra tại cơ sở khám chữa bệnh tuyến chỉ định hoặc hiệu thuốc tuyến chỉ định. Có thể dùng để trả phần chi phí do người tham gia bảo hiểm và vợ chồng, cha mẹ, con cái tự bỏ ra khi khám chữa bệnh tại cơ sở tuyến chỉ định, cũng như phần tiền tự bỏ ra khi mua thuốc, thiết bị y tế, vật tư y tế tại hiệu thuốc tuyến chỉ định. Cần thử nghiệm dùng tài khoản cá nhân cho khoản đóng bảo hiểm y tế cơ bản cư dân đô thị và nông thôn của vợ chồng, cha mẹ, con cái. Tài khoản cá nhân không được dùng cho chi phí y tế cộng đồng, tập thể dục, hay các khoản chi tiêu dùng chăm sóc sức khoẻ không thuộc phạm vi bảo hiểm. +- Ghi chú: Rất nhiều người có vài nghìn tệ nằm trong tài khoản cá nhân, trong khi người cao tuổi trong nhà tự bỏ tiền mua thuốc. Liên kết một lần thì có hiệu lực dài hạn. Hầu hết các nơi gọi là「家庭共济」, làm được ngay trong app bảo hiểm y tế địa phương. Cần chú ý một vạch đỏ ở chiều ngược lại: tài khoản cá nhân không dùng cho các khoản chi như tập thể dục, chăm sóc sức khoẻ. Mang nó tới hiệu thuốc mua thực phẩm chức năng để rút tiền mặt là gian lận bảo hiểm. + +### 14. Uống nước máy đã đun sôi, không mua nước đóng chai quanh năm + +- Chi phí: Mua một ẩm. Mỗi lần đun nước tốn một ít thời gian, cộng thêm chút tiền khí đốt hoặc tiền điện. +- Nói thẳng: Giá nước ở bậc một của Bắc Kinh quy ra 0,005 tệ một lít. Nếu tính nước đóng chai giá 1 tệ 500 ml thì ra 2 tệ một lít. Hai mức chênh nhau 400 lần. Nước máy tuân theo một tiêu chuẩn quốc gia bắt buộc, bình thường đun sôi lên uống là được. Nếu thật sự gặp sự cố nguồn nước, nước có mùi thối và đục, thì đun thêm mấy phút không giải quyết được, phải dừng uống và báo ra ngoài. +- Lợi ích: Giá nước dân dụng Bắc Kinh ở bậc một là 5 tệ một mét khối, quy ra 0,005 tệ một lít. Nếu tính nước đóng chai giá 1 tệ 500 ml thì ra 2 tệ mỗi lít, gấp 400 lần nước máy. Nước uống sinh hoạt tuân theo tiêu chuẩn quốc gia bắt buộc GB 5749-2022. Tiêu chuẩn này ban hành ngày 15 tháng 3 năm 2022, có hiệu lực từ ngày 1 tháng 4 năm 2023. Cơ quan quản lý là Cục Quản lý và Phòng ngừa bệnh quốc gia (Trung Quốc, giá nước năm 2015 và tiêu chuẩn năm 2022) +- Mức bằng chứng: C (Tranh cãi) +- Nguồn: Ủy ban Phát triển và Cải cách thành phố Bắc Kinh. Bảng giá nước sinh hoạt dân dụng. ；Cục Quản lý thị trường quốc gia (国家市场监督管理总局), Hệ thống công khai toàn văn tiêu chuẩn quốc gia. GB 5749-2022 Tiêu chuẩn vệ sinh nước uống sinh hoạt. ；Nhóm điều tra liên hợp thành phố Hàng Châu (2025). Thông báo về kết quả điều tra tình hình cấp nước bất thường ở một số khu chung cư quận Dư Hàng của thành phố Hàng Châu (ngày 23 tháng 7 năm 2025), Tân Hoa Xinh đăng lại ；Chính quyền quận Dư Hàng thành phố Hàng Châu. Dữ liệu giám sát chất lượng nước nguồn quận Dư Hàng tháng 4／5／6 năm 2025（Tập đoàn Quản lý Cấp nước Dư Hàng Hàng Châu công bố, đơn vị kiểm nghiệm là Công ty TNHH Giám sát Môi trường Gia Kiếm Hàng Châu, bảng PDF phụ lục 1, mục 24「粪大肠菌群」） 、、；Cục An ninh mạng và An ninh mạng của Bộ Công an (公安部网络安全保卫局) (2025). Bịa「thông báo tình huống công an」giả, dựng tin đồn「nước phân」, Thao Tư Hào bị bị giữ +- Ghi chú:Tranh cãi. Cái gọi là「đạt chuẩn」là kiểm tra nước ra khỏi nhà máy và nước ở cuối đường ống. Khi nguồn nước hoặc mạng lưới có sự cố, chỉ tiêu này mất kiểm soát trong thời gian ngắn. Ngày 16 tháng 7 năm 2025, nước thô dẫn vào nhà máy nước Nhân Hợp quận Dư Hàng có mùi bất thường, nước máy ở một số khu chung cư Nhân Hợp và Lưng Chủ có mùi thối. Kết luận chính thức là tảo phân huỷ trong điều kiện thiếu ô-xi, sinh ra các chất sulfide. Sau khi đổi nguồn nước, không phát hiện tổng số vi khuẩn đường ruột và Escherichia coli trong nước ra khỏi nhà máy, mọi hạng mục đều đạt tiêu chuẩn nước uống sinh hoạt GB 5749-2022. Sau đó có 7 cán bộ bị kỷ luật, tiền nước tháng 7 miễn toàn bộ, còn được đổi miếng lọc miễn phí. Nhưng cùng nguồn nước đó, chỉ số vi khuẩn đường ruột chí ỉ đã đi lên liên tục mấy tháng trước sự cố. Cửa rút nước Dư Hành Vĩnh Thắng tháng 4 là 10 cái/L, tháng 5 là 7300 cái/L, tháng 6 là 16000 cái/L. Giới hạn của loại III trong tiêu chuẩn nước mặt GB 3838 là 10000 cái/L. Cùng tháng, hồ Tứ Lĩnh 28000 cái/L, hồ Mã Đầu Sơn 24000 cái/L cũng vượt. Đun sôi chỉ diệt được vi sinh vật, không bỏ được chất tạo mùi, cũng không bỏ được chất hoá chất hoà tan trong nước. Ngửi thấy mùi lạ, hoặc nước vàng hay đục, hãy dừng uống, báo cho doanh nghiệp cấp nước và 12345, tạm chuyển sang uống nước đóng chai, đừng trông đợi đun thêm mấy phút để áp chế. Nhưng đây là sự cố cục bộ, xảy ra thỉnh thoảng, địa phương cũng đã bồi thường theo từng đợt, cho nên vẫn không thành lý do mua nước đóng chai quanh năm. Tin trên mạng「粪水管接错」là thông báo tình huống công an bịa ra, người đăng đã bị bị giữ (xem phần 9). Giá 1 tệ 500 ml của nước đóng chai là giả định để tính sổ, hãy tính lại theo giá thật ở thành phố của bạn. Khu chung cư cũ có cơ sở cấp nước lần hai không đạt, có thể lắp thêm bộ lọc, hoặc chỉ mua nước uống trong trường hợp đó. Giá nướmỗi nơi một khác, số liệu Bắc Kinh chỉ dùng để nhìn mức độ. + +### 15. Không giao dịch cổ phiếu thường xuyên + +- Chi phí: Không tốn tiền. Chỗ khó là thấy tin là muốn vào ngay, phải nhịn lại. +- Nói thẳng: Người mua bán cổ phiếu thường xuyên nhất, trung bình mỗi năm chỉ kiếm 11.4%, cùng thời gian đó toàn thị trường là 17.9%. Mấy phần trăm chênh lệch đó gần như bị phí môi giới trả lúc mua bán ăn mất. Nếu không tính phí, người giao dịch thường và người giao dịch ít kiếm được gần như nhau. +- Lợi ích: Nghiên cứu này xem 66.465 tài khoản gia đình tại nhà môi giới chiết khấu Mỹ, trong giai đoạn 1991 đến 1996. Nhóm giao dịch thường xuyên nhất có lợi nhuận quy đổi cả năm 11.4%, tức trung bình mỗi năm kiếm 11.4%. Cùng thời gian đó toàn thị trường là 17.9%, bình quân một hộ gia đình là 16.4%. Bình quân tỷ lệ luân chuyển cổ phiếu của một hộ trong một năm là 75%, tức trong một năm bán đi ba phần tư số cổ phiếu đang cầm. Nhóm giao dịch thường xuyên và nhóm ít giao dịch chênh lệch rất nhỏ về lợi nhuận trước phí, nên khoảng cách chủ yếu đến từ chi phí giao dịch phải trả lúc mua bán (Mỹ, 1991 đến 1996) +- Mức bằng chứng: A +- Nguồn: Barber BM, Odean T (2000). Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors. The Journal of Finance 55(2):773-806. +- Ghi chú: Số liệu này là của Mỹ thập niên 1990. Cơ cấu hoa hồng và thuế đóng cửa của cổ phiếu A khác, nhưng「luân chuyển càng nhiều, lợi nhuận càng thấp」thì vẫn đúng. Không phải lời khuyên đầu tư. + +### 16. Không vay tiền để đầu tư, không dùng đòn bẩy, không mua thứ mình không hiểu + +- Chi phí: Không tốn tiền. Chỗ khó là lúc thị trường đi lên, phải nhịn không nghĩ「vay chút tiền để kiếm thêm」. +- Nói thẳng: Vay tiền mua cổ phiếu, một khi rơi xuống dưới đường mà bạn không bổ tiền được, bạn sẽ bị bán cưỡng chế. Lỗ trên sổ sách biến thành lỗ thật ngay lập tức, không còn cơ hội「等它涨回来」. Quy định giám sát nói: người mở tài khoản chưa đủ nửa năm, có tài sản bình quân trong 20 ngày giao dịch gần nhất dưới 50 vạn tệ thì không được mở tài khoản vay margin. Ngưỡng đó bản thân nó đã nói lên đây không phải dành cho người bình thường. +- Lợi ích: Đòn bẩy chính là vay tiền để mua, nó khuếch đại lỗ. Sau khi bạn mua bằng tiền vay, giá trị tài sản thế chấp có thể rơi xuống dưới đường tỷ lệ bảo đảm tối thiểu. Đường này do công ty chứng khoán vạch ra, nó quản lý đúng phần bạn đã thế chấp ở đó. Thấm xuống dưới đường đó mà bạn không bổ sung thế chấp để kéo lên được, bạn sẽ bị đóng vị thế cưỡng chế, công ty chứng khoán bán thẳng giúp bạn. Lỗ trên sổ sách biến thành lỗ thật ngay lập tức, không có cơ hội「等它涨回来」. Quy định giám sát nói rằng khách hàng「从事证券交易时间不足半年、缺乏风险承担能力、最近 20 个交易日日均证券类资产低于 50 万元」không được mở tài khoản tín dụng. Ngưỡng đó bản thân nó đã nói lên đây không phải công cụ dành cho người bình thường (Trung Quốc, 2015) +- Mức bằng chứng: B +- Nguồn: Ủy ban Chứng khoán Trung Quốc (中国证监会) (2015). Biện pháp quản lý nghiệp vụ tài chính và chứng khoán của công ty chứng khoán（证监会令第 117 号，第十二条）. ；Sở Giao dịch chứng khoán Thượng Hải. Giải thích《Quy chế thi hành giao dịch tài chính và chứng khoán của Sở Giao dịch chứng khoán Thượng Hải》(tài liệu giáo dục nhà đầu tư). +- Ghi chú: Câu「không hiểu thì không mua」không có tài liệu nào chứng minh trực tiếp, thuộc đồng thuận, nếu tính riêng thì là mức C. Dùng vay tiêu dùng hoặc rút tiền từ thẻ tín dụng để đầu tư cũng là đòn bẩy hình thức khác, lãi suất của chúng xem mục 7 của phần này (không dùng khoản trả tối thiểu trên thẻ tín dụng). Không phải lời khuyên đầu tư. + +### 17. Thay quỹ quản lý chủ động bằng quỹ chỉ số phạm vi rộng cho phần giữ dài hạn + +- Chi phí: Không tốn tiền. Cái giá phải trả là bạn từ bỏ khả năng chọn trúng quản lý quỹ nổi tiếng. Chỗ khó là khi chỉ số giảm, bạn cũng phải chịu chung, không bỏ chạy giữa chừng. +- Nói thẳng: Đống cổ phiếu mà các quỹ chủ động gộp lại mua, vốn đã rất gần cả thị trường rồi. Nhưng khoản phí quản lý họ thu thêm được trừ thẳng vào lợi nhuận bạn thực nhận. Nếu coi kết quả của quỹ là may rủi rồi tính đi tính lại để so, có rất ít quỹ nào lấy được phần kiếm thêm để bù lại đúng khoản phí mình thu. +- Lợi ích: Danh mục tổng thể của các quỹ cổ phiếu quản lý chủ động ở Mỹ gần danh mục thị trường: cổ phiếu các quỹ đó cùng nắm giữ, xấp xỉ cả thị trường. Nhưng chi phí cao của việc quản lý chủ động biến nguyên vẹn thành lợi nhuận ròng thấp hơn của nhà đầu tư. Nghiên cứu dùng mô phỏng tự lấy mẫu, xáo trộn ngẫu nhiên kết quả nhiều lần, để xem riêng may rủi thì đạt tới mức nào. Kết quả cho thấy rất ít quỹ tạo được lợi nhuận vượt trội đủ bù chi phí: phần quỹ kiếm được nhiều hơn thị trường không đủ bù khoản phí nó thu (Mỹ, công bố năm 2010). Công ty chỉ số S&P Dow Jones thống kê mỗi năm một lần: năm 2024, 65% quỹ cổ phiếu vốn lớn quản lý chủ động ở Mỹ thua chỉ số S&P 500. Bộ thống kê đó đã theo dõi 24 năm, trung bình mỗi năm 64% thua. Trong 15 năm tính đến cuối năm 2024, không nhóm nào mà đa số quỹ chủ động thắng +- Mức bằng chứng: B +- Nguồn: Fama EF, French KR (2010). Luck versus Skill in the Cross-Section of Mutual Fund Returns. The Journal of Finance 65(5):1915-1947. ；S&P Dow Jones Indices (2025). SPIVA U.S. Scorecard Year-End 2024. （trang chính thức chặn lấy tự động, số liệu đối chiếu theo bản lưu trữ ngày 2025-05-12: >）；S&P Dow Jones Indices (2025). SPIVA Asia Ex-Japan Scorecard Mid-Year 2025（Report 1a，China Large-Cap 对 S&P China A 300 的跑输比例：1 年 60.5%、3 年 69.6%、5 年 50.8%）. （trang chính thức chặn lấy tự động, đối chiếu theo bản lưu trữ）；Harvey CR, Liu Y (2022). Luck versus Skill in the Cross Section of Mutual Fund Returns: Reexamining the Evidence. The Journal of Finance, 77(3). （bên có ý kiến phản biện） +- Ghi chú:Tranh cãi. Có một bài báo kiểm tra lại kết luận này. Nó chỉ ra rằng cách kiểm tra lấy mẫu lặp lại của Fama-French lấy mẫu chưa đủ. Theo thuật toán của bài đó, ngay cả khi phần kiếm thêm của một quỹ rõ ràng là đáng kể, vẫn có thể bị phán đoán là「không phân biệt được với may rủi」. Vì thế kết luận「hầu như không quỹ nào có năng lực」có thể đã kết luận nặng quá. Số liệu ở Trung Quốc không nghiêng hẳn về một phía như ở Mỹ. Thống kê của S&P Dow Jones tính đến tháng 6 năm 2025: tỷ lệ quỹ cổ phiếu vốn lớn quản lý chủ động của A thua chỉ số S&P China A 300 là 60.5% trong 1 năm, 69.6% trong 3 năm, 50.8% trong 5 năm. Bộ thống kê này chỉ tính tới 5 năm, không có số dài hơn. Vậy nên ở Trung Quốc, việc quỹ chủ động không thắng chỉ số trong dài hạn còn chưa có kết luận, nhưng quỹ chỉ số có mức phí thấp thì vẫn đúng. Không phải lời khuyên đầu tư. +### 18. Cùng một loại quỹ thì chọn quỹ có mức phí thấp + +- Chi phí: Không tốn tiền. Trước khi mua, xem thêm một lần「管理费 + 托管费 + 销售服务费」cộng lại là bao nhiêu. +- Nói thẳng: Mức phí một năm cao hơn 1 điểm phần trăm thì 20 năm sau số tiền thực nhận giảm khoảng 18%, 30 năm sau giảm khoảng 26%. Trong cùng một loại quỹ, chọn quỹ có mức phí thấp là thứ duy nhất ở đây bạn biết trước chắc chắn. +- Lợi ích: Chênh lệch mức phí ăn mòn vốn gốc từng chút theo lãi kép, nghĩa là khoản bị trừ mỗi năm cộng dồn năm sau năm. Mức phí một năm cao hơn 1 điểm phần trăm thì giá trị cuối kỳ sau 20 năm giảm 0.99^20 ≈ 18.2%, giá trị cuối kỳ là số tiền còn nằm trong tài khoản vào thời điểm đó; sau 30 năm giảm khoảng 26%. Trung Quốc bắt đầu cải cách phí quỹ công từ năm 2023, hạ trần mức phí của quỹ lợi ích chủ động (loại chủ yếu mua cổ phiếu, do quản lý quỹ chọn cổ phiếu). Từ ngày 7 tháng 7 năm 2023, quỹ lợi ích chủ động mới đăng ký có mức phí quản lý không quá 1.2%, mức phí lưu ký không quá 0.2%. 136 công ty quỹ đã hạ cả hai khoản này của các sản phẩm cũ xuống dưới mức đó +- Mức bằng chứng: C +- Nguồn: Ủy ban Chứng khoán Trung Quốc (中国证监会) (2023). Lãnh đạo các bộ phận liên quan của Ủy ban Chứng khoán trả lời phóng viên về cải cách phí quỹ công. ；Cục Chứng khoán Xiamen (厦门证监局) (2024).「Quỹ lợi ích chủ động mới đăng ký thống nhất thực hiện trần "mức phí quản lý không quá 1.2%, mức phí lưu ký không quá 0.2%"」. ；Fama EF, French KR (2010). Xem mục 11. +- Ghi chú: Công thức này chỉ làm theo công thức, với giả định hai quỹ kiếm được như nhau trước khi trừ phí. Thực tế còn phải tính thêm phí mua, phí bán lại, và phí dịch vụ bán hàng của lô cổ phần loại C. Không phải lời khuyên đầu tư. + +### 19. Đừng đặt hết tiền vào một cổ phiếu, một nền tảng, một căn nhà + +- Chi phí: Không tốn tiền. Chỗ khó là phải chấp nhận một điều: bạn sẽ không đúng lúc nào đặt trúng vào thứ đáng nhất. +- Nói thẳng: Chỉ cần mấy thứ không cùng tăng cùng giảm tuyệt đối, chia tiền ra nhiều chỗ thì lợi nhuận trung bình không đổi, còn biên độ của tài khoản thì nhỏ đi. Ngược lại, đặt hết tiền vào một cổ phiếu, một nền tảng, một căn nhà thì nó xảy ra chuyện gì là bạn xảy ra chuyện đó. +- Lợi ích: Chỉ cần vài tài sản không cùng tăng cùng giảm tuyệt đối thì chia tiền ra, lợi nhuận kỳ vọng không đổi, biến động lại nhỏ hơn. Đây là kết luận cơ bản về lựa chọn danh mục do Markowitz đưa ra năm 1952, và mấy chục năm sau vẫn dùng nguyên. Ngược lại, đặt hết tiền vào một cổ phiếu hay một nền tảng thì phương sai của nó là toàn bộ phương sai của bạn, nó tăng bao nhiêu thì bạn tăng bấy nhiêu. Khi xảy ra chuyện không có gì phòng hạ, cũng không có tài sản nào khác bù lại cho bạn +- Mức bằng chứng: B +- Nguồn: Markowitz, H. (1952). Portfolio Selection. The Journal of Finance, 7(1), 77–91. +- Ghi chú: Xếp mức B vì bài kinh điển đó chỉ là suy luận toán học thuần tuý, không tính ra con số thực tế「phân tán giúp giảm được bao nhiêu tiền». Phân tán chỉ làm cho tài khoản biến động nhỏ đi, không bảo đảm bạn không lỗ. Mua quỹ chỉ số phạm vi rộng (xem mục 17 của phần này) vốn đã là cách phân tán tiện nhất. Không phải lời khuyên đầu tư. + +### 20. Người nộp thuế thu nhập cá nhân thì mở tài khoản lương hưu cá nhân, mỗi năm 12000 tệ được trừ trước thuế; người không nộp thuế thì mở là không đáng + +- Chi phí: Mở tài khoản không tốn tiền. Cái giá thật là số tiền đó bị khoá tới lúc bạn về hưu, giữa chừng không rút ra được. +- Nói thẳng: Mỗi tệ bạn gửi vào, số thuế tiết kiệm được bằng「thuế suất của bạn trừ 3%」. Gửi đủ 12000 tệ trong một năm, người thuế suất 10% tiết kiệm 840 tệ, người 20% tiết kiệm 2040 tệ, người 30% tiết kiệm 3240 tệ. Người vốn không nộp thuế thu nhập cá nhân không tiết kiệm được đồng nào, lúc rút ra còn phải nộp 3%, là lỗ ròng. Tiền phải khoá tới lúc về hưu mới rút được. +- Lợi ích: Số tiền gửi vào, mỗi năm tối đa 12000 tệ, được trừ thật khỏi thu nhập tổng hợp hoặc thu nhập kinh doanh: bạn gửi vào bao nhiêu thì trừ khỏi thu nhập chịu thuế bấy nhiêu. Lợi nhuận kiếm được khi đầu tư số tiền này trong tài khoản thì tạm thời không chịu thuế thu nhập cá nhân. Khi rút ra sau này thì không gộp vào thu nhập tổng hợp, mà tính thuế riêng ở mức 3%. Vì thế số tiền bạn tiết kiệm là「thuế suất biên của bạn trừ 3%」nhân với số tiền bạn thực gửi. Thuế suất biên là mức thuế áp cho mỗi phần thu nhập bạn kiếm thêm. Tính ra: người có thuế suất biên 10% mỗi năm tiết kiệm được tối đa 840 tệ, người 20% tiết kiệm 2040 tệ, người 30% tiết kiệm 3240 tệ. Người vốn không phải nộp thuế thu nhập cá nhân (thu nhập tổng hợp trong năm trừ hết các khoản rồi không tới ngưỡng), gửi tiền vào không tiết kiệm được đồng nào, lúc rút ra lại phải nộp 3%, là lỗ ròng. Hạn mức đóng mỗi năm tính cộng dồn theo dương lịch, sang năm sau tính lại, phần năm đó chưa đủ thì không chuyển sang năm kế tiếp. Chính sách này thực hiện trên toàn quốc từ ngày 15 tháng 12 năm 2024 (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Tài chính, Tổng cục Thuế (财政部、税务总局) (2024). Thông báo thực hiện chính sách ưu đãi thuế thu nhập cá nhân cho lương hưu cá nhân trên toàn quốc（Thông báo số 21 năm 2024 của Bộ Tài chính và Tổng cục Thuế, ngày 12 tháng 12 năm 2024）. ；Bộ Nhân sự và Xã hội cùng năm bộ khác (2022). Biện pháp thực hiện lương hưu cá nhân（第七、八、九、十二、十三条）. +- Ghi chú: Tài khoản này「vận hành khép kín」, tiền gửi vào thì không rút ra được. Chỉ có bốn trường hợp được nhận: đã tới tuổi nhận lương hưu cơ bản, mất hoàn toàn năng lực lao động, ra nước ngoài định cư, và các trường hợp khác theo quy định của nhà nước. Vì thế nó không phải khoản tích luỹ, mà là một sắp xếp thuế phải khoá nhiều năm. Hãy tính rõ mỗi phần thu nhập bạn kiếm thêm thì phải nộp bao nhiêu thuế, rồi quyết định gửi bao nhiêu. Đừng vì「móc lông vịt」mà khoá cả khoản dự phòng vào (khoản dự phòng xem mục 27). Tiền trong tài khoản còn phải do bạn tự chọn sản phẩm để đầu tư, và có thể lỗ. Cách chọn sản phẩm giống mục 17 và 18 của phần này (quỹ chỉ số phạm vi rộng, chọn quỹ có mức phí thấp). Chứng từ khấu trừ do Nền tảng Quản lý Thông tin Lương hưu cá nhân cấp. Bạn chọn được thời điểm khấu trừ: một là khấu trừ ngay lúc công ty trả lương và thu thuế, cách khác là khấu trừ khi tổng kết vào năm sau (xem mục 2). + +### 21. Phòng tập thì trả theo lượt hoặc theo chu kỳ ngắn, trừ khi bạn đã có hơn một năm đi đều đặn + +- Chi phí: Mua theo lượt thì giá mỗi lượt đắt hơn một chút so với chia ra thẻ cả năm. Bạn phải thành thật ước lượng mình mỗi tháng đi được mấy lần. +- Nói thẳng: Trong số liệu ba năm ở Mỹ, hội viên chọn hợp đồng phí tháng từ 70 đô la trở lên, trung bình mỗi tháng chỉ đi 4.3 lần, quy ra hơn 17 đô la mỗi lần. Thẻ 10 lượt chỉ có 10 đô la mỗi lượt. Trong thời gian còn hội viên, những người này trung bình trả nhiều hơn 600 đô la. Trừ khi bạn đã có hơn một năm đi đều đặn, mua theo lượt hoặc mua ngắn hạn là hơn. +- Lợi ích: Nghiên cứu này xem dữ liệu 3 năm của 7.752 hội viên tại 3 câu lạc bộ thể hình ở Mỹ. Hội viên chọn hợp đồng phí tháng trên 70 đô la, trung bình mỗi tháng chỉ đi 4.3 lần, quy ra hơn 17 đô la mỗi lần. Thẻ 10 lượt chỉ có 10 đô la mỗi lượt. Trong thời gian còn hội viên, những người này trung bình trả nhiều hơn 600 đô la. Ngoài ra, tỷ lệ hội viên trả theo tháng quá một năm mà chưa huỷ thẻ cao hơn 17% so với hội viên trả cả năm (Mỹ, dữ liệu bảng theo dõi ba năm) +- Mức bằng chứng: B +- Nguồn: DellaVigna S, Malmendier U (2006). Paying Not to Go to the Gym. American Economic Review 96(3):694-719. +- Ghi chú: Đây chỉ là một nghiên cứu, dùng số liệu Mỹ. Nhưng cái đằng sau thì rất phổ biến: người ta ước lượng cao hơn sức tự kiểm của mình trong tương lai. Nếu bạn đã có một năm liền đi từ hai buổi mỗi tuần trở lên, thì thẻ cả năm đúng là rẻ hơn. + +### 22. Trả tiền trước phải ký hợp đồng bằng văn bản; người bán có rủi ro phải ngừng nhận tiền, trước khi bỏ chạy bạn có quyền đòi lại số dư chưa tiêu + +- Chi phí: Không tốn tiền. Chỗ khó là kiên quyết đòi một bản hợp đồng bằng văn bản tại chỗ, không chấp nhận「nạp tiền trước đã」. +- Nói thẳng: Người bán nhận tiền trả trước của bạn thì phải ký hợp đồng bằng văn bản với bạn, ghi rõ cách hoàn tiền. Người bán gặp rủi ro kinh doanh lớn thì phải ngừng nhận tiền trả trước, muốn đóng cửa hay chuyển điểm cũng phải báo bạn trước. Lúc đó bạn có quyền yêu cầu bên đó tiếp tục phục vụ, hoặc hoàn lại số dư chưa tiêu. Người bán vi phạm các quy định này sẽ bị phạt tiền. +- Lợi ích: Quy chế quy định, người bán nhận tiền trả trước phải lập hợp đồng bằng văn bản và phải thỏa thuận trong hợp đồng cách hoàn tiền. Người bán có rủi ro kinh doanh lớn thì phải ngừng nhận tiền trả trước. Người bán quyết định đóng cửa hay chuyển địa điểm thì phải báo trước cho người tiêu dùng. Người tiêu dùng có quyền yêu cầu bên đó tiếp tục thực hiện, hoặc hoàn lại số dư chưa tiêu. Vi phạm thì có thể bị phạt bằng 1 đến 10 lần thu nhập bất hợp pháp; không có thu nhập bất hợp pháp thì phạt dưới 500.000 tệ +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2024). Quy chế thi hành Bộ luật Bảo vệ quyền lợi người tiêu dùng của Cộng hoà nhân dân Trung Quốc（国务院令第 778 号，2024 年 7 月 1 日施行）. 第二十二条：「经营者以收取预付款方式提供商品或者服务的，应当与消费者订立书面合同，约定商品或者服务的具体内容、价款或者费用、预付款退还方式、违约责任等事项。」「经营者未按照约定提供商品或者服务的，应当按照消费者的要求履行约定或者退还预付款。」「经营者出现重大经营风险，有可能影响经营者按照合同约定或者交易习惯正常提供商品或者服务的，应当停止收取预付款。经营者决定停业或者迁移服务场所的，应当提前告知消费者……消费者……有权要求经营者继续履行提供商品或者服务的义务，或者要求退还未消费的预付款余额。」— khi người bán cung cấp hàng hoá hoặc dịch vụ theo cách nhận tiền trả trước thì phải lập hợp đồng bằng văn bản với người tiêu dùng, thỏa thuận nội dung cụ thể của hàng hoá hay dịch vụ, giá hay phí, cách hoàn tiền trả trước, trách nhiệm khi vi phạm và các việc liên quan. Người bán không cung cấp hàng hoá hay dịch vụ đúng thỏa thuận thì phải theo yêu cầu của người tiêu dùng mà thực hiện, hoặc hoàn lại tiền trả trước. Người bán gặp rủi ro kinh doanh lớn, có thể ảnh hưởng tới việc cung cấp hàng hoá hay dịch vụ bình thường theo thỏa thuận hợp đồng hoặc thói quen giao dịch, thì phải ngừng nhận tiền trả trước. Người bán quyết định đóng cửa hoặc chuyển địa điểm phục vụ thì phải báo trước cho người tiêu dùng…… người tiêu dùng…… có quyền yêu cầu người bán tiếp tục nghĩa vụ cung cấp hàng hoá hay dịch vụ, hoặc yêu cầu hoàn lại số dư tiền trả trước chưa tiêu；Điều 五十 về chế tài phạt. +- Ghi chú: Quy chế này cho bạn cơ sở để kiện tụng sau, không phòng ngừa được trước. Người bán mà bỏ chạy thật thì tiền nhiều khả năng đã không còn nằm trong tài khoản. Vì thế hãy làm theo mục「phòng tập trả theo lượt hoặc theo chu kỳ ngắn」trong phần này, nạp ít vào hơn, gặp chuyện thật thì dựa vào quy chế này mà đòi quyền. Cùng quy chế đó, Điều 二十七 còn quy định: người bán nhận tiền đặt cọc phải thỏa thuận trước về cách thức, thủ tục và thời hạn hoàn lại, không được đặt điều kiện hoàn tiền phi lý. + +### 23. Món mua lớn không thiết yếu thì đặt thời gian chờ, mua hàng trên mạng thì dùng luôn quyền trả hàng trong bảy ngày không cần lý do + +- Chi phí: Không tốn tiền. Thứ bạn muốn sẽ phải chờ thêm một ngày mới có. Phí vận chuyển khi trả hàng thường do bạn tự bỏ. Chỗ khó là nhịn không đặt hàng ngay trong ngày. +- Nói thẳng: Số tiền tiết kiệm được là tiền của những món ngày hôm sau bạn đã không còn muốn. Hàng mua trên mạng, tính từ ngày bạn nhận hàng, trong bảy ngày có thể trả lại không cần nêu lý do. Có mấy loại không được trả: hàng làm riêng theo yêu cầu của bạn, hàng tươi sống và dễ hỏng, hàng số đã mở bao bì, và báo in. +- Lợi ích: Số tiền tiết kiệm là phần trong số hàng mua bốc đầu mà「ngày hôm sau đã không còn muốn」. Phần này chiếm bao nhiêu thì không có một tỷ lệ chung. Hàng mua trên mạng, tính từ ngày nhận hàng, trong bảy ngày có thể trả lại không cần lý do. Có bốn loại ngoại lệ: hàng làm riêng theo yêu cầu, hàng tươi sống dễ hỏng, hàng số đã mở bao bì, và báo in (Trung Quốc, áp dụng từ 2017) +- Mức bằng chứng: C +- Nguồn: Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2020, sửa đổi). Biện pháp tạm thời về trả hàng trong bảy ngày không cần lý do đối với hàng mua trên mạng（总局令第 31 号）. +- Ghi chú: Bản thân「thời gian chờ」là kinh nghiệm thực hành, không có tài liệu nào chứng minh trực tiếp. Khi thời gian chờ không giữ được, còn quyền trả hàng trong bảy ngày không cần lý do. Trước khi trả hàng, xác nhận hai việc: món hàng đó có nằm trong danh sách không được trả không, và bao bì còn nguyên vẹn không. + +### 24. Đừng tích trữ hàng vì giá gạch chéo và đợt khuyến mãi lớn + +- Chi phí: Không tốn tiền. Cái giá phải trả là có những mức giảm giá thật bạn cũng bỏ lỡ. Bạn cần nhớ giá thường ngày của mấy món mình hay mua. Chỗ khó là ngày khuyến mãi lớn nhịn không tích trữ. +- Nói thẳng: Người ta đánh giá một món đắt hay rẻ dựa trên con số nhìn thấy đầu tiên, sau đó điều chỉnh bao nhiêu cũng không đủ. Vì thế「原价 999 现价 499」sẽ khiến bạn ước lượng độ rẻ của nó cao hơn thực tế một cách có hệ thống. Mà theo quy định, mốc dùng để tính giá giảm chỉ là giá giao dịch thấp nhất của chính cửa hàng đó trong bảy ngày trước đợt khuyến mãi. Cái giá bị gạch chéo không nhất định đã bán thật. +- Lợi ích: Phán đoán của con người bị con số nhìn thấy đầu tiên ghim lại, sau đó điều chỉnh thường không đủ bù. Vì thế「原价 999 现价 499」sẽ nâng đánh giá của bạn về mức「rẻ hơn bao nhiêu」lên một cách có hệ thống. Quy định giám sát ấn định, mốc để tính giá giảm phải là giá giao dịch thấp nhất của cùng một cửa hàng trong bảy ngày trước đợt khuyến mãi, mức giảm bao nhiêu phải lấy giá thấp nhất đã thực sự bán ra trong bảy ngày đó của cửa hàng làm chuẩn. Cái giá bị gạch chéo không nhất định đã bán thật (Trung Quốc, áp dụng từ 2020) +- Mức bằng chứng: C +- Nguồn: Tversky A, Kahneman D (1974). Judgment under Uncertainty: Heuristics and Biases. Science 185(4157):1124-1131. ；Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2020). Quy chế tạm thời quản lý hành vi khuyến mãi（总局令第 32 号，第二十一条）. +- Ghi chú: Hiệu ứng neo là kết luận từ thí nghiệm trong phòng, chưa ai tính được tích trữ hàng thật sự lỗ bao nhiêu tiền. Tích trữ hàng còn vài khoản chi bạn sẽ không ghi vào sổ: hàng hết hạn, chỗ chứt, và「nhà có sẵn thì dùng nhiều hơn chút」. Chỉ tích trữ loại hàng thoả cả ba điều: hạn sử dụng dài, chắc chắn dùng hết, và đơn giá thật sự thấp hơn giá trước đó. +### 25. Bảo hiểm thì mua loại chỉ bảo vệ trước, coi「hoàn tiền」「chia lợi nhuận」là phần không được bảo đảm + +- Chi phí: Bảo hiểm loại chỉ bảo vệ đến hạn không hoàn lại tiền, bạn sẽ có cảm giác「tiền này mất trắng」. Phần tiết kiệm thì bạn phải tự lo. Chỗ khó là chấp nhận số tiền đó không lấy lại được. +- Nói thẳng: Phần tiền hoàn lại và phần chia lợi nhuận, công ty bảo hiểm không bảo đảm trả cho bạn. Trong tờ giấy giới thiệu bắt buộc phải chữ in đậm「未来的保单红利为非保证利益」. Số tiền phí bảo hiểm bạn trả thêm cho phần hoàn tiền, coi như bạn gửi công ty bảo hiểm giữ hộ, lợi nhuận không được bảo đảm, muốn lấy ra cũng không dễ. +- Lợi ích: Quy tắc giám sát yêu cầu tờ giấy giới thiệu của sản phẩm chia lợi nhuận phải dùng chữ in đậm cảnh báo「未来的保单红利为非保证利益，其红利分配是不确定的」. Bảng minh hoạ lợi nhuận chính là bảng tính lợi nhuận mà công ty bảo hiểm đưa ra cho bạn xem, trần chênh lệch lãi của nó bằng 4.5% trừ đi lãi suất dự kiến của sản phẩm. Lãi suất dự kiến là mức sinh lời ghi cố định trong hợp đồng. Phần phí bảo hiểm trả thêm của loại hoàn tiền và loại chia lợi nhuận, coi như tiền gửi công ty bảo hiểm giữ. Lợi nhuận không được bảo đảm, thanh khoản cũng kém, lấy ra không dễ (Trung Quốc, áp dụng từ 2023) +- Mức bằng chứng: C +- Nguồn: Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (中国银保监会) (2022). Quy tắc công khai thông tin sản phẩm bảo hiểm nhân thọ trên một năm（银保监规〔2022〕24 号）. +- Ghi chú:Tranh cãi. Với người hoàn toàn không tiết kiệm được tiền thì bảo hiểm loại hoàn tiền bắt họ đóng đúng hạn, tác dụng đó có giá trị thực tế. Lời khuyên ở trên nhắm vào người bình thường tự kiểm mình được.「Mua loại chỉ bảo vệ, lấy tiền phí tiết kiệm đi tự đầu tư」tốt hơn「mua loại hoàn tiền」bao nhiêu, phải xem từng sản phẩm, phần này không có số liệu hệ thống nào để dẫn. Không phải lời khuyên đầu tư. + +### 26. Mua đủ bảo hiểm trách nhiệm dân sự bên thứ ba: hạn mức bảo hiểm bắt buộc thống nhất toàn quốc và không cao, phần vượt ra do túi bạn trả + +- Chi phí: Bảo hiểm trách nhiệm dân sự bên thứ ba thương mại, mỗi năm vài trăm tới hơn nghìn tệ. Mức bảo hiểm mua càng cao thì tiền phí tính trên mỗi đơn vị bảo hiểm càng rẻ. +- Nói thẳng: Khi bạn có trách nhiệm, bảo hiểm bắt buộc bồi thường tối đa ba khoản: tử vong và tổn thương 18 vạn tệ, chi phí y tế 1.8 vạn tệ, tổn thất tài sản 2000 tệ. Ba khoản tính riêng, không chuyển qua lại được. Tiền bồi thường khi làm người khác bị thương rất dễ vượt các hạn mức đó, phần vượt ra bạn tự bỏ ra túi, nên phải mua đủ bảo hiểm bên thứ ba. +- Lợi ích: Bảo hiểm bắt buộc trách nhiệm cho tai nạn giao thông áp dụng hạn mức trách nhiệm thống nhất trên toàn quốc. Hạn mức chia thành bốn khoản tính riêng: tử vong và tổn thương, chi phí y tế, tổn thất tài sản, và bồi thường khi bạn không có lỗi. Sau cải cách bảo hiểm xe toàn diện năm 2020, hạn mức trách nhiệm tổng đang áp dụng là 20 vạn tệ. Trong 20 vạn tệ đó, tử vong và tổn thương 18 vạn tệ, chi phí y tế 1.8 vạn tệ, tổn thất tài sản 0.2 vạn tệ. Khi bạn không có lỗi, tử vong và tổn thương 1.8 vạn tệ, chi phí y tế 1800 tệ, tổn thất tài sản 100 tệ. Phần vượt hạn mức do người gây thiệt hại tự chịu, bên gây ra va chạm phải tự bỏ tiền. Xe không mua bảo hiểm bắt buộc thì bị giữ lại, còn bị phạt bằng 2 lần tiền phí bảo hiểm phải đóng +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2006, sửa đổi 2012). Quy chế bảo hiểm bắt buộc trách nhiệm cho tai nạn giao thông của phương tiện cơ giới（国务院令第 462 号，第 618 号令修改）第二十三条：「机动车交通事故责任强制保险在全国范围内实行统一的责任限额。责任限额分为死亡伤残赔偿限额、医疗费用赔偿限额、财产损失赔偿限额以及被保险人在道路交通事故中无责任的赔偿限额。」「责任限额由保监会会同国务院公安部门、国务院卫生主管部门、国务院农业主管部门规定。」— bảo hiểm bắt buộc này áp dụng hạn mức trách nhiệm thống nhất trên toàn quốc. Hạn mức trách nhiệm chia thành hạn mức bồi thường tử vong và tổn thương, hạn mức bồi thường chi phí y tế, hạn mức bồi thường tổn thất tài sản, và hạn mức bồi thường cho người được bảo hiểm không có lỗi trong tai nạn giao thông đường bộ. Hạn mức trách nhiệm do Ủy ban Quản lý bảo hiểm phối hợp với các bộ Công an, Y tế, Nông nghiệp của Hội đồng Nhà nước quy định. 第二十一条 về bồi thường trong hạn mức; 第三十九条 về xe không mua bảo hiểm「由公安机关交通管理部门扣留机动车……处依照规定投保最低责任限额应缴纳的保险费的2倍罚款」— cơ quan quản lý giao thông của công an giữ phương tiện lại…… và phạt bằng 2 lần tiền phí bảo hiểm phải đóng theo quy định cho mức trách nhiệm tối thiểu. ；Ủy ban Giám sát Ngân hàng và Bảo hiểm Trung Quốc (中国银行保险监督管理委员会) (2020). Ý kiến chỉ đạo về việc thực hiện cải cách bảo hiểm xe toàn diện（银保监发〔2020〕41 号，四（四）：「将交强险总责任限额从12.2万元提高到20万元，其中死亡伤残赔偿限额从11万元提高到18万元，医疗费用赔偿限额从1万元提高到1.8万元，财产损失赔偿限额维持0.2万元不变」）— nâng hạn mức trách nhiệm tổng của bảo hiểm bắt buộc từ 12.2 vạn tệ lên 20 vạn tệ, trong đó hạn mức tử vong và tổn thương từ 11 vạn tệ lên 18 vạn tệ, hạn mức chi phí y tế từ 1 vạn tệ lên 1.8 vạn tệ, hạn mức tổn thất tài sản giữ nguyên 0.2 vạn tệ. （Công báo của Hội đồng Nhà nước） +- Ghi chú: Hạn mức do cơ quan tài chính giám sát quy định và có thể điều chỉnh. Bộ số này là giá trị đang áp dụng sau cải cách bảo hiểm xe toàn diện tháng 9 năm 2020, trước khi mua bảo hiểm có thể đối chiếu lại một lần. Điểm mấu chốt là cấu trúc của nó: bảo hiểm bắt buộc bồi thường tách theo từng hạng mục, hạng mục nào chưa bồi thường hết thì phần hạn mức còn lại không chuyển sang bù cho hạng mục khác. Tổng tiền bồi thường khi làm người khác bị thương có thể vượt xa nó rất nhiều, và phần chênh lệch vượt ra không có bảo hiểm nào trả giúp bạn. Mức bảo hiểm bên thứ ba nên mua bao nhiêu, cuốn sách này không đưa ra con số. Nhưng đây là một trong ít chỗ trong sách mà「bỏ thêm vài trăm tệ để đổi lấy một việc xác suất rất nhỏ, mà nếu lãi thì nén được cả một gia đình」. + +### 27. Để dành trước khoản dự phòng bằng 3 đến 6 tháng chi phí, đặt ở chỗ rút ra được bất cứ lúc nào + +- Chi phí: Không tốn tiền. Cái giá phải trả là số tiền đó phải bỏ lỡ mức sinh lời cao hơn, chỉ có thể để ở tài khoản thanh toán, quỹ tiền tệ hoặc tiền gửi rút được bất cứ lúc nào. Chỗ khó là trong lúc tích góp phải nhịn không đụng vào nó. +- Nói thẳng: Để 3 đến 6 tháng chi phí ở chỗ rút ra được bất cứ lúc nào. Như vậy khi thất nghiệp hoặc ốm bệnh, bạn không phải vay loại tiền lãi 18% một năm, cũng không phải bán lỗ những khoản đầu tư đang cầm. Tiền gửi trong ngân hàng được bảo hiểm tiền gửi, mỗi ngân hàng bồi thường tối đa 50 vạn tệ. +- Lợi ích: Có khoản dự phòng thì khi thất nghiệp hoặc ốm bệnh bạn không phải đi vay, cũng không phải bán lỗ để cầu cạnh. Lãi suất tính theo năm của vay tiêu dùng và của khoản trả tối thiểu trên thẻ tín dụng, xem mục 7 của phần này (không dùng khoản trả tối thiểu trên thẻ tín dụng). Tiền gửi trong ngân hàng được bảo hiểm tiền gửi bảo vệ, mỗi ngân hàng bồi thường tối đa 50 vạn tệ (Trung Quốc, 2015) +- Mức bằng chứng: C +- Nguồn: Hội đồng Nhà nước (国务院) (2015). Quy chế bảo hiểm tiền gửi（国务院令第 660 号，第五条）. +- Ghi chú: Khoảng「3 đến 6 tháng」là kinh nghiệm thông dụng của mọi người, không tìm được nguồn gốc chính thức hay học thuật, nên xếp mức C. Số tiền tính theo chi phí cố định hằng tháng của bạn, không tính theo thu nhập. Không phải lời khuyên đầu tư. + +### 28. Trước khi trả nợ vay mua nhà sớm, làm một bài so sánh, đừng theo cảm tính + +- Chi phí: Không tốn tiền. Dành nửa tiếng để tính một bảng. Bạn phải thành thật ước lượng mức sinh lời ổn định sau thuế mà mình lấy được. +- Nói thẳng: Trả nợ vay mua nhà sớm nghĩa là chắc chắn kiếm được bằng lãi suất vay mua nhà của bạn. Vì thế chỉ hỏi một việc: mức sinh lời sau thuế mà bạn lấy được ổn định trong dài hạn có cao hơn lãi suất vay mua nhà không? Cao thì đừng vội trả, không cao hoặc không chắc thì hãy trả. Với khoản trả gốc và lãi bằng nhau, những năm đầu phần lãi chiếm nhiều, trả càng sớm thì tiết kiệm càng nhiều. Mục「tính thử trả sớm」trong app của ngân hàng cho ra con số trực tiếp. +- Lợi ích: Cách tính thế này. Trước hết đặt lãi suất vay mua nhà hằng năm là r, theo đúng hợp đồng của bạn. Ngày 20 tháng 8 năm 2026, LPR kỳ hạn trên 5 năm là 3.5%, lãi suất thực thi là mức LPR cộng hoặc trừ điểm. Rồi đặt g là mức sinh lời quy đổi cả năm sau thuế của một hướng đầu tư khác mà bạn có thể chọn: tỷ lệ bình quân mỗi năm kiếm được khi đổi số tiền đó sang chỗ khác. Mức sinh lời chắc chắn của việc trả nợ sớm chính là r, mỗi khoản trả bớt tức là chắc chắn kiếm được r. Nếu g lớn hơn r một cách ổn định, để tiền trong đầu tư là hơn. Nếu g nhỏ hơn r, hoặc g không chắc chắn, thì trả nợ sớm tức là khoá lại một mức r không rủi ro. Với khoản trả gốc và lãi bằng nhau, trong những năm đầu phần lãi bạn trả vào chiếm nhiều, nên trả càng sớm thì lãi tiết kiệm được càng nhiều. Điều này xem thẳng con số ở mục「tính thử trả sớm」trong app của ngân hàng (Trung Quốc, lãi suất năm 2026) +- Mức bằng chứng: C +- Nguồn: Trung tâm Liên ngân hàng liên ngành, được Ngân hàng Nhân dân Trung Quốc uỷ quyền (2026). Thông báo lãi suất báo giá thị trường vay (LPR) ngày 20 tháng 8 năm 2026. +- Ghi chú: Ở đây chỉ đưa ra cách tính, không kết luận. Trường hợp lợi nhuận đầu tư ổn định vượt lãi suất vay mua nhà là ít. Nhưng dù chọn bên nào, bạn cũng phải để dành đủ tiền mặt dự phòng. Ngoài ra hãy xem trong hợp đồng có điều khoản phạt trả nợ sớm hay không. Không phải lời khuyên đầu tư. + +### 29. Mua hàng trên mạng thì nhìn quy tắc của nền tảng và luật, không nghe người livestream, không tin「đánh giá tốt」 + +- Chi phí: Không tốn tiền. Trước khi đặt hàng, xem thêm điều khoản hậu mãi và quy tắc về tiền cọc. Sau khi nhận hàng, giữ lại đơn hàng, ảnh chụp màn hình phát lại của buổi livestream và lịch sử chat. Chỗ khó là không bị không khí trong buổi livestream dẫn dắt. +- Nói thẳng: Người bán lừa bạn thì bồi thường gấp ba mức giá bạn trả, mức nào không tới 500 tệ thì tính 500 tệ. Nền tảng biết rõ người bán xâm phạm quyền mà vẫn không xử lý thì phải chịu chung trách nhiệm. Tiền cọc nhiều nhất bằng 20% giá trị hợp đồng, bạn vi phạm thì mất tiền cọc, người bán vi phạm thì hoàn gấp đôi. Buôn bán giả và bình luận giả trong livestream thì phải chịu trách nhiệm, giá cả dựa trên dữ liệu hành vi thì bị cấm rõ ràng. +- Lợi ích: Người kinh doanh có hành vi gian lận thì bồi thường gấp ba mức giá, mức nào không tới 500 tệ thì tính 500 tệ. Nền tảng biết hoặc phải biết người bán xâm phạm quyền mà không xử lý thì phải chịu trách nhiệm liên đới, bạn có thể đòi thẳng khoản bồi thường đó từ nền tảng. Phòng livestream và người livestream không được bịa giao dịch, lượt theo dõi, lượt xem, lượt thích, cũng không được đăng thông tin giả, gây thiệt hại thì phải chịu trách nhiệm dân sự. Người tổ chức giao dịch giả, đánh giá giả để quảng bá giúp nhau thì bị phạt dưới 1 triệu tệ; tình huống nghiêm trọng thì phạt từ 1 triệu tới 2 triệu tệ, còn có thể bị thu hồi giấy phép. Tiền cọc không được vượt 20% giá trị hợp đồng. Bạn không thực hiện hợp đồng thì mất tiền cọc; người bán không thực hiện thì phải hoàn gấp đôi cho bạn. Dùng thông tin cá nhân để ra quyết định tự động thì không được đối xử bất công về giá giao dịch và các điều kiện khác, đây gọi là giá cả theo hành vi dữ liệu (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2013, sửa đổi). Bộ luật Bảo vệ quyền lợi người tiêu dùng（第四十四、五十五条）. （Cục Quản lý thị trường quốc gia đăng lại）；Cục Quản lý không gian mạng Trung Quốc cùng sáu bộ khác (国家网信办等七部门) (2021). Biện pháp quản lý tiếp thị trên mạng (thử nghiệm)（第十八、二十八条）. ；Ủy ban Thường vụ Quốc hội (2025, sửa đổi). Luật cạnh tranh không công bằng（第九、二十五条）. （Hội đồng Quỹ hưu trí xã hội quốc gia đăng lại）；Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự（第五百八十六、五百八十七条）. ；Ủy ban Thường vụ Quốc hội (2021). Luật bảo vệ thông tin cá nhân（第二十四条）. （Cục Thống kê quốc gia đăng lại）；Cục Quản lý thị trường tỉnh Sơn Đông (2025). Sáu vụ cạnh tranh không công bằng trên mạng điển hình. +- Ghi chú: Việc「nền tảng nào hiệu quả chi phí tốt nhất, dễ gặp hàng kém nhất」không có số liệu so sánh chính thức nào để dẫn, nên cuốn sách này không xếp hạng nền tảng. Chỉ tra được kết quả kiểm tra mẫu toàn bộ (xem mục 32, tra thông báo kiểm tra mẫu của quốc gia). Sau khi Luật cạnh tranh không công bằng sửa đổi năm 2025, quảng cáo giả chuyển từ Điều 8 cũ sang Điều 9. Chế tài phạt chuyển từ Điều 20 cũ sang Điều 25. Luật mới có hiệu lực từ ngày 15 tháng 10 năm 2025. Khi viết tài liệu, đừng chép số điều cũ. Trong loạt vụ án của Sơn Đông, có một công ty tổ chức「người ve chai」làm 75 giao dịch giả với số vốn hơn 68 vạn tệ, bị phát hiện, còn con số phạt thì không ghi trong thông báo. Quyền trả hàng trong bảy ngày không cần lý do, xem mục 23 của phần này (dùng luôn quyền trả hàng trong bảy ngày không cần lý do). + +### 30. Hàng mua trong livestream có vấn đề, trước hết yêu cầu nền tảng cung cấp thông tin người bán và người livestream, nền tảng phải cung cấp + +- Chi phí: Không tốn tiền. Chỉ cần đưa ra yêu cầu với nền tảng. +- Nói thẳng: Mua hàng trong livestream mà có vấn đề thì nút thắt phổ biến nhất là không tìm ra phải kiện ai. Lúc đó bạn có quyền trực tiếp yêu cầu nền tảng cung cấp thông tin của người vận hành phòng livestream và người livestream, cùng các hồ sơ kinh doanh liên quan. Nền tảng phải cung cấp. +- Lợi ích: Khi xảy ra tranh chấp tiêu dùng, bên kinh doanh của nền tảng tiếp thị trên livestream phải theo yêu cầu của người tiêu dùng, cung cấp thông tin về người vận hành phòng livestream, người làm tiếp thị livestream, cùng các hồ sơ hoạt động kinh doanh liên quan. Khó khăn phổ biến nhất khi bảo vệ quyền trên livestream là không tìm ra phải kiện ai, có được những thông tin này thì biết phải kiện ai +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2024). Quy chế thi hành Bộ luật Bảo vệ quyền lợi người tiêu dùng của Cộng hoà nhân dân Trung Quốc（国务院令第 778 号，2024 年 7 月 1 日施行）. ：「经营者通过网络直播等方式提供商品或者服务的，应当依法履行消费者权益保护相关义务。直播营销平台经营者应当建立健全消费者权益保护制度，明确消费争议解决机制。发生消费争议的，直播营销平台经营者应当根据消费者的要求提供直播间运营者、直播营销人员相关信息以及相关经营活动记录等必要信息。」「直播间运营者、直播营销人员发布的直播内容构成商业广告的，应当依照《中华人民共和国广告法》的有关规定履行广告发布者、广告经营者或者广告代言人的义务。」— bên kinh doanh cung cấp hàng hoá hay dịch vụ qua mạng livestream và các hình thức tương tự thì phải theo luật thực hiện nghĩa vụ bảo vệ quyền lợi người tiêu dùng. Bên kinh doanh của nền tảng tiếp thị livestream phải xây dựng và hoàn thiện chế độ bảo vệ quyền lợi người tiêu dùng, làm rõ cơ chế giải quyết tranh chấp tiêu dùng. Khi xảy ra tranh chấp tiêu dùng, bên kinh doanh của nền tảng tiếp thị livestream phải theo yêu cầu của người tiêu dùng, cung cấp các thông tin cần thiết như thông tin người vận hành phòng livestream, người làm tiếp thị livestream và hồ sơ hoạt động kinh doanh liên quan. Nội dung livestream do người vận hành phòng livestream, người làm tiếp thị livestream phát tạo thành quảng cáo thương mại thì phải theo các quy định của《Luật Quảng cáo Cộng hoà nhân dân Trung Quốc》mà thực hiện nghĩa vụ của bên phát hành quảng cáo, bên kinh doanh quảng cáo hoặc người làm đại diện quảng cáo. +- Ghi chú: Trước khi đặt hàng, quay màn hình lại khung hình livestream và những lời cam kết của người livestream. Có vấn đề thì trước hết yêu cầu nền tảng cung cấp thông tin người bán và người livestream, rồi quyết định khiếu nại hay khởi kiện. Cùng quy chế đó còn quy định ba việc. Một là bên kinh doanh không được tự mở rộng danh mục hàng「không áp dụng quyền trả hàng trong bảy ngày không cần lý do」. Hai là không được đặt「không áp dụng quyền trả hàng không cần lý do」thành lựa chọn mặc định. Ba là không được từ chối trả hàng cho bạn khi chưa có xác nhận của bạn. +### 31. Mua phải thực phẩm không an toàn: ngoài hoàn tiền còn đòi được mười lần giá, khoản bồi thường thêm dưới một nghìn tệ thì tính bằng một nghìn tệ + +- Chi phí: Không tốn tiền. Giữ lại thực phẩm, bao bì và hoá đơn, đừng bỏ đi cũng đừng rửa. Chụp ảnh ngay tại chỗ. Trước hết tìm người bán hoặc nền tảng, thương lượng không xong thì gọi 12315. Nếu thực sự kiện ra tòa thì phải mất mấy tháng. Khó nhất là nhịn không vứt món đồ đi trước. +- Nói thẳng: Mua phải thực phẩm quá hạn hay bị hỏng, chính món đồ, bao bì và hoá đơn là bằng chứng của bạn, đừng vội bỏ đi. Bạn có thể đòi bồi thường gấp mười lần giá hoặc gấp ba lần thiệt hại, dưới một nghìn tệ thì tính bằng một nghìn tệ. Chọn tìm người bán hay nhà sản xuất là tuỳ bạn, bên bạn tìm phải đền trước, không được đẩy sang bên kia. +- Lợi ích: Khi bị hại vì thực phẩm không đạt tiêu chuẩn an toàn thực phẩm, bạn có thể yêu cầu người kinh doanh bồi thường thiệt hại, cũng có thể yêu cầu nhà sản xuất. Bên nhận được yêu cầu bồi thường của bạn chịu trách nhiệm đền trước, phải trả tiền trước và không được đẩy trách nhiệm. Nghĩa là bạn tìm ai thì người đó phải trả tiền trước, không được đẩy bạn sang bên kia. Ngoài bồi thường thiệt hại, bạn còn có thể yêu cầu bên kia trả tiền phạt bằng mười lần giá hoặc gấp ba lần thiệt hại, và「增加赔偿的金额不足一千元的，为一千元」. Nhưng có một ngoại lệ:「食品的标签、说明书存在不影响食品安全且不会对消费者造成误导的瑕疵的除外」. Giải thích tư pháp còn quy định thêm mấy điều. Một, phải đồng thời thoả ba điều kiện. Người mua là do nhu cầu tiêu dùng cá nhân hoặc sinh hoạt gia đình. Thực phẩm không đạt tiêu chuẩn an toàn thực phẩm. Và không có bằng chứng cho thấy người đó biết ngay lúc mua. Đủ cả ba thì「以其实际支付价款为基数计算价款十倍的惩罚性赔偿金」. Hai, ai biết rõ không đạt chuẩn vẫn mua rồi đòi bồi thường thì Tòa án nhân dân chỉ hỗ trợ「在合理生活消费需要范围内」. Mua liên tục, kiện nhiều lần thì gộp tổng số lần mua vào cùng một phạm vi để xét. Ba, có mấy loại khiếm khuyết Tòa án không hỗ trợ tiền phạt, điều kiện là nó không khiến người tiêu dùng hiểu nhầm về an toàn thực phẩm. Các khiếm khuyết đó là: chữ và cỡ chữ không quy cách, có chính tả sai, thừa chữ hoặc thiếu chữ. Còn cách ghi khối lượng tịnh và quy cách không đúng quy cách. Còn thực phẩm không có yêu cầu bảo quản đặc biệt mà không ghi điều kiện bảo quản (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (2025, sửa đổi lần thứ ba). Luật An toàn thực phẩm (食品安全法, điều một trăm bốn mươi tám). (Cơ sở dữ liệu luật và quy định quốc gia, có hiệu lực từ 2025-12-1) ; Tòa án nhân dân tối cao (2024). Giải thích về một số vấn đề pháp lý liên quan đến việc áp dụng pháp luật trong các vụ án tranh chấp về tiền phạt trong lĩnh vực thực phẩm và dược phẩm (关于审理食品药品惩罚性赔偿纠纷案件适用法律若干问题的解释) (法释〔2024〕9, các điều một, sáu, bảy, tám, mười hai, mười ba, mười bốn, mười lăm và mười sáu, có hiệu lực từ 2024-8-22). (Công báo Tòa án nhân dân tối cao) +- Ghi chú: Thứ tự của bằng chứng quan trọng hơn số tiền. Chụp ảnh và quay video trước, đưa vào khung bao bì, ngày sản xuất, vật lạ và bản ghi mua hàng, rồi hãy liên hệ người bán hoặc nền tảng. Đừng bỏ đi hay rửa món đồ trước khi trao đổi với ai. 12315 làm việc hoà giải, không phải phán quyết. Người bán cứ nhất quyết không nhận thì chỉ còn đường kiện tòa, cách tính vụ kiện nhỏ xem mục 22 ở phần 8 (mua qua mạng, giao dịch đồ cũ bị lừa). Một điều nữa: ăn ra tóc, giun, thứ lạ như vậy có tính là「không đạt tiêu chuẩn an toàn thực phẩm」hay không, thường phải do cơ quan quản lý thị trường nhận định hoặc qua giám định, không phải do bạn nói. Nên câu「ăn ra một sợi tóc là chắc chắn có một nghìn tệ」không đứng vững. Gian lận với hàng hoá thường đền gấp ba lần, dưới 500 tệ thì tính bằng 500 tệ, xem mục 29 của phần này (mua qua mạng tin vào quy tắc của nền tảng và điều luật). Nếu tự làm thực phẩm để bán thì cần giấy phép nào, ghi nhãn thế nào, vì sao không được nói thuốc chữa bệnh, đều có quy định riêng. Chi tiết xem mục 8 đến 11 ở phần 12 (giấy phép thực phẩm, nhãn, vạch đỏ về công dụng chữa bệnh và vạch đỏ hình sự). Ngược lại, biến việc đòi bồi thường thành một nghề thì có vạch đỏ rõ ràng. Một, cố tình tạo ra hình ảnh giả để người bán bị coi là vi phạm pháp sản xuất kinh doanh, rồi dùng khiếu nại hay kiện tòa để đòi tiền bồi thường, thì có dấu hiệu tống tiền, Tòa phải chuyển đầu mối và hồ sơ cho cơ quan công an. Hai, kiện tòa dựa trên hình ảnh giả đã tạo thì Tòa bác yêu cầu; nếu cấu thành vụ án giả dối thì phạt tiền hoặc bị giữ tuỳ mức độ, và người bán còn có thể kiện ngược lại bạn về tội xâm phạm danh dự. Mục này thuộc hàng người hưởng là bạn và gia đình. +### 32. Mua món đồ lớn thì tra thông báo kiểm tra mẫu quốc gia, chứng nhận 3C và nhãn tiết kiệm năng lượng trước + +- Chi phí: Không tốn tiền. Vài phút, tra trên trang chính thức của Cục Quản lý thị trường quốc gia và của cơ quan chứng nhận. +- Nói thẳng: Năm 2024 kiểm tra mẫu quốc gia lấy 25250 lô sản phẩm, 14.3% không đạt. Kênh thương mại điện tử còn tệ hơn: lấy 7297 lô, 23.5% không đạt; cửa hàng thật là 15.1%. Trước khi mua món đồ lớn, dành vài phút tra thông báo kiểm tra mẫu, rồi tra xem chứng chỉ 3C có thật không. Sản phẩm nào phải dán nhãn tiết kiệm năng lượng thì không được thiếu. +- Lợi ích: Năm 2024 kiểm tra mẫu giám sát quốc gia lấy 25250 lô sản phẩm, tỷ lệ không đạt 14.3%. Trong đó kênh thương mại điện tử lấy 7297 lô, tỷ lệ không đạt 23.5%, cao hơn năm trước 1.8 điểm phần trăm. Cửa hàng thật lấy 8055 lô, tỷ lệ không đạt 15.1%. Chứng chỉ chứng nhận sản phẩm bắt buộc (3C) tra được tính xác thật ở mục「认证结果」trên trang của cơ quan chứng nhận. Nhãn tiết kiệm năng lượng được quản lý bắt buộc theo danh mục sản phẩm, trên nhãn phải có cấp tiết kiệm năng lượng, chỉ số tiết kiệm năng lượng và mã thông tin tiết kiệm năng lượng. Làm giả, dùng nhầm nhãn tiết kiệm năng lượng, hoặc phải dán mà không dán thì bị xử phạt theo Luật Tiết kiệm năng lượng (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Cục Quản lý thị trường quốc gia (2025). Thông báo về tình hình kiểm tra mẫu giám sát quốc gia chất lượng sản phẩm năm 2024. ; Ủy ban Chứng nhận và Công nhận quốc gia (国家认监委). Cổng tra「认证结果」trên trang chính thức. ; Ủy ban Phát triển và Cải cách quốc gia, Tổng cục Quản lý chất lượng, kiểm tra và kiểm dịch (2016). Biện pháp quản lý nhãn tiết kiệm năng lượng (能源效率标识管理办法) (Quy định số 35, các điều ba, bảy và hai mươi bảy). +- Ghi chú: Kiểm tra mẫu là cơ quan quản lý chọn một phần hàng hoá ra kiểm tra, kết quả không bằng tỷ lệ đạt của cả thị trường, cũng không dùng để xếp hạng nền tảng. Chuyện「tra một lần giúp mua được bao nhiêu hàng kém hơn」chưa có nghiên cứu nào, nên đặt mức B. Nền tảng tra của cơ quan chứng nhận là cx.cnca.cn, lúc kiểm chứng mục này không mở được, lối vào lấy theo trang chủ cnca.gov.cn. Số liệu trong thông báo kiểm tra mẫu cập nhật hằng năm, dùng thì tra bản mới nhất. +### 33. Vòng tay đá, ngọc, đồng hồ, đồ chơi thời trang: tính theo「số tiền đã tiêu」, không tính theo「số tiền để dành」 + +- Chi phí: Không tốn tiền. Chỉ là đừng đưa「sau này sẽ tăng giá」vào lý do mua. Khó nhất là nhịn không cộng thêm ngân sách vì「giữ giá」. +- Nói thẳng: Giá của loại này đã chứa sẵn một phần thuế tiêu thụ: đồng hồ mỗi chiếc từ 1 vạn tệ trở lên là 20%, trang sức vàng bạc bạch kim và kim cương là 5%, đồ trang sức và ngọc khác là 10%, thuế giá trị gia tăng tính riêng. Khi bán lại thì khoản thuế đó không hoàn lại. Nếu thật sự thích thì mua như một khoản chi tiêu, đừng tính như đầu tư. +- Lợi ích: Xem phía mua trước. Bảng khoản thuế và thuế suất của thuế tiêu thụ quy định thuế suất thế này.「Đồng hồ cao cấp」20%. Trong「trang sức quý và đá quý」, trang sức vàng bạc, trang sức bạch kim, kim cương và đồ trang sức kim cương 5%. Trang sức quý khác và đá quý 10%.「Đồng hồ cao cấp」được định nghĩa là「销售价格（不含增值税）每只在 10000 元（含）以上的各类手表」. Xem tiếp phía thu hồi vốn, căn cứ là Biện pháp quản lý tiệm cầm đồ. Điều ba mươi sáu: thời hạn cầm đồ tối đa không được quá 6 tháng. Điều ba mươi bảy nói về lãi suất tiền cầm, tức khoản lãi của số tiền tiệm cầm cho bạn vay. Khoản lãi đó tính theo lãi suất vay pháp định kỳ hạn 6 tháng do Ngân hàng Nhân dân Trung Quốc công bố và theo thời hạn cầm đồ. Lãi không được trừ trước, không được cắt tiền khỏi số tiền đưa cho bạn. Điều ba mươi tám: với cầm đồ có tài sản động (cầm đồng hồ, ngọc) thì phí tổng hợp hằng tháng không được vượt 42‰ tiền vốn, khoản phí tổng hợp thu thêm mỗi tháng không vượt 4.2% số tiền vay. Thế chấp bất động sản là 27‰, thế chấp quyền tài sản là 24‰. Điều bốn mươi: hết hạn mà trong 5 ngày không chuộc cũng không gia hạn thì tính là mất tài sản cầm đồ, hàng không còn chờ bạn chuộc. Điều bốn mươi ba: tài sản mất cầm đồ có giá định giá trên 3 vạn tệ thì có thể ủy thác đấu giá. Tiền bán đấu giá phải trừ phí đấu giá và tiền vốn cùng lãi trước. Phần còn lại trả lại cho người cầm đồ, thiếu thì vẫn đòi người cầm đồ bù (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2008). Quy định tạm thời của Cộng hoà Nhân dân Trung Quốc về thuế tiêu thụ (中华人民共和国消费税暂行条例) (Quy định của Hội đồng Nhà nước số 539, kèm Bảng khoản thuế và thuế suất của thuế tiêu thụ). ; Bộ Tài chính, Tổng cục Thuế (2006). Thông báo về việc điều chỉnh và hoàn thiện chính sách thuế tiêu thụ (关于调整和完善消费税政策的通知) (财税〔2006〕33). ; Bộ Thương mại, Bộ Công an (2005). Biện pháp quản lý tiệm cầm đồ (典当管理办法) (Quy định số 8 năm 2005, các điều ba mươi sáu, ba mươi bảy, ba mươi tám, bốn mươi và bốn mươi ba). +- Ghi chú: Cầm sang tiệm cầm đồ thì phải tính rõ: phí tổng hợp hằng tháng có trần bằng 42‰ số tiền vay, một năm quy ra khoảng một nửa số tiền vay, lãi vay còn phải tính riêng. Con số「một năm một nửa」là bảng tính sơ bộ lấy phí tháng nhân thẳng với 12, dùng để ước cỡ, không phải lãi suất năm chính xác. Thuế suất và mức phí đều là số được ghi rõ trong văn bản. Nhưng「loại này có tăng giá dài hạn không」không có số liệu chính thức nào để dẫn, cuốn sách này không dự báo và không bình luận loại nào. Có một lý do duy nhất cần chặn: coi việc tiêu tiền làm đầu tư. Thích, đeo lên thấy vui, mua như chi tiêu thì không có gì sai; lấy「giữ giá」「sau này bán được giá hơn」làm lý do để cộng thêm ngân sách thì đó mới là chỗ mất tiền. Vàng cũng chia hai đầu: vàng miếng đầu tư giao dịch theo trọng lượng và trang sức bán theo từng món không phải một thứ. Giá trang sức đã chứa phí thiết kế và phí công nghệ, khi thu hồi thường không tính phần đó. Giá niêm yết trên nền tảng đồ cũ không phải giá giao dịch, đừng coi là giá thị trường. +### 34. Ngọc và đá quý chỉ nhận báo cáo kiểm nghiệm có nhãn CMA, và tra cơ quan cấp giấy trên trang chính thức của cơ quan đó + +- Chi phí: Không tốn tiền. Trước khi đặt hàng, tra chứng chỉ của cơ quan cấp giấy một lần, vài phút là đủ. +- Nói thẳng: Chứng chỉ có thể làm giả, nên phải tra cơ quan đã cấp chứng chỉ. Cơ quan kiểm nghiệm phải có chứng chỉ mới được phát báo cáo ra bên ngoài, và trên báo cáo phải in nhãn CMA. Cơ quan cấp giấy công bố danh sách các cơ quan trên trang chính thức, bạn tự tra xem cơ quan đó có trong danh sách không. Trong các vụ do cơ quan nhà nước xử lý, có trường hợp「giấy chứng nhận kiểm nghiệm quốc gia」đi kèm hàng do chính bên bán tự liên hệ người làm ra. +- Lợi ích: Xem quy định trước, căn cứ là Biện pháp quản lý công nhận chứng chỉ của cơ quan kiểm nghiệm. Điều hai: công nhận chứng chỉ là việc đánh giá, cấp phép đối với cơ quan kiểm nghiệm「向社会出具具有证明作用的数据、结果」: xem xét nó trước, rồi mới cấp cho nó tư cách phát chứng chỉ. Điều mười lăm: dấu công nhận chứng chỉ gồm hai phần, mẫu hình CMA và số hiệu chứng chỉ công nhận. Điều hai mươi mốt: cơ quan phát dữ liệu, kết quả có giá trị chứng minh ra bên ngoài thì phải ghi dấu công nhận chứng chỉ trên báo cáo kiểm nghiệm. Điều hai mươi hai: cơ quan công nhận chứng chỉ phải công bố trên trang chính thức thông tin về các cơ quan đã được công nhận, kèm trạng thái chứng chỉ. Điều ba mươi tư: cơ quan không có chứng chỉ công nhận theo pháp luật mà tự ý phát dữ liệu, kết quả có giá trị chứng minh ra bên ngoài thì phạt 3 vạn tệ. Điều ba mươi bảy: làm giả, sửa đổi, dùng nhầm chứng chỉ hoặc dấu công nhận chứng chỉ thì phạt dưới 3 vạn tệ. Xem tiếp các vụ án chính thức, lấy từ loạt vụ điển hình「đập bại giả, làm sạch nguồn」trong thị trường thủ công truyền thống do Cục Quản lý thị trường quốc gia công bố ngày 2026-6-4. Một, một phòng livestream ở Đông Hải, Giang Tôy tuyên bố bán「tự nhiên, không tẩm axit, không tối ưu, hàng A của Myanmar」. Phòng livestream còn kèm「giấy chứng nhận kiểm nghiệm quốc gia」. Kết quả kiểm tra lại là đá thạch anh (nhuộm màu), ngọc phế thuỷ (xử lý), thuỷ tinh và nhựa. Tờ chứng chỉ đó do chính người bên bán liên hệ làm ra. Hai, cùng huyện có một vụ khác, thu giữ 469 sản phẩm pha lê tự nhiên nghi giả, kiểm tra ra 210 món là pha lê tự nhiên và 259 món là pha lê tổng hợp. Ba, một cửa hàng ở Trấn Bình, Hà Nam tự làm một bảng trưng bày để livestream. Trên bảng ghi「GIA CMA Trung tâm thẩm định ngọc và đá quý Trung Uy, đơn vị hợp tác được chứng nhận」. Cửa hàng này còn mua trên mạng 181 nhãn chứng nhận của「Trung tâm kiểm nghiệm chất lượng ngọc và đá quý có sự giám sát của quốc gia」. Những nhãn này bán kèm vòng tay (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Tổng cục Quản lý chất lượng, kiểm tra và kiểm dịch quốc gia (2015, ban hành; sửa đổi 2021). Biện pháp quản lý công nhận chứng chỉ của cơ quan kiểm nghiệm (检验检测机构资质认定管理办法) (các điều hai, mười lăm, hai mươi mốt, hai mươi hai, ba mươi tư và ba mươi bảy). ; Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2026). Cục Quản lý thị trường quốc gia công bố một loạt vụ điển hình「đập bại giả, làm sạch nguồn」trong thị trường thủ công truyền thống. ; Ủy ban Quản lý tiêu chuẩn quốc gia (国家标准化管理委员会). GB/T 16552—2017《Tên gọi ngọc và đá quý》(còn hiệu lực, áp dụng từ 2018-5-1). +- Ghi chú: Tên gọi của ngọc và đá quý có tiêu chuẩn quốc gia. Kết luận kiểm nghiệm được viết ở dạng có ngoặc như「石英岩玉（染色）」「翡翠（处理）」, hai chữ trong ngoặc chính là chỗ chênh lệch giá.「Hàng A」「mỏ cũ」「không tối ưu」là lời của người bán, không phải kết luận thẩm định. Phòng livestream bán đá thạch anh nhuộm màu nêu trên dùng đúng bộ từ đó. Cùng loạt vụ còn nhiều kiểu khác: bịa mỏ, thuê người đóng vai「công nhân mỏ」để livestream, thuê người tạo không khí trong phòng livestream, dựng kịch bản「cần tiền gấp」「người thân bệnh nặng」. Còn có tuyên bố hàng hoá「phòng chống và chữa thấp khớp, tăng miễn dịch, cải thiện chất lượng giấc ngủ, thông mạch máu」, loại này bị xử lý theo quảng cáo sai lệch. Cách tra chung cho món đồ lớn xem mục 32 của phần này (tra thông báo kiểm tra mẫu và chứng nhận 3C). Khi phòng livestream có chuyện thì tìm ai, xem mục 29 và 30 (mua qua mạng tin vào quy tắc nền tảng, phòng livestream có vấn đề thì yêu cầu nền tảng cung cấp thông tin). +### 35. Hộp mù, bài bốc ngẫu nhiên: tính theo「số tiền này sẽ không quay lại」 + +- Chi phí: Không tốn tiền. Chỉ là trước khi mua phải chốt ngân sách. Khó nhất là không ra được món mình muốn mà vẫn dừng tay. +- Nói thẳng: Người bán phải đặt quy tắc rút, xác suất rút và số lượng phiên bản giới hạn ở chỗ dễ thấy, không được điều khiển kết quả từ hệ thống phía sau, không được đặt hộp rỗng. Chế độ bảo đảm kết quả cùng hạn mức tiền và số lần rút, văn bản chỉ khuyến khích người bán làm. Hộp mù mua qua mạng, người bán đã nói rõ và bạn cũng đã xác nhận thì sau khi mở bao bì không được áp dụng quyền đổi trả lại trong bảy ngày không cần lý do. +- Lợi ích: Căn cứ là《Hướng dẫn (thử nghiệm) về hành vi kinh doanh hộp mù》. Điều chín liệt kê một loạt thông tin quan trọng, phải công bố ra bên ngoài theo cách dễ nhận biết, đảm bảo người tiêu dùng biết trước khi mua. Các thông tin đó là: tên hàng, loại, kiểu dáng, quy tắc rút, phân bố hàng, số lượng phát hành hàng giới hạn, xác suất rút, phạm vi giá trị hàng. Cùng điều đó quy định ba điều không được làm. Một, không được thay đổi kết quả rút bằng cách điều khiển từ hệ thống phía sau, không được tuỳ tiện điều chỉnh xác suất rút, tức là dẫn dắt tiêu thụ bằng hình thức khác. Hai, không được từ chối hoặc cố ý chậm giao hộp mù bằng cách chiết khấu, thu mua lại, đổi mới. Ba, không được đặt hộp rỗng. Điều mười một chỉ「khuyến khích」người kinh doanh xây dựng cơ chế bảo đảm kết quả, dẫn dắt tiêu thụ lý trí bằng cách ấn định thời gian rút, hạn mức tiền rút và hạn mức số lần rút. Điều mười bảy quản lý hộp mù bán qua internet. Người kinh doanh đã thông báo và nhắc nhở đầy đủ, người tiêu dùng cũng đã xác nhận khi mua từng lần, thì hộp mù đó sau khi mở bao bì có thể không áp dụng quyền đổi trả lại trong bảy ngày không cần lý do. Người kinh doanh không được dùng cách chọn sẵn để thay cho bước xác nhận của người tiêu dùng. Điều hai mươi mốt: không được làm hoặc giả làm hoạt động cờ bạc dưới danh nghĩa hộp mù. Điều hai mươi ba: không được bán hộp mù cho người chưa thành niên dưới 8 tuổi; bán cho người chưa thành niên từ 8 tuổi trở lên thì phải xác nhận theo pháp luật rằng đã có sự đồng ý của người giám hộ liên quan (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Cục Quản lý thị trường quốc gia (2023). Hướng dẫn (thử nghiệm) về hành vi kinh doanh hộp mù (盲盒经营行为规范指引（试行）) (国市监稽发〔2023〕39, các điều chín, mười một, mười bảy, hai mươi mốt và hai mươi ba). +- Ghi chú: Văn bản này là hướng dẫn, quản lý người kinh doanh phải làm gì, bản thân nó không phải căn cứ xử phạt. Thực sự có chuyện thì xử lý theo Luật Chất lượng sản phẩm, Luật Cạnh tranh không lành mạnh, Luật Bảo vệ quyền lợi người tiêu dùng. Có ba trường hợp là lý do cụ thể để bạn khiếu nại: không công bố xác suất, xác suất đã công bố không khớp với thực tế, mua phải hộp rỗng. Trước khi khiếu nại, quay video màn hình lại quy tắc trên trang và quá trình mở hộp. Việc sau khi mở bao bì không áp dụng quyền đổi trong bảy ngày không cần lý do có một tiền đề: người bán đã thông báo đầy đủ, bạn cũng đã xác nhận khi mua từng lần, và không được chọn sẵn thay bạn. Bán cho người chưa thành niên có quy định khác: người bán không được bán cho trẻ dưới 8 tuổi, từ 8 tuổi trở lên phải xác nhận đã có sự đồng ý của người giám hộ. Trẻ nạp tiền điện thoại rồi tặng thưởng thì xin hoàn tiền thế nào, xem mục 9 của phần này (trẻ dùng điện thoại nạp tiền tặng thưởng). Thời hạn tỉnh táo 24 giờ cho khoản mua lớn không thiết yếu, xem mục 23 (đặt thời hạn tỉnh táo cho khoản mua lớn không thiết yếu). Vòng lặp「mua xong nhạt dần, rồi lại mua tiếp」, xem mục 23 ở phần 6 (đừng trông chờ mua đồ để cải thiện tâm trạng). +### 36. Quan tâm thực phẩm có thêm gì thì mua loại đóng gói sẵn: quầy hàng bán lẻ không bắt buộc ghi bảng thành phần + +- Chi phí: Không tốn tiền. Cùng một món, loại đóng gói sẵn thường đắt hơn loại bán lẻ một chút. Liếc bảng thành phần, mấy giây. +- Nói thẳng: Pháp luật chỉ yêu cầu thực phẩm đóng gói sẵn phải có bảng thành phần, loại bán lẻ thì không. Quầy bán lẻ chỉ cần ghi tên, ngày sản xuất, hạn sử dụng và ai sản xuất. Nên「không ghi phụ gia」không có nghĩa là「không cho thêm phụ gia」, chỉ có nghĩa luật không bắt phải ghi. Nếu quan tâm thì mua loại đóng gói sẵn, cả bảng thành phần lẫn tên phụ gia đều in trên bao bì. +- Lợi ích: Điều sáu mươi bảy của Luật An toàn thực phẩm quy định bao bì thực phẩm đóng gói sẵn phải có nhãn. Nhãn phải ghi chín nội dung. Trong đó mục thứ hai là「成分或者配料表」, mục thứ bảy là「所使用的食品添加剂在国家标准中的通用名称」. Điều sáu mươi tám đòi hỏi với thực phẩm bán lẻ ít hơn nhiều. Nguyên văn là: 食品经营者销售散装食品，应当在散装食品的容器、外包装上标明「食品的名称、生产日期或者生产批号、保质期以及生产经营者名称、地址、联系方式等内容」— người kinh doanh thực phẩm bán thực phẩm bán lẻ phải ghi trên thùng chứa và bao bì ngoài「tên thực phẩm, ngày sản xuất hoặc lô sản xuất, hạn sử dụng, cùng tên, địa chỉ và thông tin liên hệ của người sản xuất kinh doanh」. Điều này không nhắc bảng thành phần, cũng không nhắc phụ gia. Thủy sản đông lạnh là nơi khác biệt này rõ nhất. GB 2760—2024 cho phép dùng axit photphoric và photphat làm chất giữ nước trong thủy sản đông lạnh, gọi tắt là chất giữ nước. Lượng dùng tối đa là 5.0 g/kg, tính theo gốc photphat. Dùng trong giới hạn đó là hợp pháp. Vấn đề là dùng quá liều và bọc đá để tăng trọng lượng. Sau khi chương trình 3·15 của Đài Truyền hình Trung Quốc năm 2025 làm rõ「虾仁违规超量添加保水剂（复合磷酸盐）及包冰增重」, cơ quan quản lý thị trường yêu cầu doanh nghiệp「产品标签应明确标注保水剂（如磷酸盐）成分、净含量（扣除冰衣重量）、配料表」(toàn quốc, Luật An toàn thực phẩm sửa đổi lần thứ ba năm 2025, có hiệu lực từ 2025-12-1) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (2025, sửa đổi lần thứ ba). Luật An toàn thực phẩm của Cộng hoà Nhân dân Trung Quốc (中华人民共和国食品安全法) (các điều sáu mươi bảy, sáu mươi tám, ban hành ngày 2025-9-12, có hiệu lực từ 2025-12-1). ; Ủy ban Y tế và Sức khoẻ quốc gia, Cục Quản lý thị trường quốc gia (2024). Tiêu chuẩn quốc gia an toàn thực phẩm, Tiêu chuẩn sử dụng phụ gia thực phẩm (GB 2760—2024, Phụ lục A, bảng A.1, mục「磷酸及磷酸盐」, mã nhóm thực phẩm 09.02.01 thủy sản đông lạnh; A.3 quy định phụ gia thực phẩm liệt kê trong bảng A.1 được dùng theo phạm vi sử dụng và lượng dùng tối đa quy định). ; Cục Quản lý thị trường Lư An (2025). Thư nhắc nhở doanh nghiệp về sản phẩm thuỷ sản chế biến và việc dùng chất giữ nước (ngày 2025-3-24). +- Ghi chú:「Không ghi」không có nghĩa là「không cho thêm」, chỉ có nghĩa luật không bắt phải ghi. Muốn biết thành phần thì mua loại đóng gói sẵn. Với tôm đông lạnh, cá đông lạnh, khối lượng tịnh trên bao bì là trọng lượng sau khi đã trừ lớp đá. Tính đơn giá theo đó mới là giá thật. Loại bán lẻ thì tính theo giá niêm yết, sau khi rã đông có thể co lại mất một đoạn lớn. Chất giữ nước trong giới hạn là hợp pháp, không phải vấn đề an toàn, đừng coi nó là chất độc. Thêm quá liều và bọc đá để tăng trọng lượng là vấn đề tiền: bạn trả giá thuỷ sản mà mua về nước. Ngược lại, bốn thứ phải ghi ở quầy bán lẻ cũng có thể không ghi. Tên, ngày sản xuất, hạn sử dụng, tên địa chỉ điện thoại của người sản xuất kinh doanh, thiếu một là vi phạm điều sáu mươi tám. Không mua cũng được, gọi 12315 cũng được. Khi bạn tự mở cửa hàng bán thực phẩm đóng túi thì phải ghi thế nào, xem mục 9 ở phần 12 (nhãn thực phẩm đóng gói sẵn). Mục này thuộc người hưởng là bạn và gia đình, ở hai hàng cao nhất. +### 37. Cổ phiếu lỗ cũng phải bán theo quy tắc định trước, đừng trông chờ mua thêm để hạ giá vốn mà lật lời + +- Chi phí: Không tốn tiền. Khó nhất là thừa nhận khoản này đã lỗ rồi bán nó đi. +- Nói thẳng: Bản năng của con người là cổ phiếu lên giá thì bán, cổ phiếu lỗ thì ôm chặt. Trong bản ghi bảy năm của mười nghìn tài khoản ở Mỹ, xác suất bán ra cổ phiếu đang lên giá cao hơn xác suất bán ra cổ phiếu đang giảm hơn nửa. Cùng nhóm người đó còn thích mua thêm vào cổ phiếu đang lỗ hơn. Nhưng những cổ phiếu lỗ họ giữ lại, sau đó một năm trung bình vẫn kém những cổ phiếu đã bán 3.4 điểm phần trăm. +- Lợi ích: Nghiên cứu này xem toàn bộ lịch sử giao dịch của 10,000 tài khoản tại một công ty chứng khoán chiết khấu ở Mỹ trong giai đoạn 1987 đến 1993. Tỷ lệ bán ra cổ phiếu lãi (PGR) là 0.148. Tỷ lệ bán ra cổ phiếu lỗ (PLR) là 0.098. Hai tỷ lệ chênh nhau 0.050, giá trị t vượt 35. Tính theo ngày, xác suất bán ra cổ phiếu đang lên giá cao hơn xác suất bán ra cổ phiếu đang giảm hơn 50%. Cùng bộ tài khoản đó cũng tích cực mua thêm cổ phiếu lỗ hơn. Tỷ lệ mua lại cổ phiếu lỗ là 0.135, cổ phiếu lãi là 0.094, giá trị t là 19. Nghiên cứu còn kiểm tra ý tưởng「giữ lại thì sẽ hồi giá」. Kết quả ngược lại: cổ phiếu lãi đã bán có lợi nhuận vượt trội trong năm sau, trung bình cao hơn cổ phiếu lỗ giữ lại 3.4 điểm phần trăm. Lợi nhuận vượt trội là phần nhiều hơn so với chỉ số thị trường. Tác giả còn tính một ví dụ theo số bình quân. Cùng bán đi 1000 đô la, bán cổ phiếu lỗ cho lợi nhuận một năm sau cao hơn khoảng 4.4% so với bán cổ phiếu lãi. Trong 4.4% đó có khoản tiết kiệm thuế sớm khi bán cổ phiếu lỗ. Phần thuế tính theo hệ thống thuế của Mỹ (Mỹ, 1987 đến 1993) +- Mức bằng chứng: A +- Nguồn: Odean T (1998). Are Investors Reluctant to Realize Their Losses? The Journal of Finance 53(5):1775-1798. +- Ghi chú: Mục này quản lý「có bán hay không」, mục 15 (không giao dịch cổ phiếu thường xuyên) quản lý「đừng liên tục ra tay」, hai mục không mâu thuẫn. Định quy tắc trước rồi làm theo, số lần ra tay lại ít hơn. Mua thêm để hạ giá vốn thì xét về số học là không sai, mua thêm một chút thì giá vốn trung bình thấp hơn một chút. Chỗ sai là coi đó là cách lật lời, tác dụng thực tế là làm lớn thêm số tiền bạn đặt vào một món cổ phiếu đó, xem mục 19 (đừng đặt hết tiền vào một món cổ phiếu). Đoạn về thuế tính theo hệ thống thuế Mỹ, Trung Quốc không áp dụng, phần còn lại vẫn đúng. Đây là dữ liệu tài khoản ở Mỹ thập niên 1980 đến 1990, chi phí giao dịch và hệ thống thuế của cổ phiếu A đều khác. Không phải lời khuyên đầu tư. +### 38. Lúc thị trường tăng giảm dữ dội thì bớt ra tay, đừng tăng vốn mua bán khi nóng nhất + +- Chi phí: Không tốn tiền. Khó nhất là lúc người ta khác đang kiếm tiền thì bạn không tăng cược theo. +- Nói thẳng: Sàn chứng khoán Thượng Hải đã tính sổ cho vòng tăng giảm dữ dội 2014 đến 2015 trên 40 triệu tài khoản. 85% hộ gia đình có tài khoản nhỏ nhất vì giao dịch thường xuyên mà kiếm được ít hơn 250 tỷ tệ so với cứ nắm giữ không động. Con số đó bằng khoảng ba phần mười giá trị cổ phiếu ban đầu của họ. Trong hai năm rưỡi thị trường đi ngang, chênh lệch này chỉ một phần trăm đến ba phần trăm. +- Lợi ích: Nghiên cứu dùng dữ liệu tài khoản theo ngày của Sàn chứng khoán Thượng Hải, phủ toàn bộ khoảng 40 triệu tài khoản. Thời gian mẫu từ 2014-7 đến 2015-12, tổng cộng 18 tháng. Trong 18 tháng đó chỉ số chung Thượng Hải tăng hơn 150% trước. Nó đạt đỉnh 5166.35 điểm ngày 2015-6-12, rồi lại giảm 40% vào cuối tháng 12. Tài khoản gia đình chia thành bốn nhóm theo giá trị vốn đầu kỳ. Nhóm nhỏ nhất chiếm 85% toàn bộ tài khoản gia đình, nhóm lớn nhất chiếm 0.5%. Tổng giá trị cổ phiếu đầu kỳ của hai nhóm gần bằng nhau, lần lượt là 880 tỷ tệ và 808 tỷ tệ. Kết quả là 85% nhỏ nhất mất 250 tỷ tệ vì giao dịch chủ động, 0.5% lớn nhất kiếm được 254 tỷ tệ. Ở đây「mất」là so với việc cứ nắm giữ suốt cùng giai đoạn, không làm gì. Mỗi con số bằng khoảng ba phần mười giá trị cổ phiếu đầu kỳ của nhóm đó, một bên 28%, một bên 31%. Để đối chiếu, hai năm rưỡi trước tháng 6 năm 2014 thị trường khá đi ngang. Trong thời gian đó lấy bất kỳ 18 tháng nào, chênh lệch cùng loại chỉ bằng 1% đến 3% giá trị đầu kỳ. Trong vòng tăng giá này, tài khoản gia đình trung bình khoảng ba tuần đổi hết cổ phiếu đang cầm, gần 18 lần mỗi năm (Trung Quốc, Thượng Hải, 2014 đến 2015; thời kỳ đối chiếu 2012 đến 2014) +- Mức bằng chứng: A +- Nguồn: An L, Lou D, Shi D (2022). Wealth redistribution in bubbles and crashes. Journal of Monetary Economics 126:134-153. ; Bản thảo làm việc do cùng các tác giả công bố (2021-10, số liệu của mục này đối chiếu theo đúng bản đó). +- Ghi chú: Nghiên cứu không nói đừng mua cổ phiếu. Nó tính rằng trong cùng một giai đoạn, phía tài khoản nhỏ càng giao dịch chủ động thì tiền càng chảy sang bên đối thủ, và mốc so sánh là「cứ nắm giữ không động」. Người lấy đi số tiền đó cũng không phải tổ chức, mà là nhóm cá nhân có tài khoản lớn nhất: trong thời gian này khoảng chín phần mười khối lượng giao dịch của Sàn chứng khoán Thượng Hải đến từ tài khoản nhà đầu tư nhỏ. Câu「chín phần mười nhà đầu tư nhỏ lỗ」lưu truyền trên mạng không tìm được số liệu chính thức để đối chiếu, mục này không dùng con số đó. Có thể so với tỷ lệ luân chuyển. Trong vòng tăng giá này tài khoản gia đình hoán đổi gần 18 lần mỗi năm, còn hộ gia đình Mỹ trong mục 15 (không giao dịch cổ phiếu thường xuyên) mỗi năm chỉ hoán đổi ba phần tư. Không phải lời khuyên đầu tư. +### 39. Muốn mua tài sản nước ngoài như cổ phiếu Mỹ thì đi qua quỹ QDII hoặc kênh Hồng Kông, đừng mở tài khoản ở công ty chứng khoán nước ngoài, cũng đừng dùng hạn mức đổi tiền hằng năm để mua + +- Chi phí: Không tốn tiền. Mua quỹ QDII ngay trong ứng dụng ngân hàng hoặc chứng khoán của bạn, mất mấy phút. Cái giá phải trả là số loại sản phẩm mua được ít hơn ở công ty chứng khoán nước ngoài. +- Nói thẳng: Với các công ty chứng khoán nước ngoài như Tiger, Futu, Longbridge, từ 2026-5 không được mở tài khoản cho người ở trong nước nữa. Tài khoản đã mở thì trong hai năm chỉ được bán ra và chuyển tiền ra, không được mua thêm. Hạn mức đổi tiền cũng không được dùng để mua cổ phiếu nước ngoài, người vi phạm thì năm đó và hai năm sau không được dùng hạn mức này. Muốn mua tài sản nước ngoài thì đi qua quỹ QDII hoặc kênh Hồng Kông. +- Lợi ích: Ngày 2026-5-22, Ủy ban Chứng khoán và Phái đoàn, Bộ Công nghiệp và Thông tin, Bộ Công an, Ngân hàng Nhân dân Trung Quốc, Cục Quản lý thị trường quốc gia, Cục Quản lý Tài chính, Cục Quản lý không gian mạng quốc gia và Cục Ngoại hối quốc gia, tám cơ quan ban hành《Phương án thực hiện cải tổ toàn diện hoạt động kinh doanh chứng khoán, phái đoán, quỹ ngoài quốc gia trái phép》. Yêu cầu chung là「经过2年集中整治，全面取缔境外证券期货基金经营机构非法跨境经营活动」. Cơ quan nước ngoài không được tiếp thị rách nước trong nước, không được「提供相关开户、处理交易指令、资金划转等交易服务」. Với người đã mở tài khoản, phương án viết「禁止境外机构为存量投资者在境内非法提供买入交易、转入资金等服务，只允许单向卖出交易并转出资金」. Hết hai năm thì「全面关停境内网站、交易软件及配套服务器」. Cùng ngày đó Ủy ban Chứng khoán và Phái đoàn thông báo sẽ tịch thu toàn bộ thu nhập bất hợp pháp của các chủ thể trong nước và nước ngoài liên quan của Tiger, Futu, Longbridge, và xử phạt nghiêm theo pháp luật. Câu hỏi trả lời cho báo chí đi kèm nêu rõ phải「引导境内投资者通过港股通、合格境内机构投资者（QDII）及跨境理财通等合法渠道开展境外投资」. Về phía đổi tiền,《Đơn xin mua ngoại tệ của cá nhân》của Cục Ngoại hối quốc gia ghi rõ việc mua ngoại tệ của cá nhân「不得用于境外买房、证券投资、购买人寿保险和投资性返还分红类保险等尚未开放的资本项目」. Người vi phạm bị đưa vào「关注名单」,「当年及之后连续 2 年不享有个人便利化额度」, còn bị xử phạt hành chính và chuyển sang điều tra chống rửa tiền, thông tin đưa vào hồ sơ tín dụng cá nhân (toàn quốc, từ 2026-5) +- Mức bằng chứng: A +- Nguồn: Ủy ban Chứng khoán và Phái đoàn Trung Quốc (2026-05-22). Tám cơ quan, trong đó có Ủy ban Chứng khoán và Phái đoàn, ban hành chung《Phương án thực hiện cải tổ toàn diện hoạt động kinh doanh chứng khoán, phái đoán, quỹ ngoài quốc gia trái phép》. ; Ủy ban Chứng khoán và Phái đoàn Trung Quốc (2026-05-22). Lãnh đạo các bộ phận liên quan của Ủy ban Chứng khoán và Phái đoàn trả lời báo chí về《Phương án thực hiện cải tổ toàn diện hoạt động kinh doanh chứng khoán, phái đoán, quỹ nước ngoài trái phép》. ; Ủy ban Chứng khoán và Phái đoàn Trung Quốc (2026-05-22). Ủy ban Chứng khoán và Phái đoàn xử lý nghiêm các vụ công ty như Tiger kinh doanh vượt biên giới trái phép. ; Cục Ngoại hối quốc gia (2021). Đơn xin mua ngoại tệ của cá nhân (Phụ lục 1). +- Ghi chú: Mấy văn bản này quản lý cơ quan nước ngoài, không ghi chế tánh cho chính người đầu tư. Người đã có tài khoản ở công ty chứng khoán nước ngoài thì trong hai năm bán ra và chuyển tiền về là xong. Ba kênh hợp pháp đều có điều kiện mở khác nhau, hỏi rõ ở ngân hàng hoặc công ty chứng khoán của bạn. Cổ phiếu Mỹ đã tăng nhiều trong quá khứ, không có nghĩa sau này cũng vậy. Mua quỹ QDII cũng theo cách ở mục 17 (quỹ chỉ số rộng) và mục 19 (đừng đặt hết tiền vào một món cổ phiếu, một nền tảng, một căn nhà), chọn chỉ số rộng, đừng đặt vào một thị trường duy nhất. Rút tiền ở nước ngoài có hạn mức riêng, xem mục 6 ở phần 21 (rút tiền ở nước ngoài). Không phải lời khuyên đầu tư. +### 40. Chỉ mua bảo hiểm cho khoản thiệt hại mà bạn không gánh nổi, khoản thiệt hại gánh nổi thì để quỹ dự phòng đảm nhiệm + +- Chi phí: Không tốn tiền. Việc cần làm là liệt kê một lượt các khoản thiệt hại lớn nhà bạn có thể gặp, xem khoản nào tiền tiết kiệm không gánh nổi. Một lần, nửa tiếng. +- Nói thẳng: Tiền phí bảo hiểm bạn trả không được hoàn lại hết, một phần phải trả chi phí và lợi nhuận của công ty bảo hiểm. Nên đáng mua là khoản thiệt hại hiếm khi xảy ra, một khi xảy ra thì nhà bạn không lấy ra nổi, ví dụ bệnh nặng, lái xe gây thương tích cho người khác, người kiếm tiền trong nhà mất. Khoản thiệt hại nhỏ mà tiền tiết kiệm gánh nổi thì để quỹ dự phòng đảm nhiệm. +- Lợi ích: Trong tiền phí bảo hiểm công ty thu vào, chỉ một phần được trả ra làm tiền bồi hoàn. Phần còn lại phải trả chi phí bán hàng và vận hành, cộng lợi nhuận của công ty. Cơ quan quản lý yêu cầu công ty bán bảo hiểm sức khoẻ ngắn hạn cho cá nhân, cứ nửa năm công bố một lần「综合赔付率」trên trang chính thức. Đó là tỷ lệ tiền đã trả so với tiền phí, muốn biết một công ty trả ra bao nhiêu thì tra trang của công ty. Nên với một khoản thiệt hại, mua bảo hiểm lâu dài tính trung bình là lỗ một chút. Đổi lại bạn trải một khoản chi lớn có thể dìm chết cả nhà thành một khoản nhỏ mỗi năm. Thiệt hại càng nhỏ, càng dễ bạn trả nổi, phần đổi này càng không đáng. Theo tiêu chí đó, mấy món cuốn sách đã viết là đáng mua. Bảo hiểm y tế một năm xem mục 20 ở phần 7 (bảo hiểm y tế một năm). Người có xe xem mục 26 của phần này (bảo hiểm trách nhiệm bên thứ ba). Người đi nước ngoài xem mục 4 ở phần 21 (y tế ở nước ngoài). Nhà có người sống dựa vào thu nhập của bạn thì xem mục 41 của phần này (bảo hiểm nhân thọ có thời hạn) (toàn quốc, từ 2021) +- Mức bằng chứng: C +- Nguồn: Văn phòng Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2021). Thông báo về việc chuẩn hoá hoạt động bảo hiểm sức khoẻ ngắn hạn (银保监办发〔2021〕7, tỷ lệ tổng hợp chi trả được công bố nửa năm một lần trên trang chính thức của công ty). +- Ghi chú: Đặt mức C vì không có số liệu chính thức nào để tính xem loại bảo hiểm nào mua là đáng, ở đây đưa ra cách cân đối. Bảo hiểm vỡ màn hình, bảo hiểm chậm chuyến bay, thiệt hại vài trăm tệ, bản thân bạn trả nổi, theo tiêu chí này không mua cũng được. Bảo hiểm bệnh hiểm nặng trả cho khoản thu nhập mấy năm ngừng làm vì bệnh, người có tiền tiết kiệm đủ gánh một hai năm ngừng việc thì có thể cân nhắc không mua. Bảo hiểm y tế cơ bản là bảo hiểm xã hội, không theo cách cân đối này, vẫn phải đóng, xem mục 9 ở phần 7 (bảo hiểm y tế cư dân). Khoản thiệt hại nhỏ dựa vào đâu, xem mục 27 của phần này (quỹ dự phòng). Mục này thuộc người hưởng là bạn và gia đình. +### 41. Nhà có người sống dựa vào thu nhập của bạn thì mua bảo hiểm nhân thọ có thời hạn cho người kiếm tiền trước, đừng mua cho con trước + +- Chi phí: Một năm vài trăm đến vài nghìn tệ, tuỳ tuổi, mức bảo hiểm và số năm mua (ước tính sơ bộ của tác giả, không có số liệu chính thức). Trước khi mua phải khai báo trung thực tình trạng sức khoẻ, một lần mất nửa tiếng. +- Nói thẳng: Bảo hiểm nhân thọ có thời hạn chỉ bảo vệ trong số năm cố định, trong những năm đó nếu người đó mất thì nhà nhận được một khoản tiền. Nó dành cho người thân sống dựa vào thu nhập của bạn, ví dụ tiền vay mua nhà chưa trả hết, con còn nhỏ. Không ai sống dựa vào bạn thì không cần. Con không kiếm tiền, bảo hiểm tử vát mua cho con còn bị luật giới hạn mức bồi thường. +- Lợi ích: Cơ quan quản lý định nghĩa bảo hiểm nhân thọ là bảo hiểm nhân thọ lấy tuổi thọ con người làm đối tượng bảo hiểm, chia thành bảo hiểm nhân thọ có thời hạn, bảo hiểm nhân thọ trọn đời, bảo hiểm viên mãn và các loại khác. Bảo hiểm nhân thọ có thời hạn lấy việc người được bảo hiểm tử vong làm điều kiện trả tiền bảo hiểm, thời hạn bảo hiểm là số năm cố định. Bảo hiểm nhân thọ trọn đời bảo vệ đến hết đời. Bảo hiểm tai nạn chỉ trả khi tai nạn dẫn đến tử vong hoặc tàn tật, chết vì bệnh thì không trả, không thay thế được bảo hiểm nhân thọ. Hợp đồng lấy tử vong làm điều kiện trả tiền phải được chính người được bảo hiểm đồng ý và chấp nhận mức bảo hiểm, nếu không thì hợp đồng vô hiệu. Cha mẹ mua bảo hiểm tử vát cho con chưa thành niên không chịu ràng buộc này, nhưng tổng tiền bảo hiểm trả do con tử vong không được vượt hạn mức do cơ quan quản lý quy định. Mức bảo hiểm của bảo hiểm nhân thọ thường ở mức hàng trăm nghìn tệ, nên lợi ích đặt là lớn. Mức bảo hiểm nên mua bao nhiêu không có cách tính chính thức. Cách của tác giả là: tiền vay mua nhà chưa trả hết, cộng chi phí nuôi con đến khi trưởng thành, cộng thêm vài năm chi phí sinh hoạt (toàn quốc, sửa đổi 2015) +- Mức bằng chứng: C +- Nguồn: Ủy ban Giám sát và Quản lý bảo hiểm Trung Quốc (2011). Biện pháp quản lý điều khoản bảo hiểm và mức phí bảo hiểm của công ty bảo hiểm nhân thọ (人身保险公司保险条款和保险费率管理办法) (Quy định của Ủy ban Giám sát và Quản lý bảo hiểm số 3 năm 2011, các điều tám và mười hai). ; Ủy ban Thường vụ Quốc hội (2015, sửa đổi). Luật Bảo hiểm của Cộng hoà Nhân dân Trung Quốc, các điều ba mươi ba và ba mươi tư. +- Ghi chú: Đặt mức C vì phía chính thức chỉ có định nghĩa và hạn mức, còn「bảo hiểm người lớn trước」suy ra từ「ai mất thu nhập thì nhà không sống nổi」. Con gặp chuyện thì nhà không mất đi một khoản thu nhập. Người kiếm tiền trong nhà mất thì tiền vay mua nhà và chi phí nuôi con vẫn còn. Khi tiền e hẹp, mua bảo hiểm nhân thọ có thời hạn và bảo hiểm y tế một năm cho người kiếm tiền trước, rồi mới đến bảo hiểm y tế cho con. Khai báo trung thực và giữ nguyên việc gia hạn xem thế nào, xem mục 20 ở phần 7 (bảo hiểm y tế một năm). Mục này thuộc hàng người hưởng thứ ②, là vợ hoặc chồng và con của bạn. +### 42. Ký bảo hiểm nhân thọ trên một năm rồi hối hận, hủy trong thời hạn cân nhắc, tiền phí trả lại gần hết + +- Chi phí: Không tốn tiền. Gọi đường dây khách hàng của công ty bảo hiểm hoặc ra quầy, mất mười mấy phút. Khó nhất là sau khi ký, dành 15 ngày để bình tĩnh đọc hợp đồng một lượt. +- Nói thẳng: Bảo hiểm nhân thọ trên một năm đều có thời hạn cân nhắc. Mua qua ngân hàng thì 15 ngày kể từ khi nhận được bản hợp đồng, trong thời gian đó hủy bảo hiểm bị trừ nhiều nhất 10 tệ phí giấy tờ. Qua thời hạn cân nhắc rồi mới hủy thì chỉ lấy lại được phần tiền trong hợp đồng quy định hoàn lại, thường ít hơn số tiền đã trả. Công ty bảo hiểm gọi điện hỏi thăm bạn có hiểu rõ không, chưa hiểu thì đừng nói là hiểu rõ. +- Lợi ích: Với bảo hiểm nhân thọ có thời hạn bảo hiểm trên một năm, cơ quan quản lý yêu cầu ghi rõ thời hạn cân nhắc tính từ ngày nào và kéo dài mấy ngày. Quyền của người mua bảo hiểm trong thời hạn cân nhắc cũng phải ghi rõ. Thời hạn cân nhắc của bảo hiểm sức khoẻ dài hạn không được ít hơn 15 ngày. Với bảo hiểm mua qua ngân hàng có thời hạn trên một năm, hợp đồng đều phải quy định thời hạn cân nhắc 15 ngày. Nó tính từ ngày bạn nhận hợp đồng bảo hiểm và ký nhận bằng văn bản, theo ngày lịch. Trên bìa loại hợp đồng bảo hiểm này phải in câu nhắc:「您在收到保险合同后15日内有全额退保（扣除不超过10元的工本费）的权利。超过15日退保有损失。」Qua thời hạn cân nhắc, bạn vẫn có thể hủy hợp đồng bất cứ lúc nào. Công ty bảo hiểm phải trong 30 ngày hoàn lại giá trị tiền mặt của hợp đồng theo quy định trong hợp đồng (phần tiền trong hợp đồng được hoàn lại). Cảnh báo rủi ro của cơ quan quản lý tài chính nói, hủy bảo hiểm nhân thọ dài hạn sau thời hạn cân nhắc thì giá trị tiền mặt thường thấp hơn tổng tiền phí đã trả. Với hợp đồng mới bảo hiểm nhân thọ trên một năm, công ty bảo hiểm phải gọi điện hỏi thăm bạn trong thời hạn cân nhắc. Cuộc gọi hỏi thăm phải xác nhận bạn có biết thiệt hại khi hủy bảo hiểm và quyền trong thời hạn cân nhắc hay không (toàn quốc, từ 2010 và 2019) +- Mức bằng chứng: A +- Nguồn: Văn phòng Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2019). Biện pháp quản lý hoạt động ngân hàng làm đại lý bảo hiểm (商业银行代理保险业务管理办法) (银保监办发〔2019〕179, các điều ba mươi sáu, sáu mươi bảy và sáu mươi tám). ; Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2019). Biện pháp quản lý bảo hiểm sức khoẻ (银保监会令 2019 年第 3 号) (điều mười lăm). ; Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2022). Quy tắc công bố thông tin sản phẩm bảo hiểm nhân thọ trên một năm (银保监规〔2022〕24, bản mô tả sản phẩm phải nêu thời điểm bắt đầu, số ngày và quyền của thời hạn cân nhắc). ; Ủy ban Giám sát và Quản lý bảo hiểm Trung Quốc (2010). Quy định về các dịch vụ cơ bản trong hoạt động bảo hiểm nhân thọ (人身保险业务基本服务规定) (Quy định số 4 năm 2010 của Ủy ban Giám sát và Quản lý bảo hiểm, điều mười lăm). ; Ủy ban Thường vụ Quốc hội (2015, sửa đổi). Luật Bảo hiểm của Cộng hoà Nhân dân Trung Quốc, các điều mười lăm và bốn mươi bảy. ; Cục Giám sát Tài chính địa phương Ninh Hạ (2026-06-16). Cảnh báo rủi ro về việc phòng chống đại lý hủy bảo hiểm trái phép. +- Ghi chú: Thời hạn cân nhắc là quy định dành cho bảo hiểm nhân thọ trên một năm. Bảo hiểm y tế một năm, bảo hiểm tai nạn, quy định không yêu cầu, có hay không phải xem điều khoản. Mua không qua ngân hàng thì số ngày theo điều khoản trong hợp đồng, nhận bản hợp đồng bảo hiểm thì mở mục「犹豫期」trước. Bảo hiểm tích lũy, bảo hiểm chia lợi nhuận số tiền phí phải trả lớn, thời hạn cân nhắc đáng dùng nhất, xem mục 25 của phần này (bảo hiểm tiêu dùng). Lợi ích đặt là「vừa」vì thứ tiết kiệm được là phần chênh lệch giữa tiền phí đã trả và giá trị tiền mặt, thường từ vài trăm đến vài nghìn tệ, không có số liệu chính thức. Mục này thuộc người hưởng là chính bạn. +### 43. Trước khi ký ở quầy ngân hàng, xem kỹ trong tay mình là tiền gửi hay bảo hiểm + +- Chi phí: Không tốn tiền. Xem thêm hai phút bìa hợp đồng trước khi ký. +- Nói thẳng: Ngân hàng có thể bán hộ bảo hiểm, nhưng không được nói thành tiền gửi. Bìa hợp đồng bảo hiểm phải in chữ lớn「保险合同」và tên công ty bảo hiểm, không được có chữ「存款」「储蓄」. Trên bìa là công ty bảo hiểm thì số tiền đó không phải tiền gửi, hủy sau thời hạn cân nhắc sẽ lỗ. Người trên 65 tuổi mua ở ngân hàng, nguyên tắc chỉ mua loại có lợi nhuận xác định. +- Lợi ích: Quy định ghi rõ, ngân hàng không được bán lẫn bảo hiểm với tiền gửi tiết kiệm, quỹ, sản phẩm quản lý tài sản của ngân hàng. Không được đơn giản so lợi nhuận của bảo hiểm với tiền gửi, không được nói lợi nhuận không chắc chắn thành lợi nhuận được bảo đảm. Giấy tờ bảo hiểm và tài liệu quảng bá không được dùng tên và biểu trưng của ngân hàng, không được có chữ「存款」「储蓄」「与银行共同推出」. Bìa hợp đồng bảo hiểm phải in「保险合同」bằng chữ không nhỏ hơn cỡ 72 và in tên công ty bảo hiểm bằng chữ không nhỏ hơn cỡ số hai. Có ba nhóm người mua bảo hiểm, nguyên tắc chỉ được bán sản phẩm có quyền lợi bảo hiểm xác định: người mua bảo hiểm trên 65 tuổi; người mua sản phẩm trả góp hằng kỳ trên 60 tuổi; người ghi thu nhập năm thấp hơn thu nhập bình quân tiêu dùng của người dân địa phương. Hợp đồng của ba nhóm người này còn phải chuyển sang công ty bảo hiểm thẩm định thủ công, không được tự động xuất hồ sơ tại điểm giao dịch. Nhân viên công ty bảo hiểm không được bán bảo hiểm trong khu vực ngân hàng. Bảo hiểm trên một năm bán qua ngân hàng có thời hạn cân nhắc 15 ngày, hủy sau thời hạn cân nhắc thì có thiệt hại (toàn quốc, từ 2019-10) +- Mức bằng chứng: B +- Nguồn: Văn phòng Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2019). Biện pháp quản lý hoạt động ngân hàng làm đại lý bảo hiểm (商业银行代理保险业务管理办法) (银保监办发〔2019〕179, các điều ba mươi, ba mươi mốt, ba mươi hai, ba mươi sáu, bốn mươi hai và bốn mươi chín). +- Ghi chú: Đặt mức B vì quy định rất rõ ràng, nhưng không có số liệu chính thức về「người bị đánh lừa mất bao nhiêu tiền」. Người lớn tuổi trong nhà dễ gặp nhất khi đi gửi tiền ở ngân hàng, đi cùng thì giúp xem bìa. Đã ký rồi thì trong 15 ngày hủy theo mục 42 của phần này (thời hạn cân nhắc). Thấy mình bị đánh lừa thì khiếu nại thế nào, xem mục 44 của phần này (đại lý hủy bảo hiểm). Mục này thuộc người hưởng là bạn và người lớn tuổi trong nhà. +### 44. Muốn hủy bảo hiểm thì tự tìm công ty bảo hiểm, đừng tìm「đại lý hủy bảo hiểm」, thấy mình bị đánh lừa thì gọi 12378 khiếu nại + +- Chi phí: Không tốn tiền. Tự gọi đường dây khách hàng của công ty bảo hiểm hoặc ra quầy, đi thêm một hai chút so với giao cho bên trung gian. +- Nói thẳng: Người quảng cáo trên mạng「hoàn tiền toàn bộ, không thành công không thu phí」phần lớn là đại lý hủy bảo hiểm trái phép. Họ muốn lấy từ ba phần mười đến năm phần mười số tiền hoàn lại, còn đòi giấy tờ tùy thân, thẻ ngân hàng và mã xác nhận của bạn. Đã hủy bảo hiểm thì sự bảo vệ cũng mất. Thấy mình bị đánh lừa thì trước tiên gọi khách hàng của công ty bảo hiểm, không hợp thì gọi 12378, miễn phí suốt quá trình. +- Lợi ích: Tháng 2026-2, Cục Quản lý Tài chính, Văn phòng Quản lý không gian mạng Trung ương, Bộ Công an, Ngân hàng Nhân dân Trung Quốc và Ủy ban Chứng khoán và Phái đoàn Trung Quốc cùng ban hành cảnh báo rủi ro. Cảnh báo nói những cách nói như「全额退保」là thông tin sai sự thật, không phù hợp với Luật Bảo hiểm và các luật quy định khác. Người làm「bảo vệ quyền lợi hộ」thu phí dịch vụ cao, bạn muốn dừng giữa chừng còn có thể bị kiện vì「违约」. Họ còn thu thập thẻ SIM, thẻ ngân hàng, hợp đồng bảo hiểm, địa chỉ nhà, trường học của con bạn. Số liệu đó bán lại thì bạn có thể gặp lừa đảo qua điện thoại và tiêu ý trái phép trên thẻ. Cảnh báo tháng 6-2026 của Cục Giám sát Tài chính địa phương Ninh Hạ viết chi tiết hơn. Loại trung gian này thỏa thuận là nếu hủy bảo hiểm thành công thì thu phí đại lý từ 30% đến 50% số tiền hủy bảo hiểm, nếu bỏ dở giữa chừng thì phải trả phạt nặng. Họ dạy người ta bịa lý do như bán hàng gây hiểu lầm, ký hộ để đi khiếu nại. Người làm theo có thể vi phạm Luật Xử phạt quản lý trật tự công cộng, nghiêm trọng hơn có thể cấu thành đồng phạm vu khống oan hoặc tống tiền. Hủy bảo hiểm thì trách nhiệm bảo hiểm kết thúc ngay. Sau này mua bảo hiểm cùng loại có thể tăng giá hoặc bị từ chối vì đã lớn tuổi hoặc đã phát hiện vấn đề sức khoẻ. Nếu thật sự bị đánh lừa thì trước tiên khiếu nại và thương lượng với khách hàng chính thức của công ty bảo hiểm, không hợp thì gọi 12378, đường dây khiếu nại và bảo vệ quyền lợi người tiêu dùng ngân hàng và bảo hiểm, miễn phí suốt quá trình. Nếu chỉ khó khăn trả tiền phí trong thời gian ngắn, trước tiên hỏi công ty bảo hiểm có làm được giảm mức đóng một lần, vay theo hợp đồng bảo hiểm hay không. Giảm mức đóng một lần nghĩa là sau này không phải đóng tiếp, mức bảo hiểm giảm theo tỷ lệ (toàn quốc, 2026) +- Mức bằng chứng: B +- Nguồn: Cục Quản lý Tài chính, Văn phòng Quản lý không gian mạng Trung ương, Bộ Công an, Ngân hàng Nhân dân Trung Quốc và Ủy ban Chứng khoán và Phái đoàn Trung Quốc (2026-02-06). Cảnh báo rủi ro về việc cảnh giác «video ngắn và livestream của các đối tượng "bảo vệ quyền lợi" trái pháp» (关于警惕不法「代理维权」短视频及直播陷阱的风险提示, đăng lại trên trang của Văn phòng Quản lý không gian mạng Trung ương). ; Cục Giám sát Tài chính địa phương Ninh Hạ (2026-06-16). Cảnh báo rủi ro về việc phòng chống đại lý hủy bảo hiểm trái phép. ; Cục Giám sát Tài chính tỉnh Hồ (2025-07-23). Cảnh báo rủi ro về việc phòng chống ngành công ngầm bảo hiểm「hủy bảo hiểm」trái phép (关于防范非法「代理退保」黑灰产的风险提示). +- Ghi chú: Cục Giám sát Tài chính tỉnh Hồ năm 2025 đã thông báo một vụ. Một người phụ nữ bị người「bạn」khuyên đi đại lý hủy bảo hiểm, đã nộp「phí tư vấn」, cuối cùng vẫn hủy theo giá trị tiền mặt. Ba tháng sau khi hủy, bà được chẩn đoán ung thư vú, hợp đồng bảo hiểm bệnh hiểm nặng đó đã mất. Có hủy hay không phải tự tính: bảo vệ còn dùng được mà không trả nổi thì trước tiên hỏi giảm mức đóng một lần. Tự hủy trong thời hạn cân nhắc, xem mục 42 của phần này (thời hạn cân nhắc). Mục này thuộc người hưởng là chính bạn. +### 45. Ghi người thụ hưởng là một người cụ thể trên hợp đồng bảo hiểm, có chuyện thì nhớ đi nhận tiền: bảo hiểm nhân thọ trong 5 năm, bảo hiểm khác trong 2 năm phải thực hiện yêu cầu + +- Chi phí: Không tốn tiền. Khi mua bảo hiểm thì điền một lần, sau này đổi người thụ hưởng phải báo công ty bảo hiểm bằng văn bản. +- Nói thẳng: Không ghi người thụ hưởng, hoặc người thụ hưởng đã mất trước, thì tiền bảo hiểm tính là tài sản thừa kế, phải dùng để trả nợ của người đã mất trước. Ghi người thụ hưởng là「vợ, họ tên」, sau đó ly hôn, lúc có chuyện thì coi như không ghi. Người nhà phải biết có hợp đồng bảo hiểm này: bảo hiểm nhân thọ phải yêu cầu trong 5 năm kể từ khi biết có chuyện, bảo hiểm khác là 2 năm. +- Lợi ích: Luật Bảo hiểm quy định người thụ hưởng do người được bảo hiểm hoặc người mua bảo hiểm chỉ định. Người mua bảo hiểm chỉ định thì phải được người được bảo hiểm đồng ý. Có ba trường hợp tiền bảo hiểm tính là tài sản thừa kế của người được bảo hiểm, chia theo quy tắc thừa kế. Một, không chỉ định người thụ hưởng, hoặc chỉ định không rõ ràng. Hai, người thụ hưởng chết trước người được bảo hiểm, và không có người thụ hưởng khác. Ba, người thụ hưởng bị mất hoặc từ bỏ quyền thụ hưởng theo pháp luật, và không có người thụ hưởng khác. Khi người thụ hưởng và người được bảo hiểm chết trong cùng một sự việc, không phân biệt được trước sau thì mặc định người thụ hưởng chết trước. Số tiền tính là tài sản thừa kế phải trả nợ thuế và nợ của người đã mất trước, phần còn lại mới chia. Giải thích của Tòa án nhân dân tối cao ghi rõ: ghi người thụ hưởng là「họ tên kèm quan hệ」, đến lúc có chuyện mà quan hệ đã thay đổi thì coi như không chỉ định người thụ hưởng. Đổi người thụ hưởng mà không báo công ty bảo hiểm thì không có hiệu lực với công ty bảo hiểm. Thời hiệu khởi kiện để nhận tiền: bảo hiểm nhân thọ là 5 năm, bảo hiểm khác là 2 năm, đều tính từ ngày bạn biết hoặc phải biết có chuyện. Mức bảo hiểm của bảo hiểm nhân thọ thường ở mức hàng trăm nghìn tệ, nên lợi ích đặt là lớn (toàn quốc, sửa đổi 2015, có hiệu lực từ 2021) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (2015, sửa đổi). Luật Bảo hiểm của Cộng hoà Nhân dân Trung Quốc, các điều hai mươi sáu, ba mươi chín, bốn mươi mốt và bốn mươi hai. ; Tòa án nhân dân tối cao (2020, sửa đổi). Giải thích về một số vấn đề liên quan đến việc áp dụng Luật Bảo hiểm của Cộng hoà Nhân dân Trung Quốc, phần ba (关于适用《中华人民共和国保险法》若干问题的解释（三）), điều chín, khoản hai, điểm ba, và điều mười. ; Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc, Điều một nghìn một trăm năm mươi chín. +- Ghi chú: Hết thời hiệu khởi kiện thì công ty bảo hiểm có quyền từ chối trả tiền, ra tòa cũng khó lấy lại. Nên người nhà ít nhất phải biết bạn đã mua bảo hiểm ở những công ty nào. Sau khi ly hôn, kết hôn lại, sinh con, quay lại đổi người thụ hưởng một lần. Sau khi người đó mất, đi lấy tiền ở từng nơi, xem mục 9 ở phần 25 (đi lấy ở từng nơi). Dùng cái chết để trả nợ thì không đi được, xem mục 13 ở phần 29 (trả nợ). Mục này thuộc hàng người hưởng thứ ②, là vợ hoặc chồng và con của bạn. \ No newline at end of file diff --git "a/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" "b/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" new file mode 100644 index 00000000..55e7dc51 --- /dev/null +++ "b/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" @@ -0,0 +1,213 @@ +[← 回总目录](../README.md) + +# 9. 普通人容易踩的法律红线 + +这一节算两笔账：你的人身自由，和你的钱。下面这些事，普通人常以为「没什么」，实际会被拘留、罚款，重的还会判刑。每条都给出法条里的具体数字，以及官方渠道公布过的案例。本节引的治安管理处罚法是 2025 年修订版，2026 年 1 月 1 日起施行，条款号和旧法不一样。刑法引的是北京市公安交管局转载的「根据刑法修正案（十一）修正」的整合文本，正文里会注明对应的修正案。条目的顺序，按「最容易不小心踩到」加「后果最重」从高到低排。 + +[← Về mục lục](../README.md) + +# 9. Vạch đỏ pháp lý dân dân thường dễ vấp + +Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của bạn. Với những việc dưới đây, người bình thường hay nghĩ「chẳng có gì」, thực tế lại bị giữ, bị phạt tiền, nặng hơn còn bị kết án. Mỗi mục đưa ra con số cụ thể trong điều luật, cùng các vụ án từng được công bố qua kênh chính thức. Luật Xử phạt quản lý trật tự công cộng dẫn trong phần này là bản sửa đổi năm 2025, có hiệu lực từ ngày 1 tháng 1 năm 2026, số điều khác với luật cũ. Bộ luật Hình sự dẫn ở đây là văn bản hợp nhất「được sửa đổi theo Bản sửa đổi (mười một) của Bộ luật Hình sự」do Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại, và mỗi mục ghi rõ bản sửa đổi tương ứng. Các mục được xếp theo「dễ vô tình vấp nhất」cộng「hậu quả nặng nhất」, từ cao xuống thấp. + +### 1. Không chuyển tiếp tin về thiên tai, dịch bệnh hay sự cố công an không rõ thật giả trong nhóm, không chỉnh sửa ảnh, không dùng AI dựng ảnh hiện trường giả + +- Chi phí: Không tốn tiền. Trước khi chuyển tiếp, nhìn qua một lần, xem nó có phải thông báo chính thức không. Không chắc thì đừng chuyển, cũng đừng sửa ảnh, đừng thêm「theo lời đồn」. Chỗ khó là nhịn được cái thói muốn làm người báo tin đầu tiên. +- Nói thẳng: Một tin về thiên tai, dịch bệnh hay sự cố công an, biết là giả mà vẫn chuyển đi thì bị phạt nặng y như tự bịa ra. Không phải bạn bịa ra thì cũng vậy: bị giữ 5 đến 10 ngày, còn phạt tiền 1000 tệ. Làm mất trật tự xã hội thì phạt tù dưới 3 năm, gây hậu quả nghiêm trọng thì phạt tù 3 đến 7 năm. Ảnh đã chỉnh sửa và ảnh hiện trường giả làm bằng AI đều tính là bạn tự bịa. +- Lợi ích: Cố tình tung tin đồn, báo sai nguy cơ, dịch bệnh, thiên tai hay sự cố công an, làm mất trật tự công cộng thì bị giữ 5 đến 10 ngày. Có thể còn phạt tiền dưới 1000 tệ. Mức nhẹ hơn là bị giữ dưới 5 ngày, hoặc phạt tiền dưới 1000 tệ. Bịa nguy cơ, dịch bệnh, thiên tai hay sự cố công an giả rồi đưa lên mạng để lan truyền thì phạt tù dưới 3 năm. Biết là giả mà cố tình truyền ra cũng thuộc mức này. Điều kiện là làm mất trật tự công cộng ở mức nặng, cách điều luật viết là「严重扰乱社会秩序」. Gây hậu quả nghiêm trọng thì phạt tù 3 đến 7 năm (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第二十九条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). 刑法修正案（九）（第三十二条，即刑法第二百九十一条之一第二款）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百九十一条之一）. （Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại） +- Ghi chú: Biết là giả mà vẫn chuyển tiếp, với tự bịa ra thì công an dùng chung một điều luật. Câu「tôi chỉ chuyển lại thôi」không được miễn trừ. Ảnh chỉnh sửa và ảnh hiện trường giả do AI tạo ra cũng tính là bạn tự bịa. Bộ Công an (公安部) năm 2025 đã công bố nhiều đợt vụ án điển hình về tin đồn mạng liên quan tới thiên tai và nguy cơ. Lúc viết phần này trang của Bộ Công an không mở được, nên không dẫn vụ án cụ thể. + +### 2. Không đăng, không chuyển tiếp ảnh, biểu tượng cảm xúc và bình luận xúc phạm liệt sĩ, không đùa cợt với cờ đỏ và quốc ca + +- Chi phí: Không tốn tiền. Với nội dung liên quan tới liệt sĩ, cờ đỏ, quốc huy, quốc ca thì không bình luận, không sửa ảnh, không chuyển tiếp. Chỗ khó là khi người ta trong nhóm đùa tiếp, nhịn không nói theo một câu. +- Nói thẳng: Đùa cợt về liệt sĩ, cờ đỏ, quốc ca thì nhẹ bị giữ 5 đến 10 ngày, hoặc phạt tiền 1000 đến 3000 tệ. Nặng hơn thì bị giữ 10 đến 15 ngày. Đủ cấu thành tội phạm thì phạt tù dưới 3 năm. Có người đăng 2 tin vu khống Nguyễn Long Bình trong nhóm WeChat của dân làng có 499 người, bị phạt kỉ luật 6 tháng, không vào tù nhưng bị hạn chế đi lại. +- Lợi ích: Dùng lời miệt thị, bịa đặt sự thật để xâm phạm tên, hình ảnh, danh dự, vinh dự của liệt sĩ, làm tổn hại lợi ích công cộng thì phạt theo mức sau. Bị giữ 5 đến 10 ngày, hoặc phạt tiền 1000 đến 3000 tệ. Xúc phạm, phủ nhận công trạng và tinh thần của liệt sĩ, làm ra rồi lan truyền lời lẽ, hình ảnh, âm thanh và video tô điểm cho cuộc chiến xâm lược, làm mất trật tự công cộng, cũng thuộc mức này. Mức nặng hơn là bị giữ 10 đến 15 ngày, còn có thể phạt tiền dưới 5000 tệ. Đến mức「情节严重」thì đủ cấu thành tội phạm, phạt tù dưới 3 năm. Xúc phạm cờ đỏ, quốc huy ở nơi công cộng, hoặc sửa lời, nhạc quốc ca, hoặc hát quốc ca theo cách méo mó, xúc phạm, khi nghiêm trọng thì cũng phạt tù dưới 3 năm. Vụ án chính thức: Tiêu Nhậm đăng 2 tin vu khống, xúc phạm Nguyễn Long Bình trong nhóm WeChat dân làng có 499 người, bị phạt kỉ luật 6 tháng (kỉ luật là hình phạt không giam vào tù nhưng bị hạn chế đi lại). Trùu Nhậm đăng trên Weibo có hơn 250 vạn người theo dõi 2 nội dung bóp méo thành tích của cán bộ, chiến binh giữ biên giới, bị phạt 8 tháng, còn phải xin lỗi công khai trên báo chí toàn quốc (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第三十五条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 英雄烈士保护法（第二十二、二十六条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第三十五条，即刑法第二百九十九条之一）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百九十九条、第二百九十九条之一）. ；Tòa án nhân dân tối cao (最高人民法院) (2022). 涉英烈权益保护十大典型案例（案例二、案例三）. +- Ghi chú: Điều này của Luật Xử phạt quản lý trật tự công cộng chỉ được thêm vào khi sửa đổi năm 2025. Trước đó, việc này chỉ có thể xử lý theo Bộ luật Hình sự, hoặc do viện kiểm sát khởi xướng vụ án dân sự vì lợi ích công cộng. Nhóm WeChat vài trăm người, trong mắt tòa án vẫn là nơi công cộng. + +### 3. Xem xong trên trang nước ngoài thì đóng lại, không chụp màn hình, không chuyển tiếp, không chuyển vào nhóm, cũng không nói theo ở phần bình luận + +- Chi phí: Không tốn tiền. Đóng trang là xong việc. Không lưu trữ, không chuyển tiếp, không chuyển sang nền tảng trong nước. Không dùng tài khoản trùng với danh tính của bạn để đăng nội dung chính trị lên nền tảng nước ngoài lâu dài. Cũng không nhận tiền bài và trợ cấp do tổ chức, đoàn thể, cá nhân nước ngoài cho. Chỗ khó là nhìn thấy nội dung gây khó chịu mà nhịn không chụp, không chia sẻ. +- Nói thẳng: Các điều luật này phạt hành động truyền ra ngoài, không phạt hành động xem. Bạn chuyển tiếp, chuyển vào nhóm, hay nói theo ở phần bình luận thì khi xử lý vụ án đều bị tính là phát tán. Nếu đến mức kêu gọi lật đổ, kêu gọi chia cắt thì phạt tù dưới 5 năm, người cầm đầu thì trên 5 năm. Sách và âm thanh video về chủ nghĩa khủng bố, cực đoan là ngoại lệ: biết là loại đó mà vẫn giữ trong tay, khi nghiêm trọng thì phạt tù dưới 3 năm. +- Lợi ích: Dùng bịa đặt tin đồn, vu khống hoặc cách khác để kêu gọi người ta lật đổ chế độ nhà nước, lật đổ chế độ xã hội chủ nghĩa thì phạt tù dưới 5 năm. Người cầm đầu (điều luật gọi là「首要分子」) hoặc người tội lớn thì phạt tù trên 5 năm, mức tù cao nhất là 15 năm. Kêu gọi chia cắt quốc gia, phá vỡ thống nhất quốc gia thì mức tù y như trên. Cùng tổ chức, đoàn thể, cá nhân nước ngoài làm hai việc đó thì phạt nặng hơn. Tổ chức, đoàn thể hoặc cá nhân trong nước và nước ngoài bỏ tiền trợ cấp cho người khác làm thì người trực tiếp chịu trách nhiệm bị phạt tù dưới 5 năm, nghiêm trọng thì trên 5 năm. Trộm, dò xét, mua, cung cấp trái phép quốc mật hoặc tình báo cho tổ chức, đoàn thể, nhân sự nước ngoài thì phạt tù 5 đến 10 năm. Trong mục này, nghiêm trọng đặc biệt thì trên 10 năm hoặc tù chung thân, nhẹ hơn thì dưới 5 năm. Tội phạm gây nguy hại an ninh quốc gia đều bị tước quyền chính trị 1 đến 5 năm, trong số năm đó bị cắt quyền bầu cử và quyền làm công chức. Có thể phạt kèm tịch thu tài sản, đưa tài sản cá nhân về của nhà nước. Phần xử phạt hành chính cũng có quy định. Không cá nhân, tổ chức nào được dùng mạng để làm hoạt động gây nguy hại an ninh, vinh dự và lợi ích quốc gia. Cũng không được kêu gọi lật đổ chế độ nhà nước, lật đổ chế độ xã hội chủ nghĩa. Không được kêu gọi chia cắt quốc gia, phá vỡ thống nhất quốc gia. Đăng hoặc truyền tải loại thông tin này thì bị xử phạt theo luật và quy chế hành chính liên quan (toàn quốc). Có một loại nữa thì thành tội mà không cần truyền ra. Biết là vật phẩm tuyên truyền chủ nghĩa khủng bố, cực đoan mà vẫn nắm giữ trái phép, đủ mức nghiêm trọng là vào tội. Vật phẩm ở đây gồm sách, tư liệu âm thanh hình ảnh hoặc đồ vật khác. Mức tù là dưới 3 năm, gồm tù giam, giam giữ (nhốt ngắn hạn) hoặc kỉ luật (không giam nhưng bị hạn chế đi lại). Còn phạt kèm hoặc chỉ phạt tiền, tức là vừa kết án vừa phạt tiền, cũng có thể chỉ phạt tiền. Căn cứ là điều một trăm hai mươi, khoản sáu của Bộ luật Hình sự, do Bản sửa đổi (chín) thêm vào. +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (1997). 刑法（第四十五、五十五、五十六、一百零三、一百零五、一百零六、一百零七、一百一十一、一百一十三条，这几条未经修正案修改）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第十三条第二款、第七十一条第一项）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). 刑法修正案（九）（新增第一百二十条之六）. 国家信访局转载全文 +- Ghi chú: Các điều luật dẫn ở mục này phạt hành vi kêu gọi người khác, truyền ra ngoài, cung cấp cho người khác, chỉ xem thì không nằm trong đó. Tài liệu chủ nghĩa khủng bố, cực đoan là ngoại lệ mà điều luật ghi rõ, chỉ riêng việc giữ trong tay đã đủ để định tội (điều một trăm hai mươi, khoản sáu), nên loại này tuyệt đối không được lưu trữ. Nhìn thấy thì dừng lại tại đó: không chụp màn hình lưu, không chuyển tiếp, không bình luận, không rủ người khác cùng chuyển. Chuyển tiếp và nói theo ở phần bình luận, khi xử lý vụ án đều bị tính là phát tán, câu「tôi chỉ chuyển lại thôi」không được miễn trừ. Nội dung lấy từ nền tảng nước ngoài còn có thể bị phạt nặng hơn. Không chắc thật giả thì càng đừng chuyển theo, xem mục 1 của phần này (không chuyển tiếp tin không rõ thật giả). Công cụ vượt tường bị phạt thế nào, xem mục 11 ở phần 11. Chuyển nội dung loại này ra ngoài không mang lại lợi ích thực chất nào, cái giá lại là mức tù, tước quyền chính trị và tịch thu tài sản. Vụ án cùng loại công khai tra cứu được chỉ thấy ở báo chí nước ngoài và văn bản do gia đình công bố, ví dụ một blogger chuyên đăng bình luận chính trị trên nền tảng nước ngoài bị phán năm 2023 là 7 năm. Cuốn sách này không xin được bản án có thể đối chiếu qua kênh chính thức, nên chỉ viết theo điều luật, không kèm vụ án. + +### 4. Video khiêu dâm thì tự xem ở nhà, đừng gửi vào nhóm, đừng bán「tư liệu」, đừng lập nhóm + +- Chi phí: Không tốn tiền. Không chuyển tiếp, không lưu vào tệp trong nhóm, cũng đừng đi làm chủ nhóm hay quản trị viên của loại nhóm đó. +- Nói thẳng: Tự xem không nằm trong điều khoản xử phạt. Gửi vào nhóm một lần là phát tán cho cả nhóm: bị giữ 10 đến 15 ngày, còn phạt tiền dưới 5000 tệ, nghiêm trọng thì phạt tù dưới 2 năm. Bán「tư liệu」để kiếm tiền thì phạt tù dưới 3 năm, nặng nhất là tù chung thân. Lập nhóm phát tán trên 30 người thì người lập, người làm quản trị, người chủ động đưa vào đều bị tính là phạm tội. +- Lợi ích: Phát tán thông tin khiêu dâm trên mạng thì bị giữ 10 đến 15 ngày, còn có thể phạt tiền dưới 5000 tệ. Mức nhẹ hơn là bị giữ dưới 5 ngày, hoặc phạt tiền 1000 đến 3000 tệ. Nội dung có liên quan đến người chưa thành niên thì phạt nặng hơn. Phát tán vật phẩm khiêu dâm đến mức「情节严重」thì phạt tù dưới 2 năm. Bán, truyền tải để kiếm tiền thì phạt tù dưới 3 năm kèm phạt tiền. Trong mục này, nghiêm trọng thì phạt tù 3 đến 10 năm, đặc biệt nghiêm trọng thì trên 10 năm hoặc tù chung thân. Lập một nhóm chủ yếu dùng để phát tán thông tin khiêu dâm điện tử, số thành viên từ 30 người trở lên thì xử lý theo tội phạm. Gây hậu quả nghiêm trọng thì cũng vậy. Người lập nhóm, người làm quản trị và người chủ động đưa nội dung vào đều bị quy kết theo tội phát tán vật phẩm khiêu dâm (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第八十条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第三百六十三、三百六十四条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2010). 关于办理利用互联网、移动通讯终端、声讯台制作、复制、出版、贩卖、传播淫秽电子信息刑事案件具体应用法律若干问题的解释（二）（第三条）. +- Ghi chú: Cả hai bộ luật này đều phạt hành vi「phát tán」, tự xem riêng không nằm trong điều khoản. Nhưng gửi vào nhóm một lần đã là phát tán cho cả nhóm.「情节严重」cần đến bao nhiêu vật thì quy định trong giải thích của tòa án, phần này chưa đối chiếu con số cụ thể. + +### 5. Việc「làm thêm」bảo bạn dùng thẻ của mình để nhận tiền, rút tiền, chuyển khoản thì không làm, dù được trả bao nhiêu hoa hồng + +- Chi phí: Không tốn tiền. Bỏ loại việc「trả tiền trong ngày, mang thẻ đi ngân hàng」. +- Nói thẳng: Việc làm thêm này kiếm được vài trăm tệ, cái giá phải trả là một hồ sơ tội phạm. Có người mang theo giấy tờ tùy thân và 1 thẻ ngân hàng tới địa điểm chỉ định, phối hợp chuyển khoản rút tiền, kiếm được hơn 500 tệ. Viện kiểm sát không khởi tố anh ta, công an vẫn bắt giữ 10 ngày, phạt tiền 1000 tệ. Người khác giao thẻ ngân hàng và thẻ SIM của mình lẫn của người khác cho bọn nhận tiền rút tiền, bị phạt tù 2 năm 3 tháng, phạt tiền 2.2 vạn tệ. +- Lợi ích: Biết là tiền và vật phẩm đến từ tội phạm mà vẫn giúp chuyển tay, mua lại, bán ra, hoặc giúp che giấu bằng cách khác thì phạt tù dưới 3 năm, kèm phạt tiền hoặc chỉ phạt tiền (vừa kết án vừa phạt tiền, cũng có thể chỉ phạt tiền). Vụ án chính thức: Vương Nhậm Hào theo chỉ thị mang giấy tờ tùy thân và 1 thẻ ngân hàng tới địa điểm chỉ định, phối hợp chuyển khoản rút tiền, kiếm được hơn 500 tệ. Viện kiểm sát không khởi tố anh ta, công an bắt giữ 10 ngày theo Luật Chống lừa đảo viễn thông và mạng, kèm phạt tiền 1000 tệ. Mãn Nhậm Hào giao thẻ ngân hàng và thẻ SIM của mình lẫn của người khác cho bọn nhận tiền, chuyển khoản, rút tiền, bị phạt tù 2 năm 3 tháng, kèm phạt tiền 2.2 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第三百一十二条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 检察机关依法惩治电信网络诈骗及其关联犯罪典型案例（案例九）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 依法惩治掩饰、隐瞒犯罪所得、犯罪所得收益犯罪典型案例（案例五）. +- Ghi chú: Cho vay thẻ ngân hàng, giúp người khác「chạy tiền」gì hậu quả thế nào, xem mục 8 ở phần 8 (cho mượn thẻ ngân hàng). Mục này nói về việc tự rút tiền, chuyển khoản. Câu「tôi không biết tiền đó từ đâu ra」khi xét xử không nhất thiết được chấp nhận. Mức công vượt mức bất thường, bắt bạn quét mặt để phối hợp, giao dịch nhìn thế nào cũng thấy không bình thường, chừng đó là đủ để kết luận bạn thực ra biết. Kết quả nhẹ nhất cũng là bị giữ 10 ngày, cộng thêm một hồ sơ tội phạm. +### 6. Có người rủ bạn làm「bọc vật liệu」để vay tiền, chia hoa hồng theo khoản vay, thì không làm việc nào cả + +- Chi phí: Không tốn tiền. Từ chối thẳng. Không cho mượn giấy tờ tùy thân và lịch sử tín dụng của chính mình. Không ký hợp đồng vay tiền chưa đọc, cũng không đi thay người khác ra ngân hàng ký trực tiếp. +- Nói thẳng: Giúp người khác「bọc vật liệu」để vay tiền rồi chia tiền, bị xử lý theo tội lừa đảo. Có một nhóm 126 người, bọc người không có việc làm thành「khách hàng chất lượng」, lừa vay ở hơn 80 ngân hàng số tiền hơn 3000 vạn tệ. Người ra đứng ký được chia ba phần mươi khoản vay. Trong 80 người đã bị xét, có 76 người bị phạt tù từ 1 năm 4 tháng đến 6 năm 6 tháng. Tiền vay nằm dưới tên ai thì nợ là của người đó. +- Lợi ích: Dùng thủ đoạn lừa đảo để lấy từ ngân hàng hoặc cơ quan tài chính khác các khoản vay, hợp đồng thanh toán, thư tín dụng, bảo lãnh. Gây thiệt hại lớn cho ngân hàng hoặc cơ quan tài chính khác thì phạt tù dưới 3 năm. Còn phạt kèm hoặc chỉ phạt tiền, tức là vừa kết án vừa phạt tiền, cũng có thể chỉ phạt tiền. Thiệt hại đặc biệt lớn, hoặc có tình tiết khác đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm kèm phạt tiền. Mốc công an khởi xướng điều tra là thiệt hại trực tiếp từ 50 vạn tệ trở lên. Vốn dĩ không định trả, lừa vay bằng lý do dự án nhập vốn bịa, hợp đồng kinh tế giả, chứng từ giả hoặc chứng thư quyền sở hữu giả thì là tội lừa vay tiền ngân hàng. Mục này lừa được 5 vạn tệ là khởi xướng, phạt tù dưới 5 năm kèm phạt tiền 2 vạn đến 20 vạn tệ. Số tiền lớn hoặc có tình tiết nghiêm trọng khác thì phạt tù 5 đến 10 năm, đặc biệt lớn thì trên 10 năm cho tới tù chung thân. Làm giả, sửa, mua bán văn bản, giấy tờ, con dấu của cơ quan nhà nước thì phạt tù dưới 3 năm, nghiêm trọng thì 3 đến 10 năm. Vụ án chính thức: một nhóm 126 người, tại 16 tỉnh thành bọc người không có việc làm thành「khách hàng chất lượng」, lừa vay ở hơn 80 chi nhánh ngân hàng số tiền hơn 3000 vạn tệ. Người「trắng tài khoản」ra vay được chia ba phần mươi khoản vay. Viện kiểm sát đã truy cứu 105「trắng tài khoản」. Trong 80 người bị xét xử sơ thẩm năm 2026, 76 người bị phạt tù từ 1 năm 4 tháng đến 6 năm 6 tháng về tội lừa vay tiền ngân hàng, kèm phạt tiền. Bốn người khác vì rút tiền hộ bọn, bị gộp nhiều tội lừa vay tiền ngân hàng và rửa tiền để tính chung thời gian tù. Bốn người này bị phạt tù từ 2 năm 4 tháng đến 7 năm 6 tháng (toàn quốc). Tội lừa vay tiền trong mục này theo văn bản Bản sửa đổi (mười một) của Bộ luật Hình sự, áp dụng từ tháng 3 năm 2021. +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第十一条，修改刑法第一百七十五条之一）. ；Quốc hội Nhân dân (全国人大) (1997). 刑法（第八十八、一百九十三、二百八十条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部) (2022). 关于公安机关管辖的刑事案件立案追诉标准的规定（二）（第二十二、四十五条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). 骗贷流水线背后藏洗钱黑手（检察日报）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). 出境入境管理法（第十二条）. （Cục Quản lý Nhập cảnh Xuất cảnh quốc gia）；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 修正). 民事诉讼法（第二百六十六条）. （Cục Quản lý thị trường tỉnh Quý Châu đăng lại） +- Ghi chú: Câu「lúc đó tôi định trả」không được chấp nhận trong vụ án đó. Cơ quan kiểm sát làm rõ những người này không có việc làm, không có tài sản cố định, vốn không thể trả nổi, tiền chia được lại tiêu hết, nên kết luận vốn dĩ không định trả. Người rút tiền hộ bọn còn có thể bị cộng thêm tội rửa tiền. Tiền vay nằm dưới tên ai, hợp đồng ai ký, người đó là người vay. Cho dù không cấu thành tội phạm thì khoản nợ này và dấu vết trên lịch sử tín dụng vẫn là của bạn. Cầm tiền rồi chạy ra nước ngoài không giải quyết được: sau khi khởi xướng mà trốn, dù bao nhiêu năm vẫn truy được. Người đã kết án mà chưa thi hành xong, đang bị coi là bị can hoặc bị cáo, đều không được xuất cảnh. Vụ án dân sự chưa kết thúc thì tòa án có thể không cho bạn xuất cảnh; người bị thi hành án không thi hành thì tòa án cũng hạn chế xuất cảnh. Rút tiền, chuyển khoản hộ người khác xem mục 5 của phần này (việc「làm thêm」bảo bạn dùng thẻ của mình nhận tiền); bị người khác mượn tên vay tiền thì phát hiện bằng cách nào, xem mục 9 ở phần 8. + +### 7. Nhặt được thẻ ngân hàng thì không rút tiền; nhặt được điện thoại, ví thì liên lạc chủ nhân hoặc giao cho công an, không giữ lại + +- Chi phí: Không tốn tiền, mất mấy phút. Gọi điện tìm chủ nhân, hoặc mang đồ đến đồn công an. +- Nói thẳng: Nhặt được đồ giữ lại dùng, số tiền đến mức「số tiền lớn」mà vẫn không chịu trả lại thì phạt tù dưới 2 năm, số tiền rất lớn thì phạt tù 2 đến 5 năm. Dùng thẻ ngân hàng nhặt được để rút tiền ở ATM còn nặng hơn, bị tính là mạo danh thẻ tín dụng của người khác; số tiền lớn thì phạt tù dưới 5 năm, còn phạt tiền 2 vạn đến 20 vạn tệ. Riêng trường hợp rút tiền, công an khởi xướng trực tiếp, không cần chủ nhân tự kiện. +- Lợi ích: Nhặt được đồ người khác đánh rơi thì phải trả lại cho chủ nhân, báo kịp để họ đến nhận, hoặc giao nộp cho công an và cơ quan có liên quan. Giữ đồ người khác bỏ lại để dùng, số tiền đến mức「số tiền lớn」mà vẫn không chịu trả lại thì cấu thành tội chiếm đoạt, phạt tù dưới 2 năm. Số tiền rất lớn thì phạt tù 2 đến 5 năm. Nhặt được thẻ tín dụng của người khác rồi dùng ở máy ATM thuộc「mạo danh thẻ tín dụng của người khác」. Đủ cấu thành tội phạm thì xử lý theo tội lừa đảo thẻ tín dụng: số tiền lớn thì phạt tù dưới 5 năm, kèm phạt tiền 2 vạn đến 20 vạn tệ. Số tiền rất lớn hoặc có tình tiết nghiêm trọng khác thì phạt tù 5 đến 10 năm, kèm phạt tiền 5 vạn đến 50 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 民法典（第三百一十四条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第一百九十六、二百七十条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2008). 关于拾得他人信用卡并在自动柜员机（ATM 机）上使用的行为如何定性问题的批复. +- Ghi chú: Tội chiếm đoạt là「chỉ xử lý khi bị kiện」, phải do chủ nhân tự ra tòa khởi kiện, công an không chủ động xử lý. Nhưng dùng thẻ nhặt được để rút tiền thì khác, không cần chủ nhân kiện, công an khởi xướng trực tiếp.「Số tiền lớn」cụ thể là bao nhiêu do giải thích của tòa án quy định, phần này chưa đối chiếu. Sau khi giao nộp đồ cho cơ quan có liên quan, khi chủ nhân đến nhận thì phải trả phí giữ giữ (điều ba trăm mười bảy của Bộ luật Dân sự). + +### 8. Ưu đãi thì lấy bằng chính danh tính thật của bạn; không mua số ảo để đăng ký hàng loạt tài khoản nhỏ, không dựa lỗ hổng để lấy lặp + +- Chi phí: Không tốn tiền. Bỏ loại việc trong「nhóm nhiệm vụ háo mùa」. Thấy nền tảng có lỗ hổng thì đừng đụng vào. Chỗ khó là nhịn được cái thói coi ưu đãi là khoản thu nhập có thể nhân bản hàng loạt. +- Nói thẳng: Lấy ưu đãi bằng chính danh tính thật của bạn, một tài khoản một lần thì không nằm trong phạm vi xử phạt. Đăng ký hàng loạt người dùng giả, tự đặt hàng rồi tự nhận đơn để lấy nhiều lần thì bị xử lý như lừa đảo: lừa được 3000 tệ là đủ phạt tù dưới 3 năm, từ 3 vạn tệ trở lên thì phạt tù 3 đến 10 năm. Có tài xế dùng số ảo giá 0.1 tệ một cái để quét mã ưu đãi, rút tiền hơn 9 vạn tệ, cả 12 người cùng vụ đều bị kết án. +- Lợi ích: Lừa tài sản của người khác hoặc của nhà nước, số tiền lớn thì phạt tù dưới 3 năm, kèm phạt tiền hoặc chỉ phạt tiền (vừa kết án vừa phạt tiền, cũng có thể chỉ phạt tiền). Số tiền rất lớn, hoặc có tình tiết nghiêm trọng khác thì phạt tù 3 đến 10 năm kèm phạt tiền. Số tiền đặc biệt lớn, hoặc có tình tiết đặc biệt nghiêm trọng khác thì trên 10 năm hoặc tù chung thân, kèm phạt tiền hoặc tịch thu tài sản. Ba mức số tiền lớn, rất lớn, đặc biệt lớn, mỗi tỉnh ấn định tiêu chuẩn địa phương trong khoảng dưới đây. Lớn là từ 3000 tệ đến hơn 1 vạn tệ. Rất lớn là từ 3 vạn tệ đến hơn 10 vạn tệ, đặc biệt lớn là hơn 50 vạn tệ. Dạy người khác cách làm, bán「hướng dẫn」thì phạt tù dưới 5 năm theo tội truyền cách phạm tội, nghiêm trọng thì 5 đến 10 năm, đặc biệt nghiêm trọng thì trên 10 năm hoặc tù chung thân. Đầu vào đầu ra trong việc mua bán số điện thoại, mã xác minh thì phạt tù dưới 3 năm theo tội xâm phạm thông tin cá nhân của công dân, đặc biệt nghiêm trọng thì 3 đến 7 năm. Dùng hệ thống tài khoản hàng loạt, hệ thống tự động đổi địa chỉ mạng, cùng nền tảng tiếp nhận hàng loạt mã xác minh qua tin nhắn, bản thân đã bị Luật Chống lừa đảo viễn thông và mạng cấm rõ ràng ở điều mười bốn. Mua bán, cho thuê, cho mượn trái phép thẻ điện thoại và tài khoản internet, giúp người khác vượt kiểm tra định danh, nằm ở điều ba mươi mốt. Vi phạm điều mười bốn thì toàn bộ số tiền kiếm được bị tịch thu, rồi phạt gấp 1 đến 10 lần số tiền kiếm được. Không kiếm được tiền, hoặc kiếm được dưới 5 vạn tệ thì phạt tiền dưới 50 vạn tệ, nghiêm trọng còn bị giữ dưới 15 ngày. Vi phạm điều ba mươi mốt thì mức phạt tiền cũng gấp 1 đến 10 lần, không kiếm được tiền hoặc kiếm được dưới 2 vạn tệ thì phạt tiền dưới 20 vạn tệ, nghiêm trọng cũng bị giữ dưới 15 ngày. Tham gia lừa đảo viễn thông và mạng mà chưa đến mức kết án thì bị giữ 10 đến 15 ngày. Số tiền kiếm được bị tịch thu, rồi phạt gấp 1 đến 10 lần. Không kiếm được tiền hoặc kiếm được dưới 1 vạn tệ thì phạt tiền dưới 10 vạn tệ. Vụ án chính thức một: tài xế giao đồ ăn Lưu Nhậm Hào mua số điện thoại ảo giá 0.1 tệ một cái, đăng ký người dùng mới của một nền tảng. Anh ta nhận mã giảm giá người mới giảm 29 trừ 15 để đặt hàng, rồi bán đảo hàng cho cửa hàng nhỏ để rút tiền. Anh ta còn dùng số ảo mời lẫn nhau để nhận thưởng từ 2 tệ đến 20 tệ. Anh ta và tài xế Đàm Nhậm Hào lừa được lần lượt hơn 9 vạn tệ và hơn 7 vạn tệ. Viện kiểm sát quận Dương Phố Thượng Hải khởi tố 12 người về tội lừa đảo, tòa án phạt tù từ 3 năm 6 tháng cho tới giam giữ 3 tháng. Vụ án chính thức hai: Vương Nhậm lợi dụng khoảng cách thời gian giữa hoàn tiền của cửa hàng và việc thu hồi xu xổ số ảo. Anh ta hoàn 1156 món, rút 224 món đồ chơi tỉ lệ, số tiền hơn 24.9 vạn tệ. Anh ta còn bán lại trên nền tảng đồ cũ, tòa án phạt tù 3 năm hoãn 4 năm về tội lừa đảo, kèm phạt tiền 3 vạn tệ. Trong thời gian xét xử anh ta trả lại tiền lỗi và nộp trước tiền phạt, tổng cộng 33 vạn tệ. Vụ án chính thức ba: mã giảm giá người dùng mới và trợ cấp tài xế của một nền tảng đặt xe công nghệ bị lấy bằng đơn ảo kiểu「tự đặt xe tự nhận chuyến」, liên quan 19 tỉnh hơn 70 thành phố, số tiền hơn 200 vạn tệ. Viện kiểm sát quận Long Thành, thành phố Triệu Dương, tỉnh Liao Ninh khởi tố 163 người, tất cả đều bị kết án có tội. Hai người chủ mưng Hoàng Nhậm, Thái Nhậm bị gộp ba tội lừa đảo, xâm phạm thông tin cá nhân của công dân và truyền cách phạm tội để tính chung thời gian tù, mỗi người phạt tù 3 năm 6 tháng. Ngoài ra 34 tên đồng phạm ở mức nhẹ, viện kiểm sát quyết định không khởi tố, nhưng vẫn chuyển cho công an xử phạt hành chính (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百五十三条之一、第二百六十六条、第二百九十五条）. （Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại）；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理诈骗刑事案件具体应用法律若干问题的解释（法释〔2011〕7 号，第一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). 反电信网络诈骗法（第十四、三十一、三十八、四十二、四十四条）. ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第九百八十五条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 剑指电信网络诈骗｜不法分子盯上了电商经济的漏洞（检察日报）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 如此「薅羊毛」涉嫌诈骗（检察日报）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2024). 辽宁朝阳龙城区：办理系列电信网络诈骗案（法治日报）. +- Ghi chú: Ranh giới không nằm ở chỗ「háo được bao nhiêu」mà ở chỗ có bịa đặt hay không. Dùng chính danh tính thật của bạn tham gia hoạt động, một tài khoản lấy một lần, mã ưu đãi dùng cho bản thân mình, đó là ưu đãi nền tảng cho bạn. Để lấy nhiều hơn mà đăng ký người dùng giả, bịa giao dịch, giấu việc mình đã hoàn tiền, là khiến đối phương tưởng là thật rồi mới giao đồ, bị xác định là lừa đảo, số tiền chỉ quyết định mức tù. Số tiền không đủ mức kết án vẫn có hậu quả: nền tảng có thể khoá tài khoản, thu hồi ưu đãi, còn có thể yêu cầu bạn hoàn lại theo nguyên tắc「tiền không được hưởng thì phải trả lại」. Tham gia lừa đảo viễn thông mà chưa tới mức phạm tội thì vẫn còn mức bị giữ và phạt tiền.「Trong nhóm có người dẫn, trả tiền theo đơn」không làm thay đổi bản chất của phần bạn tham gia. Dạy người khác cách làm, bán lại tài khoản và số điện thoại là những tội khác, nhiều tội sẽ cộng lại. Người hưởng lợi chính là bạn. Tự viết script, bán script xem mục 3 ở phần 11; bán thẻ, bán tài khoản xem mục 8 ở phần 8 (cho mượn thẻ ngân hàng) và mục 16 của phần này (không cho mượn giấy tờ tùy thân). +### 9. Không ném bất cứ thứ gì ra ngoài từ ban công, cửa sổ; đầu thuốc lá, túi rác, chai rượu đều tính + +- Chi phí: Không tốn tiền, cũng không mất thời gian. Đầu thuốc lá, túi rác, chai rượu đều cho vào thùng rác trong nhà, không thả ra từ cửa sổ và ban công. +- Nói thẳng: Ném đồ xuống dưới, dù không trúng người vẫn bị giữ 5 ngày, phạt tiền 1000 tệ; nghiêm trọng hơn thì bị giữ 10 đến 15 ngày. Trúng người thì đổi tội khác: có người uống rượu ở tầng 21 ném chai rượu rỗng, ly thuỷ tinh xuống sân trường dưới, trúng học sinh 13 tuổi gây thương tích nặng, bị phạt tù 10 năm. Người khác ném 8 viên gạch từ tầng 32 làm chết 1 người, năm 2024 bị phê duyệt án tử hình. +- Lợi ích: Ném đồ từ trên cao hoặc từ chỗ nào cao khác xuống, chỉ cần có nguy cơ làm thương người, làm hỏng tài sản hoặc gây nguy hiểm cho an toàn công cộng thì đã có thể bị giữ dưới 5 ngày, hoặc phạt tiền dưới 1000 tệ. Nghiêm trọng thì bị giữ 10 đến 15 ngày, còn có thể phạt tiền dưới 1000 tệ. Nghiêm trọng tới mức cấu thành tội ném đồ từ trên cao thì phạt tù dưới 1 năm, kèm phạt tiền hoặc chỉ phạt tiền (vừa kết án vừa phạt tiền, cũng có thể chỉ phạt tiền). Gây thương tích hoặc gây chết người thì xử lý theo tội nặng hơn. Vụ án chính thức: Lý Nhậm Khôi uống rượu buổi sáng ở tầng 21, ném chai rượu rỗng, ly thuỷ tinh xuống sân trường dưới, trúng học sinh 13 tuổi gây thương tích nặng mức hai, bị phạt tù 10 năm về tội gây nguy hiểm cho an toàn công cộng bằng phương pháp nguy hiểm. Chu Nhậm ném 8 viên gạch từ tầng 32 xuống đám đông, làm chết 1 người, năm 2024 Tòa án nhân dân tối cao phê duyệt án tử hình (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第四十三条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第三十三条，即刑法第二百九十一条之二）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百九十一条之二）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). 依法惩治危害公共安全犯罪典型案例（案例一）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 专访最高法副院长李勇：让高空抛物行为人都能被处罚. +- Ghi chú: Không trúng người vẫn có thể bị giữ, điều luật chỉ đòi hỏi「có nguy cơ」. Bộ luật Dân sự còn một quy định bồi thường dân sự riêng về việc ném đồ từ trên cao (điều một nghìn hai trăm năm mươi tư). Lúc viết phần này không mở được nguyên văn nên không dẫn. + +### 10. Không mua súng nhựa, súng bi, súng hơi trên mạng; không mua「súng đồ chơi」không rõ nguồn gốc + +- Chi phí: Không tốn tiền. Muốn bắn súng thì đến phòng bắn súng uy tín. +- Nói thẳng: Súng bi, súng hơi mua trên mạng, nếu năng lượng ở miệng nòng đạt 1.8 joule/cm² thì bị coi là súng. Giữ súng trong tay mà không có giấy phép thì phạt tù dưới 3 năm, nghiêm trọng thì phạt tù 3 đến 7 năm. Có một cặp vợ chồng bán「súng đồ chơi」, trong 43 khẩu có 18 khẩu được định giá là súng; vì không chứng minh được họ biết nên không bị khởi tố. +- Lợi ích: Giữ, tàng trữ súng mà không có giấy phép thì phạt tù dưới 3 năm, nghiêm trọng thì phạt tù 3 đến 7 năm. Súng không theo tiêu chuẩn, không bắn được đạn tiêu chuẩn, chỉ cần năng lượng ở miệng nòng lớn hơn hoặc bằng 1.8 joule/cm² thì đều bị định là súng. Năng lượng ở miệng nòng là năng lượng mà mỗi centimet vuông đạn mang theo lúc rời nòng. Nhiều khẩu súng bi, súng hơi mua trên mạng vượt quá con số này. Từ năm 2018, Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao yêu cầu với súng hơi có năng lượng thấp hơn phải đánh giá tổng hợp về ngoại hình, vật liệu, giá, mục đích sử dụng, bản thân có biết hay không, rồi mới quyết định có định tội hay không. Vụ án chính thức: vợ chồng Triệu Nhậm Hào bán「súng đồ chơi」tại chợ, thu giữ 43 khẩu, trong đó 18 khẩu được định giá đạt tiêu chuẩn súng. Vì không có chứng cứ đủ để chứng minh họ biết đó là súng, theo nguyên tắc chứng cứ còn nghi ngờ thì không khởi tố (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第一百二十八条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2018). 关于涉以压缩气体为动力的枪支、气枪铅弹刑事案件定罪量刑问题的批复（法释〔2018〕8 号）. ；Tòa án nhân dân tối cao (最高人民法院) (2018). 《批复》的理解与适用. +- Ghi chú: Chỗ mấu chốt khiến vụ án đó không khởi tố là xác định họ「không biết」. Người trưởng thành cố ý lên mạng mua súng bi, súng hơi, sau đó nói không biết đó là súng thì rất khó thành lập. Nách và dao quản lý thế nào, phần này chưa đối chiếu nguyên văn nên không viết. + +### 11. Mua máy bay không cân thì đăng ký tên thật trước; không bay ở sân bay, khu quân sự và không gian bị kiểm soát của thành phố; không phá giới hạn chiều cao + +- Chi phí: Đăng ký tên thật không tốn tiền, mất mấy phút là xong. Trước mỗi lần cất cánh, dùng kênh chính thức tra xem khu vực bạn định bay có phải không gian bị kiểm soát hay không. +- Nói thẳng: Bay mà chưa đăng ký tên thật thì phạt tiền dưới 200 tệ, nghiêm trọng thì phạt tiền 2000 đến 2 vạn tệ. Bay ở không gian bị kiểm soát mà không được phép thì phạt tiền dưới 500 tệ, nghiêm trọng thì tịch thu máy bay, còn phạt tiền 1000 đến 1 vạn tệ. Mức nặng hơn còn bị giữ 5 đến 10 ngày. Có người trả tiền để phá giới hạn chiều cao, bay lên gần 3000 mét chụp tầng mây rồi đăng lên mạng, bị giữ 5 ngày. +- Lợi ích: Bay mà không đăng ký tên thật thì bị yêu cầu chấp hành điều chỉnh, còn có thể phạt tiền dưới 200 tệ, nghiêm trọng thì phạt tiền 2000 đến 2 vạn tệ. Máy bay không cân, nhẹ, nhỏ mà chưa được phép bay ở không gian bị kiểm soát thì bị yêu cầu dừng bay, còn có thể phạt tiền dưới 500 tệ. Trong mục này, nghiêm trọng thì tịch thu máy bay, kèm phạt tiền 1000 đến 1 vạn tệ. Vi phạm quy định quản lý không gian không có giấy phép mà mức nặng hơn thì bị giữ 5 đến 10 ngày. Vụ án chính thức: Lý Nhậm Hào trả tiền phá giới hạn chiều cao, bay máy bay lên gần 3000 mét chụp tầng mây rồi đăng lên, bị giữ 5 ngày. Trương Nhậm Hào phá giới hạn chiều cao và vùng cấm bay giúp người khác trên 20 lượt, kiếm được hơn 1.5 vạn tệ. Anh ta bị quy kết theo tội cung cấp chương trình, công cụ xâm nhập, kiểm soát trái phép hệ thống thông tin máy tính. Bị phạt tù 6 tháng, hoãn thi hành 1 năm, kèm phạt tiền (toàn quốc, áp dụng từ tháng 1 năm 2024) +- Mức bằng chứng: A +- Nguồn: Hội đồng Bộ trưởng (国务院), Ủy ban Trung ương Quân đội (中央军委) (2023). 无人驾驶航空器飞行管理暂行条例（国令第 761 号，第十、十九、四十七、五十一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第四十六条）. ；Sở Công an tỉnh Hồ Bắc (湖北省公安厅) (2026). 公安部公布依法打击无人机「黑飞」违法犯罪八起典型案例（转载人民公安报）. +- Ghi chú: Máy bay không cân và nhẹ bay ở không gian được phép bay thì không cần phê duyệt. Không cân là loại máy bay trọng lượng cất cánh dưới 0.25 ki lô gam. Không gian được phép bay là nơi ngoài không gian bị kiểm soát; còn khu vực quanh sân bay, cơ sở quân sự, cơ quan chính phủ thì phần lớn thuộc không gian bị kiểm soát. Trang nguyên văn của Bộ Công an không mở được lúc viết phần này, nên phần này dẫn trang Sở Công an tỉnh Hồ Bắc đăng lại. + +### 12. Không chụp lén người khác, không lắp camera trong phòng thuê, nhà nghỉ, phòng khách sạn + +- Chi phí: Không tốn tiền, cũng không mất thời gian. Không chụp lén người khác, cũng không lắp camera trong phòng thuê, nhà nghỉ, phòng khách sạn. +- Nói thẳng: Chụp lén, bức tườm, phát tán đời tư của người khác thì bị giữ dưới 5 ngày hoặc phạt tiền dưới 1000 tệ, mức nặng hơn thì bị giữ 5 đến 10 ngày. Dùng thiết bị chuyên chụp lén, gây hậu quả nghiêm trọng thì phạt tù dưới 2 năm. Có hai người lắp thiết bị trong phòng của ba khách sạn để quay từ xa, lần lượt bị phạt tù 1 năm 3 tháng và 1 năm. Lắp camera trong nhà mình không sao, lắp trong phòng người khác ở thì sao. +- Lợi ích: Bức tườm, chụp lén, nghe lén, phát tán đời tư của người khác thì bị giữ dưới 5 ngày, hoặc phạt tiền dưới 1000 tệ. Mức nặng hơn thì bị giữ 5 đến 10 ngày, còn có thể phạt tiền dưới 1000 tệ. Sử dụng trái phép thiết bị chuyên nghe lén, chụp lén, gây hậu quả nghiêm trọng thì phạt tù dưới 2 năm. Thiết bị chuyên nghe lén, chụp lén là thiết bị thiết kế riêng để nghe lén, chụp lén. Vụ án chính thức: Niếm Nhậm Bình, Niếm Nhậm Kiến mua thiết bị chụp lén trên mạng, lắp vào nhiều phòng của ba khách sạn để quay từ xa, lần lượt bị phạt tù 1 năm 3 tháng và 1 năm (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第五十条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). 依法惩治非法生产、销售、使用窃听、窃照设备犯罪典型案例（案例二）. +- Ghi chú: Hình chụp lén lấy ra bán hoặc phát tán thì xử lý theo tội khác, ví dụ sản xuất, buôn bán vật phẩm khiêu dâm để phi pháp, trong cùng loại vụ án có người bị phạt 10 năm. Lắp camera trong nhà mình không sao, lắp trong phòng người khác dùng thì sao. + +### 13. Chơi mậu thẩu, bài poker thì được; không ăn phần trăm, không làm cái, không tổ chức bàn thu tiền, không chơi cờ bạc trên mạng + +- Chi phí: Không tốn tiền. Người thân chơi giải trí số tiền nhỏ thì vẫn chơi. Không ăn phần trăm từ bàn, không làm cái, không tổ chức bàn thu tiền. Chỗ khó là khi có người rủ rê thì nhịn không đi nhận. +- Nói thẳng: Người thân chơi bài số tiền nhỏ thì pháp luật không quản. ăn phần trăm từ bàn, làm cái, tổ chức bàn thu tiền thì bản chất đã khác: tham gia cờ bạc, số tiền cược lớn thì bị giữ dưới 5 ngày hoặc phạt tiền dưới 1000 tệ. Tụ tập cờ bạc, sống bằng cờ bạc thì phạt tù dưới 3 năm, mở sòng bạc thì phạt tù dưới 5 năm. Bao nhiêu là「số tiền cược lớn」do từng nơi tự quy định, không có con số thống nhất toàn quốc. +- Lợi ích: Vì tiền mà tạo điều kiện thuận lợi cho cờ bạc, hoặc tham gia cờ bạc, số tiền cược lớn thì bị giữ dưới 5 ngày, hoặc phạt tiền dưới 1000 tệ. Nghiêm trọng thì bị giữ 10 đến 15 ngày, kèm phạt tiền 1000 đến 5000 tệ. Vì tiền mà tụ tập người lại chơi cờ bạc, hoặc sống bằng cờ bạc thì phạt tù dưới 3 năm kèm phạt tiền. Mở sòng bạc thì phạt tù dưới 5 năm kèm phạt tiền, nghiêm trọng thì phạt tù 5 đến 10 năm (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第八十二条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第三百零三条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第三十六条，开设赌场刑期提高到五年以下、五年以上十年以下）. +- Ghi chú: Bao nhiêu là「số tiền cược lớn」không có con số thống nhất toàn quốc, từng nơi tự quy định, nên mục này chỉ ở mức B. Ăn phần trăm từ bàn, thu tiền hộ nền tảng cờ bạc trực tuyến, khi xét xử thường bị xác định là mở sòng bạc. Nguyên văn giải thích của tòa án liên quan phần này chưa đối chiếu. Cũng không tìm được vụ án chính thức nào. +### 14. Không ăn thịt hoang dã, không mua bán, không nuôi động vật được bảo vệ làm thú cưng + +- Chi phí: Không tốn tiền. Thú cưng chỉ nên mua giống có nguồn gốc hợp pháp, nhân giống nhân tạo. +- Nói thẳng: Ăn động vật được bảo vệ cấp quốc gia thì tịch thu vật, rồi phạt gấp 2 đến 20 lần giá trị. Mua, vận chuyển, bán động vật quý hiếm, nguy cấp thì phạt tù dưới 5 năm, nghiêm trọng thì phạt tù 5 đến 10 năm. Động vật nhân giống nhân tạo, mua bán làm thú cưng thì thường không phải tội phạm; có người bán hơn 40 con vẹt nhân giống nhân tạo mà không bị khởi tố. Động vật bắt ngoài hoang dã, không rõ nguồn gốc thì không nằm trong nhóm đó. +- Lợi ích: Ăn động vật hoang dã được bảo vệ cấp quốc gia, hoặc mua để ăn, thì tịch thu vật, rồi phạt tiền gấp 2 đến 20 lần giá trị của nó. Nhân giống động vật hoang dã được bảo vệ cấp quốc gia mà không có giấy phép thì tịch thu, kèm phạt gấp 1 đến 10 lần. Mua, vận chuyển, bán trái phép động vật hoang dã quý, nguy cấp được bảo vệ cấp quốc gia và sản phẩm của chúng thì phạt tù dưới 5 năm kèm phạt tiền. Trong mục này, nghiêm trọng thì phạt tù 5 đến 10 năm, đặc biệt nghiêm trọng thì trên 10 năm. Săn bắt, mua, vận chuyển, bán động vật hoang dã trên cạn khác để ăn, khi nghiêm trọng thì phạt tù dưới 3 năm. Vụ án chính thức: người nuôi Vương Nhậm Hào cùng 2 người khác bán hơn 40 con vẹt mào Phi Điều nhân giống nhân tạo (động vật được bảo vệ cấp hai), tổng giá chưa tới 400 tệ, năm 2021 viện kiểm sát quyết định không khởi tố. Từ năm 2022, Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao làm rõ: động vật nhân giống nhân tạo đã thành công nghệ, đã đủ quy mô, bán và vận chuyển làm thú cưng thì thường không xử lý theo tội phạm (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). 野生动物保护法（2022 年修订，第三十一、五十一、五十三条）. （Bộ Môi trường và Tự nhiên đăng lại）；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第三百四十一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第四十一条，即刑法第三百四十一条第三款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 关于办理破坏野生动物资源刑事案件适用法律若干问题的解释（法释〔2022〕12 号，第十三条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 鹦鹉小案 民生大案. +- Ghi chú: Câu「thường không xử lý là tội phạm」chỉ nói tới nhóm nhân giống nhân tạo, nuôi làm thú cưng. Động vật bắt ngoài hoang dã, không rõ nguồn gốc thì không áp dụng. Loài nào được tính là động vật bảo vệ cấp quốc gia thì theo danh mục. Mua trước thì tra danh mục đã, đỡ hơn là sau khi có chuyện mới giải thích. + +### 15. Đòi nợ thì không giữ người, không giam người, không đuổi theo tới nhà ở + +- Chi phí: Không tốn tiền. Đối phương nợ tiền không trả thì ra tòa khởi kiện. Giấy nhận nợ viết thế nào, xem mục 18 ở phần 8 (giấy nhận nợ). Chỗ khó là khi đối phương chống trả thì nhịn không ra tay, không đi tới nhà họ. +- Nói thẳng: Tiền là họ thực sự nợ bạn, bạn cũng không được giữ người, không được giam người. Điều luật ghi riêng: vì đòi nợ mà giữ người, giam người thì vẫn xử lý như giam giữ trái phép. Nhẹ thì bị giữ 10 đến 15 ngày, kèm phạt tiền 1000 đến 2000 tệ. Cấu thành tội giam giữ trái phép thì phạt tù dưới 3 năm, làm thương tích nặng thì phạt tù 3 đến 10 năm, làm chết thì trên 10 năm. Đòi nợ thì ra tòa khởi kiện. +- Lợi ích: Tự ý hạn chế tự do cá nhân của người khác, tự ý vào nhà người khác, hoặc tự ý khám người khác thì bị giữ 10 đến 15 ngày, kèm phạt tiền 1000 đến 2000 tệ. Mức nhẹ hơn thì bị giữ 5 đến 10 ngày, kèm phạt tiền dưới 1000 tệ. Giam giữ trái pháp thì phạt tù dưới 3 năm, giam giữ (nhốt ngắn hạn) hoặc kỉ luật (không giam nhưng bị hạn chế đi lại) hoặc tước quyền chính trị. Làm thương tích nặng thì phạt tù 3 đến 10 năm, làm chết thì trên 10 năm. Vì đòi nợ mà giữ người, giam người thì phạt theo đúng bộ quy tắc đó, kể cả khi khoản nợ là thật (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第四十七条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百三十八条）. +- Ghi chú: Câu「anh ta đúng là nợ tôi tiền」không có tác dụng trong điều này. Điều luật ghi riêng, vì đòi nợ mà giữ người, giam người cũng xử lý như giam giữ trái phép. Trên mạng tòa án của Trung Quốc có nhiều vụ án đều là giữ người để đòi một khoản nợ hợp pháp, bị phạt tù từ 7 tháng đến 1 năm. Lúc viết phần này trang đó không mở được nên không dẫn. + +### 16. Không cho mượn giấy tờ tùy thân, không dùng giấy tờ của người khác, cũng không dùng giấy tờ người khác để đăng ký, mở thẻ, mua vé + +- Chi phí: Không tốn tiền. Có người mượn thì từ chối là xong. Nếu phải đưa bản sao thì ghi rõ bản sao đó chỉ dùng cho việc nào lên nó. +- Nói thẳng: Cho mượn, cho thuê, chuyển nhượng giấy tờ tùy thân của mình cho người khác thì bị cảnh cáo và phạt tiền dưới 200 tệ. Mạo danh giấy tờ tùy thân của người khác, hoặc dùng giấy tờ lấy được bằng lừa đảo thì phạt tiền 200 đến 1000 tệ, hoặc bị giữ trong 10 ngày. Ở những nơi bắt buộc có danh tính thật như mua vé tàu, ở khách sạn mà dùng giấy tờ giả hoặc giấy tờ của người khác, nghiêm trọng thì bị phạt giam giữ hoặc kỉ luật. +- Lợi ích: Cho thuê, cho mượn, chuyển nhượng thẻ căn cưởc dân của mình cho người khác thì bị cảnh cáo và phạt tiền dưới 200 tệ. Mạo danh giấy tờ tùy thân của người khác, hoặc dùng giấy tờ tùy thân lấy được bằng lừa đảo thì phạt tiền 200 đến 1000 tệ, hoặc bị giữ dưới 10 ngày. Mua, bán, sử dụng giấy tờ tùy thân làm giả hoặc sửa đổi cũng thuộc mức này. Có những nơi theo quy định nhà nước bắt buộc phải xuất trình chứng minh danh tính. Ở những nơi đó mà dùng giấy tờ làm giả, sửa đổi hoặc lấy của người khác thì nghiêm trọng là phải kết án. Giấy tờ gồm thẻ căn cưởc, hộ chiếu, thẻ bảo hiểm xã hội, giấy phép lái xe. Hình phạt là giam giữ (nhốt ngắn hạn) hoặc kỉ luật (không giam nhưng bị hạn chế đi lại), còn phạt kèm hoặc chỉ phạt tiền, tức là vừa kết án vừa phạt tiền, cũng có thể chỉ phạt tiền (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2011). 居民身份证法（2011 年修正，第十六、十七条）. （Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại）；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十条之一）. +- Ghi chú: Cho giấy tờ tùy thân ra, người ta có thể dùng nó để đăng ký công ty, mở thẻ ngân hàng. Hậu quả của hai việc đó xem mục 28 và 8 ở phần 8 (đừng làm người đại diện danh nghĩa, đừng cho mượn thẻ ngân hàng). Dùng giấy tờ của người khác để mua vé tàu, ở khách sạn, thuộc loại「hoạt động phải cung cấp chứng minh danh tính」mà điều luật nói đến. Mục này không tìm được vụ án chính thức nào. + +### 17. Khi bị công an chặn hoặc đưa đi thì hợp tác, không đẩy, không đánh, không giật máy quay của người thi hành; đã uống rượu thì càng phải nhịn + +- Chi phí: Không tốn tiền. Tại hiện trường thì không ra tay. Có ý kiến thì để sau hãy nói: khiếu nại, xin xét lại hành chính, ra tòa kiện. Chỗ khó là nhịn được cái bực trong mấy phút bị chặn, nhất là khi đã uống rượu. +- Nói thẳng: Cản trở việc thi hành công vụ tại hiện trường thì bị cảnh cáo hoặc phạt tiền dưới 500 tệ, nghiêm trọng thì bị giữ 5 đến 10 ngày. Ra tay với công an là mức khác: cắn, tát, đá đấm, ôm đi đánh, ném đồ vật, gây thương tích nhẹ trở lên thì tính là tấn công bạo lực công an, phạt tù dưới 3 năm. Dùng súng, dao quản hoặc lái xe đâm thì phạt tù 3 đến 7 năm. Vật tay, vùng ra như thế sức chống đối cực thường không tính. +- Lợi ích: Cản trở cán bộ nhà nước làm công vụ đúng quy định thì bị cảnh cáo hoặc phạt tiền dưới 500 tệ. Nghiêm trọng thì bị giữ 5 đến 10 ngày, còn có thể phạt tiền dưới 1000 tệ. Cản trở công an nhân dân thì phạt nặng hơn. Tấn công bạo lực công an nhân dân đang thi hành công vụ đúng quy định thì phạt tù dưới 3 năm. Dùng súng, dao quản, lái xe đâm, những cách nguy hiểm nghiêm trọng cho an toàn thân thể của công an thì phạt tù 3 đến 7 năm. Cắn, tát, đá đấm, ôm đi đánh, ném vật gây thương tích nhẹ trở lên mới tính là「tấn công bạo lực」. Vật tay, vùng ra, đá chân, thuộc sức chống đối cực thường thì không tính (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第六十一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第三十一条，即刑法第二百七十七条第五款）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百七十七条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 关于办理袭警刑事案件适用法律若干问题的解释（高检发释字〔2025〕1 号，第一、八、十二条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 袭警罪入刑一月 检察机关批捕 405 人. +- Ghi chú: Năm 2021 Viện kiểm sát nhân dân tối cao thông báo, trong tháng đầu tiên sau khi tội tấn công công an được định tội riêng đã phê chuẩn bắt 405 người. Trong 42 vụ ở tỉnh Giang Tô, 47.6% do gây rối sau khi uống rượu. Trong 16 vụ ở tỉnh Hồ Bắc có 12 vụ chỉ ở mức đấm đá, cắn xé. Ra tay với công an hỗ trỗ thì xử lý theo tội gây cản trở công vụ. Mục này không tìm được vụ án chính thức nào ghi rõ hoàn cảnh cá nhân. + +### 18. Người kia chưa đủ 14 tuổi thì không được quan hệ tình dục; câu「cô ấy đồng ý」không phải là lý do + +- Chi phí: Không tốn tiền. Không chắc tuổi đối phương thì không quan hệ tình dục. +- Nói thẳng: Người kia chưa đủ 14 tuổi, dù cô ấy có đồng ý hay không, vẫn bị xử lý nặng hơn theo tội cưỡng ép, mức tù cơ bản là 3 đến 10 năm. Người kia chưa đủ 12 tuổi thì bị đồng nhất là bạn「lẽ ra phải biết」. Từ 12 đến 14 tuổi thì có thể nhìn ra khả năng là chưa thành niên qua thể chất phát triển, thái độ, cách ăn mặc; bạn vẫn làm thì bị xác định là「biết」. Nói thật là không biết thì tòa án xử rất chặt. +- Lợi ích: Quan hệ tình dục với người chưa thành niên dưới 14 tuần tuổi thì xử lý theo tội cưỡng ép và phạt nặng hơn, mức tù cơ bản 3 đến 10 năm. Cách hiểu của Tòa án nhân dân tối cao là thế này: dùng vũ lực, đe dọa, các biện pháp cưỡng chế thì dù có biết người kia chưa thành niên hay không, vẫn định tội cưỡng ép. Không dùng biện pháp cưỡng chế thì với người dưới 12 tuần tuổi đều đồng nhất là「lẽ ra phải biết」. Từ 12 đến 14 tuần tuổi, nhìn qua thể chất phát triển, lời nói thái độ, cách ăn mặc mà thấy có thể là chưa thành niên, bạn vẫn làm thì xác định là「biết」. Ngoại lệ「thật sự không biết」được nắm rất chặt (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百三十六条）. ；Tòa án nhân dân tối cao (最高人民法院) (2015). 关于「与十四周岁以下幼女发生性关系是否要明知才构成犯罪」的答复. +- Ghi chú: Nguyên văn ý kiến của Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao cùng Bộ Công an, Bộ Giáo dục về xử lý vụ án xâm hại tình dục người chưa thành niên, cùng trang trả lời năm 2003 của Tòa án nhân dân tối cao, đều không mở được lúc viết phần này. Vì vậy tiêu chuẩn xác định「biết」dẫn từ câu trả lời trên trang của Tòa án nhân dân tối cao. Cách nói trong điều luật là「chưa thành niên nữ」. Mục này không phân biệt người ra tay là nam hay nữ. +### 19. Trước khi ra tay, tính rõ cái giá: bị giữ, phải bồi, kết án + +- Chi phí: Không tốn tiền, cũng không mất thời gian. Khó chỉ ở mấy giây trước lúc ra tay. +- Nói thẳng: Đánh một lần thì bị giữ 5 đến 10 ngày, cộng phạt tiền 500 đến 1000 tệ. Đánh thành thương tích nhẹ là đã thành vụ án hình sự, phạt tù dưới 3 năm; đánh thành thương tích nặng thì phạt tù 3 đến 10 năm. Tiền thì tính riêng: có người làm gãy ba xương sườn đối phương, bồi 11.3 vạn tệ để thỏa thuận, nhờ đó không bị khởi tố. Người khác làm gãy xương quay đối phương, không bồi, bị phạt mười tháng. +- Lợi ích: Đánh người hoặc cố ý gây thương tích thì bị giữ 5 đến 10 ngày, kèm phạt tiền 500 đến 1000 tệ. Đánh theo nhóm, đánh người tàn tật, phụ nữ mang thai, người già, trẻ nhỏ, đánh nhiều lần, hoặc một lần đánh nhiều người thì bị giữ 10 đến 15 ngày, kèm phạt tiền 1000 đến 2000 tệ. Đánh thành thương tích nhẹ là tội cố ý gây thương tích, phạt tù dưới 3 năm, giam giữ (nhốt ngắn hạn) hoặc kỉ luật (không giam nhưng bị hạn chế đi lại). Đánh thành thương tích nặng thì phạt tù 3 đến 10 năm. Về mặt dân sự còn phải bồi chi phí y tế, lương mất công (tiền lương bị bỏ lỡ), chi phí chăm sóc, chi phí đi lại, tiền ăn ăn ở trong bệnh viện, chi phí dinh dưỡng. Đánh thành tàn tật thì còn phải bồi tiền bồi thường tàn tật. Cách tính tiền bồi thường tàn tật là thu nhập bình quân mà dân số đô thị của nơi có tòa án xét xử nhân với số năm của năm trước nhân với 20 năm. Người bị thương trên 60 tuổi thì mỗi thêm một tuổi trừ đi một năm, trên 75 tuổi tính là 5 năm. Năm 2022 các viện kiểm sát toàn quốc nhận hơn 7 vạn vụ án thương tích nhẹ, đa phần do bất hòa gia đình, hàng xóm hoặc sự cố bất ngờ. Vụ án điển hình của Viện kiểm sát nhân dân tối cao: Lư Nhậm làm gãy ba xương sườn đối phương (thương tích nhẹ mức hai), bồi 11.3 vạn tệ để đạt thoả thuận, nhờ đó được không khởi tố. Thạch Nhậm làm gãy xương quay đối phương (thương tích nhẹ mức hai), không bồi, bị phạt tù mười tháng (toàn quốc, Luật Xử phạt quản lý trật tự công cộng áp dụng từ ngày 1 tháng 1 năm 2026) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第九、五十一条）. ；Quốc hội Nhân dân (全国人大) (1997). 刑法（第二百三十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2022 修正). 关于审理人身损害赔偿案件适用法律若干问题的解释（法释〔2022〕14 号，第六至十二、十五条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部) (2023). 关于依法妥善办理轻伤害案件的指导意见. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 检察机关依法妥善办理轻伤害案件典型案例. +- Ghi chú: Cách tránh xung đột, vì sao ra tay trước là thiệt, ranh giới tự vệ ở đâu, xem mục 10 ở phần 8 (xung đột và tự vệ). Cái giá của việc ra tay với người khác chỉ để trút giận, xem mục 11 đến 14 ở phần 8. Đánh nhau vì bất hòa dân sự, mức nhẹ hơn thì công an có thể hoà giải. Đánh thành thương tích nhẹ mà thương lượng xong với đối phương và đã trả thật thì có thể nhẹ hơn, thậm chí không khởi tố, nhưng đồng tiền bồi thường không được thiếu một đồng nào. Bị thương tới mức nào thì gọi là thương tích nhẹ do tiêu chuẩn định giá quy định, phần này chưa đối chiếu. Lương mất công, chi phí chăm sóc tính theo thu nhập thực tế và số ngày bị chậm, không có con số thống nhất. + +### 20. Sinh con ra không nuôi nổi thì lối thoát hợp pháp chỉ có một: đăng ký nhận nuôi ở cơ quan dân vụ; nhận tiền rồi đưa con cho người khác có thể bị kết án về tội bán trẻ em, bỏ trẻ không chăm thì là tội bỏ bê + +- Chi phí: Không tốn tiền. Đăng ký nhận nuôi ở cơ quan dân vụ cấp tỉnh trở lên thì phải nộp hồ sơ, còn phải chờ xét duyệt và đánh giá nuôi dưỡng, trước sau cũng tốn thời gian. +- Nói thẳng: Không nuôi nổi thì có lối thoát, nhưng lối thoát hợp pháp chỉ có một: ra cơ quan dân vụ đăng ký nhận nuôi, quan hệ nuôi dưỡng chỉ lập được từ ngày đăng ký. Tư cho con người khác ngoài mối quan hệ, mà nhất là có nhận tiền, thì có thể từ「cho con nuôi」thành tội bán trẻ em, mức tù khởi điểm là trên 5 năm. Bỏ con rồi không chăm, cố tình không nuôi dưỡng, khi nghiêm trọng thì là tội bỏ bê, phạt tù dưới 5 năm. +- Lợi ích: Điều hai trăm bốn mươi của Bộ luật Hình sự:「拐卖妇女、儿童的，处五年以上十年以下有期徒刑，并处罚金」. Có tình tiết tăng nặng thì phạt tù từ mười năm trở lên hoặc tù chung thân, kèm phạt tiền hoặc tịch thu tài sản.「情节特别严重的，处死刑，并处没收财产」. Ý kiến về xử lý theo pháp luật tội bán phụ nữ, trẻ em nói ở mục 17 cách phân biệt bán con ruột với cho con nuôi trong dân gian. Tiêu chuẩn để xét là「关键在于行为人是否具有非法获利的目的」, tức là xem có lợi ích kiếm tiền từ việc này hay không. Có ba trường hợp có thể xác định là bán con ruột, quy kết theo tội bán phụ nữ, trẻ em. Thứ nhất là「将生育作为非法获利手段，生育后即出卖子女的」. Thứ hai là「明知对方不具有抚养目的，或者根本不考虑对方是否具有抚养目的，为收取钱财将子女『送』给他人的」. Thứ ba là「为收取明显不属于『营养费』『感谢费』的巨额钱财将子女『送』给他人的」. Ngược lại, không vì tiền, thực sự vì khó khăn sinh hoạt mà tự ý cho con nuôi, chỉ nhận ít tiền nuôi dưỡng, tiền cảm ơn thì thuộc cho con nuôi trong dân gian, không quy kết theo tội bán trẻ em. Xem tiếp điều hai trăm sáu mốt của Bộ luật Hình sự.「对于年老、年幼、患病或者其他没有独立生活能力的人，负有扶养义务而拒绝扶养，情节恶劣的，处五年以下有期徒刑、拘役或者管制」. Điều một nghìn không trăm chín mươi tư của Bộ luật Dân sự quy định chỉ có ba loại người được làm người cho con nuôi: người giám hộ trẻ mồ côi, cơ quan phúc lợi xã hội trẻ em, và cha mẹ ruột có khó khăn đặc biệt không có khả năng nuôi dưỡng. Điều một nghìn một trăm linh năm:「收养应当向县级以上人民政府民政部门登记。收养关系自登记之日起成立」 +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (1997). 刑法（第二百四十、二百六十一条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部), Bộ Tư pháp (司法部) (2010). 关于依法惩治拐卖妇女儿童犯罪的意见（法发〔2010〕7 号，第 17 条）. Trang chính thức của Tòa án nhân dân tối cao ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第一千零九十四、一千一百零五条）. Chính quyền thành phố Diêm City đăng lại toàn văn phần nhận nuôi +- Ghi chú: Chuyện có nên sinh con hay không không bàn ở đây. Sau khi con sinh ra, đi theo thứ tự này là tiện nhất. Bước một, sinh con ở bệnh viện, khám thai và nằm viện đẻ xem phần 27. Giấy chứng nhận sinh con là tiền đề của mọi thủ tục về sau. Bước hai, muốn cho con nuôi thì ra thẳng cơ quan dân vụ cấp tỉnh trở lên, không qua trung gian. Tiền đã đi qua tay là bản chất có thể đổi. Trong vòng một năm sau sinh, nếu xuất hiện「ý nghĩ tự làm hại mình hoặc hại đứa trẻ」thì xử lý như cấp cứu. Danh mục những lúc phải tới bệnh viện ngay trong thai kỳ và sau sinh xem mục 7 ở phần 27. Tái khám sau 42 ngày cũng chính là sàng lọc trầm cảm sau sinh, xem mục 16 ở phần 27. Có ý nghĩ tự sát thì gọi 12356, xem mục 25 ở phần 1. Mục này không kèm vụ án: các vụ án điển hình chính thức về bán con ruột, lượt này chưa đối chiếu từng vụ, nên chỉ dẫn nguyên văn điều luật và ý kiến. + +### 21. Không bịa tai nạn, không phóng đại thiệt hại để lừa bồi thường: đó là tội lừa đảo bảo hiểm, và người làm chứng, sửa xe, giám định giúp bạn cùng bị tính + +- Chi phí: Không tốn tiền, cũng không mất thời gian, không làm là xong. +- Nói thẳng: Bịa tai nạn, phóng đại thiệt hại để lừa bồi thường là tội lừa đảo bảo hiểm, số tiền lớn thì phạt tù dưới 5 năm, còn phạt tiền 1 vạn đến 10 vạn tệ. Người làm chứng, giám định, định giá tài sản giúp bạn mà cố ý đưa ra chứng từ giả thì bị quy kết là đồng phạm cùng bạn. Trong một vụ án ở Tuyên Thành, tỉnh An Huy, 8 người bị phạt tù từ 2 đến 4 năm, mỗi người phạt tiền 1 vạn tệ. +- Lợi ích: Điều một trăm chín mươi tám của Bộ luật Hình sự liệt kê năm trường hợp. Một: bên mua bảo hiểm cố ý dựng nên đối tượng bảo hiểm, bịa ra một đối tượng tham gia bảo hiểm không hề tồn tại. Hai: bên mua bảo hiểm, bên được bảo hiểm hoặc người hưởng bồi thường bịa nguyên nhân của tai nạn bảo hiểm đã xảy ra, hoặc phóng đại mức độ thiệt hại. Ba: bịa ra một tai nạn bảo hiểm chưa từng xảy ra. Bốn: bên mua bảo hiểm, bên được bảo hiểm cố ý gây ra tai nạn bảo hiểm tổn thất tài sản. Năm: bên mua bảo hiểm, người hưởng bồi thường cố ý gây ra cái chết, thương tích hoặc bệnh tật cho bên được bảo hiểm. Số tiền lớn thì phạt tù dưới 5 năm hoặc giam giữ, kèm phạt tiền từ hơn 1 vạn tệ đến dưới 10 vạn tệ. Số tiền rất lớn hoặc có tình tiết nghiêm trọng khác thì phạt tù 5 đến 10 năm, kèm phạt tiền từ 2 vạn tệ đến 20 vạn tệ. Số tiền đặc biệt lớn hoặc có tình tiết đặc biệt nghiêm trọng khác thì phạt tù trên 10 năm, kèm phạt tiền từ 2 vạn tệ đến 20 vạn tệ hoặc tịch thu tài sản. Khoản cuối của điều này quy cho người giúp việc: người giám định, người làm chứng, người định giá tài sản trong vụ bảo hiểm. Họ cố ý lập chứng từ giả, tạo điều kiện cho người khác lừa bảo hiểm thì bị xử lý như đồng phạm của tội lừa đảo bảo hiểm, quy kết cùng người lừa. Điều hai mươi bảy của Luật Bảo hiểm: không xảy ra tai nạn bảo hiểm mà báo là có thì công ty bảo hiểm có quyền huỷ hợp đồng, và không hoàn tiền phí. Dùng chứng từ làm giả, sửa đổi, hoặc bịa nguyên nhân tai nạn, phóng đại mức thiệt hại thì công ty bảo hiểm không bồi phần khai báo thừa, phần đã bồi thì phải hoàn lại hoặc bồi thường. Vụ án: trong vụ Trương Nhậm Tùng ở Tuyên Thành, tỉnh An Huy, Vương Tín Hàm cùng 7 người nữa, tổng cộng 8 người, bị phạt tù từ 4 năm xuống 2 năm về tội lừa đảo bảo hiểm, đều bị phạt tiền 1 vạn tệ. Vụ kiện dân sự giả bịa ra trong vụ án đó có thiệt hại được công ty bảo hiểm xác định là 2 vạn tệ, các bên dựa vào hợp đồng chuyển nhượng xe làm giả để có tư cách khởi kiện, tòa án buộc bồi hơn 7 vạn tệ. Bản án dân sự đã có hiệu lực đó bị xét lại và hủy ngày 26 tháng 9 năm 2019. Biện pháp chống gian lận bảo hiểm yêu cầu: cả hai đầu bán hợp đồng và bồi thường (đầu ký kết bảo hiểm và đầu giải quyết bồi thường) đều phải kiểm tra thông tin rủi ro. Còn phải thăm dò xây dựng nền tảng chia sẻ thông tin chống gian lận đa ngành, trung tâm tình báo chống gian lận. Các nền tảng này tập trung rà soát dữ liệu nghi ngờ gian lận, chuyển dấu vết cho hiệp hội ngành, tổ chức chống gian lận và cơ quan bảo hiểm để kiểm chứng. Nghi ngờ phạm tội thì báo công an kịp thời. Cục Quản lý Tài chính ngân hàng và các cơ quan chi nhánh của nó phải cùng công an, viện kiểm sát nhân dân, tòa án nhân dân xây dựng cơ chế kết nối giữa quản lý hành chính và tư pháp hình sự (toàn quốc, ban hành ngày 31 tháng 7 năm 2024) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会). 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第一百九十八条. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015 修正). 中华人民共和国保险法第二十七条. ；Cục Quản lý Tài chính ngân hàng quốc gia (国家金融监督管理总局) (2024). 关于印发《反保险欺诈工作办法》的通知. ；Viện kiểm sát nhân dân tỉnh An Huy đăng lại (安徽省人民检察院转载) 《法制日报》(2020-01-13). 杀妻骗保牵出案中案. +- Ghi chú: Mấy việc dưới đây tưởng vô hại đều nằm trong năm trường hợp đó. Xảy ra tai nạn nhỏ thì tìm tiệm sửa xe「báo hơn một chút」, tính luôn vết thương vết hư cũ vào lần này, đâm nhẹ vào nhau rồi mới báo, không có chuyện gì mà bịa ra một chuyện. Bảo hiểm xe mua thế nào, xem mục 26 ở phần 5 (mua đủ bảo hiểm trách nhiệm bên thứ ba). Sau khi mua bảo hiểm cho người nhà mà cố ý gây ra cái chết cho họ, điều luật ghi rõ phải gộp nhiều tội, xem mục 38 ở phần 8 (mua bảo hiểm cho người nhà). Bảo hiểm y tế là bộ quy tắc khác, quẹt thẻ bảo hiểm y tế trống hoặc rút tiền từ tài khoản cá nhân của bảo hiểm y tế thì xử lý như lừa đảo, xem ghi chú của mục 13 ở phần 5 +### 22. Không bán cơ quan của mình, cũng đừng đi tìm người hiến cho người khác: thận bán được hơn 2 vạn tệ, cùng một quả bán cho bệnh nhân tới 20 vạn tệ, tiền bị tịch thu lại còn phạt theo giao dịch 10 đến 20 lần + +- Chi phí: Không tốn tiền, cũng không tốn thời gian, chỉ cần đừng làm là được. +- Nói thẳng: Ở Bắc Kinh có một vụ, người bán thận chỉ nhận được hơn 2 vạn tệ, còn đúng quả thận đó bán cho bệnh nhân giá 18 vạn đến 22 vạn tệ. Tiền bán được sẽ bị tịch thu, sau đó còn phạt thêm 10 đến 20 lần số giao dịch. Cái giá của cơ thể sau khi mất một quả thận, xem mục 35 ở phần 1 (quả thận còn lại làm việc thay hai quả). +- Lợi ích: 《人体器官捐献和移植条例》第五条：「任何组织或者个人不得以任何形式买卖人体器官，不得从事与买卖人体器官有关的活动」— mọi tổ chức, cá nhân đều không được mua bán cơ quan người dưới bất kỳ hình thức nào, cũng không được làm việc liên quan đến việc mua bán đó. 第十条：「任何组织或者个人不得获取未满 18 周岁公民的活体器官用于移植」— không được lấy cơ quan sống của công dân chưa đủ 18 tuổi để ghép. 第十一条：「活体器官的接受人限于活体器官捐献人的配偶、直系血亲或者三代以内旁系血亲」— người nhận chỉ được là vợ, chồng, họ hàng trực hệ hoặc họ hàng chiềng trong vòng ba đời của người hiến. Điều ba mươi bảy của条例 nêu cách phạt việc mua bán cơ quan người: mua bán cơ quan người hoặc làm việc liên quan, thì bị cơ quan y tế chính quyền nhân dân cấp huyện trở lên điều tra xử lý. Trước tiên tịch thu thu nhập bất hợp pháp, cất hết tiền có được từ giao dịch. Sau đó phạt theo số giao dịch, từ 10 lần trở lên đến 20 lần trở xuống. Cán bộ y tế tham gia thì bị thu hồi chứng chỉ hành nghề và cấm làm y tế vĩnh viễn.Hành vi đã đủ yếu tố hình sự thì truy cứu trách nhiệm hình sự theo luật (toàn quốc, có hiệu lực từ ngày 1 tháng 5 năm 2024,《人体器官移植条例》cũ đồng thời hết hiệu lực). 刑法第二百三十四条之一：「组织他人出卖人体器官的，处五年以下有期徒刑，并处罚金；情节严重的，处五年以上有期徒刑，并处罚金或者没收财产」— người tổ chức người khác bán cơ quan người thì phạt tù dưới năm năm kèm phạt tiền; tình tiết nặng thì phạt tù từ năm năm trở lên, kèm phạt tiền hoặc tịch thu tài sản. Còn ba trường hợp, xử theo 刑法第二百三十四条、第二百三十二条: một là tự ý lấy cơ quan của người đó khi chưa được người đó đồng ý; hai là lấy cơ quan của người chưa đủ mười tám tuổi; ba là ép hoặc lừa người khác hiến cơ quan. Các trường hợp này bị xử về tội cố ý gây thương tích hoặc tội giết người. 案例一：vụ Trịnh Vĩ và 15 người, tòa nhân dân trung cấp cấp 1 thành phố Bắc Kinh tuyên phúc thẩm ngày 20 tháng 8 năm 2014, liên quan 51 quả thận, tổng số tiền hơn 1034 vạn tệ. Người bán thận (người hiến)「换来的仅仅是 2 万到 2.5 万元不等的报酬」— chỉ nhận được khoản công từ 2 đến 2.5 vạn tệ. Bên bệnh nhân chờ thận (bên nhận) đưa giá「以前是一个肾 18 万或 19 万，后来是 21 万或 22 万」— trước đây một quả 18 vạn hoặc 19 vạn, sau lên 21 vạn hoặc 22 vạn. Nơi lấy thận là một tòa nhà bốn tầng thuê mỗi tháng 7500 tệ, y tá trưởng của băng đảng khai nơi đó「脏乱不堪，连抢救设备都没有」— bẩn thỉu không chịu nổi, ngay cả thiết cấp cứu cũng không có. Người tổ chức Trịnh Vĩ bị phạt tù 12 năm về tội tổ chức mua bán cơ quan người, 14 người còn lại bị phạt từ 3 năm rưỡi đến 9 năm. 案例二：vụ「thiếu niên bán thận mua iPhone」do Viện kiểm sát nhân dân tối cao công bố. Vương mỗ 17 tuổi, tối ngày 28 tháng 4 năm 2011 bị cắt thận phải ở Trấn Châu. Nơi phẫu thuật là một khoa do người khác thuê phụ trách trong một bệnh viện không có tư cách ghép. Ngày 2 tháng 5 ra viện, anh ta nhận 2.2 vạn tệ, bên nhận trả cho môi giới 15 vạn tệ và 10.000 USD, chủ mưu thu lời hơn 5.6 vạn tệ. Sau phẫu thuật Vương mỗ bị suy chức năng thận,「伤情构成重伤、三级伤残」— thương tích nặng, thương tật mức 3 +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2023). Biện pháp hiến tặng và ghép bộ phận cơ thể người（国务院令第 767 号，第五、十、十一、三十七条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会). Bộ luật Hình sự nước Cộng hòa Nhân dân Trung Hoa（经刑法修正案十一修正的整合文本）第二百三十四条之一. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2012-04-09). Vụ「thiếu niên bán thận mua iPhone」đưa ra tòa. ；Trang web Tòa nhân dân trung cấp thành phố Lang Phong, Hà Bắc chuyển lại (2014-09-02). Phúc thẩm vụ mua bán thận lớn nhất cả nước, người tổ chức lấy 51 quả thận, lợi nhuận lên tới hàng chục triệu tệ. +- Ghi chú: Người được lợi là chính bạn. Giao dịch bề ngoài là「một lần nhận hai vạn」, thực ra là đổi một cơ quan lấy một chiếc điện thoại. Kết luận giám định trong vụ thiếu niên bán thận là thương tích nặng, thương tật mức 3, nghỉ vài tháng không khỏi. Trong vụ đó, viện kiểm sát xác định 5 bị cáo vi phạm 刑法第二百三十四条第二款, truy tố về tội cố ý gây thương tích. Một trong các lý do là nhân viên y tế không kiểm tra「người hiến」có phải vị thành niên hay không trước khi cắt thận phải. Hôm nay chuyện đó tái diễn, người đi lôi kéo bị xử về tội tổ chức mua bán cơ quan người, còn việc lấy cơ quan của người chưa đủ mười tám tuổi vẫn bị xử về tội cố ý gây thương tích theo 第二百三十四条. Phẫu thuật của các môi giới đen làm trong nhà thuê hoặc nơi mượn, ngoài việc khớp bệnh thì không có xét nghiệm nào, cũng không có tái khám sau mồ, xảy ra biến chứng thì không tìm ai chịu trách nhiệm. Thực sự muốn hiến cho người thân, đi theo con đường của 第十一条: làm ở cơ sở y tế có tư cách ghép, qua hội đồng đạo đức duyệt, không thu tiền suốt quá trình. Hiến cơ quan sau khi chết là无偿, không nằm trong mục này. Cái giá cơ thể phải trả sau khi mất một quả thận, xem mục 35 ở phần 1 (quả thận còn lại làm việc thay hai quả). Đến mức nghèo không còn tiền mà phải tính chuyện bán cơ quan, trước tiên xem phần 7 (các kênh cứu trợ: bảo hiểm thất nghiệp, trợ cấp tạm thời, trợ cấp người có thu nhập thấp). Bẫy vay nóng mạng và cho vay「包装材料」, xem mục 6 của phần này. + +### 23. Đừng đi mại dâm: mức mặc định là giam 10 đến 15 ngày, không phải phạt chút tiền là xong + +- Chi phí: Không tốn tiền, cũng không tốn thời gian, chỉ cần đừng làm là được. +- Nói thẳng: Bị bắt không phải là phạt chút tiền là xong. Pháp luật đặt mức mặc định là giam 10 đến 15 ngày, còn có thể phạt thêm tối đa 5000 tệ. Chỉ tình tiết nhẹ mới là giam dưới 5 ngày hoặc phạt tối đa 1000 tệ. Ngoài ra, biết mình đã mắc giang mai, lậu mà vẫn đi mua dâm thì là tội phạm, phạt dưới 5 năm. +- Lợi ích: 《治安管理处罚法》第七十八条：卖淫、嫖娼的，处十日以上十五日以下拘留，可以并处五千元以下罚款。情节较轻的，处五日以下拘留或者一千元以下罚款。在公共场所拉客招嫖的，处五日以下拘留或者一千元以下罚款。Nghĩa là: bán dâm hay mua dâm, mức phạt mặc định là giam từ mười ngày đến mười lăm ngày, còn có thể phạt tối đa năm nghìn tệ. Tình tiết nhẹ hơn thì giam dưới năm ngày hoặc phạt tối đa một nghìn tệ. Lôi kéo khách mua dâm ở nơi công cộng cũng phạt mức này. 第七十九条：引诱、容留、介绍他人卖淫的，处十日以上十五日以下拘留，可以并处五千元以下罚款。情节较轻的，处五日以下拘留或者一千元以上二千元以下罚款。Nghĩa là: dụ dỗ, chứa chấp, mai mối cho người khác bán dâm thì mức thấp nhất cũng giam mười ngày đến mười lăm ngày kèm phạt tối đa năm nghìn tệ; tình tiết nhẹ thì giam dưới năm ngày hoặc phạt từ một nghìn đến hai nghìn tệ. 第八十七条：旅馆业、饮食服务业、文化娱乐业、出租汽车业等单位的人员，在公安机关查处卖淫、嫖娼活动时为违法犯罪行为人通风报信的，或者以其他方式为上述活动提供条件的，处十日以上十五日以下拘留。Nghĩa là: nhân viên khách sạn, quán ăn, quán hát, taxi và nơi tương tự mà báo tin cho người vi phạm khi công an đi kiểm tra, hoặc giúp các hoạt động đó bằng cách khác, thì bị giam mười ngày đến mười lăm ngày. 《刑法》第三百六十条是传播性病罪：明知自己患有梅毒、淋病等严重性病卖淫、嫖娼的，处五年以下有期徒刑、拘役或者管制，并处罚金（toàn quốc, có hiệu lực từ ngày 1 tháng 1 năm 2026 theo《治安管理处罚法》） +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2025). Luật Trật tự xử phạt hành chính（2025 年修订，第七十八、七十九、八十七条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc. Bộ luật Hình sự nước Cộng hòa Nhân dân Trung Hoa（经刑法修正案十一修正的整合文本）第三百六十条. （trang web Cục Giao thông Công an thành phố Bắc Kinh chuyển lại）；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (2015). Sửa đổi Bộ luật Hình sự（九）（第四十三条，删去刑法第三百六十条第二款）. +- Ghi chú: Ở đây giam là giam hành chính, người bị giam phải vào trại giam, không phải phạt xong là đi. Tìm người giới thiệu, cung cấp chỗ cho người khác thì bị phạt nặng như chính mình đi mua dâm, khoản 第七十九条 lo chính việc này. Người làm ở khách sạn, KTV báo tin cho khách thì cũng bị giam 10 đến 15 ngày theo 第八十七条. 刑法第三百六十条 trước đây còn có 第二款「嫖宿幼女罪」, 刑法修正案（九）2015 年 xóa rồi, giờ chuyện này trực tiếp bị xử về tội hiếp dâm với khung nặng hơn, xem mục 18 của phần này (đối phương chưa đủ 14 tuổi thì không được quan hệ). Biển hiệu「上门保健」「按摩推拿」không làm thay đổi bản chất, xử theo việc thực sự xảy ra. Rủi ro mắc bệnh lây truyền qua đường tình dục và HIV, xem mục 30 ở phần 1 (dùng bao cao su suốt quá trình quan hệ tình dục) và mục 38 ở phần 13 (trong 72 giờ đi lấy thuốc ngăn chặn). Những việc phải chú ý ở bản thân nơi vui chơi, xem phần 22. diff --git "a/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" "b/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" new file mode 100644 index 00000000..15a04476 --- /dev/null +++ "b/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" @@ -0,0 +1,388 @@ +[← 回总目录](../README.md) + +# 13. 紧急情况：先做什么 + +本节讲出了事先做什么、再做什么、什么不要做。急救和野外的条目算的是特定死因和活下来的比例。威胁和骗局的条目算的是钱和人身自由。两类数字不互相换算。条目的顺序按「这事多常见 × 做对做错差多少条命」从高到低排。这些条目首先是给你自己和配偶、直系亲属用的。心脏骤停有 79.2% 发生在家里，你最可能救的就是家里人。对朋友同事次之。对陌生人也不是一点好处没有，但这一档最低，还可能被讹、被卷进案子。所以这类条目把保护自己的动作和免责的条款写在前面。 + +本节只收几分钟内要做的动作。认出骗局、被人拿把柄要钱，处置要按天算，不在这一节。反诈的硬规则和七种高发骗局见第 8 节第 3 条（反诈硬规则）。被人拿隐私照片、裸聊录像要钱见第 8 节第 32 条（对方拿把柄来要钱）。钱已经转出去了，立刻打 110 或 96110 要求止付，见第 8 节第 2 条（止付）。 + +[← Về mục lục](../README.md) + +# 13. Tình huống khẩn cấp: làm gì trước + +Phần này nói trước khi xảy ra chuyện thì làm gì, sau đó làm gì, và cái gì không được làm. Các mục cấp cứu và ngoài hoang tính tỷ lệ tử vong theo từng nguyên nhân cụ thể và tỷ lệ sống sót. Các mục về mối đe dọa và lừa đảo tính tiền và tự do cá nhân. Hai bộ số này không quy đổi qua lại. Thứ tự các mục xếp từ cao xuống thấp theo「việc này xảy ra thường cỡ nào × làm đúng làm sai khác nhau mấy mạng người」. Trước hết các mục này dành cho bạn, người phối ngẫu và người thân trực hệ. 79.2% các ca ngừng tim xảy ra ở trong nhà, nên người bạn có khả năng cứu cao nhất là người trong nhà bạn. Bạn bè và đồng nghiệp thứ sau. Người xa lạ cũng không hoàn toàn vô ích, nhưng nhóm đó thấp nhất, lại còn có thể bị tống tiền hoặc bị kéo vào một vụ án. Nên các mục loại này đặt trước những bước tự bảo vệ và các điều khoản miễn trách. + +Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận ra lừa đảo, hoặc bị người ta cầm những thứ bắt để tống tiền, thì xử lý tính theo ngày, không thuộc phần này. Quy tắc cứng chống lừa đảo và bảy kiểu lừa đảo hay gặp, xem mục 3 ở phần 8 (quy tắc cứng chống lừa đảo). Bị tống tiền bằng ảnh riêng tư, clip nhạy cảm, xem mục 32 ở phần 8 (đối phương cầm nhúng để tống tiền). Tiền đã chuyển đi rồi thì gọi ngay 110 hoặc 96110 yêu cầu phong tỏa thanh toán, xem mục 2 ở phần 8 (phong tỏa thanh toán). + +### 1. Người bị ngã và không thở: ấn mạnh vào giữa ngực, bảo người bên cạnh gọi 120 và đi lấy AED + +- Chi phí: Không tốn tiền. Khóa hồi sinh tim phổi của Hội Chữ thập đỏ hoặc của bệnh viện kéo dài nửa ngày đến một ngày, học một lần là đủ dùng nhiều năm. Không học vẫn làm được: chồng hai tay đặt vào giữa đường nối hai núm vú, ấn mạnh xuống khoảng 2 lần mỗi giây, đừng dừng. +- Nói thẳng: Người bạn có khả năng cao nhất phải ấn chính là người nhà bạn: 79.2% các ca ngừng tim xảy ra ở trong nhà. Có người ấn, tỷ lệ sống sót đến lúc xuất viện tăng từ 3.9% lên 16.1%. Gọi không đáp, không thở thì ấn luôn. Vị trí là giữa đường nối hai núm vú, chồng hai tay ấn mạnh, 2 lần mỗi giây, không dừng. Bảo người thứ hai gọi 120 và đi lấy AED, tay bạn không dừng. +- Lợi ích: Gộp 79 nghiên cứu, 142.740 người lại với nhau (phân tích tổng hợp): trong số người tim ngừng ngoài bệnh viện (ngừng tim ngoài bệnh viện), tổng cộng chỉ 7.6% sống sót đến lúc xuất viện. Khi có người chứng kiến làm hồi sinh tim phổi, tỷ lệ sống sót đến lúc xuất viện tăng từ 3.9% lên 16.1%. 53% có người bên cạnh nhìn thấy lúc bị lên cơn, thật sự được người chứng kiến ấn ngực chỉ 32%. Sổ đăng ký BASIC-OHCA của Trung Quốc ghi 38.227 ca cấp cứu được tiếp nhận. Trong đó 30.282 ca (79.2%) ngừng tim xảy ra tại nhà. Trong nhóm này, tỷ lệ được người chứng kiến ấn ngực là 20.3%, sống sót đến lúc xuất viện là 1.2%. Thứ tự là ấn ngực trước, rồi bảo người thứ hai gọi điện và đi lấy AED. Máy bật lên thì làm theo giọng nói của nó +- Mức bằng chứng: A +- Nguồn: Sasson C và cộng sự (2010). Predictors of survival from out-of-hospital cardiac arrest: a systematic review and meta-analysis. Circulation: Cardiovascular Quality and Outcomes. ; Zheng J và cộng sự (2023). Incidence, process of care, and outcomes of out-of-hospital cardiac arrest in China: a prospective study of the BASIC-OHCA registry. The Lancet Public Health. ; Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一百八十四条）. +- Ghi chú: Đừng bấm huyệt nhân trung trước, đừng đi tìm người thân trước, đừng quay video trước. Cũng đừng vì sợ bẻ gãy xương sườn mà không dám ấn mạnh. Điều 184 của Bộ luật Dân sự ghi rõ: người tự nguyện cứu hộ khẩn cấp mà làm người được cứu bị tổn thương thì người cứu hộ không chịu trách nhiệm dân sự. + +### 2. Người cao tuổi ngã, người bị ngã: quỳ xuống gọi họ, gọi 120, đừng vội đỡ dậy; với người xa lạ thì bước đi cũng hợp pháp, mà đã dừng lại thì đừng bê người bằng tay + +- Chi phí: Không tốn tiền. Quỳ xuống hỏi một câu, gọi một cuộc điện thoại, mất mấy phút. Khó nhất là nhịn cái tay đưa ra đỡ. Với người xa lạ còn phải tính thêm một khoản: gọi 120 sẽ để lại số của bạn, có thể bị gọi lại, có thể bị hỏi một câu. Nếu bị người nhà của họ nhắm vào thì kiện tụng là chuyện tính từ mấy tháng trở lên. Và dù cuối cùng tòa xử bạn thắng, phí luật sư vẫn phải tự bỏ ra. +- Nói thẳng: Chủ đạo của mục này là người cao tuổi trong nhà và người quen xung quanh. Trước hết xem có thở không, không thở thì ấn ngực như mục 1 (ấn mạnh vào giữa ngực). Có thở thì cũng đừng bế dậy: người đó có thể bị đột quỵ hoặc ngất trước rồi mới ngã, dựng cứng sẽ nặng thêm. Quỳ xuống gọi họ, hỏi họ nhớ mình ngã thế nào không. Đau lưng hông, giống gãy xương thì tuyệt đối không bê, gọi 120 ngay tại chỗ. +- Lợi ích: 《老年人跌倒干预技术指南》của Bộ Y tế (卫生部) viết rõ hai câu. Một câu là「跌倒是我国 65 岁以上老年人伤害死亡的首位原因」. Câu kia là「发现老年人跌倒，不要急于扶起，要分情况进行处理」. Người không tỉnh, gọi ngay số điện thoại cấp cứu. Người nôn, xoay đầu sang một bên và dọn miệng mũi. Người ngừng thở ngừng tim, làm nén ngực ngoài cơ thể ngay lập tức.「如需搬动，应保证平稳，尽量平卧」. Người còn tỉnh, hỏi trước họ có nhớ diễn biến lúc ngã không,「如不能记起，可能为晕厥或脑血管意外」. Đau đầu dữ dội, hoặc méo miệng, nói khó, tay chân không có lực thì「立即扶起老年人可能加重脑出血或脑缺血，使病情加重」. Có dấu hiệu gãy xương, hoặc đau lưng hông, mất kiểm soát tiểu tiện đại tiện thì「如无相关专业知识，不要随便搬动」. Tất cả các trường hợp trên đều phải gọi ngay số điện thoại cấp cứu. Người cao tuổi muốn tự đứng dậy thì có thể giúp họ đứng lên từ từ,「坐、卧休息并观察，确认无碍后方可离开」. Nói đến chuyện「sợ bị vu khống」thì pháp luật quy định thế này. Điều thứ chín mươi của《最高人民法院关于适用〈中华人民共和国民事诉讼法〉的解释》có hai câu. Câu đầu là「当事人对自己提出的诉讼请求所依据的事实……应当提供证据加以证明」. Câu thứ hai là「在作出判决前，当事人未能提供证据或者证据不足以证明其事实主张的，由负有举证证明责任的当事人承担不利的后果」. Nguyên văn Điều một nghìn một trăm chín mươi tám của Bộ luật Dân sự là:「宾馆、商场、银行、车站、机场、体育场馆、娱乐场所等经营场所、公共场所的经营者、管理者或者群众性活动的组织者，未尽到安全保障义务，造成他人损害的，应当承担侵权责任。」Điều một trăm tám mươi tư của Bộ luật Dân sự:「因自愿实施紧急救助行为造成受助人损害的，救助人不承担民事责任」. Tòa án nhân dân tối cao đã công bố một vụ án điển hình. Chủ hiệu thuốc làm hồi sinh tim phổi cho một người bị ngừng tim và bẻ gãy xương sườn của người đó. Người được cứu nằm viện 18 ngày, yêu cầu bồi thường hơn 9 nghìn tệ. Tòa phán quyết người cứu hộ không chịu trách nhiệm dân sự +- Mức bằng chứng: B +- Nguồn: Bộ Y tế (卫生部) (2011). 老年人跌倒干预技术指南（「老年人跌倒后的处理」）. Bản PDF đầy đủ ở Trung tâm kiểm soát và phòng ngừa bệnh Quảng Châu ，phần xử lý tại hiện trường có đầy đủ ở bản đăng lại của Chính phủ quận Quý Nam ; Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一百八十三、一百八十四条）. ; Tòa án nhân dân tối cao (最高人民法院) (2023). 人民法院抓实公正与效率践行社会主义核心价值观典型案例（案例十三，齐某某诉孙某某健康权纠纷案，辽宁省康平县人民法院）. ; Tòa án nhân dân tối cao (最高人民法院) (2022 年第二次修正，2022 年 4 月 10 日施行). 关于适用《中华人民共和国民事诉讼法》的解释（第九十条）. ; Điều một nghìn một trăm chín mươi tám của Bộ luật Dân sự như trên, bản đầy đủ do Viện kiểm sát nhân dân tối cao đăng lại ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 年修正，2024 年 1 月 1 日施行). Bộ luật Tố tụng dân sự, 民事诉讼法（第一百五十二、一百六十四条）. Bản đầy đủ đăng lại ở Ủy ban Phát triển và cải cách Thượng Hải ; Hội đồng Nhà nước (国务院) (2006). 诉讼费用交纳办法（国务院令第 481 号，第六、二十九条）. ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，2026 年 1 月 1 日起施行，第二十六条第一项、第五十条第五项）. +- Ghi chú: Để ở mức B là vì hướng dẫn chỉ nói cách xử lý, không có con số「làm theo thì giảm được bao nhiêu ca tử vong」. Điều 184 miễn trách cho tổn thương phát sinh trong lúc cứu người, điều kiện là bạn tự nguyện và tình huống khẩn cấp. Những việc như「tự ý nhét người vào xe riêng chở đi bệnh viện」nó không lo, chờ 120 thì chắc hơn. Gọi 120 xong đừng tút máy, làm theo lời người trực tổng đài. Bộ phán đoán này trước hết dành cho người cao tuổi trong nhà và người quen. Đổi sang người không quen gặp ngoài đường thì bước đi không vi phạm luật, và cũng tiết kiệm nhất: pháp luật không bắt bạn phải cứu. Cái giá phải trả nếu dừng lại thì phải tính từng bước: thời gian, để lại số điện thoại, bị người nhà của họ tìm đến, bị kiện ra tòa, ồn đến cơ quan nơi làm việc, bị quay video và đăng lên mạng. Xem chi tiết [docs/遇到陌生人出事该不该停.md](../docs/遇到陌生人出事该不该停.md). Nếu quyết định dừng lại thì việc duy nhất bắt buộc phải làm là**đừng bê người bằng tay**, lý do nằm ở y học: dựng cứng có thể làm nặng thêm xuất huyết não hoặc tổn thương cột sống. Có một ngoại lệ: nếu nơi đó do bạn kinh doanh, bạn quản lý, hoặc hoạt động do bạn tổ chức thì bước đi có thể phải chịu trách nhiệm. Nghĩa vụ bảo đảm an toàn theo Điều 1198 của Bộ luật Dân sự đòi bạn gánh một phần trách nhiệm an toàn cho những người ở trong khu vực đó. Bản thân chuyện phòng ngã xem mục 13 ở phần 1 (tập thăng bằng và sức chân). Triệu chứng đến muộn sau khi người cao tuổi bị va đầu, xem mục 10 của phần này. Chuyện tiền sau khi cứu người mà bị thương, xem mục 39 của phần này. Bị công kích trên mạng thì lấy bằng chứng và báo công thế nào, xem mục 16 ở phần 8. + +### 3. Miệng lệch, một bên tay không có lực, nói không rõ: gọi 120 ngay, đừng chờ, đừng tự lái xe đi + +- Chi phí: Không tốn tiền. Ba động tác, xem một lần là biết: bảo người đó cười, giơ hai tay ngang thân, nói một câu trọn vẹn. +- Nói thẳng: Miệng lệch, một bên không giơ nổi tay lên, nói không rõ, gọi 120 ngay. Dùng thuốc mở mạch máu trong 3 giờ đầu kể từ lúc khởi phát, khả năng hồi phục đến mức tự lo cơ bản cao khoảng 1.75 lần so với không dùng thuốc. Kéo dài đến 3 đến 4.5 giờ thì chỉ còn 1.26 lần. Quá 4.5 giờ thì không thấy khác biệt. Ghi lại thời điểm chính xác bắt đầu có triệu chứng. Không cho uống nước, không cho uống thuốc, không tự lái xe. +- Lợi ích: Gộp dữ liệu cá nhân của 16 thử nghiệm ngẫu nhiên về can thiệp mạch bằng thuốc tiêm tĩnh mạch, trên 6756 bệnh nhân (phân tích tổng hợp). Can thiệp mạch bằng thuốc tiêm tĩnh mạch là dùng thuốc để mở mạch máu bị tắc. Dùng alteplase trong 3 giờ đầu kể từ lúc khởi phát, tỷ số hồi phục đến mức tự lo cơ bản (thang điểm Rankin cải tiến 0–1) là 1.75, 95% CI 1.35–2.27 (khoảng tin cậy). Khả năng đạt được kết quả đó cao hơn khoảng 1.75 lần so với không can thiệp mạch. Kéo dài đến 3 đến 4.5 giờ là 1.26 (1.05–1.51). Quá 4.5 giờ là 1.15 (0.95–1.40), xét thống kê đã không thấy khác biệt +- Mức bằng chứng: A +- Nguồn: Emberson J, Lees KR, Lyden P, et al. (2014). Effect of treatment delay, age, and stroke severity on the effects of intravenous thrombolysis with alteplase for acute ischaemic stroke: a meta-analysis of individual patient data from randomised trials. Lancet, 384(9958), 1929-1935. +- Ghi chú: Can thiệp mạch càng sớm càng hiệu quả, quá 4.5 giờ thì gần như không dùng được. Gọi 120 thay vì tự lái xe, vì xe cứu thương đưa thẳng tới bệnh viện có trung tâm đột quỵ và còn báo trước để bên đó chuẩn bị. Ghi lại thời điểm chính xác bắt đầu có triệu chứng, bác sĩ dùng nó để quyết định có can thiệp mạch được không. Không cho uống nước, không cho uống thuốc, đừng nghĩ「nghỉ một lát xem sao」. + +### 4. Chóng mặt xoay tròn không đứng vững, nhìn thấy hai hình, một mắt tối đi, hoặc chỉ tay không trúng mũi mình: vẫn coi như đột quỵ và gọi 120 + +- Chi phí: Không tốn tiền. Ngoài ba động tác「miệng lệch, giơ tay, nói」(mục 3, miệng lệch, nói không rõ) thì thêm hai động tác nữa. Một là bảo người đó đứng thẳng nhắm mắt, xem có chệnh không, rồi hỏi người đó nhìn vật có thành hai hình không, có mất một mảng không. Động tác còn lại là khi bạn nghi ngờ thì giơ ngón tay chỉ vào mũi, nhắm mắt chỉ lại một lần. +- Nói thẳng: Chỉ nhìn ba dấu hiệu「miệng lệch, giơ tay, nói」thì hơn một phần mười số ca đột quỵ sẽ bị bỏ sót, vì có ca đột quỵ lúc đầu vừa không méo miệng vừa không yếu tay. Tính thêm đứng không vững, nhìn thành hai hình hoặc mất một mảng vào thì tỷ lệ bị bỏ sót giảm còn 4.4%. Chóng mặt xoay tròn không đứng vững, nhìn thấy hai hình, chỉ tay không trúng mũi mình, thì vẫn gọi 120, vẫn ghi lại thời điểm. +- Lợi ích: Trung tâm đột quỵ của Đại học Kentucky, Mỹ, năm 2014 rà lại hồ sơ bệnh của 736 ca đột quỵ thiếu máu cấp (tắc mạch máu não gây đột quỵ). Trong đó 14.1% bệnh nhân lúc khởi phát hoàn toàn không có ba triệu chứng miệng lệch, tay yếu, nói không rõ (FAST). Tính thêm「đi loạng, chân không có lực, triệu chứng thị giác」vào khám sàng lọc thì tỷ lệ bị bỏ sót giảm từ 14.1% xuống 4.4% +- Mức bằng chứng: B +- Nguồn: Aroor S, Singh R, Goldstein LB (2017). BE-FAST (Balance, Eyes, Face, Arm, Speech, Time): Reducing the Proportion of Strokes Missed Using the FAST Mnemonic. Stroke, 48(2), 479-481. +- Ghi chú: Mạch máu ở vùng tiểu não và thân não bị tắc (nhồi máu hệ tuần hoàn sau), thường không méo miệng, không yếu tay. Nó chỉ là chóng mặt xoay tròn, đi lệch, nhìn thấy hai hình, ăn uống bị sặc, dễ bị nhầm là bệnh cổ sống, hạ đường huyết, hay chứng rối loạn tiền đình rồi chậm trễ. Chỉ tay không trúng và nhắm mắt không đứng vững là dấu hiệu tiểu não bị ảnh hưởng. Nhưng không động tác nào loại trừ được đột quỵ: xuất hiện đột ngột, không đỡ thì gọi 120, và ghi lại thời điểm chính xác lúc bị. Nghiên cứu đó chỉ xem hồ sơ của một bệnh viện, tác giả tự ghi còn phải kiểm chứng lại. + +### 5. Một mắt tối đi đột ngột như kéo xuống rèm, dù chỉ vài phút rồi tự hết thì vẫn đi cấp cứu ngay trong ngày theo đường đột quỵ + +- Chi phí: Không tốn tiền. Đánh giá cấp cứu và kiểm tra mạch máu chiếm nửa ngày đến một ngày. +- Nói thẳng: Một mắt tối đi đột ngột như kéo xuống rèm là mạch máu đáy mắt bị tắc, tắc càng lâu thì phần võng mạc cứu lại được càng ít. Nó cũng là một dạng thiếu máu não, mấy ngày sau dễ đột quỵ. Người được đánh giá và bắt đầu điều trị ngay trong ngày thì tỷ lệ tái phát đột quỵ trong 90 ngày giảm từ 10.3% xuống 2.1%. Hãy đi cấp cứu trong ngày ở nơi có trung tâm đột quỵ, đừng chờ ngày hôm sau để xếp hàng khám mắt. +- Lợi ích: Tuyên bố khoa học năm 2021 của Hiệp hội Tim Hoa Kỳ về tắc động mạch trung tâm võng mạc (mạch máu chính đáy mắt bị tắc) viết rõ:「急性视网膜中央动脉阻塞是急症」. Hệ thống cấp cứu phải nhận diện nó như một ca cấp cứu và xếp theo mức nghiêm trọng. Máu bị cắt càng lâu, phần võng mạc cứu lại được càng ít. Nghiên cứu EXPRESS của Oxford xem hai nhóm bệnh nhân: đột quỵ nhỏ, và thiếu máu não tự hết triệu chứng (y học gọi là cơn thiếu máu não thoáng qua). Sau khi nhóm này chuyển sang đánh giá và bắt đầu điều trị ngay trong ngày, tỷ lệ tái phát đột quỵ trong 90 ngày giảm từ 10.3%（32/310）xuống 2.1%（6/281） +- Mức bằng chứng: B +- Nguồn: Mac Grory B, Schrag M, Biousse V, et al. (2021). Management of Central Retinal Artery Occlusion: A Scientific Statement From the American Heart Association. Stroke, 52(6), e282-e294. ; Rothwell PM, Giles MF, Chandratheva A, et al. (2007). Effect of urgent treatment of transient ischaemic attack and minor stroke on early recurrent stroke (EXPRESS study): a prospective population-based sequential comparison. Lancet, 370(9596), 1432-1442. +- Ghi chú: Để ở mức B là vì EXPRESS không phải thử nghiệm chia nhóm ngẫu nhiên. Nó so hai giai đoạn trước và sau khi đổi cách làm trên cùng một dân số, giữa hai giai đoạn đó các điều kiện khác cũng có thể đã thay đổi, không tách được. Mắt tối đi vài phút rồi tự hết là dạng dễ nhầm nhất với「mệt quá」「tụt đường huyết」rồi bỏ qua. Đó là một dạng cơn thiếu máu não thoáng qua, mấy ngày sau đó đúng là lúc đột quỵ xuất hiện nhiều nhất. Một mắt mất thị lực mà không đau, không đỏ, không chảy nước mắt thì càng phải coi trọng. Đừng chờ ngày hôm sau xếp hàng khám mắt ngoại trú, đi thẳng vào cấp cứu ở bệnh viện có trung tâm đột quỵ. + +### 6. Một mắt sưng đau, đỏ, nhìn đèn thấy một vòng cầu vồng, kèm đau đầu buồn nôn, nôn ói: đi cấp cứu mắt trong ngày + +- Chi phí: Không tốn tiền. Cấp cứu mắt đo nhãn áp và soi tiền đồ chiếm nửa ngày. +- Nói thẳng: Đây là cơn glô-cô-mơ phát cấp. Dịch nước trong mắt không thoát ra được, nhãn áp bị đẩy lên đột ngột. Kéo dài bao lâu thì dây thần kinh thị giác bị tổn thương nặng dần, tổn thương rồi thì không hồi lại được. Người bệnh trong thời gian phong tỏa ở Vũ Hán phải chờ trung bị 241 giờ mới được điều trị, tỷ lệ mù là 21.87%. Năm sau cùng kỳ không phong tỏa, chờ trung bị 121 giờ, tỷ lệ mù là 7.84%. +- Lợi ích: Một bệnh viện mắt ở Vũ Hán so hai giai đoạn bệnh nhân glô-cô-mơ góc đóng nguyên phát cấp. Trong 76 ngày phong tỏa năm 2020 có 54 người, 64 mắt; cùng kỳ năm 2021 có 46 người, 51 mắt; đặc điểm dân số hai nhóm tương đương. Từ lúc xuất hiện triệu chứng đến lúc được điều trị, nhóm phong tỏa là 241.84±211.95 giờ, nhóm đối chứng là 121.53±96.12 giờ (P=0.001). Nhãn áp lúc đến khám là 52.63±12.45 mmHg so với 45.16±9.79 mmHg (P=0.001). Tỷ lệ mù là 21.87% so với 7.84%. Tổn thương dây thần kinh thị giác do glô-cô-mơ gặp ở 20/64 mắt (31.25%) so với 7/51 mắt (13.73%) (P=0.03). Một bệnh viện ở Hàn Quốc rà lại 50 con mắt từng lên cơn một lần rồi sau đó bị lấy thủy tinh thể. Sau một năm, 25 con mắt (50%) còn lại tổn khuyết thị trường ở các mức độ khác nhau. Khoảng thời gian từ lúc có triệu chứng đến lúc hạ được nhãn áp càng dài thì khả năng để lại tổn khuyết càng lớn (P=0.005). Một nghiên cứu tiền cứu khác cho 43 người có tiền đồ nông nằm sấp trong phòng tối 1 giờ. Nhãn áp ở cả hai mắt đều tăng rõ rệt (P<0.01). Ở con mắt chủ yếu dùng hằng ngày, mức tăng trung vị là 3.60 mmHg, con mắt còn lại tăng 2.70 mmHg (P<0.05) +- Mức bằng chứng: B +- Nguồn: Zhou L, Wu S, Wang Y, Bao X, Peng T, Luo W, Ortega-Usobiaga J (2022). Clinical presentation of acute primary angle closure during the COVID-19 epidemic lockdown. Frontiers in Medicine, 9, 1078237. ; Sung MS, Kim HJ, Park SW (2023). Predictors of long-term visual field outcome after an episode of acute primary angle closure. Clinical & Experimental Ophthalmology, 51(4), 291-299. ; Wang J, Wang J, Ng TK, Huang C (2025). Asymmetric intraocular pressure changes in dominant and contralateral eyes: the dark room prone provocative test. Seminars in Ophthalmology, 40(4), 325-331. +- Ghi chú: Có hai lý do để ở mức B. Một là hai nhóm ở Vũ Hán được so bằng cách đọc lại hồ sơ bệnh cũ, bản thân việc phong tỏa còn tạo ra những khác biệt khác, không tách được khỏi yếu tố「kéo dài」. Hai là nhóm ở Hàn Quốc chỉ có 50 con mắt, cũng là nghiên cứu hồi cứu. Yếu tố kích thích không phải là nhìn màn hình, mà là hai việc chồng lên nhau: ở chỗ tối đồng tử giãn ra, còn cúi đầu hoặc nằm sấp lâu lại đẩy thủy tinh thể ra phía trước. Người vốn đã có tiền đồ nông thì dễ bị kích thích như vậy, nên nhóm thật sự có nguy cơ cao là người trên 50 tuổi, người viễn thị, và người có người thân trong nhà từng bị glô-cô-mơ góc đóng. Người trẻ thức cả đêm chơi game hầu như không lên cơn như vậy. Sai lệch dễ gặp nhất lúc lên cơn là nhầm sang viêm dạ dày ruột, đau nửa đầu, hoặc huyết áp cao, vì đau đầu và nôn biểu hiện rõ hơn triệu chứng ở mắt. Một mắt mất thị lực mà không đau không đỏ là chuyện khác, xem mục 5 (một mắt tối đi đột ngột). Kính chống ánh sáng xanh có ích không, xem mục 16 ở phần 6 (kính chống ánh sáng xanh). + +### 7. Đau tức ngực kiểu bóp nghẽn, khó thở, cứng ngực quá 15 phút không dịu: gọi 120, đừng chịu đựng và đừng tự lái xe + +- Chi phí: Không tốn tiền. +- Nói thẳng: Đau tức ngực kiểu bóp nghẽn, khó thở, cứng ngực quá 15 phút không dịu thì gọi 120. Đến bệnh viện mà nong mạch máu trong vòng 55 phút thì tỷ lệ tử vong trong 30 ngày là 2.8%. Quá 55 phút thì 9.7%. Mỗi đoạn chậm trễ là thêm một mảng cơ tim chết hoại. Đừng tự lái xe, mất ý thức trên đường là một tai nạn thứ hai. +- Lợi ích: Israel từng làm một khảo sát toàn quốc về hội chứng mạch vành cấp. Đối tượng phân tích là bệnh nhân nhồi máu cơ tim có đoạn ST nâng lên, một loại nhồi máu cơ tim. Tính khoảng cách từ lúc vào bệnh viện đến lúc dùng bóng nong mạch máu bị tắc (nong bóng). Khoảng cách ≤55 phút thì tử vong trong 30 ngày là 5/178（2.8%）. Khoảng cách >55 phút thì tử vong trong 30 ngày là 15/155（9.7%） +- Mức bằng chứng: B +- Nguồn: Karkabi B, Meir G, Zafrir B, et al. (2021). Door-to-balloon time and mortality in patients with ST-elevation myocardial infarction. European Heart Journal - Quality of Care and Clinical Outcomes. +- Ghi chú: Nghiên cứu đó so đoạn bị chậm sau khi đã vào bệnh viện, không phải đoạn từ lúc khởi phát đến lúc vào bệnh viện. Nhưng lý do thì vẫn vậy: chậm thêm một chút nghĩa là thêm một mảng cơ tim chết hoại. Cơn đau có thể nằm ở bụng trên, ở lưng, ở cằm, cũng có thể chỉ là tê một bên cánh tay. Người bị tiểu đường và người cao tuổi có thể không đau, chỉ thấy khó thở. Đừng tự lái xe, mất ý thức trên đường là một tai nạn thứ hai. + +### 8. Đau dữ dội như bị xé toạc, lan từ ngực ra sau lưng và xuống thắt lưng: gọi 120 và nói rõ「cơn đau đang di chuyển」 + +- Chi phí: Không tốn tiền. +- Nói thẳng: Đau dữ dội như bị xé từ ngực lan ra sau lưng và xuống thắt lưng, thì lúc gọi 120 cứ nói thẳng「như bị xé, cơn đau đang đi xuống」. Bệnh này không mổ, chỉ uống thuốc thì tỷ lệ tử vong 58%. Cách xử lý lại ngược hẳn với nhồi máu cơ tim: bác sĩ phải nghĩ đến nó thì mới làm CT cải tiến, chứ không cho bạn dùng thuốc chống đông và làm tan cục máu. Nói thêm là huyết áp hai tay chênh nhau nhiều, một bên không sờ thấy mạch. +- Lợi ích: Sổ đăng ký quốc tế về bóc tách động mạch cấp (IRAD) thu 464 ca từ 12 trung tâm. Bóc tách động mạch là căn bệnh đau dữ dội kiểu xé ngực lưng. Trong các triệu chứng người bệnh kể ra trước tiên, phổ biến nhất là đau nhói dữ dội khởi phát đột ngột. Tỷ lệ tử vong toàn nguyên trong thời gian nằm viện là 27.4%. Bóc tách loại A không mổ, chỉ uống thuốc, tỷ lệ tử vong là 58% +- Mức bằng chứng: B +- Nguồn: Hagan PG, Nienaber CA, Isselbacher EM, et al. (2000). The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. JAMA, 283(7), 897-903. +- Ghi chú: Cách xử lý bóc tách động mạch ngược hẳn với nhồi máu cơ tim. Nhồi máu cơ tim cần thuốc chống đông, thuốc làm tan cục máu, tức là thuốc làm cho máu không đông lại; dùng thuốc đó lên bóc tách thì hậu quả là máu chảy qua chỗ rách. Vì vậy đến bệnh viện hãy chủ động nói「xé như thế, cơn đau đang đi xuống」để bác sĩ nghĩ đến việc chụp CT cải tiến. Huyết áp hai tay chênh nhau nhiều, một bên không sờ thấy mạch là dấu hiệu. Người kiểm soát huyết áp không tốt, người mắc hội chứng Marfan, người trong nhà từng bị phình động mạch hoặc bóc tách động mạch thì nguy cơ cao. Cũng đừng tự lái xe, cũng đừng chịu đựng đến ngày hôm sau. + +### 9. Đột nhiên một cơn「đau đầu nặng nhất đời», trong vòng một giờ đã tới lúc đau nhất: đi cấp cứu ngay để chụp CT sọ + +- Chi phí: Đi cấp cứu bản thân không tốn bao nhiêu tiền. Một lần chụp CT sọ ở khoa cấp cứu vài trăm tệ. +- Nói thẳng: Cần xem là phải nhanh bao lâu thì đau tới mức nặng nhất. Với đau đầu dữ dội mà trong một giờ đã đau tới đỉnh, khoảng 6% là do mạch máu ngoài não vỡ. Trên 40 tuổi, đau cổ hoặc cứng cổ, có người nhìn thấy bạn ngất, khởi phát khi đang hoạt động, bùng nổ trong tích tắc, không cúi cổ xuống được, trúng một dấu hiệu là đi cấp cứu chụp CT sọ. Người đi khám đa số không có gì, nhưng bỏ sót một lần rồi tái xuất huyết là chết. +- Lợi ích: 10 khoa cấp cứu ở Canada làm một nghiên cứu mẫu tiền cứu trên 2131 ca「đau đầu dữ dội đạt đến mức đau nhất trong một giờ」, chốt trước tiêu chí rồi theo dõi ghi nhận suốt quá trình. Trong đó 132 ca (6.2%) là xuất huyết dưới nhện (mạch máu ngoài não vỡ). Quy tắc SAH Ottawa có sáu điều: từ 40 tuổi trở lên, đau cổ hoặc cứng cổ, có người nhìn thấy ngất. Ba điều còn lại là khởi phát khi đang hoạt động, khởi phát bùng nổ thành một thể, và khi bác sĩ khám thì cổ không cúi xuống được. Khớp bất kỳ điều nào thì phải kiểm tra. Quy tắc này lọc ra được toàn bộ số người bị xuất huyết, độ nhạy 100%, 95% CI 97.2%–100.0%（khoảng tin cậy）. Cái giá phải trả là trong số người không bị xuất huyết chỉ 15.3% bị nó loại ra (độ đặc hiệu 15.3%) +- Mức bằng chứng: A +- Nguồn: Perry JJ, Stiell IG, Sivilotti MLA, et al. (2013). Clinical decision rules to rule out subarachnoid hemorrhage for acute headache. JAMA, 310(12), 1248-1255. +- Ghi chú: Cần xem không phải「đau nặng cỡ nào」mà là「bao lâu thì đau tới đỉnh」; đau đầu bùng nổ trong vài giây đến một phút mới là dấu hiệu nguy hiểm, đau nửa đầu tăng dần thì không phải. Kèm nôn, cứng cổ, sợ ánh sáng, ngất thoáng qua thì càng phải coi trọng. Độ đặc hiệu chỉ 15.3%, nghĩa là đa số người đi kiểm tra theo quy tắc này thì kiểm tra xong không có gì. Nhưng bỏ sót một lần rồi tái xuất huyết là chết, nên đáng để kiểm tra. Đừng uống viên thuốc giảm đau để ép rồi đi ngủ tiếp. + +### 10. Người cao tuổi va đầu rồi hai ba tuần đến vài tháng sau mà đi loạng, uể oải, ngủ gật hoặc một bên không có lực: đi chụp CT sọ + +- Chi phí: Một lần chụp CT sọ vài trăm tệ. +- Nói thẳng: Sau khi người cao tuổi va đầu, máu thường đến tuần thứ ba mới tích lại. Lúc vừa va thì có thể hoàn toàn không sao, đến nỗi người đó không nhớ là mình đã va. Sau khi mổ, tỷ lệ tái phát tới được 33%, tỷ lệ tử vong tới được 32%. Thấy người cao tuổi lú lẫn bỗng, đi loạng bước, lúc nào cũng muốn ngủ, một tay không có lực thì trước hết quay lại hỏi trong hai ba tháng qua có ngã hay va đầu không, rồi mới đi chụp CT. +- Lợi ích: Chi hội Phẫu thuật Thần kinh của Hiệp hội Y khoa Trung Quốc (中华医学会神经外科学分会) cùng các cơ quan khác đã công bố《慢性硬膜下血肿药物治疗中国专家共识》. Máu tụ dưới màng cứng mạn là một vũng máu tích dần giữa các màng não. Cách gọi chuyên môn là tổn thương chiếm chỗ mạn do máu tích giữa màng nhện và màng cứng. Nó「通常在颅脑外伤后第三周形成」, hay gặp ở người cao tuổi. Sau khi mổ, tỷ lệ tái phát có thể tới 33%, tỷ lệ tử vong có thể tới 32%. Ở bệnh nhân trên 90 tuổi, nhìn chung chỉ 24% hồi phục tốt. Rà soát hệ thống: toàn cầu mỗi năm có 3.39 đến 39.1 ca mới trên 100.000 người, tỷ lệ tăng theo tuổi, trên 80 tuổi có thể gấp ba. Nguyên nhân thường gặp nhất là chấn thương và ngã, sau đó là nghiện rượu lâu năm, dùng thuốc chống đông máu và bạo lực. Rà soát y học người cao tuổi: biểu hiện lâm sàng「典型包括意识模糊、认知下降、步态障碍或嗜睡，可能与其他老年综合征相似」 +- Mức bằng chứng: B +- Nguồn: Zhang J và cộng sự (2021). Expert consensus on drug treatment of chronic subdural hematoma. Chinese Neurosurgical Journal, 7(1), 47. ; Dziho A và cộng sự (2025). Global prevalence and incidence of chronic subdural hematoma: A systematic review. Brain and Spine, 5, 105893. ; Mathon B, Shotar E (2026). Diagnostic and therapeutic management of chronic subdural hematoma in elderly patients. Gériatrie et Psychologie Neuropsychiatrie du Vieillissement, 24(1), 30-41. +- Ghi chú: Lúc va thì có thể hoàn toàn không sao, đến nỗi người cao tuổi còn không nhớ là mình đã va đầu. Nên trong nhà thấy「dạo này bỗng lú lẫn, đi loạng bước, lúc nào cũng muốn ngủ, một tay không có lực」thì trước hết quay lại hỏi trong hai ba tháng qua có ngã, va đầu không, rồi mới đi chụp CT. Đừng quy thẳng là「sao trí nhớ」hay「di chứng của lần đột quỵ trước」, quy sai là chậm. Người cao tuổi uống aspirin, clopidogrel, warfarin hoặc thuốc chống đông máu đường uống thế hệ mới có nguy cơ cao hơn, uống rượu nhiều lâu năm cũng vậy. Phòng ngã xem mục 13 ở phần 1 (tập thăng bằng và sức chân). + +### 11. Một chân sưng, căng, ấn vào thì đau: đi khám càng sớm càng tốt; thêm khó thở đột ngột hoặc đau ngực thì gọi 120 ngay + +- Chi phí: Một lần siêu âm mạch máu chân (siêu âm tĩnh mạch chi dưới) một hai trăm tệ. +- Nói thẳng: Một chân sưng đột ngột, to hơn chân kia một vòng, ấn vào thì đau, hãy đi siêu âm tĩnh mạch chân càng sớm càng tốt. Nếu sau đó xuất hiện khó thở đột ngột, đau ngực khi hít sâu, nhịp tim nhanh, choáng, thì cục máu có thể đã chạy tới phổi, gọi 120 ngay. Người được xác định thuyết phối khối phổi thì 3 tháng tử vong 17.4%. Ngồi yên quá 4 giờ thì nguy cơ gần như tăng gấp đôi, nhưng tính ra trên một người vẫn rất thấp. +- Lợi ích: Sổ đăng ký hợp tác quốc tế về thuyết phối khối phổi (ICOPER) thu 2454 ca thuyết phối khối phổi cấp đã được xác định từ 52 bệnh viện ở 7 nước. Thuyết phối khối phổi là tình trạng cục máu bịch tắc động mạch phổi. Trong nhóm đó, tỷ lệ tử vong trong 3 tháng không rõ nguyên nhân là 17.4%（426/2454）. Dự án WRIGHT của Tổ chức Y tế Thế giới: chuyến đi liên tục trên 4 giờ làm nguy cơ máu tụ trong tĩnh mạch rồi cục máu chạy tới phổi (thuyết tắc mạch tĩnh mạch)「大约翻倍」. Nhưng tính xác suất trên một người vẫn rất thấp, ngồi không động quá 4 giờ thì「约为六千分之一」 +- Mức bằng chứng: B +- Nguồn: Goldhaber SZ, Visani L, De Rosa M (1999). Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet, 353(9162), 1386-1389. ; WHO (2007). Study results released on travel and blood clots. +- Ghi chú: Hai chân cùng sưng thì phần lớn là vấn đề của tim, gan, thận hay chức năng tĩnh mạch. Một chân sưng đột ngột, to hơn chân kia một vòng, mới là dáng vẻ điển hình của huyết khối tĩnh mạch sâu. Dễ xảy ra với người đi xe dài, đi máy bay, nằm liệt sau mổ hoặc sau gãy xương, bó bột, mang thai và trước sau khi sinh, bị ung thư, dùng thuốc tránh thai. Cục máu bong ra rồi bịt động mạch phổi chính là thuyết phối khối phổi: khó thở đột ngột, hít một hơi sâu thì đau ngực hơn, nhịp tim nhanh, choáng, ho ra máu (ho máu). Đến mức đó thì gọi 120 ngay. 17.4% là tỷ lệ tử vong của các ca thuyết phối khối phổi đã được xác định, không phải「chân sưng thì nguy cơ lớn cỡ này」, đừng dùng con số đó để tự dọn mình. + +### 12. Máu chảy nhiều: dùng tay ấn thật chặt vào vết thương, tứ chi ấn không được thì dùng gar, đồng thời gọi 120 + +- Chi phí: Một gar kiểu xoay vặn vài chục đến hơn một trăm tệ, để trong xe và trong túi cấp cứu. Học một lần cách「buộc trên vết thương, vặn đến khi ngừng máu, ghi giờ」, mất mấy phút. +- Nói thẳng: Trước hết dùng tay hoặc quần áo ấn thật chặt, tứ chi ấn không được thì dùng gar, rồi gọi 120, đã buộc thì không nới. Trong số thương vận chiến trường, người được buộc gar trước khi sốc thì sống sót chín phần mười, người buộc sau khi sốc thì chỉ sống sót một phần mười, cả 5 người không buộc đều chết. Buộc trên vết thương, vặn đến khi ngừng máu, ghi lại giờ. Đừng dùng dây giày hay sợi dây thay thế. Ở cổ và thân mình chỉ được ấn, không được buộc. +- Lợi ích: 11 trung tâm cấp cứu chấn thương cấp một tại bang Texas, Mỹ, trong 6 năm thu 1.026 ca chấn thương mạch máu tứ chi, trong đó 181 ca (17.6%) được buộc gar trước khi tới bệnh viện. Tính thẳng không điều chỉnh gì, tỷ lệ tử vong ở nhóm có buộc gar là 3.9%, nhóm không buộc là 5.2%. Sau khi đã cân bằng các điều kiện khác (hiệu chỉnh đa yếu tố), việc buộc gar vẫn liên quan riêng biệt đến việc sống sót. OR hiệu chỉnh là 5.86, khoảng tin cậy 95% CI 1.41 đến 24.47. Ở bệnh viện chiến trường Baghdad với 232 thương vận: người buộc gar trước khi sốc thì tỷ lệ sống sót 90%, người buộc sau khi sốc chỉ còn 10%. Cả 5 người cần gar mà không buộc đều chết, người đã buộc thì tỷ lệ sống sót 87%. Thứ tự là: ấn trực tiếp, quần áo hay khăn đều được; tứ chi ấn không được thì dùng gar; gọi 120; không nới +- Mức bằng chứng: A +- Nguồn: Teixeira PGR và cộng sự (2018). Civilian Prehospital Tourniquet Use Is Associated with Improved Survival in Patients with Peripheral Vascular Injury. Journal of the American College of Surgeons. ; Kragh JF và cộng sự (2009). Survival with emergency tourniquet use to stop bleeding in major limb trauma. Annals of Surgery. ; Chi hội Bác sĩ Cấp cứu của Hiệp hội Bác sĩ Trung Quốc (中国医师协会急诊医师分会) và cộng sự (2020). 止血带的急诊应用专家共识. 中华急诊医学杂志 29(6):773-779：「院前急救时，止血带一旦使用则不建议松开，除非得到可以替代的彻底止血的救治」「最长使用时间不应超过2 h。但如果客观情况无法到医院救治或者无替代止血办法，则在得到正规救援前不解除止血带」. ，bản đầy đủ +- Ghi chú: Đừng lấy dây, dây giày, những thứ mảnh làm gar, vì không siết được động mạch và còn cứa rách da. Đã buộc gar thì đừng cứ vài phút lại nới ra「xả máu». Ở cổ và thân mình chỉ được ấn, không được buộc. Trong số liệu chiến trường đó, bản thân cái gar không làm ai bị cắt chi, có 4 ca xuất hiện tê liệt thần kinh tạm thời. Ý kiến chung của các chuyên gia cấp cứu Trung Quốc nói cùng ý: đã dùng gar trước bệnh viện thì không nên nới, trừ khi có cách khác thay thế để cầm máu triệt để. Chảy máu ở tay trên buộc vào một phần ba trên của tay trên, chảy máu ở đùi buộc vào gần hông. Không buộc ở cẳng tay, bắp chân, khuỷu tay, khuỷu chân. Ý kiến chung khuyến nghị tổng thời gian từ lúc trước bệnh viện tới sau bệnh viện cố không quá 2 giờ, nhưng nếu không tới được bệnh viện và không có cách nào khác để cầm máu thì không mở ra cho tới khi có cứu hộ chính quy. Gar tạm làm từ dây lưng, cà vạt, tay áo thì đứng một mình khó cầm được máu, phải kèm một que để vặn siết. + +### 13. Bị chó, mèo cắn hoặc cào rách da: xịt xen kẽ nước xà phòng và nước chảy trong 15 phút, trong ngày đi tiêm vắc-xin + +- Chi phí: Rửa không tốn tiền. Vắc-xin dại cả liệu trình vài trăm tệ. Mức phơi nhiễm độ ba còn phải tiêm thêm huyết thanh miễn dịch bệnh dại (tiêm thẳng kháng thể đã có sẵn), vài trăm đến hơn một nghìn tệ. +- Nói thẳng: Bệnh dại một khi đã lên cơn thì tỷ lệ tử vong gần 100%, rửa vết thương cộng với tiêm vắc-xin là con đường duy nhất. Xịt xen kẽ nước xà phòng và nước sạch chảy trong 15 phút, trong ngày đi tiêm vắc-xin. Chỗ bị cắn thủng, chỗ da rách bị liếm, thuộc mức phơi nhiễm độ ba, phải tiêm thêm huyết thanh miễn dịch.「Con nuôi trong nhà, đã tiêm rồi」không làm thay đổi cách xử lý, mức độ xếp theo vết thương chứ không theo xuất thân của con chó. +- Lợi ích: Nguyên văn quy chuẩn năm 2023 của Cục Kiểm soát bệnh quốc gia (国家疾控局) và Ủy ban Quốc gia Y tế (国家卫健委) ghi:「狂犬病病死率几乎达 100%」. Cách xử lý là「用肥皂水（或者其他弱碱性清洁剂）和一定压力的流动清水交替彻底冲洗所有咬伤和抓伤处约 15 分钟」. Da lộ ra bị cắn nhẹ, bị cào mà không chảy máu, là mức phơi nhiễm II, cần tiêm vắc-xin. Vết cắn xuyên thủng, da rách bị liếm, là mức phơi nhiễm III,「应处置伤口并注射狂犬病被动免疫制剂和接种狂犬病疫苗」. WHO: mỗi năm trên toàn cầu khoảng 59.000 người chết vì bệnh dại, 99% do chó lây +- Mức bằng chứng: A +- Nguồn: Văn phòng Tổng hợp Cục Kiểm soát bệnh quốc gia và Văn phòng Ủy ban Quốc gia Y tế (国家疾控局综合司、国家卫生健康委办公厅) (2023). 关于印发狂犬病暴露预防处置工作规范（2023 年版）的通知. ; WHO. 狂犬病 实况报道. +- Ghi chú: Đừng vì「con nuôi trong nhà, đã tiêm rồi」mà bỏ qua. Quy chuẩn xem vết thương ở mức nào (mức độ phơi nhiễm), không xem con chó từ đâu ra. Đừng tự băng vết thương. Phương pháp quan sát mười ngày chỉ quyết định có dừng tiêm hay không, không quyết định có bắt đầu tiêm hay không. Bị động vật hoang dã, bị dơi cắn thì xử lý y hệt. + +### 14. Bị bỏng: xả ngay bằng nước mát chảy trong 20 phút, đừng bôi kem đánh răng hay nước tương + +- Chi phí: Không tốn tiền. Tốn mất 20 phút. +- Nói thẳng: Trong 3 giờ đầu sau khi bị thương mà xả nước mát chảy đủ 20 phút thì khả năng về sau phải ghép da chỉ còn khoảng sáu phần mười so với không xả. Là nước mát chứ không phải nước đá, chườm đá làm tổn thương nặng thêm. Quần áo dính vào vết thương thì đừng giật ra, xả xong dùng vải sạch phủ lên rồi đi bệnh viện. Kem đánh răng, nước tương, dầu mè và bột đều đừng bôi. Nghĩ ra muộn thì trong ba giờ đó xả bù vẫn còn tác dụng. +- Lợi ích: Australia theo dõi 2495 trẻ em bị bỏng. Ở những trẻ được xả nước mát chảy trong 20 phút trong 3 giờ đầu sau khi bị thương, tỷ số odds về sau phải ghép da là 0.6, 95% CI 0.4–0.8（khoảng tin cậy）, tức chỉ còn khoảng sáu phần mười so với nhóm không xả như vậy. Kết quả gộp 7 nghiên cứu cũng tìm ra điều tương tự:「伤后三小时内 20 分钟凉流动水显著降低了需要植皮和手术处理创面的可能」 +- Mức bằng chứng: A +- Nguồn: Griffin BR, Frear CC, Babl F, Oakley E, Kimble RM (2020). Cool Running Water First Aid Decreases Skin Grafting Requirements in Pediatric Burns: A Cohort Study of Two Thousand Four Hundred Ninety-five Children. Annals of Emergency Medicine, 75(1), 75-85. ；Griffin B, Cabilan CJ, Ayoub B, et al. (2022). The effect of 20 minutes of cool running water first aid within three hours of thermal burn injury on patient outcomes: A systematic review and meta-analysis. Australasian Emergency Care, 25(4), 367-376. +- Ghi chú: Là nước mát chứ không phải nước đá, chườm đá làm tổn thương nặng thêm. Quần áo dính vào vết thương thì đừng giật ra, xả xong dùng vải sạch phủ lên rồi đi bệnh viện. Kem đánh răng, nước tương, dầu mè và bột đều đừng bôi, vì làm bác sĩ khó đánh giá độ sâu và dễ nhiễm trùng. Trong ba giờ đầu xả bù vẫn còn tác dụng. + +### 15. Nổi mề đay toàn thân, khó thở hoặc choáng váng: xử lý như sốc phản vệ, gọi 120 ngay và nói rõ chuyện gì + +- Chi phí: Không tốn tiền. Người đã biết mình dị ứng nặng thì phải tốn tiền mua bút adrenaline, và đó là thuốc kê đơn. +- Nói thẳng: Đột nhiên nổi mề đay toàn thân, kèm khó thở hoặc choáng váng, thì xử lý như sốc phản vệ. Gọi 120 ngay, nói thẳng「怀疑过敏性休克」 — nghi ngờ sốc phản vệ — và bảo xe cứu thương mang adrenaline đến. Thuốc chống dị ứng và hormone tác dụng chậm, không thay được adrenaline. Người đã từng bị thì xin bác sĩ kê bút tiêm tự động, tiêm vào phần giữa phía ngoài đùi, xuyên qua cả quần vẫn tiêm được. +- Lợi ích: Adrenaline là thuốc lựa chọn đầu tiên (thuốc tuyến một) cho sốc phản vệ mức II trở lên, và hướng dẫn xếp nó là khuyến cáo mạnh. Thuốc chống dị ứng và hormone tác dụng chậm, không đứng được chỗ của nó +- Mức bằng chứng: B (khuyến cáo mạnh trong hướng dẫn, nhưng mức bằng chứng nền thấp) +- Nguồn: Hiệp hội Giáo dục Y khoa Trung Quốc (中国医药教育协会) và cộng sự (2022). Hướng dẫn thực hành lâm sàng cấp cứu sốc phản vệ (2020). Frontiers in Pharmacology: 「Epinephrine is the first-line medicine in GRADE II to IV anaphylaxis」「Intramuscular epinephrine should be injected in the mid-anterolateral thigh」, liều tiêm bắp tối đa ở người lớn 0.5 mg, dưới 14 tuổi là 0.3 mg, không có phản ứng thì lặp lại mỗi 5–15 phút. ；Muraro A và cộng sự (2022). EAACI guidelines: Anaphylaxis (2021 update). Allergy. +- Ghi chú: Bản hướng dẫn dị ứng này (EAACI) tự viết rằng trong lĩnh vực đó「证据大多停留在很低的等级」 — bằng chứng phần lớn dừng ở mức rất thấp. Nó xếp khuyến cáo mạnh vì lợi ích lớn hơn rủi ro nhiều, không phải vì có thử nghiệm lớn. Adrenaline ở Trung Quốc là thuốc kê đơn, người thường không có trong tay. Việc làm được là gọi 120, nói「怀疑过敏性休克」 — nghi ngờ sốc phản vệ — và bảo xe cứu thương mang thuốc đến. Người đã từng dị ứng nặng thì xin bác sĩ kê bút tiêm tự động, và học cách tiêm vào phần giữa phía ngoài đùi (phía ngoại, hơi chếch trước), xuyên qua cả quần vẫn tiêm được. + +### 16. Người ta co giật rồi ngã xuống: dọn chỗ xung quanh, cho nằm nghiêng, nhìn đồng hồ đếm thời gian, không đưa gì vào miệng + +- Chi phí: Không tốn tiền. +- Nói thẳng: Dời các vật cứng quanh đó đi, cho nằm nghiêng với miệng hướng xuống, và nhìn đồng hồ ghi thời gian. Đừng ấn giữ, đừng cạy miệng, đừng nhét thứ gì vào miệng, làm vậy chỉ làm tổn thương răng và hàm. Cơn co giật kéo dài quá 5 phút không dừng, dừng xong lại lên rất sớm, sau cơn không gọi tỉnh được hoặc khó thở, xảy ra dưới nước, hoặc đây là cơn đầu tiên, đều phải gọi xe cứu thương. Chỉ cho uống nước khi người đó đã tỉnh hẳn. +- Lợi ích: Tránh tổn thương răng và hàm do cạy miệng và nhét vật cứng vào. Tránh gãy xương do ấn giữ quá mức. Cơn co giật không dừng sau 5 phút là vạch đỏ bắt buộc phải gọi xe cứu thương +- Mức bằng chứng: B +- Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Cấp cứu khi người bị động kinh co giật: 「Don't hold the person down or stop their movements」「Don't put anything in their mouth. This can hurt their teeth or jaw」「Don't offer water or food until they are fully alert」「Time the seizure to track if it lasts more than 5 minutes. If it does, seek immediate medical attention or call 911」 — đừng ấn giữ người đó xuống hay chặn cử động của họ; đừng đưa bất cứ thứ gì vào miệng, làm vậy có thể làm tổn thương răng hoặc hàm; chỉ cho ăn uống khi họ đã hoàn toàn tỉnh táo; đếm thời gian cơn co giật để biết có kéo dài quá 5 phút không, nếu có thì tìm cấp cứu y tế ngay hoặc gọi 911. +- Ghi chú: Trang đó còn liệt thêm vài trường hợp phải gọi xe cứu thương ngay: lên cơn rất sớm sau khi một cơn vừa kết thúc, sau cơn không gọi tỉnh được hoặc khó thở, bị thương trong lúc co giật, xảy ra dưới nước, và cơn đầu tiên. Cơn dừng thì cho nằm nghiêng với miệng hướng xuống, chờ người đó tỉnh hẳn rồi hãy cho nước. Trong nhà có người bị động kinh thì đây là những thao tác dùng đến nhiều nhất. Làm vậy cho người xa lạ thì khả năng đổi lại cho bạn nhỏ hơn, nhưng bạn cũng không phải chịu trách nhiệm (Điều 184 của Bộ luật Dân sự, xem mục 2 của phần này, đừng vội vàng dựng người đó dậy). + +### 17. Người bị tiểu đường đột nhiên run, đổ mồ hôi lạnh, nói không rõ: cho 15 gram đường trước, 15 phút sau đo lại + +- Chi phí: Không tốn tiền. Mang theo vài viên đường vuông hoặc một hộp nhỏ nước trái cây. +- Nói thẳng: Cho 15 gram đường trước, đường vuông hay một hộp nhỏ nước trái cây đều được. 15 phút sau đo lại một lần nữa, vẫn thấp thì làm lại. Đường huyết dưới 3.0 mmol/L là hạ đường huyết mà y học phải để ý. Khi tới mức phải nhờ người khác xử lý giúp thì có thể tử vong. Người không tỉnh thì đừng cho ăn uống lấy miếng nào, sẽ hút vào khí quản, gọi thẳng 120. +- Lợi ích: Đường huyết dưới 3.0 mmol/L là hạ đường huyết có ý nghĩa trên lâm sàng. Hạ đường huyết kèm thay đổi ý thức hoặc hành vi, phải nhờ người khác giúp xử lý, là hạ đường huyết nặng, có thể tử vong +- Mức bằng chứng: A +- Nguồn: Hiệp hội Đái thái đường Hoa Kỳ (美国糖尿病学会) (2025). Tiêu chuẩn chẩn đoán và điều trị đái thường 2025, chương 6 Mục tiêu đường huyết và hạ đường huyết. Diabetes Care: phân cấp「Level 1: Glucose <70 mg/dL (<3.9 mmol/L) and ≥54 mg/dL (≥3.0 mmol/L)」「Level 2: Glucose <54 mg/dL (<3.0 mmol/L)」「Level 3: A severe event characterized by altered mental and/or physical status requiring assistance for treatment of hypoglycemia, irrespective of glucose level」; xử lý「For most individuals, 15 g carbohydrates should be ingested」「Fifteen minutes after initial treatment, repeat the treatment if hypoglycemia persists」 — với đa số người, nên nạp 15 g carbohydrate; 15 phút sau khi xử lý lần đầu, lặp lại nếu hạ đường huyết còn. +- Ghi chú: Không cho người không tỉnh ăn uống vào, sẽ hút vào khí quản, gọi 120 thẳng. Cùng bộ tiêu chuẩn đó còn khuyên người dùng insulin và người hay bị hạ đường huyết nên có sẵn glucagon. Nguyên văn là「All individuals treated with insulin or who are at high risk of hypoglycemia...should be prescribed glucagon」 — mọi người được điều trị bằng insulin hoặc có nguy cơ cao bị hạ đường huyết… nên được kê glucagon. Dạng xịt mũi và dạng tiêm đã đóng sẵn thuốc dùng tiện hơn dạng phải pha lúc đó. Những cách này mặc định là chuẩn bị cho chính bạn hoặc người bị tiểu đường trong nhà bạn, và đường bạn mang theo cũng là mang cho họ. + +### 18. Người ta bị điện giật: cắt điện trước, rồi dùng gậy khô hay thứ gì tương tự gạp nguồn điện ra, tuyệt đối không kéo bằng tay + +- Chi phí: Không tốn tiền. +- Nói thẳng: Cắt điện trước, hoặc dùng gậy khô hay thứ gì tương tự để gạp nguồn điện ra, rồi mới chạm vào người đó. Kéo bằng tay thì một người bị thương thành hai. Sau khi ngắt khỏi nguồn điện mà người đó không thở, ấn ngay theo mục 1 của phần này (ấn mạnh vào giữa ngực), đừng sợ là chấn thương điện mà không dám chạm vào. Đường cao thế rớt xuống đất hay khu vực gần thiết bị biến áp thì không tự xử lý, lùi ra xa chờ ngành điện cắt nguồn. +- Lợi ích: Tránh người cứu nạn cùng bị điện giật, khiến một người bị thương thành hai. Sau khi ngắt khỏi nguồn điện mà người đó không thở, làm hồi sinh tim phổi theo mục 1 của phần này (ấn mạnh vào giữa ngực). Các bước cấp cứu của Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ là: nhìn trước, đừng chạm, vì người đó có thể vẫn đang nối với nguồn điện. Cắt được điện thì cắt, không cắt được thì dùng bìa carton, nhựa hay gỗ, những thứ không dẫn điện, để gạp nguồn điện ra. Sau khi ngắt khỏi nguồn điện, nếu nhịp thở hoặc mạch ngừng, làm hồi sinh tim phổi ngay. Một bài tổng quan nói người bị ngừng tim sau điện giật rồi được cứu sống thường có tiên lượng khá +- Mức bằng chứng: B +- Nguồn: US CDC. What to Do to Protect Yourself From Electrical Hazards (mục First aid). ；Spies C, Trohman RG (2006). Narrative review: Electrocution and life-threatening electrical injuries. Annals of Internal Medicine 145(7):531-537. +- Ghi chú: 「先断电再碰人」 — cắt điện trước rồi mới chạm vào người — là quy tắc thuần vì lợi ích cho bản thân: kéo bằng tay thì người thứ hai ngã xuống chính là bạn. Đường cao thế rớt xuống đất hay khu vực gần thiết bị biến áp thì không tự xử lý, lùi ra xa chờ ngành điện cắt nguồn. Ngừng tim do điện giật thì ấn theo cách thường lệ, khử rung theo cách thường lệ, đừng vì đó là「电伤」mà không dám làm. + +### 19. Báo động carbon monoxide kêu, hoặc cả nhà cùng đau đầu buồn nôn: ra ngoài trước rồi mới gọi + +- Chi phí: Không tốn tiền. +- Nói thẳng: Báo động kêu, hoặc cả nhà cùng đau đầu, buồn nôn, choáng váng, thì đưa tất cả mọi người ra ngoài trước, rồi mới gọi. Đừng ở lại trong nhà đi tìm nguyên nhân, cũng đừng quay lại đóng van. Triệu chứng giống cảm lạnh, và「出门就好转」 — ra ngoài là khỏe hơn — là dấu hiệu rõ nhất. Năm 2018 Trung Quốc báo cáo 11523 người chết vì ngộ độc carbon monoxide, nhiều nhất vào mùa đông, phần lớn chết ngay trong nhà. +- Lợi ích: Hệ thống giám sát nguyên nhân tử vong của Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc báo cáo 11523 ca tử vong vì ngộ độc carbon monoxide trong năm 2018. Nhiều nhất là tháng 1 (2845 ca), tháng 2 (2321 ca) và tháng 12 (2010 ca). Trong ba tháng mùa đông đó, từ 66% đến 73% số ca chết xảy ra ngay trong nhà. Đây là số được báo lên hệ thống, không phải suy ra từ dân số cả nước. Ở Hoa Kỳ, hơn 400 người mỗi năm chết vì ngộ độc carbon monoxide ngoài ý muốn, không liên quan đến hỏa hoạn. Hơn 10.000 người vào phòng cấp cứu, hơn 14.000 người nhập viện +- Mức bằng chứng: B +- Nguồn: You J, Liu J, Zhou M (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly 2(2):32: 「In 2018, there were 11,523 deaths caused by carbon monoxide poisoning reported in China」 — năm 2018, Trung Quốc báo cáo 11.523 ca tử vong vì ngộ độc carbon monoxide. ；Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Ngộ độc carbon monoxide: 「More than 400 Americans die from unintentional CO poisoning not linked to fires」「More than 100,000 visit an emergency department」「More than 14,000 are hospitalized」 — hơn 400 người Mỹ chết vì ngộ độc CO ngoài ý muốn không liên quan đến hỏa hoạn; hơn 100.000 người đến phòng cấp cứu; hơn 14.000 người nhập viện, các triệu chứng thường gặp「headache, dizziness, weakness, upset stomach, vomiting, chest pain, and confusion」 — đau đầu, chóng mặt, yếu, rối loạn tiêu hoá, nôn, đau ngực và lú lẫn. ；Ủy ban An toàn Sản phẩm Tiêu dùng Hoa Kỳ. Trung tâm Thông tin Carbon Monoxide: 「If you suspect CO poisoning, get outside to fresh air immediately, and then call 911」 — nếu nghi ngờ ngộ độc CO, hãy ra ngoài hít không khí tươi ngay, rồi mới gọi 911. +- Ghi chú: Triệu chứng giống cảm lạnh, lại giống say rượu. Cả nhà cùng khó chịu, ra ngoài là khỏe, đó là dấu hiệu rõ nhất. Đừng ở lại trong nhà đi tìm nguyên nhân, đừng quay lại đóng van, đưa người ta ra ngoài rồi mới gọi. Lắp báo động thế nào, xem mục 3 ở phần 1. +### 20. Uống nhầm chất tẩy rửa, thuốc trừ sâu hoặc thuốc: đừng gây nôn, mang theo vỏ chai đi viện ngay; dính vào mắt hoặc da thì xải bằng nhiều nước sạch trong 15 phút + +- Chi phí: Không tốn tiền. Để chất tẩy rửa, thuốc sát trùng, thuốc trừ sâu và thuốc ở chỗ cao trẻ không với tới, hoặc trong tủ có khoá. Tuyệt đối không rót sang chai đồ uống. +- Nói thẳng: Uống nhầm chất tẩy rửa, thuốc trừ sâu hoặc thuốc thì đừng gây nôn, nôn ra thì chất lỏng ăn mòn sẽ cháy lại thực quản một lần nữa. Trước hết cho uống một lượng sữa hoặc lòng trắng trứng vừa phải, mang theo vỏ chai và tờ hướng dẫn đi bệnh viện ngay. Dính vào da hoặc mắt thì xải bằng nhiều nước sạnh đủ 15 phút, mắt thì phải kéo mí ra mà xải, đừng xoa. Dùng thuốc sát trùng chung với thuốc tẩy bồn cầu sẽ sinh khí clo. +- Lợi ích: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (中国疾病预防控制中心职业卫生与中毒控制所) đưa ra bốn cách xử lý. Nuốt vào miệng (hấp thu qua đường tiêu hoá): cho uống ngay một lượng sữa hoặc lòng trắng trứng vừa phải,「切记，不要催吐」 — nhớ kỹ, đừng gây nôn. Đi viện càng nhanh càng tốt, mang theo vỏ bao bì hoặc tờ hướng dẫn,「以便于医生快速判断成分并制定救治方案」 — để bác sĩ nhanh chóng xác định thành phần và lập phương án điều trị. Hít vào (hấp thu qua đường hô hấp): chuyển ngay ra nơi thông thoáng, nới lỏng cổ áo để đường hô hấp thông; nếu ho dữ dội, thở gấp hoặc khó thở thì đi viện ngay. Tiếp xúc qua da: xải ngay vùng bị tiếp xúc bằng nhiều nước sạch trong ít nhất 15 phút, chú ý không để nước xải bắn vào mắt hoặc phần da khác còn nguyên vẹn. Tiếp xúc vào mắt: xải ngay mắt bằng nhiều nước sạch hoặc nước muối sinh lý trong ít nhất 15 phút. Khi xải phải kéo mí mắt ra, để nước tiếp xúc đủ với nhãn cầu và mặt trong mi mắt,「切勿揉搓眼睛」 — tuyệt đối không xoa mắt. Xải xong đi khoa mắt ngay. Sản phẩm nguy hiểm thì nhận theo độ pH. pH<3 là thuốc tẩy bồn cầu, thuốc tẩy cặn, thuốc lau sàn, thuốc đánh bóng thép không gỉ. Một số thuốc thông ống nước siêu mạnh còn chứa axit sunfuric loãng. pH>11 là thuốc thông ống nước, thuốc tẩy bếp siêu mạnh, bột rửa bát, muối thông ống, bột tẩy quần áo. Còn một cặp phối hợp gây họa ngay tại chỗ: dùng thuốc sát trùng chung với thuốc tẩy bồn cầu chứa axit clohydric sẽ sinh khí clo, kích ứng đường hô hấp, gây ho dữ dội và thở gấp, thậm chí đe doạ tính mạng +- Mức bằng chứng: B +- Nguồn: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (2025). Nhắc nhở an toàn kỳ nghỉ hè: cảnh giác chất tẩy rửa axit mạnh và bazơ mạnh. +- Ghi chú: Để ở mức B là vì tài liệu phổ biến chính thức chỉ nói cách xử lý và cách nhận ra, không có con số「làm theo thì giảm được bao nhiêu vụ」. Lý do không gây nôn là chất lỏng ăn mòn nôn ra sẽ cháy lại thực quản, cái ống từ cổ xuống dạ dày, một lần nữa. Không chắc mình uống nhầm thứ gì thì mang vỏ chai, tờ hướng dẫn cùng chất nôn đi bệnh viện. Phòng hơn cấp cứu nhiều: cất theo nhóm, không để lẫn với thức ăn đồ uống, dùng xong vặn kín nắp rồi vứt vỏ chai, tuyệt đối không đổ vào chai nước khoáng. Khi dùng thì đeo găng cao su và kính bảo hộ, mở cửa sổ thông gió, không trộn các loại với nhau. Về carbon monoxide xem mục 19 của phần này, về bỏng xem mục 14 của phần này. + +### 21. Axit, bazơ hay hóa chất khác văng lên người: cởi ngay quần áo bị nhiễm, xải bằng nhiều nước sạch chảy, mắt thì kéo mí ra mà xải, xải đủ thời gian rồi hãy đi + +- Chi phí: Không tốn tiền. Ngày đầu đi làm hãy xem vị trí thiết bị rửa mắt và vòi xải khẩn cấp, đừng đợi xảy ra chuyện rồi mới đi tìm. +- Nói thẳng: Ba bước. Bước một, cởi ngay quần áo, giày, tất dính chất; chỗ quần áo úp sát vào người thì tổn thương nặng nhất. Bước hai, xải bằng nhiều nước sạch chảy, da và mắt đều ít nhất 15 phút, Sở Quản lý Khẩn cấp tỉnh Hunan đưa ra 20 đến 30 phút, lấy con số dài hơn. Bước ba, mắt thì kéo mí ra mà xải, xải xong đi thẳng khoa mắt. Dùng nước ở nhiệt độ phòng, không dùng nước nóng. +- Lợi ích: Ba bước tại hiện trường. Bước một, cởi ngay quần áo, giày, tất bị nhiễm: chất độc vẫn tiếp tục hấp thu qua da, chỗ bị quần áo úp sát thì tổn thương nặng nhất. Bước hai, xải bằng nhiều nước sạch chảy. Mức sàn mà Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc đưa ra là ít nhất 15 phút cho cả da và mắt. Sở Quản lý Khẩn cấp tỉnh Hunan đưa ra 20 đến 30 phút để xải vết thương,「强烈的化学品要更长」 — hoá chất ăn mòn mạnh cần lâu hơn. Lấy con số dài hơn trong hai mức đó. Bước ba, với mắt thì phải「撑开眼睑，确保水流充分接触眼球及眼睑内侧」 — kéo mí mắt ra, đảm bảo nước tiếp xúc đủ với nhãn cầu và mặt trong mi mắt — và「切勿揉搓眼睛」 — tuyệt đối không xoa mắt — rồi đi khoa mắt ngay. Dùng nước ở nhiệt độ phòng hoặc ấm nhẹ, cấm nước nóng. Vết thương mới không bôi dầu mỡ, thuốc đỏ hay bất cứ thứ gì, cũng không bọc bằng vải bẩn, vì thứ bôi lên sẽ làm bác sĩ khó đánh giá độ sâu. Về mặt pháp lý,「化学性皮肤灼伤」 và「化学性眼部灼伤」 — bỏng hoá chất ở da và ở mắt — đều có trong danh mục bệnh nghề nghiệp theo luật, trường hợp xảy ra trong công việc thì áp dụng bệnh nghề nghiệp và tai nạn lao động. Khi xảy ra tai nạn cấp tính do yếu tố nguy hại nghề nghiệp, người sử dụng lao động「应当及时组织救治、进行健康检查和医学观察，所需费用由用人单位承担」 — có nghĩa tổ chức cứu chữa kịp thời, làm kiểm tra sức khoẻ và quan sát y tế, các khoản chi phí do người sử dụng lao động chịu (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (2025). Nhắc nhở an toàn kỳ nghỉ hè: cảnh giác chất tẩy rửa axit mạnh và bazơ mạnh. ；Sở Quản lý Khẩn cấp tỉnh Hunan (湖南省应急管理厅) (2015). Xử lý và điều trị khẩn cấp các tai nạn ngộ độc hoá chất thường gặp. ；Ủy ban Sức khoẻ Quốc gia cùng ba bộ khác (2024). Phân loại và danh mục bệnh nghề nghiệp (có hiệu lực từ ngày 1 tháng 8 năm 2025). ；Ủy ban Thường vụ Quốc hội (2018). Luật Phòng chống và kiểm soát bệnh nghề nghiệp (Điều ba mươi bảy). (Chính phủ thành phố Bắc Kinh đăng lại) +- Ghi chú: Để ở mức B là vì chuyện xải bao lâu thì hai tài liệu chính thức khác nhau: một viết 15 phút, một viết 20 đến 30 phút, không có một con số chuẩn quốc gia thống nhất. Đi bệnh viện thì mang theo tên hoá chất, vỏ bao bì, hoặc bảng dữ liệu an toàn (SDS), bác sĩ cần nó để lập phương án. Đừng ném đại một loại axit hay bazơ khác vào để「中和」 — trung hoà: hai thứ trộn vào sinh nhiệt, lượng cho vào không kiểm soát được. «Dung dịch trung hoà» trong tài liệu của Hunan nói đến loại nước xải chuyên dụng mà đơn vị chuẩn bị sẵn, không phải lấy thứ có trong tay ở hiện trường ra dội vào nhau. Bột khô như vôi bột, xi măng bột thì quét sạch trước rồi mới xải nước (có nơi trang ứng cứu cứu nói cả axit sunfuric đậm đặc cũng phải lau khô trước, mỗi nơi nói một kiểu, cuốn sách này không viết). Về chất tẩy rửa trong nhà và uống nhầm xem mục 20 của phần này. Về phòng vệ trước khi đi làm và khám sức khoẻ xem mục 10 và 11 ở phần 19 (tờ thông báo nguy hại trước khi đi làm, và thiết bị bảo hộ đơn vị phải cấp). + +### 22. Chóng mặt, buồn nôn, không ra mồ hôi hoặc không tỉnh khi trời nóng: chuyển ngay ra chỗ mát, cởi áo quần xả nước để hạ nhiệt; người không tỉnh thì không cho uống nước, gọi 120 + +- Chi phí: Không tốn tiền. +- Nói thẳng: Da nóng đỏ, nói không rõ, lú lẫn là kiệt sức vì nóng nặng. Từ chóng mặt tới hôn mê có thể chỉ mất mấy chục phút, tỷ lệ tử vong cực cao. Chuyển ngay ra nơi mát và thông thoáng, cởi áo quần, hạ nhiệt bằng nước mát và khăn ướt, đồng thời gọi 120. Người không tỉnh thì đừng cho một ngụm nước nào, sẽ bị sặc. +- Lợi ích: US National Weather Service: kiệt sức vì nóng nặng là「严重的医疗急症」 — một cấp cứu y khoa nghiêm trọng. Biểu hiện là thân nhiệt trên 103°F (39.4°C), lú lẫn, nói không rõ, da nóng và đỏ. Cách xử lý là gọi cấp cứu ngay, chuyển người đó ra chỗ mát, hạ nhiệt bằng nước mát và khăn ướt,「不要给液体」 — không cho uống bất cứ thứ gì. Kiệt sức vì nóng biểu hiện ở mồ hôi đầm, mệt rã rời, da ẩm lạnh, mạch nhanh và yếu; xử lý là chuyển ra chỗ mát, nới lỏng quần áo, cho uống từng ngụm nhỏ; nôn, bệnh nặng thêm, hoặc quá 1 giờ không giảm thì đi viện. Cục Kiểm soát bệnh tỉnh Vân Nam khuyên: chuyển ngay ra nơi mát và thông thoáng, hạ nhiệt bằng khăn lạnh ướt, người còn tỉnh thì cho uống đồ uống có chút muối, triệu chứng nặng thì gọi số cấp cứu ngay; kiệt sức vì nóng nặng「病死率极高」 — tỷ lệ tử vong cực cao. Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc khuyên: hạ nhiệt trước, vận chuyển sau, hạ nhiệt trong「黄金半小时」 — nửa giờ vàng — sau khi người bệnh mất ý thức. Nếu có điều kiện, người trẻ và trung niên khoẻ mạnh không có bệnh nền thì hạ nhiệt nhanh bằng cách ngâm nước lạnh; người lớn tuổi và người có bệnh nền thì xịt nước lên da để hạ nhiệt bằng bay hơi. Hướng dẫn bệnh nhiệt của Hiệp hội Y khoa Hoang dã, cập nhật năm 2024, nói ngâm nước lạnh là cách điều trị quan trọng nhất với kiệt sức vì nóng nặng và hạ nhiệt nhanh nhất. Không có đá thì dùng loại nước mát nhất trong tay; không ngâm được thì xịt nước và quạt, tốc độ chậm khoảng một nửa so với ngâm +- Mức bằng chứng: B +- Nguồn: US National Weather Service. Heat Related Illnesses. ; Cục Kiểm soát bệnh tỉnh Vân Nam (云南省疾控局) (2026). Có mấy triệu chứng sau thì bạn có thể đang say nắng, 出现这些症状，你可能是中暑了. ; Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc. Thời tiết nóng, Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc nhắc phòng say nắng: 「如有条件可用冷水浸泡的方法对无基础疾病的青壮年快速降温。老人和有基础疾病的患者可向皮肤喷水通过蒸发降温。降温应遵循"先降温，后转运" 的原则」 — nếu có điều kiện, dùng cách ngâm nước lạnh để hạ nhiệt nhanh cho người trẻ và trung niên khoẻ mạnh không có bệnh nền; người lớn tuổi và người có bệnh nền thì xịt nước lên da để hạ nhiệt bằng bay hơi; hạ nhiệt phải theo nguyên tắc "hạ nhiệt trước, vận chuyển sau". ; Ủy ban Sức khoẻ Quốc gia (2010). Những điểm chính về phòng say nắng khi nóng. ; Eifling KP và cộng sự (2024). Wilderness Medical Society Clinical Practice Guidelines for the Prevention and Treatment of Heat Illness: 2024 Update. Wilderness & Environmental Medicine 35(1_suppl):112S-127S. (trang nội dung từ chối cho công cụ tự động tải về, nên khuyến cáo cụ thể được đối chiếu theo bản tóm tắt cập nhật trên trang của Hiệp hội Y khoa Hoang dã: 「cold water immersion … is the most important method of treating severe heat illness」「If CWI is unavailable, evaporative cooling is the recommended second-line treatment. … generally half as fast as CWI」 — ngâm nước lạnh là cách quan trọng nhất để điều trị kiệt sức vì nóng nặng; nếu không ngâm được, hạ nhiệt bằng bay hơi là cách tuyến hai được khuyến cáo, chậm hơn khoảng một nửa)；Lipman GS và cộng sự (2019). Cùng bộ hướng dẫn đó, bản 2019. +- Ghi chú: Đừng đợi xem「缓一缓」 — dịu đi một chút — rồi mới làm: kiệt sức vì nóng nặng có thể đi từ chóng mặt tới hôn mê trong mấy chục phút. Không đổ nước vào miệng người không tỉnh, sẽ bị sặc. Chỗ đặt túi đá thì hai nguồn nói khác nhau. Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc nói đặt ở cổ, nách và bẹn; Hiệp hội Y khoa Hoang dã nói các nghiên cứu không cho thấy đặt ở mấy chỗ đó có tác dụng hạ nhiệt. Khuyến cáo của họ là: nếu chỉ có túi đá thì phủ kín cả người, túi đá không đủ thì đặt ở má, lòng bàn tay và lòng bàn chân. Cả hai đều nói nếu có điều kiện thì dùng ngâm nước lạnh. + +### 23. Người bị say nắng mà còn tỉnh: uống nhiều lần mỗi lần ít, nước muối loãng mát hoặc nước thể thao; đừng uống hùng hục, đừng uống rượu và đồ uống nhiều đường + +- Chi phí: Không tốn tiền. Một chai nước thể thao vài tệ. +- Nói thẳng: Chỉ người còn tỉnh mới uống. Nước muối loãng mát hoặc nước thể thao, uống nhiều lần mỗi lần ít, mỗi giờ 500 đến 1000 mililit. Đừng uống hùng hục, nôn ra rồi hút vào khí quản còn nguy hiểm hơn. Không uống rượu và đồ uống nhiều đường, chúng khiến cơ thể mất thêm nhiều nước. Trước hết lau người bằng nước mát, đặt túi đá ở cổ, nách và bẹn để hạ nhiệt, xong mới bù nước. Người không tỉnh, nói không rõ, không cho một ngụm nào. +- Lợi ích: Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc đã ban hành Hướng dẫn phòng ngừa và xử lý khẩn cấp say nắng cho công chúng (bản 2014). Trong đó viết「至少每小时喝 2~4 杯凉水（500~1000ml），水温不宜过高，饮水应少量多次」 — mỗi giờ uống ít nhất 2~4 cốc nước mát (500~1000ml), nhiệt độ nước không nên quá cao, phải uống nhiều lần mỗi lần ít. Cũng viết「不要饮用含酒精或大量糖分的饮料。这些饮料会导致失去更多的体液。同时，还应避免饮用过凉的冰冻饮料」 — không uống đồ uống chứa cồn hoặc nhiều đường; các đồ uống đó làm cơ thể mất nhiều dịch hơn, đồng thời nên tránh uống đồ uống đông lạnh quá lạnh. Với kiệt sức vì nóng nặng thì ghi rõ「不要给病人喝水」 — không cho bệnh nhân uống nước. US NIOSH với kiệt sức vì nóng thì「鼓励小口频繁地喝凉水」 — nên uống nước mát từng ngụm nhỏ và thường xuyên; với chuột rút vì nóng thì mỗi 15 đến 20 phút bù một lần bằng nước thể thao có đường và chất điện giải,「不要用盐片」 — không dùng viên muối +- Mức bằng chứng: B +- Nguồn: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (2014). Hướng dẫn phòng ngừa và xử lý khẩn cấp say nắng cho công chúng (bản 2014). ; US CDC/NIOSH. Heat-Related Illnesses. +- Ghi chú: Sau khi mồ hôi đầm mà chỉ uống nước trắng thì muối trong máu bị loãng (hạ natri máu), nên nước phải có chút muối. Uống hùng hục một lúc dễ nôn, thứ nôn hút vào khí quản chết nhanh hơn cả say nắng. 「别喝冰镇的」 — đừng uống đá lạnh — là khuyến cáo trong bản hướng dẫn năm 2014 của Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc, lý do là sợ co thắt dạ dày, không có bằng chứng định lượng. Đừng hiểu nhầm thành「không được dùng nước lạnh」. xem mục 22 của phần này (khi chóng mặt buồn nôn dưới trời nóng thì cởi áo quần xả nước hạ nhiệt) nói về mấy cách hạ nhiệt. Đó là lau bằng nước mát, ngâm nước lạnh, và đặt túi đá ở cổ, nách, bẹn. Chính mấy cách này là hàng đầu trong những cách hạ nhiệt mà nguồn chính thức khuyến cáo, và chúng đặt lên người chứ không rót vào miệng. Hạ nhiệt luôn đứng trước bù nước. Người không tỉnh, nói không rõ, không cho một ngụm nào. +### 24. Khi hoảng hoạt: bò sát đất, sờ cửa rồi mới mở, cửa nóng thì đừng mở, đi cầu thang chứ đừng đi thang máy, ra ngoài rồi đừng quay lại + +- Chi phí: Lắp trong nhà 1 đến 2 báo động khói độc lập, vài chục tệ. Xem trước hai lối ra của hành lang. +- Nói thẳng: Nằm sát xuống đất, bò về phía lối ra từ dưới lớp khói. Trước khi mở cửa, sờ tay nắm và cánh cửa, nóng hoặc khe cửa đang khét khói thì đừng mở cửa đó, đi đường khác. Mở cửa chậm lại, thấy khói đặc thì đóng ngay. Đi cầu thang chứ đừng đi thang máy, mất điện là mắc kẹt trong đó. Ra ngoài rồi đừng quay lại để lấy đồ. Áo quần cháy thì dừng lại, nằm sấp xuống, lăn đi, dùng tay che mặt. +- Lợi ích: Nguyên văn của FEMA Hoa Kỳ về thoát hiểm:「趴到地上，从烟下面爬向出口」 — nằm sát xuống đất, bò về phía lối ra từ dưới lớp khói. Trước khi mở cửa, sờ tay nắm và cánh cửa,「如果是热的，或者门缝有烟，别开这扇门，用第二条路」 — nếu nóng, hoặc khe cửa có khói, thì đừng mở cửa đó, hãy dùng đường thứ hai. Mở cửa chậm lại, thấy khói đặc thì đóng ngay. Áo quần cháy thì「停下、倒地、打滚」 — dừng, nằm sấp xuống, lăn — và dùng tay che mặt. Không ra được thì đóng cửa, dùng vải hoặc băng keo bịt khe cửa và lỗ thông gió, rồi gọi điện thoại báo vị trí. Nhắc nhở tự cứu trong hoảng hoạn của Bộ Quản lý Khẩn cấp cũng nhất quán: khi thoát hiểm không đi thang máy, lửa chưa lớn thì cố chạy xuống dưới, dùng khăn ướt hoặc khẩu trang che mũi miệng, bò sấp hoặc cúi người rút ra +- Mức bằng chứng: B +- Nguồn: US FEMA Ready.gov. Home Fires. ；Bộ Quản lý Khẩn cấp (应急管理部) (2019). Cách tự cứu khi hoảng hoạt ở nơi công cộng: 「逃生时不要乘坐电梯，当火势不大时，要尽量往楼层下面跑」「可用湿毛巾或口罩蒙鼻，匍匐或弯腰撤离」 — khi thoát hiểm không được đi thang máy, lửa chưa lớn thì cố chạy xuống dưới các tầng; có thể dùng khăn ướt hoặc khẩu trang che mũi, bò sấp hoặc cúi người rút ra. ；Sư đoàn Cứu hoả và Cứu hộ tỉnh Quý Châu (2020). Kiến thức cứu hoả, ba điểm chính khi thoát hiểm: 「湿毛巾对折3次（叠成8层），除烟率比较高，但烟和毒气浓度过高时，过滤效果就无法保证了。另外，毛巾过湿会导致呼吸困难」 — khăn ướt gấp đôi 3 lần (thành 8 lớp) thì lọc khói khá tốt, nhưng khi nồng độ khói và khí độc quá cao thì không còn bảo đảm tác dụng lọc; thêm nữa, khăn quá ướt sẽ gây khó thở. +- Ghi chú: Đừng đi thang máy, mất điện là mắc kẹt trong đó. Đừng quay lại lấy tài sản. Đừng chạy lên trên, trừ khi đã xác nhận cửa trên nóc mở. Đừng nhảy lầu. Khăn ướt phải gấp đôi 3 lần thành 8 lớp, đừng tẩm quá nhiều nước, quá ướt thì ngược lại khó thở. Khi khói và khí độc quá đặc thì khăn cũng chắn không được. + +### 25. Thấy người đang đuối nước: gọi người và báo 110/120, đưa sào hoặc thả vật nổi, tuyệt đối không xuống nước + +- Chi phí: Không tốn tiền. Khó là nhịn cơn thèm nhảy xuống. +- Nói thẳng: Người đang đuối nước sẽ theo bản năng ấn người cứu xuống dưới nước, và người xuống nước cũng có thể chết đuối. Thứ tự là: hô to gọi người → báo 110/120 → đưa sào dài, dây → thả phao cứu sinh, xô rỗng, chai nước khoáng → người lên bờ mà không thở thì ấn ngực ngay. Không nắm tay thành dây chuyền rồi xuống nước. Mỗi năm trên thế giới có khoảng 30 vạn người chết đuối, đây là nguyên nhân tử vong thứ tư ở trẻ từ 1 đến 4 tuổi. +- Lợi ích: WHO: mỗi năm trên thế giới có khoảng 30 vạn người chết đuối, 92% ở các nước thu nhập trung bình và thấp. Đuối nước là nguyên nhân tử vong thứ tư ở trẻ từ 1 đến 4 tuổi, thứ ba ở trẻ từ 5 đến 14 tuổi. Can thiệp WHO khuyến cáo là「安全救援和复苏培训」 — đào tạo cứu hộ an toàn và hồi sinh, chẳng hạn là học trước cách cứu mà không xuống nước. Thứ tự là: hô to gọi người → gọi điện thoại → đưa sào dài, dây, thả phao cứu sinh, xô rỗng, chai nước khoáng → người lên bờ sau mà không thở thì ấn ngực ngay +- Mức bằng chứng: C +- Nguồn: WHO. Đuối nước, thông tin thực trạng. +- Ghi chú: Lý do để ở mức C là quy tắc「không xuống nước」 không tìm được nguyên văn chính thức, chỉ có số tử vong của WHO và hướng「cứu hộ an toàn」. Đừng nắm tay thành dây chuyền rồi xuống nước. Chưa qua huấn luyện thì đừng đi kéo người, người đuối nước sẽ ấn người cứu xuống dưới. 「Không xuống nước」 cũng là quy tắc thuần vì lợi ích cho bản thân: người xuống nước tự chết đuối, còn người ở trong nước thì ai cũng thế. Người bạn thật sự cần cứu phần lớn là con bạn, cách phòng nằm ở xem mục 12 ở phần 1 (trẻ ở gần nước không được rời mắt). + +### 26. Người bị nghẹn không nói nổi: đứng sau lưng, vỗ lưng 5 lần rồi đẩy bụng 5 lần, ngã xuống thì làm hồi sinh tim phổi + +- Chi phí: Không tốn tiền. Xem lại động tác một lần: một tay nắm lại đặt trên rốn, tay kia bọc lấy, đẩy nhanh vào trong và lên trên. +- Nói thẳng: Trước hết hỏi「bạn bị nghẹn phải không, nói được không」. Người còn ho mạnh, còn ra tiếng thì đừng can thiệp, cứ để họ ho. Người không nói nổi thì đứng sau lưng, vỗ lưng 5 lần rồi đẩy bụng 5 lần, lặp lại cho tới khi vật ra ngoài hoặc người đó ngã xuống. Ngã xuống thì làm hồi sinh tim phổi, đồng thời gọi người báo số cấp cứu. Phụ nữ có thai và người quá béo thì đẩy vào ngực. Đừng thò tay vào miệng móc thứ không nhìn thấy. +- Lợi ích: Nguyên văn của Thư viện Y khoa Quốc gia Hoa Kỳ: trước hết hỏi「bạn bị nghẹn phải không, nói được không」.「能用力咳嗽、能说话的不要动手」 — người còn ho mạnh, còn nói được thì đừng can thiệp, bản thân cơn ho đã phần lớn tự đẩy vật ra. Người không nói nổi thì「反复做 5 次拍背接 5 次腹部冲击，直到异物出来或人失去意识」 — lặp lại vỗ lưng 5 lần rồi đẩy bụng 5 lần cho tới khi vật ra ngoài hoặc người mất ý thức.「失去意识就开始心肺复苏」 — mất ý thức thì bắt đầu hồi sinh tim phổi — và gọi người báo số cấp cứu cùng lúc. Phụ nữ có thai và người béo thì làm đẩy vào ngực +- Mức bằng chứng: B +- Nguồn: US National Library of Medicine, MedlinePlus. Choking – adult or child over 1 year. ；Ủy ban Sức khoẻ thành phố Phúc Châu (福州市卫生健康委员会) (2026). Phương pháp Heimlich đã cập nhật: 「1岁以上幼儿、儿童及成人：从旧版单一腹部冲击改为"5次背部拍击+5次腹部冲击"的循环」「中国红十字系统急救培训一直都是按"5次背部拍击+5次腹部冲击"循环进行的方式进行教学」 — trẻ nhỏ từ 1 tuổi trở lên, trẻ em và người lớn: chuyển từ cách chỉ đẩy bụng của bản cũ sang vòng lặp "5 lần vỗ lưng + 5 lần đẩy bụng"; đào tạo cấp cứu của hệ thống Hội Chữ thập đỏ Trung Quốc vẫn luôn dạy theo vòng lặp "5 lần vỗ lưng + 5 lần đẩy bụng". +- Ghi chú: Đừng thò tay vào miệng móc thứ không nhìn thấy, sẽ đẩy sâu hơn. Đừng vỗ vào người còn ho được. Nghẹn thường xảy ra nhất ở bàn ăn nhà bạn, nên người được lợi chủ yếu là người trong nhà. Cách dạy ở Trung Quốc cũng là bộ này: Trung tâm Đào tạo Cứu hộ Khẩn cấp của Hội Chữ thập đỏ tỉnh Phúc Kiến nói rằng với trẻ trên 1 tuổi và người lớn, hệ thống Hội Chữ thập đỏ Trung Quốc vẫn dạy theo vòng lặp vỗ lưng 5 lần rồi đẩy bụng 5 lần. Nhiều hướng dẫn cũ trên mạng chỉ dạy đẩy bụng, đó là bản cũ. + +### 27. Lạc đường hoặc xe hỏng ở sa mạc Gobi, sa mạc, vùng không người ở: ở lại tại chỗ hoặc cạnh xe, báo vị trí trước, rồi tìm bóng mát và giữ ấm, tiết kiệm nước, đừng đi lang thang + +- Chi phí: Trước khi vào vùng không người ở, báo cho ai đó biết tuyến đường, điểm đến, ngày quay lại, loại xe và biển số. Mỗi người mỗi ngày bắt đầu từ 4 lít nước. Máy liên lạc vệ tinh hay thiết bị tin nhắn ngắn BeiDou giá hơn nghìn tệ, thuê một lần vài trăm tệ. Khó là giữ được nguyên tắc「ở lại tại chỗ, đừng đi lang thang」. +- Nói thẳng: Ở lại tại chỗ hoặc cạnh xe, vì xe là thứ che nắng lớn nhất. Thứ tự là: có tín hiệu thì gửi vị trí trước rồi mới gọi điện thoại → tìm bóng mát và giữ ấm → tiết kiệm nước, ban ngày không di chuyển, đêm mát rồi hãy đi → đừng đi lang thang → phát tín hiệu bằng gương, khói, ba đống lửa. Trước khi vào, báo cho ai đó biết tuyến đường, điểm đến, ngày quay lại và loại xe, biển số. +- Lợi ích: Trang sa mạc Mojave của Cục Công viên Quốc gia Hoa Kỳ có ba câu nguyên văn. Một câu là「每人每天至少 1 加仑（约 3.8 升）水，徒步和骑行者每人每天 2 加仑」 — mỗi người mỗi ngày ít nhất 1 gallon (khoảng 3.8 lít) nước, người đi bộ và đi xe đạp thì mỗi người mỗi ngày 2 gallon. Một câu là「告诉别人你的路线、目的地、返回日期、车型和车牌」 — báo cho ai đó biết tuyến đường, điểm đến, ngày quay lại, loại xe và biển số. Còn một câu là「手机信号时有时无」 — tín hiệu điện thoại lúc có lúc không. Nguyên văn ở Thung lũng Chết:「热的时候不要在谷底和低海拔徒步」 — khi trời nóng thì đừng đi bộ ở đáy thung lũng và vùng thấp, vùng hoang dã「几乎没有手机信号」 — gần như không có tín hiệu điện thoại, và khuyên mang theo điện thoại vệ tinh. Nguyên văn ở trang lái xe trong sa mạc của Mojave:「出了紧急情况，待在车旁是你能做的最重要的事」 — khi gặp khẩn cấp, ở lại cạnh xe là việc quan trọng nhất bạn làm được. Trang đó còn nói việc này không phải lúc nào cũng xảy ra, nhưng quả là có người đã chết vì phơi nắng nóng hoặc rét trên đường quay lại tuyến đường cái. Thứ tự ưu tiên chia làm năm bước. Bước một là báo an và toạ độ, có tín hiệu thì gửi vị trí trước rồi mới gọi điện thoại. Bước hai là tìm bóng mát và giữ ấm, xe là vật che chắn lớn nhất. Bước ba là nước, ban ngày không di chuyển, đêm mát rồi hãy đi. Bước bốn là đừng đi lang thang. Bước năm là phát tín hiệu. Nhắc nhở của Cục Quản lý Khẩn cấp thành phố Trùng Khánh: ba đống lửa xếp thành hình tam giác là tín hiệu cầu cứu được thế giới công nhận. Ở gần thì hét「ba tiếng ngắn, ba tiếng dài, rồi ba tiếng ngắn」, cách một phút thì lặp lại. Có nắng thì dùng gương, nắp hộp thực phẩm, những thứ bóng, để phản xạ. Bộ Văn hoá và Du lịch nhắc khách tự lái xe cẩn thận khi tới vùng không người ở, đi nhiều xe càng tốt, cần thì thuê điện thoại vệ tinh +- Mức bằng chứng: C +- Nguồn: US National Park Service. Mojave National Preserve Safety. ; US National Park Service. Death Valley Safety. ; US National Park Service. Desert Driving Safety. ; Cục Quản lý Khẩn cấp thành phố Trùng Khánh (重庆市应急管理局) (2024). Khoa học khẩn cấp, những tín hiệu cầu cứu bạn cần biết khi ở ngoài trời (bản tranh dài): 「燃放三堆火焰是国际通行的求救信号，将火堆摆成三角形」「三声短三声长，再三声短；间隔1分钟之后再重复」 — đốt ba đống lửa là tín hiệu cầu cứu quốc tế, xếp ba đống lửa thành hình tam giác; ba tiếng ngắn ba tiếng dài rồi ba tiếng ngắn, cách 1 phút thì lặp lại. ; Bộ Văn hoá và Du lịch (文化和旅游部) (2018). Bộ Văn hoá và Du lịch nhắc du khách cân nhắc khi chọn du lịch khám phá ngoài trời: 「谨慎前往无人区和人烟稀少区」「尽量多车同行，提前准备对讲机，必要时租用卫星电话」 — cẩn thận khi tới vùng không người ở và vùng ít dân cư; nên đi nhiều xe cùng nhau, chuẩn bị bộ đàm trước, cần thì thuê điện thoại vệ tinh. +- Ghi chú: Để ở mức C là vì tất cả những thứ này là nhắc nhở chính thức, không phải số liệu nghiên cứu. Lượng nước uống, việc báo trước, vùng phủ tín hiệu, ở lại cạnh xe và tín hiệu cầu cứu đều là nguyên văn chính thức. Câu「ba tiếng một nhóm」 được truyền rộng không tìm được nguyên văn chính thức, nên đã viết lại thành ba đống lửa và ba tiếng ngắn, ba tiếng dài, ba tiếng ngắn của phần chính thức. Đừng đi bộ tìm đường vào buổi trưa. Đừng uống nước tiểu. Đừng vì tiết kiệm nước mà không uống. + +### 28. Run, nói không rõ, đi lảo đảo là thân nhiệt tụt: thay quần áo ướt, quấn lại, làm ấm thân trước, uống đồ uống nóng có đường, không uống rượu + +- Chi phí: Một tấm chăn cách nhiệt nhôm vài tệ, cộng thêm một bộ quần áo khô để dự phòng. Biết nhận ra dấu hiệu thì không tốn tiền. +- Nói thẳng: Thân nhiệt tụt dưới 35℃ là thân nhiệt tụt: người đó run, nói không rõ, đi lảo đảo, không suy nghĩ rõ. Thay quần áo ướt, dùng chăn cách nhiệt hoặc quần áo khô quấn lại, làm ấm thân trước chứ đừng làm ấm tay chân trước. Cho đồ uống nóng có đường, không cho rượu, rượu dập cơn run vì lạnh. Hết run không phải là khỏe hơn, mà là nặng thêm. Không còn ý thức thì làm hồi sinh tim phổi. +- Lợi ích: Nguyên văn của US National Weather Service: thân nhiệt dưới 95°F (35°C) là thân nhiệt tụt,「觉得冷、迟钝、想不清楚，立刻看医生」 — thấy lạnh, lơ chậm, không suy nghĩ rõ thì đi khám ngay. Cách xử lý là「立即就医，把人移到有暖气的室内，先从身体中心开始复温」 — đi viện ngay, chuyển người đó vào trong nhà có hệ thống sưởi, bắt đầu làm ấm lại từ trung tâm cơ thể — và cho đồ uống nóng.「酒精会减少寒战」 — rượu làm giảm cơn run vì lạnh — nên không uống.「失去意识做心肺复苏」 — mất ý thức thì làm hồi sinh tim phổi. Phòng lạnh thì làm ấm bằng nước ấm hoặc hơi ấm của cơ thể,「不要用热水或壁炉一类辐射热」 — không dùng nước nóng hay nhiệt bức xạ kiểu lò sưởi. Hướng dẫn thân nhiệt tụt năm 2019 của Hiệp hội Y khoa Hoang dã là bộ chỉ dẫn hành nghề cho lĩnh vực này +- Mức bằng chứng: B +- Nguồn: US National Weather Service. During Extreme Cold. ; Dow J và cộng sự (2019). Wilderness Medical Society Clinical Practice Guidelines for the Out-of-Hospital Evaluation and Treatment of Accidental Hypothermia: 2019 Update. Wilderness & Environmental Medicine. +- Ghi chú: Đừng xoa tay chân. Đừng ngâm bồn nước nóng. Đừng cứng kéo hất người đó đi, người thân nhiệt tụt nặng thì tim dễ ngừng. Đừng vì「hết run」mà tưởng đã khỏe, không run (hết cơn run vì lạnh) là nặng thêm. Toàn văn hướng dẫn của Hiệp hội Y khoa Hoang dã phải đăng ký mới đọc được, lần này chỉ đối chiếu phần tóm tắt. +### 29. Bị rắn cắn: ngồi yên, tháo nhẫn và đồng hồ, đưa đi bệnh viện tiêm huyết thanh kháng nọc càng nhanh càng tốt; không rạch, không hút, không buộc chặt + +- Chi phí: Không tốn tiền. Biết trước bệnh viện nào gần đó có huyết thanh kháng nọc. Khó là nhịn cơnghè tự xử lý vết thương. +- Nói thẳng: Ngồi yên, cố định chi thể bị cắn. Tháo nhẫn và đồng hồ trước, vì chỗ đó sẽ sưng lên. Đưa ngay tới bệnh viện có huyết thanh kháng nọc, huyết thanh là thứ duy nhất đảo ngược được phần lớn tác dụng độc. Không rạch, không hút bằng miệng, không chườm đá, không dùng dây buộc tới bầm tím. Đừng uống rượu, đừng chạy. Chụp được tấm ảnh con rắn từ xa thì chụp, đừng đi bắt. +- Lợi ích: WHO có ba câu nguyên văn. Một câu là「立即完全固定被咬的身体部位，并迅速送到最近的医疗机构就医」 — cố định ngay và hoàn toàn phần cơ thể bị cắn, rồi nhanh chóng đưa tới cơ sở y tế gần nhất. Một câu là「避免使用止血带，并避免切割伤口」 — tránh dùng gar, và tránh rạch vết thương. Còn một câu là「用针对当地蛇种的适当抗蛇毒血清进行治疗」 — điều trị bằng huyết thanh kháng nọc thích hợp cho loài rắn ở địa phương.「高质量抗蛇毒血清是防止或逆转蛇咬伤的大部分毒性作用的最有效治疗方法」 — huyết thanh kháng nọc chất lượng cao là cách điều trị hiệu quả nhất để ngăn hay đảo ngược phần lớn tác dụng độc của vết cắn rắn. Mỗi năm trên thế giới có khoảng 81.000 đến 138.000 người chết vì rắn cắn, số người phải cắt chiết cùng số tàn tật vĩnh viễn khoảng gấp 3 lần số tử vong. Hướng dẫn xử lý rắn cắn của Trung Quốc năm 2024 ước tính mỗi năm trong nước có 25 vạn đến 28 vạn ca rắn độc cắn, người bị cắn phần lớn trên 50 tuổi. Các bệnh viện ở lưu vực Trường Giang và 11 tỉnh phía nam tiếp nhận 40817 ca rắn cắn trong giai đoạn 2017 đến 2021, có 23 người tử vong trong thời gian nằm viện (0.06%) +- Mức bằng chứng: B +- Nguồn: WHO. Cắn động vật, thông tin thực trạng. ; WHO (2023). Ngộ độc do rắn cắn, thông tin thực trạng. ; Lai R và cộng sự (2024). The Chinese guideline for management of snakebites. World Journal of Emergency Medicine 15(5):333-355: 「An estimated 250,000–280,000 cases of venomous snakebites occur in China each year, affecting mainly patients aged ≥50 years」 — ước tính mỗi năm ở Trung Quốc có 250.000 đến 280.000 ca rắn độc cắn, chủ yếu ở nhóm bệnh nhân từ 50 tuổi trở lên. ; Yan S và cộng sự (2026). Clinical epidemiology of snakebite envenoming in hospitals 11 provinces of Yangtze River Basin and southern regions of China. PLoS Neglected Tropical Diseases: 「withdrawal (4346, 10.65%), and death (23, 0.06%)」 — tự xuất viện (4346, 10.65%) và tử vong (23, 0.06%). +- Ghi chú: Chụp được ảnh con rắn thì chụp, đừng đi bắt. Đừng uống rượu, đừng chạy, đừng hút bằng miệng, đừng chườm đá, đừng dùng dây buộc tới bầm tím. Trung Quốc không có thống kê chính thức số ca chết vì rắn cắn trên toàn quốc. Tỷ lệ tử vong nội trú ở 11 tỉnh đó trông thấp, nhưng có hơn một phần mười bệnh nhân tự xuất viện khi chưa điều trị xong, họ ra sao sau đó thì không nằm trong thống kê. + +### 30. Khi động đất, trước hết xem mình đang ở trong loại nhà nào: nhà xây đúng tiêu chuẩn chống chấn thì nằm tại chỗ, nhà cũ có thể sập và cửa cách vài bước thì ra ngoài + +- Chi phí: Không tốn tiền. Xem trước nhà mình ra cửa mất mấy bước, đi đường nào, cái bàn nào thì chắc. +- Nói thẳng: Trước hết xem mình đang ở trong loại nhà nào. Nhà xây đúng quy chuẩn từ sau năm 2001, muốn ra ngoài phải đi xuống cầu thang, đi qua sảnh, thì nằm tại chỗ: che đầu cổ, chui xuống dưới cái bàn chắc, nắm chặt chân bàn. Nhà một tầng, nhà gạch cũ, nhà tự xây không có thiết kế chống chấn thì cửa ở ngay vài bước, ra ngoài. Mọi trường hợp đều không nhảy lầu, không đi thang máy, không đi xuống cầu thang khi nhà đang rung. +- Lợi ích: Câu tắc trong video phổ biến của Cục Địa chấn Trung Quốc《地震来了，是躲还是跑？》là「震时就近躲避，震后迅速撤离」 — lúc rung thì tránh tại chỗ, xong rung thì rút ra ngay — và video nói rõ giới hạn「这里指的是在符合抗震设防要求的建筑内」 — ở đây nói là bên trong công trình đạt yêu cầu thiết kế chống chấn. Lý do là loại công trình này「整体垮塌的可能性较小」 — ít có khả năng sụp đổ toàn bộ. Nguyên văn nói「坠落的建筑构件及装饰物品」 — cấu kiện kết cấu và đồ trang trí rơi xuống —「才是对生命安全最大的威胁」 — mới là mối đe doạ lớn nhất với tính mạng. Còn「震时无保护措施的盲目乱跑」 — mù quáng chạy loạn lúc rung mà không có biện pháp bảo vệ —「反而容易被坠落的天花板、吊灯（扇）等物品砸伤」 — lại dễ bị trần nhà, đèn chùm (quạt trần) rơi xuống đập vào. Cách tránh là「用手或者其他软物保护好头颈部」 — dùng tay hoặc vật mềm khác bảo vệ kỹ đầu và cổ —「并牢牢地抓住桌腿」 — và nắm chặt chân bàn; không có bàn thì「迅速贴紧承重墙蹲伏」 — nhanh chóng áp sát tường chịu lực rồi ngồi xuống —「避开外墙、窗户、阳台」 — tránh tường ngoài, cửa sổ, ban công. 「如果在户外」 thì「应该第一时间疏散到空旷的地方避震」 — nếu ở ngoài trời thì phải sơ tán ra khoảng trống ngay lập tức để tránh chấn. Nhìn ngược lại, động đất năm 1988 ở Armenia là các tòa nhà panel bê tông cốt thép kiểu Xô Viết sập hàng loạt. Nghiên cứu hồi sát của dịp đó viết rằng「第一下震动之后离开建筑是保护性行为」 — rời khỏi công trình ngay sau cú rung đầu tiên là hành vi bảo vệ. Khả năng bị thương của người ở trong nhà gấp 4.40 lần người chạy ra ngoài (tỷ số khả năng 4.40), 95% CI 2.24–8.71 (khoảng tin cậy). Khả năng bị thương ở trong công trình gấp 12.20 lần ở ngoài trời (3.62–63.79). Ở tòa nhà cao hơn 5 tầng, khả năng bị thương gấp 3.65 lần ở tòa nhào thấp hơn (2.12–6.33) +- Mức bằng chứng: B +- Nguồn: Cục Địa chấn Trung Quốc (do Sở Địa chấn tỉnh An Huy sản xuất). 地震来了，是躲还是跑？（video phổ biến, bản chép lời thoại phụ đề）. ；Armenian HK, Noji EK, Oganesian AP (1992). A case-control study of injuries arising from the earthquake in Armenia, 1988. Bulletin of the World Health Organization, 70(2), 251-257. ；US FEMA Ready.gov. Earthquakes. ；Mahue-Giangreco M, Mack W, Seligson H, Bourque LB (2001). Risk factors associated with moderate and serious injuries attributable to the 1994 Northridge Earthquake. Annals of Epidemiology, 11(5), 347-357. +- Ghi chú:Tranh cãi. Hai kết luận nói về hai loại nhà khác nhau. Chính Cục Địa chấn Trung Quốc cũng viết điều kiện của việc「tránh tại chỗ」 là「符合抗震设防要求的建筑内」. Nó còn yêu cầu「因地制宜」「因人而异」 — tuỳ tình huống, tuỳ người — nên phải xem cả ba việc: nhà có chịu được động đất không, bạn đang đứng ở vị trí nào, bản thân có chạy nổi không. Loạt nhà ở Armenia sập cả khối, không có chỗ để tránh, ra ngoài mới là lựa chọn hơn. FEMA Hoa Kỳ khuyên cho nhà ở trong nước họ là「就地趴下、护住头颈、抓牢，不要往外跑」, thương tích ở đó chủ yếu đến từ vật rơi và do ngã khi di chuyển trong lúc rung. Nghiên cứu Northridge còn chỉ ra rằng đưa tay ra đỡ đồ, hay vịn vào thứ「trông chắc」, lại làm nặng thêm thương tích. Vì vậy chỉ hỏi「chạy hay không」 thì vô ích, phải đánh giá「tòa nhà này có sập không, bạn có kịp ra khoảng trống trong mấy giây không」. Nhà xây đúng quy chuẩn từ sau năm 2001, cần đi xuống cầu thang hoặc đi qua sảnh thì nằm tại chỗ. Nhà một tầng, nhà gạch cũ, nhà tự xây không có thiết kế chống chấn thì cửa ở ngay vài bước, ra ngoài. Nghiên cứu ở Armenia chỉ so người sống sót bị thương với không bị thương, không tính người tử vong, cũng không nói có trừ ảnh hưởng của độ cao tầng hay không. Người chạy ra ngoài được vốn đã có thể sống ở tầng thấp, số đó phải tính vào. Mọi trường hợp đều không nhảy lầu, không đi thang máy, không đi xuống cầu thang khi rung. Ra ngoài rồi thì tránh xa nhà cao lớn, cầu vượt, cột điện, tường kính và biển quảng cáo. + +### 31. Gặp gấu, lợn rừng, bầy chó hoang: đừng chạy, đừng hét, đừng nhìn thẳng vào mắt, nghiêng người lùi ra chậm + +- Chi phí: Không tốn tiền. Vào núi thì mang theo một cây gậy leo núi, đi theo nhóm, đi có tiếng động cho con vật biết bạn tới trước. Khó là đứng yên mà không chạy. +- Nói thẳng: Đừng chạy, gấu chạy nhanh như ngựa đua cả lúc lên dốc lẫn lúc xuống dốc. Cũng đừng trèo cây, gấu đen và gấu nâu đều trèo được. Đứng yên, dùng giọng trầm nói liên tục với nó, đợi nó không động đậy rồi hãy nghiêng người lùi ra chậm. Gấu nâu vồ thì nằm sấp xuống giả vờ chết, hai tay khoanh sau gáy, hai chân dang ra. Gấu đen vồ thì không giả chết, chạy vào xe hoặc vào nhà, không thoát thì cầm thứ bên cạnh phản công. Bị chó cào cắn thì rửa và tiêm theo mục 13 của phần này. +- Lợi ích: Cục Công viên Quốc gia Hoa Kỳ có sáu câu nguyên văn về gấu. Một câu là「不要跑，熊上下坡都能跑得像赛马一样快」 — đừng chạy, gấu lên dốc xuống dốc đều chạy nhanh như ngựa đua. Một câu là「保持冷静，大多数熊不想攻击你，只想被放着不管」 — giữ bình tĩnh, phần lớn gấu không muốn tấn công bạn, chỉ muốn được để yên. Một câu là「用低沉的声音持续跟熊说话」 — dùng giọng trầm nói liên tục với gấu. Một câu là「熊不动时，侧着身慢慢退开，这样既能盯着熊又不会绊倒」 — khi gấu không động đậy thì nghiêng người lùi ra chậm, vừa nhìn chừng được con gấu vừa không bị vấp. Một câu là「让自己显得尽量大」 — hãy làm cho mình trông càng to càng tốt. Còn một câu là「不要爬树，灰熊和黑熊都会爬树」 — đừng trèo cây, gấu xám và gấu đen đều trèo cây. Bị gấu nâu tấn công thì「装死，趴着双手扣在颈后、双腿分开」 — giả chết, nằm sấp xuống, hai tay khoanh sau gáy, hai chân dang ra. Bị gấu đen tấn công thì「不要装死，逃向车或房子，逃不了就用手边任何东西反击」 — đừng giả chết, chạy về phía xe hoặc nhà, không thoát thì phản công bằng bất cứ thứ gì bên cạnh. Bị chó cắn thì xử lý theo mục 13 của phần này +- Mức bằng chứng: B +- Nguồn: US National Park Service. Staying Safe Around Bears. ；Văn phòng Tổng hợp Cục Kiểm soát bệnh quốc gia, Văn phòng Ủy ban Sức khoẻ Quốc gia (国家疾控局综合司、国家卫生健康委办公厅) (2023). Quy định công tác phòng ngừa và xử lý sau phơi nhiễm bệnh dại (bản 2023). +- Ghi chú: Mức B chỉ áp cho gấu. Riêng「đừng chạy, lùi chậm, tìm chỗ cao hoặc vật cản」 với lợn rừng và chó hoang là kinh nghiệm, tính là mức C, lần này không tìm được nhắc nhở chính thức của Cục Quản lý Lâm nghiệp và Phát triển Quốc gia cùng các cục lâm nghiệp địa phương. Đừng dùng thức ăn dỗ, đừng áp sát chụp ảnh, đừng dắt chó vào núi có lợn rừng. Ở Trung Quốc gấu chủ yếu là gấu đen và gấu nâu, giả chết hay không giả chết thì tách ra xem theo hai loại đó. + +### 32. Gặp giông ngoài hoang dã: chui vào xe có nóc cứng hoặc vào công trình, tránh đường sống lưng, cây đứng một mình, mặt nước và kim loại + +- Chi phí: Không tốn tiền. Cái giá phải trả là lịch trình bị gián đoạn hơn nửa giờ. Khó là khi tiếng sấm còn xa thì đã chịu quay về. +- Nói thẳng: Ngoài trời trong vùng có giông dông thì không có chỗ nào an toàn. Chui vào xe có nóc cứng hoặc vào công trình; xe mui trần, xe máy, lều trại, chòi, dưới gốc cây, dưới phần đá nhô ra đều không tính là chỗ che. Tránh bãi trống, đỉnh núi và đường sống lưng, cây đứng một mình, mặt nước và lan can sắt. Cả nhóm không đi được thì tản ra đứng. Nghe tiếng sấm là rút, sau tiếng sấm cuối cùng 30 phút mới coi là an toàn. +- Lợi ích: Ngoài trời trong vùng có giông dông thì không có chỗ nào an toàn. Sau tiếng sấm cuối cùng 30 phút mới coi là an toàn +- Mức bằng chứng: B +- Nguồn: US National Weather Service. An toàn khi có sét ngoài trời: 「There is no safe place outside when thunderstorms are in the area」「Avoid open fields, the top of a hill or a ridge top. Stay away from tall, isolated trees or other tall objects」「Stay away from water, wet items, such as ropes, and metal objects, such as fences and poles」 — khi có giông dông trong khu vực thì không có chỗ nào ngoài trời là an toàn; tránh bãi trống, đỉnh đồi hoặc đường sống lưng, tránh xa cây cao đứng lẻ hoặc vật cao khác; tránh xa nước, vật ướt như dây thừng, và vật kim loại như hàng rào, cột; khi thật sự không đi được thì「If you are in a group, spread out to avoid the current traveling between group members」 — nếu đi thành nhóm, hãy tản ra để tránh dòng điện lan trong nhóm; muốn hoạt động lại thì「Wait 30 minutes until after the last rumble of thunder」 — chờ 30 phút sau tiếng sấm cuối cùng. +- Ghi chú: Lều trại, chòi, dưới gốc cây, dưới phần đá nhô ra đều không tính là chỗ che. Xe có nóc cứng thì tính, xe mui trần và xe máy thì không. Nghe tiếng sấm là bắt đầu rút, đừng đợi mưa rơi xuống mới đi. +### 33. Lên cao nguyên thì tăng dần theo độ cao ngủ (đêm đó ngủ ở cao bao nhiêu); triệu chứng nặng thì hạ xuống, đừng gắng chịu + +- Chi phí: Tốn thêm một đến hai ngày để thích nghi. Thuốc phòng ngừa là thuốc kê đơn, phải hỏi bác sĩ trước khi xuất phát. Khó là lịch trình chật mà vẫn chừa ra mấy ngày đó. +- Nói thẳng: Cái cần tính là ngủ ở cao bao nhiêu, không phải ban ngày đã lên tới cao bao nhiêu. Ngủ ở trên 2450 mét là đã có nguy cơ bệnh độ cao. Sau khi vượt 3000 mét, độ cao ngủ mỗi đêm chỉ nên tăng thêm 500 mét, cứ lên thêm 1000 mét thì chừa thêm một ngày để thích nghi. Đau đầu kèm buồn nôn, đi lảo đảo, ý thức thay đổi, hoặc không hẳn mình cũng thở không nổi thì hãy đi xuống. Hạ xuống đứng trước hít oxy và uống thuốc. +- Lợi ích: Ngủ ở trên 2450 mét là đã có nguy cơ bệnh độ cao. Phù não cao nguyên và phù phổi (tích dịch trong não và phổi), không hạ xuống vùng thấp kịp thì chết +- Mức bằng chứng: A +- Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Sách vàng, du lịch độ cao lớn và bệnh độ cao: 「Any unacclimatized traveler proceeding to a sleeping altitude of ≥2,450 m (≥8,000 ft)—and sometimes lower—is at risk for altitude illness」 — bất kỳ người đi du lịch chưa thích nghi nào tới độ cao ngủ ≥2.450 m (≥8.000 ft), đôi khi còn thấp hơn, đều có nguy cơ bị bệnh độ cao; tốc độ leo「500 m (1,650 ft) per night in sleeping altitude once above 3,000 m (9,800 ft)」 — 500 m (1.650 ft) mỗi đêm tính theo độ cao ngủ, một khi đã vượt 3.000 m (9.800 ft), mỗi lần lên thêm 1000 mét thì cộng thêm một ngày thích nghi; thuốc phòng ngừa acetazolamide「125 mg twice a day; 250 mg twice a day if >100 kg body weight」; nếu「If symptoms become worse despite rest or treatment at the same elevation」 thì phải hạ xuống. ；Hiệp hội Y khoa Hoang dã (2024). Hướng dẫn thực hành lâm sàng về phòng ngừa, chẩn đoán và điều trị bệnh độ cao cấp tính (cập nhật 2024). Wilderness & Environmental Medicine. +- Ghi chú: Cái cần tính là「ngủ ở cao bao nhiêu」, không phải ban ngày đã lên tới cao bao nhiêu. Đau đầu kèm buồn nôn, đi lảo đảo, ý thức thay đổi, hoặc không hẳn mình cũng thở không nổi, đều là tín hiệu phải đi xuống. Hạ xuống được ưu tiên trước hít oxy và uống thuốc, vừa hít oxy vừa leo tiếp thì làm sai. + +### 34. Bị ve cắn: dùng nhíp mỏm nhọn áp sát da rồi kéo ra thật đều, đừng bôi dầu, đừng đốt bằng lửa + +- Chi phí: Một cây nhíp mỏm nhọn. +- Nói thẳng: Dùng nhíp mỏm nhọn kẹp vào phần đầu của con ve, áp sát da, rồi kéo thẳng ra ngoài với lực đều đặn. Đừng bôi dầu, đừng đốt bằng lửa, đừng xoay rồi giật, làm vậy sẽ kích thích nó đẩy dịch có mầm bệnh vào da, và dễ làm phần miệng cắn đang cắm trong thịt nằm lại. Rút xong thì ghi lại ngày tháng. Trong vài tuần sau đó mà sốt, nổi ban, mệt rã rời thì khi đi khám hãy nói trước là đã bị ve cắn. +- Lợi ích: Bôi dầu, đốt bằng lửa, xoay rồi giật sẽ kích thích con ve, đẩy dịch có mầm bệnh vào da. Làm vậy cũng dễ làm phần miệng cắn đang cắm trong thịt nằm lại bên trong. Hội chứng sốt kèm giảm tiểu cầu mà ve truyền ở Trung Quốc: năm 2024 cả nước báo cáo 6602 ca, 306 ca tử vong, đều là mức cao nhất kể từ khi phát hiện bệnh này. Giai đoạn 2010 đến 2024, tỷ lệ tử vong trung bình là 4.79%, càng lớn tuổi càng cao, từ 85 tuổi trở lên là 9.15% +- Mức bằng chứng: B +- Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Sau khi bị ve cắn: 「Grasp the tick as close to the skin's surface as possible using clean fine-tipped tweezers」「Pull tick away from the skin with steady, even pressure」「Do not use petroleum jelly, heat, nail polish, or other substances to try and make the tick detach from the skin. This may agitate the tick and force infected fluid from the tick into the skin」 — dùng nhíp sạch mỏm nhọn kẹp ve sát bề mặt da nhất có thể; kéo ve ra khỏi da bằng lực đều đặn; không dùng dầu mỏ, nhiệt, sơn móng tay hay vật khác để làm ve rời da, vì làm vậy có thể kích thích ve và đẩy dịch nhiễm từ ve vào da. ；Ren Q và cộng sự (2026). Epidemiological characteristics of severe fever with thrombocytopenia syndrome (SFTS) in China, 2010–2024: a fifteen-year national surveillance analysis. BMC Infectious Diseases 26:1497: 「The number of reported cases (6,602 cases) and deaths (306 cases) in 2024 were both the highest levels since the discovery of this disease」「The average CFR was 4.79%, and the CFR increased with age, reaching 9.15% among those aged 85 years and above」 — số ca báo cáo (6.602 ca) và số ca tử vong (306 ca) trong năm 2024 đều ở mức cao nhất kể từ khi phát hiện bệnh này; tỷ lệ tử vong trung bình là 4.79% và tăng theo tuổi, đạt 9.15% ở nhóm từ 85 tuổi trở lên. +- Ghi chú: Rút xong thì ghi lại ngày tháng. Trong vài tuần sau đó mà sốt, nổi ban, mệt rã rời thì khi đi khám hãy nói trước là đã bị ve cắn. Bác sĩ không hỏi mà bạn không nói thì dễ bị điều trị như bị cảm. + +### 35. Nước ngoài hoang dã thì cứ đun sôi rồi uống, sôi thêm một phút là đủ, không cần nấu lâu hơn vì ở cao + +- Chi phí: Phải mang thêm bộ bếp và nhiên liệu. +- Nói thẳng: Mọi vi sinh vật gây bệnh qua đường uống đều chết trong vài giây ở nhiệt độ sôi. Sôi thêm một phút chỉ là khoảng dự phòng cho lúc mắt nhìn nhầm, không phải vì ở cao mà phải nấu lâu hơn. Không đun được thì dùng máy lọc hoặc viên khử trùng, mỗi thứ có những mầm bệnh mà thứ kia không chặn được, đun được thì cứ đun. Dòng suối nhìn trong, nước tuyết cũng có thể có ký sinh trùng. +- Lợi ích: Mọi vi sinh vật gây bệnh qua đường uống đều bị giết (bị vô hiệu hoá) trong vài giây khi nước được đun sôi. 「高海拔水烧不到位」 — nước ở cao không đun đủ — là một hiểu lầm phổ biến +- Mức bằng chứng: B +- Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Sách vàng, khử trùng nước cho người đi du lịch: 「All organisms that may cause illness from drinking water are killed within seconds at boiling temperature」「CDC recommends that travelers boil water for a full minute to account for user variability in identifying boiling points and to add a margin of safety」「Although the boiling point for water decreases with increasing elevation, at common travel elevations the temperature needed to achieve boiling is still well above the temperature required to inactivate enteric pathogens」 — mọi sinh vật có thể gây bệnh khi uống nước đều bị giết trong vài giây ở nhiệt độ sôi; CDC khuyên người đi du lịch đun nước sôi trọn một phút, để tính tới việc người dùng nhận ra điểm sôi không giống nhau và để chừa biên an toàn; tuy điểm sôi của nước giảm khi độ cao tăng, ở những độ cao du lịch thông thường thì nhiệt độ cần thiết để đun sôi vẫn cao hơn nhiều nhiệt độ cần thiết để vô hiệu hoá mầm bệnh đường ruột. +- Ghi chú: Cùng trang đó còn viết rằng không nhất thiết phải đun thật sôi: giữ ở 60 ℃ trong 30 phút cũng hiệu quả như nhau, còn đun sôi chỉ là điểm kết thúc nhìn thấy được mà không cần nhiệt kế. Không đun được thì dùng máy lọc hoặc viên khử trùng hoá chất, mỗi thứ có những mầm bệnh mà thứ kia không chặn được, đun được thì cứ đun. Dòng suối nhìn trong, nước tuyết cũng có thể có ký sinh trùng, đừng đánh giá bằng mắt. + +### 36. Bị người lạ đòi tiền ngoài hoang đương: đưa tiền cho họ, không ra tay, ghi nhớ đặc điểm, thoát ra thì báo công an + +- Chi phí: Tiền mặt trong người. Khó là kìm được cơn tức. +- Nói thẳng: Đưa tiền cho họ, không ra tay, không cãi lại. Bên kia liều mạng vì mấy trăm tệ mà đánh cướp tài sản: từ ba năm đến mười năm, xâm nhà, mang súng, gây thương tích nặng hoặc làm chết người thì từ mười năm trở lên, cho tới án tử hình. Họ cầm tiền rồi quay đi, bạn đuổi lên đánh thì không còn tính là tự vệ. Thoát ra thì ghi ngay đặc điểm ngoại hình, giọng nói, biển số xe và hướng chạy trốn, rồi gọi 110. +- Lợi ích: Mạng người đáng hơn tiền, bên kia liều mạng vì mấy trăm tệ mà phạm tội nặng. Điều hai trăm sáu mươi ba của Bộ luật Hình sự quy định: khi cướp tài sản công cộng hoặc tư nhân bằng bạo lực, ép buộc hay hình thức khác thì phạt tù từ ba năm đến mười năm kèm phạt tiền. Các trường hợp mang súng, xâm nhà, gây thương tích nặng hoặc chết người thì phạt tù từ mười năm trở lên cho tới án tử hình. Tự vệ theo Điều hai mươi của Bộ luật Hình sự chỉ bảo vệ việc「chấm dứt hành vi xâm phạm đang diễn ra」. Bên kia cầm tiền rồi quay đi, bạn đuổi lên đánh thì không còn là tự vệ. Thoát ra thì ghi ngay đặc điểm ngoại hình, giọng nói, biển số xe, hướng chạy trốn, rồi gọi 110 +- Mức bằng chứng: C +- Nguồn: Quốc hội (2020). Bộ luật Hình sự, 刑法（第二十条、第二百六十三条）. (bản tổng hợp do Cục Quản lý giao thông thuộc Công an thành phố Bắc Kinh đăng lại) +- Ghi chú: Lý do để ở mức C là mệnh đề「đưa tiền mà không chống trả thì an toàn hơn」 không có tài liệu trực tiếp, điều luật chỉ nói rủi ro của bên kia và giới hạn của bạn. Đừng đi theo ra chỗ vắng để「rút tiền」. Đừng đưa mật khẩu điện thoại. Đừng lên xe của họ. Trả tiền trước chỉ áp dụng cho tình huống bị chặn trực diện, tính mạng bị đe doạ, thoát ra rồi vẫn báo công an. Bị người ta dùng ảnh riêng tư hoặc video bạn trong lúc trò chuyện tình cảm đòi tiền thì ngược lại, không đưa một đồng nào, xem mục 32 ở phần 8 (đối phương cầm điểm yếu đòi tiền). + +### 37. Gặp một nhóm đánh nhau: lùi ra rồi bỏ đi, đừng vào rách, đừng đứng xem, đừng nhặt đồ nằm trên mặt đất; nếu báo công an thì lùi tới khoảng cách an toàn rồi gọi 110 + +- Chi phí: Không tốn tiền. Khó là nhịn cả hai cơn thèm: bước vào rách, và đứng lại xem náo nhiệt. Sau khi báo công an thì có thể bị gọi lại làm một lần biên bản, tốn nhiều nhất là nửa ngày. +- Nói thẳng: Bạn động vào một cái là việc rách đánh trở thành tham gia đánh nhau. Tham gia tụ chế đánh nhau thì phạt tù dưới 3 năm; đông người, ở trên đường, cầm đồ vũ khí thì phạt tù từ 3 đến 10 năm. Đứng xem cũng dễ bị tổn thương do nhầm. Việc nên làm là lùi ra rồi bỏ đi. Nếu muốn báo công an thì lùi tới khoảng cách an toàn rồi gọi 110, nói rõ địa điểm, có mấy người, có ai cầm dao gậy không, có ai nằm bất động không. +- Lợi ích: Điều hai trăm chín mươi hai của Bộ luật Hình sự:「聚众斗殴的，对首要分子和其他积极参加的，处三年以下有期徒刑、拘役或者管制」 — trong tụ chế đánh nhau, người chủ chốt và những người khác tích cực tham gia thì phạt tù dưới ba năm, kỉ luật hoặc điều tra không giam giữ. Tụ chế đánh nhau nhiều lần, số người đông, quy mô lớn, ảnh hưởng xấu đến xã hội, gây trật tự xã hội nghiêm trọng ở nơi công cộng hoặc tuyến giao thông chính, hoặc mang vũ khí, thì「处三年以上十年以下有期徒刑」 — phạt tù từ ba năm đến mười năm. Tụ chế đánh nhau「聚众斗殴，致人重伤、死亡的」 — gây thương tích nặng hoặc chết — thì bị truy cứu theo tội cố ý gây thương tích, cố ý giết người. Luật Xử phạt quản lý trật tự công cộng cũng có điều khoản xử phạt. Đánh người khác thì bị giữ 5 đến 10 ngày và phạt tiền 500 đến 1000 tệ. Họp nhóm đánh nhau thì xử lý theo tội gây rối, tìm cớ tranh cãi, tối đa bị giữ 15 ngày và phạt tiền 2000 tệ. Theo mục 5 của Ý kiến chỉ đạo về việc áp dụng đúng đắn hệ thống tự vệ (关于依法适用正当防卫制度的指导意见), hành vi xâm phạm「既包括针对本人的不法侵害，也包括危害国家、公共利益或者针对他人的不法侵害」 — không chỉ gồm xâm phạm nhắm vào bản thân mình, mà còn gồm xâm phạm gây hại cho quốc gia, lợi ích công cộng hoặc cho người khác. Theo mục 9 của ý kiến chỉ đạo đó: một bên ra tay trước mà dùng thủ đoạn quá đà, hoặc một bên ra tay trước rồi vẫn tiếp tục xâm phạm dù bên kia đang cố tránh xung đột, thì bên đánh lại thường được xác định là hành vi tự vệ +- Mức bằng chứng: B +- Nguồn: Quốc hội (1997). Bộ luật Hình sự, 刑法（第二百三十二、二百三十四、二百九十二、三百零五、三百一十条）. ；Ủy ban Thường vụ Quốc hội (2025). Luật Xử phạt quản lý trật tự công cộng, 治安管理处罚法（2025 年修订，第三十、五十一条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部) (2020). Ý kiến chỉ đạo về việc áp dụng đúng đắn hệ thống tự vệ, 关于依法适用正当防卫制度的指导意见（法发〔2020〕31 号, mục 5 và 9）. ；Quốc hội (2018). Bộ luật Tố tụng hình sự, 刑事诉讼法（2018 年修订，第六十二、一百一十、一百一十一、一百九十三条）. Toàn văn đăng lại trên Viện kiểm sát nhân dân dân tộc trực thuộc +- Ghi chú: Để ở mức B là vì phần hậu quả có văn bản luật cứng, nhưng bước「lùi ra, báo công an, không vào rách」 thì không có nguyên văn chính thức, phải suy ra từ điều luật và từ cách xác định đánh nhau hai bên. 「Tôi đi khuyên can thôi」 không phải lá chắn, khi xác định người ta xem bạn có ra tay không, đánh vào ai. Bên kia đã nằm xuống hoặc đã bỏ chạy, bạn vẫn đuổi lên đánh thì không còn tính là tự vệ. Thấy một người bị vây đánh một chiều thì về mặt pháp lý bạn thật sự có quyền ngăn lại. Tự vệ không chỉ bảo vệ bản thân mình mà cả người khác. Nhưng vào bằng tay trần thì nhiều khi chỉ thêm một người bị thương. Gọi người, quay video từ xa, báo công an thì hữu ích hơn. 「Không vào, không đứng xem」 chỉ có lợi cho bạn. Lợi ích lớn nhất của việc gọi 110 rơi vào người đang nằm dưới đất, đó là bậc thấp nhất trong nhóm người được lợi mà cuốn sách này nói tới; có bỏ thời gian hay không thì bạn tự cân. Trước khi cân, cần biết mấy điều. Nghĩa vụ làm chứng vốn đã có: Điều sáu mươi hai của Bộ luật Tố tụng hình sự viết rõ rằng ai biết tình hình vụ án đều có nghĩa vụ làm chứng, không gọi cái điện thoại đó thì nghĩa vụ cũng không biến mất. Có thương tích nặng hoặc có người chết thì cảnh sát lấy hình giám sát vẫn tìm ra người có mặt, đứng xem thì ngược lại dễ bị ghi sổ nhất. Người báo việc không muốn công khai tên thì cơ quan xử lý「应当为他保守秘密」 — phải giữ kín danh tính cho họ (Điều một trăm mười một). Nói rõ địa điểm, số người, có cầm đồ vũ khí hay không là có thể treo máy, không cần ở lại tại chỗ. Bước tới mức bị nhắc nhở hoặc bị giữ dưới mười ngày là khi tòa triệu tập bạn ra tòa, bạn không đưa ra được lý do chính đáng mà vẫn không chịu làm chứng (Điều một trăm chín mươi ba), người đi đường hiếm khi tới bước này. Không báo công an không vi phạm luật: câu「有权利也有义务」 ở Điều một trăm mười của Bộ luật Tố tụng hình sự không có điều khoản phạt, và Bộ luật Hình sự cũng không có tội phạt cho người bình thường biết mà không báo, nên bỏ đi là lựa chọn hợp pháp. Sau khi bỏ đi chỉ còn hai vạch đỏ. Một là giúp giấu người hoặc làm chứng giả để bao bọc sau đó (Điều ba trăm mười của Bộ luật Hình sự, dưới 3 năm, trường hợp nghiêm trọng từ 3 đến 10 năm). Hai là cố tình khai chứng giả khi được yêu cầu làm chứng (Điều ba trăm linh năm, dưới 3 năm, trường hợp nghiêm trọng từ 3 đến 7 năm). Im lặng và nói đúng sự thật thì không sao, bịa chuyện thì mới có chuyện. Bản thân bạn bị kéo vào xung đột, xem mục 10 ở phần 8; bị người lạ đòi tiền, xem mục 36 của phần này; chuyện tiền sau khi cứu người mà bị thương, xem mục 39 của phần này. +### 38. Có thể đã bị phơi nhiễm HIV: trong 72 giờ đi lấy thuốc ngăn chặn, càng sớm càng tốt + +- Chi phí: Một liệu trình 28 ngày thuốc giá hơn nghìn tệ, ở phần lớn các nơi phải tự trả. Tối đó phải chạy tới bệnh viện một chuyến. Khó là thừa nhận ngay tại chỗ rằng mình có thể đã bị lây, rồi ra khỏi nhà ngay. +- Nói thẳng: Nghi ngờ mình đã bị lây HIV thì 72 giờ là mốc cứng, tốt nhất là trong 2 giờ đã uống được thuốc. Thuốc phải uống liên tục 28 ngày. Quá 72 giờ là hết đường, nên đừng đợi tới hôm sau mới quyết. +- Lợi ích: Chuyên mục phổ biến kiến thức phòng HIV của Chính phủ thành phố Bắc Kinh nói: thuốc phòng ngừa sau khi đã tiếp xúc gọi là phòng ngừa sau phơi nhiễm (viết tắt PEP), và nên「尽可能在 2 小时内进行预防性用药，最好在 24 小时内、不超过 72 小时」 — càng sớm càng tốt thì dùng thuốc phòng ngừa trong 2 giờ, tốt nhất trong 24 giờ, không quá 72 giờ. Còn phải「连续服用 28 天」. Quá 72 giờ thì không còn khuyến cáo bắt đầu, đó là một mốc thời gian cứng +- Mức bằng chứng: B +- Nguồn: Chính phủ thành phố Bắc Kinh. Trước và sau khó phân biệt? Sáu câu hỏi sáu câu trả lời để hiểu toàn diện phòng ngừa trước và sau phơi nhiễm HIV. ；Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc, Trung tâm Kiểm soát bệnh lây qua đường tình dục và HIV (2020). Hướng dẫn kỹ thuật phòng ngừa sau phơi nhiễm HIV (bản thử nghiệm). +- Ghi chú: Lấy thuốc ở đâu: cơ sở y tế chuyên HIV do Trung tâm Kiểm soát bệnh địa phương chỉ định, thường nằm ở khoa Bệnh truyền nhiễm (phòng khám bệnh lây qua đường tình dục và HIV). Nhiều thành phố đã có phòng khám mở 24 giờ, số liệu công khai của Trùng Khánh là 29 phòng khám ở 26 quận huyện, trong đó 14 phòng mở suốt ngày. Trước khi đi, gọi điện thoại Trung tâm Kiểm soát bệnh địa phương hỏi chỗ nào có thuốc, đừng phí thời gian ở khoa cấp cứu. Các tình huống được tính là「phơi nhiễm」 gồm quan hệ tình dục không dùng biện pháp bảo vệ, bao cao su rách, bị kim đâm, máu dính lên niêm mạc; có dùng thuốc hay không thì bác sĩ quyết định. Thuốc phải uống đủ 28 ngày, bỏ mấy lần thì hiệu quả giảm rõ rệt. Sau đó theo lệnh bác sĩ, xét lại ở tuần thứ 4 đến 6 và ở tháng thứ 3. Thuốc ngăn chặn chỉ là biện pháp khắc phục; phòng ngừa thường ngày và xét nghiệm xem mục 30 và 31 ở phần 1 (dùng bao cao su suốt quá trình, có quan hệ nguy cơ cao thì đi xét). + +### 39. Cứu người mà bị thương, mất tiền: trước hết tìm người gây hại và bảo hiểm y tế, rồi hãy đi khai để được xác nhận hành động nghĩa + +- Chi phí: Không tốn tiền. Khai báo thì phải nộp hồ sơ, chờ kết quả xác nhận, chạy vài chuyến, mất vài tuần đến vài tháng. +- Nói thẳng: Cứu người mà thiệt hại cho bản thân thì không phải tự nhận xui xẻo. Người bị thương khi làm việc nghĩa thì cứu chữa trước, tính tiền sau. Phí y tế do người gây hại trả, người đó bỏ trốn hoặc không đủ khả năng thì đi theo bảo hiểm y tế cơ bản. Không có người gây hại thì người bạn cứu, theo luật, phải cho bạn bồi thường tương xứng. Đồng thời đi khai xác nhận hành động nghĩa, nếu bị thương thì thực hiện chế độ bảo hiểm tai nạn lao động. +- Lợi ích: Điều một trăm tám mươi ba của Bộ luật Dân sự:「因保护他人民事权益使自己受到损害的，由侵权人承担民事责任，受益人可以给予适当补偿。没有侵权人、侵权人逃逸或者无力承担民事责任，受害人请求补偿的，受益人应当给予适当补偿」 — khi vì bảo vệ quyền lợi dân sự của người khác mà bản thân bị tổn hại thì người gây hại chịu trách nhiệm dân sự, người hưởng lợi có thể cho bồi thường tương xứng; không có người gây hại, người gây hại trốn đi hoặc không đủ khả năng chịu trách nhiệm dân sự, mà người bị hại yêu cầu bồi thường thì người hưởng lợi phải cho bồi thường tương xứng. Công văn số 39 năm 2012 của Văn phòng Nội các quốc gia nói xử lý người bị thương khi làm việc nghĩa thế nào. Cơ sở y tế phải lập kênh thông xanh,「坚持『先救治、后收费』的原则」. Nguyên văn về khoản phí y tế là「因紧急救治发生的医疗费用，有加害人或责任人的，由加害人或责任人承担；无加害人或责任人以及加害人或责任人逃逸或者无力承担的，按规定通过基本医疗保障制度解决」 — phí y tế phát sinh do cứu chữa khẩn cấp, có người gây hại hoặc người chịu trách nhiệm thì do người đó chịu; không có người gây hại hoặc người chịu trách nhiệm, hoặc người đó trốn đi hay không đủ khả năng chịu, thì giải quyết theo quy định qua hệ thống bảo hiểm y tế cơ bản. Người chết thì「符合烈士评定条件的，依法评定为烈士」 — nếu đủ điều kiện được phong hàm anh thì được xếp là liệt sĩ theo luật; 「属于视同工伤情形的，享受一次性工亡补助金以及相当于本人 40 个月工资的遗属特别补助金」 — nếu thuộc trường hợp coi như tai nạn lao động thì được hưởng trợ cấp tử vong một lần cùng khoản trợ cấp đặc biệt cho người thân, bằng tiền lương 40 tháng của người đó. Người bị tàn tật thì「符合享受工伤保险待遇条件的，依据《工伤保险条例》落实相应待遇；不符合的，由民政部门评定伤残等级并落实相应待遇」 — nếu đủ điều kiện hưởng chế độ bảo hiểm tai nạn lao động thì thực hiện chế độ tương ứng theo Quy định về bảo hiểm tai nạn lao động; không đủ thì cơ quan dân sự xếp cấp độ tàn tật và thực hiện chế độ tương ứng. Về việc làm thì「优先纳入就业援助，予以重点支持」 — ưu tiên đưa vào hỗ trợ việc làm, có hỗ trợ trọng điểm, chức danh công ích thì「优先安排符合条件的见义勇为人员」 — ưu tiên xếp người làm việc nghĩa đủ điều kiện. Về nhà ở thì「优先纳入住房保障体系，优先配租、配售保障性住房或发放住房租赁补贴」 — ưu tiên đưa vào hệ thống bảo đảm nhà ở, ưu tiên cho thuê, bán nhà ở bảo đảm hoặc cấp trợ cấp thuê nhà +- Mức bằng chứng: B +- Nguồn: Quốc hội (2020). Bộ luật Dân sự, 民法典（第一百八十三条）. ；Văn phòng Nội các quốc gia chuyển đạt Bộ Dân vụ và các bộ khác (国务院办公厅转发民政部等部门) (2012). Ý kiến về tăng cường bảo vệ quyền lợi của người làm việc nghĩa, 关于加强见义勇为人员权益保护的意见（国办发〔2012〕39 号）. +- Ghi chú: Để ở mức B là vì các chế độ được hưởng bị văn bản của Trung ương ấn định, nhưng thưởng bao nhiêu, đi xác nhận thế nào thì do điều lệ về việc nghĩa của từng tỉnh quy định, số ở mỗi nơi mỗi khác, lần này chưa kiểm tra từng tỉnh. Thường thì khai với cơ quan công an nơi xảy ra chuyện hoặc với đơn vị phụ trách việc nghĩa (quỹ). Có được xác nhận hay không phụ thuộc bằng chứng: giấy biên nhận báo việc, hình giám sát tại hiện trường, tên và số điện thoại của người chứng kiến, hồ sơ bệnh án, tất cả đều lấy ngay tại chỗ. Nếu phải ra tòa thì quyền lợi dân sự của người làm việc nghĩa vốn nằm trong phạm vi trợ giúp pháp lý miễn phí. Mức trợ cấp tử vong một lần theo tiêu chuẩn toàn quốc và cách quy đổi mức tàn tật thành tiền, xem mục 15 và 16 ở phần 19. Cách xin trợ giúp pháp lý, xem mục 3 ở phần 7. + +### 40. Dao, thép, thuỷ tinh cắm vào người: ấn quanh vật thể để cầm máu, đừng rút nó ra + +- Chi phí: Không tốn tiền. Khó là nhịn cơn bản năng「rút ra trước」, và ngồi giữ người đó chờ xe cứu thương. +- Nói thẳng: Rút ra có thể gây chảy máu ồ ạt, và cũng khiến bác sĩ khó đánh giá tổn thương sâu bao nhiêu. Cách làm: gọi 120, dùng tay hoặc miếng bông để ấn quanh vật thể mà cầm máu, đệm cho vật thể nằm yên không rung. Vết thương ở bụng mà lộ ruột ra thì đừng nhét lại, phủ một miếng gạc sạch ướt lên. Vết thương ở ngực mà đang sủi bọt thì đừng bịt kín bằng nilon, bịt kín có thể làm xẹp phổi. +- Lợi ích: Hướng dẫn cấp cứu quốc tế năm 2025 của Hiệp hội Liên hợp Chữ thập Đỏ và Trăng Lưỡi liền quốc tế:「急救的人应当压住扎进身体的异物（比如刀）周围，把异物稳住，不要把它拔出来」 — người cứu nạn nên ấn quanh vật thể cắm vào cơ thể (ví dụ con dao), giữ vật thể ổn định, không rút ra. 「不要从腹部或胸部的伤口里拔出任何扎入的异物，拔出来可能造成大出血，也会让医生更难判断伤得有多重」 — không rút bất cứ vật thể nào đang cắm ở vết thương bụng hoặc ngực ra, vì rút ra có thể gây chảy máu ồ ạt và khiến bác sĩ khó đánh giá tổn thương nặng bao nhiêu. 「扎在身体里的异物应当稳住；异物随脉搏跳动的，就让它跳，松松地稳住即可」 — vật thể cắm trong cơ thể phải được giữ ổn định; nếu vật thể nhảy theo mạch thì cứ để nó nhảy, giữ lỏng là được. 「急救的人不应把露出来的内脏塞回体内」 — người cứu nạn không nên nhét ruột lộ ra trở lại vào trong. Vết thương thành ngực hở không chảy máu ra ngoài thì「可以敞着、不加敷料，因为把伤口封死可能让空气积在胸腔里，使肺部分或完全塌陷（气胸）」 — có thể để hở, không băng bó, vì bịt kín vết thương có thể làm khí tích trong lồng ngực, khiến một phần hoặc toàn bộ phổi xẹp (tràn khí). Mọi vết thương xuyên thấm vùng ngực và bụng đều phải gọi cấp cứu ngay +- Mức bằng chứng: B +- Nguồn: International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, trang 256–257, 267–268. +- Ghi chú: Để ở mức B là vì đây là cách làm được hướng dẫn công nhận, không có thử nghiệm chia nhóm ngẫu nhiên, tất nhiên không thể chia người bị thương thành hai nhóm「rút」 và「không rút」 để thử. Ngoại lệ duy nhất là khi vật thể chặn khí quản, người đó không thở nổi. Tới bệnh viện thì đi thẳng tới cái bàn phân loại cấp thiết trước cửa khoa cấp cứu (bàn tiếp đón và phân loại), báo「vết thương xuyên thấm」. Đây là mức phải vào phòng cấp cứu ngay, đừng xếp hàng ở quầy đăng ký khám (xem mục 8 ở phần 24). Thật sự bị xuyên thấm thì gọi 120, xe cứu thương sẽ báo trước bệnh viện để chuẩn bị tiếp nhận. Chảy máu ồ ạt thì ấn thế nào, đặt gar thế nào, xem mục 12 của phần này. + +### 41. Nghi ngờ gãy xương thì đừng cử động chỗ đó nữa, đỡ cho cố định, chườm lạnh trong vòng 20 phút, đừng tự nắn lại + +- Chi phí: Không tốn tiền. Tìm một tấm ván cứng, tạp chí cuộn lại hoặc quần áo làm chỗ tựa. Chỗ chườm lạnh phải có một lớp vải ngăn ở giữa. +- Nói thẳng: Chân tay bị thương thì cứ coi là gãy xương trước, đứng được, cử động được cũng không loại trừ được, chỉ X quang mới xác định. Để vùng bị thương giữ nguyên vị trí, dùng tay đỡ, hoặc dùng khăn tam giác, ván cứng chống theo đúng vị trí cũ. Chườm lạnh càng sớm càng tốt, mỗi lần không quá 20 phút. Đừng tự nắn lại. Xương gãy xuyên ra da thì cầm máu trước, phủ gạc sạch, đừng ấn xương trở lại. +- Lợi ích: Hiệp hội Liên hợp Chữ thập Đỏ và Trăng Lưỡi liền quốc tế đã ban hành hướng dẫn cấp cứu quốc tế năm 2025. Trong đó viết「手脚上的任何一处伤，都应当先按可能的骨折来处理」 — bất kỳ chỗ nào trên chân tay bị thương, đều nên xử lý trước như là gãy xương có thể. Còn viết「确诊骨折唯一准确的办法是拍 X 光。拿不准就当骨折办」 — cách duy nhất chính xác để xác nhận gãy xương là chụp X quang, không chắc thì cứ coi như gãy xương. Mục đích của việc cố định là「尽量减少活动和疼痛」 — hạn chế cử động và đau nhất có thể. Chườm lạnh thì「一次最多连续 20 分钟，这样可以减轻疼痛；冷得比这更久可能损伤皮肤」 — mỗi lần tối đa 20 phút liên tục, như vậy giảm đau, chườm lâu hơn thế có thể làm tổn thương da. 「伤到股骨这类长骨时，光是骨头本身出的血就可能致命」 — khi bị thương ở xương dài như xương đùi, chỉ riêng máu chảy ra từ xương đã có thể gây tử vong. Khi đau hoặc sưng rõ, có dấu hiệu sốc, gãy xương đùi hoặc chi thể bị sai vị trí rõ ràng thì nên gọi cấp cứu. Gãy xương cong thành góc thì phải có đồng thời hai điều kiện mới nghĩ tới việc nắn về nguyên vị trí ngay tại chỗ (复位). Một là「地处偏远（或者条件有限），而且成角处下方的肢体发凉、发白，或者急救的人没法固定好以便搬动」 — ở nơi xa xôi (hoặc điều kiện hạn chế), và phần chi thể dưới chỗ gãy góc bị lạnh, tái, hoặc người cứu nạn không cố định được để cử động. Hai là người cứu nạn đã qua huấn luyện. Với gãy xương hở (xương gãy xuyên ra da) thì việc trước tiên là cầm máu và phòng nhiễm trùng +- Mức bằng chứng: B +- Nguồn: International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, trang 301–302. +- Ghi chú: Để ở mức B là vì đây đều là những cách làm được công nhận. Kéo giãn cố định cho vết gãy xương đùi trước khi đưa đi bệnh viện thì chỉ có một nghiên cứu hồi sát bệnh án cũ làm căn cứ, độ tin cậy của bằng chứng rất thấp. Kết quả là lượng máu truyền trong 24 giờ ít hơn, thời gian nằm viện ngắn hơn, còn khác biệt về tử vong và biến chứng nhỏ đến mức không nói rõ được. Còn về việc nên nắn về nguyên vị trí trước hay cố định nguyên trạng thì hẳn là không tìm thấy bằng chứng. Ba hiểu lầm thường gặp cần sửa: đứng được đi được không có nghĩa là không gãy xương; chỉ dùng lạnh không dùng nóng, cũng đừng luân phiên nóng lạnh; chấn thương trẹo không phải「vận động một chút là khỏi」, cứ nghỉ trước. Người lớn tuổi ngã mà nghi ngờ gãy xương thì có những điều cấm khi di chuyển, xem mục 2 của phần này. Sau khi bó bột hoặc nằm liệt giường mà một chân sưng lên thì cảnh giác huyết khối, xem mục 11 của phần này. Sau khi chữa xong có cần làm giám định mức tàn tật, có cần làm giấy chứng nhận người khuyết tật không, xem mục 10 và 11 ở phần 24. + +### 42. Sau khi bị cưỡng ép tình dục: tới nơi an toàn rồi gọi 110; trước khi khám lạm dụng thì đừng tắm, đừng giặt quần áo, đừng dọn phòng, trong 72 giờ hãy tới bệnh viện + +- Chi phí: Báo công an không tốn tiền. Thuốc tránh thai khẩn cấp là thuốc không kê đơn, mua ở hiệu thuốc là được. Thuốc ngăn chặn một liệu trình giá hơn nghìn tệ, có uống hay không thì bác sĩ quyết định, xem mục 38 của phần này (trong 72 giờ đi lấy thuốc ngăn chặn). Khó là lúc đó rất khó mở miệng, lại rất muốn tắm ngay. +- Nói thẳng: Dấu vết còn trên người, trên quần áo và trong phòng là bằng chứng quan trọng nhất để nhận diện bên kia. Vì vậy trước khi khám lạm dụng thì đừng tắm, cũng đừng giặt quần áo, đừng dọn phòng, gọi 110 càng sớm càng tốt. Thuốc tránh thai khẩn cấp có ở hiệu thuốc, uống càng sớm càng hiệu quả. Trong 72 giờ hãy tới bệnh viện, để bác sĩ quyết định có uống thuốc ngăn chặn không. +- Lợi ích: Kiểm sát viên của Viện kiểm sát nhân dân thành phố Bắc Kinh nói:「无论是未成年人还是成年人，遇到侵害，都应当第一时间拨打110报警」 — dù là người chưa thành niên hay người lớn, khi gặp xâm hại thì phải gọi 110 báo ngay lập tức. Báo công an và khám chữa kịp thời thì「便于侦查机关及时勘查现场、提取物证痕迹、固定证据」 — thuận tiện cho cơ quan điều tra khám hiện trường kịp thời, thu dấu vật chứng và vết tích, cố định bằng chứng. Sổ tay phòng chống xâm hại tình dục trên Nhân Dân Nhật Báo mà Sở Dân vụ tỉnh Quý Châu đăng lại viết:「一旦不幸遭遇性侵害，最重要的事情就是留存证据」 — một khi không may gặp xâm hại tình dục thì việc quan trọng nhất là lưu lại bằng chứng. Cần giữ「体液、毛发、皮屑、通信记录等证据」 — dịch vật cơ thể, lông tóc, vảy da, lịch sử liên lạc và các bằng chứng khác — và「被侵害后，不要擅自洗澡、洗衣物、收拾房间、扔东西」 — sau khi bị xâm hại thì không tự ý tắm, không giặt quần áo, không dọn phòng, không vứt bỏ đồ. Với người quen gây ra chuyện thì giữ lại lịch sử cuộc gọi và liên lạc. Bên kia sau đó đe doạ hay mua chuộc thì cũng phải để lại bằng chứng. Viện kiểm sát nhân dân quận Tây thành phố Bắc Kinh viết:「在接受身体检查之前尽量不要洗澡」 — trước khi thăm khám thể chất nên hạn chế tắm. Tổ chức Y tế Thế giới nói thuốc tránh thai khẩn cấp dùng được khi「女性在未采取有效的避孕保护措施时遭受性侵犯」 — phụ nữ bị xâm hại tình dục khi chưa có biện pháp tránh thai bảo vệ có hiệu quả. Khuyến cáo dùng trong 5 ngày, càng sớm càng hiệu quả. Thuốc tránh thai khẩn cấp chứa levonorgestrel (một loại nội tiết tố thai khoa), tỷ lệ mang thai ở phụ nữ đã dùng là 1.2% đến 2.1%. Viên levonorgestrel bán ở hiệu thuốc trong nước là thuốc không kê đơn. Ý kiến của Ủy ban Sức khoẻ thành phố Vũ Hán: sau quan hệ tình dục không có biện pháp bảo vệ thì「72小时以内，服药越早，预防妊娠效果越好」 — trong 72 giờ, uống thuốc càng sớm thì hiệu quả phòng mang thai càng tốt. Thuốc ngăn chặn HIV phải uống trong 2 giờ nếu có thể, muộn nhất không quá 72 giờ (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Viện kiểm sát nhân dân thành phố Bắc Kinh (2024). Nhìn phim qua lăng kính pháp lý丨xem《九部的检察官》, người chưa thành niên gặp xâm hại trái pháp nên tự bảo vệ thế nào? (đăng lại từ Đài Phát thanh Trung Quốc của Tổng cục Phát thanh và Truyền hình quốc gia). ；Sở Dân vụ tỉnh Quý Châu (2024). Sổ tay phòng chống xâm hại tình dục cho người chưa thành niên, phát cho phụ huynh và các cháu! (đăng lại từ Nhân Dân Nhật Báo). ；Viện kiểm sát nhân dân quận Tây thành phố Bắc Kinh (2018). Vì sao họ không còn im lặng nữa…… ；Tổ chức Y tế Thế giới (2021). Tránh thai khẩn cấp (thông tin thực trạng). ；Ủy ban Sức khoẻ thành phố Vũ Hán (2020). Dùng thuốc tránh thai khẩn cấp an toàn và hợp lý. ；Chính phủ thành phố Bắc Kinh. Trước và sau khó phân biệt? Sáu câu hỏi sáu câu trả lời để hiểu toàn diện phòng ngừa trước và sau phơi nhiễm HIV. +- Ghi chú: Để ở mức B là vì đây là kinh nghiệm điều tra và khuyến cáo dùng thuốc của cơ quan chính thức, không có con số「làm theo thì tỷ lệ kết án cao hơn bao nhiêu」. Người được lợi chủ yếu là bạn và gia đình. Con số 72 giờ của thuốc tránh thai khẩn cấp là theo hướng dẫn trong nước, Tổ chức Y tế Thế giới nói dùng được trong 5 ngày, nhưng càng muộn càng tệ, đừng kéo dài. Đã có thai rồi thì thuốc tránh thai khẩn cấp không có tác dụng. Khi báo công an thì nói đúng sự thật, không giấu, không thổi phồng. Bên kia sau đó tìm bạn thương lượng, đưa tiền hay đe doạ thì đều để lại hồ sơ, đừng xoá. Nghĩ kỹ rồi báo cũng được, nhưng bằng chứng sẽ mất dần theo thời gian. Báo hay không báo thì cả hai trường hợp đều phải tới bệnh viện trong hạn. Chuyện ghi âm, xem mục 41 ở phần 8 (ghi âm). Bên kia sau đó dùng ảnh hoặc chuyện báo công an để dọa bạn thì cũng là phạm pháp, xem mục 32 ở phần 8 (đối phương dọa bằng việc báo công an, gửi ảnh để đòi tiền). Người bạn gặp chuyện này thì hãy đi cùng họ để báo công an và tới bệnh viện, đừng khuyên họ tắm hay nghỉ ngơi trước. \ No newline at end of file diff --git "a/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" "b/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" new file mode 100644 index 00000000..208a5dd4 --- /dev/null +++ "b/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" @@ -0,0 +1,92 @@ +[← 回总目录](../README.md) + +# 14. 账号与信息安全 + +口径：金钱与个人信息。账号被别人登进去，丢的先是钱。对方还会用你的账号去骗你通讯录里的人。你的身份也等于被人拿走了。 + +[← Về mục lục](../README.md) + +# 14. Tài khoản và bảo mật thông tin + +Thước đo: tiền và thông tin cá nhân. Khi tài khoản của bạn bị người khác đăng nhập vào, thứ mất trước là tiền. Bên kia còn dùng tài khoản của bạn để lừa những người trong danh bạ. Danh tính của bạn cũng coi như đã bị lấy mất. + +### 1. Bật xác thực hai lớp cho tất cả tài khoản email, thanh toán và mạng xã hội, ưu tiên xác nhận bằng cửa sổ bật lên trên điện thoại, mã tin nhắn chỉ để sau + +- Chi phí: Không tốn tiền. Mỗi tài khoản bật một lần, hai ba phút là xong. +- Nói thẳng: Xác thực hai lớp nghĩa là lúc đăng nhập, ngoài mật khẩu còn phải xác nhận thêm một lần xem có đúng là bạn không. Cách hiện cửa sổ bật lên trên điện thoại rồi bạn bấm xác nhận chặn được hơn chín phần mười các vụ đánh cắp tài khoản nhắm đường. Cách hỏi đáp những câu kiểu「lần trước bạn đăng nhập ở đâu」「email dự phòng của bạn là gì」thì chỉ chặn được khoảng một phần mười các vụ đánh cắp tài khoản nhắm đường. +- Lợi ích: Google thống kê 350.000 lần đánh cắp tài khoản thật (còn gọi là chiếm đoạt tài khoản). Một nhóm xác minh bằng thiết bị, ví dụ cửa sổ bật lên trên điện thoại để bạn bấm xác nhận, hoặc cắm một khóa bảo mật. Nhóm này chặn được「超过 94% 源于钓鱼的劫持尝试和 100% 的自动化劫持尝试」. Phishing là lừa bạn nhập mật khẩu vào trang web giả, còn tự động hóa là máy lấy mật khẩu bị lộ ra rồi thử hàng loạt. Nhóm còn lại xác minh bằng câu hỏi, ví dụ hỏi lần trước bạn đăng nhập ở đâu, email dự phòng là gì. Nhóm này「只拦下了低至 10% 的钓鱼劫持和 73% 的自动化劫持」 +- Mức bằng chứng: A +- Nguồn: Doerfler P, Thomas K, Marincenko M, et al. (2019). Evaluating Login Challenges as a Defense Against Account Takeover. The World Wide Web Conference (WWW '19). +- Ghi chú: Cùng nghiên cứu đó cũng thấy loại xác thực này đôi khi chặn nhầm chính người dùng thật. 52% người dùng thật lần đầu không đăng nhập được, nhưng cuối cùng 97% vẫn vào được. Hãy bật tính năng này cho email trước, vì phần lớn tài khoản khác đều dùng email để tìm lại mật khẩu. + +### 2. Đặt mật khẩu riêng cho email, không dùng lại ở bất kỳ trang nào khác + +- Chi phí: Không tốn tiền. Lưu vào trình quản lý mật khẩu thì không phải tự nhớ. Khó là bỏ được thói quen dùng một mật khẩu ở khắp mọi nơi. +- Nói thẳng: Mật khẩu bị lộ ở trang khác sẽ được dùng để đăng nhập thẳng vào email của bạn. Một khi email bị người ta đăng nhập vào, mọi tài khoản nào dùng email để tìm lại mật khẩu đều mất theo. Nên mật khẩu email phải là một mật khẩu riêng, chỗ khác không dùng lại. +- Lợi ích: Lấy tài khoản và mật khẩu bị lộ ở nơi khác rồi thử đăng nhập vào từng cái một, cách đó gọi là tấn công dồn mật khẩu, và là cách tấn công ít tốn công nhất. Email của bạn cũng bị thử kiểu đó. Một khi email bị người ta đăng nhập vào, mọi tài khoản dùng nó để tìm lại mật khẩu đều mất theo. Cục An ninh mạng và hạ tầng Hoa Kỳ khuyến nghị: mỗi tài khoản một mật khẩu mạnh khác nhau, ít nhất 16 ký tự, giao cho trình quản lý mật khẩu lưu +- Mức bằng chứng: C +- Nguồn: US CISA. Use Strong Passwords. +- Ghi chú: Không nhớ được thì dùng trình quản lý mật khẩu có sẵn trong trình duyệt. Nó giữ mật khẩu từng trang giúp bạn, tốt hơn nhiều so với dùng chung một mật khẩu khắp mọi nơi. Đừng lưu mật khẩu trong mục yêu thích của WeChat hay trong ứng dụng ghi chú. + +### 3. Đặt mật khẩu khóa màn hình điện thoại, đặt mã PIN cho thẻ SIM + +- Chi phí: Không tốn tiền. Mật khẩu khóa màn hình và mã PIN mỗi thứ đặt một lần là xong. +- Nói thẳng: Thẻ SIM là cái thẻ nhỏ trong điện thoại, mã xác minh tin nhắn nhận qua đó. Điện thoại bị mất, người nhặt được sẽ rút thẻ ra, cắm sang điện thoại khác để nhận mã xác minh, rồi lần lượt đặt lại các tài khoản của bạn. Đặt mã PIN cho thẻ thì khi thẻ cắm sang máy khác, muốn bật máy lên phải nhập mã trước, đường đó bị cắt. +- Lợi ích: Sau khi điện thoại bị mất, đường nhanh nhất của người nhặt được là cắm thẻ SIM vào điện thoại khác để nhận mã xác minh tin nhắn của bạn. Có mã rồi thì họ đặt lại từng tài khoản của bạn một cách. Thẻ SIM đã đặt mã PIN thì cắm sang máy khác muốn bật máy lên phải nhập mật khẩu trước, người nhặt được không dùng nó để nhận mã xác minh nữa +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp +- Ghi chú: Mã PIN nằm ở mục「khóa thẻ SIM」trong phần cài đặt của điện thoại. Mã ban đầu khi mới xuất xưởng thường là 1234 hoặc 0000. Nhập sai ba lần liên tiếp thì phải dùng mã PUK do nhà mạng cung cấp mới mở được. Nên sau khi đặt xong, ghi mã đó ra giấy trước. + +### 4. Điện thoại bị mất thì làm theo đúng thứ tự này: báo mất thẻ SIM, khóa từ xa, đổi mật khẩu, báo công an, phong tỏa thẻ ngân hàng + +- Chi phí: Không tốn tiền. Làm trọn bộ mất mười mấy phút. +- Nói thẳng: Thứ tự quan trọng hơn tốc độ. Bước đầu tiên là báo mất thẻ SIM, đường sống của mã xác minh bị cắt. Tiếp theo khóa điện thoại từ xa, rồi dùng máy tính đổi mật khẩu email và thanh toán, sau đó báo công an và lấy biên nhận, cuối cùng tùy tình hình mà phong tỏa thẻ ngân hàng. Điện thoại không ở trên người thì gọi số chăm sóc khách hàng của nhà mạng bằng điện thoại của người khác cũng báo mất được. +- Lợi ích: Thứ tự quan trọng hơn tốc độ. Bước đầu tiên, báo mất thẻ SIM, người khác không nhận được mã xác minh của bạn. Bước thứ hai, khóa điện thoại từ xa và xóa sạch nội dung trong máy. Bước thứ ba, đổi mật khẩu email và thanh toán từ máy tính. Bước thứ tư, báo công an và lấy biên nhận. Cuối cùng, phong tỏa thẻ ngân hàng khi cần. Ủy ban Truyền thông Liên bang Hoa Kỳ cũng nhắc: dù nghĩ chỉ là đánh rơi thì vẫn nên khóa từ xa. Bị trộm thì báo công an ngay, nêu rõ hãng và số IMEI (mã định danh của máy), và báo ngay cho nhà mạng +- Mức bằng chứng: C +- Nguồn: US FCC. Protect Your Smart Device. ；thứ tự các bước là kinh nghiệm của tác giả；làm lại giấy tờ tùy thân xem phần 7, vay với danh tính người khác xem mục về tín dụng ở phần 8 +- Ghi chú: Nhớ trước số chăm sóc khách hàng của ba nhà mạng: Mobile 10086, Unicom 10010, Telecom 10000. Số điện thoại của bạn đăng ký ở thành phố nào thì ghi lại luôn, nhân viên tổng đài sẽ hỏi. Gọi chăm sóc khách hàng bằng điện thoại của người khác vẫn báo mất được. + +### 5. Thẻ bị dùng ăn cắp: báo mất và phong tỏa trước, báo công an sau, rồi yêu cầu ngân hàng bồi thường; chứng minh「là chính bạn đã dùng thẻ」là trách nhiệm của ngân hàng + +- Chi phí: Không tốn tiền. Thấy trên thẻ có động tĩnh bất thường thì báo mất hoặc phong tỏa ngay. Giữ lại biên bản báo công an, biên nhận báo mất, và các thông báo giao dịch ngân hàng gửi tới. Nếu thẻ vẫn còn trong tay bạn thì tới gần nơi đó làm một giao dịch nhỏ như tra cứu số dư hay rút tiền, để lại chứng từ chứng minh lúc xảy ra việc thẻ thật nằm trong tay bạn. Khó là nhịn đừng cãi lại với nhân viên tổng đài trước, cứ báo mất trước đã. +- Nói thẳng: Khi thẻ bị người khác dùng ăn cắp, bạn không cần chứng minh「đây không phải tôi dùng」. Ngược lại, ngân hàng phải đưa ra chứng cứ chứng minh giao dịch là chính chủ thẻ đã thực hiện, không đưa ra được thì phải bồi thường cho bạn. Điều kiện là bạn báo mất và phong tỏa ngay khi phát hiện. Kéo dài không báo mất thì phần thiệt hại phát sinh thêm phải tự gánh chịu. +- Lợi ích: Quy định của Tòa án nhân dân tối cao tách rõ「ai phải đưa ra chứng cứ」. Bạn nói đây là thẻ giả bị sao chép, hoặc bị ăn cắp qua mạng, thì bạn phải đưa ra chứng cứ trước. Thẻ giả bị sao chép là thẻ đã bị người ta làm bản sao để dùng. Những thứ sau dùng để chứng minh được: văn bản pháp lý đã có hiệu lực, thẻ thật nằm ở đâu lúc giao dịch, giao dịch xảy ra ở đâu. Còn có sao kê giao dịch của tài khoản, thông báo giao dịch, biên bản báo công an, biên nhận báo mất. **Ngược lại, khi ngân hàng phát hành thẻ hoặc tổ chức thanh toán phi ngân hàng (thanh toán bên thứ ba) nói giao dịch là chủ thẻ tự thực hiện, hoặc được chủ thẻ ủy quyền, thì chúng phải đưa ra chứng cứ**. Sau khi bạn báo cho ngân hàng, nếu ngân hàng không kiểm tra kịp thời, hoặc không cung cấấp và lưu giữ kịp thời chứng từ giao dịch và hình ảnh giám sát, dẫn đến không lấy được chứng cứ, thì hậu quả không đưa ra được chứng cứ do ngân hàng gánh chịu. Sau khi xác định có căn cứ: chủ thẻ tín dụng (thẻ tiết kiệm) có thể yêu cầu ngân hàng phát hành trả lại tiền gửi và lãi tiền gửi bị dùng ăn cắp, đồng thời bồi thường tổn thất. Chủ thẻ tín dụng có thể yêu cầu hoàn lại khoản tiền vượng hạn bị trừ cùng lãi và phạt vi phạm hợp đồng, đồng thời bồi thường tổn thất; ngân hàng ngược lại yêu cầu bạn trả lại khoản tiền vượng hạn đó thì tòa không hỗ trợ. Bạn còn có thể yêu cầu ngân hàng phát hành kịp thời xóa bản ghi tín dụng xấu tương ứng（toàn quốc, có hiệu lực từ ngày 25 tháng 5 năm 2021） +- Mức bằng chứng: A +- Nguồn: Tòa án nhân dân tối cao (最高人民法院) (2021). 关于审理银行卡民事纠纷案件若干问题的规定（第四、五、七、十四、十五条）. +- Ghi chú: Có hai trường hợp bạn phải tự gánh chịu. Một là thẻ ngân hàng, mật khẩu, mã xác minh bạn giữ không cẩn thận, bản thân bạn có lỗi (nguyên văn là「未尽妥善保管义务具有过错」), lỗi bao nhiêu thì gánh tới bấy nhiêu. Vì vậy đừng nói mật khẩu cho người khác, đừng chuyển mã xác minh cho người khác (xem mục 1, xác thực hai lớp nên dùng xác nhận bằng cửa sổ trên điện thoại trước). Hai là không báo mất kịp thời, để tổn thất tiếp tục nới rộng, phần phát sinh thêm phải tự gánh. Vì vậy bước đầu tiên luôn là báo mất và phong tỏa, đừng gọi tổng đài mà cãi trước. Các tổ chức thanh toán bên thứ ba cũng áp dụng bộ quy tắc này. Trong tài liệu quảng bá của họ có ghi「先行赔付」, nếu cam kết cụ thể và rõ ràng thì có thể yêu cầu họ bồi thường trước theo đúng điều đó. Tiền bị người ta lừa bạn tự chuyển đi thì phải đi theo cách khác, xem mục 2 ở phần 8 (phát hiện bị lừa thì gọi ngay 110 hoặc 96110 yêu cầu phong tỏa thanh toán). + +### 6. Cứ một thời gian lại xem thiết bị đã đăng nhập và các ứng dụng đã cấp quyền của tài khoản, cái nào không dùng thì dọn đi + +- Chi phí: Không tốn tiền. Mỗi lần xem mất mấy phút. Khó là không có ai nhắc, phải tự nhớ để xem. +- Nói thẳng:「Thiết bị đã đăng nhập」là những chiếc điện thoại và máy tính hiện vẫn dùng được tài khoản của bạn. Người đánh cắp tài khoản thường nằm im một thời gian rồi mới ra tay. Trong danh sách mà hiện ra thiết bị bạn không nhận ra, hoặc có phần mềm đã lâu không dùng vẫn còn nối vào tài khoản của bạn, thì thấy là đăng xuất khỏi toàn bộ rồi đổi mật khẩu. +- Lợi ích: Khi đánh cắp tài khoản, người ta thường không ra tay ngay, họ nằm im một thời gian đã. Danh sách thiết bị đã đăng nhập trong tài khoản ghi lại những chiếc điện thoại và máy tính hiện còn dùng được tài khoản đó. Danh sách ứng dụng đã cấp quyền ghi lại những phần mềm bên ngoài mà bạn đã cho phép đăng nhập bằng tài khoản này. Thiết bị lạ trong danh sách, và phần mềm bên thứ ba đã lâu không dùng, là những dấu vết dễ phát hiện nhất +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp +- Ghi chú: WeChat, Alipay, email, tài khoản Apple và tài khoản Android đều có mục này. Thấy thiết bị không nhận ra thì bấm đăng xuất khỏi toàn bộ, rồi đổi mật khẩu. + +### 7. Đừng bấm「đồng ý tất cả」chỉ để dùng ứng dụng: với thông tin không bắt buộc, bạn không đồng ý thì họ không được phép từ chối dịch vụ + +- Chi phí: Không tốn tiền. Khó là nhịn không bấm「đồng ý tất cả」. +- Nói thẳng: Ứng dụng xin thông tin của bạn, nếu không phải thông tin bắt buộc để cung cấp dịch vụ đó, mà bạn không đồng ý thì nó không được vì thế mà không cho bạn dùng. Nó cũng chỉ được thu thập trong phạm vi cần thiết. Bản đồ cần vị trí của bạn là bắt buộc, đèn pin xin danh bạ thì không phải. +- Lợi ích: Luật ghi rõ hai điều. Một là không được vì cá nhân không đồng ý, hoặc rút lại sự đồng ý, mà từ chối cung cấp sản phẩm hay dịch vụ; trừ khi xử lý các thông tin đó là điều kiện bắt buộc để cung cấp dịch vụ. Hai là việc thu thập phải giới hạn ở phạm vi tối thiểu để đạt mục đích xử lý, chỉ được lấy những gì thật sự dùng đến +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). 个人信息保护法. 中国人大网. ：第六条「收集个人信息，应当限于实现处理目的的最小范围，不得过度收集个人信息」；第十六条「个人信息处理者不得以个人不同意处理其个人信息或者撤回同意为由，拒绝提供产品或者服务；处理个人信息属于提供产品或者服务所必需的除外」；第十五条「基于个人同意处理个人信息的，个人有权撤回其同意。个人信息处理者应当提供便捷的撤回同意的方式」 +- Ghi chú: Tiêu chí để đánh giá là thông tin đó có bắt buộc với dịch vụ này hay không. Bản đồ cần vị trí là bắt buộc, đèn pin xin danh bạ thì không. Cài ứng dụng xong, vào trang quyền ứng dụng trong phần cài đặt điện thoại và tắt những quyền không cần. Đến lúc thật sự dùng thì chọn chỉ cho phép một lần này. + +### 8. Bạn có quyền xem, sao chép, sửa và xóa thông tin cá nhân của mình, bị từ chối thì có thể kiện + +- Chi phí: Không tốn tiền. Chỉ khi bên kéo dài không làm thì mới cần khiếu nại hoặc kiện. Đi thật sự ra tòa thì tính từ mấy tháng trở lên, phí luật sư còn phải tự bỏ ra, nên khiếu nại trước hơn là hờn. Khó là khi bên kéo dài, phải đi đòi nhiều lần. +- Nói thẳng: Bạn có quyền yêu cầu doanh nghiệp cho bạn xem, sao chép, sửa và xóa thông tin của chính mình. Khi dịch vụ dừng, khi thời hạn lưu giữ hết, khi bạn rút lại sự đồng ý, doanh nghiệp vốn phải tự xóa. Nếu nó từ chối bạn thì phải nêu lý do; không làm thì bạn có thể kiện thẳng ra tòa. Đóng tài khoản và xóa thông tin là hai việc khác nhau, đóng tài khoản rồi vẫn phải yêu cầu xóa riêng. +- Lợi ích: Có vài trường hợp doanh nghiệp phải tự chủ động xóa: dịch vụ đã dừng, thời hạn lưu giữ đã hết, bạn rút lại sự đồng ý, mục đích thu thập ban đầu đã đạt được, và các trường hợp tương tự. Nó không xóa thì bạn có quyền yêu cầu nó xóa. Nó từ chối bạn thực hiện các quyền này thì phải nêu lý do. Bạn có thể trực tiếp kiện ra tòa +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). 个人信息保护法. 中国人大网. ：第四十五条「个人有权向个人信息处理者查阅、复制其个人信息……个人请求查阅、复制其个人信息的，个人信息处理者应当及时提供」；第四十六条更正、补充权；第四十七条列了五种应当主动删除的情形，含「（一）处理目的已实现、无法实现或者为实现处理目的不再必要」「（二）个人信息处理者停止提供产品或者服务，或者保存期限已届满」「（三）个人撤回同意」，「个人信息处理者未删除的，个人有权请求删除」；第五十条「个人信息处理者应当建立便捷的个人行使权利的申请受理和处理机制。拒绝个人行使权利的请求的，应当说明理由」「个人可以依法向人民法院提起诉讼」 +- Ghi chú: Đóng tài khoản và xóa thông tin cá nhân là hai việc khác nhau, sau khi đóng tài khoản vẫn phải yêu cầu xóa riêng. Trước khi đổi điện thoại hoặc bán điện thoại cũ, hãy đăng xuất hết các tài khoản và gỡ liên kết trên máy cũ, rồi mới khôi phục cài đặt gốc. Luật trao cho bạn quyền xóa sau sự việc, nó không lấy lại được thứ đã lộ ra ngoài. + +### 9. Quét mặt không phải thứ bạn buộc phải đồng ý: đã có cách khác thì không được chỉ bắt bạn quét mặt, bạn không đồng ý thì họ phải đưa ra cách khác + +- Chi phí: Không tốn tiền. Khi bị yêu cầu quét mặt, hãy hỏi một câu「có cách xác minh nào khác không」. Bên kia nói không thì yêu cầu họ đưa ra một cách. Khó là phải mở miệng hỏi trực tiếp. +- Nói thẳng: Chừng nào còn cách khác để làm cùng một việc đó, bên kia không được chỉ bắt bạn quét mặt. Bạn không đồng ý quét mặt thì họ phải đưa ra cách khác như quẹt thẻ, mật khẩu hay giấy tờ tùy thân, và cũng không được dùng câu「vậy thì không làm được giấy tờ」để ép bạn. Ở phòng khách sạn, phòng tắm công cộng, phòng thay đồ, nhà vệ sinh, không ai được lắp thiết bị nhận diện khuôn mặt. +- Lợi ích: 人脸识别技术应用安全管理办法 ghi rõ:「实现相同目的或者达到同等业务要求，存在其他非人脸识别技术方式的，不得将人脸识别技术作为唯一验证方式。个人不同意通过人脸信息进行身份验证的，应当提供其他合理、便捷的方式。」Quy chế còn ghi:「任何组织和个人不得以办理业务、提升服务质量等为由，误导、欺诈、胁迫个人接受人脸识别技术验证个人身份。」Khi xử lý thông tin khuôn mặt dựa trên sự đồng ý của bạn thì phải có「在充分知情的前提下自愿、明确作出的单独同意」: hỏi riêng về đúng việc này, và bạn gật đầu riêng thì mới tính. Bạn có quyền rút lại sự đồng ý, bên xử lý phải đưa ra cách rút lại thuận tiện. Xử lý thông tin khuôn mặt của người chưa đủ mười bốn tuổi phải có sự đồng ý của cha mẹ hoặc người giám hộ khác. Lắp thiết bị nhận diện khuôn mặt ở nơi công cộng thì phải「应当为维护公共安全所必需」, và phải đặt biển báo hiển thị rõ ràng. Bên trong các không gian riêng tư nằm trong nơi công cộng như phòng khách sạn, phòng tắm công cộng, phòng thay đồ công cộng, nhà vệ sinh công cộng, không tổ chức hay cá nhân nào được lắp. Thông tin khuôn mặt phải được lưu trong thiết bị nhận diện khuôn mặt, không được truyền ra ngoài qua mạng internet. Có hai trường hợp ngoại lệ: luật hoặc quy định hành chính có quy định khác, hoặc đã có sự đồng ý riêng（toàn quốc, có hiệu lực từ ngày 1 tháng 6 năm 2025） +- Mức bằng chứng: A +- Nguồn: Văn phòng Quốc gia Quản lý Thông tin Internet (国家互联网信息办公室) và Bộ Công an (公安部) (2025). 人脸识别技术应用安全管理办法（第 19 号令，第十条、十二条、十三条，2025 年 6 月 1 日起施行）. +- Ghi chú: Hay gặp nhất là cổng vào chung cư, nền tảng cho thuê nhà, phòng tập, khách sạn bắt bạn đăng ký khuôn mặt. Khi bên kia nói「hệ thống chỉ hỗ trợ quét mặt」thì đọc nguyên văn quy chế cho họ nghe. Nguyên văn là:「实现相同目的或者达到同等业务要求，存在其他非人脸识别技术方式的，不得将人脸识别技术作为唯一验证方式」. Rồi yêu cầu họ đưa ra cách khác như quẹt thẻ, mật khẩu hay giấy tờ tùy thân. Vẫn không đưa thì phản ánh với cơ quan quản lý thông tin mạng địa phương. Nhà nước có quy định riêng về xác minh danh tính bằng khuôn mặt, ví dụ một số tình huống tài chính và hành chính công, thì làm theo các quy định đó. Khác biệt lớn nhất giữa khuôn mặt và mật khẩu là khi đã lộ thì không đổi lại được, nên nó đáng được thận trọng hơn mật khẩu. Đơn vị lưu thông tin khuôn mặt của trên 100.000 người thì phải báo cáo với cơ quan quản lý thông tin mạng từ cấp tỉnh trở lên trong 30 ngày làm việc, đây cũng là một cách hỏi xem bên kia có đàng hoàng hay không. Quyền xem, sửa, xóa thông tin cá nhân của chính mình xem mục 8. diff --git "a/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" "b/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" new file mode 100644 index 00000000..6921442d --- /dev/null +++ "b/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" @@ -0,0 +1,86 @@ +[← 回总目录](../README.md) + +# 15. 租房与买房 + +口径：金钱。租房最容易亏钱的地方是押金和中介。买房最费钱的地方是贷款利息，还有卖方知道、你不知道的那些事。 + +### 1. Số tiền cọc, thời điểm hoàn trả và các trường hợp bị giữ lại phải ghi vào hợp đồng + +- Chi phí: Không tốn tiền. Lúc ký hợp đồng mất thêm mười phút. Chỗ khó là phải nêu thẳng trước mặt chủ nhà +- Nói thẳng: Bao nhiêu tiền cọc, khi nào được trả lại, trường hợp nào bị giữ lại, ba điều đó phải ghi vào hợp đồng trước khi ký. Lý do giữ tiền mà hợp đồng không ghi thì chủ nhà không được dùng để giữ tiền cọc của bạn khi bạn trả nhà. +- Lợi ích: Quy định hành chính viết rõ: 「出租人收取押金的，应当在住房租赁合同中约定押金的数额、返还时间以及扣减押金的情形等事项。除住房租赁合同约定的情形以外，出租人无正当理由不得扣减押金。」 — chủ nhà có thu tiền cọc thì phải ghi trong hợp đồng thuê nhà số tiền cọc, thời điểm hoàn trả và các trường hợp được giữ lại tiền cọc. Ngoài những trường hợp hợp đồng thuê nhà đã quy định, chủ nhà không có lý do chính đáng thì không được giữ tiền cọc. Ba việc phải ghi vào hợp đồng: thu bao nhiêu tiền cọc, khi nào trả lại, trường hợp nào được giữ lại. Lý do không có trong hợp đồng thì chủ nhà không được dùng để giữ tiền của bạn +- Mức bằng chứng: A +- Nguồn: Ủy ban Nhà nước (国务院) (2025). Quy định quản lý thuê nhà ở, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十条）. +- Ghi chú: Ngày trả nhà, chụp ảnh và quay video cùng chủ nhà. Ghi lại chỉ số đồng hồ nước, đồng hồ điện, đồng hồ gas, tường và sàn cũng chụp lại. Tiền cọc bị giữ mà không có lý do thì trước hết khiếu nại với cơ quan quản lý nhà ở. Số tiền không lớn thì kiện theo thủ tục vụ án nhỏ, xem phần 8 + +### 2. Có người cắt nước cắt điện, đổi ổ khóa, đến tận cửa đe dọa để đuổi bạn đi: báo công an và giữ chứng cứ, quy định cấm đúng những cách này + +- Chi phí: Không tốn tiền. Một cuộc gọi báo công an, làm được ngay tại chỗ +- Nói thẳng: Chủ nhà không được cắt nước cắt điện, đổi ổ khóa hay đến tận cửa đe dọa bạn để buộc bạn hủy hợp đồng hoặc dọn ra, quy định cấm rõ ràng. Gặp chuyện này thì báo công an trước. Biên bản xử lý hiện trường là chứng cứ để sau này bạn yêu cầu bồi thường. Đừng tự đấm lại đối phương. +- Lợi ích: Quy định hành chính viết rõ: 「出租人不得采取暴力、威胁或者其他非法方式迫使承租人解除住房租赁合同或者腾退租赁住房。」 — chủ nhà không được dùng bạo lực, đe dọa hay bất kỳ cách trái pháp nào khác để buộc người thuê hủy hợp đồng thuê nhà hoặc dọn ra khỏi nhà thuê. 腾退 nghĩa là bắt bạn dọn ra. Gặp trường hợp này thì báo công an trước và giữ lại chứng cứ. Đừng tự đấm lại đối phương +- Mức bằng chứng: A +- Nguồn: Ủy ban Nhà nước (国务院) (2025). Quy định quản lý thuê nhà ở, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十二条）. +- Ghi chú: Báo công an là để có biên bản xử lý, sau này yêu cầu bồi thường thì đó là chứng cứ. Đồng thời báo cơ quan quản lý nhà ở địa phương + +### 3. Môi giới không được thu hộ và chi hộ tiền thuê cùng tiền cọc, tiền phải vào thẳng tay chủ nhà + +- Chi phí: Không tốn tiền. Kiểm tra người nhận tiền trước khi trả, mất vài phút. Chỗ khó là giữ vững khi môi giới bảo bạn chuyển tiền vào cho nó +- Nói thẳng: Theo quy định, tiền thuê và tiền cọc không được môi giới thu hộ và chi hộ, tiền phải trả thẳng cho chủ nhà. Tiền đi qua tay môi giới một lần là thêm một lần nó ôm tiền rồi bỏ trốn. Nếu nó bỏ trốn thật, bạn mất trắng tiền thuê và chủ nhà vẫn đuổi bạn đi. Trước khi trả tiền, hãy kiểm tra người nhận có đúng là tên trên giấy chứng nhận quyền sở hữu không. +- Lợi ích: Quy định hành chính liệt kê 「代收、代付住房租金、押金」, tức thu hộ và chi hộ tiền thuê nhà và tiền cọc, vào những việc cơ quan môi giới bất động sản không được làm, nên môi giới không được thu tiền thuê và tiền cọc thay chủ nhà. Tiền đi qua tay môi giới một lần là thêm một lần nó ôm tiền rồi bỏ trốn +- Mức bằng chứng: A +- Nguồn: Ủy ban Nhà nước (国务院) (2025). Quy định quản lý thuê nhà ở, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第二十五条）. +- Ghi chú: Trước khi trả tiền, hãy kiểm tra người nhận có đúng là tên trên giấy chứng nhận quyền sở hữu không. Nếu không phải chính chủ thì hỏi rõ và xin một văn bản ủy quyền bằng văn bản. Môi giới chỉ được thu phí môi giới cho việc se tiếp hai bên + +### 4. Thuê dài hạn căn hộ có dịch vụ thì xem tài khoản giám sát tiền trước, đừng trả cả năm một lúc để mua rẻ + +- Chi phí: Không tốn tiền. Xem một lần mất hơn mười phút +- Nói thẳng: Trả cả năm một lúc chỉ được giảm giá chút ít. Công ty nào phá sản thì tiền cọc cùng mấy chục tháng tiền thuê còn lại mất sạch. Ràng buộc vay mua tiền thuê còn tệ hơn: căn hộ bỏ trốn thì khoản vay vẫn phải trả hết theo từng tháng. Trước khi ký, hãy xem họ đã công bố tài khoản giám sát tiền thuê chưa. +- Lợi ích: Quy định hành chính yêu cầu doanh nghiệp thuê nhà ở kinh doanh theo hình thức cho thuê lại 「设立住房租赁资金监管账户并向社会公示」, lập tài khoản giám sát tiền thuê nhà và công bố ra xã hội. Kinh doanh theo hình thức cho thuê lại là doanh nghiệp thuê nguyên căn từ chủ nhà rồi cho thuê lại cho bạn. Khoản giảm giá nhỏ khi trả cả năm một lúc không bù được rủi ro mất cả tiền cọc lẫn tiền thuê còn lại khi doanh nghiệp bỏ trốn. Ai đã ký thêm vay mua tiền thuê thì rủi ro càng lớn +- Mức bằng chứng: A +- Nguồn: Ủy ban Nhà nước (国务院) (2025). Quy định quản lý thuê nhà ở, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十九条）. +- Ghi chú: Lúc ký, xem có bị gắn khoản vay hay không. Nếu một ứng dụng trả góp đưa cả năm tiền một lượt cho căn hộ và bạn trả lại theo từng tháng, thì căn hộ bỏ trốn bạn vẫn phải trả tiếp. Loại hợp đồng đó đừng ký + +### 5. Nhà bị bán trong thời gian thuê, hợp đồng thuê vẫn có hiệu lực, không phải dọn + +- Chi phí: Không tốn tiền +- Nói thẳng: Nhà bị bán trong thời gian thuê thì hợp đồng thuê của bạn vẫn có hiệu lực. Chủ nhà mới phải cho bạn ở đến hết hạn theo hợp đồng cũ. Không phải dọn, cũng không phải thương lại giá. Điều kiện là bạn đang ở hợp pháp trong đó, nên hãy giữ hợp đồng, lịch sử chuyển khoản và bằng chứng về thời điểm bạn dọn vào. +- Lợi ích: Bộ luật Dân sự có một quy tắc tên là 「买卖不破租赁」, mua bán không phá vỡ hợp đồng thuê. Nguyên văn điều khoản là: 租赁物在租赁期限内发生所有权变动的，不影响租赁合同的效力 — vật thuê đổi chủ sở hữu trong thời hạn thuê thì không ảnh hưởng hiệu lực của hợp đồng thuê. Nghĩa là trong thời hạn thuê mà nhà đã đổi chủ thì hợp đồng đã ký vẫn còn hiệu lực. Chủ nhà mới phải cho bạn ở đến hết hạn theo hợp đồng cũ +- Mức bằng chứng: A +- Nguồn: Ủy ban Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự, 民法典（第七百二十五条）. +- Ghi chú: Điều kiện là bạn đã dọn vào ở hợp pháp, cách nói trong điều khoản là chiếm hưởng hợp pháp. Hãy giữ hợp đồng, lịch sử chuyển khoản và bằng chứng về thời điểm dọn vào + +### 6. Trước khi ký, kiểm tra giấy chứng nhận quyền sở hữu và tình trạng thế chấp, mọi khoản tiền đều chuyển khoản kèm ghi chú mục đích + +- Chi phí: Không tốn tiền. Xem một lần mất nửa tiếng +- Nói thẳng: Hai cách mất sạch tiền thường gặp nhất là trả tiền thuê cho người không phải chủ sở hữu, và thuê một căn nhà đã thế chấp cho ngân hàng từ trước rồi sau đó bị phong tỏa. Trước khi ký, hãy tra quyền sở hữu và thế chấp một lần tại trung tâm đăng ký bất động sản. Mọi khoản tiền đều chuyển khoản, ghi chú 「tiền thuê căn nhà X tháng Y」, khi có tranh chấp thì đó là chứng cứ trực tiếp. +- Lợi ích: Hai cách mất sạch tiền thường gặp nhất là trả tiền thuê cho người không phải chủ sở hữu, và thuê một căn nhà đã thế chấp cho ngân hàng từ trước rồi sau đó bị phong tỏa. Khi chuyển khoản, ghi chú 「tiền thuê căn nhà X tháng Y」, khi có tranh chấp thì đó là chứng cứ trực tiếp +- Mức bằng chứng: C +- Nguồn: kinh nghiệm của tác giả, không có tài liệu trực tiếp; cùng lập luận về giữ chứng cứ và ghi chú khi chuyển khoản, xem hai mục về tiền cước và giấy nợ trong phần 8 +- Ghi chú: Thông tin quyền sở hữu có thể tra tại trung tâm đăng ký bất động sản địa phương. Lúc ký, bảo chủ nhà trực tiếp đến dự, nếu không đến được thì xin một văn bản ủy quyền. Nếu chủ nhà thứ hai cho thuê lại cho bạn thì cần có văn bản đồng ý bằng văn bản của chủ nhà đầu tiên + +### 7. Nhà cũ mà môi giới thu hộ tiền mua nhà thì bắt buộc phải đi qua tài khoản tiền giao dịch riêng do môi giới mở ở ngân hàng + +- Chi phí: Không tốn tiền. Chỗ khó là kiên trì đòi dùng tài khoản riêng, có khi phải tranh cãi với môi giới mấy câu +- Nói thẳng: Tiền mua nhà là vài trăm nghìn tệ đến vài triệu tệ, là khoản đơn lẻ lớn nhất trong cuốn sách này. Khi môi giới thu hộ, theo quy định phải đi qua tài khoản riêng mà môi giới mở ở ngân hàng, không chuyển bằng WeChat vào tài khoản cá nhân của môi giới viên. Ngoài ra, hai môi giới cùng làm một hợp đồng chỉ được thu một khoản hoa hồng, còn thủ tục vay và thủ tục sang tên phải ký hợp đồng riêng và nói rõ chi phí trước. +- Lợi ích: Quy định viết rõ ràng: khi các bên tham gia giao dịch bất động sản thỏa thuận để cơ quan môi giới bất động sản thu hộ và chi hộ tiền giao dịch thì phải chuyển qua tài khoản tiền thanh toán giao dịch khách hàng chuyên dùng mà cơ quan đó mở ở ngân hàng. Nghĩa là tiền mua nhà phải vào tài khoản riêng có sự giám sát, không phải tài khoản của chính môi giới, cũng không phải tài khoản riêng của một môi giới viên nào +- Mức bằng chứng: A +- Nguồn: Bộ Xây dựng và Đô thị (住房城乡建设部), Ủy ban Cải cách và Phát triển Quốc gia và Bộ Nhân sự và Xã hội (人力资源社会保障部) (2011). Biện pháp quản lý môi giới bất động sản, 房地产经纪管理办法（令第 8 号，2016 年第 29 号令修改）第二十四条：「房地产交易当事人约定由房地产经纪机构代收代付交易资金的，应当通过房地产经纪机构在银行开设的客户交易结算资金专用存款账户划转交易资金。」 第十八条「房地产经纪服务实行明码标价制度……在经营场所醒目位置标明房地产经纪服务项目、服务内容、收费标准」；第十九条「两家或者两家以上房地产经纪机构合作开展同一宗房地产经纪业务的，只能按照一宗业务收取佣金，不得向委托人增加收费」；第十七条 về thủ tục vay, đăng ký bất động sản và các dịch vụ khác「应当向委托人说明服务内容、收费标准等情况，经委托人同意后，另行签订合同」. +- Ghi chú: Tiền mua nhà là vài trăm nghìn tệ đến vài triệu tệ, là khoản đơn lẻ lớn nhất trong cuốn sách này, đừng chuyển bằng WeChat cho môi giới viên. Còn hai kiểu thu thêm tiền thường gặp. Một là hai môi giới cùng làm một hợp đồng, kiểu này chỉ được thu một khoản hoa hồng. Hai là thủ tục vay và thủ tục sang tên, đó là dịch vụ khác, phải ký hợp đồng riêng và nói trước chi phí. Mọi khoản tiền đều chuyển khoản kèm ghi chú mục đích, xem mục 6 của phần này (trước khi ký, kiểm tra giấy chứng nhận quyền sở hữu và tình trạng thế chấp). + +### 8. Đừng thuê nhà bị chia ngăn: đơn vị cho thuê nhỏ nhất là căn phòng thiết kế ban đầu, bếp, nhà vệ sinh, ban công không được ở + +- Chi phí: Không tốn tiền. Cái giá phải trả là với cùng ngân sách thì hoặc ở xa hơn một chút, hoặc ở chật hơn một chút +- Nói thẳng: Theo quy định, đơn vị cho thuê nhỏ nhất là căn phòng thiết kế sẵn. Bếp, nhà vệ sinh, ban công và hầm chứa dưới lòng đất không được dùng để ở. Loại nhà này bị kiểm tra thì bạn có thể phải dọn ra. Ngoài ra, trong thời gian thuê chủ nhà cũng không được tự ý tăng giá thuê. +- Lợi ích: Quy định viết rõ ràng, đơn vị cho thuê nhỏ nhất là căn phòng thiết kế ban đầu, không được chia một phòng thành mấy phòng rồi cho thuê riêng. Diện tích xây dựng tính trên mỗi người không được thấp hơn tiêu chuẩn tối thiểu do địa phương quy định. Bếp, nhà vệ sinh, ban công và hầm chứa dưới lòng đất không được cho thuê để ở. Trong thời hạn thuê, chủ cho thuê cũng không được tự ý tăng giá thuê +- Mức bằng chứng: A +- Nguồn: Bộ Xây dựng và Đô thị (住房和城乡建设部) (2010). Biện pháp quản lý cho thuê nhà ở thương mại, 商品房屋租赁管理办法（令第 6 号）第八条：「出租住房的，应当以原设计的房间为最小出租单位，人均租住建筑面积不得低于当地人民政府规定的最低标准。厨房、卫生间、阳台和地下储藏室不得出租供人员居住。」 第九条：「出租人应当按照合同约定履行房屋的维修义务并确保房屋和室内设施安全……房屋租赁合同期内，出租人不得单方面随意提高租金水平。」 +- Ghi chú: Khi việc chia ngăn trái phép bị kiểm tra, người phải dọn ra là người thuê, tiền cọc và tiền thuê đã trả thường không lấy lại được. An toàn cũng kém: tường chia ngăn hay chặn lối thoát hiểm, một căn cho nhiều người ở lại dùng chung một dây điện. Khi xem nhà, hãy để ý ba điều: bức tường có phải xây thêm sau không, phòng có cửa sổ không, đồng hồ điện chịu được nhiều người như vậy không. + +### 9. Hộ khẩu thành phố muốn chuyển ra nông thôn ở thì chỉ thuê nhà nông thôn, đừng mua đất nền và nhà trên đất nền + +- Chi phí: Không tốn tiền. Cái giá phải trả là căn nhà không thuộc về bạn, một hợp đồng thuê dài nhất 20 năm, hết hạn phải thương lại với chủ nhà +- Nói thẳng: Người có hộ khẩu thành phố không được ra nông thôn mua đất nền, nhà của nông dân hay 「小产权房」, văn bản của nhà nước cấm rõ ràng. Người trong lành bảo 「chỉ cần viết hợp đồng là được」 cũng không được, tiền đã đưa thì căn nhà vẫn không thuộc về bạn. Muốn ở nông thôn lâu dài thì ký hợp đồng thuê, một lần tối đa 20 năm, hết hạn rồi ký tiếp. +- Lợi ích: Văn bản của Văn phòng Ủy ban Nhà nước (国务院办公厅) viết rõ ràng: 「城镇居民不得到农村购买宅基地、农民住宅或『小产权房』。」 — cư dân thành phố không được ra nông thôn mua đất nền, nhà ở của nông dân hay 「小产权房」. Cùng văn bản đó còn quy định, nông dân đã bán hoặc cho thuê nhà ở thì đơn xin đất nền sau đó không được duyệt. Văn bản năm 2019 của Văn phòng Ban Trung ương công tác nông thôn và Bộ Nông nghiệp và Nông thôn nhắc lại: 「严禁城镇居民到农村购买宅基地。」 — nghiêm cấm cư dân thành phố ra nông thôn mua đất nền. Cùng văn bản đó nêu con đường hợp pháp: cư dân thành phố có thể thuê nhà nông thôn để ở hoặc để kinh doanh, thời hạn hợp đồng thuê không được quá hai mươi năm, hết hạn thì hai bên có thể thỏa thuận khác +- Mức bằng chứng: A +- Nguồn: Văn phòng Ủy ban Nhà nước (国务院办公厅) (2007). Thông báo về việc thực hiện nghiêm các quy phạm pháp và chính sách liên quan đến đất đai phục đồng nông thôn, 关于严格执行有关农村集体建设用地法律和政策的通知（国办发〔2007〕71 号）. ；Văn phòng Ban Trung ương công tác nông thôn và Bộ Nông nghiệp và Nông thôn (中央农村工作领导小组办公室、农业农村部) (2019). Thông báo về việc tăng cường quản lý đất nền nông thôn, 关于进一步加强农村宅基地管理的通知（中农发〔2019〕11 号）phần năm và phần sáu. +- Ghi chú: Đất nền là mảnh đất làng phân cho dân của chính làng đó để xây nhà, và chỉ phân cho người trong làng. Khi thuê nhà nông thôn, hãy ký hợp đồng bằng văn bản với hộ nông dân cho thuê, ghi rõ thời hạn thuê, tiền thuê, có được cải tạo hay xây thêm không, và tiền cải tạo tính thế nào khi hết hạn. Văn bản quy định thời hạn thuê không quá 20 năm, đừng ký loại 「thuê 50 năm」「thuê 70 năm」, và đừng trả tiền thuê nhiều năm một lúc. Văn bản còn cấm xuống nông thôn dùng đất nền xây biệt thự sân và hội câu lạc tư nhân, đừng bỏ tiền cho dân làng 「hợp xây」. diff --git "a/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" new file mode 100644 index 00000000..ba189ec8 --- /dev/null +++ "b/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -0,0 +1,86 @@ +[← 回总目录](../README.md) + +# 16. 得了慢性病之后怎么活 + +口径：总死亡率与金钱。查出慢性病以后往往还有几十年要过，本节算的是这几十年怎么管。 + +### 1. Uống thuốc đủ theo đơn của bác sĩ, đừng thấy khá hơn là ngừng + +- Chi phí: Không tốn tiền, ngược lại còn tiết kiệm tiền chữa biến chứng về sau. Mỗi ngày uống thuốc chiếm chưa tới vài phút, chỗ khó là ngày nào cũng uống đủ. +- Nói thẳng: Người uống đủ thuốc theo lời bác sĩ có khả năng chết trong cùng khoảng thời gian chỉ bằng khoảng một nửa người uống thuốc đại đoạn. Con số này phải tính giảm. Cả người uống đủ cả thuốc giả cũng có tỷ lệ tử vong thấp, nghĩa là một phần lợi ích đến từ 「người kiên trì uống thuốc vốn đã chú ý sức khoẻ hơn」. Nhưng cái hại của việc ngừng thuốc thì chắc chắn, đặc biệt với thuốc hạ huyết áp, thuốc chống đông máu và thuốc chống thải ghép. +- Lợi ích: 21 nghiên cứu, 46847 người gộp lại (phân tích tổng hợp): so với người uống thuốc đại đoạn, người uống đủ theo đơn có khả năng chết trong cùng khoảng thấp hơn khoảng 44%. Giá trị gốc là tỷ số khả năng 0.56, 95% CI 0.50–0.63 (khoảng tin cậy). Chỉ tính những điều trị có tác dụng thật thì thấp hơn khoảng 45% (0.55, 0.49–0.62) +- Mức bằng chứng: A +- Nguồn: Simpson SH, Eurich DT, Majumdar SR, et al. (2006). A meta-analysis of the association between adherence to drug therapy and mortality. BMJ, 333(7557), 15. +- Ghi chú:Tranh cãi. Đây cũng là con số cần tính giảm nhất trong cuốn sách. Trong cùng phân tích tổng hợp đó, cả người uống đủ cả thuốc giả (placebo) cũng có tỷ lệ tử vong thấp (0.56, 0.43–0.74). Điều đó cho thấy phần lớn lợi ích đến từ 「người kiên trì uống thuốc vốn đã chú ý sức khoẻ hơn」. Tác giả gọi yếu tố bị trộn vào đó là hiệu ứng người tuân thủ khỏe mạnh. Nhưng cái hại của bản thân việc ngừng thuốc thì rõ ràng, đặc biệt với thuốc hạ huyết áp, thuốc chống đông máu và thuốc chống thải ghép + +### 2. Làm giấy công nhận bệnh mạn tính ngoại trú trước, rồi đăng ký khám chữa tại nơi khác, để huyết áp cao, tiểu đường, xạ trị hoá chất, lọc máu và chống thải ghép được quyết toán ngay tại nơi khác + +- Chi phí: Không tốn tiền. Trước khi đi, làm một lần đăng ký khám chữa bệnh tại nơi khác trên ứng dụng bảo hiểm y tế, làm xong ngay trên điện thoại. +- Nói thẳng: Huyết áp cao, tiểu đường, xạ trị hoá chất ngoại trú cho ung thư ác tính, lọc máu cho người bị suy thận giai đoạn cuối, chống thải ghép sau cấy ghép cơ quan, năm loại này khám ngoại trú ở nơi khác cũng dùng thẳng bảo hiểm y tế được. Không phải tự bỏ tiền trước rồi về nơi tham gia bảo hiểm để hoàn. Nhưng phải làm giấy công nhận bệnh mạn tính ngoại trú ở nơi tham gia bảo hiểm trước, rồi mới đăng ký khám chữa tại nơi khác. Hai việc đó làm đủ thì mới có tác dụng. +- Lợi ích: Văn bản của Cục Quản lý bảo hiểm y tế quốc gia (国家医保局) nói là đã làm được bước này. Nguyên văn là 「基本实现高血压、糖尿病、恶性肿瘤门诊放化疗、尿毒症透析、器官移植术后抗排异治疗等 5 种门诊慢特病费用跨省直接结算县域可及」. 「县域可及」 nghĩa là ở cấp huyện cũng làm được. Văn bản còn trên nền 5 loại này lại bổ sung thêm 5 loại +- Mức bằng chứng: A +- Nguồn: Văn phòng Cục Quản lý bảo hiểm y tế quốc gia và Văn phòng Bộ Tài chính (国家医保局办公室、财政部办公厅) (2024). Thông báo về việc mở rộng có trật tự và an toàn danh mục bệnh mạn tính ngoại trú được quyết toán trực tiếp liên tỉnh, 关于稳妥有序扩大跨省直接结算门诊慢特病病种范围的通知（医保办发〔2024〕19 号）. +- Ghi chú: Điều kiện là làm đủ cả hai việc: làm giấy công nhận bệnh mạn tính ngoại trú ở nơi tham gia bảo hiểm trước, rồi đăng ký khám chữa tại nơi khác. Năm loại bệnh bổ sung đã mở đến đâu thì mỗi nơi tiến độ khác nhau, cứ theo nơi bảo hiểm y tế địa phương công bố + +### 3. Tái khám theo khoảng cách bác sĩ dặn, ghi chỉ số mỗi lần vào cùng một quyển sổ + +- Chi phí: Mỗi lần xét nghiệm vài chục đến vài trăm tệ. Đi bệnh viện một chuyến cộng thời gian chờ kết quả thì mất một ít thời gian. Chỗ khó là cứ sau một khoảng lại phải đi một lần, dễ hoãn lần này sang lần khác. +- Nói thẳng: Bệnh mạn tính cần xem sự thay đổi trong một khoảng thời gian, không phải con số của một lần khám. Hãy ghi ngày, chỉ số, loại thuốc và liều đang dùng vào cùng một quyển sổ. Khi chuyển bệnh viện thì lấy ra, không phải làm lại xét nghiệm và bỏ tiền lần nữa. +- Lợi ích: Bệnh mạn tính cần xem chỉ số thay đổi thế nào trong một khoảng thời gian, chỉ nhìn một lần thì chưa đủ. Ghi chung một chỗ còn có một lợi ích nữa: khi chuyển bệnh viện không phải xét nghiệm lại, cũng không phải bỏ tiền lần nữa +- Mức bằng chứng: C +- Nguồn: kinh nghiệm của tác giả, không có tài liệu trực tiếp +- Ghi chú: Mỗi lần ghi tối thiểu bốn thứ: ngày, giá trị chỉ số, thuốc đang uống, liều thuốc. Chụp ảnh phiếu xét nghiệm bằng điện thoại cũng được, nhưng phải để cùng một chỗ, đừng rải mỗi thứ một nơi + +### 4. Đừng bỏ điều trị chính thức chỉ để thử phương thuốc dân gian hay thực phẩm bảo vệ sức khoẻ + +- Chi phí: Không tốn tiền, còn tiết kiệm tiền mua thực phẩm bảo vệ sức khoẻ. Không mất thời gian. Chỗ khó là giữ vững khi người khác liên tục khuyên bạn thử. +- Nói thẳng: Cái hại của việc ngừng thuốc thì chắc chắn, lợi ích của phương thuốc dân gian và thực phẩm bảo vệ sức khoẻ thì không chắc. Đó là đổi một tổn thất chắc chắn lấy một lợi ích không chắc. Muốn thử thực phẩm bổ sung thì nói với bác sĩ điều trị trước. Có thực phẩm bổ sung làm thay đổi nồng độ thuốc chống đông máu và thuốc chống thải ghép trong máu. +- Lợi ích: Cái hại của việc ngừng thuốc thì chắc chắn, lợi ích của phương thuốc dân gian thì không chắc. Cuốn sách này liệt kê một loạt loại sản phẩm thường vô dụng ở phần 6 +- Mức bằng chứng: C +- Nguồn: kinh nghiệm của tác giả, không có tài liệu trực tiếp; bằng chứng liên quan xem phần 6 +- Ghi chú: Muốn thử thực phẩm bổ sung thì nói với bác sĩ điều trị trước. Có thực phẩm bổ sung tương tác với thuốc kê đơn, chẳng hạn làm thay đổi nồng độ thuốc chống đông máu và thuốc chống thải ghép trong máu + +### 5. Bệnh mạn tính đã ổn định thì ở bệnh viện cộng đồng một lần lấy được thuốc dùng 12 tuần + +- Chi phí: Không tốn tiền. Đến trạm y tế cộng đồng hỏi một câu có kê được đơn dài hạn không, chỉ mất một câu. +- Nói thẳng: Người bệnh đã ổn định, thuốc cũng không đổi, ở trạm y tế cộng đồng một lần lấy đủ thuốc dùng 12 tuần. Bỏ được nửa ngày mỗi tháng đi bệnh viện lớn xếp hàng khám. Trong cùng nơi tham gia bảo hiểm, tỷ lệ hoàn tiền khi khám ở bệnh viện tuyến cơ sở thường cao hơn bệnh viện tuyến trên khoảng mười điểm phần trăm. +- Lợi ích: Không phải mỗi tháng đi bệnh viện lớn xếp hàng lấy thuốc. Trong cùng nơi tham gia bảo hiểm (còn gọi là khu vực điều hòa), tỷ lệ hoàn tiền khi nằm viện ở bệnh viện tuyến cơ sở về nguyên tắc cao hơn tuyến trên khoảng 10 điểm phần trăm +- Mức bằng chứng: A +- Nguồn: Văn phòng Ủy ban Nhà nước (国务院办公厅) (2026). Một số biện pháp thúc đẩy xây dựng hệ thống chẩn đoán điều trị phân tầng. ：「对于符合条件的慢性病患者，基层医疗卫生机构单次可开具不超过12周用药的长期处方。」 — với người bị bệnh mạn tính đủ điều kiện, cơ sở y tế tuyến cơ sở có thể kê một lần đơn dài hạn tối đa 12 tuần thuốc.；cùng văn bản 「原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距」, và yêu cầu bệnh viện tuyến trên mở phòng khám bệnh mạn tính thường gặp như huyết áp cao, tiểu đường, bệnh phổi tắc nghẽn mạn tính tại tuyến cơ sở +- Ghi chú: 「符合条件」 do bác sĩ đánh giá, thường là người đã chẩn đoán rõ và phác đồ dùng thuốc cũng ổn định. Thuốc vẫn là thuốc đó, cái tiết kiệm là nửa ngày chạy bệnh viện mỗi tháng và phí khám. Ở tuyến cơ sở không có loại thuốc bạn cần, có thể hỏi xem có đăng ký báo thiếu thuốc và vận chuyển không. + +### 6. Trước khi đăng ký bác sĩ gia đình ở cộng đồng, hỏi rõ cái nào vào bảo hiểm y tế, cái nào tự trả + +- Chi phí: Không tốn tiền. Hỏi thêm một câu trước khi ký, mất một phút. +- Nói thẳng: Gói dịch vụ cơ bản trong hợp đồng đăng ký đi qua bảo hiểm y tế theo quy định. Tiền của gói dịch vụ cá nhân hoá thì bạn phải tự trả toàn bộ. Khi có người chào bán 「gói nâng cấp」, hãy hỏi rõ đó là gói cơ bản hay gói cá nhân hoá, đã đăng ký hay chưa, một năm bao nhiêu tiền. +- Lợi ích: Gói dịch vụ cơ bản theo quy định do bảo hiểm y tế trả. Tiền của gói dịch vụ cá nhân hoá thì bạn trả toàn bộ, gói này còn phải đăng ký với cơ quan y tế và sức khoẻ cấp huyện +- Mức bằng chứng: A +- Nguồn: Văn phòng Ủy ban Nhà nước (国务院办公厅) (2026). Một số biện pháp thúc đẩy xây dựng hệ thống chẩn đoán điều trị phân tầng. ：「加强基层门诊付费与签约服务政策联动，基本服务包按规定纳入医保支付；个性化服务包由签约基层医疗卫生机构按程序向县级卫生健康部门备案，费用由个人支付。」 — tăng cường liên kết giữa thanh toán khám ngoại trú tuyến cơ sở và chính sách dịch vụ theo hợp đồng đăng ký, gói dịch vụ cơ bản được đưa vào bảo hiểm y tế trả theo quy định; gói dịch vụ cá nhân hoá do cơ sở y tế tuyến cơ sở đã ký đăng ký với cơ quan y tế và sức khoẻ cấp huyện theo thủ tục, chi phí do cá nhân trả. +- Ghi chú: Giá trị của việc đăng ký nằm ở chỗ có người theo dõi bệnh của bạn lâu dài, không nằm ở chỗ gói có bao nhiêu mục. Khi có người chào bán 「gói nâng cấp」, hãy hỏi ba câu: đây là gói cơ bản hay gói cá nhân hoá, đã đăng ký chưa, một năm bao nhiêu tiền. + +### 7. Xác định tiểu đường thì đi khám đáy mắt một lần, sau đó khám theo khoảng cách bác sĩ dặn; mỗi năm khám chân một lần + +- Chi phí: Một lần khám đáy mắt có giãn đồng tử vài chục đến vài trăm tệ. Khám chân thường đã nằm trong tiền khám, không phải trả thêm. Chỗ khó là lúc mắt và chân chưa khó chịu gì vẫn phải nhớ đi khám. +- Nói thẳng: Tiểu đường làm hỏng đáy mắt, làm hỏng chân, gần như không có triệu chứng gì trước khi hỏng tới mức không nhìn thấy và loét. Sớm phát hiện thì còn chữa được. Người bị tiểu đường typ 2 nên khám đáy mắt có giãn đồng tử một lần ngay lúc xác định chẩn đoán, người typ 1 bắt đầu khám 5 năm sau khi bệnh khởi phát. Người đã khám một hai lần không thấy vấn đề và đường huyết cũng đạt chuẩn thì có thể giãn ra một đến hai năm một lần. Chân thì khám mỗi năm một lần. +- Lợi ích: Tiểu đường làm hỏng đáy mắt gọi là bệnh lõ mạch võng mạc, làm hỏng chân gọi là chân tiểu đường. Hai thứ này gần như không có triệu chứng trước khi có chuyện. Sớm phát hiện thì còn chữa được. Đợi tới lúc không nhìn thấy nữa hoặc chân bắt đầu loét thì đã muộn +- Mức bằng chứng: A +- Nguồn: Hiệp hội Đái thái Hoa Kỳ (美国糖尿病学会) (2026). Tiêu chuẩn chẩn đoán và điều trị đái thái 2026·chương 12 về bệnh lõ mạch võng mạc, bệnh thần kinh và chăm sóc bàn chân. Diabetes Care：「People with type 2 diabetes should have an initial dilated and comprehensive eye examination by an ophthalmologist or optometrist at the time of the diabetes diagnosis.」「Adults with type 1 diabetes should have an initial dilated and comprehensive eye examination by an ophthalmologist or optometrist 5 years after the onset of diabetes.」「If there is no evidence of retinopathy from one or more annual eye exams and glycemic indicators are within the goal range, then screening every 1–2 years may be considered.」「Perform a comprehensive foot evaluation at least annually to identify risk factors for ulcers and amputations.」 +- Ghi chú: Người typ 2 phải khám ngay lúc xác định chẩn đoán, người typ 1 bắt đầu khám 5 năm sau khi bệnh khởi phát. Khác biệt là typ 2 thường đã tăng âm thầm mấy năm. Người đã khám một hai lần không thấy vấn đề và đường huyết cũng đạt chuẩn thì có thể giãn ra 1 đến 2 năm một lần. Người cảm giác ở chân bị tê dần thì mỗi ngày tự nhìn lòng bàn chân, soi bằng gương cũng được. + +### 8. Đã từng bị sỏi thận thì uống nước mỗi ngày 2,5–3 lít, hạ muối xuống dưới 6 g + +- Chi phí: Không tốn tiền. Biến việc uống nước bất cứ lúc nào thành thói quen. Ngày ra nhiều mồ hôi thì còn phải uống nhiều hơn, chỗ khó là ngày nào cũng nhớ. +- Nói thẳng: Người lần đầu bị sỏi canxi mà không làm gì thì trong năm năm có khoảng một phần tư tái phát. Người uống đủ nước mỗi ngày thì trong năm năm chỉ tái phát hơn một phần mười, giảm hơn một nửa. Hướng dẫn của Anh đưa ra lượng là mỗi ngày uống 2.5 đến 3 lít nước, muối không quá 6 g. +- Lợi ích: Một thử nghiệm đối chứng kéo dài 5 năm, chia ngẫu nhiên thành hai nhóm, trên 199 người lần đầu bị sỏi canxi mà không tìm ra nguyên nhân (còn gọi là sỏi canxi nguyên ốc). Một nhóm uống ít nhất 2 lít nước mỗi ngày, nhóm kia không can thiệp gì. Trong 5 năm, nhóm uống nhiều nước tái phát 12/99 (12.1%), nhóm không can thiệp 27/100 (27.0%), P=0.008. P càng nhỏ càng cho thấy khác biệt giữa hai nhóm không phải ngẫu nhiên. Trung bình bao lâu lại tái phát: 38.7±13.2 tháng so với 25.1±16.4 tháng, P=0.016. Lượng nước tiết trong 24 giờ đo lúc đầu ở người bị sỏi thấp hơn hẳn người không bị sỏi: nam giới là 1,057±238 mL so với 1,401±562 mL (P<0.0001), nữ giới là 990±230 mL so với 1,239±440 mL (P<0.001). Hướng dẫn NICE NG118 điều 1.8.1 khuyên người trưởng thành uống 2.5 đến 3 lít nước mỗi ngày, trẻ em và thanh niên uống 1 đến 2 lít theo tuổi. Pha thêm nước chanh tươi vào, đừng uống nước có ga, người trưởng thành uống muối không quá 6 g mỗi ngày +- Mức bằng chứng: A +- Nguồn: Borghi L, Meschi T, Amato F, Briganti A, Novarini A, Giannini A (1996). Urinary volume, water and recurrences in idiopathic calcium nephrolithiasis: a 5-year randomized prospective study. The Journal of Urology, 155(3), 839-843. ；Viện Quốc gia về Chất lượng và Chăm sóc Sức khoẻ (National Institute for Health and Care Excellence) (2019). Renal and ureteric stones: assessment and management. NICE guideline NG118, điều 1.8.1. +- Ghi chú: Tái phát giảm khoảng 55%, lại không tốn tiền, theo cách xếp hạng của cuốn sách thì đây là mục hiệu quả chi phí cao nhất của phần này. Cái cần theo dõi là lượng nước tiết trong 24 giờ, mục tiêu là trên 2 đến 2,5 lít, không phải nhìn vào lượng đã uống vào. Người ra nhiều mồ hôi, người làm việc trong môi trường nóng thì phải uống nhiều hơn. Thử nghiệm đó làm trên người bị sỏi canxi nguyên ốc lần phát đầu tiên. Thành phần sỏi không chỉ một loại, sỏi axit uric, sỏi nhiễm trùng và sỏi xystin mỗi loại có cách xử lý riêng. Người đã từng tái phát thì nên làm một lần phân tích thành phần sỏi và xét nghiệm nước tiểu 24 giờ, rồi mới quyết định có thêm thuốc hay không. Đừng kiêng hẳn canxi chỉ để phòng sỏi, xem mục 21 ở phần 6. Khi đau quặn lưng bụng dữ dội thì phải đến bệnh viện, uống nhiều nước không đẩy được. Máu tiểu không đau thì còn có thứ khác cần xét, xem mục 27 ở phần 1 + +### 9. Xác định bệnh gout thì uống thuốc hạ acid uric dài hạn, hạ acid uric máu xuống dưới 360 µmol/L rồi giữ nguyên ở đó + +- Chi phí: Allopurinol vài tệ đến vài chục tệ mỗi tháng. Lúc mới đầu mỗi vài tuần xét nghiệm acid uric máu để chỉnh liều, ổn định rồi thì mỗi vài tháng xét nghiệm một lần. 3 đến 6 tháng đầu còn phải uống kèm một thuốc phòng ngừa cơn. Chỗ khó là lúc không đau vẫn phải uống liên tục. +- Nói thẳng: Gout cần uống thuốc hạ acid uric dài hạn, hạ acid uric xuống dưới 360 µmol/L, không thể chỉ uống thuốc giảm đau lúc đau. Người bệnh được y tá chỉnh liều theo kết quả xét nghiệm thì sau hai năm có 95% đạt chuẩn, còn người khám theo kiểu cũ chỉ 30%. Đạt chuẩn rồi vẫn không được ngừng thuốc. Nhịn ăn không hạ được xuống mức đó, bỏ rượu cũng chỉ giảm 1.6 mg/dL. +- Lợi ích: Một thử nghiệm đối chứng chia ngẫu nhiên thành hai nhóm, gồm 517 người lớn từng lên cơn trong 12 tháng trở lại. Một nhóm do y tá phụ trách, giải thích rõ cho bệnh nhân bệnh là gì, rồi chỉnh acid uric theo chỉ số cho tới khi đạt chuẩn. Nhóm kia khám như cũ với bác sĩ đa khoa. Sau 2 năm, tỷ lệ acid uric máu hạ xuống dưới <360 µmol/L (6 mg/dL) là 95% so với 30%. Khả năng đạt chuẩn gấp 3.18 lần nhóm thường (RR 3.18, 95% CI 2.42–4.18, đây là khoảng tin cậy của con số), P<0.0001. Ở năm thứ 2, tần suất lên cơn, u tophi và chất lượng sống, vài chỉ số phụ này cũng tốt hơn nhóm thường. Mỗi năm tuổi thọ tăng thêm, tính cả phần quy đổi theo chất lượng sống (số năm sống điều chỉnh theo chất lượng sống), tốn £5,066. Cùng thử nghiệm đó còn làm nghiên cứu theo dõi, phát bảng hỏi cho 438 người, thu về 82%. Cả hai nhóm đều tự báo cáo số lần lên cơn trong 12 tháng qua. Nhóm đạt chuẩn có trung vị 0 lần (nửa số người ở giữa rơi vào 0–0 lần). Nhóm thường là 1 lần (0–3), P<0.001. Khoảng trong ngoặc này gọi là khoảng tứ phân vị. Tỷ lệ vẫn còn uống thuốc hạ acid uric, nhóm đạt chuẩn gấp 1.19 lần nhóm thường (RR hiệu chỉnh 1.19, 1.09–1.30). Hướng dẫn gout 2020 của Hiệp hội Thấp khớp Hoa Kỳ mạnh mẽ khuyến nghị bắt đầu điều trị hạ acid uric cho ba nhóm người: người có u tophi dưới da, người chụp chiếu thấy gout làm xói mòn xương, người lên cơn ≥2 lần mỗi năm. Cách làm là theo dõi liên tục acid uric máu, chỉnh liều theo kết quả cho tới khi đạt chuẩn, mục tiêu <6 mg/dL. Thuốc lựa chọn đầu tiên là allopurinol, người bị bệnh thận mạn tính từ giai đoạn 3 trở lên cũng dùng nó trước, liều khởi đầu ≤100 mg/ngày. Lúc bắt đầu cho kèm thuốc chống viêm để phòng ngừa, uống ít nhất 3 đến 6 tháng. Cùng hướng dẫn đó còn dẫn một nhóm bệnh nhân được theo dõi. Sau khi kiểm soát tốt trong thời gian dài, những người này ngừng thuốc, và sau khi ngừng acid uric máu vẫn giữ <7 mg/dL. Trong số họ, người không lên cơn lần nào trong suốt 5 năm theo dõi chỉ chiếm 13% (27/211). Về ăn uống, người hạn chế rượu hoặc bỏ rượu có acid uric máu thấp hơn người không hạn chế 1.6 mg/dL, một phần bia làm acid uric máu tăng 0.16 mg/dL. Ăn uống lành mạnh, ăn theo kiểu Địa Trung Hải và ăn theo DASH thì hiệu quả càng nhỏ hơn. Có khác biệt về việc chọn thuốc hạ acid uric nào là thuốc đầu tiên. Hướng dẫn của Hiệp hội Thấp khớp Hoa Kỳ mạnh mẽ khuyến nghị allopurinol là thuốc đầu tiên, lý do là khi dùng đủ liều thì có tác dụng, dung nạp tốt, an toàn và rẻ. Hướng dẫn còn khuyến nghị có điều kiện: người đang uống febuxostat mà có tiền sử tim mạch, hoặc vừa có biến cố tim mạch, thì nếu có thuốc khác để thay thế nên thay. Khuyến nghị này khớp với cảnh báo hộp đen mà FDA gắn cho febuxostat. Một căn cứ là thử nghiệm CARES, 6190 người bị gout kèm bệnh tim mạch được chia ngẫu nhiên uống febuxostat hoặc allopurinol. Rủi ro tử vong toàn nguyên ở nhóm febuxostat gấp 1.22 lần nhóm allopurinol (HR 1.22, 95% CI 1.01–1.47, cao hơn khoảng 22%). Tử vong do tim mạch gấp 1.34 lần (1.03–1.73, cao hơn khoảng 34%). Thử nghiệm FAST ở châu Âu cho kết luận khác. Thử nghiệm này gồm 6128 người bệnh trên 60 tuổi có yếu tố nguy cơ tim mạch, sự kiện tim mạch chính ở febuxostat không nhiều hơn allopurinol (HR 0.85, 0.70–1.03). Tỷ lệ tử vong là 7.2% ở nhóm febuxostat và 8.6% ở nhóm allopurinol. Người uống thuốc làm tăng đào thải acid uric qua thận, Hiệp hội Thấp khớp Hoa Kỳ không khuyến nghị kiềm hóa nước tiểu, vì không có bằng chứng cho là có hiệu quả và độ chắc chắn của bằng chứng rất thấp. Hướng dẫn đồng thời yêu cầu dặn những người này uống nhiều nước, và người đã biết có sỏi thận hoặc bệnh thận mạn tính từ giai đoạn 3 trở lên thì không được dùng loại thuốc này +- Mức bằng chứng: A +- Nguồn: Doherty M, Jenkins W, Richardson H, et al. (2018). Efficacy and cost-effectiveness of nurse-led care involving education and engagement of patients and a treat-to-target urate-lowering strategy versus usual care for gout: a randomised controlled trial. Lancet, 392(10156), 1403-1412. ；Abhishek A, Jenkins W, La-Crette J, Fernandes G, Doherty M (2020). Nurse-led care is preferred over GP-led care of gout and improves gout outcomes: results of Nottingham Gout Treatment Trial follow-up study. Rheumatology, 59(3), 575-579. ；FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760. ；White WB, Saag KG, Becker MA, et al.; CARES Investigators (2018). Cardiovascular Safety of Febuxostat or Allopurinol in Patients with Gout. New England Journal of Medicine, 378(13), 1200-1210. ；Mackenzie IS, Ford I, Nuki G, et al.; FAST Study Group (2020). Long-term cardiovascular safety of febuxostat compared with allopurinol in patients with gout (FAST): a multicentre, prospective, randomised, open-label, non-inferiority trial. Lancet, 396(10264), 1745-1757. ；Chi nhánh Nội tiết của Hiệp hội Y khoa Trung Quốc (中华医学会内分泌学分会) (2020). Hướng dẫn chẩn đoán và điều trị tăng axit uric máu và gout tại Trung Quốc (2019). 中华内分泌代谢杂志, 36(1), 1-13. +- Ghi chú:Tranh cãi: nên dùng thuốc hạ acid uric nào thì hướng dẫn của các nước khác nhau. Hướng dẫn Hoa Kỳ chọn allopurinol làm thuốc đầu tiên, người có bệnh tim mạch thì đổi được thì đừng dùng febuxostat, nhưng thử nghiệm FAST ở châu Âu không thấy febuxostat nguy hiểm hơn. Ở trong nước vẫn mua được benzbromarone, đó là thuốc làm tăng đào thải acid uric. Hướng dẫn của Hiệp hội Y khoa Trung Quốc (bản 2019 và bản cập nhật 2024) có khuyến nghị riêng về chọn thuốc và về kiềm hóa nước tiểu. Khi uống benzbromarone có cần thêm natri hydrocarbonat để kiềm hóa nước tiểu không thì hướng dẫn Hoa Kỳ cho là không có bằng chứng, cứ theo ý bác sĩ của bạn, uống nhiều nước thì không được bỏ. **Người Hán trước khi dùng allopurinol nên xét HLA-B\*5801**. Đây là một mã gen bẩm sinh phụ trách phản ứng miễn dịch, xét một lần thì kết quả không đổi suốt đời. Người mang gen này chiếm 7.4% trong người Hán, người Hàn Quốc và người Thái, còn người da trắng và người gốc Tây Ban Nha chỉ 0.7%. Người châu Á và người không thuộc chủng tộc Phi sau khi dùng allopurinol có nguy cơ bị hội chứng quá mẫn với allopurinol gấp 3 lần người da trắng, bệnh này nặng thì da bong tróc diện rộng, có thể đe dọa tính mạng. Hướng dẫn gout 2020 của Hiệp hội Thấp khớp Hoa Kỳ khuyến nghị với hai nhóm này là trước khi dùng thuốc có thể cân nhắc xét trước. **Mới bắt đầu hạ acid uric lại dễ kích thích cơn hơn**, nên hướng dẫn yêu cầu rõ ràng phải uống kèm một thuốc chống viêm phòng ngừa trong 3 đến 6 tháng đầu. 「Uống thuốc hạ acid uric xong đau hơn nên bỏ thuốc」 là cách thất bại phổ biến nhất của việc này. Đau lên thì phải làm là tăng thuốc chống viêm theo đơn bác sĩ, không phải ngừng thuốc hạ acid uric. Ngưỡng đạt chuẩn mỗi nơi viết khác nhau: <6 mg/dL quy đổi khoảng 360 µmol/L, phiếu xét nghiệm ở Trung Quốc thường dùng µmol/L. Thăm khám thấy acid uric cao mà chưa từng lên cơn là chuyện khác, xem mục 19 ở phần 6. Về đồ uống có đường và rượu, xem mục 7 và 20 ở phần 2 (không uống đồ uống có đường, uống ít rượu hoặc không uống) diff --git "a/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" "b/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" new file mode 100644 index 00000000..d2176d9a --- /dev/null +++ "b/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" @@ -0,0 +1,83 @@ +[← 回总目录](../README.md) + +# 17. 家里有老人 + +口径：金钱与人身自由。家里有老人，多数麻烦不是钱不够，是没提前把话说清楚、把字签好。最后一条讲长期卧床的压疮，按死亡率算账，不和本节其他条目比较。 + +[← Về mục lục](../README.md) + +# 17. Nhà có người già + +Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn rắc rối không phải vì thiếu tiền, mà là chưa nói cho rõ và chưa ký từ trước. Mục cuối nói về loét áp lực cho người nằm lâu, tính theo tỷ lệ tử vong, không so với các mục khác trong phần này. + +### 1. Trong khi người già còn minh mẫn, hãy viết giấy chỉ định trước người giám hộ tương lai + +- Chi phí: Đi công chứng một bản, vài trăm tệ. Nói với người già một lần là định được. Khó ở chỗ phải mở lời: phải nói ra thành lời cái ngày ông ấy không còn tự lo cho mình được nữa +- Nói thẳng: Trong khi người già vẫn còn minh mẫn, bỏ vài trăm tệ ra công chứng một bản văn bản viết tay, ghi rõ sau này ai làm người giám hộ cho ông ấy. Không có bản đó, đến lúc ông ấy không tự quyết định được nữa thì ai thay mặt ký, ai quản tiền của ông ấy, sẽ do toà án chỉ định. Người trong nhà không đồng ý với nhau thì còn kéo dài thêm. +- Lợi ích: Bộ luật Dân sự quy định, người trưởng thành còn minh mẫn, tự quyết định được (pháp luật gọi là 「người trưởng thành có đầy đủ năng lực hành vi dân sự」), có thể thỏa trước với người khác về việc sau này ai làm người giám hộ cho mình. Người thỏa thuận có thể là thân thích, cũng có thể là cá nhân hoặc tổ chức khác muốn làm người giám hộ. Thỏa xong phải 「chỉ định người giám hộ của mình bằng văn bản」. Về sau ông ấy không tự quyết định được nữa, hoặc chỉ quyết định được một phần (pháp luật gọi là 「mất hoặc mất một phần năng lực hành vi dân sự」), thì người đã ghi trong văn bản đứng ra quản. Không có bản văn bản này, về sau ai thay mặt ký, ai quản tiền của ông ấy, phải đi qua toà án chỉ định +- Mức bằng chứng: A +- Nguồn: Ủy ban Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự, 民法典（第三十三条）. +- Ghi chú: Bản văn bản này gọi là giám hộ tự nguyện bằng văn bản, tức người do mình chỉ định từ trước. Người này đứng trước thứ tự giám hộ mặc định mà pháp luật đặt ra. Người già sống một mình, con cái ở nơi khác, nhà có quan hệ phức tạp thì càng cần làm trước. Nên đi công chứng ở văn phòng công chứng + +### 2. Hãy lập di chúc, nhớ bản lập sau gạt bản lập trước, di chúc công chứng không còn được ưu tiên + +- Chi phí: Tự tay viết một bản di chúc, không tốn tiền. Đi công chứng, vài trăm đến hơn nghìn tệ. Khó ở chỗ phải nói chuyện với người nhà về chuyện sau khi mình chết +- Nói thẳng: Pháp luật công nhận sáu loại di chúc, nhưng viết sai hình thức là cả bản vô hiệu, đó là chỗ hỏng phổ biến nhất. Loại tự tay viết thì chính ông ấy phải viết tay từ đầu đến cuối, ký tên, ghi rõ năm tháng ngày. Lập mấy bản mà nội dung mâu thuẫn nhau thì lấy bản lập sau cùng. Từ năm 2021, bản công chứng cũng không còn tự động đứng trước nữa. +- Lợi ích: Bộ luật Dân sự công nhận sáu hình thức di chúc: tự tay viết, nhờ người viết thay, in bằng máy, ghi âm ghi hình, nói miệng, công chứng. Mỗi hình thức có thủ tục bắt buộc riêng (pháp luật gọi là 「yếu tố hình thức」). Nếu lập mấy bản mà nội dung mâu thuẫn thì lấy bản lập sau cùng. Từ năm 2021, bản công chứng không còn tự động đứng trước nữa +- Mức bằng chứng: A +- Nguồn: Ủy ban Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千一百三十四条至第一千一百四十二条）. +- Ghi chú: Di chúc tự tay viết phải do chính người viết viết tay từ đầu đến cuối, ký tên, ghi rõ năm tháng ngày. Di chúc in bằng máy cần hơn hai người chứng kiến. Người lập di chúc và người chứng kiến phải ký vào từng trang và ghi rõ năm tháng ngày. Người thừa kế, người được ghi vào di chúc để chia đồ, và thân thích của họ đều không được làm người chứng kiến. Viết sai hình thức là cả bản vô hiệu, đó là chỗ hỏng phổ biến nhất + +### 3. Tiền của người già để riêng một tài khoản, chi tiêu lớn thì đặt quy tắc hai người xác nhận + +- Chi phí: Không tốn tiền. Nói chuyện với người nhà một lần, đặt ra một con số là được. Khó ở chỗ mở lời thế nào, đừng để người già thấy bạn đang đề phòng ông ấy +- Nói thẳng: Kẻ lừa người già mua thực phẩm chức năng, đồ sưu tầm, sản phẩm tài chính thì cách nào cũng như nhau: trước hết dành vài tháng làm quen, rồi thu mẻ một lượt. Thỏa trước một quy tắc: chi tiêu vượt một con số thì phải báo với ai đó trước. Việc này đẩy thời điểm móc ví lùi lại, qua được lúc bốc đồng nhất. +- Lợi ích: Các vụ lừa người già mua thực phẩm chức năng, đồ sưu tầm, sản phẩm tài chính đều dùng một cách: dành vài tháng tích đủ niềm tin, rồi thu mẻ một lượt. Thỏa trước 「chi tiêu vượt một con số thì phải báo với ai đó trước」. Việc này đẩy quyết định móc ví lùi lại, qua được lúc bốc đồng nhất +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; về nguyên tắc chống lừa đảo chung xem phần 8 và phần 13 +- Ghi chú: Đừng nói chuyện này là 「sợ ông bị lừa」. Hãy nói là 「tiền trong nhà ghi sổ chung」, người già dễ chấp nhận hơn. Cũng đừng giao hết tiền của người già cho con cái giữ, như thế sẽ nảy sinh một loại tranh chấp khác + +### 4. Cho người già một câu để nói ra bất cứ lúc nào mà thoát khỏi việc + +- Chi phí: Không tốn tiền. Nói chuyện với người già một lần, vài phút là đủ +- Nói thẳng: Bán hàng và lừa đảo đều dựa vào áp lực tại chỗ, khiến người ta ngại từ chối ngay trước mặt. Thỏa trước với người già một câu: 「Tôi phải về hỏi người nhà đã」. Ông ấy nói ra lúc nào cũng được, ngay tại chỗ có chỗ để xuống. Cách này công dụng hơn là giải thích đạo lý sau đó. +- Lợi ích: Bán hàng và lừa đảo đều dựa vào gây áp lực tại chỗ. Thỏa trước câu 「tôi phải về hỏi người nhà đã」cho người già có chỗ để xuống. Hiệu quả hơn là giải thích đạo lý sau đó +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp +- Ghi chú: Dùng cùng quy tắc xác nhận của hai người ở mục 3 của phần này (tiền của người già để riêng một tài khoản). Việc chống ngã trong nhà làm luôn tiện tay, xem phần 1. Câu này để người già có chỗ xuống, để ông ấy khỏi phải cãi lại ngay tại chỗ, không phải để đề phòng ông ấy. Bị đám bán hàng vây, khó nhất là mở lời từ chối trước mặt cả một căn phòng, còn khó hơn là phân biệt thật giả + +### 5. Mấy thứ「đầu tư dưỡng già」yêu cầu người già nộp tiền trước thì đừng đụng: mua thẻ, mua giường, mua căn hộ dưỡng già, dưỡng già du lịch, mua sản phẩm cho người già đều là một kiểu huy động vốn trái phép + +- Chi phí: Không tốn tiền. Đọc mấy cái tên dưới đây cho người già nghe một lượt, vài phút. Sau này gặp thì từ chối thẳng: không đi nghe giảng, không đi tham quan, không để lại số điện thoại. Khó ở chỗ nếu người già đã đi mấy lần, đã quen người thì khó khuyên lại +- Nói thẳng: Mấy món dưỡng già yêu cầu người già nộp tiền trước mà lại hứa sinh lời thì đều là một lối lừa như nhau: mua thẻ, mua giường, mua căn hộ dưỡng già, dưỡng già du lịch, mở hội nghị bán sản phẩm cho người già. Theo quy định, tiền mất kiểu này chỉ tự mình chịu. Có người thu 136 triệu tệ nhờ đặt giường căn hộ dưỡng già, hại hơn 2.100 người. Chỉ bước móc ví là chặn được. +- Lợi ích: Bộ Dân vụ liệt kê năm dạng biểu hiện, nhận cái tên đáng tin hơn nhận mặt người. Dạng thứ nhất, mượn danh tổ chức dịch vụ dưỡng già: không có thực thể, thuê tạm một chỗ, tuyên bố đã ký hợp tác phối hợp với cơ sở dưỡng già để 「đội lốt」bên thật. Dạng thứ hai, dưới danh cung cấp 「dịch vụ dưỡng già」: bán 「thẻ quý khách」「thẻ hội viên」「thẻ trả trước」, bắt người ta trả trước 「phí dịch vụ dưỡng già」, dịch vụ hứa hẹn rõ ràng vượt quá năng lực cung ứng giường của chính mình. Dạng thứ ba, dưới danh đầu tư 「dự án dưỡng già」. Cái tên thường gặp là góp vốn vào cơ sở dưỡng sinh dưỡng già và bán căn hộ dưỡng già không có thật. Còn có cho thuê giường dưỡng dài hạn và 「ngân hàng thời gian」cùng nhau dưỡng già. Cái cờ tung ra là bán hoàn vốn, thuê lại sau bán, cam định mua lại. Dạng thứ tư, dưới danh bán 「sản phẩm cho người già」: mua lại hàng, gửi bán hộ, tiêu dùng hoàn lợi, kiểm tra sức khoẻ miễn phí, tặng quà, bán hàng qua hội nghị, hội thảo dưỡng sinh, khám miễn phí của chuyên gia. Dạng thứ năm, dưới danh hưởng 「dưỡng già du lịch」: du lịch giá rẻ hoặc miễn phí, nạp tiền hoàn lợi, tích điểm dưỡng già. Tiền không đòi lại được thì luật đã ghi rõ. Điều hai mươi lăm của Bộ luật Phòng ngừa và xử lý huy động vốn trái phép nêu 「因参与非法集资受到的损失，由集资参与人自行承担」, tiền mất chỉ tự mình chịu. Trong các vụ cơ quan công bố, Đới Mỗ Bình và những người khác huy động vốn trái phép 136 triệu tệ dưới danh đặt giường căn hộ dưỡng già, nạn nhân hơn 2.100 người, Đới Mỗ Bình bị phán 15 năm về tội lừa đảo huy động vốn, còn bị phạt 400.000 tệ. Tào Mỗ Minh huy động vốn lừa đảo 13.207 triệu tệ dưới danh đầu tư ngành dưỡng già, liên quan hơn 110.000 người, bị phán chung thân, tước quyền chính trị suốt đời, tịch thu toàn bộ tài sản cá nhân. Bán thực phẩm chức năng qua hội nghị thì khép tội lừa đảo. Nhi Mỗ Mung dùng yến tiệc miễn phí, tặng quà, xe riêng đưa đón, lôi người già đi nghe giảng. Diễn giả dùng 「nhân thiết giả, ca giả, mặc cả hỏng」bịa ra công dụng phòng chống ung thư. Người già bị dụ mua 「二氢槲皮素」(dihydroquercetin) với giá gấp nhiều lần vốn. Doanh số hơn 620.000 tệ. Nhi Mỗ Mung bị phán 10 năm 3 tháng, còn bị phạt 120.000 tệ （toàn quốc, bộ luật có hiệu lực từ ngày 1 tháng 5 năm 2021） +- Mức bằng chứng: A +- Nguồn: Bộ Dân vụ (民政部) (2023). Cảnh báo rủi ro về huy động vốn trái phép trong lĩnh vực dưỡng già, 关于养老服务领域非法集资的风险提示. （đăng trên Trang web chính phủ Trung Quốc）；Bộ Dân vụ (民政部) và Ủy ban Giám sát Ngân hàng và Bảo hiểm Trung Quốc (中国银保监会) (2021). Cảnh báo rủi ro về huy động vốn trái phép trong lĩnh vực dưỡng già, 关于养老领域非法集资的风险提示. ；Quốc vụ viện (国务院) (2021). Bộ luật Phòng ngừa và xử lý huy động vốn trái phép（số 737 của Quốc vụ viện, Điều hai, hai mươi lăm và ba mươi）, 防范和处置非法集资条例（国务院令第 737 号，第二、二十五、三十条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). Các vụ án điển hình của cơ quan kiểm sát đập phá và chấn chỉnh tội phạm lừa đảo nhằm vào người già, 检察机关打击整治养老诈骗犯罪典型案例. ；Tòa án nhân dân tối cao (最高人民法院) (2025). Các vụ án điển hình trừng trị lừa đảo liên quan sinh kế theo pháp luật（vụ năm, vụ lừa đảo của Nhi Mỗ Mung）, 依法惩治涉民生领域诈骗犯罪典型案例（案例五 倪某荣诈骗案）. +- Ghi chú: Phán đoán không dựa vào 「cái này trông có vẻ chính quy」, mà dựa vào 「có bắt nộp trước không, có hứa trả lời không」. Cơ sở dưỡng già vốn là bán dịch vụ, trong lời mà nói đến đầu tư, hoàn lợi, thu hồi vốn thì đừng tin. Bị lôi đi nghe giảng, đi du lịch, đi khám miễn phí thì bản thân không vi phạm pháp, cũng rất khó chặn. Chặn được là bước móc ví, dùng cùng quy tắc xác nhận của hai người ở mục 3. Nếu đã nộp tiền rồi, thì gọi ngay 110 hoặc 96110 yêu cầu ngừng chi trả theo mục 2 ở phần 8, đồng thời báo việc cho bộ phận đầu mối ở địa phương phụ trách xử lý huy động vốn trái phép. Dưỡng già dựa vào nhà là lối đi khác, xem mục 6 （bảo hiểm ngược chiều thế chấp nhà ở cho người già） + +### 6. Trừ bảo hiểm ngược chiều thế chấp nhà ở cho người già của công ty bảo hiểm, mấy thứ「dưỡng già dựa vào nhà」còn lại đừng đụng, tuyệt đối không thế chấp nhà đi mua sản phẩm tài chính + +- Chi phí: Không tốn tiền. Sổ đỏ trong nhà đừng giao cho bất kỳ ai đến tận cửa dưới danh 「dưỡng già dựa vào nhà」. Các thỏa thuận thế chấp, bảo đảm, vay nợ thì trước khi ký đưa con cái xem một lượt. Khó ở chỗ người đến tận cửa thường rất nhiệt tình, đi lại nhiều chuyến, người già ngại từ chối +- Nói thẳng: Dưỡng già dựa vào nhà chính quy chỉ có một loại, tức bảo hiểm ngược chiều thế chấp nhà ở của công ty bảo hiểm, vẫn đang thí điểm. Loại bán tận cửa thì chẳng liên quan gì, là lừa người già thế chấp nhà, rồi dùng tiền vay được mua sản phẩm tài chính mà nó giới thiệu. Có người đến lúc nhà bị thế chấp cũng không biết, cuối cùng mất nhà, lại còn gánh khoản vay. +- Lợi ích: Bộ phận bảo vệ quyền lợi người tiêu dùng của Ủy ban Giám sát Ngân hàng và Bảo hiểm Trung Quốc đã nói rõ hai loại thật giả. Dưỡng già dựa vào nhà chính quy chỉ có một loại, gọi là 「bảo hiểm ngược chiều thế chấp nhà ở cho người già」. Người già có bất động sản hợp pháp đầy đủ, đem nhà thế chấp cho công ty bảo hiểm. Nhà vẫn do mình ở, mình dùng, mình thu tiền thuê. Muốn bán hay muốn xử lý thì phải được người nhận thế chấp （công ty bảo hiểm）đồng ý. Người già nhận lương hưu theo thỏa thuận, cho đến khi mất. Nghề này hiện vẫn ở giai đoạn thí điểm, người làm khá ít. Người xấu nói 「dưỡng già dựa vào nhà」, với bảo hiểm ngược chiều thế chấp nhà ở mà nhà nước đang thí điểm thì 「hoàn toàn không liên quan」, là 「mượn chính sách nhà nước để tạo thế tuyên truyền cho hoạt động huy động vốn trái phép」. Cách làm là lừa người già làm thế chấp nhà, rồi dùng tiền vay mua sản phẩm tài chính nó giới thiệu, 「thường là Ponzi lấy tiền mới trả nợ cũ」. Hậu quả bản văn viết là 「có người tham gia thậm chí còn không biết nhà mình bị thế chấp, cuối cùng mất nhà, lại còn gánh nợ」. Trong bản cảnh báo còn đưa một đường để xem mức sinh lời. Nguyên lời là 「理财产品如果承诺收益率超过 6% 就要打问号，超过 8% 就很危险，10% 以上就要准备损失全部本金，『保本高收益』就是金融诈骗」, tức sản phẩm tài chính mà hứa mức sinh lời trên 6% thì phải nghi, trên 8% thì rất nguy hiểm, trên 10% thì phải chuẩn bị mất hết vốn gốc, 「bảo toàn vốn lợi nhuận cao」chính là lừa đảo tài chính. Cùng bản cảnh báo đó còn yêu cầu 「chú trọng khâu ký kết hợp đồng, không ký vào hợp đồng trắng」 （toàn quốc, tháng 6 năm 2021） +- Mức bằng chứng: B +- Nguồn: Bộ phận bảo vệ quyền lợi người tiêu dùng của Ủy ban Giám sát Ngân hàng và Bảo hiểm Trung Quốc (中国银保监会消费者权益保护局) (2021). Cảnh báo về cảnh giác với lừa đảo tài chính dưới danh 「đầu tư dưỡng già」「dưỡng già dựa vào nhà」, 关于警惕「投资养老」「以房养老」金融诈骗的风险提示（2021 年第三期）. （Sở Tài chính địa phương tỉnh Hồ Nam chuyển lại）；Bộ Dân vụ (民政部) và Ủy ban Giám sát Ngân hàng và Bảo hiểm Trung Quốc (中国银保监会) (2021). Cảnh báo rủi ro về huy động vốn trái phép trong lĩnh vực dưỡng già（dạng thứ tư: huy động vốn trái phép dưới danh tuyên bố 「dưỡng già dựa vào nhà」）, 关于养老领域非法集资的风险提示（第四种表现形式：以宣称「以房养老」为名非法集资）. +- Ghi chú: Chấm mức B là vì những lời này xuất phát từ bản cảnh báo rủi ro của cơ quan giám quản và từ phán đoán kinh nghiệm, không phải số liệu nghiên cứu. Ba đường 6%、8%、10% là đường tham chiếu cơ quan giám quản đưa cho người tiêu dùng, không phải tiêu chuẩn pháp luật đặt ra. Ủy ban Giám sát Ngân hàng và Bảo hiểm Trung Quốc đã hợp nhất vào Tổng cục Giám sát Tài chính quốc gia, trang phát hành gốc không mở được, ở đây trích trang chuyển lại của sở giám quản tài chính tỉnh. Mất tiền trên sản phẩm tài chính còn là chuyện nhỏ, sau khi nhà bị thế chấp người già vẫn còn nợ một khoản vay, tổn thất lớn hơn vốn gốc nhiều. Quy tắc chung về ký kết và hợp đồng trắng xem mục 17 ở phần 8, sau khi bị lừa làm sao để ngừng chi trả xem mục 2 ở phần 8 + +### 7. Người già trong nhà nằm liệt giường lâu ngày hoặc mất khả năng tự chăm sóc nặng, hãy tới cơ quan bảo hiểm y tế nơi tham gia bảo hiểm để xin bảo hiểm chăm sóc dài hạn; nó không chỉ dành cho người già + +- Chi phí: Không tốn tiền. Nộp một hồ sơ tại cơ quan bảo hiểm y tế nơi tham gia bảo hiểm. Làm thêm một lần đánh giá mức mất khả năng. Ai đánh giá lần đầu đã đạt thì phí đánh giá do quỹ chi trả, không phải tự bỏ ra +- Nói thẳng: Bảo hiểm chăm sóc dài hạn trả giúp bạn phí thuê người chăm sóc. Nếu trong nhà có người mất khả năng tự chăm sóc nặng, kéo dài trên nửa năm, hãy tới cơ quan bảo hiểm y tế nơi tham gia bảo hiểm để xin. Phí chăm sóc không cần tạm ứng trước, người tham gia bảo hiểm theo diện công nhân viên được trả khoảng bảy phần mười, người tham gia bảo hiểm theo diện dân cư không có việc làm được trả khoảng năm phần mười, tiền giường và tiền ăn thì không trả. Nó không có ngạch tuổi, người trẻ mất khả năng tự chăm sóc nặng cũng xin được +- Lợi ích: Chi phí dịch vụ chăm sóc dài hạn đủ điều kiện không đặt mức đồng chi trả. Mức đồng chi trả nghĩa là phải tự ứng trước đủ một con số rồi quỹ mới bắt đầu hoàn, ở đây không phải ứng trước. Người tham gia bảo hiểm theo chính sách công nhân viên được quỹ hoàn khoảng 70%. Người tham gia bảo hiểm theo chính sách dân cư không có việc làm được hoàn khoảng 50%. Số tiền cao nhất được hoàn trong một năm (hạn mức chi trả tối đa hằng năm) không vượt quá 50% thu nhập bình quân của dân cư không có việc làm ở khu vực chi trả thống nhất của địa phương trong năm trước +- Mức bằng chứng: A +- Nguồn: Cục Quản lý bảo hiểm y tế quốc gia và bảy cơ quan khác (2026). 关于印发《加快建立长期护理保险制度实施方案》的通知. : đối tượng bảo hiểm là 「按规定参保缴费且失能状态长期持续（一般为6个月以上），经申请通过评估认定的失能人员」, 「制度起步阶段保障重度失能人员」; 「待遇享受不设起付标准。符合规定的长期护理服务费用，按未就业城乡居民参保政策参保的，基金支付比例为50%左右；按单位职工参保政策参保的，基金支付比例为70%左右，退休人员享受单位职工参保待遇」; 「参保人员基金年度最高支付限额不超过统筹地区上年度城乡居民人均可支配收入的50%」; 「待遇享受固定等待期原则上按照6个月设置」; xem thêm Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước 《关于加快建立长期护理保险制度的意见》. +- Ghi chú: Mất khả năng phải 「长期持续（一般为6个月以上）」, nằm viện mấy tuần thì chưa đủ điều kiện xin. Trước khi được hưởng còn phải qua 6 tháng chờ. Các khoản không tính là chăm sóc như tiền giường, tiền ăn thì bảo hiểm chăm sóc dài hạn không trả. Phần chi phí y tế vốn phải do bảo hiểm y tế trả thì bảo hiểm chăm sóc dài hạn cũng không trả. Người đang nhận tiền chăm sóc dưỡng về tai nạn lao động thì không được hưởng trùng. Khoản tiền này và khoản trợ cấp chăm sóc trong hai khoản trợ cấp dành cho người khuyết tật là hai thủ tục, hai khoản tiền, không xung đột, xem mục 8 ở phần 7 (hai khoản trợ cấp). Chăm sóc tại nhà và tại cộng đồng được ưu ái hơn trong tỷ lệ hoàn, không nhất thiết phải vào cơ sở. Chính sách này từng địa phương triển khai theo lô, tiêu chuẩn đánh giá mức mất khả năng cũng khác nhau. Gọi 12393 hoặc số của cơ quan thực hiện bảo hiểm y tế nơi tham gia bảo hiểm, hỏi địa phương đã bắt đầu chưa và đặt lịch đánh giá thế nào. Đánh giá không đạt thì cũng đừng bỏ cuộc, khi tình trạng mất khả năng nặng hơn thì có thể xin lại. + +### 8. Khi trong nhà có người nằm liệt giường lâu ngày, hãy coi loét áp lực là kẻ thù số một: dùng nệm hơi điện, xoay người đúng giờ, mỗi ngày soi một lần chỗ xương nhô ra + +- Chi phí: Nệm hơi điện dùng trong nhà loại luân phiên áp lực, vài trăm tệ trở lên. Mỗi lần xoay người chỉ mất vài phút. Chỗ khó là cứ hai ba tiếng phải xoay một lần, ban đêm cũng không dừng, một người không gánh nổi +- Nói thẳng: Người già gãy xương hông mà đã có loét áp lực trước khi mổ thì khả năng tử vong trong 30 ngày cao hơn khoảng hai phần mười. Đổi sang nệm hơi điện thì người mắc loét mới ít hơn khoảng sáu phần mười. Xoay người đúng giờ và làm thật sự đúng thì ít hơn khoảng bảy phần mười. Đừng để người bệnh nằm sấp lâu. Mỗi ngày soi một lần xương cùng, gót chân, hông, vai lưng, thấy vết đỏ ấn không mất màu thì trong ngày gọi y tá hoặc bác sĩ +- Lợi ích: 19520 bệnh nhân phẫu thuật gãy xương hông (NSQIP Hoa Kỳ, 2016 đến 2019). Người đã có loét áp lực trước khi mổ thì tỷ lệ tử vong trong 30 ngày cao hơn. Sau khi tính cả các yếu tố khác thì mối liên hệ này vẫn còn (OR 1.2, nguy cơ tăng 21%, P=0.004). Những người đó còn dễ gặp các vấn đề khác: huyết khối tĩnh mạch sâu (OR 1.59), viêm phổi (OR 1.39), nhập viện lại không theo kế hoạch (OR 1.43), số ngày nằm viện trung bình dài thêm 0.4 ngày. 324 người già nằm viện (tuổi trung bình 86) được theo dõi trung vị 18 tháng. Loét áp lực có thể dự đoán riêng cái chết dài hạn (HR 1.500, 95% CI 1.059 đến 2.126, P=0.020). Dự đoán riêng nghĩa là sau khi loại các yếu tố khác, bản thân nó dự đoán được. 65 thử nghiệm ngẫu nhiên gộp lại (phân tích tổng hợp mạng lưới). Người ta so mặt phẳng khí lưu động chủ động bằng điện với nệm giường bệnh viện tiêu chuẩn (mặt phẳng đỡ giường bệnh viện tiêu chuẩn). Mặt phẳng khí lưu động chủ động bằng điện chính là loại nệm hơi điện. Người dùng loại này thì ít bị loét áp lực hơn (RR 0.42, 95% CI 0.29 đến 0.63, mức chắc chắn trung bình). Mặt phẳng khí lưu động hỗn hợp bằng điện là RR 0.22 (0.07 đến 0.66). Tổng quan Cochrane 11 thử nghiệm với 4462 người (cập nhật năm 2026). So xoay 2 tiếng một lần với 4 tiếng một lần thì không nói rõ khác biệt (RR 1.05, 95% CI 0.79 đến 1.39, mức chắc chắn cực thấp). So sánh giữa các khoảng xoay người khác cũng chẳng chắc hơn. Trong thử nghiệm 1226 bệnh nhân ICU, cảnh báo thời gian thực bằng cảm biến đeo được cộng với tín hiệu hình ảnh giúp thực sự xoay 2 tiếng một lần. Tỷ lệ loét áp lực thấp hơn chăm sóc thường lệ (RR 0.28, 95% CI 0.10 đến 0.75, mức chắc chắn trung bình). Trong thử nghiệm 116 người, nhóm nằm sấp có tỷ lệ loét áp lực cao hơn nhóm nằm ngửa (RR 4.55, 95% CI 2.31 đến 8.98) +- Mức bằng chứng: B +- Nguồn: Porter SB, Pla R, Chow JH, et al. (2022). Preoperative Pressure Ulcers, Mortality, and Complications in Older Hip Fracture Surgery Patients. JAAOS Global Research & Reviews, 6(11). ；Ottaviani S, Rondanina E, Longo E, 等 (2026). Impact of pressure ulcers and frailty on long-term mortality: a prospective cohort study of hospitalized older adults. European Geriatric Medicine, 17(2), 665-672. ；Shi C, Dumville JC, Cullum N (2018). Support surfaces for pressure ulcer prevention: A network meta-analysis. PLoS ONE, 13(2), e0192707. ；Latimer SL, Chaboyer WP, Probst S, 等 (2026). Repositioning for pressure injury prevention in adults. Cochrane Database of Systematic Reviews, 6, CD009958. +- Ghi chú: Chấm B vì ba lý do. Một, hai nghiên cứu tỷ lệ tử vong chỉ thấy 「có loét áp lực thì sau đó tình hình tệ hơn」, điều đó không có nghĩa là 「phòng được loét áp lực thì giảm được tỷ lệ tử vong」. Hai, kết luận của mấy thử nghiệm ngẫu nhiên về khoảng xoay người rất không chắc. Ba, bằng chứng về nệm hơi điện chắc hơn một chút, nhưng nghiên cứu so đệm giường bệnh viện, áp sang nệm dùng ở nhà là suy rộng ra. Soi mỗi ngày xương cùng, gót chân, hông, vai lưng, xem có 「vết đỏ ấn xuống không mất màu」 không, đây là việc chăm sóc thường lệ, không có tài liệu riêng. Dịch vụ chăm sóc mà bảo hiểm chăm sóc dài hạn hoàn trả, xem mục 7 của phần này. Một chân sưng đột ngột thì xử lý như huyết khối tĩnh mạch sâu, xem mục 11 ở phần 13. Khoảng thời gian nằm liệt giường lâu sau khi ngã hoặc chấn thương nặng, xem mục 34 ở phần 1. diff --git "a/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" "b/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" new file mode 100644 index 00000000..2f206001 --- /dev/null +++ "b/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" @@ -0,0 +1,102 @@ +[← 回总目录](../README.md) + +# 20. 刚出生的孩子怎么带 + +口径：婴儿死亡率与金钱。这一节只收关系到孩子能不能活下来、或者你要花多少钱的事，而且只收差别足够大、又有硬证据的。怎么哄、怎么带这些育儿细节，本书不写。出生之前和出院当天要办的事，见第 27 节。 + +### 1. 让宝宝仰着睡，睡硬的平面，同房不同床，床上不放任何软东西 + +- Chi phí: Không tốn tiền. Phải đổi nệm sang loại cứng, vài trăm tệ. Chỗ khó là phải giữ vững trước ý kiến của bậc trên, theo đúng cách này từ đầu đến cuối +- Nói thẳng: Cho con nằm ngửa mà ngủ, trên một mặt phẳng cứng. Người lớn và con ngủ cùng một phòng, nhưng không ngủ cùng một giường. Trên giường không đặt gối, chăn, tấm chắn cạnh nôi, gấu bông, gối chỉnh hình. Cũi nôi và ghế trẻ em nghiêng chỉ để ngồi, không dùng làm giường ngủ. Riêng nước Mỹ, mỗi năm có khoảng 3500 trẻ chết trong lúc ngủ. +- Lợi ích: Mỗi năm ở Mỹ có khoảng 3500 trẻ sơ sinh chết vì những nguyên nhân liên quan đến giấc ngủ. Trong đó có hội chứng đột tử ở trẻ sơ sinh, các ca không rõ nguyên nhân, và trẻ bị nghẹt thở do tai nạn trên giường. Môi trường ngủ an toàn do Hiệp hội Nhi khoa Hoa Kỳ đưa ra là 「仰卧位；使用坚实、不倾斜的睡眠表面；同房不同床；避免柔软的寝具和过热」, tức nằm ngửa; dùng mặt ngủ cứng, không nghiêng; cùng phòng khác giường; tránh chăn ga mềm và tránh nóng quá. Còn vài việc cũng làm giảm rủi ro: cho bú mẹ, không cho con tiếp xúc với nicotine và rượu, tiêm phòng đúng lịch, dùng núm vú. +- Mức bằng chứng: A +- Nguồn: Moon RY, Carlin RF, Hand I, Task Force on Sudden Infant Death Syndrome and the Committee on Fetus and Newborn (2022). Sleep-Related Infant Deaths: Updated 2022 Recommendations for Reducing Infant Deaths in the Sleep Environment. Pediatrics, 150(1), e2022057990. +- Ghi chú: 「床上不放任何软东西」nghĩa là không đặt gối, chăn, tấm chắn cạnh nôi, gấu bông, gối chỉnh hình. Cả gối cho bú trông có vẻ an toàn cũng không đặt. Cũi nôi và ghế trẻ em nghiêng không phải chỗ ngủ. Trong phần này, chuyện này chênh lệch nhiều nhất, và cũng cần giữ vững trước áp lực của bậc trên nhất +### 2. Tiêm mũi đầu tiên vaccine viêm gan B trong vòng 24 giờ sau khi sinh + +- Chi phí: Không tốn tiền. Vaccine viêm gan B thuộc vaccine trong lịch tiêm chuẩn của quốc gia, vốn đã miễn phí. Ngày con sinh ở bệnh viện đã tiêm rồi, không phải chạy đi tiêm thêm +- Nói thẳng: Cần tiêm mũi đầu tiên vaccine viêm gan B trong vòng 24 giờ sau khi con chào đời. Trong số người nhiễm viêm gan B ở giai đoạn sơ sinh và trẻ nhỏ, khoảng 95% chuyển thành viêm gan mạn tính. Người nhiễm sau khi đã trưởng thành thì đa phần tự khỏi. Vì vậy mũi tiêm này càng sớm càng có ích. Nếu mẹ có kháng nguyên bề mặt viêm gan B dương tính, con còn phải tiêm thêm globulin miễn dịch viêm gan B cùng lúc. +- Lợi ích: Tổ chức Y tế Thế giới nói: 「所有婴儿都应在出生后尽早（24 小时内）接种乙肝疫苗。」, tức mọi trẻ sơ sinh đều nên tiêm vaccine viêm gan B càng sớm càng tốt sau khi sinh (trong vòng 24 giờ). Lý do là 「婴儿期和幼儿期感染约有 95% 会发展为慢性肝炎」, tức khoảng 95% người nhiễm ở giai đoạn sơ sinh và trẻ nhỏ sẽ tiến triển thành viêm gan mạn tính. Người nhiễm sau khi đã trưởng thành đa phần tự thành (không điều trị vẫn tự khỏi) +- Mức bằng chứng: A +- Nguồn: World Health Organization. Hepatitis B fact sheet. +- Ghi chú: Nếu mẹ có kháng nguyên bề mặt viêm gan B dương tính, sau khi sinh còn phải tiêm cho con một mũi globulin miễn dịch viêm gan B. Chuyện này phải làm rõ ngay ở lần khám thai, đừng đợi sinh xong mới hỏi. Xem phần 1 về những xét nghiệm tầm soát bà mẹ cần làm +### 3. Tiêm đủ các loại vaccine trong lịch tiêm chuẩn của quốc gia, toàn bộ miễn phí, lỡ mất thì chỉ bù những liều chưa tiêm + +- Chi phí: Không tốn tiền. Luật viết rõ 「接种单位接种免疫规划疫苗不得收取任何费用」, tức cơ sở tiêm chủng không được thu bất kỳ khoản phí nào khi tiêm vaccine trong lịch tiêm chuẩn. Phải chạy vài lần phòng tiêm chủng theo đúng tuổi tháng. Giữ kỹ giấy chứng nhận tiêm chủng (预防接种证), đừng đánh mất +- Nói thẳng: Vaccine trong lịch tiêm chuẩn của quốc gia miễn phí suốt quá trình, cơ sở tiêm chủng không được thu tiền. Khi con vào nhà trẻ, vào trường, giấy chứng nhận tiêm chủng đều bị kiểm tra, mũi bị bỏ sót sẽ bị phát hiện và bù thêm. Lỡ mất thì chỉ bù những liều chưa tiêm, không phải tiêm lại từ đầu, đổi nhà sản xuất vẫn tiêm tiếp được. Với eczema, mề đay, viêm mũi dị ứng, hen suyễn, chỉ cần bệnh ổn định là tiêm bình thường, vàng da sơ sinh cũng vậy. +- Lợi ích: Khi con vào nhà trẻ, vào trường, cơ sở giáo dục và trường học phải kiểm tra giấy chứng nhận tiêm chủng. Phát hiện vaccine trong lịch tiêm chuẩn của quốc gia chưa tiêm đúng quy định thì phải báo cơ sở tiêm chủng, và phối hợp thúc giám hộ bổ sung. Vì vậy những mũi bỏ sót không thể kéo dài, khi vào trường chắc chắn bị kiểm tra ra. Hiện áp dụng là phiên bản 2026. Vaccine viêm gan C và virus gây u nhú (HPV) đã được đưa vào lịch tiêm chuẩn của quốc gia. Vaccine bạch hậu, ho gà, u dị và tải đổi tuổi bắt đầu tiêm từ 3 tháng tuổi thành 2 tháng tuổi, tổng cộng 5 liều, mỗi liều tiêm 1 lần ở 2 tháng tuổi, 4 tháng tuổi, 6 tháng tuổi, 18 tháng tuổi và 6 tuổi. Trẻ đúng tuổi ở Tây Tạng, Thanh Hải, Tân Cương và Binh đoàn Xây dựng Thủ công Tân Cương, từ tháng 3 năm 2026 cũng tiêm vaccine viêm não Nhật Bản thường quy. Một số vaccine có thể tiêm cùng một ngày: các vaccine trong lịch tiêm chuẩn của quốc gia ở giai đoạn hiện tại đều có thể tiêm đồng thời theo lịch tiêm hoặc theo nguyên tắc bù liều. Vaccine tiêm bắp từ hai loại trở lên phải tiêm ở các vị trí khác nhau, 「严禁将两种或多种疫苗混合吸入同一支注射器内接种」, cấm tuyệt đối trộn hai hay nhiều vaccine vào cùng một bơm tiêm để tiêm. Vaccine mũi tiêm không tiêm cùng ngày, giữa các vaccine giảm độc liều sống tiêm bắp từ hai loại trở lên phải cách nhau không dưới 28 ngày. Khoảng cách giữa vaccine bất hoạt, vaccine giảm độc liều sống dạng uống và vaccine khác không bị hạn chế. Nguyên tắc bù liều cho trẻ dưới 18 tuổi là càng sớm càng tốt, 「只需补种未完成的剂次，无需重新开始全程接种」, tức chỉ cần bù phần liều chưa hoàn thành, không phải tiêm lại toàn bộ, cùng một loại vaccine đổi nhà sản xuất thì tiêm tiếp được (toàn quốc, phiên bản 2026) +- Mức bằng chứng: A +- Nguồn: 全国人大常委会 (2019). 疫苗管理法（第四十八、四十九条）. （Ủy ban Thường vụ Đại hội Đại biểu Quốc dân, Cục Y tế và Sức khoẻ tỉnh Vân Nam đăng lại）；国家疾病预防控制局、国家卫生健康委 (2026). 《国家免疫规划疫苗儿童免疫程序及说明（2026 年版）》解读问答. +- Ghi chú: Hiểu lầm phổ biến nhất là 「孩子体质不好，先别打」, tức con có thể chất yếu, thôi chưa tiêm. Cách hiểu chính thức là: cái gọi là 「过敏性体质」, người trong nhà có bệnh dị ứng, con trước đây từng dị ứng với thức ăn hoặc thuốc, đều không phải lý do không tiêm. Viêm da cơ địa (eczema), mề đay, viêm mũi dị ứng, hen phế quản, chỉ cần bệnh ổn định là tiêm bình thường. Đang uống thuốc chống dị ứng, hoặc đang dùng corticoid dạng hít, vẫn tiêm được. Vàng da sinh lý và vàng da do sữa mẹ ở trẻ sơ sinh cũng tiêm bình thường. Chỉ có hai trường hợp thật sự không tiêm: trước đây từng bị phản ứng dị ứng nặng cấp thỉnh khi tiêm cùng một loại vaccine, hoặc dị ứng nặng với một thành phần đã biết có trong vaccine. Trẻ bị suy giảm miễn nhiễm, cùng trẻ đang hóa trị, xạ trị hoặc đang dùng thuốc ức chế miễn dịch, phải đánh giá riêng từng trẻ, đến phòng tiêm chủng nói rõ tình hình rồi mới quyết định. Xem mục 2 của phần này về mũi vaccine viêm gan B đầu tiên. Vaccine ngoài lịch tiêm chuẩn, do người dân tự trả, thu thêm tiền vaccine và phí dịch vụ tiêm, không liên quan đến phần miễn phí nói ở đây. +### 4. Sáu tháng đầu chỉ cho bú mẹ, không cần cho uống nước, từ 6 tháng thêm thức ăn dặm và tiếp tục cho bú mẹ + +- Chi phí: Không tốn tiền, còn tiết kiệm tiền sữa công thức. Tốn là thời gian, cho bú mỗi ngày chiếm không ít. Chỗ khó là phải cho suốt, 6 tháng đầu không được ngắt +- Nói thẳng: Bắt đầu cho bú trong giờ đầu tiên sau khi con sinh. Sáu tháng đầu chỉ cho bú mẹ, không cần cho uống nước. Từ khi đủ 6 tháng tuổi thì thêm thức ăn dặm, mỗi ngày phải có thức ăn động vật giàu sắt, tiếp tục cho bú mẹ đến 2 tuổi hoặc lâu hơn. Sữa mẹ không đủ hoặc không thể cho bú thì dùng sữa công thức, không cần tự trách. Cho ăn gì ảnh hưởng tới an nguy của con, kém xa so với tư thế ngủ. +- Lợi ích: Khuyến nghị của Tổ chức Y tế Thế giới là 「在出生后第一个小时内开始母乳喂养」, tức bắt đầu cho bú mẹ trong giờ đầu tiên sau khi sinh. Sáu tháng đầu 「纯母乳喂养，即不提供其他任何食物或液体，包括水」, tức cho bú mẹ hoàn toàn, không cung cấp thêm bất kỳ thức ăn hay chất lỏng nào khác, kể cả nước. Sau đó 「从 6 月龄起开始添加安全充足的辅食，同时继续母乳喂养到 2 岁或以上」, tức từ 6 tháng tuổi bắt đầu thêm thức ăn dặm an toàn đủ chất, đồng thời tiếp tục cho bú mẹ đến 2 tuổi hoặc hơn. Hướng dẫn dịch vụ của Ủy ban Quốc gia Y tế và Sức khoẻ cũng khuyến nghị cho bú mẹ hoàn toàn từ 0 đến 6 tháng tuổi. Hướng dẫn đó còn yêu cầu từ khi đủ 6 tháng tuổi 「每天的辅食必须包含含铁丰富的动物性食物」, tức thức ăn dặm hằng ngày phải có thức ăn động vật giàu sắt. +- Mức bằng chứng: A +- Nguồn: World Health Organization. Breastfeeding. ；Văn phòng Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委办公厅) (2024). 婴幼儿营养喂养评估服务指南（试行）（国卫办妇幼函〔2024〕452 号）. +- Ghi chú: Sữa mẹ không đủ, hoặc cơ thể không cho phép cho bú, thì dùng sữa công thức. Chênh lệch giữa cho bú mẹ và cho bú sữa công thức, nhỏ hơn nhiều so với mục ngủ an toàn. Số lần thêm thức ăn dặm: 6 đến 8 tháng tuổi mỗi ngày 1 đến 2 lần, 9 đến 12 tháng tuổi mỗi ngày 2 đến 3 lần. Những gì con ăn trong một ngày phải phủ ít nhất bốn trong bảy nhóm. Khi con bị eczema nặng hoặc dị ứng trứng, có nên né đậu phộng hay không, xem mục 12 của phần này (đừng né đậu phộng). +### 5. Pha sữa công thức bằng nước trên 70 ℃, pha xong để nguội rồi cho uống, phần còn lại thì đổi đi + +- Chi phí: Không tốn tiền. Mỗi lần phải chờ thêm vài phút cho sữa nguội. Chỗ khó là lúc con đói khóc, vẫn phải chờ nhiệt độ nước giảm xuống rồi mới pha +- Nói thẳng: Sữa công thức không vô trùng, có thể mang một loại vi khuẩn tên là Cronobacter sakazakii (阪崎克罗诺杆菌). Trẻ sơ sinh nhiễm phải, tỷ lệ tử vong được báo cáo nằm trong khoảng bốn phần đến tám phần mười. Khi pha, đổ nước nóng vào bình trước, chờ nhiệt độ nước xuống còn khoảng 73.8 ℃ rồi mới cho bột vào. Lúc đó sữa vẫn trên 70 ℃, giết được loại vi khuẩn này xuống dưới một phần trăm nghìn mức ban đầu. Pha xong để nguội rồi cho uống, phần còn lại thì đổi đi. +- Lợi ích: Sữa công thức không vô trùng, có thể mang Cronobacter sakazakii (阪崎克罗诺杆菌). Tỷ lệ tử vong ở trẻ sơ sinh sau khi nhiễm được báo cáo là 40% đến 80%. Nghiên cứu thực nghiệm cho thấy: đổ nước nóng vào bình trước, chờ nhiệt độ nước xuống còn khoảng 73.8 ℃ rồi mới cho bột vào, ở đa số trường hợp nhiệt độ sữa vẫn còn trên 70 ℃. Cách này làm giảm loại vi khuẩn đó trên 5 bậc log, xuống dưới một phần trăm nghìn mức ban đầu +- Mức bằng chứng: A +- Nguồn: Beary MA, Daly SE, Baker J, Snyder AB (2025). Assessing Hot Water Reconstitution Instructions and Labeling of Powdered Infant Formula to Ensure Cronobacter spp. Reduction. Journal of Food Protection, 88(9), 100571. ；Sima WG, Legesse T, Girma S, et al. (2025). Emerging microbial risks: Cronobacter sakazakii in powdered infant formula for infants under six months of age in Ethiopia. BMC Microbiology, 26, 307. ；World Health Organization & FAO (2007). Safe preparation, storage and handling of powdered infant formula: guidelines. +- Ghi chú: Ngưỡng 70 ℃ đến từ hướng dẫn vệ sinh công cộng. Nguyên văn của Beary và cộng sự là 「针对高危婴儿照护者的公共卫生指南建议用加热到至少 70 ℃（158 ℉）的水冲调配方奶粉以灭活微生物」, tức hướng dẫn vệ sinh công cộng cho người chăm trẻ sơ sinh nguy cơ cao khuyến nghị pha sữa công thức bằng nước đun nóng đến ít nhất 70 ℃ (158 ℉) để tiêu diệt vi sinh vật. Trẻ sinh non, trẻ cân nặng thấp và trẻ dưới 2 tháng tuổi có nguy cơ cao nhất. Đừng dùng nước để trong bình giữ nhiệt suốt ngày để pha. Cũng đừng để sữa đã pha nằm ở nhiệt độ phòng chờ bữa sau +### 6. Không cho con dưới 1 tuổi ăn mật ong + +- Chi phí: Không tốn tiền +- Nói thẳng: Con dưới 1 tuổi, đừng cho ăn mật ong. Nước mật ong, mật ong trộn vào thức ăn dặm, bánh ngọt có mật ong, đều tính. Trong mật ong có thể có vi khuẩn gây ngộ độc botulinum. Hệ vi sinh trong bụng người lớn nén được nó, trẻ sơ sinh không nén được. Dưa muối tự làm và thực phẩm lên men tự làm cũng cùng lý do đó. +- Lợi ích: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ nói 「不要给 1 岁以下的儿童喂蜂蜜」, tức đừng cho trẻ dưới 1 tuổi ăn mật ong, vì 「蜂蜜可能含有引起肉毒中毒的细菌」, tức mật ong có thể chứa vi khuẩn gây ngộ độc botulinum. Hệ vi sinh ruột của người trưởng thành nén được bào tử loại này, là trạng thái ngủ của vi khuẩn, trẻ sơ sinh không nén được +- Mức bằng chứng: A +- Nguồn: US CDC. Preventing Botulism. +- Ghi chú: Những thứ không được cho ăn gồm nước mật ong, thức ăn dặm trộn mật ong, bánh ngọt có mật ong. Cùng lý do đó, dưa muối tự làm và thực phẩm lên men tự làm cũng đừng cho trẻ sơ sinh ăn +### 7. Mũi tiêm vitamin K lúc sinh nhất định phải tiêm + +- Chi phí: 0 tệ đến vài chục tệ. Khoản tiền này thường đã nằm trong phí nằm viện sinh con, không phải trả riêng +- Nói thẳng: Trẻ sơ sinh dự trữ rất ít vitamin K trong cơ thể, không bù thì có thể xuất huyết, loại xuất hiện muộn thường là xuất huyết trong não. Ở nơi chưa tiêm rộng rãi mũi này, cứ 10 vạn trẻ sơ sinh thì có 478 trẻ bị xuất huyết kiểu này. Tiêm mũi đó, ước tính giảm được hơn bảy phần tám. Bệnh viện chính quy trong nước tiêm thường quy, sinh tại nhà hoặc tại cơ sở không chính quy thì phải tự hỏi xem đã tiêm chưa. +- Lợi ích: Trẻ sơ sinh dự trữ rất ít vitamin K trong cơ thể. Không bù có thể xuất hiện xuất huyết do thiếu vitamin K, loại xuất hiện muộn thường biểu hiện là xuất huyết trong sọ (xuất huyết não). Khuyến nghị đồng thuận của Bỉ là 「出生时肌肉注射 1 或 2 毫克维生素 K 一次」, tức tiêm bắp một liều 1 hoặc 2 miligam vitamin K vào lúc sinh. Ở những vùng không có phòng ngừa có hệ thống, tỷ lệ xuất hiện xuất huyết do thiếu vitamin K là 478 ca trên 100.000 người, con số này lấy từ 42.000 trẻ sơ sinh ở Tashkent. Tác giả ước tính, làm phòng ngừa có thể làm tỷ lệ này giảm 「cao hơn 78.5%」 +- Mức bằng chứng: B +- Nguồn: Fiesack S, Smits A, Rayyan M, et al. (2021). Nutrients, 13(11), 4109. ；Tursunov D, Yoshida Y, Yrysov K, et al. (2018). Nagoya Journal of Medical Science, 80(1), 11. +- Ghi chú: Sinh con ở bệnh viện chính quy tại Trung Quốc, bệnh viện sẽ tiêm mũi này thường quy, không phải tự lo. Sinh tại nhà, hoặc tại cơ sở không chính quy, phải tự đi xác nhận. Dạng uống kém hiệu quả hơn dạng tiêm, ở đây tiêm là tiêm vào bắp +### 8. Trẻ dưới 3 tháng tuổi, nhiệt độ chạm 38 ℃ thì đi bệnh viện luôn, không quan sát ở nhà + +- Chi phí: Khám bệnh cộng xét nghiệm vài trăm tệ +- Nói thẳng: Con dưới 3 tháng tuổi, nhiệt độ chạm 38 ℃ thì đi bệnh viện luôn. Đừng quan sát ở nhà, cũng đừng cho thuốc hạ sốt rồi chờ thêm. Ở tuổi này con có thể mắc nhiễm khuẩn nặng mà chỉ biểu hiện là sốt. Vì vậy thấy 「nhìn tinh thần vẫn được」không phải là lý do để chờ ở nhà. Trẻ mới sinh chưa đầy 8 ngày tuổi còn không nằm trong phạm vi của hướng dẫn nào, càng phải đi ngay. +- Lợi ích: Hiệp hội Nhi khoa Hoa Kỳ có hướng dẫn riêng cho trẻ sơ sinh đủ tháng, trông như còn khoẻ, từ 8 đến 60 ngày tuổi. Hướng dẫn này định nghĩa sốt là 「体温 ≥38.0 ℃」. Nó chia thành ba nhóm 8 đến 21 ngày tuổi, 22 đến 28 ngày tuổi, 29 đến 60 ngày tuổi, mỗi nhóm một cách xử lý. Trẻ dưới 8 ngày tuổi thậm chí không nằm trong phạm vi hướng dẫn này, nghĩa là giai đoạn đó càng phải đi khám ngay +- Mức bằng chứng: A +- Nguồn: Pantell RH, Roberts KB, Adams WG, et al. (2021). Evaluation and Management of Well-Appearing Febrile Infants 8 to 60 Days Old. Pediatrics, 148(2), e2021052228. +- Ghi chú: Ở tuổi này, mắc nhiễm khuẩn nặng con cũng có thể chỉ biểu hiện là sốt. Nhìn 「nhìn tinh thần vẫn được」không phải lý do để quan sát ở nhà. Cũng đừng tự cho thuốc hạ sốt trước rồi chờ thêm +### 9. Dù mệt mỏi, giận dữ đến đâu cũng đừng lắc đứa trẻ + +- Chi phí: Không tốn tiền. Chỗ khó là khi chính mình quá mệt, quá giận thì phải dừng tay +- Nói thẳng: Đầu trẻ sơ sinh to, cổ yếu. Lắc mạnh khiến trẻ xuất huyết trong não và trong mắt, có thể cướp mạng, cũng có thể để lại tàn tật cả đời. Vẗ nhẹ để ru con ngủ thì không tính, nguy hiểm là cái lắc mạnh không kìm được. Khi sắp không kiềm chế nổi, hãy đặt con lại vào nôi, bước ra vài phút, để con khóc một lát còn an toàn hơn là bế con lắc. +- Lợi ích: Đầu trẻ sơ sinh chiếm tỷ lệ lớn, cơ cổ yếu. Lắc dữ dẫn đến xuất huyết trong sọ và xuất huyết võng mạc (xuất huyết trong não và trong mắt), có thể gây chết, có thể để lại tàn tật trọn đời. Đung đưa nhẹ để ru ngủ không tính, nguy hiểm là đung đưa dữ mất kiểm soát. Y học gọi loại chấn thương do lắc, đâm như vậy là 「虐待性头部外伤」. Đây là nguyên nhân tử vong đứng đầu do chấn thương đầu ở trẻ dưới 2 tuổi. Ở Mỹ, từ 1999 đến 2014, nó gây ra gần 2250 trẻ dưới 5 tuổi tử vong +- Mức bằng chứng: B +- Nguồn: Spies EL, Klevens J (2016). Fatal Abusive Head Trauma Among Children Aged <5 Years — United States, 1999–2014. MMWR 65(20):505-509. ；Choudhary AK 等 (2018). Consensus statement on abusive head trauma in infants and young children. Pediatric Radiology 48(8):1048-1065. +- Ghi chú: Khi thật sự đến lúc sắp mất kiểm soát, trước tiên hãy đặt con lại vào nôi, bước ra vài phút. Để con khóc một lát an toàn hơn nhiều so với bế con trên tay lắc. Người một mình chăm con, ngủ thiếu ngủ lâu ngày, phải nhớ điều này nhất. Xem phần 3 về cách xử lý cảm xúc của chính bạn +### 10. Bỉ thì không xem nhãn hiệu, xem ba điều: vừa thân không, có thay đủ tần suất không, có bị thông báo kiểm tra mẫu rồi không + +- Chi phí: Mỗi tháng từ hai ba trăm tới hơn một nghìn tệ, tuỳ bạn dùng phân khúc nào +- Nói thẳng: Không cơ quan chính thức nào xếp hạng 「thương hiệu nào là tốt nhất」. Chỉ kiểm tra được hai thứ: thông báo kiểm tra chất lượng sản phẩm của Cục Quản lý thị trường quốc gia, và kết quả điều tra của cơ quan quản lý với từng sự việc cụ thể. Đắt không có nghĩa là an toàn, nhập khẩu không có nghĩa là đạt chuẩn. Trước hết mua bao nhỏ thử một tuần, xem con có hăm lỡ hay hằn ràng không, hợp rồi mới mua số lượng lớn. +- Lợi ích: Không có cơ quan chính thức nào công bố bảng xếp hạng 「thương hiệu nào là tốt nhất」, nên cẩm nang này không đề xuất thương hiệu. Thông tin kiểm chứng được chỉ có hai loại. Một là thông báo kiểm tra chất lượng sản phẩm của Cục Quản lý thị trường quốc gia. Hai là kết quả điều tra của cơ quan quản lý với từng sự việc cụ thể. Sau khi tháng 6 năm 2026 「vấn đề nhiễm propionamide trong bỉ trẻ em」được quan tâm, Cục Quản lý thị trường quốc gia dẫn đầu thành lập nhóm điều tra liên bộ để kiểm chứng +- Mức bằng chứng: C +- Nguồn: Cục Quản lý thị trường quốc gia (市场监管总局) (2026-06-22). 市场监管总局牵头成立联合调查组核查「婴幼儿纸尿裤甲酰胺问题」：「针对媒体反映的"婴幼儿纸尿裤甲酰胺问题"，市场监管总局、工业和信息化部、国家卫生健康委、国家疾控局高度重视，成立联合调查组，核查婴幼儿纸尿裤甲酰胺有关问题，并依法依规处理。有关情况将及时公布。」. ；Cổng thông tin Chính phủ Trung Quốc đăng lại. +- Ghi chú: Khi thành lập, nhóm điều tra liên bộ chỉ nói 「有关情况将及时公布」. TODO（cần đối chiếu: kết luận của cuộc điều tra. Tính đến ngày 30 tháng 9 năm 2026, trên trang chính thức của Cục Quản lý thị trường quốc gia và trong các thông báo kiểm tra mẫu giám sát quốc gia các năm gần đây vẫn không có kết luận nào về bỉ）. Cách mua thực tế: đắt không có nghĩa là an toàn, nhập khẩu không có nghĩa là đạt chuẩn. Trước hết mua bao nhỏ thử một tuần, xem con có hăm lỡ hay hằn ràng không, hợp rồi mới mua số lượng lớn. Hăm lỡ chủ yếu do thay không đủ thường xuyên và giữ không thông thoáng, quan hệ với thương hiệu không lớn như tưởng. Xem phần 5 về cách tra thông báo kiểm tra mẫu +### 11. Với đồ lớn thì cứ theo thứ tự 「mượn, mua đồ cũ, mua mới」, đừng mua đủ một lần + +- Chi phí: Không tốn thêm tiền, số tiền tiết kiệm được thường tính bằng nghìn tệ. Chỗ khó là nhịn không mua đủ một lần +- Nói thẳng: Xe đẩy, cũi cưng, khăn quấn, đồ chơi đều dùng chưa lâu, đồ cũ rất nhiều. Hỏi mượn trước, mượn không được thì mua đồ cũ, không còn cách nào mới mua mới, thường tiết kiệm được hơn một nghìn tệ. Ghế an toàn và nệm thì đừng mua đồ cũ, tổn hại sau một cú va chạm trên ghế không nhìn thấy bằng mắt. Với những khoản tiền lớn như trung tâm chăm sóc sau sinh, lớp tăng cường sớm, hãy bình tĩnh 24 giờ rồi mới quyết. +- Lợi ích: Xe đẩy, cũi cưng, khăn quấn, đồ chơi có thời gian sử dụng ngắn, nguồn cung trên thị trường đồ cũ lớn hơn nhiều so với nhu cầu. Hai thứ thật sự không nên tiết kiệm và không nên mua đồ cũ là ghế an toàn và nệm. Tổn hại kết cấu của ghế an toàn sau một cú va chạm không nhìn thấy bằng mắt +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; xem phần 1 về bằng chứng của ghế an toàn, và phần 5 về chi tiêu bốc đầu +- Ghi chú: Trung tâm chăm sóc sau sinh, lớp tăng cường sớm, bể bơi trẻ em, cẩm nang này không khuyến nghị cũng không phủ nhận. Nhưng đều là một khoản tiền lớn có thể chi hoặc không, và thời hạn bình tĩnh 24 giờ nói ở mục 5 của phần này áp dụng cho chúng +### 12. Con bị eczema nặng hoặc dị ứng trứng, đừng né đậu phộng, sớm thêm theo hướng dẫn của bác sĩ, nhưng tuyệt đối không cho ăn hạt cả nguyên + +- Chi phí: Một hộp sốt đậu phộng vài chục tệ. Cho ăn vài lần một tuần, làm luôn thôi. Trước khi thêm phải đưa con đi khám bác sĩ một lần để đánh giá. +- Nói thẳng: Trẻ sơ sinh bị eczema nặng hoặc dị ứng trứng, cho ăn đậu phộng đều đặn từ 4 đến 11 tháng tuổi, đến 5 tuổi tỷ lệ dị ứng đậu phộng chỉ 1.9%, trẻ né hoàn toàn là 13.7%. Với nhóm trẻ mà xét nghiệp da đã có phản ứng nhẹ, nhóm ăn là 10.6%, nhóm né là 35.3%. Trước khi thêm, hãy đưa con đi khám bác sĩ. Tuyệt đối không cho ăn hạt đậu phộng nguyên, sẽ nghẹn vào khí quản. +- Lợi ích: Thử nghiệm chia nhóm ngẫu nhiên LEAP của Anh. 640 trẻ sơ sinh được đưa vào nghiên cứu, điều kiện là bị eczema nặng, dị ứng trứng, hoặc cả hai. Khi nhập mẫu trẻ từ 4 đến 11 tháng tuổi. Chia ngẫu nhiên thành hai nhóm, một nhóm ăn đậu phộng đều đặn, một nhóm né hoàn toàn, đến 60 tháng tuổi thì xem tỷ lệ dị ứng đậu phộng. Trong 530 trẻ có xét nghiệm da âm tính với đậu phộng lúc nhập mẫu, nhóm né 13.7% dị ứng, nhóm ăn 1.9%（P<0.001）. Trong 98 trẻ đã dương tính yếu lúc nhập mẫu, nhóm né 35.3%, nhóm ăn 10.6%（P=0.004）. Biến cố bất lợi nghiêm trọng không khác nhau giữa hai nhóm. +- Mức bằng chứng: A +- Nguồn: Du Toit G, Roberts G, Sayre PH, 等 (2015). Randomized trial of peanut consumption in infants at risk for peanut allergy. New England Journal of Medicine, 372(9), 803-813. ；Văn phòng Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委办公厅) (2020). 婴幼儿喂养健康教育核心信息. ；Perkin MR 等 (2016). Randomized Trial of Introduction of Allergenic Foods in Breast-Fed Infants. New England Journal of Medicine, 374(18), 1733-1743. （thử nghiệm EAT với trẻ sơ sinh thường nêu trong ghi chú） +- Ghi chú: **Dạng thức chỉ có thể là sốt đậu phộng pha loãng, hoặc bột đậu phộng trộn vào thức ăn dặm, tuyệt đối không cho hạt đậu phộng nguyên.** Thông tin cốt lõi về nuôi dưỡng của Ủy ban Quốc gia Y tế và Sức khoẻ nói rõ: 「整粒花生、坚果、果冻等食物易吸入气管，引起窒息，婴幼儿应当避免食用」, tức hạt đậu phộng nguyên, hạt cây, thạch và các thức ăn tương tự rất dễ bị hút vào khí quản, gây ngạt thở, trẻ sơ sinh và trẻ nhỏ nên tránh ăn. Bị nghẹn thì làm gì, xem mục 26 ở phần 13 (có người bị nghẹn không nói nổi). Chỉ nhóm trẻ nguy cơ cao mới áp dụng: trẻ bị eczema nặng, hoặc đã dị ứng trứng. **Trước khi thêm phải đi khám bác sĩ để đánh giá, đừng tự thử ở nhà**: trước khi nhập mẫu, thử nghiệm LEAP làm xét nghiệm da cho từng trẻ, trẻ có phản ứng da lớn hơn 4 milimét bị loại khỏi thử nghiệm, không cho ăn. Với trẻ thường, bằng chứng yếu hơn nhiều. Trong một thử nghiệm khác gồm 1162 trẻ sơ sinh bú mẹ bình thường, tính theo nhóm thì nhóm cho sớm tỷ lệ dị ứng thức ăn là 5.6%, nhóm thường là 7.1%, chênh lệch không có ý nghĩa (P=0.32); chỉ trong nhóm trẻ thực sự làm theo mới thấy khác biệt (2.4% so với 7.3%). Cách tính này dễ nói quá hiệu quả, chính tác giả nói tính theo nhóm không chứng minh được hiệu lực, nhưng cũng cho thấy cho sớm là an toàn. Văn bản nuôi dưỡng chính thức của Trung Quốc hiện chưa có lập trường về 「nên cho sớm hay nên tránh thức ăn dễ dị ứng」, nên mục này viết theo bằng chứng thử nghiệm quốc tế. Về thời điểm cũng phải để ý: LEAP bắt đầu từ 4 tháng tuổi, còn Trung Quốc bắt đầu thêm thức ăn dặm khi đủ 6 tháng tuổi, xem mục 4 của phần này (sáu tháng đầu chỉ cho bú mẹ). Trẻ nguy cơ cao bắt đầu từ mấy tháng tuổi và bắt đầu thế nào, nghe bác sĩ. \ No newline at end of file diff --git "a/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" "b/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" new file mode 100644 index 00000000..bc9c5e59 --- /dev/null +++ "b/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" @@ -0,0 +1,94 @@ +[← 回总目录](../README.md) + +# 21. 出国、旅行与境外安全 + +这一节算的是钱和人身自由，有几条还关系到人身安全。到了国外，管你的是所在国的法律，不是中国的。中国使领馆能帮你到哪一步，有明确的界线。 + +### 1. Trước khi đi, tra cảnh báo an toàn của Bộ Ngoại giao một lần, chỗ nào ghi 「暂勿前往」thì đừng đến + +- Chi phí: Không tốn tiền. Mở Trang web Dịch vụ Lãnh sự Trung Quốc (中国领事服务网) xem qua một lượt, mất vài phút là xong. +- Nói thẳng: Bộ Ngoại giao chia rủi ro an toàn ở nước ngoài thành bốn bậc, bậc cao nhất gọi là 「暂勿前往」. Từ ngày 15 tháng 9 năm 2026, nghĩa vụ tránh các nước và vùng lãnh thổ nguy hiểm đã được ghi vào quy chếnh hành chính. Khi làm giấy tờ và qua cửa xuất cảnh, nhân viên còn nhắc bạn, và khi cần sẽ khuyên bạn đừng đi. Trước khi đi, xem lượt trang đó, mất vài phút. +- Lợi ích: Bộ Ngoại giao công bố cảnh báo an toàn ở nước ngoài theo mức rủi ro. Từ thấp đến cao, tạm chia làm bốn bậc: 「注意安全」, 「加强安全防范」, 「谨慎前往」, 「暂勿前往」. Quy định mới có hiệu lực từ ngày 15 tháng 9 năm 2026 cũng đưa nghĩa vụ của cá nhân vào quy chếnh hành chính. Nguyên văn là: 「中国公民应当关注国外安全提醒和旅游目的地安全风险提示，避免前往及驻留高风险国家或者地区」, tức công dân Trung Quốc nên để ý cảnh báo an toàn ở nước ngoài và cảnh báo rủi ro an toàn của điểm đến du lịch, tránh đi và lưu trú ở nước hoặc vùng lãnh thổ nguy hiểm. Quy định mới đồng thời yêu cầu cơ quan quản lý nhập cảnh, cơ quan cấp giấy tờ nhập xuất cảnh và quản lý kiểm tra biên giới, khi cấp giấy và kiểm tra xuất cảnh thì nhắc những người sắp đến khu vực nguy hiểm. Nguyên văn khoản này là: 「对准备前往风险等级为最高级别或者严重危及人身安全案件突发高发的国家或者地区的中国公民，必要时应当劝阻其前往」, tức với công dân Trung Quốc đang chuẩn bị đến nước hoặc vùng lãnh thổ có mức rủi ro cao nhất, hoặc nơi các vụ nghiêm trọng đe dọa an toàn cá nhân xảy ra bùng nổ thường xuyên, cơ quan chức trách nên khuyên can họ đừng đi khi cần thiết. +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). 关于出境入境管理的规定（国令第 841 号）. （Điều hai）；Hội đồng Nhà nước (国务院) (2023). Quy chếnh bảo vệ và hỗ trợ lãnh sự, 领事保护与协助条例（国令第 763 号）. （Điều mười chín và hai mươi mốt）；Vụ Lãnh sự, Bộ Ngoại giao. Cảnh báo an toàn, 安全提醒. +- Ghi chú: Cảnh báo đổi theo tình hình, cẩm nang này không chép lại danh sách, trước khi đi bạn tự tra trang đó một lần. Tính đến ngày 7 tháng 9 năm 2026, mục đó vẫn còn các cảnh báo loại 「暂勿前往」này: Eswatini (2026-08-25), Palestine (2023-10-10), Sudan (2023-04-17). Bốn cảnh báo nữa là Afghanistan (2023-02-17), Syria (2023-01-04), Peru (2022-12-29), Somalia (2022-11-24). Ngoài ra còn có cảnh báo riêng cho khu vực miền bắc Myanmar và một số tỉnh của Cộng hòa Dân chủ Congo. Đại lý du lịch cũng có nghĩa vụ thông báo. Trước khi đi theo đoàn, bạn có thể yêu cầu bên đó bằng văn bản nói rõ rủi ro của điểm đến +### 2. Lưu 12308 và số điện thoại bảo vệ lãnh sự của đại sứ đoàn địa phương vào điện thoại, rồi chép một bản bỏ vào ví, đừng đợi sự cố rồi mới tìm + +- Chi phí: Không tốn tiền. Lưu hai số, rồi chép một bản, mất vài phút. +- Nói thẳng: Khi có sự cố chỉ có một số gọi đi. Ở trong nước gọi 12308, ở nước ngoài gọi +86-10-12308. Từ khi mở đường dây nóng này đã tiếp hơn 300 vạn cuộc gọi. Cùng thời gian đó, Bộ Ngoại giao và các cơ quan đại diện ở nước ngoài xử lý hơn 50 vạn vụ bảo vệ lãnh sự. +- Lợi ích: Bộ Ngoại giao có một đường dây nóng khẩn cấp bảo vệ và phục vụ lãnh sự toàn cầu, có người trực 24 giờ. Ở trong nước gọi 12308. Ở nước ngoài gọi +86-10-12308, hoặc +86-10-65612308. Từ khi mở năm 2014 đến nay, đường dây nóng đã tiếp hơn 300 vạn cuộc gọi. Cùng thời gian đó, Bộ Ngoại giao và các cơ quan đại diện ở nước ngoài xử lý hơn 50 vạn vụ bảo vệ lãnh sự các loại +- Mức bằng chứng: A +- Nguồn: Bộ Ngoại giao và Đại sứ toàn tháp Trung Quốc tại Myanmar (外交部、中国驻缅甸使领馆) (2023). 关于当前缅北地区安全局势的领事提醒. ；Văn phòng Thông tin của Hội đồng Nhà nước (国务院新闻办) (2023). 国务院政策例行吹风会介绍领事保护与协助有关情况. +- Ghi chú: Đại sứ hoặc đại sứ toàn tháp tại nước đó còn có số bảo vệ lãnh sự riêng, thường nhanh hơn tổng đài. Trước khi đi, tra theo từng nước trên Trang web Dịch vụ Lãnh sự Trung Quốc, lưu số đó cùng số tổng đài vào điện thoại. Mất điện thoại hoặc điện thoại hết pin thì số lưu trong đó cũng vô dụng. Vì vậy nên chép tay một bản bỏ vào ví, hoặc viết ra một bản đưa cho người nhà +### 3. Biết trước bảo vệ lãnh sự làm được gì và không làm được gì: có thể thăm, không thể cứu ra, chi phí vẫn tự bỏ + +- Chi phí: Không tốn tiền. Đọc kỹ ranh giới này mất vài phút. +- Nói thẳng: Bạn bị bắt ở nước ngoài, đại sứ đoàn làm được: xác minh tình hình với nơi đó, yêu cầu xử lý theo pháp luật và công bằng, đi thăm bạn, ngồi dự xét xử. Nó không thay đổi phán quyết thay bạn, cũng không đưa người ra khỏi nước đó. Chi phí ăn ở, đi lại, y tế do người khác trả trước, cuối cùng vẫn do bạn trả. Nên trước khi đi nước ngoài, chuẩn bị tiền và bảo hiểm, đừng coi 「có sự cố thì nhà nước sẽ gánh」là một kế hoạch. +- Lợi ích: Khi bạn bị bắt, đại sứ đoàn làm được những việc sau. Xác minh tình hình với các cơ quan có liên quan của nước sở tại, yêu cầu xử lý theo pháp luật và công bằng. Nguyên văn quy chếnh là 「按照驻在国法律和我国与驻在国缔结或者共同参加的国际条约对其进行探视或者与其联络，了解其相关需求，要求驻在国有关部门给予该中国公民人道主义待遇和公正待遇」, tức căn cứ pháp luật nước sở tại và các hiệp định mà hai nước ký hoặc cùng tham gia, để thăm người đó hoặc liên lạc với người đó, nắm nhu cầu của họ, và yêu cầu cơ quan có liên quan của nước sở tại đối xử với công dân Trung Quốc đó nhân đạo và công bằng. Nói gọn là đi thăm bạn, liên lạc với bạn, hỏi bạn cần gì, theo pháp luật địa phương và các hiệp định hai nước ký, và yêu cầu nơi đó đối xử với bạn theo tiêu chuẩn nhân đạo và công bằng. Khi xét xử, đại sứ đoàn có thể ngồi dự và có thể yêu cầu bảo vệ quyền tố tụng của bạn. Khi bạn bị thương hoặc gặp thiên tai, nó có thể thúc ép chính quyền địa phương cứu hộ khẩn cấp, và giúp gia đình lo hậu quả. Về chi phí thì quy chếnh nói rõ không kém gì: 「中国公民、法人、非法人组织在领事保护与协助过程中，得到第三方提供的食宿、交通、医疗等物资和服务的，应当支付应由其自身承担的费用」, tức công dân, pháp nhân, tổ chức phi pháp nhân Trung Quốc trong quá trình được bảo vệ và hỗ trợ lãnh sự mà nhận vật tư, dịch vụ như ăn ở, đi lại, y tế do bên thứ ba cung cấp thì phải trả phần chi phí do mình phải chịu. Ăn ở, đi lại, y tế người khác trả trước thì phần bạn phải trả vẫn là bạn trả +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2023). Quy chếnh bảo vệ và hỗ trợ lãnh sự, 领事保护与协助条例（国令第 763 号）. （Điều chín, mười, mười bốn, hai mươi sáu） +- Ghi chú: Đại sứ đoàn không trả tiền phạt thay bạn, không trả viện phí thay bạn, không mua vé máy bay thay bạn, cũng không đổi được kết luận của tòa án nước sở tại. Nó cho bạn là thông tin về luật sư, thông dịch, y tế, nơi tang lễ tại chỗ, và giúp bạn liên lạc với người thân, giúp bạn tìm đường khiếu kiện. Nên trước khi đi nước ngoài, tự chuẩn bị tiền và bảo hiểm, đừng coi 「có sự cố thì nhà nước sẽ gánh」là một kế hoạch +### 4. Mua một gói bảo hiểm có bảo hiểm y tế nước ngoài và vận chuyển y tế, đừng chỉ mua bảo hiểm chuyến bay hoãn + +- Chi phí: Một gói bảo hiểm du lịch nước ngoài ngắn hạn, từ vài chục tới vài trăm tệ. +- Nói thẳng: Tiền khám chữa bệnh ở nước ngoài tự túi. Bảo hiểm y tế trong nước về cơ bản không chi trả, đại sứ đoàn cũng không ứng trước. Thứ thật sự có thể làm sập một gia đình không phải là phí khám ngoại trú, mà là nằm viện sau chấn thương nặng, cộng với việc đưa người về nước, việc đó cần máy bay riêng hoặc hộ tống y tế. Khi mua bảo hiểm du lịch, xem chủ yếu hai hạn mức: chi phí y tế nước ngoài, vận chuyển và hồi hương y tế khẩn cấp. +- Lợi ích: Tiền khám chữa bệnh ở nước ngoài phải tự bỏ ra. Bảo hiểm y tế trong nước về cơ bản không hoàn trả ở nước ngoài. Bảo vệ lãnh sự cũng không ứng trước khoản tiền này (xem mục 3 của phần này, bảo vệ lãnh sự làm được gì và không làm được gì, căn cứ là Điều hai mươi sáu của quy chếnh). Thứ thật sự làm sập một gia đình không phải là phí khám ngoại trú, mà là nằm viện sau chấn thương nặng và vận chuyển y tế. Chuyển người bệnh từ bệnh viện nước ngoài về nước phải dùng máy bay riêng hoặc hộ tống y tế, thuộc nhóm đắt nhất trong các khoản tự bỏ. Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ nói: vận chuyển khẩn cấp người bệnh từ nơi xa cách về bệnh viện tốt, nếu tự bỏ thì có thể tốn hơn 100.000 đô la Mỹ. Khi mua bảo hiểm, xem trước hai hạn mức. Một là hạn mức chi phí y tế nước ngoài. Hai là hạn mức vận chuyển và hồi hương y tế khẩn cấp, quản lý khoản cao nhất được bồi thường cho việc chuyển viện sau tai nạn và đưa người bệnh về nước +- Mức bằng chứng: C +- Nguồn: Hội đồng Nhà nước (国务院) (2023). Quy chếnh bảo vệ và hỗ trợ lãnh sự, 领事保护与协助条例（国令第 763 号）. （Điều hai mươi sáu）；Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ (美国疾病控制与预防中心). Travelers' Health: Travel Insurance：「Medical evacuation insurance covers emergency transportation from a remote area to a high-quality hospital, which could otherwise cost more than $100,000.」. ；phần còn lại là kinh nghiệm của tác giả, không có tài liệu trực tiếp +- Ghi chú: Con số 100.000 đô la là mức sàn, không phải một khoảng chi phí. Không tìm thấy thống kê khoảng chi phí công bố của cơ quan chính thức hay của ngành, nên cẩm nang này không đưa ra khoảng cụ thể. Trước khi mua, xác nhận ba điều. Một, nước bạn đến có nằm trong phạm vi bảo hành không. Hai, các hoạt động rủi ro cao như lặn, trượt tuyết, đi xe máy có bị loại trừ không. Ba, việc bồi thường là công ty bảo hiểm quyết toán thẳng với bệnh viện, hay bạn tạm ứng trước rồi về sau quyết toán lại. Đi nước không cần visa thì vẫn phải mua +### 5. 「 tuyển dụng lương cao ở nước ngoài」 thì coi như lừa đảo, bị lừa đi làm lừa đảo qua điện thoại, về nước còn bị hạn chế xuất cảnh + +- Chi phí: Không tốn tiền. Chỗ khó là tự tay gạt bỏ một công việc trông rất tốt. +- Nói thẳng: Lương tháng cao đến mức phi lý, không hỏi bằng cấp hay kinh nghiệm, không ký hợp đồng chính quy, bảo bạn tập hợp ở thành phố biên giới trước, còn giữ hộ hộ chiếu của bạn. Đi làm việc đó thì cả người lẫn hộ chiếu đều không nằm trong tay bạn. Khi làm giấy tờ sai sự thật lúc xuất cảnh, kiểm tra biên giới sẽ chặn bạn không cho đi. Tham gia hành vi phạm pháp ở nước ngoài, về nước rồi còn có thể bị cấm xuất cảnh 6 tháng đến 3 năm. +- Lợi ích: Cách gọi chính thức về loại vụ án này rất thẳng thắn. Những năm gần đây xuất hiện 「上当受骗出境或者虚构事由非法出境从事跨境赌博、电信网络诈骗等违法活动，严重危害人民群众生命和财产安全」, tức bị lừa ra nước ngoài, hoặc bịa lý do để lén ra nước ngoài, sang đó làm cờ bạc xuyên biên giới, lừa đảo qua mạng viễn thông và các hoạt động phạm pháp khác, gây nguy hiểm nghiêm trọng tới tính mạng và tài sản của quần chúng. Nói gọn là bị lừa đi, hoặc bịa chuyện trốn ra ngoài để làm cờ bạc và lừa đảo qua điện thoại. Quy định mới siết chặt cả hai đầu. Đầu ra nước: 「申请出境入境、停留居留的事由应当真实、合法」, tức lý do xin xuất cảnh, nhập cảnh, lưu trú phải đúng sự thật và hợp pháp. Nộp giấy tờ giả hoặc nói dối, cơ quan quản lý nhập cảnh có quyền không cấp giấy tờ cho bạn, không cho bạn xuất cảnh. Đầu về nước: 「中国公民在境外从事违法犯罪活动，危害国家安全和利益的，可以……决定自回国之日起 6 个月至 3 年以内不准其出境」, tức công dân Trung Quốc thực hiện hành vi vi phạm pháp luật ở nước ngoài, gây nguy hiểm cho an ninh và lợi ích quốc gia, có thể 「……quyết định tính từ ngày về nước, không cho xuất cảnh trong thời gian 6 tháng đến 3 năm」. Người bị xử phạt hành chính giam giữ vì tình tiết lấy giấy tờ gian lận hoặc xuất nhập cảnh trái phép thì cũng bị cấm xuất cảnh 6 tháng đến 3 năm +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). 关于出境入境管理的规定（国令第 841 号）. （Điều ba, bốn）；Bộ Tư pháp, Bộ Công an, Cục Quản lý nhập cảnh quốc gia (司法部、公安部、国家移民局) (2026). 就《国务院关于出境入境管理的规定》答记者问. +- Ghi chú: Các dấu hiệu nhận biết về cơ bản là cố định: lương tháng cao hơn nhiều so với cùng vị trí, không yêu cầu bằng cấp hay kinh nghiệm, không ký hợp đồng chính quy, bảo tập hợp ở thành phố biên giới trước, làm thủ tục visa thay bạn và giữ hộ chiếu của bạn. Khu vực miền bắc Myanmar đã có cảnh báo lãnh sự riêng. Nếu thật sự đến nơi đó, tự do cá nhân và hộ chiếu đều không nằm trong tay bạn, muốn đi cũng không đi được +### 6. Rút tiền ở nước ngoài một năm không được quá 100.000 tệ nhân dân tệ, tính gộp tất cả thẻ mang tên bạn + +- Chi phí: Không tốn tiền. Trước khi đi, ước tính chuyến này cần rút bao nhiêu tiền mặt, phần còn lại dùng thẻ. Chỗ khó là đổi thói quen, được dùng thẻ thì đừng rút tiền mặt. +- Nói thẳng: Hạn mức rút tiền mặt ở nước ngoài không tính từng thẻ một, mà tính tất cả thẻ ngân hàng trong tên bạn (kể cả thẻ phụ) cộng lại, mỗi năm dương lịch 100.000 tệ nhân dân tệ. Rút hết hạn mức thì phần thời gian còn lại của năm đó và cả năm sau không được rút tiền mặt ở nước ngoài. Chỉ có rút tiền mặt bị dừng, chi tiêu bằng thẻ không bị ảnh hưởng, và cũng không chiếm hạn mức mua ngoại tệ 50.000 đô la Mỹ mỗi năm của bạn. +- Lợi ích: Căn cứ là Thông tư của Cục Quản lý Ngoại hối nhà nước về việc quản lý giao dịch rút tiền mặt số lớn bằng thẻ ngân hàng ở nước ngoài (关于规范银行卡境外大额提取现金交易的通知, 汇发〔2017〕29 号, có hiệu lực từ ngày 1 tháng 1 năm 2018). Điều một: 「个人持境内银行卡在境外提取现金，本人名下银行卡（含附属卡）合计每个自然年度不得超过等值 10 万元人民币。超过年度额度的，本年及次年将被暂停持境内银行卡在境外提取现金」, tức cá nhân dùng thẻ ngân hàng trong nước rút tiền mặt ở nước ngoài, tổng số tiền của tất cả thẻ ngân hàng trong tên mình (kể cả thẻ phụ) cộng lại không được vượt quá 100.000 tệ nhân dân tệ tương đương trong mỗi năm dương lịch. Vượt quá hạn mức năm thì việc rút tiền mặt ở nước ngoài bằng thẻ ngân hàng trong nước bị đình chỉ trong năm đó và năm kế tiếp. Khoản thứ hai của cùng điều đó: 「个人不得通过借用他人银行卡或出借本人银行卡等方式规避或协助规避境外提取现金管理」, tức cá nhân không được vay thẻ ngân hàng của người khác hoặc cho mượn thẻ của mình để trốn hoặc để giúp người khác trốn việc quản lý rút tiền mặt ở nước ngoài. Điều hai: mỗi ngày cơ quan ngoại hối gửi danh sách đình chỉ tới các ngân hàng phát hành thẻ, ngân hàng phát hành thẻ phải từ 17 giờ theo giờ Bắc Kinh trong ngày, không chậm hơn thế, chặn việc người có trong danh sách đình chỉ rút tiền mặt ở nước ngoài bằng thẻ của ngân hàng mình. Điều ba: người bị đưa vào danh sách đình chỉ có thể mang giấy tờ tùy thân hợp lệ của mình đến chi nhánh cơ quan ngoại hối để tra chi tiết rút tiền mặt ở nước ngoài của bản thân. Điều bốn: hạn mức mỗi thẻ ngoại tệ mỗi ngày, điều chỉnh từ mức tương đương 1000 đô la Mỹ lên mức tương đương 10.000 tệ nhân dân tệ; thẻ nhân dân tệ vẫn giữ mức mỗi thẻ mỗi ngày không quá mức tương đương 10.000 tệ nhân dân tệ. Điều sáu: ngân hàng phát hành thẻ, tổ chức thanh toán thẻ nhân dân tệ trong nước và cá nhân vi phạm các quy định của thông tư này bị xử phạt theo các quy định liên quan của Quy chếnh quản lý ngoại hối của Cộng hòa Nhân dân Trung Quốc. Câu trả lời của cơ quan ngoại hối cho báo chí còn làm rõ vài điểm. Thông tư không ảnh hưởng tới việc cá nhân dùng thẻ chi tiêu ở nước ngoài. Các giao dịch thường ngày do du lịch, công tác, du học ở nước ngoài như ăn ở, đi lại, mua sắm, đều có thể thanh toán bằng thẻ ngân hàng. Các chi tiêu đó cũng không chiếm hạn mức mua ngoại tệ thuận tiện hằng năm 50.000 đô la Mỹ của cá nhân. Hạn mức năm do cơ quan ngoại hối tính bằng cách tổng hợp dữ liệu của các tổ chức tài chính phát hành thẻ, vì kiểm soát tức thời sẽ làm chậm phản hồi giao dịch, 「外汇局采用延时控制手段」, tức không chặn tại chỗ. Theo thống kê, năm 2016 có 81% số lần rút tiền mặt ở nước ngoài bằng thẻ ngân hàng trong nước dưới 30.000 tệ nhân dân tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Cục Quản lý Ngoại hối nhà nước (国家外汇管理局) (2017). 关于规范银行卡境外大额提取现金交易的通知（汇发〔2017〕29 号, Điều một, hai, ba, bốn, sáu, bảy, có hiệu lực từ ngày 1 tháng 1 năm 2018）. ；Cục Quản lý Ngoại hối nhà nước (国家外汇管理局) (2017). 外汇局有关负责人就规范银行卡境外大额提取现金交易有关问题答记者问. ；Tập đoàn UnionPay Quốc tế Trung Quốc (中国银联国际). Rút tiền tại ATM ở nước ngoài, 境外ATM取款. +- Ghi chú: Hạn mức tính theo năm dương lịch, ngày 1 tháng 1 bắt đầu lại từ đầu. Chỉ có việc rút tiền mặt ở nước ngoài bị dừng, chi tiêu bằng thẻ không bị ảnh hưởng. Nếu thật sự cần dùng nhiều tiền mặt ở nước ngoài thì theo Biện pháp quản lý ngoại hối của cá nhân (个人外汇管理办法), làm thủ tục mua ngoại tệ trước, rồi mang tiền ngoại tệ mặt ra khỏi nước. Bản thân việc rút tiền còn có hai cái bẫy, trang 「境外ATM取款」của UnionPay Quốc tế nói rõ. Một, thẻ có hai nhãn ở một số ATM tại một số nước không chọn được kênh UnionPay để rút. Hai, một số ATM ở nước ngoài hiện trên màn hình khoản phí thêm (surcharge), khoản tiền đó do tổ chức thanh toán tại chỗ thu, 「与银联无关」, không liên quan gì đến UnionPay. Cùng trang đó còn nhắc 「由于信用卡取款需支付透支利息，建议持卡人境外取款使用银联借记卡」, tức vì rút tiền bằng thẻ tín dụng phải trả lãi vượt hạn mức, nên khuyến nghị chủ thẻ rút tiền ở nước ngoài bằng thẻ ghi nợ UnionPay. Thẻ bị mất, bị nuốt hoặc bị dùng gian lận thì làm gì, xem mục 5 ở phần 14 (thẻ bị dùng gian lận thì báo mất và đóng băng trước, rồi báo công an) +### 7. Chụp ảnh hộ chiếu, visa, giấy tùy thân lưu một bản trên đám mây, mất rồi báo công an trước rồi xin giấy tờ du lịch + +- Chi phí: Không tốn tiền. Trước khi đi, chụp vài bức ảnh. +- Nói thẳng: Mất hộ chiếu, đại sứ đoàn có thể cấp cho bạn một giấy tờ du lịch dùng để về nước. Điều kiện là bạn chứng minh được mình là ai. Có ảnh trang thông tin, ảnh trang visa và lịch sử nhập cảnh trong tay thì việc cấp lại nhanh hơn nhiều. Bạn còn phải báo công an tại chỗ trước để lấy biên nhận, kiểm tra xuất cảnh và bồi thường bảo hiểm đều cần tờ này. Điện thoại thường mất cùng hộ chiếu, đừng chỉ lưu ảnh trong điện thoại. +- Lợi ích: Mất hộ chiếu, đại sứ toàn tháp ở nước ngoài có thể cấp cho bạn giấy tờ du lịch dùng để về nước, nhưng cần bạn chứng minh danh tính trước. Có ảnh rõ trang thông tin hộ chiếu, ảnh trang visa và lịch sử nhập cảnh trong tay thì tốc độ cấp lại chênh lệch rất lớn. Đồng thời bạn phải báo cô an tại chỗ trước để lấy biên nhận báo án. Nhiều nước yêu cầu tài liệu này khi kiểm tra xuất cảnh và khi bồi thường bảo hiểm. Cửa vào để tìm đại sứ đoàn chính là hai số bảo vệ lãnh sự đã lưu ở mục 2 của phần này +- Mức bằng chứng: C +- Nguồn: Hội đồng Nhà nước (国务院) (2023). Quy chếnh bảo vệ và hỗ trợ lãnh sự, 领事保护与协助条例（国令第 763 号）. （Điều mười sáu）；kinh nghiệm của tác giả, không có tài liệu trực tiếp +- Ghi chú: Cách lưu là chụp xong lưu vào hai nơi khác nhau, ví dụ một bản trên ổ đám mây, một bản gửi cho một người nhà. Ngoài ra in một bản giấy, cất chỗ khác với bản gốc. Chụp luôn thẻ ngân hàng và đơn thuốc thường dùng. Điện thoại và hộ chiếu thường mất cùng nhau, chỉ lưu trong điện thoại thì coi như không lưu +### 8. Tự lái xe ở nước ngoài, trước tiên xác nhận nơi đó có công nhận giấy phép lái xe Trung Quốc không, 「giấy phép lái xe quốc tế」bán trên mạng đa phần là giấy vô dụng + +- Chi phí: Không tốn tiền. Trước khi đi, tra quy định của điểm đến một lần. +- Nói thẳng: 「Giấy phép lái xe quốc tế」do bên ngoài làm hộ trên mạng không có giá trị ở đại lập thổ Trung Quốc. Tờ đó là bản dịch được các bên ký Công ước Geneva 1949 công nhận lẫn nhau, và đại lập thổ Trung Quốc không có trong 103 bên ký. Nên ở trong nước, ai làm hộ cũng không có hiệu lực pháp lý. Có dùng giấy phép lái xe Trung Quốc để lái xe ở nước đó hay không phải xem quy định riêng của điểm đến, chưa xác nhận rõ thì đừng lái. +- Lợi ích: Cái gọi là 「giấy phép lái xe quốc tế」(IDP) là bản dịch được các nước ký Công ước về Giao thông đường bộ năm 1949 công nhận lẫn nhau. Công ước này ký tại Geneva ngày 19 tháng 9 năm 1949, có hiệu lực từ ngày 26 tháng 3 năm 1952. Hiện có 103 bên ký, tức các nước và vùng lãnh thổ đã ký. Đại lập thổ Trung Quốc không nằm trong danh sách đó. Vì vậy Trung Quốc không cấp loại giấy phép lái xe quốc tế này. Bất kỳ cơ quan nào 「làm hộ giấy phép lái xe quốc tế」trong nước đều không tạo ra hiệu lực pháp lý. Có dùng giấy phép lái xe Trung Quốc để lái xe hay không phụ thuộc quy định riêng của nước hoặc vùng lãnh thổ đến: nơi thì công nhận giấy phép lái xe Trung Quốc kèm bản dịch có chứng thực, nơi thì bắt phải đổi lấy giấy phép lái xe của nước đó, nơi thì hoàn toàn không công nhận +- Mức bằng chứng: A +- Nguồn: United Nations Treaty Collection. Convention on Road Traffic, Geneva, 19 September 1949 — status of treaties. +- Ghi chú: Ở phần lớn các nước, lái xe không giấy phép là tội phạm hình sự, hoặc bị phạt nặng. Hơn nữa khi xảy ra sự cố, bảo hiểm nhiều khả năng từ chối bồi thường, đó mới là chi phí thật. Có hai cách tra: tìm 「Lưu ý dành cho công dân Trung Quốc」của nước đến trên Trang web Dịch vụ Lãnh sự Trung Quốc, hoặc bảo công ty cho thuê xe đưa bạn một xác nhận bằng văn bản. Hồng Kông, Macao, Đài Loan cùng một số nước có cách bố trí riêng +### 9. Tìm môi giới làm visa, du học, định cư, trước tiên hỏi nó đã đăng ký với cơ quan quản lý nhập cảnh chưa + +- Chi phí: Không tốn tiền. Hỏi thêm một câu xem nó đã đăng ký chưa, rồi giữ lại tài liệu bằng văn bản. +- Nói thẳng: Từ ngày 15 tháng 9 năm 2026, môi giới làm visa, du học, định cư phải đăng ký với cơ quan quản lý nhập cảnh. Những câu 「chắc chắn qua」, 「bảo đảm cấp visa」, 「hồ sơ chúng tôi làm giúp」 vốn nằm trong danh sách bị cấm rõ ràng. Bạn tự điền sai, nhiều nhất lần này bị từ chối visa. Dùng hồ sơ giả mà bị xác định là khai sai, vết bẩn sẽ nằm lại trong lịch sử nhập cảnh của bạn về sau. +- Lợi ích: Năm 2018 đã bỏ giấy phép đầu tư cơ sở môi giới nhập cảnh tư nhân, tức mở môi giới không cần xin phép trước. Sau đó số lượng cơ sở tăng nhanh, và xuất hiện tình trạng 「không rõ số lượng cơ sở, một số cơ sở làm ăn trái phép, gây hại quyền lợi của người xuất nhập cảnh」, đến mức cơ quan chính thức không đếm nổi có bao nhiêu. Từ ngày 15 tháng 9 năm 2026, loại môi giới này chuyển sang quản lý theo đăng ký. Cơ sở phải báo với cơ quan quản lý nhập cảnh nơi đó trong vòng 15 ngày kể từ ngày thành lập. Cơ sở đã hoạt động trước đó phải đăng ký bổ sung trong vòng 90 ngày kể từ ngày quy chếnh có hiệu lực. Các hành vi bị cấm rõ ràng gồm mấy loại. Một là 「发布虚假信息，或者通过夸大宣传、误导性宣传等方式招徕服务对象」, tức công bố thông tin sai, hoặc tuyển khách bằng quảng cáo ngoại trộng, quảng cáo gây hiểu lầm. Hai là 「提供或者协助提供虚假材料，协助他人违规办理签证、停留居留证件、护照等出境入境证件或者手续」, tức cung cấp hoặc giúp cung cấp tài liệu giả, giúp người khác làm trái phép các thủ tục về visa, giấy lưu trú, hộ chiếu và các giấy tờ nhập xuất cảnh khác. Ba là 「泄露、出售、非法提供……个人隐私或者个人信息」, tức làm lộ, bán hoặc cung cấp trái phép thông tin cá nhân. Bốn là 「组织或者协助他人从事跨境违法犯罪活动」, tức tổ chức hoặc giúp người khác thực hiện hành vi phạm pháp xuyên biên giới. Doanh nghiệp, tổ chức ở nước ngoài không được cung cấp loại dịch vụ môi giới này tại lãnh thổ Trung Quốc +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). 关于出境入境管理的规定（国令第 841 号）. （Điều bảy, tám, mười, mười hai, mười ba） +- Ghi chú: Môi giới nào lúc nào cũng dùng 「chắc chắn qua」, 「bảo đảm cấp visa」, 「hồ sơ chúng tôi làm giúp」 thì bản thân nó đã làm việc bị cấm rõ ràng. Một khi bị phát hiện, vết bẩn nằm lại ở lịch sử visa của bạn. Bạn tự điền hồ sơ mà có vấn đề, nhiều nhất lần này bị từ chối visa. Dùng hồ sơ giả mà bị xác định là khai sai thì hậu quả nặng hơn nhiều. Tư vấn chính sách và tra cứu thông tin không thu tiền thì không tính là dịch vụ môi giới +### 10. Đừng làm thay thư mời cho người lạ, cũng đừng để ai làm thủ tục nhập xuất cảnh dưới danh nghĩa của bạn + +- Chi phí: Không tốn tiền. Chỗ khó là từ chối ngay trước mặt một lần tình nghĩa. +- Nói thẳng: Làm thay thư mời hoặc hồ sơ xin cho người khác mà nội dung không đúng, người bị phạt là người làm. Cá nhân bị phạt từ 5000 tệ đến 1 vạn tệ, có thu nhập bất hợp pháp thì thu giữ cả. Đơn vị bị phạt từ 1 vạn đến 5 vạn tệ, người trực tiếp xử lý còn bị phạt thêm từ 5000 đến 1 vạn tệ. Một lần tình nghĩa, đổi lấy việc bạn tự bỏ tiền ra trả. +- Lợi ích: Người làm thư mời phải chịu trách nhiệm về việc nội dung có thật không. Nguyên văn là: 「个人为他人申请出境入境、停留居留出具虚假邀请函件或者其他申请材料的，由移民管理机构处 5000 元以上 1 万元以下罚款；有违法所得的，没收违法所得」, tức cá nhân làm thư mời giả hoặc tài liệu xin việc khác giả cho người khác xin xuất nhập cảnh, lưu trú thì bị cơ quan quản lý nhập cảnh phạt từ 5000 tệ đến 1 vạn tệ; có thu nhập bất hợp pháp thì thu giữ thu nhập bất hợp pháp. Đơn vị làm tài liệu giả bị phạt từ 1 vạn đến 5 vạn tệ. Đồng thời người chịu trách nhiệm trực tiếp và người chịu trách nhiệm trực tiếp khác, tức lãnh đạo quản lý việc đó và người trực tiếp làm, còn bị phạt từ 5000 tệ đến 1 vạn tệ +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). 关于出境入境管理的规定（国令第 841 号）. （Điều ba, mười một） +- Ghi chú: Đây cùng loại rủi ro với mục 「không giúp người lạ mang đồ」ở phần 8: hồ sơ xuất dưới danh nghĩa của bạn thì hậu quả rơi vào bạn. Công chức và quân nhân muốn qua môi giới để làm trái phép quốc tịch nước ngoài hoặc tư cách cư trú ở nước ngoài thì môi giới không được làm, còn phải báo cáo cơ quan giám sát +### 11. Trước khi xuất cảnh, gửi lịch trình, nơi ở và người đi cùng cho một người nhà, hẹn trước bao lâu liên lạc một lần + +- Chi phí: Không tốn tiền. Trước khi đi, gửi một tin nhắn. +- Nói thẳng: Khi mất liên lạc ở nước ngoài, người nhà có thể nhờ đại sứ đoàn giúp, đại sứ đoàn sẽ báo cho gia đình biết cách báo công an tại chỗ, và sau khi lập hồ sơ còn thúc ép xử lý. Điều kiện là ở trong nước có người biết bạn đáng lẽ ở đâu, đáng lẽ liên lạc vào lúc nào. Gửi số hiệu chuyến bay, địa chỉ nơi ở, số điện thoại người đi cùng và tần suất liên lạc đã hẹn cho một người nhà là đủ. +- Lợi ích: Khi công dân Trung Quốc mất tích ở nước ngoài, người thân có thể nhờ cơ quan ngoại giao ở nước ngoài giúp. Nguyên văn quy chếnh là 「驻外外交机构应当提供当地报警方式及其他获取救助的信息」, tức cơ quan ngoại giao ở nước ngoài phải cung cấp cách báo công an tại chỗ cùng các thông tin khác về cách tiếp cận cứu hộ. Sau khi công an địa phương lập hồ sơ, đại sứ đoàn sẽ thúc họ xử lý kịp thời. Các trợ giúp này có khởi động được hay không phụ thuộc vào việc ở trong nước có ai biết bạn đáng lẽ ở đâu, đáng lẽ liên lạc vào lúc nào +- Mức bằng chứng: C +- Nguồn: Hội đồng Nhà nước (国务院) (2023). Quy chếnh bảo vệ và hỗ trợ lãnh sự, 领事保护与协助条例（国令第 763 号）. （Điều mười ba）；kinh nghiệm của tác giả, không có tài liệu trực tiếp +- Ghi chú: Chỉ cần viết rõ bốn thứ: số hiệu chuyến bay, địa chỉ nơi ở, tên và số điện thoại người đi cùng, tần suất liên lạc đã hẹn. Nếu đi nơi sóng yếu hoặc tình hình an ninh kém, hãy viết thêm câu 「bao lâu không có tin thì báo công an」. Tất cả các thủ tục cứu hộ nói trên đều phải có người nhà biết lịch trình của bạn thì mới bắt đầu được \ No newline at end of file diff --git "a/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" "b/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" new file mode 100644 index 00000000..e0f7303f --- /dev/null +++ "b/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" @@ -0,0 +1,94 @@ +[← 回总目录](../README.md) + +# 22. 怎么放松：娱乐场所和减压 + +前半节讲娱乐场所里哪些钱花得冤、安全出口在哪，口径是金钱和人身自由。后半节讲怎么减压，口径是精力和总死亡率。两半的数字不放在一起换算。第 3 节已经写过情绪低落时先做哪几件事，这里给更具体的做法。 + +### 1. Vào KTV, quán bar, phòng giải đố trước khi vào hãy nhìn lối thoát hiểm ở đâu, bị khóa hoặc bị chặn thì đổi chỗ + +- Chi phí: Không tốn tiền. Lúc bước vào, ngẩng lên nhìn 10 giây. +- Nói thẳng: Theo quy định, khi đang hoạt động, lối thoát hiểm và lối ra khẩn cấp của nơi giải trí phải thông suốt. Không được chặn, không được khóa, không được xếp đồ chắn đường, cũng không được che chỉ dẫn. Quán Internet cũng không được bịt cửa khóa cửa sổ. Không làm được thì là vi phạm pháp, bạn có thể báo việc. Phòng nhà ở dưới lòng đất, chỉ có một lối ra, hành lang chất đầy đồ vật, thì đổi chỗ. +- Lợi ích: Đây là nghĩa vụ pháp lý của người kinh doanh. Không làm được thì là vi phạm pháp, bạn có thể báo việc. Nguyên văn quy chếnh: 「营业期间，娱乐场所应当保证疏散通道和安全出口畅通，不得封堵、锁闭疏散通道和安全出口，不得在疏散通道和安全出口设置栅栏等影响疏散的障碍物」, tức trong thời gian hoạt động, nơi giải trí phải bảo đảm lối sơ tán và lối ra khẩn cấp thông suốt, không được chặn hay khóa lối sơ tán và lối ra khẩn cấp, không được dựng rào chắn hay vật cản ảnh hưởng tới sơ tán trên các lối đó. Quy chếnh còn yêu cầu 「应当在疏散通道和安全出口设置明显指示标志，不得遮挡、覆盖指示标志」, tức phải đặt biển chỉ dẫn rõ ràng trên lối sơ tán và lối ra khẩn cấp, không được che hay phủ biển chỉ dẫn. Quán Internet cũng vậy, bị yêu cầu 「营业期间禁止封堵或者锁闭门窗、安全疏散通道和安全出口」, tức trong thời gian hoạt động cấm chặn hoặc khóa cửa sổ, lối sơ tán an toàn và lối ra khẩn cấp, đồng thời 「禁止明火照明和吸烟」, tức cấm dùng lửa trần để chiếu sáng và cấm hút thuốc. +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2006). Quy chếnh quản lý nơi giải trí, 娱乐场所管理条例（国令第 458 号, Điều hai mươi, hai mươi mốt）. ；Hội đồng Nhà nước (国务院) (2002). Quy chếnh quản lý nơi kinh doanh dịch vụ truy cập Internet, 互联网上网服务营业场所管理条例（国令第 363 号, Điều hai mươi mốt）. +- Ghi chú: Gặp bất kỳ điều nào sau đây thì đổi chỗ: phòng nhà ở dưới lòng đất, chỉ có một lối ra, hành lang chất đầy đồ vật, đèn chỉ dẫn lối ra khẩn cấp không sáng. Loại chỗ này khi xảy ra cháy thì khói nặng, hành lang hẹp, người đông, mà phần lớn người đã uống rượu. Thời gian để bạn chạy ra tính bằng giây. Số điện thoại báo việc theo quy chếnh phải được ghi trên biển cảnh báo trong chỗ +### 2. Gọi bảng giá trước khi gọi món, nơi giải trí bắt buộc niêm yết giá, không chấp nhận 「chơi xong mới tính」 + +- Chi phí: Không tốn tiền. Một lần đòi bảng giá. Chỗ khó là phải mở miệng hỏi giá trước mặt người ta, hơi khó ngại. +- Nói thẳng: Theo quy định, nơi giải trí bán dịch vụ và bán hàng đều phải niêm yết giá, còn phải chủ động đưa bảng giá cho bạn xem. Cái gọi là 「rượu giá cắt cổ」hầu như xuất hiện ở nơi đã không xem bảng giá rồi tính tiền theo giá nói miệng. Vào phòng nhà ở, hãy xin bảng giá rồi chụp ảnh. Gọi món xong thì xin một bản xác nhận. Có tranh chấp khi tính tiền thì gọi 12315 ngay tại chỗ. +- Lợi ích: Nguyên văn quy chếnh: 「娱乐场所提供娱乐服务项目和出售商品，应当明码标价，并向消费者出示价目表；不得强迫、欺骗消费者接受服务、购买商品。」, tức nơi giải trí cung cấp dịch vụ giải trí và bán hàng phải niêm yết giá và đưa bảng giá ra cho người tiêu dùng; không được ép buộc hay lừa dỗi người tiêu dùng chấp nhận dịch vụ, mua hàng. Vi phạm pháp luật về giá thì bị xử phạt theo pháp luật. Phần lớn tranh chấp kiểu 「rượu giá cắt cổ」xảy ra ở loại chỗ không đưa bảng giá ra rồi tính tiền theo giá nói miệng +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2006). Quy chếnh quản lý nơi giải trí, 娱乐场所管理条例（国令第 458 号, Điều hai mươi chín, năm mươi ba）. +- Ghi chú: Cách làm: vào phòng nhà ở hãy xin một bảng giá, bản giấy hay điện tử đều được, cầm được là chụp ảnh ngay. Gọi món xong thì xin một bản đã xác nhận. Món 「pha chế」hay 「đĩa trái cây」mà nhân viên giới thiệu, cứ hỏi giá trước rồi hãy gọi. Có tranh chấp khi tính tiền thì gọi 12315 ngay tại chỗ. Đừng trả tiền trước rồi mới quyền lực, đã trả tiền thì việc chứng minh khó hơn nhiều +### 3. Trong chỗ vui có người đưa 「thứ」thì lập tức rời đi, chứa chỗ và cung cấp đều không phải 「giúp bạn bè」 + +- Chi phí: Không tốn tiền. Đứng dậy và đi ra. Chỗ khó là phải rời bàn ngay trước mặt cả nhóm. +- Nói thẳng: Nơi giải trí không được buôn bán ma túy, không được cung cấp ma túy, cũng không được để người ta hút ma túy ở bên trong, vi phạm thì bị đình chỉ hoạt động 3 đến 6 tháng, nặng hơn thì thu hồi giấy phép. Bản thân bạn hút thì bị xử phạt hành chính. Để người ta 「dùng một lần」trong phòng nhà ở bạn đặt hoặc nơi bạn ở chính là chứa chỗ cho người khác hút ma túy, thuộc tội phạm hình sự. Có người lôi ra bột không rõ, viên nén hay điếu thuốc điện tử mời bạn thử, thì lập tức đi. +- Lợi ích: Quy chếnh cấm rõ nơi giải trí 「贩卖、提供毒品，或者组织、强迫、教唆、引诱、欺骗、容留他人吸食、注射毒品」, tức buôn bán, cung cấp ma túy, hoặc tổ chức, ép buộc, xúi giục, dỗ dỗ, lừa dỗi, chứa chỗ cho người khác hút hoặc tiêm ma túy. Người làm nghề cũng không được hút ma túy, không được tạo điều kiện cho khách làm các việc trên. Vi phạm thì cơ quan công an thu giữ thu nhập bất hợp pháp và tài sản phi pháp, yêu cầu đình chỉ hoạt động 3 đến 6 tháng. Tình tiết nặng thì thu hồi giấy phép kinh doanh nơi giải trí. Quy chếnh còn yêu cầu nơi giải trí treo biển cảnh báo chống ma túy, cấm cờ bạc, cấm mại dâm ở vị trí dễ thấy trong sảnh, phòng nhà ở và phòng riêng, đồng thời ghi số điện thoại báo việc +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2006). Quy chếnh quản lý nơi giải trí, 娱乐场所管理条例（国令第 458 号, Điều mười bốn, ba mươi, bốn mươi hai）. +- Ghi chú: Chỉ cần có người lấy ra bột không rõ, viên nén hay điếu thuốc điện tử và mời bạn 「thử thử」là rời đi ngay, đừng ở lại xem náo nhiệt. Rủi ro bạn tự gánh chịu còn nặng hơn nơi đó. Bản thân hút ma túy vốn đã bị xử phạt hành chính. Để bạn bè 「dùng một lần ở đây」trong phòng nhà ở bạn đặt hoặc nơi ở của bạn thì cấu thành chứa chỗ cho người khác hút ma túy, là tội phạm hình sự +### 4. Không ăn đồ ngọt và đồ ăn vặt người lạ đưa, không uống thức uống đã rời khỏi tầm mắt, không nhận điếu thuốc điện tử người khác đưa + +- Chi phí: Không tốn tiền. Người ta đưa thì nhất quyết không nhận. Ly của bạn đã rời khỏi tầm mắt thì đổi ly mới. +- Nói thẳng: Ma túy ngày nay không còn hình dạng của ma túy. Thuốc lá điện tử 「gây say」trộn chất tổng hợp cần sa trông gần giống thuốc lá điện tử thường, có người chuyên đưa cho người ta thử trong quán bar, KTV và khách sạn. Người lừa người khác hút bị giam giữ hành chính 10 đến 15 ngày, phạt 1000 đến 5000 tệ; người bị lừa hút cũng bị giam giữ 10 đến 15 ngày. Thức uống đã rời khỏi tầm mắt thì đổi ly mới. +- Lợi ích: Ma túy ngày nay không còn hình dạng của ma túy. Nhóm chất tổng hợp cần sa từ ngày 1 tháng 7 năm 2021 bị quản lý trọn nhóm, cả nhóm này được coi là ma túy. Có người trộn nó vào thuốc lá điện tử, làm thành thuốc lá điện tử 「gây say」. Trong vụ án được Viện kiểm sát nhân dân tối cao đưa ra, công an tài văn mô tả rằng thuốc lá này 「可以通过添加食用颜料变成各种颜色、各种口味」, tức có thể biến thành mọi màu, mọi vị bằng cách thêm màu ăn được, và 「在外观上和普通电子烟基本一致」, tức bên ngoài gần như không khác thuốc lá điện tử thường. Kẻ bán rao răm trong quán bar, KTV, khách sạn, mời gọi bạn bè tới, xúi người ta hút. Trong bài đưa tin, trong các vụ hút ma túy được xử lý ở địa phương trong năm đó, người chưa thành niên chiếm gần bốn phần mười. Ethinolate, tên hoá chất là 依托咪酯, từ ngày 1 tháng 10 năm 2023 được đưa vào danh mục chất tâm thần loại hai. Lừa người khác hút vốn đã là một mức vi phạm: kẻ dỗ dỗ, xúi giục, lừa dỗi hoặc ép buộc người khác hút, tiêm ma túy bị giam giữ hành chính 10 đến 15 ngày kèm phạt 1000 đến 5000 tệ. Thành tội phạm thì bị xử 3 năm trở xuống, tình tiết nặng thì 3 đến 7 năm, có ép buộc thì 3 đến 10 năm, thủ đạo với người chưa thành niên thì nặng hơn. Hậu quả cũng rơi vào người bị lừa: hút, tiêm ma túy bị giam giữ 10 đến 15 ngày và có thể phạt dưới 3000 tệ. Tình tiết nhẹ thì bị giam giữ dưới 5 ngày hoặc phạt dưới 1000 tệ. Còn có thể bị đồng thời cấm vào nơi giải trí trong 6 tháng đến 1 năm. Nghiện rồi thì bị cấu lệnh cai ma túy tại cộng đồng 3 năm. Không chịu cai tại cộng đồng, hoặc hút lại trong thời gian cai thì bị cai ma túy cách ly bắt buộc 2 năm, dài nhất có thể kéo dài thêm 1 năm (toàn quốc, Luật Xử phạt hành chính về trật tự công cộng, áp dụng từ tháng 1 năm 2026) +- Mức bằng chứng: A +- Nguồn: Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 江苏苏州吴江区：办案检察官揭秘披着潮流外衣的新型毒品. ；Cục Quản lý Dược phẩm nhà nước, Bộ Công an, Ủy ban Quốc gia Y tế và Sức khoẻ (国家药监局、公安部、国家卫生健康委) (2023). 关于调整麻醉药品和精神药品目录的公告（2023 年第 120 号）. （Cục Quản lý Dược phẩm Thượng Hải đăng lại, trang gốc nmpa.gov.cn trả về 412 trên máy này）；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2025). Luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订, Điều tám mươi tư, tám mươi lăm）. ；Đại hội Đại biểu Quốc dân (全国人大) (1997). Bộ luật Hình sự, 刑法（Điều ba trăm năm mươi ba）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2007). Luật chống ma túy, 禁毒法（Điều ba mươi ba, ba mươi tám, bốn mươi bảy）. （Chính phủ tỉnh Quảng Đông đăng lại） +- Ghi chú: Không chỉ nơi giải trí, cửa trường học, tụ họp bạn bè, xe đường dài cũng vậy. Cho đồ ngọt, cho thức uống hay cho điếu thuốc điện tử thì đều thế. Sau khi bị cho thuốc thường đi kèm cướp đoạt, xâm hại tình dục và tống tiền, phần này không có thống kê chính thức công khai, thuộc phán đoán theo kinh nghiệm. Thức uống chỉ cần rời khỏi tầm mắt là đổi ly mới, dễ hơn nhiều so với truy đòi sau. Trong chỗ vui có người lấy ra bột không rõ, viên nén, điếu thuốc điện tử thì làm gì, xem mục 3 của phần này. Giúp người ta mang đồ xem mục 29 ở phần 8. +### 5. Vào quán Internet dùng giấy tùy thân của bạn, đừng cho mượn, cũng đừng dùng giấy của người khác + +- Chi phí: Không tốn tiền. +- Nói thẳng: Theo quy định, quán Internet phải kiểm tra và ghi đăng ký giấy tùy thân của người lên mạng, còn phải ghi lại thông tin lên mạng, lưu giữ không dưới 60 ngày. Cơ quan công an và cơ quan văn hoá có thể tra bất cứ lúc nào. Trong khoảng thời gian đó quán Internet không được sửa, cũng không được xoá. Dùng giấy của ai lên mạng thì trong khoảng thời gian đó, việc làm trên máy đó ghi vào tên người đó. +- Lợi ích: Nguyên văn quy chếnh: 「互联网上网服务营业场所经营单位应当对上网消费者的身份证等有效证件进行核对、登记，并记录有关上网信息。登记内容和记录备份保存时间不得少于 60 日」, tức đơn vị kinh doanh nơi cung cấp dịch vụ truy cập Internet phải kiểm tra, đăng ký giấy tùy thân và các giấy tờ hợp lệ khác của người tiêu dùng lên mạng, và ghi lại thông tin liên quan tới việc lên mạng. Nội dung đăng ký và bản sao lưu hồ sơ phải lưu giữ không dưới 60 ngày. Khi cơ quan quản lý văn hoá và cơ quan công an tra cứu theo đúng pháp luật, quán Internet phải cung cấp. Trong thời hạn lưu giữ không được sửa hay xoá. Vì vậy dùng giấy của ai lên mạng thì trong khoảng thời gian đó, việc làm trên máy đó ghi vào tên người đó +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2002). Quy chếnh quản lý nơi kinh doanh dịch vụ truy cập Internet, 互联网上网服务营业场所管理条例（国令第 363 号, Điều hai mươi ba）. +- Ghi chú: Đây cùng một việc với mục 「không cho mượn giấy tùy thân」ở phần 9. Quán Internet không được phục vụ người chưa thành niên, ở cửa ra vào chỗ dễ thấy phải treo biển cấm người chưa thành niên. Nơi giải trí có ca hát cũng không phục vụ người chưa thành niên. Giúp người chưa thành niên 「quẹt giấy vào quán Internet」thì bị phạt là quán Internet, nhưng dữ liệu thì nằm lại trên giấy tờ của bạn +### 6. Đừng đến nơi chơi luận án hay phòng giải đố mở trong tòa nhà dân cư hoặc dưới tầng hầm một + +- Chi phí: Không tốn tiền. Trước khi đặt, hỏi một câu địa chỉ ở tầng mấy. +- Nói thẳng: Thông tư của năm 2022 của năm bộ ngành ghi rõ, địa điểm giải trí theo kịch bản không được mở trong tòa nhà dân cư, cũng không được mở dưới tầng hầm một. Loại chỗ này tối, vách ngăn nhiều, đạo cụ dễ cháy, cửa thường phải giải câu đố mới mở được. Một khi xảy ra cháy, chạy ra khó hơn cửa hàng thường rất nhiều. Trước khi đặt, xem địa chỉ và tầng trên trang đánh giá, không hợp lệ thì đổi chỗ. +- Lợi ích: Thông tư riêng của năm bộ ngành năm 2022 viết rất cụ thể: 「剧本娱乐经营场所不得设在居民楼内、建筑物地下一层以下（不含地下一层）等地」, tức nơi kinh doanh giải trí theo kịch bản không được đặt trong tòa nhà dân cư, hay ở dưới tầng hầm một của công trình (không tính tầng hầm một). Thông tư còn yêu cầu nơi đó thực hiện các quy định của Luật an toàn sản xuất và Luật phòng cháy chữa cháy, 「常态化开展火灾风险自知、自查、自改，提高紧急情况下的组织疏散逃生和初起火灾扑救能力」, tức thường xuyên tự biết, tự kiểm tra, tự sửa rủi ro cháy, nâng cao khả năng tổ chức sơ tán thoát hiểm và dập tắt cháy khi mới bắt đầu trong tình huống khẩn cấp. Cùng thông tư đó còn yêu cầu kịch bản phải có ghi chú về độ tuổi phù hợp. Bối cảnh không phù hợp với người chưa thành niên thì không được cho người chưa thành niên vào. Ngoài ngày lễ tết theo luật định, ngày nghỉ và kỳ nghỉ hè, không được cung cấp hoạt động giải trí theo kịch bản cho người chưa thành niên +- Mức bằng chứng: A +- Nguồn: Bộ Văn hoá và Du lịch, Bộ Công an, Bộ Xây dựng và Nhà ở, Bộ Quản lý Khẩn cấp, Cục Quản lý thị trường quốc gia (文化和旅游部、公安部、住房和城乡建设部、应急管理部、市场监管总局) (2022). 关于加强剧本娱乐经营场所管理的通知. +- Ghi chú: Đặc điểm của loại chỗ này là tối, vách ngăn nhiều, đạo cụ dễ cháy, cửa thường phải 「giải câu đố」mới mở được. Một khi xảy ra cháy, khó chạy ra hơn cửa hàng thường nhiều. Trước khi đặt, ngay trên trang đánh giá đã thấy được tầng, cũng thấy được có phải tòa nhà dân cư không. Không hợp lệ thì đổi chỗ luôn +### 7. Tâm trạng tệ thì đi bộ hoặc chạy, cỡ tác động chống trầm cảm (độ lớn hiệu quả) tỷ lệ thuận với cường độ + +- Chi phí: Không tốn tiền. Mỗi lần khoảng nửa tiếng. Chỗ khó là lúc tâm trạng tệ lại càng không muốn động, phải tự đẩy mình một cái. +- Nói thẳng: Khi tâm trạng tệ, đi bộ nhanh hoặc chạy chậm cho cải thiện rõ nhất. Kế tiếp là yoga, tập tạp, thái quyền và công năng, mấy cách này chênh lệch không nhiều. Cường độ càng cao thì hiệu quả càng tốt, tập tạp và yoga dễ duy trì nhất. Chất lượng bằng chứng ở lĩnh vực này hơi thấp, điều chắc chắn là vận động hơn không vận động, hãy chọn cái mà bạn duy trì nổi. +- Lợi ích: Đây là kết quả gộp 218 thử nghiệm chia nhóm ngẫu nhiên. Liên quan tới 495 nhóm, 14170 người tham gia đạt tiêu chuẩn lâm sàng trầm cảm nặng. Thuật toán cho các loại tập luyện so với nhau (phân tích tổng hợp mạng Bayes). Đối chứng dùng để so là chăm sóc thường và viên giả dược (đối chứng chủ động). Độ lớn hiệu quả dùng Hedges' g, số càng âm thì triệu chứng càng cải thiện. Đi bộ nhanh hoặc chạy chậm −0.62（95% khoảng tin cậy −0.80 đến −0.45, nghĩa là giá trị thật nằm khoảng đó）. Yoga −0.55（−0.73 đến −0.36）. Tập tạp −0.49（−0.69 đến −0.29）. Kết hợp sức bền −0.43（−0.61 đến −0.24）. Thái quyền hoặc công năng −0.42（−0.65 đến −0.21）. Nguyên văn nói 「运动的效果与所处方的强度成正比」, tức hiệu quả tập luyện tỷ lệ thuận với cường độ được kê: cường độ càng cao, hiệu quả càng tốt. Tập tạp và yoga được chấp nhận nhiều nhất, dễ duy trì nhất +- Mức bằng chứng: A +- Nguồn: Noetel M, Sanders T, Gallardo-Gómez D, et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ, 384, e075847. +- Ghi chú:Tranh cãi. Chính tác giả ghi rõ, trong 218 nghiên cứu chỉ có 1 nghiên cứu đủ cẩn thận, khó bị thiết kế làm sai lệch. Theo cách chấm điểm thông dụng, đi bộ nhanh và chạy chậm chỉ ở mức 「thấp」đáng tin, còn lại là 「rất thấp」. Vận động cũng không che được với người tham gia, người ta biết mình đang tập thì dễ cảm thấy khá hơn, không bỏ được phần kỳ vọng đó. Hướng kết luận vẫn dùng được: vận động hơn không vận động, cường độ đừng quá thấp, chọn cái mình duy trì nổi. Đây là phần mở rộ của mục 「khi tâm trạng tệ thì vận động trước đã」ở phần 3, không thay thế việc điều trị đúng quy trình sau khi đã chẩn đoán +### 8. Khi lo âu nhiều tới mức ảnh hưởng cuộc sống, lớp giảm căng thông chánh niệm hiệu quả không kém thuốc chống lo âu thường dùng + +- Chi phí: Đăng ký lớp học tốn một ít tiền. Khóa học 8 tuần, mỗi tuần học một buổi, còn phải luyện ở nhà mỗi ngày. Chỗ khó là luyện liên tục 8 tuần, ngày nào cũng phải luyện. +- Nói thẳng: Ở người được chẩn đoán rối loạn lo âu, học lớp giảm căng thông chánh niệm 8 tuần thì mức cải thiện triệu chứng gần như giống uống thuốc chống lo âu escitalopram. Người uống thuốc có gần tám phần mười gặp tác dụng phụ, có người phải bỏ dở. Người học lớp khoảng một phần rưỡi gặp tác dụng phụ, không ai phải bỏ. Khóa học phải là khóa chuẩn hoá 8 tuần, một ứng dụng thiền nào cũng tính là không. +- Lợi ích: Đây là thử nghiệm do 3 trung tâm y tế học thuật ở thành phố của Mỹ thực hiện, mục đích là xem học lớp có kém hơn uống thuốc không, tức thử nghiệm chứng nhận phi tương đương có nhóm đối chứng ngẫu nhiên. 276 người lớn được chẩn đoán rối loạn lo âu chia ngẫu nhiên theo tỉ lệ 1:1 thành hai nhóm bằng nhau. Một nhóm học lớp giảm căng thông chánh niệm 8 tuần. Một nhóm uống escitalopram (liều linh hoạt 10 đến 20 mg). 208 người làm hết khóa. Lúc kết thúc, tính theo điểm mức độ lo âu do bác sĩ chấm (Thang đo ấn tượng lâm sàng toàn thể, CGI), nhóm chánh niệm giảm 1.35, nhóm thuốc giảm 1.43. Chênh lệch giữa hai nhóm là −0.07 (95% CI −0.38 đến 0.23, nghĩa là chênh lệch thật nằm khoảng đó; P=0.65, nghĩa là chênh lệch nhỏ này không phân biệt được với không có chênh lệch). Cận dưới của khoảng này không vượt qua đường 「không kém hơn đáng kể」đã vạch sẵn (giá trị ngưỡng phi tương đương −0.495). Chênh lệch tác dụng phụ rất lớn: nhóm thuốc có 78.6% gặp ít nhất một tác dụng phụ liên quan tới thử nghiệm, 10 người phải bỏ dở. Nhóm chánh niệm là 15.4%, không ai phải bỏ +- Mức bằng chứng: A +- Nguồn: Hoge EA, Bui E, Mete M, Dutton MA, Baker AW, Simon NM (2023). Mindfulness-based stress reduction vs escitalopram for the treatment of adults with anxiety disorders: a randomized clinical trial. JAMA Psychiatry, 80(1), 13-21. +- Ghi chú: Thử nghiệm dùng khóa giảm căng thông chánh niệm chuẩn hoá 8 tuần (MBSR), một ứng dụng thiền nào cũng không thể ngang với nó. Nó dành cho người đã đạt tiêu chuẩn chẩn đoán rối loạn lo âu, nếu cần đi khám bác sĩ thì vẫn phải đi. Nếu vì lo tác dụng phụ của thuốc mà bạn cứ trì hoãn không điều trị, đây là một lựa chọn có sức bằng chứng tương đương +### 9. Muốn bình tĩnh lại ngay tại chỗ thì dùng 5 phút 「thở dài vòng lặp」: hít vào hai đoạn, kéo dài hơi thở ra + +- Chi phí: Không tốn tiền. Mỗi ngày 5 phút. +- Nói thẳng: Làm bài tập thở 5 phút mỗi ngày, kiên trì một tháng, mức cải thiện tâm trạng và độ chậm của hơi thở đều tốt hơn thiền chánh niệm dài tương ứng. Trong các phương pháp thở, 「thở dài vòng lặp」có hơi thở ra kéo dài là tốt nhất. Cách làm là hít vào bằng mũi một hơi, rồi hít thêm một hơi nhỏ cho phổi đầy, sau đó thở ra chậm bằng miệng cho hết. Đây chỉ là một thử nghiệm, tốt hay không do người đó tự báo, nhưng không tốn tiền. +- Lợi ích: Đây là thử nghiệm đối chứng ngẫu nhiên làm tại nhà, so ba bài tập thở 5 phút mỗi ngày với thiền chánh niệm cùng thời lượng, kéo dài 1 tháng. Ba phương pháp thở là: thở dài vòng lặp (nhấn mạnh kéo dài hơi thở ra), thở hộp (hít, nín, thở đều nhau), và thở tăng cường luân phiên có nín (hít dài, thở ngắn). Gom dữ liệu của từng người, từng lần vào tính chung (mô hình hiệu ứng ngẫu nhiên), bài tập thở đều hơn thiền chánh niệm, nhất là thở dài vòng lặp lấy hơi thở ra là chính. Nó tốt hơn ở cả cải thiện tâm trạng (P<0.05, nghĩa là chênh lệch này khó do tình cờ) và giảm tần suất thở (P<0.05) +- Mức bằng chứng: B +- Nguồn: Balban MY, Neri E, Kogon MM, et al. (2023). Brief structured respiration practices enhance mood and reduce physiological arousal. Cell Reports Medicine, 4(1), 100895. +- Ghi chú: Cách làm là hít vào bằng mũi một hơi, hít thêm một hơi nhỏ trong thời gian ngắn cho phổi đầy, rồi thở ra chậm và dài bằng miệng cho hết, lặp lại trong 5 phút. Chỉ có một thử nghiệm này, làm ở nhà, tốt hay không do người đó tự điền bảng hỏi báo, số người cũng không nhiều. Hiệu quả chỉ hơn thiền một chút, không giải quyết được vấn đề. Nhưng nó không tốn tiền, chỉ 5 phút, làm được ở bất cứ lúc nào, hiệu quả chi phí cao +### 10. Coi 「gặp gỡ với người khác đều đặn」là một khoản chi cho sức khoẻ, đừng chỉ tìm người khi tâm trạng tệ + +- Chi phí: Tốn một ít tiền, còn phải bỏ thời gian đi gặp. Chỗ khó là lúc tâm trạng tốt vẫn phải chủ động rủ người. +- Nói thẳng: Người ít giao lưu với mọi người có xác suất chết cùng kỳ cao hơn khoảng ba phần mười. Người cảm thấy cô đơn cao hơn khoảng một phần tư, người sống một mình cao hơn khoảng ba phần mười, người dưới 65 tuổi rõ hơn. Đây đều là quan sát theo dõi, cũng có thể vì người sức khoẻ vốn đã yếu thì dễ sống một mình hơn. +- Lợi ích: Đây là kết quả gộp 70 nghiên cứu ghi nhận trước rồi theo dõi sau (phân tích tổng hợp), tổng cộng 3.400.000 người tham gia, theo dõi trung bình 7 năm. Cô lập xã hội, cảm giác cô đơn, sống một mình đều liên quan tới việc tỷ lệ tử vong tăng. Sau khi trừ các yếu tố gây nhiễu như tuổi, tình trạng sức khoẻ, tỉ số odds bình quân có trọng số lần lượt là: cô lập xã hội 1.29, cô đơn 1.26, sống một mình 1.32. Quy đổi ra thì xác suất chết cùng kỳ cao hơn lần lượt khoảng ba phần mười, khoảng một phần tư, khoảng ba phần mười. Độ lớn hiệu quả của ba cái này gần bằng nhau, và lớn hơn trong nhóm dưới 65 tuổi so với nhóm người cao tuổi +- Mức bằng chứng: A +- Nguồn: Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D (2015). Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspectives on Psychological Science, 10(2), 227-237. +- Ghi chú:Tranh cãi. Đây đều là mối liên hệ quan sát theo dõi một nhóm người, không phải thử nghiệm đối chứng có nhóm, nên không tách được nguyên nhân với hậu quả. Người sức khoẻ vốn đã yếu thì dễ sống một mình hơn, ít giao lưu hơn. Tác giả coi các quan hệ xã hội là một yếu tố sức khoẻ cùng cỡ với hút thuốc và béo phì, nhưng không có thử nghiệm ngẫu nhiên nào chứng minh 「tăng giao lưu xã hội có giảm tỷ lệ tử vong」. Mục này không mâu thuẫn với mục 「giảm bớt những quan hệ làm bạn hao hụt sức lực」ở phần 3: mục đó chặt bỏ những quan hệ làm bạn hao sức, mục này giữ lại vài mối quan hệ giúp bạn hồi phục sức lực +### 11. Ở gần cây xanh thì cứ dùng, đi bộ, đi dạo, ngồi nghỉ đều tính + +- Chi phí: Không tốn tiền. Tiện đường thì làm luôn. +- Nói thẳng: Cây xanh quanh chỗ ở càng tốt trong bán kính 500 mét thì xác suất chết cùng kỳ càng thấp, tính mỗi 0.1 điểm chỉ số thực vật thì thấp hơn khoảng 4%. Đo là cây xanh quanh chỗ ở, không phải số lần bạn đi công viên. Ở khu cây xanh tốt, người ở thường giàu hơn, không khí cũng tốt hơn. Vì vậy đừng hiểu là đi công viên là sống thêm, khi chọn chỗ ở thì tính cả cây xanh quanh đó vào. +- Lợi ích: Đây là kết quả gộp 9 nghiên cứu theo dõi nhóm người nhiều năm (phân tích tổng hợp các nghiên cứu mẫu theo thời gian), liên quan tới 7 quốc gia, 8.320.000 người. Đo mức cây xanh quanh chỗ ở trong bán kính 500 mét, bằng chỉ số thực vật chuẩn hoá (NDVI, con số quy đổi độ phủ xanh mặt đất). Mỗi 0.1 điểm chỉ số cây xanh tăng thêm, tỉ số nguy cơ gộp của tỷ lệ tử vong toàn nguyên là 0.96 (95% CI 0.94–0.97), xác suất chết cùng kỳ thấp hơn khoảng 4%, giá trị thật nằm khoảng thấp hơn 3% tới thấp hơn 6%. Trong 9 nghiên cứu thì có 7 nghiên cứu thấy cây xanh tốt hơn thì tỷ lệ tử vong thấp hơn, và chênh lệch cũng khó do tình cờ. 2 nghiên cứu còn lại không thấy liên hệ +- Mức bằng chứng: A +- Nguồn: Rojas-Rueda D, Nieuwenhuijsen MJ, Gascon M, Perez-Leon D, Mudu P (2019). Green spaces and mortality: a systematic review and meta-analysis of cohort studies. The Lancet Planetary Health, 3(11), e469-e477. +- Ghi chú:Tranh cãi. Kết quả các nghiên cứu rất khác nhau (I²=95%, số này càng lớn thì các nghiên cứu càng không khớp nhau). Hơn nữa nó đo mức cây xanh quanh nơi ở, không liên quan gì tới số lần bạn đi công viên. Người ở nơi cây xanh tốt thường cũng giàu hơn, không khí tốt hơn, hay vận động hơn, mấy thứ đó không tách được với cây xanh thì cái nào là nguyên nhân. Đừng đọc thành 「đi công viên là sống thêm」, phần dùng được là tính cả cây xanh quanh nhà vào lúc chọn chỗ ở, chi phí bằng không. Con số NDVI này, đất trống và mặt bê tông gần 0, thảm thực vật dày gần 1. Nên 「0.1」xấp xỉ khác biệt giữa dãy phố trọc và dãy phố có hàng cây bàng đầy đủ \ No newline at end of file diff --git "a/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" "b/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" new file mode 100644 index 00000000..a9946a4a --- /dev/null +++ "b/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" @@ -0,0 +1,73 @@ +[← 回总目录](../README.md) + +# 28. 别为了外形把身体搞坏 + +本节说的是「为了改变外形而对身体动手」的几种做法：饿肚子、吃药、打针、开刀。这几件事，决定多半是和别人比较之后做的，代价却全落在自己身上，而且相当一部分改不回来。 + +本节不评价任何人想把自己变成什么样子，那是各人自己的事。同一个目标，可以找「有处方、有资质、有复查」的，也可以「网购、美容院、自己加量」，本节只比这两种做法的风险差多少。算账主要看身体上实际出了什么事，比如失明、住院、死亡。医美那两条还涉及非法行医的刑事责任。 + +### 1. Đừng kiểm soát cân nặng bằng ăn kiêng cực đoan, nhịn ăn hay tự nôn; muốn giảm thì giảm bằng vận động + +- Chi phí: Không tốn tiền. Ngược lại còn tiết kiệm được tiền mua sản phẩm giảm cân. +- Nói thẳng: Người mắc chán ăn tâm thần có xác suất chết xấp xỉ 5.9 lần người cùng tuổi. Theo dõi 1000 người trong một năm thì chết 5.1 người. Trong số người chết, cứ năm người thì một người chết tự tử. Lối vào của nó là ăn kiêng: theo dõi học sinh 44 trường trung học trong ba năm, nữ sinh ăn kiêng nghiêm trọng có xác suất mắc rối loạn ăn uống gấp 18 lần nữ sinh không kiêng. Muốn kiểm soát cân nặng thì vận động nhiều hơn, đừng ăn ít đến mức đói. +- Lợi ích: Tổng hợp 36 nghiên cứu. Nhóm chán ăn tâm thần được theo dõi tổng cộng 166.642 người-năm (tổng số năm theo dõi của từng người). Tỷ lệ tử vong chuẩn hoá của họ là 5.86, tức xác suất chết xấp xỉ 5.9 lần người cùng tuổi. Cứ 1000 người theo dõi trong một năm thì có 5.1 ca tử vong, tức 1000 người theo dõi một năm chết 5.1 người. Trong số người chết, khoảng 1/5 chết tự tử. Tỷ lệ tử vong chuẩn hoá của chứng ăn vô độ (bulimia) là 1.93, của rối loạn ăn uống khác là 1.92, tức mỗi loại xấp xỉ 1.9 lần người cùng tuổi. Nghiên cứu khác làm ở 44 trường trung học tại Victoria, Úc, theo dõi học sinh 14–15 tuổi trong 3 năm, cùng một nhóm học sinh được khảo sát 6 đợt trong ba năm đó. Nữ sinh mắc rối loạn ăn uống mới là 21.8 ca trên 1000 người theo dõi một năm, tức 1000 nữ sinh theo dõi một năm xuất hiện 21.8 ca mới. Nữ sinh ăn kiêng nghiêm trọng có nguy cơ gấp 18 lần người không kiêng, người ăn kiêng mức vừa là 5 lần. Phân tích sau đó hiệu chỉnh cho việc đã từng ăn kiêng và triệu chứng tâm thần, loại bỏ ảnh hưởng của「trước đây đã từng ăn kiêng」và「vốn đã có triệu chứng tâm thần」. Sau khi loại bỏ, BMI (chỉ số béo gầy tính bằng trọng lượng chia chiều cao bình phương), lượng vận động và giới tính không còn dự báo rối loạn ăn uống mới. Tác giả kết luận「ở thanh thiếu niên, dùng vận động thay vì hạn chế ăn để kiểm soát cân nặng có vẻ làm giảm nguy cơ rối loạn ăn uống」 +- Mức bằng chứng: A +- Nguồn: Arcelus J và cộng sự (2011). Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry. ；Patton GC và cộng sự (1999). Onset of adolescent eating disorders: population based cohort study over 3 years. BMJ. +- Ghi chú: Ở đây「ăn kiêng nghiêm trọng」nghĩa là duy trì lâu dài mức ăn thấp hơn nhiều so với nhu cầu cơ thể, không phải bỏ một bữa hay bỏ ăn khuya. Nếu đã tự nôn, ăn xong lại tự trách, lên cân rồi không dừng được thì đi khoa tâm thần hoặc khoa tâm lý lâm sàng của bệnh viện đa khoa. Đây là bệnh có cách điều trị đúng quy trình, không phải vấn đề ý chí. Quan hệ giữa BMI và tỷ lệ tử vong xem mục 33 ở phần 2; mục đó nói về khoảng cân nặng dài hạn, không nói về biện pháp giảm cân. +### 2. Trước khi tiêm, cấy chỉ hay phẫu thuật, kiểm tra hai thứ: giấy phép cơ sở có ghi「y học thẩm mỹ」không, và người thao tác có phải bác sĩ chủ trị không + +- Chi phí: Không tốn tiền. Kiểm tra hai loại giấy tờ này chỉ mất vài phút. Cơ sở phải treo Giấy phép hành nghề cơ sở y tế ở đại sảnh. Văn bằng của bác sĩ tra được trên hệ thống tra cứu bác sĩ hành nghề của ngành y tế. +- Nói thẳng: Đâm kim, mổ hay dùng thiết bị y tế làm rách da hay đi vào trong cơ thể thì cũng tính là khám bệnh. Chỉ được làm ở cơ sở mà giấy phép có ghi「y học thẩm mỹ」, và người thao tác phải là bác sĩ hành nghề đã đăng ký. Làm ở tiệm làm đẹp, studio hay phòng khách sạn là hành nghề y tế bất hợp pháp. Người thao tác tình hình nghiêm trọng thì bị phạt tù dưới 3 năm; làm người khác bị thương thì 3 đến 10 năm; gây chết người thì từ 10 năm trở lên. +- Lợi ích: Điều hai của《Biện pháp quản lý dịch vụ y học thẩm mỹ》（医疗美容服务管理办法）cho định nghĩa y học thẩm mỹ. Nguyên văn là「运用手术、药物、医疗器械以及其他具有创伤性或者侵入性的医学技术方法对人的容貌和人体各部位形态进行的修复与再塑」, tức là sửa chữa và tạo hình lại diện mạo cùng hình dạng các bộ phận của cơ thể bằng phẫu thuật, thuốc, thiết bị y tế và các biện pháp y học mang tính tổn thương hoặc xâm lấn. Nói cho dễ hiểu, chỉ cần làm rách da, đưa kim vào, hay đưa thứ gì đó vào trong cơ thể, thì đã tính là đi khám bệnh. Điều tám và Điều hai mươi bốn yêu cầu phải có Giấy phép hành nghề cơ sở y tế trước, đồng thời được cơ quan đăng ký chấp thuận chuyên khoa y học thẩm mỹ, thì mới được triển khai. Giấy phép không ghi「y học thẩm mỹ」thì không được làm. Điều mười một yêu cầu bác sĩ chủ trị (người trực tiếp thao tác) có đủ tư cách bác sĩ hành nghề và đã đăng ký. Làm phẫu thuật thẩm mỹ còn phải có ít nhất 6 năm công tác lâm sàng trong chuyên ngành liên quan, nha khoa thẩm mỹ từ 5 năm trở lên, y học thẩm mỹ truyền thống và da liễu thẩm mỹ mỗi bên từ 3 năm trở lên. Điều hai mươi yêu cầu trước khi thao tác phải thông báo bằng văn bản về chỉ định và chống chỉ định (trường hợp nào được làm, trường hợp nào không được làm), rủi ro y tế và những điểm cần lưu ý, rồi lấy chữ ký đồng ý.《Bộ luật Hình sự》Điều ba trăm ba mươi sáu điều chỉnh tội hành nghề y tế bất hợp pháp. Hành nghề y tế bất hợp pháp là người không có tư cách hành nghề y tế lại đi tiêm, mổ, khám bệnh cho người khác. Tình hình nghiêm trọng thì phạt tù không quá 3 năm, cải tạo không giam giữ hoặc quản chế, và phạt tiền hoặc chỉ phạt tiền. Gây tổn hại nghiêm trọng sức khỏe người đến khám thì phạt tù từ 3 đến 10 năm và phạt tiền. Gây người đến khám tử vong thì phạt tù từ 10 năm trở lên và phạt tiền +- Mức bằng chứng: A +- Nguồn: Bộ Y tế (卫生部) (2002). Biện pháp quản lý dịch vụ y học thẩm mỹ（医疗美容服务管理办法）(Số 19 của Bộ Y tế, các điều hai, tám, mười một, mười sáu, hai mươi và hai mươi bốn). ；Quốc hội Nhân dân Trung Quốc (全国人大) (1997). Bộ luật Hình sự（刑法）(điều ba trăm ba mươi sáu). +- Ghi chú: Cùng bộ biện pháp đó, Điều mười sáu còn quy định dịch vụ y học thẩm mỹ chỉ được làm ở cơ sở y tế thẩm mỹ, hoặc ở bệnh viện, phòng khám có khoa y học thẩm mỹ. Nên việc đến nhà tiêm, hay làm lén trong studio, chỉ xét về địa điểm đã không hợp quy rồi. Điều hai mươi nêu rõ, người chưa thành niên hay người chưa tự quyết định được mà làm y học thẩm mỹ thì phải có đồng ý của người giám hộ, tức là bố mẹ ký. Chăm sóc mặt, massage, làm móng thuộc thẩm mỹ đời sống, không chịu bộ quy tắc này. Nhưng chỉ cần động đến kim và dao mổ thì đã là y học thẩm mỹ. +### 3. Khi tiêm filler vào mặt, tránh lưng mũi, khoảng giữa hai lông mày và trán; đừng tiêm ở chỗ không chính quy chỉ để tiết kiệm tiền + +- Chi phí: Không tốn tiền. Chỉ là đổi chỗ tiêm, hoặc không tiêm. +- Nói thẳng: Trong số người bị mù sau khi tiêm filler, hơn một nửa tiêm ở mũi, kế đến là giữa hai lông mày, trán và rãnh mũi má, hơn tám phần mười dùng axit hyaluronic. Người bị mù chỉ có hai phần mười hồi phục thị lực hoàn toàn, hơn một phần mười hồi phục một phần, và gần hai phần mười còn bị tổn thương não. Hiện chưa có cách cấp cứu nào ổn định hữu hiệu, nên chỉ có cách tránh trước các vùng này. +- Lợi ích: Thống kê các ca được báo cáo trên tạp chí y khoa toàn cầu. Từ 2015-1 đến 2018-9 có 48 ca mất thị lực một phần hoặc hoàn toàn sau khi tiêm filler. Vùng xảy ra nhiều nhất là mũi, chiếm 56.3%. Tiếp theo là giữa hai lông mày 27.1%, trán 18.8%, rãnh mũi má (nếp gấp khi cười) 14.6%. 81.3% ca dùng chất làm đầy là axit hyaluronic. 43.8% xuất hiện thay đổi da. 18.8% xuất hiện biến chứng hệ thần kinh trung ương (não có vấn đề). 10 ca (20.8%) hồi phục thị lực hoàn toàn, 8 ca (16.7%) hồi phục một phần, số còn lại không hồi phục. Tác giả kết luận cách xử lý giữa các nơi rất khác nhau, chưa có phương pháp điều trị nào được chứng minh là liên tục có hiệu quả +- Mức bằng chứng: B +- Nguồn: Beleznay K và cộng sự (2019). Update on Avoiding and Treating Blindness From Fillers: A Recent Review of the World Literature. Aesthet Surg J. +- Ghi chú: Những con số này được đếm từng ca một từ các ca báo cáo trên tạp chí. Trong đó chỉ có số người bị hại, không có tổng số lần tiêm. Nên không tính được「tiêm một lần thì xác suất mù là bao nhiêu」, chỉ thấy được vùng nào hay xảy ra sự cố và sau đó về cơ bản không cứu được. Chứng nhận mức B cũng vì thế. Việc cơ sở chính quy làm để giảm rủi ro vốn là quy trình thường gặp: trước tiên soi rõ mạch máu dưới chỗ định tiêm, dùng kim đầu tròn thay vì kim nhọn, vừa rút kim vừa đẩy một lượng nhỏ, và chuẩn bị sẵn hyaluronidase. Mỗi khâu đều đòi hỏi người thao tác là bác sĩ được đào tạo, đó cũng là lý do đừng tìm kênh không chính quy. Trước khi làm, hãy đối chiếu mục 2 của phần này (kiểm tra giấy phép cơ sở và bác sĩ chủ trị) một lần. +### 4. Đừng mua thuốc giảm cân, cà phê giảm cân, kẹo giảm cân hay bánh mận enzyme hứa「giảm nhanh」 + +- Chi phí: Không tốn tiền. Ngược lại còn tiết kiệm tiền mua các sản phẩm này. Khó ở chỗ phải chấp nhận rằng không có chuyện「tuần nào cũng giảm được vài cân」. +- Nói thẳng: Trong nhóm sản phẩm này, thành phần bị phát hiện lén thêm thường gặp nhất là sibutramine. Nó vốn là thuốc giảm cân, đến năm 2010 thì bị rút khỏi thị trường Trung Quốc. Người thừa cân mắc bệnh tim hoặc tiểu đường dùng hơn ba năm thì nhồi máu cơ tim không tử vong cao khoảng ba phần mười, đột quỵ cao khoảng ba phần sáu. Cân nặng chỉ giảm thêm được 1.7 kg. Lúc kiểm tra đột xuất phát hiện thì bạn đã uống vào rồi. +- Lợi ích: Thử nghiệm SCOUT tuyển 9.804 người. Tất cả đều trên 55 tuổi, thừa cân hoặc béo phì, và mắc bệnh tim mạch hoặc tiểu đường typ 2. Họ được bốc thăm chia thành nhóm dùng thuốc và nhóm không dùng, trung bình điều trị 3.4 năm. Điểm cuối chính là mấy sự cố xấu đã định sẵn từ trước: nhồi máu cơ tim không tử vong, đột quỵ không tử vong, ngừng tim rồi được cứu lại, và tử vong tim mạch. Tần suất các sự cố này là 11.4% vs 10.0%. HR 1.16 (95% CI 1.03–1.31, P=0.02), tức người dùng thuốc có xác suất xảy ra cao hơn khoảng một phần sáu. Giá trị thật nằm trong khoảng cao hơn 3% đến cao hơn 31%. Nhồi máu cơ tim không tử vong 4.1% vs 3.2%, HR 1.28 (1.04–1.57), cao hơn khoảng ba phần mười. Đột quỵ không tử vong 2.6% vs 1.9%, HR 1.36 (1.04–1.77), cao hơn khoảng ba phần sáu. Tử vong tim mạch và tử vong mọi nguyên nhân không tăng; mọi nguyên nhân nghĩa là cộng tất cả các nguyên nhân lại. Sau khi bốc thăm, nhóm sibutramine trung bình giảm thêm 1.7 kg. Cục Quản lý Thực phẩm và Dược phẩm Trung Quốc cũ căn cứ đánh giá an toàn cho rằng「rủi ro của điều trị giảm cân lớn hơn lợi ích」, tháng 10 năm 2010 đã thông báo ngừng sản xuất, kinh doanh và sử dụng chế phẩm sibutramine cùng nguyên liệu. Sibutramine và phenolphthalein đều bị cấm thêm vào thực phẩm, kể cả thực phẩm chức năng +- Mức bằng chứng: A +- Nguồn: James WPT và cộng sự (2010). Effect of sibutramine on cardiovascular outcomes in overweight and obese subjects. NEJM. ；Ủy ban Giám sát thị trường thành phố Thiên Tân (天津市市场监督管理委员会). Cảnh báo「chất sibutramine hydrochloride」và「phenolphthalein」bị thêm bất hợp pháp vào một số thực phẩm（警惕部分食品中违法添加「盐酸西布曲明」和「酚酞」）. +- Ghi chú: SCOUT tuyển người vốn đã mắc bệnh tim hoặc tiểu đường, tức nhóm nguy cơ cao. Thử nghiệm này không chứng minh người trẻ khoẻ mạnh ăn vào cũng xảy ra đúng như thế. Nhưng chính vì không phân biệt được ai là người nguy cơ cao, cơ quan quản lý mới cho nó rút khỏi thị trường luôn. Thành phần hay bị lén thêm nữa là phenolphthalein, vốn là thuốc nhuận tràng, ăn lâu dài nhiều sẽ làm rối loạn điện chất trong cơ thể và gây rối loạn nhịp tim. Cách phân biệt: thuốc giảm cân đàng hoàng là thuốc kê đơn, số phép tra được ở nhà thuốc và bệnh viện. Sản phẩm bán dưới danh「cà phê」「kẹo」「mận」「thay bữa」mà hứa tuần giảm vài cân thì theo định nghĩa không được có tác dụng dược, mà có tác dụng tức là đã trộn thêm thứ gì đó. Nguyên văn 国食药监办〔2010〕432 号 nằm trên trang của Cục Quản lý dược phẩm quốc gia, máy này mở luôn báo lỗi 412 nên phải trích lại trang của cơ quan quản lý cấp tỉnh. Hai câu「rủi ro lớn hơn lợi ích」và「tháng 10 năm 2010 ngừng sản xuất, kinh doanh, sử dụng」có thể đối chiếu từng chữ ở đó. +### 5. Đừng dùng steroid đồng hóa (「thuốc tiêm tăng cơ」「thuốc uống」) để tăng cơ + +- Chi phí: Không tốn tiền, lại còn tiết kiệm tiền thuốc. Khó ở chỗ phải chấp nhận cơ bắp mọc chậm. +- Nói thẳng: Nam giới có kết quả steroid dương tính trong kiểm tra doping ở phòng gym thì nguy cơ chết gấp 3 lần người cùng tuổi, và số lần đến bệnh viện mỗi năm nhiều hơn gấp đôi. Viêm da, vú to ở nam và rối loạn cương, cứ mười người dùng thuốc thì ít nhất một người mắc. Loại thuốc này ở Trung Quốc là thuốc kê đơn, nên kênh nào bán được cho bạn thì bản thân nó đã vi phạm pháp. +- Lợi ích: Đan Mạch quay lại tra hồ sơ, cho mỗi người dùng thuốc một đối chứng cùng tuổi. Người dùng là 545 nam giới có kết quả steroid androgenic dương tính trong kiểm tra doping phòng gym từ 2006-01-03 đến 2018-03-01, ghép với 5.450 đối chứng. Lại dùng thêm 644 người từ chối kiểm tra và 6.440 đối chứng để soát lại một lần nữa. Kết quả: tỷ lệ tử vong của người dùng gấp 3 lần đối chứng, HR 3.0 (95% CI 1.3–7.0), mức gấp thật nằm trong khoảng từ 1.3 đến 7. Số lần nhập viện hoặc khám mỗi năm trung vị là 0.81 vs 0.36 (P<0.0001). Trung vị nghĩa là một nửa người nhiều hơn con số này, một nửa ít hơn. Giá trị P này cho thấy chênh lệch gần như không thể là ngẫu nhiên. Viêm da (mụn), vú to ở nam và rối loạn cương đều ảnh hưởng hơn 10% người dùng, tức cứ mười người dùng thì ít nhất một người mắc, tất cả đều cao rõ rệt so với đối chứng (P<0.0001). Lần soát hồ sơ thứ hai cho kết quả nhất quán.《Biện pháp chống doping》Điều bảy: Nhà nước quản lý chặt các chất cấm có trong danh mục doping, bất kỳ tổ chức, cá nhân nào cũng không được sản xuất, kinh doanh, nhập khẩu, xuất khẩu bất hợp pháp. Cùng biện pháp đó quy định chế phẩm đồng hóa protein và hormone peptide phải quản lý như thuốc kê đơn, tức chỉ mua được bằng đơn thuốc của bác sĩ;「蛋白同化制剂」là cách luật pháp gọi nhóm thuốc này +- Mức bằng chứng: A +- Nguồn: Horwitz H và cộng sự (2019). Health consequences of androgenic anabolic steroid use. J Intern Med. ；Quốc vụ viện (国务院) (2004). Biện pháp chống doping（反兴奋剂条例）(Số 398 của Quốc vụ viện, điều bảy và điều chín). +- Ghi chú:Tranh cãi. Đây là nghiên cứu theo dõi một nhóm người rồi xem kết quả, không phải thử nghiệm chia nhóm ngẫu nhiên. Người dùng thuốc vốn có thể cũng mạo hiểm hơn ở mặt khác, con số 3 lần không thể đổ hết cho thuốc. Mức gấp thật có thể thấp tới 1.3, cũng có thể cao tới 7.0, biên độ rất rộng. Nhưng「viêm da, vú to ở nam và rối loạn cương mỗi thứ đều có trên một phần mười người dùng mắc」là tác dụng phụ nhìn thấy được bằng mắt, ít bị ảnh hưởng bởi kiểu thiên lệch này. Trong luật, nhóm thuốc này được viết là「蛋白同化制剂」, thấy bốn chữ đó là nói đến nó. Bột protein thường và creatine không nằm trong mục này. Rắc rối là sản phẩm「tuyên bố tăng cơ rất nhanh」có thể lén thêm thành phần. Cách phân biệt giống hệt mục 4 (đừng mua thuốc giảm cân hứa giảm nhanh): thuốc đàng hoàng là thuốc kê đơn, số phép tra được. Bán dưới danh「bột tăng cơ」「thực phẩm bổ sung」mà hứa mấy tuần có kết quả, nếu thật sự hữu hiệu thì là đã trộn thêm gì đó. + +### 6. Đã định dùng thuốc giảm cân thì đến bệnh viện lấy đơn thuốc, đừng mua ở cửa hàng trực tuyến mà không cần đơn + +- Chi phí: Một lần khám, cộng đi lại hai chiều. Mấy chục tệ, nửa ngày. +- Nói thẳng: Thuốc kê đơn bắt buộc phải bán bằng đơn thuốc, bán trên mạng cũng vậy: phải có đơn thật và định danh thật. Cửa hàng không hỏi khám, không cần đơn, đặt hàng là gửi ngay thì hoặc là bán không phải thứ nó nói, hoặc là cửa hàng đó vi phạm quy định. Khi xảy ra sự cố, không ai chịu trách nhiệm về liều lượng và chống chỉ định của bạn. Một lần khám mất mấy chục tệ, là có bác sĩ帮你 giữ cửa ải này. +- Lợi ích:《Biện pháp quản lý giám sát chất lượng kinh doanh và sử dụng thuốc》（药品经营和使用质量监督管理办法）yêu cầu doanh nghiệp bán lẻ thuốc phải tuân thủ chế độ quản lý phân loại thuốc kê đơn và thuốc không kê đơn. Chế độ đó chia thuốc thành hai loại: một loại bắt buộc mua bằng đơn của bác sĩ, loại kia mua được tự mình. Nhà thuốc phải bán thuốc kê đơn theo đơn thuốc theo quy định, giữ đơn không dưới năm năm. Khi dược sĩ hoặc nhân viên kỹ thuật dược khác không có mặt thì phải treo bảng thông báo, chưa qua họ kiểm tra thì không được bán thuốc kê đơn. Thuốc kê đơn không được bày bán mở, tức là đặt trên kệ để khách tự lấy. Điều chín của《Biện pháp giám sát quản lý kinh doanh thuốc trên mạng》（药品网络销售监督管理办法）quy định bán thuốc kê đơn cho cá nhân qua mạng thì phải bảo đảm nguồn đơn thật, đáng tin cậy và thực hiện định danh thật. Điều mười yêu cầu hiển thị nổi bật thông tin cảnh báo rủi ro dưới mỗi trang hiển thị thuốc. Đơn chưa qua kiểm tra thì không được hiển thị hướng dẫn sử dụng, cũng không được cung cấp dịch vụ mua, tức là trang không nên có cả nút mua. Điều tám quy định thuốc Nhà nước quản lý đặc biệt không được kinh doanh qua mạng. Nhóm thuốc này bao gồm vaccine, chế phẩm máu, thuốc gây nghiện, thuốc tâm thần, thuốc độc dùng trong y tế, thuốc phóng xạ và hóa chất tiền chất thuộc nhóm thuốc +- Mức bằng chứng: B +- Nguồn: Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2023). Biện pháp quản lý giám sát chất lượng kinh doanh và sử dụng thuốc（药品经营和使用质量监督管理办法）(Số 84 của Cục). ；Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2022). Biện pháp giám sát quản lý kinh doanh thuốc trên mạng（药品网络销售监督管理办法）(Số 58 của Cục, từ điều tám đến điều mười). +- Ghi chú: Pháp lệnh có thể đối chiếu từng chữ, nhưng「uống theo đơn thì so với tự mua tự uống sẽ bớt được bao nhiêu sự cố」chưa ai nghiên cứu định lượng, nên ghi mức B vì「khó định lượng」. Steroid và hormone giới tính nói ở mục 5 và 7 của phần này cũng đi theo cùng bộ quy tắc đó. Đến lúc thật sự phải nhờ thuốc để giảm cân, bác sĩ sẽ xem chỉ số khối cơ thể, người có bệnh gì khác không, có tình trạng nào không dùng được thuốc không, rồi mới quyết định uống hay không, uống loại nào. Cái phán đoán đó không mua được. +### 7. Chỉ dùng thuốc hormone giới tính khi bác sĩ kê đơn và có tái khám định kỳ; đừng mua trên mạng, đừng tự tăng liều + +- Chi phí: Khám định kỳ, cộng lấy máu xét nghiệm lại. Một năm mấy trăm tệ, mất mấy nửa ngày. +- Nói thẳng: Mục này không bàn có nên dùng hay không, chỉ bàn dùng thế nào. Người chuyển giới nữ dùng estrogen thì huyết khối trong mạch rõ ràng nhiều hơn người cùng tuổi: mỗi 1000 người trong hai năm thêm khoảng 4 ca, trong tám năm thêm khoảng 17 ca. Loại rủi ro này phải trông vào tái khám định kỳ; mua thuốc trên mạng, tự chỉnh liều thì không ai kiểm tra giúp bạn. +- Lợi ích: Các nhà nghiên cứu theo dõi một nhóm người lấy từ hồ sơ bệnh án điện tử của hệ thống y tế Kaiser Permanente: 2.842 người chuyển giới nữ và 2.118 người chuyển giới nam. Thời gian theo dõi trung bình lần lượt là 4.0 năm và 3.6 năm. Mỗi người được ghép với khoảng 10 lần số đối chứng đồng tính theo năm sinh, chủng tộc, điểm khám và thời điểm vào nghiên cứu, tổng cộng 48.686 nam đồng tính và 48.775 nữ đồng tính. Thuyên tắc tắc mạch tĩnh mạch (huyết khối hình thành trong mạch và chặn nó lại) ở người chuyển giới nữ nhiều hơn đối chứng. So với nam đồng tính, ở 2 năm là 4.1 ca trên 1000 người (95% CI 1.6–6.7). Ở 8 năm là 16.7 ca (6.4–27.5). Tính ra, cứ 1000 người thì hai năm thêm 4.1 ca, tám năm thêm 16.7 ca. So với nữ đồng tính, hai con số là 3.4 (1.1–5.6) và 13.7 (4.1–22.7). Đột quỵ thiếu máu (loại đột quỵ do mạch bị tắc) và nhồi máu cơ tim nhìn chung gần nhau giữa các nhóm. Nhưng ở người chuyển giới nữ chỉ bắt đầu điều trị hormone trong thời gian theo dõi, chênh lệch về thuyên tắc tĩnh mạch và đột quỵ thiếu máu rõ rệt hơn. Bằng chứng về người chuyển giới nam chưa đủ để kết luận. Tác giả kết luận「kết quả này có thể cho thấy cần theo dõi lâu dài tác dụng phụ của estrogen lên mạch máu trong điều trị hormone cho người chuyển giới」 +- Mức bằng chứng: A +- Nguồn: Getahun D và cộng sự (2018). Cross-sex Hormones and Acute Cardiovascular Events in Transgender Persons: A Cohort Study. Ann Intern Med. ；Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2022). Biện pháp giám sát quản lý kinh doanh thuốc trên mạng（药品网络销售监督管理办法）(Số 58 của Cục, điều chín). +- Ghi chú: Các nhà nghiên cứu tự nêu chỗ yếu của nghiên cứu này: có bao nhiêu người tự mua thuốc ngoài hệ thống bệnh viện thì họ không tra được, nên đường ranh「có dùng thuốc hay không」vốn không cắt được sạch. Vì thế càng nên đưa việc dùng thuốc vào tái khám định kỳ chính quy. Nghiên cứu này chỉ xem mạch máu có xảy ra sự cố hay không; nó không trả lời có nên dùng điều trị hormone hay dùng nó mua được gì, đó là bộ bằng chứng khác. Bất kỳ chỗ nào dùng hormone giới tính cũng thế, gồm thuốc tránh sĩ, điều trị hormone giai đoạn mãn kinh, và thay thế testosterone. Ai mua trên mạng, chỉnh liều theo bài viết trên mạng, và không bao giờ kiểm tra chỉ số đông máu, gan, thận thì rủi ro nằm ở đoạn không ai nhìn, không liên quan tới việc vì sao dùng thuốc. Đường đúng ở Trung Quốc là khoa nội tiết, hoặc khoa chuyên ngành tương ứng của cơ sở y tế chính quy: đánh giá, kê đơn, tái khám. +### 8. Nếu「cảm thấy mình xấu」đã đến mức soi gương lặp đi lặp lại, muốn sửa đi sửa lại, thì làm một đánh giá về hình thể trước khi nói đến phẫu thuật + +- Chi phí: Một lần khám ở khoa tâm thần hoặc khoa tâm lý lâm sàng, mấy chục tệ. Khó ở chỗ phải đi được chuyến đó trước khi đặt lịch phẫu thuật. +- Nói thẳng: Trong số người đi phẫu thuật tạo hình, khoảng 15% mắc rối loạn hình thể, tức phóng đại một khuyết điểm mà người khác hầu như không nhìn thấy đến mức ảnh hưởng cuộc sống. Ở người đến khoa da liễu để làm đẹp khoảng 13%, cả hai nhóm đều hơn bảy phần mười là nữ. Người như vậy xong một chỗ lại nhắm sang chỗ tiếp theo, phẫu thuật không giải quyết được. Hãy khám một lần trước để xem có phải trường hợp của mình không, rồi mới tính đến mổ. +- Lợi ích: Theo quy tắc, gom đủ 33 nghiên cứu rồi tổng hợp tính. Ở người khám phẫu thuật tạo hình, tỷ lệ mắc rối loạn hình thể (phóng đại khuyết điểm mà người khác hầu như không nhìn thấy đến mức ảnh hưởng cuộc sống) là 15.04%, các nghiên cứu lẻ nằm trong khoảng 2.21%–56.67%. Quy ra số người, cứ 100 người đi phẫu thuật thì khoảng 15 người符合, còn riêng từng nghiên cứu thì từ 2 đến 57 người. Nhóm này tuổi trung bình 34.54±12.41 tuổi, con số sau là biên độ dao động, 74.38% là nữ. Người khám da liễu là 12.65%, các nghiên cứu lẻ trong khoảng 4.52%–35.16%, tuổi trung bình 27.79±9.03 tuổi, 76.09% là nữ. Tác giả đề nghị bác sĩ phẫu thuật tạo hình và bác sĩ da liễu chủ động đánh giá, nhận ra người có nguy cơ cao, và sắp xếp vài khoa cùng theo dõi +- Mức bằng chứng: B +- Nguồn: Ribeiro RVE (2017). Prevalence of Body Dysmorphic Disorder in Plastic Surgery and Dermatology Patients: A Systematic Review with Meta-Analysis. Aesthetic Plast Surg. +- Ghi chú: Tỷ lệ các nghiên cứu đo được chạy từ 2.21% đến 56.67%, chênh lệch rất lớn. Lý do là bảng hỏi và nhóm khảo sát đều khác nhau. Nên chỉ đọc được「tỷ lệ này không thấp, đáng để sàng lọc một lần trước」, không coi 15.04% là con số chính xác, ghi mức B cũng vì thế. Hầu hết người muốn làm thẩm mỹ không có vấn đề tâm lý. Mục này nhắm vào nhóm nhỏ「lặp lại, kéo dài, đã ảnh hưởng đến công việc và giao tiếp」, phẫu thuật không giải quyết được vấn đề của họ. Dấu hiệu để phán đoán có mấy cái: mất một phần lớn buổi ngày để kiểm tra hoặc che một chỗ, vì thế mà tránh giao tiếp hoặc chụp ảnh, xong một chỗ lại nhắm sang chỗ tiếp theo. diff --git a/web/package.json b/web/package.json index 4dbf0b28..b14f5083 100644 --- a/web/package.json +++ b/web/package.json @@ -9,7 +9,8 @@ "build": "astro build && node scripts/copy-assets.mjs", "preview": "astro preview", "check": "tsx src/lib/parity.ts", - "verify": "npm run check && npm run build && tsx src/lib/render-check.ts" + "verify": "npm run check && npm run build && npm run check-lang && tsx src/lib/render-check.ts", + "check-lang": "tsx src/lib/lang-switcher-check.ts" }, "dependencies": { "astro": "7.3.4" diff --git a/web/src/lib/astro-page.ts b/web/src/lib/astro-page.ts index f5e9fd09..4e7ae29b 100644 --- a/web/src/lib/astro-page.ts +++ b/web/src/lib/astro-page.ts @@ -2,14 +2,14 @@ // // 分工（别串了）： // · 壳（SEO 头、hreflang、JSON-LD、界面文案、site-config、跳语言页、sitemap、robots） -// 全部由 tools/site/build.mjs 出，Astro 只是调 renderPage/chooserPage/sitemapXml/robotsTxt。 +// 全部由 tools/site/build.mjs 出，Astro 只是调 renderPage/chooserJump/sitemapXml/robotsTxt。 // 域名、文案、统计数字改那边，这里自动跟上。 // · 预渲染（卡片、侧栏目录、术语表）和语料内联由 web/ 出（render.ts + 下面这个文件）。 // 爬虫看到的全文从这里来。 // // 注入全走字符串锚点，和 tools/offline/build.mjs 一个路数：锚点找不到就抛错， // 不要静默发出半成品——少一段预渲染，页面看起来正常但爬虫少看几百条。 -import { renderPage, chooserPage, sitemapXml, robotsTxt, SITE } from '../../../tools/site/build.mjs'; +import { renderPage, chooserJump, sitemapXml, robotsTxt } from '../../../tools/site/build.mjs'; import { locale, readRepo } from './site'; import { loadLocale } from './load'; import { buildGloss } from './parse'; @@ -92,9 +92,9 @@ export function localePageParts(code) { return { ...splitDoc(html), ...rest }; } -/** 跳语言小页三段。 */ -export function chooserPageParts() { - return splitDoc(chooserPage(SITE)); +/** ?lang=xx 老链接的跳转脚本，整段塞进根页面的 head。没带 lang 就不跳。 */ +export function chooserJumpParts() { + return chooserJump(); } -export { chooserPage, sitemapXml, robotsTxt }; +export { chooserJump, sitemapXml, robotsTxt }; diff --git a/web/src/lib/lang-switcher-check.ts b/web/src/lib/lang-switcher-check.ts new file mode 100644 index 00000000..90842260 --- /dev/null +++ b/web/src/lib/lang-switcher-check.ts @@ -0,0 +1,98 @@ +// 语言下拉的落点检查。跑在构建产物（web/dist）上： +// +// npm run check-lang +// +// 查一件事：每种语言的页面上，换语言选项指向的地址是不是「同一个目录换成那种语言」， +// 而且当前的筛选参数（?q=…）和定位锚点（#e-13-1）都跟着过去。要求 4 就是这一条。 +// +// 为什么单独抠这一段跑：页内那段 buildLangPicker 读的是 location，浏览器才有。 +// 这里给它一份假的 location，把它从构建产物里原样抠出来执行一遍—— +// 抠的是发出去的那份代码本身，不是照着抄的一份，抄错了就测不出来。 +// +// 曾经的错：选项地址写的是 (here === SITE.dir ? here : SITE.dir)，SITE.dir 是 +// 「这一页所在的目录」，跟「要去的那种语言」无关。结果从 /zh/ 选英文，跳到 /zh/， +// 还是中文；换语言等于没换。这个检查就是为了不让它再溜过去。 +import { readFileSync, existsSync } from 'node:fs'; +import { site } from './site'; + +const DIST = new URL('../../dist/', import.meta.url); +// 站点根在 Pages 上带的路径前缀（/HowToLiveBetter/ 这种）。页面自己那份 +// site-config 里没有 canonical，根从 site.json 取——换域名只改那一处，这里自动跟上。 +const PREFIX = new URL(site.site + '/').pathname; + +const read = p => readFileSync(p, 'utf8'); +const configOf = file => + JSON.parse(/id="site-config" type="application\/json">([\s\S]*?)<\/script>/.exec(read(file))[1]); +// 构建产物里没有的页面（比如那种语言还没翻完）就不测，不是错 +const pages = [['index.html', ''], ['en/index.html', 'en/'], ['vi/index.html', 'vi/'], ['zh/index.html', 'zh/']] + .filter(([file]) => existsSync(new URL(file, DIST))); + +let bad = 0; +const fail = what => { bad++; console.log(` ✗ ${what}`); }; + +// 每种语言的正文页都该有这一段，且几份完全一致——不一致说明有一份没重新构建。 +// 别名页（/en/ 现在是转发页）不算：它就是一块跳转页，本来就没有语言下拉。 +const bodyPages = pages.filter(([file]) => file.endsWith('index.html') && file !== 'en/index.html'); +let body = null; +for (const [file] of bodyPages) { + const m = /function buildLangPicker\(\)\{[\s\S]*?\n\}/.exec(read(new URL(file, DIST))); + if (!m) { fail(`${file} 里没有 buildLangPicker，语言下拉不会建`); continue; } + if (body === null) body = m[0]; + else if (body !== m[0]) fail(`${file} 里的 buildLangPicker 和根上那份不是同一段（构建产物不是一起出的？）`); +} +if (body === null) { console.log('构建产物里没有 buildLangPicker'); process.exit(1); } + +// 假 DOM：够 buildLangPicker 用就行（它只用 createElement/getElementById/addEventListener） +const opts = []; +const sel = { appendChild: o => opts.push(o), addEventListener(){}, closest: () => ({}), selectedOptions: [] }; +const document = { getElementById: id => (id === 'lang' ? sel : null), createElement: () => ({ dataset: {} }) }; +const location = { pathname: '', search: '', hash: '' }; +const run = new Function('document', 'location', 'SITE', body + '; buildLangPicker();'); + +// 每页测两种落点：干净的，和带着筛选参数+锚点定位到某一条的 +const CASES = [ + ['', '', ''], + ['?q=travel', '', '#e-13-1'], +]; + +for (const [file] of pages) { + if (file === 'en/index.html') { // 别名转发页，没有语言下拉可测 + const a = read(new URL(file, DIST)); + if (!/http-equiv="refresh"/.test(a)) fail('en/index.html 不是转发页，/en/ 的老链接会 404'); + else console.log(' ✓ en/：别名转发页，带着 ?q= 和锚点回根上'); + continue; + } + const cfg = configOf(new URL(file, DIST)); + cfg.offline = false; + const prefix = PREFIX; + const here = `${prefix}${cfg.dir || ''}`; + for (const [search, , hash] of CASES) { + opts.length = 0; + location.pathname = here; + location.search = search; + location.hash = hash; + try { run(document, location, cfg); } + catch (e) { fail(`${cfg.code} 页面上跑 buildLangPicker 抛错：${e.message}`); continue; } + + for (const o of opts) { + if (!o.dataset.url) { // 当前语言，不给链接是对的 + if (o.value !== cfg.code) fail(`${cfg.code} 页面上「${o.value}」没有链接，它不是当前语言`); + continue; + } + const want = new URL(prefix + (cfg.langDirs?.[o.value] ?? o.value + '/'), 'https://x' + here).href; + const got = new URL(o.dataset.url, 'https://x' + here).href; + if (got !== want + search + hash) + fail(`${here} 选「${o.value}」：应该是 ${want + search + hash}，实际 ${got}`); + // 落点那条在这个语言里得存在：// 下要有页面（别名页也算） + const path = new URL(got).pathname; + const file = path.slice(prefix.length).replace(/^\/+/, ''); + const target = (file || 'index.html').endsWith('/') ? file + 'index.html' : file; + if (!existsSync(new URL(target, DIST)) && !existsSync(new URL(file, DIST))) + fail(`${here} 选「${o.value}」指向 ${path}，构建产物里没有那一页`); + } + } + console.log(` ✓ ${cfg.code}（${here || '/'}）：${cfg.locales.length} 种语言，换语言换目录，?q= 和锚点都带着`); +} + +if (bad) { console.log(`\n语言下拉有 ${bad} 处不对`); process.exit(1); } +console.log('\n语言下拉落点都对：换语言换目录，筛选参数和定位锚点都跟着走。'); diff --git a/web/src/pages/en.astro b/web/src/pages/en.astro index e0c91909..cd1a78da 100644 --- a/web/src/pages/en.astro +++ b/web/src/pages/en.astro @@ -1,8 +1,37 @@ --- -// 英文主页（默认语言）。整页由 tools/site/build.mjs 的壳 + web 的预渲染拼出来， -// 见 src/lib/astro-page.ts 的分工说明。这里只写骨架，内容原样塞进去。 -import { localePageParts } from '../lib/astro-page'; -const { lang, head, body } = localePageParts('en'); +// /en/ 现在是英文的别名地址，正文在根上（locales.json 里 en 的 dir 是空串）。 +// README 和别处早就散着 /en/ 的链接，不给这一页就是一批 404。 +// meta refresh 带着查询串和锚点一起去根上，关掉 JS 也照样跳； +// canonical 指根上那一页：别名不是另一本书，搜索引擎按 canonical 归并。 +import { SITE } from '../../../tools/site/build.mjs'; +const root = SITE + '/'; --- - + + + + +HowToLiveBetter + + + + + + +
    +

    The English edition now lives at the site root.

    +Continue to the English edition +
    + diff --git a/web/src/pages/index.astro b/web/src/pages/index.astro index 4ae417e5..e316784e 100644 --- a/web/src/pages/index.astro +++ b/web/src/pages/index.astro @@ -1,8 +1,11 @@ --- -// 根目录跳语言的小页：和原来 tools/site/build.mjs 生成的 index.html 同一份内容， -// 只是换成 Astro 出。noindex，爬虫不收它，读者按浏览器语言分流。 -import { chooserPageParts } from '../lib/astro-page'; -const { lang, head, body } = chooserPageParts(); +// 根 URL 发的是英文（默认语言住���根上，locales.json 里 en 的 dir 是空串）。 +// 语言选择在这一页的下拉里：换语言保留当前的筛选参数和 #e-xx-y 锚点。 +// 问「/?lang=vi」这种老链接的（README 和别处都这么写过）走下面的跳转脚本， +// 命中不了就回英文——英文是默认那一种。 +import { localePageParts, chooserJumpParts } from '../lib/astro-page'; +const { lang, head, body } = localePageParts('en'); +const jump = chooserJumpParts(); --- - + From fdb2f172e7b2b6d5833387a3212341975a6533e9 Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 12:24:44 +0700 Subject: [PATCH 13/25] =?UTF-8?q?=E8=8B=B1=E6=96=87=E6=9C=80=E5=90=8E?= =?UTF-8?q?=E4=B8=A4=E8=8A=82=EF=BC=8833=20=E6=AE=8B=E7=96=BE=E4=B9=8B?= =?UTF-8?q?=E5=90=8E=E6=80=8E=E4=B9=88=E6=B4=BB=E3=80=8126=20=E5=81=9A?= =?UTF-8?q?=E4=B8=80=E4=B8=AA=E7=BD=91=E7=AB=99=E6=88=96=E5=B9=B3=E5=8F=B0?= =?UTF-8?q?=EF=BC=89=E7=BF=BB=E5=AE=8C=E5=B9=B6=E5=90=88=E5=85=A5=EF=BC=9A?= =?UTF-8?q?EN=2034/34=20=E5=85=A8=E4=B9=A6?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- ...31\346\210\226\345\271\263\345\217\260.md" | 112 ++++++++++ ...16\346\200\216\344\271\210\346\264\273.md" | 206 ++++++++++++++++++ ...22\344\270\215\345\210\222\347\256\227.md" | 65 ++++++ 3 files changed, 383 insertions(+) create mode 100644 "en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" create mode 100644 "en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" create mode 100644 "vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" diff --git "a/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" "b/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" new file mode 100644 index 00000000..bb162362 --- /dev/null +++ "b/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" @@ -0,0 +1,112 @@ +[← 回总目录](../README.md) + +# 26. 做一个网站或平台：资质、备案和服务器 + +你想做一个网站或者 App。本节回答四件事：它在法律上算哪种生意，要办哪张证，服务器放在哪里，用户在上面发的东西和收的钱谁来负责。口径是人身自由与金钱。条目的顺序是：先写会坐牢的，再写要先办证的，然后是平台每天都要做的事，最后一条是选服务器的经验，那条只能给判断、给不出数字。你是受雇的技术人员，红线见第 11 节。公司怎么注册、怎么报税，见第 12 节，本节不重复。各类证的对照表和选服务器的决策表，见 [docs/做平台要办哪些证.md](../docs/做平台要办哪些证.md)。 + +[← Back to the contents](../README.md) + +# 26. Running a website or platform: licences, filings, servers + +You want to build a website or an app. This section answers four questions: what kind of business it is in law, which licence you need, where the servers sit, and who is answerable for what users post and for the money taken in. The measures are personal freedom and money. The items come in this order: first the ones that mean prison, then the ones that need a licence before you start, then what the platform has to do every day, and last the experience behind choosing a server, which can give you a judgement but no figures. If you are a technical employee working for somebody else, the red lines are in section 11. How a company is registered and how the tax is filed are in section 12, and they are not repeated here. The table comparing the licences and the decision table for choosing a server are in [docs/做平台要办哪些证.md](../docs/做平台要办哪些证.md). + +### 1. The platform should not take the buyer's money and pass it on to the seller; let the money go straight through a licensed payment institution + +- Cost: No money. It takes some time to connect the product of a licensed payment institution. A licensed payment institution is a company that has a payment licence from the central bank. With its split-settlement or guaranteed-transaction function, the buyer's money goes straight to the seller and never touches the platform's own account +- In plain language: If the buyer's money comes into your account first and you pass it on to the seller, you are doing payment business. That business needs the approval of the People's Bank of China, and it needs 100 million yuan of paid-in capital to be shown for real. Do it without the approval and you are shut down, the money you took is confiscated and you are fined on top of that, with the boss and the person handling it fined separately. In the serious cases there is a prison sentence as well. There is only one safe way: the money must not come into your account. +- Benefit: The payment business the regulations describe is 「根据收款人或者付款人提交的电子支付指令转移货币资金」 — transferring money according to the electronic payment instruction submitted by the payee or the payer. In plain terms, it means moving money for somebody else on somebody else's instruction. To do that business a company 「应当经中国人民银行批准，取得支付业务许可」 — must be approved by the People's Bank of China and hold a payment business licence. The registered capital is at least 100 million yuan and it must be paid in cash: that 100 million yuan has to go into the company's account in real money. 「未经依法批准，擅自设立非银行支付机构、从事或者变相从事支付业务的，由中国人民银行依法予以取缔，没收违法所得」 — anyone who, without approval as required by law, sets up a non-bank payment institution, or engages in or takes on payment business without it, is shut down by the People's Bank of China and has the illegal gains confiscated. Anyone who does this without the approval is shut down at once and the money they took is taken away. Where the illegal gains are 500,000 yuan or more, there is a further fine of 1 to 5 times the illegal gains. Where there are no illegal gains, or fewer than 500,000 yuan, the fine is 500,000 to 2 million yuan. The legal representative and the person directly responsible are fined 100,000 to 500,000 yuan each. The Criminal Law also counts 「非法从事资金支付结算业务」 — carrying out fund payment and settlement business without a licence — as operating without a licence. Where the circumstances are serious it carries 5 years or less, and in the especially serious cases 5 years or more. What counts as serious is drawn in the interpretation of the Supreme People's Court (最高人民法院) and the Supreme People's Procuratorate (最高人民检察院). The line is 5 million yuan or more in the value of the illegal business, or 100,000 yuan or more in illegal gains (nationwide, from May 2024) +- Evidence grade: A +- Sources: State Council (国务院) (2023). Regulations on the Supervision and Administration of Non-Bank Payment Institutions, 非银行支付机构监督管理条例（国务院令第 768 号，第二、六、八、二十七、四十七条）. ; People's Bank of China (中国人民银行) (2024). Detailed Rules for the Implementation of the Regulations on the Supervision and Administration of Non-Bank Payment Institutions, 非银行支付机构监督管理条例实施细则（中国人民银行令〔2024〕第 4 号，第六十八条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百二十五条第三项）. ; Supreme People's Court (最高人民法院) and Supreme People's Procuratorate (最高人民检察院) (2019). Interpretation on Several Issues Concerning the Application of Law in Handling Criminal Cases of Illegal Fund Payment and Settlement Services and Illegal Foreign Exchange Trading, 关于办理非法从事资金支付结算业务、非法买卖外汇刑事案件适用法律若干问题的解释（第一、三、四条）. +- Notes: Designing the money so that it never comes into your account is the only safe way. The trade calls this kind of collection-and-payment-on-behalf 「secondary clearing」, but the word is not in any official document. The detailed rules use the wording 「变相从事支付业务」 — taking on payment business in a disguised way, that is, moving money for others on the user's instruction without the approval. The interpretation of the Supreme People's Court (最高人民法院) and the Supreme People's Procuratorate (最高人民检察院) lists only three situations: cashing out through fabricated transactions, cashing out through a company account, and cashing out through cheques. It does not name a platform collecting and paying on behalf of others, so one cannot say that a platform doing that 「certainly amounts to a crime」 + +### 2. A website that charges money needs a licence, matching transactions is a different licence, and a free site only needs a filing + +- Cost: The filing costs nothing. The Value-Added Telecommunications Business Licence (增值电信业务经营许可证) costs money, and it decides 「向用户收钱的网站能不能开」 — whether a website that charges its users is allowed to open. To get that licence you need a company first. A business operating only inside one province needs registered capital of at least 1 million yuan. A business operating across provinces needs at least 10 million yuan. The review takes 60 days, and the licence is valid for 5 years +- In plain language: A website that charges its users has to get the Value-Added Telecommunications Business Licence first. That licence is only issued to companies; an individual cannot get it. A website that does not charge only has to file, and the filing is free. Open a site and charge money without the licence and the income is confiscated and you are fined, and in the serious cases the site is closed. Whether your platform falls into the class that matches buyers with sellers is not decided for you in this item, and the reason is in the notes. +- Benefit: The regulations say 「国家对经营性互联网信息服务实行许可制度；对非经营性互联网信息服务实行备案制度。未取得许可或者未履行备案手续的，不得从事互联网信息服务」 — the state runs a licensing system for commercial internet information services and a filing system for non-commercial ones, and without the licence or without the filing you may not carry on internet information services at all. A site that charges has to get the licence first; a site that does not charge has to file first; with neither, you may not open. What counts as commercial? The regulations call it 「通过互联网向上网用户有偿提供信息或者网页制作等服务活动」 — providing information or web page production to internet users for money. In simple terms, it means charging the users. The class that matches transactions is online data processing and transaction processing, abbreviated in the trade as EDI. The regulations define it as 「利用各种与公用通信网或互联网相连的数据与交易/事务处理应用平台，通过公用通信网或互联网为用户提供在线数据处理和交易/事务处理的业务」. It also 「包括交易处理业务、电子数据交换业务和网络/电子设备数据处理业务」. In other words, a website that processes data online for users, or that matches and processes transactions between buyers and sellers. Operating without the licence, or operating beyond the scope of the licence, means the illegal gains are confiscated and a fine of 3 to 5 times the illegal gains is imposed. Operating beyond the scope means holding the licence and doing things that go beyond what it says. Where there are no illegal gains, or fewer than 50,000 yuan, the fine is 100,000 to 1 million yuan. Where the circumstances are serious, the website is ordered closed (nationwide) +- Evidence grade: A +- Sources: State Council (国务院) (2000). Measures for the Administration of Internet Information Services, 互联网信息服务管理办法（国务院令第 292 号，2024 年 12 月 6 日第二次修订，第三、四、七、十九条）. ; State Council (国务院) (2024). Decision on Amending and Repealing Certain Administrative Regulations, 关于修改和废止部分行政法规的决定（国务院令第 797 号，第十八项）. ; State Council (国务院) (2000). Regulations on Telecommunications, 电信条例（2016 年第二次修订，第七、九、十三、六十九条）. ; Ministry of Industry and Information Technology (工业和信息化部) (2017). Measures for the Administration of Licences for the Conduct of Telecommunications Business, 电信业务经营许可管理办法（工信部令第 42 号，第六、九、十一、十四条）. +- Notes: Whether what you are doing is in the EDI class is for you to work out by matching the definition in the Catalogue of Telecommunications Business (《电信业务分类目录》). No official document says in so many words that an e-commerce platform with matching and fund handling must hold an EDI licence. The service guide of the Ministry of Industry and Information Technology (工业和信息化部) only says 「按照业务界定申请相应的电信业务经营许可」 — apply for the telecommunications business licence that matches how the business is defined. So whether you need one has to be judged from the definition yourself. One more thing: an individual cannot get this licence; the applicant has to be a company established in accordance with the law + +### 3. Streaming needs the Internet Culture Business Licence, and the licence for audio-visual programmes is basically out of reach for a private company + +- Cost: The Internet Culture Business Licence (网络文化经营许可证), called the web culture licence in short, decides 「能不能靠直播、游戏解说这类网络表演赚钱」 — whether you may earn money from online performances such as livestreaming or game commentary. It is approved by the provincial cultural administrative department, which must give a result within 20 days of accepting the application, and it is valid for 3 years, with the renewal filed 30 days before it expires. The Audio-Visual Programme Licence (视听节目许可) decides 「能不能在网上放节目」 — whether you may put programmes online, and it requires the applicant to be a wholly state-owned or a state-controlled entity +- In plain language: Livestreaming and game commentary need the Internet Culture Business Licence from the provincial cultural department. Online performance has to be written into the scope of business on the licence. A result comes within 20 days of the application, and the licence lasts 3 years. You have to keep the videos you have streamed for at least 60 days yourself. Open up without the licence and you are told to stop and warned, then fined up to 30,000 yuan, and if you do not stop you go on a blacklist. +- Benefit: A unit doing online performance has to get the licence first. The regulations say 「应当根据《互联网文化管理暂行规定》，向省级文化行政部门申请取得《网络文化经营许可证》，许可证的经营范围应当明确包括网络表演」. Playing games in front of an audience and commenting on them are managed on the same footing, and the regulations say 「参照本办法进行管理」. Video material of all online performances 「保存时间不得少于60日」 — must be kept for no less than 60 days, so you keep the videos you have streamed yourself for the full 60 days. Doing commercial internet culture business without approval gets an order to stop and a warning, then a fine of up to 30,000 yuan. Anyone told to stop and who goes on is put on the cultural market blacklist. Putting programmes online is a different licence: 「应当取得《信息网络传播视听节目许可证》」. The applicant 「具备法人资格，为国有独资或国有控股单位」 — must have the status of a legal person and be a wholly state-owned or a state-controlled entity, so it has to be a unit wholly owned or controlled by the state, and a private company basically cannot get it. Anyone who does it without the licence is warned and ordered to put it right, and may be fined up to 30,000 yuan. Where the circumstances are serious it is punished under the Regulations on the Administration of Broadcasting and Television (广播电视管理条例) (nationwide) +- Evidence grade: A +- Sources: Ministry of Culture (文化部) (2016). Measures for the Administration of the Business of Online Performance, 网络表演经营活动管理办法（文市发〔2016〕33 号，第二、四、六、十三、二十条）. ; Ministry of Culture (文化部) (2011). Interim Provisions on the Administration of Internet Culture, 互联网文化管理暂行规定（文化部令第 51 号，2017 年经文化部令第 57 号修订，第七、八、九、十六、二十一、二十八条）. ; National Radio and Television Administration (广电总局) and Ministry of Information Industry (信息产业部) (2007). Provisions on the Administration of Internet Audio-Visual Programme Services, 互联网视听节目服务管理规定（第 56 号令，2015 年修订，第二、七、八、二十四条）. ; Cyberspace Administration of China (国家网信办) and six other departments (2021). Guiding Opinions on Strengthening the Standardised Management of Online Livestreaming, 关于加强网络直播规范管理工作的指导意见（国信办发文〔2021〕3 号，第 9 项）. +- Notes: Which business goes with which licence is set out most clearly in the document of the seven departments. A platform that earns money from livestreaming needs the web culture licence plus the ICP filing, and the ICP filing is the registration the website does with the Ministry of Industry and Information Technology (工业和信息化部). Putting programmes online needs either the audio-visual programme licence or the completed registration. Publishing news needs the Internet News Information Service Licence. One more thing: the Interim Provisions on the Administration of Internet Culture (互联网文化管理暂行规定) reprinted on the government website in 2025 are the unrevised 2011 text, so quote the version currently in force from the Ministry of Culture and Tourism (文化和旅游部) + +### 4. A server inside China needs a filing, and the access provider must itself hold the value-added telecommunications licence + +- Cost: The filing costs nothing. It is submitted together with the server at the provider, and it usually takes a few days to a few weeks. Ten minutes is enough to check once whether a provider is qualified +- In plain language: A server inside China has to be filed before the site opens. The filing is free and goes in with the server at the provider, and it generally takes a few days to a few weeks. Without the filing you are ordered to put it right within a set period and fined 10,000 yuan, and if you do not put it right the site is closed. Not putting the filing number on the home page is a further fine of 5,000 to 50,000 yuan. +- Benefit: The regulations say 「未经备案，不得在中华人民共和国境内从事非经营性互联网信息服务」 — without a filing you may not carry out non-commercial internet information services inside the People's Republic of China. A website that does not charge its users still has to file first if it sits inside China. A site with no filing is ordered to put it right within a set period and is fined 10,000 yuan. Anyone who refuses to put it right has the site closed. The provider who connects your server to the network is covered too: it 「不得为未经备案的组织或者个人从事非经营性互联网信息服务提供互联网接入服务」. A provider that takes the money for you is covered in the same way. Such a provider 「不得为未依法取得经营许可证或者履行非经营性互联网信息服务备案手续的单位或者个人提供接入或者代收费等服务」. The meaning is that a proper provider may not take on a site that has neither the licence nor the filing. Then there are the two services the regulations call internet data centre business, which is colocation and cloud hosting, and internet access service business, which is broadband access. Both need a licence in their own right, so the provider must hold a licence first. A website whose home page does not show the filing number is ordered to put it right and fined 5,000 to 50,000 yuan (nationwide) +- Evidence grade: A +- Sources: Ministry of Information Industry (信息产业部) (2005). Measures for the Filing of Non-Commercial Internet Information Services, 非经营性互联网信息服务备案管理办法（信产部令第 33 号，2024 年经工信部令第 68 号修订，第五、十三、十八、二十二条）. ; Ministry of Industry and Information Technology (工业和信息化部) (2017). Measures for the Administration of Licences for the Conduct of Telecommunications Business, 电信业务经营许可管理办法（工信部令第 42 号，第二十、二十四条）. ; Ministry of Industry and Information Technology (工业和信息化部) (2015). Catalogue of Telecommunications Business (2015 edition), 电信业务分类目录（2015 年版，B11、B14、B21、B25）. +- Notes: Check whether the provider holds a licence before you talk about the price. To check a licence, use the MIIT integrated telecommunications business market management information system, tsm.miit.gov.cn, and search by the full name of the company or by the licence number. To check a website filing, use beian.miit.gov.cn. A 「data centre」 without the matching licence leaves you not knowing even who to appeal to on the day it fails. Putting the machine outside China does get round the filing. But where the users are in China and the money is in China, not one of the obligations written in items 5 to 10 in this section goes away, and you add the cost of sending data out of the country. +### 5. Let users sell on the platform and you have to verify, register, report the information and keep it for three years + +- Cost: No money. You have to verify the sellers' qualifications in the back office and build a registration file. Report the tax information once a quarter and keep the transaction data for three years. The hard part is that this goes on all year; it is not something you do once +- In plain language: If you let other people sell on your platform, you must first verify the seller's identity, address, contact details and licences, then register the seller and open a file. Once the file exists, you go back and check again at least once every six months. Who the seller is goes to the market regulation authorities, and the tax position goes to the tax authorities once a quarter. Keep the goods and the transaction records for three years. Miss one item and you are first ordered to put it right, and if you do not, the fine is 20,000 to 100,000 yuan. +- Benefit: The law sets out a stretch of what the platform has to do. The words are 「应当要求申请进入平台销售商品或者提供服务的经营者提交其身份、地址、联系方式、行政许可等真实信息，进行核验、登记，建立登记档案，并定期核验更新」. Before a seller comes on the platform they hand over their identity, address, contact details and the licences they are supposed to hold. You check them, register them and open a file, and then after a while you have to go back and check again. How often is 「定期」 — regularly? The departmental rule is more exact: 「至少每六个月核验更新一次」 — at least once every six months. You also have to report the identity information of the sellers on the platform to the market regulation authorities. The tax authorities get a report too, in the words 「依照税收征收管理法律、行政法规的规定，向税务部门报送平台内经营者的身份信息和与纳税有关的信息」. So who the seller is goes to the market regulation authorities, and the seller's identity and tax position go to the tax authorities. Information on goods and services and information on transactions 「保存时间自交易完成之日起不少于三年」 — must be kept for no less than three years counting from the day the transaction is completed. If you fall short on any of the above, you are ordered to put it right within a set period. If you do not, the fine is 20,000 to 100,000 yuan. Where the circumstances are serious, you are ordered to suspend business for rectification and fined a further 100,000 to 500,000 yuan. The tax information has to be reported 「于季度终了的次月内」 — within the month following the end of the quarter. Missing that deadline carries the same fine (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). E-Commerce Law, 电子商务法（第十、二十七、二十八、三十一、八十条）. ; State Administration for Market Regulation (市场监管总局) (2021). Measures for the Supervision of Online Trading, 网络交易监督管理办法（市场监管总局令第 37 号，第八、二十四、二十五、三十一条）. ; State Council (国务院) (2025). Provisions on the Reporting of Tax-Related Information by Internet Platform Enterprises, 互联网平台企业涉税信息报送规定（国务院令第 810 号，第二、三、四、十条）. ; State Taxation Administration (国家税务总局) (2025). Announcement on Matters Concerning the Reporting of Tax-Related Information by Internet Platform Enterprises, 关于互联网平台企业报送涉税信息有关事项的公告（税务总局公告 2025 年第 15 号）. +- Notes: The seller has to register as a market entity as well. The law lists a few cases where you do not: 「个人销售自产农副产品、家庭手工业产品，个人利用自己的技能从事依法无须取得许可的便民劳务活动和零星小额交易活动」 — a person selling their own agricultural and sideline products or home handicrafts, and a person using their own skills in convenience service activities that need no licence under the law and in occasional small-value transactions. 「零星小额」 — occasional and small in value — has a fixed figure: Article 8 of the Measures for the Supervision of Online Trading sets it at an annual turnover of 100,000 yuan. On the platform side there are two more things to do: mark the registered sellers and the unregistered ones separately, and do not allow the ratings left by consumers to be deleted + +### 6. You are answerable for what users post: a review system, a reporting channel, and stop the transmission and report at once when you find something unlawful + +- Cost: Money is needed to pay people to do the reviewing, and it is the platform's largest ongoing expense. The reporting channel and a record of every action taken both have to be kept. The hard part is that somebody has to watch it every day; one day off and something can go wrong +- In plain language: When something goes wrong with what a user has posted, the platform is answerable. You have to build this set of rules: registration, review of posts, review of comments, a daily sweep, and what to do when something goes wrong. Put the reporting channel somewhere the user can see at once. When you find unlawful information, stop the transmission at once, keep the records and report it to the authority in charge. If you do nothing, the fine is 50,000 to 500,000 yuan, and more in the serious cases. +- Benefit: The regulations say the platform 「应当履行信息内容管理主体责任」 — must carry the main responsibility for the content, so what users post is the platform's own work to manage. What you have to set up is 「健全用户注册、账号管理、信息发布审核、跟帖评论审核、版面页面生态管理、实时巡查、应急处置和网络谣言、黑色产业链信息处置等制度」 — a system covering user registration, account management, the review of what is posted, the review of replies and comments, the management of the ecology of the pages and layouts, real-time patrolling, emergency handling, and the handling of online rumours and information about criminal supply chains. From registration, posting and comments to the daily sweep and the handling of an incident, every step needs a rule written down. You also 「应当在显著位置设置便捷的投诉举报入口」 — must put an easy complaint and reporting channel in a prominent position. When you find unlawful information you 「应当依法立即采取处置措施，保存有关记录，并向有关主管部门报告」. The Cybersecurity Law is more exact. The words are 「发现法律、行政法规禁止发布或者传输的信息的，应当立即停止传输该信息，采取消除等处置措施，防止信息扩散，保存有关记录，并向有关主管部门报告」. Anyone who fails to do this is ordered to put it right, is warned and is named in a public notice, and may be fined 50,000 to 500,000 yuan. Anyone who refuses to put it right, or where the circumstances are serious, is fined 500,000 to 2 million yuan, and may be ordered to suspend business for rectification, to close the website or the application, or to have the licence or the business licence revoked. The person directly responsible is fined 50,000 to 200,000 yuan (nationwide) +- Evidence grade: A +- Sources: Cyberspace Administration of China (国家互联网信息办公室) (2019). Provisions on the Governance of the Ecology of Network Information Content, 网络信息内容生态治理规定（网信办令第 5 号，第八、九、十、十六、三十七条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第四十九、六十九条）. ; State Council (国务院) (2000). Measures for the Administration of Internet Information Services, 互联网信息服务管理办法（第十五、十六、二十三条）. +- Notes: You do not have to think of the review rules yourself; copy them. The nine kinds of content that may not be posted are set out clearly in Article 15 of the Measures for the Administration of Internet Information Services. If you livestream there are three more things: you have to 「建立直播内容审核平台」 — set up a platform for reviewing the content of the stream, you have to be able to cut off a livestream that is running at any moment, which the article calls 「即时阻断」 — instant blocking, and you have to keep the content of the stream and the logs for 60 days. The video material of an online performance is likewise kept for no less than 60 days + +### 7. If you provide information posting or instant messaging, you must require real identity information from users + +- Cost: Money is needed to connect a real-name verification channel, and these channels charge per check. Design the product on the principle of a real name in the back office and a free choice in the front: store the real name in the back office, show a nickname on the page +- In plain language: If your site lets people post, or lets them chat, you have to ask for the user's real identity when they take out the service. A user who does not give it may not be allowed to use it. The way to do it is to store the real name in the back office and let the user pick the nickname shown on the page; the phone number is enough for the check. If you do not take real names, you are first ordered to put it right. If you do not, the fine is 50,000 to 500,000 yuan, and the site may be closed. +- Benefit: The law says 「网络运营者为用户办理网络接入、域名注册服务，办理固定电话、移动电话等入网手续，或者为用户提供信息发布、即时通讯等服务，在与用户签订协议或者确认提供服务时，应当要求用户提供真实身份信息。用户不提供真实身份信息的，网络运营者不得为其提供相关服务」. If you did not ask for the real name, or the user gave none and you let them use the service anyway, you are ordered to put it right. If you refuse to put it right, or where the circumstances are serious, the fine is 50,000 to 500,000 yuan, and you may be ordered to suspend the relevant business, to suspend business for rectification, to close the website or the application, or to have the licence or the business licence revoked. The person directly responsible is fined 10,000 to 100,000 yuan. Livestreaming services follow the principle of a real name in the back office and a free choice in the front, with 「基于移动电话号码等方式的真实身份信息认证」 — verification of real identity information by means such as a mobile telephone number. The meaning is that the platform keeps the user's real name in the back office, the user decides the nickname shown on the page, and the check rests on a phone number or the like. Livestreaming and selling on livestreaming is stricter: 「基于身份证件信息、统一社会信用代码等真实身份信息认证」 — verification by real identity information such as identity document details and the unified social credit code. The unified social credit code is the string of characters on the business licence (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十六、六十四条；2016 年文本为第二十四、六十一条）. ; Cyberspace Administration of China (国家互联网信息办公室) (2016). Provisions on the Administration of Internet Livestreaming Services, 互联网直播服务管理规定（第十二、十六条）. ; Cyberspace Administration of China (国家网信办) and six other departments (2021). Interim Measures for the Administration of Online Livestreaming Marketing, 网络直播营销管理办法（试行）（国信办发文〔2021〕5 号，第八条）. +- Notes: The article numbers have moved. The Cybersecurity Law was amended in 2025 and comes into force on 2026-1-1. The clause on real names moved from Article 24 to Article 26, and the penalties moved from Article 61 to Article 64. The wording did not change. If the material in your hands still cites the old numbers, they have to be corrected + +### 8. Do not open livestreaming accounts for people under 16, and handle tipping money in bands by age + +- Cost: No money. Add one step at each of the two points, registration and payment: first check the age, then confirm that the guardian agreed +- In plain language: A person under 16 cannot register a livestreaming account with you. Someone who has reached 16 must have their identity verified, and must have the consent of their parents or other guardian. Tipping money falls into three bands: under 8 no tipping at all; 8 to 16 needs the guardian's consent; over 16 needs the guardian's consent or a check of proof of income. Where a child tips money behind the back of the guardian and the guardian demands it back, the court rules that it is returned. +- Benefit: The law says 「网络直播服务提供者不得为未满十六周岁的未成年人提供网络直播发布者账号注册服务；为年满十六周岁的未成年人提供网络直播发布者账号注册服务时，应当对其身份信息进行认证，并征得其父母或者其他监护人同意」. The regulations also require 「建立网络直播发布者真实身份信息动态核验机制」 — a mechanism for checking the real identity of the people who livestream, again at any time. They also require you to 「合理限制不同年龄阶段未成年人在使用其服务中的单次消费数额和单日累计消费数额」 — to set reasonable limits on a single payment and on the daily total for minors at each age band. Tipping now falls into three bands: 「不得向八周岁以下的未成年人提供打赏服务。向八周岁至十六周岁的未成年人提供打赏服务，应征得其监护人同意。向十六周岁以上的未成年人提供打赏服务，应征得其监护人同意或核验其收入证明材料」. There is one more clause: where a minor with limited capacity for civil conduct tips money without the guardian's consent and the guardian asks for the money back, the court supports it and the money can be recovered. A minor with limited capacity for civil conduct is one who cannot yet act entirely on their own in law (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Law on the Protection of Minors, 未成年人保护法（第七十四、七十六条）. ; State Council (国务院) (2023). Regulations on the Protection of Minors Online, 未成年人网络保护条例（国务院令第 766 号，第三十一、四十四条）. ; General Office of the Cyberspace Administration of China (中央网信办秘书局) (2026). Notice on Strengthening the Standardised Management of Online Livestreaming Tipping, 关于加强网络直播打赏规范管理的通知（第 7 项）. ; Supreme People's Court (最高人民法院) (2020). Guiding Opinions on the Proper Handling of Civil Cases Related to the Novel Coronavirus Epidemic under the Law, 关于依法妥善审理涉新冠肺炎疫情民事案件若干问题的指导意见（二）（法发〔2020〕17 号，第九条）. +- Notes: If a refund really is disputed, handle it in the direction that is best for the minor. The wording on this has changed. The opinions of the four departments in 2022 and the notice of the National Radio and Television Administration (广电总局) in 2020 both said that minors may not tip at all. The new notice of April 2026 changed it to bands at 8 and 16, but did not say that the earlier two documents are repealed, and the authorities have not said how the two fit together. So the sentence 「no minor may ever tip」 is no longer accurate. +### 9. Handle an infringement notice in time, and if nothing follows within 15 days of passing on the counter-notice, put the content back + +- Cost: No money. You need a fixed process: receive the notice, pass it to the seller, take the seller's counter-notice, and publish the result. Keep a record at every step +- In plain language: When somebody sends you an infringement notice with preliminary evidence attached, you have to delete the material, block it or take the link down at once, and at the same time pass the notice to the seller. If you sit on it, you have to pay the extra part of the loss together with the seller. If the seller counters, you pass the counter-notice to the other side and wait 15 days. If they neither complain to an authority nor sue, you put the material back. +- Benefit: The rights holder 「有权通知电子商务平台经营者采取删除、屏蔽、断开链接、终止交易和服务等必要措施。通知应当包括构成侵权的初步证据」 — is entitled to tell the operator of the e-commerce platform to take the necessary measures, such as deleting, blocking, cutting the link, ending the transaction and ending the service, and the notice has to include preliminary evidence of the infringement. The rights holder is the owner of the trademark, the copyright or the patent. A sentence on its own is not enough; the preliminary evidence has to go with it. After the platform receives the notice, it 「应当及时采取必要措施，并将该通知转送平台内经营者；未及时采取必要措施的，对损害的扩大部分与平台内经营者承担连带责任」. If you sit on it, you share the liability with the seller for the extra part of the loss. After the seller has countered, the platform 「在转送声明到达知识产权权利人后十五日内，未收到权利人已经投诉或者起诉通知的，应当及时终止所采取的措施」 — once the counter-notice has reached the rights holder, if within 15 days no notice comes that the rights holder has complained to an authority or sued, the platform must end the measures it took at once. The meaning is that you pass the seller's counter-notice to the rights holder and wait 15 days. If they neither complain to an authority nor sue, you put back what you deleted or took down. A platform that knows or should know of an infringement and does nothing shares the liability with the infringer. The intellectual property authority can also order you to put it right within a set period; if you do not, the fine is 50,000 to 500,000 yuan, and where the circumstances are serious it is 500,000 to 2 million yuan. The other way round, 「恶意发出错误通知，造成平台内经营者损失的，加倍承担赔偿责任」 — anyone who maliciously sends a false notice and causes the seller a loss pays double the damages (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). E-Commerce Law, 电子商务法（第四十二、四十三、四十四、四十五、八十四条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千一百九十五、一千一百九十六、一千一百九十七条）. +- Notes: If you are not an e-commerce platform but an ordinary website, the notice-and-takedown rules of the Civil Code apply. The difference is how long you have to wait after the other side counters: the E-Commerce Law gives a clear 15 days, while the Civil Code only says 「合理期限」 — a reasonable period. Publishing the notice you received, the seller's counter-notice and the result of your handling is an express requirement of Article 44 of the E-Commerce Law + +### 10. Do not put users' information outside the country in passing; transfers out of China have statutory conditions and a headcount threshold + +- Cost: No money. The least trouble is to keep the user data inside China. Where it really has to go out, use the standard contract or get certified +- In plain language: Sending information about users inside China to machines outside the country is what the law calls a transfer out of China. Having the server outside China does not mean the data has not gone out. To take it out you must use one of four routes: a security assessment by the cyberspace administration authority, certification by a professional body, a contract signed on the standard contract, or whatever else the law provides. You also have to tell the users who receives the information outside China and what it is used for, and you have to ask them about that one thing on its own. +- Benefit: Handing information about users inside China to a company outside China, or storing it on a machine outside China, is in law 「向境外提供个人信息」 — providing personal information abroad, and it 「应当具备」 — must have one of four conditions. First, a security assessment organised by the cyberspace administration authority. Second, certification for the protection of personal information by a professional body. Third, a contract signed with the recipient outside China on the standard contract drawn up by the cyberspace administration authority. Fourth, other conditions set by laws and administrative regulations. At the same time you have to tell the users the name of the recipient outside China, its contact details, the purpose of the processing, the way it is processed, the kinds of personal information, and how the user exercises their own rights. And you have to 「取得个人的单独同意」 — obtain the separate consent of the individual, by asking the user about this one thing on its own. The thresholds are set by the total number of people. An ordinary company is, in the regulations, a data handler other than an operator of critical information infrastructure. Such companies have one line below which nothing has to be done: 「自当年1月1日起累计向境外提供不满10万人个人信息（不含敏感个人信息）的」 — where, counted from 1 January of that year, fewer than 100,000 people's personal information other than sensitive personal information has been provided abroad. Below that line, all three of declaring for an assessment, signing the standard contract and being certified are waived. At 100,000 people or more but fewer than 1,000,000, you must sign the standard contract or pass the certification. At 1,000,000 or more, or where sensitive personal information reaches 10,000 people or more, you must declare for a security assessment (nationwide, from 2024-3) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Personal Information Protection Law, 个人信息保护法（第三十八、三十九、四十条）. ; Cyberspace Administration of China (国家互联网信息办公室) (2024). Provisions on Promoting and Standardising the Cross-Border Flow of Data, 促进和规范数据跨境流动规定（网信办令第 16 号，第三、四、五、七、八条）. ; Cyberspace Administration of China (国家互联网信息办公室) (2022). Measures for the Security Assessment of Data Transfers Out of China, 数据出境安全评估办法（网信办令第 11 号，第四条）. +- Notes: Having the server outside China is not the same as 「no data has gone out」. A user registers inside China and you send their information to a machine outside China, and that is a transfer out. There is one class that does not have to take the three routes of the security assessment, the certification and the standard contract: cross-border shopping, cross-border delivery and booking a flight or a hotel, because these are 「为订立、履行个人作为一方当事人的合同」 — transfers genuinely necessary to conclude or perform a contract to which the individual is a party. One more thing to watch is how the number of people is counted: it is counted 「自当年 1 月 1 日起累计」 — from 1 January of that year and accumulated, not over the twelve months counting back from today + +### 11. Choosing a server: first ask whether you can put up with the downtime, then whether anybody maintains it, and compare prices last + +- Cost: Cloud servers from the main cloud providers run from several hundred to several thousand yuan a month. A VPS from a small provider starts at a few tens of yuan a month. A VPS is a small virtual server. Colocation for a dedicated server starts at several thousand yuan a year, and the hardware you have to buy yourself +- In plain language: The three tiers differ not in price but in what you are holding on the day something fails. A large provider has a filing channel, publishes its qualifications, and when something breaks there is a notice, there is a compensation clause in the contract, and there is somebody you can find. A small provider costs half as much, but when it oversells, folds or is cut off upstream you do not even know who to complain to. For a static site with no user data, a VPS costing a few tens of yuan is enough. +- Benefit: The three tiers differ not in price but in what you are holding on the day something fails. The main cloud providers have a filing channel and publish their qualifications; when there is a fault there is a notice, the contract carries a compensation clause, and there is somebody you can find. A small provider can be half the price, but when it oversells, folds or is cut off upstream you have nobody to appeal to, so before you negotiate the price go to the MIIT system and check whether it holds a Value-Added Telecommunications Business Licence. Colocation for a dedicated server only makes sense when both conditions hold: bandwidth or hardware is clearly cheaper, and you have your own person to maintain it. How to choose: for a static site with no user data, the cheapest VPS is enough. For a site with user registration and transactions, choose a cloud server that scales on demand and takes snapshots, and keep the backups in at least two places +- Evidence grade: C +- Sources: the author's own experience, no direct literature; the statutory requirements on the qualifications of a provider and on the filing are in item 4 in this section, and the duty to keep the logs for 6 months and the graded protection duties are in item 16 in section 11 +- Notes: That is a judgement from experience, with no literature behind it. Only one thing can really be checked: the provider's licence. If you want to save money, save on the specifications and on the bandwidth, not on the qualifications + +## Licence + +The text is published under [CC BY 4.0](../LICENSE). If you reprint or adapt it, say where it came from and link the original. The code is under [MIT](../LICENSE-CODE). \ No newline at end of file diff --git "a/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" new file mode 100644 index 00000000..a3a7aa03 --- /dev/null +++ "b/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -0,0 +1,206 @@ +[← 回总目录](../README.md) + +# 33. 残疾之后怎么活 + +这一节写的是残疾已经发生之后的日子。怎么防住致残，在第 1 节和第 13 节，这里不重复。 + +四种口径分开算，不互相换算。急症和并发症那几条按死亡率算。补贴、待遇、打官司那几条按金钱算。上学、办事那几条按时间算。精神障碍住院、行为能力和监护那几条按人身自由算。 + +世界卫生组织的说明书给了三个数字。全球约有 13 亿人存在显著功能障碍，占世界人口的 16%，大约每六个人里有一个。部分残疾人比没有残疾的人早死最多 20 年。残疾人得抑郁、哮喘、糖尿病、卒中、肥胖或者口腔健康不良的风险是两倍。这些差距里，很大一部分不是残疾本身造成的，是看不上病、上不了学、找不到工作、出不了门造成的。所以这一节的条目，一半是身体，一半是手续和权利。 + +已经写在别处的五件事，这里只指路，不重复。残疾人证怎么办、七个类别和一到四级怎么评，见第 24 节第 11 条（残疾人证）。伤残鉴定要等到什么时候做，见第 24 节第 10 条（伤残鉴定）。残疾人两项补贴怎么领，见第 7 节第 8 条（两项补贴）。工伤的劳动能力鉴定和伤残等级换算成多少钱，见第 19 节第 15 条（劳动能力鉴定）。重度失能的长期护理保险怎么申请，见第 17 节第 7 条（长期护理保险）。那一条写在「家里有老人」那节，但这项待遇没有年龄门槛，年轻人一样能领。 + +本节大部分条目的受益人是你自己和家人，属于最高的两档。涉及别人的地方在备注里单独写。 + +[← Back to the contents](../README.md) + +# 33. Living after a disability + +This section is about the days after a disability has already happened. How to keep a disability from happening is in section 1 and section 13, and is not repeated here. + +Four measures are counted separately and are not converted into one another. The emergency and complication items are counted by mortality. The subsidies, benefits and lawsuits items are counted by money. The schooling and paperwork items are counted by time. The psychiatric admission, capacity and guardianship items are counted by personal freedom. + +The World Health Organization's briefing gives three numbers. Worldwide about 1.3 billion people have a significant functional disability, which is 16% of the world's population, roughly one in every six people. Some disabled people die up to 20 years earlier than people without a disability. Disabled people run twice the risk of depression, asthma, diabetes, stroke, obesity or poor oral health. A large part of those gaps is not caused by the disability itself, but by not being able to see a doctor, go to school, find work or get out of the house. So half the items in this section are about the body, and half about procedures and rights. + +Five things are already written elsewhere; here they are only pointed to, not repeated. How to get a disability certificate, and how the seven categories and grades one to four are assessed: see item 11 in section 24 (disability certificate). When to have the disability assessment done: see item 10 in section 24 (disability assessment). How to claim the two disability subsidies: see item 8 in section 7 (the two subsidies). For a work injury, the labour capacity assessment and how much a disability grade converts to in money: see item 15 in section 19 (labour capacity assessment). How to apply for long-term care insurance with a severe loss of function: see item 7 in section 17 (long-term care insurance). That item is written in the section on when there are elderly at home, but this benefit has no age limit, and young people can claim it too. + +Most items in this section benefit you and your family, which is the top two grades. Where they affect other people, that is written separately in the notes. + +### 1. After a spinal cord injury, a sudden severe headache and sweating: sit the person up straight first, loosen the clothing, and call 120 at the same time + +- Cost: No money. Three moves, done in a few tens of seconds. +- In plain language: After a spinal cord injury a section of the body's alarm wiring is cut. A full bladder or tight clothing may go unnoticed by the person, while the blood pressure surges. The signal is a sudden severe headache and sweating. Sit the person up straight, loosen anything tight, check whether they are holding urine, and call 120 at the same time. +- Benefit: This kind of attack is called autonomic dysreflexia. A systematic review searched four databases, Medline, CINAHL, EMBASE and PsycINFO, and included 31 studies, 6 of which were randomised controlled trials. The review states that the non-drug treatment during an attack is three steps: put the patient in an upright position, loosen tight clothing, and remove the trigger. The basis for these three steps is clinical consensus and physiological data, and they count as level 5 evidence. A blood-pressure drug is used only when the blood pressure stays high; for prazosin there is level 1 evidence, and for nifedipine and prostaglandin E2 level 2 evidence. The review also states that strategies to prevent attacks mainly target the two common triggers of the genitourinary system and of surgery, and that most are supported only by level 4 and level 5 evidence +- Evidence grade: B +- Sources: Krassioukov A, Warburton DE, Teasell R, Eng JJ; Spinal Cord Injury Rehabilitation Evidence Research Team (2009). A systematic review of the management of autonomic dysreflexia after spinal cord injury. Archives of Physical Medicine and Rehabilitation, 90(4):682-695. +- Notes: The grade is B because those three moves at the scene have no randomised controlled trial behind them, only consensus and physiological reasoning. The review itself says that most of these measures are supported only by uncontrolled studies. The drugs are the doctor's business; do not give a blood-pressure drug at home yourself. Laying the person flat will push the blood pressure higher, so sit them up straight. Family and care workers need to know these three steps too. Do not just ride it out as an ordinary headache. + +### 2. The first ten years after becoming disabled are the high-risk period for suicide; do not treat it as something to think your way out of + +- Cost: No money. Make one phone call, or register once at a hospital. +- In plain language: In the first years after becoming disabled, it is common for you or your family to break down emotionally; that is not a question of willpower. Long-term follow-up at two spinal cord injury centres in the United Kingdom found that this group dies by suicide at five times the ordinary rate. More than nine in ten suicides happen in the first ten years after the injury. Where the injury itself was a suicide attempt, the risk is higher again. +- Benefit: The study followed 2304 patients with traumatic spinal cord injury who were newly admitted between 1991 and 2010, survived the first year after the injury, and had a neurological deficit at discharge, following them to the end of 2014. In 63 of the cases (2.7%) the cause of the injury was itself a suicide attempt. By the end of the follow-up 533 people had died, of whom 4.2% died by suicide, and 91% of the suicides happened in the first ten years after the injury. The age-standardised suicide mortality was 62.5 per 100,000 people per year (95% CI 36.4–88.6, credible range), five times the rate of the general population of England and Wales in 2014 (12.2 per 100,000 people). For the people whose injury came from a suicide attempt, compared with other causes of injury, the odds ratio for death was 4.32 (P<0.001) and for suicide 9.46 (P<0.001) +- Evidence grade: A +- Sources: Savic G, DeVivo MJ, Frankel HL, Jamous MA, Soni BM, Charlifue S (2018). Suicide and traumatic spinal cord injury - a cohort study. Spinal Cord, 56(1):2-6. +- Notes: This study looked only at spinal cord injury, so applying it to blindness, amputation, burns and other causes of disability is an extrapolation. What it can tell you is the time window: the risk does not end at discharge, and you have to watch it for ten years. To have somebody to talk to, call 12356: see item 11 in section 29 (12356). Do not stock sleeping pills and pesticide at home: see item 25 in section 1 (thoughts of suicide). If grief and mood are stuck in the same place after half a year, register at a psychiatry or clinical psychology department: see item 8 in section 29 (grief past half a year). + +### 3. Psychiatric admission follows the principle of voluntariness; only two circumstances allow it without the person's consent + +- Cost: No money. Remember the two circumstances so that you can say them when you need them. +- In plain language: One word from the family or the employer cannot keep a person in a psychiatric hospital. The law recognises only two situations: the person has already harmed themselves or is at risk of doing so, and the person has already endangered others or is at risk of doing so. The first requires a guardian's consent before the person may be admitted. If the diagnosis is disputed, a fresh diagnosis may be requested within three days, and after that the person can commission an appraisal themselves. +- Benefit: Article thirty of the Mental Health Law (《精神卫生法》) states 「精神障碍的住院治疗实行自愿原则」, that is, admission for the treatment of a mental disorder follows the principle of voluntariness. Inpatient treatment may be carried out only where the diagnostic conclusion and the assessment of the condition show that the patient has a severe mental disorder and one of the following applies. One: 「已经发生伤害自身的行为，或者有伤害自身的危险的」, that is, the person has already carried out an act of self-harm, or is at risk of self-harm. Two: 「已经发生危害他人安全的行为，或者有危害他人安全的危险的」, that is, the person has already endangered the safety of others, or is at risk of doing so. Article thirty-one: in the first case the medical institution may admit the person only with the guardian's consent, and may not admit them if the guardian does not agree. Article thirty-two: in the second case, where the person or the guardian objects, a fresh diagnosis may be requested; the request must be made within three days of receiving the diagnostic conclusion, and must be made by two practising psychiatrists other than the doctor who made the first diagnosis; where there is still an objection to the fresh diagnosis, the person may commission an appraisal body lawfully qualified to appraise to carry out a medical appraisal of the mental disorder. Article thirty-five: where the fresh diagnosis or the appraisal report shows that the person cannot be established as having a severe mental disorder, or that inpatient treatment is not needed, the medical institution may not carry out inpatient treatment. Article seventy-eight: a person who deliberately has a patient without a mental disorder sent to a medical institution for treatment as though they had one bears liability in accordance with the law (nationwide, in force from 2013-5-1, amended 2018) +- Evidence grade: A +- Sources: Mental Health Law of the People's Republic of China（中华人民共和国精神卫生法）(2018 amendment, articles thirty, thirty-one, thirty-two, thirty-five and seventy-eight). +- Notes: The other way round, not just anybody can be sent in either. Article twenty-eight provides that where a person suspected of having a mental disorder has already carried out an act of self-harm or of endangering another's safety, or is at risk of doing so, their close relatives, their work unit or the local public security organ must immediately take measures to stop it and send them to a medical institution for a diagnosis of mental disorder. A medical institution that receives such a person may not refuse to make a diagnosis. So "sent in for a diagnosis" and "kept in for admission" are two doors: the first is low, the second is high. Article seventy-five lists the medical institution's own liability; carrying out restraint or isolation against the rules, or diagnosing a person without a mental disorder as having one, leads to discipline up to withdrawal of the practising certificate. What to do when somebody near you has an urge to hurt someone: see section 8. + +### 4. The person who cares long-term for a disabled family member needs somebody watching their own health + +- Cost: No money. Hand a share of the caring work to somebody else, and get a check-up yourself. +- In plain language: A carer collapsing is not self-indulgence. A United States study followed more than eight hundred older people for four years. The ones who were caring for a disabled spouse and also reported mental strain had about sixty per cent higher death risk than people who did not care for anyone. Carers who did not feel strained showed no such rise. So what needs watching is not the caring itself, but that long-term tension. +- Benefit: The study was a prospective population cohort, run in four communities in the United States between 1993 and 1998, with an average follow-up of 4.5 years. There were 392 carers and 427 non-carers, aged 66 to 96, all living with a spouse. Within four years 103 people (12.6%) died. After adjusting for socio-demographic factors, existing illness and subclinical cardiovascular disease, people who were caring and reported mental or emotional strain had a death risk 63% higher than the non-caring controls (RR 1.63, 95% CI 1.00–2.65, credible range). Carers who did not report strain (RR 1.08, 95% CI 0.61–1.90) and people whose spouse was disabled but who did not do the caring themselves (RR 1.37, 95% CI 0.73–2.58) showed no rise in adjusted mortality +- Evidence grade: A +- Sources: Schulz R, Beach SR (1999). Caregiving as a risk factor for mortality: the Caregiver Health Effects Study. JAMA, 282(23):2215-2219. +- Notes:Disputed: The lower bound of the credible range in this study sits exactly on 1.00, so the possibility of "no difference" has not been ruled out. The population is older Americans living with a spouse, caring for an older spouse who is disabled; applying it to people caring for a younger disabled relative is an extrapolation. What it can tell you is the direction: the people who find it a strain run higher risk. Hand over as much of the caring as you can; ask the local federation of disabled persons and the civil affairs department about respite and day-care services. The carer should not stop their own chronic illness medication: see item 1 in section 16 (take the medicine as prescribed). The person who benefits from this item is yourself. + +### 5. If you sit in a wheelchair long-term, switch to a pressure-relief cushion and check your sitting bones and tailbone every day + +- Cost: A pressure-relief cushion runs from a few hundred yuan to over a thousand. One look a day, one minute. +- In plain language: An ordinary foam cushion presses on the skin long enough to break it, and once broken it heals slowly and easily gets infected. Older people in nursing homes who use pressure-relief cushions have clearly fewer pressure ulcers over the sitting bones, but once the tailbone is counted in there is no difference. The wheelchair itself also has to be fitted to the person's measurements; do not settle for a standard model. +- Benefit: This randomised clinical trial was done in 12 nursing homes, recruiting 232 residents between 2004 and 2008 who were over 65, sat in a wheelchair more than 6 hours a day, and scored below 18 on the Braden scale. Everyone was first fitted with a well-fitting wheelchair, then randomised into two groups. One group used a skin-protection cushion (air, viscous fluid with foam, or gel with foam), the other a 7.6 cm thick segmented foam cushion. Follow-up was 6 months, or until a pressure ulcer appeared. Pressure ulcers at the ischial tuberosity: 8 cases (6.7%) in the segmented foam group, 1 case (0.9%) in the skin-protection cushion group, P=0.04. Counting ulcers at the ischium and the sacrococcygeal area together: 21 cases (17.6%) in the segmented foam group, 12 cases (10.6%) in the skin-protection cushion group, P=0.14, which showed no difference +- Evidence grade: B +- Sources: Brienza D, Kelsey S, Karg P, et al. (2010). A randomized clinical trial on preventing pressure ulcers with wheelchair seat cushions. Journal of the American Geriatrics Society, 58(12):2308-2314. +- Notes: There are two reasons for the grade B. First, the sample was only 232 people, the P value of 0.04 on the primary endpoint had just crossed the line, and the secondary endpoint showed no difference. Second, the population is very old nursing-home residents, so applying it to young wheelchair users is an extrapolation. In the trial everyone was first fitted with a well-fitting wheelchair, so "fits" and "cushion" work together; do not change only the cushion. For pressure ulcers during long bed rest, an air mattress and turning on a schedule: see item 8 in section 17 (pressure ulcers). How to apply for the subsidy for fitting basic assistive devices: see item 7 in this section (ask everything at the county federation). + +### 6. Do not buy therapies and devices that "can cure paralysis, blindness or deafness" + +- Cost: No money, it saves money. The hard part is that the family is holding on to hope and cannot be talked out of it. +- In plain language: The Advertising Law has settled the wording. Advertisements for medical treatment, drugs and medical devices may not guarantee the effect, may not state a cure rate or an effective rate, and may not use people giving their own testimony. So "stand up after three courses of treatment" or "how many people have already recovered" is itself an unlawful advertisement, and there is no need to examine the theory it puts forward. Something that is not a medical device may not even mention treating illness. +- Benefit: Article sixteen of the Advertising Law (《广告法》) provides that advertisements for medical treatment, drugs and medical devices may not contain the following: 「（一）表示功效、安全性的断言或者保证；（二）说明治愈率或者有效率；（三）与其他药品、医疗器械的功效和安全性或者其他医疗机构比较；（四）利用广告代言人作推荐、证明」, that is, assertions or guarantees of efficacy or safety; statements of a cure rate or an effective rate; comparisons with the efficacy or safety of other drugs or devices, or with other medical institutions; and recommendations or endorsements by advertising spokespeople. Article seventeen provides that, apart from advertisements for medical treatment, drugs and medical devices, 「禁止其他任何广告涉及疾病治疗功能，并不得使用医疗用语或者易使推销的商品与药品、医疗器械相混淆的用语」, that is, no other advertisement may concern the treatment of disease, and none may use medical terms, or terms likely to confuse the goods being promoted with drugs or medical devices. Article four provides that advertisements may not contain false or misleading content, may not deceive or mislead consumers, and the advertiser is responsible for the truth of the content (nationwide, amended 2021) +- Evidence grade: A +- Sources: Advertising Law of the People's Republic of China（中华人民共和国广告法）(2021 amendment, articles four, sixteen and seventeen). +- Notes: There is only one test: see whether it dares to promise everything. The rehabilitation department of a proper hospital will not promise you that you will stand up after a number of courses of treatment. When you meet this kind of advertising, report it to 12315 or to the local market regulator. Health supplements use the same script: see item 10 in section 6 (health supplements). If you have already bought and want your money back, follow the method for online shopping and pre-paid consumption: see item 29 in section 5 (online shopping follows the platform's rules). The people who benefit from this item are yourself and your family. + +### 7. Once the disability certificate is issued, go to the county federation of disabled persons and ask about everything you can apply for in one visit + +- Cost: No money. One trip to the county federation and a morning of asking. There is more than one department to visit afterwards. +- In plain language: The certificate itself pays nothing; it is a key. What hangs off it is several separate benefits, and each one you have to apply for yourself — none of them comes to you on its own. On the first visit to the federation, go through the list of items one by one; that saves trips back and forth later. +- Benefit: The certificate unlocks at least six benefits, each with its own basis. One: the living allowance for persons with disabilities in difficulty and the nursing allowance for persons with severe disabilities — see item 8 in section 7 (the two allowances). Two: rehabilitation assistance for children with disabilities — see item 8 in this section (rehabilitation assistance). Three: a subsidy for fitting basic assistive devices, based on Article twenty-six of the Regulations on Disability Prevention and Rehabilitation of Persons with Disabilities (《残疾预防和残疾人康复条例》). Four: a subsidy for home accessibility modifications — see item 9 in this section (ramps, handrails and bathroom modifications). Five: quota hiring and employment services — see item 10 in this section (declare the certificate). Six: a personal income tax reduction — see item 11 in this section (personal income tax reduction). Article twenty-six of the Regulations also states that the state provides basic rehabilitation services to poor and severely disabled people in urban and rural areas through key rehabilitation projects, and 「按照国家有关规定对基本型辅助器具配置给予补贴」, that is, subsidises the fitting of basic assistive devices in accordance with relevant state provisions. Article twenty provides that the relevant departments of people's governments at county level and above, and the federations of disabled persons, shall use community resources to set up rehabilitation premises or purchase services, and carry out community rehabilitation work such as rehabilitation guidance, daily-living-skills training, rehabilitation nursing, assistive-device fitting, information consultation and referrals (nationwide, in force from 2017-7-1) +- Evidence grade: A +- Sources: State Council (2017). Regulations on Disability Prevention and Rehabilitation of Persons with Disabilities (《残疾预防和残疾人康复条例》) (articles twenty and twenty-six). +- Notes: The process of getting the certificate itself: see item 11 in section 24 (disability certificate). Most subsidy standards and specific measures are set at the provincial level, and the same benefit can differ several times over between two provinces. Write down the name of each item when you ask, then go home and look up the original document again on the local federation's website and the government website, by name; do not rely on what the person at the counter tells you. Do not expect to ask everything in one visit — go back and ask again after a year, policies change. + +### 8. If the child is under 7 and has a disability or autism, apply to the county federation for rehabilitation assistance + +- Cost: No money. Submit one application at the county federation, then choose one institution from the designated list. +- In plain language: The state has built a rehabilitation assistance system for children with disabilities and children with autism aged 0 to 6, covering surgery, assistive devices and rehabilitation training. Families in difficulty are served first; where local finances allow, the age range and the household income conditions may be relaxed. This is the window in which rehabilitation works best — do not wait until it is time for school. Institutions must be chosen from the federation's designated list. +- Benefit: The State Council's Opinions on Establishing a Rehabilitation Assistance System for Children with Disabilities (《国务院关于建立残疾儿童康复救助制度的意见》) states that the beneficiaries are eligible children aged 0–6 with visual, hearing, speech, physical or intellectual disabilities, and children with autism. These include children with disabilities from families on the urban or rural minimum living allowance, from registered poor households, children with disabilities raised in child welfare institutions, disabled orphans, children with disabilities within the scope of support for persons in extreme poverty, and children with disabilities from other financially strained families. The specific criteria for "other financially strained families" are laid down by local people's governments at county level and above. The document also states 「有条件的地区，可扩大残疾儿童康复救助年龄范围，也可放宽对救助对象家庭经济条件的限制」, that is, areas with the means may widen the age range of the assistance and relax the household income conditions for beneficiaries. Funding standards are set by people's governments at county level and above according to local finances, the number of beneficiaries and the category of disability, with a dynamic adjustment mechanism. The higher-level basis is Article twenty-six of the Regulations on Disability Prevention and Rehabilitation of Persons with Disabilities: the state establishes a rehabilitation assistance system for children with disabilities, 「逐步实现0—6岁视力、听力、言语、肢体、智力等残疾儿童和孤独症儿童免费得到手术、辅助器具配置和康复训练等服务」, that is, gradually ensuring that children aged 0–6 with visual, hearing, speech, physical or intellectual disabilities and children with autism receive surgery, assistive-device fitting and rehabilitation training free of charge (nationwide, from 2018) +- Evidence grade: A +- Sources: State Council (2018). Opinions on Establishing a Rehabilitation Assistance System for Children with Disabilities (《国务院关于建立残疾儿童康复救助制度的意见》) (Guo Fa [2018] No. 20). ; State Council (2017). Regulations on Disability Prevention and Rehabilitation of Persons with Disabilities (《残疾预防和残疾人康复条例》) (article twenty-six). +- Notes: The assistance standards are set by each province; first search the local federation's website for 「残疾儿童康复救助实施办法」. The designated-institution list is also in the federation's hands. The document requires the federation, together with the relevant departments, to tighten supervision of entry and exit for designated rehabilitation institutions, establish regular inspection and comprehensive evaluation mechanisms, and set up a blacklist. For an institution not on the list you pay every yuan yourself, and if something goes wrong it is hard to hold anyone to account. The beneficiary of this item is your child, which puts it in the second tier. Institutions that promise to "guarantee a cure" are handled under item 6 in this section (can cure paralysis). + +### 9. Ramps, handrails and bathroom modifications at home can be subsidised by a government at county level or above + +- Cost: Applying costs nothing. Most of the work itself is subsidised, but you may still have to cover part of it. You will have to visit the federation and the subdistrict office. +- In plain language: The Law on Barrier-free Environment (《无障碍环境建设法》) has a clause: governments at county level and above shall support and guide barrier-free home modifications, and shall give appropriate subsidies to qualifying households with disabled or elderly members. These projects are generally led by the federation and run on annual quotas, so register early. How much is paid and which items are covered is decided locally. +- Benefit: Article nineteen of the Law on Barrier-free Environment: 「县级以上人民政府应当支持、指导家庭无障碍设施改造。对符合条件的残疾人、老年人家庭应当给予适当补贴。」, that is, people's governments at county level and above shall support and guide barrier-free home modifications, and shall give appropriate subsidies to qualifying households with disabled or elderly members. Article eighteen provides that, for existing residential buildings, residential areas, public buildings, public places, transport facilities, urban and rural roads and the like that do not meet the barrier-free facility engineering construction standards, people's governments at county level and above shall, in light of actual circumstances, formulate targeted barrier-free facility modification plans and organise their implementation. Article sixty-two provides that any organisation or individual has the right to put opinions and suggestions to the relevant competent departments of the government on strengthening and improving the barrier-free environment, and to complain or report behaviour that violates this Law; a competent department that receives such a complaint or report 「应当及时处理并予以答复」, that is, shall handle it promptly and reply. Article sixty-three provides that where conduct in violation of this Law harms the public interest, the people's procuratorate may raise a procuratorial suggestion or bring public interest litigation (nationwide, in force from 2023-9-1) +- Evidence grade: B +- Sources: Barrier-free Environment Construction Law of the People's Republic of China（中华人民共和国无障碍环境建设法）(articles eighteen, nineteen, sixty-two and sixty-three). +- Notes: The grade is B because the law only writes 「应当给予适当补贴」 — shall give appropriate subsidies; how much is paid and who counts as qualifying is handed entirely to the localities, with no nationally uniform number. First search the local federation's website for 「困难重度残疾人家庭无障碍改造」. Quotas are mostly allocated by year, so queue early. Adding a lift to a residential building and modifying stairwells is a different process, through the housing and urban-rural development department. Where the barrier-free facilities in a public place fall short, complain to the competent department under Article sixty-two — it has to answer you. + +### 10. When job hunting, say up front that you hold the certificate; hiring you offsets a payment for the firm + +- Cost: No money. One extra sentence in the interview. The hard part is bringing it up. +- In plain language: The proportion of disabled employees a work unit arranges may not be lower than 1.5% of its total on-the-job staff, with the exact proportion set by the province. Fall short and the unit has to pay the disability employment security fund on the head-count gap, each head calculated on the unit's average annual staff wage of the previous year. So hiring you is not just a favour — it saves real money. Say clearly in the interview that you hold the certificate, and the other side can do the arithmetic itself. +- Benefit: Article eight of the Regulations on Employment of Persons with Disabilities (《残疾人就业条例》): 「用人单位安排残疾人就业的比例不得低于本单位在职职工总数的1.5%。具体比例由省、自治区、直辖市人民政府根据本地区的实际情况规定。」, that is, the proportion of disabled employees a unit arranges may not be lower than 1.5% of its own total on-the-job staff, and the specific proportion is set by the people's government of the province, autonomous region or municipality according to local circumstances. Article nine: a unit that fails to reach the proportion set by the provincial people's government where it is located shall pay the disability employment security fund. Article eight of the Measures for the Collection and Use of the Disability Employment Security Fund (《残疾人就业保障金征收使用管理办法》) gives the formula: 「保障金年缴纳额=（上年用人单位在职职工人数×所在地省、自治区、直辖市人民政府规定的安排残疾人就业比例-上年用人单位实际安排的残疾人就业人数）×上年用人单位在职职工年平均工资」, that is, annual fund payable = (previous year's on-the-job staff × the disabled-employment proportion set by the provincial government − previous year's disabled employees actually hired) × previous year's average annual staff wage. Article sixteen: for a small or micro enterprise within 3 years of business registration that has not reached the required proportion of disabled employment and whose total on-the-job staff is 20 or fewer (20 included), the fund is waived. Article thirty-six of the Law on the Protection of Persons with Disabilities (《残疾人保障法》) provides that the state gives tax preferences in accordance with the law to work units that reach or exceed the required proportion of disabled employment, or that employ disabled people centrally (nationwide, in force from 2007-5-1) +- Evidence grade: A +- Sources: State Council (2007). Regulations on Employment of Persons with Disabilities (《残疾人就业条例》) (articles eight and nine). ; Ministry of Finance, State Taxation Administration and China Disabled Persons' Federation (2015). Measures for the Collection and Use of the Disability Employment Security Fund (《残疾人就业保障金征收使用管理办法》) (Cai Shui [2015] No. 72, articles six, eight and sixteen). ; Law of the People's Republic of China on the Protection of Persons with Disabilities（中华人民共和国残疾人保障法）(article thirty-six). +- Notes: For a small or micro enterprise registered within the last three years with 20 staff or fewer, this reason does not hold — do not count on it. The fund's reduction and exemption policies have been adjusted year after year; what applies is whatever the Ministry of Finance announces that year. Also, quota hiring requires real employment. Taking a wage for a nominal post, or letting a unit use you to offset the fund, is fraud, and you would be dragged into it too. Recognition as a person having difficulty finding work, and the social insurance allowance: see item 12 in section 7 (persons having difficulty finding work). Vocational training allowance: see item 13 in section 7 (vocational training allowance). The person who benefits from this item is yourself. + +### 11. Personal income tax can be reduced for disabled people; phone the provincial tax bureau to find out by how much + +- Cost: No money. Tick one box when filing, or submit filing materials once. +- In plain language: The Individual Income Tax Law says the income of disabled people may be taxed at a reduced rate. But how much and for how long is left to provincial governments, so it differs from province to province. Some provinces set an annual reduction cap, others reduce by a percentage. The way to find out is to call 12366 and ask the local tax bureau, or look up the reduction items in the Individual Income Tax app. +- Benefit: Article five of the Individual Income Tax Law (《个人所得税法》): 「有下列情形之一的，可以减征个人所得税，具体幅度和期限，由省、自治区、直辖市人民政府规定，并报同级人民代表大会常务委员会备案：（一）残疾、孤老人员和烈属的所得；（二）因自然灾害遭受重大损失的。」, that is, individual income tax may be reduced in either of the following cases, the extent and duration set by the people's government of the province, autonomous region or municipality and filed with the standing committee of the people's congress at the same level: first, the income of persons with disabilities, elderly persons living alone and dependants of martyrs; second, major losses suffered through natural disaster. Article thirty-six of the Law on the Protection of Persons with Disabilities separately provides that the state gives tax preferences in accordance with the law to disabled persons engaged in individual business, and waives administrative and institutional fees (nationwide, in force from 2019-1-1) +- Evidence grade: A +- Sources: Individual Income Tax Law of the People's Republic of China（中华人民共和国个人所得税法）(2018 amendment, article five). ; Law of the People's Republic of China on the Protection of Persons with Disabilities（中华人民共和国残疾人保障法）(article thirty-six). +- Notes: The benefit is graded "small" because the law itself gives no number; in most provinces the annual reduction comes to a few hundred to a few thousand yuan. It is not much money, but one filing keeps working for years, so it is worth the trip. How much your own province reduces is decided only by the local tax bureau's announcement; comparison tables passed around online are out of date year-round. If you run a small shop of your own, the administrative and institutional fees are waived under Article thirty-six of the Law on the Protection of Persons with Disabilities — raise it when you apply for the licence. + +### 12. Guide dogs may be taken into public places and public transport; blind people ride city buses free + +- Cost: No money. If you are stopped, you have to open your mouth and quote the statute, and there may be a standoff for a while. +- In plain language: The law states that public places and public transport must make things convenient for disabled people who carry guide dogs, hearing-assistance dogs or assistance dogs. Blind people with valid ID ride city buses, the metro and ferries free, and disabled people may carry their necessary assistive devices free. When you are stopped, state these two provisions; if that does not work, complain. +- Benefit: Article forty-six of the Law on Barrier-free Environment (《无障碍环境建设法》): 「公共场所经营管理单位、交通运输设施和公共交通运输工具的运营单位应当为残疾人携带导盲犬、导听犬、辅助犬等服务犬提供便利。」, that is, operators of public places, of transport facilities and of public transport vehicles shall provide convenience for disabled people carrying service dogs such as guide dogs, hearing-assistance dogs and assistance dogs. Article thirty-five provides that government service hotlines and emergency call systems — for police, fire, traffic accidents, medical emergencies and the like — shall gradually have barrier-free functions such as voice, large print, Braille and one-touch calling. Article sixty-two provides that any organisation or individual has the right to complain or report behaviour that violates this Law, and the competent department shall handle it promptly and reply. Article fifty of the Law on the Protection of Persons with Disabilities (《残疾人保障法》): 「盲人持有效证件免费乘坐市内公共汽车、电车、地铁、渡船等公共交通工具。盲人读物邮件免费寄递。」, that is, blind people with valid ID ride city buses, trolleybuses, the metro, ferries and other public transport free, and mail containing reading materials for blind people is sent free. The same article also provides that people's governments at county level and above shall give disabled people convenience and preferential terms when they use public transport, and that disabled people may carry their necessary assistive devices free (nationwide, in force from 2023-9-1) +- Evidence grade: A +- Sources: Barrier-free Environment Construction Law of the People's Republic of China（中华人民共和国无障碍环境建设法）(articles thirty-five, forty-six and sixty-two). ; Law of the People's Republic of China on the Protection of Persons with Disabilities（中华人民共和国残疾人保障法）(article fifty). +- Notes: The law says "provide convenience", with no penalty attached, so in a standoff on the spot the other side may not give way straight away. Keep a screenshot of the statute on your phone — that works better than reciting article numbers. If it cannot be settled on the spot, complain under Article sixty-two; the competent department has to answer you. Railways and civil aviation have their own rules; call customer service before you travel. The "valid ID" accepted for free travel is not recognised identically everywhere — ask the local federation which kind counts. + +### 13. A candidate with a disability sitting the national college entrance examination can apply for reasonable accommodation; Braille papers add half again to the exam time + +- Cost: No money. You have to hand in one written application within the time the province sets, and then turn up in person for confirmation +- In plain language: A candidate with a disability sitting the national college entrance examination can apply for a Braille paper, a large-print paper, exemption from the foreign language listening test, a separate exam room, being allowed to bring in a hearing aid or a wheelchair, and extra time. None of it is granted automatically. You have to hand in a written application within the time the province sets, with a disability certificate of the second generation or above attached. If you disagree with the result, you can ask the provincial education department for a review +- Benefit: Article five of the Provisions on the Administration of the National Unified Examination for Admission to Regular Higher Education Institutions by Candidates with Disabilities《残疾人参加普通高等学校招生全国统一考试管理规定》lists twelve forms of reasonable accommodation. They include: giving candidates with a visual disability the current Braille paper, a large-print paper or an ordinary paper; exempting candidates with a hearing disability from the foreign language listening test; letting candidates with a visual disability bring in a Braille pen, a Braille writing slate, a Braille typewriter with no storage function, an electronic video magnifier with no storage function, a white cane, an optical magnifier and other assistive devices; letting candidates with a hearing disability bring in a hearing aid or a cochlear implant; letting candidates with limited mobility use a wheelchair or a walking aid, and bringing their own special desk and chair where they have special needs. The size of the extension is in Article five, item six: 「使用盲文试卷的视力残疾考生的考试时间，在该科目规定考试总时长的基础上延长50%；使用大字号试卷或普通试卷的视力残疾考生、因脑瘫或其他疾病引起的上肢无法正常书写或无上肢考生等书写特别困难考生的考试时间，在该科目规定考试总时长的基础上延长30%」. There is also priority entry to the examination venue and room, a separate standardised examination room, and the provision of guiding assistants and sign language interpreters. Article seven sets out the application procedure: the candidate lodges a formal written application at the time, place and in the manner set by the provincial education examination authority, and provides their own disability certificate of the second generation or above; an expert panel confirms it on the spot and makes an overall assessment, and the provincial education examination authority issues a Notification. Article eight says that a candidate who objects to the Notification may apply in writing to the provincial education administrative department for a review. Article nine says that where the foreign language listening test is exempted, the foreign language score is worked out as 「笔试成绩×外语科总分值/笔试部分总分值」. Article eighteen says that candidates with disabilities taking other state education examinations who need reasonable accommodation may apply these provisions by reference (nationwide, issued and in force from 2017-4-7) +- Evidence grade: A +- Sources: Ministry of Education (教育部) and China Disabled Persons' Federation (中国残联) (2017). Provisions on the Administration of the National Unified Examination for Admission to Regular Higher Education Institutions by Candidates with Disabilities (Teaching〔2017〕4, Articles five, seven, eight, nine and eighteen)《残疾人参加普通高等学校招生全国统一考试管理规定（教学〔2017〕4 号，第五条、第七条、第八条、第九条、第十八条）》. +- Notes: The application dates are set by each province and written into that year's rules for signing up for the examination. Miss the date and there is no second one. So check the provincial education examination authority's website in the first term of the final year of high school. Article eighteen also covers the postgraduate entrance examination, the adult higher education entrance examination and other state education examinations, which may apply by reference. The rule that a school must not refuse admission during the school years is item 14 in this section (schools must not refuse admission). The person who benefits from this entry is you or your child + +### 14. A disabled child applying for a school place: the school must not refuse; when the child cannot get to school the education authority arranges teaching at home + +- Cost: No money. You have to hand in one application and there may be one assessment. When the school makes excuses, you go up to the education authority +- In plain language: Kindergartens and all kinds of schools must not refuse admission for a disabled child who meets the conditions, and ordinary vocational schools are no different. Where a child needs one-to-one care and cannot get to school, the county education bureau arranges teaching at home or remote teaching, and the child still has a student record. Admission and transfer are decided by the education bureau after an assessment by an expert committee it convenes. The school does not get to decide on its own +- Benefit: Article seven of the Regulations on Education for Persons with Disabilities《残疾人教育条例》says: 「学前教育机构、各级各类学校及其他教育机构应当依照本条例以及国家有关法律、法规的规定，实施残疾人教育；对符合法律、法规规定条件的残疾人申请入学，不得拒绝招收。」 Article twenty-nine says: 「普通职业学校不得拒绝招收符合国家规定的录取标准的残疾人入学，普通职业培训机构应当积极招收残疾人入学。」 The regulations also say that where a disabled child or young person of school age needs one-to-one care and cannot attend school, the education administrative department of the county-level people's government makes the arrangements as it sees fit, delivers compulsory education by providing teaching at home or distance education, and keeps the child on the student register. Article twenty says the education administrative department of the county-level people's government shall, together with the health administrative department, the civil affairs department and the disabled persons' federation, set up an expert committee on education for persons with disabilities made up of specialists in education, psychology, rehabilitation and social work. That committee assesses the physical condition, the ability to receive an education and the ability to adapt to school life and study, and puts forward recommendations on admission and transfer. The assessment result is private to the child, and the units and staff involved have a duty of confidentiality. Article twenty-three says that a disabled student taking classes alongside other students in an ordinary school may follow the ordinary compulsory education curriculum, curriculum standards and textbooks, but that the learning requirements may be given some flexibility. Article fifty-two says that candidates with disabilities who need the necessary support and reasonable accommodation for a state education examination may apply (nationwide, in force from 2017-5-1) +- Evidence grade: A +- Sources: State Council (国务院) (2017). Regulations on Education for Persons with Disabilities (Articles seven, twenty, twenty-three, twenty-nine and fifty-two)《残疾人教育条例（第七条、第二十条、第二十三条、第二十九条、第五十二条）》. +- Notes: A school saying 「我们这儿没条件」 in conversation does not count. Make it give a written reply, and take that reply to the county education bureau. The assessment is done by an expert committee the education bureau commissions. The words in Article twenty-three, 「学习要求可以有适度弹性」, are useful: you can cite them when a school tries to push a child out on the grounds of results. For keeping evidence of bullying at school and which procedure the school must follow, see item 3 in section 30 (school bullying). For reasonable accommodation in the national college entrance examination see item 13 in this section (reasonable accommodation). The person who benefits from this entry is your child + +### 15. You can still get a driving licence without the right leg or without both legs; the vehicle category is called C5 + +- Cost: The tuition is about the same as for an ordinary licence, a few thousand yuan in most places. The car has to be converted to manual controls. A medical check every three years +- In plain language: Someone who has lost the right leg or both legs, or the use of them, but can sit upright unaided and whose upper limbs meet the conditions, can take the test for the special automatic car licence for disabled drivers, code C5. Someone missing one hand can apply as well. People with vision loss in one eye and people with a hearing loss each have vehicle categories they can take. The medical examination has to be done at a designated hospital recognised by the province. +- Benefit: Annex 1 of the Provisions on the Application for and Use of Motor Vehicle Driving Licences《机动车驾驶证申领和使用规定》sets out the physical conditions in detail. Item 8: 「右下肢、双下肢缺失或者丧失运动功能但能够自主坐立，且上肢符合本项第5目规定的，可以申请残疾人专用小型自动挡载客汽车准驾车型的机动车驾驶证。一只手掌缺失，另一只手拇指健全，其他手指有两指健全，上肢和手指运动功能正常，且下肢符合本项第6目规定的，可以申请残疾人专用小型自动挡载客汽车准驾车型的机动车驾驶证」. Item 6: someone missing the left lower limb alone, or the use of it, but with a normal right lower limb, may apply for the small automatic car category. On vision: 「单眼视力障碍，优眼裸视力或者矫正视力达到对数视力表5.0以上，且水平视野达到150度的」 may apply for the categories small car, small automatic car, low-speed goods vehicle, three-wheeled vehicle, and the special small automatic passenger car for disabled drivers. On hearing: a person with a hearing loss who can tell where a sound is coming from while wearing a hearing device may apply for the categories small car and small automatic car, and must wear the hearing device while driving. Article eighty-five says that the proof of physical condition submitted when a licence for the special small automatic passenger car for disabled drivers is issued must come from a designated medical institution recognised by the provincial health administrative department. A holder of this vehicle category must have a medical examination every three years and submit the proof of physical condition within thirty days after the end of the scoring period. Article seventy-eight says that while driving, the disabled motor vehicle sign must be put on the body of the vehicle as prescribed (nationwide, amended by Ministry of Public Security Order No. 172, effective 1 January 2025) +- Evidence grade: A +- Sources: Ministry of Public Security Order No. 172 (公安部令第 172 号). Provisions on the Application for and Use of Motor Vehicle Driving Licences (Articles seventy-eight and eighty-five, Annex 1)《机动车驾驶证申领和使用规定（第七十八条、第八十五条、附件 1）》. +- Notes: A much lower bar is the motorised wheelchair for disabled people. Article one hundred and nineteen of the Road Traffic Safety Law《道路交通安全法》puts it among non-motor vehicles, so no licence is needed. Article fifty-eight says that a motorised wheelchair for disabled people may not exceed fifteen kilometres per hour in a non-motor vehicle lane. A C5 car needs a hand-operated accelerator and brake. Have the conversion done by a qualified manufacturer, and then go to the vehicle administration office to change the registration. If you do not submit the proof of physical condition within one year after three scoring periods end, the licence is cancelled + +### 16. Do your rehabilitation in a proper institution that has a rehabilitation medicine department, and let the therapist set the intensity; more is not better + +- Cost: Inpatient rehabilitation costs several hundred yuan a day, and the part you pay yourself varies from place to place. The time works out in months. You have to practise every day, and it is very dull +- In plain language: Taking the person home, putting them in bed and waiting for natural recovery is the commonest mistake. After a stroke, organised care by a multidisciplinary team leaves more people alive and living at home a year later. But do not add to it on your own either. Starting earlier, more often and doing more within 24 hours of a stroke gives worse results at three months. How much to practise, and when, is for the rehabilitation department and the therapist to decide +- Benefit: The Cochrane systematic review included 29 trials and 5902 participants, comparing organised inpatient stroke unit care with other forms of care. At the last follow-up (median 1 year) the odds ratio for an adverse outcome (death, disability or the need for institutional care) was 0.77 (95% CI 0.69–0.87, credible range), for death it was 0.76 (95% CI 0.66–0.88), and for death or dependence on others 0.75 (95% CI 0.66–0.85), with moderate quality of evidence. Turned into absolute numbers, for every 100 people who get stroke unit care there are 2 more who survive, 6 more who live at home and 6 more who manage their own lives. The AVERT trial, on the other hand, randomly allocated 2104 patients across 56 acute stroke units in 5 countries, comparing 「常规照护」 with 「常规照护加超早期高强度活动」. At three months a good outcome (modified Rankin Scale 0 to 2) was seen in 480 people (46%) in the very early group and 525 people (50%) in the usual care group, adjusted odds ratio 0.73 (95% CI 0.59–0.90, P=0.004). Deaths were 88 against 72, with no difference shown (odds ratio 1.34, 95% CI 0.93–1.93, P=0.113) +- Evidence grade: A +- Sources: Langhorne P, Ramachandra S; Stroke Unit Trialists' Collaboration (2020). Organised inpatient (stroke unit) care for stroke: network meta-analysis. Cochrane Database of Systematic Reviews, 4:CD000197. ; AVERT Trial Collaboration group (2015). Efficacy and safety of very early mobilisation within 24 h of stroke onset (AVERT): a randomised controlled trial. Lancet, 386(9988):46-55. +- Notes:Disputed: both pieces of evidence come from stroke, and applying them to spinal cord injury, amputation, burns and other causes of disability is an extrapolation. What the AVERT result shows is that 「越早越猛」 does not hold, not that early activity is useless. In the trial the great majority of people in both groups were out of bed within 24 hours; what differed was the frequency and the total amount. When you choose an institution, check whether it has a rehabilitation medicine department and full-time rehabilitation therapists and occupational therapists. Do not judge it by the two characters 「康复」 on the sign. For referral between the grassroots level and the top-tier hospital, and how the deductible is worked out, see item 1 in section 24 (referral step by step). For community-based rehabilitation and assistive devices see item 7 in this section (ask at the county federation of disabled persons) + +### 17. If your hearing is going, get a hearing aid fitted; the reason is to hear, not to prevent dementia + +- Cost: A hearing aid runs from several thousand to several tens of thousands of yuan, and most of it is yours to pay. It has to be fitted, and you go back several times for adjustments +- In plain language: The claim that 「戴助听器能防痴呆」 does not hold up at the moment. A three-year trial of close to a thousand people showed no difference overall. A difference appeared only among those whose risk of cognitive decline was already higher. So the reason to get a hearing aid is to hear now, to follow conversation and to hear cars when you cross the road, not to protect your brain decades from now. +- Benefit: The ACHIEVE trial was carried out in 4 communities in the United States and included 977 people aged 70 to 84 with untreated hearing loss and no obvious cognitive impairment. They were randomised into two groups, one receiving a hearing intervention (hearing counselling and the fitting of hearing aids) and one receiving health education as a control, with follow-up every 6 months. The primary endpoint was the change in overall cognitive score over 3 years. In the main analysis pooling the two source populations there was no significant difference between the groups: −0.200 (95% CI −0.256 to −0.144) in the hearing intervention group, −0.202 (95% CI −0.258 to −0.145) in the control group, a difference of 0.002 (95% CI −0.077 to 0.081), P=0.96. A pre-specified sensitivity analysis showed that the effect of the intervention differed between the two source populations (interaction P=0.010). The participants who came from that long-term cardiovascular health study were older, had more risk factors for cognitive decline and had lower baseline cognitive scores. The authors' conclusion is that a hearing intervention may reduce the 3-year change in cognition among people at higher risk of decline, and does not among those at lower risk. The trial did not report serious adverse events attributed to the study +- Evidence grade: A +- Sources: Lin FR, Pike JR, Albert MS, et al.; ACHIEVE Collaborative Research Group (2023). Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): a multicentre, randomised controlled trial. Lancet, 402(10404):786-797. +- Notes:Disputed: the claim you often see online, that hearing aids cut cognitive decline by nearly half, comes from a subgroup inside this trial. It is not the main result, and the main result is that there was no difference between the two groups. 「配助听器」 and 「防痴呆」 have to be judged separately: the first is worth doing, the second should not be used as the reason. A candidate with a hearing disability sitting the national college entrance examination can be exempted from the foreign language listening test; see item 13 in this section (reasonable accommodation). A person with a hearing loss must wear a hearing device while driving; see item 15 in this section (C5) +### 18. To have an adult declared 「管不了自己的事」 you apply to the court; the family's word does not count + +- Cost: An application to the court carries a court fee, usually tens of yuan up to a hundred yuan. The assessment fee is separate, usually several thousand yuan, and the applicant advances it. The whole thing takes several months. +- In plain language: Whether an adult is 「无民事行为能力」 or 「限制民事行为能力」 is for the court alone to determine. The family, the employer and the residents' committee cannot decide it themselves. An interested party or a relevant organisation has to apply to the court. The other way round, if an adult's intelligence or mental health recovers, an application can be made for a determination that he has regained capacity. This procedure is a restriction and a protection at the same time. +- Benefit: Article twenty-one of the Civil Code《民法典》says: 「不能辨认自己行为的成年人为无民事行为能力人，由其法定代理人代理实施民事法律行为。」 Article twenty-two says: 「不能完全辨认自己行为的成年人为限制民事行为能力人，实施民事法律行为由其法定代理人代理或者经其法定代理人同意、追认；但是，可以独立实施纯获利益的民事法律行为或者与其智力、精神健康状况相适应的民事法律行为。」 Article twenty-four says: 「不能辨认或者不能完全辨认自己行为的成年人，其利害关系人或者有关组织，可以向人民法院申请认定该成年人为无民事行为能力人或者限制民事行为能力人。」 The same article also provides that once a determination has been made, on application by the person himself, by an interested party or by a relevant organisation, the people's court may, according to the state of his recovery in intelligence and mental health, determine that he has regained capacity for civil conduct as a person with limited capacity or as a person with full capacity. Relevant organisations include: residents' committees, villagers' committees, schools, medical institutions, women's federations, disabled persons' federations, lawfully established organisations for the elderly, and the civil affairs departments. Article twenty-three says that the guardian of a person with no capacity for civil conduct or with limited capacity is his legal representative (nationwide, in force from 2021-1-1) +- Evidence grade: A +- Sources: Civil Code of the People's Republic of China (Articles twenty-one, twenty-two, twenty-three and twenty-four)《中华人民共和国民法典（第二十一条、第二十二条、第二十三条、第二十四条）》. +- Notes: The application goes through the court's special procedure, which is faster than an ordinary civil case but still takes several months. You have to submit a diagnosis from a hospital, and in most cases the court will also commission a judicial assessment. The assessment fee is usually several thousand yuan, advanced by the applicant. Signing for someone, or disposing of his property, without going through this procedure may later be found invalid, and a flat that has been sold may have to be returned. One more point to keep in mind: limited capacity for civil conduct does not mean nothing can be done. Things that match his intelligence and mental condition, and things that bring him pure benefit, he can do himself. Do not act for him across the board. For how the guardian is determined see item 19 in this section (guardian) + +### 19. A guardian for an adult is determined in the order set by law; if the person under guardianship injures someone else, the guardian pays + +- Cost: No money. Where there is a dispute you go to the residents' committee, the villagers' committee, the civil affairs department or the court +- In plain language: When an adult loses capacity for civil conduct, the guardian follows in order: the spouse, parents and children, other close relatives, and last any other person or organisation willing to take the role. Where they dispute it, the residents' committee, the villagers' committee or the civil affairs department designates the guardian, and anyone who objects can go to the court. While the person is still of sound mind he can nominate in writing who his future guardian will be. If the person under guardianship injures someone else, the guardian pays. +- Benefit: Article twenty-eight of the Civil Code《民法典》says that an adult with no capacity for civil conduct or with limited capacity shall have as guardian, in order, one of the following who has the capacity to be a guardian: 「（一）配偶；（二）父母、子女；（三）其他近亲属；（四）其他愿意担任监护人的个人或者组织，但是须经被监护人住所地的居民委员会、村民委员会或者民政部门同意。」 Article thirty-one says that where there is a dispute over the determination of the guardian, the residents' committee, the villagers' committee or the civil affairs department of the place where the person under guardianship resides designates the guardian, and anyone who objects to the designation may apply to the people's court for a designation, or apply to the court directly. In making the designation it 「应当尊重被监护人的真实意愿，按照最有利于被监护人的原则」. Where the person's rights and interests were left unprotected before the designation, the residents' committee, the villagers' committee, an organisation provided for by law, or the civil affairs department acts as temporary guardian. Article thirty-three says: 「具有完全民事行为能力的成年人，可以与其近亲属、其他愿意担任监护人的个人或者组织事先协商，以书面形式确定自己的监护人」. Article one thousand one hundred and eighty-eight says: 「无民事行为能力人、限制民事行为能力人造成他人损害的，由监护人承担侵权责任。监护人尽到监护职责的，可以减轻其侵权责任。」 Where the person has property, the compensation is paid out of his own property first and the guardian pays what is left. Article one thousand one hundred and eighty-nine says that where a guardian entrusts the duties of guardianship to another person, the guardian still bears tort liability, and the person entrusted bears the corresponding liability where he is at fault (nationwide, in force from 2021-1-1) +- Evidence grade: A +- Sources: Civil Code of the People's Republic of China (Articles twenty-eight, thirty-one, thirty-three, one thousand one hundred and eighty-eight and one thousand one hundred and eighty-nine)《中华人民共和国民法典（第二十八条、第三十一条、第三十三条、第一千一百八十八条、第一千一百八十九条）》. +- Notes: Article thirty-three is voluntary guardianship. Fixing the person in writing while he is still of sound mind saves most of the disputes that follow; for how to write it see item 1 in section 17 (voluntary guardianship). Be careful with the compensation part: handing the person over to a care home or a relative to look after him does not take the responsibility off the guardian, and the person entrusted answers for his own fault only. Guardian liability insurance for the person under guardianship is therefore worth having. Article seventy-eight of the Mental Health Law《精神卫生法》also says that a guardian who abandons a patient, or who fails in his duties and causes harm, bears compensation liability under the law + +### 20. Refused a job, kept off the permanent staff or underpaid because of a disability: you can sue, but work out the cost of the process first + +- Cost: Legal fees run from several thousand to several tens of thousands of yuan, and you pay them yourself. An ordinary procedure at first instance takes 6 months or more, and can be extended. You have to preserve the evidence yourself. +- In plain language: The law forbids discrimination against disabled people in recruitment, confirmation of permanent status, promotion, grading of professional title, pay and social insurance. You can complain to the competent authority, and you can sue in the court. But if you win the case, the other side does not pay your legal fees. So complain first, apply for legal aid at the same time, and only then think about going to court. +- Benefit: Article three of the Law on the Protection of Persons with Disabilities《残疾人保障法》says: 「禁止基于残疾的歧视。禁止侮辱、侵害残疾人。」 Article thirty-eight says: 「在职工的招用、转正、晋级、职称评定、劳动报酬、生活福利、休息休假、社会保险等方面，不得歧视残疾人。」 The same article also provides that the unit where a disabled employee works shall, according to the characteristics of the disabled employee, provide suitable working conditions and labour protection, and shall, as actually necessary, adapt the workplace, the equipment and the living facilities. Article sixty-four says: 「违反本法规定，在职工的招用等方面歧视残疾人的，由有关主管部门责令改正；残疾人劳动者可以依法向人民法院提起诉讼。」 Article four of the Regulations on the Employment of Persons with Disabilities《残疾人就业条例》says: 「禁止在就业中歧视残疾人。」 Article thirteen says that an employer shall not discriminate against disabled employees in promotion, advancement, grading of professional title, pay, social insurance, living benefits and the like. Article forty-two of the Legal Aid Law《法律援助法》says that an applicant who has documents proving that he belongs to a specific group such as 「无固定生活来源的未成年人、老年人、残疾人等特定群体」 is exempted from any check on whether he is financially in difficulty. Article forty-five says that where a legal aid institution provides legal aid services to elderly people or to persons with disabilities, it shall provide accessible facilities, equipment and services as the circumstances require (nationwide) +- Evidence grade: A +- Sources: Law of the People's Republic of China on the Protection of Persons with Disabilities (Articles three, thirty-eight and sixty-four)《中华人民共和国残疾人保障法（第三条、第三十八条、第六十四条）》. ; State Council (国务院) (2007). Regulations on the Employment of Persons with Disabilities (Articles four and thirteen)《残疾人就业条例（第四条、第十三条）》. ; Legal Aid Law of the People's Republic of China (Articles forty-two and forty-five)《中华人民共和国法律援助法（第四十二条、第四十五条）》. +- Notes: The cost of the process has to be worked out first. An ordinary procedure at first instance takes 6 months or more and can be extended; a simplified procedure takes 3 months. Legal fees are not among the litigation costs listed in Article six of the Regulations on the Payment of Litigation Costs, and the 「败诉方负担」 in Article twenty-nine does not include legal fees. So take the two cheap routes first: complain to the human resources and social security department or to the disabled persons' federation, which must order the discrimination stopped; and apply for legal aid at the same time, because Article forty-two exempts disabled people from the check on financial hardship. Collecting evidence is the hardest part. When you are turned down at an interview the other side seldom leaves a written reason, so keep a recording and keep the messages. Once you are employed, the employment dispute goes to labour arbitration; for keeping evidence and the time limit see item 8 in section 19 (keep the evidence before you leave). The person who benefits from this entry is you yourself \ No newline at end of file diff --git "a/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" new file mode 100644 index 00000000..e5fea729 --- /dev/null +++ "b/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -0,0 +1,65 @@ +[← 回总目录](../README.md) + +# 18. 养孩子划不划算 + +这一节算的是钱和时间两笔账。和第 10 节算结婚账一样，这里只把账拆开给你看，不替你下结论。已经怀孕、要照流程一步步办事的，看第 27 节。 + +[← Về mục lục](../README.md) + +# 18. Nuôi con có đáng không + +Mục này tính hai cuốn sổ: tiền và thời gian. Giống phần 10 tính chi phí cưới, ở đây chỉ bóc tách từng khoản ra cho bạn xem, không thay bạn kết luận. Đã mang thai và cần đi theo từng bước thủ tục thì xem phần 27. + +### 1. Hãy cộng vào trước các khoản được nhận: trợ cấp nuôi con của Nhà nước mỗi đứa trẻ mỗi năm 3.600 tệ, trả đến khi đủ 3 tuổi + +- Chi phí: Không tốn tiền. Tới nơi có hộ khẩu của đứa trẻ để xin, chạy một lượt là xong +- Nói thẳng: Đứa trẻ sinh ra từ sau ngày 1 tháng 1 năm 2025 và chưa đủ 3 tuổi thì nhận được khoản tiền này. Một người trong hai người làm cha mẹ tới nơi có hộ khẩu của đứa trẻ để xin. Mỗi đứa trẻ mỗi năm 3.600 tệ, nhận liên tục đến khi đủ 3 tuổi. Khoản tiền này không phải chịu thuế thu nhập cá nhân. Khi xin các khoản trợ cấp như trợ cấp người có thu nhập thấp thì nó cũng không tính vào thu nhập của gia đình bạn +- Lợi ích: Phương án do Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Hội đồng Nhà nước ban hành quy định: trẻ sơ sinh và trẻ nhỏ sinh từ ngày 1 tháng 1 năm 2025 và dưới 3 tuổi thì chi trả theo 「mỗi đứa trẻ mỗi năm 3.600 tệ」. Một người trong hai người làm cha mẹ, hoặc người giám hộ khác của đứa trẻ đi xin. Khoản trợ cấp này 「không phải chịu thuế thu nhập cá nhân」. Khi xét các khoản trợ cấp như trợ cấp người có thu nhập thấp thì nó 「không tính vào thu nhập của gia đình hoặc cá nhân」 +- Mức bằng chứng: A +- Nguồn: Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước (2025). 育儿补贴制度实施方案. +- Ghi chú: Đây là mức sàn thống nhất toàn quốc. Có tỉnh, thành phố có thêm trợ cấp địa phương riêng, có thể nhận cộng với khoản này. Ba năm cộng lại 10.800 tệ, trong chi phí nuôi con chỉ là một phần rất nhỏ, đừng lấy nó làm căn cứ để quyết định có sinh hay không + +### 2. Nghỉ thai sản 98 ngày, trợ cấp thai sản do quỹ bảo hiểm thai sản chi trả theo lương bình quân hằng tháng của công nhân viên năm trước của đơn vị + +- Chi phí: Không tốn tiền. Đơn vị sử dụng lao động lo khai báo, bạn không phải chạy đi đâu +- Nói thẳng: Khi sinh con, ít nhất được nghỉ thai sản 98 ngày, trong đó trước ngày sinh có thể nghỉ trước 15 ngày. Sinh khó thì cộng thêm 15 ngày. Sinh thêm đứa con nữa thì cộng thêm 15 ngày. Khoản tiền chi trả trong thời gian đó gọi là trợ cấp thai sản, do quỹ bảo hiểm thai sản chi trả. Đơn vị không đóng bảo hiểm thai sản cho bạn thì do đơn vị tự chi trả +- Lợi ích: Số ngày nghỉ thai sản do Quốc vụ viện quy định. 「Nữ công nhân viên sinh con được nghỉ thai sản 98 ngày, trong đó trước ngày sinh có thể nghỉ 15 ngày; sinh khó thì cộng thêm 15 ngày nghỉ thai sản; sinh đa thai, mỗi lần sinh thêm 1 đứa trẻ thì cộng thêm 15 ngày nghỉ thai sản.」 Khoản tiền nhận trong thời gian nghỉ thai sản gọi là trợ cấp thai sản. Người đã tham gia bảo hiểm thai sản thì 「chi trả theo mức lương bình quân hằng tháng của công nhân viên năm trước của đơn vị sử dụng lao động, do quỹ bảo hiểm thai sản chi trả」. Người không tham gia bảo hiểm thì đơn vị sử dụng lao động chi trả cho bạn theo mức lương trước khi nghỉ thai sản +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2012). Quy định đặc biệt về bảo vệ lao động nữ công nhân viên (Quốc vụ lệnh số 619, Điều bảy, Điều tám). +- Ghi chú: 98 ngày là mức tối thiểu do Nhà nước đặt ra. Điều lệ dân số và kế hoạch hoá của các tỉnh thường còn cộng thêm hơn 60 ngày nghỉ thưởng cho việc sinh con, tính cả thường là từ 158 ngày. Lương trong thời gian nghỉ thưởng do bên nào chi trả thì quy định khác nhau ở mỗi tỉnh + +### 3. Biết điều này: không được giảm lương hoặc sa thải vì mang thai, vì sinh con, vì cho con bú + +- Chi phí: Không tốn tiền +- Nói thẳng: Trong thời gian mang thai, sinh con và cho con bú, nếu đơn vị giảm lương của bạn hoặc sa thải bạn thì đều là vi phạm pháp, bị phạt mỗi người từ 1.000 đến 5.000 tệ. Trẻ chưa đầy một tuổi thì mỗi ngày bạn có 1 giờ cho con bú, đơn vị cũng không được bố trí bạn làm thêm giờ hay ca đêm. Cách đòi bồi thường xem phần 19 +- Lợi ích: Quốc vụ viện quy định: 「Đơn vị sử dụng lao động không được vì nữ công nhân viên mang thai, sinh con, cho con bú mà giảm lương, sa thải, hoặc hủy hợp đồng lao động hay hợp đồng thuê bao với nữ công nhân viên.」 Trong ba thời gian đó đơn vị không được giảm tiền của bạn, cũng không được sa thải bạn hay hủy hợp đồng. Nữ công nhân viên đang cho con dưới 1 tuổi bú thì mỗi ngày có 1 giờ cho con bú. Đơn vị cũng không được kéo dài giờ làm của bạn, không được bố trí ca đêm. Đơn vị vi phạm những điều trên thì bị phạt mỗi người từ 1.000 đến 5.000 tệ +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2012). Quy định đặc biệt về bảo vệ lao động nữ công nhân viên (Quốc vụ lệnh số 619, Điều năm, Điều chín, Điều mười ba). +- Ghi chú: Nếu thật sự bị sa thải vi phạm pháp thì đường đòi bồi thường xem phần 19. Trọng tâm khi lấy chứng cứ là thông báo bằng văn bản về việc chuyển vị trí, cắt giảm lương và lịch sử trò chuyện, phải giữ kỹ hai thứ đó + +### 4. Cuốn sổ thời gian tính theo 「thời gian của ai bị chiếm mất」, không tính theo 「có vất vả không」 + +- Chi phí: Không tốn tiền, cái mất là thời gian. Trước khi trẻ đủ 3 tuổi thì về cơ bản thời gian bị chiếm hết cả ngày. Chỗ khó là chuyện này lặp lại ngày này sang ngày khác, không thể cắn răng chịu mấy ngày rồi qua +- Nói thẳng: Quy đổi số giờ trông trẻ mỗi ngày và số lần bị đánh thức giữa đêm thành thời gian. Rồi nhân với mức lương theo giờ của người làm việc đó. Con số tính ra thường lớn hơn nhiều so với tiền sữa công thức và tã. Vì thế nên chia việc từ trước khi sinh +- Lợi ích: Quy đổi thời gian trông trẻ mỗi ngày và số lần bị đánh thức giữa đêm thành giờ. Rồi nhân với mức lương theo giờ của người làm việc đó. Con số thu được thường lớn hơn nhiều so với tiền sữa công thức và tã. Việc trông trẻ, nấu ăn, dậy đêm đi vệ sinh dù không ai trả tiền vẫn được ghi vào chi phí kiểu đó. Cách tính giống bộ tính việc nhà ở phần 10. Chia việc cho xong rồi hãy quyết định có sinh con hay không +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; thước đo xem phần 10 +- Ghi chú: Không cần tính số chính xác lắm. Cái hữu ích là hai người cùng nhìn một bảng để bàn chia việc trước khi sinh, khỏi sinh xong rồi cãi nhau + +### 5. Cuốn sổ tiền chia làm ba đoạn để tính: 0 đến 3 tuổi, giáo dục bắt buộc, sau giáo dục bắt buộc + +- Chi phí: Bao nhiêu tiền thì không có con số cố định, khác nhau theo thành phố và theo cách nuôi, chênh nhau rất xa. Chỗ khó là phải bình tĩnh tách ba đoạn sổ sách ra tính riêng, không thể đoán bừa một con số bằng cảm tính +- Nói thẳng: Tiền ở giai đoạn 0 đến 3 tuổi chủ yếu chi vào cách nuôi: tự nuôi, nhờ ông bà, hay thuê người nuôi. Từ tiểu học đến trung học thì chủ yếu chi vào nhà ở và ngoại khóa. Sau đó thì chủ yếu chi vào con đường thi cử. Cách chi của ba đoạn hoàn toàn khác nhau, gộp vào nhau chỉ cho ra một tổng số đáng sợ +- Lợi ích: Khoản lớn nhất trong giai đoạn 0 đến 3 tuổi là cách chăm sóc: tự nuôi, nhờ ông bà hay thuê người nuôi. Khoản lớn nhất trong giai đoạn giáo dục bắt buộc (tiểu học đến trung học) là nhà ở và chi phí ngoại khóa. Sau đó thì khoản lớn nhất là con đường học vấn. Ba đoạn đó chi tiền vào hai chỗ hoàn toàn khác nhau, gộp lại tính thì chỉ cho ra một tổng số đáng sợ +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp +- Ghi chú: Trước hết cộng vào các khoản chắc chắn nhận được như trợ cấp nuôi con, trợ cấp thai sản, bảo hiểm y tế hoàn trả, rồi xem còn thiếu bao nhiêu. Đừng dựa vào những con số 「nuôi một đứa con tốn bao nhiêu triệu」 lan truyền trên mạng, loại số đó không rõ thước đo, bạn không biết nó tính những gì và bỏ sót những gì + +### 6. Sinh vì người lớn, sinh vì hôn nhân, sinh vì lúc về già, mỗi kiểu ghi một cuốn sổ riêng + +- Chi phí: Không tốn tiền, cái mất là một cuộc nói chuyện nghiêm túc. Chỗ khó là phải nói thẳng ra, không vòng vo +- Nói thẳng: Người khác muốn bạn sinh con thì có thể tính là một yếu tố cân nhắc của bạn. Nhưng thời gian và tiền thì ghi vào sổ của chính bạn. 「Nuôi con khi về già」 thì phải tính riêng một cuốn, vì nó đánh cược vào năng lực tài chính và ý chí của một người khác hai mươi năm sau. Hãy viết cả hậu quả của việc không sinh ra, rồi đối chiếu hai bên +- Lợi ích: Cách nghĩ giống phần 10. Kỳ vọng của người khác có thể tính là một yếu tố cân nhắc của bạn. Nhưng chi phí thì ghi vào sổ của chính bạn, không ghi vào sổ của họ. 「Nuôi con khi về già」 càng phải tính riêng một cuốn, vì nó giả định năng lực tài chính và ý chí của một người khác hai mươi năm sau +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; xem phân tích cùng loại ở phần 10 +- Ghi chú: Hãy viết cả 「hậu quả của việc không sinh」 rồi đặt cạnh 「hậu quả của việc sinh」 để đối chiếu From 18e336ce2c4f95336ae93273954d762a7aa8073f Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 12:35:22 +0700 Subject: [PATCH 14/25] =?UTF-8?q?=E8=8A=82=E6=A0=87=E9=A2=98=E5=92=8C?= =?UTF-8?q?=E5=AF=BC=E8=AF=BB=E4=B9=9F=E8=A6=81=E7=BF=BB=EF=BC=9Ajoin=20?= =?UTF-8?q?=E4=B8=8D=E5=86=8D=E6=8B=BF=E4=B8=AD=E6=96=87=E5=8E=9F=E6=96=87?= =?UTF-8?q?=E7=9B=96=E5=9C=A8=E8=AF=91=E6=96=87=E5=89=8D=E9=9D=A2=EF=BC=8C?= =?UTF-8?q?=E6=A3=80=E6=9F=A5=E5=8A=A0=E4=B8=8A=E8=BF=99=E4=B8=A4=E6=9D=A1?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 节的条目一直是逐条校对的，节首那几行（回目录、节标题、导读）却没人管—— join.mjs 拿中文原文那一份无条件拼在第 0 块前面（tools/i18n/join.mjs:92）。 结果：en 34 节里 30 节的标题是翻好的，却被当成「多出来的」删掉，中文那份留着； 剩下的连节名带导读都是中文。vi 更糟，19 节的节首根本没翻。 页面上看得见：英文站每个小节标题是中文的，有几节渲染两遍（中文在前、英文在后）， 侧栏目录也是中文节名。 改三处： · join.mjs 拼的时候不再碰译文块的内容。第 0 块自带节首就用它自带的； 真的没有（少数块是这样切的）才补，补的是中文那份并且报出来。 · structure.mjs 加 compareSection：节号、节标题跟原文一字不差就是没翻， 导读里剔掉照抄的法条名书名之后还剩汉字就是整段没翻。 · check.mjs 和 join.mjs 都调它，CI 跑的就是 check.mjs。 顺带把 01-04 的节名对齐 en/README.md 的目录（全站同一个说法）， 并把这四节的导读翻好。英文 34/34 节现在节首节身都干净。 --- ...15\350\246\201\346\227\251\346\255\273.md" | 4 +- ...01\346\205\242\346\205\242\346\255\273.md" | 4 +- ...52\350\264\271\347\262\276\345\212\233.md" | 4 +- ...52\350\264\271\346\227\266\351\227\264.md" | 4 +- ...01\346\265\252\350\264\271\351\222\261.md" | 6 - ...15\351\235\242\346\270\205\345\215\225.md" | 6 - ...31\346\200\216\344\271\210\346\264\273.md" | 6 - ...61\346\220\255\350\277\233\345\216\273.md" | 6 - ...25\345\276\213\347\272\242\347\272\277.md" | 6 - ...22\344\270\215\345\210\222\347\256\227.md" | 6 - ...51\347\232\204\347\272\242\347\272\277.md" | 6 - ...16\345\201\232\347\224\237\346\204\217.md" | 6 - ...47\346\200\245\346\203\205\345\206\265.md" | 8 - ...41\346\201\257\345\256\211\345\205\250.md" | 6 - ...77\344\270\216\344\271\260\346\210\277.md" | 6 - ...16\346\200\216\344\271\210\346\264\273.md" | 6 - ...14\346\234\211\350\200\201\344\272\272.md" | 6 - ...22\344\270\215\345\210\222\347\256\227.md" | 6 - ...14\345\222\214\345\267\245\344\274\244.md" | 6 - ...20\346\200\216\344\271\210\345\270\246.md" | 6 - ...03\345\244\226\345\256\211\345\205\250.md" | 6 - ...16\344\271\210\346\224\276\346\235\276.md" | 6 - ...00\350\203\275\345\210\222\347\256\227.md" | 6 - "en/book/24-\347\234\213\347\227\205.md" | 8 - ...01\345\212\236\344\273\200\344\271\210.md" | 8 - ...31\346\210\226\345\271\263\345\217\260.md" | 6 - ...25\345\222\214\347\224\237\344\272\247.md" | 8 - ...53\344\275\223\346\220\236\345\235\217.md" | 8 - ...23\345\207\273\344\271\213\345\220\216.md" | 8 - ...16\347\232\204\345\255\251\345\255\220.md" | 8 - ...52\345\207\240\346\235\241\350\267\257.md" | 10 - ...72\345\233\275\347\225\231\345\255\246.md" | 10 - ...16\346\200\216\344\271\210\346\264\273.md" | 18 +- ...53\345\220\203\345\207\272\344\272\213.md" | 12 -- tools/i18n/check.mjs | 2 + tools/i18n/join.mjs | 38 ++-- tools/i18n/structure.mjs | 33 +++ ...31\346\200\216\344\271\210\346\264\273.md" | 191 +++++++++++++++++ ...25\345\276\213\347\272\242\347\272\277.md" | 6 - ...51\347\232\204\347\272\242\347\272\277.md" | 161 +++++++++++++++ ...47\346\200\245\346\203\205\345\206\265.md" | 8 - ...41\346\201\257\345\256\211\345\205\250.md" | 6 - ...14\346\234\211\350\200\201\344\272\272.md" | 6 - ...22\344\270\215\345\210\222\347\256\227.md" | 6 - ...16\347\232\204\345\255\251\345\255\220.md" | 141 +++++++++++++ ...16\346\200\216\344\271\210\346\264\273.md" | 193 ++++++++++++++++++ ...53\345\220\203\345\207\272\344\272\213.md" | 92 +++++++++ 47 files changed, 842 insertions(+), 277 deletions(-) create mode 100644 "vi/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" create mode 100644 "vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" create mode 100644 "vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" create mode 100644 "vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" create mode 100644 "vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" diff --git "a/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" "b/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" index ddd2c44a..ec5783ea 100644 --- "a/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" +++ "b/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 1. 不要早死 +# 1. Do not die young -这一节讲三件事：车祸、火灾、中毒这类意外和伤害死亡；少数几种证据最硬的疫苗和筛查；身体出了什么信号就该去查一查。收益那一栏的数字，来自随机分组的试验、跟踪一大批人很多年的研究，或者官方统计。成本那一栏是作者按市面价格的粗略估计，只用来互相比较，不要拿去引用。 +This section covers three things: accidental and violent deaths such as car crashes, fires and poisoning; the handful of vaccines and screenings with the hardest evidence behind them; and which signals from your body mean you should go and get checked. The figures in the benefit column come from trials that randomised people into groups, from studies that followed a large number of people for many years, or from official statistics. The cost column is the author's rough estimate at going-market prices, there for comparing options against each other; do not quote them. ### 1. Wear a seat belt, in the front and in the back diff --git "a/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" "b/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" index f455d5ac..f64a2d4d 100644 --- "a/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" +++ "b/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 2. 不要慢慢死 +# 2. Do not die slowly -本节只收两类慢性风险：一类对总死亡率影响大，一类证据硬。条目按性价比从高到低排。有几条标了「随机试验」，意思是把人随机分成两组来比，这种结果最可信。其余数字都来自只跟踪记录、不做分组的研究。这类研究算出的 HR、RR 等风险比里混着两种干扰：这些人本来就不一样，有人是先病了才这么做。所以这些数字只说明方向和大概多大，不代表照做就能降这么多。 +This section takes in only two kinds of chronic risk: the ones that move overall mortality a long way, and the ones with hard evidence behind them. The items are in order from best value to worst. A few are marked "randomised trial", meaning people were split at random into two groups and compared; that is the most trustworthy kind of result. The remaining figures come from studies that only follow people and record what happens, without splitting them into groups. The risk ratios such as HR and RR that those studies produce are mixed up with two kinds of confounding: the people were not alike to begin with, and in some cases they only started doing the thing after they became ill. So those figures show direction and rough size only; they do not mean that doing this brings your risk down by exactly that much. ### 1. Quit smoking, the earlier the better diff --git "a/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" "b/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" index d4ea82c0..6212fbf3 100644 --- "a/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" +++ "b/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 3. 不要浪费精力 +# 3. Do not waste energy -本节只算精力和时间账。看的是四样东西：你每天能用的注意力、反应快不快、出错多不多、被打断以后要多久才回得来。所有数字都是直接测出来的，来自实验，或者来自跟踪一群人做的记录（队列研究）。这些数字不换算成寿命，也不换算成死亡率。本节多数条目的证据等级是 B。这个领域本来就这样，不代表建议不成立。第 20 条讲的是和警察、医生、柜员打交道时该有的预期，引的是法规，不是实验，所以等级记 C。第 15 条（把悲观念头当症状看）和第 23 条（把「别人要求我完美」当症状看）引的是死亡率与自杀风险的数字，口径和本节其他条目不一样，不放在一起比较。 +This section counts only willpower and time. It looks at four things: how much attention you have available each day, how fast you react, how many mistakes you make, and how long it takes to get back after being interrupted. Every figure is measured directly, either in an experiment or in records following a group of people (a cohort study). These figures are not converted into lifespan and not into mortality either. Most items in this section are graded B. That is how this field is; it does not mean the advice does not hold. Item 20 is about what to expect when dealing with police, doctors and clerks: it cites regulations rather than experiments, so it is graded C. Item 15 (treating "it will definitely get worse" thoughts as a symptom) and item 23 (treating "other people demand perfection from me" as a symptom) quote mortality and suicide-risk figures, a different measure from the rest of this section, so they are not compared with the others. ### 1. Turn off notifications you do not need, and keep the phone out of sight while you work diff --git "a/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" "b/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" index 9b14a7d9..8f886d8b 100644 --- "a/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" +++ "b/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 4. 不要浪费时间 +# 4. Do not waste time -本节只算时间账。看的是每天、每周、每年能省回多少小时，或者一件事按时做完的比例能提高多少。这里不把时间换算成寿命，也不换算成钱。涉及钱的地方只给算法，不给结论。本节多数条目的证据是单独一项研究，或者是时间利用调查的总量统计，所以等级以 B、C 为主。 +This section counts only time. It looks at how many hours you get back per day, per week and per year, or how much the share of things finished on time can go up. Time is not converted into lifespan here, and not into money either. Where money comes up it gives you the method and not the conclusion. Most items in this section rest on a single study, or on aggregate figures from time-use surveys, so the grades are mostly B and C. ### 1. Turn "I mean to do it" into "when, where, and if this comes up then that" diff --git "a/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" "b/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" index 321477e4..279b1c93 100644 --- "a/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" +++ "b/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 5. 不要浪费钱 - -这一节只算钱：能省下多少，每年收益差多少，一笔钱长期平均下来能拿回多少。这些不换算成寿命或时间。涉及投资的条目只讲怎么算账，只用核实过的历史数据，都不构成投资建议。 - [← Back to the contents](../README.md) # 5. Do not waste money diff --git "a/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" "b/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" index 73fff460..93655d1e 100644 --- "a/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" +++ "b/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 6. 反面清单 - -本节收的是「看起来便宜又有效」的东西。翻开原始研究之后，它们的好处接近零、等于零，有的还是负的。排的顺序看两件事：大家在这上面花的钱多不多，支持它的证据硬不硬。越靠前的越值得先停掉。 - [← Back to the contents](../README.md) # 6. The reverse list diff --git "a/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" "b/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" index 719d8d66..24918080 100644 --- "a/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" +++ "b/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 7. 没钱的时候怎么活 - -这一节算的是钱和保障。能领多少钱、能免多少钱、能得到什么服务、要满足什么条件，都写在里面，不折算成寿命。政策每年都在变，各地也不一样，所以金额和门槛都以你所在地当年公布的为准。条目的排序是：不花钱、手续少、见效快的排在前面。要攒长期资格的、各地差别大的排在后面。只凭经验、没有官方文件的放在最后。 - [← Back to the contents](../README.md) # 7. How to get by with no money diff --git "a/en/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" "b/en/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" index 562144f8..f85d7bc9 100644 --- "a/en/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" +++ "b/en/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 8. 别把自己搭进去：法律与财产安全 - -这一节算钱和人身自由，不换算成寿命。讲的是三件事：法律规定你要担什么责任、照着做能少赔多少钱、出了事你在程序上有哪些权利。条款号都和官方原文核对过。排序的办法是：一次出事损失最大、做起来又最省事的放最前面；需要提前安排的放中间；只有经验、没有官方文件撑着的放最后。 - [← Back to the contents](../README.md) # 8. Do not put yourself at risk: law and property safety diff --git "a/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" "b/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" index 13ec789f..a3031767 100644 --- "a/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" +++ "b/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 9. 普通人容易踩的法律红线 - -这一节算两笔账：你的人身自由，和你的钱。下面这些事，普通人常以为「没什么」，实际会被拘留、罚款，重的还会判刑。每条都给出法条里的具体数字，以及官方渠道公布过的案例。本节引的治安管理处罚法是 2025 年修订版，2026 年 1 月 1 日起施行，条款号和旧法不一样。刑法引的是北京市公安交管局转载的「根据刑法修正案（十一）修正」的整合文本，正文里会注明对应的修正案。条目的顺序，按「最容易不小心踩到」加「后果最重」从高到低排。 - [← Back to the contents](../README.md) # 9. Legal lines ordinary people cross diff --git "a/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" index fed4a1a1..28f71aa9 100644 --- "a/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" +++ "b/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 10. 恋爱和结婚划不划算 - -本节把「结婚划不划算」拆成钱、时间、健康、法律风险和关系质量几笔账，分开算。每笔账只用能查到出处的数字，出处限于官方统计，或者荟萃分析（把多项研究的数据汇总起来重算一遍）。这里不做道德判断，也不替你下结论。健康那几笔按死亡风险算，时间那几笔按每天多少分钟算，法律那几笔按法条怎么写算。这三种算法之间不互相换算。拆账方法、填空清单和常见误区见 [docs/结婚划不划算.md](../docs/结婚划不划算.md)。 - [← Back to the contents](../README.md) # 10. Is dating and marriage worth it diff --git "a/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" "b/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" index 1f2d985f..fee69314 100644 --- "a/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" +++ "b/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 11. 程序员和技术人容易踩的红线 - -本节口径是人身自由与金钱，换回的是「别被关」和「别赔钱」。写代码、做技术的人，常把下面这些事当成「只是技术问题」，可它们对应的是刑法里的具体罪名，或者民事上要赔的具体一笔钱。每条都给出法条里的数字，以及最高人民检察院、最高人民法院官网公布的案例。刑法引用的是北京市公安交管局转载的「根据刑法修正案（十一）修正」整合文本。条目顺序按「技术人最容易以为没事 × 后果最重」从高到低排。第 1 条是通用的一条，讲动手前怎么判断，以及出事之后先做什么。 - [← Back to the contents](../README.md) # 11. Lines programmers and tech workers cross diff --git "a/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" "b/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" index 67c5c8c0..edcb4a76 100644 --- "a/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" +++ "b/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 12. 创业与做生意：别把家底赔进去 - -这一节只算钱和法律责任，不折算成寿命。开一家店、办一家公司，哪一步会把你自己的钱、家里的钱、配偶的钱一起赔进去？法律怎么规定，能不能提前隔开？最容易把家底赔进去、又最省事就能避开的条目，放在最前面。所以签字类的条目靠前，要自己动手做的方法靠后。只凭经验写、没有文献支撑的，标 C 级。 - [← Back to the contents](../README.md) # 12. Starting a business: do not lose the family savings diff --git "a/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" "b/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" index 14961e32..9ffd35a4 100644 --- "a/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" +++ "b/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" @@ -1,11 +1,3 @@ -[← 回总目录](../README.md) - -# 13. 紧急情况：先做什么 - -本节讲出了事先做什么、再做什么、什么不要做。急救和野外的条目算的是特定死因和活下来的比例。威胁和骗局的条目算的是钱和人身自由。两类数字不互相换算。条目的顺序按「这事多常见 × 做对做错差多少条命」从高到低排。这些条目首先是给你自己和配偶、直系亲属用的。心脏骤停有 79.2% 发生在家里，你最可能救的就是家里人。对朋友同事次之。对陌生人也不是一点好处没有，但这一档最低，还可能被讹、被卷进案子。所以这类条目把保护自己的动作和免责的条款写在前面。 - -本节只收几分钟内要做的动作。认出骗局、被人拿把柄要钱，处置要按天算，不在这一节。反诈的硬规则和七种高发骗局见第 8 节第 3 条（反诈硬规则）。被人拿隐私照片、裸聊录像要钱见第 8 节第 32 条（对方拿把柄来要钱）。钱已经转出去了，立刻打 110 或 96110 要求止付，见第 8 节第 2 条（止付）。 - [← Back to the contents](../README.md) # 13. Emergencies: what to do first diff --git "a/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" "b/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" index 3b9f7205..1c9fbac5 100644 --- "a/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" +++ "b/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 14. 账号与信息安全 - -口径：金钱与个人信息。账号被别人登进去，丢的先是钱。对方还会用你的账号去骗你通讯录里的人。你的身份也等于被人拿走了。 - [← Back to the contents](../README.md) # 14. Accounts and information security diff --git "a/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" "b/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" index cb5b4ffa..39e5c6d9 100644 --- "a/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" +++ "b/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 15. 租房与买房 - -口径：金钱。租房最容易亏钱的地方是押金和中介。买房最费钱的地方是贷款利息，还有卖方知道、你不知道的那些事。 - [← Back to the contents](../README.md) # 15. Renting and buying a home diff --git "a/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" index 82e9354d..21f1415c 100644 --- "a/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" +++ "b/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 16. 得了慢性病之后怎么活 - -口径：总死亡率与金钱。查出慢性病以后往往还有几十年要过，本节算的是这几十年怎么管。 - [← Back to the contents](../README.md) # 16. Living with a chronic illness diff --git "a/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" "b/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" index 581d3212..e5d1f91c 100644 --- "a/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" +++ "b/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 17. 家里有老人 - -口径：金钱与人身自由。家里有老人，多数麻烦不是钱不够，是没提前把话说清楚、把字签好。最后一条讲长期卧床的压疮，按死亡率算账，不和本节其他条目比较。 - [← Back to the contents](../README.md) # 17. When there are elderly at home diff --git "a/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" index b4f1aa84..c860be89 100644 --- "a/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" +++ "b/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 18. 养孩子划不划算 - -这一节算的是钱和时间两笔账。和第 10 节算结婚账一样，这里只把账拆开给你看，不替你下结论。已经怀孕、要照流程一步步办事的，看第 27 节。 - [← Back to the contents](../README.md) # 18. Is raising children worth it diff --git "a/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" "b/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" index 34d12df2..f6e187c1 100644 --- "a/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" +++ "b/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 19. 在职、离职和工伤 - -这一节算的是钱：钱怎么算、字怎么签、期限怎么算。前三条讲你还在岗时就该拿到手的钱，中间六条讲离职，最后七条讲职业病和工伤。工伤的钱比被裁大一个量级，期限也更硬。职业病造成的损伤换不回来。 - [← Back to the contents](../README.md) # 19. Working, leaving, and work injuries diff --git "a/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" "b/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" index 8b9b9f01..dc112540 100644 --- "a/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" +++ "b/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 20. 刚出生的孩子怎么带 - -口径：婴儿死亡率与金钱。这一节只收关系到孩子能不能活下来、或者你要花多少钱的事，而且只收差别足够大、又有硬证据的。怎么哄、怎么带这些育儿细节，本书不写。出生之前和出院当天要办的事，见第 27 节。 - [← Back to the contents](../README.md) # 20. Caring for a newborn diff --git "a/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" "b/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" index 74aa29fa..00811c4b 100644 --- "a/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" +++ "b/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 21. 出国、旅行与境外安全 - -这一节算的是钱和人身自由，有几条还关系到人身安全。到了国外，管你的是所在国的法律，不是中国的。中国使领馆能帮你到哪一步，有明确的界线。 - [← Back to the contents](../README.md) # 21. Travel and safety abroad diff --git "a/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" "b/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" index 1dbeb008..4e966744 100644 --- "a/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" +++ "b/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 22. 怎么放松：娱乐场所和减压 - -前半节讲娱乐场所里哪些钱花得冤、安全出口在哪，口径是金钱和人身自由。后半节讲怎么减压，口径是精力和总死亡率。两半的数字不放在一起换算。第 3 节已经写过情绪低落时先做哪几件事，这里给更具体的做法。 - [← Back to the contents](../README.md) # 22. Relaxing: venues and stress relief diff --git "a/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" "b/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" index fbffffc3..cf9aa9ba 100644 --- "a/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" +++ "b/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 23. 学什么技能划算 - -这一节算的是钱和时间，第 2 条另外算死亡率。前六条回答一个问题：还要不要接着读书。第 1 条先说法律划掉了哪一段选项，第 2 条算读书和寿命的关系，第 3 条看全国人的学历都在哪一档。第 4 条讲读不起的时候有哪些资助，第 5 条讲没考上普高还有什么路，第 6 条把「读书还是打工」算成一道题，给一个自己动手算的方法。后七条回答另一个问题：决定学点东西之后怎么挑。中国按工种分的工资统计，从公开的官方渠道很难逐项核实。所以这里只写查得到出处的几件事：多读一年书工资大概多多少（也叫教育回报率）、证书合不合法怎么验、培训补贴从哪里来、挑技能时看哪个维度更不容易被顶掉。具体某个工种一个月挣多少，本书不给数字。第 14 到 19 条回答第三个问题：定下来学什么之后，怎么学才省时间。这六条算的是时间，不和钱做换算。第 20 到 23 条回答第四个问题：已经在做专业技术工作了，职称从哪里报、怎么考、别踩什么坑、评上值多少。 - [← Back to the contents](../README.md) # 23. Which skills are worth learning diff --git "a/en/book/24-\347\234\213\347\227\205.md" "b/en/book/24-\347\234\213\347\227\205.md" index 4f8cdfa2..ebfae8cd 100644 --- "a/en/book/24-\347\234\213\347\227\205.md" +++ "b/en/book/24-\347\234\213\347\227\205.md" @@ -1,11 +1,3 @@ -[← 回总目录](../README.md) - -# 24. 看病：怎么少花钱少走弯路 - -口径：金钱与时间。这一节讲看病本身怎么走：先去哪家医院，怎么往上转，钱能少花在哪。另外讲急诊怎么排队，重伤治完之后要做哪些鉴定、办哪些证。能领什么待遇见第 7 节。慢性病长期怎么管见第 16 节。急症现场先做什么见第 13 节。 - -本节主要依据国务院办公厅 2026 年 4 月印发的《关于加快建设分级诊疗体系的若干措施》。分级诊疗就是按病情轻重决定先去哪一级医院。这份文件把「去基层」和「省钱」直接挂上了钩。但各地落地进度不同，正文写的是国家层面的规则。具体到你所在的统筹地区，要以当地医保和卫健部门的说法为准。统筹地区就是你的医保参保地，通常是一个市。 - [← Back to the contents](../README.md) # 24. Medical care: spending less and going fewer rounds diff --git "a/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" "b/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" index 01e8670d..b3f137f9 100644 --- "a/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" +++ "b/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" @@ -1,11 +1,3 @@ -[← 回总目录](../README.md) - -# 25. 人走了以后要办什么 - -口径：金钱。本节只讲手续怎么办、钱怎么拿，包括人刚走时先做什么、哪些费用有清单可以对。还有哪些账户里的钱能取回来，哪些事近亲属有权出面办。前五条按事情发生的先后排，依次是现场、死亡证明、遗体和火化、对死因有异议怎么办、销户。后面几条按性价比排。老人生前该做的安排，比如意定监护、遗嘱、账户，见第 17 节。因工死亡的三笔待遇见第 19 节。 - -殡葬这部分依据新版《殡葬管理条例》，2026 年 3 月 30 日起施行。养老保险的丧葬补助金和遗属抚恤金给多少，由人力资源社会保障部门另外规定。这次没拿到能逐字核对的原文，所以正文不写具体金额。 - [← Back to the contents](../README.md) # 25. What to do after someone dies diff --git "a/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" "b/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" index bb162362..ec9ff15a 100644 --- "a/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" +++ "b/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 26. 做一个网站或平台：资质、备案和服务器 - -你想做一个网站或者 App。本节回答四件事：它在法律上算哪种生意，要办哪张证，服务器放在哪里，用户在上面发的东西和收的钱谁来负责。口径是人身自由与金钱。条目的顺序是：先写会坐牢的，再写要先办证的，然后是平台每天都要做的事，最后一条是选服务器的经验，那条只能给判断、给不出数字。你是受雇的技术人员，红线见第 11 节。公司怎么注册、怎么报税，见第 12 节，本节不重复。各类证的对照表和选服务器的决策表，见 [docs/做平台要办哪些证.md](../docs/做平台要办哪些证.md)。 - [← Back to the contents](../README.md) # 26. Running a website or platform: licences, filings, servers diff --git "a/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" "b/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" index 9f73acc5..7e5dff49 100644 --- "a/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" +++ "b/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" @@ -1,11 +1,3 @@ -[← 回总目录](../README.md) - -# 27. 怀孕和生产：从发现怀孕到出院办证 - -这一节前半段算的是孕妇和胎儿的死亡风险。后半段算的是钱和跑腿时间。两种账不放在一起换算。内容只讲「什么时候做什么」。孕期营养、胎教这类没有硬证据、或者作用很小的事，这里不写。生育津贴、产假和养孩子的钱账见第 18 节，孩子出生之后怎么带见第 20 节，急症见第 13 节。 - -世界卫生组织给的数字：2023 年，全球约有 26 万名女性死于怀孕和生产。其中约 75% 死于这五类原因：产后大出血、感染、妊娠期高血压疾病（怀孕后血压升高的一类病，包括子痫前期和子痫）、分娩时的并发症、不安全的人工流产。世界卫生组织同时说，「大多数孕产妇死亡是可以预防的，因为预防和处理并发症的医疗手段都是已知的」。这一节里作用最大的几条，就是把这几类问题提前拦下来。 - [← Back to the contents](../README.md) # 27. Pregnancy and childbirth: from test to paperwork diff --git "a/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" "b/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" index c5288c92..86af9faa 100644 --- "a/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" +++ "b/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" @@ -1,11 +1,3 @@ -[← 回总目录](../README.md) - -# 28. 别为了外形把身体搞坏 - -本节说的是「为了改变外形而对身体动手」的几种做法：饿肚子、吃药、打针、开刀。这几件事，决定多半是和别人比较之后做的，代价却全落在自己身上，而且相当一部分改不回来。 - -本节不评价任何人想把自己变成什么样子，那是各人自己的事。同一个目标，可以找「有处方、有资质、有复查」的，也可以「网购、美容院、自己加量」，本节只比这两种做法的风险差多少。算账主要看身体上实际出了什么事，比如失明、住院、死亡。医美那两条还涉及非法行医的刑事责任。 - [← Back to the contents](../README.md) # 28. Do not wreck your body for the sake of looks diff --git "a/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" "b/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" index d74082ab..46d5a6a2 100644 --- "a/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" +++ "b/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" @@ -1,11 +1,3 @@ -[← 回总目录](../README.md) - -# 29. 遭遇重大打击之后 - -亲人去世、被告知重病、失业、离婚。这几件事在研究里有个共同点：出事的不只是心情，还有身体。风险最高的那一段，集中在头几周到头一年。这一节不讲怎么「走出来」，只讲这段时间里哪些事必须有人盯住，哪些决定必须往后放，什么时候该去挂号。多数条目算的是死亡概率。谈花钱和待遇的四条算的是钱：别一上来就花钱做哀伤咨询（第 9 条）、打 12356 和挂心理门诊（第 11 条）、不可逆的大决定往后推（第 12 条）、别拿死当还债的办法（第 13 条）。两类数字不互相换算。没有亲人也没有朋友可找的，「找个人盯住」这句话怎么落地见第 6 条。 - -亲人去世当时的手续、遗体和该取的钱见第 25 节。老人生前的监护与遗嘱见第 17 节。失业能领什么见第 7 节。情绪低落时的运动与光照见第 3 节和第 22 节。有自杀念头时先打 12356 见第 1 节第 25 条、念头的时间尺度见第 1 节第 32 条、救回来之后的后遗症见第 1 节第 33 条。这些本节都不重复。 - [← Back to the contents](../README.md) # 29. After a severe blow diff --git "a/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" "b/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" index 5b6ffe30..7533f38c 100644 --- "a/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" +++ "b/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" @@ -1,11 +1,3 @@ -[← 回总目录](../README.md) - -# 30. 上学以后的孩子 - -第 20 节讲的是孩子刚出生那一年。第 18 节算的是养孩子要花多少钱、多少时间。中间 3 到 18 岁这十几年，写在这一节。分龄管屏幕那一条从 0 岁写起，因为官方的分档就是 0～3 岁、3～6 岁、中小学三档，拆开写反而不好用。本节回答的是：「别耽误学习」和「身体」撞在一起的时候，哪些事不能等。这些事有时间窗口，过了就补不回来。学业该不该那么卷，本节不讨论，只给窗口和数字。 - -前八条和最后两条算的是死亡和身体上真落下的损害。第 9 条（不买「治愈近视」的产品）算钱。第 10 条（睡眠作业体育排名的明文规定）和第 11 条（休学保留学籍）算时间。溺水、交通、头盔、HPV 疫苗见第 1 节。婴幼儿见第 20 节。孩子被骗和游戏充值见第 5 节。未成年人自己别踩的红线见第 9 节。急症怎么认、怎么叫车见第 13 节。家长去世后的孩子见第 29 节。这些都不在本节重复。 - [← Back to the contents](../README.md) # 30. School-age children diff --git "a/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" "b/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" index f3f6bd37..2859191f 100644 --- "a/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" +++ "b/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" @@ -1,13 +1,3 @@ -[← 回总目录](../README.md) - -# 31. 十八岁之后有哪几条路 - -这一节算的是钱和时间，拒服兵役那一条还关系到人身自由。这是一张路径地图，不是某一条路的说明书。本节先把各条路的法定门槛摆在一起，再逐条写那些知道了才用得上的东西：当兵能拿到什么，进体制内除了考公还有哪些口子，高考填志愿时哪两个专业直接对着编制，学历后来还能不能补，想出国打工该找什么样的公司，不进单位的人社保怎么办，没本钱起步有没有国家替你出一部分利息的钱可借。 - -十八岁之后能走的路，远不止这一节写到的这些。这里只收有全国层面的公开文件、门槛和待遇能逐条核对的那几条。进国企、进工厂、当船员、做自由职业、接着读高中读大学，这些路也能走。只是它们没有统一文件可查，本书不凭印象替它们写条件。 - -选哪条路是你自己的事。本节只提供两样东西：门槛写在哪份文件里，走上去之后有哪些期限和资格过期就作废。读书还是打工那道题怎么算、技能怎么挑，见第 23 节。进了单位之后的加班费、工伤和离职补偿，见第 19 节。开店办公司的本钱和退场，见第 12 节。出国留学见第 32 节。失业之后的保障见第 7 节。这些都不重复。 - [← Back to the contents](../README.md) # 31. What are the paths after eighteen diff --git "a/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" "b/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" index e9fa611c..2b0014b8 100644 --- "a/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" +++ "b/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" @@ -1,13 +1,3 @@ -[← 回总目录](../README.md) - -# 32. 出国留学：身份、打工、保险和回国认证 - -本节算钱和人身自由两笔账，不讨论该不该出国，也不比较学校好坏。下面四件事不知道就要吃亏：签证身份怎么才算没断、打工时数的上限在哪、哪几笔保险和费用是必须交的、回国以后这张文凭能不能被认。 - -覆盖美国、加拿大、英国、澳大利亚四个国家，数字一国一国列。来源是各国移民局和中国教育部的官方网页。**各国的留学政策改得很勤，比中国的法规还勤。本节所有数字的截至日期是 2026 年 9 月。出发前和每次续签前，都照来源栏的链接自己再查一遍。** - -人在境外出了事怎么找使领馆、12308 能管什么、境外医疗和转运保险怎么买，见第 21 节。读书还是打工那道账，见第 23 节。账号被盗见第 14 节，电信诈骗见第 8 节。这些都不重复写。 - [← Back to the contents](../README.md) # 32. Studying abroad: status, work, insurance and certification on return diff --git "a/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" index a3a7aa03..80313bbd 100644 --- "a/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" +++ "b/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -1,17 +1,3 @@ -[← 回总目录](../README.md) - -# 33. 残疾之后怎么活 - -这一节写的是残疾已经发生之后的日子。怎么防住致残，在第 1 节和第 13 节，这里不重复。 - -四种口径分开算，不互相换算。急症和并发症那几条按死亡率算。补贴、待遇、打官司那几条按金钱算。上学、办事那几条按时间算。精神障碍住院、行为能力和监护那几条按人身自由算。 - -世界卫生组织的说明书给了三个数字。全球约有 13 亿人存在显著功能障碍，占世界人口的 16%，大约每六个人里有一个。部分残疾人比没有残疾的人早死最多 20 年。残疾人得抑郁、哮喘、糖尿病、卒中、肥胖或者口腔健康不良的风险是两倍。这些差距里，很大一部分不是残疾本身造成的，是看不上病、上不了学、找不到工作、出不了门造成的。所以这一节的条目，一半是身体，一半是手续和权利。 - -已经写在别处的五件事，这里只指路，不重复。残疾人证怎么办、七个类别和一到四级怎么评，见第 24 节第 11 条（残疾人证）。伤残鉴定要等到什么时候做，见第 24 节第 10 条（伤残鉴定）。残疾人两项补贴怎么领，见第 7 节第 8 条（两项补贴）。工伤的劳动能力鉴定和伤残等级换算成多少钱，见第 19 节第 15 条（劳动能力鉴定）。重度失能的长期护理保险怎么申请，见第 17 节第 7 条（长期护理保险）。那一条写在「家里有老人」那节，但这项待遇没有年龄门槛，年轻人一样能领。 - -本节大部分条目的受益人是你自己和家人，属于最高的两档。涉及别人的地方在备注里单独写。 - [← Back to the contents](../README.md) # 33. Living after a disability @@ -138,7 +124,7 @@ Most items in this section benefit you and your family, which is the top two gra - Cost: No money. You have to hand in one written application within the time the province sets, and then turn up in person for confirmation - In plain language: A candidate with a disability sitting the national college entrance examination can apply for a Braille paper, a large-print paper, exemption from the foreign language listening test, a separate exam room, being allowed to bring in a hearing aid or a wheelchair, and extra time. None of it is granted automatically. You have to hand in a written application within the time the province sets, with a disability certificate of the second generation or above attached. If you disagree with the result, you can ask the provincial education department for a review -- Benefit: Article five of the Provisions on the Administration of the National Unified Examination for Admission to Regular Higher Education Institutions by Candidates with Disabilities《残疾人参加普通高等学校招生全国统一考试管理规定》lists twelve forms of reasonable accommodation. They include: giving candidates with a visual disability the current Braille paper, a large-print paper or an ordinary paper; exempting candidates with a hearing disability from the foreign language listening test; letting candidates with a visual disability bring in a Braille pen, a Braille writing slate, a Braille typewriter with no storage function, an electronic video magnifier with no storage function, a white cane, an optical magnifier and other assistive devices; letting candidates with a hearing disability bring in a hearing aid or a cochlear implant; letting candidates with limited mobility use a wheelchair or a walking aid, and bringing their own special desk and chair where they have special needs. The size of the extension is in Article five, item six: 「使用盲文试卷的视力残疾考生的考试时间，在该科目规定考试总时长的基础上延长50%；使用大字号试卷或普通试卷的视力残疾考生、因脑瘫或其他疾病引起的上肢无法正常书写或无上肢考生等书写特别困难考生的考试时间，在该科目规定考试总时长的基础上延长30%」. There is also priority entry to the examination venue and room, a separate standardised examination room, and the provision of guiding assistants and sign language interpreters. Article seven sets out the application procedure: the candidate lodges a formal written application at the time, place and in the manner set by the provincial education examination authority, and provides their own disability certificate of the second generation or above; an expert panel confirms it on the spot and makes an overall assessment, and the provincial education examination authority issues a Notification. Article eight says that a candidate who objects to the Notification may apply in writing to the provincial education administrative department for a review. Article nine says that where the foreign language listening test is exempted, the foreign language score is worked out as 「笔试成绩×外语科总分值/笔试部分总分值」. Article eighteen says that candidates with disabilities taking other state education examinations who need reasonable accommodation may apply these provisions by reference (nationwide, issued and in force from 2017-4-7) +- Benefit: Article five of the Provisions on the Administration of the National Unified Examination for Admission to Regular Higher Education Institutions by Candidates with Disabilities《残疾人参加普通高等学校招生全国统一考试管理规定》lists twelve forms of reasonable accommodation. They include: giving candidates with a visual disability the current Braille paper, a large-print paper or an ordinary paper; exempting candidates with a hearing disability from the foreign language listening test; letting candidates with a visual disability bring in a Braille pen, a Braille writing slate, a Braille typewriter with no storage function, an electronic video magnifier with no storage function, a white cane, an optical magnifier and other assistive devices; letting candidates with a hearing disability bring in a hearing aid or a cochlear implant; letting candidates with limited mobility use a wheelchair or a walking aid, and bringing their own special desk and chair where they have special needs. The size of the extension is in Article five, item six: 「使用盲文试卷的视力残疾考生的考试时间，在该科目规定考试总时长的基础上延长50%；使用大字号试卷或普通试卷的视力残疾考生、因脑瘫或其他疾病引起的上肢无法正常书写或无上肢考生等书写特别困难考生的考试时间，在该科目规定考试总时长的基础上延长30%」. There is also priority entry to the examination venue and room, a separate standardised examination room, and the provision of guiding assistants and sign language interpreters. Article seven sets out the application procedure: the candidate lodges a formal written application at the time, place and in the manner set by the provincial education examination authority, and provides their own disability certificate of the second generation or above; an expert panel confirms it on the spot and makes an overall assessment, and the provincial education examination authority issues a Notification《告知书》. Article eight says that a candidate who objects to the Notification may apply in writing to the provincial education administrative department for a review. Article nine says that where the foreign language listening test is exempted, the foreign language score is worked out as 「笔试成绩×外语科总分值/笔试部分总分值」. Article eighteen says that candidates with disabilities taking other state education examinations who need reasonable accommodation may apply these provisions by reference (nationwide, issued and in force from 2017-4-7) - Evidence grade: A - Sources: Ministry of Education (教育部) and China Disabled Persons' Federation (中国残联) (2017). Provisions on the Administration of the National Unified Examination for Admission to Regular Higher Education Institutions by Candidates with Disabilities (Teaching〔2017〕4, Articles five, seven, eight, nine and eighteen)《残疾人参加普通高等学校招生全国统一考试管理规定（教学〔2017〕4 号，第五条、第七条、第八条、第九条、第十八条）》. - Notes: The application dates are set by each province and written into that year's rules for signing up for the examination. Miss the date and there is no second one. So check the provincial education examination authority's website in the first term of the final year of high school. Article eighteen also covers the postgraduate entrance examination, the adult higher education entrance examination and other state education examinations, which may apply by reference. The rule that a school must not refuse admission during the school years is item 14 in this section (schools must not refuse admission). The person who benefits from this entry is you or your child @@ -203,4 +189,4 @@ Most items in this section benefit you and your family, which is the top two gra - Benefit: Article three of the Law on the Protection of Persons with Disabilities《残疾人保障法》says: 「禁止基于残疾的歧视。禁止侮辱、侵害残疾人。」 Article thirty-eight says: 「在职工的招用、转正、晋级、职称评定、劳动报酬、生活福利、休息休假、社会保险等方面，不得歧视残疾人。」 The same article also provides that the unit where a disabled employee works shall, according to the characteristics of the disabled employee, provide suitable working conditions and labour protection, and shall, as actually necessary, adapt the workplace, the equipment and the living facilities. Article sixty-four says: 「违反本法规定，在职工的招用等方面歧视残疾人的，由有关主管部门责令改正；残疾人劳动者可以依法向人民法院提起诉讼。」 Article four of the Regulations on the Employment of Persons with Disabilities《残疾人就业条例》says: 「禁止在就业中歧视残疾人。」 Article thirteen says that an employer shall not discriminate against disabled employees in promotion, advancement, grading of professional title, pay, social insurance, living benefits and the like. Article forty-two of the Legal Aid Law《法律援助法》says that an applicant who has documents proving that he belongs to a specific group such as 「无固定生活来源的未成年人、老年人、残疾人等特定群体」 is exempted from any check on whether he is financially in difficulty. Article forty-five says that where a legal aid institution provides legal aid services to elderly people or to persons with disabilities, it shall provide accessible facilities, equipment and services as the circumstances require (nationwide) - Evidence grade: A - Sources: Law of the People's Republic of China on the Protection of Persons with Disabilities (Articles three, thirty-eight and sixty-four)《中华人民共和国残疾人保障法（第三条、第三十八条、第六十四条）》. ; State Council (国务院) (2007). Regulations on the Employment of Persons with Disabilities (Articles four and thirteen)《残疾人就业条例（第四条、第十三条）》. ; Legal Aid Law of the People's Republic of China (Articles forty-two and forty-five)《中华人民共和国法律援助法（第四十二条、第四十五条）》. -- Notes: The cost of the process has to be worked out first. An ordinary procedure at first instance takes 6 months or more and can be extended; a simplified procedure takes 3 months. Legal fees are not among the litigation costs listed in Article six of the Regulations on the Payment of Litigation Costs, and the 「败诉方负担」 in Article twenty-nine does not include legal fees. So take the two cheap routes first: complain to the human resources and social security department or to the disabled persons' federation, which must order the discrimination stopped; and apply for legal aid at the same time, because Article forty-two exempts disabled people from the check on financial hardship. Collecting evidence is the hardest part. When you are turned down at an interview the other side seldom leaves a written reason, so keep a recording and keep the messages. Once you are employed, the employment dispute goes to labour arbitration; for keeping evidence and the time limit see item 8 in section 19 (keep the evidence before you leave). The person who benefits from this entry is you yourself \ No newline at end of file +- Notes: The cost of the process has to be worked out first. An ordinary procedure at first instance takes 6 months or more and can be extended; a simplified procedure takes 3 months. Legal fees are not among the litigation costs listed in Article six of the Regulations on the Payment of Litigation Costs《诉讼费用交纳办法》, and the 「败诉方负担」 in Article twenty-nine does not include legal fees. So take the two cheap routes first: complain to the human resources and social security department or to the disabled persons' federation, which must order the discrimination stopped; and apply for legal aid at the same time, because Article forty-two exempts disabled people from the check on financial hardship. Collecting evidence is the hardest part. When you are turned down at an interview the other side seldom leaves a written reason, so keep a recording and keep the messages. Once you are employed, the employment dispute goes to labour arbitration; for keeping evidence and the time limit see item 8 in section 19 (keep the evidence before you leave). The person who benefits from this entry is you yourself \ No newline at end of file diff --git "a/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" "b/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" index c22a121c..74e9ea9b 100644 --- "a/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" +++ "b/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" @@ -1,15 +1,3 @@ -[← 回总目录](../README.md) - -# 34. 家里的常备药别吃出事 - -退烧药、止痛药、感冒药、胃药、止泻药，很多在药店就能自己买。不用处方能买，不等于怎么吃都行。这一节不讲每种病该吃什么药，只挑一个动作就能避开重后果的几件事。 - -所有条目按死亡率口径算，包括肝衰竭、胃出血、新生儿肾衰竭这类健康终点。收益大多是「少出一次严重事故」，研究给不出「做了能少几成」的数字，收益量级是按后果轻重判断的，每条备注里写了凭什么定。 - -已经写在别处的，这里只指路。误服了药或清洁剂先做什么，见第 13 节第 20 条（误服）。慢性病的药按医嘱吃满，见第 16 节第 1 条（药按医嘱吃满）。网上买处方药要先过处方审核，见第 28 节第 6 条（减肥药就去医院拿处方）。 - -本节条目的受益人是你自己和家人，属于最高的两档。 - [← Back to the contents](../README.md) # 34. Do not mishandle the medicines at home diff --git a/tools/i18n/check.mjs b/tools/i18n/check.mjs index f4ed88b1..44746324 100644 --- a/tools/i18n/check.mjs +++ b/tools/i18n/check.mjs @@ -47,6 +47,8 @@ for (const code of codes){ const dst = entrySkeleton(read(dstPath), S); for (const e of dst.entries) e._locale = code; if (src.n !== dst.n) report(`${code} book/${f}`, `节号是 ${dst.n}，原文 ${src.n}`); + // 节标题和导读：条目之外，节首那几行也得翻 + for (const w of STRUCT.compareSection(src, dst, `${code} book/${f}`)) problems.push(w); if (src.entries.length !== dst.entries.length) report(`${code} book/${f}`, `条目数 ${dst.entries.length}，原文 ${src.entries.length}`); for (let i = 0; i < Math.min(src.entries.length, dst.entries.length); i++) diff --git a/tools/i18n/join.mjs b/tools/i18n/join.mjs index 4c53f2f6..e01a1baf 100644 --- a/tools/i18n/join.mjs +++ b/tools/i18n/join.mjs @@ -58,6 +58,12 @@ for (const srcPath of files){ const a = entrySkeleton(want, strings(SOURCE.code)); const b = entrySkeleton(got, strings(locale)); for (const e of b.entries) e._locale = locale; // 字段名按译文那门语言认 + // 第 0 块自带节首，节号/标题/导读在这一块里校（译文块是整节的一小块，不是整节）。 + // 拿整节那一份（src + 第 0 块合起来）当基准：want 只是条目，节首在 want 里没有。 + if (k === 0 && /^#{1,2} \d+\. /m.test(got)){ + const secA = entrySkeleton(src.head.join('\n') + '\n' + want, strings(SOURCE.code)); + for (const w of STRUCT.compareSection(secA, b, `${locale} ${name} 块 0 节首`)){ bad++; trouble++; console.error(' ' + w); } + } for (const [i, x] of a.entries.entries()){ const y = b.entries[i]; const tag = `${name} 块 ${k} 第 ${x.n} 条`; @@ -68,28 +74,22 @@ for (const srcPath of files){ bad++; trouble++; console.error(` ${dir}/${parts[k]}：${b.entries.length} 条，原文这块 ${a.entries.length} 条`); } - // 译文块里如果又带了一遍节首（回目录那行 + 「# N. 节名」+ 导读），拼出来就是两份， - // 页面上那一节连同它的条目渲染两遍——en 和 vi 五个文件都中过（2026-09-30 迁移到 - // Astro 时预渲染和运行时解析出来的节数对不上，查出来的）。 + // 节首（回目录那行 + 「# N. 节名」 + 导读）归第 0 块管，译文该翻的就在第 0 块里。 // - // 判据按内容，不按行数：节首的每一行都逐行对上（导读在译文里可能已经翻过， - // 对不上就整块留着，宁可多渲染一遍导读也不能把译文正文吃掉）。 + // 这里只判「有没有」，不判「翻没翻」——翻没翻是 check.mjs 的事（它拿节标题跟原文 + // 逐字比，一模一样就是没翻）。曾经在这里反过来做：拿原文节首的每一行去译文块里 + // 逐行找，对上了就把译文那份节首去掉、再把中文节首拼上去。结果是译文里翻好的节名 + // 被当成「多出来的」删掉，中文那份留着——en 和 vi 一半的节标题是中文的， + // 有的还多出一份（中文在前、翻好的在后），页面上那一节连同条目渲染两遍。 + // 拼的时候不该拿原文去覆盖译文：原文那一份在 book/ 里摆着，译文自己带节首才对。 let body = got; - if (k === 0 && src.head.length){ - const headLines = src.head.join('\n').split('\n').map(l => l.trim()).filter(Boolean); - const bodyLines = got.split('\n'); - let consumed = 0; - for (const hl of headLines){ - const at2 = bodyLines.findIndex((l, i) => i >= consumed && l.trim() === hl); - if (at2 < 0) { consumed = -1; break; } // 对不上：整块保留 - consumed = at2 + 1; - } - if (consumed > 0 && /^#{1,2} \d+\. /.test(headLines.find(l => /^#{1,2} \d+\. /.test(l)) || '')){ - body = bodyLines.slice(consumed).join('\n').replace(/^\n+/, ''); - console.log(` ${name} 块 0：译文里带了节首（${headLines.length} 行），已去掉，之前会重复渲染一次`); - } + if (k === 0 && src.head.length && !/^#{1,2} \d+\. /m.test(got)){ + // 第 0 块没带节首（少数块是这样切的）：补上，但补的是中文原文那份，所以报出来。 + // 不报的话 check.mjs 只会说「节标题跟原文一样」，看不出是这里补上去的。 + body = src.head.join('\n') + '\n' + got; + console.error(` ${name} 块 0：译文块里没有节首，先补了中文原文那份——节名和导读得在 .i18n/${locale}/${name.replace(/\.md$/, '')}/part00.md 里翻好`); } - pieces.push(k === 0 ? src.head.join('\n') + '\n' + body : body); + pieces.push(body); } // 写进仓库里的译文目录：contentPath，不是 L.dir。L.dir 是页面上线上的地址， // 英文那份是空串（页面在根上），拿它拼就是 book/ —— 中文原文那一份。 diff --git a/tools/i18n/structure.mjs b/tools/i18n/structure.mjs index 693f3386..2cc701b4 100644 --- a/tools/i18n/structure.mjs +++ b/tools/i18n/structure.mjs @@ -162,6 +162,39 @@ function label(S, f) { return S.fields[f].replace(/[：:]$/, ''); } */ export const STRUCT = { ALWAYS, FIELDS, + /** 整节的判据：节号、节标题、导读。条目逐条比是 compare()。 + * + * 节标题和导读以前没人查，join.mjs 又拿中文原文那一份盖在译文前面（2026-10-02 修）， + * 结果 en 和 vi 一半的节标题是中文的，有的还多出一份，页面上那一节渲染两遍。 + * 这两条都是机器查得出来的：节标题跟原文一字不差就是没翻；导读里只要还有 + * 整段没翻的中文就是没翻完。 */ + compareSection(a, b, tag) { + const out = []; + const han = s => /[㐀-鿿]/.test(s); // 汉字。引号里照抄的法条名不算漏译 + if (a.n !== b.n) out.push(`${tag}：节号是 ${b.n}，原文 ${a.n}`); + if (!b.title.trim()) out.push(`${tag}：节标题是空的`); + else if (a.title && b.title === a.title) + out.push(`${tag}：节标题「${b.title}」跟中文原文一字不差，这一行没翻`); + + // 导读逐段比。节首的导读是散文，但「整段还是中文」是查得出来的：把一行里 + // 照抄的中文（法条名、书名）剔掉之后，剩下的还是汉字，就是没翻。 + // 只在原文有导读时查，段落数不强制——译文可以合并或拆开自然段。 + if (a.intro.length && !b.intro.length) out.push(`${tag}：导读不见了，原文有 ${a.intro.length} 段`); + for (const line of b.intro){ + if (!han(line)) continue; // 整段没有汉字，不用往下剔了 + // 照抄的部分：法条名、书名、括注、链接。剩下的还有汉字，就是整段还没翻。 + // 只剔「紧贴汉字、含汉字」的这些形式——英文里写 (breaking line) 是普通括注， + // 一并剔掉会把真漏译的段落放过去。 + const s = line + .replace(/^\[.*\]\(.*\)$/, '') // 回目录那行 + .replace(/「[^」]*」|『[^』]*』|《[^》]*》/g, '') // 照抄的法条名、书名 + .replace(/\[[^\]]*\]\([^)]*\)/g, '') // 照抄的链接 + .replace(/[（(][^)）]*[㐀-鿿][^)）]*[）)]/g, '') // 含汉字的括注 + .trim(); + if (han(s)) out.push(`${tag}：导读里这一段还没翻：「${line.slice(0, 30)}…」`); + } + return out; + }, compare(a, b, tag) { const out = []; const S = strings(SOURCE.code); diff --git "a/vi/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" "b/vi/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" new file mode 100644 index 00000000..4a8be53d --- /dev/null +++ "b/vi/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" @@ -0,0 +1,191 @@ +[← Về mục lục](../README.md) + +# 7. Sống thế nào khi hết tiền + +Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được miễn bao nhiêu, được dịch vụ gì, phải đáp ứng điều kiện gì, tất cả đều ghi ở đây, không quy đổi thành tuổi thọ. Chính sách đổi mỗi năm, mỗi nơi mỗi khác, nên số tiền và ngưỡng đều theo những gì địa phương của bạn công bố trong năm đó. Các mục được sắp như sau: mục không tốn tiền, thủ tục ít, có tác dụng nhanh thì đứng trước. Mục phải tích lũy điều kiện lâu dài, mỗi nơi khác nhau nhiều thì đứng sau. Mục chỉ dựa vào kinh nghiệm, không có văn bản chính thức thì để cuối cùng. + +### 1. Mất việc thì xin trợ cấp bảo hiểm thất nghiệp trực tuyến trước + +- Chi phí: Không tốn tiền. Nộp đơn bằng điện thoại được, cả ba cửa đều được: ứng dụng「掌上 12333」, tiểu trình thẻ bảo hiểm xã hội điện tử trong WeChat hoặc Alipay, và nền tảng dịch vụ bảo hiểm xã hội công cộng quốc gia. Có thẻ bảo hiểm xã hội hoặc căn cước trong tay là làm được, phần lớn các nơi không phải chạy tận nơi. Điều kiện trước là đã làm xong đăng ký thất nghiệp. +- Nói thẳng: Đơn vị và bản thân bạn đóng đủ 1 năm, việc mất việc không phải do bạn chủ động xin thôi, làm xong đăng ký thất nghiệp thì nhận được tiền. Đóng đủ 1 năm đến dưới 5 năm, nhận nhiều nhất 12 tháng. Đủ 5 năm đến dưới 10 năm, nhiều nhất 18 tháng. Từ 10 năm trở lên, nhiều nhất 24 tháng. Trong thời gian nhận tiền, bảo hiểm thất nghiệp đóng thay phần bảo hiểm y tế người lao động cho bạn. +- Lợi ích: Phải đáp ứng đồng thời ba điều kiện. Một là đơn vị và bản thân bạn đã đóng bảo hiểm thất nghiệp tổng cộng đủ 1 năm. Hai là việc mất việc không phải do bạn chủ động xin thôi (pháp luật gọi là「非因本人意愿中断就业」). Ba là đã làm đăng ký thất nghiệp và có ý định tìm việc. Thời gian được nhận tính theo số năm đã đóng. Tổng cộng đủ 1 năm đến dưới 5 năm, dài nhất 12 tháng. Đủ 5 năm đến dưới 10 năm, dài nhất 18 tháng. Từ 10 năm trở lên, dài nhất 24 tháng. Số tiền mỗi tháng do chính quyền tỉnh quy định, mỗi tỉnh khác nhau. Số tiền thấp hơn lương tối thiểu địa phương, cao hơn trợ cấp người có thu nhập thấp ở thành phố. Mục tiêu của nhà nước là dần đưa lên tới 90% mức lương tối thiểu. Trong thời gian nhận tiền trợ cấp thất nghiệp, bạn vẫn tham gia bảo hiểm y tế người lao động và được hưởng quyền lợi bảo hiểm y tế cơ bản. Khoản phí bảo hiểm y tế cơ bản phải đóng được lấy từ quỹ bảo hiểm thất nghiệp, bạn không phải đóng (toàn quốc, quy định áp dụng từ 1999) +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (1999). Quy định về bảo hiểm thất nghiệp（失业保险条例）（国务院令第 258 号，第十四、十七、十八条）. ；Bộ Nhân sự và Xã hội (人力资源社会保障部). Cẩm nang thủ tục xin trợ cấp bảo hiểm thất nghiệp trên Nền tảng dịch vụ chính phủ nhân sự và bảo hiểm xã hội toàn quốc（全国人社政务服务平台·失业保险待遇申领办事指南）. ；Cục Việc làm thuộc Ủy ban Cải cách và Phát triển quốc gia (国家发展改革委就业司) (2022). Mức trợ cấp thất nghiệp sẽ dần nâng lên bằng 90% lương tối thiểu（失业保险金标准，将逐步提高至最低工资的 90%）. ；Sở Nhân sự và Xã hội tỉnh Liêu Ninh (辽宁省人力资源和社会保障厅) (2019). Thông báo về việc điều chỉnh tiếp mức trợ cấp thất nghiệp（关于进一步调整失业保险金标准的通知）：「为落实《人力资源社会保障部财政部关于调整失业保险金标准的指导意见》(人社部发〔2017〕71号〕"逐步将失业保险金标准提高到最低工资标准的90%"等要求」. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2010). Luật Bảo hiểm xã hội（社会保险法）（第四十八条）. （Kho luật và chính sách Cục Tính thuế quốc gia） +- Ghi chú: Ai tự chủ động nghỉ việc thì không tính là「非因本人意愿」, không nhận được. Trong thời gian nhận trợ cấp, bảo hiểm y tế người lao động do bảo hiểm thất nghiệp đóng thay. Số tiền mỗi tỉnh chênh lệch rất nhiều, theo những gì nơi bạn tham gia bảo hiểm công bố. Cụm「最低工资 90%」lấy từ ý kiến chỉ đạo năm 2017 của Bộ Nhân sự và Xã hội cùng Bộ Tài chính（人社部发〔2017〕71 号）, nguyên văn là「逐步将失业保险金标准提高到最低工资标准的90%」. Đó là mục tiêu, không phải hiện nay đã trả theo 90%. Trang gốc trên cổng Bộ Nhân sự và Xã hội không mở được trên máy này, nên câu đó được đối chiếu theo trích dẫn trong văn bản của Sở Nhân sự và Xã hội tỉnh Liêu Ninh. + +### 2. Bị nợ lương thì khiếu nại với thanh tra lao động trước, rồi mới xin hoà giải lao động, cả hai đường đều không tốn tiền, phần lớn vụ có kết quả trong vài tháng + +- Chi phí: Không tốn tiền, khiếu nại và hoà giải đều không thu phí. Cần chuẩn bị hợp đồng lao động, bảng lương, sổ chấm công, lịch sử chuyển khoản và tin nhắn, chạy tới cục nhân sự xã hội hoặc ủy ban hoà giải 1 đến 3 chuyến. Xin hoà giải có thời hạn 1 năm (pháp luật gọi là thời hiệu khởi kiện). Người còn làm ở đơn vị đó mà bị nợ lương thì không bị giới hạn 1 năm này. Người đã nghỉ việc thì phải nộp trong vòng 1 năm kể từ ngày nghỉ. Khó ở chỗ phải chạy đi chạy lại, phải tìm từng mẩu chứng cứ, dễ nản giữa chừng. +- Nói thẳng: Cả hai đường đều không mất tiền. Gọi 12333 hoặc khiếu nại với bộ phận thanh tra lao động ở cục nhân sự xã hội, sau khi khởi kiện phải xác minh xong trong 60 ngày làm việc, có thể buộc công ty trả tiền trong thời hạn, quá hạn không trả thì còn phải bồi thêm 50% đến 100%. Không được thì xin hoà giải lao động, kết thúc trong 45 ngày. Nhưng「kết thúc」không có nghĩa là「tiền về tài khoản」: công ty không còn tài sản, chủ doanh nghiệp bỏ trốn, thắng kiện cũng có thể không được tiền. +- Lợi ích: Bước đầu là khiếu nại. Gọi 12333, hoặc khiếu nại với bộ phận thanh tra lao động ở cục nhân sự xã hội địa phương. Pháp luật quy định xác minh xong trong 60 ngày làm việc kể từ ngày khởi kiện, vụ phức tạp thì được kéo dài thêm 30 ngày làm việc. Xác minh đúng thì có thể buộc công ty trả tiền trong thời hạn. Quá hạn vẫn không trả thì phải trả thêm một khoản bồi thường bằng 50% đến 100% số tiền phải trả. Năm 2024 các cơ quan thanh tra trên toàn quốc xử lý 13.7 vạn vụ, thu hồi 86.5 tỷ tệ tiền lương cho 65.4 vạn người. Trong tay có giấy nợ hoặc bảng lương, sự việc nợ lương rõ ràng, còn có thể xin tòa cấp cấp xét cấp lệnh chi trả. Lệnh chi trả là chỉ thị bắt công ty trả tiền do tòa ban, tòa phải ban trong 15 ngày kể từ ngày nhận đơn. Bước thứ hai là xin hoà giải. Pháp luật quy định kết thúc trong 45 ngày kể từ ngày nhận đơn, có thể kéo dài thêm 15 ngày. Người đòi lương có thể xin thi hành trước: không chờ xét xong toàn bộ, thi hành trước một phần. Số tiền nợ không vượt quá 12 lần lương tối thiểu theo tháng của địa phương thì thuộc loại một trọng một quyết: phán quyết của trọng tài là kết quả cuối, công ty không được khởi kiện. Năm 2024 toàn quốc tiếp nhận 425.7 vạn vụ, giải quyết xong 415.6 vạn vụ. Tổng tiền giải quyết 934.7 tỷ tệ. Tỷ lệ thành công hòa giải 79.6%, tỷ lệ kết thúc ở trọng tài 98.2%, tỷ lệ phán quyết không đi tiếp 73.7%. Người ta nói việc này phải kéo dài「nửa năm một năm」, nói đến phần vụ sau khi qua trọng tài lại kiện ra tòa: xét sơ thẩm 6 tháng (loại đơn giản là 3 tháng), cộng thêm xét phúc âm 3 tháng. Năm 2024 tòa xét xong 8.8 vạn vụ đòi tiền công lao động, thu hồi 285.4 tỷ tệ, 998 người bị kết án về tội không chịu trả tiền công lao động (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Nhân sự và Xã hội (人力资源社会保障部) (2025). Báo cáo thống kê công trình nhân sự và bảo hiểm xã hội năm 2024. ；Tòa án nhân dân tối cao (最高人民法院) (2025). Báo cáo công tác của Tòa án nhân dân tối cao. ；Quốc vụ viện (国务院) (2004). Quy định về thanh tra bảo đảm lao động（劳动保障监察条例）（国务院令第 423 号，第九、十七、二十六条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007). Luật Hoà giải và trọng tài lao động（劳动争议调解仲裁法）（主席令第八十号，第二十七、四十三、四十四、四十七、五十、五十三条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2012, sửa đổi). Luật Hợp đồng lao động（劳动合同法）（第三十、八十五条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2023, sửa đổi). Luật Tố tụng dân sự（民事诉讼法）（第一百零九、一百五十二、一百六十四、一百八十三、二百二十五、二百二十七条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2011). Bản sửa đổi (tám) Bộ luật Hình sự（刑法修正案（八））第四十一条, tức là 刑法第二百七十六条之一 của Bộ luật Hình sự. ；Tòa án nhân dân tối cao (最高人民法院) (2013). Giải thích về một số vấn đề pháp lý khi xét các vụ án hình sự về việc không chịu trả tiền công lao động（法释〔2013〕3 号，第三条）. ；Quốc vụ viện (国务院) (2019). Quy định bảo đảm việc trả lương cho công nhân nông nghiệp（保障农民工工资支付条例）（国务院令第 724 号，第十、二十六、三十、三十一、三十二、四十一条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2021). Luật Trợ giúp pháp lý（法律援助法）（第三十一、四十二条）. +- Ghi chú: Khiếu nại và hoà giải đều có ích, nhưng phải phân biệt「kết thúc」với「tiền về tài khoản」. Tỷ lệ kết thúc 98.2% chỉ nói thủ tục đã đi hết. Công ty không có tài sản, chủ doanh nghiệp bỏ trốn thì thắng kiện vẫn không được tiền. Tòa án nhân dân tối cao chỉ công bố tổng số tiền đã thi hành được (2.3 nghìn tỷ tệ năm 2024) và nói「执行完毕率保持高位」, còn cuối cùng bao nhiêu tiền nợ lương thật sự vào túi người bị thiếu thì không có số liệu chính thức. Cụm「仲裁终结率」trong báo cáo không được định nghĩa, mục này hiểu là「sau khi có phán quyết thì không còn đi kiện ra tòa」. Nợ lương đến mức nào thì đủ đi khởi kiện hình sự. Một là nợ một người từ 3 tháng trở lên, số tiền từ 5000 đến 2 vạn tệ trở lên. Trường hợp khác là nợ từ 10 người trở lên, tổng cộng từ 3 vạn đến 10 vạn tệ trở lên. Với hai trường hợp này, đã bị buộc trả mà vẫn không trả thì đủ đi khởi kiện. Tòa án cấp cao và viện kiểm sát của từng tỉnh ấn định và công bố con số cụ thể trong khoảng đó, nên trước khi khiếu nại nên tra con số của tỉnh mình, và nêu ra tận mặt khi khiếu nại. Công nhân nông nghiệp tại công trình còn có quy định riêng về tài khoản lương riêng và việc nhà thầu chính phải trả trước, nên tìm nhà thầu chính còn hơn tìm đội trưởng. Công nhân vào thành phố làm việc xin trợ giúp pháp lý để đòi lương thì không bị xét khó khăn tài chính. Nguyên văn Luật Hợp đồng lao động và Luật Tố tụng dân sự được trích từ trang sao chép của Cục Tính thuế và Cục Quản lý thị trường tỉnh Quý Châu (đều là gov.cn), vì trang của Đại biểu Quốc hội Trung Quốc không mở được. Bản sửa đổi (tám) Bộ luật Hình sự không tải được trên trang Đại biểu Quốc hội Trung Quốc, nên đã dùng curl mở ra để đối chiếu. + +### 3. Không đủ tiền kiện tụng thì xin trợ giúp pháp lý, các vụ đòi lương, cấp dưỡng, tai nạn lao động vốn đã nằm trong phạm vi + +- Chi phí: Không tốn tiền. Trợ giúp pháp lý không thu một đồng nào của người được trợ giúp. +- Nói thẳng: Đòi lương, đòi tiền cấp dưỡng, đòi trợ cấp bảo hiểm xã hội hoặc trợ cấp người có thu nhập thấp, bị thiệt hại khi làm việc nghĩa cứu người, đều có thể xin trợ giúp pháp lý miễn phí. Gọi 12348, hoặc đến trung tâm trợ giúp pháp lý địa phương. Bây giờ không phải về quê gia đình mở giấy chứng minh khó khăn tài chính nữa, bản thân người khai đúng tình hình là được. Bị can có thể bị kết án tử hình, và bị can mù, khiếm thính, khiếm mù hoặc chưa thành niên, đều được chỉ định luật sư. +- Lợi ích: Quy định của Quốc vụ viện ghi rõ trợ giúp pháp lý dân sự gồm sáu loại việc. Nguyên văn quy định là:「依法请求国家赔偿的；请求给予社会保险待遇或者最低生活保障待遇的；请求发给抚恤金、救济金的；请求给付赡养费、抚养费、扶养费的；请求支付劳动报酬的；主张因见义勇为行为产生的民事权益的」. Nói trắng ra là: đòi bồi thường của nhà nước, đòi trợ cấp bảo hiểm xã hội hoặc trợ cấp người có thu nhạp thấp, đòi tiền từ nạn và tiền trợ cấp, đòi tiền cấp dưỡng cấp dưỡng nuôi dưỡng, đòi tiền lương, bị thiệt hại khi làm việc nghĩa cứu người. Ở phần hình sự, bị can có thể bị kết án tử hình, cùng bị can mù, khiếm thính, khiếm mù và chưa thành niên, không xét tình hình tài chính, đều được chỉ định luật sư +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2003). Quy định về trợ giúp pháp lý（法律援助条例）（国令第 385 号，第十条、第十一条、第十二条）. ；Lãnh đạo Bộ Tư pháp (司法部有关负责人) (2021). Trả lời phóng viên về việc thực thi Luật Trợ giúp pháp lý. +- Ghi chú: Luật Trợ giúp pháp lý có hiệu lực từ ngày 1 tháng 1 năm 2022, có một thay đổi. Trước đây phải「提交经济困难证明」. Nay đổi thành bản thân người khai đúng tình hình tài chính của mình là được, không phải về quê mở giấy nữa. Cửa vào là trung tâm trợ giúp pháp lý và trung tâm dịch vụ pháp lý công cộng ở địa phương, cũng có thể gọi 12348. + +### 4. Không còn đường lui thì đến trạm cứu hộ, ở ăn ở và có vé về quê + +- Chi phí: Không tốn tiền. Tự đến trạm quản lý cứu hộ địa phương xin giúp, cũng có thể nói với công an hoặc bộ phận dân vụ để họ đưa bạn đi. Khi vào phải khai trung thực họ tên và đồ đạc mang theo. Khó ở chỗ phải nói với người khác rằng mình không có tiền, không có chỗ đi. +- Nói thẳng: Trạm cứu hộ lo ăn, lo ở, bị bệnh thì đưa đi bệnh viện, còn giúp liên hệ người nhà. Không có tiền về nhà thì cấp vé. Thường ở không quá 10 ngày. Đến tự nguyện, đi tự nguyện, không phải trại tạm giam, cũng không phát tiền mặt. Vé cấp theo nơi đăng ký hộ khẩu hoặc nơi cư trú, trước khi đi hãy nghĩ kỹ mình về đâu. +- Lợi ích: Trạm cứu hộ cung cấp thức ăn và chỗ ở đáp ứng yêu cầu vệ sinh. Ở trong trạm mà phát bệnh thì đưa đi bệnh viện chữa. Trạm còn giúp liên hệ với người thân và cơ quan. Không có tiền mua vé về nơi đăng ký hộ khẩu hoặc nơi cư trú thì cấp chứng từ đi xe (đi tàu), tức là vé tàu xe. Nơi đăng ký hộ khẩu là quê, nơi cư trú là chỗ đang ở. Thời hạn cứu hộ thường không quá 10 ngày. Chờ người nhà đón, chờ điều trị, chờ sắp xếp thì có thể kéo dài. Công an và nhân viên cơ quan hành chính khác khi thực hiện công vụ mà gặp người lang thang xin ăn thì phải báo cho người đó biết có thể đi xin giúp. Với người khuyết tật, chưa thành niên, người cao tuổi và người khó di chuyển, phải hướng dẫn và đưa đi (toàn quốc, từ năm 2003) +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2003). Biện pháp cứu hộ và quản lý người lang thang xin ăn không nơi nương tựa ở thành phố（城市生活无着的流浪乞讨人员救助管理办法）（国务院令第 381 号，第五、六、七条）. ；Bộ Dân vụ (民政部) (2003). Quy tắc chi tiết về thi hành Biện pháp cứu hộ và quản lý người lang thang xin ăn không nơi nương tựa ở thành phố（城市生活无着的流浪乞讨人员救助管理办法实施细则）（民政部令第 24 号，第十一、十二条）. +- Ghi chú: Xin giúp và rời đi đều tự nguyện, không phải trại tạm giam. Trạm cho ăn ở cơ bản, không phát tiền mặt. Chứng từ đi xe cấp theo nơi đăng ký hộ khẩu hoặc nơi cư trú, trước khi xin giúp hãy nghĩ kỹ mình đi đâu. +### 5. Tìm việc thì dùng dịch vụ việc làm công cộng miễn phí và thị trường việc lẻ trước, đừng qua môi giới thu phí + +- Chi phí: Không tốn tiền. Đi đăng ký ở cơ quan dịch vụ việc làm công cộng cấp huyện trở lên, ví dụ trung tâm dịch vụ việc làm, thị trường lao động. Thị trường việc lẻ địa phương cũng đăng ký được. Không biết địa chỉ thì gọi 12333 hỏi. +- Nói thẳng: Trung tâm dịch vụ việc làm, thị trường lao động, những cơ quan công này, việc giới thiệu việc làm, hướng nghiệp, tư vấn chính sách, làm đăng ký việc làm và đăng ký thất nghiệp đều không thu tiền. Pháp luật còn cấm họ dùng các dịch vụ này để kiếm tiền. Đăng ký và đăng tin ở thị trường việc lẻ cũng miễn phí. Làm luôn đăng ký thất nghiệp cho xong, lĩnh tiền thất nghiệp và xác nhận người khó khăn việc làm đều cần đến nó. +- Lợi ích: Pháp luật quy định mấy thứ sau đây đều miễn phí: tư vấn chính sách việc làm; thông tin nơi nào đang tuyển, nơi nào thiếu người; hướng nghiệp và giới thiệu việc làm; làm đăng ký việc làm và đăng ký thất nghiệp; giúp việc làm cho người khó khăn việc làm. Những cơ quan này không được kinh doanh, không được lấy các dịch vụ này ra kiếm tiền. Thông tin tìm việc, tuyển dụng việc lẻ, việc đăng ký và đăng tin cũng được cung cấp miễn phí cho xã hội, đã nằm trong phạm vi dịch vụ thông tin việc làm công cộng. Với người chờ việc lâu (tức là lâu không nhận được việc), hộ thu nhập thấp và người làm việc lẻ khó khăn khác, còn phải tăng cường hỗ trợ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007, sửa đổi 2015). Luật thúc đẩy việc làm（第三十五条）. ；Bộ Nhân sự và Xã hội và bốn cơ quan khác (人力资源社会保障部等五部门) (2022). Ý kiến về việc tăng cường xây dựng thị trường việc lẻ, hoàn thiện dịch vụ tìm việc và tuyển dụng（人社部发〔2022〕38 号）. +- Ghi chú: Đăng ký thất nghiệp là thủ tục trước khi lĩnh bảo hiểm thất nghiệp và xác nhận người khó khăn việc làm, làm luôn cho xong. Làm việc lẻ trả theo ngày thì đối chiếu tiền công ngay tại chỗ, giữ lại cách liên hệ của đối phương và bản ghi công việc. + +### 6. Gặp biến cố đột ngột thì xin trợ cấp tạm thời trước + +- Chi phí: Không tốn tiền. Nộp đơn lên chính quyền xã, thị trấn hoặc ủy ban quản lý khu phố, nói rõ ở nhà đã xảy ra chuyện gì. Ví dụ hỏa hoạn, tai nạn giao thông, người nhà đột ngột mắc bệnh nặng, hoặc chi tiêu bắt buộc hằng ngày đột ngột tăng lên. Sau khi thẩm định và niêm yết công khai, phòng dân sự cấp huyện phê duyệt. Chuẩn bị giấy tờ theo danh sách địa phương đưa, thường phải đi 1 đến 2 lần. +- Nói thẳng: Nhà đột ngột gặp chuyện, như hỏa hoạn, tai nạn, người nhà đột ngột mắc bệnh nặng, nhất thời không gánh nổi sinh hoạt cơ bản, thì có thể xin trợ cấp tạm thời lên chính quyền xã, thị trấn hoặc ủy ban khu phố. Không yêu cầu bạn đã là hộ nghèo. Thường phát một lần, số tiền mỗi nơi chênh lệch rất lớn. +- Lợi ích: Hai loại hộ gia đình có thể nhận trợ cấp tạm thời. Loại một là hộ vì sự cố bất ngờ, thành viên gia đình đột ngột mắc bệnh nặng, nhất thời không gánh nổi các chi phí cơ bản như ăn ở và khám bệnh. Loại hai là hộ nghèo mà chi tiêu bắt buộc đột ngột tăng, vượt khả năng của mình. Chuyện nào được cứu, cho bao nhiêu tiền, do chính quyền địa phương cấp huyện trở lên tự quyết định và công bố (chế độ toàn quốc, số tiền do địa phương quyết định, áp dụng từ 2014) +- Mức bằng chứng: B +- Nguồn: Quốc vụ viện (国务院) (2014, sửa đổi 2019). Biện pháp tạm thời về cứu trợ xã hội（国务院令第 649 号，第四十七、四十八、四十九条）. +- Ghi chú: Không yêu cầu bạn đã là hộ nghèo. Thường phát một lần, số tiền các nơi chênh lệch nhiều, lấy số địa phương công bố trong năm đó làm chuẩn. + +### 7. Thu nhập dưới mức trợ cấp người có thu nhập thấp địa phương thì đi xin trợ cấp + +- Chi phí: Không tốn tiền. Các thành viên gia đình sống chung nộp đơn bằng văn bản lên chính quyền xã, thị trấn hoặc ủy ban khu phố nơi đăng ký hộ khẩu. Tự làm bất tiện thì có thể ủy quyền cho ủy ban dân cư hoặc ủy ban khu phố làm thay. Phải chịu kiểm tra thu nhập và tài sản của gia đình. Thường đi 2 đến 3 lần, mang căn cước, sổ hộ khẩu và chứng minh thu nhập, tài sản. Khó ở chỗ phải bày hết của cải ra cho người kiểm tra, còn phải chờ niêm yết công khai. +- Nói thẳng: Tổng thu nhập cả nhà chia cho số người, dưới mức trợ cấp địa phương, tài sản trong nhà cũng đúng quy định, thì xin được. Cuối 2024, cả nước mức trợ cấp thành phố bình quân 798.1 tệ mỗi người mỗi tháng, nông thôn 593.9 tệ. Phải về nơi đăng ký hộ khẩu để xin. Có nhà, có xe, có tiền tiết kiệm, hoặc con cái có khả năng nuôi bạn, có thể không được duyệt. +- Lợi ích: Tổng thu nhập cả nhà chia cho số người, dưới mức trợ cấp địa phương, tài sản trong nhà cũng nằm trong quy định, thì nhận được trợ cấp. Cuối 2024, cả nước mức trợ cấp thành phố bình quân 798.1 tệ một người một tháng, nông thôn 593.9 tệ một người một tháng. Thành phố trợ cấp 625.0 vạn người, nông thôn trợ cấp 3361.5 vạn người. Mức trợ cấp tính như sau: lấy chi tiêu bình quân đầu người của cư dân địa phương trong năm trước, nhân với một tỷ lệ định lượng, sau một khoảng thời gian điều chỉnh lại một lần. Nói trắng ra là lấy số tiền bình quân mỗi người ở địa phương tiêu trong năm trước, trừ đi một phần rồi ra (bình quân cả nước, mỗi nơi một khác) +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2014, sửa đổi 2019). Biện pháp tạm thời về cứu trợ xã hội（国务院令第 649 号，第九、十、十一条）. ；Cục Thống kê Quốc gia (国家统计局) (2025). Các ngành xã hội phát triển tốt, phúc lợi người dân không ngừng tăng — báo cáo thống kê xã hội dân sinh kể từ「十四五」. ；Bộ Dân vụ và ba cơ quan khác (民政部等四部门) (2024). Ý kiến hướng dẫn về việc xác định và điều chỉnh chuẩn trợ cấp người có thu nhập thấp（民发〔2024〕16 号）. +- Ghi chú: Trợ cấp là tuyến giữ cuối cùng, phải về nơi đăng ký hộ khẩu để xin. Nhà, xe, tiền tiết kiệm trong nhà vượt chuẩn địa phương, hoặc con cái có khả năng nuôi bạn, đều có thể không được duyệt. Duyệt rồi còn có lợi khác: đóng bảo hiểm y tế cư dân được hỗ trợ (bảo hiểm y tế cư dân, xem mục 9), khám bệnh đi được trợ cấp y tế (trợ cấp y tế, xem mục 10), xin trợ cấp pháp lý không kiểm tra khó khăn tài chính (trợ cấp pháp lý, xem mục 3). + +### 8. Người có giấy chứng nhận người khuyết tật thì đi xin hai khoản trợ cấp cho người khuyết tật + +- Chi phí: Không tốn tiền. Mang giấy chứng nhận người khuyết tật, đến ủy ban khu phố hoặc xã, thị trấn nộp đơn một lần là được. +- Nói thẳng: Người khuyết tật trong hộ nghèo có thể nhận trợ cấp sinh hoạt. Mức khuyết tật là cấp nhất hoặc cấp nhị, lại cần chăm sóc lâu dài, thì nhận được trợ cấp chăm sóc. Hai khoản nhận cùng lúc. Chuẩn do từng tỉnh tự đặt, chênh lệch lớn, hỏi cụ thể ở ủy ban khu phố. Mang giấy chứng nhận người khuyết tật thế hệ hai, đến ủy ban khu phố hoặc xã, thị trấn làm một lần. +- Lợi ích: Quốc vụ viện ấn định hai khoản trợ cấp. Khoản thứ nhất là trợ cấp sinh hoạt cho người khuyết tật khó khăn, đối tượng ghi trong văn bản là「低保家庭中的残疾人，有条件的地方可逐步扩大到低收入残疾人及其他困难残疾人」— nghĩa là trước tiên bảo đảm người khuyết tật trong hộ nghèo, địa phương có dư lực thì mở rộng ra. Khoản thứ hai là trợ cấp chăm sóc cho người khuyết tật nặng, đối tượng ghi trong văn bản là「残疾等级被评定为一级、二级且需要长期照护的重度残疾人」— người khuyết tật nặng được đánh giá là cấp nhất hoặc cấp nhị và cần chăm sóc lâu dài. Hai khoản này nhận cùng lúc +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2015). Ý kiến về việc thiết lập toàn diện chế độ trợ cấp sinh hoạt cho người khuyết tật khó khăn và trợ cấp chăm sóc cho người khuyết tật nặng（国发〔2015〕52 号）. +- Ghi chú: Chuẩn do chính quyền tỉnh đặt, mỗi nơi chênh lệch lớn, mỗi tháng từ vài chục tệ đến vài trăm tệ. Khi xin mang theo giấy chứng nhận người khuyết tật thế hệ hai. Người nào bất tiện thì ủy quyền cho người nhà làm thay + +### 9. Đừng để bảo hiểm y tế cư dân bị đứt quãng mỗi năm 400 tệ, hộ khó khăn được miễn giảm + +- Chi phí: Phần bạn tự đóng, mỗi người mỗi năm không dưới 400 tệ. Đây là tiêu chuẩn quốc gia năm 2024、2025, mỗi nơi có thể cao hơn mức này. Phải đóng trong đúng khoảng thời gian thu phí tập trung cố định mỗi năm. Người đặc biệt khó khăn được hỗ trợ toàn bộ, khoản này do nhà nước trả hết. Đối tượng trợ cấp người có thu nhập thấp được hỗ trợ định mức, nhà nước trả một phần cố định. +- Nói thẳng: Cá nhân mỗi năm đóng từ 400 tệ, ngân sách còn hỗ trợ thêm theo đầu người chừng 700 tệ. Người đặc biệt khó khăn được hỗ trợ toàn bộ, đối tượng trợ cấp người có thu nhập thấp được hỗ trợ định mức. Đứt đóng là có cái giá: không tham gia trong kỳ tập trung, hoặc không đóng liên tục, thì sau khi đóng xong có 3 tháng khám bệnh không được chi trả, mỗi năm đứt thêm về nguyên tắc còn cộng thêm 1 tháng. +- Lợi ích: Ngoài phần tự đóng, ngân sách còn hỗ trợ thêm theo đầu người. Năm 2024 không dưới 670 tệ, năm 2025 cao hơn năm trước 30 tệ, không dưới 700 tệ, phần cá nhân vẫn không dưới 400 tệ (医保发〔2025〕22 号, theo văn bản của tỉnh chuyển tiếp). Trong báo cáo chính phủ năm 2026 nói sẽ tăng thêm 24 tệ. Đối tượng trợ cấp người có thu nhập thấp và người đặc biệt khó khăn, chi phí y tế đúng quy định, sau khi bảo hiểm y tế thanh toán phần còn lại, vẫn có thể đi trợ cấp y tế chi trả lần nữa với tỷ lệ không dưới 70%. Cái giá của việc đứt đóng là thời gian chờ. Không tham gia trong kỳ thu phí tập trung hoặc không đóng liên tục, sau khi đóng xong có 3 tháng thời gian chờ cố định, khoảng thời gian này khám bệnh không được chi trả. Mỗi năm đứt thêm, về nguyên tắc cộng thêm 1 tháng (áp dụng từ 2025, toàn quốc) +- Mức bằng chứng: A +- Nguồn: Cục Quản lý bảo hiểm y tế quốc gia (国家医保局) (2024). Giải thích chính sách về Thông báo liên quan đến công tác bảo hiểm y tế cơ bản cho cư dân đô thị và nông thôn năm 2024（医保发〔2024〕19 号）. ；Cục Bảo hiểm y tế tỉnh Hà Bắc và các cơ quan khác (河北省医疗保障局等) (2025). Thông báo về việc làm tốt công tác bảo hiểm y tế cơ bản cho cư dân đô thị và nông thôn năm 2025（冀医保发〔2025〕6 号，地方文件）. ；Cục Bảo hiểm y tế tỉnh An Hui và các cơ quan khác (安徽省医疗保障局等) (2025). Thông báo về việc làm tốt công tác bảo hiểm y tế cơ bản cho cư dân đô thị và nông thôn năm 2025（淮北市医保局转载）：「按照国家医保局、财政部、国家税务总局《关于做好2025年城乡居民基本医疗保障有关工作的通知》（医保发〔2025〕22号）要求」「居民医保人均财政补助标准较上年提高30元，达到每人每年不低于700元」. ；Văn phòng Quốc vụ viện (国务院办公厅) (2021). Ý kiến về việc hoàn thiện chế độ bảo hiểm y tế và trợ cấp cho bệnh hiểm nghèo nặng（国办发〔2021〕42 号）. ；Văn phòng Quốc vụ viện (国务院办公厅) (2024). Ý kiến hướng dẫn về việc hoàn thiện cơ chế lâu dài tham gia bảo hiểm y tế cơ bản（国办发〔2024〕38 号）. ；Cục Quản lý bảo hiểm y tế quốc gia (国家医保局) (2026). Báo cáo chính phủ năm 2026 nhắc đến các công tác bảo hiểm y tế này. +- Ghi chú: Thông báo năm 2025 của Cục Quản lý bảo hiểm y tế quốc gia（医保发〔2025〕22 号）không được đăng công khai trên trang của Cục và trang Chính phủ Trung Quốc. Hai con số 700 tệ và 400 tệ trùng nhau trong văn bản của Hà Bắc và An Hui, bản An Hui ghi rõ là theo thông báo này mà đặt. Trong thời gian lĩnh bảo hiểm thất nghiệp, bảo hiểm y tế do quỹ bảo hiểm thất nghiệp đóng thay, không cần tham gia trùng. + +### 10. Mắt bệnh nặng thì đi theo bảo hiểm y tế, bảo hiểm bệnh hiểm nghèo, trợ cấp y tế và đăng ký khám chữa bệnh ngoài địa phương, không đụng tới vay tiền trực tuyến + +- Chi phí: Không tốn tiền. Thứ tự là cố định, làm từng bước một. Bước một, trước khi nằm viện ở tỉnh khác, đăng ký qua ứng dụng nhỏ「国家异地就医备案」trong tài khoản WeChat chính thức của Cục Quản lý bảo hiểm y tế quốc gia（国家医保局）. Bước hai, khi ra viện thì quyết toán thẳng. Bước ba, sau khi bảo hiểm chi trả mà phần bạn tự bỏ vẫn quá nặng, xin trợ cấp y tế ở chính quyền xã hoặc văn phòng đường phố. Bước bốn, người còn đi làm thì xin nghỉ phép theo thời gian y tế; thời gian y tế là khoảng nghỉ việc vì bệnh mà không bị sa thải; thuộc tai nạn lao động thì đi theo bảo hiểm tai nạn lao động. Bước năm, không trả nổi thẻ tín dụng thì chủ động tìm ngân hàng thương lượng trả góp. Bước sáu, thật sự không đủ thì lên nền tảng quyên góp từng cá nhân do Bộ Dân vụ chỉ định để gây quỹ. Giấy tờ chủ yếu là giấy chẩn đoán, bảng chi phí, phiếu quyết toán, phải chạy 2 đến 4 lượt. Khó ở chỗ đang nằm bệnh mà vẫn phải đi từng lượt làm thủ tục. +- Nói thẳng: Thứ tự là cố định. Trước khi nằm viện ở tỉnh khác thì đăng ký trong tài khoản WeChat của Cục Quản lý bảo hiểm y tế quốc gia（国家医保局）, ra viện thì quyết toán thẳng, không phải tạm ứng toàn bộ trước. Sau khi bảo hiểm chi trả mà bạn vẫn không trả nổi, xin trợ cấp y tế ở chính quyền xã hoặc đường phố. Người còn đi làm có thời gian y tế từ 3 đến 24 tháng tính theo thâm niên, trong khoảng đó công ty không được sa thải bạn. Đừng đụng tới vay tiền trực tuyến và vay ngoài lãi cao. +- Lợi ích: Nhà nước gộp bảo hiểm y tế cơ bản, bảo hiểm bệnh hiểm nghèo và trợ cấp y tế gọi là「三重制度」, từng tầng đỡ cho tầng dưới. Người thuộc diện trợ cấp người có thu nhập thấp và người khó khăn đặc biệt, chi phí y tế đúng quy định được trợ cấp không dưới 70%. Người bệnh nặng vốn không nghèo, chữa bệnh rồi mới nghèo, có thể tự nêu đơn xin trợ cấp. Đã đăng ký trước khi nằm viện ở tỉnh khác thì ra viện quyết toán thẳng, không phải tạm ứng toàn bộ rồi quay về nơi tham gia bảo hiểm để hoàn. Nghỉ việc vì bệnh có thời gian y tế từ 3 đến 24 tháng, cụ thể bao lâu tùy thâm niên. Trong khoảng đó công ty không được chấm dứt hợp đồng lao động, tiền lương nghỉ bệnh không được thấp hơn 80% mức lương tối thiểu địa phương. Công nhân bị tai nạn lao động có thời gian giữ lương điều trị, tức khoảng chữa thương không đi làm, lương và phúc lợi cũ vẫn được trả, công ty trả theo tháng, thường không quá 12 tháng. Nợ thẻ tín dụng vượt khả năng trả nợ mà bạn vẫn muốn trả, có thể thương lượng với ngân hàng một khoản trả góp theo điều chỉnh riêng, xếp lại kế hoạch trả nợ, dài nhất 5 năm, đã ký thỏa thuận thì ngân hàng phải ngừng đòi. Nền tảng quyên góp từng cá nhân hỗ trợ người góp tiền trên mạng phải do cơ quan dân vụ của Quốc vụ viện chỉ định. Nền tảng phải xác minh thông tin nhờ cứu có thật hay không. Tiền quyên góp chỉ được chuyển cho chính người nhờ cứu hoặc tài khoản do người đó cung cấp (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Văn phòng Quốc vụ viện (国务院办公厅) (2021). Ý kiến về việc hoàn thiện chế độ bảo hiểm và trợ cấp y tế cho bệnh hiểm nghèo nặng（关于健全重特大疾病医疗保险和救助制度的意见）（国办发〔2021〕42 号）. ；Cục Quản lý bảo hiểm y tế quốc gia (国家医保局) (2025). Khám chữa bệnh ngoài địa phương ở tỉnh khác thì hoàn thế nào? ；Bộ Lao động (劳动部) (1994). Quy định thời gian y tế cho công nhân doanh nghiệp mắc bệnh hoặc bị thương không do lao động（企业职工患病或非因工负伤医疗期规定）（劳部发〔1994〕479 号，第二、三条）. ；Bộ Lao động (劳动部) (1995). Ý kiến về việc thực hiện Luật Lao động（关于贯彻执行《劳动法》若干问题的意见）（劳部发〔1995〕309 号，mục 59）. （Sở Nhân sự và Xã hội tỉnh Cát Lâm đăng lại）；Quốc vụ viện (国务院) (2010, sửa đổi). Quy định bảo hiểm tai nạn lao động（工伤保险条例）（国务院令第 586 号，第三十、三十三条）. ；Ủy ban Giám sát ngân hàng (银监会) (2011). Biện pháp quản lý hoạt động thẻ tín dụng của ngân hàng thương mại（商业银行信用卡业务监督管理办法）（银监会令 2011 年第 2 号，第七十条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2023). Quyết định sửa đổi Luật Từ thiện（关于修改慈善法的决定）（第一百二十四条）. ；Bộ Dân vụ và bốn cơ quan khác (民政部等五部门) (2024). Biện pháp quản lý nền tảng dịch vụ nhờ cứu cá nhân qua mạng（个人求助网络服务平台管理办法）（令第 75 号，第三、十一、十三、十七条）. +- Ghi chú: Tỷ lệ trợ cấp y tế bao nhiêu, mỗi năm tối đa bao nhiêu, do nơi bạn tham gia bảo hiểm tự quy định. Thời gian y tế chia sáu bậc, tuỳ bạn đã làm việc tổng cộng bao nhiêu năm và ở đơn vị này bao lâu. Hết thời gian y tế thì công ty có thể chấm dứt hợp đồng một cách hợp pháp, nên xem trước mình đang ở bậc nào. Người cần cấp cứu cũng nằm trong danh sách phải đăng ký. Trang của Cục Quản lý bảo hiểm y tế không nói cấp cứu có được tính là đăng ký tự động không, có đăng ký bổ sung sau được không, nên gọi điện hỏi nơi tham gia bảo hiểm. Đừng đụng tới vay tiền trực tuyến và vay ngoài lãi cao. Tòa án chỉ bảo vệ tới mức bằng 4 lần LPR kỳ hạn một năm (không vay lãi cao, xem mục 15). Phần vượt quá mức đó không được bảo vệ, khoản gốc vẫn phải trả. Nằm trên giường bệnh cũng không có sức để đánh kiện kiểu đó. Bệnh nặng cần cấp cứu là chuyện khác: không có tiền vẫn phải gọi 120, bệnh viện không được từ chối, đẩy hoặc chậm trễ cứu chữa, phần cấp cứu đó do quỹ cứu hộ khẩn cấp bệnh tật chi trả, xem mục 9 ở phần 24 (cấp cứu cũng phải cứu trước). Trang gốc của văn bản 劳部发〔1995〕309 号 không mở được trên cổng Bộ Nhân sự và Xã hội, nội dung lấy từ trang sao chép của Sở Nhân sự và Xã hội tỉnh. + +### 11. Mất thẻ căn cước thì đi làm lại ngay, gấp thì xin thẻ tạm trước + +- Chi phí: Phải nộp phí cấp giấy tờ. Đây là khoản thu theo quy định, thu bao nhiêu do nhà nước quy định, theo mức niêm yết công khai địa phương. Nộp đơn tại công an xã, thị trấn nơi bạn đăng ký hộ khẩu thường trú, chạy 1 lượt. +- Nói thẳng: Thẻ làm lại, tính từ ngày nộp đơn thì cấp giấy trong 60 ngày. Gấp thì tại chỗ xin thêm một thẻ căn cước tạm, cấp trong 3 ngày, giá trị 3 tháng. Không có thẻ căn cước thì tìm việc, thuê chỗ ở, mở tài khoản, nhận trợ cấp đều không làm được, nó là điều kiện tiên quyết của mọi thủ tục. +- Lợi ích: Tính từ ngày bạn nộp《居民身份证申领登记表》, cơ quan công an phải cấp giấy trong 60 ngày. Trong khoảng thời gian cấp mới, cấp đổi hoặc cấp lại, nếu gấp dùng giấy thì có thể xin một thẻ căn cước tạm. Cơ quan công an phải cấp trong 3 ngày kể từ ngày nhận đơn, giá trị 3 tháng (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2003). Luật Thẻ căn cước cư dân（居民身份证法）（主席令第四号，第十二、二十条）. ；Bộ Công an (公安部) (2005). Biện pháp quản lý thẻ căn cước tạm của cư dân（临时居民身份证管理办法）（公安部令第 78 号，第二、七、九、十二、十七条）. +- Ghi chú: Không có thẻ căn cước thì tìm việc, thuê chỗ ở, mở tài khoản, nhận trợ cấp đều không làm được, đây là điều kiện tiên quyết của mọi thủ tục. Luật Thẻ căn cước cư dân có sửa đổi năm 2011, mục này dẫn bản công bố năm 2003, điều khoản phải theo bản đang hiệu hành. Có làm ở nơi khác được không thì theo thông báo của công an địa phương. + +### 12. Làm xong đăng ký thất nghiệp thì hãy xin công nhận là người khó khăn việc làm, để nhận trợ cấp bảo hiểm xã hội hoặc làm công việc công ích + +- Chi phí: Không tốn tiền. Nộp đơn xin công nhận ở cơ quan dịch vụ việc làm công cộng cấp đường phố hoặc cộng đồng. Điều kiện công nhận mỗi nơi khác nhau, thường gặp là tuổi đã lớn, thất nghiệp lâu ngày, thuộc hộ có thu nhập thấp, người khuyết tật, hộ không có ai có việc làm. Phải làm xong đăng ký thất nghiệp thì mới xin được. Khó ở chỗ phải tự đối chiếu từng điều kiện một, và tiền bảo hiểm xã hội còn phải tạm ứng trước. +- Nói thẳng: Được công nhận rồi, bạn tự đóng bảo hiểm xã hội thì nhận được trợ cấp, nguyên tắc không quá hai phần ba mức bạn thật sự đóng, dài nhất 3 năm. Được đơn vị tuyển dụng hoặc được bố trí vào công việc công ích thì phần bảo hiểm xã hội do đơn vị đóng cũng có trợ cấp. Lương của công việc công ích tham chiếu lương tối thiểu địa phương. Loại công việc này tiền không nhiều, nhưng có bảo hiểm xã hội, hợp để làm cầu nối. Trợ cấp bảo hiểm xã hội phần lớn đóng trước, hoàn sau, nên phải tạm ứng tiền trước. +- Lợi ích: Được công nhận rồi, người không có đơn vị, tự đóng bảo hiểm xã hội (gọi là việc làm tự do) thì nhận được trợ cấp bảo hiểm xã hội. Mức trợ cấp nguyên tắc không quá 2/3 mức bạn thật sự đóng, dài nhất 3 năm. Được đơn vị tuyển dụng hoặc được bố trí vào công việc công ích thì đơn vị nhận trợ cấp theo số tiền bảo hiểm hưu trí, y tế, thất nghiệp đã đóng thật. Công việc công ích là công việc do chính quyền bỏ tiền ra mở, trợ cấp của nó tham chiếu lương tối thiểu địa phương. Công việc công ích do chính quyền mở thì ưu tiên bố trí người khó khăn việc làm. Ngoài ra còn có trợ cấp tìm việc một lần, phát cho sinh viên tốt nghiệp. Phạm vi là sinh viên tốt nghiệp thuộc hộ có thu nhập thấp, hộ không có ai có việc làm, hoặc người khó khăn đặc biệt. Sinh viên khuyết tật, sinh viên từng nhận vay vốn học bổng quốc gia cũng tính. Khoản tiền này nhận trong năm học cuối cùng (khung toàn quốc, điều kiện công nhận và mức tiền do địa phương quy định) +- Mức bằng chứng: B +- Nguồn: Bộ Tài chính và Bộ Nhân sự và Xã hội (财政部、人力资源社会保障部) (2023, sửa đổi). Biện pháp quản lý quỹ trợ cấp việc làm（就业补助资金管理办法）（财社〔2017〕164 号，2023 年 12 月修订印发）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007, sửa đổi 2015). Luật thúc đẩy việc làm（就业促进法）（第五十二、五十三条）. +- Ghi chú: Lương công việc công ích không cao, nhưng có bảo hiểm xã hội, hợp để làm cầu nối. Trợ cấp bảo hiểm xã hội phần lớn đóng trước hoàn sau, phải tạm ứng tiền trước. Điều kiện cụ thể hỏi 12333 hoặc đường phố. + +### 13. Trong thời gian thất nghiệp thì đi nhận trợ cấp đào tạo nghề, trợ cấp thực tập việc làm và trợ cấp bảo hiểm xã hội, đừng tự bỏ tiền học khóa + +- Chi phí: Không tốn tiền. Bản thân việc đào tạo đã chiếm không ít thời gian của bạn. Khó ở chỗ phải học liên tục hết khóa, đỗ thi lấy chứng chỉ thì mới được trợ cấp. +- Nói thẳng: Đào tạo không phải tự bỏ tiền. Học và có chứng chỉ theo quy định thì nhận được trợ cấp đào tạo nghề, mỗi người nhiều nhất 3 lần. Sinh viên tốt nghiệp chưa tìm được việc trong vòng hai năm sau khi ra trường, và thanh niên thất nghiệp từ 16 đến 24 tuổi, có thể tham gia thực tập việc làm, trong thời gian thực tập nhận trợ cấp sinh hoạt cơ bản. Trước hết vào cổng Bộ Nhân sự và Xã hội địa phương tra「职业培训补贴目录」, đừng tự bỏ tiền đăng ký khóa. +- Lợi ích: Quy chế của Bộ Tài chính cùng Bộ Nhân sự và Xã hội liệt kê một loạn khoản quỹ trợ cấp việc làm, đó là khoản ngân quỹ dành riêng để trợ cấp việc làm. Học đào tạo và có chứng chỉ đúng quy định thì được trợ cấp đào tạo nghề, mỗi người nhiều nhất 3 lần. Trợ cấp cho người khó khăn việc làm tự đóng bảo hiểm xã hội (việc làm tự do) dài nhất 3 năm; cách tuổi về hưu theo pháp luật chưa đầy 5 năm thì kéo dài đến khi về hưu. Trợ cấp cho sinh viên tốt nghiệp tự đóng bảo hiểm xã hội dài nhất 2 năm, được doanh nghiệp tuyển thì dài nhất 1 năm. Sinh viên tốt nghiệp chưa tìm được việc trong vòng 2 năm sau khi ra trường, và thanh niên thất nghiệp từ 16 đến 24 tuổi, có thể tham gia thực tập việc làm, trong thời gian thực tập nhận trợ cấp sinh hoạt cơ bản +- Mức bằng chứng: A +- Nguồn: Bộ Tài chính và Bộ Nhân sự và Xã hội (财政部、人力资源社会保障部) (2023). Biện pháp quản lý quỹ trợ cấp việc làm（就业补助资金管理办法）（第五条、第七条、第十条）. +- Ghi chú: Danh mục và mức trợ cấp cụ thể do từng tỉnh quy định, trước hết tra「职业培训补贴目录」trên cổng Bộ Nhân sự và Xã hội địa phương. Sinh viên tốt nghiệp thuộc hộ có thu nhập thấp, hộ không có ai có việc làm, còn sinh viên đang nợ vay vốn học bổng hoặc sinh viên khuyết tật, có trợ cấp tìm việc, khởi nghiệp một lần. + +### 14. Tìm việc đừng chỉ nộp hồ sơ, hãy dùng cách có hệ thống: học kỹ thuật, đặt mục tiêu, nhờ người khác giúp + +- Chi phí: Không tốn tiền. Mỗi ngày phải dành 1 đến 2 giờ, làm liên tục mấy tuần. Khóa học có phí thì không cần đăng ký, thay bằng tư vấn nghề miễn phí của cơ quan dịch vụ việc làm công cộng. Khó ở chỗ phải duy trì hằng ngày khi không ai giám sát, lại còn phải chủ động mở lời nhờ người khác giúp. +- Nói thẳng: Người từng tham gia tư vấn tìm việc có xác suất tìm được việc gấp 2.67 lần người không tham gia. Cái thật sự có tác dụng gồm mấy thứ: học kỹ thuật tìm việc cụ thể, trình bày rõ về bản thân, đặt mục tiêu rõ ràng, chủ động tiếp cận, mở lời nhờ người khác giúp. Nhờ người khác giúp một cách cụ thể là nhờ người quen giới thiệu nội bộ. +- Lợi ích: Gộp 47 nghiên cứu tư vấn tìm việc có đối chứng lại để tính. Các nghiên cứu này là thử nghiệm hoặc thiết kế bán thực nghiệm, đều chia nhóm để so sánh. Kết quả là nhóm được tư vấn có xác suất tìm được việc gấp 2.67 lần nhóm không được tư vấn (OR 2.67, tức là tỷ số xác suất). Thành phần có tác dụng trong tư vấn gồm mấy thứ. Một là dạy kỹ thuật tìm việc. Hai là cải thiện cách tự thể hiện (trình bày rõ về bản thân). Ba là nâng cao năng lực bản thân (tin rằng mình làm được). Bốn là khuyến khích chủ động tiếp cận. Năm là đặt mục tiêu. Sáu là tìm hỗ trợ xã hội (mở lời nhờ người khác giúp). Tư vấn vừa dạy kỹ năng vừa tạo động lực thì hiệu quả lớn nhất +- Mức bằng chứng: A +- Nguồn: Liu S, Huang JL, Wang M (2014). Effectiveness of job search interventions: a meta-analytic review. Psychological Bulletin 140(4): 1009–1041. +- Ghi chú: Phần lớn các nghiên cứu này làm ở châu Âu và châu Mỹ, dạng là các khóa nhóm hoàn chỉnh. Chuyển sang Trung Quốc là mang kết quả từ nơi khác áp vào, không phải đo trong nước. Tìm hỗ trợ xã hội một cách cụ thể là nhờ người quen giới thiệu và nội bộ, đừng ngượng ngùng không dám mở lời. +### 15. Không nộp tiền cọc, không giữ giấy tờ, không ký「培训贷」, không vào đa cấp, không vay lãi cao + +- Chi phí: Không tốn tiền. Gặp ba kiểu lời mời「nộp tiền trước rồi lên ca」「đào tạo trước rồi xếp việc」「kéo người khác vào」 thì từ chối ngay tại chỗ. Giữ lại lịch sử chat và chứng từ chuyển khoản. Khó ở chỗ càng thiếu tiền càng muốn đánh cược, nói「không」 ngay tại chỗ không phải dễ. +- Nói thẳng: Đơn vị sử dụng lao động không được giữ thẻ căn cước của bạn, không được thu tiền cọc, cũng không được đổi tên khác để thu tiền. Môi giới không giới thiệu được việc cho bạn thì phải hoàn lại phí môi giới. 培训贷 là khoản phải nộp tiền đào tạo trước rồi mới lên ca, đa cấp là kiểu tính tiền theo số người kéo vào, cả hai đều là hình thức lừa đảo được nhà nước nêu đích danh. Phần lãi vay vượt bốn lần LPR kỳ hạn một năm thì tòa án không hỗ trợ, nhưng khoản gốc vẫn phải trả. +- Lợi ích: Công ty không được giữ thẻ căn cước và các giấy tờ khác của bạn, không được bắt bạn tìm người bảo lãnh, cũng không được đổi tên khác để thu tiền hay vật. Vi phạm thì bị buộc trả lại, kèm phạt từ 500 tệ đến 2000 tệ cho mỗi người. Môi giới việc làm cũng không được giữ giấy tờ, không được thu tiền cọc, không giới thiệu được việc thì phải hoàn phí môi giới. Giả danh tuyển dụng để dụ dỗ bạn nộp tiền đào tạo trước hoặc vay tiền trước, đó là kiểu lừa đảo「招转培」「培训贷」mà Bộ Nhân sự và Xã hội cùng bốn cơ quan khác đã cảnh báo đích danh. Kiểu tính hoa hồng theo số người kéo vào, phải nộp tiền mới được tham gia thì thuộc đa cấp, người tham gia có thể bị phạt dưới 2000 tệ. Phần lãi vay ngoài vượt quá 4 lần LPR kỳ hạn một năm tính tại thời điểm ký hợp đồng thì tòa án không hỗ trợ. Quy tắc mới này thay thế cách cũ chia ba đoạn theo hai mức 24%, 36% (toàn quốc, áp dụng từ tháng 8 năm 2020) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007). Luật Hợp đồng lao động（劳动合同法）（主席令第六十五号，第九、八十四条）. ；Bộ Lao động và Bảo hiểm xã hội (劳动保障部) (2007). Quy định về dịch vụ việc làm và tuyển dụng việc làm（就业服务与就业管理规定）（劳动保障部令第 28 号，第十四、五十五、五十八条）. ；Bộ Nhân sự và Xã hội cùng bốn cơ quan khác (人力资源社会保障部等五部门) (2026). Cảnh báo nguy cơ「招转培」「培训贷」và các hình thức tương tự. ；Quốc vụ viện (国务院) (2005). Quy định cấm đa cấp（禁止传销条例）（国务院令第 444 号，第二、七、二十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Quyết định sửa đổi Quy định về một số vấn đề pháp lý khi xét vụ án vay ngoài（关于修改《关于审理民间借贷案件适用法律若干问题的规定》的决定）（法释〔2020〕6 号，第二十六条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Điều chỉnh trần lãi vay ngoài theo giới hạn bảo vệ của tòa thành「4 lần LPR kỳ hạn một năm」. +- Ghi chú: Mức 4 lần LPR là trần lãi mà tòa án sẵn sàng bảo vệ, không phải「vay rồi không phải trả」, khoản gốc vẫn phải trả. Bản giải thích tư pháp này tháng 12 năm 2020 lại sửa một lần, thứ tự điều khoản có thay đổi, phải theo bản đang hiệu hành. Càng gấp dùng tiền thì càng dễ bị những chiêu này nhắm vào. Trước hết hãy đi những đường chính thống ở đầu phần này: trợ cấp bảo hiểm thất nghiệp, khiếu nại thanh tra lao động, trợ giúp pháp lý, trạm cứu hộ, trợ cấp tạm thời. + +### 16. Khó ở nhà thì xếp hàng chờ nhà ở công lập trước + +- Chi phí: Nộp đơn không thu phí. Nộp hồ sơ ở cơ quan bảo đảm nhà ở cấp thành phố, huyện, đồng thời bằng văn bản đồng ý cho họ kiểm tra những gì bạn khai. Thiếu giấy tờ thì họ nói một lượt cần bổ sung mấy thứ. Thời hạn xếp hàng chờ nhà gọi là thời gian chờ, thường không quá 5 năm. +- Nói thẳng: Phải đồng thời thoả ba điều kiện: ở địa phương không có nhà, hoặc diện tích nhà thấp hơn tiêu chuẩn; thu nhập và tài sản thấp hơn tiêu chuẩn; người lao động từ nơi khác đến còn phải có việc làm ổn định tại địa phương đủ số năm quy định. Thời gian chờ do địa phương công bố. Đây là nước xa, không cứu được ngọn lửa gần, nhưng xếp hàng không tốn tiền, đăng ký sớm thì được đến sớm. +- Lợi ích: Đủ cả ba điều kiện thì được nộp đơn. Một là ở địa phương không có nhà ở, hoặc diện tích nhà thấp hơn tiêu chuẩn quy định. Hai là thu nhập và tài sản thấp hơn tiêu chuẩn quy định. Ba là người lao động từ nơi khác đến phải có việc làm ổn định tại địa phương đủ số năm quy định. Điều kiện cụ thể do cơ quan bảo đảm nhà ở của các thành phố trực thuộc trung ương và của thành phố, huyện quy định rồi công bố. Người đã ghi tên vào danh sách chờ được gọi là đối tượng chờ, cơ quan bảo đảm nhà ở phải sắp xếp nguồn nhà cho họ trong thời gian chờ (khung toàn quốc, ngưỡng và tiền thuê do địa phương quy định, áp dụng từ 2012) +- Mức bằng chứng: B +- Nguồn: Bộ Xây dựng và Nhà ở (住房和城乡建设部) (2012). Biện pháp quản lý nhà ở công lập cho thuê（公共租赁住房管理办法）（住房和城乡建设部令第 11 号，第七、八、十条）. +- Ghi chú: Đây là nước xa, không cứu được ngọn lửa gần, nhưng xếp hàng không tốn tiền, đăng ký sớm thì được đến sớm. Các nơi còn có loại khác như nhà ở cho thuê có bảo đảm, quy tắc khác nhau, vào trang của Sở Xây dựng và Nhà ở địa phương tra. + +### 17. Ép chi phí ở và ăn xuống mức thấp nhất: thuê chung theo tháng thay vì thuê theo ngày, tự nấu ăn và bếp ăn trợ giúp thay cho đồ ăn giao + +- Chi phí: Tìm nhà phải mất 1 đến 3 ngày. Tiền cọc thường bằng 1 tháng tiền thuê (ước tính sơ bộ của tác giả). Tự nấu ăn thì mỗi ngày chiếm 30 đến 60 phút. Khó ở chỗ phải tự làm ngày này qua ngày khác, kiềm chế không gọi đồ ăn. +- Nói thẳng: Trong chi tiêu bình quân đầu người cả nước năm 2025, ăn chiếm 29.3%, ở chiếm 21.7%, hai khoản cộng lại quá nửa, nên ép hai khoản này là hiệu quả nhất. Theo kinh nghiệm của tác giả, nhà thuê theo ngày và nhà thuê theo giờ quy ra cả tháng thì đắt hơn thuê chung; tự nấu ăn rẻ hơn gọi đồ ăn. Khi tìm việc phải hỏi rõ có bao ăn ở không, việc mà bao ăn ở thì coi như xoá sạch hai khoản lớn kia. +- Lợi ích: Năm 2025, dân cư toàn quốc bình quân mỗi người mỗi năm chi 29476 tệ (gọi là chi tiêu bình quân đầu người). Trong đó ăn uống 8631 tệ, chiếm 29.3%. Ở ở 6397 tệ, chiếm 21.7%. Hai khoản cộng lại quá nửa, nên ép hai khoản này là hiệu quả nhất. Dịch vụ bếp ăn trợ giúp cho người cao niên mở cho toàn thể người cao niên, hình thức gồm bếp ăn người cao niên, bàn ăn người cao niên, điểm cấp cơm; người cao niên khó khăn có khoản trợ cấp khác nhau. Làm chỗ dẫn: tính đến ngày 1 tháng 1 năm 2026, mức lương tối thiểu tháng bậc một thấp nhất là ở Thanh Hải 2080 tệ, cao nhất là ở Thượng Hải 2740 tệ (toàn quốc, 2025 đến 2026) +- Mức bằng chứng: B +- Nguồn: Cục Thống kê Quốc gia (国家统计局) (2026). Tình hình thu nhập và chi tiêu của dân cư năm 2025. ；Bộ Dân vụ và mười cơ quan khác (民政部等十一部门) (2023). Phương án hành động tích cực phát triển dịch vụ bếp ăn trợ giúp cho người cao niên（积极发展老年助餐服务行动方案）（民发〔2023〕58 号）. ；Sở Nhân sự và Xã hội tỉnh Tứ Xuyên (四川省人力资源和社会保障厅) (2026). Mức lương tối thiểu ở các tỉnh, vùng tự trị, thành phố trực thuộc trung ương trên toàn quốc（截至 2026 年 1 月 1 日，转载人社部官网表）. +- Ghi chú: Nhà thuê theo ngày và nhà thuê theo giờ quy ra tháng thì đắt hơn thuê chung rất nhiều, đây là đánh giá theo kinh nghiệm, không có số liệu chính thức. Điểm cấp cơm cho người cao niên có mở cho người không cao niên không, giá bao nhiêu, mỗi nơi khác nhau. Khi tìm việc phải hỏi rõ có bao ăn ở không, việc bao ăn ở thì xoá thẳng hai khoản lớn kia. Trang bảng gốc của Bộ Nhân sự và Xã hội không mở được, số liệu lấy từ trang sao chép của Sở Nhân sự và Xã hội tỉnh Tứ Xuyên. + +### 18. Bảo hiểm xã hội đứt quãng thì đừng hoảng: lương hưu tính theo số năm tích luỹ, bảo hiểm y tế nộp bù theo quy định + +- Chi phí: Không tốn tiền. Bảo hiểm hưu trí cư dân và bảo hiểm y tế cư dân có thể tự đóng theo năm. Bảo hiểm xã hội của người lao động bị đứt thì không cần bỏ tiền tìm người đứng tên hộ cho「đừng đứt kỳ」. +- Nói thẳng: Lương hưu tính theo số năm tích luỹ, không tính theo số năm liên tục, đứt giữa chừng không bị xoá, đến tuổi về hưu mà tích luỹ đủ 15 năm thì nhận hằng tháng. Bảo hiểm y tế người lao động đã đóng đủ số năm thì sau khi về hưu không phải đóng nữa, số năm mỗi nơi khác nhau, gọi 12393 hỏi nơi tham gia bảo hiểm. Thật sự bị ảnh hưởng là những điều kiện đòi hỏi đóng liên tục như mua nhà, lập hộ tịch, điểm cư dân. +- Lợi ích: Bảo hiểm hưu trí cơ bản tính theo tổng số năm đã đóng, đứt giữa chừng không bị xoá. Đến tuổi về hưu theo pháp luật, tích luỹ đủ 15 năm thì nhận lương hưu hằng tháng. Chưa đủ thì tiếp tục đóng cho tới khi đủ 15 năm. Người đổi chỗ làm việc (cách nói chính thức là khác vùng điều hành) thì quan hệ bảo hiểm hưu trí đi theo người, số năm đóng tiếp tục cộng dồn. Bảo hiểm y tế người lao động đã đóng tới số năm quốc gia quy định thì sau khi về hưu không phải đóng nữa. Cái giá của việc đứt bảo hiểm y tế cư dân là thời gian chờ: sau khi đóng lại có một khoảng thời gian khám bệnh không được chi trả (bảo hiểm y tế cư dân, xem mục 9) (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2010, sửa đổi 2018). Luật Bảo hiểm xã hội（社会保险法）（第十六、十九、二十七条）. ；Văn phòng Quốc vụ viện (国务院办公厅) (2024). Ý kiến hướng dẫn về việc hoàn thiện cơ chế lâu dài tham gia bảo hiểm y tế cơ bản（关于健全基本医疗保险参保长效机制的指导意见）（国办发〔2024〕38 号）. +- Ghi chú: Số năm bảo hiểm y tế người lao động không phải đóng nữa sau khi về hưu, luật viết là「quy định của nhà nước」, số mỗi nơi thực hiện khác nhau, nam nữ cũng thường khác nhau, gọi 12393 hoặc hỏi Cục Quản lý bảo hiểm y tế nơi tham gia. Đứt đóng chủ yếu ảnh hưởng các điều kiện ở địa phương đòi hỏi đóng liên tục đủ số tháng, như mua nhà, lập hộ tịch, điểm cư dân, các quy tắc này do từng nơi tự quy định, tra ở địa phương. Bỏ tiền tìm người đứng tên hộ đóng hộ, tiền có thể mất trắng, cũng có rủi ro pháp lý, không nên làm. + +### 19. Từng vào tù, từng phá sản, từng bị đưa vào danh sách không thi hành án, pháp luật vẫn còn đường bắt đầu lại, trước hết làm hết thủ tục + +- Chi phí: Không thu phí. Người vừa hết án tù thì đến văn phòng tư pháp hoặc đường phố nơi đăng ký hộ khẩu hay nơi đang ở báo cáo, xin điều trợ cấp tổ chức lại cuộc sống (việc chính quyền sắp xếp và giúp đỡ). Người đã bị đưa vào danh sách người bị thi hành án không thi hành thì trả xong phần phải trả, rồi xin tòa án thi hành án xoá tên. Cá nhân tham gia bảo hiểm đủ 3 năm ở Thâm Quyến thì có thể xin phá sản cá nhân. Mỗi bước đều phải chuẩn bị giấy tờ, trình bày đối diện, rồi còn có thời gian quan sát 1 đến 3 năm. Khó ở chỗ phải kể lại chuyện đã qua nhiều lần, và thời gian quan sát vẫn phải chịu. +- Nói thẳng: Người vừa hết án tù không bị phân biệt đối xử về việc làm, học tập, bảo hiểm xã hội, người khó khăn về sinh hoạt thì do chính quyền giúp sắp xếp. Người khi phạm tội chưa đầy 18 tuổi, bị phán không quá 5 năm thì lý lịch được niêm phong, khi nhập quân hay tìm việc không phải khai. Thời hạn trong danh sách không thi hành án là 2 năm, trả xong tiền thì tòa án phải xoá tên trong 3 ngày làm việc. Tiền án của người thành niên không niêm phong, khi nhập quân hay xin việc phải khai trung thực. +- Lợi ích: Người vừa hết án tù theo pháp luật có quyền ngang bằng công dân khác, không bị phân biệt đối xử về việc làm, học tập và hưởng bảo hiểm xã hội. Người khó khăn về sinh hoạt do chính quyền nơi đăng ký hộ khẩu hay nơi đang ở giúp sắp xếp, cứu giúp và hỗ trợ theo pháp luật (Luật Cải tạo nhà tù sửa đổi năm 2026, áp dụng từ ngày 1 tháng 11). Văn bản năm 2010 của Trung ương yêu cầu tăng cường đào tạo nghề cho họ, đưa người đủ điều kiện vào diện trợ cấp người có thu nhập thấp, và lập các cơ sở tạm trú chuyển tiếp. Người khi phạm tội chưa đầy 18 tuổi và bị phán không quá 5 năm thì lý lịch phạm tội được niêm phong, khi nhập quân hay tìm việc không phải khai; người đó tự xin giấy chứng nhận không có tiền án thì được cấp trong 3 ngày làm việc. Thời hạn bị đưa vào danh sách không thi hành án là 2 năm. Trả xong tiền, hoặc đã hoà giải với bên kia và các trường hợp tương tự, thì tòa án phải xoá tên trong 3 ngày làm việc. Quy định về phá sản cá nhân của Khu kinh tế đặc biệt Thâm Quyến là văn bản phá sản cá nhân duy nhất hiện có, tuyên bố phá sản rồi chịu đủ 3 năm thời gian quan sát thì phần nợ chưa trả có thể được xoá. Doanh nghiệp pháp nhân không trả được nợ đến hạn, lại nợ nhiều hơn tài sản có (gọi là vốn không đủ nợ), thì có thể xin tái cơ cấu, hoà giải hoặc tuyên bố phá sản thanh toán, các khoản nợ chấm dứt cùng doanh nghiệp này (toàn quốc, phá sản cá nhân chỉ có ở Thâm Quyến) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2026). Luật Cải tạo nhà tù（监狱法）（2026 年修订，第五十九、六十、六十二条）. ；Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Quốc vụ viện (中共中央办公厅、国务院办公厅) (2010). Thông báo chuyển tải Ý kiến của Ủy ban Trung ương xây dựng và quản lý an ninh xã hội về việc tăng cường công tác tổ chức lại cuộc sống và giúp đỡ cho người vừa hết án tù và được giải tội（转发《中央社会治安综合治理委员会关于进一步加强刑满释放解除劳教人员安置帮教工作的意见》的通知）（中办发〔2010〕5 号）. （Chính quyền quận Tư Dương, thành phố Dương Châu đăng lại）；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2018). Luật Tố tụng hình sự（刑事诉讼法）（2018 年修正，第二百八十六条）. ；Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao, Bộ Công an và Bộ Tư pháp (2022). Biện pháp thi hành niêm phong lý lịch phạm tội của người chưa thành niên（关于未成年人犯罪记录封存的实施办法）（第九、十五条）. （Sở Công an tỉnh Hồ Nam đăng lại）；Tòa án nhân dân tối cao (最高人民法院) (2017, sửa đổi). Một số quy định về công bố thông tin người bị thi hành án không thi hành（关于公布失信被执行人名单信息的若干规定）（第二、十条）. ；Ủy ban Thường vụ Đại biểu Nhân dân thành phố Thâm Quyến (深圳市人大常委会) (2020). Quy định về phá sản cá nhân của Khu kinh tế đặc biệt Thâm Quyến（深圳经济特区个人破产条例）（第二、二十三、九十五、九十七、一百零三条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2006). Luật Phá sản doanh nghiệp（企业破产法）（主席令第五十四号，第二、七条）. +- Ghi chú: Tiền án của người thành niên không niêm phong. Điều 100 của Bộ luật Hình sự đòi hỏi phải khai trung thực việc mình đã bị xử phạt hình sự khi nhập quân hay tìm việc. Các nghề công chức, thẩm phán, luật sư, giáo viên và việc thẩm tra lý lịch chính trị có hạn chế riêng theo pháp luật. Bản đang hiệu hành của Luật Cải tạo nhà tù trước ngày 1 tháng 11 năm 2026 tương ứng với điều 37, 38, số điều lấy theo bản có hiệu lực. Văn bản 中办发〔2010〕5 号 không tìm thấy trang gốc trên cổng Chính phủ trung ương, lấy từ trang sao chép của chính quyền quận. Trong thời gian quan sát của phá sản cá nhân thì không được đi hạng nhất máy bay, không được tiêu vào cao, không được mua nhà; các khoản nợ như bồi thường do cố ý xâm phạm thân thể người khác thuộc 8 loại không được xoá; phần miễn trừ giành được bằng gian lận còn có thể bị thu hồi. Được xoá khỏi danh sách không thi hành án không có nghĩa lịch sử tín dụng cũng sạch theo, đó là hai hệ thống khác nhau. +### 20. Trước khi mắc bệnh nặng, mua thêm một năm bảo hiểm chi phí y tế hoặc bảo hiểm bệnh hiểm nghèo ngoài bảo hiểm y tế cơ bản, xem kỹ bốn chữ「保证续保」 + +- Chi phí: Người trẻ mua bảo hiểm chi phí y tế bồi hoàn theo chi phí, hiệu lực một năm, mỗi năm vài trăm tệ. Bồi hoàn theo chi phí nghĩa là chi trả theo đúng số tiền bạn đã bỏ ra. Bảo hiểm bệnh hiểm nghèo mỗi năm từ vài nghìn tệ trở lên (tác giả ước lượng, không có số chính thức). Trước khi mua bạn phải khai trung thực tình trạng sức khỏe, mất chừng nửa tiếng một lần. +- Nói thẳng: Bảo hiểm này bù cho phần bạn phải tự trả sau khi bảo hiểm y tế đã thanh toán, và cả khoản thu nhập mất khi bạn nghỉ làm. Trước khi mua, tìm cho ra bốn chữ「保证续保」. Không có bốn chữ đó thì sang năm sản phẩm có thể ngừng bán hoặc tăng giá. Tình trạng sức khỏe phải khai trung thực, giấu thì khi xảy ra sự cố sẽ bị từ chối bồi hoàn. Trước tiên đóng bảo hiểm y tế cơ bản đã, rồi hãy cân nhắc mục này. +- Lợi ích: Cơ quan quản lý chia sản phẩm này thành hai loại. Bảo hiểm chi phí y tế chi trả theo đúng số tiền y tế bạn đã bỏ ra. Bảo hiểm bệnh hiểm nghèo, khi được chẩn đoán đúng một trong những bệnh ghi trong hợp đồng, sẽ trả thẳng một khoản theo mức đã thỏa thuận. Cả hai loại đều bù cho phần bạn phải tự trả sau khi bảo hiểm y tế đã thanh toán, và cho khoản thu nhập mất khi nghỉ làm. Thời hạn bảo hiểm từ một năm trở xuống mà điều khoản không ghi「保证续保」thì thuộc bảo hiểm sức khỏe ngắn hạn. Quá một năm, hoặc có ghi「保证续保」, mới là bảo hiểm sức khỏe dài hạn. Nghĩa của「保证续保」là: đến kỳ gia hạn, công ty phải tiếp tục bảo hiểm theo đúng điều khoản cũ và mức phí đã thỏa thuận. Công ty bảo hiểm không được gợi ý cùng một người mua lặp lại bảo hiểm chi phí y tế có chức năng bảo vệ giống hoặc tương tự. Mua hai hợp đồng, cùng một khoản tiền cũng chỉ được chi trả một lần. Định nghĩa bệnh của bảo hiểm bệnh hiểm nghèo có quy chuẩn thống nhất trong ngành, gồm 28 bệnh nặng và 3 bệnh nhẹ. Ung thư ác tính, nhồi máu cơ tim cấp, di chứng sau đột quỵ là những bệnh cốt lõi (toàn quốc, áp dụng từ 2019、2020) +- Mức bằng chứng: B +- Nguồn: Cục Giám soát bảo hiểm và ngân hàng (银保监会) (2019). Biện pháp quản lý kinh doanh bảo hiểm sức khỏe（银保监会令 2019 年第 3 号，第二、四、四十一条）. ；Hiệp hội Bảo hiểm Trung Quốc và Hiệp hội Y khoa Trung Quốc (中国保险行业协会、中国医师协会) (2020). Hỏi đáp về quy chuẩn sử dụng định nghĩa bệnh của bảo hiểm bệnh hiểm nghèo（2020 年修订版）. ；Cục Giám soát bảo hiểm và ngân hàng (银保监会) (2020). Thông báo về việc sử dụng《Bảng tỷ lệ mắc bệnh hiểm nghèo của ngành bảo hiểm nhân thọ Trung Quốc（2020）》（银保监发〔2020〕51 号）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2015 sửa đổi). Luật Bảo hiểm nước Cộng hòa Nhân dân Trung Hoa第十六条（如实告知与合同成立二年后不得解除）. +- Ghi chú: Chọn mức B là vì phía chính phủ chỉ đưa ra định nghĩa và quy tắc, không có con số「mua rồi tiết kiệm được bao nhiêu」.「百万医疗险」chỉ là tên bán hàng, về pháp luật nó là bảo hiểm sức khỏe ngắn hạn. Điều khoản không ghi「保证续保」thì sang năm có thể ngừng bán hoặc tăng giá, trước khi mua hãy tìm bốn chữ đó trong điều khoản. Hai hợp đồng bảo hiểm chi phí y tế chi trả theo chi phí thực tế, cùng một khoản tiền không được chi trả hai lần, đừng mua chồng lên nhau. Đã có bệnh mà vẫn mua và không khai trung thực, khi xảy ra sự cố sẽ bị từ chối bồi hoàn. Khai trung thực có thời hạn hai năm: hợp đồng tròn hai năm, công ty bảo hiểm không thể lấy lý do bạn không khai trung thực để hủy hợp đồng, xảy ra việc là phải bồi hoàn. Trong hai năm mà bị phát hiện cố ý không khai, việc xảy ra trước khi hợp đồng bị hủy sẽ không được bồi hoàn, tiền phí cũng không hoàn. Những khoản thiệt hại nào xứng đáng để bảo hiểm gánh, xem mục 40 ở phần 5 (khoản thiệt hại không gánh nổi). Trước tiên đóng bảo hiểm y tế cư dân (mục 9) đã, rồi hãy cân nhắc mục này. Quy chuẩn định nghĩa bệnh do hiệp hội ngành đưa ra dưới sự hướng dẫn của Cục Giám soát bảo hiểm và ngân hàng, không phải văn bản của chính phủ. + +### 21. Thật sự không có chỗ ngủ, nơi mở cửa 24 giờ chỉ cứu vãn được một đêm, ưu tiên đến trạm cứu hộ + +- Chi phí: Khu chờ sân bay, nhà ga thường không mất tiền. Quán ăn nhanh, quán internet thì phải tiêu ít tiền. Ngủ không ngon, đồ đạc có nguy cơ bị lấy cắp. Khó nhất là cả đêm không dám ngủ thật. +- Nói thẳng: Khu chờ sân bay, nhà ga gánh được một đêm, có nhà vệ sinh cũng có ổ cắm điện. Nhưng không có bất kỳ số liệu chính thức nào nói như vậy là an toàn. Giữ đồ sát người, không nhận đồ ăn thức uống người lạ đưa, đừng vào góc khuất. Dưới cầu, ven sông, công trường ban đêm không ngủ được. Trạm cứu hộ nuôi ăn nuôi ở miễn phí, ưu tiên đến đó trước. +- Lợi ích: Sống qua được một đêm, có nhà vệ sinh, sạc được điện. Không có số liệu chính thức +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có văn bản chính thức +- Ghi chú: Rủi ro an toàn là có thật. Giữ đồ sát người, không nhận đồ uống và đồ ăn người lạ đưa, tránh góc khuất. Ban đêm đừng ngủ ở dưới cầu, ven sông, công trường. Ngủ ngoài trời trong thời tiết cực đoan có thể mất mạng. Trạm cứu hộ ở mục 4 nuôi ăn nuôi ở miễn phí, đến đó trước. Một số nơi nửa đêm sẽ dọn người, đừng xung đột với bảo vệ. diff --git "a/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" "b/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" index 55e7dc51..6ac65fa5 100644 --- "a/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" +++ "b/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 9. 普通人容易踩的法律红线 - -这一节算两笔账：你的人身自由，和你的钱。下面这些事，普通人常以为「没什么」，实际会被拘留、罚款，重的还会判刑。每条都给出法条里的具体数字，以及官方渠道公布过的案例。本节引的治安管理处罚法是 2025 年修订版，2026 年 1 月 1 日起施行，条款号和旧法不一样。刑法引的是北京市公安交管局转载的「根据刑法修正案（十一）修正」的整合文本，正文里会注明对应的修正案。条目的顺序，按「最容易不小心踩到」加「后果最重」从高到低排。 - [← Về mục lục](../README.md) # 9. Vạch đỏ pháp lý dân dân thường dễ vấp diff --git "a/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" "b/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" new file mode 100644 index 00000000..e016ca6f --- /dev/null +++ "b/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" @@ -0,0 +1,161 @@ +[← 回总目录](../README.md) + +# 11. 程序员和技术人容易踩的红线 + +本节口径是人身自由与金钱，换回的是「别被关」和「别赔钱」。写代码、做技术的人，常把下面这些事当成「只是技术问题」，可它们对应的是刑法里的具体罪名，或者民事上要赔的具体一笔钱。每条都给出法条里的数字，以及最高人民检察院、最高人民法院官网公布的案例。刑法引用的是北京市公安交管局转载的「根据刑法修正案（十一）修正」整合文本。条目顺序按「技术人最容易以为没事 × 后果最重」从高到低排。第 1 条是通用的一条，讲动手前怎么判断，以及出事之后先做什么。 + +[← Về mục lục](../README.md) + +# 11. Vạch đỏ lập trình viên và người làm kỹ thuật dễ vấp + +Thước đo của phần này là tự do cá nhân và tiền, đổi lại bạn nhận được「đừng bị giam」và「đừng phải bồi tiền」. Người viết mã, làm kỹ thuật thường coi những việc dưới đây chỉ là「vấn đề kỹ thuật」, nhưng chúng ứng với những tội danh cụ thể trong Bộ luật Hình sự, hoặc với một khoản tiền bồi thường dân sự cụ thể. Mỗi mục đưa ra con số trong điều luật, cùng các vụ án do Viện kiểm sát nhân dân tối cao và Tòa án nhân dân tối cao công bố trên trang chính thức. Bộ luật Hình sự được dẫn là văn bản hợp nhất「được sửa đổi theo Bản sửa đổi (mười một) của Bộ luật Hình sự」do Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại. Các mục được xếp theo「người làm kỹ thuật dễ tưởng không có chuyện × hậu quả nặng nhất」, từ cao xuống thấp. Mục 1 là mục chung, nói về cách đánh giá trước khi ra tay, và làm gì ngay sau khi có chuyện. + +### 1. Trước khi ra tay hỏi ba câu: đã làm tổn hại lợi ích của ai, bên kia có khả năng truy cứu đến mức nào, bạn để lại bao nhiêu chứng cứ; bị truy cứu thì tìm luật sư hình sự ngay + +- Chi phí: Không tốn tiền. Trước khi ra tay, đổi câu hỏi「kỹ thuật có làm được không」thành câu hỏi khác: nhà sản xuất, nền tảng, công ty có đi báo việc không. Chỗ khó là nhịn được cái ý「chuyện nhỏ này không ai hỏi đâu」, vì nhật ký máy chủ, lịch sử thanh toán, lịch sử nhóm trò chuyện, khi xét xử đều mặc định được truy xuất. +- Nói thẳng: Trước khi ra tay, hỏi một câu: nhà sản xuất, nền tảng, công ty có đi báo việc không. Nhật ký máy chủ, lịch sử thanh toán, lịch sử nhóm trò chuyện đều truy xuất được khi xét xử. Thật sự bị bắt giữ hình sự thì trong tối đa 37 ngày có quyết định bắt hay không. Từ lần thẩm vấn đầu tiên bạn đã có thể mời luật sư, và 37 ngày đó là lúc tìm luật sư hình sự hữu ích nhất. +- Lợi ích: Bất kỳ đơn vị hay cá nhân nào phát hiện sự thật phạm tội đều có quyền báo việc. Cơ quan xử lý cho là có sự thật phạm tội thì phải khởi xướng. Sau khi bị bắt giữ hình sự, công an phải trong 3 ngày trình xin phê chuẩn bắt (xin viện kiểm sát bắt chính thức). Trường hợp đặc biệt có thể kéo dài thêm 1 đến 4 ngày. Lưu trú nhiều nơi (liên tục phạm tội ở các địa phương khác nhau) thì kéo dài được tới 30 ngày. Viện kiểm sát dùng thêm 7 ngày để quyết định. Cộng lại tối đa 37 ngày để quyết định có bắt hay không. Từ lần thẩm vấn đầu tiên, hoặc từ lúc bị áp dụng biện pháp cưỡng chế (bị giữ, bị bắt, các loại tương tự), bạn đã có thể ủy thác luật sư. Khi vụ án còn đang trong giai đoạn công an điều tra thì chỉ có thể ủy thác luật sư. Luật sư yêu cầu gặp thì trại giam sắp xếp chậm nhất trong 48 giờ. Nếu có thể bị phạt kỉ luật, giam giữ, những hình phạt nhẹ hơn ngồi tù, hoặc chỉ bị phạt kèm mà không phạt chính, thì có thể được bảo lãnh tại nhà: nộp một khoản tiền bảo lãnh hoặc tìm một người bảo lãnh, về nhà chờ xét xử. Bảo lãnh tại nhà dài nhất 12 tháng (toàn quốc) +- Mức bằng chứng: C +- Nguồn: Quốc hội Nhân dân (全国人大) (2018). 刑事诉讼法（2018 年修正，第三十四、三十九、六十七、七十九、九十一、一百一十、一百一十二条）. +- Ghi chú: Con số 37 ngày đó tính từ lúc bị giữ đến lúc viện kiểm sát quyết định có phê chuẩn bắt hay không, không phải trần thời gian giam. Sau khi phê chuẩn bắt còn có thời hạn giam trong giai đoạn điều tra, phải tính tiếp. Luật sư rành nghề cũng hữu ích nhất đúng trong 37 ngày này: gặp thân nhân, nộp một bản ý kiến「không nên phê chuẩn bắt」cho viện kiểm sát, xin bảo lãnh tại nhà. Đợi tới lúc bị khởi tố rồi mới tìm luật sư là muộn. Các mục còn lại của phần này nói cùng một việc: bên kia có lý do để báo việc hay không. + +### 2. Không làm, không bán phần mềm gian lập trong game và công cụ hỗ trợ, kể cả công cụ AI「chỉ chụp màn hình, không đụng bộ nhớ」 + +- Chi phí: Không tốn tiền. Cái giá là từ bỏ loại việc nhận đơn đó và thu nhập từ「tiện ích nhỏ」. Chỗ khó là khoản tiền đó đến nhanh, lại không thấy ai sẽ truy cứu. +- Nói thẳng: Bán phần mềm gian lập cho 20 người dùng, hoặc thu được 5000 tệ, là đủ phạt tù dưới 3 năm, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Có người làm phần mềm xuyên tường, ngắm tự động, 3 tháng bán được hơn 8 vạn, bị phạt tù 1 năm hoãn thi hành 1 năm, phạt tiền 5000 tệ. Có người lấy chương trình phía máy chủ để mở máy chủ riêng, thu hơn 2000 vạn tệ tiền nạp, nặng nhất bị phạt tù 4 năm, tiền phạt tối đa 1000 vạn tệ. +- Lợi ích: Cung cấp chương trình, công cụ chuyên dùng để xâm nhập, kiểm soát trái phép hệ thống thông tin máy tính, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền. Phạt kèm là vừa kết án vừa phạt tiền, chỉ phạt tiền là chỉ phạt tiền thôi. Đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm kèm phạt tiền. Ngưỡng của「nghiêm trọng」là cung cấp công cụ cho trên 20 người dùng, hoặc thu nhập bất hợp pháp trên 5000 tệ. Thu nhập bất hợp pháp là số tiền kiếm được nhờ việc đó. Cố ý vòng qua hoặc phá hỏng biện pháp bảo vệ kỹ thuật mà chủ thể quyền thiết lập còn có thể cấu thành tội xâm phạm quyền tác giả, phạt tù dưới 3 năm, nghiêm trọng thì phạt tù 3 đến 10 năm. Vụ án chính thức: Vương Nhậm Hào cùng những người khác sản xuất, bán phần mềm gian lập, thực hiện xuyên tường và ngắm tự động bằng cách đọc và sửa bộ nhớ của trò chơi. Trong 3 tháng họ bán được hơn 8 vạn tệ, lợi nhuận hơn 3 vạn tệ. Bị phạt tù 1 năm hoãn thi hành 1 năm, phạt tiền 5000 tệ. Hoãn thi hành 1 năm nghĩa là đã kết án 1 năm nhưng trước mắt không vào tù; trong năm đó không tái phạm thì không phải ngồi tù. Hạng Nhậm Hào cùng những người khác bán loạt phần mềm gian lập「TZ」, doanh thu hơn 20 vạn tệ, Hạng Nhậm Hào bị phạt tù 3 năm hoãn thi hành 3 năm, phạt tiền 2 vạn tệ. Lưu Nhậm Hào cùng những người khác có được chương trình phía máy chủ của trò chơi, sửa tham số để mở máy chủ riêng, thu hơn 2000 vạn tệ tiền nạp. Bị phạt tù từ 4 năm xuống 2 năm 6 tháng, phạt tiền từ 1000 vạn tệ xuống 130 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第三款、第二百一十七条第六项）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第三条）. （Cục Công an thành phố Thâm Quyến đăng lại）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). 上海静安：严惩制售游戏外挂黑灰产共护清朗网络空间. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 2020 年度检察机关保护知识产权典型案例（案例五）. +- Ghi chú: Các vụ án gian lập đã công bố đều là loại đọc ghi bộ nhớ trò chơi, vòng qua chống gian lận. Chưa thấy vụ án chính thức nào cho công cụ hỗ trợ「chỉ chụp màn hình, chỉ mô phỏng phím bấm」. Nhưng vụ án đặt vé ở mục 3 của phần này cho thấy không đụng bộ nhớ, chỉ gửi yêu cầu, vẫn có thể bị quy kết là tội thu thập dữ liệu hệ thống thông tin máy tính trái phép. Cuối cùng có phải tội phạm hay không, tính theo tội nào, do cơ quan xử lý quyết định trên cơ sở hồ sơ nhà sản xuất nộp lúc báo việc, không phải do tác giả quyết. Bán cho 20 người dùng là đã đủ mức「nghiêm trọng」, mà mức đó rất nhỏ. + +### 3. Không viết, không bán script đặt vé hộ, mua nhanh, đặt hàng ảo, quét ưu đãi, kể cả loại「tự động bấm nút」 + +- Chi phí: Không tốn tiền, cũng không tốn thời gian. +- Nói thẳng: Script không sửa bộ nhớ, không bẻ khóa, chỉ gửi yêu cầu nhanh hơn người, vẫn bị kết án. Kiếm được 5000 tệ hoặc làm bên kia thiệt hại 1 vạn tệ là đủ. Có người dùng tài khoản 12306 mua được để đặt hộ hơn 200 vé, thu hơn 2 vạn tệ, bị phạt tù 1 năm 6 tháng hoãn thi hành 1 năm 6 tháng, phạt tiền 5000 tệ. Bán phần mềm đặt vé trên cửa hàng trực tuyến thì bị buộc bồi 2 vạn tệ về cạnh tranh không lành mạnh. +- Lợi ích: Xâm nhập hệ thống thông tin máy tính, hoặc dùng kỹ thuật khác để lấy dữ liệu trong đó, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Ngưỡng của「nghiêm trọng」là thu nhập bất hợp pháp trên 5000 tệ, hoặc gây thiệt hại trên 1 vạn tệ. Vụ án chính thức: có người dùng phần mềm gian lập nhập hàng loạt tài khoản 12306 mua được, gửi yêu cầu mua vé dày đặc bằng nhiều luồng. Người đó đặt hộ hơn 200 vé tàu, lợi nhuận hơn 2 vạn tệ. Tòa án quy tội thu thập dữ liệu hệ thống thông tin máy tính trái phép, phạt tù 1 năm 6 tháng hoãn thi hành 1 năm 6 tháng, phạt tiền 5000 tệ. Trương Nhậm Hào bán trên cửa hàng trực tuyến phần mềm đặt vé nhắm vào một ứng dụng bán vé, bị xác định là cạnh tranh không lành mạnh, buộc bồi 2 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). 上海静安：查办利用外挂软件批量代抢火车票牟利案. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 2024 年人民法院知识产权典型案例（案例七，抢票软件不正当竞争案）. +- Ghi chú: Script trong vụ án đặt vé không sửa bộ nhớ, không bẻ khóa, chỉ「nhanh hơn người」, vẫn bị kết án theo điều 285 của Bộ luật Hình sự. Nhận tiền đặt hộ và bán script đều là hành vi hưởng lợi. Trường hợp tự dùng một lần thì không tìm được vụ án chính thức nào. + +### 4. Trình thu thập dữ liệu thì chỉ lấy trang công khai không cần đăng nhập, không vòng biện pháp chống thu thập, không đụng thông tin cá nhân, dữ liệu đã lấy thì không bán + +- Chi phí: Không tốn tiền. Trình thu thập dữ liệu chỉ là chương trình tự động mở trang, lấy dữ liệu trên đó xuống. Trước khi lấy, xem qua tệp robots và hướng dẫn cấp quyền của giao diện; robots là tệp trang web đặt ở thư mục gốc, ghi rõ những trang nào không cho phép lấy. Dữ liệu phải đăng nhập, phải có token (một chuỗi chứng thư quyền tạm), phải bẻ chữ ký mới có được thì không lấy. Chỗ khó là loại dữ liệu này về mặt kỹ thuật lấy là có, nhịn lại hoàn toàn dựa vào ý thức. +- Nói thẳng: Dữ liệu phải đăng nhập, phải có token, phải bẻ chữ ký mới có được, thuộc về vòng qua lớp bảo vệ, phạt tù dưới 3 năm, nghiêm trọng thì phạt tù 3 đến 7 năm. Lấy phải thông tin cá nhân còn phiền phức hơn. Hành trạng, thông tin liên lạc, thông tin tín dụng, thông tin tài sản 50 mục là đã tính là nghiêm trọng. Lưu trú, sức khoẻ, giao dịch 500 mục, loại khác 5000 mục, hoặc thu được 5000 tệ, cũng vậy. +- Lợi ích: Vòng qua lớp bảo vệ để lấy dữ liệu trong hệ thống thì mức tù giống mục 3 của phần này (script đặt vé, mua nhanh, đặt hàng ảo), phạt tù dưới 3 năm, hoặc 3 đến 7 năm. Thu thập, bán hoặc cung cấp trái phép thông tin cá nhân của công dân thì đạt bất kỳ mốc nào dưới đây là「nghiêm trọng」: hành trạng, nội dung liên lạc, tín dụng, thông tin tài sản trên 50 mục; lưu trú, lịch sử liên lạc, sức khoẻ, thông tin giao dịch trên 500 mục; thông tin cá nhân khác trên 5000 mục; thu nhập bất hợp pháp trên 5000 tệ. Nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Vụ án chính thức: nhân viên phát triển phần mềm Lý Nhậm Hào giấu chức năng đánh cắp thư viện ảnh trong điện thoại vào dưới lớp vỏ phần mềm「kiểm tra nhan sắc đẹp」, lấy được 1751 ảnh khuôn mặt, còn bán lại dữ liệu từ kho dữ liệu thu thập thông tin, bị phạt tù 3 năm hoãn thi hành 3 năm. Kho dữ liệu thu thập thông tin là cơ sở dữ liệu gom thông tin cá nhân bị rò rỉ khắp nơi. Có công ty mạng dùng trình thu thập dữ liệu để lấy dữ liệu cửa hàng và đơn hàng của một nền tảng giao đồ ăn, gây thiệt hại hơn 4 vạn tệ, bị truy cứu theo tội thu thập dữ liệu hệ thống thông tin máy tính trái phép (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百五十三条之一）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2017). 关于办理侵犯公民个人信息刑事案件适用法律若干问题的解释（第三、四、五条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 检察机关依法惩治侵犯公民个人信息犯罪典型案例（案例二）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例一）. +- Ghi chú: Giải thích của tòa án năm 2017 quy định rằng「không có sự đồng ý của chủ thể, đưa thông tin đã thu thập hợp pháp cho người khác」cũng là「cung cấp」. Trong công ty, xuất bảng người dùng cho đối tác bên ngoài cũng rơi vào câu đó. 5000 mục chỉ là lượng của một bảng tính nhỏ. + +### 5. Không nhận đơn phát triển ứng dụng cờ bạc, lừa đảo, khiêu dâm; không dựng kênh thanh toán, không mở khoá tài khoản cho「khách hàng」 + +- Chi phí: Không tốn tiền. Cái giá là từ bỏ loại việc thuê ngoài có giá cao mà yêu cầu lại nói không rõ. Chỗ khó là số tiền bên kia trả thường cao hơn hẳn một đơn bình thường. +- Nói thẳng: Nhận đơn có giá cao, yêu cầu mơ hồ như vậy thì coi như đồng phạm của nhóm cờ bạc. Xử lý theo tội mở sòng bạc thì phạt tù dưới 5 năm, nghiêm trọng thì 5 đến 10 năm, nặng hơn hẳn so với chỉ giúp một tay thì tối đa 3 năm. Một ứng dụng đoán cược thể thao điện tử có 28 người viết mã bị phạt tù từ 1 năm đến 3 năm 6 tháng. Người đi mở khoá tài khoản QQ cho nhóm lừa đảo bị phạt tù 1 năm 6 tháng, phạt tiền 2 vạn tệ. +- Lợi ích: Biết người khác đang dùng mạng để phạm tội mà vẫn cung cấp hỗ trợ kỹ thuật, hỗ trợ thanh toán quyết toán, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền. Đồng thời cấu thành tội phạm khác thì xử lý theo tội nặng hơn. Mở sòng bạc thì phạt tù dưới 5 năm kèm phạt tiền, nghiêm trọng thì phạt tù 5 đến 10 năm. Vụ án chính thức: tập đoàn Trương Thịnh phát triển game cờ bạc rồi「đóng gói bán ra nước ngoài」, thu phí dựng trang, phí báo cáo. Số tiền liên quan gần 70 tỷ tệ. Trương Thịnh bị phạt tù 9 năm 9 tháng, phạt tiền 5000 vạn tệ. Vương Cường cùng 79 người phụ trách thuê ngoài phát triển bị phạt tù từ 6 tháng đến 9 năm 9 tháng. Trần Đông cùng những người khác phát triển, vận hành ứng dụng đoán cược thể thao điện tử, nhóm kỹ thuật viết mã ở Hải Nam, 28 người bị xử theo tội mở sòng bạc, phạt tù từ 3 năm 6 tháng đến 1 năm không đều, kèm phạt tiền. Trương Nhậm Hào mở xưởng đi mở khoá các tài khoản QQ bị khoá cho nhóm lừa đảo, bị phạt tù 1 năm 6 tháng, phạt tiền 2 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十七条之二、第三百零三条第二款）. ；Viện kiện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 建赌博网站并出售，涉案近 70 亿元. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 上海闵行区：办理一起新型开设网络赌场案. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 依法惩治帮助信息网络犯罪活动及相关犯罪典型案例（案例一）. +- Ghi chú:「Biết» không cần bên kia nói ra mục đích sử dụng. Giá đưa ra cao hơn hẳn giá thị trường, chức năng là nạp, rút, cộng tỷ lệ cược, khách không chịu nói dùng để làm gì, khi xét xử đều dùng để suy ra bạn biết rõ trong lòng. Nhân sự kỹ thuật phát triển ứng dụng cờ bạc bị xử lý như đồng phạm của tội mở sòng bạc, không thuộc mức trần 3 năm của tội giúp đỡ. Tội giúp đỡ là tên rút gọn của「tội giúp đỡ hoạt động tội phạm trên thông tin mạng」. +### 6. Khi nghỉ việc, bàn giao lại quyền truy cập tài khoản; không xoá cơ sở dữ liệu, không để lại cửa hậu, không đổi mật khẩu để khoá hệ thống, kể cả khi công ty nợ bạn tiền + +- Chi phí: Không tốn tiền. Công ty nợ lương bạn thì đi hoà giải lao động để đòi lại, đừng tự ra tay (hoà giải lao động xem ở phần 7). Chỗ khó là lúc nóng giận mà nhịn không đụng vào hệ thống. +- Nói thẳng: Tài khoản, mật khẩu do công ty cấp không có nghĩa là nghỉ việc rồi vẫn dùng được; đăng nhập vào là kiểm soát trái phép, phạt tù dưới 3 năm. Xoá dữ liệu làm hệ thống không chạy được thì phạt tù dưới 5 năm, hậu quả đặc biệt nghiêm trọng thì phạt tù trên 5 năm. Có một kỹ sư cao cấp bị sa thải, dùng tài khoản quản trị cũ đăng nhập vào máy chủ công ty để xoá dữ liệu và nhật ký, thiệt hại hơn 12 vạn tệ, bị phạt tù 3 năm hoãn thi hành 5 năm, phạt tiền 3 vạn tệ. +- Lợi ích: Kiểm soát trái phép hệ thống thông tin máy tính, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Xoá, sửa chức năng hoặc dữ liệu của hệ thống làm hệ thống không chạy bình thường, hậu quả nghiêm trọng thì phạt tù dưới 5 năm, đặc biệt nghiêm trọng thì trên 5 năm. Vụ án chính thức: kỹ sư cao cấp công nghệ thông tin Lữ Nhậm Hào sau khi bị sa thải đã dùng tài khoản quản trị cũ đăng nhập vào máy chủ dùng chung của công ty. Anh ta đổi mật khẩu, xoá dữ liệu và nhật ký, thiệt hại hơn 12 vạn tệ. Bị phạt tù 3 năm hoãn thi hành 5 năm, phạt tiền 3 vạn tệ. Nhân viên công ty phần mềm Bạch Nhậm Hào khi viết chương trình nâng cấp cố ý chèn mã sai, khiến hệ thống của khách không đăng nhập được, thiệt hại hơn 4 vạn tệ, bị truy cứu theo tội phá hoại hệ thống thông tin máy tính (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百八十六条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2024). 检察机关「依法惩治网络犯罪 助力网络空间综合治理」典型案例（案例六）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例七）. +- Ghi chú: Tài khoản, mật khẩu do công ty cấp không có nghĩa là nghỉ việc rồi vẫn dùng được. Dùng tài khoản cũ đăng nhập, coi như「kiểm soát trái phép」. Thiệt hại tính theo chi phí khôi phục dữ liệu, vài vạn tệ là đủ để khởi xướng. + +### 7. Nghỉ việc không mang mã nguồn, danh sách khách hàng, tài liệu kỹ thuật đi; không tải lên ổ đĩa cá nhân, không dùng lại ở công ty sau + +- Chi phí: Không tốn tiền. Trước khi nghỉ việc, xoá sạch tài liệu công ty trên máy tính cá nhân và trên ổ đĩa mạng. Chỗ khó là có tệp do mình tự viết từng dòng, không nỡ. +- Nói thẳng: Mã do bạn tự viết thì quyền tác giả vẫn thuộc công ty; danh sách khách hàng cũng là bí mật kinh doanh; mang đi là lấy đồ của công ty. Có nhân viên nghỉ việc mang mã nguồn hệ thống kiểm tra an ninh sang công ty mới, bản thân bị phạt tù 5 năm, phạt tiền 50 vạn tệ, công ty mới bị phạt tiền 400 vạn tệ. Có người khai thác lỗ hổng hệ thống để tải hơn 3.3 vạn tệp thiết kế, bị phạt tù 1 năm 10 tháng. Công ty còn có thể kiện bạn riêng. +- Lợi ích: Vi phạm nghĩa vụ bảo mật, tiết lộ bí mật kinh doanh hoặc tự dùng lấy, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 10 năm kèm phạt tiền. Luật Chống cạnh tranh không lành mạnh còn liệt kê hai hành vi là xâm phạm: lấy bí mật kinh doanh bằng「xâm nhập điện tử」(dùng biện pháp kỹ thuật để xâm nhập hệ thống); xúi giục, giúp người khác vi phạm nghĩa vụ bảo mật. Vụ án chính thức: Tôn Nhậm Minh cùng những nhân viên nghỉ việc khác mang mã nguồn hệ thống kiểm tra an ninh và thuật toán lõi đi, dùng vào sản phẩm của công ty mới; Tôn Nhậm Minh bị phạt tù 5 năm, phạt tiền 50 vạn tệ, công ty mới bị phạt tiền 400 vạn tệ. Thiên Nhậm Hào khai thác lỗ hổng hệ thống để tải hơn 3.3 vạn tệp thiết kế, bị phạt tù 1 năm 10 tháng, phạt tiền 10 vạn tệ. Kỹ sư Dương Nhậm Hào lưu mã nguồn các mô-đun của ứng dụng vào máy tính cá nhân và lên đám mây, sau khi nhảy sang công ty khác thì làm ứng dụng cùng loại, bị truy cứu theo tội xâm phạm bí mật kinh doanh (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百一十九条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 反不正当竞争法（2025 年修订，第十条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关知识产权保护典型案例（案例七）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2020). 2019 年度检察机关保护知识产权典型案例（案例十六）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例五）. +- Ghi chú: Mã「do mình tự viết」thì quyền tác giả vẫn thuộc công ty, điểm này xem mục 13 của phần này (mã viết bằng nguồn lực của công ty thuộc về công ty). Mang đi là lấy đồ của công ty. Danh sách khách hàng là thông tin kinh doanh, cũng là bí mật kinh doanh. Ngoài án phạt tù, công ty còn có thể kiện riêng đòi bồi thường dân sự. + +### 8. Không chạy chương trình của mình trên máy tính, máy chủ, camera của người khác; không dùng máy của công ty để đào mỏ + +- Chi phí: Không tốn tiền, cũng không tốn thời gian. +- Nói thẳng: Kiểm soát máy tính, máy chủ, camera của người khác mà không được phép, tới 20 máy là phạt tù dưới 3 năm, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Không đủ cấu thành tội phạm thì vẫn bị giữ dưới 5 ngày, còn có thể phạt tiền 5 vạn đến 50 vạn tệ. Người đã từng chịu hình phạt hình sự thì không được làm các chức vụ then chốt về quản trị an ninh mạng và vận hành mạng cả đời. +- Lợi ích: Kiểm soát trái phép hệ thống thông tin máy tính, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Kiểm soát trái phép trên 20 máy là「nghiêm trọng」. Chưa đủ cấu thành tội phạm thì công an tịch thu thu nhập bất hợp pháp, bắt giữ dưới 5 ngày, và có thể phạt tiền 5 vạn đến 50 vạn tệ. Mức nặng hơn thì bị giữ 5 đến 15 ngày, và có thể phạt tiền 10 vạn đến 100 vạn tệ. Người đã chịu xử phạt hành chính thì trong 5 năm, người đã chịu hình phạt hình sự thì cả đời, không được làm các chức vụ then chốt về quản trị an ninh mạng và vận hành mạng. Vụ án chính thức: Lý Nhậm Hào mua địa chỉ IP của camera mạng, dùng phần mềm kiểm soát 105 camera để theo dõi đời tư, bị phạt tù 3 năm hoãn thi hành 5 năm, phạt tiền 2 vạn tệ. Ngô Nhậm Hào cùng những người khác cài mã độc vào bo mạch chủ điện thoại, kiểm soát hơn 330 vạn máy điện thoại để gửi lại mã xác minh, bị phạt tù từ 2 năm đến 4 năm 6 tháng (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十九、六十六条；2016 年文本为第二十七、六十三条，终身禁业为修正后的第六十六条第三款）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 山东滨州开发区：办理一起非法控制计算机信息系统案. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 充分发挥检察职能 推进网络空间治理典型案例（案例六）. +- Ghi chú: Dùng máy chủ công ty để đào mỏ thì không tìm được vụ án nào do Viện kiểm sát nhân dân tối cao, Tòa án nhân dân tối cao công bố; mục này viết theo điều luật về「kiểm soát hệ thống của người khác khi không được phép」.「Cả đời không được làm chức vụ then chốt」là hậu quả kèm theo trong Luật An ninh mạng, ngoài mức phạt tiền và án tù, với người làm kỹ thuật thì nặng hơn mức phạt tiền. + +### 9. Không thử hệ thống của người khác khi không có văn bản ủy quyền;「vì thiện chí」và「báo cáo sau」đều không phải lý do thoát tội + +- Chi phí: Không tốn tiền. Trước khi tham gia SRC của hãng hoặc chương trình thử nghiệm cộng đồng, phải có văn bản ghi rõ phạm vi ủy quyền, danh sách mục tiêu và khoảng thời gian được thử. SRC là trung tâm ứng cứu sự cố an ninh mạng do chính hãng lập ra, chuyên nhận lỗ hổng từ bên ngoài báo lên. Xác minh xong lỗ hổng có thật thì dừng. Chỗ khó là bỏ được cái thói「thử luôn một cái cho tiện». +- Nói thẳng: Có kết án hay không phụ thuộc vào việc có ủy quyền không và lấy được bao nhiêu, chứ không phụ thuộc vào ý bạn. Có 500 bộ tài khoản, mật khẩu thường, hoặc thu được 5000 tệ, hay làm bên kia thiệt hại 1 vạn tệ, là đủ phạt tù dưới 3 năm. Để chứng minh lỗ hổng mà tiện tay tải một bảng người dùng thì thường một lần đã vượt 500 bộ. Đã có tài khoản, mật khẩu trong tay mà dùng ra ngoài phạm vi ủy quyền thì cũng là xâm nhập. +- Lợi ích: Hệ thống của công ty và cá nhân thông thường thuộc các hệ thống thông tin máy tính ngoài lĩnh vực công việc nhà nước, xây dựng quốc phòng, khoa học kỹ thuật tiên tiến. Xâm nhập loại hệ thống này, hoặc dùng biện pháp kỹ thuật khác để lấy dữ liệu được lưu, xử lý, truyền trong đó, khi nghiêm trọng thì phạt tù dưới 3 năm hoặc giam giữ, phạt kèm hoặc chỉ phạt tiền. Đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm kèm phạt tiền.「Nghiêm trọng」có vài tiêu chuẩn: lấy được thông tin xác thực danh tính của dịch vụ tài chính trên mạng như thanh toán, giao dịch chứng khoán, giao dịch hợp đồng tương lai (tài khoản, mật khẩu, các loại tương tự) từ 10 bộ trở lên; lấy được thông tin xác thực danh tính khác từ 500 bộ trở lên; kiểm soát trái phép hệ thống thông tin máy tính trên 20 máy; thu nhập bất hợp pháp trên 5000 tệ, hoặc gây thiệt hại kinh tế trên 1 vạn tệ. Gấp từ 5 lần trở lên so với các con số trên là「đặc biệt nghiêm trọng」. Vụ án chính thức: vụ án hướng dẫn số 36 của viện kiểm sát xác lập một câu:「超出授权范围使用账号、密码登录计算机信息系统，属于侵入计算机信息系统的行为」, nghĩa là vốn đã có tài khoản, mật khẩu nhưng dùng ra ngoài phạm vi ủy quyền thì bản thân hành vi đó đã là xâm nhập. Ba người trong vụ án dùng tài khoản, mật khẩu và mã thông báo nắm được trong công việc, đăng nhập từ địa điểm khác vào hệ thống nội bộ của công ty, tải dữ liệu nằm ngoài phạm vi công việc, rồi bán ra trên mạng, thu nhập bất hợp pháp 37000 tệ. Ba người lần lượt bị phạt tù 4 năm, 3 năm 9 tháng, 4 năm, mỗi người phạt tiền 4 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第一款、第二款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. （Cục Công an thành phố Thâm Quyến đăng lại）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2017). 第九批指导性案例（检例第 36 号，卫梦龙、龚旭、薛东东非法获取计算机信息系统数据案）. +- Ghi chú: Vụ án hướng dẫn số 36 vốn là vụ bán dữ liệu để hưởng lợi. Dẫn nó chỉ để lấy câu「đăng nhập ra ngoài phạm vi ủy quyền cũng là xâm nhập」, không phải nói thử nghiệm thiện chí bị phạt nặng như vậy. Bạn định làm gì, sau đó có báo cáo hay không, đều không xoá được tội danh, chỉ có thể ảnh hưởng tới mức tù và có bị khởi tố hay không. Hãng cảm ơn trước rồi quay lại báo việc, hai việc đó không mâu thuẫn, cảm ơn không phải là ủy quyền. Lúc viết phần này, không tìm được trên trang Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao vụ án thử nghiệm thiện chí nào đối chiếu từng chữ, nên chỉ viết theo điều luật và tiêu chuẩn xử phạt. Muốn làm thì đi theo SRC công khai của hãng, hoặc có văn bản ủy thác đo đạc, ghi rõ phạm vi, mục tiêu và khoảng thời gian vào hồ sơ. Chứng minh lỗ hổng tồn tại bằng ít bằng chứng nhất, đừng lấy dữ liệu hàng loạt. Mức chưa đủ cấu thành tội phạm vẫn phải chịu xử phạt hành chính, kèm cấm hành nghề, mức phạt tiền và thời hạn giống mục 8 của phần này (không chạy chương trình của mình trên máy của người khác). Sau khi phát hiện lỗ hổng thì xử lý thế nào, xem mục 10 của phần này (báo cáo lỗ hổng theo quy định). +### 10. Báo cáo lỗ hổng theo quy định; trước khi vá xong không công khai chi tiết, không phát hành công cụ khai thác, không đưa ra nước ngoài + +- Chi phí: Không tốn tiền. Báo lỗ hổng cho nhà cung cấp sản phẩm, hoặc cho nền tảng chính thức. Trước khi hãng vá xong, không đăng chi tiết, cũng không đăng mã kiểm chứng. +- Nói thẳng: Cá nhân cũng nằm trong phạm vi điều chỉnh của Quy định về quản lý lỗ hổng an ninh sản phẩm mạng. Trước khi nhà sản xuất đưa ra cách vá, bạn không được công khai lỗ hổng, cũng không được công khai chi tiết lỗ hổng của hệ thống đang dùng. Không phát hành chương trình, công cụ chuyên khai thác lỗ hổng; khi công khai phải nói kèm cách vá. Lỗ hổng chưa công khai thì không được đưa cho tổ chức, cá nhân nước ngoài ngoài nhà sản xuất. +- Lợi ích: Quy định này quản lý ba nhóm đối tượng: nhà cung cấp sản phẩm mạng trong nước, đơn vị vận hành mạng, và các tổ chức hoặc cá nhân tham gia hoạt động phát hiện, thu thập, công khai lỗ hổng. Không tổ chức, cá nhân nào được dùng lỗ hổng để làm hoạt động gây nguy hại an ninh mạng, không được thu thập, bán, công khai thông tin lỗ hổng trái phép. Công khai thông tin lỗ hổng ra xã hội thì phải tuân thủ năm điều. Một, không được công khai trước khi nhà cung cấp sản phẩm có biện pháp vá. Hai, không được công khai tình hình chi tiết rằng hệ thống mạng, hệ thống thông tin và thiết bị mà đơn vị vận hành mạng đang dùng có lỗ hổng. Ba, không được cố tình phóng đại mức nguy hại và rủi ro, không được dùng thông tin lỗ hổng để làm nội dung độc hại hoặc lừa đảo. Bốn, không được công khai hoặc cung cấp chương trình, công cụ chuyên dùng để khai thác lỗ hổng gây nguy hại an ninh mạng. Năm, khi công khai phải công khai kèm biện pháp vá hoặc phòng ngừa. Ngoài ra, thông tin lỗ hổng chưa công khai thì không được cung cấp cho tổ chức, cá nhân nước ngoài ngoài nhà cung cấp sản phẩm. Quy định khuyến khích báo lỗ hổng cho bốn nền tảng chính thức. Một là nền tảng chia sẻ thông tin mối đe doạ và lỗ hổng an ninh mạng của Bộ Công nghiệp và Thông tin. Hai là nền tảng lỗ hổng của Trung tâm Thông báo Quốc gia về an ninh mạng và thông tin. Ba là nền tảng lỗ hổng của Trung tâm Xử lý Khẩn cấp Kỹ thuật Mạng máy tính quốc gia. Bốn là kho lỗ hổng của Trung tâm Đánh giá An ninh thông tin Trung Quốc. Vi phạm quy định mà thu thập, công khai thông tin lỗ hổng thì Bộ Công nghiệp và Thông tin, Bộ Công an xử lý theo trách nhiệm của mình. Trường hợp cấu thành tình huống theo quy định của Luật An ninh mạng thì xử phạt theo luật đó. Công khai thông tin an ninh mạng như lỗ hổng hệ thống ra xã hội thì phải tuân thủ quy định của nhà nước. Vi phạm thì cơ quan quản lý có liên quan yêu cầu chấp hành điều chỉnh, cảnh cáo, có thể phạt tiền từ hơn 1 vạn tệ đến dưới 10 vạn tệ. Không chấp hành điều chỉnh hoặc nghiêm trọng thì phạt tiền từ 10 vạn tệ đến 100 vạn tệ. Còn có thể bị yêu cầu tạm dừng hoạt động liên quan, đình chỉ hoạt động để cải tổ, đóng website hoặc ứng dụng. Cũng có thể bị thu hồi giấy phép hoạt động liên quan hoặc thu hồi giấy đăng ký kinh doanh. Người chịu trách nhiệm trực tiếp và người liên quan trực tiếp khác bị phạt tiền từ hơn 1 vạn tệ đến dưới 10 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Công nghiệp và Thông tin (工业和信息化部), Văn phòng Quốc gia Quản lý Thông tin Internet (国家互联网信息办公室), Bộ Công an (公安部) (2021). 网络产品安全漏洞管理规定（工信部联网安〔2021〕66 号，第二、四、九、十、十四条，2021 年 9 月 1 日施行）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十八、六十五条；2016 年文本为第二十六、六十二条，规定第十四条援引的是 2016 年文本的条号）. +- Ghi chú: Quy định nói rõ cá nhân nằm trong đối tượng quản lý, câu「tôi chỉ là người nghiệp dư」không nằm trong ngoại lệ. Gửi lỗ hổng của hệ thống trong nước ra nền tảng thưởng lỗ hổng ở nước ngoài thì đúng vào câu「lỗ hổng chưa công khai không được đưa ra nước ngoài」. Có đủ tư cách để thử hay không, xem mục 9 của phần này (không thử hệ thống của người khác khi không có văn bản ủy quyền); còn sau khi có lỗ hổng thì xử lý thế nào là chuyện khác. Lỗ hổng tìm ra hợp pháp cũng có thể bị phạt vì cách công khai; lỗ hổng tìm ra khi chưa được ủy quyền thì cả khâu thử lẫn khâu công khai đều có vấn đề. Trong các kênh báo cáo, nền tảng của Bộ Công nghiệp và Thông tin và Trung tâm ứng cứu sự cố mạng quốc gia (CNCERT) được dùng nhiều nhất. Quy định không tách riêng trường hợp「đã báo hãng mà hãng không vá」; cách làm thực tế là để lại hồ sơ quá trình thương lượng và báo cáo, đồng thời báo cho nền tảng chính thức. + +### 11. Không bán công cụ vượt tường, không bán tài khoản VPN, không dựng loại nút truy cập đó cho người khác + +- Chi phí: Không tốn tiền, cũng không tốn thời gian. +- Nói thẳng: Tự dựng đường truyền để ra nước ngoài, hoặc dùng kênh quốc tế ngoài kênh công cộng của nhà nước, công an có thể yêu cầu dừng kết nối, cảnh cáo, phạt tiền dưới 1.5 vạn tệ và tịch thu thu nhập bất hợp pháp. Kinh doanh theo cách đó thì cơ quan xử lý có thể xử lý theo tội kinh doanh phi pháp, nghiêm trọng thì phạt tù dưới 5 năm. Mục này nói về việc bán và dựng nút truy cập. Người cá nhân tự dùng thì mức phạt cao nhất là cảnh cáo cộng phạt tiền. +- Lợi ích: Kết nối trực tiếp ra quốc tế thì bắt buộc phải dùng kênh ra vào quốc tế do mạng viễn thông công cộng của nhà nước cung cấp. Không cá nhân, tổ chức nào được tự lập hoặc dùng kênh khác: không được tự dựng đường truyền ra nước ngoài, cũng không được dùng đường do người khác dựng. Vi phạm thì công an yêu cầu dừng kết nối, cảnh cáo, có thể phạt kèm tiền dưới 1.5 vạn tệ, tịch thu thu nhập bất hợp pháp. Kinh doanh theo cách đó thì cơ quan xử lý có thể xử lý theo tội kinh doanh phi pháp. Nghiêm trọng thì phạt tù dưới 5 năm, phạt kèm hoặc chỉ phạt tiền bằng 1 đến 5 lần thu nhập bất hợp pháp. Đặc biệt nghiêm trọng thì phạt tù trên 5 năm (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Hội đồng Bộ trưởng (国务院) (1997). 计算机信息网络国际联网管理暂行规定（第六、十四条）. （Ủy ban Nông nghiệp và Nông thôn thành phố Trùng Khánh đăng lại）；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百二十五条第四项）. +- Ghi chú: Lúc viết phần này, trên trang Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao không tìm được vụ án vượt tường nào được công bố, cách xử lý ở các nơi cũng rất khác nhau, nên để mức B. Người cá nhân tự dùng thì mức phạt nặng nhất là cảnh cáo cộng phạt tiền như điều mười bốn của quy định tạm thời viết. + +### 12. Đã ký thoả thuận không cạnh tranh, sau khi nghỉ việc công ty không trả bồi thường hằng tháng thì gửi văn bản đòi; quá 3 tháng không trả thì có thể chấm dứt; với chức vụ không tiếp cận bí mật kinh doanh thì có thể yêu cầu xác nhận điều khoản không có hiệu lực + +- Chi phí: Không tốn tiền. Lúc nghỉ việc giữ lại bản gốc thoả thuận không cạnh tranh, cùng chứng cứ công ty không trả tiền hằng tháng. +- Nói thẳng: Không cạnh tranh nghĩa là trong một thời gian sau khi nghỉ việc không được đi làm cho công ty cùng ngành, tối đa 2 năm, công ty phải trả bồi thường hằng tháng. Không thỏa thuận con số thì trả bằng 30% lương trung bình 12 tháng trước khi nghỉ việc, không thấp hơn mức lương tối thiểu địa phương. Do lý do từ phía công ty mà 3 tháng không trả thì bạn có thể yêu cầu chấm dứt. Từ năm 2025, với chức vụ không biết, không tiếp cận bí mật kinh doanh thì có thể yêu cầu xác nhận điều khoản không có hiệu lực. +- Lợi ích: Không cạnh tranh chỉ ràng buộc lãnh đạo, nhân sự kỹ thuật cao cấp, và những người khác có nghĩa vụ bảo mật. Thời hạn tối đa 2 năm. Công ty phải trả bồi thường kinh tế hằng tháng trong thời hạn đó. Không thỏa thuận số tiền thì trả bằng 30% lương trung bình 12 tháng trước khi nghỉ việc, đồng thời không thấp hơn mức lương tối thiểu địa phương. Do lý do từ phía công ty mà 3 tháng không trả thì người lao động có thể yêu cầu chấm dứt ràng buộc này. Công ty chủ động chấm dứt thì phải trả thêm 3 tháng bồi thường. Từ năm 2025, người lao động ở vị trí không biết, không tiếp cận bí mật kinh doanh có thể yêu cầu xác nhận điều khoản không có hiệu lực. Phần thỏa thuận vượt quá phần bí mật mà người đó thực sự biết thì vô hiệu. Vụ án chính thức: một công ty dược kiện giám đốc công nghệ cao cấp Trịnh Nhậm Hào, đòi bồi thường vi phạm 7.1 triệu tệ. Tòa án xác định sản phẩm của công ty mới không thay thế được sản phẩm công ty cũ, phạm vi không cạnh tranh phải giới hạn ở phần bí mật mà ông thực sự biết, nên bác toàn bộ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). 劳动合同法（2012 年修正，第二十三、二十四条）. （Cục Quản lý Khẩn cấp thành phố Phụ Thuận đăng lại）；Tòa án nhân dân tối cao (最高人民法院) (2020). 关于审理劳动争议案件适用法律问题的解释（一）（第三十六至四十条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 关于审理劳动争议案件适用法律问题的解释（二）（法释〔2025〕12 号，第十三至十五条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 劳动争议典型案例（案例四）. +- Ghi chú: Ngược lại, nếu thoả thuận có hiệu lực, công ty cũng đã trả tiền hằng tháng, mà bạn vẫn đi làm cho công ty cạnh tranh thì phải trả phạt vi phạm theo thoả thuận. Trả xong còn có thể bị yêu cầu giữ hết phần thời gian không cạnh tranh còn lại. Thỏa thuận ký trong thời gian còn làm việc, công ty không trả thêm tiền thì vẫn có hiệu lực. + +### 13. Mã viết trong giờ làm việc, bằng nguồn lực của công ty thì thuộc công ty; dự án nguồn mở của bạn thì dùng thời gian, thiết bị của bạn, đừng trộn mã công ty + +- Chi phí: Không tốn tiền. Khi nhận việc, đọc kỹ điều khoản quyền sở hữu trí tuệ trong hợp đồng lao động. Chỗ khó là dự án cá nhân và công việc của công ty thường nằm chung một máy tính, cần tách ra. +- Nói thẳng: Phần mềm chủ yếu viết bằng thiết bị, mạng nội bộ hoặc dữ liệu kinh doanh của công ty, và do công ty chịu trách nhiệm thì quyền tác giả thuộc công ty, bạn chỉ giữ quyền ghi tên, hợp đồng có thỏa thuận khác thì cũng vậy. Viết sau giờ làm việc mà dùng máy tính công ty hoặc dữ liệu kinh doanh thì vẫn có thể là của công ty. Phát minh làm ra bằng nguồn lực công ty thì bằng sáng chế cũng thuộc công ty. +- Lợi ích: Tác phẩm tạo ra để hoàn thành nhiệm vụ công việc của đơn vị là tác phẩm công vụ (sản phẩm làm ra khi đi làm). Trường hợp thông thường quyền tác giả thuộc tác giả, đơn vị có quyền ưu tiên sử dụng trong phạm vi công việc, và tác giả không được cho phép bên thứ ba sử dụng theo cùng cách trong 2 năm. Nhưng có hai loại khác nhau. Một là các tác phẩm như thiết kế công trình, phần mềm, chủ yếu sử dụng điều kiện vật chất kỹ thuật của đơn vị và do đơn vị chịu trách nhiệm. Hai là tác phẩm mà hợp đồng thỏa thuận thuộc đơn vị. Hai loại này tác giả chỉ giữ quyền ghi tên. Điều kiện vật chất kỹ thuật của đơn vị là thiết bị, mạng nội bộ, dữ liệu kinh doanh của công ty. Sáng chế hoàn thành khi thực hiện nhiệm vụ của đơn vị, hoặc chủ yếu sử dụng điều kiện vật chất kỹ thuật của đơn vị, là phát minh công vụ, quyền xin cấp bằng sáng chế và bằng sáng chế nhận được đều thuộc đơn vị (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 著作权法（2020 年修正，第十八条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 专利法（2020 年修正，第六条）. +- Ghi chú: Phần mềm bị xếp thẳng vào loại「quyền tác giả thuộc đơn vị」, lập trình viên hầu như không rơi vào nhóm「quyền tác giả thuộc tác giả, đơn vị ưu tiên sử dụng trong 2 năm」. Việc「viết sau giờ làm việc」mà dùng máy tính, mạng nội bộ hoặc dữ liệu kinh doanh của công ty, vẫn có thể bị tính là chủ yếu làm ra bằng thiết bị và tư liệu của đơn vị. Muốn làm sạch dự án cá nhân thì dùng máy của mình, không đụng kho mã công ty, không trùng với công việc công ty. + +### 14. Làm sản phẩm bằng mã nguồn mở như GPL thì mở mã theo giấy phép yêu cầu hoặc đổi sang giấy phép khác, đừng coi là「không ai quản」 + +- Chi phí: Không tốn tiền. Trước khi dùng mã nguồn mở của người khác, xem giấy phép của nó (điều kiện sử dụng tác giả ghi rõ). Dòng GPL yêu cầu sản phẩm có dùng nó cũng phải công khai mã nguồn, xung đột với việc bán mã đóng, xung đột thì đổi sang giấy phép khác. Chỗ khó là bỏ một thư viện đã quen dùng. +- Nói thẳng: Xâm phạm quyền tác giả thì bồi theo thiệt hại của bên kia hoặc số tiền bạn thu được, cố ý và nghiêm trọng thì có thể bị phạt bồi gấp 1 đến 5 lần. Có công ty thuê nhân viên cũ của đối thủ, làm ra phần mềm giống nhau về căn bản, và biện bằng phần mềm của đối phương dùng GPLv2, lẽ ra phải mở mã. Tòa án nhân dân tối cao cho rằng đối phương có mở mã hay không không ảnh hưởng tới quyền tác giả của nó, vẫn phán phải chấm dứt vi phạm và bồi 50 vạn tệ. +- Lợi ích: Xâm phạm quyền tác giả thì bồi theo thiệt hại thực tế của chủ quyền, hoặc theo số tiền người vi phạm thu được. Nếu cả hai đều không tính được thì căn cứ vào số tiền phải trả nếu mua hợp lệ giấy phép đó. Cố ý vi phạm và nghiêm trọng thì có thể bị phạt bồi gấp 1 đến 5 lần. Gây tổn hại lợi ích công cộng thì còn có thể tịch thu thu nhập bất hợp pháp và phạt tiền. Vụ án chính thức: (2021) 最高法知民终 51 号. Bị đơn thuê nhân viên cũ của nguyên đơn, phần mềm cổng phát triển mà bên bị đơn phát triển giống bản chất phần mềm của nguyên đơn. Bên bị đơn biện bảo phần mềm của nguyên đơn dựa trên GPLv2, vốn dĩ phải mở mã. Tòa án nhân dân tối cao xác định「开发者自身是否违反 GPLv2 协议和是否享有软件著作权是两个独立问题」, phán phải chấm dứt vi phạm, bồi thường 50 vạn tệ (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 著作权法（2020 年修正，第五十三、五十四条）. ；Tòa Xử quyền sở hữu trí tuệ của Tòa án nhân dân tối cao (最高人民法院知识产权法庭) (2022). 涉「开源软件」著作权侵权案（（2021）最高法知民终 51 号）. +- Ghi chú: Vụ án này cho thấy không thể lấy「đối phương dùng mã GPL」làm lá chắn cho việc bạn sao chép. Ngược lại, tác giả GPL đã kiện bên thương mại không mở mã và trong nước đã có án đã có hiệu lực, nhưng phần này không tìm được nguyên văn trên trang Tòa án nhân dân tối cao nên không dẫn. Để ở mức B là vì việc có tuân thủ giấy phép nguồn mở hay không chỉ có thể dựa vào hợp đồng và các quy định chung trong luật quyền tác giả, không có điều luật chuyên biệt nào. + +### 15. Làm ứng dụng chỉ thu thập thông tin cá nhân thật cần cho nghiệp vụ, dưới 14 tuổi phải có đồng ý của người giám hộ, chính sách bảo mật viết rõ + +- Chi phí: Không tốn tiền. Trước khi đưa lên vận hành, đối chiếu danh sách quyền ứng dụng xin với chính sách bảo mật, quyền không cần thì đừng xin. Chỗ khó là bên sản phẩm luôn muốn giữ thêm mấy thứ「sau này có thể dùng đến」. +- Nói thẳng: Chỉ thu thập thông tin thật cần cho nghiệp vụ, người chưa đủ 14 tuổi thì phải có đồng ý của người giám hộ. Vi phạm thì trước hết bị yêu cầu chấp hành điều chỉnh, ứng dụng có thể bị ngừng. Không chấp hành điều chỉnh thì phạt tiền dưới 100 vạn tệ, người phụ trách phạt tiền 1 vạn đến 10 vạn tệ. Nghiêm trọng thì phạt tiền dưới 50 triệu tệ hoặc 5% doanh thu năm trước, người phụ trách phạt tiền 10 vạn đến 100 vạn tệ; người viết mã và làm sản phẩm cũng có thể tính là người phụ trách. +- Lợi ích: Xử lý thông tin cá nhân phải có mục đích rõ ràng, hợp lý, trực tiếp liên quan đến mục đích đó, và phải dùng cách ảnh hưởng ít nhất tới quyền lợi cá nhân. Xử lý thông tin cá nhân phải được cá nhân đồng ý, hoặc phải thuộc trường hợp theo quy định pháp luật. Người chưa đủ 14 tuần tuổi thì bắt buộc phải được cha mẹ hoặc người giám hộ đồng ý. Vi phạm thì cơ quan quản lý yêu cầu chấp hành điều chỉnh, cảnh cáo, tịch thu thu nhập bất hợp pháp, với ứng dụng vi phạm thì yêu cầu tạm dừng hoặc chấm dứt cung cấp dịch vụ. Không chấp hành điều chỉnh thì phạt tiền dưới 100 vạn tệ, người chịu trách nhiệm trực tiếp và người liên quan trực tiếp khác phạt tiền 1 vạn đến 10 vạn tệ. Nghiêm trọng thì phạt tiền dưới 50 triệu tệ, hoặc không quá 5% doanh thu năm trước. Người chịu trách nhiệm phạt tiền 10 vạn đến 100 vạn tệ. Những người này còn có thể bị cấm giữ chức vụ giám đốc, giám sát, lãnh đạo cấp cao của doanh nghiệp liên quan trong một thời hạn nhất định. Bán hoặc cung cấp thông tin cá nhân mà cấu thành tội phạm thì xem mục 4 của phần này (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). 个人信息保护法（第六、十三、三十一、六十六条）. +- Ghi chú:「Người chịu trách nhiệm trực tiếp」trong thực tế hay rơi vào người viết mã và làm sản phẩm. Bộ Công nghiệp và Thông tin định kỳ công bố danh sách gỡ ứng dụng vi phạm, phần này chưa đối chiếu được đợt công bố nào nên không dẫn. +### 16. Trước khi website, ứng dụng đưa vào vận hành phải đăng ký ICP; giữ nhật ký trên 6 tháng trở lên theo yêu cầu phân cấp bảo vệ + +- Chi phí: Đăng ký miễn phí, đi theo quy trình của nhà cung cấp điện toán đám mây thì mất mấy tuần. Đánh giá phân cấp bảo vệ thì tính phí theo cấp độ. Phân cấp bảo vệ gọi tắt là phân cấp bảo vệ, là bộ biện pháp quản lý an toàn mà nhà nước yêu cầu website và hệ thống thực hiện theo mức độ quan trọng; website nhỏ thì thường không cần đánh giá. +- Nói thẳng: Đăng ký không mất tiền, đi theo quy trình của nhà cung cấp điện toán đám mây thì mất mấy tuần. Máy chủ trong nước mà không đăng ký thì tên miền không phân giải được, không chấp hành điều chỉnh thì bị đóng website. Trang chủ không ghi số đăng ký thì phạt tiền 5000 tệ đến 5 vạn tệ. Đăng ký nói không thu phí mà thực tế có thu tiền thì phạt tiền 10 vạn đến 100 vạn tệ. Nhật ký phải giữ đủ 6 tháng, không chấp hành điều chỉnh hoặc để xảy ra hậu quả thì phạt tiền 1 vạn đến 10 vạn tệ. +- Lợi ích: Dịch vụ thông tin trên internet mang tính kinh doanh thì phải được cấp phép, không mang tính kinh doanh thì phải đăng ký. Không có giấy phép, hoặc không đăng ký thì không được làm. Không đăng ký thì cơ quan quản lý viễn thông yêu cầu chấp hành điều chỉnh, không chấp hành thì yêu cầu đóng website. Kinh doanh khi chưa được phép thì tịch thu thu nhập bất hợp pháp, kèm phạt tiền gấp 3 đến 5 lần thu nhập bất hợp pháp, hoặc phạt tiền 10 vạn đến 100 vạn tệ. Trang chủ website không ghi số đăng ký thì phạt tiền 5000 tệ đến 5 vạn tệ. Đơn vị vận hành mạng phải theo hệ thống phân cấp bảo vệ xây dựng quản lý an toàn, phòng chống tấn công, lưu nhật ký mạng không ít hơn 6 tháng, phân loại, sao lưu, mã hoá dữ liệu. Không thực hiện thì bị yêu cầu chấp hành điều chỉnh, cảnh cáo. Không chấp hành điều chỉnh hoặc để xảy ra hậu quả thì phạt tiền 1 vạn đến 10 vạn tệ, người chịu trách nhiệm trực tiếp phạt tiền 5000 tệ đến 5 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Hội đồng Bộ trưởng (国务院) (2011). 互联网信息服务管理办法（2011 年修订，第四、十九、二十二条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十三、六十一条；2016 年文本为第二十一、五十九条）. +- Ghi chú: Máy chủ trong nước mà không đăng ký thì tên miền không phân giải được. Có hai cái bẫy thường gặp: đăng ký là cá nhân không kinh doanh nhưng thực tế có thu tiền; đổi tên miền hoặc đổi chủ thể mà không đi làm thay đổi đăng ký. Giữ nhật ký 6 tháng vừa là nghĩa vụ vừa là cách tự bảo vệ, khi có chuyện thì chứng minh được không phải mình làm. + +### 17. Trước khi cung cấp dịch vụ trí tuệ nhân tạo tạo sinh ra công chúng, phải làm đánh giá an toàn và đăng ký thuật toán theo quy định; nội dung tạo sinh phải gắn nhãn + +- Chi phí: Không tốn tiền, nhưng phải đi qua quy trình của cơ quan quản lý không gian thông tin, xong cũng tốn thời gian. Công cụ nhỏ do cá nhân làm, chỉ cần không mở cho công chúng thì không kích hoạt bộ này. +- Nói thẳng: Dịch vụ trí tuệ nhân tạo tạo sinh hướng tới công chúng, có thể ảnh hưởng dư luận hoặc vận động người ta, thì trước khi đưa vào vận hành phải làm đánh giá an toàn, đăng ký thuật toán; có thay đổi hoặc dừng thì phải làm thủ tục thay đổi, huỷ đăng ký. Hình ảnh và video tạo sinh phải gắn nhãn. Công cụ nhỏ bạn tự dùng, không mở cho công chúng thì không cần lo mấy điều này. Vi phạm thì xử phạt theo Luật An ninh mạng và các luật khác. +- Lợi ích: Cung cấp dịch vụ trí tuệ nhân tạo tạo sinh có tính chất dư luận hoặc có khả năng vận động xã hội thì phải tiến hành đánh giá an toàn theo quy định của nhà nước, đăng ký thuật toán theo quy định quản lý thuật toán đề xuất, và làm các thủ tục thay đổi, huỷ đăng ký về sau. Tính chất dư luận hoặc khả năng vận động xã hội là khả năng ảnh hưởng tới nhận thức của công chúng, hoặc có thể vận động, tổ chức người ta. Bên cung cấp phải gắn nhãn cho nội dung tạo sinh như hình ảnh, video. Quy chế này áp dụng từ ngày 15 tháng 8 năm 2023 (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Văn phòng Quốc gia Quản lý Thông tin Internet (国家互联网信息办公室) và sáu bộ ngành khác (2023). 生成式人工智能服务管理暂行办法（第十二、十七条）. +- Ghi chú: Bản quy chế này không tự viết phạt bao nhiêu tiền. Vi phạm thì phải quay lại Luật An ninh mạng, Luật An toàn dữ liệu, Luật Bảo vệ thông tin cá nhân để tìm mức phạt, nên để ở mức B. Vụ án dân sự về việc nội dung tạo bằng AI xâm phạm quyền tác giả, quyền với hình ảnh của người khác, phần này chưa đối chiếu nguyên văn trên trang Tòa án nhân dân tối cao nên không dẫn. \ No newline at end of file diff --git "a/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" "b/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" index 15a04476..895abf9f 100644 --- "a/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" +++ "b/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" @@ -1,11 +1,3 @@ -[← 回总目录](../README.md) - -# 13. 紧急情况：先做什么 - -本节讲出了事先做什么、再做什么、什么不要做。急救和野外的条目算的是特定死因和活下来的比例。威胁和骗局的条目算的是钱和人身自由。两类数字不互相换算。条目的顺序按「这事多常见 × 做对做错差多少条命」从高到低排。这些条目首先是给你自己和配偶、直系亲属用的。心脏骤停有 79.2% 发生在家里，你最可能救的就是家里人。对朋友同事次之。对陌生人也不是一点好处没有，但这一档最低，还可能被讹、被卷进案子。所以这类条目把保护自己的动作和免责的条款写在前面。 - -本节只收几分钟内要做的动作。认出骗局、被人拿把柄要钱，处置要按天算，不在这一节。反诈的硬规则和七种高发骗局见第 8 节第 3 条（反诈硬规则）。被人拿隐私照片、裸聊录像要钱见第 8 节第 32 条（对方拿把柄来要钱）。钱已经转出去了，立刻打 110 或 96110 要求止付，见第 8 节第 2 条（止付）。 - [← Về mục lục](../README.md) # 13. Tình huống khẩn cấp: làm gì trước diff --git "a/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" "b/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" index 208a5dd4..8e15b8de 100644 --- "a/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" +++ "b/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 14. 账号与信息安全 - -口径：金钱与个人信息。账号被别人登进去，丢的先是钱。对方还会用你的账号去骗你通讯录里的人。你的身份也等于被人拿走了。 - [← Về mục lục](../README.md) # 14. Tài khoản và bảo mật thông tin diff --git "a/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" "b/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" index d2176d9a..62728e60 100644 --- "a/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" +++ "b/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 17. 家里有老人 - -口径：金钱与人身自由。家里有老人，多数麻烦不是钱不够，是没提前把话说清楚、把字签好。最后一条讲长期卧床的压疮，按死亡率算账，不和本节其他条目比较。 - [← Về mục lục](../README.md) # 17. Nhà có người già diff --git "a/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" index e5fea729..8a103fcc 100644 --- "a/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" +++ "b/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -1,9 +1,3 @@ -[← 回总目录](../README.md) - -# 18. 养孩子划不划算 - -这一节算的是钱和时间两笔账。和第 10 节算结婚账一样，这里只把账拆开给你看，不替你下结论。已经怀孕、要照流程一步步办事的，看第 27 节。 - [← Về mục lục](../README.md) # 18. Nuôi con có đáng không diff --git "a/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" "b/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" new file mode 100644 index 00000000..8fa6b52f --- /dev/null +++ "b/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" @@ -0,0 +1,141 @@ +[← 回总目录](../README.md) + +# 30. 上学以后的孩子 + +第 20 节讲的是孩子刚出生那一年。第 18 节算的是养孩子要花多少钱、多少时间。中间 3 到 18 岁这十几年，写在这一节。分龄管屏幕那一条从 0 岁写起，因为官方的分档就是 0～3 岁、3～6 岁、中小学三档，拆开写反而不好用。本节回答的是：「别耽误学习」和「身体」撞在一起的时候，哪些事不能等。这些事有时间窗口，过了就补不回来。学业该不该那么卷，本节不讨论，只给窗口和数字。 + +前八条和最后两条算的是死亡和身体上真落下的损害。第 9 条（不买「治愈近视」的产品）算钱。第 10 条（睡眠作业体育排名的明文规定）和第 11 条（休学保留学籍）算时间。溺水、交通、头盔、HPV 疫苗见第 1 节。婴幼儿见第 20 节。孩子被骗和游戏充值见第 5 节。未成年人自己别踩的红线见第 9 节。急症怎么认、怎么叫车见第 13 节。家长去世后的孩子见第 29 节。这些都不在本节重复。 + +### 1. Trẻ bảo đau dữ dội bất ngờ và đau hơn nữa thì đưa đi bệnh viện ngay, đừng để cháu nhịn đến tan học + +- Chi phí: Không tốn tiền. Xin nghỉ nửa ngày, đưa trẻ đăng ký khám cấp cứng. +- Nói thẳng: Có những bệnh được tính theo giờ. Bé trai đột nhiên đau dữ dội một bên bìu là xoắn tinh hoàn, tức mạch máu nối với tinh hoàn bị xoắn lại, máu không còn chảy tới. Xử lý trong vòng 6 giờ thì 97.2% tinh hoàn được giữ lại. Kéo dài đến 19 đến 24 giờ thì chỉ còn 42.5%. Cơn đau kiểu này không đợi được tan học, cũng không đợi nổi hết kỳ thi. +- Lợi ích: Theo đúng quy trình, người ta tìm đủ 30 bộ báo cáo ca bệnh, 2116 người bị xoắn tinh hoàn, rồi gộp lại xem chung. Trong đó 1283 ca được thống kê theo từng nhóm 6 giờ. Tính từ lúc bắt đầu đau, trong 0 đến 6 giờ tỷ lệ giữ được tinh hoàn là 97.2%. Từ 7 đến 12 giờ là 79.3%. Từ 13 đến 18 giờ là 61.3%. Từ 19 đến 24 giờ là 42.5%. Từ 25 đến 48 giờ là 24.4%. Quá 48 giờ là 7.4%. Gộp thành các nhóm lớn thì trong 12 giờ là 90.4%, từ 13 đến 24 giờ là 54.0%, quá 24 giờ là 18.1%. Kết luận của tác giả là thời gian còn cứu được có thể dài hơn mức 6 đến 8 giờ thường nói. Nên đã đau lâu thì vẫn phải đi xử lý, đừng vì nghĩ「chẳng qua đã muộn」mà không đi +- Mức bằng chứng: A +- Nguồn: Mellick LB, et al. (2019). A Systematic Review of Testicle Survival Time After a Torsion Event. Pediatric Emergency Care. +- Ghi chú: Nói trước với bé trai một câu: nếu bỗng đau dữ dội bên dưới thì nói ngay, đừng ngượng. Xoắn tinh hoàn hay gặp ở các bé trai tuổi thanh niên, nhiều người nhịn vì khó mở lời. Những bệnh khác cũng tính theo giờ như vậy là viêm ruột thừa cấp (kéo dài sẽ thủng), nhiễm toan ceton do đái tháo đường, viêm màng não. Phần này chưa kiểm chứng từng bệnh một xem trễ mấy giờ thì hỏng đến đâu. Dấu hiệu chung của chúng là bắt đầu đột ngột, liên tục nặng thêm, khác hẳn mọi lần trước. Nhận biết cấp cứu và gọi xe cứu thương, xem phần 13. + +### 2. Không nên vì「chờ thi xong」mà trì hoãn việc chữa trị cần làm, có những cửa sổ thời gian đi theo tuổi xương chứ không theo lịch thi + +- Chi phí: Không tốn tiền, chỉ phải quyết định làm sớm hay làm muộn. Nghỉ và tái khám tốn thời gian, tiền chữa trị tính riêng. Chỗ khó là đai nẹo bắt đeo mỗi ngày ít nhất 18 giờ, trẻ phải chịu suốt thời gian đó. +- Nói thẳng: Vẹo cột sống là cột sống cong sang một bên. Đai nẹo là một chiếc áo cứng đeo ngoài người, giữ cho cột sống thẳng. Đeo khi xương còn đang lớn, 75% trẻ không cong đến mức phải mổ; nhóm chỉ quan sát mà không đeo thì chỉ 42%. Thời gian đeo được đi theo lúc xương ngừng lớn, không đi theo kỳ thi vào trung học phổ thông hay vào đại học, nên đợi thi xong thường chỉ còn cách mổ. +- Lợi ích: Một thử nghiệm do nhiều bệnh viện cùng làm, tuyển 242 bệnh nhân vẹo cột sống tự nhiên ở tuổi thanh niên đủ điều kiện đeo đai nẹo. Trong đó 116 người được chia nhóm bằng bốc thăm, 126 người tự chọn có đeo hay không. Cách tính thành bại đã định trước: độ cong lên tới 50 độ trở lên coi là thất bại chữa trị; khi xương ngừng lớn mà chưa tới số độ đó coi là thành công chữa trị. Tính theo nhóm được phân lúc đầu, không quan tâm sau này có thực sự đeo đủ hay không. Nhóm đeo đai nẹo bốc thăm có tỷ lệ thành công 75%, nhóm chỉ quan sát bốc thăm là 42% (OR 4.11, 95% CI 1.85–9.16), cơ hội thành công cao khoảng 4 lần so với chỉ quan sát. Gộp nhóm bốc thăm với nhóm tự chọn thì 72% so với 48% (sau khi đã bù lại khác biệt vốn có giữa hai nhóm, OR 1.93, 1.08–3.46, vẫn khoảng 1.9 lần). Đeo càng nhiều giờ một ngày thì tỷ lệ thành công càng cao (P<0.001, mối quan hệ này khó có thể là ngẫu nhiên). Thử nghiệm bị dừng sớm vì đai nẹo thật sự có hiệu quả +- Mức bằng chứng: A +- Nguồn: Weinstein SL, et al. (2013). Effects of bracing in adolescents with idiopathic scoliosis. New England Journal of Medicine. +- Ghi chú: Ở đây lấy vẹo cột sống làm ví dụ vì nó có thử nghiệm chia nhóm ngẫu nhiên, có cửa sổ thời gian rõ ràng, và vừa là một trong các mục chính của khám sức khoẻ học sinh (xem mục 7 của phần này). Cùng loại còn có lách, nhược thị, sâu răng, và những ca mổ bệnh tim bẩm sinh có thể chọn thời điểm. Điểm chung của chúng là làm càng muộn thì hiệu quả càng kém, hoặc cái giá càng lớn. Cửa sổ thời gian cụ thể thì hỏi bác sĩ chuyên khoa. Sợ ảnh hưởng giờ học, trong giai đoạn giáo dục bắt buộc có thể xin nghỉ học, tối đa 1 năm, học vị được giữ, xem mục 11 của phần này. + +### 3. Trẻ bị bắt nạt thì báo ngay trong ngày cho nhà trường và yêu cầu xử lý bằng văn bản, có đánh nhau, cướp tiền, tung tin thì báo công an + +- Chi phí: Không tốn tiền. Tìm giáo viên chủ nhiệm, tìm cả tổ chuyên trách xử lý bạo lực học đường của trường. Muốn báo công an thì gọi 110. Chỗ khó là phải chịu được lời khuyên kiểu「đừng làm ầm lên」. +- Nói thẳng: Trẻ bị bạn cùng bắt nạt có xác suất nảy sinh ý nghĩ tự sát cao khoảng 2.2 lần so với trẻ không bị bắt nạt, và đã từng có hành vi tự sát khoảng 2.6 lần. Nhà trường nhận được báo cáo phải điều tra ngay, xác định là bắt nạt thì phải kỷ luật; trường hợp nặng thì không được giấu, phải báo công an. Vì vậy hãy báo trong ngày, yêu cầu đi qua thủ tục xác định, và để lại hồ sơ bằng văn bản. +- Lợi ích: Gộp 34 nghiên cứu, 284.375 trẻ em và thanh thiếu niên lại. Bị bạn cùng bắt nạt có liên quan đến ý nghĩ tự sát, OR 2.23 (95% CI 2.10–2.37), cơ hội xuất hiện ý nghĩ đó cao khoảng 2.2 lần so với trẻ không bị bắt nạt. Thêm 9 nghiên cứu khác, 70.102 người, bắt nạt có liên quan đến tự sát không thành công, OR 2.55 (1.95–3.34), khoảng 2.6 lần. Giới tính, tuổi và chất lượng của chính các nghiên cứu khác nhau, kết quả vẫn như nhau. Bắt nạt trên mạng liên quan mạnh hơn bắt nạt ngoài đời. Quy định về bảo vệ học sinh vị thành niên《未成年人学校保护规定》第二十一条 liệt kê năm nhóm hành vi: đánh, đá, tát, cào, cắn, xô đẩy, kéo giật; chửi bới, chế giễu, đặt biệt danh xúc phạm; cướp đoạt, ép buộc hoặc cố ý phá hủy tài sản; cố ý loại trừ và cô lập; bịa sự thật trên mạng để vu khống hoặc tung tin giả. Cùng điều đó quy định: bên có lợi thế về tuổi, thể chất hoặc số người cố ý hoặc ác ý làm các việc đó, gây tổn thương thân thể, mất tài sản hoặc tổn hại tinh thần thì có thể được xác định là bắt nạt. 第二十二条 yêu cầu giáo viên và nhân viên nhà trường khi thấy học sinh có biểu hiện tâm lý bất thường hoặc bị tổn thương thân thể thì phải kịp thời tìm hiểu và báo cáo. 第二十三条 yêu cầu nhà trường nhận báo cáo thì điều tra ngay, trình cơ quan xử lý bạo lực học đường để xác định và xử lý, đồng thời thông báo cha mẹ tham gia. Khi xác định cấu thành bắt nạt thì phải áp dụng hình thức giáo dục kỷ luật hoặc kỷ luật. Hành vi bắt nạt nghiêm trọng vi phạm quản lý trật tự công cộng hoặc bị nghi ngờ phạm tội thì không được giấu, phải kịp thời báo cơ quan công an và cơ quan quản lý giáo dục (toàn quốc, áp dụng từ ngày 1 tháng 9 năm 2021) +- Mức bằng chứng: A +- Nguồn: van Geel M, et al. (2014). Relationship between peer victimization, cyberbullying, and suicide in children and adolescents: a meta-analysis. JAMA Pediatrics. ；Bộ Giáo dục (教育部) (2021). Quy định về bảo vệ học sinh vị thành niên, 未成年人学校保护规定（教育部令第 50 号，第二十一、二十二、二十三条）. +- Ghi chú: Trên người trẻ có vết thương không giải thích được, bỗng nhiên không chịu đi học, đồ đạc bị mất đi đi lại, trước hết hãy hỏi theo các dấu hiệu nêu trong 第二十二条. Bản phân tích gộp nhiều nghiên cứu này chỉ ghi nhận và theo dõi, không có nhóm đối chứng, nên không chứng minh chỉ riêng bắt nạt đã gây tự sát, nhưng hướng và độ lớn của rủi ro thì rõ ràng. 《Phương án tăng cường quản lý toàn diện bạo lực học đường đối với học sinh tiểu học và trung học》《加强中小学生欺凌综合治理方案》còn yêu cầu phân biệt nghiêm ngặt bạo lực với chuyện đánh nhau vui đùa giữa học sinh. Vì vậy nhà trường có nghĩa vụ lập kết luận, không thể cho xong bằng một câu「trẻ con chơi thôi mà」. + +### 4. Cho trẻ ở ngoài trời đủ 2 giờ mỗi ngày, đây là cách phòng cận thị duy nhất hiện có thử nghiệm ngẫu nhiên đứng sau + +- Chi phí: Không tốn tiền. Nhưng mỗi ngày phải dành ra thời gian, phần lớn là phải bóc ra từ thời gian làm bài và học online. Chỗ khó là làm được mỗi ngày, không phải làm mấy lần là xong. +- Nói thẳng: Lớp một tiểu học thêm mỗi ngày 40 phút ngoài trời, sau ba năm số trẻ mới cận thị giảm từ khoảng bốn phần mười xuống khoảng ba phần mười. Cơ quan y tế đưa ra mức là mỗi ngày ban ngày ở ngoài trời không dưới 2 giờ, hoặc cộng lại trong tuần là 14 giờ. Trời âm u vẫn tính, giờ giải lao ra ngoài đứng cũng tính. +- Lợi ích: Học sinh lớp một ở 12 trường tiểu học thành phố Quảng Châu được chia nhóm bằng bốc thăm theo từng trường (952 người ở trường can thiệp, 951 người ở trường đối chứng, tuổi trung bình 6.6 tuổi). Cách can thiệp là mỗi ngày học thêm 1 tiết hoạt động ngoài trời 40 phút, đồng thời khuyến khích cha mẹ đưa trẻ ra ngoài nhiều hơn ngoài giờ học. Trong 3 năm, tỷ lệ cận thị mới ở nhóm can thiệp là 30.4%, tức 259 ca trong số 853 người đủ điều kiện phân tích; nhóm đối chứng là 39.5%, tức 287 ca trong số 726 người. Chênh lệch giữa hai nhóm là -9.1 điểm phần trăm (95% CI -14.1 đến -4.1, P<0.001), chênh lệch thật nằm trong khoảng thấp hơn 4 đến 14 điểm phần trăm. Mức tăng độ trong 3 năm (thay đổi khúc xạ cầu tương đương) là -1.42 D so với -1.59 D, chênh 0.17 D (P=0.04). Nhóm can thiệp tăng độ ít hơn một chút. D là đơn vị đo độ. Sự dài ra của trục mắt (chiều dài từ trước ra sau của nhãn cầu) thì không thấy khác biệt giữa hai nhóm (P=0.07). 《Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên》《防控儿童青少年近视核心知识十条》yêu cầu trẻ em và thanh thiếu niên mỗi ngày ban ngày hoạt động ngoài trời không dưới 2 giờ, hoặc cộng lại trong tuần không dưới 14 giờ. Tài liệu ghi rõ kể cả trời âm u thì hoạt động ngoài trời vẫn có tác dụng bảo vệ, và yêu cầu tránh khung giờ nắng nóng gay gắt sau buổi trưa +- Mức bằng chứng: A +- Nguồn: He M, et al. (2015). Effect of Time Spent Outdoors at School on the Development of Myopia Among Children in China: A Randomized Clinical Trial. JAMA. ；Văn phòng của Ủy ban Quốc gia Y tế (国家卫生健康委办公厅) (2023). Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên, 防控儿童青少年近视核心知识十条（国卫办妇幼函〔2023〕278 号）. +- Ghi chú: Thử nghiệm này chỉ kéo dài 3 năm, và trẻ tham gia đều bắt đầu từ 6 tuổi. Chênh lệch 9.1 điểm phần trăm về tỷ lệ cận thị chỉ đủ để một phần trẻ không bị cận thị, không bảo đảm ai là không bị. Cùng bộ Mười điểm cốt lõi đó còn đưa ra vài cách làm. Đọc và viết thì giữ「cách mặt giấy một chân, cách mắt một nắm tay, cách đầu bút một ngón tay」. Sau 20 phút nhìn gần, phải nhìn xa từ 20 giây trở lên. Học sinh tiểu học và trung học dùng màn hình điện tử không phải để học thì mỗi lần không quá 15 phút, cộng lại trong ngày không quá 1 giờ.「Ở ngoài trời」và「học thể dục」không phải một chuyện, giờ giải lao ra ngoài đứng cũng được tính. + +### 5. Không cho màn hình ở tuổi 0 đến 3, hạn chế tối đa ở tuổi 3 đến 6, học sinh tiểu học và trung học dùng màn hình không phải để học thì mỗi ngày không quá 1 giờ + +- Chi phí: Không tốn tiền. Khó ở chỗ phải lấy màn hình ra khỏi tay trẻ, và cũng khó ở chỗ bố mẹ tự bớt xem trước mặt con. +- Nói thẳng: Cơ quan y tế quốc gia yêu cầu không cho màn hình ở tuổi 0 đến 3, hạn chế tối đa ở tuổi 3 đến 6, học sinh tiểu học và trung học dùng màn hình không phải để học thì mỗi lần không quá 15 phút, mỗi ngày không quá 1 giờ. Lý do là nhìn gần lâu sẽ khiến cận thị tới sớm. Bằng chứng về màn hình và cận thị không chắc bằng, cái thật sự thử ra hiệu quả là mỗi ngày 2 giờ ở ngoài trời. +- Lợi ích: Điều bốn của《Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên》《防控儿童青少年近视核心知识十条》viết nguyên văn:「0～3岁婴幼儿不使用手机、平板、电脑等视屏类电子产品；3～6岁幼儿尽量避免其接触和使用手机、平板、电脑等视屏类电子产品；中小学生非学习目的使用电子屏幕单次时长不宜超过15分钟，每天累计时长不宜超过1小时」, tức trẻ sơ sinh và trẻ nhỏ 0 đến 3 tuổi không dùng điện thoại, máy tính bảng, máy tính; trẻ 3 đến 6 tuổi nên tránh tiếp xúc và dùng các thiết bị đó; học sinh tiểu học và trung học dùng màn hình không phải để học thì mỗi lần không nên quá 15 phút, tổng trong ngày không nên quá 1 giờ. Cùng điều đó nói rõ「长时间近距离使用电子视屏类产品，易消耗儿童远视储备量，是儿童青少年近视早发、高发的重要原因」, tức dùng sản phẩm màn hình ở khoảng cách gần trong thời gian dài dễ làm hao hụt dự trữ thị lực của trẻ, là một nguyên nhân quan trọng khiến cận thị ở trẻ em và thanh thiếu niên tới sớm và xuất hiện nhiều. Bản triển khai năm 2018 của tám bộ, trong đó có Bộ Giáo dục, đưa ra cùng yêu cầu với phía bố mẹ, và bổ sung một câu「年龄越小，连续使用电子产品的时间应越短」, tức tuổi càng nhỏ thì thời gian dùng liên tục thiết bị điện tử càng phải ngắn. Bằng chứng nghiên cứu yếu hơn nhiều: một tổng quan hệ thống thu 33 nghiên cứu, 11 nghiên cứu được đưa vào phép gộp, đối tượng từ 3 tháng tuổi đến 33 tuổi. Chỉ tính điện thoại và máy tính bảng thì tỷ số odds cận thị là 1.26 (95% CI 1.00 đến 1.60, I²=77%). Gộp cả máy tính thì là 1.77 (1.28 đến 2.45, I²=87%). Kết luận của tác giả chỉ viết「可能相关」, và nói rõ trong 33 nghiên cứu không nghiên cứu nào có phép đo thời gian dùng màn hình đáng tin cậy +- Mức bằng chứng: B +- Nguồn: Văn phòng của Ủy ban Quốc gia Y tế (国家卫生健康委办公厅) (2023). Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên, 防控儿童青少年近视核心知识十条（国卫办妇幼函〔2023〕278 号）. ; Bộ Giáo dục và bảy bộ khác (教育部等八部门) (2018). Phương án triển khai phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên toàn diện, 综合防控儿童青少年近视实施方案（教体艺〔2018〕3 号）. ; Foreman J, Salim AT, Praveen A, et al. (2021). Association between digital smart device use and myopia: a systematic review and meta-analysis. The Lancet Digital Health, 3(12), e806-e818. ; Madigan S, McArthur BA, Anhorn C, Eirich R, Christakis DA (2020). Associations between screen use and child language skills: a systematic review and meta-analysis. JAMA Pediatrics, 174(7), 665-675. +- Ghi chú: Cho mức B là vì cái cơ quan đưa ra là quy định chia theo nhóm tuổi, không phải con số đo từ thử nghiệm. Nghiên cứu cho tỷ số odds có cận dưới vừa đúng ở 1.00, nên tác giả chỉ dám viết「可能相关」. Vì vậy cứ coi đây là một quy tắc gia đình dễ thực hiện, đừng hiểu là「không xem màn hình thì không cận thị」. Cái có thử nghiệm ngẫu nhiên bảo chứng là mục 4 (2 giờ ở ngoài trời). Trẻ mẫu giáo cũng cần khám mắt: cùng tài liệu đó ghi rõ 1～3 tuổi, 4～6 tuổi và sau 7 tuổi đều nên làm khám khúc xạ định kỳ để xem còn lại bao nhiêu dự trữ thị lực, cách khám xem mục 12 của phần này (khám khúc xạ sau khi nhỏ mắt). Quan hệ giữa màn hình và khả năng ngôn ngữ của trẻ cũng đã có phân tích tổng hợp: 42 nghiên cứu, 18905 người, dùng màn hình lâu hơn thì khả năng ngôn ngữ yếu hơn (r=−0.14, 95% CI −0.18 đến −0.10). Nhưng xem chung, xem chương trình có nội dung giáo dục lại ra tương quan thuận (r=0.16 và 0.13). Đây là tương quan chứ không phải nhân quả, chỉ cho thấy dùng màn hình thay cho việc nói chuyện với trẻ là không đáng. + +### 6. Hãy coi cận thị là một tổn thương còn lại: độ càng cao thì sau này càng nhiều bong võng mạc, thoái hoa điểm vàng và glôcôm + +- Chi phí: Không tốn tiền. Bản thân cận thị không cần làm gì, việc phải làm nằm ở mục 4 và 5 của phần này (ngoài trời 2 giờ, quản lý màn hình theo tuổi). +- Nói thẳng: Cận thị không tự khỏi, độ cũng không giảm lại được. Nó không chỉ là chuyện「đeo kính」. So với người không cận thị, khả năng bị thoái hoa hoàng điểm liên quan cận thị cao hơn mười mấy lần ở mức nhẹ, cao hơn hàng trăm lần ở mức trên 600 độ. Bong võng mạc cao khoảng 3 lần ở mức nhẹ, khoảng 13 lần ở mức trên 600 độ. Nên trẻ càng cận thị sớm, độ tăng càng nhanh thì nguy cơ phải gánh về sau càng lớn. +- Lợi ích: Một tổng quan hệ thống và phân tích tổng hợp gom các nghiên cứu trước tháng 6 năm 2019, chia theo độ thành ba nhóm: nhẹ (−0.5 đến −3.00 D), trung bình (−3.00 đến −6.00 D), cao (≤−6.00 D, trên 600 độ). So với người không cận thị, tỷ số odds thoái hoa hoàng điểm liên quan cận thị lần lượt là 13.57 (95% CI 6.18 đến 29.79), 72.74 (33.18 đến 159.48), 845.08 (230.05 đến 3104.34). Bong võng mạc lần lượt là 3.15 (1.92 đến 5.17), 8.74 (7.28 đến 10.50), 12.62 (6.65 đến 23.94). Đục thủy tinh sau bao lên lần lượt là 1.56 (1.32 đến 1.84), 2.55 (1.98 đến 3.28), 4.55 (2.66 đến 7.75). Glôcôm góc mở, nhẹ là 1.59 (1.33 đến 1.91), gộp trung bình và cao là 2.92 (1.89 đến 4.52). Tỷ số odds tổn thương thị lực ở người trên 60 tuổi, ba nhóm lần lượt là 1.71 (1.07 đến 2.74), 5.54 (3.12 đến 9.85), 87.63 (34.50 đến 222.58). Cách nói của《Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên》cũng giống: trên 600 độ là cận thị cao, trong nhóm này tỷ lệ mắc bệnh mắt gây mù như đục thủy tinh, glôcôm góc mở, thoái hoa hoàng điểm, bong võng mạc, bệnh thị giác thần kinh liên quan cận thị rõ ràng cao hơn các nhóm khác, và「近视可防可控不可逆」 +- Mức bằng chứng: A +- Nguồn: Haarman AEG, Enthoven CA, Tideman JWL, Tedja MS, Verhoeven VJM, Klaver CCW (2020). The complications of myopia: a review and meta-analysis. Investigative Ophthalmology & Visual Science, 61(4), 49. ; Văn phòng của Ủy ban Quốc gia Y tế (国家卫生健康委办公厅) (2023). Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên, 防控儿童青少年近视核心知识十条（国卫办妇幼函〔2023〕278 号）. +- Ghi chú: Những nghiên cứu được gom vào đây đều là chỉ ghi nhận và theo dõi, tỷ số odds lớn đến mức bất thường. Khoảng tin cậy của mục thoái hoa hoàng điểm ở cận thị cao trải từ 230 đến 3104, nghĩa là các nghiên cứu bên dưới rất khác nhau, con số này chỉ xem được độ lớn chứ không dùng làm giá trị chính xác. Phần lớn các biến chứng này chỉ lộ ra khi đã trung niên hoặc cao niên. Bản thân cận thị không phải là tàn tật, đa số người đeo kính là nhìn bình thường. Nhưng cận thị không phải là chuyện cứ mua một cặp kính là xong, đáng bỏ công sức để hoãn lại và kiềm soát nó. Sau khi xác định chẩn đoán thì tái khám theo bao lâu xem mục 12 của phần này (khám khúc xạ sau khi nhỏ mắt). Đừng mua sản phẩm quảng cáo là chữa khỏi, xem mục 9 của phần này (chữa khỏi cận thị). + +### 7. Báo cáo khám sức khoẻ học sinh hằng năm phải tự đọc một lượt, mục bất thường thì đưa đi khám trong năm đó + +- Chi phí: Không tốn tiền. Tiền khám sức khoẻ học sinh trong giai đoạn giáo dục bắt buộc do ngân sách công của trường chi trả, bố mẹ không phải bỏ thêm ra. +- Nói thẳng: Trường tiểu học và trung học tổ chức khám sức khoẻ học sinh một lần mỗi năm. Những mục cần chú ý trên báo cáo là thừa cân, béo phì, suy dinh dưỡng, bất thường cong cột sống, nhìn hở, sâu răng, và đây là những mục dễ bị lỡ qua nhất. Đừng chỉ liếc qua chiều cao cân nặng rồi cất tờ giấy đi. Có mục nào bất thường thì đưa trẻ đi khám chuyên khoa trong năm đó. +- Lợi ích:《Biện pháp quản lý khám sức khoẻ học sinh (bản 2021)》《中小学生健康体检管理办法（2021 年版）》quy định trường tiểu học và trung học tổ chức 1 lần khám sức khoẻ cho học sinh trong trường mỗi năm. Khám ngoại bao gồm đầu, cổ, ngực, cột sống, tứ chi, da, hạch bạch huyết. Khám mắt bao gồm bên ngoài mắt, thị lực xa, độ khúc xạ (độ cận thị và lõm mắt). Báo cáo phát cho từng cá nhân phải có kết quả tìm được, đánh giá chung và khuyến nghị hướng dẫn sức khoẻ. Trong đó thừa cân, béo phì, suy dinh dưỡng, bất thường cong cột sống, nhìn hở, sâu răng (sâu răng) phải là trọng tâm hướng dẫn. Cơ quan quản lý giáo dục phải lập hồ sơ và theo dõi tiếp đối với học sinh có vấn đề sức khoẻ (ghi lại, rồi theo dõi sau). Phí khám sức khoẻ học sinh trong giai đoạn giáo dục bắt buộc do ngân sách công của trường chi trả, bố mẹ không phải bỏ thêm ra (toàn quốc, ban hành tháng 9 năm 2021) +- Mức bằng chứng: B +- Nguồn: Ủy ban Quốc gia Y tế và Bộ Giáo dục (国家卫生健康委、教育部) (2021). Biện pháp quản lý khám sức khoẻ học sinh tiểu học và trung học (bản 2021), 中小学生健康体检管理办法（2021 年版）（国卫医发〔2021〕29 号）. +- Ghi chú: Cho mức B vì đây là văn bản quy định trường phải làm thế nào, không phải nghiên cứu đo hiệu quả. Nó bảo đảm「mỗi năm khám, khám đúng những mục này, thấy bất thường thì phải theo dõi」, chứ không bảo đảm khám ra xong sẽ có người đứng ra xử lý. Khám sức khoẻ chỉ là sàng lọc một lượt, không phải chẩn đoán. Nhìn hở thì đi mắt làm khám khúc xạ sau khi nhỏ mắt (mục 12 của phần này). Bất thường cong cột sống thì đi khoa chỉnh hình hoặc khoa ngoại cột sống (mục 2 của phần này, đừng trì hoãn chữa trị vì chờ thi xong). Hai bước này đều phải bố mẹ tự đưa trẻ đi. + +### 8. Cho trẻ 12 đến 18 tuổi làm một lần sàng lọc trầm cảm, đừng lấy bài tâm lý học đường làm chẩn đoán + +- Chi phí: 0 đến mấy chục tệ, điền một bảng khảo sát sàng lọc ở trạm y tế cộng đồng hoặc phòng khám bệnh viện. Hai đường dây 12355 và 12356 gọi thì không mất tiền. +- Nói thẳng: Hướng dẫn phòng ngừa của Mỹ khuyến nghị sàng lọc trầm cảm nặng cho trẻ 12 đến 18 tuổi, bằng chứng ở trẻ 11 tuổi trở xuống chưa đủ. Tự sát là nguyên nhân tử vong thứ hai ở trẻ 10 đến 19 tuổi. Bài tâm lý học đường hằng năm chỉ là khảo sát toàn trường, không thay được chẩn đoán. Trẻ nói khó chịu, hoặc ăn, ngủ, sở thích, thành tích đổi bất thường thì đăng ký thẳng khoa Tâm thần hoặc khoa Tâm lý lâm sàng. +- Lợi ích: Nhóm dịch vụ phòng ngừa của Mỹ ra tuyên bố khuyến nghị năm 2022. Tuyên bố khuyến nghị sàng lọc rối loạn trầm cảm nặng cho thanh niên 12 đến 18 tuổi (khuyến nghị loại B). Loại B nghĩa là mức chắc chắn trung bình, tính ra lợi ích cũng trung bình, đáng làm. Sàng lọc trầm cảm nặng cho trẻ 11 tuổi trở xuống thì bằng chứng không đủ (tuyên bố I, tức chưa đủ bằng chứng, không đưa ra khuyến nghị). Sàng lọc riêng nguy cơ tự sát cho trẻ em và thanh thiếu niên thì cả bằng chứng về lợi ích lẫn bất lợi đều không đủ (tuyên bố I). Tài liệu chỉ ra tự sát là nguyên nhân tử vong thứ hai ở nhóm tuổi 10 đến 19. Tài liệu còn chỉ ra trầm cảm nặng ở trẻ em và thanh thiếu niên liên quan mật thiết đến nhiều vấn đề về sau. Các vấn đề đó gồm trầm cảm tái phát khi đã trưởng thành, rối loạn tâm thần khác, và tăng nguy cơ có ý nghĩ tự sát, tự sát không thành công, tự sát tử vong. 《Mười biện pháp tăng cường công tác sức khoẻ tâm lý cho học sinh tiểu học và trung học》《进一步加强中小学生心理健康工作十条措施》yêu cầu xây dựng hệ thống ba cấp gồm giáo viên chủ nhiệm, giáo viên tâm lý chuyên trách và kiêm nhiệm, bác sĩ tâm thần. Mỗi học sinh được đo tâm lý không quá 1 lần mỗi năm học. Tài liệu còn yêu cầu thúc đẩy cơ chế giới thiệu khám chữa bệnh tâm thần và học lại sau khi khỏi bệnh. Nó yêu cầu phát huy vai trò của bàn dịch vụ thanh niên 12355 và đường dây trợ giúp tâm lý 12356 (toàn quốc, ban hành tháng 10 năm 2025) +- Mức bằng chứng: B +- Nguồn: US Preventive Services Task Force (2022). Screening for Depression and Suicide Risk in Children and Adolescents: US Preventive Services Task Force Recommendation Statement. JAMA. ；Văn phòng của Bộ Giáo dục (教育部办公厅) (2025). Mười biện pháp tăng cường công tác sức khoẻ tâm lý cho học sinh tiểu học và trung học, 进一步加强中小学生心理健康工作十条措施（教基厅〔2025〕2 号）. +- Ghi chú: Cho mức B vì nguyên bản đưa ra là xếp hạng mức khuyến nghị, không phải hiệu quả quy đổi được thành số. Đó còn là khuyến nghị của Mỹ, trong nước không có khuyến nghị sàng lọc chính thức tương ứng. Kết quả sàng lọc dương tính chỉ nói là cần đánh giá lại một lần, không có nghĩa là đã chẩn đoán. Khi trẻ có ý nghĩ tự sát thì xử lý xem mục 25 ở phần 1. Cách xử lý ở trong nhà xem phần 29. + +### 9. Đừng mua sản phẩm và dịch vụ quảng cáo là có thể「chữa khỏi cận thị」「giảm độ」 + +- Chi phí: Không tốn tiền, ngược lại còn tiết kiệm được khoản tiền đó. +- Nói thẳng: Cục Quản lý thị trường quốc gia nói thẳng「在目前医疗技术条件下，近视不能治愈」, tức với kỹ thuật y tế hiện nay cận thị không thể chữa khỏi. Quảng cáo「恢复」「降低度数」「近视治愈」「度数修复」loại này thì phải xử lý nghiêm. Thấy các từ đó là loại thẳng, không cần so giá nữa. Máy bảo vệ mắt là thiết bị y tế, sàn thương mại điện tử có trách nhiệm kiểm tra người bán có đủ điều kiện bán không. +- Lợi ích: Thông báo về đợt chuyên đợc xử lý của Văn phòng Cục Quản lý thị trường quốc gia, từ tháng 11 năm 2021 đến tháng 3 năm 2022, nói rõ:「在目前医疗技术条件下，近视不能治愈」. Thông báo yêu cầu xử lý nghiêm theo pháp luật đối với hành vi tiếp thị sản phẩm phòng và kiểm soát cận thị cho trẻ em và thanh thiếu niên có cách diễn đạt gây hiểu lầm như「康复」「恢复」「降低度数」「近视治愈」「近视克星」「度数修复」. Thông báo yêu cầu tăng cường trấn áp, và cái nó quản là hành vi mượn danh y học cổ truyền Trung Quốc, hoặc lợi dụng lý thuyết kỹ thuật y học cổ truyền để tiếp thị quảng cáo lừa dối người tiêu dùng. Thông báo còn yêu cầu đánh nghiêm những loại quảng cáo sai lệch và vi phạm: dùng tên hoặc hình ảnh người bệnh để so sánh hiệu quả trước và sau chữa trị, và cam kết chắc chắn về độ an toàn, hiệu quả của phương pháp chữa trị. Thông báo yêu cầu sàn thương mại điện tử thực hiện nghiêm trách nhiệm thẩm định điều kiện kinh doanh đối với thiết bị y tế như máy bảo vệ mắt và hàng đặc biệt liên quan (xem rõ người bán có đủ điều kiện bán không) (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Văn phòng Cục Quản lý thị trường quốc gia (市场监管总局办公厅) (2021). Thông báo về việc tổ chức đợt chuyên đợc xử lý tiếp thị và quảng cáo thương mại trái phép với sản phẩm phòng và kiểm soát cận thị cho trẻ em và thanh thiếu niên. +- Ghi chú: Cho mức B vì căn cứ là cách xác định của cơ quan quản lý khi xử lý vi phạm, không phải số đo từ thử nghiệm đối chứng sản phẩm. Nhưng câu「近视不能治愈」thì cơ quan y tế và cơ quan quản lý nói nhất quán: trong《Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên》viết là「近视可防可控不可逆」. Hai việc thật sự có bằng chứng nằm ở mục 4 của phần này (ngoài trời) và mục 12 của phần này (khám khúc xạ và đeo kính đúng quy định, tái khám). Kính chắn ánh sáng xanh xem phần 6. + +### 10. Giấc ngủ, bài tập, thể dục và xếp hạng đều có quy định bằng văn bản, trường không làm được thì có thể kiến nghị + +- Chi phí: Không tốn tiền. Biết nội dung quy định là đủ, khi cần thì phản ánh với nhà trường hoặc cơ quan quản lý giáo dục địa phương. Chỗ khó là phải mở lời nêu ra. +- Nói thẳng: Văn bản ghi rõ học sinh tiểu học ngủ 10 giờ mỗi ngày, học sinh trung học 9 giờ, học sinh trung học phổ thông 8 giờ. Tiết học buổi sáng ở tiểu học thường không sớm hơn 8:20, ở trung học không sớm hơn 8:00. Năm 2025 lại quy định mỗi ngày hoạt động thể chất không dưới 2 giờ, không được xếp hạng học sinh theo điểm thi. Con ngủ thiếu nhiều ngày liên tục thì lấy mấy quy định này ra đối chiếu trước, đừng vội trách con dây dưa. +- Lợi ích: Thông báo về việc tăng cường quản lý giấc ngủ học sinh tiểu học và trung học quy định: học sinh tiểu học phải ngủ đủ 10 giờ mỗi ngày, học sinh trung học 9 giờ, học sinh trung học phổ thông 8 giờ. Giờ bắt đầu tiết học buổi sáng ở tiểu học thường không sớm hơn 8:20, ở trung học không sớm hơn 8:00. Trường không được yêu cầu học sinh đến trường sớm để tham gia hoạt động giáo dục dạy học thống nhất. Học sinh tiểu học đi ngủ thường không muộn hơn 21:20, học sinh trung học không muộn hơn 22:00, học sinh trung học phổ thông không muộn hơn 23:00. Cơ sở đào tạo ngoài trường phải kết thúc buổi đào tạo không muộn hơn 20:30, lớp đào tạo phát trực tuyến kết thúc không muộn hơn 21:00. Từ 22:00 đến 8:00 hôm sau không được cung cấp dịch vụ trò chơi cho người chưa thành niên (toàn quốc, ban hành tháng 3 năm 2021). Mười biện pháp tăng cường công tác sức khoẻ tâm lý cho học sinh tiểu học và trung học có thêm mấy quy định: thực hiện đầy đủ thời gian hoạt động thể chất tổng hợp mỗi ngày của học sinh tiểu học và trung học không dưới 2 giờ, khuyến khích thực hiện「giờ giải lao 15 phút」. Văn bản còn quy định không được xếp hạng học sinh theo điểm thi, nghiêm cấm giao bài tập lặp lại máy móc hoặc mang tính trừng phạt, khuyến khích mỗi tuần dành một ngày「không có bài tập」 (toàn quốc, ban hành tháng 10 năm 2025) +- Mức bằng chứng: B +- Nguồn: Văn phòng Bộ Giáo dục (教育部办公厅) (2021). Thông báo về việc tăng cường quản lý giấc ngủ học sinh tiểu học và trung học（教基厅函〔2021〕11 号）. ；Văn phòng Bộ Giáo dục (教育部办公厅) (2025). Mười biện pháp tăng cường công tác sức khoẻ tâm lý cho học sinh tiểu học và trung học（教基厅〔2025〕2 号）. +- Ghi chú: Cả hai đều là văn bản của cơ quan chủ quản. Mức độ các trường các nơi thực hiện chênh nhau rất nhiều, nên mục này để mức B. Chúng có ích là để bạn biết việc nào trường vốn đã phải làm, để khi đàm phán bạn có căn cứ trong tay. Về giấc ngủ của người lớn, xem phần 2 và 3. + +### 11. Con không chịu nổi thì có thể xin nghỉ học, trường bắt buộc giữ học vị cho cháu, tối đa 1 năm + +- Chi phí: Không tốn tiền. Cha mẹ hoặc người giám hộ hợp pháp khác nộp đơn lên nhà trường. Chỗ khó là phải chấp nhận cho con dừng học một thời gian. +- Nói thẳng: Nghỉ học có thủ tục chính thức. Cha mẹ hoặc người giám hộ khác nộp đơn, nhà trường xác nhận rồi báo lên cấp trên để phê duyệt. Một lần tối đa 1 năm, hết hạn còn nghỉ tiếp thì phải nộp đơn lại. Trong thời gian nghỉ học, nhà trường phải giữ học vị cho con, khi trở lại học thì nối tiếp. Cách làm cụ thể do từng tỉnh quy định, hỏi phòng giáo vụ của trường hoặc sở giáo dục địa phương. +- Lợi ích: Quy định quản lý học vị học sinh tiểu học và trung học《中小学生学籍管理办法》Điều mười sáu quy định học sinh nghỉ học thì làm thế nào. Việc học sinh nghỉ học do cha mẹ hoặc người giám hộ hợp pháp khác nộp đơn. Nhà trường xác nhận rồi báo lên cơ quan quản lý học vị cấp trên để phê duyệt. Khi trở lại học, nhà trường phải làm thủ tục liên quan kịp thời. Thời hạn một lần xin nghỉ học không quá 1 năm, hết hạn mà thật sự còn nghỉ tiếp thì phải nộp đơn xin lại. Trong thời gian học sinh nghỉ học, nhà trường phải giữ học vị cho học sinh (tư cách và chỗ đứng của học sinh được giữ lại). Cách làm cụ thể về nghỉ học và trở lại học do cơ quan quản lý giáo dục cấp tỉnh quy định (toàn quốc, ban hành tháng 1 năm 2025). Mười biện pháp tăng cường công tác sức khoẻ tâm lý cho học sinh tiểu học và trung học yêu cầu đẩy nhanh việc xây dựng cơ chế giới thiệu chuyển tuyến khám bệnh tâm thần và trở lại học sau khi hồi phục, hướng dẫn nhà trường xây dựng kế hoạch ứng phó với khủng hoảng tâm lý bất thường xảy ra trong trường +- Mức bằng chứng: B +- Nguồn: Bộ Giáo dục (教育部) (2025). Quy định quản lý học vị học sinh tiểu học và trung học《中小学生学籍管理办法》（教基〔2025〕1 号，Điều mười sáu）. ；Văn phòng Bộ Giáo dục (教育部办公厅) (2025). Mười biện pháp tăng cường công tác sức khoẻ tâm lý cho học sinh tiểu học và trung học（教基厅〔2025〕2 号）. +- Ghi chú: Trước khi đi làm thủ tục thì hỏi rõ ba việc: cần xin loại giấy tờ y tế nào, khoảng thời gian nghỉ học thì học vị được ghi thế nào trong hệ thống, khi trở lại học là về nguyên lớp hay xuống một lớp. Để mức B là vì điều kiện nghỉ học, hồ sơ phải nộp và cách nối lại khi trở lại học đều do cơ quan giáo dục cấp tỉnh tự quy định, phần cấp quốc gia chỉ dựng khung. Quy định về bảo vệ học sinh vị thành niên《未成年人学校保护规定》còn yêu cầu trường giáo dục bắt buộc lập hồ sơ báo cáo cho học sinh bỏ học, nghỉ học và nghỉ phép dài ngày, nên nghỉ học là việc trường đã làm qua, không phải mở ngoại lệ riêng cho bạn. + +### 12. Khám ra nhìn kém thì ra bệnh viện khám khúc xạ sau khi nhỏ mắt, sau đó tái khám theo khoảng cách bác sĩ chỉ định + +- Chi phí: Phí khám và phí kính từ mấy chục tới mấy trăm tệ. Sau khi nhỏ mắt, trẻ sợ ánh sáng và nhìn gần bị mờ, phải nửa ngày đến một hai ngày mới hết. +- Nói thẳng: Khám khúc xạ sau khi nhỏ mắt là nhỏ một loại thuốc nhỏ mắt trước, làm cơ điều tiêu trong mắt thả lỏng, rồi mới đo độ. Việc này làm ở bệnh viện, và là cách chẩn đoán cận thị chính xác nhất. Cơ quan y tế nói rõ loại thuốc này không gây tổn thương cho con mắt khỏe mạnh, ngừng thuốc là mắt trở lại bình thường. Đừng lấy một lần đo bằng máy tính ở tiệm kính mắt làm chẩn đoán, cũng đừng vì sợ「độ tăng」mà không cho con đeo kính. +- Lợi ích: 《Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên》《防控儿童青少年近视核心知识十条》giải thích rõ khám khúc xạ sau khi nhỏ mắt là gì. Trước tiên dùng thuốc làm liệt cơ thần kinh thị giác, rồi đo khúc xạ. Thuốc làm liệt cơ thần kinh thị giác là thuốc nhỏ mắt làm cơ điều tiêu trong mắt thả lỏng. Khám khúc xạ sau khi nhỏ mắt là đo khúc xạ bằng y học tại bệnh viện, là「tiêu chuẩn vàng」để chẩn đoán cận thị (cách chính xác nhất). Thuốc nhỏ mắt không gây tổn thương cho con mắt khỏe mạnh, sau khi dùng có thể xuất hiện tạm thời sợ ánh sáng và nhìn gói không rõ, ngừng thuốc là trở lại bình thường. Đã chẩn đoán cận thị thì phải can thiệp khoa học kịp thời, đeo kính và các cách khác để nắn chỉnh. Đeo kính đúng số có thể nắn chỉnh hiệu quả tình trạng nhìn mờ và làm chậm tiến trình cận thị. Người đeo kính mà thị lực bình thường thì trẻ mẫu giáo và học sinh tiểu học tái khám mỗi 3 đến 6 tháng một lần, học sinh trung học và trung học phổ thông tái khám mỗi 6 đến 12 tháng một lần, tuỳ tình hình cụ thể mà quyết định có đổi kính hay không. Trẻ em ở độ tuổi 1 đến 3, 4 đến 6, và sau 7 tuổi nên được khám khúc xạ định kỳ (đo độ), theo dõi dự trữ thị lực còn lại (toàn quốc, ban hành tháng 7 năm 2023) +- Mức bằng chứng: B +- Nguồn: Văn phòng của Ủy ban Quốc gia Y tế (国家卫生健康委办公厅) (2023). Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên, 防控儿童青少年近视核心知识十条（国卫办妇幼函〔2023〕278 号）. +- Ghi chú: Để mức B là vì đây là cách làm do tài liệu phổ biến chính thức đưa ra, không phải nghiên cứu gốc có số liệu cụ thể. Cận thị nặng sẽ kéo theo những bệnh như bong võng mạc, nên tái khám không chỉ để đổi kính. Kết quả「nhìn kém」trong khám sức khoẻ học sinh chỉ là kết quả sàng lọc ban đầu, vẫn còn mang con đi bệnh viện làm một lần khám mắt đầy đủ, xem mục 7 của phần này. + +### 13. Sau khi răng cối chính mọc lên thì làm hàn khe răng + +- Chi phí: Mỗi răng mấy chục tới một hai trăm tệ. Không gây tê, cũng không phải mài răng, mỗi lần mấy phút. Nhiều nơi có chương trình miễn phí cho trẻ đủ tuổi. +- Nói thẳng: Hàn khe răng là dùng nhựa phủ kín các rãnh ở mặt nhai của răng lớn, thức ăn vụn không mắc vào được. Làm xong hai năm, cơ hội bị sâu răng còn khoảng một phần tám so với không hàn. Không hàn thì 16% mặt răng bị sâu, hàn rồi chỉ khoảng 5%. Sau bốn năm vẫn còn tác dụng. +- Lợi ích: 38 thử nghiệm, 7924 trẻ em từ 5 đến 16 tuổi, Cochrane tìm đủ rồi gộp lại tính. So sánh chất hàn răng loại nhựa với không hàn. Ở răng cối chính thứ nhất của trẻ 5 đến 10 tuổi, sau 24 tháng thì OR bị sâu răng là 0.12 (95% CI 0.08–0.19), cơ hội bị sâu răng khoảng một phần tám so với không hàn. Con số này từ 7 thử nghiệm, 1548 người được chia nhóm bằng bốc thăm, 1322 người cuối cùng được khám, chất lượng bằng chứng ở mức trung bình. Nếu nhóm đối chứng có 16% mặt răng bị sâu trong 24 tháng thì sau khi hàn là 5.2% (3.13%–7.37%). Khi nhóm đối chứng là 40% thì là 6.25% (3.84%–9.63%). Khi nhóm đối chứng là 70% thì là 19% (12.3%–27.2%). Sau 48 đến 54 tháng thì OR là 0.21 (0.16–0.28), vẫn còn khoảng một phần năm so với không hàn. Kết luận của tác giả là chất hàn răng loại nhựa ở 24 tháng làm giảm sâu răng so với không hàn từ 11% đến 51%. Trong 4 thử nghiệm có ghi nhận tác dụng phụ thì không thấy tác dụng phụ nào. Về việc loại vật liệu hàn khác như kính ion thì tốt hơn nhau ở chỗ nào, bằng chứng chưa đủ +- Mức bằng chứng: A +- Nguồn: Ahovuo-Saloranta A, et al. (2017). Pit and fissure sealants for preventing dental decay in permanent teeth. Cochrane Database of Systematic Reviews. +- Ghi chú: Răng cối chính thứ nhất (răng lớn không thay nữa) thường mọc khoảng 6 tuổi, răng thứ hai khoảng 12 tuổi. Hàn phải làm lúc răng vừa mọc, chưa hỏng, đã sâu rồi thì chỉ còn vá. Nhiều khu vực đưa hàn khe răng cho trẻ đủ tuổi vào danh sách chương trình miễn phí, hỏi trung tâm kiểm soát bệnh địa phương hoặc trạm y tế cộng đồng. Trong phần này không tìm được văn bản phạm vi toàn quốc nào đối chiếu từng chữ, nên không viết phạm vi bảo hiểm và nhóm tuổi. Sâu răng cũng là một trong các mục trọng tâm hướng dẫn trong khám sức khoẻ học sinh, xem mục 7 của phần này. + +### 14. Con thích chơi game thì xem giấc ngủ, bài tập và ra ngoài vận động có bị chen mất không, đừng chỉ nhìn thời gian chơi + +- Chi phí: Không tốn tiền. Mỗi tuần nên biết một lần con đi ngủ lúc mấy giờ, bài tập có nộp không, có ra ngoài vận động không. +- Nói thẳng: Số thật sự đủ chuẩn để gọi là「rối loạn vì game」xấp xỉ hai ba người trong mỗi trăm người. Ngưỡng trong bảng hỏi đặt thấp, lại chỉ khảo sát thiếu niên thì tỷ lệ sẽ thổi lên cao. Xem số liệu thật của các công ty game thì chơi nhiều hay ít gần như không liên quan tới tâm trạng tốt xấu. Người chưa thành niên theo quy định chỉ chơi được từ 8 giờ đến 9 giờ tối vào thứ Sáu, thứ Bảy, chủ nhật và ngày lễ. +- Lợi ích: Một phân tích tổng hợp gom 53 nghiên cứu từ 17 nước, 226247 người. Tỷ lệ mắc rối loạn vì game là 3.05% (95% CI 2.38–3.91, đây là khoảng tin cậy). Chỉ tính những nghiên cứu lấy mẫu nghiêm hơn thì là 1.96%. Kết quả giữa các nghiên cứu chênh nhau rất xa, 77% phần chênh lệch là do dùng bảng hỏi nào. Mẫu là thiếu niên, điểm ngưỡng thấp, mẫu nhỏ, đều làm tỷ lệ lệch lên cao. Một nghiên cứu khác lấy được dữ liệu của 7 công ty game, nối thời lượng chơi thật trong sáu tuần của 38935 người chơi với mức hạnh phúc tự điền của họ. Gần như không tìm được bằng chứng cho thấy thời lượng chơi ảnh hưởng tới hạnh phúc. Cục Quản lý Báo chín và Xuất bản quy định doanh nghiệp game chỉ được cung cấp 1 giờ dịch vụ cho người chưa thành niên vào thứ Sáu, thứ Bảy, chủ nhật và ngày lễ hợp pháp, mỗi ngày từ 20 giờ đến 21 giờ (toàn quốc / quốc tế, từ năm 2021) +- Mức bằng chứng: B +- Nguồn: Stevens MW, Dorstyn D, Delfabbro PH, King DL. (2021). Global prevalence of gaming disorder: A systematic review and meta-analysis. Australian and New Zealand Journal of Psychiatry, 55(6), 553–568. ；Vuorre M, Johannes N, Magnusson K, Przybylski AK. (2022). Time spent playing video games is unlikely to impact well-being. Royal Society Open Science, 9(7), 220411. ；Cục Quản lý Báo chín và Xuất bản (国家新闻出版署) (2021). Thông báo về việc quản lý chặt hơn nữa, ngăn người chưa thành niên nghiện game trực tuyến（国新出发〔2021〕14 号）. +- Ghi chú: Để mức B là vì tỷ lệ mắc thay đổi quá nhiều theo bảng hỏi, và nghiên cứu về hạnh phúc chỉ theo dõi sáu tuần. Cái giá phổ biến nhất của game là chen mất giấc ngủ và thời gian ở ngoài trời. Yêu cầu rõ ràng về giấc ngủ xem mục 10 của phần này（giấc ngủ, bài tập, thể dục）, thời gian ở ngoài trời xem mục 4 của phần này（2 giờ ở ngoài trời mỗi ngày）, thời lượng dùng màn hình xem mục 5 của phần này（không phải để học thì mỗi ngày không quá 1 giờ）. Nạp tiền và hoàn tiền xem mục 9 ở phần 5（trẻ nạp tiền và tặng quà trên điện thoại）. Kết quả học tập, giấc ngủ và quan hệ với người nhà cùng sụp xuống mà khuyên cũng không ngừng được, lúc đó đưa con đi khoa tâm thần hoặc khoa tâm lý trẻ em. Người hưởng lợi là con. +### 15. Trẻ nói mình thích người cùng giới, đừng mắng, đừng đuổi khỏi nhà, đừng đưa đi「cải tận」: thái độ của gia đình có liên quan đến chuyện trẻ có tự sát hay không + +- Chi phí: Không tốn tiền. Chỗ khó là bố mẹ phải xử lý cảm xúc của mình trước, và lúc đó không nói lời độc ác. +- Nói thẳng: Trong học sinh trung học trong nước, trẻ thích người cùng giới hoặc thích cả hai giới có xác suất thử tự sát trong một năm khoảng gấp 3 đến 5.5 lần so với bạn cùng tuổi. Nghiên cứu ở nước ngoài thấy trẻ tuổi thanh niên bị gia đình phản đối dữ dội thì sau này nhiều người đã thử tự sát hơn hẳn. Trẻ bị bố mẹ đưa đi「cải tận」 thì tỷ lệ thử tự sát cao nhất. +- Lợi ích: Một khảo sát trong nước phủ 506 trường trung học ở 7 tỉnh, gồm 150822 học sinh từ năm nhất đến năm ba. Trong vòng một năm qua, ở nhóm nam chỉ thích giới tính khác là 2.2%, chỉ thích cùng giới là 6.9%, thích cả hai là 12.2%. Ở nữ tương ứng là 3.1%, 8.9%, 10.9%. Tính ra khoảng gấp 3 đến 5.5 lần so với bạn học chỉ thích giới tính khác. Người nghiên cứu đã trừ đi tuổi, áp lực học tập, hoàn cảnh kinh tế gia đình và bị bắt nạt. Sau đó với hai nhóm nam đó, OR là 3.13 (95% CI 2.28–4.28) và 3.83 (2.85–5.14), cao khoảng hai đến ba lần. Một phân tích tổng hợp gom 35 nghiên cứu tính ra OR của thanh niên thiểu số từng thử tự sát là 3.50 (2.98–4.12). Một nghiên cứu ở Mỹ phỏng vấn 224 người đồng tính và song tính 21 đến 25 tuổi, nhờ họ nhớ lại phản ứng của gia đình hồi tuổi thanh niên. So với bạn cùng tuổi mà gia đình hầu như không phản đối, nhóm bị gia đình phản đối nặng có OR đã thử tự sát là 8.4. Trong một nghiên cứu khác, với trường hợp cha mẹ tự cố thay đổi xu hướng tính dục của con, OR đã thử tự sát là 3.08 (1.39–6.83). Với trường hợp cha mẹ còn đưa con đi gặp nhà trị liệu hoặc người tôn giáo, là 5.07 (2.38–10.79) (Trung Quốc / nhiều nước / Mỹ) +- Mức bằng chứng: B +- Nguồn: Huang Y, Li P, Guo L, et al. (2018). Sexual minority status and suicidal behaviour among Chinese adolescents: a nationally representative cross-sectional study. BMJ Open, 8(8), e020969. ；di Giacomo E, Krausz M, Colmegna F, Aspesi F, Clerici M. (2018). Estimating the Risk of Attempted Suicide Among Sexual Minority Youths: A Systematic Review and Meta-analysis. JAMA Pediatrics, 172(12), 1145–1152. ；Ryan C, Huebner D, Diaz RM, Sanchez J. (2009). Family rejection as a predictor of negative health outcomes in white and Latino lesbian, gay, and bisexual young adults. Pediatrics, 123(1), 346–352. ；Ryan C, Toomey RB, Diaz RM, Russell ST. (2020). Parent-Initiated Sexual Orientation Change Efforts With LGBT Adolescents: Implications for Young Adult Mental Health and Adjustment. Journal of Homosexuality, 67(2), 159–173. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (sửa đổi năm 2024). Luật bảo vệ người chưa thành niên（未成年人保护法）（第十七条）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2015). Luật chống bạo hành gia đình（反家庭暴力法）（第二条）. +- Ghi chú: Cho mức B vì đây đều là khảo sát một lần, mức độ phản đối của gia đình phải nhớ lại khi đã trưởng thành, nên chỉ cho thấy có liên quan chứ không cho biết ai gây ra ai. Bố mẹ không cần phải đồng ý ngay, chỉ cần đừng làm việc làm tổn hại con. Về mặt pháp luật, cha mẹ không được ngược đãi, bỏ rơi con, cũng không được áp dụng bạo hành gia đình với con. Chửi bới thường xuyên, đe dọa cũng tính là bạo hành gia đình. Trẻ tâm trạng u buồn, nói không muốn sống thì gọi 12356 trước, xem mục 25 ở phần 1 (khi có ý nghĩ tự sát thì gọi 12356). Cũng có thể đi khoa Tâm thần theo mục 8 của phần này (làm một lần sàng lọc trầm cảm cho trẻ 12 đến 18 tuổi). Vì sao đừng đụng vào「cải tận」, xem mục 28 ở phần 6 (không làm cải tận xu hướng tính dục). Người hưởng lợi là đứa trẻ, và cả bạn. diff --git "a/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" new file mode 100644 index 00000000..f9658fde --- /dev/null +++ "b/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -0,0 +1,193 @@ +[← Về mục lục](../README.md) + +# 33. Sống thế nào sau khi bị khuyết tật + +Phần này viết về những ngày sau khi đã bị khuyết tật. Cách phòng ngừa gây khuyết tật nằm ở phần 1 và phần 13, ở đây không lặp lại. + +Bốn kiểu tính riêng nhau, không quy đổi cho nhau. Các mục về cấp chứng và biến chứng tính theo tỷ lệ tử vong. Các mục về trợ cấp, quyền lợi và kiện tụng tính theo tiền. Các mục về đi học và làm thủ tục tính theo thời gian. Các mục về nhập viện điều trị rối loạn tâm thần, năng lực hành vi và giám hộ tính theo tự do cá nhân. + +Tờ hướng dẫn của Tổ chức Y tế Thế giới đưa ra ba con số. Trên toàn cầu có khoảng 1,3 tỷ người bị suy giảm chức năng đáng kể, chiếm 16% dân số thế giới, trung bình cứ sáu người thì có một. Một số người khuyết tật chết sớm hơn người không khuyết tật tới 20 năm. Nguy cơ người khuyết tật mắc trầm cảm, hen, đái tháo đường, đột quỵ, béo phì hoặc sức khỏe miệng kém gấp hai lần. Trong những chênh lệch này, phần lớn không do bản thân khuyết tật gây ra, mà do không được khám bệnh, không được đến trường, không tìm được việc, không ra khỏi cửa. Nên các mục trong phần này một nửa là thân thể, một nửa là thủ tục và quyền lợi. + +Năm việc đã viết ở chỗ khác, ở đây chỉ dẫn đường, không lặp lại. Cách làm giấy chứng nhận người khuyết tật, bảy nhóm và cách chấm từ cấp 1 đến cấp 4, xem mục 11 ở phần 24 (giấy chứng nhận người khuyết tật). Thời điểm nào được làm giám định thương tật, xem mục 10 ở phần 24 (giám định thương tật). Cách nhận hai khoản trợ cấp cho người khuyết tật, xem mục 8 ở phần 7 (hai khoản trợ cấp). Khảo sát năng lực lao động của tai nạn lao động và thương tật đổi ra bao nhiêu tiền, xem mục 15 ở phần 19 (khảo sát năng lực lao động). Cách nộp đơn xin bảo hiểm chăm sóc dài hạn cho người mất năng lực nặng, xem mục 7 ở phần 17 (bảo hiểm chăm sóc dài hạn). Mục đó nằm ở phần 17 (Khi trong nhà có người già), nhưng khoản quyền lợi này không có giới hạn độ tuổi, người trẻ vẫn nhận được. + +Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia đình bạn, thuộc hai mức cao nhất. Chỗ nào liên quan đến người khác thì ghi riêng trong Ghi chú. + +### 1. Sau chấn thương tủy sống mà đột nhiên đau đầu dữ dội và ra mồ hôi: đỡ ngồi ngay lên, nới lỏng quần áo, đồng thời gọi 120 + +- Chi phí: Không tốn tiền. Ba động tác, vài chục giây là xong. +- Nói thẳng: Ở người bị chấn thương tủy sống, đường báo động trong cơ thể đứt mất một đoạn. Bàng quang đang bí, quần áo đang chật, bản thân có thể không cảm nhận được, nhưng huyết áp lại tăng vọt. Dấu hiệu là đột nhiên đau đầu dữ dội và ra mồ hôi. Lúc đó đỡ người ấy ngồi ngay lên, nới lỏng quần áo đang bó, xem có phải đang bí tiểu không, đồng thời gọi 120. +- Lợi ích: Cơn này gọi là phản xạ thần kinh tự động bất thường. Một bài tổng hệ thống đã tìm bốn cơ sở dữ liệu Medline, CINAHL, EMBASE và PsycINFO, đưa vào 31 nghiên cứu, trong đó 6 là thử nghiệm đối chứng ngẫu nhiên. Bài tổng kết nêu rõ, xử lý không dùng thuốc ngay lúc cơn gồm ba bước: đặt bệnh nhân ở tư thế ngồi dựng lên, nới lỏng quần áo bó sát, loại bỏ kích thích gây cơn. Ba bước này dựa trên đồng thuận lâm sàng và số liệu sinh lý, thuộc bằng chứng cấp độ 5. Chỉ khi huyết áp tăng cao kéo dài mới dùng thuốc hạ áp, trong đó prazosin có bằng chứng cấp độ 1, nifedipin và prostaglandin E2 có bằng chứng cấp độ 2. Bài tổng kết cũng nêu rõ, chiến lược phòng cơn chủ yếu nhắm hai nhóm nguyên nhân thường gặp là hệ tiết niệu sinh dục và phẫu thuật, phần lớn chỉ có bằng chứng cấp độ 4 và cấp độ 5 chống đỡ +- Mức bằng chứng: B +- Nguồn: Krassioukov A, Warburton DE, Teasell R, Eng JJ; Spinal Cord Injury Rehabilitation Evidence Research Team (2009). A systematic review of the management of autonomic dysreflexia after spinal cord injury. Archives of Physical Medicine and Rehabilitation, 90(4):682-695. +- Ghi chú: Chọn B là vì ba động tác tại chỗ đó không có thử nghiệm đối chứng ngẫu nhiên phía sau, chỉ có đồng thuận và suy luận sinh lý. Bản thân bài tổng kết cũng nói, phần lớn các cách này chỉ có nghiên cứu không đối chứng. Phần dùng thuốc là việc của bác sĩ, ở nhà không tự cho người ta thuốc hạ áp. Để người nằm thẳng xuống thì huyết áp còn cao hơn, nên phải đỡ ngồi dậy. Người nhà và người chăm cũng phải biết ba bước này. Đừng coi đó là đau đầu thường rồi bỏ qua. + +### 2. Mười năm đầu sau khi bị khuyết tật là giai đoạn nguy cơ tự tử cao, đừng tưởng chỉ cần nghĩ thoáng là xong + +- Chi phí: Không tốn tiền. Gọi một cuộc điện thoại, hoặc đi đăng ký một lần khám. +- Nói thẳng: Vài năm đầu sau khi bị khuyết tật, bản thân hoặc người nhà sụp đổ cảm xúc là chuyện thường gặp, không phải vấn đề ý chí. Theo dõi lâu dài tại hai trung tâm chấn thương tủy sống ở Anh cho thấy, nhóm người này có tỷ lệ tự tử gấp năm lần người bình thường. Hơn chín phần mười số ca tự tử xảy ra trong mười năm đầu sau chấn thương. Người mà nguyên nhân chấn thương chính là tự tử, nguy cơ còn cao hơn một đoạn. +- Lợi ích: Nghiên cứu theo dõi 2304 bệnh nhân chấn thương tủy sống do chấn thương, nhập viện mới từ năm 1991 đến năm 2010, sống qua năm đầu sau chấn thương, lúc xuất viện còn khiếm khuyết thần kinh, theo dõi đến cuối năm 2014. Trong đó 63 ca (2.7%) nguyên nhân chấn thương chính là tự tử chưa thành. Đến cuối thời gian theo dõi tổng cộng 533 người chết, trong đó 4.2% chết do tự tử, 91% số ca tự tử xảy ra trong mười năm đầu sau chấn thương. Tỷ lệ tử vong do tự tử chuẩn hóa theo tuổi là 62.5 ca mỗi năm trên 100.000 người (95% CI 36.4–88.6, khoảng tin cậy), gấp năm lần tỷ lệ tự tử của dân số chung nước Anh và Xứ Wales năm 2014 (12.2 ca trên 100.000 người). Nhóm bị thương do tự tử chưa thành, so với các nguyên nhân thương tật khác, odds ratio tử vong là 4.32 (P<0.001), odds ratio tự tử là 9.46 (P<0.001) +- Mức bằng chứng: A +- Nguồn: Savic G, DeVivo MJ, Frankel HL, Jamous MA, Soni BM, Charlifue S (2018). Suicide and traumatic spinal cord injury - a cohort study. Spinal Cord, 56(1):2-6. +- Ghi chú: Nghiên cứu này chỉ xem chấn thương tủy sống, áp sang mù lòa, cắt cụt, bỏng và các nguyên nhân khuyết tật khác thì là suy diễn lan. Nó cho biết được khung thời gian: nguy cơ không hết sau khi ra viện, phải theo dõi mười năm. Cần người bên cạnh nói chuyện thì gọi 12356, xem mục 11 ở phần 29 (12356). Đừng tích thuốc ngủ và thuốc trừ sâu trong nhà, xem mục 25 ở phần 1 (ý nghĩ tự tử). Đau buồn và cảm xúc mắc kẹt sau nửa năm vẫn thế thì đi đăng ký khoa tâm thần hoặc tâm lý lâm sàng, xem mục 8 ở phần 29 (sau nửa năm vẫn còn đau buồn). + +### 3. Nhập viện điều trị rối loạn tâm thần theo nguyên tắc tự nguyện, chỉ có hai trường hợp được làm mà không cần sự đồng ý của bản thân + +- Chi phí: Không tốn tiền. Ghi nhớ hai trường hợp, đến lúc cần là nói được ngay. +- Nói thẳng: Một câu của người nhà hoặc cơ quan không giữ được ai đó ở lại bệnh viện tâm thần. Pháp luật chỉ công nhận hai trường hợp: đã tự gây hại cho mình hoặc có nguy cơ đó, đã gây nguy hại cho người khác hoặc có nguy cơ đó. Trường hợp thứ nhất cần người giám hộ đồng ý mới được tiếp nhận. Không đồng ý với chẩn đoán thì trong ba ngày có thể yêu cầu chẩn đoán lại, sau đó còn tự uỷ quyền cho bên giám định. +- Lợi ích: Luật Sức khỏe Tâm thần (精神卫生法) Điều ba mươi nêu rõ「精神障碍的住院治疗实行自愿原则」— việc điều trị nội trú rối loạn tâm thần theo nguyên tắc tự nguyện. Chỉ khi chẩn đoán và đánh giá tình trạng bệnh xác định người khám là bệnh nhân rối loạn tâm thần nặng, đồng thời thuộc một trong các trường hợp sau, mới được điều trị nội trú. Thứ nhất「已经发生伤害自身的行为，或者有伤害自身的危险的」— đã có hành vi tự gây hại cho bản thân, hoặc có nguy cơ tự hại. Thứ hai「已经发生危害他人安全的行为，或者有危害他人安全的危险的」— đã có hành vi gây nguy hại cho an toàn của người khác, hoặc có nguy cơ đó. Điều ba mươi mốt: thuộc trường hợp thứ nhất thì chỉ khi người giám hộ đồng ý, cơ sở y tế mới được điều trị nội trú; người giám hộ không đồng ý thì không được làm. Điều ba mươi hai: thuộc trường hợp thứ hai mà bản thân hoặc người giám hộ không đồng ý, có thể yêu cầu chẩn đoán lại, phải nộp trong ba ngày kể từ ngày nhận được kết luận chẩn đoán, do hai bác sĩ hành nghề tâm thần ngoài bác sĩ chẩn đoán lần đầu thực hiện; vẫn không đồng ý với kết luận chẩn đoán lại thì có thể tự uỷ quyền cho tổ chức giám định có tư cách hành nghề theo pháp luật để làm giám định y khoa về rối loạn tâm thần. Điều ba mươi lăm: kết luận chẩn đoán lại hoặc báo cáo giám định cho thấy không xác định được là bệnh nhân rối loạn tâm thần nặng, hoặc bệnh nhân không cần điều trị nội trú, thì cơ sở y tế không được tiến hành điều trị nội trú. Điều bảy mươi tám: cố ý đưa người không bị rối loạn tâm thần vào cơ sở y tế với danh nghĩa rối loạn tâm thần để điều trị, thì phải bồi thường theo pháp luật (toàn quốc, có hiệu lực từ ngày 1 tháng 5 năm 2013, sửa đổi năm 2018) +- Mức bằng chứng: A +- Nguồn: Luật Sức khỏe Tâm thần của Cộng hoà Nhân dân Trung Quốc (中华人民共和国精神卫生法)（2018 年修正，第三十条、第三十一条、第三十二条、第三十五条、第七十八条）. +- Ghi chú: Ngược lại, cũng không phải ai cũng bị cấm đưa đi khám. Điều 28 quy định, người nghi bị rối loạn tâm thần đã có hành vi tự gây hại cho bản thân hoặc gây nguy hại cho an toàn người khác, hoặc có nguy cơ đó, thì người thân thích, cơ quan công tác, công an địa phương phải ngay lập tức áp dụng biện pháp ngăn chặn, và đưa người đó đến cơ sở y tế để chẩn đoán rối loạn tâm thần. Cơ sở y tế khi tiếp nhận người đưa đến thì không được từ chối chẩn đoán. Nên「đưa đi chẩn đoán」và「giữ lại nhập viện」là hai cánh cửa, cánh trước thấp, cánh sau cao. Điều 75 liệt kê trách nhiệm của cơ sở y tế: thực hiện cách ly cưỡng chế trái quy định, chẩn đoán người không bị rối loạn tâm thần thành rối loạn tâm thần, thì bị xử lý cho đến khi bị thu hồi chứng chỉ hành nghề. Người bên cạnh có xung lực gây hại thì làm gì, xem phần 8. + +### 4. Người chăm lâu dài người khuyết tật trong nhà, sức khỏe của chính mình cũng phải có người nhìn + +- Chi phí: Không tốn tiền. Giao bớt phần việc chăm sóc đi một phần, tự đi khám sức khỏe một lần. +- Nói thẳng: Người chăm sóc kiệt sức không phải là làm quá. Một nghiên cứu ở Mỹ theo dõi hơn tám trăm người già trong bốn năm. Nhóm vừa chăm phối ngẫu mất năng lực vừa nói mình mệt mỏi tinh thần, nguy cơ tử vong cao hơn người không chăm khoảng sáu phần mười. Nhóm chăm nhưng không thấy mệt thì không có sự tăng này. Nên thứ phải theo dõi không phải việc chăm, mà là sự căng thẳng kéo dài đó. +- Lợi ích: Nghiên cứu là cohort dân số theo dõi trước, từ năm 1993 đến năm 1998, theo dõi trung bình 4.5 năm, làm ở bốn cộng đồng tại Hoa Kỳ. 392 người chăm sóc và 427 người không chăm sóc, tuổi từ 66 đến 96, đều sống cùng phối ngẫu. Trong bốn năm có 103 người (12.6%) chết. Sau khi điều chỉnh yếu tố xã hội nhân khẩu, bệnh đã có và bệnh tim mạch cận lâm sàng, người vừa chăm vừa nói mình bị căng thẳng tinh thần hoặc cảm xúc có nguy cơ tử vong cao hơn đối chứng không chăm 63% (RR 1.63, 95% CI 1.00–2.65, khoảng tin cậy). Nhóm chăm nhưng không báo cáo căng thẳng (RR 1.08, 95% CI 0.61–1.90) và nhóm phối ngẫu mất năng lực nhưng bản thân không chăm (RR 1.37, 95% CI 0.73–2.58), tỷ lệ tử vong sau điều chỉnh đều không tăng +- Mức bằng chứng: A +- Nguồn: Schulz R, Beach SR (1999). Caregiving as a risk factor for mortality: the Caregiver Health Effects Study. JAMA, 282(23):2215-2219. +- Ghi chú:Tranh cãi: Giới dưới của khoảng tin cậy của nghiên cứu này đúng 1.00, khả năng「không có khác biệt» chưa bị loại hẳn. Đối tượng là người già Mỹ sống cùng phối ngẫu, người được chăm là phối ngẫu già mất năng lực, áp sang người chăm thân nhân khuyết tật trẻ tuổi thì là suy diễn lan. Nó cho biết được hướng: nhóm mệt thì nguy cơ cao hơn. Công việc chăm giao được thì giao bớt, dịch vụ nuôi dưỡng và chăm sóc ban ngày hỏi hội người khuyết tật địa phương và phòng dân vụ. Người chăm sóc đừng tự ngừng thuốc cho bệnh mạn của mình, xem mục 1 ở phần 16 (uống thuốc đủ theo chỉ định). Người được lợi của mục này là bạn. + +### 5. Người ngồi xe lăn lâu thì đổi đệm ngồi sang loại giảm áp, mỗi ngày xem một lượt xương ngồi và xương cùng + +- Chi phí: Một chiếc đệm giảm áp giá vài trăm tệ đến hơn một nghìn tệ. Mỗi ngày xem một lượt, một phút. +- Nói thẳng: Đệm bọt thường ngồi lâu sẽ làm căng da, vỡ ra thì lành rất lâu, dễ nhiễm trùng. Người già ở viện dưỡng lão dùng đệm giảm áp thì loét ở xương ngồi ít hơn rõ rệt, nhưng nếu tính cả xương cùng thì không thấy khác biệt. Xe lăn cũng phải theo kích cỡ của người, đừng mua một kiểu đại trà rồi thôi. +- Lợi ích: Thử nghiệm lâm sàng ngẫu nhiên này làm ở 12 viện dưỡng lão, từ năm 2004 đến năm 2008 tuyển 232 cư dân trên 65 tuổi, mỗi ngày ngồi xe lăn hơn 6 giờ, điểm Braden dưới 18. Tất cả đều được lắp xe lăn vừa vặn trước, rồi chia ngẫu nhiên thành hai nhóm. Một nhóm dùng đệm bảo vệ da (bọt khí, chất lỏng dính cộng bọt, hoặc gel cộng bọt), một nhóm dùng đệm bọt cắt khúc dày 7.6 cm. Theo dõi 6 tháng hoặc cho đến khi xuất hiện loét. Loét ở vùng mỏm ngồi, nhóm bọt cắt khúc 8 ca (6.7%), nhóm đệm bảo vệ da 1 ca (0.9%), P=0.04. Tính gộp loét ở xương ngồi và vùng cùng cụt, nhóm bọt cắt khúc 21 ca (17.6%), nhóm đệm bảo vệ da 12 ca (10.6%), P=0.14, không ra khác biệt +- Mức bằng chứng: B +- Nguồn: Brienza D, Kelsey S, Karg P, et al. (2010). A randomized clinical trial on preventing pressure ulcers with wheelchair seat cushions. Journal of the American Geriatrics Society, 58(12):2308-2314. +- Ghi chú: Chọn B có hai lý do. Thứ nhất mẫu chỉ có 232 người, P của điểm chính 0.04 vừa vượt qua, điểm phụ không ra khác biệt. Thứ hai đối tượng là cư dân cao tuổi ở viện dưỡng lão, áp sang người dùng xe lăn trẻ tuổi thì là suy diễn lan. Trong thử nghiệm tất cả đều được lắp xe lăn vừa vặn trước, nên「vừa vặn」và「đệm」cùng có tác dụng, đừng chỉ đổi đệm. Phòng loét cho người nằm lâu: đệm khí và lật người đúng giờ, xem mục 8 ở phần 17 (loét). Cách nộp đơn xin trợ cấp lắp thiết bị hỗ trợ kiểu cơ bản, xem mục 7 của phần này (hỏi hết ở hội người khuyết tật cấp huyện). + +### 6. Đừng mua các liệu pháp và thiết bị「chữa khỏi bại liệt, mù lòa, điếc」 + +- Chi phí: Không tốn tiền, lại còn tiết kiệm được tiền. Khó là người nhà còn ôm hy vọng, khuyên không nổi. +- Nói thẳng: Luật Quảng cáo đã nói chết. Quảng cáo y tế, thuốc, thiết bị y tế không được đảm bảo hiệu quả, không được nói tỷ lệ chữa khỏi hay tỷ lệ hiệu quả, không được tìm người ra kể. Nên「ba đợt điều trị đứng dậy」「bao nhiêu người đã khỏi」thì bản thân nó đã là quảng cáo trái phép, không cần nghiên cứu lý do nó nêu. Thứ không phải thiết bị y tế thì ngay cả nhắc đến chữa bệnh cũng bị cấm. +- Lợi ích: Luật Quảng cáo Điều mười sáu quy định, quảng cáo y tế, thuốc, thiết bị y tế không được chứa nội dung sau:「（一）表示功效、安全性的断言或者保证；（二）说明治愈率或者有效率；（三）与其他药品、医疗器械的功效和安全性或者其他医疗机构比较；（四）利用广告代言人作推荐、证明」— khẳng định hay bảo đảm công dụng, an toàn; nêu tỷ lệ chữa khỏi hay tỷ lệ hiệu quả; so sánh với thuốc, thiết bị y tế khác hay cơ sở y tế khác; dùng người phát ngôn quảng cáo để giới thiệu, chứng minh. Điều mười bảy quy định, ngoài quảng cáo y tế, thuốc, thiết bị y tế,「禁止其他任何广告涉及疾病治疗功能，并不得使用医疗用语或者易使推销的商品与药品、医疗器械相混淆的用语」— cấm mọi quảng cáo khác đề cập đến chức năng chữa bệnh, và không được dùng ngôn ngữ y khoa hay ngôn ngữ dễ làm hàng hóa đang chào bán bị nhầm với thuốc, thiết bị y tế. Điều bốn quy định, quảng cáo không được chứa nội dung giả dối hoặc gây hiểu lầm, không được lừa dối, gây hiểu lầm cho người tiêu dùng, người quảng cáo phải chịu trách nhiệm về tính chân thực của nội dung quảng cáo (toàn quốc, sửa đổi năm 2021) +- Mức bằng chứng: A +- Nguồn: Luật Quảng cáo của Cộng hoà Nhân dân Trung Quốc (中华人民共和国广告法)（2021 年修正，第四条、第十六条、第十七条）. +- Ghi chú: Cách phán đoán chỉ một: xem nó có dám nói hết hay không. Khoa phục hồi chức năng của bệnh viện chính quy không hứa với bạn mấy đợt điều trị là đứng dậy được. Gặp loại quảng cáo này thì báo 12315 hoặc cơ quan quản lý thị trường địa phương. Bộ thoại quảng cáo bảo phẩm cũng cùng một lối, xem mục 10 ở phần 6 (bảo phẩm). Đã mua rồi muốn đòi lại tiền, làm theo cách mua hàng qua mạng và tiêu dùng trả trước, xem mục 29 ở phần 5 (mua qua mạng theo quy tắc của nền tảng). Người được lợi của mục này là bạn và gia đình. + +### 7. Sau khi giấy chứng nhận người khuyết tật đã được cấp, đến hội người khuyết tật cấp huyện hỏi hết những khoản có thể xin trong một lần + +- Chi phí: Không tốn tiền. Đi một chuyến tới hội người khuyết tật cấp huyện, hỏi cả buổi sáng. Sau đó còn phải đi nhiều cơ quan nữa, không chỉ một. +- Nói thẳng: Bản thân giấy chứng nhận người khuyết tật không phát tiền, nó là một chìa khóa. Phía sau nó là mấy khoản quyền lợi tách bạch với nhau, mỗi khoản đều phải tự nộp đơn, không khoản nào tự đến. Lần đầu đến hội người khuyết tật, hãy hỏi lần lượt từng mục trong danh sách, sẽ đỡ phải chạy đi chạy lại mấy chuyến. +- Lợi ích: Giấy chứng nhận mở ra ít nhất sáu khoản quyền lợi, mỗi khoản có căn cứ riêng. Một là trợ cấp sinh hoạt cho người khuyết tật khó khăn và trợ cấp chăm sóc cho người khuyết tật nặng, xem mục 8 ở phần 7 (hai khoản trợ cấp). Hai là trợ giúp phục hồi chức năng cho trẻ khuyết tật, xem mục 8 của phần này (trợ giúp phục hồi chức năng). Ba là trợ cấp lắp thiết bị hỗ trợ kiểu cơ bản, căn cứ là Điều hai mươi sáu của Quy định về Phòng ngừa khuyết tật và Phục hồi chức năng cho người khuyết tật (《残疾预防和残疾人康复条例》). Bốn là trợ cấp cải tạo tiện nghi không vật cản trong gia đình, xem mục 9 của phần này (làm lối đi, lan can và cải tạo phòng tắm). Năm là tuyển dụng theo tỷ lệ và dịch vụ việc làm, xem mục 10 của phần này (chủ động nói ra mình có giấy). Sáu là giảm thuế thu nhập cá nhân, xem mục 11 của phần này (giảm thuế cá nhân). Điều hai mươi sáu của Quy định này còn viết rằng nhà nước thông qua thực hiện các dự án phục hồi chức năng trọng điểm để cung cấp dịch vụ phục hồi chức năng cơ bản cho người khuyết tật nghèo ở đô thị và nông thôn, cho người khuyết tật nặng,「按照国家有关规定对基本型辅助器具配置给予补贴」— tức là trợ cấp cho việc lắp thiết bị hỗ trợ kiểu cơ bản theo quy định của nhà nước. Điều hai mươi quy định các cơ quan liên quan của chính quyền nhân dân cấp huyện trở lên và hội người khuyết tật phải tận dụng nguồn lực cộng đồng, lập địa điểm phục hồi chức năng hoặc mua dịch vụ, để làm công tác phục hồi chức năng tại cộng đồng như hướng dẫn phục hồi, huấn luyện khả năng sinh hoạt hằng ngày, chăm sóc phục hồi, lắp thiết bị hỗ trợ, tư vấn thông tin và chuyển giới thiệu (toàn quốc, có hiệu lực từ ngày 1 tháng 7 năm 2017) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2017). Quy định về Phòng ngừa khuyết tật và Phục hồi chức năng cho người khuyết tật (《残疾预防和残疾人康复条例》)（第二十条、第二十六条）. +- Ghi chú: Quy trình làm giấy chứng nhận người khuyết tật, xem mục 11 ở phần 24 (giấy chứng nhận người khuyết tật). Tiêu chuẩn trợ cấp và cách làm cụ thể phần lớn do cấp tỉnh quy định, cùng một khoản ở hai tỉnh có thể chênh nhau mấy lần. Khi hỏi, ghi lại tên từng khoản một, về nhà tra lại văn bản gốc theo tên trên trang chính thức của hội người khuyết tật địa phương và trên cổng thông tin điện tử của chính quyền, đừng chỉ nghe lời nói miệng bên quầy. Đừng trông chờ hỏi trọn trong một lần, cứ sau một năm hỏi lại, chính sách có thay đổi. + +### 8. Trẻ dưới 7 tuổi có khuyết tật hoặc tự kỷ, đến hội người khuyết tật cấp huyện xin trợ giúp phục hồi chức năng + +- Chi phí: Không tốn tiền. Nộp một đơn ở hội người khuyết tật cấp huyện, rồi chọn một cơ sở trong danh sách chỉ định. +- Nói thẳng: Nhà nước lập một chế độ trợ giúp phục hồi chức năng cho trẻ khuyết tật và trẻ tự kỷ từ 0 đến 6 tuổi, gồm phẫu thuật, thiết bị hỗ trợ và tập luyện phục hồi chức năng. Trước hết là gia đình khó khăn được hưởng, nơi tài chính địa phương cho phép thì có thể nới lỏng cả về tuổi lẫn điều kiện kinh tế của gia đình. Đây là giai đoạn phục hồi chức năng cho kết quả tốt nhất, đừng kéo đến lúc đáng đi học mới nói. Cơ sở phải chọn trong danh sách chỉ định của hội. +- Lợi ích: Ý kiến của Hội đồng Nhà nước về việc lập chế độ trợ giúp phục hồi chức năng cho trẻ khuyết tật (《国务院关于建立残疾儿童康复救助制度的意见》) nêu rằng đối tượng được trợ giúp là trẻ khuyết tật và trẻ tự kỷ từ 0—6 tuổi thỏa điều kiện, bị khuyết tật về thị giác, thính giác, ngôn ngữ, vận động, trí tuệ. Gồm trẻ khuyết tật thuộc gia đình được trợ cấp sinh hoạt tối thiểu ở đô thị và nông thôn, thuộc gia đình hộ nghèo được lập sổ, trẻ khuyết tật do cơ sở bảo trì trẻ em nuôi dưỡng, trẻ mồ côi khuyết tật, trẻ khuyết tật thuộc phạm vi được trợ cấp nuôi dưỡng người khó khăn khắc nghẹt nhất, cùng trẻ khuyết tật thuộc các gia đình khó khăn về tài chính khác. Cách xác định cụ thể gia đình khó khăn về tài chính khác do chính quyền địa phương cấp huyện trở lên quy định. Văn bản còn viết「有条件的地区，可扩大残疾儿童康复救助年龄范围，也可放宽对救助对象家庭经济条件的限制」— nơi có điều kiện thì có thể mở rộng phạm vi tuổi được trợ giúp, cũng có thể nới lỏng hạn chế về điều kiện kinh tế của gia đình người được hưởng. Tiêu chuẩn bảo đảm kinh phí do chính quyền địa phương cấp huyện trở lên căn cứ tình hình tài chính địa phương, số lượng đối tượng bảo đảm, loại khuyết tật để xếp loại xác định, và có cơ chế điều chỉnh theo thời. Căn cứ cấp cao hơn là Điều hai mươi sáu của Quy định về Phòng ngừa khuyết tật và Phục hồi chức năng cho người khuyết tật: nhà nước lập chế độ trợ giúp phục hồi chức năng cho trẻ khuyết tật,「逐步实现0—6岁视力、听力、言语、肢体、智力等残疾儿童和孤独症儿童免费得到手术、辅助器具配置和康复训练等服务」— tức từng bước để trẻ khuyết tật và trẻ tự kỷ từ 0—6 tuổi bị khuyết tật về thị giác, thính giác, ngôn ngữ, vận động, trí tuệ được miễn phí phẫu thuật, lắp thiết bị hỗ trợ và tập luyện phục hồi chức năng (toàn quốc, từ năm 2018) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2018). Ý kiến về việc lập chế độ trợ giúp phục hồi chức năng cho trẻ khuyết tật (《国务院关于建立残疾儿童康复救助制度的意见》)(国发〔2018〕20 号). ；Hội đồng Nhà nước (国务院) (2017). Quy định về Phòng ngừa khuyết tật và Phục hồi chức năng cho người khuyết tật (《残疾预防和残疾人康复条例》)（第二十六条）. +- Ghi chú: Tiêu chuẩn trợ giúp do từng tỉnh quy định, trước hết tra trên trang chính thức của hội người khuyết tật địa phương「残疾儿童康复救助实施办法」. Danh sách cơ sở chỉ định cũng nằm trong tay hội. Văn bản yêu cầu hội phối hợp với các cơ quan liên quan siết chặt quản lý việc cơ sở phục hồi chức năng chỉ định được vào và được ra, lập cơ chế kiểm tra định kỳ và đánh giá tổng hợp, và lập danh sách cấm. Cơ sở không có trong danh sách thì bạn phải tự trả toàn bộ, có chuyện gì cũng khó truy trách nhiệm. Người được lợi của mục này là con bạn, thuộc mức thứ hai. Cơ sở nào hứa「chắc chắn khỏi」thì xử lý theo mục 6 của phần này (chữa khỏi bại liệt). + +### 9. Lối đi, lan can và cải tạo phòng tắm trong nhà, có thể xin trợ cấp từ chính quyền cấp huyện trở lên + +- Chi phí: Nộp đơn không tốn tiền. Phần cải tạo phần lớn có trợ cấp, bạn có thể phải bỏ thêm một phần. Phải đi hội người khuyết tật và ủy ban phường. +- Nói thẳng: Luật Xây dựng môi trường không vật cản (《无障碍环境建设法》) có một điều: chính quyền cấp huyện trở lên phải hỗ trợ, hướng dẫn cải tạo tiện nghi không vật cản trong gia đình, và phải trợ cấp mức độ thích hợp cho gia đình có người khuyết tật, người già thỏa điều kiện. Loại dự án này thường do hội người khuyết tật phụ trách, làm theo hạn ngạch từng năm, nên phải đi đăng ký sớm. Trợ cấp bao nhiêu, sửa mấy hạng mục thì địa phương quy định. +- Lợi ích: Điều mười chín của Luật Xây dựng môi trường không vật cản:「县级以上人民政府应当支持、指导家庭无障碍设施改造。对符合条件的残疾人、老年人家庭应当给予适当补贴。」— tức chính quyền nhân dân cấp huyện trở lên phải hỗ trợ, hướng dẫn cải tạo tiện nghi không vật cản trong gia đình, và phải trợ cấp mức độ thích hợp cho gia đình có người khuyết tật, người già thỏa điều kiện. Điều mười tám quy định, đối với công trình ở, khu ở, công trình công cộng, nơi công cộng, phương tiện vận tải, đường giao thông đô thị và nông thôn đã có sẵn mà không đạt tiêu chuẩn kỹ thuật xây dựng công trình tiện nghi không vật cản, chính quyền nhân dân cấp huyện trở lên phải căn cứ tình hình thực tế để lập kế hoạch cải tạo tiện nghi có trọng điểm và tổ chức thực hiện. Điều sáu mươi hai quy định, mọi tổ chức và cá nhân có quyền đưa ý kiến, kiến nghị với các cơ quan chủ trì liên quan của chính quyền về việc tăng cường và cải thiện xây dựng môi trường không vật cản, và khiếu nại, báo việc về hành vi vi phạm luật này; cơ quan chủ trì nhận được khiếu nại, báo việc「应当及时处理并予以答复」— phải xử lý kịp thời và trả lời. Điều sáu mươi ba quy định, đối với hành vi vi phạm luật này gây thiệt hại cho lợi ích công cộng, Viện kiểm sát nhân dân có thể đưa ra kiến nghị của Viện kiểm sát hoặc khởi kiện vì lợi ích công cộng (toàn quốc, có hiệu lực từ ngày 1 tháng 9 năm 2023) +- Mức bằng chứng: B +- Nguồn: Luật Xây dựng môi trường không vật cản của Cộng hoà Nhân dân Trung Quốc (中华人民共和国无障碍环境建设法)（第十八条、第十九条、第六十二条、第六十三条）. +- Ghi chú: Để ở mức B là vì luật chỉ viết「应当给予适当补贴」, trợ cấp bao nhiêu, ai thỏa điều kiện thì giao hết cho địa phương, không có con số thống nhất cả nước. Trước hết tra trên trang chính thức của hội người khuyết tật địa phương「困难重度残疾人家庭无障碍改造」. Hạn ngạch phần lớn chia theo năm, phải xếp hàng sớm. Lắp thang máy trong khu chung cư, cải tạo hành lang là một quy trình khác, đi theo cơ quan nhà ở và đô thị. Tiện nghi không vật cản ở nơi công cộng không đạt thì khiếu nại với cơ quan chủ trì theo Điều sáu mươi hai, cơ quan đó phải trả lời bạn. + +### 10. Khi tìm việc, chủ động nói ra mình có giấy, doanh nghiệp tuyển bạn thì được trừ một khoản tiền + +- Chi phí: Không tốn tiền. Thêm một câu trong buổi phỏng vấn. Khó là mở lời. +- Nói thẳng: Tỷ lệ người khuyết tật mà đơn vị sắp xếp không được thấp hơn 1.5% tổng số công nhân viên đang làm việc, tỷ lệ cụ thể do tỉnh quy định. Không đạt thì phải nộp quỹ bảo đảm việc làm cho người khuyết tật theo số người thiếu, một người tính theo mức lương bình quân năm của công nhân viên năm trước của đơn vị. Nên tuyển bạn không chỉ là nhường người, mà là tiết kiệm tiền thật. Nói rõ chuyện bạn có giấy trong buổi phỏng vấn, đối phương tự tính được món này có lời hay không. +- Lợi ích: Điều tám của Quy định về Việc làm cho người khuyết tật (《残疾人就业条例》):「用人单位安排残疾人就业的比例不得低于本单位在职职工总数的1.5%。具体比例由省、自治区、直辖市人民政府根据本地区的实际情况规定。」— tức tỷ lệ người khuyết tật mà đơn vị sắp xếp không được thấp hơn 1.5% tổng số công nhân viên đang làm việc của chính đơn vị; tỷ lệ cụ thể do chính quyền nhân dân tỉnh, khu tự trị, thành phố trực thuộc trung ương quy định theo tình hình thực tế địa phương. Điều chín: đơn vị không đạt tỷ lệ do chính quyền nhân dân cấp tỉnh nơi đơn vị đặt trụ sở quy định thì phải nộp quỹ bảo đảm việc làm cho người khuyết tật. Điều tám của Biện pháp thu, chi quỹ bảo đảm việc làm cho người khuyết tật (《残疾人就业保障金征收使用管理办法》) đưa ra công thức:「保障金年缴纳额=（上年用人单位在职职工人数×所在地省、自治区、直辖市人民政府规定的安排残疾人就业比例-上年用人单位实际安排的残疾人就业人数）×上年用人单位在职职工年平均工资」— tức số tiền phải nộp trong năm bằng (số công nhân viên đang làm việc năm trước × tỷ lệ việc làm cho người khuyết tật do chính quyền tỉnh, khu tự trị, thành phố trực thuộc trung ương nơi đó quy định − số người khuyết tật đơn vị thực sự sắp xếp năm trước) × lương bình quân năm của công nhân viên đang làm việc năm trước. Điều mười sáu: kể từ ngày đăng ký kinh doanh, trong vòng 3 năm, doanh nghiệp vừa và nhỏ sắp xếp việc làm cho người khuyết tật chưa đạt tỷ lệ quy định, có tổng số công nhân viên đang làm việc từ 20 người trở xuống (gồm cả 20 người) thì được miễn quỹ. Điều ba mươi sáu của Luật Bảo vệ Người khuyết tật (《残疾人保障法》) quy định, nhà nước cho các đơn vị sắp xếp việc làm cho người khuyết tật đạt hoặc vượt tỷ lệ quy định, hoặc sắp xếp tập trung, được hưởng ưu đãi thuế theo pháp luật (toàn quốc, Quy định về Việc làm cho người khuyết tật có hiệu lực từ ngày 1 tháng 5 năm 2007) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2007). Quy định về Việc làm cho người khuyết tật (《残疾人就业条例》)(第八条、第九条). ；Bộ Tài chính, Cục Tổng cục Thuế nhà nước, Hiệp hội Người khuyết tật Trung Quốc (中国残联) (2015). Biện pháp thu, chi quỹ bảo đảm việc làm cho người khuyết tật (《残疾人就业保障金征收使用管理办法》)(财税〔2015〕72 号，第六条、第八条、第十六条). ；Luật Bảo vệ Người khuyết tật của Cộng hoà Nhân dân Trung Quốc (中华人民共和国残疾人保障法)（第三十六条）. +- Ghi chú: Với doanh nghiệp vừa và nhỏ mới đăng ký dưới ba năm, dưới 20 người, lý do này không thành lập, đừng trông chờ. Chính sách giảm, miễn quỹ bảo đảm mấy năm nay cứ điều chỉnh, cụ thể theo thông báo của Bộ Tài chính trong năm. Ngoài ra, yêu cầu việc làm theo tỷ lệ là việc làm thật. Đứng tên hưởng lương, để đơn vị lấy bạn để khấu trừ quỹ bảo đảm, là hành vi lừa đảo, bạn cũng bị kéo vào. Xác định là người khó khăn trong việc làm và trợ cấp bảo hiểm xã hội, xem mục 12 ở phần 7 (người khó khăn trong việc làm). Trợ cấp đào tạo nghề, xem mục 13 ở phần 7 (trợ cấp đào tạo nghề). Người được lợi của mục này là bạn. + +### 11. Thuế thu nhập cá nhân của người khuyết tật có thể được giảm, giảm bao nhiêu thì gọi điện hỏi cơ quan thuế tỉnh + +- Chi phí: Không tốn tiền. Kê khai thì chọn một mục, hoặc nộp một bộ hồ sơ khai báo một lần. +- Nói thẳng: Luật Thuế thu nhập cá nhân viết rằng thu nhập của người khuyết tật có thể được giảm thuế. Nhưng giảm bao nhiêu, giảm bao lâu, luật giao cho chính quyền tỉnh quy định, nên mỗi tỉnh một khác. Tỉnh nào đặt hạn mức miễn giảm theo năm, tỉnh nào giảm theo tỷ lệ. Cách làm là gọi 12366 hỏi cơ quan thuế địa phương, hoặc tìm mục miễn giảm trong ứng dụng Thuế thu nhập cá nhân. +- Lợi ích: Điều năm của Luật Thuế thu nhập cá nhân (《个人所得税法》):「有下列情形之一的，可以减征个人所得税，具体幅度和期限，由省、自治区、直辖市人民政府规定，并报同级人民代表大会常务委员会备案：（一）残疾、孤老人员和烈属的所得；（二）因自然灾害遭受重大损失的。」— tức thuế thu nhập cá nhân có thể được giảm trong hai trường hợp sau, mức độ và thời hạn do chính quyền nhân dân tỉnh, khu tự trị, thành phố trực thuộc trung ương quy định và báo cáo Ủy ban Thường vụ Quốc hội cùng cấp: (một) thu nhập của người khuyết tật, người cao niên sống cô đơn và thân nhân liệt sĩ; (hai) thiệt hại lớn do thiên tai gây ra. Điều ba mươi sáu của Luật Bảo vệ Người khuyết tật (《残疾人保障法》) quy định thêm, nhà nước cho người khuyết tật làm kinh doanh cá thể được hưởng ưu đãi thuế theo pháp luật, và được miễn các loại phí hành chính, phí sự nghiệp công (toàn quốc, Luật Thuế thu nhập cá nhân có hiệu lực từ ngày 1 tháng 1 năm 2019) +- Mức bằng chứng: A +- Nguồn: Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc (中华人民共和国个人所得税法)（2018 年修正，第五条）. ；Luật Bảo vệ Người khuyết tật của Cộng hoà Nhân dân Trung Quốc (中华人民共和国残疾人保障法)（第三十六条）. +- Ghi chú: Đặt mức lợi ích là「nhỏ」, vì bản thân luật không đưa ra con số, phần lớn các tỉnh miễn giảm mỗi năm ở mức vài trăm đến vài nghìn tệ. Số tiền không nhiều, nhưng làm một lần thì dùng được nhiều năm, đáng đi. Cụ thể tỉnh của bạn giảm bao nhiêu thì chỉ thông báo của cơ quan thuế địa phương mới có giá trị, bảng đối chiếu được chuyền tay trên mạng thường xuyên lỗi thời. Người tự mở cửa hàng nhỏ thì các loại phí hành chính, phí sự nghiệp công được miễn theo Điều ba mươi sáu của Luật Bảo vệ Người khuyết tật, lúc xin giấy phép nhớ nêu ra. + +### 12. Chó dẫn đường cho người mù được vào nơi công cộng và phương tiện giao thông công cộng, người mù đi xe buýt trong thành phố miễn phí + +- Chi phí: Không tốn tiền. Bị chặn thì phải mở miệng viện dẫn điều luật, có khi phải cãi lại một lúc. +- Nói thẳng: Pháp luật quy định nơi công cộng và giao thông công cộng phải tạo điều kiện thuận tiện cho người khuyết tật mang theo chó dẫn đường, chó hỗ trợ nghe, chó hỗ trợ. Người mù có giấy tờ hợp lệ được đi xe buýt trong thành phố, tàu điện ngầm và phà miễn phí, người khuyết tật được mang miễn phí thiết bị hỗ trợ cần thiết đi kèm. Bị chặn thì nói ra hai điều đó, không được thì khiếu nại. +- Lợi ích: Điều bốn mươi sáu của Luật Xây dựng môi trường không vật cản (《无障碍环境建设法》):「公共场所经营管理单位、交通运输设施和公共交通运输工具的运营单位应当为残疾人携带导盲犬、导听犬、辅助犬等服务犬提供便利。」— tức đơn vị quản lý nơi công cộng, cơ sở vận tải và đơn vị vận hành phương tiện giao thông công cộng phải tạo điều kiện thuận tiện cho người khuyết tật mang theo chó phục vụ như chó dẫn đường, chó hỗ trợ nghe, chó hỗ trợ. Điều ba mươi lăm quy định, đường dây nóng phục vụ công việc của cơ quan hành chính và các hệ thống gọi cấp cứu như báo động cầu cứu, ứng cứu hoả hoạn, tai nạn giao thông, cấp cứu y tế, dần dần phải có chức năng không vật cản như giọng nói, chữ lớn, chữ Braille, gọi bằng một nút. Điều sáu mươi hai quy định, mọi tổ chức và cá nhân có quyền khiếu nại, báo việc về hành vi vi phạm luật này, cơ quan chủ trì phải xử lý kịp thời và trả lời. Điều năm mươi của Luật Bảo vệ Người khuyết tật (《残疾人保障法》):「盲人持有效证件免费乘坐市内公共汽车、电车、地铁、渡船等公共交通工具。盲人读物邮件免费寄递。」— tức người mù có giấy tờ hợp lệ được đi xe buýt, xe điện, tàu điện ngầm, phà và các phương tiện giao thông công cộng khác trong thành phố miễn phí, và bưu kiện có tài liệu đọc cho người mù được chuyển miễn phí. Cùng điều khoản đó còn quy định, chính quyền cấp huyện trở lên phải tạo điều kiện thuận tiện và ưu đãi cho người khuyết tật khi đi phương tiện giao thông công cộng, người khuyết tật được miễn phí mang theo thiết bị hỗ trợ cần thiết (toàn quốc, Luật Xây dựng môi trường không vật cản có hiệu lực từ ngày 1 tháng 9 năm 2023) +- Mức bằng chứng: A +- Nguồn: Luật Xây dựng môi trường không vật cản của Cộng hoà Nhân dân Trung Quốc (中华人民共和国无障碍环境建设法)（第三十五条、第四十六条、第六十二条）. ；Luật Bảo vệ Người khuyết tật của Cộng hoà Nhân dân Trung Quốc (中华人民共和国残疾人保障法)（第五十条）. +- Ghi chú: Luật viết là「提供便利」, không kèm chế tài, nên lúc cãi lại tại chỗ phía bên kia không nhất định nhường ngay. Lưu trong điện thoại một ảnh chụp điều luật, hữu ích hơn học thuộc số điều. Không giải quyết được tại chỗ thì khiếu nại theo Điều sáu mươi hai, cơ quan chủ trì phải trả lời. Đường sắt và hàng không có quy định riêng, trước khi đi hãy gọi số chăm sóc khách hàng xác nhận. Giấy tờ「hợp lệ」để được đi miễn phí, mỗi nơi công nhận không hẳn giống nhau, hãy hỏi hội người khuyết tật địa phương xem loại nào được công nhận. + +### 13. Người khuyết tật dự thi cao học có thể xin các hỗ trợ hợp lý, dùng đề Braille thì thời gian thi tăng gấp rưỡi + +- Chi phí: Không tốn tiền. Phải nộp một đơn bằng văn bản trong thời hạn tỉnh quy định, còn phải đến tận nơi để xác nhận. +- Nói thẳng: Thí sinh khuyết tật dự thi cao học có thể xin đề Braille, đề chữ lớn, miễn phần thi nghe ngoại ngữ, phòng thi riêng, mang máy trợ thính và xe lăn vào phòng thi, thời gian thi cũng được kéo dài. Những thứ này không tự động có, phải nộp đơn bằng văn bản đúng thời hạn tỉnh quy định, kèm giấy chứng nhận người khuyết tật từ thế hệ thứ hai trở lên. Không đồng ý với kết quả thì có thể xin cơ quan giáo dục tỉnh xét lại. +- Lợi ích: Điều năm của Quy định quản lý việc người khuyết tật tham dự kỳ thi tuyển sinh quốc gia vào các trường đại học thường dân (《残疾人参加普通高等学校招生全国统一考试管理规定》) liệt kê mười hai hạng mục hỗ trợ hợp lý. Gồm cấp đề Braille hiện hành, đề chữ lớn hoặc đề thường cho thí sinh khuyết tật về thị giác; miễn trừ phần thi nghe ngoại ngữ cho thí sinh khuyết tật về thính giác; cho phép thí sinh khuyết tật về thị giác mang theo bút Braille, bảng ghi Braille, máy đánh máy Braille không có chức năng lưu trữ, máy phóng hình không có chức năng lưu trữ, gậy tờ, kính lúp quang học và các thiết bị hỗ trợ khác; cho phép thí sinh khuyết tật về thính giác mang máy trợ thính, cấy ghép ốc tai nhân tạo; cho phép thí sinh di chuyển khó khăn dùng xe lăn, khung đi bộ, có nhu cầu đặc biệt thì được mang bàn ghế riêng. Mức kéo dài ghi ở khoản sáu của Điều năm:「使用盲文试卷的视力残疾考生的考试时间，在该科目规定考试总时长的基础上延长50%；使用大字号试卷或普通试卷的视力残疾考生、因脑瘫或其他疾病引起的上肢无法正常书写或无上肢考生等书写特别困难考生的考试时间，在该科目规定考试总时长的基础上延长30%」— thời gian thi của thí sinh khuyết tật về thị giác dùng đề Braille được kéo dài thêm 50% trên tổng thời gian thi quy định của môn đó; thời gian thi của thí sinh khuyết tật về thị giác dùng đề chữ lớn hoặc đề thường, của thí sinh viết rất khó như liệt não, hoặc do bệnh khác gây ra không viết nổi bằng tay trên, hoặc không có tay trên, được kéo dài thêm 30% trên tổng thời gian thi quy định của môn đó. Ngoài ra còn có ưu tiên vào khu thi, phòng thi chuẩn hoá riêng, bố trí cán bộ hướng dẫn hỗ trợ và thông dịch viên ngôn ngữ ký hiệu. Điều bảy quy định thủ tục xin: thí sinh nộp đơn bằng văn bản chính thức theo thời gian, địa điểm, cách thức mà cơ quan thi cấp tỉnh quy định, kèm giấy chứng nhận người khuyết tật từ thế hệ thứ hai trở lên của bản thân; nhóm chuyên gia xác nhận tại chỗ và đánh giá tổng hợp, cơ quan thi cấp tỉnh cấp «Thông báo». Điều tám quy định, thí sinh có ý kiến khác về «Thông báo» thì có thể đề nghị bằng văn bản xét lại tới cơ quan quản lý giáo dục cấp tỉnh. Điều chín quy định, với trường hợp miễn trừ phần thi nghe ngoại ngữ, điểm môn ngoại ngữ tính theo «điểm bài viết × tổng điểm môn ngoại ngữ ÷ tổng điểm phần viết». Điều mười tám quy định, người khuyết tật tham dự các kỳ thi giáo dục quốc gia khác mà cần hỗ trợ hợp lý thì có thể tham chiếu quy định này mà thực hiện (toàn quốc, ban hành và có hiệu lực ngày 7 tháng 4 năm 2017) +- Mức bằng chứng: A +- Nguồn: Bộ Giáo dục (教育部), Hiệp hội Người khuyết tật Trung Quốc (中国残联) (2017). Quy định quản lý việc người khuyết tật tham dự kỳ thi tuyển sinh quốc gia vào các trường đại học thường dân (《残疾人参加普通高等学校招生全国统一考试管理规定》)(教学〔2017〕4 号，第五条、第七条、第八条、第九条、第十八条). +- Ghi chú: Thời hạn xin do từng tỉnh quy định, ghi trong phương pháp đăng ký dự thi cao học của năm đó, hết hạn là mất. Nên học kỳ một năm ba của lớp mười hai phải lên trang chính thức của cơ quan thi cấp tỉnh xác nhận thời hạn. Điều mười tám còn quy định, kỳ thi sau đại học, thi tuyển sinh người lớn và các kỳ thi giáo dục quốc gia khác cũng có thể xin tham chiếu. Quy định trong giai đoạn học đường rằng trường không được từ chối nhận, xem mục 14 của phần này (trường học không được từ chối nhận). Người được lợi của mục này là bạn hoặc con bạn. + +### 14. Trẻ khuyết tật xin nhập học, trường học không được từ chối; trẻ không đến được trường thì giáo dục sở bố trí dạy đến nhà + +- Chi phí: Không tốn tiền. Phải nộp một đơn, có thể phải qua một lần đánh giá. Khi trường từ chối thì phải kêu lên phòng giáo dục. +- Nói thẳng: Trường mầm non và các loại trường khác không được từ chối nhận trẻ khuyết tật hợp điều kiện vào học, trường dạy nghề thường cũng vậy. Trẻ cần người chăm riêng, không đến được trường, thì giáo dục huyện bố trí dạy đến nhà hoặc học từ xa, vẫn có hồ sơ học tập. Việc nhập học và chuyển trường do giáo dục sở mời hội đồng chuyên gia đánh giá rồi quyết định, không phải trường tự ý quyết. +- Lợi ích: Điều bảy của Quy định về Giáo dục người khuyết tật (《残疾人教育条例》):「学前教育机构、各级各类学校及其他教育机构应当依照本条例以及国家有关法律、法规的规定，实施残疾人教育；对符合法律、法规规定条件的残疾人申请入学，不得拒绝招收。」— tức cơ sở giáo dục mầm non, các trường học mọi cấp mọi loại và cơ sở giáo dục khác phải thực hiện giáo dục cho người khuyết tật theo quy định này và theo pháp luật, quy định của nhà nước; không được từ chối nhận người khuyết tật hợp điều kiện theo pháp luật, quy định xin nhập học. Điều hai mươi chín:「普通职业学校不得拒绝招收符合国家规定的录取标准的残疾人入学，普通职业培训机构应当积极招收残疾人入学。」— trường dạy nghề thường không được từ chối nhận người khuyết tật đạt chuẩn tuyển sinh do nhà nước quy định, cơ sở đào tạo nghề thường phải tích cực nhận người khuyết tật. Quy định còn viết, trẻ em, thiếu niên khuyết tật đúng tuổi cần người chăm riêng, không đến trường học được, thì cơ quan quản lý giáo dục của chính quyền huyện điều phối, thực hiện giáo dục bắt buộc bằng cách dạy đến nhà hoặc giáo dục từ xa, và đưa vào quản lý hồ sơ học tập. Điều hai mươi quy định, cơ quan quản lý giáo dục của chính quyền huyện phải phối hợp với cơ quan y tế, cơ quan dân vụ, hội người khuyết tật để lập một hội đồng chuyên gia về giáo dục người khuyết tật gồm các chuyên gia về giáo dục, tâm lý, phục hồi chức năng và công tác xã hội, hội đồng đó đánh giá tình trạng sức khoẻ thể chất, khả năng tiếp thụ giáo dục và khả năng thích nghi với đời sống học tập ở trường rồi đề nghị về nhập học, chuyển trường; kết quả đánh giá thuộc quyền riêng tư của trẻ, các cơ quan và nhân viên có nghĩa vụ giữ bí mật. Điều hai mươi ba quy định, học sinh khuyết tật học chung lớp ở trường thường có thể áp dụng chương trình, tiêu chuẩn, giáo trình của giáo dục bắt buộc thường, nhưng yêu cầu học tập đối với họ có thể linh hoạt ở mức độ thích hợp. Điều năm mươi hai quy định, người khuyết tật tham dự kỳ thi quốc gia mà cần điều kiện hỗ trợ và hỗ trợ hợp lý thì có thể đề nghị (toàn quốc, có hiệu lực từ ngày 1 tháng 5 năm 2017) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2017). Quy định về Giáo dục người khuyết tật (《残疾人教育条例》)(第七条、第二十条、第二十三条、第二十九条、第五十二条). +- Ghi chú: Trường nói miệng「chỗ chúng tôi không có điều kiện」thì không được coi là. Bắt nó trả lời bằng văn bản, cầm bản trả lời đó đi tìm giáo dục huyện. Việc đánh giá do giáo dục sở ủy thác cho hội đồng chuyên gia. Câu「yêu cầu học tập có thể linh hoạt ở mức độ thích hợp」trong Điều hai mươi ba rất hữu ích, gặp trường dùng thành tích làm lý do đuổi học thì viện dẫn. Bị bắt nạt ở trường thì lưu chứng cứ thế nào, trường phải đi qua bước nào, xem mục 3 ở phần 30 (bắt nạt học đường). Hỗ trợ hợp lý cho kỳ thi cao học, xem mục 13 của phần này (hỗ trợ hợp lý). Người được lợi của mục này là con bạn. + +### 15. Mất chân phải hoặc cả hai chân dưới vẫn thi được giấy phép lái xe, loại xe được phép lái gọi là C5 + +- Chi phí: Học phí tầm như giấy phép lái xe thường, ở hầu hết các nơi vài nghìn tệ. Xe phải độ lại sang điều khiển tay. Cứ ba năm khám sức khoẻ một lần. +- Nói thẳng: Người mất chân phải hoặc cả hai chân dưới, không cử động được, nhưng tự ngồi vững được và phần thân trên đủ điều kiện, thì được thi giấy phép lái xe hơi nhỏ số hộc tự động dành riêng cho người khuyết tật, mã C5. Người mất một bàn tay cũng xin được. Người khuyết thị lực một mắt, người khuyết thính giác có các loại xe riêng để thi. Khám sức khoẻ phải làm ở bệnh viện chuyên khoa do tỉnh công nhận. +- Lợi ích: Phụ lục 1 của Quy định về việc đăng ký và sử dụng giấy phép lái xe cơ giới (《机动车驾驶证申领和使用规定》) viết khá chi tiết về điều kiện thể chất. Khoản 8:「右下肢、双下肢缺失或者丧失运动功能但能够自主坐立，且上肢符合本项第5目规定的，可以申请残疾人专用小型自动挡载客汽车准驾车型的机动车驾驶证。一只手掌缺失，另一只手拇指健全，其他手指有两指健全，上肢和手指运动功能正常，且下肢符合本项第6目规定的，可以申请残疾人专用小型自动挡载客汽车准驾车型的机动车驾驶证」— người mất chân phải, mất cả hai chân dưới hoặc mất chức năng vận động nhưng ngồi đứng tự được, đồng thời phần thân trên đáp ứng khoản 5 của cùng mục này, thì có thể xin giấy phép lái xe cơ giới loại xe hành khách nhỏ, số hộc tự động dành riêng cho người khuyết tật; người mất một bàn tay, bàn tay còn lại có ngón cái lành, các ngón tay khác có hai ngón lành, chức năng vận động của tay và ngón tay bình thường, đồng thời phần dưới đáp ứng khoản 6 của cùng mục này, thì cũng có thể xin giấy phép lái xe cơ giới loại xe hành khách nhỏ, số hộc tự động dành riêng cho người khuyết tật. Khoản 6: người chỉ mất hoặc mất chức năng vận động chân dưới trái nhưng chân phải bình thường thì có thể xin loại xe hành khách nhỏ, số hộc tự động. Về thị lực:「单眼视力障碍，优眼裸视力或者矫正视力达到对数视力表5.0以上，且水平视野达到150度的」, tức khuyết thị lực một mắt, thị lực trần hoặc thị lực đã điều chỉnh của mắt tốt đạt 5.0 trở lên trên bảng thị lực logarit, và tầm nhìn ngang đạt 150 độ, có thể xin loại xe ô tô nhỏ, xe ô tô nhỏ số hộc tự động, xe tải nhỏ chạy chậm, xe ba bánh, xe hành khách nhỏ số hộc tự động dành riêng cho người khuyết tật. Về thính giác: người khuyết thính giác nhưng đeo thiết bị trợ thính thì phân biệt được hướng âm thanh thì có thể xin loại xe ô tô nhỏ, xe ô tô nhỏ số hộc tự động, khi lái xe phải đeo thiết bị trợ thính. Điều tám mươi lăm quy định, giấy chứng nhận điều kiện thể chất nộp khi làm thủ tục giấy phép lái xe hành khách nhỏ, số hộc tự động dành riêng cho người khuyết tật phải do cơ sở y tế chuyên khoa được cơ quan quản lý y tế cấp tỉnh công nhận cấp. Người có loại xe được phép lái đó phải khám sức khoẻ ba năm một lần, nộp giấy chứng nhận điều kiện thể chất trong vòng ba mươi ngày sau khi kết thúc chu kỳ tính điểm. Điều bảy mươi tám quy định, khi lái xe phải đặt đúng quy định biển hiệu dành riêng cho xe của người khuyết tật trên thân xe (toàn quốc, Quy định sửa đổi theo lệnh Bộ Công an số 172, có hiệu lực từ ngày 1 tháng 1 năm 2025) +- Mức bằng chứng: A +- Nguồn: Lệnh Bộ Công an số 172 (公安部令第 172 号). Quy định về việc đăng ký và sử dụng giấy phép lái xe cơ giới (《机动车驾驶证申领和使用规定》)(第七十八条、第八十五条、附件 1). +- Ghi chú: Rẻ hơn nhiều là xe lăn điện dành cho người khuyết tật. Điều một trăm mười chín của Luật An toàn giao thông đường bộ (《道路交通安全法》) xếp nó vào nhóm phương tiện không động cơ, nên không cần giấy phép lái xe; Điều năm mươi tám quy định, xe lăn điện dành cho người khuyết tật chạy trong làn phương tiện không động cơ thì tốc độ tối đa không được quá mười lăm ki lô mét một giờ. Xe loại C5 phải lắp bước ga và phanh tay, độ xe phải tìm nhà sản xuất có đủ điều kiện, độ xong đi đăng ký thay đổi tại cơ quan quản lý phương tiện. Ba chu kỳ tính điểm trôi qua mà trong vòng một năm không nộp giấy chứng nhận điều kiện thể chất thì giấy phép lái xe bị thu hồi. + +### 16. Phục hồi chức năng thì làm ở cơ sở chính quy có khoa phục hồi chức năng, cường độ nghe theo chuyên gia trị liệu, không phải càng mạnh càng tốt + +- Chi phí: Phục hồi chức năng nội trú mỗi ngày vài trăm tệ, phần tự trả thì tuỳ địa phương. Thời gian tính theo tháng. Tập luyện mỗi ngày, rất chán. +- Nói thẳng: Đưa người bệnh về nhà nằm chờ tự hồi phục là sai lầm phổ biến nhất. Sau đột quỵ, chăm sóc có tổ chức bởi nhóm đa chuyên ngành cho ra nhiều người hơn còn sống sau một năm và còn ở được tại nhà. Nhưng cũng đừng tự ý tăng cường, sau đột quỵ trong vòng 24 giờ mà tập sớm hơn, thường xuyên hơn, nhiều hơn thì kết quả sau ba tháng lại kém hơn. Tập bao nhiêu, tập lúc nào thì nghe khoa phục hồi chức năng và chuyên gia trị liệu. +- Lợi ích: Một bài tổng hợp Cochrane đưa vào 29 thử nghiệm, 5902 người tham gia, so sánh chăm sóc nội trú có tổ chức theo đơn vị đột quỵ với các cách chăm sóc khác. Khi theo dõi tới lần cuối (trung vị 1 năm), tỷ lệ bất lợi (tử vong, tàn tật, hoặc cần chăm sóc trong cơ sở) có odds ratio 0.77 (95% CI 0.69–0.87, khoảng tin cậy), tử vong có odds ratio 0.76 (95% CI 0.66–0.88), tử vong hoặc phải dựa vào người khác có odds ratio 0.75 (95% CI 0.66–0.85), chất lượng bằng chứng ở mức trung bình. Quy đổi ra số tuyệt đối, cứ 100 người được chăm sóc theo đơn vị đột quỵ thì có thêm 2 người sống, thêm 6 người ở được tại nhà, thêm 6 người tự lo được việc sinh hoạt. Mặt khác, thử nghiệm AVERT phân ngẫu nhiên 2104 bệnh nhân tại 56 đơn vị đột quỵ cấp tính ở 5 nước, so sánh「chăm sóc thường quy」với「chăm sóc thường quy cộng vận động cường độ cao siêu sớm」. Ở thời điểm ba tháng, nhóm kết quả tốt (điểm 0—2 trên thang Rankin cải biến) có 480 người (46%) ở nhóm siêu sớm, 525 người (50%) ở nhóm thường quy, odds ratio sau điều chỉnh 0.73 (95% CI 0.59–0.90, P=0.004). Tử vong là 88 người so với 72 người, không ra khác biệt (odds ratio 1.34, 95% CI 0.93–1.93, P=0.113) +- Mức bằng chứng: A +- Nguồn: Langhorne P, Ramachandra S; Stroke Unit Trialists' Collaboration (2020). Organised inpatient (stroke unit) care for stroke: network meta-analysis. Cochrane Database of Systematic Reviews, 4:CD000197. ；AVERT Trial Collaboration group (2015). Efficacy and safety of very early mobilisation within 24 h of stroke onset (AVERT): a randomised controlled trial. Lancet, 386(9988):46-55. +- Ghi chú:Tranh cãi: Cả hai bằng chứng này đều đến từ đột quỵ, áp sang chấn thương tủy sống, cắt cụt, bỏng và các nguyên nhân khuyết tật khác thì là suy diễn lan. Kết quả của AVERT nói「càng sớm càng mạnh」không đúng, không phải nói vận động sớm là vô ích. Trong thử nghiệm, đại đa số người ở cả hai nhóm đều đã xuống giường trong vòng 24 giờ, khác nhau là tần suất và tổng lượng. Khi chọn cơ sở, hãy xem nó có khoa phục hồi chức năng không, có chuyên gia trị liệu phục hồi chức năng và chuyên gia hoạt động trị liệu (occupational therapist) làm việc toàn thời gian không, đừng chỉ nhìn hai chữ「phục hồi」trên biển hiệu. Chuyển viện giữa tuyến cơ sở và tuyến trên, mức đồng chi trả tính thế nào, xem mục 1 ở phần 24 (chuyển viện theo từng tuyến). Phục hồi chức năng tại cộng đồng và lắp thiết bị hỗ trợ, xem mục 7 của phần này (hỏi hết ở hội người khuyết tật cấp huyện). + +### 17. Nghe kém thì đi lắp máy trợ thính, lý do lắp là để nghe thấy, không phải để phòng bệnh Alzheimer + +- Chi phí: Máy trợ thính giá vài nghìn đến vài chục nghìn tệ, phần lớn phải tự trả. Phải đo khớp âm thanh, còn phải chỉnh vài lần xa nhà. +- Nói thẳng: Câu「đeo máy trợ thính phòng bệnh Alzheimer」hiện chưa đứng vững. Một thử nghiệm gần một nghìn người trong ba năm, xét chung không ra khác biệt. Chỉ trong nhóm vốn đã có nguy cơ suy giảm nhận thức cao hơn mới thấy khác biệt. Nên lý do để lắp máy trợ thính là nghe thấy ngay lúc này, nói chuyện được với người, khi qua đường nghe thấy xe, không phải để giữ não cho mấy chục năm sau. +- Lợi ích: Thử nghiệm ACHIEVE làm ở 4 cộng đồng tại Hoa Kỳ, đưa vào 977 người từ 70 đến 84 tuổi, bị mất thính giác chưa được điều trị, không có rối loạn nhận thức rõ rệt. Ngẫu nhiên chia hai nhóm, một nhóm nhận can thiệp về thính giác (tư vấn thính giác và lắp máy trợ thính), một nhóm nhận giáo dục sức khoẻ làm nhóm chứng, theo dõi 6 tháng một lần. Điểm chính là thay đổi điểm nhận thức toàn cục trong 3 năm. Trong phân tích chính gộp hai nhóm dân số, hai nhóm không có khác biệt có ý nghĩa: nhóm can thiệp thính giác -0.200 (95% CI -0.256 đến -0.144), nhóm chứng -0.202 (95% CI -0.258 đến -0.145), chênh lệch 0.002 (95% CI -0.077 đến 0.081), P=0.96. Một phân tích độ nhạy định sẵn cho thấy hiệu quả của can thiệp có khác nhau giữa hai nhóm dân số (tương tác P=0.010). Người tham dự từ nghiên cứu quan sát dài hạn về sức khoẻ tim mạch lâu dài lớn tuổi hơn, có nhiều yếu tố nguy cơ suy giảm nhận thức hơn, điểm nhận thức nền thấp hơn. Kết luận của nhóm nghiên cứu là can thiệp thính giác có thể giảm mức thay đổi nhận thức trong ba năm ở nhóm nguy cơ suy giảm nhận thức cao, còn ở nhóm nguy cơ thấp thì không. Thử nghiệm không báo cáo biến chứng nặng nào do nghiên cứu +- Mức bằng chứng: A +- Nguồn: Lin FR, Pike JR, Albert MS, et al.; ACHIEVE Collaborative Research Group (2023). Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): a multicentre, randomised controlled trial. Lancet, 402(10404):786-797. +- Ghi chú:Tranh cãi: Câu「máy trợ thính làm giảm gần một nửa suy giảm nhận thức」hay thấy trên mạng đến từ một nhóm con trong thử nghiệm này, không phải kết quả chính của nó, kết quả chính là hai nhóm không có khác biệt.「Lắp máy trợ thính」và「phòng bệnh Alzheimer」phải tách ra mà xem: cái trước thì nên làm, cái sau thì đừng lấy làm lý do. Thí sinh khuyết thị giác dự thi cao học được miễn phần thi nghe ngoại ngữ, xem mục 13 của phần này (hỗ trợ hợp lý). Người khuyết thính giác lái xe thì phải đeo thiết bị trợ thính, xem mục 15 của phần này (C5). + +### 18. Nói một người lớn「không quản được việc của mình」thì phải xin tòa án công nhận, người nhà nói không tính + +- Chi phí: Nộp đơn ra tòa thì phải trả phí tố tụng, thường vài chục đến một trăm tệ. Phí giám định tính riêng, thường vài nghìn tệ, người xin trả trước. Đi hết thủ tục mất mấy tháng. +- Nói thẳng: Một người lớn có「không có năng lực pháp lý dân sự」hay「năng lực pháp lý dân sự bị hạn chế」hay không, chỉ tòa án mới công nhận được. Người nhà, cơ quan, tổ dân phố không tự quyết được. Phải do người có lợi ích liên quan hoặc tổ chức liên quan nộp đơn ra tòa. Ngược lại, khi sức khoẻ trí tuệ hoặc tâm thần hồi phục thì cũng có thể xin công nhận người đó đã hồi phục năng lực pháp lý dân sự. Thủ tục này vừa là hạn chế, vừa là bảo vệ. +- Lợi ích: Điều hai mươi mối của Bộ luật Dân sự (《民法典》):「不能辨认自己行为的成年人为无民事行为能力人，由其法定代理人代理实施民事法律行为。」— người lớn không nhận biết được hành vi của mình là người không có năng lực pháp lý dân sự, do người đại diện theo pháp luật của họ thay họ thực hiện giao dịch pháp lý dân sự. Điều hai mươi hai:「不能完全辨认自己行为的成年人为限制民事行为能力人，实施民事法律行为由其法定代理人代理或者经其法定代理人同意、追认；但是，可以独立实施纯获利益的民事法律行为或者与其智力、精神健康状况相适应的民事法律行为。」— người lớn không nhận biết được hoàn toàn hành vi của mình là người có năng lực pháp lý dân sự bị hạn chế, việc thực hiện giao dịch pháp lý dân sự phải do người đại diện theo pháp luật của họ thay, hoặc phải được người đó đồng ý, công nhận hậu nững; tuy nhiên người đó có thể tự thực hiện những giao dịch pháp lý dân sự thuần lợi cho mình, hoặc những giao dịch pháp lý dân sự phù hợp với tình trạng trí tuệ, sức khoẻ tâm thần của mình. Điều hai mươi bốn:「不能辨认或者不能完全辨认自己行为的成年人，其利害关系人或者有关组织，可以向人民法院申请认定该成年人为无民事行为能力人或者限制民事行为能力人。」— người lớn không nhận biết, hoặc không nhận biết hoàn toàn hành vi của mình thì người có lợi ích liên quan hoặc tổ chức liên quan có thể xin Tòa án nhân dân công nhận người lớn đó là người không có năng lực pháp lý dân sự hoặc có năng lực pháp lý dân sự bị hạn chế. Cùng điều khoản đó còn quy định, sau khi được công nhận, theo đơn của bản thân người đó, người có lợi ích liên quan hoặc tổ chức liên quan, Tòa án nhân dân có thể căn cứ tình trạng trí tuệ, sức khoẻ tâm thần đã hồi phục mà công nhận người đó hồi phục thành người có năng lực pháp lý dân sự bị hạn chế hoặc có năng lực pháp lý dân sự đầy đủ. Tổ chức liên quan gồm: Ủy ban nhân dân khu phố, Ủy ban nhân dân làng, trường học, cơ sở y tế, hội liên hiệp phụ nữ, hội người khuyết tật, tổ chức người cao niên lập hợp pháp, cơ quan dân vụ. Điều hai mươi ba quy định, người giám hộ của người không có năng lực pháp lý dân sự và của người có năng lực pháp lý dân sự bị hạn chế chính là người đại diện theo pháp luật của họ (toàn quốc, có hiệu lực từ ngày 1 tháng 1 năm 2021) +- Mức bằng chứng: A +- Nguồn: Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc (中华人民共和国民法典)（第二十一条、第二十二条、第二十三条、第二十四条）. +- Ghi chú: Thủ tục xin đi theo thủ tục đặc biệt của tòa án, nhanh hơn vụ án dân sự thường nhưng vẫn mất mấy tháng. Phải nộp giấy chứng nhận chẩn đoán của bệnh viện, tòa thường còn ủy thác cho bên giám định tư pháp, phí giám định thường vài nghìn tệ, người xin trả trước. Không qua thủ tục này mà ký thay, phân phối tài sản của người khác thì về sau có thể bị xác nhận là vô hiệu, nhà đã bán cũng có thể phải hoàn trả. Còn một điểm phải nhớ: năng lực pháp lý dân sự bị hạn chế không có nghĩa là không làm được gì. Những việc phù hợp với trí tuệ, tình trạng tâm thần của họ, cùng những việc thuần lợi cho họ, họ tự làm được, đừng thay tất cả. Người giám hộ xác định thế nào, xem mục 19 của phần này (người giám hộ). + +### 19. Người giám hộ của người lớn được xác định theo thứ tự luật định; người được giám hộ làm hại người khác thì người giám hộ bồi thường + +- Chi phí: Không tốn tiền. Khi có tranh chấp thì phải tìm Ủy ban nhân dân khu phố, Ủy ban nhân dân làng, cơ quan dân vụ hoặc tòa án để giải quyết. +- Nói thẳng: Sau khi người lớn mất năng lực hành vi, người giám hộ được xếp theo thứ tự: vợ chồng, cha mẹ và con cái, người thân thích khác, cuối cùng là cá nhân hoặc tổ chức khác sẵn lòng đảm nhiệm. Tranh chấp thì Ủy ban nhân dân khu phố, Ủy ban nhân dân làng hoặc cơ quan dân vụ chỉ định, không đồng ý thì kiện ra tòa. Khi người đó còn minh mẫn thì có thể chỉ định bằng văn bản người giám hộ về sau. Người được giám hộ làm hại người khác thì người giám hộ bồi thường. +- Lợi ích: Điều hai mươi tám của Bộ luật Dân sự: người lớn không có năng lực pháp lý dân sự hoặc có năng lực pháp lý dân sự bị hạn chế thì người giám hộ do những người có năng lực giám hộ dưới đây theo thứ tự đảm nhiệm:「（一）配偶；（二）父母、子女；（三）其他近亲属；（四）其他愿意担任监护人的个人或者组织，但是须经被监护人住所地的居民委员会、村民委员会或者民政部门同意。」— (một) vợ chồng; (hai) cha mẹ, con cái; (ba) người thân thích khác; (bốn) cá nhân hoặc tổ chức khác sẵn lòng làm người giám hộ, nhưng phải được Ủy ban nhân dân khu phố, Ủy ban nhân dân làng nơi người được giám hộ cư trú hoặc cơ quan dân vụ đồng ý. Điều ba mươi mốt: có tranh chấp về việc xác định người giám hộ thì Ủy ban nhân dân khu phố, Ủy ban nhân dân làng nơi người được giám hộ cư trú hoặc cơ quan dân vụ chỉ định, không đồng ý với chỉ định thì có thể xin Tòa án nhân dân chỉ định, cũng có thể trực tiếp xin Tòa án nhân dân. Khi chỉ định thì「应当尊重被监护人的真实意愿，按照最有利于被监护人的原则」— phải tôn trọng ý muốn thật của người được giám hộ, theo nguyên tắc có lợi nhất cho người được giám hộ. Trước khi chỉ định mà quyền lợi rơi vào trạng thái không ai bảo vệ thì Ủy ban nhân dân khu phố, Ủy ban nhân dân làng, tổ chức liên quan theo luật định hoặc cơ quan dân vụ làm người giám hộ tạm thời. Điều ba mươi ba:「具有完全民事行为能力的成年人，可以与其近亲属、其他愿意担任监护人的个人或者组织事先协商，以书面形式确定自己的监护人」— người lớn có năng lực pháp lý dân sự đầy đủ có thể thỏa thuận trước với người thân thích, cá nhân hoặc tổ chức khác sẵn lòng làm người giám hộ, rồi xác định người giám hộ của mình bằng văn bản. Điều một nghìn một trăm tám mươi tám:「无民事行为能力人、限制民事行为能力人造成他人损害的，由监护人承担侵权责任。监护人尽到监护职责的，可以减轻其侵权责任。」— người không có năng lực pháp lý dân sự, người có năng lực pháp lý dân sự bị hạn chế gây thiệt hại cho người khác thì người giám hộ chịu trách nhiệm về tội phạm gây thiệt hại. Người giám hộ đã làm tròn nghĩa vụ giám hộ thì có thể giảm trách nhiệm đó. Người có tài sản thì chi phí bồi thường lấy từ tài sản của chính người đó, phần thiếu thì người giám hộ bồi thường. Điều một nghìn một trăm tám mươi chín: người giám hộ ủy thác nghĩa vụ giám hộ cho người khác thì người giám hộ vẫn phải chịu trách nhiệm về tội phạm gây thiệt hại, người được ủy thác có lỗi thì chịu trách nhiệm tương ứng (toàn quốc, có hiệu lực từ ngày 1 tháng 1 năm 2021) +- Mức bằng chứng: A +- Nguồn: Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc (中华人民共和国民法典)（第二十八条、第三十一条、第三十三条、第一千一百八十八条、第一千一百八十九条）. +- Ghi chú: Điều ba mươi ba chính là giám hộ tự nguyện bằng văn bản. Khi người đó còn minh mẫn thì dùng văn bản chốt người được chọn, tiết kiệm được phần lớn tranh chấp về sau, cụ thể viết thế nào xem mục 1 ở phần 17 (giám hộ tự nguyện bằng văn bản). Đoạn bồi thường thì phải cẩn thận: giao người cho cơ sở nuôi dưỡng hay bà con trông coi thì trách nhiệm vẫn ở người giám hộ, người được ủy thác chỉ chịu trách nhiệm tương ứng khi có lỗi. Nên mua cho người được giám hộ một bảo hiểm trách nhiệm người giám hộ là có ý nghĩa. Điều bảy mươi tám của Luật Sức khỏe Tâm thần còn quy định, người giám hộ bỏ rơi người bệnh hoặc không làm tròn nghĩa vụ giám hộ mà gây thiệt hại thì chịu trách nhiệm bồi thường theo pháp luật. + +### 20. Bị từ chối tuyển, không được chuyển hợp đồng chính thức, bị trả thiếu lương vì khuyết tật thì có thể kiện, nhưng phải tính trước chi phí của cả thủ tục + +- Chi phí: Phí luật sư vài nghìn đến vài chục nghìn tệ, tự bỏ ra. Bản án sơ thẩm theo thủ tục thường từ 6 tháng trở lên, có thể kéo dài. Chứng cứ phải tự giữ. +- Nói thẳng: Pháp luật cấm phân biệt đối xử với người khuyết tật trong việc tuyển dụng, chuyển hợp đồng chính thức, thăng tiến, xét chức danh, thù lao, bảo hiểm xã hội. Bạn có thể khiếu nại với cơ quan chủ trì, cũng có thể kiện ra tòa. Nhưng thắng kiện xong thì phía bên kia không trả phí luật sư cho bạn. Vì vậy hãy khiếu nại trước, đồng thời xin trợ giúp pháp lý, rồi mới nghĩ tới kiện. +- Lợi ích: Điều ba của Luật Bảo vệ Người khuyết tật (《残疾人保障法》):「禁止基于残疾的歧视。禁止侮辱、侵害残疾人。」— cấm phân biệt đối xử dựa trên khuyết tật. Cấm xúc phạm, xâm phạm người khuyết tật. Điều ba mươi tám:「在职工的招用、转正、晋级、职称评定、劳动报酬、生活福利、休息休假、社会保险等方面，不得歧视残疾人。」— ở các mặt tuyển dụng, chuyển hợp đồng chính thức, thăng tiến, xét chức danh, thù lao lao động, phúc lợi sinh hoạt, nghỉ ngơi, bảo hiểm xã hội thì không được phân biệt đối xử với người khuyết tật. Cùng điều khoản đó còn quy định, đơn vị có công nhân viên khuyết tật phải căn cứ đặc điểm của họ mà tạo điều kiện lao động và bảo vệ lao động thích hợp, và tuỳ nhu cầu thực tế mà cải tạo nơi làm việc, thiết bị lao động và cơ sở sinh hoạt. Điều sáu mươi bốn:「违反本法规定，在职工的招用等方面歧视残疾人的，由有关主管部门责令改正；残疾人劳动者可以依法向人民法院提起诉讼。」— vi phạm luật này mà phân biệt đối xử với người khuyết tật trong việc tuyển dụng công nhân viên thì bị cơ quan chủ trì liên quan yêu cầu sửa; người lao động khuyết tật có quyền khởi kiện ra Tòa án nhân dân theo pháp luật. Điều bốn của Quy định về Việc làm cho người khuyết tật (《残疾人就业条例》):「禁止在就业中歧视残疾人。」— cấm phân biệt đối xử với người khuyết tật trong việc làm. Điều mười ba: đơn vị sử dụng lao động không được phân biệt đối xử với công nhân viên khuyết tật về chức vụ, thăng tiến, xét chức danh, thù lao, bảo hiểm xã hội, phúc lợi sinh hoạt. Điều bốn mươi hai của Luật Trợ giúp pháp lý quy định, người xin có tài liệu chứng minh thuộc「nhóm nhân dân đặc biệt không có nguồn sống cố định như trẻ vị thành niên, người cao niên, người khuyết tật」thì được miễn kiểm tra hoàn cảnh khó khăn tài chính. Điều bốn mươi lăm quy định, cơ quan trợ giúp pháp lý có cung cấp dịch vụ trợ giúp pháp lý cho người cao niên, người khuyết tật thì phải tuỳ tình hình thực tế mà cung cấp thiết bị và dịch vụ không vật cản (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Luật Bảo vệ Người khuyết tật của Cộng hoà Nhân dân Trung Quốc (中华人民共和国残疾人保障法)（第三条、第三十八条、第六十四条）. ；Hội đồng Nhà nước (国务院) (2007). Quy định về Việc làm cho người khuyết tật (《残疾人就业条例》)(第四条、第十三条). ；Luật Trợ giúp pháp lý của Cộng hoà Nhân dân Trung Quốc (中华人民共和国法律援助法)（第四十二条、第四十五条）. +- Ghi chú: Chi phí của cả thủ tục phải tính trước. Bản án sơ thẩm theo thủ tục thường từ 6 tháng trở lên, có thể kéo dài; thủ tục đơn giản là 3 tháng. Phí luật sư không nằm trong khoản phí tố tụng liệt kê ở Điều sáu của Biện pháp thu phí tố tụng, cơ chế「bên thua kiện chịu」ở Điều hai mươi chín không bao gồm phí luật sư. Vì vậy hãy đi trước hai con đường rẻ tiền: khiếu nại với cơ quan nhân sự và xã hội hoặc hội người khuyết tật, để cơ quan chủ trì yêu cầu sửa; đồng thời xin trợ giúp pháp lý, Điều bốn mươi hai miễn cho người khuyết tật việc kiểm tra hoàn cảnh khó khăn tài chính. Thu thập chứng cứ là khó nhất. Khi bị từ chối phỏng vấn, phía bên kia hiếm khi để lại lý do bằng văn bản, ghi âm được thì ghi, giữ được lịch sử trò chuyện thì giữ. Người đã vào làm thì đi theo tranh chấp lao động ra hoà giải lao động, cách lưu chứng cứ và thời hiệu xem mục 8 ở phần 19 (lưu chứng cứ trước khi nghỉ việc). Người được lợi của mục này là bạn. diff --git "a/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" new file mode 100644 index 00000000..0dc5a8b5 --- /dev/null +++ "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" @@ -0,0 +1,92 @@ +[← Về mục lục](../README.md) + +# 34. Đừng để thuốc trong nhà gây tai nạo + +Thuốc hạ sốt, thuốc giảm đau, thuốc cảm cúm, thuốc dạ dày, thuốc chống tiêu chảy, nhiều loại mua ở hiệu thuốc là tự mua được. Mua không cần đơn thuốc không có nghĩa là uống thế nào cũng được. Phần này không nói mỗi bệnh nên uống thuốc gì, chỉ chọn ra mấy việc chỉ cần một hành động là tránh được hậu quả nặng. + +Tất cả các mục tính theo tỷ lệ tử vong, gồm cả những kết cục sức khoẻ như suy gan, xuất huyết dạ dày, suy thận ở trẻ sơ sinh. Phần lớn lợi ích là「tránh được một vụ tai nạn nghiêm trọng」, nghiên cứu không đưa ra con số「làm rồi giảm được bao nhiêu phần trăm」, mức lợi ích được định theo mức độ nặng của hậu quả, mỗi mục ghi rõ trong Ghi chú căn cứ để định. + +Những việc đã viết ở chỗ khác, ở đây chỉ dẫn đường. Nuốt nhầm thuốc hoặc chất tẩy rửa thì làm gì trước, xem mục 20 ở phần 13（nuốt nhầm）. Thuốc bệnh mạn uống đủ theo chỉ định, xem mục 1 ở phần 16（uống thuốc đủ theo chỉ định）. Mua thuốc có đơn trên mạng phải qua duyệt đơn trước, xem mục 6 ở phần 28（thuốc giảm cân thì đi bệnh viện lấy đơn）. + +Người được lợi của các mục trong phần này là bạn và gia đình bạn, thuộc hai mức cao nhất. + +### 1. Trước khi uống hai loại thuốc cảm cúm hoặc thuốc giảm đau cùng lúc, xem bảng thành phần, paracetamol chỉ được chiếm một chỗ + +- Chi phí: Không tốn tiền. Nhìn bảng thành phần trên mỗi hộp một chút, một hai phút. +- Nói thẳng: Rất nhiều thuốc cảm cúm có pha paracetamol, uống hai hộp cùng lúc là uống gấp đôi, quá liều sẽ hại gan. Ở Hoa Kỳ, gần ba phần mười số người bị suy gan cấp tính do chính chất này đã chết trước khi kịp ghép gan. Tờ hướng dẫn trong nước khuyên mỗi ngày tối đa 2 g, trong thời gian uống thuốc cũng đừng uống rượu. +- Lợi ích: 22 bệnh viện ở Hoa Kỳ đăng ký liên tục 662 ca suy gan cấp tính trong 6 năm. Trong đó 275 ca (42%) do paracetamol gây ra, liều đã nuốt có trung vị 24 g. Trong 275 người này, 131 người (48%) ăn nhầm quá liều, không phải cố ý. Trong nhóm ăn nhầm quá liều, 38% còn uống kèm hai loại thuốc trở lên có chứa thành phần này. Trong 275 người đó, 27% chết trước khi kịp ghép gan, 8% được ghép gan. Năm 2020 Cục Quản lý Dược phẩm nhà nước yêu cầu tờ hướng dẫn thuốc không cần đơn phải ghi: dùng quá liều gây tổn thương gan nặng, liều tối đa trong một ngày không nên quá 2 g. Tờ hướng dẫn còn phải ghi, nên tránh dùng kèm với thuốc có paracetamol hoặc thuốc giảm đau hạ sốt khác. Quy định Hoa Kỳ yêu cầu loại thuốc không cần đơn này ghi ở cảnh báo thứ nhất ba tình huống gây hại gan: vượt quá liều tối đa trong ngày, uống chung với thuốc khác có paracetamol, và trong thời gian uống thuốc mỗi ngày uống trên 3 cốc rượu. +- Mức bằng chứng: B +- Nguồn: Larson AM, Polson J, Fontana RJ, et al.; Acute Liver Failure Study Group (2005). Acetaminophen-induced acute liver failure: results of a United States multicenter, prospective study. Hepatology, 42(6):1364-1372. ；Cục Quản lý Dược phẩm nhà nước (国家药监局) (2020). Thông báo về việc sửa đổi tờ hướng dẫn dạng thường và dạng giải phóng của paracetamol (2020年第15号) và phụ lục 2《对乙酰氨基酚常释制剂非处方药说明书修订要求》(Cục Quản lý Dược phẩm tỉnh Hồ Nam (湖南省药品监督管理局) đăng lại). ；21 CFR 201.326(a)(1)(iii)(A) Liver warning. +- Ghi chú: Paracetamol còn gọi là扑热息痛. Tên thuốc cảm cúm phối hợp có chữ「氨酚」trong tên, ví dụ 氨酚伪麻、氨咖黄敏, đều có thành phần này. Không chắc thì cứ mang mấy hộp thuốc tới dược sĩ ở hiệu thuốc. Để ở mức B là vì nhóm số liệu Hoa Kỳ là đăng ký ca bệnh, không có nhóm chứng, tính không ra xem bảng thành phần giúp bớt được bao nhiêu sự cố. Đặt lợi ích là「lớn」là theo hậu quả: trong số ca suy gan cấp tính có gần ba phần mười chết thẳng. + +### 2. Trẻ sốt không dùng aspirin, nimesulide và metamizole, thuốc cảm cúm có chứa metamizole cũng không cho trẻ vị thành niên uống + +- Chi phí: Không tốn tiền. Khi mua thuốc hạ sốt thì nhìn tên thuốc gốc và bảng thành phần. +- Nói thẳng: Trẻ sốt, ba loại thuốc này đều đừng dùng. Aspirin có liên quan tới hội chứng Reye, bệnh này tổn thương não và gan, theo thống kê ở Hoa Kỳ khoảng ba phần mười số trẻ mắc chết, sau khi cảnh báo được ban hành số ca giảm hơn chín phần mười. Nimesulide (尼美舒利) cấm dùng cho trẻ dưới 12 tuổi, metamizole (安乃近) cấm dùng cho người dưới 18 tuổi. +- Lợi ích: Hội chứng Reye gây bệnh não và thoái hóa mỡ gan, thường xuất hiện sau khi mắc cúm hoặc thủy đậu. Trong hệ thống giám sát toàn quốc của Hoa Kỳ từ 1981 đến 1997 có 1207 ca báo cáo ở người dưới 18 tuổi. 82% số trẻ bệnh đo được acid salicylic trong máu, đây là thành phần của aspirin và các thuốc cùng nhóm. Tỷ lệ chết chung 31%. Cảnh báo về nhóm thuốc chứa acid salicylic bắt đầu ban hành từ năm 1980. Năm đó có 555 ca báo cáo, là đỉnh, từ năm 1987 mỗi năm không quá 36 ca. Năm 2011 Cục Quản lý Thực phẩm và Dược phẩm nhà nước quy định, dạng uống của nimesulide cấm dùng cho trẻ dưới 12 tuổi. Năm 2020 Cục Quản lý Dược phẩm nhà nước quy định, viên metamizole cấm dùng cho trẻ vị thành niên dưới 18 tuổi. Các dạng phối hợp có chứa metamizole như viên phối hợp thanh hao và metamizole, viên và viên nang trọng cảm cũng cấm dùng cho người dưới 18 tuổi. +- Mức bằng chứng: B +- Nguồn: Belay ED, Bresee JS, Holman RC, Khan AS, Shahriari A, Schonberger LB (1999). Reye's syndrome in the United States from 1981 through 1997. New England Journal of Medicine, 340(18):1377-1382. ；Thông báo về việc tăng cường quản lý việc sử dụng dạng uống của nimesulide (国食药监安〔2011〕209号). ；Cục Quản lý Dược phẩm nhà nước (国家药监局) (2020). Thông báo về việc sửa đổi tờ hướng dẫn các chế phẩm liên quan đến metamizole (2020年第34号) và phụ lục 1 đến 3. +- Ghi chú: Metamizole với người lớn cũng không phải lựa chọn đầu tiên. Tờ hướng dẫn ghi rõ nó có thể gây mất bạch cầu hạt, thiếu máu do tủy xương và sốc phản vệ, chỉ dùng trong thời gian ngắn khi bệnh nặng, cấp tốc mà không có thuốc nào khác hiệu quả. Tên thuốc trọng cảm nghe như thuốc cảm cúm thường, bên trong có metamizole. Trẻ nên dùng thuốc hạ sốt gì, liều bao nhiêu theo cân nặng, hỏi bác sĩ nhi khoa hoặc dược sĩ. Trẻ dưới 3 tháng sốt thì đi thẳng bệnh viện, xem mục 8 ở phần 20 (38 ℃ thì đi thẳng bệnh viện). Để ở mức B là vì số liệu về hội chứng Reye đến từ báo cáo giám sát, việc số ca giảm và việc ban hành cảnh báo chỉ là quan hệ trước sau, không phải thử nghiệm đối chứng. Đặt lợi ích là「lớn」, vì sau khi có cảnh báo số ca mỗi năm rơi từ 555 xuống dưới 36. + +### 3. Người trên 60 tuổi, đã từng chảy máu dạ dày, đang dùng thuốc chống đông máu hoặc thuốc nội tiết thì hỏi bác sĩ trước khi uống thuốc giảm đau như ibuprofen + +- Chi phí: Không tốn tiền. Hỏi bác sĩ hoặc dược sĩ một lần. +- Nói thẳng: Nhóm thuốc giảm đau như ibuprofen, naproxen, aspirin làm hại dạ dày. Trong các thử nghiệm liều cao, người uống ibuprofen có nguy cơ xuất huyết, thủng dạ dày cao khoảng 4 lần người không uống. Người lớn tuổi, đã từng chảy máu dạ dày, đang uống thuốc chống đông máu hoặc thuốc nội tiết, uống rượu mỗi ngày thì nguy cơ cao hơn. +- Lợi ích: Nhóm thuốc này gọi là thuốc chống viêm không steroid, thường gặp có ibuprofen, naproxen, diclofenac. Một phân tích tổng hợp gộp 280 thử nghiệm ngẫu nhiên của nhóm thuốc này so với giả dược, tổng cộng 124513 người. So với giả dược, nguy cơ thủng, tắc nghẽn hoặc xuất huyết tiêu hoá trên: ibuprofen RR 3.97 (95% CI 2.22–7.10, khoảng 4 lần), naproxen 4.22 (2.71–6.56), diclofenac 1.89 (1.16–3.09). Tất cả các thuốc trong nhóm này làm nguy cơ suy tim mạch tăng khoảng gấp đôi. Quy định Hoa Kỳ yêu cầu thuốc không cần đơn thuộc nhóm này phải ghi cảnh báo xuất huyết dạ dày. Cảnh báo liệt kê sáu nhóm người có nguy cơ cao hơn: trên 60 tuổi, đã từng loét dạ dày hoặc chảy máu, đang dùng thuốc chống đông máu hoặc thuốc nội tiết, đang dùng kèm thuốc khác thuộc nhóm này, trong thời gian uống thuốc mỗi ngày uống trên 3 cốc rượu, uống nhiều hơn hoặc lâu hơn hướng dẫn. +- Mức bằng chứng: A +- Nguồn: Coxib and traditional NSAID Trialists' (CNT) Collaboration; Bhala N, Emberson J, Merhi A, et al. (2013). Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. Lancet, 382(9894):769-779. ；21 CFR 201.326(a)(2)(iii)(A) Stomach bleeding warning. +- Ghi chú: Kết luận của bài báo nói về liều cao. Rủi ro khi thỉnh thoảng uống một hai ngày theo đúng liều trong tờ hướng dẫn thì bài này không đưa riêng con số. Rất nhiều thuốc cảm cúm phối hợp cũng có thành phần thuộc nhóm này, uống chung với thuốc giảm đau là bị trùng, hãy xem bảng thành phần trước. Bác sĩ bảo bạn uống aspirin liều nhỏ lâu dài để phòng huyết khối thì đừng tự ngừng, cũng đừng tự thêm một loại thuốc giảm đau nữa. + +### 4. Trẻ dưới 2 tuổi bị cảm, đừng tự cho uống thuốc cảm cúm phối hợp và thuốc ho + +- Chi phí: Không tốn tiền, lại còn tiết kiệm được tiền mua thuốc. +- Nói thẳng: Cục Quản lý Dược phẩm nhà nước yêu cầu 14 loại thuốc cảm cúm phối hợp dành cho trẻ phải có cảnh báo: không nên để bố mẹ tự cho trẻ dưới 2 tuổi dùng, phải dùng dưới sự hướng dẫn của bác sĩ hoặc dược sĩ. Trong các thử nghiệm làm trên trẻ, thuốc ho và thuốc chống dị ứng đều không tốt hơn giả dược. +- Lợi ích: Năm 2021, thông báo số 57 của Cục Quản lý Dược phẩm nhà nước sửa đổi tờ hướng dẫn của 14 chế phẩm, gồm dung dịch uống paracetamol và pseudoephedrine, hạt paracetamol và chlorpheniramine dành cho trẻ, hạt paracetamol và dextromethorphan dành cho trẻ. Cảnh báo thêm vào là「不建议家长或监护人自行给2岁以下婴幼儿使用本品，应在医师或药师的指导下使用」— không nên bố mẹ hoặc người giám hộ tự cho trẻ sơ sinh và trẻ nhỏ dưới 2 tuổi dùng sản phẩm này, phải dùng dưới sự hướng dẫn của bác sĩ hoặc dược sĩ. Mục lưu ý thêm「应严格按照药品说明书用法用量使用，避免用药过量」— phải dùng đúng liều theo tờ hướng dẫn, tránh dùng quá liều. Cũng thêm「应避免合并使用含有相同或相似活性成份的抗感冒药」— nên tránh dùng kèm thuốc cảm cúm có cùng hoặc gần giống thành phần hoạt chất. Một bài tổng hợp Cochrane đưa vào 29 thử nghiệm đối chứng ngẫu nhiên, trong đó 10 thử nghiệm làm trên trẻ. Trong các thử nghiệm trên trẻ, không nhóm thuốc nào tốt hơn giả dược: thuốc ho, thuốc kháng histamine (thành phần chống dị ứng), thuốc kháng histamine phối hợp chất làm thông mũi, thuốc ho phối hợp thuốc giãn phế quản. 21 nghiên cứu có báo cáo tác dụng phụ, chế phẩm chứa thuốc kháng histamine và dextromethorphan thì tác dụng phụ nhiều hơn. +- Mức bằng chứng: B +- Nguồn: Cục Quản lý Dược phẩm nhà nước (国家药监局) (2021). Thông báo về việc sửa đổi tờ hướng dẫn của 14 chế phẩm như dung dịch uống paracetamol và pseudoephedrine (2021年第57号) và phụ lục. ；Smith SM, Schroeder K, Fahey T (2014). Over-the-counter (OTC) medications for acute cough in children and adults in community settings. Cochrane Database of Systematic Reviews, (11):CD001831. +- Ghi chú: Trong 14 loại này, nhiều tên có chữ「氨酚」, đều chứa paracetamol, uống chung với thuốc hạ sốt là bị trùng, xem mục 1 của phần này (paracetamol). Trong bài tổng hợp đó có một thử nghiệm thấy mật ong tốt hơn giả dược, nhưng trẻ dưới 1 tuổi thì không cho uống mật ong, xem mục 6 ở phần 20 (mật ong). Để ở mức B là vì trong mỗi nhóm thuốc ở bài tổng hợp đều số thử nghiệm ít, tác giả không gộp được thành con số. Đặt lợi ích là「vừa」là theo phán đoán: nhóm thuốc này chưa có lợi ích nào được chứng minh ở trẻ, rủi ro chính là quá liều và trùng thành phần. + +### 5. Sau tuần thứ 20 mang thai, đừng tự uống thuốc giảm đau như ibuprofen + +- Chi phí: Không tốn tiền. +- Nói thẳng: Sau tuần thứ 20 mang thai, phần lớn nước ối do thận của thai nhi tạo ra. Lúc đó uống thuốc nhóm ibuprofen có thể làm hại thận thai nhi, làm nước ối ít đi. Ở Hoa Kỳ đã có báo cáo trẻ sơ sinh chết vì nguyên nhân này. Liều nhỏ aspirin do bác sĩ kê không nằm trong phạm vi này. +- Lợi ích: Ngày 10 tháng 10 năm 2020, Cục Quản lý Thực phẩm và Dược phẩm Hoa Kỳ (FDA) ban hành cảnh báo an toàn. Dùng thuốc chống viêm không steroid từ khoảng tuần thứ 20 mang thai trở đi có thể gây vấn đề về thận ở thai nhi, dẫn tới nước ối ít. Cảnh báo áp dụng cho cả thuốc có đơn lẫn thuốc không cần đơn, gồm aspirin, ibuprofen, naproxen, diclofenac, celecoxib. FDA rà soát báo cáo biến chứng đến tháng 7 năm 2017, tìm được 35 ca nước ối ít hoặc suy thận ở trẻ sơ sinh, toàn bộ đều là kết cục nặng. Trong đó 5 ca trẻ sơ sinh chết, đều kèm suy thận ở trẻ sơ sinh. Trong tài liệu, phần lớn các ca nước ối ít hồi phục trong vòng 72 giờ đến 6 ngày sau khi ngừng thuốc. FDA khuyên phụ nữ mang thai từ tuần thứ 20 trở đi không dùng nhóm thuốc này, trừ khi bác sĩ khuyên dùng rõ ràng. Liều nhỏ 81 mg aspirin do bác sĩ kê để điều trị vấn đề riêng trong thai kỳ thì không chịu hạn chế này. Khi cần giảm đau hạ sốt, FDA nhắc có thể dùng paracetamol hoặc thuốc khác, hãy hỏi dược sĩ hoặc bác sĩ trước. +- Mức bằng chứng: B +- Nguồn: U.S. Food and Drug Administration (2020-10-15). FDA recommends avoiding use of NSAIDs in pregnancy at 20 weeks or later because they can result in low amniotic fluid. Drug Safety Communication. +- Ghi chú: FDA nhắc, rất nhiều thuốc không cần đơn trị cảm cúm, cúm, mất ngủ cũng chứa các thành phần thuộc nhóm này, phải xem bảng thành phần. Nhãn thuốc không cần đơn ở Hoa Kỳ trước đây chỉ cảnh báo ba tháng cuối thai kỳ, lần này kéo mốc lên khoảng tuần thứ 20. Liều nhỏ aspirin phòng sản giật thai kỳ, xem mục 5 ở phần 27 (liều nhỏ aspirin). Để ở mức B là vì số liệu đến từ báo cáo biến chứng, không biết có bao nhiêu người uống thuốc mà không sao. Đặt lợi ích là「vừa」là theo phán đoán: kết cục thì nặng, nhưng xảy ra ít. + +### 6. Tự mua omeprazole khi bị trào ngược và nóng rát dạ dày, nhiều nhất 7 ngày; khó nuốt, nôn ra máu, phân đen thì đi thẳng bệnh viện + +- Chi phí: Không tốn tiền. Uống đúng số ngày ghi trên tờ hướng dẫn. +- Nói thẳng: Từ năm 2022, omeprazol dạng viên nén bao phim chống mua không cần đơn thuốc nữa, nhưng tờ hướng dẫn ghi rõ tự uống không quá 7 ngày, trong vòng hai tháng đừng uống lại. Khi khó nuốt hoặc đau khi nuốt, nôn ra máu, đi ngoài ra máu hoặc phân đen thì đừng tự uống, đây có thể là dấu hiệu của tình trạng nghiêm trọng. Thuốc này còn có thể che mất triệu chứng ung thư dạ dày, làm chậm chẩn đoán. +- Lợi ích: Năm 2022, thông báo số 68 của Cục Quản lý Dược phẩm nhà nước chuyển omeprazol dạng viên nén bao phim chống từ thuốc có đơn sang thuốc không cần đơn, đồng thời ban hành tờ hướng dẫn mẫu. Chỉ định ghi trong tờ mẫu là「胃酸过多引起的烧心和反酸症状的短期缓解」— làm dịu tạm thời triệu chứng nóng rát và trào ngược do dạ dày tăng acid. Mục lưu ý ghi rõ: không dùng quá 7 ngày, triệu chứng không giảm thì phải hỏi bác sĩ hoặc dược sĩ. Trong vòng hai tháng không được uống lại, triệu chứng tái diễn thì phải đi khám ngay. Khó nuốt hoặc đau khi nuốt, nôn ra máu, đi ngoài ra máu hoặc phân đen thì không dùng, đây có thể là dấu hiệu của tình trạng nghiêm trọng. Khi xuất hiện các triệu chứng báo động như sụt cân rõ rệt, nôn lặp, khó nuốt, nôn ra máu hoặc phân đen thì phải loại trừ bệnh ác tính trước, vì việc điều trị có thể che mất triệu chứng và làm chậm chẩn đoán. Trên 55 tuổi mà có triệu chứng mới hoặc triệu chứng thay đổi thì phải hỏi bác sĩ. Người đang dùng clopidogrel thì tránh uống omeprazol cùng lúc. +- Mức bằng chứng: B +- Nguồn: Cục Quản lý Dược phẩm nhà nước (国家药监局) (2022). Thông báo về việc chuyển omeprazol dạng viên nén bao phim chống từ thuốc có đơn sang thuốc không cần đơn (2022年第68号) và phụ lục 2《非处方药说明书范本》. +- Ghi chú: Clopidogrel là thuốc chống đông máu mà người đã đặt stent, từng bị nhồi máu cơ tim thường phải dùng. Omeprazol làm nó yếu đi, người đang uống thì hỏi bác sĩ trước. Để ở mức B là vì đây là những gì tờ hướng dẫn quy định, không có nghiên cứu nào tính được làm theo sẽ bớt được bao nhiêu sự cố. Đặt lợi ích là「vừa」là theo phán đoán: nó ngăn việc kéo dài xuất huyết dạ dày, ung thư dạ dày, nhưng phần lớn chứng trào ngược, nóng rát dạ dày không nghiêm trọng. + +### 7. Cảm cúm thường thì đừng đi xin bác sĩ kê kháng sinh + +- Chi phí: Không tốn tiền, lại còn tiết kiệm được tiền thuốc. +- Nói thẳng: Cảm cúm mà uống kháng sinh thì khỏi không nhanh hơn là không uống. Người lớn uống kháng sinh có nguy cơ tác dụng phụ cao hơn hơn hai lần người không uống. Với viêm mũi mủ ra dịch mà kéo dài chưa đến 10 ngày, uống kháng sinh cũng không có lợi ích chắc chắn. +- Lợi ích: Một bài tổng hợp Cochrane gộp 6 thử nghiệm ngẫu nhiên, 1147 người. So với giả dược, tỷ lệ không khỏi hoặc triệu chứng kéo dài ở người cảm cúm uống kháng sinh không có khác biệt rõ rệt (RR 0.83, 95% CI 0.60–1.14, khác biệt không có ý nghĩa). Nguy cơ tác dụng phụ gấp 1.8 lần nhóm giả dược (95% CI 1.01–3.21). Trong đó người lớn gấp 2.62 lần (1.32–5.18), trẻ không tăng (0.91, 0.51–1.63). Với viêm mũi mủ cấp ra dịch mủ dưới 10 ngày, uống kháng sinh cũng không có lợi ích rõ ràng (RR 0.73, 0.47–1.13). Tác dụng phụ thì nhiều hơn (RR 1.46, 1.10–1.94, cao hơn khoảng 46%). Kết luận của tác giả là: cả hai trường hợp đều không nên dùng kháng sinh thường quy. +- Mức bằng chứng: A +- Nguồn: Kenealy T, Arroll B (2025). Antibiotics for the common cold and acute purulent rhinitis. Cochrane Database of Systematic Reviews, 11:CD000247. +- Ghi chú: Phiên bản này phát hành lại năm 2025, phần tóm tắt ghi rõ đã tìm tài liệu đến năm 2013. Kháng sinh ở nước ta là thuốc có đơn, hiệu thuốc phải có đơn mới bán, kháng sinh còn dư trong nhà cũng đừng tự lấy uống. Đặt lợi ích là「vừa」là theo phán đoán: nó ngăn chặn tác dụng phụ, không phải tử vong. + +### 8. Khi tiêu chảy thì uống muối bù điện giải uống để bù nước trước, trẻ dưới 5 tuổi không cho thuốc cầm tiêu chảy + +- Chi phí: Một hộp muối bù điện giải uống vài nghìn đến mười mấy nghìn tệ, nên có sẵn một hộp trong nhà. +- Nói thẳng: Khi tiêu chảy, điều quan trọng nhất là bù nước, không phải cầm tiêu chảy. Pha đúng theo tờ hướng dẫn rồi uống muối bù điện giải uống, đừng thay bằng nước trái cây hay nước giải khát, quá ngọt lại càng khiến người ta mất nước hơn. Trẻ dưới 5 tuổi không cho thuốc cầm tiêu chảy như loperamid, nặng thì gây tắc ruột, có thể tử vong. +- Lợi ích: Sổ tay hướng dẫn điều trị tiêu chảy của Tổ chức Y tế Thế giới viết rằng thuốc「chống tiêu chảy」và thuốc chống nôn không có lợi ích thực tế với tiêu chảy cấp hoặc kéo dài ở trẻ em. Chúng không ngăn được mất nước, cũng không cải thiện dinh dưỡng. Một số có tác dụng phụ nguy hiểm, đôi khi tử vong, tuyệt đối không cho trẻ dưới 5 tuổi dùng. Nhóm thuốc làm chậm nhu động ruột như loperamid có thể gây tắc ruột do liệt, rất nặng và có thể tử vong, lại còn kéo dài thời gian nhiễm trùng. Sổ tay còn viết, đồ uống nhiều đường sẽ hút nước trong cơ thể vào ruột, gây mất nước tăng natri, ví dụ như nước giải khát và nước trái cây đóng bình bán sẵn. Sổ tay của Tổ chức Y tế Thế giới khuyến nghị công thức nhẹ thẩm thấu, tổng độ thẩm thấu 245 mOsm/L. So với công thức cũ, nó làm giảm 33% số lần truyền tĩnh mạch ngoài dự kiến. Một bài tổng hợp Cochrane gộp 8 thử nghiệm, kết quả tương tự (OR 0.59, 95% CI 0.45–0.79, ít hơn khoảng bốn phần mười). +- Mức bằng chứng: A +- Nguồn: World Health Organization (2005). The treatment of diarrhoea: a manual for physicians and other senior health workers, 4th revision. ；Hahn S, Kim S, Garner P (2002). Reduced osmolarity oral rehydration solution for treating dehydration caused by acute diarrhoea in children. Cochrane Database of Systematic Reviews, (1):CD002847. +- Ghi chú: Khi mua thì xem độ thẩm thấu ghi trên tờ hướng dẫn, chọn công thức nhẹ thẩm thấu. Sổ tay viết rằng, trẻ đi ngoài kèm máu phần lớn là kiết lỵ, phải đi khám bác sĩ và dùng kháng sinh, không thể chỉ bù nước ở nhà. Đặt lợi ích là「vừa」là theo phán đoán: con số trong tay so hai loại muối bù nước với nhau, không phải có uống với không uống. + +### 9. Thuốc giảm đau trong một tháng đừng uống quá 15 ngày, thuốc giảm đau phối hợp đừng quá 10 ngày, không thì đau đầu có thể do thuốc gây ra + +- Chi phí: Không tốn tiền. Phải ghi nhớ số ngày đã uống thuốc, rồi nhịn không uống bừa. +- Nói thẳng: Đau đầu tái diễn mà cứ ngày nào cũng uống thuốc giảm đau thì có thể càng uống càng đau. Phân loại quốc tế gọi đó là đau đầu do lạm dụng thuốc. Thuốc giảm đau thường uống từ 15 ngày trở lên trong một tháng, thuốc giảm đau phối hợp từ 10 ngày trở lên, kéo dài quá 3 tháng thì đã đủ để chẩn đoán. Phần lớn người bệnh cải thiện sau khi ngừng thuốc dùng quá. +- Lợi ích: Phân loại quốc tế của Hiệp hội Đau nửa đầu Quốc tế đưa ra tiêu chuẩn chẩn đoán trong Phân loại quốc tế các bệnh đau nửa đầu, xuất bản lần thứ 3. Người vốn đã bị đau nửa đầu, mỗi tháng đau từ 15 ngày trở lên, đồng thời dùng quá liều thuốc trị đau nửa đầu một cách đều đặn quá 3 tháng. Với nhóm thuốc giảm đau không phải opioid như paracetamol, aspirin, ibuprofen, ngưỡng quá liều là từ 15 ngày mỗi tháng trở lên. Với thuốc giảm đau phối hợp, ngưỡng là từ 10 ngày mỗi tháng trở lên. Thuốc giảm đau phối hợp là loại gộp từ hai thành phần có tác dụng giảm đau trở lên, hoặc có thêm thành phần hỗ trợ như caffein, ví dụ paracetamol phối hợp codein. Nhiều loại thuốc giảm đau không phải opioid dùng luân phiên thì số ngày cộng lại. Tài liệu chú giải phân loại viết rằng, bằng chứng dịch tễ học ở nhiều nước cho thấy trong nhóm người đau đầu từ 15 ngày mỗi tháng trở lên, hơn một nửa thuộc loại đau đầu này. Phần lớn người bệnh cải thiện sau khi ngừng thuốc quá liều, và việc điều trị phòng đau nửa đầu cũng hiệu quả hơn. +- Mức bằng chứng: B +- Nguồn: Headache Classification Committee of the International Headache Society (2018). The International Classification of Headache Disorders, 3rd edition. Cephalalgia, 38(1):1-211. ；Bản trực tuyến ICHD-3, mục 8.2 Medication-overuse headache (gồm 8.2.3、8.2.5). +- Ghi chú: Viên giảm đau mà tên có chữ「咖」thường đã thêm caffein, tính theo thuốc phối hợp thì 10 ngày. Trong một tháng số ngày đau đầu càng tăng dần thì đi khám khoa thần kinh, đừng chỉ tăng liều thuốc. Để ở mức B là vì tiêu chuẩn phân loại đưa ra là ngưỡng chẩn đoán, không phải thử nghiệm đối chứng. Đặt lợi ích là「vừa」là theo phán đoán: nó đổi lại là những ngày không đau đầu, không dính gì đến tử vong. From e07a7cb8e8da55d2e43b6732cb19bc8b73d7707c Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 12:40:40 +0700 Subject: [PATCH 15/25] =?UTF-8?q?=E8=B6=8A=E5=8D=97=E6=96=87=E7=AC=AC=2031?= =?UTF-8?q?=20=E8=8A=82=EF=BC=88=E5=8D=81=E5=85=AB=E5=B2=81=E4=B9=8B?= =?UTF-8?q?=E5=90=8E=E6=9C=89=E5=93=AA=E5=87=A0=E6=9D=A1=E8=B7=AF=EF=BC=89?= =?UTF-8?q?=E7=AD=89=E7=BF=BB=E5=AE=8C=E5=B9=B6=E5=90=88=E5=85=A5=EF=BC=9A?= =?UTF-8?q?VI=2027/34?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 两处判据的修法，都在 structure.mjs： · 争议标记那条正则补 trimStart。越南文「- Ghi chú: Tranh cãi.」冒号后有空格， ^Tranh cãi 匹配不上，一条真标了争议的条目被报成「原文标了争议，译文没标」。 · 中文数字法条编号纳入允许名单。「第一千零四十五条」在原文里是汉字，numbersOf 读不出来，越南文按 BRIEF 写成「Điều 1045」就成了「多了原文没有的数字」。 英文因为把编号拼成 one thousand and forty-five 反而躲过去了，纯属侥幸—— 现在两种写法都算原文就有的那个数。并排写法「第一千零五十二、一千零五十三条」 也认，「第」只出现在头一个。 另外把 vi 15、16 两节切块时漏掉的节首补进 part00（含回目录那行），把 18、31 的 导读翻好。 --- tools/i18n/structure.mjs | 51 ++- ...15\350\246\201\346\227\251\346\255\273.md" | 4 +- ...01\346\205\242\346\205\242\346\255\273.md" | 4 +- ...52\350\264\271\347\262\276\345\212\233.md" | 4 +- ...52\350\264\271\346\227\266\351\227\264.md" | 4 +- ...01\346\265\252\350\264\271\351\222\261.md" | 4 +- ...61\346\220\255\350\277\233\345\216\273.md" | 395 ++++++++++++++++++ ...22\344\270\215\345\210\222\347\256\227.md" | 183 ++++++++ ...51\347\232\204\347\272\242\347\272\277.md" | 192 +++++---- ...41\346\201\257\345\256\211\345\205\250.md" | 2 +- ...77\344\270\216\344\271\260\346\210\277.md" | 6 +- ...16\346\200\216\344\271\210\346\264\273.md" | 6 +- ...14\346\234\211\350\200\201\344\272\272.md" | 2 +- ...22\344\270\215\345\210\222\347\256\227.md" | 46 +- ...14\345\222\214\345\267\245\344\274\244.md" | 156 +++++++ ...20\346\200\216\344\271\210\345\270\246.md" | 4 +- ...03\345\244\226\345\256\211\345\205\250.md" | 4 +- ...16\344\271\210\346\224\276\346\235\276.md" | 4 +- "vi/book/24-\347\234\213\347\227\205.md" | 113 +++++ ...01\345\212\236\344\273\200\344\271\210.md" | 96 +++++ ...53\344\275\223\346\220\236\345\235\217.md" | 7 +- ...16\347\232\204\345\255\251\345\255\220.md" | 7 +- ...52\345\207\240\346\235\241\350\267\257.md" | 150 +++++++ ...16\346\200\216\344\271\210\346\264\273.md" | 94 ++--- ...53\345\220\203\345\207\272\344\272\213.md" | 2 +- 25 files changed, 1335 insertions(+), 205 deletions(-) create mode 100644 "vi/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" create mode 100644 "vi/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" create mode 100644 "vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" create mode 100644 "vi/book/24-\347\234\213\347\227\205.md" create mode 100644 "vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" create mode 100644 "vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" diff --git a/tools/i18n/structure.mjs b/tools/i18n/structure.mjs index 2cc701b4..7f0f83a0 100644 --- a/tools/i18n/structure.mjs +++ b/tools/i18n/structure.mjs @@ -27,6 +27,45 @@ const SCALE_WORDS = [ ]; const SCALE_VALUES = SCALE_WORDS.map(([v]) => String(v)); +// 中文数字转阿拉伯数字：法条编号在原文里是汉字（「第一千零四十五条」），译文按 BRIEF +// 要求写成「Điều 1045」——那 1045 是原文就有的那个数，不是译文自己加的。 +// 不转换的话，越南文这一栏每条法条都报「多了原文没有的数字」， +// 英文因为把编号拼成 "one thousand and forty-five" 反而躲过去了，纯属侥幸。 +const CN_DIGITS = { 零: 0, 〇: 0, 一: 1, 二: 2, 两: 2, 三: 3, 四: 4, 五: 5, 六: 6, 七: 7, 八: 8, 九: 9 }; +const CN_UNITS = { 十: 10, 百: 100, 千: 1000 }; +function cnNumber(s) { + // 「一千零四十五」= 1045：「零」不占位，只表示这一段没读数 + let total = 0, section = 0, cur = 0; + for (const ch of s){ + if (ch in CN_DIGITS){ cur = CN_DIGITS[ch]; continue; } + const u = CN_UNITS[ch]; + if (!u) return null; // 万、亿之类这里不拆（法条编号里不出现） + // 「十五」开头没有「一」，十就是 10 不是 0+10 + section += (cur === 0 && u === 10 ? 1 : cur) * u; + cur = 0; + } + total = section + cur; + return total > 0 ? String(total) : null; +} + +/** 一栏里所有「第<中文数字>条 / 编 / 号 / 款」这类法条编号换算出来的数字。 + * + * 一条法条可以并排写：「第一千零五十二、一千零五十三条规定」——「第」只出现在头一个， + * 后面几个是顿号连着的。所以先把「第…条」整串切出来，再按顿号/逗号逐个读。 */ +function cnNumbersOf(field) { + const out = new Set(); + const D = '[零〇一二三四五六七八九十百千]'; + for (const m of String(field).matchAll(new RegExp(`第${D}{1,8}[条款项号章节][^。；;]{0,40}`, 'g'))){ + for (const p of m[0].slice(1).split(/[、,，]/)){ + const digits = new RegExp(`^${D}{1,8}`).exec(p.trim()); + if (!digits) continue; + const v = cnNumber(digits[0]); + if (v) out.add(v); + } + } + return [...out]; +} + function numbersOf(text) { const s = String(text); const norm = n => n.replace(/[.,](?=\d{3}\b)/g, ''); @@ -149,7 +188,10 @@ function decorate(e, S) { e.refs = refsOf(all, S.xref); e.links = all.match(/https?:\/\//g)?.length ?? 0; e.order = FIELDS.filter(f => e.fields[f] !== undefined); - e.dispute = new RegExp('^' + S.dispute).test(e.fields.note ?? ''); + // trimStart：字段名和内容之间可能有空格（「- Ghi chú: Tranh cãi. …」，越南文那边 + // 就是这样写的），不带这个 trim 的话 ^Tranh cãi 匹配不上，一条争议条目会被报成 + // 「原文标了争议，译文没标」。锚点还是 ^，正文中间提到争议不算。 + e.dispute = new RegExp('^' + S.dispute).test((e.fields.note ?? '').trimStart()); e.todo = S.todo.some(w => all.includes(w) || (e.fields.cost ?? '').includes(w)); } @@ -222,10 +264,13 @@ export const STRUCT = { // 数字：收益、成本、来源三栏逐个对。全书可核对性的底座，一个都不许动。 // 备注是散文，「40 例」这类个数和量级本来就随语言变（「several dozen」）， // 数字不许动的是前三个栏（CLAUDE.md 的原话：收益栏是全书可核对性的底座）。 - // 唯一的例外是量级换算：源文「167 万」在英文里只能写成 1,670,000，那不算改了数字。 + // 两个例外： + // ① 量级换算：源文「167 万」在英文里只能写成 1,670,000，那不算改了数字。 + // ② 中文数字：源文「第一千零四十五条」按 BRIEF 要写成「Điều 1045」， + // 那个 1045 是原文就有的，放进允许名单（见 cnNumber）。 for (const f of ['gain', 'cost', 'src']){ const A = a.numbers[f], B = b.numbers[f]; - const allowed = new Set([...A.ns, ...A.scale]); + const allowed = new Set([...A.ns, ...A.scale, ...cnNumbersOf(a.fields[f] ?? '')]); const miss = [...new Set(A.ns.filter(x => !B.ns.includes(x)))]; const extra = [...new Set(B.ns.filter(x => !allowed.has(x)))]; if (miss.length) out.push(`${tag}：${label(D, f)}栏少了原文的数字 ${miss.join('、')}`); diff --git "a/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" "b/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" index bb64bd25..030f1475 100644 --- "a/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" +++ "b/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 1. 不要早死 +# 1. Đừng chết sớm -这一节讲三件事：车祸、火灾、中毒这类意外和伤害死亡；少数几种证据最硬的疫苗和筛查；身体出了什么信号就该去查一查。收益那一栏的数字，来自随机分组的试验、跟踪一大批人很多年的研究，或者官方统计。成本那一栏是作者按市面价格的粗略估计，只用来互相比较，不要拿去引用。 +Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như tai nạn xe, hỏa hoạn, ngộ độ; vài loại vắc-xin và xét sàng có bằng chứng chắc nhất; và những dấu hiệu cơ thể báo hiệu nên đi khám. Số trong cột Lợi ích lấy từ thử nghiệm chia ngẫu nhiên, từ các nghiên cứu theo dõi một nhóm người trong nhiều năm, hoặc từ thống kê chính thức. Cột Chi phí là ước tính thô của tác giả theo giá thị trường, chỉ để so sánh các cách với nhau, đừng trích dẫn. ### 1. Cài dây đai an toàn, cả hàng trước lẫn hàng sau diff --git "a/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" "b/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" index 7b8d7a84..4ff2f52e 100644 --- "a/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" +++ "b/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 2. 不要慢慢死 +# 2. Đừng chết dần -本节只收两类慢性风险：一类对总死亡率影响大，一类证据硬。条目按性价比从高到低排。有几条标了「随机试验」，意思是把人随机分成两组来比，这种结果最可信。其余数字都来自只跟踪记录、不做分组的研究。这类研究算出的 HR、RR 等风险比里混着两种干扰：这些人本来就不一样，有人是先病了才这么做。所以这些数字只说明方向和大概多大，不代表照做就能降这么多。 +Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ tử vong chung dịch chuyển nhiều, và loại có bằng chứng chắc. Các mục xếp theo hiệu quả chi phí từ cao xuống thấp. Có mấy mục ghi「thử nghiệm ngẫu nhiên」, nghĩa là chia người thành hai nhóm ngẫu nhiên rồi so sánh, kết quả kiểu đó đáng tin nhất. Số còn lại đến từ các nghiên cứu chỉ theo dõi và ghi lại, không chia nhóm. Những tỷ số rủi ro như HR, RR mà các nghiên cứu đó tính ra bị hai loại nhiễu trộn vào: vốn dĩ những người đó đã khác nhau, và có người chỉ bắt đầu làm việc đó sau khi đã mắc bệnh. Vì vậy những con số này chỉ nói hướng và đại khái cỡ bao nhiêu, không có nghĩa làm theo là giảm được đúng bấy nhiêu. ### 1. Bỏ thuốc lá, càng sớm càng tốt diff --git "a/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" "b/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" index ea16f0a0..c8971de1 100644 --- "a/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" +++ "b/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 3. 不要浪费精力 +# 3. Đừng phí sức -本节只算精力和时间账。看的是四样东西：你每天能用的注意力、反应快不快、出错多不多、被打断以后要多久才回得来。所有数字都是直接测出来的，来自实验，或者来自跟踪一群人做的记录（队列研究）。这些数字不换算成寿命，也不换算成死亡率。本节多数条目的证据等级是 B。这个领域本来就这样，不代表建议不成立。第 20 条讲的是和警察、医生、柜员打交道时该有的预期，引的是法规，不是实验，所以等级记 C。第 15 条（把悲观念头当症状看）和第 23 条（把「别人要求我完美」当症状看）引的是死亡率与自杀风险的数字，口径和本节其他条目不一样，不放在一起比较。 +Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng chú ý bạn dùng được mỗi ngày, phản xạ nhanh bao nhiêu, sai sót bao nhiêu, và sau khi bị cắt ngang thì bao lâu mới quay lại được. Mọi con số đều đo trực tiếp, từ thí nghiệm, hoặc từ hồ sơ theo dõi một nhóm người (nghiên cứu mạch). Những số này không quy đổi ra tuổi thọ, cũng không quy đổi ra tỷ lệ tử vong. Phần lớn các mục trong phần này xếp B. Lĩnh vực này vốn vậy, không có nghĩa lời khuyên không đúng. Mục 20 nói về điều nên mong đợi khi làm việc với cảnh sát, bác sĩ, thủ quầy: nó dẫn quy định chứ không phải thí nghiệm, nên xếp C. Mục 15 (coi ý nghĩ「chuyện đó chắc chắn sẽ tệ hơn」là một triệu chứng) và mục 23 (coi ý nghĩ「người khác đòi mình phải hoàn hảo」là một triệu chứng) dẫn số về tỷ lệ tử vong và nguy cơ tự sát, khác đơn vị đo với các mục khác trong phần này, nên không so với nhau. ### 1. Tắt các thông báo không cần thiết, và để điện thoại ngoài tầm nhìn khi làm việc diff --git "a/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" "b/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" index e9c0548e..39302372 100644 --- "a/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" +++ "b/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 4. 不要浪费时间 +# 4. Đừng phí thời gian -本节只算时间账。看的是每天、每周、每年能省回多少小时，或者一件事按时做完的比例能提高多少。这里不把时间换算成寿命，也不换算成钱。涉及钱的地方只给算法，不给结论。本节多数条目的证据是单独一项研究，或者是时间利用调查的总量统计，所以等级以 B、C 为主。 +Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi năm lấy lại được bao nhiêu giờ, hoặc tỷ lệ việc giao đúng hạn có thể tăng thêm bao nhiêu. Ở đây không quy đổi thời gian ra tuổi thọ, cũng không quy đổi ra tiền. Chỗ nào có nhắc tiền thì chỉ đưa cách tính, không đưa kết luận. Phần lớn bằng chứng ở phần này là một nghiên cứu đơn lẻ, hoặc là số tổng từ khảo sát sử dụng thời gian, nên mức xếp chủ yếu là B và C. ### 1. Viết「định làm」thành「mấy giờ, ở đâu, gặp gì thì làm gì」 diff --git "a/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" "b/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" index bc2fe9ff..8cedda52 100644 --- "a/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" +++ "b/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 5. 不要浪费钱 +# 5. Đừng phí tiền -这一节只算钱：能省下多少，每年收益差多少，一笔钱长期平均下来能拿回多少。这些不换算成寿命或时间。涉及投资的条目只讲怎么算账，只用核实过的历史数据，都不构成投资建议。 +Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm lợi ích chênh lệch bao nhiêu, một khoản tiền dài hạn tính ra trung bình lấy lại được bao nhiêu. Không quy đổi những cái đó ra tuổi thọ hay thời gian. Chỗ nào có nhắc tiền chỉ đưa cách tính, không đưa kết luận. ### 1. Tắt hết tự động gia hạn, đến hạn thì tự gia hạn diff --git "a/vi/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" "b/vi/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" new file mode 100644 index 00000000..4c2d537e --- /dev/null +++ "b/vi/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" @@ -0,0 +1,395 @@ +[← Về mục lục](../README.md) + +# 8. Đừng để bản thân rơi vào rủi ro: luật pháp và tài sản + +Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tuổi thọ. Nói về ba việc: pháp luật quy định bạn phải gánh trách nhiệm gì, làm theo thì bồi thường bớt được bao nhiêu, và khi xảy ra chuyện bạn có những quyền gì trong thủ tục. Mọi số điều đều đã đối chiếu với bản chính thức. Cách xếp là: việc mỗi lần xảy ra thiệt hại lớn nhất mà làm lại dễ nhất thì để trước; việc cần chuẩn bị từ trước thì để giữa; việc chỉ có kinh nghiệm, không có văn bản chính thức chống lưng thì để sau. + +### 1. Tai nạn giao thông: dừng xe, cứu người, báo công an, đừng bỏ chạy + +- Chi phí: Không tốn tiền. Dừng xe, bật đèn báo khẩn cấp, cứu người, gọi báo công an, đợi cảnh sát giao thông nhiều nhất tốn một hai tiếng. Tai nạn nhỏ mà không ai bị thương và hai bên cũng không tranh cãi về sự thật thì chụp ảnh trước. Ghi thêm họ tên, số điện thoại, biển số, số giấy phép lái xe, số giấy chứng nhận bảo hiểm và chỗ bị va vào của đối phương. Hai bên ký xong là rời hiện trường được, rồi tự thương lượng. +- Nói thẳng: Bỏ chạy thì tai nạn này mặc định tính là bạn hoàn toàn có trách nhiệm. Giấy phép lái xe bị thu, cả đời không thi lại được. Còn một tầng thực tế hơn: bảo hiểm thiệt hại xe và bảo hiểm trách nhiệm bên thứ ba không bao trả cho việc bỏ trốn, hàng chục vạn tệ bồi thường bạn phải tự lo hết. Ở lại hiện trường cứu người, báo công an, nhiều nhất tốn một hai tiếng. +- Lợi ích: Bên bỏ chạy phải gánh toàn bộ trách nhiệm cho tai nạn này. Chỉ khi bạn chứng minh được bên kia cũng có lỗi thì trách nhiệm mới giảm được một phần. Bỏ chạy còn bị tước giấy phép lái xe, cả đời không thi lại được. Nếu tai nạn này cấu thành tội gây tai nạn giao thông thì bỏ chạy bị phạt 3 đến 7 năm. Nếu vì bỏ chạy mà khiến người chết thì phạt trên 7 năm. Bảo hiểm cũng không lo: trong bảo hiểm xe thương mại, bảo hiểm thiệt hại xe và bảo hiểm trách nhiệm bên thứ ba gặp 「交通肇事逃逸」「故意破坏、伪造现场、毁灭证据」 thì nhất luật không bồi thường, hàng chục vạn tệ bồi thường bạn phải tự bỏ ra hết (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). Pháp luật An toàn giao thông đường bộ, 道路交通安全法（2021 年修订，第七十、一百零一条）. （bản đăng lại của Cục Cảnh sát giao thông Công an Bắc Kinh）；Hội đồng Nhà nước (国务院) (2017). Pháp luật thực thi về An toàn giao thông đường bộ, 道路交通安全法实施条例（国务院令第 405 号，2017 年修订，第八十六、九十二条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第一百三十三条）. ；Hiệp hội Bảo hiểm Trung Quốc (中国保险行业协会) (2020). Điều khoản mẫu bảo hiểm xe thương mại bản 2020, 机动车商业保险示范条款（2020 版）（第九、二十二条）. +- Ghi chú: Có người bị thương thì cứu người trước, rồi mới báo công an. Nếu vì cứu người mà buộc phải di chuyển hiện trường thì đánh dấu lại chỗ cũ của xe và của người.「Va chạm một cái không có cảm giác nên lái đi」về sau cũng có thể bị tính là bỏ trốn, không chắc thì báo công an. Mục bảo hiểm trích điều khoản mẫu, đó là mẫu thống nhất của ngành bảo hiểm, bạn thực sự bồi thường được bao nhiêu thì lấy hợp đồng bảo hiểm trong tay mình làm chuẩn. + +### 2. Phát hiện bị lừa thì gọi ngay 110 hoặc 96110 yêu cầu dừng chi trả, đừng tự mình đi tra xét trước + +- Chi phí: Không tốn tiền. Gọi một cuộc điện thoại, nói rõ thời điểm chuyển khoản, số tiền và tài khoản của đối phương. Giữ lại ảnh chụp màn hình chuyển khoản và lịch sử trò chuyện. +- Nói thẳng: Khoảng thời gian ngắn ngay sau khi tiền vừa chuyển đi là cửa sổ duy nhất. Trong khoảng đó cảnh sát có thể dừng chi trả khẩn cấp và đóng băng nhanh, ngân hàng phải phối hợp. Chờ tiền bị chuyển vòng vòng rồi chia nhỏ thì cơ bản đuổi không về nữa. Nên gọi 110 hoặc 96110 trước, đừng tự mình đi tra xét, đi hỏi trước. Số 96110 gọi đến là cảnh sát đang can bạn, hãy nghe máy. +- Lợi ích: Cơ quan công an có quyền tra cứu tức thì, dừng chi trả khẩn cấp và đóng băng nhanh với số tiền liên quan vụ án. Ngân hàng và tổ chức thanh toán phải phối hợp. Số tiền bị đóng băng thì theo thủ tục trả lại cho bị hại. Một khi tiền đã bị chuyển đi và chia nhỏ thì xác suất đuổi về giảm mạnh, nên dừng chi trả càng sớm càng hiệu quả. 96110 là đường dây cảnh báo lừa đảo thống nhất toàn quốc, cuộc gọi đến là cảnh sát đang can bạn, nhất định phải nghe (toàn quốc, từ tháng 12 năm 2022) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). Pháp luật Chống lừa đảo viễn thông và mạng, 反电信网络诈骗法（第二十、三十四条）. ；Công an tỉnh Phúc Kiến (福建省公安厅) (2023). 96110 来电，请务必接听. （trang của công an cấp tỉnh） +- Ghi chú: Bất kỳ ai tự xưng là công an, viện kiểm sát, tòa án mà bảo bạn chuyển tiền vào 「tài khoản an toàn」 thì đều là lừa đảo. Pháp luật chỉ viết 「建立止付冻结制度」, tức là bảo ngân hàng giữ tiền lại. Đuổi được bao nhiêu còn tùy lúc bạn báo công an thì tiền còn trong tài khoản hay không. Mô tả về Ứng dụng 「国家反诈中心」 và 96110 trích từ trang của Công an tỉnh Phúc Kiến, trang gốc trên website Bộ Công an không mở được. + +### 3. Hãy nhớ quy tắc chống lừa đảo: đừng tin vội cuộc gọi, đừng lộ thông tin, đừng bấm link, chuyển khoản thì kiểm lại nhiều lần, bảy loại lừa đảo phổ biến nhất đều có hình dáng này + +- Chi phí: Không tốn tiền. Hãy nhớ 「ba không một nhiều」, đọc qua bảy loại dưới đây, cài thêm Ứng dụng 「国家反诈中心」, số 96110 gọi đến thì nhất định phải nghe. Bất cứ ai bảo bạn chuyển tiền, chia sẻ màn hình, tải App lạ, bấm link, báo mã xác thực thì cúp máy trước, rồi gọi số điện thoại chính thức để kiểm tra. Khó ở chỗ đối phương thường hối rất gấp, bạn phải nhịn, chịu cúp máy trước. +- Nói thẳng: Bất cứ ai bảo bạn chuyển tiền, chia sẻ màn hình, tải App lạ, bấm link, báo mã xác thực thì cúp máy trước, rồi gọi lại bằng số chính thức để kiểm tra. Chạy đơn hoàn tiền, đầu tư tài chính mạng giả mạo, vay tiền online, giả mạo nhân viên chăm sóc khách hàng, giả mạo công an viện kiểm sát tòa án,杀猪盘, dụ dọc tình dục chạy đơn — bảy loại này phổ biến nhất, tất cả đều muốn bạn chuyển tiền trước. Nhận ra trước thì rẻ hơn nhiều so với việc đi đòi tiền sau. +- Lợi ích: Pháp luật yêu cầu chính phủ và các bộ ngành liên quan nói cho mọi người biết chiêu thức của các loại lừa đảo viễn thông và mạng. Công an cùng bộ phận tài chính, viễn thông và quản lý không gian mạng xây dựng hệ thống cảnh báo, can ngăn, thấy ai sắp bị lừa thì can ngăn kịp thời. 96110 chính là đường dây cảnh báo, can ngăn thống nhất toàn quốc, cuộc gọi đến thì phải nghe. Bộ Công an báo cáo tháng 6 năm 2026: 10 loại lừa đảo xảy ra nhiều nhất chiếm 85% toàn bộ vụ án lừa đảo viễn thông. Chạy đơn hoàn tiền xảy ra nhiều nhất, chiếm 25%. Đầu tư tài chính mạng giả mạo gây thiệt hại lớn nhất, chiếm khoảng 40% tổng thiệt hại. Công an tỉnh Phúc Kiến phân tích đặc điểm nhận diện từng loại như sau. Chạy đơn hoàn tiền: 「người kêu bạn cùng kiếm tiền thì khả năng cao là muốn kiếm tiền của bạn」, không làm hết danh sách nhiệm vụ thì không trả vốn. Đầu tư tài chính giả mạo: trong nhóm có 「水军」 khoe lợi nhuận, trước cho chút tiền hoàn rồi bảo bạn bỏ thêm, rút tiền thì phải 「nộp thuế」. Vay tiền online: 「vay tiền không cần nộp tiền đặt cọc, không cần làm sổ dòng tiền ngân hàng, gặp kiểu này chắc chắn là giả」. Giả mạo nhân viên chăm sóc khách hàng: nói bạn mở thành viên nên sẽ tự động trừ tiền, ảnh hưởng tín dụng, ép bạn chuyển tiền để hủy. Giả mạo công an, viện kiểm sát, tòa án: 「cơ quan công an, viện kiểm sát, tòa án không có所谓『tài khoản an toàn』, bất kỳ『cơ quan công an, viện kiểm sát, tòa án』nào làm vụ, ghi lời khai qua điện thoại, QQ, WeChat, mạng Internet đều là lừa đảo」. 杀猪盘: 「người『tình nhân』chưa từng gặp mặt, chỉ thấy qua ảnh hoặc video thì không được轻信」, mãi không video không gặp mặt nhưng luôn có việc gấp cần tiền. Dụ dọc tình dục chạy đơn: hẹn gặp mặt trước để 「làm nhiệm vụ」, tiền càng nộp càng nhiều. Nhận ra trước thì rẻ hơn nhiều so với việc dừng chi trả sau. Dừng chi trả đuổi được bao nhiêu tùy vào lúc báo công an thì tiền còn trong tài khoản hay không, cách dừng chi trả xem mục 2 của phần này（gọi 110 hoặc 96110 yêu cầu dừng chi trả）（toàn quốc, từ tháng 12 năm 2022） +- Mức bằng chứng: B +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). Pháp luật Chống lừa đảo viễn thông và mạng, 反电信网络诈骗法（第八、二十、三十四条）. ；Bộ Công an (公安部) (2026). Họp báo Bộ Công an: thông báo tình hình, đặc điểm mới của tội phạm lừa đảo viễn thông và mạng hiện nay và những điểm chính trong phòng chống（ngày 12 tháng 6 năm 2026）：「以上这10类案件就占了全部电诈案件的85%。其中，刷单返利诈骗发案最高，占所有电诈案件的25%；虚假网络投资理财诈骗损失最高，约占全部电诈案件损失的40%」. ；Công an tỉnh Phúc Kiến (福建省公安厅) (2023). 96110 来电，请务必接听. （trang của công an cấp tỉnh, có khẩu quyết 「来电不轻信，信息不透露，链接不点击，转账多核实」 và lời cảnh báo 「警惕 192 开头的诈骗电话」）；Công an tỉnh Phúc Kiến (福建省公安厅) (2023 到 2024). Chuyên mục kỹ năng nhận biết lừa đảo：刷单 ；投资理财 ；贷款 ；冒充客服 ；冒充公检法 ；杀猪盘 ；色诱刷单 +- Ghi chú: Tự xưng công an, nhân viên chăm sóc khách hàng, lãnh đạo, quản trị rủi ro của nền tảng mà bảo bạn 「phối hợp điều tra」「hủy thành viên」「chạy đơn hoàn tiền」「thông tin nội bộ đầu tư」 thì đều là một khuôn như nhau. Vay thật không cần nộp tiền đặt cọc, công an viện kiểm sát tòa án không có 「tài khoản an toàn」, nhớ hai câu này là chặn được hơn một nửa. Xếp mức B vì khẩu quyết này chỉ là cách tuyên truyền của phía chính thức, không có con số 「nhớ rồi thì bị lừa ít đi bao nhiêu」. Khẩu quyết và cách phân tích bảy loại lấy từ trang của công an cấp tỉnh. 10 loại lừa đảo xảyra nhiều mà Bộ Công an báo cáo năm 2026 còn bao gồm giả mạo nhân viên chăm sóc khách hàng của sàn thương mại điện tử và giao hàng, tín dụng vay tiền, giao dịch ảo trong game online, kết bạn, hẹn hò online, giả mạo lãnh đạo, người quen và đổi, hoàn vé máy bay, khuôn đều như nhau, cuối cùng đều bắt bạn chuyển tiền. + +### 4. Thấy mặt trong video, nghe giọng trong điện thoại đều không tính là đã kiểm chứng; liên quan đến chuyển khoản thì cúp máy trước, dùng số cũ trong danh bạ của mình gọi lại + +- Chi phí: Không tốn tiền. Gọi thêm một cuộc điện thoại, đợi thêm vài phút. Hàng ngày少 đăng video thấy rõ mặt và đoạn thu âm dài lên tài khoản công khai. Khó ở chỗ đối phương thường hối rất gấp, bạn phải nhịn không chuyển khoản ngay tại chỗ. +- Nói thẳng: Thấy mặt, nghe giọng, đều không tính là kiểm chứng. Liên quan đến chuyển khoản thì cúp máy trước, rồi gọi lại bằng số cũ đã lưu trong danh bạ của mình, hoặc nhờ người quen chung hỏi giúp một câu. Đối phương nói 「tín hiệu kém」「đang họp」, không chịu video trực tiếp thì coi luôn là lừa đảo. +- Lợi ích: Cục An ninh mạng Bộ Công an đưa ra ba cách. Cách thứ nhất là kiểm chứng bằng động tác: nhờ đối phương làm một động tác quay đầu liền mạch, hoặc úp bàn tay che kín mặt rồi nhanh chóng bỏ ra. Văn bản gốc nói 「công nghệ AI hiện tại khi mô phỏng những che phủ khuôn mặt phức tạp và thay đổi ánh sáng như vậy容易出现卡顿、模糊或扭曲」. Cách thứ hai là nghe chi tiết giọng nói: 「giọng do AI tạo đôi khi thiếu nhịp thở tự nhiên, lên xuống cảm xúc, hoặc mang cảm giác cơ khí nhẹ, tiếng ồn nền bất thường」. Cách thứ ba là kiểm chứng đa kênh. Văn bản gốc nói đây là 「chiêu cốt lõi nhất」: 「sau khi cúp máy hãy gọi thẳng vào số điện thoại đối phương thường dùng mà bạn quen biết（đừng gọi lại số cuộc gọi đáng ngờ）, hoặc nhờ người thứ ba mà hai bên cùng quen biết kiểm chứng gián tiếp」. Khẩu quyết là 「đừng轻信 video giọng nói, kiểm chứng bằng động tác thì soát lại nhiều lần」. Về mặt pháp luật, đổi mặt đổi giọng早已 bị cấm用于 lừa đảo. Các dịch vụ sinh khổi mặt người, thay mặt người, giọng nói tổng hợp, mô phỏng giọng nói thì nền tảng phải gắn nhãn nổi bật. Tổ chức, cá nhân nào cũng không được恶意 xóa, sửa,伪造, giấu nhãn này. Nhưng tuân quy là nền tảng, kẻ lừa đảo sẽ không gắn, nên 「không thấy nhãn AI」không thể反过来 thành bằng chứng đối phương là người thật. Cục Cảnh sát hình sự Bộ Công an từng công bố một vụ án. Băng nhóm 14 người bị phát hiện có hơn 5 vạn video mặt người chuyển động tổng hợp, hơn 50 vạn thông tin cá nhân công dân. Cảnh sát còn tịch thu 60 máy chủ đám mây (toàn quốc, nghĩa vụ gắn nhãn có hiệu lực từ ngày 1 tháng 9 năm 2025) +- Mức bằng chứng: B +- Nguồn: Cục An ninh mạng Bộ Công an (公安部网安局) (2025). 眼见≠为实，警惕 AI 新型诈骗！（bản đăng lại của Công an tỉnh Phúc Kiến）；Cục Quản lý không g mạng quốc gia, Bộ Công nghiệp và Công nghệ Thông tin, Bộ Công an (国家互联网信息办公室、工业和信息化部、公安部) (2022). Quy định quản lý hợp thành sâu cho dịch vụ thông tin Internet, 互联网信息服务深度合成管理规定（第六、十六、十七条，2023 年 1 月 10 日施行）. ；Cục Quản lý không g mạng quốc gia, Bộ Công nghiệp và Công nghệ Thông tin, Bộ Công an, Tổng cục Phát thanh Truyền hình Quốc gia (国家互联网信息办公室、工业和信息化部、公安部、国家广播电视总局) (2025). Biện pháp gắn nhãn nội dung sinh, hợp thành bằng trí tuệ nhân tạo, 人工智能生成合成内容标识办法（第十条，2025 年 9 月 1 日施行）. ；Cục Cảnh sát hình sự Bộ Công an, Thời sự央视 (公安部刑侦局、央视新闻) (2026). Công an tìm thấy hơn 5 vạn video mặt người chuyển động tổng hợp, ai đang 「复制」khuôn mặt bạn？（bản đăng lại của Công an tỉnh Phúc Kiến） +- Ghi chú: Khi gọi lại phải dùng số cũ đã lưu trong danh bạ của mình. Đừng gọi lại đúng số hiển thị trên màn hình nhận cuộc gọi, cũng đừng dùng số mới đối phương gửi trong lúc chat. Đối phương lấy cớ 「tín hiệu kém」「đang họp」mà không chịu video trực tiếp, không chịu làm động tác chỉ định thì coi luôn là lừa đảo. Cách kiểm chứng bằng động tác chỉ là phụ, AI càng ngày càng giỏi, chỗ chắc chắn duy nhất là cách kiểm chứng thứ ba. Đã chuyển tiền rồi, xem mục 2 của phần này（gọi ngay 110 yêu cầu dừng chi trả）. Muốn không để khuôn mặt và giọng nói của mình bị thu thập thì xem phần 14. Xếp mức B vì phía chính thức đưa ra cách nhận biết và ví dụ, không có con số 「kiểm chứng theo cách này xong thì tỷ lệ bị lừa giảm bao nhiêu」. Website chính thức của Bộ Công an (mps.gov.cn) lần này vẫn sập toàn bộ bằng mã 521, không mở được; văn bản gốc của Cục An ninh mạng và Cục Cảnh sát hình sự Bộ Công an lấy từ trang đăng lại của Công an tỉnh Phúc Kiến. + +### 5. Bị buộc tội hoặc bị triệu tập thì mời luật sư trước, không thương lượng riêng, không xóa dữ liệu + +- Chi phí: Phí luật sư từ vài nghìn tệ trở lên. Người khó khăn về kinh tế có thể xin luật sư miễn phí từ tổ chức trợ giúp pháp lý. Khi bị hỏi thì chỉ trả lời những câu liên quan đến vụ án, phần còn lại giao cho luật sư. Khó là giữ được bình tĩnh tại chỗ, không vội tự mình giải thích. +- Nói thẳng: Lần đầu bị gọi lên hỏi, bạn đã có thể mời luật sư rồi. Luật sư yêu cầu gặp bạn thì trại giam thu xếp chậm nhất trong 48 tiếng, cuộc nói chuyện không bị nghe. Những câu không liên quan đến vụ án này bạn có thể không trả lời. +- Lợi ích: Từ lần hỏi cung đầu tiên, hoặc từ khi bị áp dụng biện pháp bắt buộc（bị hạn chế tự do cá nhân）, bạn có quyền uỷ quyền cho luật sư. Trong đoạn vụ án còn nằm ở công an để điều tra thì chỉ được uỷ quyền cho luật sư, không được nhờ người ngoài luật sư làm người bào chữa. Luật sư đề nghị gặp thì trại giam thu xếp chậm nhất trong 48 tiếng, cuộc gặp không bị nghe lén. Không được ép bất kỳ ai tự chứng minh mình có tội. Những câu không liên quan đến vụ án này bạn có quyền từ chối trả lời. Triệu tập, bắt đi hỏi cung một lần nhiều nhất 12 tiếng, vụ đặc biệt nghiêm trọng, phức tạp nhiều nhất 24 tiếng. Triệu tập là báo bạn đi hỏi cung, bắt đi hỏi cung là dùng sức ép đưa bạn đi. Không được triệu tập liên tục, biến chỗ ở thành nơi giam giữ. Bịa đặt sự thật mà tố cáo người khác, muốn người đó bị truy cứu trách nhiệm hình sự thì phạt dưới 3 năm, gây hậu quả nghiêm trọng thì phạt 3 đến 10 năm. Bị giữ, bị bắt rồi mà vụ việc bị hủy, không truy tố hay tuyên vô tội thì có thể xin bồi thường nhà nước, mỗi ngày tính theo mức lương bình quân đầu người của người lao động cả nước năm trước (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). Bộ luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第三十四、三十五、三十九、五十二、一百一十九、一百二十条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百四十三条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). Luật Bồi thường nhà nước, 国家赔偿法（2012 年修正，第十七、三十三条）. （bản đăng lại của Cục Thống kê Quốc gia） +- Ghi chú: Đừng lén đi gặp đối phương, cũng đừng đi tìm người nhà của họ. Bất kỳ cuộc 「thương lượng riêng」nào cũng có thể bị hiểu là bạn đã nhận. Lịch sử trò chuyện, lịch sử gọi điện, lịch sử hành trình là vật liệu để bạn tự chứng minh mình trong sạch, xóa đi là tự xóa cả chứng cứ của mình. Nhớ nhầm, việc báo việc sau khi điều tra không thành thì không tính là tội tố cáo vu khống. Mục này áp dụng cho người bị buộc tội, không phân nam nữ. Ba mục triển khai xem mục 33 của phần này（trách nhiệm của bên bịa đặt sự thật）、mục 34 của phần này（không đủ chứng cứ thì phải tuyên vô tội và khiếu nại, xử lại）、mục 35 của phần này（bồi thường nhà nước sau khi bị giam tính theo ngày） + +### 6. Sau khi xảy ra chuyện hãy chủ động nói rõ: đầu thú nhiều nhất giảm được bốn phần mười hình phạt, con đường 「né thời hiệu truy tố」 đã bị chặn + +- Chi phí: Không tốn tiền. Tự mình đến cơ quan công an đầu thú, hoặc ở nguyên chỗ gọi điện báo công an, rồi kể lại chuyện một cách trung thực. Khó là chống đỡ được ý nghĩ 「chờ thêm xem, biết đâu không ai điều tra」. +- Nói thẳng: Đầu thú nhiều nhất giảm được bốn phần mười hình phạt, tình tiết nhẹ hơn thì có thể không phạt. Mới chỉ khai trung thực sau khi người ta tìm đến, nhiều nhất cũng giảm được hai phần mười. Bỏ chạy, không nhận, những thứ này không đổi lại được gì. Trốn vài năm cho vụ án hết thời hiệu cũng không thành: vụ đã khởi kiện rồi thì khoảng thời gian bạn trốn không tính vào thời hiệu truy tố. +- Lợi ích: Điều sáu mươi bảy của Bộ luật Hình sự: sau khi phạm tội mà tự mình đầu thú, khai trung thực tội trạng của mình thì là đầu thú. Người đầu thú có thể được xử nhẹ hơn hoặc giảm hình phạt, trong đó tội nhẹ hơn thì có thể miễn hình phạt. Không đầu thú nhưng khai trung thực tội trạng của mình（thú tội）thì có thể được xử nhẹ hơn; vì khai trung thực mà tránh được hậu quả đặc biệt nghiêm trọng thì có thể giảm hình phạt. Hướng dẫn định lượng hình sự do Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao ban hành đã nêu rõ mức độ. Án cơ sở là mức án tính ra trước theo tình tiết vụ án, các phần trăm bên dưới đều giảm trên mức đó. Người đầu thú có thể giảm án cơ sở tới 40%, tội nhẹ hơn thì giảm trên 40% hoặc miễn hình phạt theo pháp luật. Trong thú tội, khai trung thực tội trạng của mình thì giảm tới 20%, khai trung thực loại tội nặng hơn mà cơ quan tư pháp chưa nắm được thì giảm 10% đến 30%, vì khai trung thực mà tránh được hậu quả đặc biệt nghiêm trọng thì giảm 30% đến 50%. Nhận tội, nhận hình phạt thì giảm tới 30%, đồng thời có tình tiết đầu thú, thú tội quan trọng, nộp lại tài sản phạm tội, bồi thường, lượng thứ thì giảm tới 60%, tội nhẹ hơn thì giảm trên 60% hoặc miễn hình phạt theo pháp luật. Thời hiệu truy tố chia theo mức án tối đa theo pháp luật thành bốn bậc: 5 năm、10 năm、15 năm、20 năm（Điều tám mươi bảy của Bộ luật Hình sự）. Nhưng Điều tám mươi tám của Bộ luật Hình sự ghi rõ, sau khi viện kiểm sát, công an, cơ quan an ninh quốc gia khởi tố điều tra, hoặc sau khi tòa án nhận vụ việc, mà trốn điều tra hoặc trốn xét xử thì không bị giới hạn bởi thời hiệu truy tố (toàn quốc, mức độ định lượng hình sự có hiệu lực từ ngày 1 tháng 7 năm 2021) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大). Bộ luật Hình sự nước Cộng hòa Nhân dân Trung Hoa, bản hợp nhất sau khi sửa đổi lần thứ mười một, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第六十七、八十七、八十八条. ；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). Hướng dẫn định lượng hình sự cho các tội phổ biến（thử điểm）, 关于常见犯罪的量刑指导意见（试行）（法发〔2021〕21 号，2021 年 7 月 1 日施行）三（六）（七）（十四）. Toàn văn được Viện kiểm sát nhân dân thành phố Trương Dịch tỉnh Cam Túc đăng lại. ；Tòa án nhân dân tối cao (最高人民法院) (2010). Một số vấn đề cụ thể về xử lý đầu thú và lập công, 关于处理自首和立功若干具体问题的意见（法发〔2010〕60 号）一. Công báo Tòa án nhân dân tối cao. +- Ghi chú: Người được lợi của mục này chính là bạn. Đầu thú chỉ là tình tiết giảm nhẹ, tội vẫn bị định, giảm bao nhiêu do tòa án quyết định, hướng dẫn định lượng hình sự yêu cầu siết chặt mức giảm nhẹ với tội bạo lực nghiêm trọng và tội ma túy. Tai nạn giao thông có cách tính riêng: ở lại hiện trường, cứu người, báo với cơ quan công an thì tính là tự đầu thú, nhưng vì những việc này vốn là nghĩa vụ pháp lý nên mức giảm nhẹ phải siết chặt; chạy rồi sau đó đầu thú vẫn tính là đầu thú, chỉ là phải lấy khung hình phạt pháp luật nặng hơn làm cơ sở, rồi mới quyết định có giảm không, giảm bao nhiêu, cách làm xem mục 1 của phần này（tai nạn giao thông thì dừng xe, cứu người, báo công an trước）. Mời luật sư trước và khai trung thực không mâu thuẫn, xem mục 5 của phần này（bị buộc tội thì mời luật sư trước）: luật sư giúp bạn đối chiếu xem bản ghi lời khai viết đúng không, không phải giúp bạn bịa ra cách nói. + +### 7. Đã uống rượu thì đừng cầm vô lăng, kể cả xe điện và 「chỉ dời xe một chút」 + +- Chi phí: Người đưa đón thay thế hoặc đi taxi vài chục tệ. Trước khi tụ họp thì thống nhất trước hôm nay ai không uống. Khó là khi bị ép uống rượu ở chỗ đông người thì phản lại. +- Nói thẳng: Uống rượu rồi lái xe, thu giấy phép 6 tháng, phạt một hai nghìn. Vi phạm lần nữa thì giữ cộng thêm tước giấy phép. Say rượu lái xe thì tước thẳng, 5 năm không thi lại được, còn bị truy cứu trách nhiệm hình sự về tội lái xe nguy hiểm, sẽ để lại án tích, ảnh hưởng việc kiểm tra chính trị của bạn và một số nghề nghiệp. Bảo hiểm cũng không lo: uống rượu rồi lái xe thì bảo hiểm xe nhất luật không bồi thường. 「Chỉ dời xe một chút」cũng tính là lái xe. +- Lợi ích: Uống rượu rồi lái xe, tạm giữ giấy phép lái xe 6 tháng, phạt 1000 đến 2000 tệ. Vi phạm lần nữa thì bị giữ dưới 10 ngày và tước giấy phép lái xe. Say rượu lái xe thì tước thẳng giấy phép lái xe, 5 năm không được thi lại, đồng thời truy cứu trách nhiệm hình sự về tội lái xe nguy hiểm, phạt quản thúc và phạt tiền, quản thúc là giam giữ ngắn hạn. Người lái xe kinh doanh thì nặng hơn: uống rượu lái xe bị giữ 15 ngày, phạt 5000 tệ, cấm lái 5 năm; say rượu lái xe cấm lái 10 năm, và cả đời không được lái xe kinh doanh nữa. Uống rượu hoặc say rượu lái xe gây tai nạn nghiêm trọng, cấu thành tội phạm thì cấm lái cả đời. Bảo hiểm xe thương mại nhất luật không bồi thường cho việc lái xe sau khi uống rượu, dùng ma túy (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). Pháp luật An toàn giao thông đường bộ, 道路交通安全法（2021 年修订，第九十一条）. （bản đăng lại của Cục Cảnh sát giao thông Công an Bắc Kinh）；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). Sửa đổi Bộ luật Hình sự（九）, 刑法修正案（九）（第八条，即刑法第一百三十三条之一）. ；Hiệp hội Bảo hiểm Trung Quốc (中国保险行业协会) (2020). Điều khoản mẫu bảo hiểm xe thương mại bản 2020, 机动车商业保险示范条款（2020 版）（第九、二十二条）. +- Ghi chú: Say rượu lái xe là tội hình sự, sẽ để lại án tích, ảnh hưởng việc kiểm tra chính trị của bản thân và một số tư cách nghề nghiệp. 「Uống rượu」và 「say rượu」được kẻ đường theo nồng độ cồn trong máu, ranh giới cụ thể do tiêu chuẩn quốc gia quy định. Phần này chưa đối chiếu con số đó, đừng ước lượng cảm tính xem mình đã tới mức nào. + +### 8. Đừng cho ai mượn thẻ ngân hàng, sim điện thoại hay tài khoản thanh toán, 「cho thuê tài khoản nhận tiền」 không phải việc làm thêm + +- Chi phí: Không tốn tiền. Có người nói「mượn thẻ đi qua tài khoản một chuyến, cho bạn vài trăm tệ」, từ chối là được. +- Nói thẳng: Cho thuê, cho mượn, bán thẻ ngân hàng, sim điện thoại, tài khoản thanh toán cho người khác, bản thân việc này đã vi phạm pháp. Phạt thu nhập bất hợp pháp từ 1 đến 10 lần, không có thu nhập thì phạt tối đa 20 vạn tệ, tình tiết nghiêm trọng thì bị giữ 15 ngày, còn bị ghi vào hồ sơ tín dụng, hạn chế chức năng tài khoản. Nếu biết đối phương mang đi phạm tội mà vẫn giúp thu tiền, chuyển tiền thì là tội giúp sức phạm tội, phạt dưới 3 năm. +- Lợi ích: Cho thuê, cho mượn, mua bán thẻ điện thoại, tài khoản ngân hàng, tài khoản thanh toán, tài khoản Internet, bản thân việc này đã vi phạm pháp. Tịch thu thu nhập bất hợp pháp, rồi phạt từ 1 đến 10 lần thu nhập bất hợp pháp. Không có thu nhập bất hợp pháp thì phạt tối đa 20 vạn tệ. Tình tiết nghiêm trọng thì bị giữ dưới 15 ngày. Còn có thể bị ghi vào hồ sơ tín dụng, hạn chế chức năng tài khoản, ngừng nghiệp vụ ngoài quầy（chỉ còn làm được ở quầy）. Biết người khác dùng mạng để phạm tội mà vẫn cung cấp giúp loại trợ giúp như thanh toán, quyết toán, tình tiết nghiêm trọng thì cấu thành tội giúp sức hoạt động phạm tội thông tin mạng（gọi tắt là帮信罪）, phạt dưới 3 năm và phạt tiền (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). Pháp luật Chống lừa đảo viễn thông và mạng, 反电信网络诈骗法（第三十一、四十四条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). Sửa đổi Bộ luật Hình sự（九）, 刑法修正案（九）（第二十九条，即刑法第二百八十七条之二）. +- Ghi chú: Nói「tôi không biết nó mang thẻ đi làm gì」không chặn được. Khi phá án người ta sẽ suy ngược từ hoàn cảnh bên ngoài xem bạn có biết không: từng nhận tiền cảm ơn, dòng tiền rõ ràng bất thường, đều có thể bị tính là bạn biết. Thẻ bị người khác dùng để nhận tiền赃 thì cho dù cuối cùng bạn không cấu thành tội phạm, tài khoản vẫn bị đóng băng, bản thân vẫn bị xử kỷ luật. + +### 9. Mỗi năm miễn phí kiểm tra báo cáo tín dụng của mình hai lần, xem có khoản vay và thẻ nào không phải mình mở không + +- Chi phí: Không tốn tiền. Kiểm tra trên mạng, mười phút là xong. +- Nói thẳng: Mỗi năm bạn có thể kiểm tra miễn phí báo cáo tín dụng của mình trên mạng hai lần, mười phút là xong. Dùng website chính thức của Trung tâm tín dụng Ngân hàng Nhân dân Trung Hoa hoặc ứng dụng ngân hàng trên điện thoại để kiểm tra, đừng dùng ứng dụng 「kiểm tra tín dụng」của bên thứ ba, loại ứng dụng đó đang thu thập thông tin của bạn. Dấu hiệu xấu lưu 5 năm kể từ ngày việc đó kết thúc, qua rồi thì phải xóa. Có ý kiến khác thì có thể nêu, trong 20 ngày phải trả lời bạn bằng văn bản. +- Lợi ích: Pháp luật hành chính viết「chủ thể thông tin cá nhân có quyền miễn phí hai lần mỗi năm lấy báo cáo tín dụng của mình」, mỗi năm bạn có thể kiểm tra miễn phí bản thân hai lần. Thời hạn lưu trữ thông tin xấu là「kể từ ngày hành vi hoặc sự kiện xấu kết thúc là 5 năm; quá 5 năm thì phải xóa」. Đối với báo cáo có ý kiến khác thì bạn có thể nêu ra, tổ chức tín dụng hoặc cơ quan cung cấp thông tin này phải trả lời bạn bằng văn bản trong 20 ngày kể từ ngày nhận được ý kiến khác +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2013). Điều lệ quản lý ngành tín dụng, 征信业管理条例（国令第 631 号，第十六条、第十七条、第二十五条）. +- Ghi chú: Thẻ căn cước bị người khác chiếm dụng để vay online, mở thẻ, bản thân thường phải đến lúc mua nhà vay vốn bị từ chối mới phát hiện, lúc đó đã kéo dài mấy năm rồi. Cổng kiểm tra là website chính thức của Trung tâm tín dụng Ngân hàng Nhân dân Trung Hoa, hoặc ứng dụng ngân hàng trên điện thoại. Đừng dùng ứng dụng 「kiểm tra tín dụng」của bên thứ ba, loại ứng dụng đó bản thân nó đang thu thập thông tin của bạn + +### 10. Xung đột thì báo công an trước, đừng ra tay, người ra tay trước gần như chắc chắn thiệt + +- Chi phí: Không tốn tiền. Lùi đến chỗ đông người rồi báo công an, để hành động của đối phương nằm trong camera giám sát. Khó là nhịn không cãi lại, không đẩy. +- Nói thẳng: Đánh người, bị giữ 5 đến 10 ngày, phạt năm trăm đến một nghìn. Nhóm đánh, đánh người già, phụ nữ, trẻ em, người khuyết tật, nhiều lần đánh người, bị giữ 10 đến 15 ngày, phạt một hai nghìn. Đánh gây thương tích nhẹ là vụ án hình sự, phạt dưới 3 năm, thương tích nặng phạt 3 đến 10 năm. Hai người đánh nhau không tính là phòng vệ chính đáng, người ra tay trước gần như chắc chắn thiệt. Lùi đến chỗ đông người rồi báo công an, so với việc đánh lại thì hữu dụng hơn. +- Lợi ích: Đánh người hoặc cố ý gây thương tích thì bị giữ 5 đến 10 ngày, đồng thời phạt 500 đến 1000 tệ. Nhóm đánh, đánh người già, phụ nữ, trẻ em, người khuyết tật, nhiều lần đánh người hoặc một lần đánh nhiều người thì bị giữ 10 đến 15 ngày, đồng thời phạt 1000 đến 2000 tệ. Đánh nhau theo nhóm xử lý theo tội gây rối trật tự công cộng, tối đa bị giữ 15 ngày và phạt 2000 tệ, gây rối trật tự công cộng là vô cớ sinh sự, ồn ào gây chuyện. Cố ý gây thương tích thân thể người khác, cấu thành tội phạm thì phạt dưới 3 năm; đánh người gây thương tích nặng thì phạt 3 đến 10 năm. Hai người đánh nhau, pháp luật gọi là ẩu đả lẫn nhau, không tính là phòng vệ chính đáng. Nhưng nếu đối phương ra tay trước, thủ đoạn lại rõ ràng quá mức, hoặc bạn đã cố gắng né mà đối phương vẫn đánh tiếp thì bạn đánh lại thường bị coi là phòng vệ. Đánh lại quá mức mới gọi là phòng vệ vượt quá, phải thỏa mãn đồng thời hai điều: rõ ràng vượt quá giới hạn cần thiết, đồng thời gây thiệt hại lớn. Cho dù thật sự vượt quá thì cũng nên giảm nhẹ hoặc miễn hình phạt (toàn quốc, Pháp luật Xử phạt hành chính về trật tự công cộng có hiệu lực từ ngày 1 tháng 1 năm 2026) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). Pháp luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订，第三十、五十一条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二十、二百三十四条）. ；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院)、Bộ Công an (公安部) (2020). Hướng dẫn về việc áp dụng đúng pháp luật chế độ phòng vệ chính đáng, 关于依法适用正当防卫制度的指导意见（法发〔2020〕31 号，mục 6, 9, 11 và 14）. +- Ghi chú: Sau khi Pháp luật Xử phạt hành chính về trật tự công cộng sửa đổi năm 2025 thì số điều thay đổi hết, đánh người từ Điều bốn mươi ba của pháp cũ chuyển thành Điều năm mươi mốt, tài liệu trích số điều cũ đã lỗi thời.「Đánh gây thương tích nhẹ là phải phạt」thì thương tích nhẹ tính thế nào, do tiêu chuẩn giám định quy định, phần này chưa đối chiếu. Đối phương đã ngã, hoặc đã chạy thì xâm hại coi như dừng, bạn đuổi theo đánh tiếp thì không tính là phòng vệ. + +### 11. Xâm hại không tránh được thì có thể đánh lại, nhưng chỉ đánh đúng người đang ra tay, người ấy dừng thì bạn dừng + +- Chi phí: Phí luật sư từ vài nghìn tệ trở lên, người khó khăn về kinh tế có thể xin trợ giúp pháp lý. Cái giá主要 nằm ở事后: thường bị khởi tố với tội cố ý gây thương tích trước, phải phối hợp điều tra, thậm chí còn bị giam hơn chục ngày. Khó là trong khoảng ngày này giữ cho được bình tĩnh, không乱. +- Nói thẳng: Khi đối phương đang xâm hại bạn hoặc người khác mà bạn ra tay ngăn cản, đánh nó bị thương thì không chịu trách nhiệm hình sự. Khi đối phương đang đả thương, giết người, cướp, hiếp, bắt cóc thì đánh nó bị thương hay đánh chết cũng không tính là vượt quá. Nhưng chỉ được đánh đúng người đang ra tay, người ấy dừng thì bạn dừng. Cuối cùng nếu kết luận đứng về phía bạn, bạn vẫn có thể bị khởi tố trước, bị giam vài ngày trước. +- Lợi ích: Pháp luật quy định thế nào là phòng vệ chính đáng. Quốc gia, lợi ích công cộng, thân thể, tài sản và quyền khác của chính bạn hay người khác đang bị xâm hại trái pháp. Lúc đó ra tay ngăn cản, đánh người đang xâm hại thì thuộc phòng vệ chính đáng, không chịu trách nhiệm hình sự. Cùng điều đó còn viết:「正当防卫明显超过必要限度造成重大损害的，应当负刑事责任，但是应当减轻或者免除处罚」, tức phòng vệ chính đáng rõ ràng vượt quá giới hạn cần thiết mà gây thiệt hại lớn thì vẫn chịu trách nhiệm hình sự, nhưng nên giảm nhẹ hoặc miễn hình phạt. Còn một khoản nữa:「对正在进行行凶、杀人、抢劫、强奸、绑架以及其他严重危及人身安全的暴力犯罪，采取防卫行为，造成不法侵害人伤亡的，不属于防卫过当，不负刑事责任」, tức đối với hành vi bạo lực đang tiến hành mà xâm hại nghiêm trọng đến tính mạng, an toàn thân thể, ra tay phòng vệ mà gây thương vong cho người xâm hại thì không phải phòng vệ vượt quá. Hướng dẫn của Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao và Bộ Công an nói ranh giới kỹ hơn. Muốn kết luận là phòng vệ vượt quá thì phải có đồng thời hai điều「rõ ràng vượt quá giới hạn cần thiết」và「gây thiệt hại lớn」, thiếu một cũng không được.「Gây thiệt hại lớn」nghĩa là đánh người xâm hại thành thương nặng hoặc đánh chết, gây thương tích nhẹ và nhẹ hơn thì không tính. Chỉ khi người xâm hại thật sự đã mất khả năng xâm hại, hoặc thật sự đã bỏ cuộc xâm hại thì mới coi là xâm hại kết thúc. Phòng vệ phải nhằm vào chính người xâm hại trái pháp. Đối với xâm hại trái pháp hết sức nhẹ, khi nhìn ra được chỉ là xâm hại nhẹ mà dùng cách đủ để gây thương nặng hoặc chết người để ngăn cản thì không công nhận là hành động phòng vệ. Chi phí trong quá trình có ví dụ chính thức. Án chỉ đạo số 47, vụ于海明: tối 27 tháng 8 năm 2018, công an khởi tố điều tra「vụ cố ý gây thương tích」. Ngày 1 tháng 9, Công an thành phố Côn Sơn căn cứ khoản ba của Điều hai mươi Bộ luật Hình sự công nhận là phòng vệ chính đáng, hủy vụ việc. Án chỉ đạo số 45, vụ陈某: một sinh viên bị 9 người vây đánh, cầm dao đánh trả khiến 3 người bị thương nặng. Công an khởi tố điều tra với tội danh cố ý gây thương tích, giam giữ hình sự anh ta. Cơ quan kiểm sát công nhận thuộc phòng vệ chính đáng, không phê chuẩn bắt, công an mới thả người. Sau khi bị giam giữ mà công an cho rằng cần bắt thì phải trong 3 ngày đề nghị viện kiểm sát xét duyệt phê chuẩn（gửi viện kiểm sát xem có cho bắt không）. Trường hợp đặc biệt có thể kéo dài 1 đến 4 ngày. Đối tượng tình nghi nghiêm trọng phạm tội lưu động, nhiều lần, theo nhóm thì kéo dài tới 30 ngày. Viện kiểm sát lại ra quyết định trong 7 ngày (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二十条）. ；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院)、Bộ Công an (公安部) (2020). Hướng dẫn về việc áp dụng đúng pháp luật chế độ phòng vệ chính đáng, 关于依法适用正当防卫制度的指导意见（法发〔2020〕31 号，mục 5, 6, 7, 10, 11, 13, 14 và 19）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2018). 第十二批指导性案例（检例第 45 号陈某正当防卫案、检例第 47 号于海明正当防卫案）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). Bộ luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第九十一条）. Toàn văn do Viện kiểm sát Khu tự trị dân tộc Hồi Ninh Hạ đăng lại +- Ghi chú: Việc mặc định vẫn là mục 10 nói về lùi ra báo công an（xung đột thì báo công an trước, đừng ra tay）. Hai người đánh nhau không tính là phòng vệ, đối phương ngã rồi hoặc chạy rồi mà bạn còn đuổi đánh cũng không tính. Chỉ đến lúc không còn đường lui mới xét chuyện đánh lại. Sau đó có ba việc phải làm. Thứ nhất, sau khi báo công an thì đừng rời hiện trường. Thứ hai, giữ ngay camera và nhân chứng, hướng dẫn yêu cầu công an đi lấy录像录音 của hiện trường ngay trong giờ đầu tiên, hỏi nhân chứng ngay trong giờ đầu tiên. Thứ ba, trước khi bị cảnh sát hỏi chính thức thì mời luật sư trước, xem mục 5（bị buộc tội thì mời luật sư trước）. Cuối cùng vụ việc bị hủy, không truy tố hay tuyên vô chất thì những ngày bị giam có thể xin bồi thường nhà nước theo từng ngày, xem mục 35（bồi thường nhà nước sau khi bị giam）. Ra tay phòng vệ cho người khác cũng được pháp luật cho phép. Nhưng tay không đi can vào vụ đánh nhau giữa người lạ thì rủi ro phải tính riêng, xem mục 37 ở phần 13（chứng kiến一群人đánh nhau）. Người được lợi của mục này chủ yếu là chính bạn và người thân. + +### 12. Mắc mứa với ai — bị nợ lương, bị sa thải, bị lừa tiền — thì đi khiếu nại, hoà giải lao động, khởi kiện, đừng đi tính sổ với người ta + +- Chi phí: Không tốn tiền, khiếu nại và hoà giải lao động đều không thu phí. Phải chạy một hai chuyến, chờ vài tháng. Khó là trong vài tháng đó nhịn không ra tay. +- Nói thẳng: Đi con đường theo pháp luật, kết quả xấu nhất là tiền đòi không về. Đi trả thù thì kết quả là tiền vẫn đòi không về, còn phải tự lấy mạng đổi. Vụ ở无锡: 徐某金 không lấy được bằng tốt nghiệp, lại không vừa ý với tiền công thực tập, cầm dao gây án khiến 8 người chết 17 người bị thương,被判死刑 và đã thi hành. Nợ lương thì đi khiếu nại và hoà giải lao động, năm 2024 cả nước nhờ con đường này追回 lương cho 654.000 người. +- Lợi ích: Con đường theo pháp luật có số chính thức. Năm 2024 cả nước thanh tra lao động xử lý 137.000 vụ,追回 lương cho 654.000 người với 8.650.000.000 tệ. Còn có 998 người bị xử tù vì tội không trả thù lao lao động. Nợ lương thì khiếu nại với thanh tra lao động thế nào, xin hoà giải lao động thế nào, xem mục 2 ở phần 7. Bảng giá của con đường trả thù viết ở Điều hai trăm ba mươi hai Bộ luật Hình sự:「故意杀人的，处死刑、无期徒刑或者十年以上有期徒刑；情节较轻的，处三年以上十年以下有期徒刑」. Ví dụ chính thức: 徐某金 thi không đạt, không lấy được bằng tốt nghiệp, lại không vừa ý với tiền công thực tập. Ngày 16 tháng 11 năm 2024, anh ta về Viện Cao đẳng Công nghệ Ngôn tình无锡 cầm dao gây án, khiến 8 người chết, 17 người bị thương. Tòa sơ thẩm xử tội cố ý giết người với hình phạt tử hình, t剥夺政治权利终身, ngày 20 tháng 1 năm 2025 được Tòa án nhân dân tối cao chuẩn准 thi hành. Người anh ta không vừa ý là trường học và khoản tiền công đó, nhưng người bị hại lại là bạn học, khoản tiền công đó anh ta cũng không nhận được một xu (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百三十二条）. ；Tân Hoa xã (新华社) (2024). 江苏宜兴一校园发生持刀伤人事件致 8 死 17 伤. （chuyển bản thông báo của cảnh sát Nghi Hưng, chứa số người thương vong và nguyên văn động cơ）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 无锡校园持刀伤人案罪犯徐加金被执行死刑. ；Bộ Nhân lực和社会保障 (人力资源社会保障部) (2025). 2024 年度人力资源和社会保障事业发展统计公报. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 最高人民法院工作报告（「对驾车冲撞行人的樊维秋、校园持刀行凶的徐加金等判处死刑」）. +- Ghi chú: Thực tập sinh còn đi học thì thường không tính là đã thiết lập quan hệ lao động với đơn vị. Tiền công thực tập đòi không được thì chỉ còn cách nhờ trường đứng ra xử lý, hoặc ra tòa khởi kiện, không đi được con đường hoà giải lao động. Bị nợ lương thì khiếu nại cụ thể thế nào, hoà giải thế nào, xem mục 2 ở phần 7. Đường đỏ khi đi đòi nợ xem mục 15 ở phần 9（đòi nợ thì không bắt giữ người, không giam người）. Đi con đường theo pháp luật cũng có thể đòi không về tiền, vì vụ việc kết thúc không đồng nghĩa tiền đã về tài khoản. Đi trả thù thì tiền vẫn đòi không về, còn phải tự lấy mạng đổi. + +### 13. Dù căm đến đâu cũng đừng ra tay với người không liên quan: lái xe đâm vào đám đông, gây án ở nơi công cộng thì xử tội gây nguy hiểm cho an toàn công cộng bằng phương pháp nguy hiểm, khởi tố từ ba năm, có người chết thì tử hình + +- Chi phí: Không tốn tiền. Khó là trong lúc nóng giận nén được ý nghĩ này lại. +- Nói thẳng: Lái xe đâm vào đám đông, gây án ở nơi công cộng thì xử tội gây nguy hiểm cho an toàn công cộng bằng phương pháp nguy hiểm. Một người cũng không trúng thì vẫn bắt đầu từ ba năm. Chết người thì trên mười năm, chung thân hoặc tử hình. Vụ ở Chu Hải: 樊维秋 không vừa ý việc chia tài sản khi ly hôn, lái xe lao vào đám người đang tập luyện ở Trung tâm Thể thao, khiến 35 người chết 43 người bị thương,被判死刑 và đã thi hành. +- Lợi ích: Điều một trăm mười bốn của Bộ luật Hình sự:「放火、决水、爆炸以及投放毒害性、放射性、传染病病原体等物质或者以其他危险方法危害公共安全，尚未造成严重后果的，处三年以上十年以下有期徒刑」. Một người cũng không trúng thì vẫn bắt đầu từ ba năm. Điều một trăm mười lăm: gây thương nặng, chết người, hoặc làm tài sản công, tư chịu thiệt hại lớn thì phạt trên mười năm, chung thân hoặc tử hình. Ví dụ chính thức: 樊维秋 vì hôn nhân tan vỡ, cuộc sống thất ý, lại không vừa ý kết quả chia tài sản khi ly hôn,「quyết định dùng cách lái xe đâm vào đám đông để phát泄 riêng」. Tối ngày 11 tháng 11 năm 2024, anh ta lái xe lao vào đám người đang tập luyện ở Trung tâm Thể thao thành phố Chu Hải, khiến 35 người chết, 43 người bị thương. Tòa sơ thẩm xử tội gây nguy hiểm cho an toàn công cộng bằng phương pháp nguy hiểm với hình phạt tử hình, t剥夺政治权利终身, ngày 20 tháng 1 năm 2025 được Tòa án nhân dân tối cao chuẩn准 thi hành. Người bị đâm là những người chạy bộ không quen anh ta (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第一百一十四、一百一十五条）. ；Tân Hoa xã (新华社) (2024). 广东珠海驾车撞人事件造成 35 人死亡 43 人受伤. （chuyển bản thông báo của Công an thành phố Chu Hải）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2024). 樊维秋以危险方法危害公共安全案一审宣判，判处死刑，剥夺政治权利终身. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 珠海驾车撞人案罪犯樊维秋被执行死刑. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 法律：治安管理处罚法（2025 年修订，第二十九、五十条）. +- Ghi chú: 「Suy cho cùng tôi cũng không muốn sống nữa」trong định lượng hình sự không phải là mức giảm. Báo cáo công tác năm 2025 của Tòa án nhân dân tối cao nói về loại vụ này là「đối với những kẻ tội ác tày trời thách thức đường lối pháp luật và đạo đức, kiên quyết xử lý nghiêm, nặng và nhanh theo pháp luật」. Chỉ mới nói suông cũng đã vi phạm pháp rồi. Nói suông về việc phóng hỏa, nổ bom, đầu độc这类危害公共安全的事，把公共秩序搅乱了的，拘留 5 đến 10 ngày, còn phạt được dưới 1000 tệ. Viết thư đe dọa, hoặc dùng cách khác đe dọa tính mạng thân thể người khác, cũng vi phạm pháp. Loại này phải bị giữ dưới 5 ngày, hoặc phạt dưới 1000 tệ（Pháp luật Xử phạt hành chính về trật tự công cộng có hiệu lực từ ngày 1 tháng 1 năm 2026）. Đã nghĩ đến bước này rồi thì xem mục 14（xuất hiện ý nghĩ 「kéo một người chết theo」thì xử như cấp cứu）. + +### 14. Nổi lên ý nghĩ 「kéo một người chết theo」「cùng chết」thì xử như cấp cứu: rời hiện trường, đưa chìa khóa xe và dao cho người khác, gọi 12356 + +- Chi phí: Không tốn tiền. Khó là thừa nhận mình lúc này không ổn, cũng khó là đưa đồ vào tay người khác. +- Nói thẳng: Loại ý nghĩ này từ lúc nảy ra đến lúc ra tay thì cửa sổ rất ngắn. Trong 82 người tự tử không thành, 47.6% nói từ lần đầu nảy ý đến khi ra tay không quá 10 phút. Nên việc cần làm không phải 「nghĩ cho thông」, mà là trước hết tách vô lăng, dao, van ga, dây thừng và bản thân ra một bước,熬过 mười phút này. Rồi gọi 12356. +- Lợi ích: Cửa sổ từ lúc nảy ý đến lúc ra tay của loại ý nghĩ này rất ngắn. Trong phỏng vấn 82 người tự tử không thành trong vòng ba ngày nhập viện, 47.6% nói từ lúc ý nghĩ lần đầu xuất hiện đến khi làm không quá 10 phút. Một cuộc tổng quan hệ thống 10 năm（gộp các nghiên cứu trong mười năm lại soát一遍）cho rằng bằng chứng về việc hạn chế phương tiện gây tử vong đang ngày càng mạnh. Sau khi kiểm soát thuốc giảm đau, số ca tự tử liên quan giảm 43%. Sau khi thêm hàng rào ở các điểm nhảy lầu, tự tử giảm 86%（79% đến 91%）. Tách vô lăng, dao, van ga, dây thừng ra một bước, chờ的就是 mười phút này qua đi. 12356 là đường dây trợ giúp tâm lý thống nhất toàn quốc, mỗi ngày nghe máy không dưới 18 tiếng (toàn quốc, từ ngày 1 tháng 5 năm 2025) +- Mức bằng chứng: B +- Nguồn: Deisenhammer EA 等 (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry. ；Zalsman G 等 (2016). Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry. ；Ủy ban Y tế và Sức khỏe Quốc gia (国家卫生健康委) (2024). 关于应用「12356」全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. +- Ghi chú: Lý do xếp mức B là cả hai nghiên cứu này đều lấy từ tài liệu研究自杀，lấy áp vào「suy nghĩ muốn làm hại người khác」thì không có nghiên cứu trực tiếp. Nghiên cứu của Deisenhammer mới chỉ có 82 người, mới làm ở một bệnh viện. Trạng thái không ổn thì đừng lái xe, cũng đừng đi ca trực nào mà phải chịu trách nhiệm an toàn cho người khác, bàn giao ca trước đã. Muốn làm hại chính mình thì xem mục 25 ở phần 1（có ý nghĩ tự tử thì gọi 12356）. Tâm trạng xuống thấp thì làm gì trước, xem mục 19 ở phần 3. + +### 15. Người thân cận nói 「không ai muốn sống yên」, 「tao đưa con đi cùng」 thì đừng coi là lời giận: người thân gần có thể đưa họ đi khám, công an đã nhận báo cáo cũng phải xử lý + +- Chi phí: Phí đăng ký và khám chẩn đoán vài chục đến vài trăm tệ. Đưa người đó đi một chuyến bệnh viện thì mất nửa ngày đến một ngày. Khó là mở lời thuyết phục họ đi cùng. +- Nói thẳng: Người trong nhà nói 「không ai muốn sống yên」 hoặc 「tao đưa con đi cùng」 thì người thân gần có thể đưa thẳng họ tới bệnh viện, bệnh viện không được từ chối chẩn đoán. Đã ra tay đánh nhau rồi, hoặc sắp xảy ra chuyện, thì nhập viện không cần có sự đồng ý của chính người đó. Không phân biệt được là lời giận hay ý thật thì gọi 12356 hỏi trước, hoặc báo 110 cho công an tới nơi. Giấu hộ họ không phải là giúp họ. +- Lợi ích: Điều hai mươi tám khoản hai của Luật Sức khoẻ Tâm thần viết: 「疑似精神障碍患者发生伤害自身、危害他人安全的行为，或者有伤害自身、危害他人安全的危险的，其近亲属、所在单位、当地公安机关应当立即采取措施予以制止，并将其送往医疗机构进行精神障碍诊断」, nghĩa là người nghi mắc rối loạn tâm thần đã làm hại bản thân hoặc đe dọa an toàn của người khác, hoặc có nguy cơ như vậy, thì người thân gần, đơn vị nơi họ làm việc, công an địa phương phải lập tức biện pháp ngăn lại và đưa họ tới cơ sở y tế để chẩn đoán rối loạn tâm thần. Cùng điều đó còn quy định: 「医疗机构接到送诊的疑似精神障碍患者，不得拒绝为其作出诊断」, cơ sở y tế nhận người nghi mắc rối loạn tâm thần được đưa đến thì không được từ chối chẩn đoán. Điều hai mươi chín yêu cầu bệnh viện giữ người được đưa đến loại này lại trong viện, chỉ định ngay bác sĩ tâm thần hành nghề để chẩn đoán và kết luận kịp thời. Điều ba mươi quy định nhập viện điều trị theo nguyên tắc tự nguyện. Nhưng có một ngoại lệ: kết luận chẩn đoán cho thấy người đó bị rối loạn tâm thần nặng và 「已经发生危害他人安全的行为，或者有危害他人安全的危险的」, tức là đã hành động hoặc có nguy cơ đe dọa an toàn của người khác. Loại này phải được nhập viện điều trị, không cần người đó đồng ý. Chỉ dọa gây hại cho người khác thì đã có thể báo 110 trực tiếp. Viết thư đe dọa, hoặc dùng cách khác để đe dọa an toàn cá nhân của người khác thì bị giữ dưới 5 ngày, hoặc phạt dưới 1000 tệ. Dọa sẽ thực hiện tội phạm gây nguy hiểm cho an toàn công cộng như gây hoả hoạn, gây nổ, thả vật nguy hiểm, lại làm xáo trộn trật tự công cộng thì bị giữ 5 đến 10 ngày, đồng thời có thể phạt dưới 1000 tệ (toàn quốc, Pháp luật Xử phạt hành chính về trật tự công cộng có hiệu lực từ ngày 1 tháng 1 năm 2026) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). Pháp luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订，第二十九、五十条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012, sửa đổi 2018). Luật Sức khoẻ Tâm thần, 精神卫生法（第二十八、二十九、三十、三十五条）. ；Cục Điều tra Tai nạn Hàng không (BEA), Pháp (2016). Final Report BEA2015-0125.en: Accident on 24 March 2015 at Prads-Haute-Bléone to the Airbus A320-211 registered D-AIPX operated by Germanwings. +- Ghi chú: Xét theo người được lợi, mục này tính tới vợ chồng, cha mẹ, con cái sống cùng bạn. 「Không ai muốn sống yên」 thường đã bao gồm cả bạn, và pháp luật trao quyền đưa đi khám đúng cho người thân gần, người ngoài muốn quản cũng không được, chỉ có thể báo công an. Đưa đi khám chỉ dùng cho tín hiệu nguy hiểm có thật, lời giận lúc cãi nhau không tính, cũng không dùng nó để nhốt người nhà không nghe lời vào bệnh viện. Cùng bộ luật đó còn viết: sau khi khám lại hoặc giám định mà xác định người đó không bị rối loạn tâm thần nặng thì bệnh viện không được giữ họ nhập viện（Điều ba mươi lăm）. Không phân biệt được thì gọi 12356 hỏi trước, hoặc báo 110 cho công an tới nơi. Kết luận điều tra chuyến bay 9525 của Germanwings là: không có cách nào để chính quyền và chủ lao động ngăn được phi hành đoàn phó bay vào ngày hôm đó. Nguyên nhân là 「无论是他本人，还是任何其他人——医生、同事或者家属——都没有告知他们」. Cả 150 người trên máy đều tử vong, gồm 2 phi hành đoàn, 4 tiếp viên và 144 hành khách. + +### 16. Trên mạng thì đừng chửi người, đừng bịa tin, đừng chia sẻ tin chưa kiểm chứng; bị tấn công trên mạng thì lưu bằng chứng rồi báo công an + +- Chi phí: Không tốn tiền. Trước khi đăng thì nghĩ thêm mười giây. Bị tấn công thì chụp màn hình, quay video, lưu lại đường dẫn và ID tài khoản, trước hết khiếu nại với nền tảng, rồi báo công an. Khó là nhịn không chửi lại lúc bị chửi. +- Nói thẳng: Chửi công khai người khác, hoặc bịa tin vu khống, thì bị giữ dưới 5 ngày, hoặc phạt dưới 1000 tệ, tình tiết nặng hơn thì bị giữ 5 đến 10 ngày. Tình tiết nghiêm trọng thì cấu thành tội xúc phạm, tội vu khống, phạt dưới 3 năm. Chia sẻ lại không miễn trách. Ngược lại, khi bạn bị tấn công trên mạng thì tội xúc phạm, vu khống nguyên tắc phải tự bạn ra tòa kiện, nên việc chụp màn hình, quay video, lưu đường dẫn chỉ làm được nếu tự bạn làm. +- Lợi ích: Công khai xúc phạm người khác, hoặc bịa sự thật để vu khống người khác, thì bị giữ dưới 5 ngày, hoặc phạt dưới 1000 tệ; tình tiết nặng hơn thì bị giữ 5 đến 10 ngày. Nhiều lần gửi tin nhắn xúc phạm đe dọa, quấy rối, theo dõi, chụp lén rồi tung tin tức riêng tư của người khác thì cũng phạt như vậy. Cố ý tung tin đồn, làm xáo trộn trật tự công cộng thì bị giữ 5 đến 10 ngày. Tình tiết nghiêm trọng thì cấu thành tội xúc phạm, tội vu khống, phạt dưới 3 năm. Vu khống qua mạng mà người bị hại khó thu thập chứng cứ thì có thể xin tòa yêu cầu công an phối hợp (toàn quốc, Pháp luật Xử phạt hành chính về trật tự công cộng có hiệu lực từ ngày 1 tháng 1 năm 2026) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). Pháp luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订，第二十九、五十条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百四十六条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). Sửa đổi Bộ luật Hình sự（九）, 刑法修正案（九）（第十六条，增加刑法第二百四十六条第三款）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). 刑法修正案（九）（第二百四十六条新增第三款）. Toàn văn đăng lại bởi Vụ Xử lý thư khiếu nại và thư án dân chủng quốc gia (国家信访局) +- Ghi chú: Với hai tội xúc phạm và vu khống, nguyên tắc là 「có người báo thì mới xử lý」, nghĩa là phải tự bạn ra tòa kiện. Công an thường không tự khởi kiện, chứng cứ chỉ tự bạn tích góp. Nhưng có hai ngoại lệ. Một là trừ 「trường hợp gây tổn hại nghiêm trọng đến trật tự xã hội và lợi ích quốc gia」, trường hợp này công an có thể trực tiếp làm việc đó (khoản hai Điều hai trăm bốn mươi sáu của Bộ luật Hình sự). Hai là bị vu khống trên mạng mà bản thân thu thập chứng cứ thật sự khó, tòa có thể yêu cầu công an phụ giúp điều tra (khoản ba cùng điều đó, thêm vào bằng Sửa đổi thứ chín Bộ luật Hình sự). Đã chia sẻ lại thì vẫn phải chịu trách nhiệm, biết là giả mà vẫn phát ra ngoài thì vẫn tính là 「bịa sự thật」. Bản thân bị tấn công trên mạng rồi xử lý thế nào, xem mục 37 của phần này（bị tấn công trên mạng: chặn bảo vệ và lưu giữ bằng chứng trước, rồi chọn cách）. Số điều của tội xúc phạm, vu khống đổi từ Điều bốn mươi hai của luật cũ sang Điều năm mươi của luật mới. + +### 17. Đọc hết tờ giấy trước khi ký, không ký hộ người khác, không ký trên giấy trắng + +- Chi phí: Không tốn tiền. Dành thêm vài phút, đọc từ đầu đến cuối hợp đồng, giấy mượn, giấy bảo lãnh, hồ sơ nhận việc. Chỗ không hiểu thì hỏi tại chỗ, rồi chụp ảnh lưu lại. Khó là đọc chậm rãi trước mặt đối phương. +- Nói thẳng: Đã ký là có hiệu lực. 「Không đọc kỹ」「Tưởng là thứ khác」 gần như không đứng được trước tòa. Vậy nên dành thêm vài phút đọc từ đầu đến cuối, không hiểu thì hỏi tại chỗ, chụp ảnh lưu lại. Không ký hộ người khác, không ký trên giấy trắng, bản sao thẻ căn cước ghi rõ mục đích sử dụng. Chữ ký điện tử và xác nhận bằng nhận diện khuôn mặt cũng được tính là ký. +- Lợi ích: Hợp đồng được lập theo pháp luật thì được pháp luật bảo vệ, ràng buộc cả hai bên đã ký. Đã ký thì phải thực hiện đầy đủ theo thỏa thuận. 「Không đọc kỹ」「Tưởng là thứ khác」 rất khó đứng được trước tòa. Không ký thì không phải gánh món nợ và trách nhiệm không thuộc về mình (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第四百六十五、五百零九条）. +- Ghi chú: 「Không ký hộ người khác, không ký giấy trắng, ghi rõ mục đích sử dụng trên bản sao thẻ căn cước」 là cách làm theo kinh nghiệm, không có văn bản chính thức riêng nào. Chúng cùng một mục đích: đừng để chữ ký của bạn bị ghép vào một hợp đồng có hiệu lực pháp lý mà bạn lại không tránh được phải thừa nhận. Chữ ký điện tử và xác nhận bằng nhận diện khuôn mặt cũng tính là ký. + +### 18. Cho vay thì viết giấy mượn cho đủ, trước khi bảo lãnh cho người khác hãy nghĩ xem mình có sẵn lòng trả thay không + +- Chi phí: Không tốn tiền. Dùng một tờ giấy ghi rõ người cho vay, người vay, số tiền, lãi suất, thời hạn và cách trả, hai bên cùng ký. Tiền chuyển qua tài khoản để lại dấu vết, không đưa tiền mặt. +- Nói thẳng: Giấy mượn phải ghi đủ: người cho vay, người vay, số tiền, lãi suất, thời hạn, cách trả, hai bên cùng ký. Tiền chuyển qua tài khoản, không đưa tiền mặt, ghi đủ thì mới dễ khởi kiện. Khi bảo lãnh cho người khác thì xem kỹ trong đó có chữ 「liên」 hay không. Bảo lãnh chung nghĩa là phải đợi người có nợ kiện người vay trước, thi hành án không ra tiền, rồi mới đến lượt bạn. Đã viết bảo lãnh liên đới thì người có nợ kiệm thẳng bạn. +- Lợi ích: Hợp đồng vay phải có hình thức văn bản, nội dung thường gồm loại khoản vay, loại tiền, mục đích vay, số tiền, lãi suất, thời hạn và cách trả. Ghi đủ thì mới dễ khởi kiện. Hợp đồng bảo lãnh không ghi rõ bảo lãnh theo cách nào, hoặc ghi không rõ ràng, thì tính là bảo lãnh chung nghĩa. Người bảo lãnh chung nghĩa có thể chờ trước: người đòi nợ phải khởi kiện hoặc trọng tài trước, rồi xin thi hành án tài sản của người vay, thi hành xong vẫn không đủ thì mới đến lượt bạn trả thay. Ngược lại, đã ký 「bảo lãnh liên đới」 thì không có lớp bảo vệ đó, người đòi nợ có thể trực tiếp kiện bạn (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第六百六十八、六百八十一、六百八十六、六百八十七条）. +- Ghi chú: Phần lãi vượt quá 4 lần LPR thì tòa không bảo vệ, xem phần 7. Ghi số thẻ căn cước và ghi 「vay」 trên giấy mượn là cách làm theo kinh nghiệm. Trên giấy bảo lãnh có chữ 「liên」 hay không, quyết định bạn đứng sau người vay hay đứng ngang hàng với họ để bị đòi, phải đọc kỹ trước khi ký. + +### 19. Bảo vệ quyền lợi có thời hạn: thời hiệu khởi kiện dân sự là 3 năm, hoà giải lao động là 1 năm, đối phương chỉ cần một câu 「quá thời hiệu」 là đủ + +- Chi phí: Không tốn tiền. Đòi nợ thì đừng chỉ nghĩ trong đầu, dùng WeChat, tin nhắn, email hoặc thư bảo đảm để lại dấu vết. +- Nói thẳng: Thời hiệu khởi kiện dân sự là 3 năm, tính từ ngày bạn biết quyền lợi bị xâm phạm. Hết thời hiệu thì tòa không tự động xử lý, nhưng đối phương chỉ cần nói một câu 「quá thời hiệu」 ở tòa là xong: tiền vẫn còn, chỉ là đòi lại không được. Một tin nhắn đòi nợ có để lại dấu vết là đủ để 3 năm đó tính lại từ đầu. Tranh chấp lao động là bộ khác: thời hiệu trọng tài là 1 năm. +- Lợi ích: Bộ luật Dân sự quy định 「thời hiệu khởi kiện để yêu cầu tòa bảo vệ quyền lợi dân sự là ba năm」. Điểm tính từ 「ngày người có quyền biết hoặc phải biết quyền lợi dân sự bị xâm phạm và biết bên có nghĩa vụ」, tức là từ ngày bạn biết mình thiệt hại và biết phải tìm ai. Lâu nhất không quá 20 năm kể từ ngày quyền lợi bị xâm phạm. Hết thời hiệu thì 「bên có nghĩa vụ có thể đưa ra phòng vấn không thực hiện nghĩa vụ」. Tòa không tự rà thời hiệu, nhưng đối phương nhắc ở tòa là có hiệu lực, tiền vẫn còn, chỉ là đòi lại không được. Có bốn việc làm gián đoạn thời hiệu, gián đoạn một lần là 3 năm tính lại từ đầu. Một là bạn yêu cầu người nợ của bạn thực hiện nghĩa vụ. Hai là đối phương đồng ý thực hiện. Ba là bạn khởi kiện hoặc xin trọng tài. Bốn là các trường hợp khác có hiệu lực tương đương. Đã thỏa thuận trả góp thì tính từ ngày kỳ trả cuối cùng đến hạn. Có bốn loại yêu cầu không bị thời hiệu chi phối: yêu cầu ngừng xâm phạm, loại trừ trở ngại, loại bỏ nguy cơ; đòi lại bất động sản của mình và tài sản động đã đăng ký; tiền nuôi dưỡng, tiền phụng dưỡng, tiền trợ cấp; các trường hợp khác theo quy định của pháp luật. Tranh chấp lao động là bộ khác: thời hiệu trọng tài 1 năm, tính từ ngày biết hoặc phải biết quyền lợi bị xâm phạm. Nhưng 「khi phát sinh tranh chấp do bị nợ lương trong thời gian quan hệ lao động còn tồn tại thì người lao động xin trọng tài không bị giới hạn bởi thời hiệu trọng tài quy định tại khoản một điều này; nhưng khi quan hệ lao động chấm dứt thì phải đưa ra trong vòng một năm kể từ ngày quan hệ lao động chấm dứt」 (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一百八十八、一百八十九、一百九十二、一百九十三、一百九十五、一百九十六、一百九十七条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). Luật Hoà giải và Trọng tài tranh chấp lao động, 劳动争议调解仲裁法（第二十七条）. （bản đăng lại của Sở Nhân sự và Xã hội thành phố An Khang） +- Ghi chú: Đây có lẽ là việc rẻ nhất cả sách: một tin nhắn đòi nợ có dấu vết là 3 năm tính lại từ đầu; không làm gì thì có thể mất trắng quyền thắng kiện. Thỏa thuận trước 「tôi không viện dẫn thời hiệu」 thì không có giá trị, hai bên thỏa thuận kéo dài hay rút ngắn 3 năm cũng không có giá trị. Ngược lại, quá thời hiệu mà đối phương vẫn đồng ý trả thì không được viện dẫn thời hiệu để từ chối trả. Đã tự nguyện trả rồi thì cũng không được đòi lại. Giấy mượn và bảo lãnh xem mục 18 của phần này（giấy mượn và bảo lãnh）. Giai đoạn bị thi hành án xem mục 20 của phần này（bị khởi kiện, bị thi hành án thì báo cáo tài sản thật）. Tiền làm thêm giờ và tiền nghỉ phép năm chưa dùng thì đi theo bộ hoà giải lao động, xem mục 1 và 2 ở phần 19. +### 20. Bị khởi kiện, bị thi hành án thì báo cáo tài sản thật, trả được bao nhiêu thì trả bấy nhiêu, đừng chuyển nhà và tiền cho người thân hoặc công ty + +- Chi phí: Không tốn tiền. Báo cáo tài sản theo yêu cầu của tòa: hiện có gì, và trong vòng một năm trước ngày nhận thông báo thi hành án có gì. Trả được thì trả trước. Không trả nổi thì thương lượng hoà giải, thương lượng trả góp với người xin thi hành án, người đó chính là bên đòi nợ. +- Nói thẳng: Ghi tên nhà vào công ty, chuyển tiền cho người thân, đúng là những tình huống luật quy định là tội không thi hành án, phạt dưới 3 năm, tình tiết đặc biệt nghiêm trọng phạt 3 đến 7 năm. Người thân và công ty giúp giấu tiền, sang tên tài sản thì bị tính là đồng bọn, tiền vẫn bị truy thu. Không báo cáo hoặc báo cáo tài sản không thật thì cá nhân bị phạt dưới 10 vạn tệ, bị giữ 15 ngày. Thật sự không có tiền thì không cấu thành tội này. +- Lợi ích: Chuyển tài sản bản thân đã là tình huống bị quy thành tội. Ủy ban Thường vụ Quốc hội từng ban hành giải thích về Điều ba trăm mười ba của Bộ luật Trị phạt và Cải cách Dân chủ nước Cộng hòa Nhân dân Trung Hoa. Giải thích đó liệt kê thẳng 「người bị thi hành án giấu, chuyển, cố ý hủy hoại tài sản hoặc chuyển nhượng tài sản vô giá, hoặc chuyển nhượng tài sản với giá thấp bất thường rõ rệt, khiến bản án, quyết định không thi hành được」 vào 「có khả năng thi hành mà cố không thi hành, tình tiết nghiêm trọng」. Giấu tiền, chuyển tiền, cố ý phá nát đồ đạc của mình, cho không, bán rẻ đi, khiến bản án không thi hành được, đều bị tính. Hai tòa cao nhất (Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao) năm 2024 lại bổ sung mười tình huống. Trong đó có cách xử lý quyền lợi tài sản của mình bằng hoà giải giả, chuyển nhượng giả. Cũng có cách mua tài sản của người khác bằng mức giá cao bất thường rõ rệt, bảo lãnh cho nợ của người khác. Tất cả đều là cố ý làm cạn tài sản mà bạn có thể dùng để trả nợ. Tội không thi hành án phạt dưới 3 năm. Tình tiết đặc biệt nghiêm trọng thì phạt 3 đến 7 năm và phạt tiền. Đặc biệt nghiêm trọng là các tình huống: dùng vụ án giả, trọng tài giả, công chứng giả để cản trở thi hành án. Tụ tập đông người xông vào hiện trường thi hành án. Vây khốn, giữ người, đánh người thi hành án. Vì không thi hành án mà người xin thi hành án tự sát, tự làm hại. Người ngoài vụ không phải bị can, biết rõ tình hình mà vẫn thông đồng, giúp giấu tiền, sang tên thì bị trừng phạt như đồng bọn trong tội không thi hành án. Tài sản bị xử lý trái pháp phải được truy thu hoặc buộc hoàn trả. Đã chuyển tài sản sau khi bị can nhận thông báo có thư, trước khi bản án có hiệu lực thì cũng bị tính. Về mặt dân sự, cho không hay chuyển nhượng với giá thấp bất thường rõ rệt thì bên đòi nợ có thể xin tòa hủy giao dịch đó. Không báo cáo hoặc báo cáo tài sản không thật thì tòa có thể phạt tiền, với cá nhân dưới 10 vạn tệ, cũng có thể bị giữ dưới 15 ngày (toàn quốc, giải thích của hai tòa cao nhất có hiệu lực từ tháng 12 năm 2024) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2002). Giải thích về Điều ba trăm mười ba của Bộ luật Trị phạt và Cải cách Dân chủ nước Cộng hòa Nhân dân Trung Hoa, 关于《中华人民共和国刑法》第三百一十三条的解释. ；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2024). Giải thích một số vấn đề về áp dụng pháp luật khi xét các vụ án hình sự không thi hành án, quyết định（法释〔2024〕13 号，第三、四、五、六、八、十一、十二条）. Trang công bố của Tòa án nhân dân tối cao, đã đối chiếu toàn văn với bản đăng lại của Nhà tù thành phố Quảng Châu, tỉnh Quảng Đông. 、；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). 刑法修正案（九）（第三十九条，修改刑法第三百一十三条）. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第五百三十八、五百三十九、五百四十条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 修正). Pháp luật Tố tụng Dân sự, 民事诉讼法（第一百一十四、一百一十八、二百五十二条）. 贵州省市场监管局转载： +- Ghi chú: Câu 「ghi tên nhà vào công ty làm nhà ở cho nhân viên, chuyển hết tiền cho người thân」 đang lưu truyền trên mạng không phải lỗ hổng, mà đúng là trường hợp cho không và chuyển nhượng giả mà giải thích của tòa nêu đích danh. Người thân và công ty giúp bạn thao tác bị tính là đồng bọn, tài sản phải truy thu thì vẫn bị truy thu. Ngược lại, thật sự không có tiền thì không cấu thành tội này. Để biết mình có phải 「có tiền mà không trả」 không, trước hết phải trừ khoản sinh hoạt cần thiết cho bạn và gia đình bạn phải nuôi. Trước khi viện kiệm sát khởi tố, bạn trả lại một phần hoặc toàn bộ mà tình tiết lại nhẹ thì có thể không khởi tố. Trả trước khi có bản án sơ thẩm thì có thể được giảm nhẹ, thậm chí miễn phạt. Bị đưa vào danh sách, bị hạn chế tiêu dùng rồi làm sao, xem mục 21 của phần này（bị hạn chế tiêu dùng hoặc bị đưa vào danh sách không thi hành án）. Thi hành xong thì khôi phục thế nào, xem mục 19 ở phần 7（sau khi lỡ bị đưa vào danh sách thì làm lại thế nào）. + +### 21. Bị hạn chế tiêu dùng hoặc bị đưa vào danh sách không thi hành án, trước hết xem mình bị đưa vào theo điều khoản nào, có thể sửa thì xin sửa + +- Chi phí: Không tốn tiền. Đến tòa thi hành án hỏi rõ mình bị đưa vào theo khoản nào. Chuẩn bị chứng tứ đã trả tiền, đã làm theo bản án（chứng tứ thi hành）, hoặc chuẩn bị tài liệu bảo lãnh. Món chi cần cho sinh hoạt và kinh doanh thì phải xin từng lần một. Khó là phải tự đi nhiều chuyến, xin nhiều lần. +- Nói thẳng: Trước hết làm rõ bạn bị hạn chế tiêu dùng hay bị đưa vào danh sách không thi hành án, đây là hai việc khác nhau. Danh sách không thi hành án chỉ có sáu tình huống được đưa vào, thời hạn 2 năm. Đã thi hành xong cùng sáu tình huống khác, tòa phải xoá trong 3 ngày làm việc. Hạn chi tiêu cấm chín loại tiêu dùng, đi tàu cao tốc tên đầu chữ G cũng tính vào. Phần cần cho sinh hoạt hoặc kinh doanh thì xin từng lần một. +- Lợi ích: Chỉ có sáu tình huống mới bị đưa vào danh sách không thi hành án. Một là có khả năng thi hành mà cố không thi hành. Hai là dùng chứng cứ giả, đe dọa bạo lực để cản trở việc chống thi hành án. Ba là dùng vụ án giả, trọng tài giả, hoặc giấu, chuyển tài sản để né thi hành án. Bốn là vi phạm quy định báo cáo tài sản. Năm là vi phạm lệnh hạn chế tiêu dùng. Sáu là không có lý do chính đáng mà cố không thi hành thoả thuận hoà giải trong thi hành án. Đã cung cấp bảo lãnh đầy đủ và hiệu lực, hoặc tài sản đã bị phong tỏa, giữ, đóng băng đủ để thanh toán thì tòa không được đưa vào theo khoản một（có khả năng thi hành mà cố không thi hành）. Người bị thi hành án là người chưa thành niên thì không được đưa vào. Thời hạn đưa vào là 2 năm. Dùng đe dọa bạo lực để cản trở chống thi hành án mà tình tiết nghiêm trọng, hoặc có nhiều hành vi không thi hành án thì có thể kéo dài 1 đến 3 năm. Đã thi hành xong, thoả thuận hoà giải đã thi hành xong, tòa quyết định chấm dứt thi hành án cùng sáu tình huống khác, tòa phải xoá trong 3 ngày làm việc. Thi hành tích cực hoặc chủ động sửa hành vi không thi hành án thì còn có thể được xoá sớm. Người tự nhiên bị hạn chế tiêu dùng không được làm chín loại tiêu dùng cao và tiêu dùng ngoài sinh hoạt, việc làm. Một là đi máy bay, nằm khoang mềm tàu hỏa, khoang hạng hai trở lên của tàu thủy. Hai là ở khách sạn, nhà nghỉ, hội quán, sân golf hạng sao trở lên và tiêu nhiều. Ba là mua bất động sản, hoặc xây mới, mở rộng, sửa sang nhà ở cao cấp. Bốn là thuê văn phòng cao cấp, khách sạn, căn hộ để làm việc. Năm là mua xe không thiết yếu cho kinh doanh. Sáu là du lịch, nghỉ dưỡng. Bảy là con cái học trường tư thục thu phí cao. Tám là dùng tiền phí lớn để mua sản phẩm tài chính, bảo hiểm. Chín là ngồi bất kỳ chỗ nào trên tàu cao tốc tên đầu chữ G, hoặc chỗ hạng nhất trở lên trên các tàu cao tốc khác. Vì sinh hoạt hoặc kinh doanh mà phải làm một trong những việc đó thì phải xin tòa, được duyệt mới được làm. Cung cấp bảo lãnh thật và hiệu lực, hoặc được người xin thi hành án đồng ý thì tòa có thể dỡ hạn chế tiêu dùng (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Tòa án nhân dân tối cao (最高人民法院) (2017 修正). Một số quy định về công bố thông tin về người bị thi hành án không thi hành án, 关于公布失信被执行人名单信息的若干规定（第一、二、三、四、十条）. ；Tòa án nhân dân tối cao (最高人民法院) (2015 修正). Một số quy định về hạn chế tiêu dùng cao của người bị thi hành án và các tiêu dùng liên quan, 关于限制被执行人高消费及有关消费的若干规定（法释〔2015〕17 号，第三、八、九条）：「（九）乘坐G字头动车组列车全部座位、其他动车组列车一等以上座位等其他非生活和工作必需的消费行为」. +- Ghi chú: Hạn chi tiêu là hạn tiêu dùng cao. Quy định này ra đời năm 2010, năm 2015 được sửa bằng 法释〔2015〕17 号, tên đổi thành 《……及有关消费的若干规定》, mục này viết theo văn bản hiện hành sau sửa đổi. Lần sửa đổi năm 2015 có ba điểm đáng lưu ý. Một là thêm loại thứ chín, mọi chỗ ngồi trên tàu cao tốc đều tính. Hai là bỏ lớp giới hạn 「bằng tài sản của mình」, trước đây viết là 「không được có các hành vi sau trả phí bằng tài sản của mình」. Ba là đơn vị bị hạn chế tiêu dùng thì người thực sự kiểm soát cũng bị hạn chế, nhưng tiêu dùng tư, dùng tiền cá nhân thì có thể xin tòa thi hành án. Danh sách không thi hành án và hạn chế tiêu dùng là hai việc khác nhau, có thể bạn chỉ dính vào một cái. Cả hai cũng không có nghĩa là lịch sử tín dụng bị bẩn, bị xoá khỏi danh sách không sửa lịch sử tín dụng, xem mục 19 ở phần 7（sau khi lỡ bị đưa vào danh sách thì làm lại thế nào）. Vợ chồng, con cái dùng tiền mình tự kiếm được thì có bị hạn chế không, văn bản hiện hành không nói rõ, gặp thì hỏi tòa thi hành án. + +### 22. Mua sắm trên mạng, giao dịch đồ cũ bị lừa thì khiếu nại nền tảng trước, báo công an sau, rồi mới tính có đáng kiện hay không + +- Chi phí: Khiếu nại nền tảng và báo công an đều không tốn tiền. Khi khởi kiện, số tiền tranh chấp dưới 1 vạn tệ thì phí tiếp nhận 50 tệ; đi theo thủ tục giản dị, hoà giải hoặc rút đơn thì giảm thêm một nửa. Khởi kiện, xét xử, thi hành án mỗi bước ít nhất nửa ngày đến một ngày, khởi kiện trên mạng thì bớt được chuyến đi chạy về. Phải lưu đơn hàng, lịch sử trò chuyện, lịch sử chuyển tiền và phiếu vận chuyển. Khó là đừng bỏ cuộc giữa đường khi phải đi lại nhiều lần. +- Nói thẳng: Trước hết tìm nền tảng. Nền tảng có ký quỹ và quy tắc xử phạt, xử lý nhanh nhất. Rồi báo công an: bị lừa trên 3000 tệ là đủ ngưỡng khởi kiện tội lừa đảo, không đủ chuẩn hình sự thì xử lý theo pháp luật xử phạt hành chính, bị giữ 5 đến 10 ngày. Muốn kiện tụng thì tranh chấp mua sắm trên mạng có thể khởi kiện tại tòa nơi nhận hàng, số tiền nhỏ thì áp dụng thủ tục tụng thọp, xét xử xong trong 2 tháng, khởi kiện ngay trên mạng. +- Lợi ích: Lừa đảo số tiền từ 3000 tệ đến trên 1 vạn tệ thì cấu thành 「số tiền lớn」 của tội lừa đảo, phạt dưới 3 năm, đồng thời phạt tiền hoặc chỉ phạt tiền, mỗi tỉnh ấn định tiêu chuẩn địa phương trong khung này. Không đủ chuẩn hình sự thì xử lý theo pháp luật xử phạt hành chính: bị giữ 5 đến 10 ngày, hoặc phạt dưới 2000 tệ. Tình tiết nặng hơn thì bị giữ 10 đến 15 ngày, đồng thời có thể phạt dưới 3000 tệ. Về mặt dân sự, tranh chấp hợp đồng mua sắm trên mạng có thể khởi kiện tại tòa nơi nhận hàng, không cần chạy tới nơi cư trú của người bán. Số tiền tranh chấp dưới 50% mức lương bình quân năm của người có việc làm trong tỉnh của năm trước thì áp dụng thủ tục tụng thọp. Tụng thọp là bản án sơ thẩm chung cứu（xét xử xong là có hiệu lực, không được kháng nghị）. Khởi kiện xong phải xét xử trong 2 tháng, thủ tục giản dị là 3 tháng. Tố tụng dân sự có thể tiến hành trên các nền tảng trên mạng, hiệu lực như ngoài đời (toàn quốc, pháp luật tố tụng dân sự từ tháng 1 năm 2024, pháp luật xử phạt hành chính về trật tự công cộng từ tháng 1 năm 2026) +- Mức bằng chứng: A +- Nguồn: Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). Giải thích một số vấn đề về áp dụng cụ thể pháp luật khi xét các vụ án lừa đảo, 关于办理诈骗刑事案件具体应用法律若干问题的解释（法释〔2011〕7 号，第一条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百六十六条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). Pháp luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订，第五十八条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 修正). Pháp luật Tố tụng Dân sự, 民事诉讼法（第十六、一百六十四、一百六十五、一百六十八条）. （bản đăng lại của Ủy ban Phát triển và Cải cách Thượng Hải）；Tòa án nhân dân tối cao (最高人民法院) (2022 修正). Giải thích về việc áp dụng Pháp luật Tố tụng Dân sự, 关于适用《民事诉讼法》的解释（第二十条）. ；Hội đồng Nhà nước (国务院) (2006). Quy định về thu phí tố tụng, 诉讼费用交纳办法（国务院令第 481 号，第十三、十五、十六条）. +- Ghi chú: Đáng kiện hay không thì tính như sau: lấy mức lương theo giờ sau thuế của bạn nhân với số giờ phải bỏ ra, cộng thêm phí tiếp nhận của tòa. Khởi kiện, xét xử, có thể còn phải xin thi hành án, mỗi việc tính nửa ngày đến một ngày. Tính ra rồi so với số tiền bạn muốn đòi lại, rồi trừ bớt, ước lượng xem thắng kiện thì có thi hành được tiền thật không. Có kiện hay không, tự bạn quyết theo bảng tính đó. Sau khi pháp luật xử phạt hành chính về trật tự công cộng sửa đổi năm 2025 thì số điều đổi hết, trộm cắp và lừa đảo từ Điều bốn mươi chín của luật cũ sang Điều năm mươi tám. Sau khi pháp luật tố tụng dân sự sửa đổi năm 2023 thì thủ tục tụng thọp nằm ở Điều trăm sáu mươi lăm. Hai chỗ này đừng chép tài liệu cũ. 「Khiếu nại nền tảng trước」 là kinh nghiệm: nền tảng có ký quỹ và quy tắc xử phạt, xử lý nhanh nhất. 法释〔2011〕7 号 trên trang của viện kiệm sát không ghi số hiệu văn bản, số hiệu còn phải làm rõ với trang bản gốc của Tòa án nhân dân tối cao. + +### 23. Tặng tiền lớn khi yêu đương và trong hôn nhân, nghĩ kỹ trước khi chuyển, đã chuyển thì nguyên tắc không đòi lại được + +- Chi phí: Không tốn tiền. Trước khi chuyển khoản lớn thì nghĩ một câu: đó là vay hay tặng. Là vay thì viết giấy mượn, ghi chú 「vay」 trong phần chuyển tiền. Khó là mở lời nói câu đó. +- Nói thẳng: Tiền đã chuyển thì nguyên tắc không đòi lại được. Món quà đã tặng chỉ hủy được khi chưa chuyển quyền sở hữu sang tên người kia. Sau khi sang tên xong, chỉ còn ba trường hợp hủy được: người nhận nghiêm trọng xâm phạm bạn hoặc người thân gần của bạn, có nghĩa vụ nuôi dưỡng bạn mà không nuôi dưỡng, không thực hiện nghĩa vụ đã thoả thuận trong hợp đồng tặng. Vậy nên trước khi chuyển khoản lớn hãy nghĩ kỹ, đó là vay hay tặng. +- Lợi ích: Món quà đã tặng thì hủy được trước khi quyền tài sản chuyển sang tên người kia, pháp luật gọi mốc đó là thời điểm quyền tài sản chuyển giao. Chuyển tiền, sang tên xong thì chỉ còn ba trường hợp hủy được: bên nhận nghiêm trọng xâm phạm bạn hoặc người thân gần của bạn; có nghĩa vụ nuôi dưỡng bạn mà không nuôi dưỡng; không thực hiện nghĩa vụ đã thoả thuận trong hợp đồng tặng. Quà tặng, tiền mừng có giá trị không lớn dịp lễ, dịp sinh nhật có ý nghĩa kỷ niệm, cùng các chi tiêu hằng ngày để thể hiện hay vun bồi tình cảm, không thuộc tiền cưới, khi chia tay không phải hoàn lại theo kiểu tiền cưới (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第六百五十八、六百六十三条）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). Một số quy định về áp dụng pháp luật khi xét các vụ án tranh chấp liên quan đến tiền cưới, 关于审理涉彩礼纠纷案件适用法律若干问题的规定（法释〔2024〕1 号，第三条）. +- Ghi chú: 「Những số tiền đặc biệt như 520, 1314 thường được xác định là tặng」 là cách nói thường gặp trong thực tiễn, phần này không tìm được bản gốc chính thức của Tòa án nhân dân tối cao, nên không coi là kết luận. Món quà đã công chứng tại văn phòng công chứng thì không thể nói muốn rút lại là rút lại. Một khoản chuyển tiền là vay hay là tặng thì cuối cùng vẫn phải xem chứng cứ, một câu 「khoản này tôi cho bạn mượn trước」 trong lịch sử trò chuyện hữu ích hơn nhiều so với tranh cãi sau đó. +### 24. Tiền cưới thì chuyển khoản và ghi chú, giữ lại lịch sử trò chuyện; cưới để đòi tiền thì đòi lại được + +- Chi phí: Không tốn tiền. Tiền cưới, ba món vàng, tiền đổi lời đều nên chuyển khoản và ghi rõ mục đích trong phần ghi chú. Chuyện chỉ nói miệng thì để lại dấu vết trong lịch sử trò chuyện, hoặc nhờ người nhà có mặt xác nhận. +- Nói thẳng: Cưới để đòi tài sản là pháp luật cấm, tòa hỗ trợ yêu cầu hoàn trả. Không đăng ký, đã đăng ký nhưng thật sự không sống chung, hoặc đã đưa tiền cưới trước khi cưới khiến bên đưa tiền gặp khó khăn về sinh hoạt, thì có thể yêu cầu hoàn trả. Đã đăng ký và sống chung thì thường không hoàn trả. Vì vậy tiền cưới, ba món vàng, tiền đổi lời đều nên chuyển khoản và ghi chú mục đích, bên đưa tiền thì giữ bằng chứng, nam nữ như nhau. +- Lợi ích: Pháp luật cấm cưới để đòi tài sản. Khi một bên cưới để đòi tài sản dưới danh nghĩa tiền cưới mà bên kia yêu cầu hoàn trả thì tòa hỗ trợ. Có ba tình huống có thể yêu cầu hoàn trả: không làm đăng ký kết hôn; đã đăng ký nhưng thật sự không sống chung; đã đưa tiền cưới trước khi cưới và bên đưa tiền gặp khó khăn về sinh hoạt. Đã đăng ký và sống chung thì thường không hoàn trả. Nhưng nếu thời gian sống chung ngắn, số tiền cưới lại quá cao thì tòa căn cứ vào số tiền thật sự đã dùng vào việc gì, có đồ theo vợ hay không, có mang thai sinh con hay không, hai bên có lỗi hay không, để xác định tỷ lệ hoàn trả. Không đăng ký nhưng đã sống chung thì cũng xử lý theo tỷ lệ như vậy (toàn quốc, từ tháng 2 năm 2024) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零四十二条）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). Một số quy định về áp dụng pháp luật khi xét các vụ án tranh chấp liên quan đến tiền cưới, 关于审理涉彩礼纠纷案件适用法律若干问题的规定（法释〔2024〕1 号，第二、五、六条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Giải thích (một) về việc áp dụng phần Gia đạo và Hôn nhân của Bộ luật Dân sự, 关于适用《民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第五条）. +- Ghi chú: Tiền cưới có tính là 「số tiền quá cao」 không thì phải xem thu nhập khả dụng bình quân đầu người ở nơi bên đưa tiền và hoàn cảnh gia đình họ, cả nước không có một con số thống nhất. Phải trả lại bao nhiêu thì thẩm phán quyết từng vụ. Bên đưa tiền và bên nhận tiền áp dụng cùng một quy tắc, bên đưa tiền thì giữ bằng chứng. + +### 25. Tài sản trước khi cưới không phải sợ, việc thêm tên vợ chồng và tiền của cha mẹ thì viết rõ trước + +- Chi phí: Tự viết thỏa thuận tài sản bằng văn bản thì không tốn tiền, ra công chứng thì phí riêng. Khi cha mẹ đóng tiền thì nhờ họ viết một dòng giải trình bằng văn bản. Cả việc này tốn mấy tiếng một lần. Khó là mở lời bàn chuyện tiền với người nhà. +- Nói thẳng: Tài sản trước khi cưới là tài sản cá nhân của bạn, không biến thành tài sản chung chỉ vì cưới lâu năm. Vợ chồng có thể bằng văn bản thỏa thuận tài sản trước và sau khi cưới thuộc về mỗi người. Cần để ý một điểm: nói sẽ thêm tên người kia vào nhà thì trước khi đăng ký sang tên vẫn rút lại được, đăng ký xong là của người kia. Khi ly hôn, bên giấu tiền, chuyển tiền, tiêu tán hoặc bịa nợ thì có thể được chia ít hoặc không được chia. +- Lợi ích: Tài sản trước khi cưới là tài sản cá nhân, không biến thành tài sản chung chỉ vì cưới lâu năm. Vợ chồng có thể bằng văn bản thỏa thuận: tài sản trước khi cưới và thu nhập sau khi cưới thuộc về từng người, thuộc về chung, hoặc một phần thuộc chung, thỏa thuận này ràng buộc cả hai bên. Cha mẹ đóng tiền mua nhà trước khi cưới thì coi là tặng cho con mình, trừ khi nói rõ là tặng cho cả hai người. Cha mẹ đóng tiền sau khi cưới thì xử lý theo thỏa thuận, không có thỏa thuận thì xử lý theo nguyên tắc tài sản chung. Nếu trước khi cưới hoặc trong hôn nhân có thỏa thuận tặng bất động sản của một bên cho bên kia, hoặc để thành của hai người cùng sở hữu thì bên tặng có thể rút lại trước khi đăng ký sang tên, đăng ký xong là của người kia. Khi ly hôn, bên giấu, chuyển, bán, tiêu tán tài sản chung, hoặc bịa nợ thì có thể được chia ít hoặc không được chia (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零六十三、一千零六十五、一千零九十二条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Giải thích (một) về việc áp dụng phần Gia đạo và Hôn nhân của Bộ luật Dân sự, 关于适用《民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第二十九、三十一、三十二条）. +- Ghi chú: Thỏa thuận tài sản có hiệu lực với cả nam lẫn nữ. Muốn dùng nó để chống lại nợ của chồng với người ngoài thì điều kiện là chủ nợ phải biết hai vợ chồng có thỏa thuận đó. Người đó không biết thì không ràng buộc được họ. Thỏa thuận không công chứng vẫn có hiệu lực, công chứng chỉ để sau này khỏi tranh cãi về chữ ký và ngày tháng. + +### 26. Trước khi cưới xem tình trạng hôn nhân của đối phương, lấy cưới làm vỏ để đòi tiền thì truy cứu tội lừa đảo + +- Chi phí: Không tốn tiền. Trước khi đăng ký thì xem thẻ căn cước và sổ hộ khẩu của đối phương, phần tình trạng hôn nhân. Giao dịch tài sản lớn thì lưu lại lịch sử chuyển tiền và lịch sử trò chuyện. +- Nói thẳng: Hôn nhân lần hai thì vô hiệu. Đã có vợ chồng mà vẫn đi cưới, hoặc biết đối phương có vợ chồng mà vẫn cưới thì phạt dưới 2 năm. Lấy cưới làm vỏ để đòi tiền thì truy cứu theo tội lừa đảo: số tiền lớn phạt dưới 3 năm, số tiền rất lớn phạt 3 đến 10 năm, số tiền đặc biệt lớn phạt trên 10 năm hoặc chung thân. Trước khi đăng ký, xem thẻ căn cước và sổ hộ khẩu của đối phương, phần tình trạng hôn nhân. +- Lợi ích: Hôn nhân lần hai thì vô hiệu. Đã có vợ chồng mà vẫn cưới lần hai, hoặc biết đối phương có vợ chồng mà vẫn cưới thì phạt dưới 2 năm. Lừa đảo tài sản của công lẫn tư với số tiền lớn thì phạt dưới 3 năm và phạt tiền. Số tiền rất lớn thì phạt 3 đến 10 năm. Số tiền đặc biệt lớn thì phạt trên 10 năm hoặc chung thân, đồng thời phạt tiền hoặc tịch thu tài sản (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零五十一条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百五十八、二百六十六条）. +- Ghi chú: 「Lừa cưới」 bản thân không phải một tội danh. Có theo tội lừa đảo hay không thì phải xem ngay từ đầu người kia đã có ý định chiếm đoạt tiền, đã bịa lời để lấy tiền, và còn phải để công an khởi kiện điều tra. Có đủ chuẩn 「số tiền lớn」 không thì số tiền do giải thích tư pháp quy định, các nơi có thể điều chỉnh, phần này chưa làm rõ. Người bị lừa cưới không phân nam nữ, có cả vụ bị lừa tiền cưới lẫn vụ bị lừa đồ theo vợ. + +### 27. Sửa chữa tại nhà, kiểm định xe, sửa sang nhà thì hỏi giá trước khi để ai đó làm, và đứng cạnh xem suốt quá trình + +- Chi phí: Không tốn tiền. Trước khi làm thì hỏi rõ cách tính tiền và sẽ thay những bộ phận nào, chụp ảnh báo giá lại, hoặc bảo bên kia gửi bằng văn bản. Sửa xong phải có chứng tứ sửa chữa và hoá đơn, giữ lại bộ phận cũ. Còn phải đứng cạnh xem một hai tiếng. Khó là giữ được sự chú ý suốt quá trình, không rời đi. +- Nói thẳng: Sửa thiết bị điện tử gia dụng phải niêm yết cách tính tiền, phải có chứng tứ sửa chữa và hoá đơn, không được báo hỏng sai, không được cố ý thay bộ phận còn tốt, tình tiết nghiêm trọng thì phạt dưới 3 vạn tệ. Mọi người bán đều phải niêm yết giá công khai, không được tăng giá ngoài giá niêm yết. Vậy nên trước khi làm phải hỏi rõ giá và bộ phận sẽ thay, sửa xong phải có chứng tứ và hoá đơn, giữ lại bộ phận cũ. Bị thu thêm thì gọi 12315 trước. +- Lợi ích: Người sửa chữa thiết bị điện tử gia dụng phải niêm yết công khai hạng mục dịch vụ và phụ kiện, cách tính tiền, tiêu chuẩn chất lượng, thời hạn bảo hành và số điện thoại khiếu nại. Phải đưa cho người tiêu dùng chứng tứ dịch vụ sửa chữa và hoá đơn thu tiền, và ghi thật hạng mục sửa chữa cùng chi tiết. Không được ghi thêm, phóng đại, bịa hạng mục sửa chữa, không được báo sai bộ phận hỏng, không được cố ý thay bộ phận còn tốt. Tình tiết nghiêm trọng thì phạt dưới 3 vạn tệ. Mọi người bán đều phải niêm yết giá công khai, không được tăng giá ngoài giá niêm yết, không được thu bất kỳ khoản phí nào chưa niêm yết. Dùng giá rẻ để dụ dỗ, dùng giá cao để quyết toán, và báo giả giảm giá đều là gian lận giá. Người bán phải ghi rõ bằng cách dễ thấy hạng mục, nội dung, giá và cách tính của dịch vụ, không được quảng cáo sai lệch hoặc gây hiểu nhầm (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Bộ Thương mại (商务部) (2012). Quy định quản lý ngành dịch vụ sửa chữa thiết bị điện tử gia dụng, 家电维修服务业管理办法（商务部令 2012 年第 7 号，第五、七、九、十四条）. ；Tổng cục Quản lý thị trường quốc gia (市场监管总局) (2022). Quy định về niêm yết giá công khai và cấm gian lận giá, 明码标价和禁止价格欺诈规定（市场监管总局令第 56 号，第五、八、十九条）. ；Hội đồng Nhà nước (国务院) (2024). Quy định thực thi Luật Bảo vệ quyền lợi người tiêu dùng, 消费者权益保护法实施条例（国务院令第 778 号，第九、十条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2013 修正). Luật Bảo vệ quyền lợi người tiêu dùng, 消费者权益保护法（第八、十条）：「消费者有权根据商品或者服务的不同情况，要求经营者提供……服务的内容、规格、费用等有关情况」「有权拒绝经营者的强制交易行为」. （Cơ sở dữ liệu luật pháp quốc gia） +- Ghi chú: Bị thu thêm thì gọi 12315 trước, việc xử phạt do cơ quan thương mại và quản lý thị trường thực hiện, bạn không phải tự mình ra tòa. Để xếp mức B là vì quy định dù là chính thức nhưng 「đứng cạnh xem tiết kiệm được bao nhiêu」 không có con số, chủ yếu dựa vào kinh nghiệm. Hỏi giá, giữ chứng tứ, giữ bộ phận cũ thì ai cũng nên làm như nhau, không cần nhìn mặt người. Quy định về thiết bị gia dụng chỉ áp dụng cho sửa chữa thiết bị gia dụng; kiểm định xe và sửa sang nhà thì áp dụng quy định niêm yết giá công khai và quy định thực thi. Luật Bảo vệ quyền lợi người tiêu dùng ở cấp trên cũng ghi hai quyền này. Điều tám nói bạn có quyền yêu cầu người bán cho biết nội dung, quy cách, chi phí của dịch vụ. Điều mười nói bạn có quyền từ chối giao dịch cưỡng ép. + +### 28. Đừng làm 「người đứng tên」, đừng đưa thẻ căn cước cho người khác để đăng ký công ty + +- Chi phí: Không tốn tiền. Có người nhờ bạn 「đứng tên thôi, không phải quản gì」 thì từ chối là được. +- Nói thẳng: Pháp luật mặc định người đại diện theo pháp luật là người thật sự điều hành, 「tôi chỉ đứng tên, không biết gì」 không phải lý do miễn trách. Người đại diện theo pháp luật thi hành công việc mà gây thiệt hại cho người khác thì công ty phải bồi thường trước, bồi thường xong có thể truy đòi ngược lại bạn nếu bạn có lỗi. Đã đứng tên rồi thì Luật Công ty cho phép bạn từ chối, công ty phải chỉ định người đại diện theo pháp luật mới trong 30 ngày. +- Lợi ích: Người đại diện theo pháp luật do giám đốc hoặc quản lý thật sự điều hành công việc công ty đảm nhiệm, pháp luật mặc định chính bạn là người điều hành. Người đại diện theo pháp luật thi hành công việc mà gây thiệt hại cho người khác thì công ty bồi thường trước, bồi thường xong có thể truy đòi ngược lại người đại diện theo pháp luật có lỗi（lấy lại khoản tiền đó）. Giám đốc, lãnh đạo công ty có nghĩa vụ trung thành và nghĩa vụ cần cù đối với công ty（trung thành với công ty, làm việc tận tâm）. Thi hành công việc mà gây thiệt hại cho người khác, bản thân lại cố ý hoặc có lỗi nặng thì bản thân cũng phải tự bỏ tiền bồi thường. 「Không biết」 trong mắt pháp luật không phải lý do miễn trách (toàn quốc, từ tháng 7 năm 2024) +- Mức bằng chứng: B +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Luật Công ty, 公司法（2023 年修订，第十、十一、一百八十、一百九十一条）. +- Ghi chú: Nguyên văn Luật Công ty chỉ nói trách nhiệm thuộc về ai. Hậu quả thường gặp hơn của việc đứng tên là: khi công ty bị thi hành án, người đại diện theo pháp luật cũng bị hạn chế tiêu dùng, thuế và xử phạt hành chính cũng có thể kéo bạn vào. Những hậu quả đó đến từ pháp luật và quy chế thi hành án khác, phần này chưa đối chiếu nguyên văn nên xếp mức B. Đã đứng tên rồi thì Luật Công ty cho phép bạn từ chối, công ty phải chỉ định người đại diện theo pháp luật mới trong 30 ngày. + +### 29. Ra nước ngoài thì đừng mang đồ giúp người lạ, đừng nhận hộ bưu kiện không rõ nguồn gốc + +- Chi phí: Không tốn tiền. Từ chối là được. +- Nói thẳng: Lực lượng lãnh sự của Bộ Ngoại giao khuyên công dân đi nước ngoài: đừng mang hành lý đồ vật giúp người khác, nhất là hàng cấm, hoặc thứ mà bạn không hiểu rõ. Trong hành lý mà tìm thấy ma túy hoặc hàng vận chuyển lậu thì 「tôi không biết trong đó có gì」 là bạn phải tự chứng minh. Nhận hộ bưu kiện trong nước cũng vậy, nhất là loại bảo bạn ứng tiền trước rồi hoàn lại sau. +- Lợi ích: Lực lượng lãnh sự của Bộ Ngoại giao khuyên công dân đi nước ngoài 「xin hãy tuyệt đối tránh mang hành lý đồ vật giúp người khác, nhất là hàng cấm hoặc hàng không rõ đặc điểm」. Làm vậy là để tránh phải gánh trách nhiệm pháp lý khi vô tình mang hàng cấm giúp người khác. Trong hành lý mà tìm thấy ma túy hoặc hàng vận chuyển lậu thì 「không biết trong đó là gì」 là bạn phải tự chứng minh +- Mức bằng chứng: B +- Nguồn: Lực lượng lãnh sự Bộ Ngoại giao (外交部领事司) (2023). Những lời khuyên dành cho công dân Trung Quốc chuẩn bị ra nước ngoài, 对准备出国的中国公民有哪些建议. +- Ghi chú: Trang nhắc nhở chính thức của Tổng cục Hải quan và Bộ Công an về việc này không mở được, mục này chỉ dẫn trang của Lực lượng lãnh sự Bộ Ngoại giao. Và nguyên văn chỉ nêu nguyên tắc, không nêu điều luật, nên xếp mức B. Nhận hộ bưu kiện trong nước cũng vậy, nhất là loại bảo bạn ứng tiền trước rồi hoàn lại sau. + +### 30. Nuôi chó thì phải dắt dây: không dắt dây mà xảy ra chuyện thì chịu trách nhiệm không lỗi (dù bạn có lỗi hay không cũng phải bồi thường), bồi thường đến cùng + +- Chi phí: Một sợi dây vài chục tệ. Khó là mỗi lần ra ngoài đều nhớ dắt dây. +- Nói thẳng: Chó cắn người thì nguyên tắc bạn phải bồi thường dù có lỗi hay không. Chỉ khi chứng minh được bên bị hại cố ý hoặc có lỗi nặng thì mới có thể bồi thường ít hơn hoặc không phải bồi thường. Mà nếu bạn vi phạm quy định, không dắt dây thì chỉ còn lối thoát 「bên bị hại cố ý」 mới được giảm nhẹ, không dắt dây tức là chặn luôn lối thoát duy nhất. Chó vật nuôi bị cấm gây thương tích cho người thì không có lý do miễn trừ nào. +- Lợi ích: Động vật do mình nuôi gây thiệt hại cho người khác thì người nuôi hoặc người quản lý nó lúc đó phải bồi thường, pháp luật gọi là phải chịu trách nhiệm cáo buộc. Chỉ khi chứng minh được bên bị hại cố ý gây chuyện, hoặc có lỗi nặng thì mới không phải hoặc không phải bồi thường đầy đủ. Mà vi phạm quy định quản lý, không có biện pháp an toàn, ví dụ như phải dắt dây mà không dắt thì chỉ còn 「bên bị hại cố ý」 mới được giảm nhẹ. Chó vật nuôi bị cấm như chó dữ gây thương tích cho người thì không có lý do miễn trừ nào +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典. Toàn văn đăng lại bởi Viện kiểm sát nhân dân tối cao. 第一千二百四十五条：「饲养的动物造成他人损害的，动物饲养人或者管理人应当承担侵权责任；但是，能够证明损害是因被侵权人故意或者重大过失造成的，可以不承担或者减轻责任。」第一千二百四十六条：「违反管理规定，未对动物采取安全措施造成他人损害的，动物饲养人或者管理人应当承担侵权责任；但是，能够证明损害是因被侵权人故意造成的，可以减轻责任。」第一千二百四十七条：「禁止饲养的烈性犬等危险动物造成他人损害的，动物饲养人或者管理人应当承担侵权责任。」第一千二百四十九条：遗弃、逃逸的动物在此期间造成损害，由原饲养人或者管理人承担责任 +- Ghi chú: Tiền bồi thường vết cắn của chó gồm tiền khám chữa, tiền công mất việc, vắc-xin dại và globulin miễn dịch, một lần trên vạn tệ là chuyện thường thấy. Vết thương nặng, hoặc trúng phải trẻ em thì số tiền còn cao hơn. Bỏ mặc cũng không thoát trách nhiệm: bị cắn trong thời gian bị bỏ rơi, đào thoi thì vẫn tính vào người nuôi trước đó. Sau khi bị cắn xử lý thế nào, xem phần 13. +### 31. Đối phương uống tới mất ý thức, đã ngủ, hoặc có thể dưới 14 tuổi thì không quan hệ tình dục + +- Chi phí: Không tốn tiền. Cái giá phải trả là dừng lại ngay lúc đó. Khó là dừng được tại chỗ. +- Nói thẳng: Đối phương uống tới mất ý thức hoặc đã ngủ thì không có năng lực đồng ý, quan hệ tình dục lúc đó theo pháp luật cũng là cưỡng ép, hình phạt căn cứ là 3 đến 10 năm. Sau đó gần như không thể chứng minh đối phương đã đồng ý lúc đó. Cách làm đáng tin duy nhất là không quan hệ tình dục, và khi tiệc tàn thì giao người đó cho bạn bè hoặc người nhà của chính họ. Cách xác định tuổi, xem mục 18 ở phần 9（chưa đủ 14 tuổi）. +- Lợi ích: Điều hai trăm ba mươi sáu của Bộ luật Hình sự: người cưỡng ép phụ nữ bằng bạo lực, uy hiếp hoặc phương tiện khác thì phạt tù từ ba năm đến mười năm. 「Phương tiện khác」 không giới hạn ở việc ra tay; đối phương mất ý thức, đang ngủ không thể đồng ý có hiệu lực thì rơi vào bốn chữ đó. Cưỡng ép phụ nữ trong tình huống nghiêm trọng, cưỡng ép nhiều phụ nữ và các trường hợp tương tự thì phạt tù từ mười năm trở lên, chung thân hoặc tử hình. Cùng điều đó còn quy định, quan hệ tình dục với thiếu nữ chưa đủ mười bốn tuổi thì xử theo tội cưỡng ép và phạt nặng hơn, cách xác định thiếu nữ xem mục 18 ở phần 9（chưa đủ 14 tuổi）. Bài phổ biến pháp luật của Viện kiểm sát nhân dân quận Tây Thành, Bắc Kinh nói rằng người 「麻醉、灌醉后实施性侵的，同样构成强奸罪」 cũng cấu thành tội cưỡng ép. Điểm mấu chốt là 「có trái với ý chí thật bên trong của nạn nhân hay không」, còn nạn nhân có chống cự hay không lúc đó 「并非关键」. Điều hai trăm ba mươi bảy của Bộ luật Hình sự quy định cưỡng ép sách tế: cưỡng ép sách tế người khác bằng bạo lực, uy hiếp hoặc phương tiện khác thì phạt tù dưới năm năm, hoặc giam (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百三十六条）. ；Quốc hội (全国人大) (2020 年修正). Bộ luật Hình sự, 刑法（第二百三十七条）. （bản đăng lại và hợp nhất của Cục Quản lý giao thông đường bộ Công an Bắc Kinh）；Viện kiểm sát nhân dân quận Tây Thành, Bắc Kinh (北京市西城区人民检察院) (2018). 她们为什么不再沉默……. +- Ghi chú: Xếp mức B là vì điều luật chỉ có bốn chữ 「phương tiện khác」. 「Có ép rượu rồi quan hệ tình dục vẫn là cưỡng ép」 lấy từ bài phổ biến pháp luật của viện kiểm sát, không phải giải thích tư pháp. Điều về tội cưỡng ép chỉ nói về phụ nữ. Nam giới bị xâm hại kiểu này thì truy cứu theo tội cưỡng ép sách tế, nên đối phương là nam hay nữ thì cũng không được đụng vào. Nếu bạn là bên bị hại thì làm gì trước, xem mục 42 ở phần 13（sau khi bị xâm hại tính dục）. Khi bạn tự uống tới mất trí nhớ thì hai rủi ro bị buộc tội và bị tống tiền cùng tồn tại, xem mục 32 của phần này（đối phương nắm nhược điểm đòi tiền thì báo công an）. + +### 32. Sau khi quan hệ tình dục, quay video trần, đối phương dọa gọi công an, gửi ảnh, báo với cơ quan của bạn để đòi tiền thì không đưa một đồng nào, không xoá một bản ghi nào, báo công an ngay + +- Chi phí: Không tốn tiền. Chụp màn hình tài khoản, lịch sử trò chuyện và mã nhận tiền của đối phương, mang theo điện thoại tới trạm công an. Khó là nhịn được ý nghĩ 「cho tiền để xong chuyện」. +- Nói thẳng: Dọa bằng 「tôi sẽ gọi công an」「tôi sẽ gửi ảnh cho cơ quan của bạn」 để đòi tiền là tống tiền. Không đưa một đồng nào, không xoá một bản ghi nào, báo công an ngay tại chỗ, đã đưa khoản đầu tiên thì sẽ có khoản thứ hai. Đối phương thật sự tung ảnh ra thì đó là thêm một tội cho họ. Tống tiền trên 2000 đến 5000 tệ thì phạt dưới 3 năm, 3 vạn đến trên 10 vạn tệ thì phạt 3 đến 10 năm. +- Lợi ích: Điều hai trăm bảy mươi tư của Bộ luật Hình sự: tống tiền tài sản của công lẫn tư, số tiền lớn hoặc tống tiền nhiều lần thì phạt tù dưới ba năm, giam hoặc quản túc, đồng thời phạt tiền hoặc chỉ phạt tiền. Giam là giam giữ ngắn hạn, quản túc là không giam nhưng hạn chế tự do, đồng thời phạt tiền hoặc chỉ phạt tiền là phạt thêm tiền hoặc chỉ phạt tiền. Số tiền rất lớn hoặc có tình tiết nghiêm trọng khác thì phạt tù từ ba năm đến mười năm, đồng thời phạt tiền. Số tiền đặc biệt lớn hoặc có tình tiết đặc biệt nghiêm trọng thì phạt tù từ mười năm trở lên, đồng thời phạt tiền. Hai tòa cao nhất còn có một giải thích: Giải thích một số vấn đề về áp dụng pháp luật khi xét các vụ án tống tiền, 关于办理敲诈勒索刑事案件适用法律若干问题的解释（法释〔2013〕10 号, có hiệu lực từ ngày 27 tháng 4 năm 2013）. Điều một của giải thích đó ấn định ba mức tiền. Tống tiền tài sản công tư trị giá từ hai nghìn tệ đến trên năm nghìn tệ thì xác định là 「số tiền lớn」. Từ ba vạn tệ đến trên mười vạn tệ thì xác định là 「số tiền rất lớn」. Từ ba mươi vạn tệ đến trên năm mươi vạn tệ thì xác định là 「số tiền đặc biệt lớn」. Điều ba: tống tiền từ ba lần trở lên trong vòng hai năm thì phải xác định là 「tống tiền nhiều lần」 (toàn quốc). Điều một nghìn không trăm hai mươi của Bộ luật Dân sự: không tổ chức hay cá nhân nào được xâm phạm quyền riêng tư của người khác bằng cách dò xét, quấy rối, làm lộ, công khai. Điều một nghìn không trăm ba mươi ba cấm quấy rối sự yên tĩnh đời sống riêng tư bằng điện thoại, tin nhắn, công cụ nhắn tin tức thì. Vậy nên đối phương thật sự tung ảnh ra thì đó là thêm một tội cho họ. Nguyên văn nhắc nhở của Sở Công an tỉnh Phúc Kiến với các vụ tương tự là: 「一定要保存好证据，及时拨打 110，切勿继续转账」 +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020 年修正). Bộ luật Hình sự, 刑法（第二百七十四条）. （bản đăng lại và hợp nhất của Cục Quản lý giao thông đường bộ Công an Bắc Kinh）；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2013). Giải thích một số vấn đề về áp dụng pháp luật khi xét các vụ án tống tiền, 关于办理敲诈勒索刑事案件适用法律若干问题的解释（法释〔2013〕10 号）. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零三十二、一千零三十三条）. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千一百九十五条）：「权利人有权通知网络服务提供者采取删除、屏蔽、断开链接等必要措施。通知应当包括构成侵权的初步证据及权利人的真实身份信息」「未及时采取必要措施的，对损害的扩大部分与该网络用户承担连带责任」. （Cơ sở dữ liệu luật pháp quốc gia）；Cục Công an tỉnh Phúc Kiến (福建省公安厅) (2024). Vì một tin nhắn 「约炮」 mà tôi gánh trên 23 vạn tệ nợ. ；Báo Kiểm sát (检察日报) (2018). Hai lần bị kết án không hối hận, đăng ảnh riêng tư để đe dọa người trên mạng（bản đăng lại trên trang Viện kiểm sát nhân dân quận Cố Lâu, thành phố Hà Châu）. ；Báo Kiểm sát (检察日报) (2022). Đàn ông dọa công khai ảnh trần để ép 10 phụ nữ quan hệ tình dục, sàng lọc dữ liệu lớn 「báo công an」！（trên trang Viện kiểm sát nhân dân tối cao）. +- Ghi chú: Ba mức tiền do mỗi tỉnh ấn định trong khung mà giải thích tư pháp đưa ra. 2000 đến trên 5000 tệ tính là 「số tiền lớn」. 3 vạn đến trên 10 vạn tệ tính là 「số tiền rất lớn」. 30 vạn đến trên 50 vạn tệ tính là 「số tiền đặc biệt lớn」. Tỉnh của bạn rơi vào con số nào thì theo tiêu chuẩn địa phương. Tống tiền ba lần trở lên trong hai năm thì mỗi lần số tiền không lớn cũng vẫn tính là 「nhiều lần」. Đừng tự xoá lịch sử trò chuyện, xoá ảnh, huỷ tài khoản, chặn là xong, đó là tự xoá bằng chứng của chính mình, xem mục 5 của phần này（bị triệu tập thì mời luật sư trước, không xoá bản ghi）. Bị khởi kiện điều tra thật thì những bản ghi đó đồng thời là tài liệu để bạn tự chứng minh. Cũng đừng thử 「thương lượng chia nhiều khoản」. Tống tiền bằng quay video trần phần lớn do nhóm ở nước ngoài làm, họ vừa quay màn hình vừa dùng mã độc để lấy danh bạ của bạn. 「Gửi cho cả gia đình」 là câu đàm kỳ; sau khi báo công an vẫn khởi kiện được. Dọa bằng 「tôi sẽ nói với người nhà bạn là bạn thích giống tính」 để đòi tiền cũng là tống tiền. Ở thành phố Diêm Trung, tỉnh Giang Tô có người hẹn gặp qua ứng dụng hẹn hò người đồng giới rồi đòi tiền như vậy, hơn 14.500 tệ; người bị hại báo công an thì họ bị khởi kiện, trước đó họ đã bị phạt tù hai lần vì cùng thủ đoạn. Dùng ảnh trần ép người ta quan hệ tình dục thì truy cứu theo tội cưỡng ép. Ở huyện Tượng Sơn, tỉnh Chiêu có người lừa được ảnh trần của nhiều phụ nữ trên nền tảng xã hội, rồi dọa tung ảnh, xác minh có 10 người bị hại. Một người nhận được tin đe dọa đã báo công an, hai người chưa từng gặp mặt, viện kiểm sát vẫn xử lý là cưỡng ép chưa thành công, cuối cùng khởi tố theo tội cưỡng ép và tội cưỡng ép sách tế. Vậy nên đừng gửi ảnh trần cho người quen qua mạng. Ngược lại, nếu bạn là bên bị hại mà đi đòi bồi thường, số tiền lớn cũng không có nghĩa là tống tiền, xem mục 36 của phần này（bản thân là bên bị hại mà đi đòi bồi thường）. Ảnh thật sự bị đăng lên mạng thì có thể yêu cầu nền tảng xoá. Điều một nghìn một trăm chín mươi lăm của Bộ luật Dân sự quy định bạn có quyền yêu cầu nền tảng xoá, chặn, ngắt liên kết, trong đơn phải kèm chứng cứ sơ bộ về hành vi xâm phạm và thông tin nhận dạng thật của bạn. Nền tảng nhận thông báo mà xử lý không kịp thì phần thiệt hại mở rộng do nó và người đăng cùng chịu trách nhiệm liên đới. + +### 33. Bị người khác bịa sự thật báo cáo thì có thể yêu cầu truy cứu: đủ mức xử phạt hành chính thì bị giữ từ 5 ngày, đủ cấu thành tội thì phạt dưới 3 năm + +- Chi phí: Không tốn tiền. Đến cơ quan công an báo việc, nói rõ đối phương đã bịa sự thật. Khó là đưa ra được thứ chứng minh họ cố tình bịa. +- Nói thẳng: Bịa sự thật để vu khống người khác, muốn khiến họ bị truy cứu trách nhiệm hình sự hoặc bị xử phạt hành chính, bản thân việc đó đã vi phạm pháp. Trường hợp này bị giữ dưới 5 ngày, hoặc phạt dưới 1000 tệ. Tình tiết nặng hơn thì bị giữ 5 đến 10 ngày. Đủ cấu thành tội phạm thì là tội vu khống, phạt dưới 3 năm. Nhưng nhớ nhầm, báo nhầm thì không tính, muốn truy cứu đối phương thì bạn phải chứng minh họ cố tình vu. +- Lợi ích: Điều năm mươi của Pháp luật Xử phạt hành chính về trật tự công cộng（sửa đổi ngày 27 tháng 6 năm 2025, có hiệu lực từ ngày 1 tháng 1 năm 2026）quy định về vu khống. Bịa sự thật vu khống người khác, cố làm cho người đó bị truy cứu hình sự hoặc bị xử phạt hành chính thì phải xử phạt. Phạt giam dưới năm ngày, hoặc phạt tiền dưới một nghìn tệ. Tình tiết nặng hơn thì phạt giam từ năm ngày đến mười ngày, có thể kèm phạt tiền dưới một nghìn tệ. Điều hai trăm bốn mươi ba của Bộ luật Hình sự quy định tội vu khống. Bịa sự thật vu khống người khác, cố ý làm cho người đó bị truy cứu hình sự, tình tiết nghiêm trọng thì phạt tù dưới ba năm, giam hoặc quản túc. Gây hậu quả nghiêm trọng thì phạt tù từ ba năm đến mười năm. Cán bộ, công chức cơ quan nhà nước phạm tội trong khoản trên bị phạt nặng hơn, phạt nặng hơn là phán trong cùng khung hình phạt nhưng chọn mức nặng hơn. Không phải cố ý vu khống mà báo nhầm, hoặc báo cáo sai sự thật thì không áp dụng hai khoản trên. Nhớ nhầm, báo nhầm thì không truy cứu theo hai khoản đó (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). Pháp luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订，第五十条）. （bản đăng lại toàn văn của Cục Công an Bắc Kinh）；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百四十三条）. +- Ghi chú: Muốn truy cứu đối phương thì chỉ chứng minh 「tôi không làm」 chưa đủ, còn phải chứng minh họ 「bịa sự thật」 và 「cố ý vu khống」. Phần sau khó chứng minh hơn nhiều, không chắc truy cứu được. Cả nước không có số liệu công khai về việc khởi kiện và xử lý vu khống, cuốn sách này tra không ra, nên không viết tỷ lệ, cũng không bàn phân bố theo giới tính. Sau khi bị buộc tội thì phải đưa chứng cứ và tìm cách cứu vãn thế nào, xem mục 34 và 35 của phần này（không đủ chứng cứ thì phải tuyên vô tội, bồi thường nhà nước）. + +### 34. Không đủ chứng cứ thì vốn đã phải tuyên vô tội, lời khai bị ép phải bị loại; đã tuyên thì còn khiếu nại và xử lại + +- Chi phí: Đi bộ thủ tục thì không tốn tiền, nhưng khiếu nại và xét lại đều rất tốn thời gian. Thường phải thuê luật sư, phí luật sư vài nghìn tệ trở lên. Người khó khăn về kinh tế có thể xin trợ giúp pháp lý miễn phí, xem mục 5 của phần này（bị buộc tội thì mời luật sư trước）. Khó là chờ qua từng vòng mà không bỏ cuộc. +- Nói thẳng: Ngưỡng để kết tội nằm trong Pháp luật Tố tụng Hình sự: chứng cứ kết tội phải được xác minh là đúng theo thủ tục luật định, và phải loại trừ nghi vấn hợp lý khi cân nhắc toàn bộ chứng cứ của vụ án. Không làm được thì phải tuyên vô tội. Lời khai có bắt ép, lời khai của nhân chứng có bạo lực, uy hiếp đều phải bị loại, không được dùng làm căn cứ của bản án. Sau khi bản án có hiệu lực, bạn và người đại diện hợp pháp, người thân gần của bạn vẫn có thể khiếu nại. +- Lợi ích: Điều năm mươi lăm của Pháp luật Tố tụng Hình sự: chứng cứ phải đúng và đủ, phải đáp ứng các điều kiện sau — chứng cứ kết tội đều đã được xác minh đúng theo thủ tục luật định, và đánh giá toàn bộ chứng cứ của vụ án đã loại trừ nghi vấn hợp lý. Loại trừ nghi vấn hợp lý nghĩa là cân nhắc chung toàn bộ chứng cứ của vụ án, sự thật được xác định không còn một khả năng nào khác hợp lý. Điều năm mươi sáu: lời khai của bị can, bị cáo thu thập bằng các biện pháp phi pháp như bắt ép, cùng lời khai của nhân chứng, lời kể của bị hại thu thập bằng bạo lực, uy hiếp hay biện pháp phi pháp khác thì phải bị loại. Trong lúc điều tra, thẩm vấn, xét xử mà phát hiện chứng cứ phải loại thì phải loại theo pháp luật, không được làm căn cứ cho ý kiến khởi tố, quyết định khởi tố và bản án. Điểm (ba) của Điều hai trăm: không đủ chứng cứ, không thể kết tội bị cáo thì phải ra phán quyết vô tội với lý do không đủ chứng cứ, tội bị cáo bị cáo buộc không thành lập. Điều hai trăm năm mươi hai mở đường cho khiếu nại. Đương sự và người đại diện hợp pháp, người thân gần của họ, với bản án, quyết định đã có hiệu lực pháp luật, có thể đề nghị tòa án hoặc viện kiểm sát. Nhưng khiếu nại không làm dừng việc thi hành bản án, quyết định. Điều hai trăm năm mươi ba liệt kê các trường hợp tòa án phải xét xử lại (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). Pháp luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第五十五、五十六、二百、二百五十二、二百五十三条）. +- Ghi chú: Đây là những đường ranh giới mà pháp luật viết ra, còn kết quả thì không được bảo đảm. Cuốn sách này không có số liệu chính thức đối chiếu được về tỷ lệ tuyên vô tội trong các vụ án hình sự ở Trung Quốc hay tỷ lệ khiếu nại được xét lại, nên không viết tỷ lệ, cũng không bình luận vụ cụ thể. Có ba việc bạn tự làm được. Một, sau lần hỏi cung đầu tiên thì ủy quyền cho luật sư, xem mục 5 của phần này（bị buộc tội thì mời luật sư trước）. Hai, đọc hết bản ghi từng chữ rồi hãy ký. Ba, đừng nhận tội việc mình chưa làm để 「ra tù sớm」, nhận tội nhận phạt thì chỉ được giảm nhẹ, không tránh được trách nhiệm. Đủ điều kiện luật định thì tòa phải xét xử lại. +### 35. Bị giữ rồi vụ được rút, không khởi tố hoặc tuyên vô tội thì đi xin bồi thường nhà nước, tiền tính theo ngày + +- Chi phí: Không tốn tiền. Gửi đơn yêu cầu bồi thường đến cơ quan đã ra quyết định bắt giữ, bắt, khởi tố hoặc tuyên án. +- Nói thẳng: Bị bắt giữ, bị bắt rồi vụ được rút, không khởi tố hoặc tuyên vô tội thì có thể xin bồi thường nhà nước. Số ngày bị giam được bồi thường theo ngày, từ ngày 18 tháng 5 năm 2026 là 495.94 tệ một ngày, một năm khoảng 18 vạn tệ. Hậu quả nghiêm trọng thì cộng thêm tiền bùi đắp tổn thương tâm. Tội nhẹ chỉ vì tình tiết nhẹ mà không khởi tố thì không bồi thường. Quá thời hiệu truy tố, chưa đủ tuổi, hoặc tự khai sai thì cũng không bồi thường. +- Lợi ích: Điều mười bảy của Luật Bồi thường nhà nước liệt kê vài tình huống mà người bị hại có quyền được bồi thường. Thứ nhất là bắt giữ: bắt giữ công dân trái quy định của Pháp luật Tố tụng Hình sự, hoặc bắt giữ hợp pháp nhưng quá thời hạn luật định. Sau đó có quyết định rút vụ, không khởi tố, hoặc bản án tuyên vô tội và chấm dứt truy cứu trách nhiệm hình sự. Thứ hai là bắt: sau khi áp dụng biện pháp bắt đối với công dân, có quyết định rút vụ, không khởi tố, hoặc bản án tuyên vô tội và chấm dứt truy cứu trách nhiệm hình sự. Thứ ba là xét lại: vụ án được xét lại theo thủ tục giám sát xét xử mà bản án cũ đã được thi hành thì tuyên vô tội. Thủ tục giám sát xét xử là thủ tục lấy bản án đã có hiệu lực ra xét lại. Điều ba mươi ba: khi xâm phạm tự do cá nhân của công dân thì tiền bồi thường mỗi ngày tính theo mức lương bình quân ngày của công nhân viên trong năm trước của nhà nước. Điều ba mươi lăm: khi gây tổn thương tâm thì phải xoá tác động của hành vi xâm phạm, khôi phục danh dự, xin lỗi cho người bị hại trong phạm vi hành vi xâm phạm ảnh hưởng, hậu quả nghiêm trọng thì phải trả tiền bùi đắp tổn thương tâm tương ứng. Điều mười chín liệt kê sáu tình huống nhà nước không bồi thường. Một, tự khai sai hoặc tạo chứng cứ có tội khác nên bị giam hoặc bị phạt tù. Hai, người không chịu trách nhiệm hình sự theo Điều mười bảy, Điều mười tám của Bộ luật Hình sự bị giam, ví dụ chưa đến tuổi chịu trách nhiệm hình sự, hoặc người bị bệnh tâm thần gây hại khi không phân biệt, không kiềm chế được hành vi của mình. Ba, người mà Pháp luật Tố tụng Hình sự quy định không truy cứu trách nhiệm hình sự bị giam. Điều này gồm tình tiết rõ ràng là nhỏ, hậu quả không lớn, không coi là tội phạm, đã quá thời hiệu truy tố, bị cáo đã tử vong. Cũng gồm tội nhẹ mà viện kiểm sát quyết định không khởi tố, tức là không khởi tố có điều kiện thường gọi là không khởi tố tương đối (cũng gọi là không khởi tố tuỳ quy chế). Sau khi hết thời gian thử thách không khởi tố có điều kiện với người chưa thành niên, và sau khi hoà giải với bị hại rồi không khởi tố, cũng nằm trong đó. Bốn, hành vi cá nhân của cán bộ xử án không liên quan đến việc thực hiện chức trách. Năm, thiệt hại do hành vi cố ý của chính mình như tự làm hại, tự gây thương tích. Sáu, các trường hợp khác theo quy định của pháp luật. Điều bảy của giải thích tư pháp hình sự năm 2015 của hai tòa cao nhất xác nhận lại: người không chịu trách nhiệm hình sự và người bị không khởi tố tương đối bị giam thì nhà nước không bồi thường. Nhưng sau khi khởi tố mà tòa tuyên oan tù giam, tù có thời hạn, tù chung thân và đã thi hành thì sau khi bản án có hiệu lực, thời gian bị giam tiếp vẫn phải bồi thường. Điều tám của giải thích đó: cơ quan xử án lấy lý do 「bạn tự khai sai」「bạn tự làm hại bản thân」 để không bồi thường thì cơ quan xử án phải chứng minh. Viện kiểm sát nhân dân tối cao đã ra thông báo tháng 5 năm 2026. Cục Thống kê Quốc gia công bố ngày 15 tháng 5 năm 2026: mức lương bình quân năm 2025 của người có việc làm tại đơn vị thành phố không tự trả cả nước khoảng 129.000 tệ（con số chính xác 129441 tệ）. Mức lương bình quân ngày tính ra là 495.94 tệ. Từ ngày 18 tháng 5 năm 2026, khi các cấp viện kiểm sát nhân dân ra quyết định bồi thường nhà nước thì tiền bồi thường cho việc xâm phạm tự do cá nhân của công dân tính 495.94 tệ mỗi ngày. Mức trước đó là 475.52 tệ mỗi ngày, áp dụng từ tháng 5 năm 2025 (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012 年修正). Luật Bồi thường nhà nước, 国家赔偿法（第十七、十九、三十三、三十五条）. （bản đăng lại của Cục Thống kê Quốc gia）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). Tiêu chuẩn mới nhất về tiền bồi thường xâm phạm tự do cá nhân của công dân. ；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2015). Giải thích một số vấn đề về áp dụng pháp luật khi xét các vụ án bồi thường hình sự, 关于办理刑事赔偿案件适用法律若干问题的解释（法释〔2015〕24 号，第七、八条）. ；Quốc hội (全国人大) (2018). Pháp luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第十六、一百七十七、一百八十一、二百八十四、二百九十条）. . Điều mười chín của Luật Bồi thường nhà nước và giải thích đó dẫn số điều của Pháp luật Tố tụng Hình sự năm 2012（Điều mười lăm, khoản hai Điều trăm bảy mươi ba, khoản hai Điều hai trăm bảy mươi ba, Điều hai trăm bảy mươi chín）, sau khi sửa đổi năm 2018 tương ứng là Điều mười sáu, khoản hai Điều trăm bảy mươi bảy, khoản hai Điều hai trăm tám mươi tư, Điều hai trăm chín mươi +- Ghi chú: Tiêu chuẩn tính theo ngày này được cập nhật mỗi tháng 5 theo mức lương bình quân năm của đơn vị thành phố không tự trả do Cục Thống kê Quốc gia công bố. Khi dùng thì tra con số của năm đó, đừng áp thẳng con số trong mục này. Nhận được quyết định không khởi tố, trước hết xem căn cứ ghi là gì. Ghi 「không có sự thật phạm tội」 hoặc 「không đủ chứng cứ」 thì có thể xin bồi thường. Ghi 「tội nhẹ」, hoặc đã quá thời hiệu truy tố thì nguyên tắc những ngày bị giam không được bồi thường. Không đồng ý với việc không khởi tố tương đối thì sau khi nhận quyết định có thể khiếu nại với viện kiểm sát trong 7 ngày. Tiền bùi đắp tổn thương tâm do cơ quan xử án ấn định theo hậu quả, không có công thức thống nhất. Việc bồi thường thuộc về cơ quan đã ra quyết định giam bạn（công an, viện kiểm sát hoặc tòa án）, đi tìm cơ quan đó trước. Không đồng ý về cách xử lý của nó thì theo thủ tục của Luật Bồi thường nhà nước đi lên cấp trên. Cách tính thật thì xem quyết định trong vụ Quốc Lợi ở mục 36 của phần này（bản thân là bên bị hại mà đi đòi bồi thường）. Quốc Lợi bị giam 1826 ngày. Tiền bồi thường tính theo tiêu chuẩn của năm ra quyết định, không theo năm bị giam, dùng mức lương bình quân ngày của công nhân viên cả nước năm 2016 là 258.89 tệ. Phần hạn chế tự do cá nhân được bồi thường khoảng 473.000 tệ（472733.14 tệ）, cộng thêm tiền bùi đắp tổn thương tâm khoảng 165.000 tệ（165450 tệ）. Mười vạn tệ tiền bùi đắp tổn thương tâm mà bản thân xin được 8888 vạn tệ, phần vượt không được chấp nhận. Cũng nên biết những khoản nào không được bồi thường: quyết định đó xác định tiền công mất việc, bồi thường về sức khoẻ, tổn thất tài sản, phí luật sư, tiền sinh hoạt trong trại tạm giam đều không thuộc phạm vi bồi thường của nhà nước. + +### 36. Bản thân là bên bị hại mà đi đòi bồi thường: gọi 12315, khởi kiện hoặc thuê luật sư; đừng một mình đi gặp đối phương, đừng gộp 「cho tiền」 với 「tôi không đăng」 thành một câu + +- Chi phí: Không tốn tiền. Gọi 12315, ra tòa khởi kiện, hoặc ủy thác luật sư đưa ra yêu cầu. Khi gặp nói chuyện thì ghi âm toàn bộ. Khó là nhịn được, không tự đi nói chuyện riêng với đối phương. +- Nói thẳng: Bản thân là bên bị hại mà đòi bồi thường bên xâm phạm thì có căn cứ pháp lý, không phải vì 「đòi nhiều」 là tự động thành tống tiền. Nhưng cái giá có thể rất lớn: Quốc Lợi đầu tiên bị tuyên năm năm tù vì tội tống tiền, sau đó mới được xử lại và tuyên vô tội. Vậy nên đi con đường có để lại dấu vết, đừng một mình tới nơi đối phương chỉ định để nói chuyện, đừng gộp 「cho tiền」 và 「tôi sẽ không đăng」 thành một điều kiện để nói ra. +- Lợi ích: Vụ án chính thức: sau khi năm 2008 cơ quan có liên quan công bố một số lô sữa bột 「施恩」 có chứa melamine, kết quả kiểm tra của con gái Quốc Lợi là 「双肾中央集合系统内可见数个点状强回声」. Quốc Lợi nhiều lần yêu cầu bồi thường từ nhà phân phối và công ty Thận Nhiệt, rồi đưa tin ra báo chí. Tháng 6 năm 2009 hai bên đạt thoả thuận, công ty Thận Nhiệt bồi thường 40 vạn tệ, Quốc Lợi viết văn bản xác nhận không còn truy cứu và từ bỏ yêu cầu bồi thường. Sau đó công ty báo công an, Quốc Lợi bị Tòa án nhân dân huyện Triều An tuyên năm năm tù vì tội tống tiền. Viện kiểm sát nhân dân tỉnh Quảng Đông đưa ra đề nghị xét lại. Đề nghị cho rằng 「chứng cứ hiện có không đủ để chứng minh Quốc Lợi có chủ ý chiếm đoạt」. Là người giám hộ hợp pháp, Quốc Lợi 「có quyền yêu cầu nhà sản xuất sữa bồi thường, hành vi đòi bồi thường của họ có căn cứ pháp lý」. Đề nghị còn viết, 「người tiêu dùng có quyền phê phán, chỉ trích qua các phương tiện truyền thông đại chúng hành vi làm tổn hại quyền lợi hợp pháp của họ」. Ngày 7 tháng 4 năm 2017, Tòa án nhân dân tỉnh Quảng Đông xét lại tuyên Quốc Lợi vô tội. Lý do là 「sự thật bản án xác định Quốc Lợi nhằm chiếm đoạt vô đoạt, dùng phương pháp đe dọa, uy hiếp để cưỡng ép chiếm đoạt tài sản thì không rõ ràng, chứng cứ không đủ, bản chất hành vi của Quốc Lợi không vượt quá phạm vi tranh chấp dân sự」. Quốc Lợi bị giam 1826 ngày, từ ngày 23 tháng 7 năm 2009 đến ngày 22 tháng 7 năm 2014 (Quảng Đông, xét lại năm 2017, tuyên vô tội) +- Mức bằng chứng: B +- Nguồn: Viện kiểm sát nhân dân tỉnh Quảng Đông (广东省人民检察院) (2017). Đề nghị của Viện kiểm sát nhân dân tỉnh Quảng Đông đề nghị sửa án, tuyên Quốc Lợi vô tội. （bản đăng lại của Viện kiểm sát nhân dân thành phố Hà Nguyên）；Ủy ban Bồi thường của Tòa án nhân dân tỉnh Quảng Đông (广东省高级人民法院赔偿委员会) (2018). Vụ Quốc Lợi xin bồi thường nhà nước với lý do được xét lại tuyên vô tội, Tòa án nhân dân trung cấp thành phố Triều Châu, tỉnh Quảng Đông bồi thường（（2018）粤委赔 11 号）. +- Ghi chú: Xếp mức B là vì đây chỉ là một vụ, không phải tỷ lệ có thể khái quát. Nó đồng thời nói rõ hai điều: bản thân việc đòi bồi thường là hợp pháp, và bạn vẫn có thể bị công an lập vụ để điều tra. Cách đi rủi ro thấp nhất là giao việc đòi bồi thường cho luật sư, hoặc ra tòa khởi kiện trực tiếp, bản thân không tiếp xúc riêng với đối phương. Đặc biệt cần cẩn trọng khi đối phương hẹn bạn tới nơi của họ để 「nói chuyện để giải quyết」, việc báo công an và khởi kiện thường xảy ra tại nơi của đối phương. Đã ký thoả thuận hoà giải thì làm theo thoả thuận, ký xong lại mở một vụ đòi bồi thường lớn khác thì dễ bị hiểu là 「đe dọa」. Sau khi bị khởi kiện thì xử lý thế nào, xem mục 5, 34 và 35 của phần này（mời luật sư trước, không đủ chứng cứ thì phải tuyên vô tội, bồi thường nhà nước）. Trường hợp đối phương nắm nhược điểm đòi tiền, xem mục 32 của phần này（đối phương dọa gọi công an, gửi ảnh để đòi tiền）. + +### 37. Bị tấn công trên mạng: bật chống bảo vệ và lưu giữ bằng chứng trước, rồi chọn trong ba con đường nền tảng, lệnh cấm, báo công an + +- Chi phí: Không tốn tiền. Tại chỗ dành mười mấy phút để chống bảo vệ và lấy chứng cứ. Nếu đi tới bước xin lệnh cấm hoặc tự khởi kiện thì phải tới tòa, có thể còn phải thuê luật sư. Khó là khi bị chửi thì đi lưu giữ bằng chứng trước, không đi cãi lại. +- Nói thẳng: Tại chỗ làm ba việc trước: bật phần chống bảo vệ của nền tảng, lưu nội dung bằng công cụ lấy chứng cứ nhanh, rồi báo cáo qua lối vào nhanh. Sau đó có thể xin tòa ra lệnh cấm để đối phương phải dừng tay trước. Cũng có thể tự ra tòa kiện, thu thập chứng cứ khó thì xin công an phối hợp. Hậu quả đặc biệt nghiêm trọng, ảnh hưởng lại xấu thì viện kiểm sát khởi tố, bạn không phải tự đưa ra chứng cứ. +- Lợi ích: Điều 15 của 法发〔2023〕14 号 nói về lệnh cấm. Chủ thể quyền lợi có chứng cứ chứng minh bên thực hiện, hoặc sắp thực hiện, hành vi vi phạm pháp xâm phạm quyền nhân thân của họ. Không ngăn kịp thì quyền lợi hợp pháp của chủ thể sẽ bị thiệt hại khó khắc phục. Chủ thể quyền lợi xin theo Điều chín trăm chín mươi bảy của Bộ luật Dân sự thì tòa án có thể theo pháp luật ra lệnh cấm xâm phạm quyền nhân thân. Lệnh cấm xâm phạm quyền nhân thân là một mệnh lệnh của tòa buộc đối phương dừng tay trước. Điều 11 của giải thích đó nói về việc thu thập chứng cứ. Trong vụ tự khởi kiện, 「被害人提供证据确有困难的，可以要求公安机关提供协助」. Cơ quan công an phải kịp thời làm rõ chủ thể hành vi, thu thập chứng cứ về tình hình lan truyền và mức độ ảnh hưởng. Nhà cung cấp dịch vụ Internet phải cung cấp hỗ trợ kỹ thuật cần thiết. 「经公安机关协助取证，达到自诉案件受理条件的，人民法院应当决定立案」. Không thu thập được thì cơ quan công an phải giải trình bằng văn bản. Vụ tự khởi kiện là vụ do bên bị hại tự ra tòa kiện. Điều 12 của giải thích đó liệt kê năm tình huống phải xác định là 「nghiêm trọng gây tổn hại trật tự xã hội」 mà khoản hai Điều hai trăm bốn mươi sáu của Bộ luật Hình sự nói đến. Một, gây hậu quả nghiêm trọng như bên bị hại hoặc người thân gần của họ bị rối loạn tâm thần, tự sát, và ảnh hưởng xã hội xấu. Hai, tùy tiện lấy công chúng thường làm đối tượng xâm phạm, thông tin lan truyền rộng rãi gây ra rất nhiều bình luận tục trào mời và độc ác, và ảnh hưởng xã hội xấu. Ba, xúc phạm vu khống nhiều người, hoặc nhiều lần tung ra, và ảnh hưởng xã hội xấu. Bốn, tổ chức, chỉ đạo người ta tung ra hàng loạt trên nhiều nền tảng, và ảnh hưởng xã hội xấu. Năm, các trường hợp khác. Điều 4 của giải thích đó: tổ chức 「人肉搜索」, thu thập trái phép thông tin cá nhân công dân và công bố cho số đông không xác định, tình tiết nghiêm trọng thì định tội và phạt theo tội xâm phạm thông tin cá nhân công dân. Điều hai mươi ba của Quy định quản lý thông tin bạo lực trên mạng yêu cầu nền tảng cung cấp các lựa chọn chống bảo vệ sau. Chặn người dùng lạ hoặc một người dùng cụ thể. Giới hạn phạm vi hiển thị của thông tin bạn đăng. Cấm chia sẻ lại hoặc bình luận. Chỉ nhận tin nhắn riêng của bạn bè, hoặc từ chối nhận mọi tin nhắn riêng. Điều hai mươi lăm yêu cầu có chức năng lấy chứng cứ nhanh. Điều hai mươi sáu yêu cầu đặt lối vào khiếu nại, báo cáo nhanh ở vị trí dễ thấy. Điều mười lăm yêu cầu nền tảng phát hiện nghi vấn vi phạm pháp luật thì báo công an kịp thời +- Mức bằng chứng: A +- Nguồn: Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院)、Bộ Công an (公安部) (2023). Hướng dẫn về việc trị phạt theo pháp luật các hành vi vi phạm pháp phạm tội bạo lực trên mạng, 关于依法惩治网络暴力违法犯罪的指导意见（法发〔2023〕14 号, mục 4, 11, 12 và 15）. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第九百九十七条）. Toàn văn đăng lại bởi Viện kiểm sát nhân dân tối cao ；Văn phòng Quản lý thông tin Internet quốc gia (国家互联网信息办公室)、Bộ Công an (公安部)、Bộ Văn hóa và Du lịch (文化和旅游部)、Tổng cục Phát thanh Truyền hình Quốc gia (国家广播电视总局) (2024). Quy định quản lý thông tin bạo lực trên mạng, 网络暴力信息治理规定（第十五、二十三、二十五、二十六条, có hiệu lực từ ngày 1 tháng 8 năm 2024）. Công báo Hội đồng Nhà nước +- Ghi chú: Xếp thứ tự theo 「nhanh và rẻ」: nền tảng nhanh nhất, rẻ nhất, lưu giữ bằng chứng và báo cáo trước. Tòa chậm nhất và đắt nhất, cuối cùng mới quyết định có kiện hay không. Cả ba con đường đều phải tự gánh chi phí trong quá trình. Lệnh cấm phải tự xin tòa. Tự khởi kiện là tự ra tòa đăng ký, tự lấy chứng cứ. Muốn viện kiểm sát tiếp tục, đổi sang tự khởi tố công thì phải trùng một trong năm tình huống nêu ở mục 12 của 法发〔2023〕14 号, không đủ thì chỉ có thể tự kiện. Có hai việc đừng làm. Một là không cãi lại, cãi lại sẽ biến bạn thành người bị xử phạt trong mục 16 của phần này（không chửi người, không bịa tin trên mạng）. Hai là không xoá nội dung trong tài khoản của bạn, đó là bằng chứng. Không chịu nổi thì gọi 12356, xem mục 25 ở phần 1. Yêu cầu nền tảng xoá thông tin cá nhân của bạn, xem mục 8 ở phần 14. Vì giúp người lạ mà bị quay, bị đăng lên mạng, xem [docs/遇到陌生人出事该不该停.md](../docs/遇到陌生人出事该不该停.md). +### 38. Con đường 「mua bảo hiểm cho người ta trước rồi mới ra tay」 bị pháp luật chặn kín từ đầu: không lấy được phân xu nào, tội cộng lại tội cố ý giết người và lừa bảo hiểm + +- Chi phí: Không tốn tiền. +- Nói thẳng: Bốn cánh cửa đều đóng. Mua bảo hiểm chết cho họ mà không báo người nhà thì hợp đồng vô hiệu. Bên mua bảo hiểm cố ý làm chết người được bảo hiểm thì công ty bảo hiểm không bồi thường. Cố ý giết người thừa kế thì mất quyền thừa kế. Bộ luật Hình sự còn tách riêng loại lừa bảo hiểm này, cộng lại với tội cố ý giết người để phạt, cao nhất là tử hình. Không lấy được phân xu nào. +- Lợi ích: Điều ba mươi tư của Luật Bảo hiểm: hợp đồng có điều kiện trả tiền bảo hiểm khi chết, nếu không được người được bảo hiểm đồng ý và công nhận số tiền bảo hiểm thì hợp đồng vô hiệu. Người được bảo hiểm là người được bảo vệ. Cha mẹ mua bảo hiểm cho con chưa thành niên thì không chịu hạn chế này. Điều ba mươi chín: khi bên mua bảo hiểm chỉ định người thụ hưởng thì phải được người được bảo hiểm đồng ý. Bên mua bảo hiểm là người bỏ tiền mua bảo hiểm, người thụ hưởng là người nhận tiền sau khi xảy ra chuyện. Điều bốn mươi ba quy định về cố ý gây hại. Bên mua bảo hiểm cố ý làm chết, làm thương tật hoặc làm bệnh cho người được bảo hiểm thì công ty bảo hiểm không phải trả tiền bảo hiểm. Bên mua bảo hiểm đã đóng đủ hơn hai năm thì hoàn lại giá trị tiền mặt cho người quyền lợi khác. Người thụ hưởng cố ý làm chết, làm thương tật, làm bệnh cho người được bảo hiểm, hoặc cố ý giết người được bảo hiểm mà không thành công thì người thụ hưởng đó mất quyền thụ hưởng. Khoản hai Điều hai mươi bảy: bên mua bảo hiểm, người được bảo hiểm cố ý tạo ra sự cố bảo hiểm thì công ty bảo hiểm có quyền hủy hợp đồng, không chịu trách nhiệm bồi thường hay trả tiền, trừ quy định tại Điều bốn mươi ba thì không hoàn tiền phí bảo hiểm. Điều một nghìn một trăm hai mươi lăm của Bộ luật Dân sự: cố ý giết người thừa kế thì mất quyền thừa kế. Khoản một điểm năm của Điều trăm chín mươi tám của Bộ luật Hình sự liệt kê 「bên mua bảo hiểm, người thụ hưởng cố ý làm chết, làm thương tật hoặc làm bệnh cho người được bảo hiểm để lừa tiền bảo hiểm」 là lừa đảo bảo hiểm. Số tiền lớn thì phạt dưới năm năm và phạt tiền từ 1 vạn đến 10 vạn tệ. Số tiền rất lớn thì phạt từ năm năm đến mười năm. Số tiền đặc biệt lớn thì phạt từ mười năm trở lên. Khoản hai: hành vi nêu trên đồng thời cấu thành tội phạm khác thì xử phạt theo quy định cộng nhiều tội, cộng nhiều tội là gộp mấy tội lại để tính án. Điều hai trăm ba mươi hai: cố ý giết người thì phạt tử hình, chung thân hoặc tù từ mười năm trở lên. Vụ việc là vụ Trương Tùng Thông ở Tuyên Thành, tỉnh An Huy. Rạng sáng ngày 18 tháng 7 năm 2016, anh lái xe lao xuống ao, vợ chết đuối. Chưa đầy nửa tháng trước khi xảy ra chuyện, anh đã mua bảo hiểm cho vợ 4 hợp đồng, mỗi hợp đồng 1 triệu tệ. Ngày 25 tháng 7 anh bị bắt. Tòa phạt theo cộng nhiều tội tội cố ý giết người và lừa đảo bảo hiểm, kết án tử hình treo và phạt tiền 5 vạn tệ +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015 修正). Luật Bảo hiểm của nước Cộng hòa Nhân dân Trung Hoa, 中华人民共和国保险法第二十七、三十四、三十九、四十三条. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự của nước Cộng hòa Nhân dân Trung Hoa, 中华人民共和国民法典第一千一百二十五条. ；Ủy ban Thường vụ Quốc hội (全国人大常委会). Bộ luật Hình sự của nước Cộng hòa Nhân dân Trung Hoa, bản hợp nhất sau khi sửa đổi lần thứ mười một, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第一百九十八条、第二百三十二条. ；Báo Pháp luật (《法制日报》) (2020-01-13), đăng lại bởi Viện kiểm sát nhân dân tỉnh An Huy (安徽省人民检察院转载). Giết vợ lừa bảo hiểm lộ ra vụ án trong vụ án（vụ án điển hình về giám sát vụ án dân sự giả của Viện kiểm sát nhân dân tỉnh An Huy）. +- Ghi chú: Người được lợi của mục này chính là bạn, mục đích là tránh tử hình hoặc chung thân. Số người chết và bản án trong vụ án lấy từ nguyên văn vụ án điển hình do cơ quan kiểm sát công bố, phần này không viết chi tiết nào ngoài các bước điều tra. Bịa chuyện tai nạn, thổi phóng thành tổn thất, các kiểu lừa bảo hiểm không liên quan đến mạng người vẫn là tội phạm, kể cả người làm chứng giúp thì cũng bị tính, xem mục 21 ở phần 9（bịa chuyện tai nạn, thổi phóng thành tổn thất để lừa bồi thường）. Mua bảo hiểm cho người nhà vốn là chuyện bình thường, cái cần ngăn là ý nghĩ 「mua bảo hiểm xong rồi mới ra tay」. Xử lý cơn thôi thúc muốn làm hại người nhà như một trường hợp cấp cứu, xem mục 14 và 15 của phần này（ý nghĩ tự nhiên xuất hiện trong mình, người bên cạnh nói lời đe dọa）. + +### 39. Báo công an thì lấy biên nhận tiếp nhận ngay tại chỗ, không khởi kiện thì phải có văn bản: trong 7 ngày xin xét lại, thêm 7 ngày xin xét lại lần hai, viện kiểm sát có thể yêu cầu công an khởi kiện + +- Chi phí: Không tốn tiền. Tại chỗ hỏi thêm một câu, xin được biên nhận tiếp nhận vụ án. Sau đó theo các mốc ngày nêu trên mà đi hỏi tiến độ. Khó là phải nhiều lần mở miệng hỏi. +- Nói thẳng: Báo công an thì phải lấy được biên nhận tiếp nhận ngay tại chỗ, đó là chứng cứ duy nhất cho thấy bạn đã báo. Công an quyết định không khởi kiện thì phải đưa cho bạn văn bản từ chối khởi kiện. Không đồng ý thì trong 7 ngày xin xét lại, không đồng ý nữa thì thêm 7 ngày xin xét lại lần hai. Bạn cũng có thể tìm viện kiểm sát, nhờ nó giám sát việc khởi kiện của công an. +- Lợi ích: Điều trăm bảy mươi mốt của Quy định về trình tự xử lý vụ án hình sự của cơ quan công an（sau khi sửa bằng Lệnh Bộ Công an số 159）quy định về biên nhận tiếp nhận vụ án. Nguyên văn là: 「公安机关接受案件时，应当制作受案登记表和受案回执，并将受案回执交扭送人、报案人、控告人、举报人」. Điều trăm bảy mươi tư: 「对接受的案件，或者发现的犯罪线索，公安机关应当迅速进行审查」. Điều trăm bảy mươi tám quy định về việc khởi kiện hay không. Sau khi tiếp nhận vụ án và xét xét, thấy có sự thật phạm tội cần truy cứu trách nhiệm hình sự và thuộc thẩm quyền của mình thì được khởi kiện sau khi người phụ trách cơ quan công an cấp tỉnh trở lên phê duyệt. Không khởi kiện thì vẫn phải đưa bạn một tờ giấy: 「对有控告人的案件，决定不予立案的，公安机关应当制作不予立案通知书，并在三日以内送达控告人」. Điều trăm bảy mươi chín quy định về khi không đồng ý thì làm sao. Nguyên văn là: 「控告人对不予立案决定不服的，可以在收到不予立案通知书后七日以内向作出决定的公安机关申请复议；公安机关应当在收到复议申请后三十日以内作出决定」. Tiếp theo viết: 「控告人对不予立案的复议决定不服的，可以在收到复议决定书后七日以内向上一级公安机关申请复核；上一级公安机关应当在收到复核申请后三十日以内作出决定。对上级公安机关撤销不予立案决定的，下级公安机关应当执行」. Vụ án lớn, phức tạp thì có thể kéo dài thời hạn xét lại, xét lại lần hai, nhưng 「不得超过三十日，并书面告知申请人」. Điều trăm mười hai của Pháp luật Tố tụng Hình sự: thấy có sự thật phạm tội cần truy cứu trách nhiệm hình sự thì phải khởi kiện; không khởi kiện thì 「将不立案的原因通知控告人。控告人如果不服，可以申请复议」. Điều trăm mười ba: 「被害人认为公安机关对应当立案侦查的案件而不立案侦查，向人民检察院提出的，人民检察院应当要求公安机关说明不立案的理由。人民检察院认为公安机关不立案理由不能成立的，应当通知公安机关立案，公安机关接到通知后应当立案」. Điều sáu mươi mốt của Quy định về trình tự xử lý vụ án hành chính của cơ quan công an cũng có yêu cầu như vậy. Vụ án thuộc thẩm quyền của đơn vị thì 「应当立即调查处理，制作受案登记表和受案回执，并将受案回执交报案人、控告人、举报人、扭送人」. Thời hạn xử lý vụ án hành chính 「自受理之日起不得超过 30 日；案情重大、复杂的，经上一级公安机关批准，可以延长 30 日」, thời gian giám định không tính vào thời hạn giải quyết. Điều chín của Quy định về thanh tra cơ quan công an quy định về việc khiếu nại công an. Nguyên văn là: 「督察机构对群众投诉的正在发生的公安机关及其人民警察违法违纪行为，应当及时出警，按照规定给予现场处置，并将处理结果及时反馈投诉人」（toàn quốc） +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2018). Pháp luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第一百一十二、一百一十三条）. ；Bộ Công an (公安部) (2020). Quyết định sửa đổi Quy định về trình tự xử lý vụ án hình sự của cơ quan công an（公安部令第 159 号）và toàn văn sau khi sửa. ；Bộ Công an (公安部) (2013/2019 修正). Quy định về trình tự xử lý vụ án hành chính của cơ quan công an（公安部令第 125 号, quyết định sửa đổi số 149 号）. 、；Hội đồng Nhà nước (国务院) (2011). Quy định về thanh tra cơ quan công an（国务院令第 603 号，第九条）. +- Ghi chú: Người được lợi của mục này chính là bạn. Báo công an không nhất thiết sẽ được xử lý, nhưng có xử lý hay không, xử lý trong bao lâu thì đều có tiêu chuẩn và thời hạn bằng văn bản. Ghi nhớ ba thao tác theo thứ tự. Một, lấy biên nhận tiếp nhận ngay tại chỗ, ghi lại số vụ án và đơn vị tiếp nhận. Hai, hết thời hạn mà chưa có kết quả thì theo đúng thời hạn nêu trên mà đi hỏi tiến độ, hỏi 「vụ án đang ở bước nào」, đừng hỏi 「sao không ai xử lý tôi」. Ba, nhận được văn bản từ chối khởi kiện thì xem lý do trước, trong 7 ngày xin xét lại, thêm 7 ngày xin xét lại lần hai, đồng thời có thể đề nghị Viện kiểm sát nhân dân giám sát việc khởi kiện, ba con đường không mâu thuẫn nhau. Vụ án hành chính là kiểu đánh nhau, trộm nhỏ. Thời hạn giải quyết tính từ ngày tiếp nhận, không quá 30 ngày. Vụ án lớn, phức tạp thì được cơ quan công an cấp trên phê duyệt để kéo dài thêm 30 ngày. Thời gian giám định không tính vào đó. Thái độ phục vụ tệ thì khiếu nại qua thanh tra, vụ án vẫn được đuổi theo đúng trình tự khởi kiện. Nhặt được đồ rơi, tranh chấp hàng xóm, kiểu không phải vụ án hình sự, thì cơ quan công an xử theo vụ án hành chính hoặc hoà giải, thậm chí chỉ đăng ký ghi nhận, đừng hiểu là 「báo công an mà không ai xử lý」. Sau khi bị lừa đảo thì quy trình dừng chi trả thế nào, xem mục 2 của phần này（gọi 110 ngay và yêu cầu dừng chi trả）. Bị tấn công trên mạng xem mục 37 của phần này（bị tấn công trên mạng: lưu giữ bằng chứng trước rồi chọn cách）. + +### 40. Đừng tặng tiền, tặng thẻ cho người xử án, người thi hành: hối người thì bản thân cũng bị phạt, hối cán bộ giám sát, cán bộ thi hành hành chính, nhân viên tư pháp thì còn bị phạt nặng hơn + +- Chi phí: Không tốn tiền. Chỉ là không làm. +- Nói thẳng: 「Một ít tiền để xong chuyện」 trong Bộ luật Hình sự là một tội độc lập, người bị bắt không chỉ là kẻ nhận tiền. Hối người thì phạt dưới 3 năm. Tình tiết nghiêm trọng thì phạt 3 đến 10 năm, đặc biệt nghiêm trọng thì phạt trên 10 năm hoặc chung thân. Hối cán bộ giám sát, cán bộ thi hành hành chính, nhân viên tư pháp vẫn là nhóm bị phạt nặng hơn theo văn bản. Bị ép mà đưa tiền, lại không nhận được lợi ích phi pháp thì không tính là hối người. +- Lợi ích: Điều ba trăm tám mươi chín của Bộ luật Hình sự định nghĩa hối người: 「为谋取不正当利益，给予国家工作人员以财物的，是行贿罪」, tức là để mong lợi ích phi pháp mà đưa của cải cho cán bộ, công chức nhà nước thì là tội hối người. Cùng điều còn viết: 「在经济往来中，违反国家规定，给予国家工作人员以财物，数额较大的，或者违反国家规定，给予国家工作人员以各种名义的回扣、手续费的，以行贿论处」, tức là trong quan hệ kinh tế mà vi phạm quy định của nhà nước, đưa cho cán bộ, công chức nhà nước tài sản có số tiền lớn, hoặc đưa hoa hồng, phí thủ dụch dưới danh nghĩa khác thì xử theo tội hối người. Điều đó cũng để một ngoại lệ: 「因被勒索给予国家工作人员以财物，没有获得不正当利益的，不是行贿」, tức là vì bị ép mà đưa tài sản cho cán bộ, công chức nhà nước mà không nhận được lợi ích phi pháp thì không phải hối người. Mức phạt ghi ở Điều ba trăm chín mươi của Bộ luật Hình sự（sửa bằng Sửa đổi thứ mười hai Bộ luật Hình sự, có hiệu lực từ ngày 1 tháng 3 năm 2024）. Nguyên văn là: 「对犯行贿罪的，处三年以下有期徒刑或者拘役，并处罚金；因行贿谋取不正当利益，情节严重的，或者使国家利益遭受重大损失的，处三年以上十年以下有期徒刑，并处罚金；情节特别严重的，或者使国家利益遭受特别重大损失的，处十年以上有期徒刑或者无期徒刑，并处罚金或者没收财产」. Có bảy tình huống phải phạt nặng hơn: 「（一）多次行贿或者向多人行贿的；（二）国家工作人员行贿的；（三）在国家重点工程、重大项目中行贿的；（四）为谋取职务、职级晋升、调整行贿的；（五）对监察、行政执法、司法工作人员行贿的；（六）在生态环境、财政金融、安全生产、食品药品、防灾救灾、社会保障、教育、医疗等领域行贿，实施违法犯罪活动的；（七）将违法所得用于行贿的」. Khoản cuối để lối giảm nhẹ khi chủ động khai báo: 「行贿人在被追诉前主动交待行贿行为的，可以从轻或者减轻处罚。其中，犯罪较轻的，对调查突破、侦破重大案件起关键作用的，或者有重大立功表现的，可以减轻或者免除处罚」(toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会). Bộ luật Hình sự của nước Cộng hòa Nhân dân Trung Hoa, bản hợp nhất sau khi sửa đổi lần thứ mười một, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第三百八十九条. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Sửa đổi Bộ luật Hình sự của nước Cộng hòa Nhân dân Trung Hoa（十二）（điểm năm, sửa Điều ba trăm chín mươi của Bộ luật Hình sự; có hiệu lực từ ngày 1 tháng 3 năm 2024）. +- Ghi chú: Người được lợi của mục này chính là bạn. Đừng trông chờ hối một lần là né được phạt. Bên kia có đồng nghiệp, có sổ sách, có camera giám sát và thanh tra, thứ bạn đưa ra chính là bằng chứng khi họ bị điều tra, cũng là bằng chứng tội của bạn, hơn nữa loại hối người này là nhóm bị phạt nặng hơn theo văn bản. Thật sự gặp người đòi lợi ích thì đi theo con đường ở mục 39 của phần này（báo công an thì lấy biên nhận tiếp nhận ngay tại chỗ）: giữ lại biên nhận tiếp nhận và chứng từ, báo lên cơ quan thanh tra hoặc cơ quan kỷ luật đảng. Khoản ba Điều ba trăm tám mươi chín của Bộ luật Hình sự đặt điều kiện miễn trách là 「bị ép và không nhận được lợi ích phi pháp」, một khi bạn nhân cơ hội đó để lấy thứ vốn không lấy được thì điều kiện không còn. Phong bao đỏ trong bệnh viện thuộc bộ quy tắc khác, xem mục 12 ở phần 24. +### 41. Cuộc gọi và cuộc gặp có thể đổi thái độ thì bật ghi âm ngay: cuộc nói chuyện bạn tham gia thì không cần xin phép đối phương trước + +- Chi phí: Không tốn tiền. Bật ghi âm cuộc gọi trên điện thoại hoặc dùng ứng dụng ghi âm là được. Ghi từ đầu đến cuối, xong thì giữ nguyên tệp gốc, không cắt ghép, không ghi đè. +- Nói thẳng: Chuyện vay tiền, nợ tiền, bồi thường, hoàn tiền, tranh chấp lao động, nói rõ trên điện thoại mà không để lại gì thì vài tháng sau hai bên có thể kể mỗi người một cách. Cuộc nói chuyện bạn tham gia thì ghi lại được, không báo trước cho đối phương cũng không vì thế mà một mình không làm được chứng cứ. Cái không được làm là nghe lén, theo dõi đời tư của người khác, xông vào nhà họ để ghi trộm, hoặc đe dọa đối phương phải nói. +- Lợi ích: Điều sáu mươi sáu của Pháp luật Tố tụng Dân sự liệt kê tư liệu hình ảnh, âm thanh và dữ liệu điện tử đều là chứng cứ. Giải thích của Tòa án nhân dân tối cao về Pháp luật Tố tụng Dân sự, Điều trăm linh sáu, chỉ loại trừ chứng cứ thu thập bằng ba phương pháp: làm tổn hại nghiêm trọng quyền lợi hợp pháp của người khác, vi phạm quy định cấm của pháp luật, vi phạm nghiêm trọng trật tự công cộng và đạo đức. 「Không báo trước cho bên kia」 không nằm trong ba phương pháp đó. Quy định của Tòa án nhân dân tối cao về chứng cứ, Điều mười lăm: dùng tệp ghi âm, ghi hình làm chứng cứ thì phải nộp phương tiện gốc có lưu tệp đó, ví dụ chiếc điện thoại ban đầu. Điều chín mươi: tư liệu hình ảnh, âm thanh và dữ liệu điện tử có điểm đáng ngờ thì không được dùng một mình để xác định sự thật. Một bài viết của thẩm phán đăng lại trên trang Tòa án nhân dân tối cao cũng nói như vậy: tệp ghi âm phải được thu thập hợp pháp, nội dung thật, không cắt ghép, và phải liên quan tới vụ án (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 年修正, có hiệu lực từ ngày 1 tháng 1 năm 2024). Pháp luật Tố tụng Dân sự, 民事诉讼法（第六十六条）. Ủy ban Phát triển và Cải cách thành phố Thượng Hải đăng lại toàn văn ；Tòa án nhân dân tối cao (最高人民法院) (2022 年第二次修正). Giải thích về việc áp dụng Pháp luật Tố tụng Dân sự của nước Cộng hòa Nhân dân Trung Hoa, 关于适用《中华人民共和国民事诉讼法》的解释（第一百零六条）. ；Tòa án nhân dân tối cao (最高人民法院) (2019 年修正). Một số quy định về chứng cứ trong tố tụng dân sự, 关于民事诉讼证据的若干规定（第十四、十五、九十条）. ；Lưu Đan Ni (2023). Pháp luật trong phim 「Cô Ấy Đã Biến Mất」 (báo Tòa án nhân dân, đăng lại trên trang Tòa án nhân dân tối cao). +- Ghi chú: Người được lợi là bạn và gia đình bạn. Trong tệp ghi âm nên để người, việc, số tiền và thời gian khớp nhau. Đừng chỉ hỏi 「khi nào bạn trả tôi」, có thể hỏi tự nhiên 「khoản 5 vạn tệ bạn cho tôi mượn năm ngoái, bạn định trả khi nào」. Tệp ghi âm đưa cho tòa, hội đồng trọng tài hoặc công an dùng thì được, đừng đăng lên mạng, công khai lời của người khác có thể dính vào tranh chấp quyền riêng tư và danh dự, xem mục 16 của phần này（không chửi người, không bịa tin trên mạng）. Có tệp ghi âm không có nghĩa là chắc thắng, tòa còn cân nhắc chứng cứ khác xem sự thật và mức độ liên quan. Tệp ghi âm chỉ giải quyết vấn đề có dùng được để chứng minh hay không, có thật sự kiện tụng, tốn bao lâu thì phải tính riêng. Tài liệu cần lưu trước khi nghỉ việc xem mục 8 ở phần 19（trước khi nghỉ việc thì lưu phiếu lương, bản chấm công）. Căn cứ của mục này là quy tắc tố tụng dân sự, cách dùng chứng cứ trong vụ án hình sự và hành chính có quy định khác. + +### 42. Tại hiện trường có chuyện thì chụp toàn cảnh trước, chụp quan hệ vị trí sau, cuối cùng chụp hư hại và vết thương, ảnh gốc và video gốc đừng xoá + +- Chi phí: Không tốn tiền. Dùng điện thoại quay liên tục vài chục giây tới vài phút. Trước hết bảo đảm an toàn cho bạn, không làm chậm việc cứu người và xử lý. +- Nói thẳng: Chỉ chụp cận cảnh một vết xước trên xe hay một món đồ trên mặt đất thì vài ngày sau không ai biết nó vốn nằm ở đâu. Hãy quay một đoạn toàn cảnh, lấy vào khung ngã ba, biển số nhà, xe, người và xung quanh, rồi chụp vị trí tương đối của chúng, cuối cùng chụp chi tiết. Phải di chuyển đồ vật hay xe thì nên chụp trước. Giữ ảnh gốc và video gốc trong điện thoại, đừng chỉ giữ bản đã qua ứng dụng nhắn tin. +- Lợi ích: Điều sáu mươi sáu của Pháp luật Tố tụng Dân sự liệt kê vật chứng, tư liệu hình ảnh, âm thanh và dữ liệu điện tử đều là chứng cứ. Điều mười bốn của quy định về chứng cứ của Tòa án nhân dân tối cao liệt kê hình ảnh, âm thanh, video là dữ liệu điện tử. Điều mười lăm yêu cầu nộp bản gốc của dữ liệu điện tử, dùng video làm chứng cứ thì phải nộp phương tiện gốc. Ảnh và video giữ lại được trạng thái hiện trường sắp thay đổi, rồi đối chiếu với biên bản báo công an, hồ sơ sửa chữa, hồ sơ bệnh án, lời kể của nhân chứng. Thứ tự 「toàn cảnh, vị trí, chi tiết」 là kinh nghiệm, pháp luật không quy định (toàn quốc) +- Mức bằng chứng: C +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 年修正, có hiệu lực từ ngày 1 tháng 1 năm 2024). Pháp luật Tố tụng Dân sự, 民事诉讼法（第六十六条）. Ủy ban Phát triển và Cải cách thành phố Thượng Hải đăng lại toàn văn ；Tòa án nhân dân tối cao (最高人民法院) (2019 年修正). Một số quy định về chứng cứ trong tố tụng dân sự, 关于民事诉讼证据的若干规定（第十四、十五条）. +- Ghi chú: Người được lợi là bạn và gia đình bạn. Để xếp mức C là vì thứ tự chụp chỉ là kinh nghiệm, điều luật chỉ quy định ảnh, video có thể làm chứng cứ và phải nộp bản gốc. Có người bị thương, hoả hoạn, rò rỉ khí hoặc còn nguy hiểm khác thì cứu người, tránh đi, đừng để chuyện chụp ảnh làm chậm. Việc phải làm tại hiện trường tai nạn giao thông xem mục 1 của phần này（tai nạn giao thông thì dừng xe, cứu người, báo công an trước）. + +### 43. Bị bạo hành gia đình: trước hết báo công an để có biên bản ra hiện trường, rồi ra tòa xin lệnh bảo vệ an toàn cá nhân, không cần ly hôn trước, cũng không mất phí + +- Chi phí: Không tốn tiền. Xin lệnh bảo vệ an toàn cá nhân không phải nộp phí tố tụng, cũng không phải đặt bảo lãnh. Cái phải bỏ ra là thời gian báo công an, khám thương, viết đơn. Khó nhất là mở lời với người nhà. +- Nói thẳng: Trong số phụ nữ bị giết ở khắp thế giới, hơn một phần ba do chồng hoặc bạn đời ra tay, thường là kết cục của bạo hành kéo dài. Trong số phụ nữ bị đánh, người từng bị bóp cổ có xác suất bị giết sau cao hơn hơn 7 lần so với người chưa từng bị bóp cổ. Vậy nên bị đánh thì báo công an, rồi ra tòa xin lệnh bảo vệ. Tòa thường phải trả lời trong 72 giờ. +- Lợi ích: Trước hết nói về rủi ro. Gộp dữ liệu của 66 nước lại（tổng quan hệ thống kê）. Trong các vụ phụ nữ bị giết, 38.6% do bạn đời thân mật gây ra, ở nam giới là 6.3%. Tác giả nói loại án mạng người như vậy thường là bước cuối của một thời gian dài ngược đãi. Một nghiên cứu khác so sánh 506 vụ giết người và cố ý giết người, cùng 427 phụ nữ bị đánh nhưng không xảy ra chuyện đó. Người trước đây từng bị bạn đời bóp cổ có xác suất bị cố ý giết cao gấp 6.70 lần so với người chưa từng bị bóp cổ (OR 6.70, 95% CI 3.91–11.49). Xác suất bị giết cao gấp 7.48 lần (OR 7.48, 95% CI 4.53–12.35). Nói tiếp về pháp luật. Người bị hại và người thân gần có quyền báo công an. Cơ quan công an phải tới nơi kịp thời, ngăn bạo hành gia đình, điều tra thu thập chứng cứ, phối hợp chăm sóc y tế và khám thương. Tình tiết nhẹ thì cơ quan công an có thể ra văn bản cảnh cáo đối phương. Chịu bạo hành gia đình hoặc đối mặt với nguy cơ bạo hành gia đình có thật thì xin tòa lệnh bảo vệ an toàn cá nhân. Tòa phải ra quyết định trong 72 giờ sau khi tiếp nhận, tình huống khẩn cấp thì trong 24 giờ. Lệnh có thể cấm đối phương gây bạo hành, cấm quấy rối, theo dõi, tiếp xúc bạn và người thân gần của bạn, còn có thể bắt đối phương dọn ra khỏi nhà bạn. Lệnh có hiệu lực tối đa 6 tháng, trước khi hết hạn thì có thể xin gia hạn. Xin lệnh không phải điều kiện là phải ly hôn. Tòa cho rằng bạo hành gia đình 「存在较大可能性」 là có thể cấp. Đối phương vi phạm lệnh thì tòa có thể phạt dưới 1000 tệ, bị giữ dưới 15 ngày. Cấu thành tội phạm thì xử theo tội không thi hành án, quyết định của tòa. Xúc phạm, đe dọa, theo dõi, quấy rối thường xuyên, cùng với bị bỏ đói, bị đóng băng, đều tính là bạo hành gia đình. +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). Luật Chống bạo hành gia đình, 反家庭暴力法（第二、十三、十五、十六、十九、二十三、二十八、二十九、三十、三十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2022). Một số quy định về áp dụng pháp luật khi xét các vụ án lệnh bảo vệ an toàn cá nhân, 关于办理人身安全保护令案件适用法律若干问题的规定（法释〔2022〕17 号，第一、三、六、十、十二条）. ；Tòa án nhân dân tối cao (最高人民法院) (2016). Phúc trả về các vấn đề thủ tục về vụ án lệnh bảo vệ an toàn cá nhân（法释〔2016〕15 号，第一、二项）. Toàn văn đăng lại bởi Tòa án nhân dân huyện Bình Nguyên ；Stöckl H và cộng sự (2013). Mức độ phổ biến toàn cầu của giết người bạn đời thân mật: một tổng quan hệ thống kê. Lancet. ；Glass N và cộng sự (2008). Bóp cổ không gây tử vong là một yếu tố nguy cơ quan trọng với giết người phụ nữ. Journal of Emergency Medicine. +- Ghi chú: Người được lợi chủ yếu là bạn và gia đình bạn. Tỷ lệ các vụ giết người là số liệu nước ngoài, nước ta không công bố thống kê cùng cách đo. Cần tích góp chứng cứ: biên bản ra hiện trường, văn bản cảnh cáo, kết quả khám thương, hồ sơ khám bệnh, bản cam kết do đối phương viết, lịch sử trò chuyện và tệp ghi âm đều dùng được. Tệp ghi âm xem mục 41 của phần này（ghi âm）. Xin lệnh bảo vệ không cần thuê luật sư, thường vài ngày là có kết quả. Muốn ly hôn hoặc muốn đòi bồi thường là một vụ kiện khác, bản án sơ thẩm theo thủ tục thông thường tốn ít nhất 6 tháng, phí luật sư tự bỏ ra. Người bị bạo hành gia đình có thể xin trợ giúp pháp lý, tòa cũng có thể thu phí tố tụng chậm, giảm hoặc miễn. Lệnh bảo vệ không có nghĩa là an toàn cá nhân đã được bảo đảm, trước mắt có nguy hiểm thì gọi 110 trước. Người bị ép hoặc hoảng sợ, không thể tự xin, thì người thân gần, cơ quan công an, hội liên hiệp phụ nữ, ủy ban nhân dân khu phố, ủy ban thôn có thể xin thay. Nếu người bị hại là bạn bè hoặc đồng nghiệp thì bạn có thể đưa họ đi báo công an, đi bệnh viện. Thấy trong nhà người lạ đang có người bị đánh thì gọi 110 là đủ. Đừng tự xông vào can ngăn, lý do xem mục 10 của phần này（xung đột thì báo công an trước）. + +### 44. Người nhà cờ bạc nợ nần, đừng vội trả thay: nợ cờ bạc pháp luật không bảo vệ, tiền vay để cờ bạc cũng không phải nợ chung của vợ chồng + +- Chi phí: Không tốn tiền. Khó là giữ được sự kiên nhẫn khi chủ nợ tìm đến cửa, người nhà khóc lóc cầu xin mà vẫn không rút tiền ra. Chủ nợ thật sự kiện thì bạn phải ra tòa. +- Nói thẳng: Số tiền nợ trên bàn cờ bạc thì tòa không bảo vệ. Chủ nợ biết người kia vay tiền để cờ bạc mà vẫn cho vay thì số tiền đó tòa cũng không bảo vệ. Một bên vay tiền để cờ bạc thì không tính là khoản nợ hai vợ chồng cùng trả. Chưa qua xét xử thì tòa không thể bắt bên không vay phải chịu trách nhiệm hộ. +- Lợi ích: Thông báo năm 2017 của Tòa án nhân dân tối cao viết bốn điểm. Nợ phát sinh trong các hoạt động vi phạm pháp luật như cờ bạc, nghiện ma túy thì không được pháp luật bảo vệ. Chủ nợ biết hoặc phải biết người kia vay tiền để cờ bạc, nghiện ma túy mà vẫn cho vay thì không được pháp luật bảo vệ. Một bên vay tiền dưới tên mình rồi dùng vào việc vi phạm pháp luật, yêu cầu xử như nợ chung của vợ chồng thì tòa không chấp nhận. Chưa qua thủ tục xét xử thì không thể yêu cầu bên không vay chịu trách nhiệm dân sự (toàn quốc, từ năm 2017) +- Mức bằng chứng: B +- Nguồn: Tòa án nhân dân tối cao (最高人民法院) (2017). Thông báo về việc xử lý đúng quy củ các vụ án liên quan đến nợ vợ chồng, 关于依法妥善审理涉及夫妻债务案件有关问题的通知（法〔2017〕48 号，第二、四条）. +- Ghi chú: Để xếp mức B là vì đây là văn bản nội bộ của tòa, không có con số, chủ nợ biết hay không biết cũng phải xem chứng cứ. Thông báo này ban hành theo Luật Hôn nhân và gia đình thời đó, sau khi Bộ luật Dân sự có hiệu lực thì nhận diện nợ chung của vợ chồng thế nào, xem mục 12 ở phần 10（vợ chồng vay một khoản lớn）. Khoản vay mà chủ nợ không biết là để cờ bạc thì vẫn phải trả, chỉ là người vay tự trả. Muốn chứng minh số tiền đã đi vào cờ bạc thì cần có sao kê chuyển tiền, lịch sử trên nền tảng cờ bạc, hoặc hồ sơ xử lý của công an, nên giữ lại từ trước. Chủ nợ thật sự kiện thì bản án sơ thẩm theo thủ tục thông thường tốn ít nhất 6 tháng, thủ tục giản dị là 3 tháng. Phí luật sư mỗi người trả phần mình, bên thua không gánh. Chủ nợ tìm người vây khốn, giữ người thì xem mục 15 ở phần 9（đòi nợ không giữ người）, báo công an ngay. Nếu chính người đó đã cờ bạc tới mức muốn chết thì xem mục 37 ở phần 1（cờ bạc tới mức vay tiền rồi vẫn cờ thì gọi 12356 trước）. Người được lợi chủ yếu là vợ chồng và người thân trực hệ. \ No newline at end of file diff --git "a/vi/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/vi/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" new file mode 100644 index 00000000..3d79d203 --- /dev/null +++ "b/vi/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -0,0 +1,183 @@ +[← Về mục lục](../README.md) + +# 10. Yêu đương và kết hôn có đáng không + +Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bảng tính riêng: tiền, thời gian, sức khoẻ, rủi ro pháp lý và chất lượng quan hệ, rồi tính từng cái một. Mỗi bảng chỉ dùng con số có tra ra được nguồn, và nguồn chỉ giới hạn ở thống kê chính thức hoặc một tổng hợp phân tích (meta-analysis) gộp dữ liệu của nhiều nghiên cứu rồi tính lại. Không có đánh giá đạo đức nào ở đây. + +### 1. Quen thêm nhiều người thay vì bám lấy một người: có hợp nhau hay không thì không đoán được trước khi gặp + +- Chi phí: Không tốn tiền. Thời gian dùng để quen thêm vài người, đừng nghĩ đi nghĩ lại về cùng một người. Khó nhất là nhịn không bám lấy một người. +- Nói thẳng: Hai nghiên cứu speed date bắt người tham gia điền hơn 100 câu hỏi chọn bạn đời trước khi gặp. Câu hỏi chỉ tính ra được ai nói chung dễ thích người khác, ai nói chung được nhiều người thích, không tính ra được giữa hai người này có nảy sinh cảm tình hay không. Có hợp nhau hay không thì gặp mới biết. Cho nên quen thêm vài người đáng hơn việc đặt cược hết vào một người. +- Lợi ích: Trong hai nghiên cứu speed date, người tham gia trước khi gặp điền hơn 100 bảng tự đánh giá liên quan đến chọn bạn đời (bảng hỏi tự chấm điểm). Các nhà nghiên cứu dùng mô hình rừng ngẫu nhiên (một thuật toán tìm quy tắc) để tính. Có thể tính ra「một người nói chung dễ thích người khác đến đâu」, những câu hỏi này giải thích được 4%–18% phương sai trong đó. Phương sai là sự khác biệt giữa người với người, câu hỏi chỉ giải thích được một phần nhỏ. Cũng tính ra được「một người nói chung được nhiều người thích đến đâu」, giải thích được 7%–27%. Nhưng「giữa hai người cụ thể này có nảy sinh cảm tình hay không」không tính ra được, dù gộp đặc điểm và sở thích trước buổi hẹn thế nào cũng không tính ra được. Cùng nhóm nghiên cứu này trước đó cũng làm một thí nghiệm speed date. Người tham gia tự báo loại hình lý tưởng trước buổi hẹn, ví dụ coi trọng ngoại hình hay coi trọng triển vọng kinh tế. Sở thích đó không dự báo được mức quan tâm của họ đối với một người cụ thể tại chỗ +- Mức bằng chứng: A +- Nguồn: Joel S, Eastwick PW, Finkel EJ (2017). Is romantic desire predictable? Machine learning applied to initial romantic attraction. Psychological Science, 28(10), 1478-1489. ; Eastwick PW, Finkel EJ (2008). Sex differences in mate preferences revisited: do people know what they initially desire in a romantic partner? Journal of Personality and Social Psychology, 94(2), 245-264. +- Ghi chú: Hai nghiên cứu chỉ nói「trước khi gặp thì không tính ra được」, không nói「cố thế nào cũng vô ích」. Có hợp nhau hay không thì gặp mới biết, cho nên bỏ thời gian quen thêm vài người, mỗi người gặp thật sự một lần, đáng hơn việc đặt cược hết vào một người. Đối tượng nghiên cứu đều là sinh viên đại học Âu Mỹ. Môi trường yêu đương ở Trung Quốc khác, áp lên mình thì phải giảm kỳ vọng. + +### 2. Dừng lại sau khi đối phương từ chối rõ ràng: cứ bám theo tiếp là vi phạm trật tự an ninh, không phải「chân thành」 + +- Chi phí: Không tốn tiền, cũng không tốn thời gian. Khó nhất là nhịn: không nhắn tin nữa, không đi đợi nữa. +- Nói thẳng: Đối phương đã nói rõ là từ chối, mà bạn vẫn nhắn tin vẫn đi đợi, thì từ tháng 1 năm 2026 trở đi là vi phạm trật tự an ninh. Có thể bị giữ dưới 5 ngày, hoặc phạt dưới 1000 tệ. Tình tiết khá nặng thì giữ từ 5 đến 10 ngày. Công an còn có thể ra lệnh cấm bạn tiếp cận đối phương, vi phạm thì giữ thêm từ 5 đến 10 ngày. +- Lợi ích: Luật Xử phạt hành chính về trật tự an ninh công có hiệu lực từ ngày 1 tháng 1 năm 2026 tách việc bám theo, theo dõi thành một điều riêng. Căn cứ ghi là「多次发送淫秽、侮辱、恐吓等信息或者采取滋扰、纠缠、跟踪等方法，干扰他人正常生活的」. Hành vi này xử phạt giữ người dưới năm ngày, hoặc phạt tiền dưới một nghìn tệ. Tình tiết khá nặng thì giữ người từ năm đến mười ngày, đồng thời có thể phạt thêm dưới một nghìn tệ. Điều này còn thêm một tầng: được người phụ trách cơ quan công an phê duyệt, có thể lệnh cho người vi phạm trong một thời hạn nhất định không được tiếp cận người bị xâm hại. Vi phạm lệnh cấm tiếp cận thì giữ người từ năm đến mười ngày nữa +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025 sửa đổi). Luật Xử phạt hành chính về trật tự an ninh công, 治安管理处罚法（第五十条第一款第五项、第二款）. （Viện kiểm sát nhân dân tối cao đăng lại） +- Ghi chú:「Bám theo đến cùng rồi đối phương sẽ cảm động」không có bất kỳ nghiên cứu nào ủng hộ, nhưng lệnh giữ người và lệnh cấm tiếp cận thì ghi trắng đen trong văn bản. Hồ sơ để lại còn ảnh hưởng đến thi công chức, thi biên chế, kiểm tra chính trị, và cả một số chứng chỉ hành nghề của vài ngành. Đường ranh giữa theo đuổi một người và quấy rối một người, không nhìn bạn nghĩ gì, mà nhìn đối phương đã từng nói rõ từ chối hay chưa. + +### 3. Muốn biết đối phương có hứng thú hay không thì xem hành vi, đừng xem「tín hiệu」: ai chủ động, ai sắp xếp, ai theo sau + +- Chi phí: Không tốn tiền, cũng không tốn bao nhiêu thời gian. Khó nhất là đổi thói quen: đừng nghĩ đi nghĩ lại nét mặt và giọng điệu của đối phương, thay vào đó ghi lại sự thật. +- Nói thẳng: Cho 52 cặp nam nữ lạ nhau nói chuyện chừng mười phút, rồi mỗi bên tự đánh giá xem đối phương có đang thả thính không. Kết quả là「 đối phương không thả thính」đoán khá đúng, còn「đối phương đang thả thính」lại không nhận ra. Chuyển sang 261 người xem video rồi phán, kết luận y vậy. Cho nên đừng đoán nét mặt và giọng điệu, hãy xem hành vi: ai chủ động mở lời, ai sắp xếp lần gặp sau, khi bạn không đẩy tiếp thì đối phương có theo sau không. +- Lợi ích: Hai nghiên cứu đo độ chính xác của người khi nhận ra thả thính. Nghiên cứu thứ nhất cho 52 cặp khác giới không quen nhau nói chuyện với nhau từ 10 đến 12 phút. Sau khi nói xong, mỗi người tự báo mình có đang thả thính không, và báo xem cảm thấy đối phương có thả thính không. Kết quả là「những tương tác không có thả thính thì phán đúng hơn, còn tương tác có thả thính lại phán kém hơn」. Nghiên cứu thứ hai dùng 26 đoạn video mỗi đoạn một phút, cho 261 người quan sát phán, kết luận nhất quán. Người ta nhìn ra được khá đáng tin「đối phương không thả thính」, nhưng rất khó nhìn ra「đối phương đang thả thính」 +- Mức bằng chứng: B +- Nguồn: Hall JA, Xing C, Brooks S (2015). Accurately detecting flirting: error management theory, the traditional sexual script, and flirting base rate. Communication Research, 42(7), 939-958. ; Perilloux C, Kurzban R (2015). Do men overperceive women's sexual interest? Psychological Science, 26(1), 70-77. （quan điểm bị tranh cãi） +- Ghi chú:Tranh cãi. Có nghiên cứu khác cho rằng không phải đàn ông đánh giá quá cao ý định tình dục của phụ nữ, mà là phụ nữ khi điền问卷 báo ý định của mình thấp đi. Rốt cuộc ai phán đúng hơn thì vẫn chưa có kết luận. Nhưng hai phe đều đồng ý một chuyện: suy ngược ra suy nghĩ của đối phương qua những tín hiệu nhỏ như nét mặt, giọng điệu thì độ chính xác rất thấp. Hành vi nhìn thấy được thì tin cậy hơn nhiều: đối phương có chủ động mở lời không, có sắp xếp lần gặp sau không, khi bạn không đẩy tiếp thì đối phương có theo sau không. + +### 4. Quan hệ tốt hay không chủ yếu do cảm nhận của chính bạn trong quan hệ đó, không phải do điều kiện của đối phương + +- Chi phí: Không tốn tiền, cũng không tốn bao nhiêu thời gian. Khó nhất là đổi góc nhìn: cất danh sách chọn bạn đời đi, thay vào đó xem cách hai người sống cùng nhau. +- Nói thẳng: Điều nói lên quan hệ tốt hay không rõ nhất là cảm nhận của chính bạn, không phải điều kiện của đối phương. Ví dụ bạn thấy đối phương nghiêm túc hay không, có coi trọng bạn hay không, hai người cãi nhau nhiều hay không. Cảm nhận của chính bạn hữu dụng hơn lời của bạn đời từ hai đến bốn lần. Nhưng quan hệ này về sau khá lên hay xấu đi thì hầu như bộ câu hỏi nào cũng không tính ra được. +- Lợi ích: Phân tích máy học trên 29 phòng thí nghiệm và 43 bộ dữ liệu cặp đôi theo dõi dài hạn, tức là ghi lại cùng một nhóm cặp đôi suốt quá trình. Những mục dự báo chất lượng quan hệ mạnh nhất đều đến từ cảm nhận của chính người trong cuộc đối với quan hệ này: cảm nhận được sự cam kết của đối phương, sự coi trọng, mức hài lòng tình dục, cảm nhận đối phương hài lòng, xung đột. Trong các mục liên quan đến cá nhân, mạnh nhất là mức hài lòng cuộc sống, cảm xúc tiêu cực, trầm cảm, gắn kết tránh né, gắn kết lo âu. Các mục liên quan đến quan hệ này lúc mới bắt đầu khảo sát giải thích được nhiều nhất 45% phương sai. Phương sai là độ khác biệt về chất lượng quan hệ giữa các cặp với nhau, 45% là phần nói rõ được bằng các mục đó. Đến lúc nghiên cứu kết thúc thì giải thích được nhiều nhất 18%. Và sức dự báo của「các mục tự báo」gấp từ hai đến bốn lần so với「các mục do bạn đời báo」. Ngoài cảm nhận quan hệ do mình báo ra, nếu thêm khác biệt cá nhân và báo cáo của bạn đời vào thì không thêm được sức dự báo nào +- Mức bằng chứng: A +- Nguồn: Joel S, Eastwick PW, Allison CJ, et al. (2020). Machine learning uncovers the most robust self-report predictors of relationship quality across 43 longitudinal couples studies. Proceedings of the National Academy of Sciences, 117(32), 19061-19071. +- Ghi chú: Chính nghiên cứu đó cũng thấy rằng quan hệ này trong thời gian nghiên cứu khá lên hay xấu đi thì hầu như không tính ra được bằng bất kỳ nhóm chỉ số tự đánh giá nào trong bộ câu hỏi. Cho nên để phán quan hệ hiện tại tốt hay không, hỏi cảm nhận của chính mình chính xác hơn việc đối chiếu danh sách điều kiện của đối phương, nhưng đừng mong dự báo được nó sẽ ra sao sau này. Mục「tình cảm đừng chỉ hỏi có hay không」của phần này nói đúng chuyện ấy, ở đây liệt kê cụ thể là những mục nào. + +### 5. Muốn kéo gần khoảng cách thì thay nhau trả lời một loạt câu hỏi từ nông đến sâu, đừng học thoại + +- Chi phí: Không tốn tiền. Tốn 45 phút. +- Nói thẳng: Hai người lạ thay nhau trả lời 36 câu hỏi từ nông đến sâu, sau 45 phút thì thân nhau hơn nhiều so với người chỉ tán gẫu chừng đó thời gian. Thứ tạo ra hiệu quả là việc thay nhau nói thật, nói dần vào chỗ sâu, không phải nhờ thoại. Điều này chỉ nói lên lúc đó thân hơn, không có nghĩa là phát triển được thành quan hệ dài hạn. +- Lợi ích: Chọn hai người không quen nhau ghép cặp, để họ thay nhau trả lời ba nhóm câu hỏi dần dần đi sâu, tổng cộng 36 câu. Câu hỏi bắt đầu từ「bạn nhất muốn mời ai ăn cơm」đến「lần cuối cùng bạn khóc trước mặt người khác là khi nào」. Mức thân thiết sau 45 phút cao hơn hẳn nhóm đối chứng tán gẫu cùng thời lượng. Cũng cao hơn mức điểm trung bình mà sinh viên trong nghiên cứu đó tự chấm cho quan hệ thân thiết nhất ngoài đời của mình +- Mức bằng chứng: B +- Nguồn: Aron A, Melinat E, Aron EN, Vallone RD, Bator RJ (1997). The experimental generation of interpersonal closeness: a procedure and some preliminary findings. Personality and Social Psychology Bulletin, 23(4), 363-377. +- Ghi chú: Mới chỉ làm thí nghiệm một lần, đối tượng là sinh viên đại học, chỉ đo mức thân thiết trong thời gian ngắn, không đo xem có phát triển thành quan hệ dài hạn hay không. Mức thân thiết trong thí nghiệm đến từ việc hai người thay nhau, nói dần vào chỗ sâu về tình hình thật của mình, không dựa vào thoại, chiêu thức hay biểu hiện đơn phía. Cho nên lúc trò chuyện có thể「tôi nói một chút thật, rồi hỏi lại một câu sâu bằng nhau」, hai bên thay phiên. + +### 6. Yêu xa không kém yêu gần, chỗ khó thật sự là ba tháng đầu sau khi dọn về sống cùng + +- Chi phí: Tốn tiền vé đường đi, còn tốn thời gian đường đi đường về. Số lần gặp ít hơn yêu gần. Khó nhất là chịu đựng lâu dài việc không gặp được mặt. +- Nói thẳng: Yêu xa trong thời gian chia đôi chỗ lại ổn hơn, vì gặp ít nên dễ tưởng đối phương đẹp hơn thực tế. Chỗ khó thật sự nằm ở sau khi dọn về sống cùng. Trong dữ liệu theo dõi, khoảng một nửa cặp yêu xa đi được đến bước sống cùng thành phố, nửa còn lại chia tay ngay trong thời gian chia đôi. Trong nửa đoàn tụ ấy, một phần ba chia tay trong vòng 3 tháng. +- Lợi ích: Hai nghiên cứu có kết luận nhất quán. Yêu xa trong thời gian chia đôi lại ổn định hơn yêu gần. Lý do là trong quan hệ yêu xa, mức tưởng đối phương đẹp hơn thực tế cao hơn hẳn, nghiên cứu đo các mục là méo mó do lý tưởng hóa, tình yêu lãng mạn, hồi ức về quan hệ, cảm nhận nhất quán. Và gặp càng ít thì tưởng càng đẹp. Nhưng sự ổn định này chỉ đúng khi hai bên vẫn còn chia đôi, một khi chuyển sang cùng thành phố, cặp yêu xa lại dễ chia tay hơn. Tỉ lệ do nghiên cứu theo dõi yêu xa chuyển sang cùng thành phố đưa ra là: khoảng một nửa cặp yêu xa đi được đến bước dọn về sống cùng, nửa còn lại chia tay ngay trong thời gian chia đôi. Trong nửa đoàn tụ ấy, một phần ba chia tay trong vòng 3 tháng sau khi đoàn tụ. Khoảng cách giữa hai lần gặp trong thời gian chia đôi càng dài, lý tưởng hóa càng mạnh, sau khi đoàn tụ càng không ổn +- Mức bằng chứng: B +- Nguồn: Stafford L, Merolla AJ, Castle JD (2006). When long-distance dating partners become geographically close. Journal of Social and Personal Relationships, 23(6), 901-919. ; Stafford L, Merolla AJ (2007). Idealization, reunions, and stability in long-distance dating relationships. Journal of Social and Personal Relationships, 24(1), 37-54. ; Dargie E, và cộng sự (2015). Journal of Sex & Marital Therapy, 41(2), 181-202. （nghiên cứu yêu xa nói tới trong ghi chú） +- Ghi chú: Đối tượng là quan hệ yêu đương của sinh viên đại học Mỹ, không gồm hôn nhân, cũng khá xưa, đừng áp thẳng tỉ lệ tuyệt đối vào mình. Nghiên cứu khác khảo sát 717 người yêu xa và 425 người yêu gần. Chất lượng hai loại quan hệ chênh lệch rất nhỏ, quan hệ tốt hay không chủ yếu do hoàn cảnh riêng của hai người và chính bản thân quan hệ đó. Cho nên khi quyết định có yêu xa hay không, đừng lấy「yêu xa chắc chắn chia tay」làm lý do. Điều cần chuẩn bị trước là mấy tháng sau khi đoàn tụ, lúc đó không gian tự chủ giảm đi, hiểu nhau nhiều hơn, xung đột và ghen cũng nhiều hơn. + +### 7. Xem số liệu đăng ký trước, nghe lời bố mẹ sau: tỷ lệ kết hôn, tỷ lệ ly hôn tính theo cách nào + +- Chi phí: Không tốn tiền. Đọc đoạn đăng ký hôn nhân trong báo cáo thống kê hằng năm của Bộ Dân vụ, mất vài phút. +- Nói thẳng: Năm 2024 cả nước đăng ký kết hôn 610.6 vạn cặp, ít hơn năm trước hai phần trăm, ly hôn 351.3 vạn cặp. Trên mạng nói tỷ lệ ly hôn quá nửa, là chia hai con số đó cho nhau, khoảng 57.5%. Nhưng người ly hôn và người kết hôn không phải cùng một nhóm, không thể đọc thành một nửa số hôn nhân sẽ ly hôn. Tỷ lệ kết hôn 4.3‰, tỷ lệ ly hôn 2.5‰ trong báo cáo thống kê là chia cho dân số cả nước. +- Lợi ích: Năm 2024 cả nước đăng ký kết hôn đúng pháp luật 610.6 vạn cặp, giảm 20.5% so với năm trước, tỷ lệ kết hôn 4.3‰. Cùng năm thực hiện thủ tục ly hôn đúng pháp luật 351.3 vạn cặp, trong đó ly hôn đăng ký tại cơ quan dân vụ 262.2 vạn cặp, ly hôn do tòa án xét xử và hoà giản 89.1 vạn cặp, tỷ lệ ly hôn 2.5‰. Báo cáo thống kê ghi rõ: tỷ lệ kết (ly) hôn = số cặp kết (ly) hôn trong năm / dân số bình quân trong năm × 1000‰. Đây là「tỷ lệ thô」, mẫu số là toàn bộ dân số, không phải dân số đã kết hôn +- Mức bằng chứng: A +- Nguồn: Bộ Dân vụ (民政部) (2025). Báo cáo thống kê công trình dân vụ năm 2024（五（二）1 婚姻登记服务、注释 5）, bản trên trang và bản PDF. 、 +- Ghi chú: Câu「tỷ lệ ly hôn trên 50%」hay gặp trên mạng là chia số cặp ly hôn trong năm cho số cặp kết hôn trong năm, 351.3 / 610.6 ≈ 57.5%. Cái này gọi là「tỷ lệ ly trên kết」. Hai số chia cho nhau không phải cùng một nhóm người, báo cáo thống kê cũng không có chỉ số này, nên không thể đọc thành「một nửa số hôn nhân sẽ ly hôn」. Muốn tính「xác suất cuối cùng một cuộc hôn nhân là ly hôn」 thì phải theo dõi cùng một nhóm người kết hôn từ đầu đến cuối, loại số liệu chính thức như vậy phần này không tìm thấy. + +### 8. Tính cả lợi ích sức khoẻ vào, nhưng phải hạ giá theo dữ liệu chỉ quan sát + +- Chi phí: Không tốn tiền, cũng không tốn thời gian. Chỉ cần nhớ một điều: các con số này nói là người đã kết hôn có tỷ lệ tử vong thấp hơn, đó là hai việc cùng xảy ra, không phải kết hôn làm bạn sống lâu hơn. +- Nói thẳng: Người đã kết hôn có xác suất tử vong cùng kỳ thấp hơn người chưa kết hôn khoảng 12%, người chưa bao giờ kết hôn cao hơn người đã kết hôn khoảng 24%. Nhưng đó chỉ là hai việc cùng xảy ra, không phải kết hôn gây ra. Người vốn khoẻ, thu nhập cao, giao tiếp nhiều thì vừa dễ kết hôn hơn, vừa dễ sống lâu hơn. Các con số này phần lớn từ châu Âu và châu Mỹ, áp vào trong nước chỉ nên coi là đơn vị cỡ. +- Lợi ích: Một phân tích tổng hợp về người cao niên, gộp nhiều nghiên cứu rồi tính lại, tổng cộng 53 nhóm so sánh, hơn 250.000 người. Người đã kết hôn so với người chưa kết hôn, tử vong toàn nguyên RR 0.88, 95% CI 0.85–0.91 (RR là tỷ số nguy cơ, CI là khoảng tin cậy), xác suất tử vong cùng kỳ thấp hơn khoảng 12%. Trong cùng nghiên cứu, người gả chồng sớm 1.11, ly hôn hoặc sống ly tân 1.16, chưa kết hôn 1.11, cao hơn lần lượt khoảng 11%, 16%, 11%. Không có khác biệt giữa nam và nữ. Ước tính thận trọng nhất là RR 0.94, chỉ thấp hơn khoảng 6%. Một phân tích tổng hợp khác so người chưa kết hôn (chưa bao giờ kết hôn) với người đã kết hôn, dùng 95 tài liệu, hơn 500.000.000 người. Kết quả HR 1.24 (1.19–1.30), xác suất tử vong cùng kỳ cao hơn khoảng 24%. Tuổi càng cao thì chênh lệch càng nhỏ. Còn một phân tích tổng hợp tách theo giới tính, 21 nghiên cứu, hơn 7.890.000 người: nam nữ đều vậy, tình trạng chưa kết hôn đi kèm với tử vong toàn nguyên, tử vong do ung thư, tử vong do tim mạch. Ở phía nam thì chênh lệch về tử vong do tim mạch và tử vong toàn nguyên lớn hơn. Nam chưa kết hôn so với nữ chưa kết hôn, tử vong do đột quỵ nhiều hơn 31%, tử vong toàn nguyên nhiều hơn 9% +- Mức bằng chứng: A (Tranh cãi) +- Nguồn: Manzoli L, et al. (2007). Marital status and mortality in the elderly: a systematic review and meta-analysis. Soc Sci Med 64:77–94. ；Roelfs DJ, et al. (2011). The rising relative risk of mortality for singles: meta-analysis and meta-regression. Am J Epidemiol 174(4):379–389. ；Wang Y, et al. (2020). Sex differences in the association between marital status and the risk of cardiovascular, cancer, and all-cause mortality: a systematic review and meta-analysis of 7,881,040 individuals. Glob Health Res Policy 5:4. +- Ghi chú:Tranh cãi ở chỗ người kết hôn vốn đã được tuyển chọn trước: người khoẻ, thu nhập cao, giao tiếp nhiều thì vừa dễ kết hôn hơn, vừa dễ sống lâu hơn. Cả ba phân tích tổng hợp đều dùng dữ liệu quan sát trên nhóm dân cư có sẵn, chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. Chính tác giả cũng nói có thiên lệch công bố hoặc hạn chế về phương pháp. Roelfs còn thấy nghiên cứu càng làm nghiêm ngặt thì chênh lệch tính ra càng nhỏ. Các con số này chủ yếu lấy từ dân cư châu Âu và châu Mỹ, áp vào trong nước chỉ nên coi là đơn vị cỡ. + +### 9. Bảng tính thời gian thì tính theo「việc không được trả công」, nói rõ phân công rồi hãy đi đăng ký + +- Chi phí: Không tốn tiền. Trước khi cưới dành một hai tiếng, nói rõ việc nhà, chăm sóc người lớn tuổi trẻ nhỏ, ai phải nhường việc, tức các việc không được trả công, thành phân công rõ ràng, sau đó làm theo phân công. Khó ở chỗ phải nói thẳng ra, đừng kéo dài không nói. +- Nói thẳng: Trong khảo sát toàn quốc về sử dụng thời gian, các việc không được trả công như việc nhà và chăm sóc người thân, nam bình quân mỗi ngày 1 giờ 52 phút, nữ 3 giờ 29 phút. Khảo sát không chia nhóm theo đã kết hôn hay chưa, nên「sau khi kết hôn việc nhà tăng thêm bao nhiêu」không có số liệu chính thức, chênh lệch giới tính này chỉ nên coi là mức trần. Phân công có thể thỏa thuận thành một kiểu khác trước khi đi đăng ký. +- Lợi ích: Khảo sát toàn quốc về sử dụng thời gian năm 2024. Việc không được trả công là việc nhà, chăm sóc người thân, các việc không có công trả. Người trong ngày có làm loại việc này thì bình quân mỗi ngày 2 giờ 45 phút, trong đó nam 1 giờ 52 phút, nữ 3 giờ 29 phút. Trong một ngày, tỷ lệ người có làm loại việc này là nam 67.5%, nữ 83.9%. Việc nhà của dân cư bình quân mỗi ngày 1 giờ 17 phút, tính riêng người có làm thì 1 giờ 59 phút, ít hơn 28 phút so với năm 2018. Chăm sóc người thân bình quân mỗi ngày 30 phút, tính riêng người có làm thì 1 giờ 46 phút. Số liệu năm 2018 là: việc nhà nam 45 phút, nữ 2 giờ 6 phút; chăm sóc con nam 17 phút, nữ 53 phút +- Mức bằng chứng: A +- Nguồn: Cục Thống kê Quốc gia (国家统计局) (2024). Hai báo cáo công bố của khảo sát toàn quốc về sử dụng thời gian lần thứ ba（第三次全国时间利用调查公报（第二号、第三号））. 、；Cục Thống kê Quốc gia (国家统计局) (2019). Báo cáo khảo sát toàn quốc về sử dụng thời gian năm 2018. +- Ghi chú: Cả hai khảo sát chỉ chia nhóm theo giới tính, thành thị hay nông thôn, độ tuổi, không chia theo đã kết hôn hay chưa. Nên「sau khi kết hôn việc nhà tăng thêm bao nhiêu」không có số liệu chính thức, chỉ có thể dùng chênh lệch giới tính làm mức trần. Số liệu là bình quân cả nước, phân công của hai người bạn hoàn toàn có thể thỏa thuận thành một kiểu khác. + +### 10. Bảng tính tiền thì xem quy tắc mặc định của pháp luật trước, rồi mới quyết định có cần thỏa thuận bằng văn bản không + +- Chi phí: Không tốn tiền. Đọc mấy điều về tài sản trong phần hôn nhân và gia đình của Bộ luật Dân sự. Nếu cần thì hai người ký một thỏa thuận tài sản bằng văn bản, pháp luật chỉ yêu cầu viết ra giấy, không cần công chứng. Khó ở chỗ phải nói trước chuyện tiền. +- Nói thẳng: Không ký thỏa thuận bằng văn bản thì theo pháp luật mặc định, lương, thưởng, thu nhập kinh doanh và đầu tư, tài sản thừa kế hoặc được tặng sau khi cưới đều thuộc sở hữu chung của hai người. Tài sản trước khi cưới, tài sản mà di chúc hoặc hợp đồng tặng ghi rõ chỉ cho một người thì mới là tài sản riêng. Hai người có thể thỏa thuận bằng văn bản là mỗi người giữ phần của mình, đã ký thì có tính ràng buộc. Bên nào phải nhiều hơn vì chăm con, chăm người lớn tuổi thì lúc ly hôn có thể yêu cầu bồi thường. +- Lợi ích: Quy tắc mặc định khi không có thỏa thuận khác là: lương, thưởng, tiền công lao động, thu nhập kinh doanh và đầu tư, thu nhập từ quyền sở hữu trí tuệ, tài sản thừa kế hoặc được tặng sau khi cưới thuộc tài sản chung của vợ chồng（第一千零六十二条）. Tài sản trước khi cưới, bồi thường tổn thương thân thể, tài sản mà di chúc hoặc hợp đồng tặng ghi rõ chỉ thuộc về một người, đồ dùng sinh hoạt riêng của một người, là tài sản riêng（第一千零六十三条）. Hai bên có thể thỏa thuận bằng văn bản rằng tài sản trước và sau khi cưới thuộc riêng từng người, thuộc chung, hoặc một phần riêng một phần chung, thỏa thuận này có tính ràng buộc pháp lý đối với cả hai bên（第一千零六十五条）. Bên nào phải nhiều nghĩa vụ hơn vì nuôi con, chăm người lớn tuổi, hỗ trợ công việc của bên kia thì lúc ly hôn có quyền yêu cầu bồi thường（第一千零八十八条） +- Mức bằng chứng: A +- Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人民代表大会) (2020). Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（中华人民共和国民法典）（2021 年 1 月 1 日起施行，第一千零六十二、一千零六十三、一千零六十五、一千零八十八条）. Cơ sở dữ liệu văn pháp quốc gia ；toàn văn điều khoản xem ở bản Công báo của Tòa án nhân dân tối cao đăng lại +- Ghi chú: Tiền cước, đám cưới, nhà cưới không có thống kê chính thức, mục này không viết số. Khoản bồi thường ở điều 1088 cho bao nhiêu thì trước hết hai người tự bàn, thỏa thuận không xong thì tòa án quyết, không có công thức cố định. Thỏa thuận tài sản bằng văn bản dùng được cho cả hai bên: bên thu nhập cao dùng được, bên làm nhiều việc nhà và chăm sóc không được trả công cũng dùng được. + +### 11. Bố mẹ góp tiền mua nhà, lúc chuyển khoản thì viết rõ là cho hay là vay + +- Chi phí: Không tốn tiền. Ngày góp tiền thì viết một trang giải trình, ghi rõ khoản tiền đó tặng cho ai hay cho vay ai, bố mẹ góp tiền và con cái đều ký tên. Nếu ghi là khoản vay thì để cả hai vợ chồng ký, phần ghi chú trong chuyển khoản cũng viết cùng câu đó. +- Nói thẳng: Tiền bố mẹ bỏ ra mua nhà trước khi cưới, mặc định chỉ cho con mình. Tiền bỏ ra sau khi cưới mà không có thỏa thuận thì tính là tài sản chung của vợ chồng. Từ tháng 2 năm 2025, bố mẹ một bên góp toàn bộ sau khi cưới mà không viết rõ chỉ cho con mình thì lúc ly hôn tòa án có thể xử nhà cho con của bên góp tiền, rồi quyết định có bồi thường bên kia hay không. Nếu tính là khoản vay thì ngày đó viết rõ, cả hai bên ký tên. +- Lợi ích: Tiền bố mẹ góp cho cả hai bên để mua nhà trước khi cưới thì được xác định là tặng riêng cho con cái của họ. Trừ khi bố mẹ nói rõ là tặng cho cả hai bên. Tiền góp sau khi cưới thì có thỏa thuận thì theo thỏa thuận. Không có thỏa thuận hoặc thỏa thuận không rõ thì xử lý theo nguyên tắc tài sản chung của vợ chồng. Từ ngày 1 tháng 2 năm 2025 còn thêm một quy định. Trong thời gian hôn nhân, bố mẹ một bên góp toàn bộ, mà hợp đồng tặng không viết rõ chỉ tặng cho con mình, thì lúc chia tài sản khi ly hôn tòa án có thể xử nhà thuộc về con của bên góp tiền. Xử về bên đó rồi thì căn cứ mấy yếu tố để quyết định có bồi thường bên kia hay không, bồi bao nhiêu. Các yếu tố đó là sống chung và có con chung, lỗi khi ly hôn, mức đóng góp cho gia đình, và giá thị trường của căn nhà tại thời điểm ly hôn. Bố mẹ một bên góp một phần, hoặc bố mẹ cả hai bên đều góp, thì căn cứ nguồn và tỷ lệ góp để xử về một bên, đồng thời bồi thường hợp lý (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Tòa án nhân dân tối cao (最高人民法院) (2020). Giải thích (một) về việc áp dụng phần hôn nhân và gia đình của Bộ luật Dân sự（关于适用《民法典》婚姻家庭编的解释（一））（法释〔2020〕22 号，第二十九条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). Giải thích (hai) về việc áp dụng phần hôn nhân và gia đình của Bộ luật Dân sự（关于适用《民法典》婚姻家庭编的解释（二））（法释〔2025〕1 号，第八条）. ；Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第一千零六十二条第一款第四项）. +- Ghi chú: Viết rõ thì cả hai bên đều dùng được. Bố mẹ bỏ tiền không cần trông chờ tới lúc ly hôn mới bù thêm một tờ giấy vay nữa để giữ phần vốn góp. Sau khi giải thích (hai) có hiệu lực, việc tiền do ai bỏ ra đã là căn cứ để chia nhà, tên đứng trên giấy chứng nhận không còn quyết định thuộc về ai. Bên nhận tiền góp cũng không phải chờ tới lúc ly hôn mới biết khoản tiền đó tính là gì. Muốn làm theo kiểu khoản vay thì làm thật như khoản vay: hai bên nói rõ là vay, thỏa thuận cách trả, chuyển khoản ghi rõ, cần thì để lại chứng từ trả nợ. Thiếu một mắt xích nào thì lên tòa phải giải thích. +### 12. Khoản vay lớn của một bên vợ chồng, nếu bạn không ký và không xác nhận sau, không tự động thành nợ của bạn + +- Chi phí: Không tốn tiền. Đừng ký「người vay chung」「người bảo lãnh」trên giấy vay hay giấy bảo đảnh của vợ chồng bạn. Cũng đừng viết trong tin nhắn kiểu「tiền này chúng tôi sẽ trả」, câu đó có thể bị hiểu là bạn đã xác nhận khoản nợ đó sau. Khó ở chỗ phải nhịn không đáp lời khi bị đòi tiền. +- Nói thẳng: Khoản tiền vợ chồng bạn vay sau lưng bạn, chỉ cần nó rõ ràng vượt quá chi phí sinh hoạt hằng ngày của gia đình, mà bạn không ký, không xác nhận sau, thì không tự động thành nợ của bạn. Bên đòi tiền phải tự chứng minh khoản tiền đó dùng cho sinh hoạt chung hoặc kinh doanh chung. Nhưng khoản tiền mua nhà, nếu quả thực đã biến thành căn nhà hai người cùng ở hoặc cùng sở hữu, thì bên kia thường chứng minh được tới mức đó. +- Lợi ích: Những khoản vay dưới đây thuộc nợ chung của vợ chồng. Một là hai vợ chồng cùng ký tên. Hai là một bên xác nhận sau, tức là nhận khoản nợ đó về sau, cả hai người đều thừa nhận. Ba là một bên trong thời gian hôn nhân, với danh nghĩa cá nhân, vay để đáp ứng nhu cầu sinh hoạt hằng ngày của gia đình. Khoản tiền một bên vay với danh nghĩa cá nhân mà rõ ràng vượt quá chi phí sinh hoạt hằng ngày của gia đình thì không thuộc nợ chung của vợ chồng. Ngoại lệ là khi chủ nợ (bên đòi tiền) chứng minh được khoản tiền đó dùng cho sinh hoạt chung của vợ chồng, cho sản xuất kinh doanh chung, hoặc là khoản nợ mà cả hai người đều nhận. Khoản tiền mua nhà rõ ràng vượt quá nhu cầu sinh hoạt hằng ngày của gia đình thì nghĩa vụ chứng minh thuộc về bên chủ nợ, bên đòi tiền phải tự đưa ra chứng cứ. Nhưng nếu khoản tiền đó quả thực đã biến thành căn nhà hai vợ chồng cùng ở hoặc cùng sở hữu thì chủ nợ thường chứng minh được tới mức「dùng cho sinh hoạt chung của vợ chồng」. Vì vậy bước hữu ích hơn là bước ở mục 11 của phần này (bố mẹ góp tiền thì viết rõ là vay hay là tặng) (toàn quốc, áp dụng từ 2021) +- Mức bằng chứng: A +- Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第一千零六十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020, sửa đổi lần thứ hai). Quy định về một số vấn đề pháp lý khi xét các vụ án vay ngoài（关于审理民间借贷案件适用法律若干问题的规定）（第十五、十六条）. （Công báo của Tòa án nhân dân tối cao） +- Ghi chú: Lúc ly hôn chia tài sản, bố mẹ bên kia bất ngờ đưa ra một hợp đồng vay chỉ có chữ ký của vợ chồng bạn thì phải xem thành hai bước. Bước một xem khoản vay đó có thật là đã xảy ra hay không. Bạn nói tiền chưa từng được đưa, mà lý lẽ lại nói nối, thì tòa án sẽ xét kết hợp nhiều yếu tố. Xem trước số tiền vay, tiền được đưa bằng cách nào, năng lực tài chính của hai bên; sau đó xem hai bên vẫn giao dịch ra sao, tài sản có biến động tương ứng không, lời khai của người làm chứng thế nào. Bên kia chỉ đưa ra được chứng từ chuyển khoản thì vẫn phải tự chứng minh khoản vay đó có thật. Bước hai, khoản vay đã thành lập thì mới xem có phải nợ chung không. Quy tắc chung về giấy vay và giấy bảo đảnh xem mục 18 ở phần 8. + +### 13. Đi đăng ký chỉ cần thẻ căn cước cộng một bản cam kết có chữ ký, không cần sổ hộ khẩu, không thu tiền, cấp giấy ngay tại chỗ + +- Chi phí: Không tốn tiền. Cả hai người đều phải có mặt. Thường xong trong vài chục phút. +- Nói thẳng: Từ tháng 5 năm 2025, đi đăng ký chỉ cần hai thứ: thẻ căn cước của chính mình, cộng một bản cam kết có chữ ký. Trong cam kết ghi rõ bản thân không có người vợ chồng, và với người kia không có quan hệ huyết thống trực tiếp, cũng không có quan hệ huyết thống bên thứ ba trở lên tính từ ông bà. Không cần mang sổ hộ khẩu, cũng không phải quay về nơi đăng ký hộ khẩu của một bên để làm. Đăng ký không thu đồng nào, đủ điều kiện thì cấp giấy ngay tại chỗ. +- Lợi ích: Quy định đăng ký hôn nhân mới áp dụng từ ngày 10 tháng 5 năm 2025 đã sửa bộ giấy tờ người dân trong nước phải nộp khi kết hôn. Điều khoản viết rõ chỉ cần xuất trình hai thứ:「(một) thẻ căn cước của chính mình; (hai) bản cam kết có chữ ký xác nhận bản thân không có người vợ chồng và không có quan hệ huyết thống trực tiếp, quan hệ huyết thống bên thứ ba trở lên tính từ ông bà với người kia」. Sổ hộ khẩu không còn là giấy tờ bắt buộc. Cơ quan đăng ký là「cơ quan dân vụ của chính quyền cấp huyện, hoặc chính quyền xã (thị trấn) do chính quyền tỉnh, vùng tự trị, thành phố trực thuộc trung ương xác định theo nguyên tắc thuận tiện cho nhân dân」. Hạn chế「tại nơi đăng ký hộ khẩu thường trú của một bên vợ chồng」trong quy định cũ đã được bỏ. Quy định còn viết rằng「cơ quan đăng ký hôn nhân khi tiến hành đăng ký hôn nhân không được thu phí」, và「đối với vợ chồng đủ điều kiện kết hôn thì phải đăng ký ngay tại chỗ, cấp giấy chứng nhận kết hôn」 +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2025). Quy định đăng ký hôn nhân（婚姻登记条例）（国令第 804 号，第二、六、八、十条）. +- Ghi chú: Có bốn trường hợp cơ quan đăng ký không làm, và chỉ bốn thôi: chưa đến tuổi kết hôn theo pháp luật, không phải cả hai đều hoàn toàn tự nguyện, một bên hoặc cả hai bên đã có người vợ chồng, hai người là quan hệ huyết thống trực tiếp hoặc huyết thống bên thứ ba trở lên tính từ ông bà. Những khoản có thể thu tiền chỉ là dịch vụ phụ như chụp ảnh, sao chép. Bản thân việc đăng ký không thu đồng nào, gặp khoản gọi là「phí cấp giấy」「phí làm gấp」 thì từ chối thẳng và khiếu nại. Thời hạn 30 ngày suy nghĩ khi đăng ký ly hôn xem ở mục tính chi phí thoát ra trong phần này. + +### 14. Trước khi cưới nên làm một lần khám sức khoẻ tiền hôn nhân, mắc bệnh hiểm nghèo nặng thì phải nói thật với bên kia trước khi đăng ký + +- Chi phí: Phần lớn các nơi miễn phí hoặc giảm phí cho người đúng tuổi. Tự trả cũng chỉ vài trăm tệ. Mất nửa ngày. +- Nói thẳng: Mắc bệnh hiểm nghèo nặng mà không nói thật với bên kia trước khi đăng ký thì bên kia có thể xin tòa án hủy cuộc hôn nhân đó. Phải đưa ra trong vòng một năm kể từ ngày biết. Bản thân việc khám sức khoẻ tiền hôn nhân từ 2003 không còn là điều kiện phải có trước khi cấp giấy, có đi không là tuỳ chọn. Chi phí tuỳ chính sách địa phương, mất nửa ngày. +- Lợi ích: Hậu quả pháp lý của việc giấu bệnh là rõ ràng. Quy định đăng ký hôn nhân mới quy định「một bên vợ chồng mắc bệnh hiểm nghèo nặng thì phải nói thật với bên kia trước khi đăng ký kết hôn; không nói thật thì bên kia có thể căn cứ điều 第一千零五十三条 của Bộ luật Dân sự xin Tòa án nhân dân hủy cuộc hôn nhân」. Nhà nước coi việc khám sức khoẻ tiền hôn nhân là chốt chặn đầu tiên trong phòng chống dị tật bẩm sinh. Mục tiêu đang đặt ra là「giữ tỷ lệ khám y tế trước khi cưới và tỷ lệ người thuộc nhóm mục tiêu được khám sức khoẻ trước khi mang thai lần lượt ở mức 70% và trên 80%」. Văn bản còn yêu cầu「đẩy việc bố trí cơ sở khám y tế trước khi cưới ở gần nơi đăng ký hôn nhân để thuận tiện cho nhân dân」, và đẩy mạnh「dịch vụ một cửa cho đăng ký hôn nhân, khám y tế trước khi cưới, hướng dẫn sinh dưỡng」. Bộ luật Dân sự quy định một bên mắc bệnh hiểm nghèo nặng thì phải nói thật trước khi đăng ký kết hôn, không nói thật thì bên kia có thể xin Tòa án nhân dân hủy cuộc hôn nhân. Yêu cầu hủy cuộc hôn nhân thì phải đưa ra trong vòng một năm kể từ ngày biết hoặc phải biết lý do hủy +- Mức bằng chứng: B +- Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第一千零五十三条）. ；Quốc vụ viện (国务院) (2025). Quy định đăng ký hôn nhân（婚姻登记条例）（国令第 804 号，第十二条）. ；Văn phòng Ủy ban Quốc gia Y tế và Sức khoẻ dân dân (国家卫生健康委办公厅) (2023). Kế hoạch nâng cao năng lực phòng chống dị tật bẩm sinh（出生缺陷防治能力提升计划（2023—2027 年））（国卫办妇幼发〔2023〕9 号）. ；Ủy ban Quốc gia Y tế và Sức khoẻ dân dân cùng bốn cơ quan khác (国家卫生健康委等五部门) (2020). Thông báo về việc tăng cường chăm sóc sức khoẻ tiền hôn nhân（关于加强婚前保健工作的通知）（国卫妇幼函〔2020〕205 号）：「积极倡导各地结合实际推行免费婚前检查」. +- Ghi chú: Khám sức khoẻ tiền hôn nhân từ 2003 không còn là bước phải làm trước khi cấp giấy, có đi không là tuỳ chọn. Nhưng quy định「mắc bệnh hiểm nghèo nặng mà không nói thật」thật sự trao cho bên kia quyền ra tòa án hủy cuộc hôn nhân đó, phải đưa ra trong vòng một năm kể từ ngày biết. Cả nước không có quy định thống nhất rằng khám là miễn phí. Thông báo năm 2020 của Ủy ban Quốc gia Y tế và Sức khoẻ dân dân cùng bốn cơ quan khác chỉ viết「积极倡导各地结合实际推行免费婚前检查」. Vì vậy có miễn phí không, khám những hạng mục nào thì do từng nơi tự quy định, hỏi tại quầy một cửa của cơ quan bảo vệ bà mẹ và trẻ em địa phương hoặc của văn phòng đăng ký hôn nhân. + +### 15. Đừng chỉ hỏi giá trị tình cảm「có hay không」, hãy xem chất lượng quan hệ + +- Chi phí: Không tốn tiền. Trước khi cưới, dùng mấy lần cãi vã, mấy tháng sống chung để xem chất lượng quan hệ, đừng chỉ đối chiếu danh sách điều kiện. Khó ở chỗ phải nhịn không kết luận chỉ từ danh sách. +- Nói thẳng: Chất lượng hôn nhân càng cao thì người ta càng khoẻ, nhưng mối liên hệ này rất yếu, gần bằng mức ảnh hưởng của thói quen như ăn uống tới sức khoẻ. Muốn xem quan hệ có tốt không thì phải nhìn qua mấy lần cãi vã, mấy tháng sống chung, đừng chỉ đối chiếu danh sách điều kiện. Cuộc hôn nhân tệ có gây hại cho sức khoẻ nhiều hơn việc không kết hôn không, chưa có dữ liệu nào so trực tiếp. +- Lợi ích: Phân tích tổng hợp 126 nghiên cứu, hơn 72.000 người cho thấy chất lượng hôn nhân càng cao thì sức khoẻ càng tốt. Cỡ tác động r nằm trong khoảng 0.07 đến 0.21. r là mức độ hai việc biến đổi cùng nhau, 0 là không liên quan gì, 1 là đồng biến hoàn toàn. Trong đó r = 0.11 với nguy cơ tử vong, r = −0.13 với phản ứng tim mạch khi xảy ra xung đột. Cỡ này tương đương với ảnh hưởng của các hành vi sức khoẻ như ăn uống tới sức khoẻ. Trong các nghiên cứu có kiểm tra rõ khác biệt giới tính, hầu như không thấy nam nữ khác nhau. Một tổng quan trong lĩnh vực tim mạch chỉ ra rằng mức độ không hài lòng với hôn nhân và chất lượng hôn nhân có ảnh hưởng rõ rệt tới nguy cơ tim mạch +- Mức bằng chứng: B (Tranh cãi) +- Nguồn: Robles TF, et al. (2014). Marital quality and health: a meta-analytic review. Psychol Bull 140(1):140–187. ；Dhindsa DS, et al. (2020). Marital status and outcomes in patients with cardiovascular disease. Trends Cardiovasc Med 30(4):215–220. +- Ghi chú:Tranh cãi ở chỗ mối liên hệ này vốn đã rất yếu. Một phần kết quả có thiên lệch công bố. Hơn nữa phần lớn nghiên cứu chỉ khảo sát tại một thời điểm, không thấy rõ ai dẫn tới ai, chính Robles trong bài gốc cũng liệt kê các hạn chế đó. Phần này không tìm thấy phân tích tổng hợp nào so trực tiếp tỷ lệ tử vong giữa「hôn nhân chất lượng thấp」với「chưa kết hôn」, nên「hôn nhân tệ tệ hơn không kết hôn」hiện chỉ là suy luận, không phải kết luận bằng số. + +### 16. Tính chi phí thoát ra: ly hôn thuận lợi có 30 ngày suy nghĩ, ly hôn kiện tụng có điều kiện theo pháp luật + +- Chi phí: Không tốn tiền. Trước khi cưới dành chút thời gian đọc: nhỡ phải thoát ra thì mất bao lâu, đi theo thủ tục nào. Khó ở chỗ phải nghĩ chuyện chia tay trước khi cưới. +- Nói thẳng: Ly hôn thuận lợi không phải ký xong là xong. Sau khi nộp đơn trong 30 ngày, bất kỳ bên nào cũng rút lại được. Hết hạn thì vẫn phải trong 30 ngày, hai người tự đến nhận giấy chứng nhận ly hôn, không đến thì coi như rút đơn. Bên kia không đồng ý thì chỉ còn đi kiện. Tòa án hoà giải trước, phải có bạo lực gia đình, cưới hai vợ, sống ly tân đủ hai năm và các trường hợp tương tự, hoà giải không xong mới xử cho ly hôn. +- Lợi ích: Ly hôn thuận lợi phải ký thỏa thuận ly hôn bằng văn bản và tự đến cơ quan đăng ký hôn nhân nộp đơn（第一千零七十六条）. Từ ngày cơ quan đăng ký nhận được đơn, trong 30 ngày bất kỳ bên nào cũng có thể rút lại. Sau khi hết hạn đó, trong 30 ngày, hai bên phải tự đến xin cấp giấy chứng nhận ly hôn, không xin thì coi như rút đơn（第一千零七十七条）. Một bên không đồng ý thì đi kiện, tòa án phải hoà giải trước. Có cưới hai vợ hoặc sống chung với người khác, bạo lực gia đình hoặc ngược đãi, bỏ rơi, cờ bạc ma túy nhiều lần không cải, sống ly tân đủ hai năm vì bất hoà và các trường hợp tương tự, hoà giải không có hiệu quả thì phải chấp thuận cho ly hôn. Đã xử không cho ly hôn rồi lại sống ly tân đủ một năm, kiện lại thì phải chấp thuận（第一千零七十九条）. Năm 2024 cả nước ly hôn 351.3 vạn cặp, trong đó 89.1 vạn cặp đi theo phán quyết và hoà giải của tòa án +- Mức bằng chứng: A +- Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人民代表大会) (2020). Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（中华人民共和国民法典）（第一千零七十六、一千零七十七、一千零七十九条）. ；Bộ Dân vụ (民政部) (2020). Thông báo về việc triển khai các quy định liên quan đến đăng ký hôn nhân trong Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（关于贯彻落实《中华人民共和国民法典》中有关婚姻登记规定的通知）（民发〔2020〕116 号）. ；Bộ Dân vụ (民政部) (2025). Báo cáo thống kê công trình dân vụ năm 2024. Nguồn tham khảo giống mục 7 của phần này（结婚率、离婚率各是什么口径） +- Ghi chú: Thời hạn suy nghĩ chỉ áp dụng cho ly hôn thuận lợi. Ly hôn kiện tụng không có thời hạn suy nghĩ, nhưng kéo dài hơn. Phí tố tụng, phí luật sư mỗi nơi khác nhau, mục này không viết số. Chi phí thoát ra cao không có nghĩa là không nên cưới, nhưng trước khi cưới thì nên biết những điều này. + +### 17. Riêng phần「kết hôn vì bố mẹ」thì ghi một sổ riêng, đừng trộn vào sổ của bạn + +- Chi phí: Không tốn tiền. Trên giấy chia làm hai cột. Một cột là trong cuộc hôn nhân đó những lợi ích nào là của bạn, mấy mục ở đầu phần này tính được. Cột kia là những lợi ích nào là của bố mẹ: yên tâm, thể diện, không còn bị thúc cưới. Khó ở chỗ phải để cột của bố mẹ riêng ra, không gộp vào cột của mình. +- Nói thẳng: Phần lợi ích của bố mẹ, tức là yên tâm, thể diện, không còn bị thúc cưới, là có thật. Nhưng nó không tự động thành lợi ích sức khoẻ hay chất lượng quan hệ của bạn. Còn ba sổ là thời gian, tiền và chi phí thoát ra thì do bạn tự gánh hết. Chia hai cột trên giấy, viết rõ lợi ích nào của mình, lợi ích nào của bố mẹ, sổ này có đáng hay không thì bạn nhìn là biết. +- Lợi ích: Lợi ích của bố mẹ là có thật. Nhưng nó không tự động chuyển thành lợi ích sức khoẻ của bạn (mục 8 của phần này) hay chất lượng quan hệ (mục 15 của phần này). Còn sổ thời gian (mục 9 của phần này), sổ tiền (mục 10 đến 12 của phần này) và chi phí thoát ra (mục 16 của phần này) thì do hai vợ chồng tự gánh. Tách sổ ra rồi,「hiệu quả chi phí」của việc kết hôn như làm nhiệm vụ là cao hay thấp thì bạn tự thấy được +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, thuộc loại gợi ý về phương pháp; phần dữ liệu xem các mục trước trong phần này +- Ghi chú: Cột của bố mẹ cũng có thể tính là một khoản rất nặng, đánh đổi thế nào thì bạn tự quyết. Danh sách điền chỗ trống xem tại [docs/结婚划不划算.md](../docs/结婚划不划算.md). +### 18. Coi quan hệ tốt xấu là một sổ sức khoẻ: cãi nhau thì nói vào việc, không nói vào người, đừng chửi, đừng chế giễu + +- Chi phí: Không tốn tiền. Khi cãi nhau chỉ nói chuyện đang xảy ra trước mắt, không chửi, không chế giễu, không đào lại chuyện cũ. Khó ở chỗ phải kìm miệng lúc nóng máu. +- Nói thẳng: Vợ chồng có quan hệ tốt thì thân thể khoẻ hơn một chút, nguy cơ tử vong cũng thấp hơn một chút. Mối liên hệ này không lớn, gần bằng mức ảnh hưởng của ăn uống tới sức khoẻ. Trong một thí nghiệm, vợ chồng mà lúc cãi nhau luôn mang thái độ thù địch thì vết thương nhỏ do thí nghiệm tạo ra chỉ lành nhanh bằng sáu phần mười tốc độ của vợ chồng bình tĩnh. +- Lợi ích: Gộp 126 nghiên cứu, hơn 72.000 người lại tính (phân tích tổng hợp). Chất lượng hôn nhân càng cao thì thân thể càng tốt, hệ số tương quan r nằm trong khoảng .07 đến .21. Hệ số tương quan 0 là không liên quan gì, 1 là đồng biến hoàn toàn. Trong đó tương quan với việc nguy cơ tử vong thấp hơn là r = .11. Cãi nhau thì phản ứng tim mạch nhỏ hơn, r = -.13. Tác giả nói các tác động này đều nhỏ, gần bằng mức liên hệ giữa thói quen sức khoẻ như ăn uống với sức khoẻ. Một thí nghiệm giao cắt khác dùng 42 cặp vợ chồng khoẻ, mỗi cặp nằm viện hai lần, mỗi lần 24 giờ. Một lần để nói chuyện hỗ trợ lẫn nhau, một lần để bàn một việc có bất đồng. Sau khi cãi nhau xong, vết thương bọng nước do thí nghiệm tạo ra lành chậm hơn. Vợ chồng thể hiện thái độ thù địch nhiều hơn ở cả hai lần thì tốc độ lành vết thương chỉ bằng 60% tốc độ của vợ chồng ít thù địch hơn. Sáng hôm sau khi cãi nhau, các chỉ số viêm trong máu của họ (IL-6, yếu tố hoại tử u) cũng tăng nhiều hơn +- Mức bằng chứng: B +- Nguồn: Robles TF, Slatcher RB, Trombello JM, McGinn MM (2014). Marital quality and health: a meta-analytic review. Psychological Bulletin. ；Kiecolt-Glaser JK 等 (2005). Hostile marital interactions, proinflammatory cytokine production, and wound healing. Archives of General Psychiatry. +- Ghi chú: Lý do để mức B là: phần lớn nghiên cứu được gộp chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. Nên không phân biệt được quan hệ xấu khiến người ta bị bệnh, hay thân thể xấu khiến quan hệ xấu đi. Có nghiên cứu sau khi trừ tuổi và thu nhập thì mối liên hệ vẫn còn. Thí nghiệm vết thương chỉ có 42 cặp vợ chồng. Cách làm「nói vào việc không nói vào người, không chửi không chế giễu」là tác giả suy ra từ「hành vi thù địch」trong thí nghiệm, không có nghiên cứu nào so trực tiếp. Cãi nhau dữ dội mà không sửa được thì hai người cùng đi làm tư vấn hôn nhân và gia đình, điểm này cũng không có tài liệu riêng. Cãi nhau mà đánh nhau, hoặc lạm ngạng, đe dọa lâu dài, thì không còn là chuyện giao tiếp nữa, đó là bạo lực gia đình, xem mục 43 ở phần 8 (bị bạo lực gia đình). Muốn cân nhắc có nên đi không thì xem chi phí thoát ra ở mục 16 của phần này (chi phí thoát ra). + +### 19. Đôi bạn cùng giới nên tranh thủ lúc cả hai còn minh mẫn, làm xong giấy uỷ quyền, giám hộ tự nguyện bằng văn bản và di chúc: về mặt pháp luật hai người không phải người thân, không làm thì không có quyền ký và không có quyền thừa kế + +- Chi phí: Giám hộ tự nguyện bằng văn bản và di chúc thì làm ở văn phòng công chứng, phí công chứng mỗi nơi khác nhau, vài trăm tệ trở lên. Di chúc tự tay viết thì không tốn tiền. Trước sau phải dành một hai buổi nửa ngày. Khó ở chỗ hai người phải ngồi lại nói chuyện bệnh tật và cái chết. +- Nói thẳng: Người thân theo pháp luật là vợ chồng, cha mẹ, con cái, anh chị em, bạn đời không phải. Không làm giấy tờ trước thì khi một bên nặng bệnh, bên kia không quyết được thay. Khi một bên qua đời, bên kia không thừa kế được phần nào. Ba bản giấy tờ mỗi bản phụ trách một đoạn: khi còn sống thì dựa vào giấy uỷ quyền, khi mất năng lực tâm thần thì dựa vào giám hộ tự nguyện bằng văn bản, sau khi mất thì dựa vào di chúc. +- Lợi ích: Bộ luật Dân sự liệt kê người thân gồm vợ chồng, cha mẹ, con cái, anh chị em, ông bà nội, ông bà ngoại, cháu nội, cháu ngoại（第一千零四十五条）. Người thừa kế theo pháp luật trước hết là vợ chồng, con cái, cha mẹ, sau đó là anh chị em, ông bà nội, ông bà ngoại（第一千一百二十七条）. Bạn đời không có trong cả hai danh sách. Phẫu thuật, xét nghiệm chuyên sâu, điều trị đặc biệt thì theo quy định khác（第一千二百一十九条）. Những việc không thể hoặc không nên giải thích cho bệnh nhân thì nhân viên y tế phải giải thích với「người thân」của bệnh nhân và lấy được sự đồng ý. Có ba đường để bù vào chỗ thiếu. Một là quy định cho phép người thành niên và「cá nhân khác sẵn lòng làm người giám hộ」thỏa thuận trước bằng văn bản về người giám hộ（第三十三条）. Khi bản thân mất hoặc mất một phần năng lực chủ động dân sự thì người đó làm người giám hộ, người giám hộ là đại diện theo pháp luật. Hai là quy định giấy uỷ quyền bằng văn bản phải ghi rõ tên người đại diện, nội dung đại diện, phạm vi và thời hạn, có chữ ký hoặc con dấu của chính bên ủy quyền（第一百六十五条）. Chủ uỷ quyền qua đời thì việc uỷ quyền chấm dứt. Ba là quy định cho phép lập di chúc, tặng tài sản cá nhân cho cá nhân ngoài những người thừa kế theo pháp luật（第一千一百三十三条）. Một bài viết của Sở Tư pháp thành phố Ích Châu đăng trên trang của Sở Tư pháp tỉnh Hồ Bắc nói rõ rằng đôi bạn cùng giới có thể làm công chứng thỏa thuận giám hộ tự nguyện bằng văn bản, chỉ định người kia làm người giám hộ của mình sau này (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第二十三、三十三、一百六十五、一百七十三、一千零四十五、一千一百二十四、一千一百二十七、一千一百三十三至一千一百四十、一千一百四十二、一千二百一十九条）. （Cơ sở dữ liệu văn pháp quốc gia）；Sở Tư pháp thành phố Ích Châu (宜昌市司法局) (2021). Công chứng giám hộ tự nguyện bằng văn bản: con đường công chứng tới cầu vồng. Mục「实践荟萃」trên trang của Sở Tư pháp tỉnh Hồ Bắc. +- Ghi chú: Vị thế của người thừa kế của bạn đời là「người được tặng theo di chúc」, không phải người thừa kế. Biết có di chúc như vậy thì trong 60 ngày phải nói rõ là nhận, không nói là coi như từ bỏ. Bản thân bạn đời và những người có lợi ích liên quan với họ không được làm người chứng kiến di chúc. Đã lập nhiều di chúc thì lấy bản cuối cùng, xem mục 2 ở phần 17 (di chúc sau thì đảo ngược di chúc trước). Người giám hộ tự nguyện có ký trực tiếp được giấy đồng ý phẫu thuật không thì mỗi bệnh viện làm khác nhau, hỏi trước bệnh viện bạn hay đi. Người thân không công nhận các giấy tờ này thì phải ra tòa: xét sơ thẩm theo thủ tục thông thường từ 6 tháng trở lên, phí luật sư tự bỏ ra. Vì vậy giấy tờ càng nên làm ở văn phòng công chứng. Nhà mua chung, tiền gửi chung, thì viết thêm một thỏa thuận bằng văn bản theo số tiền góp thật. Làm giám hộ tự nguyện bằng văn bản thì xem mục 1 ở phần 17 (chỉ định bằng văn bản người giám hộ trong tương lai). Người được lợi là chính bạn và bạn đời của bạn. + +### 20. Đừng kết hôn mà giấu xu hướng tính dục, cũng đừng coi hôn nhân hình thức là cách cho nhanh: về mặt pháp luật đó là một cuộc hôn nhân thật, muốn thoát ra chỉ có thể ly hôn + +- Chi phí: Không tốn tiền. Khó ở chỗ phải chịu được áp lực bố mẹ thúc cưới, có thể phải gánh nhiều năm. +- Nói thẳng: Kết hôn mà giấu xu hướng tính dục thì về mặt pháp luật không phải là hôn nhân vô hiệu, cũng không hủy được, muốn thoát ra chỉ có ly hôn. Hôn nhân hình thức cũng là hôn nhân thật, tiền kiếm được sau khi cưới, nợ mắc sau khi cưới đều tính cho hai vợ chồng. Bên bị giấu khi ly hôn có thể yêu cầu chia nhiều tài sản hơn, bồi thường tổn thương tinh thần, nhưng tòa án có chấp nhận hay không thì chưa có kết luận. +- Lợi ích: Bộ luật Dân sự quy định hôn nhân vô hiệu chỉ có ba trường hợp: cưới hai vợ, có quan hệ huyết thống cấm kết hôn, chưa đến tuổi kết hôn theo pháp luật（第一千零五十一条）. Hôn nhân có thể hủy được chỉ có hai trường hợp: bị ép cưới, giấu bệnh hiểm nghèo nặng trước khi cưới（第一千零五十二、一千零五十三条）. Yêu cầu xác nhận hôn nhân vô hiệu với lý do ngoài ba trường hợp đó thì tòa án xử bác（第十七条 của Giải thích (một) về việc áp dụng phần hôn nhân và gia đình của Bộ luật Dân sự）. Vì vậy muốn thoát ra chỉ có thể ly hôn. Hôn nhân hình thức đã đăng ký vẫn là hôn nhân hợp pháp. Lương, thưởng, tiền công lao động sau khi cưới là tài sản chung của vợ chồng（第一千零六十二条）. Khoản nợ vay có cả hai vợ chồng cùng ký, hoặc một bên vay để chi phí sinh hoạt hằng ngày của gia đình, là nợ chung của vợ chồng（第一千零六十四条）. Bồi thường tổn thương khi ly hôn chỉ áp dụng cho cưới hai vợ, sống chung với người khác, bạo lực gia đình, ngược đãi bỏ rơi thành viên gia đình, và「lỗi lớn khác」（第一千零九十一条）. Giải thích (một) giới hạn「sống chung với người khác」là sống chung với「người khác giới tính ngoài hôn nhân」（第二条 của Giải thích (một)）. Vì vậy quan hệ đồng giới chỉ có thể được tranh đòi theo「lỗi lớn khác」. Phần bồi thường gồm cả bồi thường tổn thương tinh thần, bên bị giấu phải nêu ra cùng đơn kiện ly hôn, nếu không xử cho ly hôn thì không hỗ trợ. Sau khi Bộ luật Dân sự có hiệu lực, không tìm thấy phán quyết nào của tòa án được công bố cho loại vụ này. Thời kỳ luật cũ, một bài viết đăng năm 2013 trên trang của Tòa án nhân dân tỉnh Giang Tô bàn về vụ vợ kiện chồng vì chồng giấu chuyện đồng tính luyến. Bài viết nêu hai cách hiểu. Một cách cho rằng đồng tính luyến không thuộc lỗi theo pháp luật thời đó, không nên hỗ trợ chia nhiều tài sản cùng bồi thường tổn thương tinh thần. Tác giả đồng tình với cách còn lại: coi bên bị giấu là「bên không có lỗi」, khi chia tài sản nên quan tâm đến cô, và còn phải bồi thường tiền an ủi tinh thần (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第一千零五十一至一千零五十四、一千零六十二、一千零六十四、一千零七十九、一千零八十七、一千零九十一条）. （Cơ sở dữ liệu văn pháp quốc gia）；Tòa án nhân dân tối cao (最高人民法院) (2020). Giải thích (một) về việc áp dụng phần hôn nhân và gia đình của Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（关于适用《中华人民共和国民法典》婚姻家庭编的解释（一））（法释〔2020〕22 号，第二、十七、八十六至九十条）. ；Zhu Laikuan, Wang Hongjuan (2013). Chồng đồng tính luyến dẫn tới ly hôn thì có thể yêu cầu bồi thường tổn thương hay không? Trang mạng Tòa án Giang Tô. +- Ghi chú: Phần chia nhiều tài sản và bồi thường tổn thương tinh thần chỉ có một bài phân tích, bài đó không viết vụ cuối cùng xử ra sao, nên có nhận được hay không còn tuỳ từng tòa. Bộ luật Dân sự thêm vào luật cũ một mục「lỗi lớn khác」, cho bên bị giấu một cửa để tranh đòi. Bản thân điều khoản là mức A. Hai bên của hôn nhân hình thức đều có bạn đời cùng giới thì giữa hai bạn đời không được luật hôn nhân bảo vệ, tài sản phải viết riêng cho rõ, xem mục 19 của phần này (đôi bạn cùng giới làm giám hộ tự nguyện bằng văn bản và di chúc). Ly hôn hình thức cũng phải qua 30 ngày suy nghĩ hoặc ra tòa, xem mục 16 của phần này (ly hôn thuận lợi có 30 ngày suy nghĩ). Ly hôn thuận lợi mà viết là từ bỏ quyền bồi thường thì sau đó kiện sẽ không được hỗ trợ. Người được lợi trước hết là người vợ chồng bị giấu, sau đó là chính bạn: một khi chuyện giấu bị phát hiện thì ly hôn, chia tài sản, nuôi con sẽ dính vào nhau cùng lúc. \ No newline at end of file diff --git "a/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" "b/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" index e016ca6f..e7c6f456 100644 --- "a/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" +++ "b/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" @@ -1,161 +1,155 @@ -[← 回总目录](../README.md) - -# 11. 程序员和技术人容易踩的红线 - -本节口径是人身自由与金钱，换回的是「别被关」和「别赔钱」。写代码、做技术的人，常把下面这些事当成「只是技术问题」，可它们对应的是刑法里的具体罪名，或者民事上要赔的具体一笔钱。每条都给出法条里的数字，以及最高人民检察院、最高人民法院官网公布的案例。刑法引用的是北京市公安交管局转载的「根据刑法修正案（十一）修正」整合文本。条目顺序按「技术人最容易以为没事 × 后果最重」从高到低排。第 1 条是通用的一条，讲动手前怎么判断，以及出事之后先做什么。 - [← Về mục lục](../README.md) # 11. Vạch đỏ lập trình viên và người làm kỹ thuật dễ vấp -Thước đo của phần này là tự do cá nhân và tiền, đổi lại bạn nhận được「đừng bị giam」và「đừng phải bồi tiền」. Người viết mã, làm kỹ thuật thường coi những việc dưới đây chỉ là「vấn đề kỹ thuật」, nhưng chúng ứng với những tội danh cụ thể trong Bộ luật Hình sự, hoặc với một khoản tiền bồi thường dân sự cụ thể. Mỗi mục đưa ra con số trong điều luật, cùng các vụ án do Viện kiểm sát nhân dân tối cao và Tòa án nhân dân tối cao công bố trên trang chính thức. Bộ luật Hình sự được dẫn là văn bản hợp nhất「được sửa đổi theo Bản sửa đổi (mười một) của Bộ luật Hình sự」do Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại. Các mục được xếp theo「người làm kỹ thuật dễ tưởng không có chuyện × hậu quả nặng nhất」, từ cao xuống thấp. Mục 1 là mục chung, nói về cách đánh giá trước khi ra tay, và làm gì ngay sau khi có chuyện. +Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại là「đừng bị bắt giam」và「đừng phải bồi thường」. Người viết mã, người làm kỹ thuật thường coi những việc dưới đây chỉ là「vấn đề kỹ thuật」, nhưng chúng ứng với những tội danh cụ thể trong Bộ luật Hình sự, hoặc với một khoản tiền phải bồi thường cụ thể trong quan hệ dân sự. Mỗi mục đưa ra con số ghi trong điều luật, cùng những vụ án đã được Viện kiểm sát nhân dân tối cao và Tòa án nhân dân tối cao công bố trên trang chính thức. Phần Bộ luật Hình sự dẫn theo bản hợp nhất「đã sửa đổi theo Bản sửa đổi (mười một) của Bộ luật Hình sự」mà Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại. Thứ tự các mục xếp từ cao xuống thấp theo「thứ kỹ sư dễ cho là không sao × hậu quả nặng nhất」. Mục 1 là mục chung, nói cách đánh giá trước khi làm và việc phải làm đầu tiên sau khi có chuyện. -### 1. Trước khi ra tay hỏi ba câu: đã làm tổn hại lợi ích của ai, bên kia có khả năng truy cứu đến mức nào, bạn để lại bao nhiêu chứng cứ; bị truy cứu thì tìm luật sư hình sự ngay +### 1. Trước khi làm, hỏi ba câu: đã hại lợi ích của ai, đối phương có khả năng truy cứu đến đâu, mình để lại bằng chứng bao nhiêu; bị truy cứu thì tìm luật sư hình sự ngay -- Chi phí: Không tốn tiền. Trước khi ra tay, đổi câu hỏi「kỹ thuật có làm được không」thành câu hỏi khác: nhà sản xuất, nền tảng, công ty có đi báo việc không. Chỗ khó là nhịn được cái ý「chuyện nhỏ này không ai hỏi đâu」, vì nhật ký máy chủ, lịch sử thanh toán, lịch sử nhóm trò chuyện, khi xét xử đều mặc định được truy xuất. -- Nói thẳng: Trước khi ra tay, hỏi một câu: nhà sản xuất, nền tảng, công ty có đi báo việc không. Nhật ký máy chủ, lịch sử thanh toán, lịch sử nhóm trò chuyện đều truy xuất được khi xét xử. Thật sự bị bắt giữ hình sự thì trong tối đa 37 ngày có quyết định bắt hay không. Từ lần thẩm vấn đầu tiên bạn đã có thể mời luật sư, và 37 ngày đó là lúc tìm luật sư hình sự hữu ích nhất. -- Lợi ích: Bất kỳ đơn vị hay cá nhân nào phát hiện sự thật phạm tội đều có quyền báo việc. Cơ quan xử lý cho là có sự thật phạm tội thì phải khởi xướng. Sau khi bị bắt giữ hình sự, công an phải trong 3 ngày trình xin phê chuẩn bắt (xin viện kiểm sát bắt chính thức). Trường hợp đặc biệt có thể kéo dài thêm 1 đến 4 ngày. Lưu trú nhiều nơi (liên tục phạm tội ở các địa phương khác nhau) thì kéo dài được tới 30 ngày. Viện kiểm sát dùng thêm 7 ngày để quyết định. Cộng lại tối đa 37 ngày để quyết định có bắt hay không. Từ lần thẩm vấn đầu tiên, hoặc từ lúc bị áp dụng biện pháp cưỡng chế (bị giữ, bị bắt, các loại tương tự), bạn đã có thể ủy thác luật sư. Khi vụ án còn đang trong giai đoạn công an điều tra thì chỉ có thể ủy thác luật sư. Luật sư yêu cầu gặp thì trại giam sắp xếp chậm nhất trong 48 giờ. Nếu có thể bị phạt kỉ luật, giam giữ, những hình phạt nhẹ hơn ngồi tù, hoặc chỉ bị phạt kèm mà không phạt chính, thì có thể được bảo lãnh tại nhà: nộp một khoản tiền bảo lãnh hoặc tìm một người bảo lãnh, về nhà chờ xét xử. Bảo lãnh tại nhà dài nhất 12 tháng (toàn quốc) +- Chi phí: Không tốn tiền. Trước khi làm, đổi câu hỏi「kỹ thuật có làm được không」thành câu khác: nhà sản xuất, nền tảng, công ty có đi báo việc không. Chỗ khó là nhịn được ý nghĩ「việc nhỏ này không ai để ý」, vì nhật ký máy chủ, lịch sử thanh toán, lịch sử chat nhóm đều mặc định được lấy ra khi điều tra +- Nói thẳng: Trước khi làm, hỏi một câu: nhà sản xuất, nền tảng, công ty có đi báo việc không. Nhật ký máy chủ, lịch sử thanh toán, lịch sử chat nhóm đều lấy ra được. Bị giữ hình sự thì trong tối đa 37 ngày sẽ quyết định bắt hay không bắt bạn. Từ lần hỏi đầu tiên bạn đã có thể mời luật sư. 37 ngày đó là lúc tìm luật sư hình sự hữu ích nhất. +- Lợi ích: Bất kỳ tổ chức hoặc cá nhân nào phát hiện sự việc phạm tội đều có quyền báo việc. Cơ quan điều tra cho là có sự việc phạm tội thì phải khởi kiện. Sau khi bị giữ hình sự, công an phải trong 3 ngày xin phê chuẩn bắt (xin viện kiểm sát bắt chính thức). Trường hợp đặc biệt có thể kéo dài thêm 1 đến 4 ngày. Hành vi gây ra nhiều vụ ở các địa phương khác nhau liên tiếp thì kéo dài được đến 30 ngày. Viện kiểm sát có thêm 7 ngày để quyết định. Cộng lại tối đa 37 ngày để quyết định bắt hay không. Từ lần thẩm vấn đầu tiên, hoặc từ lúc bị áp dụng biện pháp cưỡng chế (bị giữ, bị bắt thuộc loại đó), bạn đã có thể ủy thác luật sư. Khi vụ án còn đang ở giai đoạn công an điều tra thì chỉ được ủy thác luật sư. Luật sư yêu cầu được gặp, trại tạm giam bố trí chậm nhất trong 48 giờ. Nếu có thể chỉ bị phạt loại nhẹ hơn cải tù, ví dụ quản thích hay giam giữ, hoặc chỉ bị phạt kèm mà không phạt tù chính, thì được bảo lãnh tại ngoại: nộp một khoản tiền bảo lãnh hoặc tìm một người bảo lãnh, rồi về nhà chờ kết quả điều tra. Bảo lãnh tại ngoại tối đa 12 tháng (toàn quốc) - Mức bằng chứng: C - Nguồn: Quốc hội Nhân dân (全国人大) (2018). 刑事诉讼法（2018 年修正，第三十四、三十九、六十七、七十九、九十一、一百一十、一百一十二条）. -- Ghi chú: Con số 37 ngày đó tính từ lúc bị giữ đến lúc viện kiểm sát quyết định có phê chuẩn bắt hay không, không phải trần thời gian giam. Sau khi phê chuẩn bắt còn có thời hạn giam trong giai đoạn điều tra, phải tính tiếp. Luật sư rành nghề cũng hữu ích nhất đúng trong 37 ngày này: gặp thân nhân, nộp một bản ý kiến「không nên phê chuẩn bắt」cho viện kiểm sát, xin bảo lãnh tại nhà. Đợi tới lúc bị khởi tố rồi mới tìm luật sư là muộn. Các mục còn lại của phần này nói cùng một việc: bên kia có lý do để báo việc hay không. +- Ghi chú: 37 ngày đó tính từ lúc bị giữ đến lúc viện kiểm sát quyết định phê chuẩn bắt hay không, không phải trần của thời gian giam. Sau khi phê chuẩn bắt còn có thời hạn giam trong giai đoạn điều tra, rồi tính tiếp từ đó. Luật sư rành việc cũng hữu ích nhất đúng trong 37 ngày ấy: gặp người bị bắt, nộp cho viện kiểm sát một ý kiến「không nên phê chuẩn bắt」, xin bảo lãnh tại ngoại. Đợi đến lúc khởi tố rồi mới tìm luật sư thì đã muộn. Các mục còn lại của phần này nói về cùng một điều: đối phương có lý do để báo việc hay không. -### 2. Không làm, không bán phần mềm gian lập trong game và công cụ hỗ trợ, kể cả công cụ AI「chỉ chụp màn hình, không đụng bộ nhớ」 +### 2. Không làm và không bán hack game, kể cả công cụ hỗ trợ bằng AI「chỉ đọc màn hình, không đụng bộ nhớ」 -- Chi phí: Không tốn tiền. Cái giá là từ bỏ loại việc nhận đơn đó và thu nhập từ「tiện ích nhỏ」. Chỗ khó là khoản tiền đó đến nhanh, lại không thấy ai sẽ truy cứu. -- Nói thẳng: Bán phần mềm gian lập cho 20 người dùng, hoặc thu được 5000 tệ, là đủ phạt tù dưới 3 năm, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Có người làm phần mềm xuyên tường, ngắm tự động, 3 tháng bán được hơn 8 vạn, bị phạt tù 1 năm hoãn thi hành 1 năm, phạt tiền 5000 tệ. Có người lấy chương trình phía máy chủ để mở máy chủ riêng, thu hơn 2000 vạn tệ tiền nạp, nặng nhất bị phạt tù 4 năm, tiền phạt tối đa 1000 vạn tệ. -- Lợi ích: Cung cấp chương trình, công cụ chuyên dùng để xâm nhập, kiểm soát trái phép hệ thống thông tin máy tính, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền. Phạt kèm là vừa kết án vừa phạt tiền, chỉ phạt tiền là chỉ phạt tiền thôi. Đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm kèm phạt tiền. Ngưỡng của「nghiêm trọng」là cung cấp công cụ cho trên 20 người dùng, hoặc thu nhập bất hợp pháp trên 5000 tệ. Thu nhập bất hợp pháp là số tiền kiếm được nhờ việc đó. Cố ý vòng qua hoặc phá hỏng biện pháp bảo vệ kỹ thuật mà chủ thể quyền thiết lập còn có thể cấu thành tội xâm phạm quyền tác giả, phạt tù dưới 3 năm, nghiêm trọng thì phạt tù 3 đến 10 năm. Vụ án chính thức: Vương Nhậm Hào cùng những người khác sản xuất, bán phần mềm gian lập, thực hiện xuyên tường và ngắm tự động bằng cách đọc và sửa bộ nhớ của trò chơi. Trong 3 tháng họ bán được hơn 8 vạn tệ, lợi nhuận hơn 3 vạn tệ. Bị phạt tù 1 năm hoãn thi hành 1 năm, phạt tiền 5000 tệ. Hoãn thi hành 1 năm nghĩa là đã kết án 1 năm nhưng trước mắt không vào tù; trong năm đó không tái phạm thì không phải ngồi tù. Hạng Nhậm Hào cùng những người khác bán loạt phần mềm gian lập「TZ」, doanh thu hơn 20 vạn tệ, Hạng Nhậm Hào bị phạt tù 3 năm hoãn thi hành 3 năm, phạt tiền 2 vạn tệ. Lưu Nhậm Hào cùng những người khác có được chương trình phía máy chủ của trò chơi, sửa tham số để mở máy chủ riêng, thu hơn 2000 vạn tệ tiền nạp. Bị phạt tù từ 4 năm xuống 2 năm 6 tháng, phạt tiền từ 1000 vạn tệ xuống 130 vạn tệ (toàn quốc) +- Chi phí: Không tốn tiền. Cái giá phải trả là bỏ các đơn thuộc loại đó cùng thu nhập từ「tiện ích nhỏ」. Chỗ khó là số tiền đó đến nhanh, lại không thấy ai sẽ truy cứu +- Nói thẳng: Bán hack đủ 20 lượt người dùng, hoặc kiếm được 5000 tệ, là đủ phạt tù dưới 3 năm, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Có người làm hack nhìn xuyên tường và ngắm bắn, 3 tháng bán hơn 8 vạn, bị phạt tù 1 năm hoãn 1 năm, phạt tiền 5000 tệ. Có người dùng chương trình phía máy chủ để mở server riêng, thu hơn 2000 vạn tệ tiền nạp, nặng nhất phạt tù 4 năm, tiền phạt tối đa 1000 vạn tệ. +- Lợi ích: Cung cấp chương trình, công cụ chuyên dùng để xâm nhập, kiểm soát trái phép hệ thống máy tính, khi đặc biệt nghiêm trọng thì phạt tù dưới 3 năm, kèm phạt tiền hoặc chỉ phạt tiền. Kèm phạt tiền là vừa kết án vừa phạt tiền, chỉ phạt tiền là chỉ phạt tiền. Đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm kèm phạt tiền. Ngưỡng「情节严重」là cung cấp công cụ cho 20 lượt người dùng trở lên, hoặc thu nhập bất hợp pháp 5000 tệ trở lên. Thu nhập bất hợp pháp là số tiền kiếm được nhờ việc đó. Cố tình vòng qua hoặc phá hủy biện pháp bảo vệ công nghệ mà chủ sở hữu quyền thiết lập, còn có thể cấu thành tội xâm phạm quyền tác giả, phạt tù dưới 3 năm, nặng thì 3 đến 10 năm. Vụ án chính thức: Vương Mộc và những người khác sản xuất bán hack, dựa vào việc đọc và sửa bộ nhớ trò chơi để làm hack nhìn xuyên tường và ngắm bắn. 3 tháng bán hơn 8 vạn tệ, lợi nhuận hơn 3 vạn tệ. Phạt tù 1 năm hoãn 1 năm, phạt tiền 5000 tệ. Hoãn 1 năm nghĩa là đã kết án 1 năm nhưng trước mắt không vào tù, trong năm đó không tái phạm thì không phải đi tù. Hạng Mộc và những người khác bán bộ hack「TZ」, doanh thu hơn 20 vạn tệ, Hạng Mộc bị phạt tù 3 năm hoãn 3 năm, phạt tiền 2 vạn tệ. Lưu Mộc và những người khác có được chương trình phía máy chủ của trò chơi, sửa tham số để mở server riêng, thu hơn 2000 vạn tệ tiền nạp. Phạt tù 4 năm đến 2 năm 6 tháng, phạt tiền 1000 vạn đến 130 vạn tệ (toàn quốc) - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第三款、第二百一十七条第六项）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第三条）. （Cục Công an thành phố Thâm Quyến đăng lại）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). 上海静安：严惩制售游戏外挂黑灰产共护清朗网络空间. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 2020 年度检察机关保护知识产权典型案例（案例五）. -- Ghi chú: Các vụ án gian lập đã công bố đều là loại đọc ghi bộ nhớ trò chơi, vòng qua chống gian lận. Chưa thấy vụ án chính thức nào cho công cụ hỗ trợ「chỉ chụp màn hình, chỉ mô phỏng phím bấm」. Nhưng vụ án đặt vé ở mục 3 của phần này cho thấy không đụng bộ nhớ, chỉ gửi yêu cầu, vẫn có thể bị quy kết là tội thu thập dữ liệu hệ thống thông tin máy tính trái phép. Cuối cùng có phải tội phạm hay không, tính theo tội nào, do cơ quan xử lý quyết định trên cơ sở hồ sơ nhà sản xuất nộp lúc báo việc, không phải do tác giả quyết. Bán cho 20 người dùng là đã đủ mức「nghiêm trọng」, mà mức đó rất nhỏ. +- Ghi chú: Các vụ án hack đã công bố đều là loại đọc ghi bộ nhớ trò chơi, vượt cơ chế chống gian lận. Chưa thấy vụ án chính thức nào về công cụ「chỉ chụp màn hình, chỉ mô phỏng phím」. Nhưng vụ giành vé ở mục 3 cho thấy không đụng bộ nhớ, chỉ gửi yêu cầu, vẫn đủ để kết tội lấy dữ liệu trái phép. Cuối cùng có thành tội phạm hay không, tính theo tội nào, do cơ quan điều tra quyết định theo hồ sơ nhà sản xuất nộp lúc báo việc, không phải do tác giả quyết. Bán đủ 20 lượt người dùng đã là「情节严重」, mức đó rất nhỏ. -### 3. Không viết, không bán script đặt vé hộ, mua nhanh, đặt hàng ảo, quét ưu đãi, kể cả loại「tự động bấm nút」 +### 3. Không viết và không bán script giành vé, đua đặt hàng, tạo đơn ảo, mò hàng miễn phí, kể cả loại chỉ「tự động bấm nút」 -- Chi phí: Không tốn tiền, cũng không tốn thời gian. -- Nói thẳng: Script không sửa bộ nhớ, không bẻ khóa, chỉ gửi yêu cầu nhanh hơn người, vẫn bị kết án. Kiếm được 5000 tệ hoặc làm bên kia thiệt hại 1 vạn tệ là đủ. Có người dùng tài khoản 12306 mua được để đặt hộ hơn 200 vé, thu hơn 2 vạn tệ, bị phạt tù 1 năm 6 tháng hoãn thi hành 1 năm 6 tháng, phạt tiền 5000 tệ. Bán phần mềm đặt vé trên cửa hàng trực tuyến thì bị buộc bồi 2 vạn tệ về cạnh tranh không lành mạnh. -- Lợi ích: Xâm nhập hệ thống thông tin máy tính, hoặc dùng kỹ thuật khác để lấy dữ liệu trong đó, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Ngưỡng của「nghiêm trọng」là thu nhập bất hợp pháp trên 5000 tệ, hoặc gây thiệt hại trên 1 vạn tệ. Vụ án chính thức: có người dùng phần mềm gian lập nhập hàng loạt tài khoản 12306 mua được, gửi yêu cầu mua vé dày đặc bằng nhiều luồng. Người đó đặt hộ hơn 200 vé tàu, lợi nhuận hơn 2 vạn tệ. Tòa án quy tội thu thập dữ liệu hệ thống thông tin máy tính trái phép, phạt tù 1 năm 6 tháng hoãn thi hành 1 năm 6 tháng, phạt tiền 5000 tệ. Trương Nhậm Hào bán trên cửa hàng trực tuyến phần mềm đặt vé nhắm vào một ứng dụng bán vé, bị xác định là cạnh tranh không lành mạnh, buộc bồi 2 vạn tệ (toàn quốc) +- Chi phí: Không tốn tiền, cũng không mất thời gian +- Nói thẳng: Script không sửa bộ nhớ, không bẻ khóa, chỉ gửi yêu cầu nhanh hơn người thì vẫn bị kết tội, kiếm được 5000 tệ hoặc làm bên kia thiệt hại 1 vạn tệ là đủ. Có người dùng tài khoản 12306 mua được để giành vé thay, giành hơn 200 vé, kiếm hơn 2 vạn tệ, bị phạt tù 1 năm 6 tháng hoãn 1 năm 6 tháng, phạt tiền 5000 tệ. Bán phần mềm giành vé trên cửa hàng trực tuyến thì bị buộc bồi thường 2 vạn tệ theo tội cạnh tranh không lành mạnh. +- Lợi ích: Xâm nhập hệ thống máy tính, hoặc dùng kỹ thuật khác để lấy dữ liệu trong đó, khi đặc biệt nghiêm trọng thì phạt tù dưới 3 năm, kèm phạt tiền hoặc chỉ phạt tiền, đặc biệt nghiêm trọng hơn thì 3 đến 7 năm. Ngưỡng「情节严重」là thu nhập bất hợp pháp 5000 tệ trở lên, hoặc thiệt hại 1 vạn tệ trở lên. Vụ án chính thức: có người dùng hack nhập hàng loạt tài khoản 12306 mua được, rồi gửi yêu cầu mua vé dồn dập bằng nhiều luồng. Anh ta giành hơn 200 vé tàu, lợi nhuận hơn 2 vạn tệ. Tòa án quy theo tội chiếm dụng dữ liệu hệ thống máy tính trái phép, phạt tù 1 năm 6 tháng hoãn 1 năm 6 tháng, phạt tiền 5000 tệ. Trịnh Mộc bán phần mềm giành vé nhắm vào một ứng dụng bán vé, bị kết luận cấu thành cạnh tranh không lành mạnh, bồi thường 2 vạn tệ (toàn quốc) - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). 上海静安：查办利用外挂软件批量代抢火车票牟利案. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 2024 年人民法院知识产权典型案例（案例七，抢票软件不正当竞争案）. -- Ghi chú: Script trong vụ án đặt vé không sửa bộ nhớ, không bẻ khóa, chỉ「nhanh hơn người」, vẫn bị kết án theo điều 285 của Bộ luật Hình sự. Nhận tiền đặt hộ và bán script đều là hành vi hưởng lợi. Trường hợp tự dùng một lần thì không tìm được vụ án chính thức nào. +- Ghi chú: Script trong vụ giành vé không sửa bộ nhớ, không bẻ khóa, chỉ「nhanh hơn người」, vẫn bị kết tội theo Điều 285 của Bộ luật Hình sự. Nhận tiền để giành vé thay và bán script đều là hưởng lợi phi pháp. Trường hợp chỉ tự dùng một lần thì không tìm thấy vụ án chính thức nào. -### 4. Trình thu thập dữ liệu thì chỉ lấy trang công khai không cần đăng nhập, không vòng biện pháp chống thu thập, không đụng thông tin cá nhân, dữ liệu đã lấy thì không bán +### 4. Crawler chỉ lấy trang công khai không cần đăng nhập, không vòng cơ chế chống thu thập, không đụng thông tin cá nhân, dữ liệu đã lấy thì không bán -- Chi phí: Không tốn tiền. Trình thu thập dữ liệu chỉ là chương trình tự động mở trang, lấy dữ liệu trên đó xuống. Trước khi lấy, xem qua tệp robots và hướng dẫn cấp quyền của giao diện; robots là tệp trang web đặt ở thư mục gốc, ghi rõ những trang nào không cho phép lấy. Dữ liệu phải đăng nhập, phải có token (một chuỗi chứng thư quyền tạm), phải bẻ chữ ký mới có được thì không lấy. Chỗ khó là loại dữ liệu này về mặt kỹ thuật lấy là có, nhịn lại hoàn toàn dựa vào ý thức. -- Nói thẳng: Dữ liệu phải đăng nhập, phải có token, phải bẻ chữ ký mới có được, thuộc về vòng qua lớp bảo vệ, phạt tù dưới 3 năm, nghiêm trọng thì phạt tù 3 đến 7 năm. Lấy phải thông tin cá nhân còn phiền phức hơn. Hành trạng, thông tin liên lạc, thông tin tín dụng, thông tin tài sản 50 mục là đã tính là nghiêm trọng. Lưu trú, sức khoẻ, giao dịch 500 mục, loại khác 5000 mục, hoặc thu được 5000 tệ, cũng vậy. -- Lợi ích: Vòng qua lớp bảo vệ để lấy dữ liệu trong hệ thống thì mức tù giống mục 3 của phần này (script đặt vé, mua nhanh, đặt hàng ảo), phạt tù dưới 3 năm, hoặc 3 đến 7 năm. Thu thập, bán hoặc cung cấp trái phép thông tin cá nhân của công dân thì đạt bất kỳ mốc nào dưới đây là「nghiêm trọng」: hành trạng, nội dung liên lạc, tín dụng, thông tin tài sản trên 50 mục; lưu trú, lịch sử liên lạc, sức khoẻ, thông tin giao dịch trên 500 mục; thông tin cá nhân khác trên 5000 mục; thu nhập bất hợp pháp trên 5000 tệ. Nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Vụ án chính thức: nhân viên phát triển phần mềm Lý Nhậm Hào giấu chức năng đánh cắp thư viện ảnh trong điện thoại vào dưới lớp vỏ phần mềm「kiểm tra nhan sắc đẹp」, lấy được 1751 ảnh khuôn mặt, còn bán lại dữ liệu từ kho dữ liệu thu thập thông tin, bị phạt tù 3 năm hoãn thi hành 3 năm. Kho dữ liệu thu thập thông tin là cơ sở dữ liệu gom thông tin cá nhân bị rò rỉ khắp nơi. Có công ty mạng dùng trình thu thập dữ liệu để lấy dữ liệu cửa hàng và đơn hàng của một nền tảng giao đồ ăn, gây thiệt hại hơn 4 vạn tệ, bị truy cứu theo tội thu thập dữ liệu hệ thống thông tin máy tính trái phép (toàn quốc) +- Chi phí: Không tốn tiền. Crawler là chương trình tự động mở trang web rồi gom dữ liệu trên đó. Trước khi lấy, xem qua tệp robots và hướng dẫn cấp quyền của giao diện; robots là tệp website đặt ở thư mục gốc, ghi rõ những trang nào không được lấy. Dữ liệu phải đăng nhập, phải có token (một chuỗi giấy thông hành tạm), phải bẻ chữ ký mới có được thì đừng lấy. Chỗ khó là loại dữ liệu đó về kỹ thuật cứ thế là có, nhịn lại hoàn toàn dựa vào tự kiểm +- Nói thẳng: Dữ liệu phải đăng nhập, phải có token, phải bẻ chữ ký mới có được thuộc loại vòng qua bảo vệ, phạt tù dưới 3 năm, nặng thì 3 đến 7 năm. Lấy trúng thông tin cá nhân còn phiền hơn. Dữ liệu hành trình, liên lạc, tín dụng, tài sản 50 mục là đã「情节严重」. Lưu trú, sức khoẻ, giao dịch 500 mục, loại khác 5000 mục, hoặc kiếm được 5000 tệ, cũng vậy. +- Lợi ích: Lấy dữ liệu trong hệ thống bằng cách vòng qua bảo vệ thì mức phạt tù giống mục 3 (script giành vé, đua đặt hàng, tạo đơn ảo), phạt tù dưới 3 năm, hoặc 3 đến 7 năm. Thu thập, bán hoặc cung cấp trái phép thông tin cá nhân của công dân, chạm bất kỳ ngưỡng nào dưới đây là「情节严重」: hành trình, nội dung liên lạc, tín dụng, tài sản 50 mục trở lên; lưu trú, thông báo cuộc gọi, sức khoẻ, giao dịch 500 mục trở lên; thông tin cá nhân loại khác 5000 mục trở lên; thu nhập bất hợp pháp 5000 tệ trở lên. Nghiêm trọng thì phạt tù dưới 3 năm, kèm phạt tiền hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì 3 đến 7 năm. Vụ án chính thức: Lý Mộc, nhân viên phát triển phần mềm, ngụy trang chức năng đánh cắp thư viện ảnh điện thoại thành phần mềm「đo mức độ hấp dẫn», lấy được 1751 ảnh khuôn mặt, còn bán lại dữ liệu trong kho dữ liệu tổng hợp, bị phạt tù 3 năm hoãn 3 năm. Kho dữ liệu tổng hợp là cơ sở dữ liệu gom thông tin cá nhân bị rò rỉ ở nhiều nơi. Một công ty công nghệ mạng dùng crawler để lấy dữ liệu cửa hàng và đơn hàng của nền tảng giao đồ ăn, gây thiệt hại hơn 4 vạn tệ, bị truy cứu theo tội chiếm dụng dữ liệu hệ thống máy tính trái pháp (toàn quốc) - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百五十三条之一）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2017). 关于办理侵犯公民个人信息刑事案件适用法律若干问题的解释（第三、四、五条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 检察机关依法惩治侵犯公民个人信息犯罪典型案例（案例二）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例一）. -- Ghi chú: Giải thích của tòa án năm 2017 quy định rằng「không có sự đồng ý của chủ thể, đưa thông tin đã thu thập hợp pháp cho người khác」cũng là「cung cấp」. Trong công ty, xuất bảng người dùng cho đối tác bên ngoài cũng rơi vào câu đó. 5000 mục chỉ là lượng của một bảng tính nhỏ. +- Ghi chú: Giải thích tư pháp năm 2017 quy định「未经本人同意，把合法收集来的信息给别人」cũng tính là「提供」. Trong công ty mà xuất bảng người dùng cho đối tác bên ngoài thì cũng rơi vào câu đó. 5000 mục chỉ bằng một bảng nhỏ. -### 5. Không nhận đơn phát triển ứng dụng cờ bạc, lừa đảo, khiêu dâm; không dựng kênh thanh toán, không mở khoá tài khoản cho「khách hàng」 +### 5. Không nhận đơn phát triển ứng dụng cờ bạc, lừa đảo, khiêu dâm, không dựng cổng thanh toán cho「khách hàng», không mở khoá tài khoản bị khoá -- Chi phí: Không tốn tiền. Cái giá là từ bỏ loại việc thuê ngoài có giá cao mà yêu cầu lại nói không rõ. Chỗ khó là số tiền bên kia trả thường cao hơn hẳn một đơn bình thường. -- Nói thẳng: Nhận đơn có giá cao, yêu cầu mơ hồ như vậy thì coi như đồng phạm của nhóm cờ bạc. Xử lý theo tội mở sòng bạc thì phạt tù dưới 5 năm, nghiêm trọng thì 5 đến 10 năm, nặng hơn hẳn so với chỉ giúp một tay thì tối đa 3 năm. Một ứng dụng đoán cược thể thao điện tử có 28 người viết mã bị phạt tù từ 1 năm đến 3 năm 6 tháng. Người đi mở khoá tài khoản QQ cho nhóm lừa đảo bị phạt tù 1 năm 6 tháng, phạt tiền 2 vạn tệ. -- Lợi ích: Biết người khác đang dùng mạng để phạm tội mà vẫn cung cấp hỗ trợ kỹ thuật, hỗ trợ thanh toán quyết toán, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền. Đồng thời cấu thành tội phạm khác thì xử lý theo tội nặng hơn. Mở sòng bạc thì phạt tù dưới 5 năm kèm phạt tiền, nghiêm trọng thì phạt tù 5 đến 10 năm. Vụ án chính thức: tập đoàn Trương Thịnh phát triển game cờ bạc rồi「đóng gói bán ra nước ngoài」, thu phí dựng trang, phí báo cáo. Số tiền liên quan gần 70 tỷ tệ. Trương Thịnh bị phạt tù 9 năm 9 tháng, phạt tiền 5000 vạn tệ. Vương Cường cùng 79 người phụ trách thuê ngoài phát triển bị phạt tù từ 6 tháng đến 9 năm 9 tháng. Trần Đông cùng những người khác phát triển, vận hành ứng dụng đoán cược thể thao điện tử, nhóm kỹ thuật viết mã ở Hải Nam, 28 người bị xử theo tội mở sòng bạc, phạt tù từ 3 năm 6 tháng đến 1 năm không đều, kèm phạt tiền. Trương Nhậm Hào mở xưởng đi mở khoá các tài khoản QQ bị khoá cho nhóm lừa đảo, bị phạt tù 1 năm 6 tháng, phạt tiền 2 vạn tệ (toàn quốc) +- Chi phí: Không tốn tiền. Cái giá là bỏ mấy đơn thuê ngoài giá cao mà yêu cầu lại nói không rõ. Chỗ khó là bên kia thường trả nhiều hơn hẳn một đơn bình thường +- Nói thẳng: Đơn giá cao, yêu cầu mơ hồ như thế mà nhận vào thì coi như đồng phạm của bọn cờ bạc. Theo tội lập casino thì phạt tù dưới 5 năm, nặng thì 5 đến 10 năm, nặng hơn mức chỉ giúp việc tối đa 3 năm. Có một ứng dụng dự đoán thể thao điện tử, 28 người viết code bị phạt tù từ 1 năm đến 3 năm 6 tháng. Người mở khoá tài khoản QQ cho bọn lừa đảo thì bị phạt tù 1 năm 6 tháng, phạt tiền 2 vạn tệ. +- Lợi ích: Biết rõ người khác đang dùng mạng để phạm tội mà vẫn cung cấp hỗ trợ kỹ thuật, thanh toán, quyết toán các loại giúp việc, khi đặc biệt nghiêm trọng thì phạt tù dưới 3 năm, kèm phạt tiền hoặc chỉ phạt tiền. Đồng thời còn cấu thành tội phạm khác thì xét theo tội nào nặng hơn. Lập casino phạt tù dưới 5 năm kèm phạt tiền, nghiêm trọng thì 5 đến 10 năm. Vụ án chính thức: Tập đoàn Trương Thịnh sau khi phát triển game cờ bạc thì「đóng gói toàn mạng bán ra nước ngoài」, thu tiền dựng trang và tiền báo cáo. Số tiền liên quan gần 7 tỷ tệ. Trương Thịnh bị phạt tù 9 năm 9 tháng, phạt tiền 5000 vạn tệ. Vương Cường cùng 79 người phụ trách giao khoán phát triển bị phạt tù từ 6 tháng đến 9 năm 9 tháng. Trần Đông và những người khác phát triển, vận hành ứng dụng dự đoán thể thao điện tử, nhóm kỹ thuật viết code tại Hải Nam, 28 người bị phạt tù từ 3 năm 6 tháng đến 1 năm theo tội lập casino, kèm phạt tiền. Trương Mộc mở xưởng mở khoá tài khoản QQ bị khoá cho bọn lừa đảo, bị phạt tù 1 năm 6 tháng, phạt tiền 2 vạn tệ (toàn quốc) - Mức bằng chứng: A -- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十七条之二、第三百零三条第二款）. ；Viện kiện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 建赌博网站并出售，涉案近 70 亿元. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 上海闵行区：办理一起新型开设网络赌场案. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 依法惩治帮助信息网络犯罪活动及相关犯罪典型案例（案例一）. -- Ghi chú:「Biết» không cần bên kia nói ra mục đích sử dụng. Giá đưa ra cao hơn hẳn giá thị trường, chức năng là nạp, rút, cộng tỷ lệ cược, khách không chịu nói dùng để làm gì, khi xét xử đều dùng để suy ra bạn biết rõ trong lòng. Nhân sự kỹ thuật phát triển ứng dụng cờ bạc bị xử lý như đồng phạm của tội mở sòng bạc, không thuộc mức trần 3 năm của tội giúp đỡ. Tội giúp đỡ là tên rút gọn của「tội giúp đỡ hoạt động tội phạm trên thông tin mạng」. -### 6. Khi nghỉ việc, bàn giao lại quyền truy cập tài khoản; không xoá cơ sở dữ liệu, không để lại cửa hậu, không đổi mật khẩu để khoá hệ thống, kể cả khi công ty nợ bạn tiền +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十七条之二、第三百零三条第二款）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 建赌博网站并出售，涉案近 70 亿元. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 上海闵行区：办理一起新型开设网络赌场案. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 依法惩治帮助信息网络犯罪活动及相关犯罪典型案例（案例一）. +- Ghi chú: 「明知」không cần bên kia nói ra mục đích sử dụng. Báo giá cao hơn hẳn giá thị trường, chức năng là nạp, rút và tỷ lệ thắng, khách không chịu nói dùng để làm gì, thì khi điều tra đều được dùng để suy ra rằng bạn biết rõ. Kỹ thuật viên phát triển ứng dụng cờ bạc bị xử lý theo tội đồng phạm của tội lập casino, không phải mức trần 3 năm của tội giúp tin. Tội giúp tin là tên gọi rút gọn của「帮助信息网络犯罪活动罪」. +### 6. Rời công ty thì bàn giao xong quyền truy cập tài khoản, không xoá cơ sở dữ liệu, không để lại backdoor, không đổi mật khẩu để khoá hệ thống, dù công ty còn nợ bạn -- Chi phí: Không tốn tiền. Công ty nợ lương bạn thì đi hoà giải lao động để đòi lại, đừng tự ra tay (hoà giải lao động xem ở phần 7). Chỗ khó là lúc nóng giận mà nhịn không đụng vào hệ thống. -- Nói thẳng: Tài khoản, mật khẩu do công ty cấp không có nghĩa là nghỉ việc rồi vẫn dùng được; đăng nhập vào là kiểm soát trái phép, phạt tù dưới 3 năm. Xoá dữ liệu làm hệ thống không chạy được thì phạt tù dưới 5 năm, hậu quả đặc biệt nghiêm trọng thì phạt tù trên 5 năm. Có một kỹ sư cao cấp bị sa thải, dùng tài khoản quản trị cũ đăng nhập vào máy chủ công ty để xoá dữ liệu và nhật ký, thiệt hại hơn 12 vạn tệ, bị phạt tù 3 năm hoãn thi hành 5 năm, phạt tiền 3 vạn tệ. -- Lợi ích: Kiểm soát trái phép hệ thống thông tin máy tính, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Xoá, sửa chức năng hoặc dữ liệu của hệ thống làm hệ thống không chạy bình thường, hậu quả nghiêm trọng thì phạt tù dưới 5 năm, đặc biệt nghiêm trọng thì trên 5 năm. Vụ án chính thức: kỹ sư cao cấp công nghệ thông tin Lữ Nhậm Hào sau khi bị sa thải đã dùng tài khoản quản trị cũ đăng nhập vào máy chủ dùng chung của công ty. Anh ta đổi mật khẩu, xoá dữ liệu và nhật ký, thiệt hại hơn 12 vạn tệ. Bị phạt tù 3 năm hoãn thi hành 5 năm, phạt tiền 3 vạn tệ. Nhân viên công ty phần mềm Bạch Nhậm Hào khi viết chương trình nâng cấp cố ý chèn mã sai, khiến hệ thống của khách không đăng nhập được, thiệt hại hơn 4 vạn tệ, bị truy cứu theo tội phá hoại hệ thống thông tin máy tính (toàn quốc) +- Chi phí: Không tốn tiền. Công ty nợ bạn lương thì đi hoà giải lao động để đòi lại, đừng tự làm (hoà giải lao động xem phần 7). Chỗ khó là lúc nóng giận phải nhịn không đụng vào hệ thống +- Nói thẳng: Mật khẩu tài khoản do công ty cấp không có nghĩa sau khi nghỉ việc vẫn dùng được, đăng nhập vào là kiểm soát trái pháp, phạt tù dưới 3 năm. Xoá dữ liệu làm hệ thống không chạy được thì phạt tù dưới 5 năm, hậu quả đặc biệt nghiêm trọng thì 5 năm trở lên. Có một kỹ sư cao cấp bị sa thải, dùng tài khoản quản trị cũ đăng nhập máy chủ công ty để xoá dữ liệu và nhật ký, gây thiệt hại hơn 12 vạn tệ, bị phạt tù 3 năm hoãn 5 năm, phạt tiền 3 vạn tệ. +- Lợi ích: Kiểm soát trái phép hệ thống máy tính, khi đặc biệt nghiêm trọng thì phạt tù dưới 3 năm, kèm phạt tiền hoặc chỉ phạt tiền, đặc biệt nghiêm trọng hơn thì 3 đến 7 năm. Xoá, sửa chức năng hoặc dữ liệu của hệ thống làm cho hệ thống không chạy bình thường, hậu quả nghiêm trọng thì phạt tù dưới 5 năm, đặc biệt nghiêm trọng thì 5 năm trở lên. Vụ án chính thức: kỹ sư IT cao cấp Lý Mộc bị sa thải, dùng tài khoản quản trị cũ đăng nhập máy chủ dùng chung của công ty. Anh ta đổi mật khẩu, xoá dữ liệu và nhật ký, gây thiệt hại hơn 12 vạn tệ. Phạt tù 3 năm hoãn 5 năm, phạt tiền 3 vạn tệ. Nhân viên công ty phần mềm Bạch Mộc khi viết chương trình nâng cấp đã cố tình chèn mã lỗi, khiến hệ thống của khách không đăng nhập được, thiệt hại hơn 4 vạn tệ, bị truy cứu theo tội phá hoại hệ thống máy tính (toàn quốc) - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百八十六条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2024). 检察机关「依法惩治网络犯罪 助力网络空间综合治理」典型案例（案例六）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例七）. -- Ghi chú: Tài khoản, mật khẩu do công ty cấp không có nghĩa là nghỉ việc rồi vẫn dùng được. Dùng tài khoản cũ đăng nhập, coi như「kiểm soát trái phép」. Thiệt hại tính theo chi phí khôi phục dữ liệu, vài vạn tệ là đủ để khởi xướng. +- Ghi chú: Mật khẩu tài khoản do công ty cấp không có nghĩa sau khi nghỉ việc vẫn dùng được. Dùng tài khoản cũ để đăng nhập là「非法控制」. Thiệt hại tính theo chi phí khôi phục dữ liệu, mấy vạn tệ là đủ để khởi kiện. -### 7. Nghỉ việc không mang mã nguồn, danh sách khách hàng, tài liệu kỹ thuật đi; không tải lên ổ đĩa cá nhân, không dùng lại ở công ty sau +### 7. Rời công ty không mang mã nguồn, danh sách khách hàng, tài liệu kỹ thuật đi, không đăng lên ổ đĩa mạng cá nhân, không dùng lại ở công ty sau -- Chi phí: Không tốn tiền. Trước khi nghỉ việc, xoá sạch tài liệu công ty trên máy tính cá nhân và trên ổ đĩa mạng. Chỗ khó là có tệp do mình tự viết từng dòng, không nỡ. -- Nói thẳng: Mã do bạn tự viết thì quyền tác giả vẫn thuộc công ty; danh sách khách hàng cũng là bí mật kinh doanh; mang đi là lấy đồ của công ty. Có nhân viên nghỉ việc mang mã nguồn hệ thống kiểm tra an ninh sang công ty mới, bản thân bị phạt tù 5 năm, phạt tiền 50 vạn tệ, công ty mới bị phạt tiền 400 vạn tệ. Có người khai thác lỗ hổng hệ thống để tải hơn 3.3 vạn tệp thiết kế, bị phạt tù 1 năm 10 tháng. Công ty còn có thể kiện bạn riêng. -- Lợi ích: Vi phạm nghĩa vụ bảo mật, tiết lộ bí mật kinh doanh hoặc tự dùng lấy, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 10 năm kèm phạt tiền. Luật Chống cạnh tranh không lành mạnh còn liệt kê hai hành vi là xâm phạm: lấy bí mật kinh doanh bằng「xâm nhập điện tử」(dùng biện pháp kỹ thuật để xâm nhập hệ thống); xúi giục, giúp người khác vi phạm nghĩa vụ bảo mật. Vụ án chính thức: Tôn Nhậm Minh cùng những nhân viên nghỉ việc khác mang mã nguồn hệ thống kiểm tra an ninh và thuật toán lõi đi, dùng vào sản phẩm của công ty mới; Tôn Nhậm Minh bị phạt tù 5 năm, phạt tiền 50 vạn tệ, công ty mới bị phạt tiền 400 vạn tệ. Thiên Nhậm Hào khai thác lỗ hổng hệ thống để tải hơn 3.3 vạn tệp thiết kế, bị phạt tù 1 năm 10 tháng, phạt tiền 10 vạn tệ. Kỹ sư Dương Nhậm Hào lưu mã nguồn các mô-đun của ứng dụng vào máy tính cá nhân và lên đám mây, sau khi nhảy sang công ty khác thì làm ứng dụng cùng loại, bị truy cứu theo tội xâm phạm bí mật kinh doanh (toàn quốc) +- Chi phí: Không tốn tiền. Trước khi nghỉ việc, xoá sạch tài liệu công ty trong máy tính cá nhân và trên ổ đĩa mạng. Chỗ khó là có tệp do mình tự tay viết từng dòng, không nỡ +- Nói thẳng: Mã nguồn do bạn tự viết cũng thuộc quyền tác giả của công ty, danh sách khách hàng cũng là bí mật thương mại, mang đi là lấy đồ của công ty. Có nhân viên nghỉ việc mang mã nguồn hệ thống an ninh sang công ty mới, bản thân bị phạt tù 5 năm, phạt tiền 50 vạn tệ, công ty mới bị phạt tiền 400 vạn tệ. Có người lợi lỗ hổng hệ thống để tải hơn 3.3 vạn tệ liệu thiết kế, bị phạt tù 1 năm 10 tháng. Công ty còn có thể kiện bạn đòi bồi thường riêng. +- Lợi ích: Vi phạm nghĩa vụ bảo mật, tiết lộ bí mật thương mại hoặc tự lấy để dùng, khi đặc biệt nghiêm trọng thì phạt tù dưới 3 năm, kèm phạt tiền hoặc chỉ phạt tiền, đặc biệt nghiêm trọng hơn thì 3 đến 10 năm kèm phạt tiền. Luật Chống cạnh tranh không lành mạnh còn liệt kê hai việc là xâm phạm: lấy bí mật thương mại bằng「xâm nhập điện tử」(dùng kỹ thuật để xâm nhập hệ thống); xúi giục, giúp người khác vi phạm nghĩa vụ bảo mật. Vụ án chính thức: Tôn Nhậm Minh và những nhân viên nghỉ việc khác mang mã nguồn hệ thống an ninh cùng thuật toán lõi đi, dùng cho sản phẩm của công ty mới, Tôn Nhậm Minh bị phạt tù 5 năm, phạt tiền 50 vạn tệ, công ty mới bị phạt tiền 400 vạn tệ. Điền Mộc lợi lỗ hổng hệ thống để tải hơn 3.3 vạn tệ liệu thiết kế, bị phạt tù 1 năm 10 tháng, phạt tiền 10 vạn tệ. Kỹ sư Dương Mộc lưu mã nguồn các mô-đun ứng dụng vào máy tính cá nhân và đám mây, sau khi nhảy sang công ty khác lại làm ứng dụng cùng loại, bị truy cứu theo tội xâm phạm bí mật thương mại (toàn quốc) - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百一十九条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 反不正当竞争法（2025 年修订，第十条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关知识产权保护典型案例（案例七）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2020). 2019 年度检察机关保护知识产权典型案例（案例十六）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例五）. -- Ghi chú: Mã「do mình tự viết」thì quyền tác giả vẫn thuộc công ty, điểm này xem mục 13 của phần này (mã viết bằng nguồn lực của công ty thuộc về công ty). Mang đi là lấy đồ của công ty. Danh sách khách hàng là thông tin kinh doanh, cũng là bí mật kinh doanh. Ngoài án phạt tù, công ty còn có thể kiện riêng đòi bồi thường dân sự. +- Ghi chú: 「自己写的代码」cũng thuộc quyền tác giả của công ty, điểm này xem mục 13 (mã nguồn viết bằng nguồn lực công ty thì thuộc công ty). Mang đi là lấy đồ của công ty. Danh sách khách hàng là thông tin kinh doanh, cũng là bí mật thương mại. Ngoài án tù, công ty còn có thể kiện bồi thường dân sự riêng. -### 8. Không chạy chương trình của mình trên máy tính, máy chủ, camera của người khác; không dùng máy của công ty để đào mỏ +### 8. Không chạy chương trình của bạn trên máy tính, máy chủ, camera của người khác, không dùng máy của công ty để đào mỏ -- Chi phí: Không tốn tiền, cũng không tốn thời gian. -- Nói thẳng: Kiểm soát máy tính, máy chủ, camera của người khác mà không được phép, tới 20 máy là phạt tù dưới 3 năm, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Không đủ cấu thành tội phạm thì vẫn bị giữ dưới 5 ngày, còn có thể phạt tiền 5 vạn đến 50 vạn tệ. Người đã từng chịu hình phạt hình sự thì không được làm các chức vụ then chốt về quản trị an ninh mạng và vận hành mạng cả đời. -- Lợi ích: Kiểm soát trái phép hệ thống thông tin máy tính, khi nghiêm trọng thì phạt tù dưới 3 năm, phạt kèm hoặc chỉ phạt tiền, đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm. Kiểm soát trái phép trên 20 máy là「nghiêm trọng」. Chưa đủ cấu thành tội phạm thì công an tịch thu thu nhập bất hợp pháp, bắt giữ dưới 5 ngày, và có thể phạt tiền 5 vạn đến 50 vạn tệ. Mức nặng hơn thì bị giữ 5 đến 15 ngày, và có thể phạt tiền 10 vạn đến 100 vạn tệ. Người đã chịu xử phạt hành chính thì trong 5 năm, người đã chịu hình phạt hình sự thì cả đời, không được làm các chức vụ then chốt về quản trị an ninh mạng và vận hành mạng. Vụ án chính thức: Lý Nhậm Hào mua địa chỉ IP của camera mạng, dùng phần mềm kiểm soát 105 camera để theo dõi đời tư, bị phạt tù 3 năm hoãn thi hành 5 năm, phạt tiền 2 vạn tệ. Ngô Nhậm Hào cùng những người khác cài mã độc vào bo mạch chủ điện thoại, kiểm soát hơn 330 vạn máy điện thoại để gửi lại mã xác minh, bị phạt tù từ 2 năm đến 4 năm 6 tháng (toàn quốc) +- Chi phí: Không tốn tiền, cũng không mất thời gian +- Nói thẳng: Kiểm soát máy tính, máy chủ, camera của người khác mà không được phép, tới 20 máy là phạt tù dưới 3 năm, đặc biệt nghiêm trọng thì 3 đến 7 năm. Không đủ thành tội phạm thì vẫn bị giữ dưới 5 ngày, còn có thể phạt tiền 5 vạn đến 50 vạn tệ. Người đã chịu hình phạt hình sự thì suốt đời không được làm vị trí then chốt trong quản lý an ninh mạng và vận hành mạng. +- Lợi ích: Thực hiện kiểm soát trái phép đối với hệ thống máy tính, khi đặc biệt nghiêm trọng thì phạt tù dưới 3 năm, kèm phạt tiền hoặc chỉ phạt tiền, đặc biệt nghiêm trọng hơn thì 3 đến 7 năm. Kiểm soát trái phép từ 20 máy trở lên đã là「情节严重」. Chưa đủ thành tội phạm thì công an tịch thu thu nhập bất hợp pháp, bị giữ dưới 5 ngày, và có thể phạt tiền 5 vạn đến 50 vạn tệ. Tình tiết nặng hơn thì bị giữ 5 đến 15 ngày, và có thể phạt tiền 10 vạn đến 100 vạn tệ. Người đã chịu xử phạt hành chính thì 5 năm, đã chịu hình phạt hình sự thì suốt đời, không được làm vị trí then chốt trong quản lý an ninh mạng và vận hành mạng. Vụ án chính thức: Lý Mộc mua địa chỉ IP của camera mạng, dùng phần mềm kiểm soát 105 camera để theo dõi riêng tư, bị phạt tù 3 năm hoãn 5 năm, phạt tiền 2 vạn tệ. Ngô Mộc và những người khác cài mã độc vào bo mạch chủ điện thoại, kiểm soát hơn 330 vạn chiếc điện thoại để gửi mã xác minh về, bị phạt tù 2 năm đến 4 năm 6 tháng (toàn quốc) - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十九、六十六条；2016 年文本为第二十七、六十三条，终身禁业为修正后的第六十六条第三款）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 山东滨州开发区：办理一起非法控制计算机信息系统案. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 充分发挥检察职能 推进网络空间治理典型案例（案例六）. -- Ghi chú: Dùng máy chủ công ty để đào mỏ thì không tìm được vụ án nào do Viện kiểm sát nhân dân tối cao, Tòa án nhân dân tối cao công bố; mục này viết theo điều luật về「kiểm soát hệ thống của người khác khi không được phép」.「Cả đời không được làm chức vụ then chốt」là hậu quả kèm theo trong Luật An ninh mạng, ngoài mức phạt tiền và án tù, với người làm kỹ thuật thì nặng hơn mức phạt tiền. +- Ghi chú: Dùng máy chủ công ty để đào mỏ thì không tìm thấy vụ án nào được Viện kiểm sát nhân dân tối cao hay Tòa án nhân dân tối cao công bố, nên mục này viết theo điều luật về「kiểm soát hệ thống của người khác khi không được phép」. 「终身不得从事关键岗位」là hậu quả kèm theo mà Luật An ninh mạng đặt ra ngoài khoản phạt tiền và án tù, với người làm kỹ thuật thì nặng hơn khoản phạt tiền. -### 9. Không thử hệ thống của người khác khi không có văn bản ủy quyền;「vì thiện chí」và「báo cáo sau」đều không phải lý do thoát tội +### 9. Không có văn bản cho phép thì không thử hệ thống của người khác,「vì thiện ý」và「báo cáo sau đó」đều không phải lý do để thôi tội -- Chi phí: Không tốn tiền. Trước khi tham gia SRC của hãng hoặc chương trình thử nghiệm cộng đồng, phải có văn bản ghi rõ phạm vi ủy quyền, danh sách mục tiêu và khoảng thời gian được thử. SRC là trung tâm ứng cứu sự cố an ninh mạng do chính hãng lập ra, chuyên nhận lỗ hổng từ bên ngoài báo lên. Xác minh xong lỗ hổng có thật thì dừng. Chỗ khó là bỏ được cái thói「thử luôn một cái cho tiện». -- Nói thẳng: Có kết án hay không phụ thuộc vào việc có ủy quyền không và lấy được bao nhiêu, chứ không phụ thuộc vào ý bạn. Có 500 bộ tài khoản, mật khẩu thường, hoặc thu được 5000 tệ, hay làm bên kia thiệt hại 1 vạn tệ, là đủ phạt tù dưới 3 năm. Để chứng minh lỗ hổng mà tiện tay tải một bảng người dùng thì thường một lần đã vượt 500 bộ. Đã có tài khoản, mật khẩu trong tay mà dùng ra ngoài phạm vi ủy quyền thì cũng là xâm nhập. -- Lợi ích: Hệ thống của công ty và cá nhân thông thường thuộc các hệ thống thông tin máy tính ngoài lĩnh vực công việc nhà nước, xây dựng quốc phòng, khoa học kỹ thuật tiên tiến. Xâm nhập loại hệ thống này, hoặc dùng biện pháp kỹ thuật khác để lấy dữ liệu được lưu, xử lý, truyền trong đó, khi nghiêm trọng thì phạt tù dưới 3 năm hoặc giam giữ, phạt kèm hoặc chỉ phạt tiền. Đặc biệt nghiêm trọng thì phạt tù 3 đến 7 năm kèm phạt tiền.「Nghiêm trọng」có vài tiêu chuẩn: lấy được thông tin xác thực danh tính của dịch vụ tài chính trên mạng như thanh toán, giao dịch chứng khoán, giao dịch hợp đồng tương lai (tài khoản, mật khẩu, các loại tương tự) từ 10 bộ trở lên; lấy được thông tin xác thực danh tính khác từ 500 bộ trở lên; kiểm soát trái phép hệ thống thông tin máy tính trên 20 máy; thu nhập bất hợp pháp trên 5000 tệ, hoặc gây thiệt hại kinh tế trên 1 vạn tệ. Gấp từ 5 lần trở lên so với các con số trên là「đặc biệt nghiêm trọng」. Vụ án chính thức: vụ án hướng dẫn số 36 của viện kiểm sát xác lập một câu:「超出授权范围使用账号、密码登录计算机信息系统，属于侵入计算机信息系统的行为」, nghĩa là vốn đã có tài khoản, mật khẩu nhưng dùng ra ngoài phạm vi ủy quyền thì bản thân hành vi đó đã là xâm nhập. Ba người trong vụ án dùng tài khoản, mật khẩu và mã thông báo nắm được trong công việc, đăng nhập từ địa điểm khác vào hệ thống nội bộ của công ty, tải dữ liệu nằm ngoài phạm vi công việc, rồi bán ra trên mạng, thu nhập bất hợp pháp 37000 tệ. Ba người lần lượt bị phạt tù 4 năm, 3 năm 9 tháng, 4 năm, mỗi người phạt tiền 4 vạn tệ (toàn quốc) +- Chi phí: Không tốn tiền. Trước khi tham gia SRC của nhà sản xuất hoặc chương trình thử nghiệm công khai, phải có văn bản ghi rõ phạm vi được phép, danh sách mục tiêu và khung thời gian được thử. SRC là trung tâm ứng cứu an ninh do chính nhà sản xuất lập, chuyên nhận lỗ hổng báo từ bên ngoài. Xác minh tới mức lỗ hổng đúng là tồn tại thì dừng. Chỗ khó là từ bỏ thói「thử luôn một cái cho vui」 +- Nói thẳng: Có bị kết tội hay không phụ thuộc vào việc đã có văn bản cho phép chưa và đã lấy được bao nhiêu, không xét ý định của bạn. Lấy được 500 bộ tài khoản mật khẩu thường, hoặc kiếm được 5000 tệ, làm bên kia thiệt hại 1 vạn tệ, đã đủ phạt tù dưới 3 năm. Để chứng minh lỗ hổng mà tiện tay xuất một bảng người dùng thì một lần là vượt 500 bộ. Có tài khoản mật khẩu trong tay mà dùng ngoài phạm vi được phép thì cũng là xâm nhập. +- Lợi ích: Hệ thống của công ty và cá nhân thông thường thuộc loại hệ thống máy tính ngoài phạm vi công việc quốc gia, xây dựng quốc phòng và khoa học công nghệ tiên tiến. Xâm nhập loại hệ thống này, hoặc dùng kỹ thuật khác để lấy dữ liệu lưu trữ, xử lý, truyền trong đó, khi đặc biệt nghiêm trọng thì phạt tù dưới 3 năm hoặc giam giữ, kèm phạt tiền hoặc chỉ phạt tiền. Đặc biệt nghiêm trọng thì 3 đến 7 năm kèm phạt tiền. 「情节严重」có vài tiêu chí: lấy được thông tin nhận dạng của các dịch vụ tài chính qua mạng như thanh toán, giao dịch chứng khoán, giao dịch hợp đồng tương lai (tài khoản mật khẩu các loại) 10 bộ trở lên; lấy được thông tin nhận dạng khác 500 bộ trở lên; kiểm soát trái phép hệ thống máy tính 20 máy trở lên; thu nhập bất hợp pháp 5000 tệ trở lên, hoặc gây thiệt hại kinh tế 1 vạn tệ trở lên. Gấp 5 lần các con số trên là「情节特别严重」. Vụ án chính thức: Bản án hướng dẫn số 36 của viện kiểm sát xác lập một câu:「vượt ra ngoài phạm vi được phép mà dùng tài khoản, mật khẩu để đăng nhập hệ thống máy tính là hành vi xâm nhập hệ thống máy tính」. Nghĩa là bạn vốn đã có tài khoản và mật khẩu, nhưng dùng ngoài phạm vi được phép thì bản thân hành vi đó đã là xâm nhập. Ba người trong vụ đó dùng tài khoản, mật khẩu và mã thông báo nắm được trong công việc, đăng nhập từ xa vào hệ thống nội bộ công ty, tải dữ liệu ngoài phạm vi công việc, rồi bán ra mạng, thu nhập bất hợp pháp 37000 tệ. Ba người bị phạt tù lần lượt 4 năm, 3 năm 9 tháng, 4 năm, mỗi người phạt tiền 4 vạn tệ (toàn quốc) - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第一款、第二款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. （Cục Công an thành phố Thâm Quyến đăng lại）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2017). 第九批指导性案例（检例第 36 号，卫梦龙、龚旭、薛东东非法获取计算机信息系统数据案）. -- Ghi chú: Vụ án hướng dẫn số 36 vốn là vụ bán dữ liệu để hưởng lợi. Dẫn nó chỉ để lấy câu「đăng nhập ra ngoài phạm vi ủy quyền cũng là xâm nhập」, không phải nói thử nghiệm thiện chí bị phạt nặng như vậy. Bạn định làm gì, sau đó có báo cáo hay không, đều không xoá được tội danh, chỉ có thể ảnh hưởng tới mức tù và có bị khởi tố hay không. Hãng cảm ơn trước rồi quay lại báo việc, hai việc đó không mâu thuẫn, cảm ơn không phải là ủy quyền. Lúc viết phần này, không tìm được trên trang Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao vụ án thử nghiệm thiện chí nào đối chiếu từng chữ, nên chỉ viết theo điều luật và tiêu chuẩn xử phạt. Muốn làm thì đi theo SRC công khai của hãng, hoặc có văn bản ủy thác đo đạc, ghi rõ phạm vi, mục tiêu và khoảng thời gian vào hồ sơ. Chứng minh lỗ hổng tồn tại bằng ít bằng chứng nhất, đừng lấy dữ liệu hàng loạt. Mức chưa đủ cấu thành tội phạm vẫn phải chịu xử phạt hành chính, kèm cấm hành nghề, mức phạt tiền và thời hạn giống mục 8 của phần này (không chạy chương trình của mình trên máy của người khác). Sau khi phát hiện lỗ hổng thì xử lý thế nào, xem mục 10 của phần này (báo cáo lỗ hổng theo quy định). -### 10. Báo cáo lỗ hổng theo quy định; trước khi vá xong không công khai chi tiết, không phát hành công cụ khai thác, không đưa ra nước ngoài +- Ghi chú: Bản án hướng dẫn số 36 vốn là vụ bán dữ liệu để hưởng lợi. Dẫn nó chỉ để nói câu「đăng nhập vượt quá phạm vi được phép cũng là xâm nhập」, không phải bảo thử nghiệm thiện ý sẽ bị kết án nặng như nhau. Ý bạn định làm gì, sau đó có báo cáo hay không, không xoá được tội danh, chỉ có thể ảnh hưởng tới mức phạt và việc có khởi tố hay không. Nhà sản xuất cảm ơn trước rồi báo việc sau, hai việc đó không mâu thuẫn, cảm ơn không phải là cho phép. Khi viết phần này, trên trang của Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao không tìm thấy vụ án thử nghiệm thiện ý nào đối chiếu từng chữ, nên chỉ viết theo điều luật và tiêu chuẩn xử phạt. Muốn làm thì đi theo SRC công khai của nhà sản xuất, hoặc theo hợp đồng đo thử bằng văn bản, ghi rõ phạm vi, mục tiêu và khung thời gian vào hồ sơ. Chứng minh lỗ hổng bằng lượng bằng chứng tối thiểu, không lấy dữ liệu hàng loạt. Mức không đủ thành tội phạm vẫn chịu xử phạt hành chính và cấm hành nghề, mức phạt tiền và thời hạn thì giống mục 8 (không chạy chương trình của bạn trên máy của người khác). Sau khi phát hiện lỗ hổng thì xử lý thế nào, xem mục 10 (báo lỗ hổng theo quy định). +### 10. Báo lỗ hổng theo quy định, trước khi vá không công bố chi tiết, không đưa công cụ khai thác, không đưa ra nước ngoài -- Chi phí: Không tốn tiền. Báo lỗ hổng cho nhà cung cấp sản phẩm, hoặc cho nền tảng chính thức. Trước khi hãng vá xong, không đăng chi tiết, cũng không đăng mã kiểm chứng. -- Nói thẳng: Cá nhân cũng nằm trong phạm vi điều chỉnh của Quy định về quản lý lỗ hổng an ninh sản phẩm mạng. Trước khi nhà sản xuất đưa ra cách vá, bạn không được công khai lỗ hổng, cũng không được công khai chi tiết lỗ hổng của hệ thống đang dùng. Không phát hành chương trình, công cụ chuyên khai thác lỗ hổng; khi công khai phải nói kèm cách vá. Lỗ hổng chưa công khai thì không được đưa cho tổ chức, cá nhân nước ngoài ngoài nhà sản xuất. -- Lợi ích: Quy định này quản lý ba nhóm đối tượng: nhà cung cấp sản phẩm mạng trong nước, đơn vị vận hành mạng, và các tổ chức hoặc cá nhân tham gia hoạt động phát hiện, thu thập, công khai lỗ hổng. Không tổ chức, cá nhân nào được dùng lỗ hổng để làm hoạt động gây nguy hại an ninh mạng, không được thu thập, bán, công khai thông tin lỗ hổng trái phép. Công khai thông tin lỗ hổng ra xã hội thì phải tuân thủ năm điều. Một, không được công khai trước khi nhà cung cấp sản phẩm có biện pháp vá. Hai, không được công khai tình hình chi tiết rằng hệ thống mạng, hệ thống thông tin và thiết bị mà đơn vị vận hành mạng đang dùng có lỗ hổng. Ba, không được cố tình phóng đại mức nguy hại và rủi ro, không được dùng thông tin lỗ hổng để làm nội dung độc hại hoặc lừa đảo. Bốn, không được công khai hoặc cung cấp chương trình, công cụ chuyên dùng để khai thác lỗ hổng gây nguy hại an ninh mạng. Năm, khi công khai phải công khai kèm biện pháp vá hoặc phòng ngừa. Ngoài ra, thông tin lỗ hổng chưa công khai thì không được cung cấp cho tổ chức, cá nhân nước ngoài ngoài nhà cung cấp sản phẩm. Quy định khuyến khích báo lỗ hổng cho bốn nền tảng chính thức. Một là nền tảng chia sẻ thông tin mối đe doạ và lỗ hổng an ninh mạng của Bộ Công nghiệp và Thông tin. Hai là nền tảng lỗ hổng của Trung tâm Thông báo Quốc gia về an ninh mạng và thông tin. Ba là nền tảng lỗ hổng của Trung tâm Xử lý Khẩn cấp Kỹ thuật Mạng máy tính quốc gia. Bốn là kho lỗ hổng của Trung tâm Đánh giá An ninh thông tin Trung Quốc. Vi phạm quy định mà thu thập, công khai thông tin lỗ hổng thì Bộ Công nghiệp và Thông tin, Bộ Công an xử lý theo trách nhiệm của mình. Trường hợp cấu thành tình huống theo quy định của Luật An ninh mạng thì xử phạt theo luật đó. Công khai thông tin an ninh mạng như lỗ hổng hệ thống ra xã hội thì phải tuân thủ quy định của nhà nước. Vi phạm thì cơ quan quản lý có liên quan yêu cầu chấp hành điều chỉnh, cảnh cáo, có thể phạt tiền từ hơn 1 vạn tệ đến dưới 10 vạn tệ. Không chấp hành điều chỉnh hoặc nghiêm trọng thì phạt tiền từ 10 vạn tệ đến 100 vạn tệ. Còn có thể bị yêu cầu tạm dừng hoạt động liên quan, đình chỉ hoạt động để cải tổ, đóng website hoặc ứng dụng. Cũng có thể bị thu hồi giấy phép hoạt động liên quan hoặc thu hồi giấy đăng ký kinh doanh. Người chịu trách nhiệm trực tiếp và người liên quan trực tiếp khác bị phạt tiền từ hơn 1 vạn tệ đến dưới 10 vạn tệ (toàn quốc) +- Chi phí: Không tốn tiền. Báo lỗ hổng cho nhà cung cấp sản phẩm, hoặc cho nền tảng chính thức. Trước khi nhà sản xuất vá xong, không đưa chi tiết, cũng không đưa mã dùng để kiểm chứng +- Nói thẳng: Cá nhân cũng nằm trong phạm vi《Quy định quản lý lỗ hổng bảo mật của sản phẩm mạng》. Trước khi nhà sản xuất đưa ra cách vá, bạn không được công bố lỗ hổng, cũng không được công bố chi tiết lỗ hổng của hệ thống đang dùng. Không được đưa chương trình và công cụ chuyên khai thác lỗ hổng, khi công bố thì phải nói rõ cách vá. Lỗ hổng chưa công bố thì không được đưa cho tổ chức và cá nhân ở nước ngoài ngoài nhà sản xuất. +- Lợi ích: Quy định này quản lý ba nhóm chủ thể: nhà cung cấp sản phẩm mạng trong nước, đơn vị vận hành mạng, và tổ chức hoặc cá nhân làm hoạt động phát hiện, thu thập, công bố lỗ hổng. Không tổ chức hay cá nhân nào được dùng lỗ hổng để làm hoạt động gây hại an ninh mạng, không được thu thập, bán, công bố thông tin lỗ hổng trái phép. Công bố thông tin lỗ hổng ra xã hội thì phải giữ năm điều. Một, không được công bố trước khi nhà cung cấp sản phẩm đưa ra biện pháp vá. Hai, không được công bố tình hình chi tiết về lỗ hổng của mạng, hệ thống thông tin và thiết bị mà đơn vị vận hành mạng đang dùng. Ba, không được cố tình thổi phồng mức độ nguy hại và rủi ro, không được dùng thông tin lỗ hổng để bóp méo tin độc hoặc lừa đảo. Bốn, không được công bố hoặc cung cấp chương trình, công cụ chuyên dùng để khai thác lỗ hổng gây hại an ninh mạng. Năm, khi công bố thì phải công bố kèm biện pháp vá hoặc phòng ngừa. Ngoài ra, thông tin lỗ hổng chưa công bố thì không được cung cấp cho tổ chức hoặc cá nhân ở nước ngoài ngoài nhà cung cấp sản phẩm. Quy định khuyến khích báo lỗ hổng cho bốn nền tảng chính thức. Một là Nền tảng chia sẻ thông tin mối đe dọa và lỗ hổng an ninh mạng của Bộ Công nghiệp và Thông tin (工业和信息化部). Hai là nền tảng lỗ hổng của Trung tâm Thông báo Quốc gia về An nạo mạng và Thông tin. Ba là nền tảng lỗ hổng của Trung tâm Điều phối Xử lý Khẩn cấp Kỹ thuật Mạng Máy tính Quốc gia. Bốn là kho lỗ hổng của Trung tâm Đánh giá An ninh Thông tin Trung Quốc. Vi phạm quy định mà thu thập, công bố thông tin lỗ hổng thì Bộ Công nghiệp và Thông tin, Bộ Công an xử lý theo đúng trách nhiệm của mình. Cấu thành tình huống quy định trong Luật An ninh mạng thì phạt theo luật đó. Công bố ra xã hội thông tin an ninh mạng như lỗ hổng hệ thống thì phải tuân theo quy định của Nhà nước. Vi phạm thì cơ quan quản lý liên quan yêu cầu sửa, đưa ra cảnh cáo, có thể phạt tiền từ 1 vạn đến 10 vạn tệ. Không sửa hoặc tình tiết nghiêm trọng thì phạt tiền từ 10 vạn đến 100 vạn tệ. Còn có thể bị yêu cầu dừng hoạt động liên quan, đình chỉ kinh doanh để chỉnh phục, đóng website hoặc ứng dụng. Cũng có thể bị thu hồi giấy phép kinh doanh liên quan hoặc giấy phép đăng ký kinh doanh. Người phụ trách trực tiếp và người chịu trách nhiệm trực tiếp khác bị phạt tiền từ 1 vạn đến 10 vạn tệ (toàn quốc) - Mức bằng chứng: A -- Nguồn: Bộ Công nghiệp và Thông tin (工业和信息化部), Văn phòng Quốc gia Quản lý Thông tin Internet (国家互联网信息办公室), Bộ Công an (公安部) (2021). 网络产品安全漏洞管理规定（工信部联网安〔2021〕66 号，第二、四、九、十、十四条，2021 年 9 月 1 日施行）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十八、六十五条；2016 年文本为第二十六、六十二条，规定第十四条援引的是 2016 年文本的条号）. -- Ghi chú: Quy định nói rõ cá nhân nằm trong đối tượng quản lý, câu「tôi chỉ là người nghiệp dư」không nằm trong ngoại lệ. Gửi lỗ hổng của hệ thống trong nước ra nền tảng thưởng lỗ hổng ở nước ngoài thì đúng vào câu「lỗ hổng chưa công khai không được đưa ra nước ngoài」. Có đủ tư cách để thử hay không, xem mục 9 của phần này (không thử hệ thống của người khác khi không có văn bản ủy quyền); còn sau khi có lỗ hổng thì xử lý thế nào là chuyện khác. Lỗ hổng tìm ra hợp pháp cũng có thể bị phạt vì cách công khai; lỗ hổng tìm ra khi chưa được ủy quyền thì cả khâu thử lẫn khâu công khai đều có vấn đề. Trong các kênh báo cáo, nền tảng của Bộ Công nghiệp và Thông tin và Trung tâm ứng cứu sự cố mạng quốc gia (CNCERT) được dùng nhiều nhất. Quy định không tách riêng trường hợp「đã báo hãng mà hãng không vá」; cách làm thực tế là để lại hồ sơ quá trình thương lượng và báo cáo, đồng thời báo cho nền tảng chính thức. +- Nguồn: Bộ Công nghiệp và Thông tin (工业和信息化部), Văn phòng Quản lý mạng quốc gia (国家互联网信息办公室), Bộ Công an (公安部) (2021). 网络产品安全漏洞管理规定（工信部联网安〔2021〕66 号，第二、四、九、十、十四条，2021 年 9 月 1 日施行）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十八、六十五条；2016 年文本为第二十六、六十二条，规定第十四条援引的是 2016 年文本的条号）. +- Ghi chú: Quy định ghi rõ cá nhân vào đối tượng bị quản lý,「tôi chỉ là người nghiệp dư」không nằm trong ngoại lệ. Gửi lỗ hổng của hệ thống trong nước lên nền tảng thưởng lỗ hổng ở nước ngoài thì rơi đúng vào câu「lỗ hổng chưa công bố không được cung cấp ra nước ngoài」. Có đủ tư cách để thử hay không, xem mục 9 (không có văn bản cho phép thì không thử hệ thống của người khác), đó là chuyện khác với sau khi có lỗ hổng thì xử lý thế nào. Lỗ hổng tìm ra hợp pháp vẫn có thể bị phạt vì cách công bố; lỗ hổng tìm ra khi chưa được phép thì cả khâu thử lẫn khâu công bố đều có vấn đề. Trong các kênh báo cáo, nền tảng của Bộ Công nghiệp và Thông tin và Trung tâm ứng cứu Internet quốc gia (CNCERT) được dùng nhiều nhất. Quy định không tách riêng trường hợp「đã báo nhà sản xuất mà nhà sản xuất không vá」, nên cách làm thực tế là để lại hồ sơ quá trình thương lượng và báo cáo, đồng thời báo cho nền tảng chính thức. -### 11. Không bán công cụ vượt tường, không bán tài khoản VPN, không dựng loại nút truy cập đó cho người khác +### 11. Không bán công cụ vượt tường, không bán tài khoản VPN, không dựng loại nút mạng đó cho người khác -- Chi phí: Không tốn tiền, cũng không tốn thời gian. -- Nói thẳng: Tự dựng đường truyền để ra nước ngoài, hoặc dùng kênh quốc tế ngoài kênh công cộng của nhà nước, công an có thể yêu cầu dừng kết nối, cảnh cáo, phạt tiền dưới 1.5 vạn tệ và tịch thu thu nhập bất hợp pháp. Kinh doanh theo cách đó thì cơ quan xử lý có thể xử lý theo tội kinh doanh phi pháp, nghiêm trọng thì phạt tù dưới 5 năm. Mục này nói về việc bán và dựng nút truy cập. Người cá nhân tự dùng thì mức phạt cao nhất là cảnh cáo cộng phạt tiền. -- Lợi ích: Kết nối trực tiếp ra quốc tế thì bắt buộc phải dùng kênh ra vào quốc tế do mạng viễn thông công cộng của nhà nước cung cấp. Không cá nhân, tổ chức nào được tự lập hoặc dùng kênh khác: không được tự dựng đường truyền ra nước ngoài, cũng không được dùng đường do người khác dựng. Vi phạm thì công an yêu cầu dừng kết nối, cảnh cáo, có thể phạt kèm tiền dưới 1.5 vạn tệ, tịch thu thu nhập bất hợp pháp. Kinh doanh theo cách đó thì cơ quan xử lý có thể xử lý theo tội kinh doanh phi pháp. Nghiêm trọng thì phạt tù dưới 5 năm, phạt kèm hoặc chỉ phạt tiền bằng 1 đến 5 lần thu nhập bất hợp pháp. Đặc biệt nghiêm trọng thì phạt tù trên 5 năm (toàn quốc) +- Chi phí: Không tốn tiền, cũng không mất thời gian +- Nói thẳng: Tự dựng đường truyền để ra nước ngoài, hoặc dùng kênh quốc tế ngoài kênh công cộng của Nhà nước, công an có thể yêu cầu dừng kết nối mạng, đưa ra cảnh cáo, phạt tiền dưới 1.5 vạn tệ và tịch thu thu nhập bất hợp pháp. Kinh doanh dựa trên đó thì cơ quan điều tra có thể xử lý theo tội kinh doanh phi pháp, đặc biệt nghiêm trọng thì phạt tù dưới 5 năm. Mục này nói về việc bán và dựng nút mạng. Cá nhân tự dùng thì mức phạt cao nhất là cảnh cáo cộng phạt tiền. +- Lợi ích: Kết nối trực tiếp với mạng quốc tế thì phải dùng kênh ra vào quốc tế do mạng viễn thông công cộng của Nhà nước cung cấp. Không tổ chức hay cá nhân nào được tự lập hoặc dùng kênh khác: không được tự dựng đường truyền để ra nước ngoài, cũng không được dùng đường do người khác dựng. Vi phạm thì công an yêu cầu dừng kết nối mạng, đưa ra cảnh cáo, có thể phạt tiền kèm dưới 1.5 vạn tệ, tịch thu thu nhập bất hợp pháp. Kinh doanh dựa trên đó thì cơ quan điều tra có thể xử lý theo tội kinh doanh phi pháp. Đặc biệt nghiêm trọng thì phạt tù dưới 5 năm, kèm phạt tiền bằng 1 đến 5 lần thu nhập bất hợp pháp, hoặc chỉ phạt tiền. Đặc biệt nghiêm trọng hơn thì 5 năm trở lên (toàn quốc) - Mức bằng chứng: B -- Nguồn: Hội đồng Bộ trưởng (国务院) (1997). 计算机信息网络国际联网管理暂行规定（第六、十四条）. （Ủy ban Nông nghiệp và Nông thôn thành phố Trùng Khánh đăng lại）；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百二十五条第四项）. -- Ghi chú: Lúc viết phần này, trên trang Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao không tìm được vụ án vượt tường nào được công bố, cách xử lý ở các nơi cũng rất khác nhau, nên để mức B. Người cá nhân tự dùng thì mức phạt nặng nhất là cảnh cáo cộng phạt tiền như điều mười bốn của quy định tạm thời viết. +- Nguồn: Quốc vụ viện (国务院) (1997). 计算机信息网络国际联网管理暂行规定（第六、十四条）. （Ủy ban Nông nghiệp và Nông thôn thành phố Trùng Khánh đăng lại）；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百二十五条第四项）. +- Ghi chú: Khi viết phần này, trên trang của Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao không tìm thấy vụ án vượt tường nào được công bố, cách xử lý ở các địa phương cũng khác nhau khá nhiều, nên để mức B. Cá nhân tự dùng thì mức phạt nặng nhất là cảnh cáo cộng phạt tiền quy định ở khoản 14 của Quy định tạm thời. -### 12. Đã ký thoả thuận không cạnh tranh, sau khi nghỉ việc công ty không trả bồi thường hằng tháng thì gửi văn bản đòi; quá 3 tháng không trả thì có thể chấm dứt; với chức vụ không tiếp cận bí mật kinh doanh thì có thể yêu cầu xác nhận điều khoản không có hiệu lực +### 12. Đã ký thỏa thuận không cạnh tranh, sau khi nghỉ việc công ty không trả bồi thường hằng tháng thì gửi văn bản yêu cầu, đủ 3 tháng không trả thì có thể chấm dứt; vị trí chưa từng tiếp cận bí mật thương mại thì có thể yêu cầu xác nhận điều khoản không có hiệu lực -- Chi phí: Không tốn tiền. Lúc nghỉ việc giữ lại bản gốc thoả thuận không cạnh tranh, cùng chứng cứ công ty không trả tiền hằng tháng. -- Nói thẳng: Không cạnh tranh nghĩa là trong một thời gian sau khi nghỉ việc không được đi làm cho công ty cùng ngành, tối đa 2 năm, công ty phải trả bồi thường hằng tháng. Không thỏa thuận con số thì trả bằng 30% lương trung bình 12 tháng trước khi nghỉ việc, không thấp hơn mức lương tối thiểu địa phương. Do lý do từ phía công ty mà 3 tháng không trả thì bạn có thể yêu cầu chấm dứt. Từ năm 2025, với chức vụ không biết, không tiếp cận bí mật kinh doanh thì có thể yêu cầu xác nhận điều khoản không có hiệu lực. -- Lợi ích: Không cạnh tranh chỉ ràng buộc lãnh đạo, nhân sự kỹ thuật cao cấp, và những người khác có nghĩa vụ bảo mật. Thời hạn tối đa 2 năm. Công ty phải trả bồi thường kinh tế hằng tháng trong thời hạn đó. Không thỏa thuận số tiền thì trả bằng 30% lương trung bình 12 tháng trước khi nghỉ việc, đồng thời không thấp hơn mức lương tối thiểu địa phương. Do lý do từ phía công ty mà 3 tháng không trả thì người lao động có thể yêu cầu chấm dứt ràng buộc này. Công ty chủ động chấm dứt thì phải trả thêm 3 tháng bồi thường. Từ năm 2025, người lao động ở vị trí không biết, không tiếp cận bí mật kinh doanh có thể yêu cầu xác nhận điều khoản không có hiệu lực. Phần thỏa thuận vượt quá phần bí mật mà người đó thực sự biết thì vô hiệu. Vụ án chính thức: một công ty dược kiện giám đốc công nghệ cao cấp Trịnh Nhậm Hào, đòi bồi thường vi phạm 7.1 triệu tệ. Tòa án xác định sản phẩm của công ty mới không thay thế được sản phẩm công ty cũ, phạm vi không cạnh tranh phải giới hạn ở phần bí mật mà ông thực sự biết, nên bác toàn bộ (toàn quốc) +- Chi phí: Không tốn tiền. Khi nghỉ việc, giữ lại bản gốc thỏa thuận không cạnh tranh, cùng bằng chứng công ty không trả tiền hằng tháng +- Nói thẳng: Không cạnh tranh nghĩa là sau khi nghỉ việc, một khoảng thời gian không được đi làm cho công ty cùng nghề, tối đa 2 năm, công ty phải trả bồi thường hằng tháng. Không thỏa thuận con số thì trả bằng 30% mức lương trung bình của 12 tháng trước khi nghỉ việc, không thấp hơn mức lương tối thiểu địa phương. Do lý do của công ty mà 3 tháng chưa trả, bạn có thể yêu cầu chấm dứt. Từ năm 2025, vị trí chưa từng tiếp cận bí mật thương mại thì có thể yêu cầu xác nhận điều khoản không có hiệu lực. +- Lợi ích: Ràng buộc không cạnh tranh chỉ áp dụng với lãnh đạo, nhân viên kỹ thuật cao cấp, và người khác có nghĩa vụ bảo mật. Thời hạn tối đa 2 năm. Công ty phải trả bồi thường kinh tế hằng tháng trong suốt thời hạn đó. Không thỏa thuận con số bồi thường thì trả bằng 30% mức lương trung bình của 12 tháng trước khi nghỉ việc, và không thấp hơn mức lương tối thiểu địa phương. Vì lý do của chính công ty mà 3 tháng chưa trả, người lao động có thể yêu cầu chấm dứt ràng buộc không cạnh tranh này. Công ty chủ động chấm dứt thì phải trả thêm 3 tháng bồi thường. Từ năm 2025, người lao động ở vị trí không biết, không tiếp cận bí mật thương mại thì có thể yêu cầu xác nhận điều khoản không có hiệu lực. Phần vượt ra ngoài phạm vi bí mật mà người đó thực sự biết thì vô hiệu. Vụ án chính thức: một công ty dược phẩm kiện giám đốc công nghệ cao cấp Trịnh Mộc, đòi bồi thường vi phạm 710 vạn tệ. Tòa án xác định sản phẩm của công ty mới của ông không thay thế được sản phẩm của công ty cũ, phạm vi không cạnh tranh phải giới hạn ở phần bí mật ông thực sự biết, nên bác toàn bộ (toàn quốc) - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). 劳动合同法（2012 年修正，第二十三、二十四条）. （Cục Quản lý Khẩn cấp thành phố Phụ Thuận đăng lại）；Tòa án nhân dân tối cao (最高人民法院) (2020). 关于审理劳动争议案件适用法律问题的解释（一）（第三十六至四十条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 关于审理劳动争议案件适用法律问题的解释（二）（法释〔2025〕12 号，第十三至十五条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 劳动争议典型案例（案例四）. -- Ghi chú: Ngược lại, nếu thoả thuận có hiệu lực, công ty cũng đã trả tiền hằng tháng, mà bạn vẫn đi làm cho công ty cạnh tranh thì phải trả phạt vi phạm theo thoả thuận. Trả xong còn có thể bị yêu cầu giữ hết phần thời gian không cạnh tranh còn lại. Thỏa thuận ký trong thời gian còn làm việc, công ty không trả thêm tiền thì vẫn có hiệu lực. +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). 劳动合同法（2012 年修正，第二十三、二十四条）. （Cục Quản lý ứng cứu thành phố Phụ Thuận đăng lại）；Tòa án nhân dân tối cao (最高人民法院) (2020). 关于审理劳动争议案件适用法律问题的解释（一）（第三十六至四十条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 关于审理劳动争议案件适用法律问题的解释（二）（法释〔2025〕12 号，第十三至十五条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 劳动争议典型案例（案例四）. +- Ghi chú: Ngược lại, thỏa thuận có hiệu lực, công ty cũng đã trả tiền hằng tháng, mà bạn vẫn đi làm cho công ty cạnh tranh thì phải nộp phạt vi phạm theo điều khoản đã thoả thuận. Nộp xong còn có thể bị yêu cầu giữ hết phần thời gian không cạnh tranh còn lại. Thỏa thuận ký trong thời gian còn đi làm mà công ty không trả thêm tiền thì vẫn có hiệu lực. -### 13. Mã viết trong giờ làm việc, bằng nguồn lực của công ty thì thuộc công ty; dự án nguồn mở của bạn thì dùng thời gian, thiết bị của bạn, đừng trộn mã công ty +### 13. Mã nguồn viết trong giờ làm việc hoặc bằng nguồn lực công ty thì thuộc công ty, dự án mã nguồn mở của bạn thì tự làm bằng thời gian và thiết bị của bạn, không trộn mã công ty -- Chi phí: Không tốn tiền. Khi nhận việc, đọc kỹ điều khoản quyền sở hữu trí tuệ trong hợp đồng lao động. Chỗ khó là dự án cá nhân và công việc của công ty thường nằm chung một máy tính, cần tách ra. -- Nói thẳng: Phần mềm chủ yếu viết bằng thiết bị, mạng nội bộ hoặc dữ liệu kinh doanh của công ty, và do công ty chịu trách nhiệm thì quyền tác giả thuộc công ty, bạn chỉ giữ quyền ghi tên, hợp đồng có thỏa thuận khác thì cũng vậy. Viết sau giờ làm việc mà dùng máy tính công ty hoặc dữ liệu kinh doanh thì vẫn có thể là của công ty. Phát minh làm ra bằng nguồn lực công ty thì bằng sáng chế cũng thuộc công ty. -- Lợi ích: Tác phẩm tạo ra để hoàn thành nhiệm vụ công việc của đơn vị là tác phẩm công vụ (sản phẩm làm ra khi đi làm). Trường hợp thông thường quyền tác giả thuộc tác giả, đơn vị có quyền ưu tiên sử dụng trong phạm vi công việc, và tác giả không được cho phép bên thứ ba sử dụng theo cùng cách trong 2 năm. Nhưng có hai loại khác nhau. Một là các tác phẩm như thiết kế công trình, phần mềm, chủ yếu sử dụng điều kiện vật chất kỹ thuật của đơn vị và do đơn vị chịu trách nhiệm. Hai là tác phẩm mà hợp đồng thỏa thuận thuộc đơn vị. Hai loại này tác giả chỉ giữ quyền ghi tên. Điều kiện vật chất kỹ thuật của đơn vị là thiết bị, mạng nội bộ, dữ liệu kinh doanh của công ty. Sáng chế hoàn thành khi thực hiện nhiệm vụ của đơn vị, hoặc chủ yếu sử dụng điều kiện vật chất kỹ thuật của đơn vị, là phát minh công vụ, quyền xin cấp bằng sáng chế và bằng sáng chế nhận được đều thuộc đơn vị (toàn quốc) +- Chi phí: Không tốn tiền. Khi nhận việc, đọc kỹ điều khoản sở hữu trí tuệ trong hợp đồng lao động. Chỗ khó là dự án cá nhân và việc của công ty thường nằm chung một máy, phải tách ra +- Nói thẳng: Phần mềm chủ yếu viết ra dựa trên thiết bị, mạng nội bộ hoặc dữ liệu nghiệp vụ của công ty, do công ty chịu trách nhiệm, thì quyền tác giả thuộc công ty, bạn chỉ giữ quyền ghi tên tác giả, hợp đồng có thỏa thuận như vậy thì cũng vậy. Viết sau giờ làm mà có dùng máy tính công ty hoặc dữ liệu nghiệp vụ thì vẫn có thể tính là của công ty. Sáng chế làm ra bằng nguồn lực công ty thì bằng sáng chế cũng thuộc công ty. +- Lợi ích: Tác phẩm sáng tạo để hoàn thành nhiệm vụ công việc của đơn vị là tác phẩm công sức (sản phẩm làm ra khi đi làm). Trường hợp thông thường quyền tác giả thuộc tác giả, đơn vị có quyền ưu tiên sử dụng trong phạm vi công việc, và trong 2 năm tác giả không được cho phép bên thứ ba sử dụng cùng cách. Nhưng có hai loại là khác. Một là thiết kế kỹ thuật, phần mềm sáng tạo chủ yếu dựa vào điều kiện vật chất kỹ thuật của đơn vị và do đơn vị chịu trách nhiệm. Loại khác là tác phẩm mà hợp đồng thỏa thuận thuộc đơn vị. Hai loại này tác giả chỉ giữ quyền ghi tên tác giả. Điều kiện vật chất kỹ thuật của đơn vị là thiết bị, mạng nội bộ, dữ liệu nghiệp vụ của công ty. Sáng chế hoàn thành khi thực hiện nhiệm vụ của đơn vị, hoặc chủ yếu dựa vào điều kiện vật chất kỹ thuật của đơn vị, thì là sáng chế công sức, quyền xin cấp bằng sáng chế và bằng sáng chế nhận được đều thuộc đơn vị (toàn quốc) - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 著作权法（2020 年修正，第十八条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 专利法（2020 年修正，第六条）. -- Ghi chú: Phần mềm bị xếp thẳng vào loại「quyền tác giả thuộc đơn vị」, lập trình viên hầu như không rơi vào nhóm「quyền tác giả thuộc tác giả, đơn vị ưu tiên sử dụng trong 2 năm」. Việc「viết sau giờ làm việc」mà dùng máy tính, mạng nội bộ hoặc dữ liệu kinh doanh của công ty, vẫn có thể bị tính là chủ yếu làm ra bằng thiết bị và tư liệu của đơn vị. Muốn làm sạch dự án cá nhân thì dùng máy của mình, không đụng kho mã công ty, không trùng với công việc công ty. +- Ghi chú: Phần mềm bị xếp thẳng vào loại「quyền tác giả thuộc đơn vị」, lập trình viên hầu như không rơi vào loại「quyền tác giả thuộc tác giả, đơn vị ưu tiên sử dụng trong hai năm」. 「Viết sau giờ làm」mà có dùng máy tính công ty, mạng nội bộ hoặc dữ liệu nghiệp vụ công ty thì vẫn có thể bị tính là chủ yếu làm ra bằng thiết bị và tài liệu của đơn vị. Muốn làm sạch dự án cá nhân thì dùng máy của mình, không đụng kho mã công ty, không trùng với công việc công ty. -### 14. Làm sản phẩm bằng mã nguồn mở như GPL thì mở mã theo giấy phép yêu cầu hoặc đổi sang giấy phép khác, đừng coi là「không ai quản」 +### 14. Làm sản phẩm bằng mã nguồn mở như GPL, thì mở mã theo yêu cầu giấy phép hoặc đổi giấy phép, đừng coi là「không ai quản」 -- Chi phí: Không tốn tiền. Trước khi dùng mã nguồn mở của người khác, xem giấy phép của nó (điều kiện sử dụng tác giả ghi rõ). Dòng GPL yêu cầu sản phẩm có dùng nó cũng phải công khai mã nguồn, xung đột với việc bán mã đóng, xung đột thì đổi sang giấy phép khác. Chỗ khó là bỏ một thư viện đã quen dùng. -- Nói thẳng: Xâm phạm quyền tác giả thì bồi theo thiệt hại của bên kia hoặc số tiền bạn thu được, cố ý và nghiêm trọng thì có thể bị phạt bồi gấp 1 đến 5 lần. Có công ty thuê nhân viên cũ của đối thủ, làm ra phần mềm giống nhau về căn bản, và biện bằng phần mềm của đối phương dùng GPLv2, lẽ ra phải mở mã. Tòa án nhân dân tối cao cho rằng đối phương có mở mã hay không không ảnh hưởng tới quyền tác giả của nó, vẫn phán phải chấm dứt vi phạm và bồi 50 vạn tệ. -- Lợi ích: Xâm phạm quyền tác giả thì bồi theo thiệt hại thực tế của chủ quyền, hoặc theo số tiền người vi phạm thu được. Nếu cả hai đều không tính được thì căn cứ vào số tiền phải trả nếu mua hợp lệ giấy phép đó. Cố ý vi phạm và nghiêm trọng thì có thể bị phạt bồi gấp 1 đến 5 lần. Gây tổn hại lợi ích công cộng thì còn có thể tịch thu thu nhập bất hợp pháp và phạt tiền. Vụ án chính thức: (2021) 最高法知民终 51 号. Bị đơn thuê nhân viên cũ của nguyên đơn, phần mềm cổng phát triển mà bên bị đơn phát triển giống bản chất phần mềm của nguyên đơn. Bên bị đơn biện bảo phần mềm của nguyên đơn dựa trên GPLv2, vốn dĩ phải mở mã. Tòa án nhân dân tối cao xác định「开发者自身是否违反 GPLv2 协议和是否享有软件著作权是两个独立问题」, phán phải chấm dứt vi phạm, bồi thường 50 vạn tệ (toàn quốc) +- Chi phí: Không tốn tiền. Trước khi dùng mã nguồn mở của người khác, xem giấy phép của nó (điều kiện sử dụng mà tác giả viết ra). Họ GPL yêu cầu sản phẩm có dùng nó cũng phải mở mã nguồn ra, xung đột với việc đóng mã rồi bán tiền, xung đột thì đổi. Chỗ khó là từ bỏ một thư viện đã dùng quen +- Nói thẳng: Xâm phạm quyền tác giả thì bồi thường theo thiệt hại của bên kia hoặc theo số tiền bạn kiếm được, cố ý và đặc biệt nghiêm trọng thì có thể bị phạt tới mức 1 đến 5 lần. Có công ty thuê nhân viên cũ của đối thủ, làm ra phần mềm về cơ bản giống nhau, rồi biện rằng phần mềm của bên kia dùng GPLv2, lẽ ra phải mở mã. Tòa án nhân dân tối cao cho rằng việc bên kia có mở mã hay không không ảnh hưởng tới quyền tác giả của nó, vẫn kết án buộc ngừng xâm phạm, bồi thường 50 vạn tệ. +- Lợi ích: Xâm phạm quyền tác giả thì bồi thường theo thiệt hại thực tế của chủ sở hữu quyền, hoặc theo số tiền người xâm phạm kiếm được. Cả hai đều không tính được thì căn cứ số tiền phải trả nếu mua bình thường giấy phép này. Cố ý xâm phạm và đặc biệt nghiêm trọng thì có thể bị phạt bồi thường 1 đến 5 lần. Làm tổn hại lợi ích công cộng thì còn có thể tịch thu thu nhập bất hợp pháp và phạt tiền. Vụ án chính thức: (2021) 最高法知民终 51 号. Bên cáo tuyển nhân viên cũ của nguyên đơn, phần mềm cổng phát triển ra về cơ bản giống phần mềm của nguyên đơn. Bên cáo biện bạo rằng phần mềm của nguyên đơn dựa trên GPLv2, vốn đã phải mở mã. Tòa án nhân dân tối cao xác định「việc bản thân nhà phát triển có vi phạm thỏa thuận GPLv2 hay không và việc có sở hữu quyền tác giả phần mềm là hai vấn đề độc lập」, kết án buộc ngừng xâm phạm, bồi thường 50 vạn tệ (toàn quốc) - Mức bằng chứng: B -- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 著作权法（2020 年修正，第五十三、五十四条）. ；Tòa Xử quyền sở hữu trí tuệ của Tòa án nhân dân tối cao (最高人民法院知识产权法庭) (2022). 涉「开源软件」著作权侵权案（（2021）最高法知民终 51 号）. -- Ghi chú: Vụ án này cho thấy không thể lấy「đối phương dùng mã GPL」làm lá chắn cho việc bạn sao chép. Ngược lại, tác giả GPL đã kiện bên thương mại không mở mã và trong nước đã có án đã có hiệu lực, nhưng phần này không tìm được nguyên văn trên trang Tòa án nhân dân tối cao nên không dẫn. Để ở mức B là vì việc có tuân thủ giấy phép nguồn mở hay không chỉ có thể dựa vào hợp đồng và các quy định chung trong luật quyền tác giả, không có điều luật chuyên biệt nào. +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 著作权法（2020 年修正，第五十三、五十四条）. ；Tòa chuyên môn về sở hữu trí tuệ của Tòa án nhân dân tối cao (最高人民法院知识产权法庭) (2022). 涉「开源软件」著作权侵权案（（2021）最高法知民终 51 号）. +- Ghi chú: Vụ án này cho thấy「bên kia dùng mã GPL」không thể dùng làm lá chắn khi bạn sao chép. Ngược lại, tác giả GPL đã kiện bên thương mại đóng mã ở trong nước và đã có bản án có hiệu lực, phần này không tìm thấy văn bản trên trang của Tòa án nhân dân tối cao nên không dẫn. Để mức B là vì có tuân thủ giấy phép mã nguồn mở hay không chỉ dựa được vào hợp đồng và các quy định chung trong luật bản quyền, không có điều luật riêng nào. -### 15. Làm ứng dụng chỉ thu thập thông tin cá nhân thật cần cho nghiệp vụ, dưới 14 tuổi phải có đồng ý của người giám hộ, chính sách bảo mật viết rõ +### 15. Ứng dụng chỉ thu thập thông tin cá nhân cần thiết cho nghiệp vụ, dưới 14 tuổi phải có sự đồng ý của người giám hộ, chính sách bảo mật phải viết rõ -- Chi phí: Không tốn tiền. Trước khi đưa lên vận hành, đối chiếu danh sách quyền ứng dụng xin với chính sách bảo mật, quyền không cần thì đừng xin. Chỗ khó là bên sản phẩm luôn muốn giữ thêm mấy thứ「sau này có thể dùng đến」. -- Nói thẳng: Chỉ thu thập thông tin thật cần cho nghiệp vụ, người chưa đủ 14 tuổi thì phải có đồng ý của người giám hộ. Vi phạm thì trước hết bị yêu cầu chấp hành điều chỉnh, ứng dụng có thể bị ngừng. Không chấp hành điều chỉnh thì phạt tiền dưới 100 vạn tệ, người phụ trách phạt tiền 1 vạn đến 10 vạn tệ. Nghiêm trọng thì phạt tiền dưới 50 triệu tệ hoặc 5% doanh thu năm trước, người phụ trách phạt tiền 10 vạn đến 100 vạn tệ; người viết mã và làm sản phẩm cũng có thể tính là người phụ trách. -- Lợi ích: Xử lý thông tin cá nhân phải có mục đích rõ ràng, hợp lý, trực tiếp liên quan đến mục đích đó, và phải dùng cách ảnh hưởng ít nhất tới quyền lợi cá nhân. Xử lý thông tin cá nhân phải được cá nhân đồng ý, hoặc phải thuộc trường hợp theo quy định pháp luật. Người chưa đủ 14 tuần tuổi thì bắt buộc phải được cha mẹ hoặc người giám hộ đồng ý. Vi phạm thì cơ quan quản lý yêu cầu chấp hành điều chỉnh, cảnh cáo, tịch thu thu nhập bất hợp pháp, với ứng dụng vi phạm thì yêu cầu tạm dừng hoặc chấm dứt cung cấp dịch vụ. Không chấp hành điều chỉnh thì phạt tiền dưới 100 vạn tệ, người chịu trách nhiệm trực tiếp và người liên quan trực tiếp khác phạt tiền 1 vạn đến 10 vạn tệ. Nghiêm trọng thì phạt tiền dưới 50 triệu tệ, hoặc không quá 5% doanh thu năm trước. Người chịu trách nhiệm phạt tiền 10 vạn đến 100 vạn tệ. Những người này còn có thể bị cấm giữ chức vụ giám đốc, giám sát, lãnh đạo cấp cao của doanh nghiệp liên quan trong một thời hạn nhất định. Bán hoặc cung cấp thông tin cá nhân mà cấu thành tội phạm thì xem mục 4 của phần này (toàn quốc) +- Chi phí: Không tốn tiền. Trước khi đưa lên, đối chiếu danh sách quyền truy cập mà ứng dụng xin với chính sách bảo mật, quyền không cần thì đừng xin. Chỗ khó là bên sản phẩm luôn muốn giữ lại thêm vài dữ liệu「có thể sau này cần dùng」 +- Nói thẳng: Chỉ thu thập thông tin cần thiết cho nghiệp vụ, dưới 14 tuổi phải có sự đồng ý của người giám hộ. Vi phạm thì trước hết bị yêu cầu sửa, ứng dụng có thể bị dừng. Không sửa thì phạt tiền dưới 100 vạn tệ, người chịu trách nhiệm phạt tiền 1 vạn đến 10 vạn tệ. Nghiêm trọng thì phạt tiền dưới 5000 vạn tệ hoặc 5% doanh thu năm trước, người chịu trách nhiệm phạt tiền 10 vạn đến 100 vạn tệ, người viết mã và làm sản phẩm cũng có thể tính là người chịu trách nhiệm. +- Lợi ích: Xử lý thông tin cá nhân phải có mục đích rõ ràng hợp lý, trực tiếp liên quan đến mục đích đó, và phải dùng cách ảnh hưởng tới quyền lợi cá nhân nhỏ nhất. Xử lý thông tin cá nhân phải có sự đồng ý của cá nhân, hoặc thuộc trường hợp quy định theo luật. Trẻ chưa đủ 14 tuổi thì bắt buộc có sự đồng ý của cha mẹ hoặc người giám hộ. Vi phạm thì cơ quan quản lý yêu cầu sửa, đưa ra cảnh cáo, tịch thu thu nhập bất hợp pháp, yêu cầu ứng dụng vi phạm tạm dừng hoặc chấm dứt dịch vụ. Không sửa thì phạt tiền dưới 100 vạn tệ, người phụ trách trực tiếp và người chịu trách nhiệm trực tiếp khác phạt tiền 1 vạn đến 10 vạn tệ. Nghiêm trọng thì phạt tiền dưới 5000 vạn tệ, hoặc dưới 5% doanh thu năm trước. Người chịu trách nhiệm phạt tiền 10 vạn đến 100 vạn tệ. Những người này còn có thể bị cấm giữ chức giám đốt, giám sát, lãnh đạo của doanh nghiệp liên quan trong một khoảng thời gian nhất định. Bán hoặc cung cấp thông tin cá nhân mà cấu thành tội phạm thì xem mục 4 (toàn quốc) - Mức bằng chứng: B - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). 个人信息保护法（第六、十三、三十一、六十六条）. -- Ghi chú:「Người chịu trách nhiệm trực tiếp」trong thực tế hay rơi vào người viết mã và làm sản phẩm. Bộ Công nghiệp và Thông tin định kỳ công bố danh sách gỡ ứng dụng vi phạm, phần này chưa đối chiếu được đợt công bố nào nên không dẫn. -### 16. Trước khi website, ứng dụng đưa vào vận hành phải đăng ký ICP; giữ nhật ký trên 6 tháng trở lên theo yêu cầu phân cấp bảo vệ +- Ghi chú: 「直接责任人员」trong thực tế thường rơi vào người viết mã và làm sản phẩm. Bộ Công nghiệp và Thông tin định kỳ công bố danh sách gỡ ứng dụng vi phạm, phần này không đối chiếu được đợt công bố cụ thể nào nên không dẫn. +### 16. Website, ứng dụng trước khi đưa lên phải đăng ký ICP, giữ nhật ký tối thiểu 6 tháng theo yêu cầu bảo vệ theo cấp độ -- Chi phí: Đăng ký miễn phí, đi theo quy trình của nhà cung cấp điện toán đám mây thì mất mấy tuần. Đánh giá phân cấp bảo vệ thì tính phí theo cấp độ. Phân cấp bảo vệ gọi tắt là phân cấp bảo vệ, là bộ biện pháp quản lý an toàn mà nhà nước yêu cầu website và hệ thống thực hiện theo mức độ quan trọng; website nhỏ thì thường không cần đánh giá. -- Nói thẳng: Đăng ký không mất tiền, đi theo quy trình của nhà cung cấp điện toán đám mây thì mất mấy tuần. Máy chủ trong nước mà không đăng ký thì tên miền không phân giải được, không chấp hành điều chỉnh thì bị đóng website. Trang chủ không ghi số đăng ký thì phạt tiền 5000 tệ đến 5 vạn tệ. Đăng ký nói không thu phí mà thực tế có thu tiền thì phạt tiền 10 vạn đến 100 vạn tệ. Nhật ký phải giữ đủ 6 tháng, không chấp hành điều chỉnh hoặc để xảy ra hậu quả thì phạt tiền 1 vạn đến 10 vạn tệ. -- Lợi ích: Dịch vụ thông tin trên internet mang tính kinh doanh thì phải được cấp phép, không mang tính kinh doanh thì phải đăng ký. Không có giấy phép, hoặc không đăng ký thì không được làm. Không đăng ký thì cơ quan quản lý viễn thông yêu cầu chấp hành điều chỉnh, không chấp hành thì yêu cầu đóng website. Kinh doanh khi chưa được phép thì tịch thu thu nhập bất hợp pháp, kèm phạt tiền gấp 3 đến 5 lần thu nhập bất hợp pháp, hoặc phạt tiền 10 vạn đến 100 vạn tệ. Trang chủ website không ghi số đăng ký thì phạt tiền 5000 tệ đến 5 vạn tệ. Đơn vị vận hành mạng phải theo hệ thống phân cấp bảo vệ xây dựng quản lý an toàn, phòng chống tấn công, lưu nhật ký mạng không ít hơn 6 tháng, phân loại, sao lưu, mã hoá dữ liệu. Không thực hiện thì bị yêu cầu chấp hành điều chỉnh, cảnh cáo. Không chấp hành điều chỉnh hoặc để xảy ra hậu quả thì phạt tiền 1 vạn đến 10 vạn tệ, người chịu trách nhiệm trực tiếp phạt tiền 5000 tệ đến 5 vạn tệ (toàn quốc) +- Chi phí: Đăng ký miễn phí, đi theo quy trình của nhà cung cấp điện toán đám mây mất khoảng mấy tuần. Đánh giá bảo vệ theo cấp độ thì thu phí theo cấp. Bảo vệ theo cấp độ gọi tắt là đánh giá cấp độ, là một bộ biện pháp quản lý an toàn mà Nhà nước phân loại theo mức quan trọng của hệ thống và yêu cầu website, hệ thống phải thực hiện; website nhỏ thường không cần đánh giá +- Nói thẳng: Đăng ký không mất tiền, đi quy trình của nhà cung cấp đám mây mất khoảng mấy tuần. Máy chủ trong nước mà không đăng ký thì tên miền không phân giải được, không sửa thì bị đóng website. Không ghi số đăng ký trên trang chủ thì phạt tiền 5000 tệ đến 5 vạn tệ. Đăng ký nói không thu phí mà thực tế lại thu tiền thì phạt tiền 10 vạn đến 100 vạn tệ. Nhật ký phải giữ đủ 6 tháng, không sửa hoặc đã để lại hậu quả thì phạt tiền 1 vạn đến 10 vạn tệ. +- Lợi ích: Dịch vụ thông tin Internet có tính thương mại thì áp dụng giấy phép, không thương mại thì áp dụng đăng ký. Chưa có giấy phép hoặc chưa đăng ký thì không được làm. Chưa đăng ký thì cơ quan quản lý viễn thông yêu cầu sửa, không sửa thì yêu cầu đóng website. Kinh doanh mà chưa có giấy phép thì tịch thu thu nhập bất hợp pháp, phạt tiền bằng 3 đến 5 lần thu nhập bất hợp pháp, hoặc phạt tiền 10 vạn đến 100 vạn tệ. Trang chủ website không ghi số đăng ký thì phạt tiền 5000 tệ đến 5 vạn tệ. Đơn vị vận hành mạng phải theo hệ thống bảo vệ theo cấp độ (đánh giá cấp độ) để xây dựng quy chế quản lý an toàn, phòng chống tấn công, giữ nhật ký mạng không dưới 6 tháng, sao lưu và mã hoá dữ liệu theo phân loại. Không thực hiện thì yêu cầu sửa, đưa ra cảnh cáo. Không sửa hoặc đã gây hậu quả thì phạt tiền 1 vạn đến 10 vạn tệ, người phụ trách phạt tiền 5000 tệ đến 5 vạn tệ (toàn quốc) - Mức bằng chứng: A -- Nguồn: Hội đồng Bộ trưởng (国务院) (2011). 互联网信息服务管理办法（2011 年修订，第四、十九、二十二条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十三、六十一条；2016 年文本为第二十一、五十九条）. -- Ghi chú: Máy chủ trong nước mà không đăng ký thì tên miền không phân giải được. Có hai cái bẫy thường gặp: đăng ký là cá nhân không kinh doanh nhưng thực tế có thu tiền; đổi tên miền hoặc đổi chủ thể mà không đi làm thay đổi đăng ký. Giữ nhật ký 6 tháng vừa là nghĩa vụ vừa là cách tự bảo vệ, khi có chuyện thì chứng minh được không phải mình làm. +- Nguồn: Quốc vụ viện (国务院) (2011). 互联网信息服务管理办法（2011 年修订，第四、十九、二十二条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十三、六十一条；2016 年文本为第二十一、五十九条）. +- Ghi chú: Máy chủ trong nước mà không đăng ký thì tên miền không phân giải được ngay. Có hai cái hố thường gặp: đăng ký là cá nhân phi thương mại nhưng thực tế lại thu tiền; đổi tên miền hoặc đổi chủ thể mà không đi làm thủ tục thay đổi đăng ký. Giữ nhật ký 6 tháng vừa là nghĩa vụ vừa là cách tự bảo vệ, khi có chuyện thì chứng minh được không phải mình làm. -### 17. Trước khi cung cấp dịch vụ trí tuệ nhân tạo tạo sinh ra công chúng, phải làm đánh giá an toàn và đăng ký thuật toán theo quy định; nội dung tạo sinh phải gắn nhãn +### 17. Trước khi cung cấp dịch vụ AI tạo sinh cho công chúng, phải đánh giá an toàn và đăng ký thuật toán theo quy định, nội dung tạo ra phải gắn nhãn -- Chi phí: Không tốn tiền, nhưng phải đi qua quy trình của cơ quan quản lý không gian thông tin, xong cũng tốn thời gian. Công cụ nhỏ do cá nhân làm, chỉ cần không mở cho công chúng thì không kích hoạt bộ này. -- Nói thẳng: Dịch vụ trí tuệ nhân tạo tạo sinh hướng tới công chúng, có thể ảnh hưởng dư luận hoặc vận động người ta, thì trước khi đưa vào vận hành phải làm đánh giá an toàn, đăng ký thuật toán; có thay đổi hoặc dừng thì phải làm thủ tục thay đổi, huỷ đăng ký. Hình ảnh và video tạo sinh phải gắn nhãn. Công cụ nhỏ bạn tự dùng, không mở cho công chúng thì không cần lo mấy điều này. Vi phạm thì xử phạt theo Luật An ninh mạng và các luật khác. -- Lợi ích: Cung cấp dịch vụ trí tuệ nhân tạo tạo sinh có tính chất dư luận hoặc có khả năng vận động xã hội thì phải tiến hành đánh giá an toàn theo quy định của nhà nước, đăng ký thuật toán theo quy định quản lý thuật toán đề xuất, và làm các thủ tục thay đổi, huỷ đăng ký về sau. Tính chất dư luận hoặc khả năng vận động xã hội là khả năng ảnh hưởng tới nhận thức của công chúng, hoặc có thể vận động, tổ chức người ta. Bên cung cấp phải gắn nhãn cho nội dung tạo sinh như hình ảnh, video. Quy chế này áp dụng từ ngày 15 tháng 8 năm 2023 (toàn quốc) +- Chi phí: Không tốn tiền, nhưng phải đi quy trình của cơ quan quản lý mạng, xin xong mất một khoảng thời gian. Công cụ nhỏ do cá nhân làm, chỉ cần không mở cho công chúng thì không kích hoạt bộ quy định này +- Nói thẳng: Dịch vụ AI tạo sinh hướng tới công chúng, có khả năng ảnh hưởng dư luận hoặc vận động người ta, thì trước khi đưa lên phải đánh giá an toàn, đăng ký thuật toán; có sửa đổi hoặc dừng thì còn phải làm thủ tục thay đổi, huỷ bỏ. Ảnh và video do mô hình tạo ra phải gắn nhãn. Công cụ nhỏ tự dùng, không mở cho công chúng thì không phải lo. Vi phạm thì bị phạt theo Luật An ninh mạng và các luật khác. +- Lợi ích: Cung cấp dịch vụ trí tuệ nhân tạo tạo sinh có tính chất dư luận hoặc có khả năng vận động xã hội thì phải tiến hành đánh giá an toàn theo quy định của Nhà nước, đăng ký thuật toán theo quy định quản lý thuật toán đề xuất, cùng các thủ tục thay đổi, huỷ bỏ về sau. Tính chất dư luận hoặc khả năng vận động xã hội là khả năng ảnh hưởng tới nhận thức của công chúng, hoặc khả năng tổ chức vận động người ta. Bên cung cấp phải gắn nhãn cho nội dung do mô hình tạo ra như ảnh, video. Biện pháp này có hiệu lực từ ngày 15 tháng 8 năm 2023 (toàn quốc) - Mức bằng chứng: B -- Nguồn: Văn phòng Quốc gia Quản lý Thông tin Internet (国家互联网信息办公室) và sáu bộ ngành khác (2023). 生成式人工智能服务管理暂行办法（第十二、十七条）. -- Ghi chú: Bản quy chế này không tự viết phạt bao nhiêu tiền. Vi phạm thì phải quay lại Luật An ninh mạng, Luật An toàn dữ liệu, Luật Bảo vệ thông tin cá nhân để tìm mức phạt, nên để ở mức B. Vụ án dân sự về việc nội dung tạo bằng AI xâm phạm quyền tác giả, quyền với hình ảnh của người khác, phần này chưa đối chiếu nguyên văn trên trang Tòa án nhân dân tối cao nên không dẫn. \ No newline at end of file +- Nguồn: Văn phòng Quản lý mạng quốc gia (国家互联网信息办公室) và sáu cơ quan khác (2023). 生成式人工智能服务管理暂行办法（第十二、十七条）. +- Ghi chú: Biện pháp này không tự ghi mức phạt tiền là bao nhiêu. Vi phạm thì phải quay lại Luật An ninh mạng, Luật An toàn dữ liệu, Luật Bảo vệ Thông tin cá nhân để tìm mức phạt, nên để mức B. Vụ án dân sự về nội dung do AI tạo ra xâm phạm quyền tác giả, quyền hình ảnh của người khác, phần này không đối chiếu được văn bản trên trang của Tòa án nhân dân tối cao nên không dẫn. \ No newline at end of file diff --git "a/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" "b/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" index 8e15b8de..2477b6a2 100644 --- "a/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" +++ "b/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" @@ -1,6 +1,6 @@ [← Về mục lục](../README.md) -# 14. Tài khoản và bảo mật thông tin +# 14. Tài khoản và an toàn thông tin Thước đo: tiền và thông tin cá nhân. Khi tài khoản của bạn bị người khác đăng nhập vào, thứ mất trước là tiền. Bên kia còn dùng tài khoản của bạn để lừa những người trong danh bạ. Danh tính của bạn cũng coi như đã bị lấy mất. diff --git "a/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" "b/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" index 6921442d..ff79fa77 100644 --- "a/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" +++ "b/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" @@ -1,8 +1,8 @@ -[← 回总目录](../README.md) +[← Về mục lục](../README.md) -# 15. 租房与买房 +# 15. Thuê nhà và mua nhà -口径：金钱。租房最容易亏钱的地方是押金和中介。买房最费钱的地方是贷款利息，还有卖方知道、你不知道的那些事。 +Cột đo của phần này là tiền. Chỗ dễ lỗ nhất khi thuê nhà là tiền cọc và môi giới. Chỗ tốn tiền nhất khi mua nhà là lãi vay, cộng thêm những việc bên bán biết mà bên mua không biết. ### 1. Số tiền cọc, thời điểm hoàn trả và các trường hợp bị giữ lại phải ghi vào hợp đồng diff --git "a/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" index ba189ec8..9ad619bb 100644 --- "a/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" +++ "b/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -1,8 +1,8 @@ -[← 回总目录](../README.md) +[← Về mục lục](../README.md) -# 16. 得了慢性病之后怎么活 +# 16. Sống sau khi mắc bệnh mạn tính -口径：总死亡率与金钱。查出慢性病以后往往还有几十年要过，本节算的是这几十年怎么管。 +Cột đo của phần này là tỷ lệ tử vong chung và tiền. Sau khi phát hiện bệnh mạn tính thường còn phải sống hàng chục năm, phần này tính mấy chục năm ấy nên quản lý thế nào. ### 1. Uống thuốc đủ theo đơn của bác sĩ, đừng thấy khá hơn là ngừng diff --git "a/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" "b/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" index 62728e60..48de50e0 100644 --- "a/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" +++ "b/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" @@ -1,6 +1,6 @@ [← Về mục lục](../README.md) -# 17. Nhà có người già +# 17. Khi nhà có người lớn tuổi Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn rắc rối không phải vì thiếu tiền, mà là chưa nói cho rõ và chưa ký từ trước. Mục cuối nói về loét áp lực cho người nằm lâu, tính theo tỷ lệ tử vong, không so với các mục khác trong phần này. diff --git "a/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" index 8a103fcc..0d2eb596 100644 --- "a/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" +++ "b/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -2,39 +2,39 @@ # 18. Nuôi con có đáng không -Mục này tính hai cuốn sổ: tiền và thời gian. Giống phần 10 tính chi phí cưới, ở đây chỉ bóc tách từng khoản ra cho bạn xem, không thay bạn kết luận. Đã mang thai và cần đi theo từng bước thủ tục thì xem phần 27. +Phần này tính hai thứ: tiền và thời gian. Giống phần 10 tính chuyện kết hôn, ở đây chỉ tách bảng tính ra để bạn tự xem, không thay bạn kết luận. Đã mang thai và đang làm theo từng bước thủ tục, xem phần 27. -### 1. Hãy cộng vào trước các khoản được nhận: trợ cấp nuôi con của Nhà nước mỗi đứa trẻ mỗi năm 3.600 tệ, trả đến khi đủ 3 tuổi +### 1. Cộng trước vào các khoản được nhận: trợ cấp nuôi con của Nhà nước mỗi đứa trẻ mỗi năm 3600 tệ, trả đến khi đủ 3 tuổi -- Chi phí: Không tốn tiền. Tới nơi có hộ khẩu của đứa trẻ để xin, chạy một lượt là xong -- Nói thẳng: Đứa trẻ sinh ra từ sau ngày 1 tháng 1 năm 2025 và chưa đủ 3 tuổi thì nhận được khoản tiền này. Một người trong hai người làm cha mẹ tới nơi có hộ khẩu của đứa trẻ để xin. Mỗi đứa trẻ mỗi năm 3.600 tệ, nhận liên tục đến khi đủ 3 tuổi. Khoản tiền này không phải chịu thuế thu nhập cá nhân. Khi xin các khoản trợ cấp như trợ cấp người có thu nhập thấp thì nó cũng không tính vào thu nhập của gia đình bạn -- Lợi ích: Phương án do Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Hội đồng Nhà nước ban hành quy định: trẻ sơ sinh và trẻ nhỏ sinh từ ngày 1 tháng 1 năm 2025 và dưới 3 tuổi thì chi trả theo 「mỗi đứa trẻ mỗi năm 3.600 tệ」. Một người trong hai người làm cha mẹ, hoặc người giám hộ khác của đứa trẻ đi xin. Khoản trợ cấp này 「không phải chịu thuế thu nhập cá nhân」. Khi xét các khoản trợ cấp như trợ cấp người có thu nhập thấp thì nó 「không tính vào thu nhập của gia đình hoặc cá nhân」 +- Chi phí: Không tốn tiền. Đi tới nơi có hộ khẩu của đứa trẻ để xin, chạy một lượt là xong +- Nói thẳng: Đứa trẻ sinh ra từ ngày 1 tháng 1 năm 2025 trở đi và chưa đủ 3 tuổi thì nhận được khoản tiền này. Một người trong hai người làm cha mẹ đi tới nơi có hộ khẩu của đứa trẻ để xin. Mỗi đứa trẻ mỗi năm 3600 tệ, nhận liên tục đến khi đủ 3 tuổi. Khoản tiền này không phải chịu thuế thu nhập cá nhân. Khi xin các khoản trợ cấp như trợ cấp người có thu nhập thấp thì nó cũng không tính vào thu nhập của gia đình bạn. +- Lợi ích: Phương án do Văn phòng Trung ương Đảng Cộng sản Trung Quốc (中共中央办公厅), Văn phòng Hội đồng Nhà nước (国务院办公厅) ban hành quy định: trẻ sơ sinh, trẻ nhỏ sinh từ ngày 1 tháng 1 năm 2025 và dưới 3 tuổi thì chi trả theo「mỗi đứa trẻ mỗi năm 3600 tệ」. Một người trong hai người làm cha mẹ, hoặc người giám hộ khác của đứa trẻ, đi xin. Khoản trợ cấp này「không phải chịu thuế thu nhập cá nhân」. Khi xét các khoản trợ cấp như trợ cấp người có thu nhập thấp thì nó「không tính vào thu nhập của gia đình hoặc cá nhân」 - Mức bằng chứng: A -- Nguồn: Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước (2025). 育儿补贴制度实施方案. -- Ghi chú: Đây là mức sàn thống nhất toàn quốc. Có tỉnh, thành phố có thêm trợ cấp địa phương riêng, có thể nhận cộng với khoản này. Ba năm cộng lại 10.800 tệ, trong chi phí nuôi con chỉ là một phần rất nhỏ, đừng lấy nó làm căn cứ để quyết định có sinh hay không +- Nguồn: Văn phòng Trung ương Đảng Cộng sản Trung Quốc (中共中央办公厅), Văn phòng Hội đồng Nhà nước (国务院办公厅) (2025). 育儿补贴制度实施方案. +- Ghi chú: Đây là mức sàn thống nhất toàn quốc. Có tỉnh, thành có thêm trợ cấp địa phương riêng, có thể nhận cộng với khoản này. Ba năm cộng lại 10800 tệ, trong chi phí nuôi con chỉ chiếm một phần rất nhỏ, đừng lấy nó làm căn cứ để quyết định có sinh hay không ### 2. Nghỉ thai sản 98 ngày, trợ cấp thai sản do quỹ bảo hiểm thai sản chi trả theo lương bình quân hằng tháng của công nhân viên năm trước của đơn vị -- Chi phí: Không tốn tiền. Đơn vị sử dụng lao động lo khai báo, bạn không phải chạy đi đâu -- Nói thẳng: Khi sinh con, ít nhất được nghỉ thai sản 98 ngày, trong đó trước ngày sinh có thể nghỉ trước 15 ngày. Sinh khó thì cộng thêm 15 ngày. Sinh thêm đứa con nữa thì cộng thêm 15 ngày. Khoản tiền chi trả trong thời gian đó gọi là trợ cấp thai sản, do quỹ bảo hiểm thai sản chi trả. Đơn vị không đóng bảo hiểm thai sản cho bạn thì do đơn vị tự chi trả -- Lợi ích: Số ngày nghỉ thai sản do Quốc vụ viện quy định. 「Nữ công nhân viên sinh con được nghỉ thai sản 98 ngày, trong đó trước ngày sinh có thể nghỉ 15 ngày; sinh khó thì cộng thêm 15 ngày nghỉ thai sản; sinh đa thai, mỗi lần sinh thêm 1 đứa trẻ thì cộng thêm 15 ngày nghỉ thai sản.」 Khoản tiền nhận trong thời gian nghỉ thai sản gọi là trợ cấp thai sản. Người đã tham gia bảo hiểm thai sản thì 「chi trả theo mức lương bình quân hằng tháng của công nhân viên năm trước của đơn vị sử dụng lao động, do quỹ bảo hiểm thai sản chi trả」. Người không tham gia bảo hiểm thì đơn vị sử dụng lao động chi trả cho bạn theo mức lương trước khi nghỉ thai sản +- Chi phí: Không tốn tiền. Đơn vị sử dụng lao động đi khai báo, bạn không phải chạy đi đâu +- Nói thẳng: Khi sinh con, ít nhất được nghỉ thai sản 98 ngày, trong đó trước ngày sinh có thể nghỉ trước 15 ngày. Sinh khó thì cộng thêm 15 ngày. Sinh thêm một đứa thì cộng thêm 15 ngày. Khoản tiền chi trả trong thời gian đó gọi là trợ cấp thai sản, do quỹ bảo hiểm thai sản chi trả. Đơn vị không đóng bảo hiểm thai sản cho bạn thì do đơn vị tự chi trả. +- Lợi ích: Số ngày nghỉ thai sản do Quốc vụ viện (国务院) quy định. 「Nữ công nhân viên sinh con được nghỉ thai sản 98 ngày, trong đó trước ngày sinh có thể nghỉ 15 ngày; sinh khó thì cộng thêm 15 ngày nghỉ thai sản; sinh đa thai, mỗi lần sinh thêm 1 đứa trẻ thì cộng thêm 15 ngày nghỉ thai sản.」 Khoản tiền nhận trong thời gian nghỉ thai sản gọi là trợ cấp thai sản. Người đã tham gia bảo hiểm thai sản thì「chi trả theo mức lương bình quân hằng tháng của công nhân viên năm trước của đơn vị sử dụng lao động, do quỹ bảo hiểm thai sản chi trả」. Người không tham gia bảo hiểm thì đơn vị sử dụng lao động chi trả cho bạn theo mức lương trước khi nghỉ thai sản - Mức bằng chứng: A -- Nguồn: Quốc vụ viện (国务院) (2012). Quy định đặc biệt về bảo vệ lao động nữ công nhân viên (Quốc vụ lệnh số 619, Điều bảy, Điều tám). -- Ghi chú: 98 ngày là mức tối thiểu do Nhà nước đặt ra. Điều lệ dân số và kế hoạch hoá của các tỉnh thường còn cộng thêm hơn 60 ngày nghỉ thưởng cho việc sinh con, tính cả thường là từ 158 ngày. Lương trong thời gian nghỉ thưởng do bên nào chi trả thì quy định khác nhau ở mỗi tỉnh +- Nguồn: Quốc vụ viện (国务院) (2012). 女职工劳动保护特别规定（国令第 619 号，第七条、第八条）. +- Ghi chú: 98 ngày là mức tối thiểu do Nhà nước đặt ra. Điều lệ dân số và kế hoạch hoá của các tỉnh thường còn cộng thêm hơn 60 ngày nghỉ thưởng cho việc sinh con, tính cả thường là từ 158 ngày. Lương trong thời gian nghỉ thưởng do bên nào chi trả thì quy định ở mỗi tỉnh khác nhau ### 3. Biết điều này: không được giảm lương hoặc sa thải vì mang thai, vì sinh con, vì cho con bú - Chi phí: Không tốn tiền -- Nói thẳng: Trong thời gian mang thai, sinh con và cho con bú, nếu đơn vị giảm lương của bạn hoặc sa thải bạn thì đều là vi phạm pháp, bị phạt mỗi người từ 1.000 đến 5.000 tệ. Trẻ chưa đầy một tuổi thì mỗi ngày bạn có 1 giờ cho con bú, đơn vị cũng không được bố trí bạn làm thêm giờ hay ca đêm. Cách đòi bồi thường xem phần 19 -- Lợi ích: Quốc vụ viện quy định: 「Đơn vị sử dụng lao động không được vì nữ công nhân viên mang thai, sinh con, cho con bú mà giảm lương, sa thải, hoặc hủy hợp đồng lao động hay hợp đồng thuê bao với nữ công nhân viên.」 Trong ba thời gian đó đơn vị không được giảm tiền của bạn, cũng không được sa thải bạn hay hủy hợp đồng. Nữ công nhân viên đang cho con dưới 1 tuổi bú thì mỗi ngày có 1 giờ cho con bú. Đơn vị cũng không được kéo dài giờ làm của bạn, không được bố trí ca đêm. Đơn vị vi phạm những điều trên thì bị phạt mỗi người từ 1.000 đến 5.000 tệ +- Nói thẳng: Trong thời gian mang thai, sinh con, cho con bú, nếu đơn vị giảm lương của bạn hoặc sa thải bạn thì đều là vi phạm pháp, bị phạt tiền mỗi người từ 1000 đến 5000 tệ. Trẻ chưa đầy một tuổi thì mỗi ngày bạn có 1 giờ cho con bú, đơn vị cũng không được bố trí bạn làm thêm giờ hay ca đêm. Cách đòi bồi thường xem phần 19. +- Lợi ích: Quốc vụ viện (国务院) quy định: 「Đơn vị sử dụng lao động không được vì nữ công nhân viên mang thai, sinh con, cho con bú mà giảm lương, sa thải, hoặc hủy hợp đồng lao động hay hợp đồng thuê bao với nữ công nhân viên.」 Trong ba thời gian đó đơn vị không được giảm tiền của bạn, cũng không được sa thải bạn hay hủy hợp đồng. Nữ công nhân viên cho con dưới 1 tuổi bú thì mỗi ngày có 1 giờ cho con bú. Đơn vị cũng không được kéo dài giờ làm của bạn, không được bố trí ca đêm. Đơn vị vi phạm những điều trên thì bị phạt tiền mỗi người từ 1000 đến 5000 tệ - Mức bằng chứng: A -- Nguồn: Quốc vụ viện (国务院) (2012). Quy định đặc biệt về bảo vệ lao động nữ công nhân viên (Quốc vụ lệnh số 619, Điều năm, Điều chín, Điều mười ba). +- Nguồn: Quốc vụ viện (国务院) (2012). 女职工劳动保护特别规定（国令第 619 号，第五条、第九条、第十三条）. - Ghi chú: Nếu thật sự bị sa thải vi phạm pháp thì đường đòi bồi thường xem phần 19. Trọng tâm khi lấy chứng cứ là thông báo bằng văn bản về việc chuyển vị trí, cắt giảm lương và lịch sử trò chuyện, phải giữ kỹ hai thứ đó -### 4. Cuốn sổ thời gian tính theo 「thời gian của ai bị chiếm mất」, không tính theo 「có vất vả không」 +### 4. Cuốn sổ thời gian tính theo「thời gian của ai bị chiếm mất」, không tính theo「có vất vả không」 - Chi phí: Không tốn tiền, cái mất là thời gian. Trước khi trẻ đủ 3 tuổi thì về cơ bản thời gian bị chiếm hết cả ngày. Chỗ khó là chuyện này lặp lại ngày này sang ngày khác, không thể cắn răng chịu mấy ngày rồi qua -- Nói thẳng: Quy đổi số giờ trông trẻ mỗi ngày và số lần bị đánh thức giữa đêm thành thời gian. Rồi nhân với mức lương theo giờ của người làm việc đó. Con số tính ra thường lớn hơn nhiều so với tiền sữa công thức và tã. Vì thế nên chia việc từ trước khi sinh +- Nói thẳng: Quy đổi số giờ trông trẻ mỗi ngày và số lần bị đánh thức giữa đêm thành thời gian. Rồi nhân với mức lương theo giờ của người làm việc đó. Con số tính ra thường lớn hơn nhiều so với tiền sữa công thức và tã. Vì thế nên chia việc từ trước khi sinh. - Lợi ích: Quy đổi thời gian trông trẻ mỗi ngày và số lần bị đánh thức giữa đêm thành giờ. Rồi nhân với mức lương theo giờ của người làm việc đó. Con số thu được thường lớn hơn nhiều so với tiền sữa công thức và tã. Việc trông trẻ, nấu ăn, dậy đêm đi vệ sinh dù không ai trả tiền vẫn được ghi vào chi phí kiểu đó. Cách tính giống bộ tính việc nhà ở phần 10. Chia việc cho xong rồi hãy quyết định có sinh con hay không - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; thước đo xem phần 10 @@ -42,18 +42,18 @@ Mục này tính hai cuốn sổ: tiền và thời gian. Giống phần 10 tín ### 5. Cuốn sổ tiền chia làm ba đoạn để tính: 0 đến 3 tuổi, giáo dục bắt buộc, sau giáo dục bắt buộc -- Chi phí: Bao nhiêu tiền thì không có con số cố định, khác nhau theo thành phố và theo cách nuôi, chênh nhau rất xa. Chỗ khó là phải bình tĩnh tách ba đoạn sổ sách ra tính riêng, không thể đoán bừa một con số bằng cảm tính -- Nói thẳng: Tiền ở giai đoạn 0 đến 3 tuổi chủ yếu chi vào cách nuôi: tự nuôi, nhờ ông bà, hay thuê người nuôi. Từ tiểu học đến trung học thì chủ yếu chi vào nhà ở và ngoại khóa. Sau đó thì chủ yếu chi vào con đường thi cử. Cách chi của ba đoạn hoàn toàn khác nhau, gộp vào nhau chỉ cho ra một tổng số đáng sợ +- Chi phí: Tốn bao nhiêu tiền thì không có con số cố định, khác nhau theo thành phố và theo cách nuôi, chênh nhau rất xa. Chỗ khó là phải bình tĩnh tách ba đoạn sổ ra tính riêng, không thể đoán bừa một con số bằng cảm tính +- Nói thẳng: Tiền ở giai đoạn 0 đến 3 tuổi chủ yếu chi vào cách nuôi: tự nuôi, nhờ ông bà, hay thuê người nuôi. Từ tiểu học đến trung học thì chủ yếu chi vào nhà ở và ngoại khóa. Sau đó thì chủ yếu chi vào con đường học vấn. Cách chi của ba đoạn hoàn toàn khác nhau, gộp vào nhau chỉ cho ra một tổng số đáng sợ. - Lợi ích: Khoản lớn nhất trong giai đoạn 0 đến 3 tuổi là cách chăm sóc: tự nuôi, nhờ ông bà hay thuê người nuôi. Khoản lớn nhất trong giai đoạn giáo dục bắt buộc (tiểu học đến trung học) là nhà ở và chi phí ngoại khóa. Sau đó thì khoản lớn nhất là con đường học vấn. Ba đoạn đó chi tiền vào hai chỗ hoàn toàn khác nhau, gộp lại tính thì chỉ cho ra một tổng số đáng sợ - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp -- Ghi chú: Trước hết cộng vào các khoản chắc chắn nhận được như trợ cấp nuôi con, trợ cấp thai sản, bảo hiểm y tế hoàn trả, rồi xem còn thiếu bao nhiêu. Đừng dựa vào những con số 「nuôi một đứa con tốn bao nhiêu triệu」 lan truyền trên mạng, loại số đó không rõ thước đo, bạn không biết nó tính những gì và bỏ sót những gì +- Ghi chú: Trước hết cộng vào các khoản chắc chắn nhận được như trợ cấp nuôi con, trợ cấp thai sản, bảo hiểm y tế hoàn trả, rồi xem còn thiếu bao nhiêu. Đừng dựa vào những con số「nuôi một đứa con tốn bao nhiêu triệu」lan truyền trên mạng, loại số đó không rõ thước đo, bạn không biết nó tính những gì và bỏ sót những gì ### 6. Sinh vì người lớn, sinh vì hôn nhân, sinh vì lúc về già, mỗi kiểu ghi một cuốn sổ riêng - Chi phí: Không tốn tiền, cái mất là một cuộc nói chuyện nghiêm túc. Chỗ khó là phải nói thẳng ra, không vòng vo -- Nói thẳng: Người khác muốn bạn sinh con thì có thể tính là một yếu tố cân nhắc của bạn. Nhưng thời gian và tiền thì ghi vào sổ của chính bạn. 「Nuôi con khi về già」 thì phải tính riêng một cuốn, vì nó đánh cược vào năng lực tài chính và ý chí của một người khác hai mươi năm sau. Hãy viết cả hậu quả của việc không sinh ra, rồi đối chiếu hai bên -- Lợi ích: Cách nghĩ giống phần 10. Kỳ vọng của người khác có thể tính là một yếu tố cân nhắc của bạn. Nhưng chi phí thì ghi vào sổ của chính bạn, không ghi vào sổ của họ. 「Nuôi con khi về già」 càng phải tính riêng một cuốn, vì nó giả định năng lực tài chính và ý chí của một người khác hai mươi năm sau +- Nói thẳng: Người khác muốn bạn sinh con thì có thể tính là một yếu tố cân nhắc của bạn. Nhưng thời gian và tiền thì ghi vào sổ của chính bạn. 「Nuôi con khi về già」phải tính riêng một cuốn, vì nó đánh cược vào năng lực tài chính và ý chí của một người khác hai mươi năm sau. Hãy viết cả hậu quả của việc không sinh ra, rồi đối chiếu hai bên. +- Lợi ích: Cách nghĩ giống phần 10. Kỳ vọng của người khác có thể tính là một yếu tố cân nhắc của bạn. Nhưng chi phí thì ghi vào sổ của chính bạn, không ghi vào sổ của họ. 「Nuôi con khi về già」càng phải tính riêng một cuốn, vì nó giả định năng lực tài chính và ý chí của một người khác hai mươi năm sau - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; xem phân tích cùng loại ở phần 10 -- Ghi chú: Hãy viết cả 「hậu quả của việc không sinh」 rồi đặt cạnh 「hậu quả của việc sinh」 để đối chiếu +- Ghi chú: Hãy viết cả「hậu quả của việc không sinh」rồi đặt cạnh「hậu quả của việc sinh」để đối chiếu \ No newline at end of file diff --git "a/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" "b/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" new file mode 100644 index 00000000..7ffa1cd1 --- /dev/null +++ "b/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" @@ -0,0 +1,156 @@ +[← Về mục lục](../README.md) + +# 19. Đi làm, nghỉ việc và tai nạn lao động + +Phần này tính tiền: tiền thế nào được tính, chữ ký thế nào, hạn thế nào. Ba mục đầu nói về phần còn giữ được của người đang làm việc, mục 4 đến 8 nói về cách nghỉ việc cho đúng, từ mục 9 trở đi nói về tai nạn lao động. + +### 1. Tiền làm thêm giờ tính theo ba mức 1.5 lần, 2 lần, 3 lần; không trả thì báo thanh tra lao động, quá hạn không trả còn phải cộng thêm 50% đến 100% + +- Chi phí: Không tốn tiền. Tiện tay lưu một bản bản chấm điểm giờ ra vào, lịch trực và đơn xin làm thêm giờ. Tin nhắn phân công việc của sếp trong nhóm chat thì chụp màn hình lưu lại. Chỗ khó là phải nhớ lưu mỗi tháng. +- Nói thẳng: Làm thêm giờ trong ngày thì trả gấp 1.5 lần lương. Ngày nghỉ lễ mà không được bù ngày nghỉ thì gấp 2 lần. Ngày lễ tết theo luật như Tết Nguyên Đán, Tết Nguyên Tiêu, Quốc tế Lao động, Quốc khánh thì gấp 3 lần, và không được lấy ngày nghỉ bù để thay. Công ty không trả thì báo thanh tra lao động. Sau khi bị yêu cầu trả trong thời hạn mà vẫn không trả, công ty còn phải trả thêm từ một nửa đến một lần khoản còn nợ. +- Lợi ích: Luật quy định giờ làm việc bình thường không quá 8 giờ mỗi ngày, tính trung bình không quá 44 giờ mỗi tuần. Làm thêm giờ trong ngày thì thường không quá 1 giờ mỗi ngày. Có lý do đặc biệt phải làm thêm thì không quá 3 giờ mỗi ngày, và cộng lại trong tháng không quá 36 giờ. Tiền làm thêm giờ chia ba mức. Bố trí bạn làm thêm giờ trong ngày thì trả không dưới 150% lương. Bố trí bạn đi làm ngày nghỉ mà không bù được ngày nghỉ thì không dưới 200%. Bố trí bạn đi làm ngày lễ tết theo luật thì không dưới 300%. Riêng mức ngày lễ tết, trong luật không có lựa chọn「lấy ngày nghỉ bù để thay cho tiền」. Đơn vị bố trí làm thêm giờ mà không trả tiền thì bị cơ quan quản lý lao động buộc phải trả đủ trong thời hạn. Quá thời hạn vẫn không trả thì bị buộc phải trả thêm cho người lao động một khoản bồi thường, từ 50% đến 100% trên số tiền phải trả, tính ngoài khoản còn nợ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 劳动法（2018 年第二次修正，第三十六、四十一、四十三、四十四条）. （Cục Quản lý thị trường quốc gia đăng lại）；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 劳动合同法（第八十五条第三项）. +- Ghi chú: Ngày nghỉ và ngày lễ tết theo luật phải tính riêng. Ngày lễ tết theo luật cả năm chỉ có Tết Nguyên Đán, Tết Nguyên Tiêu, Quốc tế Lao động, Quốc khánh và mấy ngày lễ hợp pháp khác. Ngày nghỉ liên tục do điều chỉnh lịch nghỉ để bù thì tính là ngày nghỉ, trả 200%. Hai câu「công ty quy định làm thêm giờ phải được duyệt」và「tôi tự nguyện ở lại」không có nghĩa công ty được phép không trả tiền, nhưng chúng làm bạn khó lấy bằng chứng hơn, nên phải giữ lại lệnh phân công và hồ sơ duyệt. Công ty muốn áp dụng hai chế độ giờ làm đặc biệt là tính giờ tổng hợp và giờ không xác định thì phải được cơ quan quản lý lao động chấp thuận trước, không tự quyết được. Thời hiệu khởi kiện khi còn tranh cãi xem mục 19 ở phần 8. + +### 2. Nghỉ phép năm tính theo thâm niên cộng dồn: 5, 10, 15 ngày; nghỉ không được thì đổi thành tiền bằng 300% lương ngày + +- Chi phí: Không tốn tiền. Cộng lại số năm bạn đã làm ở tất cả các đơn vị. Sổ đóng bảo hiểm xã hội chính là bằng chứng. Chỗ khó là phải ký được tờ「tự nguyện không nghỉ phép năm」khi công ty đưa ra. +- Nói thẳng: Cộng dồn thâm niên ở mọi đơn vị đủ 1 năm đến chưa đủ 10 năm thì mỗi năm có 5 ngày nghỉ phép năm. Đủ 10 năm là 10 ngày, đủ 20 năm là 15 ngày. Đổi việc thì vẫn cộng dồn tiếp như cũ. Đơn vị không sắp xếp được mà bạn đồng ý không nghỉ thì mấy ngày đó trả gấp 3 lần lương ngày. Ký tờ「tự nguyện không nghỉ phép năm」xong thì chỉ còn lương bình thường. +- Lợi ích: Làm việc liên tục đủ 12 tháng là có nghỉ phép năm. Số ngày tính theo thời gian làm việc **cộng dồn**, cộng cả số năm đã làm ở các đơn vị khác. Đủ 1 năm đến chưa đủ 10 năm là 5 ngày, đủ 10 năm đến chưa đủ 20 năm là 10 ngày, đủ 20 năm là 15 ngày. Ngày lễ tết theo luật của nhà nước và ngày nghỉ không tính vào số ngày nghỉ phép năm. Đơn vị thật sự vì công việc không sắp xếp được, lại có bạn đồng ý không nghỉ, thì mấy ngày đáng lẽ phải nghỉ mà không nghỉ được thì「按照该职工日工资收入的 300% 支付年休假工资报酬」, tức là trả tiền lương nghỉ phép năm bằng ba lần thu nhập lương ngày của người lao động đó. Trong 300% đó đã có phần lương của một ngày làm việc bình thường vốn phải có, nên phần thực tế cộng thêm là 200%. Lương ngày = lương tháng của bạn ÷ 21.75. Lương tháng tính theo 12 tháng trước khi trả khoản tiền này, bỏ tiền làm thêm giờ rồi lấy trung bình tháng. Năm mới vào đơn vị thì quy đổi theo số ngày lịch còn lại trong đơn vị đó. Khi nghỉ việc thì quy đổi và quyết toán theo thời gian đã làm trong năm đó, đã nghỉ nhiều hơn thì không trừ lại. Đơn vị vừa không sắp xếp nghỉ, vừa không trả tiền thì bị buộc phải sửa trong thời hạn. Quá thời hạn vẫn không sửa thì ngoài khoản lương nghỉ phép năm đó còn phải trả thêm một khoản bồi thường bằng đúng số tiền đó (toàn quốc, có hiệu lực từ ngày 1 tháng 1 năm 2008) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2007). 职工带薪年休假条例（国务院令第 514 号，第二、三、四、五、七条）. ；Bộ Nhân sự và Xã hội (2008). 企业职工带薪年休假实施办法（人社部令第 1 号，第三、四、五、十、十一、十二、十五条）. +- Ghi chú: Thời gian làm việc cộng dồn phải cộng cả các đơn vị, đã từng đổi việc thì đừng tính theo thâm niên ở đơn vị hiện tại. **Chỉ khi người lao động「tự nguyện không nghỉ vì lý do của bản thân và có văn bản」thì đơn vị mới chỉ phải trả lương bình thường**, nên đừng ký tờ「tự nguyện không nghỉ phép năm」. Nghỉ phép việc tự sự tích luỹ từ 20 ngày trở lên mà đơn vị không trừ lương thì năm đó không được nghỉ phép năm. Nghỉ ốm vượt số tháng tương ứng thì cũng không được: thâm niên chưa đủ 10 năm là 2 tháng, từ 10 đến 20 năm là 3 tháng, trên 20 năm là 4 tháng. Phép thăm viếng, phép cưới tang, phép thai sản và thời gian lương được trả khi ngừng việc vì tai nạn lao động không tính vào nghỉ phép năm. + +### 3. Thời gian thử việc có trần theo luật, lương không được dưới 80%, và chỉ được thỏa thuận một lần + +- Chi phí: Không tốn tiền. Trước khi ký, dành vài phút để đối chiếu thời hạn ghi trong hợp đồng với độ dài thời gian thử việc. +- Nói thẳng: Hợp đồng dưới hoặc bằng 1 năm thì thời gian thử việc tối đa 1 tháng. Hợp đồng từ 1 đến 3 năm thì tối đa 2 tháng. Hợp đồng trên 3 năm thì tối đa 6 tháng. Cùng một công ty chỉ thử việc bạn một lần, chuyển sang vị trí khác sau khi hết thử việc rồi thử lần nữa là vi phạm pháp. Lương thời gian thử việc không được thấp hơn mức lương thấp nhất cùng vị trí, cũng không được thấp hơn 80% lương hợp đồng quy định, và cũng không được thấp hơn lương tối thiểu tại địa phương. +- Lợi ích: Hợp đồng trên 3 tháng đến chưa đủ 1 năm thì thời gian thử việc không quá 1 tháng. Hợp đồng trên 1 năm đến chưa đủ 3 năm thì không quá 2 tháng. Hợp đồng trên 3 năm, cùng hợp đồng không ghi ngày hết hạn, thì không quá 6 tháng. Hợp đồng có thời hạn là làm xong một việc, hoặc hợp đồng dưới 3 tháng, thì không được thỏa thuận thời gian thử việc. 「同一用人单位与同一劳动者只能约定一次试用期」, tức là cùng một công ty chỉ thử một lần với cùng một người. Hợp đồng chỉ ghi thời gian thử việc thì thời gian thử việc đó không thành lập, và khoảng thời gian ấy chính là thời hạn của hợp đồng lao động. Lương thời gian thử việc「不得低于本单位相同岗位最低档工资或者劳动合同约定工资的百分之八十，并不得低于用人单位所在地的最低工资标准」, tức là phải bằng ít nhất tám phần trăm mức lương thấp nhất cùng vị trí, đồng thời không được thấp hơn mức lương tối thiểu tại nơi đơn vị sử dụng lao động đặt trụ sở. Thỏa thuận thời gian thử việc dài hơn trần theo luật mà bạn đã đi làm thật, thì đơn vị phải trả thêm một khoản bồi thường cho khoảng thời gian vượt quá, tính theo mức lương tháng của bạn sau khi hết thời gian thử việc. Trong thời gian thử việc, đơn vị muốn chấm dứt hợp đồng chỉ được theo các trường hợp luật liệt kê, và phải nói rõ lý do với người lao động (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 劳动合同法（第十九、二十、二十一、八十三条）. +- Ghi chú: Ba cách làm sai thường gặp nhất. Một là hợp đồng chỉ ký 1 năm nhưng lại thỏa thuận thời gian thử việc 3 tháng. Hai là sau khi hết thử việc thì chuyển vị trí, rồi lại có một thời gian thử việc nữa. Ba là「thời gian thử việc không đóng bảo hiểm xã hội」, nghĩa vụ đóng bảo hiểm xã hội tính từ ngày đầu tiên đi làm, không liên quan gì đến thử việc, nghĩa vụ của phía sử dụng lao động xem mục 16 ở phần 12. Bị sa thải trong thời gian thử việc không có nghĩa là mất trắng, phải xem lý do công ty đưa ra có đúng là các trường hợp luật cho phép sa thải hay không. Bị chấm dứt sai luật thì tính 2N theo mục 6 của phần này. + +### 4. Bị sa thải thì tính N trước: cứ đủ mỗi năm một tháng lương, chưa đủ sáu tháng thì tính nửa tháng + +- Chi phí: Không tốn tiền. Bấm máy tính một lần, mất mười phút. +- Nói thẳng: Khoản bồi thường kinh tế công ty phải trả khi bạn bị sa thải tính thế này: đủ một năm thì một tháng lương, đủ nửa năm đến chưa đủ một năm thì tính như một năm, chưa đủ nửa năm thì nửa tháng. Mức lương tháng ở đây là trung bình 12 tháng trước khi rời đi, có tính cả thưởng, trợ cấp, phụ cấp. Lương vượt 3 lần mức lương tháng bình quân của năm trước tại địa phương thì tính trần ở 3 lần, số năm công chỉ tính tối đa 12 năm. +- Lợi ích: Khoản bồi thường kinh tế theo hợp đồng lao động chính là khoản tiền đơn vị phải trả khi bạn bị sa thải, mọi người thường gọi là N. Cách tính là：「每满一年支付一个月工资……六个月以上不满一年的，按一年计算；不满六个月的，向劳动者支付半个月工资」, tức là đủ một năm thì một tháng lương, từ sáu tháng đến chưa đủ một năm thì tính như một năm, chưa đủ sáu tháng thì trả nửa tháng lương. Lương tháng vượt ba lần lương tháng bình quân của người lao động năm trước tại địa phương thì tính trần ở ba lần, và thâm niên chỉ tính tối đa đến mười hai năm +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第四十七条）. +- Ghi chú: Lương tháng ở đây là lương trung bình mười hai tháng trước khi hợp đồng lao động chấm dứt. Thưởng, trợ cấp và phụ cấp đều phải tính vào, không chỉ tính lương cơ bản. Đây là điểm công ty dễ làm thấp nhất. + +### 5. Công ty không báo trước 30 ngày mà cho bạn nghỉ thì còn phải trả thêm một tháng lương + +- Chi phí: Không tốn tiền. +- Nói thẳng: Công ty không báo trước 30 ngày bằng văn bản mà bắt bạn đi thì phải trả thêm một tháng lương ngoài N, đó là cái người ta gọi là N+1. Khi thương lượng, hãy tách tháng lương đó thành một khoản riêng, đừng để công ty gộp vào trong N. +- Lợi ích: Hợp đồng lao động quy định khi đơn vị sử dụng lao động chấm dứt hợp đồng theo điều này thì có hai cách. Cách một là「提前三十日以书面形式通知劳动者本人」. Cách hai là「额外支付劳动者一个月工资后，可以解除劳动合同」. Muốn bạn đi ngay thì phải bỏ thêm một tháng lương. Tháng lương thêm đó dân gọi là tiền báo trước, tính ngoài N +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第四十条）. +- Ghi chú: Vì vậy kết quả thường gặp của một cuộc sa thải hợp pháp là N+1. Khi thương lượng, tính riêng N và 1, đừng để công ty gộp tháng lương thêm vào trong N. + +### 6. Công ty chấm dứt hợp đồng sai luật thì tiền bồi thường bằng hai lần mức bồi thường kinh tế + +- Chi phí: Không tốn tiền. Đi qua hoà giải lao động phải chờ mấy tháng. Chỗ khó là mấy tháng đó bạn phải theo suốt, không được bỏ giữa chừng. +- Nói thẳng: Nếu công ty chấm dứt hợp đồng sai luật thì tiền bồi thường gấp đôi mức bồi thường kinh tế, tức là cái người ta gọi là 2N. Có phải chấm dứt sai luật hay không phải xem từng trường hợp. Hãy hỏi trợ giúp pháp lý hoặc luật sư cho rõ trước, rồi mới quyết định là thương lượng với công ty hay đi qua hoà giải lao động. +- Lợi ích: Hợp đồng lao động quy định, khi đơn vị sử dụng lao động vi phạm luật này mà chấm dứt hoặc kết thúc hợp đồng lao động thì「应当依照本法第四十七条规定的经济补偿标准的二倍向劳动者支付赔偿金」, tức là phải trả cho người lao động khoản tiền bồi thường bằng hai lần mức bồi thường kinh tế mà luật đó quy định. Cách tính là tính ra một khoản theo tiêu chuẩn bồi thường kinh tế trước, rồi nhân đôi +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第八十七条）. +- Ghi chú: Đây là cái dân gọi là 2N. Cái gì tính là chấm dứt sai luật thì phải xem từng trường hợp. Hãy hỏi trợ giúp pháp lý hoặc luật sư trước, rồi mới quyết định thương lượng hay đi khởi kiện, các bước đi xem phần 7. + +### 7. Đừng ký「tự nguyện nghỉ việc vì lý do cá nhân」, ký là mất trắng cả khoản N + +- Chi phí: Không tốn tiền. Chỗ khó là phải chịu được áp lực bị bảo ký ngay tại chỗ. +- Nói thẳng: Ký xác nhận「tự nguyện nghỉ việc vì lý do cá nhân」thì bồi thường kinh tế và tiền bảo hiểm thất nghiệp thường cùng mất trắng. Chính bạn đề nghị nghỉ việc vốn không thuộc các trường hợp phải bồi thường, trừ lỗi hợp pháp của công ty như nợ lương, không đóng bảo hiểm xã hội. Không cần ký trong ngày, để nguôi một đêm rồi hãy quyết. +- Lợi ích: Hợp đồng lao động giới hạn các trường hợp phải trả bồi thường kinh tế vào đúng những loại luật liệt kê. Người lao động tự đề nghị nghỉ việc thì không nằm trong đó, trừ khi do lỗi hợp pháp của công ty như nợ lương, không đóng bảo hiểm xã hội. Ký「tự nguyện nghỉ việc」xong thì bồi thường kinh tế và tiền bảo hiểm thất nghiệp thường cùng mất trắng +- Mức bằng chứng: B +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第三十八条、第四十六条）. +- Ghi chú: Câu công ty hay dùng là「ký trước để làm thủ tục, phần bồi thường tính riêng」. Muốn nhận thì ghi thẳng vào văn kiện ngay tại chỗ, không ghi thì không ký. Không ký trong ngày không có bất kỳ hậu quả pháp lý nào, để nguôi một đêm rồi hãy quyết. + +### 8. Trước khi nghỉ việc, lưu lại phiếu lương, bản chấm công, hợp đồng lao động, sổ bảo hiểm xã hội và lịch sử chat + +- Chi phí: Không tốn tiền. Dành nửa tiếng để lưu hồ sơ. Chỗ khó là phải nhớ lưu trước khi đi. +- Nói thẳng: Trả lại máy tính và tài khoản là phiếu lương, bản chấm công, hợp đồng, sổ bảo hiểm xã hội và lịch sử chat không lấy lại được nữa. Khi khởi kiện, lương bao nhiêu, có làm thêm giờ hay không, vẫn phải dựa vào hồ sơ trong tay bạn. Hãy lưu hồ sơ quan hệ lao động của chính bạn, đừng mang mã nguồn và danh sách khách hàng của công ty đi. +- Lợi ích: Một khi trả lại máy tính và tài khoản thì những hồ sơ đó không lấy lại được nữa. Khi khởi kiện, việc ai phải đưa bằng chứng tuy nghiêng về phía đơn vị hơn một chút, nhưng lương bao nhiêu, có làm thêm giờ hay không thì vẫn dựa vào bằng chứng trong tay bạn +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; các bước bảo vệ quyền lợi xem phần 7 +- Ghi chú: Hãy lưu hồ sơ quan hệ lao động của chính bạn. Đừng tiện tay mang mã nguồn, danh sách khách hàng và tài liệu kỹ thuật của công ty đi, rủi ro của việc mang đi xem phần 11. +### 9. Ngay sau khi nghỉ việc, làm hai việc: đăng ký thất nghiệp để lĩnh tiền bảo hiểm thất nghiệp, và đọc kỹ hợp đồng cạnh tranh + +- Chi phí: Không tốn tiền. +- Nói thẳng: Ngay sau khi nghỉ việc, làm hai việc. Một là đi đăng ký thất nghiệp và lĩnh tiền bảo hiểm thất nghiệp. Hai là đọc lại hợp đồng cạnh tranh. Tiền bảo hiểm thất nghiệp tính từ ngày làm thủ tục đăng ký, đăng ký muộn một tháng là mất một tháng tiền. Chỉ người mất việc vì lý do không phải do mình mới được lĩnh, đó cũng là một lý do nữa cho việc không nên ký theo mục 7 của phần này (đừng ký「tự nguyện nghỉ việc vì lý do cá nhân」). +- Lợi ích: Để lĩnh tiền bảo hiểm thất nghiệp có bốn điều kiện: đã đóng bảo hiểm thất nghiệp đủ 1 năm, mất việc vì lý do không phải do mình muốn đi (không thất nghiệp do chủ ý của bản thân), đã làm thủ tục đăng ký thất nghiệp, và vẫn đang đi tìm việc. Số năm đã đóng quyết định lâu nhất là được lĩnh 12, 18 hay 24 tháng. Cách đăng ký trên mạng xem mục 1 ở phần 7 (mất việc thì đăng ký tiền bảo hiểm thất nghiệp trên mạng trước). Hạn chế cạnh tranh dài nhất là 2 năm. Trong 2 năm đó công ty phải trả bạn bồi thường hằng tháng. Hợp đồng không ghi là bao nhiêu thì tính bằng 30% lương tháng trung bình 12 tháng trước khi nghỉ việc, và không được thấp hơn lương tối thiểu tại địa phương. Chỉ khi công ty liên tục 3 tháng không tráng vì lý do từ phía mình thì bạn mới được yêu cầu chấm dứt thoả thuận đó, xem mục 12 ở phần 11 (đã ký hợp đồng cạnh tranh mà công ty không trả bồi thường thì làm gì). Hạn chế cạnh tranh có thời hạn, nhưng trước ngày hết hạn thì vẫn phải giữ đúng thoả thuận +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; những con số dẫn ra từ Ủy ban Nhà nước (国务院) (1999). 失业保险条例（国务院令第 258 号，第十四、十六、十七条：「失业保险金自办理失业登记之日起计算」）. và Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). 劳动合同法（2012 年修正，第二十三、二十四条）. （Sở Quản lý Khẩn cấp thành phố Phụ Thuận đăng lại） +- Ghi chú: Để ở mức C là vì hành động「ngày nghỉ việc đi làm ngay」không có tài liệu trực tiếp, còn những con số trong mục Nguồn đều là nguyên văn quy định. Tiền bảo hiểm thất nghiệp không hết hạn mất tác dụng, nhưng nó tính từ ngày làm thủ tục đăng ký, nên làm sớm một ngày là có thêm một ngày tiền. Hồ sơ và cửa đăng ký trên mạng xem phần 7. Mức bồi thường và điều kiện chấm dứt hạn chế cạnh tranh xem phần 11. Đã ký「tự nguyện nghỉ việc vì lý do cá nhân」thì không những không có N, tiền bảo hiểm thất nghiệp cũng mất luôn, xem mục 7 của phần này (đừng ký「tự nguyện nghỉ việc vì lý do cá nhân」). + +### 10. Trước khi vào chỗ làm việc có bụi, tiếng ồn, hóa chất, xem hợp đồng có ghi nguy cơ nghề nghiệp không; ba lần khám sức khoẻ nghề nghiệp do đơn vị sắp xếp và trả tiền + +- Chi phí: Không tốn tiền. Khi ký hợp đồng, đọc hết mục「职业病危害告知」từng điều một. Báo cáo mỗi lần khám sức khoẻ thì giữ lại một bản. Chỗ khó là phải hỏi thẳng mặt và xin được hồ sơ ngay tại chỗ. +- Nói thẳng: Khi ký hợp đồng, đơn vị phải ghi thật vào hợp đồng vị trí này có nguy cơ nghề nghiệp gì, bảo vệ bằng cách nào, được hưởng ưu đãi gì. Giấu bạn thì bạn có quyền từ chối làm việc đó, công ty không được vì thế mà sa thải bạn. Ba lần khám sức khoẻ nghề nghiệp gồm trước khi nhận việc, trong thời gian làm việc, và khi rời khỏi công việc, do đơn vị sắp xếp và trả tiền. Không làm khám sức khoẻ khi rời việc thì đơn vị không được chấm dứt hợp đồng với bạn. +- Lợi ích: Luật phòng và chữa bệnh nghề nghiệp quy định, lúc ký hợp đồng lao động, đơn vị phải nói rõ bốn điều: làm việc này có thể sinh ra nguy cơ nghề nghiệp nào, để lại hậu quả gì, phòng ngừa bằng cách nào, được hưởng ưu đãi gì. Nguyên văn điều luật là「如实告知劳动者，并在劳动合同中写明，不得隐瞒或者欺骗」, tức là phải nói thật với người lao động và ghi vào hợp đồng lao động, được giấu hay lừa dối. Sau khi ký hợp đồng mà lại điều chuyển bạn sang việc có hại chưa nói trong hợp đồng thì phải nói thật lại, và bàn với bạn để sửa điều khoản hợp đồng. Đơn vị không làm thì「người lao động có quyền từ chối làm công việc có nguy cơ nghề nghiệp, đơn vị sử dụng lao động không được vì thế mà chấm dứt hợp đồng lao động đã ký với người lao động」. Khám sức khoẻ nghề nghiệp có ba lần: trước khi nhận việc, trong thời gian làm việc, và khi rời việc. 「Chi phí khám sức khoẻ nghề nghiệp do đơn vị sử dụng lao động chịu」, tiền do đơn vị đó trả, kết quả phải báo bằng văn bản cho chính người đó. Người chưa làm khám trước khi nhận việc thì đơn vị không được sắp xếp làm công việc có tiếp xúc với nguy cơ nghề nghiệp. Người vốn không hợp với loại việc đó (có chống chỉ định nghề nghiệp) cũng không được sắp xếp làm các công việc bị chống chỉ định đó. Khám ra tổn thương sức khoẻ có liên quan đến công việc này thì phải điều người đó ra khỏi vị trí cũ và sắp xếp lại cho tử tế. 「Đối với người lao động chưa làm khám sức khoẻ nghề nghiệp trước khi rời việc thì không được chấm dứt hoặc kết thúc hợp đồng lao động đã ký với người đó」. Khi nghỉ việc thì bạn「có quyền yêu cầu bản sao hồ sơ theo dõi sức khoẻ nghề nghiệp của mình, đơn vị sử dụng lao động phải cung cấp thật và miễn phí, và đóng dấu lên bản sao đó」. Hồ sơ theo dõi sức khoẻ nghề nghiệp đó, đơn vị phải sao chép thật một bản miễn phí cho bạn, rồi đóng dấu lên bản sao (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 职业病防治法（2018 年修正，第三十三、三十五、三十六条）. （Chính phủ thành phố Bắc Kinh đăng lại） +- Ghi chú: Khi phỏng vấn, hỏi ba câu là đo được đại thể chỗ này có nghiêm không: vị trí này có nguy cơ nghề nghiệp không, hợp đồng có ghi không, tiền khám trước khi nhận việc do ai trả. Không trả lời được, hoặc trả lời「làm thử đã」, thì tự rủi ro. Khám sức khoẻ khi rời việc là cái dễ bị bỏ qua nhất: không làm thì đơn vị không được chấm dứt hoặc kết thúc hợp đồng lao động với bạn. Chính cuộc khám này sau này còn là hồ sơ then chốt để xin chẩn đoán bệnh nghề nghiệp, bản sao hồ sơ cũng vậy, phải xin được trước khi đi. Bản thân bệnh nghề nghiệp thì áp dụng như tai nạn lao động, các ưu đãi xem từ mục 12 trở đi. + +### 11. Tổn thương do bụi, tiếng ồn, độc chất hoá học thì không đảo ngược: đơn vị phải cấp phương tiện bảo vệ, công việc không có bảo vệ thì được từ chối + +- Chi phí: Không tốn tiền. Chỗ khó là phải đeo khẩu trang chống bụi hạt, nút bịt tai, kính bảo hộ suốt thời gian, còn phải từ chối công việc không có bảo vệ, cái giá là đồng nghiệp có thể nói bạn「nhiều chuyện」. +- Nói thẳng: Bệnh bụi phổi, điếc do tiếng ồn, bỏng mắt do hoá chất đều là bệnh nghề nghiệp theo luật, phổi và thính lực hỏng thì không chữa lại được. Bạn có quyền biết vị trí này có nguy cơ gì, và có quyền đòi đơn vị cấp thiết bị và phương tiện bảo vệ. Công việc không có bảo vệ thì bạn không làm cũng được. Đơn vị vì thế mà giảm lương hoặc sa thải bạn thì đều không có hiệu lực. +- Lợi ích: Từ ngày 1 tháng 8 năm 2025, bệnh nghề nghiệp theo luật gồm 12 nhóm lớn 135 loại, trong đó có 4 điều khoản mở. Nhóm đứng đầu là bệnh phổi bụi nghề nghiệp. Trong đó có bệnh phổi do silic, do bụi than, do bụi than chì, do bụi than đen, do bụi asbest, do bụi đá phiến. Rồi đến bệnh phổi do bụi xi măng, do bụi mica, do bụi gốm, do bụi nhôm, do hàn điện, do bụi luyện kim, cộng thêm một điều khoản mở. Bệnh tai mũi họng và miệng nghề nghiệp có điếc do tiếng ồn và điếc do sốc. Bệnh mắt nghề nghiệp có bỏng mắt do hoá chất và viêm mắt do ánh điện. Những tổn thương này không đảo ngược, không có cách chữa nào lấy lại phổi và thính lực. Luật đưa ra bốn quyền có thể dùng ngay tại chỗ. Một là biết nơi làm việc có những nguy cơ nghề nghiệp nào, gây hậu quả gì, phải áp dụng biện pháp bảo vệ nào. Hai là yêu cầu đơn vị cấp thiết bị bảo vệ và phương tiện bảo vệ cá nhân đạt chuẩn, và cải thiện điều kiện làm việc. Ba là phê bình, kiện cáo và tố cáo những cách làm vi phạm quy định phòng chữa bệnh nghề nghiệp, cũng như cách làm nguy hại tới tính mạng và sức khoẻ. Bốn là「từ chối chỉ đạo sai quy định và từ chối bị ép làm công việc không có biện pháp phòng chữa bệnh nghề nghiệp」, chỉ đạo sai quy định và công việc không có bảo vệ thì bạn không làm cũng được. Và「nếu vì người lao động thực hiện hợp pháp quyền lợi của mình mà giảm lương, giảm ưu đãi hoặc chấm dứt, kết thúc hợp đồng lao động đã ký với họ thì hành vi đó vô hiệu」. Dùng bốn quyền này mà bị giảm lương, giảm ưu đãi hoặc sa thải thì những cách đó không có hiệu lực (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Quốc gia Y tế, Bộ Nhân sự và Xã hội, Cục Quốc gia Kiểm soát và Phòng ngừa bệnh, Tổng Liên đoàn Lao động (2024). 职业病分类和目录（2025 年 8 月 1 日起实施）. ；cách tính 12 nhóm lớn 135 loại trong danh mục mới xem Ủy ban Quốc gia Y tế (2024). 新版目录将职业病调整为 12 大类 135 种. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 职业病防治法（第三十九条）. （Chính phủ thành phố Bắc Kinh đăng lại） +- Ghi chú: Phải chọn đúng khẩu trang, khẩu trang chống bụi hạt mới chặn được bụi, khẩu trang vải thường và khẩu trang y tế dùng một lần không chặn được. Chỗ làm việc có tiếng ồn mà đơn vị không phát nút bịt tai thì tự mua, vài tệ một cặp, mất thính lực thì không bù lại được. 「Làm hai năm kiếm đủ tiền rồi đi」không đúng với bệnh bụi phổi. Bệnh phổi do silic có thể khởi phát nhiều năm sau khi ngừng làm việc với bụi, và vẫn tiếp tục nặng thêm. Nên khám sức khoẻ khi rời việc lại càng quan trọng (xem mục 10 của phần này, ba lần khám sức khoẻ nghề nghiệp do đơn vị sắp xếp và trả tiền). Danh mục mới còn thêm hai nhóm. Một là bệnh cơ xương khớp nghề nghiệp, trong đó có hội chứng ống cổ tay, chỉ áp dụng cho công nhân sản xuất dùng cổ tay lặp lại lâu hoặc dùng sức. Hai là rối loạn tâm thần và hành vi nghề nghiệp, trong đó có rối loạn căng thẳng sau chấn thương, chỉ áp dụng cho công an nhân dân, nhân viên y tế, lực lượng cứu hộ và các lực lượng ứng cứu khẩn cấp có tham gia xử lý sự cố. Xử lý tại chỗ khi hoá chất bắn vào người xem mục 21 ở phần 13. + +### 12. Bị thương lúc làm việc, hay bị xe đâm trên đường đi làm: việc đầu tiên là xin xác định tai nạn lao động, đơn vị không báo thì bạn tự báo + +- Chi phí: Không tốn tiền, chỉ mất một chuyến tới cơ quan bảo hiểm xã hội. Mang theo ba loại giấy tờ: đơn xin xác định tai nạn lao động, giấy chứng minh quan hệ lao động, giấy chẩn đoán y tế. +- Nói thẳng: Bị thương lúc làm việc là tai nạn lao động. Bị xe đâm trên đường đi làm mà phần lỗi chính không ở bạn thì cũng là tai nạn lao động. Nhưng không xin xác định tai nạn lao động thì tiền khám chữa, lương được trả trong thời gian điều trị, và tiền trợ cấp thương tật đều mất trắng. Đơn vị phải báo trong 30 ngày. Đơn vị không báo thì bạn, hoặc người nhà và công đoàn, có thể tự báo với cơ quan bảo hiểm xã hội trong vòng một năm. +- Lợi ích: Không xin xác định tai nạn lao động thì tiền khám chữa, lương được trả trong thời gian điều trị (lương thời gian ngừng việc được trả), và tiền trợ cấp thương tật đều mất trắng. Đơn vị quá 30 ngày không báo thì trong thời gian đó, các khoản chi phí về ưu đãi tai nạn lao động phải trả thì chính đơn vị đó chịu +- Mức bằng chứng: A +- Nguồn: Ủy ban Nhà nước (国务院) (2010 修订). 工伤保险条例（国务院令第 586 号）第十四条列了七种「应当认定为工伤」，含「（六）在上下班途中，受到非本人主要责任的交通事故或者城市轨道交通、客运轮渡、火车事故伤害的」；第十七条「所在单位应当自事故伤害发生之日或者被诊断、鉴定为职业病之日起30日内，向统筹地区社会保险行政部门提出工伤认定申请」「用人单位未按前款规定提出工伤认定申请的，工伤职工或者其近亲属、工会组织在事故伤害发生之日或者被诊断、鉴定为职业病之日起1年内，可以直接向用人单位所在地统筹地区社会保险行政部门提出工伤认定申请」「用人单位未在本条第一款规定的时限内提交工伤认定申请，在此期间发生符合本条例规定的工伤待遇等有关费用由该用人单位负担」；第十九条「职工或者其近亲属认为是工伤，用人单位不认为是工伤的，由用人单位承担举证责任」；第二十条「自受理工伤认定申请之日起60日内作出工伤认定的决定」. +- Ghi chú: Một năm là thời hạn cứng, quá hạn là không còn đường điều. Đơn vị nói không phải tai nạn lao động thì đơn vị phải đưa bằng chứng. Nhưng bạn cũng phải giữ hồ sơ: ảnh chụp ngày xảy ra, thông tin liên lạc của đồng nghiệp, hồ sơ khám chữa, bản chấm công. Điều mười sáu của nghị định liệt kê ba trường hợp không công nhận: cố ý phạm tội, say rượu hoặc nghiện ma túy, tự làm hại bản thân hoặc tự sát. Điều mười lăm lại có「视同工伤」, tức là coi như tai nạn lao động: chết vì bệnh phát đột ngột trong giờ làm và tại chỗ làm, hoặc chết sau cấp cứu không cứu được trong 48 giờ. + +### 13. Đừng tin「leo tới bàn giấy là được coi là tai nạn lao động」: thấy khó chịu đột ngột thì gọi 120 trước, đừng chạy tới chấm công + +- Chi phí: Không tốn tiền. +- Nói thẳng: Khó chịu đột ngột mà chạy tới công ty không giúp bạn nhận thêm đồng nào. Coi như tai nạn lao động đòi hỏi người đó phát bệnh trong giờ làm việc và tại chỗ làm việc, phát bệnh trên đường chưa tới nơi thì vốn không tính. Cấp cứu quá 48 giờ thì ngược lại cũng không tính. Mấy chục phút chờ để khớp điều kiện là đổi lấy tỷ lệ sống sót của chính bạn. +- Lợi ích: Coi như tai nạn lao động đòi hỏi người đó「chết vì bệnh phát đột ngột trong giờ làm và tại chỗ làm」, hoặc chết sau cấp cứu không cứu được trong 48 giờ. Trì hoãn việc khám chữa để gom đủ hai điều kiện đó thì cái bạn trả là cơ hội sống sót của mình, mà khoản bồi thường đổi lại còn không chắc nhận được +- Mức bằng chứng: A +- Nguồn: Điều mười lăm, khoản một, mục (一) của Quy định về bảo hiểm tai nạn lao động:「在工作时间和工作岗位，突发疾病死亡或者在48小时之内经抢救无效死亡的」视同工伤；Điều mười bốn, mục (六) chỉ xếp「受到非本人主要责任的交通事故或者城市轨道交通、客运轮渡、火车事故伤害的」khi đi trên đường là tai nạn lao động, phát bệnh đột ngột giữa đường thì không nằm trong đó. +- Ghi chú: Hai điều kiện phải có cùng lúc: đang trong giờ làm, và đang ở chỗ làm việc. Phát bệnh trên đường chưa tới nơi thì không áp dụng. Phát bệnh lúc đã đến chỗ rồi được đưa về nhà hoặc đưa đi bệnh viện, chết sau cấp cứu không cứu được trong 48 giờ thì vẫn có thể áp dụng. Người ta xem lúc phát bệnh người đó ở đâu, không xem cuối cùng chết ở đâu. Video ngắn giải thích điều này thành「leo tới công ty là có tiền」, hướng ngược hẳn. Cái quyết định kết cục là mấy chục phút sau khi phát bệnh có đi bệnh viện không. Ngoài ra 48 giờ là ranh giới cứng, cấp cứu quá 48 giờ thì ngược lại không tính. + +### 14. Đơn vị không đóng bảo hiểm tai nạn lao động cho bạn thì ưu đãi tai nạn lao động vẫn có, và do đơn vị trả đủ theo đúng tiêu chuẩn + +- Chi phí: Không tốn tiền. Đơn vị không công nhận thì phải đi qua hoà giải lao động, thậm chí ra tòa, ít nhất mấy tháng. Chỗ khó là mấy tháng đó phải theo suốt. +- Nói thẳng: Đơn vị không đóng bảo hiểm tai nạn lao động cho bạn thì ưu đãi tai nạn lao động vẫn có. Chỉ khác ở chỗ đổi thành đơn vị tự trả đủ các khoản và theo đúng tiêu chuẩn, không bớt một phần nào. Không tham gia bảo hiểm là đơn vị bị phạt thêm: bị buộc đóng bù, bị tính phí trễ nộp theo ngày là năm phần mười nghìn, quá hạn còn bị phạt bằng một đến ba lần số tiền chậm nộp. 「Công ty không mua bảo hiểm tai nạn lao động nên không đền được」là lời nói sai. +- Lợi ích: Quy định ghi rõ: đơn vị không đóng bảo hiểm tai nạn lao động cho người lao động, khi người đó bị tai nạn lao động thì chính đơn vị đó trả các khoản tiền theo đúng khoản ưu đãi và tiêu chuẩn quy định. Phải trả những khoản nào, mỗi khoản bao nhiêu, giống hệt người đã đóng bảo hiểm, chỉ khác là đơn vị tự bỏ tiền ra +- Mức bằng chứng: A +- Nguồn: Điều sáu mươi hai, khoản hai của Quy định về bảo hiểm tai nạn lao động:「依照本条例规定应当参加工伤保险而未参加工伤保险的用人单位职工发生工伤的，由该用人单位按照本条例规定的工伤保险待遇项目和标准支付费用。」Khoản một cùng điều: bị buộc phải tham gia và đóng bù trong thời hạn,「自欠缴之日起，按日加收万分之五的滞纳金；逾期仍不缴纳的，处欠缴数额1倍以上3倍以下的罚款」. +- Ghi chú: 「Công ty không mua bảo hiểm tai nạn lao động cho bạn nên không đền được」là câu nói sai phổ biến nhất trong loại việc này. Không tham gia bảo hiểm chỉ khiến đơn vị bị phạt thêm, ưu đãi của bạn không giảm. +### 15. Khi tình trạng thương tích đã ổn định thì đi đánh giá năng lực lao động, hạng khuyết tật quy đổi thẳng thành tiền + +- Chi phí: Không tốn tiền. Đi một lượt quy trình đánh giá năng lực lao động, trước sau phải chờ một thời gian +- Nói thẳng: Sau khi tình trạng thương tích ổn định thì đi đánh giá năng lực lao động, hạng đánh giá quy đổi thẳng thành tiền. Tiền trợ cấp khuyết tật một lần tính theo lương của chính bạn: hạng năm 18 tháng, hạng sáu 16 tháng, hạng bảy 13 tháng, hạng tám 11 tháng, hạng chín 9 tháng, hạng mười 7 tháng. Trước khi có kết quả đánh giá, đừng ký thoả thuận 「giải quyết một lần」 +- Lợi ích: Tiền trợ cấp khuyết tật một lần được quy đổi theo hạng khuyết tật thành mấy tháng lương, rồi trả một lần. Hạng năm là 18 tháng, hạng sáu là 16 tháng, hạng bảy là 13 tháng lương của chính bạn. Hạng tám là 11 tháng, hạng chín là 9 tháng, hạng mười là 7 tháng lương của chính bạn +- Mức bằng chứng: A +- Nguồn: Quy định về bảo hiểm tai nạn lao động, 工伤保险条例 điều hai mươi mốt: 「经治疗伤情相对稳定后存在残疾、影响劳动能力的，应当进行劳动能力鉴定」; điều hai mươi hai: 「劳动功能障碍分为十个伤残等级，最重的为一级，最轻的为十级」; điều ba mươi sáu, chế độ cho hạng năm và hạng sáu, gồm tiền trợ cấp khuyết tật một lần 18/16 tháng lương của chính bạn và trợ cấp khuyết tật hằng tháng bằng 70%/60% lương của chính bạn; điều ba mươi bảy, chế độ cho hạng bảy đến hạng mười, gồm 13/11/9/7 tháng lương của chính bạn. +- Ghi chú: Hạng năm, hạng sáu giữ quan hệ lao động, người đó vẫn là công nhân viên của đơn vị, hằng tháng nhận trợ cấp khuyết tật. Người ở hạng bảy đến hạng mười, khi hợp đồng hết hạn hoặc khi chính họ đề nghị chấm dứt thì có thêm hai khoản tiền: tiền trợ cấp y tế tai nạn lao động một lần do quỹ chi trả, tiền trợ cấp việc làm cho người khuyết tật một lần do đơn vị chi trả, mức bao nhiêu do tỉnh quyết định. Trước khi có kết quả đánh giá thì đừng ký thoả thuận 「giải quyết một lần」 + +### 16. Phải phân biệt ba khoản tiền khi chết vì tai nạn lao động: tiền trợ cấp mai táng, tiền trợ cấp hưu trí người được nuôi dưỡng, tiền trợ cấp một lần khi chết vì tai nạn lao động + +- Chi phí: Không tốn tiền, làm một bộ thủ tục giấy tờ +- Nói thẳng: Chết vì tai nạn lao động là ba khoản tiền, đừng chỉ thương lượng một khoản. Tiền trợ cấp mai táng bằng 6 tháng lương bình quân theo tháng của công nhân viên năm trước tại địa phương. Tiền trợ cấp hưu trí người được nuôi dưỡng trả theo tháng theo lương của người đã mất, vợ hoặc chồng mỗi tháng 40%, người thân khác mỗi người mỗi tháng 30%. Tiền trợ cấp một lần khi chết vì tai nạn lao động bằng 20 lần thu nhập bình quân của dân cư đô thị cả nước năm trước, cả nước chỉ có một con số này +- Lợi ích: Khoản thứ nhất là tiền trợ cấp mai táng, bằng 6 tháng lương bình quân theo tháng của công nhân viên năm trước ở nơi tham gia bảo hiểm (khu vực chi trả thống nhất). Khoản thứ hai là tiền trợ cấp hưu trí người được nuôi dưỡng, trả theo tháng theo lương của người đã mất, vợ hoặc chồng mỗi tháng 40%, người thân khác mỗi người mỗi tháng 30%. Khoản thứ ba là tiền trợ cấp một lần khi chết vì tai nạn lao động, bằng 20 lần thu nhập bình quân của dân cư đô thị cả nước năm trước, cả nước chỉ có một con số này, không chia theo vùng. Với người chết vì tai nạn lao động năm 2026, khoản này là khoảng 1.130.000 tệ (con số chính xác 1130040 tệ). Cách tính là thu nhập bình quân của dân cư đô thị cả nước năm trước là 56502 tệ nhân với 20 +- Mức bằng chứng: A +- Nguồn: Quy định về bảo hiểm tai nạn lao động, 工伤保险条例第三十九条: 「（一）丧葬补助金为6个月的统筹地区上年度职工月平均工资；（二）供养亲属抚恤金按照职工本人工资的一定比例发给由因工死亡职工生前提供主要生活来源、无劳动能力的亲属。标准为：配偶每月40%，其他亲属每人每月30%，孤寡老人或者孤儿每人每月在上述标准的基础上增加10%……（三）一次性工亡补助金标准为上一年度全国城镇居民人均可支配收入的20倍。」. ；Cục Thống kê Quốc gia (国家统计局) (2026). Tình hình thu nhập và chi tiêu của dân cư năm 2025, 2025 年居民收入和消费支出情况. +- Ghi chú: Tiền trợ cấp một lần khi chết vì tai nạn lao động là mức thống nhất cả nước, số tiền cập nhật hằng năm theo số liệu Cục Thống kê Quốc gia công bố. Khi thương lượng bồi thường, không chấp nhận lý do 「tiêu chuẩn địa phương của chỗ chúng tôi thấp」. + +### 17. Ở công ty lâu ngày bị bắt nạt, mắng, gây khó, đừng gồng chịu: ghi lại để có chứng cứ, rồi tuỳ tính chất mà khiếu nại, báo công an hoặc đi hoà giải + +- Chi phí: Không tốn tiền. Phải tốn thời gian ghi chép: ngày tháng, người có mặt, nguyên văn lời của đối phương, ảnh chụp màn hình trò chuyện và giữ bản gốc email. Đi hoà giải lao động, ra tòa phải mất mấy tháng. Đổi việc thì cái giá cao nhất, đi hay không thì bạn tự tính +- Nói thẳng: Người bị bắt nạt ở công ty thì về sau nguy cơ mắc bệnh tim, đột quỵ cao hơn khoảng sáu phần mười, bị bắt nạt càng nhiều thì nguy cơ càng cao. Người bị bắt nạt thì xác suất nảy sinh ý nghĩ tự sát vài năm sau cao gấp khoảng hai lần người khác. Chịu mắng, chịu khí ở nơi làm việc không phải chuyện nhỏ mà nhịn một chút là qua +- Lợi ích: Về sức khoẻ, ba nhóm theo dõi của Thụy Điển và Đan Mạch gộp lại gồm 79.201 người đang làm việc từ 18 đến 65 tuổi. Ban đầu không ai mắc bệnh tim mạch, theo dõi trung bình 12.4 năm. 9% nói rằng trong năm qua họ bị bắt nạt ở nơi làm việc. Người bị bắt nạt thì nguy cơ mắc bệnh tim mạch (bệnh mạch vành và bệnh mạch não) cao hơn khoảng 59% (HR 1.59, 95% CI 1.28–1.98). Người từng gặp bạo lực ở nơi làm việc thì cao hơn khoảng 25% (HR 1.25, 1.12–1.40). Cả hai đều là gặp càng nhiều thì nguy cơ càng cao. Một khảo sát theo dõi lấy mẫu ngẫu nhiên toàn quốc của Na Uy, 1846 người đang làm việc, đo hai ba năm một lần, tổng cộng ba lần. Người bị bắt nạt thì xác suất sau đó nảy sinh ý nghĩ tự sát cao khoảng gấp 2 lần người khác (OR 2.05, 95% CI 1.08–3.89). Ngược lại, người đã có ý nghĩ tự sát trước đó thì sau đó không dễ bị bắt nạt hơn. Về mặt pháp lý, nước ta không có luật riêng cho 「bắt nạt tại nơi làm việc」, phải đối chiếu với từng hành vi cụ thể. Đơn vị xúc phạm, phạt thể xác, đánh đập, lục soát hoặc giam giữ trái phép người lao động thì bị xử phạt hành chính, gây thiệt hại thì phải bồi thường, cấu thành tội phạm thì truy cứ trách nhiệm hình sự. Khi đơn vị dùng bạo lực, đe dọa cưỡng ép bạn làm việc thì bạn có thể hủy hợp đồng ngay, không cần báo trước, và vẫn được nhận bồi thường kinh tế. Đơn vị phải có biện pháp phòng ngừa và ngăn chặn việc ai đó lợi dụng chức vụ, quan hệ trên dưới để quấy rối tình dục. Người bị hại có thể yêu cầu người quấy rối chịu trách nhiệm dân sự. Hoà giải lao động phải kết thúc trong 45 ngày kể từ ngày tiếp nhận, vụ phức tạp thì kéo dài thêm tối đa 15 ngày +- Mức bằng chứng: B +- Nguồn: Xu T 等 (2019). Workplace bullying and workplace violence as risk factors for cardiovascular disease: a multi-cohort study. European Heart Journal. ；Nielsen MB, Nielsen GH, Notelaers G, Einarsen S (2015). Workplace Bullying and Suicidal Ideation: A 3-Wave Longitudinal Norwegian Study. American Journal of Public Health. ；Ủy ban Thường vụ Quốc hội nhân dân (全国人大常委会) (2007). Bộ luật Hợp đồng lao động, 劳动合同法（第三十八条第二款、第四十六条、第八十八条）. ；Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零一十条）, bản toàn văn do Viện kiểm sát nhân dân tối cao đăng lại ；Ủy ban Thường vụ Quốc hội nhân dân (全国人大常委会) (2022, sửa đổi). Luật bảo vệ quyền lợi phụ nữ, 妇女权益保障法（第二十五条）. ；Ủy ban Thường vụ Quốc hội nhân dân (全国人大常委会) (2007). Luật hoà giải và trọng tài tranh chấp lao động, 劳动争议调解仲裁法（第四十三条）, bản toàn văn do Sở Nhân sự và Xã hội thành phố An Khang đăng lại +- Ghi chú: Lý do chấm B: các con số sức khoẻ đến từ theo dõi ghi chép, không phải thử nghiệm chia nhóm, và 「bị bắt nạt」 là do bản thân người bị hại tự khai. Có bốn bước làm được ngay. Bước một, từ hôm nay ghi chép lại, xin phép ghi âm xem mục 41 ở phần 8 (ghi âm). Bước hai, phản ánh bằng văn bản qua kênh chính thức của đơn vị, giữ lại bằng chứng đã phản ánh. Bước ba, bị đánh, bị nhốt, bị xúc phạm trước đông người thì báo công an. Bước bốn, bị ép nghỉ vì nợ lương, không đóng bảo hiểm xã hội thì làm theo mục 7 của phần này (đừng ký đơn tự nguyện nghỉ việc) và mục 8 của phần này (lưu chứng cứ trước khi nghỉ việc). Cách tính bồi thường kinh tế xem mục 4 của phần này (bị sa thải thì tính rõ N trước). Bị coi thường, bị loại trừ, bị gây khó đơn thuần thì phần lớn không khớp với điều luật nào, đường đi chủ yếu là khiếu nại nội bộ, xin đổi vị trí và đổi việc. Hoà giải lao động tính từ ngày tiếp nhận tối đa 60 ngày, ra tòa phân thủ cấp một tối thiểu 6 tháng, phí luật sư tự bỏ ra. Khi đã không chịu nổi nữa, gọi 12356 trước, xem mục 25 ở phần 1 (12356). Không phản đối việc chỉ vì khó chịu là nghỉ việc, cái giá của việc đổi việc thì bạn tự tính. diff --git "a/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" "b/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" index 2f206001..c36645f5 100644 --- "a/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" +++ "b/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 20. 刚出生的孩子怎么带 +# 20. Chăm trẻ mới sinh -口径：婴儿死亡率与金钱。这一节只收关系到孩子能不能活下来、或者你要花多少钱的事，而且只收差别足够大、又有硬证据的。怎么哄、怎么带这些育儿细节，本书不写。出生之前和出院当天要办的事，见第 27 节。 +Cột đo của phần này là tỷ lệ tử vong trẻ sơ sinh và tiền. Phần này chỉ nhận những mục liên quan đến việc đứa trẻ có sống sót hay không, hoặc đến việc có bị gì về sau hay không. Chỗ nào có nhắc tiền chỉ đưa cách tính, không đưa kết luận. ### 1. 让宝宝仰着睡，睡硬的平面，同房不同床，床上不放任何软东西 diff --git "a/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" "b/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" index bc9c5e59..c9e67ca6 100644 --- "a/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" +++ "b/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 21. 出国、旅行与境外安全 +# 21. Đi nước ngoài và an toàn ở nước ngoài -这一节算的是钱和人身自由，有几条还关系到人身安全。到了国外，管你的是所在国的法律，不是中国的。中国使领馆能帮你到哪一步，有明确的界线。 +Cột đo của phần này là tiền và tự do. Các mục xếp theo hiệu quả chi phí từ cao xuống thấp: mua bảo hiểm đúng chỗ, chọn chuyến bay và khách sạn, xử lý việc xảy ra ở nước ngoài, kể cả thị thực. ### 1. Trước khi đi, tra cảnh báo an toàn của Bộ Ngoại giao một lần, chỗ nào ghi 「暂勿前往」thì đừng đến diff --git "a/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" "b/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" index e0f7303f..9e64f88a 100644 --- "a/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" +++ "b/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" @@ -1,8 +1,8 @@ [← 回总目录](../README.md) -# 22. 怎么放松：娱乐场所和减压 +# 22. Thả lỏng: nơi chốn và cách giảm căng thẳng -前半节讲娱乐场所里哪些钱花得冤、安全出口在哪，口径是金钱和人身自由。后半节讲怎么减压，口径是精力和总死亡率。两半的数字不放在一起换算。第 3 节已经写过情绪低落时先做哪几件事，这里给更具体的做法。 +Cột đo của phần này là sức khoẻ và thời gian. Các mục xếp theo hiệu quả chi phí từ cao xuống thấp, và mỗi mục nói một hoạt động giải trí cụ thể: ở đâu, tốn bao nhiêu tiền, tốn bao nhiêu thời gian, và bằng chứng đằng sau là gì. ### 1. Vào KTV, quán bar, phòng giải đố trước khi vào hãy nhìn lối thoát hiểm ở đâu, bị khóa hoặc bị chặn thì đổi chỗ diff --git "a/vi/book/24-\347\234\213\347\227\205.md" "b/vi/book/24-\347\234\213\347\227\205.md" new file mode 100644 index 00000000..f84a89e1 --- /dev/null +++ "b/vi/book/24-\347\234\213\347\227\205.md" @@ -0,0 +1,113 @@ +[← Về mục lục](../README.md) + +# 24. Khám bệnh: làm sao để ít tốn tiền và đỡ đi vòng + +Thước đo: tiền và thời gian. Phần này nói cách đi khám: vào bệnh viện nào trước, muốn lên tuyến trên thì làm thế nào, tiền có thể bớt chi ở đâu. Phần này còn nói cấp cứu xếp hàng thế nào, sau khi chữa chấn thương nặng cần làm những phép giám định nào, xin những giấy tờ nào. Xem phần 7 về những khoản trợ cấp có thể nhận. Xem phần 16 về cách quản lý bệnh mạn tính lâu dài. Xem phần 13 về việc đầu tiên phải làm ngay tại hiện trường. + +Phần này chủ yếu dựa vào《Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng》(关于加快建设分级诊疗体系的若干措施), do Văn phòng Hội đồng Nhà nước (国务院办公厅) ban hành tháng 4 năm 2026. Chẩn đoán và điều trị phân tầng nghĩa là căn cứ bệnh nặng hay nhẹ để quyết định vào bệnh viện cấp nào trước. Văn bản này gắn thẳng「đi tuyến cơ sở」với「tiết kiệm tiền」. Nhưng tiến độ thực hiện ở mỗi nơi khác nhau, phần chữ trong sách đưa ra quy định ở tầm quốc gia. Riêng với vùng gộp quỹ của bạn, cứ theo cách nói của cơ quan bảo hiểm y tế và cơ quan y tế địa phương. Vùng gộp quỹ là nơi bạn tham gia bảo hiểm y tế, thường là một thành phố. + +### 1. Bệnh thường ngày xem ở tuyến cơ sở trước, chuyển tuyến lên từng bậc, mức đồng chi trả nội trú vẫn tính tiếp + +- Chi phí: Không tốn tiền. Phải đến tuyến cơ sở một lần trước. Chỗ khó là kiềm chế không đi lấy số khám chuyên gia ngay. +- Nói thẳng: Đi tuyến cơ sở trước, rồi từ đó chuyển lên bệnh viện lớn để nằm viện, phần tiền bạn tự trả chỉ phải ra một lần. Tự đi thẳng bệnh viện cấp ba nằm viện rồi sau đó chuyển về tuyến cơ sở tập luyện phục hồi, phần đó có thể phải trả hai lần. Được bệnh viện lớn chuyển về tuyến cơ sở nằm tiếp thì cùng một bệnh trong suốt khoảng thời gian ấy cũng không tính lại lần nữa. +- Lợi ích: Bệnh nhân có bảo hiểm y tế được chuyển tuyến từng bậc qua tuyến cơ sở thì mức đồng chi trả nội trú ở bệnh viện tuyến trên được tính liên tục. Mức đồng chi trả là phần mỗi lần nẩm viện bạn tự trả trước cho đến khi bảo hiểm y tế mới bắt đầu hoàn. Tính liên tục nghĩa là hai bệnh viện trước sau gộp lại chỉ tính một lần, không phải trả lại từ đầu. Bệnh nhân nằm viện được bệnh viện tuyến trên chuyển ngược xuống tuyến cơ sở thì trong cùng một chu kỳ bệnh không còn đặt mức đồng chi trả riêng nữa +- Mức bằng chứng: A +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「统筹地区内经基层医疗卫生机构逐级转诊的参保患者，在上级医院的住院起付线可连续计算；由上级医院下转至基层医疗卫生机构的住院患者，同一疾病周期内不再另设住院起付线。」 +- Ghi chú: Tự đi thẳng bệnh viện cấp ba nằm viện rồi chuyển về tuyến cơ sở tập luyện phục hồi, phần tiền đó có thể phải trả hai lần, đi theo chuyển tuyến thì chỉ tính một lần. Quy định chuyển tuyến do từng tỉnh, từng thành phố và từng liên kết y tế tự đặt ra, trước khi đi nên hỏi tuyến cơ sở trước. + +### 2. Trong cùng một vùng gộp quỹ (nơi tham gia bảo hiểm y tế), bệnh viện cấp càng thấp thì tỷ lệ hoàn càng cao, chênh nhau khoảng 10 điểm phần trăm + +- Chi phí: Không tốn tiền. Chỗ khó là kiềm chế không chạy lên bệnh viện lớn. +- Nói thẳng: Cùng một khoản viện phí, bệnh viện cấp càng thấp thì được hoàn càng nhiều. Các nơi phải kéo giãn khoảng cách này ra, mỗi bậc lên cao hơn chênh nhau khoảng mười điểm phần trăm. Việc có thể xong ở tuyến cơ sở mà chạy lên bệnh viện lớn thì cùng khoản chi phí đó bạn phải tự trả thêm khoảng một phần mười. Phần quỹ bảo hiểm y tế tăng thêm trong năm cũng nghiêng về tuyến cơ sở. +- Lợi ích: Văn bản yêu cầu các nơi kéo giãn mức hoàn phí nội trú giữa các cơ sở y tế khác cấp, về nguyên tắc mỗi bậc chênh nhau khoảng 10 điểm phần trăm. Với khoảng cách đó, cùng một khoản viện phí mỗi bậc bệnh viện lên cao hơn, bạn phải tự trả thêm khoảng một phần mười. Quỹ bảo hiểm y tế tăng thêm trong năm cũng nghiêng về tuyến cơ sở +- Mức bằng chứng: A +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「因地制宜适当拉开参保人员在不同等级医疗卫生机构的住院报销水平，原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距。」「具备条件的地方，可按分级诊疗导向酌情拉开参保人员在不同等级医疗卫生机构的门诊报销水平。」 +- Ghi chú: Cùng một khoản viện phí, ở tuyến cơ sở và ở bệnh viện cấp ba, phần bạn tự trả có thể chênh nhau vài nghìn tệ. Bệnh nào xong được ở tuyến cơ sở thì đừng chạy lên, và không phải bệnh nào cũng cần xem ở tuyến cơ sở. + +### 3. Muốn đi bệnh viện lớn, đi theo chuyển tuyến từ tuyến cơ sở hoặc trung tâm chuyển tuyến của bệnh viện, đừng tìm kẻ bán số khám + +- Chi phí: Không tốn tiền. Phải làm thêm một thủ tục chuyển tuyến. +- Nói thẳng: Bệnh viện lớn phải dành ra một phần số khám và giường bệnh, dành riêng cho người được chuyển lên từ tuyến cơ sở. Bệnh viện còn phải lập trung tâm chuyển tuyến, hoặc chỉ định một bộ phận cố định phụ trách việc này, đến năm 2027 thì phủ hết. Không lấy được số khám không có nghĩa là không còn đường. Số khám mua của kẻ bán số khám ngoài chuyện đắt, còn có thể đưa bạn vào khoa không đúng chuyên. +- Lợi ích: Bệnh viện dẫn dắt trong liên kết y tế kiểu chặt phải dành cho tuyến cơ sở một tỷ lệ số khám và giường bệnh nhất định, và tiếp khám kịp thời cho bệnh nhân được chuyển tuyến từ tuyến cơ sở. Liên kết y tế kiểu chặt là nhóm cơ sở gồm vài bệnh viện và cơ sở y tế cộng đồng buộc lại với nhau để hợp tác. Nên khi chuyển lên từ tuyến cộng đồng sẽ có số khám và giường riêng, không chen chúc với người giành số khám ngoài cửa. Cơ sở y tế phải lập trung tâm chuyển tuyến hoặc chỉ định một bộ phận cố định, đến năm 2027 thì phủ hết +- Mức bằng chứng: A +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「牵头医院要为基层医疗卫生机构预留一定比例的号源和床位，并及时接诊经基层转诊患者。」「医疗机构要强化转诊服务统一管理，设立转诊中心或指定固定部门承担患者转诊服务工作，到2027年实现全覆盖。」 +- Ghi chú: Số khám dành riêng là dành cho bệnh nhân chuyển tuyến, không giành được số khám thì đi theo chuyển tuyến. Tìm kẻ bán số khám ngoài chuyện đắt, còn có thể khiến bạn rơi vào khoa không đúng chuyên. Không tìm thấy trung tâm chuyển tuyến thì gọi đường dây dịch vụ của bệnh viện hỏi. + +### 4. Trước khi đi khám chữa bệnh ở tỉnh khác, hỏi một câu ở nơi bạn đang sống, tính cần thiết về nguyên tắc phải có bác sĩ phó trưởng khoa đánh giá + +- Chi phí: Không tốn tiền. Hỏi thêm một câu trước khi lên đường. +- Nói thẳng: Có đi khám chữa bệnh ở nơi khác hay không, về nguyên tắc do bác sĩ phó trưởng khoa trở lên ở bệnh viện cấp hai, cấp ba đánh giá. Mức hoàn khi đi khám chữa bệnh tạm thời ở nơi khác khác với bệnh viện cùng cấp tại nơi tham gia bảo hiểm, cụ thể theo quy định địa phương. Hỏi một câu ở nơi bạn đang sống trước khi lên đường còn hơn đến nơi khác mới biết không hoàn được. +- Lợi ích: Tính cần thiết khi khám chữa bệnh vượt vùng gộp quỹ, ra khỏi tỉnh, về nguyên tắc do người có chức danh phó trưởng khoa trở lên ở bệnh viện cấp hai, cấp ba đánh giá. Vượt vùng gộp quỹ nghĩa là ra khỏi nơi bạn tham gia bảo hiểm y tế. Có đi nơi khác khám hay không phải do bác sĩ đủ bậc quyết định. Mức hoàn của người tạm sang tỉnh khác khám chữa bệnh giữ mức chênh lệch hợp lý với cơ sở cùng cấp tại nơi tham gia bảo hiểm. Khám cùng một bệnh ở nơi khác thì số tiền hoàn về ít hơn một đoạn +- Mức bằng chứng: A +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「原则上由二、三级医院副主任医师及以上职称人员评估患者跨统筹地区、跨省异地就医的必要性。」「进一步强化异地就医直接结算管理服务，跨省临时外出就医人员报销水平与参保地同级别医疗机构报销水平保持合理差异。」 +- Ghi chú: 「临时外出就医」 và「长期异地居住」 là hai kiểu khai báo khác nhau, mức hoàn cũng khác nhau. Cách khai báo thế nào, những loại bệnh nào được thanh toán trực tiếp, xem phần 16. Đừng chạy tới nơi khác rồi mới nghĩ tới chuyện hoàn. + +### 5. Đừng còn lấy bệnh viện cấp ba làm chỗ lấy thuốc, bệnh mạn tính ở thời kỳ ổn định thì quay về tuyến cơ sở + +- Chi phí: Không tốn tiền. Chỗ khó là bỏ được thói quen「chỉ tin bệnh viện lớn」. +- Nói thẳng: Bệnh viện cấp ba về sau chủ yếu lo bệnh cấp nguy kịch và bệnh khó chẩn đoán. Khám lại bệnh thường ngày, và khám ngoại trú thường cho bệnh mạn tính đã ổn định, sẽ bị giảm dần. Khám ngoại trú thường của cùng một đội chuyên gia sẽ chuyển xuống mở ở tuyến cộng đồng. Còn lấy bệnh viện lớn làm chỗ lấy thuốc thì sau này càng khó lấy số khám. +- Lợi ích: Bệnh viện cấp ba phải tập trung vào bệnh cấp nguy kịch và bệnh khó, phức tạp, giảm dần khám ngoại trú thường cho bệnh thường ngày tái khám và bệnh mạn tính đã chẩn đoán rõ, ổn định. Số khám của loại khám ngoại trú thường ấy sẽ bị thu dần tuỳ tình hình. Bệnh viện tuyến trên sẽ kéo dài khám ngoại trú thường của đội chuyên gia xuống tuyến cơ sở, khám ngoại trú thường của cùng một đội chuyên gia sẽ mở ở tuyến cộng đồng +- Mức bằng chứng: A +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「三级医院要聚焦急危重症和疑难复杂疾病，加强转诊会诊和住院服务，逐步酌减常见病复诊和诊断明确、病情稳定的慢性病等普通门诊。」「紧密型医联体内上级医院要在基层医疗卫生机构开设高血压、糖尿病、慢性阻塞性肺疾病等常见病、慢性病门诊，提供诊疗和健康管理服务，并将专家团队普通门诊向基层医疗卫生机构延伸。」 +- Ghi chú: Đây là hướng chính sách, không phải lời khuyên. Sau này khám ngoại trú thường ở bệnh viện lớn sẽ càng khó lấy số. Bệnh mạn tính ở thời kỳ ổn định, ở tuyến cơ sở có thể kê thuốc một lần đủ 12 tuần, xem phần 16. + +### 6. Mỗi lần khám xong, tự giữ lại một bộ hồ sơ bệnh án, kết quả xét nghiệm và ảnh + +- Chi phí: Không tốn tiền. Bệnh viện được thu phí sao chép theo quy định, mức phí phải công khai. Chỗ khó là nhớ đi sao chép sau mỗi lần khám. +- Nói thẳng: Quyền tra cứu và sao chép hồ sơ bệnh án của chính bạn là quyền mà luật trao cho bạn. Hồ sơ khám ngoại trú, sổ nhập viện, phiếu xét nghiệm, ảnh, hồ sơ phẫu thuật và gây mê, tư liệu bệnh lý, bảng chi phí đều sao chép được. Bệnh viện có một cửa sổ riêng tiếp nhận, chỉ được thu phí sao chép. Đổi bệnh viện, tìm bác sĩ khác xem lại, xin bảo hiểm thương mại, khi thật sự có tranh chấp, đều dựa vào bộ hồ sơ này. +- Lợi ích: Quy định nêu rõ quyền tra cứu, sao chép hồ sơ bệnh án là quyền của bệnh nhân. Phạm vi gồm hồ sơ khám ngoại trú, sổ nhập viện, phiếu xét nghiệm, ảnh y khoa, hồ sơ phẫu thuật và gây mê, tư liệu bệnh lý, chi phí khám chữa bệnh và toàn bộ tài liệu thuộc hồ sơ bệnh án. Đổi bệnh viện, tìm bác sĩ thứ hai xem lại, xin bảo hiểm thương mại, khi thật sự có tranh chấp, đều dựa vào những tài liệu này +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2018). Quy chế phòng ngừa và xử lý tranh chấp y tế（国务院令第 701 号）Điều mười sáu: 「患者有权查阅、复制其门诊病历、住院志、体温单、医嘱单、化验单（检验报告）、医学影像检查资料、特殊检查同意书、手术同意书、手术及麻醉记录、病理资料、护理记录、医疗费用以及国务院卫生主管部门规定的其他属于病历的全部资料。」「患者要求复制病历资料的，医疗机构应当提供复制服务，并在复制的病历资料上加盖证明印记……可以收取工本费，收费标准应当公开。」「患者死亡的，其近亲属可以依照本条例的规定，查阅、复制病历资料。」Điều mười lăm「任何单位和个人不得篡改、伪造、隐匿、毁灭或者抢夺病历资料。」；Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委), Cục Quản lý Y học cổ truyền quốc gia (国家中医药管理局) (2013). Quy định quản lý hồ sơ bệnh án của cơ sở y tế (bản 2013): cơ sở y tế「应当指定部门或者专（兼）职人员负责受理复制病历资料的申请」, việc sao chép thực hiện khi người xin có mặt và có đóng dấu chứng thực. +- Ghi chú: Sao chép có cửa sổ riêng, không cần đi xin bác sĩ. Danh sách cần sao chép thì cứ theo Điều mười sáu của《Quy chế phòng ngừa và xử lý tranh chấp y tế》đọc. Trước là hồ sơ khám ngoại trú, sổ nhập viện, bảng nhiệt độ, phiếu y lệnh, phiếu xét nghiệm, tư liệu ảnh. Sau là hồ sơ phẫu thuật và gây mê, tư liệu bệnh lý, hồ sơ chăm sóc, bảng chi phí khám chữa bệnh. Lấy xong xếp vào một thư mục theo thứ tự thời gian. Trong đó quan trọng nhất là tóm tắt lúc ra viện, biên bản phẫu thuật, kết quả giải phẫu bệnh và ảnh then chốt. Chỉ số qua các lần của bệnh mạn tính thì ghi vào cùng một quyển sổ, xem phần 16. + +### 7. Có nghi ngờ về chẩn đoán và điều trị thì yêu cầu niêm phong hồ sơ ngay tại chỗ, hai bên cùng có mặt, lập biên bản, mỗi bên giữ một bản + +- Chi phí: Không tốn tiền. Nêu yêu cầu niêm phong với bệnh viện, rồi chờ cả hai bên đến làm thủ tục. Chỗ khó là nói được câu đó trước mặt bác sĩ. +- Nói thẳng: Có nghi ngờ về cách điều trị thì có thể yêu cầu niêm phong hồ sơ bệnh án. Việc niêm phong phải có cả bệnh nhân và cơ sở y tế cùng có mặt, lập biên bản, ký và đóng dấu, mỗi bên giữ một bản. Hồ sơ chưa viết xong thì niêm phong trước phần đã viết xong. Nghi ngờ do truyền dịch, truyền máu, tiêm, dùng thuốc thì túi dịch, ống thuốc và bộ truyền dịch cùng phải được niêm phong. +- Lợi ích: Quy chế quy định「发生医疗纠纷需要封存、启封病历资料的，应当在医患双方在场的情况下进行」. Phần được niêm phong có thể là bản gốc, cũng có thể là bản sao, do cơ sở y tế giữ. Hồ sơ chưa hoàn thành mà cần niêm phong thì niêm phong phần đã hoàn thành trước, phần làm sau thì niêm phong tiếp. Cơ sở y tế phải lập biên bản danh mục hồ sơ đã niêm phong, để hai bên ký hoặc đóng dấu, mỗi bên giữ một bản. Nghi ngờ truyền dịch, truyền máu, tiêm, dùng thuốc gây hậu quả xấu thì hai bên phải cùng niêm phong, mở niêm phong vật thể tại hiện trường. Cần xét nghiệm thì hai bên cùng giao cho một cơ sở có đủ điều kiện xét nghiệm. Không giao được thì cơ quan quản lý y tế của chính quyền cấp huyện nơi cơ sở y tế đang đặt trụ sở chỉ định. Nghi ngờ truyền máu gây hậu quả xấu mà cần niêm phong máu thì cơ sở y tế phải báo trạm máu cử người đến. Còn một ràng buộc cứng:「任何单位和个人不得篡改、伪造、隐匿、毁灭或者抢夺病历资料。」Sau khi niêm phong mà tranh chấp đã giải quyết, hoặc đã quá 3 năm không ai yêu cầu giải quyết nữa, cơ sở y tế có thể tự mở niêm phong (toàn quốc, áp dụng từ ngày 1 tháng 10 năm 2018) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2018). Quy chế phòng ngừa và xử lý tranh chấp y tế（国务院令第 701 号，第十五、二十四、二十五条）. +- Ghi chú: Niêm phong và sao chép là hai việc khác nhau. Sao chép là việc nên làm khi bình thường, xem mục 6 (tự giữ hồ sơ và ảnh). Niêm phong là khi xảy ra tranh chấp thì ghim bằng chứng lại, để chống việc sau đó bổ sung viết hoặc sửa. Nghi ngờ do truyền dịch, truyền máu, dùng thuốc thì ngoài hồ sơ còn phải niêm phong cả túi dịch, ống thuốc và bộ truyền dịch, đừng để người ta dọn đi mà vứt bỏ. Cơ sở y tế phải công bố ở vị trí dễ thấy các đường giải quyết tranh chấp y tế, thủ tục và cách liên hệ, và lập một bộ phận tiếp nhận khiếu nại thống nhất. Có khiếu nại thì tìm bộ phận đó, đừng đi lý lẽ với bác sĩ trực. +### 8. Chấn thương nặng thì đi thẳng bàn tiền phân loại của khoa cấp cứu, đừng xếp hàng ở cửa sổ lấy số khám + +- Chi phí: Không tốn tiền. Vào cửa đi thêm vài bước để tìm bàn tiền phân loại. Nói rõ bằng một câu triệu chứng nguy kịch nhất. +- Nói thẳng: Cấp cứu không xếp theo thứ tự đến trước, mà xếp theo bệnh nặng nhẹ thành bốn bậc. Chỗ xếp bậc là bàn tiền phân loại ở cửa vào khoa cấp cứu, không phải cửa sổ lấy số khám. Không qua bàn phân loại thì trong hệ thống của bạn không có bậc nào, chấn thương nặng đến mấy cũng chỉ đi theo hàng chậm. Đến nơi hãy báo cáo trước triệu chứng nguy hiểm nhất: chảy máu nhiều, thở không ra, ý thức không tỉnh, trên người có vật đâm, đè nặng ở ngực. +- Lợi ích: Chỉ tiêu chất lượng cấp cứu bản 2024 của Ủy ban Quốc gia Y tế (国家卫生健康委) quy định tiền phân loại ở khoa cấp cứu phải「按病情危重程度将患者分为四级」。「I 级为急危患者，需要立即得到救治」。「II 级为急重患者，往往评估与救治同时进行」。「III 级为急症患者，需要短时间内得到诊治」。「IV 级为非急症患者或亚急症患者，在普通诊疗区顺序就诊」. Tiền phân loại nghĩa là vào khoa cấp cứu thì có người chuyên trách đánh giá bệnh nặng nhẹ trước, rồi chốt thứ tự khám. Quy chuẩn thi hành của Bắc Kinh nêu yêu cầu đáp ứng của từng bậc. Cấp 1「立即进行评估和救治，安排患者进入复苏室或抢救室」. Cấp 2「立即监护生命体征，10 分钟内得到救治，安排患者进入抢救室」. Cấp 3「先于 4 级患者优先诊治，安排患者在普通诊疗区候诊；若候诊时间大于 30 分钟，需再次评估」. Cấp 4「顺序就诊，除非病情变化，否则候诊时间较长；若候诊时间大于 4 小时，可再次评估」. Khi phân loại phải「坚持急危重症优先就诊原则」, ngay cửa vào khoa cấp cứu phải có khu vực tiền phân loại ở vị trí dễ thấy (bốn bậc là chuẩn toàn quốc, thời gian đáp ứng là chuẩn của Bắc Kinh) +- Mức bằng chứng: A +- Nguồn: Văn phòng Ủy ban Quốc gia Y tế (国家卫生健康委办公厅) (2024). Thông báo in chỉ tiêu chất lượng khám chữa bệnh cho 6 chuyên khoa trong đó có cấp cứu (bản 2024)（国卫办医政函〔2024〕150 号）, phụ lục 1, chỉ số bốn「急诊分级分诊执行率」. ；Ủy ban Y tế và Sức khoẻ thành phố Bắc Kinh (北京市卫生健康委员会) (2019). Phương án Bắc Kinh tăng cường tiền phân loại phân bậc ở khoa cấp cứu và phụ lục《Chuẩn phân loại phân bậc tiền kiểm ở khoa cấp cứu của các bệnh viện Bắc Kinh (thử nghiệm)》: 「坚持急危重症优先就诊原则。按照患者病情的严重程度，遵循从重到轻、从病情迅速变化到相对稳定的原则，合理安排患者就诊顺序。」「在急诊科入口的显著位置，设置急诊预检分诊区域」. +- Ghi chú: Chia bốn bậc là thống nhất toàn quốc. Mỗi bậc phải chờ bao lâu thì địa phương tự quy định, bản của Bắc Kinh là bản có thể đối chiếu từng chữ. Xếp bậc dựa vào mạch, huyết áp, độ bão hòa oxy và triệu chứng, không dựa vào cảm giác đau của bạn. Chỉ tiêu cứng của bậc 1 có mấy điều. Một là nhịp tim lớn hơn 180 lần/phút, hoặc nhỏ hơn 40 lần/phút. Hai là huyết áp tâm thu thấp hơn 70 mmHg, tâm thu là chỉ số huyết áp cao. Ba là độ bão hòa oxy dưới 80%. Bốn là sốc, nhồi máu cơ tim xác định, rối loạn ý thức cấp. Thấy bàn phân loại xếp bạn thấp hơn thể nào thì bổ sung, nói rõ triệu chứng nguy hiểm nhất. Trong lúc chờ mà bệnh thay đổi thì lập tức quay lại bàn phân loại đánh giá lại. Khi thật sự nguy kịch thì gọi 120 còn nhanh hơn tự đến bệnh viện, bác sĩ ngoài bệnh viện sẽ xếp một bậc theo cùng bộ tiêu chí và báo trước bệnh viện để chuẩn bị tiếp đón (xem phần 13). + +### 9. Không có tiền, không có giấy tờ, không nói rõ mình là ai, cấp cứu vẫn phải cứu trước + +- Chi phí: Không tốn tiền. +- Nói thẳng: Nhà nước có một quỹ cứu hộ khẩn cấp khi bệnh, dùng riêng để trả chi phí cấp cứu cho hai loại người: người không xác định được danh tính, và người không đủ tiền trả. Bệnh viện không được từ chối hay trì hoãn cấp cứu, trạm cấp cứu cũng không được không xuất xe vì tiền. Vậy nên đừng ở nhà chịu đựng vì không có tiền, cũng đừng sợ nợ tiền mà không gọi xe cứu thương. Khoản tiền này chỉ lo phần cấp cứu, vào viện sau vẫn theo bảo hiểm y tế và trợ cấp y tế. +- Lợi ích: Đối tượng được trợ cấp của chế độ cứu hộ khẩn cấp khi bệnh là「在中国境内发生急重危伤病、需要急救但身份不明确或无力支付相应费用的患者」. Quỹ trả hai loại chi phí: 「无法查明身份患者所发生的急救费用」 và「身份明确但无力缴费的患者所拖欠的急救费用」. Văn bản còn ghi「各级各类医疗机构及其工作人员必须及时、有效地对急重危伤患者施救，不得以任何理由拒绝、推诿或拖延救治」. Cấp cứu ngoài bệnh viện cũng có quy định: số gọi cấp cứu y tế ngoài bệnh viện toàn quốc là「120」. Trạm cấp cứu (trung tâm)「不得因指挥调度原因拒绝、推诿或者延误院前医疗急救服务」, và cũng「不得因费用问题拒绝或者延误院前医疗急救服务」. Chuyển bệnh nhân tuân theo「就近、就急、满足专业需要、兼顾患者意愿的原则」(toàn quốc, từ năm 2013) +- Mức bằng chứng: A +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2013). Ý kiến chỉ đạo về việc xây dựng chế độ cứu hộ khẩn cấp khi bệnh（国办发〔2013〕15 号）. ；Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生和计划生育委员会) (2014). Biện pháp quản lý cấp cứu y tế ngoài bệnh viện（委令第 3 号，第十三条、第二十二条、第二十三条、第二十五条）. +- Ghi chú: Quỹ chỉ trả phần「急救费用」, lại là cứu trước rồi sau đó cơ sở y tế mới xin quỹ, bạn không phải ra cửa sổ nhận tiền. Có bảo hiểm y tế thì vẫn dùng bảo hiểm y tế trước như bình thường. Thủ tục xin và hạn mức ở mỗi tỉnh khác nhau. Trong nguyên tắc chuyển bệnh,「就近、就急」đứng trước「患者意愿」, nên xe cứu thương không nhất thiết đưa bạn tới bệnh viện bạn chỉ định, lúc nguy kịch thì lấy bệnh viện gần nhất xử lý được là chính. Phần bạn tự trả về sau mà vẫn quá nặng thì làm trợ cấp y tế, xem mục 10 ở phần 7 (trợ cấp y tế). + +### 10. Giám định thương tật phải đợi điều trị kết thúc rồi hãy làm, làm sớm thì cấp độ bị chấp thấp + +- Chi phí: Phí giám định vài trăm đến hai nghìn tệ, thường do bên xin đơn trả trước. Phải đợi điều trị kết thúc, hoặc hiệu quả lâm sàng ổn định, thường là ba đến sáu tháng sau khi bị thương, có khi lâu hơn. Chỗ khó là lúc thương tích chưa lành đã muốn chốt sớm, phải kiềm chế. +- Nói thẳng: Đền bao nhiêu phần lớn tùy cấp độ được xếp, còn cấp độ thì tùy còn lại bao nhiêu tổn hương chức năng sau khi điều trị. Tiêu chuẩn ghi rõ phải làm sau khi điều trị kết thúc, hoặc khi hiệu quả điều trị lâm sàng ổn định. Xương chưa lành, đinh chưa lấy ra mà đi giám định thì người ta đánh giá đúng cái trạng thái lúc đó, tức là bạn tự ép cấp độ của mình xuống. Gãy xương thường đã lành hẳn mà không ảnh hưởng chức năng thì không được xếp cấp. +- Lợi ích: 《Phân cấp mức độ tàn tật do tổn thương cơ thể》quy định thời điểm giám định「应在原发性损伤及其与之确有关联的并发症治疗终结或者临床治疗效果稳定后进行鉴定」. Tiêu chuẩn「将人体损伤致残程度划分为 10 个等级，从一级（人体致残率 100%）到十级（人体致残率 10%），每级致残率相差 10%」. Căn cứ xếp là「人体组织器官结构破坏、功能障碍及其对医疗、护理的依赖程度，适当考虑由于残疾引起的社会交往和心理因素影响」. Trong mười cấp, các khoản liên quan tới gãy xương gồm mấy điều. Một là「一椎体压缩性骨折（压缩程度达 1/3）或者粉碎性骨折」. Hai là「骨盆两处以上骨折或者粉碎性骨折，畸形愈合」. Ba là「四肢任一大关节（踝关节除外）功能丧失 25% 以上」. Cùng một vị trí và cùng một tính chất tàn tật thì không tính trùng. Bị tàn tật từ hai chỗ trở lên thì phải ghi rõ từng cấp (toàn quốc, áp dụng từ ngày 1 tháng 1 năm 2017) +- Mức bằng chứng: A +- Nguồn: Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao, Bộ Công an, Bộ An ninh quốc gia, Bộ Tư pháp (最高人民法院、最高人民检察院、公安部、国家安全部、司法部) (2016). Thông báo công bố《Phân cấp mức độ tàn tật do tổn thương cơ thể》(ban hành ngày 18 tháng 4 năm 2016, áp dụng từ ngày 1 tháng 1 năm 2017), điều 4.2, 4.4, 4.5, 5.10.6, 6.2 của tiêu chuẩn. +- Ghi chú: Ba bộ xếp tàn tật không dùng chung, đừng lẫn. Bị đánh, tai nạn giao thông và các tổn thương cá nhân khác thì dùng《Phân cấp mức độ tàn tật do tổn thương cơ thể》của giám định tư pháp, để tính tiền bồi thường tàn tật (xem phần 9). Tai nạn lao động thì làm khảo sát năng lực lao động, dùng bộ mười cấp tàn tật khác, và bộ đó quy định quyền lợi tai nạn lao động (xem phần 19). Muốn hưởng chính sách cho người khuyết tật thì phải xin riêng giấy chứng nhận người khuyết tật (xem mục 11, giấy chứng nhận người khuyết tật). Trước khi giám định, chuẩn bị đủ hồ sơ bệnh án, biên bản phẫu thuật, ảnh chụp lại (xem mục 6, tự giữ hồ sơ và ảnh), thiếu ảnh là nguyên nhân bị chấp cấp độ thấp nhất. Khi lần thương này lẫn với bệnh mạn tính sẵn có thì ý kiến giám định còn phải ghi rõ lần thương này chiếm bao nhiêu phần trong hậu quả tàn tật. Ô mục đó ảnh hưởng trực tiếp tới số tiền đền. + +### 11. Sau khi điều trị xác nhận còn tổn hương chức năng, đến hiệp hội người khuyết tật cấp huyện nơi đăng ký hộ khẩu xin giấy chứng nhận người khuyết tật + +- Chi phí: Bản giấy không thu phí sao chép. Phí đánh giá và phí ảnh về nguyên tắc do người xin tự trả, khó khăn thì xin giảm trừ. Mang theo giấy tờ tùy thân, sổ hộ khẩu và 3 tấm ảnh thẻ nền trắng cỡ hai inch đến hiệp hội người khuyết tật cấp huyện nơi đăng ký hộ khẩu. Kết quả đạt thì công bố tại thôn (cộng đồng) trong 5 ngày làm việc, hiệp hội xét xong trong 10 ngày làm việc. Chỗ khó là phải đi mấy chuyến và chờ khoảng thời gian ấy. +- Nói thẳng: Giấy chứng nhận người khuyết tật là ngưỡng để nhận quyền lợi của người khuyết tật, không phải giấy chứng minh mức độ thương tích. Chia thành bảy nhóm: thị lực, thính lực, ngôn ngữ, vận động, trí tuệ, tâm thần và đa dạng, mỗi nhóm từ cấp một đến cấp bốn. Quy trình là xin, tiếp nhận, đánh giá ở cơ sở được chỉ định, công bố, phê duyệt, cấp giấy, thời hạn mười năm, hết hạn đổi miễn phí. Không đạt thì là không đạt, đây không phải việc có thể nhờ người quen mà xin được. +- Lợi ích: Giấy chứng nhận người khuyết tật「是认定残疾人及其残疾类别、残疾等级的合法凭证，是残疾人依法享有国家和地方政府优惠政策的重要依据」. Tiêu chuẩn đánh giá là tiêu chuẩn quốc gia《Phân loại và phân cấp tàn tật》(GB/T 26341-2010). Làm giấy giữ nguyên tắc tự nguyện xin và quản lý theo địa phương. Cả bảy nhóm tàn tật về thị lực, thính lực, ngôn ngữ, vận động, trí tuệ, tâm thần và đa dạng đều có thể xin, cấp độ tàn tật từ một đến bốn. Hiệp hội người khuyết tật cấp huyện phụ trách tiếp nhận và cấp giấy, căn cứ kết luận đánh giá của bệnh viện hoặc cơ sở chuyên môn do tỉnh chỉ định. Kết luận đánh giá phù hợp tiêu chuẩn thì công bố tại thôn (cộng đồng) trong năm ngày làm việc, hiệp hội người khuyết tật cấp huyện「在十个工作日内审核完毕」.「办理残疾人证不收取工本费」. Phí đánh giá và phí ảnh「原则上由申请人个人自理；有条件的地方可由当地财政予以补贴，对特殊困难的申请人应协调有关部门予以减免」. Thời hạn của giấy là mười năm, hết hạn đổi miễn phí. Loại tàn tật hoặc cấp độ thay đổi thì có thể xin đánh giá lại. Có ý kiến khác về kết luận đánh giá thì trong mười ngày làm việc đến hiệp hội người khuyết tật cấp thành phố xin đánh giá lại. Vẫn còn ý kiến khác thì Ủy ban chuyên gia đánh giá tàn tật cấp tỉnh đưa ra kết luận cuối cùng (toàn quốc, áp dụng từ ngày 1 tháng 1 năm 2018) +- Mức bằng chứng: A +- Nguồn: Hiệp hội Người khuyết tật Trung Quốc (中国残疾人联合会), Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生和计划生育委员会) (2017). Biện pháp quản lý giấy chứng nhận người khuyết tật nước Cộng hoà nhân dân Trung Quốc（中华人民共和国残疾人证管理办法）(ban hành ngày 27 tháng 5 năm 2017, áp dụng từ ngày 1 tháng 1 năm 2018; Điều hai, ba, năm, bảy, chín, mười bảy, mười chín, hai mươi hai, hai mươi lăm). +- Ghi chú: Có làm hay không thì tính tiền trước đã. Bản giấy không trực tiếp phát tiền. Nó nối tới hai khoản trợ cấp cho người khuyết tật: người khuyết tật trong gia đình nhận trợ cấp thu nhập thấp nhận khoản trợ cấp sinh hoạt, người khuyết tật cấp một, cấp hai cần chăm sóc dài hạn nhận khoản trợ cấp chăm sóc, xem mục 8 ở phần 7 (hai khoản trợ cấp). Giấy còn nối tới các ưu đãi phục hồi chức năng, dụng cụ trợ giúp, thuế và phí, việc làm và giao thông ở các nơi. Tiêu chuẩn ở mỗi tỉnh khác nhau rất nhiều, hãy tra trên trang chính thức của hiệp hội người khuyết tật địa phương cho rõ rồi hãy quyết định có đi chuyến này không. Nó khác với cấp độ tàn tật của giám định tư pháp và với khảo sát năng lực lao động cho tai nạn lao động, đó là ba bộ riêng, không thay thế nhau (xem mục 10, giám định thương tật phải đợi điều trị kết thúc rồi hãy làm). Giấy chỉ dùng cho chính người đó, không cho mượn. Tình trạng tàn tật không còn phù hợp tiêu chuẩn thì bị thu hồi, sau khi thu hồi một năm không được xin lại. Người tàn tật đa dạng thì lấy loại nặng nhất trong các loại để xếp cấp. +### 12. Cảm ơn bác sĩ đã cứu bạn thì dùng thư cảm ơn, bằng khen và đánh giá mức hài lòng, đừng dùng bao lì xì: quy chuẩn cấm tài vật, không cấm lòng biết ơn + +- Chi phí: Không tốn tiền. Viết một lá thư cảm ơn, hoặc điền một lần trên nền tảng đánh giá mức hài lòng, mười mấy phút. +- Nói thẳng: Chặn tiền trước không mua được quyền ưu tiên, số khám, giường bệnh, thuốc khan hiếm và lịch phẫu thuật đều bị cấm dùng để đổi lấy lợi ích. Sau khi chữa xong mà muốn cảm ơn bác sĩ cứu mạng thì cảm ơn được, nhưng đừng tặng thuốc lá, rượu hay bao lì xì, bác sĩ mà nhận là vi phạm. Có ba cách cảm ơn thật sự rơi vào tay bác sĩ: điền nghiêm túc một lần đánh giá mức hài lòng, viết thư cảm ơn có tên hoặc tặng một bằng khen, và tái khám đúng lời bác sĩ dặn. +- Lợi ích: 《Chín nguyên tắc về liêm chính khi làm việc của nhân viên cơ sở y tế》(国卫医发〔2021〕41 号) điều khoản thứ bảy: 「维护诊疗秩序，不破坏就医公平。坚持平等原则，共建公平就医环境。严禁利用号源、床源、紧缺药品耗材等医疗资源或者检查、手术等诊疗安排收受好处、损公肥私」. Điều khoản thứ tám:「共建和谐关系，不收受患方『红包』。恪守医德、严格自律」. Cùng điều khoản đó còn ghi「严禁索取或者收受患者及其亲友的礼品、礼金、消费卡和有价证券、股权、其他金融产品等财物」. Những bữa ăn, vui chơi do bên bệnh nhân trả tiền cũng vậy:「严禁参加其安排、组织或者支付费用的宴请或者旅游、健身、娱乐等活动安排」. Văn bản đồng thời yêu cầu cơ sở y tế thực hiện trách nhiệm chính, tiến hành giám sát và「坚决查处违规行为」. Đánh giá mức hài lòng đã được đưa vào công tác đánh giá hiệu quả của bệnh viện công lập cấp ba. Căn cứ là《Ý kiến của Văn phòng Hội đồng Nhà nước về tăng cường công tác đánh giá hiệu quả bệnh viện công lập cấp ba》(国办发〔2019〕4 号). Văn bản nói:「三级公立医院绩效考核指标体系由医疗质量、运营效率、持续发展、满意度评价等 4 个方面的指标构成」. Văn bản còn ghi:「医院满意度由患者满意度和医务人员满意度两部分组成……」. Sau đó nói tiếp, phải「通过门诊患者、住院患者和医务人员满意度评价，衡量患者获得感及医务人员积极性」. Bộ đánh giá này còn có nền tảng thống nhất toàn quốc:「国家建立公立医院满意度管理制度……」. Mốc thời gian là:「2019 年 3 月底前，全国三级公立医院全部纳入国家卫生健康委满意度调查平台。各地要应用国家卫生健康委满意度调查平台，将调查结果纳入三级公立医院绩效考核」. Bệnh viện còn phải quản lý khiếu nại. Căn cứ là《Biện pháp quản lý khiếu nại của cơ sở y tế》(国家卫生健康委令第 3 号, áp dụng từ năm 2019), Điều hai. Quản lý khiếu nại là việc bệnh nhân「提出意见、建议或者投诉请求」về hành vi khám chữa bệnh, quản lý y tế, an toàn chất lượng khám chữa bệnh. Bệnh viện phải chịu trách nhiệm điều tra, xử lý và phản hồi.「二级以上医疗机构应当设置医患关系办公室或者指定部门（以下统称投诉管理部门）统一承担投诉管理工作」. Nhiệm vụ của bộ phận tiếp nhận khiếu nại gồm「统一受理投诉，调查、核实投诉事项，提出处理意见，及时答复患者」. Cơ sở y tế「应当鼓励工作人员主动收集患者对医疗服务、医疗质量安全等方面的意见和建议」(toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Quốc gia Y tế (国家卫生健康委), Cục Quản lý bảo hiểm y tế quốc gia (国家医保局), Cục Quản lý Y học cổ truyền quốc gia (国家中医药局) (2021). Thông báo in Chín nguyên tắc về liêm chính khi làm việc của nhân viên cơ sở y tế（国卫医发〔2021〕41 号）. Công báo của Hội đồng Nhà nước ；Văn phòng Hội đồng Nhà nước (国务院办公厅) (2019). Ý kiến về tăng cường công tác đánh giá hiệu quả bệnh viện công lập cấp ba（国办发〔2019〕4 号）. ；Ủy ban Quốc gia Y tế (国家卫生健康委) (2019). Biện pháp quản lý khiếu nại của cơ sở y tế（委令第 3 号）. +- Ghi chú: Người hưởng lợi là bạn, và cũng là người đã cứu bạn. Làm việc cứu người cả đêm rồi muốn cảm ơn là chuyện thường tình, cảm ơn được, chỉ cần đổi cách thức. Bao lì xì, thuốc lá và rượu đẩy rủi ro sang phía người nhận, còn thư cảm ơn, bằng khen và đánh giá mức hài lòng là những cách mà chế độ công nhận. Đã tặng mà bên nhận đẩy lại thì đừng hiểu là họ thấy ít hoặc làm bộ, đó là việc họ bắt buộc phải làm theo quy định. Ngoài ra cần phân biệt hai việc: tặng bao lì xì cho bác sĩ thuộc kỷ luật ngành và do bệnh viện quản lý, tặng tài vận cho công chức thụ lý và công chức thi hành thì là tội hối lộ theo Bộ luật Hình sự, hai việc không cùng cấp độ, cái sau xem mục 40 ở phần 8. Việc viết thư cảm ơn, tặng bằng khen thì bệnh viện thường nhận, nhưng câu「会进个人和科室记录」là kinh nghiệm thực tế, không có quy định nào viết thế. Có căn cứ trong văn bản là việc kết quả đánh giá mức hài lòng được đưa vào công tác đánh giá hiệu quả bệnh viện công lập cấp ba. Có nghi ngờ về bản thân việc khám chữa bệnh thì yêu cầu niêm phong hồ sơ ngay tại chỗ (xem mục 7 của phần này, niêm phong hồ sơ), đồng thời giữ lại hồ sơ và ảnh (mục 6, tự giữ hồ sơ và ảnh) \ No newline at end of file diff --git "a/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" "b/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" new file mode 100644 index 00000000..52b42379 --- /dev/null +++ "b/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" @@ -0,0 +1,96 @@ +[← Về mục lục](../README.md) + +# 25. Sau khi người thân mất thì phải làm gì + +Thước đo: tiền. Phần này chỉ nói thủ tục làm thế nào, tiền lấy ra sao, gồm lúc vừa mất thì làm gì trước, những khoản chi phí nào có bảng giá để đối chiếu. Phần này còn nói những khoản tiền nằm trong các tài khoản khác nhau có lấy lại được không, những việc nào người thân gần được đứng ra làm. Năm mục đầu xếp theo thứ tự sự việc xảy ra, lần lượt là hiện trường, giấy chứng minh tử vong, thi thể và thiêu, bất đồng về nguyên nhân tử vong thì xử lý sao, và huỷ đăng ký hộ khẩu. Các mục sau xếp theo hiệu quả chi phí. Những việc nên làm khi người thân còn sống, như giám hộ tự nguyện bằng văn bản, di chúc, tài khoản, xem phần 17. Ba khoản trợ cấp khi chết vì tai nạn lao động xem phần 19. + +Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, áp dụng từ ngày 30 tháng 3 năm 2026. Mức trợ cấp mai táng và trợ cấp tử vong trong bảo hiểm hưu trí bằng bao nhiêu thì do cơ quan nhân sụ và xã hội quy định riêng. Lần này không lấy được văn bản gốc đối chiếu từng chữ, nên phần chữ trong sách không viết con số cụ thể. + +### 1. Người mất ở nhà hay ngoài đời, trước hết phân biệt tử vong bình thường với tử vong bất thường: tai nạn, chết không rõ nguyên nhân, người sống một mình lâu ngày mới bị phát hiện thì báo công an ngay, thi thể và hiện trường đừng đụng vào + +- Chi phí: Không tốn tiền. Chỉ cần gọi một cuộc. +- Nói thẳng: Chết do tai nạn, không rõ vì sao chết, sống một mình lâu ngày mới được phát hiện thì chỉ công an mới cấp được giấy chứng minh tử vong. Hiện trường đã bị đụng vào thì việc khó nói rõ, giấy cũng khó cấp. Không có tờ giấy đó thì sau này thiêu, huỷ hộ khẩu, bồi hoàn bảo hiểm đều kẹt. Không phân biệt được thuộc loại nào thì gọi 110 trước. Cảnh sát tới nơi kiểm tra xong vẫn cấp giấy cho bạn. +- Lợi ích: Điều tám khoản hai của《Quy chế đăng ký hộ khẩu》ghi:「公民因意外事故致死或者死因不明，户主、发现人应当立即报告当地公安派出所或者乡、镇人民委员会」. Nghĩa là xảy ra tai nạn, hoặc không biết người đó chết thế nào, thì chủ hộ và người phát hiện phải báo ngay công an phường xí phong hoặc chính quyền xã, thị trấn. Điều hai mươi ba của《Quy chế quản lý tang chế》ghi:「自然人正常死亡的，医疗卫生机构应当及时出具死亡证明。涉及公安机关依法处置且未经医疗卫生机构救治的非正常死亡案（事）件的逝者，由公安机关出具死亡证明」. Nghĩa là chết bệnh bình thường thì cơ sở y tế đã khám cho người đó cấp giấy. Người chết không qua cơ sở y tế cứu chữa mà lại thuộc diện công an xử lý thì công an cấp. Giấy chứng minh tử vong bất thường chỉ công an mới cấp được. Hiện trường đã bị đụng vào thì vừa không nói rõ được vừa khó cấp. Sau này thiêu, huỷ hộ khẩu, bồi hoàn bảo hiểm, tất cả đều kẹt ở tờ giấy này +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (1958). Quy chế đăng ký hộ khẩu nước Cộng hoà nhân dân Trung Quốc（中华人民共和国户口登记条例）(Điều tám). (đăng trên cổng thông tin điện tử của Chính phủ tỉnh Quảng Đông)；Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, áp dụng từ ngày 30 tháng 3 năm 2026, Điều hai mươi ba）. +- Ghi chú: Nằm liệt giường lâu ngày, tử vong tại nhà thì tính là tử vong bình thường. Có thể tìm cơ sở đã khám cho người đó lúc còn sống để cấp giấy, cũng có thể tìm trạm y tế cộng đồng hoặc trạm y tế xã ở khu vực bạn ở. Giấy chứng minh tử vong thì ai cứu chữa người đó sẽ cấp, xem mục 2 (giấy chứng minh tử vong). Không phân biệt được thuộc loại nào thì cứ xử lý như tử vong bất thường, gọi 110 trước, cảnh sát tới nơi điều tra xong vẫn cấp được giấy. Nếu dọn sạch hiện trường rồi mới báo công an thì không ai giúp được bạn. Đừng tự di chuyển thi thể, đừng rửa và thay đồ cho người đó, cũng đừng cất đi lọ thuốc và mẩu giấy. + +### 2. Giấy chứng minh tử vong là chìa khoá cho mọi việc phía sau: ai cứu chữa thì người đó cấp, tử vong bình thường tại nhà thì tìm cơ sở y tế cộng đồng, ký trong một ngày + +- Chi phí: Không tốn tiền. +- Nói thẳng: Mất ở bệnh viện thì tìm cơ sở đã cứu chữa xin giấy chứng minh tử vong. Mất ở nhà, chết bình thường thì tìm trạm y tế cộng đồng hoặc trạm y tế xã ở khu vực bạn ở. Nguyên nhân tử vong đã rõ thì ký trong một ngày. Huỷ hộ khẩu, thiêu, rút quỹ tích góp nhà ở, nhận quyền lợi bảo hiểm xã hội, bồi hoàn, làm thừa kế, việc nào cũng cần nó. Lấy được rồi thì sao chép thêm vài bản. +- Lợi ích:《Quy định quản lý thông tin tử vong dân số (thử nghiệm)》nói: chết trong cơ sở y tế thì do「负责救治的医疗卫生机构签发」, mất ở bệnh viện hoặc trạm y tế thì tìm cơ sở đã cứu chữa. Chết bình thường tại nhà hoặc nơi khác thì「由本辖区社区卫生服务机构或乡镇（街道）卫生院签发」, tìm trạm y tế cộng đồng, trạm y tế xí phong hoặc trạm y tế đường phố ở khu vực bạn ở. Cơ quan công an tư pháp kết luận là tử vong bất thường thì do cơ quan công an tư pháp xử lý theo quy định và thủ tục hiện hành. Thông báo năm 2026 của năm bộ nói: 《Giấy chứng minh tử vong》「纸质版与电子证照具有同等法律效力」, bản điện tử có hiệu lực như bản giấy. Tử vong bình thường đã rõ nguyên nhân thì phải ký trong một ngày kể từ khi người đó mất. Giấy bị thất lạc thì có thể xin cấp lại một lần ở cơ sở đã ký. Thông tin cơ bản có sai thì có thể căn cứ hồ sơ xin ký lại một lần +- Mức bằng chứng: A +- Nguồn: Văn phòng Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委办公厅) (2014). Quy định quản lý thông tin tử vong dân số (thử nghiệm)（国卫办规划发〔2014〕68 号, Điều chín）. (Sở Dân vụ thành phố Thâm Quyến đăng lại)；Ủy ban Quốc gia Y tế, Bộ Công an, Bộ Dân vụ, Cục Quản lý Y học cổ truyền quốc gia, Cục Quản lý và Phòng ngừa bệnh quốc gia (2026). Thông báo về tăng cường đăng ký thông tin giấy chứng minh tử vong và quản lý giấy chứng nhận điện tử của dân chúng（国卫规划发〔2026〕5 号）. (Ủy ban Y tế và Sức khoẻ tỉnh Phúc Kiến đăng lại) +- Ghi chú: Huỷ hộ khẩu, thiêu, rút quỹ tích góp nhà ở, nhận quyền lợi bảo hiểm xã hội, bồi hoàn bảo hiểm, làm thừa kế, việc nào cũng cần tờ giấy này. Lấy xong thì sao chép thêm vài bản, nơi nào cấp được giấy chứng nhận điện tử thì xin luôn. Chỉ có một lần để xin cấp lại, đừng đánh mất bản gốc. Tên hay số giấy tờ tùy thân bị sai thì sửa ngay tại chỗ, vì đến lúc làm thừa kế mới phát hiện thì mọi cửa sổ phải làm lại từ đầu. + +### 3. Vận chuyển, lưu giữ và thiêu thi thể chỉ làm được ở nhà tang lễ, kho lưu thi thể của bệnh viện không làm tang nghiệp, lưu giữ thường không quá 3 ngày + +- Chi phí: Không tốn tiền. +- Nói thẳng: Vận chuyển, lưu giữ, bảo quản, hoàn thiện thi thể và thiêu chỉ nhà tang lễ được làm. Kho lưu thi thể của bệnh viện không làm tang nghiệp, nhiều nhất là cho đặt tạm thi thể. Nhà tang lễ lưu giữ thường không quá 3 ngày, muốn giữ thêm phải đi gia hạn. Thiêu thì cần giấy chứng minh tử vong và giấy xác nhận thiêu. Sau khi thiêu, tờ giấy chứng minh thiêu đó khi làm thừa kế, bồi hoàn vẫn cần. +- Lợi ích: Điều hai mươi lăm của《Quy chế quản lý tang chế》ghi:「遗体接运、存放、防腐、整容、火化服务，由殡仪馆专门负责提供」. Mấy việc này nơi khác không làm được. Điều hai mươi tư:「医疗卫生机构的太平间不得外包，不得开展殡仪服务，可以为遗体暂时停放提供便利服务」. Kho lưu thi thể bệnh viện chỉ đặt tạm, không lo việc tang. Điều hai mươi bảy:「殡仪馆存放遗体一般不超过 3 日，需要延期存放的，丧属或者遗体移交方应当在殡仪馆办理延期存放手续」. Muốn giữ thêm vài ngày thì ra nhà tang lễ làm thủ tục gia hạn. Điều hai mươi tám:「殡仪馆应当凭死亡证明和火化确认书火化遗体，并出具火化证明」. Thiếu một trong hai giấy là không thiêu được, thiêu xong bạn được một tờ giấy chứng minh thiêu. Điều hai mươi chín:「在实行火葬的地区，遗体应当就地、就近在殡仪馆火化」, phải tìm nhà tang lễ gần nơi người đó mất để thiêu +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, áp dụng từ ngày 30 tháng 3 năm 2026, Điều tám, chín, hai mươi tư, hai mươi lăm, hai mươi bảy, hai mươi tám, hai mươi chín）. +- Ghi chú: Bọn「trọn gói」Ghé tới sau khi người mất ở bệnh viện thì phần lớn không phải người của nhà tang lễ. Bản thân việc vận chuyển thuộc mục cơ sở, có giá niêm yết, xem mục 6 (bảng giá mục cơ sở). Giấy chứng minh thiêu cất giữ kỹ, làm thừa kế, bảo hiểm, di dời tro cốt đều cần. Ở địa phương thiêu hay chôn đất do chính quyền cấp tỉnh quy định. Phong tục tang chế của dân tộc thiểu số được bảo vệ theo Điều chín của quy chế. Ai đáng hoả thiêu lại chôn đất, hoặc chôn, đắp mộ ngoài nghĩa trang và khu vực an táng sinh thái, thì cơ quan dân vụ sẽ yêu cầu sửa lại trong thời hạn (sửa lại cho đúng trong hạn). + +### 4. Bất đồng về nguyên nhân tử vong thì trong 48 giờ phải yêu cầu mổ xác, đồng thời niêm phong hồ sơ bệnh án + +- Chi phí: Không tốn tiền. Chỗ khó là lúc ấy còn nhớ ra mà yêu cầu. +- Nói thẳng: Có nghi ngờ về nguyên nhân tử vong thì phải yêu cầu mổ xác trong 48 giờ sau khi người đó mất. Bệnh viện có điều kiện bảo quản xác thì có thể kéo dài đến 7 ngày. Mổ xác còn phải có người thân gần ký đồng ý. Không ký, hoặc kéo dài quá thời hạn, thì trách nhiệm không làm rõ được nguyên nhân tử vong thuộc về bên kéo dài. Cùng lúc yêu cầu mổ xác thì yêu cầu niêm phong hồ sơ bệnh án. Thi thể mà đã thiêu thì không mổ xác được nữa. +- Lợi ích: Thời hạn mổ xác nằm ở Điều hai mươi sáu của《Quy chế phòng ngừa và xử lý tranh chấp y tế》. Nguyên văn: 「患者死亡，医患双方对死因有异议的，应当在患者死亡后 48 小时内进行尸检；具备尸体冻存条件的，可以延长至 7 日」. Nên muốn làm rõ nguyên nhân tử vong thì phải mổ xác trong 48 giờ sau khi người đó mất, bệnh viện có điều kiện đông lạnh thì nhiều nhất kéo dài đến 7 ngày. Cùng điều khoản đó còn viết: 「尸检应当经死者近亲属同意并签字，拒绝签字的，视为死者近亲属不同意进行尸检」. Người nhà không ký tức là coi như không đồng ý. Rồi viết: 「不同意或者拖延尸检，超过规定时间，影响对死因判定的，由不同意或者拖延的一方承担责任」. Ai không đồng ý, ai kéo dài quá hạn thì phần sau không làm rõ được nguyên nhân tử vong thuộc về người đó. Điều hai mươi tư: việc niêm phong, mở niêm phong hồ sơ「应当在医患双方在场的情况下进行」. Đóng bìn niêm phong và mở ra, bạn và bệnh viện đều phải có mặt +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2018). Quy chế phòng ngừa và xử lý tranh chấp y tế（国务院令第 701 号, Điều hai mươi tư, hai mươi sáu）. +- Ghi chú: Qua thời hạn rồi mới nói「lúc đầu tôi đã thấy không ổn」thì trách nhiệm lại rơi vào bên kéo dài. Thi thể đã thiêu thì không còn khả năng mổ xác, nên chuyện này phải chốt trước khi liên hệ nhà tang lễ. Cách niêm phong hồ sơ bệnh án xem phần 24. + +### 5. Huỷ đăng ký hộ khẩu có thời hạn theo luật: ở thành phố thì trước khi an táng, ở nông thôn thì trong một tháng + +- Chi phí: Không tốn tiền. Mang theo giấy chứng minh tử vong và sổ hộ khẩu, ra công an phường xí phong một chuyến. +- Nói thẳng: Ở thành phố phải huỷ hộ khẩu trước khi an táng, ở nông thôn thì trong một tháng. Mang giấy chứng minh tử vong và sổ hộ khẩu ra công an phường xí phóng làm. Kéo dài không làm thì chuyển nhượng thừa kế, nhận quyền lợi, rút bảo hiểm đều vướng vào tình trạng hộ khẩu. +- Lợi ích: Điều tám khoản một của《Quy chế đăng ký hộ khẩu》ghi:「公民死亡，城市在葬前，农村在一个月以内，由户主、亲属、抚养人或者邻居向户口登记机关申报死亡登记，注销户口。公民如果在暂住地死亡，由暂住地户口登记机关通知常住地户口登记机关注销户口」. Nghĩa là ở thành phố làm trước ngày chôn, ở nông thôn làm trong một tháng. Chủ hộ, người thân, người nuôi dưỡng hoặc hàng xóm đều có thể ra cơ quan đăng ký hộ khẩu (công an phường xí phóng) báo tử vong, xin huỷ hộ khẩu. Người chết khi đang tạm trú ở nơi khác thì công an nơi tạm trú báo công an nơi có hộ khẩu để huỷ +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (1958). Quy chế đăng ký hộ khẩu nước Cộng hoà nhân dân Trung Quốc（中华人民共和国户口登记条例）(Điều tám). (đăng trên cổng thông tin điện tử của Chính phủ tỉnh Quảng Đông) +- Ghi chú: Cứ để hộ khẩu còn nguyên thì chuyển nhượng thừa kế, nhận quyền lợi, rút bảo hiểm đều bị kẹt, các cửa sổ này đều xem tình trạng hộ khẩu. Lương hưu trí mà vẫn được phát thì đó là chiếm đoạt, sớm muộn cũng phải hoàn lại. Ra công an phường xí phóng tiện thể hỏi luôn: quyền lợi bảo hiểm xã hội và bảo hiểm y tế dừng thế nào, xin trợ cấp mai táng ra sao. Mấy khoản tiền đó xem mục 9 (số dư quỹ tích góp nhà ở, quyền lợi bảo hiểm xã hội). + +### 6. Dịch vụ tang chế chia thành mục cơ sở và mục ngoài cơ sở, mục cơ sở có bảng giá, mức thu do pháp luật quy định + +- Chi phí: Không tốn tiền. Trước khi lên đường, dành vài phút xem bảng giá. +- Nói thẳng: Vận chuyển thi thể, lưu giữ, lễ khí chung, thiêu, gửi tro cốt, an táng sinh thái đều nằm trong bảng mục cơ sở của Nhà nước. An táng bằng hài chương tro cốt do đơn vị công lập cung cấp cũng nằm trong đó. Giá của mấy mục này do quy định quyết định. Ngoài bảng giá thì không được tự lập mục khác để thu tiền. Trước khi lên đường, hãy hỏi rõ mục nào thuộc mục cơ sở, tiêu chuẩn giá ở địa phương là bao nhiêu, rồi hãy bàn phần còn lại. +- Lợi ích: Vận chuyển thi thể, lưu giữ thi thể, lễ khí chung, thiêu thi thể, gửi tro cốt, an táng sinh thái đều nằm trong bảng mục cơ sở của Nhà nước. An táng bằng hài chương tro cốt do cơ sở tang chế do Nhà nước tổ chức cũng nằm trong bảng. Mức thu của mấy mục này do pháp luật quy định, giá theo quy định, không phải nhà tang lễ tự nói. Danh mục ngoài bảng cơ sở do cơ quan dân vụ cấp phường trở lên, tức cấp địa phương và trở lên, quyết định +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, được Hội đồng Nhà nước sửa đổi thông qua tại phiên họp thường vụ lần thứ 72 ngày 14 tháng 11 năm 2025,「自2026年3月30日起施行」）. Điều năm: 「殡葬服务分为基础项目和非基础项目，收费政策由国务院发展改革部门、财政部门会同国务院民政等部门制定。」「国家制定殡葬服务基础项目清单，将遗体接运、遗体存放、遗体告别、遗体火化、骨灰寄存、生态安葬以及政府举办的殡葬服务机构提供的骨灰格位安葬等纳入清单范围」；Điều năm mươi「殡葬服务实行清单化管理并动态调整，禁止在清单之外设立项目、收取费用。」 +- Ghi chú: Tỉnh, khu tự trị, thành phố trực thuộc có thể thêm mục vào bảng của Nhà nước. Hãy hỏi rõ mục nào thuộc mục cơ sở, tiêu chuẩn thu ở địa phương là bao nhiêu, rồi hãy bàn phần còn lại. Giá mục trong bảng bị định theo quy định, ngoài bảng thì bạn chọn, phân biệt hai loại này thì phần bị thu thêm mới có căn cứ mà nói. Muốn làm tang nghiệm cho người thân cho ra dáng không có gì sai, cuốn sách này không khuyên bạn tiết kiệm, tiền tiêu ở đâu thì bạn tự quyết định. + +### 7. Gặp ép mua kèm, thêm tiền ngoài giá niêm yết, tách một dịch vụ thành mấy dịch vụ để thu tiền thì ghi lại và khiếu nại + +- Chi phí: Không tốn tiền. Chỗ khó là lúc ấy vẫn còn chịu làm rõ. +- Nói thẳng: Mấy việc sau đều là hành vi vi phạm giá mà quy chế nêu đích danh: không niêm yết bảng giá, thêm tiền ngoài mức niêm yết, kèm bán tang phẩm, ép bạn chấp nhận dịch vụ có thu phí của nơi khác. Tách một dịch vụ ra mấy dịch vụ để thu tiền cũng là tăng giá hình thức. Tại chỗ hãy đòi bảng giá trước, giữ lại chứng từ thu tiền và tin nhắn. Sau đó khiếu nại vẫn kịp. Kiểm tra giá thì tìm quản lý thị trường, quản lý dịch vụ tang chế thì tìm cơ quan dân vụ. +- Lợi ích: Quy chế liệt kê năm loại hành vi vi phạm giá, dưới đây nói bốn loại. Không niêm yết bảng giá, thu nhiều hơn mức niêm yết thì thuộc không niêm yết rõ ràng hoặc thêm tiền ngoài mức niêm yết. Không mua cái này thì không làm cái kia thì thuộc buộc bán hoặc gắn điều kiện bất hợp lý để bán hàng. Ép bạn dùng dịch vụ có thu phí của nơi khác thì thuộc bắt buộc, hoặc bắt buộc hình thức, chấp nhận dịch vụ có thu phí của bên thứ ba. Đổi tên để thu nhiều hơn thì thuộc tăng mức thu hình thức. Ngoài ra còn cấm「分解项目、扩大范围收费等形式变相提高收费标准」. Tách một dịch vụ thành mấy dịch vụ để thu riêng, hoặc mở rộng phạm vi thu tiền ra, đều nằm trong danh sách cấm +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, được Hội đồng Nhà nước sửa đổi thông qua tại phiên họp thường vụ lần thứ 72 ngày 14 tháng 11 năm 2025,「自2026年3月30日起施行」）. Điều bốn mươi hai: 「殡葬服务机构不得违反有关规定擅自设立收费项目，或者采取分解项目、扩大范围收费等形式变相提高收费标准。」Điều bốn mươi ba liệt kê hành vi vi phạm giá có「（一）不按规定明码标价，或者在标价之外加价或者收取未标明的费用」「（三）采取捆绑、附加不合理条件或者不合理限制等方式，强制或者变相强制销售丧葬用品、提供殡葬相关服务并收取费用」「（四）强制或者变相强制丧属接受第三方的有偿服务，或者利用相关服务网点重复收取费用」；Điều năm mươi một yêu cầu cơ quan dân vụ lập cơ chế chia sẻ thông tin giám sát với đường dây nóng phục vụ hành chính công,「对丧属反映的殡葬服务问题，应当会同有关部门及时予以处理并将处理结果告知丧属」 +- Ghi chú: Xử lý hành vi vi phạm giá thuộc cơ quan quản lý thị trường, bản thân dịch vụ tang chế thuộc cơ quan dân vụ. Tại chỗ hãy đòi một tờ bảng giá, giữ chứng từ thu tiền và tin nhắn, sau đó ra khiếu nại vẫn kịp. +### 8. Môi giới tang chế, môi giới mua tang phẩm, tổ chức và dẫn dắt tang sự phải đăng ký với cơ quan dân vụ cấp huyện + +- Chi phí: Không tốn tiền. Hỏi một câu xem đã đăng ký hay chưa. +- Nói thẳng: Môi giới các loại như đại diện tang chế, mua giúp tang phẩm, tổ chức và dẫn dắt tang sự, tư vấn thông tin phải đăng ký với cơ quan dân vụ cấp huyện. Bên không đăng ký thì không vào được hệ thống dịch vụ tang chế của cơ quan dân vụ. Mỗi nơi có tự tra cứu được hay không thì khác nhau, cách chắc nhất là gọi hỏi cơ quan dân vụ cấp huyện. Bên dùng thủ đoạn lừa đảo để thu tiền lung tung, hoặc bán lại thông tin người đã mất, sẽ bị xử lý theo pháp luật. +- Lợi ích: Bên môi giới không đăng ký thì không nằm trong hệ thống dịch vụ tang chế do cơ quan dân vụ quản lý. Bên dùng thủ đoạn lừa đảo để thu phí sai quy định, xúi giục làm tang nghiệp tốn kém, ép chấp nhận dịch vụ, bán lại thông tin người đã mất, sẽ bị xử lý theo pháp luật, còn bị ghi dấu trong hệ thống và công bố ra xã hội, ai cũng thấy +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, được Hội đồng Nhà nước sửa đổi thông qua tại phiên họp thường vụ lần thứ 72 ngày 14 tháng 11 năm 2025,「自2026年3月30日起施行」）. Điều bốn mươi mốt: 「除殡葬服务机构外，从事殡葬服务代理、用品代购、策划主持、信息咨询等殡葬相关服务活动的组织和个人，应当向县级人民政府民政部门备案。」「对存在以欺骗等手段违规收费、诱导大操大办、强迫接受服务、倒卖逝者信息等损害丧属权益行为的依法处置，并在殡葬服务信息系统中进行标注，向社会公开。」 +- Ghi chú: 「倒卖逝者信息」là cái bẫy phổ biến nhất trong nghề này. Vừa làm xong thủ tục lại nhận ngay một loạt cuộc gọi quảng cáo thì phần lớn là thông tin đã bị chuyển nhượng. Nền tảng hành hương trực tuyến cũng chịu quản lý như vậy. Điều bốn mươi lăm của quy chế không cho phép nó xúi giục tiêu và cũng không cho phép xử lý thông tin người đã mất trái phép. + +### 9. Tiền nằm rải rác các nơi thì phải đi lấy từng chỗ: số dư quỹ tích góp nhà ở, quyền lợi bảo hiểm xã hội, quyền lợi tai nạn lao động + +- Chi phí: Không tốn tiền, nhưng phải đi mấy cửa sổ. Trước khi đi chuẩn bị giấy chứng minh tử vong và giấy chứng minh quan hệ thân nhân. +- Nói thẳng: Số dư quỹ tích góp nhà ở thì người thừa kế được rút, khoản này không nằm trong tài khoản ngân hàng nên dễ bị bỏ sót nhất. Số tiền còn lại trong tài khoản bảo hiểm hưu trí cá nhân cũng được thừa kế. Chết vì bệnh thì người nhà được hưởng thêm trợ cấp mai táng và trợ cấp tử vong từ 3 đến 24 tháng. Chết vì tai nạn lao động thì có thêm ba khoản của tai nạn lao động. +- Lợi ích: Số dư tài khoản quỹ tích góp nhà ở, tức số tiền còn lại trong tài khoản quỹ của người đó, người thừa kế và người được tặng theo di chúc được rút. Người được tặng theo di chúc là người được ghi trong di chúc, người mà người đó để lại tài sản cho. Chết vì tai nạn lao động thì có thêm ba khoản: trợ cấp mai táng, trợ cấp cho thân nhân được trợ cấp và trợ cấp tử vong do tai nạn lao động một lần. Số tiền còn lại trong tài khoản bảo hiểm hưu trí cá nhân, Luật bảo hiểm xã hội ghi rõ có thể thừa kế. Người tham gia bảo hiểm hưu trí người lao động chết vì bệnh hoặc không phải vì tai nạn lao động thì người nhà còn được nhận trợ cấp mai táng và trợ cấp tử vong, gọi chung là quyền lợi cho thân nhân. Công thức cho người lao động doanh nghiệp thống nhất trên cả nước. Trợ cấp mai táng bằng 2 lần mức thu nhập chi tiêu bình quân mỗi người mỗi tháng của dân cư đô thị năm trước của tỉnh. Trợ cấp tử vong lấy mức thu nhập hằng tháng đó làm căn cứ, trả theo thời gian đóng bảo hiểm, từ 3 đến 24 tháng: dưới 5 năm là 3 tháng, từ 5 đến 10 năm là 6 tháng, từ 10 đến 15 năm là 9 tháng, trên 15 năm thì cứ thêm 1 năm thì cộng 1 tháng, nhiều nhất 24 tháng. Người đã nghỉ hưu thì cứ nhận 1 năm lương hưu trí thì bớt 1 tháng, ít nhất 9 tháng +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý quỹ tích góp nhà ở（国务院令第 844 号 sửa đổi, áp dụng từ ngày 20 tháng 9 năm 2026）, khoản cuối Điều hai mươi tư. : 「职工死亡或者被宣告死亡的，职工的继承人、受遗赠人可以提取职工住房公积金账户内的存储余额；无继承人也无受遗赠人的，职工住房公积金账户内的存储余额纳入住房公积金的增值收益。」；ba khoản trợ cấp ở Điều ba mươi chín của《Quy chế bảo hiểm tai nạn lao động》xem phần 19. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018 sửa đổi). Luật bảo hiểm xã hội（Điều mười bốn, mười bảy）: 「个人死亡的，个人账户余额可以继承」「参加基本养老保险的个人，因病或者非因工死亡的，其遗属可以领取丧葬补助金和抚恤金」. (Cơ sở dữ liệu văn bản quốc pháp)；Bộ Nhân sụ và Xã hội, Bộ Tài chính (人力资源社会保障部、财政部) (2021). Biện pháp tạm thời về quyền lợi cho thân nhân trong bảo hiểm hưu trí cơ sở của người lao động doanh nghiệp（人社部发〔2021〕18 号, áp dụng từ ngày 1 tháng 9 năm 2021, Điều bốn, năm）, bản gốc không đăng công khai, văn bản được đối chiếu theo trích dẫn nguyên văn trong quyết định giải quyết khiếu nại hành chính của Chính phủ quận Hoa Đô, thành phố Quảng Châu（花都府行复〔2022〕226 号）: 「丧葬补助金的标准，按照参保人员死亡时本省……上一年度城镇居民月人均可支配收入的2倍计算」「缴费年限不满5年的，发放月数为3个月……缴费年限30年以上的，按照30年计算，发放月数最高为24个月」. ；Sở Nhân sụ và Xã hội tỉnh Phúc Kiến (福建省人力资源和社会保障厅). Câu hỏi thường gặp: 「每领取1年基本养老金减少1个月，发放月数最低为9个月」「累计缴费年限不足5年的，其遗属待遇（含丧葬补助金和抚恤金）标准不得超过其个人缴费之和」. +- Ghi chú: Khoản quỹ tích góp nhà ở hay bị bỏ sót nhất, vì nó không nằm trong tài khoản ngân hàng và cũng không tự động vào phần chia tài sản. Số tiền còn lại trong tài khoản bảo hiểm hưu trí cá nhân và quyền lợi cho thân nhân là hai khoản tính riêng và phát riêng, phải ra cửa sổ bảo hiểm xã hội nơi tham gia bảo hiểm làm cùng một lúc. Mỗi nơi đòi giấy tờ khác nhau. Thời gian đóng dưới 5 năm thì tổng quyền lợi cho thân nhân không vượt quá số tiền bản thân người đó đã đóng. Công thức này lấy từ biện pháp về bảo hiểm hưu trí người lao động doanh nghiệp, còn quyền lợi cho thân nhân trong bảo hiểm hưu trí của cơ quan, tổ chức, đơn vị công lập và bảo hiểm hưu trí dân cư thì hỏi tại cửa sổ. Yêu cầu về hình thức của di chúc và thừa kế xem phần 17. + +### 10. Tài khoản và thông tin cá nhân của người đã mất, người thân gần có quyền yêu cầu tra cứu, sao chép, sửa, xoá + +- Chi phí: Không tốn tiền. Chỉ khi nền tảng không hợp tác mới phải khiếu nại hoặc ra tòa. +- Nói thẳng: Sau khi người đó mất, người thân gần vì lợi ích hợp pháp và chính đáng của mình có thể yêu cầu nền tảng tra cứu, sao chép, sửa, xoá thông tin cá nhân của người đã mất. Nền tảng mà từ chối thì phải nói lý do. Bạn cũng có thể ra tòa kiện. Có một ngoại lệ: trước khi mất người đó đã sắp xếp khác, ví dụ đã đặt người thừa kế tài khoản hoặc đã để lại ý muốn rõ ràng thì theo cách sắp xếp đó. +- Lợi ích: Luật viết rõ: người tự nhiên mất, thì người thân gần vì lợi ích hợp pháp, chính đáng của mình có thể thực hiện các quyền như tra cứu, sao chép, sửa, xoá đối với thông tin cá nhân liên quan của người đã mất. Người thân gần có thể yêu cầu nền tảng đưa thông tin đó ra cho xem, sao chép một bản, sửa chỗ sai, xoá đi +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). Luật bảo vệ thông tin cá nhân. Trang Quốc hội nhân dân. Điều bốn mươi chín: 「自然人死亡的，其近亲属为了自身的合法、正当利益，可以对死者的相关个人信息行使本章规定的查阅、复制、更正、删除等权利；死者生前另有安排的除外。」；Điều năm mươi「个人信息处理者应当建立便捷的个人行使权利的申请受理和处理机制。拒绝个人行使权利的请求的，应当说明理由」「个人可以依法向人民法院提起诉讼」 +- Ghi chú: Trong điều luật có một câu「死者生前另有安排的除外」, người đã từng đặt người thừa kế tài khoản hoặc để lại ý muốn rõ ràng thì làm theo điều người đó đã để lại. Vậy nên khi người đó còn tỉnh, đáng viết ra tài khoản và mật khẩu nằm ở đâu, xem phần 17. Huỷ tài khoản và xoá thông tin cá nhân là hai việc khác nhau, xem phần 14. \ No newline at end of file diff --git "a/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" "b/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" index a9946a4a..6e667795 100644 --- "a/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" +++ "b/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" @@ -1,10 +1,9 @@ [← 回总目录](../README.md) -# 28. 别为了外形把身体搞坏 +# 28. Đừng phá hỏng cơ thể vì ngoại hình -本节说的是「为了改变外形而对身体动手」的几种做法：饿肚子、吃药、打针、开刀。这几件事，决定多半是和别人比较之后做的，代价却全落在自己身上，而且相当一部分改不回来。 - -本节不评价任何人想把自己变成什么样子，那是各人自己的事。同一个目标，可以找「有处方、有资质、有复查」的，也可以「网购、美容院、自己加量」，本节只比这两种做法的风险差多少。算账主要看身体上实际出了什么事，比如失明、住院、死亡。医美那两条还涉及非法行医的刑事责任。 +Phần này nói về mấy cách làm thay đổi ngoại hình bằng cách đụng vào cơ thể: nhịn ăn, uống thuốc, tiêm, mổ. Mấy việc này, phần lớn quyết định làm sau khi so với người khác, cái giá lại dồn hết vào bản thân mình, và một phần khá lớn không sửa lại được. +Phần này không đánh giá ai muốn mình thành hình dạng nào; đó là việc của từng người. Cùng một mục tiêu, có thể tìm chỗ「có đơn thuốc, có giấy phép, có theo dõi lại」, cũng có thể「mua qua mạng, đến thẩm mỹ viện, tự tăng liều」; phần này chỉ so hai cách đó khác nhau về rủi ro bao nhiêu. Cách tính chủ yếu xem cơ thể thực sự bị gì, như mù mắt, nằm viện, tử vong. Hai mục về thẩm mỹ còn liên quan tới trách nhiệm hình sự của hành nghề y không đúng phép. ### 1. Đừng kiểm soát cân nặng bằng ăn kiêng cực đoan, nhịn ăn hay tự nôn; muốn giảm thì giảm bằng vận động diff --git "a/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" "b/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" index 8fa6b52f..24352835 100644 --- "a/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" +++ "b/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" @@ -1,10 +1,9 @@ [← 回总目录](../README.md) -# 30. 上学以后的孩子 +# 30. Trẻ đang đi học -第 20 节讲的是孩子刚出生那一年。第 18 节算的是养孩子要花多少钱、多少时间。中间 3 到 18 岁这十几年，写在这一节。分龄管屏幕那一条从 0 岁写起，因为官方的分档就是 0～3 岁、3～6 岁、中小学三档，拆开写反而不好用。本节回答的是：「别耽误学习」和「身体」撞在一起的时候，哪些事不能等。这些事有时间窗口，过了就补不回来。学业该不该那么卷，本节不讨论，只给窗口和数字。 - -前八条和最后两条算的是死亡和身体上真落下的损害。第 9 条（不买「治愈近视」的产品）算钱。第 10 条（睡眠作业体育排名的明文规定）和第 11 条（休学保留学籍）算时间。溺水、交通、头盔、HPV 疫苗见第 1 节。婴幼儿见第 20 节。孩子被骗和游戏充值见第 5 节。未成年人自己别踩的红线见第 9 节。急症怎么认、怎么叫车见第 13 节。家长去世后的孩子见第 29 节。这些都不在本节重复。 +Mục 20 nói về năm đầu tiên khi đứa trẻ vừa mới sinh. Mục 18 tính câu hỏi nuôi con có đáng không. +Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật sự rơi vào thân thể. Mục 9 (kỳ vọng khi làm việc với cảnh sát, bác sĩ, thủ quầy) dẫn quy định chứ không phải thí nghiệm, nên xếp C. Các mục còn lại xếp B, vì chủ yếu dựa vào một nghiên cứu hoặc khảo sát tự báo cáo. ### 1. Trẻ bảo đau dữ dội bất ngờ và đau hơn nữa thì đưa đi bệnh viện ngay, đừng để cháu nhịn đến tan học diff --git "a/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" "b/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" new file mode 100644 index 00000000..35bf8b98 --- /dev/null +++ "b/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" @@ -0,0 +1,150 @@ +[← Về mục lục](../README.md) + +# 31. Sau mười tám tuổi có những con đường nào + +Phần này tính tiền và thời gian; riêng mục về không đi nghĩa vụ quân sự còn liên quan tới tự do. Đây là một bản đồ các con đường, không phải hướng dẫn chi tiết cho một con đường nào. Phần này đặt các ngưỡng pháp lý của từng đường cạnh nhau trước, rồi lần lượt viết những điều mà biết rồi mới dùng được: đi nghĩa vụ quân sự thì được gì, vào làm viên chức trong hệ thống nhà nước ngoài kỳ thi công chức còn những lối nào, khi điền nguyện thi đại học thì hai ngành nào đối thẳng với chỗ có biên chế, bằng học sau này còn bổ sung được không, muốn đi làm ở nước ngoài thì tìm công ty kiểu nào, người không vào đơn vị làm bảo hiểm xã hội thế nào, không có vốn khởi đầu thì có khoản vay nào nhà nước góp một phần lãi. +Những con đường có thể đi sau mười tám tuổi còn nhiều hơn những gì phần này viết. Ở đây chỉ nhận những đường có văn bản công khai ở tầm cả nước, các ngưỡng và đãi ngộ kiểm chế được từng điều. Vào doanh nghiệp nhà nước, vào nhà máy, đi làm thuyền viên, làm tự do, học tiếp trung học rồi đại học — những đường đó cũng đi được. Chỉ là chúng không có một văn bản thống nhất để tra, sách này không viết điều kiện cho chúng theo cảm tính. +Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai thứ: ngưỡng nằm trong văn bản nào, và sau khi đi lên đó có những hạn chót nào, chứng chỉ nào hết hạn thì mất. Câu hỏi học hay đi làm tính thế nào, chọn kỹ năng ra sao, xem phần 23. Sau khi vào đơn vị, tiền làm thêm giờ, tai nạn lao động và bồi thường khi nghỉ việc, xem phần 19. Vốn và lối thoát khi mở cửa hàng hay lập công ty, xem phần 12. Du học, xem phần 32. Bảo đảm sau khi thất nghiệp, xem phần 7. Những phần đó không lặp lại ở đây. + +### 1. Đối chiếu điều kiện trước đã: yêu cầu về tuổi và trình độ của mười hai con đường đều được viết thành văn bản + +- Chi phí: Không tốn tiền. Dành mười phút đối chiếu một lượt, xem mình có đủ điều kiện không. +- Nói thẳng: Điều kiện về tuổi và trình độ của mười hai con đường đều được viết trong văn bản. Đi làm công ăn thường phải đủ mười sáu tuổi. Đi nghĩa vụ quân sự là nam giới đủ mười tám tuổi, người tốt nghiệp đại học có thể nới đến hai mươi bốn tuổi. Thi tuyển công chức cần từ mười tám đến ba mươi lăm tuổi, trình độ từ cao đẳng trở lên. Còn thi tự học và đăng ký làm hộ kinh doanh cá thể thì không xem trình độ, cũng không xem tuổi. +- Lợi ích: Đi nghĩa vụ quân sự thì xem Luật nghĩa vụ quân sự (兵役法):「年满十八周岁的男性公民，应当被征集服现役；当年未被征集的，在二十二周岁以前仍可以被征集服现役。普通高等学校毕业生的征集年龄可以放宽至二十四周岁，研究生的征集年龄可以放宽至二十六周岁」, tức nam công dân đủ mười tám tuổi thì phải được truy xuất nhập ngũ; người không được truy xuất trong năm đó thì trước hai mươi hai tuổi vẫn có thể được truy xuất; tuổi truy xuất của người tốt nghiệp đại học thường có thể nới đến hai mươi bốn tuổi, của nghiên cứu sinh nới đến hai mươi sáu tuổi. Thi tuyển công chức thì xem Điều mười tám của Quy định tuyển dụng công chức《公务员录用规定》. Điều kiện dự thi gồm「年龄为十八周岁以上，三十五周岁以下」「具有大学专科以上文化程度」, tức từ mười tám tuổi trở lên và dưới ba mươi lăm tuổi, có trình độ từ cao đẳng trở lên. Hai điều kiện này có thể được điều chỉnh cho phù hợp nếu được cơ quan quản lý công chức cấp tỉnh trở lên phê duyệt. Điều mười chín của quy định đó liệt kê năm loại người không được dự thi. Ví dụ người từng bị kết án hình sự vì tội phạm, người bị sa thải khỏi chức vụ công, người bị đưa vào danh sách cấm đoán để trừng phạt liên ngành vì mất uy tín. Tuyển lính cứu hộ thì xem《Biện pháp tuyển lính trong đội cứu hộ và cứu nạn tổng hợp quốc gia》《国家综合性消防救援队伍消防员招录办法》. Điều năm:「年龄为 18 周岁以上、22 周岁以下」「具有高中以上文化程度」. Điều sáu:「大学专科以上学历人员、解放军和武警部队退役士兵……年龄可以放宽至 24 周岁」. Quân nhân văn chứng thì xem Điều chín của《Quy chế nhân sự văn chứng Quân giải phóng nhân dân Trung Quốc》《中国人民解放军文职人员条例》, điều kiện cơ bản có「年满 18 周岁」. Điều mười chín của quy chế đó ghi rõ đối với chức vụ cấp bậc thấp thì tuổi tối đa khi tuyển lần đầu là 35 tuổi. Cơ quan hành chính sự vệ tổ chức tuyển dụng công chức thì xem Điều tám của《Quy chế quản lý nhân sự của cơ quan hành chính sự vệ》《事业单位人事管理条例》:「事业单位新聘用工作人员，应当面向社会公开招聘」. Hộ kinh doanh cá thể thì xem《Quy chế thúc đẩy phát triển hộ kinh doanh cá thể》《促进个体工商户发展条例》:「有经营能力的公民……依法登记为个体工商户的，适用本条例」. Thi tự học thì xem《Quy chế lâm thời về thi tự học bậc đại học》《高等教育自学考试暂行条例》. Điều ba:「中华人民共和国公民，不受性别、年龄、民族、种族和已受教育程度的限制，均可依照本条例的规定参加高等教育自学考试」. Điều hai mươi:「各级各类全日制学校的在校生不得报考」. Sinh viên chính quy có học phí do Nhà nước bảo trả thì xem Điều hai của《Biện pháp thực hiện chế độ giáo dục có học phí cho sinh viên sư phạm liên thông bậc đại học và sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục》《教育部直属师范大学本研衔接师范生公费教育实施办法》. Bộ này chỉ tuyển ở sáu trường đại học sư phạm trực thuộc Bộ. Cụ thể là Đại học Bắc Kinh, Đại học Hoa Đông, Đại học Đông Bắc, Đại học Hoa Trung, Đại học Thiển Tây và Đại học Tây Nam. Bậc đại học và bậc sau đại học thiết kế liền nhau, chia thành từng đoạn đánh giá, học xong bậc đại học thì tiếp tục học sau đại học. Học phí và tiền ở trong suốt thời gian học do ngân sách trung ương chi trả, ngoài ra còn có trợ cấp sinh hoạt. Điều năm của biện pháp đó quy định「招收公费师范生实行提前批次录取」, tức nhận sinh viên có học phí theo lô nhập trước các lô bình thường. Sinh viên y khoa định hướng thì xem《Thông báo về việc làm tốt công tác bố trí việc làm và quản lý thực hiện cam kết cho sinh viên y khoa được đào tạo miễn phí theo đơn đặt hàng nông thôn》《关于做好农村订单定向免费培养医学生就业安置和履约管理工作的通知》. Nó viết:「在录取后、获得入学通知书前，须与培养学校和定向就业所在地的县级卫生健康行政部门……人力资源社会保障部门签署协议」. Đi làm công ăn ở nước ngoài thì xem《Quy chế quản lý hợp tác lao động ngoài biên giới》《对外劳务合作管理条例》. Điều năm: muốn làm hợp tác lao động ngoài biên giới thì phải được cơ quan quản lý thương mại phê duyệt và có tư cách kinh doanh. Điều bảy khoản bốn:「未依法取得对外劳务合作经营资格证书并办理登记，不得从事对外劳务合作」. Công ty không có giấy phép đó mà tổ chức người đi làm công ăn nước ngoài thì bản thân nó đã vi phạm pháp. Điều tám khoản hai:「任何单位和个人不得以商务、旅游、留学等名义组织劳务人员赴国外工作」. Nhận việc qua sàn và làm việc tự do không có ngưỡng trình độ theo luật; căn cứ tự tham gia bảo hiểm xã hội và bảo hiểm tai nạn lao động xem mục 11 và 12 của phần này. Luật Giáo dục bắt buộc quy định「禁止用人单位招用应当接受义务教育的适龄儿童、少年」(toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2021). Luật nghĩa vụ quân sự（兵役法）（第二十条）. （đăng trên Bộ Quốc phòng）；Ban Tổ chức Trung ương Đảng Cộng sản Trung Quốc (中共中央组织部) (2019). Quy định tuyển dụng công chức（公务员录用规定）（第十八、十九条）. （đăng trên Cổng Đảng viên）；Bộ Nhân sự và Xã hội, Bộ Môi trường và Tự nhiên (人力资源社会保障部、应急管理部) (2021). Biện pháp tuyển lính trong đội cứu hộ và cứu nạn tổng hợp quốc gia（国家综合性消防救援队伍消防员招录办法）（人社部发〔2021〕58 号，第五、六条）. ；Hội đồng Nhà nước, Ủy ban Quân sự Trung ương (国务院、中央军委) (2022). Quy chế nhân sự văn chứng Quân giải phóng nhân dân Trung Quốc（中国人民解放军文职人员条例）（sửa đổi lần hai, Lệnh số 757, 第九、十九条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Hội đồng Nhà nước (国务院) (2014). Quy chế quản lý nhân sự của cơ quan hành chính sự vệ（事业单位人事管理条例）（Hội đồng Nhà nước Lệnh số 652, 第八条）. ；Hội đồng Nhà nước (国务院) (2022). Quy chế thúc đẩy phát triển hộ kinh doanh cá thể（促进个体工商户发展条例）（Hội đồng Nhà nước Lệnh số 755, 第二条）. ；Hội đồng Nhà nước (国务院) (1988, sửa đổi 2014). Quy chế lâm thời về thi tự học bậc đại học（高等教育自学考试暂行条例）（第三、二十条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2018). Luật Giáo dục bắt buộc（义务教育法）（第十四条）. （đăng trên Bộ Giáo dục）；Văn phòng Hội đồng Nhà nước (国务院办公厅) (2024). Thông báo chuyển tiếp các biện pháp của Bộ Giáo dục và các bộ về chế độ giáo dục có học phí cho sinh viên sư phạm liên thông bậc đại học và sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục（国办发〔2024〕27 号，第二、五条）. ；Ủy ban Quốc gia Y tế và 7 bộ khác (国家卫生健康委等 7 部门) (2019). Thông báo về việc làm tốt công tác bố trí việc làm và quản lý thực hiện cam kết cho sinh viên y khoa được đào tạo miễn phí theo đơn đặt hàng nông thôn. ；Hội đồng Nhà nước (国务院) (2012). Quy chế quản lý hợp tác lao động ngoài biên giới（对外劳务合作管理条例）（Hội đồng Nhà nước Lệnh số 620, 第五、七、八条）. +- Ghi chú: Đủ điều kiện rồi còn phải qua mấy chặng nữa là điều kiện chính trị, khám sức khoẻ và thẩm tra chính trị. Mấy chặng đó không có tiêu chuẩn chấm điểm công khai, tra văn bản cũng không có. Cửa vào tuyển sinh liên thông của trường trung cấp nghề, trường kỹ nghề và giáo dục nghề xem mục 5 ở phần 23; trợ cấp khi gia đình không gánh nổi học phí xem mục 4 ở phần 23. Hộ kinh doanh cá thể không xem trình độ, nhưng mở cửa hàng không dễ. Vốn, bảo lãnh cá nhân và chi phí rút lui khi đóng cửa xem phần 12. Bảo lãnh cá nhân nghĩa là công ty không trả được thì chính bạn phải trả. Tiền khởi đầu có vay được khoản vay mà Nhà nước góp một phần lãi suất không, xem mục 16 của phần này. Hai con đường sinh viên sư phạm có học phí và sinh viên y khoa định hướng đã được chốt vào ngày điền nguyện vọng kỳ thi gaokao, cái giá nằm ở mục 13 của phần này. Điều kiện của việc đi làm công ăn nước ngoài không nằm ở bạn mà ở chỗ công ty có đủ tư cách hay không, cách nhận biết xem mục 14 của phần này. Con đường nào đáng hơn thì phần này không so sánh. + +### 2. Năm thứ mười tám tuổi phải đăng ký nghĩa vụ quân sự trước ngày 31 tháng 10; quân nhân lính nghĩa vụ nhập ngũ đúng hai năm + +- Chi phí: Không tốn tiền. Đăng ký thì điền một lần trên mạng, vài phút là xong. Nếu thật sự đi nhập ngũ thì cái giá là hai năm toàn thời gian, trong hai năm đó không làm được việc khác song song. +- Nói thẳng: Đăng ký nghĩa vụ quân sự là nghĩa vụ theo luật, không phải đăng ký tham gia quân đội, đã đăng ký cũng không có nghĩa là phải đi. Nam công dân đủ mười tám tuổi trong năm đó phải đăng ký trên mạng xong trước ngày 31 tháng 10. Không đăng ký thì chính quyền huyện sẽ yêu cầu sửa trong thời hạn, sửa không thì cưỡng chức buộc thi hành nghĩa vụ quân sự và phạt tiền. Nếu thật sự đi nhập ngũ, quân nhân lính nghĩa vụ nhập ngũ đúng hai năm. +- Lợi ích: Phần đăng ký thì xem《Quy định về công tác đăng ký nghĩa vụ quân sự》《兵役登记工作规定》. Nó viết:「县、自治县、不设区的市、市辖区人民政府兵役机关应当于每年 10 月 31 日前，组织本行政区域内当年 12 月 31 日以前年满 18 周岁的男性公民进行初次兵役登记」, tức cơ quan nghĩa vụ quân sự của chính quyền huyện, huyện tự trị, thành phố không thuộc khu vực và quận phải tổ chức đăng ký nghĩa vụ quân sự lần đầu cho nam công dân trong địa bàn mà trước ngày 31 tháng 12 của năm đó đã đủ 18 tuổi, trước ngày 31 tháng 10 hằng năm. Cùng quy định đó còn viết:「初次兵役登记主要采取网络登记的方式进行，由公民通过兵役登记网络平台自主完成」, tức đăng ký lần đầu chủ yếu làm trên mạng, người dân tự hoàn thành qua cổng đăng ký nghĩa vụ quân sự. Không đăng ký thì sao, xem khoản một Điều năm mươi bảo của Luật nghĩa vụ quân sự. Khoản này quản lý「拒绝、逃避兵役登记的」, tức người từ chối hoặc trốn đăng ký nghĩa vụ quân sự. Hình phạt là「由县级人民政府责令限期改正；逾期不改正的，由县级人民政府强制其履行兵役义务，并处以罚款」, tức chính quyền huyện yêu cầu sửa trong thời hạn; quá hạn không sửa thì chính quyền huyện cưỡng chức buộc thi hành nghĩa vụ quân sự và phạt tiền. Luật nghĩa vụ quân sự còn viết「义务兵服现役的期限为二年」, tức thời hạn quân nhân lính nghĩa vụ nhập ngũ là hai năm. 「根据军队需要和本人自愿，可以征集年满十七周岁未满十八周岁的公民服现役」, tức tuỳ nhu cầu của quân đội và tự nguyện của bản thân, có thể truy xuất công dân đủ mười bảy tuổi chưa đủ mười tám tuổi nhập ngũ. Đối với quân nhân cấp bậc:「服现役满十二年或者符合国家规定的其他条件的，由安置地的县级以上地方人民政府安排工作」, tức đã phục vụ đủ mười hai năm hoặc thỏa mãn điều kiện khác theo quy định của Nhà nước thì chính quyền địa phương cấp huyện trở lên nơi sắp xếp sẽ bố trí việc làm. 「服现役满三十年或者年满五十五周岁或者符合国家规定的其他条件的，作退休安置」, tức đã phục vụ đủ ba mươi năm, hoặc đủ năm mươi lăm tuổi, hoặc thỏa mãn điều kiện khác theo quy định của Nhà nước thì sắp xếp theo chế độ nghỉ hưu. Sắp xếp việc làm nghĩa là chính quyền cho một công việc, sắp xếp nghỉ hưu nghĩa là làm thủ tục nghỉ hưu. 《Quy chế công tác truy xuất quân nhân》《征兵工作条例》viết「优先保证普通高等学校毕业生和对政治、身体条件或者专业技能有特别要求的兵员征集」, tức ưu tiên bảo đảm truy xuất người tốt nghiệp đại học thường và những quân nhân có yêu cầu đặc biệt về chính trị, thể chất hay kỹ năng chuyên môn. Quy chế đó còn viết「应征公民为普通高等学校的全日制在校生、应届毕业生的，可以在入学前户籍所在地或者学校所在地应征」, tức công dân được truy xuất là học sinh toàn thời gian hoặc người sắp tốt nghiệp của trường đại học thường thì có thể nhập ngũ tại nơi có hộ khẩu trước khi nhập học, hoặc tại nơi trường đang học (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Văn phòng Hội đồng Nhà nước, Văn phòng Ủy ban Quân sự Trung ương (国务院办公厅、中央军委办公厅) (2025). Quy định về công tác đăng ký nghĩa vụ quân sự（兵役登记工作规定）（国办发〔2025〕41 号）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2021). Luật nghĩa vụ quân sự（兵役法）（sửa đổi năm 2021，第二十、二十五、二十六、五十四、五十七条）. （đăng trên Bộ Quốc phòng）；Hội đồng Nhà nước, Ủy ban Quân sự Trung ương (国务院、中央军委) (2023). Quy chế công tác truy xuất quân nhân（征兵工作条例）（Hội đồng Nhà nước, Ủy ban Quân sự Trung ương Lệnh số 759，第五条）. +- Ghi chú: Đăng ký và nhập ngũ là hai chuyện, cái giá chênh nhau rất xa. Loạt hình phạt cấm xuất cảnh trong hai năm chỉ áp cho người sau khi truy xuất mà đổi ý, xem mục 3 của phần này (từ chối nghĩa vụ quân sự). Phụ nữ không bắt buộc đăng ký nghĩa vụ quân sự lần đầu, chỉ bị truy xuất tuỳ nhu cầu quân đội. Quân nhân thi đỗ trường quân, quân nhân có thành tích tốt được cấp quân sĩ (tiến lên quân sĩ), điều kiện đều nằm trong quy chế nội bộ của quân đội, ra ngoài tra không được. Luật nghĩa vụ quân sự chỉ viết hai câu: sĩ quan xuất phái chọn từ「表现优秀的现役士兵」, và việc tuyển học viên của trường quân không chịu hạn chế tuổi truy xuất. Cụ thể điểm chấm và số suất, sách này không nêu. + +### 3. Sau khi được truy xuất mà từ chối nghĩa vụ quân sự thì trong hai năm không được xuất cảnh, không được học tiếp, và không vào được cơ quan công chức hay doanh nghiệp nhà nước + +- Chi phí: Không tốn tiền. Cái giá phải trả trước khi đăng ký, vì phải nghĩ rõ có đi hay không. +- Nói thẳng: Đã đăng ký rồi lại từ chối đi nhập ngũ thì hậu quả là một loạt. Không được tuyển làm công chức, doanh nghiệp nhà nước và cơ quan hành chính sự vệ cũng không tuyển bạn. Trong hai năm không được xuất cảnh, không được học tiếp, không được quay lại trường học lại. Bạn còn bị đưa vào danh sách chủ thể mất uy tín nghiêm trọng trong việc thực hiện nghĩa vụ quốc phòng. Vì vậy「cứ đăng ký trước, không được thì rút về」là một ý tưởng đắt đỏ. +- Lợi ích: Khoản một, điểm hai Điều năm mươi bảo của Luật nghĩa vụ quân sự nói về「应征公民拒绝、逃避征集服现役的」, tức công dân được truy xuất từ chối hoặc trốn truy xuất nhập ngũ. Hình phạt cũng là「由县级人民政府责令限期改正；逾期不改正的，由县级人民政府强制其履行兵役义务，并处以罚款」. Cùng điều đó, khoản hai viết:「有前款第二项行为，拒不改正的，不得录用为公务员或者参照《中华人民共和国公务员法》管理的工作人员，不得招录、聘用为国有企业和事业单位工作人员，两年内不准出境或者升学复学，纳入履行国防义务严重失信主体名单实施联合惩戒。」, tức hành vi ở điểm hai của khoản trên mà cứ không chịu sửa thì không được tuyển làm công chức hay làm người lao động được quản lý theo Luật Công chức, không được tuyển hay bổ nhiệm làm viên chức của doanh nghiệp nhà nước và cơ quan hành chính sự vệ, trong hai năm không được xuất cảnh hoặc học tiếp, và bị đưa vào danh sách chủ thể mất uy tín nghiêm trọng trong việc thực hiện nghĩa vụ quốc phòng để trừng phạt liên ngành. Trừng phạt liên ngành nghĩa là nhiều cơ quan dựa vào cùng một danh sách để cùng thực hiện các hạn chế kể trên. Điều năm mươi tám gắn bộ hình phạt đó với người đã nhập ngũ. Quân nhân nhằm trốn nghĩa vụ quân sự mà từ chối thực hiện nhiệm vụ hoặc bỏ trốn khỏi đơn vị thì hậu quả tuỳ kết quả xử lý. Người bị quân đội xóa tên, bị khai trừ quân tịch hoặc bị truy trách nhiệm hình sự thì áp dụng hình phạt tại khoản hai Điều năm mươi bảo. Trong đó người bị xóa tên thì còn bị phạt tiền (toàn quốc, áp dụng từ ngày 1 tháng 10 năm 2021) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2021). Luật nghĩa vụ quân sự（兵役法）（sửa đổi năm 2021，第五十七、五十八条）. （đăng trên Bộ Quốc phòng） +- Ghi chú: Ở đây tính là tự do cá nhân và điều kiện làm việc, không phải tiền. Loạt hình phạt này chỉ áp cho hai loại người. Một là「应征公民拒绝、逃避征集服现役的」. Hai là người nhập ngũ rồi bị xóa tên, bị khai trừ quân tịch, bị truy trách nhiệm hình sự. Người chỉ không làm đăng ký nghĩa vụ quân sự thì không nằm trong đó, đăng ký nghĩa vụ quân sự xem mục 2 của phần này. Điều kiện để phạt là「拒不改正」, do chính quyền huyện xác định và xử phạt. Phạt bao nhiêu thì các nơi tự quy định, sách này không viết. Có đi hay không thì bạn tự quyết, nhưng phải chốt trước khi đăng ký. + +### 4. Hai ưu đãi rõ ràng nhất cho sinh viên đại học đi nhập ngũ: học phí được bù mỗi năm tối đa 20000 tệ, học vị giữ trong hai năm, và có chương trình riêng cho thi cao học + +- Chi phí: Không tốn tiền. Sau khi giải ngũ thì nộp hồ sơ theo yêu cầu của trường, phải chạy một hai chuyến. +- Nói thẳng: Học sinh đang học, người sắp tốt nghiệp và tân sinh viên đi nhập ngũ thì trong thời gian phục vụ, học phí do Nhà nước chi trả. Sinh viên đại học hoặc cao đẳng mỗi người mỗi năm tối đa 20000 tệ, học sau đại học mỗi người mỗi năm tối đa 25000 tệ, hai năm thì nhiều nhất bốn vạn hoặc năm vạn. Học vị được giữ lại cho bạn, trong hai năm sau khi giải ngũ vẫn được nhập học hoặc học lại. Thi cao học lại có một chương trình riêng chỉ tuyển sinh viên đại học đã giải ngũ. +- Lợi ích: Thông báo năm 2024 của Bộ Tài chính và ba bộ khác nói sinh viên các trường đại học đi nghĩa vụ quân sự có ba ưu đãi: bù học phí, hoàn trả khoản vay học phí của Nhà nước, và giảm học phí. Hoàn trả nghĩa là Nhà nước trả thay khoản vay học phí đó. Mức của cả ba được「相应调整为本专科学生每人每年最高不超过 20000 元、研究生每人每年最高不超过 25000 元」, tức điều chỉnh thành mức sinh viên đại học hoặc cao đẳng mỗi người mỗi năm không quá 20000 tệ, học sau đại học mỗi người mỗi năm không quá 25000 tệ. Mức mới áp dụng từ học kỳ mùa thu năm 2024. Bù cụ thể bao nhiêu thì xem câu này:「按学生实际缴纳的学费或获得的国家助学贷款（包括本金及其全部偿还之前产生的利息）两者金额较高者执行」, tức tính theo số tiền lớn hơn trong hai khoản là học phí học sinh thật sự đã đóng hoặc khoản vay học phí của Nhà nước được cấp (kể cả gốc và toàn bộ lãi phát sinh trước khi trả hết). Có thêm câu「超出标准部分不予补偿、代偿或减免」, tức phần vượt quá mức không được bù, hoàn trả hay giảm. Phần học vị thì xem《Luật bảo đảm cho cựu chiến binh》. Nó viết:「现役军人入伍前已被普通高等学校录取或者是正在普通高等学校就学的学生，服现役期间保留入学资格或者学籍，退役后两年内允许入学或者复学，可以按照国家有关规定转入本校其他专业学习」, tức học sinh đã được trường đại học thường tuyển, hoặc đang học ở trường đại học thường, trước khi nhập ngũ thì trong thời gian phục vụ được giữ quyền nhập học hoặc học vị; sau khi giải ngũ, trong hai năm vẫn được nhập học hoặc học lại; và có thể chuyển sang học chuyên ngành khác của cùng trường theo quy định của Nhà nước. Luật đó còn viết:「高等学校根据国家统筹安排，可以通过单列计划、单独招生等方式招考退役军人」, tức trường đại học có thể dùng chỉ tiêu tách riêng, tuyển sinh riêng để tuyển cựu chiến binh theo kế hoạch Nhà nước sắp xếp. Phần thi cao học thì xem thông báo năm 2026 của Văn phòng Bộ Giáo dục. Chương trình tên là「退役大学生士兵计划」, tức chương trình dành cho sinh viên đại học đã giải ngũ. Thông báo viết:「在全国硕士研究生招生计划总规模内单列下达、专项使用，不得调整为普通计划或其他专项计划。招生单位自主确定……初试成绩要求（一般不高于本单位相应学科专业普通计划考生初试成绩要求）……同等条件下服役时间长的考生优先录取」, tức chỉ tiêu được tách riêng trong tổng quy mô tuyển sinh cao học toàn quốc và dùng riêng, không được chuyển thành chỉ tiêu bình thường hay chương trình riêng khác; cơ quan tuyển sinh tự quyết định yêu cầu về điểm kỳ thi đầu (về nguyên tắc không cao hơn yêu cầu điểm kỳ thi đầu của thí sinh chỉ tiêu bình thường cùng ngành chuyên ngành tương ứng); cùng điều kiện thì thí sinh phục vụ lâu hơn được tuyển trước. Tách riêng và dùng riêng nghĩa là phần chỉ tiêu đó được chia ra riêng, chỉ dùng để tuyển sinh viên đại học đã giải ngũ. Năm 2026 có 532 trường đại học thường đảm nhận chương trình này (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Tài chính, Bộ Giáo dục, Ngân hàng Nhân dân Trung Quốc, Cục Quản lý Tài chính (财政部、教育部、中国人民银行、金融监管总局) (2024). Thông báo về việc điều chỉnh hoàn thiện chính sách khoản vay học phí của Nhà nước（关于调整完善国家助学贷款有关政策的通知）（财教〔2024〕188 号）. ；Bộ Tài chính và bốn bộ khác (财政部等五部门) (2019). Quy chế quản lý quỹ hỗ trợ học sinh（学生资助资金管理办法）（财科教〔2019〕19 号，第五条）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2020). Luật bảo đảm cho cựu chiến binh（退役军人保障法）（第三十四、三十五条）. （đăng trên Bộ Quốc phòng）；Văn phòng Bộ Giáo dục (教育部办公厅) (2025). Thông báo về việc làm tốt công tác tuyển sinh theo chỉ tiêu chuyên biệt cho cao học năm 2026 dành cho sinh viên đại học đã giải ngũ（关于做好 2026 年退役大学生士兵专项硕士研究生招生计划招生工作的通知）（教学厅〔2025〕8 号）. ；Bộ Công tác cựu chiến binh (退役军人事务部) (2022). 23 điểm chính sách đào tạo cho quân nhân giải ngũ:「从2022年招生起，高职（专科）毕业生及在校生（含高校新生）应征入伍，退役后完成高职（专科）学业的，申请专升本，免于参加文化课考试」（theo 教学厅〔2021〕8 号）. ；Bộ Giáo dục (教育部) (2026). Quy định quản lý công tác tuyển sinh cao học toàn quốc năm 2027（2027 年全国硕士研究生招生工作管理规定）（教学〔2026〕2 号，第二十二、二十四、二十五条），đăng trên Cổng tuyển sinh cao học. ；Bộ Giáo dục (教育部) (2024). Quy định quản lý công tác tuyển sinh cao học toàn quốc năm 2025（2025 年全国硕士研究生招生工作管理规定）（教学〔2024〕4 号，第二十二条）. +- Ghi chú: Còn hai cửa nữa. Một là lên đại học cử nhân không phải thi môn văn hoá: sinh viên cao đẳng đi nhập ngũ, sau khi giải ngũ học xong cao đẳng thì có thể xin miễn thi môn văn hoá để lên đại học, trường kiểm tra kỹ năng nghề nghiệp rồi tuyển theo thứ tự. Cửa này bắt đầu từ tuyển sinh năm 2022, cách làm cụ thể xem thông báo năm đó của tỉnh. Hai là cộng 10 điểm ở kỳ thi đầu: sinh viên đại học đã giải ngũ mà trong vòng 3 năm sau khi giải ngũ đi thi cao học thì tổng điểm kỳ thi đầu cộng thêm 10 điểm. Đã đăng ký chương trình riêng thì không cộng điểm nữa, hai ưu đãi không cùng hưởng được. Người từng lập công chiến công từ bậc hai trở lên thì có thể xin miễn kỳ thi đầu để học thạc sĩ. Danh sách trường có chương trình riêng thay đổi mỗi năm, năm 2025 là 523 trường. Mức học phí đã điều chỉnh hai lần: năm 2019 là 8000/12000, mùa thu năm 2023 đổi thành 16000/20000, mùa thu năm 2024 đổi thành 20000/25000. Bản tóm tắt năm 2022 của Bộ Công tác cựu chiến binh vẫn ghi bản đầu tiên, đừng chép lại nguyên văn. + +### 5. Sau khi giải ngũ chỉ có sáu trường hợp được chính quyền bố trí việc làm; người tự tìm việc phải đến đăng ký trong vòng 30 ngày + +- Chi phí: Không tốn tiền. Mang theo giấy chứng nhận giải ngũ, đến cơ quan công tác cựu chiến binh ở nơi sắp xếp làm một chuyến. +- Nói thẳng: Chuyện「đi quân được xếp việc」lâu nay đã không còn phổ biến. Giờ chỉ có sáu trường hợp do chính quyền bố trí việc làm, còn quân nhân lính nghĩa vụ phục vụ hai năm rồi giải ngũ thì đa số không nằm trong sáu trường hợp đó. Nhóm không được chính quyền bố trí gọi là tự tìm việc, và có một thời hạn cứng: sau khi rời khỏi quân đội phải đến đăng ký trong vòng 30 ngày. Quá 30 ngày mà không có lý do chính đáng thì coi như bạn tự bỏ chế độ sắp xếp. +- Lợi ích: Điều hai mươi lăm của《Quy chế sắp xếp cựu chiến binh》《退役军人安置条例》liệt kê sáu trường hợp do chính quyền nơi sắp xếp bố trí việc làm. Nguyên văn là:「（一）军士服现役满 12 年的；（二）服现役期间个人获得勋章、荣誉称号的；（三）服现役期间个人荣获三等战功、二等功以上奖励的；（四）服现役期间个人获得一级表彰的；（五）因战致残被评定为 5 级至 8 级残疾等级的；（六）是烈士子女的」, tức một là quân nhân cấp bậc đã phục vụ đủ 12 năm; hai là trong thời gian phục vụ, cá nhân được tặng huân chương hoặc danh hiệu tôn vinh; ba là trong thời gian phục vụ, cá nhân được tuyên dương chiến công cấp ba hoặc được khen thưởng từ cấp hai trở lên; bốn là trong thời gian phục vụ, cá nhân được tuyên dương cấp một; năm là bị thương vì chiến đấu mà xếp hạng khuyết tật từ cấp 5 đến cấp 8; sáu là là con của liệt sĩ. Quy chế đó còn viết:「安置地人民政府应当在接收退役军士和义务兵的 6 个月内完成安排退役军士和义务兵工作的任务」, tức chính quyền nơi sắp xếp phải hoàn thành việc sắp xếp quân nhân cấp bậc và quân nhân lính nghĩa vụ giải ngũ trong vòng 6 tháng kể từ khi tiếp nhận họ. Đơn vị sử dụng lao động「应当按照规定与其签订不少于 3 年的中长期聘用合同或者劳动合同。其中，企业接收军龄 10 年以上的退役军士的，应当与其签订无固定期限劳动合同」, tức phải ký với họ hợp đồng thuê mướn dài hạn hoặc hợp đồng lao động không dưới 3 năm; trong đó, doanh nghiệp tiếp nhận quân nhân cấp bậc có thời gian phục vụ từ 10 năm trở lên thì phải ký hợp đồng lao động không xác định thời hạn. Hợp đồng không xác định thời hạn là hợp đồng không ghi ngày hết hạn. Trong lúc chờ được sắp xếp việc làm thì「按照当地月最低工资标准逐月发放生活补助」, tức mỗi tháng phát trợ cấp sinh hoạt theo mức lương tối thiểu hằng tháng tại địa phương. Người không lên việc đúng quy định vì lý do không phải từ phía mình thì「按照不低于本单位同等条件人员平均工资 80% 的标准，逐月发放生活费直至上岗为止」, tức mỗi tháng phát chi phí sinh hoạt theo mức không thấp hơn 80% mức lương trung bình của người cùng điều kiện trong cùng đơn vị, cho tới khi lên việc. Những trường hợp không được sắp xếp theo hình thức bố trí việc làm gồm bị khai trừ Đảng phái và từng bị kết án hình sự. Thời hạn của nhóm tự tìm việc xem Điều năm mươi mốt:「自主就业的退役军士和义务兵，应当自被批准退出现役之日起 30 日内，到安置地人民政府退役军人工作主管部门报到。无正当理由不按照规定时间报到超过 30 日的，视为放弃安置待遇。」, tức quân nhân cấp bậc và quân nhân lính nghĩa vụ giải ngũ tự tìm việc thì phải từ ngày được duyệt rời khỏi quân đội, đến cơ quan quản lý công tác cựu chiến binh của chính quyền nơi sắp xếp, trong vòng 30 ngày; không có lý do chính đáng mà quá 30 ngày không đến đúng thời hạn quy định thì coi như từ bỏ chế độ sắp xếp (toàn quốc, áp dụng từ ngày 1 tháng 9 năm 2024) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước, Ủy ban Quân sự Trung ương (国务院、中央军委) (2024). Quy chế sắp xếp cựu chiến binh（退役军人安置条例）（Hội đồng Nhà nước, Ủy ban Quân sự Trung ương Lệnh số 787，công bố ngày 29 tháng 7 năm 2024, áp dụng từ ngày 1 tháng 9；第二十五、三十一、三十二、三十五、五十一、七十五条）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2020). Luật bảo đảm cho cựu chiến binh（退役军人保障法）（第十三、十四、十五条）. （đăng trên Bộ Quốc phòng） +- Ghi chú: Đúng ngày quy chế này áp dụng, 《Quy chế sắp xếp quân nhân giải ngũ》năm 2011（国务院、中央军委令第 608 号）bị bãi bỏ. Rất nhiều hướng dẫn trên mạng vẫn dẫn quy chế cũ, cách diễn đạt điều kiện đã thay đổi, bản cũ viết là「平时荣获二等功以上奖励或者战时荣获三等功以上奖励」. Thời hạn 30 ngày chỉ áp cho nhóm tự tìm việc, các hình thức sắp xếp khác thì「按照规定时间」mà đến đăng ký. Khi đăng ký sẽ được cấp thẻ ưu đãi cựu chiến binh, các khoản trợ cấp giải ngũ, hỗ trợ việc làm khởi nghiệp và trợ cấp học vấn về sau đều nối tiếp từ đó. + +### 6. Ba khoản sau khi giải ngũ: tiền trở cấp một lần, thời gian phục vụ tính vào thâm niên công, và tự làm ăn thì mỗi năm giảm 20000 tệ thuế + +- Chi phí: Không tốn tiền. Khi giải ngũ, quân đội chuyển quan hệ bảo hiểm xã hội và tiền của bạn ra. Bạn đến cơ quan bảo hiểm xã hội ở nơi sắp xếp (cửa sổ làm bảo hiểm xã hội) để nối tiếp. +- Nói thẳng: Sau khi giải ngũ có ba khoản. Một là tiền trở cấp một lần cho người tự tìm việc, trả theo thời gian phục vụ, người có công còn được tăng thêm. Hai là mấy năm đi quân không phải khoảng trống của bảo hiểm xã hội, thời gian phục vụ được tính vào thâm niên công, thời gian đóng bảo hiểm cộng với trước nhập ngũ và sau giải ngũ. Ba là thuế: người tự mở cửa hàng kinh doanh thì trong 3 năm, mỗi năm giảm được tối đa 20000 tệ thuế. +- Lợi ích: Khoản thứ nhất xem《Quy chế sắp xếp cựu chiến binh》. Nó viết:「对自主就业的退役军士和义务兵，根据其服现役年限发放一次性退役金」, tức quân nhân cấp bậc và quân nhân lính nghĩa vụ giải ngũ tự tìm việc thì được trả tiền trở cấp một lần tuỳ thời gian phục vụ. Tiền do ai chi trả:「由中央财政专项安排，具体标准由国务院退役军人工作主管部门、中央军事委员会政治工作部门会同国务院财政部门……确定，并适时调整」, tức do ngân sách trung ương sắp xếp riêng, mức cụ thể do cơ quan quản lý cựu chiến binh của Hội đồng Nhà nước và cơ quan chính trị quân sự Trung ương cùng Bộ Tài chính xác định, và điều chỉnh theo thời. Tỷ lệ tăng chia thành năm bậc. Được tặng huân chương hoặc danh hiệu tôn vinh thì tăng 25%, chiến công cấp một hoặc tuyên dương cấp một thì 20%. Chiến công cấp hai, huân chương cấp một hoặc tuyên dương cấp hai thì 15%. Chiến công cấp ba hoặc chiến công cấp hai thì 10%, chiến công cấp bốn hoặc chiến công cấp ba thì 5%. Quy chế đó còn viết:「退役军人未及时就业的，可以依法向户口所在地人力资源社会保障部门申领失业保险待遇，服现役年限视同参保缴费年限，但是以退休、供养方式安置的退役军人除外」, tức cựu chiến binh chưa có việc làm kịp thì có thể theo luật xin trợ cấp bảo hiểm thất nghiệp ở cơ quan bảo hiểm xã hội nơi có hộ khẩu, thời gian phục vụ được tính như thời gian đóng bảo hiểm, trừ cựu chiến binh được sắp xếp theo chế độ nghỉ hưu hoặc nuôi dưỡng. Tính như thời gian đóng bảo hiểm nghĩa là mấy năm đó bạn chưa đóng tiền thì vẫn tính như đã đóng. Khoản thứ hai xem《Luật bảo đảm cho cựu chiến binh》:「退役军人服现役年限计算为工龄，退役后与所在单位工作年限累计计算」, tức thời gian phục vụ được tính vào thâm niên công, sau khi giải ngũ thì cộng dồn với số năm làm việc ở đơn vị đó. Thâm niên công là số năm đã tham gia làm việc, mấy năm phục vụ cộng với các năm sau giải ngũ sẽ cộng tiếp vào. Luật đó còn viết rằng thời gian phục vụ cộng với thời gian đóng bảo hiểm hưu trí, y tế, thất nghiệp trước nhập ngũ và sau giải ngũ thì「依法合并计算」, và「退役的军士和义务兵服现役经历视为基层工作经历」, tức thời gian đóng trước nhập ngũ và sau giải ngũ được cộng hợp pháp, kinh nghiệm phục vụ của quân nhân cấp bậc và quân nhân lính nghĩa vụ giải ngũ được coi là kinh nghiệm làm việc ở cơ sở. 《Luật bảo hiểm quân nhân》:「军人服现役年限视同职工基本医疗保险缴费年限，与入伍前和退出现役后参加职工基本医疗保险的缴费年限合并计算」, tức thời gian phục vụ được tính như thời gian đóng bảo hiểm y tế cơ bản của người lao động, cộng với thời gian đóng trước nhập ngũ và sau khi rời khỏi quân đội. Khoản thứ ba xem thông báo số 14 năm 2023 của Bộ Tài chính, Cục Thuế và Bộ Công tác cựu chiến binh. Thông báo viết về trường hợp quân nhân giải ngũ tự tìm việc làm kinh doanh cá thể:「在 3 年（36 个月，下同）内按每户每年 20000 元为限额依次扣减其当年实际应缴纳的增值税、城市维护建设税、教育费附加、地方教育附加和个人所得税。限额标准最高可上浮 20%」, tức trong 3 năm (36 tháng, các mục sau cũng vậy), theo mức mỗi hộ mỗi năm 20000 tệ, trừ dần vào các khoản thuế mà trong năm đó phải nộp theo thứ tự: thuế giá trị gia tăng, thuế xây dựng và bảo trì đô thị, phụ thu giáo dục, phụ thu giáo dục địa phương và thuế thu nhập cá nhân; mức tối đa có thể nâng lên tới 20%. Trừ dần nghĩa là trừ từ các khoản thuế phải nộp trong năm đó xuống theo thứ tự cho tới hết mức, không phải tiền được phát cho bạn. Đơn vị sử dụng lao động thì「定额标准为每人每年 6000 元，最高可上浮 50%」, tức mức cố định mỗi người mỗi năm 6000 tệ, có thể nâng lên tối đa 50%. Thời hạn áp dụng là từ ngày 1 tháng 1 năm 2023 đến ngày 31 tháng 12 năm 2027. Đến hạn mà chưa đủ 3 năm thì vẫn được hưởng tiếp cho tới khi đủ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước, Ủy ban Quân sự Trung ương (国务院、中央军委) (2024). Quy chế sắp xếp cựu chiến binh（退役军人安置条例）（Hội đồng Nhà nước, Ủy ban Quân sự Trung ương Lệnh số 787，第二十一、二十二、七十四、八十一、八十五条）. ；Bộ Công tác cựu chiến binh (退役军人事务部) (2020). Mười câu hỏi về tiền trở cấp một lần (đăng lại từ《Cựu chiến binh Trung Quốc》số 10 năm 2020):「基本标准。士兵服现役每满一年，发给4500元」「退役金标准适时调整」. ；Sở Công tác cựu chiến binh tỉnh Quý Châu (贵州省退役军人事务厅) (2024). Tăng tiền trở cấp một lần, có thay đổi quan trọng!：「现在执行的标准是：士兵服现役每满一年，发给4500元」. ；Chính quyền quận Mộc Bình, thành phố Yên Đài (烟台市牟平区人民政府) (2025). Trích trong bản trả lời《Thông báo của Sở Công tác cựu chiến binh và Sở Tài chính thành phố Yên Đài về việc điều chỉnh mức trợ cấp kinh tế một lần cho quân nhân giải ngũ tự tìm việc》：「2024年退出现役的每服役一年补助6000元」. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2020). Luật bảo đảm cho cựu chiến binh（退役军人保障法）（第四十二、四十四、五十条）. （đăng trên Bộ Quốc phòng）；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2012). Luật bảo hiểm quân nhân（军人保险法）（第十六、二十三条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Bộ Tài chính, Cục Thuế, Bộ Công tác cựu chiến binh (财政部、税务总局、退役军人事务部) (2023). Thông báo về chính sách thuế hỗ trợ thêm cho quân nhân giải ngũ tự tìm việc khởi nghiệp và làm việc（关于进一步扶持自主就业退役士兵创业就业有关税收政策的公告）（2023 年第 14 号）. +- Ghi chú: Quy chế chỉ viết rằng số tiền do cơ quan chủ quản quyết định riêng. Con số chính thức tra được là câu trong bài đăng năm 2020 trên trang Bộ Công tác cựu chiến binh「服现役每满一年，发给4500元」, cùng bài đó còn viết「适时调整」. Tỷ lệ tăng trong bài đó đã không khớp với quy chế đang có hiệu lực, phải theo quy chế. Bài đăng tháng 8 năm 2024 của Sở Công tác cựu chiến binh tỉnh Quý Châu, dưới quy chế mới, vẫn viết「现在执行的标准是：士兵服现役每满一年，发给4500元」. TODO（cần đối chiếu：không tìm được văn bản tiêu chuẩn chính thức, chỉ xác nhận được tới tháng 8 năm 2024 cơ quan cấp tỉnh vẫn ghi như vậy）. Ngoài ra đừng lẫn「trợ cấp kinh tế một lần」của địa phương với khoản này. Đó là một khoản tiền khác do các nơi tự quy định và tự chi trả, ví dụ Yên Đài năm 2024 trợ cấp mỗi năm phục vụ 6000 tệ. Số mới trong các hướng dẫn trên mạng phần lớn không có nguồn, có cái lại là gộp hai khoản đó vào nhau. Trong thời gian quân nhân lính nghĩa vụ phục vụ, gia đình còn có thể nhận một khoản ưu đãi gia đình riêng. Luật nghĩa vụ quân sự chỉ viết「标准由地方人民政府制定，中央财政给予定额补助」, cụ thể bao nhiêu thì hỏi chính quyền huyện nơi chấp thuận cho bạn nhập ngũ. Khoản thuế là mỗi năm giảm được tối đa bao nhiêu tệ thuế. Nếu việc kinh doanh không lên thì không dùng hết, nên cả mục này chỉ để mức lợi ích「vừa」. 《Quy chế thúc đẩy việc làm và khởi nghiệp cho cựu chiến binh》áp dụng từ ngày 1 tháng 8 năm 2026 bổ sung một điều: chứng chỉ nghề nghiệp của người lao động có tay nghề thi đỗ trong thời gian phục vụ thì sau khi giải ngũ có giá trị trên toàn quốc. + +### 7. Muốn vào hệ thống cơ quan công chức không chỉ có thi cao công chức: làm đủ hai năm ở chương trình phục vụ cơ sở thì dự được kế hoạch tuyển định hướng dành riêng cho 10% mỗi năm + +- Chi phí: Không tốn tiền. Cái giá là hai năm phục vụ ở cơ sở cấp xã. Hai năm đó bạn nhận trợ cấp ngân sách, không phải lương thị trường, chỗ khó là chịu được hai năm ấy. +- Nói thẳng: Muốn vào hệ thống cơ quan công chức thì không chỉ có thi cao công chức. Làm đủ hai năm chương trình phục vụ cơ sở kiểu「三支一扶」, kết quả đánh giá đạt, là dự được kế hoạch tuyển định hướng chỉ dành cho nhóm này. Mỗi năm các tỉnh phải dành khoảng 10％ chỉ tiêu tuyển công chức cho người tham gia các chương trình phục vụ cơ sở kiểu「三支一扶」. Trong ba năm, thi cao học còn được cộng 10 điểm tổng điểm kỳ thi đầu. Cái giá là hai năm đó chỉ nhận trợ cấp ngân sách. +- Lợi ích: Thông báo của kế hoạch「三支一扶」đợt thứ tư viết「服务期为两年」, tức thời gian phục vụ là hai năm. Nó còn viết:「各省（区、市）每年应拿出公务员考录计划的 10％左右，面向『三支一扶』计划等服务基层项目人员定向考录。各省（区、市）县乡基层事业单位公开招聘时，应根据本地区实际拿出一定数量或比例的岗位，对『三支一扶』服务期满考核合格的人员进行专项招聘……聘用后可以不再约定试用期」, tức mỗi năm mỗi tỉnh (khu vực, thành phố) phải dành khoảng 10％ chỉ tiêu tuyển công chức để tuyển định hướng người tham gia chương trình phục vụ cơ sở kiểu「三支一扶」; khi các tỉnh (khu vực, thành phố) tuyển công chức công lập ở cấp huyện, xã, thôn, thì phải dành ra một số lượng hoặc tỷ lệ vị trí thực tế của địa phương để tuyển đặc biệt những người tham gia「三支一扶」đủ hạn và đạt kết quả đánh giá; sau khi bổ nhiệm thì có thể không còn quy định thử việc. Tuyển định hướng và tuyển đặc biệt nghĩa là phần chỉ tiêu đó chỉ thi cho người từng làm chương trình phục vụ cơ sở, người khác không thi được. Cùng thông báo đó còn viết:「期满考核合格的『三支一扶』人员，三年内参加全国硕士研究生招生考试的，初试总分加 10 分，同等条件下优先录取……高职（高专）毕业生期满且考核合格的，可免试入读成人高等学历教育专科起点本科」, tức người tham gia「三支一扶」đủ hạn và đạt kết quả đánh giá thì trong ba năm dự thi tuyển sinh cao học toàn quốc được cộng 10 điểm tổng điểm kỳ thi đầu và cùng điều kiện thì được tuyển trước; tốt nghiệp cao đẳng (cao đẳng nghề) đủ hạn và đạt kết quả đánh giá thì được miễn thi vào học chương trình chính quy bậc đại học của giáo dục tiếp tục sau đại học. Miễn thi vào học nghĩa là không phải thi lại, vào học thẳng bậc đại học. Thông báo còn viết:「参加『三支一扶』计划前无工作经历的人员期满且考核合格的，两年内在参加机关和企事业单位考录（招聘）、自主创业、落户、升学等方面可同等享受应届毕业生相关政策。『三支一扶』人员在基层服务年限计算为工龄，其参加工作时间按其到基层报到之日起算」, tức người trước khi tham gia「三支一扶」chưa có kinh nghiệm làm việc, sau khi đủ hạn và đạt kết quả đánh giá thì trong hai năm được hưởng ngang nhau chính sách dành cho người mới tốt nghiệp trong các việc dự tuyển công chức của cơ quan và đơn vị, tự lập nghiệp, lập hộ khẩu, học tiếp; thời gian phục vụ ở cơ sở được tính vào thâm niên công, thời gian làm việc tính từ ngày đến cơ sở báo cáo. Mức trợ cấp của ngân sách trung ương là khu vực phía Tây mỗi người mỗi năm 3 vạn tệ, khu vực trung 2.4 vạn tệ, khu vực phía Đông 1.2 vạn tệ. Ý kiến của Văn phòng Trung ương Đảng và Văn phòng Hội đồng Nhà nước viết:「参加基层服务项目前无工作经历的人员服务期满且考核合格后 2 年内，在参加机关事业单位考录（招聘）、各类企业吸纳就业、自主创业、落户、升学等方面可同等享受应届高校毕业生的相关政策」, tức người trước khi tham gia chương trình phục vụ cơ sở chưa có kinh nghiệm làm việc, sau khi đủ hạn và đạt kết quả đánh giá thì trong 2 năm được hưởng ngang nhau chính sách dành cho người mới tốt nghiệp đại học trong các việc dự tuyển của cơ quan công chức, các loại doanh nghiệp tiếp nhận người lao động, tự lập nghiệp, lập hộ khẩu, học tiếp. Ý kiến đó còn viết:「省级以上机关录用公务员，除特殊职位外，按照有关规定一律从具有 2 年以上基层工作经历的人员中考录」, tức cơ quan cấp tỉnh trở lên tuyển công chức, trừ các chức vụ đặc biệt, phải tuyển từ người có từ 2 năm kinh nghiệm làm việc ở cơ sở. Một con đường khác là Kế hoạch phía Tây. Nó do「共青团中央、教育部、财政部、人力资源社会保障部共同组织实施」, tức Ban Thanh đoàn Trung ương Đảng, Bộ Giáo dục, Bộ Tài chính và Bộ Nhân sự và Xã hội cùng tổ chức thực hiện. Nó tuyển sinh viên đại học mới tốt nghiệp hoặc học sau đại học đang học, để「到西部基层从事为期 1—3 年」tình nguyện ở cơ sở phía Tây trong thời hạn từ 1 đến 3 năm (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ban Tổ chức Trung ương Đảng Cộng sản Trung Quốc, Bộ Nhân sự và Xã hội và mười bộ (中共中央组织部、人力资源社会保障部等十部门) (2021). Thông báo về việc thực hiện kế hoạch「三支一扶」đợt thứ tư cho sinh viên, học viên tốt nghiệp（关于实施第四轮高校毕业生「三支一扶」计划的通知）（人社部发〔2021〕32 号，第七、八、九、十一、十二、十三条）. ；Văn phòng Trung ương Đảng, Văn phòng Hội đồng Nhà nước (中共中央办公厅、国务院办公厅) (2017). Ý kiến về việc tiếp tục hướng dẫn và khuyến khích sinh viên, học viên tốt nghiệp đi làm việc ở cơ sở（关于进一步引导和鼓励高校毕业生到基层工作的意见）（第五、十六、十八条）. ；Văn phòng quản lý dự án Kế hoạch tình nguyện sinh viên đại học phía Tây (全国大学生志愿服务西部计划项目管理办公室) (2026). Thỏa thuận tuyển sinh Kế hoạch tình nguyện sinh viên đại học phía Tây năm học 2026—2027. +- Ghi chú: Kế hoạch「三支一扶」đợt thứ tư áp dụng cho giai đoạn 2021 đến 2025. Lần này không tìm được văn bản của đợt thứ năm và thông báo năm 2026, trước khi đăng ký hãy tra thông báo tuyển sinh của năm đó. Năm 2025 ngân sách trung ương hỗ trợ tuyển 34430 người. Nguyên văn phương án thực hiện của Kế hoạch phía Tây không tra được trên mạng, trang chính thức mở mục chính sách không ra nội dung. Các ưu đãi sau khi đủ hạn thì theo văn bản của Văn phòng Trung ương Đảng và Văn phòng Hội đồng Nhà nước, vốn quản lý mọi chương trình phục vụ cơ sở. Hai năm đó chỉ nhận trợ cấp ngân sách chứ không phải lương thị trường, cái giá này phải tính cùng với kế hoạch tuyển định hướng sau hai năm. + +### 8. Giáo viên vị trí đặc biệt hết hạn ba năm, đạt kết quả đánh giá và còn muốn ở lại thì phải được chuyển thành biên chế đúng thời hạn + +- Chi phí: Không tốn tiền. Cái giá là ba năm dạy ở trường giáo dục bắt buộc nông thôn, trước khi vào làm còn phải thi lấy giấy chứng nhận giáo viên. Chỗ khó là ở được ba năm ấy. +- Nói thẳng: Kế hoạch vị trí đặc biệt là một trong số ít con đường mà văn bản ghi thẳng「期满入編」, tức hết hạn thì chuyển thành biên chế. Điều kiện là bằng đại học trở lên, chủ yếu là ngành sư phạm, không quá 30 tuổi, và có giấy chứng nhận giáo viên tương ứng. Thời gian phục vụ ba năm, ba năm đó mức lương và đãi ngộ giống giáo viên biên chế của trường công lập địa phương. Hết hạn ba năm, đạt kết quả đánh giá và còn muốn ở lại thì văn bản yêu cầu chuyển biên chế đúng thời hạn. +- Lợi ích: Thông báo kế hoạch vị trí đặc biệt năm 2024 ghi điều kiện tuyển dụng:「要求本科及以上学历，以师范类专业为主，小学阶段可适当招聘师范高等专科学校毕业生」「年龄不超过 30 周岁」, tức yêu cầu bằng đại học trở lên, chủ yếu là ngành sư phạm, ở bậc tiểu học có thể tuyển thêm tốt nghiệp trường cao đẳng sư phạm; tuổi không quá 30. Còn phải có giấy chứng nhận giáo viên tương ứng. Thông báo còn viết:「参加过『大学生志愿服务西部计划』、有从教经历的志愿者和参加过半年以上实习支教的师范院校毕业生同等条件下优先录取」「支持鼓励符合条件的退役军人到中小学任教，可综合考虑服役年限等因素对退役军人相应放宽年龄限制」, tức tình nguyện đã tham gia「Kế hoạch tình nguyện sinh viên đại học phía Tây」hoặc có kinh nghiệm dạy, và sinh viên tốt nghiệp trường sư phạm đã đi thực tập hỗ trợ giảng dạy nửa năm trở lên thì cùng điều kiện được tuyển trước; khuyến khích và hỗ trợ cựu chiến binh đủ điều kiện đến dạy ở tiểu học và trung học, có thể xét tổng hợp thời gian phục vụ để nới hạn tuổi cho họ. Phần đãi ngộ và vào biên chế nằm ở mấy câu này.「从 2024 年 1 月 1 日起，提高特岗教师工资性补助标准，中央财政……中部地区由年人均 3.52 万元提高到 3.88 万元，西部地区由 3.82 万元提高到 4.18 万元」「同等条件下在薪酬待遇、职称评聘、评先评优、年度考核等方面享受与当地公办学校在编教师同等待遇」「保证三年服务期满、考核合格且愿意留任的特岗教师及时入编并落实工作岗位，做好相关人事、工资关系等接转工作，连续计算工龄、教龄，不再实行试用期」, tức từ ngày 1 tháng 1 năm 2024 nâng mức trợ cấp lương cho giáo viên vị trí đặc biệt, ngân sách trung ương ở khu vực trung nâng mức bình quân mỗi người mỗi năm từ 3.52 vạn tệ lên 3.88 vạn tệ, khu vực phía Tây từ 3.82 vạn tệ lên 4.18 vạn tệ; cùng điều kiện thì về lương, chức danh, khen thưởng và đánh giá năm được hưởng đãi ngộ như giáo viên biên chế của trường công lập địa phương; bảo đảm giáo viên vị trí đặc biệt đủ hạn ba năm, đạt kết quả đánh giá và muốn ở lại thì chuyển biên chế đúng thời hạn và được bố trí công việc, làm thủ tục chuyển quan hệ nhân sự và lương, thâm niên công và thâm niên dạy được cộng liên tục, không còn áp dụng thử việc (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Văn phòng Bộ Giáo dục, Văn phòng Bộ Tài chính (教育部办公厅、财政部办公厅) (2024). Thông báo về việc triển khai kế hoạch vị trí giáo viên đặc biệt ở trường giáo dục bắt buộc nông thôn năm 2024（关于做好 2024 年农村义务教育阶段学校教师特设岗位计划实施工作的通知）（教师厅〔2024〕1 号）. +- Ghi chú: Mức lợi ích để「lớn」vì hết hạn mà chuyển biên chế tức là có một công việc biên chế dài hạn, giá trị vượt hẳn mấy vạn tệ. Bản mới nhất tra được là thông báo năm 2024, các năm 2025 và 2026 chưa vào kho chính sách. Vị trí cụ thể, mức trợ cấp và tuổi thì theo thông báo năm đó và phương án thực hiện của tỉnh. Trung ương dựng khung chung, chi tiết do tỉnh quy định. Giấy chứng nhận giáo viên phải thi trước, không thể vào làm rồi mới bổ sung. + +### 9. Lính cứu hộ và nhân sự văn chứng quân đội là hai cửa tuyển công khai khác, ngưỡng thấp hơn thi công chức + +- Chi phí: Không tốn tiền. Từ khi đăng ký đến khi được tuyển phải qua khám sức khoẻ, thẩm tra chính trị, kiểm tra thể lực, kiểm tra tâm lý và phỏng vấn, mất mấy tháng. Chỗ khó là từng chặt một loại người. +- Nói thẳng: Cả hai con đường này đều không chặn ở ngưỡng「trình độ cao đẳng trở lên」. Lính cứu hộ tuyển người 18 đến 22 tuổi, trình độ trung học trở lên; người có trình độ cao đẳng trở lên và quân nhân giải ngũ trong quân đội hay bộ đội cảnh sát vũ trang được nới đến 24 tuổi. Nhân sự văn chứng quân đội chỉ yêu cầu đủ mười tám tuổi, chức vụ cấp bậc thấp tuyển lần đầu tối đa 35 tuổi. Cái giá chung là thẩm tra chính trị và khám sức khoẻ loại bớt người. +- Lợi ích: Phần lính cứu hộ xem《Biện pháp tuyển lính trong đội cứu hộ và cứu nạn tổng hợp quốc gia》. Điều kiện cơ bản ở Điều năm gồm「年龄为 18 周岁以上、22 周岁以下」「具有高中以上文化程度」. Điều sáu:「大学专科以上学历人员、解放军和武警部队退役士兵、具有 2 年以上灭火救援实战经验的政府专职消防队员和政府专职林业扑火队员，年龄可以放宽至 24 周岁；对消防救援工作急需的特殊专业人才，经应急管理部批准年龄可以进一步放宽，原则上不超过 28 周岁」, tức người có trình độ cao đẳng trở lên, quân nhân giải ngũ trong quân đội hay bộ đội cảnh sát vũ trang, lính cứu hộ chuyên trách của chính phủ và lính chuyên trách dập lửa rừng của chính phủ có từ 2 năm kinh nghiệm thực chiến chữa cháy cứu hộ thì nới tuổi đến 24; nhân sự chuyên môn đặc biệt mà công tác cứu hộ cần gấp, được Bộ Môi trường và Tự nhiên phê duyệt thì nới tuổi thêm, về nguyên tắc không quá 28. Điều bảy:「消防员面向社会公开招录，主要从本省级行政区域常住人口中招录，根据需要也可以面向其他省份招录」, tức lính cứu hộ tuyển công khai trước công chúng, chủ yếu từ dân số thường trú trong địa bàn hành chính cùng cấp, tuỳ nhu cầu thì có thể tuyển cả từ các tỉnh khác. Điều chín quy định thủ tục tuyển là「宣传动员、组织报名、资格审查、体格检查、政治考核、体能测试和岗位适应性测试、心理测试和面试、公示、录用」, tức tuyên truyền vận động, tổ chức đăng ký, thẩm tra điều kiện, kiểm tra thể chất, thẩm tra chính trị, kiểm tra thể lực và kiểm tra mức độ phù hợp với vị trí, kiểm tra tâm lý và phỏng vấn, công khai kết quả, tuyển dụng. Phần nhân sự văn chứng quân đội xem Điều mười tám của《Quy chế nhân sự văn chứng Quân giải phóng nhân dân Trung Quốc》:「军队实行公开招考、直接引进、专项招录相结合的文职人员招录聘用制度」, tức quân đội thực hiện chế độ tuyển dụng nhân sự văn chứng kết hợp thi đấu công khai, giới thiệu trực tiếp và tuyển đặc biệt. Đó là ba đường: thi công khai để lấy người, giới thiệu trực tiếp người có sẵn, và tuyển riêng cho nhóm đối tượng xác định. Điều kiện cơ bản ở Điều chín gồm「年满 18 周岁」「符合军队招录聘用文职人员的政治条件」「符合岗位要求的文化程度、专业水平和工作能力」, tức đủ mười tám tuổi, đáp ứng điều kiện chính trị để quân đội tuyển nhân sự văn chứng, và có trình độ văn hóa, trình độ chuyên môn, năng lực làm việc phù hợp với yêu cầu vị trí. Điều mười chín ghi rõ đối với chức vụ cấp bậc thấp thì tuổi tối đa khi tuyển lần đầu là 35 tuổi. Quy chế đó còn ghi「文职人员岗位应当优先招录聘用符合条件的退役军人」, tức vị trí nhân sự văn chứng phải ưu tiên tuyển và bổ nhiệm cựu chiến binh đủ điều kiện (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Nhân sự và Xã hội, Bộ Môi trường và Tự nhiên (人力资源社会保障部、应急管理部) (2021). Biện pháp tuyển lính trong đội cứu hộ và cứu nạn tổng hợp quốc gia（国家综合性消防救援队伍消防员招录办法）（人社部发〔2021〕58 号，第五、六、七、九条）. ；Hội đồng Nhà nước, Ủy ban Quân sự Trung ương (国务院、中央军委) (2022). Quy chế nhân sự văn chứng Quân giải phóng nhân dân Trung Quốc（中国人民解放军文职人员条例）（sửa đổi lần hai, Hội đồng Nhà nước, Ủy ban Quân sự Trung ương Lệnh số 757，áp dụng từ ngày 1 tháng 1 năm 2023；第九、十八、十九、二十条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia） +- Ghi chú: 《Quy chế thúc đẩy việc làm và khởi nghiệp cho cựu chiến binh》áp dụng từ ngày 1 tháng 8 năm 2026 nêu thêm hai việc. Một là tuyển lính cứu hộ phải dành một tỷ lệ nhất định, chuyên tuyển quân nhân cấp bậc và quân nhân lính nghĩa vụ giải ngũ. Hai là khi tuyển lực lượng trợ giúp công an (cảnh sát phụ), cùng điều kiện thì ưu tiên tuyển cựu chiến binh. Số suất, vị trí và điều kiện cụ thể hằng năm của hai con đường này đều theo thông báo năm đó, sách này chỉ viết những ngưỡng cố định không đổi trong quy chế. Lính cứu hộ là vị trí trong đội kỷ luật, trước khi vào hãy nghĩ kỹ rủi ro nghề nghiệp. + +### 10. Bằng cấp không phải việc mua một lần là xong: thi tự học không giới hạn tuổi cũng không xem trình độ đã học, thi lớn và đại học mở đều bổ sung được + +- Chi phí: Phí đăng ký và giáo trình mỗi năm mấy trăm tệ. Cái giá là học tự thọn lâu dài, hoặc đi học buổi tối; bậc đại học học song song với việc làm (không rời việc) tối thiểu phải học hai năm rưỡi. Chỗ khó là phải kiên trì lâu dài. +- Nói thẳng: Năm mười tám tuổi không học được thì phía sau vẫn còn ba cửa chính thức: thi tự học, thi lớn cho người lớn, và đại học mở. Thi tự học không giới hạn tuổi, cũng không xem bạn đã học tới bậc nào, nhưng học sinh toàn thời gian thì không được thi. Thi lớn cho người lớn thì phải có bằng tốt nghiệp trung học hoặc cao đẳng. Cả ba đường đều chậm, nhưng không đường nào xem kết quả năm mười tám tuổi của bạn thế nào. +- Lợi ích: Thi tự học thì xem《Quy chế lâm thời về thi tự học bậc đại học》. Điều ba:「中华人民共和国公民，不受性别、年龄、民族、种族和已受教育程度的限制，均可依照本条例的规定参加高等教育自学考试」, tức công dân Trung Quốc, không phân biệt giới tính, tuổi, dân tộc, chủng tộc và trình độ đã được giáo dục, đều có thể dự thi tự học bậc đại học theo quy định của quy chế này. Điều hai mươi:「各级各类全日制学校的在校生不得报考」, tức học sinh của mọi loại trường toàn thời gian không được dự thi. Điều hai mươi lăm nêu ba điều kiện để được cấp bằng: học xong và đạt tất cả các môn, hoàn thành luận văn tốt nghiệp hoặc nhiệm vụ thực hành, và đạt kết quả nhận xét về đạo đức. Làm đủ cả ba thì được cấp chứng chỉ tốt nghiệp. Điều đó còn viết「获得专科（基础科）或本科毕业证书者，国家承认其学历」, tức ai có bằng cao đẳng (phần cơ sở) hoặc bằng đại học thì Nhà nước công nhận trình độ đó. Điều ba mươi hai nêu mức lương của người có bằng:「非在职人员录用后，与普通高等学校同类毕业生相同」, tức người không đang làm việc sau khi được tuyển thì như người tốt nghiệp cùng loại của trường đại học thường. Điều hai mươi mốt của《Luật Giáo dục bậc cao》:「国家实行高等教育自学考试制度，经考试合格的，发给相应的学历证书或者其他学业证书」, tức Nhà nước thực hiện chế độ thi tự học bậc đại học, ai đỗ thì được cấp chứng chỉ trình độ hoặc chứng chỉ học vụ tương ứng. Thi lớn cho người lớn thì xem《Cách thức tuyển sinh các trường đại học dành cho người lớn toàn quốc năm 2025》. Đối tượng dự thi là「国家承认学历的各类高、中等学校在校生以外的从业人员和社会其他人员」, tức người lao động và người dân khác ngoài học sinh các trường trung học, trung cấp chính quy thừa nhận trình độ của Nhà nước. 「报考高起本或高起专的考生应高级中等教育学校毕业或者具有同等学力。报考专升本的考生必须是已取得……专科毕业证书、本科结业证书或以上证书的人员」, tức thí sinh dự thi khóa đại học hoặc cao đẳng bắt đầu từ trung học thì phải tốt nghiệp trường trung học phổ thông hoặc có trình độ tương đương; thí sinh dự thi bậc đại học phải là người đã có bằng cao đẳng, bằng hoàn thành bậc đại học hoặc giấy chứng nhận cao hơn. 「脱产最短学习时间为：高起本四年、高起专和专升本两年，非脱产最短学习时间为：高起本五年、高起专和专升本两年半」, tức thời gian học ngắn nhất khi rời việc học: khóa đại học bắt đầu từ trung học bốn năm, khóa cao đẳng bắt đầu từ trung học và khóa đại học hai năm; khi học song song với việc làm: khóa đại học bắt đầu từ trung học năm năm, khóa cao đẳng bắt đầu từ trung học và khóa đại học hai năm rưỡi. Rời việc học là bỏ việc ra trường học cả ngày, không rời việc là vừa đi làm vừa học. Đại học mở thì xem《Ý kiến của Bộ Giáo dục về việc làm tốt đại học mở》. Nó viết:「开放大学实行注册入学，学生修完课程并获得相应学分，即可获得课程证书，修满规定的学分并达到相关要求，即可获得相应的学历与非学历证书，符合学位授予条件的应授予相应学位」, tức đại học mở nhận đăng ký theo kiểu đăng ký trước, học sinh học xong môn và đạt đủ số tín chỉ thì có chứng chỉ môn học, tích lũy đủ số tín chỉ quy định và đạt các yêu cầu liên quan thì có chứng chỉ trình độ và chứng chỉ không phải trình độ, ai đủ điều kiện cấp bằng thì được cấp bằng tương ứng (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (1988, sửa đổi 2014). Quy chế lâm thời về thi tự học bậc đại học（高等教育自学考试暂行条例）（第三、二十、二十五、三十二条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2018). Luật Giáo dục bậc cao（高等教育法）（第二十一条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Văn phòng Bộ Giáo dục (教育部办公厅) (2025). Thông báo về việc làm tốt công tác tuyển sinh các trường đại học dành cho người lớn toàn quốc năm 2025（关于做好 2025 年全国成人高校招生工作的通知）（教学厅〔2025〕6 号）kèm phụ lục《Cách thức tuyển sinh các trường đại học dành cho người lớn toàn quốc năm 2025》. ；Bộ Giáo dục (教育部) (2016). Ý kiến về việc làm tốt đại học mở（关于办好开放大学的意见）（教职成〔2016〕2 号）. ；Bộ Giáo dục (教育部) (2007). Thông báo về việc làm tốt hơn nữa việc quản lý các hình thức tuyển sinh của trường đại học（关于进一步做好高等学校各类招生管理工作的通知）（教发〔2007〕13 号）：「普通专升本招生计划按不超过当年应届普通高职（专科）毕业生5%的比例安排」. ；Cổng Chính phủ Trung Quốc (中国政府网) (2020). Năm nay mở rộng cao học 18.9 vạn, đại học cử nhân 32.2 vạn, hỗ trợ việc làm cho sinh viên tốt nghiệp, Phó Bộ trưởng Bộ Giáo dục Ô Thiết Hoa：「扩大普通高校专升本的规模，同比争取增加32.2万人」. ；Sở Khảo thử tỉnh Phúc Kiến (福建省教育考试院) (2026). Khi chuyển tiếp có dẫn「教育部高校学生司《关于做好2026年普通高等学校专升本考试招生工作的通知》（教学司函〔2025〕53号）」. +- Ghi chú: Còn một hình thức đại học cử nhân dành cho sinh viên tốt nghiệp cao đẳng toàn thời gian. Văn bản công khai ở cấp quốc gia là thông báo năm 2007 của Bộ Giáo dục, trong đó ghi kế hoạch đại học cử nhân không quá 5% số sinh viên tốt nghiệp cao đẳng của năm đó. Từ năm 2020 Bộ Giáo dục mở rộng quy mô đại học cử nhân, năm đó tăng thêm 32.2 vạn người so với năm trước, mức 5% thực tế không còn được áp dụng. Công tác cụ thể hằng năm do Cục Quản lý Sinh viên trường đại học của Bộ Giáo dục phát công văn, năm 2026 là 教学司函〔2025〕53 号. Công văn đó không công bố ra ngoài, chỉ thấy được số văn bản trong các văn bản mà tỉnh chuyển tiếp. Đối tượng tuyển do từng tỉnh tự quy định, về cơ bản chỉ nhận tốt nghiệp cao đẳng toàn thời gian của chính tỉnh trong năm đó, nên phần nội dung không viết điều kiện của nó. Từ mùa thu năm 2025, giáo dục tiếp tục bậc đại học không còn gọi là「函授」「业余」mà thống nhất gọi là「非脱产」. Giáo dục tiếp tục bậc đại học chính là loại học bổ sung trình độ sau này, còn dùng tên cũ trong quảng cáo thì phải cảnh giác. Bổ sung trình độ và「học kỹ năng gì thì đáng tiền」là hai việc khác nhau, về sau xem phần 23. Chứng chỉ bán「lấy nhanh」và chứng chỉ chếch mạch xem mục 8 ở phần 23. + +### 11. Không vào cơ quan công chức thì làm việc linh hoạt: hưu trí và y tế phải tự tham gia bảo hiểm ở nơi làm việc, hạn chế về hộ khẩu đã được nới + +- Chi phí: Hưu trí và y tế bạn phải đóng toàn bộ, tính theo mức đóng của địa phương, mỗi tháng hơn nghìn tệ. Mức đóng là con số dùng làm căn cứ tính phí bảo hiểm. Đây là khoản chi cố định dễ bị tính sót nhất khi làm việc linh hoạt, chỗ khó là đều đặn đóng hằng tháng. +- Nói thẳng: Không vào cơ quan không có nghĩa không còn chuyện bảo hiểm xã hội, chỉ là không ai đóng giúp bạn nửa kia. Văn bản yêu cầu các nơi nới hạn chế về hộ khẩu đối với người làm việc linh hoạt khi tham gia bảo hiểm ở nơi làm việc, vài thành phố rất lớn có thể nới dần. Vì vậy đừng hiểu「tự do」là không phải lo bảo hiểm xã hội. Để hở mà không đóng thì lương hưu và bảo hiểm y tế đều phải tính theo luật của người đóng gián đoạn. +- Lợi ích: Phần tham gia bảo hiểm thì xem《Ý kiến chỉ đạo về việc bảo vệ quyền lợi lao động của người lao động có hình thức việc làm mới》. Nó viết:「各地要放开灵活就业人员在就业地参加基本养老、基本医疗保险的户籍限制，个别超大型城市难以一步实现的，要结合本地实际，积极创造条件逐步放开。组织未参加职工基本养老、职工基本医疗保险的灵活就业人员，按规定参加城乡居民基本养老、城乡居民基本医疗保险，做到应保尽保。督促企业依法参加社会保险。企业要引导和支持不完全符合确立劳动关系情形的新就业形态劳动者根据自身情况参加相应的社会保险」, tức các nơi phải nới hạn chế về hộ khẩu đối với người làm việc linh hoạt khi tham gia bảo hiểm hưu trí cơ bản và bảo hiểm y tế cơ bản ở nơi làm việc; thành phố rất lớn khó mở ngay trong một bước thì phải căn cứ thực tế địa phương, tích cực tạo điều kiện để nới dần; tổ chức người làm việc linh hoạt chưa tham gia bảo hiểm hưu trí cơ bản và bảo hiểm y tế cơ bản của người lao động thì tham gia bảo hiểm hưu trí cơ bản và bảo hiểm y tế cơ bản của dân cư địa phương theo quy định, đảm bảo ai cũng được tham gia; thúc ép doanh nghiệp tham gia bảo hiểm xã hội theo luật; doanh nghiệp phải hướng dẫn và hỗ trợ người lao động có hình thức việc làm mới không hoàn toàn thuộc trường hợp xác lập quan hệ lao động tham gia bảo hiểm xã hội tương ứng theo tình hình thực tế của họ. Cùng văn bản đó chia việc dùng lao động thành ba loại:「符合确立劳动关系情形的，企业应当依法与劳动者订立劳动合同。不完全符合确立劳动关系情形但企业对劳动者进行劳动管理……指导企业与劳动者订立书面协议……个人依托平台自主开展经营活动、从事自由职业等，按照民事法律调整双方的权利义务」, tức thuộc trường hợp xác lập quan hệ lao động thì doanh nghiệp phải ký hợp đồng lao động với người lao động theo luật; không hoàn toàn thuộc trường hợp xác lập quan hệ lao động nhưng doanh nghiệp có quản lý lao động thì hướng dẫn doanh nghiệp ký thỏa thuận bằng văn bản với người lao động; người cá nhân dựa vào sàn tự chủ kinh doanh, làm nghề tự do thì quyền và nghĩa vụ của hai bên được điều chỉnh theo luật dân sự (toàn quốc, từ năm 2021) +- Mức bằng chứng: A +- Nguồn: Bộ Nhân sự và Xã hội và tám bộ (人力资源社会保障部等八部门) (2021). Ý kiến chỉ đạo về việc bảo vệ quyền lợi lao động của người lao động có hình thức việc làm mới（关于维护新就业形态劳动者劳动保障权益的指导意见）（人社部发〔2021〕56 号，第二、八条）. ；Văn phòng Hội đồng Nhà nước (国务院办公厅) (2020). Ý kiến về việc hỗ trợ việc làm linh hoạt theo nhiều kênh（关于支持多渠道灵活就业的意见）（国办发〔2020〕27 号）. +- Ghi chú: Nguyên văn là「放开」hạn chế về hộ khẩu và vẫn để lỗ thông cho các thành phố rất lớn, không phải「toàn quốc đã bỏ hẳn」. Đến mức nào thì hỏi thành phố mà bạn sẽ tham gia bảo hiểm. Mức lợi ích để「lớn」vì đóng gián đoạn thì mất chính các chế độ hưu trí và y tế, giá trị vượt hẳn mấy vạn tệ. Nhưng chi phí cũng rõ ràng, tự đóng toàn bộ mỗi tháng hơn nghìn tệ. Sau khi đóng gián đoạn bảo hiểm xã hội thì khắc phục thế nào, số năm cộng dồn ra sao, xem mục 18 ở phần 7 (đóng gián đoạn bảo hiểm xã hội). Ngoài ra, bán nông sản, hàng tiêu dùng ở địa điểm và khung giờ do chính quyền quy định, hoặc dùng tay nghề của mình làm việc lợi ích không cần giấy phép, thì không phải đăng ký giấy phép kinh doanh. Điều này nằm trong 国办发〔2020〕27 号. + +### 12. Giao đồ ăn, chạy xe ô tô công nghệ, chở hàng trong thành phố: sàn trả phí bảo hiểm tai nạn lao động theo từng đơn, bạn không phải đóng + +- Chi phí: Không tốn tiền, khoản tiền đó sàn đóng. Việc bạn phải làm là xem sàn mình nhận việc có nằm trong phạm vi thí điểm không. Có hai cách: tìm「职业伤害保障」trong trang bảo vệ hoặc quyền lợi của ứng dụng, hoặc gọi điện cho cơ quan bảo hiểm xã hội của tỉnh hỏi xem sàn đó đã đăng ký chưa. +- Nói thẳng: Người giao đồ ăn, chạy xe ô tô công nghệ, chở hàng trong thành phố thì sàn phải đóng phí bảo hiểm tai nạn lao động theo từng đơn thay bạn, bạn không phải đồng nào, nhận đơn nào thì được bảo hiểm đơn đó. Đây vẫn là thí điểm, trước khi nhận việc hãy xem tỉnh của bạn và sàn có nằm trong phạm vi không. Người đã có quan hệ lao động với sàn thì theo bảo hiểm tai nạn lao động, không theo bộ này. +- Lợi ích: Chế độ bảo đảm này xem《Biện pháp bảo hiểm tai nạn lao động cho người lao động có hình thức việc làm mới (thử nghiệm)》. Điều hai viết:「平台企业应当依照本办法规定参加新就业形态人员职业伤害保障……为通过平台注册并接单，以平台企业名义提供出行、即时配送和同城货运等劳动并获得报酬或者收入的新就业形态人员缴纳职业伤害保障费，实现每单必保、每人必保。适用《工伤保险条例》规定的劳动者，企业应当依法为其参加工伤保险，不适用本办法」, tức doanh nghiệp sàn phải theo quy định của biện pháp này tham gia bảo hiểm tai nạn lao động cho người lao động có hình thức việc làm mới; đóng phí bảo hiểm tai nạn lao động cho người lao động đăng ký qua sàn và nhận đơn, cung cấp lao động vận tải, giao hàng tức thì và chở hàng trong thành phố dưới danh nghĩa doanh nghiệp sàn và nhận thù lao hoặc thu nhập, đạt mỗi đơn đều có bảo hiểm, mỗi người đều có bảo hiểm; người lao động thuộc phạm vi Quy chế bảo hiểm tai nạn lao động thì doanh nghiệp phải cho họ tham gia bảo hiểm tai nạn lao động theo luật, không áp dụng biện pháp này. Điều chín:「试行期间，出行行业按照每单 0.01 元执行；即时配送行业按照每单 0.07 元、0.25 元执行；同城货运行业按照每单 0.18 元执行」, tức trong thời gian thử nghiệm, ngành vận tải tính 0.01 tệ mỗi đơn; ngành giao hàng tức thì tính 0.07 tệ và 0.25 tệ mỗi đơn; ngành chở hàng trong thành phố tính 0.18 tệ mỗi đơn. Phần mở rộng thí điểm thì xem đoạn này:「2025 年 7 月 1 日起，在已先行试点 7 个省份和 7 个平台企业的基础上……增加天津、河北、辽宁、浙江、安徽、福建、山东、湖北、广西、宁夏等 10 个省份……2026 年，推动职业伤害保障试点在全国 31 个省份和新疆生产建设兵团实施；将出行、即时配送和同城货运三个行业的平台企业总体纳入职业伤害保障试点范围」, tức từ ngày 1 tháng 7 năm 2025, trên cơ sở đã thí điểm trước ở 7 tỉnh và 7 doanh nghiệp sàn thì bổ sung thêm 10 tỉnh gồm Thiên Tân, Hà Bắc, Liêu Niinh, Chiêu An, Phúc Kiến, Sơn Đông, Hồ Bắc, Quảng Tây, Ninh Hạ; năm 2026 thúc đẩy thí điểm bảo hiểm tai nạn lao động triển khai ở cả 31 tỉnh và Tân Cương Binh đoàn Xây dựng Thủ công; đưa doanh nghiệp sàn của cả ba ngành vận tải, giao hàng tức thì và chở hàng trong thành phố vào phạm vi thí điểm bảo hiểm tai nạn lao động (thí điểm, mở rộng dần theo từng năm) +- Mức bằng chứng: A +- Nguồn: Bộ Nhân sự và Xã hội và chín bộ (人力资源社会保障部等九部门) (2025). Thông báo về việc mở rộng thí điểm bảo hiểm tai nạn lao động cho người lao động có hình thức việc làm mới（关于扩大新就业形态人员职业伤害保障试点的通知）（人社部发〔2025〕24 号）kèm phụ lục《Biện pháp bảo hiểm tai nạn lao động cho người lao động có hình thức việc làm mới (thử nghiệm)》第二、九条. +- Ghi chú: Đây vẫn là thí điểm, chưa phải chế độ chính thức có ở toàn quốc. Trước khi nhận việc hãy xác nhận tỉnh của bạn và sàn có nằm trong phạm vi không; nếu không thì chỉ còn lớp bảo hiểm thương mại mua bằng tiền của bạn. Một văn bản khác là《Ý kiến chỉ đạo về việc bảo vệ quyền lợi lao động của người lao động có hình thức việc làm mới》còn nêu thêm vài yêu cầu. Sàn phải trả mức thù lao không thấp hơn lương tối thiểu địa phương cho người làm việc đúng quy định, phải trả đủ và đúng hạn, không được khấu trừ hoặc nợ đãng không lý do. Ngày lễ theo luật còn phải trả mức thù lao hợp lý. Số văn bản của tài liệu này là 人社部发〔2021〕56 号, mục làm việc linh hoạt cũng dẫn nó, xem mục 11 của phần này. Khi có sự cố thì làm theo bộ này trước, đừng tự thỏa thuận riêng với điểm giao. Phần bảo hiểm y tế và hưu trí vẫn phải tự tham gia, xem mục 11 của phần này. + +### 13. Hai lối chốt được biên chế ngay ngày điền nguyện vọng thi đại học: sinh viên sư phạm hưởng kinh phí và sinh viên y khoa định hướng, cái giá là 6 năm thực hiện nghĩa vụ + +- Chi phí: Không tốn tiền. Học phí và tiền ăn ở do nhà nước trả, còn được trợ cấp sinh hoạt. Cái giá là sau khi tốt nghiệp, 6 năm không tự do chọn việc, vi phạm thì phải hoàn tiền, nộp tiền phạt và bị ghi vào lịch sử tín dụng. Chỗ khó là 6 năm đó không đổi được nơi. +- Nói thẳng: Cả hai lối đều là nhà nước trả trước tiền cho bạn, sau khi tốt nghiệp bạn đi làm đúng chỗ đã ghi trong thỏa thuận trong 6 năm. Sinh viên sư phạm hưởng kinh phí chỉ tuyển ở sáu trường đại học sư phạm trực thuộc Bộ Giáo dục, nhận ở đợt xét tuyển sớm. Sinh viên y khoa định hướng phải ký thỏa thuận ba bên trước khi nhận được giấy báo nhập học. Vi phạm thì phải hoàn tiền, nộp tiền phạt, và còn bị ghi vào lịch sử tín dụng. Nghĩ kỹ có chịu được 6 năm đó không rồi hãy điền nguyện vọng. +- Lợi ích: Sinh viên sư phạm hưởng kinh phí xem Biện pháp giáo dục kinh phí cho sinh viên sư phạm liên thông từ đại học lên sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục《教育部直属师范大学本研衔接师范生公费教育实施办法》（năm 2024, biện pháp năm 2018 bị bãi bỏ đồng thời）. Điều hai: ngân sách trung ương chịu học phí và tiền ăn ở trong thời gian học sinh viên đang học, đồng thời cấp trợ cấp sinh hoạt. Điều ba: chủ yếu đào tạo cho các địa (thành phố, châu, liên minh) trở xuống, ngoài thủ phủ tỉnh ở vùng trung tây và vùng tây. Biện pháp còn ghi「不得定向到直辖市、计划单列市或省会城市主城区任教（五个自治区、陆地边境省份、海南省、贵州省、青海省除外）」, tức không được định hướng đi dạy ở thành phố trực thuộc trung ương, thành phố tách riêng theo kế hoạch hay khu trung tâm thủ phủ tỉnh (trừ năm khu tự trị, các tỉnh giáp biên đất liền, tỉnh Hải Nam, tỉnh Quý Châu, tỉnh Thanh Hải). Điều chín: sau khi tốt nghiệp thì về chung địa (thành phố, châu, liên minh) định hướng theo tỉnh có hộ khẩu để dạy ở trường tiểu học và trung học,「承诺从事中小学教育工作 6 年以上」, tức cam kết làm công tác giáo dục tiểu học và trung học trên 6 năm. Ai làm việc ở trường đô thị thì「应到农村义务教育学校任教服务至少 1 年」, tức phải đi dạy phục vụ tại trường giáo dục bắt buộc nông thôn ít nhất 1 năm. Điều mười ba: không thực hiện nghĩa vụ hoặc thời hạn thực hiện không đủ thì chịu trách nhiệm theo thỏa thuận. Một trong các hậu quả là「违约行为纳入信用记录，并及时与全国信用信息共享平台共享，违约相关材料归入本人人事档案」, tức hành vi vi phạm được đưa vào lịch sử tín dụng, kịp thời chia sẻ với nền tảng chia sẻ thông tin tín dụng toàn quốc, hồ sơ liên quan tới vi phạm đưa vào hồ sơ nhân sự của bản thân. Câu nữa là「履约任务完成前不得到定向地（市、州、盟）之外从教，不得被招录（聘用）为其他公务人员。完成履约任务后即修复相关信用」, tức chưa hoàn thành nghĩa vụ thì không được đi dạy ngoài địa (thành phố, châu, liên minh) định hướng, không được tuyển dụng (thuê mời) làm công chức khác; hoàn thành nghĩa vụ thì khôi phục lại uy tín tín dụng liên quan. Trong thời gian chưa đủ thời hạn dịch vụ thỏa thuận thì không xin được chức vụ công lập khác; đủ thời hạn thì lịch sử tín dụng mới được khôi phục. Điều sáu: sau khi nhận tuyển, nếu qua khảo sát thấy không phù hợp để đi dạy thì trong với 1 năm trả lại khoản kinh phí giáo dục đã hưởng và chuyển sang chuyên ngành không phải sư phạm. Điều mười bốn: không có giấy chứng nhận trình độ và bằng thạc sĩ sư phạm cùng giấy chứng nhận giáo viên trong thời hạn học quy định thì xử lý như vi phạm, hoàn tiền và nộp tiền phạt. Điều hai mươi mốt: các nơi giải quyết biên chế cần thiết để sinh viên sư phạm hưởng kinh phí đến dạy ở trường tiểu học và trung học, trong tổng số biên chế đơn vị sự nghiệp hiện có. Sinh viên y khoa định hướng miễn phí đào tạo theo đơn đặt của nông thôn xem một thông báo khác. Sau khi nhận tuyển, trước khi nhận được giấy báo nhập học, phải ký thỏa thuận với trường đào tạo và với sở y tế, sở nhân sự và xã hội cấp huyện nơi định hướng việc làm. Trong thỏa thuận phải「承诺毕业后到定向农村基层医疗卫生机构服务 6 年」, tức cam kết sau khi tốt nghiệp phục vụ 6 năm tại cơ sở y tế cơ sở nông thôn định hướng. Thực hiện nghĩa vụ là đi đủ thời hạn ở nơi đã ghi trong thỏa thuận. Sau khi thực hiện nghĩa vụ và báo cáo đi làm, phải học 3 năm đào tạo chuyên khoa bác sĩ nội trú theo tiêu chuẩn hoá toàn diện, gọi tắt là đào tạo chuẩn hoá. Đào tạo chuẩn hoá là giai đoạn sau khi tốt nghiệp, ở trong bệnh viện tiếp tục rèn tay nghề theo tiêu chuẩn thống nhất. Trong thời gian đào tạo, quan hệ nhân sự và lương cũ giữ nguyên. Hồ sơ vi phạm đưa vào hồ sơ nhân sự cá nhân, và là căn cứ đánh giá uy tín khi tuyển dụng (tuyển việc) công khai ở cơ quan hành chính, đơn vị sự nghiệp như cơ sở y tế, khi nhận đào tạo chuyên khoa bác sĩ nội trú, khi nhận học cao học. Người cố ý vi phạm sẽ bị công bố toàn quốc trên trang「信用中国」, và chịu trừng phạt chung vì mất uy tín. Thời gian đào tạo bị vi phạm thì không tính vào thời hạn phục vụ. Người không hoàn tiền và tiền phạn theo thỏa thuận thì cơ quan y tế cấp huyện truy thu qua đường pháp lý (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Văn phòng Hội đồng Nhà nước (2024). Thông báo chuyển Biện pháp giáo dục kinh phí cho sinh viên sư phạm liên thông từ đại học lên sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục và các cơ quan（国办发〔2024〕27 号，điều hai, ba, sáu, chín, mười ba, mười bốn, hai mươi mốt）. ；Ủy ban Quốc gia Y tế và Sức khoẻ, Văn phòng Tổ chức cán bộ trung ương, Ủy ban Phát triển và Cải cách quốc gia, Bộ Giáo dục, Bộ Tài chính, Bộ Nhân sự và Xã hội, Cục Quản lý y dược trung ương (2019). Thông báo về việc làm tốt công tác bố trí việc làm và quản lý nghĩa vụ thực hiện cho sinh viên y khoa định hướng miễn phí đào tạo theo đơn đặt của nông thôn. +- Ghi chú: Biện pháp năm 2024 đổi sinh viên sư phạm hưởng kinh phí thành「本研衔接」, nghĩa là học tiếp thạc sĩ sư phạm ngay sau đại học. Biện pháp cũ năm 2018 đã bị bãi bỏ, câu「tốt nghiệp đại học là dạy thẳng」còn truyền trên mạng là cách nói cũ. Đã ký thỏa thuận trước khi biện pháp ban hành thì theo thỏa thuận cũ.「Có biên chế」không có nghĩa「vào làm rồi thì không phải lo gì」: biên chế phải bù ra từ tổng số hiện có, vị trí cụ thể do thành phố cấp huyện tổ chức tuyển riêng cho trường sử dụng, bạn và trường cùng chọn nhau. Sinh viên y khoa định hướng hết thời hạn phục vụ mà muốn đi thì được, văn bản ghi「服务期满，对自主择业的应予同意」, tức hết thời hạn phục vụ thì phải chấp thuận cho người tự chọn việc. Người đã vi phạm cũng có cách xoá hồ sơ vi phạm, cái giá là làm đủ theo thỏa thuận cũ, hoặc chuyển sang cơ sở khác ở nông thôn trong tỉnh rồi làm đủ ít nhất 6 năm, tiền và tiền phạt đã nộp thì không hoàn lại. Với sinh viên y khoa còn phải cộng thêm 3 năm đào tạo chuẩn hoá, cả chu kỳ dài hơn ngành sư phạm. Mỗi tỉnh còn có kế hoạch riêng về sinh viên sư phạm hưởng kinh phí và sinh viên y khoa định hướng, điều kiện do tỉnh quy định, trước khi điền nguyện vọng hãy xem văn bản tuyển sinh năm đó của tỉnh. + +### 14. Muốn đi làm ở nước ngoài, trước hết tra công ty này có giấy phép hợp tác lao động xuất khẩu hay không: họ không được thu tiền ký quỹ của bạn + +- Chi phí: Không tốn tiền. Trước khi đi, bỏ mười mấy phút kiểm tra tư cách công ty, rồi đọc hết hai bản hợp đồng. +- Nói thẳng: Chuyện đi làm ở nước ngoài, pháp luật quản không phải bạn mà là công ty tổ chức đưa bạn đi. Chỉ doanh nghiệp đã có giấy chứng nhận hợp tác lao động xuất khẩu và đã đăng ký mới làm được. Nghe câu「xin visa du lịch sang trước rồi đổi sang visa lao động」chính câu đó cho thấy đối phương đang vi phạm. Công ty cũng không được tuỳ danh nghĩa nào mà thu tiền ký quỹ của bạn, hay bắt bạn đặt tài sản làm bảo đảm. +- Lợi ích: Quy định về hợp tác lao động xuất khẩu《对外劳务合作管理条例》Điều năm: làm hợp tác lao động xuất khẩu phải được cơ quan quản lý thương mại cấp tỉnh hoặc thành phố trực thuộc tỉnh phê duyệt và được cấp tư cách kinh doanh. Tư cách kinh doanh đó chính là giấy chứng nhận hợp tác lao động xuất khẩu, không có thì không được tổ chức người ra nước ngoài làm việc. Điều bảy: danh sách giấy chứng nhận báo lên cơ quan quản lý thương mại của Hội đồng Nhà nước, và thông báo cho các đại sứ, lãnh sự của Trung Quốc ở nước ngoài. Điều này còn ghi「未依法取得对外劳务合作经营资格证书并办理登记，不得从事对外劳务合作」, tức chưa có giấy chứng nhận hợp tác lao động xuất khẩu và đăng ký theo pháp luật thì không được làm hợp tác lao động xuất khẩu. Điều tám:「任何单位和个人不得以商务、旅游、留学等名义组织劳务人员赴国外工作」, tức không cá nhân, tổ chức nào được tổ chức người lao động ra nước ngoài làm việc với danh nghĩa thương mại, du lịch, du học. Điều chín: doanh nghiệp phải trong vòng 5 ngày làm việc kể từ ngày đăng ký nộp khoản「不低于 300 万元人民币的对外劳务合作风险处置备用金」tại ngân hàng chỉ định, tức khoản tiền dự phòng xử lý rủi ro hợp tác lao động xuất khẩu không dưới 3 triệu tệ, danh sách nộp tiền công bố cho xã hội. Điều mười: khoản dự phòng dùng để bồi thường các khoản phí dịch vụ phải trả lại, thù lao lao động và thiệt hại phải đền theo pháp luật mà doanh nghiệp từ chối nhận hoặc không có khả năng nhận. Khoản đó còn dùng cho「因发生突发事件，劳务人员回国或者接受紧急救助所需费用」, tức chi phí để người lao động về nước hoặc được cứu trợ khẩn cấp khi xảy ra sự cố bất ngờ. Điều mười một: không được tổ chức người lao động ra nước ngoài làm công việc liên quan tới cờ bạc, hoạt động tình dục. Điều mười hai: không tổ chức đào tạo thì không được tổ chức đi nước ngoài. Điều mười ba: phải mua bảo hiểm tai nạn cá nhân cho người lao động trong thời gian làm việc ở nước ngoài. Điều hai mươi mốt, hai mươi ba: phải ký hợp đồng bằng văn bản riêng với nhà tuyển dụng nước ngoài và riêng với người lao động, không ký thì không được tổ chức đi nước ngoài. Điều hai mươi lăm:「不得以任何名目向劳务人员收取押金或者要求劳务人员提供财产担保」, tức không được tuỳ danh nghĩa nào thu tiền ký quỹ từ người lao động, hay yêu cầu người lao động đặt tài sản làm bảo đảm. Điều hai mươi sáu: trong vòng 10 ngày làm việc kể từ ngày ký hợp đồng phải báo đăng ký với cơ quan quản lý thương mại. Điều hai mươi chín: quyền lợi thực tế của người lao động không khớp với hợp đồng thì doanh nghiệp phải hỗ trợ yêu cầu bồi thường; không được bồi thường đầy đủ thì có quyền yêu cầu doanh nghiệp chịu trách nhiệm bồi thường. Điều ba mươi mốt: nước hoặc vùng lãnh thổ có tình hình an ninh khó bảo đảm an toàn cá nhân thì doanh nghiệp không được tổ chức người lao động đến đó. Điều ba mươi bảy: cơ quan quản lý thương mại lập cơ chế lưu hồ sơ tín dụng xấu và công bố. Chế tài phạt nằm ở các điều sau. Điều ba mươi chín: không đủ tư cách làm hợp tác lao động xuất khẩu thì bị xử lý, cấm đoạt, cấu thành tội phạm thì truy cứu trách nhiệm hình sự. Điều bốn mươi: tổ chức người đi nước ngoài với danh nghĩa thương mại, du lịch, du học, cho vay tư cách, tổ chức làm công việc liên quan cờ bạc và tình dục thì thu hồi giấy chứng nhận hợp tác lao động xuất khẩu. Điều bốn mươi ba: không ký hợp đồng, hoặc che giấu thông tin, cung cấp thông tin giả để tổ chức người đi nước ngoài thì phạt từ 10 vạn tệ đến 20 vạn tệ, còn phải phạt người phụ trách chính từ 2 vạn tệ đến 5 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (2012). Quy định về hợp tác lao động xuất khẩu（国务院令第 620 号，điều năm, bảy, tám, chín, mười, mười một, mười hai, mười ba, hai mươi mốt, hai mươi ba, hai mươi lăm, hai mươi sáu, hai mươi chín, ba mươi mốt, ba mươi bảy, ba mươi chín, bốn mươi, bốn mươi ba）. +- Ghi chú: Quy trình đúng là thế này: công ty ký hợp đồng hợp tác lao động bằng văn bản với nhà tuyển dụng nước ngoài trước, rồi ký hợp đồng dịch vụ bằng văn bản hoặc hợp đồng lao động với bạn. Hợp đồng phải ghi rõ nội dung công việc, thời hạn, mức thù lao, thời giờ làm, ngày phép, bảo hiểm xã hội và rủi ro an toàn cá nhân. Trong vòng 10 ngày làm việc sau khi ký, công ty phải báo đăng ký với cơ quan quản lý thương mại. Công ty còn phải bố trí cho bạn học kỹ năng, phòng chống an toàn, ngoại ngữ và pháp luật tập quán địa phương, và mua bảo hiểm tai nạn cá nhân cho thời gian làm việc ở nước ngoài. Đối phương không cho xem hợp đồng, đòi tiền ký quỹ, thúc bạn đi trước rồi bổ sung thủ tục sau, chính là loại công ty phải tránh. Chiêu tuyển dụng lương cao ở nước ngoài và khu công nghiệp ứng dụng lừa đảo chuyển tiền xem mục 5 ở phần 21, những cảnh báo an toàn phải tra trước khi đi và 12308 nằm ở phần 21, không lặp lại. Quy định không đặt ngưỡng học vấn hay tuổi cho người đi làm, ngưỡng nằm ở tư cách của công ty và ở việc nước nơi đi làm có cấp phép hay không. Công ty chỉ được ký hợp đồng với doanh nghiệp hay tổ chức đã được nước đó phê duyệt, không được ký với cá nhân ở nước ngoài. Khoản dự phòng do cơ quan chủ quản dùng khi công ty không trả, bạn không tự rút được. Nên khi thật sự có chuyện thì phải khiếu nại lên cơ quan quản lý thương mại trước, Điều hai mươi trao cho bạn quyền khiếu nại đó. Gửi hơn 100 người sang cùng một nước hoặc vùng lãnh thổ thì phải bố trí quản lý đi kèm và báo đăng ký với đại sứ, lãnh sự. Dự án ít người không có lớp bảo đảm đó, càng phải tự giữ hợp đồng và biên nhận chuyển tiền. + +### 15. Làm việc từ xa cho công ty ở nước ngoài tại nhà, khoản tiền này tính là thu nhập trong nước theo luật thuế: thuế cá nhân tự kê, chuyển tiền về đi theo khoản「职工报酬」 + +- Chi phí: Không tốn tiền. Mỗi năm làm một lần tổng hợp quyết toán thuế, thời gian là từ ngày 1 tháng 3 đến ngày 30 tháng 6 của năm sau. Khi chuyển tiền về vượt hạn mức cả năm thì còn phải chuẩn bị hợp đồng lao động và chứng minh thu nhập. Chỗ khó là khoản thuế này bạn phải tự nhớ đi kê. +- Nói thẳng: Khoản tiền đó tính trong nước hay ngoài nước phụ thuộc bạn làm việc ở đâu, không phụ thuộc tiền chuyển vào từ đâu. Người ở trong nước làm việc từ xa cho công ty ở nước ngoài thì tính là thu nhập trong nước, phải nộp thuế ở trong nước. Công ty nước ngoài không có cơ quan tại Trung Quốc thì không ai khấu trừ thay bạn, khoản thuế này bạn phải tự đi kê. Tiền chuyển về đi theo khoản「职工报酬」. +- Lợi ích: Khoản tiền đó có tính là thu nhập trong nước hay không thì xem Quy định thi hành Luật Thuế thu nhập cá nhân（中国 个人所得税法实施条例）（国务院令第 707 号，áp dụng từ ngày 1 tháng 1 năm 2019）Điều ba. Quy định ghi:「除国务院财政、税务主管部门另有规定外，下列所得，不论支付地点是否在中国境内，均为来源于中国境内的所得：（一）因任职、受雇、履约等在中国境内提供劳务取得的所得」, tức trừ trường hợp cơ quan tài chính, thuế của Hội đồng Nhà nước có quy định khác, thì các khoản thu nhập sau đây, bất kể chỗ trả tiền có ở trong nước Trung Quốc hay không, đều là thu nhập phát sinh từ trong nước Trung Quốc: một là thu nhập từ việc cung cấp lao động trong nước Trung Quốc do giữ chức vụ, được thuê, thực hiện hợp đồng. Điều sáu của cùng quy định đó định nghĩa hai khoản thuế. Một là lương, công:「工资、薪金所得，是指个人因任职或者受雇取得的工资、薪金、奖金、年终加薪、劳动分红、津贴、补贴以及与任职或者受雇有关的其他所得」, tức thu nhập lương, công là thu nhập cá nhân có được do giữ chức vụ hoặc được thuê gồm lương, công, thưởng, lương tăng cuối năm, chia lợi nhuận lao động, trợ cấp, bồi đắm và các thu nhập khác liên quan tới chức vụ hoặc quan hệ thuê mướn. Khoản kia là thù lao lao động:「劳务报酬所得，是指个人从事劳务取得的所得，包括从事设计、装潢、安装、制图、化验、测试、医疗、法律、会计、咨询、讲学、翻译、审稿、书画、雕刻、影视、录音、录像、演出、表演、广告、展览、技术服务、介绍服务、经纪服务、代办服务以及其他劳务取得的所得」, tức thu nhập từ lao động là thu nhập cá nhân có được khi làm lao động, gồm thu nhập từ thiết kế, trang trí, lắp đặt, vẽ kỹ thuật, thí nghiệm, kiểm nghiệm, y tế, pháp lý, kế toán, tư vấn, giảng dạy, phiên dịch, duyệt bản thảo, thư pháp, khắc, phim ảnh, thu âm, quay video, biểu diễn, trình diễn, quảng cáo, triển lãm, dịch vụ kỹ thuật, dịch vụ giới thiệu, dịch vụ môi giới, dịch vụ hội đồng và các loại lao động khác. Điều sáu của《Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc》ghi cách tính thu nhập chịu thuế:「居民个人的综合所得，以每一纳税年度的收入额减除费用六万元以及专项扣除、专项附加扣除和依法确定的其他扣除后的余额，为应纳税所得额」, tức thu nhập tổng hợp của cá nhân cư trú là số dư của thu nhập mỗi năm tính thuế trừ đi khoản chi phí sáu vạn tệ cùng các khoản khấu trừ riêng, khấu trừ riêng bổ sung và khoản khấu trừ khác do pháp luật quy định. Cùng điều đó còn ghi「劳务报酬所得、稿酬所得、特许权使用费所得以收入减除百分之二十的费用后的余额为收入额」, tức thu nhập từ thù lao lao động, thù lao tác phẩm và phí bản quyền thì lấy thu nhập trừ 20% chi phí, phần còn lại là thu nhập tính thuế. Câu trừ 20% chỉ liệt kê ba khoản này, lương công không nằm trong đó. Vì vậy lương công tính toàn bộ vào thu nhập tính thuế, hai khoản thuế tính ra khác nhau. Điều chín:「个人所得税以所得人为纳税人，以支付所得的单位或者个人为扣缴义务人」, tức thuế thu nhập cá nhân lấy người có thu nhập làm người nộp thuế, lấy đơn vị hoặc cá nhân trả tiền làm người khấu trừ. Điều mười:「有下列情形之一的，纳税人应当依法办理纳税申报：……（二）取得应税所得没有扣缴义务人；……（四）取得境外所得」, tức có một trong các trường hợp sau thì người nộp thuế phải kê khai thuế theo pháp luật:……hai là có thu nhập chịu thuế nhưng không có người khấu trừ……bốn là có thu nhập từ nước ngoài. Điều mười một:「居民个人取得综合所得，按年计算个人所得税……需要办理汇算清缴的，应当在取得所得的次年三月一日至六月三十日内办理汇算清缴」, tức cá nhân cư trú có thu nhập tổng hợp thì tính thuế thu nhập cá nhân theo năm……trường hợp phải tổng hợp quyết toán thì phải làm trong thời gian từ ngày 1 tháng ba đến ngày ba mươi tháng sáu của năm sau năm có thu nhập. Chuyển tiền về thì xem Quy định quản lý ngoại hối cá nhân（中国 个人外汇管理办法）（中国人民银行令〔2006〕第 3 号，áp dụng từ ngày 1 tháng 2 năm 2007）. Điều chín:「对个人结汇和境内个人购汇实行年度总额管理。年度总额内的，凭本人有效身份证件在银行办理；超过年度总额的，经常项目项下凭本人有效身份证件和有交易额的相关证明等材料在银行办理」, tức việc cá nhân chuyển tiền ra nước ngoài và cá nhân trong nước mua ngoại hối quản lý theo hạn mức cả năm. Trong hạn mức cả năm thì tự mang giấy tờ tùy thân hợp lệ ra ngân hàng làm; vượt hạn mức cả năm thì trong nhóm tài khoản thường, mang giấy tờ tùy thân hợp lệ và các chứng từ có giao dịch như chứng minh liên quan ra ngân hàng làm. Điều bảy:「银行和个人在办理个人外汇业务时，应当遵守本办法的相关规定，不得以分拆等方式逃避限额监管，也不得使用虚假商业单据或者凭证逃避真实性管理」, tức ngân hàng và cá nhân khi làm nghiệp vụ ngoại hối cá nhân phải tuân thủ các quy định liên quan của quy định này, không được chia nhỏ để né tránh việc kiểm soát hạn mức, cũng không được dùng chứng từ thương mại hay giấy tờ giả để né tránh kiểm soát tính xác thực. Quy định thi hành về quản lý ngoại hối cá nhân（汇发〔2007〕1 号）Điều hai:「对个人结汇和境内个人购汇实行年度总额管理。年度总额分别为每人每年等值 5 万美元」, tức việc cá nhân chuyển tiền ra nước ngoài và cá nhân trong nước mua ngoại hối quản lý theo hạn mức cả năm; hạn mức cả năm là mỗi người mỗi năm tương đương 5 vạn đô la Mỹ. Điều mười:「境内个人经常项目项下非经营性结汇超过年度总额的，凭本人有效身份证件及以下证明材料在银行办理：……（七）职工报酬：雇佣合同及收入证明」, tức cá nhân trong nước chuyển tiền ra nước ngoài phi kinh doanh trong nhóm tài khoản thường mà vượt hạn mức cả năm thì mang giấy tờ tùy thân hợp lệ cùng các chứng từ sau ra ngân hàng làm:……bảy là thù lao lao động: hợp đồng lao động và chứng minh thu nhập (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (2018). Quy định thi hành Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc（国务院令第 707 号，điều ba, sáu）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (2018). Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc（sửa đổi lần thứ bảy năm 2018，điều sáu, chín, mười, mười một）. ；Kho văn bản quy phạm Tổng cục Thuế. Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc（điều sáu）. ；Ngân hàng Nhân dân Trung Quốc (2006). Quy định quản lý ngoại hối cá nhân（中国人民银行令〔2006〕第 3 号，điều bảy, chín）. ；Cục Quản lý Ngoại hối (2007). Quy định thi hành về quản lý ngoại hối cá nhân（汇发〔2007〕1 号，điều hai, mười）. +- Ghi chú: Tính theo khoản thuế nào thì tuỳ quan hệ giữa bạn và bên kia. Ký hợp đồng lao động, công việc nghe phân công của bên kia thì tính là lương công. Tự nhận việc theo từng đồ án, giao kết quả thì tính là thù lao lao động. Hai cách tính ra thuế khác nhau, trước khi kê hãy xác định mình thuộc loại nào. Khi chuyển tiền về, một năm trong hạn tương đương 5 vạn đô la Mỹ thì chỉ cần mang giấy tờ tùy thân ra ngân hàng chuyển; phần vượt hạn thì phải đưa hợp đồng lao động và chứng minh thu nhập cho ngân hàng. Đừng đi tìm người chia nhỏ khoản tiền để né hạn mức. Trên đây nói trường hợp người ở trong nước, không xuất cảnh. Thật sự xuất cảnh sang nước ngoài làm việc cho nhà tuyển dụng nước ngoài thì là bộ khác, xem mục 14 của phần này（muốn đi làm ở nước ngoài, trước hết tra công ty này có giấy phép hợp tác lao động xuất khẩu hay không）. Bốn việc sau đây cuốn sách này tạm chưa viết. Chúng đều có thể có chi phí thật, nhưng hiện chưa tìm được căn cứ nào đối chiếu từng chữ. Một là khoản thu nhập này có phải nộp thuế giá trị gia tăng không, có phải xuất hoá đơn thay không. Hai là làm lâu dài như vậy có phải đăng ký giấy phép kinh doanh hay đăng ký làm hộ kinh doanh cá thể không. Ba là có thể theo luật lao động yêu cầu nhà tuyển dụng nước ngoài thực hiện quyền lợi không, ví dụ ngày phép năm, tiền làm thêm giờ, bồi thường khi chấm dứt. Bốn là chuyển dữ liệu ra nước ngoài và bảo mật. +### 16. Không vào làm trong đơn vị mà tự làm ăn, tiền khởi nghiệp thì xem khoản vay có bảo đảm cho vốn khởi nghiệp: cá nhân tối đa 30 vạn, ngân sách trả một nửa tiền lãi + +- Chi phí: Không tốn tiền. Đi hỏi thì không mất tiền. Trước hết chạy một lần cơ quan Nhân sự và Xã hội để thẩm định điều kiện, rồi tới từ chức bảo lãnh và ngân hàng thực hiện (ngân hàng cho vay) làm thủ tục mỗi nơi một lần. Chỗ khó là tiền vay mượn được rồi vẫn phải trả. +- Nói thẳng: Nếu cứ phải vay mà khởi nghiệp thì trước hết xem khoản này, có ngân sách trả phụ lãi. Cá nhân vay tối đa 30 vạn tệ, tối đa 3 năm, tiền lãi do ngân sách trả một nửa. Có một điều kiện cứng hay bị bỏ qua: lúc xin, bạn và vợ hoặc chồng không được có khoản vay nào khác, chỉ có các khoản không tính là khoản vay khác gồm khoản vay học tập, tín dụng vi mô cho người thoát nghèo, vay mua nhà, vay mua xe và tín dụng tiêu dùng nhỏ dưới 5 vạn tệ. +- Lợi ích: Điều kiện xin của cá nhân nằm ở Quy định quản lý quỹ đặc biệt phát triển tài chính toàn diện《普惠金融发展专项资金管理办法》（财金〔2023〕75 号）Điều sáu. Điều kiện thứ nhất là thuộc nhóm việc làm trọng điểm. Văn bản liệt kê người đã đăng ký thất nghiệp trong đô thị, người gặp khó khăn trong việc làm (gồm người khuyết tật), cựu chiến binh, người vừa chấm thúc án tù. Còn có người mới tốt nghiệp đại học (gồm sinh viên làm cán bộ ở nông thôn và sinh viên về nước học), công nhân và người thất nghiệp ở doanh nghiệp thuộc diện giảm thừa công suất. Còn có nông dân về quê khởi nghiệp, chủ cửa hàng trên mạng, người thoát nghèo, nông dân ở nông thôn tự khởi nghiệp. Điều kiện thứ hai là không có khoản vay nào khác:「除助学贷款、脱贫人口小额信贷、住房贷款、购车贷款、5 万元以下小额消费贷款（含信用卡消费）以外，申请人提交创业担保贷款申请时，本人及其配偶无其他贷款」, tức ngoài các khoản vay học tập, tín dụng vi mô cho người thoát nghèo, vay mua nhà, vay mua xe, tín dụng tiêu dùng nhỏ dưới 5 vạn tệ (gồm chi tiêu bằng thẻ tín dụng), thì người xin nộp đơn xin khoản vay có bảo đảm cho vốn khởi nghiệp không có khoản vay nào khác, bản thân và vợ hoặc chồng đều vậy. Điều chín:「对符合条件的创业担保贷款，财政部门给予贷款实际利率 50% 的财政贴息。对展期、逾期的创业担保贷款，财政部门原则上不予贴息」, tức với khoản vay có bảo đảm cho vốn khởi nghiệp đủ điều kiện, cơ quan tài chính trả phụ lãi bằng 50% lãi suất thực tế của khoản vay. Với khoản vay được gia hạn hoặc quá hạn, cơ quan tài chính nguyên tắc không trả phụ lãi. Phụ lãi nghĩa là ngân sách trả thay một nửa tiền lãi. Gia hạn là kéo dài thêm sau khi đến hạn, quá hạn là đã quá hạn mà chưa trả, hai trường hợp đó ngân sách không trả thay nữa. Điều mười: khoản vay có bảo đảm cho vốn khởi nghiệp do cá nhân xin dưới 10 vạn tệ thì nguyên tắc bỏ yêu cầu bảo đảm ngược, không phải tìm người khác hay dùng tài sản làm bảo đảm cho cơ quan bảo lãnh. Điều mười một: khoản vay có bảo đảm cho vốn khởi nghiệp cá nhân nằm trong phạm vi hỗ trợ trợ cấp thì「额度不超过 30 万元，贷款期限不超过 3 年」, tức hạn mức không quá 30 vạn tệ, thời hạn vay không quá 3 năm. Người cùng góp vốn khởi nghiệp thì có thể nâng lên tương ứng, tối đa không quá 110% tổng các mức trần của hạn mức cá nhân. Lãi suất tối đa của khoản vay được hỗ trợ trợ cấp chia ba bậc: vùng đông là LPR+50BPs, vùng trung tây LPR+150BPs, vùng thoát nghèo LPR+250BPs. LPR là lãi suất báo giá thị trường của ngân hàng, BPs là đơn vị đo lãi suất, cộng thêm càng nhiều thì lãi càng cao. Người「还款积极、带动就业能力强、创业项目好」, tức trả nợ tích cực, năng lực tạo việc làm mạnh, dự án khởi nghiệp tốt, thì sau khi đến hạn vẫn xin tiếp được, tổng số lần không quá 3 lần. Điều tám: người xin trước hết nộp đơn xin thẩm định điều kiện tới cơ quan Nhân sự và Xã hội địa phương, rồi nộp hồ sơ tới cơ quan bảo lãnh và ngân hàng thực hiện. Điều mười hai: được Ủy ban nhân dân cấp tỉnh đồng ý thì các nơi có thể mở rộng phạm vi hỗ trợ, nâng hạn mức và tỷ lệ phụ lãi (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Tài chính (2023). Thông báo ban hành Quy định quản lý quỹ đặc biệt phát triển tài chính toàn diện《普惠金融发展专项资金管理办法》（财金〔2023〕75 号，第六、八、九、十、十一、十二条）. +- Ghi chú: Ngân sách trả một nửa tiền lãi không làm thay đổi việc「tiền vay thì phải trả」. Lập trường của sách ở mục 1 ở phần 12 không đổi: khởi nghiệp chỉ lấy số tiền mà lỗ còn chịu được, không đụng tới tài sản gia đình, không đặt bảo lãnh cá nhân cho khoản vay của công ty, vợ hoặc chồng càng không ký theo. Nếu bạn đã quyết định vay thì khoản vay chính sách này là mức lãi thấp nhất, đừng đụng vào tín dụng tiêu dùng và vay vốn ngoài ngân hàng. Hạn mức và phạm vi áp dụng các tỉnh có thể nâng lên, thực tế duyệt được bao nhiêu thì theo cách nói năm đó của cơ quan Nhân sự và Xã hội địa phương. Thông báo「cá nhân tối đa 20 vạn」thời dịch bệnh năm 2020 là văn bản tạm thời, bây giờ xem quy định này. Mở cửa hàng là đăng ký hộ kinh doanh cá thể hay lập công ty, giấy phép, kê khai thuế và rút lui xem phần 12. Trợ cấp đào tạo, trợ cấp bảo hiểm xã hội và thực tập nghề trong thời gian thất nghiệp xem mục 13 ở phần 7 (trợ cấp đào tạo nghề). \ No newline at end of file diff --git "a/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" index f9658fde..13fef8cc 100644 --- "a/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" +++ "b/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -1,6 +1,6 @@ [← Về mục lục](../README.md) -# 33. Sống thế nào sau khi bị khuyết tật +# 33. Sống sau khi bị khuyết tật Phần này viết về những ngày sau khi đã bị khuyết tật. Cách phòng ngừa gây khuyết tật nằm ở phần 1 và phần 13, ở đây không lặp lại. @@ -120,74 +120,74 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Nguồn: Luật Xây dựng môi trường không vật cản của Cộng hoà Nhân dân Trung Quốc (中华人民共和国无障碍环境建设法)（第三十五条、第四十六条、第六十二条）. ；Luật Bảo vệ Người khuyết tật của Cộng hoà Nhân dân Trung Quốc (中华人民共和国残疾人保障法)（第五十条）. - Ghi chú: Luật viết là「提供便利」, không kèm chế tài, nên lúc cãi lại tại chỗ phía bên kia không nhất định nhường ngay. Lưu trong điện thoại một ảnh chụp điều luật, hữu ích hơn học thuộc số điều. Không giải quyết được tại chỗ thì khiếu nại theo Điều sáu mươi hai, cơ quan chủ trì phải trả lời. Đường sắt và hàng không có quy định riêng, trước khi đi hãy gọi số chăm sóc khách hàng xác nhận. Giấy tờ「hợp lệ」để được đi miễn phí, mỗi nơi công nhận không hẳn giống nhau, hãy hỏi hội người khuyết tật địa phương xem loại nào được công nhận. -### 13. Người khuyết tật dự thi cao học có thể xin các hỗ trợ hợp lý, dùng đề Braille thì thời gian thi tăng gấp rưỡi +### 13. Thí sinh khuyết tật dự kỳ thi tuyển sinh đại học có thể xin biện pháp hỗ trợ hợp lý, dùng đề braille thì kéo dài thời gian thi thêm một nửa -- Chi phí: Không tốn tiền. Phải nộp một đơn bằng văn bản trong thời hạn tỉnh quy định, còn phải đến tận nơi để xác nhận. -- Nói thẳng: Thí sinh khuyết tật dự thi cao học có thể xin đề Braille, đề chữ lớn, miễn phần thi nghe ngoại ngữ, phòng thi riêng, mang máy trợ thính và xe lăn vào phòng thi, thời gian thi cũng được kéo dài. Những thứ này không tự động có, phải nộp đơn bằng văn bản đúng thời hạn tỉnh quy định, kèm giấy chứng nhận người khuyết tật từ thế hệ thứ hai trở lên. Không đồng ý với kết quả thì có thể xin cơ quan giáo dục tỉnh xét lại. -- Lợi ích: Điều năm của Quy định quản lý việc người khuyết tật tham dự kỳ thi tuyển sinh quốc gia vào các trường đại học thường dân (《残疾人参加普通高等学校招生全国统一考试管理规定》) liệt kê mười hai hạng mục hỗ trợ hợp lý. Gồm cấp đề Braille hiện hành, đề chữ lớn hoặc đề thường cho thí sinh khuyết tật về thị giác; miễn trừ phần thi nghe ngoại ngữ cho thí sinh khuyết tật về thính giác; cho phép thí sinh khuyết tật về thị giác mang theo bút Braille, bảng ghi Braille, máy đánh máy Braille không có chức năng lưu trữ, máy phóng hình không có chức năng lưu trữ, gậy tờ, kính lúp quang học và các thiết bị hỗ trợ khác; cho phép thí sinh khuyết tật về thính giác mang máy trợ thính, cấy ghép ốc tai nhân tạo; cho phép thí sinh di chuyển khó khăn dùng xe lăn, khung đi bộ, có nhu cầu đặc biệt thì được mang bàn ghế riêng. Mức kéo dài ghi ở khoản sáu của Điều năm:「使用盲文试卷的视力残疾考生的考试时间，在该科目规定考试总时长的基础上延长50%；使用大字号试卷或普通试卷的视力残疾考生、因脑瘫或其他疾病引起的上肢无法正常书写或无上肢考生等书写特别困难考生的考试时间，在该科目规定考试总时长的基础上延长30%」— thời gian thi của thí sinh khuyết tật về thị giác dùng đề Braille được kéo dài thêm 50% trên tổng thời gian thi quy định của môn đó; thời gian thi của thí sinh khuyết tật về thị giác dùng đề chữ lớn hoặc đề thường, của thí sinh viết rất khó như liệt não, hoặc do bệnh khác gây ra không viết nổi bằng tay trên, hoặc không có tay trên, được kéo dài thêm 30% trên tổng thời gian thi quy định của môn đó. Ngoài ra còn có ưu tiên vào khu thi, phòng thi chuẩn hoá riêng, bố trí cán bộ hướng dẫn hỗ trợ và thông dịch viên ngôn ngữ ký hiệu. Điều bảy quy định thủ tục xin: thí sinh nộp đơn bằng văn bản chính thức theo thời gian, địa điểm, cách thức mà cơ quan thi cấp tỉnh quy định, kèm giấy chứng nhận người khuyết tật từ thế hệ thứ hai trở lên của bản thân; nhóm chuyên gia xác nhận tại chỗ và đánh giá tổng hợp, cơ quan thi cấp tỉnh cấp «Thông báo». Điều tám quy định, thí sinh có ý kiến khác về «Thông báo» thì có thể đề nghị bằng văn bản xét lại tới cơ quan quản lý giáo dục cấp tỉnh. Điều chín quy định, với trường hợp miễn trừ phần thi nghe ngoại ngữ, điểm môn ngoại ngữ tính theo «điểm bài viết × tổng điểm môn ngoại ngữ ÷ tổng điểm phần viết». Điều mười tám quy định, người khuyết tật tham dự các kỳ thi giáo dục quốc gia khác mà cần hỗ trợ hợp lý thì có thể tham chiếu quy định này mà thực hiện (toàn quốc, ban hành và có hiệu lực ngày 7 tháng 4 năm 2017) +- Chi phí: Không tốn tiền. Phải nộp một đơn bằng văn bản trong thời hạn tỉnh quy định, còn phải đến trực tiếp để xác nhận. +- Nói thẳng: Thí sinh khuyết tật dự kỳ thi tuyển sinh đại học có thể xin đề braille, đề chữ lớn, miễn phần thi nghe ngoại ngữ, phòng thi riêng, mang máy trợ thính và xe lăn vào phòng, thời gian thi cũng được kéo dài. Những thứ này không tự động có, phải nộp đơn bằng văn bản đúng thời hạn tỉnh quy định, kèm giấy chứng nhận người khuyết tật từ thế hệ thứ hai trở lên. Không đồng ý với kết quả thì có thể xin sở giáo dục tỉnh xét lại. +- Lợi ích: Điều năm của Quy định về việc tham gia kỳ thi tuyển sinh quốc gia vào trường đại học thường chính của người khuyết tật (残疾人参加普通高等学校招生全国统一考试管理规定) liệt kê mười hai biện pháp hỗ trợ hợp lý. Gồm cấp đề braille hiện hành, đề chữ lớn hoặc đề thường cho thí sinh khiếm thị; miễn phần thi nghe ngoại ngữ cho thí sinh khiếm thính; cho phép thí sinh khiếm thị mang theo bút braille, bảng viết braille, máy đánh máy braille không có chức năng lưu trữ, máy phóng điện tử không có chức năng lưu trữ, gậy tầm nhìn, kính lúp quang học và các thiết bị hỗ trợ khác; cho phép thí sinh khiếm thính mang theo máy trợ thính, cấy ghép ốc tai nhân tạo; cho phép thí sinh khó di chuyển dùng xe lăn, khung đi chỗ nâng đỡ, có nhu cầu đặc biệt thì được mang bàn ghế đặc biệt của mình. Mức kéo dài thời gian thi nằm ở điểm sáu của điều năm:「使用盲文试卷的视力残疾考生的考试时间，在该科目规定考试总时长的基础上延长50%；使用大字号试卷或普通试卷的视力残疾考生、因脑瘫或其他疾病引起的上肢无法正常书写或无上肢考生等书写特别困难考生的考试时间，在该科目规定考试总时长的基础上延长30%」— thời gian thi của thí sinh khiếm thị dùng đề braille được kéo dài thêm 50% trên tổng thời gian thi quy định của môn đó; thời gian thi của thí sinh khiếm thị dùng đề chữ lớn hoặc đề thường, và của thí sinh viết chữ đặc biệt khó như bị liệt não hay do bệnh khác khiến hai chi trên không viết bình thường, hoặc không có hai chi trên, được kéo dài thêm 30% trên tổng thời gian thi quy định của môn đó. Ngoài ra còn có ưu tiên khi vào khu vực thi, phòng thi chuẩn hoá riêng, bố trí người hướng dẫn hỗ trợ và thông ngôn viên ngôn ngữ ký hiệu. Điều bảy quy định thủ tục xin: thí sinh phải đưa ra đơn chính thức bằng văn bản theo thời gian, địa điểm, hình thức do cơ quan kiểm tra giáo dục cấp tỉnh quy định, và cung cấp giấy chứng nhận người khuyết tật thế hệ thứ hai trở lên của mình; nhóm chuyên gia xác nhận tại chỗ và đánh giá tổng hợp, cơ quan kiểm tra giáo dục cấp tỉnh cấp Thông báo. Điều tám quy định, thí sinh có ý kiến khác với Thông báo thì có thể đưa đơn xét lại bằng văn bản tới cơ quan quản lý giáo dục cấp tỉnh. Điều chín quy định, với môn ngoại ngữ được miễn phần nghe thì điểm tính theo công thức「笔试成绩×外语科总分值/笔试部分总分值」. Điều mười tám quy định, người khuyết tật tham gia kỳ thi giáo dục quốc gia khác cần biện pháp hỗ trợ hợp lý thì được áp dụng theo quy định này (toàn quốc, ban hành và có hiệu lực ngày 7 tháng 4 năm 2017) - Mức bằng chứng: A -- Nguồn: Bộ Giáo dục (教育部), Hiệp hội Người khuyết tật Trung Quốc (中国残联) (2017). Quy định quản lý việc người khuyết tật tham dự kỳ thi tuyển sinh quốc gia vào các trường đại học thường dân (《残疾人参加普通高等学校招生全国统一考试管理规定》)(教学〔2017〕4 号，第五条、第七条、第八条、第九条、第十八条). -- Ghi chú: Thời hạn xin do từng tỉnh quy định, ghi trong phương pháp đăng ký dự thi cao học của năm đó, hết hạn là mất. Nên học kỳ một năm ba của lớp mười hai phải lên trang chính thức của cơ quan thi cấp tỉnh xác nhận thời hạn. Điều mười tám còn quy định, kỳ thi sau đại học, thi tuyển sinh người lớn và các kỳ thi giáo dục quốc gia khác cũng có thể xin tham chiếu. Quy định trong giai đoạn học đường rằng trường không được từ chối nhận, xem mục 14 của phần này (trường học không được từ chối nhận). Người được lợi của mục này là bạn hoặc con bạn. +- Nguồn: Bộ Giáo dục (教育部), Hội Liên hiệp Người khuyết tật Trung Quốc (中国残联) (2017). Quy định về việc tham gia kỳ thi tuyển sinh quốc gia vào trường đại học thường chính của người khuyết tật (教学〔2017〕4 号，第五条、第七条、第八条、第九条、第十八条). +- Ghi chú: Thời hạn xin do từng tỉnh quy định, ghi trong quy chế đăng ký thi tuyển sinh của năm đó, trễ là mất. Vì vậy học kỳ một lớp mười hai phải vào trang của cơ quan kiểm tra giáo dục cấp tỉnh để xác nhận thời hạn. Điều mười tám còn quy định, các kỳ thi giáo dục quốc gia khác như thi sau đại học, thi tốt nghiệp trung học phổ thông người lớn cũng được áp dụng khi xin. Quy định ở giai đoạn đi học rằng nhà trường không được từ chối, xem mục 14 của phần này (nhà trường không được từ chối). Người hưởng mục này là bạn hoặc con bạn. -### 14. Trẻ khuyết tật xin nhập học, trường học không được từ chối; trẻ không đến được trường thì giáo dục sở bố trí dạy đến nhà +### 14. Trẻ khuyết tật xin nhập học, nhà trường không được từ chối; trẻ không đến được trường thì phòng giáo dục sắp xếp giảng viên đến tận nhà -- Chi phí: Không tốn tiền. Phải nộp một đơn, có thể phải qua một lần đánh giá. Khi trường từ chối thì phải kêu lên phòng giáo dục. -- Nói thẳng: Trường mầm non và các loại trường khác không được từ chối nhận trẻ khuyết tật hợp điều kiện vào học, trường dạy nghề thường cũng vậy. Trẻ cần người chăm riêng, không đến được trường, thì giáo dục huyện bố trí dạy đến nhà hoặc học từ xa, vẫn có hồ sơ học tập. Việc nhập học và chuyển trường do giáo dục sở mời hội đồng chuyên gia đánh giá rồi quyết định, không phải trường tự ý quyết. -- Lợi ích: Điều bảy của Quy định về Giáo dục người khuyết tật (《残疾人教育条例》):「学前教育机构、各级各类学校及其他教育机构应当依照本条例以及国家有关法律、法规的规定，实施残疾人教育；对符合法律、法规规定条件的残疾人申请入学，不得拒绝招收。」— tức cơ sở giáo dục mầm non, các trường học mọi cấp mọi loại và cơ sở giáo dục khác phải thực hiện giáo dục cho người khuyết tật theo quy định này và theo pháp luật, quy định của nhà nước; không được từ chối nhận người khuyết tật hợp điều kiện theo pháp luật, quy định xin nhập học. Điều hai mươi chín:「普通职业学校不得拒绝招收符合国家规定的录取标准的残疾人入学，普通职业培训机构应当积极招收残疾人入学。」— trường dạy nghề thường không được từ chối nhận người khuyết tật đạt chuẩn tuyển sinh do nhà nước quy định, cơ sở đào tạo nghề thường phải tích cực nhận người khuyết tật. Quy định còn viết, trẻ em, thiếu niên khuyết tật đúng tuổi cần người chăm riêng, không đến trường học được, thì cơ quan quản lý giáo dục của chính quyền huyện điều phối, thực hiện giáo dục bắt buộc bằng cách dạy đến nhà hoặc giáo dục từ xa, và đưa vào quản lý hồ sơ học tập. Điều hai mươi quy định, cơ quan quản lý giáo dục của chính quyền huyện phải phối hợp với cơ quan y tế, cơ quan dân vụ, hội người khuyết tật để lập một hội đồng chuyên gia về giáo dục người khuyết tật gồm các chuyên gia về giáo dục, tâm lý, phục hồi chức năng và công tác xã hội, hội đồng đó đánh giá tình trạng sức khoẻ thể chất, khả năng tiếp thụ giáo dục và khả năng thích nghi với đời sống học tập ở trường rồi đề nghị về nhập học, chuyển trường; kết quả đánh giá thuộc quyền riêng tư của trẻ, các cơ quan và nhân viên có nghĩa vụ giữ bí mật. Điều hai mươi ba quy định, học sinh khuyết tật học chung lớp ở trường thường có thể áp dụng chương trình, tiêu chuẩn, giáo trình của giáo dục bắt buộc thường, nhưng yêu cầu học tập đối với họ có thể linh hoạt ở mức độ thích hợp. Điều năm mươi hai quy định, người khuyết tật tham dự kỳ thi quốc gia mà cần điều kiện hỗ trợ và hỗ trợ hợp lý thì có thể đề nghị (toàn quốc, có hiệu lực từ ngày 1 tháng 5 năm 2017) +- Chi phí: Không tốn tiền. Phải nộp một đơn, có thể phải qua một lần đánh giá. Khi nhà trường đẩy trách nhiệm thì phải lên tới phòng giáo dục. +- Nói thẳng: Trường mầm non và mọi loại trường không được từ chối trẻ khuyết tật đủ điều kiện xin nhập học, trường dạy nghề thường cũng vậy. Trẻ cần người chăm riêng, không đến được trường thì phòng giáo dục huyện sắp xếp giảng viên đến tận nhà hoặc dạy từ xa, vẫn có học phạm. Việc nhập học và chuyển trường do phòng giáo dục nhờ ủy ban chuyên gia đánh giá rồi quyết định, không phải nhà trường tự nói. +- Lợi ích: Điều bảy của Quy định về giáo dục người khuyết tật (残疾人教育条例):「学前教育机构、各级各类学校及其他教育机构应当依照本条例以及国家有关法律、法规的规定，实施残疾人教育；对符合法律、法规规定条件的残疾人申请入学，不得拒绝招收。」— các cơ sở giáo dục mầm non, trường học các cấp các loại và cơ sở giáo dục khác phải thực hiện giáo dục người khuyết tật theo quy định này và theo quy định pháp luật của nhà nước; không được từ chối tiếp nhận người khuyết tật đủ điều kiện theo quy định pháp luật xin nhập học. Điều hai mươi chín:「普通职业学校不得拒绝招收符合国家规定的录取标准的残疾人入学，普通职业培训机构应当积极招收残疾人入学。」— trường dạy nghề thường không được từ chối tiếp nhận người khuyết tật đạt chuẩn tuyển sinh do nhà nước quy định; cơ sở đào tạo nghề thường phải tích cực tuyển người khuyết tật. Quy định còn nói, trẻ em và thiếu niên khuyết tật đúng tuổi cần người chăm riêng, không thể đến trường học, thì do cơ quan quản lý giáo dục của chính quyền cấp huyện sắp xếp, thực hiện giáo dục bắt buộc bằng cách cho giảng viên đến tận nhà hoặc giáo dục từ xa, đồng thời đưa vào quản lý học phạm. Điều hai mươi quy định, cơ quan quản lý giáo dục của chính quyền cấp huyện phải phối hợp với cơ quan y tế, cơ quan dân vụ, hội liên hiệp người khuyết tật để lập ủy ban chuyên gia về giáo dục người khuyết tật gồm các chuyên gia về giáo dục, tâm lý, phục hồi chức năng, công tác xã hội; ủy ban đó đánh giá tình trạng thể chất, khả năng tiếp nhận giáo dục và khả năng thích nghi với đời sống học tập ở trường, đồng thời đưa ra ý kiến về nhập học, chuyển trường; kết quả đánh giá là thông tin riêng của trẻ, các cơ quan và cán bộ liên quan có nghĩa vụ giữ bí mật. Điều hai mươi ba quy định, học sinh khuyết tật học chung lớp ở trường thường có thể áp dụng chương trình, chuẩn và sách giáo khoa của giáo dục bắt buộc thường, nhưng yêu cầu học tập đối với họ có thể có mức độ linh hoạt. Điều năm mươi hai quy định, người khuyết tật tham gia kỳ thi giáo dục quốc gia cần điều kiện hỗ trợ và biện pháp hỗ trợ hợp lý thì có thể đưa đơn (toàn quốc, có hiệu lực từ ngày 1 tháng 5 năm 2017) - Mức bằng chứng: A -- Nguồn: Hội đồng Nhà nước (国务院) (2017). Quy định về Giáo dục người khuyết tật (《残疾人教育条例》)(第七条、第二十条、第二十三条、第二十九条、第五十二条). -- Ghi chú: Trường nói miệng「chỗ chúng tôi không có điều kiện」thì không được coi là. Bắt nó trả lời bằng văn bản, cầm bản trả lời đó đi tìm giáo dục huyện. Việc đánh giá do giáo dục sở ủy thác cho hội đồng chuyên gia. Câu「yêu cầu học tập có thể linh hoạt ở mức độ thích hợp」trong Điều hai mươi ba rất hữu ích, gặp trường dùng thành tích làm lý do đuổi học thì viện dẫn. Bị bắt nạt ở trường thì lưu chứng cứ thế nào, trường phải đi qua bước nào, xem mục 3 ở phần 30 (bắt nạt học đường). Hỗ trợ hợp lý cho kỳ thi cao học, xem mục 13 của phần này (hỗ trợ hợp lý). Người được lợi của mục này là con bạn. +- Nguồn: Hội đồng Nhà nước (国务院) (2017). Quy định về giáo dục người khuyết tật (残疾人教育条例，第七条、第二十条、第二十三条、第二十九条、第五十二条). +- Ghi chú: Nhà trường nói miệng「chỗ chúng tôi không có điều kiện」thì không được. Yêu cầu nhà trường trả lời bằng văn bản, cầm bản trả lời đó đến phòng giáo dục cấp huyện. Việc đánh giá do phòng giáo dục ủy thác cho ủy ban chuyên gia. Câu「yêu cầu học tập có thể có mức độ linh hoạt」trong điều hai mươi ba rất có ích, gặp nhà trường dùng kết quả học tập làm lý do khuyên bỏ học thì dẫn câu này. Con bị bắt nạt ở trường thì lấy bằng chứng thế nào, nhà trường bắt buộc phải đi qua bước nào, xem mục 3 ở phần 30 (bắt nạt học đường). Biện pháp hỗ trợ hợp lý cho kỳ thi tuyển sinh đại học xem mục 13 của phần này (biện pháp hỗ trợ hợp lý). Người hưởng mục này là con bạn. -### 15. Mất chân phải hoặc cả hai chân dưới vẫn thi được giấy phép lái xe, loại xe được phép lái gọi là C5 +### 15. Mất hoặc không cử động được hai chi dưới bên phải vẫn thi được giấy phép lái xe, loại xe được phép lái gọi là C5 -- Chi phí: Học phí tầm như giấy phép lái xe thường, ở hầu hết các nơi vài nghìn tệ. Xe phải độ lại sang điều khiển tay. Cứ ba năm khám sức khoẻ một lần. -- Nói thẳng: Người mất chân phải hoặc cả hai chân dưới, không cử động được, nhưng tự ngồi vững được và phần thân trên đủ điều kiện, thì được thi giấy phép lái xe hơi nhỏ số hộc tự động dành riêng cho người khuyết tật, mã C5. Người mất một bàn tay cũng xin được. Người khuyết thị lực một mắt, người khuyết thính giác có các loại xe riêng để thi. Khám sức khoẻ phải làm ở bệnh viện chuyên khoa do tỉnh công nhận. -- Lợi ích: Phụ lục 1 của Quy định về việc đăng ký và sử dụng giấy phép lái xe cơ giới (《机动车驾驶证申领和使用规定》) viết khá chi tiết về điều kiện thể chất. Khoản 8:「右下肢、双下肢缺失或者丧失运动功能但能够自主坐立，且上肢符合本项第5目规定的，可以申请残疾人专用小型自动挡载客汽车准驾车型的机动车驾驶证。一只手掌缺失，另一只手拇指健全，其他手指有两指健全，上肢和手指运动功能正常，且下肢符合本项第6目规定的，可以申请残疾人专用小型自动挡载客汽车准驾车型的机动车驾驶证」— người mất chân phải, mất cả hai chân dưới hoặc mất chức năng vận động nhưng ngồi đứng tự được, đồng thời phần thân trên đáp ứng khoản 5 của cùng mục này, thì có thể xin giấy phép lái xe cơ giới loại xe hành khách nhỏ, số hộc tự động dành riêng cho người khuyết tật; người mất một bàn tay, bàn tay còn lại có ngón cái lành, các ngón tay khác có hai ngón lành, chức năng vận động của tay và ngón tay bình thường, đồng thời phần dưới đáp ứng khoản 6 của cùng mục này, thì cũng có thể xin giấy phép lái xe cơ giới loại xe hành khách nhỏ, số hộc tự động dành riêng cho người khuyết tật. Khoản 6: người chỉ mất hoặc mất chức năng vận động chân dưới trái nhưng chân phải bình thường thì có thể xin loại xe hành khách nhỏ, số hộc tự động. Về thị lực:「单眼视力障碍，优眼裸视力或者矫正视力达到对数视力表5.0以上，且水平视野达到150度的」, tức khuyết thị lực một mắt, thị lực trần hoặc thị lực đã điều chỉnh của mắt tốt đạt 5.0 trở lên trên bảng thị lực logarit, và tầm nhìn ngang đạt 150 độ, có thể xin loại xe ô tô nhỏ, xe ô tô nhỏ số hộc tự động, xe tải nhỏ chạy chậm, xe ba bánh, xe hành khách nhỏ số hộc tự động dành riêng cho người khuyết tật. Về thính giác: người khuyết thính giác nhưng đeo thiết bị trợ thính thì phân biệt được hướng âm thanh thì có thể xin loại xe ô tô nhỏ, xe ô tô nhỏ số hộc tự động, khi lái xe phải đeo thiết bị trợ thính. Điều tám mươi lăm quy định, giấy chứng nhận điều kiện thể chất nộp khi làm thủ tục giấy phép lái xe hành khách nhỏ, số hộc tự động dành riêng cho người khuyết tật phải do cơ sở y tế chuyên khoa được cơ quan quản lý y tế cấp tỉnh công nhận cấp. Người có loại xe được phép lái đó phải khám sức khoẻ ba năm một lần, nộp giấy chứng nhận điều kiện thể chất trong vòng ba mươi ngày sau khi kết thúc chu kỳ tính điểm. Điều bảy mươi tám quy định, khi lái xe phải đặt đúng quy định biển hiệu dành riêng cho xe của người khuyết tật trên thân xe (toàn quốc, Quy định sửa đổi theo lệnh Bộ Công an số 172, có hiệu lực từ ngày 1 tháng 1 năm 2025) +- Chi phí: Học phí tương tự giấy phép lái xe thường, ở hầu hết các nơi vài nghìn tệ. Xe phải đội lại thành điều khiển bằng tay. Cứ ba năm kiểm tra sức khoẻ một lần. +- Nói thẳng: Người mất hoặc không cử động được chi dưới bên phải hoặc cả hai chi dưới, nhưng tự ngồi vững được và hai chi trên đủ điều kiện thì có thể thi giấy phép lái xe ô tô nhỏ số tự động dành riêng cho người khuyết tật, mã C5. Người mất một bàn tay cũng xin được. Người khiếm thị một mắt, người khiếm thính có các loại xe riêng để thi. Kiểm tra sức khoẻ phải làm ở bệnh viện chuyên khoa được tỉnh công nhận. +- Lợi ích: Phụ lục 1 của Quy định về việc cấp và sử dụng giấy phép lái xe cơ giới (机动车驾驶证申领和使用规定) viết khá chi tiết về điều kiện thể chất. Mục số 8:「右下肢、双下肢缺失或者丧失运动功能但能够自主坐立，且上肢符合本项第5目规定的，可以申请残疾人专用小型自动挡载客汽车准驾车型的机动车驾驶证。一只手掌缺失，另一只手拇指健全，其他手指有两指健全，上肢和手指运动功能正常，且下肢符合本项第6目规定的，可以申请残疾人专用小型自动挡载客汽车准驾车型的机动车驾驶证」— mất hoặc mất khả năng vận động hai chi dưới bên phải hoặc cả hai, nhưng tự ngồi đứng được, và hai chi trên đáp ứng quy định tại mục năm của cùng mục này, thì có thể xin giấy phép lái xe cơ giới với loại xe được phép lái là ô tô con nhỏ số tự động dành riêng cho người khuyết tật; mất một bàn tay, tay còn lại có ngón cái lành, trong các ngón tay còn lại có hai ngón lành, chức năng vận động của hai chi trên và ngón tay bình thường, và hai chi dưới đáp ứng quy định tại mục sáu của cùng mục này, thì cũng có thể xin loại giấy phép lái xe cơ giới như trên. Mục sáu: chỉ mất hoặc mất khả năng vận động chi dưới bên trái, nhưng chi dưới bên phải bình thường thì có thể xin loại xe ô tô nhỏ số tự động. Về thị lực:「单眼视力障碍，优眼裸视力或者矫正视力达到对数视力表5.0以上，且水平视野达到150度的」— khiếm thị một mắt, thị lực tốt hơn không kính hoặc sau nhiễu đạt từ 5.0 trở lên trên bảng thị lực loga, và tầm nhìn ngang đạt 150 độ, thì có thể xin các loại ô tô nhỏ, ô tô nhỏ số tự động, xe tải chạy chậm, xe ba bánh và ô tô con nhỏ số tự động dành riêng cho người khuyết tật. Về thính giác: người khiếm thính nhưng mang thiết bị trợ thính vẫn xác định được hướng của nguồn âm thanh thì có thể xin loại ô tô nhỏ và ô tô nhỏ số tự động, khi lái phải mang thiết bị trợ thính. Điều tám mươi lăm quy định, giấy chứng minh điều kiện thể chất nộp khi làm thủ tục giấy phép lái ô tô con nhỏ số tự động dành riêng cho người khuyết tật phải do cơ sở y tế chuyên khoa được cơ quan quản lý sức khoẻ cấp tỉnh công nhận cấp. Người có loại xe được phép lái này phải kiểm tra sức khoẻ mỗi ba năm một lần, nộp giấy chứng minh điều kiện thể chất trong vòng ba mươi ngày sau khi hết chu kỳ điểm. Điều bảy mươi tám quy định, khi lái phải đặt đúng quy định biển hiệu dành riêng cho xe cơ giới của người khuyết tật trên thân xe (toàn quốc, Lệnh số 172 của Bộ Công an sửa đổi, có hiệu lực từ ngày 1 tháng 1 năm 2025) - Mức bằng chứng: A -- Nguồn: Lệnh Bộ Công an số 172 (公安部令第 172 号). Quy định về việc đăng ký và sử dụng giấy phép lái xe cơ giới (《机动车驾驶证申领和使用规定》)(第七十八条、第八十五条、附件 1). -- Ghi chú: Rẻ hơn nhiều là xe lăn điện dành cho người khuyết tật. Điều một trăm mười chín của Luật An toàn giao thông đường bộ (《道路交通安全法》) xếp nó vào nhóm phương tiện không động cơ, nên không cần giấy phép lái xe; Điều năm mươi tám quy định, xe lăn điện dành cho người khuyết tật chạy trong làn phương tiện không động cơ thì tốc độ tối đa không được quá mười lăm ki lô mét một giờ. Xe loại C5 phải lắp bước ga và phanh tay, độ xe phải tìm nhà sản xuất có đủ điều kiện, độ xong đi đăng ký thay đổi tại cơ quan quản lý phương tiện. Ba chu kỳ tính điểm trôi qua mà trong vòng một năm không nộp giấy chứng nhận điều kiện thể chất thì giấy phép lái xe bị thu hồi. +- Nguồn: Lệnh số 172 của Bộ Công an (公安部令第 172 号). Quy định về việc cấp và sử dụng giấy phép lái xe cơ giới (机动车驾驶证申领和使用规定，第七十八条、第八十五条、附件 1). +- Ghi chú: Loại xe dễ hơn nhiều là xe lăn có động cơ dành cho người khuyết tật. Điều 119 của Luật An toàn giao thông đường bộ xếp loại xe này vào nhóm phương tiện không có động cơ, nên không cần giấy phép lái xe; Điều 58 quy định, xe lăn có động cơ dành cho người khuyết tật khi chạy trong làn xe tự động không được vượt quá tốc độ tối đa mười lăm ki lô mét một giờ. Xe mã C5 phải lắp bước ga và phanh cơ, việc đội xe phải tìm nhà sản xuất có giấy phép, đội xong đi đăng ký đổi thông tin tại cơ quan quản lý phương tiện. Không nộp giấy chứng minh điều kiện thể chất trong vòng một năm sau khi hết ba chu kỳ điểm thì giấy phép lái xe sẽ bị thu hồi. -### 16. Phục hồi chức năng thì làm ở cơ sở chính quy có khoa phục hồi chức năng, cường độ nghe theo chuyên gia trị liệu, không phải càng mạnh càng tốt +### 16. Phục hồi chức năng làm ở cơ sở chính quy có khoa phục hồi chức năng, cường độ nghe người trị liệu, tập càng nặng càng không tốt -- Chi phí: Phục hồi chức năng nội trú mỗi ngày vài trăm tệ, phần tự trả thì tuỳ địa phương. Thời gian tính theo tháng. Tập luyện mỗi ngày, rất chán. -- Nói thẳng: Đưa người bệnh về nhà nằm chờ tự hồi phục là sai lầm phổ biến nhất. Sau đột quỵ, chăm sóc có tổ chức bởi nhóm đa chuyên ngành cho ra nhiều người hơn còn sống sau một năm và còn ở được tại nhà. Nhưng cũng đừng tự ý tăng cường, sau đột quỵ trong vòng 24 giờ mà tập sớm hơn, thường xuyên hơn, nhiều hơn thì kết quả sau ba tháng lại kém hơn. Tập bao nhiêu, tập lúc nào thì nghe khoa phục hồi chức năng và chuyên gia trị liệu. -- Lợi ích: Một bài tổng hợp Cochrane đưa vào 29 thử nghiệm, 5902 người tham gia, so sánh chăm sóc nội trú có tổ chức theo đơn vị đột quỵ với các cách chăm sóc khác. Khi theo dõi tới lần cuối (trung vị 1 năm), tỷ lệ bất lợi (tử vong, tàn tật, hoặc cần chăm sóc trong cơ sở) có odds ratio 0.77 (95% CI 0.69–0.87, khoảng tin cậy), tử vong có odds ratio 0.76 (95% CI 0.66–0.88), tử vong hoặc phải dựa vào người khác có odds ratio 0.75 (95% CI 0.66–0.85), chất lượng bằng chứng ở mức trung bình. Quy đổi ra số tuyệt đối, cứ 100 người được chăm sóc theo đơn vị đột quỵ thì có thêm 2 người sống, thêm 6 người ở được tại nhà, thêm 6 người tự lo được việc sinh hoạt. Mặt khác, thử nghiệm AVERT phân ngẫu nhiên 2104 bệnh nhân tại 56 đơn vị đột quỵ cấp tính ở 5 nước, so sánh「chăm sóc thường quy」với「chăm sóc thường quy cộng vận động cường độ cao siêu sớm」. Ở thời điểm ba tháng, nhóm kết quả tốt (điểm 0—2 trên thang Rankin cải biến) có 480 người (46%) ở nhóm siêu sớm, 525 người (50%) ở nhóm thường quy, odds ratio sau điều chỉnh 0.73 (95% CI 0.59–0.90, P=0.004). Tử vong là 88 người so với 72 người, không ra khác biệt (odds ratio 1.34, 95% CI 0.93–1.93, P=0.113) +- Chi phí: Nằm viện phục hồi chức năng mỗi ngày vài trăm tệ, phần tự trả thay đổi tuỳ địa phương. Thời gian tính theo tháng. Ngày nào cũng phải tập, rất chán. +- Nói thẳng: Đưa người bệnh về nhà nằm chờ tự hồi phục là sai lầm thường gặp nhất. Sau đột quỵ, chăm sóc có tổ chức bởi nhóm đa chuyên ngành cho nhiều người hơn còn sống và còn ở được tại nhà sau một năm. Nhưng cũng đừng tự tăng cường độ, tập sớm hơn, nhiều hơn, nhiều lần hơn trong vòng 24 giờ đầu sau đột quỵ thì kết quả sau ba tháng lại kém hơn. Tập bao nhiêu, tập lúc nào thì nghe khoa phục hồi chức năng và người trị liệu. +- Lợi ích: Tổng quan hệ thống Cochrane đưa vào 29 thử nghiệm, 5902 người tham gia, so sánh chăm sóc có tổ chức trong khoa đột quỵ với các cách chăm sóc khác. Tại lần theo dõi cuối (trung vị 1 năm), tỷ số khả năng của kết cục xấu (tử vong, tàn tật hoặc phải vào cơ sở chăm sóc) là 0.77 (95% CI 0.69–0.87, khoảng tin cậy), tỷ số khả năng của tử vong là 0.76 (95% CI 0.66–0.88), tỷ số khả năng của tử vong hoặc phụ thuộc người khác là 0.75 (95% CI 0.66–0.85), chất lượng bằng chứng ở mức trung bình. Quy đổi ra tuyệt đối thì cứ 100 người được chăm sóc trong khoa đột quỵ thì có thêm 2 người sống sót, thêm 6 người ở được tại nhà, thêm 6 người tự lo. Mặt khác, thử nghiệm AVERT phân bổ ngẫu nhiên 2104 bệnh nhân ở 56 khoa đột quỵ cấp cứu tại 5 nước, so sánh「chăm sóc thông thường」với「chăm sóc thông thường cộng vận động sớm cường độ cao」. Sau ba tháng, kết cục tốt (0—2 điểm theo thang Rankin cải tiến) là nhóm vận động cực sớm 480 người (46%), nhóm thông thường 525 người (50%), tỷ số khả năng sau điều chỉnh 0.73 (95% CI 0.59–0.90, P=0.004). Tử vong là 88 người so với 72 người, không ra khác biệt (tỷ số khả năng 1.34, 95% CI 0.93–1.93, P=0.113) - Mức bằng chứng: A - Nguồn: Langhorne P, Ramachandra S; Stroke Unit Trialists' Collaboration (2020). Organised inpatient (stroke unit) care for stroke: network meta-analysis. Cochrane Database of Systematic Reviews, 4:CD000197. ；AVERT Trial Collaboration group (2015). Efficacy and safety of very early mobilisation within 24 h of stroke onset (AVERT): a randomised controlled trial. Lancet, 386(9988):46-55. -- Ghi chú:Tranh cãi: Cả hai bằng chứng này đều đến từ đột quỵ, áp sang chấn thương tủy sống, cắt cụt, bỏng và các nguyên nhân khuyết tật khác thì là suy diễn lan. Kết quả của AVERT nói「càng sớm càng mạnh」không đúng, không phải nói vận động sớm là vô ích. Trong thử nghiệm, đại đa số người ở cả hai nhóm đều đã xuống giường trong vòng 24 giờ, khác nhau là tần suất và tổng lượng. Khi chọn cơ sở, hãy xem nó có khoa phục hồi chức năng không, có chuyên gia trị liệu phục hồi chức năng và chuyên gia hoạt động trị liệu (occupational therapist) làm việc toàn thời gian không, đừng chỉ nhìn hai chữ「phục hồi」trên biển hiệu. Chuyển viện giữa tuyến cơ sở và tuyến trên, mức đồng chi trả tính thế nào, xem mục 1 ở phần 24 (chuyển viện theo từng tuyến). Phục hồi chức năng tại cộng đồng và lắp thiết bị hỗ trợ, xem mục 7 của phần này (hỏi hết ở hội người khuyết tật cấp huyện). +- Ghi chú:Tranh cãi: Cả hai bằng chứng này đều đến từ đột quỵ, áp sang chấn thương tủy sống, cắt chiết, bỏng và các nguyên nhân khuyết tật khác thì là suy diễn lan. Kết quả AVERT nói rằng「càng sớm càng mạnh」không đúng, không phải là「vận động sớm không có ích」. Trong thử nghiệm, đa số người ở cả hai nhóm đều xuống giường trong vòng 24 giờ, khác nhau ở tần suất và tổng lượng. Khi chọn cơ sở, xem nó có khoa phục hồi chức năng không, có nhân viên chuyên trách về phục hồi chức năng và trị liệu hoạt động tự do không, đừng chỉ nhìn hai chữ「phục hồi」trên biển hiệu. Chuyển viện giữa tuyến cơ sở và bệnh viện hạng A, mức đồng chi trả tính thế nào, xem mục 1 ở phần 24 (chuyển viện theo từng bậc). Phục hồi chức năng cộng đồng và lắp thiết bị hỗ trợ xem mục 7 của phần này (hỏi hết ở hội người khuyết tật cấp huyện). -### 17. Nghe kém thì đi lắp máy trợ thính, lý do lắp là để nghe thấy, không phải để phòng bệnh Alzheimer +### 17. Nghe kém thì đi lắp máy trợ thính, lý do lắp là để nghe rõ, không phải để phòng chứng mất trí nhớ -- Chi phí: Máy trợ thính giá vài nghìn đến vài chục nghìn tệ, phần lớn phải tự trả. Phải đo khớp âm thanh, còn phải chỉnh vài lần xa nhà. -- Nói thẳng: Câu「đeo máy trợ thính phòng bệnh Alzheimer」hiện chưa đứng vững. Một thử nghiệm gần một nghìn người trong ba năm, xét chung không ra khác biệt. Chỉ trong nhóm vốn đã có nguy cơ suy giảm nhận thức cao hơn mới thấy khác biệt. Nên lý do để lắp máy trợ thính là nghe thấy ngay lúc này, nói chuyện được với người, khi qua đường nghe thấy xe, không phải để giữ não cho mấy chục năm sau. -- Lợi ích: Thử nghiệm ACHIEVE làm ở 4 cộng đồng tại Hoa Kỳ, đưa vào 977 người từ 70 đến 84 tuổi, bị mất thính giác chưa được điều trị, không có rối loạn nhận thức rõ rệt. Ngẫu nhiên chia hai nhóm, một nhóm nhận can thiệp về thính giác (tư vấn thính giác và lắp máy trợ thính), một nhóm nhận giáo dục sức khoẻ làm nhóm chứng, theo dõi 6 tháng một lần. Điểm chính là thay đổi điểm nhận thức toàn cục trong 3 năm. Trong phân tích chính gộp hai nhóm dân số, hai nhóm không có khác biệt có ý nghĩa: nhóm can thiệp thính giác -0.200 (95% CI -0.256 đến -0.144), nhóm chứng -0.202 (95% CI -0.258 đến -0.145), chênh lệch 0.002 (95% CI -0.077 đến 0.081), P=0.96. Một phân tích độ nhạy định sẵn cho thấy hiệu quả của can thiệp có khác nhau giữa hai nhóm dân số (tương tác P=0.010). Người tham dự từ nghiên cứu quan sát dài hạn về sức khoẻ tim mạch lâu dài lớn tuổi hơn, có nhiều yếu tố nguy cơ suy giảm nhận thức hơn, điểm nhận thức nền thấp hơn. Kết luận của nhóm nghiên cứu là can thiệp thính giác có thể giảm mức thay đổi nhận thức trong ba năm ở nhóm nguy cơ suy giảm nhận thức cao, còn ở nhóm nguy cơ thấp thì không. Thử nghiệm không báo cáo biến chứng nặng nào do nghiên cứu +- Chi phí: Máy trợ thính vài nghìn tới vài chục nghìn tệ, phần lớn phải tự trả. Cần đo khớp, còn phải chỉnh vài lần. +- Nói thẳng: Câu「đeo máy trợ thính phòng chứng mất trí nhớ」hiện chưa đứng vững. Một thử nghiệm ba năm gần một nghìn người, nhìn chung không ra khác biệt. Chỉ thấy khác biệt ở nhóm vốn đã có nguy cơ suy giảm nhận thức cao hơn. Nên lý do lắp máy trợ thính là nghe rõ ngay lúc này, nói chuyện được với người, nghe thấy xe khi qua đường, không phải để bảo vệ bộ não vài chục năm sau. +- Lợi ích: Thử nghiệm ACHIEVE làm ở 4 cộng đồng tại Hoa Kỳ, đưa vào 977 người từ 70 đến 84 tuổi bị mất thính lực chưa được điều trị và không có rối loạn nhận thức rõ rệt. Chia ngẫu nhiên thành hai nhóm, một nhóm nhận can thiệp về thính giác (tư vấn thính giác và lắp máy trợ thính), một nhóm nhận giáo dục sức khoẻ làm đối chứng, theo dõi mỗi 6 tháng một lần. Điểm chính là thay đổi điểm nhận thức toàn phần sau 3 năm. Trong phân tích chính gộp hai quần thể nguồn, hai nhóm không khác biệt có ý nghĩa: nhóm can thiệp thính giác −0.200 (95% CI −0.256 đến −0.144), nhóm đối chứng −0.202 (95% CI −0.258 đến −0.145), chênh lệch 0.002 (95% CI −0.077 đến 0.081), P=0.96. Một phân tích độ nhạy cảy định sẵn cho thấy hiệu quả của can thiệp khác nhau giữa hai quần thể nguồn (tương tác P=0.010). Người tham gia đến từ nghiên cứu quan sát dài hạn về sức khoẻ tim mạch thì lớn tuổi hơn, có nhiều yếu tố nguy cơ suy giảm nhận thức hơn, điểm nhận thức nền thấp hơn. Kết luận của nhóm nghiên cứu là can thiệp thính giác có thể giảm mức thay đổi nhận thức trong ba năm ở nhóm có nguy cơ suy giảm cao, còn ở nhóm nguy cơ thấp thì không. Thử nghiệm không báo cáo biến cố bất lợi nghiêm trọng nào do thử nghiệm gây ra - Mức bằng chứng: A - Nguồn: Lin FR, Pike JR, Albert MS, et al.; ACHIEVE Collaborative Research Group (2023). Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): a multicentre, randomised controlled trial. Lancet, 402(10404):786-797. -- Ghi chú:Tranh cãi: Câu「máy trợ thính làm giảm gần một nửa suy giảm nhận thức」hay thấy trên mạng đến từ một nhóm con trong thử nghiệm này, không phải kết quả chính của nó, kết quả chính là hai nhóm không có khác biệt.「Lắp máy trợ thính」và「phòng bệnh Alzheimer」phải tách ra mà xem: cái trước thì nên làm, cái sau thì đừng lấy làm lý do. Thí sinh khuyết thị giác dự thi cao học được miễn phần thi nghe ngoại ngữ, xem mục 13 của phần này (hỗ trợ hợp lý). Người khuyết thính giác lái xe thì phải đeo thiết bị trợ thính, xem mục 15 của phần này (C5). +- Ghi chú:Tranh cãi: Câu「máy trợ thính làm giảm gần một nửa suy giảm nhận thức」hay gặp trên mạng đến từ một nhóm con của thử nghiệm này, không phải kết quả chính; kết quả chính là hai nhóm không khác biệt.「Lắp máy trợ thính」và「phòng chứng mất trí nhớ」phải tách ra: cái trước nên làm, cái sau đừng dùng làm lý do. Thí sinh khiếm thính dự kỳ thi tuyển sinh đại học được miễn phần thi nghe ngoại ngữ, xem mục 13 của phần này (biện pháp hỗ trợ hợp lý). Người khiếm thính lái xe phải mang thiết bị trợ thính, xem mục 15 của phần này (C5). -### 18. Nói một người lớn「không quản được việc của mình」thì phải xin tòa án công nhận, người nhà nói không tính +### 18. Nói một người thành niên「không quản lý được việc của mình」thì phải xin tòa án công nhận, người nhà nói không tính -- Chi phí: Nộp đơn ra tòa thì phải trả phí tố tụng, thường vài chục đến một trăm tệ. Phí giám định tính riêng, thường vài nghìn tệ, người xin trả trước. Đi hết thủ tục mất mấy tháng. -- Nói thẳng: Một người lớn có「không có năng lực pháp lý dân sự」hay「năng lực pháp lý dân sự bị hạn chế」hay không, chỉ tòa án mới công nhận được. Người nhà, cơ quan, tổ dân phố không tự quyết được. Phải do người có lợi ích liên quan hoặc tổ chức liên quan nộp đơn ra tòa. Ngược lại, khi sức khoẻ trí tuệ hoặc tâm thần hồi phục thì cũng có thể xin công nhận người đó đã hồi phục năng lực pháp lý dân sự. Thủ tục này vừa là hạn chế, vừa là bảo vệ. -- Lợi ích: Điều hai mươi mối của Bộ luật Dân sự (《民法典》):「不能辨认自己行为的成年人为无民事行为能力人，由其法定代理人代理实施民事法律行为。」— người lớn không nhận biết được hành vi của mình là người không có năng lực pháp lý dân sự, do người đại diện theo pháp luật của họ thay họ thực hiện giao dịch pháp lý dân sự. Điều hai mươi hai:「不能完全辨认自己行为的成年人为限制民事行为能力人，实施民事法律行为由其法定代理人代理或者经其法定代理人同意、追认；但是，可以独立实施纯获利益的民事法律行为或者与其智力、精神健康状况相适应的民事法律行为。」— người lớn không nhận biết được hoàn toàn hành vi của mình là người có năng lực pháp lý dân sự bị hạn chế, việc thực hiện giao dịch pháp lý dân sự phải do người đại diện theo pháp luật của họ thay, hoặc phải được người đó đồng ý, công nhận hậu nững; tuy nhiên người đó có thể tự thực hiện những giao dịch pháp lý dân sự thuần lợi cho mình, hoặc những giao dịch pháp lý dân sự phù hợp với tình trạng trí tuệ, sức khoẻ tâm thần của mình. Điều hai mươi bốn:「不能辨认或者不能完全辨认自己行为的成年人，其利害关系人或者有关组织，可以向人民法院申请认定该成年人为无民事行为能力人或者限制民事行为能力人。」— người lớn không nhận biết, hoặc không nhận biết hoàn toàn hành vi của mình thì người có lợi ích liên quan hoặc tổ chức liên quan có thể xin Tòa án nhân dân công nhận người lớn đó là người không có năng lực pháp lý dân sự hoặc có năng lực pháp lý dân sự bị hạn chế. Cùng điều khoản đó còn quy định, sau khi được công nhận, theo đơn của bản thân người đó, người có lợi ích liên quan hoặc tổ chức liên quan, Tòa án nhân dân có thể căn cứ tình trạng trí tuệ, sức khoẻ tâm thần đã hồi phục mà công nhận người đó hồi phục thành người có năng lực pháp lý dân sự bị hạn chế hoặc có năng lực pháp lý dân sự đầy đủ. Tổ chức liên quan gồm: Ủy ban nhân dân khu phố, Ủy ban nhân dân làng, trường học, cơ sở y tế, hội liên hiệp phụ nữ, hội người khuyết tật, tổ chức người cao niên lập hợp pháp, cơ quan dân vụ. Điều hai mươi ba quy định, người giám hộ của người không có năng lực pháp lý dân sự và của người có năng lực pháp lý dân sự bị hạn chế chính là người đại diện theo pháp luật của họ (toàn quốc, có hiệu lực từ ngày 1 tháng 1 năm 2021) +- Chi phí: Xin với tòa án thì phải nộp phí tụng, thường vài chục tệ đến một trăm tệ. Phí giám định tính riêng, thường vài nghìn tệ, do người nộp đơn tạm trả trước. Đi hết cảnh thì mất mấy tháng. +- Nói thẳng: Một người thành niên có「không có năng lực hành vi dân sự」hay「có năng lực hành vi dân sự bị giới hạn」hay không, chỉ tòa án mới công nhận được. Người nhà, đơn vị, ủy ban nhân dân khu phố không tự nói quyết được. Phải do người có lợi ích liên quan hoặc tổ chức có liên quan nộp đơn với tòa án. Ngược lại, khi trí tuệ hoặc sức khoẻ tâm thần đã hồi phục thì cũng có thể xin công nhận là người đó đã khôi phục năng lực hành vi. Thủ tục này vừa là hạn chế, vừa là bảo vệ. +- Lợi ích: Điều hai mươi mốt của Bộ luật Dân sự:「不能辨认自己行为的成年人为无民事行为能力人，由其法定代理人代理实施民事法律行为。」— người thành niên không nhận biết được hành vi của mình là người không có năng lực hành vi dân sự, việc thực hiện giao dịch pháp luật dân sự do người đại diện theo pháp luật của họ thay. Điều hai mươi hai:「不能完全辨认自己行为的成年人为限制民事行为能力人，实施民事法律行为由其法定代理人代理或者经其法定代理人同意、追认；但是，可以独立实施纯获利益的民事法律行为或者与其智力、精神健康状况相适应的民事法律行为。」— người thành niên không nhận biết được hoàn toàn hành vi của mình là người có năng lực hành vi dân sự bị giới hạn, việc thực hiện giao dịch pháp luật dân sự do người đại diện theo pháp luật của họ thay hoặc phải được người đó chấp thuận rồi công nhận sau; nhưng họ có thể tự thực hiện giao dịch pháp luật dân sự thuần lợi, hoặc giao dịch phù hợp với trình độ trí tuệ và tình trạng sức khoẻ tâm thần của mình. Điều hai mươi tư:「不能辨认或者不能完全辨认自己行为的成年人，其利害关系人或者有关组织，可以向人民法院申请认定该成年人为无民事行为能力人或者限制民事行为能力人。」— người thành niên không nhận biết được hoặc không nhận biết được hoàn toàn hành vi của mình, người có lợi ích liên quan hoặc tổ chức có liên quan có thể xin Tòa án nhân dân công nhận người thành niên đó là người không có năng lực hành vi dân sự hoặc có năng lực hành vi dân sự bị giới hạn. Cùng điều đó còn quy định, sau khi đã được công nhận thì theo đơn của bản thân người đó, của người có lợi ích liên quan hoặc của tổ chức có liên quan, Tòa án nhân dân có thể căn cứ tình trạng trí tuệ và sức khoẻ tâm thần đã hồi phục để công nhận họ đã khôi phục thành người có năng lực hành vi dân sự bị giới hạn hoặc người có năng lực hành vi dân sự đầy đủ. Tổ chức có liên quan gồm: ủy ban nhân dân khu phố, ủy ban nhân dân làng, trường học, cơ sở y tế, hội liên hiệp phụ nữ, hội liên hiệp người khuyết tật, tổ chức người cao tuổi lập theo pháp luật, cơ quan dân vụ và các cơ quan khác. Điều hai mươi ba quy định, người giám hộ của người không có năng lực hành vi dân sự và của người có năng lực hành vi dân sự bị giới hạn là người đại diện theo pháp luật của họ (toàn quốc, có hiệu lực từ ngày 1 tháng 1 năm 2021) - Mức bằng chứng: A -- Nguồn: Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc (中华人民共和国民法典)（第二十一条、第二十二条、第二十三条、第二十四条）. -- Ghi chú: Thủ tục xin đi theo thủ tục đặc biệt của tòa án, nhanh hơn vụ án dân sự thường nhưng vẫn mất mấy tháng. Phải nộp giấy chứng nhận chẩn đoán của bệnh viện, tòa thường còn ủy thác cho bên giám định tư pháp, phí giám định thường vài nghìn tệ, người xin trả trước. Không qua thủ tục này mà ký thay, phân phối tài sản của người khác thì về sau có thể bị xác nhận là vô hiệu, nhà đã bán cũng có thể phải hoàn trả. Còn một điểm phải nhớ: năng lực pháp lý dân sự bị hạn chế không có nghĩa là không làm được gì. Những việc phù hợp với trí tuệ, tình trạng tâm thần của họ, cùng những việc thuần lợi cho họ, họ tự làm được, đừng thay tất cả. Người giám hộ xác định thế nào, xem mục 19 của phần này (người giám hộ). +- Nguồn: Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc (中华人民共和国民法典，第二十一条、第二十二条、第二十三条、第二十四条). +- Ghi chú: Đơn này đi theo thủ tục đặc biệt của tòa án, nhanh hơn vụ án dân sự thường, nhưng vẫn mất mấy tháng. Phải nộp giấy chứng nhận chẩn đoán của bệnh viện, phần lớn tòa án còn ủy thác cho giám định tư pháp, phí giám định thường vài nghìn tệ, do người nộp đơn tạm trả trước. Không qua thủ tục này mà ký thay hoặc xử lý tài sản của người khác thì về sau có thể bị công nhận là vô hiệu, căn nhà đã bán cũng có thể phải hoàn trả. Còn một điểm phải nhớ: có năng lực hành vi dân sự bị giới hạn không có nghĩa là không làm được gì cả. Việc phù hợp với trình độ trí tuệ, tình trạng tâm thần của họ, và việc thuần lợi, thì họ tự làm được, đừng thay tất cả. Người giám hộ được xác định thế nào thì xem mục 19 của phần này (người giám hộ). -### 19. Người giám hộ của người lớn được xác định theo thứ tự luật định; người được giám hộ làm hại người khác thì người giám hộ bồi thường +### 19. Người giám hộ của người thành niên được xác định theo thứ tự luật định; người được giám hộ gây hại cho người khác thì người giám hộ bồi thường -- Chi phí: Không tốn tiền. Khi có tranh chấp thì phải tìm Ủy ban nhân dân khu phố, Ủy ban nhân dân làng, cơ quan dân vụ hoặc tòa án để giải quyết. -- Nói thẳng: Sau khi người lớn mất năng lực hành vi, người giám hộ được xếp theo thứ tự: vợ chồng, cha mẹ và con cái, người thân thích khác, cuối cùng là cá nhân hoặc tổ chức khác sẵn lòng đảm nhiệm. Tranh chấp thì Ủy ban nhân dân khu phố, Ủy ban nhân dân làng hoặc cơ quan dân vụ chỉ định, không đồng ý thì kiện ra tòa. Khi người đó còn minh mẫn thì có thể chỉ định bằng văn bản người giám hộ về sau. Người được giám hộ làm hại người khác thì người giám hộ bồi thường. -- Lợi ích: Điều hai mươi tám của Bộ luật Dân sự: người lớn không có năng lực pháp lý dân sự hoặc có năng lực pháp lý dân sự bị hạn chế thì người giám hộ do những người có năng lực giám hộ dưới đây theo thứ tự đảm nhiệm:「（一）配偶；（二）父母、子女；（三）其他近亲属；（四）其他愿意担任监护人的个人或者组织，但是须经被监护人住所地的居民委员会、村民委员会或者民政部门同意。」— (một) vợ chồng; (hai) cha mẹ, con cái; (ba) người thân thích khác; (bốn) cá nhân hoặc tổ chức khác sẵn lòng làm người giám hộ, nhưng phải được Ủy ban nhân dân khu phố, Ủy ban nhân dân làng nơi người được giám hộ cư trú hoặc cơ quan dân vụ đồng ý. Điều ba mươi mốt: có tranh chấp về việc xác định người giám hộ thì Ủy ban nhân dân khu phố, Ủy ban nhân dân làng nơi người được giám hộ cư trú hoặc cơ quan dân vụ chỉ định, không đồng ý với chỉ định thì có thể xin Tòa án nhân dân chỉ định, cũng có thể trực tiếp xin Tòa án nhân dân. Khi chỉ định thì「应当尊重被监护人的真实意愿，按照最有利于被监护人的原则」— phải tôn trọng ý muốn thật của người được giám hộ, theo nguyên tắc có lợi nhất cho người được giám hộ. Trước khi chỉ định mà quyền lợi rơi vào trạng thái không ai bảo vệ thì Ủy ban nhân dân khu phố, Ủy ban nhân dân làng, tổ chức liên quan theo luật định hoặc cơ quan dân vụ làm người giám hộ tạm thời. Điều ba mươi ba:「具有完全民事行为能力的成年人，可以与其近亲属、其他愿意担任监护人的个人或者组织事先协商，以书面形式确定自己的监护人」— người lớn có năng lực pháp lý dân sự đầy đủ có thể thỏa thuận trước với người thân thích, cá nhân hoặc tổ chức khác sẵn lòng làm người giám hộ, rồi xác định người giám hộ của mình bằng văn bản. Điều một nghìn một trăm tám mươi tám:「无民事行为能力人、限制民事行为能力人造成他人损害的，由监护人承担侵权责任。监护人尽到监护职责的，可以减轻其侵权责任。」— người không có năng lực pháp lý dân sự, người có năng lực pháp lý dân sự bị hạn chế gây thiệt hại cho người khác thì người giám hộ chịu trách nhiệm về tội phạm gây thiệt hại. Người giám hộ đã làm tròn nghĩa vụ giám hộ thì có thể giảm trách nhiệm đó. Người có tài sản thì chi phí bồi thường lấy từ tài sản của chính người đó, phần thiếu thì người giám hộ bồi thường. Điều một nghìn một trăm tám mươi chín: người giám hộ ủy thác nghĩa vụ giám hộ cho người khác thì người giám hộ vẫn phải chịu trách nhiệm về tội phạm gây thiệt hại, người được ủy thác có lỗi thì chịu trách nhiệm tương ứng (toàn quốc, có hiệu lực từ ngày 1 tháng 1 năm 2021) +- Chi phí: Không tốn tiền. Khi có tranh chấp thì phải đến ủy ban nhân dân khu phố, ủy ban nhân dân làng, cơ quan dân vụ hoặc tòa án để giải quyết. +- Nói thẳng: Người thành niên mất năng lực hành vi thì người giám hộ được xác định theo thứ tự: người phối ngẫu, cha mẹ và con cái, người thân thích khác, cuối cùng là người khác hoặc tổ chức sẵn sàng làm. Tranh chấp thì ủy ban nhân dân khu phố, ủy ban nhân dân làng hoặc cơ quan dân vụ chỉ định; không chịu thì kiện ra tòa án. Khi người đó còn minh mẫn thì có thể chỉ định bằng văn bản người giám hộ cho tương lai. Người được giám hộ gây hại cho người khác thì người giám hộ bồi thường. +- Lợi ích: Điều hai mươi tám của Bộ luật Dân sự: người thành niên không có năng lực hành vi dân sự hoặc có năng lực hành vi dân sự bị giới hạn, do những người có năng lực giám hộ sau đây theo thứ tự làm người giám hộ:「（一）配偶；（二）父母、子女；（三）其他近亲属；（四）其他愿意担任监护人的个人或者组织，但是须经被监护人住所地的居民委员会、村民委员会或者民政部门同意。」— (một) người phối ngẫu; (hai) cha mẹ, con cái; (ba) người thân thích khác; (bốn) người hay tổ chức khác sẵn sàng làm người giám hộ, nhưng phải được ủy ban nhân dân khu phố, ủy ban nhân dân làng nơi người được giám hộ cư trú hoặc cơ quan dân vụ đồng ý. Điều ba mươi mốt: khi có tranh chấp về việc xác định người giám hộ thì ủy ban nhân dân khu phố, ủy ban nhân dân làng nơi người được giám hộ cư trú hoặc cơ quan dân vụ chỉ định; không chịu chỉ định thì có thể xin Tòa án nhân dân chỉ định, cũng có thể trực tiếp nộp đơn với Tòa án nhân dân. Khi chỉ định thì「应当尊重被监护人的真实意愿，按照最有利于被监护人的原则」— phải tôn trọng ý chí thật của người được giám hộ, theo nguyên tắc có lợi nhất cho người được giám hộ. Trước khi có chỉ định mà quyền lợi ở trạng thái không ai bảo vệ thì ủy ban nhân dân khu phố, ủy ban nhân dân làng, tổ chức có liên quan theo pháp luật hoặc cơ quan dân vụ làm người giám hộ tạm thời. Điều ba mươi ba:「具有完全民事行为能力的成年人，可以与其近亲属、其他愿意担任监护人的个人或者组织事先协商，以书面形式确定自己的监护人」— người thành niên có năng lực hành vi dân sự đầy đủ có thể thỏa thuận trước với người thân thích, hoặc với người hay tổ chức khác sẵn sàng làm người giám hộ, để bằng văn bản xác định người giám hộ của chính mình. Điều một nghìn một trăm tám mươi tám:「无民事行为能力人、限制民事行为能力人造成他人损害的，由监护人承担侵权责任。监护人尽到监护职责的，可以减轻其侵权责任。」— người không có năng lực hành vi dân sự và người có năng lực hành vi dân sự bị giới hạn gây hại cho người khác thì người giám hộ chịu trách nhiệm vi phạm; người giám hộ đã làm tròn nghĩa vụ giám hộ thì có thể giảm trách nhiệm vi phạm. Người đó có tài sản thì phí bồi thường chi từ tài sản của bản thân, phần thiếu thì người giám hộ bồi thường. Điều một nghìn một trăm tám mươi chín: người giám hộ ủy thác nghĩa vụ giám hộ cho người khác thì người giám hộ vẫn phải chịu trách nhiệm vi phạm, người được ủy thác có lỗi thì chịu trách nhiệm tương ứng (toàn quốc, có hiệu lực từ ngày 1 tháng 1 năm 2021) - Mức bằng chứng: A -- Nguồn: Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc (中华人民共和国民法典)（第二十八条、第三十一条、第三十三条、第一千一百八十八条、第一千一百八十九条）. -- Ghi chú: Điều ba mươi ba chính là giám hộ tự nguyện bằng văn bản. Khi người đó còn minh mẫn thì dùng văn bản chốt người được chọn, tiết kiệm được phần lớn tranh chấp về sau, cụ thể viết thế nào xem mục 1 ở phần 17 (giám hộ tự nguyện bằng văn bản). Đoạn bồi thường thì phải cẩn thận: giao người cho cơ sở nuôi dưỡng hay bà con trông coi thì trách nhiệm vẫn ở người giám hộ, người được ủy thác chỉ chịu trách nhiệm tương ứng khi có lỗi. Nên mua cho người được giám hộ một bảo hiểm trách nhiệm người giám hộ là có ý nghĩa. Điều bảy mươi tám của Luật Sức khỏe Tâm thần còn quy định, người giám hộ bỏ rơi người bệnh hoặc không làm tròn nghĩa vụ giám hộ mà gây thiệt hại thì chịu trách nhiệm bồi thường theo pháp luật. +- Nguồn: Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc (第二十八条、第三十一条、第三十三条、第一千一百八十八条、第一千一百八十九条). +- Ghi chú: Điều ba mươi ba chính là giám hộ tự nguyện bằng văn bản. Khi người đó còn minh mẫn thì dùng văn bản chốt người được chọn, sau này khỏi phải tranh chấp phần lớn, viết cụ thể thế nào thì xem mục 1 ở phần 17 (giám hộ tự nguyện bằng văn bản). Đoạn bồi thường phải cẩn thận: giao người cho cơ sở nuôi dưỡng hay người thân chăm sóc thì trách nhiệm vẫn nằm ở người giám hộ, người được ủy thác chỉ chịu trách nhiệm tương ứng khi bản thân có lỗi. Vì vậy mua cho người được giám hộ một bảo hiểm trách nhiệm người giám hộ là có nghĩa. Điều bảy mươi tám của Luật Sức khỏe Tâm thần còn quy định, người giám hộ bỏ rơi người bệnh hoặc không làm tròn nghĩa vụ giám hộ mà gây hại thì phải chịu trách nhiệm bồi thường theo pháp luật. -### 20. Bị từ chối tuyển, không được chuyển hợp đồng chính thức, bị trả thiếu lương vì khuyết tật thì có thể kiện, nhưng phải tính trước chi phí của cả thủ tục +### 20. Bị từ chối tuyển, không được chuyển biên chế, bị trả lương thiếu vì khuyết tật thì có thể kiện, nhưng hãy tính chi phí trong quá trình trước -- Chi phí: Phí luật sư vài nghìn đến vài chục nghìn tệ, tự bỏ ra. Bản án sơ thẩm theo thủ tục thường từ 6 tháng trở lên, có thể kéo dài. Chứng cứ phải tự giữ. -- Nói thẳng: Pháp luật cấm phân biệt đối xử với người khuyết tật trong việc tuyển dụng, chuyển hợp đồng chính thức, thăng tiến, xét chức danh, thù lao, bảo hiểm xã hội. Bạn có thể khiếu nại với cơ quan chủ trì, cũng có thể kiện ra tòa. Nhưng thắng kiện xong thì phía bên kia không trả phí luật sư cho bạn. Vì vậy hãy khiếu nại trước, đồng thời xin trợ giúp pháp lý, rồi mới nghĩ tới kiện. -- Lợi ích: Điều ba của Luật Bảo vệ Người khuyết tật (《残疾人保障法》):「禁止基于残疾的歧视。禁止侮辱、侵害残疾人。」— cấm phân biệt đối xử dựa trên khuyết tật. Cấm xúc phạm, xâm phạm người khuyết tật. Điều ba mươi tám:「在职工的招用、转正、晋级、职称评定、劳动报酬、生活福利、休息休假、社会保险等方面，不得歧视残疾人。」— ở các mặt tuyển dụng, chuyển hợp đồng chính thức, thăng tiến, xét chức danh, thù lao lao động, phúc lợi sinh hoạt, nghỉ ngơi, bảo hiểm xã hội thì không được phân biệt đối xử với người khuyết tật. Cùng điều khoản đó còn quy định, đơn vị có công nhân viên khuyết tật phải căn cứ đặc điểm của họ mà tạo điều kiện lao động và bảo vệ lao động thích hợp, và tuỳ nhu cầu thực tế mà cải tạo nơi làm việc, thiết bị lao động và cơ sở sinh hoạt. Điều sáu mươi bốn:「违反本法规定，在职工的招用等方面歧视残疾人的，由有关主管部门责令改正；残疾人劳动者可以依法向人民法院提起诉讼。」— vi phạm luật này mà phân biệt đối xử với người khuyết tật trong việc tuyển dụng công nhân viên thì bị cơ quan chủ trì liên quan yêu cầu sửa; người lao động khuyết tật có quyền khởi kiện ra Tòa án nhân dân theo pháp luật. Điều bốn của Quy định về Việc làm cho người khuyết tật (《残疾人就业条例》):「禁止在就业中歧视残疾人。」— cấm phân biệt đối xử với người khuyết tật trong việc làm. Điều mười ba: đơn vị sử dụng lao động không được phân biệt đối xử với công nhân viên khuyết tật về chức vụ, thăng tiến, xét chức danh, thù lao, bảo hiểm xã hội, phúc lợi sinh hoạt. Điều bốn mươi hai của Luật Trợ giúp pháp lý quy định, người xin có tài liệu chứng minh thuộc「nhóm nhân dân đặc biệt không có nguồn sống cố định như trẻ vị thành niên, người cao niên, người khuyết tật」thì được miễn kiểm tra hoàn cảnh khó khăn tài chính. Điều bốn mươi lăm quy định, cơ quan trợ giúp pháp lý có cung cấp dịch vụ trợ giúp pháp lý cho người cao niên, người khuyết tật thì phải tuỳ tình hình thực tế mà cung cấp thiết bị và dịch vụ không vật cản (toàn quốc) +- Chi phí: Phí luật sư vài nghìn tới vài chục nghìn tệ, do bạn tự trả. Bản án sơ thẩm theo thủ tục thông thường kéo dài từ 6 tháng, có thể kéo dài hơn. Bằng chứng bạn phải tự giữ. +- Nói thẳng: Pháp luật cấm phân biệt đối xử với người khuyết tật trong việc tuyển dụng, chuyển biên chế, thăng chức, xếp chức danh, lương, bảo hiểm xã hội. Bạn có thể khiếu nại với cơ quan chủ quản, cũng có thể kiện ra tòa án. Nhưng thắng kiện thì đối phương không trả phí luật sư hộ bạn. Vì vậy hãy khiếu nại trước, đồng thời xin trợ giúp pháp lý, rồi mới cân nhắc kiện. +- Lợi ích: Điều ba của Luật Bảo đảm người khuyết tật:「禁止基于残疾的歧视。禁止侮辱、侵害残疾人。」— cấm phân biệt đối xử dựa trên khuyết tật; cấm xúc phạm, xâm phạm người khuyết tật. Điều ba mươi tám:「在职工的招用、转正、晋级、职称评定、劳动报酬、生活福利、休息休假、社会保险等方面，不得歧视残疾人。」— trong việc tuyển dụng, chuyển biên chế, thăng chức, xếp chức danh, lương lao động, phúc lợi sinh hoạt, nghỉ ngơi, bảo hiểm xã hội cho công nhân viên, không được phân biệt đối xử với người khuyết tật. Cùng điều đó còn quy định, đơn vị có công nhân viên khuyết tật phải căn cứ đặc điểm của họ để cung cấp điều kiện lao động và bảo vệ lao động thích hợp, và tuỳ nhu cầu thực tế để cải tạo nơi làm việc, thiết bị lao động và cơ sở sinh hoạt. Điều sáu mươi tư:「违反本法规定，在职工的招用等方面歧视残疾人的，由有关主管部门责令改正；残疾人劳动者可以依法向人民法院提起诉讼。」— vi phạm luật này mà phân biệt đối xử với người khuyết tật trong việc tuyển dụng công nhân viên và các mặt khác, thì bị cơ quan chủ quản có liên quan buộc phải sửa chữa; người lao động khuyết tật có thể kiện Tòa án nhân dân theo pháp luật. Điều bốn của Quy định về việc làm của người khuyết tật:「禁止在就业中歧视残疾人。」— cấm phân biệt đối xử với người khuyết tật trong việc làm. Điều mười ba của quy định đó quy định, đơn vị sử dụng lao động không được phân biệt đối xử với công nhân viên khuyết tật trong việc thăng chức, thăng cấp, xếp chức danh, lương, bảo hiểm xã hội, phúc lợi sinh hoạt. Điều bốn mươi hai của Luật Trợ giúp pháp lý quy định, người nộp đơn có giấy tờ chứng minh thuộc「无固定生活来源的未成年人、老年人、残疾人等特定群体」thì được miễn kiểm tra tình trạng khó khăn về tài chính. Điều bốn mươi lăm quy định, cơ sở trợ giúp pháp lý có cung cấp dịch vụ trợ giúp pháp lý cho người cao tuổi, người khuyết tật thì phải theo tình hình thực tế mà cung cấp thiết bị và dịch vụ không vướng (toàn quốc) - Mức bằng chứng: A -- Nguồn: Luật Bảo vệ Người khuyết tật của Cộng hoà Nhân dân Trung Quốc (中华人民共和国残疾人保障法)（第三条、第三十八条、第六十四条）. ；Hội đồng Nhà nước (国务院) (2007). Quy định về Việc làm cho người khuyết tật (《残疾人就业条例》)(第四条、第十三条). ；Luật Trợ giúp pháp lý của Cộng hoà Nhân dân Trung Quốc (中华人民共和国法律援助法)（第四十二条、第四十五条）. -- Ghi chú: Chi phí của cả thủ tục phải tính trước. Bản án sơ thẩm theo thủ tục thường từ 6 tháng trở lên, có thể kéo dài; thủ tục đơn giản là 3 tháng. Phí luật sư không nằm trong khoản phí tố tụng liệt kê ở Điều sáu của Biện pháp thu phí tố tụng, cơ chế「bên thua kiện chịu」ở Điều hai mươi chín không bao gồm phí luật sư. Vì vậy hãy đi trước hai con đường rẻ tiền: khiếu nại với cơ quan nhân sự và xã hội hoặc hội người khuyết tật, để cơ quan chủ trì yêu cầu sửa; đồng thời xin trợ giúp pháp lý, Điều bốn mươi hai miễn cho người khuyết tật việc kiểm tra hoàn cảnh khó khăn tài chính. Thu thập chứng cứ là khó nhất. Khi bị từ chối phỏng vấn, phía bên kia hiếm khi để lại lý do bằng văn bản, ghi âm được thì ghi, giữ được lịch sử trò chuyện thì giữ. Người đã vào làm thì đi theo tranh chấp lao động ra hoà giải lao động, cách lưu chứng cứ và thời hiệu xem mục 8 ở phần 19 (lưu chứng cứ trước khi nghỉ việc). Người được lợi của mục này là bạn. +- Nguồn: Luật Bảo đảm người khuyết tật của Cộng hoà Nhân dân Trung Quốc (第三条、第三十八条、第六十四条). ；Hội đồng Nhà nước (国务院) (2007). Quy định về việc làm của người khuyết tật (残疾人就业条例，第四条、第十三条). ；Luật Trợ giúp pháp lý của Cộng hoà Nhân dân Trung Quốc (法律援助法，第四十二条、第四十五条). +- Ghi chú: Chi phí trong quá trình phải tính trước. Bản án sơ thẩm theo thủ tục thông thường kéo dài từ 6 tháng, có thể kéo dài hơn; thủ tục giản lược là 3 tháng. Phí luật sư không nằm trong các khoản phí tụng mà điều sáu của Biện pháp quản lý việc nộp phí tụng liệt kê; câu「bên thua kiện chịu phí」trong điều hai mươi chín không gồm phí luật sư. Vì vậy hãy đi trước hai đường rẻ tiền: khiếu nại với cơ quan nhân sự và xã hội hoặc hội người khuyết tật, để cơ quan chủ quản buộc phải sửa chữa; đồng thời xin trợ giúp pháp lý, điều bốn mươi hai miễn kiểm tra tình trạng khó khăn tài chính cho người khuyết tật. Thu thập chứng cứ là khó nhất. Khi bị từ chối phỏng vấn, phía bên kia hiếm khi để lại lý do bằng văn bản, ghi âm được, giữ lại được lịch sử trò chuyện thì cứ giữ. Đã đi làm rồi thì đi theo hoà giải lao động vì tranh chấp lao động, cách giữ chứng cứ và thời hiệu xem mục 8 ở phần 19 (giữ chứng cứ trước khi nghỉ việc). Người hưởng mục này là bạn. diff --git "a/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" index 0dc5a8b5..0684c81e 100644 --- "a/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" +++ "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" @@ -1,6 +1,6 @@ [← Về mục lục](../README.md) -# 34. Đừng để thuốc trong nhà gây tai nạo +# 34. Đừng dùng thuốc trong nhà sai cách Thuốc hạ sốt, thuốc giảm đau, thuốc cảm cúm, thuốc dạ dày, thuốc chống tiêu chảy, nhiều loại mua ở hiệu thuốc là tự mua được. Mua không cần đơn thuốc không có nghĩa là uống thế nào cũng được. Phần này không nói mỗi bệnh nên uống thuốc gì, chỉ chọn ra mấy việc chỉ cần một hành động là tránh được hậu quả nặng. From 4ef30ed2cfce987b8745dba05de207539251356f Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 12:42:50 +0700 Subject: [PATCH 16/25] =?UTF-8?q?=E8=8A=82=E6=A0=87=E9=A2=98=E7=BB=9F?= =?UTF-8?q?=E4=B8=80=E7=94=A8=20README=20=E7=9B=AE=E5=BD=95=E9=87=8C?= =?UTF-8?q?=E9=82=A3=E4=B8=80=E4=BB=BD=EF=BC=8C=E6=A3=80=E6=9F=A5=E9=87=8C?= =?UTF-8?q?=E5=AF=B9=E4=B8=80=E9=81=8D?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 页面的侧栏目录读的是 <语言>/README.md，译本 book/ 里的「# N. 节名」是另一处。 en 有 1 节、vi 有 3 节两处说法不同（en 32 出国留学、vi 16 慢性病、vi 24 看病、 vi 25 人走了以后），读者的两处搜索对不上。 check.mjs 从 README 的目录一节读出节名，compareSection 多带一个 toc 参数， 节标题跟目录不一致就报。对不上是真的会报的（试过把 en 32 的节标题换掉，check 立刻报出来，不是静悄悄地过）。 --- ...72\345\233\275\347\225\231\345\255\246.md" | 2 +- tools/i18n/check.mjs | 18 +- tools/i18n/structure.mjs | 7 +- ...16\345\201\232\347\224\237\346\204\217.md" | 210 ++++++++++++++++++ ...53\345\220\203\345\207\272\344\272\213.md" | 62 +++--- 5 files changed, 264 insertions(+), 35 deletions(-) create mode 100644 "vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" diff --git "a/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" "b/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" index 2b0014b8..00f24979 100644 --- "a/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" +++ "b/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" @@ -1,6 +1,6 @@ [← Back to the contents](../README.md) -# 32. Studying abroad: status, work, insurance and certification on return +# 32. Studying abroad: status, work, insurance, certification This section measures money and personal freedom. It does not discuss whether you should go abroad, and it does not compare schools. Four things below cost you if you do not know them: when your visa status counts as unbroken, where the ceiling on working hours is, which insurances and fees you must pay, and whether this diploma is recognised once you are back home. diff --git a/tools/i18n/check.mjs b/tools/i18n/check.mjs index 44746324..1ac037e5 100644 --- a/tools/i18n/check.mjs +++ b/tools/i18n/check.mjs @@ -30,6 +30,19 @@ const codes = only < 0 ? LOCALES.map(l => l.code) : [process.argv[only + 1]]; const problems = []; const report = (where, what) => problems.push(`${where}：${what}`); +/** 那门语言 README「目录」一节的节名，按节文件名（01-不要早死）索引。 + 页面的侧栏目录读的是 README，译本 book/ 里的节标题是另一处，两处该说同一句话。 */ +function tocTitles(code) { + const p = contentPath(code, 'README.md'); + if (!existsSync(resolve(ROOT, p))) return {}; + const S = strings(code); + const out = {}; + for (const m of read(p).matchAll(/^\s*\d+\.\s+\[([^\]]+)\]\((book\/[^)]+\.md)\)/gm)){ + out[m[2].slice('book/'.length).replace(/\.md$/, '')] = m[1]; + } + return out; +} + for (const code of codes){ if (code === SOURCE.code) continue; const L = LOCALE(code); @@ -37,6 +50,7 @@ for (const code of codes){ const srcDir = (SOURCE.contentDir ? SOURCE.contentDir + '/' : '') + 'book/'; const files = readdirSync(resolve(ROOT, srcDir)).filter(f => f.endsWith('.md')).sort(); + const toc = tocTitles(code); const have = []; for (const f of files){ // 译文所在的仓库目录：contentPath，不是 L.dir（那是页面上线上的地址，英文是空串） @@ -47,8 +61,8 @@ for (const code of codes){ const dst = entrySkeleton(read(dstPath), S); for (const e of dst.entries) e._locale = code; if (src.n !== dst.n) report(`${code} book/${f}`, `节号是 ${dst.n}，原文 ${src.n}`); - // 节标题和导读：条目之外，节首那几行也得翻 - for (const w of STRUCT.compareSection(src, dst, `${code} book/${f}`)) problems.push(w); + // 节标题和导读：条目之外，节首那几行也得翻。节标题还要跟 README 目录里那一份一致 + for (const w of STRUCT.compareSection(src, dst, `${code} book/${f}`, toc[f.replace(/\.md$/, '')])) problems.push(w); if (src.entries.length !== dst.entries.length) report(`${code} book/${f}`, `条目数 ${dst.entries.length}，原文 ${src.entries.length}`); for (let i = 0; i < Math.min(src.entries.length, dst.entries.length); i++) diff --git a/tools/i18n/structure.mjs b/tools/i18n/structure.mjs index 7f0f83a0..5b723a8a 100644 --- a/tools/i18n/structure.mjs +++ b/tools/i18n/structure.mjs @@ -210,13 +210,18 @@ export const STRUCT = { * 结果 en 和 vi 一半的节标题是中文的，有的还多出一份，页面上那一节渲染两遍。 * 这两条都是机器查得出来的：节标题跟原文一字不差就是没翻；导读里只要还有 * 整段没翻的中文就是没翻完。 */ - compareSection(a, b, tag) { + // `toc` 是那门语言 README 目录里的节名。给了就要求译本的节标题跟它一字不差： + // 页面侧栏读的是 README，book 文件的节标题是另一处，同一节说两句话的话读者的 + // 两处搜索对不上，这层该拦。 + compareSection(a, b, tag, toc) { const out = []; const han = s => /[㐀-鿿]/.test(s); // 汉字。引号里照抄的法条名不算漏译 if (a.n !== b.n) out.push(`${tag}：节号是 ${b.n}，原文 ${a.n}`); if (!b.title.trim()) out.push(`${tag}：节标题是空的`); else if (a.title && b.title === a.title) out.push(`${tag}：节标题「${b.title}」跟中文原文一字不差，这一行没翻`); + else if (toc && b.title !== toc) + out.push(`${tag}：节标题「${b.title}」跟目录里的「${toc}」不一致`); // 导读逐段比。节首的导读是散文，但「整段还是中文」是查得出来的：把一行里 // 照抄的中文（法条名、书名）剔掉之后，剩下的还是汉字，就是没翻。 diff --git "a/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" "b/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" new file mode 100644 index 00000000..d023a16b --- /dev/null +++ "b/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" @@ -0,0 +1,210 @@ +[← Về mục lục](../README.md) + +# 12. Khởi nghiệp: đừng làm mất khoản tiết kiệm của gia đình + +Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi ra tuổi thọ. Khi mở một cửa hàng hay lập một công ty, bước nào có thể khiến tiền của bạn, tiền trong nhà và tiền của người phối ngẫu cùng mất đi? Luật quy định thế nào, có tách ra trước được không? Những mục dễ làm mất vốn của cả nhà nhất lại dễ tránh nhất thì đặt trước. Vì vậy các mục phải ký giấy tờ nằm trước, các cách phải tự làm theo nằm sau. Mục nào chỉ dựa vào kinh nghiệm, không có tài liệu chống đỡ, đánh dấu mức C. + +### 1. Chỉ dùng số tiền mà lỡ mất trắng vẫn chịu được để khởi nghiệp, không đụng vốn của cả nhà, không vay tiền để mở hàng + +- Chi phí: Không tốn tiền. Trước khi bắt đầu, hãy tự đặt cho mình một con số: mất trắng số tiền đó cũng không ảnh hưởng tiền trả góp, nuôi con hay khám bệnh. Vượt quá con số đó thì không đầu tư. Tiền của người nhà và tiền vay mượn không tính vào con số đó. Khó nhất là lúc công việc có dấu hiệu lên, nhịn không thêm tiền vào nữa. +- Nói thẳng: Nợ của một hộ kinh doanh cá thể thì phải lấy tiền và tài sản đứng tên bạn để trả. Không phân biệt được là cả nhà cùng làm hay không, thì lấy tài sản của cả nhà. Lập công ty cũng không chắc an toàn: khi công ty không trả nổi nợ, phần vốn bạn đã cam kết góp mà chưa đến hạn sẽ bị yêu cầu góp sớm. Cho nên góp bao nhiêu thì căn cứ vào「lỡ mất trắng vẫn còn trả góp, khám bệnh」. +- Lợi ích: Lỗ không chỉ là mất tiền. Nợ của một hộ kinh doanh cá thể, nếu việc kinh doanh do một mình bạn làm, thì lấy tiền và tài sản đứng tên bạn trả (chịu trách nhiệm bằng tài sản cá nhân). Nếu cả nhà cùng làm, hoặc không phân biệt được ai làm, thì lấy tiền và tài sản của cả nhà trả (chịu trách nhiệm bằng tài sản gia đình). Nợ có chữ ký của cả hai vợ chồng là nợ chung, hai người cùng trả. Một bên có đồng ý sau đó cũng được tính là. Một bên vay tiền dưới tên mình, chỉ cần chủ nợ chứng minh được số tiền đó dùng vào việc kinh doanh hai vợ chồng cùng làm, thì hai người vẫn cùng trả. Về phía công ty cũng có một quy tắc: khi công ty không trả được nợ đến hạn, chủ nợ có thể yêu cầu cổ phần đã cam kết góp trong vốn đăng ký mà chưa đến hạn nộp (vốn góp được cam kết, 认缴出资) góp sớm (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第五十六、一千零六十四条）. ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Luật Công ty, 公司法（2023 年修订，第四十七、五十四条）. +- Ghi chú: 「亏得起」 nghĩa là: mất trắng số tiền đó, bạn cũng không phải bán nhà, không phải vay tiền người thân để lấp lỗ hổng. Có bao nhiêu doanh nghiệp sống sót, trung bình sống được mấy năm, loại thống kê chính thức đó lần này không tìm được bản gốc để đối chiếu trên trang Cục Thống kê Quốc gia và Cục Quản lý thị trường quốc gia. Vì vậy phần chính văn không ghi con số, chỉ ghi hậu quả pháp lý, để ở mức B. + +### 2. Đừng ký bảo đảm cá nhân cho khoản vay của công ty, người phối ngẫu càng đừng ký theo + +- Chi phí: Không tốn tiền. Khi ngân hàng, nhà cung cấp hay chủ nhà bảo bạn ký「bảo đảm liên đới cá nhân của cổ phân」hoặc đòi「chữ ký của người phối ngẫu」, hãy dừng lại. Ký trang đó có nghĩa là bạn tự tay trao đi lớp bảo vệ「nợ của công ty không tìm đến cá nhân」(từ bỏ trách nhiệm hữu hạn). Không ký được thì đừng ký. Bắt buộc phải ký thì chỉ ký「bảo đảm chung」, và ghi rõ số tiền nhiều nhất bạn bảo đảm. Khó nhất là bên kia hay nói「không ký thì không cho vay」, bạn phải chịu được. +- Nói thẳng: Cái bạn ký là「bảo đảm liên đới」: công ty không trả được, chủ nợ có thể bỏ qua công ty và đòi thẳng bạn. Ghi「bảo đảm chung」thì chủ nợ phải kiện công ty ra tòa hoặc trọng tài trước, bán xong tài sản công ty còn thiếu bao nhiêu thì mới đến phần bạn. Người phối ngẫu cùng ký, tài sản của hai nhà cùng mất, chữ「hữu hạn」của công ty hữu hạn không còn tác dụng với bạn. +- Lợi ích: Câu chữ trên giấy bảo đảm nghĩa là「người nợ không trả, bạn trả thay」. Ghi「bảo đảm liên đới」thì chủ nợ không cần tìm công ty, có thể đến tận nhà bạn đòi. Ghi「bảo đảm chung」thì chủ nợ phải kiện công ty hoặc trọng tại trước, rồi mới xin tòa thi hành án cưỡng chế bán tài sản công ty trừ nợ. Bán xong còn thiếu bao nhiêu thì bạn mới phải bù. Nợ có chữ ký của cả hai vợ chồng là nợ chung, hai người cùng trả. Một bên có đồng ý sau đó cũng được tính là. Người phối ngẫu đã ký thì tài sản của hai nhà cùng mất (toàn quốc, có hiệu lực từ năm 2021) +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第六百八十一、六百八十七、六百八十八、一千零六十四条）. +- Ghi chú: Ngân hàng cho vay doanh nghiệp nhỏ và siêu nhỏ gần như lúc nào cũng bắt cổ phân công ty ký bảo đảm liên đới. Vẫn vay được như thường, chỉ là trước khi ký phải hiểu rõ trang đó trao đi cái gì, rồi giữ số tiền bảo đảm trong mức「lỡ mất trắng」của mục 1 (chỉ dùng số tiền mà lỡ mất trắng vẫn chịu được để khởi nghiệp). Quy tắc chung về phiếu vay và giấy bảo đảm, xem mục 18 ở phần 8. + +### 3. Chọn đúng loại hình kinh doanh trước khi mở hàng: hộ cá thể và cổ phần công chung phải trả đến cùng, chỉ công ty hữu hạn mới「hữu hạn」 + +- Chi phí: Đăng ký làm online trên cổng dịch vụ công của địa phương, mỗi lần mất mấy tiếng. Sau khi lập công ty hữu hạn thì hằng tháng phải làm sổ sách và kê khai thuế, đây là khoản chi phí lâu dài: tự làm thì tốn thời gian, thuê người làm thay thì tốn tiền, giá ở mỗi nơi khác nhau. +- Nói thẳng: Hộ cá thể và cổ phần công chung phải trả đến cùng, nợ bao nhiêu trả bấy nhiêu. Cổ phân của công ty hữu hạn chỉ phải đến mức số tiền đã cam kết lúc đăng ký. Nhưng cam kết bao nhiêu thì trong 5 năm phải thật sự đưa ra bấy nhiêu: điền 1.000.000 tệ là có thể nợ 1.000.000 tệ. Dùng chung sổ sách của công ty và của nhà thì chữ「hữu hạn」cũng mất hiệu lực. +- Lợi ích: Nợ của một hộ kinh doanh cá thể phải lấy tiền và tài sản của cá nhân trả, hoặc lấy tiền và tài sản của cả nhà trả. Cổ phần công chung trong doanh nghiệp hợp tác thông thường chịu trách nhiệm liên đới không giới hạn đối với nợ của công ty hợp tác. Nghĩa là nợ bao nhiêu trả bấy nhiêu, chưa trả hết thì chưa xong. Chủ nợ còn có thể bám vào đúng một người, đòi người đó trả toàn bộ. Công ty hợp tác hữu hạn thì khác: cổ phần hữu hạn bên trong chỉ phải đến mức số tiền mình cam kết góp (以认缴出资为限). Cổ phân của công ty trách nhiệm hữu hạn cũng chỉ phải đến mức số tiền mình cam kết góp (以认缴出资额为限). Nhưng chữ「hữu hạn」mất hiệu lực trong ba trường hợp. Một là số tiền cam kết phải thật sự góp đủ trong 5 năm kể từ ngày công ty thành lập, và khi công ty không trả được nợ đến hạn thì chủ nợ có thể yêu cầu góp sớm. Hai là số tiền đã góp vào rồi lại rút ra (抽逃出资) thì phải hoàn lại. Ba là lấy công ty làm vỏ để trốn nợ. Công ty chỉ có một cổ phân, lại không chứng minh được tiền của công ty và tiền của bạn là hai khoản riêng biệt, cũng tính vào trường hợp thứ ba. Dưới hai cách làm này, nợ của công ty bạn cũng phải trả cùng (chịu trách nhiệm liên đới, 承担连带责任)（toàn quốc, có hiệu lực từ tháng 7 năm 2024） +- Mức bằng chứng: A +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第五十六条）. ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2006). Luật Doanh nghiệp hợp tác, 合伙企业法（2006 年修订，第二条）. ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Luật Công ty, 公司法（2023 年修订，第四、二十三、四十七、五十三、五十四条）. +- Ghi chú: Điều kiện của chữ「hữu hạn」là công và tư phải phân minh: tài khoản công ty và tài khoản cá nhân không dùng chung, không dùng tiền công ty trả chi phí trong nhà, sổ sách giữ đầy đủ. Vốn đăng ký đừng điền số lớn cho đẹp mắt. Cam kết góp bao nhiêu thì trong 5 năm phải thật sự đưa ra bấy nhiêu. Điền 1.000.000 tệ là tạo ra một khoản nợ tiềm tàng 1.000.000 tệ. + +### 4. Đừng làm cổ phân danh nghĩa, đừng giữ cổ phần thay người khác + +- Chi phí: Không tốn tiền. Có người nói「cho mượn giấy tờ tùy thân để làm cổ phân」hay「cổ phần tạm đứng tên bạn」, từ chối hết. +- Nói thẳng: Tên bạn ghi trong sổ đăng ký thì về mặt pháp luật bạn là cổ phân. Số tiền cam kết góp lúc đăng ký phải góp đúng hạn, không góp nổi thì ngoài bù số thiếu còn phải bồi thường thiệt hại cho công ty. Khi lập công ty mà cổ phân khác không thật sự đưa tiền vào, thiếu bao nhiêu bạn cũng phải trả bấy nhiêu. Người khác làm ăn, nợ lại ghi tên bạn. +- Lợi ích: Cổ phân ghi trong sổ đăng ký chính là cổ phân về mặt pháp luật. Bạn phải góp đủ số tiền đã cam kết, không thiếu một xu, theo thời hạn điều lệ công ty quy định (góp đầy đủ vốn góp được cam kết, 足额缴纳认缴出资). Không góp đúng hạn thì ngoài bù số thiếu còn phải bồi thường thiệt hại cho công ty. Khi lập công ty mà có cổ phân không thật sự đưa tiền vào, các cổ phân khác có mặt lúc lập công ty cùng chia số thiếu đó (chịu trách nhiệm liên đới, 承担连带责任). Khi công ty không trả được nợ đến hạn, chủ nợ có thể yêu cầu góp sớm số tiền chưa đến hạn. Người thật sự làm ăn không có trong sổ đăng ký, mà nợ lại ghi tên bạn (toàn quốc, có hiệu lực từ tháng 7 năm 2024) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Luật Công ty, 公司法（2023 年修订，第四十七、四十九、五十、五十四条）. ; Tòa án nhân dân tối cao (最高人民法院) (2020 修正). 关于适用《中华人民共和国公司法》若干问题的规定（三）（法释〔2011〕3 号，2014 年法释〔2014〕2 号、2020 年法释〔2020〕18 号两次修正，第二十四、二十五条）：「如无法律规定的无效情形，人民法院应当认定该合同有效」「实际出资人未经公司其他股东半数以上同意，请求公司变更股东……人民法院不予支持」「名义股东处分股权造成实际出资人损失，实际出资人请求名义股东承担赔偿责任的，人民法院应予支持」— Nói thẳng: hợp đồng có hiệu lực trừ khi luật có lý do vô hiệu; yêu cầu đổi đăng ký cổ phân mà không được hơn một nửa cổ phân khác đồng ý thì tòa không hỗ trợ; cổ phân danh nghĩa chuyển nhượng cổ phần gây thiệt hại cho nhà đầu tư thật thì tòa hỗ trợ yêu cầu bồi thường của nhà đầu tư thật với cổ phân danh nghĩa. （Cơ sở dữ liệu luật và quy định quốc gia, 国家法律法规数据库）; Tòa án nhân dân tối cao (最高人民法院) (2025). 关于适用《中华人民共和国公司法》若干问题的解释（征求意见稿）第九十条. +- Ghi chú: Giữa cổ phân danh nghĩa và người thật sự bỏ tiền, thỏa thuận giữ hộ có hiệu lực nếu không rơi vào lý do vô hiệu. Nhưng người thật sự bỏ tiền muốn đổi tên đăng ký sang mình thì phải được hơn một nửa cổ phân khác đồng ý. Người đứng tên danh nghĩa bán hoặc thế chấp cổ phần thì tòa xử lý theo quy định thiện chí của Bộ luật Dân sự. Bên mua không biết, giá hợp lý, đã làm thủ tục đăng ký thì người bỏ tiền không lấy lại được cổ phần, chỉ có thể đòi người đứng tên danh nghĩa bồi thường. Tháng 9 năm 2025 đã có bản dự thảo giải thích mới thay cho bản giải thích tư pháp này, bản chính thức có hiệu lực thì toàn văn bị bãi bỏ, lấy bản sửa đổi năm 2020 còn hiệu lực tính đến tháng 9 năm 2026 làm chuẩn. Rủi ro làm đại diện pháp lý danh nghĩa, xem mục 28 ở phần 8. + +### 5. Trước khi nhượng quyền, kiểm tra hồ sơ đăng ký ở Bộ Thương mại, lấy bản công bố bằng văn bản, ghi「thời gian cân nhắc」vào hợp đồng + +- Chi phí: Không tốn tiền. Vào website「商业特许经营信息管理系统」của Bộ Thương mại, kiểm tra bên thương hiệu đã đăng ký hay chưa. Yêu cầu bên kia viết ra giấy tờ tất cả những điều phải nói trước khi ký, đưa bạn cùng bản hợp đồng. Mất mấy tiếng là xong. +- Nói thẳng: Theo quy định, bên thương hiệu phải tự mở 2 cửa hàng trực tiếp và vận hành đủ 1 năm, sau khi ký còn phải đăng ký với cơ quan thương mại trong 15 ngày. Trước khi ký, bên đó phải đưa cho bạn văn bản giấy trình bày 12 nội dung, gồm phí nhượng quyền, tiền bảo đảm hoàn lại thế nào, các cửa hàng hiện có có lãi hay không. Không tra được hồ sơ đăng ký, không chịu đưa hồ sơ, đòi nộp tiền mới cho xem hợp đồng thì đi luôn. +- Lợi ích: Quy định yêu cầu bên thương hiệu (trong quy định gọi là bên nhượng quyền, 特许人) phải có ít nhất 2 cửa hàng trực tiếp do chính mình mở và tự vận hành, vận hành trên 1 năm. Tính từ ngày ký hợp đồng nhượng quyền đầu tiên, trong 15 ngày bên đó phải đăng ký với cơ quan quản lý thương mại có thẩm quyền. Không đăng ký thì phạt từ 1 vạn đến 5 vạn tệ. Quá hạn vẫn không đăng ký thì phạt từ 5 vạn đến 10 vạn tệ và phải công bố ra bên ngoài. Trước khi ký, bên đó phải đưa cho bạn văn bản giấy trình bày tất cả những điều phải nói, tổng cộng 12 nội dung. Trong đó có phí nhượng quyền thu theo những dạng nào, mỗi dạng bao nhiêu tiền; tiền bảo đảm được hoàn lại trong trường hợp nào; hiện có bao nhiêu cửa hàng nhượng quyền, kinh doanh ra sao. Còn có bản tóm tắt báo cáo kiểm toán đã kiểm tra sổ sách trong 2 năm gần nhất, và các vụ kiện, trọng tài trong 5 năm gần nhất. Hợp đồng còn phải ghi rõ trong một khoảng thời gian sau khi ký, bên nhượng quyền có thể tự một mình quyết định rút lui (đơn phương chấm dứt, 单方解除). Bên thương hiệu giấu không nói, hoặc đưa thông tin giả, thì bên nhượng quyền có thể chấm dứt hợp đồng (toàn quốc, có hiệu lực từ tháng 5 năm 2007) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2007). 商业特许经营管理条例（国务院令第 485 号，第七、八、十二、二十二、二十三、二十五条）. ; Bộ Thương mại (商务部) (2012). 商业特许经营信息披露管理办法（商务部令 2012 年第 2 号，第五、九条）. ; Bộ Thương mại (商务部). 商业特许经营信息管理系统. +- Ghi chú: Không tra được hồ sơ đăng ký, không chịu đưa hồ sơ công bố, đòi nộp tiền mới cho xem hợp đồng thì đi luôn.「Thời gian cân nhắc」là số ngày sau khi ký vẫn có thể tự một mình quyết định rút lui. Số ngày do hợp đồng quy định, nghị định chỉ nói「một khoảng thời gian nhất định」, nên khi ký hãy yêu cầu ghi rõ số ngày vào. Với những chỗ bên thương hiệu công bố như「lượng bán trung bình, lợi nhuận gộp và lợi nhuận thuần của các cửa hàng nhượng quyền hiện có」, phải hỏi con số lấy từ đâu, quy chế bắt bên thương hiệu nêu nguồn số liệu. + +### 6. Trước khi đăng ký, chốt tên, địa điểm kinh doanh, phạm vi kinh doanh và vốn đăng ký, đủ hồ sơ thì lấy được giấy phép ngay tại chỗ + +- Chi phí: Bước đăng ký không tốn tiền. Phí đăng ký doanh nghiệp và hộ kinh doanh cá thể đã ngừng thu từ ngày 1 tháng 1 năm 2015. Tự làm trên cổng dịch vụ công của địa phương, hoặc đến trung tâm một cửa, mỗi lần vài tiếng. Khắc con dấu, thuê người làm kế toán thay, thuê một địa chỉ đăng ký (quản lý địa chỉ) là mua dịch vụ ngoài thị trường, không phải khoản thu của nhà nước. +- Nói thẳng: Bản thân việc đăng ký không tốn tiền, phí đăng ký đã ngừng thu từ năm 2015. Hồ sơ phải nộp chỉ có năm loại: đơn xin, giấy tờ chứng minh thân phận hoặc tư cách, giấy tờ về địa điểm kinh doanh, điều lệ hoặc hợp đồng thành lập (hộ cá thể không phải nộp), cộng với những gì quy định pháp luật nêu thêm. Đủ hồ sơ thì cấp giấy phép ngay tại chỗ, không cấp được ngay thì xong trong 3 ngày làm việc. Khoản tiền bị thu phần lớn là phí chạy việc của bên làm thay. +- Lợi ích: Quy định ghi rõ hồ sơ phải nộp là năm loại.「(một) đơn xin」.「(hai) giấy tờ chứng minh tư cách của người xin, giấy tờ chứng minh thân phận của cá nhân」.「(ba) giấy tờ liên quan đến nơi ở hoặc địa điểm kinh doanh chính」.「(bốn) điều lệ của công ty, pháp nhân không phải công ty, hợp tác xã nông nghiệp của nông dân (liên kết), hoặc hợp đồng thành lập của doanh nghiệp hợp tác」.「(năm) các tài liệu khác mà pháp luật, quy định hành chính và cơ quan quản lý thị trường của Quốc vụ viện quy định phải nộp」. Nói thẳng ra: một đơn xin; một giấy chứng minh bạn là ai hoặc có tư cách làm việc này; một giấy chứng minh cửa hàng hay văn phòng là của bạn dùng; điều lệ hoặc hợp đồng thành lập; cộng với những gì quy định pháp luật nêu thêm. Công ty còn phải nộp giấy tờ bổ nhiệm và giấy tờ chứng minh thân phận của người đại diện theo pháp luật, giám đốc, giám sát và cán bộ quản lý cấp cao. Hộ cá thể không phải nộp điều lệ. Tên「do người xin tự khai theo luật」, tự nghĩ, tự khai. Vốn đăng ký「áp dụng chế độ đăng ký theo cam kết」, trước hết ghi số tiền cam kết góp, không phải nộp ngay tại chỗ. Phạm vi kinh doanh chọn trong danh mục cơ quan đăng ký công khai. Hồ sơ đầy đủ, đúng thể thức thì「được xác nhận và đăng ký ngay tại chỗ」. Không xong được ngay thì xong trong「3 ngày làm việc」, vụ phức tạp thì thêm 3 ngày làm việc.「Ngày ký giấy phép là ngày thành lập của chủ thể kinh doanh」: ngày in trên giấy phép chính là ngày công ty hay cửa hàng được thành lập (toàn quốc, từ tháng 3 năm 2022) +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2021). Quy định về đăng ký chủ thể kinh doanh（市场主体登记管理条例）（国务院令第 746 号，第十、十三、十四、十六、十九、二十一条）. ; Cục Quản lý thị trường quốc gia (市场监管总局) (2022). Quy tắc chi tiết về thi hành Quy định đăng ký chủ thể kinh doanh（市场监管总局令第 52 号，第十二、二十五、二十六条）. ; Bộ Tài chính, Ủy ban Cải cách và Phát triển quốc gia (财政部、国家发展改革委) (2014). Thông báo về việc bãi bỏ, dừng thu và miễn thu một loạt khoản thu dịch vụ hành chính công（财税〔2014〕101 号，附件 1）. +- Ghi chú: Bản thân việc đăng ký không thu phí, khoản bị thu phần lớn là phí chạy việc của bên làm thay. Giá khắc con dấu do thị trường quyết định. Văn phòng Quốc vụ viện ra lệnh「严禁指定公章制作单位制作公章」, nên có người nói「bắt buộc phải đến chỗ được chỉ định để khắc con dấu」thì bạn không cần phải nghe. Vốn đăng ký điền bao nhiêu không ảnh hưởng tới mặt tiền, chỉ quyết định giới hạn trách nhiệm của bạn, và phải nộp đủ trong 5 năm, xem mục 3 của phần này (chọn đúng loại chủ thể trước khi khai trương). Phạm vi kinh doanh có mục cần giấy phép thì chưa có giấy thì không được khai trương, xem mục 7 của phần này (ngành cần giấy phép). + +### 7. Ngành cần giấy phép thì chưa có giấy thì không khai trương + +- Chi phí: Không tốn tiền. Trước khi khai trương, tra rõ ngành bạn định làm có phải xin giấy phép không. Làm ăn uống, bán thực phẩm thì phải xin giấy phép kinh doanh thực phẩm. Chỉ bán thực phẩm đóng gói sẵn do người khác làm thì không cần giấy phép, chuyển sang kê khai. Có giấy rồi hãy mở cửa kinh doanh. +- Nói thẳng: Không có giấy mà khai trương thì nhẹ là bị buộc dừng kinh doanh, thu giữ toàn bộ số tiền đã thu, kèm phạt dưới 1 vạn tệ. Nếu bán hàng độc quyền, độc bán, hoặc thuộc các loại chứng khoán, hợp đồng tương lai, bảo hiểm, thanh toán và quyết toán vốn, tình tiết nặng thì cấu thành tội kinh doanh phi pháp, phạt tù từ 5 năm trở xuống. Trang trí xong mới xin giấy, trước khi giấy về đã mở bán thử, là chỗ vấp nhiều nhất. +- Lợi ích: Kinh doanh mà mở cửa khi chưa có giấy phép thì do bộ phận mà pháp luật chỉ định xử lý. Kinh doanh không có giấy phép mà pháp luật không ghi rõ phạt thế nào thì bị buộc dừng lại, thu giữ số tiền đã thu (thu nhập bất hợp pháp), rồi phạt dưới 1 vạn tệ. Còn một mức nặng hơn. Kinh doanh hàng hoặc dịch vụ pháp luật quy định phải độc quyền, độc bán hoặc hạn chế buôn bán mà chưa được phép. Làm những việc chứng khoán, hợp đồng tương lai, bảo hiểm, thanh toán và quyết toán vốn mà chưa được phê duyệt. Hai trường hợp này, tình tiết nặng thì cấu thành tội kinh doanh phi pháp. Phạt tù từ 5 năm trở xuống hoặc phạt tù ngắn hạn (giam giữ ngắn), kèm phạt tiền bằng 1 đến 5 lần số tiền đã thu. Tình tiết đặc biệt nặng thì phạt tù từ 5 năm trở lên (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2017). Biện pháp xử lý kinh doanh không phép không giấy phép（无证无照经营查处办法）（国务院令第 684 号，第五、六、十三条）. ；Cục Quản lý thị trường quốc gia (市场监管总局) (2023). Biện pháp quản lý giấy phép và kê khai kinh doanh thực phẩm（市场监管总局令第 78 号，第四条）. ；Đại biểu Quốc hội (全国人大) (1997). Bộ luật Hình sự（第二百二十五条）. +- Ghi chú: Điều kiện cấp giấy phép trong các lĩnh vực tài chính, y tế, giáo dục đào tạo có luật riêng, mục này chỉ đối chiếu quy tắc xử lý chung và điều khoản của Bộ luật Hình sự. Cửa hàng trà sữa, quán ăn nhỏ thuộc dịch vụ ăn uống, phải xin giấy phép kinh doanh thực phẩm. Trang trí xong mới xin giấy, trước khi giấy về đã mở bán thử, là chỗ vấp thường gặp. Ngành thực phẩm chia làm mấy bậc, bán thịt rau tươi có phải xin giấy không, xem mục 8 (làm thực phẩm thì xem trước mình ở bậc nào). + +### 8. Làm thực phẩm thì xem trước mình ở bậc nào: sản xuất và ăn uống phải có giấy phép, chỉ bán đóng gói sẵn thì kê khai, bán thịt rau tươi không cần giấy + +- Chi phí: Bản thân giấy phép không thu tiền. Nhưng cơ sở và thiết bị phải đạt, kiểm tra thực tế cộng với thủ tục xin giấy mất mấy tuần. Bán thịt thì còn phải giữ giấy kiểm dịch và giấy xác nhận chất lượng thịt cho từng lô. +- Nói thẳng: Tự làm đồ rồi đem bán, như quán ăn nhỏ, đồ ủ muối, bánh nướng, rau củ muối tự làm, đều phải xin giấy phép. Chỉ bán thực phẩm đóng gói sẵn do người khác làm thì không cần giấy phép, chỉ cần kê khai với cơ quan quản lý thị trường từ cấp huyện trở lên. Bán thịt tươi, rau củ thì không cần giấy. Không có giấy mà làm thì mức phạt thấp nhất cũng từ 5 vạn đến 10 vạn tệ, còn cao hơn vốn của đa số nghề nhỏ. +- Lợi ích: Luật An toàn thực phẩm: làm ăn thực phẩm phải có giấy phép trước. Tự sản xuất thực phẩm, bán thực phẩm, làm ăn uống, đều phải có giấy phép theo luật.「Nhưng bán nông sản thực phẩm và chỉ bán thực phẩm đóng gói sẵn thì không cần giấy phép. Chỉ bán thực phẩm đóng gói sẵn thì phải kê khai với cơ quan quản lý, giám sát an toàn thực phẩm của chính quyền nhân dân địa phương từ cấp huyện trở lên」. Nông sản thực phẩm là thịt tươi, rau củ, loại chưa qua chế biến. Thực phẩm đóng gói sẵn là hàng người khác làm xong, bán kèm cả bao bì.「Quy tắc cụ thể về quản lý xưởng sản xuất, chế biến thực phẩm nhỏ và rong quán thực phẩm do tỉnh, khu tự trị, thành phố trực thuộc trung ương quy định」, hai loại này do từng tỉnh tự đặt quy tắc. Làm ăn thực phẩm mà chưa có giấy phép thì tịch thu toàn bộ số tiền đã thu, thực phẩm, công cụ, thiết bị, nguyên liệu. Lô hàng này tính theo giá bán mà tổng số tiền (giá trị lô hàng) dưới 1 vạn tệ thì phạt thêm từ 5 vạn đến 10 vạn tệ. Giá trị lô hàng từ 1 vạn tệ trở lên thì phạt từ 10 lần đến 20 lần giá trị lô hàng. Quy định về giết lợn: lợn chỉ được giết ở nhà máy giết lợn do nhà nước chỉ định, kiểm dịch tập trung.「Ngoài nông thôn, cá nhân tự giết ăn tại nhà thì không áp dụng giết lợn tại điểm cố định, ngoài ra mọi tổ chức và cá nhân không được giết lợn nếu chưa qua điểm giết cố định」.「Các tổ chức, cá nhân kinh doanh sản phẩm lợn, sản xuất chế biến thực phẩm thịt, cũng như đơn vị kinh doanh ăn uống và đơn vị ăn uống tập thể, khi sản xuất kinh doanh sản phẩm lợn đều phải dùng sản phẩm lợn từ nhà máy giết lợn tại điểm cố định đã qua kiểm dịch và kiểm nghiệm chất lượng thịt」. Nên thịt lợn bạn bán và dùng đều phải từ nhà máy giết được chỉ định, phải qua cả kiểm dịch lẫn kiểm nghiệm chất lượng thịt. Tự giết lợn ở nơi không phải nhà máy được chỉ định thì bị buộc đóng cửa, tịch thu lợn, công cụ thiết bị giết mổ và số tiền đã thu, mức phạt y như bậc không có giấy mà làm thực phẩm ở trên. Luật Phòng chống dịch bệnh động vật: sản phẩm động vật bán và vận chuyển phải mang theo giấy kiểm dịch và dấu kiểm dịch. Không mang thì bị buộc sửa, rồi phạt không quá một lần giá trị lô hàng tương đương hàng đạt chuẩn cùng loại. Giải thích tư pháp: thực phẩm do xưởng nhỏ và rong quán làm, bán không đúng tiêu chuẩn an toàn thực phẩm, người mua yêu cầu bồi thường gấp mười (bồi thường mang tính răn đe, tức ngoài trả lại tiền còn phải bồi thêm khoản đó), tòa án phải hỗ trợ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2025, sửa đổi lần thứ ba). Luật An toàn thực phẩm（食品安全法）（第三十五、三十六、一百二十二条）. ；Quốc vụ viện (国务院) (2021). Quy định về quản lý giết lợn（生猪屠宰管理条例）（国务院令第 742 号，第二、二十三、三十一条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2021, sửa đổi). Luật Phòng chống dịch bệnh động vật（动物防疫法）（第五十一、一百条）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). Giải thích về một số vấn đề pháp lý khi xét các vụ án tranh chấp về bồi thường mang tính răn đe trong lĩnh vực dược phẩm và thực phẩm（法释〔2024〕9 号，第四条）. +- Ghi chú: Ngưỡng của xưởng nhỏ, nhà hàng nhỏ, rong quán thực phẩm do từng tỉnh đặt, mục này không ghi số cụ thể. Vào trang Cục Quản lý thị trường tỉnh bạn, tìm「食品小作坊小餐饮小摊贩管理条例」, xem bản của địa phương. Bán thực phẩm tự làm trên mạng không phải vì「trên mạng」mà thiếu một giấy phép, chỗ nào cần giấy phép thì vẫn phải có. Ngoài giấy phép ra còn có quy tắc chung của phần này: chủ thể và giấy phép xem ở mục 6 của phần này (chốt tên, địa điểm, phạm vi kinh doanh trước khi đăng ký), ngành khác có cần giấy phép không xem ở mục 7 của phần này (ngành cần giấy phép). Người mua có thể đòi bồi thường bao nhiêu xem ở mục 31 ở phần 5 (bồi thường gấp mười cho thực phẩm). + +### 9. Đóng túi để bán thì là thực phẩm đóng gói sẵn: trên nhãn, ngày sản xuất, hạn sử dụng, bảng thành phần thiếu mục nào cũng không được + +- Chi phí: In nhãn mỗi nhãn vài xu đến vài chục xu. Mỗi lô đều phải điền đúng ngày và thành phần, tốn một chút thời gian. +- Nói thẳng: Thực phẩm đóng vào bao bì rồi bán thì theo pháp luật gọi là thực phẩm đóng gói sẵn, trên bao bì phải có nhãn, và có chín mục bắt buộc phải ghi, trong đó có cả ngày sản xuất và hạn sử dụng. Thực phẩm bán lẻ cũng phải ghi tên, ngày, tên và địa chỉ người bán. Mục bắt buộc mà không ghi thì lô hàng dưới 1 vạn tệ phạt 5000 đến 5 vạn tệ, đủ 1 vạn thì phạt gấp 5 đến 10 lần giá trị lô hàng. +- Lợi ích: Luật An toàn thực phẩm: thực phẩm đóng gói sẵn (thực phẩm đóng vào bao bì rồi bán) phải có nhãn trên bao bì. Nhãn phải ghi tên, quy cách, khối lượng tịnh, ngày sản xuất, thành phần hoặc bảng thành phần, tên, địa chỉ, cách liên hệ của nhà sản xuất, hạn sử dụng. Còn có mã tiêu chuẩn sản phẩm, điều kiện bảo quản, tên chung trong tiêu chuẩn quốc gia của phụ gia thực phẩm dùng trong sản phẩm, số giấy phép sản xuất. Pháp luật hoặc tiêu chuẩn an toàn thực phẩm có yêu cầu ghi thêm thì cũng phải ghi. Thực phẩm bán lẻ là thực phẩm không đóng túi, cân theo nhu cầu. Bán thực phẩm bán lẻ thì trên dụng cụ chứa và bao bì ngoài cũng phải ghi: tên thực phẩm, ngày sản xuất hoặc lô sản xuất, hạn sử dụng, tên, địa chỉ, cách liên hệ của nhà sản xuất, kinh doanh. Thực phẩm đóng gói sẵn do mình làm, bán mà không có nhãn, hoặc nhãn, hướng dẫn không đúng quy định của luật này thì tịch thu số tiền đã thu và cả thực phẩm. Lô hàng này tính theo giá bán mà tổng số tiền (giá trị lô hàng) dưới một vạn tệ thì phạt thêm từ 5000 tệ đến 5 vạn tệ. Giá trị lô hàng từ một vạn tệ trở lên thì phạt từ 5 lần đến 10 lần giá trị lô hàng. Tình tiết nặng thì bị buộc dừng sản xuất, dừng kinh doanh cho đến khi thu hồi giấy phép. Nhãn, hướng dẫn chỉ có lỗi nhỏ (khiếm khuyết), không ảnh hưởng an toàn thực phẩm cũng không khiến người tiêu dùng hiểu nhầm thì trước hết buộc sửa, không sửa mới phạt dưới 2000 tệ. Bản《Quy tắc chung nhãn thực phẩm đóng gói sẵn》mới mang số GB 7718-2025, ban hành ngày 16 tháng 3 năm 2025, áp dụng từ ngày 16 tháng 3 năm 2027, giữa hai mốc có hai năm chuyển tiếp. Nó yêu cầu bố trí khu vực riêng trên bao bì, đồng thời trình bày ngày sản xuất và ngày hết hạn sử dụng. Ngày phải dùng chữ và màu có độ tương phản cao, dễ đọc, và xếp theo thứ tự năm-tháng-ngày (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2025, sửa đổi lần thứ ba). Luật An toàn thực phẩm（食品安全法）（第六十七、六十八、一百二十五条）. ；Ủy ban Quốc gia Y tế, Cục Quản lý thị trường quốc gia (国家卫生健康委员会、国家市场监督管理总局) (2025). Thông báo công bố《Quy tắc chung nhãn thực phẩm đóng gói sẵn》(GB 7718-2025) cùng 50 tiêu chuẩn an toàn thực phẩm và 9 văn bản sửa đổi（2025 年第 2 号）. （Ủy ban Y tế thành phố Lạc Sa trích lại toàn văn thông báo）；Tòa án nhân dân tối cao (最高人民法院) (2024). Giải thích về một số vấn đề pháp lý khi xét các vụ án tranh chấp về bồi thường mang tính răn đe trong lĩnh vực dược phẩm và thực phẩm（法释〔2024〕9 号，第六、八条）. （cách phân bậc giữa khiếm khuyết nhãn và thiếu mục bắt buộc）；Cục Quản lý thị trường thành phố Bắc Kinh (北京市市场监督管理局) (2026). Không tìm thấy ngày sản xuất? Nhãn thực phẩm sắp「thay mặt」! Những điểm chính trong quy định mới này. （ngày áp dụng, thời gian chuyển tiếp và yêu cầu mới về trình bày ngày） +- Ghi chú: Lỗi nhãn chia làm hai loại, phạt bao nhiêu, bồi thường bao nhiêu đều theo cách chia này. Một là lỗi nhỏ, như chính tả, cỡ chữ không quy chuẩn, không ảnh hưởng an toàn thực phẩm cũng không khiến người ta hiểu nhầm. Loại này cơ quan quản lý thị trường chỉ yêu cầu sửa, không sửa mới phạt dưới 2000 tệ, người mua cũng không đòi được bồi thường gấp mười. Loại kia là thiếu mục bắt buộc, như bỏ ngày sản xuất, hạn sử dụng. Giải thích tư pháp nói rõ đây không tính là lỗi nhỏ: phạt tiền tính theo giá trị lô hàng, dưới 1 vạn tệ thì phạt 5000 tệ đến 5 vạn tệ, từ 1 vạn tệ trở lên thì phạt gấp 5 đến 10 lần giá trị lô hàng. Người mua còn có thể đòi bạn bồi thường gấp mười giá thực phẩm, dưới 1000 tệ thì tính bằng 1000 tệ (xem mục 31 ở phần 5, bồi thường gấp mười cho thực phẩm). Yêu cầu chi tiết từng điều của GB 7718-2025, mục này mới chỉ đối chiếu được ngày và cách trình bày trong thông báo và bài giải thích chính thức, thể thức cụ thể theo nguyên văn tiêu chuẩn. +### 10. Thực phẩm thường không được nói là chữa bệnh: nhãn, hướng dẫn, quảng cáo và lời livestream đều tính + +- Chi phí: Không tốn tiền. Rà lại câu quảng cáo, trang giới thiệu sản phẩm và lời livestream. Khó ở chỗ nhịn không nói những câu như「hạ đường huyết, bao tiêu」「chữa táo bón」. +- Nói thẳng: Nhãn, hướng dẫn và quảng cáo của thực phẩm thường đều không được nói là có thể phòng bệnh, chữa bệnh. Ngoài ba loại quảng cáo y tế, dược phẩm, thiết bị y tế ra, các quảng cáo khác không được nói là chữa bệnh, cũng không được dùng từ nghe như thuốc. Phạt bằng 1 đến 3 lần chi phí quảng cáo. Không tính được chi phí quảng cáo thì phạt từ 10 vạn đến 20 vạn tệ. Tình tiết nặng còn bị thu hồi giấy phép kinh doanh. +- Lợi ích: Luật An toàn thực phẩm:「Nhãn và hướng dẫn của thực phẩm, phụ gia thực phẩm không được chứa nội dung sai lệch, không được đề cập chức năng phòng bệnh, chữa bệnh. Nhà sản xuất kinh doanh chịu trách nhiệm về nội dung nhãn, hướng dẫn do mình cung cấp」.「Nội dung quảng cáo thực phẩm phải thật, hợp pháp, không chứa nội dung sai lệch, không đề cập chức năng phòng bệnh, chữa bệnh. Nhà sản xuất kinh doanh thực phẩm chịu trách nhiệm về tính thật, hợp pháp của nội dung quảng cáo thực phẩm」. Nghĩa là trên bao bì, hướng dẫn và quảng cáo đều không được nói là có thể phòng bệnh, chữa bệnh, viết gì thì người bán tự chịu trách nịch. Thổi phóng thực phẩm trong quảng cáo, lừa người tiêu dùng thì bị phạt theo Luật Quảng cáo. Luật Quảng cáo cũng có một điều cấm.「Ngoài quảng cáo y tế, dược phẩm, thiết bị y tế, cấm mọi quảng cáo khác đề cập chức năng chữa bệnh, và không được dùng từ ngữ y tế hoặc từ ngữ dễ khiến sản phẩm đang quảng cáo bị lẫn với thuốc, thiết bị y tế」. Ngoài ba loại quảng cáo y tế, dược phẩm, thiết bị y tệ, mọi quảng cáo khác đều không được nói là chữa bệnh, cũng không được dùng từ nghe như thuốc, như thiết bị y tế. Vi phạm điều này thì bị buộc dừng quảng cáo. Còn phải yêu cầu bên đã chạy quảng cáo (bên quảng cáo) công kích làm rõ, xoá bỏ ảnh hưởng trong phạm vi đã phát. Rồi phạt tiền từ 1 lần đến 3 lần chi phí quảng cáo. Không tính được chi phí quảng cáo, hoặc báo rõ ràng thấp hơn thực tế, thì phạt từ 10 vạn đến 20 vạn tệ. Tình tiết nặng thì phạt gấp từ 3 lần đến 5 lần chi phí quảng cáo, hoặc phạt từ 20 vạn đến 100 vạn tệ, còn có thể thu hồi giấy phép kinh doanh, không làm ăn được nữa (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2025, sửa đổi lần thứ ba). Luật An toàn thực phẩm（食品安全法）（第七十一、七十三、一百二十五、一百四十条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2021, sửa đổi). Luật Quảng cáo（广告法）（第十七、五十八条）. +- Ghi chú:「Chức năng bảo vệ sức khoẻ」cũng không phải thứ thực phẩm thường được nói. Chức năng bảo vệ sức kho�è mà thực phẩm chức năng được tuyên bố chỉ được lấy từ danh mục nhà nước quy định, sản phẩm còn phải qua đăng ký hoặc kê khai, thực phẩm thường không có bước này. Người mua nhận ra loại lời này thế nào, xem mục 10 ở phần 6 (đừng bỏ nhiều tiền mua thực phẩm bảo vệ sức khoẻ để điều chỉnh cơ thể). Ngoài bao bì và quảng cáo, trang giới thiệu sản phẩm, lời livestream, văn bản đăng trên mạng xã hội cũng là nội dung quảng cáo, đăng ra là để lại dấu vết. + +### 11. Ngành thực phẩm có đường ranh giới hình sự: bán thịt chết bệnh, hàng vượt tiêu chuẩn là đủ thành tội, trộn thêm nguyên liệu độc hại thì không tính tiền, khởi tội năm năm + +- Chi phí: Không tốn tiền. Làm theo bộ mục 8 của phần này (làm thực phẩm thì xem trước mình ở bậc nào): giữ giấy kiểm dịch và chứng từ nhập hàng, bảng thành phần chỉ đưa loại có tra được số hiệu phê duyệt. +- Nói thẳng: Mấy mục trước phạt tiền, mục này bắt người. Bán thịt chết bệnh, thịt kiểm nghiệm không đạt, hoặc hàng nông dư lưu, dư lượng độc tố nặng, chỉ cần「đủ sức」khiến người ta ăn vào bị ngộ độc nặng là đủ thành tội, không cần thật sự có người ăn hỏng, phạt tù từ 3 năm trở xuống. Trộn nguyên liệu không phải thực phẩm có độc, gây hại vào thực phẩm thì khởi tội 5 năm trở xuống, không xét bạn đã bán được bao nhiêu tiền. +- Lợi ích: Điều một trăm bốn mươi ba của Bộ luật Hình sự quy định mức này. Thực phẩm bạn làm, bán không đúng tiêu chuẩn an toàn thực phẩm. Chỉ cần「đủ sức gây ra tai nạn ngộ độc thực phẩm nghiêm trọng hoặc bệnh do thực phẩm nghiêm trọng khác」là đủ thành tội. Phạt tù từ 3 năm trở xuống hoặc phạt tù ngắn hạn (giam giữ ngắn), kèm phạt tiền. Gây tổn hại nghiêm trọng cho sức khoẻ con người, hoặc có tình tiết nghiêm trọng khác, thì phạt tù từ 3 năm đến 7 năm và phạt tiền. Hậu quả đặc biệt nghiêm trọng thì phạt tù từ 7 năm trở lên cho tới tù chung thân, phạt tiền hoặc tịch thu toàn bộ tài sản. Điều một trăm bốn mươi tư: trộn nguyên liệu không phải thực phẩm có độc, gây hại vào thực phẩm bạn đang làm, bán, hoặc biết đã trộn mà vẫn bán, thì phạt tù từ 5 năm trở xuống và phạt tiền. Gây tổn hại nghiêm trọng cho sức khoẻ con người, hoặc có tình tiết nghiêm trọng khác, thì phạt tù từ 5 năm đến 10 năm và phạt tiền. Gây chết người, hoặc có tình tiết đặc biệt nghiêm trọng khác, thì xử theo Điều một trăm bốn mươi mốt (từ 10 năm trở lên, tù chung thân hoặc tử hình). Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao đã ban hành một bản giải thích (giải thích của hai tòa). Bản nêu rõ những tình huống nào được coi là「đủ sức gây ra tai nạn ngộ độc thực phẩm nghiêm trọng hoặc bệnh do thực phẩm nghiêm trọng khác」. Một là「chứa chất gây ô nhiễm như vi sinh vật gây bệnh vượt quá giới hạn tiêu chuẩn nghiêm trọng, dư lượng thuốc trừ sâu, dư lượng thuốc thú y, độc tố sinh học, kim loại nặng」. Hai là「là thịt và sản phẩm từ gia cầm, thú, thuỷ sản chết bệnh, chết không rõ nguyên nhân hoặc không đạt kiểm dịch, kiểm nghiệm」. Ba là「thuộc loại nhà nước cấm sản xuất, bán để phòng chống bệnh dịch và các nhu cầu đặc biệt khác」. Bốn là「thực phẩm công thức dinh dưỡng cho mục đích y tế đặc biệt, thực phẩm chính, phụ dành cho trẻ nhỏ có thành phần dinh dưỡng không đúng tiêu chuẩn an toàn thực phẩm nghiêm trọng」. Dùng phụ gia liều quá, dùng sai chỗ, đến mức gây ra hậu quả nêu trên, thì cũng bị quy cho tội theo Điều một trăm bốn mươi ba. Cái gì là「nguyên liệu không phải thực phẩm có độc, gây hại」không phải do bạn quyết định. Một loại là chất mà pháp luật cấm cho thêm vào thực phẩm. Loại khác là chất trong danh sách công bố của các bộ ngành của Quốc vụ viện, như thuốc trừ sâu bị cấm, thuốc không được dùng trên gia súc nuôi ăn. Còn một tình huống nữa cũng tính là「tình tiết nghiêm trọng khác」ở Điều một trăm bốn mươi tư. Làm, bán trên 10 vạn tệ mà chưa đến 20 vạn tệ. Địa điểm là khuôn viên trường tiểu học, trung xác thu học, cơ sở chăm sóc người cao tuổi và xung quanh. Đối tượng bán là người chưa thành niên, người cao tuổi. Giết lợn tự lập ở nơi không phải nhà máy giết được chỉ định để buôn bán, tình tiết nặng thì xử theo tội kinh doanh phi pháp. Khi mức tiền trên 10 vạn tệ, hoặc thu được trên 5 vạn tệ, thì coi là「tình tiết nghiêm trọng」（toàn quốc） +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020, sửa đổi). Bộ luật Hình sự（第一百四十一、一百四十三、一百四十四条）. （Cơ sở dữ liệu luật và quy phạm quốc gia）；Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao (最高人民法院、最高人民检察院) (2021). Giải thích về một số vấn đề pháp lý khi xử lý các vụ án hình sự gây hại an toàn thực phẩm（法释〔2021〕24 号，第一、二、五、七、九、十七、十八条，2022 年 1 月 1 日起施行）. （Cục Quản lý thị trường tỉnh Hà Bắ trích lại toàn văn）、（Sở Tư pháp tỉnh Liao Ninh trích lại toàn văn, hai bản đối chiếu từng chữ khớp nhau） +- Ghi chú: Ranh giới với ba mục trước là có đụng tới an toàn thực phẩm hay không. Không xin giấy, nhãn thiếu mục, thổi tác dụng chữa bệnh thì bị phạt tiền. Bán thịt chết bệnh, hàng vượt tiêu chuẩn, thêm phụ gia bừa bãi, trộn nguyên liệu không phải thực phẩm thì đi vào thủ tục hình sự. Phòng thủ tiện nhất vẫn là bộ động tác ở mục 8 của phần này (làm thực phẩm thì xem trước mình ở bậc nào). Thịt chỉ nhập từ nhà máy giết cố định. Mỗi lô giữ giấy kiểm dịch và giấy xác nhận chất lượng thịt. Thành phần chỉ dùng loại tra được số hiệu phê duyệt. Thứ được gọi là「thịt xử lý giá rẻ」「phế tấm không biết từ đâu」thì hay xảy ra sự cố nhất. Lần này trang Công báo Tòa án nhân dân tối cao trả về lỗi 502, nên đã dẫn hai bản trích lại chính thức của tỉnh và đối chiếu từng chữ với nhau. + +### 12. Có giấy phép là bắt đầu có nghĩa vụ kê khai, không có doanh thu vẫn phải kê khai đúng hạn (nộp một tờ toàn số không) + +- Chi phí: Không tốn tiền. Tự kê khai theo tháng hoặc theo quý trên cổng thuế điện tử, mỗi lần vài phút. Thuê người làm kế toán thay là khoản chi khác ngoài thị trường. Đơn vị đã thuê người làm còn phải đăng ký bảo hiểm xã hội trong 30 ngày kể từ ngày thành lập. Khó ở chỗ những tháng không có việc cũng phải nhớ kê khai đúng hạn. +- Nói thẳng: Có giấy phép là phải kê khai thuế đúng hạn. Không kiếm được đồng nào vẫn phải kê khai, chỉ là các con số trên bảng điền toàn số không. Không kê khai thì trước hết phạt dưới 2000 tệ, tình tiết nặng phạt từ 2000 tệ đến 1 vạn tệ. Liên tục ba tháng không kê khai bất kỳ loại thuế nào, hệ thống sẽ tự động xếp bạn vào diện không bình thường, không phát được hoá đơn. +- Lợi ích: 「Người nộp thuế trong thời hạn nộp thuế không có số thuế phải nộp thì vẫn phải làm thủ tục kê khai thuế theo quy định」: kỳ này không phải nộp đồng thuế nào, bảng kê khai vẫn phải nộp. Không kê khai đúng hạn thì「cơ quan thuế buộc sửa trong thời hạn, có thể phạt dưới hai nghìn tệ; tình tiết nặng có thể phạt từ hai nghìn tệ đến một vạn tệ」. Không kê khai mà nộp thiếu thuế thì phải truy thu, cộng thêm tiền chậm nộp tính theo ngày, kèm phạt「từ năm mươi phần trăm đến năm lần số thuế nộp thiếu」. Tiền chậm nộp cộng theo tỷ lệ năm phần nghìn mỗi ngày. 「Liên tục ba tháng không kê khai thuế với bất kỳ loại thuế nào, hệ thống quản lý thuế sẽ tự động xác định là hộ không bình thường và dừng việc dùng sổ và hoá đơn」. Ba tháng không kê khai gì, hệ thống sẽ tự xếp bạn vào diện không bình thường, không lấy được và không phát được hoá đơn. Kê khai bù và chấp nhận xử phạt thì sẽ tự động gỡ bỏ. Đơn vị đã thuê người làm thì「phải trong vòng ba mươi ngày kể từ ngày thành lập……nộp đơn đăng ký bảo hiểm xã hội với cơ quan bảo hiểm xã hội địa phương」（toàn quốc） +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2002). Quy tắc chi tiết về thi hành Luật Quản lý thuế thu（税收征收管理法实施细则）（国务院令第 362 号，第三十二、七十五条）. ; Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2015). Luật Quản lý thuế thu（税收征收管理法）（2015 年修正，第二十五、三十二、六十二、六十四条）. ; Cục Tính thuế quốc gia (国家税务总局) (2019). Thông báo về một số việc trong quản lý thuế（国家税务总局公告 2019 年第 48 号，第三条）. ; Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2018). Luật Thuế thu nhập cá nhân（个人所得税法）（2018 年修正，第十二条）. ; Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2018). Luật Bảo hiểm xã hội（社会保险法）（2018 年修正，第五十七条）. +- Ghi chú: Sau「đăng ký một mã nhiều giấy」không phải đăng ký thuế riêng nữa. Lấy mã tín dụng xã hội thống nhất trên giấy phép để đăng nhập cổng thuế điện tử, xác nhận thông tin đăng ký, rồi kê khai những loại thuế mình phải nộp. Hộ kinh doanh cá thể và doanh nghiệp tư nhân nộp thuế thu nhập cá nhân theo thu nhập kinh doanh. Trước hết nộp một khoản「trong vòng mười lăm ngày sau khi kết thúc tháng hoặc quý」, rồi「làm quyết toán trước ngày ba mươi mốt tháng ba năm sau năm thu được tiền」(tính cả năm, trả nhiều hoàn ít). Không khai trương vẫn phải kê khai, kê khai toàn số không cũng là kê khai. Để yên ba tháng là thành hộ không bình thường, hoá đơn bị dừng, muốn đăng ký huỷ cũng không đi được thủ tục rút gọn, xem mục 23 của phần này (lỗ thì rút lui theo thủ tục). + +### 13. Hoá đơn chỉ xuất theo giao dịch thật, người nộp thuế quy mô nhỏ hãy dùng hết ngưỡng miễn thuế + +- Chi phí: Không tốn tiền. Không mua hoá đơn, không bán hoá đơn, cũng không giúp người khác「lập sổ qua hoá đơn」. Xuất hoá đơn và kê khai thuế theo việc kinh doanh thật sự xảy ra. +- Nói thẳng: Nếu bạn là người nộp thuế quy mô nhỏ nộp thuế theo phương pháp giản lược, một tháng bán chưa đến 10 vạn, một quý bán chưa đến 30 vạn thì thuế giá trị gia tăng được miễn. Vốn phải nộp 3% thì cùng kỳ cũng giảm còn nộp 1%. Ngược lại,「tìm người mua hoá đơn để bù chi phí」chính là xuất hoá đơn khống, khởi tội từ 3 năm trở xuống, còn phạt từ 2 vạn đến 20 vạn tệ. +- Lợi ích: Bậc quy mô nhỏ, nộp thuế theo phương pháp giản lược, tên gọi là người nộp thuế quy mô nhỏ. Pháp luật viết là「chưa đạt tới ngưỡng khởi điểm thì miễn thuế giá trị gia tăng」. Đường này gọi là ngưỡng khởi điểm. Từ ngày 1 tháng 1 năm 2026 đến ngày 31 tháng 12 năm 2027, ngưỡng là 10 vạn tệ tính theo tháng, 30 vạn tệ tính theo quý, 1000 tệ tính theo lần hoặc theo ngày, bán không đạt đường này thì không phải nộp. Trong cùng khoảng thời gian đó, phần vốn nộp 3% giảm còn nộp 1%, trừ cho bán, cho thuê nhà và chuyển nhượng quyền sử dụng đất. Nói ngược lại, xuất hoá đơn giá trị gia tăng chuyên dụng không có giao dịch mua bán thật, hoặc loại hoá đơn khác có thể đem khấu trừ thuế, thì phạt tù từ 3 năm trở xuống hoặc phạt tù ngắn hạn, kèm phạt từ 2 vạn đến 20 vạn tệ. Xuất cho người khác, tự xuất cho mình, nhờ người khác xuất cho mình, giới thiệu người khác xuất, đều là xuất khống. Liên quan số thuế lớn thì phạt tù từ 3 năm đến 10 năm. Số tiền khổng lồ thì phạt tù từ 10 năm trở lên hoặc tù chung thân (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2024). Luật Thuế giá trị gia tăng（增值税法）（2026 年 1 月 1 日施行，第二十三条）. ；Bộ Tài chính, Cục Tính thuế (财政部、税务总局) (2026). Thông báo về việc chuyển tiếp các chính sách ưu đãi thuế giá trị gia tăng sau khi Luật Thuế giá trị gia tăng có hiệu lực（财政部 税务总局公告 2026 年第 10 号，第一、二、六条）. ；Đại biểu Quốc hội (全国人大) (1997). Bộ luật Hình sự（第二百零五条）. +- Ghi chú: Luật Thuế giá trị gia tăng có hiệu lực từ ngày 1 tháng 1 năm 2026. Hai thông báo giảm thuế cho người nộp thuế quy mô nhỏ trước đây, số 2023 年第 19 号 và số 2023 年第 1 号, theo đó ngừng thi hành hoặc bãi bỏ. Ngưỡng số tiền không đổi, nhưng căn cứ đã thay, mục này đã cập nhật theo văn bản mới. Trang nguyên văn Bộ luật Hình sự là bản năm 1997, Điều 205 sau này đã được sửa đổi về các bậc phạt tù, mục này chỉ ghi những bậc đã đối chiếu và không bị sửa. Miễn là thuế giá trị gia tăng, thuế thu nhập tính riêng. 「Tìm người mua hoá đơn để bù chi phí」là xuất hoá đơn khống, không phải tiết kiệm thuế. +### 14. Ai tự xưng cơ quan thuế đòi bạn nộp tiền khai thuế đều là lừa đảo, làm thủ tục thuế chỉ đi qua cổng thuế điện tử và 12366 + +- Chi phí: Không tốn tiền. Nhận cuộc gọi, tin nhắn như「税务抽查」「账户异常」「不报税就罚款」, hoặc có người kết bạn thì cắt máy, không trả lời. Rồi gọi 12366, hoặc gọi số cố định của cơ quan thuế quản lý để kiểm chứng. +- Nói thẳng: Cơ quan thuế không gọi điện xin mật khẩu thẻ ngân hàng, mật khẩu thanh toán hay mã xác minh tin nhắn, cũng không đưa một tài khoản để bạn chuyển thuế vào. Người đến kiểm tra sổ sách thật thì phải xuất trình thông báo kiểm tra và giấy chứng nhận kiểm tra. Nhận cuộc gọi kiểu này thì cắt máy trước, rồi gọi 12366 để kiểm chứng. Đã chuyển tiền rồi thì gọi ngay 110 hoặc 96110 để yêu cầu dừng chi trả. +- Lợi ích: Nguyên văn khuyến cáo của Cục Thuế nhà nước: 「税务部门不会通过微信小程序、电话、短信等方式索要银行卡密码、支付密码、账户余额及与银行卡收付款项相关的短信验证码」— cơ quan thuế không đòi mật khẩu thẻ ngân hàng, mật khẩu thanh toán, số dư tài khoản hay mã xác minh tin nhắn liên quan đến giao dịch thu chi qua thẻ ngân hàng bằng tiểu ứng WeChat, điện thoại, tin nhắn. 「一些不法分子冒充税务干部以提供『帮办纳税申报』『税收优惠代办』『政策辅导』等为名接触纳税人」— có kẻ giả làm cán bộ thuế, tiếp cận người nộp thuế với danh nghĩa「thay bạn khai thuế」「làm thủ tục ưu đãi thuế」「hướng dẫn chính sách」. Muốn kiểm chứng thì gọi「12366 纳税缴费服务热线或主管税务机关固定电话」— đường dây nóng 12366 hoặc số cố định của cơ quan thuế quản lý. Người đến kiểm tra sổ sách thật (kiểm tra hình sự thuế) thì「必须出示检查通知书和检查证」— bắt buộc xuất trình thông báo kiểm tra và giấy chứng nhận kiểm tra. 「税款只能通过电子税务局或者在办税厅缴纳，税务部门不会提供指定账号」— tiền thuế chỉ nộp qua cổng thuế điện tử hoặc tại sảnh nộp thuế, cơ quan thuế không cung cấp tài khoản chỉ định. Cơ quan trung gian làm thay việc thuế cũng có một điều. 「以税务机关和税务人员的名义敲诈纳税人、扣缴义务人的」, phạt tiền tùy mức độ. Giả danh cơ quan thuế và cán bộ thuế để tống tiền người nộp thuế và đơn vị khấu trừ, nộp thay là tình huống này. Giả danh công chức nhà nước để khoe khoang lừa đảo thì bị giam từ 10 ngày đến 15 ngày, còn có thể phạt thêm tối đa 1000 tệ. Vừa đủ cấu thành tội lừa đảo vừa đủ cấu thành tội khoe khoang lừa đảo thì「依照处罚较重的规定定罪处罚」(điều nào phạt nặng hơn thì xử theo điều đó) (toàn quốc). +- Mức bằng chứng: A +- Nguồn: Cục Thuế nhà nước (国家税务总局) (2025). 税务部门提醒广大纳税人——增强防范意识 警惕互联网涉税诈骗. ; Sở Thuế thành phố Dương Châu (国家税务总局扬州市税务局) (2023). 扬州防范涉税诈骗宣传篇. ; Cục Thuế nhà nước (国家税务总局) (2025). 涉税专业服务管理办法（试行）（国家税务总局令第 58 号，第三十二条第七项、第三十三条）. ; Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，2026 年 1 月 1 日施行，第六十二条）. ; Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部) (2016). 关于办理电信网络诈骗等刑事案件适用法律若干问题的意见（法发〔2016〕32 号）. +- Ghi chú: Kiểu lừa này nhắm vào hộ kinh doanh cá thể và công ty nhỏ vừa đăng ký. Lời dẫn thường là「系统显示你没报税，今天不处理就罚款」, rồi dẫn bạn sang tài khoản chính thức hay tiểu ứng của chúng, thu một lần vài trăm tệ. Trong khi bảu kê đó vốn miễn phí, tự làm chỉ mất vài phút. Thuê tổ chức ngoài thị trường làm sổ sách kế toán thay không vi phạm pháp luật, nhưng đó không phải cơ quan nhà nước, cũng không dùng danh nghĩa cơ quan thuế để đòi bạn. Đã trả tiền hoặc chuyển khoản rồi thì làm theo mục 2 ở phần 8 (dừng chi trả khi bị lừa), gọi ngay 110 hoặc 96110 để yêu cầu dừng chi trả. + +### 15. Nhận tiền thì viết「定金」chứ đừng viết「订金」, tiền phạt vi phạm phải ghi rõ con số, hạn thanh toán thì soi như cho vay + +- Chi phí: Không tốn tiền. Trong hợp đồng viết đúng hai chữ「定金」, số tiền không quá 20% tổng giá trị hợp đồng. Nhận tiền rồi mới giao hàng. Trước khi đồng ý cho bên kia lấy hàng trước, trả tiền sau (cho bán hàng có hạn nợ), hãy tra đối tác một lượt. +- Nói thẳng: Chỉ khi viết đúng hai chữ「定金」thì mới có quy tắc trả gấp đôi: bạn vi phạm thì số tiền đó không lấy lại được; bên kia vi phạm thì phải trả lại gấp đôi. Phần vượt quá 20% giá trị hợp đồng thì không tính. Viết thành「订金」thì pháp luật không có bộ quy tắc đó. Cho bán hàng có hạn nợ nghĩa là cho vay tiền không lãi; bên kia phá sản thì thành nợ xấu. +- Lợi ích: Tiền đặt chỉ có hiệu lực khi tiền thật sự được giao cho bên kia (thành lập từ lúc thực tế giao). Bên đặt tiền vi phạm thì không lấy lại được số tiền đó. Bên nhận tiền đặt vi phạm thì phải trả lại nguyên số cũ, rồi bồi thêm một phần bằng nhau (trả lại gấp đôi). Phần vượt 20% tổng giá trị hợp đồng (giá trị đối tượng của hợp đồng chính) không tính là tiền đặt. Tiền phạt vi phạm thỏa thuận cao hơn hoặc thấp hơn thiệt hại thực tế quá nhiều thì tòa án có thể theo yêu cầu của một bên mà hạ xuống hoặc nâng lên. Hợp đồng ghi cả tiền phạt vi phạm lẫn tiền đặt, thật sự xảy ra vi phạm thì chỉ chọn một, không thể đòi cả hai (toàn quốc, từ năm 2021). +- Mức bằng chứng: A +- Nguồn: Quốc hội Nhân dân (全国人大) (2020). 民法典（第五百八十五、五百八十六、五百八十七、五百八十八条）. +- Ghi chú:「订金」trong Bộ luật Dân sự không có bộ quy tắc như tiền đặt, thường chỉ được coi là khoản trả trước. Đây là cách nói lưu hành trong thực tiễn, phần này chưa kiểm chứng án lệ. Hạn thanh toán là cho vay tiền của mình không lãi cho bên kia, bên kia phá sản thì thành nợ xấu. Trước khi cho bán hàng có hạn nợ, vào Hệ thống công bố thông tin tín dụng doanh nghiệp quốc gia tra hồ sơ của bên kia, xem đã từng bị khởi kiện hay chưa. + +### 16. Người lao động tháng đầu phải ký hợp đồng bằng văn bản, 30 ngày đăng ký bảo hiểm xã hội + +- Chi phí: Phần do công ty đóng trong bảo hiểm xã hội thì phải nộp đều, tỷ lệ khác nhau theo địa phương. Bản thân việc ký hợp đồng không tốn tiền, dùng thẳng mẫu hợp đồng của cơ quan nhân sự và xã hội địa phương là được. +- Nói thẳng: Người đã đi làm đủ một tháng mà chưa ký hợp đồng bằng văn bản thì từ tháng thứ hai, mỗi tháng phải trả lương gấp đôi, tính liền đến đủ một năm. Bảo hiểm xã hội phải đăng ký trong 30 ngày kể từ ngày bắt đầu dùng lao động, không đăng ký thì phạt bằng 1 đến 3 lần số phí bảo hiểm xã hội phải đóng. Pháp luật không có mục「员工自愿放弃社保」. +- Lợi ích: Người đã đi làm đủ một tháng, chưa đủ một năm mà chưa ký hợp đồng bằng văn bản thì mỗi tháng phải trả lương gấp đôi. Kể từ ngày đầu tiên người lao động đi làm, phải đăng ký bảo hiểm xã hội cho họ trong 30 ngày. Không đăng ký thì trước hết bị buộc sửa. Quá hạn thì phạt bằng 1 đến 3 lần số phí bảo hiểm xã hội phải đóng, người trực tiếp làm việc đó bị phạt thêm từ 500 tệ đến 3000 tệ. Bảo hiểm xã hội không nộp đúng hạn, không nộp đủ số thì tính tiền chậm nộp (滞纳金) mỗi ngày năm phần mười nghìn kể từ ngày bị nợ. Quá hạn vẫn không nộp thì phạt tiếp bằng 1 đến 3 lần số nợ. Đơn vị phải tự khai, nộp đủ tiền đúng hạn. Chỉ sự kiện bất khả kháng như thiên tai mới cho phép nộp muộn hoặc nộp thiếu. Không có mục「员工自愿放弃」(toàn quốc). +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007). 劳动合同法（第十、八十二条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2010). 社会保险法（2018 年修正，第五十八、六十、八十四、八十六条）. +- Ghi chú: Lương gấp đôi tính từ tháng thứ hai cho tới đủ một năm, kéo dài bao lâu thì tốn bấy nhiêu. Nhân viên chưa ký hợp đồng, chưa đóng bảo hiểm xã hội, một khi xảy ra tai nạn lao động, hoặc sau khi nghị việc đi qua hòa giải lao động, thì khoản nộp bù cộng tiền phạt cộng tiền bồi thường, thường nhiều hơn hẳn số tiền đã tiết kiệm được. + +### 17. Lương trả theo hợp đồng và theo pháp luật, tăng lương, cho vay, thưởng đều phải qua văn bản + +- Chi phí: Không tốn tiền. Viết rõ mức lương trong hợp đồng. Tăng lương, thưởng, ứng trước, cho vay, đều ký giấy để lại. Không hứa miệng. Cũng không tự bỏ tiền túi ra trả lương thay công ty. Khó là người quen mở lời mà vẫn theo đúng quy tắc này. +- Nói thẳng: Lương phải trả đúng hạn, đủ số, theo hợp đồng và theo pháp luật. Nợ lương thì nhân viên có thể trực tiếp xin tòa ra lệnh thanh toán. Tăng lương, thưởng, ứng trước chỉ nói miệng thì khi tranh cãi, bên không đưa được bằng chứng sẽ chịu thiệt. Tự bỏ tiền túi cho nhân viên vay thì phải viết giấy vay, ghi rõ số tiền, lãi suất và thời hạn trả. +- Lợi ích: Những nội dung bắt buộc phải ghi trong hợp đồng lao động gồm cách trả lương, cách đóng bảo hiểm xã hội. Đơn vị phải theo hợp đồng và quy định của nhà nước, trả lương đúng hạn và đủ số. Kéo dài không trả thì nhân viên có thể trực tiếp xin tòa án ra lệnh thanh toán (văn bản của tòa buộc người nợ phải trả). Ông chủ bỏ tiền của mình cho nhân viên vay thì đó là một hợp đồng vay. Trừ khi giữa hai cá nhân có thỏa thuận khác, việc vay phải lập thành văn bản, ghi rõ vay bao nhiêu, tính lãi thế nào, khi nào trả, trả ra sao. Thưởng chỉ được hứa miệng không có giấy văn, khi tranh cãi thì bên không đưa được bằng chứng chịu thiệt (toàn quốc). +- Mức bằng chứng: C +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007). 劳动合同法（第十七、三十条）. ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第四百六十九、六百六十八条）. +- Ghi chú: Không phải là không giúp, mà là giúp cũng phải theo thể chế. Ăn huệ ngoài thể chế một khi thành thói quen, muốn dừng lại sẽ sinh tranh cãi. Vay riêng lẫn vào sổ công ty, công ty lỗ thì khoản nợ xấu rơi vào túi bạn. Phần điều luật đã đối chiếu, còn lời khuyên là kinh nghiệm, nên để mức C. + +### 18. Bán trước rồi mới sản xuất: dùng bán trước và đơn nhỏ để kiểm chứng có người mua, rồi mới bỏ tiền vào sản xuất + +- Chi phí: Vài chục đến vài trăm tệ, làm một mẫu hoặc một trang. Rồi đi tìm vài chục khách thật, bảo họ trả trước hoặc đặt đơn nhỏ. Cách này chậm hơn vài tuần so với khoảng sản xuất ngay. Khó là nhịt không làm hàng trước. +- Nói thẳng: Có người theo dõi 116 công ty mới ở Italia khoảng một năm: người khởi nghiệp đặt giả thuyết trước rồi lấy phản hồi thị trường để kiểm chứng làm ăn tốt hơn, cũng sẵn sàng đổi hướng kịp thời hơn, mà không bỏ cuộc sớm hơn. Nên nếu không ai chịu trả trước, không ai đặt đơn nhỏ, đừng tin「làm ra có người mua」. +- Lợi ích: Một thử nghiệm chia nhóm ngẫu nhiên theo dõi 116 doanh nghiệp khởi nghiệp ở Italia khoảng một năm. Thử nghiệm chia nhóm ngẫu nhiên nghĩa là chia người ra hai nhóm ngẫu nhiên, chỉ đào tạo một nhóm, rồi so kết quả hai nhóm. Kết quả là: người khởi nghiệp được đào tạo, ra quyết định theo cách「đặt giả thuyết trước, rồi dùng phản hồi thị trường kiểm chứng nghiêm ngặt」thì kết quả kinh doanh tốt hơn, cũng dễ rẽ hướng kịp thời đổi phương án khác, tỷ lệ bỏ cuộc sớm không cao hơn. Cách giải thích của tác giả là, làm như vậy giảm xác suất tiếp tục đổ tiền vào những dự án trông có vẻ triển vọng nhưng thật ra không làm nổi (dự án dương giả) (Italia, công bố năm 2020). +- Mức bằng chứng: B +- Nguồn: Camuffo A, Cordova A, Gambardella A, Spina C (2020). A Scientific Approach to Entrepreneurial Decision Making: Evidence from a Randomized Control Trial. Management Science 66(2):564-586. +- Ghi chú: Chỉ có một nghiên cứu như vậy, cỡ mẫu nhỏ, lại làm ở Italia, nên để mức B. Tên bài, nguồn xuất bản và phần tóm tắt đã đối chiếu qua Crossref và Semantic Scholar, còn các con số hiệu quả trong toàn văn thì phần này chưa kiểm chứng. Xét tới thao tác cụ thể: nếu không ai chịu trả trước, không ai đặt đơn nhỏ, đừng tin「làm ra có người mua」. + +### 19. Làm xong mẫu thì so qua danh sách sản xuất hàng loạt, rồi hãy nói chuyện khoảng sản xuất + +- Chi phí: Vài trăm tệ cho một lần thử sản xuất số lượng nhỏ. Rồi dành vài tiếng để tính rõ mấy khoản: tỷ lệ thành phẩm (trong 100 sản phẩm thì bán được mấy cái), số lượng đặt hàng tối thiểu và hạn thanh toán của nhà cung cấp, tồn kho chiếm bao nhiêu tiền, bao bì và vận chuyển, chứng nhận và giấy phép, nhãn hiệu. Khó là nhịt không bỏ qua bước này để khoảng sản xuất ngay. +- Nói thẳng: Khác biệt giữa một mẫu và một nghìn cái nằm ở chỗ tiền bị kẹp ở đâu. Tỷ lệ thành phẩm quyết định chi phí thật. Số lượng đặt hàng tối thiểu và hạn thanh toán quyết định bạn phải ứng trước bao nhiêu. Tồn kho quyết định tiền quay lại lúc nào. Nếu sản phẩm nằm trong danh mục chứng nhận bắt buộc thì không chứng nhận, không gắn nhãn chứng nhận thì không được xuất kho và bán. +- Lợi ích: Khác biệt giữa một mẫu và một nghìn cái nằm ở chỗ tiền bị kẹp ở đâu. Tỷ lệ thành phẩm quyết định chi phí thật, tức trong một lô sản xuất ra có bao nhiêu phần trăm đạt và bán được. Nhà cung cấp bắt buộc bạn đặt tối thiểu bao nhiêu (số lượng đặt hàng tối thiểu), giao hàng xong bao lâu thì nhận được tiền (hạn thanh toán), hai khoản này quyết định bạn phải ứng trước bao nhiêu. Tồn kho quyết định tiền quay lại lúc nào. Sản phẩm nằm trong danh mục chứng nhận sản phẩm bắt buộc phải đạt chứng nhận trước, và gắn nhãn hiệu chứng nhận lên sản phẩm, mới được xuất kho và bán. Chưa chứng nhận mà tự ý xuất kho bán thì xử phạt theo nghị định về chứng nhận và công nhận. Kinh doanh ăn uống và bán thực phẩm phải xin giấy phép kinh doanh thực phẩm (toàn quốc). +- Mức bằng chứng: C +- Nguồn: Cục Quản lý chất lượng và kiểm định quốc gia (质检总局) (2009). 强制性产品认证管理规定（质检总局令第 117 号，第二、四十九条）. ；Cục Quản lý thị trường quốc gia (市场监管总局) (2023). 食品经营许可和备案管理办法（市场监管总局令第 78 号，第四条）. +- Ghi chú: Bản danh sách tự thân là phương pháp kinh nghiệm, phần điều luật đã đối chiếu nguyên văn. Trước khi đưa vào sản xuất, vào hệ thống tra cứu nhãn hiệu của Cục Sở hữu trí tuệ quốc gia xem đã có nhãn hiệu gần giống được đăng ký chưa, rồi hãy nộp đăng ký nhãn hiệu của mình. Vạch đỏ khi dùng nhãn hiệu và hình ảnh của người khác, xem mục 20 và 21 (mua hàng phải giữ chứng từ, hình ảnh phải tự làm hoặc mua giấy phép). Danh mục chứng nhận bắt buộc căn cứ bản Cục Quản lý thị trường quốc gia công bố mới nhất. + +### 20. Mua hàng thì mỗi lô giữ lại chứng từ và thông tin bên bán, giá mua thấp hẳn so với thị trường thì không nhận: hàng giả do nhân viên mua, người bị kết án là chủ doanh nghiệp + +- Chi phí: Không tốn tiền. Mỗi lô hàng giữ lại mấy thứ: hợp đồng mua hàng, hóa đơn, chứng từ thanh toán, giấy phép kinh doanh và giấy uỷ quyền của bên bán. Rồi đặt một quy tắc nội bộ: ai mua hàng thì người đó ký tên, giá mua thấp hơn rõ rệt so với giá thị trường thì không nhận. Khó là bỏ qua những lô giá rẻ trông rất có lãi. +- Nói thẳng: Bán hàng giả, thu được 3 vạn tệ, hoặc bán được 5 vạn tệ, là đã đủ để kết án, phạt tù dưới 3 năm. Hàng còn nằm trong kho chưa bán đi cũng tính. 「我不知道是假的」không phải do bạn quyết: mua giá thấp hẳn so với thị trường mà không nêu được lý do thì bị coi là bạn đã biết. Giữ hợp đồng mua hàng, hóa đơn, chứng từ thanh toán và thông tin bên bán, thì mới có đường miễn bồi thường. +- Lợi ích: Bán hàng hóa biết là hàng đăng ký nhãn hiệu giả, khi thu nhập bất hợp pháp đủ 3 vạn tệ, hoặc số tiền bán ra đủ 5 vạn tệ, thì tới ngưỡng phải kết án. Phạt tù dưới 3 năm, kèm phạt tiền, hoặc chỉ phạt tiền. Đạt gấp mười ngưỡng đó thì phạt tù 3 đến 10 năm kèm phạt tiền. Hàng còn nằm trong kho chưa bán cũng tính: giá trị hàng đạt từ ba lần tiêu chuẩn số tiền bán ra trở lên thì vẫn cấu thành tội. 「明知」không cần bạn tự thừa nhận. Giải thích của Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao năm 2025 liệt kê năm tình huống có thể kết luận trực tiếp rằng bạn đã biết, trừ khi bạn đưa được bằng chứng chứng minh thật sự không biết. Một là biết nhãn hiệu đăng ký trên hàng hóa đã bị sửa, đổi hoặc che. Hai là làm giả, sửa giấy uỷ quyền của chủ sở hữu nhãn hiệu, hoặc biết giấy đó bị làm giả, sửa. Ba là đã bị xử phạt hình sự hoặc hành chính vì bán hàng giả, rồi lại bán cùng loại. Bốn là không có lý do chính đáng mà mua hoặc bán với giá thấp hẳn so với giá thị trường. Năm là sau khi cơ quan quản lý hành chính hoặc cơ quan tư pháp phát hiện, chuyển hoặc tiêu huỷ hàng vi phạm cùng chứng từ kế toán, hoặc đưa ra chứng minh giả. Công ty phạm điều này thì「对单位判处罚金，并对其直接负责的主管人员和其他直接责任人员」xử phạt theo cùng điều — tức công ty nộp phạt tiền, người quyết định và người thực hiện bị xử theo cùng điều. Nên hàng do nhân viên mua, người bị phạt và bị kết án vẫn là người quyết định. Ngược lại, Luật Thương hiệu để lại một lối thoát: khi bán mà thật sự không biết hàng này xâm phạm nhãn hiệu đăng ký của người khác, thì「能证明该商品是自己合法取得并说明提供者的，不承担赔偿责任」— ai chứng minh được mình có hàng hóa đó một cách hợp pháp và nêu ra nhà cung cấp thì không phải bồi thường. Có được miễn bồi thường hay không, tất cả quyết định ở việc có chứng từ và thông tin bên bán không; không giữ thì không dùng được điều này (toàn quốc, giải thích có hiệu lực từ ngày 26 tháng 4 năm 2025). +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). 刑法修正案（十一）（第十七、二十四项，即刑法第二百一十四、二百二十条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 关于办理侵犯知识产权刑事案件适用法律若干问题的解释（法释〔2025〕5 号，2025 年 4 月 26 日起施行）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2019). 商标法（2019 年修正，第五十七、六十四条；2026 年 6 月 26 日修订后为第七十二、七十八条，自 2027 年 1 月 1 日起施行）. 、 +- Ghi chú: Mỹ phẩm, đồ uống, hàng mẹ và bé, hàng số, quần áo là những nhóm hàng hay xảy ra vụ này. Điểm chung là cùng một mã hàng có khoảng chênh giá rất lớn. Những cách gọi như「渠道货」「尾单」「海外免税」đều không thay thế được chứng từ. Ngoài kết án còn một mức xử phạt hành chính: khi xác định xâm phạm, hàng vi phạm và dụng cụ phải bị thu giữ, doanh thu kinh doanh phi pháp từ 5 vạn tệ trở lên còn có thể phạt kèm tối đa 5 lần. Đã nhận hàng mới phát hiện sai thì dừng bán ngay, niêm phong lại, rồi liên lạc hãng hoặc cơ quan quản lý thị trường. Đừng bán tiếp, đừng tiêu huỷ hàng và sổ sách. Tiêu huỷ chứng cứ rơi đúng vào tình huống thứ năm「bị suy đoán là đã biết hàng giả」nêu trên. Dùng hình ảnh của người khác, xem mục 21 (hình ảnh thì tự làm hoặc mua giấy phép). Có một kiểu chơi nhắm thẳng vào điều này: có người chủ động mang hàng tới, nói lô bút hay món nhỏ khác này cho bạn miễn phí, để trong cửa hàng bán rồi chia tiền. Vài ngày sau có người tới mua, lấy ngay bằng chứng tại chỗ, rồi đòi bồi thường với lý do xâm phạm. Hàng cho không, đủ ba điều kiện của đường miễn bồi thường đều thiếu: không có hợp đồng mua hàng và chứng từ thanh toán, không nói được nhà cung cấp thật, giá thấp bất thường. Điều「合法取得并说明提供者」không dùng được vào lúc nào cả. Nên hàng được mang tận nơi, không kèm chứng từ, chỉ để lại một tài khoản tin nhắn, thì tuyệt đối không đưa lên kệ. Định bán hộ thì ký hợp đồng bán hộ bằng văn bản, giữ bản sao giấy tờ tùy thân và giấy phép kinh doanh của bên kia, ghi rõ hàng thuộc ai, xâm phạm thì tính vào ai. + +### 21. Hình ảnh trên hàng hóa, bao bì, treo cổ và ảnh quảng bá thì hoặc tự làm hoặc mua giấy phép, đổi màu, thêm biểu tượng không tính là「đã sửa」 + +- Chi phí: Giấy phép dùng ảnh từ thư viện chính hãng rẻ nhất vài chục tệ một ảnh. Thuê người thiết kế riêng, vài trăm đến vài nghìn tệ. Mỗi ảnh dùng trên hàng hóa đều giữ lại một bằng chứng cấp quyền và biên lai mua. Khó là không dùng ảnh tìm đại ý trên mạng. +- Nói thẳng: Đổi màu trên ảnh của người khác, đổi nền, thêm biểu tượng của bạn, vẫn là dùng tác phẩm của người khác mà không được phép. Sao chép tác phẩm nghệ thuật của người khác để kiếm tiền, thu được 3 vạn tệ, làm đến 5 vạn tệ, hoặc in đủ 500 bản, thì cấu thành tội vi phạm bản quyền, phạt tù dưới 3 năm. Với nhãn hiệu,「像到容易混淆」bản thân nó đã là xâm phạm. +- Lợi ích: Luật Bản quyền liệt kê「未经著作权人许可，以改编、翻译、注释等方式使用作品」cùng với việc sao chép trực tiếp là hành vi xâm phạm. Vậy nên sửa màu trên ảnh gốc, đổi nền, thêm biểu tượng của bạn vẫn là dùng tác phẩm của người khác không được phép. Ngưỡng phải kết án nằm ở đây: để kiếm tiền, không được chủ sở hữu bản quyền (chủ sở hữu quyền tác giả) cho phép mà in tác phẩm nghệ thuật ra bán (sao chép phát hành). Đủ một trong ba con số thì cấu thành tội. Thu nhập bất hợp pháp đủ 3 vạn tệ. Hoặc giao dịch này làm đến 5 vạn tệ (số tiền kinh doanh phi pháp). Hoặc tổng số bản in đủ 500 bản (tờ). Chạm một trong ba con số đó là cấu thành tội vi phạm bản quyền. Phạt tù dưới 3 năm, kèm phạt tiền, hoặc chỉ phạt tiền. Đạt gấp mười ngưỡng thì phạt tù 3 đến 10 năm. Chỉ bán hàng in sẵn xâm phạm của người khác, thu được từ 5 vạn tệ trở lên, hoặc bán ra từ 10 vạn tệ trở lên thì xử theo tội bán hàng xâm phạm bản sao có bản quyền, phạt tù dưới 5 năm. Nhãn hiệu cũng vậy: dùng nhãn hiệu gần giống nhãn hiệu đăng ký trên cùng một loại hàng hóa, hoặc dùng nhãn hiệu giống hay gần giống trên hàng hóa tương tự,「容易导致混淆的」bản thân đã nằm trong phạm vi xâm phạm. 「近似」là tình huống xâm phạm luật ghi rõ, sửa qua chút vẫn là xâm phạm (toàn quốc, giải thích có hiệu lực từ ngày 26 tháng 4 năm 2025). +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). 著作权法（2020 年修正，第五十二条第六项、第五十三条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). 刑法修正案（十一）（第二十、二十一项，即刑法第二百一十七、二百一十八条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 关于办理侵犯知识产权刑事案件适用法律若干问题的解释（法释〔2025〕5 号）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2019). 商标法（2019 年修正，第五十七条第二项；2026 年修订后为第七十二条第二项）. +- Ghi chú: Có ba chỗ hay vấp. Một là in hình nhân vật hoạt hình, hình ngôi sao tìm được trên mạng lên áo, cốc, ốp điện thoại. Hai là phông chữ: phông dùng thương mại phần lớn phải có giấy phép riêng, xem trước miễn phí không có nghĩa dùng thương mại miễn phí. Ba là bản thiết kế làm cho khách có lẫn tài nguyên chưa được cấp quyền, trách nhiệm vẫn thuộc về người bán hàng. Muốn gọn thì dùng thư viện ảnh có giấy phép thương mại, và lưu lại trang cấp quyền. Hoặc bắt bên thiết kế ghi rõ trong hợp đồng tài nguyên lấy từ đâu, cấp quyền tới mức nào, có sự cố thì bên thiết kế chịu trách nhiệm. Ngoài kết án còn có bồi thường dân sự và phạt hành chính: gây tổn hại lợi ích công cộng thì có thể thu giữ thu nhập bất hợp pháp, và phạt gấp 1 đến 5 lần doanh thu kinh doanh phi pháp. Trước khi đưa vào sản xuất thì tra nhãn hiệu, xem mục 19 (làm xong mẫu thì so qua danh sách sản xuất hàng loạt). Ngược lại, hình ảnh của bạn bị người khác dùng thì cũng theo bộ quy tắc này mà bảo vệ. + +### 22. Ở đúng vị trí của mình thì làm theo quy chế, ý tốt ngoài quy chế thì nghĩ trước trách nhiệm thuộc về ai + +- Chi phí: Không tốn tiền. Khi làm việc thì theo quy chế, theo y lệnh, theo quy trình. Không quyết thay người khác khi vượt quá thẩm quyền của mình. Trước khi giúp, tự hỏi một câu: có việc thì ai chịu trách nhiệm. Khó là chống lại yêu cầu「linh động một chút」tại chỗ. +- Nói thẳng: Lấy y tá làm ví dụ: nghị định ghi rõ, phát hiện bệnh nhân chuyển biến nguy kịch thì phải báo bác sĩ ngay; phát hiện y lệnh vi phạm quy định thì phải nêu với bác sĩ đã ra y lệnh. Mọi vị trí khác cũng vậy, trước hết phải thấy quy chế của mình đánh dấu「phải báo cáo」ở đâu. Pháp luật cũng không bắt người ta nhìn người sắp chết mà không cứu: tự nguyện ra cứu khẩn cấp mà làm người được cứu bị thương thì không phải bồi tiền. +- Lợi ích: Lấy y tá làm ví dụ, quy định ghi rõ giới hạn nhiệm vụ như sau. Phát hiện bệnh chuyển biến nguy kịch thì báo bác sĩ ngay. Trong tình huống khẩn cấp, để cứu người bệnh sắp tử vong thì có thể tiến hành cấp cứu khẩn cấp cần thiết trước. Phát hiện y lệnh vi phạm pháp luật, hoặc vi phạm quy chuẩn kỹ thuật chẩn đoán trị bệnh thì phải nêu với bác sĩ, cần thì báo trưởng khoa. Bộ luật Dân sự có quy định khác: tự nguyện ra cứu khẩn cấp mà khiến người được cứu bị thương thì bên cứu không phải bồi tiền (không chịu trách nhiệm dân sự) (toàn quốc). +- Mức bằng chứng: B +- Nguồn: Quốc vụ viện (国务院) (2008). 护士条例（国务院令第 517 号，第十七条）. ；Quốc vụ viện (国务院) (2020). 护士条例（2020 年修订，第十七条）. （Cơ sở dữ liệu luật pháp quốc gia）；Quốc hội Nhân dân (全国人大) (2020). 民法典（第一百八十四条）. +- Ghi chú: Ở trong phạm vi công việc thì làm theo quy chế, cứu người trong tình huống khẩn cấp thì được pháp luật bảo vệ. Còn những việc khác, trước hết phải phân biệt đó là thẩm quyền của ai, trách nhiệm của ai, đừng dùng thiện ý riêng thay cho quyết định của hệ thống. Nghị định về y tá năm 2020 được sửa đổi bằng lệnh của Quốc vụ viện số 726, chỉ sửa mấy điều về đăng ký, Điều 17 không đổi một chữ. Điều trong Luật Bác sĩ về việc hành nghề theo quy chuẩn chẩn đoán trị bệnh, lượt này không mở được nguyên văn chính thức nên không dẫn. + +### 23. Lỗ thua thì rút lui theo thủ tục: hủy được theo lối giản dị thì hủy, vốn không đủ trả nợ thì đi thủ tục phá sản, đừng để yên + +- Chi phí: Tự làm thủ tục hủy đăng ký giản dị được: đăng lên Hệ thống công bố thông tin tín dụng doanh nghiệp quốc gia 20 ngày. Còn nợ chưa trả xong thì phải làm rõ sổ sách trước, thu cái phải thu, trả cái phải trả, bước này gọi là thanh lý. Đi thủ tục phá sản thì phải xin tòa án, mất mấy tháng. Khó là lỗ thua rồi không muốn đối mặt, kéo dài không làm. +- Nói thẳng: Không nợ ai, không nợ lương, không nợ thuế, không nợ bảo hiểm xã hội, thì toàn thể nhà đầu tư viết một bản cam kết, công bố 20 ngày không có ai phản đối thì hủy đăng ký giản dị được. Hộ kinh doanh cá thể không phải công bố, 10 ngày không có phản đối thì hủy thẳng. Không trả nổi nợ thì xin phá sản. Để yên, 2 năm không báo cáo thường niên lại không liên lạc được thì bị thu hồi giấy phép, người đại diện theo pháp luật có trách nhiệm không được làm người đại diện trong 3 năm. +- Lợi ích: Không có nợ của mình, và cũng không ai nợ mình, hoặc đã trả xong hết. Không nợ lương, không nợ bảo hiểm xã hội, không nợ thuế. Toàn thể nhà đầu tư đã viết cam kết bằng văn bản. Đủ mấy điều đó, đăng công bố 20 ngày không có ai phản đối thì đi được thủ tục hủy đăng ký giản dị. Hộ kinh doanh cá thể không phải công bố, 10 ngày không có ý kiến phản đối của cơ quan liên quan thì hủy thẳng. Công ty không trả nổi khoản nợ đến hạn, số nợ lớn hơn số tài sản đang có (vốn không đủ trả nợ) thì có thể xin tòa án mở thủ tục tái cơ cấu, thỏa thuận hoặc phá sản thanh lý. Công ty đã giải thể mà chưa thanh lý, số nợ lớn hơn số tài sản, thì người phụ trách thanh lý phải xin thủ tục phá sản thanh lý. Hậu quả của việc để yên là: không công bố báo cáo thường niên đúng hạn thì bị đưa vào danh sách bất thường trong hoạt động. Liên tiếp 2 năm không báo cáo thường niên, đã vào danh sách bất thường mà vẫn không sửa, lại không liên lạc được theo địa chỉ đã đăng ký thì bị thu hồi giấy phép kinh doanh. Người làm đại diện theo pháp luật của doanh nghiệp bị thu hồi giấy phép và chịu trách nhiệm cá nhân thì, kể từ ngày bị thu hồi, 3 năm không được làm đại diện theo pháp luật, giám đốc, giám sát, lãnh đạo cấp cao của công ty. Đã vào danh sách bất thường trong hoạt động thì ngay cả thủ tục hủy đăng ký giản dị cũng không đi được (toàn quốc). +- Mức bằng chứng: A +- Nguồn: Quốc vụ viện (国务院) (2021). 市场主体登记管理条例（国务院令第 746 号，第三十一、三十二、三十三条）. ；Sáu bộ ngành gồm Cục Quản lý thị trường quốc gia (市场监管总局) v.v. (2025). 企业注销指引（2025 年修订）（2025 年第 52 号公告）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2006). 企业破产法（第二、七条）. ；Quốc vụ viện (国务院) (2024 修订). 企业信息公示暂行条例（国务院令第 777 号修订，第十八条）：「企业因连续2年未按规定报送年度报告被列入经营异常名录未改正，且通过登记的住所或者经营场所无法取得联系的，由县级以上市场监督管理部门吊销营业执照」. （Cơ sở dữ liệu luật pháp quốc gia）；Quốc vụ viện (国务院) (2024). 关于修改和废止部分行政法规的决定（国务院令第 777 号）. ；市场主体登记管理条例（第十二条第四项）与公司法（2023 年修订，第一百七十八条第一款第四项）：担任因违法被吊销营业执照的企业的法定代表人并负有个人责任的，自被吊销营业执照之日起未逾 3 年不得担任法定代表人和董事、监事、高级管理人员. ； +- Ghi chú: Nghị định tạm về công bố thông tin doanh nghiệp năm 2024 được sửa bằng lệnh của Quốc vụ viện số 777, có hiệu lực từ ngày 1 tháng 5, từ đó không sửa thêm. Điều 17 của bản cũ「经营异常满 3 年列入严重违法企业名单」đã bị xóa. Danh sách doanh nghiệp vi phạm nghiêm trọng, mất uy tín hiện nay quản lý doanh nghiệp báo cáo thường niên gian dối, tình tiết nặng. Trên mạng rất nhiều lời nói vẫn theo bản cũ, đừng sao chép. Chế độ phá sản cá nhân (kể cả hộ kinh doanh cá thể) hiện mới chỉ có vài địa phương thí điểm, phần này chưa kiểm chứng nguyên văn nên không viết. diff --git "a/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" index 0684c81e..9abb2265 100644 --- "a/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" +++ "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" @@ -1,59 +1,59 @@ [← Về mục lục](../README.md) -# 34. Đừng dùng thuốc trong nhà sai cách +# 34. Đừng để thuốc trong nhà gây ra chuyện -Thuốc hạ sốt, thuốc giảm đau, thuốc cảm cúm, thuốc dạ dày, thuốc chống tiêu chảy, nhiều loại mua ở hiệu thuốc là tự mua được. Mua không cần đơn thuốc không có nghĩa là uống thế nào cũng được. Phần này không nói mỗi bệnh nên uống thuốc gì, chỉ chọn ra mấy việc chỉ cần một hành động là tránh được hậu quả nặng. +Thuốc hạ sốt, thuốc giảm đau, thuốc cảm cúm, thuốc dạ dày, thuốc chống tiêu chảy, nhiều loại tự mua được ở hiệu thuốc. Mua được mà không cần đơn không có nghĩa uống thế nào cũng được. Phần này không nói bệnh gì thì uống thuốc nào, chỉ chọn ra mấy việc chỉ cần một động tác là tránh được hậu quả nặng. -Tất cả các mục tính theo tỷ lệ tử vong, gồm cả những kết cục sức khoẻ như suy gan, xuất huyết dạ dày, suy thận ở trẻ sơ sinh. Phần lớn lợi ích là「tránh được một vụ tai nạn nghiêm trọng」, nghiên cứu không đưa ra con số「làm rồi giảm được bao nhiêu phần trăm」, mức lợi ích được định theo mức độ nặng của hậu quả, mỗi mục ghi rõ trong Ghi chú căn cứ để định. +Tất cả mục trong phần này tính theo thước đo tỷ lệ tử vong, gồm cả những kết cục sức khoẻ như suy gan, xuất huyết dạ dày, suy thận ở trẻ sơ sinh. Phần lớn lợi ích là「được một lần tai nạn nghiêm trọng bớt đi」, nghiên cứu không đưa ra con số「làm thì giảm được mấy phần trăm」, cỡ lợi ích được xếp theo mức độ nặng của hậu quả, và mỗi mục ghi rõ trong Ghi chú căn cứ để xếp. -Những việc đã viết ở chỗ khác, ở đây chỉ dẫn đường. Nuốt nhầm thuốc hoặc chất tẩy rửa thì làm gì trước, xem mục 20 ở phần 13（nuốt nhầm）. Thuốc bệnh mạn uống đủ theo chỉ định, xem mục 1 ở phần 16（uống thuốc đủ theo chỉ định）. Mua thuốc có đơn trên mạng phải qua duyệt đơn trước, xem mục 6 ở phần 28（thuốc giảm cân thì đi bệnh viện lấy đơn）. +Những việc đã viết ở chỗ khác thì ở đây chỉ dẫn đường. Uống nhầm thuốc hoặc chất tẩy rửa thì làm gì trước, xem mục 20 ở phần 13 (uống nhầm). Thuốc bệnh mạn thì uống đủ theo đơn, xem mục 1 ở phần 16 (uống thuốc đủ theo chỉ định). Mua thuốc có đơn trên mạng thì phải qua kiểm tra đơn trước, xem mục 6 ở phần 28 (thuốc giảm cân thì đến bệnh viện lấy đơn). -Người được lợi của các mục trong phần này là bạn và gia đình bạn, thuộc hai mức cao nhất. +Mục trong phần này có người hưởng là bạn và gia đình bạn, thuộc hai bậc cao nhất. -### 1. Trước khi uống hai loại thuốc cảm cúm hoặc thuốc giảm đau cùng lúc, xem bảng thành phần, paracetamol chỉ được chiếm một chỗ +### 1. Trước khi uống hai loại thuốc cảm cúm hoặc giảm đau cùng lúc, hãy xem bảng thành phần, paracetamol chỉ được chiếm một chỗ -- Chi phí: Không tốn tiền. Nhìn bảng thành phần trên mỗi hộp một chút, một hai phút. -- Nói thẳng: Rất nhiều thuốc cảm cúm có pha paracetamol, uống hai hộp cùng lúc là uống gấp đôi, quá liều sẽ hại gan. Ở Hoa Kỳ, gần ba phần mười số người bị suy gan cấp tính do chính chất này đã chết trước khi kịp ghép gan. Tờ hướng dẫn trong nước khuyên mỗi ngày tối đa 2 g, trong thời gian uống thuốc cũng đừng uống rượu. -- Lợi ích: 22 bệnh viện ở Hoa Kỳ đăng ký liên tục 662 ca suy gan cấp tính trong 6 năm. Trong đó 275 ca (42%) do paracetamol gây ra, liều đã nuốt có trung vị 24 g. Trong 275 người này, 131 người (48%) ăn nhầm quá liều, không phải cố ý. Trong nhóm ăn nhầm quá liều, 38% còn uống kèm hai loại thuốc trở lên có chứa thành phần này. Trong 275 người đó, 27% chết trước khi kịp ghép gan, 8% được ghép gan. Năm 2020 Cục Quản lý Dược phẩm nhà nước yêu cầu tờ hướng dẫn thuốc không cần đơn phải ghi: dùng quá liều gây tổn thương gan nặng, liều tối đa trong một ngày không nên quá 2 g. Tờ hướng dẫn còn phải ghi, nên tránh dùng kèm với thuốc có paracetamol hoặc thuốc giảm đau hạ sốt khác. Quy định Hoa Kỳ yêu cầu loại thuốc không cần đơn này ghi ở cảnh báo thứ nhất ba tình huống gây hại gan: vượt quá liều tối đa trong ngày, uống chung với thuốc khác có paracetamol, và trong thời gian uống thuốc mỗi ngày uống trên 3 cốc rượu. +- Chi phí: Không tốn tiền. Nhìn bảng thành phần của từng hộp thuốc một lượt, một hai phút. +- Nói thẳng: Nhiều thuốc cảm cúm có trộn paracetamol. Uống hai hộp cùng lúc tức là uống gấp đôi, quá liều thì hỏng gan. Ở Hoa Kỳ, trong số người bị suy gan cấp do paracetamol, gần ba phần mười chết trước khi kịp ghép gan. Hướng dẫn trong nước khuyên mỗi ngày tối đa 2 g, và trong lúc uống thuốc đừng uống rượu. +- Lợi ích: Mỹ có 22 bệnh viện ghi nhận liên tục trong 6 năm 662 ca suy gan cấp. Trong đó 275 ca (42%) do paracetamol, liều đã nuốt có trung vị 24 g. Trong 275 người đó, 131 người (48%) uống quá liều do không cẩn thận, không phải cố ý. Trong nhóm uống quá liều do không cẩn thận, 38% uống đồng thời hai loại thuốc trở lên có chứa thành phần này. Trong 275 người đó, 27% chết trước khi kịp ghép gan, 8% được ghép gan. Năm 2020 Cục Quản lý Dược phẩm Quốc gia yêu cầu hướng dẫn thuốc không kê đơn phải ghi rõ: dùng quá liều gây tổn thương gan nặng, liều tối đa mỗi ngày không nên vượt quá 2 g. Hướng dẫn còn phải ghi, nên tránh dùng chung với thuốc có paracetamol hoặc thuốc giảm đau hạ nhiệt khác. Quy định Mỹ yêu cầu loại thuốc không kê đơn này ghi ở cảnh báo thứ nhất ba tình huống gây hại gan: vượt quá liều tối đa trong một ngày, uống chung với thuốc khác có paracetamol, và uống từ 3 ly rượu trở lên mỗi ngày trong lúc dùng thuốc. - Mức bằng chứng: B -- Nguồn: Larson AM, Polson J, Fontana RJ, et al.; Acute Liver Failure Study Group (2005). Acetaminophen-induced acute liver failure: results of a United States multicenter, prospective study. Hepatology, 42(6):1364-1372. ；Cục Quản lý Dược phẩm nhà nước (国家药监局) (2020). Thông báo về việc sửa đổi tờ hướng dẫn dạng thường và dạng giải phóng của paracetamol (2020年第15号) và phụ lục 2《对乙酰氨基酚常释制剂非处方药说明书修订要求》(Cục Quản lý Dược phẩm tỉnh Hồ Nam (湖南省药品监督管理局) đăng lại). ；21 CFR 201.326(a)(1)(iii)(A) Liver warning. -- Ghi chú: Paracetamol còn gọi là扑热息痛. Tên thuốc cảm cúm phối hợp có chữ「氨酚」trong tên, ví dụ 氨酚伪麻、氨咖黄敏, đều có thành phần này. Không chắc thì cứ mang mấy hộp thuốc tới dược sĩ ở hiệu thuốc. Để ở mức B là vì nhóm số liệu Hoa Kỳ là đăng ký ca bệnh, không có nhóm chứng, tính không ra xem bảng thành phần giúp bớt được bao nhiêu sự cố. Đặt lợi ích là「lớn」là theo hậu quả: trong số ca suy gan cấp tính có gần ba phần mười chết thẳng. +- Nguồn: Larson AM, Polson J, Fontana RJ, et al.; Acute Liver Failure Study Group (2005). Acetaminophen-induced acute liver failure: results of a United States multicenter, prospective study. Hepatology, 42(6):1364-1372. ；Thông báo của Cục Quản lý Dược phẩm Quốc gia về việc sửa đổi hướng dẫn sử dụng dược phẩm paracetamol phóng thích và phóng thích kéo dài (2020年第15号) kèm phụ lục 2《对乙酰氨基酚常释制剂非处方药说明书修订要求》, đăng lại bởi Sở Quản lý Dược phẩm tỉnh Hunan (湖南省药品监督管理局). ；21 CFR 201.326(a)(1)(iii)(A) Liver warning. +- Ghi chú: Paracetamol còn gọi là 扑热息痛. Tên thông thường của thuốc cảm cúm phối hợp có hai chữ「氨酚」, như 氨酚伪麻, 氨咖黄敏, đều chứa thành phần này. Không chắc thì mang cả mấy hộp thuốc đến hiệu thuốc cho dược sĩ xem. Để ở mức B là vì nhóm số liệu Mỹ là đăng ký ca bệnh, không có nhóm đối chứng, không tính được xem bảng thành phần thì bớt được bao nhiêu sự cố. Cỡ lợi ích đặt là「lớn」là theo hậu quả: gần ba phần mườu ca suy gan cấp chết thẳng. -### 2. Trẻ sốt không dùng aspirin, nimesulide và metamizole, thuốc cảm cúm có chứa metamizole cũng không cho trẻ vị thành niên uống +### 2. Trẻ bị sốt đừng dùng aspirin, nimesulide và metamizole, thuốc cảm cúm chứa metamizole cũng không cho người chưa thành niên uống -- Chi phí: Không tốn tiền. Khi mua thuốc hạ sốt thì nhìn tên thuốc gốc và bảng thành phần. -- Nói thẳng: Trẻ sốt, ba loại thuốc này đều đừng dùng. Aspirin có liên quan tới hội chứng Reye, bệnh này tổn thương não và gan, theo thống kê ở Hoa Kỳ khoảng ba phần mười số trẻ mắc chết, sau khi cảnh báo được ban hành số ca giảm hơn chín phần mười. Nimesulide (尼美舒利) cấm dùng cho trẻ dưới 12 tuổi, metamizole (安乃近) cấm dùng cho người dưới 18 tuổi. -- Lợi ích: Hội chứng Reye gây bệnh não và thoái hóa mỡ gan, thường xuất hiện sau khi mắc cúm hoặc thủy đậu. Trong hệ thống giám sát toàn quốc của Hoa Kỳ từ 1981 đến 1997 có 1207 ca báo cáo ở người dưới 18 tuổi. 82% số trẻ bệnh đo được acid salicylic trong máu, đây là thành phần của aspirin và các thuốc cùng nhóm. Tỷ lệ chết chung 31%. Cảnh báo về nhóm thuốc chứa acid salicylic bắt đầu ban hành từ năm 1980. Năm đó có 555 ca báo cáo, là đỉnh, từ năm 1987 mỗi năm không quá 36 ca. Năm 2011 Cục Quản lý Thực phẩm và Dược phẩm nhà nước quy định, dạng uống của nimesulide cấm dùng cho trẻ dưới 12 tuổi. Năm 2020 Cục Quản lý Dược phẩm nhà nước quy định, viên metamizole cấm dùng cho trẻ vị thành niên dưới 18 tuổi. Các dạng phối hợp có chứa metamizole như viên phối hợp thanh hao và metamizole, viên và viên nang trọng cảm cũng cấm dùng cho người dưới 18 tuổi. +- Chi phí: Không tốn tiền. Khi mua thuốc hạ sốt thì nhìn tên thông thường và bảng thành phần. +- Nói thẳng: Trẻ bị sốt thì cả ba loại thuốc này đều đừng dùng. Aspirin có liên quan đến hội chứng Reye, căn bệnh này tổn thương não và gan; theo thống kê Mỹ khoảng ba phần mười trẻ mắc bệnh chết, sau khi cảnh báo phát ra số ca giảm hơn chín phần mười. Nimesulide cấm dưới 12 tuổi, metamizole cấm dưới 18 tuổi. +- Lợi ích: Hội chứng Reye gây bệnh não và thoái hoá mỡ gan, thường xuất hiện sau khi bị cúm hoặc thủy đậu. Trong hệ thống giám sát toàn quốc Mỹ từ năm 1981 đến 1997 có 1207 ca ở người dưới 18 tuổi. 82% trẻ bệnh đo được acid salicylic trong máu, đây là thành phần của aspirin và nhóm thuốc cùng loại. Tỷ lệ tử vong toàn phần 31%. Từ năm 1980 bắt đầu phát cảnh báo cho nhóm thuốc chứa acid salicylic. Năm đó có 555 ca, là đỉnh; từ năm 1987 mỗi năm không quá 36 ca. Năm 2011 Cục Quản lý Dược phẩm và Thực phẩm Quốc gia quy định cấm dùng dạng uống của nimesulide cho trẻ dưới 12 tuổi. Năm 2020 Cục Quản lý Dược phẩm Quốc gia quy định cấm viên nén metamizole cho trẻ em và thanh thiếu niên dưới 18 tuổi. Các viên phối hợp chứa metamizole như 青蒿安乃近片, 重感灵片 và 重感灵胶囊 cũng bị cấm cho người dưới 18 tuổi. - Mức bằng chứng: B -- Nguồn: Belay ED, Bresee JS, Holman RC, Khan AS, Shahriari A, Schonberger LB (1999). Reye's syndrome in the United States from 1981 through 1997. New England Journal of Medicine, 340(18):1377-1382. ；Thông báo về việc tăng cường quản lý việc sử dụng dạng uống của nimesulide (国食药监安〔2011〕209号). ；Cục Quản lý Dược phẩm nhà nước (国家药监局) (2020). Thông báo về việc sửa đổi tờ hướng dẫn các chế phẩm liên quan đến metamizole (2020年第34号) và phụ lục 1 đến 3. -- Ghi chú: Metamizole với người lớn cũng không phải lựa chọn đầu tiên. Tờ hướng dẫn ghi rõ nó có thể gây mất bạch cầu hạt, thiếu máu do tủy xương và sốc phản vệ, chỉ dùng trong thời gian ngắn khi bệnh nặng, cấp tốc mà không có thuốc nào khác hiệu quả. Tên thuốc trọng cảm nghe như thuốc cảm cúm thường, bên trong có metamizole. Trẻ nên dùng thuốc hạ sốt gì, liều bao nhiêu theo cân nặng, hỏi bác sĩ nhi khoa hoặc dược sĩ. Trẻ dưới 3 tháng sốt thì đi thẳng bệnh viện, xem mục 8 ở phần 20 (38 ℃ thì đi thẳng bệnh viện). Để ở mức B là vì số liệu về hội chứng Reye đến từ báo cáo giám sát, việc số ca giảm và việc ban hành cảnh báo chỉ là quan hệ trước sau, không phải thử nghiệm đối chứng. Đặt lợi ích là「lớn」, vì sau khi có cảnh báo số ca mỗi năm rơi từ 555 xuống dưới 36. +- Nguồn: Belay ED, Bresee JS, Holman RC, Khan AS, Shahriari A, Schonberger LB (1999). Reye's syndrome in the United States from 1981 through 1997. New England Journal of Medicine, 340(18):1377-1382. ；Thông tư về việc tăng cường quản lý việc sử dụng dạng uống của nimesulide (国食药监安〔2011〕209号). ；Thông báo của Cục Quản lý Dược phẩm Quốc gia về việc sửa đổi hướng dẫn sử dụng cho các chế phẩm liên quan đến metamizole (2020年第34号) kèm phụ lục 1 đến 3. +- Ghi chú: Metamizole với người lớn cũng không phải lựa chọn hàng đầu. Hướng dẫn ghi rõ nó có thể gây mất bạch cầu hạt, thiếu máu bất tái sinh và sốc dị ứng, chỉ dùng ngắn hạn khi bệnh nặng cấp tính mà không có thuốc hiệu quả khác. 重感灵 nghe tên như thuốc cảm cúm thường, bên trong có metamizole. Trẻ nên dùng thuốc hạ sốt nào, liều bao nhiêu theo cân nặng thì hỏi bác sĩ nhi khoa hoặc dược sĩ. Trẻ dưới 3 tháng bị sốt thì đi thẳng bệnh viện, xem mục 8 ở phần 20 (38 ℃ thì đi thẳng bệnh viện). Để ở mức B là vì số liệu hội chứng Reye đến từ báo cáo giám sát, việc số ca giảm và việc phát cảnh báo chỉ là quan hệ trước sau, không phải thử nghiệm đối chứng. Cỡ lợi ích đặt là「lớn」, vì sau khi có cảnh báo số ca rơi từ 555 ca mỗi năm xuống dưới 36 ca. -### 3. Người trên 60 tuổi, đã từng chảy máu dạ dày, đang dùng thuốc chống đông máu hoặc thuốc nội tiết thì hỏi bác sĩ trước khi uống thuốc giảm đau như ibuprofen +### 3. Người trên 60 tuổi, đã từng xuất huyết dạ dày, đang uống thuốc chống đông hoặc hormone, trước khi uống thuốc giảm đau như ibuprofen hãy hỏi bác sĩ - Chi phí: Không tốn tiền. Hỏi bác sĩ hoặc dược sĩ một lần. -- Nói thẳng: Nhóm thuốc giảm đau như ibuprofen, naproxen, aspirin làm hại dạ dày. Trong các thử nghiệm liều cao, người uống ibuprofen có nguy cơ xuất huyết, thủng dạ dày cao khoảng 4 lần người không uống. Người lớn tuổi, đã từng chảy máu dạ dày, đang uống thuốc chống đông máu hoặc thuốc nội tiết, uống rượu mỗi ngày thì nguy cơ cao hơn. -- Lợi ích: Nhóm thuốc này gọi là thuốc chống viêm không steroid, thường gặp có ibuprofen, naproxen, diclofenac. Một phân tích tổng hợp gộp 280 thử nghiệm ngẫu nhiên của nhóm thuốc này so với giả dược, tổng cộng 124513 người. So với giả dược, nguy cơ thủng, tắc nghẽn hoặc xuất huyết tiêu hoá trên: ibuprofen RR 3.97 (95% CI 2.22–7.10, khoảng 4 lần), naproxen 4.22 (2.71–6.56), diclofenac 1.89 (1.16–3.09). Tất cả các thuốc trong nhóm này làm nguy cơ suy tim mạch tăng khoảng gấp đôi. Quy định Hoa Kỳ yêu cầu thuốc không cần đơn thuộc nhóm này phải ghi cảnh báo xuất huyết dạ dày. Cảnh báo liệt kê sáu nhóm người có nguy cơ cao hơn: trên 60 tuổi, đã từng loét dạ dày hoặc chảy máu, đang dùng thuốc chống đông máu hoặc thuốc nội tiết, đang dùng kèm thuốc khác thuộc nhóm này, trong thời gian uống thuốc mỗi ngày uống trên 3 cốc rượu, uống nhiều hơn hoặc lâu hơn hướng dẫn. +- Nói thẳng: Nhóm thuốc giảm đau như ibuprofen, naproxen, aspirin gây hại dạ dày. Trong thử nghiệm liều cao, người uống ibuprofen có nguy cơ xuất huyết dạ dày, thủng dạ dày cao khoảng 4 lần so với người không uống. Người lớn tuổi, đã từng xuất huyết dạ dày, đang uống thuốc chống đông hoặc hormone, uống rượu mỗi ngày thì nguy cơ cao hơn. +- Lợi ích: Nhóm thuốc này gọi là thuốc chống viêm không steroid, thường gặp là ibuprofen, naproxen, diclofenac. Một phân tích tổng hợp gộp 280 thử nghiệm ngẫu nhiên của nhóm thuốc này với giả dược, tổng cộng 124513 người. So với giả dược, nguy cơ thủng, tắc hoặc xuất huyết đường tiêu hoá trên: ibuprofen RR 3.97 (95% CI 2.22–7.10, khoảng 4 lần), naproxen 4.22 (2.71–6.56), diclofenac 1.89 (1.16–3.09). Tất cả thuốc trong nhóm này làm nguy cơ suy tim khoảng gấp đôi. Quy định Mỹ yêu cầu thuốc không kê đơn loại này phải ghi cảnh báo xuất huyết dạ dày. Cảnh báo liệt kê sáu nhóm người có nguy cơ cao hơn: trên 60 tuổi, từng có loét dạ dày hoặc xuất huyết dạ dày, đang uống thuốc chống đông hoặc hormone, đang uống thuốc khác cùng nhóm, uống từ 3 ly rượu trở lên mỗi ngày trong lúc dùng thuốc, và dùng nhiều hơn hoặc lâu hơn hướng dẫn. - Mức bằng chứng: A - Nguồn: Coxib and traditional NSAID Trialists' (CNT) Collaboration; Bhala N, Emberson J, Merhi A, et al. (2013). Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. Lancet, 382(9894):769-779. ；21 CFR 201.326(a)(2)(iii)(A) Stomach bleeding warning. -- Ghi chú: Kết luận của bài báo nói về liều cao. Rủi ro khi thỉnh thoảng uống một hai ngày theo đúng liều trong tờ hướng dẫn thì bài này không đưa riêng con số. Rất nhiều thuốc cảm cúm phối hợp cũng có thành phần thuộc nhóm này, uống chung với thuốc giảm đau là bị trùng, hãy xem bảng thành phần trước. Bác sĩ bảo bạn uống aspirin liều nhỏ lâu dài để phòng huyết khối thì đừng tự ngừng, cũng đừng tự thêm một loại thuốc giảm đau nữa. +- Ghi chú: Kết luận của bài báo nói về cách dùng liều cao. Rủi ro khi thỉnh thoảng uống một hai ngày, đúng liều trong hướng dẫn, bài này không đưa con số riêng. Nhiều thuốc cảm cúm phối hợp cũng chứa thành phần nhóm này, uống cùng thuốc giảm đau thì thành phần bị lặp, xem bảng thành phần trước. Nếu bác sĩ bảo bạn uống aspirin liều thấp lâu dài để phòng huyết khối thì đừng tự ngừng, cũng đừng tự thêm một loại thuốc giảm đau nữa. -### 4. Trẻ dưới 2 tuổi bị cảm, đừng tự cho uống thuốc cảm cúm phối hợp và thuốc ho +### 4. Trẻ dưới 2 tuổi bị cảm cúm, đừng tự cho uống thuốc cảm cúm phối hợp và thuốc ho -- Chi phí: Không tốn tiền, lại còn tiết kiệm được tiền mua thuốc. -- Nói thẳng: Cục Quản lý Dược phẩm nhà nước yêu cầu 14 loại thuốc cảm cúm phối hợp dành cho trẻ phải có cảnh báo: không nên để bố mẹ tự cho trẻ dưới 2 tuổi dùng, phải dùng dưới sự hướng dẫn của bác sĩ hoặc dược sĩ. Trong các thử nghiệm làm trên trẻ, thuốc ho và thuốc chống dị ứng đều không tốt hơn giả dược. -- Lợi ích: Năm 2021, thông báo số 57 của Cục Quản lý Dược phẩm nhà nước sửa đổi tờ hướng dẫn của 14 chế phẩm, gồm dung dịch uống paracetamol và pseudoephedrine, hạt paracetamol và chlorpheniramine dành cho trẻ, hạt paracetamol và dextromethorphan dành cho trẻ. Cảnh báo thêm vào là「不建议家长或监护人自行给2岁以下婴幼儿使用本品，应在医师或药师的指导下使用」— không nên bố mẹ hoặc người giám hộ tự cho trẻ sơ sinh và trẻ nhỏ dưới 2 tuổi dùng sản phẩm này, phải dùng dưới sự hướng dẫn của bác sĩ hoặc dược sĩ. Mục lưu ý thêm「应严格按照药品说明书用法用量使用，避免用药过量」— phải dùng đúng liều theo tờ hướng dẫn, tránh dùng quá liều. Cũng thêm「应避免合并使用含有相同或相似活性成份的抗感冒药」— nên tránh dùng kèm thuốc cảm cúm có cùng hoặc gần giống thành phần hoạt chất. Một bài tổng hợp Cochrane đưa vào 29 thử nghiệm đối chứng ngẫu nhiên, trong đó 10 thử nghiệm làm trên trẻ. Trong các thử nghiệm trên trẻ, không nhóm thuốc nào tốt hơn giả dược: thuốc ho, thuốc kháng histamine (thành phần chống dị ứng), thuốc kháng histamine phối hợp chất làm thông mũi, thuốc ho phối hợp thuốc giãn phế quản. 21 nghiên cứu có báo cáo tác dụng phụ, chế phẩm chứa thuốc kháng histamine và dextromethorphan thì tác dụng phụ nhiều hơn. +- Chi phí: Không tốn tiền, lại còn tiết kiệm tiền mua thuốc. +- Nói thẳng: Cục Quản lý Dược phẩm Quốc gia yêu cầu 14 loại thuốc cảm cúm phối hợp dành cho trẻ phải có cảnh báo: không khuyến nghị cha mẹ tự cho trẻ dưới 2 tuổi dùng, phải dùng dưới sự hướng dẫn của bác sĩ hoặc dược sĩ. Trong các thử nghiệm thực hiện trên trẻ, thuốc ho và thuốc chống dị ứng đều không tốt hơn giả dược. +- Lợi ích: Năm 2021 thông báo số 57 của Cục Quản lý Dược phẩm Quốc gia sửa đổi hướng dẫn của 14 chế phẩm, gồm 氨酚麻美口服溶液, 小儿氨酚黄那敏颗粒, 小儿氨酚烷胺颗粒 và các loại khác. Cảnh báo mới thêm vào là「不建议家长或监护人自行给2岁以下婴幼儿使用本品，应在医师或药师的指导下使用」— không khuyến nghị cha mẹ hoặc người giám hộ tự cho trẻ sơ sinh và trẻ nhỏ dưới 2 tuổi dùng sản phẩm này, phải dùng dưới sự hướng dẫn của bác sĩ hoặc dược sĩ. Mục lưu ý thêm「应严格按照药品说明书用法用量使用，避免用药过量」— phải dùng đúng cách và đúng liều ghi trong hướng dẫn, tránh dùng quá liều. Còn thêm「应避免合并使用含有相同或相似活性成份的抗感冒药」— nên tránh dùng đồng thời thuốc cảm cúm khác có thành phần hoạt chất giống nhau hoặc tương tự. Một tổng quan hệ thống Cochrane đưa vào 29 thử nghiệm đối chứng ngẫu nhiên, trong đó 10 thử nghiệm thực hiện trên trẻ. Trong các thử nghiệm ở trẻ, các nhóm thuốc này đều không tốt hơn giả dược: thuốc ho, thuốc kháng histamine (thành phần chống dị ứng), thuốc kháng histamine kết hợp chất làm thông mũi, thuốc ho kết hợp thuốc giãn phế quản. Có 21 nghiên cứu báo cáo tác dụng phụ, chế phẩm chứa thuốc kháng histamine và dextromethorphan ghi nhận nhiều tác dụng phụ hơn. - Mức bằng chứng: B -- Nguồn: Cục Quản lý Dược phẩm nhà nước (国家药监局) (2021). Thông báo về việc sửa đổi tờ hướng dẫn của 14 chế phẩm như dung dịch uống paracetamol và pseudoephedrine (2021年第57号) và phụ lục. ；Smith SM, Schroeder K, Fahey T (2014). Over-the-counter (OTC) medications for acute cough in children and adults in community settings. Cochrane Database of Systematic Reviews, (11):CD001831. -- Ghi chú: Trong 14 loại này, nhiều tên có chữ「氨酚」, đều chứa paracetamol, uống chung với thuốc hạ sốt là bị trùng, xem mục 1 của phần này (paracetamol). Trong bài tổng hợp đó có một thử nghiệm thấy mật ong tốt hơn giả dược, nhưng trẻ dưới 1 tuổi thì không cho uống mật ong, xem mục 6 ở phần 20 (mật ong). Để ở mức B là vì trong mỗi nhóm thuốc ở bài tổng hợp đều số thử nghiệm ít, tác giả không gộp được thành con số. Đặt lợi ích là「vừa」là theo phán đoán: nhóm thuốc này chưa có lợi ích nào được chứng minh ở trẻ, rủi ro chính là quá liều và trùng thành phần. +- Nguồn: Thông báo của Cục Quản lý Dược phẩm Quốc gia về việc sửa đổi hướng dẫn sử dụng cho 14 chế phẩm như 氨酚麻美口服溶液 (2021年第57号) kèm phụ lục. ；Smith SM, Schroeder K, Fahey T (2014). Over-the-counter (OTC) medications for acute cough in children and adults in community settings. Cochrane Database of Systematic Reviews, (11):CD001831. +- Ghi chú: Trong 14 loại này, nhiều tên có chữ「氨酚」, cũng chứa paracetamol, uống cùng thuốc hạ sốt thì thành phần bị lặp, xem mục 1 của phần này (paracetamol). Trong bài tổng quan đó có một thử nghiệm thấy mật ong tốt hơn giả dược, nhưng trẻ chưa đầy 1 tuổi không được cho ăn mật ong, xem mục 6 ở phần 20 (mật ong). Để ở mức B là vì mỗi nhóm thuốc trong tổng quan đều có ít thử nghiệm, tác giả không gộp được ra con số. Cỡ lợi ích đặt là「vừa」là đánh giá: chưa có bằng chứng lợi ích nào ở trẻ, rủi ro chủ yếu là quá liều và thành phần bị lặp. -### 5. Sau tuần thứ 20 mang thai, đừng tự uống thuốc giảm đau như ibuprofen +### 5. Sau tuần thứ 20 của thai kỳ, đừng tự uống thuốc giảm đau như ibuprofen - Chi phí: Không tốn tiền. -- Nói thẳng: Sau tuần thứ 20 mang thai, phần lớn nước ối do thận của thai nhi tạo ra. Lúc đó uống thuốc nhóm ibuprofen có thể làm hại thận thai nhi, làm nước ối ít đi. Ở Hoa Kỳ đã có báo cáo trẻ sơ sinh chết vì nguyên nhân này. Liều nhỏ aspirin do bác sĩ kê không nằm trong phạm vi này. -- Lợi ích: Ngày 10 tháng 10 năm 2020, Cục Quản lý Thực phẩm và Dược phẩm Hoa Kỳ (FDA) ban hành cảnh báo an toàn. Dùng thuốc chống viêm không steroid từ khoảng tuần thứ 20 mang thai trở đi có thể gây vấn đề về thận ở thai nhi, dẫn tới nước ối ít. Cảnh báo áp dụng cho cả thuốc có đơn lẫn thuốc không cần đơn, gồm aspirin, ibuprofen, naproxen, diclofenac, celecoxib. FDA rà soát báo cáo biến chứng đến tháng 7 năm 2017, tìm được 35 ca nước ối ít hoặc suy thận ở trẻ sơ sinh, toàn bộ đều là kết cục nặng. Trong đó 5 ca trẻ sơ sinh chết, đều kèm suy thận ở trẻ sơ sinh. Trong tài liệu, phần lớn các ca nước ối ít hồi phục trong vòng 72 giờ đến 6 ngày sau khi ngừng thuốc. FDA khuyên phụ nữ mang thai từ tuần thứ 20 trở đi không dùng nhóm thuốc này, trừ khi bác sĩ khuyên dùng rõ ràng. Liều nhỏ 81 mg aspirin do bác sĩ kê để điều trị vấn đề riêng trong thai kỳ thì không chịu hạn chế này. Khi cần giảm đau hạ sốt, FDA nhắc có thể dùng paracetamol hoặc thuốc khác, hãy hỏi dược sĩ hoặc bác sĩ trước. +- Nói thẳng: Sau tuần thứ 20 của thai kỳ, phần lớn nước ối do thận của thai nhi tạo ra. Lúc đó uống ibuprofen và nhóm thuốc này có thể làm tổn thương thận thai nhi và làm nước ối giảm. Trong số báo cáo cùng loại mà Hoa Kỳ nhận được có ca trẻ sơ sinh chết. Liều aspirin thấp do bác sĩ kê không nằm trong phạm vi này. +- Lợi ích: Tháng 10 năm 2020 Cục Quản lý Thực phẩm và Dược phẩm Hoa Kỳ (FDA) phát cảnh báo an toàn. Dùng thuốc chống viêm không steroid sau khoảng tuần thứ 20 của thai kỳ có thể gây vấn đề về thận ở thai nhi, dẫn tới thiếu nước ối. Cảnh báo áp dụng cho cả thuốc kê đơn và thuốc không kê đơn, gồm aspirin, ibuprofen, naproxen, diclofenac, celecoxib. FDA rà soát báo cáo bất lợi đến tháng 7 năm 2017, tìm thấy 35 ca thiếu nước ối hoặc suy thận ở trẻ sơ sinh, toàn bộ đều là kết cục nặng. Trong đó 5 ca trẻ sơ sinh chết, đều kèm suy thận ở trẻ sơ sinh. Trong tài liệu, phần lớn các ca thiếu nước ối hồi phục trong vòng 72 giờ đến 6 ngày sau khi ngừng thuốc. FDA khuyến nghị phụ nữ mang thai không dùng nhóm thuốc này sau tuần thứ 20, trừ khi bác sĩ khuyên rõ ràng như vậy. Liều aspirin 81 mg do bác sĩ kê để điều trị một số vấn đề riêng trong thai kỳ không bị giới hạn này. Khi cần giảm đau và hạ sốt, FDA nhắc có thể dùng paracetamol hoặc thuốc khác, hãy hỏi dược sĩ hoặc bác sĩ trước. - Mức bằng chứng: B - Nguồn: U.S. Food and Drug Administration (2020-10-15). FDA recommends avoiding use of NSAIDs in pregnancy at 20 weeks or later because they can result in low amniotic fluid. Drug Safety Communication. -- Ghi chú: FDA nhắc, rất nhiều thuốc không cần đơn trị cảm cúm, cúm, mất ngủ cũng chứa các thành phần thuộc nhóm này, phải xem bảng thành phần. Nhãn thuốc không cần đơn ở Hoa Kỳ trước đây chỉ cảnh báo ba tháng cuối thai kỳ, lần này kéo mốc lên khoảng tuần thứ 20. Liều nhỏ aspirin phòng sản giật thai kỳ, xem mục 5 ở phần 27 (liều nhỏ aspirin). Để ở mức B là vì số liệu đến từ báo cáo biến chứng, không biết có bao nhiêu người uống thuốc mà không sao. Đặt lợi ích là「vừa」là theo phán đoán: kết cục thì nặng, nhưng xảy ra ít. +- Ghi chú: FDA nhắc, nhiều thuốc không kê đơn điều trị cảm cúm, cúm, mất ngủ cũng chứa thành phần nhóm này, phải xem bảng thành phần. Nhãn thuốc không kê đơn ở Mỹ trước đây chỉ cảnh báo cho ba tháng cuối thai kỳ, lần này đưa mốc về sớm hơn khoảng tuần thứ 20. Aspirin liều thấp phòng tiền sản vô mạch, xem mục 5 ở phần 27 (aspirin liều thấp). Để ở mức B là vì số liệu đến từ báo cáo bất lợi, không biết có bao nhiêu người uống thuốc mà không sao. Cỡ lợi ích đặt là「vừa」là đánh giá: kết cục nặng, nhưng hiếm gặp. ### 6. Tự mua omeprazole khi bị trào ngược và nóng rát dạ dày, nhiều nhất 7 ngày; khó nuốt, nôn ra máu, phân đen thì đi thẳng bệnh viện From 469fa3f935b37ea5e4ee342aa4384d5f145aac77 Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 12:47:28 +0700 Subject: [PATCH 17/25] =?UTF-8?q?=E8=B6=8A=E5=8D=97=E6=96=87=E7=AC=AC=2006?= =?UTF-8?q?=20=E8=8A=82=EF=BC=88=E5=8F=8D=E9=9D=A2=E6=B8=85=E5=8D=95?= =?UTF-8?q?=EF=BC=89=E7=AD=89=E7=BF=BB=E5=AE=8C=E5=B9=B6=E5=90=88=E5=85=A5?= =?UTF-8?q?=EF=BC=9AVI=2030/34=EF=BC=9B=E8=8A=82=E6=A0=87=E9=A2=98?= =?UTF-8?q?=E8=A6=81=E6=94=B9=E6=BA=90=E7=A0=81=E9=82=A3=E4=B8=80=E4=BB=BD?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 两件事： 一是节标题只改 book/ 里的不够——join 是从 .i18n 的块 0 重生成的，只改 book/ 的话 下次 join 就被覆盖回去。2026-10-02 改完 en 32、vi 16/24/25，第二天 main 的 CI 就报 「节标题跟目录不一致」，本地当时是绿的（.i18n 已被 gitignore，改的那一份没进库）。 README 目录和 .i18n 块 0 两处都得改，这条记进 tools/i18n/README.md 了。 二是 vi 06 三处：节首导读还是中文；条 12 原文「1 小时后」的数字被写成「một giờ」， 条 15 原文「200 余万元」「10 万元」被写成「2 triệu」「100.000」——量级换算是允许的， 但原文没有量级词时就得照抄那三个数，否则收益栏可核对性断了。 --- tools/i18n/README.md | 5 + ...15\351\235\242\346\270\205\345\215\225.md" | 257 ++++++++++++++++++ ...16\345\201\232\347\224\237\346\204\217.md" | 48 ++-- ...16\346\200\216\344\271\210\346\264\273.md" | 2 +- ...14\345\222\214\345\267\245\344\274\244.md" | 2 +- ...00\350\203\275\345\210\222\347\256\227.md" | 209 ++++++++++++++ "vi/book/24-\347\234\213\347\227\205.md" | 2 +- ...01\345\212\236\344\273\200\344\271\210.md" | 2 +- ...52\345\207\240\346\235\241\350\267\257.md" | 49 ++-- ...53\345\220\203\345\207\272\344\272\213.md" | 46 ++-- 10 files changed, 547 insertions(+), 75 deletions(-) create mode 100644 "vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" create mode 100644 "vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" diff --git a/tools/i18n/README.md b/tools/i18n/README.md index a405bb95..d96f95c1 100644 --- a/tools/i18n/README.md +++ b/tools/i18n/README.md @@ -30,6 +30,11 @@ node tools/site/build.mjs `join` 就能看到还差哪几块。`join-ready.sh` 用的就是这个判据——一个块里还带 `- 成本：`， 就是这块还没翻。 +**节标题改哪一份**：两边都要改，`<语言>/README.md` 的目录里一份、`.i18n/<语言>/<节名>/part00.md` +的块 0 里一份。`book/` 那份是 join 从块 0 合出来的，改了它，下次 join 就被覆盖回去 +（2026-10-02 改完 book/ 里三节，第二天 CI 就报「节标题跟目录不一致」——就是只改了一半）。 +`check.mjs` 拿 README 目录里那份跟译本节标题对字，两边得说同一句话。 + **仓库里的路径一律走 `contentPath(code, rel)`（`tools/site/locales.mjs`），别用 `dir`。** `dir` 是页面上线的目录，英文那份是空串（页面在根 URL 上）；`contentDir` 才是正文在仓库里 的位置（英文是 `en`）。拿 `dir` 拼路径会拼到 `book/`——中文原文那一份，`join.mjs` 那样写 diff --git "a/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" "b/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" new file mode 100644 index 00000000..b85565ad --- /dev/null +++ "b/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" @@ -0,0 +1,257 @@ +[← Về mục lục](../README.md) + +# 6. Danh sách phản tác dụng + +Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở nghiên cứu gốc ra thì lợi ích của chúng gần bằng không, bằng không, có cái còn âm. Thứ tự xếp nhìn hai thứ: người ta đã bỏ bao nhiêu tiền vào đó, và bằng chứng đỡ nó có chắc không. Càng lên trước càng đáng ngừng trước. + +### 1. Đừng uống vitamin tổng hợp để kéo dài sống hoặc phòng bệnh tim mạch + +- Chi phí: Khoảng 0.5 đến 2 tệ mỗi ngày, 200 đến 700 tệ mỗi năm. Uống không tốn thời gian, chỉ một viên. Uống lâu dễ sinh phụ thuộc, thấy ngày nào không uống là như bỏ sót một việc. +- Nói thẳng: 14.000 bác sĩ nam uống vitamin tổng hợp trong mười một năm. Số người gặp sự cố nặng ở tim và mạch máu, và số người chết, bằng nhóm không uống. Gộp chín thử nghiệm, hơn 50.000 người, cũng không thấy ảnh hưởng nào tới tử vong. Không mua thì tiết kiệm được hai trăm đến bảy trăm tệ mỗi năm. Người thực sự bị thiếu một chất dinh dưỡng, và phụ nữ có thai bổ sung axit folic, không nằm trong mục này. +- Lợi ích: Thử nghiệm PHS II tuyển 14.641 bác sĩ nam, theo dõi 11.2 năm (đây là trung vị, một nửa dài hơn, một nửa ngắn hơn). Tỷ lệ gặp sự cố tim mạch ở nhóm uống và nhóm không uống bằng nhau. Sự cố tim mạch chính HR 1.01 (HR là tỷ số nguy cơ, 1 nghĩa là hai bên như nhau). 95% CI 0.91–1.10 là khoảng tin cậy của con số này. Tử vong cũng vậy, tử vong toàn nguyên HR 0.94 (0.88–1.02). Cơ quan chấm điểm biện pháp phòng ngừa của Hoa Kỳ USPSTF, năm 2022 gộp 9 thử nghiệm đối chứng chia nhóm ngẫu nhiên, 51.550 người, kết quả là multivitamin không liên quan gì tới tử vong toàn nguyên. Xếp hạng I, nghĩa là bằng chứng chưa đủ, không nói rõ được. Nó có vẻ đáng tiền vì nghe 「một viên bổ sung đủ mọi chất dinh dưỡng」 rất thuận tai, và một ngày chỉ tốn mấy chục xu +- Mức bằng chứng: A +- Nguồn: Sesso HD et al. (2012). Multivitamins in the prevention of cardiovascular disease in men: the Physicians' Health Study II randomized controlled trial. JAMA. ；US Preventive Services Task Force (2022). Vitamin, Mineral, and Multivitamin Supplementation to Prevent Cardiovascular Disease and Cancer. JAMA. ；Gaziano JM 等 (2012). Multivitamins in the Prevention of Cancer in Men. JAMA. （备注里那项癌症结果） +- Ghi chú:Tranh cãi: Trong cùng nhóm bác sĩ nam đó, nghiên cứu còn xem tới ung thư. Nhóm uống có tỷ lệ mắc ung thư thấp hơn khoảng 8% (0,86–0,998, P=0,04, vừa đủ vượt ngưỡng 「không giống ngẫu nhiên」). Tỷ lệ chết vì ung thư thấp hơn khoảng 12% (0,77–1,01). Khoảng này vượt qua 1, nên có thể chỉ là ngẫu nhiên. Lợi ích nhỏ này rất nhỏ, và chỉ thấy ở bác sĩ nam. Người thực sự bị thiếu chất dinh dưỡng, phụ nữ có thai (bổ sung axit folic), và người ăn quá thiếu lệch, không tính vào đây, cứ bổ sung theo lời bác sĩ + +### 2. Đừng uống dầu cá thông thường để phòng bệnh tim mạch + +- Chi phí: Khoảng 1 đến 3 tệ mỗi ngày, 400 đến 1.000 tệ mỗi năm. Uống không tốn thời gian. +- Nói thẳng: Hai thử nghiệm lớn lần lượt có 26.000 người và 15.000 người bị đái tháo đường. Người uống 1 gram dầu cá mỗi ngày, số gặp sự cố nặng ở tim mạch và số chết, đều bằng nhóm không uống. 「Ăn cá có lợi」 không bằng 「uống viên dầu cá có lợi」. Không mua thì tiết kiệm được bốn trăm đến một nghìn tệ mỗi năm. +- Lợi ích: Thử nghiệm VITAL có 25.871 người, uống 1 gram dầu cá mỗi ngày, theo dõi 5.3 năm (trung vị). Cả sự cố tim mạch lẫn tử vong, hai bên đều không thấy khác biệt. Sự cố tim mạch chính HR 0.92 (HR là tỷ số nguy cơ, 1 nghĩa là hai bên như nhau). 95% CI 0.80–1.06 là khoảng tin cậy của con số này. Nó vượt qua 1, nghĩa là không thấy khác biệt. Tử vong toàn nguyên HR 1.02 (0.90–1.15). Thử nghiệm ASCEND có 15.480 người bị đái tháo đường, cũng uống 1 gram mỗi ngày, trung bình theo dõi 7.4 năm, vẫn không thấy khác biệt. Sự cố mạch máu nghiêm trọng RR 0.97 (0.87–1.08). Tử vong toàn nguyên RR 0.95 (0.86–1.05). Nó có vẻ đáng tiền vì nhiều người coi thẳng 「ăn cá có lợi」 thành 「uống viên dầu cá có lợi」 +- Mức bằng chứng: A +- Nguồn: Manson JE et al. (2019). Marine n-3 Fatty Acids and Prevention of Cardiovascular Disease and Cancer. NEJM. ；ASCEND Study Collaborative Group (2018). Effects of n-3 Fatty Acid Supplements in Diabetes Mellitus. NEJM. ；Bhatt DL 等 (2019). Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia. New England Journal of Medicine. （备注里那项高纯度处方鱼油试验） +- Ghi chú:Tranh cãi: Có một thử nghiệm thực sự có hiệu quả. Thử nghiệm REDUCE-IT có 8.179 người vốn đã mắc bệnh tim mạch hoặc đái tháo đường, đều đang dùng statin, mỡ máu triglyceride vẫn cao (135–499 mg/dL). Họ uống mỗi ngày 4 g thuốc EPA tinh khiết, tỷ lệ gặp sự cố là 17,2% so với 22,0%. Nguy cơ thấp hơn khoảng một phần tư (HR 0,75, 0,68–0,83). Nhưng đó là thuốc kê đơn liều cao kê cho bệnh nhân nguy cơ cao, không giống người thường uống dầu cá dạng thực phẩm chức năng. Ở đây cũng không tính việc ăn cá. + +### 3. Đừng bổ sung vitamin D cho người không thiếu vitamin D chỉ để kéo dài sống + +- Chi phí: Khoảng 0.2 đến 0.5 tệ mỗi ngày, 100 đến 200 tệ mỗi năm. Số tiền không nhiều, cái giá thật là tưởng rằng mình đã phòng ngừa xong. +- Nói thẳng: Người vốn không thiếu vitamin D mà bổ sung trong 5 năm, số mắc ung thư, số mắc bệnh tim mạch và số chết, đều bằng nhóm không bổ sung. Người xét ra thực sự thiếu, người bị loãng xương, người lâu không hền được nắng, không nằm trong mục này, họ bổ sung theo lời bác sĩ. +- Lợi ích: Thử nghiệm VITAL có 25.871 người, uống 2000 IU mỗi ngày, theo dõi 5.3 năm (trung vị). Ung thư, bệnh tim mạch và tử vong đều không thấy khác biệt. Loại ung thư dễ di căn HR 0.96 (HR là tỷ số nguy cơ, 1 nghĩa là hai bên như nhau; 95% CI 0.88–1.06 là khoảng tin cậy). Sự cố tim mạch chính HR 0.97 (0.85–1.12). Tử vong toàn nguyên HR 0.99 (0.87–1.12). Thử nghiệm D-Health có 21.315 người Australia trên 60 tuổi. Họ uống 60.000 IU mỗi tháng, theo dõi 5 năm. Mục này cũng không thấy khác biệt, tử vong toàn nguyên HR 1.04 (0.93–1.18). Tỷ lệ tử vong là 5.3% so với 5.1%. Nó có vẻ đáng tiền vì có nghiên cứu chỉ theo dõi ghi nhận, không can thiệp gì, thấy người có mức vitamin D thấp chết nhiều hơn, điều đó bị hiểu nhầm thành 「bổ sung vào là có ích」 +- Mức bằng chứng: A +- Nguồn: Manson JE et al. (2019). Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease. NEJM. ；Neale RE et al. (2022). The D-Health Trial: a randomised controlled trial of the effect of vitamin D on mortality. Lancet Diabetes Endocrinol. +- Ghi chú:Tranh cãi: Cả hai thử nghiệm đều không cố ý tuyển người thực sự thiếu vitamin D, nên kết luận chỉ áp cho 「người vốn không thiếu」. Người xét ra thực sự thiếu, người bị loãng xương, người lâu không hền nắng, trẻ nhỏ, vẫn bổ sung theo lời bác sĩ. Ở đây chỉ xem tử vong toàn nguyên, ung thư và bệnh tim mạch, những thứ khác không xem + +### 4. Đừng uống thực phẩm bổ sung chống oxy hóa để phòng ung thư (β-carotene, vitamin E, vitamin A) + +- Chi phí: Khoảng 0.5 đến 2 tệ mỗi ngày. Với người hút thuốc, cái giá thật là nguy cơ ung thư phổi tăng theo. +- Nói thẳng: Mục này không phải chỉ vô ích. Gộp 78 thử nghiệm, gần 300.000 người, người uống thực phẩm bổ sung chống oxy hóa lại có tỷ lệ tử vong cao hơn một chút. Trong hai thử nghiệm lớn chỉ tuyển người hút thuốc, β-carotene làm tỷ lệ mắc ung thư phổi tăng 18% đến 28%, tử vong tăng 8% đến 17%. Người hút thuốc và người từng tiếp xúc với bụi amianhạt thì đặc biệt đừng uống. +- Lợi ích: Cochrane gộp 78 thử nghiệm đối chứng chia nhóm ngẫu nhiên, 296.707 người. Người uống thực phẩm bổ sung chống oxy hóa có tử vong toàn nguyên RR 1.02 (RR là tỷ số nguy cơ, 1 nghĩa là hai bên như nhau; 95% CI 0.98–1.05 là khoảng tin cậy). Chỉ xem nhóm thử nghiệm có thiết kế chặt hơn, ít lệch hơn, RR 1.04 (1.01–1.07), cao hơn một chút. Riêng β-carotene RR 1.05 (1.01–1.09), riêng vitamin E RR 1.03 (1.00–1.05). Thử nghiệm ATBC có 29.133 nam giới hút thuốc, uống 20 mg mỗi ngày, tỷ lệ mắc ung thư phổi cao hơn 18% (3%–36%), tử vong toàn nguyên cao hơn 8% (1%–16%). Thử nghiệm CARET có 18.314 người hút thuốc hoặc từng tiếp xúc với bụi amianhạt. Nguy cơ mắc ung thư phổi của họ cao hơn khoảng 28% (RR 1.28, 1.04–1.57). Nguy cơ tử vong cao hơn khoảng 17% (tử vong toàn nguyên RR 1.17, 1.03–1.33). Nó có vẻ đáng tiền vì chuỗi suy luận 「chống oxy hóa tức là chống lão hoá, chống lão hoá thì phòng được ung thư」, mỗi bước nghe ra đều như hiển nhiên +- Mức bằng chứng: A +- Nguồn: Bjelakovic G et al. (2012). Antioxidant supplements for prevention of mortality in healthy participants and patients with various diseases. Cochrane Database Syst Rev. ；The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group (1994). The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. NEJM. ；Omenn GS et al. (1996). Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. NEJM. +- Ghi chú: Cơ quan chính thức của Hoa Kỳ chấm điểm các biện pháp phòng ngừa USPSTF, năm 2022 xếp β-carotene và vitamin E ở mức D, nghĩa là khuyến nghị không dùng. Người hút thuốc và người từng tiếp xúc với bụi amianhạt đã chịu hại vì thế. Chất chống oxy hóa có sẵn trong rau củ quả không tính vào đây, cứ ăn. + +### 5. Đừng trông chờ glucosamine/chondroitin chữa viêm khớp gối + +- Chi phí: Khoảng 2 đến 5 tệ mỗi ngày, 700 đến 1.800 tệ mỗi năm. +- Nói thẳng: Người bị thoái hóa khớp gối uống thuốc giả 24 tuần, cũng có sáu phần trăm nói đau ít hơn. Người uống glucosamine hoặc chondroitin chỉ cao hơn con số đó bốn năm điểm phần trăm, nhỏ đến mức có thể chỉ là ngẫu nhiên. Một năm bảy trăm đến một nghìn tám trăm tệ, thứ bạn mua về cơ bản chỉ là tác dụng tâm lý của sáu phần trăm đó. +- Lợi ích: Thử nghiệm GAIT có 1.583 người bị thoái hóa khớp gối, uống trong 24 tuần. Trong nhóm uống thuốc an toàn (thuốc giả), có 60.1% nói cơn đau giảm. Glucosamine chỉ cao hơn 3.9 điểm phần trăm (P=0.30). Chondroitin cao hơn 5.3 điểm phần trăm (P=0.17). Dùng cả hai cao hơn 6.5 điểm phần trăm (P=0.09). Mấy chênh lệch này đều quá nhỏ, không loại trừ được khả năng chỉ là ngẫu nhiên. Thuốc thật dùng làm đối chứng trong cùng thử nghiệm, celecoxib, cao hơn 10.0 điểm phần trăm (P=0.008). Nó có vẻ đáng tiền vì nghe 「bổ sung nguyên liệu làm sụn」 đúng bệnh, và chỉ uống thuốc giả đã có sáu phần trăm nói hiệu quả +- Mức bằng chứng: A +- Nguồn: Clegg DO et al. (2006). Glucosamine, chondroitin sulfate, and the two in combination for painful knee osteoarthritis. NEJM. +- Ghi chú:Tranh cãi: Trong cùng thử nghiệm đó, ở nhóm lúc đầu đã đau nặng, tỷ lệ hiệu quả khi dùng cả hai là 79,2%, nhóm đối chứng là 54,3% (P=0,002). Nhưng tác giả nói đây là thứ lục ra từ dữ liệu sau khi thử nghiệm kết thúc, chỉ có thể coi là đầu mối, không phải kết luận. Cách làm thực sự có bằng chứng cho viêm khớp gối là giảm cân và tập cơ, xem phần 2 + +### 6. Đừng dựa vào vitamin C để phòng cảm cúm + +- Chi phí: Một viên effervescent khoảng 1 đến 2 tệ, 300 đến 700 tệ mỗi năm. +- Nói thẳng: Uống vitamin C hằng ngày về cơ bản không làm bạn ít bị cảm cúm hơn. Bắt đầu uống khi đã bị cảm thì tác dụng lúc có lúc không. Uống hằng ngày rút ngắn một lần cảm cúm khoảng 8%, tức khó chịu ít hơn nửa ngày, đáng hay không với ba trăm đến năm trăm tệ một năm thì bạn tự tính. Vận độn viên marathon và lính hành quân ở vùng khí lạnh là ngoại lệ, ở họ số người uống rồi bị cảm giảm được một nửa. +- Lợi ích: Cochrane gộp 29 thử nghiệm đối chứng, 11.306 người. Người thường uống vitamin C mỗi ngày, gần như không giảm số lần bị cảm. Tỷ lệ bị cảm RR 0.97 (RR là tỷ số nguy cơ, 1 nghĩa là hai bên như nhau). 95% CI 0.94–1.00 là khoảng tin cậy. Với các thử nghiệm bắt đầu uống sau khi đã bị cảm, nguyên văn viết 「chưa thấy tác dụng nhất quán」: có thử nghiệm có hiệu quả, có thử nghiệm không. Trong chính tổng quan đó, uống dài ngày rút ngắn thời gian khó chịu khoảng 8% (người lớn, 95% CI 3%–12%), trẻ em 14% (7%–21%). Vận độn viên marathon, người trượt tuyết và lính hành quân ở vùng khí lạnh là ngoại lệ. Trong 5 thử nghiệm, 598 người, số người bị cảm giảm khoảng một nửa. Tỷ lệ mắc RR 0.48 (95% CI 0.35–0.64). Nó có vẻ đáng tiền vì rẻ, chua ngọt dễ uống, và mấy chục năm nay câu 「vitamin C phòng cảm cúm」 đã lan truyền khắp nơi +- Mức bằng chứng: A +- Nguồn: Hemilä H, Chalker E (2013). Vitamin C for preventing and treating the common cold. Cochrane Database Syst Rev. +- Ghi chú:Tranh cãi: Trong chính tổng quan đó, uống hằng ngày quả thật rút ngắn thời gian khó chịu khi bị cảm. Người lớn ngắn hơn khoảng 8% (3%–12%), trẻ em ngắn hơn 14% (7%–21%). Tính ra mỗi lần cảm bệnh ít hơn khoảng nửa ngày. Vận độn viên marathon, người trượt tuyết, lính hành quân ở vùng khí lạnh là ngoại lệ. Trong 5 thử nghiệm, 598 người, số người bị cảm giảm khoảng một nửa (RR 0.48, 0.35–0.64). Đa số không thuộc hai nhóm này + +### 7. Đừng tự làm 「PET-CT toàn thân」 hay 「gói xét nghiệm chất chỉ dấu ung thư」 khi không có triệu chứng + +- Chi phí: Một lần PET-CT toàn thân khoảng 7000 đến 10000 tệ, cộng thêm một lần nhiễm xạ. Gói chất chỉ dấu ung thư vài trăm tệ. Cái giá thật nằm ở sau đó: nếu ra một kết quả 「bất thường」 thật ra chỉ là hốt, gọi là dương tính giả, phần tiếp theo là chụp lại, lấy mẫu mô xét nghiệm, mổ, tiền và sự khổ nằm hết ở đoạn đó. +- Nói thẳng: Người không có triệu chứng đi khám, thứ nhiều nhất tìm ra chính là chuyện hốt. Trong số người làm CT toàn thân, hơn tám phần trăm ít nhất ra một kết quả 「bất thường」, gần bốn phần trăm bị gọi quay lại chụp lại, nhưng đa số là tổn thương lành tính. Phụ nữ xét chất chỉ dấu ung thư buồng trứng, số chết vì ung thư buồng trứng lại cao hơn một chút, còn có người vì hốt mà phải mổ oan một lần. +- Lợi ích: Cơ quan chấm điểm biện pháp phòng ngừa của Hoa Kỳ USPSTF xếp mức D cho việc dùng CA-125 và siêu âm sàng lọc ung thư buồng trứng ở phụ nữ không có triệu chứng, nghĩa là khuyến nghị không làm. Căn cứ là thử nghiệm PLCO. Tỷ lệ chết vì ung thư buồng trứng, nhóm sàng lọc 0.34%, nhóm thường 0.29%. RR 1.18 (95% CI 0.82–1.71). Xét ra không có lợi ích. Tỷ lệ phải mổ oan vì chẩn đoán sai (dương tính giả) trong các thử nghiệm là 0.2%–3.25%, trong số đó tối đa 15% gặp biến chứng phẫu thuật nặng. Một nghiên cứu khác làm CT toàn thân cho 1.192 người không có triệu chứng. Trong đó 86% ít nhất ra một kết quả 「bất thường」, 37% được khuyên xét tiếp. Nguyên văn viết 「phần lớn các phát hiện được mô tả là lành tính」. Nó có vẻ đáng tiền vì 「một lần quét cả người, phát hiện sớm chữa sớm」 quá đúng trực giác +- Mức bằng chứng: A +- Nguồn: US Preventive Services Task Force (2018). Screening for Ovarian Cancer: US Preventive Services Task Force Recommendation Statement. JAMA（cùng khuyến nghị đó còn có trang chính thức của USPSTF）. 、；Furtado CD et al. (2005). Whole-body CT screening: spectrum of findings and recommendations in 1192 patients. Radiology. +- Ghi chú: Phần chất chỉ dấu ung thư có thử nghiệm đối chứng ngẫu nhiên đứng sau. Phần CT toàn thân và PET-CT chỉ có loại theo dõi 「xem người đã làm thì sau đó ra sao」, không có nhóm đối chứng, nên bằng chứng yếu hơn một bậc, tương đương mức B. Vài sàng lọc thật sự có bằng chứng, như ung thư ruột giàu, ung thư cổ tử cung, ung thư vú, và người nguy cơ cao làm CT ngực liều thấp, là chuyện khác, xem phần 1. Người đã có triệu chứng, hoặc người đã xác định ung thư mà làm PET-CT, đó là khám bệnh chứ không phải sàng lọc, không so được + +### 8. Đừng trông chờ vòng tay hay đồng hồ đo chuyển động giúp bạn giảm cân + +- Chi phí: Thiết bị 200 đến 2000 tệ. Ngoài ra mỗi ngày còn phải phân tâm nhìn mấy con số đó. +- Nói thẳng: Người thừa cân làm cùng chế độ ăn và vận động, người đeo vòng tay giảm 3.5 kilo trong hai năm, còn người không đeo lại giảm 5.9 kilo. Muốn giảm cân, tiền mua thiết bị không giúp được gì, cái thật sự có tác dụng là bản thân việc vận động. +- Lợi ích: Thử nghiệm IDEA có 471 người lớn thừa cân hoặc béo phì, theo dõi 24 tháng. Mọi người làm cùng chế độ ăn và vận động, khác nhau chỉ ở chỗ có đeo thiết bị đeo được hay không. Nhóm đeo giảm 3.5 kilo (95% CI 2.6–4.5, đây là khoảng tin cậy). Nhóm không đeo giảm 5.9 kilo (5.0–6.8), nhiều hơn. Hai nhóm chênh 2.4 kilo (1.0–3.7, P=0.002). Nó có vẻ đáng tiền vì câu 「phải đo mới quản lý được」 nghe rất có lý, và thiết bị chỉ phải mua một lần +- Mức bằng chứng: B +- Nguồn: Jakicic JM et al. (2016). Effect of Wearable Technology Combined With a Lifestyle Intervention on Long-term Weight Loss: The IDEA Randomized Clinical Trial. JAMA. +- Ghi chú: Ở đây chỉ có một thử nghiệm, lại dùng thiết bị đeo ở cánh tay kiểu đầu thập niên 2010, nên không thể nói 「mọi thiết bị đeo đều vô ích」. Bản thân việc vận động có hiệu quả, xem phần 2. Kết luận chỉ một câu: muốn giảm cân, thiết bị không phải khoản tiền bắt buộc phải chi + +### 9. Đừng trả tiền bản giá cao hơn để mua thực phẩm hữu cơ vì 「khỏe hơn」 + +- Chi phí: Thường là 1 đến 3 lần giá thực phẩm thường cùng loại. +- Nói thẳng: Không có bằng chứng đáng tin nào cho rằng thực phẩm hữu cơ giàu dinh dưỡng hơn. Nghiên cứu thật sự xem sức khỏe sau khi người ta ăn chỉ có 3 nghiên cứu, đều không thấy khác biệt. Tỷ lệ phát hiện dư lượng thuốc trừ sâu trên nông sản hữu cơ thấp hơn 30 điểm phần trăm, nhưng có phát hiện không có nghĩa là vượt ngưỡng. Ăn nhiều rau quả vốn đã có ích, có phải hữu cơ hay không không ảnh hưởng điều đó. +- Lợi ích: Có một tổng quan hệ thống gộp 17 nghiên cứu làm trên người và 223 nghiên cứu đo thành phần dinh dưỡng của thực phẩm. Kết luận là 「thiếu bằng chứng đáng tin cho thấy thực phẩm hữu cơ giàu dinh dưỡng hơn đáng kể so với thực phẩm thông thường」. Trong đó chỉ 3 nghiên cứu xem sức khỏe sau khi ăn (xem dị ứng và nhiễm trùng có triệu chứng), hai bên không thấy khác biệt. Tỷ lệ phát hiện dư lượng thuốc trừ sâu trên nông sản hữu cơ thấp hơn 30 điểm phần trăm, nhưng có phát hiện không có nghĩa là vượt ngưỡng. Nó có vẻ đáng tiền vì 「không có thuốc trừ sâu thì an toàn hơn, an toàn hơn thì khỏe hơn」, mỗi bước nghe ra đều như hiển nhiên +- Mức bằng chứng: B +- Nguồn: Smith-Spangler C et al. (2012). Are organic foods safer or healthier than conventional alternatives?: a systematic review. Ann Intern Med. +- Ghi chú: Nghiên cứu xem thẳng sức khỏe sau khi ăn rất ít. Vì vậy kết luận là 「không có bằng chứng nào cho thấy nó tốt hơn」, chứ không phải 「đã chứng minh nó không có ích」. Ảnh hưởng tới môi trường, phúc lợi động vật, khẩu vị, ở đây không bàn. Ăn nhiều rau quả vốn đã có ích, có phải hữu cơ hay không không ảnh hưởng điều đó + +### 10. Đừng bỏ tiền lớn mua thực phẩm chức năng, cao thuốc bắc, đồ bồi bổ để 「điều chỉnh cơ thể」 + +- Chi phí: Từ vài trăm tệ tới hàng chục nghìn tệ, tùy loại. Sau khi đóng thành hộp quà tặng, giá thường nhân lên mấy lần. +- Nói thẳng: Cơ quan quản lý thực phẩm chức năng quy định, trên bao bì phải in một câu: thực phẩm chức năng không phải là thuốc, không thay thế thuốc điều trị bệnh. Những thành phần phổ biến như đa vitamin, dầu cá, vitamin D, sáu mục đầu phần này đã lần lượt tra rồi, đều không thấy tác dụng. Khoản tiền này phần lớn là để mua cho cha mẹ, mang tới tận nhà. Tấm lòng đó không có gì đáng nói, chỉ là chọn nhầm món. +- Lợi ích: Cục Quản lý thị trường quốc gia quy định, mặt trước bao bì thực phẩm chức năng phải ghi bằng diện tích không nhỏ hơn 20% một câu: 「thực phẩm chức năng không phải là thuốc, không thay thế thuốc điều trị bệnh」. Cùng quy định đó còn khóa công dụng của thực phẩm chức năng trong phạm vi 「bổ sung chất dinh dưỡng trong khẩu phần, duy trì hoặc cải thiện trạng thái sức khỏe cơ thể, hoặc giảm yếu tố nguy cơ gây bệnh」. Chính cơ quan quản lý không công nhận nó chữa được bệnh. Vài thành phần phổ biến nhất trong đó, đa vitamin, dầu cá, vitamin D, chất chống oxy hóa, glucosamine, vitamin C, cho ra kết quả gì trong thử nghiệm đối chứng chia nhóm ngẫu nhiên, 6 mục đầu phần này đã liệt kê từng mục. Nó có vẻ đáng tiền vì câu 「không chữa bệnh, chỉ điều chỉnh」 đúng là nói gì cũng không sai, và không ai chứng minh được nó sai. Mà khi tặng quà mà bỏ số tiền này thì dựa vào tình cảm chứ không dựa vào bằng chứng +- Mức bằng chứng: C +- Nguồn: Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2019). 市场监管总局就《保健食品标注警示用语指南》和《保健食品原料目录与保健功能目录管理办法》有关情况举办专题新闻发布会. +- Ghi chú: Kết luận là 「không có bằng chứng ủng hộ」, chứ không phải 「đã chứng minh không có tác dụng」, nên chỉ cho mức C. Sản phẩm ghi rõ thành phần thì tra nghiên cứu tương ứng theo mấy mục trước. Người xét ra thực sự thiếu dinh dưỡng, người đang trị bệnh, cứ theo lời bác sĩ. Cái cần bỏ là sản phẩm, không phải tấm lòng. Người già tự mua và đang ăn, miễn là không dùng nó thay thuốc đang uống và không sinh vấn đề gì, thì không cần lên tận nhà cãi lại, việc sắp xếp vaccine và xét nghiệm sàng lọc cần làm cấp thiết hơn. Đổi số tiền đó sang những thứ sau, từng mục đều tra được trong sách. Đưa người già đi tiêm vaccine cúm hằng năm, xem mục 20 ở phần 1. Vaccine thủy đậu cho người trên 50 tuổi xem mục 21 ở phần 1. Vaccine phế cầu cho người trên 65 tuổi xem mục 22 ở phần 1. Mua một máy đo huyết áp, nhắc người đó uống đủ thuốc hạ huyết áp theo đơn và đưa huyết áp về mức mục tiêu, xem mục 7 ở phần 1. Sửa phòng tắm và cầu thang trong nhà, cùng tập thăng bằng và sức chân, xem mục 13 ở phần 1. Đến tuổi cần sàng lọc ung thư, đưa người đó đi làm một lần, xem mục 17 đến 19 ở phần 1. Người già đã nằm liệt giường lâu, phòng loét ép và bảo hiểm chăm sóc dài hạn xem mục 7 và 8 ở phần 17. Muốn mang quà tới nhà thì đổi sang trái cây, gạo, thứ người đó thích, thực tế hơn cao thuốc. Cái thật sự cần chặn là hai loại: một là đầu tư tiền vào chế độ dưỡng lão phải trả trước, hai là loại sẽ thay thế thuốc, xem mục 5 ở phần 17 + +### 11. Đừng uống men vi sinh dài ngày khi ruột không có vấn đề + +- Chi phí: Khoảng 2 đến 5 tệ mỗi ngày, 700 đến 1800 tệ mỗi năm. +- Nói thẳng: Gộp 45 thử nghiệm làm trên người lớn khỏe mạnh, uống men vi sinh không làm hệ vi sinh trong ruột thay đổi lâu dài, ngừng là trở về như cũ, mỡ máu cũng không cải thiện. Trường hợp cụ thể như kháng sinh gây tiêu chảy thì có bằng chứng riêng, không nằm trong mục này. +- Lợi ích: Một tổng quan gộp 45 thử nghiệm làm trên người lớn khỏe mạnh, kết luận là 「chưa chứng minh men vi sinh tạo ra thay đổi bền vững của hệ vi sinh ruột, cũng chưa cải thiện mỡ máu ở người lớn khỏe mạnh」. Vi khuẩn trong ruột quả thật có thay đổi, nhưng chỉ tạm thời, ngừng là trở lại. Nó có vẻ đáng tiền vì chữ 「sức khỏe đường ruột」 rộng quá, mặc vào đâu cũng không sai, và mỗi lọ hằng ngày đơn giá rất thấp +- Mức bằng chứng: B +- Nguồn: Khalesi S et al. (2019). A review of probiotic supplementation in healthy adults: helpful or hype? Eur J Clin Nutr. +- Ghi chú: Cùng tổng quan đó còn nói ở người khỏe mạnh có một vài cải thiện nhỏ, như phân thành hình bài hơn, mềm vừa phải, và vi khuẩn lactobacillus trong âm đạo nữ tăng lên. Tiêu chảy do kháng sinh, cũng như một chủng vi khuẩn cụ thể dùng cho một bệnh cụ thể, mỗi cái có bằng chứng riêng, không so được. Ở đây chỉ nói về người khỏe mạnh uống hằng ngày như thực phẩm chức năng + +### 12. Đừng ép mình tắm nước lạnh để 「tăng cường miễn dịch」 + +- Chi phí: Không tốn tiền. Mỗi ngày phải chịu khó chịu 30 đến 90 giây, muốn duy trì vào mùa đông thì cần ý chí. Người có bệnh tim mạch còn phải thận trọng, kích thích lạnh có rủi ro với họ. +- Nói thẳng: Trong thử nghiệm 3.018 người, người tắm nước lạnh xin nghỉ việc ít đi 29%, nhưng tổng số ngày nghỉ thì bằng nhóm khác, nên không nói là ít bệnh hơn. Năm 2025 gộp 11 thử nghiệm lại, sau khi xả nước lạnh, chỉ số viêm trong cơ thể tăng lên trong thời gian ngắn, chỉ số miễn dịch không thay đổi. Thích tắm thì cứ tắm, chỉ đừng coi nó là một khoản đầu tư sức khỏe. +- Lợi ích: Bên Hà Lan có một thử nghiệm đối chứng chia nhóm ngẫu nhiên, 3.018 người, tắm nước nóng xong rồi xả nước lạnh 30 giây, 60 giây hoặc 90 giây. Số lần xin nghỉ việc của họ giảm 29% (IRR 0.71, đây là tỷ số số lần). Nhưng nguyên văn ghi 「số ngày nghỉ việc không khác biệt có ý nghĩa giữa các nhóm」: tổng số ngày bị bệnh, hai bên bằng nhau. Chất lượng cuộc sống, hiệu suất làm việc, lo âu cũng không thấy khác biệt có ý nghĩa. Năm 2025 có một tổng quan hệ thống gộp 11 thử nghiệm chia nhóm ngẫu nhiên, 3.177 người. Ngay sau khi xả nước lạnh và 1 giờ sau đó, chỉ số viêm trong cơ thể tăng lên. Chỉ số miễn dịch không thay đổi. Mức căng thẳng chỉ giảm ở mốc 12 giờ. Tác giả tự nói bằng chứng 「bị giới hạn bởi số thử nghiệm chia nhóm ngẫu nhiên ít và cỡ mẫu nhỏ」, nghĩa là loại thử nghiệm này quá ít và mỗi thử nghiệm cũng ít người. Nó có vẻ đáng tiền vì không tốn tiền, và trực giác 「chịu khổ là có hiệu quả」 rất thuận +- Mức bằng chứng: B +- Nguồn: Buijze GA et al. (2016). The Effect of Cold Showering on Health and Work: A Randomized Controlled Trial. PLOS ONE. ；Cain T et al. (2025). Effects of cold-water immersion on health and wellbeing: A systematic review and meta-analysis. PLOS ONE. +- Ghi chú:Tranh cãi: Số lần xin nghỉ việc quả thật giảm 29%, chênh lệch này không giống ngẫu nhiên. Nhưng tổng số ngày bị bệnh không giảm, nên giống 「có bệnh vẫn đi làm」 hơn là thật sự ít bệnh. Vận độn viên dùng nước lạnh để hồi phục sau buổi tập, là chuyện khác, ở đây không bàn. Tắm nước lạnh bản thân không gây hại cho người khỏe mạnh, chỉ đừng coi nó là một khoản đầu tư sức khỏe + +### 13. Đừng mua các sản phẩm giải độc, thụt ruột, enzyme, nước kiềm + +- Chi phí: Vài trăm tới vài nghìn tệ. Loại dùng nhịn đói để 「giải độc」 còn có thể làm đường huyết tụt và thiếu dinh dưỡng. +- Nói thẳng: Ngành giải độc làm ăn rất lớn, nhưng bằng chứng lâm sàng đứng sau rất ít, và không ai làm thử nghiệm nghiêm túc với các chế độ ăn giải độc bán ngoài thị trường. Có người đi tra xem chế độ ăn kiềm tính, nước kiềm có liên quan gì tới ung thư, chỉ tìm được 1 nghiên cứu đạt chuẩn, và cũng không thấy liên quan. +- Lợi ích: Kết luận của tổng quan đó là 「mặc dù ngành giải độc phát triển rầm rộ, bằng chứng lâm sàng ủng hộ những chế độ ăn này rất ít」, và 「không có thử nghiệm đối chứng ngẫu nhiên nào cho các chế độ ăn giải độc thương mại」. Một tổng quan hệ thống khác chuyên tra chế độ ăn kiềm tính, nước kiềm và ung thư: rà 8.278 tài liệu, chỉ 1 tài đủ chuẩn đưa vào, kết quả cũng không thấy liên quan. Tác giả viết 「việc khuyến khích công chúng dùng chế độ ăn kiềm và nước kiềm để phòng hoặc điều trị ung thư là không hợp lý」. Nó có vẻ đáng tiền vì hai tiền đề 「trong cơ thể có độc tố cần đào ra」 và 「tính chất toan kiềm sẽ gây ung thư」 nghe rất giống sinh lý học +- Mức bằng chứng: C +- Nguồn: Klein AV, Kiat H (2015). Detox diets for toxin elimination and weight management: a critical review of the evidence. J Hum Nutr Diet. ；Fenton TR, Huang T (2016). Systematic review of the association between dietary acid load, alkaline water and cancer. BMJ Open. +- Ghi chú: Chỉ có thể nói 「không có bằng chứng ủng hộ」, nên chỉ cho mức C. Chất thải chuyển hóa sinh ra trong cơ thể vốn do gan và thận xử lý. Chỉ cần hai cơ quan này không bệnh thì không cần làm thêm điều gì để 「giải độc」 + +### 14. Đừng ép mình uống nước chỉ để đủ 「8 ly nước mỗi ngày」 + +- Chi phí: Không tốn tiền. Tốn một chút thời gian và sức lực để nhớ uống, ban đêm cũng phải đi tiểu nhiều hơn. +- Nói thẳng: Có người chuyên truy nguồn gốc của câu 「8 ly nước mỗi ngày」, không tìm được nghiên cứu nào ủng hộ nó. Thức ăn và các loại đồ uống khác vốn đã mang theo lượng nước lớn, nên không cần uống nhiều như vậy. Khát thì uống, nhìn màu nước tiểu là đủ. Người làm việc nắng nóng, vận động mạnh, từng bị sỏi thận, và người già không thấy khát nữa thì cần chủ động uống. +- Lợi ích: Có một bài tổng quan tài liệu chuyên tra câu 「8×8」, tức mỗi ngày 8 ly, mỗi ly 8 once nước. Kết luận của nó là 「không tìm thấy bất kỳ nghiên cứu khoa học nào ủng hộ 8×8」. Nó còn viết rằng 「khảo sát mức ăn uống của hàng nghìn người lớn cho thấy không cần lượng nước lớn như vậy」, vì thức ăn và đồ uống khác đã cung cấp phần lớn nước. Câu này có vẻ đáng tiền vì nước không mất tiền, con số 8 thì tròn trặn, mà 「uống nhiều nước」 thì nói ở đâu cũng không thấy sai +- Mức bằng chứng: C +- Nguồn: Valtin H (2002). "Drink at least eight glasses of water a day." Really? Is there scientific evidence for "8 x 8"? Am J Physiol Regul Integr Comp Physiol. +- Ghi chú: Cái cần bỏ là con số 「mỗi ngày phải đủ mấy ly」, không phải nói uống nhiều nước là có hại. Khát thì uống, nhìn màu nước tiểu, thế là đủ. Người làm việc trong nhiệt độ cao, người vận động mạnh, người từng bị sỏi thận, và người lớn tuổi không thấy khát thì cần chủ động uống, không so được + +### 15. Đừng bỏ tiền xem bói, xem tarot, xem chiêm tinh rồi ra quyết định + +- Chi phí: Không tốn tiền. Chỗ khó là từ bỏ một lối tắt để ra quyết định, sau đó phải tự nghĩ. +- Nói thẳng: Có một thí nghiệm lớp học kinh điển: bản 「phân tích tính cách cá nhân」 học sinh nhận được thực ra ai cũng giống nhau, nhưng ai cũng thấy rất chính xác. Xem bói trông chính xác là vì thế. Số lần nhà chiêm tinh chọn trúng chỉ ngang đoán mò. Cái giá không chỉ mấy chục tệ: một nhóm xem bói ở huyện Tư, tỉnh Hà Nam, thu tiền hương khói và 「gói hỗ trợ」 cao nhất 3888 tệ, kẻ chủ mưu bị án mười một năm. +- Lợi ích: Thí nghiệm lớp học kinh điển: bản 「phân tích cá nhân」 mà học sinh nhận sau khi làm bài trắc nghiệm tính cách thật ra giống nhau ở mọi người, thế mà ai cũng thấy chính xác. Kết luận của tác giả là 「việc người được phân tích hoặc người phân tích chấp nhận không chứng minh được lời giải thích là đúng」. 39 học sinh chấm độ chính xác của bản phân tích này từ 0 đến 5 điểm, 34 người chấm 4 hoặc 5 điểm, tính theo phân bố trong bảng thì trung bình khoảng 4.26 điểm. Điều đó giải thích vì sao xem bói lúc nào cũng trông chính xác, hiện tượng này gọi là hiệu ứng Barnum. Chiêm tinh cũng từng qua kiểm tra mù đôi, kết quả đăng trên Nature. Nhà chiêm tinh phải chọn từ ba kết quả trắc nghiệm tính cách bản nào khớp với lá số. Trong 116 bản, họ chọn trúng 40 bản, đoán mò thuần cũng trúng được 38.5 bản, còn nhà chiêm tinh tự dự trước sẽ trúng 58.5 bản. Vụ án chính thức: Viện Kiểm sát nhân dân huyện Tư, tỉnh Hà Nam đã thông báo một nhóm lừa đảo xem bói trên mạng, trước hết dùng xem tướng tay miễn phí để kéo người vào, rồi thu 「tiền hương khói」 và 「gói hỗ trợ」 cao nhất 3888 tệ. Hơn 50 nạn nhân, thu hồi được hơn 200 vạn tệ tiền chiếm đoạt, kẻ chủ mưu bị án mười một năm tù và phạt 10 vạn tệ tiền phạt +- Mức bằng chứng: B +- Nguồn: Forer BR (1949). The fallacy of personal validation: A classroom demonstration of gullibility. Journal of Abnormal and Social Psychology 44(1):118-123（表 1 评分分布：2 分 1 人、3 分 4 人、4 分 18 人、5 分 16 人，N=39）. ；Carlson S (1985). A double-blind test of astrology. Nature 318:419-425（表 2：首选 116 份，挑中 40 份，随机期望 38.5，占星师预计 58.5；「The data are consistent with chance, inconsistent with astrological hypothesis」）. ，第二作者主页全文 ；最高人民检察院 (2026). 河南息县：历经三年持续追踪打掉一批网络算命诈骗团伙. +- Ghi chú: Cả hai bài đều chỉ là một nghiên cứu riêng lẻ. Bài của Forer không ghi thẳng điểm trung bình, 4.26 là tính từ phân bố trong bảng 1 của nguyên văn. Ở đây chỉ nói 「nó dự đoán không chính xác」, không đánh giá tín ngưỡng nào. Bỏ ra mấy chục tệ để giải trí thì không tính, dùng nó để quyết định nghỉ việc, kết hôn, đầu tư mới là chỗ lỗ hơn tiền. + +### 16. Đừng mua kính chống ánh sáng xanh để 「bảo vệ thị lực」, cũng đừng tin 「nhìn màn hình vài tháng là mắt hỏng」, nhưng mắt sưng đau đỏ thì coi là cấp cứu + +- Chi phí: Không tốn tiền, còn tiết kiệm được phần giá hơn vài trăm tới hơn nghìn tệ gắn thêm trên mặt kính. +- Nói thẳng: Kính lọc ánh sáng xanh không có tác dụng, không mua thì tiết kiệm được vài trăm tới hơn nghìn tệ tiền chênh lệch. So với kính thường, người đeo có mắt mỏi hay không, nhìn được bao xa cũng không khác, và không tìm được bằng chứng nào nó bảo vệ đáy mắt. Nhìn màn hình lâu thấy mắt rát và khô thì khỏi lại được, thứ thật sự gây hại mắt là ánh hàn hồ, tia cực tím, laser, và cơn glôcôm cấp nặng. +- Lợi ích: Tổng quan hệ thống Cochrane năm 2023 gom 17 thử nghiệm đối chứng chia nhóm ngẫu nhiên, mỗi thử nghiệm 5 đến 156 người, thời gian theo dõi từ chưa tới 1 ngày tới 5 tuần. Vì dữ liệu quá ít, quy mô mỗi thử nghiệm khác nhau, thời gian theo dõi cũng không thống nhất, tác giả không gộp chúng thành một tổng số. Mọi kết quả dưới đây so với kính không lọc ánh sáng xanh. Khi theo dõi chưa tới một tuần, mức mỏi mắt người bệnh tự báo 「có thể không khác biệt」, độ chắc của bằng chứng thấp. Trong một thử nghiệm 120 người, hai nhóm chênh 9.76 đơn vị. Khoảng tin cậy của chênh lệch này là 95% CI −33.95 đến 53.47. Khoảng đó vượt qua cả âm lẫn dương, nghĩa là không thấy khác biệt. Tần số thấp nhấp nháy ranh giới là một chỉ số khách quan đo mức mỏi mắt, hai nhóm 「có thể gần như không khác biệt」, độ chắc thấp. Thị lực tốt nhất sau khi đeo kính có kính 「chắc là không ảnh hưởng」. Hai nhóm chênh 0.00 logMAR (95% CI −0.02 đến 0.02). Mục này chỉ có 1 nghiên cứu, 156 người, độ chắc trung bình. Ảnh hưởng tới độ tỉnh táo ban ngày thì không nói rõ được (2 thử nghiệm, 42 người, độ chắc rất thấp). Bài tổng quan này cũng không tìm được bằng chứng ủng hộ việc nó bảo vệ vùng hoàng thể mắt +- Mức bằng chứng: A +- Nguồn: Singh S, Downie LE, Anderson AJ, Keller PR, White SJ, Ang M, Wolffsohn JS (2023). Blue-light filtering spectacle lenses for visual performance, sleep, and macular health in adults. Cochrane Database of Systematic Reviews, 8, CD013244. +- Ghi chú: Nhìn màn hình gây mỏi mắt và khô mắt, biểu hiện là mắt rát, mắt khô, nhìn lâu thì mờ. Nghỉ một lát, chớm mắt nhiều hơn là khỏi. Nó có hồi phục được, không có nghĩa là 「mắt đã hỏng」. Câu này là đồng thuận của ngành mắt, không phải kết luận của bài tổng quan Cochrane trong cột nguồn, nên chỉ tính là mức C. Thứ thật sự gây tổn thương mắt không hồi phục được trong thời gian ngắn là nguồn sáng mạnh như ánh hàn hồ, tia cực tím, laser. Mắt bị ánh hàn hồ làm tổn thương gọi là 「viêm mắt do ánh sáng điện」, là bệnh nghề nghiệp theo luật. Cách chặn là kính bảo hộ và mặt nạ, không phải kính chống ánh sáng xanh, xem mục 11 ở phần 19 (phòng chống bụi, tiếng ồn, chất độc hoá học). Thức khuya dậy nhìn màn hình thì cái cần phòng không phải bản thân màn hình, mà có một ngoại lệ: mắt sưng đau đỏ, nhìn đèn có một vòng cầu vồng, kèm đau đầu, buồn nôn, nôn ói, đó là cơn glôcôm góc kín cấp tính, vài ngày là có thể ép hỏng dây thị giác, ngày đó phải đi cấp cứng mắt, xem mục 6 ở phần 13 (cấp cứu mắt). Ở chỗ tối đồng tử giãn, cúi đầu lâu đẩy thể thủy tinh về phía trước, hai việc đó làm nên thể bệnh. Người có nguy cơ cao là trên 50 tuổi, viễn thị, tiền đình nông. Khô mắt do màn hình kéo dài sẽ thành bệnh mạn tính phải đi khám, không phải ngủ một giấc là khỏi. Mắt cứ khó chịu, hoặc thị lực thật sự giảm, hãy đi khám mắt tìm nguyên nhân, đừng tự đổi mặt kính là xong. Trẻ em và thanh niên phòng cận thị thế nào, xem mục 4, 12 và 9 ở phần 30 (mỗi ngày 2 giờ ngoài trời, khúc xạ sau khi nhỏ mắt, không mua sản phẩm 「chữa khỏi cận thị」). + +### 17. Đừng coi hiến máu là cách giữ sức khoẻ: 「giải độc, hạ mỡ máu, phòng nhồi máu cơ tim」 không có thử nghiệm chia nhóm ngẫu nhiên nào đỡ, hiến thường xuyên thì đúng là dễ thiếu sắt + +- Chi phí: Không tốn tiền. Một lần hiến máu, gồm điền giấy tờ, lấy máu và ngồi lại quan sát nghỉ ngơi, mất khoảng một giờ. Mấy tuần sau đó cơ thể vẫn đang bù lại lượng sắt đã mất. +- Nói thẳng: Nói hiến máu giải độc, hạ mỡ máu, phòng nhồi máu cơ tim thì không đáng tin. Người hiến máu đều đặn vốn đã khoẻ hơn. Bỏ yếu tố đó đi, mức giảm tỷ lệ tử vong do hiến thêm một lần mỗi năm còn 7.5%, thay vì gần hai phần mươi. Trong thử nghiệm thật sự chia người ra so sánh, số người chết không khác nhau. Ngược lại, hiến quá thường xuyên thì đúng là khiến người ta mệt và sợ lạnh, và nguyên nhân là thiếu sắt. +- Lợi ích: Nghiên cứu Kuopio ở Phần Lan theo dõi 2862 nam giới từ 42 đến 60 tuổi, trung bình gần 9 năm. Trong 24 tháng trước khi bắt đầu thống kê có 153 người từng hiến máu, trong đó 1 người (0.7%) bị nhồi máu cơ tim cấp. Trong 2529 người không hiến máu có 316 người bị (12.5%), chênh lệch giữa hai tỷ lệ P<0.0001. Sau khi trừ ảnh hưởng của tuổi và các yếu tố nguy cơ bệnh mạch vành khác, người từng hiến máu có nguy cơ nhồi máu cơ tim thấp hơn khoảng 88%. Tỷ số nguy cơ tương đối 0.12 (95% CI 0.02–0.86, P=0.035). Nhưng đây chỉ là theo dõi ghi nhận, không chia nhóm đối chứng. Cơ sở dữ liệu SCANDAT ở Bắc Âu có 1.182.495 người hiến máu, cộng dồn 9.526.627 người-năm theo dõi, trong đó 15.401 người chết. Khi chỉ trừ các đặc điểm như tuổi và giới tính, hiến thêm một lần mỗi năm gắn với tỷ lệ tử vong thấp hơn 18.6% (95% CI 16.8%–20.4%). Khi trừ thêm lớp 「luôn khoẻ mới được phép hiến liên tục」, mức giảm còn 7.5% (5.7%–9.4%). Kết luận của tác giả là 「không thể coi là bằng chứng chắc chắn có lợi cho sức khoẻ」. Thử nghiệm đối chứng chia nhóm ngẫu nhiên FeAST tuyển 1277 người bị bệnh động mạch chi dưới có triệu chứng, chia ngẫu nhiên thành hai nhóm, một nhóm mở máu định kỳ mỗi 6 tháng để giảm sắt, một nhóm làm đối chứng. Chỉ số chính là tử vong toàn nguyên, 125/636 (20%) so với 148/641 (23%). HR 0.85 (HR là tỷ số nguy cơ, 1 nghĩa là hai bên như nhau). 95% CI 0.67–1.08, P=0.17. Tử vong cộng với nhồi máu cơ tim và đột quỵ ở người còn sống, gộp lại là 180 (28%) so với 205 (32%), HR 0.88 (0.72–1.07, P=0.20). Cả hai đều không thấy khác biệt. Thử nghiệm chia nhóm ngẫu nhiên INTERVAL tuyển 45.263 người hiến máu toàn phần. Những người này chia thành ba khoảng cách hiến máu. Nam giới là 12 tuần, 10 tuần, 8 tuần, nữ giới là 16 tuần, 14 tuần, 12 tuần. Rút ngắn khoảng cách thì lượng máu thu được trong hai năm tăng rõ (nhóm nam 8 tuần nhiều hơn nhóm 12 tuần 1.69 đơn vị, 95% CI 1.59–1.80). Chất lượng cuộc sống, hoạt động thể chất, chức năng nhận thức ở các nhóm đều không thấy khác biệt. Nhưng các khó chịu do hiến máu thì nhiều hơn, gồm mệt, khó thở, chóng mắt, hoa mắt và hội chứng chân không yên (ban đêm khó chịu trong chân đến nỗi nằm không yên). Đồng thời trung bình hemoglobin và ferritin thấp hơn, số lần hoãn hiến vì hemoglobin thấp cũng nhiều hơn (từng mục P<0.0001). Thiếu sắt ảnh hưởng tới khả năng tạo nhiệt của cơ thể: 10 phụ nữ thiếu máu do thiếu sắt, 8 người đã cạn dự trữ sắt nhưng chưa thiếu máu, 12 phụ nữ đối chứng, ngâm trong nước 28 ℃ trong 100 phút. Nhóm thiếu máu có nhiệt độ trực tràng 36.0±0.2 ℃, nhóm đối chứng 36.2±0.1 ℃ (P=0.001). Mức tiêu thụ oxy 5.28±0.26 so với 5.99±0.29 mL·min⁻¹·kg⁻¹ (P=0.04). Hai loại hooc môn tuyến giáp là thyroxine và triiodothyronine cũng thấp hơn (P<0.002). Cách nói này trông có vẻ đúng vì người hiến máu vốn đã là người khoẻ qua sàng lọc nhiều lần, còn giả thuyết 「bỏ bớt sắt là bảo vệ tim」 thì chưa được kiểm chứng trong thử nghiệm chia nhóm ngẫu nhiên +- Mức bằng chứng: A +- Nguồn: Salonen JT, Tuomainen TP, Salonen R, Lakka TA, Nyyssönen K (1998). Donation of blood is associated with reduced risk of myocardial infarction. The Kuopio Ischaemic Heart Disease Risk Factor Study. American Journal of Epidemiology, 148(5), 445-451. ；Ullum H, Rostgaard K, Kamper-Jørgensen M, et al. (2015). Blood donation and blood donor mortality after adjustment for a healthy donor effect. Transfusion, 55(10), 2479-2485. ；Zacharski LR, Chow BK, Howes PS, et al. (2007). Reduction of iron stores and cardiovascular outcomes in patients with peripheral arterial disease: a randomized controlled trial. JAMA, 297(6), 603-610. ；Di Angelantonio E, Thompson SG, Kaptoge S, et al. (2017). Efficiency and safety of varying the frequency of whole blood donation (INTERVAL): a randomised trial of 45 000 donors. Lancet, 390(10110), 2360-2371. ；Beard JL, Borel MJ, Derr J (1990). Impaired thermoregulation and thyroid function in iron-deficiency anemia. American Journal of Clinical Nutrition, 52(5), 813-819. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (1997). Luật Hiến máu Cộng hòa Nhân dân Trung Hoa（中华人民共和国献血法）（第二、九、十四条）. ；Ủy ban Quốc gia Y tế và Sức khoẻ cùng các cơ quan khác (国家卫生健康委等) (2024-01-08). 关于进一步做好无偿献血者激励奖励工作的通知. +- Ghi chú: Cái cần bỏ là 「đi hiến máu để tốt cho cơ thể mình」, không phải bản thân việc hiến máu. Máu dùng cho người khác, theo thang người hưởng của cuốn sách thì rơi vào bậc thứ ④ (người xa lạ). Phần quay lại được cho chính bạn là phần nhà nước dành riêng. Điều 14 của Luật Hiến máu (献血法) quy định người hiến máu tự nguyện dùng máu trong bệnh viện thì được miễn các khoản phí lấy máu, bảo quản, tách chiết và kiểm nghiệm, còn vợ chồng và người thân trực hệ được miễn hoặc giảm theo quy định của chính quyền cấp tỉnh. Thông báo tháng 1 năm 2024 của Ủy ban Quốc gia Y tế và Sức khoẻ ghi rõ 28 tỉnh đã áp dụng việc miễn trực tiếp phí dùng máu cho người hiến tự nguyện và người thân khi xuất viện tại bệnh viện trong tỉnh. Người nhận được Giải thưởng công đóng hiến máu tự nguyện còn được hưởng chính sách 「ba miễn」 tại địa phương: miễn phí đi phương tiện công cộng, miễn phí vào các công viên do nhà nước đầu tư, miễn phí khám chữa bệnh ngoại trú thường ở bệnh viện công lập. Muốn hiến thì theo đúng khoảng cách luật định. Điều 9 của Luật Hiến máu (献血法) quy định mỗi lần thường 200 ml, tối đa không quá 400 ml. Hai lần cách nhau không ít hơn 6 tháng. Tuổi khuyến nghị là từ 18 đến 55 tuổi. Đừng rút ngắn khoảng cách tới mức tối thiểu chỉ để lấy danh dự hay chỉ tiêu tập thể. Phụ nữ, người cân nhẹ, người kinh nguyệt nhiều vốn đã thiếu sắt, người hiến nhiều lần có thể tự trả tiền xét ferritin một lần. Sau khi hiến mà cứ mệt, sợ lạnh, hoặc buổi tối nằm xuống khó chịu trong chân đến nỗi phải cử động (gọi là hội chứng chân không yên), đừng tự đoán là 「tổn khí」, hãy xét công thức máu và ferritin. Nếu thiếu sắt thì bổ sung theo lời bác sĩ, đồng thời kéo dài khoảng cách hiến máu. Phần bằng chứng người về việc sợ lạnh chỉ có 10 người, dùng để giải thích vì sao lại vậy, không phải nói có bao nhiêu người bị như thế + +### 18. Đừng dựa vào gói khám sức khoẻ tổng quát để phòng bệnh, hãy làm đúng những xét nghiệm sàng lọc có bằng chứng theo tuổi và giới tính + +- Chi phí: Không tốn tiền, còn tiết kiệm được vài trăm đến vài nghìn tệ gói khám mỗi năm. Đổi lại bạn phải tự nhớ xét những mục nào, mỗi mục cách nhau bao lâu thì làm lại. +- Nói thẳng: Người khám sức khoẻ tổng quát định kỳ và người không khám, tỷ lệ tử vong như nhau, số người chết vì bệnh tim mạch và vì ung thư cũng bằng nhau. Làm đúng những xét nghiệm sàng lọc có bằng chứng theo tuổi và giới tính là đủ. Từ 35 tuổi trở lên, mỗi năm được đo huyết áp miễn phí một lần. Từ 65 tuổi trở lên, mỗi năm có một lần khám sức khoẻ miễn phí. +- Lợi ích: Bản cập nhật năm 2019 của tổng quan hệ thống Cochrane gồm 17 thử nghiệm đối chứng chia nhóm ngẫu nhiên. Khi tính tỷ lệ tử vong toàn nguyên, gộp 11 thử nghiệm trong đó, 233.298 người tham gia, 21.535 ca tử vong. Nhóm khám sức khoẻ tổng quát y hệt nhóm đối chứng, RR 1.00. RR là tỷ số nguy cơ, 1 nghĩa là hai bên như nhau. 95% CI 0.97–1.03 là khoảng tin cậy. Khi tính tử vong do bệnh tim mạch, gộp 9 thử nghiệm, 170.227 người, 6.237 ca tử vong, RR 1.05 (95% CI 0.94–1.16). Khi tính tử vong do ung thư, gộp 8 thử nghiệm, 139.290 người, 3.663 ca tử vong, RR 1.01 (95% CI 0.92–1.12). Cả hai cũng không thấy khác biệt. Kết luận của tác giả là 「khám sức khoẻ tổng quát khó có ích」. Chương trình dịch vụ công cộng y tế cơ bản quốc gia cho cư dân từ 35 tuổi trở lên đo huyết áp miễn phí một lần mỗi năm, và cho người cao tuổi từ 65 tuổi trở lên khám sức khoẻ miễn phí một lần mỗi năm. Lần khám miễn phí này gồm thăm khám thể chất, công thức máu, xét nghiệm nước tiểu, đường huyết lúc nhịn đói và điện tâm đồ. Chức năng gan xét alanin aminotransferase, aspartat aminotransferase, bilirubin toàn phần. Chức năng thận xét creatinin huyết thanh, urê máu. Mỡ máu xét tổng cholesterol, triglyceride, cholesterol mật độ thấp, cholesterol mật độ cao. Cuối cùng còn có siêu âm bụng gan mật tụy lách (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Krogsbøll LT, Jørgensen KJ, Gøtzsche PC (2019). General health checks in adults for reducing morbidity and mortality from disease. Cochrane Database of Systematic Reviews, 1, CD009009. ；Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委) (2017). 国家基本公共卫生服务规范（第三版）（老年人健康管理服务、高血压患者健康管理服务）. +- Ghi chú: Cái cần bỏ là 「lấy gói khám hằng năm làm biện pháp phòng bệnh」, không phải nói đừng làm xét nghiệm nào cả. Thêm nữa, từ 35 tuổi trở lên mỗi năm đo huyết áp miễn phí một lần, từ 65 tuổi trở lên mỗi năm khám sức khoẻ miễn phí một lần, vốn đã do chương trình dịch vụ công cộng y tế cơ bản quốc gia cấp, không tốn tiền túi mình. Mục nào có lý do đáng xét thì cứ xét. Huyết áp, đường huyết, viêm gan B, xem mục 7, 8 và 14 ở phần 1. Vú, cổ tử cung, trực tràng và đại tràng, xem mục 17, 18 và 19 ở phần 1. Vi khuẩn Helicobacter pylori và CT liều thấp, xem mục 23 và 24 ở phần 1. Đã từng có hành vi nguy cơ cao thì đi xét, xem mục 31 ở phần 1. Từ bao nhiêu tuổi bắt đầu xét, mỗi bao lâu xét lại, đều ghi ở đó. Khám do cơ quan tổ chức, người khác trả tiền thì cứ đi. Cái cần chặn là tự bỏ tiền mua gói và tự thêm mục. Người đã mắc bệnh mạn tính, người có triệu chứng, hoặc xét nghiệm bác sĩ chỉ định theo bệnh thì không tính vào đây, đó là khám bệnh chứ không phải sàng lọc. Mục không có bằng chứng thường gặp nhất trong gói là chất chỉ dấu ung thư và chụp hình toàn thân, xem mục 7 của phần này (gói PET-CT toàn thân và chất chỉ dấu ung thư). Xét ra acid uric máu cao mà chưa từng đau, hay sỏi mật mà chưa từng đau, xem mục 19 của phần này (acid uric cao không triệu chứng) và mục 20 của phần này (sỏi mật không triệu chứng). Ngoài ra, phần lớn thử nghiệm Cochrane thu thập đều làm ở các nước thu nhập cao, nơi khám bệnh vốn đã thuận tiện, nên khu vực thiếu nguồn lực y tế không nên sao chép thẳng kết luận này. + +### 19. Đừng bắt đầu uống thuốc hạ acid uric chỉ vì xét sức khoẻ ra acid uric cao mà chưa từng đau lần nào + +- Chi phí: Không tốn tiền, còn tiết kiệm được tiền thuốc dài hạn và tiền tái khám định kỳ, đồng thời tránh được một rủi ro dùng thuốc không cần thiết. +- Nói thẳng: Người có acid uric vượt chuẩn nhưng chưa từng lên cơn gout và cũng chưa có u tophi, Hiệp hội Thấp khớp Hoa Kỳ khuyên chưa bắt đầu uống thuốc hạ acid uric. Phải có 24 người uống đủ 3 năm thì ngăn được một lần lên cơn gout. Muốn nhờ đó bảo vệ thận thì cũng hỏng, chức năng thận ở người uống thuốc vẫn tụt xuống, nhanh y hệt người uống thuốc giả. Đã lên cơn thật rồi hãy điều trị. +- Lợi ích: Hướng dẫn gout năm 2020 của Hiệp hội Thấp khớp Hoa Kỳ có nói tới tình trạng tăng acid uric máu không triệu chứng. Đó là khi acid uric máu >6.8 mg/dL, trước đó chưa từng lên cơn gout, dưới da cũng chưa có u tophi. Hướng dẫn khuyến nghị có điều kiện rằng những người này không nên bắt đầu uống bất kỳ loại thuốc hạ acid uric nào. Các thuốc gồm allopurinol, febuxostat và probenecid. Mức chắc chắn của bằng chứng được xếp là 「cao」. Chú thích dưới bảng ghi rõ 「theo nguy cơ quy doanh, phải điều trị 24 người bệnh trong 3 năm mới ngăn được một lần lên cơn gout」. Thử nghiệm đối chứng chia nhóm ngẫu nhiên CKD-FIX tuyển 363 người. Họ mắc bệnh thận mạn tính giai đoạn 3 hoặc 4, không có tiền sử gout nhưng có nguy cơ tiến triển. Một nhóm uống allopurinol 100–300 mg mỗi ngày, một nhóm uống thuốc giả. Thử nghiệm kéo dài 104 tuần. eGFR là chỉ số phản ánh chức năng thận, đơn vị là mL/min/1.73 m². Một nhóm thay đổi mỗi năm -3.33 (95% CI -4.11 đến -2.55). Nhóm kia thay đổi mỗi năm -3.23 (-3.98 đến -2.47). Hai nhóm chênh nhau -0.10 (-1.18 đến 0.97), P=0.85, tốc độ suy giảm chức năng thận như nhau. Biến cố bất lợi nghiêm trọng 46% so với 44%. Thử nghiệm đối chứng chia nhóm ngẫu nhiên PERL tuyển 530 người bị đái thái tuổi 1 kèm bệnh thận do đái thái ở giai đoạn sớm và trung bình, uống allopurinol 3 năm, rồi ngừng thuốc 2 tháng để rửa hết tác dụng thuốc. Allopurinol quả thật đã hạ acid uric máu từ 6.1 xuống 3.9 mg/dL. Nhưng sau khi ngừng thuốc, chỉ số chức năng thận GFR của hai nhóm chỉ chênh 0.001 mL/min/1.73 m². Khoảng tin cậy của chênh lệch này là 95% CI -1.9 đến 1.9, P=0.99. Lượng albumin lọt ra trong nước tiểu ở nhóm uống allopurinol lại cao hơn 40% (0 đến 80). Nó trông có lợn vì trực giác 「chỉ số vượt chuẩn thì phải uống thuốc ép xuống」 quá thuận lợi +- Mức bằng chứng: A +- Nguồn: FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760（表 1 及其脚注）. ；Badve SV, Pascoe EM, Tiku A, et al. (2020). Effects of Allopurinol on the Progression of Chronic Kidney Disease. New England Journal of Medicine, 382(26), 2504-2513. ；Doria A, Galecki AT, Spino C, et al. (2020). Serum Urate Lowering with Allopurinol and Kidney Function in Type 1 Diabetes. New England Journal of Medicine, 382(26), 2493-2503. +- Ghi chú: Cái cần bỏ là 「chưa từng lên cơn mà đã uống thuốc hạ acid uric cả đời」, không phải nói acid uric cao thì không sao. Cùng hướng dẫn đó có liệt kê ngoại lệ. Dưới đây là vài trường hợp có thể cân nhắc bắt đầu uống. Một là lần lên cơn đầu tiên đi kèm bệnh thận mạn tính giai đoạn 3 trở lên. Hai là acid uric máu >9 mg/dL (khoảng 535 µmol/L). Ba là từng bị sỏi đường tiết niệu. Người đã từng lên cơn, người có u tophi, người chụp chiếu thấy xương đã bị xói mòn thì nên uống thuốc. Cụ thể xem mục 9 ở phần 16 (chẩn đoán gout thì uống thuốc hạ acid uric dài hạn). Trong dân tộc Hán, khoảng 7.4% người mang một kiểu gen gọi là HLA-B\*5801, còn người da trắng chỉ 0.7%. Người mang kiểu gen này uống allopurinol dễ bị phát ban thuốc nặng đe dọa tính mạng, nguy cơ hội chứng quá mẫn ở người châu Á gấp 3 lần người da trắng. Đây vốn là loại thuốc có thể chưa cần uống, không thiếu gì một rủi ro lần này. Nếu thật sự muốn bắt đầu uống, hãy đi xét kiểu gen trước, xem mục 9 ở phần 16 (chẩn đoán gout thì uống thuốc hạ acid uric dài hạn). Hai thử nghiệm về thận đó làm trên người bệnh thận mạn tính và người bị đái thái tuổi 1, không thể mở rộng thành 「acid uric cao hoàn toàn vô hại với thận」. Chúng chỉ bác bỏ một điều, là 「uống thuốc hạ acid uric để bảo vệ thận」 + +### 20. Đừng cắt túi mật phòng bệnh chỉ vì xét sức khoẻ ra sỏi mật mà chưa từng đau lần nào + +- Chi phí: Không tốn tiền, còn tiết kiệm được một ca mổ dưới gây mê toàn thân và một lần nằm viện. Đổi lại bạn phải nhớ vài dấu hiệu nào vừa xuất hiện là phải đi bệnh viện ngay. +- Nói thẳng: Phần lớn người có sỏi trong túi mật nhưng chưa từng đau sẽ không bao giờ đau. Italy theo dõi 580 người như vậy gần chín năm, 78% không hề có triệu chứng. Một nhóm 135 người ở Hoa Kỳ được theo dõi khoảng bốn năm, chỉ 7% phải mổ. Hướng dẫn chẩn đoán và điều trị của Anh viết thẳng: chỉ phát hiện tình cờ trong túi mật có sỏi, không có triệu chứng thì không cần điều trị. +- Lợi ích: Khảo sát dân số MICOL của Italy siêu âm cho 11.229 người từ 29 đến 69 tuổi. Có 856 người được phát hiện sỏi mật, trong đó 580 người không triệu chứng, theo dõi trung bình 8.7 năm. Trong 580 người đó, 453 người (78.1%) không hề có triệu chứng, 61 người (10.5%) có triệu chứng nhẹ, 66 người (11.4%) có triệu chứng nặng. Trong suốt thời gian theo dõi có tổng cộng 189 ca cắt túi mật, trong đó 41.3% làm trên người không triệu chứng. Kết luận của tác giả là 「điều trị chờ xem vẫn là cách xử lý hiệu quả với phần lớn bệnh nhân」, điều trị chờ xem nghĩa là quan sát trước, không mổ. Một HMO ở Hoa Kỳ có số liệu theo dõi, HMO là một loại cơ sở y tế quản lý theo hợp đồng, 691 người được chẩn đoán sỏi mật, theo dõi trung bình 78±61.6 tháng. Trong đó 135 người không triệu chứng được theo dõi 58±50.2 tháng (trung vị 46.3 tháng), chỉ 10% xuất hiện triệu chứng sỏi mật, 7% phải mổ. Trong tổng 50 ca tử vong của cả nhóm chỉ có 2 ca liên quan đến đường mật, và cả hai đều ở nhóm có triệu chứng. Kết luận của tác giả là 「bệnh nhân có sỏi im lặng không cần mổ trước khi xuất hiện triệu chứng」, sỏi im lặng là sỏi không kêu, không lên cơn. Hướng dẫn CG188 của NICE tại điều 1.2.1 viết như sau. 「对胆囊和胆道均正常、偶然发现无症状胆囊结石的人，应告知其在出现症状前不需要治疗」 — với người có túi mật và đường mật bình thường, phát hiện tình cờ sỏi túi mật không triệu chứng, cần thông báo rằng họ không cần điều trị trước khi xuất hiện triệu chứng. Cắt bỏ phòng bệnh trông có lợn vì trực giác 「đã sớm muộn cũng đau, thành cắt một lần lúc còn trẻ」 quá thuận lợi +- Mức bằng chứng: A +- Nguồn: Festi D, Reggiani ML, Attili AF, et al. (2010). Natural history of gallstone disease: Expectant management or active treatment? Results from a population-based cohort study. Journal of Gastroenterology and Hepatology, 25(4), 719-724. ；McSherry CK, Ferstenberg H, Calhoun WF, Lahman E, Virshup M (1985). The natural history of diagnosed gallstone disease in symptomatic and asymptomatic patients. Annals of Surgery, 202(1), 59-63. ；Viện Quốc gia về Chất lượng và Chăm sóc Sức khoẻ (National Institute for Health and Care Excellence) (2014). Gallstone disease: diagnosis and management. NICE guideline CG188，第 1.2.1、1.3.1 条. +- Ghi chú: Ở đây chỉ nói một tình huống: sỏi nằm trong túi mật và chưa từng đau. **Sỏi ở đường mật chính là chuyện khác.** Điều 1.3.1 của cùng hướng dẫn đó ghi rõ, sỏi ở đường mật chính dù có triệu chứng hay không thì vẫn phải làm sạch sỏi trong đường mật và cắt túi mật. Gặp các tình huống sau thì phải đi bệnh viện ngay trong ngày, đừng cứ dựa vào kết luận này mà chịu đựng: đau quặn liên tục vùng thượng bụng phải hay vùng bụng trên, sốt kèm rét run, lòng trắng mắt hoặc da vàng, nước tiểu đậm màu. Khi bác sĩ xem xét kết quả chụp rồi khuyên mổ, hãy hỏi rõ lý do nào, ví dụ đường mật bị ảnh hưởng, có polyp kèm theo, thành túi mật bất thường, đừng dùng kết luận này để bác bỏ bác sĩ. Nó chỉ chặn một việc là 「không lý do nào, chỉ vì phát hiện sỏi thì cắt」. Hai nghiên cứu nêu trên đều chỉ theo dõi một nhóm người rồi nhìn về sau, không chia nhóm đối chứng. Nhưng chúng muốn trả lời là 「khể đụng vào nó thì sẽ ra sao」 chứ không phải 「cách điều trị nào tốt hơn」, dùng ở đây là hợp lý. Ngoài ra hai nghiên cứu này theo dõi từ 4 đến 9 năm, theo dõi lâu hơn thì tỷ lệ xuất hiện triệu chứng sẽ tăng lên + +### 21. Đừng bỏ canxi hoàn toàn để phòng sỏi thận + +- Chi phí: Không tốn tiền. Ngược lại còn tiện hơn, không phải hằng ngày nhớ món nào có canxi. +- Nói thẳng: Người từng bị sỏi canxi oxalat thường được khuyên ăn ít canxi, kết quả thử nghiệm lại ngược hẳn. Nam giới bị sỏi tái phát nhiều lần, ăn canxi bình thường đồng thời giảm đạm động vật và muối, trong năm năm số người tái phát chỉ bằng một nửa nhóm ăn ít canxi. Hướng dẫn của Anh cũng ghi không nên hạn chế canxi. +- Lợi ích: Có một thử nghiệm đối chứng chia nhóm ngẫu nhiên kéo dài 5 năm, tuyển 120 nam giới bị sỏi canxi oxalat tái phát nhiều lần, đồng thời lượng canxi trong nước tiểu lại cao (gọi là tăng canxi niệu không rõ nguyên nhân). Họ được chia ngẫu nhiên thành hai nhóm. Một nhóm ăn canxi bình thường, mỗi ngày 30 mmol. Nhóm này đồng thời ăn ít đạm động vật (mỗi ngày 52 g), ăn ít muối (mỗi ngày 50 mmol natri clorua). Nhóm kia ăn chế độ ăn ít canxi truyền thống, mỗi ngày 10 mmol. Trong 5 năm, số ca tái phát là 12/60 so với 23/60. Tỷ số nguy cơ tương đối chưa hiệu chỉnh là 0.49 (95% CI 0.24–0.98, đây là khoảng tin cậy), P=0.04, nguy cơ tái phát ở nhóm ăn canxi bình thường chỉ bằng một nửa. Trong thời gian theo dõi, lượng canxi trong nước tiểu của cả hai nhóm đều giảm khoảng 170 mg mỗi ngày. Nhưng lượng oxalat thải ra trong nước tiểu thì khác. Nhóm ăn ít canxi tăng trung bình 5.4 mg mỗi ngày, nhóm ăn canxi bình thường giảm trung bình 7.2 mg mỗi ngày. Hướng dẫn NG118 của NICE tại điều 1.8.1 ghi rõ không nên hạn chế lượng canxi nạp vào mỗi ngày. Người lớn giữ ở mức 700 đến 1,200 mg mỗi ngày. Trẻ em và thanh niên 350 đến 1,000 mg mỗi ngày tuỳ theo tuổi. Bỏ canxi trông có lợn vì trực giác 「sỏi làm bằng canxi, vậy ăn ít canxi đi」 quá thuận lợi +- Mức bằng chứng: A +- Nguồn: Borghi L, Schianchi T, Meschi T, et al. (2002). Comparison of two diets for the prevention of recurrent stones in idiopathic hypercalciuria. New England Journal of Medicine, 346(2), 77-84. ；Viện Quốc gia về Chất lượng và Chăm sóc Sức khoẻ (National Institute for Health and Care Excellence) (2019). Renal and ureteric stones: assessment and management. NICE guideline NG118，第 1.8.1 条. +- Ghi chú:Tranh cãi: Thử nghiệm này chỉ làm trên nam giới, và những người đó đều bị sỏi canxi oxalat tái phát kèm lượng canxi trong nước tiểu cao, không thể áp thẳng cho mọi loại sỏi. Thêm nữa nhóm thử nghiệm thay đổi cả ba thứ cùng lúc: ăn canxi bình thường, ăn ít đạm động vật, ăn ít muối, nên không thể gán toàn bộ lợi ích cho một việc 「không bỏ canxi」 là được. Ít muối và ít đạm động vật cũng đóng góp phần của mình. Cái cần bỏ là 「chủ động loại bỏ những thứ chứa canxi」, không phải nói 「nên bổ sung thêm canxi」. Ở đây nói tới canxi trong ba bữa ăn hằng ngày, canxi bổ sung là chuyện khác, người từng bị sỏi có nên uống canxi bổ sung không thì hỏi bác sĩ điều trị. Mục uống nhiều nước thì xem mục 8 ở phần 16 (sau sỏi thận uống nước tới 2.5–3 lít mỗi ngày) + +### 22. Đừng mua tinh thể, chuỗi tay, tượng Pixiu chỉ để 「đổi vận」「kéo tài」「dưỡng người」 + +- Chi phí: Không tốn tiền. Việc phải làm chỉ là đừng tính 「đổi vận」「dưỡng người」 vào lý do mua nó. Chỗ khó là khi mọi người đều đeo, đều nói là linh, mà bạn không theo. +- Nói thẳng: Nói đeo vào thì đổi được vận, kéo được tài, dưỡng được người thì không có bằng chứng nào để kiểm chứng. Nó cùng một lý với chuyện xem bói trông chính xác: lời nói đủ chung chung, bạn tự điền vào. Khi người bán nói tới công năng chữa bệnh thì đó là quảng cáo sai sự thật, năm 2026 có phòng phát trực tiếp bị xử lý vì việc này. Mua làm đồ trang trí thì không sao. +- Lợi ích: Thí nghiệm lớp học kinh điển cho thấy ai nhận bản 「phân tích cá nhân」 chung chung giống nhau đều cho là rất chính xác, hiện tượng đó gọi là hiệu ứng Barnum. Nó giải thích cảm giác 「đeo vào thấy mọi việc thuận lợi」 đến từ đâu. Kết luận của tác giả đó là 「việc người được phân tích chấp nhận không chứng minh được lời giải thích là đúng」. Vụ việc chính thức: Cục Quản lý thị trường quốc gia công bố ngày 4 tháng 6 năm 2026 một loạt vụ điển hình 「đánh giả, làm sạch nguồn gốc」 trong các thị trường thủ công truyền thống. Trong đó có một cửa hàng trang sức ở Tứ Hội, Quảng Đông, bịa chất liệu sản phẩm trong phòng phát trực tiếp. Cửa hàng còn bịa công năng 「phòng trị thấp khớp, tăng cường miễn dịch, cải thiện chất lượng giấc ngủ, thông mạch máu」 cho sản phẩm. Cửa hàng dựng kịch bản để đánh lừa, lừa dối người tiêu dùng. Việc đó vi phạm Điều 9, khoản đầu của Luật Chống cạnh tranh không công (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Forer BR (1949). The fallacy of personal validation: A classroom demonstration of gullibility. Journal of Abnormal and Social Psychology. ；Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2026). 市场监管总局公布一批传统工艺市场「打假清源」典型案例. +- Ghi chú: Kết luận chỉ có một: công năng của nó không thể kiểm chứng, và không đo được khả năng dự báo nào. Không đánh giá tín ngưỡng hay phong tục nào, cũng không nhắm vào các mục đích tặng quà, lưu niệm, truyền lại. Cần tách hai việc: công năng mà người bán tuyên bố có quy định văn bản làm căn cứ, có thể đi khiếu nại; còn các lời như 「đổi vận」 vừa không có nghiên cứu, vừa không chứng minh được là giả, chỉ có thể tự kiềm chế bằng ngân sách. Trước khi mua, hãy coi nó là món trang sức thường, tự hỏi một câu 「nếu nó không thay đổi gì cả, giá này tôi có mua không」. Cách kiểm tra chất liệu và giấy chứng nhận, vì sao coi nó là đầu tư là không đáng, xem mục 33 và 34 ở phần 5 (chuỗi tay và đồng hồ đắt tiền tính theo số tiền đã tiêu, đá quý thì nhận báo cáo CMA). Bỏ tiền xem bói thì xem mục 15 của phần này (xem bói, tarot, chiêm tinh). + +### 23. Đừng trông chờ mua sắm để cải thiện tâm trạng hay cảm giác về vị thế của mình + +- Chi phí: Không tốn tiền. Chỗ khó là lúc tâm trạng tệ mà nhịn không đặt hàng. +- Nói thẳng: Người coi trọng tiền và vật chất càng nặng thì tự đánh giá mình càng không hạnh phúc, nhưng liên hệ rất yếu, và không phân biệt được cái nào dẫn tới cái nào. Có nghiên cứu còn thấy bỏ tiền để mua một trải nghiệm thì vui hơn mua đồ. Nên lúc tâm trạng tệ, bỏ tiền cho một chuyến đi chơi hay một bữa ăn cùng người khác, có khi hữu ích hơn là đầy giỏ hàng. +- Lợi ích: Một phân tích tổng hợp (gộp nhiều nghiên cứu lại) gồm 259 mẫu độc lập, 753 cỡ hiệu ứng. Kết quả là người càng coi trọng vật chất thì hạnh phúc tự báo càng thấp. Trên loại thang đo đa chiều dùng phổ biến nhất, hệ số tương quan r = -0.19 (sau khi hiệu chỉnh ρ = -0.24). Trên thang chỉ đo một mặt là 「coi trọng tiền bạc」 thì liên hệ yếu hơn, r = -0.08 đến -0.11. Tách theo loại kết quả đang xem thì mức độ khác nhau. Liên hệ mạnh nhất với hành vi sức khoẻ và chi tiêu phiêu lưu cùng đánh giá tiêu cực về bản thân, r = -0.28 đến -0.44. Liên hệ yếu nhất với hài lòng cuộc sống và tâm trạng tiêu cực, r = -0.13 đến -0.15. Tác giả còn làm phân tích trung gian (xem nó truyền qua mắt xích gì), kết quả gợi ý liên hệ nghịch này có thể do nhu cầu tâm lý chưa được đáp ứng. Một nghiên cứu khác: trong hai cuộc khảo sát trên các nhóm người khác nhau, người trả lời đều nói tiền bỏ ra để có một trải nghiệm sống làm mình vui hơn tiền bỏ ra để mua đồ. Trong thí nghiệm phòng thí nghiệm sau đó, hồi tưởng lại lúc vui sau khi bỏ tiền cho một trải nghiệm cũng cao hơn hồi tưởng lại lúc bỏ tiền mua đồ (không áp dụng được tiêu chuẩn toàn quốc, cả hai nghiên cứu đều dùng mẫu ở nước ngoài) +- Mức bằng chứng: B +- Nguồn: Dittmar H, Bond R, Hurst M, Kasser T (2014). The relationship between materialism and personal well-being: A meta-analysis. Journal of Personality and Social Psychology, 107(5), 879-924. ；Van Boven L, Gilovich T (2003). To do or to have? That is the question. Journal of Personality and Social Psychology, 85(6), 1193-1202. +- Ghi chú:Tranh cãi: Phần lớn nghiên cứu được thu thập hỏi một lớp người tại cùng một thời điểm, xem hai thứ có khớp nhau không, không phân biệt được cái nào trước cái nào. Chính tác giả cũng viết trong kết luận rằng muốn thấy rõ chiều hướng thì cần thêm thử nghiệm và nghiên cứu theo dõi dài hạn. Vậy không thể đọc thành 「mua đồ làm người ta không vui」, chỉ có thể đọc thành 「trông chờ mua đồ để thành vui thì không mấy khả thi」. Hơn nữa liên hệ rất yếu: r = -0.19 chỉ giải thích được vài phần trăm khác biệt, giữa người với người rất khác nhau. Cả hai nghiên cứu đều dùng mẫu ở nước ngoài, không có dữ liệu cùng loại cho người Trung Quốc để viện dẫn. Ở đây không phản đối việc bỏ tiền cho thứ mình thích, cái bị phản đối là coi mua sắm là cách điều chỉnh tâm trạng, vì làm thế thì lần sau tâm trạng tệ bạn vẫn đặt hàng. Cụ thể tiết kiệm tiền thì xem phần 5. Làm gì khi tâm trạng tệ mà thật sự có bằng chứng, xem nửa sau của phần 22 và phần 29. Khoản ngân sách cộng thêm để 「mạnh hơn người ta một bậc」, xem mục 24 của phần này (dời lên một bậc). + +### 24. Đừng bỏ thêm tiền đổi nhà, đổi xe, đổi giới xung quanh chỉ để 「dời lên một bậc trong nhóm người xung quanh」 + +- Chi phí: Không tốn tiền. Việc phải làm chỉ là trước khi cộng thêm ngân sách, hỏi thêm một câu: khoản chênh lệch này mua chính cái đồ, hay mua vị trí của tôi trong nhóm người đó. Chỗ khó là khi mọi người xung quanh đều đổi, mà bạn không theo. +- Nói thẳng: So với người có thu nhập ngang nhau, hàng xóm càng kiếm được nhiều thì người đó càng báo cáo hạnh phúc thấp hơn. Điều này chỉ xảy ra với hàng xóm có cùng trình độ học vật và hay gặp gỡ. Nên khoản tiền bỏ thêm để dọn vào nơi mà ai cũng dư dả hơn mình, cái mua được là đẩy mình xuống một bậc. Nhưng khi thu nhập của cả hai cùng tăng, người ta vẫn thấy dễ sống hơn, mức tuyệt đối vẫn có tác dụng. +- Lợi ích: Khảo sát Gia đình và Hộ gia đình toàn quốc Hoa Kỳ (NSFH) phỏng vấn lại cùng một lứ người trong hai vòng. Tác giả đối chiếu bộ dữ liệu này với thu nhập trung bình địa phương. Thu nhập trung bình địa phương (PUMA) được suy ra từ mẫu 5% của tổng điều tra dân số năm 1990. Sau khi trừ thu nhập của chính người đó và một loạt đặc điểm cá nhân, kết quả là: thu nhập của hàng xóm càng cao thì hạnh phúc tự báo của bản thân càng thấp. Hiệu ứng lớn, đổi cách tính vẫn giữ nguyên, và rất có ý nghĩa. Cách diễn đạt của tác giả là mức hạnh phúc giảm do 「thu nhập hàng xóm tăng」 và do 「thu nhập của bản thân giảm cùng mức đó」 nằm ở cùng một bậc lớn. Khi đổi sang so một người với chính họ trước và sau (gọi là thêm hiệu ứng cố định cá nhân), hệ số giữ nguyên dấu và gần nguyên giá trị, chỉ có mức ý nghĩa giảm xuống còn 10%. Khi giới hạn lại để chỉ so trong cùng một tiểu bang và cùng một vòng khảo sát (gọi là hiệu ứng cố định tiểu bang và vòng khảo sát), kết quả hầu như không đổi. Điều đó cho thấy hiệu ứng không phải do 「người hay chuyển tới nơi nghèo vốn đã vui hơn」, cũng không phải do một yếu tố chưa đo được ở cấp tiểu bang. Hiệu ứng mạnh hơn ở người hay gặp hàng xóm, không có ý nghĩa ở người chỉ gặp bạn bè ngoài khu phố. Người thuê nhà và người có nhà không khác nhau, điều này không ủng hộ cách giải thích 「chỗ đắt thì giá nhà cao, nên số tiền thực chi tiêu được còn ít」. Hạnh phúc chỉ giảm theo thu nhập của hàng xóm cùng trình độ học vật, gần như không phản ứng với thu nhập của hàng xóm khác trình độ. Hơn nữa nó chủ yếu truyền qua 「mức hài lòng với tình cảnh vật chất của bản thân」, không phải qua mức hài lòng với các mặt khác như đời sống gia đình. Cuối cùng, nếu thu nhập của bản thân và hàng xóm cùng tăng theo cùng tỷ lệ, hạnh phúc vẫn tăng. +- Mức bằng chứng: B +- Nguồn: Luttmer EFP (2005). Neighbors as Negatives: Relative Earnings and Well-Being. The Quarterly Journal of Economics, 120(3), 963–1002. ；Luttmer EFP (2004). Neighbors as Negatives: Relative Earnings and Well-Being. NBER Working Paper No. 10667（phần diễn giải và cỡ lớn được viện dẫn ở mục này được đối chiếu từng chữ theo bản này）. +- Ghi chú: Có vài lý do đặt mức B: dữ liệu đến từ bảng hỏi có phỏng vấn lại nhiều lần cùng một lứ người, chỉ có đúng nghiên cứu này, mẫu lại ở nước ngoài; kết quả đo là hạnh phúc do người trả lời tự báo, không phải tiền. Tác giả đúng là đã dùng thống kê để loại vài loại nhiễu, ví dụ so một người trước và sau, so trong cùng tiểu bang và cùng vòng, nhưng xét cho cùng không phải thử nghiệm ngẫu nhiên. Người Trung Quốc cũng không có dữ liệu cùng loại để viện dẫn. Cỡ lợi ích đặt 「vừa」 là theo đúng cách tính của mục 23 của phần này (đừng trông chờ mua sắm cải thiện tâm trạng). Số tiền tiết kiệm được đúng là có thể lên tới mức vạn tệ. Nhưng 「tiết kiệm được khoản này đổi lại bao nhiêu hạnh phúc」 thì không có con số nào đỡ, nên không tính là 「lớn」 ở mức vạn tệ. Ở đây không phản đối việc bỏ tiền cho căn nhà rộng hơn, khu chung cư yên tĩnh hơn, quãng đường đi làm ngắn hơn, những thứ đó mua chính cái đồ. Cái cần chặn là lý do 「đồng nghiệp đều đổi, mình không được kém họ」. Xếp ưu tiên khi chọn chỗ ở thì xem mục 18 ở phần 4 (khi chọn chỗ ở, đặt thời gian đi làm lên đầu). Mua sắm để điều chỉnh tâm trạng thì xem mục 23 của phần này (đừng trông chờ mua sắm cải thiện tâm trạng). Trên mạng cứ so lên xem mình ở đâu thì xem mục 21 ở phần 3 (đừng coi 「người ta sống thế nào」 là bài đọc bắt buộc mỗi ngày) + +### 25. Đừng tin 「ý chí giống cơ bắp, dùng một lúc là hết」 + +- Chi phí: Không tốn tiền. Cái cần sửa chỉ là một cách nói. +- Nói thẳng: Cách nói phổ biến là làm xong một việc tốn sức tự kiểm soát thì việc kế tiếp dễ phá quy tắc hơn, nên phải dùng ý chí tiết kiệm. 23 phòng thí nghiệm, 2141 người làm lại thí nghiệm này theo cùng một bộ phương án, kết quả gần như bằng không. Đừng dùng 「hôm nay ý chí đã hết」 để giải thích cho bản thân nữa, cũng đừng bỏ tiền mua khoá học dạy cách tích trữ ý chí. +- Lợi ích: Một nghiên cứu lặp lại có đăng ký trước, nhiều phòng thí nghiệm, 23 phòng thí nghiệm, tổng cộng 2141 người tham gia, làm lại mô hình nhiệm vụ tuần tự theo bộ phương án chuẩn hoá của Sripada và cộng sự: làm trước một nhiệm vụ tiêu tốn sức tự kiểm soát, rồi làm nhiệm vụ thứ hai cũng cần sức tự kiểm soát. Kết quả gộp cho thấy hiệu ứng hao mòn bản thân rất nhỏ, và 95% khoảng tin cậy có chứa 0 (d = 0.04, 95% CI −0.07 đến 0.15). Tác giả dựa vào đó bàn về việc mô hình 「tự kiểm soát là một nguồn lực hữu hạn」 còn đứng vững hay không. +- Mức bằng chứng: A +- Nguồn: Hagger MS, Chatzisarantis NLD, Alberts H, et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science, 11(4), 546–573. +- Ghi chú: Cái cần bỏ là mô hình 「tự kiểm soát là một nguồn lực bị hao mòn hết」, không phải phủ nhận 「người ta sẽ mệt」. Ngủ thiếu và làm việc liên tục quá lâu khiến hiệu suất giảm, đó là chuyện khác, xử lý theo mấy mục về giấc ngủ và giờ làm ở phần 3. Cách làm thật sự có thử nghiệm chia nhóm ngẫu nhiên đỡ là đổi môi trường và đổi cách viết, xem mục 10 ở phần 4 (đưa món không muốn chạm tới ra xa) và mục 1 ở phần 4 (viết thành 「mấy giờ, ở đâu, gặp gì thì làm gì」). Những mục khác của phần này tính bằng tiền, riêng mục này tính bằng thời gian và sức lực, hai bên không quy đổi qua lại. + +### 26. Đừng trông chờ ăn sáng hay chế độ nhịn ăn 16:8 giúp giữ cân nặng, hãy chọn khung giờ ăn nào bạn giữ được lâu dài + +- Chi phí: Không tốn tiền. Phần tiết kiệm được là chút sức lực mà mỗi ngày bạn phải tranh luận với giờ ăn. +- Nói thẳng: Cách nói phổ biến là bỏ bữa sáng sẽ khiến người ta ăn bùng ở bữa trưa và bữa tối, tổng cả ngày lại nhiều hơn. Trong thử nghiệm thì không phải: người không ăn sáng mỗi ngày ăn ít hơn hơn hai trăm kcal, cân nặng còn nhẹ hơn một chút. So với chỉ ăn ít đi, chế độ 16:8 sau một năm cũng không khác biệt về cân nặng. Người lâu ngày không ăn sáng quả thật mắc bệnh tim mạch nhiều hơn khoảng một phần mươi bảy, còn tỷ lệ tử vong toàn nguyên thì không thấy khác biệt. +- Lợi ích: Xem trước lời nói 「bỏ bữa sáng sẽ khiến người ta ăn nhiều hơn về sau」. Một bài phân tích tổng hợp gộp 13 thử nghiệm ngẫu nhiên. Người được xếp vào nhóm ăn sáng mỗi ngày ăn vào nhiều hơn nhóm không ăn sáng 259.79 kcal (95% CI 78.87–440.71, đây là khoảng tin cậy). Cân nặng cao hơn 0.44 kg (0.07–0.82). Người không ăn sáng không bù lại ở bữa trưa và bữa tối. Các thử nghiệm này theo dõi rất ngắn, theo cân nặng trung bình 7 tuần, theo lượng ăn trung bình 2 tuần. Xem tiếp chế độ 16:8, tức chỉ ăn trong một khoảng giờ cố định. Một thử nghiệm cho 116 người lớn thừa cân làm trong 12 tuần. Nhóm chỉ ăn từ 12 giờ trưa tới 8 giờ tối và nhóm ăn đều ba bữa mỗi ngày chênh nhau −0.26 kg (−1.30 đến 0.78), không có ý nghĩa. Một thử nghiệm khác làm ở Quảng Châu, 139 người bệnh béo phì, theo dõi 12 tháng. Ăn trong khung giờ kèm kiểm soát calo giảm nhiều hơn chỉ kiểm soát calo 1.8 kg (−4.0 đến 0.4), cũng không có ý nghĩa. Cuối cùng xem bỏ bữa sáng lâu dài có gây hại không. Trong tổng hợp 2.380.000 người, người không ăn sáng có nguy cơ bệnh tim mạch cao hơn khoảng 17% (OR 1.17, 1.09–1.26), tử vong do tim mạch cao hơn khoảng 49% (OR 1.49, 1.20–1.84). Nhưng đây là những nghiên cứu chỉ ghi nhận, không chia nhóm. Trong nghiên cứu gốc theo dõi từ 17 đến 23 năm, tỷ số nguy cơ tử vong toàn nguyên là 1.19, khoảng tin cậy 0.99 đến 1.42, vượt qua 1, không tính là khác biệt. +- Mức bằng chứng: A +- Nguồn: Sievert K, Hussain SM, Page MJ, 等 (2019). Effect of breakfast on weight and energy intake: systematic review and meta-analysis of randomised controlled trials. BMJ. ；Lowe DA, Wu N, Rohdin-Bibby L, 等 (2020). Effects of Time-Restricted Eating on Weight Loss and Other Metabolic Parameters in Women and Men With Overweight and Obesity. JAMA Internal Medicine. ；Liu D, Huang Y, Huang C, 等 (2022). Calorie Restriction with or without Time-Restricted Eating in Weight Loss. New England Journal of Medicine. ；Zhang H, Zhang S, Liu Y, 等 (2025). The association between skipping breakfast and cardiovascular disease: a meta analysis. Frontiers in Cardiovascular Medicine. ；Rong S, Snetselaar LG, Xu G, 等 (2019). Association of Skipping Breakfast With Cardiovascular and All-Cause Mortality. Journal of the American College of Cardiology. ；Fernandes-Alves D, Teixeira GP, Guimarães KC, Crispim CA (2026). Systematic Review and Meta-analysis of Randomized Clinical Trials Comparing Time-Restricted Eating With and Without Caloric Restriction for Weight Loss. Nutrition Reviews. ；Sun T, Zhang L, Lu Y, 等 (2024). Non-linear relationship between the first meal time of the day and gallstone incidence in American adults. Frontiers in Nutrition. ；Liu T, Wang Y, Wang X, 等 (2023). Habitually Skipping Breakfast Is Associated with the Risk of Gastrointestinal Cancers: Evidence from the Kailuan Cohort Study. Journal of General Internal Medicine. +- Ghi chú:Tranh cãi. Con số 17% về bệnh tim mạch đến từ những nghiên cứu chỉ ghi nhận, không chia nhóm. Người không ăn sáng vốn đã dễ hút thuốc, uống rượu, ngồi nhiều, thu nhập thấp hơn, mà không hiệu chỉnh sạch được. Trong 17% đó có bao nhiêu là do bữa sáng, nói không rõ. Chỉ số tử vong toàn nguyên vốn đã vượt qua 1. Chế độ 16:8 cũng còn tranh cãi: gộp 30 thử nghiệm, 1341 người, với điều kiện hai nhóm ăn bằng nhau, nhóm ăn trong khung giờ giảm nhiều hơn 1.46 kg (−2.65 đến −0.26), đổi lại mất thêm 0.41 kg khối lượng cơ. Nhưng phần lớn các thử nghiệm này rất ngắn, thử nghiệm dài và nghiêm nhất là thử nghiệm mười hai tháng thì nói không có khác biệt. Nhịn ăn lâu còn liên quan tới túi mật. Trong một khảo sát cắt ngang 6.547 người, bữa đầu tiên trong ngày muộn hơn mỗi giờ một chút thì tỷ lệ phát hiện sỏi mật cao hơn khoảng 5% (OR 1.05, 1.02–1.08); người ăn bữa đầu muộn từ 9 giờ tới 14 giờ cao hơn khoảng 49% so với người ăn sớm hơn (OR 1.49, 1.24–1.77). Cơ chế là mật đọng trong túi mật càng lâu càng dễ kết tinh thành sỏi. Nhưng đây là khảo sát tại một thời điểm, cũng có thể là do đã có vấn đề túi mật rồi mới đổi thói quen ăn. Nghiên cứu mẫu Kailuan ở Trung Quốc theo dõi 5.6 năm, có 369 ca ung thư đường tiêu hoá: người không ăn sáng có nguy cơ ung thư đại tràng và trực tràng gấp 2.32 lần (1.34–4.01), ung thư túi mật và đường mật ngoài gan gấp 5.43 lần (1.34–21.93). Khoảng tin cậy của con số sau rộng tới 21.93, nghĩa là số ca quá ít, chỉ coi là manh mối. Gộp lại, xét riêng chuyện giữ cân nặng thì giờ ăn ảnh hưởng rất nhỏ, đừng phí tâm lực vì nó. Còn chuyện bỏ hẳn bữa sáng lâu dài có tốt không, bằng chứng hiện có toàn là tương quan, không chứng minh được nhân quả, nhưng hướng thì nhất quán nghiêng về bất lợi; muốn an toàn thì ăn bình thường, chỉ đừng trông nó giúp giảm cân. Nguyên nhân chính của viêm dạ dày và loét dạ dày không phải là bụng rỗng, mà là Helicobacter pylori và uống thuốc giảm đau lâu dài; mục nào cần xét thì xem mục 23 ở phần 1 (xét Helicobacter pylori, dương tính thì diệt). Xét sức khoẻ ra sỏi mật mà chưa từng đau thì làm gì, xem mục 20 của phần này (xét sức khoẻ ra sỏi mật mà chưa từng đau, đừng cắt túi mật phòng bệnh). Muốn giữ cân nặng, thứ có bằng chứng là ăn gì và ăn bao nhiêu, xem mục 28 ở phần 2 (mỗi ngày đủ 5 phần rau quả), mục 29 ở phần 2 (ăn ít thực phẩm chế biến công nghiệp) và mục 33 ở phần 2 (giữ BMI ở 20–25). Người bị đái thái, người đang uống thuốc hạ đường huyết, hay phụ nữ mang thai thì không áp dụng, bỏ một bữa có thể hạ đường huyết, hãy làm theo chỉ định của bác sĩ. Nhịn ăn cực đoan và nôn kích thực là chuyện khác, xem mục 1 ở phần 28 (đừng dùng ăn kiêu cực đoan, nhịn ăn hoặc nôn kích thực để giữ cân nặng). + +### 27. Đừng tin 「thủ dâm tổn hại sức khoẻ」「nhịn thái dâm mới được cải thiện」, điều cần xem thật sự là có làm lỡ giấc ngủ, công việc và quan hệ hay không + +- Chi phí: Không tốn tiền. Phần tiết kiệm được là tiền mua phương thuốc 「bổ thận」 và khoá học 「cai thái dâm», cùng cả mối lo ây tự trách mình lâu dài. +- Nói thẳng: Không có nghiên cứu nào nói thủ dâm tổn hại sức khoẻ. Ngược lại, một nghiên cứu lớn ở Hoa Kỳ thấy nam giới xuất tinh nhiều lần mắc ung thư tuyến tiền liệt ít hơn khoảng hai phần mươi. Một nghiên cứu ở Anh thấy người quan hệ nhiều ở tuổi từ hai mươi đến ba mươi có nguy cơ cao hơn, nên cả hai hướng đều không phải lý do để tập thể dục. Nhiều người cho rằng mình 「nghiện xem nội dung khiêu dâm」, phần lớn là vì việc đó xung đột với đạo đức của chính mình. +- Lợi ích: Xem phần thể chất trước. Một nghiên cứu ở Hoa Kỳ theo dõi 31925 nam giới tới năm 2010. Mỗi người tự báo số lần xuất tinh mỗi tháng, rồi chia nhóm theo số lần. So với người 4 đến 7 lần mỗi tháng, người 21 lần trở lên mỗi tháng có ung thư tuyến tiền liệt HR 0.81 (95% CI 0.72–0.92, thấp hơn khoảng 19%), tính theo số lần ở tuổi 20 đến 29. Tính theo số lần ở tuổi 40 đến 49, HR 0.78 (0.69–0.89, thấp hơn khoảng 22%). Phần giảm chủ yếu là ung thư tuyến tiền liệt nguy cơ thấp. Hướng ngược lại là một nghiên cứu bệnh chứng đối chứng ở Anh, so 431 người bệnh được chẩn đoán trước 60 tuổi với 409 người đối chứng. Người thủ dâm nhiều ở tuổi từ hai mươi đến ba mươi có nguy cơ cao hơn, người thủ dâm nhiều ở tuổi năm mươi mấy thì nguy cơ thấp hơn. Xem tiếp nội dung khiêu dâm. Một tổng quan hệ thống kèm phân tích tổng hợp cho rằng nhiều người cho là mình 「nghiện rồi」, phần lớn là vì hành vi không khớp với điều mình tin, chứ không chỉ vì xem nhiều (Hoa Kỳ / Anh) +- Mức bằng chứng: B +- Nguồn: Rider JR, Wilson KM, Sinnott JA, et al. (2016). Ejaculation Frequency and Risk of Prostate Cancer: Updated Results with an Additional Decade of Follow-up. European Urology, 70(6), 974–982. ；Dimitropoulou P, Lophatananon A, Easton D, et al. (2009). Sexual activity and prostate cancer risk in men diagnosed at a younger age. BJU International, 103(2), 178–185. ；Grubbs JB, Perry SL, Wilt JA, Reid RC. (2019). Pornography Problems Due to Moral Incongruence: An Integrative Model with a Systematic Review and Meta-Analysis. Archives of Sexual Behavior, 48(2), 397–415. +- Ghi chú:Tranh cãi. Hai nghiên cứu về ung thư tuyến tiền liệt đều chỉ ghi nhận, không có nhóm đối chứng, số lần lại do người tham gia nhớ lại, nên kết luận chỉ có thể dừng ở 「không thấy bằng chứng tổn hại」, không thể ngược lại khuyên nên làm nhiều hơn. Chính tác giả nghiên cứu ở Anh cũng nói, đoạn tuổi năm mươi mấy có thể là đã bị bệnh rồi mới ít xuất tinh hơn. Tóm tắt bài nghiên cứu về nội dung khiêu dâm không đưa ra cỡ hiệu ứng, nên ở đây chỉ ghi hướng. Lợi ích đặt là nhỏ, vì phần giảm chủ yếu là ung thư tuyến tiền liệt nguy cơ thấp, không phải tử vong. Dấu hiệu thật sự nên đi khoa tâm thần hoặc khoa tâm lý lâm sàng là: thức khuya vì chuyện này, mất công, ảnh hưởng quan hệ với bạn đời, muốn dừng mà không dừng được. Thức khuya thì xem mục 9 ở phần 3 (đừng thức khuya vì game, video ngắn, nội dung khiêu dâm). Lo xem nội dung khiêu dâm nhiều dẫn tới rối loạn cương, trước hết hãy đi xét theo mục 28 ở phần 1 (rối loạn cương thì đi xét tim mạch trước). Gửi vào nhóm hoặc bán tài nguyên thì hậu quả pháp lý, xem mục 4 ở phần 9 (xem video khiêu dâm ở nhà thì phần mình tự chịu). Người hưởng là chính bạn. + +### 28. Đừng bỏ tiền làm 「điều chỉnh xu hướng tính dục」「chữa chứng đồng tính」, cũng đừng đưa người nhà đi + +- Chi phí: Không tốn tiền. Phần tiết kiệm được là tiền cho các cơ sở điều chỉnh, phòng khám tâm lý và khoá học 「cai đồng tính」. Chỗ khó là người nhà vẫn nhất quy định phải đưa đi. +- Nói thẳng: Nhóm chuyên gia của Hiệp hội Tâm lý Hoa Kỳ, sau khi đọc nghiên cứu, nói loại điều chỉnh này không đổi được xu hướng tính dục, còn có thể gây hại. Trong một khảo sát ở Hoa Kỳ, người từng trải qua điều chỉnh có ý nghĩ tự sát nhiều hơn gần chín phần mươi. Ở Trung Quốc có bệnh viện tâm thích dùng danh nghĩa 「rối loạn xu hướng tình dục」 để ép nhận bệnh một người đàn ông, bị tòa xét là xâm phạm, bồi 5000 tệ. +- Lợi ích: Báo cáo của nhóm công tác năm 2009 của Hiệp hội Tâm lý Hoa Kỳ nói nỗ lực đổi xu hướng tính dục 「khó có thể thành công, và có một mức rủi ro gây hại nhất định」. Với việc đưa trẻ vị thành niên vào các chương trình điều chỉnh bắt buộc hoặc nằm viện, báo cáo khuyên cả cha mẹ lẫn đứa trẻ đều không nên cân nhắc. Một khảo sát toàn quốc có tính đại diện ở Hoa Kỳ phỏng vấn 1518 người lớn thuộc nhóm thiểu số tính dục, khoảng 7% từng trải qua điều chỉnh, trong đó tám phần mười đến từ người có đạo. Sau khi trừ tuổi, các trải nghiệm bất hạnh thời thơ ấu và những yếu tố khác, người từng trải qua điều chỉnh có OR từng có ý nghĩ tự sát trong đời là 1.92 (95% CI 1.01–3.64, cao hơn khoảng chín phần mươi). OR đã lên kế hoạch tự sát là 1.75 (1.01–3.06). OR thử tự sát không bị thương hoặc bị thương nhẹ là 1.88 (1.01–3.50). Một nghiên cứu khác phỏng vấn 245 người trẻ từ 21 đến 25 tuổi. Trong số người không bị điều chỉnh ở tuổi vị thành niên, 22.0% đã thử tự sát. Người mà cha mẹ tự tìm cách đổi xu hướng tính dục của con thì chiếm 48.1%. Người còn bị cha mẹ đưa đi tìm nhà trị liệu hoặc người tôn giáo thì chiếm 62.8%. Bảng phân loại bệnh ICD-10 của Tổ chức Y tế Thế giới ghi rõ: 「bản thân xu hướng tính dục không được coi là một rối loạn」. Năm 2017 Tòa án huyện Dịch Thành, trấn Trực Mã, tỉnh Hà Nam xét một vụ: bệnh viện tâm thích dùng danh nghĩa 「rối loạn xu hướng tình dục」 để ép nhận bệnh một người đàn ông. Tòa cho rằng ông 「bệnh ở mức trung bình, cũng không có hành vi hoặc nguy cơ tự sát, làm hại người, phá tài sản, không đáp ứng điều kiện phải điều trị bắt buộc」. Tòa buộc bệnh viện công khai xin lỗi và bồi 5000 tệ bù tổn thương tinh thần (Hoa Kỳ / Trung Quốc) +- Mức bằng chứng: B +- Nguồn: American Psychological Association Task Force on Appropriate Therapeutic Responses to Sexual Orientation (2009). Report of the Task Force on Appropriate Therapeutic Responses to Sexual Orientation. ；Blosnich JR, Henderson ER, Coulter RWS, Goldbach JT, Meyer IH. (2020). Sexual Orientation Change Efforts, Adverse Childhood Experiences, and Suicide Ideation and Attempt Among Sexual Minority Adults, United States, 2016–2018. American Journal of Public Health, 110(7), 1024–1030. ；Ryan C, Toomey RB, Diaz RM, Russell ST. (2020). Parent-Initiated Sexual Orientation Change Efforts With LGBT Adolescents: Implications for Young Adult Mental Health and Adjustment. Journal of Homosexuality, 67(2), 159–173. ；World Health Organization. ICD-10 Version:2019, F66 Psychological and behavioural disorders associated with sexual development and orientation. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2018, sửa đổi). Luật Sức khoẻ Tâm thần（精神卫生法）（第二十三、三十、七十八条）. ；Trang web Viện kiểm sát nhân dân Khu tự trị Quảng Tây đăng lại bài của Tân Kinh Báo (2017). 驻马店「同性恋强制治疗案」终审. ；Phản biện: Sullins DP. (2022). Sexual Orientation Change Efforts Do Not Increase Suicide: Correcting a False Research Narrative. Archives of Sexual Behavior, 51, 3377–3393. ；Blosnich JR, Coulter RWS, Henderson ER, et al. (2023). Correcting a False Research Narrative: A Commentary on Sullins (2022). Archives of Sexual Behavior, 52(3), 885–888. +- Ghi chú:Tranh cãi. Sullins năm 2022 tính lại trên cùng bộ dữ liệu đó, và trừ đi những ý nghĩ tự sát đã có sẵn trước khi điều chỉnh. Tính như vậy thì việc điều chỉnh và ý nghĩ tự sát không còn tương quan dương (OR 0.44, 0.20–0.94). Cơ quan nơi Sullins làm việc có lập trường chống đồng tính, và tác giả gốc cũng đã viết phản hồi. Vậy mức độ gây hại còn tranh cãi, nhưng không thấy bằng chứng phản biện cho điểm 「không đổi được」. Đặt mức B vì các con số về hậu quả đều đến từ khảo sát hồi tưởng sau sự việc. Về mặt pháp lý, điều trị nằm viện theo nguyên tắc tự nguyện, chỉ khi có rối loạn tâm thần nặng và có nguy cơ làm hại bản thân hoặc người khác thì mới ép nhận bệnh. Nhân viên tâm lý không được thực hành trị liệu tâm lý. Cố ý đưa người không bị rối loạn tâm thần vào bệnh viện để điều trị thì phải chịu trách nhiệm bồi thường. Người đau khổ vì chuyện này lâu dài có thể đi khoa tâm thần hoặc khoa tâm lý lâm sàng, mục tiêu là giảm đau khổ. Con cái nói ra thì gia đình nên làm gì, xem mục 15 ở phần 30 (con nói mình thích người cùng giới). Người hưởng là chính bạn và người nhà. diff --git "a/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" "b/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" index d023a16b..ac321ff8 100644 --- "a/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" +++ "b/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" @@ -173,38 +173,38 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Nguồn: Cục Quản lý chất lượng và kiểm định quốc gia (质检总局) (2009). 强制性产品认证管理规定（质检总局令第 117 号，第二、四十九条）. ；Cục Quản lý thị trường quốc gia (市场监管总局) (2023). 食品经营许可和备案管理办法（市场监管总局令第 78 号，第四条）. - Ghi chú: Bản danh sách tự thân là phương pháp kinh nghiệm, phần điều luật đã đối chiếu nguyên văn. Trước khi đưa vào sản xuất, vào hệ thống tra cứu nhãn hiệu của Cục Sở hữu trí tuệ quốc gia xem đã có nhãn hiệu gần giống được đăng ký chưa, rồi hãy nộp đăng ký nhãn hiệu của mình. Vạch đỏ khi dùng nhãn hiệu và hình ảnh của người khác, xem mục 20 và 21 (mua hàng phải giữ chứng từ, hình ảnh phải tự làm hoặc mua giấy phép). Danh mục chứng nhận bắt buộc căn cứ bản Cục Quản lý thị trường quốc gia công bố mới nhất. -### 20. Mua hàng thì mỗi lô giữ lại chứng từ và thông tin bên bán, giá mua thấp hẳn so với thị trường thì không nhận: hàng giả do nhân viên mua, người bị kết án là chủ doanh nghiệp +### 20. Mỗi lô hàng nhập về để lại một bộ hoá đơn và thông tin nhà cung cấp, giá nhập thấp hơn hẳn giá thị trường thì không nhập: hàng giả do nhân viên mua, người bị kết án là chủ -- Chi phí: Không tốn tiền. Mỗi lô hàng giữ lại mấy thứ: hợp đồng mua hàng, hóa đơn, chứng từ thanh toán, giấy phép kinh doanh và giấy uỷ quyền của bên bán. Rồi đặt một quy tắc nội bộ: ai mua hàng thì người đó ký tên, giá mua thấp hơn rõ rệt so với giá thị trường thì không nhận. Khó là bỏ qua những lô giá rẻ trông rất có lãi. -- Nói thẳng: Bán hàng giả, thu được 3 vạn tệ, hoặc bán được 5 vạn tệ, là đã đủ để kết án, phạt tù dưới 3 năm. Hàng còn nằm trong kho chưa bán đi cũng tính. 「我不知道是假的」không phải do bạn quyết: mua giá thấp hẳn so với thị trường mà không nêu được lý do thì bị coi là bạn đã biết. Giữ hợp đồng mua hàng, hóa đơn, chứng từ thanh toán và thông tin bên bán, thì mới có đường miễn bồi thường. -- Lợi ích: Bán hàng hóa biết là hàng đăng ký nhãn hiệu giả, khi thu nhập bất hợp pháp đủ 3 vạn tệ, hoặc số tiền bán ra đủ 5 vạn tệ, thì tới ngưỡng phải kết án. Phạt tù dưới 3 năm, kèm phạt tiền, hoặc chỉ phạt tiền. Đạt gấp mười ngưỡng đó thì phạt tù 3 đến 10 năm kèm phạt tiền. Hàng còn nằm trong kho chưa bán cũng tính: giá trị hàng đạt từ ba lần tiêu chuẩn số tiền bán ra trở lên thì vẫn cấu thành tội. 「明知」không cần bạn tự thừa nhận. Giải thích của Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao năm 2025 liệt kê năm tình huống có thể kết luận trực tiếp rằng bạn đã biết, trừ khi bạn đưa được bằng chứng chứng minh thật sự không biết. Một là biết nhãn hiệu đăng ký trên hàng hóa đã bị sửa, đổi hoặc che. Hai là làm giả, sửa giấy uỷ quyền của chủ sở hữu nhãn hiệu, hoặc biết giấy đó bị làm giả, sửa. Ba là đã bị xử phạt hình sự hoặc hành chính vì bán hàng giả, rồi lại bán cùng loại. Bốn là không có lý do chính đáng mà mua hoặc bán với giá thấp hẳn so với giá thị trường. Năm là sau khi cơ quan quản lý hành chính hoặc cơ quan tư pháp phát hiện, chuyển hoặc tiêu huỷ hàng vi phạm cùng chứng từ kế toán, hoặc đưa ra chứng minh giả. Công ty phạm điều này thì「对单位判处罚金，并对其直接负责的主管人员和其他直接责任人员」xử phạt theo cùng điều — tức công ty nộp phạt tiền, người quyết định và người thực hiện bị xử theo cùng điều. Nên hàng do nhân viên mua, người bị phạt và bị kết án vẫn là người quyết định. Ngược lại, Luật Thương hiệu để lại một lối thoát: khi bán mà thật sự không biết hàng này xâm phạm nhãn hiệu đăng ký của người khác, thì「能证明该商品是自己合法取得并说明提供者的，不承担赔偿责任」— ai chứng minh được mình có hàng hóa đó một cách hợp pháp và nêu ra nhà cung cấp thì không phải bồi thường. Có được miễn bồi thường hay không, tất cả quyết định ở việc có chứng từ và thông tin bên bán không; không giữ thì không dùng được điều này (toàn quốc, giải thích có hiệu lực từ ngày 26 tháng 4 năm 2025). +- Chi phí: Không tốn tiền. Mỗi lô hàng đều giữ lại mấy thứ: hợp đồng mua, hoá đơn, chứng từ chuyển khoản, giấy phép kinh doanh và giấy uỷ quyền của nhà cung cấp. Rồi đặt một quy tắc nội bộ: ai mua hàng thì người đó ký, giá thấp hơn giá thị trường rõ rệt thì không mua. Khó ở chỗ phải bỏ qua mấy lô hàng rẻ trông rất có lãi. +- Nói thẳng: Bán hàng giả, thu được 3 vạn, hoặc bán đến mức 5 vạn, là đủ bị kết án, phạt tù từ 3 năm trở xuống. Hàng nằm trong kho chưa bán ra cũng tính.「Tôi không biết là hàng giả」không phải do bạn quyết định: giá nhập thấp hơn giá thị trường rõ rệt mà không nói được lý do thì bị coi là bạn biết. Giữ hợp đồng mua, hoá đơn, chứng từ chuyển khoản và thông tin nhà cung cấp thì mới có đường miễn bồi thường. +- Lợi ích: Biết là hàng giả nhãn hiệu đã đăng ký mà vẫn bán, số tiền thu được (thu nhập bất hợp pháp) đủ 3 vạn tệ, hoặc số tiền bán ra đủ 5 vạn tệ, thì đã tới ngưỡng phải kết án. Phạt tù từ 3 năm trở xuống, kèm phạt tiền, hoặc chỉ phạt tiền. Đạt mười lần mức này thì phạt tù từ 3 năm đến 10 năm và phạt tiền. Hàng vẫn còn trong kho chưa bán cũng tính: giá trị lô hàng đạt từ ba lần mức giá bán trở lên thì cũng cấu thành tội. Từ「biết」không cần bạn tự nhận. Bản giải thích tư pháp năm 2025 của Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao liệt kê năm tình huống có thể trực tiếp coi là bạn biết, trừ khi bạn đưa được bằng chứng chứng minh thật sự không biết. Một là biết nhãn hiệu đăng ký trên hàng đã bị sửa, đổi hoặc phủ. Hai là làm giả, sửa giấy uỷ quyền của chủ sở hữu nhãn hiệu, hoặc biết giấy đó bị làm giả, sửa. Ba là đã chịu xử lý hình sự hoặc xử phạt hành chính vì bán hàng giả, lại bán tiếp loại hàng đó. Bốn là không có lý do chính đáng mà mua hoặc bán với giá thấp hơn hẳn giá thị trường. Năm là sau khi cơ quan thực thi hành chính hoặc cơ quan tư pháp phát hiện, chuyển đi, huỷ hàng vi phạm và chứng từ kế toán, hoặc đưa ra chứng minh giả. Công ty phạm tội này thì「phạt tiền cho đơn vị, và xử phạt những người phụ trách trực tiếp cùng những người chịu trách nhiệm trực tiếp khác」theo cùng điều đó. Nghĩa là công ty bị phạt tiền, còn người quyết định và người thực hiện thì bị xử theo cùng điều. Vậy hàng do nhân viên mua, người bị phạt và bị kết án vẫn là người quyết định. Ngược lại, Luật Nhãn hiệu để lại một lối thoát: lúc bán mà thật sự không biết hàng đó vi phạm nhãn hiệu đã đăng ký của người khác,「có thể chứng minh mình có mua hàng đó một cách hợp pháp và nêu tên người cung cấp thì không phải chịu trách nhiệm bồi thường」. Có được miễn bồi thường hay không, tất cả tuỳ vào có hoá đơn và thông tin nhà cung cấp hay không, không giữ thì không dùng được điều này (toàn quốc, bản giải thích tư pháp có hiệu lực từ ngày 26 tháng 4 năm 2025) - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). 刑法修正案（十一）（第十七、二十四项，即刑法第二百一十四、二百二十条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 关于办理侵犯知识产权刑事案件适用法律若干问题的解释（法释〔2025〕5 号，2025 年 4 月 26 日起施行）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2019). 商标法（2019 年修正，第五十七、六十四条；2026 年 6 月 26 日修订后为第七十二、七十八条，自 2027 年 1 月 1 日起施行）. 、 -- Ghi chú: Mỹ phẩm, đồ uống, hàng mẹ và bé, hàng số, quần áo là những nhóm hàng hay xảy ra vụ này. Điểm chung là cùng một mã hàng có khoảng chênh giá rất lớn. Những cách gọi như「渠道货」「尾单」「海外免税」đều không thay thế được chứng từ. Ngoài kết án còn một mức xử phạt hành chính: khi xác định xâm phạm, hàng vi phạm và dụng cụ phải bị thu giữ, doanh thu kinh doanh phi pháp từ 5 vạn tệ trở lên còn có thể phạt kèm tối đa 5 lần. Đã nhận hàng mới phát hiện sai thì dừng bán ngay, niêm phong lại, rồi liên lạc hãng hoặc cơ quan quản lý thị trường. Đừng bán tiếp, đừng tiêu huỷ hàng và sổ sách. Tiêu huỷ chứng cứ rơi đúng vào tình huống thứ năm「bị suy đoán là đã biết hàng giả」nêu trên. Dùng hình ảnh của người khác, xem mục 21 (hình ảnh thì tự làm hoặc mua giấy phép). Có một kiểu chơi nhắm thẳng vào điều này: có người chủ động mang hàng tới, nói lô bút hay món nhỏ khác này cho bạn miễn phí, để trong cửa hàng bán rồi chia tiền. Vài ngày sau có người tới mua, lấy ngay bằng chứng tại chỗ, rồi đòi bồi thường với lý do xâm phạm. Hàng cho không, đủ ba điều kiện của đường miễn bồi thường đều thiếu: không có hợp đồng mua hàng và chứng từ thanh toán, không nói được nhà cung cấp thật, giá thấp bất thường. Điều「合法取得并说明提供者」không dùng được vào lúc nào cả. Nên hàng được mang tận nơi, không kèm chứng từ, chỉ để lại một tài khoản tin nhắn, thì tuyệt đối không đưa lên kệ. Định bán hộ thì ký hợp đồng bán hộ bằng văn bản, giữ bản sao giấy tờ tùy thân và giấy phép kinh doanh của bên kia, ghi rõ hàng thuộc ai, xâm phạm thì tính vào ai. +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). Bản sửa đổi (mười một) Bộ luật Hình sự（刑法修正案（十一））（第十七、二十四项，即刑法第二百一十四、二百二十条）. ；Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao (最高人民法院、最高人民检察院) (2025). Giải thích về một số vấn đề pháp lý khi xử lý các vụ án hình sự xâm phạm quyền sở hữu trí tuệ（法释〔2025〕5 号，2025 年 4 月 26 日起施行）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2019). Luật Nhãn hiệu（商标法）（2019 年修正，第五十七、六十四条；2026 年 6 月 26 日修订后为第七十二、七十八条，自 2027 年 1 月 1 日起施行）. 、 +- Ghi chú: Mỹ phẩm, đồ uống, hàng mẹ và bé, hàng điện tử, quần áo là những nhóm hàng hay xảy ra vụ việc loại này. Điểm chung là cùng một mã hàng có khoảng chênh giá rất lớn. Các nói「hàng kênh」「hàng tồn cuối」「hàng miễn thuế nước ngoài」đều không thay thế được hoá đơn. Ngoài phạt tù còn có một mức xử phạt hành chính: xác định được vi phạm thì tịch thu hàng vi phạm và công cụ, doanh thu vi phạm từ 5 vạn tệ trở lên còn có thể bị phạt kèm không quá 5 lần. Đã nhập hàng về mới phát hiện sai thì dừng bán ngay, niêm phong lại, rồi liên hệ hãng hoặc cơ quan quản lý thị trường. Đừng bán tiếp, đừng huỷ hàng và sổ sách. Huỷ chứng cứ đúng là rơi vào tình huống thứ năm「bị coi là bạn biết hàng giả」nêu trên. Dùng hình ảnh của người khác xem ở mục 21 của phần này (hình ảnh thì tự làm hoặc mua giấy uỷ quyền). Có một kiểu lừa đảo nhắm vào đúng điều này: có người chủ động mang đến, nói mẻ bút hay món nhỏ khác cho bạn miễn phí, đặt trong cửa hàng bán rồi chia tiền. Vài ngày sau có người đến mua, lấy bằng chứng tại chỗ, rồi đòi bồi thường vì vi phạm. Hàng cho không thì thiếu đúng cả ba điều kiện để được miễn bồi thường: không có hợp đồng mua và chứng từ chuyển khoản, không nói được nhà cung cấp thật, giá rõ ràng bất thường. Không mục「có mua hợp pháp và nêu tên nhà cung cấp」nào dùng được. Vì vậy ai chủ động giao hàng tới, không đưa hoá đơn, chỉ để lại một tài khoản WeChat thì đều không lên kệ. Thật sự muốn làm đại lý, hãy ký hợp đồng đại lý bằng văn bản, giữ lại bản sao căn cước và giấy phép kinh doanh của bên kia, ghi rõ hàng thuộc về ai, vi phạm thì tính vào ai. -### 21. Hình ảnh trên hàng hóa, bao bì, treo cổ và ảnh quảng bá thì hoặc tự làm hoặc mua giấy phép, đổi màu, thêm biểu tượng không tính là「đã sửa」 +### 21. Hình ảnh trên hàng, bao bì, treo thẻ và ảnh quảng cáo thì tự làm hoặc mua giấy uỷ quyền, đổi màu, thêm biểu tượng không tính là「đã sửa」 -- Chi phí: Giấy phép dùng ảnh từ thư viện chính hãng rẻ nhất vài chục tệ một ảnh. Thuê người thiết kế riêng, vài trăm đến vài nghìn tệ. Mỗi ảnh dùng trên hàng hóa đều giữ lại một bằng chứng cấp quyền và biên lai mua. Khó là không dùng ảnh tìm đại ý trên mạng. -- Nói thẳng: Đổi màu trên ảnh của người khác, đổi nền, thêm biểu tượng của bạn, vẫn là dùng tác phẩm của người khác mà không được phép. Sao chép tác phẩm nghệ thuật của người khác để kiếm tiền, thu được 3 vạn tệ, làm đến 5 vạn tệ, hoặc in đủ 500 bản, thì cấu thành tội vi phạm bản quyền, phạt tù dưới 3 năm. Với nhãn hiệu,「像到容易混淆」bản thân nó đã là xâm phạm. -- Lợi ích: Luật Bản quyền liệt kê「未经著作权人许可，以改编、翻译、注释等方式使用作品」cùng với việc sao chép trực tiếp là hành vi xâm phạm. Vậy nên sửa màu trên ảnh gốc, đổi nền, thêm biểu tượng của bạn vẫn là dùng tác phẩm của người khác không được phép. Ngưỡng phải kết án nằm ở đây: để kiếm tiền, không được chủ sở hữu bản quyền (chủ sở hữu quyền tác giả) cho phép mà in tác phẩm nghệ thuật ra bán (sao chép phát hành). Đủ một trong ba con số thì cấu thành tội. Thu nhập bất hợp pháp đủ 3 vạn tệ. Hoặc giao dịch này làm đến 5 vạn tệ (số tiền kinh doanh phi pháp). Hoặc tổng số bản in đủ 500 bản (tờ). Chạm một trong ba con số đó là cấu thành tội vi phạm bản quyền. Phạt tù dưới 3 năm, kèm phạt tiền, hoặc chỉ phạt tiền. Đạt gấp mười ngưỡng thì phạt tù 3 đến 10 năm. Chỉ bán hàng in sẵn xâm phạm của người khác, thu được từ 5 vạn tệ trở lên, hoặc bán ra từ 10 vạn tệ trở lên thì xử theo tội bán hàng xâm phạm bản sao có bản quyền, phạt tù dưới 5 năm. Nhãn hiệu cũng vậy: dùng nhãn hiệu gần giống nhãn hiệu đăng ký trên cùng một loại hàng hóa, hoặc dùng nhãn hiệu giống hay gần giống trên hàng hóa tương tự,「容易导致混淆的」bản thân đã nằm trong phạm vi xâm phạm. 「近似」là tình huống xâm phạm luật ghi rõ, sửa qua chút vẫn là xâm phạm (toàn quốc, giải thích có hiệu lực từ ngày 26 tháng 4 năm 2025). +- Chi phí: Giấy uỷ quyền trong thư viện ảnh chính hãng bắt đầu vài chục tệ một ảnh. Thuê người thiết kế riêng thì vài trăm đến vài nghìn tệ. Mỗi ảnh dùng trên hàng đều giữ lại một chứng từ uỷ quyền và biên la mua. Khó ở chỗ nhịn không dùng ảnh tìm được trên mạng. +- Nói thẳng: Sửa màu trên ảnh của người khác, đổi nền, thêm biểu tượng của mình, vẫn là dùng tác phẩm của người khác mà không được phép. Sao chép tác phẩm nghệ thuật của người khác để kiếm tiền, thu được 3 vạn, làm đến 5 vạn, hoặc in đủ 500 bản, thì cấu thành tội xâm phạm quyền tác giả, phạt tù từ 3 năm trở xuống. Về nhãn hiệu,「giống đến mức dễ gây nhầm lẫn」bản thân đã là vi phạm. +- Lợi ích: Luật Sở hữu trí tuệ đặt「sử dụng tác phẩm khi không được chủ sở hữu quyền tác giả cho phép, thông qua việc sửa đổi, dịch, chú thích」ngang hàng với sao chép trực tiếp là hành vi xâm phạm. Vậy sửa màu trên ảnh gốc, đổi nền, thêm biểu tượng của mình vẫn là dùng tác phẩm của người khác không được phép. Ngưỡng phải kết án nằm ở đây: để kiếm tiền, không được chủ sở hữu quyền tác giả của tác phẩm cho phép mà in tác phẩm nghệ thuật ra bán (sao chép phát hành). Đủ ba con số để thành tội. Số tiền thu được (thu nhập bất hợp pháp) đủ 3 vạn tệ. Hoặc mức kinh doanh phi pháp đạt 5 vạn tệ. Hoặc số bản in ra cộng dồn đủ 500 phần (bản). Chạm một trong ba con số đó là cấu thành tội xâm phạm quyền tác giả. Phạt tù từ 3 năm trở xuống, kèm phạt tiền, hoặc chỉ phạt tiền. Đạt gấp mười lần thì phạt tù từ 3 năm đến 10 năm. Chỉ bán hàng in sẵn có sẵn vi phạm, thu trên 5 vạn tệ, hoặc bán ra trên 10 vạn tệ, thì xử theo tội bán hàng làm bản sao vi phạm, phạt tù từ 5 năm trở xuống. Nhãn hiệu cũng vậy: dùng nhãn hiệu gần giống với nhãn hiệu đã đăng ký trên cùng loại hàng, hoặc dùng nhãn hiệu giống hay gần giống trên hàng tương tự,「dễ gây nhầm lẫn」đã nằm trong phạm vi xâm phạm.「Gần giống」là tình huống xâm phạm được luật ghi rõ, sửa một chút vẫn là vi phạm (toàn quốc, bản giải thích tư pháp có hiệu lực từ ngày 26 tháng 4 năm 2025) - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). 著作权法（2020 年修正，第五十二条第六项、第五十三条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). 刑法修正案（十一）（第二十、二十一项，即刑法第二百一十七、二百一十八条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 关于办理侵犯知识产权刑事案件适用法律若干问题的解释（法释〔2025〕5 号）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2019). 商标法（2019 年修正，第五十七条第二项；2026 年修订后为第七十二条第二项）. -- Ghi chú: Có ba chỗ hay vấp. Một là in hình nhân vật hoạt hình, hình ngôi sao tìm được trên mạng lên áo, cốc, ốp điện thoại. Hai là phông chữ: phông dùng thương mại phần lớn phải có giấy phép riêng, xem trước miễn phí không có nghĩa dùng thương mại miễn phí. Ba là bản thiết kế làm cho khách có lẫn tài nguyên chưa được cấp quyền, trách nhiệm vẫn thuộc về người bán hàng. Muốn gọn thì dùng thư viện ảnh có giấy phép thương mại, và lưu lại trang cấp quyền. Hoặc bắt bên thiết kế ghi rõ trong hợp đồng tài nguyên lấy từ đâu, cấp quyền tới mức nào, có sự cố thì bên thiết kế chịu trách nhiệm. Ngoài kết án còn có bồi thường dân sự và phạt hành chính: gây tổn hại lợi ích công cộng thì có thể thu giữ thu nhập bất hợp pháp, và phạt gấp 1 đến 5 lần doanh thu kinh doanh phi pháp. Trước khi đưa vào sản xuất thì tra nhãn hiệu, xem mục 19 (làm xong mẫu thì so qua danh sách sản xuất hàng loạt). Ngược lại, hình ảnh của bạn bị người khác dùng thì cũng theo bộ quy tắc này mà bảo vệ. +- Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). Luật Bảo hộ quyền tác giả（著作权法）（2020 年修正，第五十二条第六项、第五十三条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). Bản sửa đổi (mười một) Bộ luật Hình sự（刑法修正案（十一））（第二十、二十一项，即刑法第二百一十七、二百一十八条）. ；Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao (最高人民法院、最高人民检察院) (2025). Giải thích về một số vấn đề pháp lý khi xử lý các vụ án hình sự xâm phạm quyền sở hữu trí tuệ（法释〔2025〕5 号）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2019). Luật Nhãn hiệu（商标法）（2019 年修正，第五十七条第二项；2026 年修订后为第七十二条第二项）. +- Ghi chú: Có ba chỗ hay vấp. Một là in hình nhân vật hoạt hình, hình người nổi tiếng tìm được trên mạng lên quần áo, cốc, vỏ điện thoại. Hai là phông chữ: phông dùng thương mại phần lớn phải có giấy uỷ quyền riêng, xem trước miễn phí không có nghĩa dùng thương mại miễn phí. Ba là bản thiết kế làm cho khách có lẫn tư liệu chưa được uỷ quyền, trách nhiệm vẫn thuộc về người bán hàng. Muốn làm gọn thì dùng thư viện ảnh có giấy uỷ quyền thương mại, và lưu lại trang giấy uỷ quyền. Hoặc bắt bên thiết kế ghi rõ trong hợp đồng tư liệu lấy từ đâu, uỷ quyền trong phạm vi nào, có sự cố thì bên thiết kế chịu trách nhiệm. Ngoài phạt tù còn có bồi thường dân sự và phạt hành chính: gây tổn hại lợi ích công cộng thì có thể tịch thu thu nhập bất hợp pháp, và phạt gấp từ 1 đến 5 lần doanh thu vi phạm. Trước khi sản xuất, tra nhãn hiệu xem ở mục 19 của phần này (làm xong mẫu thì so qua danh sách sản xuất hàng loạt). Ngược lại, hình ảnh của bạn bị người khác dùng thì cũng theo bộ quy tắc này mà bảo vệ quyền lợi. -### 22. Ở đúng vị trí của mình thì làm theo quy chế, ý tốt ngoài quy chế thì nghĩ trước trách nhiệm thuộc về ai +### 22. Ở đúng vị trí thì làm theo quy trình, còn cái thiện ý ngoài quy trình thì nghĩ kỹ trách nhiệm thuộc về ai -- Chi phí: Không tốn tiền. Khi làm việc thì theo quy chế, theo y lệnh, theo quy trình. Không quyết thay người khác khi vượt quá thẩm quyền của mình. Trước khi giúp, tự hỏi một câu: có việc thì ai chịu trách nhiệm. Khó là chống lại yêu cầu「linh động một chút」tại chỗ. -- Nói thẳng: Lấy y tá làm ví dụ: nghị định ghi rõ, phát hiện bệnh nhân chuyển biến nguy kịch thì phải báo bác sĩ ngay; phát hiện y lệnh vi phạm quy định thì phải nêu với bác sĩ đã ra y lệnh. Mọi vị trí khác cũng vậy, trước hết phải thấy quy chế của mình đánh dấu「phải báo cáo」ở đâu. Pháp luật cũng không bắt người ta nhìn người sắp chết mà không cứu: tự nguyện ra cứu khẩn cấp mà làm người được cứu bị thương thì không phải bồi tiền. -- Lợi ích: Lấy y tá làm ví dụ, quy định ghi rõ giới hạn nhiệm vụ như sau. Phát hiện bệnh chuyển biến nguy kịch thì báo bác sĩ ngay. Trong tình huống khẩn cấp, để cứu người bệnh sắp tử vong thì có thể tiến hành cấp cứu khẩn cấp cần thiết trước. Phát hiện y lệnh vi phạm pháp luật, hoặc vi phạm quy chuẩn kỹ thuật chẩn đoán trị bệnh thì phải nêu với bác sĩ, cần thì báo trưởng khoa. Bộ luật Dân sự có quy định khác: tự nguyện ra cứu khẩn cấp mà khiến người được cứu bị thương thì bên cứu không phải bồi tiền (không chịu trách nhiệm dân sự) (toàn quốc). +- Chi phí: Không tốn tiền. Khi làm việc thì theo quy định, theo chỉ định của bác sĩ, theo quy trình. Không thay người khác quyết định việc vượt ngoài thẩm quyền của mình. Trước khi giúp, tự hỏi một câu: có chuyện thì ai chịu trách nhiệm. Khó ở chỗ phải từ chối tận thời lời「linh động một chút». +- Nói thẳng: Lấy y tá làm ví dụ: quy định ghi rõ, phát hiện bệnh nhân lâm nguy thì phải báo bác sĩ ngay; phát hiện chỉ định điều trị vi phạm quy định thì phải nêu với bác sĩ đã kê đơn. Các vị trí khác cũng vậy, trước hết phải xem quy trình của mình đánh dấu「phải báo cáo」ở đâu. Pháp luật cũng không bắt phải nhìn người sắp chết mà không cứu: tự nguyện ra cứu gấp mà làm người được cứu bị thương thì không phải bồi thường tiền. +- Lợi ích: Lấy y tá làm ví dụ, quy định ghi rõ ranh giới công vụ như sau. Phát hiện tình hình bệnh lâm nguy thì báo bác sĩ ngay. Trong tình huống khẩn cấp để cứu bệnh nhân nguy kịch, có thể thực hiện trước các biện pháp cấp cứu cần thiết. Phát hiện chỉ định điều trị vi phạm pháp luật, hoặc vi phạm quy chuẩn kỹ thuật khám chữa bệnh thì phải nêu với bác sĩ, cần thì báo trưởng khoa. Bộ luật Dân sự có quy định khác: tự nguyện ra cứu gấp mà làm người được cứu bị thương thì bên cứu không phải bồi thường tiền (không chịu trách nhiệm dân sự) (toàn quốc) - Mức bằng chứng: B -- Nguồn: Quốc vụ viện (国务院) (2008). 护士条例（国务院令第 517 号，第十七条）. ；Quốc vụ viện (国务院) (2020). 护士条例（2020 年修订，第十七条）. （Cơ sở dữ liệu luật pháp quốc gia）；Quốc hội Nhân dân (全国人大) (2020). 民法典（第一百八十四条）. -- Ghi chú: Ở trong phạm vi công việc thì làm theo quy chế, cứu người trong tình huống khẩn cấp thì được pháp luật bảo vệ. Còn những việc khác, trước hết phải phân biệt đó là thẩm quyền của ai, trách nhiệm của ai, đừng dùng thiện ý riêng thay cho quyết định của hệ thống. Nghị định về y tá năm 2020 được sửa đổi bằng lệnh của Quốc vụ viện số 726, chỉ sửa mấy điều về đăng ký, Điều 17 không đổi một chữ. Điều trong Luật Bác sĩ về việc hành nghề theo quy chuẩn chẩn đoán trị bệnh, lượt này không mở được nguyên văn chính thức nên không dẫn. +- Nguồn: Quốc vụ viện (国务院) (2008). Quy định về y tá（护士条例）（国务院令第 517 号，第十七条）. ；Quốc vụ viện (国务院) (2020). Quy định về y tá（护士条例）（2020 年修订，第十七条）. （Cơ sở dữ liệu luật và quy phạm quốc gia）；Đại biểu Quốc hội (全国人大) (2020). Bộ luật Dân sự（第一百八十四条）. +- Ghi chú: Trong phạm vi công vụ thì làm theo quy trình, cứu người trong tình huống khẩn cấp được pháp luật bảo vệ. Còn những việc khác, trước hết phải phân định đó là quyền của ai, trách nhiệm của ai, đừng dùng thiện ý riêng của mình thay cho quyết định mang tính quy trình. Quy định về y tá năm 2020 được sửa đổi bằng Quy định của Quốc vụ viện số 726, chỉ sửa mấy điều về đăng ký, Điều 17 không đổi một chữ. Các điều về việc hành nghề theo quy chuẩn khám chữa bệnh trong Luật Y khoa, lần này không mở được nguyên văn chính thức nên không dẫn. -### 23. Lỗ thua thì rút lui theo thủ tục: hủy được theo lối giản dị thì hủy, vốn không đủ trả nợ thì đi thủ tục phá sản, đừng để yên +### 23. Lỗ thì rút lui theo thủ tục: huỷ được kiểu rút gọn thì huỷ, vỡ nợ thì đi phá sản, đừng để yên -- Chi phí: Tự làm thủ tục hủy đăng ký giản dị được: đăng lên Hệ thống công bố thông tin tín dụng doanh nghiệp quốc gia 20 ngày. Còn nợ chưa trả xong thì phải làm rõ sổ sách trước, thu cái phải thu, trả cái phải trả, bước này gọi là thanh lý. Đi thủ tục phá sản thì phải xin tòa án, mất mấy tháng. Khó là lỗ thua rồi không muốn đối mặt, kéo dài không làm. -- Nói thẳng: Không nợ ai, không nợ lương, không nợ thuế, không nợ bảo hiểm xã hội, thì toàn thể nhà đầu tư viết một bản cam kết, công bố 20 ngày không có ai phản đối thì hủy đăng ký giản dị được. Hộ kinh doanh cá thể không phải công bố, 10 ngày không có phản đối thì hủy thẳng. Không trả nổi nợ thì xin phá sản. Để yên, 2 năm không báo cáo thường niên lại không liên lạc được thì bị thu hồi giấy phép, người đại diện theo pháp luật có trách nhiệm không được làm người đại diện trong 3 năm. -- Lợi ích: Không có nợ của mình, và cũng không ai nợ mình, hoặc đã trả xong hết. Không nợ lương, không nợ bảo hiểm xã hội, không nợ thuế. Toàn thể nhà đầu tư đã viết cam kết bằng văn bản. Đủ mấy điều đó, đăng công bố 20 ngày không có ai phản đối thì đi được thủ tục hủy đăng ký giản dị. Hộ kinh doanh cá thể không phải công bố, 10 ngày không có ý kiến phản đối của cơ quan liên quan thì hủy thẳng. Công ty không trả nổi khoản nợ đến hạn, số nợ lớn hơn số tài sản đang có (vốn không đủ trả nợ) thì có thể xin tòa án mở thủ tục tái cơ cấu, thỏa thuận hoặc phá sản thanh lý. Công ty đã giải thể mà chưa thanh lý, số nợ lớn hơn số tài sản, thì người phụ trách thanh lý phải xin thủ tục phá sản thanh lý. Hậu quả của việc để yên là: không công bố báo cáo thường niên đúng hạn thì bị đưa vào danh sách bất thường trong hoạt động. Liên tiếp 2 năm không báo cáo thường niên, đã vào danh sách bất thường mà vẫn không sửa, lại không liên lạc được theo địa chỉ đã đăng ký thì bị thu hồi giấy phép kinh doanh. Người làm đại diện theo pháp luật của doanh nghiệp bị thu hồi giấy phép và chịu trách nhiệm cá nhân thì, kể từ ngày bị thu hồi, 3 năm không được làm đại diện theo pháp luật, giám đốc, giám sát, lãnh đạo cấp cao của công ty. Đã vào danh sách bất thường trong hoạt động thì ngay cả thủ tục hủy đăng ký giản dị cũng không đi được (toàn quốc). +- Chi phí: Huỷ kiểu rút gọn bạn tự làm được: niêm yết trên Hệ thống công bố thông tin tín dụng doanh nghiệp quốc gia trong 20 ngày. Còn nợ chưa trả hết thì phải làm rõ sổ sách trước, cái phải thu thì thu, cái phải trả thì trả, bước này gọi là thanh lễ. Đi phá sản thì nộp đơn xin tòa, mất mấy tháng. Khó ở chỗ lỗ rồi không muốn đối mặt, cứ kéo dài không làm. +- Nói thẳng: Không nợ ai, không nợ lương, thuế, bảo hiểm xã hội, thì các nhà đầu tư cùng viết một bản cam kết, niêm yết 20 ngày không ai phản đối là huỷ được kiểu rút gọn. Hộ cá thể không phải niêm yết, 10 ngày không phản đối thì huỷ luôn. Không trả nổi nợ thì nộp đơn xin phá sản. Để yên, 2 năm không nộp báo cáo thường niên mà liên lạc không được, thì bị thu hồi giấy phép, người đại diện theo pháp luật có trách nhiệm không được làm người đại diện theo pháp luật trong 3 năm. +- Lợi ích: Không nợ người khác và người khác không nợ mình, hoặc đã trả hết. Không nợ lương, không nợ bảo hiểm xã hội, không nợ thuế. Các nhà đầu tư đã viết cam kết bằng văn bản. Đủ mấy điều đó thì niêm yết 20 ngày không ai phản đối là đi được thủ tục huỷ kiểu rút gọn. Hộ cá thể không phải niêm yết, cơ quan có liên quan trong 10 ngày không phản đối thì huỷ luôn. Công ty không trả nổi khoản nợ đến hạn mà số nợ nhiều hơn số tài sản có (vỡ nợ), có thể xin tòa tái cơ cấu, thỏa thuận hoặc thanh lễ phá sản. Đã giải thể mà chưa thanh lễ, nợ nhiều hơn tài sản có, thì người phụ trách thanh lễ nên nộp đơn xin thanh lễ phá sản. Hậu quả khi để yên thì thế này: không niêm yết báo cáo thường niên đúng hạn thì bị đưa vào danh mục bất thường kinh doanh. Liên tục 2 năm không nộp báo cáo thường niên, đã vào danh mục bất thường kinh doanh mà không sửa, lại không liên lạc được ở địa chỉ đã đăng ký, thì bị thu hồi giấy phép kinh doanh. Người làm người đại diện theo pháp luật của doanh nghiệp bị thu hồi giấy phép và chịu trách nhiệm cá nhân, kể từ ngày bị thu hồi giấy phép trong 3 năm không được làm người đại diện theo pháp luật, giám đốc, giám sát, cán bộ quản lý của công ty. Đã vào danh mục bất thường kinh doanh thì kể cả huỷ kiểu rút gọn cũng không đi được (toàn quốc) - Mức bằng chứng: A -- Nguồn: Quốc vụ viện (国务院) (2021). 市场主体登记管理条例（国务院令第 746 号，第三十一、三十二、三十三条）. ；Sáu bộ ngành gồm Cục Quản lý thị trường quốc gia (市场监管总局) v.v. (2025). 企业注销指引（2025 年修订）（2025 年第 52 号公告）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2006). 企业破产法（第二、七条）. ；Quốc vụ viện (国务院) (2024 修订). 企业信息公示暂行条例（国务院令第 777 号修订，第十八条）：「企业因连续2年未按规定报送年度报告被列入经营异常名录未改正，且通过登记的住所或者经营场所无法取得联系的，由县级以上市场监督管理部门吊销营业执照」. （Cơ sở dữ liệu luật pháp quốc gia）；Quốc vụ viện (国务院) (2024). 关于修改和废止部分行政法规的决定（国务院令第 777 号）. ；市场主体登记管理条例（第十二条第四项）与公司法（2023 年修订，第一百七十八条第一款第四项）：担任因违法被吊销营业执照的企业的法定代表人并负有个人责任的，自被吊销营业执照之日起未逾 3 年不得担任法定代表人和董事、监事、高级管理人员. ； -- Ghi chú: Nghị định tạm về công bố thông tin doanh nghiệp năm 2024 được sửa bằng lệnh của Quốc vụ viện số 777, có hiệu lực từ ngày 1 tháng 5, từ đó không sửa thêm. Điều 17 của bản cũ「经营异常满 3 年列入严重违法企业名单」đã bị xóa. Danh sách doanh nghiệp vi phạm nghiêm trọng, mất uy tín hiện nay quản lý doanh nghiệp báo cáo thường niên gian dối, tình tiết nặng. Trên mạng rất nhiều lời nói vẫn theo bản cũ, đừng sao chép. Chế độ phá sản cá nhân (kể cả hộ kinh doanh cá thể) hiện mới chỉ có vài địa phương thí điểm, phần này chưa kiểm chứng nguyên văn nên không viết. +- Nguồn: Quốc vụ viện (国务院) (2021). Quy định về đăng ký chủ thể kinh doanh（市场主体登记管理条例）（国务院令第 746 号，第三十一、三十二、三十三条）. ；Cục Quản lý thị trường quốc gia và năm cơ quan khác (市场监管总局等六部门) (2025). Cẩm nang huỷ doanh nghiệp (bản sửa đổi năm 2025)（2025 年第 52 号公告）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2006). Luật Phá sản doanh nghiệp（企业破产法）（第二、七条）. ；Quốc vụ viện (国务院) (2024, sửa đổi). Quy chế tạm thời về công khai thông tin doanh nghiệp（企业信息公示暂行条例）（国务院令第 777 号修订，第十八条）：「企业因连续2年未按规定报送年度报告被列入经营异常名录未改正，且通过登记的住所或者经营场所无法取得联系的，由县级以上市场监督管理部门吊销营业执照」. （Cơ sở dữ liệu luật và quy phạm quốc gia）；Quốc vụ viện (国务院) (2024). Quyết định sửa đổi và bãi bỏ một số quy chế hành chính（国务院令第 777 号）. ；Quy định về đăng ký chủ thể kinh doanh（第十二条第四项）与 Luật Công ty (2023, sửa đổi, Điều một trăm bảy mươi tám khoản một điểm bốn): người làm người đại diện theo pháp luật của doanh nghiệp bị thu hồi giấy phép kinh doanh vì vi phạm pháp luật mà chịu trách nhiệm cá nhân, kể từ ngày bị thu hồi giấy phép kinh doanh chưa quá 3 năm thì không được làm người đại diện theo pháp luật, giám đốc, giám sát, cán bộ quản lý cấp cao. ； +- Ghi chú: Quy chế tạm thời về công khai thông tin doanh nghiệp năm 2024 được sửa đổi bằng Quy định của Quốc vụ viện số 777, có hiệu lực từ ngày 1 tháng 5, sau đó không sửa thêm. Điều 17 của bản cũ「经营异常满 3 年列入严重违法企业名单」đã bị bỏ. Danh sách doanh nghiệp vi phạm nghiêm trọng, mất uy tín hiện nay quản lý doanh nghiệp bịa báo cáo thường niên, tình tiết nặng. Nhiều cách nói trên mạng vẫn là bản cũ, đừng sao chép. Chế độ phá sản cá nhân (kể cả hộ cá thể) hiện mới có ở vài địa phương thí điểm, phần này không đối chiếu nguyên văn nên không ghi. \ No newline at end of file diff --git "a/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" index 9ad619bb..6ca98bbf 100644 --- "a/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" +++ "b/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -1,6 +1,6 @@ [← Về mục lục](../README.md) -# 16. Sống sau khi mắc bệnh mạn tính +# 16. Sống với bệnh mãn tính Cột đo của phần này là tỷ lệ tử vong chung và tiền. Sau khi phát hiện bệnh mạn tính thường còn phải sống hàng chục năm, phần này tính mấy chục năm ấy nên quản lý thế nào. diff --git "a/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" "b/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" index 7ffa1cd1..68e07118 100644 --- "a/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" +++ "b/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" @@ -2,7 +2,7 @@ # 19. Đi làm, nghỉ việc và tai nạn lao động -Phần này tính tiền: tiền thế nào được tính, chữ ký thế nào, hạn thế nào. Ba mục đầu nói về phần còn giữ được của người đang làm việc, mục 4 đến 8 nói về cách nghỉ việc cho đúng, từ mục 9 trở đi nói về tai nạn lao động. +Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký thế nào, thời hạn tính thế nào. Ba mục đầu nói về số tiền bạn phải có trong tay khi còn đang làm việc. Sáu mục giữa nói về việc nghỉ việc. Bảy mục cuối nói về bệnh nghề nghiệp và tai nạn lao động. Tiền của tai nạn lao động lớn hơn tiền của sa thải một bậc, thời hạn cũng cứng hơn. Tổn thương do bệnh nghề nghiệp thì không hoàn lại được. ### 1. Tiền làm thêm giờ tính theo ba mức 1.5 lần, 2 lần, 3 lần; không trả thì báo thanh tra lao động, quá hạn không trả còn phải cộng thêm 50% đến 100% diff --git "a/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" "b/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" new file mode 100644 index 00000000..9187388e --- /dev/null +++ "b/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" @@ -0,0 +1,209 @@ +[← Về mục lục](../README.md) + +# 23. Học kỹ năng nào thì đáng + +Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ tử vong. Sáu mục đầu trả lời một câu hỏi: có nên đi học tiếp không. Mục 1 nói trước luật đã gạch bỏ lựa chọn nào, mục 2 tính quan hệ giữa đi học và tuổi thọ, mục 3 xem trình độ học vấn của dân chúng cả nước nằm ở mức nào. Mục 4 nói khi không đủ tiền đi học thì có những khoản hỗ trợ nào, mục 5 nói không đỗ được trường phổ thông thì còn đường nào, mục 6 biến câu hỏi「học hay đi làm」thành một bài toán và đưa cách tự tính. Bảy mục sau trả lời một câu hỏi khác: đã quyết định học một thứ thì chọn thế nào. Thống kê lương theo ngành nghề ở Trung Quốc rất khó đối chiếu từng nghề qua các kênh chính thức công khai. Nên ở đây chỉ ghi những điều tra cứu được nguồn: đọc thêm một năm thì lương hơn khoảng bao nhiêu, còn gọi là mức hoàn lại của việc học, kiểm tra chứng chỉ có hợp pháp không bằng cách nào, trợ cấp đào tạo lấy từ đâu, và khi chọn kỹ năng thì xem chiều nào ít bị thay thế hơn. Riêng một nghề cụ thể kiếm được bao nhiêu trong một tháng, cuốn sách này không đưa ra con số. Từ mục 14 đến mục 19 trả lời câu hỏi thứ ba: đã chốt học cái gì thì học thế nào cho tiết kiệm thời gian. Sáu mục đó tính thời gian, không quy đổi sang tiền. Từ mục 20 đến mục 23 trả lời câu hỏi thứ tư: đang làm công việc chuyên môn thì chức danh nghề nghiệp khai ở đâu, thi như thế nào, tránh bẫy nào, và lấy được chức danh thì đáng bao nhiêu. + +### 1. Dưới 16 tuổi không có lựa chọn「đi làm công ăn」: đơn vị thuê bạn bị phạt 5000 tệ mỗi tháng, đơn vị nào vẫn thuê thì là việc chui + +- Chi phí: Không tốn tiền. Tốn một phút để tính xem mình bao nhiêu tuổi. +- Nói thẳng: Học xong phổ thông cơ sở thường là 15 đến 16 tuổi, vừa nằm ngay trên ranh giới này. Dưới 16 tuổi mà đơn vị thuê bạn là vi phạm pháp, mỗi lao động trẻ em dùng một người bị phạt 5000 tệ mỗi tháng. Nên đơn vị nào vẫn lấy bạn là đơn vị biết rõ vi phạm mà vẫn dùng. Lỡ có chuyện thì tiền khám chữa và tiền bồi thường chỉ có thể đòi chính đơn vị đó, nó cố không trả hoặc bỏ chạy thì không ai đứng ra đỡ thay. +- Lợi ích: Luật Giáo dục bắt buộc (义务教育法) quy định ở Điều mười một:「凡年满六周岁的儿童，其父母或者其他法定监护人应当送其入学接受并完成义务教育」. Điều mười bốn:「禁止用人单位招用应当接受义务教育的适龄儿童、少年」. Điều năm mươi tám: trẻ em đủ tuổi đi học mà không có lý do chính đáng, không cho đi học đúng quy định luật này, thì chính quyền xã hoặc giáo dục của chính quyền huyện địa phương「给予批评教育，责令限期改正」. Quy định cấm sử dụng lao động trẻ em (禁止使用童工规定) Điều hai: cơ quan nhà nước, tổ chức xã hội, doanh nghiệp và đơn vị sự nghiệp, đơn vị phi lợi nhuận tư nhân, hay hộ kinh doanh cá thể「均不得招用不满16周岁的未成年人」. Điều sáu:「按照每使用一名童工每月处5000元罚款的标准给予处罚」, quá hạn không sửa thì phạt mỗi người mỗi tháng 1 vạn tệ. Điều bảy: giới thiệu lao động trẻ em đi làm thì「按照每介绍一人处5000元罚款」. Điều mười: lao động trẻ em ốm hay bị thương thì người sử dụng lao động「应当负责送到医疗机构治疗，并负担治疗期间的全部医疗和生活费用」, thương tật hoặc tử vong thì còn phải bồi thường một lần. Điều mười ba chỉ chừa một ngoại lệ: đơn vị văn hoá và thể thao, có sự đồng ý của cha mẹ hoặc người giám hộ khác, được thuê người làm nghệ thuật chuyên nghiệp và vận động viên dưới 16 tuổi (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 义务教育法（2018 年第二次修正，第十一、十四、五十八条）. （Bộ Giáo dục đăng lại）；Ủy ban Nhà nước (国务院) (2002). 禁止使用童工规定（国务院令第 364 号，第二、六、七、十、十三条）. +- Ghi chú: Người hưởng lợi là bạn. Sau 16 tuổi thì đi làm công là hợp pháp. Nhưng những chỗ nhận một người 16 tuổi không học vấn đúng là loại mục 10 của phần này nói tới (chọn kỹ năng chống bị máy thay), quy trình cố định, dễ bị máy đẩy ra nhất. Tiền làm thêm giờ, nghỉ phép năm, ưu đãi tai nạn lao động mà phần 19 nói tới, đều đặt điều kiện là có quan hệ lao động và có bằng chứng. Việc chui không có cả hai, nên số tiền đó cũng không lấy được. + +### 2. Câu hỏi đi học có ích không cũng nên tính vào sổ tử vong: mỗi năm học thêm, nguy cơ tử vong ở tuổi trưởng thành giảm khoảng 1.9% + +- Chi phí: Không tốn tiền. Cái giá phải trả là thời gian và sức lực của mấy năm học thêm, phần học phí xem mục 4 của phần này (học phí và học bổng trường trung cấp chuyên nghiệp). Chỗ khó là phải học liên tục, không bỏ giữa chừng. +- Nói thẳng: Mỗi năm học thêm, nguy cơ tử vong sau khi thành người lớn trung bình thấp hơn khoảng 1.9%. Rõ nhất là người từ 18 đến 49 tuổi, mỗi năm thêm thì thấp hơn khoảng 2.9%; người trên 70 tuổi chỉ thấp hơn 0.8%. Người đọc được suốt thường vốn đã có điều kiện tốt, trong đó bao nhiêu là do chính việc học thì không tách được. +- Lợi ích: Gộp lại toàn bộ các nghiên cứu trước đây rồi tính chung (tổng quan hệ thống và phân tích tổng hợp). Khoảng tra cứu từ tháng 1 năm 1980 đến tháng 5 năm 2023, cuối cùng chọn 603 nghiên cứu, 59 nước, 70 địa điểm, 10355 quan sát. Giáo dục và tỷ lệ tử vong ở tuổi trưởng thành (tử vong tính chung mọi nguyên nhân) có quan hệ đáp lượng, đọc càng nhiều thì giảm càng nhiều. Mỗi năm học thêm, nguy cơ tử vong giảm trung bình 1.9% (95% khoảng không chắc chắn 1.8–2.0, giá trị thật nằm gần trong khoảng đó). Người từ 18 đến 49 tuổi mỗi năm thêm giảm 2.9% (2.8–3.0). Người trên 70 tuổi giảm 0.8% (0.6–1.0). Không thấy khác biệt giữa nam và nữ. Xét theo mức phát triển của từng nước (chỉ số dân số xã hội) cũng không thấy khác biệt +- Mức bằng chứng: A (Tranh cãi) +- Nguồn: IHME-CHAIN Collaborators (2024). Effects of education on adult mortality: a global systematic review and meta-analysis. The Lancet Public Health, 9(3), e155-e165. +- Ghi chú: Tranh cãi. Chính các tác giả cũng báo cáo thiên lệch công bố (p<0.0001), nghĩa là nghiên cứu cho kết quả đẹp dễ được đăng hơn. Kết quả giữa các nghiên cứu cũng chênh nhau khá xa. Các nghiên cứu được chọn đều là theo dõi quan sát, không phải thử nghiệm chia nhóm. Số năm đi học bị trộn lẫn với điều kiện gia đình, địa phương, nghề nghiệp, nên không tách được trong 1.9% đó bao nhiêu do chính việc học và bao nhiêu do「người đọc được suốt vốn đã có điều kiện tốt」. Vì vậy nó nói khá chắc rằng học nhiều hơn và sống lâu hơn là hai việc cùng hướng, chứ không có nghĩa bạn đi học là sẽ kéo dài 1.9%. Người hưởng lợi là bạn. + +### 3. Trước khi kết luận học vấn đã mất giá, xem cơ cấu học vấn cả nước: cứ 100.000 người thì chỉ 15467 người học trình độ đại học + +- Chi phí: Không tốn tiền. Chỉ cần nhìn các con số bên dưới. +- Nói thẳng: Kết quả tổng điều tra dân số năm 2020 là: cứ 100.000 người ở cả nước thì chỉ 15467 người học tới cao đẳng trở lên, khoảng 15%. Người học hết phổ thông cơ sở có 34507 người, là nhóm đông nhất. Người từ 15 tuổi trở lên bình quân mới học 9.91 năm, vừa qua mốc tốt nghiệp phổ thông cơ sở. +- Lợi ích: Mốc thời gian chuẩn của tổng điều tra dân số lần thứ bảy là ngày 1 tháng 11 năm 2020. Trong dân số cả nước, người có trình độ đại học (cao đẳng trở lên) có 218360767 người. Trung cấp (kể cả trường trung cấp chuyên nghiệp) 213005258 người. Phổ thông cơ sở 487163489 người. Tiểu học 349658828 người. So với tổng điều tra dân số lần thứ sáu năm 2010, cứ 100.000 người thì số người có trình độ đại học tăng từ 8930 người lên 15467 người. Trung cấp tăng từ 14032 người lên 15088 người. Phổ thông cơ sở giảm từ 38788 người xuống 34507 người. Tiểu học giảm từ 26779 người xuống 24767 người. Số năm học trung bình của người từ 15 tuổi trở lên tăng từ 9.08 năm lên 9.91 năm. Tỷ lệ mù chữ giảm từ 4.08% xuống 2.67% +- Mức bằng chứng: A +- Nguồn: Cục Thống kê Quốc gia, Văn phòng Ban Chỉ đạo Tổng điều tra dân số lần thứ bảy của Hội đồng Nhà nước (2021). 第七次全国人口普查公报（第六号）——人口受教育情况. +- Ghi chú: Người hưởng lợi là bạn. Các con số này chỉ nói học vấn có hiếm hay dày trong dân chúng. Với một người cụ thể thì đi học đáng không, phải tự tính theo cách làm ở mục 6 của phần này (học hay đi làm thì tính thế nào). Còn phải để ý: tổng điều tra đếm cả toàn bộ dân số, kể cả người già và trẻ em. Tỷ lệ người trẻ học tới đại học cao hơn con số này nhiều, so với người cùng tuổi sẽ thổi phóng lợi thế của mình. Ngược lại, dùng nó để bác câu「đi học vô ích」thì hợp lý, vì mẫu số chính là cả lực lượng lao động đang giành chỗ với bạn. Người hưởng lợi là bạn. + +### 4. Trường hợp「không đủ tiền để đi học」thì tính theo chính sách trước đã: học phí trường trung cấp chuyên nghiệp đã miễn phần lớn, học bổng 2300 tệ, học vay tối đa 2 vạn tệ một năm + +- Chi phí: Không tốn tiền. Trước khi khai giảng, hỏi một lần bộ phận tài trợ học sinh của trường, nộp một lần hồ sơ. +- Nói thẳng: Trong trường trung cấp chuyên nghiệp công lập, học sinh nông thôn, học sinh học chuyên ngành liên quan nông nghiệp ở thành phố, và học sinh gia đình khó khăn đã được miễn học phí. Học bổng của trường trung cấp chuyên nghiệp và trường phổ thông trung học, từ học kỳ xuân năm 2025 là 2300 tệ một năm. Lên đại học thì học bổng là 3700 tệ một năm, học vay tối đa 2 vạn tệ một năm. Thiếu tiền thì có thể trì hoãn, cũng có thể tìm cách khác, hiếm khi đến mức phải bỏ học đi làm ngay lúc này. +- Lợi ích: Chính sách miễn học phí trường trung cấp chuyên nghiệp bắt đầu từ học kỳ thu năm 2012. Nguyên văn là「对公办中等职业学校全日制正式学籍一、二、三年级在校生中所有农村（含县镇）学生、城市涉农专业学生和家庭经济困难学生免除学费（艺术类相关表演专业学生除外）」. Năm 2024 có điều chỉnh chính sách tài trợ học sinh. Học bổng quốc gia cho học sinh đại học tăng từ 3300 tế lên 3700 tệ mỗi người mỗi năm. Học bổng khuyến khích tài năng của nhà nước tăng từ 5000 tệ lên 6000 tệ. Hai khoản này thực hiện từ học kỳ thu năm 2024 (học bổng khuyến khích tài năng áp dụng từ năm 2024). Học bổng quốc gia của trường trung cấp chuyên nghiệp và của trường phổ thông trung học đều tăng từ 2000 tệ lên 2300 tệ mỗi người mỗi năm, thực hiện từ học kỳ xuân năm 2025. Học vay quốc gia cũng được điều chỉnh, từ học kỳ thu năm 2024. Học sinh đại học và cao đẳng「每人每年申请贷款额度由不超过16000元提高至不超过20000元」. Học sinh sau đại học「由不超过20000元提高至不超过25000元」. Lãi suất bằng lãi suất định giá tham chiếu của thị trường vay cùng kỳ cùng bậc (LPR) trừ 70 điểm cơ bản, tức thấp hơn 0.7 điểm phần trăm (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Tài chính, Ủy ban Phát triển và Cải cách, Bộ Giáo dục, Bộ Nhân sự và Xã hội (2012). 关于扩大中等职业教育免学费政策范围 进一步完善国家助学金制度的意见（财教〔2012〕376 号）. ；Các cục chủ trì của Bộ Tài chính, Bộ Giáo dục, Bộ Nhân sự và Xã hội (2024). 就调整学生资助政策有关问题答记者问. ；Bộ Tài chính, Bộ Giáo dục, Ngân hàng Nhà nước, Cục Quản lý Tài chính (2024). 关于调整完善国家助学贷款有关政策的通知（财教〔2024〕188 号）. +- Ghi chú: Người hưởng lợi là bạn và con cái. Miễn học phí cụ thể gồm những ai, học bổng chia mấy bậc, đều do tỉnh quy định. Trường tư thục được hỗ trợ theo mức trường công lập cùng chuyên ngành tại địa phương, phần chênh lệch vẫn phải tự trả, nên trước khi nhập học phải đối chiếu văn bản địa phương một lần. Học vay là khoản vay chứ không phải trợ cấp, ra trường phải trả lại. Nhưng nó tách「lúc này không đủ tiền học phí」và「mấy năm này có đáng không」thành hai việc riêng, đừng vội lấy thiếu tiền làm kết luận. Ở phần 7 đã viết về trợ cấp đào tạo và trợ cấp tạm thời trong thời gian thất nghiệp, nhà thật sự mất nguồn thu nhập thì hãy đi xin những khoản đó trước. + +### 5. Không đỗ được trường phổ thông không phải là hết đường: trường trung cấp chuyên nghiệp có tuyển sinh liên thông và thi riêng, tuyển dụng lao động kỹ thuật còn được hạ yêu cầu học vấn + +- Chi phí: Không tốn tiền. Trước khi điền nguyện vọng vào trường phổ thông, tra cổng tuyển sinh giáo dục nghề nghiệp của tỉnh một lần. +- Nói thẳng: Luật Giáo dục nghề nghiệp (职业教育法) năm 2022 viết rõ: giáo dục nghề nghiệp và giáo dục phổ thông có vị thế quan trọng như nhau. Trường trung cấp chuyên nghiệp có thể tuyển sinh liên thông với trường cao đẳng nghề, trường cao đẳng nghề còn phải chừa chỉ tiêu tuyển sinh hoặc thi riêng để dành cho học sinh trung cấp chuyên nghiệp. Đơn vị sử dụng lao động cũng không được đặt ngưỡng chặn bạn vì bạn học trường nghề. Điểm không đủ để vào trường phổ thông nghĩa là đổi một đường, không phải dừng lại ở trình độ phổ thông cơ sở. +- Lợi ích: Điều ba của Luật Giáo dục nghề nghiệp (có hiệu lực từ ngày 1 tháng 5 năm 2022):「职业教育是与普通教育具有同等重要地位的教育类型」. Điều ba mươi bảy: trường trung cấp chuyên nghiệp「可以按照国家有关规定，在有关专业实行与高等职业学校教育的贯通招生和培养」. Trường cao đẳng nghề「可以按照国家有关规定，采取文化素质与职业技能相结合的考核方式招收学生」. Cơ quan quản lý giáo dục từ cấp tỉnh trở lên phải lập cổng tuyển sinh giáo dục nghề nghiệp thống nhất, gom và công bố trường, ngành học và tình hình tuyển sinh, đồng thời cung cấp dịch vụ tra cứu đăng ký dự thi. Điều năm mươi ba:「职业学校学生在升学、就业、职业发展等方面与同层次普通学校学生享有平等机会」. Trường cao đẳng nghề và trường đại học công lập có đào tạo nghề nghiệp「应当在招生计划中确定相应比例或者采取单独考试办法，专门招收职业学校毕业生」. 「用人单位不得设置妨碍职业学校毕业生平等就业、公平竞争的报考、录用、聘用条件」. Với vị trí trong đơn vị sự nghiệp tuyển dụng công khai có yêu cầu trình độ kỹ năng nghề, thì「可以适当降低学历要求」(toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). 职业教育法（2022 年修订，第三、三十七、五十三条）. （Bộ Giáo dục đăng lại） +- Ghi chú: Người hưởng lợi là bạn và con cái. Đây là những lối đi theo luật, không phải bảo đảm trúng tuyển. Đào tạo liên thông gồm những ngành nào, tuyển sinh riêng chừa bao nhiêu tỷ lệ, kỳ thi nghề nghiệp học mấy môn, đều do tỉnh quy định. Trước khi điền nguyện vọng, nhất định phải tra cổng tuyển sinh giáo dục nghề nghiệp và chỉ tiêu tuyển sinh của tỉnh mình năm đó, đừng suy từ kinh nghiệm của tỉnh khác. Khi chọn ngành ở trường trung cấp chuyên nghiệp thì lượt qua một lần theo chiều của mục 10 của phần này (chọn kỹ năng chống bị máy thay): ngành phải tới hiện trường, môi trường làm việc không theo khuôn thì ít bị thay thế hơn. Chọn trường thì xem trường đó có nằm trong chỉ tiêu tuyển sinh giáo dục nghề nghiệp của tỉnh mình năm đó không. Giấy chứng chỉ trường cấp thì kiểm tra theo ba bước ở mục 8 của phần này (kiểm tra chứng chỉ có trong danh mục không). Ai quảng cáo「có chỗ làm chắc chắn」,「đảm bảo đỗ」thì đừng tin. +### 6. Biến câu hỏi「học hay đi làm」thành một bài toán: ba năm lương kiếm được sớm, đổi lấy chênh lệch thu nhập mỗi năm của mấy chục năm sau + +- Chi phí: Không tốn tiền. Ngồi tính một lần, khoảng nửa tiếng. +- Nói thẳng: Trải hai lựa chọn ra giấy rồi so với nhau. Đi làm sớm thì chỉ kiếm được ba năm lương ở vị trí không yêu cầu cao, và chỉ một lần đó. Học tiếp thì đổi lại phần thu nhập mỗi năm tăng thêm, rồi còn phải nhân với ba mươi tới bốn mươi năm còn lại. Ở các đô thị Trung Quốc, đọc thêm một năm thì thu nhập cao hơn trung bình 4.0% (năm 1988) tới 10.2% (năm 2001). Số liệu đã cũ, chỉ để tham khảo đại thể. +- Lợi ích: Đây là ước tính cho các đô thị Trung Quốc trong giai đoạn 1988 tới 2001. Đọc thêm một năm thì thu nhập cao hơn bao nhiêu, đó gọi là mức hoàn lại của việc học. Nguyên văn là「从 1988 年每受教育一年 4.0% 大幅上升到 2001 年的 10.2%」. Mức tăng chủ yếu xảy ra sau năm 1992. Nó phản ánh khoảng chênh lệch thu nhập giữa người học hết đại học và người không học (mức tăng lương do học đại học) đang nới rộng ra. Đây là đại lượng tại Trung Quốc dùng được khi quy đổi「đọc thêm một năm」thành chênh lệch thu nhập mỗi năm. Mức trung bình toàn cầu xem mục 7 của phần này (mức hoàn lại trung bình toàn cầu của một năm học thêm) +- Mức bằng chứng: B +- Nguồn: Zhang J, Zhao Y, Park A, Song X (2005). Economic returns to schooling in urban China, 1988 to 2001. Journal of Comparative Economics, 33(4), 730-752. +- Ghi chú: Trước khi tính, hãy trừ các khoản miễn học phí và học bổng nói ở mục 4 của phần này (học phí và học bổng trường trung cấp chuyên nghiệp), và tính thêm cái khoản ở mục 2 của phần này (một năm học thêm và tỷ lệ tử vong). Số liệu chỉ tới năm 2001, chỉ phủ vùng thành thị, không dùng thẳng cho hiện nay được. Mức trung bình toàn cầu 9% ở mục 7 của phần này cũng vậy. Tính thật theo cách này thì kết luận thay đổi theo từng người. Học phí cao, chuyên ngành không khớp với vị trí, số năm học lại dài, thì tính ra có thể thật sự không đáng, nhưng điều đó chỉ nói riêng chương trình đó không đáng, không nói học vô ích. Còn vài điều không đưa vào công thức được: ngưỡng học vấn theo luật, xem mục 5 của phần này (tuyển sinh liên thông và thi riêng của trường trung cấp chuyên nghiệp); khả năng chống bị thay thế, xem mục 10 của phần này (chọn kỹ năng chống bị máy thay); và khả năng đổi ngành. Người hưởng lợi là bạn. + +### 7. Ghi trước đường cơ sở: một năm học thêm, mức hoàn lại tư nhân trung bình toàn cầu (phần rơi vào thu nhập của chính mình) khoảng 9% mỗi năm + +- Chi phí: Không tốn tiền. Ghi nhớ con số này, coi nó là thước đo cho mọi phương án「học một chút」. +- Nói thẳng: Tính theo bình quân toàn cầu, đọc thêm một năm thì thu nhập cá nhân cao hơn khoảng 9% mỗi năm, và mức hoàn lại của nữ luôn cao hơn nam. Con số này nói về đi học chính quy, đăng một lớp không áp thẳng vào được. Dùng nó làm thước đo: một khoá học bắt bạn bỏ ra 3 vạn tệ và nửa năm, thì hãy hỏi mức thu nhập nó tạo ra có tương xứng với một năm đi học không. +- Lợi ích: Một tổng quan gồm 139 nước, 1120 ước tính, phủ giai đoạn 1950 tới 2014, tính ra:「多受一年教育的全球平均私人回报率是每年 9%」. Mức hoàn lại tư nhân là phần rơi vào chính thu nhập của người học: đọc thêm một năm thì thu nhập cao hơn trung bình khoảng 9%. Mức hoàn lại tư nhân của giáo dục đại học thì đang tăng. Mức hoàn lại xã hội của giáo dục (phần rơi vào cả xã hội chứ không phải cá nhân) vẫn rất cao. Mức hoàn lại trung bình của nữ vẫn cao hơn nam +- Mức bằng chứng: A +- Nguồn: Psacharopoulos G, Patrinos HA (2018). Returns to investment in education: a decennial review of the global literature. Education Economics, 26(5), 445-458. ；Bài làm việc nghiên cứu chính sách của Ngân hàng Thế giới 8402. +- Ghi chú: Đây là mức trung bình toàn cầu của「đọc thêm một năm」, không phải mức lợi ích của「đăng một lớp」, cũng không phải con số cụ thể của Trung Quốc, chỉ dùng làm đại lượng tham chiếu. Công dụng của nó là cho mọi phương án đào tạo một thước đo: một khoá học bắt bạn bỏ ra 3 vạn tệ và nửa năm thì phải hỏi nó có tạo ra mức thay đổi thu nhập tương đương một năm giáo dục chính quy không. Còn phải để ý, mức trung bình toàn cầu che khuất chênh lệch rất lớn giữa ngành và giữa các nước. + +### 8. Trước khi trả tiền thi chứng chỉ, kiểm tra chứng chỉ đó có trong Danh mục chứng chỉ nghề nghiệp quốc gia hoặc trong danh sách cơ quan đánh giá đã đăng ký với Bộ Nhân sự và Xã hội không + +- Chi phí: Không tốn tiền. Kiểm tra một lần, mất mười mấy phút. +- Nói thẳng: Có cơ sở nhân danh nghề mới, mức kỹ năng để mở lớp lung tung, thu tiền sai, cấp chứng chỉ ẩu, có cả cơ sở giả mạo cơ quan có thẩm quyền. Bộ Nhân sự và Xã hội đã có đợt xử lý riêng về việc này. Trước khi đăng ký, kiểm tra ba bước: nghề này có trong Danh mục chứng chỉ nghề nghiệp quốc gia không, cơ quan cấp chứng chỉ có phải cơ quan đánh giá đã đăng ký với cơ quan nhân sự không, mã số chứng chỉ tra trên nền tảng chính thức có ra không. Không tra được thì tiền đăng ký là mất trắng. +- Lợi ích: Bộ Nhân sự và Xã hội đã mở đợt xử lý toàn quốc riêng cho việc này. Những sai sót văn bản nêu tên gồm「一些机构和单位以新职业职业资格、职业技能等级等名义随意举办培训、评价、发证活动，乱收费、滥发证；一些机构和单位虚假或夸大宣传，甚至假冒权威机构名义组织培训、评价、颁发证书等」, tức là tổ chức, đơn vị lấy danh nghĩa chứng chỉ nghề mới hay mức kỹ năng nghề để tuỳ tiện mở đào tạo, đánh giá, cấp chứng chỉ, thu tiền lung tung, cấp quá đà, còn có nơi quảng cáo sai hoặc phóng đại, thậm chí giả mạo cơ quan có thẩm quyền để tổ chức đào tạo, đánh giá, cấp chứng chỉ. Biện pháp xử lý gồm phối hợp với các cơ quan quản lý không gian mạng, công an, thị trường để điều tra,「严厉打击假冒鉴定机构、人招部门备案评价机构的评价发证活动」, tức là đánh mạnh vào hoạt động đánh giá cấp chứng chỉ của cơ quan giả mạo cơ quan chuyên môn và của cơ quan đánh giá đã đăng ký. Quảng cáo không hợp lệ trên trang web và nền tảng mới thì đóng theo luật. Văn bản đồng thời yêu cầu「引导公众了解证书查询渠道和方式，判断证书合法合规性和含金量」 +- Mức bằng chứng: A +- Nguồn: Bộ Nhân sự và Xã hội (2022). 关于开展技术技能类「山寨证书」专项治理工作的通知. +- Ghi chú: Kiểm tra theo đúng thứ tự này. Trước hết xem nghề đó có trong《国家职业资格目录》hay không, không có trong danh mục thì không phải chứng chỉ nghề nghiệp quốc gia. Rồi xem cơ quan cấp chứng chỉ có phải cơ quan đánh giá đã đăng ký với cơ quan nhân sự không. Cuối cùng vào nền tảng tra chứng chỉ đánh giá nhân tài kỹ năng chính thức, tra mã số chứng chỉ. Tra không ra thì đừng đăng ký. Loại nào có chữ「quốc tế」,「cao cấp」,「đăng ký」trong tên, hoặc hứa「đảm bảo đỗ」,「treo giấy phép có thu nhập」, thì rủi ro cao nhất. + +### 9. Đào tạo nên đi qua kênh trợ cấp của nhà nước trước, đừng vừa học vừa tự bỏ tiền đăng lớp thương mại + +- Chi phí: Không tốn tiền, hoặc giảm được rất nhiều. Bản thân khoá đào tạo thì vẫn tốn thời gian. Chỗ khó là nhịn được cái thói hứa điện thoại là bỏ tiền đăng lớp, thay vào đó tra danh mục trợ cấp một lần đã. +- Nói thẳng: Từ năm 2025 đến cuối năm 2027, nhà nước dự định tổ chức hơn 3000 vạn lượt đào tạo có trợ cấp. Tiền lấy từ nhiều nguồn: quỹ trợ cấp việc làm, quỹ bảo hiểm thất nghiệp, kinh phí phát triển ngành, quỹ hợp tác đông tây, kinh phí đào tạo người lao động của doanh nghiệp. Nghề càng thiếu người, chi phí đào tạo càng cao thì trợ cấp càng nhiều. Người đang làm việc thì hỏi đơn vị trước, người khác thì lấy danh mục trợ cấp ở cơ quan nhân sự địa phương ra xem trước, rồi mới quyết định học cái gì. +- Lợi ích: Quy mô ở tầm quốc gia là rõ ràng: từ năm 2025 đến cuối năm 2027,「聚焦高精尖产业与急需行业、就业重点群体等开展补贴性培训 3000 万人次以上」. Nguồn tiền được gộp thành một: quỹ trợ cấp việc làm, quỹ bảo hiểm thất nghiệp, kinh phí phát triển ngành, kinh phí hợp tác đông tây, kinh phí đào tạo người lao động của doanh nghiệp. Mức trợ cấp cao hay thấp tuỳ theo mức khan hiếm cấp bách của nghề và chi phí đào tạo, mỗi nghề mỗi khác. Với doanh nghiệp còn yêu cầu「按规定足额提取和使用职工教育经费，确保 60% 以上用于一线职工教育和培训」 +- Mức bằng chứng: A +- Nguồn: Nhóm dẫn đầu xúc tiến việc làm và bảo vệ lao động của Hội đồng Nhà nước (2025). 关于开展大规模职业技能提升培训行动的指导意见. +- Ghi chú: Cách nhận và nhận được bao nhiêu thì do từng tỉnh quy định. Cửa vào là mục「职业培训补贴目录」trên trang cơ quan nhân sự địa phương và cơ sở thực hành công cộng. Ở phần 7 đã viết về trợ cấp đào tạo nghề trong thời gian thất nghiệp (mỗi người tối đa được hưởng 3 lần) và trợ cấp bảo hiểm xã hội. Người đang làm việc thì hỏi kinh phí đào tạo người lao động của đơn vị trước. Tra danh mục trợ cấp rồi mới chọn học gì, tiết kiệm hơn nhiều so với chọn xong rồi đi tìm trợ cấp. + +### 10. Khi chọn kỹ năng, xem trước cái「có phải làm bằng tay, có phải phán đoán tại chỗ không」, loại này khó bị tự động hoá thay nhất + +- Chi phí: Không tốn tiền. Trước khi chọn, lượt việc bạn định học qua chiều này một lượt. +- Nói thẳng: Theo ước tính của Mỹ, khoảng 47% vị trí có nguy cơ khá cao bị máy tính đảm nhiệm, còn tính theo từng nhiệm vụ cụ thể thì tỷ lệ thấp hơn nhiều. Có ba chỗ máy chưa vượt qua được: độ khéo léo của đôi tay, khả năng nghĩ ra cái mới, và khả năng ứng xử với con người. Nên việc phải tới hiện trường, môi trường không theo khuôn, phải phán đoán tại chỗ thì khó bị thay hơn, còn vị trí quy trình cố định, chủ yếu xử lý thông tin thì nguy cơ cao nhất. +- Lợi ích: Ước tính từng nghề một trong 702 nghề tại Mỹ, xem「công việc này giao cho máy tính được với xác suất bao nhiêu」(đánh giá xác suất khả năng tính toán hoá), kết quả là khoảng 47% vị trí việc làm rơi vào khoảng nguy cơ cao. Trong mô hình, ba chỗ máy hiện chưa vượt qua quyết định mức nguy cơ cao hay thấp, tác giả gọi đó là「bottleneck kỹ thuật」. Một là cảm giác và thao tác, gồm độ khéo léo của ngón tay, độ khéo léo của bàn tay, làm việc trong không gian hẹp và ở tư thế khó chịu. Hai là trí tuệ sáng tạo, tức khả năng nghĩ ra cái mới. Ba là trí tuệ xã hội, tức khả năng ứng xử với con người. Vì vậy công việc có môi trường không theo khuôn, cần thao tác tay và phán đoán tại chỗ, xác suất bị thay thế luôn thấp hơn, đây không phải chuyện tình cờ của vài nghề riêng lẻ. Vị trí có quy trình cố định, chủ yếu xử lý thông tin, có nguy cơ cao nhất +- Mức bằng chứng: B +- Nguồn: Frey CB, Osborne MA (2017). The future of employment: how susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, 254-280. +- Ghi chú: Tranh cãi. Nghiên cứu này là một bản thảo bài báo năm 2013 chưa công bố chính thức, và điểm nó chấm là「cả nghề có bị máy đảm nhiệm được không」. Sau đó có nghiên cứu tách ra từng nhiệm vụ cụ thể để ước tính, kết quả tỷ lệ nguy cơ cao thấp hơn nhiều. Nghiên cứu cũng không nói chuyện này sẽ xảy ra vào năm nào. Dùng nó thì đừng nhìn con số 47%, hãy nhìn ba chỗ nó chỉ ra là máy vướng. Việc lắp đặt, sửa chữa, chăm sóc tại hiện trường chống bị thay hơn công việc giấy tờ thuần tuý, lý do nằm ở đó. Vì vậy khi「học một kỹ năng」thì nên chọn loại cần tới hiện trường, đối diện môi trường không theo khuôn. + +### 11. Trước khi học, xác nhận kỹ năng này có đánh giá theo cấp bậc không, cấp bậc có đổi ra tiền được không + +- Chi phí: Không tốn tiền. Chỉ cần tra một lần. +- Nói thẳng: Nhà nước đang đẩy「thợ bậc tám mới」, yêu cầu doanh nghiệp gắn lương theo vị trí với bậc kỹ năng, và người lao động thể hiện tốt còn được công nhận bậc ngay mà không cần xét thâm niên, cấp bậc, tư cách. Đây là hướng chính sách, mỗi địa phương mỗi doanh nghiệp khác nhau rất nhiều. Kỹ năng nào có tiêu chuẩn nghề nghiệp quốc gia và lối đi theo bậc, thì sau này mới có chuyện tăng lương theo bậc, nên học trước phải tra xem có không. +- Lợi ích: Nhà nước đang đẩy hệ thống đánh giá kỹ năng nghề của「thợ bậc tám mới」. Văn bản yêu cầu gắn bậc với lương, nguyên văn là「强化技能导向的薪酬分配制度……指导企业实行岗位薪酬与岗位价值、技能等级双挂钩的分配办法」. Các địa phương phải「加大高级工以上评聘力度和批次」. Doanh nghiệp đủ điều kiện có thể tự đánh giá bậc kỹ năng cho người lao động của mình. Với người lao động thể hiện tốt thì「可打破工作年限、等级、资历等条件，直接认定相应职业技能等级」 +- Mức bằng chứng: B +- Nguồn: Nhóm dẫn đầu xúc tiến việc làm và bảo vệ lao động của Hội đồng Nhà nước (2025). 关于开展大规模职业技能提升培训行动的指导意见. ；Bộ Nhân sự và Xã hội (2021). 「技能中国行动」实施方案. +- Ghi chú: Đây là hướng chính sách, không phải con số lương đã thực hiện được, mức độ thực thi khác nhau rất nhiều giữa các địa phương và doanh nghiệp. Phần làm theo được là: khi chọn kỹ năng thì xem nó có tiêu chuẩn nghề nghiệp quốc gia và lối đi theo bậc tương ứng không, có lối đi theo bậc thì sau này mới nói chuyện tăng lương theo bậc. Kỹ năng không có bậc, cũng không có ngưỡng vào nghề thì học xong chỉ có thể tự đi tìm việc. + +### 12. Đừng coi「trên mạng nói lương cao」là thiếu người, hãy tra danh mục nghề khan hiếm và danh mục trợ cấp tại địa phương + +- Chi phí: Không tốn tiền. Tra hai văn bản của địa phương. +- Nói thẳng: Mỗi nơi cho nghề nào nhiều trợ cấp đào tạo hơn là tuỳ mức khan hiếm cấp bách và chi phí đào tạo, nên mức trợ cấp nhìn được chỗ đó thiếu người hay không. Câu「nghề này kiếm được bao nhiêu một tháng」trên mạng phần lớn là giá niêm yết của nền tảng tuyển dụng hoặc lời tự kể của cá nhân. Bạn thấy phần lớn là người đã kiếm được tiền, và không tính giờ làm, đường đi, bệnh nghề nghiệp, mùa cao mùa thấp. +- Lợi ích: Chính mức trợ cấp là cách nhà nước định giá「những nghề nào đang thiếu người」. Chính sách yêu cầu「健全根据职业（工种）急需紧缺程度、培训成本等情况确定的差异化补贴机制」, các địa phương căn cứ đó xác định phạm vi các chương trình đào tạo kỹ năng nghề có trợ cấp tại địa phương. Nên cùng một nghề mà mức trợ cấp chênh nhau bao nhiêu giữa các thành phố thì phản ánh đúng mức khan hiếm tại chỗ đó +- Mức bằng chứng: C +- Nguồn: Nhóm dẫn đầu xúc tiến việc làm và bảo vệ lao động của Hội đồng Nhà nước (2025). 关于开展大规模职业技能提升培训行动的指导意见. ；Cục Thống kê Quốc gia (2026). 2025年城镇单位就业人员年平均工资情况：「规模以上企业就业人员年平均工资为106080元，其中，中层及以上管理人员210016元，专业技术人员155491元，办事人员和有关人员94936元，社会生产服务和生活服务人员79857元，生产制造及有关人员80739元」. ；Trung tâm Hướng dẫn kỹ thuật đào tạo việc làm của Bộ Nhân sự và Xã hội (2023). 2022年第四季度全国招聘大于求职「最缺工」的100个职业排行（中国就业网）. +- Ghi chú: Lương chính thức chia theo nghề chỉ chia tới năm nhóm vị trí lớn, không có nghề cụ thể. Số liệu năm 2025 của Cục Thống kê Quốc gia là doanh nghiệp quy mô trở lên có lương năm trung bình khoảng 10.6 vạn tệ (con số chính xác 106080 tệ). Quản lý cấp trung trở lên khoảng 21 vạn tệ (con số chính xác 210016 tệ), nhân viên kỹ thuật chuyên môn khoảng 15.5 vạn tệ (con số chính xác 155491 tệ), nhân viên văn phòng 94936 tệ, người sản xuất 80739 tệ, người dịch vụ sản xuất và sinh hoạt 79857 tệ. Bảng「100 nghề thiếu người nhất」của Bộ Nhân sự và Xã hội tra được trên Cổng việc làm Trung Quốc, kỳ mới nhất dừng ở quý bốn năm 2022, lập từ số liệu tuyển dụng và xin việc của các cơ quan dịch vụ việc làm công cộng ở 102 thành phố cả nước. Câu「nghề này kiếm được bao nhiêu một tháng」lan truyền trên mạng gần như đến từ giá niêm yết của nền tảng tuyển dụng hoặc lời tự kể của cá nhân. Phần bạn nhìn thấy phần lớn là nhóm đã kiếm được tiền, và không ai tính giờ làm, đường đi, bệnh nghề nghiệp, mùa cao mùa thấp cả. Muốn xem nhu cầu thật thì vào trang cơ quan nhân sự địa phương tra danh mục trợ cấp, rồi tra bảng xếp hạng nhu cầu vị trí của cơ quan dịch vụ việc làm công cộng. Cuối cùng hãy ra công trường, nhà máy, cửa hàng hỏi người đang làm việc. +### 13. Với cùng số tiền và thời gian, ưu tiên chương trình ngắn hạn học xong là vào việc được ngay + +- Chi phí: Phải đóng học phí khoá ngắn hạn, số tiền không nhiều. Về thời gian thì phải bỏ ra một khoảng. Chỗ khó là khi học được nửa chừng chưa thấy kết quả thì dừng lại đúng điều kiện đã hẹn trước. +- Nói thẳng: Chính sách khuyến khích loại đào tạo「ngắn, nhanh, hiệu quả」giúp người học có việc trong thời gian ngắn, cùng chuyên ngành nhỏ mà các trường nghề mở ra. Chọn chương trình thì chỉ nhìn một điều: học xong có một vị trí cụ thể để nộp hồ sơ không. Học cả năm, tốt nghiệp rồi còn phải「tích lũy thêm kinh nghiệm」thì thời gian bỏ ra thực tế vượt xa giá niêm yết. Trước khi bắt đầu, viết sẵn điều kiện dừng: học tới mức nào mà vẫn chưa tìm được việc thì dừng. +- Lợi ích: Quan điểm chính sách cũng như vậy: hỗ trợ「民办培训机构开展市场化的高质量职业技能培训项目和能短期实现就业的『短平快』式职业技能培训项目」. Đồng thời thúc các trường nghề「聚焦就业市场急需知识和技能……开设微专业和职业能力培训课程」 +- Mức bằng chứng: C +- Nguồn: Nhóm dẫn đầu xúc tiến việc làm và bảo vệ lao động của Hội đồng Nhà nước (2025). 关于开展大规模职业技能提升培训行动的指导意见. +- Ghi chú: Tiêu chí đánh giá chỉ một: học xong có một vị trí cụ thể để nộp hồ sơ không. Học cả năm, tốt nghiệp rồi còn phải「tích lũy thêm kinh nghiệm」thì chi phí thực tế cao hơn giá niêm yết nhiều. Điều kiện dừng mà phần 4 nói tới cũng áp dụng: trước khi bắt đầu, viết rõ「học tới mức nào mà vẫn chưa tìm được việc thì dừng」. + +### 14. Học xong thì đóng sách lại tự kiểm một lượt, đừng đọc lại từ đầu + +- Chi phí: Không tốn tiền, tổng thời gian cũng không phải tăng. Đổi khoảng thời gian trước đây dùng để đọc lại thành đóng sách viết ra hoặc làm bài. Chỗ khó là bài tự kiểm lộ ra ngay chỗ chưa nhớ, thấy nặng hơn đọc lại. +- Nói thẳng: Đọc xong đọc lại thấy rất quen nhưng thật ra không nhớ chắc. Đóng sách tự kiểm một lượt thì hiệu quả rõ hơn hẳn, thêm bài kiểm tra nhỏ trong lớp làm điểm tăng ở mức trung bình. Khi xếp hạng mười cách học phổ biến, tự kiểm nằm ở bậc cao nhất. Đọc xong một mục thì viết lại các ý chính ra giấy, rồi mới lật trở lại đối chiếu. +- Lợi ích: Phân tích tổng hợp gồm 222 nghiên cứu độc lập với 48478 học sinh: bài kiểm tra trong lớp làm thành tích học tập của học sinh tăng lên, cỡ hiệu quả ở mức trung bình (g = 0.499). Cỡ hiệu quả phụ thuộc vào mấy điều kiện: nhóm đối chứng học theo cách nào, dạng bài kiểm tra có giống kỳ thi cuối không, tư liệu có khớp không, có cho phản hồi sửa sai không, bài kiểm tra lặp lại mấy lần, test ở đâu và lúc nào, can thiệp kéo dài bao lâu, dùng thiết kế thử nghiệm nào. Một tổng quan khác đánh giá từng cách một trong mười cách học phổ biến, xếp「bài kiểm tra có luyện tập」và「luyện phân tán」vào bậc hiệu quả cao nhất (high utility), vì chúng hợp với người học ở mọi lứa tuổi và mọi trình độ, và đã được kiểm chứng trong nhiều dạng đánh giá cũng như trong thực tế giáo dục. +- Mức bằng chứng: A +- Nguồn: Yang C, Luo L, Vadillo MA, Yu R, Shanks DR (2021). Testing (quizzing) boosts classroom learning: A systematic and meta-analytic review. Psychological Bulletin, 147(4), 399–435. ；Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 14(1), 4–58. +- Ghi chú: Tự kiểm là để buộc bản thân bày ra thứ đang nằm trong đầu, không phải để chấm điểm. Cách làm: đọc xong một mục, đóng sách viết lại các ý chính, rồi lật trở lại đối chiếu. Có phản hồi sửa sai thì hiệu quả hơn, nên sau khi đối chiếu phải nhìn rõ sai ở đâu. Mức lợi ích ghi「lớn」là căn cứ cỡ hiệu quả và thang đánh giá. Hai tài liệu bằng chứng này đo điểm số, không đo số giờ, nên không áp ngưỡng tiết kiệm thời gian. Tự kiểm và mục 15 của phần này (chia cùng thời gian ra mấy ngày) là hai cách được xếp hạng cao nhất trong cùng một tổng quan đó, nên dùng kèm nhau. + +### 15. Chia cùng thời gian đó ra mấy ngày, đừng học một lạch xong + +- Chi phí: Không tốn tiền, tổng thời gian cũng không đổi. Chỉ có cách xếp lịch thay đổi: cùng 3 giờ đó, chia thành ba ngày mỗi ngày 1 giờ. Chỗ khó là phải bắt đầu trước vài ngày, không dồn đến đêm cuối cùng. +- Nói thẳng: Cùng một khoảng thời gian đó, học chia ra mấy ngày thì nhớ lâu hơn học một lạch xong. Có 317 thử nghiệm đang đo đúng chuyện này và kết luận giống nhau. Còn một quy tắc rất tiện: bạn cần nhớ đến bao lâu thì khoảng cách ôn lại phải dài bấy nhiêu. Ngày mai thi thì hôm nay ôn, một tháng sau thi thì cứ vài ngày quét lại một lần. +- Lợi ích: Một phân tích tổng hợp gom 184 bài báo, 317 thử nghiệm, 839 lần đánh giá hiệu quả của luyện phân tán. Nó so sánh hai việc: học liền một mạch và học chia vài lần (hiệu ứng khoảng cách), cùng khoảng cách ngắn và khoảng cách dài (hiệu ứng độ trễ). Kết luận là khoảng cách ôn lại và thời gian cần nhớ cùng quyết định tỷ lệ nhớ ở bài kiểm tra cuối: muốn nhớ lâu hơn bao nhiêu thì khoảng cách ôn cho tỷ lệ nhớ tốt nhất cũng dài hơn bấy nhiêu. Một tổng quan khác đánh giá từng cách một trong mười cách học phổ biến, xếp luyện phân tán vào bậc hiệu quả cao nhất (high utility). +- Mức bằng chứng: A +- Nguồn: Cepeda NJ, Pashler H, Vul E, Wixted JT, Rohrer D (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354–380. ；Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques. Psychological Science in the Public Interest, 14(1), 4–58. +- Ghi chú: Dùng chung với mục 14 của phần này (đóng sách tự kiểm) thì tốt nhất: bản thân bài tự kiểm cũng chia ra mấy ngày làm. Phân tích tổng hợp của Cepeda chủ yếu dùng tư liệu nhớ danh sách từ, nên không đưa ra số ngày chung cho nội dung phức tạp. 「Càng thi xa thì khoảng cách càng dài」là quy tắc tác giả đưa ra, không kèm lịch cụ thể. Ôn dồn vào đêm trước khi thi thì hữu ích ngay tại chỗ, vấn đề là quên nhanh. + +### 16. Đừng coi gạch chân, đọc lại, viết tóm tắt là cách học chính + +- Chi phí: Không tốn tiền, ngược lại còn tiết kiệm thời gian. Chỗ khó là mấy cách đó dễ làm nhất, và cũng tạo cảm giác「tôi đang học」mạnh nhất. +- Nói thẳng: Học sinh dùng nhiều nhất là gạch chân và đọc lại, nhưng trong bảng xếp hạng mười cách học phổ biến thì cả hai nằm ở bậc thấp nhất. Viết tóm tắt, phương pháp ghi nhớ từ khoá, tưởng tượng hình ảnh đi kèm bài đọc cũng ở bậc thấp nhất. Chúng không hoàn toàn vô dụng, chỉ là điều kiện để có tác dụng quá hẹp, đổi cùng số thời gian sang tự kiểm thì đáng hơn. +- Lợi ích: Một tổng quan đánh giá mười cách học phổ biến, xếp hạng theo「hiệu quả có lan được sang điều kiện học khác, đặc điểm học sinh khác, tư liệu khác, dạng đánh giá khác hay không」. Xếp bậc cao nhất là bài kiểm tra có luyện tập và luyện phân tán. Xếp bậc trung bình là đặt câu hỏi tinh vi, tự giải thích, luyện xen kẽ. Năm cách ở bậc thấp nhất là: viết tóm tắt, gạch chân (tức kẻ dòng), phương pháp ghi nhớ từ khoá, tưởng tượng hình ảnh đi kèm bài đọc, và đọc lại. Tổng quan ghi rõ: phần lớn học sinh nói họ dùng đọc lại và gạch chân, nhưng hai cách đó không ổn định trong việc nâng điểm, nên nên thay bằng cách khác, chẳng hạn dùng bài kiểm tra có luyện tập thay cho đọc lại. +- Mức bằng chứng: A +- Nguồn: Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 14(1), 4–58. +- Ghi chú: Kẻ dòng vẫn dùng được để「đánh dấu chỗ lát nữa sẽ tự kiểm」, cái sai là coi việc kẻ dòng là học. Tổng quan cũng nói bậc thấp nhất không có nghĩa là bằng không, chỉ là điều kiện để có tác dụng hẹp và còn cần nhiều nghiên cứu hơn. Cách thay thế xem mục 14 của phần này (đóng sách tự kiểm) và mục 15 của phần này (chia cùng thời gian ra mấy ngày). + +### 17. Luyện xen kẽ nhiều dạng bài, đừng làm liên hai mươi bài cùng một dạng + +- Chi phí: Không tốn tiền. Thay vì làm bài tập theo thứ tự dạng, hãy làm theo thứ tự đảo. Chỗ khó là luyện xen kẽ lúc đó tỷ lệ đúng thấp hơn, có cảm giác như lùi. +- Nói thẳng: Làm một loại bài liền một dãy thì lúc đó rất thuận tay, nhưng hết dãy thì không nhất thiết còn phân biệt được phải dùng cách nào. Luyện xen kẽ vài dạng bài thì hiệu quả dài hạn tốt hơn, ở mức trung bình. Cách này cũng kén tư liệu: luyện hình ảnh và luyện phân loại là hiệu quả nhất, bài toán nhỏ hơn một chút, còn học từ vựng thì không bằng luyện theo khối. +- Lợi ích: Một phân tích tổng hợp nhiều tầng gồm 59 nghiên cứu, 238 cỡ hiệu quả, 158 mẫu. Hiệu quả chung của luyện xen kẽ so với luyện theo khối ở mức trung bình (Hedges' g = 0.42). Xét theo tư liệu: nhóm vẽ là tốt nhất (g = 0.67), các tư liệu hình ảnh khác cũng khá; nhiệm vụ toán học có hiệu quả nhỏ hơn (g = 0.34); văn bản giải thích và tư liệu vị giác không có hiệu quả chung đáng kể; với từ vựng thì luyện theo khối lại tốt hơn (g = −0.39). Phân tích hồi quy cho thấy: các nhóm càng giống nhau, bên trong cùng một nhóm càng khác nhau, tư liệu càng phức tạp thì lợi ích của luyện xen kẽ càng lớn. +- Mức bằng chứng: A +- Nguồn: Brunmair M, Richter T (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators. Psychological Bulletin, 145(11), 1029–1052. +- Ghi chú: Tổng quan dẫn ở mục 16 của phần này (đừng coi gạch chân là cách học chính) xếp luyện xen kẽ vào bậc trung bình, vì lúc đó số nghiên cứu chưa nhiều. Xem cả hai thì cách dùng là: nội dung cần「nhận ra đây là dạng nào, phải dùng cách nào」 thì luyện xen kẽ, chẳng hạn dạng bài toán, ca bệnh, áp dụng điều luật; nội dung cần học thuộc thì luyện theo khối, chẳng hạn từ vựng, định nghĩa. + +### 18. Đừng chọn cách học theo kiểu「tôi là người học bằng mắt, người kia là người học bằng tai」 + +- Chi phí: Không tốn tiền. Chỉ là không chọn tư liệu, không chọn khoá học theo cách phân loại đó nữa, và không bỏ tiền làm bài kiểm tra kiểu học nữa. +- Nói thẳng: Nhiều người tin rằng mỗi người có phong cách học riêng, như học bằng mắt, học bằng tai, và dạy đúng phong cách thì học mới tốt. Nghiên cứu kiểm tra nghiêm ngặt chuyện này rất ít, gần như không có nghiên cứu nào ủng hộ, có vài nghiên cứu còn ra kết quả ngược lại. Mỗi người thật sự có sở thích khác nhau, nhưng không có bằng chứng nào cho thấy dạy theo sở thích thì học tốt hơn, nên không cần bỏ tiền làm bài kiểm tra phong cách học. +- Lợi ích: Một tổng quan được giao đánh giá bằng chứng về phong cách học chỉ ra: muốn chứng minh「dạy theo phong cách học」có hiệu quả thì phải chia học sinh theo phong cách trước, rồi chia ngẫu nhiên học sinh mỗi nhóm vào các cách dạy khác nhau, cuối cùng cho tất cả thi cùng một bài kiểm tra. Dấu hiệu chứng minh có hiệu quả là xuất hiện một tương tác đặc biệt: cách dạy hiệu quả nhất với phong cách này lại không phải hiệu quả nhất với phong cách kia. Kết luận của tổng quan là gần như không tìm thấy tương tác đó; tài liệu về phong cách học rất nhiều, nhưng số nghiên cứu dùng thiết kế thử nghiệm đạt yêu cầu rất ít, và có vài nghiên cứu còn cho kết quả trái hẳn với「giả thuyết khớp». Từ đó các tác giả cho rằng bằng chứng hiện có không đủ để đưa bài kiểm tra phong cách học vào thực hành giáo dục. +- Mức bằng chứng: A +- Nguồn: Pashler H, McDaniel M, Rohrer D, Bjork R (2008). Learning Styles: Concepts and Evidence. Psychological Science in the Public Interest, 9(3), 105–119. +- Ghi chú: Tổng quan đồng thời nói, điều đó không có nghĩa mọi phiên bản của lý thuyết phong cách học đều đã bị kiểm tra và bị bác bỏ, nhiều phiên bản chưa hề được kiểm tra. Cái cần tránh là bỏ tiền làm bài kiểm tra phong cách học và chọn lớp đào tạo theo phong cách. Tiêu chí chọn đào tạo xem mục 9 của phần này (đào tạo nên đi qua kênh trợ cấp của nhà nước) và mục 13 của phần này (ưu tiên chương trình ngắn hạn học xong là vào việc được ngay). +### 19. Phần cần học thuộc thì tự ra câu hỏi theo cách dùng về sau để tự kiểm, đừng đọc thuộc một lượt là xong + +- Chi phí: Không tốn tiền, cũng không tốn thêm thời gian. Cách luyện đổi thôi: thay「đọc thuộc một lượt nguyên văn」bằng「tự ra câu hỏi theo cách dùng về sau để tự kiểm」. Chỗ khó là phải nghĩ trước ra những thứ đó về sau sẽ dùng thế nào. +- Nói thẳng: Đóng sách tự kiểm thì ích lợi không dừng ở mấy chữ đã đọc thuộc. Cộng lại 192 lần so sánh và hơn một vạn người thì cách đó dùng được cho câu hỏi đổi dạng, cho bài toán ứng dụng và bài suy luận, cỡ hiệu quả trung bình hơi nhỏ. Nhưng có hai trường hợp gần như không dùng được: đảo câu hỏi và đáp án, và nội dung chỉ từng thấy trong sách mà chưa bao giờ bị kiểm. +- Lợi ích: Một phân tích tổng hợp gom các nghiên cứu hơn 40 năm, gồm 67 bài đã công bố và chưa công bố, 122 thử nghiệm, 192 cỡ hiệu quả chuyển đổi, tổng cộng 10382 người. So với nhóm đối chứng「xem lại một lượt mà không kiểm」thì học đạt được nhờ bài kiểm tra có luyện tập thật sự chuyển sang được tình huống khác, cỡ hiệu quả d=0.40 (95% CI 0.31 đến 0.50). Mấu chuyển đổi mạnh nhất là: đổi dạng câu hỏi để kiểm, kiểm bài toán ứng dụng và bài suy luận, câu hỏi kiểu chẩn đoán y khoa, kiểm bằng những từ dẫn nghĩa gần giống nhau. Mấu chuyển đổi yếu nhất là: đảo kích thích và phản ứng, tư liệu gặp lúc mới học nhưng chưa từng bị kiểm, câu hỏi dạng bài mẫu. Trong các điều kiện điều tiết, dạng câu trả lời lúc luyện và lúc kiểm có giống nhau không, đã từng luyện gọi lại với phản hồi chi tiết chưa, và lần kiểm đầu tiên đã trả lời đúng chưa, đều ảnh hưởng mạnh tới việc có xuất hiện chuyển đổi tích cực hay không. Tác giả còn dùng hai cách hiệu chỉnh thiên lệch công bố, hiệu quả của các điều kiện điều tiết bị ảnh hưởng rất ít, nhưng ước lượng điểm cắt bị hạ rõ rệt: khi không có điều kiện nào vừa nêu thì thường không thấy chuyển đổi tích cực +- Mức bằng chứng: A +- Nguồn: Pan SC, Rickard TC (2018). Transfer of test-enhanced learning: meta-analytic review and synthesis. Psychological Bulletin, 144(7), 710-756. +- Ghi chú: Từ「học thuộc」thường gộp hai việc vào một: xem lại tư liệu (đọc lại, gạch chân), và chủ động gọi lại sau khi đóng sách. Loại trước được xếp hạng thấp nhất trong tổng quan xếp hạng mười cách học, muốn xem chính phần xếp hạng thì xem mục 16 của phần này (đừng coi gạch chân, đọc lại, viết tóm tắt là cách học chính). Cần làm là loại sau: chủ động gọi lại sau khi đóng sách, cụ thể xem mục 14 của phần này (đóng sách tự kiểm), và phải chia ra mấy ngày làm, xem mục 15 của phần này (chia ra mấy ngày). Cách làm là nghĩ trước ra về sau sẽ dùng thế nào: phải chọn ra trong bài thi, hay phải phán đoán tại hiện trường, rồi ra câu hỏi theo đúng dạng đó để tự kiểm. Đáng làm thêm một lượt kiểm đảo chiều, vì trong phân tích tổng hợp này chuyển đổi theo chiều đảo là yếu nhất, cả hai chiều đều phải luyện tới. + +### 20. Khi xét chức danh nghề nghiệp, trước hết xem mình ở chuỗi nào, bậc nào, rồi tìm cửa khai theo tính chất đơn vị + +- Chi phí: Không tốn tiền. Tra thông báo khai của tỉnh mình năm đó một lần, khoảng nửa tiếng. +- Nói thẳng: Chức danh nghề nghiệp cả nước chia thành 27 chuỗi, mỗi chuỗi chia làm bậc sơ cấp, trung cấp, cao cấp. Thường phải khai từng bậc một. Người làm việc trong đơn vị thì đơn vị xét hồ sơ, công bố rồi chuyển lên trên. Người ở doanh nghiệp tư nhân thì ra điểm tiếp nhận hồ sơ xét chức danh, hoặc nhờ cơ quan dịch vụ nhân sự, hiệp hội nghề nghiệp. Người làm tự do thì dự xét của cơ quan nhân sự địa phương tổ chức. +- Lợi ích: Cả nước có 27 chuỗi chức danh nghề nghiệp, ví dụ kinh tế, kỹ thuật, y tế, xuất bản. Chức danh chia ba tầng cao cấp, trung cấp, sơ cấp. Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội áp dụng cho doanh nghiệp, đơn vị sự nghiệp, tổ chức xã hội, tổ chức kinh tế cá thể, và cả với người làm tự do. Người khai thường phải khai từng bậc một. Đơn vị nơi bạn làm việc phải xét hồ sơ và công bố tại đơn vị, thời gian công bố không dưới 5 ngày làm việc. Doanh nghiệp tư nhân và người làm tự do có thể nhờ cơ quan đại diện nhân sự xét, công bố, giới thiệu. Các địa phương phải đặt điểm tiếp nhận hồ sơ xét chức danh ở khu công nghệ, khu công nghiệp cao và nơi tương tự, cũng có thể tiếp nhận qua cơ quan dịch vụ nhân sự, hiệp hội thương mại, hội liên hiệp nghề nghiệp và hiệp hội khoa học. Người làm tự do đủ điều kiện thì theo nguyên tắc thuộc địa phương nào thì dự xét chức danh do cơ quan nhân sự địa phương tổ chức. Hồ sơ tra được trên mạng thì không được bắt bạn mở giấy tờ khác. Hồ sơ thiếu thì bên xét phải nói một lần là thiếu gì. Người ở doanh nghiệp tư nhân thường khai tại nơi có quan hệ lao động. Người được cử đi công tác ở tỉnh khác liên tục đủ một năm thì được khai xét chức danh tại nơi được cử đi. Người đã nghỉ hưu không được khai. Người lao động của đơn vị sự nghiệp đang bị kỷ luật từ mức khiển trách trở lên thì không được khai trong thời gian bị kỷ luật (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội (人力资源和社会保障部令第 40 号) (2019). 职称评审管理暂行规定. ：第二条、第十三条、第十四条、第十五条、第十六条、第十七条、第十八条；Văn phòng Bộ Nhân sự và Xã hội (2020). 关于进一步做好民营企业职称工作的通知. ：「在专业技术人才密集的创业孵化基地、高新技术开发区、科技园区等地设立职称申报受理服务点，或通过人才中介服务机构、工商联、行业协会商会、学会等社会组织受理民营企业专业技术人才职称申报」「民营企业专业技术人才被派驻外地连续工作一年以上的，可按有关规定在派驻地申报职称评审」；Tân Hoa Xinh (2024). 职称评审进入高峰，信息如何查？有何新变化？ ：「我国共设有经济、工程、卫生、出版等27个职称系列」「符合条件的自由职业者，可根据属地原则申报参加当地人社部门组织的职称评审」；Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「不将论文作为评价应用型人才的限制性条件」「对职称外语和计算机应用能力考试不作统一要求」「符合条件的高技能人才，可参加工程系列专业技术人才职称评审」「公务员不得参加专业技术人才职称评审」 +- Ghi chú: Cần chuẩn bị những hồ sơ nào thì cả nước không có một danh sách thống nhất. Mỗi chuỗi, mỗi tỉnh hằng năm đều phát thông báo khai, trong đó ghi danh mục hồ sơ, thời gian khai và điều kiện về thành tích. Nên bước đầu tiên là vào trang cơ quan nhân sự tỉnh tìm thông báo của năm đó. Bài báo không còn bắt buộc phải có: ý kiến cải cách năm 2016 yêu cầu không đặt bài báo làm điều kiện cứng với nhân tài ứng dụng, và không đặt yêu cầu thống nhất cho kỳ thi ngoại ngữ chức danh cũng như ứng dụng máy tính. Người lao động kỹ thuật cũng có đường: nhân tài kỹ năng cao đủ điều kiện được dự xét chức danh chuỗi kỹ thuật. Công chức không được dự xét chức danh. + +### 21. Chức danh sơ cấp và trung cấp ở ngành kế toán lấy bằng thi quốc gia thống nhất, đăng ký theo học vấn và số năm công + +- Chi phí: Phải đóng phí đăng ký theo tiêu chuẩn của tỉnh. Ôn thi chiếm mấy tháng thời gian ngoài giờ làm việc. +- Nói thẳng: Có những chuỗi lấy chức danh sơ cấp và trung cấp bằng cách thi, thi đỗ là xong, không xét lại nữa. Kế toán là một trường hợp: kế toán trợ viên cần tốt nghiệp trung học trở lên, kế toán viên cần bằng đại học và làm kế toán đủ 4 năm, hoặc bằng cao đẳng và đủ 5 năm. Người đã có một số chứng chỉ nghề nghiệp cũng có thể được công nhận là có chức danh tương ứng. +- Lợi ích: Ý kiến cải cách năm 2016 quy định, với chuyên ngành thực hiện kỳ thi quốc gia thống nhất cho chức danh sơ cấp và trung cấp thì không tiếp tục xét hoặc công nhận tương ứng. Ở những lĩnh vực mà chức danh gắn chặt với chứng chỉ nghề nghiệp, có chứng chỉ nghề nghiệp thì được công nhận là có chức danh tương ứng, và còn dùng làm điều kiện để khai bậc cao hơn. Lấy ngành kế toán làm ví dụ. Kế toán trợ viên và kế toán viên thực hiện kỳ thi quốc gia thống nhất. Kế toán trưởng thì kết hợp thi và xét, chức danh cao cấp nhất thường chỉ xét. Điều kiện dự thi như sau. Kế toán trợ viên: tốt nghiệp trung học trở lên, gồm trường trung cấp chuyên nghiệp, trường nghề, trường kỹ thuật. Kế toán viên: bằng tiến sĩ; hoặc thạc sĩ, làm kế toán đủ 1 năm; hoặc bằng cử nhân, tốt nghiệp lớp sau đại học, làm đủ 2 năm; hoặc đại học, làm đủ 4 năm; hoặc cao đẳng, làm đủ 5 năm. Người có bằng thạc sĩ kế toán, tiến sĩ kế toán thì khi thi kế toán viên được miễn môn Quản trị tài chính (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「专业技术人才取得职业资格即可认定其具备相应系列和层级的职称，并可作为申报高一级职称的条件。初级、中级职称实行全国统一考试的专业不再进行相应的职称评审或认定。」；Bộ Nhân sự và Xã hội, Bộ Tài chính (2019). 关于深化会计人员职称制度改革的指导意见. ：「助理会计师、会计师实行全国统一的会计专业技术资格考试」「高级会计师采取考试与评审相结合方式，正高级会计师一般采取评审方式」；Bộ Tài chính và cộng sự (2024). 关于做好会计专业学位与会计专业技术资格衔接有关工作的通知. +- Ghi chú: Chuỗi của bạn là thi, là xét, hay kết hợp thi và xét thì xem bản「hướng dẫn về đẩy mạnh cải cách hệ thống chức danh ×××」của chính chuỗi đó. Bộ Nhân sự và Xã hội cùng các cơ quan quản lý ngành đã phát hành theo từng chuỗi, tra được trong kho chính sách trên trang Chính phủ Trung Quốc. Thời gian đăng ký thi và các môn thi thì theo thông báo thi của năm đó. Công chức được dự thi kỳ thi kế toán, nhưng không được dự xét chức danh kế toán. + +### 22. Đừng thuê trung gian xét thay, đừng mua bài báo viết hộ, đừng bịa hồ sơ: bị phát hiện thì chức danh bị thu hồi, lưu vào hồ sơ lý tính 3 năm + +- Chi phí: Không tốn tiền. Chỗ khó là nhịn được quảng cáo「bỏ tiền là đảm bảo đỗ」. +- Nói thẳng: Chức danh có được bằng hồ sơ giả, bài báo thuê viết, hoặc chiếm đoạt kết quả của người khác thì bị phát hiện là thu hồi. Bản ghi mất uy tín bị lưu 3 năm, và đưa vào nền tảng tín dựng toàn quốc. Cam kết bạn ký lúc khai mà không đúng thì 3 năm sau không được khai nữa. Cơ quan nhân sự còn chuyên điều tra quảng cáo sai, xét giả và chứng chỉ giả của các trung gian. +- Lợi ích: Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội quy định, người cung cấp hồ sơ giả, chiếm đoạt tác phẩm và kết quả học thuật của người khác, hoặc dùng thủ đoạn bất chính khác để lấy chức danh thì chức danh bị thu hồi. Đồng thời ghi vào kho hồ sơ lý tính xét chức danh, đưa vào nền tảng chia sẻ thông tin tín dụng toàn quốc, thời hạn ghi là 3 năm. Phương pháp giám sát năm 2024 nêu tên mấy loại vi phạm của người khai: cam kết sai và thông tin bịa, bài báo bịa hoặc thuê viết, thành tích bịa, và nói nhờ gây áp lực. Lúc khai phải ký cam kết cá nhân, nếu cam kết không đúng thì 3 năm không được khai. Cơ quan nhân sự phối hợp với công an, cơ quan quản lý không gian mạng và thị trường để điều tra trang web giả, quảng cáo sai, bẫy hợp đồng, xét giả và chứng chỉ giả của trung gian. Vụ nghiêm trọng mà nghi ngờ phạm tội thì chuyển cơ quan có thẩm quyền xử lý (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội (人力资源和社会保障部令第 40 号) (2019). 职称评审管理暂行规定. ：第三十九条；Bộ Nhân sự và Xã hội (2024). 职称评审监管暂行办法. ：第五条、第十五条、第十七条、第二十七条；Văn phòng Bộ Nhân sự và Xã hội (2022). 关于进一步做好职称评审工作的通知. ：「未经备案的职称评审委员会，评审结果不纳入全国职称评审信息查询验证系统」；Tân Hoa Xinh (2024). 职称评审进入高峰，信息如何查？有何新变化？ ：「大家可通过www.12333.gov.cn网页、手机12333客户端、人力资源社会保障部门户网站、“人力资源和社会保障部”微信公众号等多种渠道进行登录」 +- Ghi chú: Có một cách phân biệt thật giả: chức danh do hội đồng xét hợp lệ cấp thì tra được trên nền tảng tra chứng chỉ xét chức danh toàn quốc. Cửa vào là trang 12333, ứng dụng 12333 và trang của Bộ Nhân sự và Xã hội. Hội đồng xét chưa đăng ký mà cấp thì hệ thống đó tra không ra. Vì vậy loại hứa「đảm bảo đỗ」,「không cần bạn dự thi」,「không tra được chứng chỉ vẫn dùng được」thì cơ bản có thể kết luận là có vấn đề. Người hưởng lợi chính là bạn, mức lợi ích định theo cách「tránh bị thu hồi chức danh và tránh 3 năm lưu vào hồ sơ mất uy tín」. + +### 23. Có chức danh không có nghĩa tăng lương: hỏi rõ đơn vị xét và tuyển theo tỷ lệ vị trí, hay xét rồi chưa chắc được tuyển + +- Chi phí: Không tốn tiền. Hỏi một lần bộ phận nhân sự của đơn vị là xong. +- Nói thẳng: Chức danh là căn cứ để tuyển dụng, đánh giá và thăng chức, nhưng được xét và được tuyển là hai chuyện. Đơn vị sự nghiệp thường xét trong tỷ lệ vị trí, được xét thì phần lớn được tuyển. Đơn vị không thực hiện quản lý theo vị trí thì có thể tách việc xét và việc tuyển, đã xét thì đơn vị không nhất thiết cho đúng vị trí và mức lương tương ứng. Đổi đơn vị, đổi địa phương thì chức danh có thể phải xét lại hoặc xác nhận. +- Lợi ích: Điều hai của mệnh lệnh số 40 của Bộ Nhân sự và Xã hội: kết quả xét chức danh là căn cứ quan trọng cho việc tuyển dụng, đánh giá, thăng tiến của nhân tài chuyên môn. Ý kiến cải cách năm 2016 quy định, đối với đơn vị sự nghiệp thực hiện quản lý theo vị trí đầy đủ thì tổ chức xét chức danh trong tỷ lệ cơ cấu vị trí. Đơn vị không thực hiện quản lý theo vị trí, cùng chuỗi có tính phổ thông cao và nghề mới, thì có thể tách việc xét và việc tuyển. Nhân tài chuyên môn chuyển vùng, chuyển đơn vị thì chức danh được xét lại hoặc xác nhận theo thẩm quyền quản lý, trừ khi nhà nước có quy định khác (toàn quốc) +- Mức bằng chứng: B +- Nguồn: Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội (人力资源和社会保障部令第 40 号) (2019). 职称评审管理暂行规定. ：第二条、第二十九条；Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「对于全面实行岗位管理、专业技术人才学术技术水平与岗位职责密切相关的事业单位，一般应在岗位结构比例内开展职称评审。对于不实行岗位管理的单位，以及通用性强、广泛分布在各社会组织的职称系列和新兴职业，可采用评聘分开方式。」 +- Ghi chú: Để ở mức B là vì văn bản chỉ viết cách nối việc xét với việc tuyển, không có con số thống nhất cả nước cho「xét được rồi tăng bao nhiêu». Phần làm theo được là: ở đơn vị sự nghiệp và doanh nghiệp nhà nước, hỏi bộ phận nhân sự về số vị trí trống và quy tắc tuyển của đơn vị. ở doanh nghiệp tư nhân, hỏi người chủ xem họ có công nhận tấm bằng đó không, có tăng tiền không. \ No newline at end of file diff --git "a/vi/book/24-\347\234\213\347\227\205.md" "b/vi/book/24-\347\234\213\347\227\205.md" index f84a89e1..8fdef1d3 100644 --- "a/vi/book/24-\347\234\213\347\227\205.md" +++ "b/vi/book/24-\347\234\213\347\227\205.md" @@ -1,6 +1,6 @@ [← Về mục lục](../README.md) -# 24. Khám bệnh: làm sao để ít tốn tiền và đỡ đi vòng +# 24. Khám bệnh: tốn ít tiền hơn và đi lại ít hơn Thước đo: tiền và thời gian. Phần này nói cách đi khám: vào bệnh viện nào trước, muốn lên tuyến trên thì làm thế nào, tiền có thể bớt chi ở đâu. Phần này còn nói cấp cứu xếp hàng thế nào, sau khi chữa chấn thương nặng cần làm những phép giám định nào, xin những giấy tờ nào. Xem phần 7 về những khoản trợ cấp có thể nhận. Xem phần 16 về cách quản lý bệnh mạn tính lâu dài. Xem phần 13 về việc đầu tiên phải làm ngay tại hiện trường. diff --git "a/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" "b/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" index 52b42379..9312ebfa 100644 --- "a/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" +++ "b/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" @@ -1,6 +1,6 @@ [← Về mục lục](../README.md) -# 25. Sau khi người thân mất thì phải làm gì +# 25. Sau khi có người qua đời Thước đo: tiền. Phần này chỉ nói thủ tục làm thế nào, tiền lấy ra sao, gồm lúc vừa mất thì làm gì trước, những khoản chi phí nào có bảng giá để đối chiếu. Phần này còn nói những khoản tiền nằm trong các tài khoản khác nhau có lấy lại được không, những việc nào người thân gần được đứng ra làm. Năm mục đầu xếp theo thứ tự sự việc xảy ra, lần lượt là hiện trường, giấy chứng minh tử vong, thi thể và thiêu, bất đồng về nguyên nhân tử vong thì xử lý sao, và huỷ đăng ký hộ khẩu. Các mục sau xếp theo hiệu quả chi phí. Những việc nên làm khi người thân còn sống, như giám hộ tự nguyện bằng văn bản, di chúc, tài khoản, xem phần 17. Ba khoản trợ cấp khi chết vì tai nạn lao động xem phần 19. diff --git "a/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" "b/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" index 35bf8b98..2a56dec8 100644 --- "a/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" +++ "b/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" @@ -114,37 +114,38 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Nguồn: Bộ Nhân sự và Xã hội và chín bộ (人力资源社会保障部等九部门) (2025). Thông báo về việc mở rộng thí điểm bảo hiểm tai nạn lao động cho người lao động có hình thức việc làm mới（关于扩大新就业形态人员职业伤害保障试点的通知）（人社部发〔2025〕24 号）kèm phụ lục《Biện pháp bảo hiểm tai nạn lao động cho người lao động có hình thức việc làm mới (thử nghiệm)》第二、九条. - Ghi chú: Đây vẫn là thí điểm, chưa phải chế độ chính thức có ở toàn quốc. Trước khi nhận việc hãy xác nhận tỉnh của bạn và sàn có nằm trong phạm vi không; nếu không thì chỉ còn lớp bảo hiểm thương mại mua bằng tiền của bạn. Một văn bản khác là《Ý kiến chỉ đạo về việc bảo vệ quyền lợi lao động của người lao động có hình thức việc làm mới》còn nêu thêm vài yêu cầu. Sàn phải trả mức thù lao không thấp hơn lương tối thiểu địa phương cho người làm việc đúng quy định, phải trả đủ và đúng hạn, không được khấu trừ hoặc nợ đãng không lý do. Ngày lễ theo luật còn phải trả mức thù lao hợp lý. Số văn bản của tài liệu này là 人社部发〔2021〕56 号, mục làm việc linh hoạt cũng dẫn nó, xem mục 11 của phần này. Khi có sự cố thì làm theo bộ này trước, đừng tự thỏa thuận riêng với điểm giao. Phần bảo hiểm y tế và hưu trí vẫn phải tự tham gia, xem mục 11 của phần này. -### 13. Hai lối chốt được biên chế ngay ngày điền nguyện vọng thi đại học: sinh viên sư phạm hưởng kinh phí và sinh viên y khoa định hướng, cái giá là 6 năm thực hiện nghĩa vụ +### 13. Hai con đường chốt được chỗ làm việc có biên chế ngay từ ngày điền nguyện vọng kỳ thi gaokao: sinh viên sư phạm có học phí và sinh viên y khoa định hướng, cái giá là thực hiện cam kết 6 năm -- Chi phí: Không tốn tiền. Học phí và tiền ăn ở do nhà nước trả, còn được trợ cấp sinh hoạt. Cái giá là sau khi tốt nghiệp, 6 năm không tự do chọn việc, vi phạm thì phải hoàn tiền, nộp tiền phạt và bị ghi vào lịch sử tín dụng. Chỗ khó là 6 năm đó không đổi được nơi. -- Nói thẳng: Cả hai lối đều là nhà nước trả trước tiền cho bạn, sau khi tốt nghiệp bạn đi làm đúng chỗ đã ghi trong thỏa thuận trong 6 năm. Sinh viên sư phạm hưởng kinh phí chỉ tuyển ở sáu trường đại học sư phạm trực thuộc Bộ Giáo dục, nhận ở đợt xét tuyển sớm. Sinh viên y khoa định hướng phải ký thỏa thuận ba bên trước khi nhận được giấy báo nhập học. Vi phạm thì phải hoàn tiền, nộp tiền phạt, và còn bị ghi vào lịch sử tín dụng. Nghĩ kỹ có chịu được 6 năm đó không rồi hãy điền nguyện vọng. -- Lợi ích: Sinh viên sư phạm hưởng kinh phí xem Biện pháp giáo dục kinh phí cho sinh viên sư phạm liên thông từ đại học lên sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục《教育部直属师范大学本研衔接师范生公费教育实施办法》（năm 2024, biện pháp năm 2018 bị bãi bỏ đồng thời）. Điều hai: ngân sách trung ương chịu học phí và tiền ăn ở trong thời gian học sinh viên đang học, đồng thời cấp trợ cấp sinh hoạt. Điều ba: chủ yếu đào tạo cho các địa (thành phố, châu, liên minh) trở xuống, ngoài thủ phủ tỉnh ở vùng trung tây và vùng tây. Biện pháp còn ghi「不得定向到直辖市、计划单列市或省会城市主城区任教（五个自治区、陆地边境省份、海南省、贵州省、青海省除外）」, tức không được định hướng đi dạy ở thành phố trực thuộc trung ương, thành phố tách riêng theo kế hoạch hay khu trung tâm thủ phủ tỉnh (trừ năm khu tự trị, các tỉnh giáp biên đất liền, tỉnh Hải Nam, tỉnh Quý Châu, tỉnh Thanh Hải). Điều chín: sau khi tốt nghiệp thì về chung địa (thành phố, châu, liên minh) định hướng theo tỉnh có hộ khẩu để dạy ở trường tiểu học và trung học,「承诺从事中小学教育工作 6 年以上」, tức cam kết làm công tác giáo dục tiểu học và trung học trên 6 năm. Ai làm việc ở trường đô thị thì「应到农村义务教育学校任教服务至少 1 年」, tức phải đi dạy phục vụ tại trường giáo dục bắt buộc nông thôn ít nhất 1 năm. Điều mười ba: không thực hiện nghĩa vụ hoặc thời hạn thực hiện không đủ thì chịu trách nhiệm theo thỏa thuận. Một trong các hậu quả là「违约行为纳入信用记录，并及时与全国信用信息共享平台共享，违约相关材料归入本人人事档案」, tức hành vi vi phạm được đưa vào lịch sử tín dụng, kịp thời chia sẻ với nền tảng chia sẻ thông tin tín dụng toàn quốc, hồ sơ liên quan tới vi phạm đưa vào hồ sơ nhân sự của bản thân. Câu nữa là「履约任务完成前不得到定向地（市、州、盟）之外从教，不得被招录（聘用）为其他公务人员。完成履约任务后即修复相关信用」, tức chưa hoàn thành nghĩa vụ thì không được đi dạy ngoài địa (thành phố, châu, liên minh) định hướng, không được tuyển dụng (thuê mời) làm công chức khác; hoàn thành nghĩa vụ thì khôi phục lại uy tín tín dụng liên quan. Trong thời gian chưa đủ thời hạn dịch vụ thỏa thuận thì không xin được chức vụ công lập khác; đủ thời hạn thì lịch sử tín dụng mới được khôi phục. Điều sáu: sau khi nhận tuyển, nếu qua khảo sát thấy không phù hợp để đi dạy thì trong với 1 năm trả lại khoản kinh phí giáo dục đã hưởng và chuyển sang chuyên ngành không phải sư phạm. Điều mười bốn: không có giấy chứng nhận trình độ và bằng thạc sĩ sư phạm cùng giấy chứng nhận giáo viên trong thời hạn học quy định thì xử lý như vi phạm, hoàn tiền và nộp tiền phạt. Điều hai mươi mốt: các nơi giải quyết biên chế cần thiết để sinh viên sư phạm hưởng kinh phí đến dạy ở trường tiểu học và trung học, trong tổng số biên chế đơn vị sự nghiệp hiện có. Sinh viên y khoa định hướng miễn phí đào tạo theo đơn đặt của nông thôn xem một thông báo khác. Sau khi nhận tuyển, trước khi nhận được giấy báo nhập học, phải ký thỏa thuận với trường đào tạo và với sở y tế, sở nhân sự và xã hội cấp huyện nơi định hướng việc làm. Trong thỏa thuận phải「承诺毕业后到定向农村基层医疗卫生机构服务 6 年」, tức cam kết sau khi tốt nghiệp phục vụ 6 năm tại cơ sở y tế cơ sở nông thôn định hướng. Thực hiện nghĩa vụ là đi đủ thời hạn ở nơi đã ghi trong thỏa thuận. Sau khi thực hiện nghĩa vụ và báo cáo đi làm, phải học 3 năm đào tạo chuyên khoa bác sĩ nội trú theo tiêu chuẩn hoá toàn diện, gọi tắt là đào tạo chuẩn hoá. Đào tạo chuẩn hoá là giai đoạn sau khi tốt nghiệp, ở trong bệnh viện tiếp tục rèn tay nghề theo tiêu chuẩn thống nhất. Trong thời gian đào tạo, quan hệ nhân sự và lương cũ giữ nguyên. Hồ sơ vi phạm đưa vào hồ sơ nhân sự cá nhân, và là căn cứ đánh giá uy tín khi tuyển dụng (tuyển việc) công khai ở cơ quan hành chính, đơn vị sự nghiệp như cơ sở y tế, khi nhận đào tạo chuyên khoa bác sĩ nội trú, khi nhận học cao học. Người cố ý vi phạm sẽ bị công bố toàn quốc trên trang「信用中国」, và chịu trừng phạt chung vì mất uy tín. Thời gian đào tạo bị vi phạm thì không tính vào thời hạn phục vụ. Người không hoàn tiền và tiền phạn theo thỏa thuận thì cơ quan y tế cấp huyện truy thu qua đường pháp lý (toàn quốc) +- Chi phí: Không tốn tiền. Cả học phí và tiền ở trong cũng do Nhà nước chi trả, lại còn có trợ cấp sinh hoạt. Cái giá là 6 năm sau khi tốt nghiệp không được tự do chọn việc, vi phạm thì phải hoàn tiền, nộp phạt và bị ghi vào lịch sử tín dụng. Chỗ khó là trong 6 năm ấy không đổi được nơi. +- Nói thẳng: Cả hai con đường đều là Nhà nước trả hộ bạn trước, sau khi tốt nghiệp bạn đi làm đúng chỗ đã cam kết trong 6 năm. Sinh viên sư phạm có học phí chỉ tuyển ở sáu trường đại học sư phạm trực thuộc Bộ Giáo dục, nhận theo lô trước. Sinh viên y khoa định hướng phải ký thỏa thuận ba bên trước khi nhận được giấy báo nhập học. Vi phạm thì hoàn tiền, nộp phạt, còn bị ghi vào lịch sử tín dụng. Hãy nghĩ kỹ chuyện có chịu được 6 năm ấy không rồi mới điền nguyện vọng. +- Lợi ích: Phần sinh viên sư phạm có học phí thì xem《Biện pháp thực hiện chế độ giáo dục có học phí cho sinh viên sư phạm liên thông bậc đại học và sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục》(năm 2024, biện pháp năm 2018 bãi bỏ cùng lúc). Điều hai: học phí và tiền ở trong trong thời gian học do ngân sách trung ương chi trả, có trợ cấp sinh hoạt. Điều ba: chủ yếu đào tạo ở các địa phương cấp địa (tỉnh, khu, thành phố, châu) và cấp dưới, ngoài các thủ phố của khu vực trung và phía Tây. Điều đó còn viết「不得定向到直辖市、计划单列市或省会城市主城区任教（五个自治区、陆地边境省份、海南省、贵州省、青海省除外）」, tức không được định hướng dạy ở khu chính thành của thành phố trực thuộc trung ương, thành phố tách riêng theo kế hoạch, hoặc thủ phủ tỉnh (trừ năm khu tự trị, các tỉnh biên giới trên mặt đất, tỉnh Hải Nam, tỉnh Quý Châu, tỉnh Thanh Hải). Điều chín: sau khi tốt nghiệp thì về dạy ở trường tiểu học và trung học của tỉnh gốc theo định hướng đã cam kết,「承诺从事中小学教育工作 6 年以上」, tức cam kết dạy ở trường tiểu học và trung học từ 6 năm trở lên. Người đi làm ở trường ở đô thị thì「应到农村义务教育学校任教服务至少 1 年」, tức phải dạy ở trường giáo dục bắt buộc nông thôn ít nhất 1 năm. Điều mười ba: không thực hiện hoặc không đủ thời gian thực hiện thì chịu trách nhiệm theo thỏa thuận. Một hậu quả là「违约行为纳入信用记录，并及时与全国信用信息共享平台共享，违约相关材料归入本人人事档案」, tức hành vi vi phạm được ghi vào lịch sử tín dụng và kịp thời chia sẻ với nền tảng thông tin tín dụng toàn quốc, hồ sơ liên quan đến vi phạm đưa vào hồ sơ nhân sự của chính người đó. Câu khác là「履约任务完成前不得到定向地（市、州、盟）之外从教，不得被招录（聘用）为其他公务人员。完成履约任务后即修复相关信用」, tức chưa hoàn thành nghĩa vụ thì không được dạy ở ngoài địa phương định hướng, không được tuyển hay bổ nhiệm làm công chức khác; hoàn thành nghĩa vụ thì lịch sử tín dụng liên quan được khôi phục. Chưa đủ thời gian phục vụ ghi trong thỏa thuận thì không dự được chức vụ công lập khác, đủ thời gian thì lịch sử tín dụng mới trở lại. Điều sáu: sau khi nhập học mà xét thấy không phù hợp làm giáo viên thì trong 1 năm nhập học phải hoàn lại chi phí giáo dục có học phí đã được hưởng, và chuyển sang chuyên ngành phi sư phạm. Điều mười bốn: không có bằng thạc sĩ giáo dục và giấy chứng nhận giáo viên trong thời gian học quy định thì coi như vi phạm, hoàn tiền và nộp phạt. Điều hai mươi mốt: các địa phương phải giải quyết chỗ làm việc cần thiết cho sinh viên sư phạm có học phí dạy ở trường tiểu học và trung học trong tổng số biên chế sự vệ hiện có. Phần sinh viên y khoa định hướng miễn phí theo đơn đặt hàng nông thôn thì xem một thông báo khác. Sau khi nhập học, trước khi nhận được giấy báo nhập học, phải ký thỏa thuận với trường đào tạo và cơ quan quản lý y tế, nhân sự xã hội cấp huyện nơi định hướng việc làm. Trong thỏa thuận phải「承诺毕业后到定向农村基层医疗卫生机构服务 6 年」, tức cam kết sau khi tốt nghiệp sẽ làm việc 6 năm tại cơ sở y tế cơ sở nông thôn được chỉ định. Thực hiện cam kết nghĩa là đi đúng chỗ đã hẹn làm đủ thời gian. Sau khi đến báo cáo và làm việc, phải tham gia đào tạo chuyên khoa bác sĩ đa khoa hồi sức theo tiêu chuẩn thống nhất trong 3 năm, gọi tắt là đào tạo quy chuẩn. Đào tạo quy chuẩn là giai đoạn sau khi tốt nghiệp tiếp tục rèn nghề trong bệnh viện theo tiêu chuẩn thống nhất. Trong thời gian đào tạo thì quan hệ nhân sự và lương cũ giữ nguyên. Hồ sơ vi phạm đưa vào hồ sơ nhân sự cá nhân và làm căn cứ đánh giá lòng tin trong tuyển dụng công khai của các cơ quan hành chính sự vệ như cơ quan y tế, trong tuyển dục bác sĩ đa khoa hồi sức và trong tuyển sinh sau đại học. Người vi phạm cố ý sẽ được niêm yết toàn quốc qua trang「信用中国」, và bị trừng phạt liên ngành vì mất uy tín. Thời gian đào tạo bị vi phạm thì không tính vào thời gian phục vụ. Người không hoàn lại phí và tiền phạt theo thỏa thuận thì cơ quan quản lý y tế cấp huyện truy thu qua đường pháp lý (toàn quốc) - Mức bằng chứng: A -- Nguồn: Văn phòng Hội đồng Nhà nước (2024). Thông báo chuyển Biện pháp giáo dục kinh phí cho sinh viên sư phạm liên thông từ đại học lên sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục và các cơ quan（国办发〔2024〕27 号，điều hai, ba, sáu, chín, mười ba, mười bốn, hai mươi mốt）. ；Ủy ban Quốc gia Y tế và Sức khoẻ, Văn phòng Tổ chức cán bộ trung ương, Ủy ban Phát triển và Cải cách quốc gia, Bộ Giáo dục, Bộ Tài chính, Bộ Nhân sự và Xã hội, Cục Quản lý y dược trung ương (2019). Thông báo về việc làm tốt công tác bố trí việc làm và quản lý nghĩa vụ thực hiện cho sinh viên y khoa định hướng miễn phí đào tạo theo đơn đặt của nông thôn. -- Ghi chú: Biện pháp năm 2024 đổi sinh viên sư phạm hưởng kinh phí thành「本研衔接」, nghĩa là học tiếp thạc sĩ sư phạm ngay sau đại học. Biện pháp cũ năm 2018 đã bị bãi bỏ, câu「tốt nghiệp đại học là dạy thẳng」còn truyền trên mạng là cách nói cũ. Đã ký thỏa thuận trước khi biện pháp ban hành thì theo thỏa thuận cũ.「Có biên chế」không có nghĩa「vào làm rồi thì không phải lo gì」: biên chế phải bù ra từ tổng số hiện có, vị trí cụ thể do thành phố cấp huyện tổ chức tuyển riêng cho trường sử dụng, bạn và trường cùng chọn nhau. Sinh viên y khoa định hướng hết thời hạn phục vụ mà muốn đi thì được, văn bản ghi「服务期满，对自主择业的应予同意」, tức hết thời hạn phục vụ thì phải chấp thuận cho người tự chọn việc. Người đã vi phạm cũng có cách xoá hồ sơ vi phạm, cái giá là làm đủ theo thỏa thuận cũ, hoặc chuyển sang cơ sở khác ở nông thôn trong tỉnh rồi làm đủ ít nhất 6 năm, tiền và tiền phạt đã nộp thì không hoàn lại. Với sinh viên y khoa còn phải cộng thêm 3 năm đào tạo chuẩn hoá, cả chu kỳ dài hơn ngành sư phạm. Mỗi tỉnh còn có kế hoạch riêng về sinh viên sư phạm hưởng kinh phí và sinh viên y khoa định hướng, điều kiện do tỉnh quy định, trước khi điền nguyện vọng hãy xem văn bản tuyển sinh năm đó của tỉnh. +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2024). Thông báo chuyển tiếp các biện pháp của Bộ Giáo dục và các bộ về chế độ giáo dục có học phí cho sinh viên sư phạm liên thông bậc đại học và sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục（国办发〔2024〕27 号，第二、三、六、九、十三、十四、二十一条）. ；Ủy ban Quốc gia Y tế, Văn phòng Tổ chức cán bộ trung ương, Ủy ban Phát triển và Cải cách quốc gia, Bộ Giáo dục, Bộ Tài chính, Bộ Nhân sự và Xã hội, Cục Y học cổ truyền quốc gia (国家卫生健康委、中央编办、国家发展改革委、教育部、财政部、人力资源社会保障部、国家中医药局) (2019). Thông báo về việc làm tốt công tác bố trí việc làm và quản lý thực hiện cam kết cho sinh viên y khoa được đào tạo miễn phí theo đơn đặt hàng nông thôn. +- Ghi chú: Bản năm 2024 đã đổi sinh viên sư phạm có học phí thành hình thức「liên thông», tức học xong bậc đại học thì học tiếp thạc sĩ giáo dục. Biện pháp cũ năm 2018 đã bãi bỏ, cái「tốt nghiệp đại học là dạy thẳng」còn truyền trên mạng là cách nói cũ. Người đã ký thỏa thuận trước khi bản này ban hành thì theo thỏa thuận cũ.「Có biên chế」không có nghĩa là「vào là xong」: biên chế phải bố trí được trong tổng số hiện có, vị trí cụ thể do thành phố tổ chức trường tuyển dụng riêng, bạn và trường cùng chọn nhau. Sinh viên y khoa định hướng hết hạn mà muốn đi thì được, văn bản ghi「服务期满，对自主择业的应予同意」. Người đã vi phạm cũng có cách xoá hồ sơ vi phạm, cái giá là theo thỏa thuận cũ, hoặc đổi sang cơ sở khác trong tỉnh và làm đủ ít nhất 6 năm nữa, phí và tiền phạt đã nộp không hoàn lại. Sinh viên y khoa còn phải tính thêm thời gian đào tạo quy chuẩn 3 năm, nên toàn chu kỳ dài hơn ngành sư phạm. Mỗi tỉnh còn có kế hoạch riêng cho sinh viên sư phạm có học phí và sinh viên y khoa định hướng, điều kiện do tỉnh quy định, trước khi điền nguyện vọng hãy xem tài liệu tuyển sinh năm đó của tỉnh mình. -### 14. Muốn đi làm ở nước ngoài, trước hết tra công ty này có giấy phép hợp tác lao động xuất khẩu hay không: họ không được thu tiền ký quỹ của bạn +### 14. Muốn đi làm công ăn nước ngoài, trước hết tra công ty này có giấy phép hợp tác lao động ngoài biên giới không: công ty không được đòi bạn đặt tiền -- Chi phí: Không tốn tiền. Trước khi đi, bỏ mười mấy phút kiểm tra tư cách công ty, rồi đọc hết hai bản hợp đồng. -- Nói thẳng: Chuyện đi làm ở nước ngoài, pháp luật quản không phải bạn mà là công ty tổ chức đưa bạn đi. Chỉ doanh nghiệp đã có giấy chứng nhận hợp tác lao động xuất khẩu và đã đăng ký mới làm được. Nghe câu「xin visa du lịch sang trước rồi đổi sang visa lao động」chính câu đó cho thấy đối phương đang vi phạm. Công ty cũng không được tuỳ danh nghĩa nào mà thu tiền ký quỹ của bạn, hay bắt bạn đặt tài sản làm bảo đảm. -- Lợi ích: Quy định về hợp tác lao động xuất khẩu《对外劳务合作管理条例》Điều năm: làm hợp tác lao động xuất khẩu phải được cơ quan quản lý thương mại cấp tỉnh hoặc thành phố trực thuộc tỉnh phê duyệt và được cấp tư cách kinh doanh. Tư cách kinh doanh đó chính là giấy chứng nhận hợp tác lao động xuất khẩu, không có thì không được tổ chức người ra nước ngoài làm việc. Điều bảy: danh sách giấy chứng nhận báo lên cơ quan quản lý thương mại của Hội đồng Nhà nước, và thông báo cho các đại sứ, lãnh sự của Trung Quốc ở nước ngoài. Điều này còn ghi「未依法取得对外劳务合作经营资格证书并办理登记，不得从事对外劳务合作」, tức chưa có giấy chứng nhận hợp tác lao động xuất khẩu và đăng ký theo pháp luật thì không được làm hợp tác lao động xuất khẩu. Điều tám:「任何单位和个人不得以商务、旅游、留学等名义组织劳务人员赴国外工作」, tức không cá nhân, tổ chức nào được tổ chức người lao động ra nước ngoài làm việc với danh nghĩa thương mại, du lịch, du học. Điều chín: doanh nghiệp phải trong vòng 5 ngày làm việc kể từ ngày đăng ký nộp khoản「不低于 300 万元人民币的对外劳务合作风险处置备用金」tại ngân hàng chỉ định, tức khoản tiền dự phòng xử lý rủi ro hợp tác lao động xuất khẩu không dưới 3 triệu tệ, danh sách nộp tiền công bố cho xã hội. Điều mười: khoản dự phòng dùng để bồi thường các khoản phí dịch vụ phải trả lại, thù lao lao động và thiệt hại phải đền theo pháp luật mà doanh nghiệp từ chối nhận hoặc không có khả năng nhận. Khoản đó còn dùng cho「因发生突发事件，劳务人员回国或者接受紧急救助所需费用」, tức chi phí để người lao động về nước hoặc được cứu trợ khẩn cấp khi xảy ra sự cố bất ngờ. Điều mười một: không được tổ chức người lao động ra nước ngoài làm công việc liên quan tới cờ bạc, hoạt động tình dục. Điều mười hai: không tổ chức đào tạo thì không được tổ chức đi nước ngoài. Điều mười ba: phải mua bảo hiểm tai nạn cá nhân cho người lao động trong thời gian làm việc ở nước ngoài. Điều hai mươi mốt, hai mươi ba: phải ký hợp đồng bằng văn bản riêng với nhà tuyển dụng nước ngoài và riêng với người lao động, không ký thì không được tổ chức đi nước ngoài. Điều hai mươi lăm:「不得以任何名目向劳务人员收取押金或者要求劳务人员提供财产担保」, tức không được tuỳ danh nghĩa nào thu tiền ký quỹ từ người lao động, hay yêu cầu người lao động đặt tài sản làm bảo đảm. Điều hai mươi sáu: trong vòng 10 ngày làm việc kể từ ngày ký hợp đồng phải báo đăng ký với cơ quan quản lý thương mại. Điều hai mươi chín: quyền lợi thực tế của người lao động không khớp với hợp đồng thì doanh nghiệp phải hỗ trợ yêu cầu bồi thường; không được bồi thường đầy đủ thì có quyền yêu cầu doanh nghiệp chịu trách nhiệm bồi thường. Điều ba mươi mốt: nước hoặc vùng lãnh thổ có tình hình an ninh khó bảo đảm an toàn cá nhân thì doanh nghiệp không được tổ chức người lao động đến đó. Điều ba mươi bảy: cơ quan quản lý thương mại lập cơ chế lưu hồ sơ tín dụng xấu và công bố. Chế tài phạt nằm ở các điều sau. Điều ba mươi chín: không đủ tư cách làm hợp tác lao động xuất khẩu thì bị xử lý, cấm đoạt, cấu thành tội phạm thì truy cứu trách nhiệm hình sự. Điều bốn mươi: tổ chức người đi nước ngoài với danh nghĩa thương mại, du lịch, du học, cho vay tư cách, tổ chức làm công việc liên quan cờ bạc và tình dục thì thu hồi giấy chứng nhận hợp tác lao động xuất khẩu. Điều bốn mươi ba: không ký hợp đồng, hoặc che giấu thông tin, cung cấp thông tin giả để tổ chức người đi nước ngoài thì phạt từ 10 vạn tệ đến 20 vạn tệ, còn phải phạt người phụ trách chính từ 2 vạn tệ đến 5 vạn tệ (toàn quốc) +- Chi phí: Không tốn tiền. Trước khi xuất hành dành mười mấy phút để kiểm tra tư cách của công ty, rồi đọc kỹ cả hai bản hợp đồng. +- Nói thẳng: Chuyện đi làm công ăn nước ngoài, pháp luật không quản lý bạn mà quản lý công ty tổ chức cho bạn đi. Chỉ doanh nghiệp đã có giấy chứng nhận tư cách kinh doanh hợp tác lao động ngoài biên giới và đã đăng ký mới làm được việc này. Nghe câu「trước tiên ký visa du lịch qua rồi đổi sang visa lao động」thì bản thân câu đó đã chứng minh phía bên kia đang vi phạm pháp. Công ty cũng không được đòi bạn đặt tiền dưới bất kỳ danh nghĩa nào, hay bắt bạn dùng tài sản làm bảo đảm. +- Lợi ích: Điều năm của《Quy chế quản lý hợp tác lao động ngoài biên giới》: muốn làm hợp tác lao động ngoài biên giới thì phải được cơ quan quản lý thương mại cấp tỉnh hoặc thành phố trực thuộc phường phê duyệt và có tư cách kinh doanh. Tư cách kinh doanh chính là giấy chứng nhận tư cách kinh doanh hợp tác lao động ngoài biên giới đó, không có nó thì không được tổ chức người đi làm công ăn ở nước ngoài. Điều bảy: danh sách giấy phép phải báo cơ quan quản lý thương mại của Hội đồng Nhà nước và thông báo cho đại sứ quán và lãnh sự quán Trung Quốc ở nước ngoài. Điều này còn viết「未依法取得对外劳务合作经营资格证书并办理登记，不得从事对外劳务合作」. Điều tám:「任何单位和个人不得以商务、旅游、留学等名义组织劳务人员赴国外工作」. Điều chín: doanh nghiệp phải trong 5 ngày làm việc kể từ ngày đăng ký nộp tại ngân hàng chỉ định khoản「不低于 300 万元人民币的对外劳务合作风险处置备用金」, tức khoản dự phòng xử lý rủi ro hợp tác lao động ngoài biên giới không dưới 300 vạn tệ nhân dân tệ, danh sách nộp tiền được công bố xã hội. Điều mười: khoản dự phòng dùng để chi trả các khoản phí dịch vụ phải hoàn trả, tiền công và tổn thương phải bồi thường theo luật mà doanh nghiệp từ chối hoặc không có khả năng chi trả. Nó còn dùng cho「因发生突发事件，劳务人员回国或者接受紧急救助所需费用」, tức khoản chi phí để người lao động về nước hoặc được cứu hộ khẩn cấp khi xảy ra sự cố. Điều mười một: không được tổ chức người lao động sang nước ngoài làm công việc liên quan đến cờ bạc và tình dục. Điều mười hai: không tổ chức đào tạo thì không được tổ chức đi nước ngoài. Điều mười ba: phải mua bảo hiểm tai nạn cá nhân cho người lao động trong thời gian làm việc ở nước ngoài. Điều hai mươi mốt và hai mươi ba: phải ký hợp đồng bằng văn bản với nhà tuyển dụng nước ngoài và với người lao động, không ký thì không được tổ chức đi nước ngoài. Điều hai mươi lăm:「不得以任何名目向劳务人员收取押金或者要求劳务人员提供财产担保」. Điều hai mươi sáu: trong 10 ngày làm việc kể từ ngày ký hợp đồng thì báo cơ quan quản lý thương mại để đăng ký. Điều hai mươi chín: quyền lợi thực tế của người lao động không khớp với hợp đồng thì doanh nghiệp phải hỗ trợ khiếu nại; không được bồi thường đúng phải thì có quyền yêu cầu doanh nghiệp chịu trách nhiệm bồi thường. Điều ba mươi mốt: nếu tình hình an ninh của nước hoặc vùng đó không bảo đảm được an toàn cho con người thì doanh nghiệp không được tổ chức người lao động đi. Điều ba mươi bảy: cơ quan quản lý thương mại lập hệ thống hồ sơ và công bố uy tín xấu. Hình phạt nằm ở các điều sau. Điều ba mươi chín: không có tư cách làm hợp tác lao động ngoài biên giới thì bị kiểm tra, đình chỉ; nếu cấu thành tội phạm thì truy trách nhiệm hình sự. Điều bốn mươi: tổ chức đi nước ngoài mang danh nghĩa thương mại, du lịch, du học, cho vay tư cách, hoặc tổ chức làm công việc liên quan đến cờ bạc tình dục thì thu hồi giấy chứng nhận tư cách kinh doanh. Điều bốn mươi ba: không ký hợp đồng, hoặc che giấu thông tin, cung cấp thông tin giả để tổ chức đi nước ngoài thì phạt từ 10 vạn tệ đến 20 vạn tệ, đồng thời phạt người chịu trách nhiệm chính từ 2 vạn tệ đến 5 vạn tệ (toàn quốc) - Mức bằng chứng: A -- Nguồn: Hội đồng Nhà nước (2012). Quy định về hợp tác lao động xuất khẩu（国务院令第 620 号，điều năm, bảy, tám, chín, mười, mười một, mười hai, mười ba, hai mươi mốt, hai mươi ba, hai mươi lăm, hai mươi sáu, hai mươi chín, ba mươi mốt, ba mươi bảy, ba mươi chín, bốn mươi, bốn mươi ba）. -- Ghi chú: Quy trình đúng là thế này: công ty ký hợp đồng hợp tác lao động bằng văn bản với nhà tuyển dụng nước ngoài trước, rồi ký hợp đồng dịch vụ bằng văn bản hoặc hợp đồng lao động với bạn. Hợp đồng phải ghi rõ nội dung công việc, thời hạn, mức thù lao, thời giờ làm, ngày phép, bảo hiểm xã hội và rủi ro an toàn cá nhân. Trong vòng 10 ngày làm việc sau khi ký, công ty phải báo đăng ký với cơ quan quản lý thương mại. Công ty còn phải bố trí cho bạn học kỹ năng, phòng chống an toàn, ngoại ngữ và pháp luật tập quán địa phương, và mua bảo hiểm tai nạn cá nhân cho thời gian làm việc ở nước ngoài. Đối phương không cho xem hợp đồng, đòi tiền ký quỹ, thúc bạn đi trước rồi bổ sung thủ tục sau, chính là loại công ty phải tránh. Chiêu tuyển dụng lương cao ở nước ngoài và khu công nghiệp ứng dụng lừa đảo chuyển tiền xem mục 5 ở phần 21, những cảnh báo an toàn phải tra trước khi đi và 12308 nằm ở phần 21, không lặp lại. Quy định không đặt ngưỡng học vấn hay tuổi cho người đi làm, ngưỡng nằm ở tư cách của công ty và ở việc nước nơi đi làm có cấp phép hay không. Công ty chỉ được ký hợp đồng với doanh nghiệp hay tổ chức đã được nước đó phê duyệt, không được ký với cá nhân ở nước ngoài. Khoản dự phòng do cơ quan chủ quản dùng khi công ty không trả, bạn không tự rút được. Nên khi thật sự có chuyện thì phải khiếu nại lên cơ quan quản lý thương mại trước, Điều hai mươi trao cho bạn quyền khiếu nại đó. Gửi hơn 100 người sang cùng một nước hoặc vùng lãnh thổ thì phải bố trí quản lý đi kèm và báo đăng ký với đại sứ, lãnh sự. Dự án ít người không có lớp bảo đảm đó, càng phải tự giữ hợp đồng và biên nhận chuyển tiền. +- Nguồn: Hội đồng Nhà nước (国务院) (2012). Quy chế quản lý hợp tác lao động ngoài biên giới（对外劳务合作管理条例）（Hội đồng Nhà nước Lệnh số 620，第五、七、八、九、十、十一、十二、十三、二十一、二十三、二十五、二十六、二十九、三十一、三十七、三十九、四十、四十三条）. +- Ghi chú: Trình tự làm đúng phải thế này: công ty trước hết ký hợp đồng hợp tác lao động bằng văn bản với nhà tuyển dụng nước ngoài, rồi ký hợp đồng dịch vụ bằng văn bản hoặc hợp đồng lao động với bạn. Hợp đồng phải ghi rõ nội dung công việc, thời hạn, mức thù lao, thời giờ làm việc, ngày nghỉ, bảo hiểm xã hội và rủi ro an toàn cá nhân. Sau khi ký, trong 10 ngày làm việc công ty phải báo cơ quan quản lý thương mại để đăng ký. Công ty còn bắt buộc phải sắp xếp cho bạn học kỹ năng, phòng chống xâm hại, ngoại ngữ và pháp luật tập quán địa phương, đồng thời mua cho bạn bảo hiểm tai nạn cá nhân trong thời gian làm việc ở nước ngoài. Bên kia không cho bạn xem hợp đồng, đòi tiền đặt, hay thúc bạn đi trước rồi bổ sung thủ tục sau thì chính là loại công ty phải tránh. Bảng lừa đảo tuyển dụng lương cao ở nước ngoài và khu công nghiệp lừa đảo xem mục 5 ở phần 21; nhắc nhở an toàn cần tra trước khi xuất hành và mã 12308 nằm ở phần 21, không lặp lại. Quy chế không đặt ngưỡng trình độ hay tuổi cho người đi làm công ăn, ngưỡng nằm ở tư cách của công ty và ở việc nước đi có cấp phép hay không. Công ty chỉ được ký hợp đồng với doanh nghiệp hoặc tổ chức được nước đó cho phép, không được ký với cá nhân ở nước ngoài. Khoản dự phòng do cơ quan chủ quản dùng vào khi công ty không trả nợ, bạn không tự rút được. Nên khi thật sự có sự cố vẫn phải khiếu nại với cơ quan quản lý thương mại trước, Điều hai mươi trao cho bạn quyền khiếu nại đó. Dự án cử trên 100 người sang cùng một nước hoặc vùng thì phải cử quản lý đi kèm và báo đại sứ quán, lãnh sự quán để đăng ký. Dự án ít người không có lớp bảo đảm đó, càng phải tự giữ hợp đồng và chứng tử chuyển khoản cho kỹ. -### 15. Làm việc từ xa cho công ty ở nước ngoài tại nhà, khoản tiền này tính là thu nhập trong nước theo luật thuế: thuế cá nhân tự kê, chuyển tiền về đi theo khoản「职工报酬」 +### 15. Làm việc từ xa cho công ty nước ngoài ngay tại nhà, khoản tiền này theo luật thuế tính là thu nhập trong nước: thuế thu nhập cá nhân tự kê khai, tiền về nước đi theo khoản「thù lao lao động」 -- Chi phí: Không tốn tiền. Mỗi năm làm một lần tổng hợp quyết toán thuế, thời gian là từ ngày 1 tháng 3 đến ngày 30 tháng 6 của năm sau. Khi chuyển tiền về vượt hạn mức cả năm thì còn phải chuẩn bị hợp đồng lao động và chứng minh thu nhập. Chỗ khó là khoản thuế này bạn phải tự nhớ đi kê. -- Nói thẳng: Khoản tiền đó tính trong nước hay ngoài nước phụ thuộc bạn làm việc ở đâu, không phụ thuộc tiền chuyển vào từ đâu. Người ở trong nước làm việc từ xa cho công ty ở nước ngoài thì tính là thu nhập trong nước, phải nộp thuế ở trong nước. Công ty nước ngoài không có cơ quan tại Trung Quốc thì không ai khấu trừ thay bạn, khoản thuế này bạn phải tự đi kê. Tiền chuyển về đi theo khoản「职工报酬」. -- Lợi ích: Khoản tiền đó có tính là thu nhập trong nước hay không thì xem Quy định thi hành Luật Thuế thu nhập cá nhân（中国 个人所得税法实施条例）（国务院令第 707 号，áp dụng từ ngày 1 tháng 1 năm 2019）Điều ba. Quy định ghi:「除国务院财政、税务主管部门另有规定外，下列所得，不论支付地点是否在中国境内，均为来源于中国境内的所得：（一）因任职、受雇、履约等在中国境内提供劳务取得的所得」, tức trừ trường hợp cơ quan tài chính, thuế của Hội đồng Nhà nước có quy định khác, thì các khoản thu nhập sau đây, bất kể chỗ trả tiền có ở trong nước Trung Quốc hay không, đều là thu nhập phát sinh từ trong nước Trung Quốc: một là thu nhập từ việc cung cấp lao động trong nước Trung Quốc do giữ chức vụ, được thuê, thực hiện hợp đồng. Điều sáu của cùng quy định đó định nghĩa hai khoản thuế. Một là lương, công:「工资、薪金所得，是指个人因任职或者受雇取得的工资、薪金、奖金、年终加薪、劳动分红、津贴、补贴以及与任职或者受雇有关的其他所得」, tức thu nhập lương, công là thu nhập cá nhân có được do giữ chức vụ hoặc được thuê gồm lương, công, thưởng, lương tăng cuối năm, chia lợi nhuận lao động, trợ cấp, bồi đắm và các thu nhập khác liên quan tới chức vụ hoặc quan hệ thuê mướn. Khoản kia là thù lao lao động:「劳务报酬所得，是指个人从事劳务取得的所得，包括从事设计、装潢、安装、制图、化验、测试、医疗、法律、会计、咨询、讲学、翻译、审稿、书画、雕刻、影视、录音、录像、演出、表演、广告、展览、技术服务、介绍服务、经纪服务、代办服务以及其他劳务取得的所得」, tức thu nhập từ lao động là thu nhập cá nhân có được khi làm lao động, gồm thu nhập từ thiết kế, trang trí, lắp đặt, vẽ kỹ thuật, thí nghiệm, kiểm nghiệm, y tế, pháp lý, kế toán, tư vấn, giảng dạy, phiên dịch, duyệt bản thảo, thư pháp, khắc, phim ảnh, thu âm, quay video, biểu diễn, trình diễn, quảng cáo, triển lãm, dịch vụ kỹ thuật, dịch vụ giới thiệu, dịch vụ môi giới, dịch vụ hội đồng và các loại lao động khác. Điều sáu của《Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc》ghi cách tính thu nhập chịu thuế:「居民个人的综合所得，以每一纳税年度的收入额减除费用六万元以及专项扣除、专项附加扣除和依法确定的其他扣除后的余额，为应纳税所得额」, tức thu nhập tổng hợp của cá nhân cư trú là số dư của thu nhập mỗi năm tính thuế trừ đi khoản chi phí sáu vạn tệ cùng các khoản khấu trừ riêng, khấu trừ riêng bổ sung và khoản khấu trừ khác do pháp luật quy định. Cùng điều đó còn ghi「劳务报酬所得、稿酬所得、特许权使用费所得以收入减除百分之二十的费用后的余额为收入额」, tức thu nhập từ thù lao lao động, thù lao tác phẩm và phí bản quyền thì lấy thu nhập trừ 20% chi phí, phần còn lại là thu nhập tính thuế. Câu trừ 20% chỉ liệt kê ba khoản này, lương công không nằm trong đó. Vì vậy lương công tính toàn bộ vào thu nhập tính thuế, hai khoản thuế tính ra khác nhau. Điều chín:「个人所得税以所得人为纳税人，以支付所得的单位或者个人为扣缴义务人」, tức thuế thu nhập cá nhân lấy người có thu nhập làm người nộp thuế, lấy đơn vị hoặc cá nhân trả tiền làm người khấu trừ. Điều mười:「有下列情形之一的，纳税人应当依法办理纳税申报：……（二）取得应税所得没有扣缴义务人；……（四）取得境外所得」, tức có một trong các trường hợp sau thì người nộp thuế phải kê khai thuế theo pháp luật:……hai là có thu nhập chịu thuế nhưng không có người khấu trừ……bốn là có thu nhập từ nước ngoài. Điều mười một:「居民个人取得综合所得，按年计算个人所得税……需要办理汇算清缴的，应当在取得所得的次年三月一日至六月三十日内办理汇算清缴」, tức cá nhân cư trú có thu nhập tổng hợp thì tính thuế thu nhập cá nhân theo năm……trường hợp phải tổng hợp quyết toán thì phải làm trong thời gian từ ngày 1 tháng ba đến ngày ba mươi tháng sáu của năm sau năm có thu nhập. Chuyển tiền về thì xem Quy định quản lý ngoại hối cá nhân（中国 个人外汇管理办法）（中国人民银行令〔2006〕第 3 号，áp dụng từ ngày 1 tháng 2 năm 2007）. Điều chín:「对个人结汇和境内个人购汇实行年度总额管理。年度总额内的，凭本人有效身份证件在银行办理；超过年度总额的，经常项目项下凭本人有效身份证件和有交易额的相关证明等材料在银行办理」, tức việc cá nhân chuyển tiền ra nước ngoài và cá nhân trong nước mua ngoại hối quản lý theo hạn mức cả năm. Trong hạn mức cả năm thì tự mang giấy tờ tùy thân hợp lệ ra ngân hàng làm; vượt hạn mức cả năm thì trong nhóm tài khoản thường, mang giấy tờ tùy thân hợp lệ và các chứng từ có giao dịch như chứng minh liên quan ra ngân hàng làm. Điều bảy:「银行和个人在办理个人外汇业务时，应当遵守本办法的相关规定，不得以分拆等方式逃避限额监管，也不得使用虚假商业单据或者凭证逃避真实性管理」, tức ngân hàng và cá nhân khi làm nghiệp vụ ngoại hối cá nhân phải tuân thủ các quy định liên quan của quy định này, không được chia nhỏ để né tránh việc kiểm soát hạn mức, cũng không được dùng chứng từ thương mại hay giấy tờ giả để né tránh kiểm soát tính xác thực. Quy định thi hành về quản lý ngoại hối cá nhân（汇发〔2007〕1 号）Điều hai:「对个人结汇和境内个人购汇实行年度总额管理。年度总额分别为每人每年等值 5 万美元」, tức việc cá nhân chuyển tiền ra nước ngoài và cá nhân trong nước mua ngoại hối quản lý theo hạn mức cả năm; hạn mức cả năm là mỗi người mỗi năm tương đương 5 vạn đô la Mỹ. Điều mười:「境内个人经常项目项下非经营性结汇超过年度总额的，凭本人有效身份证件及以下证明材料在银行办理：……（七）职工报酬：雇佣合同及收入证明」, tức cá nhân trong nước chuyển tiền ra nước ngoài phi kinh doanh trong nhóm tài khoản thường mà vượt hạn mức cả năm thì mang giấy tờ tùy thân hợp lệ cùng các chứng từ sau ra ngân hàng làm:……bảy là thù lao lao động: hợp đồng lao động và chứng minh thu nhập (toàn quốc) +- Chi phí: Không tốn tiền. Mỗi năm quyết toán thuế một lần, thời gian là từ ngày 1 tháng 3 đến ngày 30 tháng 6 năm sau. Khi vượt tổng hạn mức cả năm thì còn phải chuẩn bị hợp đồng lao động và giấy chứng minh thu nhập. Chỗ khó là khoản thuế này bạn phải tự nhớ đi kê khai. +- Nói thẳng: Khoản tiền đó tính là trong nước hay nước ngoài, xét chỗ bạn làm việc, không phải chỗ tiền được chuyển vào. Người ở trong nước làm việc từ xa cho công ty nước ngoài thì tính là thu nhập trong nước, phải nộp thuế ở trong nước. Công ty nước ngoài không có cơ quan ở Trung Quốc thì không ai khấu trừ thay bạn, khoản thuế này bạn phải tự kê khai. Tiền chuyển vào đi theo khoản「职工报酬」, tức thù lao lao động. +- Lợi ích: Việc khoản tiền đó có tính là thu nhập trong nước không thì xem Điều ba của《Quy chế thi hành Luật thuế thu nhập cá nhân》(Hội đồng Nhà nước Lệnh số 707, áp dụng từ ngày 1 tháng 1 năm 2019). Nó viết:「除国务院财政、税务主管部门另有规定外，下列所得，不论支付地点是否在中国境内，均为来源于中国境内的所得：（一）因任职、受雇、履约等在中国境内提供劳务取得的所得」, tức trừ trường hợp cơ quan tài chính, thuế của Hội đồng Nhà nước có quy định khác, thì các khoản thu nhập sau đây, không phân biệt nơi trả có ở trong nước Trung Quốc hay không, đều là thu nhập phát sinh trong nước Trung Quốc; một là thu nhập từ việc cung cấp lao động trong nước Trung Quốc do nhận chức vụ, được thuê mướn, thực hiện hợp đồng. Cùng quy chế đó ở Điều sáu định nghĩa hai khoản thu. Một là tiền lương, tiền công:「工资、薪金所得，是指个人因任职或者受雇取得的工资、薪金、奖金、年终加薪、劳动分红、津贴、补贴以及与任职或者受雇有关的其他所得」, tức thu nhập tiền lương, tiền công là khoản cá nhân nhận được do giữ chức vụ hoặc được thuê mướn, gồm tiền lương, tiền công, thưởng, lương thêm cuối năm, chia lợi nhuận lao động, trợ cấp, bồi thường và các khoản thu nhập khác liên quan đến việc giữ chức vụ hoặc được thuê mướn. Khoản còn lại là thù lao lao động:「劳务报酬所得，是指个人从事劳务取得的所得，包括从事设计、装潢、安装、制图、化验、测试、医疗、法律、会计、咨询、讲学、翻译、审稿、书画、雕刻、影视、录音、录像、演出、表演、广告、展览、技术服务、介绍服务、经纪服务、代办服务以及其他劳务取得的所得」, tức thu nhập từ lao động là thu nhập cá nhân nhận khi làm lao động, gồm thiết kế, trang trí, lắp đặt, vẽ bản đồ, thử nghiệm, kiểm nghiệm, y tế, pháp lý, kế toán, tư vấn, giảng dạy, dịch thuật, phản biên, thư pháp, khắc, điện ảnh, thu âm, quay video, biểu diễn, trình diễn, quảng cáo, triển lãm, dịch vụ kỹ thuật, dịch vụ giới thiệu, dịch vụ môi giới, dịch vụ hộ, và lao động khác. Điều sáu của《Luật thuế thu nhập cá nhân》viết cách tính thu nhập chịu thuế:「居民个人的综合所得，以每一纳税年度的收入额减除费用六万元以及专项扣除、专项附加扣除和依法确定的其他扣除后的余额，为应纳税所得额」, tức thu nhập tổng hợp của cá nhân cư trú là số thu nhập của năm thuế trừ đi khoản chi phí 六万元 và các khoản trừ riêng, trừ riêng bổ sung và khoản trừ khác được xác định theo luật. Cùng điều đó còn viết「劳务报酬所得、稿酬所得、特许权使用费所得以收入减除百分之二十的费用后的余额为收入额」, tức thu nhập từ thù lao lao động, thù lao viết bài và tiền bản quyền sử dụng thì lấy thu nhập trừ 20% chi phí rồi phần còn lại làm thu nhập. Câu trừ 20% chỉ liệt kê ba khoản này, tiền lương không nằm trong đó. Vì vậy tiền lương tính nguyên số vào thu nhập, hai khoản thu cho ra số thuế khác nhau. Điều chín:「个人所得税以所得人为纳税人，以支付所得的单位或者个人为扣缴义务人」, tức thuế thu nhập cá nhân lấy người có thu nhập làm người nộp thuế, lấy đơn vị hoặc cá nhân trả thu nhập làm người khấu trừ. Điều mười:「有下列情形之一的，纳税人应当依法办理纳税申报：……（二）取得应税所得没有扣缴义务人；……（四）取得境外所得」, tức có một trong các trường hợp sau thì người nộp thuế phải kê khai theo luật: hai là có thu nhập chịu thuế mà không có người khấu trừ; bốn là có thu nhập từ nước ngoài. Điều mười một:「居民个人取得综合所得，按年计算个人所得税……需要办理汇算清缴的，应当在取得所得的次年三月一日至六月三十日内办理汇算清缴」, tức thu nhập tổng hợp của cá nhân cư trú thì tính thuế thu nhập cá nhân theo năm; thu nhập nào phải quyết toán thì thực hiện trong khoảng từ ngày 1 tháng 3 đến hết tháng 6 của năm sau. Phần tiền chuyển vào thì xem《Quy chế quản lý ngoại hối cá nhân》(Lệnh của Ngân hàng Nhân dân Trung Quốc〔2006〕số 3, áp dụng từ ngày 1 tháng 2 năm 2007). Điều chín:「对个人结汇和境内个人购汇实行年度总额管理。年度总额内的，凭本人有效身份证件在银行办理；超过年度总额的，经常项目项下凭本人有效身份证件和有交易额的相关证明等材料在银行办理」, tức áp dụng hạn mức tổng cả năm cho việc cá nhân bán ngoại hối và người trong nước mua ngoại hối; trong hạn mức thì ở ngân hàng chỉ cần giấy tùy thân có hiệu lực của chính mình; vượt hạn mức thì ở ngân hàng phải có giấy tùy thân có hiệu lực của chính mình cùng các chứng từ liên quan có số tiền giao dịch và tương tự. Điều bảy:「银行和个人在办理个人外汇业务时，应当遵守本办法的相关规定，不得以分拆等方式逃避限额监管，也不得使用虚假商业单据或者凭证逃避真实性管理」, tức ngân hàng và cá nhân khi làm nghiệp vụ ngoại hối cá nhân phải tuân thủ các quy định của quy chế này, không được chia nhỏ để né kiểm soát hạn mức, cũng không được dùng chứng từ thương mại hoặc giấy tờ giả để né kiểm soát tính xác thực. Điều hai của《Quy chế thi hành chi tiết về quản lý ngoại hối cá nhân》(汇发〔2007〕1 号):「对个人结汇和境内个人购汇实行年度总额管理。年度总额分别为每人每年等值 5 万美元」, tức áp dụng hạn mức tổng cả năm cho việc cá nhân bán ngoại hối và người trong nước mua ngoại hối, hạn mức mỗi người mỗi năm tương đương 5 vạn đô la. Điều mười:「境内个人经常项目项下非经营性结汇超过年度总额的，凭本人有效身份证件及以下证明材料在银行办理：……（七）职工报酬：雇佣合同及收入证明」, tức người trong nước bán ngoại hối phi kinh doanh dưới hạn mức tổng cả năm thì ở ngân hàng cần giấy tùy thân có hiệu lực của chính mình và các chứng từ sau; bảy là thù lao lao động, gồm hợp đồng lao động và giấy chứng minh thu nhập (toàn quốc) - Mức bằng chứng: A -- Nguồn: Hội đồng Nhà nước (2018). Quy định thi hành Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc（国务院令第 707 号，điều ba, sáu）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (2018). Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc（sửa đổi lần thứ bảy năm 2018，điều sáu, chín, mười, mười một）. ；Kho văn bản quy phạm Tổng cục Thuế. Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc（điều sáu）. ；Ngân hàng Nhân dân Trung Quốc (2006). Quy định quản lý ngoại hối cá nhân（中国人民银行令〔2006〕第 3 号，điều bảy, chín）. ；Cục Quản lý Ngoại hối (2007). Quy định thi hành về quản lý ngoại hối cá nhân（汇发〔2007〕1 号，điều hai, mười）. -- Ghi chú: Tính theo khoản thuế nào thì tuỳ quan hệ giữa bạn và bên kia. Ký hợp đồng lao động, công việc nghe phân công của bên kia thì tính là lương công. Tự nhận việc theo từng đồ án, giao kết quả thì tính là thù lao lao động. Hai cách tính ra thuế khác nhau, trước khi kê hãy xác định mình thuộc loại nào. Khi chuyển tiền về, một năm trong hạn tương đương 5 vạn đô la Mỹ thì chỉ cần mang giấy tờ tùy thân ra ngân hàng chuyển; phần vượt hạn thì phải đưa hợp đồng lao động và chứng minh thu nhập cho ngân hàng. Đừng đi tìm người chia nhỏ khoản tiền để né hạn mức. Trên đây nói trường hợp người ở trong nước, không xuất cảnh. Thật sự xuất cảnh sang nước ngoài làm việc cho nhà tuyển dụng nước ngoài thì là bộ khác, xem mục 14 của phần này（muốn đi làm ở nước ngoài, trước hết tra công ty này có giấy phép hợp tác lao động xuất khẩu hay không）. Bốn việc sau đây cuốn sách này tạm chưa viết. Chúng đều có thể có chi phí thật, nhưng hiện chưa tìm được căn cứ nào đối chiếu từng chữ. Một là khoản thu nhập này có phải nộp thuế giá trị gia tăng không, có phải xuất hoá đơn thay không. Hai là làm lâu dài như vậy có phải đăng ký giấy phép kinh doanh hay đăng ký làm hộ kinh doanh cá thể không. Ba là có thể theo luật lao động yêu cầu nhà tuyển dụng nước ngoài thực hiện quyền lợi không, ví dụ ngày phép năm, tiền làm thêm giờ, bồi thường khi chấm dứt. Bốn là chuyển dữ liệu ra nước ngoài và bảo mật. -### 16. Không vào làm trong đơn vị mà tự làm ăn, tiền khởi nghiệp thì xem khoản vay có bảo đảm cho vốn khởi nghiệp: cá nhân tối đa 30 vạn, ngân sách trả một nửa tiền lãi +- Nguồn: Hội đồng Nhà nước (国务院) (2018). Quy chế thi hành Luật thuế thu nhập cá nhân của Trung Quốc（中华人民共和国个人所得税法实施条例）（Hội đồng Nhà nước Lệnh số 707，第三、六条）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人民代表大会常务委员会) (2018). Luật thuế thu nhập cá nhân của Trung Quốc（中华人民共和国个人所得税法）（sửa đổi lần thứ bảy năm 2018，第六、九、十、十一条）. ；Kho văn bản chính sách Cục Thuế quốc gia. Luật thuế thu nhập cá nhân của Trung Quốc（Điều 6）. ；Ngân hàng Nhân dân Trung Quốc (中国人民银行) (2006). Quy chế quản lý ngoại hối cá nhân（个人外汇管理办法）（Lệnh của Ngân hàng Nhân dân Trung Quốc〔2006〕số 3，第七、九条）. ；Cục Quản lý ngoại hối quốc gia (国家外汇管理局) (2007). Quy chế thi hành chi tiết về quản lý ngoại hối cá nhân（个人外汇管理办法实施细则）（汇发〔2007〕1 号，第二、十条）. +- Ghi chú: Tính theo khoản thu nào thì xét quan hệ giữa bạn và bên kia. Ký hợp đồng lao động, công việc nghe phân công của họ thì tính là tiền lương, tiền công. Tự nhận việc theo dự án, giao kết quả thì tính là thù lao lao động. Hai cách cho ra số thuế khác nhau, trước khi kê khai hãy xác định mình thuộc loại nào. Khi tiền chuyển vào, trong hạn mức tương đương 5 vạn đô la mỗi năm thì chỉ cần giấy tùy thân ở ngân hàng; phần vượt hạn mức thì phải đưa hợp đồng lao động và giấy chứng minh thu nhập cho ngân hàng. Đừng tìm người chia nhỏ để né hạn mức. Những điều viết trên đây nói về trường hợp ở trong nước, không xuất cảnh. Thật sự xuất cảnh đi làm công ăn cho nhà tuyển dụng nước ngoài là bộ khác, xem mục 14 của phần này (muốn đi làm công ăn nước ngoài, trước hết tra công ty này có giấy phép hợp tác lao động ngoài biên giới không). Bốn việc dưới đây sách này tạm chưa viết. Chúng đều có thể có chi phí thực tế, nhưng hiện chưa tra được căn cứ nào đối chiếu từng chữ. Một là khoản thu nhập này có phải nộp thuế giá trị gia tăng không, có phải xuất hoá đơn thay người nộp thuế không. Hai là làm lâu dài như vậy có phải đăng ký giấy phép kinh doanh hay đăng ký làm hộ kinh doanh cá thể không. Ba là có thể theo luật lao động yêu cầu nhà tuyển dụng nước ngoài thực hiện quyền lợi như nghỉ phép năm, tiền làm thêm giờ, bồi thường khi chấm dứt hay không. Bốn là dữ liệu xuất khẩu và bảo mật + +### 16. Không vào cơ quan mà tự làm ăn, tiền khởi đầu hãy xem khoản vay bảo đảm khởi nghiệp: cá nhân tối đa 30 vạn, Nhà nước trợ một nửa lãi suất -- Chi phí: Không tốn tiền. Đi hỏi thì không mất tiền. Trước hết chạy một lần cơ quan Nhân sự và Xã hội để thẩm định điều kiện, rồi tới từ chức bảo lãnh và ngân hàng thực hiện (ngân hàng cho vay) làm thủ tục mỗi nơi một lần. Chỗ khó là tiền vay mượn được rồi vẫn phải trả. -- Nói thẳng: Nếu cứ phải vay mà khởi nghiệp thì trước hết xem khoản này, có ngân sách trả phụ lãi. Cá nhân vay tối đa 30 vạn tệ, tối đa 3 năm, tiền lãi do ngân sách trả một nửa. Có một điều kiện cứng hay bị bỏ qua: lúc xin, bạn và vợ hoặc chồng không được có khoản vay nào khác, chỉ có các khoản không tính là khoản vay khác gồm khoản vay học tập, tín dụng vi mô cho người thoát nghèo, vay mua nhà, vay mua xe và tín dụng tiêu dùng nhỏ dưới 5 vạn tệ. -- Lợi ích: Điều kiện xin của cá nhân nằm ở Quy định quản lý quỹ đặc biệt phát triển tài chính toàn diện《普惠金融发展专项资金管理办法》（财金〔2023〕75 号）Điều sáu. Điều kiện thứ nhất là thuộc nhóm việc làm trọng điểm. Văn bản liệt kê người đã đăng ký thất nghiệp trong đô thị, người gặp khó khăn trong việc làm (gồm người khuyết tật), cựu chiến binh, người vừa chấm thúc án tù. Còn có người mới tốt nghiệp đại học (gồm sinh viên làm cán bộ ở nông thôn và sinh viên về nước học), công nhân và người thất nghiệp ở doanh nghiệp thuộc diện giảm thừa công suất. Còn có nông dân về quê khởi nghiệp, chủ cửa hàng trên mạng, người thoát nghèo, nông dân ở nông thôn tự khởi nghiệp. Điều kiện thứ hai là không có khoản vay nào khác:「除助学贷款、脱贫人口小额信贷、住房贷款、购车贷款、5 万元以下小额消费贷款（含信用卡消费）以外，申请人提交创业担保贷款申请时，本人及其配偶无其他贷款」, tức ngoài các khoản vay học tập, tín dụng vi mô cho người thoát nghèo, vay mua nhà, vay mua xe, tín dụng tiêu dùng nhỏ dưới 5 vạn tệ (gồm chi tiêu bằng thẻ tín dụng), thì người xin nộp đơn xin khoản vay có bảo đảm cho vốn khởi nghiệp không có khoản vay nào khác, bản thân và vợ hoặc chồng đều vậy. Điều chín:「对符合条件的创业担保贷款，财政部门给予贷款实际利率 50% 的财政贴息。对展期、逾期的创业担保贷款，财政部门原则上不予贴息」, tức với khoản vay có bảo đảm cho vốn khởi nghiệp đủ điều kiện, cơ quan tài chính trả phụ lãi bằng 50% lãi suất thực tế của khoản vay. Với khoản vay được gia hạn hoặc quá hạn, cơ quan tài chính nguyên tắc không trả phụ lãi. Phụ lãi nghĩa là ngân sách trả thay một nửa tiền lãi. Gia hạn là kéo dài thêm sau khi đến hạn, quá hạn là đã quá hạn mà chưa trả, hai trường hợp đó ngân sách không trả thay nữa. Điều mười: khoản vay có bảo đảm cho vốn khởi nghiệp do cá nhân xin dưới 10 vạn tệ thì nguyên tắc bỏ yêu cầu bảo đảm ngược, không phải tìm người khác hay dùng tài sản làm bảo đảm cho cơ quan bảo lãnh. Điều mười một: khoản vay có bảo đảm cho vốn khởi nghiệp cá nhân nằm trong phạm vi hỗ trợ trợ cấp thì「额度不超过 30 万元，贷款期限不超过 3 年」, tức hạn mức không quá 30 vạn tệ, thời hạn vay không quá 3 năm. Người cùng góp vốn khởi nghiệp thì có thể nâng lên tương ứng, tối đa không quá 110% tổng các mức trần của hạn mức cá nhân. Lãi suất tối đa của khoản vay được hỗ trợ trợ cấp chia ba bậc: vùng đông là LPR+50BPs, vùng trung tây LPR+150BPs, vùng thoát nghèo LPR+250BPs. LPR là lãi suất báo giá thị trường của ngân hàng, BPs là đơn vị đo lãi suất, cộng thêm càng nhiều thì lãi càng cao. Người「还款积极、带动就业能力强、创业项目好」, tức trả nợ tích cực, năng lực tạo việc làm mạnh, dự án khởi nghiệp tốt, thì sau khi đến hạn vẫn xin tiếp được, tổng số lần không quá 3 lần. Điều tám: người xin trước hết nộp đơn xin thẩm định điều kiện tới cơ quan Nhân sự và Xã hội địa phương, rồi nộp hồ sơ tới cơ quan bảo lãnh và ngân hàng thực hiện. Điều mười hai: được Ủy ban nhân dân cấp tỉnh đồng ý thì các nơi có thể mở rộng phạm vi hỗ trợ, nâng hạn mức và tỷ lệ phụ lãi (toàn quốc) +- Chi phí: Không tốn tiền. Hỏi miễn phí. Trước hết chạy một lần cơ quan bảo hiểm xã hội để thẩm định điều kiện, rồi làm thủ tục một lần ở tổ chức bảo đảm và một lần ở ngân hàng chi nhánh (ngân hàng cho vay). Chỗ khó là tiền vay cuối cùng vẫn phải trả. +- Nói thẳng: Nếu nhất định phải vay tiền để khởi đầu, hãy xem khoản có lãi suất được ngân sách trợ. Cá nhân vay tối đa 30 vạn tệ, dài nhất 3 năm, lãi suất do Nhà nước trợ một nửa. Có một điều kiện cứng hay bị bỏ qua: lúc nộp hồ sơ, bạn và người thân phối ngữu không được có khoản vay nào khác, trừ khoản vay học phí, tín dụng nhỏ cho hộ thoát nghèo, vay mua nhà, vay mua xe và các khoản vay tiêu dùng nhỏ dưới 5 vạn tệ. +- Lợi ích: Điều kiện cho cá nhân nằm ở Điều sáu của《Quy chế quản lý quỹ tài chính phát triển tài chính bao trùm》《普惠金融发展专项资金管理办法》（财金〔2023〕75 号）. Điều kiện thứ nhất là phải thuộc nhóm việc làm trọng điểm. Văn bản liệt kê người đã đăng ký thất nghiệp ở thành phố, người gặp khó khăn trong việc làm (kể cả người khuyết tật), cựu chiến binh, người vừa hết thời hạn hình sự. Còn có sinh viên, học viên tốt nghiệp (kể cả cán bộ làm việc ở nông thôn của trường đại học và sinh viên về nước du học), công nhân và người thất nghiệp ở doanh nghiệp cần giải phóng công suất thừa. Còn có nông dân về quê khởi nghiệp, chủ cửa hàng trực tuyến, hộ thoát nghèo, nông dân tự khởi nghiệp ở nông thôn và những người khác. Điều kiện thứ hai là không được có khoản vay nào khác:「除助学贷款、脱贫人口小额信贷、住房贷款、购车贷款、5 万元以下小额消费贷款（含信用卡消费）以外，申请人提交创业担保贷款申请时，本人及其配偶无其他贷款」, tức trừ khoản vay học phí, tín dụng nhỏ cho hộ thoát nghèo, vay mua nhà, vay mua xe và khoản vay tiêu dùng nhỏ dưới 5 vạn tệ (kể cả tiêu dùng bằng thẻ tín dụng), thì lúc người đề nghị nộp hồ sơ vay bảo đảm khởi nghiệp, bản thân người đó và người thân phối ngữu không có khoản vay nào khác. Điều chín:「对符合条件的创业担保贷款，财政部门给予贷款实际利率 50% 的财政贴息。对展期、逾期的创业担保贷款，财政部门原则上不予贴息」, tức với khoản vay bảo đảm khởi nghiệp đủ điều kiện, cơ quan tài chính trợ phần lãi bằng 50% lãi suất thực tế của khoản vay; với khoản vay được gia hạn hoặc quá hạn thì về nguyên tắc không trợ lãi. Trợ lãi nghĩa là Nhà nước trợ một nửa lãi suất. Gia hạn là đến hạn rồi kéo dài thêm, quá hạn là tới hạn không trả, hai trường hợp đó Nhà nước không trợ nữa. Điều mười: khoản vay bảo đảm khởi nghiệp dưới 10 vạn tệ mà cá nhân đề nghị thì về nguyên tắc bỏ yêu cầu bảo đảnh ngược, không cần tìm người khác hay dùng tài sản làm bảo đảm cho tổ chức bảo đảm. Điều mười một: khoản vay bảo đảm khởi nghiệp của cá nhân thuộc phạm vi hỗ trợ thưởng khuyến khích「额度不超过 30 万元，贷款期限不超过 3 年」, tức hạn mức không quá 30 vạn tệ, thời hạn vay không quá 3 năm. Người hợp tác kinh doanh thì có thể nâng mức, tối đa không quá 110% tổng hạn mức cá nhân. Trần lãi suất của khoản vay có hỗ trợ thưởng chia làm ba bậc: khu vực phía Đông là LPR+50BPs, khu vực trung và phía Tây là LPR+150BPs, khu vực thoát nghèo là LPR+250BPs. LPR là lãi suất định giá của thị trường tín dụng ngân hàng, BPs là đơn vị đo lãi suất, cộng thêm bao nhiêu thì lãi cao bấy nhiêu. Người「还款积极、带动就业能力强、创业项目好」thì sau khi đến hạn vẫn được nộp hồ sơ tiếp, tổng số lần không quá 3 lần. Điều tám: người đề nghị trước hết nộp hồ sơ thẩm định điều kiện với cơ quan bảo hiểm xã hội địa phương, rồi nộp hồ sơ với tổ chức bảo đảm và ngân hàng chi nhánh. Điều mười hai: được Hội đồng Nhà nước cấp tỉnh đồng ý thì các nơi có thể mở rộng phạm vi hỗ trợ, nâng hạn mức và tỷ lệ trợ lãi (toàn quốc) - Mức bằng chứng: A -- Nguồn: Bộ Tài chính (2023). Thông báo ban hành Quy định quản lý quỹ đặc biệt phát triển tài chính toàn diện《普惠金融发展专项资金管理办法》（财金〔2023〕75 号，第六、八、九、十、十一、十二条）. -- Ghi chú: Ngân sách trả một nửa tiền lãi không làm thay đổi việc「tiền vay thì phải trả」. Lập trường của sách ở mục 1 ở phần 12 không đổi: khởi nghiệp chỉ lấy số tiền mà lỗ còn chịu được, không đụng tới tài sản gia đình, không đặt bảo lãnh cá nhân cho khoản vay của công ty, vợ hoặc chồng càng không ký theo. Nếu bạn đã quyết định vay thì khoản vay chính sách này là mức lãi thấp nhất, đừng đụng vào tín dụng tiêu dùng và vay vốn ngoài ngân hàng. Hạn mức và phạm vi áp dụng các tỉnh có thể nâng lên, thực tế duyệt được bao nhiêu thì theo cách nói năm đó của cơ quan Nhân sự và Xã hội địa phương. Thông báo「cá nhân tối đa 20 vạn」thời dịch bệnh năm 2020 là văn bản tạm thời, bây giờ xem quy định này. Mở cửa hàng là đăng ký hộ kinh doanh cá thể hay lập công ty, giấy phép, kê khai thuế và rút lui xem phần 12. Trợ cấp đào tạo, trợ cấp bảo hiểm xã hội và thực tập nghề trong thời gian thất nghiệp xem mục 13 ở phần 7 (trợ cấp đào tạo nghề). \ No newline at end of file +- Nguồn: Bộ Tài chính (财政部) (2023). Thông báo ban hành《Quy chế quản lý quỹ tài chính phát triển tài chính bao trùm》（关于印发《普惠金融发展专项资金管理办法》的通知）（财金〔2023〕75 号，第六、八、九、十、十一、十二条）. +- Ghi chú: Nhà nước trợ một nửa lãi suất không làm thay đổi chuyện「tiền vay thì phải trả». Lập trường ở mục 1 ở phần 12 của sách này không đổi: khởi nghiệp thì chỉ lấy số tiền mà mất thì chịu được, không đụng vào vốn gia đình, không ký bảo đảm cá nhân cho khoản vay của công ty, người thân phối ngữu càng không ký theo. Nếu bạn đã quyết định sẽ vay, thì khoản vay chính sách loại này có lãi thấp nhất, đừng đụng vào tín dụng tiêu dùng và cho vay phi chính thức. Hạn mức và phạm vi áp dụng các tỉnh có thể nâng lên, thực tế được duyệt bao nhiêu thì theo cách nói năm đó của cơ quan bảo hiểm xã hội địa phương. Thông báo lúc dịch bệnh năm 2020 ghi「个人最高 20 万」là văn bản tạm thời, giờ hãy theo quy chế này. Mở cửa hàng thì đăng ký làm hộ kinh doanh cá thể hay lập công ty, giấy phép, kê khai thuế và rút lui, xem phần 12. Trợ cấp đào tạo, trợ cấp bảo hiểm xã hội và thực tập việc làm khi thất nghiệp, xem mục 13 ở phần 7 (trợ cấp đào tạo nghề). diff --git "a/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" index 9abb2265..200a57dc 100644 --- "a/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" +++ "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" @@ -1,6 +1,6 @@ [← Về mục lục](../README.md) -# 34. Đừng để thuốc trong nhà gây ra chuyện +# 34. Đừng dùng thuốc trong nhà sai cách Thuốc hạ sốt, thuốc giảm đau, thuốc cảm cúm, thuốc dạ dày, thuốc chống tiêu chảy, nhiều loại tự mua được ở hiệu thuốc. Mua được mà không cần đơn không có nghĩa uống thế nào cũng được. Phần này không nói bệnh gì thì uống thuốc nào, chỉ chọn ra mấy việc chỉ cần một động tác là tránh được hậu quả nặng. @@ -55,38 +55,38 @@ Mục trong phần này có người hưởng là bạn và gia đình bạn, th - Nguồn: U.S. Food and Drug Administration (2020-10-15). FDA recommends avoiding use of NSAIDs in pregnancy at 20 weeks or later because they can result in low amniotic fluid. Drug Safety Communication. - Ghi chú: FDA nhắc, nhiều thuốc không kê đơn điều trị cảm cúm, cúm, mất ngủ cũng chứa thành phần nhóm này, phải xem bảng thành phần. Nhãn thuốc không kê đơn ở Mỹ trước đây chỉ cảnh báo cho ba tháng cuối thai kỳ, lần này đưa mốc về sớm hơn khoảng tuần thứ 20. Aspirin liều thấp phòng tiền sản vô mạch, xem mục 5 ở phần 27 (aspirin liều thấp). Để ở mức B là vì số liệu đến từ báo cáo bất lợi, không biết có bao nhiêu người uống thuốc mà không sao. Cỡ lợi ích đặt là「vừa」là đánh giá: kết cục nặng, nhưng hiếm gặp. -### 6. Tự mua omeprazole khi bị trào ngược và nóng rát dạ dày, nhiều nhất 7 ngày; khó nuốt, nôn ra máu, phân đen thì đi thẳng bệnh viện +### 6. Trào ngược và ợ nóng thì tự mua omeprazole, uống tối đa 7 ngày; khó nuốt, nôn ra máu, phân đen thì đi thẳng bệnh viện -- Chi phí: Không tốn tiền. Uống đúng số ngày ghi trên tờ hướng dẫn. -- Nói thẳng: Từ năm 2022, omeprazol dạng viên nén bao phim chống mua không cần đơn thuốc nữa, nhưng tờ hướng dẫn ghi rõ tự uống không quá 7 ngày, trong vòng hai tháng đừng uống lại. Khi khó nuốt hoặc đau khi nuốt, nôn ra máu, đi ngoài ra máu hoặc phân đen thì đừng tự uống, đây có thể là dấu hiệu của tình trạng nghiêm trọng. Thuốc này còn có thể che mất triệu chứng ung thư dạ dày, làm chậm chẩn đoán. -- Lợi ích: Năm 2022, thông báo số 68 của Cục Quản lý Dược phẩm nhà nước chuyển omeprazol dạng viên nén bao phim chống từ thuốc có đơn sang thuốc không cần đơn, đồng thời ban hành tờ hướng dẫn mẫu. Chỉ định ghi trong tờ mẫu là「胃酸过多引起的烧心和反酸症状的短期缓解」— làm dịu tạm thời triệu chứng nóng rát và trào ngược do dạ dày tăng acid. Mục lưu ý ghi rõ: không dùng quá 7 ngày, triệu chứng không giảm thì phải hỏi bác sĩ hoặc dược sĩ. Trong vòng hai tháng không được uống lại, triệu chứng tái diễn thì phải đi khám ngay. Khó nuốt hoặc đau khi nuốt, nôn ra máu, đi ngoài ra máu hoặc phân đen thì không dùng, đây có thể là dấu hiệu của tình trạng nghiêm trọng. Khi xuất hiện các triệu chứng báo động như sụt cân rõ rệt, nôn lặp, khó nuốt, nôn ra máu hoặc phân đen thì phải loại trừ bệnh ác tính trước, vì việc điều trị có thể che mất triệu chứng và làm chậm chẩn đoán. Trên 55 tuổi mà có triệu chứng mới hoặc triệu chứng thay đổi thì phải hỏi bác sĩ. Người đang dùng clopidogrel thì tránh uống omeprazol cùng lúc. +- Chi phí: Không tốn tiền. Uống đúng số ngày ghi trong hướng dẫn. +- Nói thẳng: Viên nén bao phủ ruột omeprazole từ năm 2022 mua không cần đơn, nhưng hướng dẫn ghi rõ tự uống không quá 7 ngày, trong vòng hai tháng đừng uống lại. Khi khó nuốt hoặc đau khi nuốt, nôn ra máu, đi ngoài ra máu hoặc phân đen thì đừng tự uống, đây có thể là dấu hiệu của tình trạng nặng. Thuốc này còn có thể che mất triệu chứng ung thư dạ dày, làm chậm chẩn đoán. +- Lợi ích: Thông báo số 68 năm 2022 của Cục Quản lý Dược phẩm Quốc gia chuyển viên nén bao phủ ruột omeprazole từ thuốc kê đơn sang thuốc không kê đơn, đồng thời ban hành mẫu hướng dẫn. Mẫu hướng dẫn ghi chỉ định là「胃酸过多引起的烧心和反酸症状的短期缓解」— làm dịu ngắn hạn triệu chứng ợ nóng và trào ngược do dạ dày tăng nhiều acid. Mục lưu ý ghi rõ: không dùng quá 7 ngày, triệu chứng không giảm thì hỏi bác sĩ hoặc dược sĩ. Không được uống lại trong vòng hai tháng, triệu chứng tái diễn thì đi khám ngay. Không dùng khi khó nuốt hoặc đau khi nuốt, nôn ra máu, đi ngoài ra máu hoặc phân đen, vì đây có thể là dấu hiệu của tình trạng nặng. Khi có triệu chứng cảnh báo như sút cân rõ, nôn lặp, khó nuốt, nôn ra máu hoặc phân đen thì phải loại trừ ung thư trước, vì việc điều trị có thể che mất triệu chứng và làm chậm chẩn đoán. Người trên 55 tuổi mà có triệu chứng mới hoặc triệu chứng thay đổi thì hỏi bác sĩ. Người đang uống clopidogrel thì tránh uống omeprazole cùng lúc. - Mức bằng chứng: B -- Nguồn: Cục Quản lý Dược phẩm nhà nước (国家药监局) (2022). Thông báo về việc chuyển omeprazol dạng viên nén bao phim chống từ thuốc có đơn sang thuốc không cần đơn (2022年第68号) và phụ lục 2《非处方药说明书范本》. -- Ghi chú: Clopidogrel là thuốc chống đông máu mà người đã đặt stent, từng bị nhồi máu cơ tim thường phải dùng. Omeprazol làm nó yếu đi, người đang uống thì hỏi bác sĩ trước. Để ở mức B là vì đây là những gì tờ hướng dẫn quy định, không có nghiên cứu nào tính được làm theo sẽ bớt được bao nhiêu sự cố. Đặt lợi ích là「vừa」là theo phán đoán: nó ngăn việc kéo dài xuất huyết dạ dày, ung thư dạ dày, nhưng phần lớn chứng trào ngược, nóng rát dạ dày không nghiêm trọng. +- Nguồn: Thông báo của Cục Quản lý Dược phẩm Quốc gia về việc chuyển viên nén bao phủ ruột omeprazole từ thuốc kê đơn sang thuốc không kê đơn (2022年第68号) kèm phụ lục 2《非处方药说明书范本》. +- Ghi chú: Clopidogrel là thuốc chống huyết khối mà người đã đặt stent, đã từng bị nhồi máu cơ tim hay dùng. Omeprazole làm nó yếu đi, người đang uống thì hỏi bác sĩ trước. Để ở mức B là vì đây chỉ là quy định trong hướng dẫn, không có nghiên cứu nào tính được làm theo thì bớt được bao nhiêu sự cố. Cỡ lợi ích đặt là「vừa」là đánh giá: nó ngăn việc kéo dài xuất huyết dạ dày và ung thư dạ dày, nhưng phần lớn triệu chứng trào ngược và ợ nóng không nặng. -### 7. Cảm cúm thường thì đừng đi xin bác sĩ kê kháng sinh +### 7. Cảm cúm thường đừng xin bác sĩ kê kháng sinh -- Chi phí: Không tốn tiền, lại còn tiết kiệm được tiền thuốc. -- Nói thẳng: Cảm cúm mà uống kháng sinh thì khỏi không nhanh hơn là không uống. Người lớn uống kháng sinh có nguy cơ tác dụng phụ cao hơn hơn hai lần người không uống. Với viêm mũi mủ ra dịch mà kéo dài chưa đến 10 ngày, uống kháng sinh cũng không có lợi ích chắc chắn. -- Lợi ích: Một bài tổng hợp Cochrane gộp 6 thử nghiệm ngẫu nhiên, 1147 người. So với giả dược, tỷ lệ không khỏi hoặc triệu chứng kéo dài ở người cảm cúm uống kháng sinh không có khác biệt rõ rệt (RR 0.83, 95% CI 0.60–1.14, khác biệt không có ý nghĩa). Nguy cơ tác dụng phụ gấp 1.8 lần nhóm giả dược (95% CI 1.01–3.21). Trong đó người lớn gấp 2.62 lần (1.32–5.18), trẻ không tăng (0.91, 0.51–1.63). Với viêm mũi mủ cấp ra dịch mủ dưới 10 ngày, uống kháng sinh cũng không có lợi ích rõ ràng (RR 0.73, 0.47–1.13). Tác dụng phụ thì nhiều hơn (RR 1.46, 1.10–1.94, cao hơn khoảng 46%). Kết luận của tác giả là: cả hai trường hợp đều không nên dùng kháng sinh thường quy. +- Chi phí: Không tốn tiền, lại còn tiết kiệm tiền thuốc. +- Nói thẳng: Cảm cúm uống kháng sinh không khỏi nhanh hơn là không uống. Ở người lớn, nguy cơ tác dụng phụ khi uống cao hơn hai lần so với không uống. Ngạt mũi ra mủ dưới 10 ngày thì uống kháng sinh cũng không có lợi ích chắc chắn. +- Lợi ích: Một tổng quan hệ thống Cochrane gộp 6 thử nghiệm ngẫu nhiên, 1147 người. So sánh cảm cúm uống kháng sinh với uống giả dược, tỷ lệ không khỏi hoặc triệu chứng kéo dài không khác biệt rõ rệt (RR 0.83, 95% CI 0.60–1.14, chênh lệch không có ý nghĩa). Nguy cơ tác dụng phụ cao bằng 1.8 lần nhóm giả dược (95% CI 1.01–3.21). Trong đó ở người lớn là 2.62 lần (1.32–5.18), ở trẻ không tăng (0.91, 0.51–1.63). Với viêm mũi mủ cấp ở người ngạt mũi ra mủ dưới 10 ngày, uống kháng sinh cũng không có lợi ích rõ ràng (RR 0.73, 0.47–1.13). Tác dụng phụ lại nhiều hơn (RR 1.46, 1.10–1.94, cao hơn khoảng 46%). Kết luận của tác giả là: cả hai trường hợp này đều không nên dùng kháng sinh thường quy. - Mức bằng chứng: A - Nguồn: Kenealy T, Arroll B (2025). Antibiotics for the common cold and acute purulent rhinitis. Cochrane Database of Systematic Reviews, 11:CD000247. -- Ghi chú: Phiên bản này phát hành lại năm 2025, phần tóm tắt ghi rõ đã tìm tài liệu đến năm 2013. Kháng sinh ở nước ta là thuốc có đơn, hiệu thuốc phải có đơn mới bán, kháng sinh còn dư trong nhà cũng đừng tự lấy uống. Đặt lợi ích là「vừa」là theo phán đoán: nó ngăn chặn tác dụng phụ, không phải tử vong. +- Ghi chú: Bản này phát hành lại năm 2025, phần tóm tắt ghi rõ đã tìm tài liệu đến năm 2013. Kháng sinh ở trong nước là thuốc kê đơn, hiệu thuốc phải bán theo đơn, kháng sinh còn dư trong nhà cũng đừng tự lấy ra uống. Cỡ lợi ích đặt là「vừa」là đánh giá: nó ngăn tác dụng phụ, không ngăn tử vong. -### 8. Khi tiêu chảy thì uống muối bù điện giải uống để bù nước trước, trẻ dưới 5 tuổi không cho thuốc cầm tiêu chảy +### 8. Tiêu chảy thì uống dung dịch bù nước đường uống trước, trẻ dưới 5 tuổi không cho thuốc cầm tiêu chảy -- Chi phí: Một hộp muối bù điện giải uống vài nghìn đến mười mấy nghìn tệ, nên có sẵn một hộp trong nhà. -- Nói thẳng: Khi tiêu chảy, điều quan trọng nhất là bù nước, không phải cầm tiêu chảy. Pha đúng theo tờ hướng dẫn rồi uống muối bù điện giải uống, đừng thay bằng nước trái cây hay nước giải khát, quá ngọt lại càng khiến người ta mất nước hơn. Trẻ dưới 5 tuổi không cho thuốc cầm tiêu chảy như loperamid, nặng thì gây tắc ruột, có thể tử vong. -- Lợi ích: Sổ tay hướng dẫn điều trị tiêu chảy của Tổ chức Y tế Thế giới viết rằng thuốc「chống tiêu chảy」và thuốc chống nôn không có lợi ích thực tế với tiêu chảy cấp hoặc kéo dài ở trẻ em. Chúng không ngăn được mất nước, cũng không cải thiện dinh dưỡng. Một số có tác dụng phụ nguy hiểm, đôi khi tử vong, tuyệt đối không cho trẻ dưới 5 tuổi dùng. Nhóm thuốc làm chậm nhu động ruột như loperamid có thể gây tắc ruột do liệt, rất nặng và có thể tử vong, lại còn kéo dài thời gian nhiễm trùng. Sổ tay còn viết, đồ uống nhiều đường sẽ hút nước trong cơ thể vào ruột, gây mất nước tăng natri, ví dụ như nước giải khát và nước trái cây đóng bình bán sẵn. Sổ tay của Tổ chức Y tế Thế giới khuyến nghị công thức nhẹ thẩm thấu, tổng độ thẩm thấu 245 mOsm/L. So với công thức cũ, nó làm giảm 33% số lần truyền tĩnh mạch ngoài dự kiến. Một bài tổng hợp Cochrane gộp 8 thử nghiệm, kết quả tương tự (OR 0.59, 95% CI 0.45–0.79, ít hơn khoảng bốn phần mười). +- Chi phí: Một hộp dung dịch bù nước đường uống vài tệ đến hơn mười tệ, nên để sẵn một hộp trong nhà. +- Nói thẳng: Khi tiêu chảy, điều quan trọng nhất là bù nước, không phải cầm tiêu chảy. Pha dung dịch bù nước đường uống theo hướng dẫn, đừng thay bằng nước trái cây hay nước giải khát, quá ngọt lại càng khiến người bệnh thiếu nước hơn. Trẻ dưới 5 tuổi không cho loperamide và nhóm thuốc cầm tiêu chảy này, nặng thì gây tắc ruột, có thể tử vong. +- Lợi ích: Sổ tay điều trị tiêu chảy của Tổ chức Y tế Thế giới ghi rõ, thuốc「cầm tiêu chảy」và thuốc chống nôn không có lợi ích thực tế với tiêu chảy cấp hoặc tiêu chảy kéo dài ở trẻ. Chúng không ngăn được mất nước, cũng không cải thiện dinh dưỡng. Có những tác dụng phụ nguy hiểm, đôi khi gây chết, tuyệt đối không cho trẻ dưới 5 tuổi dùng. Nhóm thuốc làm chậm nhu động ruột như loperamide có thể gây tắc ruột liệt nặng, có thể tử vong, còn kéo dài nhiễm trùng. Sổ tay còn ghi, đồ uống nhiều đường sẽ hút nước trong cơ thể vào ruột, gây mất nước do natri cao, ví dụ nước giải khát và nước trái cây đóng bình bán sẵn. Công thức mà sổ tay của WHO khuyến nghị là công thức nhẩm, tổng áp suất thẩm thấm 245 mOsm/L. So với công thức cũ, nó làm số lần truyền dịch tĩnh mạch ngoài dự kiến giảm 33%. Một tổng quan Cochrane gộp 8 thử nghiệm, kết quả gần tương tự (OR 0.59, 95% CI 0.45–0.79, ít hơn khoảng bốn phần mười). - Mức bằng chứng: A - Nguồn: World Health Organization (2005). The treatment of diarrhoea: a manual for physicians and other senior health workers, 4th revision. ；Hahn S, Kim S, Garner P (2002). Reduced osmolarity oral rehydration solution for treating dehydration caused by acute diarrhoea in children. Cochrane Database of Systematic Reviews, (1):CD002847. -- Ghi chú: Khi mua thì xem độ thẩm thấu ghi trên tờ hướng dẫn, chọn công thức nhẹ thẩm thấu. Sổ tay viết rằng, trẻ đi ngoài kèm máu phần lớn là kiết lỵ, phải đi khám bác sĩ và dùng kháng sinh, không thể chỉ bù nước ở nhà. Đặt lợi ích là「vừa」là theo phán đoán: con số trong tay so hai loại muối bù nước với nhau, không phải có uống với không uống. +- Ghi chú: Khi mua, xem áp suất thẩm thấm ghi trên hướng dẫn, chọn công thức nhẩm. Sổ tay ghi rõ, trẻ đi ngoài có máu phần lớn là kiết lị, phải đi khám bác sĩ dùng kháng sinh, không thể chỉ bù nước ở nhà. Cỡ lợi ích đặt là「vừa」là đánh giá: con số trong tay so hai loại dung dịch bù nước với nhau, không phải so có uống với không uống. -### 9. Thuốc giảm đau trong một tháng đừng uống quá 15 ngày, thuốc giảm đau phối hợp đừng quá 10 ngày, không thì đau đầu có thể do thuốc gây ra +### 9. Thuốc giảm đau trong một tháng đừng uống quá 15 ngày, thuốc giảm đau phối hợp đừng quá 10 ngày, nếu không đau đầu có thể do thuốc -- Chi phí: Không tốn tiền. Phải ghi nhớ số ngày đã uống thuốc, rồi nhịn không uống bừa. -- Nói thẳng: Đau đầu tái diễn mà cứ ngày nào cũng uống thuốc giảm đau thì có thể càng uống càng đau. Phân loại quốc tế gọi đó là đau đầu do lạm dụng thuốc. Thuốc giảm đau thường uống từ 15 ngày trở lên trong một tháng, thuốc giảm đau phối hợp từ 10 ngày trở lên, kéo dài quá 3 tháng thì đã đủ để chẩn đoán. Phần lớn người bệnh cải thiện sau khi ngừng thuốc dùng quá. -- Lợi ích: Phân loại quốc tế của Hiệp hội Đau nửa đầu Quốc tế đưa ra tiêu chuẩn chẩn đoán trong Phân loại quốc tế các bệnh đau nửa đầu, xuất bản lần thứ 3. Người vốn đã bị đau nửa đầu, mỗi tháng đau từ 15 ngày trở lên, đồng thời dùng quá liều thuốc trị đau nửa đầu một cách đều đặn quá 3 tháng. Với nhóm thuốc giảm đau không phải opioid như paracetamol, aspirin, ibuprofen, ngưỡng quá liều là từ 15 ngày mỗi tháng trở lên. Với thuốc giảm đau phối hợp, ngưỡng là từ 10 ngày mỗi tháng trở lên. Thuốc giảm đau phối hợp là loại gộp từ hai thành phần có tác dụng giảm đau trở lên, hoặc có thêm thành phần hỗ trợ như caffein, ví dụ paracetamol phối hợp codein. Nhiều loại thuốc giảm đau không phải opioid dùng luân phiên thì số ngày cộng lại. Tài liệu chú giải phân loại viết rằng, bằng chứng dịch tễ học ở nhiều nước cho thấy trong nhóm người đau đầu từ 15 ngày mỗi tháng trở lên, hơn một nửa thuộc loại đau đầu này. Phần lớn người bệnh cải thiện sau khi ngừng thuốc quá liều, và việc điều trị phòng đau nửa đầu cũng hiệu quả hơn. +- Chi phí: Không tốn tiền. Phải ghi nhớ số ngày uống thuốc, và phải kiềm chế không tự tiện uống. +- Nói thẳng: Đau đầu tái diễn, ngày nào cũng uống thuốc giảm đau thì có thể đau hơn. Phân loại đau đầu quốc tế gọi đó là đau đầu do lạm dụng thuốc. Thuốc giảm đau thường uống trên 15 ngày trong một tháng, thuốc giảm đau phối hợp trên 10 ngày, liên tục quá 3 tháng là đủ để tính. Đa số người khỏi hơn sau khi ngừng thuốc dùng quá. +- Lợi ích: Bản xếp loại các bệnh đau đầu của Hiệp hội Đau đầu Quốc tế,《国际头痛疾病分类》phiên bản thứ 3, đưa ra tiêu chuẩn chẩn đoán. Người vốn đã bị đau đầu, mỗi tháng đau đầu trên 15 ngày, đồng thời lạm dụng có quy tắc thuốc trị đau đầu quá 3 tháng. Với nhóm thuốc giảm đau không phải chất opioid như paracetamol, aspirin, ibuprofen thì ngưỡng quá liều là 15 ngày trở lên mỗi tháng. Ngưỡng của thuốc giảm đau phối hợp là 10 ngày trở lên mỗi tháng. Thuốc giảm đau phối hợp là khi hai loại trở lên có tác dụng giảm đau gộp lại, hoặc kèm thành phần hỗ trợ như caffeine, ví dụ paracetamol phối hợp codein. Nhiều loại thuốc giảm đau không phải opioid dùng luân phiên thì số ngày cộng lại. Phần giải thích phân loại ghi rõ, bằng chứng dịch tễ từ nhiều nước cho thấy, trong số người đau đầu trên 15 ngày mỗi tháng thì hơn một nửa thuộc loại đau đầu này. Đa số người bệnh khỏi hơn sau khi ngừng thuốc dùng quá, và việc điều trị phòng đau đầu cũng hiệu quả hơn. - Mức bằng chứng: B -- Nguồn: Headache Classification Committee of the International Headache Society (2018). The International Classification of Headache Disorders, 3rd edition. Cephalalgia, 38(1):1-211. ；Bản trực tuyến ICHD-3, mục 8.2 Medication-overuse headache (gồm 8.2.3、8.2.5). -- Ghi chú: Viên giảm đau mà tên có chữ「咖」thường đã thêm caffein, tính theo thuốc phối hợp thì 10 ngày. Trong một tháng số ngày đau đầu càng tăng dần thì đi khám khoa thần kinh, đừng chỉ tăng liều thuốc. Để ở mức B là vì tiêu chuẩn phân loại đưa ra là ngưỡng chẩn đoán, không phải thử nghiệm đối chứng. Đặt lợi ích là「vừa」là theo phán đoán: nó đổi lại là những ngày không đau đầu, không dính gì đến tử vong. +- Nguồn: Headache Classification Committee of the International Headache Society (2018). The International Classification of Headache Disorders, 3rd edition. Cephalalgia, 38(1):1-211. ；ICHD-3 bản trực tuyến 8.2 Medication-overuse headache (gồm 8.2.3, 8.2.5). +- Ghi chú: Viên giảm đau mà tên có chữ「咖」thì hầu như đã thêm caffeine, tính như thuốc phối hợp, tức 10 ngày. Số ngày đau đầu trong tháng càng tăng dần thì đi khám khoa thần kinh, đừng chỉ dựa vào tăng liều. Để ở mức B là vì tiêu chuẩn phân loại đưa ra là ngưỡng chẩn đoán, không phải thử nghiệm đối chứng. Cỡ lợi ích đặt là「vừa」là đánh giá: nó đổi lại là những ngày ít đau đầu hơn, không liên quan tới tử vong. From 0208ae13b9f278378fcf479c549642780b3d3487 Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 12:54:43 +0700 Subject: [PATCH 18/25] =?UTF-8?q?=E8=B6=8A=E5=8D=97=E6=96=87=E5=85=A8?= =?UTF-8?q?=E4=B9=A6=E7=BF=BB=E5=AE=8C=EF=BC=9A34/34=20=E8=8A=82=EF=BC=8C6?= =?UTF-8?q?41=20=E6=9D=A1=EF=BC=8C=E4=B8=89=E7=A7=8D=E8=AF=AD=E8=A8=80?= =?UTF-8?q?=E6=9D=A1=E7=9B=AE=E6=95=B0=E4=B8=80=E8=87=B4?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit vi 29 三处数字：条 1「每 1394 个」写成了「1.394」（原文没那个千分位，多一个数）； 条 2 原文「607 万多人（6,073,240 人）」，量级词在，所以两个数都该在，译文把「607 万」 写成「6.070.000」多出一个数、把括号里那个漏了；条 3「逾 2000 万人」的 20.000.000 是允许的量级换算，没动。 顺带修 vi 03 句子里夹的三个中文碎片（普遍/恰/不大），book/ 和 .i18n 两处都改了。 现在三种语言的页面各是 641 条、34 节，check.mjs 和 render-check 都逐条核过： 收益/成本/来源三栏的数字一个不多一个不少、URL 逐字节相同、成本标签注释照抄、 条目标题没有一个还是中文。 --- ...52\350\264\271\347\262\276\345\212\233.md" | 8 +- ...15\351\235\242\346\270\205\345\215\225.md" | 6 +- ...00\350\203\275\345\210\222\347\256\227.md" | 60 +++---- ...31\346\210\226\345\271\263\345\217\260.md" | 106 ++++++++++++ ...25\345\222\214\347\224\237\344\272\247.md" | 151 ++++++++++++++++++ ...23\345\207\273\344\271\213\345\220\216.md" | 122 ++++++++++++++ ...72\345\233\275\347\225\231\345\255\246.md" | 98 ++++++++++++ 7 files changed, 514 insertions(+), 37 deletions(-) create mode 100644 "vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" create mode 100644 "vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" create mode 100644 "vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" create mode 100644 "vi/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" diff --git "a/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" "b/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" index c8971de1..6d41ad06 100644 --- "a/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" +++ "b/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" @@ -185,16 +185,16 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c ### 22. Khi thấy「ai cũng thấy mình mất mặt」thì chia đôi con số ước lượng đó - Chi phí: Không tốn tiền. Chỉ là tự hỏi thêm một câu lúc căng thẳng: họ thật sự nhìn thấy sao. -- Nói thẳng: Bước vào phòng có người mặc chiếc T-shirt đáng xấu hổ, người mặc nghĩ gần một nửa số người nhận ra ai được in trên áo, thực tế chưa đến một phần tư. Mặc chiếc T-shirt thấy tử tế thì con số ước lượng gấp sáu lần thực tế. Sau khi trò chuyện, người ta cũng普遍 đánh giá thấp mức đối phương thích mình. Vậy khi thấy「ai cũng nhớ mình mất mặt」, chia đôi vẫn còn cao. -- Lợi ích: Thí nghiệm một: cho người tham dự mặc T-shirt in hình Barry Manilow, bước vào phòng có người tham dự khác. Người mặc ước trung bình 46% người chứng kiến chỉ đúng người in trên áo. Người chứng kiến đúng thực tế là 23%, ước lượng恰 gấp đôi thực tế. Tính từng lượt, ước lượng trừ thực tế, trung bình cao hơn 23 điểm phần trăm (95% CI 9–38, khoảng tin cậy). Loại hai lượt chỉ có hai và ba người chứng kiến ra thì là 27 điểm phần trăm (11–43). Thí nghiệm hai: đổi sang mặc T-shirt do người tham dự tự chọn, tự thấy tích cực. Ước trung bình 48%, người chứng kiến đúng thực tế 8%, gấp sáu lần. Một loạt nghiên cứu khác có ba bối cảnh: người lạ ghép cặp trò chuyện, bạn cùng phòng năm nhất, hội trưởng thành cho người lớn. Kết quả là người ta đánh giá thấp mức đối phương thích mình, và cũng đánh giá thấp mức đối phương enjoyed cuộc trò chuyện. Tác giả gọi đó là「khoảng cách thích」. Nó có trong các cuộc trò chuyện dài ngắn khác nhau, và tồn tại hàng tháng giữa bạn cùng phòng năm nhất. +- Nói thẳng: Bước vào phòng có người mặc chiếc T-shirt đáng xấu hổ, người mặc nghĩ gần một nửa số người nhận ra ai được in trên áo, thực tế chưa đến một phần tư. Mặc chiếc T-shirt thấy tử tế thì con số ước lượng gấp sáu lần thực tế. Sau khi trò chuyện, người ta cũng đánh giá thấp mức đối phương thích mình. Vậy khi thấy「ai cũng nhớ mình mất mặt」, chia đôi vẫn còn cao. +- Lợi ích: Thí nghiệm một: cho người tham dự mặc T-shirt in hình Barry Manilow, bước vào phòng có người tham dự khác. Người mặc ước trung bình 46% người chứng kiến chỉ đúng người in trên áo. Người chứng kiến đúng thực tế là 23%, ước lượng gấp đôi thực tế. Tính từng lượt, ước lượng trừ thực tế, trung bình cao hơn 23 điểm phần trăm (95% CI 9–38, khoảng tin cậy). Loại hai lượt chỉ có hai và ba người chứng kiến ra thì là 27 điểm phần trăm (11–43). Thí nghiệm hai: đổi sang mặc T-shirt do người tham dự tự chọn, tự thấy tích cực. Ước trung bình 48%, người chứng kiến đúng thực tế 8%, gấp sáu lần. Một loạt nghiên cứu khác có ba bối cảnh: người lạ ghép cặp trò chuyện, bạn cùng phòng năm nhất, hội trưởng thành cho người lớn. Kết quả là người ta đánh giá thấp mức đối phương thích mình, và cũng đánh giá thấp mức đối phương enjoyed cuộc trò chuyện. Tác giả gọi đó là「khoảng cách thích」. Nó có trong các cuộc trò chuyện dài ngắn khác nhau, và tồn tại hàng tháng giữa bạn cùng phòng năm nhất. - Mức bằng chứng: B - Nguồn: Gilovich T, Medvec VH, Savitsky K (2000). The spotlight effect in social judgment: An egocentric bias in estimates of the salience of one's own actions and appearance. Journal of Personality and Social Psychology, 78(2), 211–222. ; Boothby EJ, Cooney G, Sandstrom GM, Clark MS (2018). The Liking Gap in Conversations: Do People Like Us More Than We Think? Psychological Science, 29(11), 1742–1756. - Ghi chú: Lý do chấm B: đều là nghiên cứu thí nghiệm trong phòng lab và nghiên cứu ở trường, bối cảnh là mặc quần áo và nói chuyện với người lạ. Chuyển sang「bị đồng nghiệp bàn tán」「nói sai trên họp」thì không có bằng chứng trực tiếp. Những con số này chỉ cho hướng ước cao và độ lớn đại khái, không dùng làm hệ số được. Ý kiến người khác cũng không phải thứ gì cũng quan trọng. Chỗ thật sự đang đánh giá bạn, như phỏng vấn, bảo vệ luận văn, đánh giá, báo cáo khám sức khỏe, thì đánh giá là thật, đã thế thì chuẩn bị. Cái cần giảm chỉ là cảm giác「mọi người đều đang nhìn mình」trong đời thường, chỗ đó thứ phải đối phó là chính thiên lệch ước lượng của bạn. ### 23. Xem câu「người khác đòi tôi hoàn hảo」là triệu chứng, không phải sự thật - Chi phí: Không tốn tiền. Khó ở chỗ xem câu nghĩ đó là thứ cần xử lý, chứ không phải là phép đo chính xác yêu cầu thật của người khác. -- Nói thẳng: Người luôn thấy「người quanh tôi đòi tôi hoàn hảo, không đạt thì sẽ bị phủ nhận」thì dễ có ý nghĩ tự tắt hơn. Mối liên hệ giữa「tôi tự đòi mình hoàn hảo」và ý nghĩ tự tắt yếu hơn nhiều, chỉ bằng khoảng một phần ba cách thứ nhất. Trừ những ý nghĩ đã có sẵn, cảm giác thứ nhất vẫn liên quan đến việc ý nghĩ tăng lên sau đó, chỉ là độ lớn不大. -- Lợi ích: Gộp nhiều nghiên cứu (phân tích tổng hợp): 45 nghiên cứu, N = 11.747, đối tượng là sinh viên đại học, sinh viên y, người lớn trong cộng đồng và bệnh nhân tâm thần, dùng mô hình biến thiên ngẫu nhiên. Sau đây là tương quan trung bình có trọng số với ý nghĩ tự tắt. Lo lắng hoàn hảo r = .28 (95% CI .24–.32, khoảng tin cậy; 42 nghiên cứu, 7936 người). Lo lắng hoàn hảo là một tập hợp, gồm完美主义 do xã hội quy định, lo mắc lỗi, nghi ngờ hành động của bản thân, khoảng cách giữa thực tế và tiêu chuẩn. Trong đó xã hội quy định tính riêng cũng là r = .28 (.25–.32, 30 nghiên cứu, 3640 người). Vươn tới hoàn hảo r = .10 (.07–.13, 31 nghiên cứu, 4588 người), nó là tập hợp của hoàn hảo tự hướng và tiêu chuẩn cá nhân. Cha mẹ phê bình r = .20 (.11–.29). Sang mối liên hệ với hành động tự sát. Lo lắng hoàn hảo r = .12 (.07–.17, 15 nghiên cứu, 5275 người). Hoàn hảo do xã hội quy định r = .19 (.08–.29, 8 nghiên cứu, 689 người). Phần theo dõi theo thời gian: sau khi trừ ý nghĩ tự sát đã có lúc đầu, hoàn hảo do xã hội quy định vẫn biến thiên cùng chiều với ý nghĩ tự sát ở lần theo dõi sau, độ lớn不大. Hoàn hảo tự hướng và hướng người khác đều không thấy liên hệ. +- Nói thẳng: Người luôn thấy「người quanh tôi đòi tôi hoàn hảo, không đạt thì sẽ bị phủ nhận」thì dễ có ý nghĩ tự tắt hơn. Mối liên hệ giữa「tôi tự đòi mình hoàn hảo」và ý nghĩ tự tắt yếu hơn nhiều, chỉ bằng khoảng một phần ba cách thứ nhất. Trừ những ý nghĩ đã có sẵn, cảm giác thứ nhất vẫn liên quan đến việc ý nghĩ tăng lên sau đó, chỉ là độ lớn không lớn. +- Lợi ích: Gộp nhiều nghiên cứu (phân tích tổng hợp): 45 nghiên cứu, N = 11.747, đối tượng là sinh viên đại học, sinh viên y, người lớn trong cộng đồng và bệnh nhân tâm thần, dùng mô hình biến thiên ngẫu nhiên. Sau đây là tương quan trung bình có trọng số với ý nghĩ tự tắt. Lo lắng hoàn hảo r = .28 (95% CI .24–.32, khoảng tin cậy; 42 nghiên cứu, 7936 người). Lo lắng hoàn hảo là một tập hợp, gồm完美主义 do xã hội quy định, lo mắc lỗi, nghi ngờ hành động của bản thân, khoảng cách giữa thực tế và tiêu chuẩn. Trong đó xã hội quy định tính riêng cũng là r = .28 (.25–.32, 30 nghiên cứu, 3640 người). Vươn tới hoàn hảo r = .10 (.07–.13, 31 nghiên cứu, 4588 người), nó là tập hợp của hoàn hảo tự hướng và tiêu chuẩn cá nhân. Cha mẹ phê bình r = .20 (.11–.29). Sang mối liên hệ với hành động tự sát. Lo lắng hoàn hảo r = .12 (.07–.17, 15 nghiên cứu, 5275 người). Hoàn hảo do xã hội quy định r = .19 (.08–.29, 8 nghiên cứu, 689 người). Phần theo dõi theo thời gian: sau khi trừ ý nghĩ tự sát đã có lúc đầu, hoàn hảo do xã hội quy định vẫn biến thiên cùng chiều với ý nghĩ tự sát ở lần theo dõi sau, độ lớn không lớn. Hoàn hảo tự hướng và hướng người khác đều không thấy liên hệ. - Mức bằng chứng: B - Nguồn: Smith MM, Sherry SB, Chen S, Saklofske DH, Mushquash C, Flett GL, Hewitt PL (2018). The perniciousness of perfectionism: A meta-analytic review of the perfectionism-suicide relationship. Journal of Personality, 86(3), 522–542. - Ghi chú: Lý do chấm B: phần lớn nghiên cứu đưa vào là bảng hỏi một lần, không thấy trước sau. Độ mạnh tương quan ở mức trung bình (r khoảng .28), đại khái chỉ giải thích được vài phần trăm đến một phần mười chênh lệch. Thứ đo là ý nghĩ tự sát và hành động tự sát, không phải cái chết本身, nên quy mức lợi ích nhỏ với lý do「chỉ có điểm cuối thay thế」. Chính tác giả cũng nói nhóm lo lắng hoàn hảo và hành động tự sát này hình như có thiên lệch xuất bản. Thiên lệch xuất bản nghĩa là nghiên cứu kết quả không đẹp thì không được in. Bù phần nghiên cứu mất ước tính vào thì hướng kết luận không đổi. Tiêu chuẩn cao本身 không phải vấn đề, cũng không vì thế mà hạ yêu cầu. Khác nhau ở chỗ tiêu chuẩn đến từ đâu, và khi không đạt có thấy「mình thế là xong」hay không. Nó cũng như dự đoán bi quan ở mục 15 (xem câu「chắc chắn mọi chuyện sẽ tệ hơn」là triệu chứng), là thứ xử lý được như triệu chứng. Có ý nghĩ tự tắt thì gọi 12356 trước và chuyển phương tiện gây tử vong ra xa, xem mục 25 ở phần 1. Người quanh bạn bộc lộ ý nghĩ này thì bạn làm được gì, xem mục 15 ở phần 8. Sàng lọc trầm cảm cho trẻ xem mục 8 ở phần 30. diff --git "a/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" "b/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" index b85565ad..a09e3cc1 100644 --- "a/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" +++ "b/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" @@ -236,7 +236,7 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Lợi ích: Xem trước lời nói 「bỏ bữa sáng sẽ khiến người ta ăn nhiều hơn về sau」. Một bài phân tích tổng hợp gộp 13 thử nghiệm ngẫu nhiên. Người được xếp vào nhóm ăn sáng mỗi ngày ăn vào nhiều hơn nhóm không ăn sáng 259.79 kcal (95% CI 78.87–440.71, đây là khoảng tin cậy). Cân nặng cao hơn 0.44 kg (0.07–0.82). Người không ăn sáng không bù lại ở bữa trưa và bữa tối. Các thử nghiệm này theo dõi rất ngắn, theo cân nặng trung bình 7 tuần, theo lượng ăn trung bình 2 tuần. Xem tiếp chế độ 16:8, tức chỉ ăn trong một khoảng giờ cố định. Một thử nghiệm cho 116 người lớn thừa cân làm trong 12 tuần. Nhóm chỉ ăn từ 12 giờ trưa tới 8 giờ tối và nhóm ăn đều ba bữa mỗi ngày chênh nhau −0.26 kg (−1.30 đến 0.78), không có ý nghĩa. Một thử nghiệm khác làm ở Quảng Châu, 139 người bệnh béo phì, theo dõi 12 tháng. Ăn trong khung giờ kèm kiểm soát calo giảm nhiều hơn chỉ kiểm soát calo 1.8 kg (−4.0 đến 0.4), cũng không có ý nghĩa. Cuối cùng xem bỏ bữa sáng lâu dài có gây hại không. Trong tổng hợp 2.380.000 người, người không ăn sáng có nguy cơ bệnh tim mạch cao hơn khoảng 17% (OR 1.17, 1.09–1.26), tử vong do tim mạch cao hơn khoảng 49% (OR 1.49, 1.20–1.84). Nhưng đây là những nghiên cứu chỉ ghi nhận, không chia nhóm. Trong nghiên cứu gốc theo dõi từ 17 đến 23 năm, tỷ số nguy cơ tử vong toàn nguyên là 1.19, khoảng tin cậy 0.99 đến 1.42, vượt qua 1, không tính là khác biệt. - Mức bằng chứng: A - Nguồn: Sievert K, Hussain SM, Page MJ, 等 (2019). Effect of breakfast on weight and energy intake: systematic review and meta-analysis of randomised controlled trials. BMJ. ；Lowe DA, Wu N, Rohdin-Bibby L, 等 (2020). Effects of Time-Restricted Eating on Weight Loss and Other Metabolic Parameters in Women and Men With Overweight and Obesity. JAMA Internal Medicine. ；Liu D, Huang Y, Huang C, 等 (2022). Calorie Restriction with or without Time-Restricted Eating in Weight Loss. New England Journal of Medicine. ；Zhang H, Zhang S, Liu Y, 等 (2025). The association between skipping breakfast and cardiovascular disease: a meta analysis. Frontiers in Cardiovascular Medicine. ；Rong S, Snetselaar LG, Xu G, 等 (2019). Association of Skipping Breakfast With Cardiovascular and All-Cause Mortality. Journal of the American College of Cardiology. ；Fernandes-Alves D, Teixeira GP, Guimarães KC, Crispim CA (2026). Systematic Review and Meta-analysis of Randomized Clinical Trials Comparing Time-Restricted Eating With and Without Caloric Restriction for Weight Loss. Nutrition Reviews. ；Sun T, Zhang L, Lu Y, 等 (2024). Non-linear relationship between the first meal time of the day and gallstone incidence in American adults. Frontiers in Nutrition. ；Liu T, Wang Y, Wang X, 等 (2023). Habitually Skipping Breakfast Is Associated with the Risk of Gastrointestinal Cancers: Evidence from the Kailuan Cohort Study. Journal of General Internal Medicine. -- Ghi chú:Tranh cãi. Con số 17% về bệnh tim mạch đến từ những nghiên cứu chỉ ghi nhận, không chia nhóm. Người không ăn sáng vốn đã dễ hút thuốc, uống rượu, ngồi nhiều, thu nhập thấp hơn, mà không hiệu chỉnh sạch được. Trong 17% đó có bao nhiêu là do bữa sáng, nói không rõ. Chỉ số tử vong toàn nguyên vốn đã vượt qua 1. Chế độ 16:8 cũng còn tranh cãi: gộp 30 thử nghiệm, 1341 người, với điều kiện hai nhóm ăn bằng nhau, nhóm ăn trong khung giờ giảm nhiều hơn 1.46 kg (−2.65 đến −0.26), đổi lại mất thêm 0.41 kg khối lượng cơ. Nhưng phần lớn các thử nghiệm này rất ngắn, thử nghiệm dài và nghiêm nhất là thử nghiệm mười hai tháng thì nói không có khác biệt. Nhịn ăn lâu còn liên quan tới túi mật. Trong một khảo sát cắt ngang 6.547 người, bữa đầu tiên trong ngày muộn hơn mỗi giờ một chút thì tỷ lệ phát hiện sỏi mật cao hơn khoảng 5% (OR 1.05, 1.02–1.08); người ăn bữa đầu muộn từ 9 giờ tới 14 giờ cao hơn khoảng 49% so với người ăn sớm hơn (OR 1.49, 1.24–1.77). Cơ chế là mật đọng trong túi mật càng lâu càng dễ kết tinh thành sỏi. Nhưng đây là khảo sát tại một thời điểm, cũng có thể là do đã có vấn đề túi mật rồi mới đổi thói quen ăn. Nghiên cứu mẫu Kailuan ở Trung Quốc theo dõi 5.6 năm, có 369 ca ung thư đường tiêu hoá: người không ăn sáng có nguy cơ ung thư đại tràng và trực tràng gấp 2.32 lần (1.34–4.01), ung thư túi mật và đường mật ngoài gan gấp 5.43 lần (1.34–21.93). Khoảng tin cậy của con số sau rộng tới 21.93, nghĩa là số ca quá ít, chỉ coi là manh mối. Gộp lại, xét riêng chuyện giữ cân nặng thì giờ ăn ảnh hưởng rất nhỏ, đừng phí tâm lực vì nó. Còn chuyện bỏ hẳn bữa sáng lâu dài có tốt không, bằng chứng hiện có toàn là tương quan, không chứng minh được nhân quả, nhưng hướng thì nhất quán nghiêng về bất lợi; muốn an toàn thì ăn bình thường, chỉ đừng trông nó giúp giảm cân. Nguyên nhân chính của viêm dạ dày và loét dạ dày không phải là bụng rỗng, mà là Helicobacter pylori và uống thuốc giảm đau lâu dài; mục nào cần xét thì xem mục 23 ở phần 1 (xét Helicobacter pylori, dương tính thì diệt). Xét sức khoẻ ra sỏi mật mà chưa từng đau thì làm gì, xem mục 20 của phần này (xét sức khoẻ ra sỏi mật mà chưa từng đau, đừng cắt túi mật phòng bệnh). Muốn giữ cân nặng, thứ có bằng chứng là ăn gì và ăn bao nhiêu, xem mục 28 ở phần 2 (mỗi ngày đủ 5 phần rau quả), mục 29 ở phần 2 (ăn ít thực phẩm chế biến công nghiệp) và mục 33 ở phần 2 (giữ BMI ở 20–25). Người bị đái thái, người đang uống thuốc hạ đường huyết, hay phụ nữ mang thai thì không áp dụng, bỏ một bữa có thể hạ đường huyết, hãy làm theo chỉ định của bác sĩ. Nhịn ăn cực đoan và nôn kích thực là chuyện khác, xem mục 1 ở phần 28 (đừng dùng ăn kiêu cực đoan, nhịn ăn hoặc nôn kích thực để giữ cân nặng). +- Ghi chú: Tranh cãi. Con số 17% về bệnh tim mạch đến từ những nghiên cứu chỉ ghi nhận, không chia nhóm. Người không ăn sáng vốn đã dễ hút thuốc, uống rượu, ngồi nhiều, thu nhập thấp hơn, mà không hiệu chỉnh sạch được. Trong 17% đó có bao nhiêu là do bữa sáng, nói không rõ. Chỉ số tử vong toàn nguyên vốn đã vượt qua 1. Chế độ 16:8 cũng còn tranh cãi: gộp 30 thử nghiệm, 1341 người, với điều kiện hai nhóm ăn bằng nhau, nhóm ăn trong khung giờ giảm nhiều hơn 1.46 kg (−2.65 đến −0.26), đổi lại mất thêm 0.41 kg khối lượng cơ. Nhưng phần lớn các thử nghiệm này rất ngắn, thử nghiệm dài và nghiêm nhất là thử nghiệm mười hai tháng thì nói không có khác biệt. Nhịn ăn lâu còn liên quan tới túi mật. Trong một khảo sát cắt ngang 6.547 người, bữa đầu tiên trong ngày muộn hơn mỗi giờ một chút thì tỷ lệ phát hiện sỏi mật cao hơn khoảng 5% (OR 1.05, 1.02–1.08); người ăn bữa đầu muộn từ 9 giờ tới 14 giờ cao hơn khoảng 49% so với người ăn sớm hơn (OR 1.49, 1.24–1.77). Cơ chế là mật đọng trong túi mật càng lâu càng dễ kết tinh thành sỏi. Nhưng đây là khảo sát tại một thời điểm, cũng có thể là do đã có vấn đề túi mật rồi mới đổi thói quen ăn. Nghiên cứu mẫu Kailuan ở Trung Quốc theo dõi 5.6 năm, có 369 ca ung thư đường tiêu hoá: người không ăn sáng có nguy cơ ung thư đại tràng và trực tràng gấp 2.32 lần (1.34–4.01), ung thư túi mật và đường mật ngoài gan gấp 5.43 lần (1.34–21.93). Khoảng tin cậy của con số sau rộng tới 21.93, nghĩa là số ca quá ít, chỉ coi là manh mối. Gộp lại, xét riêng chuyện giữ cân nặng thì giờ ăn ảnh hưởng rất nhỏ, đừng phí tâm lực vì nó. Còn chuyện bỏ hẳn bữa sáng lâu dài có tốt không, bằng chứng hiện có toàn là tương quan, không chứng minh được nhân quả, nhưng hướng thì nhất quán nghiêng về bất lợi; muốn an toàn thì ăn bình thường, chỉ đừng trông nó giúp giảm cân. Nguyên nhân chính của viêm dạ dày và loét dạ dày không phải là bụng rỗng, mà là Helicobacter pylori và uống thuốc giảm đau lâu dài; mục nào cần xét thì xem mục 23 ở phần 1 (xét Helicobacter pylori, dương tính thì diệt). Xét sức khoẻ ra sỏi mật mà chưa từng đau thì làm gì, xem mục 20 của phần này (xét sức khoẻ ra sỏi mật mà chưa từng đau, đừng cắt túi mật phòng bệnh). Muốn giữ cân nặng, thứ có bằng chứng là ăn gì và ăn bao nhiêu, xem mục 28 ở phần 2 (mỗi ngày đủ 5 phần rau quả), mục 29 ở phần 2 (ăn ít thực phẩm chế biến công nghiệp) và mục 33 ở phần 2 (giữ BMI ở 20–25). Người bị đái thái, người đang uống thuốc hạ đường huyết, hay phụ nữ mang thai thì không áp dụng, bỏ một bữa có thể hạ đường huyết, hãy làm theo chỉ định của bác sĩ. Nhịn ăn cực đoan và nôn kích thực là chuyện khác, xem mục 1 ở phần 28 (đừng dùng ăn kiêu cực đoan, nhịn ăn hoặc nôn kích thực để giữ cân nặng). ### 27. Đừng tin 「thủ dâm tổn hại sức khoẻ」「nhịn thái dâm mới được cải thiện」, điều cần xem thật sự là có làm lỡ giấc ngủ, công việc và quan hệ hay không @@ -245,7 +245,7 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Lợi ích: Xem phần thể chất trước. Một nghiên cứu ở Hoa Kỳ theo dõi 31925 nam giới tới năm 2010. Mỗi người tự báo số lần xuất tinh mỗi tháng, rồi chia nhóm theo số lần. So với người 4 đến 7 lần mỗi tháng, người 21 lần trở lên mỗi tháng có ung thư tuyến tiền liệt HR 0.81 (95% CI 0.72–0.92, thấp hơn khoảng 19%), tính theo số lần ở tuổi 20 đến 29. Tính theo số lần ở tuổi 40 đến 49, HR 0.78 (0.69–0.89, thấp hơn khoảng 22%). Phần giảm chủ yếu là ung thư tuyến tiền liệt nguy cơ thấp. Hướng ngược lại là một nghiên cứu bệnh chứng đối chứng ở Anh, so 431 người bệnh được chẩn đoán trước 60 tuổi với 409 người đối chứng. Người thủ dâm nhiều ở tuổi từ hai mươi đến ba mươi có nguy cơ cao hơn, người thủ dâm nhiều ở tuổi năm mươi mấy thì nguy cơ thấp hơn. Xem tiếp nội dung khiêu dâm. Một tổng quan hệ thống kèm phân tích tổng hợp cho rằng nhiều người cho là mình 「nghiện rồi」, phần lớn là vì hành vi không khớp với điều mình tin, chứ không chỉ vì xem nhiều (Hoa Kỳ / Anh) - Mức bằng chứng: B - Nguồn: Rider JR, Wilson KM, Sinnott JA, et al. (2016). Ejaculation Frequency and Risk of Prostate Cancer: Updated Results with an Additional Decade of Follow-up. European Urology, 70(6), 974–982. ；Dimitropoulou P, Lophatananon A, Easton D, et al. (2009). Sexual activity and prostate cancer risk in men diagnosed at a younger age. BJU International, 103(2), 178–185. ；Grubbs JB, Perry SL, Wilt JA, Reid RC. (2019). Pornography Problems Due to Moral Incongruence: An Integrative Model with a Systematic Review and Meta-Analysis. Archives of Sexual Behavior, 48(2), 397–415. -- Ghi chú:Tranh cãi. Hai nghiên cứu về ung thư tuyến tiền liệt đều chỉ ghi nhận, không có nhóm đối chứng, số lần lại do người tham gia nhớ lại, nên kết luận chỉ có thể dừng ở 「không thấy bằng chứng tổn hại」, không thể ngược lại khuyên nên làm nhiều hơn. Chính tác giả nghiên cứu ở Anh cũng nói, đoạn tuổi năm mươi mấy có thể là đã bị bệnh rồi mới ít xuất tinh hơn. Tóm tắt bài nghiên cứu về nội dung khiêu dâm không đưa ra cỡ hiệu ứng, nên ở đây chỉ ghi hướng. Lợi ích đặt là nhỏ, vì phần giảm chủ yếu là ung thư tuyến tiền liệt nguy cơ thấp, không phải tử vong. Dấu hiệu thật sự nên đi khoa tâm thần hoặc khoa tâm lý lâm sàng là: thức khuya vì chuyện này, mất công, ảnh hưởng quan hệ với bạn đời, muốn dừng mà không dừng được. Thức khuya thì xem mục 9 ở phần 3 (đừng thức khuya vì game, video ngắn, nội dung khiêu dâm). Lo xem nội dung khiêu dâm nhiều dẫn tới rối loạn cương, trước hết hãy đi xét theo mục 28 ở phần 1 (rối loạn cương thì đi xét tim mạch trước). Gửi vào nhóm hoặc bán tài nguyên thì hậu quả pháp lý, xem mục 4 ở phần 9 (xem video khiêu dâm ở nhà thì phần mình tự chịu). Người hưởng là chính bạn. +- Ghi chú: Tranh cãi. Hai nghiên cứu về ung thư tuyến tiền liệt đều chỉ ghi nhận, không có nhóm đối chứng, số lần lại do người tham gia nhớ lại, nên kết luận chỉ có thể dừng ở 「không thấy bằng chứng tổn hại」, không thể ngược lại khuyên nên làm nhiều hơn. Chính tác giả nghiên cứu ở Anh cũng nói, đoạn tuổi năm mươi mấy có thể là đã bị bệnh rồi mới ít xuất tinh hơn. Tóm tắt bài nghiên cứu về nội dung khiêu dâm không đưa ra cỡ hiệu ứng, nên ở đây chỉ ghi hướng. Lợi ích đặt là nhỏ, vì phần giảm chủ yếu là ung thư tuyến tiền liệt nguy cơ thấp, không phải tử vong. Dấu hiệu thật sự nên đi khoa tâm thần hoặc khoa tâm lý lâm sàng là: thức khuya vì chuyện này, mất công, ảnh hưởng quan hệ với bạn đời, muốn dừng mà không dừng được. Thức khuya thì xem mục 9 ở phần 3 (đừng thức khuya vì game, video ngắn, nội dung khiêu dâm). Lo xem nội dung khiêu dâm nhiều dẫn tới rối loạn cương, trước hết hãy đi xét theo mục 28 ở phần 1 (rối loạn cương thì đi xét tim mạch trước). Gửi vào nhóm hoặc bán tài nguyên thì hậu quả pháp lý, xem mục 4 ở phần 9 (xem video khiêu dâm ở nhà thì phần mình tự chịu). Người hưởng là chính bạn. ### 28. Đừng bỏ tiền làm 「điều chỉnh xu hướng tính dục」「chữa chứng đồng tính」, cũng đừng đưa người nhà đi @@ -254,4 +254,4 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Lợi ích: Báo cáo của nhóm công tác năm 2009 của Hiệp hội Tâm lý Hoa Kỳ nói nỗ lực đổi xu hướng tính dục 「khó có thể thành công, và có một mức rủi ro gây hại nhất định」. Với việc đưa trẻ vị thành niên vào các chương trình điều chỉnh bắt buộc hoặc nằm viện, báo cáo khuyên cả cha mẹ lẫn đứa trẻ đều không nên cân nhắc. Một khảo sát toàn quốc có tính đại diện ở Hoa Kỳ phỏng vấn 1518 người lớn thuộc nhóm thiểu số tính dục, khoảng 7% từng trải qua điều chỉnh, trong đó tám phần mười đến từ người có đạo. Sau khi trừ tuổi, các trải nghiệm bất hạnh thời thơ ấu và những yếu tố khác, người từng trải qua điều chỉnh có OR từng có ý nghĩ tự sát trong đời là 1.92 (95% CI 1.01–3.64, cao hơn khoảng chín phần mươi). OR đã lên kế hoạch tự sát là 1.75 (1.01–3.06). OR thử tự sát không bị thương hoặc bị thương nhẹ là 1.88 (1.01–3.50). Một nghiên cứu khác phỏng vấn 245 người trẻ từ 21 đến 25 tuổi. Trong số người không bị điều chỉnh ở tuổi vị thành niên, 22.0% đã thử tự sát. Người mà cha mẹ tự tìm cách đổi xu hướng tính dục của con thì chiếm 48.1%. Người còn bị cha mẹ đưa đi tìm nhà trị liệu hoặc người tôn giáo thì chiếm 62.8%. Bảng phân loại bệnh ICD-10 của Tổ chức Y tế Thế giới ghi rõ: 「bản thân xu hướng tính dục không được coi là một rối loạn」. Năm 2017 Tòa án huyện Dịch Thành, trấn Trực Mã, tỉnh Hà Nam xét một vụ: bệnh viện tâm thích dùng danh nghĩa 「rối loạn xu hướng tình dục」 để ép nhận bệnh một người đàn ông. Tòa cho rằng ông 「bệnh ở mức trung bình, cũng không có hành vi hoặc nguy cơ tự sát, làm hại người, phá tài sản, không đáp ứng điều kiện phải điều trị bắt buộc」. Tòa buộc bệnh viện công khai xin lỗi và bồi 5000 tệ bù tổn thương tinh thần (Hoa Kỳ / Trung Quốc) - Mức bằng chứng: B - Nguồn: American Psychological Association Task Force on Appropriate Therapeutic Responses to Sexual Orientation (2009). Report of the Task Force on Appropriate Therapeutic Responses to Sexual Orientation. ；Blosnich JR, Henderson ER, Coulter RWS, Goldbach JT, Meyer IH. (2020). Sexual Orientation Change Efforts, Adverse Childhood Experiences, and Suicide Ideation and Attempt Among Sexual Minority Adults, United States, 2016–2018. American Journal of Public Health, 110(7), 1024–1030. ；Ryan C, Toomey RB, Diaz RM, Russell ST. (2020). Parent-Initiated Sexual Orientation Change Efforts With LGBT Adolescents: Implications for Young Adult Mental Health and Adjustment. Journal of Homosexuality, 67(2), 159–173. ；World Health Organization. ICD-10 Version:2019, F66 Psychological and behavioural disorders associated with sexual development and orientation. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2018, sửa đổi). Luật Sức khoẻ Tâm thần（精神卫生法）（第二十三、三十、七十八条）. ；Trang web Viện kiểm sát nhân dân Khu tự trị Quảng Tây đăng lại bài của Tân Kinh Báo (2017). 驻马店「同性恋强制治疗案」终审. ；Phản biện: Sullins DP. (2022). Sexual Orientation Change Efforts Do Not Increase Suicide: Correcting a False Research Narrative. Archives of Sexual Behavior, 51, 3377–3393. ；Blosnich JR, Coulter RWS, Henderson ER, et al. (2023). Correcting a False Research Narrative: A Commentary on Sullins (2022). Archives of Sexual Behavior, 52(3), 885–888. -- Ghi chú:Tranh cãi. Sullins năm 2022 tính lại trên cùng bộ dữ liệu đó, và trừ đi những ý nghĩ tự sát đã có sẵn trước khi điều chỉnh. Tính như vậy thì việc điều chỉnh và ý nghĩ tự sát không còn tương quan dương (OR 0.44, 0.20–0.94). Cơ quan nơi Sullins làm việc có lập trường chống đồng tính, và tác giả gốc cũng đã viết phản hồi. Vậy mức độ gây hại còn tranh cãi, nhưng không thấy bằng chứng phản biện cho điểm 「không đổi được」. Đặt mức B vì các con số về hậu quả đều đến từ khảo sát hồi tưởng sau sự việc. Về mặt pháp lý, điều trị nằm viện theo nguyên tắc tự nguyện, chỉ khi có rối loạn tâm thần nặng và có nguy cơ làm hại bản thân hoặc người khác thì mới ép nhận bệnh. Nhân viên tâm lý không được thực hành trị liệu tâm lý. Cố ý đưa người không bị rối loạn tâm thần vào bệnh viện để điều trị thì phải chịu trách nhiệm bồi thường. Người đau khổ vì chuyện này lâu dài có thể đi khoa tâm thần hoặc khoa tâm lý lâm sàng, mục tiêu là giảm đau khổ. Con cái nói ra thì gia đình nên làm gì, xem mục 15 ở phần 30 (con nói mình thích người cùng giới). Người hưởng là chính bạn và người nhà. +- Ghi chú: Tranh cãi. Sullins năm 2022 tính lại trên cùng bộ dữ liệu đó, và trừ đi những ý nghĩ tự sát đã có sẵn trước khi điều chỉnh. Tính như vậy thì việc điều chỉnh và ý nghĩ tự sát không còn tương quan dương (OR 0.44, 0.20–0.94). Cơ quan nơi Sullins làm việc có lập trường chống đồng tính, và tác giả gốc cũng đã viết phản hồi. Vậy mức độ gây hại còn tranh cãi, nhưng không thấy bằng chứng phản biện cho điểm 「không đổi được」. Đặt mức B vì các con số về hậu quả đều đến từ khảo sát hồi tưởng sau sự việc. Về mặt pháp lý, điều trị nằm viện theo nguyên tắc tự nguyện, chỉ khi có rối loạn tâm thần nặng và có nguy cơ làm hại bản thân hoặc người khác thì mới ép nhận bệnh. Nhân viên tâm lý không được thực hành trị liệu tâm lý. Cố ý đưa người không bị rối loạn tâm thần vào bệnh viện để điều trị thì phải chịu trách nhiệm bồi thường. Người đau khổ vì chuyện này lâu dài có thể đi khoa tâm thần hoặc khoa tâm lý lâm sàng, mục tiêu là giảm đau khổ. Con cái nói ra thì gia đình nên làm gì, xem mục 15 ở phần 30 (con nói mình thích người cùng giới). Người hưởng là chính bạn và người nhà. diff --git "a/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" "b/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" index 9187388e..def46825 100644 --- "a/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" +++ "b/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" @@ -70,7 +70,7 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Chi phí: Không tốn tiền. Kiểm tra một lần, mất mười mấy phút. - Nói thẳng: Có cơ sở nhân danh nghề mới, mức kỹ năng để mở lớp lung tung, thu tiền sai, cấp chứng chỉ ẩu, có cả cơ sở giả mạo cơ quan có thẩm quyền. Bộ Nhân sự và Xã hội đã có đợt xử lý riêng về việc này. Trước khi đăng ký, kiểm tra ba bước: nghề này có trong Danh mục chứng chỉ nghề nghiệp quốc gia không, cơ quan cấp chứng chỉ có phải cơ quan đánh giá đã đăng ký với cơ quan nhân sự không, mã số chứng chỉ tra trên nền tảng chính thức có ra không. Không tra được thì tiền đăng ký là mất trắng. -- Lợi ích: Bộ Nhân sự và Xã hội đã mở đợt xử lý toàn quốc riêng cho việc này. Những sai sót văn bản nêu tên gồm「一些机构和单位以新职业职业资格、职业技能等级等名义随意举办培训、评价、发证活动，乱收费、滥发证；一些机构和单位虚假或夸大宣传，甚至假冒权威机构名义组织培训、评价、颁发证书等」, tức là tổ chức, đơn vị lấy danh nghĩa chứng chỉ nghề mới hay mức kỹ năng nghề để tuỳ tiện mở đào tạo, đánh giá, cấp chứng chỉ, thu tiền lung tung, cấp quá đà, còn có nơi quảng cáo sai hoặc phóng đại, thậm chí giả mạo cơ quan có thẩm quyền để tổ chức đào tạo, đánh giá, cấp chứng chỉ. Biện pháp xử lý gồm phối hợp với các cơ quan quản lý không gian mạng, công an, thị trường để điều tra,「严厉打击假冒鉴定机构、人招部门备案评价机构的评价发证活动」, tức là đánh mạnh vào hoạt động đánh giá cấp chứng chỉ của cơ quan giả mạo cơ quan chuyên môn và của cơ quan đánh giá đã đăng ký. Quảng cáo không hợp lệ trên trang web và nền tảng mới thì đóng theo luật. Văn bản đồng thời yêu cầu「引导公众了解证书查询渠道和方式，判断证书合法合规性和含金量」 +- Lợi ích: Bộ Nhân sự và Xã hội đã mở đợt xử lý toàn quốc riêng cho việc này. Những sai sót văn bản nêu tên gồm「一些机构和单位以新职业职业资格、职业技能等级等名义随意举办培训、评价、发证活动，乱收费、滥发证；一些机构和单位虚假或夸大宣传，甚至假冒权威机构名义组织培训、评价、颁发证书等」, tức là tổ chức, đơn vị lấy danh nghĩa chứng chỉ nghề mới hay mức kỹ năng nghề để tuỳ tiện mở đào tạo, đánh giá, cấp chứng chỉ, thu tiền lung tung, cấp quá đà, còn có nơi quảng cáo sai hoặc phóng đại, thậm chí giả mạo cơ quan có thẩm quyền để tổ chức đào tạo, đánh giá, cấp chứng chỉ. Biện pháp xử lý gồm phối hợp với các cơ quan quản lý không gian mạng, công an, thị trường để điều tra,「严厉打击假冒鉴定机构、人社部门备案评价机构的评价发证活动」, tức là đánh mạnh vào hoạt động đánh giá cấp chứng chỉ của cơ quan giả mạo cơ quan chuyên môn và của cơ quan đánh giá đã đăng ký. Quảng cáo không hợp lệ trên trang web và nền tảng mới thì đóng theo luật. Văn bản đồng thời yêu cầu「引导公众了解证书查询渠道和方式，判断证书合法合规性和含金量」 - Mức bằng chứng: A - Nguồn: Bộ Nhân sự và Xã hội (2022). 关于开展技术技能类「山寨证书」专项治理工作的通知. - Ghi chú: Kiểm tra theo đúng thứ tự này. Trước hết xem nghề đó có trong《国家职业资格目录》hay không, không có trong danh mục thì không phải chứng chỉ nghề nghiệp quốc gia. Rồi xem cơ quan cấp chứng chỉ có phải cơ quan đánh giá đã đăng ký với cơ quan nhân sự không. Cuối cùng vào nền tảng tra chứng chỉ đánh giá nhân tài kỹ năng chính thức, tra mã số chứng chỉ. Tra không ra thì đừng đăng ký. Loại nào có chữ「quốc tế」,「cao cấp」,「đăng ký」trong tên, hoặc hứa「đảm bảo đỗ」,「treo giấy phép có thu nhập」, thì rủi ro cao nhất. @@ -163,47 +163,47 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Pashler H, McDaniel M, Rohrer D, Bjork R (2008). Learning Styles: Concepts and Evidence. Psychological Science in the Public Interest, 9(3), 105–119. - Ghi chú: Tổng quan đồng thời nói, điều đó không có nghĩa mọi phiên bản của lý thuyết phong cách học đều đã bị kiểm tra và bị bác bỏ, nhiều phiên bản chưa hề được kiểm tra. Cái cần tránh là bỏ tiền làm bài kiểm tra phong cách học và chọn lớp đào tạo theo phong cách. Tiêu chí chọn đào tạo xem mục 9 của phần này (đào tạo nên đi qua kênh trợ cấp của nhà nước) và mục 13 của phần này (ưu tiên chương trình ngắn hạn học xong là vào việc được ngay). -### 19. Phần cần học thuộc thì tự ra câu hỏi theo cách dùng về sau để tự kiểm, đừng đọc thuộc một lượt là xong +### 19. Muốn học thuộc, hãy ra đề theo cách dùng về sau rồi tự kiểm tra, đừng chỉ đọc thuộc một lượt là xong -- Chi phí: Không tốn tiền, cũng không tốn thêm thời gian. Cách luyện đổi thôi: thay「đọc thuộc một lượt nguyên văn」bằng「tự ra câu hỏi theo cách dùng về sau để tự kiểm」. Chỗ khó là phải nghĩ trước ra những thứ đó về sau sẽ dùng thế nào. -- Nói thẳng: Đóng sách tự kiểm thì ích lợi không dừng ở mấy chữ đã đọc thuộc. Cộng lại 192 lần so sánh và hơn một vạn người thì cách đó dùng được cho câu hỏi đổi dạng, cho bài toán ứng dụng và bài suy luận, cỡ hiệu quả trung bình hơi nhỏ. Nhưng có hai trường hợp gần như không dùng được: đảo câu hỏi và đáp án, và nội dung chỉ từng thấy trong sách mà chưa bao giờ bị kiểm. -- Lợi ích: Một phân tích tổng hợp gom các nghiên cứu hơn 40 năm, gồm 67 bài đã công bố và chưa công bố, 122 thử nghiệm, 192 cỡ hiệu quả chuyển đổi, tổng cộng 10382 người. So với nhóm đối chứng「xem lại một lượt mà không kiểm」thì học đạt được nhờ bài kiểm tra có luyện tập thật sự chuyển sang được tình huống khác, cỡ hiệu quả d=0.40 (95% CI 0.31 đến 0.50). Mấu chuyển đổi mạnh nhất là: đổi dạng câu hỏi để kiểm, kiểm bài toán ứng dụng và bài suy luận, câu hỏi kiểu chẩn đoán y khoa, kiểm bằng những từ dẫn nghĩa gần giống nhau. Mấu chuyển đổi yếu nhất là: đảo kích thích và phản ứng, tư liệu gặp lúc mới học nhưng chưa từng bị kiểm, câu hỏi dạng bài mẫu. Trong các điều kiện điều tiết, dạng câu trả lời lúc luyện và lúc kiểm có giống nhau không, đã từng luyện gọi lại với phản hồi chi tiết chưa, và lần kiểm đầu tiên đã trả lời đúng chưa, đều ảnh hưởng mạnh tới việc có xuất hiện chuyển đổi tích cực hay không. Tác giả còn dùng hai cách hiệu chỉnh thiên lệch công bố, hiệu quả của các điều kiện điều tiết bị ảnh hưởng rất ít, nhưng ước lượng điểm cắt bị hạ rõ rệt: khi không có điều kiện nào vừa nêu thì thường không thấy chuyển đổi tích cực +- Chi phí: Không tốn tiền, cũng không tốn thêm thời gian. Chỉ có cách luyện là thay đổi: đổi「đọc thuộc một lượt」thành「ra đề theo cách dùng về sau rồi tự kiểm tra」. Khó ở chỗ phải nghĩ trước sau này dùng những thứ này vào việc gì. +- Nói thẳng: Đóng sách ra bài tự kiểm tra, tác dụng không chỉ dừng ở vài chữ đã học thuộc. Tính chung 192 so sánh với hơn mười nghìn người, cách này còn dùng được cho bài đổi dạng câu hỏi, cho bài ứng dụng và bài suy luận, mức độ vừa nhỏ. Nhưng có hai trường hợp gần như vô ích: đảo câu hỏi với đáp án, và nội dung chỉ từng thấy trong sách mà chưa bao giờ bị kiểm tra. +- Lợi ích: Một phân tích tổng hợp gom nghiên cứu hơn 40 năm, gồm 67 bài đã xuất bản và chưa xuất bản, 122 thử nghiệm, 192 hiệu ứng chuyển đổi, tổng cộng 10382 người. So với nhóm đối chứng「đọc lại một lượt nhưng không bị kiểm tra」, học tập do bài kiểm tra có luyện tập mang lại thực sự chuyển sang được tình huống khác, hiệu ứng d=0.40 (95% CI 0.31 đến 0.50). Mấy trường hợp chuyển đổi mạnh nhất là: đổi sang dạng câu hỏi khác, hỏi bài ứng dụng và bài suy luận, câu hỏi kiểu chẩn đoán y khoa, và hỏi bằng những từ gợi ý có nghĩa gần giống. Mấy trường hợp chuyển đổi yếu nhất là: đảo kích thích với đáp ứng, tư liệu từng thấy lúc học đầu nhưng chưa bị kiểm tra, và câu hỏi dạng ví dụ mẫu. Trong các yếu tố điều tiết, dạng trả lời lúc luyện và lúc kiểm tra có giống nhau không, đã từng làm bài nhớ lại kèm phản hồi chi tiết chưa, và lần kiểm tra đầu có trả lời đúng không, đều ảnh hưởng mạnh tới việc có xuất hiện chuyển đổi tích cực hay không. Tác giả còn dùng hai cách hiệu chỉnh thiên lệch công bố, hiệu ứng của các yếu tố điều tiết chịu ảnh hưởng rất nhỏ, nhưng ước lượng điểm cắt bị hạ xuống rõ rệt: khi không có yếu tố nào trong số đó, ta thường không thấy chuyển đổi tích cực - Mức bằng chứng: A - Nguồn: Pan SC, Rickard TC (2018). Transfer of test-enhanced learning: meta-analytic review and synthesis. Psychological Bulletin, 144(7), 710-756. -- Ghi chú: Từ「học thuộc」thường gộp hai việc vào một: xem lại tư liệu (đọc lại, gạch chân), và chủ động gọi lại sau khi đóng sách. Loại trước được xếp hạng thấp nhất trong tổng quan xếp hạng mười cách học, muốn xem chính phần xếp hạng thì xem mục 16 của phần này (đừng coi gạch chân, đọc lại, viết tóm tắt là cách học chính). Cần làm là loại sau: chủ động gọi lại sau khi đóng sách, cụ thể xem mục 14 của phần này (đóng sách tự kiểm), và phải chia ra mấy ngày làm, xem mục 15 của phần này (chia ra mấy ngày). Cách làm là nghĩ trước ra về sau sẽ dùng thế nào: phải chọn ra trong bài thi, hay phải phán đoán tại hiện trường, rồi ra câu hỏi theo đúng dạng đó để tự kiểm. Đáng làm thêm một lượt kiểm đảo chiều, vì trong phân tích tổng hợp này chuyển đổi theo chiều đảo là yếu nhất, cả hai chiều đều phải luyện tới. +- Ghi chú: Từ「học thuộc」thường gộp lẫn hai việc: đọc lại tư liệu (đọc đi đọc lại, gạch chân), và chủ động gọi lại khi đã đóng sách. Loại trước được xếp hạng thấp nhất trong tổng quan xếp hạng mười cách học thường dùng, muốn xem bản thân phần xếp hạng thì xem mục 16 (đừng coi gạch chân, đọc lại, viết tóm tắt là cách học chính). Cái cần làm là loại sau: chủ động gọi lại khi đã đóng sách, cụ thể xem mục 14 (đóng sách tự kiểm tra), và phải chia ra mấy ngày, xem mục 15 (chia ra mấy ngày). Cách làm là nghĩ trước sau này dùng vào việc gì: là phải chọn ra trên bài thi, hay phải phán đoán tại chỗ, rồi ra đề đúng kiểu đó để tự kiểm tra. Ngược lại, kiểm tra theo chiều đảo cũng đáng làm, vì trong phân tích tổng hợp này chiều đảo có chuyển đổi yếu nhất, cả hai chiều đều phải luyện tới. -### 20. Khi xét chức danh nghề nghiệp, trước hết xem mình ở chuỗi nào, bậc nào, rồi tìm cửa khai theo tính chất đơn vị +### 20. Xét chức danh nghề nghiệp: trước hết xem mình thuộc chuyên môn nào, bậc nào, rồi tìm cửa đăng ký theo tính chất đơn vị -- Chi phí: Không tốn tiền. Tra thông báo khai của tỉnh mình năm đó một lần, khoảng nửa tiếng. -- Nói thẳng: Chức danh nghề nghiệp cả nước chia thành 27 chuỗi, mỗi chuỗi chia làm bậc sơ cấp, trung cấp, cao cấp. Thường phải khai từng bậc một. Người làm việc trong đơn vị thì đơn vị xét hồ sơ, công bố rồi chuyển lên trên. Người ở doanh nghiệp tư nhân thì ra điểm tiếp nhận hồ sơ xét chức danh, hoặc nhờ cơ quan dịch vụ nhân sự, hiệp hội nghề nghiệp. Người làm tự do thì dự xét của cơ quan nhân sự địa phương tổ chức. -- Lợi ích: Cả nước có 27 chuỗi chức danh nghề nghiệp, ví dụ kinh tế, kỹ thuật, y tế, xuất bản. Chức danh chia ba tầng cao cấp, trung cấp, sơ cấp. Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội áp dụng cho doanh nghiệp, đơn vị sự nghiệp, tổ chức xã hội, tổ chức kinh tế cá thể, và cả với người làm tự do. Người khai thường phải khai từng bậc một. Đơn vị nơi bạn làm việc phải xét hồ sơ và công bố tại đơn vị, thời gian công bố không dưới 5 ngày làm việc. Doanh nghiệp tư nhân và người làm tự do có thể nhờ cơ quan đại diện nhân sự xét, công bố, giới thiệu. Các địa phương phải đặt điểm tiếp nhận hồ sơ xét chức danh ở khu công nghệ, khu công nghiệp cao và nơi tương tự, cũng có thể tiếp nhận qua cơ quan dịch vụ nhân sự, hiệp hội thương mại, hội liên hiệp nghề nghiệp và hiệp hội khoa học. Người làm tự do đủ điều kiện thì theo nguyên tắc thuộc địa phương nào thì dự xét chức danh do cơ quan nhân sự địa phương tổ chức. Hồ sơ tra được trên mạng thì không được bắt bạn mở giấy tờ khác. Hồ sơ thiếu thì bên xét phải nói một lần là thiếu gì. Người ở doanh nghiệp tư nhân thường khai tại nơi có quan hệ lao động. Người được cử đi công tác ở tỉnh khác liên tục đủ một năm thì được khai xét chức danh tại nơi được cử đi. Người đã nghỉ hưu không được khai. Người lao động của đơn vị sự nghiệp đang bị kỷ luật từ mức khiển trách trở lên thì không được khai trong thời gian bị kỷ luật (toàn quốc) +- Chi phí: Không tốn tiền. Tra thông báo đăng ký của năm đó ở tỉnh mình, khoảng nửa tiếng. +- Nói thẳng: Cả nước chia chức danh nghề nghiệp thành 27 chuyên môn, mỗi chuyên môn chia làm bậc sơ cấp, bậc trung cấp, bậc cao cấp. Thường phải xét lên từng bậc một. Người làm việc trong đơn vị thì đơn vị thẩm định, công khai kết quả rồi xét lên trên. Người ở doanh nghiệp tư nhân thì đến điểm tiếp nhận đăng ký chức danh nghề nghiệp, hoặc nhờ cơ quan môi giới nhân lực, hiệp hội nghề nghiệp đăng ký. Người làm nghề tự do thì đăng ký dự đánh giá do cơ quan nhân sự và xã hội địa phương tổ chức. +- Lợi ích: Cả nước đặt 27 chuyên môn chức danh nghề nghiệp, ví dụ kinh tế, kỹ thuật, y tế, xuất bản. Chức danh chia ba tầng cao cấp, trung cấp, sơ cấp. Lệnh số 40 của Bộ Nhân sự và Xã hội áp dụng cho doanh nghiệp, cơ quan, đơn vị, tổ chức xã hội, tổ chức kinh tế cá thể, và cả người làm nghề tự do. Người đăng ký thường phải xét lên từng bậc. Đơn vị nơi bạn làm việc phải thẩm định hồ sơ và công khai trong nội bộ đơn vị, thời gian công khai không ít hơn 5 ngày làm việc. Doanh nghiệp tư nhân và người làm nghề tự do có thể nhờ cơ quan đại diện nhân sự thay mình thẩm định, công khai, giới thiệu. Các địa phương phải đặt điểm tiếp nhận đăng ký chức danh nghề nghiệp ở khu công nghệ, khu phát triển cao cấp và nơi tương tự, cũng có thể tiếp nhận đăng ký qua cơ quan môi giới nhân lực, hiệp hội nghề nghiệp, hiệp hội khoa học. Người làm nghề tự do đủ điều kiện thì theo địa bàn dự đánh giá do cơ quan nhân sự và xã hội địa phương tổ chức. Tư liệu nào kiểm chứng được trên mạng thì không được yêu cầu bạn xin thêm giấy tờ riêng. Hồ sơ thiếu thì bên xét đánh giá phải nói một lần những gì còn thiếu. Người của doanh nghiệp tư nhân thường đăng ký tại nơi có quan hệ lao động. Người được cử đi làm việc nơi khác liên tục đủ một năm thì được đăng ký tại nơi được cử đi. Người đã nghỉ hưu không được đăng ký. Công chức, viên chức cơ quan, đơn vị bị kỷ luật từ mức kỷ luật có ghi nhận lỗi trở lên thì không được đăng ký trong thời gian chịu kỷ luật (toàn quốc) - Mức bằng chứng: A -- Nguồn: Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội (人力资源和社会保障部令第 40 号) (2019). 职称评审管理暂行规定. ：第二条、第十三条、第十四条、第十五条、第十六条、第十七条、第十八条；Văn phòng Bộ Nhân sự và Xã hội (2020). 关于进一步做好民营企业职称工作的通知. ：「在专业技术人才密集的创业孵化基地、高新技术开发区、科技园区等地设立职称申报受理服务点，或通过人才中介服务机构、工商联、行业协会商会、学会等社会组织受理民营企业专业技术人才职称申报」「民营企业专业技术人才被派驻外地连续工作一年以上的，可按有关规定在派驻地申报职称评审」；Tân Hoa Xinh (2024). 职称评审进入高峰，信息如何查？有何新变化？ ：「我国共设有经济、工程、卫生、出版等27个职称系列」「符合条件的自由职业者，可根据属地原则申报参加当地人社部门组织的职称评审」；Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「不将论文作为评价应用型人才的限制性条件」「对职称外语和计算机应用能力考试不作统一要求」「符合条件的高技能人才，可参加工程系列专业技术人才职称评审」「公务员不得参加专业技术人才职称评审」 -- Ghi chú: Cần chuẩn bị những hồ sơ nào thì cả nước không có một danh sách thống nhất. Mỗi chuỗi, mỗi tỉnh hằng năm đều phát thông báo khai, trong đó ghi danh mục hồ sơ, thời gian khai và điều kiện về thành tích. Nên bước đầu tiên là vào trang cơ quan nhân sự tỉnh tìm thông báo của năm đó. Bài báo không còn bắt buộc phải có: ý kiến cải cách năm 2016 yêu cầu không đặt bài báo làm điều kiện cứng với nhân tài ứng dụng, và không đặt yêu cầu thống nhất cho kỳ thi ngoại ngữ chức danh cũng như ứng dụng máy tính. Người lao động kỹ thuật cũng có đường: nhân tài kỹ năng cao đủ điều kiện được dự xét chức danh chuỗi kỹ thuật. Công chức không được dự xét chức danh. +- Nguồn: Lệnh số 40 của Bộ Nhân sự và Xã hội (2019). 职称评审管理暂行规定. ：第二条、第十三条、第十四条、第十五条、第十六条、第十七条、第十八条；Văn phòng Bộ Nhân sự và Xã hội (2020). 关于进一步做好民营企业职称工作的通知. ：「在专业技术人才密集的创业孵化基地、高新技术开发区、科技园区等地设立职称申报受理服务点，或通过人才中介服务机构、工商联、行业协会商会、学会等社会组织受理民营企业专业技术人才职称申报」「民营企业专业技术人才被派驻外地连续工作一年以上的，可按有关规定在派驻地申报职称评审」；Tân Hoa Xã (2024). 职称评审进入高峰，信息如何查？有何新变化？ ：「我国共设有经济、工程、卫生、出版等27个职称系列」「符合条件的自由职业者，可根据属地原则申报参加当地人社部门组织的职称评审」；Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「不将论文作为评价应用型人才的限制性条件」「对职称外语和计算机应用能力考试不作统一要求」「符合条件的高技能人才，可参加工程系列专业技术人才职称评审」「公务员不得参加专业技术人才职称评审」 +- Ghi chú: Cần chuẩn bị những tài liệu nào thì cả nước không có danh mục thống nhất. Mỗi chuyên môn, mỗi tỉnh hằng năm lại ra thông báo đăng ký, danh mục tài liệu, thời gian đăng ký và điều kiện thành tích đều ghi trong đó. Nên bước đầu tiên là tìm thông báo của năm đó trên trang Bộ Nhân sự và Xã hội của tỉnh. Luận án không còn bắt buộc phải có: ý kiến cải cách năm 2016 yêu cầu không lấy luận án làm ngưỡng cứng với nhân tài ứng dụng, kỳ thi ngoại ngữ chức danh và tin học cũng không có yêu cầu thống nhất. Người lao động kỹ thuật cũng có đường: lao động kỹ năng cao đủ điều kiện có thể dự đánh giá chức danh nghề nghiệp thuộc chuyên môn kỹ thuật. Công chức không được dự đánh giá chức danh nghề nghiệp. -### 21. Chức danh sơ cấp và trung cấp ở ngành kế toán lấy bằng thi quốc gia thống nhất, đăng ký theo học vấn và số năm công +### 21. Chức danh sơ cấp, trung cấp ở kế toán thì lấy bằng kỳ thi thống nhất toàn quốc, đăng ký theo trình độ học vấn và số năm làm việc -- Chi phí: Phải đóng phí đăng ký theo tiêu chuẩn của tỉnh. Ôn thi chiếm mấy tháng thời gian ngoài giờ làm việc. -- Nói thẳng: Có những chuỗi lấy chức danh sơ cấp và trung cấp bằng cách thi, thi đỗ là xong, không xét lại nữa. Kế toán là một trường hợp: kế toán trợ viên cần tốt nghiệp trung học trở lên, kế toán viên cần bằng đại học và làm kế toán đủ 4 năm, hoặc bằng cao đẳng và đủ 5 năm. Người đã có một số chứng chỉ nghề nghiệp cũng có thể được công nhận là có chức danh tương ứng. -- Lợi ích: Ý kiến cải cách năm 2016 quy định, với chuyên ngành thực hiện kỳ thi quốc gia thống nhất cho chức danh sơ cấp và trung cấp thì không tiếp tục xét hoặc công nhận tương ứng. Ở những lĩnh vực mà chức danh gắn chặt với chứng chỉ nghề nghiệp, có chứng chỉ nghề nghiệp thì được công nhận là có chức danh tương ứng, và còn dùng làm điều kiện để khai bậc cao hơn. Lấy ngành kế toán làm ví dụ. Kế toán trợ viên và kế toán viên thực hiện kỳ thi quốc gia thống nhất. Kế toán trưởng thì kết hợp thi và xét, chức danh cao cấp nhất thường chỉ xét. Điều kiện dự thi như sau. Kế toán trợ viên: tốt nghiệp trung học trở lên, gồm trường trung cấp chuyên nghiệp, trường nghề, trường kỹ thuật. Kế toán viên: bằng tiến sĩ; hoặc thạc sĩ, làm kế toán đủ 1 năm; hoặc bằng cử nhân, tốt nghiệp lớp sau đại học, làm đủ 2 năm; hoặc đại học, làm đủ 4 năm; hoặc cao đẳng, làm đủ 5 năm. Người có bằng thạc sĩ kế toán, tiến sĩ kế toán thì khi thi kế toán viên được miễn môn Quản trị tài chính (toàn quốc) +- Chi phí: Phí đăng ký trả theo tiêu chuẩn của từng tỉnh. Ôn thi thì chiếm mấy tháng thời gian ngoài giờ làm việc. +- Nói thẳng: Có chuyên môn mà chức danh sơ cấp, trung cấp là do thi, thi đợt là có, không còn xét đánh giá riêng nữa. Kế toán là một ví dụ: kế toán trợ viên cần tốt nghiệp trung học trở lên, kế toán viên cần bằng cử nhân đã làm kế toán đủ 4 năm, bằng cao đẳng đủ 5 năm. Người đã có một số chứng chỉ nghề nghiệp cũng có thể được coi là đã có chức danh tương ứng. +- Lợi ích: Ý kiến cải cách năm 2016 quy định, những chuyên môn mà chức danh sơ cấp, trung cấp thi theo kỳ thi thống nhất toàn quốc thì không còn xét đánh giá hay công nhận tương ứng. Ở những lĩnh vực chức danh nghề nghiệp gắn chặt với chứng chỉ nghề nghiệp, ai đã có chứng chỉ nghề nghiệp thì được công nhận là có chức danh tương ứng, và còn làm điều kiện đăng ký bậc cao hơn. Lấy kế toán làm ví dụ. Kế toán trợ viên và kế toán viên thi theo kỳ thi thống nhất toàn quốc. Kế toán trưởng thì kết hợp thi và xét đánh giá, kế toán trưởng cấp cao thường chỉ xét đánh giá. Điều kiện dự thi như sau. Kế toán trợ viên: tốt nghiệp trung học trở lên, gồm trường trung cấp, trường kỹ nhuận, trường kỹ thuật. Kế toán viên: bằng tiến sĩ; hoặc bằng thạc sĩ, làm kế toán đủ 1 năm; hoặc bằng cử nhân thứ hai, hoàn thành lớp sau đại học, đủ 2 năm; hoặc bằng cử nhân, đủ 4 năm; hoặc bằng cao đẳng, đủ 5 năm. Người có bằng thạc sĩ hay tiến sĩ chuyên ngành kế toán thì khi thi kế toán viên được miễn thi môn《财务管理》(toàn quốc) - Mức bằng chứng: A -- Nguồn: Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「专业技术人才取得职业资格即可认定其具备相应系列和层级的职称，并可作为申报高一级职称的条件。初级、中级职称实行全国统一考试的专业不再进行相应的职称评审或认定。」；Bộ Nhân sự và Xã hội, Bộ Tài chính (2019). 关于深化会计人员职称制度改革的指导意见. ：「助理会计师、会计师实行全国统一的会计专业技术资格考试」「高级会计师采取考试与评审相结合方式，正高级会计师一般采取评审方式」；Bộ Tài chính và cộng sự (2024). 关于做好会计专业学位与会计专业技术资格衔接有关工作的通知. -- Ghi chú: Chuỗi của bạn là thi, là xét, hay kết hợp thi và xét thì xem bản「hướng dẫn về đẩy mạnh cải cách hệ thống chức danh ×××」của chính chuỗi đó. Bộ Nhân sự và Xã hội cùng các cơ quan quản lý ngành đã phát hành theo từng chuỗi, tra được trong kho chính sách trên trang Chính phủ Trung Quốc. Thời gian đăng ký thi và các môn thi thì theo thông báo thi của năm đó. Công chức được dự thi kỳ thi kế toán, nhưng không được dự xét chức danh kế toán. +- Nguồn: Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「专业技术人才取得职业资格即可认定其具备相应系列和层级的职称，并可作为申报高一级职称的条件。初级、中级职称实行全国统一考试的专业不再进行相应的职称评审或认定。」；Bộ Nhân sự và Xã hội, Bộ Tài chính (2019). 关于深化会计人员职称制度改革的指导意见. ：「助理会计师、会计师实行全国统一的会计专业技术资格考试」「高级会计师采取考试与评审相结合方式，正高级会计师一般采取评审方式」；Bộ Tài chính và các cơ quan khác (2024). 关于做好会计专业学位与会计专业技术资格衔接有关工作的通知. +- Ghi chú: Chuyên môn của bạn là thi, là xét đánh giá, hay kết hợp cả hai thì xem bản「Hướng dẫn để cải cách hệ thống chức danh nghề nghiệp ××」của chuyên môn đó. Bộ Nhân sự và Xã hội cùng các cơ quan chủ quản ngành đã ban hành theo từng chuyên môn, tra được trong kho chính sách của Cổng Chính phủ Trung Quốc. Thời gian đăng ký và các môn thi theo thông báo thi của năm đó. Công chức được dự kỳ thi kế toán, nhưng không được dự đánh giá chức danh nghề nghiệp kế toán. -### 22. Đừng thuê trung gian xét thay, đừng mua bài báo viết hộ, đừng bịa hồ sơ: bị phát hiện thì chức danh bị thu hồi, lưu vào hồ sơ lý tính 3 năm +### 22. Đừng nhờ trung gian xét thay, đừng mua dịch vụ viết thay luận án, đừng làm giả hồ sơ: bị phát hiện thì huỷ chức danh và lưu hồ sơ lưu tín 3 năm -- Chi phí: Không tốn tiền. Chỗ khó là nhịn được quảng cáo「bỏ tiền là đảm bảo đỗ」. -- Nói thẳng: Chức danh có được bằng hồ sơ giả, bài báo thuê viết, hoặc chiếm đoạt kết quả của người khác thì bị phát hiện là thu hồi. Bản ghi mất uy tín bị lưu 3 năm, và đưa vào nền tảng tín dựng toàn quốc. Cam kết bạn ký lúc khai mà không đúng thì 3 năm sau không được khai nữa. Cơ quan nhân sự còn chuyên điều tra quảng cáo sai, xét giả và chứng chỉ giả của các trung gian. -- Lợi ích: Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội quy định, người cung cấp hồ sơ giả, chiếm đoạt tác phẩm và kết quả học thuật của người khác, hoặc dùng thủ đoạn bất chính khác để lấy chức danh thì chức danh bị thu hồi. Đồng thời ghi vào kho hồ sơ lý tính xét chức danh, đưa vào nền tảng chia sẻ thông tin tín dụng toàn quốc, thời hạn ghi là 3 năm. Phương pháp giám sát năm 2024 nêu tên mấy loại vi phạm của người khai: cam kết sai và thông tin bịa, bài báo bịa hoặc thuê viết, thành tích bịa, và nói nhờ gây áp lực. Lúc khai phải ký cam kết cá nhân, nếu cam kết không đúng thì 3 năm không được khai. Cơ quan nhân sự phối hợp với công an, cơ quan quản lý không gian mạng và thị trường để điều tra trang web giả, quảng cáo sai, bẫy hợp đồng, xét giả và chứng chỉ giả của trung gian. Vụ nghiêm trọng mà nghi ngờ phạm tội thì chuyển cơ quan có thẩm quyền xử lý (toàn quốc) +- Chi phí: Không tốn tiền. Khó ở chỗ nhịn được quảng cáo「trả tiền là bảo đảm đỗ」. +- Nói thẳng: Chức danh nghề nghiệp mà có được nhờ tài liệu giả, nhờ thuê người viết luận án, nhờ chiếm đoạt kết quả nghiên cứu thì bị phát hiện là huỷ ngay. Bản ghi mất uy tín lưu lại 3 năm, và còn đưa vào nền tảng tín dụng toàn quốc. Bản cam kết ký lúc đăng ký mà không đúng thì 3 năm sau không được đăng ký nữa. Cơ quan nhân sự và xã hội còn xử lý riêng quảng cáo sai, đánh giá giả, chứng chỉ giả của các trung gian. +- Lợi ích: Lệnh số 40 của Bộ Nhân sự và Xã hội quy định, người dùng tài liệu giả, chiếm đoạt tác phẩm và kết quả học thuật của người khác, hoặc dùng thủ đoạn không chính đáng khác để có được chức danh nghề nghiệp thì chức danh bị huỷ. Đồng thời bị ghi vào kho hồ sơ lưu tín xét đánh giá chức danh nghề nghiệp, đưa vào nền tảng chia sẻ thông tin tín dụng toàn quốc, thời hạn lưu 3 năm. Biện pháp giám sát năm 2024 nêu tên mấy loại vi phạm của người đăng ký: cam kết giả và thông tin bịa đặt, luận án làm giá hoặc thuê viết, thành tích bịa đặt, nể nang quan hệ xin xỏ. Lúc đăng ký phải ký bản cam kết cá nhân, cam kết không đúng thì 3 năm sau không được đăng ký. Cơ quan nhân sự và xã hội phối hợp với công an, cục quản lý không gian mạng, cơ quan quản lý thị trường xử lý website giả, quảng cáo sai, bẫy hợp đồng, đánh giá giả và chứng chỉ giả của các trung gian. Vụ việc nghiêm trọng, nghi có dấu hiệu phạm tội thì chuyển cho cơ quan có thẩm quyền xử lý (toàn quốc) - Mức bằng chứng: A -- Nguồn: Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội (人力资源和社会保障部令第 40 号) (2019). 职称评审管理暂行规定. ：第三十九条；Bộ Nhân sự và Xã hội (2024). 职称评审监管暂行办法. ：第五条、第十五条、第十七条、第二十七条；Văn phòng Bộ Nhân sự và Xã hội (2022). 关于进一步做好职称评审工作的通知. ：「未经备案的职称评审委员会，评审结果不纳入全国职称评审信息查询验证系统」；Tân Hoa Xinh (2024). 职称评审进入高峰，信息如何查？有何新变化？ ：「大家可通过www.12333.gov.cn网页、手机12333客户端、人力资源社会保障部门户网站、“人力资源和社会保障部”微信公众号等多种渠道进行登录」 -- Ghi chú: Có một cách phân biệt thật giả: chức danh do hội đồng xét hợp lệ cấp thì tra được trên nền tảng tra chứng chỉ xét chức danh toàn quốc. Cửa vào là trang 12333, ứng dụng 12333 và trang của Bộ Nhân sự và Xã hội. Hội đồng xét chưa đăng ký mà cấp thì hệ thống đó tra không ra. Vì vậy loại hứa「đảm bảo đỗ」,「không cần bạn dự thi」,「không tra được chứng chỉ vẫn dùng được」thì cơ bản có thể kết luận là có vấn đề. Người hưởng lợi chính là bạn, mức lợi ích định theo cách「tránh bị thu hồi chức danh và tránh 3 năm lưu vào hồ sơ mất uy tín」. +- Nguồn: Lệnh số 40 của Bộ Nhân sự và Xã hội (2019). 职称评审管理暂行规定. ：第三十九条；Bộ Nhân sự và Xã hội (2024). 职称评审监管暂行办法. ：第五条、第十五条、第十七条、第二十七条；Văn phòng Bộ Nhân sự và Xã hội (2022). 关于进一步做好职称评审工作的通知. ：「未经备案的职称评审委员会，评审结果不纳入全国职称评审信息查询验证系统」；Tân Hoa Xã (2024). 职称评审进入高峰，信息如何查？有何新变化？ ：「大家可通过www.12333.gov.cn网页、手机12333客户端、人力资源社会保障部门户网站、“人力资源和社会保障部”微信公众号等多种渠道进行登录」 +- Ghi chú: Có một cách phân biệt thật giả: chức danh nghề nghiệp do đánh giá đúng quy trình thì tra được trên nền tảng tra chức danh nghề nghiệp toàn quốc. Cửa vào là trang 12333, ứng dụng 12333 trên điện thoại, và trang Bộ Nhân sự và Xã hội. Chức danh do hội đồng xét đánh giá chưa đăng ký chấp nhận thì hệ thống này tra không ra. Nên hứa「bảo đảm đỗ」,「không cần bản thân tham dự」,「tra không thấy chứng chỉ vẫn dùng được」 thì về cơ bản có thể kết luận là có vấn đề. Người hưởng lợi chủ yếu là bạn, cỡ lợi ích được định theo「tránh bị huỷ chức danh và bị ghi mất uy tín 3 năm」. -### 23. Có chức danh không có nghĩa tăng lương: hỏi rõ đơn vị xét và tuyển theo tỷ lệ vị trí, hay xét rồi chưa chắc được tuyển +### 23. Có chức danh nghề nghiệp không có nghĩa lương tăng: trước hết hỏi rõ đơn vị xét theo tỷ lệ vị trí, hay xét rồi chưa chắc dùng -- Chi phí: Không tốn tiền. Hỏi một lần bộ phận nhân sự của đơn vị là xong. -- Nói thẳng: Chức danh là căn cứ để tuyển dụng, đánh giá và thăng chức, nhưng được xét và được tuyển là hai chuyện. Đơn vị sự nghiệp thường xét trong tỷ lệ vị trí, được xét thì phần lớn được tuyển. Đơn vị không thực hiện quản lý theo vị trí thì có thể tách việc xét và việc tuyển, đã xét thì đơn vị không nhất thiết cho đúng vị trí và mức lương tương ứng. Đổi đơn vị, đổi địa phương thì chức danh có thể phải xét lại hoặc xác nhận. -- Lợi ích: Điều hai của mệnh lệnh số 40 của Bộ Nhân sự và Xã hội: kết quả xét chức danh là căn cứ quan trọng cho việc tuyển dụng, đánh giá, thăng tiến của nhân tài chuyên môn. Ý kiến cải cách năm 2016 quy định, đối với đơn vị sự nghiệp thực hiện quản lý theo vị trí đầy đủ thì tổ chức xét chức danh trong tỷ lệ cơ cấu vị trí. Đơn vị không thực hiện quản lý theo vị trí, cùng chuỗi có tính phổ thông cao và nghề mới, thì có thể tách việc xét và việc tuyển. Nhân tài chuyên môn chuyển vùng, chuyển đơn vị thì chức danh được xét lại hoặc xác nhận theo thẩm quyền quản lý, trừ khi nhà nước có quy định khác (toàn quốc) +- Chi phí: Không tốn tiền. Chỉ cần hỏi một lần bộ phận nhân sự của đơn vị. +- Nói thẳng: Chức danh nghề nghiệp là căn cứ để sử dụng, đánh giá, thăng tiến, nhưng được xét và được sử dụng là hai việc khác nhau. Cơ quan, đơn vị thường xét trong tỷ lệ vị trí, được xét thì phần lớn được sử dụng. Đơn vị không thực hiện quản lý theo vị trí thì có thể tách việc xét và việc sử dụng, được xét thì đơn vị không nhất định cho đúng vị trí và mức lương tương ứng. Đổi đơn vị, đổi địa phương thì chức danh nghề nghiệp có thể phải xét lại hoặc xác nhận. +- Lợi ích: Điều 2 của Lệnh số 40 của Bộ Nhân sự và Xã hội: kết quả xét đánh giá chức danh nghề nghiệp là căn cứ quan trọng để sử dụng, đánh giá, thăng tiến nhân tài chuyên môn. Ý kiến cải cách năm 2016 quy định, cơ quan, đơn vị thực hiện đầy đủ quản lý theo vị trí thì thường xét chức danh nghề nghiệp trong tỷ lệ cơ cấu vị trí. Đơn vị không thực hiện quản lý theo vị trí, cùng chuyên môn phổ biến rộng và nghề mới, thì có thể tách việc xét và việc sử dụng. Khi nhân tài chuyên môn chuyển vùng, chuyển đơn vị thì chức danh nghề nghiệp được xét lại hoặc xác nhận theo thẩm quyền quản lý, trừ trường hợp nhà nước có quy định khác (toàn quốc) - Mức bằng chứng: B -- Nguồn: Mệnh lệnh số 40 của Bộ Nhân sự và Xã hội (人力资源和社会保障部令第 40 号) (2019). 职称评审管理暂行规定. ：第二条、第二十九条；Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「对于全面实行岗位管理、专业技术人才学术技术水平与岗位职责密切相关的事业单位，一般应在岗位结构比例内开展职称评审。对于不实行岗位管理的单位，以及通用性强、广泛分布在各社会组织的职称系列和新兴职业，可采用评聘分开方式。」 -- Ghi chú: Để ở mức B là vì văn bản chỉ viết cách nối việc xét với việc tuyển, không có con số thống nhất cả nước cho「xét được rồi tăng bao nhiêu». Phần làm theo được là: ở đơn vị sự nghiệp và doanh nghiệp nhà nước, hỏi bộ phận nhân sự về số vị trí trống và quy tắc tuyển của đơn vị. ở doanh nghiệp tư nhân, hỏi người chủ xem họ có công nhận tấm bằng đó không, có tăng tiền không. \ No newline at end of file +- Nguồn: Lệnh số 40 của Bộ Nhân sự và Xã hội (2019). 职称评审管理暂行规定. ：第二条、第二十九条；Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「对于全面实行岗位管理、专业技术人才学术技术水平与岗位职责密切相关的事业单位，一般应在岗位结构比例内开展职称评审。对于不实行岗位管理的单位，以及通用性强、广泛分布在各社会组织的职称系列和新兴职业，可采用评聘分开方式。」 +- Ghi chú: Mức bằng chứng đặt là B vì văn bản chỉ nói cách nối việc xét với việc sử dụng, không có con số thống nhất toàn quốc cho「được xét thì tăng bao nhiêu」. Phần làm theo được là: ở cơ quan, đơn vị và doanh nghiệp nhà nước, trước hết hỏi bộ phận nhân sự về số vị trí trống và quy tắc sử dụng của đơn vị mình. Ở doanh nghiệp tư nhân, trước hết hỏi người chủ xem họ có công nhận chứng chỉ này không, và có tăng lương không. \ No newline at end of file diff --git "a/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" "b/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" new file mode 100644 index 00000000..c9d35181 --- /dev/null +++ "b/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" @@ -0,0 +1,106 @@ +[← Về mục lục](../README.md) + +# 26. Làm một trang web hoặc nền tảng: giấy phép, đăng ký, máy chủ + +Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả lời bốn câu hỏi: về mặt pháp lý nó là loại kinh doanh nào, cần giấy phép nào, máy chủ đặt ở đâu, và ai chịu trách nhiệm cho những gì người dùng đăng lên cùng khoản tiền thu vào. Thước đo là tự do cá nhân và tiền. Thứ tự các mục là: trước hết là mục có thể khiến bạn vào tù, tiếp theo là mục phải xin giấy phép trước khi bắt đầu, rồi đến những việc nền tảng phải làm hằng ngày, cuối cùng là kinh nghiệm chọn máy chủ, mục đó chỉ đưa ra được phán đoán chứ không đưa ra được con số. Nếu bạn là kỹ sư được thuê, các vạch đỏ nằm ở phần 11. Công ty đăng ký thế nào và kê thuế ra sao nằm ở phần 12, phần này không lặp lại. Bảng đối chiếu các loại giấy phép và bảng quyết định chọn máy chủ nằm ở [docs/做平台要办哪些证.md](../docs/做平台要办哪些证.md). + +### 1. Để tiền không đi vào tài khoản của nền tảng rồi mới chuyển cho người bán, hãy cho tiền đi thẳng qua tổ chức thanh toán được cấp phép + +- Chi phí: Không tốn tiền. Mất thời gian kết nối sản phẩm của một tổ chức thanh toán được cấp phép. Tổ chức thanh toán được cấp phép là công ty đã có giấy phép thanh toán của Ngân hàng Nhân dân Trung Quốc. Dùng chức năng chia tiền hoặc giao dịch có bảo lãnh của họ, tiền của người mua chuyển thẳng cho người bán, không đi vào tài khoản của chính nền tảng +- Nói thẳng: Tiền của người mua vào tài khoản của bạn trước, rồi bạn chuyển tiếp cho người bán, vậy là bạn đang làm ngành thanh toán. Làm ngành này phải được Ngân hàng Nhân dân Trung Quốc phê duyệt, và phải có thật một trăm triệu tệ vốn điều lệ thực góp. Không có phê duyệt mà làm thì bị đóng cửa, tiền kiếm được bị thu hồi, rồi còn bị phạt tiền; chủ doanh nghiệp và người trực tiếp xử lý bị phạt riêng. Nặng hơn thì còn bị kết án. Cách an toàn chỉ có một: tiền không đi vào tài khoản của bạn. +- Lợi ích: Ngành thanh toán mà quy chế nói tới là「根据收款人或者付款人提交的电子支付指令转移货币资金」— chuyển tiền theo lệnh thanh toán điện tử do người nhận tiền hoặc người trả tiền nộp lên. Nói ngắn gọn là thay người khác chuyển tiền theo mệnh lệnh của họ. Muốn làm ngành này thì「应当经中国人民银行批准，取得支付业务许可」— phải được Ngân hàng Nhân dân Trung Quốc phê duyệt và có giấy phép kinh doanh thanh toán. Vốn đăng ký thấp nhất là 100 triệu tệ, và phải là vốn tiền mặt thực góp; 100 triệu tệ đó phải đưa thật vào tài khoản công ty.「未经依法批准，擅自设立非银行支付机构、从事或者变相从事支付业务的，由中国人民银行依法予以取缔，没收违法所得」— ai không được phê duyệt mà tự lập tổ chức thanh toán phi ngân hàng, hoặc làm hay ngụy trang làm ngành thanh toán, sẽ bị Ngân hàng Nhân dân Trung Quốc thu tịch và tịch thu thu nhập bất hợp pháp. Làm việc này mà không có phê duyệt thì bị đóng cửa ngay, tiền kiếm được bị thu hồi. Thu nhập bất hợp pháp từ 50 vạn tệ trở lên thì phạt thêm từ 1 đến 5 lần thu nhập bất hợp pháp. Không có thu nhập bất hợp pháp, hoặc dưới 50 vạn tệ, thì phạt 50 vạn đến 200 vạn tệ. Người đại diện theo pháp luật và người trực tiếp chịu trách nhiệm bị phạt riêng từ 10 vạn đến 50 vạn tệ. Bộ luật Hình sự còn coi「非法从事资金支付结算业务」— làm nghiệp vụ chi trả, thanh toán tiền không phép — là kinh doanh phi pháp. Nghiêm trọng thì phạt tù dưới 5 năm, đặc biệt nghiêm trọng thì phạt tù từ 5 năm trở lên. Điều gì được tính là「情节严重」, giải thích của Tòa án nhân dân tối cao (最高人民法院) và Viện kiểm sát nhân dân tối cao (最高人民检察院) đã vạch đường. Đường này là số tiền kinh doanh phi pháp từ 500 vạn tệ trở lên, hoặc thu nhập bất hợp pháp từ 10 vạn tệ trở lên (toàn quốc, từ tháng 5 năm 2024) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2023). 非银行支付机构监督管理条例（国务院令第 768 号，第二、六、八、二十七、四十七条）. ；Ngân hàng Nhân dân Trung Quốc (中国人民银行) (2024). 非银行支付机构监督管理条例实施细则（中国人民银行令〔2024〕第 4 号，第六十八条）. ；Quốc hội Nhân dân (全国人大) (1997). 刑法（第二百二十五条第三项）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2019). 关于办理非法从事资金支付结算业务、非法买卖外汇刑事案件适用法律若干问题的解释（第一、三、四条）. +- Ghi chú: Thiết kế dòng tiền để nó không đi vào tài khoản của bạn là cách an toàn duy nhất. Trên thị trường gọi kiểu thu hộ rồi chi hộ như vậy là「二清」, nhưng từ này không có trong bất kỳ văn bản chính thức nào. Quy chế thi hành dùng cách nói「变相从事支付业务」— làm ngành thanh toán theo hình thức ngụy trang, tức là không được phê duyệt mà vẫn thay người dùng chuyển tiền theo mệnh lệnh của họ. Giải thích của Tòa án nhân dân tối cao (最高人民法院) và Viện kiểm sát nhân dân tối cao (最高人民检察院) chỉ liệt kê ba trường hợp: dựng giao dịch giả để rút tiền mặt, dùng tài khoản của đơn vị để rút tiền mặt, dùng séc để rút tiền mặt. Nó không nêu đích danh việc nền tảng thu hộ rồi chi hộ, nên không thể nói rằng nền tảng thu hộ chi hộ「一定构成犯罪」 + +### 2. Trang web thu tiền thì cần giấy phép, bên trung gian kết nối giao dịch là loại giấy phép khác, còn trang web miễn phí chỉ cần đăng ký khai báo + +- Chi phí: Đăng ký khai báo không tốn tiền. Giấy phép kinh doanh dịch vụ viễn thông giá trị gia tăng (增值电信业务经营许可证) thì có tốn tiền, và nó quyết định「向用户收钱的网站能不能开」— một trang web có thu tiền người dùng có được mở hay không. Muốn có giấy phép đó thì phải có công ty trước. Chỉ hoạt động trong một tỉnh thì vốn đăng ký không thấp hơn 100 vạn tệ. Hoạt động xuyên tỉnh thì không thấp hơn 1000 vạn tệ. Thời gian thẩm định 60 ngày, giấy phép có hiệu lực 5 năm +- Nói thẳng: Trang web thu tiền người dùng thì phải xin giấy phép kinh doanh dịch vụ viễn thông giá trị gia tăng trước. Giấy phép này chỉ cấp cho công ty, cá nhân không xin được. Trang web không thu tiền chỉ cần đăng ký khai báo, và đăng ký khai báo không tốn tiền. Không có giấy phép mà mở trang web thu tiền thì bị thu hồi thu nhập rồi phạt tiền, nặng hơn thì đóng cửa trang web. Nền tảng của bạn có thuộc loại trung gian kết nối giao dịch hay không, mục này không kết luận thay bạn, lý do nằm ở phần ghi chú. +- Lợi ích: Nguyên văn quy chế là「国家对经营性互联网信息服务实行许可制度；对非经营性互联网信息服务实行备案制度。未取得许可或者未履行备案手续的，不得从事互联网信息服务」— nhà nước áp dụng chế độ cấp phép với dịch vụ thông tin Internet có tính thương mại, và chế độ đăng ký khai báo với dịch vụ không tính thương mại; chưa có giấy phép hoặc chưa làm thủ tục đăng ký khai báo thì không được làm dịch vụ thông tin Internet. Thu tiền thì phải có giấy phép trước, không thu tiền thì phải đăng ký khai báo trước, thiếu cả hai thì không được mở. Thế nào gọi là có tính thương mại? Quy chế nói là「通过互联网向上网用户有偿提供信息或者网页制作等服务活动」— cung cấp có trả tiền thông tin hoặc dịch vụ làm trang web cho người dùng qua Internet. Nói ngắn gọn là thu tiền người dùng. Loại trung gian kết nối giao dịch gọi là xử lý dữ liệu và xử lý giao dịch trực tuyến, trong ngành gọi tắt là EDI. Quy chế định nghĩa nó là「利用各种与公用通信网或互联网相连的数据与交易/事务处理应用平台，通过公用通信网或互联网为用户提供在线数据处理和交易/事务处理的业务」. Nó còn「包括交易处理业务、电子数据交换业务和网络/电子设备数据处理业务」. Nói thật là trang web làm thay người dùng xử lý dữ liệu trực tuyến, hoặc làm trung gian và xử lý giao dịch giữa người bán và người mua. Không có giấy phép mà kinh doanh, hoặc kinh doanh vượt phạm vi, đều bị tịch thu thu nhập bất hợp pháp rồi phạt từ 3 đến 5 lần thu nhập bất hợp pháp. Vượt phạm vi nghĩa là đã có giấy phép nhưng làm việc nằm ngoài phạm vi ghi trên giấy phép. Không có thu nhập bất hợp pháp, hoặc thu nhập bất hợp pháp dưới 5 vạn tệ, thì phạt 10 vạn đến 100 vạn tệ. Nghiêm trọng thì bị buộc đóng cửa trang web (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Hội đồng Nhà nước (国务院) (2000). 互联网信息服务管理办法（国务院令第 292 号，2024 年 12 月 6 日第二次修订，第三、四、七、十九条）. ；Hội đồng Nhà nước (国务院) (2024). 关于修改和废止部分行政法规的决定（国务院令第 797 号，第十八项）. ；Hội đồng Nhà nước (国务院) (2000). 电信条例（2016 年第二次修订，第七、九、十三、六十九条）. ；Bộ Công nghiệp và Thông tin (工业和信息化部) (2017). 电信业务经营许可管理办法（工信部令第 42 号，第六、九、十一、十四条）. +- Ghi chú: Việc bạn làm có thuộc loại EDI hay không thì tự đối chiếu với định nghĩa trong《电信业务分类目录》. Không có câu nào của cơ quan chính thức nói thẳng rằng「带撮合和资金处理的电商平台必须办 EDI 许可」. Hướng dẫn nghiệp vụ của Bộ Công nghiệp và Thông tin (工业和信息化部) chỉ nói「按照业务界定申请相应的电信业务经营许可」— xin giấy phép kinh doanh viễn thông tương ứng với cách định nghĩa ngành nghề. Nên có phải xin hay không thì chỉ tự suy ra từ định nghĩa. Còn một điểm: cá nhân không xin được giấy phép này, người xin phải là công ty được thành lập theo pháp luật + +### 3. Phát trực tiếp thì cần giấy phép kinh doanh văn hóa trên mạng, còn giấy phép chương trình nghe nhìn thì công ty tư nhân gần như không xin được + +- Chi phí: Giấy phép kinh doanh văn hóa trên mạng (网络文化经营许可证), gọi tắt là giấy phép văn hóa mạng, quyết định「能不能靠直播、游戏解说这类网络表演赚钱」— có được kiếm tiền từ biểu diễn trên mạng như phát trực tiếp hay thuyết minh trò chơi hay không. Cơ quan quản lý văn hóa cấp tỉnh phê duyệt, phải có kết quả trong 20 ngày kể từ khi nhận hồ sơ, giấy phép có hiệu lực 3 năm, phải đi gia hạn trước khi hết hạn 30 ngày. Giấy phép chương trình nghe nhìn (视听节目许可) quyết định「能不能在网上放节目」— có được phát chương trình trên mạng hay không, và nó đòi hỏi người xin là đơn vị sở hữu toàn phần của nhà nước hoặc đơn vị do nhà nước kiểm soát phần lớn +- Nói thẳng: Phát trực tiếp hay thuyết minh trò chơi thì cần giấy phép kinh doanh văn hóa trên mạng do cơ quan văn hóa cấp tỉnh cấp. Trên giấy phép, phạm vi kinh doanh phải có ghi biểu diễn trên mạng. Trong 20 ngày kể từ khi nhận hồ sơ sẽ có kết quả, giấy phép dùng 3 năm. Video đã phát thì bạn tự giữ đủ 60 ngày. Không có giấy phép mà mở, sẽ bị yêu cầu dừng, bị cảnh cáo, rồi phạt tiền dưới 3 vạn tệ; không dừng thì vào danh sách đen. +- Lợi ích: Đơn vị làm biểu diễn trên mạng phải xin giấy phép trước. Nguyên văn quy chế là「应当根据《互联网文化管理暂行规定》，向省级文化行政部门申请取得《网络文化经营许可证》，许可证的经营范围应当明确包括网络表演」. Chơi trò chơi cho người ta xem hoặc thuyết minh cũng bị quản lý theo bộ này, nguyên văn quy chế là「参照本办法进行管理」. Toàn bộ tư liệu video của các buổi biểu diễn trên mạng「保存时间不得少于60日」— thời gian lưu giữ không được ít hơn 60 ngày, nên video đã phát bạn phải tự giữ đủ 60 ngày. Không được phê duyệt mà làm hoạt động văn hóa Internet có tính thương mại thì bị yêu cầu dừng, bị cảnh cáo, rồi phạt tiền dưới 3 vạn tệ. Bị yêu cầu dừng mà vẫn không dừng thì bị đưa vào danh sách đen của thị trường văn hóa. Phát chương trình trên mạng là một giấy phép khác:「应当取得《信息网络传播视听节目许可证》」. Người xin phải「具备法人资格，为国有独资或国有控股单位」— có tư cách pháp nhân và là đơn vị sở hữu toàn phần của nhà nước hoặc do nhà nước kiểm soát phần lớn, tức là phải là đơn vị do nhà nước làm chủ sở hữu toàn phần hoặc phần lớn, công ty tư nhân về cơ bản không xin được. Không có giấy phép mà làm thì bị cảnh cáo, bị buộc sửa, có thể còn phạt tiền dưới 3 vạn tệ. Nghiêm trọng thì xử phạt theo Quy chế quản lý phát thanh truyền hình và truyền thanh (广播电视管理条例) (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Văn hóa (文化部) (2016). 网络表演经营活动管理办法（文市发〔2016〕33 号，第二、四、六、十三、二十条）. ；Bộ Văn hóa (文化部) (2011). 互联网文化管理暂行规定（文化部令第 51 号，2017 年经文化部令第 57 号修订，第七、八、九、十六、二十一、二十八条）. ；Cục Quản lý Phát thanh truyền hình (广电总局), Bộ Công nghiệp và Thông tin (信息产业部) (2007). 互联网视听节目服务管理规定（第 56 号令，2015 年修订，第二、七、八、二十四条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家网信办) và sáu cơ quan khác (2021). 关于加强网络直播规范管理工作的指导意见（国信办发文〔2021〕3 号，第 9 项）. +- Ghi chú: Ngành nghề nào hợp với giấy phép nào thì văn bản của bảy cơ quan đó viết rõ nhất. Nền tảng kiếm tiền bằng phát trực tiếp cần giấy phép văn hóa mạng cộng thêm đăng ký ICP, đăng ký ICP là thủ tục đăng ký trang web với Bộ Công nghiệp và Thông tin (工业和信息化部). Phát chương trình trên mạng thì cần giấy phép chương trình nghe nhìn, hoặc hoàn tất đăng ký. Đăng tin tức thì cần giấy phép dịch vụ thông tin tin tức Internet. Ngoài ra,《互联网文化管理暂行规定》mà trang chính phủ Trung Quốc đăng lại năm 2025 là bản gốc năm 2011 chưa sửa đổi, khi trích dẫn phải theo văn bản đang có hiệu lực của Bộ Văn hóa và Du lịch (文化和旅游部) + +### 4. Máy chủ đặt trong nước thì phải đăng ký khai báo, và nhà cung cấp dịch vụ truy cập phải tự có giấy phép viễn thông giá trị gia tăng + +- Chi phí: Đăng ký khai báo không tốn tiền. Hồ sơ nộp cùng máy chủ tại nhà cung cấp, thường phải chờ mấy ngày đến mấy tuần. Kiểm tra một lần nhà cung cấp đã đủ điều kiện hay chưa thì mười phút là đủ +- Nói thẳng: Máy chủ đặt trong nước thì trước khi mở trang web phải đăng ký khai báo. Đăng ký khai báo không tốn tiền, hồ sơ nộp cùng máy chủ tại nhà cung cấp, thường mấy ngày đến mấy tuần. Không đăng ký khai báo thì bị buộc sửa trong thời hạn rồi phạt tiền 1 vạn tệ, không sửa thì đóng cửa trang web. Không đặt số đăng ký trên trang chủ thì còn phạt thêm 5000 tệ đến 5 vạn tệ. +- Lợi ích: Nguyên văn quy chế là「未经备案，不得在中华人民共和国境内从事非经营性互联网信息服务」— chưa đăng ký khai báo thì không được làm dịch vụ thông tin Internet không tính thương mại bên trong lãnh thổ nước Cộng hòa Nhân dân Trung Quốc. Trang web không thu tiền người dùng, chỉ cần đặt ở trong nước, vẫn phải đăng ký khai báo trước. Không đăng ký khai báo thì bị buộc sửa trong thời hạn rồi phạt tiền 1 vạn tệ. Không chịu sửa thì đóng cửa trang web. Nhà cung cấp giúp bạn đưa máy chủ lên mạng cũng nằm trong phạm vi này: nó「不得为未经备案的组织或者个人从事非经营性互联网信息服务提供互联网接入服务」— không được cung cấp dịch vụ truy cập Internet cho tổ chức hoặc cá nhân chưa đăng ký khai báo mà làm dịch vụ thông tin Internet không tính thương mại. Nhà cung cấp thu tiền hộ bạn cũng vậy. Quy chế nói nhà cung cấp loại này「不得为未依法取得经营许可证或者履行非经营性互联网信息服务备案手续的单位或者个人提供接入或者代收费等服务」— nghĩa là nhà cung cấp đàng hoàng không được nhận trang web vừa không có giấy phép vừa chưa đăng ký khai báo. Ngoài ra, hosting máy chủ và máy chủ ảo trong quy chế gọi là ngành trung tâm dữ liệu Internet, còn truy cập băng rộng gọi là ngành dịch vụ truy cập Internet, bản thân hai ngành này cũng phải có giấy phép, nên nhà cung cấp phải có giấy phép trước. Trang web không đặt số đăng ký trên trang chủ thì bị buộc sửa rồi phạt tiền 5000 tệ đến 5 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Công nghiệp và Thông tin (信息产业部) (2005). 非经营性互联网信息服务备案管理办法（信产部令第 33 号，2024 年经工信部令第 68 号修订，第五、十三、十八、二十二条）. ；Bộ Công nghiệp và Thông tin (工业和信息化部) (2017). 电信业务经营许可管理办法（工信部令第 42 号，第二十、二十四条）. ；Bộ Công nghiệp và Thông tin (工业和信息化部) (2015). 电信业务分类目录（2015 年版，B11、B14、B21、B25）. +- Ghi chú: Trước khi bàn giá, kiểm tra xem nhà cung cấp đã có giấy phép chưa. Tra giấy phép thì dùng Hệ thống thông tin quản lý tổng hợp thị trường kinh doanh viễn thông của Bộ Công nghiệp và Thông tin, tsm.miit.gov.cn, tìm theo tên đầy đủ của công ty hoặc theo số giấy phép. Tra đăng ký khai báo trang web thì dùng beian.miit.gov.cn. Một「机房」không có giấy phép tương ứng thì đến ngày xảy ra sự cố bạn cũng không biết kêu ai. Đặt máy ở nước ngoài thì đúng là tránh được đăng ký khai báo. Nhưng khi người dùng ở trong nước và tiền ở trong nước thì nghĩa vụ nào ghi ở mục 5 đến 10 của phần này cũng không bớt, lại cộng thêm một tầng chi phí chuyển dữ liệu ra nước ngoài +### 5. Cho người dùng lên bán hàng thì nền tảng phải kiểm tra đăng ký, báo thông tin và lưu giữ ba năm + +- Chi phí: Không tốn tiền. Phải làm thủ tục kiểm tra tư cách người bán trong hệ thống nền trả, còn phải lập một hồ sơ đăng ký. Mỗi quý báo một lần thông tin liên quan thuế, dữ liệu giao dịch giữ ba năm. Chỗ khó là đây là việc lặp lại hằng năm, không phải làm một lần là xong. +- Nói thẳng: Cho người khác lên nền tảng của bạn bán hàng thì trước hết bạn phải kiểm tra danh tính, địa chỉ, thông tin liên lạc và giấy phép của người bán, rồi đăng ký và lập hồ sơ. Lập hồ sơ xong thì ít nhất sáu tháng sau phải quay lại kiểm tra một lần. Người bán là ai thì báo cho cơ quan quản lý thị trường, tình hình kê thuế thì mỗi quý báo cho cơ quan thuế. Thông tin hàng hóa và bản ghi giao dịch giữ ba năm. Bỏ sót một việc thì trước hết bị buộc sửa, không sửa thì phạt tiền 2 vạn đến 10 vạn tệ. +- Lợi ích: Luật có một đoạn yêu cầu với nền tảng. Nguyên văn là「应当要求申请进入平台销售商品或者提供服务的经营者提交其身份、地址、联系方式、行政许可等真实信息，进行核验、登记，建立登记档案，并定期核验更新」— phải yêu cầu người kinh doanh xin vào nền tảng để bán hàng hoặc cung cấp dịch vụ nộp đúng thông tin thật gồm danh tính, địa chỉ, thông tin liên lạc và giấy phép hành chính, rồi kiểm tra, đăng ký, lập hồ sơ đăng ký, và định kỳ kiểm tra cập nhật lại. Trước khi người bán vào bán, anh ta phải nộp danh tính, địa chỉ, thông tin liên lạc và giấy phép mà anh ta phải có. Bạn kiểm tra một lượt, đăng ký và lập hồ sơ, sau đó còn phải quay lại kiểm tra sau một thời gian.「定期」là bao lâu? Quy chế của cơ quan có quyền nói rõ hơn:「至少每六个月核验更新一次」— ít nhất sáu tháng kiểm tra cập nhật một lần. Bạn còn phải báo thông tin danh tính của người kinh doanh trong nền tảng cho cơ quan quản lý thị trường. Phía thuế cũng phải báo, nguyên văn là「依照税收征收管理法律、行政法规的规定，向税务部门报送平台内经营者的身份信息和与纳税有关的信息」— theo luật và pháp lệnh về quản lý thuế, báo thông tin danh tính của người kinh doanh trong nền tảng và thông tin liên quan thuế cho cơ quan thuế. Nên người bán là ai thì báo cho cơ quan quản lý thị trường, còn danh tính và tình hình kê thuế của người bán thì báo cho cơ quan thuế. Thông tin hàng hóa, dịch vụ và giao dịch「保存时间自交易完成之日起不少于三年」— thời gian lưu giữ tính từ ngày giao dịch hoàn tất, không được ít hơn ba năm, tức là tính từ ngày một giao dịch xong là dữ liệu phải lưu ít nhất ba năm. Bất kỳ việc nào ở trên mà không làm thì bị buộc sửa trong thời hạn. Quá hạn không sửa thì phạt tiền 2 vạn đến 10 vạn tệ. Nghiêm trọng thì bị buộc đình chỉ kinh doanh để chỉnh phục, rồi phạt tiền 10 vạn đến 50 vạn tệ. Thông tin liên quan thuế phải「于季度终了的次月内」báo, tức là phải báo xong trong tháng kế tiếp sau khi kết thúc quý. Không báo đúng hạn thì mức phạt y hệt (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 电子商务法（第十、二十七、二十八、三十一、八十条）. ；Cục Quản lý Thị trường Quốc gia (市场监管总局) (2021). 网络交易监督管理办法（市场监管总局令第 37 号，第八、二十四、二十五、三十一条）. ；Hội đồng Nhà nước (国务院) (2025). 互联网平台企业涉税信息报送规定（国务院令第 810 号，第二、三、四、十条）. ；Cục Quản lý Thuế Nhà nước (国家税务总局) (2025). 关于互联网平台企业报送涉税信息有关事项的公告（税务总局公告 2025 年第 15 号）. +- Ghi chú: Người bán cũng phải tự đi đăng ký chủ thể kinh doanh. Luật liệt kê vài trường hợp không phải đăng ký:「个人销售自产农副产品、家庭手工业产品，个人利用自己的技能从事依法无须取得许可的便民劳务活动和零星小额交易活动」.「零星小额」có một con số cụ thể, Điều 8 của Quy định quản lý và giám sát giao dịch trên mạng ấn định là giá trị giao dịch trong một năm là 10 vạn tệ. Phía nền tảng còn phải làm hai việc: đánh dấu riêng người bán đã đăng ký và chưa đăng ký, và không cho xóa đánh giá của người tiêu dùng + +### 6. Nội dung người dùng đăng lên là bạn phải quản: có cơ chế duyệt, có cửa sổ báo việc, thấy nội dung vi phạm thì dừng truyền ngay và báo cáo + +- Chi phí: Phải tốn tiền thuê người duyệt nội dung, đây là khoản chi lớn nhất và lâu dài nhất của nền tảng. Cửa sổ báo việc và bản ghi của mỗi lần xử lý đều phải để lại dấu vết. Chỗ khó là ngày nào cũng phải có người trực, ngừng một ngày là có thể xảy ra sự cố. +- Nói thẳng: Người dùng đăng cái gì, xảy ra chuyện thì nền tảng chịu trách nhiệm. Bạn phải dựng được bộ quy chế này: đăng ký, duyệt bài đăng, duyệt bình luận, tuần tra hằng ngày, xử lý khi có sự cố. Đặt cửa sổ báo việc ở chỗ dễ thấy trên trang. Thấy thông tin vi phạm thì dừng truyền ngay, lưu bản ghi đầy đủ, báo cơ quan có thẩm quyền. Không làm thì phạt tiền 5 vạn đến 50 vạn tệ, nghiêm trọng hơn thì phạt nặng hơn. +- Lợi ích: Quy chế nói nền tảng「应当履行信息内容管理主体责任」— phải chịu trách nhiệm chính về quản lý nội dung thông tin, nội dung người dùng đăng lên do nền tảng tự chịu trách nhiệm. Cụ thể phải「健全用户注册、账号管理、信息发布审核、跟帖评论审核、版面页面生态管理、实时巡查、应急处置和网络谣言、黑色产业链信息处置等制度」— xây dựng và hoàn thiện các quy chế về đăng ký người dùng, quản lý tài khoản, duyệt thông tin đăng lên, duyệt bình luận và bài trả lời, quản lý môi trường trang, tuần tra theo thời gian thực, xử lý khẩn cấp, và xử lý thông tin tin đồn trên mạng, chuỗi sản xuất đen. Từ đăng ký, đăng bài, bình luận, đến tuần tra hằng ngày và xử lý sự cố, mỗi khâu phải có quy tắc viết ra. Ngoài ra còn phải「应当在显著位置设置便捷的投诉举报入口」— đặt cửa sổ khiếu nại và báo việc thuận tiện ở vị trí dễ thấy, nút báo việc phải để ở chỗ người dùng nhìn thấy và bấm được ngay. Thấy thông tin vi phạm thì「应当依法立即采取处置措施，保存有关记录，并向有关主管部门报告」— phải xử lý ngay theo pháp luật, lưu các bản ghi liên quan và báo cơ quan có thẩm quyền. Luật An ninh mạng nói cụ thể hơn. Nguyên văn là「发现法律、行政法规禁止发布或者传输的信息的，应当立即停止传输该信息，采取消除等处置措施，防止信息扩散，保存有关记录，并向有关主管部门报告」— thấy thông tin mà luật và pháp lệnh hành chính cấm đăng hoặc truyền thì phải dừng truyền ngay, thực hiện biện pháp xử lý như xóa bỏ để ngăn thông tin lan rộng, lưu các bản ghi liên quan và báo cơ quan có thẩm quyền. Không làm thì bị buộc sửa, bị cảnh cáo, bị công bố, có thể còn phạt tiền 5 vạn đến 50 vạn tệ. Không chịu sửa, hoặc tình tiết nghiêm trọng, thì phạt tiền 50 vạn đến 200 vạn tệ, còn có thể bị buộc đình chỉ kinh doanh để chỉnh phục, đóng cửa trang web hoặc ứng dụng, thu hồi giấy phép hoặc giấy chứng nhận đăng ký kinh doanh. Người trực tiếp chịu trách nhiệm bị phạt riêng 5 vạn đến 20 vạn tệ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2019). 网络信息内容生态治理规定（网信办令第 5 号，第八、九、十、十六、三十七条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第四十九、六十九条）. ；Hội đồng Nhà nước (国务院) (2000). 互联网信息服务管理办法（第十五、十六、二十三条）. +- Ghi chú: Quy tắc duyệt không cần tự nghĩ, cứ chép theo. Chín loại nội dung không được đăng, Điều 15 của Quy định quản lý dịch vụ thông tin Internet liệt kê rõ mồn một. Người phát trực tiếp còn phải làm thêm ba việc: phải「建立直播内容审核平台」— dựng nền tảng duyệt nội dung phát trực tiếp; phải cắt được một buổi đang phát bất cứ lúc nào, điều luật gọi việc đó là「即时阻断」; và phải lưu nội dung phát trực tiếp cùng nhật ký trong 60 ngày. Tư liệu video của biểu diễn trên mạng cũng không được ít hơn 60 ngày + +### 7. Cung cấp dịch vụ đăng thông tin và nhắn tin tức thì phải yêu cầu người dùng cung cấp thông tin danh tính thật + +- Chi phí: Phải kết nối một kênh xác minh tên thật, kênh loại này tính phí theo lượt. Sản phẩm nên thiết kế theo nguyên tắc「后台实名、前台自愿」— phía sau lưu tên thật, trên trang chỉ hiện biệt danh. +- Nói thẳng: Trang web của bạn cho đăng bài hoặc cho nhắn tin thì khi người dùng mở dịch vụ, bạn phải lấy thông tin danh tính thật của họ. Người không đưa thì không cho dùng. Cách làm là phía sau lưu tên thật, còn biệt danh hiện thế nào thì để họ tự quyết, xác minh bằng số điện thoại di động là được. Không làm tên thật thì trước hết bị buộc sửa. Không sửa thì phạt tiền 5 vạn đến 50 vạn tệ, còn có thể đóng cửa trang web. +- Lợi ích: Nguyên văn luật là「网络运营者为用户办理网络接入、域名注册服务，办理固定电话、移动电话等入网手续，或者为用户提供信息发布、即时通讯等服务，在与用户签订协议或者确认提供服务时，应当要求用户提供真实身份信息。用户不提供真实身份信息的，网络运营者不得为其提供相关服务」— khi bên vận hành mạng làm cho người dùng các dịch vụ kết nối mạng, đăng ký tên miền, làm thủ tục kết nối đường dây cố định, điện thoại di động, hoặc cung cấp dịch vụ đăng thông tin, nhắn tin tức thì khi ký hợp đồng hoặc xác nhận cung cấp dịch vụ, phải yêu cầu người dùng cung cấp thông tin danh tính thật. Người dùng không cung cấp thông tin danh tính thật thì bên vận hành mạng không được cung cấp dịch vụ liên quan. Không đòi tên thật, hoặc người dùng không đưa thông tin thật mà bạn vẫn cho dùng, thì bị buộc sửa. Không chịu sửa, hoặc tình tiết nghiêm trọng, thì phạt tiền 5 vạn đến 50 vạn tệ, còn có thể bị buộc dừng các ngành nghề liên quan, đình chỉ kinh doanh để chỉnh phục, đóng cửa trang web hoặc ứng dụng, thu hồi giấy phép hoặc giấy chứng nhận đăng ký kinh doanh. Người trực tiếp chịu trách nhiệm bị phạt riêng 1 vạn đến 10 vạn tệ. Dịch vụ phát trực tiếp theo nguyên tắc「后台实名、前台自愿」, thực hiện「基于移动电话号码等方式的真实身份信息认证」— xác thực thông tin danh tính thật dựa trên số điện thoại di động và những cách tương tự. Nghĩa là phía sau nền tảng lưu tên thật của người dùng, còn biệt danh hiện thế nào thì người dùng tự quyết, xác minh dựa vào số điện thoại di động hay cách tương tự. Bán hàng qua phát trực tiếp thì yêu cầu khắt hơn, là「基于身份证件信息、统一社会信用代码等真实身份信息认证」— xác thực thông tin danh tính thật dựa trên thông tin giấy tờ tùy thân, mã tín dụng xã hội thống nhất. Mã tín dụng xã hội thống nhất chính là dãy số trên giấy chứng nhận đăng ký kinh doanh (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十六、六十四条；2016 年文本为第二十四、六十一条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2016). 互联网直播服务管理规定（第十二、十六条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家网信办) và sáu cơ quan khác (2021). 网络直播营销管理办法（试行）（国信办发文〔2021〕5 号，第八条）. +- Ghi chú: Số điều đã đổi chỗ. Luật An ninh mạng sửa đổi năm 2025, có hiệu lực từ ngày 1 tháng 1 năm 2026. Điều về tên thật chuyển từ Điều 24 sang Điều 26, điều về xử phạt chuyển từ Điều 61 sang Điều 64. Văn bản của điều không đổi. Tài liệu bạn đang cầm nếu vẫn dẫn số điều cũ thì phải sửa lại + +### 8. Không mở phát trực tiếp cho người chưa đủ 16 tuổi, tiền thưởng xử lý theo bậc tuổi + +- Chi phí: Không tốn tiền. Thêm một bước vào mỗi khâu đăng ký và thanh toán: trước hết xác định tuổi, sau đó xác nhận người giám hộ đã đồng ý. +- Nói thẳng: Người chưa đủ 16 tuổi không được đăng ký tài khoản phát trực tiếp ở chỗ bạn. Đủ 16 tuổi thì phải kiểm tra danh tính, và phải có cha mẹ hoặc người giám hộ khác đồng ý. Tiền thưởng chia ba bậc: dưới 8 tuổi không được thưởng tiền; từ 8 đến 16 tuổi phải có người giám hộ đồng ý; trên 16 tuổi phải có người giám hộ đồng ý hoặc phải kiểm tra giấy tờ chứng minh thu nhập. Trẻ tự thưởng tiền sau lưng người giám hộ thì khi người giám hộ đòi, tòa sẽ xử trả lại. +- Lợi ích: Nguyên văn luật là「网络直播服务提供者不得为未满十六周岁的未成年人提供网络直播发布者账号注册服务；为年满十六周岁的未成年人提供网络直播发布者账号注册服务时，应当对其身份信息进行认证，并征得其父母或者其他监护人同意」— bên cung cấp dịch vụ phát trực tiếp không được cấp dịch vụ đăng ký tài khoản người phát trực tiếp cho người chưa thành niên chưa đủ mười sáu tuổi; khi cấp cho người chưa thành niên đã đủ mười sáu tuổi thì phải xác thực thông tin danh tính của họ và lấy sự đồng ý của cha mẹ hoặc người giám hộ khác. Nghị định còn yêu cầu「建立网络直播发布者真实身份信息动态核验机制」— lập cơ chế kiểm tra động thông tin danh tính thật của người phát trực tiếp, nghĩa là bạn phải kiểm tra lại được danh tính của người phát bất cứ lúc nào. Ngoài ra phải「合理限制不同年龄阶段未成年人在使用其服务中的单次消费数额和单日累计消费数额」— giới hạn hợp lý số tiền tiêu trong một lần và tổng số tiền tiêu trong một ngày của người chưa thành niên ở từng độ tuổi, tức là đặt trần chi tiêu cho từng giao dịch và từng ngày của người chưa thành niên theo tuổi. Tiền thưởng hiện chia ba bậc:「不得向八周岁以下的未成年人提供打赏服务。向八周岁至十六周岁的未成年人提供打赏服务，应征得其监护人同意。向十六周岁以上的未成年人提供打赏服务，应征得其监护人同意或核验其收入证明材料」— không được cung cấp dịch vụ thưởng tiền cho người chưa thành niên dưới tám tuổi; cung cấp dịch vụ thưởng tiền cho người chưa thành niên từ tám đến mười sáu tuổi thì phải lấy sự đồng ý của người giám hộ; cung cấp dịch vụ thưởng tiền cho người chưa thành niên trên mười sáu tuổi thì phải lấy sự đồng ý của người giám hộ hoặc kiểm tra giấy tờ chứng minh thu nhập của họ. Còn một điều nữa: người chưa thành niên bị hạn chế năng lực dân sự mà tự thưởng tiền khi chưa được người giám hộ đồng ý thì khi người giám hộ đòi tiền, tòa phải có quyết định cho đòi, tiền lấy lại được. Người bị hạn chế năng lực dân sự là người chưa thành niên mà theo pháp luật còn chưa tự quyết hết được cho mình (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 未成年人保护法（第七十四、七十六条）. ；Hội đồng Nhà nước (国务院) (2023). 未成年人网络保护条例（国务院令第 766 号，第三十一、四十四条）. ；Văn phòng Tổng thư ký Cục Quản lý Không gian mạng Trung Quốc (中央网信办秘书局) (2026). 关于加强网络直播打赏规范管理的通知（第 7 项）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). 关于依法妥善审理涉新冠肺炎疫情民事案件若干问题的指导意见（二）（法发〔2020〕17 号，第九条）. +- Ghi chú: Thực sự có tranh chấp hoàn tiền thì xử theo hướng có lợi nhất cho người chưa thành niên. Cách nói về việc này đã thay đổi. Ý kiến của bốn cơ quan năm 2022 và thông báo của Cục Quản lý Phát thanh truyền hình năm 2020 đều nói người chưa thành niên tuyệt đối không được thưởng tiền. Thông báo mới tháng 4 năm 2026 đổi thành chia bậc theo tám tuổi, mười sáu tuổi, nhưng không ghi rõ hai văn bản trước hết hiệu, cơ quan chính thức cũng không nói hai bên liên hệ với nhau thế nào. Nên câu「未成年人一律禁止打赏」bây giờ không còn chính xác +### 9. Nhận được thông báo vi phạm quyền thì phải xử lý kịp thời, chuyển bản giải trình mà sau 15 ngày không có động tĩnh thì khôi phục lại + +- Chi phí: Không tốn tiền. Phải dựng một quy trình cố định: nhận thông báo, chuyển cho người bán, nhận giải trình của người bán, công bố kết quả ra công khai. Mỗi bước đều lưu hồ sơ. +- Nói thẳng: Có người gửi thông báo vi phạm kèm bằng chứng ban đầu thì bạn phải kịp thời xóa nội dung, chặn hoặc gỡ liên kết, đồng thời chuyển thông báo cho người bán. Kéo dài không xử lý thì phần thiệt hại phát sinh thêm bạn phải cùng người bán chịu. Người bán giải trình thì bạn chuyển giải trình sang phía bên kia và chờ 15 ngày. Bên kia không khiếu nại cũng không khởi kiện thì bạn khôi phục nội dung trở lại. +- Lợi ích: Chủ sở hữu quyền「有权通知电子商务平台经营者采取删除、屏蔽、断开链接、终止交易和服务等必要措施。通知应当包括构成侵权的初步证据」— có quyền thông báo cho bên vận hành nền tảng thương mại điện tử thực hiện các biện pháp cần thiết như xóa, chặn, gỡ liên kết, chấm dứt giao dịch và chấm dứt dịch vụ. Thông báo phải kèm bằng chứng ban đầu chứng minh có vi phạm. Chủ sở hữu quyền là người nắm các quyền như nhãn hiệu, bản quyền, bằng sở hữu trí tuệ. Chỉ gửi một câu thì không tính, phải kèm bằng chứng ban đầu. Nền tảng nhận được thông báo thì「应当及时采取必要措施，并将该通知转送平台内经营者；未及时采取必要措施的，对损害的扩大部分与平台内经营者承担连带责任」— phải kịp thời thực hiện các biện pháp cần thiết và chuyển thông báo đó cho người kinh doanh trong nền tảng; không kịp thời thực hiện thì chịu trách nhiệm liên đới với người kinh doanh trong nền tảng cho phần thiệt hại bị mở rộng thêm. Kéo dài không xử lý thì phần thiệt hại phát sinh thêm bạn phải cùng người bán chịu. Sau khi người bán giải trình, nền tảng「在转送声明到达知识产权权利人后十五日内，未收到权利人已经投诉或者起诉通知的，应当及时终止所采取的措施」— trong 15 ngày kể từ khi bản giải trình đã tới được chủ sở hữu quyền, nếu không nhận được thông báo rằng chủ sở hữu quyền đã khiếu nại hoặc đã khởi kiện thì phải kịp thời chấm dứt các biện pháp đã áp dụng. Nghĩa là chuyển giải trình của người bán sang chủ sở hữu quyền rồi chờ 15 ngày. Bên kia không khiếu nại cũng không khởi kiện thì khôi phục lại những gì đã xóa hoặc gỡ xuống. Nền tảng biết hoặc phải biết có vi phạm mà không xử lý thì chịu trách nhiệm liên đới với bên vi phạm. Cơ quan quản lý sở hữu trí tuệ còn có thể buộc bạn sửa trong thời hạn, quá hạn không sửa thì phạt tiền 5 vạn đến 50 vạn tệ, nghiêm trọng thì phạt 50 vạn đến 200 vạn tệ. Ngược lại,「恶意发出错误通知，造成平台内经营者损失的，加倍承担赔偿责任」— cố ý gửi thông báo sai làm người kinh doanh trong nền tảng chịu thiệt hại thì chịu trách nhiệm bồi thường gấp đôi (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 电子商务法（第四十二、四十三、四十四、四十五、八十四条）. ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第一千一百九十五、一千一百九十六、一千一百九十七条）. +- Ghi chú: Bạn không phải nền tảng thương mại điện tử mà chỉ là một trang web thường thì áp dụng quy tắc thông báo và xóa của Bộ luật Dân sự. Khác biệt nằm ở chỗ sau khi bên kia giải trình thì bạn phải chờ bao lâu: Luật Thương mại điện tử có con số rõ là 15 ngày, còn Bộ luật Dân sự chỉ viết「合理期限」— một thời hạn hợp lý. Ngoài ra, việc công bố ra công khai thông báo đã nhận, giải trình của người bán và kết quả xử lý của bạn là yêu cầu rõ ràng trong Điều 44 của Luật Thương mại điện tử + +### 10. Đừng tùy tiện đặt thông tin người dùng ra nước ngoài, việc chuyển dữ liệu ra nước ngoài có điều kiện luật định và có ngưỡng số người + +- Chi phí: Không tốn tiền. Cách ít phiền nhất là giữ dữ liệu người dùng trong nước. Đúng là phải chuyển ra nước ngoài thì đi theo hợp đồng mẫu hoặc làm chứng nhận. +- Nói thẳng: Chuyển thông tin của người dùng trong nước sang máy ở nước ngoài thì đó gọi là chuyển ra nước ngoài. Máy chủ đặt ở nước ngoài không có nghĩa dữ liệu không ra nước ngoài. Muốn chuyển ra thì phải đi một trong bốn con đường: đánh giá an toàn của cơ quan quản lý không gian mạng, chứng nhận của tổ chức chuyên môn, ký hợp đồng theo hợp đồng mẫu, hoặc những gì luật quy định khác. Còn phải báo cho người dùng biết bên nhận ở nước ngoài là ai, dùng để làm gì, và phải hỏi riêng người dùng về chuyện này một lần. +- Lợi ích: Đưa thông tin của người dùng trong nước cho công ty ở nước ngoài, hoặc lưu trên máy ở nước ngoài, thì theo luật gọi là cung cấp thông tin cá nhân ra nước ngoài, và phải「应当具备」một trong bốn điều kiện. Một là đánh giá an toàn do cơ quan quản lý không gian mạng tổ chức. Hai là chứng nhận bảo vệ thông tin cá nhân do tổ chức chuyên môn thực hiện. Ba là ký hợp đồng với bên nhận ở nước ngoài theo hợp đồng mẫu do cơ quan quản lý không gian mạng ban hành. Bốn là các điều kiện khác theo luật và pháp lệnh hành chính. Đồng thời bạn phải báo cho người dùng mấy thứ sau: tên của bên nhận ở nước ngoài, thông tin liên lạc, mục đích xử lý, cách thức xử lý, các loại thông tin cá nhân, và cách người dùng thực hiện quyền của chính mình. Và phải「取得个人的单独同意」— có được sự đồng ý riêng của cá nhân, tức là hỏi riêng người dùng về chuyện này một lần. Ngưỡng cụ thể chia theo tổng số người tích luỹ. Doanh nghiệp bình thường trong quy chế gọi là bên xử lý dữ liệu khác với bên vận hành hạ tầng thông tin trọng yếu. Loại doanh nghiệp này có một đường ở dưới thì không phải làm gì, nguyên văn là「自当年1月1日起累计向境外提供不满10万人个人信息（不含敏感个人信息）的」— tính từ ngày 1 tháng 1 năm đó trở đi, đã cung cấp ra nước ngoài thông tin cá nhân của dưới 100.000 người, không gồm thông tin cá nhân nhạy cảm. Không tới được đường đó thì cả ba việc là khai báo để đánh giá, ký hợp đồng mẫu và làm chứng nhận đều miễn. Từ 100.000 người trở lên mà chưa tới 1.000.000 người thì phải ký hợp đồng mẫu hoặc vượt qua chứng nhận. Từ 1.000.000 người trở lên, hoặc thông tin cá nhân nhạy cảm đạt từ 10.000 người trở lên, thì phải khai báo để đánh giá an toàn (toàn quốc, từ tháng 3 năm 2024) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). 个人信息保护法（第三十八、三十九、四十条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2024). 促进和规范数据跨境流动规定（网信办令第 16 号，第三、四、五、七、八条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2022). 数据出境安全评估办法（网信办令第 11 号，第四条）. +- Ghi chú: Máy chủ đặt ở nước ngoài không đồng nghĩa với「数据没出境」. Người dùng đăng ký trong nước, bạn chuyển thông tin của họ sang máy ở nước ngoài, đó là chuyển ra nước ngoài. Có một loại không phải đi ba con đường đánh giá an toàn, chứng nhận và hợp đồng mẫu: mua sắm xuyên biên giới, gửi hàng xuyên biên giới, đặt vé máy bay và khách sạn, vì đó là「为订立、履行个人作为一方当事人的合同」— những chuyển dữ liệu thật sự cần thiết để lập hoặc thực hiện hợp đồng có cá nhân là một bên. Còn phải chú ý cách đếm số người: tính「自当年 1 月 1 日起累计」— từ ngày 1 tháng 1 năm đó trở đi, không phải đếm lùi mười hai tháng từ hôm nay + +### 11. Chọn máy chủ thì xem sập có chịu được không, xem có ai trực vận hành không, cuối cùng mới so giá + +- Chi phí: Máy chủ ảo của các hãng cloud lớn, mỗi tháng từ vài trăm đến vài nghìn tệ. VPS của nhà cung cấp nhỏ, mỗi tháng bắt đầu từ mấy chục tệ. VPS là một máy chủ ảo nhỏ. Thuê chỗ đặt máy chủ riêng thì mỗi năm bắt đầu từ vài nghìn tệ, và phần cứng phải tự mua. +- Nói thẳng: Ba bậc không khác nhau ở giá, mà khác nhau ở chỗ ngày xảy ra sự cố thì trong tay bạn có cái gì. Hãng lớn có kênh đăng ký khai báo, tư cách được công khai, hỏng thì có thông báo, có điều khoản bồi thường trong hợp đồng, và có người để tìm. Nhà cung cấp nhỏ rẻ hơn một nửa, nhưng khi nó bán vượt sức chứa, bỏ trốn, hoặc bị phía trên chặn thì bạn không biết kêu ai. Trang web tĩnh không có dữ liệu người dùng thì VPS mấy chục tệ là đủ dùng. +- Lợi ích: Ba bậc không khác nhau ở giá, mà khác nhau ở chỗ ngày xảy ra sự cố thì trong tay bạn có cái gì. Các hãng cloud lớn có kênh đăng ký khai báo, tư cách được công khai, có sự cố thì có thông báo, trong hợp đồng có điều khoản bồi thường, xảy ra chuyện thì tìm được người. Giá của nhà cung cấp nhỏ có thể thấp hơn một nửa, nhưng khi nó bán vượt sức chứa, bỏ trốn, hoặc bị phía trên chặn thì bạn không có đối tượng khiếu nại, nên trước khi bàn giá hãy tra hệ thống của Bộ Công nghiệp và Thông tin xem nó đã có giấy phép kinh doanh dịch vụ viễn thông giá trị gia tăng chưa. Thuê chỗ đặt máy chủ riêng chỉ đáng tiền khi cả hai điều kiện cùng đúng: băng thông hoặc chi phí phần cứng rẻ hơn hẳn, và bạn có người của mình trực vận hành. Cách chọn: trang web tĩnh không có dữ liệu người dùng thì VPS rẻ nhất là đủ. Trang web có đăng ký người dùng và giao dịch thì chọn máy chủ ảo co giãn theo nhu cầu và có ảnh chụp dự phòng, dữ liệu dự phòng để ở ít nhất hai chỗ. +- Mức bằng chứng: C +- Nguồn: kinh nghiệm của tác giả, không có tài liệu trực tiếp; yêu cầu pháp lý về tư cách của nhà cung cấp và về đăng ký khai báo nằm ở mục 4 của phần này, nghĩa vụ giữ nhật ký 6 tháng và nghĩa vụ bảo vệ theo cấp độ nằm ở mục 16 ở phần 11 +- Ghi chú: Phần trên là phán đoán từ kinh nghiệm, không có bằng chứng từ tài liệu. Chỉ có một việc kiểm chứng được thật sự: giấy phép của nhà cung cấp. Muốn tiết kiệm thì tiết kiệm ở cấu hình và băng thông, đừng tiết kiệm ở tư cách pháp lý + +## Giấy phép + +Nội dung phát hành theo [CC BY 4.0](../LICENSE). Nếu in lại hoặc sửa đổi, xin ghi nguồn và kèm liên kết bản gốc. Mã nguồn theo [MIT](../LICENSE-CODE). \ No newline at end of file diff --git "a/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" "b/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" new file mode 100644 index 00000000..3728a810 --- /dev/null +++ "b/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" @@ -0,0 +1,151 @@ +[← Về mục lục](../README.md) + +# 27. Mang thai và sinh con: từ que thử đến giấy tờ + +Nửa đầu phần này tính rủi ro tử vong của người mẹ và của thai nhi. Nửa sau tính tiền và thời gian chạy vặt. Hai thứ này không quy đổi qua lại. Nội dung chỉ nói khi nào làm việc gì. Dinh dưỡng khi mang thai, giáo dục thai nhi, loại không có bằng chứng chắc chắn hoặc tác dụng rất nhỏ, ở đây không viết. Phần tiền của trợ cấp thai sản, nghỉ phép thai sản và nuôi con xem phần 18. Cách chăm đứa trẻ sau khi sinh xem phần 20. Cấp cứu xem phần 13. + +Con số của Tổ chức Y tế Thế giới: năm 2023, khoảng 260.000 phụ nữ trên toàn thế giới chết vì mang thai và sinh con. Khoảng 75% trong số đó chết vì năm nhóm nguyên nhân sau: chảy máu nặng sau sinh, nhiễm trùng, rối loạn tăng huyết áp khi mang thai (nhóm bệnh do huyết áp tăng sau khi có thai, gồm tiền sản giật thai kỳ và sản giật), biến chứng khi sinh nở, và nạo thai không an toàn. Tổ chức Y tế Thế giới cũng nói「大多数孕产妇死亡是可以预防的，因为预防和处理并发症的医疗手段都是已知的」. Vài mục có tác dụng lớn nhất trong phần này đều nhằm chặn những nhóm bệnh này lại từ sớm. + +### 1. Bắt đầu uống 0.4 mg axit folic mỗi ngày từ trước khi mang thai, và uống liền suốt 3 tháng đầu thai kỳ + +- Chi phí: Vài chục tệ một chai. Phụ nữ nông thôn không tốn tiền, chương trình bổ sung axit folic miễn phí của nhà nước sẽ phát. Mỗi ngày một viên, chỗ khó là nhớ đều mỗi ngày +- Nói thẳng: Uống một viên axit folic mỗi ngày trước và quanh lúc mang thai giảm khoảng bảy phần mười xác suất trẻ bị khuyết tật ống thần kinh. Khuyết tật ống thần kinh là nhóm dị tật bẩm sinh như không có não, nứt đốt sống. Người đã sinh một đứa con bị như vậy thì xác suất đứa tiếp theo lại bị, xuống còn khoảng một phần ba mức cũ. +- Lợi ích: Phân tích tổng hợp của Cochrane gộp 5 thử nghiệm, 6708 lượt đẻ, và xếp chất lượng bằng chứng ở mức cao. Uống axit folic mỗi ngày trước và quanh lúc mang thai cho nguy cơ khuyết tật ống thần kinh của thai nhi là RR 0.31 (95% CI 0.17 đến 0.58, khoảng tin cậy), tức giảm khoảng bảy phần mười. Người đã có một đứa con bị khuyết tật ống thần kinh thì nguy cơ đứa tiếp theo lại bị là RR 0.34 (95% CI 0.18 đến 0.64), còn khoảng một phần ba mức cũ. Tách ra theo nhóm dân số, liều 0.4 mg và liều cao hơn không khác nhau. Uống riêng axit folic và uống trong vitamin tổng hợp cũng không khác nhau +- Mức bằng chứng: A +- Nguồn: De-Regil LM, Peña-Rosas JP, Fernández-Gaxiola AC, Rayco-Solon P (2015). Effects and safety of periconceptional oral folate supplementation for preventing birth defects. Cochrane Database of Systematic Reviews, (12), CD007950. +- Ghi chú: Điểm mấu chốt là「uống từ trước khi có thai」. Ống thần kinh của thai nhi khép lại trong vòng 28 ngày sau khi thụ thai. Đợi que thử dương rồi mới uống thì đã mất hơn nửa khoảng thời gian đó. Nên nếu định có con, cứ bắt đầu uống trước 3 tháng. Cùng nghiên cứu tổng hợp đó không thấy tác dụng rõ ràng của axit folic với khoảm miệng hở vòm, bệnh tim bẩm sinh hay sẩy thai. Đừng trông chờ nó chữa bệnh khác + +### 2. Trước tuần thai kỳ thứ 13, ra trạm y tế cộng đồng lập Sổ sức khoẻ bà mẹ và con, và dùng hết phần khám thai miễn phí + +- Chi phí: Không tốn tiền, đây thuộc chương trình dịch vụ công cộng cơ sở về y tế quốc gia. Phải chạy mấy chuyến theo số tuần thai kỳ +- Nói thẳng: Ít nhất năm lần khám thai trong thai kỳ, cộng một lần thăm khám tại nhà sau sinh và một lần khám 42 ngày sau sinh, đều do nhà nước trả. Cách nhận là trước tuần mang thai thứ 13 ra cộng đồng lập một quyển Sổ sức khoẻ bà mẹ và con《母子健康手册》. Không lập thì mấy lần miễn phí đó mất trắng. +- Lợi ích: 《国家基本公共卫生服务规范（第三版）》 có dịch vụ quản lý sức khoẻ sản phụ. Sản phụ thường trú trong địa bàn được làm các dịch vụ đó. Khám sức khoẻ 1 lần ở tam nguyệt đầu, 2 lần ở tam nguyệt giữa, 2 lần ở tam nguyệt cuối, 1 lần thăm khám sau sinh và 1 lần khám sức khoẻ 42 ngày sau sinh. Phải lập sổ trước tuần thai kỳ thứ 13, đồng thời làm lần khám tiền sản thứ nhất. Hai lần ở tam nguyệt giữa rơi vào tuần thai kỳ thứ 16 đến 20 và tuần thai kỳ thứ 21 đến 24. Hai lần ở tam nguyệt cuối rơi vào tuần thai kỳ thứ 28 đến 36 và tuần thai kỳ thứ 37 đến 40. Danh mục dịch vụ công khai của địa phương còn gồm toàn bộ xét nghiệm của lần đầu và sàng lọc trầm cảm sau sinh +- Mức bằng chứng: A +- Nguồn: 国家卫生计生委 (Uy ban Sức khoẻ và Kế hoạch Gia đình quốc gia) (2017). Quy định về dịch vụ công cộng cơ sở về y tế quốc gia, 国家基本公共卫生服务规范（第三版）. ；靖江市人民政府 (Chính quyền nhân dân thành phố Tĩnh Giang). Chương trình dịch vụ công cộng cơ sở về y tế quốc gia — dịch vụ quản lý sức khoẻ sản phụ (bảng dịch vụ công khai của địa phương), 国家基本公共卫生服务项目——孕产妇健康管理服务（地方服务清单公开）. +- Ghi chú: 5 lần đó là mức tối thiểu được miễn phí, không phải toàn bộ khám thai. Bệnh viện sẽ tuỳ tình hình làm thêm siêu âm, đường huyết, theo dõi tim thai, những khoản đó phải tự túi. Lập sổ càng sớm thì phân loại rủi ro thai kỳ càng sớm được làm ra. Phân loại rủi ro thai kỳ là xếp sản phụ theo mức rủi ro, chỉ khi phát hiện thuộc nhóm nguy cơ cao thì mới kịp chuyển sang bệnh viện đủ sức tiếp nhận. Bạn ở ngoài quận sở tại thì cứ lập sổ ở nơi đang sống, không cần về nơi đăng ký hộ khẩu + +### 3. Xét ba bệnh HIV, giang mai, viêm gan B ngay lần khám thai đầu tiên, và nếu ra kết quả vẫn có cách chặn miễn phí + +- Chi phí: Không tốn tiền, chương trình phòng lây truyền từ mẹ sang con cung cấp xét nghiệm miễn phí +- Nói thẳng: Ngay lần khám thai đầu tiên hãy xét cả ba bệnh HIV, giang mai, viêm gan B, xét nghiệm không tốn tiền của bạn. Ra kết quả dương tính cũng không phải ngõ cụt. Nhà nước cấp miễn phí thuốc và phác đồ để chặn con đường bệnh truyền từ mẹ sang con, đó gọi là chặn lây truyền mẹ con. +- Lợi ích: Lây truyền từ mẹ sang con nghĩa là bệnh truyền từ mẹ sang con. 《预防艾滋病、梅毒和乙肝母婴传播工作规范（2020 年版）》 có quy định về việc này. Vào tam nguyệt đầu hoặc lần khám tiền sản thứ nhất, bệnh viện phải thông báo các thông tin liên quan và「提供适宜、规范的免费检测」, nên ba xét nghiệm này không tốn tiền của bạn. Rồi căn cứ kết quả để tư vấn sau xét nghiệm hoặc chuyển tuyến. Ra kết quả dương tính cũng không phải ngõ cụt. Điều 44 của 《艾滋病防治条例》 quy định phải「免费提供预防艾滋病母婴传播的治疗和咨询」 cho sản phụ nhiễm HIV. Điều 43 yêu cầu có dịch vụ chặn lây, tức không để bệnh truyền sang con, rồi đến điều trị, thăm khám sau sinh, theo dõi trẻ sơ sinh và xét nghiệm. Khi mẹ dương tính với kháng nguyên bề mặt viêm gan B, trẻ sơ sinh phải tiêm đồng thời vắc-xin viêm gan B và globulin miễn dịch viêm gan B ngay khi sinh (xem mục 2 ở phần 20) +- Mức bằng chứng: A +- Nguồn: 国家卫生健康委办公厅 (Văn phòng Ủy ban Sức khoẻ quốc gia) (2020). Quy định về công tác phòng lây truyền HIV, giang mai và viêm gan B từ mẹ sang con, 预防艾滋病、梅毒和乙肝母婴传播工作规范（2020 年版）. （Ủy ban Sức khoẻ Tự trị Uy Tân Cương đăng lại）；国务院 (Hội đồng Nhà nước) (2006). Quy định về phòng ngừa và kiểm soát AIDS, 艾滋病防治条例（第四十三、四十四条）. （Ủy ban Sức khoẻ thành phố Bắc Kinh đăng lại） +- Ghi chú: Ba xét nghiệm này càng sớm càng hữu ích, vì chặn lây và điều trị đều cần thời gian. Điều 3 của 《艾滋病防治条例》 còn viết「任何单位和个人不得歧视艾滋病病毒感染者、艾滋病病人及其家属」, các quyền lợi hợp pháp như kết hôn, tìm việc, khám chữa bệnh, đi học đều được luật bảo vệ. Phòng ngừa và xét nghiệm thường ngày xem phần 1. Sau hành vi nguy cơ cao, cách chặn sau phơi nhiễm xem phần 13 + +### 4. Cả thai kỳ không hút một điếu, không uống một giọt rượu, và người trong nhà cũng đừng hút trong nhà + +- Chi phí: Không tốn tiền, còn tiết kiệm được tiền thuốc lá và rượu. Chỗ khó là khiến người ở cùng nhà cùng cai +- Nói thẳng: Hút thuốc khi mang thai làm tăng xác suất đẻ non, trẻ sinh ra quá nhỏ, khoảm miệng hở vòm. Nguy cơ chảy máu bất thường khi mang thai và khi sinh tăng thẳng gấp đôi. Nguy cơ tử vong đột ngột ở trẻ sơ sinh cũng cao hơn. Người trong nhà hút trong nhà thì tính y như vậy. Bỏ thuốc lúc nào vẫn kịp. +- Lợi ích: Cách nói của Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ: hút thuốc khi mang thai khiến đứa trẻ「足月出生也可能过小」 và「可能早产」. Nó「使孕期和分娩期异常出血的风险翻倍」. Nó làm tăng nguy cơ dị tật bẩm sinh như hở môi, hở vòm. Nó còn「损害胎儿正在发育的肺和大脑，损害可持续到儿童期」. Trẻ của người mẹ hút thuốc khi mang thai, và trẻ sau khi sinh phải hít vào khói, có nguy cơ hội chứng tử vong đột ngột ở trẻ sơ sinh cao hơn. Lời khuyên chính thức là「最好在怀孕前就戒烟，但如果已经怀孕，戒了仍然有用」 và「什么时候戒都不晚」 +- Mức bằng chứng: A +- Nguồn: Centers for Disease Control and Prevention. Smoking During Pregnancy. +- Ghi chú: Người khác hút trong nhà, khói thuốc lá phụ mà người mang thai hít vào vẫn tính với thai nhi. Nên người ở cùng nhà cũng đừng hút trong nhà. Rượu cũng vậy: không có mức uống nào được biết là an toàn, cách tiện nhất khi mang thai là không uống giọt nào. Riêng chuyện thuốc lá và rượu với người lớn, xem phần 2 + +### 5. Nếu có yếu tố nguy cơ tiền sản giật thai kỳ, từ sau tuần thai kỳ thứ 12 uống mỗi ngày một viên aspirin liều thấp + +- Chi phí: Vài chục tệ. Mỗi ngày một viên, phải uống liền tới khi sinh. Chỗ khó là không quên +- Nói thẳng: Tiền sản giật thai kỳ là một bệnh tăng huyết áp trong thai kỳ. Người thuộc nhóm nguy cơ cao là người đã mắc ở lần mang thai trước, đang mang đa thai, có tăng huyết áp mạn, có đái thường trước khi mang thai, có bệnh thận hoặc bệnh tự miễn. Người nguy cơ cao uống từ tuần thai kỳ thứ 12 mỗi ngày một viên aspirin liều thấp thì tiền sản giật thai kỳ giảm khoảng một phần rưỡi, còn đẻ non và trẻ chết quanh ngày sinh đều giảm khoảng hai phần mười. Người khác đừng tự uống. +- Lợi ích: Nhóm Dịch vụ Phòng ngừa Hoa Kỳ đưa ra khuyến nghị mức B: với người có nguy cơ cao về tiền sản giật thai kỳ, bắt đầu uống aspirin liều thấp (81 mg/ngày) sau tuần thai kỳ thứ 12 để phòng bệnh. Tiền sản giật thai kỳ là một bệnh tăng huyết áp trong thai kỳ. Gộp nhiều nghiên cứu lại, kết quả như sau, 95% CI là khoảng tin cậy. Tiền sản giật thai kỳ RR 0.85 (95% CI 0.75 đến 0.95, 16 nghiên cứu), thấp hơn khoảng 15%. Đẻ non RR 0.80 (95% CI 0.67 đến 0.95, 13 nghiên cứu), thấp hơn khoảng 20%. Tử vong chu sinh RR 0.79 (95% CI 0.66 đến 0.96, 11 nghiên cứu), thấp hơn khoảng 21%; tử vong chu sinh là cái chết trong khoảng thời gian quanh lúc sinh. Nhỏ hơn tuổi thai và cơn ứng thai kém RR 0.82 (95% CI 0.68 đến 0.99, 16 nghiên cứu), thấp hơn khoảng 18%; nghĩa là thai nhi lớn nhỏ hơn mức bình thường ở tuổi thai đó +- Mức bằng chứng: A +- Nguồn: US Preventive Services Task Force (2021). Aspirin Use to Prevent Preeclampsia and Related Morbidity and Mortality: Preventive Medication. +- Ghi chú: Các nhóm được tính là nguy cơ cao: đã mắc tiền sản giật thai kỳ ở lần mang thai trước, đang mang thai đôi hoặc đa thai, có tăng huyết áp mạn. Ngoài ra còn có đái thường tuýp 1 hoặc tuýp 2 trước khi mang thai, bệnh thận, và bệnh tự miễn như lupus ban đỏ hệ thống, hội chứng kháng phospholipid. Không thuộc mấy nhóm đó thì đừng tự uống. Thuốc này phải có bác sĩ kê đơn, khi khám thai hãy nói rõ bệnh sử để bác sĩ đánh giá. Ở trong nước bán chủ yếu là viên bao tanh 100 mg, uống theo chỉ định của bác sĩ + +### 6. Từ tuần thai kỳ thứ 24 trở đi, làm một lần sàng lọc đái thường thai kỳ, đừng ngại uống nước đường + +- Chi phí: Vài chục tới hơn một trăm tệ. Phải đến lúc nhịn đói, lấy máu mấy lần, mất nửa buổi sáng +- Nói thẳng: Đái thường thai kỳ là đường huyết cao chỉ xuất hiện sau khi mang thai. Từ tuần thai kỳ thứ 24, dành nửa buổi sáng uống một lần nước đường và lấy máu mấy lần là ra kết quả. Không xét cũng không để ý, thì thai nhi to quá, khó đẻ, phải mổ, tiền sản giật thai kỳ và hạ đường huyết ở trẻ sơ sinh đều tăng nguy cơ. +- Lợi ích: Khuyến nghị mức B của Nhóm Dịch vụ Phòng ngừa Hoa Kỳ: sàng lọc đái thường thai kỳ ở tuần thai kỳ thứ 24 trở đi với phụ nữ mang thai không có triệu chứng, tức tự thấy không khó chịu. Đái thường thai kỳ là đường huyết cao chỉ xuất hiện sau khi mang thai. Nếu chỉ sàng lọc một lần thì lần đó phải đặt ở tuần thai kỳ thứ 24 trở đi. Về việc sàng lọc trước tuần thai kỳ thứ 24 có nên không, nhóm đã kết luận rằng bằng chứng hiện có chưa đủ để phân biệt lợi và hại (mức I). Đái thường thai kỳ không được kiểm soát làm tăng nguy cơ: thai nhi quá to, khó đẻ, mổ lấy thai, tiền sản giật thai kỳ và hạ đường huyết ở trẻ sơ sinh +- Mức bằng chứng: A +- Nguồn: US Preventive Services Task Force (2021). Gestational Diabetes: Screening. +- Ghi chú: Trong số người phát hiện bệnh, đa số kiểm soát được đường huyết bằng cách điều chỉnh ăn uống và vận động, số ít phải tiêm insulin. Đừng vì sợ phải tiêm mà bỏ lần xét này. Sau khi sinh còn phải xét lại đường huyết: người từng mắc đái thường thai kỳ sau này có nguy cơ mắc đái thường tuýp 2 cao hơn rõ rệt + +### 7. Thuộc lòng danh sách「cần vào viện ngay」này, áp dụng trong thai kỳ và suốt một năm sau sinh + +- Chi phí: Không tốn tiền. Mất năm phút để ghi nhớ, tốt nhất là cho người ở cùng nhà biết luôn +- Nói thẳng: Chỉ cần một dấu hiệu nào trong danh sách là đi viện trong ngày, đừng ở nhà chờ qua đêm. Đau đầu không dịu hoặc nặng dần, nhìn thấy khác đi, sốt tới 38 ℃. Tay hoặc mặt sưng rõ, thở không nổi, đau ngực hoặc tim đập nhanh. Đau bụng dữ dội không dịu, thai ngừng cử động hoặc giảm hẳn, chảy máu hoặc rò dịch ở âm đạo. Một tay hoặc một chân sưng và đau, có ý nghĩ làm hại mình hoặc đứa trẻ. Trong vòng một năm sau sinh vẫn tính. +- Lợi ích: Các dấu hiệu cảnh báo khẩn cấp về mẹ và thai mà Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ liệt kê gồm: đau đầu không dịu hoặc nặng dần, chóng mặt hoặc ngất, nhìn thấy khác đi, thân nhiệt đạt 38 ℃ trở lên; tay hoặc mặt sưng rõ, có ý nghĩ làm hại mình hoặc đứa trẻ, khó thở, đau ngực hoặc tim đập quá nhanh; buồn nôn và nôn dữ dội, đau bụng dữ dội không dịu, thai ngừng cử động hoặc giảm hẳn trong thai kỳ; chảy máu âm đạo hoặc rò dịch trong thai kỳ, chảy máu âm đạo hoặc dịch bất thường sau sinh; một tay hoặc một chân sưng nặng, đỏ hoặc đau. Những vấn đề này đều có thể xảy ra trong vòng một năm sau sinh, không phải「hết tháng nằm hồi là xong」 +- Mức bằng chứng: A +- Nguồn: Centers for Disease Control and Prevention. Hear Her — Urgent Maternal Warning Signs. ；World Health Organization. Maternal mortality fact sheet. +- Ghi chú: Công dụng của danh sách này là tách「chịu thêm chút nữa」 khỏi「phải đi ngay」. Đau đầu kèm nhìn mờ, cộng thêm đau vùng bụng trên, là dấu hiệu hướng tới tiền sản giật thai kỳ, tức bệnh tăng huyết áp trong thai kỳ. Một bắp chân sưng và đau là dấu hiệu hướng tới huyết khối tĩnh mạch sâu (xem phần 13). Theo Tổ chức Y tế Thế giới, chảy máu nặng sau sinh có thể cướp mạng một người phụ nữ khoẻ mạnh trong vài giờ. Có một dấu hiệu là đi thẳng viện, đừng ở nhà chờ qua đêm + +### 8. Vỡ ối thì nằm ngửa tại chỗ, kê cao mông, gọi 120, đừng đi lại và đừng tắm + +- Chi phí: Không tốn tiền +- Nói thẳng: Vỡ ối rồi thì không phải xếp đồ chậm rãi rồi mới đi. Cách đúng là nằm xuống tại chỗ, kê cao mông, gọi 120. Đứng và đi lại thì nước ối thoát nhanh hơn, dây rốn có thể trôi theo ra ngoài. Đó là cấp cứu có thể lấy mạng đứa trẻ. +- Lợi ích: Cách xử lý do Ủy ban Sức khoẻ thành phố Bắc Kinh đưa ra: sau khi vỡ ối thì「马上找个地方平躺下来，防止羊水流出来」「不要随意走动」, và「立即拨打 120 急救电话」. Trong lúc chờ xe, báo người nhà mang đồ tới bệnh viện gặp mình. Tài liệu của các cơ sở y tế giải thích thêm lý do: khi đang đứng thì nước ối thoát nhanh hơn, dây rốn của thai nhi có thể trôi ra ngoài theo, đó gọi là sa dây rốn, là cấp cứu đe dọa thai nhi. Vì vậy phải nằm xuống và kê cao mông, có thể chọn tư thế nằm nghiêng bên trái +- Mức bằng chứng: B +- Nguồn: 北京市卫生健康委员会 (Ủy ban Sức khoẻ thành phố Bắc Kinh). Chủ đề cấp cứu hằng tuần: bà mẹ sắp sinh bị vỡ ối sớm thì xử lý thế nào, 每周急救话题：准妈妈羊水早破怎样办？. ；北京市顺义区妇幼保健院 (Bệnh viện Sức khoẻ bà mẹ và trẻ em quận Thuận Nghĩa, Bắc Kinh). Giai đoạn cuối thai kỳ bị vỡ ối tại nhà thì làm gì, 孕晚期在家破水怎么办？. +- Ghi chú: Ba dấu hiệu sắp sinh là co bóp tử cung đều đặn, ra máu âm đạo, và vỡ ối. Hai dấu hiệu đầu còn xếp đồ cho thoải mái, vỡ ối là dấu hiệu duy nhất đòi「nằm xuống tại chỗ rồi gọi xe」. Sau khi vỡ ối mà tắm, đi toilet, tự lái xe, đều là cách làm sai điển hình. Ghi lại thời điểm vỡ ối và màu nước ối, rồi báo bác sĩ. Nước ối vàng xanh, hoặc có lẫn hạt phân su, là dấu hiệu thai nhi bị ngạt, tức thai thiếu oxy trong tử cung. Nếu đồng thời có co bóp tử cung đều đặn, thai cử động giảm rõ, sốt kèm rét run, thì càng phải nhấn với bác sĩ rằng đây là ca cấp cứu + +### 9. Muốn giảm đau khi sinh thì nói thẳng, điều này không làm tăng nguy cơ mổ lấy thai + +- Chi phí: Vài trăm tới hơn một nghìn tệ, giá ở mỗi nơi mỗi khác. Phải hỏi trước bệnh viện và khoa gây mê có làm được không. Chỗ khó là phải tự mình mở lời +- Nói thẳng: Giảm đau khi sinh là tiêm một liều thuốc gây mê vào thắt lưng để đẩy xuống cơn đau sinh nở. Nó hiệu quả hơn các cách giảm đau khác, và không làm tăng khả năng bạn phải mổ. Câu「tiêm giảm đau xong thì đau lưng cả đời」cũng chẳng có cơ sở. Giờ nó đã có mục tính phí riêng, bệnh viện cũng sẵn sàng làm hơn trước. +- Lợi ích: Phân tích tổng hợp của Cochrane kết luận: gây mê giảm đau ngoài màng cứng hiệu quả hơn các cách không dùng ngoài màng cứng, cả trong giảm đau khi sinh lẫn nâng cao mức hài lòng của sản phụ với cách giảm đau. Giảm đau khi sinh dùng đúng là gây mê ngoài màng cứng. Cùng nghiên cứu đó còn nói「硬膜外镇痛对剖宫产风险和长期腰痛没有影响」. Hiện tượng sinh có dùng dụng cụ, tức dùng dụng cụ để đỡ đứa trẻ ra, tăng lên không còn thấy trong các nghiên cứu từ 2005 trở đi. Từ năm 2024, hướng dẫn lập danh mục giá dịch vụ y tế sản khoa của Cục Quản lý bảo hiểm y tế quốc gia tách riêng thành các mục「分娩镇痛」「导乐分娩」「亲情陪产」. Lý do đúng là trước đó chưa có mục giá riêng, nên động lực của bệnh viện trong việc cung cấp giảm đau khi sinh bị giảm +- Mức bằng chứng: A +- Nguồn: Anim-Somuah M, Smyth RMD, Cyna AM, Cuthbert A (2018). Epidural versus non-epidural or no analgesia for pain management in labour. Cochrane Database of Systematic Reviews, (5), CD000331. ；国家医疗保障局 (Cục Quản lý bảo hiểm y tế quốc gia) (2024). Tái dựng danh mục giá dịch vụ sản khoa, góp phần xây dựng xã hội thân thiện với sinh con, 重构产科服务价格项目 助力构建生育友好型社会. +- Ghi chú: Hai câu「tiêm giảm đau xong đau lưng cả đời」 và「tiêm giảm đau xong khó đẻ phải mổ」 lây lan rộng nhất đều không có cơ sở. Cochrane gộp gây mê ngoài màng cứng với các cách giảm đau khác lại, và kết luận nó không ảnh hưởng tới nguy cơ mổ lấy thai cũng không ảnh hưởng tới đau lưng lâu dài. Tác dụng phụ có thật là những cái sau: huyết áp thấp, sốt, chân mềm yếu, tiểu tiện khó, và giai đoạn một cùng giai đoạn hai của cuộc sinh kéo dài hơn, dễ phải dùng thuốc kích thích tử cung hơn. Có làm hay không thì bạn tự quyết, nhưng đừng bỏ vì lời đồn. Có làm được hay không còn tuỳ bệnh viện có bác sĩ gây mê trực 24 giờ hay không, nên hỏi rõ lúc khám thai + +### 10. Không có chỉ định y khoa thì đừng tự yêu cầu mổ lấy thai, cũng đừng mổ chỉ để chọn ngày + +- Chi phí: Không tốn tiền. Chỗ khó là chịu đựng lời của người nhà +- Nói thẳng: Cách nói của Tổ chức Y tế Thế giới là: khi tỷ lệ mổ lấy thai ở một nơi vượt quá 10%, số tử vong ở bà mẹ và trẻ sơ sinh không còn tiếp tục giảm. Chỗ nào cần mổ thì phải mổ. Mổ chỉ để chọn ngày, hoặc vì sợ đau, chỉ là phải chịu một cuộc mổ bụng lớn mà không đổi lại được gì. +- Lợi ích: Tổ chức Y tế Thế giới phát biểu năm 2015: khi tỷ lệ mổ lấy thai trong một cộng đồng tăng lên khoảng 10%, số tử vong ở sản phụ và trẻ sơ sinh giảm xuống. Vượt quá 10% thì không có bằng chứng cho thấy tỷ lệ tử vong còn cải thiện thêm. Tổ chức Y tế Thế giới đồng thời nhấn mạnh, phải cố hết sức để cung cấp mổ lấy thai cho sản phụ thật sự cần, chứ không phải để đạt một tỷ lệ nào đó +- Mức bằng chứng: A +- Nguồn: World Health Organization (2015). WHO Statement on Caesarean Section Rates（WHO/RHR/15.02）. +- Ghi chú: Khi đúng là cần về mặt y khoa, mổ lấy thai là cuộc phẫu thuật cứu mạng, cứ mổ đi, đừng chịu đựng. Cái cần tránh là mổ khi không thật sự cần về mặt y khoa, chẳng hạn vì sợ đau, vì chọn ngày đẹp, hay vì tính tuổi vào trường của con. Mổ lấy thai là phẫu thuật bụng lớn, hồi phục chậm hơn, lần mang thai sau có nguy cơ cao hơn về nhau thai bám màng, nhau cấy vào tử cung, và vỡ tử cung. Điều này ảnh hưởng trực tiếp tới việc bạn còn muốn sinh đứa thứ hai hay không + +### 11. Trước khi mang thai, xác nhận tình hình tham gia bảo hiểm thai sản, vợ chồng chưa có việc làm cũng được hoàn phí khám chữa bệnh thai sản + +- Chi phí: Không tốn tiền. Gọi một lần đường dây hỏi bảo hiểm xã hội hoặc bảo hiểm y tế +- Nói thẳng: Khi đơn vị đóng bảo hiểm thai sản, phí khám chữa bệnh sinh con và trợ cấp trong thời gian nghỉ phép thai sản đều lấy từ quỹ đó. Nếu vợ hoặc chồng chưa có việc làm, phần phí khám chữa bệnh cũng được hoàn. Đủ số tháng đóng chưa, sinh con ở nơi khác có phải đăng ký trước không, đều phải hỏi trước, sau đó không bù lại được. +- Lợi ích: Điều 54 của 《社会保险法》 quy định: nơi đơn vị sử dụng lao động đã đóng bảo hiểm thai sản thì người lao động của đơn vị được hưởng trợ cấp bảo hiểm thai sản. Nếu vợ hoặc chồng của người lao động không có việc làm thì được hưởng phần chi phí khám chữa bệnh thai sản theo quy định của nhà nước, kinh phí lấy từ quỹ bảo hiểm thai sản. Nói ngắn gọn, đơn vị đã đóng bảo hiểm thai sản thì người lao động được hưởng trợ cấp, và phí sinh con của người phối ngẫu không có việc làm cũng lấy từ quỹ đó. Trợ cấp bảo hiểm thai sản chia làm hai phần: một là chi phí khám chữa bệnh thai sản, tức phí khám chữa bệnh của riêng việc sinh con; hai là trợ cấp thai sản, tức khoản tiền trả cho bạn trong thời gian nghỉ phép thai sản. Điều 56 liệt kê ba trường hợp được nhận trợ cấp thai sản: nữ công nhân sinh con và được nghỉ phép thai sản, được nghỉ phép cho phẫu thuật kế hoạch hoá gia đình, và các trường hợp khác theo quy định của pháp luật. Trợ cấp thai sản「按照职工所在用人单位上年度职工月平均工资计发」, tức tính trên mức lương trung bình hằng tháng của người lao động tại đơn vị của bạn trong năm trước, không phải mức lương của chính bạn +- Mức bằng chứng: A +- Nguồn: 全国人大常委会 (Ủy ban Thường vụ Quốc hội) (2010). Luật Bảo hiểm xã hội của nước Cộng hoà Nhân dân Trung Quốc, 中华人民共和国社会保险法（第五十四、五十六条）. （Cục Thuế tỉnh Quảng Đông, Cục Quản lý thuế quốc gia đăng lại） +- Ghi chú: Có hai việc phải hỏi rõ trước, vì sau đó không bù lại được. Một là số tháng đóng liên tục có đủ không: mỗi nơi đều đặt ra số tháng tối thiểu, không đủ thì không được hưởng. Hai là sinh con ngoài nơi tham gia bảo hiểm thì có phải đăng ký sinh con ngoài địa phương trước không: không đăng ký thì có thể phải tự ứng tiền trước, về hoàn thủ bằng tay sau, thậm chí không hoàn được. Số ngày nghỉ phép thai sản và cách tính trợ cấp thai sản xem mục 2 ở phần 18 + +### 12. Trước khi xuất viện, làm xong Giấy khai sinh, nghĩ trước tên con và đừng viết sai chữ + +- Chi phí: Không tốn tiền. Làm luôn trong lúc làm thủ tục xuất viện +- Nói thẳng: Giấy này là điều kiện trước để đăng ký hộ khẩu, tham gia bảo hiểm và vào trường, và chỉ cơ sở đã đỡ đẻ mới có thể cấp lần đầu. Ra viện rồi mới quay lại bù, phải đi một quy trình khác và nộp thêm giấy tờ. Chữ trong tên, trước hết phải xác nhận hệ thống hộ khẩu hiện ra được. +- Lợi ích: Điều 23 của 《母婴保健法》 quy định: cơ sở y tế và người hành nghề đỡ đẻ tại nhà, theo quy định của cơ quan quản lý y tế của Hội đồng Nhà nước, phải cấp Giấy khai sinh do cơ quan thống nhất in và phát. Giấy này là điều kiện trước cho mọi việc phía sau: đăng ký hộ khẩu, tham gia bảo hiểm, làm thẻ bảo hiểm y tế, nhập trường mầm non và vào trường. Và lần cấp đầu tiên chỉ cơ sở đã đỡ đẻ mới làm được. Ra viện rồi mới quay lại bù, phải theo quy trình khác và cần thêm giấy tờ chứng minh +- Mức bằng chứng: A +- Nguồn: 全国人大常委会 (Ủy ban Thường vụ Quốc hội) (1994, sửa đổi 2017). Luật Chăm sóc bà mẹ và trẻ sơ sinh của nước Cộng hoà Nhân dân Trung Quốc, 中华人民共和国母婴保健法（第二十三条）. +- Ghi chú: Trước khi điền, đối chiếu từng chữ tên bố mẹ, số căn cước công dân và tên của trẻ. Chữ dùng trong tên phải là chữ mà hệ thống hộ khẩu hiện ra được, chữ hiếm sẽ khiến mọi việc sau này rắc rối hơn. Đổi tên hay đổi giấy đều phải qua bổ cấp hoặc cấp lại, phiền hơn nhiều so với kiểm tra tại chỗ. Cầm giấy xong thì cất riêng, giữ cẩn thận ngang với sổ hộ khẩu và căn cước công dân + +### 13. Đừng từ chối cả sàng lọc máu gót chân cho trẻ sơ sinh lẫn sàng lọc thính + +- Chi phí: Ở phần lớn các nơi miễn phí, hoặc vài chục tệ. Có mấy mục mở rộng thì phải tự túi. Sau khi sinh, chọc gót chân trẻ một lần +- Nói thẳng: Một lần lấy máu gót chân và một lần kiểm tra thính giác có thể phát hiện suy giảm tuyến giáp, bệnh phenylketonuria và khiếm thính khi trẻ vẫn trông hoàn toàn bình thường. Đợi triệu chứng lộ ra thì tổn hại trí tuệ đã không bù lại được. +- Lợi ích: Điều 3 của 《新生儿疾病筛查管理办法》 quy định:「全国新生儿疾病筛查病种包括先天性甲状腺功能减低症、苯丙酮尿症等新生儿遗传代谢病和听力障碍。」 Hai nhóm bệnh này có một điểm chung: giai đoạn đầu hoàn toàn không thấy bất thường, đến khi triệu chứng rõ thì tổn hại trí tuệ đã không đảo ngược được. Mà phát hiện sớm và can thiệp sớm thì rất hiệu quả, cách làm là bổ sung hormone tuyến giáp, uống sữa công thức chuyên biệt, đeo máy trợ thính hoặc cấy ghép ốc tai nhân tạo. Điều 10 quy định, nếu kết quả sàng lọc dương tính thì cơ sở phải kịp báo người giám hộ đi xác nhận chẩn đoán +- Mức bằng chứng: A +- Nguồn: 卫生部 (Bộ Y tế) (2009). Biện pháp quản lý sàng lọc bệnh ở trẻ sơ sinh, 新生儿疾病筛查管理办法（卫生部令第 64 号，第三、十、十一条）. +- Ghi chú: Điều 11 của quy định này yêu cầu trước khi sàng lọc phải nói thật với người giám hộ về các mục, điều kiện, cách làm, độ nhạy, tức khả năng phát hiện được bao nhiêu, và chi phí, đồng thời lấy chữ ký đồng ý. Nên bạn sẽ nhận một phiếu đồng thuận sau khi được thông tin. Trước khi ký, đọc kỹ: nhóm bệnh nào do nhà nước quy định, nhóm nào là mục bổ sung tự trả tiền. Sàng lọc lần đầu dương tính không phải là chẩn đoán, đa số xét lại một lần là loại trừ được, đừng hoảng loạn ngay ở bước này. Sàng lọc thính lần đầu không đạt thì phải đi xét lại theo thông báo, đừng tự kết luận「trẻ nghe thấy là được」 + +### 14. Có giấy khai sinh rồi thì lập bảo hiểm y tế cho con, đừng đợi đăng ký xong hộ khẩu + +- Chi phí: Phí bảo hiểm y tế người dân cư trú đóng một lần mỗi năm, vài trăm tệ +- Nói thẳng: Có giấy khai sinh là lập bảo hiểm cho con trên mạng được, không cần đợi đăng ký hộ khẩu. Đã tham gia bảo hiểm thì phí nằm viện lúc xuất viện được hoàn ngay tại chỗ. Phí nằm viện của trẻ sơ sinh không nhỏ, chênh lệch giữa có bảo hiểm và không có rất lớn. +- Lợi ích: Cách nói năm 2024 của Cổng thông tin Chính phủ Trung Quốc:「新生儿凭出生医学证明就可以在线上参保」, không cần đăng ký hộ khẩu trước.「新生儿出院以后就能够直接报销医药费用」. Cách làm trước đây là phải đăng ký hộ khẩu rồi mới được tham gia bảo hiểm, phí nằm viện phải ứng đủ trước rồi mới hoàn sau. Thời gian xử lý bình quân việc đăng ký và đóng tiền cũng đã rút từ 28.7 ngày làm việc đầu năm xuống còn 6.4 ngày làm việc +- Mức bằng chứng: A +- Nguồn: 中国政府网 (Cổng thông tin Chính phủ Trung Quốc) (2024). Họp báo định kỳ chính sách của Hội đồng Nhà nước về cơ chế dài hạn cho việc tham gia bảo hiểm y tế cơ sở, 健全基本医疗保险参保长效机制国务院政策例行吹风会. +- Ghi chú: Đẻ non, vàng da phải chiếu đèn xanh, viêm phổi trẻ sơ sinh là những lý do thường gặp khiến trẻ sơ sinh phải nằm viện, chi phí dễ động tới mấy vạn tệ, chênh lệch giữa có bảo hiểm và không có rất lớn. Quy định ở các nơi về việc phải tham gia bảo hiểm trong bao nhiêu ngày sau khi sinh thì mới được hưởng từ ngày sinh không giống nhau. Cách thực hiện cụ thể thì theo đúng lời cơ quan bảo hiểm y tế địa phương. Gọi 12393 hỏi một câu trước khi xuất viện là tiện nhất + +### 15. Trong vòng một tháng sau khi trẻ sinh, ra công an phường xã khai đăng ký sinh + +- Chi phí: Không tốn tiền. Chạy một lần tới quầy hộ khẩu ở công an phường xã +- Nói thẳng: Luật cho một tháng, không phải「khi nào rảnh thì nói」. Đăng ký hộ khẩu chặn các việc phía sau: tiêm phòng, nhập trường mầm non, vào trường và làm căn cước công dân. Càng kéo dài, càng phải bù thêm nhiều giấy tờ. +- Lợi ích: Điều 7 của 《户口登记条例》 quy định:「婴儿出生后一个月以内，由户主、亲属、抚养人或者邻居向婴儿常住地户口登记机关申报出生登记。」 Đó là thời hạn do luật đặt ra, không phải lời khuyên. Đăng ký hộ khẩu là điều kiện trước cho các việc phía sau: tiêm phòng, nhập trường mầm non, vào trường, làm căn cước công dân. Càng kéo dài, càng phải bù thêm nhiều giấy tờ +- Mức bằng chứng: A +- Nguồn: 国务院 (Hội đồng Nhà nước) (1958). Quy định về đăng ký hộ khẩu của nước Cộng hoà Nhân dân Trung Quốc, 中华人民共和国户口登记条例（第七条）. （Chính quyền nhân dân tỉnh Quảng Đông đăng lại） +- Ghi chú: Mang theo đủ những giấy này: giấy khai sinh, căn cước công dân và sổ hộ khẩu của cả hai bên, giấy đăng ký kết hôn. Đăng ký theo bố hay theo mẹ thì bạn tự chọn, nhưng chính sách hai nơi khác nhau, cả vùng trường và nơi tham gia bảo hiểm đều bị ảnh hưởng. Suy nghĩ cho xong trước khi đi, đổi lại rất phiền. Trẻ sinh ngoài hôn nhân vẫn được khai đăng ký sinh như thường + +### 16. Đừng bỏ lần tái khám 42 ngày sau sinh, đó cũng là lần sàng lọc trầm cảm sau sinh + +- Chi phí: Không tốn tiền, thuộc dịch vụ công cộng cơ sở về y tế. Mất nửa ngày. Chỗ khó là vừa hết tháng nằm hồi thì lười nhúc nhích +- Nói thẳng: Lần tái khám này không tốn tiền, gồm công thức máu, xét nghiệm dịch âm đạo thường quy, và cả một lần sàng lọc trầm cảm sau sinh. Trong vòng một năm sau sinh vẫn có thể xảy ra chuyện nghiêm trọng, đừng nghĩ hết tháng nằm hồi là đã an toàn. +- Lợi ích: Phần quản lý sức khoẻ sản phụ trong 《国家基本公共卫生服务规范（第三版）》 có sau sinh là 1 lần thăm khám sau sinh và 1 lần khám sức khoẻ 42 ngày sau sinh. Danh mục dịch vụ công khai của địa phương cho thấy lần này gồm xét nghiệm máu, xét nghiệm dịch âm đạo thường quy, và 1 lần sàng lọc trầm cảm sau sinh. Biến chứng nặng có thể xảy ra bất cứ lúc nào trong vòng một năm sau sinh. Trong danh sách「cần vào viện ngay」 ở mục 7 của phần này,「有伤害自己或伤害孩子的念头」「产后阴道出血或异常分泌物」「一侧肢体严重肿痛」 đều thuộc giai đoạn sau sinh +- Mức bằng chứng: A +- Nguồn: 国家卫生计生委 (Uy ban Sức khoẻ và Kế hoạch Gia đình quốc gia) (2017). Quy định về dịch vụ công cộng cơ sở về y tế quốc gia, 国家基本公共卫生服务规范（第三版）. ；靖江市人民政府 (Chính quyền nhân dân thành phố Tĩnh Giang). Chương trình dịch vụ công cộng cơ sở về y tế quốc gia — dịch vụ quản lý sức khoẻ sản phụ (bảng dịch vụ công khai của địa phương), 国家基本公共卫生服务项目——孕产妇健康管理服务（地方服务清单公开）. +- Ghi chú: Trầm cảm sau sinh không phải là nũng niu, người nhà khuyên một câu「cứ nghĩ thoáng đi」 thì không giải quyết được. Nó có bảng hỏi sàng lọc riêng, có cách điều trị, và có thể hồi phục. Khi có ý nghĩ tự sát thì xử lý thế nào, xem phần 1. Khi đã sinh con mà thật sự không nuôi nổi thì lối thoát hợp pháp xem mục 20 ở phần 9. Lần tái khám này còn xem vết thương đã lành thế nào, tử cung hồi phục ra sao, chức năng sàn vùng chậu có tốt không. Kế hoạch tránh thai cũng được bàn lúc này: sau sinh không có kinh nguyệt không có nghĩa là không thể có thai diff --git "a/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" "b/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" new file mode 100644 index 00000000..769dcaf8 --- /dev/null +++ "b/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" @@ -0,0 +1,122 @@ +[← Về mục lục](../README.md) + +# 29. Sau một cú đánh nặng + +Người thân qua đời, được báo mình mắc bệnh nặng, thất nghiệp, ly hôn. Trong các nghiên cứu, mấy việc này có một điểm chung: không chỉ tâm trạng mới hỏng, mà cả cơ thể nữa. Đoạn có nguy cơ cao nhất rơi vào mấy tuần đầu cho tới năm đầu. Phần này không nói làm sao để ra khỏi cảnh đó, chỉ nói trong khoảng thời gian ấy việc nào phải có người trông, quyết định nào phải đẩy lùi, và lúc nào nên đi đăng ký khám. Phần lớn các mục tính theo xác suất tử vong. Bốn mục nói về tiền và quyền lợi thì tính theo tiền: đừng vội trả tiền làm tư vấn đau buồn (mục 9), gọi 12356 và đăng ký khám tâm lý ngoại trú (mục 11), hoãn mọi quyết định lớn không thể đảo ngược (mục 12), đừng dùng cái chết để trả nợ (mục 13). Hai loại số liệu này không quy đổi qua lại. Người không có thân nhân cũng không có bạn để nhờ thì làm sao cho câu «có người trông hộ» thành hiện thực, xem mục 6. + +Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải lấy xem phần 25. Giám hộ và di chúc trước khi người thân qua đời xem phần 17. Thất nghiệp thì được nhận khoản gì xem phần 7. Vận động và ánh sáng khi tâm trạng sa sút xem phần 3 và phần 22. Có ý nghĩ tự tử thì gọi 12356 trước, xem mục 25 ở phần 1; ý nghĩ đó kéo dài bao lâu, xem mục 32 ở phần 1; di chứng còn lại sau khi được cứu, xem mục 33 ở phần 1. Những việc đó phần này không lặp lại. + +### 1. Trong mấy ngày đầu sau khi người thân qua đời, đừng ở một mình; người có bệnh tim, huyết áp cao, tiểu đường thì vẫn uống thuốc như cũ, đau tức ngực thì gọi 120 + +- Chi phí: Không tốn tiền. Mấy ngày đó đừng ở một mình, thuốc thì vẫn uống như cũ +- Nói thẳng: Ngày đầu tiên sau khi người thân qua đời, xác suất bị nhồi máu cơ tim cao gấp 21.1 lần so với bình thường, rồi giảm dần theo từng ngày. Người 60 đến 89 tuổi, trong 30 ngày sau khi người bạn đời qua đời, tỷ lệ nhồi máu cơ tim hoặc đột quỵ tăng khoảng gấp đôi, rồi giảm lại sau một tháng. Người vốn đã có bệnh tim mạch thì mấy ngày đó đừng ngừng thuốc. Đau tức ngực, khó thở, một bên tay chân yếu thì gọi 120 ngay +- Lợi ích: 1.985 bệnh nhân nhồi máu cơ tim cấp, so sánh chính người đó với các khoảng thời gian trước khi bệnh khởi phát. Trong đó 270 người (13.6%) mất người thân trong vòng 6 tháng trước khi bệnh, 19 người mất trong vòng 1 ngày trước khi nhồi máu. Trong 24 giờ sau khi người thân qua đời, tỷ lệ nhồi máu cơ tim cấp tăng 21.1 lần (95% CI 13.1–34.1, tức khoảng 13 đến 34 lần). Sau đó giảm dần theo từng ngày. Đổi sang số người thực tế: trong nhóm có nguy cơ nhồi máu cơ tim 5% trong mười năm, cứ 1394 người mới mất người thân thì thêm 1 ca. Trong nhóm có nguy cơ 20%, cứ 320 người thì thêm 1 ca. Một nghiên cứu khác ở Anh so 30.447 người 60 đến 89 tuổi mất bạn đời với 83.588 người đối chứng có điều kiện tương đương. Trong 30 ngày sau khi mất bạn đời, nhóm mất bạn đời có 50 ca (0.16%), nhóm đối chứng có 67 ca (0.08%). IRR 2.20 (1.52–3.15), tức xác suất bị bệnh cao khoảng 2.2 lần nhóm đối chứng. Trong đó nhồi máu cơ tim là 2.14 (1.20–3.81), khoảng 2.1 lần. Đột quỵ là 2.40 (1.22–4.71), khoảng 2.4 lần. Sau 30 ngày thì giảm. Trong 90 ngày còn xét thêm hai nhóm bệnh. Một là hội chứng vành cấp không nhồi máu cơ tim (tim thiếu máu đột ngột nhưng chưa đủ để nhồi máu), IRR 2.20 (1.12–4.29). Một là thuyên tắc phổi (cục máu đông nghẽn trong mạch phổi), IRR 2.37 (1.18–4.75) +- Mức bằng chứng: A +- Nguồn: Mostofsky E, et al. (2012). Risk of acute myocardial infarction after the death of a significant person in one's life: the Determinants of Myocardial Infarction Onset Study. Circulation. ；Carey IM, et al. (2014). Increased risk of acute cardiovascular events after partner bereavement: a matched cohort study. JAMA Internal Medicine. +- Ghi chú: Đưa hộp thuốc cho người khác giữ, đừng để mình tự nhớ. Mấy ngày đó đừng đi đường dài. Đừng dựa vào rượu để chịu đựng, xem phần 2. Nhồi máu cơ tim và đột quỵ nhận biết thế nào, gọi xe thế nào, xem phần 13. Cả hai đều là nghiên cứu nhìn lại dữ liệu nhóm người, không tách được trong đó có bao nhiêu người vốn đang sắp bệnh. Nhưng cả hai cùng chỉ một hướng và cùng một khoảng thời gian. Người không thể đưa hộp thuốc cho ai và chắc mấy ngày đó ở một mình trong nhà thì xem mục 6 của phần này + +### 2. Trong tuần bạn được báo mắc bệnh nặng, đừng đi lấy kết quả một mình, quyết định gì ngoài điều trị thì đẩy lùi + +- Chi phí: Không tốn tiền. Tìm một người đi cùng bạn lấy kết quả. Đẩy mấy quyết định lớn ngoài điều trị lùi vài ngày +- Nói thẳng: Trong tuần đầu sau khi có chẩn đoán ung thư, xác suất tự tử cao gấp 12.6 lần so với người chưa được chẩn đoán, xác suất chết vì bệnh tim mạch cao gấp 5.6 lần. Đến năm thứ nhất, nguy cơ tự tử giảm còn 3.1 lần, ung thư càng khó chữa thì càng rõ. Nên ngày lấy kết quả hãy đi cùng một người, về đến nhà đừng ở một mình. Chuyện nghỉ việc, bán nhà thì để đến lần khám sau +- Lợi ích: Nhật Bản theo dõi hồ sơ toàn quốc hơn 607 vạn người (6.073.240 người), giai đoạn 1991 đến 2006. So với người không bị chẩn đoán ung thư, sau khi chẩn đoán thì nguy cơ tự tử tương đối trong tuần đầu là 12.6 (95% CI 8.6–17.8), tức khoảng 12.6 lần. Hệ số thật nằm trong khoảng 8.6 đến 17.8. Tỷ lệ mắc trong tuần này là 2.50/1000 người-năm, tức 1.000 người theo dõi một năm thì có 2.5 ca. Đến năm thứ nhất giảm còn 3.1 (2.7–3.5, 0.60/1000 người-năm), khoảng 3.1 lần. Tử vong tim mạch trong tuần đầu là 5.6 (5.2–5.9, 116.80/1000 người-năm), khoảng 5.6 lần. 4 tuần đầu là 3.3 (3.1–3.4, 65.81/1000 người-năm). Nguy cơ tăng rồi nhanh chóng giảm trong năm đầu sau khi chẩn đoán. Ung thư càng khó chữa, càng ít khả năng sống thì mức tăng càng rõ. So sánh chính nhóm bệnh nhân ở các khoảng thời gian khác thì kết luận vẫn vậy +- Mức bằng chứng: A +- Nguồn: Fang F, et al. (2012). Suicide and cardiovascular death after a cancer diagnosis. New England Journal of Medicine. +- Ghi chú: Đây là số liệu của bệnh nhân ung thư. Đổi sang các bệnh nan y khác thì không có nguồn gốc tương đương. Người gặp chuyện thực ra rất ít: tỷ lệ tự tử trong tuần đầu là 2.50/1000 người-năm, tức 1.000 người theo dõi một năm thì có 2.5 ca. Viết ra không phải để dọa, mà để nói rõ mấy ngày đầu mới biết tin phải có người ở bên cạnh. Quản lý lâu dài sau khi chẩn đoán, bồi hoàn và tái khám xem phần 16 và phần 24. Không tìm được người đi cùng thì xem mục 6 của phần này (thay «người đó» bằng hàng xóm và danh sách của cộng đồng) + +### 3. Sau khi thất nghiệp, ổn định giờ sinh hoạt, bảo hiểm y tế và nhịp đi tìm việc trước, đừng ở nhà suốt ngày + +- Chi phí: Không tốn tiền. Đi đăng ký thất nghiệp, nhận tiền bảo hiểm thất nghiệp trước, rồi xếp việc tìm việc vào một lịch cố định. Khó ở chỗ không ai đuổi mà vẫn phải dậy đúng giờ +- Nói thẳng: Người thất nghiệp sau đó tử vong nhiều hơn người còn việc khoảng sáu phần mười, nam và người không nhiều tuổi nổi bật hơn, mười năm đầu cao nhất. Trừ bớt yếu tố hút thuốc uống rượu thì mức tăng giảm khoảng một phần tư, phần này mình tự kiểm soát được. Đi đăng ký thất nghiệp, nhận tiền bảo hiểm thất nghiệp trước, trong thời gian nhận thì bảo hiểm y tế do quỹ bảo hiểm thất nghiệp đóng thay +- Lợi ích: 42 nghiên cứu, 235 ước lượng nguy cơ tử vong, hơn 20.000.000 người gộp lại. Sau khi cân bằng các yếu tố như tuổi, nguy cơ tử vong do mọi nguyên nhân ở người thất nghiệp có HR trung bình 1.63, cao hơn người còn việc khoảng sáu phần mười. Nam cao hơn nữ. Ảnh hưởng rõ ở người đang ở giai đoạn đầu và giữa sự nghiệp, giai đoạn sau thì nhỏ hơn. 10 năm đầu theo dõi có nguy cơ cao nhất, sau đó giảm. Trong số các nghiên cứu có tính cả hành vi sức khoẻ như hút thuốc, uống rượu vào, HR trung bình thấp hơn 24%, tức mức tăng tính ra giảm khoảng một phần tư +- Mức bằng chứng: A +- Nguồn: Roelfs DJ, et al. (2011). Losing life and livelihood: a systematic review and meta-analysis of unemployment and all-cause mortality. Social Science & Medicine. +- Ghi chú: Đây là nghiên cứu nhìn lại dữ liệu nhóm người, không tách được chiều nhân quả. Người vốn sức khoẻ kém cũng dễ thất nghiệp hơn. Nhưng câu «sau khi trừ phần ảnh hưởng của thói quen sinh hoạt thì hiệu ứng thấp hơn 24%» cho một hướng để bắt tay. Thất nghiệp được nhận khoản gì, ở đâu nhận, bảo hiểm xã hội gia hạn thế nào thì xem phần 7. Bị sa thải thì được bồi thường bao nhiêu, đừng ký tự nguyện nghỉ việc thì xem phần 19 + +### 4. Người thân chết vì tự tử, tai nạn hoặc bị giết thì đừng trông chờ tự gượng, hãy chủ động đi tìm giúp đỡ chuyên môn + +- Chi phí: Gọi 12356 thì không tốn tiền. Muốn khám bác sĩ thì đăng ký khám tâm lý ngoại trú, giá như khám ngoại trú thường. Khó ở chỗ phải tự mở lời nhờ người +- Nói thẳng: Người mất bạn đời vì tự tử, trong năm năm sau có xác suất được chẩn đoán rối loạn tâm thần cao hơn người bình thường khoảng tám phần mười (nam), bảy phần mười (nữ). So với người mất bạn đời vì nguyên nhân khác thì cũng cao hơn từ bảy phần mười đến gấp đôi. Người mất người thân kiểu này dễ cảm thấy bị người ta lạnh nhạt, dễ cảm thấy xấu hổ. Đó là lý do phải chủ động đi tìm người, đừng đợi mình ổn rồi hãy đi +- Lợi ích: Hồ sơ toàn quốc Đan Mạch theo dõi 6.700.000 người trưởng thành, giai đoạn 1980 đến 2014, tổng cộng hơn 136.000.000 người-năm (cộng dồn số năm mỗi người được theo dõi). Ở người mất bạn đời vì bạn đời tự tử, nguy cơ được chẩn đoán mới rối loạn tâm thần trong 5 năm sau so với người bình thường là IRR 1.8 ở nam (95% CI 1.6–2.0), tức cao hơn khoảng tám phần mười. Ở nữ là 1.7 (1.6–1.8), tức cao hơn khoảng bảy phần mười. So với người mất bạn đời vì nguyên nhân tử vong khác, IRR 1.7 ở nam (1.5–1.9) và 2.0 ở nữ (1.9–2.2), tức cao hơn từ bảy phần mười đến gấp đôi. Trong nhóm này, tỷ lệ có hành vi tự tử, tỷ lệ tử vong và tỷ lệ phải nhận trợ cấp của chính quyền địa phương cũng cao hơn. Các bệnh thể chất như xơ gan, rối loạn giấc ngủ cũng tăng nguy cơ. Ngoài ra có 57 nghiên cứu rà từng yếu tố, không gộp lại để tính. Mất người thân vì tự tử có liên quan tới bốn việc sau. Một là nguy cơ bạn đời tự tử tăng. Hai là nguy cơ cha mẹ phải vào viện tâm thần sau khi con tự tử tăng. Ba là nguy cơ người mẹ tự tử sau khi người con đã trưởng thành tự tử tăng. Bốn là nguy cơ trẻ bị trầm cảm sau khi cha mẹ tự tử tăng. So với người mất người thân vì bạo lực chết vì nguyên nhân khác, người mất vì tự tử thường nói nhiều hơn rằng mình bị người ta lạnh nhạt, thấy xấu hổ +- Mức bằng chứng: A +- Nguồn: Erlangsen A, et al. (2017). Association Between Spousal Suicide and Mental, Physical, and Social Health Outcomes: A Longitudinal and Nationwide Register-Based Study. JAMA Psychiatry. ；Pitman A, et al. (2014). Effects of suicide bereavement on mental health and suicide risk. The Lancet Psychiatry. ；Kristensen P 等 (2012). Bereavement and mental health after sudden and violent losses: a review. Psychiatry. （bản tổng quan trong phần Ghi chú） +- Ghi chú: Cái chết đột ngột, bạo lực là tai nạn, bị giết, thảm hoạ. Sau loại chết này, tỷ lệ rối loạn sau chấn thương tâm lý, trầm cảm và nỗi buồn bị kẹt đều cao hơn sau khi chết bệnh. Hồi phục cũng chậm hơn. Khi bản thân có ý nghĩ tự tử thì xem mục 25 ở phần 1 + +### 5. Trong nửa năm đầu sau khi mất bạn đời, giao việc canh bữa cơm, việc ngủ, thuốc bệnh mạn tính và lịch tái khám cho một người cố định + +- Chi phí: Không tốn tiền. Tìm một người cụ thể, con cái, anh chị em hay bạn đều được, bảo người đó định kỳ đến nhà hoặc gọi điện. Khó ở chỗ phải mở lời giao việc này +- Nói thẳng: Trong nửa năm đầu sau khi mất bạn đời, xác suất tử vong cùng kỳ cao hơn người chưa mất bạn đời khoảng bốn phần mười. Sau nửa năm giảm còn cao hơn khoảng một phần rưỡi. Ở nam cao hơn khoảng hai phần ba, ở nữ nhỏ đến mức có thể chỉ là trùng hợp. Khoảng cách này không phải tự nghĩ thông là vượt qua được. Việc cần làm là ghim việc uống thuốc, ăn cơm và tái khám vào một người cụ thể, đừng trông chờ trí nhớ của mình +- Lợi ích: 15 nghiên cứu ghi sẵn rồi theo dõi tiếp, 2.263.888 người gộp lại. Trong 6 tháng đầu sau khi mất bạn đời, RR tử vong là 1.41 (95% CI 1.26–1.57), tức xác suất tử vong cùng kỳ cao hơn người chưa mất bạn đời khoảng bốn phần mười. Sau 6 tháng là 1.14 (1.10–1.18), giảm còn cao hơn khoảng một phần rưỡi. Ở nam RR 1.23 (1.18–1.28), cao hơn khoảng hai phần ba. Ở nữ là 1.04 (1.00–1.08, không có ý nghĩa thống kê, tức khoảng cách này có thể chỉ là trùng hợp). Dưới 65 tuổi và từ 65 tuổi trở lên không thấy khác biệt rõ. Một phân tích khác gộp 123 bài báo, 1377 ước lượng nguy cơ, hơn 500.000.000 người. Sau khi cân bằng khác biệt điều kiện, HR 1.23 (1.19–1.28), tức nói chung cao hơn khoảng hai phần ba. Nam còn cao hơn nữ. Ở nam 1.27 (1.19–1.35), cao hơn hai phần bảy. Ở nữ 1.15 (1.08–1.22), cao hơn một phần rưỡi +- Mức bằng chứng: A +- Nguồn: Moon JR, et al. (2011). Widowhood and mortality: a meta-analysis. PLoS ONE. ；Shor E, et al. (2012). Widowhood and mortality: a meta-analysis and meta-regression. Demography. +- Ghi chú: Đây là nghiên cứu nhìn lại dữ liệu nhóm người, không gỡ được yếu tố khác lẫn vào. Vợ chồng vốn đã giống nhau về lối sống và bệnh tật. Nhưng quy luật «nửa năm đầu cao hơn nhiều so với sau đó» giống nhau trong cả hai phân tích tổng hợp độc lập, đáng coi là một giai đoạn cần có người trông. Nam có nguy cơ cao hơn và ít kêu cầu giúp đỡ hơn. Nhà có người lớn tuổi vừa mất bạn đời thì theo phần 17, làm cho rõ tài khoản, giám hộ và di chúc. Không tìm được người như vậy (sống một mình, không còn con, con cái cũng không còn) thì giao việc này cho cộng đồng và điện thoại, xem mục 6 của phần này +### 6. Không có thân nhân cũng không có bạn, hãy thay «người trông hộ bạn» bằng ba thứ: hàng xóm có vào được nhà bạn, danh sách thăm hỏi của cộng đồng, và người liên hệ khẩn cấp trong điện thoại + +- Chi phí: Không tốn tiền. Khó ở chỗ phải tự mở lời, nói với hàng xóm và ủy ban nhân dân một câu «tôi sống một mình, trong nhà vừa xảy ra chuyện» +- Nói thẳng: Khi không có thân nhân cũng không có bạn thì cần trông hai việc: có người phát hiện bạn khi xảy ra chuyện, và ăn uống dùng thuốc đúng giờ hằng ngày. Ăn uống dùng thuốc thì tự lo, viết thành danh sách, đặt báo thức, đừng trông vào trí nhớ. Có người phát hiện thì dựa vào ba thứ: để lại cho hàng xóm một chìa khoá dự phòng, hỏi ủy ban nhân dân xem có thể vào danh sách thăm hỏi chăm sóc không, và đặt sẵn người liên hệ khẩn cấp trong điện thoại, đừng tắt máy +- Lợi ích: 44.573 bệnh nhân ngoại trú, tất cả có nguy cơ hoặc tiền sử huyết khối mạch máu do xơ vữa (bệnh mạch máu có mảng bám trong lòng mạch, dễ bị cục máu đông nghẽn). Họ được bác sĩ nhiều nước đăng ký vào chương trình quốc tế REACH để theo dõi, nhập mẫu tháng 12 năm 2003 đến tháng 12 năm 2004, theo dõi 4 năm. Trong đó 8.594 người (19%) sống một mình. Người sống một mình, trong 4 năm, tử vong do mọi nguyên nhân là 14.1%, người ở cùng nhà là 11.1%. Tử vong tim mạch 8.6% so với 6.8%. Hai chỉ số log-rank P<0.01, khác biệt khó có thể là trùng hợp. Ở các nhóm tuổi khác nhau thì mức chênh lệch cũng khác nhau (P tương tác=0.03). Từ 45 đến 65 tuổi, tử vong người sống một mình so với người ở cùng nhà là 7.7% so với 5.7%. Sau khi cân bằng các yếu tố khác, HR 1.24 (95% CI 1.01–1.51), tức cao hơn khoảng hai phần tư. Từ 66 đến 80 tuổi là 13.2% so với 12.3%, HR 1.12 (1.01–1.26), cao hơn khoảng một phần mười. Trên 80 tuổi là 24.6% so với 28.4%, HR 0.92 (0.79–1.06), không thấy tăng. Ý kiến năm 2022 của Bộ Dân vụ cùng mười bộ khác quy định phải triển khai dịch vụ thăm hỏi chăm sóc. Đối tượng là người cao niên sống một mình, không có con, con ở lại quê, mất khả năng tự chăm sóc, khuyết tật nặng, gia đình có con đông theo kế hoạch sinh và các trường hợp đặc biệt. Hình thức dịch vụ là «thăm hỏi tại nhà theo định kỳ, điện thoại, hình ảnh, giám sát từ xa». Chính quyền xã, phường điều tra, ủy ban nhân dân xóm, khu phố hỗ trợ, nắm thông tin cơ bản và «ý muốn nhận dịch vụ thăm hỏi chăm sóc», rồi đưa vào quản lý lưới ở cấp cơ sở, tức chia theo khu vực và giao người phụ trách. Đến cuối năm 2025, «bảo đảm tỷ lệ thăm hỏi hằng tháng đạt 100% với người cao niên khó khăn đặc biệt», «với người cao niên khó khăn đặc biệt mất khả năng tự chăm sóc thì mỗi tháng thăm hỏi chăm sóc không ít hơn một lần». Văn bản còn khuyến nghị trang bị trong gia đình hệ thống gọi cầu thông minh, đồng hồ nước thông minh, sản phẩm theo dõi sức khoẻ, thiết bị giám sát người cao niên, «một khi phát hiện bất thường thì cảnh báo kịp thời và đồng thời gửi thông báo cho người liên hệ khẩn cấp». Nhân viên phụ vụ phát hiện vấn đề khẩn cấp thì «phải giúp gọi số cấp cứu ngay lập tức» +- Mức bằng chứng: B +- Nguồn: Udell JA, et al. (2012). Living alone and cardiovascular risk in outpatients at risk of or with atherothrombosis. Archives of Internal Medicine. ；Bộ Dân vụ và mười bộ khác (2022). Ý kiến hướng dẫn về việc triển khai dịch vụ thăm hỏi chăm sóc người cao niên khó khăn đặc biệt（民发〔2022〕73 号）. +- Ghi chú: Cụ thể cách dùng ba thứ đó. Một, đưa chìa khoá dự phòng hoặc mật khẩu tạm của ổ khóa cho một người hàng xóm hoặc bảo vệ, nói với người đó một câu «mấy ngày này tôi ở một mình». Không cần thân tình, chỉ cần có người biết bạn ở trong nhà. Hai, đến ủy ban nhân dân cộng đồng nói mình sống một mình, hỏi xem có thể vào danh sách thăm hỏi chăm sóc không. Trong số người cao niên thì người sống một mình, không có con, mất con là nhóm được chính thức xác định làm đối tượng dịch vụ, tiện thể hỏi xem địa phương có lắp miễn phí thiết bị gọi cầu thông minh, đồng hồ nước thông minh loại đó không. Ba, đặt sẵn người liên hệ khẩn cấp và thẻ cấp cứu y tế trong điện thoại, mấy ngày đó đừng để im hay tắt máy. Cho B vì hai lý do. Một là liên hệ giữa sống một mình và tỷ lệ tử vong đến từ nghiên cứu nhìn lại dữ liệu nhóm người, không gỡ được cách giải thích ngược rằng người vốn sức khoẻ kém, điều kiện kém thì dễ sống một mình hơn, chính bài gốc cũng viết kết quả này còn chờ xác nhận. Hai là văn bản về dịch vụ thăm hỏi chăm sóc chỉ phủ người cao niên. Người chưa đầy 60 tuổi không có chế độ riêng tương ứng. Có thể trông chờ là một câu trong văn bản đó, nó yêu cầu cán bộ lưới và nhân viên xã hội «kịp thời phát hiện rủi ro khủng hoảng tâm lý như biến cố gia đình, thất nghiệp, mất đi cơ hội học» (xem mục 11 của phần này, muốn có người để nói chuyện thì gọi 12356). «Có người ở trong nhà» đáng giá bao nhiêu thì xem mục 1 ở phần 13 (ép ngực). Ngừng tim xảy ra ngoài bệnh viện thì 79.2% xảy ra ở nhà. Có người bên cạnh làm ép ngực thì tỷ lệ xuất viện còn sống là 16.1%, không có người thì chỉ 3.9%. Trong nhà không có người nghĩa là rơi vào nhóm thấp. Thêm hai việc nữa làm tiện tay. Bản thân việc cưới tang cũng phải chạy nhiều cửa sổ ở nhà tang lễ, trạm y tế cộng đồng, công an, quỹ tích góp nhà ở (xem phần 25). Xếp việc đó thành lịch cố định mỗi ngày ra ngoài một lần thì các việc cần đi trong mấy ngày đó đã có sẵn. Đường dây nóng gọi nhiều lần được, không phải chỉ gọi một lần (xem mục 11 của phần này, muốn có người để nói chuyện thì gọi 12356). Bản thân việc «sống một mình» cũng là một yếu tố nguy cơ dài hạn. Tỷ số tử vong của người sống một mình là 1.32, cao hơn khoảng ba phần mười, xem mục 10 ở phần 22 (coi việc gặp gỡ định kỳ với người khác là một khoản chi cho sức khoẻ). Nhưng ở đây chỉ nói mấy ngày trước mắt, sổ sức khoẻ dài hạn thì tính riêng + +### 7. Sau khi cha mẹ qua đời, nói thật với con, để con dự lễ tang, đừng đưa con đi nơi khác để «đổi không khí» + +- Chi phí: Không tốn tiền. Khó ở chỗ bản thân còn đang khó chịu, vẫn phải nói rõ với con +- Nói thẳng: Trẻ mất cha hoặc mẹ trước 18 tuổi, khi bước vào đầu tuổi trưởng thành có xác suất tử vong cao hơn trẻ chưa từng gặp cảnh đó khoảng năm phần mười. Cha mẹ chết vì tai nạn, tự tử thì cao hơn khoảng tám phần mười, chết vì bệnh thì cao hơn khoảng ba phần mười. Việc nhà làm được là nói thật với con, để con dự lễ tang, không cắt ngang việc học, không đưa con sang nhà bà bải để đổi không khí +- Lợi ích: Toàn bộ dân số sinh ở Đan Mạch giai đoạn 1968 đến 2008 và ở Thụy Điển giai đoạn 1973 đến 2006. Cộng thêm 89.3% người sinh ở Phần Lan giai đoạn 1987 đến 2007. Tổng cộng hơn 7.300.000 người. Trong đó 189.094 người mất cha mẹ trước 18 tuổi. Nhóm trẻ này có tỷ lệ tử vong do mọi nguyên nhân cao hơn 50% so với người chưa mất người thân (tỷ số tử vong 1.50, 95% CI 1.43–1.58, tức khoảng 1.5 lần). Cha mẹ chết vì nguyên nhân không tự nhiên như tai nạn, tự tử thì là 1.84 (1.71–2.00), cao hơn khoảng tám phần mười. Chết vì nguyên nhân tự nhiên như bệnh thì là 1.33 (1.24–1.41), cao hơn khoảng ba phần mười. Cao nhất khi nguyên nhân chết của trẻ cùng loại với nguyên nhân chết của cha mẹ +- Mức bằng chứng: B +- Nguồn: Li J, et al. (2014). Mortality after parental death in childhood: a nationwide cohort study from three Nordic countries. PLoS Medicine. +- Ghi chú: Khi nói với con đừng bịa «đi công tác xa», việc giám hộ và nuôi dưỡng phải viết ra giấy. Cho B không phải vì con số yếu. Đây là nghiên cứu toàn bộ dân số của ba nước, bản thân số liệu đủ A. Cho B là vì từ con số này đến cách làm cụ thể «nói với con thế nào, có nên cho con dự lễ tang không» không có bằng chứng thử nghiệm trực tiếp. Tác giả cũng viết rõ là thiếu dữ liệu về quan hệ cha mẹ con và lối sống của trẻ sau khi mất người thân. Mức tăng vừa có phần do thể chất sẵn có, vừa có phần do ảnh hưởng dài hạn của việc mất người thân. Chỉ định giám hộ và sắp xếp tài sản xem phần 17. Cấp dưỡng trẻ và chuyện đi học xem phần 18 + +### 8. Nỗi buồn kéo dài quá nửa năm vẫn đứng yên, ngày nào cũng không nổi, thì đi đăng ký khoa tâm thần hoặc tâm lý lâm sàng + +- Chi phí: Phí đăng ký khám cộng phí điều trị. Loại điều trị này làm theo liệu trình, trong thử nghiệm thì trung bình 19 tuần làm 16 buổi. Khó ở chỗ phải tự bước ra đăng ký số khám đó +- Nói thẳng: Cứ khoảng mười người trưởng thành mất người thân thì có một người bị kẹt: nửa năm trở lên vẫn nhớ đến mức ăn không nổi, không đi làm được, không chăm con nổi, người tê dại, thấy đời không có ý nghĩa. Đây không phải vì chưa đủ vững lòng, nó có tiêu chuẩn chẩn đoán và có điều trị riêng. Điều trị có mục tiêu hiệu quả 51%, tâm lý trị liệu thường là 28%, mà có tác dụng nhanh hơn +- Lợi ích: 14 nghiên cứu gộp lại, đối tượng là người trưởng thành mất người thân không phải vì bạo lực và chưa từng khám tâm thần. Tỷ lệ mắc rối loạn đau buồn kéo dài là 9.8% (95% CI 6.8–14.0), tức cứ 100 người trưởng thành mất người thân thì khoảng 10 người. Tuổi càng cao thì tỷ lệ càng lớn. Một nghiên cứu khác kiểm tra bộ tiêu chuẩn chẩn đoán này có chính xác không, chọn 291 người mất người thân, phỏng vấn ba lần ở 0–6, 6–12, 12–24 tháng sau khi mất. Biểu hiện cốt lõi của chẩn đoán là nỗi nhớ mãnh liệt về người đã mất. Phải cộng thêm ít nhất năm trong chín biểu hiện kia. Năm biểu hiện đầu là tê dại về cảm xúc, sốc, thấy đời không có ý nghĩa, không tin tưởng ai, đau khổ vì mất người đó. Bốn biểu hiện sau là không thể chấp nhận, không nhận ra mình phải là ai, tránh né thực tế, không thể sống tiếp. Các biểu hiện này phải xuất hiện mỗi ngày, hoặc nặng đến mức không làm được việc gì. Và vẫn còn sau khi người thân qua đời ít nhất 6 tháng, vẫn ảnh hưởng đến đời sống hằng ngày. Một thử nghiệm khác có 95 người rút thăm chia thành hai nhóm, làm 16 buổi điều trị, trung bình 19 tuần. Điều trị cho đau buồn phức tạp hiệu quả 51%. Tâm lý trị liệu quan hệ thông thường là 28% (P=0.02, khác biệt khó là trùng hợp), và có tác dụng nhanh hơn. Số người cần điều trị là 4.3, tức trung bình cứ điều trị 4.3 người thì có thêm một người khỏi +- Mức bằng chứng: A +- Nguồn: Lundorff M, et al. (2017). Prevalence of prolonged grief disorder in adult bereavement: A systematic review and meta-analysis. Journal of Affective Disorders. ；Prigerson HG, et al. (2009). Prolonged grief disorder: Psychometric validation of criteria proposed for DSM-V and ICD-11. PLoS Medicine. ；Shear MK, et al. (2005). Treatment of complicated grief: a randomized controlled trial. JAMA. +- Ghi chú: Đừng tự đếm số tháng. Cách đánh giá đúng là «đã ảnh hưởng tới ăn, ngủ, đi làm, chăm con», khớp thì đi đăng ký khám. Khoa gọi là tâm thần, tâm lý hoặc tâm lý lâm sàng, bệnh viện chuyên khoa tâm thần công lập đều có. Chẩn đoán này trong tài liệu khác gọi là đau buồn phức tạp, cũng gọi là rối loạn đau buồn kéo dài. Các tiêu chuẩn khác nhau đòi kéo dài bao lâu cũng khác nhau, có bản viết 6 tháng, có bản viết 12 tháng. Con số 9.8% đến từ 14 nghiên cứu làm khác nhau khá xa, tác giả cũng nhắc nó có thể không chuyển được sang nhóm người khác + +### 9. Đừng vội trả tiền làm tư vấn đau buồn, xem trước nỗi buồn của mình có thật sự bị kẹt không (các triệu chứng ở mục 8) + +- Chi phí: Không tốn tiền, ngược lại còn tiết kiệm được một khoản phí tư vấn +- Nói thẳng: Sắp xếp tâm lý trị liệu cho tất cả người mất người thân thì lợi ích rất nhỏ, đợi một thời gian nữa không còn thấy nữa, vì phần lớn tự khỏi. Nhưng với người đã rõ ràng không hợp thích nổi thì điều trị hiệu quả ngang với điều trị các vấn đề tâm lý khác. Vậy nên đối chiếu với mục 8 (đau buồn bị kẹt thì đi khoa tâm thần), khớp rồi hãy bỏ khoản tiền này +- Lợi ích: 61 nghiên cứu đối chứng gộp lại. Can thiệp tâm lý dành cho mọi người mất người thân thì lợi ích rất nhỏ ngay khi kết thúc điều trị, đợi thêm một thời gian thì không còn thấy lợi ích. Chỉ nhận người mất người thân «đã rõ ràng không hợp thích nổi» thì can thiệp có hiệu quả tốt ngang tâm lý trị liệu cho các vấn đề tâm lý khác. Nghiên cứu không sàng lọc trước cho kết quả kém, vì nhóm không được can thiệp vốn cũng dần tự khỏi theo thời gian. Ngoài ra có 14 thử nghiệm chia nhóm ngẫu nhiên gộp lại: điều trị cho người đã có vấn đề thì rõ ràng tốt hơn nhóm đối chứng ngay sau điều trị và cả ở lần theo dõi sau đó, để thêm thời gian thì hiệu quả còn tăng. Can thiệp phòng ngừa làm đại trà cho người chưa có vấn đề thì không có tác dụng. Một nghiên cứu khác theo dõi 205 người từ trước khi mất bạn đời cho tới 6 và 18 tháng sau. Nghiên cứu này chia ra năm đường đi. Lần lượt là đau buồn thường, đau buồn kéo dài, trầm cảm kéo dài, ngược lại là khỏe lên trong thời gian mất người thân, và sức bền (chịu được). Trong đó nhóm sức bền phổ biến nhất, nhóm đau buồn thường tương đối ít +- Mức bằng chứng: A +- Nguồn: Currier JM, et al. (2008). The effectiveness of psychotherapeutic interventions for bereaved persons: a comprehensive quantitative review. Psychological Bulletin. ；Wittouck C, et al. (2011). The prevention and treatment of complicated grief: a meta-analysis. Clinical Psychology Review. ；Bonanno GA, et al. (2002). Resilience to loss and chronic grief: a prospective study from preloss to 18-months postloss. Journal of Personality and Social Psychology. ；Stroebe M 等 (2007). Health outcomes of bereavement. Lancet. （bản tổng quan trong phần Ghi chú） +- Ghi chú: Ý là «không phải ai cũng cần», không phải «ai cũng đừng làm». Nhóm mất người thân nguy cơ cao như mục 4 (người thân chết vì tự tử, tai nạn hoặc bị giết) và nhóm đã bị kẹt như mục 8 (đau buồn bị kẹt thì đi khoa tâm thần), chính là nhóm có bằng chứng ủng hộ việc nên làm. Kết luận của một bài tổng quan trên Lancet cũng vậy: đau buồn không phải bệnh, phần lớn không cần can thiệp chuyên môn. Sức lực nên dồn vào nhóm nguy cơ cao. Cũng nên dồn vào người đã có đau buồn phức tạp, trầm cảm và rối loạn sau chấn thương tâm lý sau khi mất người thân +### 10. Sau khi sống xa nhau hoặc ly hôn, đừng để bảo hiểm y tế bị gián đoạn, đừng dựa vào rượu để chịu đựng + +- Chi phí: Không tốn tiền. Gia hạn bảo hiểm y tế, bỏ rượu. Khó ở chỗ khoảng thời gian này dễ dựa vào rượu để chịu đựng +- Nói thẳng: Người sống xa nhau hoặc ly hôn có xác suất chết sớm cao hơn rõ rệt người còn hôn nhân, nam và người trẻ hơn tăng nhiều hơn. Điều này chưa chắc do ly hôn gây ra, người ly hôn vốn đã có thể khác nhau về sức khoẻ và thu nhập. Có hai việc mình tự kiểm soát được: đừng để bảo hiểm y tế và khám sức khoẻ định kỳ bị gián đoạn, đừng dựa vào thuốc lá và rượu để chịu đựng +- Lợi ích: 32 nghiên cứu ghi sẵn rồi theo dõi tiếp gộp lại, bao phủ hơn 6.500.000 người, 160.000 ca tử vong, hơn 755.000 ca ly hôn, 11 quốc gia. So với người còn hôn nhân, người trưởng thành sống xa nhau hoặc ly hôn có nguy cơ chết sớm tăng rõ rệt. Nam và người trẻ hơn tăng rõ hơn nữ và người lớn tuổi. Tác giả bàn bốn đường có thể. Một là lựa chọn xã hội, người vốn đã khác nhau về sức khoẻ hay điều kiện thì dễ đi đến ly hôn hơn. Hai là gián đoạn nguồn lực, thu nhập và sự bảo đảm bị đứt. Ba là thay đổi hành vi sức khoẻ, thuốc lá, rượu, giờ ngủ xấu đi. Bốn là áp lực tâm lý mạn tính (căng thẳng kéo dài) +- Mức bằng chứng: A +- Nguồn: Sbarra DA, et al. (2011). Divorce and Death: A Meta-Analysis and Research Agenda for Clinical, Social, and Health Psychology. Perspectives on Psychological Science. +- Ghi chú: Bài gốc chỉ nói «tăng rõ rệt» và so sánh giữa các nhóm, không đưa ra hệ số có thể trích dẫn, nên ở đây không viết con số cụ thể. Cỡ lợi ích đặt là vừa, theo cách «biết hướng, không có mức giảm nào để áp». Tài sản, nợ và tiền cưới lúc ly hôn xem phần 8 và phần 10. Cách sắp xếp nuôi con xem phần 18 + +### 11. Muốn có người để nói chuyện thì gọi 12356, người chưa thành niên và thanh niên gọi 12355, muốn khám bác sĩ thì đăng ký khám tâm lý ngoại trú + +- Chi phí: Không tốn tiền. Cả hai đường dây nóng đều miễn phí. Khám tâm lý ngoại trú thì đăng ký như khám ngoại trú thường +- Nói thẳng: Muốn tìm người để nói chuyện thì gọi 12356, đây là đường dây nóng hỗ trợ tâm lý toàn quốc, miễn phí. Người chưa thành niên và thanh niên có 12355, nền tảng dịch vụ tổng hợp cho thanh niên. Muốn khám bác sĩ thì đăng ký khám tâm lý ở khoa tâm thần, khoa tâm lý hoặc khoa tâm lý lâm sàng. Khi có biến cố gia đình, thất nghiệp, mất đi cơ hội học thì cán bộ lưới cộng đồng và nhân viên xã hội vốn phải chủ động phát hiện +- Lợi ích: Phương án năm 2026 của Ủy ban Quốc gia Y tế và Sức khoẻ cùng 25 bộ yêu cầu hoàn thiện chế độ vận hành đường dây nóng hỗ trợ tâm lý 12356. Phải bảo đảm vận hành ổn định, nâng tỷ lệ cuộc gọi kết nối được. Phải thúc đẩy xây dựng nền tảng thông tin 12356 cấp tỉnh. Phải tăng cường xây dựng cơ chế phối hợp với các đường dây nóng dịch vụ như 110. Còn phải làm tốt nền tảng dịch vụ tổng hợp cho thanh niên 12355, và phối hợp với 12356. Đến năm 2030 có ba mục tiêu. Một là hơn 80% thôn xóm và cộng đồng đô thị có phòng tư vấn tâm lý hoặc phòng làm việc xã hội và có cung cấp dịch vụ tâm lý. Hai là mỗi cấp huyện (quận) có ít nhất 1 bệnh viện chuyên khoa tâm thần công lập có dịch vụ khám tâm lý ngoại trú. Ba là mỗi tỉnh xây xong nền tảng hỗ trợ tâm lý 12356 cấp tỉnh. Đội ngũ can thiệp khủng hoảng tâm lý và hỗ trợ tâm lý gồm bác sĩ tâm thần, kỹ thuật viên trị liệu tâm lý, nhân viên tư vấn tâm lý, chuyên gia công tác xã hội và tình nguyện viên. Phương án còn yêu cầu ba nhóm người lưu ý và phát hiện sớm. Ba nhóm này là cán bộ lưới (người phụ trách theo khu vực trong cộng đồng), chuyên gia công tác xã hội và tình nguyện viên. Họ phải kịp thời phát hiện rủi ro khủng hoảng tâm lý như biến cố gia đình, thất nghiệp, mất đi cơ hội học (toàn quốc, ban hành tháng 3 năm 2026) +- Mức bằng chứng: B +- Nguồn: Ủy ban Quốc gia Y tế và Sức khoẻ cùng 25 bộ (2026). Thông báo về ban hành phương án hoàn thiện hệ thống dịch vụ tâm lý xã hội và cơ chế can thiệp khủng hoảng（国卫医政发〔2026〕8 号）. +- Ghi chú: Cho B vì phòng tư vấn ở xóm và cộng đồng, phòng khám tâm lý cấp huyện đều là mục tiêu đến năm 2030 mới phải xây, không phải hiện giờ nơi nào cũng có. Còn 12356 và 12355 thì gọi được ngay bây giờ. Thời điểm mở 12356 và mỗi ngày nghe trong bao lâu thì xem mục 25 ở phần 1. Tư vấn tâm lý và trị liệu tâm lý không phải một. Chẩn đoán và kê đơn thuốc phải do bác sĩ tâm thần thực hiện, điều trị phải làm ở cơ sở y tế có giấy phép. Loại «tư vấn chữa lành» và «tư vấn tâm linh» tính tiền theo giờ bán ngoài không nằm trong hệ thống này + +### 12. Trong ba tháng đầu sau biến cố, mọi quyết định lớn không thể đảo ngược đều đẩy lùi + +- Chi phí: Không tốn tiền, chỉ là đẩy quyết định lùi, có thể vì thế bỏ lỡ một số cơ hội. Khó ở chỗ lúc đó người ta thấy cơ hội không đợi mình +- Nói thẳng: Khoảng thời gian này khả năng phán đoán của bạn với mọi thứ được gọi là «cơ hội» là tệ nhất. Người nói rằng liên lạc được với người đã mất, người nói rằng giúp bạn lấy lại số tiền, người khuyên bạn nghỉ việc đi nơi khác làm lại, người rủ bạn góp vốn, đều sẽ tìm đến. Việc nào số tiền vượt quá thu nhập một tháng, hoặc phải trả giá lớn nếu hối hận, đều dời lùi ba tháng. Kể cho một người không có lợi ích gì với việc này nghe trước, đừng quyết trong ngày +- Lợi ích: Tránh việc trong lúc vừa chịu đòn, phán đoán kém nhất, phải quyết định tài sản mà không thể hối hận. Những việc đó gồm bán nhà, nghỉ việc, đem một lần toàn bộ tiền bồi thường hay tiền đền bù đi đầu tư, cưới lại vội, đứng bảo lãnh cho người khác. Số tiền của những quyết định này thường từ vài chục nghìn tệ đến mức một căn nhà, vì vậy cỡ lợi ích đặt là vừa. Văn bản chính thức cũng xếp «biến cố gia đình, thất nghiệp, mất đi cơ hội học» vào loại rủi ro khủng hoảng tâm lý mà cộng đồng phải chủ động rà soát, trạng thái khoảng thời gian đó vốn đã được coi là trạng thái cần người ngoài đưa tay +- Mức bằng chứng: C +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; cách diễn đạt «biến cố gia đình, thất nghiệp, mất đi cơ hội học» như rủi ro khủng hoảng tâm lý xem Ủy ban Quốc gia Y tế và Sức khoẻ cùng 25 bộ (2026). Phương án hoàn thiện hệ thống dịch vụ tâm lý xã hội và cơ chế can thiệp khủng hoảng（国卫医政发〔2026〕8 号）. +- Ghi chú: Nếu bên cạnh không có người không có lợi ích gì với việc này thì gọi 12356 kể một lần (xem mục 11 của phần này, muốn có người để nói chuyện thì gọi 12356), hoặc tự viết ra rồi ba ngày sau đọc lại. «Ba tháng» là ranh giới do tác giả tự đặt, không có tài liệu nào chống đỡ thời hạn cụ thể này, nên ghi mức C. Những trò lừa đảo thường gặp nhắm vào người mất người thân và người già xem phần 17 (đầu tư chu cấp thu hưu, vay mua nhà bằng tiền hưu trí) và phần 6 (bói quẻ). Bẫy của việc đứng bảo lãnh và giấy nợ xem phần 8. Có vài khoản tiền phải chủ động đi làm, không nằm trong danh sách «đẩy lùi»: số dư quỹ tích góp nhà ở, quyền lợi bảo hiểm xã hội, ba khoản tiền khi chết vì tai nạn lao động, xem phần 25 và phần 19 + +### 13. Đừng dùng cái chết để trả nợ: bảo hiểm nhân thọ trong hai năm không bồi thường, tai nạn lao động không được công nhận, nợ vẫn bị trừ trước tiên từ di sản + +- Chi phí: Không tốn tiền, chỉ là làm cho số sách này rõ ràng +- Nói thẳng: Bảo hiểm nhân thọ không bồi thường khi tự tử trong hai năm đầu, chỉ hoàn lại số tiền trong hợp đồng mà được hoàn. Tai nạn lao động cũng không được công nhận, ba khoản tiền khi chết vì tai nạn lao động không có đồng nào. Nợ nợ bị trừ trước tiên từ di sản bạn để lại, người nhà chỉ nhận phần còn lại. Họ còn phải gánh sức khoẻ mà mục 4 của phần này nói (mất người thân vì tự tử hoặc bạo lực) +- Lợi ích: Điều 44 của Bộ luật Bảo hiểm quy định về hợp đồng có điều kiện trả tiền bảo hiểm khi người được bảo hiểm chết. Tính từ ngày hợp đồng thành lập hoặc ngày hiệu lực được khôi phục, nếu người được bảo hiểm tự tử trong hai năm thì công ty bảo hiểm không phải nghĩa vụ trả tiền bảo hiểm. Không tính trường hợp người được bảo hiểm tự tử khi chưa có năng lực hành vi dân sự, tức người mà pháp luật không cho chịu trách nhiệm về hành vi của chính mình. Trong trường hợp đó công ty bảo hiểm phải hoàn trả giá trị tiền mặt của hợp đồng theo thỏa thuận, tức chỉ hoàn lại số tiền trong hợp đồng được hoàn, không trả số tiền bảo hiểm. Mức bảo hiểm nhân thọ thường ở cỡ vài chục nghìn tệ, vì vậy cỡ lợi ích đặt là lớn. Điều 16, khoản ba của Quy chế bảo hiểm tai nạn lao động quy định: tự tàn tật hoặc tự tử thì không được công nhận là tai nạn lao động, cũng không được coi là tai nạn lao động. Ba khoản quyền lợi khi chết vì tai nạn lao động ở Điều 39 vì thế mất trắng cả ba. Một là trợ cấp mai táng, trả bằng 6 tháng mức lương trung bình của người lao động năm trước tại địa phương. Hai là tiền tùy táng người nuôi dưỡng. Ba là trợ cấp một lần khi chết vì tai nạn lao động, bằng 20 lần thu nhập chi tiêu bình quân đầu người dân đô thị toàn quốc năm trước. Điều 1159 của Bộ luật Dân sự quy định: khi chia di sản thì phải trừ các khoản thuế và nợ mà người để lại có nghĩa vụ theo pháp luật, tức lấy di sản trả thuế và nợ trước, phần còn lại mới chia. Điều 1161 quy định: người thừa kế trả nợ trong giới hạn giá trị thực tế của di sản được nhận, tức chỉ dùng phần được thừa kế để trả, không phải bỏ thêm ra túi mình; người từ bỏ quyền thừa kế thì không phải chịu trách nhiệm trả nợ +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (2015 修正). Bộ luật Bảo hiểm nước Cộng hoà nhân dân Trung Quốc（中华人民共和国保险法第四十四条）. ；Hội đồng Nhà nước (2010 修订). Quy chế bảo hiểm tai nạn lao động（工伤保险条例，国务院令第 586 号，第十六条、第三十九条）. ；Quốc hội (2020). Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（中华人民共和国民法典第一千一百五十九条、第一千一百六十一条）. +- Ghi chú: Ở đây chỉ tính sổ sức khoẻ về tiền. Các mục khác của phần này và của phần 1 nói là đừng đi đến bước này. Điều 44 của Bộ luật Bảo hiểm chỉ nói về hai năm đầu, viết ra là để nêu đủ điều khoản, không phải để đặt ra một mốc thời gian. Sức khoẻ của người nhà xem mục 4 của phần này (vợ chồng có người chết vì tự tử thì trong 5 năm đầu nguy cơ mắc rối loạn tâm thần mới cao hơn bảy đến tám phần mười). Khi có ý nghĩ tự tử thì làm gì xem mục 25 và 32 ở phần 1, di chứng còn lại sau khi được cứu xem mục 33 ở phần 1. Mức ba khoản tiền khi chết vì tai nạn lao động xem mục 16 ở phần 19. Thủ tục cụ thể về di sản và nợ xem mục 9 ở phần 25 (lấy từng khoản quỹ tích góp nhà ở và quyền lợi bảo hiểm xã hội) \ No newline at end of file diff --git "a/vi/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" "b/vi/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" new file mode 100644 index 00000000..fd80b982 --- /dev/null +++ "b/vi/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" @@ -0,0 +1,98 @@ +[← Về mục lục](../README.md) + +# 32. Đi học ở nước ngoài: thị thực, việc làm, bảo hiểm, chứng nhận + +Phần này tính hai khoản: tiền và tự do cá nhân. Nó không bàn chuyện có nên đi nước ngoài hay không, cũng không so trường nào tốt hơn. Bốn việc dưới đây mà không biết thì sẽ thiệt: tư cách thị thực thế nào mới coi là không bị đứt, số giờ làm thêm tối đa là bao nhiêu, những khoản bảo hiểm và phí nào bắt buộc phải trả, và tấm bằng này khi về nước có được công nhận không. + +Phần này bao gồm bốn nước: Hoa Kỳ, Canada, Anh và Úc, số liệu liệt kê theo từng nước. Nguồn là các trang chính thức của cơ quan nhập cảnh mỗi nước và của Bộ Giáo dục Trung Quốc. **Chính sách du học ở các nước này thay đổi rất thường xuyên, thường xuyên hơn cả quy định của Trung Quốc. Mọi con số trong phần này có thời điểm chốt là tháng 9 năm 2026. Trước khi lên đường và trước mỗi lần gia hạn, hãy mở lại các liên kết trong mục Nguồn và tự kiểm tra thêm một lần.** + +Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra sao, 12308 quản lý được những gì, và mua bảo hiểm y tế và bảo hiểm đưa người bệnh về nước ra sao, xem phần 21. Bài toán nên học hay nên làm, xem phần 23. Bị mất tài khoản, xem phần 14; lừa đảo qua điện thoại, xem phần 8. Những việc đó không viết lại ở đây. + +### 1. Trước khi nộp học phí, tra xem trường có nằm trong danh sách cơ sở được công nhận của Trung tâm Dịch vụ Học bổng không + +- Chi phí: Không tốn tiền. Gõ tên trường một lần vào 「认证院校查询」 (Tra cứu cơ sở được công nhận) của Trung tâm, mất vài phút là xong +- Nói thẳng: Khi về nước, nếu bạn định dùng tấm bằng này để thi công chức, vào doanh nghiệp nhà nước, làm hộ khẩu hay xét chức danh, thì phải qua một bước công nhận trước. Việc công nhận do Trung tâm Dịch vụ Học bổng của Bộ Giáo dục (教育部留学服务中心) làm. Nếu trường không có trong danh sách được công nhận, học phí và thời gian bạn bỏ ra coi như mất trắng, tấm bằng ở trong nước không dùng được. Tra danh sách không tốn tiền, tra xong rồi hãy nộp học phí. +- Lợi ích: Trung tâm Dịch vụ Học bổng của Bộ Giáo dục Trung Quốc có một cửa tra cứu cơ sở được công nhận. Bạn có thể tra theo từng nước, cũng có thể tra theo tên trường. Tra là để biết bằng cấp và học vị mà trường đó cấp có nằm trong phạm vi công nhận hay không. Ngoài ra, Mạng lưới Giám sát và Quản lý Giáo dục có yếu tố nước ngoài của Bộ Giáo dục còn công bố thông tin giám sát đối với giáo dục hợp tác Trung Quốc – nước ngoài và việc mở trường ở nước ngoài +- Mức bằng chứng: A +- Nguồn: Trung tâm Dịch vụ Học bổng của Bộ Giáo dục Trung Quốc（中国（教育部）留学服务中心）. Tra cứu cơ sở được công nhận（认证院校查询）. ；Mạng lưới Giám sát và Quản lý Giáo dục có yếu tố nước ngoài của Bộ Giáo dục（教育部教育涉外监管信息网）. +- Ghi chú: Cái mà đại gia nói miệng là 「Bộ Giáo dục công nhận」 không tính, hãy theo kết quả bạn tự tra trên mạng. Danh sách có thay đổi: lúc nhập học có mặt, lúc tốt nghiệp thì có khi không, nên trong thời gian học cứ tra lại mỗi năm. Chương trình giáo dục hợp tác Trung Quốc – nước ngoài có một hệ thống công nhận và đăng ký riêng. Người nhập học sau năm 2008 có thể dùng họ tên và số chứng minh nhân dân để tra số đăng ký của mình trên mạng giám sát giáo dục có yếu tố nước ngoài. + +### 2. Theo dõi vụ kiện về quy định F-1 mới ở Hoa Kỳ: 「tối đa bốn năm, học xong thì rời đi trong 30 ngày」 dự kiến có hiệu lực ngày 15 tháng 9 năm 2026, bị tòa án đình chỉ một ngày trước đó, hiện vẫn là 「cho tới khi học xong」 + +- Chi phí: Không tốn tiền. Chép ngày kết thúc chương trình trên I-20 vào lịch, đặt nhắc trước 90 ngày. I-20 là giấy chứng nhận đủ điều kiện du học do trường cấp cho bạn. Ngoài ra, mỗi học kỳ xem một lần thông báo của văn phòng sinh viên quốc tế của trường, để biết vụ kiện này có tiến triển gì mới không +- Nói thẳng: Hoa Kỳ định đổi F-1 thành tối đa bốn năm, học xong thì rời đi trong 30 ngày. Một ngày trước khi có hiệu lực, tòa án liên bang đã đình chỉ. Vậy hiện nay vẫn là quy định cũ: học bao lâu thì được ở bao lâu, học xong có 60 ngày để chuẩn bị rời đi. Đình chỉ không phải là bãi bỏ, chính quyền còn quyền kháng nghị, và quy định có thể đổi bất cứ lúc nào. +- Lợi ích: Bộ An ninh nội địa (DHS) công bố một quy định ngày 17 tháng 7 năm 2026, dự kiến có hiệu lực ngày 15 tháng 9 năm 2026. Nội dung là sinh viên F-1 không còn nhập cảnh theo 「cho tới khi học xong」, mà chuyển sang thời hạn cố định, tối đa bốn năm, hết hạn thì phải xin gia hạn riêng. Sau khi học xong hoặc kết thúc thực tập, chỉ còn 30 ngày để chuẩn bị rời đi. Ngày 14 tháng 9 năm 2026, thẩm phán Saylor của Tòa án Liên bang Khu vực bang Massachusetts đã hoãn hiệu lực của quy định đó theo Điều 705 của Đạo luật Thủ tục Hành chính (《行政程序法》). Việc đình chỉ áp dụng toàn quốc và cho cả quy định. Tòa án không bãi bỏ quy định, yêu cầu bãi bỏ bị từ chối, nhưng cho phép nêu lại vào lần sau. Vụ án vẫn đang được xét xử. Theo lệnh của tòa, người mang thị thực F, J và I vẫn nhập cảnh theo 「cho tới khi học xong」, thời hạn ân hạn 60 ngày để rời đi vẫn giữ nguyên, cũng không phải xin gia hạn riêng +- Mức bằng chứng: A +- Nguồn: DHS (2026). Establishing a Fixed Time Period of Admission and an Extension of Stay Procedure for Nonimmigrant Academic Students, Exchange Visitors（công bố 2026-07-17, dự kiến có hiệu lực 2026-09-15）. ；Presidents' Alliance on Higher Education and Immigration v. U.S. Department of Homeland Security, No. 1:26-cv-13799-FDS (D. Mass. Sept. 14, 2026), hồ sơ vụ án. ；Yale Office of International Students & Scholars (2026-09-14). Important Update: Court Action on the D/S Rule. ；AILA (2026). One Day Before Taking Effect, Federal Court Postpones the F, J, and I Fixed Admission Period Rule. ；8 CFR 214.2(f). +- Ghi chú: Mục này có thời điểm chốt là ngày 25 tháng 9 năm 2026. Chính quyền có thể kháng nghị lên Tòa án Phúc án Tiểu bang thứ nhất, và Bộ An ninh nội địa cũng có thể sửa quy định rồi ban hành lại. Người phải học mấy năm thì đừng mặc định quy định cũ sẽ giữ nguyên mãi. Tòa đã ấn định ngày 2 tháng 10 năm 2026 mở phiên tiếp để nghe hai bên báo cáo tiến triển, sau ngày đó xem lại thông báo của trường. Nếu quy định mới sau này có hiệu lực, muốn gia hạn thì phải xin riêng trước khi thời hạn hết, không tự động nối tiếp. Dưới quy định cũ, người bỏ học sớm có thời hạn rời đi khác với người học đủ khóa, hãy hỏi văn phòng sinh viên quốc tế của trường trước. Thị thực J nằm trong cùng quy định đó, chi tiết khác với F-1, hãy tra theo loại thị thực của bạn. Cơ sở pháp lý cho việc hoãn nằm ở mục 705 của Đạo luật Thủ tục Hành chính Hoa Kỳ, cho phép tòa tạm dừng hiệu lực của một quy định. + +### 3. Số giờ làm thêm là vạch đỏ viết cứng ở cả bốn nước Mỹ, Canada, Anh và Úc, vượt một giờ là vi phạm điều kiện thị thực + +- Chi phí: Không tốn tiền. Tự ghi lại số giờ làm theo từng tuần. Chỗ khó là chủ lao động sẽ nói 「làm thêm vài giờ chẳng ai kiểm tra」, câu đó không đáng tin +- Nói thẳng: Cả bốn nước đều đặt trần cho số giờ làm của người mang thị thực sinh viên. Vượt trần là vi phạm điều kiện thị thực, có thể bị hủy thị thực, bị trục xuất, và lần đăng ký sau sẽ bị hỏi về chuyện này. Các trần là: Hoa Kỳ 20 giờ mỗi tuần; Anh 20 giờ mỗi tuần khi học chương trình đại học và 10 giờ mỗi tuần với chương trình dưới bậc đại học; Canada 24 giờ mỗi tuần; Úc 48 giờ mỗi hai tuần. Làm hai chỗ thì phải cộng lại. +- Lợi ích: Với sinh viên F-1 ở Hoa Kỳ, làm việc trong khuôn viên trường khi học kỳ không được quá 20 giờ mỗi tuần. Việc làm bán thời gian ngoài trường đã được duyệt cũng là 20 giờ mỗi tuần. Kỳ nghỉ thì được làm toàn thời gian. Căn cứ là quy định nhập cảnh liên bang Hoa Kỳ 8 CFR 214.2(f)(9). Với thị thực sinh viên của Anh, học chương trình đại học trở lên thì trong học kỳ 20 giờ mỗi tuần, dưới bậc đại học thì 10 giờ mỗi tuần. Chương trình không toàn thời gian thì không được làm việc. Cũng không được tự kinh doanh (làm việc cho chính mình), cũng không được làm vận động viên chuyên nghiệp hay nghề giải trí. Căn cứ là phần dành cho học sinh, sinh viên trong quy tắc nhập cảnh của Anh, phụ lục Student ST26.1、ST26.5. Sinh viên toàn thời gian đủ điều kiện ở Canada được làm ngoài trường tối đa 24 giờ mỗi tuần, theo điểm 186(v) của Quy chế thi hành Luật Nhập cảnh và Người tị nạn (IRPR). Với thị thực sinh viên ở Úc, trong thời gian khóa học còn diễn ra, tối đa 48 giờ mỗi hai tuần. Thạc sĩ nghiên cứu, tiến sĩ và người phụ thuộc của họ không chịu giới hạn này +- Mức bằng chứng: A +- Nguồn: 8 CFR 214.2(f)(9). ；UK Home Office. Immigration Rules Appendix Student（ST26.1、ST26.5）. ；IRCC. Work off campus as an international student. ；Australian Government Department of Home Affairs. Student visa (subclass 500). +- Ghi chú: Việc làm ngoài trường ở Hoa Kỳ phải được duyệt trước. Khi hồ sơ xin giấy phép làm I-765 còn đang chờ xét, bạn không được bắt đầu làm. Giấy phép cũ ở Canada in số 20 giờ mỗi tuần, miễn là bạn vẫn đủ điều kiện thì thực tế làm được 24 giờ, theo quy định hiện hành. Quy định cấm tự kinh doanh ở Anh rất rộng: nhận việc riêng, mở cửa hàng trực tuyến, làm tự do đều tính vào. + +### 4. Gốc của tư cách là 「đang học toàn thời gian」: thời gian nghỉ học, bỏ học, khoảng trống khi chuyển trường đều không được làm thêm + +- Chi phí: Không tốn tiền. Trước khi ngừng lớp, xin nghỉ học hoặc đổi trường, hãy tìm văn phòng sinh viên quốc tế của trường và hỏi rõ tình trạng tư cách của bạn +- Nói thẳng: Bạn có được làm thêm hay không phụ thuộc vào việc bạn có đang học toàn thời gian hay không. Ngay khi nghỉ học, bỏ học, bị đuổi học, hoặc chuyển trường mà không đi học, quyền đi làm công ăn lương mất ngay tại chỗ. Làm tiếp lúc đó là lao động trái phép. Nhiều người nghĩ chừng nào thị thực chưa hết hạn thì không sao, nhưng tư cách đã đứt từ lâu rồi. +- Lợi ích: Canada quy định rõ không được làm ngoài trường trong thời gian nghỉ học đã được duyệt. Trong thời gian chuyển trường mà không học, cũng không được làm ngoài trường. Phải chờ học lại rồi mới bắt đầu làm việc. Ở Hoa Kỳ, quyền làm trong khuôn viên trường chỉ dành cho sinh viên F-1 có tình trạng Active trong SEVIS. SEVIS là hệ thống hồ sơ tư cách của sinh viên quốc tế, Active là đang học và còn hiệu lực. Những sinh viên này còn phải duy trì chương trình học toàn thời gian. Ở Anh, giấy phép làm phụ thuộc vào loại khóa học, chương trình không toàn thời gian thì không được làm +- Mức bằng chứng: A +- Nguồn: IRCC. Work off campus as an international student. ；U.S. Department of Homeland Security, Study in the States. Working in the United States. ；UK Home Office. Immigration Rules Appendix Student（ST26.1）. +- Ghi chú: Khi khối lượng học tập nặng và bạn muốn bớt môn, hãy hỏi cho rõ trước: bớt xuống bao nhiêu thì không còn tính là toàn thời gian. Chuyện này hỏng trước chuyện làm thêm. Bị trường đuổi học, hoặc chương trình bị chấm dứt, thì tư cách thường đứt ngay trong ngày. Đừng chờ đến khi trường gửi thư chính thức rồi mới hành động. + +### 5. Hoa Kỳ: sau khi chuyển nhà, báo địa chỉ mới trong vòng 10 ngày + +- Chi phí: Không tốn tiền. Nộp đơn đổi địa chỉ trực tuyến một lần, mất mười phút +- Nói thẳng: Ở Hoa Kỳ, khi bạn đổi chỗ ở, luật bắt bạn báo cho cơ quan nhập cảnh trong vòng 10 ngày. Không báo không phải chuyện nhỏ. Một là bản thân việc đó vi phạm pháp, hai là thư gửi cho bạn của cơ quan nhập cảnh sẽ bị thất lạc. +- Lợi ích: Quy định nhập cảnh liên bang Hoa Kỳ 8 CFR 265.1 quy định người nước ngoài phải đăng ký theo luật thì phải báo trong vòng 10 ngày kể từ khi đổi địa chỉ. Báo cho USCIS (Cơ quan Nhập cảnh và Quốc tịch Hoa Kỳ), theo cách họ yêu cầu, vừa báo việc đổi địa chỉ, vừa báo địa chỉ mới +- Mức bằng chứng: A +- Nguồn: 8 CFR 265.1. ；USCIS. AR-11, Alien's Change of Address Card. +- Ghi chú: Báo với cơ quan nhập cảnh rồi, hồ sơ SEVIS ở trường (hồ sơ tư cách sinh viên quốc tế của bạn) vẫn còn phải cập nhật riêng. Ba nước còn lại cũng đều yêu cầu đổi địa chỉ thì báo cơ quan nhập cảnh hoặc trường. Hạn chót và cách làm khác nhau theo từng nước, làm theo trang web của cơ quan nhập cảnh nước bạn đang ở. + +### 6. Trước khi lên đường và trong thời gian học, xem qua cảnh báo du học của Bộ Giáo dục một lần + +- Chi phí: Không tốn tiền. Xem một lần trước khi lên đường, sau đó mỗi học kỳ xem một lần +- Nói thẳng: Bộ Giáo dục phát cảnh báo riêng cho một nước hoặc vùng, nói rằng tình hình an ninh hoặc chính sách ở đó gần đây bất lợi cho sinh viên Trung Quốc. Đây là cảnh báo rủi ro công khai chính thức, có văn bản để tra, đáng tin hơn lời đại gia nói. Xem một lượt không tốn tiền. +- Lợi ích: Bộ Giáo dục phát cảnh báo du học qua Mạng lưới Giám sát và Quản lý Giáo dục có yếu tố nước ngoài. Năm 2025 phát hành tổng cộng 4 số. Số 1 cảnh báo rằng các dự luật giáo dục đại học do một số tiểu bang Hoa Kỳ thông qua có điều khoản tiêu cực đối với Trung Quốc (2025-04-09). Số 2 và số 3, trong hai lần cảnh báo, nói rằng các vụ án nhắm vào công dân Trung Quốc ở Philippines xảy ra nhiều và môi trường du học ở đó không tốt (2025-07-18、2025-08-30). Số 4 cảnh báo tình hình an ninh và môi trường du học ở Nhật không tốt, khuyên cân nhắc kỹ trước khi lên kế hoạch đi học ở Nhật (2025-11-16) +- Mức bằng chứng: A +- Nguồn: Mạng lưới Giám sát và Quản lý Giáo dục có yếu tố nước ngoài của Bộ Giáo dục（教育部教育涉外监管信息网）. Thông tin cảnh báo（预警信息）. ；Bộ Giáo dục phát cảnh báo du học số 4 năm 2025. ；Số 1. +- Ghi chú: Cảnh báo không phải lệnh cấm, cũng không phải bảo bạn không được đi, nó cho bạn một căn cứ chính thức để cân đối rủi ro nơi đó. Danh sách cảnh báo đổi theo tình hình, phần này không cập nhật dài hạn, cứ theo trang hiện hành trên website. An toàn cá nhân ở nước ngoài và bảo vệ lãnh sự, xem phần 21. +### 7. Australia: OSHC phải phủ trọn suốt và không được đứt quãng, xuất cảnh mà không lấy ra được bằng chứng bảo hiểm thì có thể bị từ chối nhập cảnh + +- Chi phí: Mấy nghìn đến hơn mười nghìn tệ. Mua một lần theo thời hạn visa, học mấy năm thì mua mấy năm +- Nói thẳng: Australia đã đưa bảo hiểm y tế cho sinh viên vào điều kiện của visa, không phải mua cũng được, không mua cũng được. Bảo hiểm phải phủ kín toàn bộ thời gian lưu trú. Khi gia hạn và đổi sang hợp đồng bảo hiểm mới thì không được có khoảng trống ở giữa. Lúc nhập cảnh mà không lấy ra được bằng chứng đã mua bảo hiểm thì có thể bị từ chối nhập cảnh ngay tại chỗ. Người nhà đi đón, mỗi người phải mua một hợp đồng riêng. +- Lợi ích: Bộ Nội vụ Australia quy định người nộp đơn visa sinh viên và người thân của họ phải mua bảo hiểm y tế cho sinh viên quốc tế (OSHC), và phải có suốt thời gian lưu trú ở Australia, không được gián đoạn. Bảo hiểm phải do cơ quan bảo hiểm được Australia công nhận cung cấp. Trừ các trường hợp thuộc diện được miễn mua. Giữa hợp đồng bảo hiểm mới và hợp đồng của visa trước không được có khoảng trống. Lúc nhập cảnh mà không lấy ra được chứng cứ đã tham gia bảo hiểm thì có thể bị từ chối nhập cảnh. Nhập cảnh sớm hơn ngày bắt đầu khóa học thì ngày bắt đầu bảo hiểm phải ghi là ngày thật sự đến Australia +- Mức bằng chứng: A +- Nguồn: Australian Government Department of Home Affairs. Student visa (subclass 500). +- Ghi chú: Nếu trường làm thủ thì hãy hỏi rõ tên công ty bảo hiểm, ngày bắt đầu và kết thúc hợp đồng, số hợp đồng, những thứ này phải điền trong hồ sơ xin visa. Mua ngắn hơn thì lúc gia hạn sẽ vướng, nên mua dài đến sau khi tốt nghiệp thêm một chút. + +### 8. Anh: ngoài phí visa còn một khoản phụ thu y tế tính theo năm, sinh viên mỗi năm 776 bảng Anh, nộp một lần lúc xin + +- Chi phí: Phí visa 558 bảng Anh. Phụ thu y tế 776 bảng Anh mỗi năm. Cả hai khoản đều nộp một lần lúc xin +- Nói thẳng: Sang Anh phải trả không chỉ phí visa. Còn phải nộp trước phụ thu y tế theo số năm của visa, nộp rồi mới được dùng dịch vụ y tế công của Anh. Visa hai năm thì nộp một lần 1552 bảng Anh. Cộng phí visa nữa, trước khi xuất hành khoản này phải nộp một lần hết, làm ngân sách đừng sót. +- Lợi ích: Phí visa sinh viên của Anh là 558 bảng Anh. Xin ở nước ngoài, gia hạn hoặc chuyển đổi ở trong nước đều là số này. Ngoài ra còn có phụ thu y tế, nộp rồi mới được dùng dịch vụ y tế công của Anh, tính theo số năm của visa và nộp trước một lần lúc xin. Bản thân sinh viên và người thân của họ là 776 bảng Anh mỗi năm, ví dụ visa 2 năm thì 1552 bảng Anh. Người xin thuộc nhóm khác là 1035 bảng Anh mỗi năm. Visa trên 6 tháng mà chưa đủ 1 năm thì tính tròn một năm +- Mức bằng chứng: A +- Nguồn: UK Government. Student visa. ；UK Government. Pay for UK healthcare as part of your immigration application. +- Ghi chú: Số tiền và tỷ giá đều có thể đổi, cứ tính theo con số của kỳ hiện tại trên hai trang đó. Khóa học lấy bằng, từ 18 tuổi trở lên thì về nguyên tắc ở lâu nhất 5 năm, dưới bậc đại học là 2 năm. Ước theo số năm đó xem phải nộp phụ thu bao nhiêu. + +### 9. Trước khi về nước, chừa thời gian cho việc công nhận: công nhận của Trung tâm Dịch vụ Học bổng mất 10 đến 20 ngày làm việc + +- Chi phí: Nộp hồ sơ trên mạng, nộp tiền trên mạng. Chuẩn bị hồ sơ cộng thời gian xét duyệt, tính theo ngày làm việc, tổng cộng khoảng một tháng +- Nói thẳng: Bằng tốt nghiệp từ nước ngoài muốn dùng ở nước nhà thì phải làm công nhận trình độ trước. Riêng khâu xét duyệt đã mất 10 đến 20 ngày làm việc. Gặp mùa tìm việc, hoặc hồ sơ bị trả lại thì còn lâu hơn. Tuyển sinh thu, đăng ký thi công chức, lập hộ khẩu đều có hạn, đừng đợi đến lúc cần mới bắt đầu làm. +- Lợi ích: Công nhận trình độ học vụ ngoài nước và ngoài vùng lãnh thổ làm ở sảnh giao dịch trực tuyến của Trung tâm Dịch vụ Học bổng Trung Quốc (Bộ Giáo dục). Quy trình gồm bốn bước: đăng ký và xác minh danh tính thật, nộp hồ sơ và tài liệu trực tuyến, nộp tiền trực tuyến, chờ đánh giá và xét duyệt. Thời hạn công nhận được công bố là 10 đến 20 ngày làm việc. Ngày làm việc không gồm cuối tuần và ngày lễ, đây là thời gian Trung tâm Dịch vụ Học bổng dùng để xét duyệt +- Mức bằng chứng: A +- Nguồn: Trung tâm Dịch vụ Học bổng của Trung Quốc (Bộ Giáo dục) (中国（教育部）留学服务中心). Sảnh giao dịch trực tuyến, công nhận trình độ học vụ. +- Ghi chú: Hồ sơ phải nộp gồm: bằng tốt nghiệp, hộ chiếu hoặc giấy thông hành, thẻ cư trú hoặc visa và dấu nhập cảnh, ảnh giấy tờ tùy thân, bản ủy quyền. Dữ liệu xuất nhập cảnh trong thời gian ở nước ngoài do hệ thống tự tra, nên đừng làm rách trang visa ở nước ngoài. Trên trang của Trung tâm Dịch vụ Học bổng còn có thông báo, nói giấy công nhận nào bị vô hiệu, trường nào tạm ngừng nhận hồ sơ công nhận. Trước khi làm thì xem đã có trường nào liên quan đến mình không. + +### 10. Các trường bị「siết lại việc công nhận」mỗi năm lại thêm, con đường mua bằng đã đi không còn + +- Chi phí: Không tốn tiền. Trước khi đăng ký, tra tên trường một lần trong thông báo của Trung tâm Dịch vụ Học bổng +- Nói thẳng: Trung tâm Dịch vụ Học bổng chỉ điểm một loạt trường ở nước ngoài, siết riêng việc công nhận bằng của chúng, cũng có trường bị tạm ngừng nhận hồ sơ công nhận. Những trường bị chỉ điểm thường đúng loại quảng cáo「không cần chứng chỉ ngôn ngữ, không cần xuất cảnh, vài tháng có bằng thạc sĩ」. Tra tên trường trước khi đăng ký đỡ phải khiếu nại về sau. +- Lợi ích: Trung tâm Dịch vụ Học bổng Trung Quốc (Bộ Giáo dục) liên tục phát thông báo siết lại việc đánh giá công nhận trình độ cho một số trường ở nước ngoài (bằng của những trường đó bị kiểm tra riêng thêm một lớp). Bản mới nhất là số chín (2025-10-28). Trung tâm Dịch vụ Học bổng cũng từng phát thông báo tạm ngừng nhận hồ sơ công nhận của một số trường ở nước ngoài, cũng từng cảnh báo có người lợi dụng việc công nhận trình độ học vụ ngoài nước và ngoài vùng lãnh thổ để lừa đảo +- Mức bằng chứng: A +- Nguồn: Trung tâm Dịch vụ Học bổng Trung Quốc (Bộ Giáo dục) (中国（教育部）留学服务中心). Thông báo (九) về việc siết lại việc đánh giá công nhận trình độ cho một số trường ở nước ngoài (2025-10-28). ；Trung tâm Dịch vụ Học bổng Trung Quốc (Bộ Giáo dục) (中国（教育部）留学服务中心). +- Ghi chú:「Siết lại」không có nghĩa là chắc chắn không công nhận được, nhưng bạn sẽ phải nộp thêm tài liệu và chờ lâu hơn. Mọi môi giới hứa「bảo đảm công nhận」「kênh trong nước làm gấp」đều là lừa đảo, Trung tâm Dịch vụ Học bổng đã phát cảnh báo phòng lừa đảo riêng. Thuê người khác viết hộ luận văn hoặc thay bạn học khóa online xuyên biên giới, nếu bị phát hiện thì ngoài việc bằng bị vô hiệu còn có thể ảnh hưởng đến lịch sử visa. From 694aefbb6ffc48394b7c685d754f3b4792325d8f Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 13:01:38 +0700 Subject: [PATCH 19/25] =?UTF-8?q?=E6=9D=A1=E7=9B=AE=E6=A0=87=E9=A2=98?= =?UTF-8?q?=E4=B9=9F=E7=BA=B3=E5=85=A5=E6=A3=80=E6=9F=A5=EF=BC=9A=E6=AD=A3?= =?UTF-8?q?=E6=96=87=E9=80=90=E6=A0=8F=E6=A0=A1=E4=BA=86=E5=85=AD=E9=81=8D?= =?UTF-8?q?=EF=BC=8C=E6=A0=87=E9=A2=98=E8=BF=99=E4=B8=80=E8=A1=8C=E4=BB=A5?= =?UTF-8?q?=E5=89=8D=E6=B2=A1=E4=BA=BA=E7=AE=A1?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit vi 02 第 21、22、23、25 条和 vi 20 第 1 条的标题整条还是中文，正文翻好了。 判据加在 structure.mjs 的 compare() 里：标题跟中文原文一字不差就报， 「」《》（）里照抄的法条名书名不算（BRIEF 规则 6 允许）。 试过把改好的标题换回中文，join 立刻报出来，不是静悄悄地过。 另外撤掉一个我自己写的假警报：上一版判据只按长度查「空栏」， 证据等级栏是 A/B/C 一个字母，636 条全被误报——判据太粗，量级词那些不算漏译。 现状：三种语言各 34 节 641 条，六道检查全绿。 --- tools/i18n/structure.mjs | 16 +- ...01\346\205\242\346\205\242\346\255\273.md" | 8 +- ...20\346\200\216\344\271\210\345\270\246.md" | 2 +- ...25\345\222\214\347\224\237\344\272\247.md" | 204 ++++++++---------- ...23\345\207\273\344\271\213\345\220\216.md" | 14 +- 5 files changed, 122 insertions(+), 122 deletions(-) diff --git a/tools/i18n/structure.mjs b/tools/i18n/structure.mjs index 5b723a8a..4c02da51 100644 --- a/tools/i18n/structure.mjs +++ b/tools/i18n/structure.mjs @@ -202,6 +202,16 @@ function label(S, f) { return S.fields[f].replace(/[：:]$/, ''); } * 逐条比。返回问题清单（空数组就是这块对了）。dstLabel 是给报告前缀用的语言名。 * 这一套就是 check.mjs 的判据，抽出来是为了 join 逐块校时用同一份。 */ +/** 汉字。「」《》里照抄的法条名、书名不算漏译（BRIEF 规则 6 明确允许）。 */ +const han = s => /[㐀-鿿]/.test(s); +/** 剔掉允许照抄的部分，再看还有没有汉字——有就是漏译。 */ +const titleBare = s => String(s) + .replace(/「[^」]*」|『[^』]*』|《[^》]*》/g, '') + .replace(/（[^）]*）/g, '') + .replace(/\([^)]*\)/g, '') + .replace(/]*>/g, '') + .trim(); + export const STRUCT = { ALWAYS, FIELDS, /** 整节的判据：节号、节标题、导读。条目逐条比是 compare()。 @@ -215,7 +225,6 @@ export const STRUCT = { // 两处搜索对不上，这层该拦。 compareSection(a, b, tag, toc) { const out = []; - const han = s => /[㐀-鿿]/.test(s); // 汉字。引号里照抄的法条名不算漏译 if (a.n !== b.n) out.push(`${tag}：节号是 ${b.n}，原文 ${a.n}`); if (!b.title.trim()) out.push(`${tag}：节标题是空的`); else if (a.title && b.title === a.title) @@ -248,6 +257,11 @@ export const STRUCT = { const D = strings(b._locale ?? SOURCE.code); if (a.n !== b.n) out.push(`${tag}：条号是 ${b.n}，原文 ${a.n}`); if (!b.title.trim()) out.push(`${tag}：标题是空的`); + // 标题正文不能还是中文（法条名、书名、括注里的照抄不算——那些是 BRIEF 规则 6 允许的）。 + // 条正文逐栏校了六遍，标题这一行以前没人管，翻漏了就一直漏着 + // （2026-10-02 在 vi 02 和 vi 20 各查出来几条）。 + else if (a.title && han(titleBare(b.title)) && b.title === a.title) + out.push(`${tag}：标题「${b.title}」跟中文原文一字不差，这一行没翻`); // 成本标签逐字照抄：检索页的筛选和地址栏全靠它，三种语言共用一套中文取值 if (a.tag !== b.tag) out.push(`${tag}：成本标签「${b.tag}」和原文「${a.tag}」不一样，要一字不改照抄`); diff --git "a/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" "b/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" index 4ff2f52e..e0ed8dcd 100644 --- "a/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" +++ "b/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" @@ -168,7 +168,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Wood AM và cộng sự (2018). Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. ; GBD 2016 Alcohol Collaborators (2018). Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. ; Zhao J và cộng sự (2023). Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Network Open. ; Di Castelnuovo A và cộng sự (2006). Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Archives of Internal Medicine. (phe bên kia) - Ghi chú:Tranh cãi. Phe cho rằng uống ít có lợi dựa vào một bảng gộp 34 nghiên cứu theo dõi dài hạn. Nó báo rằng người uống ít có tỷ lệ tử vong toàn nguyên thấp hơn tối đa 17%–18%. Lượng mà nó đưa ra là nam không quá 4 ly mỗi ngày, nữ không quá 2 ly. Trong mức đó, người uống nhiều hơn lại có tỷ lệ tử vong thấp hơn. Phe bên kia cho rằng nhóm "không uống" đã lẫn người bỏ rượu vì bệnh và người vốn sức khoẻ đã kém. Có một bảng gộp chỉ vào đúng những nhiễu đó, sau khi trừ đi thì tác dụng bảo vệ biến mất. Nghiên cứu Giao tải bệnh toàn cầu cũng đặt 0 là an toàn nhất. Cách đọc an toàn là: uống ít không rõ là có rút ngắn tuổi thọ hay không, uống nhiều thì chắc chắn có hại. "Bắt đầu uống rượu vì sức khoẻ" không có cơ sở. Muốn uống ít hơn thì xem mục 22 của phần này (đếm số ly trước rồi hỏi bác sĩ). Người uống rượu mỗi ngày không nên tự bỏ dứt khoát, xem mục 21 của phần này (đừng tự bỏ dứt khoát) -### 21. 每天都喝酒、一停就手抖心慌的人，别自己硬戒 +### 21. Người uống rượu mỗi ngày, ngừng lại là run tay và hồi hộp, đừng tự cai bằng sức mình - Chi phí: Khám khoa tâm thần hoặc khoa nghiện rượu một lần, mấy tệ đến mấy chục tệ. Trường hợp nặng thì phải nằm viện mấy ngày. - Nói thẳng: Người uống rượu nhiều mỗi ngày mà ngừng đột ngột sẽ có phản ứng cai rượu. Nhẹ thì run tay, ra mồ hôi, hồi hộp, mất ngủ. Nặng thì co giật hoặc lên cơn sảng run, đó là cấp cứu có thể gây chết. Na Uy theo dõi hơn 30.000 người, trong nhóm từng bị sảng run thì mỗi năm về sau có 8% chết. Muốn cai thì đi bệnh viện trước, nói rõ mỗi ngày uống bao nhiêu. @@ -177,7 +177,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Nguồn: Caputo F, Lungaro L, Costanzini A, De Giorgio R, Addolorato G (2026). Alcohol withdrawal syndrome in hospitalized patients: a practical review. European Journal of Internal Medicine, 107103. ；Bramness JG, Heiberg IH, Høye A, Rossow I (2023). Mortality and alcohol-related morbidity in patients with delirium tremens, alcohol withdrawal state or alcohol dependence in Norway: A register-based prospective cohort study. Addiction, 118(12), 2352–2359. - Ghi chú: Tự đánh giá thế nào: gần như ngày nào cũng uống, một ngày không uống thì run tay, ra mồ hôi, hồi hộp, mất ngủ, sáng dậy phải uống một chút mới dễ chịu, thì chỉ cần trúng một dấu hiệu trong số đó là đừng tự ngừng đột ngột. Bệnh viện thường dùng thuốc nhóm benzodiazepin để người bệnh chịu qua mấy ngày đó, đồng thời bổ sung thiamine (vitamin B1). Để ở mức B là vì nghiên cứu mẫu của Na Uy nói về「người từng bị sảng run sau đó thế nào」, không phải「tự cai cứng thì ra sao」, giữa hai câu hỏi đó còn là suy luận. Rượu vẫn phải cai, chỉ là cai ở bệnh viện. Mỗi tuần uống bao nhiêu thì tính là nhiều, xem mục 20 của phần này (uống ít hoặc không uống). Muốn uống ít hơn thì làm sao, xem mục 22 của phần này (đếm số ly trước rồi hỏi bác sĩ) -### 22. 想少喝酒，先把一周喝了多少数出来，再找医生聊几分钟 +### 22. Muốn uống ít rượu hơn, trước tiên tính xem một tuần uống bao nhiêu, rồi nói chuyện vài phút với bác sĩ - Chi phí: Không tốn tiền. Mỗi ngày ghi số ly vào sổ là xong. Hỏi thêm một câu ở phòng khám cũng không tốn thêm tiền. - Nói thẳng: Bác sĩ dành mấy phút để xem lượng rượu bạn uống, nói rõ rủi ro, giúp bạn đặt một mục tiêu. Sau một năm, trung bình mỗi tuần uống ít đi 20 g rượu thuần, tương đương khoảng nửa lít bia. Nói chuyện lâu hơn không có thêm lợi ích. Người đã không ngừng được thì ở nước ngoài có hai loại thuốc đã được chứng minh có tác dụng, cứ điều trị khoảng 12 người thì có thêm 1 người không quay lại kiểu uống một lần là nhiều. @@ -186,7 +186,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Nguồn: Kaner EF, Beyer FR, Muirhead C, et al. (2018). Effectiveness of brief alcohol interventions in primary care populations. Cochrane Database of Systematic Reviews, 2, CD004148. ；Jonas DE, Amick HR, Feltner C, et al. (2014). Pharmacotherapy for adults with alcohol use disorders in outpatient settings: a systematic review and meta-analysis. JAMA, 311(18), 1889–1900. - Ghi chú: Mức lợi ích ghi「nhỏ」vì cả hai nguồn bằng chứng đo lượng uống, không đo tỷ lệ tử vong, và theo quy ước của cuốn sách này, chỉ có điểm thay thế thì ghi「nhỏ」. Phần thuốc đa số thử nghiệm làm ở nước ngoài, phần lớn người tham gia đã đạt tiêu chuẩn chẩn đoán nghiện rượu, và mọi thử nghiệm đều kèm can thiệp tâm lý xã hội, nên con số là「phần lợi ích cộng thêm ngoài cái nền đó」. Về chỉ định và khả năng kê đơn của hai loại thuốc này ở Trung Quốc, cứ theo bác sĩ và tờ hướng dẫn sử dụng thuốc, đừng tự mua qua mạng. Đã xuất hiện phản ứng cai rượu thì xem mục 21 của phần này (đừng tự cai cứng). -### 23. 每天吃一小把坚果 +### 23. Mỗi ngày ăn một nắm nhỏ hạt - Chi phí: Mỗi ngày ăn 28 g, một năm khoảng hai ba trăm tệ. Ăn tiện tay, không mất thời gian. - Nói thẳng: Người ăn hạt thường xuyên có xác suất tử vong trong cùng thời gian thấp hơn người không ăn khoảng hai phần mười. Ăn một lần mỗi tuần đã thấy khác biệt, thấp hơn khoảng một phần mười. Ăn đến mức mỗi ngày một nắm thì khác biệt rõ nhất. @@ -205,7 +205,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Ghi chú: Hai phần mỗi tuần tạo ra khác biệt rất nhỏ, đừng trông chờ đổi thịt rồi sống thêm mấy năm. Cùng số tiền bỏ ra để ăn thịt thì có thể ưu tiên cá và thịt gia cầm -### 25. 把一部分精米白面换成全谷物 +### 25. Đổi một phần gạo trắng và bột mì tinh thành ngũ cốc nguyên hạt - Chi phí: Gạo lức, yến mạch, bánh mì nguyên cốc đắt hơn gạo trắng và bột mì tinh bột một chút, không mất thời gian. Chỗ khó là vị phải từ từ quen. - Nói thẳng: Người ăn thêm 90 g ngũ cốc nguyên mỗi ngày có xác suất tử vong trong cùng thời gian thấp hơn khoảng 17%. 90 g xấp xỉ ba phần. Ăn đến khoảng 200 g mỗi ngày thì con số vẫn đang giảm tiếp. diff --git "a/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" "b/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" index c36645f5..9ea837b6 100644 --- "a/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" +++ "b/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" @@ -4,7 +4,7 @@ Cột đo của phần này là tỷ lệ tử vong trẻ sơ sinh và tiền. Phần này chỉ nhận những mục liên quan đến việc đứa trẻ có sống sót hay không, hoặc đến việc có bị gì về sau hay không. Chỗ nào có nhắc tiền chỉ đưa cách tính, không đưa kết luận. -### 1. 让宝宝仰着睡，睡硬的平面，同房不同床，床上不放任何软东西 +### 1. Cho trẻ nằm ngửa, nằm trên mặt phẳng cứng, cùng phòng nhưng khác giường, trên giường không có gì mềm - Chi phí: Không tốn tiền. Phải đổi nệm sang loại cứng, vài trăm tệ. Chỗ khó là phải giữ vững trước ý kiến của bậc trên, theo đúng cách này từ đầu đến cuối - Nói thẳng: Cho con nằm ngửa mà ngủ, trên một mặt phẳng cứng. Người lớn và con ngủ cùng một phòng, nhưng không ngủ cùng một giường. Trên giường không đặt gối, chăn, tấm chắn cạnh nôi, gấu bông, gối chỉnh hình. Cũi nôi và ghế trẻ em nghiêng chỉ để ngồi, không dùng làm giường ngủ. Riêng nước Mỹ, mỗi năm có khoảng 3500 trẻ chết trong lúc ngủ. diff --git "a/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" "b/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" index 3728a810..781604f4 100644 --- "a/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" +++ "b/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" @@ -1,151 +1,135 @@ -[← Về mục lục](../README.md) +[← 回总目录](../README.md) # 27. Mang thai và sinh con: từ que thử đến giấy tờ -Nửa đầu phần này tính rủi ro tử vong của người mẹ và của thai nhi. Nửa sau tính tiền và thời gian chạy vặt. Hai thứ này không quy đổi qua lại. Nội dung chỉ nói khi nào làm việc gì. Dinh dưỡng khi mang thai, giáo dục thai nhi, loại không có bằng chứng chắc chắn hoặc tác dụng rất nhỏ, ở đây không viết. Phần tiền của trợ cấp thai sản, nghỉ phép thai sản và nuôi con xem phần 18. Cách chăm đứa trẻ sau khi sinh xem phần 20. Cấp cứu xem phần 13. +Nửa đầu phần này tính rủi ro tử vong của sản phụ và thai nhi. Nửa sau tính tiền và thời gian phải chạy vòng. Hai kiểu tài khoản không quy đổi qua lại với nhau. Nội dung chỉ nói 「làm gì khi nào」. Dinh dưỡng khi mang thai, giáo dục thai nhi — những thứ không có bằng chứng chắc chắn, hoặc tác dụng rất nhỏ — ở đây không viết. Phần tính tiền về trợ cấp sinh sản, nghỉ thai sản và nuôi con xem ở phần 18, sau khi đứa trẻ sinh ra thì chăm thế nào xem ở phần 20, trường hợp cấp cứu xem ở phần 13. -Con số của Tổ chức Y tế Thế giới: năm 2023, khoảng 260.000 phụ nữ trên toàn thế giới chết vì mang thai và sinh con. Khoảng 75% trong số đó chết vì năm nhóm nguyên nhân sau: chảy máu nặng sau sinh, nhiễm trùng, rối loạn tăng huyết áp khi mang thai (nhóm bệnh do huyết áp tăng sau khi có thai, gồm tiền sản giật thai kỳ và sản giật), biến chứng khi sinh nở, và nạo thai không an toàn. Tổ chức Y tế Thế giới cũng nói「大多数孕产妇死亡是可以预防的，因为预防和处理并发症的医疗手段都是已知的」. Vài mục có tác dụng lớn nhất trong phần này đều nhằm chặn những nhóm bệnh này lại từ sớm. - -### 1. Bắt đầu uống 0.4 mg axit folic mỗi ngày từ trước khi mang thai, và uống liền suốt 3 tháng đầu thai kỳ +Con số của Tổ chức Y tế Thế giới: năm 2023, toàn cầu có khoảng 260.000 phụ nữ chết vì mang thai và sinh con. Trong đó khoảng 75% chết vì năm nhóm nguyên nhân sau: băng huyết sau sinh, nhiễm trùng, bệnh tăng huyết áp khi mang thai (một nhóm bệnh do huyết áp tăng sau khi có thai, gồm tiền sản giật và sản giật), biến chứng khi chuyển đoạn, và phá thai không an toàn. Tổ chức Y tế Thế giới đồng thời nói 「hầu hết các ca tử vong ở sản phụ đều có thể phòng tránh được, vì các biện pháp y tế dùng để phòng và xử lý biến chứng đều đã biết」. Vài mục hiệu quả nhất trong phần này chính là chặn trước mấy nhóm vấn đề đó. +### 1. Chuẩn bị mang thai thì bắt đầu uống 0.4 miligam axit folic mỗi ngày, uống liên tới hết 3 tháng đầu thai kỳ -- Chi phí: Vài chục tệ một chai. Phụ nữ nông thôn không tốn tiền, chương trình bổ sung axit folic miễn phí của nhà nước sẽ phát. Mỗi ngày một viên, chỗ khó là nhớ đều mỗi ngày -- Nói thẳng: Uống một viên axit folic mỗi ngày trước và quanh lúc mang thai giảm khoảng bảy phần mười xác suất trẻ bị khuyết tật ống thần kinh. Khuyết tật ống thần kinh là nhóm dị tật bẩm sinh như không có não, nứt đốt sống. Người đã sinh một đứa con bị như vậy thì xác suất đứa tiếp theo lại bị, xuống còn khoảng một phần ba mức cũ. -- Lợi ích: Phân tích tổng hợp của Cochrane gộp 5 thử nghiệm, 6708 lượt đẻ, và xếp chất lượng bằng chứng ở mức cao. Uống axit folic mỗi ngày trước và quanh lúc mang thai cho nguy cơ khuyết tật ống thần kinh của thai nhi là RR 0.31 (95% CI 0.17 đến 0.58, khoảng tin cậy), tức giảm khoảng bảy phần mười. Người đã có một đứa con bị khuyết tật ống thần kinh thì nguy cơ đứa tiếp theo lại bị là RR 0.34 (95% CI 0.18 đến 0.64), còn khoảng một phần ba mức cũ. Tách ra theo nhóm dân số, liều 0.4 mg và liều cao hơn không khác nhau. Uống riêng axit folic và uống trong vitamin tổng hợp cũng không khác nhau +- Chi phí: Vài chục tệ một chai. Phụ nữ ở nông thôn không tốn tiền, chương trình bổ sung axit folic miễn phí của nhà nước sẽ phát. Mỗi ngày uống một viên, chỗ khó là nhớ đều đặn mỗi ngày. +- Nói thẳng: Uống một viên axit folic mỗi ngày trước và sau khi mang thai, xác suất đứa trẻ bị khiếm khuyết ống thần kinh giảm khoảng bảy mươi phần trăm. Khiếm khuyết ống thần kinh là nhóm dị tật bẩm sinh như thất não, nứt đốt sống. Người đã sinh một đứa trẻ như vậy thì lần sinh tiếp theo mà còn bị nữa, xác suất rơi xuống khoảng một phần ba so với trước. +- Lợi ích: Phân tích tổng hợp Cochrane gộp 5 thử nghiệm, 6708 ca sinh lại với nhau, chất lượng bằng chứng được xếp là cao. Uống axit folic mỗi ngày trước và sau khi mang thai, nguy cơ khiếm khuyết ống thần kinh ở thai nhi là RR 0.31 (95% CI 0.17 đến 0.58, khoảng tin cậy), giảm khoảng bảy mươi phần trăm. Người đã sinh con bị khiếm khuyết ống thần kinh, lần sinh tiếp theo mà lại bị nữa có nguy cơ RR 0.34 (95% CI 0.18 đến 0.64), xuống khoảng một phần ba mức cũ. Xét theo từng nhóm người thì liều 0.4 miligam và liều cao hơn hiệu quả như nhau. Chỉ uống axit folic và uống vitamin tổng hợp cũng hiệu quả như nhau. - Mức bằng chứng: A - Nguồn: De-Regil LM, Peña-Rosas JP, Fernández-Gaxiola AC, Rayco-Solon P (2015). Effects and safety of periconceptional oral folate supplementation for preventing birth defects. Cochrane Database of Systematic Reviews, (12), CD007950. -- Ghi chú: Điểm mấu chốt là「uống từ trước khi có thai」. Ống thần kinh của thai nhi khép lại trong vòng 28 ngày sau khi thụ thai. Đợi que thử dương rồi mới uống thì đã mất hơn nửa khoảng thời gian đó. Nên nếu định có con, cứ bắt đầu uống trước 3 tháng. Cùng nghiên cứu tổng hợp đó không thấy tác dụng rõ ràng của axit folic với khoảm miệng hở vòm, bệnh tim bẩm sinh hay sẩy thai. Đừng trông chờ nó chữa bệnh khác - -### 2. Trước tuần thai kỳ thứ 13, ra trạm y tế cộng đồng lập Sổ sức khoẻ bà mẹ và con, và dùng hết phần khám thai miễn phí +- Ghi chú: Điểm mấu chốt là 「bắt đầu uống trước khi mang thai」. Ống thần kinh của thai nhi đã khép lại trong vòng 28 ngày sau khi thụ thai. Đợi tới khi que thử báo có thai rồi uống thì khoảng thời gian đó đã trôi qua phần lớn. Nếu định có con thì uống sớm 3 tháng. Trong cùng nghiên cứu tổng hợp đó, axit folic không có tác dụng rõ ràng lên khe hở môi vòm hành, bệnh tim bẩm sinh hay sẩy thai, đừng trông chờ nó chữa bệnh khác. +### 2. Trước tuần thai kỳ thứ 13, ra trạm y tế cộng đồng lập Sổ sức khoẻ Mẹ và Con và dùng hết hạn mức khám thai miễn phí -- Chi phí: Không tốn tiền, đây thuộc chương trình dịch vụ công cộng cơ sở về y tế quốc gia. Phải chạy mấy chuyến theo số tuần thai kỳ -- Nói thẳng: Ít nhất năm lần khám thai trong thai kỳ, cộng một lần thăm khám tại nhà sau sinh và một lần khám 42 ngày sau sinh, đều do nhà nước trả. Cách nhận là trước tuần mang thai thứ 13 ra cộng đồng lập một quyển Sổ sức khoẻ bà mẹ và con《母子健康手册》. Không lập thì mấy lần miễn phí đó mất trắng. -- Lợi ích: 《国家基本公共卫生服务规范（第三版）》 có dịch vụ quản lý sức khoẻ sản phụ. Sản phụ thường trú trong địa bàn được làm các dịch vụ đó. Khám sức khoẻ 1 lần ở tam nguyệt đầu, 2 lần ở tam nguyệt giữa, 2 lần ở tam nguyệt cuối, 1 lần thăm khám sau sinh và 1 lần khám sức khoẻ 42 ngày sau sinh. Phải lập sổ trước tuần thai kỳ thứ 13, đồng thời làm lần khám tiền sản thứ nhất. Hai lần ở tam nguyệt giữa rơi vào tuần thai kỳ thứ 16 đến 20 và tuần thai kỳ thứ 21 đến 24. Hai lần ở tam nguyệt cuối rơi vào tuần thai kỳ thứ 28 đến 36 và tuần thai kỳ thứ 37 đến 40. Danh mục dịch vụ công khai của địa phương còn gồm toàn bộ xét nghiệm của lần đầu và sàng lọc trầm cảm sau sinh +- Chi phí: Không tốn tiền, đây thuộc dự án dịch vụ y tế công cộng cơ bản của nhà nước. Phải chạy vòng theo từng giai đoạn thai kỳ. +- Nói thẳng: Trong thai kỳ có ít nhất 5 lần khám thai, cộng thêm một lần thăm khám tại nhà sau sinh và một lần khám 42 ngày sau sinh, số tiền này do nhà nước trả. Cách để được hưởng là trước tuần thai kỳ thứ 13 ra cộng đồng lập một Sổ sức khoẻ Mẹ và Con. Không lập sổ thì mấy lần miễn phí đó mất trắng. +- Lợi ích: 《国家基本公共卫生服务规范（第三版）》 có phần quản lý sức khoẻ sản phụ. Sản phụ hộ trú trong địa bàn được dùng các dịch vụ này. 1 lần khám sức khoẻ giai đoạn đầu thai kỳ, 2 lần giữa thai kỳ, 2 lần cuối thai kỳ, 1 lần thăm khám sau sinh, 1 lần khám sức khoẻ 42 ngày sau sinh. Trước tuần thai kỳ thứ 13 phải lập sổ, đồng thời làm lần khám thai lần 1. Hai lần giữa thai kỳ rơi vào tuần thai kỳ 16 đến 20 và tuần thai kỳ 21 đến 24. Hai lần cuối thai kỳ rơi vào tuần thai kỳ 28 đến 36 và tuần thai kỳ 37 đến 40. Danh mục dịch vụ công khai của địa phương còn gồm trọn bộ xét nghiệm của lần đầu và sàng lọc trầm cảm sau sinh. - Mức bằng chứng: A -- Nguồn: 国家卫生计生委 (Uy ban Sức khoẻ và Kế hoạch Gia đình quốc gia) (2017). Quy định về dịch vụ công cộng cơ sở về y tế quốc gia, 国家基本公共卫生服务规范（第三版）. ；靖江市人民政府 (Chính quyền nhân dân thành phố Tĩnh Giang). Chương trình dịch vụ công cộng cơ sở về y tế quốc gia — dịch vụ quản lý sức khoẻ sản phụ (bảng dịch vụ công khai của địa phương), 国家基本公共卫生服务项目——孕产妇健康管理服务（地方服务清单公开）. -- Ghi chú: 5 lần đó là mức tối thiểu được miễn phí, không phải toàn bộ khám thai. Bệnh viện sẽ tuỳ tình hình làm thêm siêu âm, đường huyết, theo dõi tim thai, những khoản đó phải tự túi. Lập sổ càng sớm thì phân loại rủi ro thai kỳ càng sớm được làm ra. Phân loại rủi ro thai kỳ là xếp sản phụ theo mức rủi ro, chỉ khi phát hiện thuộc nhóm nguy cơ cao thì mới kịp chuyển sang bệnh viện đủ sức tiếp nhận. Bạn ở ngoài quận sở tại thì cứ lập sổ ở nơi đang sống, không cần về nơi đăng ký hộ khẩu - -### 3. Xét ba bệnh HIV, giang mai, viêm gan B ngay lần khám thai đầu tiên, và nếu ra kết quả vẫn có cách chặn miễn phí +- Nguồn: Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委) (2017). Quy chuẩn dịch vụ y tế công cộng cơ bản quốc gia, bản thứ ba. ；Chính quyền thành phố Tĩnh Giang (靖江市人民政府). Dự án dịch vụ y tế công cộng cơ bản quốc gia — quản lý sức khoẻ sản phụ (bảng danh mục dịch vụ công khai của địa phương). +- Ghi chú: 5 lần này là mức tối thiểu được miễn phí, không phải toàn bộ các lần khám thai. Bệnh viện sẽ tuỳ tình hình mà thêm siêu âm, đường huyết, theo dõi tim thai, những phần đó phải tự bỏ tiền ra. Lập sổ càng sớm thì phân loại nguy cơ thai kỳ càng sớm được làm ra. Phân loại nguy cơ thai kỳ là cách xếp sản phụ theo mức rủi ro cao thấp, chỉ khi xác định là nguy cơ cao thì mới chuyển sớm sang bệnh viện tiếp nhận nổi. Ở lại quận khác thì lập sổ theo nơi đang ở, không cần về hộ khẩu. +### 3. Lần khám thai đầu tiên thì loại luôn ba bệnh AIDS, giang mai, viêm gan B, có kết quả thì vẫn được chặn lây miễn phí -- Chi phí: Không tốn tiền, chương trình phòng lây truyền từ mẹ sang con cung cấp xét nghiệm miễn phí -- Nói thẳng: Ngay lần khám thai đầu tiên hãy xét cả ba bệnh HIV, giang mai, viêm gan B, xét nghiệm không tốn tiền của bạn. Ra kết quả dương tính cũng không phải ngõ cụt. Nhà nước cấp miễn phí thuốc và phác đồ để chặn con đường bệnh truyền từ mẹ sang con, đó gọi là chặn lây truyền mẹ con. -- Lợi ích: Lây truyền từ mẹ sang con nghĩa là bệnh truyền từ mẹ sang con. 《预防艾滋病、梅毒和乙肝母婴传播工作规范（2020 年版）》 có quy định về việc này. Vào tam nguyệt đầu hoặc lần khám tiền sản thứ nhất, bệnh viện phải thông báo các thông tin liên quan và「提供适宜、规范的免费检测」, nên ba xét nghiệm này không tốn tiền của bạn. Rồi căn cứ kết quả để tư vấn sau xét nghiệm hoặc chuyển tuyến. Ra kết quả dương tính cũng không phải ngõ cụt. Điều 44 của 《艾滋病防治条例》 quy định phải「免费提供预防艾滋病母婴传播的治疗和咨询」 cho sản phụ nhiễm HIV. Điều 43 yêu cầu có dịch vụ chặn lây, tức không để bệnh truyền sang con, rồi đến điều trị, thăm khám sau sinh, theo dõi trẻ sơ sinh và xét nghiệm. Khi mẹ dương tính với kháng nguyên bề mặt viêm gan B, trẻ sơ sinh phải tiêm đồng thời vắc-xin viêm gan B và globulin miễn dịch viêm gan B ngay khi sinh (xem mục 2 ở phần 20) +- Chi phí: Không tốn tiền, chương trình phòng chống lây truyền từ mẹ sang con cung cấp xét nghiệm miễn phí. +- Nói thẳng: Lần khám thai đầu tiên thì loại luôn ba bệnh AIDS, giang mai, viêm gan B, xét nghiệm không tốn tiền túi của bạn. Có kết quả dương tính cũng không phải ngõ cụt. Nhà nước cấp thuốc và phác đồ miễn phí để chặn đường bệnh đi từ mẹ sang con, việc đó gọi là chặn lây truyền mẹ con. +- Lợi ích: Lây truyền mẹ con là đường bệnh đi từ mẹ sang con. 《预防艾滋病、梅毒和乙肝母婴传播工作规范（2020 年版）》 có quy định về việc này. Vào giai đoạn đầu thai kỳ hoặc khi khám thai lần đầu, bệnh viện phải thông báo thông tin liên quan và 「提供适宜、规范的免费检测」, ba xét nghiệm này không tốn tiền túi người bệnh. Tùy kết quả lại tư vấn sau xét nghiệm hoặc chuyển tuyến. Kết quả dương tính cũng không phải ngõ cụt. Điều 44 của 《艾滋病防治条例》 quy định phải 「免费提供预防艾滋病母婴传播的治疗和咨询」 cho sản phụ nhiễm vi-rút HIV. Điều 43 yêu cầu cung cấp chặn lây (để bệnh không sang con), điều trị, thăm khám sau sinh, theo dõi trẻ sơ sinh và xét nghiệm. Khi mẹ dương tính với kháng nguyên bề mặt viêm gan B, trẻ sơ sinh phải tiêm đồng thời vaccine viêm gan B và globulin miễn dịch viêm gan B lúc mới sinh (xem mục 2 ở phần 20). - Mức bằng chứng: A -- Nguồn: 国家卫生健康委办公厅 (Văn phòng Ủy ban Sức khoẻ quốc gia) (2020). Quy định về công tác phòng lây truyền HIV, giang mai và viêm gan B từ mẹ sang con, 预防艾滋病、梅毒和乙肝母婴传播工作规范（2020 年版）. （Ủy ban Sức khoẻ Tự trị Uy Tân Cương đăng lại）；国务院 (Hội đồng Nhà nước) (2006). Quy định về phòng ngừa và kiểm soát AIDS, 艾滋病防治条例（第四十三、四十四条）. （Ủy ban Sức khoẻ thành phố Bắc Kinh đăng lại） -- Ghi chú: Ba xét nghiệm này càng sớm càng hữu ích, vì chặn lây và điều trị đều cần thời gian. Điều 3 của 《艾滋病防治条例》 còn viết「任何单位和个人不得歧视艾滋病病毒感染者、艾滋病病人及其家属」, các quyền lợi hợp pháp như kết hôn, tìm việc, khám chữa bệnh, đi học đều được luật bảo vệ. Phòng ngừa và xét nghiệm thường ngày xem phần 1. Sau hành vi nguy cơ cao, cách chặn sau phơi nhiễm xem phần 13 - -### 4. Cả thai kỳ không hút một điếu, không uống một giọt rượu, và người trong nhà cũng đừng hút trong nhà +- Nguồn: Văn phòng Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委办公厅) (2020). Quy chế công tác phòng ngừa lây truyền AIDS, giang mai và viêm gan B từ mẹ sang con, 预防艾滋病、梅毒和乙肝母婴传播工作规范（2020 年版）. （Ủy ban Y tế và Sức khoẻ Tân Lăng, 新疆维吾尔自治区卫生健康委员会, đăng lại）；Hội đồng Nhà nước (国务院) (2006). Nghị định phòng ngừa và kiểm soát AIDS, 艾滋病防治条例（第四十三、四十四条）. （Ủy ban Y tế và Sức khoẻ thành phố Bắc Kinh, 北京市卫生健康委员会, đăng lại） +- Ghi chú: Ba xét nghiệm này càng sớm càng hữu ích, vì chặn lây và điều trị đều cần thời gian. Điều 3 của nghị định còn viết 「任何单位和个人不得歧视艾滋病病毒感染者、艾滋病病人及其家属」, quyền lợi về kết hôn, tìm việc, khám chữa bệnh, học tập được pháp luật bảo vệ. Phòng ngừa và xét nghiệm thường ngày xem ở phần 1. Chặn sau phơi nhiễm sau khi có hành vi nguy cơ cao xem ở phần 13. +### 4. Suốt thai kỳ không được điếu thuốc nào, không được uống rượu nào, người trong nhà cũng đừng hút trong nhà -- Chi phí: Không tốn tiền, còn tiết kiệm được tiền thuốc lá và rượu. Chỗ khó là khiến người ở cùng nhà cùng cai -- Nói thẳng: Hút thuốc khi mang thai làm tăng xác suất đẻ non, trẻ sinh ra quá nhỏ, khoảm miệng hở vòm. Nguy cơ chảy máu bất thường khi mang thai và khi sinh tăng thẳng gấp đôi. Nguy cơ tử vong đột ngột ở trẻ sơ sinh cũng cao hơn. Người trong nhà hút trong nhà thì tính y như vậy. Bỏ thuốc lúc nào vẫn kịp. -- Lợi ích: Cách nói của Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ: hút thuốc khi mang thai khiến đứa trẻ「足月出生也可能过小」 và「可能早产」. Nó「使孕期和分娩期异常出血的风险翻倍」. Nó làm tăng nguy cơ dị tật bẩm sinh như hở môi, hở vòm. Nó còn「损害胎儿正在发育的肺和大脑，损害可持续到儿童期」. Trẻ của người mẹ hút thuốc khi mang thai, và trẻ sau khi sinh phải hít vào khói, có nguy cơ hội chứng tử vong đột ngột ở trẻ sơ sinh cao hơn. Lời khuyên chính thức là「最好在怀孕前就戒烟，但如果已经怀孕，戒了仍然有用」 và「什么时候戒都不晚」 +- Chi phí: Không tốn tiền, còn tiết kiệm tiền thuốc lá và rượu. Chỗ khó là khiến người ở cùng nhà cai theo. +- Nói thẳng: Hút thuốc khi mang thai làm xác suất sinh non, sinh ra nhỏ cân, khe hở môi vòm hành đều tăng lên. Nguy cơ chảy máu bất thường khi mang thai và khi sinh tăng gấp đôi. Nguy cơ đột tử ở trẻ sơ sinh cũng cao hơn. Người trong nhà hút thuốc trong nhà thì tính như nhau. Bỏ thuốc ở thời điểm nào vẫn còn kịp. +- Lợi ích: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ nói: hút thuốc khi mang thai khiến đứa trẻ 「足月出生也可能过小」「可能早产」. Nó làm 「使孕期和分娩期异常出血的风险翻倍」. Nó làm tăng nguy cơ dị tật bẩm sinh như hở môi, hở vòm hành. Nó còn 「损害胎儿正在发育的肺和大脑，损害可持续到儿童期」. Trẻ sơ sinh của mẹ hút thuốc khi mang thai, và trẻ sơ sinh sau khi sinh tiếp xúc với khói thuốc, có nguy cơ hội chứng đột tử cao hơn. Khuyến cáo chính thức là 「最好在怀孕前就戒烟，但如果已经怀孕，戒了仍然有用」「什么时候戒都不晚」 - Mức bằng chứng: A - Nguồn: Centers for Disease Control and Prevention. Smoking During Pregnancy. -- Ghi chú: Người khác hút trong nhà, khói thuốc lá phụ mà người mang thai hít vào vẫn tính với thai nhi. Nên người ở cùng nhà cũng đừng hút trong nhà. Rượu cũng vậy: không có mức uống nào được biết là an toàn, cách tiện nhất khi mang thai là không uống giọt nào. Riêng chuyện thuốc lá và rượu với người lớn, xem phần 2 - -### 5. Nếu có yếu tố nguy cơ tiền sản giật thai kỳ, từ sau tuần thai kỳ thứ 12 uống mỗi ngày một viên aspirin liều thấp +- Ghi chú: Người khác hút trong nhà thì khói thuốc sản phụ hít vào vẫn tính với thai nhi. Vì vậy người ở cùng nhà cũng đừng hút trong nhà. Rượu cũng vậy: không có mức uống nào được biết là an toàn, cách tiết kiệm công sức nhất khi mang thai là không uống một giọt nào. Bài toán tiền và sức lực của thuốc lá rượu với chính người lớn xem ở phần 2. +### 5. Có yếu tố nguy cơ cao tiền sản giật thì từ tuần thai kỳ thứ 12 uống mỗi ngày một viên aspirin liều thấp -- Chi phí: Vài chục tệ. Mỗi ngày một viên, phải uống liền tới khi sinh. Chỗ khó là không quên -- Nói thẳng: Tiền sản giật thai kỳ là một bệnh tăng huyết áp trong thai kỳ. Người thuộc nhóm nguy cơ cao là người đã mắc ở lần mang thai trước, đang mang đa thai, có tăng huyết áp mạn, có đái thường trước khi mang thai, có bệnh thận hoặc bệnh tự miễn. Người nguy cơ cao uống từ tuần thai kỳ thứ 12 mỗi ngày một viên aspirin liều thấp thì tiền sản giật thai kỳ giảm khoảng một phần rưỡi, còn đẻ non và trẻ chết quanh ngày sinh đều giảm khoảng hai phần mười. Người khác đừng tự uống. -- Lợi ích: Nhóm Dịch vụ Phòng ngừa Hoa Kỳ đưa ra khuyến nghị mức B: với người có nguy cơ cao về tiền sản giật thai kỳ, bắt đầu uống aspirin liều thấp (81 mg/ngày) sau tuần thai kỳ thứ 12 để phòng bệnh. Tiền sản giật thai kỳ là một bệnh tăng huyết áp trong thai kỳ. Gộp nhiều nghiên cứu lại, kết quả như sau, 95% CI là khoảng tin cậy. Tiền sản giật thai kỳ RR 0.85 (95% CI 0.75 đến 0.95, 16 nghiên cứu), thấp hơn khoảng 15%. Đẻ non RR 0.80 (95% CI 0.67 đến 0.95, 13 nghiên cứu), thấp hơn khoảng 20%. Tử vong chu sinh RR 0.79 (95% CI 0.66 đến 0.96, 11 nghiên cứu), thấp hơn khoảng 21%; tử vong chu sinh là cái chết trong khoảng thời gian quanh lúc sinh. Nhỏ hơn tuổi thai và cơn ứng thai kém RR 0.82 (95% CI 0.68 đến 0.99, 16 nghiên cứu), thấp hơn khoảng 18%; nghĩa là thai nhi lớn nhỏ hơn mức bình thường ở tuổi thai đó +- Chi phí: Vài chục tệ. Mỗi ngày một viên, phải uống liên tới ngày sinh. Chỗ khó là không quên mỗi ngày. +- Nói thẳng: Tiền sản giật là một bệnh tăng huyết áp trong thai kỳ. Người lần trước đã mắc, mang đa thai, có tăng huyết áp mạn, có đái thường trước khi mang thai, có bệnh thận hoặc bệnh tự miễn thuộc loại nguy cơ cao. Người nguy cơ cao uống mỗi ngày một viên aspirin liều thấp từ tuần thai kỳ thứ 12, tiền sản giật thấp khoảng một phần rưỡi, sinh non và trẻ chết trước hoặc sau khi sinh thấp khoảng hai phần mươi. Người khác đừng tự uống. +- Lợi ích: Nhóm dịch vụ phòng bệnh Hoa Kỳ đưa ra khuyến cáo mức B: với người có nguy cơ cao tiền sản giật, bắt đầu uống aspirin liều thấp (81 miligam/ngày) sau tuần thai kỳ thứ 12 để phòng bệnh. Tiền sản giật là một bệnh tăng huyết áp trong thai kỳ. Gộp nhiều nghiên cứu lại với nhau thì kết quả như sau, 95% CI là khoảng tin cậy. Tiền sản giật RR 0.85 (95% CI 0.75 đến 0.95, 16 nghiên cứu), thấp khoảng 15%. Sinh non RR 0.80 (95% CI 0.67 đến 0.95, 13 nghiên cứu), thấp khoảng 20%. Tử vong chu sinh RR 0.79 (95% CI 0.66 đến 0.96, 11 nghiên cứu), thấp khoảng 21%; tử vong chu sinh là cái chết trong khoảng thời gian trước và sau khi sinh. Trẻ nhỏ hơn tuổi thai／tăng trưởng trong tử cung bị hạn chế RR 0.82 (95% CI 0.68 đến 0.99, 16 nghiên cứu), thấp khoảng 18%; nói là thai nhi lớn nhỏ hơn mức trung bình so với cùng số tuần thai kỳ. - Mức bằng chứng: A - Nguồn: US Preventive Services Task Force (2021). Aspirin Use to Prevent Preeclampsia and Related Morbidity and Mortality: Preventive Medication. -- Ghi chú: Các nhóm được tính là nguy cơ cao: đã mắc tiền sản giật thai kỳ ở lần mang thai trước, đang mang thai đôi hoặc đa thai, có tăng huyết áp mạn. Ngoài ra còn có đái thường tuýp 1 hoặc tuýp 2 trước khi mang thai, bệnh thận, và bệnh tự miễn như lupus ban đỏ hệ thống, hội chứng kháng phospholipid. Không thuộc mấy nhóm đó thì đừng tự uống. Thuốc này phải có bác sĩ kê đơn, khi khám thai hãy nói rõ bệnh sử để bác sĩ đánh giá. Ở trong nước bán chủ yếu là viên bao tanh 100 mg, uống theo chỉ định của bác sĩ - -### 6. Từ tuần thai kỳ thứ 24 trở đi, làm một lần sàng lọc đái thường thai kỳ, đừng ngại uống nước đường +- Ghi chú: Các trường hợp được tính là nguy cơ cao: lần trước đã mắc tiền sản giật, mang thai đôi hoặc đa thai, có tăng huyết áp mạn; và đã có đái thường tuýp 1 hoặc tuýp 2 trước khi mang thai, có bệnh thận, hay có bệnh tự miễn như lupus ban đỏ hệ thống, hội chứng kháng phospholipid. Không thuộc mấy nhóm đó thì đừng tự uống. Thuốc này phải có bác sĩ kê đơn, lúc khám thai hãy nói rõ tiền sử bệnh để bác sĩ đánh giá. Ở trong nước bán chủ yếu là viên bao tan trong ruột 100 miligam, uống theo chỉ dẫn của bác sĩ. +### 6. Từ tuần thai kỳ thứ 24 trở đi, làm một lần sàng lọc đái thường thai kỳ, đừng chê uống nước đường phiền -- Chi phí: Vài chục tới hơn một trăm tệ. Phải đến lúc nhịn đói, lấy máu mấy lần, mất nửa buổi sáng -- Nói thẳng: Đái thường thai kỳ là đường huyết cao chỉ xuất hiện sau khi mang thai. Từ tuần thai kỳ thứ 24, dành nửa buổi sáng uống một lần nước đường và lấy máu mấy lần là ra kết quả. Không xét cũng không để ý, thì thai nhi to quá, khó đẻ, phải mổ, tiền sản giật thai kỳ và hạ đường huyết ở trẻ sơ sinh đều tăng nguy cơ. -- Lợi ích: Khuyến nghị mức B của Nhóm Dịch vụ Phòng ngừa Hoa Kỳ: sàng lọc đái thường thai kỳ ở tuần thai kỳ thứ 24 trở đi với phụ nữ mang thai không có triệu chứng, tức tự thấy không khó chịu. Đái thường thai kỳ là đường huyết cao chỉ xuất hiện sau khi mang thai. Nếu chỉ sàng lọc một lần thì lần đó phải đặt ở tuần thai kỳ thứ 24 trở đi. Về việc sàng lọc trước tuần thai kỳ thứ 24 có nên không, nhóm đã kết luận rằng bằng chứng hiện có chưa đủ để phân biệt lợi và hại (mức I). Đái thường thai kỳ không được kiểm soát làm tăng nguy cơ: thai nhi quá to, khó đẻ, mổ lấy thai, tiền sản giật thai kỳ và hạ đường huyết ở trẻ sơ sinh +- Chi phí: Vài chục tới hơn một trăm tệ. Phải đến lúc nhịn đói, lấy máu vài lần, mất nửa buổi sáng. +- Nói thẳng: Đái thường thai kỳ là tình trạng đường huyết cao xuất hiện sau khi mang thai. Sau tuần thai kỳ thứ 24, bỏ ra nửa buổi sáng để uống một lần nước đường và lấy máu vài lần là phát hiện được. Không xét nghiệm cũng không để ý thì nguy cơ thai nhi to quá, khó sinh, phải mổ, tiền sản giật và hạ đường huyết ở trẻ sơ sinh đều tăng lên. +- Lợi ích: Nhóm dịch vụ phòng bệnh Hoa Kỳ đưa ra khuyến cáo mức B: với sản phụ không triệu chứng (tự thấy không khó chịu), sàng lọc đái thường thai kỳ từ tuần thai kỳ thứ 24 trở đi. Đái thường thai kỳ là tình trạng đường huyết cao xuất hiện sau khi mang thai. Nếu chỉ sàng lọc một lần thì lần đó phải rơi vào tuần thai kỳ thứ 24 trở đi. Sàng lọc trước tuần thai kỳ thứ 24 có lợi không, kết luận của nhóm là bằng chứng hiện có chưa đủ để cân lợi và hại (mức I). Không xử lý đái thường thai kỳ thì đẩy lên nguy cơ của mấy thứ này: thai nhi quá to (thai nhi lớn quá), khó sinh, mổ lấy thai, tiền sản giật và hạ đường huyết ở trẻ sơ sinh. - Mức bằng chứng: A - Nguồn: US Preventive Services Task Force (2021). Gestational Diabetes: Screening. -- Ghi chú: Trong số người phát hiện bệnh, đa số kiểm soát được đường huyết bằng cách điều chỉnh ăn uống và vận động, số ít phải tiêm insulin. Đừng vì sợ phải tiêm mà bỏ lần xét này. Sau khi sinh còn phải xét lại đường huyết: người từng mắc đái thường thai kỳ sau này có nguy cơ mắc đái thường tuýp 2 cao hơn rõ rệt - -### 7. Thuộc lòng danh sách「cần vào viện ngay」này, áp dụng trong thai kỳ và suốt một năm sau sinh +- Ghi chú: Trong số người phát hiện ra, đa số chỉ cần điều chỉnh ăn uống và vận động là kiểm soát được đường huyết, số phải tiêm insulin là thiểu số. Đừng vì sợ phát hiện ra rồi phải tiêm thuốc mà bỏ lần này. Sau khi sinh còn phải xét lại đường huyết: người đã mắc đái thường thai kỳ thì nguy cơ mắc đái thường tuýp 2 về sau cao hơn rõ rệt. +### 7. Thuộc lòng bảng 「phải vào viện ngay」 này, trong thai kỳ và trong một năm sau khi sinh đều tính -- Chi phí: Không tốn tiền. Mất năm phút để ghi nhớ, tốt nhất là cho người ở cùng nhà biết luôn -- Nói thẳng: Chỉ cần một dấu hiệu nào trong danh sách là đi viện trong ngày, đừng ở nhà chờ qua đêm. Đau đầu không dịu hoặc nặng dần, nhìn thấy khác đi, sốt tới 38 ℃. Tay hoặc mặt sưng rõ, thở không nổi, đau ngực hoặc tim đập nhanh. Đau bụng dữ dội không dịu, thai ngừng cử động hoặc giảm hẳn, chảy máu hoặc rò dịch ở âm đạo. Một tay hoặc một chân sưng và đau, có ý nghĩ làm hại mình hoặc đứa trẻ. Trong vòng một năm sau sinh vẫn tính. -- Lợi ích: Các dấu hiệu cảnh báo khẩn cấp về mẹ và thai mà Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ liệt kê gồm: đau đầu không dịu hoặc nặng dần, chóng mặt hoặc ngất, nhìn thấy khác đi, thân nhiệt đạt 38 ℃ trở lên; tay hoặc mặt sưng rõ, có ý nghĩ làm hại mình hoặc đứa trẻ, khó thở, đau ngực hoặc tim đập quá nhanh; buồn nôn và nôn dữ dội, đau bụng dữ dội không dịu, thai ngừng cử động hoặc giảm hẳn trong thai kỳ; chảy máu âm đạo hoặc rò dịch trong thai kỳ, chảy máu âm đạo hoặc dịch bất thường sau sinh; một tay hoặc một chân sưng nặng, đỏ hoặc đau. Những vấn đề này đều có thể xảy ra trong vòng một năm sau sinh, không phải「hết tháng nằm hồi là xong」 +- Chi phí: Không tốn tiền. Dành năm phút để nhớ, tốt nhất người ở cùng nhà cũng biết. +- Nói thẳng: Chỉ cần một dấu hiệu trong danh sách này xuất hiện thì trong ngày đi viện, đừng ở nhà chờ qua đêm. Đau đầu không đỡ hoặc nặng dần, nhìn thấy hình ảnh đổi màu, sốt tới 38 ℃. Tay hoặc mặt sưng nặng, thở không ra, đau ngực hoặc tim đập nhanh. Bụng đau dữ dội không dịu, thai cử động dừng hoặc giảm rõ, chảy máu hoặc rỉ dịch ở âm đạo. Một cánh tay hoặc một chân sưng và đau, có ý nghĩ làm hại mình hoặc con. Trong một năm sau khi sinh cũng tính. +- Lợi ích: Các tín hiệu cảnh báo cấp cứu về sản phụ do Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ liệt kê gồm: đau đầu không đỡ hoặc nặng dần, hoa mắt chóng mặt hoặc ngất, nhìn thấy hình ảnh đổi màu, thân nhiệt đạt 38 ℃ hoặc cao hơn; tay hoặc mặt sưng nặng, có ý nghĩ làm hại mình hoặc con, khó thở, đau ngực hoặc nhịp tim quá nhanh; buồn nôn nôn ói nặng, bụng đau dữ dội không dịu, thai cử động dừng hoặc giảm rõ trong thai kỳ; chảy máu âm đạo hoặc rỉ dịch trong thai kỳ, sau khi sinh chảy máu âm đạo hoặc dịch tiết bất thường; một cánh tay hoặc một chân sưng nặng, đỏ hoặc đau. Những vấn đề này đều có thể xảy ra trong một năm sau khi sinh, không phải 「ra khỏi hậu sản là không sao」 - Mức bằng chứng: A -- Nguồn: Centers for Disease Control and Prevention. Hear Her — Urgent Maternal Warning Signs. ；World Health Organization. Maternal mortality fact sheet. -- Ghi chú: Công dụng của danh sách này là tách「chịu thêm chút nữa」 khỏi「phải đi ngay」. Đau đầu kèm nhìn mờ, cộng thêm đau vùng bụng trên, là dấu hiệu hướng tới tiền sản giật thai kỳ, tức bệnh tăng huyết áp trong thai kỳ. Một bắp chân sưng và đau là dấu hiệu hướng tới huyết khối tĩnh mạch sâu (xem phần 13). Theo Tổ chức Y tế Thế giới, chảy máu nặng sau sinh có thể cướp mạng một người phụ nữ khoẻ mạnh trong vài giờ. Có một dấu hiệu là đi thẳng viện, đừng ở nhà chờ qua đêm - -### 8. Vỡ ối thì nằm ngửa tại chỗ, kê cao mông, gọi 120, đừng đi lại và đừng tắm +- Nguồn: Centers for Disease Control and Prevention. Hear Her — Urgent Maternal Warning Signs. ；Tổ chức Y tế Thế giới. Maternal mortality fact sheet. +- Ghi chú: Cái lợi của bảng này là tách 「nhịn thêm một chút」 khỏi 「phải đi ngay」. Đau đầu kèm nhìn thấy mờ, cộng thêm đau vùng bụng trên, là dấu hiệu hướng tới tiền sản giật (bệnh tăng huyết áp trong thai kỳ). Một bắp chân sưng và đau là dấu hiệu hướng tới huyết khối tĩnh mạch sâu (xem ở phần 13). Băng huyết sau sinh theo cách nói của Tổ chức Y tế Thế giới có thể cướp mạng một người phụ nữ khỏe mạnh trong vài giờ. Chỉ cần một dấu hiệu là đi viện thẳng, đừng ở nhà chờ qua đêm trước. +### 8. Vỡ ối thì nằm bẹp tại chỗ, kê hông lên, gọi 120, không đi lại và không tắm - Chi phí: Không tốn tiền -- Nói thẳng: Vỡ ối rồi thì không phải xếp đồ chậm rãi rồi mới đi. Cách đúng là nằm xuống tại chỗ, kê cao mông, gọi 120. Đứng và đi lại thì nước ối thoát nhanh hơn, dây rốn có thể trôi theo ra ngoài. Đó là cấp cứu có thể lấy mạng đứa trẻ. -- Lợi ích: Cách xử lý do Ủy ban Sức khoẻ thành phố Bắc Kinh đưa ra: sau khi vỡ ối thì「马上找个地方平躺下来，防止羊水流出来」「不要随意走动」, và「立即拨打 120 急救电话」. Trong lúc chờ xe, báo người nhà mang đồ tới bệnh viện gặp mình. Tài liệu của các cơ sở y tế giải thích thêm lý do: khi đang đứng thì nước ối thoát nhanh hơn, dây rốn của thai nhi có thể trôi ra ngoài theo, đó gọi là sa dây rốn, là cấp cứu đe dọa thai nhi. Vì vậy phải nằm xuống và kê cao mông, có thể chọn tư thế nằm nghiêng bên trái +- Nói thẳng: Vỡ ối không phải là xách đồ đi dần rồi mới lên đường. Cách đúng là nằm bẹp tại chỗ, kê mông lên, gọi 120. Đứng mà đi lại thì nước ối chảy nhanh hơn, dây cuốn rốn có thể bị đẩy theo ra ngoài. Đó là cấp cứu có thể cướp mạng đứa trẻ. +- Lợi ích: Cách xử lý do Ủy ban Y tế và Sức khoẻ thành phố Bắc Kinh đưa ra: phát hiện vỡ ối thì 「马上找个地方平躺下来，防止羊水流出来」「不要随意走动」, và 「立即拨打 120 急救电话」. Trong lúc chờ xe thì báo người nhà đem đồ tới bệnh viện gặp. Tài liệu phổ biến của cơ sở y tế giải thích thêm lý do: khi đứng thì nước ối mất nhanh hơn, dây cuốn rốn của thai nhi có thể trượt ra ngoài, gọi là sa dây cuốn, đây là cấp cứu nguy kịch với thai nhi. Vì vậy cần nằm xuống, kê hông lên, có thể chọn tư thế nằm nghiêng bên trái. - Mức bằng chứng: B -- Nguồn: 北京市卫生健康委员会 (Ủy ban Sức khoẻ thành phố Bắc Kinh). Chủ đề cấp cứu hằng tuần: bà mẹ sắp sinh bị vỡ ối sớm thì xử lý thế nào, 每周急救话题：准妈妈羊水早破怎样办？. ；北京市顺义区妇幼保健院 (Bệnh viện Sức khoẻ bà mẹ và trẻ em quận Thuận Nghĩa, Bắc Kinh). Giai đoạn cuối thai kỳ bị vỡ ối tại nhà thì làm gì, 孕晚期在家破水怎么办？. -- Ghi chú: Ba dấu hiệu sắp sinh là co bóp tử cung đều đặn, ra máu âm đạo, và vỡ ối. Hai dấu hiệu đầu còn xếp đồ cho thoải mái, vỡ ối là dấu hiệu duy nhất đòi「nằm xuống tại chỗ rồi gọi xe」. Sau khi vỡ ối mà tắm, đi toilet, tự lái xe, đều là cách làm sai điển hình. Ghi lại thời điểm vỡ ối và màu nước ối, rồi báo bác sĩ. Nước ối vàng xanh, hoặc có lẫn hạt phân su, là dấu hiệu thai nhi bị ngạt, tức thai thiếu oxy trong tử cung. Nếu đồng thời có co bóp tử cung đều đặn, thai cử động giảm rõ, sốt kèm rét run, thì càng phải nhấn với bác sĩ rằng đây là ca cấp cứu - -### 9. Muốn giảm đau khi sinh thì nói thẳng, điều này không làm tăng nguy cơ mổ lấy thai +- Nguồn: Ủy ban Y tế và Sức khoẻ thành phố Bắc Kinh (北京市卫生健康委员会). Chủ đề cấp cứu hằng tuần: mẹ bầu bị vỡ ối sớm thì xử lý thế nào? ；Bệnh viện Chăm sóc Sức khoẻ Phụ nữ và Trẻ em quận Tuy Dương, thành phố Bắc Kinh (北京市顺义区妇幼保健院). Vỡ ối tại nhà vào cuối thai kỳ thì làm sao? +- Ghi chú: Ba dấu hiệu sắp sinh là co thắt tử cung đều, ra dịch hồng, vỡ ối. Hai cái trước có thể xách đồ đi từ từ, chỉ vỡ ối là 「nằm xuống gọi xe tại chỗ」. Sau khi vỡ ối mà tắm, đi vệ sinh, tự lái xe đều là cách làm sai điển hình. Ghi lại giờ vỡ ối và màu nước ối, rồi nói với bác sĩ. Nước ối vàng xanh, hoặc có hạt phân suối, là dấu hiệu thai nhi bị ngạt (thai nhi thiếu oxy trong tử cung). Nếu đồng thời có co thắt tử cung đều, thai cử động giảm rõ, sốt kèm rét run, càng phải nhấn mạnh với bác sĩ rằng đây là trường hợp cấp cứu. +### 9. Muốn đẻ không đau thì cứ nói thẳng, nó không làm tăng nguy cơ mổ lấy thai -- Chi phí: Vài trăm tới hơn một nghìn tệ, giá ở mỗi nơi mỗi khác. Phải hỏi trước bệnh viện và khoa gây mê có làm được không. Chỗ khó là phải tự mình mở lời -- Nói thẳng: Giảm đau khi sinh là tiêm một liều thuốc gây mê vào thắt lưng để đẩy xuống cơn đau sinh nở. Nó hiệu quả hơn các cách giảm đau khác, và không làm tăng khả năng bạn phải mổ. Câu「tiêm giảm đau xong thì đau lưng cả đời」cũng chẳng có cơ sở. Giờ nó đã có mục tính phí riêng, bệnh viện cũng sẵn sàng làm hơn trước. -- Lợi ích: Phân tích tổng hợp của Cochrane kết luận: gây mê giảm đau ngoài màng cứng hiệu quả hơn các cách không dùng ngoài màng cứng, cả trong giảm đau khi sinh lẫn nâng cao mức hài lòng của sản phụ với cách giảm đau. Giảm đau khi sinh dùng đúng là gây mê ngoài màng cứng. Cùng nghiên cứu đó còn nói「硬膜外镇痛对剖宫产风险和长期腰痛没有影响」. Hiện tượng sinh có dùng dụng cụ, tức dùng dụng cụ để đỡ đứa trẻ ra, tăng lên không còn thấy trong các nghiên cứu từ 2005 trở đi. Từ năm 2024, hướng dẫn lập danh mục giá dịch vụ y tế sản khoa của Cục Quản lý bảo hiểm y tế quốc gia tách riêng thành các mục「分娩镇痛」「导乐分娩」「亲情陪产」. Lý do đúng là trước đó chưa có mục giá riêng, nên động lực của bệnh viện trong việc cung cấp giảm đau khi sinh bị giảm +- Chi phí: Vài trăm tới hơn một nghìn tệ, giá ở mỗi nơi mỗi khác. Phải hỏi trước bệnh viện và khoa gây mê có làm được không. Chỗ khó là phải tự nêu ra. +- Nói thẳng: Đẻ không đau là tiêm một mũi thuốc mê vào lưng để dập cơn đau khi sinh. Nó hiệu quả hơn các cách giảm đau khác, và cũng không làm bạn dễ phải mổ hơn. 「Tiêm đẻ không đau thì đau lưng cả đời」 cũng không có cơ sở. Giờ nó đã có mục giá tính tiền riêng, bệnh viện cũng sẵn sàng làm hơn trước. +- Lợi ích: Kết luận của phân tích tổng hợp Cochrane là: gây mê ngoài màng cứng có hiệu quả hơn phương pháp không dùng ngoài màng cứng trong việc giảm đau khi sinh và nâng cao mức hài lòng của sản phụ với cách giảm đau. Đẻ không đau dùng chính gây mê ngoài màng cứng. Cùng tổng hợp đó còn viết 「硬膜外镇痛对剖宫产风险和长期腰痛没有影响」. Hiện tượng sinh có hỗ trợ bằng dụng cụ (dùng dụng cụ để kéo thai ra) tăng lên không còn thấy trong các nghiên cứu sau năm 2005. Từ năm 2024, hướng dẫn lập nhóm dự án giá dịch vụ y tế sản phụ của Cục Quản lý bảo hiểm y tế quốc gia đã tách riêng 「分娩镇痛」「导乐分娩」「亲情陪产」 thành các mục riêng. Lý do đúng là trước đây chưa có mục giá riêng, nên động lực của bệnh viện trong việc cung cấp giảm đau khi sinh bị giảm. - Mức bằng chứng: A -- Nguồn: Anim-Somuah M, Smyth RMD, Cyna AM, Cuthbert A (2018). Epidural versus non-epidural or no analgesia for pain management in labour. Cochrane Database of Systematic Reviews, (5), CD000331. ；国家医疗保障局 (Cục Quản lý bảo hiểm y tế quốc gia) (2024). Tái dựng danh mục giá dịch vụ sản khoa, góp phần xây dựng xã hội thân thiện với sinh con, 重构产科服务价格项目 助力构建生育友好型社会. -- Ghi chú: Hai câu「tiêm giảm đau xong đau lưng cả đời」 và「tiêm giảm đau xong khó đẻ phải mổ」 lây lan rộng nhất đều không có cơ sở. Cochrane gộp gây mê ngoài màng cứng với các cách giảm đau khác lại, và kết luận nó không ảnh hưởng tới nguy cơ mổ lấy thai cũng không ảnh hưởng tới đau lưng lâu dài. Tác dụng phụ có thật là những cái sau: huyết áp thấp, sốt, chân mềm yếu, tiểu tiện khó, và giai đoạn một cùng giai đoạn hai của cuộc sinh kéo dài hơn, dễ phải dùng thuốc kích thích tử cung hơn. Có làm hay không thì bạn tự quyết, nhưng đừng bỏ vì lời đồn. Có làm được hay không còn tuỳ bệnh viện có bác sĩ gây mê trực 24 giờ hay không, nên hỏi rõ lúc khám thai - -### 10. Không có chỉ định y khoa thì đừng tự yêu cầu mổ lấy thai, cũng đừng mổ chỉ để chọn ngày +- Nguồn: Anim-Somuah M, Smyth RMD, Cyna AM, Cuthbert A (2018). Epidural versus non-epidural or no analgesia for pain management in labour. Cochrane Database of Systematic Reviews, (5), CD000331. ；Cục Quản lý bảo hiểm y tế quốc gia (国家医疗保障局) (2024). Tái cấu trúc nhóm dự án giá dịch vụ sản phụ, góp phần xây dựng xã hội thân thiện với sinh sản. +- Ghi chú: Hai lời 「tiêm đẻ không đau thì đau lưng cả đời」 và 「tiêm đẻ không đau thì khó sinh phải mổ」 là hai lời đồn lan xa nhất, đều không có cơ sở. Cochrane gộp gây mê ngoài màng cứng với các cách giảm đau khác, kết luận là nó không ảnh hưởng tới nguy cơ mổ lấy thai và cũng không gây đau lưng lâu dài. Tác dụng phụ có thật là những cái này: huyết áp thấp, sốt, chân mềm yếu, không tiểu được, và giai đoạn rộn con thứ nhất và thứ hai kéo dài hơn, dễ phải dùng thuốc co tử cung hơn. Làm hay không thì bạn tự quyết, nhưng đừng vì lời đồn mà bỏ. Có làm được hay không còn phụ thuộc bệnh viện có bác sĩ gây mê trực 24 giờ hay không, hỏi rõ ngay lúc khám thai. +### 10. Không có chỉ định y tế thì đừng chủ động xin mổ lấy thai, cũng đừng mổ để chọn ngày -- Chi phí: Không tốn tiền. Chỗ khó là chịu đựng lời của người nhà -- Nói thẳng: Cách nói của Tổ chức Y tế Thế giới là: khi tỷ lệ mổ lấy thai ở một nơi vượt quá 10%, số tử vong ở bà mẹ và trẻ sơ sinh không còn tiếp tục giảm. Chỗ nào cần mổ thì phải mổ. Mổ chỉ để chọn ngày, hoặc vì sợ đau, chỉ là phải chịu một cuộc mổ bụng lớn mà không đổi lại được gì. -- Lợi ích: Tổ chức Y tế Thế giới phát biểu năm 2015: khi tỷ lệ mổ lấy thai trong một cộng đồng tăng lên khoảng 10%, số tử vong ở sản phụ và trẻ sơ sinh giảm xuống. Vượt quá 10% thì không có bằng chứng cho thấy tỷ lệ tử vong còn cải thiện thêm. Tổ chức Y tế Thế giới đồng thời nhấn mạnh, phải cố hết sức để cung cấp mổ lấy thai cho sản phụ thật sự cần, chứ không phải để đạt một tỷ lệ nào đó +- Chi phí: Không tốn tiền. Chỗ khó là chịu được lời của người nhà. +- Nói thẳng: Cách nói của Tổ chức Y tế Thế giới là khi tỷ lệ mổ lấy thai ở một địa phương vượt quá 10%, số tử vong ở mẹ và trẻ sơ sinh không còn giảm thêm. Cần mổ thì cứ mổ. Mổ chỉ để chọn ngày hoặc vì sợ đau thì chỉ là phải chịu một cuộc mổ bụng trắng công. +- Lợi ích: Tuyên bố năm 2015 của Tổ chức Y tế Thế giới: khi tỷ lệ mổ lấy thai trong một cộng đồng tăng lên khoảng 10%, số tử vong ở sản phụ và trẻ sơ sinh sẽ giảm. Sau khi vượt quá 10%, không có bằng chứng cho thấy tỷ lệ tử vong còn cải thiện thêm. Tổ chức Y tế Thế giới đồng thời nhấn mạnh phải cố hết sức cung cấp mổ lấy thai cho sản phụ thật sự cần, thay vì để đạt một tỷ lệ nào đó. - Mức bằng chứng: A -- Nguồn: World Health Organization (2015). WHO Statement on Caesarean Section Rates（WHO/RHR/15.02）. -- Ghi chú: Khi đúng là cần về mặt y khoa, mổ lấy thai là cuộc phẫu thuật cứu mạng, cứ mổ đi, đừng chịu đựng. Cái cần tránh là mổ khi không thật sự cần về mặt y khoa, chẳng hạn vì sợ đau, vì chọn ngày đẹp, hay vì tính tuổi vào trường của con. Mổ lấy thai là phẫu thuật bụng lớn, hồi phục chậm hơn, lần mang thai sau có nguy cơ cao hơn về nhau thai bám màng, nhau cấy vào tử cung, và vỡ tử cung. Điều này ảnh hưởng trực tiếp tới việc bạn còn muốn sinh đứa thứ hai hay không - -### 11. Trước khi mang thai, xác nhận tình hình tham gia bảo hiểm thai sản, vợ chồng chưa có việc làm cũng được hoàn phí khám chữa bệnh thai sản +- Nguồn: Tổ chức Y tế Thế giới (2015). WHO Statement on Caesarean Section Rates（WHO/RHR/15.02）. +- Ghi chú: Khi thật sự cần về y khoa thì mổ lấy thai là phẫu thuật cứu mạng, nên làm thì cứ làm, đừng cố chịu đựng. Cái cần tránh là không cần về y khoa mà vẫn mổ, chẳng hạn vì sợ đau, vì chọn ngày đẹp, vì tính tuổi vào trường của con. Mổ lấy thai là phẫu thuật bụng lớn, hồi phục chậm hơn, và lần sau nguy cơ nhau tiền màng, nhau cấy, vỡ tử cung cao hơn. Điều này liên quan trực tiếp tới việc bạn còn muốn sinh con thứ hai hay không. +### 11. Trước khi mang thai đã xác nhận tình trạng tham gia bảo hiểm sinh sản, người vợ không có việc làm cũng được hoàn phí y tế sinh sản -- Chi phí: Không tốn tiền. Gọi một lần đường dây hỏi bảo hiểm xã hội hoặc bảo hiểm y tế -- Nói thẳng: Khi đơn vị đóng bảo hiểm thai sản, phí khám chữa bệnh sinh con và trợ cấp trong thời gian nghỉ phép thai sản đều lấy từ quỹ đó. Nếu vợ hoặc chồng chưa có việc làm, phần phí khám chữa bệnh cũng được hoàn. Đủ số tháng đóng chưa, sinh con ở nơi khác có phải đăng ký trước không, đều phải hỏi trước, sau đó không bù lại được. -- Lợi ích: Điều 54 của 《社会保险法》 quy định: nơi đơn vị sử dụng lao động đã đóng bảo hiểm thai sản thì người lao động của đơn vị được hưởng trợ cấp bảo hiểm thai sản. Nếu vợ hoặc chồng của người lao động không có việc làm thì được hưởng phần chi phí khám chữa bệnh thai sản theo quy định của nhà nước, kinh phí lấy từ quỹ bảo hiểm thai sản. Nói ngắn gọn, đơn vị đã đóng bảo hiểm thai sản thì người lao động được hưởng trợ cấp, và phí sinh con của người phối ngẫu không có việc làm cũng lấy từ quỹ đó. Trợ cấp bảo hiểm thai sản chia làm hai phần: một là chi phí khám chữa bệnh thai sản, tức phí khám chữa bệnh của riêng việc sinh con; hai là trợ cấp thai sản, tức khoản tiền trả cho bạn trong thời gian nghỉ phép thai sản. Điều 56 liệt kê ba trường hợp được nhận trợ cấp thai sản: nữ công nhân sinh con và được nghỉ phép thai sản, được nghỉ phép cho phẫu thuật kế hoạch hoá gia đình, và các trường hợp khác theo quy định của pháp luật. Trợ cấp thai sản「按照职工所在用人单位上年度职工月平均工资计发」, tức tính trên mức lương trung bình hằng tháng của người lao động tại đơn vị của bạn trong năm trước, không phải mức lương của chính bạn +- Chi phí: Không tốn tiền. Gọi một lần đường dây tư vấn bảo hiểm xã hội hoặc bảo hiểm y tế. +- Nói thẳng: Đơn vị đã đóng bảo hiểm sinh sản thì phí y tế khi sinh con và khoản trợ cấp trong thời gian nghỉ thai sản đều lấy từ quỹ đó. Người vợ không có việc làm thì phần phí y tế cũng được hoàn. Đủ hay chưa đủ số tháng tham gia, sinh ở nơi khác có phải đăng ký trước không, đều phải hỏi trước, sau đó không bù lại được. +- Lợi ích: Điều 54 của Bộ luật Bảo hiểm xã hội: người sử dụng lao động đã nộp phí bảo hiểm sinh sản thì người lao động của họ được hưởng chế độ bảo hiểm sinh sản. Người vợ không có việc làm của người lao động được hưởng chế độ phí y tế sinh sản theo quy định của nhà nước, kinh phí lấy từ quỹ bảo hiểm sinh sản. Nói gọn, đơn vị đã đóng bảo hiểm sinh sản thì người lao động được hưởng chế độ, và phí y tế khi người vợ không có việc làm của người lao động đó sinh con cũng lấy từ quỹ ấy. Chế độ bảo hiểm sinh sản chia làm hai phần: một là phí y tế sinh sản, tức chi phí y tế của chuyện sinh con; hai là trợ cấp sinh sản, tức số tiền trả cho bạn trong thời gian nghỉ thai sản. Điều 56 liệt kê ba trường hợp được nhận trợ cấp sinh sản: nữ lao động sinh con được nghỉ thai sản, được nghỉ cho phẫu thuật kế hoạch hoá gia đình, và các trường hợp khác do pháp luật quy định. Trợ cấp sinh sản 「按照职工所在用人单位上年度职工月平均工资计发」, tính theo mức lương trung bình hằng tháng của người lao động ở đơn vị bạn trong năm trước, không phải mức lương của chính bạn. - Mức bằng chứng: A -- Nguồn: 全国人大常委会 (Ủy ban Thường vụ Quốc hội) (2010). Luật Bảo hiểm xã hội của nước Cộng hoà Nhân dân Trung Quốc, 中华人民共和国社会保险法（第五十四、五十六条）. （Cục Thuế tỉnh Quảng Đông, Cục Quản lý thuế quốc gia đăng lại） -- Ghi chú: Có hai việc phải hỏi rõ trước, vì sau đó không bù lại được. Một là số tháng đóng liên tục có đủ không: mỗi nơi đều đặt ra số tháng tối thiểu, không đủ thì không được hưởng. Hai là sinh con ngoài nơi tham gia bảo hiểm thì có phải đăng ký sinh con ngoài địa phương trước không: không đăng ký thì có thể phải tự ứng tiền trước, về hoàn thủ bằng tay sau, thậm chí không hoàn được. Số ngày nghỉ phép thai sản và cách tính trợ cấp thai sản xem mục 2 ở phần 18 - -### 12. Trước khi xuất viện, làm xong Giấy khai sinh, nghĩ trước tên con và đừng viết sai chữ +- Nguồn: Ủy ban Thường vụ Quốc hội Nhà nước (全国人大常委会) (2010). Bộ luật Bảo hiểm xã hội Trung Quốc, 中华人民共和国社会保险法（第五十四、五十六条）. （Cục Thuế tỉnh Quảng Đông, 国家税务总局广东省税务局, đăng lại） +- Ghi chú: Có hai việc phải hỏi trước, vì sau đó không bù lại được. Một là số tháng đóng liên tục có đủ không: mỗi nơi đều đặt số tháng tối thiểu, không đủ thì không được hưởng. Hai là nếu sinh ngoài nơi tham gia bảo hiểm thì có phải đăng ký sinh sản ở địa phương khác trước không: không đăng ký thì có thể phải tự ứng trước rồi về thủ công hoàn lại, thậm chí không hoàn được. Số ngày nghỉ thai sản và cách tính trợ cấp sinh sản xem ở mục 2 ở phần 18. +### 12. Trước khi xuất viện, làm Giấy chứng nhận sinh y tế, nghĩ trước tên con và đừng viết sai chữ -- Chi phí: Không tốn tiền. Làm luôn trong lúc làm thủ tục xuất viện -- Nói thẳng: Giấy này là điều kiện trước để đăng ký hộ khẩu, tham gia bảo hiểm và vào trường, và chỉ cơ sở đã đỡ đẻ mới có thể cấp lần đầu. Ra viện rồi mới quay lại bù, phải đi một quy trình khác và nộp thêm giấy tờ. Chữ trong tên, trước hết phải xác nhận hệ thống hộ khẩu hiện ra được. -- Lợi ích: Điều 23 của 《母婴保健法》 quy định: cơ sở y tế và người hành nghề đỡ đẻ tại nhà, theo quy định của cơ quan quản lý y tế của Hội đồng Nhà nước, phải cấp Giấy khai sinh do cơ quan thống nhất in và phát. Giấy này là điều kiện trước cho mọi việc phía sau: đăng ký hộ khẩu, tham gia bảo hiểm, làm thẻ bảo hiểm y tế, nhập trường mầm non và vào trường. Và lần cấp đầu tiên chỉ cơ sở đã đỡ đẻ mới làm được. Ra viện rồi mới quay lại bù, phải theo quy trình khác và cần thêm giấy tờ chứng minh +- Chi phí: Không tốn tiền. Làm luôn khi thủ tục xuất viện. +- Nói thẳng: Giấy này là điều kiện tiên quyết để đăng ký hộ khẩu, tham gia bảo hiểm, nhập học, và chỉ cơ sở đã đỡ thai mới được cấp lần đầu. Xuất viện rồi mới quay lại bù thì phải đi một quy trình khác và nộp thêm giấy tờ. Chữ trong tên nên xác nhận trước là hệ thống hộ khẩu gõ ra được. +- Lợi ích: Điều 23 của Bộ luật Chăm sóc Sức khoẻ Mẹ và Trẻ em: cơ sở y tế và người hành nghề đỡ thai tại nhà, theo quy định của cơ quan quản lý y tế của Hội đồng Nhà nước, phải cấp Giấy chứng nhận sinh y tế trẻ sơ sinh theo mẫu thống nhất. Giấy này là điều kiện tiên quyết cho mọi việc sau: đăng ký hộ khẩu, tham gia bảo hiểm, làm thẻ bảo hiểm y tế, nhập trường. Lần cấp đầu tiên chỉ cơ sở đã đỡ thai được làm. Xuất viện rồi quay lại bù thì phải đi quy trình khác, và cần thêm giấy tờ chứng minh. - Mức bằng chứng: A -- Nguồn: 全国人大常委会 (Ủy ban Thường vụ Quốc hội) (1994, sửa đổi 2017). Luật Chăm sóc bà mẹ và trẻ sơ sinh của nước Cộng hoà Nhân dân Trung Quốc, 中华人民共和国母婴保健法（第二十三条）. -- Ghi chú: Trước khi điền, đối chiếu từng chữ tên bố mẹ, số căn cước công dân và tên của trẻ. Chữ dùng trong tên phải là chữ mà hệ thống hộ khẩu hiện ra được, chữ hiếm sẽ khiến mọi việc sau này rắc rối hơn. Đổi tên hay đổi giấy đều phải qua bổ cấp hoặc cấp lại, phiền hơn nhiều so với kiểm tra tại chỗ. Cầm giấy xong thì cất riêng, giữ cẩn thận ngang với sổ hộ khẩu và căn cước công dân - -### 13. Đừng từ chối cả sàng lọc máu gót chân cho trẻ sơ sinh lẫn sàng lọc thính +- Nguồn: Ủy ban Thường vụ Quốc hội Nhà nước (全国人大常委会) (1994, sửa đổi 2017). Bộ luật Chăm sóc Sức khoẻ Mẹ và Trẻ em Trung Quốc, 中华人民共和国母婴保健法（第二十三条）. +- Ghi chú: Trước khi điền, hãy đối chiếu từng chữ tên cha mẹ, số căn cước và tên của con. Chữ dùng trong tên phải gõ ra được trong hệ thống hộ khẩu, chữ hiếm sẽ gây khó khăn khi làm việc sau này. Đổi tên và đổi giấy đều phải qua bổ cấp hoặc cấp lại, phiền hơn nhiều so với đối chiếu tại chỗ. Cầm được giấy thì cất riêng một chỗ, quan trọng ngang sổ hộ khẩu và căn cước. +### 13. Cả sàng lọc máu gót chân trẻ sơ sinh lẫn sàng lọc thính giác, đừng từ chối -- Chi phí: Ở phần lớn các nơi miễn phí, hoặc vài chục tệ. Có mấy mục mở rộng thì phải tự túi. Sau khi sinh, chọc gót chân trẻ một lần -- Nói thẳng: Một lần lấy máu gót chân và một lần kiểm tra thính giác có thể phát hiện suy giảm tuyến giáp, bệnh phenylketonuria và khiếm thính khi trẻ vẫn trông hoàn toàn bình thường. Đợi triệu chứng lộ ra thì tổn hại trí tuệ đã không bù lại được. -- Lợi ích: Điều 3 của 《新生儿疾病筛查管理办法》 quy định:「全国新生儿疾病筛查病种包括先天性甲状腺功能减低症、苯丙酮尿症等新生儿遗传代谢病和听力障碍。」 Hai nhóm bệnh này có một điểm chung: giai đoạn đầu hoàn toàn không thấy bất thường, đến khi triệu chứng rõ thì tổn hại trí tuệ đã không đảo ngược được. Mà phát hiện sớm và can thiệp sớm thì rất hiệu quả, cách làm là bổ sung hormone tuyến giáp, uống sữa công thức chuyên biệt, đeo máy trợ thính hoặc cấy ghép ốc tai nhân tạo. Điều 10 quy định, nếu kết quả sàng lọc dương tính thì cơ sở phải kịp báo người giám hộ đi xác nhận chẩn đoán +- Chi phí: Ở phần lớn các nơi miễn phí, hoặc vài chục tệ. Có mục mở rộng thì phải tự bỏ tiền ra. Trẻ sinh ra thì lấy máu gót chân một lần. +- Nói thẳng: Một lần lấy máu gót chân và một lần kiểm tra thính giác có thể phát hiện suy giáp, phenylketonuria và mất thính giác khi đứa trẻ trông hoàn toàn bình thường. Đợi đến khi triệu chứng lộ ra thì tổn hại về trí tuệ đã không còn cách nào bù lại. +- Lợi ích: Điều 3 của Quy chế quản lý sàng lọc bệnh ở trẻ sơ sinh ghi: 「全国新生儿疾病筛查病种包括先天性甲状腺功能减低症、苯丙酮尿症等新生儿遗传代谢病和听力障碍。」 Hai nhóm bệnh này có một điểm chung: giai đoạn đầu hoàn toàn không thấy bất thường, đến khi triệu chứng rõ thì tổn hại trí tuệ đã không đảo ngược được. Mà phát hiện sớm và can thiệp sớm thì hiệu quả rất tốt, cách làm là bổ sung thyroxine, cho trẻ uống công thức sữa đặc biệt, đeo máy trợ thính hoặc cấy ghép ốc tai nhân tạo. Điều 10 quy định khi kết quả sàng lọc dương tính thì cơ sở phải kịp thời báo người giám hộ đi xác nhận chẩn đoán. - Mức bằng chứng: A -- Nguồn: 卫生部 (Bộ Y tế) (2009). Biện pháp quản lý sàng lọc bệnh ở trẻ sơ sinh, 新生儿疾病筛查管理办法（卫生部令第 64 号，第三、十、十一条）. -- Ghi chú: Điều 11 của quy định này yêu cầu trước khi sàng lọc phải nói thật với người giám hộ về các mục, điều kiện, cách làm, độ nhạy, tức khả năng phát hiện được bao nhiêu, và chi phí, đồng thời lấy chữ ký đồng ý. Nên bạn sẽ nhận một phiếu đồng thuận sau khi được thông tin. Trước khi ký, đọc kỹ: nhóm bệnh nào do nhà nước quy định, nhóm nào là mục bổ sung tự trả tiền. Sàng lọc lần đầu dương tính không phải là chẩn đoán, đa số xét lại một lần là loại trừ được, đừng hoảng loạn ngay ở bước này. Sàng lọc thính lần đầu không đạt thì phải đi xét lại theo thông báo, đừng tự kết luận「trẻ nghe thấy là được」 - -### 14. Có giấy khai sinh rồi thì lập bảo hiểm y tế cho con, đừng đợi đăng ký xong hộ khẩu +- Nguồn: Bộ Y tế (卫生部) (2009). Quy chế quản lý sàng lọc bệnh ở trẻ sơ sinh, 新生儿疾病筛查管理办法（卫生部令第 64 号，第三、十、十一条）. +- Ghi chú: Điều 11 của quy chế yêu cầu trước khi sàng lọc phải nói thật với người giám hộ về các mục sàng lọc, điều kiện, cách làm, độ nhạy (phát hiện được bao nhiêu) và chi phí, rồi ký đồng ý. Vì vậy bạn sẽ nhận được một phiếu đồng ý bằng văn bản. Trước khi ký, đọc kỹ xem mục nào là bệnh do nhà nước quy định, mục nào là mục thêm tự trả tiền. Sàng lọc sơ bộ dương tính không phải là chẩn đoán, đa số làm lại một lần là loại trừ được, đừng hoảng loạn ở bước này. Sàng lọc sơ bộ thính giác không đạt thì phải đi làm lại theo thông báo, đừng tự kết luận 「thấy trẻ nghe được là xong」. +### 14. Có Giấy chứng nhận sinh y tế thì làm bảo hiểm y tế cho con ngay, đừng đợi đăng ký hộ khẩu xong -- Chi phí: Phí bảo hiểm y tế người dân cư trú đóng một lần mỗi năm, vài trăm tệ -- Nói thẳng: Có giấy khai sinh là lập bảo hiểm cho con trên mạng được, không cần đợi đăng ký hộ khẩu. Đã tham gia bảo hiểm thì phí nằm viện lúc xuất viện được hoàn ngay tại chỗ. Phí nằm viện của trẻ sơ sinh không nhỏ, chênh lệch giữa có bảo hiểm và không có rất lớn. -- Lợi ích: Cách nói năm 2024 của Cổng thông tin Chính phủ Trung Quốc:「新生儿凭出生医学证明就可以在线上参保」, không cần đăng ký hộ khẩu trước.「新生儿出院以后就能够直接报销医药费用」. Cách làm trước đây là phải đăng ký hộ khẩu rồi mới được tham gia bảo hiểm, phí nằm viện phải ứng đủ trước rồi mới hoàn sau. Thời gian xử lý bình quân việc đăng ký và đóng tiền cũng đã rút từ 28.7 ngày làm việc đầu năm xuống còn 6.4 ngày làm việc +- Chi phí: Một năm đóng phí bảo hiểm y tế cư dân một lần, vài trăm tệ. +- Nói thẳng: Có Giấy chứng nhận sinh y tế là đăng ký bảo hiểm cho con trên mạng được, không cần đợi đăng ký hộ khẩu. Đăng ký xong thì phí nằm viện lúc xuất viện hoàn được ngay. Phí nằm viện của trẻ sơ sinh không nhỏ, có bảo hiểm và không có bảo hiểm khác nhau rất nhiều. +- Lợi ích: Cách nói của Trang Chính phủ Trung Quốc năm 2024 là: 「新生儿凭出生医学证明就可以在线上参保」, không cần đăng ký hộ khẩu trước. 「新生儿出院以后就能够直接报销医药费用」. Cách làm trước đây là phải đăng ký hộ khẩu rồi mới được tham gia bảo hiểm, phí nằm viện phải ứng hết rồi mới quay lại hoàn. Thời gian xử lý trung bình cho việc đăng ký và đóng phí cũng bị rút từ 28.7 ngày làm việc vào đầu năm xuống còn 6.4 ngày làm việc. - Mức bằng chứng: A -- Nguồn: 中国政府网 (Cổng thông tin Chính phủ Trung Quốc) (2024). Họp báo định kỳ chính sách của Hội đồng Nhà nước về cơ chế dài hạn cho việc tham gia bảo hiểm y tế cơ sở, 健全基本医疗保险参保长效机制国务院政策例行吹风会. -- Ghi chú: Đẻ non, vàng da phải chiếu đèn xanh, viêm phổi trẻ sơ sinh là những lý do thường gặp khiến trẻ sơ sinh phải nằm viện, chi phí dễ động tới mấy vạn tệ, chênh lệch giữa có bảo hiểm và không có rất lớn. Quy định ở các nơi về việc phải tham gia bảo hiểm trong bao nhiêu ngày sau khi sinh thì mới được hưởng từ ngày sinh không giống nhau. Cách thực hiện cụ thể thì theo đúng lời cơ quan bảo hiểm y tế địa phương. Gọi 12393 hỏi một câu trước khi xuất viện là tiện nhất - -### 15. Trong vòng một tháng sau khi trẻ sinh, ra công an phường xã khai đăng ký sinh +- Nguồn: Trang Chính phủ Trung Quốc (中国政府网) (2024). Buổi nói chuyện định kỳ giới thiệu chính sách của Hội đồng Nhà nước về việc hoàn thiện cơ chế tham gia bảo hiểm y tế cơ bản. +- Ghi chú: Sinh non, vàng da phải chiếu ánh xanh, viêm phổi trẻ sơ sinh là những nguyên nhân thường gặp khi trẻ sơ sinh nằm viện, mỗi đợt có thể lên tới mấy vạn tệ, có bảo hiểm và không có bảo hiểm khác nhau rất nhiều. Quy định của mỗi nơi về 「trong vòng bao nhiêu ngày sau khi sinh mà đăng ký thì được hưởng chế độ tính từ ngày sinh」 không hoàn toàn giống nhau. Cụ thể thực hiện thế nào thì theo lời cơ quan bảo hiểm y tế địa phương. Gọi 12393 trước khi xuất viện hỏi một câu là tiện nhất. +### 15. Trong vòng một tháng sau khi con sinh, ra công an đường phố khai báo đăng ký sinh -- Chi phí: Không tốn tiền. Chạy một lần tới quầy hộ khẩu ở công an phường xã -- Nói thẳng: Luật cho một tháng, không phải「khi nào rảnh thì nói」. Đăng ký hộ khẩu chặn các việc phía sau: tiêm phòng, nhập trường mầm non, vào trường và làm căn cước công dân. Càng kéo dài, càng phải bù thêm nhiều giấy tờ. -- Lợi ích: Điều 7 của 《户口登记条例》 quy định:「婴儿出生后一个月以内，由户主、亲属、抚养人或者邻居向婴儿常住地户口登记机关申报出生登记。」 Đó là thời hạn do luật đặt ra, không phải lời khuyên. Đăng ký hộ khẩu là điều kiện trước cho các việc phía sau: tiêm phòng, nhập trường mầm non, vào trường, làm căn cước công dân. Càng kéo dài, càng phải bù thêm nhiều giấy tờ +- Chi phí: Không tốn tiền. Chạy một lần tới cửa sổ hộ khẩu ở công an đường phố. +- Nói thẳng: Luật cho một tháng, không phải 「rảnh thì tính」. Đăng ký hộ khẩu là điều kiện tiên quyết cho tiêm vắc-xin, nhập trường, vào lớp và làm căn cước. Càng kéo dài thì giấy tờ phải bù càng nhiều. +- Lợi ích: Điều 7 của Quy chế đăng ký hộ khẩu Trung Quốc ghi: 「婴儿出生后一个月以内，由户主、亲属、抚养人或者邻居向婴儿常住地户口登记机关申报出生登记。」 Đây là thời hạn do luật định, không phải lời khuyên. Đăng ký hộ khẩu là điều kiện tiên quyết cho các việc sau: tiêm vắc-xin, nhập trường, vào lớp, làm căn cước. Càng kéo dài thì giấy tờ phải bù càng nhiều. - Mức bằng chứng: A -- Nguồn: 国务院 (Hội đồng Nhà nước) (1958). Quy định về đăng ký hộ khẩu của nước Cộng hoà Nhân dân Trung Quốc, 中华人民共和国户口登记条例（第七条）. （Chính quyền nhân dân tỉnh Quảng Đông đăng lại） -- Ghi chú: Mang theo đủ những giấy này: giấy khai sinh, căn cước công dân và sổ hộ khẩu của cả hai bên, giấy đăng ký kết hôn. Đăng ký theo bố hay theo mẹ thì bạn tự chọn, nhưng chính sách hai nơi khác nhau, cả vùng trường và nơi tham gia bảo hiểm đều bị ảnh hưởng. Suy nghĩ cho xong trước khi đi, đổi lại rất phiền. Trẻ sinh ngoài hôn nhân vẫn được khai đăng ký sinh như thường - -### 16. Đừng bỏ lần tái khám 42 ngày sau sinh, đó cũng là lần sàng lọc trầm cảm sau sinh +- Nguồn: Hội đồng Nhà nước (国务院) (1958). Quy chế đăng ký hộ khẩu Trung Quốc, 中华人民共和国户口登记条例（第七条）. （Chính phủ tỉnh Quảng Đông, 广东省人民政府, đăng lại） +- Ghi chú: Mang đủ những giấy này: Giấy chứng nhận sinh y tế, căn cước và sổ hộ khẩu của cả hai bố mẹ, giấy đăng ký kết hôn. Đăng ký theo bố hay theo mẹ thì tự chọn được, nhưng chính sách hai nơi khác nhau, cả tuyến trường lẫn nơi tham gia bảo hiểm đều bị ảnh hưởng. Suy nghĩ cho xong trước khi đi, sửa lại rất phiền. Trẻ sinh ngoài hôn nhân thì cũng có thể khai báo đăng ký sinh. +### 16. Đừng bỏ lần tái khám 42 ngày sau khi sinh, lần đó đồng thời là sàng lọc trầm cảm sau sinh -- Chi phí: Không tốn tiền, thuộc dịch vụ công cộng cơ sở về y tế. Mất nửa ngày. Chỗ khó là vừa hết tháng nằm hồi thì lười nhúc nhích -- Nói thẳng: Lần tái khám này không tốn tiền, gồm công thức máu, xét nghiệm dịch âm đạo thường quy, và cả một lần sàng lọc trầm cảm sau sinh. Trong vòng một năm sau sinh vẫn có thể xảy ra chuyện nghiêm trọng, đừng nghĩ hết tháng nằm hồi là đã an toàn. -- Lợi ích: Phần quản lý sức khoẻ sản phụ trong 《国家基本公共卫生服务规范（第三版）》 có sau sinh là 1 lần thăm khám sau sinh và 1 lần khám sức khoẻ 42 ngày sau sinh. Danh mục dịch vụ công khai của địa phương cho thấy lần này gồm xét nghiệm máu, xét nghiệm dịch âm đạo thường quy, và 1 lần sàng lọc trầm cảm sau sinh. Biến chứng nặng có thể xảy ra bất cứ lúc nào trong vòng một năm sau sinh. Trong danh sách「cần vào viện ngay」 ở mục 7 của phần này,「有伤害自己或伤害孩子的念头」「产后阴道出血或异常分泌物」「一侧肢体严重肿痛」 đều thuộc giai đoạn sau sinh +- Chi phí: Không tốn tiền, thuộc dịch vụ y tế công cộng cơ bản. Mất nửa ngày. Chỗ khó là vừa hết hậu sản thì lười nhúc nhích. +- Nói thẳng: Lần tái khám này không tốn tiền, gồm công thức máu, soi dịch âm đạo, và cả một lần sàng lọc trầm cảm sau sinh. Trong một năm sau khi sinh đều có thể xảy ra chuyện nghiêm trọng, đừng nghĩ ra khỏi hậu sản là đã an toàn. +- Lợi ích: Trong phần quản lý sức khoẻ sản phụ của 《国家基本公共卫生服务规范（第三版）》 có, sau khi sinh, 1 lần thăm khám sau sinh và 1 lần khám sức khoẻ 42 ngày sau sinh. Danh mục dịch vụ công khai của địa phương quy định lần này gồm xét nghiệm (công thức máu, soi dịch âm đạo) và 1 lần sàng lọc trầm cảm sau sinh. Trong một năm sau khi sinh đều có thể xảy ra biến chứng nghiêm trọng. Trong bảng 「phải vào viện ngay」 ở mục 7 ở phần này, 「có ý nghĩ làm hại mình hoặc con」「chảy máu âm đạo hoặc dịch tiết bất thường sau khi sinh」「một bên chi thể sưng và đau nặng」 đều thuộc về giai đoạn sau sinh. - Mức bằng chứng: A -- Nguồn: 国家卫生计生委 (Uy ban Sức khoẻ và Kế hoạch Gia đình quốc gia) (2017). Quy định về dịch vụ công cộng cơ sở về y tế quốc gia, 国家基本公共卫生服务规范（第三版）. ；靖江市人民政府 (Chính quyền nhân dân thành phố Tĩnh Giang). Chương trình dịch vụ công cộng cơ sở về y tế quốc gia — dịch vụ quản lý sức khoẻ sản phụ (bảng dịch vụ công khai của địa phương), 国家基本公共卫生服务项目——孕产妇健康管理服务（地方服务清单公开）. -- Ghi chú: Trầm cảm sau sinh không phải là nũng niu, người nhà khuyên một câu「cứ nghĩ thoáng đi」 thì không giải quyết được. Nó có bảng hỏi sàng lọc riêng, có cách điều trị, và có thể hồi phục. Khi có ý nghĩ tự sát thì xử lý thế nào, xem phần 1. Khi đã sinh con mà thật sự không nuôi nổi thì lối thoát hợp pháp xem mục 20 ở phần 9. Lần tái khám này còn xem vết thương đã lành thế nào, tử cung hồi phục ra sao, chức năng sàn vùng chậu có tốt không. Kế hoạch tránh thai cũng được bàn lúc này: sau sinh không có kinh nguyệt không có nghĩa là không thể có thai +- Nguồn: Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委) (2017). Quy chuẩn dịch vụ y tế công cộng cơ bản quốc gia, bản thứ ba. ；Chính quyền thành phố Tĩnh Giang (靖江市人民政府). Dự án dịch vụ y tế công cộng cơ bản quốc gia — quản lý sức khoẻ sản phụ (bảng danh mục dịch vụ công khai của địa phương). +- Ghi chú: Trầm cảm sau sinh không phải là ủng ta, một lời khuyên 「cứ thấy nhẹ tâm thôi」 của người nhà cũng không giải quyết được. Nó có bảng hỏi sàng lọc riêng, có cách điều trị, và có thể hồi phục. Khi có ý nghĩ tự sát thì xử lý ra sao xem ở phần 1. Khi đứa trẻ sinh ra mà thật sự không nuôi nổi thì lối thoát hợp pháp xem ở mục 20 ở phần 9. Lần tái khám này còn xem vết thương hồi phục thế nào, tử cung hồi phục thế nào, chức năng đáy chậu có tốt không. Phương pháp tránh thai cũng được bàn ở lần này: sau khi sinh không có kinh nguyệt không có nghĩa là không thụ thai. \ No newline at end of file diff --git "a/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" "b/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" index 769dcaf8..7706ad08 100644 --- "a/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" +++ "b/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" @@ -39,7 +39,7 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Nói thẳng: Người mất bạn đời vì tự tử, trong năm năm sau có xác suất được chẩn đoán rối loạn tâm thần cao hơn người bình thường khoảng tám phần mười (nam), bảy phần mười (nữ). So với người mất bạn đời vì nguyên nhân khác thì cũng cao hơn từ bảy phần mười đến gấp đôi. Người mất người thân kiểu này dễ cảm thấy bị người ta lạnh nhạt, dễ cảm thấy xấu hổ. Đó là lý do phải chủ động đi tìm người, đừng đợi mình ổn rồi hãy đi - Lợi ích: Hồ sơ toàn quốc Đan Mạch theo dõi 6.700.000 người trưởng thành, giai đoạn 1980 đến 2014, tổng cộng hơn 136.000.000 người-năm (cộng dồn số năm mỗi người được theo dõi). Ở người mất bạn đời vì bạn đời tự tử, nguy cơ được chẩn đoán mới rối loạn tâm thần trong 5 năm sau so với người bình thường là IRR 1.8 ở nam (95% CI 1.6–2.0), tức cao hơn khoảng tám phần mười. Ở nữ là 1.7 (1.6–1.8), tức cao hơn khoảng bảy phần mười. So với người mất bạn đời vì nguyên nhân tử vong khác, IRR 1.7 ở nam (1.5–1.9) và 2.0 ở nữ (1.9–2.2), tức cao hơn từ bảy phần mười đến gấp đôi. Trong nhóm này, tỷ lệ có hành vi tự tử, tỷ lệ tử vong và tỷ lệ phải nhận trợ cấp của chính quyền địa phương cũng cao hơn. Các bệnh thể chất như xơ gan, rối loạn giấc ngủ cũng tăng nguy cơ. Ngoài ra có 57 nghiên cứu rà từng yếu tố, không gộp lại để tính. Mất người thân vì tự tử có liên quan tới bốn việc sau. Một là nguy cơ bạn đời tự tử tăng. Hai là nguy cơ cha mẹ phải vào viện tâm thần sau khi con tự tử tăng. Ba là nguy cơ người mẹ tự tử sau khi người con đã trưởng thành tự tử tăng. Bốn là nguy cơ trẻ bị trầm cảm sau khi cha mẹ tự tử tăng. So với người mất người thân vì bạo lực chết vì nguyên nhân khác, người mất vì tự tử thường nói nhiều hơn rằng mình bị người ta lạnh nhạt, thấy xấu hổ - Mức bằng chứng: A -- Nguồn: Erlangsen A, et al. (2017). Association Between Spousal Suicide and Mental, Physical, and Social Health Outcomes: A Longitudinal and Nationwide Register-Based Study. JAMA Psychiatry. ；Pitman A, et al. (2014). Effects of suicide bereavement on mental health and suicide risk. The Lancet Psychiatry. ；Kristensen P 等 (2012). Bereavement and mental health after sudden and violent losses: a review. Psychiatry. （bản tổng quan trong phần Ghi chú） +- Nguồn: Erlangsen A, et al. (2017). Association Between Spousal Suicide and Mental, Physical, and Social Health Outcomes: A Longitudinal and Nationwide Register-Based Study. JAMA Psychiatry. ；Pitman A, et al. (2014). Effects of suicide bereavement on mental health and suicide risk. The Lancet Psychiatry. ；Kristensen P và cộng sự (2012). Bereavement and mental health after sudden and violent losses: a review. Psychiatry. （bản tổng quan trong phần Ghi chú） - Ghi chú: Cái chết đột ngột, bạo lực là tai nạn, bị giết, thảm hoạ. Sau loại chết này, tỷ lệ rối loạn sau chấn thương tâm lý, trầm cảm và nỗi buồn bị kẹt đều cao hơn sau khi chết bệnh. Hồi phục cũng chậm hơn. Khi bản thân có ý nghĩ tự tử thì xem mục 25 ở phần 1 ### 5. Trong nửa năm đầu sau khi mất bạn đời, giao việc canh bữa cơm, việc ngủ, thuốc bệnh mạn tính và lịch tái khám cho một người cố định @@ -50,6 +50,7 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: A - Nguồn: Moon JR, et al. (2011). Widowhood and mortality: a meta-analysis. PLoS ONE. ；Shor E, et al. (2012). Widowhood and mortality: a meta-analysis and meta-regression. Demography. - Ghi chú: Đây là nghiên cứu nhìn lại dữ liệu nhóm người, không gỡ được yếu tố khác lẫn vào. Vợ chồng vốn đã giống nhau về lối sống và bệnh tật. Nhưng quy luật «nửa năm đầu cao hơn nhiều so với sau đó» giống nhau trong cả hai phân tích tổng hợp độc lập, đáng coi là một giai đoạn cần có người trông. Nam có nguy cơ cao hơn và ít kêu cầu giúp đỡ hơn. Nhà có người lớn tuổi vừa mất bạn đời thì theo phần 17, làm cho rõ tài khoản, giám hộ và di chúc. Không tìm được người như vậy (sống một mình, không còn con, con cái cũng không còn) thì giao việc này cho cộng đồng và điện thoại, xem mục 6 của phần này + ### 6. Không có thân nhân cũng không có bạn, hãy thay «người trông hộ bạn» bằng ba thứ: hàng xóm có vào được nhà bạn, danh sách thăm hỏi của cộng đồng, và người liên hệ khẩn cấp trong điện thoại - Chi phí: Không tốn tiền. Khó ở chỗ phải tự mở lời, nói với hàng xóm và ủy ban nhân dân một câu «tôi sống một mình, trong nhà vừa xảy ra chuyện» @@ -83,8 +84,9 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Nói thẳng: Sắp xếp tâm lý trị liệu cho tất cả người mất người thân thì lợi ích rất nhỏ, đợi một thời gian nữa không còn thấy nữa, vì phần lớn tự khỏi. Nhưng với người đã rõ ràng không hợp thích nổi thì điều trị hiệu quả ngang với điều trị các vấn đề tâm lý khác. Vậy nên đối chiếu với mục 8 (đau buồn bị kẹt thì đi khoa tâm thần), khớp rồi hãy bỏ khoản tiền này - Lợi ích: 61 nghiên cứu đối chứng gộp lại. Can thiệp tâm lý dành cho mọi người mất người thân thì lợi ích rất nhỏ ngay khi kết thúc điều trị, đợi thêm một thời gian thì không còn thấy lợi ích. Chỉ nhận người mất người thân «đã rõ ràng không hợp thích nổi» thì can thiệp có hiệu quả tốt ngang tâm lý trị liệu cho các vấn đề tâm lý khác. Nghiên cứu không sàng lọc trước cho kết quả kém, vì nhóm không được can thiệp vốn cũng dần tự khỏi theo thời gian. Ngoài ra có 14 thử nghiệm chia nhóm ngẫu nhiên gộp lại: điều trị cho người đã có vấn đề thì rõ ràng tốt hơn nhóm đối chứng ngay sau điều trị và cả ở lần theo dõi sau đó, để thêm thời gian thì hiệu quả còn tăng. Can thiệp phòng ngừa làm đại trà cho người chưa có vấn đề thì không có tác dụng. Một nghiên cứu khác theo dõi 205 người từ trước khi mất bạn đời cho tới 6 và 18 tháng sau. Nghiên cứu này chia ra năm đường đi. Lần lượt là đau buồn thường, đau buồn kéo dài, trầm cảm kéo dài, ngược lại là khỏe lên trong thời gian mất người thân, và sức bền (chịu được). Trong đó nhóm sức bền phổ biến nhất, nhóm đau buồn thường tương đối ít - Mức bằng chứng: A -- Nguồn: Currier JM, et al. (2008). The effectiveness of psychotherapeutic interventions for bereaved persons: a comprehensive quantitative review. Psychological Bulletin. ；Wittouck C, et al. (2011). The prevention and treatment of complicated grief: a meta-analysis. Clinical Psychology Review. ；Bonanno GA, et al. (2002). Resilience to loss and chronic grief: a prospective study from preloss to 18-months postloss. Journal of Personality and Social Psychology. ；Stroebe M 等 (2007). Health outcomes of bereavement. Lancet. （bản tổng quan trong phần Ghi chú） +- Nguồn: Currier JM, et al. (2008). The effectiveness of psychotherapeutic interventions for bereaved persons: a comprehensive quantitative review. Psychological Bulletin. ；Wittouck C, et al. (2011). The prevention and treatment of complicated grief: a meta-analysis. Clinical Psychology Review. ；Bonanno GA, et al. (2002). Resilience to loss and chronic grief: a prospective study from preloss to 18-months postloss. Journal of Personality and Social Psychology. ；Stroebe M và cộng sự (2007). Health outcomes of bereavement. Lancet. （bản tổng quan trong phần Ghi chú） - Ghi chú: Ý là «không phải ai cũng cần», không phải «ai cũng đừng làm». Nhóm mất người thân nguy cơ cao như mục 4 (người thân chết vì tự tử, tai nạn hoặc bị giết) và nhóm đã bị kẹt như mục 8 (đau buồn bị kẹt thì đi khoa tâm thần), chính là nhóm có bằng chứng ủng hộ việc nên làm. Kết luận của một bài tổng quan trên Lancet cũng vậy: đau buồn không phải bệnh, phần lớn không cần can thiệp chuyên môn. Sức lực nên dồn vào nhóm nguy cơ cao. Cũng nên dồn vào người đã có đau buồn phức tạp, trầm cảm và rối loạn sau chấn thương tâm lý sau khi mất người thân + ### 10. Sau khi sống xa nhau hoặc ly hôn, đừng để bảo hiểm y tế bị gián đoạn, đừng dựa vào rượu để chịu đựng - Chi phí: Không tốn tiền. Gia hạn bảo hiểm y tế, bỏ rượu. Khó ở chỗ khoảng thời gian này dễ dựa vào rượu để chịu đựng @@ -115,8 +117,8 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải ### 13. Đừng dùng cái chết để trả nợ: bảo hiểm nhân thọ trong hai năm không bồi thường, tai nạn lao động không được công nhận, nợ vẫn bị trừ trước tiên từ di sản - Chi phí: Không tốn tiền, chỉ là làm cho số sách này rõ ràng -- Nói thẳng: Bảo hiểm nhân thọ không bồi thường khi tự tử trong hai năm đầu, chỉ hoàn lại số tiền trong hợp đồng mà được hoàn. Tai nạn lao động cũng không được công nhận, ba khoản tiền khi chết vì tai nạn lao động không có đồng nào. Nợ nợ bị trừ trước tiên từ di sản bạn để lại, người nhà chỉ nhận phần còn lại. Họ còn phải gánh sức khoẻ mà mục 4 của phần này nói (mất người thân vì tự tử hoặc bạo lực) -- Lợi ích: Điều 44 của Bộ luật Bảo hiểm quy định về hợp đồng có điều kiện trả tiền bảo hiểm khi người được bảo hiểm chết. Tính từ ngày hợp đồng thành lập hoặc ngày hiệu lực được khôi phục, nếu người được bảo hiểm tự tử trong hai năm thì công ty bảo hiểm không phải nghĩa vụ trả tiền bảo hiểm. Không tính trường hợp người được bảo hiểm tự tử khi chưa có năng lực hành vi dân sự, tức người mà pháp luật không cho chịu trách nhiệm về hành vi của chính mình. Trong trường hợp đó công ty bảo hiểm phải hoàn trả giá trị tiền mặt của hợp đồng theo thỏa thuận, tức chỉ hoàn lại số tiền trong hợp đồng được hoàn, không trả số tiền bảo hiểm. Mức bảo hiểm nhân thọ thường ở cỡ vài chục nghìn tệ, vì vậy cỡ lợi ích đặt là lớn. Điều 16, khoản ba của Quy chế bảo hiểm tai nạn lao động quy định: tự tàn tật hoặc tự tử thì không được công nhận là tai nạn lao động, cũng không được coi là tai nạn lao động. Ba khoản quyền lợi khi chết vì tai nạn lao động ở Điều 39 vì thế mất trắng cả ba. Một là trợ cấp mai táng, trả bằng 6 tháng mức lương trung bình của người lao động năm trước tại địa phương. Hai là tiền tùy táng người nuôi dưỡng. Ba là trợ cấp một lần khi chết vì tai nạn lao động, bằng 20 lần thu nhập chi tiêu bình quân đầu người dân đô thị toàn quốc năm trước. Điều 1159 của Bộ luật Dân sự quy định: khi chia di sản thì phải trừ các khoản thuế và nợ mà người để lại có nghĩa vụ theo pháp luật, tức lấy di sản trả thuế và nợ trước, phần còn lại mới chia. Điều 1161 quy định: người thừa kế trả nợ trong giới hạn giá trị thực tế của di sản được nhận, tức chỉ dùng phần được thừa kế để trả, không phải bỏ thêm ra túi mình; người từ bỏ quyền thừa kế thì không phải chịu trách nhiệm trả nợ +- Nói thẳng: Bảo hiểm nhân thọ không bồi thường khi tự tử trong hai năm đầu, chỉ hoàn lại số tiền trong hợp đồng mà được hoàn. Tai nạn lao động cũng không được công nhận, ba khoản tiền khi chết vì tai nạn lao động không có một xu nào. Nợ phải trả trước ra từ di sản bạn để lại, người nhà chỉ nhận phần còn lại. Họ còn phải gánh sức khoẻ mà mục 4 của phần này nói (mất người thân vì tự tử hoặc bạo lực) +- Lợi ích: Điều 44 của Bộ luật Bảo hiểm quy định về hợp đồng có điều kiện trả tiền bảo hiểm khi người được bảo hiểm chết. Tính từ ngày hợp đồng thành lập hoặc ngày hiệu lực được khôi phục, nếu người được bảo hiểm tự tử trong hai năm thì công ty bảo hiểm không phải nghĩa vụ trả tiền bảo hiểm. Không tính trường hợp người được bảo hiểm tự tử khi chưa có năng lực hành vi dân sự, tức người mà pháp luật không cho chịu trách nhiệm về hành vi của chính mình. Trong trường hợp đó công ty bảo hiểm phải hoàn trả giá trị tiền mặt của hợp đồng theo thỏa thuận, tức chỉ hoàn lại số tiền trong hợp đồng được hoàn, không trả số tiền bảo hiểm. Mức bảo hiểm nhân thọ thường ở cỡ vài trăm nghìn tệ, vì vậy cỡ lợi ích đặt là lớn. Điều 16, khoản thứ ba của Quy chế bảo hiểm tai nạn lao động quy định: tự tàn tật hoặc tự tử thì không được công nhận là tai nạn lao động, cũng không được coi là tai nạn lao động. Ba khoản quyền lợi khi chết vì tai nạn lao động ở Điều 39 vì thế mất trắng cả ba. Một là trợ cấp mai táng, trả bằng 6 tháng mức lương trung bình của người lao động năm trước tại địa phương. Hai là tiền tùy táng người nuôi dưỡng. Ba là trợ cấp một lần khi chết vì tai nạn lao động, bằng 20 lần thu nhập chi tiêu bình quân đầu người dân đô thị toàn quốc năm trước. Điều 1159 của Bộ luật Dân sự quy định: khi chia di sản thì phải trừ các khoản thuế và nợ mà người để lại có nghĩa vụ theo pháp luật, tức lấy di sản trả thuế và nợ trước, phần còn lại mới chia. Điều 1161 quy định: người thừa kế trả nợ trong giới hạn giá trị thực tế của di sản được nhận, tức chỉ dùng phần được thừa kế để trả, không phải bỏ thêm ra túi mình; người từ bỏ quyền thừa kế thì không phải chịu trách nhiệm trả nợ - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Quốc hội (2015 修正). Bộ luật Bảo hiểm nước Cộng hoà nhân dân Trung Quốc（中华人民共和国保险法第四十四条）. ；Hội đồng Nhà nước (2010 修订). Quy chế bảo hiểm tai nạn lao động（工伤保险条例，国务院令第 586 号，第十六条、第三十九条）. ；Quốc hội (2020). Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（中华人民共和国民法典第一千一百五十九条、第一千一百六十一条）. -- Ghi chú: Ở đây chỉ tính sổ sức khoẻ về tiền. Các mục khác của phần này và của phần 1 nói là đừng đi đến bước này. Điều 44 của Bộ luật Bảo hiểm chỉ nói về hai năm đầu, viết ra là để nêu đủ điều khoản, không phải để đặt ra một mốc thời gian. Sức khoẻ của người nhà xem mục 4 của phần này (vợ chồng có người chết vì tự tử thì trong 5 năm đầu nguy cơ mắc rối loạn tâm thần mới cao hơn bảy đến tám phần mười). Khi có ý nghĩ tự tử thì làm gì xem mục 25 và 32 ở phần 1, di chứng còn lại sau khi được cứu xem mục 33 ở phần 1. Mức ba khoản tiền khi chết vì tai nạn lao động xem mục 16 ở phần 19. Thủ tục cụ thể về di sản và nợ xem mục 9 ở phần 25 (lấy từng khoản quỹ tích góp nhà ở và quyền lợi bảo hiểm xã hội) \ No newline at end of file +- Nguồn: Ủy ban Thường vụ Quốc hội (2015 sửa đổi). Bộ luật Bảo hiểm nước Cộng hoà nhân dân Trung Quốc（中华人民共和国保险法第四十四条）. ；Hội đồng Nhà nước (2010 sửa đổi). Quy chế bảo hiểm tai nạn lao động（工伤保险条例，国务院令第 586 号，第十六条、第三十九条）. ；Quốc hội (2020). Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（中华人民共和国民法典第一千一百五十九条、第一千一百六十一条）. +- Ghi chú: Ở đây chỉ tính sổ sức khoẻ về tiền. Các mục khác của phần này và của phần 1 nói là đừng đi đến bước này. Điều 44 của Bộ luật Bảo hiểm chỉ nói về hai năm đầu, viết ra là để nêu đủ điều khoản, không phải để đặt ra một mốc thời gian. Sức khoẻ của người nhà xem mục 4 của phần này (vợ chồng có người chết vì tự tử thì trong 5 năm đầu nguy cơ mắc rối loạn tâm thần mới cao hơn bảy đến tám phần mười). Khi có ý nghĩ tự tử thì làm gì xem mục 25 và 32 ở phần 1, di chứng còn lại sau khi được cứu xem mục 33 ở phần 1. Mức ba khoản tiền khi chết vì tai nạn lao động xem mục 16 ở phần 19. Thủ tục cụ thể về di sản và nợ xem mục 9 ở phần 25 (lấy từng khoản quỹ tích góp nhà ở và quyền lợi bảo hiểm xã hội) From a3eb55018a3534ede5dd7137e2de20bd20250f51 Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 13:04:07 +0700 Subject: [PATCH 20/25] =?UTF-8?q?BRIEF=20=E6=8A=8A=E5=88=9A=E8=A1=A5?= =?UTF-8?q?=E4=B8=8A=E7=9A=84=E4=B8=89=E6=9D=A1=E5=88=A4=E6=8D=AE=E5=86=99?= =?UTF-8?q?=E8=BF=9B=E5=8E=BB=EF=BC=8C=E6=A0=87=E7=82=B9=E5=92=8C=E9=87=8F?= =?UTF-8?q?=E7=BA=A7=E6=8D=A2=E7=AE=97=E5=90=84=E7=BB=99=E4=B8=80=E4=B8=AA?= =?UTF-8?q?=E6=AD=A3=E5=8F=8D=E4=BE=8B?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 翻译任务读的是 BRIEF，不是代码。这轮机器查上了三样（节标题、节导读、条目标题）， BRIEF 原来一条都没写，等于让翻的人靠撞。补上之后逐条对了：BRIEF 里写的十项要求， 代码里都有对应的判据，没有「文档写了但代码没查」。 量级换算那条原来只给了正例（167 万 → 1,670,000），没给反例，所以译文里 「10 万元 → 100.000 tệ」这种把数字改掉的做法看起来像合规。补上：原文没有量级词时 那三个数必须原样留着，中文数字法条编号才转成阿拉伯数字。争议标记也说明 越南文冒号后有空格没关系（判据加了 trimStart）。 --- tools/i18n/BRIEF.md | 58 +++++++++++++++++-- ...31\346\210\226\345\271\263\345\217\260.md" | 20 +++---- 2 files changed, 62 insertions(+), 16 deletions(-) diff --git a/tools/i18n/BRIEF.md b/tools/i18n/BRIEF.md index f32f1efe..bc1f98b3 100644 --- a/tools/i18n/BRIEF.md +++ b/tools/i18n/BRIEF.md @@ -22,6 +22,18 @@ These are machine-checked by `tools/i18n/check.mjs`. Any violation fails the bui The same number of entries, in the same order. Section headings and intro paragraphs are translated, not dropped. If a block has no `# ` section heading, do not invent one. + Everything in headings and intros is translated, including the **entry heading text** + (`### 21. 每天都喝酒、一停就手抖心慌的人，别自己硬戒` → `### 21. Người uống rượu mỗi + ngày, …`) and the **section heading** (`# 6. 反面清单` → `# 6. Danh sách phản tác dụng`). + A heading left in Chinese fails the check — it is what a reader sees first in search + results and in the sidebar. The one exception: statute and book names quoted inside + 「」 or 《》 in a heading stay as they are (rule 6). + + A section's `# N. …` line must match the title in that language's README table of + contents word for word — the page sidebar is built from the README, so the two places + have to say the same thing. Change both the README line and the block's line in + `.i18n//
    /part00.md`; the `book/` file is regenerated from the block. + 2. **Field labels** (six per entry, in this order): | Chinese | English | Vietnamese | @@ -56,6 +68,19 @@ These are machine-checked by `tools/i18n/check.mjs`. Any violation fails the bui more digits keep the shape 「约 X 万元（精确值 元）」 → "about 120,000 yuan (exact figure 1130040 yuan)" / "khoảng 120.000 tệ (con số chính xác 1130040 tệ)". + The conversion only works when the scale word is there. 「200 余万元」「10 万元」 + → "hơn 2.000.000 tệ"? No — the checker counts digit sequences, so the source's 200 + and 10 must survive as 200 and 10: write 「200 余万元」 as "200 vạn tệ trở lên" and + 「10 万元」 as "10 vạn tệ". Converting "10 万元" to "100.000" drops a digit the source + has and adds one it does not, and the Benefits column is where the book's whole + checkability rests. + + **Vietnamese only:** statute article numbers written in Chinese numerals in the source + 「第一千零四十五条」 become "Điều 1045" — the checker knows the number was in the source. + Amounts are the opposite case: 「3,018 人」 becomes "3.018 người" (thousands separator), + but 「3,177」 stays "3.177" only because that group has three digits; 「1130040」 stays + as written. Never re-group digits the source did not group. + The Notes field is prose, so "40 例" may become "several dozen". The three fields above may not. @@ -91,7 +116,20 @@ These are machine-checked by `tools/i18n/check.mjs`. Any violation fails the bui "Disputed" / "Tranh cãi". Keep the literal word "TODO" wherever it appears. The checker counts disputed entries and unverified items; the counts have to match. -8. **Register.** Short declarative sentences, one idea each, roughly 15–25 words, never + **Vietnamese only:** a space after the colon is fine — the checker skips leading + whitespace. What it does need is the marker word itself at the start: `- Ghi chú: + Tranh cãi.` reads the same as `- Ghi chú:Tranh cãi.` and both pass. What fails is a + marker in the middle of the note, or an entry the source did not dispute getting one. + +8. **No Chinese left in running prose.** Apart from the quotations rules 5 and 6 allow + (statute text, statute and book names, bibliographic identifiers like 国务院令第 768 号, + organisation names in parentheses), a Chinese word sitting inside an English or + Vietnamese sentence is a missed translation — the reader sees it. Common in practice: + a connective that came through as 本身 / 普遍 / 恰 / 口径, or a heading left untranslated. + Keep the whole phrase translated rather than splicing a Chinese word into the middle + of a Vietnamese one ("cũng普遍 đánh giá" → "cũng đánh giá"). + +9. **Register.** Short declarative sentences, one idea each, roughly 15–25 words, never over 40. Explicit subjects: you, the doctor, the court, the company. Calm. No exclamation marks. No lecturing. No AI-sounding filler: no "It is worth noting that", no "essentially", no repeated "in other words", no uplifting summary sentence at the end @@ -99,14 +137,22 @@ These are machine-checked by `tools/i18n/check.mjs`. Any violation fails the bui "output", "a shield"). Negation, scope, conditions and judgements are content — do not cut them as filler. -9. **Glossary.** Use the exact terms in `tools/i18n/glossary.en.mjs` (English) or +10. **Glossary.** Use the exact terms in `tools/i18n/glossary.en.mjs` (English) or `tools/i18n/glossary.vi.mjs` (Vietnamese). Recurring terms must be worded the same way everywhere, or the book contradicts itself. ## Checking your work After writing, the file is compared against the Chinese source: entry count, entry -numbers, field presence and order, the cost-tag comment, evidence grades, every number in -the three checked fields, cross-reference numbers, and link counts. Fix what it reports -before finishing. `node tools/i18n/join.mjs book/ --locale ` runs that -comparison for one section if you want to see it directly. +numbers, field presence and order, the cost-tag comment, evidence grades, disputed and +unverified markers, every number in the three checked fields, cross-reference numbers, +link counts — plus the three lines that are easy to leave behind: the section heading, the +section intro, and each entry's own heading. Fix everything it reports before finishing. + +`node tools/i18n/join.mjs book/ --locale ` runs that comparison for one +section. `bash tools/i18n/join-ready.sh ` joins every section you have finished and +prints the failures for the ones you have not. + +A section that reports 「切了 N 块，译文只有 M 块」 has a block file in the wrong place or +a stray extra one — list the folder and compare the entry numbers each block starts with +against the source. diff --git "a/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" "b/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" index c9d35181..b9a13c9d 100644 --- "a/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" +++ "b/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" @@ -74,33 +74,33 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 未成年人保护法（第七十四、七十六条）. ；Hội đồng Nhà nước (国务院) (2023). 未成年人网络保护条例（国务院令第 766 号，第三十一、四十四条）. ；Văn phòng Tổng thư ký Cục Quản lý Không gian mạng Trung Quốc (中央网信办秘书局) (2026). 关于加强网络直播打赏规范管理的通知（第 7 项）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). 关于依法妥善审理涉新冠肺炎疫情民事案件若干问题的指导意见（二）（法发〔2020〕17 号，第九条）. - Ghi chú: Thực sự có tranh chấp hoàn tiền thì xử theo hướng có lợi nhất cho người chưa thành niên. Cách nói về việc này đã thay đổi. Ý kiến của bốn cơ quan năm 2022 và thông báo của Cục Quản lý Phát thanh truyền hình năm 2020 đều nói người chưa thành niên tuyệt đối không được thưởng tiền. Thông báo mới tháng 4 năm 2026 đổi thành chia bậc theo tám tuổi, mười sáu tuổi, nhưng không ghi rõ hai văn bản trước hết hiệu, cơ quan chính thức cũng không nói hai bên liên hệ với nhau thế nào. Nên câu「未成年人一律禁止打赏」bây giờ không còn chính xác -### 9. Nhận được thông báo vi phạm quyền thì phải xử lý kịp thời, chuyển bản giải trình mà sau 15 ngày không có động tĩnh thì khôi phục lại +### 9. Nhận được thông báo vi phạm thì phải xử lý kịp thời, chuyển bản giải trình mà sau 15 ngày không có động tĩnh thì khôi phục lại -- Chi phí: Không tốn tiền. Phải dựng một quy trình cố định: nhận thông báo, chuyển cho người bán, nhận giải trình của người bán, công bố kết quả ra công khai. Mỗi bước đều lưu hồ sơ. +- Chi phí: Không tốn tiền. Phải dựng một quy trình cố định: nhận thông báo, chuyển cho người bán, nhận giải trình của người bán, công bố kết quả ra công khai. Mỗi bước đều lưu hồ sơ - Nói thẳng: Có người gửi thông báo vi phạm kèm bằng chứng ban đầu thì bạn phải kịp thời xóa nội dung, chặn hoặc gỡ liên kết, đồng thời chuyển thông báo cho người bán. Kéo dài không xử lý thì phần thiệt hại phát sinh thêm bạn phải cùng người bán chịu. Người bán giải trình thì bạn chuyển giải trình sang phía bên kia và chờ 15 ngày. Bên kia không khiếu nại cũng không khởi kiện thì bạn khôi phục nội dung trở lại. -- Lợi ích: Chủ sở hữu quyền「有权通知电子商务平台经营者采取删除、屏蔽、断开链接、终止交易和服务等必要措施。通知应当包括构成侵权的初步证据」— có quyền thông báo cho bên vận hành nền tảng thương mại điện tử thực hiện các biện pháp cần thiết như xóa, chặn, gỡ liên kết, chấm dứt giao dịch và chấm dứt dịch vụ. Thông báo phải kèm bằng chứng ban đầu chứng minh có vi phạm. Chủ sở hữu quyền là người nắm các quyền như nhãn hiệu, bản quyền, bằng sở hữu trí tuệ. Chỉ gửi một câu thì không tính, phải kèm bằng chứng ban đầu. Nền tảng nhận được thông báo thì「应当及时采取必要措施，并将该通知转送平台内经营者；未及时采取必要措施的，对损害的扩大部分与平台内经营者承担连带责任」— phải kịp thời thực hiện các biện pháp cần thiết và chuyển thông báo đó cho người kinh doanh trong nền tảng; không kịp thời thực hiện thì chịu trách nhiệm liên đới với người kinh doanh trong nền tảng cho phần thiệt hại bị mở rộng thêm. Kéo dài không xử lý thì phần thiệt hại phát sinh thêm bạn phải cùng người bán chịu. Sau khi người bán giải trình, nền tảng「在转送声明到达知识产权权利人后十五日内，未收到权利人已经投诉或者起诉通知的，应当及时终止所采取的措施」— trong 15 ngày kể từ khi bản giải trình đã tới được chủ sở hữu quyền, nếu không nhận được thông báo rằng chủ sở hữu quyền đã khiếu nại hoặc đã khởi kiện thì phải kịp thời chấm dứt các biện pháp đã áp dụng. Nghĩa là chuyển giải trình của người bán sang chủ sở hữu quyền rồi chờ 15 ngày. Bên kia không khiếu nại cũng không khởi kiện thì khôi phục lại những gì đã xóa hoặc gỡ xuống. Nền tảng biết hoặc phải biết có vi phạm mà không xử lý thì chịu trách nhiệm liên đới với bên vi phạm. Cơ quan quản lý sở hữu trí tuệ còn có thể buộc bạn sửa trong thời hạn, quá hạn không sửa thì phạt tiền 5 vạn đến 50 vạn tệ, nghiêm trọng thì phạt 50 vạn đến 200 vạn tệ. Ngược lại,「恶意发出错误通知，造成平台内经营者损失的，加倍承担赔偿责任」— cố ý gửi thông báo sai làm người kinh doanh trong nền tảng chịu thiệt hại thì chịu trách nhiệm bồi thường gấp đôi (toàn quốc) +- Lợi ích: Chủ sở hữu quyền「有权通知电子商务平台经营者采取删除、屏蔽、断开链接、终止交易和服务等必要措施。通知应当包括构成侵权的初步证据」— có quyền thông báo cho bên vận hành nền tảng thương mại điện tử áp dụng các biện pháp cần thiết như xóa, chặn, gỡ liên kết, chấm dứt giao dịch và chấm dứt dịch vụ. Thông báo phải kèm bằng chứng ban đầu chứng minh có vi phạm. Chủ sở hữu quyền là người nắm các quyền như nhãn hiệu, bản quyền tác giả, bằng sáng chế. Chỉ gửi một câu thì không được, phải kèm bằng chứng ban đầu. Nền tảng nhận được thông báo thì「应当及时采取必要措施，并将该通知转送平台内经营者；未及时采取必要措施的，对损害的扩大部分与平台内经营者承担连带责任」— phải kịp thời áp dụng biện pháp cần thiết và chuyển thông báo đó cho người kinh doanh trong nền tảng; không kịp thời áp dụng thì chịu trách nhiệm liên đới với người kinh doanh trong nền tảng về phần thiệt hại bị mở rộng thêm. Kéo dài không xử lý thì phần thiệt hại phát sinh thêm bạn phải cùng người bán chịu. Sau khi người bán giải trình, nền tảng「在转送声明到达知识产权权利人后十五日内，未收到权利人已经投诉或者起诉通知的，应当及时终止所采取的措施」— trong 15 ngày sau khi bản giải trình đã tới được chủ sở hữu quyền sở hữu trí tuệ, nếu không nhận được thông báo rằng chủ sở hữu quyền đã khiếu nại hoặc đã khởi kiện thì phải kịp thời chấm dứt các biện pháp đã áp dụng. Nghĩa là chuyển giải trình của người bán cho chủ sở hữu quyền rồi chờ 15 ngày. Bên kia không khiếu nại cũng không khởi kiện thì khôi phục lại những gì đã xóa hoặc gỡ xuống. Nền tảng biết hoặc phải biết có vi phạm mà không xử lý thì chịu trách nhiệm liên đới với bên vi phạm. Cơ quan quản lý sở hữu trí tuệ còn buộc bạn sửa trong thời hạn, quá hạn không sửa thì phạt tiền 5 vạn đến 50 vạn tệ, nghiêm trọng thì phạt 50 vạn đến 200 vạn tệ. Ngược lại,「恶意发出错误通知，造成平台内经营者损失的，加倍承担赔偿责任」— cố ý gửi thông báo sai làm người kinh doanh trong nền tảng chịu thiệt hại thì chịu trách nhiệm bồi thường gấp đôi (toàn quốc) - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 电子商务法（第四十二、四十三、四十四、四十五、八十四条）. ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第一千一百九十五、一千一百九十六、一千一百九十七条）. -- Ghi chú: Bạn không phải nền tảng thương mại điện tử mà chỉ là một trang web thường thì áp dụng quy tắc thông báo và xóa của Bộ luật Dân sự. Khác biệt nằm ở chỗ sau khi bên kia giải trình thì bạn phải chờ bao lâu: Luật Thương mại điện tử có con số rõ là 15 ngày, còn Bộ luật Dân sự chỉ viết「合理期限」— một thời hạn hợp lý. Ngoài ra, việc công bố ra công khai thông báo đã nhận, giải trình của người bán và kết quả xử lý của bạn là yêu cầu rõ ràng trong Điều 44 của Luật Thương mại điện tử +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 电子商务法（第四十二、四十三、四十四、四十五、八十四条）. ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第一千一百九十五、第一千一百九十六、第一千一百九十七条）. +- Ghi chú: Bạn không phải nền tảng thương mại điện tử mà chỉ là một trang web thường thì áp dụng quy tắc thông báo và gỡ bỏ của Bộ luật Dân sự. Khác biệt nằm ở chỗ sau khi bên kia giải trình thì bạn phải chờ bao lâu: Luật Thương mại điện tử ấn định rõ 15 ngày, còn Bộ luật Dân sự chỉ viết「合理期限」— một thời hạn hợp lý. Ngoài ra, việc công bố ra công khai thông báo đã nhận, giải trình của người bán và kết quả xử lý của bạn là yêu cầu rõ ràng tại Điều 44 của Luật Thương mại điện tử ### 10. Đừng tùy tiện đặt thông tin người dùng ra nước ngoài, việc chuyển dữ liệu ra nước ngoài có điều kiện luật định và có ngưỡng số người - Chi phí: Không tốn tiền. Cách ít phiền nhất là giữ dữ liệu người dùng trong nước. Đúng là phải chuyển ra nước ngoài thì đi theo hợp đồng mẫu hoặc làm chứng nhận. - Nói thẳng: Chuyển thông tin của người dùng trong nước sang máy ở nước ngoài thì đó gọi là chuyển ra nước ngoài. Máy chủ đặt ở nước ngoài không có nghĩa dữ liệu không ra nước ngoài. Muốn chuyển ra thì phải đi một trong bốn con đường: đánh giá an toàn của cơ quan quản lý không gian mạng, chứng nhận của tổ chức chuyên môn, ký hợp đồng theo hợp đồng mẫu, hoặc những gì luật quy định khác. Còn phải báo cho người dùng biết bên nhận ở nước ngoài là ai, dùng để làm gì, và phải hỏi riêng người dùng về chuyện này một lần. -- Lợi ích: Đưa thông tin của người dùng trong nước cho công ty ở nước ngoài, hoặc lưu trên máy ở nước ngoài, thì theo luật gọi là cung cấp thông tin cá nhân ra nước ngoài, và phải「应当具备」một trong bốn điều kiện. Một là đánh giá an toàn do cơ quan quản lý không gian mạng tổ chức. Hai là chứng nhận bảo vệ thông tin cá nhân do tổ chức chuyên môn thực hiện. Ba là ký hợp đồng với bên nhận ở nước ngoài theo hợp đồng mẫu do cơ quan quản lý không gian mạng ban hành. Bốn là các điều kiện khác theo luật và pháp lệnh hành chính. Đồng thời bạn phải báo cho người dùng mấy thứ sau: tên của bên nhận ở nước ngoài, thông tin liên lạc, mục đích xử lý, cách thức xử lý, các loại thông tin cá nhân, và cách người dùng thực hiện quyền của chính mình. Và phải「取得个人的单独同意」— có được sự đồng ý riêng của cá nhân, tức là hỏi riêng người dùng về chuyện này một lần. Ngưỡng cụ thể chia theo tổng số người tích luỹ. Doanh nghiệp bình thường trong quy chế gọi là bên xử lý dữ liệu khác với bên vận hành hạ tầng thông tin trọng yếu. Loại doanh nghiệp này có một đường ở dưới thì không phải làm gì, nguyên văn là「自当年1月1日起累计向境外提供不满10万人个人信息（不含敏感个人信息）的」— tính từ ngày 1 tháng 1 năm đó trở đi, đã cung cấp ra nước ngoài thông tin cá nhân của dưới 100.000 người, không gồm thông tin cá nhân nhạy cảm. Không tới được đường đó thì cả ba việc là khai báo để đánh giá, ký hợp đồng mẫu và làm chứng nhận đều miễn. Từ 100.000 người trở lên mà chưa tới 1.000.000 người thì phải ký hợp đồng mẫu hoặc vượt qua chứng nhận. Từ 1.000.000 người trở lên, hoặc thông tin cá nhân nhạy cảm đạt từ 10.000 người trở lên, thì phải khai báo để đánh giá an toàn (toàn quốc, từ tháng 3 năm 2024) +- Lợi ích: Đưa thông tin của người dùng trong nước cho công ty ở nước ngoài, hoặc lưu trên máy ở nước ngoài, thì theo luật gọi là cung cấp thông tin cá nhân ra nước ngoài, và phải「应当具备」một trong bốn điều kiện. Một là đánh giá an toàn do cơ quan quản lý không gian mạng tổ chức. Hai là chứng nhận bảo vệ thông tin cá nhân do tổ chức chuyên môn thực hiện. Ba là ký hợp đồng với bên nhận ở nước ngoài theo hợp đồng mẫu do cơ quan quản lý không gian mạng ban hành. Bốn là các điều kiện khác theo luật và pháp lệnh hành chính. Đồng thời bạn phải báo cho người dùng mấy thứ sau: tên của bên nhận ở nước ngoài, thông tin liên lạc, mục đích xử lý, cách thức xử lý, các loại thông tin cá nhân, và cách người dùng thực hiện quyền của chính mình. Và phải「取得个人的单独同意」— có được sự đồng ý riêng của cá nhân, tức là hỏi riêng người dùng về chuyện này một lần. Ngưỡng cụ thể chia theo tổng số người tích luỹ. Doanh nghiệp thường trong quy chế gọi là bên xử lý dữ liệu khác với bên vận hành hạ tầng thông tin trọng yếu. Loại doanh nghiệp này có một ngưỡng không phải làm thủ tục gì, nguyên văn là「自当年1月1日起累计向境外提供不满10万人个人信息（不含敏感个人信息）的」— tính từ ngày 1 tháng 1 năm đó trở đi, đã cung cấp ra nước ngoài thông tin cá nhân của chưa đến 10 vạn người, không gồm thông tin cá nhân nhạy cảm. Không tới được ngưỡng đó thì cả ba việc khai báo để đánh giá, ký hợp đồng mẫu và làm chứng nhận đều miễn. Từ 10 vạn người đến chưa tới 100 vạn người thì phải ký hợp đồng mẫu hoặc vượt qua chứng nhận. Từ 100 vạn người trở lên, hoặc thông tin cá nhân nhạy cảm đạt từ 1 vạn người trở lên, thì phải khai báo để đánh giá an toàn (toàn quốc, từ tháng 3 năm 2024) - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). 个人信息保护法（第三十八、三十九、四十条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2024). 促进和规范数据跨境流动规定（网信办令第 16 号，第三、四、五、七、八条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2022). 数据出境安全评估办法（网信办令第 11 号，第四条）. - Ghi chú: Máy chủ đặt ở nước ngoài không đồng nghĩa với「数据没出境」. Người dùng đăng ký trong nước, bạn chuyển thông tin của họ sang máy ở nước ngoài, đó là chuyển ra nước ngoài. Có một loại không phải đi ba con đường đánh giá an toàn, chứng nhận và hợp đồng mẫu: mua sắm xuyên biên giới, gửi hàng xuyên biên giới, đặt vé máy bay và khách sạn, vì đó là「为订立、履行个人作为一方当事人的合同」— những chuyển dữ liệu thật sự cần thiết để lập hoặc thực hiện hợp đồng có cá nhân là một bên. Còn phải chú ý cách đếm số người: tính「自当年 1 月 1 日起累计」— từ ngày 1 tháng 1 năm đó trở đi, không phải đếm lùi mười hai tháng từ hôm nay ### 11. Chọn máy chủ thì xem sập có chịu được không, xem có ai trực vận hành không, cuối cùng mới so giá -- Chi phí: Máy chủ ảo của các hãng cloud lớn, mỗi tháng từ vài trăm đến vài nghìn tệ. VPS của nhà cung cấp nhỏ, mỗi tháng bắt đầu từ mấy chục tệ. VPS là một máy chủ ảo nhỏ. Thuê chỗ đặt máy chủ riêng thì mỗi năm bắt đầu từ vài nghìn tệ, và phần cứng phải tự mua. -- Nói thẳng: Ba bậc không khác nhau ở giá, mà khác nhau ở chỗ ngày xảy ra sự cố thì trong tay bạn có cái gì. Hãng lớn có kênh đăng ký khai báo, tư cách được công khai, hỏng thì có thông báo, có điều khoản bồi thường trong hợp đồng, và có người để tìm. Nhà cung cấp nhỏ rẻ hơn một nửa, nhưng khi nó bán vượt sức chứa, bỏ trốn, hoặc bị phía trên chặn thì bạn không biết kêu ai. Trang web tĩnh không có dữ liệu người dùng thì VPS mấy chục tệ là đủ dùng. -- Lợi ích: Ba bậc không khác nhau ở giá, mà khác nhau ở chỗ ngày xảy ra sự cố thì trong tay bạn có cái gì. Các hãng cloud lớn có kênh đăng ký khai báo, tư cách được công khai, có sự cố thì có thông báo, trong hợp đồng có điều khoản bồi thường, xảy ra chuyện thì tìm được người. Giá của nhà cung cấp nhỏ có thể thấp hơn một nửa, nhưng khi nó bán vượt sức chứa, bỏ trốn, hoặc bị phía trên chặn thì bạn không có đối tượng khiếu nại, nên trước khi bàn giá hãy tra hệ thống của Bộ Công nghiệp và Thông tin xem nó đã có giấy phép kinh doanh dịch vụ viễn thông giá trị gia tăng chưa. Thuê chỗ đặt máy chủ riêng chỉ đáng tiền khi cả hai điều kiện cùng đúng: băng thông hoặc chi phí phần cứng rẻ hơn hẳn, và bạn có người của mình trực vận hành. Cách chọn: trang web tĩnh không có dữ liệu người dùng thì VPS rẻ nhất là đủ. Trang web có đăng ký người dùng và giao dịch thì chọn máy chủ ảo co giãn theo nhu cầu và có ảnh chụp dự phòng, dữ liệu dự phòng để ở ít nhất hai chỗ. +- Chi phí: Máy chủ ảo của các nhà cung cấp dịch vụ đám mây lớn, mỗi tháng từ vài trăm đến vài nghìn tệ. VPS của nhà cung cấp nhỏ, mỗi tháng bắt đầu từ mấy chục tệ. VPS là một máy chủ ảo nhỏ. Thuê chỗ đặt máy chủ riêng thì mỗi năm bắt đầu từ vài nghìn tệ, và phần cứng phải tự mua. +- Nói thẳng: Ba bậc không khác nhau ở giá, mà khác nhau ở chỗ ngày xảy ra sự cố thì trong tay bạn có cái gì. Nhà cung cấp lớn có kênh đăng ký khai báo, tư cách được công khai, hỏng thì có thông báo, có điều khoản bồi thường trong hợp đồng, và có người để tìm. Nhà cung cấp nhỏ rẻ hơn một nửa, nhưng khi nó bán vượt sức chứa, bỏ trốn, hoặc bị phía trên chặn thì bạn không biết kêu ai. Trang web tĩnh không có dữ liệu người dùng thì VPS mấy chục tệ là đủ dùng. +- Lợi ích: Ba bậc không khác nhau ở giá, mà khác nhau ở chỗ ngày xảy ra sự cố thì trong tay bạn có cái gì. Các nhà cung cấp dịch vụ đám mây lớn có kênh đăng ký khai báo, tư cách được công khai, có sự cố thì có thông báo, trong hợp đồng có điều khoản bồi thường, xảy ra chuyện thì tìm được người. Giá của nhà cung cấp nhỏ có thể thấp hơn một nửa, nhưng khi nó bán vượt sức chứa, bỏ trốn, hoặc bị phía trên chặn thì bạn không có đối tượng khiếu nại, nên trước khi bàn giá hãy tra hệ thống của Bộ Công nghiệp và Thông tin xem nó đã có giấy phép kinh doanh dịch vụ viễn thông giá trị gia tăng chưa. Thuê chỗ đặt máy chủ riêng chỉ đáng tiền khi cả hai điều kiện cùng đúng: băng thông hoặc chi phí phần cứng rẻ hơn hẳn, và bạn có người của mình trực vận hành. Cách chọn: trang web tĩnh không có dữ liệu người dùng thì VPS rẻ nhất là đủ. Trang web có đăng ký người dùng và giao dịch thì chọn máy chủ ảo co giãn theo nhu cầu và có ảnh chụp dự phòng, dữ liệu dự phòng để ở ít nhất hai chỗ. - Mức bằng chứng: C - Nguồn: kinh nghiệm của tác giả, không có tài liệu trực tiếp; yêu cầu pháp lý về tư cách của nhà cung cấp và về đăng ký khai báo nằm ở mục 4 của phần này, nghĩa vụ giữ nhật ký 6 tháng và nghĩa vụ bảo vệ theo cấp độ nằm ở mục 16 ở phần 11 - Ghi chú: Phần trên là phán đoán từ kinh nghiệm, không có bằng chứng từ tài liệu. Chỉ có một việc kiểm chứng được thật sự: giấy phép của nhà cung cấp. Muốn tiết kiệm thì tiết kiệm ở cấu hình và băng thông, đừng tiết kiệm ở tư cách pháp lý ## Giấy phép -Nội dung phát hành theo [CC BY 4.0](../LICENSE). Nếu in lại hoặc sửa đổi, xin ghi nguồn và kèm liên kết bản gốc. Mã nguồn theo [MIT](../LICENSE-CODE). \ No newline at end of file +Nội dung phát hành theo [CC BY 4.0](../LICENSE). Nếu in lại hoặc sửa đổi, xin ghi nguồn và kèm liên kết bản gốc. Mã nguồn theo [MIT](../LICENSE-CODE). From 50f3450666782ad05f5b5233d649638048b5144f Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 13:09:38 +0700 Subject: [PATCH 21/25] =?UTF-8?q?sync-stats=20=E4=B9=9F=E5=9B=9E=E5=86=99?= =?UTF-8?q?=E8=AF=91=E6=9C=AC=E7=9A=84=20README=EF=BC=9A=E4=B8=89=E4=B8=AA?= =?UTF-8?q?=20README=20=E7=9A=84=E6=95=B0=E5=AD=97=E8=AF=A5=E6=B0=B8?= =?UTF-8?q?=E8=BF=9C=E4=B8=80=E6=A0=B7?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 以前 EDITS 只列了 README.md（中文那一份），en/README.md 和 vi/README.md 里的 首屏条目数和四个徽章停在 635 / 422 / 162 / 1407，真实是 641 / 425 / 165 / 1443。 越翻越旧：每翻完一节真实数字就往前走一点，译文里的数字不动，--check 也照样绿 （它只查自己列的那几处）。 补了六条 EDITS，两种语言各三条。首屏那句的字面每门语言不同（「641 条建议」 「641 pieces of advice」「641 lời khuyên」），所以正则按各自的句子结构写； 徽章是百分号编码的 URL，三份 README 形状一样，共用一条。 --- en/README.md | 8 ++++---- tools/sync-stats.mjs | 13 +++++++++++++ vi/README.md | 8 ++++---- 3 files changed, 21 insertions(+), 8 deletions(-) diff --git a/en/README.md b/en/README.md index 8548881e..58f1bf5f 100644 --- a/en/README.md +++ b/en/README.md @@ -5,14 +5,14 @@ # A High-Return Life Guide On living longer and getting sick less, and on what to do when someone has an accident. On not wasting money, and on which things get people scammed or sued. On what you can apply for when you have no job and no money, and what registration a shop, a company or a website needs. Also on dating, marriage and children, on going abroad, and on learning a trade.
    -635 pieces of advice. Each one states what it costs, what it buys back, how strong the evidence is, and cites nothing but journal papers and official documents. +641 pieces of advice. Each one states what it costs, what it buys back, how strong the evidence is, and cites nothing but journal papers and official documents. You do not have to do all of it. This is a shortlist ordered by value for money, not a task list. Taking one or two entries with you counts. The author has not done most of them either. [![Search online](https://img.shields.io/badge/%E5%9C%A8%E7%BA%BF%E6%A3%80%E7%B4%A2-%E7%82%B9%E8%BF%99%E9%87%8C%E6%89%93%E5%BC%80-3451b2?style=flat-square)](https://streamentry.github.io/HowToLiveBetter/) -[![Items](https://img.shields.io/badge/%E6%9D%A1%E7%9B%AE-635%20%E6%9D%A1-18794e?style=flat-square)](#contents) -[![Evidence grades](https://img.shields.io/badge/%E8%AF%81%E6%8D%AE%E5%88%86%E7%BA%A7-A%20422%20%C2%B7%20B%20162%20%C2%B7%20C%2051-915930?style=flat-square)](#evidence-grades) -[![Original sources](https://img.shields.io/badge/%E5%8E%9F%E5%A7%8B%E6%96%87%E7%8C%AE-1407%20%E6%9D%A1%E9%93%BE%E6%8E%A5-565a5f?style=flat-square)](docs/核实记录/) +[![Items](https://img.shields.io/badge/%E6%9D%A1%E7%9B%AE-641%20%E6%9D%A1-18794e?style=flat-square)](#contents) +[![Evidence grades](https://img.shields.io/badge/%E8%AF%81%E6%8D%AE%E5%88%86%E7%BA%A7-A%20425%20%C2%B7%20B%20165%20%C2%B7%20C%2051-915930?style=flat-square)](#evidence-grades) +[![Original sources](https://img.shields.io/badge/%E5%8E%9F%E5%A7%8B%E6%96%87%E7%8C%AE-1443%20%E6%9D%A1%E9%93%BE%E6%8E%A5-565a5f?style=flat-square)](docs/核实记录/) [![Licence](https://img.shields.io/badge/%E8%AE%B8%E5%8F%AF-CC%20BY%204.0-565a5f?style=flat-square)](#licence) ### [Open the search page](https://streamentry.github.io/HowToLiveBetter/) · [Let an AI answer from the book (skill)](skills/life-decision-guide/README.md) diff --git a/tools/sync-stats.mjs b/tools/sync-stats.mjs index c92c1c46..52b4b268 100644 --- a/tools/sync-stats.mjs +++ b/tools/sync-stats.mjs @@ -87,6 +87,9 @@ console.log(`条目 ${entries} ｜ 节 ${sections} ｜ A ${grade.A} B ${grade.B} console.log(`性价比 极高 ${ratio['极高']}（${pct['极高']}%） 高 ${ratio['高']}（${pct['高']}%） 一般 ${ratio['一般']}（${pct['一般']}%）`); console.log(''); +// 译本的 README 也带着同一批数字（首屏那句、四个徽章）。数字口径一样，正文也一样， +// 所以这里一并回写——以前只管 README.md 那一份，译文里的数字停在某一节翻完那一刻， +// 越翻越旧（2026-10-02 en/vi 停在 635，真实 641）。三个 README 的这些数字该永远一样。 const EDITS = [ ['README.md', '首屏条目数', /(\d+) 条建议/g, `${entries} 条建议`], ['README.md', '条目徽章', /%E6%9D%A1%E7%9B%AE-(\d+)%20%E6%9D%A1/g, `%E6%9D%A1%E7%9B%AE-${entries}%20%E6%9D%A1`], @@ -99,6 +102,16 @@ const EDITS = [ ['README.md', '性价比段', /全书 (\d+) 条中性价比极高 \d+ 条（\d+%）、高 \d+ 条（\d+%）、一般 \d+ 条（\d+%）/g, `全书 ${entries} 条中性价比极高 ${ratio['极高']} 条（${pct['极高']}%）、高 ${ratio['高']} 条（${pct['高']}%）、一般 ${ratio['一般']} 条（${pct['一般']}%）`], ['README.md', '正文文件数', /正文按节拆成 (\d+) 个文件/g, `正文按节拆成 ${sections} 个文件`], + // 译本 README：首屏那句的数字、四个徽章。字面不同（英越各有各的说法），只按形状替换， + // 所以每种语言各列一条正则——数字的位置是那门语言的句子结构决定的。 + ['en/README.md', '首屏条目数', /(\d+) pieces of advice/g, `${entries} pieces of advice`], + ['en/README.md', '条目徽章', /%E6%9D%A1%E7%9B%AE-(\d+)%20%E6%9D%A1/g, `%E6%9D%A1%E7%9B%AE-${entries}%20%E6%9D%A1`], + ['en/README.md', '证据分级徽章', /A%20(\d+)%20%C2%B7%20B%20\d+%20%C2%B7%20C%20\d+/g, `A%20${grade.A}%20%C2%B7%20B%20${grade.B}%20%C2%B7%20C%20${grade.C}`], + ['en/README.md', '文献链接徽章', /-(\d+)%20%E6%9D%A1%E9%93%BE%E6%8E%A5/g, `-${links}%20%E6%9D%A1%E9%93%BE%E6%8E%A5`], + ['vi/README.md', '首屏条目数', /(\d+) lời khuyên/g, `${entries} lời khuyên`], + ['vi/README.md', '条目徽章', /%E6%9D%A1%E7%9B%AE-(\d+)%20%E6%9D%A1/g, `%E6%9D%A1%E7%9B%AE-${entries}%20%E6%9D%A1`], + ['vi/README.md', '证据分级徽章', /A%20(\d+)%20%C2%B7%20B%20\d+%20%C2%B7%20C%20\d+/g, `A%20${grade.A}%20%C2%B7%20B%20${grade.B}%20%C2%B7%20C%20${grade.C}`], + ['vi/README.md', '文献链接徽章', /-(\d+)%20%E6%9D%A1%E9%93%BE%E6%8E%A5/g, `-${links}%20%E6%9D%A1%E9%93%BE%E6%8E%A5`], ['tools/og.html', 'og 条目数', /(\d+)<\/b> 条建议/g, `${entries} 条建议`], ['tools/og.html', 'og A 级数', /A 级证据 (\d+)<\/b> 条/g, `A 级证据 ${grade.A} 条`], ['tools/og.html', 'og 链接数', /(\d+)<\/b> 条原始文献链接/g, `${links} 条原始文献链接`], diff --git a/vi/README.md b/vi/README.md index 87336cac..679d4ff2 100644 --- a/vi/README.md +++ b/vi/README.md @@ -5,14 +5,14 @@ # Cẩm Nang Sống Hiệu Quả Cao Về cách sống lâu và cách ít bệnh hơn, và về việc làm gì khi có người bị tai nạn. Về cách đừng phí tiền, và về những việc gì khiến người ta bị lừa đảo hoặc bị kiện tụng. Về bạn có thể xin được những gì khi không có việc làm và không có tiền, và một cửa hàng, một công ty hay một trang web cần đăng ký những giấy tờ gì. Cũng về yêu đương, kết hôn và con cái, về đi nước ngoài, và về học một nghề.
    -635 lời khuyên. Mỗi lời nêu rõ tốn cái gì, đổi lại được cái gì, bằng chứng chắc tới đâu, và chỉ trích dẫn bài báo bài khoa cùng tài liệu chính thức. +641 lời khuyên. Mỗi lời nêu rõ tốn cái gì, đổi lại được cái gì, bằng chứng chắc tới đâu, và chỉ trích dẫn bài báo bài khoa cùng tài liệu chính thức. Bạn không nhất thiết phải làm hết. Đây là danh sách ngắn đã xếp theo hiệu quả chi phí, không phải danh sách việc phải làm. Mang đi một hai mục là được. Tác giả cũng chưa làm phần lớn trong số đó. [![Search online](https://img.shields.io/badge/%E5%9C%A8%E7%BA%BF%E6%A3%80%E7%B4%A2-%E7%82%B9%E8%BF%99%E9%87%8C%E6%89%93%E5%BC%80-3451b2?style=flat-square)](https://streamentry.github.io/HowToLiveBetter/vi/) -[![Items](https://img.shields.io/badge/%E6%9D%A1%E7%9B%AE-635%20%E6%9D%A1-18794e?style=flat-square)](#mục-lục) -[![Evidence grades](https://img.shields.io/badge/%E8%AF%81%E6%8D%AE%E5%88%86%E7%BA%A7-A%20422%20%C2%B7%20B%20162%20%C2%B7%20C%2051-915930?style=flat-square)](#mức-bằng-chứng) -[![Original sources](https://img.shields.io/badge/%E5%8E%9F%E5%A7%8B%E6%96%87%E7%8C%AE-1407%20%E6%9D%A1%E9%93%BE%E6%8E%A5-565a5f?style=flat-square)](docs/核实记录/) +[![Items](https://img.shields.io/badge/%E6%9D%A1%E7%9B%AE-641%20%E6%9D%A1-18794e?style=flat-square)](#mục-lục) +[![Evidence grades](https://img.shields.io/badge/%E8%AF%81%E6%8D%AE%E5%88%86%E7%BA%A7-A%20425%20%C2%B7%20B%20165%20%C2%B7%20C%2051-915930?style=flat-square)](#mức-bằng-chứng) +[![Original sources](https://img.shields.io/badge/%E5%8E%9F%E5%A7%8B%E6%96%87%E7%8C%AE-1443%20%E6%9D%A1%E9%93%BE%E6%8E%A5-565a5f?style=flat-square)](docs/核实记录/) [![Licence](https://img.shields.io/badge/%E8%AE%B8%E5%8F%AF-CC%20BY%204.0-565a5f?style=flat-square)](#giấy-phép) ### [Mở trang tra cứu](https://streamentry.github.io/HowToLiveBetter/vi/) · [Để AI trả lời theo sách (skill)](skills/life-decision-guide/README.md) From ce548919348d7a5969c7134b5e3edc91df87c508 Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 13:10:13 +0700 Subject: [PATCH 22/25] =?UTF-8?q?book.yml=20=E7=9A=84=E6=B3=A8=E9=87=8A?= =?UTF-8?q?=E8=B7=9F=E4=B8=8A=EF=BC=9A=E8=AF=91=E6=96=87=E6=A3=80=E6=9F=A5?= =?UTF-8?q?=E7=8E=B0=E5=9C=A8=E8=BF=98=E6=9F=A5=E8=8A=82=E6=A0=87=E9=A2=98?= =?UTF-8?q?=E3=80=81=E5=AF=BC=E8=AF=BB=E3=80=81=E6=9D=A1=E7=9B=AE=E6=A0=87?= =?UTF-8?q?=E9=A2=98?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .github/workflows/book.yml | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/.github/workflows/book.yml b/.github/workflows/book.yml index 4d065fb9..6d417a92 100644 --- a/.github/workflows/book.yml +++ b/.github/workflows/book.yml @@ -47,8 +47,10 @@ env: jobs: # 译文检查：拿中文原文逐条对 en/ 和 vi/（tools/i18n/check.mjs，判据在 tools/i18n/structure.mjs）。 - # 查的是丢东西：节数条数、六个字段、成本标签注释、证据等级、收益/成本/来源三栏的数字、 - # 交叉引用的条号节号、链接数量。翻译 PR 在这里红就知道该翻哪一条，不用作者读译文。 + # 查的是丢东西：节数条数、六个字段、成本标签注释、证据等级、争议和待核实的标记、 + # 收益/成本/来源三栏的数字、交叉引用的条号节号、链接数量，以及三处容易漏的文字—— + # 节的标题（要与 README 目录一致）、节的导读、条目标题。 + # 翻译 PR 在这里红就知道该翻哪一条，不用作者读译文。 # 单独一个 job：译文没对齐不该连累 EPUB、PDF 和离线单文件的发布。 i18n: name: 译文检查 From 6825568dee8037526be20a4fa7a7ebd444184364 Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 13:11:20 +0700 Subject: [PATCH 23/25] =?UTF-8?q?CLAUDE.md=20=E8=B7=9F=E4=B8=8A=EF=BC=9A?= =?UTF-8?q?=E8=AF=91=E6=96=87=E7=9A=84=E4=B8=89=E5=A4=84=E6=96=B0=E5=88=A4?= =?UTF-8?q?=E6=8D=AE=E3=80=81=E6=95=B0=E5=AD=97=E7=9A=84=E5=8F=8C=E5=90=91?= =?UTF-8?q?=E8=A6=81=E6=B1=82=E3=80=81=E4=B8=89=E4=B8=AA=20README=20?= =?UTF-8?q?=E7=9A=84=E5=90=8C=E6=AD=A5?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- CLAUDE.md | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/CLAUDE.md b/CLAUDE.md index bb2bbe83..4b5f5d53 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -66,7 +66,7 @@ - **有因果、转折关系的相邻短句用「因为」「所以」「但」连起来**，单句仍不超过 50 字。 - 收益与口径是后加的两个字段，用来回答「值不值得做」（证据等级只回答「数字可不可信」，两者正交）。口径按这条主要换回什么定：死亡率（含健康终点、存活率）、金钱（含保障待遇）、时间（含精力）、自由（含法律责任、个人信息）。**不同口径之间不做比较**。 - 收益量级尽量按阈值从条目自己的「收益」栏机械套，不凭感觉：死亡率口径看相对降幅（≥20% 大 / 10–20% 中 / <10% 或只有替代终点 小）；金钱看金额（万元级 大 / 数百到数千 中 / 几十元 小）；自由看后果（避免刑责 大 / 避免拘留或行政处罚 中 / 避免民事纠纷 小）；时间看节省量（每天小时级 大 / 每周小时级 中 / 一次性 小）。数字不足以判定时才用判断，并在核实记录里写明凭什么定的。 -- 条目数、A 级条数变了要同步四处：README（正文数字与两个徽章）、`tools/site/stats.json`（检索页三种语言的数字都从这儿来）、CLAUDE.md 的目录简介，以及 **tools/og.html + 重新截图 og.png**（命令写在 og.html 文件头注释里，链接数的口径是 book/ 下所有「- 来源：」和「- 备注：」行里的 http(s) 总数）。跑 `node tools/sync-stats.mjs` 一键完成（2026-09-29 起全部用 node，原来的 sync-stats.ps1 已删，Linux、macOS 上照样跑）：它重算条目数、节数、A/B/C 数、争议数、TODO 数、链接数和性价比三档，回写 README 的九处（含「拆成 N 个文件」）、`tools/site/stats.json`、tools/og.html 的三个数字，再用全新的 user-data-dir 重出 og.png（`--no-screenshot` 可只改数字；Chrome 不在常见位置就设环境变量 `CHROME`）。**新增要随统计变的数字，就往 sync-stats.mjs 的 EDITS 表里加一行**，别在正文里另写死一处。CI 的「统计数字检查」job 跑 `node tools/sync-stats.mjs --check`，数字过时就红（2026-09-29 加，此前网页上直接合并的 PR 完全不经过这个脚本）；它只查不改，og.png 仍要本地重出。**CLAUDE.md 的目录简介脚本不碰，仍要手工改。**检索页的数字不再写死在页面源码里**（2026-09-30 多语言时改的）：模板和三份字典里写的是 `{{entries}}` 这类占位符，构建时从 `stats.json` 填——三种语言共用一份数字，谁也不会和谁差。`sync-stats.mjs` 只管把 `stats.json` 写对，网站下次构建（`web.yml` 的触发路径含正文和 `tools/site/**`）自动带上新数字，不在本地调 Astro 构建。**性价比档的规则和检索页模板（`tools/site/page.template.html`）的 COST_W、e.ratio 两行绑定，那两行改了脚本会直接报错要求同步。**每次改完条目就跑一次，和正文改动一起提交**（2026-09-18 用户定；当天早些时候的「等我说改完了再改」已作废——脚本把一次同步的开销压到几百 token，不必再攒着）。跑完不用打开 og.png：脚本自己校验图是本次写入、大小在 120KB 到 400KB，只有改过 tools/og.html 版式时才值得看一眼。汇报里也不必再写「待同步」，直接给同步后的数字。 +- 条目数、A 级条数变了要同步四处：README（正文数字与两个徽章）、`tools/site/stats.json`（检索页三种语言的数字都从这儿来）、CLAUDE.md 的目录简介，以及 **tools/og.html + 重新截图 og.png**（命令写在 og.html 文件头注释里，链接数的口径是 book/ 下所有「- 来源：」和「- 备注：」行里的 http(s) 总数）。跑 `node tools/sync-stats.mjs` 一键完成（2026-09-29 起全部用 node，原来的 sync-stats.ps1 已删，Linux、macOS 上照样跑）：它重算条目数、节数、A/B/C 数、争议数、TODO 数、链接数和性价比三档，回写 README 的九处（含「拆成 N 个文件」）、**译本 README 的首屏数字和三个徽章**（en/vi 那两份，2026-10-02 加）、`tools/site/stats.json`、tools/og.html 的三个数字，再用全新的 user-data-dir 重出 og.png（`--no-screenshot` 可只改数字；Chrome 不在常见位置就设环境变量 `CHROME`）。**新增要随统计变的数字，就往 sync-stats.mjs 的 EDITS 表里加一行**，别在正文里另写死一处。CI 的「统计数字检查」job 跑 `node tools/sync-stats.mjs --check`，数字过时就红（2026-09-29 加，此前网页上直接合并的 PR 完全不经过这个脚本）；它只查不改，og.png 仍要本地重出。**CLAUDE.md 的目录简介脚本不碰，仍要手工改。**检索页的数字不再写死在页面源码里**（2026-09-30 多语言时改的）：模板和三份字典里写的是 `{{entries}}` 这类占位符，构建时从 `stats.json` 填——三种语言共用一份数字，谁也不会和谁差。`sync-stats.mjs` 只管把 `stats.json` 写对，网站下次构建（`web.yml` 的触发路径含正文和 `tools/site/**`）自动带上新数字，不在本地调 Astro 构建。**性价比档的规则和检索页模板（`tools/site/page.template.html`）的 COST_W、e.ratio 两行绑定，那两行改了脚本会直接报错要求同步。**每次改完条目就跑一次，和正文改动一起提交**（2026-09-18 用户定；当天早些时候的「等我说改完了再改」已作废——脚本把一次同步的开销压到几百 token，不必再攒着）。跑完不用打开 og.png：脚本自己校验图是本次写入、大小在 120KB 到 400KB，只有改过 tools/og.html 版式时才值得看一眼。汇报里也不必再写「待同步」，直接给同步后的数字。 - **插入或删除条目之后，必须扫一眼 `docs/引用对照.md` 的 diff**（2026-09-19 用户问「怎么防范引用错位」时定的）。正文里 257 处「第 X 条」引用只记条号不记内容，插一条进去后面的条号集体顺延，引用却留在原地——而顺延后的条号通常**仍在范围内**，查越界一条也抓不到（第 7 节曾有 6 处指错：医疗救助指到低保、救助站指错条，全是这么来的）。`sync-stats.mjs` 会自动跑 `tools/check-refs.mjs`，把每处引用**实际指向的条目标题**连同引用处的上下文写进 `docs/引用对照.md`，该文件入库。**判据：条号没变而「指向的条目」那列变了，就是被撞歪的引用**；表里还带着引用处前十几个字，和标题一对就知道指没指对。CI 用 `node tools/check-refs.mjs --check`（有问题则退出码 1），它是 `book.yml` 里名为「交叉引用检查」的独立 job，**故意不挂在 build 里**：锚点没补好不该连累 EPUB、PDF 和离线单文件的发布，PR 上照样红、main 上该发的还是发出去。这个 job 是 2026-09-21 补的——在那之前 CLAUDE.md 写着「CI 用」但 workflow 里其实没有这一步，全靠本地 `sync-stats.ps1`，而那条链路只在 Windows 上跑、找不到 node 时还会静默跳过（脚本里就是打一句「跳过引用对照表」然后继续），外部 PR 更是完全跑不到。 @@ -115,10 +115,13 @@ ### 译文怎么维护 - **流程和判据写在 `tools/i18n/BRIEF.md`**，翻译任务的提示词和结构检查的判据是同一份。词表 `tools/i18n/glossary.{en,vi}.mjs`。 - 一次翻一节。`node tools/i18n/split.mjs book/NN-*.md --locale en` 把它切成十来个条一块放进 `.i18n/en/<节名>/partNN.md`（**已翻过的块不会被覆盖**，改参数重跑不会冲掉译文），翻译完 `node tools/i18n/join.mjs book/NN-*.md --locale en` 逐块校一遍再合成 `en/book/NN-*.md`。一轮翻好几节用 `bash tools/i18n/join-ready.sh en vi`——它按「块里还带 `- 成本：` 就是没翻」挑出翻全了的节，挨个 join。 -- `node tools/i18n/check.mjs`（CI 的「译文检查」job）拿中文原文逐条对译文：条数条号、六个字段的有无与顺序、**成本标签注释逐字照抄**、证据等级、争议与待核实的条数、**收益/成本/来源三栏的数字一个不多一个不少**（量级词换算除外，见 `tools/i18n/structure.mjs` 里 `numbersOf` 的说明）、「第 X 条」指路的条号节号、来源与备注里的链接数量。判据在 `structure.mjs`，join 和 check 共用同一份。 -- **译文不许改数字**。中文「167 万」在英文里只能写成 1,670,000、在越南文里只能写成 1.670.000，这是量级换算，`structure.mjs` 放行；除此之外数字对不上就是错。这一层机器查得出来，所以译文不需要作者审。 +- `node tools/i18n/check.mjs`（CI 的「译文检查」job）拿中文原文逐条对译文：条数条号、六个字段的有无与顺序、**成本标签注释逐字照抄**、证据等级、争议与待核实的条数、**收益/成本/来源三栏的数字一个不多一个不少**（量级词换算与中文数字法条编号除外，见 `tools/i18n/structure.mjs` 里 `numbersOf`、`cnNumbersOf` 的说明）、「第 X 条」指路的条号节号、来源与备注里的链接数量。判据在 `structure.mjs`，join 和 check 共用同一份。 +- **三处以前没人查、现在查了**：节的标题、节的导读、条目标题。`join.mjs` 以前还拿中文原文的节首无条件拼在译文前面（`tools/i18n/join.mjs` 的第 0 块那一段），结果 en 34 节里 30 节的标题被当成「多出来的」删掉、换成中文那份，页面上小节标题是中文的、有的还渲染两遍。2026-10-02 改掉（拼的时候不碰译文块），判据加在 `STRUCT.compareSection()` 和 `compare()` 里。**节标题还要跟该语言 README 目录里那一份一字不差**（页面侧栏读的是 README），两处都得改：README 那一行 + `.i18n/<语言>/<节名>/part00.md` 那一行；`book/` 里的那份是 join 生成的，改了它下次就被覆盖。 +- **译文不许改数字**。中文「167 万」在英文里只能写成 1,670,000、在越南文里只能写成 1.670.000，这是量级换算，`structure.mjs` 放行；除此之外数字对不上就是错。反过来也成立：**原文没有量级词时那三个数必须原样留着**——「10 万元」写成 100.000 是把原文的数字换掉了（收益栏的可核对性就断在那儿）。中文数字的法条编号（「第一千零四十五条」→「Điều 1045」）不算改数字。 - 越南文里 **条目 = "mục"，节 = "phần"，法条的款 = "khoản (2)"**——三个词不能混。「第 13 节」写成 "phần 13"，写成 "mục 13" 读者就当成第 13 条了。 - **各语言共用一套中文取值**：成本标签注释（`钱=0 时间=少 …`）、筛选按钮的 `data-v`、`#hash` 锚点、地址栏参数。这不是漏译，是刻意的，译文里一字不改。 +- **三个 README 的统计数字由 `sync-stats.mjs` 一起回写**（首屏条目数、条目/分级/链接三个徽章）。以前 `EDITS` 只列了中文那一份，en/vi 的数字停在某一节翻完那一刻越停越旧，而 `--check` 照样绿（它只查自己列的几处）。首屏那句三种语言字面不同（`641 条建议` / `641 pieces of advice` / `641 lời khuyên`），所以正则按各自的句子写；徽章是百分号编码的 URL，三份共用一条正则。 +- **散文里不该夹中文**。法条原文、法条名书名、机构名括注、书号（`国务院令第 768 号`）这些照抄是允许的（BRIEF 规则 6），但句子中间夹一个中文词就是漏译——越南语里出现 `cũng普遍 đánh giá` 这种，读者看得见。BRIEF 第 8 条写了，`check.mjs` 查的是节标题、导读和条目标题那三处。 README 开头「这本书想回答的问题」表**一节只占一行**（2026-09-19 用户定）：同一个链接贴两遍，读者要在同一个目的地上停两次，两行之间又没有逻辑差别。一节覆盖面宽就在一行里用逗号串几个问题、一个问号收尾（第 19 节「加班费、年休假该怎么算，被裁该拿多少补偿，上班受了伤怎么认定和拿钱？」是范式），不要拆成两行。当天把第 6、8、24 节原有的两行都并掉了，现在是 32 行对 32 节。 From 676c28a88ce7b68a66329bdc1c4d60e8b7b6798b Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 13:18:19 +0700 Subject: [PATCH 24/25] =?UTF-8?q?=E9=95=BF=E6=96=87=EF=BC=88docs/=EF=BC=89?= =?UTF-8?q?=E7=9A=84=E6=9D=A5=E6=BA=90=E6=AE=B5=E4=B9=9F=E6=8C=89=E8=A7=84?= =?UTF-8?q?=E5=88=99=206=20=E6=94=B6=E6=8B=BE=EF=BC=9A=E6=9C=BA=E6=9E=84?= =?UTF-8?q?=E5=90=8D=E8=AF=91+=E4=B8=AD=E6=96=87=E6=8B=AC=E6=B3=A8?= =?UTF-8?q?=EF=BC=8C=E4=BD=9C=E8=80=85=E5=88=97=E8=A1=A8=E7=9A=84=E3=80=8C?= =?UTF-8?q?=E7=AD=89=E3=80=8D=E6=94=B9=20et=20al.?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 长文的来源段不在 check.mjs 的判据里（那只查 book/*.md 的条目），所以没人管。 扫了一遍 en/vi 五篇长文： · 家庭应急装备清单里三个机构名（应急管理部、北京市应急管理局、全国人大常委会） 只写中文，按 BRIEF 规则 6 该「译名（中文）」。en/vi 都改了。 · 作者列表的「等」是中文的「等人」，不是书目内容，en/vi 各五处改 et al. 书目其余部分（文件标题、法条名）照抄不动——规则 6 就是这个意思，读者要能对照原文查。 --- ...45\350\243\205\345\244\207\346\270\205\345\215\225.md" | 8 ++++---- ...51\351\222\237\345\222\214\345\244\234\347\217\255.md" | 4 ++-- ...45\350\243\205\345\244\207\346\270\205\345\215\225.md" | 8 ++++---- ...51\351\222\237\345\222\214\345\244\234\347\217\255.md" | 4 ++-- 4 files changed, 12 insertions(+), 12 deletions(-) diff --git "a/en/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" "b/en/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" index 7b64c1b2..01b43a9d 100644 --- "a/en/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" +++ "b/en/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" @@ -22,11 +22,11 @@ The Ministry of Emergency Management issued a National Basic Household Emergency | 10 | Disinfecting wet wipes | 用于个人卫生清洁 — for personal hygiene | | 11 | Surgical masks | 病毒防护 — protection against viruses | -Sources: 应急管理部 (2020). 全国基础版家庭应急物资储备建议清单. +Sources: Ministry of Emergency Management (应急管理部) (2020). 全国基础版家庭应急物资储备建议清单. The local version lists more detail. The Beijing Municipal Emergency Management Bureau gives households a list in two tiers, a basic version and an extended version. The basic version is split into 「应急物品、应急工具和应急药物 3 类」, which means 3 categories, emergency supplies, emergency tools and emergency medicines. The items themselves are these: a hand-crank rechargeable torch that can receive broadcasts, a survival whistle, towels, tissues or wet wipes, a respirator, multi-function combination scissors, an emergency escape rope, a fire extinguisher or a fire blanket. Medicines and medical materials are counted separately: anti-infective and cold medicines, surgical masks, medical materials such as gauze and bandages, and iodine swabs. The extended version is split into 「食品、个人用品、逃生自救工具、医疗急救用品、重要文件资料 5 类应急物资」, which means 5 categories of emergency supplies, food, personal items, escape and self-rescue tools, first-aid supplies, and important documents. The bureau also gave three recommendations. The first is 「选购资质合法、信誉良好的生产经营企业提供的应急物资」, which means buy emergency supplies from producers and sellers who are properly licensed and have a good reputation. The second is 「优先储备基础版的应急物资品种」, which means stock the items in the basic version first. The third is 「熟悉掌握应急物资的正确使用方法，定期对应急物资状况进行检查，并及时更换已过保质期的应急物资」, which means know the correct way to use emergency supplies, check the condition of emergency supplies at regular intervals, and replace anything past its shelf life promptly. -Sources: 北京市应急管理局 (2020). 北京市修订发布居民家庭应急物资储备建议清单. ；北京市西城区人民政府转载的品种明细. +Sources: Beijing Municipal Emergency Management Bureau (北京市应急管理局) (2020). 北京市修订发布居民家庭应急物资储备建议清单. ；Beijing Xicheng District Government (北京市西城区人民政府)转载的品种明细. The national level was updated once more in 2024. The document is the Guiding Opinions on Further Strengthening the System and Capacity for Ensuring Supplies for Emergency Rescue and Disaster Relief, issued by the Office of the National Committee on Disaster Prevention, Mitigation and Relief (国防减救办发〔2024〕13 号, dated 23 September 2024). Its Annex 2 is called the Guiding Catalogue for Household Emergency Supplies Storage. The catalogue is split into 5 categories: emergency food, daily items, emergency tools, emergency medicines and medical supplies, and important documents. The basic version has 16 items, the extended version 31 items. TODO (to be verified: what the 16 and the 31 items are called. The official annex is a wps file, and there is also a PDF of it scanned into images, and this time no text could be obtained that could be checked word for word). @@ -74,7 +74,7 @@ The Beijing Municipal Emergency Management Bureau puts it this way: 「定期对 - **A packaged "emergency kit".** These kits are mostly padded out with a cheap torch and a mask. Buying the fire extinguisher, the fire blanket, the respirator and the first-aid kit separately costs less, and it is easier to check the certification one item at a time - **Food and water for more than three days.** The official list works it out at 3 days per person. Most of what you stack up beyond that ends up thrown away after it expires, see item 24 in section 5 (do not stockpile for sales promotions) -Sources: 全国人大常委会 (2021). 中华人民共和国消防法（2021 修正，第二十四条）. +Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). 中华人民共和国消防法（2021 修正，第二十四条）. ## 7. How much use this equipment really is @@ -84,6 +84,6 @@ One Cochrane systematic review pooled 98 studies and 2,605,044 people. It assess In the same review, these interventions did improve people's safety habits. More households had a working smoke alarm fitted, OR 1.81 (95% CI 1.30 to 2.52), a share about 80% higher. More had worked out an escape plan in advance, OR 2.01 (1.45 to 2.77), about double. More stored their medicines properly, OR 1.53 (1.27 to 1.84), about 50% higher. More had a fire door fitted at the head of the stairs, OR 1.61 (1.19 to 2.17), about 60% higher. -Sources: Kendrick D 等 (2012). Home safety education and provision of safety equipment for injury prevention. Cochrane Database of Systematic Reviews. +Sources: Kendrick D et al. (2012). Home safety education and provision of safety equipment for injury prevention. Cochrane Database of Systematic Reviews. So buying everything on the official list in one go costs little and counts as reasonable preparation. But do not think that having bought it you are safe. Research has repeatedly shown that three things work: fitting smoke alarms, working out in advance how to get out, and moving out of the way the things in the home that can hurt people. Either none of these three is on a shopping list, or it takes up only a small part of one. diff --git "a/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" "b/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" index 6275ecd2..8bf3e720 100644 --- "a/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" +++ "b/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" @@ -88,7 +88,7 @@ A systematic review that looked specifically for randomised trials of this kind So that is all that can be said at present. **Concentrating your eating in the daytime as far as possible during night shifts is supported by a randomised trial, but the evidence is still thin.** It costs nothing and it is not hard to do, so if you want to try it, try it. There is a second thing to try, in the next section. Use light to move the whole body clock across. -Sources: Chellappa SL 等 (2021). Daytime eating prevents internal circadian misalignment and glucose intolerance in night work. Science Advances. ；Chellappa SL 等 (2025). Daytime eating during simulated night work mitigates changes in cardiovascular risk factors: secondary analyses of a randomized controlled trial. Nature Communications. ；Wulandari F 等 (2026). Cardiometabolic status among shift workers under meal time regulation: A systematic review of randomized controlled trials. Nutrition and Health. +Sources: Chellappa SL et al. (2021). Daytime eating prevents internal circadian misalignment and glucose intolerance in night work. Science Advances. ；Chellappa SL et al. (2025). Daytime eating during simulated night work mitigates changes in cardiovascular risk factors: secondary analyses of a randomized controlled trial. Nature Communications. ；Wulandari F et al. (2026). Cardiometabolic status among shift workers under meal time regulation: A systematic review of randomized controlled trials. Nutrition and Health. ## 8. Whether the clock can be moved across @@ -128,7 +128,7 @@ A meta-analysis pooling 11 studies: light therapy gave shift workers about 32.5 Another laboratory study used intermittent bright light. In the treatment group, melatonin secretion overlapped the scheduled sleep period by 4.90 hours, against only 2.62 hours in the control group (P=0.002). The drop in cognitive test scores that showed up in the control group on the last night shift was held down in the treatment group. The authors say outright in their conclusion that this approach still has to be tested on real night-shift workers. -Sources: Zhao C, Li N, Miao W, He Y, Lin Y (2025). A systematic review and meta-analysis on light therapy for sleep disorders in shift workers. Scientific Reports. ；Lammers-van der Holst HM 等 (2021). Efficacy of intermittent exposure to bright light for treating maladaptation to night work on a counterclockwise shift work rotation. Scandinavian Journal of Work, Environment & Health. +Sources: Zhao C, Li N, Miao W, He Y, Lin Y (2025). A systematic review and meta-analysis on light therapy for sleep disorders in shift workers. Scientific Reports. ；Lammers-van der Holst HM et al. (2021). Efficacy of intermittent exposure to bright light for treating maladaptation to night work on a counterclockwise shift work rotation. Scandinavian Journal of Work, Environment & Health. ### One thing nobody has tested diff --git "a/vi/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" "b/vi/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" index fdc0e74e..5605d550 100644 --- "a/vi/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" +++ "b/vi/docs/\345\256\266\345\272\255\345\272\224\346\200\245\350\243\205\345\244\207\346\270\205\345\215\225.md" @@ -22,11 +22,11 @@ Bộ Quản lý Khẩn cấp đã ban hành một Danh mục vật tư dự tr | 10 | Khăn ướt sát trùng | 用于个人卫生清洁 — dùng để vệ sinh cá nhân | | 11 | Khẩu cầu phẫu khoa | 病毒防护 — phòng virus | -Nguồn: 应急管理部 (2020). 全国基础版家庭应急物资储备建议清单. +Nguồn: Bộ Quản lý Tình huống và Khẩn cấp (应急管理部) (2020). 全国基础版家庭应急物资储备建议清单. Bản của địa phương liệt kê chi tiết hơn. Cục Quản lý Khẩn cấp Bắc Kinh đưa cho hộ gia đình một danh mục chia hai mức: phiên bản cơ bản và phiên bản mở rộng. Phiên bản cơ bản chia thành 「应急物品、应急工具和应急药物 3 类」, tức là 3 nhóm, vật phẩm khẩn cấp, công cụ khẩn cấp và thuốc khẩn cấp. Các món cụ thể là: đèn pin quay tay sạp lại được có thể thu đài, còi cứu sinh, khăn và giấy ăn hoặc khăn ướt, mặt nạ thở, kéo đa năng gộp nhiều chức năng, dây thoát hiểm khẩn cấp, bình chữa cháy hoặc chăn chống cháy. Thuốc và vật liệu y tế tính riêng: thuốc chống nhiễm trùng và thuốc cảm cúm, khẩu cầu phẫu khoa, vật liệu y tế như gạc và băng, que bông tẩm iodine. Phiên bản mở rộng chia thành 「食品、个人用品、逃生自救工具、医疗急救用品、重要文件资料 5 类应急物资」, tức là 5 nhóm vật tư khẩn cấp, thực phẩm, đồ dùng cá nhân, công cụ thoát hiểm tự cứu, vật dụng cấp cứu y tế, và tài liệu quan trọng. Cục này còn đưa ra ba khuyến nghị. Một là 「选购资质合法、信誉良好的生产经营企业提供的应急物资」, tức là mua vật tư khẩn cấp từ doanh nghiệp sản xuất kinh doanh có giấy phép hợp lệ và uy tín tốt. Hai là 「优先储备基础版的应急物资品种」, tức là ưu tiên tích trữ các món có trong phiên bản cơ bản. Ba là 「熟悉掌握应急物资的正确使用方法，定期对应急物资状况进行检查，并及时更换已过保质期的应急物资」, tức là nắm đúng cách dùng vật tư khẩn cấp, kiểm tra tình trạng vật tư khẩn cấp theo định kỳ, và kịp thời thay vật tư khẩn cấp đã quá hạn. -Nguồn: 北京市应急管理局 (2020). 北京市修订发布居民家庭应急物资储备建议清单. ；北京市西城区人民政府转载的品种明细. +Nguồn: Cục Quản lý Tình huống và Khẩn cấp Bắc Kinh (北京市应急管理局) (2020). 北京市修订发布居民家庭应急物资储备建议清单. ；北京市西城区人民政府转载的品种明细. Ở cấp quốc gia năm 2024 lại cập nhật một lần. Văn bản là Ý kiến chỉ đạo về việc tiếp tục tăng cường hệ thống và năng lực bảo đảm vật tư cho cứu hộ và cứu thảnh, của Văn phòng Ủy ban Quốc gia phòng chống thảm họa, giảm thiểu và cứu hộ (国防减救办发〔2024〕13 号, ngày 23 tháng 9 năm 2024). Phụ lục 2 của văn bản có tên là 家庭应急物资储备指导目录. Danh mục này chia thành 5 nhóm: thực phẩm khẩn cấp, đồ dùng sinh hoạt, công cụ khẩn cấp, thuốc và vật liệu y tế khẩn cấp, và tài liệu quan trọng. Phiên bản cơ bản có 16 món, phiên bản mở rộng có 31 món. TODO (cần đối chiếu: 16 món và 31 món đó tên là gì. Phụ lục chính thức là tệp wps, có thêm một bản PDF scan thành ảnh, lần này không lấy được văn bản nào để đối chiếu từng chữ). @@ -74,7 +74,7 @@ Nguyên văn của Cục Quản lý Khẩn cấp Bắc Kinh là 「定期对应 - **Bộ "túi khẩn cấp" bán sẵn.** Trong loại túi này phần lớn chỉ có đèn pin rẻ và khẩu trang cho có. Mua riêng bình chữa cháy, chăn chữa cháy, mặt nạ thở và túi cứu thương thì rẻ hơn, và cũng dễ kiểm tra chứng nhận từng món hơn - **Tích trữ thực phẩm và nước trên ba ngày.** Danh mục chính thức tính 3 ngày mỗi người. Phần tích thêm, cuối cùng phần lớn sẽ nằm cho tới hạn rồi bỏ đi, xem mục 24 ở phần 5 (không tích trữ vì đợt khuyến mãi) -Nguồn: 全国人大常委会 (2021). 中华人民共和国消防法（2021 修正，第二十四条）. +Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2021). 中华人民共和国消防法（2021 修正，第二十四条）. ## 7. Trang bị này thực sự hữu ích đến đâu @@ -84,6 +84,6 @@ Có một tổng quan hệ thống của Cochrane gộp dữ liệu của 98 ngh Trong cùng tổng quan đó, loại can thiệp này có làm thói quen an toàn của người ta tốt lên. Nhiều hộ hơn có lắp báo động khói hoạt động bình thường, OR 1.81 (95% CI 1.30 đến 2.52), tỷ lệ cao hơn khoảng tám phần. Nhiều người hơn đã tính trước kế hoạch thoát hiểm, OR 2.01 (1.45 đến 2.77), cao hơn khoảng gấp đôi. Nhiều người hơn cất giữ thuốc đúng chỗ, OR 1.53 (1.27 đến 1.84), cao hơn khoảng năm phần. Nhiều hộ hơn lắp cửa chống cháy ở đầu cầu thang, OR 1.61 (1.19 đến 2.17), cao hơn khoảng sáu phần. -Nguồn: Kendrick D 等 (2012). Home safety education and provision of safety equipment for injury prevention. Cochrane Database of Systematic Reviews. +Nguồn: Kendrick D et al. (2012). Home safety education and provision of safety equipment for injury prevention. Cochrane Database of Systematic Reviews. Nên mua trọn danh mục chính thức trong một lần thì tốn không nhiều, cũng coi như chuẩn bị hợp lý. Nhưng đừng nghĩ mua đủ rồi là an toàn. Nghiên cứu lặp đi lặp lại chứng minh ba việc có ích: lắp báo động khói, tính trước kế hoạch để ra ngoài, và dời những món trong nhà dễ gây thương tích đi. Ba việc đó hoặc không có trong danh sách mua, hoặc chỉ chiếm một phần rất nhỏ trong danh sách. diff --git "a/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" "b/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" index fb7e5fa8..9715084d 100644 --- "a/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" +++ "b/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" @@ -88,7 +88,7 @@ Một tổng quan hệ thống đi tìm riêng loại thử nghiệm chia nhóm Nên bây giờ chỉ nói được tới đây: **Khi làm ca đêm, tập trung bữa ăn vào ban ngày càng nhiều càng tốt, có sự hỗ trợ của thử nghiệm chia nhóm ngẫu nhiên, nhưng bằng chứng vẫn mỏng.** Nó không tốn tiền, cũng không khó làm, muốn thử thì thử. Còn một cách thứ hai nữa, ở phần kế tiếp: dùng ánh sáng để chuyển cả chiếc đồng hồ sinh học sang. -Nguồn: Chellappa SL 等 (2021). Daytime eating prevents internal circadian misalignment and glucose intolerance in night work. Science Advances. ；Chellappa SL 等 (2025). Daytime eating during simulated night work mitigates changes in cardiovascular risk factors: secondary analyses of a randomized controlled trial. Nature Communications. ；Wulandari F 等 (2026). Cardiometabolic status among shift workers under meal time regulation: A systematic review of randomized controlled trials. Nutrition and Health. +Nguồn: Chellappa SL et al. (2021). Daytime eating prevents internal circadian misalignment and glucose intolerance in night work. Science Advances. ；Chellappa SL et al. (2025). Daytime eating during simulated night work mitigates changes in cardiovascular risk factors: secondary analyses of a randomized controlled trial. Nature Communications. ；Wulandari F et al. (2026). Cardiometabolic status among shift workers under meal time regulation: A systematic review of randomized controlled trials. Nutrition and Health. ## 8. Có chuyển được chiếc đồng hồ sang bên kia không @@ -128,7 +128,7 @@ Một phân tích tổng hợp gộp 11 nghiên cứu: liệu pháp ánh sáng g Một nghiên cứu phòng thí nghiệm khác dùng ánh sáng mạnh theo chu kỳ. Ở nhóm can thiệp, bài tiết melatonin trùng với khung giờ ngủ đã sắp xếp trong 4.90 giờ, nhóm đối chứng chỉ 2.62 giờ (P=0.002). Sự tụt điểm ở bài kiểm tra nhận thức xuất hiện ở nhóm đối chứng vào đêm ca cuối cùng bị kìm lại ở nhóm can thiệp. Tác giả nói thẳng trong kết luận rằng cách này còn phải kiểm chứng trên công nhân ca đêm thật. -Nguồn: Zhao C, Li N, Miao W, He Y, Lin Y (2025). A systematic review and meta-analysis on light therapy for sleep disorders in shift workers. Scientific Reports. ；Lammers-van der Holst HM 等 (2021). Efficacy of intermittent exposure to bright light for treating maladaptation to night work on a counterclockwise shift work rotation. Scandinavian Journal of Work, Environment & Health. +Nguồn: Zhao C, Li N, Miao W, He Y, Lin Y (2025). A systematic review and meta-analysis on light therapy for sleep disorders in shift workers. Scientific Reports. ；Lammers-van der Holst HM et al. (2021). Efficacy of intermittent exposure to bright light for treating maladaptation to night work on a counterclockwise shift work rotation. Scandinavian Journal of Work, Environment & Health. ### Một việc chưa ai thử From 5292debe79414d41b5852f61c39fa88f7d1553c2 Mon Sep 17 00:00:00 2001 From: j Date: Fri, 2 Oct 2026 18:59:19 +0700 Subject: [PATCH 25/25] =?UTF-8?q?=E8=AF=91=E6=96=87=E8=B7=9F=E4=B8=8A?= =?UTF-8?q?=E4=B8=AD=E6=96=87=E5=8E=9F=E6=96=87=20650=20=E6=9D=A1=EF=BC=9A?= =?UTF-8?q?en/vi=20=E5=85=A8=E4=B9=A6=E5=AF=B9=E9=BD=90=EF=BC=8C=E9=95=BF?= =?UTF-8?q?=E6=96=87=E8=A1=A5=E4=B8=89=E7=AF=87?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 上游 1d42655 合并后中文原文是 650 条，en/vi 停在 641 条。这一个 PR 把两种译文 全部对齐，并把长文那一片的检查补上。 **搬已有的译文（553 条）**：`tools/i18n/rebase.mjs`（新增）。上游删了 9 条、第 2 节 两条并成一条，条号整体顺延，所以不能按条号配——那会把译文挂到隔壁条目上，而且 结构检查不报错（顺延后的号仍在范围内）。这个脚本按条目标题配对，并把译文里的条号 和指路条号一起改成新的。改指路时卡住「这是不是条目引用」：英文的 xref 正则有一条 光秃秃的 `\bitems?\s+(\d+)`，来源栏里「第 59 条」译成 item 59 照样命中，APA 的 item 705 也命中，照单全改就是把法条条款号指到别的条目上去。卡口取旧原文那一条的 指路当白名单。搬不动的（上游删条并条后没跟着改的指路、跨删除点的区间引用、正文真 改了的条目、节标题或导读真改了的节）一节一节点名报出来，不猜。 **重翻 195 条**：新增 18 条、正文真改的 61 条、五节的导读。每条由一个子代理单独 翻，提示词自足、只许改自己那一块的未译条目。CLAUDE.md 的禁令照旧：不许改中文、 不许跑改 git 的命令、不许跑 sync-stats.mjs。 **长文（docs/）三篇新稿**：上游新增了被裁了之后先做什么、孩子出生前后要办的事、 刚确诊慢性病之后，en/docs 和 vi/docs 都没有。三篇全是条目指路，锚点必须取自该语言 book/ 里那条的真实标题（标题是各自分别翻的，和中文锚点常常不是一个说法），115 处 引用逐个查过。en/README.md 和 vi/README.md 各补三处链接（导航行、目录行、结尾 「长文另见」），grep 应为 3，已核对。 **顺带修的两处真错**： - `docs/生物钟和夜班.md` 的「第 2 节第 39 条」在英越两语还写着 item 40。正文这边 rebase 搬了，长文这边没人搬，而 2.40 已经是「买预包装食用油」。已改成 39 和 38。 - `structure.mjs` 的 cnNumbersOf 只认「第+数字+条」这一支，「第十六、十七条」里 的第二个编号读不到，译文按 BRIEF 写成 "Articles 16 and 17" 就被报成「多了原文 没有的数字」。六个条目都栽在这类枚举法条上（第一百七十一、一百七十八、一百七十九 条这种）。改成顿号后继续读。反向验过：原文没有的数字（999）和漏掉的数字照样报 出来，这一层没有被放松。 **新检查 `tools/i18n/check-docs.mjs`，挂进 CI**：check.mjs 只逐条比 book/ 里的条目， 长文一个字都不看，而长文是指路最密的地方。判据取译文引的 (节, 条) 集合跟中文原文 比——不比锚点原文，长文里锚点本来就允许截断（中文自己就在截），逐字比只会报出一堆 假警报；要比的是条号顺延。注入一处错引用验过它会红（2:44），改回来再验它会绿。 **校验**（全绿）：join --all en/vi 各 34 节 0 问题；check.mjs 两种语言 0 处； check-docs.mjs 12 篇 236 处引用 0 篇对不上；check-refs 797 处全带锚点；check-plain 650 条 0 不合格；sync-stats --check 通过；web 的四步 verify（解析一致、构建、 语言下拉落点、渲染一致）全过，三种语言各 34 节 650 条。 已知局限，没在这次改：structure.mjs 的 xref 正则不区分法条引用和条目引用，所以 来源栏里的「第 59 条」在译文里是 "item 59"（页面会把它做成一个指向第 7 节第 59 条 的链接，那一条不存在）。main 上原本就是这样，这次没有改动它，也没有变多。 --- .github/workflows/book.yml | 2 + en/README.md | 10 +- ...15\350\246\201\346\227\251\346\255\273.md" | 85 +++-- ...01\346\205\242\346\205\242\346\255\273.md" | 149 ++++---- ...52\350\264\271\347\262\276\345\212\233.md" | 103 +++--- ...52\350\264\271\346\227\266\351\227\264.md" | 24 +- ...01\346\265\252\350\264\271\351\222\261.md" | 91 ++--- ...15\351\235\242\346\270\205\345\215\225.md" | 95 +++--- ...31\346\200\216\344\271\210\346\264\273.md" | 79 +++-- ...61\346\220\255\350\277\233\345\216\273.md" | 91 ++--- ...25\345\276\213\347\272\242\347\272\277.md" | 28 +- ...22\344\270\215\345\210\222\347\256\227.md" | 61 ++-- ...51\347\232\204\347\272\242\347\272\277.md" | 33 +- ...16\345\201\232\347\224\237\346\204\217.md" | 18 - ...47\346\200\245\346\203\205\345\206\265.md" | 135 ++++---- ...41\346\201\257\345\256\211\345\205\250.md" | 7 - ...77\344\270\216\344\271\260\346\210\277.md" | 7 - ...16\346\200\216\344\271\210\346\264\273.md" | 30 +- ...14\346\234\211\350\200\201\344\272\272.md" | 17 +- ...22\344\270\215\345\210\222\347\256\227.md" | 5 - ...14\345\222\214\345\267\245\344\274\244.md" | 85 ++--- ...20\346\200\216\344\271\210\345\270\246.md" | 19 +- ...03\345\244\226\345\256\211\345\205\250.md" | 16 +- ...16\344\271\210\346\224\276\346\235\276.md" | 29 +- ...00\350\203\275\345\210\222\347\256\227.md" | 30 +- "en/book/24-\347\234\213\347\227\205.md" | 85 ++--- ...01\345\212\236\344\273\200\344\271\210.md" | 11 +- ...31\346\210\226\345\271\263\345\217\260.md" | 12 +- ...25\345\222\214\347\224\237\344\272\247.md" | 17 +- ...53\344\275\223\346\220\236\345\235\217.md" | 9 +- ...23\345\207\273\344\271\213\345\220\216.md" | 35 +- ...16\347\232\204\345\255\251\345\255\220.md" | 38 ++- ...52\345\207\240\346\235\241\350\267\257.md" | 22 +- ...72\345\233\275\347\225\231\345\255\246.md" | 11 +- ...16\346\200\216\344\271\210\346\264\273.md" | 40 +-- ...53\345\220\203\345\207\272\344\272\213.md" | 28 +- ...47\347\227\205\344\271\213\345\220\216.md" | 50 +++ ...01\345\212\236\347\232\204\344\272\213.md" | 68 ++++ ...37\345\222\214\345\244\234\347\217\255.md" | 6 +- ...10\345\201\232\344\273\200\344\271\210.md" | 54 +++ tools/i18n/README.md | 57 ++++ tools/i18n/check-docs.mjs | 74 ++++ tools/i18n/diff-delta.mjs | 164 +++++++++ tools/i18n/diff-scope.mjs | 100 ++++++ tools/i18n/put.mjs | 51 +++ tools/i18n/rebase.mjs | 320 ++++++++++++++++++ tools/i18n/structure.mjs | 20 +- vi/README.md | 10 +- ...15\350\246\201\346\227\251\346\255\273.md" | 117 ++++--- ...01\346\205\242\346\205\242\346\255\273.md" | 157 ++++----- ...52\350\264\271\347\262\276\345\212\233.md" | 106 +++--- ...52\350\264\271\346\227\266\351\227\264.md" | 35 +- ...01\346\265\252\350\264\271\351\222\261.md" | 106 +++--- ...15\351\235\242\346\270\205\345\215\225.md" | 98 +++--- ...31\346\200\216\344\271\210\346\264\273.md" | 85 +++-- ...61\346\220\255\350\277\233\345\216\273.md" | 101 +++--- ...25\345\276\213\347\272\242\347\272\277.md" | 33 +- ...22\344\270\215\345\210\222\347\256\227.md" | 82 ++--- ...51\347\232\204\347\272\242\347\272\277.md" | 42 +-- ...16\345\201\232\347\224\237\346\204\217.md" | 22 +- ...47\346\200\245\346\203\205\345\206\265.md" | 142 ++++---- ...41\346\201\257\345\256\211\345\205\250.md" | 7 - ...77\344\270\216\344\271\260\346\210\277.md" | 7 - ...16\346\200\216\344\271\210\346\264\273.md" | 30 +- ...14\346\234\211\350\200\201\344\272\272.md" | 17 +- ...22\344\270\215\345\210\222\347\256\227.md" | 7 +- ...14\345\222\214\345\267\245\344\274\244.md" | 116 +++---- ...20\346\200\216\344\271\210\345\270\246.md" | 12 +- ...03\345\244\226\345\256\211\345\205\250.md" | 14 +- ...16\344\271\210\346\224\276\346\235\276.md" | 23 +- ...00\350\203\275\345\210\222\347\256\227.md" | 35 +- "vi/book/24-\347\234\213\347\227\205.md" | 87 +++-- ...01\345\212\236\344\273\200\344\271\210.md" | 11 +- ...31\346\210\226\345\271\263\345\217\260.md" | 10 +- ...25\345\222\214\347\224\237\344\272\247.md" | 4 +- ...53\344\275\223\346\220\236\345\235\217.md" | 13 +- ...23\345\207\273\344\271\213\345\220\216.md" | 33 +- ...16\347\232\204\345\255\251\345\255\220.md" | 37 +- ...52\345\207\240\346\235\241\350\267\257.md" | 21 +- ...72\345\233\275\347\225\231\345\255\246.md" | 9 +- ...16\346\200\216\344\271\210\346\264\273.md" | 37 +- ...53\345\220\203\345\207\272\344\272\213.md" | 25 +- ...47\347\227\205\344\271\213\345\220\216.md" | 50 +++ ...01\345\212\236\347\232\204\344\272\213.md" | 68 ++++ ...37\345\222\214\345\244\234\347\217\255.md" | 6 +- ...10\345\201\232\344\273\200\344\271\210.md" | 54 +++ 86 files changed, 2596 insertions(+), 1794 deletions(-) create mode 100644 "en/docs/\345\210\232\347\241\256\350\257\212\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216.md" create mode 100644 "en/docs/\345\255\251\345\255\220\345\207\272\347\224\237\345\211\215\345\220\216\350\246\201\345\212\236\347\232\204\344\272\213.md" create mode 100644 "en/docs/\350\242\253\350\243\201\344\272\206\344\271\213\345\220\216\345\205\210\345\201\232\344\273\200\344\271\210.md" create mode 100644 tools/i18n/check-docs.mjs create mode 100644 tools/i18n/diff-delta.mjs create mode 100644 tools/i18n/diff-scope.mjs create mode 100644 tools/i18n/put.mjs create mode 100644 tools/i18n/rebase.mjs create mode 100644 "vi/docs/\345\210\232\347\241\256\350\257\212\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216.md" create mode 100644 "vi/docs/\345\255\251\345\255\220\345\207\272\347\224\237\345\211\215\345\220\216\350\246\201\345\212\236\347\232\204\344\272\213.md" create mode 100644 "vi/docs/\350\242\253\350\243\201\344\272\206\344\271\213\345\220\216\345\205\210\345\201\232\344\273\200\344\271\210.md" diff --git a/.github/workflows/book.yml b/.github/workflows/book.yml index 6d417a92..c5fcd92d 100644 --- a/.github/workflows/book.yml +++ b/.github/workflows/book.yml @@ -62,6 +62,8 @@ jobs: node-version: 22 - name: 英文与越南文对齐中文原文 run: node tools/i18n/check.mjs + - name: 长文指路的条号节号跟着原文 + run: node tools/i18n/check-docs.mjs - name: 中文文案和模板一致 run: node tools/site/extract.mjs --check diff --git a/en/README.md b/en/README.md index 90d8c681..49f32e1c 100644 --- a/en/README.md +++ b/en/README.md @@ -32,7 +32,7 @@ The AI assistant skill works with Claude Code and Codex. Once it is installed, a Long-form -[Is marriage worth it](docs/结婚划不划算.md) · [Home emergency kit checklist](docs/家庭应急装备清单.md) · [Should you stop when a stranger is in trouble](docs/遇到陌生人出事该不该停.md) · [Which licences a platform needs](docs/做平台要办哪些证.md) · [Body clock and night shifts](docs/生物钟和夜班.md) +[Is marriage worth it](docs/结婚划不划算.md) · [Home emergency kit checklist](docs/家庭应急装备清单.md) · [Should you stop when a stranger is in trouble](docs/遇到陌生人出事该不该停.md) · [Which licences a platform needs](docs/做平台要办哪些证.md) · [Body clock and night shifts](docs/生物钟和夜班.md) · [What to do first after being laid off](docs/被裁了之后先做什么.md) · [What to do before and after a child is born](docs/孩子出生前后要办的事.md) · [After a chronic illness diagnosis](docs/刚确诊慢性病之后.md) Other languages @@ -257,10 +257,10 @@ The text sticks to everyday wording wherever it can, but a few statistical terms 13. [Emergencies: what to do first](book/13-紧急情况.md): cardiac arrest, stroke and posterior circulation stroke, eye stroke, heart attack, aortic dissection, thunderclap headache, chronic subdural haematoma, pulmonary embolism, severe bleeding, bites, burns and scalds, anaphylactic shock, epilepsy, low blood sugar, electric shock, carbon monoxide, swallowed chemicals and chemical burns, objects embedded in the body, immobilising a fracture, scams, threats to privacy, heatstroke, fire, drowning, being lost, hypothermia, snake bites, earthquakes, wild animals, lightning, altitude sickness, ticks, drinking water outdoors. And whether someone can be saved at all: how to help up an elderly person who has fallen, what to do when you walk into a fight, and who pays when helping someone gets you hurt. Measure: survival and money, with the last few items also touching on personal freedom. 14. [Accounts and information security](book/14-账号与信息安全.md): second-factor verification, passwords, the SIM card, a lost phone, card fraud, logged-in devices, app permissions, face recognition, the right to check and to delete. Measure: money / personal information. 15. [Renting and buying a home](book/15-租房与买房.md): deposits, eviction by force, agents holding the money, escrow of funds, a sale does not break a lease, checking the title, a dedicated account for the money, partitioned rooms, and do not buy a rural homestead plot with an urban hukou, only rent a rural house. Measure: money. -16. [Living with a chronic illness](book/16-得了慢性病之后怎么活.md): taking the medicine, out-of-province settlement for chronic and serious outpatient disease, keeping test records, not stopping the medicine to try a folk remedy, long prescriptions, signing up with a family doctor, screening for complications, preventing kidney stones from coming back, and getting gout down to target. Measure: all-cause mortality / money. +16. [Living with a chronic illness](book/16-得了慢性病之后怎么活.md): taking the medicine, out-of-province settlement for chronic and serious outpatient disease, keeping test records, not stopping the medicine to try a folk remedy, long prescriptions, signing up with a family doctor, screening for complications, preventing kidney stones from coming back, and getting gout down to target. Measure: all-cause mortality / money. Long-form essay: [docs/刚确诊慢性病之后.md](docs/刚确诊慢性病之后.md). 17. [When there are elderly at home](book/17-家里有老人.md): voluntary guardianship, forms of will, accounts and the sales pitch, investment and reverse mortgage scams, long-term care insurance, and pressure sore prevention for someone bedridden long term. Measure: money / personal freedom, with the pressure sore item being about mortality. 18. [Is raising children worth it](book/18-养孩子划不划算.md): childcare subsidies, maternity leave and the maternity allowance, the three periods of protection, the time arithmetic, the money arithmetic. Measure: money / time. -19. [Working, leaving, and work injuries](book/19-在职离职和工伤.md): overtime pay, annual leave, probation; notification of occupational hazards and the three medical examinations, dust and noise protection; N, severance in lieu of notice, 2N, do not sign a voluntary resignation, keep your evidence; the time limit for recognising a work injury, an employer with no insurance, assessment of the ability to work, and benefits in case of death. Measure: money. +19. [Working, leaving, and work injuries](book/19-在职离职和工伤.md): overtime pay, annual leave, probation; notification of occupational hazards and the three medical examinations, dust and noise protection; N, severance in lieu of notice, 2N, do not sign a voluntary resignation, keep your evidence; the time limit for recognising a work injury, an employer with no insurance, assessment of the ability to work, and benefits in case of death. Measure: money. Long-form essay: [docs/被裁了之后先做什么.md](docs/被裁了之后先做什么.md). 20. [Caring for a newborn](book/20-刚出生的孩子怎么带.md): safe sleep, the first hepatitis B shot, the immunisation schedule, breast milk and solids, the temperature for made-up formula, honey, vitamin K, the point at which a fever means seeing a doctor, never shaking the baby, nappies and bulk buying, and introducing peanuts early for a high-risk baby to prevent allergy. Measure: infant mortality / money. 21. [Travel and safety abroad](book/21-出国旅行与境外安全.md): travel advisory levels, 12308, the limits of consular protection, overseas medical insurance, high-pay job offers abroad, the annual limit on cash withdrawals abroad, lost documents, overseas driving licences, and agent registration. Measure: money / personal freedom. 22. [Relaxing: venues and stress relief](book/22-怎么放松.md): emergency exits, displayed prices, the lines around drugs, things other people hand you, real-name registration at internet cafés, choosing the site for a murder mystery game; exercise, mindfulness, breathing, company, green space. Measure: money / personal freedom, as well as energy / all-cause mortality. @@ -268,7 +268,7 @@ The text sticks to everyday wording wherever it can, but a few statistical terms 24. [Medical care: spending less and going fewer rounds](book/24-看病.md): the tiered system and referrals, the deductible counting continuously, differences in the reimbursement rate, reserved appointment slots, assessing whether treatment away from home is necessary, keeping and sealing medical records, the four levels of emergency triage, disease emergency assistance when you cannot pay, when to have a disability assessed, how to get a disability certificate, and no red envelopes for doctors. Measure: money / time. 25. [What to do after someone dies](book/25-人走了以后要办什么.md): calling the police and the death certificate, collecting the body and cremation, objections to the cause of death and the autopsy, cancelling the household registration, the list of basic funeral items, price violations, agent registration, the housing provident fund balance and social insurance benefits, and the deceased's personal information rights. Measure: money. 26. [Running a website or platform: licences, filings, servers](book/26-做一个网站或平台.md): payment and settlement red lines, the ICP licence and filing, qualifications for live streaming and audio-visual services, platform verification and tax reporting, content governance, real-name checks, minors, notice and takedown, data export, and choosing a server. Measure: personal freedom / money. Long-form essay: [docs/做平台要办哪些证.md](docs/做平台要办哪些证.md). -27. [Pregnancy and childbirth: from test to paperwork](book/27-怀孕和生产.md): folic acid, the maternal health record and free antenatal checks, screening for the three infections and blocking transmission from mother to child, smoking and drinking in pregnancy, aspirin and gestational diabetes, the signs that mean go to hospital now, what to do when the waters break, epidural analgesia, the indications for a caesarean section, maternity insurance, the medical certificate of birth, newborn screening, insurance and household registration, and the check at 42 days after the birth. Measure: mortality and money. +27. [Pregnancy and childbirth: from test to paperwork](book/27-怀孕和生产.md): folic acid, the maternal health record and free antenatal checks, screening for the three infections and blocking transmission from mother to child, smoking and drinking in pregnancy, aspirin and gestational diabetes, the signs that mean go to hospital now, what to do when the waters break, epidural analgesia, the indications for a caesarean section, maternity insurance, the medical certificate of birth, newborn screening, insurance and household registration, and the check at 42 days after the birth. Measure: mortality and money. Long-form essay: [docs/孩子出生前后要办的事.md](docs/孩子出生前后要办的事.md). 28. [Do not wreck your body for the sake of looks](book/28-别为了外形把身体搞坏.md): extreme dieting and eating disorders, cosmetic clinics and the two certificates the attending physician must hold, facial filler and the parts of the face that go blind, weight-loss products with sibutramine added illegally, anabolic steroids, prescriptions and follow-up for weight-loss drugs and sex hormones, and body image assessment. Measure: mortality (a health outcome), with the two cosmetic medicine items also touching on personal freedom. 29. [After a severe blow](book/29-遭遇重大打击之后.md): the cardiovascular window in the first month after a bereavement, the first week after a diagnosis of serious illness, losing a job, the six months after losing a spouse, bereaved by suicide, what to do when you have neither family nor friends to take the place of that person, children whose parents have died, where to register when grief has stuck, divorce, 12356 and 12355, not making irreversible decisions in the acute stress period, and why using death to settle a debt does not work. Measure: all-cause mortality, with the four items about money and benefits counted under money. 30. [School-age children](book/30-上学以后的孩子.md): emergencies counted by the hour, not delaying treatment for the sake of exams, what to do about bullying, 2 hours outdoors a day, the school medical report, screening for depression in teenagers, products that claim to cure myopia, hard rules on sleep and homework, taking a leave of absence while keeping the student record, dilated refraction and follow-up, and fissure sealants. Measure: mortality and health outcomes, plus one or two items each on money and time. @@ -277,7 +277,7 @@ The text sticks to everyday wording wherever it can, but a few statistical terms 33. [Living after a disability](book/33-残疾之后怎么活.md): the three steps on the spot for autonomic dysreflexia, the suicide window in the ten years after a disabling injury, the principle of voluntary admission for psychiatric care and its two exceptions, the death risk of the carer, pressure-relieving cushions for a wheelchair, "healing" scams, the six benefits to ask about once you have the certificate, long-term care insurance that is not only for the elderly, rehabilitation assistance for ages 0 to 6, grants for making the home accessible, quota employment and the disability employment security fund, the individual income tax reduction, guide dogs and free travel, reasonable accommodation in the national college entrance examination, the duty not to refuse a place and home tuition, the C5 licence, how to choose a rehabilitation provider, hearing aids, assessment of capacity and guardianship. Measure: mortality / money / time / personal freedom. 34. [Do not mishandle the medicines at home](book/34-家里的常备药别吃出事.md): do not double up on paracetamol, no aspirin, nimesulide or metamizole for a child's fever, the groups at high risk of stomach trouble from ibuprofen, no compound cold medicine of your own accord under 2, no ibuprofen of your own accord after 20 weeks of pregnancy, no more than 7 days of omeprazole of your own accord, no antibiotics for a cold, rehydrate first for diarrhoea and no antidiarrhoeal for children, and more painkillers ending in more headache. Measure: mortality. -Within each section the items run from the best value for money to the worst. A section title such as "Do not die young" or "Do not waste time" names the outcome that section is trying to prevent. Whether an item is to be done or not done is settled by the item title. For the long-form essays, see [docs/家庭应急装备清单.md](docs/家庭应急装备清单.md), [docs/做平台要办哪些证.md](docs/做平台要办哪些证.md), [docs/结婚划不划算.md](docs/结婚划不划算.md), [docs/遇到陌生人出事该不该停.md](docs/遇到陌生人出事该不该停.md) and [docs/生物钟和夜班.md](docs/生物钟和夜班.md). The verification trail for every source is recorded in [docs/核实记录](docs/核实记录/). +Within each section the items run from the best value for money to the worst. A section title such as "Do not die young" or "Do not waste time" names the outcome that section is trying to prevent. Whether an item is to be done or not done is settled by the item title. For the long-form essays, see [docs/家庭应急装备清单.md](docs/家庭应急装备清单.md), [docs/做平台要办哪些证.md](docs/做平台要办哪些证.md), [docs/结婚划不划算.md](docs/结婚划不划算.md), [docs/遇到陌生人出事该不该停.md](docs/遇到陌生人出事该不该停.md) and [docs/生物钟和夜班.md](docs/生物钟和夜班.md), [docs/被裁了之后先做什么.md](docs/被裁了之后先做什么.md), [docs/孩子出生前后要办的事.md](docs/孩子出生前后要办的事.md) and [docs/刚确诊慢性病之后.md](docs/刚确诊慢性病之后.md). The verification trail for every source is recorded in [docs/核实记录](docs/核实记录/). The `index.html` in the repository root is the online search page. It filters items by keyword, section, evidence grade and the cost dimensions (money spent, time spent, willpower needed), and it reads its data straight from this file. It becomes available once GitHub Pages is switched on in the repository settings (Deploy from a branch, branch main, directory /). `tools/epub/` holds the ebook build script, and `cd tools/epub && npm ci && npm run build` produces one EPUB in `dist/` locally. GitHub Actions runs the same script after every change to the text and updates the Release. diff --git "a/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" "b/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" index ec5783ea..feeef407 100644 --- "a/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" +++ "b/en/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" @@ -28,14 +28,15 @@ This section covers three things: accidental and violent deaths such as car cras - Evidence grade: B - Sources: Marshall SW et al (1998). Fatal residential fires: who dies and who survives? JAMA. ; USFA (2022). Fatal Fires in Residential Buildings (2018-2020), Topical Fire Report Series 22(2). ; You J et al (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly. - Notes: The reason for the B: on the smoke alarm side there is only one study, and its method was to look back at households that had a fire and households that did not. It was not a randomised trial. There is no study of carbon monoxide alarms lowering mortality on their own. The benefit here is inferred from the pattern that poisoning deaths cluster in winter and in the home. -### 4. Replace gas hoses and stoves when they reach end of life, do not repipe the gas yourself, and refuse gas sales at the door +### 4. Replace gas hoses and cookers when they reach end of life, do not repipe the gas yourself, and turn down the gas company when it comes selling -- Cost: A compliant gas hose costs a few tens of yuan. Replace the stove every few years. Check whether it has passed its expiry date while you are at it. -- In plain language: A compliant hose costs a few tens of yuan. Replacing it when it reaches end of life is a duty of the user written into the Regulations on the Administration of Urban Gas. Repiping the gas yourself or taking out the gas meter is forbidden by the same regulations. Those regulations also say a gas company cannot require you to buy its designated products and services. A salesman at your door saying the hose "must be replaced" can be refused outright. -- Benefit: The regulations make it a duty of the user to "replace in good time gas appliances and connecting pipes that the state has ordered withdrawn or that have reached the end of their service life": once a stove or a hose reaches end of life, or the state withdraws it, you have to replace it. The regulations also list "installing, modifying or removing indoor gas facilities and gas metering devices without authorisation" as a forbidden act, and that covers repiping the gas and taking out the meter yourself. The regulations also state that a gas operator may not require a user to buy its designated products, and may not require a user to accept the services it provides -- Evidence grade: A -- Sources: State Council (国务院) (2010). Regulations on the Administration of Urban Gas, 城镇燃气管理条例（国务院令第 583 号）. 第二十七条: 「燃气用户应当遵守安全用气规则，使用合格的燃气燃烧器具和气瓶，及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管等」 — gas users must follow safe-use rules, use qualified gas appliances and cylinders, and replace in good time any gas appliance or connecting pipe that the state has ordered withdrawn or that has reached the end of its service life. The prohibited acts in 第二十八条 include 「（一）擅自操作公用燃气阀门」「（二）将燃气管道作为负重支架或者接地引线」「（四）擅自安装、改装、拆除户内燃气设施和燃气计量装置」 — operating a public gas valve without authorisation, using a gas pipe as a load-bearing support or as an earthing conductor, and installing, modifying or removing indoor gas facilities and gas metering devices without authorisation. 第二十条 says a gas operator may not 「（六）要求燃气用户购买其指定的产品或者接受其提供的服务」 — require a gas user to buy its designated products or accept the services it provides. 第二十九条 lets a user complain to the gas authority about charges and service, and 「有关部门应当自收到投诉之日起15个工作日内予以处理」 — the department must handle it within 15 working days of receiving the complaint. ; Shenzhen Municipal Housing and Construction Bureau (深圳市住房和建设局) (2024). Gas safety notes, 安全用气小知识（二）: 「立即关闭家里的燃气总阀门」「迅速开门、开窗进行通风」「一定不要开、关家里的电灯、抽油烟机等电器」「人员应该迅速撤离，转移到安全的地方，并且及时拨打燃气公司的报警电话」 — close the main gas valve at home, open the door and windows to ventilate, never switch the lights or the cooker hood on or off, and move people out to a safe place and call the gas company's emergency number. ; Zhangjiagang Municipal Emergency Management Bureau (张家港市应急管理局) (2024). Gas safety knowledge, 燃气安全使用知识: 「不要情急之下打开排气扇进行通风」「应远离现场之后才能打电话求助」 — do not open the exhaust fan in a hurry, and call for help only after you are clear of the scene. -- Notes: Some people come to the door in the name of a "safety inspection" and then sell a self-closing valve and an alarm. The regulations bar a gas company from making you buy its designated products. You can refuse, and you can also complain. If you smell gas, close the main gas valve first. For bottled gas, close the valve on the cylinder. Then open the door and windows to ventilate, and do not use the exhaust fan. Do not switch lights, the cooker hood or a lighter on or off. Then get everyone out and call the gas company's emergency number from a safe distance. Carbon monoxide poisoning is a separate matter, see section 13. +- Cost: A compliant gas hose costs a few tens of yuan. Replace the cooker every few years. Check whether it has passed its expiry date while you are at it. +- In plain language: A compliant hose costs a few tens of yuan. Replacing it at the end of its life is a duty of the user written into the Regulations on the Administration of Urban Gas. Repiping the gas yourself or taking out the gas meter is forbidden by the same regulations. Those regulations also say a gas company cannot require you to buy its designated products and services. A salesman at your door saying the hose "must be replaced" can be refused outright. +- Benefit: The regulations make it a duty of the user to "replace in good time gas appliances and connecting pipes that the state has ordered withdrawn or that have reached the end of their service life": once a cooker or a hose reaches end of life, or the state withdraws it, you have to replace it. The regulations also list "installing, modifying or removing indoor gas facilities and gas metering devices without authorisation" as a forbidden act, and that covers repiping the gas and taking out the meter yourself. The regulations also state that a gas operator may not require a user to buy the products it designates, and may not require a user to accept the services it provides +- Evidence grade: C +- Sources: State Council (国务院) (2010). Regulations on the Administration of Urban Gas, 城镇燃气管理条例（国务院令第 583 号）. Article 27: 「燃气用户应当遵守安全用气规则，使用合格的燃气燃烧器具和气瓶，及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管等」 — gas users must follow safe-use rules, use qualified gas appliances and cylinders, and replace in good time any gas appliance or connecting pipe that the state has ordered withdrawn or that has reached the end of its service life. The prohibited acts in Article 28 include 「（一）擅自操作公用燃气阀门」「（二）将燃气管道作为负重支架或者接地引线」「（四）擅自安装、改装、拆除户内燃气设施和燃气计量装置」 — operating a public gas valve without authorisation, using a gas pipe as a load-bearing support or as an earthing conductor, and installing, modifying or removing indoor gas facilities and gas metering devices without authorisation. Article 20 says a gas operator may not 「（六）要求燃气用户购买其指定的产品或者接受其提供的服务」 — require a gas user to buy its designated products or accept the services it provides. Article 29 lets a user complain to the gas authority about charges and service, and 「有关部门应当自收到投诉之日起15个工作日内予以处理」 — the department must handle it within 15 working days of receiving the complaint. ; Shenzhen Municipal Housing and Construction Bureau (深圳市住房和建设局) (2024). Gas safety notes, 安全用气小知识（二）: 「立即关闭家里的燃气总阀门」「迅速开门、开窗进行通风」「一定不要开、关家里的电灯、抽油烟机等电器」「人员应该迅速撤离，转移到安全的地方，并且及时拨打燃气公司的报警电话」 — close the main gas valve at home, open the door and windows to ventilate, never switch the lights or the cooker hood on or off, and move people out to a safe place and call the gas company's emergency number. ; Zhangjiagang Municipal Emergency Management Bureau (张家港市应急管理局) (2024). Gas safety knowledge, 燃气安全使用知识: 「不要情急之下打开排气扇进行通风」「应远离现场之后才能打电话求助」 — do not open the exhaust fan in a hurry, and call for help only after you are clear of the scene. +- Notes: The text of the regulation can be checked word for word, but it only says that replacing the hose on time is your duty. It does not say how many fewer accidents you get that way. No official figure was found for how much replacing hoses on a schedule cuts gas accidents, so this is graded C only. Some people come to the door in the name of a "safety inspection" and, once they have finished looking, sell a self-closing valve and an alarm. Article 20, paragraph (6) bars a gas company from making you buy the products it designates. You can refuse, and you can also complain. If you smell gas, close the main gas valve first, and for bottled gas close the valve on the cylinder. Then open the door and windows to ventilate, and do not open the exhaust fan. Do not switch lights, the cooker hood and other appliances on or off, and do not touch a lighter. Then get everyone out, leave the scene, and only then call the gas company's emergency number. Carbon monoxide poisoning is another matter, see section 13. + ### 5. Do not pick, buy or eat wild mushrooms; no local way of telling poison works - Cost: No money. The hard part is giving up a seasonal wild food. @@ -61,14 +62,15 @@ This section covers three things: accidental and violent deaths such as car cras - Evidence grade: A - Sources: Ettehad D et al (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Lu J et al (2017). Prevalence, awareness, treatment, and control of hypertension in China (China PEACE Million Persons Project). Lancet. - Notes: Measure your blood pressure once first. Of every 10 people in China with high blood pressure, fewer than 1 is under control, so that single reading is one of the highest-value actions in this book. What counts as target (130 or 140) is still argued over, but "knowing you have high blood pressure and bringing it down" is not disputed. -### 8. After 35, if you are overweight, get one fasting blood glucose test, and repeat it every three years even if it is normal +### 8. After 35, if you are overweight, get one fasting glucose test, and repeat it every three years even if it is normal - Cost: A fasting glucose test costs a dozen-odd yuan. A glycated haemoglobin test costs a few tens of yuan. One trip to the hospital to draw blood is all it takes. -- In plain language: Drawing one tube of blood for a few tens of yuan can find pre-diabetes while you have no symptoms at all. The official United States screening advice is that people aged 35 to 70 who are overweight or obese should be tested. If the result is normal, test again every three years. -- Benefit: The US Preventive Services Task Force is the body that sets official screening advice in the United States. Its current recommendation is to screen adults aged 35 to 70 who are overweight or obese for pre-diabetes and type 2 diabetes, even when they have no symptoms at all. Overweight means a body mass index ≥25, and obese means ≥30. That recommendation carries a grade B. It also states: "for adults with normal blood glucose, screening every 3 years is reasonable" +- In plain language: Drawing one tube of blood for a few tens of yuan can find prediabetes while you have no symptoms at all. The official United States screening advice is this: people aged 35 to 70 who are overweight or obese should be tested. If the result is normal, test again every three years. In the United States overweight starts at a body mass index of 25, and the line comes down to 23 for people of Asian descent; the Chinese standard counts 24 as overweight. +- Benefit: The US Preventive Services Task Force is the body that sets official screening advice in the United States. Its current recommendation is to screen adults aged 35 to 70 who are overweight or obese for prediabetes and type 2 diabetes, even when they have no symptoms at all. Overweight means a body mass index ≥25, and obese means ≥30. For Asian Americans it recommends bringing the line down to a body mass index ≥23, because at the same body mass index people of Asian descent have a higher risk of diabetes. The Chinese industry standard counts a body mass index of 24.0 to 28.0 as overweight and ≥28.0 as obese. That recommendation carries a grade B. It also states: "for adults with normal blood glucose, screening every 3 years is reasonable" - Evidence grade: A -- Sources: US Preventive Services Task Force (2021). Screening for Prediabetes and Type 2 Diabetes. -- Notes: Finding pre-diabetes does not mean you have to take drugs. A good share of people can reverse it with lifestyle changes alone. Among Chinese adults only a minority have ever known they had diabetes. Many find out only once complications appear +- Sources: US Preventive Services Task Force (2021). Screening for Prediabetes and Type 2 Diabetes. ：「at a lower BMI (≥23) if the patient is Asian American」 ; General Office of the National Health Commission (国家卫生健康委办公厅) (2024). Explanation of the Core Knowledge on Managing Residents' Weight, 2024 edition, 居民体重管理核心知识（2024 年版）释义: 「根据《成年人体重判定》（WS/T 428-2013）行业标准……24.0≤BMI＜28.0 为超重，BMI≥28.0 为肥胖」. +- Notes: Finding prediabetes does not mean you have to take drugs. A good share of people can reverse it with lifestyle changes alone. Among Chinese adults only a minority have ever known they had diabetes. Many find out only once complications appear + ### 9. Do not speed, and do not drink and drive - Cost: No money. Driving a little slower takes a few extra minutes on each trip. The hard part is drinking less at a dinner. @@ -93,23 +95,25 @@ This section covers three things: accidental and violent deaths such as car cras - Evidence grade: B - Sources: Spiegel CN, Lindaman FC (1977). Children can't fly: a program to prevent childhood morbidity and mortality from window falls. American Journal of Public Health: "Significant reduction in falls resulted, particularly in the Bronx, where reported falls declined 50 percent from 1973 to 1975"; the 1976 amendment to the New York City health code "to require that landlords provide window guards in apartments where children ten years old and younger reside". ; (2021). Unintentional Window Falls in Children and Adolescents. Academic Pediatrics: 38,840 emergency department visits recorded by the national electronic injury surveillance system from month 1 of 2007 to month 8 of 2017, "The majority of falls occurred in children under the age of 6 and were related to falls from a second story or below". - Notes: Do not put a bed, a sofa, or boxes where a child can climb up next to a window or a balcony. A window screen only keeps out mosquitoes. It comes off with one push, so do not treat it as a guard. High-rise living is not the only reason to fit one. In the US data, most falls happened on the second floor or lower. -### 12. Keep children in sight near water; life jackets for boating and open-water swimming + +### 12. Keep children in sight whenever they are near water; wear a life jacket when boating and swimming in open water - Cost: A life jacket costs 50 to 200 yuan. It also takes the attention you spend watching the child. - In plain language: A child in a life jacket who falls in the water has about half the chance of drowning. Drowning is still the leading cause of death for Chinese children aged 1 to 14. The rural drowning mortality rate is about twice the urban rate. - Benefit: A study using United States Coast Guard data matched comparable people in pairs. Among recreational boaters who fell in the water, those wearing a life jacket had about half the risk of drowning death. The adjusted value is RR 0.51, credible range 95% CI 0.35 to 0.74. In China, drowning mortality under age 20 fell from 6.60 per 100,000 people in 2013 to 3.28 per 100,000 in 2021. The rural rate is about 2 times the urban rate. Drowning is still the leading cause of death for children aged 1 to 14. In 2021, drowning caused 31.1% of injury deaths among people aged 0 to 19, and road traffic caused 27.9% - Evidence grade: A - Sources: Cummings P et al (2011). Association between wearing a personal floatation device and death by drowning among recreational boaters. Injury Prevention. ; Li Z et al (2023). Unintentional Drowning Mortality Among Individuals Under Age 20 — China, 2013–2021. China CDC Weekly. ; Zhou J et al (2024). Injury Mortality of Children and Adolescents Aged 0–19 Years — China, 2010–2021. China CDC Weekly. -- Notes: The grade A covers only the life jacket. Keeping children in sight is a general consensus. On its own it is only grade C, with no randomised trial, and there will never be one. In China, drowning deaths cluster in rural areas, in summer, and among males aged 15 to 19. +- Notes: The grade A covers only the life jacket. "Keeping children in sight" is a general consensus. On its own it is only grade C, with no randomised trial, and there will never be one. In China, drowning deaths cluster in rural areas, in summer, and among males aged 15 to 19. Going onto the ice to play in winter: see item 41 (bore a hole and measure the thickness before you step onto outdoor ice). ### 13. Over 60, train balance and leg strength; fix up the bathroom and stairs at home - Cost: Train 2 to 3 times a week, 30 minutes of tai chi or balance work each time. A non-slip mat, grab bars, and a night light in the bathroom cost a few tens to a few hundred yuan. The hard part is keeping it up for years. -- In plain language: Balance training such as tai chi two or three times a week cuts falls among older people by about 23%. The number of people who have a fall drops 15%. Non-slip surfaces and grab bars at home cut falls by another 20%. Falls are the leading cause of injury death among Chinese people over 65. More than half of them happen in the person's own home. -- Benefit: Cochrane pooled 108 trials with 23,407 people. Exercise reduced the number of falls among older people by 23%. The rate ratio for falls is RaR 0.77, credible range 95% CI 0.71 to 0.83. The number of people with a fall dropped 15% (RR 0.85). A safety assessment and home modifications reduced the fall rate by about a fifth (RR 0.81, 95% CI 0.68 to 0.97). Tai chi reduced fall risk by about 29% (RR 0.71). In China, falls are the leading cause of injury death among people over 65. Of the older people's falls monitored in 2018, 55.97% happened in the home +- In plain language: Balance training such as tai chi two or three times a week cuts falls among older people by about 23%. The number of people who have a fall drops 15%. Non-slip surfaces and grab bars at home cut falls by another two tenths. Falls are the leading cause of injury death among Chinese people over 65. More than half of them happen in the person's own home. +- Benefit: Cochrane pooled 108 trials with 23,407 people. Exercise reduced the number of falls among older people by 23%. The rate ratio for falls is RaR 0.77, credible range 95% CI 0.71 to 0.83. The number of people with a fall dropped 15% (RR 0.85). Having someone come to the home for a safety assessment and modifications reduced the fall rate by about two tenths (RR 0.81, 95% CI 0.68 to 0.97). Tai chi reduced fall risk by about 29% (RR 0.71). In China, falls are the leading cause of injury death among people over 65. Of the older people's falls monitored in 2018, 55.97% happened in the home - Evidence grade: A - Sources: Sherrington C et al (2019). Exercise for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Gillespie LD et al (2012). Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Lu Z et al (2021). Characteristics of Falls Among Older People — China, 2018. China CDC Weekly. -- Notes: This applies to people over 60 who live in their own home and not in a nursing home. The studies above count falls, not deaths. For an older person the road from a fall to a hip fracture to death is short. +- Notes: This applies to people over 60 who live in their own home and are not in a nursing home. The studies above count falls, not deaths. For an older person the road from a fall to a hip fracture and then to death is short. Not being able to see the road also means more falls; for how cataracts affect what you can see, see item 9 in section 17 (cataract surgery). Whether the bones can take a fall: see item 39 in this section (bone density test). + ### 14. Get tested for hepatitis B; if you have no antibodies, get vaccinated - Cost: One test costs a few tens of yuan. The 3 doses of hepatitis B vaccine an adult needs cost 100 to 300 yuan in total. Finish them within half a year. @@ -142,6 +146,7 @@ This section covers three things: accidental and violent deaths such as car cras - Evidence grade: A - Sources: US Preventive Services Task Force (2024). Breast Cancer: Screening. - Notes: The starting age differs by country. Some Chinese guidelines and local programmes start at 45. A parent, a child, or a sibling with breast cancer, and anyone carrying a BRCA mutation, is high risk. Those cases need a plan from a doctor, and this item does not apply to them + ### 18. Women over 30 should be screened for cervical cancer, with HPV testing first - Cost: An HPV test costs 100 to 300 yuan. If the result is negative, one test every 5 years is enough. The swab takes a few minutes. @@ -150,7 +155,6 @@ This section covers three things: accidental and violent deaths such as car cras - Evidence grade: A - Sources: Sankaranarayanan R et al (2009). HPV screening for cervical cancer in rural India. New England Journal of Medicine. - Notes: Women who have had the HPV vaccine still need screening, because the vaccine does not cover all the virus types. What makes this study unusual is that it is a randomised screening trial. It measured deaths directly, and it screened only one round. - ### 19. Start colorectal cancer screening at 45 to 50, with a faecal immunochemical test or a colonoscopy - Cost: A faecal test costs a few tens of yuan, every 1 to 2 years. A colonoscopy costs a few hundred yuan to a little over a thousand, and one every 10 years is enough if the result is negative. Set aside a day for bowel preparation before a colonoscopy. @@ -199,6 +203,7 @@ This section covers three things: accidental and violent deaths such as car cras - Evidence grade: A - Sources: National Lung Screening Trial Research Team (2011). Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. New England Journal of Medicine. - Notes: This applies only to high-risk smokers. In the same trial, 24.2% of the people who had the low-dose CT had a suspicious finding. Of those, 96.4% turned out to be false alarms (a false positive). For people who do not smoke, this test is paying money to feel anxious. + ### 25. Call 12356 when you are depressed or have suicidal thoughts, and keep no sleeping pills or pesticides at home - Cost: It costs nothing. One phone call. Also lock the medicines away and keep pesticides out of the house. @@ -207,7 +212,6 @@ This section covers three things: accidental and violent deaths such as car cras - Evidence grade: B - Sources: Zalsman G et al (2016). Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry. ; National Health Commission (国家卫生健康委) (2024). Notice on the Use of "12356" as the Nationwide Unified Psychological Assistance Hotline Number, 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. - Notes: The reason for the grade B: most of the evidence on restricting means comes from ecological studies, which compare the numbers before and after a policy. The hotline itself has no evidence of lowering mortality. The urge to kill yourself usually lasts minutes to hours. Moving lethal things one step further away is itself an intervention. For the time scale of that urge and for the long-term outcome after an attempt, see item 32 in this section (tell one person near you the moment the thought appears). For what is left behind after someone is rescued, see item 33 in this section. - ### 26. Keep a fire extinguisher, a fire blanket, escape respirators and a first aid kit, and check them once a year - Cost: A fire extinguisher, a fire blanket, one respirator per person and a first aid kit come to a few hundred yuan as a set. Buy them once, then spend ten minutes checking them every year. @@ -238,8 +242,9 @@ This section covers three things: accidental and violent deaths such as car cras - In plain language: Controlling the risk factors for cardiovascular disease improves erectile function itself. On a questionnaire score, the average gain is 2 to 3 points. The questionnaire has 25 points in total, and 22 or more counts as normal. - Benefit: A meta-analysis pooled 6 randomised controlled trials with 740 men. The interventions were lifestyle change, or drugs to control cardiovascular risk factors, with at least 6 weeks of follow-up. Here is the result: the score on the erectile function questionnaire (IIEF-5) rose by 2.66 points on average (95% CI 1.86 to 3.47, and that is the credible range). Taking out the 2 trials that used statins (143 people) leaves 4 trials based on lifestyle change alone (597 people), and they still gained 2.40 points (1.19 to 3.61). The IIEF-5 is a questionnaire of 5 questions, worth 25 points in total. The paper splits the scores into five bands: 22 to 25 points counts as no difficulty, and 5 to 7 points counts as severe - Evidence grade: A -- Sources: Gupta BP, Murad MH, Clifton MM, Prokop L, Nehra A, Kopecky SL (2011). The effect of lifestyle modification and cardiovascular risk factor reduction on erectile dysfunction: a systematic review and meta-analysis. Archives of Internal Medicine, 171(20), 1797-1803. ；Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM (1999). Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. International Journal of Impotence Research, 11(6), 319-326. -- Notes: The interventions in these 6 trials differ. Some combined weight loss with exercise, and others used drugs to control blood pressure and lipids. There were only 740 people in total, so the direction is clear but the size of the gain differs from person to person. The benefit is graded "small" because the only endpoint is a questionnaire score. There is no figure for how many illnesses people avoided later. The same three actions do have hard endpoints elsewhere: for quitting smoking see item 1 in section 2 (quitting smoking, the earlier the better), for weight loss see item 33 in section 2 (keep BMI between 20–25), and for blood pressure see item 7 in this section (measure it, and take drugs if it is high). Taking the drug is not forbidden, as long as a doctor prescribed it. "Health products" sold online often sneak these substances in at an unknown dose. How to tell them apart: see item 4 in section 28 (do not buy diet pills that promise fast weight loss). +- Sources: Gupta BP, Murad MH, Clifton MM, Prokop L, Nehra A, Kopecky SL (2011). The effect of lifestyle modification and cardiovascular risk factor reduction on erectile dysfunction: a systematic review and meta-analysis. Archives of Internal Medicine, 171(20), 1797-1803. ; Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM (1999). Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. International Journal of Impotence Research, 11(6), 319-326. +- Notes: The interventions in these 6 trials differ. Some combined weight loss with exercise, and others used drugs to control blood pressure and lipids. There were only 740 people in total, so the direction is clear but the size of the gain differs from person to person. The benefit is graded "small" because the only endpoint is a questionnaire score. There is no figure for how many illnesses people avoided later. The same three actions do have hard endpoints elsewhere: for quitting smoking see item 1 in section 2 (quitting smoking, the earlier the better), for weight loss see item 32 in section 2 (keep BMI between 20–25), and for blood pressure see item 7 in this section (measure it, and take drugs if it is high). Taking the drug is not forbidden, as long as a doctor prescribed it. "Health products" sold online often sneak these substances in at an unknown dose. How to tell them apart: see item 4 in section 28 (do not buy diet pills that promise fast weight loss). + ### 30. Use a condom for the whole of sex, and never share needles @@ -252,11 +257,12 @@ This section covers three things: accidental and violent deaths such as car cras ### 31. If you have done anything high-risk, get an HIV test once: the CDC does it free and the result stays confidential - Cost: It costs nothing. One trip to the CDC or a designated medical institution, one tube of blood. The hard part is asking for the test. -- In plain language: An HIV test at the CDC costs nothing. Under the Measures for the Administration of HIV Testing Nationwide, staff must not disclose your name, your address or your result. A positive result is not a death sentence. People who keep taking the medicine and get the virus down to an undetectable level do not pass it on to their partners. That study followed more than 700 couples and more than 70,000 condomless sex acts, and transmission between partners was zero. -- Benefit: Article twenty-three of the Regulations on the Prevention and Control of AIDS says that the health departments of local people's governments at county level and above will designate medical and health institutions. Those institutions "shall provide free counselling and preliminary screening tests to people who voluntarily seek HIV counselling and testing". Testing early does real work now. Current antiretroviral treatment can push the virus down to undetectable. The World Health Organization's line is that people on antiretroviral treatment with an undetectable viral load will not transmit HIV to their sexual partners. The PARTNER study followed 782 same-sex couples, and in every couple one partner was infected and one was not, which is called serodiscordant. The total is 1593 couple-years, which means the follow-up years of each couple added together. There were 76088 condomless sex acts during that time, and 15 new infections. After a phylogenetic analysis comparing the virus genes, all 15 of them did not come from the partner, and transmission between partners was zero. The Measures for the Administration of HIV Testing Nationwide also states that staff in HIV testing laboratories must not disclose the names, addresses or test results of people with AIDS or people who carry the virus +- In plain language: An HIV test at the CDC costs nothing. Without your own consent, no unit and no individual may disclose your name, your address, your work unit or your medical history. A positive result is not a death sentence. People who keep taking the medicine and get the virus down to an undetectable level do not pass it on to their sexual partners. That study followed more than 700 couples and more than 70,000 condomless sex acts, and transmission between partners was zero. +- Benefit: Article 23 of the Regulations on the Prevention and Control of HIV/AIDS says that the health departments of local people's governments at county level and above will designate medical and health institutions. Those institutions "shall provide free counselling and preliminary screening tests to people who voluntarily seek HIV counselling and testing". Testing early does real work now. Current antiretroviral treatment can push the virus down to undetectable. The World Health Organization's line is that people on antiretroviral treatment with an undetectable viral load will not transmit HIV to their sexual partners. The PARTNER study followed 782 same-sex couples, and in every couple one partner was infected and one was not, which is called serodifferent. The total is 1593 couple-years, which means the follow-up years of each couple added together. There were 76088 condomless sex acts during that time, and 15 new infections. After a phylogenetic analysis comparing the virus genes, all 15 of them did not come from the partner, and transmission between partners was zero. Article 39 of the same regulations states: 「未经本人或者其监护人同意，任何单位或者个人不得公开艾滋病病毒感染者、艾滋病病人及其家属的姓名、住址、工作单位、肖像、病史资料以及其他可能推断出其具体身份的信息。」 - Evidence grade: A -- Sources: Ministry of Health (卫生部) (2006). Measures for the Administration of HIV Testing Nationwide, 全国艾滋病检测工作管理办法. ; State Council (国务院) (2006). Regulations on the Prevention and Control of AIDS, 艾滋病防治条例（第三、二十三、三十八条）. (reprinted by the Beijing Municipal Health Commission) ; Rodger AJ, Cambiano V, Bruun T, et al. (2019). Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. The Lancet, 393(10189), 2428-2438. ; Guangdong CDC (广东省疾病预防控制中心). HIV testing window period, a plain-language explainer, 艾滋病检测窗口期科普. (the figures for the window period) -- Notes: The window period decides when a test counts. The figures from the Guangdong CDC are that a nucleic acid test can detect it after about 1 week, a fourth-generation combined antigen and antibody test after about 2 weeks, and a third-generation antibody test after about 3 weeks. Wait at least 2 weeks before testing, and repeat it every 2 to 4 weeks. Three months after the last high-risk behaviour, more than 99.99% of people can rule out infection. A home test kit showing positive is not a diagnosis, and it has to be confirmed at the CDC or a proper institution. Article three of the Regulations says that no unit or individual may discriminate against infected people or their families, and that marriage, employment, medical care and schooling are protected by law. Article thirty-eight says at the same time that infected people are obliged to tell their sexual partners the fact of the infection or the illness in good time. Screening for the three infections in pregnancy and blocking transmission from mother to baby: see section 27 +- Sources: State Council (国务院) (2006). Regulations on the Prevention and Control of HIV/AIDS, 艾滋病防治条例（第三、二十三、三十八、三十九条）. (reprinted by the Beijing Municipal Health Commission) ; Rodger AJ, Cambiano V, Bruun T, et al. (2019). Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. The Lancet, 393(10189), 2428-2438. ; Guangdong Provincial Center for Disease Control and Prevention (广东省疾病预防控制中心). The window period of HIV testing, a plain-language explainer, 艾滋病检测窗口期科普. (the figure for the window period; the original page has been taken down, and the link is a web archive copy dated 2026-09-15) +- Notes: The window period decides when a test counts. The figures from the Guangdong CDC are that a nucleic acid test can detect it after about 1 week, a fourth-generation combined antigen and antibody test after about 2 weeks, and a third-generation antibody test after about 3 weeks. Wait at least 2 weeks before testing, and repeat it every 2 to 4 weeks. Three months after the last high-risk behaviour, more than 99.99% of people can rule out infection. A home test kit showing positive is not a diagnosis, and it has to be confirmed at the CDC or a proper institution. Article 3 of the Regulations says that no unit and no individual may discriminate against infected people or their families, and that marriage, employment, medical care and schooling are protected by law. Article 38 says at the same time that infected people are obliged to tell their sexual partners the fact of the infection or the illness in good time. Screening for the three infections in pregnancy and blocking transmission from mother to baby: see section 27 + ### 32. Tell one person near you the moment a suicidal thought appears, and hand those few minutes to someone else - Cost: It costs nothing. The hard part is speaking up at the time, and admitting that you cannot stay on your own for the next few hours. @@ -264,7 +270,8 @@ This section covers three things: accidental and violent deaths such as car cras - Benefit: 82 people who had attempted suicide were interviewed within 3 days of admission. 47.6% of them (39 people) said that no more than 10 minutes passed from the first suicidal thought of that episode to the act itself. The people whose process ran longer scored clearly higher on the suicidal intent scale. The score on the impulsivity scale had nothing to do with how long the process took. 76.8% (63 people) said they had some contact with a partner, a family member or a friend during that time. The authors conclude that the window for intervention is usually short and that there is little room for professionals to step in at that stage, so the warning signs and the ways to intervene have to be spread to ordinary people. A separate systematic review pooled 90 follow-up studies. Among people who came to hospital after self-harm, the middle value for self-harm again without dying within 1 year was 16%. Death by suicide within 1 year was 2%. Among those followed for 9 years or more, about 7% died by suicide - Evidence grade: B - Sources: Deisenhammer EA, Ing CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry, 70(1), 19-24. ; Owens D, Horrocks J, House A (2002). Fatal and non-fatal repetition of self-harm: systematic review. British Journal of Psychiatry, 181, 193-199. -- Notes: The reason for the grade B: the Deisenhammer study had only 82 people, came from a single hospital, and relied on the survivors' own recall afterwards. The Owens review gives the middle value from 90 studies that each did things differently, not one number pooled from them. The same review also states plainly that the suicide risk of people who have harmed themselves is still several hundred times that of the general population. So "the large majority did not later die by suicide" describes this group as a whole. It is not a guarantee about any one person, and the doctor still has to be seen. Moving lethal means out of reach, and 12356: see item 25 in this section. What to do first when you feel low: see item 19 in section 3. What is left behind after someone is rescued: see item 33 in this section. What you can do when someone near you shows this kind of thought: see item 15 in section 8 +- Notes: The reason for the grade B: the Deisenhammer study had only 82 people, came from a single hospital, and relied on the survivors' own recall afterwards. The Owens review gives the middle value from 90 studies that each did things differently, not one number pooled from them. The same review also states plainly that the suicide risk of people who have harmed themselves is still several hundred times that of the general population. So "the large majority did not later die by suicide" describes this group as a whole. It is not a guarantee about any one person, and the doctor still has to be seen. Moving lethal means out of reach, and 12356: see item 25 in this section. What to do first when you feel low: see item 18 in section 3. What is left behind after someone is rescued: see item 33 in this section. What you can do when someone near you shows this kind of thought: see item 15 in section 8 + ### 33. Do not treat surviving as a safety net: after drinking pesticide or inhaling coal gas, emergency care can save the life but not the lungs or the brain @@ -315,3 +322,29 @@ This section covers three things: accidental and violent deaths such as car cras - Evidence grade: A - Sources: Grant RM, Lama JR, Anderson PL, et al. (2010). Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. New England Journal of Medicine, 363(27), 2587–2599. ; McCormack S, Dunn DT, Desai M, et al. (2016). Pre-exposure prophylaxis to prevent the acquisition of HIV-1 infection (PROUD): effectiveness results from the pilot phase of a pragmatic open-label randomised trial. The Lancet, 387(10013), 53–60. ; Molina JM, Capitant C, Spire B, et al. (2015). On-demand preexposure prophylaxis in men at high risk for HIV-1 infection. New England Journal of Medicine, 373(23), 2237–2246. ; Fonner VA, Dalglish SL, Kennedy CE, et al. (2016). Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS, 30(12), 1973–1983. ; Xu JJ, Huang XJ, Liu XC, et al. (2020). Consensus statement on human immunodeficiency virus pre-exposure prophylaxis in China. Chinese Medical Journal, 133(23), 2840–2846. (Chinese version, 《中国 HIV 暴露前预防用药专家共识》) ; World Health Organization (2019). What's the 2+1+1? Event-driven oral pre-exposure prophylaxis to prevent HIV for men who have sex with men (WHO/CDS/HIV/19.8). ; General Office of the State Council (国务院办公厅) (2024). Plan for Containing and Controlling AIDS in China (2024–2030), 中国遏制与防治艾滋病规划（2024—2030 年）. - Notes: The people in the trials were mainly men who have sex with men, so that is the group that should ask. Anyone else with high-risk sex can ask the doctor whether it suits them. Before it is prescribed you must confirm that you have not been infected. Taking it when you are already infected makes the virus resistant, and a self-test strip result is not a basis you can act on. Stopping the drug in someone with hepatitis B can bring on an attack of hepatitis, so that test comes first. The World Health Organization recommends on-demand use for men who have sex with men only, and it fits people averaging fewer than 2 times a week. The way to take it is 2 tablets some time between 2 and 24 hours before sex, then 1 tablet at 24 hours and 1 tablet at 48 hours. Women, and people with hepatitis B, can only take it every day. In the on-demand trials, the group taking the drug had more gastrointestinal and kidney side effects than the placebo group, so the check-ups cannot be skipped. China began a pilot programme among men who have sex with men in 2019, and the 2024 plan says "standardised implementation". No nationwide free policy could be found. Get it from the infectious disease department or the sexually transmitted infection and AIDS clinic of a local designated AIDS hospital, and call the local CDC first to ask which one can prescribe. It does not protect against syphilis, gonorrhoea and the like, so you still need a condom: see item 30 in this section (use a condom throughout). If high-risk sex has already happened, the remedy is item 38 in section 13 (get post-exposure prophylaxis within 72 hours). The people who benefit are you mainly, and it protects your partner as well. + +### 39. Women over 65 should get one dual-energy X-ray bone density test (DXA), and women past menopause with osteoporosis risk factors do not have to wait until 65 + +- Cost: One dual-energy X-ray bone density test in hospital, shortened to DXA, covering the lumbar spine and the hip. You lie down for a few minutes and it is done. Out of pocket it is generally one or two hundred yuan, and the price varies by place. The heel ultrasound you get at a checkup by standing on the machine can only screen, not diagnose. +- In plain language: Women over 65, and women past menopause who have osteoporosis risk factors, should have one bone density test. Assess the risk first and give the drug to the high-risk women, and hip fractures fall by about 17%. Worked out per head, among every 1000 women tested about 5 fewer have a hip fracture. All-cause mortality showed no change. +- Benefit: The recommendation the US Preventive Services Task Force issued in 2025 is this: women aged 65 and over should be screened for osteoporosis (a grade B recommendation). Women past menopause, under 65, but assessed as at high risk of fracture are also recommended for screening (grade B). For men the evidence is not enough, and it says nothing. The evidence review behind that recommendation merged 3 randomised trials with 42009 older European women. Screening works by assessing the fracture risk first, testing bone density in those at high risk, and giving the drug to those found to be at high risk. In the screening group hip fractures were about 17% lower. The original value is RR 0.83, credible range 95% CI 0.73 to 0.93. Turned into an absolute figure, screening 1000 people saves 5 hip fractures. The British SCOOP trial, one of them, took 12483 women aged 70 to 85; in the screening group hip fractures were about 28% lower (HR 0.72, 95% CI 0.59 to 0.89), with no difference in mortality. The 2017 edition of the Chinese Medical Association's diagnosis and treatment guideline recommends a bone density test for women over 65 and men over 70. Below those ages, but with one or more osteoporosis risk factors, it recommends a test as well. In the national survey of 2017 to 2018, the prevalence of osteoporosis among women over 40 was 20.6% and among men 5.0% (China) +- Evidence grade: A +- Sources: US Preventive Services Task Force, Nicholson WK, Silverstein M, et al. (2025). Screening for Osteoporosis to Prevent Fractures: US Preventive Services Task Force Recommendation Statement. JAMA. ; Kahwati LC, Kistler CE, Booth G, et al. (2025). Screening for Osteoporosis to Prevent Fractures: A Systematic Evidence Review for the US Preventive Services Task Force. JAMA. ; Shepstone L, Lenaghan E, Cooper C, et al. (2018). Screening in the community to reduce fractures in older women (SCOOP): a randomised controlled trial. The Lancet. ; Chinese Medical Association Branch of Osteoporosis and Bone Mineral Research (中华医学会骨质疏松和骨矿盐疾病分会) (2017). 原发性骨质疏松症诊疗指南(2017). 中华骨质疏松和骨矿盐疾病杂志, 10(5), 413-443 (Table 5, clinical indications for bone density measurement). ; Wang L, Yu W, Yin X, et al. (2021). Prevalence of Osteoporosis and Fracture in China: The China Osteoporosis Prevalence Study. JAMA Network Open. +- Notes: Disputed: the primary outcome SCOOP set out in advance was all osteoporotic fractures, and the screening group did not have noticeably fewer of them (HR 0.94, 95% CI 0.85 to 1.03). What fell was only hip fractures. All three trials did "assess the risk first, then test bone density", and not one compared bone density testing on its own. The evidence review also estimates that 11.8% to 24.1% of the people found were overdiagnosed, which means people who would not have broken a bone for it in their lifetime. Whether men should be tested: the United States says the evidence is not enough, the Chinese guideline sets 70 and over, so follow what your doctor advises. This item is about whether the bones can take a fall; to fall less often see item 13 (balance and leg strength), and do both. For what to do once osteoporosis has been found, see item 40 (treating osteoporosis once it is found). The people who benefit are mainly you, and your parents as well. + +### 40. Treating osteoporosis once it is found, or after a slight fall has broken a bone: ask the doctor for an osteoporosis drug, keep taking it, and do not substitute calcium for it + +- Cost: It needs a prescription, and years of use. The common ones are tablets taken once a week and injections given once a year by drip. When you start, you have to repeat the bone density test every year, and once it is stable every 1 to 2 years. The hard part is carrying on when you feel nothing wrong. +- In plain language: People already diagnosed with osteoporosis, or people who have broken a bone in a slight fall, who take the drug as the doctor plans have about half fewer hip fractures, and more than half fewer spinal fractures. Among older people who start a yearly infusion after surgery for a hip fracture, about 35% fewer break a bone again over two years. Calcium supplements alone cannot replace these drugs. +- Benefit: A fragility fracture is one that happens in a minor injury or in the course of ordinary activity. The 2017 edition of the Chinese Medical Association guideline names three groups for the drug. First, people who have had a fragility fracture of the spine or the hip. Second, people whose bone density T value is -2.5 or lower, whether or not they have broken a bone. Third, people with low bone mass who have also had a fragility fracture of the wrist, the upper arm or the pelvis, or whose calculated fracture risk is high. The guideline also states that "at present there is not yet sufficient evidence to show that simply taking calcium can replace the other osteoporosis drug treatment" (「目前尚无充分证据表明单纯补钙可以替代其他抗骨质疏松药物治疗」). Cochrane updated its review of alendronate in 2025. Among postmenopausal women who already had osteoporosis or a fracture, hip fractures were about 51% lower in the treated group. The original value is RR 0.49, credible range 95% CI 0.25 to 0.96. Symptomatic spinal fractures were about 55% lower (RR 0.45, 95% CI 0.28 to 0.73). The HORIZON trial took 2127 people who had surgery for a hip fracture, average age 74.5. Starting within 90 days of the surgery, a yearly infusion of zoledronic acid was given, with a median follow-up of 1.9 years. New fractures fell from 13.9% to 8.6%, about 35% fewer. Deaths fell from 13.3% to 9.6%, about 28% fewer. The guideline says that within a year of a hip fracture about 20% of people die of complications of one kind or another. In the national survey of 2017 to 2018, only 1.4% of women diagnosed with osteoporosis or who had broken a bone were taking the drug, and only 0.3% of men (China) +- Evidence grade: A +- Sources: Chinese Medical Association Branch of Osteoporosis and Bone Mineral Research (中华医学会骨质疏松和骨矿盐疾病分会) (2017). 原发性骨质疏松症诊疗指南(2017). 中华骨质疏松和骨矿盐疾病杂志, 10(5), 413-443 (Table 10, indications for drug treatment of osteoporosis; the section on calcium; the section on monitoring response). ; Wells GA, Hsieh SC, Peterson J, et al. (2025). Alendronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database of Systematic Reviews. ; Lyles KW, Colón-Emeric CS, Magaziner JS, et al. (2007). Zoledronic acid and clinical fractures and mortality after hip fracture. New England Journal of Medicine. ; Cummings SR, Lui LY, Eastell R, Allen IE (2019). Association Between Drug Treatments for Patients With Osteoporosis and Overall Mortality Rates: A Meta-analysis. JAMA Internal Medicine. ; Kahwati LC, Kistler CE, Booth G, et al. (2025). Screening for Osteoporosis to Prevent Fractures: A Systematic Evidence Review for the US Preventive Services Task Force. JAMA. ; Wang L, Yu W, Yin X, et al. (2021). Prevalence of Osteoporosis and Fracture in China: The China Osteoporosis Prevalence Study. JAMA Network Open. +- Notes: Disputed: a meta-analysis merging 38 trials found no sign that these drugs lower all-cause mortality (RR 0.98, 95% CI 0.91 to 1.05). Zoledronic acid on its own is not significant either (RR 0.88, 95% CI 0.68 to 1.13). The authors conclude that these drugs should only be taken to break fewer bones. The figures from Cochrane were rated low to moderate for certainty as well. In women who have not yet broken a bone and whose fracture risk is low, alendronate showed no clear effect on hip fractures (RR 0.76, 95% CI 0.43 to 1.32). Trials of a few years cannot see the rare side effects such as osteonecrosis of the jaw or atypical femoral fracture, so keep to the review dates and do not change the dose or stop on your own. Some people suit the injection, some suit the tablets, and which one is for you is the doctor's decision. How osteoporosis is found: see item 39 (bone density test). The people who benefit are mainly you, and your parents as well. +### 41. Bore a hole and measure the thickness before you step onto outdoor ice; below 15 cm do not go on, and stay off river ice and the edge of ice-fishing holes + +- Cost: The cheapest option is a proper ice rink that someone keeps an eye on, and it costs no extra money. If you insist on wild ice, take an ice drill or an ice chisel and bore a hole to measure the thickness. Then buy a pair of rescue ice picks, which are two sharp spikes with handles worn on the chest. All of that comes to a few tens up to one or two hundred yuan. Measuring every now and then along the way takes extra time. +- In plain language: The official line in China is that an adult can go skating only once the ice is 15 cm thick. The floor in the United States and Canada is 10 cm, and that counts only newly formed clear ice. White ice is only half as strong, so the thickness has to be doubled. On the same sheet of ice, the place here may be a foot thick and a few steps away only an inch or two, so you have to measure as you walk. +- Benefit: The China Meteorological Administration says in its public guidance that once the ice reaches 15 cm an adult may go skating. The ice on an open-air rink also has to reach 15 cm before it opens. The Beijing Water Authority says a proper natural ice rink needs 15 cm of ice, and then a week with the temperature held between -8℃ and 1℃ and no strong wind, before it may open. Wild ice is often full of holes cut by people fishing, and once cut it never freezes solid again. The Minnesota Department of Natural Resources guideline covers only newly formed clear ice: do not go on below 4 inches (about 10 cm), and at 4 inches you can walk and ice fish. White ice, made when snow melts and refreezes, is only about half as strong as clear ice, so double the thickness. On the same stretch of water the ice may be a foot thick here and an inch or two a few feet away. Ice on moving water is usually dangerous, and especially so near streams, bridges and culverts. Ice at the shore may also be thinner than ice further out. The US National Weather Service gives the same 4-inch floor for walking, ice fishing and skating. The Canadian Centre for Occupational Health and Safety, passing on the advice of the Canadian Red Cross, says that a few people skating or walking together need 15 cm, and a crowded skating party needs 20 cm. These are all safety guidelines. No study has worked out how many fewer deaths following them would bring (nationwide / the United States and Canada) +- Evidence grade: C +- Sources: China Meteorological Administration (中国气象局) (2019). Walking briskly on ice: how thick does the ice have to be to be safe? 想要冰上健步走，多厚的冰才安全？ ; Beijing Water Authority (北京市水务局) (2024). The proper river and lake ice rinks in Beijing have not met the standard for opening, so do not take the risk of skating on "wild ice". 北京正规河湖冰场还未达开放标准，别冒险滑「野冰」. ; Minnesota Department of Natural Resources. General ice thickness guidelines. ; Minnesota Department of Natural Resources. Ice safety. ; Minnesota Department of Natural Resources. Steps for surviving immersion. ; National Weather Service. Winter Weather Safety: Ice and Frost. ; Canadian Centre for Occupational Health and Safety. Working on or near Ice Covered Water - Basic Information. +- Notes: Every official guideline says there is no ice that is a hundred per cent safe. Meeting the thickness is only a threshold, not a guarantee. The most dangerous times are the first cold of early winter, when the ice has just formed, and the first thaw of early spring. If you do go through, do not take your clothes off: the air held in thick clothing helps keep you up. Turn back the way you came, because the ice there is the strongest. Put an arm on the ice, drive the spike into it, push with your feet and drag yourself out. Once you are out, do not stand up; roll flat away from the hole, then get warm as fast as you can, and go to hospital if you are badly frozen. If you see someone go through the ice, do not run out and pull them out yourself. What to do is in item 25 in section 13 (do not go into the water yourself when you see someone drowning): shout for help, call 110, and then hand over a pole or throw a rope. The people who benefit are mainly you, and any children you take out onto the ice. \ No newline at end of file diff --git "a/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" "b/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" index f64a2d4d..46ee8e33 100644 --- "a/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" +++ "b/en/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" @@ -86,14 +86,15 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: B - Sources: Wang K, Matsuyama Y, Kiuchi S, et al. (2026). Routine oral health practices and all-cause mortality. Journal of Dentistry. ; Ko MJ, Seo S, So JS, et al. (2026). Deteriorated oral health and function as risk factors for physical disability and mortality in community-dwelling older adults: a systematic review and meta-analysis. European Geriatric Medicine. - Notes:Disputed. These studies only followed records. People with bad teeth often have worse health and worse finances across the board. Whether bad teeth drag the body down, or poor health drags the teeth down, is not clear. But this costs almost nothing, and periodontal disease and missing teeth affect eating on their own -### 11. Walk 7000–8000 steps a day +### 11. Walk 7000–8000 steps a day, or add up 150–300 minutes of brisk walking a week -- Cost: No money. Walking for 60–90 minutes a day can be split into the commute and the shopping trip. The hard part is doing it every day. -- In plain language: People who walk 5800 steps a day have about 40% lower death rates in the same period than people who walk only 3500. At 7800 steps the fall is about 45%. Past that the figure drops a little further, then flattens out. For people over 60, 6000–8000 steps is enough. -- Benefit: Pooling 15 studies that followed a group of people, all of them only followed records with no control group. The 15 studies cover 47,471 people and 3,013 deaths. People are split into four bands by daily steps, and the median of the lowest band is 3553 steps a day. The medians of the other three bands are 5801, 7842 and 10,901 steps a day. The death risk of those three bands is about 40%, 45% and 53% lower than the lowest band (HR 0.60, 0.55, 0.47). Among people over 60, the benefit flattens out after 6000–8000 steps a day. Among people under 60 it flattens out after 8000–10000 steps. A separate pooled analysis of several studies: the effect starts at about 3867 steps a day. After that, every extra 1000 steps a day cuts the death risk by about 15%. +- Cost: It costs nothing. Counted in steps, that is 60–90 minutes of walking a day, and you can fit it into the commute and the shopping. Counted as exercise time, it is 20–45 minutes a day. The hard part is keeping it up over the years. +- In plain language: People who walk about 7800 steps a day have about a 45% lower chance of dying in the same period than people who walk only 3500. Counted by time it comes out the same: 150–300 minutes of brisk walking a week is about 30% lower than doing no exercise. Past a certain amount it stops falling any further; over 60, 6000–8000 steps is enough. Either way of counting it, hitting one of the two is enough. +- Benefit: 15 studies that followed one group of people were pooled, only recording, with no groups to compare. The 15 studies cover 47471 people and 3013 deaths. People are split into four bands by steps a day, and the median of the lowest band is 3553 steps a day. The medians of the other three are 5801, 7842 and 10901 steps a day. The death risk in those three bands is about 40%, 45% and 53% lower than in the lowest band (HR 0.60, 0.55, 0.47). Past 6000–8000 steps a day, people over 60 gain nothing more. People under 60 level off past 8000–10000 steps. A separate analysis pooled several studies: the effect starts at about 3867 steps a day. After that, every extra 1000 steps a day lowers the death risk by about 15%. Counted as exercise time: several cohort studies were pooled together, again only recording, with no groups to compare. People reaching 1–2 times the floor set by the guidelines have a death risk about 31% lower than people who do no exercise (HR 0.69). That amount is 7.5–15 MET·h a week, which comes to about 150–300 minutes a week of brisk walking. People who did not reach the floor are still about 20% lower (HR 0.80). People reaching 3–5 times the floor are about 39% lower (HR 0.61), and that is the bottom of the curve. Adding more on top does not go lower, and does no harm (past 10 times it is HR 0.69). Another pooled study using accelerometers: the quarter of people with the most moderate-to-vigorous activity have a death risk about 48% lower than the quarter with the least (HR 0.52). An accelerometer is an instrument worn on the body that measures activity directly. - Evidence grade: A -- Sources: Paluch AE et al. (2022). Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. Lancet Public Health. ; Banach M et al. (2023). The association between daily step count and all-cause and cardiovascular mortality: a meta-analysis. European Journal of Preventive Cardiology. -- Notes: The stretch from 4000 steps to 7000 steps is the best value, and you do not have to reach ten thousand. Note that these studies only followed records: the people who walked least include some who were too ill to walk at all. They walked little because they were about to die, not the other way round. So the gap these studies report is often too large. The direction is clear all the same. This item and item 14 (150–300 minutes of moderate exercise a week) are the same thing measured two ways, and hitting either one is enough. +- Sources: Paluch AE et al (2022). Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. Lancet Public Health. ; Banach M et al (2023). The association between daily step count and all-cause and cardiovascular mortality: a meta-analysis. European Journal of Preventive Cardiology. ; Arem H et al (2015). Leisure time physical activity and mortality: a detailed pooled analysis of the dose-response relationship. JAMA Internal Medicine. ; Ekelund U et al (2019). Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis. BMJ. +- Notes: Steps and exercise time are the same thing counted a different way; hitting one of the two is enough. The stretch from 4000 steps to 7000 steps is where the value is best, and there is no need to make up ten thousand steps. Note that these studies only followed records: among the people who walk least and move least there are people who were ill to begin with and could not walk. They moved less because they were near the end, not the other way round, so the gap worked out is usually too large. The studies that measured activity with instruments have short follow-up and older participants, and the real benefit is not as large as "a risk 0.52 times as high". But the trend of more movement being better is clear. + ### 12. If you have high blood pressure or high blood lipids, take the medicine as prescribed and do not stop on your own - Cost: Blood pressure pills and statins bought through the volume-based procurement scheme cost from a few yuan to a few tens of yuan a month. One dose a day takes very little time. The hard part is carrying on with them when you feel fine and think you no longer need them. @@ -102,24 +103,17 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Ettehad D et al. (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Cholesterol Treatment Trialists' (CTT) Collaboration (2010). Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170 000 participants in 26 randomised trials. Lancet. ; Chowdhury R et al. (2013). Adherence to cardiovascular therapy: a meta-analysis of prevalence and clinical consequences. European Heart Journal. - Notes: Only do this if the doctor has judged that you need the medicine. Healthy people do not. The figures for the group that keeps taking the medicine come from studies that only followed records, and people who manage to keep taking it are more disciplined to begin with, so the benefit is overstated. The figures from the two randomised trial pools above are more reliable. This section has no separate item on blood glucose: the evidence that pushing glucose lower lowers all-cause mortality is not as consistent as it is for blood pressure and blood lipids. -### 13. Sleep about 7 hours a night and keep a fixed routine +### 13. Sleep about 7 hours a night, and keep a fixed routine -- Cost: No money. You have to set aside the time to sleep, and for most people that means giving up some phone time. The hard part is fixing your bedtime and your wake-up time. -- In plain language: People who sleep under 7 hours over the long run have about 10% higher death rates in the same period. The more irregular the routine, the higher the risk. The most regular sleepers have death rates 20% to nearly 50% lower than the least regular. How regular your routine is therefore says more about your death risk than how long you sleep. -- Benefit: Pooling 16 follow-up studies, all only followed records with no control group, covering 1,380,000 people and 113,000 deaths. Short sleepers have a death risk about 12% higher (RR 1.12). Long sleepers about 30% higher (RR 1.30). A separate summary bands people by how long they sleep, and the 7-hour band has the lowest risk. Below 7 hours, every 1 hour less raises the risk about 6% (RR 1.06). Above 7 hours, every 1 hour more raises it about 13% (RR 1.13). UK Biobank has wristband records for about 61,000 people, again only followed records with no control group. People are split into five bands by how regular their routine is. The four more regular bands have a death risk 20%–48% lower than the least regular band. How regular the routine is predicts death better than how long you sleep. A separate UK Biobank study measured bedtimes with wristbands in 51,562 people, and 3,853 of them developed cardiovascular disease during the study. Social jet lag is the gap between the midpoint of sleep on a weekday and the midpoint at the weekend. People with a social jet lag of 2 hours or more have a cardiovascular risk about 30% higher (HR 1.30, 95% CI 1.11–1.54). The same association holds among people whose sleep duration is normal. -- Evidence grade: A -- Sources: Cappuccio FP et al. (2010). Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. ; Yin J et al. (2017). Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. JAHA. ; Windred DP et al. (2024). Sleep regularity is a stronger predictor of mortality risk than sleep duration: A prospective cohort study. Sleep. ; Kumar N, Krishnamurthy S (2026). Social jet lag is associated with incident cardiovascular disease independent of sleep duration and cardiac genetic risk. Journal of Internal Medicine. -- Notes: What needs changing is sleeping too little and having an irregular routine. There is no need to deliberately cut your sleep short. Long sleepers having a higher risk is most likely the cause running the other way: depression, chronic illness and sleep apnoea make people sleep more. Regularity of routine for now rests on two follow-up studies, both only followed records with no control group, so on its own it is grade B evidence. How to make up for a night up late is in item 39 in this section (sleep earlier the next night). -### 14. Build up 150–300 minutes of moderate exercise a week, and brisk walking counts - -- Cost: No money. 20–45 minutes a day. The hard part is keeping it up over the long run. -- In plain language: People who brisk-walk for 150–300 minutes a week have about 30% lower death rates in the same period than people who do not exercise. Even if you fall short of the target, the fall is about 20%. Three to five times the target brings the figure down to about 40% lower. Going further will not lower it more, but it does no harm. -- Benefit: A pooled analysis of several follow-up studies, all only followed records with no control group: people whose activity reaches 1 to 2 times the guideline minimum have a death risk about 31% lower than people who do not exercise (HR 0.69). That amount is 7.5–15 MET·h a week, which is about 150–300 minutes of brisk walking a week. People who fall short of the minimum are still about 20% lower (HR 0.80). People who reach 3 to 5 times the minimum are about 39% lower (HR 0.61), and that is where the reduction tops out. Going further will not lower it more, but it does no harm (above 10 times, HR 0.69). A separate pooled study using accelerometers: the most active quarter of people by moderate-to-vigorous activity have a death risk about 48% lower than the least active quarter (HR 0.52). An accelerometer is a device you wear that measures activity directly. +- Cost: It costs nothing. You have to set aside the time to sleep, and most people do it by turning time spent on their phone into sleep. The hard part is fixing the hour you get into bed and the hour you get up. +- In plain language: People who get less than 7 hours of sleep over the long term have about a 10% higher chance of dying in the same period. The more chaotic the routine, the higher the risk. The most regular sleepers have a death rate 20% to nearly 50% lower than the least regular. That is, whether the routine is regular tells you more about the risk of death than how long you sleep. +- Benefit: 16 cohort studies were pooled, only recording, with no groups to compare, covering 1,380,000 people and 113,000 deaths. People who slept short had a death risk about 12% higher (RR 1.12). People who slept long had it about 30% higher (RR 1.30). Another pooled analysis split people by how long they slept: the 7-hour band has the lowest risk. Every 1 hour below 7 hours raises the risk about 6% (RR 1.06). Every 1 hour above 7 hours raises it about 13% (RR 1.13). UK Biobank has wristband records for about 61,000 people, again only recording, with no groups to compare. People are put into five bands by how regular their routine is. The four more regular bands have a death risk 20%–48% lower than the least regular band. Whether the routine is regular predicts death better than how long you sleep. A separate UK Biobank study used wristbands to measure the sleep times of 51,562 people, and in that time 3,853 of them developed cardiovascular disease. Social jet lag is how much the midpoint of sleep differs between weekdays and at the weekend. People whose social jet lag reached 2 hours or more had a cardiovascular risk about 30% higher (HR 1.30, 95% CI 1.11–1.54). The same association is there in people whose sleep duration is normal. - Evidence grade: A -- Sources: Arem H et al. (2015). Leisure time physical activity and mortality: a detailed pooled analysis of the dose-response relationship. JAMA Internal Medicine. ; Ekelund U et al. (2019). Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis. BMJ. -- Notes: This item and item 11 (7000–8000 steps a day) only need one of the two. Note that these studies only followed records. The ones that measured activity with a device tracked people for a short time and the participants were mostly older, so the figures include an element of "became inactive after falling ill". The real benefit is not as large as a risk ratio of 0.52. +- Sources: Cappuccio FP et al (2010). Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. ; Yin J et al (2017). Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. JAHA. ; Windred DP et al (2024). Sleep regularity is a stronger predictor of mortality risk than sleep duration: A prospective cohort study. Sleep. ; Kumar N, Krishnamurthy S (2026). Social jet lag is associated with incident cardiovascular disease independent of sleep duration and cardiac genetic risk. Journal of Internal Medicine. +- Notes: What has to change is sleeping too little and having a chaotic routine; there is no need to deliberately cut your sleep short. People who sleep long have a higher risk, and most of that is the cause running the other way: depression, chronic illness and sleep apnoea make people sleep more. Also, a regular routine is at present supported by two cohort studies, both of which only recorded, with no groups to compare, which on their own are grade B evidence. For making up a night up late: see item 38 in this section (make up for a night up late the next night). -### 15. Play a racket sport three times a week, 45 minutes each time + +### 14. Play a racket sport three times a week, 45 minutes each time - Cost: Court hire costs a few tens of yuan each time. About 2 hours a week. The hard part is getting enough people and a court together, and turning up every week. - In plain language: People who regularly play tennis, badminton or table tennis have about half the death rate in the same period as people who do not, and cardiovascular death is about 56% lower. Swimmers are about 28% lower. Running and football showed no difference in this dataset. @@ -127,7 +121,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Oja P, Kelly P, Pedisic Z, et al. (2017). Associations of specific types of sports and exercise with all-cause and cardiovascular-disease mortality: a cohort study of 80 306 British adults. British Journal of Sports Medicine, 51(10), 812-817. - Notes:Disputed. These studies only followed records. People who take up a sport are healthier to begin with and have more social contact. The fact that running showed no difference also suggests that what matters may be which kind of person picks which sport. Do not read this item as saying running is useless, because this section has other items that go by total activity. -### 16. Add up the short bursts of effort, such as stairs and walking fast, to four or five minutes a day +### 15. Add up the short bursts of effort, such as stairs and walking fast, to four or five minutes a day - Cost: No money, and no need to carve out extra exercise time. The hard part is remembering to take the stairs and to walk faster instead of taking the easy way. - In plain language: People who do no exercise at all still have a death rate about 40% lower in the same period than people who do none at all, as long as they have three one- or two-minute bursts of effort a day. Cardiovascular death is nearly half lower. Effort here means things like taking the stairs and hurrying to get somewhere. @@ -135,7 +129,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Stamatakis E, Ahmadi MN, Gill JMR, et al. (2022). Association of wearable device-measured vigorous intermittent lifestyle physical activity with mortality. Nature Medicine, 28, 2521-2529. - Notes:Disputed. This study measured activity with wearable devices, which is more reliable than asking people to fill in a questionnaire from memory. But it still only followed records, it did not run a controlled trial, and it tracked people for 6.9 years. The subjects were people who did no exercise at all, so this item does not apply to people who already exercise regularly -### 17. Do 30–60 minutes of strength training a week +### 16. Do 30–60 minutes of strength training a week - Cost: Bodyweight squats and press-ups cost nothing. Train 1–2 times a week, 20–30 minutes each time. The hard part is fitting it into the week's schedule. - In plain language: People who do 30–60 minutes of strength training a week have a death rate 10% to 20% lower in the same period than people who do none. Doing more does not make it better. Strength work and aerobic work together work best. @@ -143,7 +137,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Momma H et al. (2022). Muscle-strengthening activities are associated with lower risk and mortality in major non-communicable diseases: a systematic review and meta-analysis of cohort studies. British Journal of Sports Medicine. - Notes: These studies only followed records, and the activity level was self-reported. The evidence that training too much makes matters worse is very weak, so there is no reason to hold back on that account. Strength training also helps older people avoid falls and keep their muscle, and those points are in section 1. -### 18. Do not sit for long stretches, get up and move from time to time +### 17. Do not sit for long stretches, get up and move from time to time - Cost: No money, and getting up a few times takes very little time. The hard part is that once you sit down you lose track of time, so set a reminder. - In plain language: The people who sit longest have a death rate in the same period 2.6 times that of the people who sit least. Sitting for a long unbroken stretch carries a separate risk on top of that. But as long as you get 60–75 minutes of moderate activity a day, most of the extra risk from sitting is cancelled out. @@ -151,7 +145,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Diaz KM et al. (2017). Patterns of Sedentary Behavior and Mortality in U.S. Middle-Aged and Older Adults: A National Cohort Study. Annals of Internal Medicine. ; Ekelund U et al. (2016). Does physical activity attenuate, or even eliminate, the detrimental association of sitting time with mortality? A harmonised meta-analysis of data from more than 1 million men and women. Lancet. - Notes: The reliable conclusion is that how much you move matters more than how long you sit, and that when activity is enough the risk from sitting largely disappears. Note that these studies only followed records, and the people who sit the most include a large number who were already ill, so a risk as high as 2.63 times clearly contains an element of "became sedentary after falling ill". In the same pooled analysis, the risk of watching more than 3 hours of television a day is present at every activity level, and the most active group only pushes that starting point out to 5 hours. -### 19. Eat less processed meat (ham, bacon, sausage, luncheon meat) +### 18. Eat less processed meat (ham, bacon, sausage, luncheon meat) - Cost: No money, and it may even save money, and it takes no time. The hard part is cutting back on the mouthful you actually like. - In plain language: People who eat the most ham, bacon and sausage have a death rate about 20% higher in the same period than people who eat the least. Every extra daily serving of processed meat adds 20% to the figure. @@ -160,31 +154,34 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Sources: Larsson SC, Orsini N (2014). Red meat and processed meat consumption and all-cause mortality: a meta-analysis. American Journal of Epidemiology. ; Schwingshackl L et al. (2017). Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. American Journal of Clinical Nutrition. ; Johnston BC et al. (2019). Unprocessed Red Meat and Processed Meat Consumption: Dietary Guideline Recommendations From the NutriRECS Consortium. Annals of Internal Medicine. (the NutriRECS guidelines, the other side of the dispute) - Notes:Disputed. The NutriRECS guidelines use a standard evidence grading system (GRADE) and rate these studies as not very reliable. They give one very weak recommendation: carry on eating the way you do now. The two sides disagree about whether the evidence is hard enough, not about the direction. No study says processed meat is good for you. Unprocessed red meat has a small effect and shows no statistically significant difference. What needs controlling is the processed meat. -### 20. Drink less alcohol, or none +### 19. Drink less alcohol, or none at all -- Cost: No money, and it saves money, and it takes no time. The hard part is the awkward moment when people at a gathering push a drink on you. -- In plain language: Past 100 g of pure alcohol a week, the more you drink the earlier you die. 100 g of pure alcohol is about 2.5 L of beer. People drinking 100–200 g a week at 40 live about half a year less; people drinking 200–350 g lose one to two years; people drinking more than 350 g lose four or five years. Judged by overall health, the best value is not drinking. -- Benefit: Pooling 83 follow-up studies, only followed records with no control group, covering 600,000 people who drink alcohol. The lowest death risk is at no more than 100 g of pure alcohol a week. That amount is about 2.5 L of 5% beer, or 300 mL of 40-proof baijiu. People drinking 100–200 g a week at 40 have a life expectancy about 6 months shorter. People drinking 200–350 g a week lose 1–2 years. People drinking more than 350 g lose 4–5 years. The source reports all three in years of life. The Global Burden of Disease study GBD 2016 adds up all kinds of health loss and puts the lowest-risk amount at 0 drinks a week. A separate pooled analysis corrected for bias in study design. People drinking 1.3–24 g a day have a death risk about 7% lower than people who never drink (RR 0.93), but that one is not statistically significant. People drinking 45–64 g a day are about 19% higher (RR 1.19). People drinking above 65 g a day are about 35% higher (RR 1.35). +- Cost: It costs nothing, it saves money, and it takes no time. The hard part is the awkward moment when someone pushes a drink on you at a gathering. +- In plain language: Once you are past 100 g of pure alcohol a week, the more you drink the earlier you die. 100 g of pure alcohol is about 2.5 L of beer. People drinking 100–200 g a week at 40 have about half a year less of life; those drinking 200–350 g have one or two years less; those past 350 g have four or five years less. Counted across overall health, the best value is not drinking. +- Benefit: 83 prospective studies were pooled, only recording, with no groups to compare, covering 600,000 people who drink. The lowest point of the death risk is at no more than 100 g of pure alcohol a week. That amount is about 2.5 L of beer at 5%, or 300 mL of 40-degree baijiu. People drinking 100–200 g a week at 40 lose about 6 months of life expectancy. People drinking 200–350 g a week lose 1–2 years. Those past 350 g lose 4–5 years. The original reports those three in years of life. The Global Burden of Disease study, GBD 2016, adds up all the different kinds of health loss together, and the amount of drinking with the lowest risk is 0 cups a week. Another pooled analysis corrected for the bias in the design of the studies. People drinking 1.3–24 g a day have a death risk about 7% lower than people who never drink in their lives (RR 0.93), but that finding showed no statistical difference. People drinking 45–64 g a day are about 19% higher (RR 1.19). People drinking more than 65 g a day are about 35% higher (RR 1.35). - Evidence grade: A -- Sources: Wood AM et al. (2018). Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. ; GBD 2016 Alcohol Collaborators (2018). Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. ; Zhao J et al. (2023). Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Network Open. ; Di Castelnuovo A et al. (2006). Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Archives of Internal Medicine. (the other side of the dispute) -- Notes:Disputed. The side that argues drinking less is good points to a pooled analysis of 34 long-term follow-up studies. It reports that light drinkers have an all-cause mortality up to 17%–18% lower. The amount it gives is no more than 4 drinks a day for men and 2 for women. Within that amount, the more people drink the lower their mortality. The other side argues that the "non-drinkers" group includes people who gave up because of illness and people who were in poor health to begin with. One pooled analysis adjusted specifically for those confounders, and once that was done the protective effect disappeared. The Global Burden of Disease study also puts 0 as the safest. The safe way to read it: drinking less will not clearly shorten your life, and drinking more certainly will. Starting to drink for health reasons has no basis. What to do if you want to drink less is in item 22 in this section (count the drinks first, then see a doctor). People who drink every day should not stop abruptly on their own, see item 21 in this section (do not stop abruptly on your own). -### 21. If you drink every day and shake or feel anxious when you stop, do not stop abruptly on your own +- Sources: Wood AM et al (2018). Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. ; GBD 2016 Alcohol Collaborators (2018). Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. ; Zhao J et al (2023). Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Network Open. ; Di Castelnuovo A et al (2006). Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Archives of Internal Medicine. (the disputed side) +- Notes: Disputed. The side that argues "drinking less does you good" is a pooled analysis of 34 long-term cohort studies. It reports that people who drink a little have an all-cause mortality lower by as much as 17%–18%. The amount it gives is no more than 4 glasses a day for men and 2 for women. Within that amount, the people who drank more had a lower mortality. The other side argues that people who gave up drinking because of illness, and people who were simply less well to begin with, are mixed into the "do not drink" group. One pooled analysis takes those confounders out, and once it does, the protective effect is gone. The Global Burden of Disease study also gives 0 as the safest. The safe way to read it: drinking less will not clearly cost you life expectancy, and drinking more certainly will. "Starting to drink for the sake of your health" has no basis. For what to do if you want to drink less, see item 21 in this section (count the drinks for the week first, then talk to a doctor). People who drink every day must not stop on their own, see item 20 in this section (do not stop on your own). + +### 20. If you drink every day and stopping makes you shake and panic, do not stop on your own -- Cost: The registration fee for psychiatry or an addiction medicine clinic runs from a few yuan to a few tens of yuan. Severe cases need a few days in hospital. -- In plain language: Someone who drinks a lot every day and stops suddenly goes into withdrawal. Mild withdrawal means shaking, sweating, anxiety and sleeplessness. Severe withdrawal means seizures or delirium tremens, which is a medical emergency that kills. Norway followed more than thirty thousand people, and of those who had delirium tremens, 8% died each year afterwards. If you want to quit, go to hospital first and say how much you drink each day. -- Benefit: A 2026 clinical review states that people who drink heavily over the long run develop withdrawal syndrome after stopping suddenly or cutting down sharply, with tremor, autonomic overactivity, anxiety, insomnia, perceptual disturbances, seizures and delirium tremens, and that in ordinary internal medicine this is 「常见且可能危及生命」 — common and potentially life-threatening. A Norwegian national register cohort of 36,287 people diagnosed with alcohol dependence, alcohol withdrawal state or delirium tremens between 2009 and 2015: the annual mortality in the delirium tremens group was 8.0%, in the withdrawal state group 5.0%, and in the alcohol dependence group 3.6%. The standardised mortality ratio for the delirium tremens group was 9.8 (95% CI 8.9–10.7, the credible range), which means the death risk is about 9.8 times that of the general population of the same age. +- Cost: One appointment in psychiatry or an addiction medicine clinic costs a few yuan to a few tens of yuan. Severe cases stay in hospital for a few days. +- In plain language: Someone who drinks heavily every day and stops suddenly gets withdrawal. The mild form is shaking, sweating, palpitations and not being able to sleep; the severe form is seizures or delirium tremens, and that is an emergency that kills. Norway followed more than 30,000 people, and among those who had delirium tremens, 8% a year died afterwards. If you want to stop, go to hospital first and say how much you drink every day. +- Benefit: A clinical review from 2026 states that people who have drunk heavily for a long time get withdrawal syndrome after stopping suddenly or cutting the amount sharply, showing tremor, autonomic overactivity, anxiety, insomnia, sensory disturbance, seizures and delirium tremens, and that in ordinary internal medicine this is "common and potentially life-threatening". A nationwide register cohort in Norway took in 36287 people diagnosed between 2009 and 2015 with alcohol dependence, alcohol withdrawal state or delirium tremens: the annual mortality was 8.0% in the delirium tremens group, 5.0% in the withdrawal state group and 3.6% in the alcohol dependence group. The standardised mortality ratio in the delirium tremens group was 9.8 (95% CI 8.9–10.7, and that is the credible range), which means the risk of death is about 9.8 times that of the general population of the same age. - Evidence grade: B - Sources: Caputo F, Lungaro L, Costanzini A, De Giorgio R, Addolorato G (2026). Alcohol withdrawal syndrome in hospitalized patients: a practical review. European Journal of Internal Medicine, 107103. ; Bramness JG, Heiberg IH, Høye A, Rossow I (2023). Mortality and alcohol-related morbidity in patients with delirium tremens, alcohol withdrawal state or alcohol dependence in Norway: A register-based prospective cohort study. Addiction, 118(12), 2352–2359. -- Notes: How to judge it yourself: if you drink almost every day, and missing a day brings shaking, sweating, anxiety and sleeplessness, and you need a drink in the morning to feel right, then any one of those means you should not stop on your own. What hospitals normally do is give benzodiazepines to get you through those few days and add thiamine (vitamin B1) at the same time. The grade is B because the Norwegian cohort says what happened later to people who had delirium tremens, not what happens to people who stop abruptly on their own, and getting from one to the other is an inference. You still need to quit, but quit in hospital instead. How much a week counts as too much is in item 20 in this section (drink less or none), and what to do if you want to drink less is in item 22 in this section (count the drinks first, then see a doctor). -### 22. To drink less, first count how much you drink in a week, then talk to a doctor for a few minutes +- Notes: How to judge it for yourself: if you drink almost every day, a day without drinking brings shaking, sweating, palpitations and not being able to sleep, and you need a drink in the morning to feel comfortable, then having any one of those is enough: do not stop abruptly on your own. What hospitals normally do is give benzodiazepines to see you through those few days, and give thiamine (vitamin B1) as well. The grade B is because the Norwegian cohort says what happened later to people who had delirium tremens, not what happens to people who stop on their own; between the two there is an inference. The drinking still has to stop, only now you stop it in hospital. For how much a week counts as too much see item 19 in this section (drink less alcohol, or none at all), and for what to do if you want to drink less see item 21 in this section (count the drinks for the week first, then talk to a doctor). + +### 21. If you want to drink less, count the drinks for the week first, then talk to a doctor for a few minutes -- Cost: No money. Counting the drinks takes one line a day. Asking the question while you are at the clinic costs nothing extra. -- In plain language: A doctor who spends a few minutes looking at how much you drink, explaining the risk and setting you a target leaves you drinking 20 grams of pure alcohol less a week on average a year later, which is about half a litre of beer. Talking for longer brings no further benefit. For people who cannot stop on their own, two drugs proved effective abroad, and for every 12 people treated about 1 more does not go back to drinking as much as before. -- Benefit: Pooling 34 randomised trials and 15,197 people, people who received a brief intervention drank 20 grams of pure alcohol less per week a year later than people who received only the minimum intervention or none (95% CI 12 to 28 grams less, the credible range; moderate-quality evidence). The baseline drinking of the participants in these studies averaged 244 grams a week. A brief intervention means advice or lifestyle counselling of no more than 5 sessions and under 60 minutes in total; talking for longer brought no extra benefit. For the drugs: pooling 122 randomised trials and 22,803 people, acamprosate means that for every 12 people treated (95% CI 8 to 26) about 1 more does not return to drinking at all. Oral naltrexone at 50 milligrams a day means that for every 12 people treated (8 to 26) about 1 more does not return to heavy drinking. +- Cost: It costs nothing. Counting the drinks takes one line a day. Asking at the clinic as it comes up does not cost extra either. +- In plain language: A doctor spends a few minutes looking at how much you drink, setting out the risk and setting you a goal, and a year later you drink 20 grams of pure alcohol a week less on average, about half a litre of beer. Talking for longer does not add more. For people who cannot stop, two drugs have been shown to work abroad: for every 12 people treated, about 1 more does not go back to drinking as much as before. +- Benefit: 34 randomised trials and 15197 people were pooled together. People who received a brief intervention drank 20 grams of pure alcohol a week less a year later than people who received the minimum intervention or none at all (95% CI 12 to 28 grams less, and that is the credible range; moderate quality evidence). The baseline drinking of the participants in these studies averaged 244 grams a week. A brief intervention means advice or lifestyle counselling of no more than 5 sessions and less than 60 minutes in total; talking for longer brought no extra benefit. On the drugs: 122 randomised trials and 22803 people were pooled together, and for every 12 people treated with acamprosate (95% CI 8 to 26) 1 more does not go back to drinking at all; with oral naltrexone at 50 milligrams a day, for every 12 people treated (8 to 26) 1 more does not go back to heavy drinking. - Evidence grade: A - Sources: Kaner EF, Beyer FR, Muirhead C, et al. (2018). Effectiveness of brief alcohol interventions in primary care populations. Cochrane Database of Systematic Reviews, 2, CD004148. ; Jonas DE, Amick HR, Feltner C, et al. (2014). Pharmacotherapy for adults with alcohol use disorders in outpatient settings: a systematic review and meta-analysis. JAMA, 311(18), 1889–1900. -- Notes: The benefit is graded "small" because both of these bodies of evidence measure how much people drink, not mortality, and by the rule of this book a surrogate endpoint alone gets "small". Most of the drug trials were done abroad, most participants already met the diagnostic criteria for alcohol dependence, and every trial also gave psychosocial support, so the figures are the extra benefit on top of that. Whether these two drugs are approved in China and whether you can be prescribed them is up to your doctor and the drug's own package information, so do not buy them yourself online. If you already have withdrawal symptoms, see item 21 in this section (do not stop abruptly on your own). -### 23. Eat a small handful of nuts every day +- Notes: The benefit is graded "small" because both bodies of evidence measure how much is drunk, not mortality, and by this book's rule a surrogate endpoint on its own is graded "small". Most of the drug trials were done abroad, the participants mostly met the diagnostic criteria for alcohol dependence, and every trial gave psychosocial support at the same time, so the figures are "the extra benefit on top of that". For the indications of these two drugs in China, and whether they can be prescribed at all, go by what the doctor and the package insert say, and do not order them yourself online. If withdrawal has already appeared, see item 20 in this section (do not stop on your own). + +### 22. Eat a small handful of nuts every day - Cost: 28 grams a day costs two or three hundred yuan a year. You eat them in passing, and it takes no time. - In plain language: People who regularly eat nuts have about 20% lower death rates in the same period than people who do not. Eating them once a week already shows a difference, about 10%. Eating a handful a day shows the clearest difference of all. @@ -192,7 +189,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Bao Y, Han J, Hu FB, et al. (2013). Association of nut consumption with total and cause-specific mortality. New England Journal of Medicine, 369(21), 2001-2011. - Notes: Buy them plain, with no added salt. Nuts are not low in calories, so do not treat them as a snack to eat freely. Note that these studies only followed a group of people over the long term, and nut eaters have better lifestyles across the board, so the figures need discounting. -### 24. Replace some of your red meat with fish and poultry +### 23. Replace some of your red meat with fish and poultry - Cost: No money, and it takes no time. This is swapping one kind of meat for another, not buying more. The hard part is shifting what you prefer. - In plain language: Eating two more servings a week of processed meat or red meat raises the death rate in the same period by about 3%. Swapping the same number of servings for fish or poultry shows no rise. The effect in this section is small, which makes this an in-passing swap rather than something to fight over. @@ -201,7 +198,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Sources: Zhong VW, Van Horn L, Greenland P, et al. (2020). Associations of Processed Meat, Unprocessed Red Meat, Poultry, or Fish Intake With Incident Cardiovascular Disease and All-Cause Mortality. JAMA Internal Medicine, 180(4), 503-512. - Notes: Two servings a week makes very little difference, so do not expect to live longer by swapping meat. When you are spending the same money on meat, you can favour fish and poultry -### 25. Replace some of your white rice and white flour with whole grains +### 24. Replace some of your white rice and white flour with whole grains - Cost: Brown rice, oats and wholemeal bread cost a little more than white rice and white flour, and take no time. The hard part is getting used to the taste. - In plain language: People who eat 90 g more whole grains a day have about 17% lower death rates in the same period. 90 g is about 3 servings. At about 200 g a day the figure is still falling. @@ -209,7 +206,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Aune D et al. (2016). Whole grain consumption and risk of cardiovascular disease, cancer, and all cause and cause specific mortality: systematic review and dose-response meta-analysis of prospective studies. BMJ. ; Schwingshackl L et al. (2017). Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. American Journal of Clinical Nutrition. - Notes: You do not have to switch everything to whole grains, because half of it already puts you in the stretch where the benefit is most visible. Note that these studies only followed records, and people who eat whole grains were healthier to begin with, so the figures are too large. Results also differ a lot between studies (I² 83%, and the higher that number, the less the studies agree). -### 26. Drink tea three or more times a week +### 25. Drink tea three or more times a week - Cost: A few tens to a few hundred yuan a year. Making a cup takes a few minutes and takes very little time. - In plain language: People with a habit of drinking tea have about 15% lower death rates in the same period. Measured from 50, they live 1.26 years longer on average, and they spend 1.41 more years free of atherosclerotic cardiovascular disease. @@ -217,7 +214,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Wang X, Liu F, Li J, et al. (2020). Tea consumption and the risk of atherosclerotic cardiovascular disease and all-cause mortality: The China-PAR project. European Journal of Preventive Cardiology, 27(18), 1956-1963. ; mycotoxins in tea: Cui P et al. (2020). Quantitative analysis and dietary risk assessment of aflatoxins in Chinese post-fermented dark tea. Food and Chemical Toxicology. ; Zhou H et al. (2022). Mycotoxins in Tea (Camellia sinensis (L.) Kuntze): Contamination and Dietary Exposure Profiling in the Chinese Population. Toxins. ; green tea extract and liver damage: Hu J et al. (2018). The safety of green tea and green tea extract consumption in adults - Results of a systematic review. Regulatory Toxicology and Pharmacology. - Notes:Disputed, and the reason is this: these studies only followed records, and Chinese men who drink tea also have a high rate of smoking and drinking. The authors have already subtracted part of that effect, but not all of it. One more thing: do not drink it scalding hot, and for the temperature of hot drinks see the item in this section on that. Some people worry about mycotoxins in tea. In China, 158 samples of dark tea have been tested and 2 of them came back with aflatoxin, and worked out against the amount of tea people drink, the intake is below the internationally acceptable level. Another study tested 352 samples of tea for 16 mycotoxins, and only ochratoxin A in dark tea came out above the limit on average. Worked out against how much tea people in China drink, neither study found a dietary risk. Keep tea somewhere dry, and do not drink it if it has got damp and gone mouldy. Green tea extract capsules are a different thing: a large dose of concentrated catechins swallowed at once can damage the liver, and brewed tea does not have that problem. So this is about brewed tea only, and do not substitute a green tea extract supplement for it. -### 27. Drink three to four cups of coffee a day, with no sugar and no cream topping +### 26. Drink three to four cups of coffee a day, with no sugar and no cream topping - Cost: Brewing it yourself costs a yuan or two a day. Making a cup takes a few minutes and very little of your time. - In plain language: People who drink three to four cups of coffee a day have about 17% lower death rates in the same period than people who do not. @@ -225,7 +222,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Poole R, Kennedy OJ, Roderick P, et al. (2017). Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ, 359, j5024. - Notes:Disputed, and the reason is this: the original authors state that the evidence comes almost entirely from follow-up records, and they also write that 「需要可靠的随机对照试验才能判断是否为因果」 — reliable randomised controlled trials are needed to tell whether the relationship is causal. In plain terms, people have to be randomly split into two groups and compared, because records alone cannot establish cause. Drink it black, because sugar and a cream topping cancel the benefit out. This does not apply to pregnant women, or to people with arrhythmia, anxiety or insomnia. -### 28. Eat 5 servings of fruit and vegetables a day (about 400 g) +### 27. Eat 5 servings of fruit and vegetables a day (about 400 g) - Cost: A few yuan to a dozen-odd yuan a day. Washing and cutting takes some time each day. The hard part is buying every day and eating enough of it every day. - In plain language: People who eat 200 g more fruit and vegetables a day have about 10% lower death rates in the same period. People who eat 5 servings a day have about 13% lower rates than people who eat only 2. Eating more than that does not lower it further. @@ -233,15 +230,16 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Aune D et al. (2017). Fruit and vegetable intake and the risk of cardiovascular disease, total cancer and all-cause mortality: a systematic review and dose-response meta-analysis of prospective studies. International Journal of Epidemiology. ; Wang DD et al. (2021). Fruit and Vegetable Intake and Mortality: Results From 2 Prospective Cohort Studies of US Men and Women and a Meta-Analysis of 26 Cohort Studies. Circulation. - Notes: 5 servings is enough, there is no need to reach 10, and the two studies agree on that point. Note that these studies only followed records and the confounding is obvious: people who eat more vegetables differ from other people in income, education and exercise. So "about 10% lower risk" is already the ceiling. -### 29. Eat less ultra-processed food (crisps, instant noodles, cakes, instant meals) +### 28. Eat less ultra-processed food (chips, instant noodles, cakes, fast food) -- Cost: No money, you only have to shop around. Either you cook yourself, or you spend time picking out food that has not been much processed, and either way it takes time every day. The hard part is fighting the habit that convenience is better. -- In plain language: People who eat the most ultra-processed food have about 20% higher death rates in the same period than people who eat the least, and cardiovascular death is 50% higher. That evidence is graded not very certain, so read the figures with a discount. -- Benefit: A review pooled several meta-analyses on top of each other, and the original studies underneath them all only followed records with no control group. Result: people who eat a lot of ultra-processed food have a death risk about 21% higher than people who eat little of it (RR 1.21), and cardiovascular death about 50% higher (RR 1.50). The evidence grades for these two are "strong indication" and "convincing" respectively. But judged by the standard evidence grading system (GRADE), the certainty is low or very low. +- Cost: It costs nothing; you only have to shop in a different place. Either you cook yourself, or you spend time picking ingredients that are barely processed, and that takes some time every day. The hard part is fighting the habit that says "convenient is better" over the long run. +- In plain language: People who eat the most ultra-processed food have about a 20% higher chance of dying in the same period than people who eat the least, and 50% higher cardiovascular mortality. But this evidence is graded as not very certain, so discount the figures. +- Benefit: An umbrella review pooled several pooled analyses together on top of each other, and the original studies underneath those analyses all only recorded, with no groups to compare. Result: people who eat a lot of ultra-processed food have a death risk about 21% higher than people who eat little (RR 1.21), and cardiovascular mortality about 50% higher (RR 1.50). Those two are graded "highly suggestive" and "convincing" respectively. But by the usual way of grading evidence (GRADE), the certainty is low or very low. - Evidence grade: A -- Sources: Lane MM et al. (2024). Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. -- Notes:Disputed on three points. First, this classification (NOVA) lumps together foods that differ a lot in nutrition. Second, it overlaps heavily with sugary drinks and processed meat (items 7 and 19), so it is not possible to tell which of them is doing the work. Third, judged by the standard evidence grading system (GRADE), the grade is low as well. So once you have done those two items, how much extra this one adds is anyone's guess. The other side's evidence is the low rating in that same review, because no original study has yet reached the opposite conclusion. So do items 7 and 19 first (stop drinking sugary drinks, eat less processed meat), then come back to this one. -### 30. Cook and heat with electricity or gas instead of coal and firewood +- Sources: Lane MM et al (2024). Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. +- Notes: Disputed, and there are three reasons. First, this classification (NOVA) puts foods that differ a great deal in nutrition into one class. Second, it overlaps heavily with sugary drinks and processed meat (items 7 and 18), so there is no way to tell which of them is doing the work. Third, graded by the usual method (GRADE), the certainty is low as well. So once you have done those two, how much more this one adds cannot be said. The evidence on the other side is the low grade the same review gives; at the moment there is no original study with the opposite conclusion. So do items 7 and 18 first (no sugary drinks, less processed meat), and then come back to this one. + +### 29. Cook and heat with electricity or gas instead of coal and firewood - Cost: A rural household has to replace the stove and add the fuel bill, which comes to a few hundred yuan a year and sometimes over a thousand. It is a one-off change and does not take time in the long run. City dwellers are basically already on electricity or gas. - In plain language: People who cook with coal or firewood have about 10% higher death rates in the same period; people who burn coal and wood for heating are about 14% higher. People who have already switched to electricity or gas have a death rate 13%–33% lower than those still burning coal and wood. @@ -250,7 +248,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Sources: Yu K et al. (2018). Association of Solid Fuel Use With Risk of Cardiovascular and All-Cause Mortality in Rural China. JAMA. ; Chen J, Hoek G (2020). Long-term exposure to PM and all-cause and cause-specific mortality: A systematic review and meta-analysis. Environment International. - Notes: These studies only followed records. Households that switched fuel are often wealthier as well, so the figure of a risk ratio of 0.67 has the family's income mixed into it. As for outdoor PM2.5, what one person can do is limited, and it comes down to moving house, wearing a mask and using a purifier. There is no research with mortality as the endpoint for air purifiers, so this section gives no figure for them. -### 31. Let hot things cool before you drink them, and skip tea, soup and coffee that are scalding +### 30. Let hot things cool before you drink them, and skip tea, soup and coffee that are scalding - Cost: No money, just two or three minutes of waiting each time. The hard part is holding back when you want a drink. - In plain language: Drinking tea that is very hot gives you 8 times the chance of oesophageal cancer that drinking warm tea does. Even at just "hot" it is 2 times. Drinking it less than two minutes after it is poured carries 5 times the risk of waiting more than four minutes. @@ -258,7 +256,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Islami F, Pourshams A, Nasrollahzadeh D, et al. (2009). Tea drinking habits and oesophageal cancer in a high risk area in northern Iran: population based case-control study. BMJ, 338, b929. ; Loomis D, Guyton KZ, Grosse Y, et al. (2016). Carcinogenicity of drinking coffee, mate, and very hot beverages. Lancet Oncology, 17(7), 877-878. - Notes: The International Agency for Research on Cancer classifies hot drinks above 65 ℃ as group 2A, which means probably carcinogenic. The same assessment says coffee itself is not carcinogenic. In China, oesophageal cancer is common in the Chaoshan area of Guangdong and around the Taihang Mountains, and that is linked to the habit of drinking while it is still scalding -### 32. Get out in the daylight, and do not avoid the sun all day +### 31. Get out in the daylight, and do not avoid the sun all day - Cost: No money. A dozen minutes a day, on the way to work and at the lunch break, is enough. - In plain language: Women who deliberately avoid the sun have a life expectancy 0.6 to 2.1 years shorter than the women who get the most sun. People who do not smoke but avoid the sun live about as long as the heaviest smokers in the group that gets the most sun. @@ -266,7 +264,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: B - Sources: Lindqvist PG, Epstein E, Nielsen K, et al. (2016). Avoidance of sun exposure as a risk factor for major causes of death: a competing risk analysis of the Melanoma in Southern Sweden cohort. Journal of Internal Medicine, 280(4), 375-387. - Notes:Disputed, and the reason is this: Sweden lies at a high latitude with little sunshine, so the conclusion cannot be carried straight over to China. And people who avoid the sun may have been frail and housebound to begin with. Also, do not go as far as getting sunburnt. You still need to cover up under the strong midday sun, because sun exposure raises the risk of skin cancer. -### 33. Keep your BMI between 20 and 25, and lose weight if you are overweight +### 32. Keep your BMI between 20 and 25, and lose weight if you are overweight - Cost: No money. Diet and exercise both take time every day. This is the item in the whole section that demands the most willpower: controlling what you eat and moving your legs over the long run, and keeping the weight off afterwards is harder than losing it. - In plain language: The death rate is lowest in the BMI 20–25 band. At a BMI of 27.5–30 it is about 20% higher, at 30–35 about 45% higher, and at 35–40 close to double. East Asian people are more sensitive to excess weight, and every 5 points of BMI raises the risk about 40%. @@ -274,7 +272,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Global BMI Mortality Collaboration (2016). Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet. ; Flegal KM et al. (2013). Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA. (the other side of the dispute) - Notes:Disputed. One side argues that being a little heavier means living longer. Their pooled figures are that overweight people have a death risk about 6% lower, and mildly obese people about 5% lower. The two sides differ mainly over whether to drop three kinds of data: people who smoke, people who are already ill, and the first few years after entry into a study. The reason to drop them is that people who are seriously ill start to lose weight first. The analysis of the 239 studies above dropped those three groups, and the risk from being overweight came back. Note also that no randomised trial has shown that losing weight by itself lowers all-cause mortality. The figures above come from comparing people of different weights side by side, so they do not mean that losing weight lowers your risk by that much. If you want to lose weight there is no need to work on mealtimes, because breakfast and 16:8 intermittent fasting bring no extra benefit, see item 26 in section 6 (do not expect breakfast or 16:8 to help you control your weight) -### 34. Eat chillies four or more times a week +### 33. Eat chillies four or more times a week - Cost: No money, and it takes no time. - In plain language: People who eat chillies four or more times a week have about 23% lower death rates in the same period than people who rarely eat them, and cardiovascular death is about a third lower. @@ -282,7 +280,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: B - Sources: Bonaccio M, Di Castelnuovo A, Costanzo S, et al. (2019). Chili Pepper Consumption and Mortality in Italian Adults. Journal of the American College of Cardiology, 74(25), 3139-3149. - Notes:Disputed, and the reason is this: this is a follow-up record study done in the context of a Mediterranean diet, and people who eat hot food may have an overall diet closer to the local tradition. Note too that people with reflux, haemorrhoids or irritable bowel will find that hot food makes the symptoms worse, so there is no need to force yourself. -### 35. Have one or two servings of milk or yoghurt a day +### 34. Have one or two servings of milk or yoghurt a day - Cost: A few yuan a day, and it takes no time. - In plain language: People who drink 2 or more servings of milk or yoghurt a day have about 17% lower death rates in the same period than people who do not. @@ -290,7 +288,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: B - Sources: Dehghan M, Mente A, Rangarajan S, et al. (2018). Association of dairy intake with cardiovascular disease and mortality in 21 countries from five continents (PURE): a prospective cohort study. Lancet, 392(10161), 2288-2297. - Notes:Disputed, and the reason is this: PURE included a large number of middle-income and low-income countries, and in those countries being able to drink milk every day already says the family has some money, and that confounding is hard to remove cleanly. People who cannot digest lactose can switch to yoghurt or low-lactose milk instead, and there is no need to force themselves to drink ordinary milk. -### 36. You do not need to give up eggs, but do not eat three or four a day +### 35. You do not need to give up eggs, but do not eat three or four a day - Cost: No money, and it takes no time. - In plain language: Every extra half a whole egg a day raises the death rate in the same period by about 7%. Swap that half for egg white, fish, poultry or nuts and the death risk for each of those is lower instead. There is no need to give up eggs, only not to eat three or four a day. @@ -298,7 +296,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: B - Sources: Zhuang P, Wu F, Mao L, et al. (2021). Egg and cholesterol consumption and mortality from cardiovascular and different causes in the United States: A population-based cohort study. PLoS Medicine, 18(2), e1003508. - Notes:Disputed, and the dispute is a large one: there are also long-term follow-up studies of large groups of people, and pooled analyses of several studies, that found no link between eggs and mortality. The US dietary guidelines have already dropped the daily limit on cholesterol. What needs avoiding is only three or four eggs a day, and one a day is nothing to worry about. -### 37. Take a bath if you can, and do not only use a shower +### 36. Take a bath if you can, and do not only use a shower - Cost: A bit more on water and electricity. Half an hour each time. - In plain language: People who bathe almost every day have a cardiovascular event risk about 28% lower than people who bathe fewer than twice a week, and brain haemorrhage about 46% lower. @@ -307,7 +305,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Sources: Ukai T, Iso H, Yamagishi K, et al. (2020). Habitual tub bathing and risks of incident coronary heart disease and stroke. Heart, 106(10), 732-737. - Notes:Disputed, and the reason is this: the result comes from a Japanese population with a bathing culture, and most families in China have no way to bathe every day. Also, do not make the water too hot and do not stay in too long. For older people and people with cardiovascular or cerebrovascular disease, water that is too hot and a bath that is too long are dangerous in themselves, and every year a large number of older people in Japan die suddenly in the bath. -### 38. Keep the nap under half an hour, do not go past an hour, and if you need one or two hours to get through the day, find out why +### 37. Keep the nap under half an hour, do not go past an hour, and if you need one or two hours to get through the day, find out why - Cost: No money. Just set an alarm before you lie down, and it takes almost no willpower. - In plain language: People who nap for no more than an hour have no higher risk of death or chronic illness, and their minds are a little sharper. People who sleep past an hour have a coronary heart disease risk about 30% higher, and diabetes and obesity about 20% higher. So set an alarm and wake yourself at half an hour. If you genuinely need one or two hours of napping to get through the day, that is a signal to have your body checked. @@ -315,7 +313,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: B - Sources: Du P, Li J, Hua Z, et al. (2026). Multiple Health Outcomes of Daytime Napping: A Comprehensive Umbrella Review. Public Health Reviews. ; Gao C, Cai R, Zheng X, et al. (2026). Objectively Measured Daytime Napping Patterns and All-Cause Mortality in Older Adults. JAMA Network Open. ; Dashti HS et al. (2021). Genetic determinants of daytime napping and effects on cardiometabolic health. Nature Communications. - Notes:Disputed. The two studies above both only followed records, with no control group. People who nap for a long time are more likely to have sleep apnoea, anaemia, an underactive thyroid, depression or some other chronic illness to begin with. So how much of the higher death rate among long sleepers is caused by the nap itself is not clear. Causal inference using genes (Mendelian randomisation, 453,000 plus 541,000 people) supports only one thing: the more often you nap, the slightly higher your blood pressure and waist. It did not find a causal effect of napping on coronary heart disease or diabetes. So do not treat "nap less" as a way to lower your death rate. The "find out why" half of this item is the author's advice and is grade C: if you are sleepy in the daytime to the point where you must sleep for a long time, check your night-time sleep first, and check for sleep apnoea if you snore or stop breathing during sleep, then check a full blood count and your thyroid. How to get more out of a short nap is in item 11 in section 3 (sleep 10 minutes when you feel sleepy in the afternoon). How long to sleep at night is in item 13 in this section (about 7 hours a night). -### 39. Make up for a night up late the next night, do not save it up for the weekend +### 38. Make up for a night up late the next night, do not save it up for the weekend - Cost: No money. The hour or two you make up is time you wanted to sleep anyway. - In plain language: People who did not make up their sleep the next night after a short night have a death risk about 15% higher than people with a regular routine, and those who cut their sleep a lot and did not make it up are about 42% higher. People who did make it up show no rise in risk. So after a night up late, go to bed early the next night. @@ -323,16 +321,17 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: B - Sources: Li X, Zhang M, Li Z, et al. (2026). Acute sleep rebound following sleep restriction is associated with reduced mortality risk. Nature Communications. - Notes: This study only followed records, with no control group. People who cannot make up their sleep may have been busier and in worse health to begin with, and that layer cannot be removed cleanly. "Neither of the two groups that made it up showed a rise in risk" only means no difference was found: their point estimates are still above 1, and making it up does not mean nothing happened. So pay back a night up late as soon as you can, and do not use that as a licence to stay up late. The make-up in this study means sleeping a little extra on the first night after a short night, and it does not include paying everything back in one go at the weekend. Turning "stay up late on weekdays, catch up at the weekend" into a weekly habit is called social jet lag, and it is itself linked to cardiovascular disease, see item 13 in this section (about 7 hours a night and a fixed routine). The rule about a fixed wake-up time at the weekend as well is in item 2 in section 3 (fixed wake-up time, at the weekend too), and it does not clash with this item: this one tells you to go to bed early that night, not to sleep until midday the next day. -### 40. The longer you work night shifts, the higher your cardiovascular risk, so change to a day post as early as you can +### 39. The longer you work night shifts, the higher your cardiovascular risk; if you can move to another post, do it early -- Cost: Moving post or changing job may mean less money, and you lose the night shift allowance. Job hunting takes time and the decision is hard. If you do this arithmetic while you are choosing a job, the cost is zero. -- In plain language: People who work night shifts have about 13% more cardiovascular disease than people who do not, and about 27% more cardiovascular death. And it grows with the years: every extra 5 years of night shifts adds about 7% more cardiovascular disease. So this is an argument counted in years, and the earlier you change post the better. As for night shifts causing cancer, the evidence is much weaker than the stories say. -- Benefit: Pooling 23 follow-up studies, all only followed records with no control group. People who work night shifts have a cardiovascular event risk about 13% higher (RR 1.13, 95% CI 1.10–1.16, the credible range). Cardiovascular death is about 27% higher (RR 1.27, 1.18–1.36). Counted in years: every extra 5 years of night shifts raises the incidence of cardiovascular disease about 7% more (RR 1.07, 1.04–1.09), and cardiovascular death about 5% more (RR 1.05, 1.03–1.06). By disease, the incidence of coronary heart disease is about 22% higher (1.22, 1.16–1.28), death from coronary heart disease about 22% higher (1.22, 1.10–1.36), death from ischaemic heart disease about 39% higher (1.39, 1.06–1.84), and death from stroke about 49% higher (1.49, 1.04–2.12). The stroke incidence figure is not significant (1.06, 0.95–1.18). +- Cost: Moving to another post or another job may mean less money, and the night shift allowance goes with it. Job hunting takes time and the decision is hard. If you weigh this up when you are choosing a job, the cost is zero. +- In plain language: People on night shifts have about 13% more cardiovascular disease than people on other shifts, and 27% more cardiovascular death. And it grows with the years: every 5 years of night shifts adds about another 7% of cardiovascular disease. So count this account by years, and move early if you can. As for night shifts causing cancer, the evidence is much weaker than the story that goes round. +- Benefit: 23 cohort studies were pooled, only recording, with no groups to compare. People on night shifts have a cardiovascular event risk about 13% higher (RR 1.13, 95% CI 1.10–1.16, and that is the credible range). Cardiovascular death is about 27% higher (RR 1.27, 1.18–1.36). By years: every 5 years of night shifts adds about another 7% to the incidence of cardiovascular disease (RR 1.07, 1.04–1.09), and about another 5% to cardiovascular death (RR 1.05, 1.03–1.06). By disease: coronary heart disease is about 22% higher (1.22, 1.16–1.28), death from coronary heart disease about 22% higher (1.22, 1.10–1.36), death from ischaemic heart disease about 39% higher (1.39, 1.06–1.84), and death from stroke about 49% higher (1.49, 1.04–2.12). The figure for the incidence of stroke is not significant (1.06, 0.95–1.18). - Evidence grade: A -- Sources: Xi J, Ma W, Tao Y, et al. (2025). Association between night shift work and cardiovascular disease: a systematic review and dose-response meta-analysis. Frontiers in Public Health. ; Esposito G, Bravi F, Santucci C, et al. (2025). Night shift work and breast cancer risk in healthcare workers: a systematic review and meta-analysis. Occupational Medicine. ; Shen QM, Li ZY, Tan YT, et al. (2026). Night shift work and risk of total and site-specific cancer: results from a prospective cohort study among Chinese men. Scandinavian Journal of Work, Environment & Health. ; Czeisler CA, Johnson MP, Duffy JF, et al. (1990). Exposure to bright light and darkness to treat physiologic maladaptation to night work. New England Journal of Medicine, 322(18), 1253-1259. -- Notes: These are all studies that only followed records, with no control group. People who work night shifts already differ from others in smoking, weight and income, and that cannot be fully adjusted away. On night shifts causing cancer, the evidence is much weaker than the stories say. In a pooled analysis of 12 studies and 12,132 cases of breast cancer, people who have worked night shifts have a risk only about 5% higher, and the credible range crosses 1 (RR 1.05, 0.96–1.14). The group with a full 20 years is about 25% higher (1.25, 1.01–1.55), but the authors' sensitivity analysis makes that one not significant, and after correcting for possible publication bias it comes out close to 1, in their words the association is 「远未确立」 — far from established. A separate study that followed Chinese men for 16.1 years and recorded 8,202 cancers found no association between night shifts and total cancer or most common cancers, with pancreatic cancer the only exception at about 59% higher in the group with 11 to 20 cumulative years (HR 1.59, 1.09–2.31). So this item does the arithmetic on the cardiovascular side, not on cancer. Not everyone should leave a night shift post, and the calculation in years is one you have to do for yourself. If you are already on nights, change the things you can change: for quitting smoking see item 1 in this section (quit smoking, the earlier the better), for blood pressure and blood lipids see item 12 in this section (take the medicine as prescribed if you have high blood pressure or high blood lipids), and for how to make up the sleep after a night shift see item 39 in this section (make up for a night up late the next night). There are two more countermeasures worth trying: concentrating your eating in the daytime, and moving your body clock across with strong light plus strict darkness during the day. The second is the only one shown to actually move the clock. Under ordinary indoor lighting, six night shifts in a row leave the clock completely unmoved. Switch to strong light of 7000 to 12000 lux plus near-total darkness during the day, and four days will turn it fully around. **The light at night has to be bright enough, the sleep during the day has to be properly dark, and you have to wear sunglasses on the way home to block the morning light. All three are required**, and leaving out the last one makes the effort pointless. Both countermeasures only reach metabolic measures and sleep; no trial has tested whether doing them lowers your cardiovascular risk. Spending your days off living a normal daytime routine pushes the clock back. For the doses, the original figures and the rest of the sources, and for how the body tells the time, see [docs/生物钟和夜班.md](../docs/生物钟和夜班.md). +- Sources: Xi J, Ma W, Tao Y, et al (2025). Association between night shift work and cardiovascular disease: a systematic review and dose-response meta-analysis. Frontiers in Public Health. ; Esposito G, Bravi F, Santucci C, et al (2025). Night shift work and breast cancer risk in healthcare workers: a systematic review and meta-analysis. Occupational Medicine. ; Shen QM, Li ZY, Tan YT, et al (2026). Night shift work and risk of total and site-specific cancer: results from a prospective cohort study among Chinese men. Scandinavian Journal of Work, Environment & Health. ; Czeisler CA, Johnson MP, Duffy JF, et al (1990). Exposure to bright light and darkness to treat physiologic maladaptation to night work. New England Journal of Medicine, 322(18), 1253-1259. +- Notes: All of these are studies that only followed records with no control group. People on night shifts differ from other people to begin with in smoking, weight and income, and that cannot be adjusted away cleanly. On night shifts causing cancer, the evidence is much weaker than the story that goes round. In a pooled analysis of 12 studies and 12,132 breast cancer cases, people who had worked night shifts were only about 5% higher, and the credible range crosses 1 (RR 1.05, 0.96–1.14). The band with 20 full years is about 25% higher (1.25, 1.01–1.55), but after the authors ran a sensitivity analysis that item is no longer significant, and correcting for possible publication bias brings it close to 1; their own words are that the association is "far from established". A separate study following Chinese men for 16.1 years with 8,202 cancer cases found no association between night shifts and total cancer or most of the common cancers; only pancreatic cancer was higher, by about 59%, in the band with 11 to 20 years accumulated (HR 1.59, 1.09–2.31). So this item is counted on cardiovascular disease, not on cancer. Not everyone should leave a night shift post; this account, counted by years, is one you have to work out for yourself. If you are already on nights, change what you can: for quitting smoking see item 1 in this section (quitting smoking, the earlier the better), for blood pressure and lipids see item 12 in this section (if you have high blood pressure or high blood lipids, take the medicine as prescribed), and for making up the sleep after a night up late see item 38 in this section (make up for a night up late the next night). There are two other measures worth trying: keeping your eating in the daytime, and shifting your body clock with bright light plus strict darkness in the day. The second is the only one shown to shift the body clock at all. Under ordinary indoor lighting, after six nights in a row the clock does not move at all; swap that for bright light of 7000 to 12000 lux with near-total darkness in the day, and in four days it has turned over completely. **Bright enough light at night, strict blackout while you sleep in the day, and sunglasses on the way home to block the morning light — all three, with none left out**; leaving out the last one makes the whole thing wasted. Both measures stop at metabolic markers and sleep: no trial has measured whether following them lowers your cardiovascular risk. Going back to a normal daytime routine on your days off pushes the clock back again. For the doses, the original figures and the rest of the sources, and for how the body tells the time, see [docs/生物钟和夜班.md](../docs/生物钟和夜班.md). + -### 41. Buy prepackaged cooking oil with an SC number, not loose oil pressed at home in a small workshop +### 40. Buy prepackaged cooking oil with an SC number, not loose oil pressed at home in a small workshop - Cost: The same money, just a different shop. All you have to do is check whether the package carries a food production licence number beginning with SC. The hard part is persuading older relatives who are used to buying home-pressed oil. - In plain language: Peanuts that go damp and mouldy grow aflatoxin, which is a proven carcinogen and does most of its damage to the liver. After Guangzhou brought small workshops under control, the share of residents who often eat home-pressed oil with abnormal liver function fell by about 35%. In Guangxi, pregnant women who ate home-pressed peanut oil had a risk about nine tenths higher of having a baby with a low birth weight, and about four fifths higher of giving birth early. @@ -340,7 +339,7 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: B - Sources: Lei J, Li Y, Wang Y, et al. (2024). The impact of small food workshops management regulations on aflatoxin B1 in home-made peanut oil and the liver function of high-consumption area residents: an interrupted time series study in Guangzhou, China. Frontiers in Public Health, 12. ; Zhong Y, Lu H, Jiang Y, et al. (2024). Effect of homemade peanut oil consumption during pregnancy on low birth weight and preterm birth outcomes: a cohort study in Southwestern China. Global Health Action, 17. ; IARC (2012). Chemical Agents and Related Occupations. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol 100F. ; National Health and Family Planning Commission (国家卫生计生委), State Food and Drug Administration (国家食品药品监督管理总局) (2017). National Food Safety Standard, Limits for Mycotoxins in Food (GB 2761-2017) - Notes: Both studies only followed records and had no control group. The endpoint in the Guangzhou study is a liver function test, not liver cancer or death. The median toxin level in home-pressed oil is in fact below the national limit for peanut oil of 20 μg/kg, and the risk comes from eating it over many years and from occasional high batches. The national limit applies to all peanut oil, but prepackaged oil has to pass production licensing and a factory release test, while small workshops are governed by the management rules of each province and are sampled far less often. The people who benefit most from this item are you and the family you eat with. -### 42. Cook with vegetable oil instead of lard and butter, you do not have to shop around between oils for health reasons, and do not expect flaxseed oil to protect your heart +### 41. Cook with vegetable oil instead of lard and butter, you do not have to shop around between oils for health reasons, and do not expect flaxseed oil to protect your heart - Cost: It does not cost much more. Vegetable oil is already the main cooking fat in most households. The hard part is dropping the habit of lard for frying and for putting over rice. - In plain language: Cutting back on saturated fats such as lard and butter and switching to vegetable oil cuts heart disease, stroke and events like them by about 17%, but shows no change in overall death risk. There is no evidence that soybean oil, corn oil and rapeseed oil differ from each other. Eating more flaxseed oil showed no drop in heart disease risk either. @@ -348,3 +347,11 @@ This section takes in only two kinds of chronic risk: the ones that move overall - Evidence grade: A - Sources: Hooper L, Martin N, Jimoh OF, et al. (2020). Reduction in saturated fat for cardiovascular disease. Cochrane Database of Systematic Reviews, (5), CD011737. ; Hooper L, Al-Khudairy L, Abdelhamid AS, et al. (2018). Omega-6 fats for the primary and secondary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews, (11), CD011094. ; Abdelhamid AS, Brown TJ, Brainard JS, et al. (2020). Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews, (3), CD003177. ; Chinese Nutrition Society (中国营养学会) (2022). Dietary Guidelines for Chinese Residents (2022). People's Medical Publishing House - Notes:Disputed: some people argue that the amount of omega-6 in soybean oil and corn oil promotes inflammation, so you should eat less and switch oils. The randomised trials in the omega-6 review above found no harm, and the authors rated the quality of that evidence low themselves. The benefit is graded "medium" on the basis of the 17% drop in cardiovascular events, since all-cause mortality is unchanged. Whatever oil you use, control the quantity: the dietary guidelines recommend 25 to 30 grams of cooking oil per person per day. +### 42. Turn the cooker hood on when you cook and when you fry, and leave it running from the moment the flame is lit until the food is done + +- Cost: If you already have a cooker hood at home, it is only a little more electricity. If you do not, buying one and having it installed costs a few hundred yuan up to over a thousand. The hard part is remembering to switch it on every time you light the fire, and not turning it off because it is noisy. +- In plain language: Chinese women who do not smoke and who never or rarely use the cooker hood when cooking have about twice the risk of lung cancer of women who always use it. Women who have lived in a home with a poorly ventilated kitchen have a risk about 50% higher as well, and this is clearest in households that burn coal. +- Benefit: 10 case-control studies were pooled together, all of them in non-smoking Chinese women. A case-control study compares people with lung cancer against people without it and then looks back at what their habits were. People exposed to cooking fumes had about 1.74 times the chance of lung cancer (odds ratio, 95% CI 1.57–1.94, the credible range). 4 of the studies compare people exposed to the fumes who did not use a cooker hood, and that is about 2.11 times (1.54–2.89). One study in Taiwan compared 1302 non-smoking women with lung cancer against 1302 women without it. It looked only at women who had cooked for more than ten years. Among those, women who had the cooker hood on for more than two thirds of those years had a risk about half that of women who had it on for less than a third (odds ratio 0.49, 0.32–0.76). A prospective study in Shanghai followed 71,000 non-smoking women from 1996 to 2009 and found 429 cases of lung cancer. Women who said they had lived in a home with a poorly ventilated kitchen had a risk about 49% higher (HR 1.49, 1.15–1.95). Those who burned coal and had poor ventilation for more than 20 years were about 2.03 times (HR 2.03, 1.35–3.05). Looking only at whether coal had ever been burned showed no difference (HR 1.03, 0.84–1.26). Nor was there a difference by which oil was used for cooking. +- Evidence grade: B +- Sources: Xue Y, Jiang Y, Jin S, Li Y (2016). Association between cooking oil fume exposure and lung cancer among Chinese nonsmoking women: a meta-analysis. OncoTargets and Therapy, 9, 2987-2992. ; Chen TY, Fang YH, Chen HL, et al (2020). Impact of cooking oil fume exposure and fume extractor use on lung cancer risk in non-smoking Han Chinese women. Scientific Reports, 10, 6774. ; Kim C, Gao YT, Xiang YB, et al (2015). Home kitchen ventilation, cooking fuels, and lung cancer risk in a prospective cohort of never smoking women in Shanghai, China. International Journal of Cancer, 136(3), 632-638. +- Notes: Disputed: the figures supporting the cooker hood come mostly from case-control studies. Patients looking back after their diagnosis tend to under-report how often they used the cooker hood, so the figures may be too large. In the prospective Shanghai study the increase in risk fell mainly on households that burned coal with poor ventilation. How much difference the hood makes in households that only burn gas, that study cannot give a separate figure for. So the evidence grade is only B, and the benefit is graded large on the basis of a risk difference of about two times. For not cooking or heating with coal and firewood, see item 29 in this section (cook and heat with electricity or gas instead of coal and firewood). The people who benefit from this item are mainly whoever does the cooking at home. diff --git "a/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" "b/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" index 6212fbf3..2b3884e7 100644 --- "a/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" +++ "b/en/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" @@ -2,7 +2,7 @@ # 3. Do not waste energy -This section counts only willpower and time. It looks at four things: how much attention you have available each day, how fast you react, how many mistakes you make, and how long it takes to get back after being interrupted. Every figure is measured directly, either in an experiment or in records following a group of people (a cohort study). These figures are not converted into lifespan and not into mortality either. Most items in this section are graded B. That is how this field is; it does not mean the advice does not hold. Item 20 is about what to expect when dealing with police, doctors and clerks: it cites regulations rather than experiments, so it is graded C. Item 15 (treating "it will definitely get worse" thoughts as a symptom) and item 23 (treating "other people demand perfection from me" as a symptom) quote mortality and suicide-risk figures, a different measure from the rest of this section, so they are not compared with the others. +This section counts energy and time only. It looks at four things: how much attention you have available each day, how fast you react, how many mistakes you make, and how long it takes to come back after an interruption. Every number here is measured directly, either in an experiment or from records kept while a group of people is followed over time (a cohort study). These numbers are not converted into lifespan, and not converted into mortality either. Most items in this section have an evidence grade of B. That is how this field is; it does not mean the advice does not hold. Item 19 sets out the expectations you should have when you deal with police, doctors and bank clerks. It cites regulations rather than experiments, so its grade is C. Item 15 (treat pessimistic thoughts as a symptom) and item 22 (treat 「others expect me to be perfect」 as a symptom) cite mortality and suicide risk figures, which are a different measure from the rest of this section and are not compared with the others. ### 1. Turn off notifications you do not need, and keep the phone out of sight while you work @@ -12,22 +12,24 @@ This section counts only willpower and time. It looks at four things: how much a - Evidence grade: B - Sources: Stothart, Mitchum & Yehnert (2015). The attentional cost of receiving a cell phone notification. Journal of Experimental Psychology: Human Perception and Performance. ; Ward, Duke, Gneezy & Bos (2017). Brain Drain: The Mere Presence of One's Own Smartphone Reduces Available Cognitive Capacity. Journal of the Association for Consumer Research. - Notes: All of these are single tasks in a laboratory. Nobody has measured what happens when you add them up day after day. If your job really has to answer at any hour, leave notifications on for a small number of contacts. -### 2. Keep a fixed wake-up time, at weekends too +### 2. Keep a fixed wake-up time, at weekends as well -- Cost: No money. Lying in less at weekends takes some willpower. -- In plain language: In the most irregular fifth, the body clock runs about 2.5 hours later than in the most regular fifth. That amounts to living in a time zone shift for years. The more regular the schedule, the better the grades. -- Benefit: 61 university students kept records for 30 days. In the most irregular fifth, the body clock (circadian rhythm) ran about 2.5 hours later than in the most regular fifth. Concretely, melatonin started being released at 00:08 against 21:32. The period of peak sleepiness also came about 1.8 hours later. The more regular the schedule, the better the academic performance (correlation coefficient r = 0.37). +- Cost: No money. Cutting back on lying in at the weekend takes some willpower. +- In plain language: In the fifth with the most irregular daily routine, the body clock runs about 2.5 hours later than in the fifth with the most regular one. That is the same as living in a time zone shift for years. The more regular the routine, the better the grades. +- Benefit: 61 university students were recorded for 30 days. In the fifth with the least regular routine, the body clock (circadian rhythm) ran about 2.5 hours later than in the fifth with the most regular one. Concretely, the time melatonin release began was 00:08 against 21:32. The stretch of greatest sleepiness was also about 1.8 hours later. The more regular the routine, the better the academic results (correlation r = 0.37). - Evidence grade: B - Sources: Phillips et al. (2017). Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sleep/wake timing. Scientific Reports. -- Notes: There is only this one study, with few participants and no control group. The authors infer that the body clock difference comes mainly from how much daylight each group gets. A fixed wake-up time therefore only works if you also get daylight in the morning. For how to recover after an occasional late night, see item 39 in section 2 (sleep earlier the following night): go to bed early that night, not until midday the next day. +- Notes: This is the only study, with few participants and no control group. The authors work out that the difference in the body clock comes mostly from how much light people get each day. A fixed wake-up time therefore only works if you get daylight in the morning. For how to make up the occasional night up late, see item 38 in section 2 (make up for a night up late the next night): the method is to go to bed early that night, not to sleep until noon the next day. + ### 3. Sleep 7 to 8 hours a night, and do not treat 6 hours as enough -- Cost: No money. You have to free 1 to 2 hours a day for sleep. Going to bed earlier takes some willpower. -- In plain language: Two weeks of 6 hours of sleep a night drops brain performance to the level of going two days and two nights without closing your eyes. The trouble is that you cannot feel it. Self-rated sleepiness rises in the first few days and then stops rising. People on 6 hours rate themselves just as sleepy as people on 4 hours. So the sentence "I get by on 6 hours" is itself a symptom of sleep loss. -- Benefit: 48 healthy adults aged 21 to 38 slept 6 hours a night for 14 days in a row. Their cognitive performance fell cumulatively, down to the level of 2 nights with no sleep at all. Self-rated sleepiness rose only in the first few days and then barely rose again. The 6-hour group and the 4-hour group did not differ in self-rated sleepiness. This was a controlled trial that randomly assigned people to two groups (an RCT). Separately, short periods of no sleep (acute sleep deprivation, no more than 48 hours) have an effect size of g = -0.776 on mind-wandering in simple attention tasks and g = -0.125 on reasoning accuracy. That part pools 70 studies. +- Cost: No money. It adds 1 to 2 hours a day for sleep. Going to bed earlier takes some willpower. +- In plain language: Two weeks of 6 hours of sleep a night puts your brain's performance at the level of going two days and two nights without closing your eyes. The trouble is that you cannot feel it. Your own sleepiness score rises for the first few days and then stops rising: people sleeping 6 hours score exactly as sleepy as people sleeping 4 hours. So 「6 hours is enough for me」 is itself a symptom of not getting enough sleep. +- Benefit: 48 healthy adults aged 21 to 38 slept 6 hours a night for 14 days in a row. Their cognitive performance declined cumulatively, down to the level of going 2 nights with no sleep at all. Their own sleepiness scores only rose in the first few days and then barely went up again. There was no difference in self-rated sleepiness between the 6-hour group and the 4-hour group. This was a controlled trial with people randomised into two groups (RCT). Separately, short periods of no sleep (acute sleep deprivation, no more than 48 hours) have an effect on simple attention lapses of g = -0.776, and on reasoning accuracy g = -0.125. That part pools 70 studies. - Evidence grade: A - Sources: Van Dongen, Maislin, Mullington & Dinges (2003). The cumulative cost of additional wakefulness. Sleep. ; Lim & Dinges (2010). A meta-analysis of the impact of short-term sleep deprivation on cognitive variables. Psychological Bulletin. -- Notes: "I don't feel tired" is not evidence that you slept enough. People who sleep 6 hours a night for years only feel used to it; their performance has not caught up. The part that pools several studies is about staying awake for a whole night. Only one randomised controlled trial supports "sleeping less over a long period". +- Notes: 「I do not feel sleepy」 is not evidence that you have slept enough. People who sleep 6 hours a night over the long run have only got used to it in their own estimation; their performance has not caught up. The part pooling several studies is about staying awake all night. 「Sleeping less over the long run」 is supported by one randomised controlled trial. How long you sleep and the risk of death is counted separately: see item 13 in section 2 (about 7 hours a night), which is measured in mortality and finds the lowest risk in the 7-hour band. + ### 4. No caffeine after 2 p.m. - Cost: No money. If you habitually drink coffee in the afternoon, holding out takes some willpower. @@ -60,6 +62,7 @@ This section counts only willpower and time. It looks at four things: how much a - Evidence grade: B - Sources: Monsell (2003). Task switching. Trends in Cognitive Sciences. ; Ophir, Nass & Wagner (2009). Cognitive control in media multitaskers. PNAS. - Notes: How much extra time switching back and forth costs varies a lot with the kind of work, so no single figure is given here. The Ophir study is a comparison at one point in time, and it cannot separate "multitasking worsens attention" from "people with worse attention multitask more". + ### 8. No light-emitting screens in the hour before bed; read paper or an e-ink screen instead - Cost: No money. Putting the phone down in the hour before bed means fighting habit over the long run, and that needs willpower. @@ -68,7 +71,6 @@ This section counts only willpower and time. It looks at four things: how much a - Evidence grade: B - Sources: Chang, Aeschbach, Duffy & Czeisler (2015). Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. PNAS. - Notes: Few participants, in a laboratory with the conditions well controlled. At everyday phone brightness and everyday usage length the effect may be smaller. How much the phone's night mode helps was not measured in this study. - ### 9. Sleep when bedtime comes; do not stay up for games, short videos or pornography - Cost: No money. The hard part is fighting the "one more episode" habit over the long run. Keeping the phone out of the bedroom is the least effort. @@ -85,14 +87,15 @@ This section counts only willpower and time. It looks at four things: how much a - Evidence grade: B - Sources: Killgore, Balkin & Wesensten (2006). Impaired decision making following 49 h of sleep deprivation. Journal of Sleep Research. - Notes: The study covers the extreme case of two nights with no sleep at all. What happens to an ordinary person who stays up until two in the morning has not been measured directly. The "do not send messages" half is the author's own experience (grade C), and sending them the next day is not too late. -### 11. If you are sleepy in the afternoon, sleep 10 minutes, not half an hour +### 11. If you feel sleepy in the afternoon, sleep 10 minutes, not half an hour -- Cost: No money. It takes 10 minutes, plus finding somewhere to lie down or lean. -- In plain language: A 10-minute nap wakes you up on the spot, and the effect lasts two and a half hours. A 5-minute nap does almost nothing. A 20-minute nap only takes effect after half an hour. A 30-minute nap leaves you groggy on waking, and performance in that spell is worse than not sleeping at all. -- Benefit: After an insufficient night, a 10-minute nap immediately improves sleepiness, fatigue, energy and cognitive performance, and part of the benefit lasts past 155 minutes. A 5-minute nap brings almost no benefit. The benefit of a 20-minute nap appears only after 35 minutes. A 30-minute nap is followed by a stretch of sleep inertia: performance is briefly worse just after waking, and improves afterwards. +- Cost: No money. 10 minutes, plus finding somewhere you can lie down or lie on your front. +- In plain language: A 10-minute nap lifts you on the spot, and the effect holds for two and a half hours. 5 minutes is basically useless. With 20 minutes you have to wait half an hour before it works. After 30 minutes you wake up groggy first, and for that stretch your performance is worse than if you had not slept at all. +- Benefit: After a night of not enough sleep, a 10-minute nap immediately improves sleepiness, fatigue, energy and cognitive performance, and part of the benefit lasts beyond 155 minutes. A 5-minute nap has almost no benefit. With 20 minutes the benefit only appears after 35 minutes. After 30 minutes there is a stretch of sleep inertia first: performance is worse just after waking, and only improves later. - Evidence grade: B - Sources: Brooks & Lack (2006). A brief afternoon nap following nocturnal sleep restriction: Which nap duration is most recuperative? Sleep. -- Notes: One laboratory study only, and all the participants were healthy young people. If you cannot sleep, there is no need to force it. There is no research support for lying down with your eyes closed for 10 minutes. Napping for more than an hour is instead linked to higher mortality and coronary heart disease risk: see item 38 in section 2 (keep the nap under half an hour). +- Notes: This is one laboratory study, and all the participants were healthy young people. If you cannot sleep, do not force it. Whether lying down with your eyes closed for 10 minutes helps is not supported by any research. A nap of more than an hour is instead tied to higher mortality and higher coronary heart disease risk: see item 37 in section 2 (keep the nap under half an hour). + ### 12. When you work in an open-plan office, use earplugs or a quiet room for work that needs memory - Cost: Earplugs cost a few tens of yuan. Booking a quiet meeting room takes a little time. @@ -125,6 +128,7 @@ This section counts only willpower and time. It looks at four things: how much a - Evidence grade: B - Sources: Whitfield JB, Zhu G, Landers JG, Martin NG (2020). Pessimism is associated with greater all-cause and cardiovascular mortality, but optimism is not protective. Scientific Reports, 10, 12609. - Notes:Disputed. This is observation only, with no groups. Some would say that people become pessimistic because they are already ill, and the authors answer that if that were true, cancer mortality would be related as well. Unmeasured factors may still be there. Optimism had no protective effect, so there is no need to force yourself into "positive thinking". What matters is not treating a pessimistic expectation as an accurate forecast of the future. + ### 16. Cut down on the relationships that drain you, and learn to refuse requests you do not want - Cost: No money. There will be some social friction in the short term, and you may lose some relationships. Saying no takes real willpower. @@ -133,16 +137,7 @@ This section counts only willpower and time. It looks at four things: how much a - Evidence grade: C - Sources: Rook (1984). The negative side of social interaction: Impact on psychological well-being. Journal of Personality and Social Psychology. - Notes: The group surveyed is unusual, and it was a one-off questionnaire, so the order of events is not visible. What it measured is well-being, not energy. Placing it in this section is the author's inference. There is no objective standard for which relationships count as "draining", so you have to judge that yourself. - -### 17. Fix decisions of low value, such as what to wear and what to eat - -- Cost: No money. At the start you spend a little time setting the default options. -- In plain language: This entry has no hard evidence behind it. The original experiment on "making choices drains willpower" did not come out in two large replications. One of them involved two thousand people and the other three thousand, with several labs working together, and the effect was basically zero. Fixing what you wear and what to eat is still worth doing, because it saves those few minutes and those few moments of dithering, not because it "saves willpower". -- Benefit: Cannot be quantified. The original experiment claimed that making choices and exercising self-control draw on one shared limited resource (ego depletion). Two large replications have since been run, both preregistered and both carried out by several labs following the original procedure, and neither reproduced the result. One involved 23 labs and 2141 people, with effect size d = 0.04, 95% CI [-0.07, 0.15] (credible range). The other involved 36 labs and 3531 people, d = 0.06. The difference is too small to count. -- Evidence grade: C -- Sources: Baumeister, Bratslavsky, Muraven & Tice (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology. ; Hagger et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science. ; Vohs et al. (2021). A Multisite Preregistered Paradigmatic Test of the Ego-Depletion Effect. Psychological Science. -- Notes:Disputed. Popular books treat "decision fatigue" as settled, but most later replications of the original experiment did not produce the same result. It is still worth doing, because it costs almost nothing and the time it saves is real. Do not expect it to improve your judgement in the afternoon. -### 18. When you are angry, walk away first and treat the other person as weather, not as an enemy +### 17. When you are angry, walk away first and treat the other person as weather, not as an enemy - Cost: No money. You have to hold back your answer on the spot, turn and walk away, and decide afterwards whether to deal with it. That takes a little practice. - In plain language: Explaining what just happened from a different angle, or taking the other person's point of view, both bring the emotion down for real, and the second works better. Suppressing it without showing anything does not work. People who punish the offender also ruminate on it for longer afterwards; people who do not retaliate move on faster. People overestimate in advance how good revenge will feel. @@ -150,32 +145,35 @@ This section counts only willpower and time. It looks at four things: how much a - Evidence grade: B - Sources: Webb TL, Miles E, Sheeran P (2012). Dealing with feeling: A meta-analysis of the effectiveness of strategies derived from the process model of emotion regulation. Psychological Bulletin. ; Carlsmith KM, Wilson TD, Gilbert DT (2008). The paradoxical consequences of revenge. Journal of Personality and Social Psychology. - Notes: The part pooling several studies measured emotional intensity, not energy, and the effect is small to moderate. The revenge part rests on a single set of experiments, so it is graded B. The empty boat is a parable from the "Mountain Tree" chapter of the Zhuangzi: you are not angry when an empty boat drifts into you, and you are angry when there is a person aboard, yet the collision is the same. That is an analogy, not evidence. If there is real loss, take the legal route in section 8 and 9, which costs far less than losing your temper on the spot. -### 19. When you are low, do the highest-value things first: move, get daylight, sleep on time, talk to someone, call 12356 +### 18. When you feel low, do the things with the best value for money first: move, get sunlight, sleep on time, talk to somebody, call 12356 -- Cost: No money, or very little. Brisk walks, jogging or strength training several times a week. Go outside into the daylight during the day. Fix your daily routine. Make a phone call when you cannot get out of it. -- In plain language: For depression, brisk walking and jogging work best, and yoga and strength training come next; the more intense the exercise, the larger the effect. Light therapy works strongly on seasonal depression, and also works on depression that is not seasonal. Self-help cognitive behavioural courses online do work, and they work better with someone guiding you when the symptoms are heavier. If you cannot get out of it, call 12356 first. -- Benefit: Pooling 218 randomised controlled trials and 14170 people (a network meta-analysis). The comparison group was people doing other activities. Against them, walking or jogging has an effect size on depression of g = -0.62 (95% confidence interval -0.80 to -0.45). Yoga gives -0.55 and strength training -0.49. The more intense the exercise, the larger the effect. Pooling the light therapy studies: effect size 0.84 for seasonal depression (0.60 to 1.08), and 0.53 for non-seasonal depression (0.18 to 0.89). A further analysis pooled the individual patient data from the studies. It looked at internet cognitive behavioural therapy (online self-help courses). Both guided and unguided beat the comparison group. For moderate to severe cases (PHQ-9 above 9), guided did better. 12356 is the national mental health assistance hotline, open at least 18 hours a day. +- Cost: No money, or very little. A brisk walk, a jog or a strength session a few times a week. Go out and get some sunlight during the day. Fix your daily routine. When you cannot see a way out, make one phone call. +- In plain language: For depression, brisk walking and jogging work best, and the harder the exercise the larger the effect; the details are in item 6 in section 22 (when your mood is bad, walk or run). Light works very well on seasonal depression, and also helps depression that is not tied to a season. Online self-help courses based on cognitive behavioural therapy do work, and they work better with someone guiding you when the symptoms are more severe. When you cannot see a way out, call 12356 first. +- Benefit: Pooling 218 randomised controlled trials and 14170 people (a network meta-analysis). The comparison group is people doing other activities. Compared with them, walking or jogging for depression has an effect size of g = -0.62 (95% CI -0.80 to -0.45). Yoga -0.55, strength training -0.49. The harder the exercise the larger the effect. Pooling several light therapy studies: effect size in seasonal depression 0.84 (0.60 to 1.08), in non-seasonal depression 0.53 (0.18 to 0.89). A separate analysis pools the patient-by-patient data from the individual studies. It looks at internet cognitive behavioural therapy (self-help courses online). Guided and unguided, both beat the comparison group. For moderate to severe depression (PHQ-9 over 9 points), the guided version is better. 12356 is the national psychological assistance hotline, open at least 18 hours a day. - Evidence grade: A -- Sources: Noetel M et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ. ; Golden RN et al. (2005). The efficacy of light therapy in the treatment of mood disorders: a review and meta-analysis of the evidence. American Journal of Psychiatry. ; Karyotaki E et al. (2021). Internet-Based Cognitive Behavioral Therapy for Depression: A Systematic Review and Individual Patient Data Network Meta-analysis. JAMA Psychiatry. ; National Health Commission (国家卫生健康委) (2024). Notice on the use of the "12356" national unified mental health assistance hotline number (国卫医政函〔2024〕259 号). -- Notes: All of these are for mild to moderate depression. For moderate to severe depression, or if you have suicidal thoughts, see a doctor, and call 12356 first (see item 25 in section 1). The exercise trials mostly had few participants, and they cannot blind people to whether they are exercising, the way a drug trial can. The light therapy research only goes as far as 2003, and the samples were small too. "Sleep on time" and "talk to someone" are the author's own experience, with no separate reference. Do not treat "move" as a demand on people with depression; ten minutes of walking counts. -### 20. Treat police, doctors and clerks as people working to rules, not as characters: what moves a case forward is paperwork and deadlines, not emotion +- Sources: Noetel M et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ. ; Golden RN et al. (2005). The efficacy of light therapy in the treatment of mood disorders: a review and meta-analysis of the evidence. American Journal of Psychiatry. ; Karyotaki E et al. (2021). Internet-Based Cognitive Behavioral Therapy for Depression: A Systematic Review and Individual Patient Data Network Meta-analysis. JAMA Psychiatry. ; National Health Commission (国家卫生健康委) (2024). Notice on the Use of the "12356" Nationwide Unified Psychological Assistance Hotline Number, 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. +- Notes: All of these are aids for mild to moderate depression. For moderate to severe depression, or if there are suicidal thoughts, go and see a doctor, and call 12356 first (see item 25 in section 1). Most of the exercise trials have few participants, and you cannot keep secret who is exercising the way you can with a drug. The light research only goes as far as 2003, and the samples are small there too. "Sleep on time" and "talk to somebody" are the author's own experience, with no separate reference. Do not treat "get moving" as a demand made of people with depression; ten minutes of walking counts. + +### 19. Treat police, doctors and bank clerks as staff who work to rules, not as characters: what moves a case along is the paperwork and the deadlines, not your feelings -- Cost: No money. What changes is your expectations and the order of your actions. The first time, you have to resist the urge to argue the case until the other side agrees. -- In plain language: Police, doctors and clerks all work to the rules of their post, and they answer to paperwork, deadlines and supervision. Whether the person is pleasant has almost nothing to do with whether your case gets done, or how many days it takes. So ask for a receipt, note the deadline, and appeal through the proper channel, instead of trying to argue until the other side agrees. -- Benefit: Change the way you handle dealings with institutions: from "beg, get angry, and keep repeating your grievance" to "take the paperwork, note the deadline, appeal through the proper channel". What you save is the time spent on repeated conversations and repeated recollection. This is a one-off adjustment of how you think, not a daily saving of time. What you can actually check are the deadlines themselves, and they are all written into the documents. For public security cases, the time limit for handling runs from the day the case is accepted and is no more than 30 days; where the case is major or complicated, it may be extended by 30 days with the approval of the public security organ at the next level up. Where a case with a complainant is decided not to be opened, the notice of refusal must be served within 3 days. The complainant may apply for reconsideration within 7 days, and the public security organ must decide within 30 days. If still not satisfied, a further 7 days may be used to apply for review by the organ at the next level up. +- Cost: No money. What changes is your expectations and the order you do things in. The first time you have to hold back the urge to argue it out until the other person agrees with you. +- In plain language: Police, doctors and bank clerks all work to the rules of their post, and they are held to account by paperwork, deadlines and supervision. Whether the person is friendly has almost nothing to do with whether your case gets done, or how many days it takes. So take the receipt, write down the deadlines and appeal through the proper channel. Do not try to argue it out until the other person agrees with you. +- Benefit: Swap how you handle institutions: from 「begging, getting angry and going over the grievance again and again」 to 「getting the paperwork, noting the deadlines and appealing through the proper channel」. What you save is the time spent repeating yourself and replaying it in your head. This is a one-off change of outlook, not a daily saving of time. What you can check are the deadlines themselves, and they are all written into the documents. A public security case is counted from the day it is accepted, and the investigation period is not more than 30 days; where the case is major or complicated it may be extended by 30 days with the approval of the public security organ one level up. Where a case with a complainant is decided not to be opened, the written notice of refusal has to be served within 3 days. The complainant may apply for reconsideration within 7 days, and the public security organ has to decide within 30 days. If still not satisfied, apply for review at the next level up within a further 7 days. - Evidence grade: C -- Sources: Ministry of Public Security (公安部) (2020). Provisions on the Procedures for Handling Criminal Cases by Public Security Organs (公安机关办理刑事案件程序规定)（公安部令第 159 号修正后，第一百七十一、一百七十八、一百七十九条）. ; Ministry of Public Security (公安部) (2013/2019 修正). Provisions on the Procedures for Handling Administrative Cases by Public Security Organs (公安机关办理行政案件程序规定)（公安部令第 125 号、第 149 号）. -- Notes: The reason for grading C: the judgement that "many people treat the person behind the counter as a character rather than as someone at work" has no literature to cite. Only the rules half can be cited: the deadlines, the paperwork and the routes of appeal are all in writing. Every figure in this entry comes from the two Ministry of Public Security regulations cited in item 39 in section 8 (what happens after you call the police). Following the procedure will not fix an unpleasant manner, and it does not guarantee a good outcome; what it changes is your expectations and your first move. Bank clerks and government service windows are the same: their operations leave a trail, they need a second pair of eyes, and errors are held against them. So what you should do is keep the slip, ask for the receipt, complain through the proper channel, and not guess on the basis of "will he cheat me". Where the lawful order of priority is written down in the open, follow it. Emergency triage is by severity of illness, not by order of arrival (item 8 in section 24, emergency pre-triage). To thank a doctor who saved your life, use a letter of thanks and the satisfaction survey, see item 12 in section 24. If you want to "give them something so they take more care", to someone handling or enforcing a case that is bribery, and the law makes it an aggravated offence, see item 40 in section 8. If you meet obvious differential treatment, complain about it, and do not treat it as an unwritten rule to copy. +- Sources: Ministry of Public Security (公安部) (2020). Provisions on the Procedures for Handling Criminal Cases by Public Security Organs, 公安机关办理刑事案件程序规定（as amended by Ministry of Public Security Order No. 159, articles 171, 178 and 179）. ; Ministry of Public Security (公安部) (2013, as amended 2019). Provisions on the Procedures for Handling Administrative Cases by Public Security Organs, 公安机关办理行政案件程序规定（Ministry of Public Security Orders No. 125 and No. 149）. +- Notes: The reason for grading C: the judgement that 「most people treat the person behind the counter as a character rather than as somebody at work」 has no literature to cite. Only the half about the rules can be cited: the deadlines, the paperwork and the channels of review are all in the open text. Every figure quoted in this entry comes from the two Ministry of Public Security regulations cited in item 39 in section 8 (the process after calling the police). Following the procedure does not control 「a bad attitude from the other side」, and it does not guarantee that your case gets done; all it changes is your expectations and your first move. Bank clerks and government service windows are the same: what they do leaves a trail, checks are made, and mistakes are held against somebody. So what you should do is keep the receipt, ask for a printed confirmation and complain through the proper channel, instead of guessing on the basis of 「whether he might cheat me」. Where there is a lawful order of priority it is written in the open: emergency care is graded by severity, not by who came first (item 7 in section 24, emergency triage). If you want to thank a doctor who saved you, go through a letter of thanks and the satisfaction survey, see item 11 in section 24. If you are thinking of 「giving a little something so they take more care」, to somebody handling a case or enforcing the law that is the crime of bribery, and the open text sets heavier punishment for it, see item 40 in section 8. When you meet obvious differential treatment, complain about it, and do not copy it as if it were an unwritten rule. -### 21. Do not treat "how other people are doing" as daily reading: limit or switch off the apps that show people your own age + +### 20. Do not treat 「how other people are doing」 as daily reading: put a limit on the apps that show you people your own age, or turn them off -- Cost: No money. Setting a time limit on an app, or closing the account, takes a few minutes. The hard part is not opening it afterwards, which means fighting habit over the long run. -- In plain language: People who habitually compare themselves online with people who are doing better tend to be in worse mood and to have lower self-esteem. People who switched Facebook off for four weeks showed a small rise in well-being, and also a drop in what they knew about current affairs. People who limited three social platforms to 10 minutes a day each had lower loneliness and depression after three weeks. -- Benefit: Pooling several studies (a meta-analysis): 54 independent samples, 94 effect sizes, N = 36,583, using a three-level random-effects model. The average correlation between online upward social comparison and psychological maladjustment is r = 0.330 (95% CI 0.289–0.370, credible range). Broken out by outcome: social-evaluative negative emotions such as fear of being looked down on 0.438, anxiety 0.382, depression 0.306, low well-being 0.268, low self-esteem 0.263. Age, cultural background and the year the data were collected do not change the strength of the association. Studies with long-term follow-up do not give stronger results than cross-sectional surveys. A separate randomised controlled trial enrolled 2,743 US Facebook users. 61% of them were willing to accept a switch-off for less than 102 dollars. They were randomly assigned to a four-week switch-off group or a control group. Switching off freed the average participant 60 minutes a day. The subjective well-being index rose by 0.09 standard deviations. Broken out: life satisfaction 0.12, anxiety 0.10, depression 0.09, happiness 0.08. These still hold after the element of chance from multiple comparisons is removed. The author gives a reference point: psychological interventions that make people more positive. Such interventions raise subjective well-being by about +0.34 standard deviations and lower depression by about -0.23 standard deviations. So this effect is roughly 25% to 40% of that of an ordinary psychological intervention. The cost over the same period: the current-affairs knowledge index fell by 0.19 standard deviations, and the degree of polarisation on political positions fell by 0.16 standard deviations. After the experiment ended, Facebook use in the switch-off group was still 12 minutes a day below the control group (0.31 standard deviations, a drop of 23%). A further randomised controlled trial enrolled 143 undergraduates, randomly split into two groups. One group limited Facebook, Instagram and Snapchat to 10 minutes a day each, and the other carried on as before. After three weeks, loneliness and depression fell clearly in the limited group. Anxiety and fear of missing out fell from their starting levels in both groups. +- Cost: No money. Setting a time limit on an app, or closing the account, takes a few minutes. The hard part is not opening it afterwards, which means fighting the habit over the long run. +- In plain language: People who habitually compare themselves online with people who are doing better tend to be in a worse mood and to have lower self-esteem. People who turned Facebook off for four weeks had a small rise in well-being, but their knowledge of current affairs fell as well. People who limited three social platforms to 10 minutes a day each had lower loneliness and lower depression three weeks later. +- Benefit: Pooling several studies (meta-analysis): 54 independent samples, 94 effect sizes, N = 36,583, using a three-level random effects model. The mean correlation between online upward social comparison and psychological maladjustment is r = 0.330 (95% CI 0.289–0.370, credible range). Split by the particular outcome. Negative emotions of the social-evaluative kind, such as fear of being looked down on, 0.438; anxiety 0.382. Depression 0.306, low well-being 0.268, low self-esteem 0.263. Age, cultural background and the year the data were collected do not change how strong the association is. Studies that follow people over the long term do not give a stronger result than surveys taken at one point in time. A separate randomised controlled trial, with 2,743 US Facebook users as participants. 61% of them agreed to turn it off for less than 102 dollars. They were randomly assigned to four weeks off or to the comparison group. For an ordinary participant, turning it off frees 60 minutes a day. The subjective well-being index rises by 0.09 standard deviations. Broken down: life satisfaction 0.12, anxiety 0.10, depression 0.09, happiness 0.08. After taking out what multiple comparisons would have turned up by chance, these still hold. The authors give a reference point: psychological interventions that make people more positive. Such interventions give about +0.34 standard deviations on subjective well-being and about -0.23 standard deviations on depression. So this effect is about 25% to 40% of that of ordinary psychological interventions. The cost over the same period: the current affairs knowledge index falls by 0.19 standard deviations, and the degree of political polarisation falls by 0.16 standard deviations. After the experiment ended, Facebook use in the group that had turned it off was still 12 minutes a day below the comparison group (0.31 standard deviations, a fall of 23%). There is also a randomised controlled trial with 143 undergraduates as participants, randomly assigned to two groups. One group limited Facebook, Instagram and Snapchat to 10 minutes a day each; the other used them as usual. Three weeks later, loneliness and depression fell clearly in the limited group. In both groups anxiety and fear of missing out were lower than at the start. - Evidence grade: A -- Sources: Lei Y, Hu S, Sun Y, Zheng L (2026). "Looking up" linked to feeling down: a meta-analysis of online upward social comparison and psychological maladjustment. Frontiers in Psychology. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2020). The Welfare Effects of Social Media. American Economic Review, 110(3), 629–676. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2019). The Welfare Effects of Social Media. NBER Working Paper No. 25514 (the specific figures quoted in this entry were checked word for word against this version; the published abstract contains no figures). ; Hunt MG, Marx R, Lipson C, Young J (2018). No More FOMO: Limiting Social Media Decreases Loneliness and Depression. Journal of Social and Clinical Psychology, 37(10), 751–768. -- Notes: The "correlation" half only shows that two things occur together, and cannot be read as one causing the other; the authors themselves say more long-term follow-up is needed. The randomly assigned half can speak about causation, but the participants were US Facebook users just before the 2018 midterm elections, and applying that to Chinese platforms is a forced fit. The Hunt study has only 143 undergraduates, and there is only that one study. The effect sizes are all small, so do not expect switching off an app to solve a mood problem. What you need to block is the "seeing how other people are doing" kind of use. Contacting family and friends, looking things up and looking for work are not part of that. If you cannot hold the line, start by turning off notifications, turning off "people you may know" and turning off recommendations on the home page. For the total screen time, see items 15 and 16 in section 4 (a hard limit on short videos, and no doomscrolling). For what to do first when you are low, see item 19 in this section (move, get daylight, sleep on time). For using purchases to rebuild a sense of identity, see item 23 in section 6. For spending more money to "move up a notch", see item 24 in section 6. -### 22. When you think everybody saw you embarrass yourself, halve the estimate +- Sources: Lei Y, Hu S, Sun Y, Zheng L (2026). "Looking up" linked to feeling down: a meta-analysis of online upward social comparison and psychological maladjustment. Frontiers in Psychology. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2020). The Welfare Effects of Social Media. American Economic Review, 110(3), 629–676. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2019). The Welfare Effects of Social Media. NBER Working Paper No. 25514 (the specific figures quoted in this entry were checked word for word against this version; the published abstract does not contain the numbers). ; Hunt MG, Marx R, Lipson C, Young J (2018). No More FOMO: Limiting Social Media Decreases Loneliness and Depression. Journal of Social and Clinical Psychology, 37(10), 751–768. +- Notes: The 「correlation」 half only says two things turn up together; it cannot be read as one causing the other, and the authors themselves say more long-term studies are needed. The randomised half can speak about cause and effect, but the participants were US Facebook users before the 2018 midterm elections, so carrying the result over to the platforms in China is a stretch. The Hunt study has only 143 undergraduates, and it stands alone. None of the effect sizes is large, so do not expect turning off an app to settle your moods. What you have to block is the 「seeing how other people are doing」 kind of use; contacting family and friends, looking things up and job hunting are not part of it. If you cannot hold the line, start with turning off notifications, turning off 「people you may know」 and turning off recommendations on the home page. For the total screen time account, see items 15 and 16 in section 4 (a hard limit on short videos, no doomscrolling). For what to do first when you feel low, see item 18 in this section (move, get sunlight, sleep on time). For buying things to get your sense of identity back, see item 23 in section 6. For spending more to 「move up a notch」, see item 24 in section 6. + +### 21. When you think everybody saw you embarrass yourself, halve the estimate - Cost: No money. You only ask yourself one extra question when you are nervous: did they really see it? - In plain language: Walk into a room of people wearing an embarrassing T-shirt, and the wearer thinks nearly half of them can recognise who is printed on it. In fact fewer than a quarter can. In a T-shirt the wearer considers decent, the estimated number is six times the real one. People also tend to underestimate how much the other person likes them after a conversation. So when you think everybody remembers you embarrassing yourself, halving it is still too high. @@ -183,7 +181,7 @@ This section counts only willpower and time. It looks at four things: how much a - Evidence grade: B - Sources: Gilovich T, Medvec VH, Savitsky K (2000). The spotlight effect in social judgment: An egocentric bias in estimates of the salience of one's own actions and appearance. Journal of Personality and Social Psychology, 78(2), 211–222. ; Boothby EJ, Cooney G, Sandstrom GM, Clark MS (2018). The Liking Gap in Conversations: Do People Like Us More Than We Think? Psychological Science, 29(11), 1742–1756. - Notes: The reason for grading B: these are small laboratory and campus studies, in settings of wearing clothes and talking to strangers. Applied to "colleagues are talking about you" or "you said the wrong thing at the meeting", there is no direct evidence. These numbers show the direction of the overestimate and the rough multiple; they cannot be used as a coefficient. Other people's opinions are not all unimportant either. Where you really are being assessed, such as an interview, a viva, an appraisal or a medical report, the assessment is real, and you should prepare. Discount only that everyday feeling of "everybody is watching me"; what you are dealing with there is your own estimation bias. -### 23. Treat "others expect me to be perfect" as a symptom, not as a fact +### 22. Treat "others expect me to be perfect" as a symptom, not as a fact - Cost: No money. The hard part is treating such a thought as something that needs handling, rather than as an accurate measurement of what others actually require. - In plain language: People who always feel that "the people around me expect me to be perfect and will reject me if I fall short" are more likely to have suicidal thoughts. The link between "I expect myself to be perfect" and suicidal thoughts is much weaker, about a third of the first. After subtracting the thoughts people already had, the first feeling is still related to a later increase in thoughts, though the size is small. @@ -191,16 +189,17 @@ This section counts only willpower and time. It looks at four things: how much a - Evidence grade: B - Sources: Smith MM, Sherry SB, Chen S, Saklofske DH, Mushquash C, Flett GL, Hewitt PL (2018). The perniciousness of perfectionism: A meta-analytic review of the perfectionism-suicide relationship. Journal of Personality, 86(3), 522–542. - Notes: The reason for grading B: most of the studies included are one-off questionnaires, so the order of events is not visible. The correlation is moderate (r of about .28) and explains only a few percent to a tenth of the difference. What it measured is suicidal ideation and suicide attempts, not death itself, so the benefit is graded small on the grounds of "surrogate endpoint only". The authors themselves say the set of perfectionism concerns and suicide attempts looks like it has publication bias. Publication bias means studies with unflattering results do not get published. Putting the estimated missing studies back in does not change the direction of the conclusion. High standards in themselves are not the problem, and you need not lower them for that reason. The difference lies in where the standard comes from, and in whether failing to meet it makes you feel "I am finished as a person". Like the pessimistic expectations in item 15 (treat "things will definitely get worse" thoughts as a symptom), this is something that can be treated as a symptom. If you have suicidal thoughts, call 12356 first and move lethal means out of reach, see item 25 in section 1. For what you can do when someone near you shows such thoughts, see item 15 in section 8. For depression screening in children, see item 8 in section 30. -### 24. When you are angry, do not let it out by smashing things, hitting a punch bag or running a lap; slow the body down first +### 23. When you are angry, do not rely on smashing things, hitting a punching bag or running a lap to let it out; slow the body down first -- Cost: No money. In the heat of the moment, spend a few minutes on something that slows your heartbeat: breathe slowly, sit down, stay still. The hard part is that what you most want to do then is exactly to let it out. -- In plain language: Hitting a punch bag, throwing things down, going for a run — these ways of letting it out that make the body more excited do not calm anger overall. Slow breathing, mindfulness and meditation, which calm the body down, clearly reduce anger and the urge to hit someone. -- Benefit: Pooling 154 studies (a meta-analysis), 184 samples in all and 10,189 people. Methods that calm the body down, such as deep breathing, mindfulness and meditation, lower anger and aggression. Effect size g = -0.63, 95% CI -0.82 to -0.43 (credible range). The result is stable across people of different sexes, races, ages and cultures. Methods that make the body more excited, such as hitting a punch bag, jogging and cycling, have no effect overall. Their effect size is g = -0.02, 95% CI -0.13 to 0.09. The authors' conclusion is that letting anger out, and going for a run, are not effective ways to calm down. +- Cost: No money. While you are angry, you swap a few minutes for something that slows the heart rate: breathing slowly, sitting down, staying still. The hard part is that letting it out is exactly what you want to do at that moment. +- In plain language: Hitting a punching bag, smashing things and going for a run raise the body's arousal, and overall they do not take the anger away. Slow breathing, mindfulness and meditation calm the body down, and they clearly push anger and the urge to hit somebody down. +- Benefit: Pooling 154 studies (meta-analysis), 184 samples in all, 10,189 people. Methods that calm the body down, such as deep breathing, mindfulness and meditation, reduce anger and aggression. Effect size g = -0.63, 95% CI -0.82 to -0.43 (credible range). The result is stable across people of different sex, race, age and culture. Methods that raise the body's arousal, such as hitting a punching bag, jogging and cycling, have no effect overall. Their effect size is g = -0.02, 95% CI -0.13 to 0.09. The authors' conclusion is that venting anger and going for a run are not effective ways of bringing anger down. - Evidence grade: A - Sources: Kjærvik SL, Bushman BJ (2024). A meta-analytic review of anger management activities that increase or decrease arousal: What fuels or douses rage? Clinical Psychology Review. -- Notes: What it measured is anger and aggression, not energy. The measure is recorded as time because that is the author's own classification: one less row to clear up after one less argument. For something you can use on the spot, see item 9 in section 22 (cyclic sighing), and item 18 in this section (walk away when you are angry). Running is not useless. Over the long run it works for low mood: see item 7 in section 22 (walk or run when you feel down). It is just not suitable for calming down on the spot. Do not drive, do not send messages and do not make decisions while angry. If somebody is hitting you, leave first and call the police. What needs handling then is not your emotions: see item 43 in section 8 (domestic violence). +- Notes: What is measured is anger and aggression, not energy. Recording the measure as time is the author's own classification: one argument you do not have means one mess you do not have to clean up. For something you can use in the moment, see item 8 in section 22 (cyclic sighing), and item 17 in this section (when you are angry, walk away first). Running is not useless. In the long run it works on a low mood: see item 6 in section 22 (when your mood is bad, walk or run). It just does not suit being used to bring the anger down on the spot. Do not drive, do not send messages and do not make decisions while angry. If somebody is hitting you, leave first and call the police. What needs handling then is not your emotions: see item 43 in section 8 (if you are being abused at home). + -### 25. When something is weighing on you, write down your thoughts and feelings over several days, but do not expect it to cure anything +### 24. When something is weighing on you, write down your thoughts and feelings over several days, but do not expect it to cure anything - Cost: No money. The usual research design is to write 3 to 4 times, for 15 to 20 minutes each time. Write about one thing that upset you, and write your deepest thoughts and feelings about it. - In plain language: Writing a bad thing down brings a benefit on average, and it is a small one. The pooled conclusions are inconsistent, and two of the pooled analyses found no effect. It costs almost nothing, so it is worth trying. If the emotional problem is serious, do not use it instead of seeing a doctor. diff --git "a/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" "b/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" index 8f886d8b..7ccb29ed 100644 --- "a/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" +++ "b/en/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" @@ -101,14 +101,15 @@ This section counts only time. It looks at how many hours you get back per day, - Evidence grade: B - Sources: Lally P, van Jaarsveld CHM, Potts HWW, Wardle J (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998–1009. - Notes: There is only this one study, 96 people, and the actions were all simple ones (such as eating a piece of fruit after a meal). How long a complicated thing takes, such as running three times a week or writing every day, has not been tested. So still finding it hard after two weeks is not evidence of failure. To make the repetition actually happen, tie the action to a fixed setting, see item 1 in this section (write "when, where, and if this comes up then that"). -### 13. When procrastination already affects your life, use self-help materials backed by randomised trials rather than insisting on finding a therapist +### 13. When procrastination has already affected your life, use self-help material backed by randomised trials rather than going out of your way to find a therapist -- Cost: A self-help book costs a few tens of yuan, and some are free. It takes about ten weeks of work: one section to read and one exercise a week. -- In plain language: Procrastination severe enough to cause suffering can be treated. Ten weeks of working through cognitive behavioural self-help material works clearly better than doing nothing, whether someone guides you or you do it entirely on your own. Guided and unguided make no visible difference, so working from the material yourself is effective too. -- Benefit: 150 people were randomly split into three groups: guided self-help with a therapist, unguided self-help, and a waiting-list control. After ten weeks, the between-group effect size on the pure procrastination scale was d = 0.70 for the guided group (95% CI 0.29–1.10, the credible range) and d = 0.50 for the unguided group (0.10–0.90); on the irrational procrastination scale they were 0.81 (0.40–1.22) and 0.69 (0.29–1.09). The share reaching a clinically meaningful improvement: 31.3% to 40.0% in the guided group and 24.0% to 36.0% in the unguided group. The two groups did not differ significantly on any outcome. +- Cost: A self-help book costs a few tens of yuan, and some are free. You put in about ten weeks: one section to read and one exercise to do each week. +- In plain language: Procrastination bad enough to be painful can be treated. Work through the cognitive-behavioural self-help material for ten weeks. Doing it with somebody guiding you and doing it entirely on your own both beat doing nothing by a clear margin. Whether somebody guides you or not makes no measurable difference, so working through it on your own works too. +- Benefit: 150 people were randomly split into three groups: self-help with a therapist guiding, self-help with no guidance, and a waiting-list control. After ten weeks the between-group effect size on the pure procrastination scale was d = 0.70 (95% CI 0.29–1.10, the credible range) in the guided group and d = 0.50 (0.10–0.90) in the unguided group; on the irrational procrastination scale the figures were 0.81 (0.40–1.22) and 0.69 (0.29–1.09) respectively. The proportion reaching a clinically meaningful improvement: 31.3% to 40.0% in the guided group, 24.0% to 36.0% in the unguided group. The two groups did not differ significantly on any outcome. - Evidence grade: A - Sources: Rozental A, Forsell E, Svensson A, Andersson G, Carlbring P (2015). Internet-based cognitive-behavior therapy for procrastination: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 83(4), 808–824. -- Notes: The participants were Swedish adults using Swedish-language material, and there is no ready-made equivalent in Chinese, so the benefit is graded "medium". The components are the usual ones: behavioural activation, behavioural experiments, stimulus control, and material on motivation and working methods. Stimulus control is item 10 in this section (move what you would rather not touch further away). When procrastination comes with clear low mood or anxiety, start with item 19 in section 3 (move, get daylight, sleep on time, talk to someone, call 12356). +- Notes: The participants were adults in Sweden and the material was in Swedish. There is no ready-made equivalent set in Chinese, so the benefit is graded "medium". The material is made up of the usual components: behavioural activation, behavioural experiments, stimulus control, and teaching about motivation and ways of working. The stimulus control part is item 10 in this section (move what you would rather not touch further away). Where the procrastination comes with clear low mood or anxiety, start with item 18 in section 3 (move, get sunlight, sleep on time, talk to somebody, call 12356). + ### 14. Use your own hourly rate to decide which housework to outsource - Cost: You have to pay for the outsourcing, and how much depends on the service you buy. You also have to spend time once working out your own hourly rate. @@ -118,14 +119,15 @@ This section counts only time. It looks at how many hours you get back per day, - Sources: Whillans, A. V., Dunn, E. W., Smeets, P., Bekkers, R., & Norton, M. I. (2017). Buying time promotes happiness. Proceedings of the National Academy of Sciences. ; National Bureau of Statistics (国家统计局) (2024). Bulletin of the Third National Time Use Survey（第三次全国时间利用调查公报（第二号））. - Notes: Work out your hourly rate first. The formula is one month's take-home income ÷ the hours you actually work in a month. Then see what it would cost to outsource this job. Outsource it when the price is below your hourly rate and you will really spend the saved time on earning money or resting. Everyone's hourly rate and every service price differs, so work out the conclusion yourself. Also, Whillans measured whether buying time made people happy, not how many hours were saved. -### 15. Set a hard limit on short videos and mindless scrolling +### 15. Set a hard limit on short videos and on aimless scrolling -- Cost: No money. The actions take only a few minutes: turn off notifications, delete the app, or set a daily time limit on it. The hard part is keeping your hand off it and resisting the tap. -- In plain language: People in China spend 5 hours 37 minutes a day online on average. Among those who actually went online that day, it was 6 hours 3 minutes. In 2018, on the same method, it was only 2 hours 42 minutes. This is the largest figure in this section. Cutting one hour a day gives you 365 hours a year. -- Benefit: Residents of China spend 5 hours 37 minutes a day online on average. Counting only those who actually went online that day, the average is 6 hours 3 minutes. 92.9% of people used the internet on the day. These are 2024 figures. In 2018, on the same method, it was only 2 hours 42 minutes. This is the largest figure in this section. Cutting one hour a day out of it gives you 365 hours a year. Separately, according to the China Internet Network Information Center (中国互联网络信息中心), by June 2025 internet users spent an average of 30.6 hours a week online, 1.9 hours more than half a year earlier. There are 1.068 billion short video users, 95.1% of internet users. +- Cost: No money. The actions take only a few minutes: turn off notifications, delete the app, or set a daily time limit on it. The hard part is holding your hand back and not opening it. +- In plain language: People in China spend 5 hours 37 minutes a day online on average. Among those who did go online that day it was 6 hours 3 minutes. In 2018 the same calculation gave only 2 hours 42 minutes. This is the largest figure in this section. Cut one hour a day and a year gives you 365 hours back. +- Benefit: Residents of China spend 5 hours 37 minutes a day online on average. Counting only the people who did go online that day, the average is 6 hours 3 minutes. 92.9% of people used the internet that day. These are 2024 figures. In 2018 the same calculation gave only 2 hours 42 minutes. This is the largest figure in this section. Cut one hour off that every day and you gain 365 hours a year. Separately, according to the China Internet Network Information Center (中国互联网络信息中心), by June 2025 internet users spent 30.6 hours a week online, 1.9 hours more than half a year earlier. There are 1,068,000,000 short-video users, 95.1% of internet users. - Evidence grade: C -- Sources: National Bureau of Statistics (国家统计局) (2024). Bulletin of the Third National Time Use Survey（第三次全国时间利用调查公报（第二号））. ; National Bureau of Statistics (国家统计局) (2019). Bulletin on the 2018 National Time Use Survey（2018年全国时间利用调查公报）. ; China Internet Network Information Center (中国互联网络信息中心) (2025). The 56th 《中国互联网络发展状况统计报告》 (Statistical Report on the Development of China's Internet). -- Notes: Time online includes work, study and socialising, and it is not all waste. It is graded C because no study has worked out how much is saved after a limit, or where that time then goes; all that is available is the aggregate total. For the "seeing how other people are doing" part of this, randomised trials have measured how much is saved and how much mood changes: see item 21 in section 3. +- Sources: National Bureau of Statistics (国家统计局) (2024). Bulletin of the Third National Time Use Survey（第三次全国时间利用调查公报（第二号））. ; National Bureau of Statistics (国家统计局) (2019). Bulletin on the 2018 National Time Use Survey（2018年全国时间利用调查公报）. ; China Internet Network Information Center (中国互联网络信息中心) (2025). The 56th Statistical Report on the Development of China's Internet, 第 56 次《中国互联网络发展状况统计报告》. +- Notes: Time online includes work, study and socialising, so not all of it is waste. The grade is C because no study has worked out how much is saved after a limit and where that saved time goes; all that is available is the aggregate total. For the "scrolling through how other people are doing" part, a randomised trial has measured how much is saved and how mood changes, see item 20 in section 3. + ### 16. No television and no doomscrolling; take in the news you need at fixed times - Cost: No money. The hard part is breaking the habit of opening it on impulse, and you may miss some breaking news. diff --git "a/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" "b/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" index 279b1c93..bd1bb604 100644 --- "a/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" +++ "b/en/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" @@ -60,6 +60,7 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: C - Sources: People's Bank of China (中国人民银行) (2020). Notice on Promoting the Market-Based Reform of Credit Card Overdraft Interest Rates（关于推进信用卡透支利率市场化改革的通知）(银发〔2020〕327). - Notes: The 18.25% is five ten-thousandths multiplied straight by 365, with no compounding. Compounded monthly it is higher. Some banks give a discount, so go by the annualised rate printed on your own statement. When you really cannot get the cash together, look for a proper channel with a lower rate first, and do not let a minimum payment roll and gather interest. + ### 8. Do not tip streamers, do not top up games, do not buy on impulse - Cost: No money. Spend a few minutes on three things: turn off password-free payment, set a limit per payment and per day, and delete the payment methods bound to the app. The hard part is resisting the tap in the moment. @@ -84,14 +85,16 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: B - Sources: Quanzhou Anti-Fraud Centre (泉州市反诈骗中心) (2025). First Lesson of the School Term: Anti-Fraud Is a Required Course for Teenagers, and These 6 Traps Must Be Avoided! (reprinted by the Fujian Provincial Public Security Department) - Notes: It is graded B because what the authorities publish is a breakdown of fraudsters' scripts and a few cases, with no case rate and no statistics on how effective the precautions are. Keep this apart from item 9 (a child tops up and tips with a phone): money a child spends on tipping and top-ups can be reclaimed under the Civil Code. Money taken by a fraudster does not fall under that rule and can only be reported and stopped through item 2 in section 8. Whether it comes back depends on whether the money is still on the account. The core script fraudsters use with children is "you have broken the rules, your parents will go to prison". Say that plainly to your child before it happens; the sentence itself is the signal that this is a scam, and it works better than asking questions afterwards. You must actually keep your promise about "tell me first, I will not shout". Children hide it for fear of being scolded, and that is the common reason the loss grew in these cases. For the adult version of the same trick, and for deepfakes made with AI, see items 3 and 4 in section 8 (hard anti-fraud rules, hang up before any transfer). -### 11. Refuse the extended warranty on electronics +### 11. Do not buy an extended warranty for electronics (an extended service plan) -- Cost: No money. The price is that after the warranty runs out you pay for any repair yourself, and that is not very likely to happen. -- In plain language: Extended warranties are usually sold at 10% to 50% of the product's original price. What the seller makes on an extended warranty is a dozen times what it makes on an ordinary product. Best Buy in the United States made more than half its profit from extended warranties in 2003, while they were only 3% to 4% of its revenue. That gap is the money you pay for nothing. -- Benefit: Extended warranties are usually priced at 10% to 50% of the product's original price. The industry estimate quoted in the study puts the average gross margin on an extended warranty at 50% to 60%: of every 100 yuan sold, 50 to 60 yuan is profit. That is about 18 times the gross margin on an ordinary product. In 2003, extended warranties were 3% to 4% of Best Buy's revenue but produced more than 50% of its profit (United States, retail data from the 2000s) +- Cost: No money. The price is that once the warranty has run out, a repair has to be paid for out of your own pocket, though that is not very likely. +- In plain language: An extended warranty usually sells for 10% to 50% of the original price, and the seller's margin on one is more than ten times what it makes on ordinary goods; that gap is the money you pay extra. For main parts such as the motor in a washing machine or a refrigerator, the state already requires a 3-year warranty. When an extended warranty is very cheap, divide the price of the warranty by the cost of one repair: the result is how likely a breakdown at least once in those years has to be before the warranty pays for itself. +- Benefit: An extended warranty is usually priced at 10% to 50% of the original price of the goods. The industry estimates cited in the research put the average gross margin on an extended warranty at 50% to 60%: of 100 yuan taken, 50 to 60 yuan is profit. That is about 18 times the gross margin on ordinary goods. In 2003 extended warranties were only 3% to 4% of Best Buy's revenue, yet they brought in more than 50% of its profit (United States, retail data from the 2000s). When you find an extended warranty priced far below this range, do the arithmetic yourself: the price of the warranty ÷ the cost of one repair = the breakdown probability needed to break even. It is only worth buying if you think the chance of at least one breakdown in those years is higher than that figure. Take a 25 yuan warranty where one repair costs 300 yuan: the chance of at least one breakdown has to be above about 8% before it pays for itself. The 300 yuan here is only an example, not a statistic. Before you do the arithmetic, subtract the part the state already covers. The state's three-guarantee period runs from the day the invoice is issued: 1 year for a whole washing machine or refrigerator, 3 years for main parts. The main parts of a washing machine are the motor, the timer, the programme controller and the capacitor; of a refrigerator, the compressor, the thermostat and so on. Any stretch of the extended warranty that overlaps the three guarantees is money spent on something you already had (nationwide, since 1995) - Evidence grade: B -- Sources: Chen T, Kalra A, Sun B (2009). Why Do Consumers Buy Extended Service Contracts? Journal of Consumer Research 36(4):611-623. -- Notes: The gross margin figure is the paper citing someone else's industry estimate (Business Week 2004), not something the paper measured. The calculation above applies to an ordinary person who can afford one repair bill. If you depend on one device to earn a living and have no spare, work it out separately. Screen insurance for phones also needs its own calculation: first work out how often you drop the phone, then judge whether the premium is worth it. +- Sources: Chen T, Kalra A, Sun B (2009). Why Do Consumers Buy Extended Service Contracts? Journal of Consumer Research 36(4):611-623. ; Ministry of Economic and Trade (国家经贸委) and others (1995). Provisions on the Responsibility for Repair, Replacement and Return of Certain Goods（部分商品修理更换退货责任规定）(国经贸〔1995〕458 号, Article eight: the three-guarantee period runs from the day the invoice is issued; annex, Catalogue of Certain Goods Subject to the Three Guarantees (First Batch)《实施三包的部分商品目录（第一批）》: washing machines and household refrigerators, whole unit 1 year, main parts 3 years). State Administration for Market Regulation, administrative normative document. +- Notes: The gross margin figure is the paper quoting somebody else's industry estimate (Business Week 2004), not something the paper measured itself. The calculation above applies to an ordinary person who can find one repair bill out of their own pocket. If you make your living from one machine and have no spare, work it out separately. Screen insurance for a cracked phone screen is a separate calculation too: first estimate how often you drop the phone, then decide whether the premium is worth it. A very cheap extended warranty cannot cost you more than a few tens of yuan. Before buying, check three things: which day it starts from (it runs from the day you buy, so the first few years overlap the three guarantees), whether the shop, the brand or a third-party company stands behind it, and whether the call-out fee and the parts fee are charged on top. Read the manual and the warranty card first: the warranty the brand promises itself may be longer than the three guarantees. + + ### 12. Prefer generic drugs that passed the consistency evaluation and drugs picked in centralised procurement - Cost: No money. When the doctor prescribes and you go to buy, ask one extra question: "do you have the centrally procured generic version?" @@ -116,14 +119,16 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: C (disputed) - Sources: Beijing Municipal Development and Reform Commission. Residential Water Price Table. ; State Administration for Market Regulation, National Standards Full-Text Disclosure System. GB 5749-2022 Hygienic Standard for Drinking Water（生活饮用水卫生标准）. ; Hangzhou Joint Investigation Group (2025). Report on the Investigation of Abnormal Water Supply in Some Communities in Yuhang District, Hangzhou（关于杭州市余杭区部分小区供水异常调查情况通报）, of 2025-7-23, reprinted by Xinhua. ; Yuhang District People's Government, Hangzhou. Yuhang District April/May/June 2025 Raw Water Quality Monitoring Data（余杭区 2025 年 4 月／5 月／6 月水源水水质监测数据）, published by Hangzhou Yuhang Water Holdings Group, tested by Hangzhou Jiashu Environmental Monitoring Co., Ltd., annex PDF table 1 item 24「粪大肠菌群」. , , ; Cybersecurity Bureau of the Ministry of Public Security (公安部网络安全保卫局) (2025). Fabricated a false「警情通报」and a false「粪水」rumour; Shao [某] was criminally detained. - Notes:Disputed: the dispute is over what "meets the standard" means. The tests cover water leaving the plant and water at the end of the pipe. When the source or the network fails, the water can fall short for a short time. On 16 July 2025 the raw water arriving at the Renhe water plant in Yuhang, Hangzhou, smelt wrong, and tap water in the Renhe and Liangzhu communities smelled. The official conclusion was that algae broke down under low oxygen and produced sulphide compounds. After the source was changed, total coliforms and Escherichia coli were not detected in the water leaving the plant, and every item met the drinking water standard GB 5749-2022. Afterwards 7 officials were held accountable, July water charges were waived, and filters were replaced free of charge. But faecal coliforms from the same source had already been climbing for several months before the incident. At the Taoxi Yongsheng intake the April figure was 10 per litre, May 7300 per litre, June 16000 per litre. The limit for Class III surface water under GB 3838 is 10000 per litre. In the same month Siling Reservoir at 28000 per litre and Mantoushan Reservoir at 24000 per litre were also over the limit. Boiling kills microorganisms. It does not remove the smelly compounds, and it does not remove chemicals dissolved in the water. If the water smells odd or looks yellow and cloudy, stop drinking it, tell the water company and call 12345, and switch to bottled water for the time being. Boiling it a few minutes longer will not help. Even so, this was a local one-off incident and the local authority compensated by the occasion, so it still is not a reason to buy bottled water all year. The "sewer pipe was connected to the wrong place" story going round is a fabricated police notice, and whoever posted it has been criminally detained (see section 9). The 1 yuan for 500 millilitres is an assumed price set so the arithmetic works; recalculate with the real price in your own city. If the secondary water supply in an old building is substandard, you can add a filter, or buy bottled water only in that situation. Water prices differ by place; the Beijing figure is only there to show the scale. -### 15. Do not trade shares often + +### 15. Do not trade shares often, and trade even less when the market soars and crashes -- Cost: No money. The hard part is that when you see news you want to act, and you have to hold back. -- In plain language: The people who trade shares most often earned 11.4% a year on average, while the market as a whole made 17.9% in the same period. The more than six percentage points in between went almost entirely into the fees paid on buying and selling. Without the fees, frequent traders and infrequent traders earned about the same. -- Benefit: The study looked at 66,465 household accounts at US discount brokerages from 1991 to 1996. The most frequent traders had an annualised return of 11.4%, that is, they earned 11.4% a year on average. The market as a whole made 17.9% over the same period, and the average household made 16.4%. The average household turned over 75% of its holdings a year, replacing three quarters of the shares it held within the year. Gross return (what is earned before fees) barely differed between the frequent and infrequent groups, so the gap comes mainly from the trading costs paid when buying and selling (United States, 1991 to 1996) +- Cost: No money. The hard part is resisting the urge to act the moment there is news, and resisting the urge to add to your position while everybody else is making money. +- In plain language: The American households that traded most often earned 11.4% a year, while the market as a whole returned 17.9% over the same period; the gap is almost all transaction costs. In China's boom and crash of 2014 to 2015, the 85% of households with the smallest accounts traded far too often. They earned 250 billion yuan less than households that simply held on, about three tenths of the money they started with. +- Benefit: This study looked at 66,465 household accounts at a US discount brokerage, from 1991 to 1996. The most frequent traders returned 11.4% a year, which is an average of 11.4% a year. The market as a whole returned 17.9% over the same period, and the average household 16.4%. The average household turned over 75% of its holdings a year, replacing three quarters of what it held. The gross return (the money made before transaction costs) barely differed between the frequent traders and the infrequent ones, so the gap comes mainly from the transaction costs paid on the way in and out (United States, 1991 to 1996). The Chinese figures: the study used daily account data from the Shanghai Stock Exchange, covering all about 40 million accounts. The sample period ran from 2014-7 to 2015-12, 18 months in all. Over those 18 months the Shanghai Composite Index first rose more than 150%. It peaked at 5166.35 on 2015-6-12 and then fell 40% by the end of December. The household accounts were split into four groups by their market value at the start. The smallest group was 85% of all household accounts and the largest 0.5%. The two groups held a similar total value of shares at the start, 880 billion yuan and 808 billion yuan respectively. The result: the smallest 85% lost 250 billion yuan through active trading, and the largest 0.5% made 254 billion yuan. The "loss" here is measured against holding the same shares through the same period and doing nothing. Each of the two figures is about three tenths of that group's starting value of shares, the first 28% and the second 31%. As a comparison, the two and a half years before June 2014 were a calm market. Take any 18 months in that period and the same difference is only 1% to 3% of the starting value. In this boom and crash, the average household turned over its holdings about every three weeks, close to 18 times a year (Shanghai, China, 2014 to 2015; comparison period 2012 to 2014) - Evidence grade: A -- Sources: Barber BM, Odean T (2000). Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors. The Journal of Finance 55(2):773-806. -- Notes: This data is from the United States in the 1990s. The structure of commission and stamp duty on A-shares is different, but the principle is the same: the more you turn over, the less you keep. None of this is investment advice. +- Sources: Barber BM, Odean T (2000). Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors. The Journal of Finance 55(2):773-806. ; An L, Lou D, Shi D (2022). Wealth redistribution in bubbles and crashes. Journal of Monetary Economics 126:134-153. ; the working paper by the same authors, released publicly（2021-10; the figures for China were checked word for word against it）. +- Notes: The US data are from the 1990s, and the commission and stamp duty structure of A-shares is different, but the principle that the more you trade the less you keep is the same. The Chinese study does not say do not buy shares. What it works out is that over the same period, the more actively the smallest accounts trade, the more of the money passes to the other party in the transaction. The people who took that money are not institutions but the households with the largest accounts: about 90% of turnover on the Shanghai Stock Exchange in that period came from retail accounts. The "90% of retail investors lose money" figure that circulates online cannot be traced to any official statistic that can be checked, so this item does not use it. Turnover can be compared across the two: in this boom and crash Chinese household accounts turned over close to 18 times a year, while the US households turned over three quarters of their holdings once a year. In the calm years the Chinese gap is only 1% to 3%, so the time to hold back hardest is when the market is hottest. None of this is investment advice. + ### 16. Do not borrow to invest, do not use leverage, do not buy what you do not understand - Cost: No money. The hard part is that in a good market you have to resist the thought of borrowing a little to earn a little more. @@ -156,6 +161,7 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: B - Sources: Markowitz, H. (1952). Portfolio Selection. The Journal of Finance, 7(1), 77–91. - Notes: It is graded B because that classic paper is pure mathematical derivation and does not work out how much spreading saves you in money. Spreading only makes the account move less; it does not promise that you will not lose. Buying a broad market index fund (see item 17 in this section) is itself the cheapest way to spread. None of this is investment advice. + ### 20. If you pay income tax, open a personal pension account: 12000 yuan a year comes off before tax. If you pay no income tax, the account does not pay - Cost: Opening the account is free. The real cost is that the money is locked until you retire and you cannot take it out in between. @@ -212,6 +218,7 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: A - Sources: State Council (国务院) (2006, amended 2012). Regulations on Compulsory Traffic Accident Liability Insurance for Motor Vehicles（机动车交通事故责任强制保险条例）(State Council Order No. 462, as amended by State Council Order No. 618) Article twenty-three：「机动车交通事故责任强制保险在全国范围内实行统一的责任限额。责任限额分为死亡伤残赔偿限额、医疗费用赔偿限额、财产损失赔偿限额以及被保险人在道路交通事故中无责任的赔偿限额。」「责任限额由保监会会同国务院公安部门、国务院卫生主管部门、国务院农业主管部门规定。」Article twenty-one, on compensation within the limits; Article thirty-nine, on having no insurance「由公安机关交通管理部门扣留机动车……处依照规定投保最低责任限额应缴纳的保险费的2倍罚款」. ; China Banking and Insurance Regulatory Commission (中国银行保险监督管理委员会) (2020). Guiding Opinions on Implementing the Comprehensive Reform of Motor Insurance（关于实施车险综合改革的指导意见）(银保监发〔2020〕41), item four, sub-item four：「将交强险总责任限额从12.2万元提高到20万元，其中死亡伤残赔偿限额从11万元提高到18万元，医疗费用赔偿限额从1万元提高到1.8万元，财产损失赔偿限额维持0.2万元不变」. (State Council Gazette) - Notes: The regulator sets the limits and they change. These figures are the values in force since the September 2020 motor insurance reform, so check them again before you buy. What matters is the structure: compulsory insurance pays item by item, and an unused balance in one item cannot be moved to cover another. The total bill for injuring someone can go far past it, and no insurance covers the difference. This book does not give a number for how much third-party cover to buy. But it is one of the few places in this book where spending several hundred yuan removes a small chance of a loss that would crush a household. + ### 27. Save 3 to 6 months of living costs as an emergency fund, and keep it where you can withdraw at any time - Cost: No money. The price is that this money gives up a higher return and can only sit in a demand account, a money market fund, or a deposit you can withdraw from at any time. The hard part is not touching it while you build it up. @@ -252,6 +259,7 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: A - Sources: Standing Committee of the National People's Congress (2025, 3rd amendment). Food Safety Law（食品安全法）(Article one hundred and forty-eight). (National Database of Laws and Regulations, in force from 2025-12-1) ; Supreme People's Court (2024). Interpretation on Several Questions of Law Concerning the Application of Law in Cases of Disputes over Punitive Compensation for Food and Drugs（关于审理食品药品惩罚性赔偿纠纷案件适用法律若干问题的解释）(法释〔2024〕9, Articles one, six, seven, eight, twelve, thirteen, fourteen, fifteen and sixteen, in force from 2024-8-22). (Gazette of the Supreme People's Court) - Notes: The order of the evidence matters more than the amount. Photograph and video first, capturing the packaging, the production date, the foreign object and the purchase record, and only then contact the merchant or the platform. Do not throw the goods away or wash them before you have spoken to anyone. What 12315 does is mediate; it does not rule. If the merchant refuses to admit anything, suing is the only road left; for how small claims are counted, see item 22 in section 8 (online shopping and second-hand transactions where you were deceived). Also note: whether a hair or an insect found in the food counts as failing food safety standards is usually for the market regulation authority to determine or for an appraisal to settle, and it is not for you to decide. So the claim that finding one hair entitles you to 1,000 yuan does not hold. Fraud in an ordinary good is compensated at three times the price, and under 500 yuan is counted as 500 yuan; see item 29 in this section (trust the platform's rules and the statute). Selling food you make yourself has its own rules on which certificates to register for, how to write the label, and why you may not claim it treats disease; for the details see items 8, 9, 10 and 11 in section 12 (food licensing, labelling, the line on therapeutic claims, and the criminal line). The other way round, making a business out of compensation claims has clear red lines. First, deliberately manufacturing a false picture that a merchant is operating unlawfully and then using complaints or lawsuits to demand money is suspected of extortion, and the court must refer the clues and materials to the police. Second, where a lawsuit is brought on the basis of a manufactured picture, the court dismisses the claim; where it constitutes false litigation, a fine or detention is imposed according to the circumstances, and the merchant may in turn sue you for infringement of reputation. The beneficiary tier for this item is you and your family. + ### 32. Before buying anything large, check the national spot-check notices, 3C certification and the energy efficiency label - Cost: No money. It takes a few minutes on the websites of the State Administration for Market Regulation and the certification body. @@ -284,6 +292,7 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: A - Sources: State Administration for Market Regulation (市场监管总局) (2023). Guiding Principles for Regulating Business Conduct in Blind Box Sales (Trial)（盲盒经营行为规范指引（试行））(国市监稽发〔2023〕39, Articles nine, eleven, seventeen, twenty-one and twenty-three). - Notes: This document is guidance about what an operator should do; by itself it is not a basis for punishment. If something really goes wrong, it is handled under the Product Quality Law, the Law Against Unfair Competition and the Consumer Rights Protection Law. There are three things you can complain about specifically: probabilities that were not published, published probabilities that do not match reality, and an empty box. Before complaining, record the rules on the page and the unboxing on video. Exclusion from the seven-day no-questions return after the seal is broken has a precondition: the merchant gave full information, you confirmed at the time of the single purchase, and a box ticked by default cannot count as your confirmation. Selling to minors is governed separately: a merchant may not sell to a child under 8, and for a minor aged 8 and above it must first confirm that the guardian's consent has been obtained. For getting a refund when a child tops up and tips with a phone, see item 9 in this section (a child tops up and tips with a phone). For the 24-hour cooling-off period on large non-essential purchases, see item 23 in this section (set a cooling-off period on large non-essential purchases). For the cycle of "it fades once I own it, so I buy again", see item 23 in section 6 (do not expect buying things to improve your mood). + ### 36. If you care what is in the food, buy prepacked: a bulk counter is not legally required to carry an ingredient list - Cost: No money. The same thing in prepacked form usually costs a little more. Reading the ingredient list takes seconds. @@ -300,15 +309,7 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: A - Sources: Odean T (1998). Are Investors Reluctant to Realize Their Losses? The Journal of Finance 53(5):1775-1798. - Notes: This item is about whether to sell; item 15 (do not trade shares often) is about not acting all the time. The two do not conflict. If you set a rule in advance and follow it, you end up acting fewer times. Adding to the position to bring the average cost down is not wrong as arithmetic: buy a little more and the average cost is a little lower. What is wrong is treating it as a way to turn things around. Its real effect is to increase the money you have tied up in this one share, see item 19 in this section (do not put all your money into one share). The tax paragraph is worked out on the US system and does not transfer to China; the rest of the reasoning still holds. This is US account data from the 1980s and 1990s, and transaction costs and tax rules on A-shares are different. None of this is investment advice. -### 38. Act less when the market surges and crashes; do not buy more and sell more at the hottest moment - -- Cost: No money. The hard part is not piling on while everyone else is making money. -- In plain language: Someone worked out the accounts of the Shanghai Stock Exchange across the 2014 to 2015 surge and crash, covering forty million accounts. The 85% of households with the smallest accounts earned 250 billion yuan less through frequent trading than by simply holding on. That is about three tenths of the market value of their shares to begin with. Over the two and a half calm years before it, the same gap was only one to three per cent. -- Benefit: The research used daily account data from the Shanghai Stock Exchange, covering all about 40 million accounts. The sample period runs from 2014-7 to 2015-12, 18 months in all. In those 18 months the Shanghai Composite Index first rose more than 150%. It peaked at 5166.35 points on 2015-6-12 and then fell 40% by the end of December. Household accounts were sorted into four groups by their market value at the start. The smallest group was 85% of all household accounts and the largest was 0.5%. The starting market value of the shares in the two groups was about the same, 880 billion yuan and 808 billion yuan respectively. The smallest 85% lost 250 billion yuan through active trading, while the largest 0.5% made 254 billion yuan. The "loss" here is measured against the same group holding throughout and doing nothing. Each figure is about three tenths of that group's starting share value: 28% for the first and 31% for the second. For comparison, the two and a half years before June 2014 were a calm market. Taking any 18 months in that period, a comparable difference is only 1% to 3% of the starting value. In this round the average household account turned over its shares about every three weeks, close to 18 times a year (Shanghai, China, 2014 to 2015; comparison period 2012 to 2014) -- Evidence grade: A -- Sources: An L, Lou D, Shi D (2022). Wealth redistribution in bubbles and crashes. Journal of Monetary Economics 126:134-153. ; the same authors' publicly available working paper (2021-10, the figures in this item were checked against it word for word). -- Notes: The study does not say do not buy shares. It works out that within the same market episode, the more actively the smaller accounts trade, the more money moves to the other side of the trade, and the benchmark for comparison is holding on without trading. The money was not taken by institutions either, but by the individuals with the largest accounts: about nine tenths of turnover on the Shanghai Stock Exchange in that period came from retail accounts. The claim going round that "90% of retail investors lose money" has no verifiable official statistics behind it, and this item does not use it. The turnover figure is there to be compared with. In that round household accounts turned over close to 18 times a year, while the US households in item 15 (do not trade shares often) turned over only three quarters of their holdings in a year. None of this is investment advice. -### 39. If you want overseas assets such as US shares, use a QDII fund or Stock Connect; do not open an account with an overseas broker, and do not use the annual foreign exchange quota +### 38. If you want overseas assets such as US shares, use a QDII fund or Stock Connect; do not open an account with an overseas broker, and do not use the annual foreign exchange quota - Cost: No money. You can buy a QDII fund in your own bank's or broker's app in a few minutes. The price is that there are fewer products available than at an overseas broker. - In plain language: Overseas brokers such as Tiger, Futu and Longbridge are no longer allowed to open accounts for people in the mainland from May 2026. For accounts already open, within two years you may only sell and transfer the money out; you may no longer buy. The foreign exchange quota may not be used to buy overseas shares either, and anyone who breaks the rules loses the quota for that year and the 2 years after. If you want overseas assets, go through a QDII fund or Stock Connect. @@ -316,15 +317,16 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: A - Sources: China Securities Regulatory Commission (中国证监会) (2026-05-22). Eight Departments Including the CSRC Jointly Issue the Implementation Plan for the Comprehensive Rectification of Illegal Cross-Border Securities, Futures and Fund Business Activities（综合整治非法跨境证券期货基金经营活动实施方案）. ; China Securities Regulatory Commission (2026-05-22). The CSRC Official in Charge Answers Reporters' Questions on the Implementation Plan for the Comprehensive Rectification of Illegal Cross-Border Securities, Futures and Fund Business Activities. ; China Securities Regulatory Commission (2026-05-22). The CSRC Seriously Investigates Cases of Illegal Cross-Border Business by Institutions Such as Tiger. ; State Administration of Foreign Exchange (国家外汇管理局) (2021). Personal Foreign Exchange Purchase Application（个人购汇申请书）(Annex 1). - Notes: These documents govern overseas institutions and say nothing about penalties for investors themselves. If you already have an account with an overseas broker, sell within the two years and bring the money back. Each of the three lawful routes has its own conditions for opening an account, so ask your own bank or broker first. The fact that US shares rose a lot in the past does not mean they will in future. Buying a QDII fund follows the same rules as item 17 in this section (a broad market index fund) and item 19 in this section (do not put all your money into one share, one platform or one property): pick a broad index and do not put everything into one market. Cash withdrawals abroad have a separate quota; see item 6 in section 21 (withdrawing cash abroad). None of this is investment advice. -### 40. Insure only the losses you could not carry yourself; cover the ones you can carry with the emergency fund +### 39. Insure only the losses you cannot carry yourself; cover the losses you can carry with an emergency fund -- Cost: No money. What you have to do is list the large losses your household could run into and see which ones savings could not absorb. It takes half an hour. -- In plain language: Not all of the premium you pay comes back as claims; some covers the insurer's costs and profit. So what is worth buying is a loss that rarely happens but that the household could not pay for if it did: a serious illness, injuring someone while driving, the loss of the person the household depends on for income. Small losses that savings can absorb go to the emergency fund. -- Benefit: Of the premiums an insurer collects, only some is paid out as claims. The rest covers selling and running the business, and the company's profit. The regulator requires a company selling short-term personal health insurance to publish an all-in loss ratio on its website every six months. It is roughly the share of premiums paid out in claims, and if you want to know how much a company pays out you can look it up on its website. So for any one loss, insurance costs you a little money on average over the long run. What you get back is turning one large expense that could crush a household into a small amount a year. The smaller the loss and the more comfortably you can afford it, the worse the trade. By that standard, the few things this book has already written about are worth buying. One-year medical insurance: see item 20 in section 7 (one-year medical insurance). If you have a car, see item 26 in this section (third-party liability cover). If you travel abroad, see item 4 in section 21 (medical care abroad). If someone at home lives on your income, see item 41 in this section (term life insurance) (nationwide, since 2021) +- Cost: No money. What you have to do is write down the big losses your household might run into and see which ones savings cannot stand. Half an hour. +- In plain language: Not all of the premium you pay comes back to you. Part of it pays the insurer's costs and profit. So what is worth buying is a loss that is rare and that the household could not pay for if it happened: a serious illness, a car accident injuring somebody, the wage-earner in the household dying. Leave the small losses that savings can stand to the emergency fund. +- Benefit: Of the premiums insurers collect, only part is paid out as claims. The rest pays selling and running costs and the company's profit. The regulator requires companies selling personal short-term health insurance to publish an overall claims ratio on their website every six months. It is roughly the share of premiums paid out in claims; to see how much a company pays out, look it up on its own website. So for any one loss, insurance costs you a little money on average over the long run. What you get in exchange is turning one large bill that could crush a household into one small payment a year. The smaller the loss and the more comfortably you can afford it, the worse the exchange. Judged that way, several things this book has already covered are worth buying. One-year medical insurance: see item 20 in section 7 (one-year medical insurance). If you have a car, see item 26 in this section (third-party liability cover). If you travel abroad, see item 4 in section 21 (overseas medical treatment). If somebody at home lives on your income, see item 40 in this section (term life insurance) (nationwide, since 2021) - Evidence grade: C -- Sources: General Office of the former China Banking and Insurance Regulatory Commission (中国银保监会办公厅) (2021). Notice on Regulating the Business of Short-Term Health Insurance（关于规范短期健康保险业务有关问题的通知）(银保监办发〔2021〕7, all-in loss ratio to be published on the company website every six months). -- Notes: It is graded C because no official figure exists for working out which insurance is good value, so what is given here is a way to decide. Screen insurance and flight delay insurance, where the loss is a few hundred yuan and you can afford it yourself, need not be bought by this standard. Critical illness insurance pays the income lost while you cannot work through an illness, so someone whose savings can cover a year or two of not working may choose not to buy it. Basic medical insurance is social insurance; you do not apply this test to it and you pay it either way, see item 9 in section 7 (resident medical insurance). For what covers small losses, see item 27 in this section (emergency fund). The beneficiary for this item is you and your family. -### 41. If someone at home lives on your income, insure the earner first with term life insurance; do not insure the child first +- Sources: General Office of the former China Banking and Insurance Regulatory Commission (中国银保监会办公厅) (2021). Notice on Regulating the Business of Short-Term Health Insurance（关于规范短期健康保险业务有关问题的通知）(银保监办发〔2021〕7 号: the overall claims ratio is disclosed on the company's website every six months). +- Notes: It is graded C because there is no official figure that works out which kind of insurance is worth buying; what this gives is a way to weigh it. Insurance against a cracked screen, against a flight delay and the like covers a loss of a few hundred yuan that you can afford; by this standard you do not have to buy it. Critical illness insurance pays the income lost during the years you are off work with an illness; if your savings can carry you through a year or two of that, you can consider not buying it. Basic medical insurance is social insurance and is not weighed by this method: you pay it either way, see item 9 in section 7 (resident medical insurance). What small losses are covered by, see item 27 in this section (emergency fund). The beneficiaries of this item are you and your family. + +### 40. If someone at home lives on your income, insure the earner first with term life insurance; do not insure the child first - Cost: Several hundred to several thousand yuan a year, depending on age, cover and the term (the author's rough estimate, no official figure). You have to disclose your health condition truthfully before taking out the policy, which takes half an hour. - In plain language: Term life insurance covers a fixed number of years, and if the person dies in that time the household receives a sum of money. It is meant for the family that lives on your income, such as when a mortgage is still outstanding or the children are still young. If nobody depends on you, you do not need it. A child earns nothing, and the law also limits how much can be paid out on a death policy for a child. @@ -332,7 +334,8 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: C - Sources: formerly the China Insurance Regulatory Commission (中国保险监督管理委员会) (2011). Measures for the Administration of Policy Terms and Premium Rates of Personal Insurance Companies（人身保险公司保险条款和保险费率管理办法）(CIRC Order No. 3 of 2011, Articles eight and twelve). ; Standing Committee of the National People's Congress (2015, amended). Insurance Law of the People's Republic of China（中华人民共和国保险法）, Articles thirty-three and thirty-four. - Notes: It is graded C because officially there are only definitions and limits, and "insure the adult first" is worked out from the question of whose income, if it stopped, the household could not survive without. When a child is hurt, the household does not lose an income. When the earner dies, the mortgage and the cost of the children are still there. When money is limited, buy term life insurance and one-year medical insurance for the earner first, and only then the child's medical insurance. For disclosure and guaranteed renewal, see item 20 in section 7 (one-year medical insurance). The beneficiary tier for this item is tier ②, your spouse and children. -### 42. Regretting personal insurance of more than one year: cancel inside the cooling-off period and almost the whole premium comes back + +### 41. Regretting personal insurance of more than one year: cancel inside the cooling-off period and almost the whole premium comes back - Cost: No money. Call the insurer's customer service line or go to the counter. It takes a bit over ten minutes. The hard part is finding a quiet half hour within 15 days of signing to read the contract. - In plain language: Every personal insurance policy of more than one year has a cooling-off period. For a policy bought at a bank it is 15 days from the day you receive the policy, and cancelling in that period costs at most 10 yuan in handling charges. Cancel after the cooling-off period and you only get back what the policy says is refundable, which is usually less than you paid. When the insurer rings to ask whether you understood, do not say you did if you did not. @@ -340,27 +343,29 @@ This section counts money only: how much you can save, how much the yearly retur - Evidence grade: A - Sources: General Office of the former China Banking and Insurance Regulatory Commission (中国银保监会办公厅) (2019). Measures for the Administration of Insurance Agency Business of Commercial Banks（商业银行代理保险业务管理办法）(银保监办发〔2019〕179, Articles thirty-six, sixty-seven and sixty-eight). ; formerly the China Banking and Insurance Regulatory Commission (2019). Measures for the Administration of Health Insurance（健康保险管理办法）(CBIRC Order No. 3 of 2019, Article fifteen). ; formerly the China Banking and Insurance Regulatory Commission (2022). Rules on Information Disclosure for Personal Insurance Products of One Year or More（一年期以上人身保险产品信息披露规则）(银保监规〔2022〕24, the product document must state when the cooling-off period starts, how many days it runs for, and what the rights are). ; formerly the China Insurance Regulatory Commission (中国保险监督管理委员会) (2010). Basic Service Provisions for Personal Insurance Business（人身保险业务基本服务规定）(CIRC Order No. 4 of 2010, Article fifteen). ; Standing Committee of the National People's Congress (2015, amended). Insurance Law of the People's Republic of China（中华人民共和国保险法）, Articles fifteen and forty-seven. ; Ningxia Financial Regulatory Bureau (宁夏金融监管局) (2026-06-16). Risk Warning on Preventing Illegal Policy Cancellation on Behalf of Customers（关于防范非法代理退保的风险提示）. - Notes: The cooling-off period is a rule set for personal insurance of more than one year. The rules do not require it for one-year medical or accident insurance, and whether there is one depends on the terms. For a policy not bought at a bank, go by the number of days in the contract and turn to the "cooling-off period" clause first when you get the policy. Savings and participating policies involve the most money, so the cooling-off period is worth using; see item 25 in this section (buy the pure protection type first). The benefit is graded "medium" because what you save is the difference between the premium paid and the cash value, commonly several hundred to several thousand yuan, and there are no official statistics. The beneficiary for this item is you yourself. -### 43. Before you sign at a bank counter, check whether what you are holding is a deposit or an insurance policy +### 42. Before you sign at the bank counter, check whether you are holding a deposit or an insurance policy -- Cost: No money. Spend two more minutes on the front of the contract before signing. -- In plain language: A bank may sell insurance on its behalf, but it may not describe it as a deposit. The cover of an insurance contract must print "insurance contract" and the name of the insurance company in large type, and may not contain the words "deposit" or "savings". If the cover names an insurance company, the money is not a deposit, and cancelling after the cooling-off period loses you money. For someone over 65 buying at a bank, in principle only products with a certain return may be sold. -- Benefit: The rules state that a bank may not sell insurance mixed in with savings deposits, funds or bank wealth management products. It may not draw a simple analogy between the return on insurance and the return on a deposit, and may not describe an uncertain return as a guaranteed return. Insurance policy documents and promotional materials may not use the bank's name or logo, and may not contain wording such as "deposit", "savings" or "jointly launched with the bank". The cover of an insurance policy must print "insurance contract" in type of no less than 72 point, and the name of the insurance company in type of no less than second size. Three kinds of buyer may in principle be sold only products whose policy benefits are certain: a policyholder over 65 years old; a buyer of a regular-premium product who is over 60 years old; and someone whose declared annual income is below the local per-capita disposable income of residents. Policies for these three kinds of buyer must also go to the insurer for manual underwriting, and may not be issued automatically at the branch. An insurer's staff may not sell insurance inside a bank branch. Insurance sold by a bank with a term of more than one year has a 15-day cooling-off period, and cancelling after it has passed means losing money (nationwide, from 2019-10) +- Cost: No money. Spend two minutes longer on the front cover of the contract before you sign. +- In plain language: A bank is allowed to sell insurance on your behalf, but not to call it a deposit. The front cover of an insurance contract must print 「保险合同」and the name of the insurance company in large type, and must not carry the words 「存款」「储蓄」. If it is an insurance company on the cover, the money is not a deposit, and cancelling after the cooling-off period means a loss. People over 65 buying at a bank may in principle only buy products whose return is certain. +- Benefit: The rules state that a bank must not sell insurance mixed together with savings deposits, funds or bank wealth-management products. It must not draw a simple analogy between the return on insurance and the return on a deposit, and must not describe an uncertain return as a guaranteed one. Insurance documents and promotional material must not use the bank's name or logo, and must not carry words such as 「存款」「储蓄」「与银行共同推出」. The front cover of a policy must print 「保险合同」in type no smaller than 72, and the name of the insurance company in type no smaller than 二号. Three kinds of person may in principle be sold only products whose benefit is certain: a policyholder over 65; a buyer of a regular-premium product who is over 60; and a person whose declared annual income is below the local per-capita disposable income of residents. Policies for these three groups must also go to the insurer for manual underwriting, and cannot be issued automatically at the branch. An insurer's staff may not sell insurance inside a bank branch. Insurance sold through a bank with an insurance period of more than one year carries a 15-day cooling-off period, and cancelling after it has run out means a loss (nationwide, from 2019-10) - Evidence grade: B -- Sources: General Office of the former China Banking and Insurance Regulatory Commission (中国银保监会办公厅) (2019). Measures for the Administration of Insurance Agency Business of Commercial Banks（商业银行代理保险业务管理办法）(银保监办发〔2019〕179, Articles thirty, thirty-one, thirty-two, thirty-six, forty-two and forty-nine). -- Notes: It is graded B because the rules are written clearly, but there is no official figure for how much a misled person loses. This is what elderly relatives most often run into when they save at a bank, so look at the cover when you go with them. If it is already signed, cancel within 15 days under item 42 in this section (the cooling-off period). For how to complain if you were misled, see item 44 in this section (policy cancellation agents). The beneficiary for this item is you and the elderly in your family. -### 44. If you want to cancel a policy, go to the insurer yourself, not to a cancellation agent, and if you were misled complain on 12378 +- Sources: General Office of the former China Banking and Insurance Regulatory Commission (中国银保监会办公厅) (2019). Measures for the Administration of Insurance Agency Business of Commercial Banks（商业银行代理保险业务管理办法）(银保监办发〔2019〕179 号, Articles thirty, thirty-one, thirty-two, thirty-six, forty-two and forty-nine). +- Notes: It is graded B because the rules are very clear, but there is no official figure for how much the people who were misled lost. This is what older people in the household are most likely to run into when they put money in a bank; go with them and take a look at the cover. If you have already signed, cancel within 15 days under item 41 in this section (cooling-off period). For how to complain if you think you were misled, see item 43 in this section (agent policy cancellation). The beneficiaries of this item are you and the older people at home. + +### 43. If you want to cancel a policy, go to the insurance company yourself; do not use an "agent policy cancellation" service, and if you think you were misled call 12378 to complain -- Cost: No money. Calling the insurer's customer service line or going to the counter yourself takes one or two more trips than handing it to an intermediary. -- In plain language: Most of the people advertising "full refund, no fee unless we succeed" online are illegal cancellation agents. They take three to five tenths of the money that comes back, and they want your identity card, your bank card and a verification code. Once the policy is cancelled, the cover is gone as well. If you think you were misled, start with the insurer's customer service line, and if you cannot settle it call 12378. The whole process is free. -- Benefit: In 2026-2 the National Financial Regulatory Administration, the Cyberspace Administration of China, the Ministry of Public Security, the People's Bank of China and the China Securities Regulatory Commission jointly issued a risk warning. It says claims such as "full refund" are false information and are inconsistent with the Insurance Law and other laws and regulations. People offering "rights protection on your behalf" charge high service fees, and if you want to drop out halfway you could end up in court for a "breach". They also collect your SIM card, your bank card, your policy details, your home address and your child's school. If that information is sold on, you may run into telecom fraud and unauthorised charges on your card. A warning from the Ningxia Financial Regulatory Bureau in 2026-6 goes into more detail. These intermediaries agree that on a successful cancellation they will take a commission of 30% to 50% of the sum refunded, and charge a heavy penalty if you abandon it halfway. They teach people to invent reasons for a complaint, such as misleading sales or a forged signature. People who follow this may breach the Law on Public Security Administration Punishments, and in serious cases may be suspected of being an accomplice to false accusation or extortion. Once a policy is cancelled, the insurer's liability ends at once. Buying similar insurance later may cost more, or be refused, because you are older or because something has been found in your health. If you really were misled, complain to and negotiate with the insurer's official customer service line first, and if you cannot settle it call 12378, the consumer complaints and rights protection hotline for banking and insurance, which is free throughout. If you simply cannot afford the premium for a while, ask the insurer first whether they offer reduced paid-up amounts or policy loans. Reduced paid-up means you stop paying and the sum insured falls in proportion (nationwide, 2026) +- Cost: No money. Calling the insurer's customer service line yourself, or going to the counter, saves you one or two extra trips compared with handing it to an intermediary. +- In plain language: Most of the "full refund of your premium, no fee if it fails" services online are illegal agent policy cancellation. They take 30% to 50% of the money returned to you, and they want your identity card, your bank card and your verification codes. Once the policy is cancelled, the cover is gone as well. If you think you were misled, go to the insurer's customer service first; if you cannot settle it, call 12378. The whole thing is free. +- Benefit: In 2026-2 the National Financial Regulatory Administration, the Cyberspace Administration of China, the Ministry of Public Security, the People's Bank of China and the China Securities Regulatory Commission jointly issued a risk warning. It says that claims such as 「全额退保」are false information and do not match the Insurance Law and other statutes and regulations. "Rights protection agents" charge high service fees, and if you want out halfway you may end up in court for "breaching the contract". They also collect your mobile phone card, your bank card, your policy, your home address and your child's school. If that information is sold on, you may run into telecom fraud and unauthorised charges on your card. A warning from the Ningxia Financial Regulatory Bureau in 2026-6 goes into more detail. These intermediaries agree to take 30% to 50% of the amount refunded as their fee, and charge a high penalty if you drop out halfway. They teach people to make up reasons such as mis-selling or a signature taken by proxy and complain on that basis. Anyone who follows them may breach the Public Security Administration Punishments Law and, in serious cases, may be an accomplice to false accusation or extortion. Cancelling a policy ends the insurer's liability at once. Buying the same kind of insurance later may cost more, or be refused, because you are older or because something has been found in your health. If you really were misled, complain and negotiate with the insurer's official customer service first; if you cannot settle it, call 12378, the bank and insurance consumer complaint hotline, which is free of charge. If you simply cannot afford the premium this month, ask the insurer first whether it offers reduced paid-up cover or a policy loan. Reduced paid-up cover means you stop paying and the sum insured comes down in proportion (nationwide, 2026) - Evidence grade: B -- Sources: National Financial Regulatory Administration, Cyberspace Administration of China, Ministry of Public Security, People's Bank of China and China Securities Regulatory Commission (2026-02-06). Risk Warning on Guarding Against Traps in Short Videos and Livestreams by Illegal "Rights Protection on Your Behalf"（关于警惕不法「代理维权」短视频及直播陷阱的风险提示）(reprinted on the CAC website). ; Ningxia Financial Regulatory Bureau (宁夏金融监管局) (2026-06-16). Risk Warning on Preventing Illegal Policy Cancellation on Behalf of Customers（关于防范非法代理退保的风险提示）. ; Hunan Financial Regulatory Bureau (湖南金融监管局) (2025-07-23). Risk Warning on Preventing Illegal "Policy Cancellation" Grey and Black Industries（关于防范非法「代理退保」黑灰产的风险提示）. -- Notes: The Hunan Financial Regulatory Bureau reported a case in 2025. A woman was persuaded by a "friend" to use an agent to cancel her policy and paid a "consulting fee"; in the end the policy was still cancelled at its cash value. Three months after cancelling, she was diagnosed with breast cancer, and the critical illness cover was gone. Whether to cancel is for you to work out: if you still need the cover and cannot pay, ask about reduced paid-up first. To cancel inside the cooling-off period yourself, see item 42 in this section (the cooling-off period). The beneficiary for this item is you yourself. -### 45. Name a specific person as the beneficiary on the policy, and claim the money when something happens: within 5 years for life insurance and 2 years for other insurance +- Sources: National Financial Regulatory Administration, Cyberspace Administration of China, Ministry of Public Security, People's Bank of China, China Securities Regulatory Commission (2026-02-06). Risk Warning on Unlawful "Rights Protection" Short Videos and Livestream Traps（关于警惕不法「代理维权」短视频及直播陷阱的风险提示）, reprinted on the Cyberspace Administration of China's website. ; Ningxia Financial Regulatory Bureau (宁夏金融监管局) (2026-06-16). Risk Warning on Preventing Illegal Policy Cancellation on Behalf of Customers（关于防范非法代理退保的风险提示）. ; Hunan Financial Regulatory Bureau (湖南金融监管局) (2025-07-23). Risk Warning on Preventing the Illegal "Agent Policy Cancellation" Grey Market（关于防范非法「代理退保」黑灰产的风险提示）. +- Notes: The Hunan Financial Regulatory Bureau reported a case in 2025. A woman was persuaded by a "friend" to go through an agent policy cancellation and paid a "consulting fee"; in the end the policy was still cancelled at its cash value. Three months after the cancellation she was diagnosed with breast cancer, and the critical illness insurance was already gone. Whether to cancel is yours to work out: if the cover is still worth having and you cannot pay, ask about reduced paid-up cover first. If you are cancelling inside the cooling-off period yourself, see item 41 in this section (cooling-off period). The beneficiary of this item is you yourself. + +### 44. Name a specific person as the beneficiary on the policy, and claim the money when something happens: within 5 years for life insurance and 2 years for other insurance - Cost: No money. Fill it in once when you take out the policy; to change the beneficiary later you must notify the insurer in writing. - In plain language: If no beneficiary is named, or the beneficiary has died first, the sum insured counts as the estate and has to be used to pay the dead person's debts first. If the beneficiary is written as "wife, [name]" and you later divorce, at the time it counts as though nothing was named. The people in your family need to know the policy exists: for life insurance you have 5 years from learning of the death to claim, and 2 years for other insurance. - Benefit: The Insurance Law provides that the beneficiary is designated by the insured person or the policyholder. Where the policyholder designates, the insured person's consent is required. There are three situations in which the sum insured counts as the estate of the insured person and is distributed under the rules of inheritance. First, no beneficiary was designated, or the designation is unclear. Second, the beneficiary died before the insured person and there is no other beneficiary. Third, the beneficiary has lawfully lost or given up the right to the benefit and there is no other beneficiary. Where the beneficiary and the insured person died in the same event and the order cannot be established, the beneficiary is presumed to have died first. Money counted as the estate must first clear the taxes and debts of the dead person, and only what is left is divided. The Supreme People's Court's interpretation states that where the beneficiary is written as "name plus relationship" and the relationship has changed by the time it is claimed, that counts as no beneficiary having been designated. Changing the beneficiary without notifying the insurance company does not take effect against the insurer. The statute of limitations for claiming the money is 5 years for life insurance and 2 years for other insurance, both counted from the day you knew or should have known of the event. The sum insured on life insurance is usually in the hundreds of thousands of yuan, so the benefit is graded large (nationwide, 2015 amendment, in force since 2021) - Evidence grade: A - Sources: Standing Committee of the National People's Congress (2015, amended). Insurance Law of the People's Republic of China（中华人民共和国保险法）, Articles twenty-six, thirty-nine, forty-one and forty-two. ; Supreme People's Court (2020, amended). Interpretation on Several Questions Concerning the Application of the Insurance Law of the People's Republic of China, Part Three（关于适用《中华人民共和国保险法》若干问题的解释（三））, Article nine, paragraph two, item three, and Article ten. ; National People's Congress (全国人大) (2020). Civil Code of the People's Republic of China（中华人民共和国民法典）, Article one thousand one hundred and fifty-nine. -- Notes: Once the statute of limitations has passed the insurer can refuse to pay, and it is hard to get the money back through a lawsuit. So at least the people in your family should know which companies you hold insurance with. After a divorce, a remarriage or the birth of a child, go back and change the beneficiary once. After someone dies, collect the money from each place one by one; see item 9 in section 25 (collecting from each place). Paying off debts with the proceeds of a life insurance policy is not a route; see item 13 in section 29 (paying off debts). The beneficiary tier for this item is tier ②, your spouse and children. \ No newline at end of file +- Notes: Once the statute of limitations has passed the insurer can refuse to pay, and it is hard to get the money back through a lawsuit. So at least the people in your family should know which companies you hold insurance with. After a divorce, a remarriage or the birth of a child, go back and change the beneficiary once. After someone dies, collect the money from each place one by one; see item 9 in section 25 (collecting from each place). Paying off debts with the proceeds of a life insurance policy is not a route; see item 13 in section 29 (paying off debts). The beneficiary tier for this item is tier ②, your spouse and children. diff --git "a/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" "b/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" index 93655d1e..50e884e3 100644 --- "a/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" +++ "b/en/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" @@ -44,6 +44,7 @@ This section collects the things that look cheap and effective. Open the origina - Evidence grade: A - Sources: Clegg DO et al. (2006). Glucosamine, chondroitin sulfate, and the two in combination for painful knee osteoarthritis. NEJM. - Notes:Disputed: in the same trial, among the people who hurt most at the start, 79.2% of those taking both improved against 54.3% in the comparison group (P=0.002). The authors say this was found after the fact in the data, so it can only be a lead, not a conclusion. The approaches to knee osteoarthritis that really have evidence behind them are losing weight and strengthening muscle; see section 2. + ### 6. Do not rely on vitamin C to prevent colds - Cost: An effervescent tablet costs about 1 to 2 yuan, 300 to 700 yuan a year. @@ -76,14 +77,15 @@ This section collects the things that look cheap and effective. Open the origina - Evidence grade: B - Sources: Smith-Spangler C et al. (2012). Are organic foods safer or healthier than conventional alternatives?: a systematic review. Ann Intern Med. - Notes: Very few studies look directly at how people's health turns out after eating it. So the conclusion is "there is no evidence that it is better", not "it has been proved to have no benefit". The effect on the environment, on animal welfare and on taste are not covered here. Eating more fruit and vegetables does work, and whether they are organic makes no difference to that. -### 10. Do not spend a lot on health products, herbal formulae or tonics to "condition your body" +### 10. Do not spend a lot of money on health supplements and other health products, herbal pastes or tonics to "look after your health" -- Cost: Anything from a few hundred yuan to more than ten thousand. Once it is put into gift packaging the price often multiplies several times over. -- In plain language: The department in charge of health foods requires the package to carry one sentence: a health food is not a medicine and cannot replace medicine in treating disease. The common ingredients, multivitamins, fish oil, vitamin D, have already been checked one by one in the first six items of this section, and none showed a benefit. Most of this money is spent on buying something to carry to your parents; the thought is not the problem, the product is simply the wrong one. -- Benefit: The State Administration for Market Regulation requires that the front of a health food package carry, in an area of no less than 20%, the sentence「保健食品不是药物，不能代替药物治疗疾病」. The same rules pin down what a health food is for, to「补充膳食营养物质、维持改善机体健康状态或者降低疾病发生风险因素」. The regulator itself does not accept that it treats disease. The ingredients it most often contains, multivitamins, fish oil, vitamin D, antioxidants, glucosamine and vitamin C, and what each of them gave in randomised controlled trials, have already been listed item by item in the first 6 entries of this section. It looks like good value because "it does not cure, it only conditions you" cannot be called wrong however you put it, and nobody can prove it false. And when you hand this over as a gift, the money comes from the relationship, not from evidence +- Cost: From several hundred yuan to more than 10,000 yuan. Once it is packaged as a gift set, the price often goes up several times over. +- In plain language: The common ingredients — multivitamins, fish oil, vitamin D — have each been checked one by one in the first six items of this section, and none of them showed a benefit. The sentence 「保健食品不是药物，不能代替药物治疗疾病」on the pack is one the authority in charge of health foods requires to be printed. This money is usually spent on buying something for your parents and carrying it to their door. The thought behind it is not the problem; the choice of product is. +- Benefit: What you get back is the money you did not spend. The reasons are in the first 6 items of this section: the commonest ingredients in health products — multivitamins, fish oil, vitamin D, antioxidants, glucosamine, vitamin C — have shown none of the expected benefits in randomised controlled trials, and the figures for each are in the Benefit field of those 6 items. The regulator's rules only say what a product may claim; they are not evidence that it does nothing. The State Administration for Market Regulation requires the front of a health food package to carry one sentence in no less than 20% of the area: 「保健食品不是药物，不能代替药物治疗疾病」. The same rules also pin down what a health food may be used for, limiting it to 「补充膳食营养物质、维持改善机体健康状态或者降低疾病发生风险因素」. It looks like good value because "it does not cure, it only looks after you" can never be called wrong, and nobody can prove otherwise. And when you spend this money on a gift, what you are relying on is affection, not evidence - Evidence grade: C -- Sources: State Administration for Market Regulation (国家市场监督管理总局) (2019). Press Conference Held by the State Administration for Market Regulation on the Guide to Warning Wording on Health Food Labels and the Measures for the Administration of the Catalogue of Health Food Ingredients and Health Functions. -- Notes: The conclusion is "not supported by evidence", not "proved to be ineffective", so this is only graded C. For a product with the ingredients written out, look up the corresponding study in the earlier items. People who come up genuinely deficient, or who are being treated, follow what the doctor says. What to reject is the product, not the thought. Where an older relative bought it themselves and is already taking it, and it has not displaced the medicine they are on and has caused no harm, there is no need to turn up and argue the point; arranging the vaccinations they should have and the screenings they are due for matters more. For the same money, the following all have a source in this book. Taking an older relative for their yearly flu shot: see item 20 in section 1. The shingles vaccine after 50: see item 21 in section 1. The pneumococcal vaccine after 65: see item 22 in section 1. Buying a blood pressure monitor and keeping them on their blood pressure tablets until the course is finished and the pressure is at target: see item 7 in section 1. Altering the bathroom and stairs at home and practising balance and leg strength with them: see item 13 in section 1. Cancer screening that comes round at their age, take them for one: see items 17, 18 and 19 in section 1. Where an older relative has been bedridden for a long time, pressure sore prevention and long-term care insurance: see items 7 and 8 in section 17. If you want to carry something over, fruit, rice and flour, or something they actually like, is more use than a herbal formula. The two things really worth stopping are this: one is retirement investment schemes where you pay money up front, and the other is products that displace medicine; see item 5 in section 17 +- Sources: State Administration for Market Regulation (国家市场监督管理总局) (2019). Press Conference Held by the State Administration for Market Regulation on the Guidelines for Warning Wording on Health Food Labels and the Administrative Measures for the Catalogue of Health Food Raw Materials and Health Functions. +- Notes: The conclusion is "there is no evidence in favour", not "it has been proved ineffective", so it is only graded C. For a product that lists its ingredients, look up the matching study as in the earlier items. If you really are short of a nutrient, or you are being treated for something, do what the doctor says. What should be rejected is the product, not the thought. If older people bought it themselves and are already taking it, as long as it has not replaced a medicine they are taking and has caused no harm, there is no need to go and argue the point; arranging the vaccinations and the screening they are due is more important. The same money spent on the following is better value, and every one of them can be traced in this book. Going with an older person for the annual flu vaccine: see item 20 in section 1. The shingles vaccine after the age of 50: see item 21 in section 1. The pneumococcal vaccine over the age of 65: see item 22 in section 1. Buying a blood pressure monitor and keeping them on their prescribed blood pressure medicine until their readings are at target: see item 7 in section 1. Modifying the bathroom and stairs at home and practising balance and leg strength with them: see item 13 in section 1. Cancer screening at the age for it: go with them once, see items 17 to 19 in section 1. For an older person already bedridden, pressure ulcer prevention and long-term care insurance: see items 7 and 8 in section 17. If you are bringing something to their door, fruit, rice and flour, or something they like to eat, is more use than a herbal paste. The two things really worth stopping are these: investing in a retirement plan where you pay money up front, and anything that would replace a medicine they are taking; see item 5 in section 17 + ### 11. Do not take probiotics long term when you have no bowel problem - Cost: About 2 to 5 yuan a day, 700 to 1800 yuan a year. @@ -92,6 +94,7 @@ This section collects the things that look cheap and effective. Open the origina - Evidence grade: B - Sources: Khalesi S et al. (2019). A review of probiotic supplementation in healthy adults: helpful or hype? Eur J Clin Nutr. - Notes: The same review does report a few small improvements in healthy people, such as more formed and suitable stools and more lactobacilli in women's vaginas. Diarrhoea caused by antibiotics, and one particular strain used for one particular condition, each have their own evidence and have to be judged separately. This item is only about healthy people taking it daily as a supplement. + ### 12. Do not force yourself into cold showers in the name of "strengthening immunity" - Cost: No money. Every shower is uncomfortable for 30 to 90 seconds, and keeping it up through winter takes willpower. People with cardiovascular disease also have to be careful, because cold stimulation is a risk for them. @@ -124,14 +127,16 @@ This section collects the things that look cheap and effective. Open the origina - Evidence grade: B - Sources: Forer BR (1949). The fallacy of personal validation: A classroom demonstration of gullibility. Journal of Abnormal and Social Psychology 44(1):118-123（表 1 评分分布：2 分 1 人、3 分 4 人、4 分 18 人、5 分 16 人，N=39）. ; Carlson S (1985). A double-blind test of astrology. Nature 318:419-425（表 2：首选 116 份，挑中 40 份，随机期望 38.5，占星师预计 58.5；「The data are consistent with chance, inconsistent with astrological hypothesis」）. , full text on the second author's homepage ; Supreme People's Procuratorate (最高人民检察院) (2026). Xixian, Henan: Three Years of Continuous Tracking Taken to a Batch of Online Fortune-Telling Fraud Gangs（河南息县：历经三年持续追踪打掉一批网络算命诈骗团伙）. - Notes: Each of those two is a single study. Forer's paper does not state the average score directly; the 4.26 is worked out from the distribution in table 1. Nothing here says it predicts accurately, and no belief is being judged. Spending a few tens of yuan on it as entertainment is not what is at issue; the losing move is using it to decide about leaving a job, getting married or investing. -### 16. Do not buy blue-light glasses to "protect your eyesight", and do not believe that months in front of a screen ruins your eyes; but treat a swollen, red eye as an emergency +### 16. Do not buy blue-light glasses to "protect your eyesight", and do not believe that "your eyes are ruined after a few months in front of a screen"; but a swollen, red eye is an emergency - Cost: No money, and you also save the several hundred to several thousand yuan added to the price of the lenses. -- In plain language: Blue-light lenses do not work, and not buying them saves you several hundred to several thousand yuan. Compared with ordinary lenses, people wearing them report the same amount of eye tiredness and see just as clearly, and no evidence has been found that they protect the back of the eye. The strain and dryness that come from looking at a screen do ease off. What really damages the eye is the arc light of a welding torch, ultraviolet light, lasers, and an acute attack of glaucoma. -- Benefit: A Cochrane systematic review from 2023 included 17 randomised controlled trials with 5 to 156 people each, followed for anything from under 1 day to 5 weeks. Because the data are so few, the trials recruited very different people, and the follow-up times do not line up, the authors did not pool them into a single total. All the results below are against lenses with no blue-light filter. With follow-up under a week, self-reported eye tiredness「可能没有差别」, with low certainty of the evidence. In one trial of 120 people the two groups differed by 9.76 units. The credible range for that difference is 95% CI −33.95 to 53.47. The range spans both signs, which means no difference can be seen. Critical flicker fusion frequency is an objective measure of eye fatigue, and on it the two groups「可能几乎没有差别」, with low certainty. Best-corrected visual acuity, meaning the best vision once the glasses correct it, was「大概没有影响」. The two groups differed by 0.00 logMAR (95% CI −0.02 to 0.02). This outcome rests on 1 study with 156 people, with moderate certainty. Whether it affects daytime alertness is unclear (2 trials, 42 people, very low certainty). The review also found no evidence that it protects the macula at the back of the eye +- In plain language: Blue-light filtering lenses do not work. Not buying them saves you the several hundred to several thousand yuan. Compared with ordinary lenses, people wearing them are neither less tired nor able to see more sharply, and no evidence has been found that they protect the back of the eye. Eye strain and dryness from looking at a screen do settle. What really damages the eye is welding arc light, ultraviolet light and laser light, and an acute attack of glaucoma. +- Benefit: The Cochrane 2023 systematic review included 17 randomised controlled trials of between 5 and 156 people each, with follow-up ranging from less than a day to 5 weeks. Because there was too little data, the trials enrolled very different people, and the follow-up times did not match, the authors did not pool them into one overall figure. The results below are all against lenses with no blue-light filter. Where follow-up was shorter than a week, self-reported eye fatigue "may make little or no difference" and the certainty of the evidence is low. In one trial of 120 people the two groups differed by 9.76 units. The credible range of that difference is 95% CI −33.95 to 53.47. A range that crosses both signs means no difference can be made out. Critical flicker fusion frequency is an objective measure of eye fatigue; the two groups "may be almost identical", with low certainty. The best corrected visual acuity "is probably not affected". The two groups differ by 0.00 logMAR (95% CI −0.02 to 0.02). This outcome rests on 1 study and 156 people, with moderate certainty. Any effect on daytime alertness could not be established (2 trials, 42 people, very low certainty). The review also found no evidence that these lenses protect the macula at the back of the eye - Evidence grade: A - Sources: Singh S, Downie LE, Anderson AJ, Keller PR, White SJ, Ang M, Wolffsohn JS (2023). Blue-light filtering spectacle lenses for visual performance, sleep, and macular health in adults. Cochrane Database of Systematic Reviews, 8, CD013244. -- Notes: What looking at a screen brings is visual fatigue and dry eye, which show up as aching, dryness and blurring after a while. Resting and blinking more eases it. It is something that eases, which is not the same as "the screen has damaged your eyes". That is the consensus of ophthalmology rather than the conclusion of the Cochrane review in the sources field, so it only counts as grade C. What really causes irreversible eye damage in a short time is an intense light source such as a welding arc, ultraviolet light or a laser. An eye burned by a welding arc is called photophthalmia, and it is a legally recognised occupational disease. What keeps it off is goggles and a face shield, not blue-light lenses; see item 11 in section 19 (protection against dust, noise, chemical poisons). For staying up all night in front of a screen, what you have to guard against is not the screen itself but one exception: an eye that is swollen and red, a rainbow halo around lights, and headache, nausea and vomiting mean an acute attack of angle-closure glaucoma, which can destroy the optic nerve within days, so go to the eye department emergency the same day; see item 6 in section 13 (eye department emergency). A dilated pupil in the dark and long periods of looking down, which pushes the lens forward, are what bring it on, and those at highest risk are people over 50, people with long-sightedness and people with a shallow anterior chamber. The dry eye that comes from screen use can, if it goes on, turn into a chronic condition that needs a doctor, and it is not fixed by a night's sleep. If your eyes keep troubling you, or your sight really is going down, go and find out why in the eye department instead of swapping lenses yourself. For how to prevent myopia in children and teenagers, see items 4, 12 and 9 in section 30 (two hours outdoors a day, dilated refraction, and not buying products that "cure" myopia). +- Notes: Looking at a screen brings eye strain and dry eye, which show up as soreness, dryness and blurred vision after a while. Resting and blinking more often eases them. That they settle does not mean "you have damaged your eyes". This is the consensus of ophthalmology, not the conclusion of the Cochrane review in the Sources field, so it is graded C only. What really causes irreversible eye damage in a short time is a strong light source such as welding arc light, ultraviolet light or laser light. Eye injury from welding arc light is called photophthalmia and is a legally recognised occupational disease. What keeps it off you is goggles and a face shield, not blue-light lenses, see item 10 in section 19 (protection against dust, noise and chemical toxins). What all-night screen work needs protection against is not the screen itself but one exception: a swollen, red eye, a rainbow ring around a light, and headache, nausea and vomiting mean an acute attack of angle-closure glaucoma, which can destroy the optic nerve within days; go to the eye emergency department the same day, see item 6 in section 13 (eye emergency). A dilated pupil in the dark and the lens being pushed forward by looking down for a long time are what trigger it; those at high risk are over 50, far-sighted, or have a shallow anterior chamber. Dry eye from the screen can also turn into a chronic condition that needs a doctor, not something a night's sleep cures. If your eyes keep bothering you, or your eyesight really is going down, go to the eye department and find the cause, rather than changing lenses and counting that as the end of it. For preventing myopia in children and teenagers, see items 4, 12 and 9 in section 30 (2 hours outdoors a day, dilated refraction testing, not buying products that "cure myopia"). + + ### 17. Do not treat blood donation as a health practice: "detox, lower cholesterol, prevent heart attacks" has no support from randomised trials, while donating too often really does run you out of iron - Cost: No money. One donation, with the forms, the blood draw and the time sitting afterwards, takes about an hour. For weeks afterwards your body is making up the iron you lost. @@ -140,14 +145,16 @@ This section collects the things that look cheap and effective. Open the origina - Evidence grade: A - Sources: Salonen JT, Tuomainen TP, Salonen R, Lakka TA, Nyyssönen K (1998). Donation of blood is associated with reduced risk of myocardial infarction. The Kuopio Ischaemic Heart Disease Risk Factor Study. American Journal of Epidemiology, 148(5), 445-451. ; Ullum H, Rostgaard K, Kamper-Jørgensen M, et al. (2015). Blood donation and blood donor mortality after adjustment for a healthy donor effect. Transfusion, 55(10), 2479-2485. ; Zacharski LR, Chow BK, Howes PS, et al. (2007). Reduction of iron stores and cardiovascular outcomes in patients with peripheral arterial disease: a randomized controlled trial. JAMA, 297(6), 603-610. ; Di Angelantonio E, Thompson SG, Kaptoge S, et al. (2017). Efficiency and safety of varying the frequency of whole blood donation (INTERVAL): a randomised trial of 45 000 donors. Lancet, 390(10110), 2360-2371. ; Beard JL, Borel MJ, Derr J (1990). Impaired thermoregulation and thyroid function in iron-deficiency anemia. American Journal of Clinical Nutrition, 52(5), 813-819. ; Standing Committee of the National People's Congress (全国人大常委会) (1997). Blood Donation Law of the People's Republic of China（中华人民共和国献血法）(Articles two, nine and fourteen). ; National Health Commission and others (国家卫生健康委等) (2024-01-08). Notice on Further Improving Incentive and Reward for Voluntary Blood Donors（关于进一步做好无偿献血者激励奖励工作的通知）. - Notes: What is being rejected is donating blood for the sake of your own health, not blood donation itself. The blood is for someone else, so on this book's beneficiary tiers it falls in tier ④ (a stranger). What does come back to you is the share the system gives you. Article fourteen of the Blood Donation Law provides that a person who has given blood unpaid is exempt from the collection, storage, separation and testing charges when using blood in hospital themselves, and that the spouse and immediate family may be exempt or pay less according to the provincial government's rules. The National Health Commission notice of January 2024 states that 28 provinces already let donors and their relatives have the cost deducted at discharge for blood used in hospitals in the same province. Those who have received the voluntary blood donation award can also, under the local "three free" policy, travel on public transport free, visit government-run parks free, and be exempt from the ordinary outpatient consultation fee at public hospitals. If you are going to give blood, keep to the statutory intervals. Article nine of the Blood Donation Law provides for 200 millilitres each time in general, and no more than 400 millilitres. The interval between the two must be at least 6 months. The recommended ages are 18 to 55. Do not shorten the interval to the minimum in order to collect an honour or a group quota. Women, people of low body weight and people with heavy menstrual bleeding are already short of iron; anyone who donates repeatedly can pay for one ferritin test. If you feel constantly tired or cold after donating, or your legs are uncomfortable at night and you cannot stop moving them (this is called restless legs), do not assume your "base qi" is damaged; have a blood count and a ferritin test. If you come up deficient, take iron as the doctor says and lengthen the interval between donations. The human evidence on feeling cold rests on 10 people; it explains why this happens, it is not a statement of how many people it happens to. -### 18. Do not rely on a general health check package for prevention; have the few screenings that do have evidence, by age and sex +### 18. Do not rely on a general health check package to prevent disease; do the few screenings with evidence behind them, by age and sex -- Cost: No money, and it saves the several hundred to several thousand yuan a year that a package costs. What you take on is the obligation to remember which tests to have and how often. -- In plain language: People who have a general health check regularly have the same mortality as people who do not, and the same number of deaths from cardiovascular disease and cancer. Having the few screenings that do have evidence, by age and sex, is enough. Anyone over 35 can have their blood pressure measured free once a year, and anyone over 65 gets a free health check once a year. -- Benefit: The 2019 update of the Cochrane systematic review included 17 randomised controlled trials. For all-cause mortality it pooled 11 of those trials, with 233,298 participants and 21,535 deaths. The group that had the general health check was the same as the comparison group, RR 1.00. RR is the risk ratio, where 1 means the two groups are the same. The 95% CI 0.97–1.03 is the credible range. For death from cardiovascular disease it pooled 9 trials with 170,227 people and 6,237 deaths, RR 1.05 (95% CI 0.94–1.16). For death from cancer it pooled 8 trials with 139,290 people and 3,663 deaths, RR 1.01 (95% CI 0.92–1.12). Neither showed a difference either. The authors' conclusion is that「通用健康检查不大可能有益」. The national basic public health service programme gives residents aged 35 and above one free blood pressure measurement a year, and gives people aged 65 and above one free health check a year. This free check covers a physical examination, a full blood count, a routine urine test, fasting blood glucose and an electrocardiogram. Liver function covers alanine aminotransferase, aspartate aminotransferase and total bilirubin. Kidney function covers serum creatinine and blood urea. Blood lipids cover total cholesterol, triglycerides, low-density lipoprotein cholesterol and high-density lipoprotein cholesterol. Finally there is an abdominal ultrasound of the liver, gallbladder, pancreas and spleen (nationwide) +- Cost: No money, and you also save the several hundred to several thousand yuan a year that the package costs. What you take on instead is remembering which tests to have and how often. +- In plain language: People who go for a general health check regularly have the same mortality as people who do not, and they die of cardiovascular disease and cancer just as often. Doing the few screenings with evidence behind them, by age and sex, is enough. Over 35 you can have your blood pressure taken free once a year; over 65 you get one free health check a year. +- Benefit: The 2019 update of the Cochrane review included 17 randomised controlled trials. For all-cause mortality it pooled 11 of those trials, 233,298 participants and 21,535 deaths. The general health check group was the same as the comparison group, RR 1.00. RR is the risk ratio, where 1 means the two groups are the same. 95% CI 0.97–1.03 is the credible range. For death from cardiovascular disease it pooled 9 trials, 170,227 people and 6,237 deaths, RR 1.05 (95% CI 0.94–1.16). For death from cancer it pooled 8 trials, 139,290 people and 3,663 deaths, RR 1.01 (95% CI 0.92–1.12). Neither showed a difference either. The authors' conclusion is that general health checks are "unlikely to be beneficial". The national basic public health service programme gives residents aged 35 and above one free blood pressure measurement a year, and gives people aged 65 and above one free health check a year. This free health check includes a physical examination, a full blood count, a routine urine test, fasting blood glucose and an electrocardiogram. Liver function covers alanine aminotransferase, aspartate aminotransferase and total bilirubin. Kidney function covers serum creatinine and blood urea nitrogen. Blood lipids cover total cholesterol, triglycerides, low-density lipoprotein cholesterol and high-density lipoprotein cholesterol. Finally there is an abdominal ultrasound of the liver, gallbladder, pancreas and spleen (nationwide) - Evidence grade: A -- Sources: Krogsbøll LT, Jørgensen KJ, Gøtzsche PC (2019). General health checks in adults for reducing morbidity and mortality from disease. Cochrane Database of Systematic Reviews, 1, CD009009. ; National Health and Family Planning Commission (国家卫生计生委) (2017). National Basic Public Health Service Specification, Third Edition（国家基本公共卫生服务规范（第三版））(health management for older people, health management for patients with hypertension). -- Notes: What is being rejected is using an annual package as a means of prevention, not doing every test there is. Separately, the free yearly blood pressure measurement over 35 and the free yearly health check over 65 are provided by the national basic public health service programme in any case, and cost you nothing. Where there is a reason to have a test, have it. Blood pressure, blood glucose and hepatitis B: see items 7, 8 and 14 in section 1. Breast, cervical and colorectal: see items 17, 18 and 19 in section 1. Helicobacter pylori and low-dose CT: see items 23 and 24 in section 1. If you have had high-risk behaviour, get tested; see item 31 in section 1. The age to start at and how often to repeat are written down there. An examination organised by your employer and paid for by someone else is worth having; what is being stopped is you paying for a package and you adding items to it. People who already have a chronic illness, people with symptoms, and tests a doctor orders for a particular condition are not covered, because that is medical care rather than screening. The items in a package with least evidence behind them are tumour markers and whole-body imaging; see item 7 in this section (whole-body PET-CT and tumour marker panels). For high uric acid that has never hurt and gallstones that have never ached, see item 19 in this section (asymptomatic high uric acid) and item 20 in this section (asymptomatic gallstones). Separately, most of the trials Cochrane included were done in high-income countries, where seeing a doctor is easy to begin with, so this conclusion should not be copied straight into a region where medical resources are tight. +- Sources: Krogsbøll LT, Jørgensen KJ, Gøtzsche PC (2019). General health checks in adults for reducing morbidity and mortality from disease. Cochrane Database of Systematic Reviews, 1, CD009009. ; National Health and Family Planning Commission (国家卫生计生委) (2017). National Basic Public Health Service Specification (Third Edition)（国家基本公共卫生服务规范（第三版））(health management for older people, health management for patients with high blood pressure). +- Notes: What is being rejected is "using an annual package as a way of preventing disease", not the claim that no test should be done. And the free blood pressure measurement over 35 and the free health check over 65 are part of the national basic public health service programme anyway, so they cost you nothing. Do have any test you have a reason to have. Blood pressure, blood glucose and hepatitis B: see items 7, 8 and 14 in section 1. Breast, cervical and colorectal: see items 17, 18 and 19 in section 1. Helicobacter pylori and low-dose CT: see items 23 and 24 in section 1. Bone density testing for women over 65: see item 39 in section 1 (bone density testing). If you have had high-risk behaviour, go and be tested: see item 31 in section 1. What age to start at and how often to repeat are written there. Check-ups arranged by your employer and paid for by somebody else are still worth going to; what should be stopped is buying a package with your own money and adding items to it. If you already have a chronic illness, if you have symptoms, or if a doctor has ordered a test for your condition, none of that counts: that is medical care, not screening. The commonest items in a package with no evidence behind them are tumour markers and whole-body imaging, see item 7 in this section (a whole-body PET-CT and tumour marker package). What to do about a high uric acid finding you have never had an attack from, or gallstones you have never felt: see item 19 in this section (asymptomatic high uric acid) and item 20 (asymptomatic gallstones). Separately, most of the trials Cochrane included were carried out in high-income countries, where seeing a doctor is easy to begin with; this conclusion should not be carried over unchanged to places where medical resources are tight. + + ### 19. Do not start urate-lowering drugs because a check-up found high uric acid and you have never had an attack - Cost: No money, and it saves the long run of medicine and repeat testing as well as the risk of a dose you did not need. @@ -156,14 +163,16 @@ This section collects the things that look cheap and effective. Open the origina - Evidence grade: A - Sources: FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760（表 1 及其脚注）. ; Badve SV, Pascoe EM, Tiku A, et al. (2020). Effects of Allopurinol on the Progression of Chronic Kidney Disease. New England Journal of Medicine, 382(26), 2504-2513. ; Doria A, Galecki AT, Spino C, et al. (2020). Serum Urate Lowering with Allopurinol and Kidney Function in Type 1 Diabetes. New England Journal of Medicine, 382(26), 2493-2503. - Notes: What is being rejected is "start lifelong urate-lowering drugs before you have ever had an attack", not the idea that high uric acid does not matter. The same guideline lists the exceptions. Starting the drug can be considered in three situations. First, a first attack that comes at the same time as stage 3 or worse chronic kidney disease. Second, blood urate above 9 mg/dL (about 535 µmol/L). Third, a history of urinary stones. People who have already had an attack, people with tophi, and people whose scans show bone already eroded should take the drug; for the details see item 9 in section 16 (once gout is diagnosed, take urate-lowering drugs long term). About 7.4% of Han Chinese carry a genotype called HLA-B\*5801, against 0.7% of white people. People with this type are more likely to develop a severe and potentially fatal drug rash on allopurinol, and the risk of hypersensitivity syndrome among Asian people is 3 times that of white people. This is a drug you can perfectly well do without for now, and there is no reason to take on that extra risk. Before you do start, have this genotype tested; see item 9 in section 16 (once gout is diagnosed, take urate-lowering drugs long term). Those two kidney trials were done in people with chronic kidney disease and type 1 diabetes, and cannot be turned into "high uric acid is completely harmless for the kidney". All they rule out is the one practice of taking a drug to lower uric acid in order to protect the kidney. -### 20. Do not remove a gallbladder preventively because a check-up found gallstones and you have never felt pain + +### 20. Do not have a preventive cholecystectomy because a check-up found gallstones you have never felt from (asymptomatic gallstones) -- Cost: No money, and it saves one operation under general anaesthetic and one hospital stay. What you take on is remembering which signs mean you have to go to hospital at once. -- In plain language: Among people with stones in the gallbladder who have never felt pain, the large majority never do. An Italian study followed 580 such people for nearly nine years and 78% had no symptoms at all throughout. A US group of 135 people followed for about four years had only 7% operated on. The British guideline puts it plainly: if stones are found in the gallbladder by chance and you have no symptoms, you do not need treatment. -- Benefit: The Italian MICOL population survey gave ultrasound scans to 11,229 people aged 29 to 69. It found 856 with gallstones, of whom 580 had no symptoms, and followed them for an average of 8.7 years. Of those 580, 453 (78.1%) had no symptoms throughout, 61 (10.5%) developed mild symptoms, and 66 (11.4%) developed severe symptoms. In total 189 cholecystectomies were done during the follow-up, and 41.3% of them were on people with no symptoms. The authors conclude that「期待治疗对多数患者仍是有效的处理方式」, where expectant management means watching and waiting rather than operating. A US HMO kept follow-up statistics, and an HMO is a managed care organisation; it followed 691 people with diagnosed gallstones for an average of 78±61.6 months. Of the 135 with no symptoms, followed for 58±50.2 months (median 46.3 months), only 10% developed symptoms of gallstones and 7% had an operation. Of the 50 deaths in the whole group, only 2 were related to the biliary tract, and both were in the group that had symptoms. The authors conclude that「有无声结石的患者不需要在出现症状前手术」, where a silent stone is one that does not speak up and does not attack. Paragraph 1.2.1 of the British NICE guideline CG188 says this.「对胆囊和胆道均正常、偶然发现无症状胆囊结石的人，应告知其在出现症状前不需要治疗」. Prophylactic removal looks like good value because of the intuition "it will hurt eventually, so cut it out while you are young" +- Cost: No money, and you also save one operation under general anaesthetic and one hospital stay. What you take on instead is remembering which signals mean you have to go to hospital at once. +- In plain language: Of the people who have stones in the gallbladder and have never felt pain, about 80% stayed without symptoms over nearly nine years. Leave the stones alone and the chance of an emergency such as cholecystitis or pancreatitis each year is about one in a thousand to three in a thousand. Removing the gallbladder is not risk-free either: about 1.5% of people have the bile duct injured. So the European and British guidelines both say that if there are no symptoms, do not operate yet. +- Benefit: The Italian MICOL population survey put 11,229 people aged 29 to 69 through an ultrasound. 856 were found to have gallstones, of whom 580 had no symptoms, and they were followed for 8.7 years on average. Of those 580 people, 453 (78.1%) stayed without symptoms, 61 (10.5%) developed mild symptoms and 66 (11.4%) developed severe symptoms. In the whole follow-up period 189 cholecystectomies were done, 41.3% of them on people with no symptoms. The authors' conclusion is that "expectant management is still an effective approach for most patients"; expectant management means watching and waiting without an operation. A US HMO kept follow-up statistics; an HMO is a managed care organisation. Of 691 people with diagnosed gallstones, followed for 78±61.6 months on average, the 135 with no symptoms were followed for 58±50.2 months (median 46.3 months), and only 10% developed symptoms of gallstones while 7% had an operation. Of the 50 deaths in the whole group only 2 had anything to do with the biliary tract, and both were in the group with symptoms. The authors' conclusion is that "patients with silent stones do not require surgery before symptoms appear"; silent stones are stones that stay quiet and do not attack. The British NICE guideline CG188 says it in these words at clause 1.2.1: "for a person with a normal gallbladder and biliary tract in whom asymptomatic gallstones are found incidentally, it should be explained that no treatment is needed until symptoms appear". The European Association for the Study of the Liver put the risks on both sides together in its 2016 gallstone guideline. In people with no symptoms, about 0.7% to 2.5% develop symptoms each year. Acute cholecystitis, acute pancreatitis, obstructive jaundice (the bile duct blocks and the person turns yellow) or cholangitis occur in 0.1% to 0.3% a year. Only after the first attack of colic do complications rise to 1% to 3% a year, so most people feel pain once before anything serious happens. On the operation side, in 51041 cholecystectomies in the Swedish national registry, 1.5% injured the bile duct. About a fifth of those (0.3%) were a partial or complete division of the bile duct. Among people whose bile duct was injured, mortality within a year was 3.9%; among those whose duct was not, 1.1%. The guideline also says that 10% to 40% of patients still have discomfort after the operation, or feel better for a while and then have it again. The guideline's conclusion is that an operation in a person with no symptoms, whether open or laparoscopic, does not lengthen life, because the risk of death and complications from the operation itself is greater than the risk of not operating. Waiting until symptoms appear also costs less in total. A 1983 modelling exercise gave specific figures: a preventive removal in a 30-year-old man costs 4 days of life on average compared with watching and waiting, and 18 days at 50. To this day no randomised trial has compared operating with not operating. Preventive removal looks good value because of the intuition that "it will hurt sooner or later, so better to take it out in one go while you are young" - Evidence grade: A -- Sources: Festi D, Reggiani ML, Attili AF, et al. (2010). Natural history of gallstone disease: Expectant management or active treatment? Results from a population-based cohort study. Journal of Gastroenterology and Hepatology, 25(4), 719-724. ; McSherry CK, Ferstenberg H, Calhoun WF, Lahman E, Virshup M (1985). The natural history of diagnosed gallstone disease in symptomatic and asymptomatic patients. Annals of Surgery, 202(1), 59-63. ; National Institute for Health and Care Excellence (2014). Gallstone disease: diagnosis and management. NICE guideline CG188, Articles 1.2.1 and 1.3.1. -- Notes: This covers one situation only: the stones are in the gallbladder and have never caused pain. **A stone in the common bile duct is a different matter.** Paragraph 1.3.1 of the same guideline says plainly that a common bile duct stone has to be cleared from the duct and the gallbladder removed, whether or not there are symptoms. If any of the following happens, go to hospital the same day and do not try to tough it out on the strength of this conclusion: persistent cramping pain in the upper right or upper abdomen, fever with chills, yellowing of the whites of the eyes or the skin, or darkening of the urine. When a doctor has looked at your imaging and recommends surgery, ask which reason they are relying on, for instance whether the bile duct is involved, whether there are polyps as well, or whether the gallbladder wall looks abnormal, and do not use this conclusion to argue against them. All it holds back is "no reason at all, cut it out simply because stones were found". The two studies above followed a group of people forwards without a control group. But what they answer is "what happens if you leave it alone", not "which treatment is better", so they fit here. Both also followed people for 4 to 9 years, and over a longer period the share who develop symptoms rises. +- Sources: Festi D, Reggiani ML, Attili AF, et al. (2010). Natural history of gallstone disease: Expectant management or active treatment? Results from a population-based cohort study. Journal of Gastroenterology and Hepatology, 25(4), 719-724. ; McSherry CK, Ferstenberg H, Calhoun WF, Lahman E, Virshup M (1985). The natural history of diagnosed gallstone disease in symptomatic and asymptomatic patients. Annals of Surgery, 202(1), 59-63. ; National Institute for Health and Care Excellence (2014). Gallstone disease: diagnosis and management. NICE guideline CG188, clauses 1.2.1 and 1.3.1. ; European Association for the Study of the Liver (2016). EASL Clinical Practice Guidelines on the prevention, diagnosis and treatment of gallstones. Journal of Hepatology, 65(1), 146-181. ; Törnqvist B, Strömberg C, Persson G, Nilsson M (2012). Effect of intended intraoperative cholangiography and early detection of bile duct injury on survival after cholecystectomy: population based cohort study. BMJ, 345, e6457. ; Ransohoff DF, Gracie WA, Wolfenson LB, Neuhauser D (1983). Prophylactic cholecystectomy or expectant management for silent gallstones. A decision analysis to assess survival. Annals of Internal Medicine, 99(2), 199-204. ; Gurusamy KS, Samraj K (2007). Cholecystectomy versus no cholecystectomy in patients with silent gallstones. Cochrane Database of Systematic Reviews, CD006230. ; the other side: Venneman NG, Buskens E, Besselink MG, et al. (2005). Small gallstones are associated with increased risk of acute pancreatitis: potential benefits of prophylactic cholecystectomy? American Journal of Gastroenterology, 100(11), 2540-2550. ; Biliary Surgery Group, Surgical Branch of the Chinese Medical Association, and Biliary Surgery Committee, Surgeons Branch of the Chinese Medical Doctor Association (2022). Expert consensus on the surgical treatment of benign gallbladder diseases (2021 edition). 中华外科杂志, 60(1), 4-9. (full text requires a subscription, not obtained) ; Same as above (2022). 《胆囊良性疾病外科治疗的专家共识（2021版）》解读, Interpretation of the Expert Consensus on the Surgical Treatment of Benign Gallbladder Diseases (2021 Edition). 中华外科杂志, 60(4), 337-342. +- Notes:Disputed. The 2021 Chinese expert consensus on biliary surgery lists some asymptomatic gallstones among the indications for surgery. An explanatory article in the same journal states that this is one of the core points of the dispute. The full text of the consensus is behind a subscription, and the exact circumstances it lists were not checked against the original (TODO: to be verified). The other objection concerns small stones: small stones under 5 millimetres are more closely related to pancreatitis. A Dutch modelling exercise says that preventive removal in these people may lengthen life or shorten it, depending on how common pancreatitis is and how deadly. The European guidelines also list exceptions: calcification of the gallbladder wall (a porcelain gallbladder), and a gallbladder polyp over 1 centimetre at the same time; these two should be removed. When you are having a different major operation on the abdomen, removing the gallbladder at the same time can be considered. This item describes one situation only: the stones are in the gallbladder and have never caused pain. **Stones in the common bile duct are a different matter.** Clause 1.3.1 of the same guideline states plainly that a common bile duct stone must be cleared from the duct and the gallbladder removed, whether there are symptoms or not. If any of the following appears, go to hospital the same day, and do not ride this conclusion out: persistent colicky pain in the right upper abdomen or upper abdomen, fever with chills, yellowing of the whites of the eyes or the skin, or dark urine. When a doctor who has looked at your images advises an operation, ask which of the reasons it is: involvement of the bile duct, a polyp growing at the same time, an abnormal gallbladder wall. Do not use this conclusion to argue against him. It only blocks "operating for no reason at all, simply because stones were found". Both studies above followed a group of people forwards without a control group. But what they have to answer is what happens if you leave the stones alone, not which treatment is better, so they fit here. Separately, the two of them followed people for 4 to 9 years; with longer follow-up the share who develop symptoms goes up + ### 21. Do not cut out calcium in order to prevent kidney stones - Cost: No money. It also saves you trouble, because you no longer have to watch what contains calcium every day. @@ -172,6 +181,7 @@ This section collects the things that look cheap and effective. Open the origina - Evidence grade: A - Sources: Borghi L, Schianchi T, Meschi T, et al. (2002). Comparison of two diets for the prevention of recurrent stones in idiopathic hypercalciuria. New England Journal of Medicine, 346(2), 77-84. ; National Institute for Health and Care Excellence (2019). Renal and ureteric stones: assessment and management. NICE guideline NG118, Article 1.8.1. - Notes:Disputed: this trial was only done in men, and all of them had recurrent calcium oxalate stones with high urinary calcium, so it cannot be applied directly to every kind of stone. The intervention group also changed three things at once, eating calcium normally, eating less animal protein and eating less salt, so not all of the benefit can be put down to not cutting out calcium; the lower salt and the less animal protein are doing work too. What is being rejected is "deliberately cut out anything containing calcium", not "you should take more calcium". What is changed here is calcium in your daily meals; calcium supplements are a separate question, and if you have had stones, ask your treating doctor whether to take them. For drinking more water, see item 8 in section 16 (after a kidney stone, drink 2.5–3 litres a day). + ### 22. Do not buy crystals, bead bracelets or pixiu for "turning your luck", "bringing in money" or "nurturing people" - Cost: No money. All you have to do is keep "turning your luck" and "nurturing people" out of your reasons for buying it. The hard part is keeping up when everyone around you is wearing one and saying it works. @@ -188,30 +198,34 @@ This section collects the things that look cheap and effective. Open the origina - Evidence grade: B - Sources: Dittmar H, Bond R, Hurst M, Kasser T (2014). The relationship between materialism and personal well-being: A meta-analysis. Journal of Personality and Social Psychology, 107(5), 879-924. ; Van Boven L, Gilovich T (2003). To do or to have? That is the question. Journal of Personality and Social Psychology, 85(6), 1193-1202. - Notes:Disputed: most of the studies included asked a group of people at one point in time whether two things moved together, which cannot show which came first. The authors say in their own conclusion that more experiments and long-term follow-up are needed to see the direction. So this cannot be read as "buying things makes people unhappy"; it can only be read as "do not expect buying things to make you happy". And the association is weak: r = -0.19 accounts for a few percentage points of difference, and people differ from each other a great deal. Both studies used foreign samples, and there is no comparable data for the Chinese population that can be cited. Nothing here argues against spending money on something you like. What is being argued against is using shopping as a way of managing your mood, because that leaves your first reaction the next time you feel low still being to order something. The specific actions for saving money are in section 5. What really has evidence behind it when you feel low is in the second half of section 22 and in section 29. For budget added in order to be "one notch above other people", see item 24 in this section (moving up one notch). -### 24. Do not spend more on a bigger house, a better car or a better circle in order to move up a notch among the people around you +### 24. Do not spend more to move up one notch among the people around you, by changing house, car or circle -- Cost: No money. All you have to do is ask one more question before adding to the budget: is this extra price buying the thing itself, or my position among these people? The hard part is not keeping up when everyone around you has changed. -- In plain language: Among people with the same income, the more a neighbour earns, the lower this person reports their own well-being. This happens only between neighbours who have the same level of education and who see each other often. So the extra money you pay to move somewhere everyone is better off than you buys you a move down a notch. Still, when incomes rise together people do feel better, so the absolute level still counts. -- Benefit: Two waves of the US National Survey of Families and Households revisited the same people. The authors matched the data to local average income. Local average income (PUMA) is estimated from the 5% sample of the 1990 census. Once the person's own income and a set of personal characteristics are taken out, the result is that the higher the neighbours' income, the lower the well-being they report. The effect is large, it is stable under a different method, and it is highly significant. What the authors write is that the fall in well-being from "neighbours' income rising" is of the same order of magnitude as the fall from "own income falling by the same amount". Once it is changed so that the same person is compared with themselves over time (this is called adding an individual fixed effect), the sign of the coefficient does not change and its size is similar; only the significance drops to the 10% level. Restricted to comparisons within the same state and the same survey wave (this is called state and wave fixed effects), the result is essentially unchanged. That shows it is not caused by "people who move to poor areas being happier to begin with", and not caused by some unmeasured factor at state level. The effect is stronger among people who often see their neighbours, and is not significant among people whose friends all live outside the neighbourhood. There is no difference between renters and homeowners, which does not support an explanation of the kind "prices are high in expensive areas, so there is less left to spend". Well-being falls only with the income of neighbours who have the same level of education, and barely reacts to the income of neighbours with a different level of education. And it works mainly through "satisfaction with my own material situation", not through satisfaction with family life or anything else. Finally, if a person's own income and the neighbours' incomes all rise by the same proportion, well-being still rises +- Cost: No money. All you have to do is ask one more question before you add to your budget: is this extra money buying the thing itself, or your position among these people? The hard part is keeping up when everybody around you has changed. +- In plain language: Comparing people with the same income, the more a neighbour earns, the lower that person reports their own sense of well-being. This only happens with neighbours who have the same education as you and whom you see often. So the extra money you spend to move into a place where everybody is better off than you buys yourself a move down a notch. But when your own income rises, you do feel better off: the absolute level still counts. +- Benefit: Two waves of the US National Survey of Families and Households (NSFH) revisited the same group of people. The author matched this data to local median income. Local median income (PUMA) is estimated from a 5% sample of the 1990 census. After subtracting the person's own income and a set of personal characteristics, the result was this: the higher a neighbour's income, the lower the person's own reported sense of well-being. The effect is large, it holds under a different specification, and it is highly significant. The author puts it this way: the drop in well-being from "a neighbour's income rising" is of the same order as the drop from "your own income falling by the same amount". Changed to comparing one person with themselves before and after (adding individual fixed effects), the sign of the coefficient is unchanged and its size is similar; only the significance falls to the 10% level. Narrowed further to comparisons within one state and one wave of the survey (adding state and wave fixed effects), the result is basically unchanged. That means it is not caused by "people who love moving to poorer places being happier to begin with", nor by some unmeasured factor at state level. The effect is stronger in people who are often in contact with their neighbours and not significant in people whose contact is with friends outside the neighbourhood. There is no difference between renters and people who own their home, which does not support the explanation "prices are high in expensive places, so there is less money to spend in reality". Sense of well-being falls only with the income of neighbours who have the same education, and barely responds to the income of neighbours with a different education. And it is carried mainly by satisfaction with one's own material circumstances, not by satisfaction with family life or anything else. Lastly, if your own income and your neighbour's income rise together by the same proportion, sense of well-being still rises - Evidence grade: B -- Sources: Luttmer EFP (2005). Neighbors as Negatives: Relative Earnings and Well-Being. The Quarterly Journal of Economics, 120(3), 963–1002. ; Luttmer EFP (2004). Neighbors as Negatives: Relative Earnings and Well-Being. NBER Working Paper No. 10667（本条引用的表述与数量级按这一版逐字核对）. -- Notes: There are several reasons for grading this B: the data come from a questionnaire that revisited the same people again and again, there is only one study, and the sample is foreign; what is being examined is the well-being people report themselves, not money. The author did use statistical means to remove some sources of interference, such as comparing the same person over time and comparing within the same state and wave, but it is still not a randomised trial. There is no comparable data for the Chinese population that can be cited. The benefit is graded "medium" because it follows the standard set in item 23 in this section (do not expect buying things to improve your mood). The money saved really could run into the tens of thousands. But there is no figure behind the claim that "saving this much would buy so much well-being", so it is not recorded as "large" at that scale. Nothing here argues against spending money on a bigger house, a quieter neighbourhood or a shorter commute; those buy the thing itself. The single reason to hold back is "everyone at work has changed, I cannot be worse off than they are". For what a home should be ranked on, see item 18 in section 4 (put commuting time near the top when you choose where to live). For buying things to adjust your mood, see item 23 in this section (do not expect buying things to improve your mood). For always comparing yourself upwards online, see item 21 in section 3 (do not treat "how other people are doing" as required daily reading). -### 25. Do not believe that willpower is like a muscle and runs out +- Sources: Luttmer EFP (2005). Neighbors as Negatives: Relative Earnings and Well-Being. The Quarterly Journal of Economics, 120(3), 963–1002. ; Luttmer EFP (2004). Neighbors as Negatives: Relative Earnings and Well-Being. NBER Working Paper No. 10667 (the wording and order of magnitude cited in this item were checked word for word against this version). +- Notes: There are several reasons for the B: the data come from a questionnaire that revisited the same group of people repeatedly, there is only this one study, and the sample is from another country; what is measured is a sense of well-being the respondent reports himself, not money. The author did use statistical methods to rule out several kinds of interference, such as comparing one person with himself over time and comparing within one state and one wave, but it is not a randomised trial after all. There is also no comparable data on the Chinese population that can be cited. The benefit is graded medium, following the standard set in item 23 in this section (do not expect buying things to improve your mood). The money saved could indeed reach tens of thousands of yuan. But there is no figure behind "how much sense of well-being that money buys", so it is not recorded as large on the tens-of-thousands scale. Nothing here argues against spending money on a bigger house, a quieter development or a shorter commute: those buy the thing itself. The one reason to stop is "my colleagues have all upgraded and I cannot be worse off than them". How a place to live should be ranked, see item 18 in section 4 (put commute time first when choosing where to live). Buying things to manage your mood, see item 23 in this section (do not expect buying things to improve your mood). Always comparing upwards online, see item 20 in section 3 (do not read "how others are doing" every day). + +### 25. Do not believe that "willpower is like a muscle and runs out after a while" -- Cost: No money. All you have to change is what you say. -- In plain language: The popular line goes: do one thing that costs self-control first and you are more likely to give in on the next, so save your willpower. 23 laboratories with 2141 people reran this experiment on a single standard procedure, and the result is almost exactly zero. Stop using "I ran out of willpower today" to explain yourself, and stop paying for courses that teach you to hoard willpower. -- Benefit: A preregistered multi-laboratory replication, with 23 laboratories and 2141 participants in all, repeated the sequential task paradigm using the standardised procedure of Sripada and colleagues: first do one task that consumes self-control, then do a second task that requires self-control. The pooled result shows that the ego-depletion effect is very small, and that the 95% credible interval includes 0 (d = 0.04, 95% CI −0.07 to 0.15). On that basis the authors discuss whether the model that self-control is a limited resource still holds +- Cost: No money. What has to change is one belief. +- In plain language: The popular line is that once you spend self-control on something, it is easier to give in next, so you should save your willpower. 23 laboratories and 2141 people redid this experiment with the same protocol, and the result is very close to zero. Stop explaining yourself with "I ran out of willpower today", and stop paying for courses that teach you how to save up willpower. +- Benefit: One preregistered multi-lab replication, 23 laboratories and 2141 participants in all, redid the sequential task paradigm following the standardised protocol of Sripada and others: first a task that consumes self-control, then a second task that requires self-control. The pooled result shows a very small ego depletion effect, and the 95% confidence interval includes 0 (d = 0.04, 95% CI −0.07 to 0.15). The authors discuss whether the model that "self-control is a limited resource" still holds in the light of that. A separate preregistered replication of the same paradigm, 36 laboratories and 3531 people, gave an effect size of d = 0.06, again too small to count. - Evidence grade: A -- Sources: Hagger MS, Chatzisarantis NLD, Alberts H, et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science, 11(4), 546–573. -- Notes: What is being rejected is the model that self-control is a resource that gets used up, not the fact that people get tired. Not enough sleep and working too long in a row do lower your efficiency, which is a different matter and belongs with the items on sleep and hours in section 3. The approaches that really do have randomised trials behind them are changing the environment and changing how you write things down; see item 10 in section 4 (move what you would rather not touch further away) and item 1 in section 4 (write "when, where, and if this comes up then that"). The other items in this section count money; this one counts time and energy, and the two are not converted into each other. -### 26. Do not expect eating breakfast or 16:8 fasting to help you control your weight; pick an eating schedule you can keep up long term +- Sources: Hagger MS, Chatzisarantis NLD, Alberts H, et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science, 11(4), 546–573. ; Vohs et al. (2021). A Multisite Preregistered Paradigmatic Test of the Ego-Depletion Effect. Psychological Science. +- Notes: What is being rejected is the model that "self-control is a resource that gets used up", not the claim that "people get tired". Not getting enough sleep and working too long in a row do lower your efficiency; that is a different matter, handled by the items on sleep and working hours in section 3. The approach that genuinely has randomised trials behind it is changing the environment and changing how you write things down, see item 10 in section 4 (move what you do not want to touch further away) and item 1 in section 4 (write it as "what time, where, and what I do when this comes up"). Fixing what you wear and what you eat is still worth doing, because it costs almost nothing. What it saves is those few minutes and those few moments of hesitation, not willpower; do not expect it to make you judge better in the afternoon. The other items in this section count money; this one counts time and energy, and the two are not converted into each other. + + +### 26. Do not expect breakfast or 16:8 to help you control your weight; pick an eating schedule you can keep up -- Cost: No money. What you save is the daily effort of arguing with yourself about meal times. -- In plain language: The popular line is that skipping breakfast makes you overeat at lunch and dinner, so you end up eating more in the day. The trials do not show that: people who skipped breakfast ate more than two hundred calories less a day and weighed very slightly less. Over a year, 16:8 made no difference in weight against simply eating less. People who skip breakfast long term really do have about a sixth more cardiovascular disease, but there is no difference in all-cause mortality. -- Benefit: First, the claim that "skipping breakfast makes you eat more later". One meta-analysis pooled 13 randomised trials. People assigned to eat breakfast took in 259.79 kcal more a day than those who did not (95% CI 78.87–440.71, the credible range). They weighed 0.44 kg more (0.07–0.82). People who skipped breakfast did not make it up at lunch or dinner. These trials all had short follow-up, an average of 7 weeks for weight and 2 weeks for intake. Next, approaches such as 16:8, where you only eat during a fixed set of hours. One trial put 116 overweight adults on it for 12 weeks. The group eating only between 12 noon and 8 in the evening weighed -0.26 kg against the group eating three regular meals a day (-1.30 to 0.78), which is not significant. Another trial, in Guangzhou, followed 139 patients with obesity for 12 months. Time-restricted eating plus calorie control lost 1.8 kg more than calorie control alone (-4.0 to 0.4), also not significant. Finally, whether skipping breakfast for a long term does harm. In a pooling of 2.38 million people, people who skipped breakfast had about 17% higher risk of cardiovascular disease (OR 1.17, 1.09–1.26) and about 49% higher cardiovascular mortality (OR 1.49, 1.20–1.84). But these are record-only studies with no grouping. In the original study with 17 to 23 years of follow-up, the risk ratio for all-cause mortality was 1.19, with a credible range of 0.99 to 1.42, which crosses 1 and so cannot be called a difference. +- Cost: No money. What you save is the energy spent every day arguing with the clock about meals. +- In plain language: The popular line is that skipping breakfast makes you overeat at lunch and dinner, so that you eat more over the day. The trials do not show that: people who skip breakfast ate a couple of hundred fewer calories a day and weighed slightly less. Compared with simply eating less, 16:8 made no difference in weight over a year. People who never eat breakfast do have about 17% more cardiovascular disease, but no difference in all-cause mortality. +- Benefit: Start with the claim that skipping breakfast makes you eat more later. One meta-analysis pooled 13 randomised trials. People assigned to eat breakfast took in 259.79 kcal a day more than those who did not (95% CI 78.87–440.71, which is the credible range). They weighed 0.44 kg more (0.07–0.82). People who skipped breakfast did not make it up at lunch or dinner. These trials were all short: an average of 7 weeks for weight, an average of 2 weeks for intake. Next, methods like 16:8, where eating is confined to a fixed number of hours. One trial put 116 overweight adults on it for 12 weeks. The group that ate only between noon and 8 in the evening differed from the group eating three regular meals a day by −0.26 kg (−1.30 to 0.78), which is not significant. Another trial, in Guangzhou, enrolled 139 patients with obesity and followed them for 12 months. Time-restricted eating on top of calorie control lost 1.8 kg more than calorie control alone (−4.0 to 0.4), again not significant. Finally, whether never eating breakfast does harm. In a pooling of 2.38 million people, people who skipped breakfast had about 17% higher risk of cardiovascular disease (OR 1.17, 1.09–1.26) and about 49% higher cardiovascular mortality (OR 1.49, 1.20–1.84). But these are record-only studies with no grouping. In the original study that followed people for 17 to 23 years, the risk ratio for all-cause mortality was 1.19, with a credible range of 0.99 to 1.42, which crosses 1 and so cannot be called a difference. - Evidence grade: A - Sources: Sievert K, Hussain SM, Page MJ, et al. (2019). Effect of breakfast on weight and energy intake: systematic review and meta-analysis of randomised controlled trials. BMJ. ; Lowe DA, Wu N, Rohdin-Bibby L, et al. (2020). Effects of Time-Restricted Eating on Weight Loss and Other Metabolic Parameters in Women and Men With Overweight and Obesity. JAMA Internal Medicine. ; Liu D, Huang Y, Huang C, et al. (2022). Calorie Restriction with or without Time-Restricted Eating in Weight Loss. New England Journal of Medicine. ; Zhang H, Zhang S, Liu Y, et al. (2025). The association between skipping breakfast and cardiovascular disease: a meta analysis. Frontiers in Cardiovascular Medicine. ; Rong S, Snetselaar LG, Xu G, et al. (2019). Association of Skipping Breakfast With Cardiovascular and All-Cause Mortality. Journal of the American College of Cardiology. ; Fernandes-Alves D, Teixeira GP, Guimarães KC, Crispim CA (2026). Systematic Review and Meta-analysis of Randomized Clinical Trials Comparing Time-Restricted Eating With and Without Caloric Restriction for Weight Loss. Nutrition Reviews. ; Sun T, Zhang L, Lu Y, et al. (2024). Non-linear relationship between the first meal time of the day and gallstone incidence in American adults. Frontiers in Nutrition. ; Liu T, Wang Y, Wang X, et al. (2023). Habitually Skipping Breakfast Is Associated with the Risk of Gastrointestinal Cancers: Evidence from the Kailuan Cohort Study. Journal of General Internal Medicine. -- Notes:Disputed. The 17% on cardiovascular disease comes from record-only studies with no grouping. People who skip breakfast were already more likely to smoke, drink, sit for long hours and earn less, and those factors cannot be fully corrected for. So there is no way to say how much of that 17% is caused by breakfast itself. The all-cause mortality figure crossed 1 in any case. There is also a dispute about 16:8: pooling 30 trials with 1341 people, and on the condition that both groups ate the same amount, the time-restricted eating group lost 1.46 kg more (-2.65 to -0.26), at the cost of 0.41 kg more lean body mass lost. But most of those trials are short, and the longest and strictest one, the twelve-month trial, found no difference. Long periods of fasting are also connected with the gallbladder. In one cross-sectional survey of 6,547 people, each hour that the first meal of the day was delayed raised the share found to have gallstones by about 5% (OR 1.05, 1.02–1.08); people who put their first meal between 9 and 14 had about 49% more than those who ate earlier (OR 1.49, 1.24–1.77). The mechanism is that the longer bile sits in the gallbladder, the more likely crystals come out of it. But this is a survey at one point in time, and it may be that the gallbladder problem came first and changed the eating habits. The Chinese Kailuan cohort followed 5.6 years and recorded 369 gastrointestinal cancers: people who skipped breakfast had 2.32 times the risk of colorectal cancer (1.34–4.01) and 5.43 times the risk of gallbladder and extrahepatic bile duct cancer (1.34–21.93). The credible range on that second figure is as wide as 21.93, which means the events are so few that it can only be treated as a lead. Taken together, for weight control the effect of meal timing is small, and it is not worth spending effort on it. As for whether skipping breakfast altogether is good for you in the long run, all the evidence is association and cannot show cause and effect, but the direction is consistently unfavourable; if you want to be safe, eat normally, just do not expect it to help you lose weight. The main causes of gastritis and gastric ulcer are not an empty stomach but Helicobacter pylori and long-term painkillers; for what to test for, see item 23 in section 1 (test for Helicobacter pylori and eradicate it if positive). For gallstones found on a check-up that have never hurt, see item 20 in this section (do not remove a gallbladder preventively because a check-up found gallstones and you have never felt pain). For weight control, what really has evidence behind it is what you eat and how much; see item 28 in section 2 (eat 5 portions of fruit and vegetables a day), item 29 in section 2 (eat less ultra-processed food) and item 33 in section 2 (keep BMI between 20–25). This does not apply to people with diabetes, people taking hypoglycaemic drugs, or people who are pregnant, since a missed meal can cause hypoglycaemia; follow the doctor's instructions. Extreme fasting and purging are a different matter; see item 1 in section 28 (do not use extreme dieting, fasting or purging to control your weight). +- Notes:Disputed. The 17% on cardiovascular disease comes from record-only studies with no grouping. People who skip breakfast are already more likely to smoke, drink, sit for long hours and have a low income, and these factors cannot be fully adjusted away. So how much of that 17% is caused by breakfast itself cannot be said. The all-cause mortality figure already crossed 1. There is a dispute about 16:8 as well: pooling 30 trials and 1341 people, and given that both groups ate the same total amount, the time-restricted eating group lost 1.46 kg more (−2.65 to −0.26), at the cost of 0.41 kg more lean body mass lost. But most of those trials are short, and the longest and strictest of them, the twelve-month trial, says there is no difference. Long fasts also have something to do with the gallbladder. In a cross-sectional survey of 6,547 people, every hour that the first meal of the day was pushed later raised the share found to have gallstones by about 5% (OR 1.05, 1.02–1.08); people whose first meal fell between 9 and 14 o'clock had about 49% more than those who ate earlier (OR 1.49, 1.24–1.77). The mechanism is that the longer bile sits in the gallbladder the more likely stones come out of it. But this is a survey taken at one point in time, and it is also possible that the gallbladder problem came first and the eating habits changed afterwards. The Kailuan cohort in China followed people for 5.6 years and recorded 369 gastrointestinal cancers: people who skipped breakfast had 2.32 times the risk of colorectal cancer (1.34–4.01) and 5.43 times the risk of gallbladder and extrahepatic bile duct cancer (1.34–21.93). The second of those has a credible range as wide as 21.93, which means too few events; it can only be read as a lead. Taken together, the effect of meal timing on weight control is small, and it is not worth spending energy on. As for whether never eating breakfast is good for you over the long run, all the existing evidence is association and cannot show cause and effect, but the direction is consistently unfavourable; if you want to be on the safe side, eat breakfast as normal, just do not expect it to help you lose weight. The main causes of gastritis and gastric ulcer are not an empty stomach but Helicobacter pylori and long-term use of painkillers; for the test, see item 23 in section 1 (test for Helicobacter pylori and eradicate it if positive). What to do about gallstones found on a check-up that have never caused pain, see item 20 in this section (a check-up found gallstones you have never felt from, do not have a preventive cholecystectomy). For weight control, what really has evidence behind it is what you eat and how much, see item 27 in section 2 (5 portions of fruit and vegetables a day), item 28 in section 2 (eat less ultra-processed food) and item 32 in section 2 (keep BMI between 20–25). This does not apply to people with diabetes, people taking diabetes medication, or people who are pregnant; a missed meal can bring hypoglycaemia, so follow the doctor's instructions. Extreme fasting and induced vomiting are a different matter, see item 1 in section 28 (do not control your weight with extreme dieting, fasting or induced vomiting). + ### 27. Do not believe that masturbation harms the body or that abstaining improves you; what matters is whether it has got in the way of sleep, work and relationships - Cost: No money. What you save is the money on "tonify the kidney" remedies and "abstinence" courses, and the long worry of blaming yourself. @@ -220,6 +234,7 @@ This section collects the things that look cheap and effective. Open the origina - Evidence grade: B - Sources: Rider JR, Wilson KM, Sinnott JA, et al. (2016). Ejaculation Frequency and Risk of Prostate Cancer: Updated Results with an Additional Decade of Follow-up. European Urology, 70(6), 974–982. ; Dimitropoulou P, Lophatananon A, Easton D, et al. (2009). Sexual activity and prostate cancer risk in men diagnosed at a younger age. BJU International, 103(2), 178–185. ; Grubbs JB, Perry SL, Wilt JA, Reid RC. (2019). Pornography Problems Due to Moral Incongruence: An Integrative Model with a Systematic Review and Meta-Analysis. Archives of Sexual Behavior, 48(2), 397–415. - Notes:Disputed. Both prostate cancer studies are record-only with no control group, and the frequency is the man's own recollection, so the conclusion can only go as far as "no evidence of harm was seen"; it cannot be turned into advice to do it more. The authors of the British study say themselves that the fifties finding may be that people became ill first and then ejaculated less often. The abstract of the pornography paper gives no effect size, so only the direction is stated here. The benefit is graded small because what falls is mainly low-risk prostate cancer, not death. The signals that really should send you to a psychiatrist or a clinical psychologist are these: staying up late over it, missing work because of it, it damaging your relationship with your partner, and not being able to stop when you want to. For staying up late, see item 9 in section 3 (do not stay up late for games, short videos or pornography). If you are worried that watching pornography has caused erectile problems, get that checked first following item 28 in section 1 (if erectile function has gone wrong, check the cardiovascular system first). For the legal consequences of posting in groups or selling material, see item 4 in section 9 (watching pornography by yourself is your own business). The beneficiary is you yourself. + ### 28. Do not pay for "sexual orientation conversion" or "homosexuality treatment", and do not send family members either - Cost: No money. What you save is the money going into conversion organisations, psychological clinics and "ex-expose yourself" businesses. The hard part is the family insisting on sending someone. @@ -227,4 +242,4 @@ This section collects the things that look cheap and effective. Open the origina - Benefit: The American Psychological Association's 2009 task force report says that efforts to change sexual orientation are「不太可能成功，并且有一定的伤害风险」. On sending a minor into a coercive or residential conversion programme, the report advises that neither parents nor children should consider it. One nationally representative American survey visited 1518 sexual minority adults, about 7% of whom had been through conversion, and eight in ten of those came from religious people. After taking out age, adverse childhood experiences and similar factors, people who had been through conversion had an odds ratio of 1.92 for suicidal ideation at some point in their life (95% CI 1.01–3.64, about ninety per cent higher). Planning suicide OR 1.75 (1.01–3.06). Suicide attempts with no injury or minor injury OR 1.88 (1.01–3.50). Another study visited 245 people aged 21 to 25. Among those not converted during puberty, 22.0% had attempted suicide. Where parents themselves tried to change their child's sexual orientation, the figure was 48.1%. Where parents also took the child to a therapist or a religious figure, it was 62.8%. The World Health Organization's classification ICD-10 states that「性取向本身不被视为一种障碍」. In 2017 a court in Yicheng District, Zhumadian, Henan heard a case in which a psychiatric hospital forcibly admitted a man in the name of "sexual preference disorder". The court found that he「病情一般，也无自杀、伤人、毁物等行为或危险，不符合应当强制治疗的条件」. The court ordered the hospital to apologise publicly and pay 5000 yuan in mental comfort compensation (United States / China) - Evidence grade: B - Sources: American Psychological Association Task Force on Appropriate Therapeutic Responses to Sexual Orientation (2009). Report of the Task Force on Appropriate Therapeutic Responses to Sexual Orientation. ; Blosnich JR, Henderson ER, Coulter RWS, Goldbach JT, Meyer IH. (2020). Sexual Orientation Change Efforts, Adverse Childhood Experiences, and Suicide Ideation and Attempt Among Sexual Minority Adults, United States, 2016–2018. American Journal of Public Health, 110(7), 1024–1030. ; Ryan C, Toomey RB, Diaz RM, Russell ST. (2020). Parent-Initiated Sexual Orientation Change Efforts With LGBT Adolescents: Implications for Young Adult Mental Health and Adjustment. Journal of Homosexuality, 67(2), 159–173. ; World Health Organization. ICD-10 Version:2019, F66 Psychological and behavioural disorders associated with sexual development and orientation. ; Standing Committee of the National People's Congress (全国人大常委会) (2018, amended). Mental Health Law（精神卫生法）(Articles twenty-three, thirty and seventy-eight). ; Guangxi Zhuang Autonomous Region Procuratorate website reprinting the Beijing News (广西壮族自治区人民检察院网站转载新京报) (2017). Final verdict in the Zhumadian "coercive treatment of homosexuality" case（驻马店「同性恋强制治疗案」终审）. ; the other side: Sullins DP. (2022). Sexual Orientation Change Efforts Do Not Increase Suicide: Correcting a False Research Narrative. Archives of Sexual Behavior, 51, 3377–3393. ; Blosnich JR, Coulter RWS, Henderson ER, et al. (2023). Correcting a False Research Narrative: A Commentary on Sullins (2022). Archives of Sexual Behavior, 52(3), 885–888. -- Notes:Disputed. Sullins 2022 recalculated the same data and took out the suicidal ideation that was already there before the conversion. Worked out that way, conversion is no longer positively associated with suicidal ideation (OR 0.44, 0.20–0.94). The organisation Sullins belongs to is opposed to homosexuality, and the original authors wrote a reply. So how much harm there is is still disputed, but no evidence has come back from the other side on the point that it cannot work. It is graded B because the harm figures all come from surveys asking people to recall afterwards. As far as the law is concerned, inpatient treatment follows the principle of consent, and only a person with a severe mental disorder who is a danger to themselves or others can be admitted involuntarily. Psychological counsellors are not allowed to carry out psychotherapy. Deliberately sending someone who is not mentally ill into hospital for treatment carries a liability to compensate. If you are in long-term pain about this, go to a psychiatric or clinical psychology department; the goal is to relieve the pain. For what a family should do once a child has said it, see item 15 in section 30 (a child says they like the same sex). The beneficiaries are you and your family. \ No newline at end of file +- Notes:Disputed. Sullins 2022 recalculated the same data and took out the suicidal ideation that was already there before the conversion. Worked out that way, conversion is no longer positively associated with suicidal ideation (OR 0.44, 0.20–0.94). The organisation Sullins belongs to is opposed to homosexuality, and the original authors wrote a reply. So how much harm there is is still disputed, but no evidence has come back from the other side on the point that it cannot work. It is graded B because the harm figures all come from surveys asking people to recall afterwards. As far as the law is concerned, inpatient treatment follows the principle of consent, and only a person with a severe mental disorder who is a danger to themselves or others can be admitted involuntarily. Psychological counsellors are not allowed to carry out psychotherapy. Deliberately sending someone who is not mentally ill into hospital for treatment carries a liability to compensate. If you are in long-term pain about this, go to a psychiatric or clinical psychology department; the goal is to relieve the pain. For what a family should do once a child has said it, see item 15 in section 30 (a child says they like the same sex). The beneficiaries are you and your family. diff --git "a/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" "b/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" index 24918080..1ddbdabc 100644 --- "a/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" +++ "b/en/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" @@ -4,22 +4,24 @@ This section counts money and protection. How much you can receive, how much you can be spared, what services you can get and what conditions you have to meet are all written down here, and nothing is converted into years of life. The policies change every year and differ from place to place, so go by the amounts and the thresholds published for your own area in that year. The items are ordered like this: those that cost no money, need little paperwork and work fast come first. Those that take years of qualifying, and those that differ a lot between places, come later. Those resting on experience alone, with no official document, are put last. -### 1. If you lose your job, apply online for unemployment insurance benefits first +### 1. If you have lost your job, apply online first for unemployment insurance benefit -- Cost: No money. You can file from your phone, and any of three channels works: the「掌上 12333」app, the electronic social security card mini-program in WeChat or Alipay, and the national social insurance public service platform. A social security card or an identity card is all you need, and in most places you do not have to turn up in person. The precondition is that you have completed unemployment registration. -- In plain language: If you and your employer have paid in for a full year and you did not resign of your own accord, completing unemployment registration lets you receive money. A full 1 year but under 5 years gives you at most 12 months. 5 years but under 10 years gives at most 18 months. 10 years or more gives at most 24 months. While you are receiving it, unemployment insurance pays your basic medical insurance for employees. -- Benefit: Three conditions have to be met together. First, you and your employer must have paid unemployment insurance for a cumulative total of 1 year. Second, the job loss must not be one you asked for yourself (the law calls this "interruption of employment not of the person's own will"). Third, you must have completed unemployment registration and be willing to look for work. How long you can receive it for depends on how long you have paid in. A cumulative total of 1 year but under 5 years gives at most 12 months. 5 years but under 10 years gives at most 18 months. 10 years or more gives at most 24 months. The monthly amount is set by the provincial government and differs by province. The amount is below the local minimum wage and above the urban means-tested income support. The national goal is to raise it gradually to 90% of the minimum wage standard. While you are receiving unemployment insurance benefits you still take part in basic medical insurance for employees and enjoy basic medical insurance benefits. The basic medical insurance contribution you would otherwise have to pay comes out of the unemployment insurance fund, and you pay nothing yourself (nationwide, regulations in force since 1999) +- Cost: No money. You can submit the application on your phone, and there are three ways in: the 「掌上 12333」 app, the electronic social insurance card mini-program in WeChat or Alipay, and the National Social Insurance Public Service Platform. A social insurance card or an identity card is enough to do it, and in most areas you do not have to go in person. The condition is that you have completed unemployment registration first. +- In plain language: If your employer and you have paid in for a full year, and you did not resign of your own accord, you can get the money once you have completed unemployment registration. 1 year to under 5 years, at most 12 months. 5 to under 10 years, at most 18 months. 10 years or more, at most 24 months. While you are receiving it, unemployment insurance pays your medical insurance for employees. +- Benefit: Three conditions have to be met at the same time. First, your employer and you have paid unemployment insurance for a cumulative 1 year or more. Second, you did not resign of your own accord (the law calls this "employment interrupted not by the person's own wishes"). Third, you have completed unemployment registration and you are willing to look for work. How long you can receive it is calculated by the years you have paid in. A cumulative 1 year to under 5 years, at most 12 months. 5 to under 10 years, at most 18 months. 10 years or more, at most 24 months. How much is paid each month is set by the provincial government, and every province differs. The amount sits below the local minimum wage and above means-tested income support in cities. The national target is to raise it gradually to 90% of the minimum wage standard. While you are receiving unemployment insurance benefit you still take part in basic medical insurance for employees and get the benefits under it. The basic medical insurance contribution that would be due is paid out of the unemployment insurance fund, and you pay none of it yourself (nationwide, the regulations in force since 1999) - Evidence grade: A -- Sources: State Council (国务院) (1999). Regulations on Unemployment Insurance（失业保险条例）(State Council Order No. 258, Articles fourteen, seventeen and eighteen). ; Ministry of Human Resources and Social Security. Guide to Applying for Unemployment Insurance Benefits on the National Human Resources and Social Security Government Service Platform（全国人社政务服务平台·失业保险待遇申领办事指南）. ; Department of Employment of the National Development and Reform Commission (2022). The unemployment insurance benefit standard will gradually be raised to 90% of the minimum wage（失业保险金标准，将逐步提高至最低工资的 90%）. ; Liaoning Provincial Department of Human Resources and Social Security (2019). Notice on Further Adjusting the Unemployment Insurance Benefit Standard（关于进一步调整失业保险金标准的通知）:「为落实《人力资源社会保障部财政部关于调整失业保险金标准的指导意见》(人社部发〔2017〕71号〕"逐步将失业保险金标准提高到最低工资标准的90%"等要求」. ; Standing Committee of the National People's Congress (2010). Social Insurance Law（社会保险法）(Article forty-eight). (State Taxation Administration law database) -- Notes: If you resigned of your own accord, that does not count as "not of your own will", and you get nothing. While you are receiving benefits, unemployment insurance pays your basic medical insurance for employees. The amounts differ a great deal between provinces, so go by what the place you are insured in publishes. "90% of the minimum wage" comes from the 2017 guiding opinion of the Ministry of Human Resources and Social Security and the Ministry of Finance (人社部发〔2017〕71 号), whose words are「逐步将失业保险金标准提高到最低工资标准的90%」. That is a goal, not what is paid now. The original page on the ministry's website would not open on this machine, so that sentence was checked against the quotation in the Liaoning provincial document. -### 2. If your wages are owed, complain to labour inspection first, then apply for labour arbitration; neither route costs anything, and most cases resolve within months +- Sources: State Council (国务院) (1999). Regulations on Unemployment Insurance（失业保险条例）(State Council Order No. 258, Articles fourteen, sixteen, seventeen and eighteen; Article sixteen:「失业保险金自办理失业登记之日起计算」). ; Ministry of Human Resources and Social Security. Guide to Applying for Unemployment Insurance Benefits, National Human Resources and Social Security Government Service Platform（全国人社政务服务平台·失业保险待遇申领办事指南）. ; Employment Department of the National Development and Reform Commission (2022). The unemployment insurance benefit standard will be raised step by step to 90% of the minimum wage（失业保险金标准，将逐步提高至最低工资的 90%）. ; Liaoning Provincial Department of Human Resources and Social Security (2019). Notice on Further Adjusting the Standards for Unemployment Insurance Benefit（关于进一步调整失业保险金标准的通知）:「为落实《人力资源社会保障部财政部关于调整失业保险金标准的指导意见》(人社部发〔2017〕71号〕"逐步将失业保险金标准提高到最低工资标准的90%"等要求」. ; Standing Committee of the National People's Congress (全国人大常委会) (2010). Social Insurance Law（社会保险法）(Article forty-eight). (reprinted by the State Taxation Administration's database of policies and regulations) +- Notes: If you resigned of your own accord, that is not "employment interrupted not by the person's own wishes" and you cannot receive anything. So do not sign "voluntary resignation for personal reasons" when you leave, see item 7 in section 19. Unemployment insurance benefit is counted from the day you complete unemployment registration, so registering a month late means losing a month of money. The first thing to do after you leave is to go and register. Registration needs a certificate of leaving issued by the employer, see item 17 in section 19 (certificate of leaving). Whether a non-compete agreement has to be honoured, and what to do when the company will not pay compensation, see item 12 in section 11 (signed a non-compete agreement). While you are receiving it, unemployment insurance pays your medical insurance for employees. The amounts differ a great deal from province to province, so go by what the place you are insured in has published. The "minimum wage 90%" comes from the 2017 guiding opinions of the Ministry of Human Resources and Social Security and the Ministry of Finance (人社部发〔2017〕71 号), which say in so many words 「逐步将失业保险金标准提高到最低工资标准的90%」. That is a target, not what is paid at 90% now. The original page on the ministry's own website would not open, so that sentence was checked against the quotation inside the Liaoning provincial department's document. + +### 2. If your wages are owed to you, complain to labour inspection first and then apply for labour arbitration; neither route costs anything, and most cases finish within a few months -- Cost: No money. Neither a complaint nor arbitration costs anything. You need to assemble the employment contract, the payslips, the attendance records, the transfer records and your chat logs, and make 1 to 3 trips to the human resources bureau or the arbitration commission. There is a 1-year time limit on applying for arbitration (the law calls it the arbitration time limit). Someone still working for the same employer with wages in arrears is not bound by that 1 year. Someone who has already left must apply within 1 year of the leaving date. The hard part is going back again and again and assembling evidence piece by piece, and losing heart in the middle. -- In plain language: Neither route costs money. Call 12333 or complain to the labour inspection office at the human resources bureau; within 60 working days of the case being opened the investigation is finished, the company can be ordered to pay by a deadline, and if it still does not pay it has to pay 50% to 100% extra. If that does not work, apply for labour arbitration, which is closed within 45 days. But "closed" does not mean "in your account": if the company has no assets and the owner has run off, you may win and still get nothing. -- Benefit: The first step is to complain. Call 12333, or complain to the labour inspection office of the local human resources bureau. The law requires the investigation to be finished within 60 working days of the case being opened, with a possible further extension of 30 working days where the circumstances are complicated. If it is substantiated, the company can be ordered to pay within a deadline. If it still does not pay by the deadline, it has to pay an additional compensation of 50% to 100% of the amount owed. In 2024 inspection agencies nationwide handled 137,000 cases and recovered 86.5 billion yuan of wages for 654,000 people. If you have an IOU or a wage sheet and the facts of the arrears are clear, you can also apply directly to a basic-level court for a payment order. A payment order is a command to pay that the court issues to the company, and the court must issue it within 15 days of accepting the application. The second step is to apply for arbitration. The law requires the case to be closed within 45 days of acceptance, with a possible further extension of 15 days. Those claiming wages can apply for advance enforcement: part of what is awarded is enforced before the whole case has been heard. Where the money owed does not exceed 12 times the local monthly minimum wage, the award is final on issue and the company cannot sue. In 2024, 4,257,000 cases were handled nationwide and 4,156,000 were closed. The amount closed was 934.7 billion yuan. The mediation success rate was 79.6%, the arbitration closure rate 98.2% and the rate of awards going no further 73.7%. When people say this drags on for "half a year to a year", they mean the cases that go from arbitration into court: 6 months at first instance (3 months for those using the summary procedure), plus 3 months at second instance. In 2024 courts concluded 88,000 cases of claims for labour remuneration and recovered 285.4 billion yuan, and 998 people were sentenced for the crime of refusing to pay labour remuneration (nationwide) +- Cost: No money; neither the complaint nor the arbitration costs anything. You need your employment contract, payslips, attendance records, transfer records and chat records, and 1 to 3 trips to the human resources and social security bureau or the arbitration commission. Applying for arbitration has a time limit of 1 year (the law calls it the arbitration limitation period). Where you are still working for this employer and your wages are being withheld, that 1-year limit does not apply. If you have already left, you must apply within 1 year of the day you left. The hard part is making the trips one after another and assembling the evidence document by document, and losing heart in between. +- In plain language: Neither route costs money. First call 12333 or go to the labour inspection office at the human resources and social security bureau and make a complaint. It has to finish the investigation within 60 working days of opening the case, and it can order the company to pay by a deadline; if the company still does not pay, it has to pay 50% to 100% on top. If that does not work, apply for labour arbitration, which finishes within 45 days. But "case closed" does not mean "the money has arrived": if the company has no assets and the owner has run off, you may win and still get nothing. +- Benefit: The first step is a complaint. Call 12333, or go to the labour inspection office of the local human resources and social security bureau and make a complaint. The law requires the investigation to be finished within 60 working days of opening the case, and where the circumstances are complicated it may be extended by a further 30 working days. If the facts are confirmed, the office can order the company to pay by a deadline. If it still does not pay by the deadline, it has to pay an extra compensation of 50% to 100% of the amount due. In 2024 inspection agencies nationwide handled 137,000 cases and recovered 8.65 billion yuan in wages for 654,000 people. If you have an IOU or a wage sheet and the unpaid wages are clear, you can also apply straight to a basic-level court for a payment order. A payment order is the court's command to the company to pay, and the court issues it within 15 days of accepting the application. The second step is to apply for arbitration. The law requires the case to be closed within 45 days of acceptance, with a possible further extension of 15 days. Somebody claiming wages can apply for advance enforcement: part of the award is enforced without waiting for the whole case to be decided. Where the amount owed is no more than 12 times the local monthly minimum wage, the award is final on arbitration and the company cannot sue. In 2024, 4,257,000 cases were handled nationwide and 4,156,000 were closed. The amount closed was 93.47 billion yuan. The mediation success rate was 79.6%, the arbitration closure rate 98.2% and the arbitration finality rate 73.7%. When people say this drags on for "half a year to a year", they mean the cases that go on to court after the arbitration: 6 months at first instance (3 months under the summary procedure), then another 3 months at second instance. In 2024 courts closed 88,000 cases of claims for labour remuneration and recovered 28.54 billion yuan, and 998 people were sentenced for the crime of refusing to pay labour remuneration (nationwide) - Evidence grade: A -- Sources: Ministry of Human Resources and Social Security (人力资源社会保障部) (2025). 2024 Annual Statistical Bulletin on the Development of Human Resources and Social Security（2024 年度人力资源和社会保障事业发展统计公报）. ; Supreme People's Court (最高人民法院) (2025). Work Report of the Supreme People's Court（最高人民法院工作报告）. ; State Council (国务院) (2004). Regulations on Labour Security Inspection（劳动保障监察条例）(State Council Order No. 423, Articles nine, seventeen and twenty-six). ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Law on Mediation and Arbitration of Labour Disputes（劳动争议调解仲裁法）(主席令第八十号, Articles twenty-seven, forty-three, forty-four, forty-seven, fifty and fifty-three). ; Standing Committee of the National People's Congress (2012, amended). Labour Contract Law（劳动合同法）(Articles thirty and eighty-five). ; Standing Committee of the National People's Congress (2023, amended). Civil Procedure Law（民事诉讼法）(Articles one hundred and nine, one hundred and fifty-two, one hundred and sixty-four, one hundred and eighty-three, two hundred and twenty-five and two hundred and twenty-seven). ; Standing Committee of the National People's Congress (2011). Amendment (Eight) to the Criminal Law（刑法修正案（八））Article forty-one, that is, Article two hundred and seventy-six, paragraph one, of the Criminal Law（刑法第二百七十六条之一）. ; Supreme People's Court (2013). Interpretation on Several Questions of Law Concerning the Application of Law in Cases of Refusal to Pay Labour Remuneration（关于审理拒不支付劳动报酬刑事案件适用法律若干问题的解释）(法释〔2013〕3, Article three). ; State Council (2019). Regulations on the Payment of Wages to Migrant Workers（保障农民工工资支付条例）(State Council Order No. 724, Articles ten, twenty-six, thirty, thirty-one, thirty-two and forty-one). ; Standing Committee of the National People's Congress (2021). Legal Aid Law（法律援助法）(Articles thirty-one and forty-two). -- Notes: A complaint and arbitration are useful, but you have to keep "the case is closed" and "the money has arrived" apart. A 98.2% closure rate only says the procedure has run its course. If the company has no assets and the owner has run off, you can win and still get nothing. The Supreme People's Court publishes only the total actually enforced (2.3 trillion yuan in 2024) and says that the "rate of completed enforcement stays high"; there is no official figure for how much of the wage arrears cases ended up in the pockets of the people who won. The bulletin does not define "rate of final arbitration awards", so this item reads it as "the award was not taken to court afterwards". Wage arrears above a certain amount are enough for criminal prosecution. One case is owing one person more than 3 months, with an amount from 5000 to 2 万元 or more. The other is owing 10 or more people, a cumulative 3 万 to 10 万元 or more. In both situations, refusal to pay after being ordered to is enough for a case to be opened. The higher courts and procuratorates of each province set and publish the specific amounts within that range, so look up your own province's figure before complaining and raise it in person. Migrant workers on a site also have a dedicated wage account and the rule that the main contractor pays first, so going to the main contractor works better than going to the foreman. Legal aid for migrant workers claiming wages does not require an economic hardship check. The texts of the Labour Contract Law and the Civil Procedure Law are quoted from the reprints on the State Taxation Administration and Guizhou provincial market regulation websites (both gov.cn), because the pages on the National People's Congress website would not open. Amendment (Eight) to the Criminal Law failed to fetch on the National People's Congress website, so it was opened with curl and checked instead. +- Sources: Ministry of Human Resources and Social Security (2025). Statistical Bulletin on the Development of Human Resources and Social Security in 2024（2024 年度人力资源和社会保障事业发展统计公报）. ; Supreme People's Court (2025). Work Report of the Supreme People's Court（最高人民法院工作报告）. ; State Council (国务院) (2004). Regulations on Labour Security Supervision（劳动保障监察条例）(State Council Order No. 423, Articles nine, seventeen and twenty-six). ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Law on Mediation and Arbitration of Labour Disputes（劳动争议调解仲裁法）(主席令第八十号, Articles twenty-seven, forty-three, forty-four, forty-seven, fifty and fifty-three). ; Standing Committee of the National People's Congress (全国人大常委会) (2012, amended). Labour Contract Law（劳动合同法）(Articles thirty, eighty-five). ; Standing Committee of the National People's Congress (全国人大常委会) (2023, amended). Civil Procedure Law（民事诉讼法）(Articles one hundred and nine, one hundred and fifty-two, one hundred and sixty-four, one hundred and eighty-three, two hundred and twenty-five and two hundred and twenty-seven). ; Standing Committee of the National People's Congress (全国人大常委会) (2011). Criminal Law Amendment (Eight), 刑法修正案（八）第四十一条, that is Article two hundred and seventy-six, paragraph one of the Criminal Law. ; Supreme People's Court (2013). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Refusal to Pay Labour Remuneration（关于审理拒不支付劳动报酬刑事案件适用法律若干问题的解释）(法释〔2013〕3 号, Article three). ; State Council (国务院) (2019). Regulations on the Payment of Wages to Migrant Workers（保障农民工工资支付条例）(State Council Order No. 724, Articles ten, twenty-six, thirty, thirty-one, thirty-two and forty-one). ; Standing Committee of the National People's Congress (全国人大常委会) (2021). Legal Aid Law（法律援助法）(Articles thirty-one and forty-two). +- Notes: A complaint and arbitration are useful, but you have to tell "case closed" apart from "the money has arrived". The 98.2% closure rate says only that the procedure has run its course. If the company has no assets and the owner has run off, you can still win and still get nothing. The Supreme People's Court publishes only the total actually enforced (2.3 trillion yuan in 2024) and "the completion rate for enforcement stays high"; there is no official figure for how much of the unpaid wages in these cases actually reached the people who were owed them. The bulletin does not define "arbitration finality rate"; this item reads it as "not taken to court again after the award". Unpaid wages above a certain amount are enough for a criminal case to be opened. One kind is owing one person for more than 3 months, in an amount from 5000 to over 20,000 yuan. The other is owing 10 people or more, totalling from 30,000 to over 100,000 yuan. In both cases, if the order to pay is not obeyed, that is enough to open a case. The higher courts and procuratorates of each province set and publish the exact figure for their own province within that range, so look up your own figure before complaining and raise it in person when you do. Migrant workers on a construction site also have a separate rule on a dedicated wage account and on the main contractor paying first; going to the main contractor works better than going to the foreman. Both routes assume there is an employing entity. A workshop or shop without a licence still counts, and you can complain and go to arbitration. Private work for a household or an individual, such as working as a nanny, is not a labour relationship and has to go to court: see item 22 in this section (first find out who employed you). Migrant workers applying for legal aid to recover wages are not asked whether they are in financial difficulty. The text of the Labour Contract Law and the Civil Procedure Law is quoted from the reprints on the State Taxation Administration's and the Guizhou provincial market regulation bureau's websites (both gov.cn), because the pages on the National People's Congress website would not open. The Criminal Law Amendment (Eight) could not be fetched from the National People's Congress website and was checked by opening it with curl instead + ### 3. If you cannot afford to sue, apply for legal aid; wage claims, support payments and work injury cases are already within scope - Cost: No money. Legal aid charges the person receiving it not one yuan. @@ -36,6 +38,7 @@ This section counts money and protection. How much you can receive, how much you - Evidence grade: A - Sources: State Council (国务院) (2003). Measures for the Assistance and Management of Homeless and Begging Persons with Nowhere to Live in Cities（城市生活无着的流浪乞讨人员救助管理办法）(State Council Order No. 381, Articles five, six and seven). ; Ministry of Civil Affairs (民政部) (2003). Detailed Implementation Rules for the Measures for the Assistance and Management of Homeless and Begging Persons with Nowhere to Live in Cities（城市生活无着的流浪乞讨人员救助管理办法实施细则）(Ministry of Civil Affairs Order No. 24, Articles eleven and twelve). - Notes: Seeking help and leaving are both voluntary; it is not a detention centre. What it provides is basic food and a bed, and no cash. The travel voucher is issued to your registered place of residence or your place of residence, so work out where you are going before you set off. + ### 5. To find work, start with free public employment services and the day labour market, not with an agency that charges - Cost: No money. Register with a public employment service body at county level or above, such as an employment service centre or a human resources market. The local day labour market will also take registrations. If you do not know the address, call 12333 and ask. @@ -52,14 +55,15 @@ This section counts money and protection. How much you can receive, how much you - Evidence grade: B - Sources: State Council (国务院) (2014, revised 2019). Interim Measures on Social Assistance（社会救助暂行办法）(State Council Order No. 649, Articles forty-seven, forty-eight and forty-nine). - Notes: You do not have to be on means-tested income support already. More often than not it is a single payment, and the amount differs a great deal from place to place, so go by what your area published that year. -### 7. If your income is below the local means-tested line, apply for means-tested income support +### 7. If your income is below the local means-tested income support line, apply for means-tested income support -- Cost: No money. A written application is made by the members of the household who live together, to the township government or sub-district office where their registered residence is. If it is awkward to do it yourself, you can ask the village committee or the residents' committee to do it for you. You have to accept a check of household income and property. It usually takes 2 to 3 trips, and you bring your identity card, household register and proof of income and property. The hard part is laying your finances open for inspection, and then waiting through the public notice. -- In plain language: Divide total family income by the number of people; if it is below the local means-tested line and the household property also falls within the rules, you can apply. At the end of 2024, urban means-tested income support averaged 798.1 yuan per person per month nationwide and rural support 593.9 yuan. You apply where your registered residence is. Having a house, a car or savings, or children who can support you, may mean you are not approved. -- Benefit: Divide total family income by the number of people; if it is below the local means-tested standard and the household property is also within the prescribed range, you can receive means-tested income support. At the end of 2024, the national urban average was 798.1 yuan per person per month and the rural average 593.9 yuan per person per month. Urban means-tested income support covered 6,250,000 people and rural support covered 33,615,000 people. The standard is set like this: take the previous year's per-capita consumption expenditure of local residents, multiply it by a set proportion, and adjust it again after an interval. Put plainly, take how much the average local resident spent the previous year and discount it (national average, different in each place) +- Cost: No money. A family member who lives with you submits one written application to the township government or the sub-district office where your household registration is. If it is awkward for you to do it yourself, you can ask the village committee or the residents' committee to do it for you. You have to accept a check of the family's income and property. It usually takes 2 to 3 trips, with your identity card, household register and proof of income and property. The hard part is laying your finances open to be checked, and then waiting through the public notice. +- In plain language: Divide the whole family's income by the number of people; if the result is below the local means-tested income support line and the family's property is also within the rules, you can apply. At the end of 2024, the urban line averaged 798.1 yuan per person per month nationwide and the rural line 593.9 yuan. What is paid is the difference between the line and what each member of the household has per month. If you own a house, a car and savings, or your children can support you, the application may be turned down. +- Benefit: Divide the whole family's income by the number of people; if the result is below the local means-tested income support standard and the family's property is also within the range set, you can receive means-tested income support. At the end of 2024, the average urban standard was 798.1 yuan per person per month and the rural standard 593.9 yuan. Urban means-tested income support covered 6,250,000 people and rural support covered 33,615,000. The payment is made monthly and is the difference: the local standard minus the household's income per head. If the local standard is 798.1 yuan and everyone in the household has 500 yuan a month, each person receives 298.1 yuan a month, not 798.1 yuan. The standard is set this way: take the local per-capita consumption expenditure of the previous year, multiply it by a set ratio, and adjust it again from time to time. Put plainly, it is what the average person locally spent the year before, worked out with a discount (national average, different in every place) - Evidence grade: A -- Sources: State Council (国务院) (2014, revised 2019). Interim Measures on Social Assistance（社会救助暂行办法）(State Council Order No. 649, Articles nine, ten and eleven). ; National Bureau of Statistics (国家统计局) (2025). Social Undertakings Developing Well and People's Wellbeing Rising: Statistical Report on Livelihood since the「十四五」period（社会事业向好发展 民生福祉不断增进——「十四五」以来社会民生统计报告）. ; Ministry of Civil Affairs and three other departments (民政部等四部门) (2024). Guiding Opinions on Further Improving the Determination and Adjustment of the Minimum Living Guarantee Standard（关于进一步做好最低生活保障标准确定调整工作的指导意见）(民发〔2024〕16). -- Notes: Means-tested income support is the last line of protection, and you have to apply where your registered residence is. If the house, car or savings at home exceed the local standard, or your children can support you, you may not be approved. Once it is approved there are other benefits as well: a subsidy towards resident medical insurance (see item 9 in this section for resident medical insurance), medical assistance when you are ill (see item 10 in this section for medical assistance), and no economic hardship check for legal aid (see item 3 in this section for legal aid). +- Sources: State Council (国务院) (2014, revised 2019). Interim Measures on Social Assistance（社会救助暂行办法）(State Council Order No. 649, Articles nine, ten and eleven; Article twelve:「县级人民政府民政部门按照共同生活的家庭成员人均收入低于当地最低生活保障标准的差额，按月发给最低生活保障金」). ; National Bureau of Statistics (2025). Social services developing well and public wellbeing rising — a statistical report on social livelihood since the 「十四五」. ; Ministry of Civil Affairs (民政部) and four other departments (2024). Guiding Opinions on Further Improving the Determination and Adjustment of the Minimum Living Guarantee Standard（关于进一步做好最低生活保障标准确定调整工作的指导意见）(民发〔2024〕16 号). +- Notes: Means-tested income support is the last line of defence, and you have to apply at your registered place of residence. If the house, car or savings at home go over the local standard, or your children are able to support you, the application may be turned down. Once it is approved there are other benefits as well: there is a subsidy for resident medical insurance (resident medical insurance: see item 9), medical care can go through medical assistance (medical assistance: see item 10), and applying for legal aid does not require a check on financial hardship (legal aid: see item 3). + ### 8. If you hold a disability certificate, apply for the two disability subsidies - Cost: No money. Take your disability certificate to the sub-district office or the township and submit one application. @@ -76,14 +80,16 @@ This section counts money and protection. How much you can receive, how much you - Evidence grade: A - Sources: National Healthcare Security Administration (国家医保局) (2024). Policy Interpretation of the Notice on Doing Well in 2024 on Basic Medical Insurance for Urban and Rural Residents（关于做好 2024 年城乡居民基本医疗保障有关工作的通知政策解读）(医保发〔2024〕19). ; Hebei Provincial Healthcare Security Administration and others (2025). Notice on Doing Well in 2025 on Basic Medical Insurance for Urban and Rural Residents（关于做好 2025 年城乡居民基本医疗保障有关工作的通知）(冀医保发〔2025〕6, a local document). ; Anhui Provincial Healthcare Security Administration and others (2025). Notice on Doing Well in 2025 on Basic Medical Insurance for Urban and Rural Residents（关于做好 2025 年城乡居民基本医疗保障有关工作的通知）, reprinted by the Huaibei Municipal Healthcare Security Bureau:「按照国家医保局、财政部、国家税务总局《关于做好2025年城乡居民基本医疗保障有关工作的通知》（医保发〔2025〕22号）要求」「居民医保人均财政补助标准较上年提高30元，达到每人每年不低于700元」. ; General Office of the State Council (国务院办公厅) (2021). Opinions on Improving the Medical Insurance and Assistance System for Major and Serious Diseases（关于健全重特大疾病医疗保险和救助制度的意见）(国办发〔2021〕42). ; General Office of the State Council (2024). Guiding Opinions on Perfecting the Long-Term Mechanism for Enrolment in Basic Medical Insurance（关于健全基本医疗保险参保长效机制的指导意见）(国办发〔2024〕38). ; National Healthcare Security Administration (2026). The 2026 Government Work Report Mentions These Healthcare Security Tasks（2026 年政府工作报告提到这些医保工作）. - Notes: The 2025 notice of the National Healthcare Security Administration (医保发〔2025〕22) has not been published on the administration's own website or on the Chinese government website. The two figures, 700 yuan and 400 yuan, agree in the documents of Hebei and Anhui, and the Anhui document states that they were set by this notice. While you are receiving unemployment insurance benefits, unemployment insurance pays your medical insurance for employees, so there is no need to enrol twice. -### 10. If you fall seriously ill, go through medical insurance, insurance for serious illness, medical assistance and out-of-area registration first, and stay away from online loans + +### 10. If you fall seriously ill, go through medical insurance, the insurance for serious illness, medical assistance and out-of-area registration first, and do not touch online loans -- Cost: No money. The order is fixed, and you work through it step by step. First, before being admitted in another province, register using the "national out-of-area medical registration" mini-program in the "National Healthcare Security Administration" WeChat account. Second, settle directly on discharge. Third, if what you still have to pay after reimbursement is too heavy, apply to the township government or the sub-district office for medical assistance. Fourth, if you are still working, take sick leave for the medical period, which is the period of being unable to work because of illness during which you cannot be dismissed; if it is a work injury, go through work injury insurance. Fifth, if you cannot pay the credit card, ask the bank to negotiate instalments. Sixth, if it is still not enough, use a personal fundraising platform designated by the Ministry of Civil Affairs. The main documents are the diagnosis certificate, the itemised bill and the settlement statement, and it usually takes 2 to 4 trips. The hard part is that you are ill and still have to go back and forth for the paperwork. -- In plain language: The order is fixed. Before being admitted in another province, register in the "National Healthcare Security Administration" WeChat account; settle directly on discharge and you do not have to pay the whole amount up front first. If you still cannot afford it after reimbursement, apply to the township or the sub-district office for medical assistance. If you are still employed, your length of service gives you a medical period of 3 to 24 months during which the company cannot dismiss you. Do not touch online loans or private high-interest borrowing. -- Benefit: The state brings basic medical insurance, insurance for serious illness and medical assistance together under what it calls the "three-tier system", which catches you layer by layer. For people on means-tested income support and people in extreme difficulty, medical expenses that meet the conditions are assisted at a proportion of no less than 70%. A severely ill patient who was not poor and has been made poor by medical bills can apply for assistance again on their own initiative. If you registered in advance for treatment in another province, you can settle directly on discharge, without paying the whole amount up front and then claiming in your home province. Being unable to work because of illness gives a medical period of 3 to 24 months, with the exact length set by length of service. During that period the company may not terminate the employment contract, and sick pay may not be below 80% of the local minimum wage. A worker with a work injury has a period of retention pay with no work, which is the time spent recovering and not working; the original wages and benefits continue to be paid, by the employer, monthly, and generally for no more than 12 months. Where a credit card balance exceeds what you can repay and you are willing to pay, you can negotiate an instalment plan with the bank and have the balance rescheduled, for up to 5 years, and once the agreement is signed the bank must stop its collection activity. A personal fundraising platform raising money for a named person must be designated by the civil affairs department of the State Council. The platform has to verify that the request for help is real. Money raised can only be paid to the person asking for help, or to an account that person provides (nationwide) +- Cost: No money. The order is fixed, one step at a time. First, before being admitted to hospital in another province, file the registration in the 「国家异地就医备案」 mini-program in the WeChat account of the National Healthcare Security Administration (国家医保局). Second, settle the bill directly on discharge. Third, if what you still have to pay after reimbursement is still too much, apply to the township government or the sub-district office for medical assistance. Fourth, if you are still working, take sick leave for the medical period; the medical period is the stretch during which you are off work because of illness and cannot be dismissed, and if it is a work injury you go through work injury insurance. Fifth, if you cannot pay off the credit card, ask the bank for an instalment plan yourself. Sixth, if it is still not enough, raise money on a personal fundraising platform designated by the Ministry of Civil Affairs. The documents are mainly the diagnosis certificate, the list of expenses and the settlement statement, and it takes 2 to 4 trips. The hard part is being ill and having to make the trips for the paperwork one after another. +- In plain language: The order is fixed. Before being admitted to hospital in another province, file the registration in the National Healthcare Security Administration WeChat account and settle directly on discharge, so you do not have to pay the whole amount up front. If you still cannot afford it after reimbursement, apply to the township or the sub-district office for medical assistance. While you are employed you have a medical period of 3 to 24 months, depending on how long you have worked, and the company cannot dismiss you during it. Do not touch online loans or private high-interest borrowing. +- Benefit: The state calls basic medical insurance, the insurance for serious illness and medical assistance together the "three-tier system", each tier catching what the tier above did not. For people on means-tested income support and people in extreme poverty, medical costs that meet the conditions are assisted at no less than 70%. A seriously ill patient who was not poor before and has been made poor by being treated can apply for further assistance on their own initiative. If you filed the registration before being admitted to hospital in another province, you can settle directly on discharge instead of paying the whole amount up front and claiming it back afterwards where you are insured. Being off work through illness carries a medical period of 3 to 24 months, the exact length set by how many years you have worked. During it the company may not terminate the employment contract, and sick pay may not be less than 80% of the local minimum wage. A worker with a work injury has a period of paid leave for recovery, which is the stretch spent healing without working; the original wages and benefits still go ahead, paid monthly by the unit, and generally for no more than 12 months. Where credit card debt goes beyond your ability to repay and you are willing to pay it, you can negotiate a personalised instalment plan with the bank, laying the debt out again as a repayment plan, for up to 5 years, and once the agreement is signed the bank has to stop chasing you for the money. A personal fundraising platform raising money online for the person themselves has to be designated by the civil affairs department of the State Council. The platform has to check whether the request for help is genuine. Money that comes in can only be paid to the person asking for help, or to an account that person has provided (nationwide) - Evidence grade: A -- Sources: General Office of the State Council (国务院办公厅) (2021). Opinions on Improving the Medical Insurance and Assistance System for Major and Serious Diseases（关于健全重特大疾病医疗保险和救助制度的意见）(国办发〔2021〕42). ; National Healthcare Security Administration (国家医保局) (2025). How Do I Get Reimbursed Directly for Treatment in Another Province?（跨省异地就医如何直接报销？）. ; Ministry of Labour (劳动部) (1994). Provisions on the Medical Period for Employees of Enterprises Who Fall Ill or Suffer Non-occupational Injury（企业职工患病或非因工负伤医疗期规定）(劳部发〔1994〕479, Articles two and three). ; Ministry of Labour (劳动部) (1995). Opinions on Certain Questions Concerning the Implementation of the Labour Law（关于贯彻执行《劳动法》若干问题的意见）(劳部发〔1995〕309, item 59). (reprinted by the Jilin provincial human resources department) ; State Council (国务院) (2010, revised). Regulations on Work Injury Insurance（工伤保险条例）(State Council Order No. 586, Articles thirty and thirty-three). ; Banking Regulatory Commission of China (银监会) (2011). Measures for the Supervision and Administration of Credit Card Business of Commercial Banks（商业银行信用卡业务监督管理办法）(CBRC Order No. 2 of 2011, Article seventy). ; Standing Committee of the National People's Congress (全国人大常委会) (2023). Decision on Amending the Charity Law（关于修改慈善法的决定）(Article one hundred and twenty-four). ; Ministry of Civil Affairs and four other departments (民政部等五部门) (2024). Measures for the Administration of Personal Fundraising Network Service Platforms（个人求助网络服务平台管理办法）(Order No. 75, Articles three, eleven, thirteen and seventeen). -- Notes: What proportion medical assistance pays and what the maximum is in a year are set by the place where you are insured. The medical period comes in six tiers, depending on how many years you have worked in total and how long you have been with this employer. Once the medical period is over, the company can lawfully end your contract, so see clearly which tier you are in. People rescued in an emergency are also on the list of those who need to register. The healthcare security administration's pages do not say whether an emergency counts as automatic registration or whether it can be done afterwards, so telephone the place where you are insured. Stay away from online loans and private high-interest borrowing. The courts protect only up to the line of 4 times the one-year LPR (for not borrowing at usurious rates, see item 15 in this section). The part above that line is not protected, and the principal still has to be repaid. Lying in a hospital bed you have no energy to bring such a lawsuit either. Acute serious illness is a different matter: you must still call 120 even if you have no money, the hospital may not refuse, pass the buck on or delay treatment, and the emergency costs for that part are paid by the emergency medical assistance fund; see item 9 in section 24 (an emergency must be treated first). The original page of Lao Bu Fa [1995] 309 would not open on the human resources ministry website, so its content was taken from a provincial reprint. +- Sources: General Office of the State Council (国务院办公厅) (2021). Opinions on Improving the Medical Insurance and Assistance System for Major and Serious Diseases（关于健全重特大疾病医疗保险和救助制度的意见）(国办发〔2021〕42 号). ; National Healthcare Security Administration (国家医保局) (2025). How do you get reimbursed directly for treatment in another province? ; Ministry of Labour (劳动部) (1994). Provisions on the Medical Period for Enterprise Employees Who Fall Ill or Are Injured away from Work（企业职工患病或非因工负伤医疗期规定）(劳部发〔1994〕479 号, Articles two and three). ; Ministry of Labour (劳动部) (1995). Opinions on Several Questions Concerning the Implementation of the Labour Law（关于贯彻执行《劳动法》若干问题的意见）(劳部发〔1995〕309 号, item 59). (reprinted by the Jilin provincial human resources and social security department) ; State Council (国务院) (2010, revised). Regulations on Work Injury Insurance（工伤保险条例）(State Council Order No. 586, Articles thirty and thirty-three). ; China Banking Regulatory Commission (银监会) (2011). Measures for the Supervision and Administration of Commercial Bank Credit Card Business（商业银行信用卡业务监督管理办法）(CBRC Order No. 2 of 2011, Article seventy). ; Standing Committee of the National People's Congress (全国人大常委会) (2023). Decision on Amending the Charity Law（关于修改慈善法的决定）(Article one hundred and twenty-four). ; Ministry of Civil Affairs (民政部) and five other departments (2024). Measures for the Administration of Personal Fundraising Service Platforms Online（个人求助网络服务平台管理办法）(Order No. 75, Articles three, eleven, thirteen and seventeen). +- Notes: What proportion medical assistance pays, and the most it pays in a year, are set by the place you are insured in. The medical period comes in six bands, decided by how many years you have worked altogether and how many you have worked at this unit. Once the medical period is over the company may lawfully terminate your contract, so work out in advance which band you are in. People taken in as an emergency are also on the list of people who have to file the registration. The National Healthcare Security Administration's page does not say whether an emergency counts as automatic registration, or whether it can be filed afterwards, so telephone the place you are insured in. Do not touch online loans or private high-interest borrowing. The courts protect lending only up to four times the one-year LPR (on not borrowing at a high rate of interest: see item 15). The part above that line is not protected, and the principal still has to be repaid. Lying in a hospital bed you have no energy to bring that kind of case anyway. A sudden serious illness or injury is a different matter: you have to call 120 whether you have money or not, and the hospital may not refuse, shirk or delay treatment, and the emergency costs for that part are paid by the emergency medical assistance fund; see item 8 in section 24 (an emergency has to be treated first). The original page of 劳部发〔1995〕309 号 would not open on the human resources ministry's own website, and the text is taken from a provincial reprint. + ### 11. If you lose your identity card, replace it at once; if you need it urgently, get a temporary identity card - Cost: There is a document handling fee. This is a statutory charge, and the amount is set by the state, so go by what your area publishes. Apply at the police station where your registered residence is; it takes 1 trip. @@ -116,6 +122,7 @@ This section counts money and protection. How much you can receive, how much you - Evidence grade: A - Sources: Liu S, Huang JL, Wang M (2014). Effectiveness of job search interventions: a meta-analytic review. Psychological Bulletin 140(4): 1009–1041. - Notes: Most of these studies were done in Europe and the United States, in the form of packaged small-group courses. Moving them to China is applying results from elsewhere, not measurements taken here. Seeking social support, in practice, means getting an acquaintance to introduce you and refer you inside; do not be embarrassed to ask. + ### 15. Do not pay deposits, do not hand over documents, do not sign a "training loan", do not join a pyramid scheme, do not borrow at usurious rates - Cost: No money. Refuse on the spot when you hear "pay first and then start work", "train first and then we find you a job", or "recruit heads". Keep the chat logs and the transfer receipts. The hard part is that the more desperate you are for money, the more you want to gamble, and saying "no" there and then is not easy. @@ -156,14 +163,16 @@ This section counts money and protection. How much you can receive, how much you - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2026). Prisons Law（监狱法）(revised 2026, Articles fifty-nine, sixty and sixty-two). ; General Offices of the CPC Central Committee and of the State Council (中共中央办公厅、国务院办公厅) (2010). Notice Forwarding the Opinions of the Central Committee on Comprehensive Governance of Public Security on Further Strengthening the Resettlement and Assistance of Persons Released After Serving a Sentence or Released from Labour Education（转发《中央社会治安综合治理委员会关于进一步加强刑满释放解除劳教人员安置帮教工作的意见》的通知）(中办发〔2010〕5). (reprinted by the Ziyang district government of Yiyang) ; Standing Committee of the National People's Congress (2018). Criminal Procedure Law（刑事诉讼法）(2018 amendment, Article two hundred and eighty-six). ; Supreme People's Court, Supreme People's Procuratorate, Ministry of Public Security and Ministry of Justice (2022). Implementation Measures for the Sealing of Criminal Records of Minors（关于未成年人犯罪记录封存的实施办法）(Articles nine and fifteen). (reprinted by the Hunan provincial public security department) ; Supreme People's Court (2017, amended). Several Provisions on the Publication of Information on the List of Judgment Defaulters（关于公布失信被执行人名单信息的若干规定）(Articles two and ten). ; Standing Committee of the Shenzhen Municipal People's Congress (深圳市人大常委会) (2020). Regulations on Personal Bankruptcy in the Shenzhen Special Economic Zone（深圳经济特区个人破产条例）(Articles two, twenty-three, ninety-five, ninety-seven and one hundred and three). ; Standing Committee of the National People's Congress (全国人大常委会) (2006). Enterprise Bankruptcy Law（企业破产法）(主席令第五十四号, Articles two and seven). - Notes: An adult's criminal record is not sealed. Article one hundred of the Criminal Law requires a truthful report of any criminal penalty when joining the army or taking a job. Civil service, judges, lawyers and teachers have further statutory restrictions in their vetting. The text of the Prisons Law in force before 1 November 2026 corresponds to Articles thirty-seven and thirty-eight, so go by the article numbers in the text as it comes into force. The original page of Zhong Ban Fa [2010] 5 could not be found on the central government website, so it was taken from the district government reprint. During the observation period for personal bankruptcy you may not take first class on a plane, spend heavily or buy a house; compensation for deliberately harming someone's person or one of the 8 other kinds of debt is not written off; and an exemption obtained by fraud may still be revoked. Being removed from the judgment defaulter list does not mean your credit record becomes clean as well; those are two separate systems. -### 20. Before serious illness strikes, add one-year medical insurance or critical illness insurance on top of basic medical insurance, and check the four words "guaranteed renewal" + +### 20. Before serious illness, put a one-year medical insurance policy or a critical illness policy in place on top of basic medical insurance, and look for the four words 「保证续保」 -- Cost: For a young person a one-year expense-reimbursement medical policy costs several hundred yuan a year. An expense-reimbursement policy is one that reimburses what you actually spent. Critical illness insurance starts at several thousand yuan a year (the author's rough estimate, no official figure). You have to disclose your health condition truthfully before taking out the policy, which takes half an hour. -- In plain language: This kind of insurance covers the part you have to pay yourself after medical insurance has reimbursed, and the loss of earnings while you cannot work. Before buying, look for the four words "guaranteed renewal": without them, the product may be withdrawn or the price raised next year. You have to disclose your health condition truthfully, and if you conceal it the claim can be refused. Put basic medical insurance in place first, then think about this. -- Benefit: The authorities divide this kind of product into two kinds. Medical insurance reimburses the medical costs you actually incurred. Disease insurance pays a lump sum in the sum stated in the contract as soon as one of the conditions listed there is diagnosed. Both cover the part you have to pay yourself after medical insurance has reimbursed, and the loss of earnings while you cannot work. Insurance with a period of one year or less, where the terms do not also state guaranteed renewal, is short-term health insurance. Only insurance of more than one year, or insurance that states guaranteed renewal, is long-term health insurance. "Guaranteed renewal" means that on renewal at the end of the period the company must continue the cover on the same terms and at the agreed rate. An insurance company may not induce the same person to buy again a expense-reimbursement medical policy with the same or a similar protection. If you buy two, the same bill can only be reimbursed once. The disease definitions for critical illness insurance follow an industry-wide standard, which includes 28 severe conditions and 3 mild conditions. Malignant tumours, acute heart attack and the after-effects of a stroke are the core conditions (nationwide, from 2019 and 2020) +- Cost: For a young person, a one-year expense-reimbursement medical insurance policy costs several hundred yuan a year. Expense reimbursement means it pays back what you have actually spent. A critical illness policy starts at several thousand yuan a year (the author's rough estimate, no official figure). Before taking out the policy you have to disclose your health condition truthfully, which takes half an hour in one go. +- In plain language: This kind of insurance makes up what you still have to pay yourself after medical insurance has reimbursed, and the earnings you lose while you are off work. Before you buy, look for the four words 「保证续保」, meaning the renewal is guaranteed: without those four words the policy may be withdrawn from sale or repriced next year. You have to disclose your health condition truthfully; if you hide it, the claim will be refused. Take out basic medical insurance first, and consider this only afterwards. +- Benefit: Officially these products come in two kinds. Medical insurance reimburses the medical costs you have actually spent. Disease insurance pays a single lump sum of the amount agreed in the contract as soon as one of the diseases named in the contract is diagnosed. Both make up what you still have to pay yourself after medical insurance has reimbursed, and the earnings you lose while you are off work. A policy whose insurance period is one year or less, and whose terms do not say the renewal is guaranteed, is short-term health insurance. Only one that runs for more than a year, or one whose terms say the renewal is guaranteed, is long-term health insurance. A guaranteed renewal means that when the policy is renewed at the end of the period, the company has to go on at the original terms and the agreed rate. An insurance company may not induce the same person to buy again an expense-reimbursement medical insurance policy with the same or a similar cover. If you buy two, the same money can only be reimbursed once. There is an industry-wide standard for the disease definitions used in a critical illness policy, covering 28 serious conditions and 3 mild ones. Malignant tumour, acute myocardial infarction and the after-effects of a stroke are the core conditions (nationwide, from 2019 and 2020) - Evidence grade: B -- Sources: formerly the China Banking and Insurance Regulatory Commission (银保监会) (2019). Measures for the Administration of Health Insurance（健康保险管理办法）(CBIRC Order No. 3 of 2019, Articles two, four and forty-one). ; China Insurance Industry Association and Chinese Medical Doctor Association (2020). Answers to Reporters' Questions on the Use Standard for the Disease Definitions of Critical Illness Insurance (2020 Revision)（重大疾病保险的疾病定义使用规范（2020 年修订版）答记者问）. ; formerly the China Banking and Insurance Regulatory Commission (2020). Notice on Matters Concerning the Use of the Empirical Incidence Rate Table for Critical Illness in Chinese Personal Insurance (2020)（关于使用《中国人身保险业重大疾病经验发生率表（2020）》有关事项的通知）(银保监发〔2020〕51). ; Standing Committee of the National People's Congress (全国人大常委会) (2015, amended). Insurance Law of the People's Republic of China（中华人民共和国保险法）, Article sixteen (truthful disclosure, and no termination after two years from the formation of the contract). -- Notes: It is graded B because the authorities have only given definitions and rules, with no figure for "how much you save by buying this". "Million-yuan medical insurance" is the name used when it is sold; as far as the rules are concerned it counts as short-term health insurance. If the terms do not say "guaranteed renewal", the product may be withdrawn or the price raised next year, so look for those four words in the terms before you buy. Two expense-reimbursement medical policies cannot reimburse the same bill twice, so do not stack them. Taking out a policy while already ill and then failing to disclose it means the claim can be refused. Truthful disclosure has a two-year limit: once the contract has existed for two years the insurer can no longer terminate it on the ground that you did not disclose, and it must pay a valid claim. Where deliberate non-disclosure is found within the two years, events occurring before the contract is terminated are not covered and the premium is not refunded. For which losses are worth insuring, see item 40 in section 5 (losses you could not carry). Put resident medical insurance in place first (item 9 in this section), then think about this. The disease definition standard is published by the industry association under the guidance of the banking and insurance regulator; it is not a government document. +- Sources: China Banking and Insurance Regulatory Commission (银保监会) (2019). Measures for the Administration of Health Insurance（健康保险管理办法）(CBIRC Order No. 3 of 2019, Articles two, four and forty-one). ; Insurance Association of China (中国保险行业协会) and Chinese Medical Doctor Association (中国医师协会) (2020). Answers to Reporters' Questions on the Specifications for the Use of Disease Definitions in Critical Illness Insurance（重大疾病保险的疾病定义使用规范（2020 年修订版）答记者问）. ; China Banking and Insurance Regulatory Commission (银保监会) (2020). Notice on Matters Concerning the Use of the Experience Rate Table for Critical Illnesses of Life Insurance in China (2020), 关于使用《中国人身保险业重大疾病经验发生率表（2020）》有关事项的通知（银保监发〔2020〕51 号）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015, amended). Insurance Law of the People's Republic of China（中华人民共和国保险法）Article sixteen (on truthful disclosure, and no termination two years after the contract is formed). +- Notes: The grade is B because the official documents give only definitions and rules, with no figure for "how much you save by having bought it". "Million-yuan medical insurance" is the name used when it is sold; under the regulations it counts as short-term health insurance. If the terms do not say 「保证续保」, it may be withdrawn from sale or repriced next year, so look for those four words in the terms before you buy. Two expense-reimbursement medical insurance policies cannot reimburse the same money twice, so do not stack them. If you already have an illness and take out a policy without disclosing it truthfully, the claim will be refused when something happens. Truthful disclosure has a two-year limit: once the contract has been in force for two years, the insurance company can no longer terminate it on the ground that you did not disclose truthfully, and it has to pay out. If deliberate non-disclosure is found within those two years, a case arising before the contract is terminated is not covered and the premium is not refunded. Which losses are worth putting insurance behind, see item 39 in section 5 (losses you cannot carry). Take out resident medical insurance (item 9) first, and consider this only afterwards. The specifications for the disease definitions are issued by the industry association under the guidance of the regulator; they are not a government document. + ### 21. If you really have nowhere to sleep, a 24-hour premises is only good for one emergency night; go to a rescue station first - Cost: The waiting areas at airports and railway stations usually cost nothing. Fast food restaurants and internet cafés require you to spend a little. You will not sleep properly and there is a risk of having your belongings stolen. The hard part is not daring to really sleep all night. @@ -171,4 +180,12 @@ This section counts money and protection. How much you can receive, how much you - Benefit: It sees you through the night, there are toilets, and you can charge your phone. No official figure - Evidence grade: C - Sources: the author's own experience, no official document -- Notes: The safety risk is real. Keep your belongings on your body, do not take drink or food offered by strangers, and avoid secluded corners. Do not sleep rough under bridges, by rivers or on building sites at night. Sleeping rough in extreme weather can kill. The rescue station in item 4 provides food and a bed free, so go there first. Some premises clear everyone out in the middle of the night; do not pick a fight with the security staff. \ No newline at end of file +- Notes: The safety risk is real. Keep your belongings on your body, do not take drink or food offered by strangers, and avoid secluded corners. Do not sleep rough under bridges, by rivers or on building sites at night. Sleeping rough in extreme weather can kill. The rescue station in item 4 provides food and a bed free, so go there first. Some premises clear everyone out in the middle of the night; do not pick a fight with the security staff. +### 22. If you are owed work pay, first find out who employed you: an unlicensed workshop or a foreman can still be taken to labour inspection, and only private work for a household or an individual goes to court + +- Cost: No money. First work out three things: whether the other side is running a shop, a factory or a business; whether the work was subcontracted down from a company or a construction site; and whether you are working in somebody's home. Whether they hold a licence you can look up on the National Enterprise Credit Information Publicity System. +- In plain language: An unlicensed workshop or shop that owes wages can still be complained about to labour inspection and taken to labour arbitration, and the owner personally can be pursued. Where a foreman subcontracted work on a construction site and does not pay, the main contractor or the building owner pays. Only private work for a household or an individual, such as working as a nanny, is not a labour dispute, and that has to go to court to be sued. +- Benefit: Article two of the Labour Contract Law limits the employing entity to organisations such as enterprises, sole proprietorships and private non-enterprises; a natural person is not among them. The judicial interpretation on labour disputes from the Supreme People's Court lists several kinds of dispute that are not labour disputes. First, disputes between a household or an individual and a domestic worker. Second, disputes between an individual craftsman and a helper or an apprentice. Third, disputes between a rural contracted household and the person it employs. These go to court as ordinary civil disputes, with the statute of limitations at 3 years. An employing entity with no business licence is not covered by that exception. Article ninety-three of the Labour Contract Law provides that where an employing entity does not have the qualification to operate lawfully and the worker has already performed labour, that entity or its investor shall pay labour remuneration, economic compensation and damages. Article twenty-nine of that interpretation provides that where a dispute arises between a worker and an employing entity that has not registered a business licence, whose business licence has been revoked, or whose term of business has expired while it still carries on, the employing entity or its investor shall be named as a party. Article thirty-three of the Regulations on Labour Security Supervision provides that where there is no business licence, or the licence has been lawfully revoked, and there is labour and employment activity, the labour and security administration department shall carry out labour security supervision. Where an individual contractor unlawfully recruits workers and causes loss to them, the organisation that awarded the contract and the individual contractor bear joint and several liability for compensation (Article ninety-four of the Labour Contract Law). Where the building owner or the general construction contractor awards or subcontracts the works to an individual or to a unit without the qualification to operate lawfully, and this leads to unpaid wages for migrant workers, the building owner or the general construction contractor shall pay (Article thirty-six of the Regulations on the Payment of Wages to Migrant Workers) (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2012, amended). Labour Contract Law（劳动合同法）(Articles two, ninety-three and ninety-four). (reprinted by the State Taxation Administration's database of policies and regulations) ; Supreme People's Court (2020). Interpretation (One) on the Application of the Law in Handling Labour Dispute Cases（关于审理劳动争议案件适用法律问题的解释（一））(法释〔2020〕26 号, Articles two and twenty-nine). ; State Council (国务院) (2004). Regulations on Labour Security Supervision（劳动保障监察条例）(State Council Order No. 423, Article thirty-three). ; State Council (国务院) (2019). Regulations on the Payment of Wages to Migrant Workers（保障农民工工资支付条例）(State Council Order No. 724, Article thirty-six). ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一百八十八条）. ; State Council (国务院) (2006). Measures for the Payment of Litigation Costs（诉讼费用交纳办法）(State Council Order No. 481, Articles six and twenty-nine). ; Standing Committee of the National People's Congress (全国人大常委会) (2023, amended). Civil Procedure Law（民事诉讼法）(Articles one hundred and fifty-two and two hundred and twenty-five). (reprinted by the Guizhou provincial market regulation bureau) +- Notes: Work through these in this order. First check whether the work came down from a company or a construction site; if it did, go to the company that awarded it or to the main contractor. The dedicated wage account on a construction site and the main contractor paying first are in item 2 (complain to labour inspection first if your wages are owed). Then check whether the other side is running a business: a shop, a factory, a workshop; whether they hold a licence or not, it goes to inspection and arbitration, and at the arbitration name the owner as a party as well. Only private work for a household or an individual goes to court. In court you bring the case and produce the evidence yourself, and at first instance the ordinary procedure takes 6 months from the start. Legal fees are not part of the litigation costs a court orders the loser to pay, so a lawyer has to be paid out of your own pocket; if you cannot afford it you can apply for legal aid, see item 3 (legal aid). If you have an IOU you can apply first for a payment order, and a small amount goes through small-claims procedure, see item 22 in section 8 (deceived in an online purchase or a second-hand transaction, small claims). The statute of limitations is 3 years, see item 19 in section 8 (enforcing your rights has a time limit). No route guarantees the money arrives; if the other side has no assets, you can win and still get nothing. diff --git "a/en/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" "b/en/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" index f85d7bc9..9e8f861a 100644 --- "a/en/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" +++ "b/en/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" @@ -12,7 +12,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Regulations on Road Traffic Safety, 道路交通安全法（2021 年修订，第七十、一百零一条）. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau) ; State Council (国务院) (2017). Regulations for the Implementation of the Regulations on Road Traffic Safety, 道路交通安全法实施条例（国务院令第 405 号，2017 年修订，第八十六、九十二条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第一百三十三条）. ; China Insurance Association (中国保险行业协会) (2020). Model Clauses for Commercial Motor Insurance, 机动车商业保险示范条款（2020 版）（第九、二十二条）. - Notes: If anyone is hurt, help them first and then call the police. If you have to move the scene in order to help someone, mark where the vehicle and the person stood to begin with. 「I brushed it and did not feel a thing, so I drove off」 can still be counted as fleeing afterwards, so call the police if you are unsure. The insurance point quotes the model clauses, which are a standard template for the industry. What your own claim actually pays out depends on the policy in your hands. - ### 2. If you find out you have been scammed, call 110 or 96110 at once and ask for the payment to be stopped. Do not investigate it yourself first - Cost: No money. Make one call and give the time of the transfer, the amount, and the account it went to. Keep the transfer screenshot and the chat records. @@ -21,7 +20,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2022). Law of the People's Republic of China on the Prevention and Control of Telecom and Online Fraud, 反电信网络诈骗法（第二十、三十四条）. ; Fujian Provincial Public Security Department (2023). 96110 来电，请务必接听. (a provincial public security page) - Notes: Anyone who claims to be from the courts, the procuratorate or the police and tells you to move money into a 「安全账户」 is running a scam. The law only says 「建立止付冻结制度」, which means letting the bank hold the money down. How much comes back to you depends on whether the money was still on the account when you reported it. The description of the 「国家反诈中心」 app and of 96110 is taken from the Fujian Provincial Public Security Department page, because the Ministry of Public Security's own page would not open. - ### 3. Learn the four hard anti-fraud rules: do not trust a caller, do not reveal information, do not open links, check again before any transfer. The seven commonest scams all have this shape - Cost: No money. Learn the 「三不一多」 rule, read through the seven categories below, and install the 「国家反诈中心」 app. Always answer 96110. If anyone asks you to transfer money, share your screen, install an unfamiliar app, open a link, or read out a verification code, hang up first and then call the official number to check. The hard part is that the other person usually pushes hard, and you have to hold out against that and hang up. @@ -30,7 +28,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: B - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2022). Law of the People's Republic of China on the Prevention and Control of Telecom and Online Fraud, 反电信网络诈骗法（第八、二十、三十四条）. ; Ministry of Public Security (公安部) (2026). Ministry of Public Security press conference: briefing on the current situation and characteristics of telecom and online fraud crimes, and the key points for preventing them (2026-6-12):「以上这10类案件就占了全部电诈案件的85%。其中，刷单返利诈骗发案最高，占所有电诈案件的25%；虚假网络投资理财诈骗损失最高，约占全部电诈案件损失的40%」. ; Fujian Provincial Public Security Department (2023). 96110 来电，请务必接听. (a provincial public security page, containing the slogan 「来电不轻信，信息不透露，链接不点击，转账多核实」 and the warning 「警惕 192 开头的诈骗电话」) ; Fujian Provincial Public Security Department (2023 to 2024). Spot-the-scam columns: 刷单 ; 投资理财 ; 贷款 ; 冒充客服 ; 冒充公检法 ; 杀猪盘 ; 色诱刷单 - Notes: Anyone claiming to be the courts, the procuratorate, the police, customer service, your leader or a platform's risk-control team, and asking you to 「配合调查」「取消会员」「刷单返利」「投资内幕」, is running the same routine. A real loan asks for no deposit, and the courts, the procuratorate and the police have no 「安全账户」. Remembering those two sentences blocks off most of it. The grade is B because this slogan is only the official publicity line, with no figure for how much less often people get scammed after learning it. The slogan and the breakdown of the seven categories are taken from a provincial public security department page. The 10 most frequent types of fraud reported by the Ministry of Public Security in 2026 also include impersonating an e-commerce or logistics customer service desk, loan and credit reporting, fake online game transactions, online dating and networking, impersonating a leader or an acquaintance, and airline ticket refunds and changes. The routine is the same in every case, and it ends with you being asked to transfer money. - ### 4. A face on a video and a voice on the phone are not verification. Hang up before any transfer and call back on an old number from your own contacts - Cost: No money. One extra phone call, a few minutes of waiting. Post fewer full-face videos and long voice messages on public accounts. The hard part is that the other person often pushes hard, and you have to hold out against transferring on the spot. @@ -48,7 +45,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). Criminal Procedure Law, 刑事诉讼法（2018 年修正，第三十四、三十五、三十九、五十二、一百一十九、一百二十条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百四十三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2012). State Compensation Law, 国家赔偿法（2012 年修正，第十七、三十三条）. (reprinted by the National Bureau of Statistics) - Notes: Do not go and find the other side privately, and do not go to their family. Any private settlement can be read as you admitting the facts. Your chat records, call records and travel records are the material that proves you innocent, and deleting them deletes your own evidence. If you got the facts wrong, or if the report you made did not hold up later, that is not the crime of framing. This item makes no distinction between men and women. For the three items it opens out into, see items 33, 34 and 35 in this section (the party that fabricated the facts, insufficiency of evidence means an acquittal with appeal and retrial, and state compensation after custody calculated by the day). - ### 6. Go and explain yourself as soon as something happens: a voluntary surrender cuts the sentence by up to 40%, and letting the case go out of time is a dead end - Cost: No money. Go yourself to the public security organ to report, or stay where you are and call the police, and then describe what happened truthfully. The hard part is holding out against the thought that you should wait a while longer and nobody will ever look into it. @@ -57,7 +53,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: National People's Congress (全国人大). Consolidated text of the Criminal Law of the People's Republic of China, as amended by Criminal Law Amendment Eleven, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第六十七、八十七、八十八条. ; Supreme People's Court and Supreme People's Procuratorate (2021). Guiding Opinions on Sentencing for Common Crimes (Trial), 关于常见犯罪的量刑指导意见（试行）（法发〔2021〕21 号，2021 年 7 月 1 日施行）三（六）（七）（十四）, full text reprinted by the Zhangye City People's Procuratorate of Gansu Province. ; Supreme People's Court (2010). Opinions on Certain Specific Questions Concerning Voluntary Surrender and Meritorious Service, 关于处理自首和立功若干具体问题的意见（法发〔2010〕60 号）一. Supreme People's Court Gazette. - Notes: The person who benefits from this item is you. A voluntary surrender is only a mitigating circumstance, the crime is still found, and the court decides how much comes off; the sentencing guidelines require leniency to be applied strictly for violent crimes and drug crimes. Traffic accidents have their own calculation. Staying at the scene, helping the injured and reporting to the public security organ counts as reporting yourself of your own accord, but because these are statutory duties in any case, leniency is applied strictly. Reporting yourself after driving away still counts as a surrender, but the heavier of the two statutory bands is used as the baseline, and only then is it decided whether to reduce it and by how much. How to do that: see item 1 in this section (after a traffic accident, stop, help the injured and call the police). Getting a lawyer first and telling the truth do not conflict: see item 5 in this section (get a lawyer first if you are accused). What the lawyer checks for you is whether the written record is right, not whether you can come up with a better story. - ### 7. After drinking, do not touch the steering wheel, not on an electric scooter and not even to move the car a short way - Cost: A little money. A hired driver or a taxi costs several tens of yuan. Agree before the gathering who is not drinking that day. The hard part is holding your ground when people at the table press you to drink. @@ -66,7 +61,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Regulations on Road Traffic Safety, 道路交通安全法（2021 年修订，第九十一条）. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau) ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（第八条，即刑法第一百三十三条之一）. ; China Insurance Association (中国保险行业协会) (2020). Model Clauses for Commercial Motor Insurance, 机动车商业保险示范条款（2020 版）（第九、二十二条）. - Notes: Drunk driving is a crime and leaves a criminal record, which affects your own political review and some professional qualifications. The line between 「饮酒」 and 「醉酒」 is drawn by the alcohol content in the blood, and the exact thresholds are set by national standards. This section has not verified those numbers, so do not estimate by feel which side of the line you are on. - ### 8. Lend your bank card, phone card or payment account to nobody. Passing transactions through your own account is not a side job - Cost: No money. If someone says 「lend me the card so I can move some money through it, I will give you a few hundred yuan」, say no. @@ -75,7 +69,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2022). Law of the People's Republic of China on the Prevention and Control of Telecom and Online Fraud, 反电信网络诈骗法（第三十一、四十四条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（第二十九条，即刑法第二百八十七条之二）. - Notes: Saying 「I did not know what he was going to do with the card」 will not hold up. Investigators reason back from the outside circumstances to whether you knew: having taken a fee, and having flows that are plainly abnormal, can each count as knowledge. If the card was used to receive the proceeds of a crime, then even if you are not finally criminally liable, the account is frozen and you are sanctioned. - ### 9. Check your own credit report twice a year for free, and look for loans and cards you did not take out - Cost: No money. You can do it online in about ten minutes. @@ -84,7 +77,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: State Council (国务院) (2013). Regulations on the Administration of Credit Reporting, 征信业管理条例（国令第 631 号，第十六条、第十七条、第二十五条）. - Notes: If your identity card has been used by somebody else to take out an online loan or a card, you often only find out when a mortgage application is refused, and by then several years have gone by. The way in is the website of the People's Bank of China Credit Reference Center, or your mobile banking app. Do not use a third-party 「查征信」 app, because those apps are collecting your information in the first place. - ### 10. If a fight breaks out, call the police and do not swing first. Whoever swings first almost always loses out - Cost: No money. Back off to somewhere with people around you and call the police, so that the other person's movements stay on the cameras. The hard part is holding back from answering back and from pushing. @@ -102,15 +94,14 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二十条）. ; Supreme People's Court, Supreme People's Procuratorate and Ministry of Public Security (2020). Guiding Opinions on the Proper Application of the Self-Defence System, 关于依法适用正当防卫制度的指导意见（法发〔2020〕31 号, items 5, 6, 7, 10, 11, 13, 14 and 19）. ; Supreme People's Procuratorate (2018). Twelfth Batch of Guiding Cases (Procuratorate Case No. 45, the Chen justifiable-defence case, and Procuratorate Case No. 47, the Yu Haiming justifiable-defence case), 第十二批指导性案例（检例第 45 号陈某正当防卫案、检例第 47 号于海明正当防卫案）. ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Criminal Procedure Law, 刑事诉讼法（2018 年修正，第九十一条）, full text reprinted by the People's Procuratorate of the Ningxia Hui Autonomous Region. - Notes: The default move is still the one in item 10 in this section, backing off and calling the police (if a fight breaks out, call the police and do not swing first). Two people hitting each other is not self-defence, and neither is chasing after the other person once they are on the ground or have run off. Only consider hitting back at the moment when there is nowhere left to retreat. There are three things to do afterwards. First, after calling the police, do not leave the scene. Second, protect the cameras and the witnesses as fast as you can; the guiding opinions require the police to fetch the video and audio recordings from the scene at the earliest opportunity and to question witnesses at the earliest opportunity. Third, before you are formally questioned by the police, get a lawyer: see item 5 in this section (get a lawyer first if you are accused). If the case is later withdrawn, if no prosecution is brought, or if you are acquitted, the days you were held can be claimed as state compensation day by day: see item 35 in this section (state compensation after custody). Defending somebody else is equally allowed by law. But walking into a fight between strangers empty-handed is a separate risk: see item 37 in section 13 (coming across a group fight). The people who benefit from this item are mainly you and your family. - -### 12. Whoever you have fallen out with — unpaid wages, dismissal, money that went missing — go through a complaint, arbitration or a lawsuit. Do not go and settle it yourself +### 12. If somebody has wronged you — unpaid wages, dismissal, money taken from you — go through a complaint, arbitration or a lawsuit, and do not go and settle it yourself -- Cost: No money. Neither a complaint nor arbitration costs anything. You have to make one or two trips and wait a few months. The hard part is holding back for those months. -- In plain language: Take the lawful route and the worst outcome is that you do not get the money back. Take revenge and the money still does not come back, and you have to answer for it with your own life. The Wuxi case: Xu Jinjin did not get his certificate and was unhappy with his internship pay, so he attacked with a knife, killing 8 and injuring 17. He was sentenced to death and the sentence has been carried out. For unpaid wages, the complaint and arbitration route recovered 8,650,000,000 yuan for 654,000 people nationwide in 2024. -- Benefit: The lawful route has official figures. In 2024 labour inspection nationwide investigated 137,000 cases and recovered 8,650,000,000 yuan in wages for 654,000 people. A further 998 people were sentenced for the crime of refusing to pay labour remuneration. How to complain to labour inspection about unpaid wages and how to apply for labour arbitration: see item 2 in section 7. The price list for the route of revenge is written into Article two hundred and thirty-two of the Criminal Law:「故意杀人的，处死刑、无期徒刑或者十年以上有期徒刑；情节较轻的，处三年以上十年以下有期徒刑」. The official case: Xu Jinjin failed his examination, did not get his certificate, and was also dissatisfied with his internship pay. On 2024-11-16 he returned to Wuxi Institute of Arts and Technology with a knife and attacked, killing 8 people and injuring 17. At first instance he was sentenced to death for intentional homicide and deprived of political rights for life, and the sentence was approved for execution by the Supreme People's Court on 2025-1-20. What he resented was the college and that payment, but the victims were his classmates, and he never received that payment either (nationwide) +- Cost: No money; neither the complaint nor the arbitration costs anything. You have to make a trip or two and then wait several months. The hard part is holding back through those months without doing anything. +- In plain language: Go by the legal route and the worst outcome is that the money does not come back. Take revenge and the money still does not come back, and on top of that you pay with your own life. The case in Wuxi: a student whose surname is Xu failed to get his diploma and was unhappy with his internship allowance, went at people with a knife, killing 8 and injuring 17; he was sentenced to death and the sentence has been carried out. Unpaid wages go to a complaint and arbitration, and in 2024 that route recovered 8.65 billion yuan for 654,000 people nationwide. +- Benefit: The legal route has official figures. In 2024 labour inspection nationwide handled 137,000 cases and recovered 8.65 billion yuan in wages for 654,000 people. Another 998 people were sentenced for the crime of refusing to pay labour remuneration. How to complain to labour inspection about unpaid wages, and how to apply for labour arbitration, see item 2 in section 7. The price list on the route of revenge is written into Article two hundred and thirty-two of the Criminal Law:「故意杀人的，处死刑、无期徒刑或者十年以上有期徒刑；情节较轻的，处三年以上十年以下有期徒刑」. The official case: the student whose surname is Xu failed his examinations, did not get his certificate of graduation, and was also unhappy with his internship allowance. On 2024-11-16 he returned to Wuxi Institute of Arts and Technology with a knife and attacked, killing 8 people and injuring 17. At first instance he was sentenced to death for intentional homicide and deprived of political rights for life, and on 2025-1-20 the Supreme People's Court approved the sentence for execution. What he was unhappy about was the school and that allowance; the people hurt were his fellow students, and he did not get a single yuan of that allowance either (nationwide) - Evidence grade: A -- Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百三十二条）. ; Xinhua News Agency (新华社) (2024). 江苏宜兴一校园发生持刀伤人事件致 8 死 17 伤. (reproducing the Yixing police statement, with the casualty figures and the motive quoted in the original words) ; Supreme People's Procuratorate (2025). 无锡校园持刀伤人案罪犯徐加金被执行死刑. ; Ministry of Human Resources and Social Security (人力资源社会保障部) (2025). 2024 年度人力资源和社会保障事业发展统计公报. ; Supreme People's Court (2025). 最高人民法院工作报告（「对驾车冲撞行人的樊维秋、校园持刀行凶的徐加金等判处死刑」）. -- Notes: An intern who is still studying does not normally count as having formed a labour relationship with the unit. If the internship pay cannot be obtained you can only ask the college to intervene, or sue in court; the labour arbitration route is not available. For how to complain and how to arbitrate over unpaid wages, see item 2 in section 7. For the line around debt collection, see item 15 in section 9 (do not detain anyone and do not lock anyone up to collect a debt). The lawful route may still fail to get the money back, because the end of a case does not mean the money has arrived. Take revenge and the money still does not come back, and you have to answer for it with your own life. +- Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百三十二条）. ; Xinhua News Agency (新华社) (2024). 江苏宜兴一校园发生持刀伤人事件致 8 死 17 伤. (reproducing the Yixing police statement, with the casualty figures and the words on the motive) ; Supreme People's Procuratorate (2025). 无锡校园持刀伤人案罪犯徐加金被执行死刑. ; Ministry of Human Resources and Social Security (2025). Statistical Bulletin on the Development of Human Resources and Social Security in 2024（2024 年度人力资源和社会保障事业发展统计公报）. ; Supreme People's Court (2025). Work Report of the Supreme People's Court（最高人民法院工作报告）（「对驾车冲撞行人的樊维秋、校园持刀行凶的徐加金等判处死刑」）. +- Notes: An intern who is still studying generally does not count as having established a labour relationship with the unit. If the internship allowance cannot be obtained, you can only ask the school to coordinate, or sue in court; the labour arbitration route is not available. If you did private work for a household or an individual and were not paid, that also has to go to court, see item 22 in section 7 (first find out who employed you). For how exactly to complain and how to go to arbitration when your wages are owed, see item 2 in section 7. For the red lines when collecting a debt, see item 15 in section 9 (do not hold anybody hostage or lock anybody up when collecting a debt). The legal route may fail to get the money back either, because a case being closed does not mean the money has arrived. Take revenge and the money still does not come back, and on top of that you pay with your own life. ### 13. Never take it out on people who have nothing to do with it: driving into a crowd and attacking in a public place are convicted as endangering public security by dangerous means, which starts at three years, and if somebody dies it is the death penalty @@ -120,15 +111,15 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第一百一十四、一百一十五条）. ; Xinhua News Agency (新华社) (2024). 广东珠海驾车撞人事件造成 35 人死亡 43 人受伤. (reproducing the Zhuhai municipal public security bureau statement) ; Supreme People's Procuratorate (2024). 樊维秋以危险方法危害公共安全案一审宣判，判处死刑，剥夺政治权利终身. ; Supreme People's Procuratorate (2025). 珠海驾车撞人案罪犯樊维秋被执行死刑. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第二十九、五十条）. - Notes: 「I am not going to live anyway」 is not a discount at sentencing. The Supreme People's Court's work report for 2025 states the line on cases like this:「对挑战法律和道德底线的罪大恶极者，坚决依法从严从重从快惩处」. Merely saying it is already illegal. Saying you will set fire to the place, blow it up or poison it, and disturbing public order, carries 5 to 10 days of detention and a fine of up to 1000 yuan. Writing a threatening letter, or threatening somebody's personal safety in some other way, is illegal in the same way, and carries up to 5 days of detention or a fine of up to 1000 yuan (the Law on Penalties for Administration of Public Security, in force from 2026-1-1). If the thought has got as far as that, see item 14 in this section (treat a thought of 「taking somebody down with me」 as an emergency). - -### 14. Treat a thought of 「taking somebody down with me」「going down together」 as an emergency: leave the scene, hand your car keys and any knife to somebody else, and call 12356 +### 14. If a thought of 「taking somebody down with me」 or of going down together comes to you, treat it as an emergency: leave the scene, hand the car keys and the knife to somebody else, and call 12356 -- Cost: No money. The hard part is admitting that you are not right at this moment, and the hard part is handing your things to somebody else. -- In plain language: The window between this kind of thought and the action is short. Among 82 people who had attempted suicide, 47.6% said that no more than 10 minutes passed between the first thought and the act. So what you have to do is not 「seeing it through」 but putting the steering wheel, the knife, the gas valve and the rope one step out of reach and getting those ten minutes past. Then call 12356. -- Benefit: The window between this kind of thought and the action is short. In interviews with 82 people who had attempted suicide, carried out within three days of admission, 47.6% said that no more than 10 minutes passed between the first thought and carrying it out. A 10-year systematic review, which pulls together the research of a decade, finds that the evidence for restricting lethal means keeps strengthening. After painkillers were brought under control, suicides involving them fell 43%. After barriers were added to the places people jump from, suicides fell 86% (79% to 91%). Putting the steering wheel, the knife, the gas valve and the rope one step out of reach is a way of waiting out those ten minutes. 12356 is the national psychological assistance hotline, answered for at least 18 hours a day (nationwide, in force from 2025-5-1) +- Cost: No money. The hard part is admitting that something is wrong with you at this moment, and the hard part is handing the things over to somebody else. +- In plain language: The window between a thought of this kind arriving and your doing something about it is very short. Of 82 people who had attempted suicide, 47.6% said that no more than 10 minutes passed between the first thought and the attempt. So what is needed is not to "think it through", but first to put one step between the steering wheel, the knife, the gas valve, the rope and yourself, and get through those ten minutes. Then call 12356. +- Benefit: The window between a thought of this kind and the attempt is very short. In interviews with 82 people who had attempted suicide, carried out within three days of admission, 47.6% said that no more than 10 minutes passed between the first thought and carrying it out. A 10-year systematic review (which pulls together and sorts through the research of ten years) concludes that the evidence on restricting lethal means keeps getting stronger. After controls on painkillers, the suicides connected with them fell by 43%. After fall points were fitted with protection, suicides fell by 86% (79% to 91%). Putting one step between the steering wheel, the knife, the gas valve, the rope and yourself is a way of waiting out those ten minutes. 12356 is the nationwide psychological assistance hotline, and it is answered for no less than 18 hours a day (nationwide, from 2025-5-1) - Evidence grade: B -- Sources: Deisenhammer EA et al. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry. ; Zalsman G et al. (2016). Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry. ; National Health Commission (国家卫生健康委) (2024). 关于应用「12356」全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. -- Notes: The grade is B because both pieces of evidence come from the literature on suicide, and there is no direct research applying them to an impulse to hurt other people. The Deisenhammer study also has only 82 people in it, done in a single hospital. When you are not in a fit state, do not drive and do not take a shift where you are responsible for somebody else's safety; hand the shift over first. If you want to hurt yourself, see item 25 in section 1 (call 12356 when you have suicidal thoughts). For what to do first when you are feeling low, see item 19 in section 3. +- Sources: Deisenhammer EA et al. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry. ; Zalsman G et al. (2016). Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry. ; National Health Commission (2024). Notice on the Application of 「12356」 as the Nationwide Psychological Assistance Hotline Number（关于应用「12356」全国统一心理援助热线电话号码的通知）(国卫医政函〔2024〕259 号). +- Notes: The grade is B because both pieces of evidence come from the literature on suicide, and there is no direct research applying them to "the urge to hurt somebody else". The Deisenhammer study also has only 82 people in it, all at a single hospital. When you are not in a fit state, do not drive, and do not go onto a shift where you are responsible for somebody else's safety either; hand the shift over first. For hurting yourself, see item 25 in section 1 (call 12356 when there are thoughts of suicide). For what to do first when your mood is low, see item 18 in section 3. + ### 15. When someone close to you says 「nobody is going to have a good time」「I am taking the children with me」, do not treat it as an angry remark: an immediate family member can take them for treatment, and the police must act on a report @@ -138,7 +129,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第二十九、五十条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2012, amended 2018). Mental Health Law, 精神卫生法（第二十八、二十九、三十、三十五条）. ; Bureau d'Enquêtes sur l'Aviation (BEA), France (2016). Final Report BEA2015-0125.en: Accident on 24 March 2015 at Prads-Haute-Bléone to the Airbus A320-211 registered D-AIPX operated by Germanwings. - Notes: Judged by who benefits, this item counts a spouse, parents or children living with you. You are usually one of the people covered by 「nobody is going to have a good time」, and the law gives the right to have somebody admitted precisely to immediate family members, so an outsider cannot act and can only report it to the police. Admission for treatment is only for a real signal of danger. Angry remarks during an argument do not count, and it cannot be used to lock a disobedient relative up in hospital. The same law also says that where a further examination or an assessment finds the person does not have a severe mental disorder, the hospital may not keep them in hospital (Article thirty-five). If you cannot tell, call 12356 and ask first, or report it to the police so an officer comes. The investigation into Germanwings flight 9525 concluded that neither the authorities nor the employer could have done anything to stop the first officer from flying that day, because 「无论是他本人，还是任何其他人——医生、同事或者家属——都没有告知他们」. All 150 people on board died: 2 pilots, 4 cabin crew and 144 passengers. - ### 16. Do not abuse people online, do not spread rumours, and do not repost what you have not checked. If you are piled on, keep the evidence first and then call the police - Cost: No money. Think for ten seconds before you post. When you are attacked, take screenshots and screen recordings, and save the links and the account IDs, complain to the platform first, and then call the police. The hard part is holding back from abusing back while you are being abused. @@ -147,7 +137,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第二十九、五十条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百四十六条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（第十六条，增加刑法第二百四十六条第三款）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（第二百四十六条新增第三款）, full text reprinted by the National Petition Office. - Notes: The two crimes of insult and defamation are in principle dealt with only on complaint, which means you have to go to court yourself. The police generally do not open a case of their own accord, and the evidence has to be gathered by you. There are two exceptions. The first is 「严重危害社会秩序和国家利益的除外」, cases which the public security organ may handle directly (the second paragraph of Article two hundred and forty-six of the Criminal Law). The second is defamation committed online where you genuinely have difficulty collecting the evidence yourself, in which case the court may require the public security organ to help with the investigation (the third paragraph of the same article, added by Criminal Law Amendment Nine). Reposting still makes you liable, and passing on something you know to be false also counts as fabricating facts. For what to do after you are piled on online, see item 37 in this section (keep the evidence first, then choose your route). The article number for insult and defamation moved from Article forty-two of the old law to Article fifty of the new one. - ### 17. Read the whole document before you sign, do not sign for other people, and do not sign a blank sheet - Cost: No money. Spend a few more minutes reading a contract, an IOU, a guarantee or an employment document from beginning to end. Ask on the spot about anything you do not understand, and take a photograph to keep. The hard part is taking the time to read it through in front of the other side. @@ -156,7 +145,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第四百六十五、五百零九条）. - Notes: 「Do not sign for other people, do not sign a blank sheet, and state the purpose on a photocopy of your identity card」 is drawn from experience, and there is no separate official document setting it out. The purpose behind all three is the same: do not let your signature be fitted into a contract that is legally valid and that you are then forced to honour. An electronic signature and a face-recognition confirmation count as signing in the same way. - ### 18. Write a clear IOU when you lend money, and before you guarantee somebody else's loan decide whether you are willing to repay it for them - Cost: No money. Use a sheet of paper to write out the lender, the borrower, the amount, the interest rate, the term and the method of repayment, and have both sides sign. Send the money by transfer so there is a trace, and do not hand over cash. @@ -165,7 +153,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第六百六十八、六百八十一、六百八十六、六百八十七条）. - Notes: On money you have lent out, a court does not protect the part of the interest above four times the LPR: see section 7. Writing your identity card number on the IOU and noting 「借款」 is drawn from experience. Whether the guarantee says 「连带」 decides whether you queue up behind the borrower or stand beside him when the money is demanded, so read it before you sign. - ### 19. Enforcing your rights has a time limit: 3 years for the statute of limitations in a civil case and 1 year for labour arbitration, and the other side only has to say 「too late」 - Cost: No money. Do not chase a debt only in your head. Use WeChat, text message, email or a registered letter so that there is a trace. @@ -183,7 +170,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2002). Interpretation of Article three hundred and thirteen of the Criminal Law of the People's Republic of China, 关于《中华人民共和国刑法》第三百一十三条的解释. ; Supreme People's Court and Supreme People's Procuratorate (2024). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Refusing to Execute Judgments and Rulings, 关于办理拒不执行判决、裁定刑事案件适用法律若干问题的解释（法释〔2024〕13 号，第三、四、五、六、八、十一、十二条）, published on the Supreme People's Court page, with the full text checked against the reprint by the Guangzhou Prison of Guangdong Province. ; ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（第三十九条，修改刑法第三百一十三条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第五百三十八、五百三十九、五百四十条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2023, amended). Civil Procedure Law, 民事诉讼法（第一百一十四、一百一十八、二百五十二条）, reprinted by the Guizhou Provincial Administration for Market Regulation: - Notes: The version circulating online of 「put the flat in the company's name as staff accommodation and transfer all the money to a relative」 is not a loophole. It is exactly the gift given away free and the sham transfer that the judicial interpretation names. Relatives and companies that help you carry it out count as accomplices, and property that can be recovered is still recovered. In the other direction, genuinely having no money is not this crime. Deciding whether you are someone who 「has the money but will not pay」 means deducting first the cost of your own living and that of the family you support. If you repay part or all of it before the procuratorate prosecutes, and the circumstances are minor, no prosecution may be brought. If you repay before the first instance judgment, a lighter punishment or even none may be imposed. For what to do after being put on the list or placed under a consumption restriction, see item 21 in this section (a consumption restriction, or being put on the judgment defaulter list). For how to recover once you have performed, see item 19 in section 7 (how to start again after being on the judgment defaulter list). - ### 21. If you are put under a consumption restriction or on the judgment defaulter list, first find out which provision put you there, and apply for correction where it can be corrected - Cost: No money. Go to the enforcement court and find out which provision you were included under. Have ready the receipts for money you have repaid and for judgment you have complied with, or have ready security documents. Consumption that living or running the business requires must be applied for one item at a time. The hard part is making the trips and lodging the applications yourself. @@ -192,7 +178,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Supreme People's Court (2017, amended). Several Provisions on the Publication of Information on the List of Judgment Defaulters, 关于公布失信被执行人名单信息的若干规定（第一、二、三、四、十条）. ; Supreme People's Court (2015, amended). Several Provisions on Restricting High Consumption and Related Consumption by Persons Subject to Enforcement, 关于限制被执行人高消费及有关消费的若干规定（法释〔2015〕17 号，第三、八、九条）：「（九）乘坐G字头动车组列车全部座位、其他动车组列车一等以上座位等其他非生活和工作必需的消费行为」. - Notes: A consumption restriction means high consumption is restricted. The provisions came out in 2010 and were amended in 2015 by 法释〔2015〕17 号, when the title was changed to 「Several Provisions on Restricting High Consumption and Related Consumption by Persons Subject to Enforcement」, and this item follows the amended text now in force. Three parts of the 2015 change matter. First, a ninth category was added, so every seat on a high-speed train counts. Second, the limitation 「paid for out of their property」 was removed: the old wording was 「不得有以下以其财产支付费用的行为」. Third, where a unit is placed under a consumption restriction, the person who actually controls it is restricted too, but private consumption and spending your own money can be applied for to the enforcement court. The judgment defaulter list and a consumption restriction are two different things, and you may only run into one of them. Neither of them means a blemish on your credit record, and being removed from the list does not change your credit record: see item 19 in section 7 (how to start again after being on the judgment defaulter list). Whether a spouse or a child spending their own earnings is restricted is not set out in the text now in force, so if you run into it, ask the enforcement court. - ### 22. If you are scammed in an online purchase or a second-hand sale, complain to the platform first, then call the police, and then work out whether it is worth suing - Cost: A little money. Neither a complaint to the platform nor a report to the police costs anything. When you sue, the acceptance fee is 50 yuan where the amount in dispute is up to 10,000 yuan, and it is halved again if the case goes through the summary procedure, is settled, or is withdrawn. Opening the case, the hearing and enforcement each take at least half a day to a day, and opening the case online saves the running about. You have to keep the order, the chat records, the transfer records and the delivery note. The hard part is not giving up halfway through all the running about. @@ -201,7 +186,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Supreme People's Court and Supreme People's Procuratorate (2011). Interpretation on Several Specific Questions Concerning the Application of the Law in Handling Criminal Cases of Fraud, 关于办理诈骗刑事案件具体应用法律若干问题的解释（法释〔2011〕7 号，第一条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百六十六条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第五十八条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2023, amended). Civil Procedure Law, 民事诉讼法（第十六、一百六十四、一百六十五、一百六十八条）. (reprinted by the Shanghai Municipal Development and Reform Commission) ; Supreme People's Court (2022, amended). Interpretation on the Application of the Civil Procedure Law, 关于适用《民事诉讼法》的解释（第二十条）. ; State Council (国务院) (2006). Measures for the Payment of Litigation Costs, 诉讼费用交纳办法（国务院令第 481 号，第十三、十五、十六条）. - Notes: Whether it is worth fighting for can be worked out like this: multiply your hourly wage after tax by the time you have to put in, then add the court's acceptance fee. Count opening the case, the hearing and, if needed, an application for enforcement at half a day to a day each. Compare the result with the amount you want back, then discount it and estimate whether a win would really be enforced. Whether to go ahead is your own decision on the basis of that arithmetic. After the 2025 revision of the Law on Penalties for Administration of Public Security every article number changed: theft and fraud moved from Article forty-nine of the old law to Article fifty-eight. After the 2023 revision of the Civil Procedure Law, small claims is in Article one hundred and sixty-five. Do not copy material citing the old numbers at either of these two places. 「Complain to the platform first」 comes from experience: a platform holds a deposit and has rules for penalties, so it deals with it fastest. The document number of 法释〔2011〕7 号 is not shown on the procuratorate's own web page and still has to be checked against the original page of the Supreme People's Court. - ### 23. Large gifts during dating and marriage: think before you hand them over, because in principle you cannot get them back afterwards - Cost: No money. Before a large transfer, ask yourself one question: is this a loan or a gift. If it is a loan, write an IOU and put 「借款」 in the transfer reference. The hard part is saying that out loud. @@ -219,15 +203,14 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零四十二条）. ; Supreme People's Court (2024). Provisions on Several Questions Concerning the Application of the Law in Handling Disputes over Bride Price, 关于审理涉彩礼纠纷案件适用法律若干问题的规定（法释〔2024〕1 号，第二、五、六条）. ; Supreme People's Court (2020). Interpretation (One) on the Application of the Marriage and Family Book of the Civil Code, 关于适用《民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第五条）. - Notes: Whether bride price counts as 「数额过高」 depends on the per-capita disposable income where the paying party lives and on that family's circumstances, and there is no single national figure. How much is to be returned is decided case by case by the judge. The rules are the same for the party who pays and the party who receives: whoever pays the money keeps the evidence. - -### 25. Property owned before the wedding is not at risk, and putting a partner's name on the flat and money from parents are worth writing down first +### 25. Do not be afraid of property owned before the marriage: write down the house being put in the other person's name, and any contribution by parents -- Cost: A little money. Drawing up a written property agreement yourself costs nothing, while notarisation costs extra. When your parents put money in, ask them to put a written statement down. The whole thing takes a few hours in one go. The hard part is talking to your family about money. -- In plain language: Property you owned before the wedding is yours alone, and it does not turn into joint property because the marriage has gone on for years. A couple may agree in writing that property before and after the wedding belongs to each of them separately. One point needs watching: if you have agreed to put the other person's name on the flat, you can still change your mind before the registration is amended, and once the registration is done it is theirs. On divorce, a party who hides money, transfers it, squanders it or fabricates debts can be given a smaller share or none at all. -- Benefit: Property owned before the wedding is personal property and does not turn into joint property because the marriage has lasted a long time. A couple may agree in writing that property before the wedding and income after it belong to each of them separately, belong to both jointly, or are partly joint, and that agreement binds both sides. Where parents put money towards a flat before the wedding, it counts as a gift to their own child unless it is made clear that it is for the couple. Money from parents after the wedding is dealt with by agreement; where there is no agreement, it is dealt with on the principle of joint property. Where, before the wedding or during the marriage, it is agreed that one party's property is gifted to the other or becomes jointly owned, the party making the gift may revoke it before the registration is amended, and after the registration it is the other party's. On divorce, a party who conceals, transfers, sells off or squanders joint property, or fabricates debts, can be given a smaller share or none at all (nationwide) +- Cost: Writing a property agreement yourself costs no money; having it notarised costs extra. When your parents pay, ask them to write down a short statement. The whole thing takes a few hours in one go. The hard part is bringing up money with your family. +- In plain language: Property you owned before the marriage is yours, and it does not become joint property just because you have been married a long time. A couple can agree in writing that property from before and after the marriage belongs to each of them. One thing to watch: if you have agreed to put the other person's name on the house, you can still change your mind before the transfer is registered; once it is registered it is theirs. On divorce, the party who hides money, transfers it away, squanders it or fabricates debts can be given less, or nothing. +- Benefit: Property owned before the marriage is personal property, and it does not become joint property just because you have been married a long time. A couple can agree in writing that property from before the marriage and income from after it belong to each of them, belong to both, or that part is joint; that agreement binds both sides. Where parents contributed to buying a home before the marriage, that counts as a gift to their own child, unless it is stated expressly that it is for the two of them. Where parents contribute after the marriage, it is handled by the agreement; where there is no agreement, it is handled on the principle of joint property. Where, before or during the marriage, one party agrees to gift a property to the other, or to make it jointly owned by the two, the party making the gift can withdraw it before the transfer is registered; after registration it is the other party's. On divorce, the party who hides, transfers, sells or squanders joint property, or fabricates debts, can be given less, or nothing (nationwide) - Evidence grade: A -- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零六十三、一千零六十五、第一千零九十二条）. ; Supreme People's Court (2020). Interpretation (One) on the Application of the Marriage and Family Book of the Civil Code, 关于适用《民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第二十九、三十一、三十二条）. -- Notes: A property agreement is equally valid for the man and the woman. If you want to use it to stand against your spouse's debts to an outsider, the creditor has to know the agreement exists. If they do not know, it does not reach them. An agreement is valid without notarisation; notarisation only saves you arguing about signatures and dates later. +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零六十三、一千零六十五、一千零九十二条）. ; Supreme People's Court (2020). Interpretation (One) on the Application of the Marriage and Family Book of the Civil Code, 关于适用《民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第二十九、三十一、三十二条）. +- Notes: A property agreement is just as valid between a man and a woman. If you want to use it against a debt your spouse owes to somebody else, the creditor has to know the agreement exists. If they do not know, it does not bind them. The agreement is valid without notarisation too; notarising only saves an argument about signatures and dates later. How to write down whether money from parents buying a home is a loan or a gift, see item 11 in section 10 (parents contribute to buying a home). ### 26. Check the other person's marital status before you marry, and money taken under the pretence of marriage is pursued as fraud @@ -237,7 +220,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零五十一条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百五十八、二百六十六条）. - Notes: 「Marriage fraud」 is not a charge in itself. Whether it can be dealt with as fraud depends on whether the other person meant from the start to take the money for themselves, told a set of lies to get it, and the case has to be opened by the public security organ for investigation. Whether an amount is 「数额较大」 is set by the judicial interpretation, which local areas may adjust, and this section has not verified it. The victims of marriage fraud are not divided by gender: there are cases of the bride price being taken and cases of the dowry being taken. - ### 27. For a home visit to repair an appliance, inspect a car or do decoration, ask the price before the work starts and stay beside the work the whole time - Cost: No money. Before the work starts, find out the charging standard and which parts are going to be replaced, and photograph the quote or ask for it in writing. After the work, get a service receipt and an invoice, and keep the old parts. You also have to stand beside the work for an hour or two. The hard part is watching the whole way through without walking away. @@ -246,7 +228,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: B - Sources: Ministry of Commerce (商务部) (2012). Measures for the Administration of the Appliance Repair Service Industry, 家电维修服务业管理办法（商务部令 2012 年第 7 号，第五、七、九、十四条）. ; State Administration for Market Regulation (市场监管总局) (2022). Provisions on Clear Price Marking and the Prohibition of Price Fraud, 明码标价和禁止价格欺诈规定（市场监管总局令第 56 号，第五、八、十九条）. ; State Council (国务院) (2024). Implementing Regulations for the Consumer Rights Protection Law of the People's Republic of China, 中华人民共和国消费者权益保护法实施条例（国务院令第 778 号，第九、十条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2013, amended). Consumer Rights Protection Law, 消费者权益保护法（第八、十条）：「消费者有权根据商品或者服务的不同情况，要求经营者提供……服务的内容、规格、费用等有关情况」「有权拒绝经营者的强制交易行为」. (National Database of Laws and Regulations) - Notes: If you have been overcharged, call 12315 first. The penalty is carried out by the commerce and the market regulation departments, and you do not have to bring a lawsuit yourself. The grade is B because the rules are official but there is no figure for 「how much less you spend by standing beside the work」, and it rests mainly on experience. Asking the price, keeping the receipt and keeping the old parts are worth doing for anybody, whoever they look like. The appliance measures only cover appliance repair; for vehicle inspection and decoration, refer to the provisions on clear price marking and to the implementing regulations. The higher-level Consumer Rights Protection Law also sets out both rights. Article eight says you have the right to require the business to explain the contents, specifications and charges of the service. Article ten says you have the right to refuse a forced transaction. - ### 28. Do not act as a nominal legal representative, and do not lend your identity card to somebody registering a company - Cost: No money. If somebody asks you to 「just be the name on it and not be involved in anything」, say no. @@ -255,7 +236,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: B - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2023). Company Law, 公司法（2023 年修订，第十、十一、一百八十、一百九十一条）. - Notes: The text of the Company Law only says who bears the liability. The more common consequence of being a name on the paper is this: when the company is enforced against by the court, the legal representative is put under a consumption restriction as well, and tax and administrative penalties may reach you too. Those consequences come from other laws and from enforcement rules, and this section has not checked the original text, so the grade is B. Once you have been named, the Company Law allows you to resign, and the company has to appoint a new legal representative within 30 days. - ### 29. Do not carry things through the border for strangers, and do not collect parcels of unknown origin on their behalf - Cost: No money. Just say no. @@ -264,7 +244,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: B - Sources: Consular Affairs Department of the Ministry of Foreign Affairs (外交部领事司) (2023). 对准备出国的中国公民有哪些建议. - Notes: The official guidance page of the General Administration of Customs and the Ministry of Public Security on this subject would not open, so this item quotes only the page of the Consular Affairs Department. The original text also gives principles and no article of law, so the grade is B. Receiving a courier parcel on somebody else's behalf in China is the same, especially when you are asked to pay first and refunded afterwards. - ### 30. A dog must be on a lead: without one, an accident is strict liability, so whether or not you are at fault you pay, and you pay to the end - Cost: A little money. A lead costs several tens of yuan. The hard part is remembering to put it on every time you go out. @@ -282,7 +261,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: B - Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百三十六条）. ; National People's Congress (全国人大) (2020, amended). Criminal Law, 刑法（第二百三十七条）. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau, consolidated text) ; People's Procuratorate of the Xicheng District of Beijing (北京市西城区人民检察院) (2018). 她们为什么不再沉默……. - Notes: The grade is B because the article says only the four characters 「其他手段」. 「Getting somebody drunk and having sex is still rape」 comes from a legal education article by a procuratorate, not from a judicial interpretation. The article on rape speaks only of women. A man attacked this way is dealt with under the crime of forced indecency, so it makes no difference whether the other person is a man or a woman. If you are the one harmed, for what to do first see item 42 in section 13 (after being sexually assaulted). If you are blackout drunk yourself, the two risks of being accused and of being blackmailed are present at the same time: see item 32 in this section (call the police if the other person holds something over you and demands money). - ### 32. After sex or a nude video chat, if the other person threatens to call the police, to send the photographs to your workplace or to tell your employer, pay nothing, delete nothing, and call the police - Cost: No money. Screenshot the other person's account, the chat records and the payment code, and take your phone to the police station. The hard part is resisting the thought of 「paying first to settle it」. @@ -291,7 +269,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: National People's Congress (全国人大) (2020, amended). Criminal Law, 刑法（第二百七十四条）. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau, consolidated text) ; Supreme People's Court and Supreme People's Procuratorate (2013). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Extortion, 关于办理敲诈勒索刑事案件适用法律若干问题的解释（法释〔2013〕10 号）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零三十二、一千零三十三条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千一百九十五条）：「权利人有权通知网络服务提供者采取删除、屏蔽、断开链接等必要措施。通知应当包括构成侵权的初步证据及权利人的真实身份信息」「未及时采取必要措施的，对损害的扩大部分与该网络用户承担连带责任」. (National Database of Laws and Regulations) ; Fujian Provincial Public Security Department (2024). 因为一条「约炮」信息，我背上 23 万的债务. ; Procuratorial Daily (检察日报) (2018). 两次获刑不思悔改　曝光隐私要挟网友（徐州市鼓楼区人民检察院网站转载）. ; Procuratorial Daily (检察日报) (2022). 男子以公布裸照威胁10名女性与其发生关系，大数据筛查「报警」！（最高人民检察院网站）. - Notes: The three bands of amount are set by each province itself, within the range the judicial interpretation gives. More than 2000 to 5000 yuan counts as 「数额较大」. More than 30,000 to 100,000 yuan counts as 「数额巨大」. More than 300,000 to 500,000 yuan counts as 「数额特别巨大」. Which figure applies in your province is settled by the local standard. Extorting three or more times within two years counts as 「repeated」 even when each amount is small. Do not delete the chat records or the photographs, cancel the account or block the person and call it a day: that deletes your own evidence as well, see item 5 in this section (get a lawyer first if you are summoned, and delete nothing). If a case really is opened, those records are at the same time the material that proves your account. Do not try to 「negotiate instalments」 either. Extortion over nude video chats is mostly done by groups outside the country, who record the screen while reading your contacts with a trojan. 「Sending it to your family」 is a line they use, and a case can still be opened after you report. Blackmailing somebody with 「I will tell your family that you like people of the same sex」 is the same kind of extortion. In Yancheng, Jiangsu, a man met someone on a dating app for men and used this to get away with 14,500 yuan; after the victim reported him he was prosecuted, and he had already been convicted twice for the same method. Forcing somebody into sex with nude photographs is pursued as rape. In Xiangshan, Zhejiang, a man tricked several women on a social platform into sending nude photographs and then threatened to publish them, and 10 victims were identified. One of them called the police as soon as she was threatened; the two had never met in person, and the procuratorate still handled it as an attempted rape, prosecuting him for rape and for forced indecency. So do not send nude photographs to people you meet online. In the other direction, if you are the victim going after the infringer for compensation, a high amount does not by itself make it extortion: see item 36 in this section (claiming compensation as the victim). If the photographs really do reach the internet, you may notify the platform to remove them. Article one thousand one hundred and ninety-five of the Civil Code gives you the right to require the platform to delete, block or disconnect the link, and the notice has to include preliminary evidence of the infringement and your real identity information. If the platform does not act in time, it and the person who posted share liability for the enlarged part of the damage. - ### 33. If somebody reports you on facts they made up, you can ask for them to be pursued: 5 days of detention where an administrative penalty applies, and 3 years or less where a crime applies - Cost: No money. Report the case to the public security organ and explain that the other party fabricated it. The hard part is producing something that shows they invented it deliberately. @@ -300,7 +277,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第五十条）. (full text reprinted by the Beijing Municipal Public Security Bureau) ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百四十三条）. - Notes: To pursue the other party, showing only 「I did not do it」 is not enough. You also have to show that they 「fabricated the facts」 and did so 「deliberately」. The second is much harder to prove, so the pursuit may not succeed. There are no published national figures for cases of false accusation that are opened and dealt with, and this book could not find any, so it gives no rate and no breakdown by gender. For the evidence and the remedies after you are accused, see items 34 and 35 in this section (insufficient evidence should mean an acquittal, and state compensation). - ### 34. Insufficient evidence should mean an acquittal, and a confession forced out of you should be excluded; and after a conviction there is still a petition and a retrial - Cost: Going through the procedure costs nothing, but a petition and a retrial both take a great deal of time. You normally need a lawyer, and legal fees start at several thousand yuan. If you cannot afford it you can apply for legal aid: see item 5 in this section (get a lawyer first if you are accused). The hard part is going on waiting round after round without giving up. @@ -318,15 +294,14 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2012, amended). State Compensation Law, 国家赔偿法（第十七、十九、三十三、三十五条）. (reprinted by the National Bureau of Statistics) ; Supreme People's Procuratorate (2026). 最高人民检察院公布最新侵犯公民人身自由的赔偿金标准. ; Supreme People's Court and Supreme People's Procuratorate (2015). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Compensation Cases, 关于办理刑事赔偿案件适用法律若干问题的解释（法释〔2015〕24 号，第七、八条）. ; National People's Congress (全国人大) (2018). Criminal Procedure Law, 刑事诉讼法（2018 年修正，第十六、一百七十七、一百八十一、二百八十四、二百九十条）. . Article nineteen of the State Compensation Law and that interpretation cite the article numbers of the 2012 Criminal Procedure Law（第十五条、第一百七十三条第二款、第二百七十三条第二款、第二百七十九条）, which after the 2018 amendment correspond in order to 第十六条、第一百七十七条第二款、第二百八十四条第二款、第二百九十条 - Notes: The standard that is calculated by the day is updated every May, using the previous year's average annual wage of employed persons in urban non-private units published by the National Bureau of Statistics. Look up the figure for the year you are using it in; do not simply apply the one written in this item. Once you have the decision not to prosecute, read the basis it gives. If it says 「no criminal facts」 or 「insufficient evidence」, you can apply for compensation. If it says 「the circumstances of the offence are trivial」, or that the limitation period has expired, in principle the days you were held cannot be compensated. If you disagree with relative non-prosecution, you may petition to the procuratorate within 7 days of receiving the decision. The payment for mental distress is set by the case-handling organ according to the consequences, and there is no single formula. The organ that owes you the money is the one that decided to detain you, whether that is the police, the procuratorate or the court, so go to that one first. If you are not satisfied with how it is dealt with, apply further up under the procedure of the State Compensation Law. For how it is actually calculated, see the decision in the Guo Li case in item 36 in this section (claiming compensation as the victim). Guo Li was held for 1826 days. The compensation is calculated at the standard of the year the decision was made, not the year he was held, and it used the 2016 national average daily wage of employees of 258.89 yuan. The part covering loss of personal freedom came to about 473,000 yuan (472733.14 yuan), and a payment for mental distress of about 165,000 yuan (165450 yuan) was added on top. The 88.88 million yuan he applied for as ten years of payment for mental distress was not supported beyond that. You also need to know what is not payable: the decision treated lost earnings, compensation for physical health, property loss, legal fees and the living allowance in the detention centre as falling outside the scope of state compensation. - -### 36. If you are the victim going after compensation, go through 12315, a lawsuit or a lawyer. Do not go alone to a meeting the other side sets, and do not tie 「pay me」 and 「I will not publish it」 into one sentence +### 36. As the victim claiming compensation, go through 12315, a lawsuit or a lawyer; do not go on your own to an appointment the other side sets, and do not tie "I will pay" to "I will not make it public" in one sentence -- Cost: No money. Go through 12315, sue in court, or hand the claim to a lawyer. Record the whole of any meeting you do have. The hard part is resisting the urge to talk to the other side on your own. -- In plain language: If you are the victim and seek compensation from the infringer, you have a legal basis, and it does not turn into extortion by itself just because 「your price is high」. But the cost can be large: Guo Li was first sentenced to five years for the crime of extortion and only later acquitted on retrial. So use a route that leaves a trace, do not go on your own to a place the other side chooses, and do not say out loud 「pay me」 and 「I will not publish it」 as one condition. -- Benefit: An official case: after the relevant departments announced in 2008 that certain batches of the 「施恩」 brand milk powder contained melamine, the examination result for Guo Li's daughter was 「双肾中央集合系统内可见数个点状强回声」. Guo Li sought compensation from the seller and from the Shide company many times, and exposed the matter to the media. In 2009-6 the two sides reached an agreement, the Shide company paid 400,000 yuan, and Guo Li produced a written statement saying that he would no longer pursue the case and gave up his claim for compensation. The company then reported to the police, and Guo Li was sentenced by the Chao'an County People's Court to five years' imprisonment for the crime of extortion. The People's Procuratorate of Guangdong Province made a recommendation for retrial. The recommendation held that 「现有证据不足以证实郭利主观上具有非法占有的目的」. As the legal guardian, Guo Li 「有权利向奶粉的生产厂家索赔，其索赔行为有法律依据」. It goes on to say that 「消费者对损害其合法权益的行为，有权通过大众传播媒介予以揭露、批评」. On 2017-4-7 the Guangdong High People's Court heard the case on retrial and acquitted Guo Li, on the ground that 「原审认定郭利以非法占有为目的，使用威胁、要挟的方法，强行索取财物行为的事实不清，证据不足，郭利的行为性质未超出民事纠纷的范畴」. Guo Li was held from 2009-7-23 to 2014-7-22, a total of 1826 days (Guangdong, acquitted on retrial in 2017) +- Cost: No money. You put it in through 12315, by suing in court, or by instructing a lawyer. If you meet the other side to talk, record the whole conversation. The hard part is holding back and not talking to the other side alone. +- In plain language: As the victim, claiming compensation from the party who infringed your rights has a legal basis, and it does not automatically become extortion just because you ask for a high amount. But the price may be high: Guo Li was first sentenced to five years for extortion and was only later acquitted on retrial. So use a route that leaves a record, do not go on your own to a place the other side names, and do not say out loud "I will pay" and "I will not make it public" as one condition. +- Benefit: The official case: in 2008, after the authorities announced that some batches of 「施恩」 brand infant formula contained melamine, the examination result for Guo Li's daughter read 「双肾中央集合系统内可见数个点状强回声」. Guo Li claimed compensation from the seller and from Shien many times, and made the matter public in the media. In 2009-6 the two sides reached an agreement: Shien paid 400,000 yuan in compensation, and Guo Li issued written material saying he would not continue to pursue the case and gave up his claim for damages. The company then reported him to the police, and the Chaonan County People's Court sentenced him to five years' imprisonment for extortion. The Guangdong Provincial Procuratorate put forward a proposal for retrial. The proposal held that 「现有证据不足以证实郭利主观上具有非法占有的目的」. As the statutory guardian, Guo Li 「有权利向奶粉的生产厂家索赔，其索赔行为有法律依据」. The proposal also says that 「消费者对损害其合法权益的行为，有权通过大众传播媒介予以揭露、批评」. On 2017-4-7 the Guangdong High People's Court acquitted Guo Li on retrial. The reason was that 「原审认定郭利以非法占有为目的，使用威胁、要挟的方法，强行索取财物行为的事实不清，证据不足，郭利的行为性质未超出民事纠纷的范畴」. Guo Li was held in custody for 1826 days, from 2009-7-23 to 2014-7-22 (Guangdong, judgment changed on retrial in 2017) - Evidence grade: B -- Sources: People's Procuratorate of Guangdong Province (广东省人民检察院) (2017). 广东省检察院再审建议对郭利案改判无罪. (reprinted by the Heyuan Municipal People's Procuratorate) ; Compensation Committee of the Guangdong High People's Court (广东省高级人民法院赔偿委员会) (2018). 郭利以再审无罪为由申请、广东省潮州市中级人民法院赔偿的国家赔偿案件（（2018）粤委赔 11 号）. -- Notes: The grade is B because this is a single case and not a proportion that can be generalised. It shows two things at once: claiming compensation is lawful, and you may still be treated by the police as a case to be opened. The lowest-risk route is to hand the claim to a lawyer, or to sue in court yourself, so that you never meet the other side alone. Be especially careful when the other side invites you to their own area to 「settle it face to face」, because the report to the police and the opening of the case usually happen on their side. Once you have signed a settlement, stick to it; signing and then raising a separate large claim is the easiest thing to read as 「coercion」. For what to do after a case is opened against you, see items 5, 34 and 35 in this section (get a lawyer first, insufficient evidence should mean an acquittal, state compensation). For the case where the other person holds something over you and demands money, see item 32 in this section (the other person threatening to call the police or to send the photographs to demand money). +- Sources: Guangdong Provincial Procuratorate (2017). 广东省检察院再审建议对郭利案改判无罪. (Guangdong Provincial Procuratorate website) ; Compensation Committee of the Guangdong High People's Court (2018). 国家赔偿案件（（2018）粤委赔 11 号）: 郭利以再审无罪为由申请、广东省潮州市中级人民法院赔偿. +- Notes: The grade is B because this is one case, not a proportion that can be generalised. It also establishes two things: claiming compensation is lawful in itself, and you may still end up being investigated by the police as a case. The lowest-risk route is to hand the claim to a lawyer, or to sue in court yourself, and not to come into contact with the other side alone. Be especially careful when the other side invites you to their own locality to "settle it face to face": the report to the police and the opening of a case usually happen where they are. Once you have signed a settlement, follow the agreement; signing it and then starting a second, much larger claim is the thing most easily read as "duress". What to do after a case has been opened, see items 5, 34 and 35 in this section (get a lawyer first, insufficient evidence should mean an acquittal, state compensation). Where the other side uses hold-up material to demand money from you, see item 32 in this section (being blackmailed with a police report or photographs). ### 37. If you are piled on online: turn on the protection, preserve the evidence first, and then choose between the three routes of the platform, an injunction and the police @@ -345,7 +320,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2015, amended). Insurance Law of the People's Republic of China, 中华人民共和国保险法第二十七、三十四、三十九、四十三条. ; National People's Congress (全国人大) (2020). Civil Code of the People's Republic of China, 中华人民共和国民法典第一千一百二十五条. ; Standing Committee of the National People's Congress (全国人大常委会). Criminal Law of the People's Republic of China, consolidated text as amended by Criminal Law Amendment Eleven, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第一百九十八条、第二百三十二条. ; Anhui Provincial People's Procuratorate, reprinting Legal Daily (2020-01-13). 杀妻骗保牵出案中案（安徽省人民检察院民事虚假诉讼监督典型案例）. - Notes: The person who benefits from this item is you, and what it avoids is the death penalty or life imprisonment. The casualties and the judgment in the case are taken from the original text of the typical case published by the procuratorate, and the entry does not set out details beyond what the investigation disclosed. Fraud against insurers that involves no loss of life, such as faking an accident or inflating the loss, is also a crime, and the people who testify for you are counted with you: see item 21 in section 9 (faking an accident or inflating the loss to claim insurance). Taking out insurance for your family is normal in itself; what has to be guarded against is the thought of 「buy the insurance, then do it」. Treat an impulse to hurt a family member as an emergency: see items 14 and 15 in this section (the thought arising in yourself, and the harsh words said by someone close). - ### 39. Ask for a case acceptance receipt when you call the police, and for a written notice if no case is opened: you can apply for reconsideration within 7 days and for review within a further 7 days, and the procuratorate can order the police to open a case - Cost: No money. Ask one extra question on the spot and get the case acceptance receipt into your hands. Afterwards chase the progress on the dates set out above. The hard part is going back to ask, again and again. @@ -354,15 +328,15 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: National People's Congress (全国人大) (2018). Criminal Procedure Law, 刑事诉讼法（2018 年修正，第一百一十二、一百一十三条）. ; Ministry of Public Security (公安部) (2020). Decision on Amending the Provisions on the Procedures for Handling Criminal Cases by Public Security Organs (Ministry of Public Security Order No. 159) and the full text as amended, 关于修改《公安机关办理刑事案件程序规定》的决定（公安部令第 159 号）及修正后全文. ; Ministry of Public Security (公安部) (2013, amended 2019). Provisions on the Procedures for Handling Administrative Cases by Public Security Organs (Ministry of Public Security Order No. 125, as amended by Order No. 149), 公安机关办理行政案件程序规定（公安部令第 125 号，第 149 号修改决定）. ; ; State Council (国务院) (2011). Regulations on the Supervision of Public Security Organs, 公安机关督察条例（国务院令第 603 号，第九条）. - Notes: The person who benefits is you. Reporting a case does not mean it will be dealt with, but whether it is dealt with and how quickly are both set out in writing, with time limits. Remember the three steps in order. First, get the case acceptance receipt on the spot and write down the case number and the unit that accepted it. Second, if nothing has come of it by the deadline, go and ask about the progress on the dates set out above, and ask 「which stage the case is at」 rather than 「why is nobody dealing with me」. Third, once you have the notice of non-acceptance, read the reason, apply for reconsideration within 7 days and for review within a further 7 days, and at the same time ask the People's Procuratorate to supervise the opening of the case; the three routes do not conflict. A public security case is the sort of thing that covers a fight or petty theft. The time limit for handling runs from the day the case is accepted and does not exceed 30 days. Where the case is major or complicated it may be extended by a further 30 days with the approval of the public security organ at the level above. Time spent on assessments does not count. If you were treated badly, complain through the supervision channel; the case itself still follows the process for opening a case. Finding lost property and disputes between neighbours were never criminal cases in the first place, and the public security organ handles them as administrative cases or through mediation, or may only register them for the record, so do not read them as 「nobody dealt with my report」. For the process for stopping a payment after you have been scammed, see item 2 in this section (call 110 at once and ask for a stop). For being piled on online, see item 37 in this section (preserve the evidence first, then choose a route). - -### 40. Do not hand money or cards to the people handling a case or enforcing the law: bribery is a crime in itself, and bribing supervisory, administrative enforcement or judicial personnel is punished more severely +### 40. Do not hand money or gift cards to the people handling a case or enforcing the law: bribery is punished for the giver too, and bribing supervision, enforcement or judicial staff is one of the cases treated as heavier - Cost: No money. It is only a matter of not doing it. -- In plain language: 「Spending a bit of money to settle it」 is a separate offence in the Criminal Law, and the person caught is not only the one who took the money. Bribery carries 3 years or less. Where the circumstances are serious it carries 3 to 10 years, and where they are especially serious, 10 years or more or life imprisonment. Bribing supervisory, administrative enforcement or judicial personnel is one of the aggravating circumstances written into the text. Money handed over after being extorted, where you obtained no improper benefit, is not bribery. -- Benefit: Article three hundred and eighty-nine of the Criminal Law defines bribery:「为谋取不正当利益，给予国家工作人员以财物的，是行贿罪」. The same article also says:「在经济往来中，违反国家规定，给予国家工作人员以财物，数额较大的，或者违反国家规定，给予国家工作人员以各种名义的回扣、手续费的，以行贿论处」. It also leaves out one exception:「因被勒索给予国家工作人员以财物，没有获得不正当利益的，不是行贿」. The sentence is set out in Article three hundred and ninety of the Criminal Law, as amended by Criminal Law Amendment Twelve and in force from 2024-3-1. The original text says:「对犯行贿罪的，处三年以下有期徒刑或者拘役，并处罚金；因行贿谋取不正当利益，情节严重的，或者使国家利益遭受重大损失的，处三年以上十年以下有期徒刑，并处罚金；情节特别严重的，或者使国家利益遭受特别重大损失的，处十年以上有期徒刑或者无期徒刑，并处罚金或者没收财产」. There are seven circumstances to be punished more heavily:「（一）多次行贿或者向多人行贿的；（二）国家工作人员行贿的；（三）在国家重点工程、重大项目中行贿的；（四）为谋取职务、职级晋升、调整行贿的；（五）对监察、行政执法、司法工作人员行贿的；（六）在生态环境、财政金融、安全生产、食品药品、防灾救灾、社会保障、教育、医疗等领域行贿，实施违法犯罪活动的；（七）将违法所得用于行贿的」. The last paragraph leaves room for a lighter sentence if you come forward yourself:「行贿人在被追诉前主动交待行贿行为的，可以从轻或者减轻处罚。其中，犯罪较轻的，对调查突破、侦破重大案件起关键作用的，或者有重大立功表现的，可以减轻或者免除处罚」(nationwide) +- In plain language: "Spend a little money and sort it out" is a separate crime under the Criminal Law, and the person caught is not only the one who took the money. Bribery carries 3 years or less. A serious case carries 3 to 10 years, and a particularly serious case 10 years or more or life imprisonment. Bribing supervision, administrative enforcement or judicial staff is one of the cases the law expressly treats as heavier. Money handed over because you were extorted, where you received no improper benefit, is not bribery. +- Benefit: Article three hundred and eighty-nine of the Criminal Law defines bribery:「为谋取不正当利益，给予国家工作人员以财物的，是行贿罪」. The same article also says:「在经济往来中，违反国家规定，给予国家工作人员以财物，数额较大的，或者违反国家规定，给予国家工作人员以各种名义的回扣、手续费的，以行贿论处」. It also leaves an exception:「因被勒索给予国家工作人员以财物，没有获得不正当利益的，不是行贿」. The sentences are in Article three hundred and ninety of the Criminal Law (amended by Criminal Law Amendment Twelve, in force from 2024-3-1). The text is:「对犯行贿罪的，处三年以下有期徒刑或者拘役，并处罚金；因行贿谋取不正当利益，情节严重的，或者使国家利益遭受重大损失的，处三年以上十年以下有期徒刑，并处罚金；情节特别严重的，或者使国家利益遭受特别重大损失的，处十年以上有期徒刑或者无期徒刑，并处罚金或者没收财产」. There are seven circumstances that count as heavier:「（一）多次行贿或者向多人行贿的；（二）国家工作人员行贿的；（三）在国家重点工程、重大项目中行贿的；（四）为谋取职务、职级晋升、调整行贿的；（五）对监察、行政执法、司法工作人员行贿的；（六）在生态环境、财政金融、安全生产、食品药品、防灾救灾、社会保障、教育、医疗等领域行贿，实施违法犯罪活动的；（七）将违法所得用于行贿的」. The last paragraph leaves room for a lighter sentence where the briber volunteers beforehand:「行贿人在被追诉前主动交待行贿行为的，可以从轻或者减轻处罚。其中，犯罪较轻的，对调查突破、侦破重大案件起关键作用的，或者有重大立功表现的，可以减轻或者免除处罚」(nationwide) - Evidence grade: A -- Sources: Standing Committee of the National People's Congress (全国人大常委会). Criminal Law of the People's Republic of China, consolidated text as amended by Criminal Law Amendment Eleven, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第三百八十九条. ; Standing Committee of the National People's Congress (全国人大常委会) (2023). Criminal Law Amendment (Twelve), 中华人民共和国刑法修正案（十二）（第五项，修改刑法第三百九十条；自 2024 年 3 月 1 日起施行）. -- Notes: The person who benefits is you. Do not expect a bribe to buy your way out of punishment. The other side has colleagues, ledgers, cameras and an inspection function, so what you hand over is the evidence against them when they are investigated, and it is evidence of your own crime as well; and this kind of bribery is one of the aggravating circumstances written into the text. If you really do meet somebody asking for a benefit, follow the route in item 39 in this section (ask for a case acceptance receipt when you call the police): keep the receipt and the documents, and report it to the supervision body or to the discipline inspection and supervision organ. The exemption in the third paragraph of Article three hundred and eighty-nine of the Criminal Law requires that you were extorted and obtained no improper benefit; the moment you use the opportunity to get hold of something you could not otherwise have got, that condition no longer holds. The red envelope given inside a hospital follows a different set of rules: see item 12 in section 24. +- Sources: Standing Committee of the National People's Congress. Criminal Law of the People's Republic of China, consolidated text as amended by Criminal Law Amendment Eleven（中华人民共和国刑法（经刑法修正案十一修正的整合文本））第三百八十九条. ; Standing Committee of the National People's Congress (2023). Criminal Law Amendment (Twelve) of the People's Republic of China（中华人民共和国刑法修正案（十二））(item five, amending Article three hundred and ninety of the Criminal Law; in force from 2024-3-1). +- Notes: The person who benefits is you yourself. Do not expect a bribe to get you off. The other side has colleagues, records, cameras and supervision inspectors: what you hand over is evidence against them when they are investigated, and it is evidence of your own crime, and this kind of bribery is expressly one of the heavier cases. If you really meet somebody demanding a favour, use the route in item 39 (ask for the case acceptance receipt when you call the police): keep the receipt and the documents, and report it to the supervision organ or the discipline inspection organ. The exemption in the third paragraph of Article three hundred and eighty-nine of the Criminal Law is "extorted and no improper benefit received"; the moment you use the opportunity to get something you could not otherwise have got, that condition is not met. Red envelopes given to hospital staff follow a different set of rules, see item 11 in section 24. + ### 41. Turn on the recorder for phone calls and meetings that could turn: you do not need the other person's consent first for a conversation you are party to @@ -372,7 +346,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2023, amended, in force 2024-1-1). Civil Procedure Law, 民事诉讼法（第六十六条）, full text reprinted by the Shanghai Municipal Development and Reform Commission. ; Supreme People's Court (2022, second amendment). Interpretation on the Application of the Civil Procedure Law of the People's Republic of China, 关于适用《中华人民共和国民事诉讼法》的解释（第一百零六条）. ; Supreme People's Court (2019, amended). Several Provisions on Evidence in Civil Litigation, 关于民事诉讼证据的若干规定（第十四、十五、九十条）. ; Liu Danni (刘丹妮) (2023). 电影《消失的她》中的法律（人民法院报，最高人民法院官网转载）. - Notes: The people who benefit are you and your family. In a recording, try to get the person, the matter, the amount and the time to line up. Do not only ask 「when are you going to give it to me」; ask it naturally as 「the 50,000 yuan you borrowed last year, when are you going to repay it」. Hand the recording to the court, the arbitration commission or the police to use; do not put it online, because publishing what other people have said can bring privacy and reputation trouble: see item 16 in this section (no abuse online, no rumours). Having a recording does not mean you will win, because the court still weighs the other evidence for truthfulness and relevance. A recording only settles what can be proved; whether to sue at all, and how long it takes, has to be worked out separately. For the material to store before you leave a job, see item 8 in section 19 (save your payslips and attendance records before you leave). The basis of this item is the rules of civil litigation; how evidence is used in criminal and administrative cases is governed separately. - ### 42. When something happens at a scene, film the wide shot first, then how things are positioned, then the damage and the injuries, and do not delete the original photographs and video - Cost: No money. Hold your phone and film continuously for several tens of seconds to a few minutes. Make sure you are safe first, and do not delay helping or dealing with it. @@ -381,7 +354,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: C - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2023, amended, in force 2024-1-1). Civil Procedure Law, 民事诉讼法（第六十六条）, full text reprinted by the Shanghai Municipal Development and Reform Commission. ; Supreme People's Court (2019, amended). Several Provisions on Evidence in Civil Litigation, 关于民事诉讼证据的若干规定（第十四、十五条）. - Notes: The people who benefit are you and your family. The grade is C because the order of filming is only drawn from experience; the article only says that photographs and video can be evidence and that the original has to be produced. If somebody is hurt, if there is a fire or a gas leak, or if any other danger is still there, help first and get clear first, and do not delay that to take photographs. For what to do at the scene of a traffic accident, see item 1 in this section (after a traffic accident, stop, help the injured and call the police). - ### 43. If you are being abused at home: call the police first so that there is a record of the officers sent, then apply to the court for a personal safety protection order; you do not have to divorce first, and it costs nothing - Cost: No money. Applying for a personal safety protection order costs no court fee, and no security has to be deposited. What it costs is the time spent calling the police, having an injury assessed and writing the application. The hardest part is saying this out loud to your own family. @@ -390,7 +362,6 @@ This section counts money and personal freedom, and is not converted into lifesp - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2015). Anti-Domestic Violence Law, 反家庭暴力法（第二、十三、十五、十六、十九、二十三、二十八、二十九、三十、三十四条）. ; Supreme People's Court (2022). Provisions on Several Questions Concerning the Application of the Law in Handling Personal Safety Protection Order Cases, 关于办理人身安全保护令案件适用法律若干问题的规定（法释〔2022〕17 号，第一、三、六、十、十二条）. ; Supreme People's Court (2016). Reply on Procedural Questions Concerning Personal Safety Protection Order Cases, 关于人身安全保护令案件相关程序问题的批复（法释〔2016〕15 号，第一、二项）, full text reprinted by the Pingyuan County People's Court. ; Stöckl H et al. (2013). The global prevalence of intimate partner homicide: a systematic review. Lancet. ; Glass N et al. (2008). Non-fatal strangulation is an important risk factor for homicide of women. Journal of Emergency Medicine. - Notes: The people who mainly benefit are you and your family. The homicide figures come from outside China, and there are no published Chinese statistics on the same measure. Build up the evidence: the record of officers sent, a warning letter, the result of an injury assessment, medical records, a written guarantee from the other person, chat records and recordings can all be used. For recordings, see item 41 in this section (recordings). A protection order does not need a lawyer, and there is usually an answer within days. Divorce or compensation is a separate case, where six months or more in the ordinary procedure at first instance is normal and legal fees are yours to pay. A victim of domestic violence may apply for legal aid, and the court may defer, reduce or waive the court fee. A protection order is not a guarantee of personal safety; if there is danger in front of you, call 110 first. Where the victim is forced or frightened and cannot apply themselves, an immediate family member, the public security organ, the women's federation, the residents' committee or the villagers' committee may apply on their behalf. If the person being abused is a friend or a colleague, you can go with them to the police and to hospital. If you see somebody in a stranger's home being beaten, calling 110 is enough. Do not step in and pull them apart yourself, and the reason is in item 10 in this section (call the police when a conflict starts). - ### 44. If a member of your family has run up gambling debts, do not rush to pay them for them: the law does not protect gambling debts, and money borrowed to gamble is not a joint marital debt - Cost: No money. The hard part is holding out when the creditor comes to the door and your family cries and pleads. If the creditor really does sue, you have to respond to the case. diff --git "a/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" "b/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" index a3031767..038a8c33 100644 --- "a/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" +++ "b/en/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" @@ -12,7 +12,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第二十九条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（第三十二条，即刑法第二百九十一条之一第二款）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百九十一条之一）. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau) - Notes: Knowingly reposting something false and writing it yourself are dealt with by the police under the same article. 「I was only reposting it」 is no excuse. Doctored pictures and fake scene images generated by AI count as writing it yourself in the same way. The Ministry of Public Security published several batches of typical cases of online rumours about disasters and dangers in 2025. The Ministry's website would not open while this section was being written, so no specific case is cited here. - ### 2. Do not post or repost pictures, stickers or comments that insult martyrs, and do not make fun of the national flag or the national anthem - Cost: No money. On content involving martyrs, the national flag, the national emblem or the national anthem: do not comment, do not doctor the picture, do not repost. The hard part is holding back from joining in when somebody in the group continues a joke. @@ -21,7 +20,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第三十五条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Law on the Protection of Heroes and Martyrs, 英雄烈士保护法（第二十二、二十六条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第三十五条，即刑法第二百九十九条之一）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百九十九条、第二百九十九条之一）. ; Supreme People's Court (2022). Ten Typical Cases Involving the Protection of the Rights and Interests of Martyrs (Case Two, Case Three), 涉英烈权益保护十大典型案例（案例二、案例三）. - Notes: This article of the Law on Penalties for Administration of Public Security was only added in the 2025 revision. Before that, the only routes were the Criminal Law, or a civil public interest litigation brought by the procuratorate. In the eyes of a court, a WeChat group of several hundred people counts as a public place. - ### 3. Close the page once you have read something on an overseas site: do not screenshot it, repost it or move it into a group chat, and do not join in under the comments - Cost: No money. Close the page and the matter is finished. Do not archive it, do not repost it, do not move it onto a domestic platform. Do not use an account that can be matched to your own identity to post political content on an overseas platform over the long term. Do not accept fees or funding from overseas institutions, organisations or individuals. The hard part is holding back from screenshotting and sharing content that startles you. @@ -30,7 +28,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第四十五、五十五、五十六、一百零三、一百零五、一百零六、一百零七、一百一十一、一百一十三条，这几条未经修正案修改）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第十三条第二款、第七十一条第一项）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Criminal Law Amendment (Nine), 刑法修正案（九）（新增第一百二十条之六）, full text reprinted by the National Petition Office. - Notes: The provisions cited in this item punish inciting others, passing things on, and providing things to other people. Merely looking is not among them. Material about terrorism and extremism is the exception written into the article, and simply holding it can already amount to a conviction (Article one hundred and twenty, item six), so this category must never be stored. Once you have seen it, that is where it ends: do not screenshot and store it, do not repost it, do not add a comment, and do not pull other people into reposting it. Both reposting and joining in under the comments count as spreading when a case is handled, and 「I was only reposting it」 is no excuse. Content that comes from an overseas platform may also attract a heavier penalty. If you are unsure whether something is true, do not join in either: see item 1 in this section (do not repost messages of unknown truth). For how the circumvention tools themselves are punished, see item 11 in section 11. Passing this kind of content on brings you no practical good, while the price is a prison sentence, deprivation of political rights and confiscation of property. Publicly checkable cases of this kind appear only in overseas media and in documents released by families, such as a blogger who had long posted political commentary on an overseas platform being sentenced to 7 years in 2023. This book has not obtained a judgment that can be checked through official channels, so it writes from the articles alone and attaches no case. - ### 4. Pornographic video is yours alone to watch: do not post it in a group chat, do not sell 「the material」, and do not set up a group - Cost: No money. Do not repost it, do not save it into a group chat's files, and do not become the owner or an administrator of that kind of group. @@ -39,7 +36,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第八十条）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第三百六十三、三百六十四条）. ; Supreme People's Court and Supreme People's Procuratorate (2010). Interpretation (Two) on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Producing, Copying, Publishing, Selling or Spreading Obscene Electronic Information by Means of the Internet, Mobile Communication Terminals and Voice Messaging Platforms（第三条）. - Notes: Both of these laws punish 「spreading」, and watching privately is not within them. But posting into a group chat once already counts as spreading to the whole group. What 「情节严重」 means in terms of quantity is set by the judicial interpretation, and this section has not verified the specific figures. - ### 5. If a 「side job」 wants you to receive money, take out cash or make transfers through your own card, do not do it whatever fee you are offered - Cost: No money. Turn down that kind of 「a few hundred a day for carrying a card to the bank」 work. @@ -57,7 +53,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第十一条，修改刑法第一百七十五条之一）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第八十八、一百九十三、二百八十条）. ; Supreme People's Procuratorate and Ministry of Public Security (2022). Provisions (Two) on the Standards for Opening Cases and Pursuing Criminal Cases under the Jurisdiction of Public Security Organs, 关于公安机关管辖的刑事案件立案追诉标准的规定（二）（第二十二、四十五条）. ; Supreme People's Procuratorate (2026). 骗贷流水线背后藏洗钱黑手（检察日报）. ; Standing Committee of the National People's Congress (全国人大常委会) (2012). Exit and Entry Administration Law, 出境入境管理法（第十二条）. (National Immigration Administration) ; Standing Committee of the National People's Congress (全国人大常委会) (2023, amended). Civil Procedure Law, 民事诉讼法（第二百六十六条）. (reprinted by the Guizhou Provincial Administration for Market Regulation) - Notes: In that case, the statement 「I meant to repay it」 was not accepted. The procuratorate found that these people had no job and no fixed assets, could not possibly repay, and had spent the money they were given, and on that basis found that they had never intended to repay. People who took out cash for the gang may also pick up a separate count of money laundering. Whoever's name the loan is in, and whoever signed the contract, is the borrower. Even where it is not a crime, the debt and the entry in your credit record are yours. Taking the money and running abroad settles nothing: once a case has been opened, however many years you stay away it can still be pursued. If you have been sentenced and have not served your sentence, or you are being treated as a suspect or a defendant, you may not leave the country. Where a civil case is not closed, the court may stop you from leaving the country; where a person under enforcement does not perform, the court may also restrict their leaving the country. For taking out cash and making transfers on somebody else's behalf, see item 5 in this section (a 「side job」 that wants you to receive money through your own card). For how to find out that somebody has taken out a loan in your name, see item 9 in section 8. - ### 7. Do not withdraw money from a bank card you pick up; contact the owner or hand a phone or wallet you find to the police, and do not pocket it - Cost: No money, and it takes a few minutes. Make one call to find the owner, or take the thing to the police station. @@ -66,7 +61,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第三百一十四条）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第一百九十六、二百七十条）. ; Supreme People's Procuratorate (2008). Reply on How to Characterise the Act of Picking Up Another Person's Credit Card and Using It on an Automated Teller Machine (ATM), 关于拾得他人信用卡并在自动柜员机（ATM 机）上使用的行为如何定性问题的批复. - Notes: The crime of embezzlement is dealt with only on complaint, which means the owner has to sue in court themselves and the police do not take it up of their own accord. But taking out cash with a card you found is different: the owner does not have to sue, and the police open the case directly. What 「数额较大」 comes to in money is set by the judicial interpretation, and this section has not verified it. After you have handed something over to the department concerned, the owner has to pay a storage fee when they come to collect it (Article three hundred and seventeen of the Civil Code). - ### 8. Claim a discount in your own real identity, do not buy virtual numbers to register batches of accounts, and do not rely on a loophole to claim it again and again - Cost: No money. Turn down the work in the 「deal groups」 that hand out wool-pulling tasks. If you find a loophole in a platform, leave it alone. The hard part is holding out against treating a discount as income you can copy in bulk. @@ -84,7 +78,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第四十三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第三十三条，即刑法第二百九十一条之二）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百九十一条之二）. ; Supreme People's Court (2024). Typical Cases of Punishing Crimes Endangering Public Security According to Law (Case One), 依法惩治危害公共安全犯罪典型案例（案例一）. ; Supreme People's Court (2025). 专访最高法副院长李勇：让高空抛物行为人都能被处罚. - Notes: You can be detained even when it hits nobody, because the article only requires 「a risk」. The Civil Code has a separate civil compensation rule on throwing objects from a height (Article one thousand two hundred and fifty-four). The original text would not open while this section was being written, so it is not quoted here. - ### 10. Do not buy imitation guns, steel pellet guns or air guns online, and do not buy a 「toy gun」 of unknown origin - Cost: No money. If you want to shoot, go to a proper shooting range. @@ -93,7 +86,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第一百二十八条）. ; Supreme People's Court and Supreme People's Procuratorate (2018). Reply on Questions Concerning Conviction and Sentencing in Criminal Cases Involving Guns and Air-gun Pellets Powered by Compressed Gas, 关于涉以压缩气体为动力的枪支、气枪铅弹刑事案件定罪量刑问题的批复（法释〔2018〕8 号）. ; Supreme People's Court (2018). 《批复》的理解与适用. - Notes: The key to that case not being prosecuted was the finding that they were 「not aware」. For an adult who deliberately goes online to buy a steel pellet gun or an air gun, afterwards saying you did not know it was a gun is very hard to establish. How crossbows and controlled knives are dealt with is not checked in this section, so nothing is written about it. - ### 11. Register a drone you have bought under your real name first, do not fly over airports, military areas or controlled airspace over a city, and do not defeat the height limit - Cost: No money. Real-name registration is free and takes a few minutes. Before every flight, check through an official channel whether the place you intend to fly over is controlled airspace. @@ -102,7 +94,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: State Council and Central Military Commission (国务院、中央军委) (2023). Interim Regulations on the Flight Administration of Unmanned Aircraft, 无人驾驶航空器飞行管理暂行条例（国令第 761 号，第十、十九、四十七、五十一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第四十六条）. ; Hubei Provincial Public Security Department (2026). 公安部公布依法打击无人机「黑飞」违法犯罪八起典型案例（转载人民公安报）. - Notes: Flying a micro or light drone in airspace where flight is permitted does not need approval. A micro drone is one whose empty weight is under 0.25 kg. The airspace where flight is permitted is everywhere outside controlled airspace, and the areas around airports, military facilities and government organs are mostly controlled airspace. The Ministry of Public Security's own page would not open while this section was being written, so the reprint by the Hubei Provincial Public Security Department is cited instead. - ### 12. Do not film people in secret, and do not install cameras in a rented flat, a guesthouse room or a hotel room - Cost: No money, and no time either. Do not film people in secret, and do not install cameras in a rented flat, a guesthouse room or a hotel room. @@ -111,7 +102,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第五十条）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十四条）. ; Supreme People's Court (2024). Typical Cases of Punishing the Crime of Illegally Producing, Selling or Using Wiretapping and Secret Photography Equipment According to Law (Case Two), 依法惩治非法生产、销售、使用窃听、窃照设备犯罪典型案例（案例二）. - Notes: Once secret footage is sold or circulated, it is handled separately as, for example, the crime of producing or selling obscene material for profit, and in the same batch of cases there are 10-year sentences. A camera in your own home does not count; a camera in a room somebody else uses does. - ### 13. Mahjong and cards are fine to play, but do not take a cut, do not be the banker, do not set up the game for a fee, and do not play online gambling - Cost: No money. Small amounts of play for fun among friends and relatives are fine. Do not take a cut of the game, do not be the banker, do not set up the game in order to collect money. The hard part is holding out when somebody else is organising it. @@ -129,7 +119,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2022). Wildlife Protection Law, 野生动物保护法（2022 年修订，第三十一、五十一、五十三条）. (reprinted by the Ministry of Ecology and Environment) ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第三百四十一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第四十一条，即刑法第三百四十一条第三款）. ; Supreme People's Court and Supreme People's Procuratorate (2022). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Damaging Wild Animal Resources, 关于办理破坏野生动物资源刑事案件适用法律若干问题的解释（法释〔2022〕12 号，第十三条）. ; Supreme People's Procuratorate (2023). 鹦鹉小案 民生大案. - Notes: 「Generally not treated as a crime」 refers only to animals bred in captivity and kept as pets. Animals taken from the wild and animals of unknown origin are not covered. Which animals count as being under national key protection is settled by the list. Checking the list before you buy is less trouble than explaining it afterwards. - ### 15. Do not detain anybody to collect a debt, do not lock anybody up, and do not follow them home and refuse to leave - Cost: No money. If the other side owes money and will not pay, sue in court. For how to write an IOU, see item 18 in section 8 (an IOU). The hard part is holding back from using force, and from following them to their home, when the other side defaults. @@ -138,15 +127,14 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第四十七条）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百三十八条）. - Notes: 「He really does owe me money」 does not help here. The article says in terms that detaining or locking somebody up in order to collect a debt is handled as unlawful detention. There are several cases on the Chinese courts' online service of people being locked up over a lawful debt, with sentences of 7 months to 1 year. That website would not open while this section was being written, so nothing is cited here. - -### 16. Do not lend your identity card to anybody, do not use somebody else's, and do not use another person's documents to register, open a bank account or buy a ticket +### 16. Do not lend out your identity card: do not use someone else's, and do not register, open a card or buy a ticket on someone else's documents -- Cost: No money. If somebody asks to borrow it, say no. If you have to hand over a photocopy, write on it which matter this one is for. -- In plain language: Lending, renting or transferring your identity card to somebody else brings a warning and a fine of up to 200 yuan. Impersonating somebody with their identity card, or using one you obtained by deception, carries a fine of 200 to 1000 yuan, or up to 10 days of detention. Using a false document or somebody else's document in a place where identity is required, such as buying a train ticket or staying in a hotel, carries criminal detention or public surveillance where the circumstances are serious. -- Benefit: Renting out, lending or transferring your resident identity card to somebody else brings a warning and a fine of up to 200 yuan. Impersonating somebody with their identity card, or using an identity card obtained by deception, carries a fine of 200 to 1000 yuan, or up to 10 days of detention. Buying, selling or using a forged or altered identity card falls in the same band. In some situations the state requires you to produce proof of identity. In those situations, using a forged or altered document, or another person's stolen document, carries a prison sentence where the circumstances are serious. Documents include the identity card, the passport, the social insurance card and the driving licence. The penalty is criminal detention, which is short-term confinement, or public surveillance, which is confinement-free but restricts your movement, and a fine is imposed in addition or instead, so the fine can come with the sentence or on its own (nationwide) +- Cost: No money. If somebody asks to borrow it, say no. If you have to hand over a photocopy, write on the copy which single thing this one is for. +- In plain language: Lending, renting or transferring your identity card to another person draws a warning and a fine of up to 200 yuan. Impersonating with someone else's identity card, or using one you obtained by deception, draws 200 to 1000 yuan, or up to 10 days of detention. Using a forged document or someone else's in a place where the rules require you to identify yourself by name — buying a train ticket, staying in a hotel — means a custodial sentence or public surveillance in serious cases. +- Benefit: Renting out, lending or transferring your resident identity card to another person draws a warning and a fine of up to 200 yuan. Impersonating with someone else's identity card, or using one obtained by deception, draws 200 to 1000 yuan, or up to 10 days of detention. Buying, selling or using a forged or altered identity card falls in the same bracket. Some activities are required by national rules to present identification. In those, using a forged, altered or stolen document means a prison sentence in serious cases. The documents covered include the identity card, passport, social security card and driving licence. The penalties are detention (short-term confinement) or public surveillance (no confinement but restricted movement), with a fine imposed alongside or separately: conviction can carry a fine, and sometimes the fine is all there is (nationwide) - Evidence grade: A -- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2011). Resident Identity Card Law, 居民身份证法（2011 年修正，第十六、十七条）. (reprinted by the Beijing Municipal Public Security Traffic Management Bureau) ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十条之一）. -- Notes: An identity card that has been lent out may be used to register a company or to get a bank card. For the consequences of both, see items 28 and 8 in section 8 (do not act as a nominal legal representative, and do not lend a bank card to somebody). Using somebody else's identity card to buy a train ticket or to stay in a hotel falls under the 「activities in which identity must be provided」 that the article describes. No official case was found for this item. +- Sources: National People's Congress (全国人大) (2011). Resident Identity Card Law（居民身份证法）(2011 amendment, Articles 16 and 17). （reprinted by the Beijing Municipal Public Security Traffic Management Bureau）；National People's Congress (全国人大) (2020). Criminal Law（刑法）(as amended by Criminal Law Amendment (Eleven), Article 280, paragraph one). ；State Council (国务院) (2002). Regulations on the Administration of Business Premises for Internet Access Services（互联网上网服务营业场所管理条例）(State Council Order No. 363, Article 23). +- Notes: An identity card that has been lent out can be used to register a company or to open a bank account; what follows from each is in items 28 and 8 in section 8 (do not act as a nominal legal representative, do not lend your bank card to anyone). Buying a train ticket or staying in a hotel with someone else's identity card is one of the "activities where identification must be provided" the law mentions. No official case was found for this item. Going to an internet cafe is the same, and uses your own card: internet cafes must check the identity card they register, keep the registered details and the browsing records for at least 60 days, and the public security and culture departments may call for them at any time. Whose card you use to get online, whatever was done on that machine in that period is recorded against that person. ### 17. Cooperate when the police stop you or take you away: do not push, do not hit, do not grab the body-worn camera, and after a drink be even more careful @@ -156,7 +144,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第六十一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第三十一条，即刑法第二百七十七条第五款）. ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百七十七条）. ; Supreme People's Court and Supreme People's Procuratorate (2025). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Assaulting the Police, 关于办理袭警刑事案件适用法律若干问题的解释（高检发释字〔2025〕1 号，第一、八、十二条）. ; Supreme People's Procuratorate (2021). 袭警罪入刑一月 检察机关批捕 405 人. - Notes: The Supreme People's Procuratorate reported in 2021 that in the first month after assaulting the police became a separate charge, it approved the arrest of 405 people. Of 42 cases in Jiangsu, 47.6% arose from making a scene after drinking. Of 16 cases in Hubei, 12 were only punching, kicking and biting. Laying hands on an auxiliary police officer is handled under the crime of obstructing official duties. No official case setting out the facts of an individual case was found for this item. - ### 18. If the other person is under 14 you must not have sex, and 「she agreed」 is not a reason - Cost: No money. If you are not sure how old the other person is, do not have sex. @@ -174,7 +161,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，第九、五十一条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百三十四条）. ; Supreme People's Court (2022, amended). Interpretation on Several Questions Concerning the Application of the Law in Trying Cases of Personal Injury Compensation, 关于审理人身损害赔偿案件适用法律若干问题的解释（法释〔2022〕14 号，第六至十二、十五条）. ; Supreme People's Procuratorate and Ministry of Public Security (2023). Guiding Opinions on Properly Handling Cases of Minor Injury According to Law, 关于依法妥善办理轻伤害案件的指导意见. ; Supreme People's Procuratorate (2023). 检察机关依法妥善办理轻伤害案件典型案例. - Notes: For how to avoid a conflict, why the first person to swing loses out, and where the boundary of self-defence lies, see item 10 in section 8 (conflict and self-defence). For the bill for laying hands on somebody out of anger, see items 11 to 14 in section 8. A fight arising from a civil dispute and of relatively minor circumstances can be mediated by the police. Where a minor injury has been caused, if you agree compensation with the other person and actually pay it, the punishment can be lighter and there may even be no prosecution, but not one yuan of the compensation can go unpaid. What counts as a minor injury is set by the assessment standards, and this section has not verified that. Lost earnings and nursing costs are calculated on actual income and on the number of days lost, and there is no single figure. - ### 20. If you cannot raise a child after it is born, registration with the civil affairs department is the only lawful way out: handing the child over for money can be punished as trafficking, and walking away is abandonment - Cost: No money. To register an adoption, go to the civil affairs department at county level or above, hand in the documents, and wait for the review and the adoption assessment, which takes some time. @@ -183,7 +169,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百四十、二百六十一条）. ; Supreme People's Court, Supreme People's Procuratorate, Ministry of Public Security and Ministry of Justice (2010). Opinions on Punishing the Crime of Trafficking Women and Children According to Law, 关于依法惩治拐卖妇女儿童犯罪的意见（法发〔2010〕7 号，第 17 条）. Supreme People's Court website. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零九十四、一千一百零五条）, full text of the adoption book reprinted by the Yancheng Municipal People's Government. - Notes: Whether to have a child at all is not discussed here. After the child is born, this order saves the most trouble. First, give birth in hospital; for the antenatal check and the hospital delivery see section 27. The medical certificate of birth is the precondition for every formality that comes after it. Second, if you are going to hand the child over, go straight to the civil affairs department at county level or above and do not go through an intermediary. The moment the money changes hands the nature of it can change. Within a year after the birth, if you find yourself with 「thoughts of harming yourself or the child」, treat it as an emergency. For the list of things that need a hospital straight away during pregnancy and after the birth, see item 7 in section 27. The check at 42 days after the birth is also a screening for postnatal depression: see item 16 in section 27. If you have suicidal thoughts, call 12356: see item 25 in section 1. This item has no case attached: the official typical cases of trafficking your own biological child were not checked case by case in this round, so only the articles and the text of the opinion are cited. - ### 21. Do not fake an accident and do not inflate a loss to claim insurance: that is the crime of insurance fraud, and the people who testify for you, repair your car or assess it count with you - Cost: No money, no time, and you simply do not do it. @@ -201,7 +186,6 @@ This section counts two things: your personal freedom and your money. In the ite - Evidence grade: A - Sources: State Council (国务院) (2023). Regulations on the Donation and Transplantation of Human Organs, 人体器官捐献和移植条例（国务院令第 767 号，第五、十、十一、三十七条）. ; Standing Committee of the National People's Congress (全国人大常委会). Criminal Law of the People's Republic of China, consolidated text as amended by Criminal Law Amendment Eleven, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第二百三十四条之一. ; Supreme People's Procuratorate (2012-04-09). 「少年卖肾买苹果手机」案诉至法院. ; Official website of the Langfang Intermediate People's Court of Hebei, reprinting (2014-09-02). 全国最大贩肾案终审 组织者摘肾 51 枚获利千万. - Notes: The person who benefits is you. On the surface this transaction is 「a one-off 20,000 yuan」, and in reality it is trading an organ for a mobile phone. The assessment in the boy kidney case found grievous bodily harm and a grade three disability, and months of rest did not mend it. In that case the prosecution found that 5 defendants had infringed Article two hundred and thirty-four, paragraph two, of the Criminal Law, and prosecuted them for intentional injury. One of the reasons was that the medical personnel did not check before the operation whether the 「donor」 was a minor, and cut out the right kidney anyway. If the same thing happens today, the people who do the recruiting are handled under the crime of organising the sale of human organs, and taking the organ of a person under eighteen years old is still convicted and punished as intentional injury under Article two hundred and thirty-four. The operations run by the black intermediaries are done in a rented flat or a borrowed space, with no assessment beyond the tissue match and no follow-up after the operation, so if a complication arises there is nobody to answer for it. If you really want to donate to a relative, take the route in Article eleven: have it done at a medical institution with transplant qualifications, reviewed by an ethics committee, and charge nothing throughout. Donating organs from your own body after death is free of charge and is outside the scope of this item. For what losing a kidney costs the body afterwards, see item 35 in section 1 (the remaining kidney does the work of two). When the shortage of money has got you to the point of considering selling an organ, look first at the assistance channels in section 7 (unemployment insurance, temporary assistance, means-tested income support). For the traps of online loans and of 「packaging the materials」 loans, see item 6 in this section. - ### 23. Do not buy sex: the default band is 10 to 15 days of detention, not a fine and nothing more - Cost: No money, no time, and you simply do not do it. diff --git "a/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" index 28f71aa9..be8c3d53 100644 --- "a/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" +++ "b/en/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -12,7 +12,6 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A - Sources: Joel S, Eastwick PW, Finkel EJ (2017). Is romantic desire predictable? Machine learning applied to initial romantic attraction. Psychological Science, 28(10), 1478-1489. ; Eastwick PW, Finkel EJ (2008). Sex differences in mate preferences revisited: do people know what they initially desire in a romantic partner? Journal of Personality and Social Psychology, 94(2), 245-264. - Notes: The two studies only show that it cannot be worked out before you meet. They do not show that no amount of effort is of any use. Whether you get on is only known once you have met, so the time is better spent meeting more people and really seeing each one of them than putting it all on one person. The people in the studies were all university students in Europe and North America. The dating environment in China is different, so discount the figures before applying them to yourself. - ### 2. Stop once the other person has clearly refused, because going on pestering them is an administrative offence, not 「sincerity」 - Cost: No money, and no time. The hard part is holding out: no more messages, and no more waiting outside. @@ -21,7 +20,6 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025, revised). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（第五十条第一款第五项、第二款）. (reprinted by the Supreme People's Procuratorate) - Notes: There is no research behind the belief that 「pursuing somebody to the end will move them」, but the detention and the no-contact order are in black and white. The record left behind can also affect sitting the civil service examination, sitting a public institution post, political review, and the professional qualifications required for some jobs. The line between pursuing somebody and harassing them does not depend on what you think, it depends on whether the other person has clearly said no. - ### 3. To judge whether the other person is interested, look at behaviour and not at 「signals」: who starts it, who arranges the next one, who follows up - Cost: No money, and not much time. The hard part is changing the habit: stop reading the other person's expressions and tone, and start recording facts instead. @@ -30,15 +28,14 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: B - Sources: Hall JA, Xing C, Brooks S (2015). Accurately detecting flirting: error management theory, the traditional sexual script, and flirting base rate. Communication Research, 42(7), 939-958. ; Perilloux C, Kurzban R (2015). Do men overperceive women's sexual interest? Psychological Science, 26(1), 70-77. (disputed side) - Notes:Disputed. Other research holds that it is not that men overestimate women's sexual interest, but that women underreport their own intentions when filling in a questionnaire. Who is the more accurate judge is still unsettled. But both sides agree on one thing: inferring another person's thoughts backwards from small cues such as expression and tone has a low accuracy rate. Visible behaviour is far more reliable: whether the other person starts things, whether they arrange the next meeting, and whether they follow up when you do not push. - ### 4. Whether a relationship is good depends mainly on how you feel in it, not on the other person's circumstances -- Cost: No money, and not much time. The hard part is changing where you look: put the checklist away and watch how things go instead. +- Cost: No money, and not much time. The hard part is changing where you look: put the checklist for choosing a partner away and watch how things go instead. - In plain language: What says most about whether a relationship is good is your own feeling in it, not the other person's circumstances. Such as whether you feel the other person is serious and whether they appreciate you, and how often the two of you argue. Your own feeling is between two and four times as good a predictor as what your partner says. But whether the relationship gets better or worse later is something no questionnaire can really work out. - Benefit: A machine learning analysis of 29 laboratories and 43 longitudinal couples datasets, where longitudinal means following the same couples and recording them all the way through. The items that predicted the quality of the relationship best all came from how the people themselves felt about the relationship: feeling the other person's commitment, feeling appreciated, sexual satisfaction, feeling that the other person is satisfied, and conflict. The strongest of the items about the individual were life satisfaction, negative mood, depression, avoidant attachment and anxious attachment. At the start of the study these items about the relationship explained at most 45% of the variance, where variance means the difference in relationship quality between one couple and another, and 45% is the part of it these items can account for. By the end of the study they explained at most 18%. And the predictive power of 「the items people report themselves」 is between two and four times that of 「the items the partner reports」. Adding individual differences and the partner's reports on top of your own reported feelings about the relationship adds no predictive power at all - Evidence grade: A - Sources: Joel S, Eastwick PW, Allison CJ, et al. (2020). Machine learning uncovers the most robust self-report predictors of relationship quality across 43 longitudinal couples studies. Proceedings of the National Academy of Sciences, 117(32), 19061-19071. -- Notes: The same study also found that whether the relationship got better or worse during the study could not really be worked out from any set of self-report measures in the questionnaire. So for judging whether the relationship in front of you is good, asking yourself how you feel is more accurate than checking the other person's list of circumstances. Do not expect to be able to predict how it will turn out later. The item in this section called 「when it comes to emotional value, do not just ask whether it is there」 is talking about the same thing; the specific items are listed here. +- Notes: The same study also found that whether the relationship got better or worse during the study could not really be worked out from any set of self-report measures in the questionnaire. So for judging whether the relationship in front of you is good, asking yourself how you feel is more accurate than checking the other person's list of circumstances. Do not expect to be able to predict how it will turn out later. For the account that ties how a relationship is going to how your health is going, see item 17 in this section (treat how a relationship is going as a health account). ### 5. To close the distance, take turns answering a set of questions that go from light to deep, and do not learn scripts @@ -48,7 +45,6 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: B - Sources: Aron A, Melinat E, Aron EN, Vallone RD, Bator RJ (1997). The experimental generation of interpersonal closeness: a procedure and some preliminary findings. Personality and Social Psychology Bulletin, 23(4), 363-377. - Notes: The experiment was only run once, on university students, and it only measured closeness over a short period. It did not measure whether the relationship can grow into a long-term one. The closeness in the experiment came from two people taking turns and going a layer at a time deeper into their own real situation, not from any script, any routine or one side performing. So when you talk, you can 「say something true yourself, and ask one question just as deep」, taking turns on both sides. - ### 6. Long distance is not worse than living in the same city, and the real hurdle is the first three months after moving in together - Cost: A fair amount of money. You have to pay for travel, and pay for the time on the way there and back. You meet less often than people in the same city. The hard part is coping for a long time with not seeing each other. @@ -66,7 +62,6 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A - Sources: Ministry of Civil Affairs (民政部) (2025). 2024 年民政事业发展统计公报（五（二）1 婚姻登记服务、注释 5），网页版与 PDF 版. ; - Notes: The 「divorce rate over 50%」 you often see online is the number of divorces in the year divided by the number of marriages in the year, 351.3 / 610.6 ≈ 57.5%. This is called the 「divorce-to-marriage ratio」. The two figures being divided are not about the same group of people, and the bulletin has no such indicator, so it cannot be read as 「half of marriages end in divorce」. To work out 「the probability that a particular marriage ends in divorce」, you have to follow the same group of married people all the way through, and this section found no such official data. - ### 8. Count the health benefit, but discount it because the data is observational - Cost: No money, and no time. You only have to remember one thing: these figures say that married people have a lower mortality rate, which is two things happening at once, not that marriage makes you live longer. @@ -75,7 +70,6 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A (disputed) - Sources: Manzoli L, et al. (2007). Marital status and mortality in the elderly: a systematic review and meta-analysis. Soc Sci Med 64:77–94. ; Roelfs DJ, et al. (2011). The rising relative risk of mortality for singles: meta-analysis and meta-regression. Am J Epidemiol 174(4):379–389. ; Wang Y, et al. (2020). Sex differences in the association between marital status and the risk of cardiovascular, cancer, and all-cause mortality: a systematic review and meta-analysis of 7,881,040 individuals. Glob Health Res Policy 5:4. - Notes:Disputed. What the dispute is about is that married people were selected from the start: people who are healthy, earn more and have more social contact are both more likely to marry and more likely to live longer. All three meta-analyses use observational data on existing groups, that is, records that were followed but not split into a control and a treatment group. The authors themselves write about publication bias and methodological limits. Roelfs also found that the more rigorous the study, the smaller the gap that comes out. These figures are mostly about European and North American populations, and for China they are a sense of scale at best. - ### 9. Work out the time account from unpaid work, and settle the division of tasks before you register - Cost: No money. Spend an hour or two before the wedding turning the unpaid work, such as housework, looking after the old and the young, and whose job will give way, into a clear division of tasks, and then work to it. The hard part is laying it out and talking it through rather than letting it drag. @@ -84,7 +78,6 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A - Sources: National Bureau of Statistics (国家统计局) (2024). 第三次全国时间利用调查公报（第二号、第三号）. ; ; National Bureau of Statistics (国家统计局) (2019). 2018 年全国时间利用调查公报. - Notes: Both surveys only break the figures down by sex, by urban or rural area and by age, and not by whether people are married. So there is no official figure for 「how much more housework there is after marriage」, and the gender gap can only be used as an upper limit. The figures are national averages, and the two of you can quite well negotiate a different division of tasks. - ### 10. Start with the default rules in the law for the money, then decide whether to put them in writing - Cost: No money. Read the articles on property in the marriage and family book of the Civil Code. If you need to, the two of you sign a written property agreement; the law only requires it to be in writing, and there is no need to have it notarised. The hard part is saying the money things out in advance. @@ -93,7 +86,6 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A - Sources: National People's Congress (全国人民代表大会) (2020). Civil Code of the People's Republic of China, 中华人民共和国民法典（2021 年 1 月 1 日起施行，第一千零六十二、一千零六十三、一千零六十五、一千零八十八条）, National Database of Laws and Regulations. ; full text of the articles reprinted in the Supreme People's Court Gazette. - Notes: There are no official figures for bride price, for the wedding or for the marital home, so this item gives no numbers. How much the compensation under Article one thousand and eighty-eight comes to is first for the two of you to talk about; if that fails, the court decides it, and there is no fixed formula. A written property agreement works for both sides: the higher earner has a use for it, and so does the party who does a lot of the unpaid work of housework and care. - ### 11. When your parents put money towards a flat, write down at the point of the transfer whether it is a loan or a gift - Cost: No money. On the day the money goes out, write a single page saying whether it is a gift to you or a loan to you, and have both the parents who put the money in and the children sign it. If it says a loan, have both of you sign it too, and write the same sentence in the transfer reference. @@ -111,7 +103,6 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零六十四条）. ; Supreme People's Court (2020, second amendment). Provisions on Several Questions Concerning the Application of the Law in Trying Cases of Private Lending, 关于审理民间借贷案件适用法律若干问题的规定（第十五、十六条）. (Supreme People's Court Gazette) - Notes: When property is divided at a divorce and the other side's parents suddenly produce a loan agreement signed only by your spouse, there are two steps. The first is to ask whether the loan actually happened. If you say the money was never handed over and can make your reasons hold together, the court will weigh several things together. First the size of the loan, how the money was handed over, and the two parties' financial positions, and then how you normally do transactions, whether there is a matching change in the property, and what the witnesses say. Where the other side can produce nothing but a transfer receipt, it is still for them to prove that the loan exists. The second step, once the loan exists, is to ask whether it is a joint debt. For the general rules on IOUs and guarantees, see item 18 in section 8 - ### 13. To register you only need your identity card and a signed statement; you do not need the household register, you pay nothing, and the certificate is issued on the spot - Cost: No money. Both of you have to be there in person. Once it is done it usually takes a few tens of minutes. @@ -120,7 +111,6 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A - Sources: State Council (国务院) (2025). Regulations on Marriage Registration, 婚姻登记条例（国令第 804 号，第二、六、八、十条）. - Notes: There are only four situations in which the registration authority will not do it: the legal age for marriage has not been reached, the two of you are not both entirely willing, one or both of you already have a spouse, or the two of you are direct blood relatives or collateral blood relatives within three generations. The only things that could cost money are side services such as photographs and photocopies. The registration itself costs nothing, so if you are asked for a 「工本费」 or a 「rush fee」, refuse and complain. For the 30-day cooling-off period for divorce registration, see the item on the cost of getting out in this section - ### 14. Have a premarital medical check once before the wedding, and if there is a serious illness you must tell the other party honestly before you register - Cost: A little money. In most areas it is free or reduced for people of the right age. Paying yourself costs only a few hundred yuan. It takes half a day. @@ -129,17 +119,7 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: B - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零五十三条）. ; State Council (国务院) (2025). Regulations on Marriage Registration, 婚姻登记条例（国令第 804 号，第十二条）. ; General Office of the National Health Commission (国家卫生健康委办公厅) (2023). 出生缺陷防治能力提升计划（2023—2027 年）（国卫办妇幼发〔2023〕9 号）. ; National Health Commission and four other departments (国家卫生健康委等五部门) (2020). Notice on Strengthening Premarital Health Care, 关于加强婚前保健工作的通知（国卫妇幼函〔2020〕205 号）：「积极倡导各地结合实际推行免费婚前检查」. - Notes: The medical check has not been a step you have to take before the certificate since 2003, and whether you go is voluntary. But the rule on「a serious illness not disclosed honestly」 does give the other party the right to go to court to have the marriage annulled, and the application has to be made within one year of finding out. There is no national rule making the check free. The 2020 notice of the National Health Commission and four other departments only says「积极倡导各地结合实际推行免费婚前检查」. So whether it is free and which items are covered are set locally; ask at the local maternal and child health centre, or at the one-stop window at the marriage registration office. - -### 15. When it comes to emotional value, do not just ask whether it is there; look at the quality of the relationship - -- Cost: No money. Before the wedding, use a few conflicts and a few months of living together to judge the quality of the relationship, and do not only check a list of circumstances. The hard part is holding out against judging from the list. -- In plain language: The better the quality of a marriage, the healthier the people in it, but the relationship is weak, on a par with the effect of habits like diet on health. To judge whether a relationship is good, use a few conflicts and a few months of living together, and do not only check a list of circumstances. Whether a bad marriage is worse for you than not marrying at all has not yet been compared directly with data. -- Benefit: A meta-analysis of 126 studies and more than 72,000 people shows that the better the quality of the marriage, the better the health. The effect size r lies between 0.07 and 0.21. The r is how far two things change together, where 0 is no relationship at all and 1 is changing in perfect step. The mortality risk is r = 0.11, and the cardiovascular reaction during conflict is r = -0.13. This is the same order of magnitude as the effect of health behaviours such as diet. In the studies that specifically tested for a difference between the sexes, essentially no difference was found between men and women. The review in the cardiovascular field states that marital dissatisfaction and marital quality have a significant effect on cardiovascular risk -- Evidence grade: B (disputed) -- Sources: Robles TF, et al. (2014). Marital quality and health: a meta-analytic review. Psychol Bull 140(1):140–187. ; Dhindsa DS, et al. (2020). Marital status and outcomes in patients with cardiovascular disease. Trends Cardiovasc Med 30(4):215–220. -- Notes:Disputed. What the dispute is about is that the association itself is weak. Some of the results show publication bias. And most of the studies surveyed at a single point in time, so they cannot show who led to what; Robles lists these limits in the original text. This section found no meta-analysis that compares the mortality rate of「a marriage of low quality」directly with that of「never married」, so「a bad marriage is worse than not marrying」is for now only an inference, not a conclusion from figures. - -### 16. Work out the cost of getting out: a divorce by agreement has a 30-day cooling-off period, and a divorce by litigation has statutory conditions +### 15. Work out the cost of getting out: a divorce by agreement has a 30-day cooling-off period, and a divorce by litigation has statutory conditions - Cost: No money. Before the wedding, spend some time reading: if you do have to get out, how long it takes and what procedure it follows. The hard part is thinking about splitting up before you have married. - In plain language: A divorce by agreement is not finished once it is signed. Within 30 days of submitting the application, either party may withdraw it. At the end of that period you also have to collect the certificate in person within 30 days, and not doing so counts as withdrawing. If the other side does not agree, you can only sue. The court tries mediation first, and where there is domestic violence, bigamy, or two years of living apart, and mediation fails, it grants the divorce. @@ -147,26 +127,27 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A - Sources: National People's Congress (全国人民代表大会) (2020). Civil Code of the People's Republic of China, 中华人民共和国民法典（第一千零七十六、一千零七十七、第一千零七十九条）. ; Ministry of Civil Affairs (民政部) (2020). Notice on Implementing the Provisions on Marriage Registration in the Civil Code of the People's Republic of China, 关于贯彻落实《中华人民共和国民法典》中有关婚姻登记规定的通知（民发〔2020〕116 号）. ; Ministry of Civil Affairs (民政部) (2025). 2024 年民政事业发展统计公报. Citation the same as item 7 in this section (what the marriage rate and the divorce rate each measure) - Notes: The cooling-off period applies only to a divorce by agreement. There is no cooling-off period for a divorce by litigation, but that takes longer. Court fees and legal fees differ from place to place, so this item gives no figures. A high cost of getting out is not a reason not to marry, but you should know it before you do. - -### 17. Keep a separate account for 「marrying because of the older generation」, and do not mix it with your own +### 16. Keep a separate account for 「marrying because of the older generation」, and do not mix it with your own - Cost: No money. Draw two columns on a piece of paper. One column is which benefits in this marriage are your own, which the earlier items in this section can work out. The other is which benefits belong to the older generation: peace of mind, face, and the pressure to marry stopping. The hard part is keeping the older generation's column separate and not merging it into your own. - In plain language: The benefits that go to the older generation, that is peace of mind, face and no more being nagged, are real. But they do not automatically become your health benefit or your relationship quality. The time account, the money account and the cost of getting out, on the other hand, are all borne by you. Draw two columns on paper and write out which benefits are yours and which belong to the older generation, and whether the deal is worth it is obvious at a glance. -- Benefit: The benefit to the older generation is real. But it does not automatically turn into your health benefit (item 8) or into relationship quality (item 15). And the time account (item 9), the money account (items 10 to 12) and the cost of getting out (item 16) are all borne by the parties themselves. Once the books are kept separately, whether「marrying to complete a task」is good value is something you can see for yourself +- Benefit: The benefit to the older generation is real. But it does not automatically turn into your health benefit (item 8) or into relationship quality (item 4, whether a relationship is good depends on how you feel in it). And the time account (item 9), the money account (items 10 to 12) and the cost of getting out (item 15) are all borne by the parties themselves. Once the books are kept separately, whether 「marrying to complete a task」 is good value is something you can see for yourself - Evidence grade: C -- Sources: The author's own experience, a methodological suggestion; for the figures see the earlier items in this section +- Sources: the author's own experience, a methodological suggestion; for the figures see the earlier items in this section - Notes: The older generation's column can also be counted as a heavy one, and how you weigh it is your own decision. For the list to fill in, see [docs/结婚划不划算.md](../docs/结婚划不划算.md) -### 18. Treat how a relationship is going as a health account: in an argument, argue about the thing and not the person, and do not insult or sneer +### 17. Treat how a relationship is going as a health account: in an argument, argue about the thing and not the person, and do not insult or sneer - Cost: No money. In an argument, stick to the thing in front of you, do not insult, do not sneer and do not bring up old grievances. The hard part is keeping your mouth shut while you are angry. - In plain language: Couples with a good relationship are somewhat healthier and have a somewhat lower risk of death. The association is small, on a par with the association between diet and health. In one experiment, couples who stayed hostile throughout their fights healed the small wounds the experiment produced at only six tenths the rate of couples who stayed calm. -- Benefit: Pooling 126 studies and more than 72,000 people, which is what a meta-analysis does. The better the quality of the marriage, the better the health, with a correlation coefficient r between .07 and .21. A correlation coefficient of 0 is no relationship at all and 1 is changing in perfect step. The correlation with a lower risk of death is r = .11. A smaller cardiovascular reaction during conflict is r = -.13. The authors say that all these effects are small, on a par with the association between health habits such as diet and health. In another crossover experiment, 42 healthy couples each stayed in hospital twice, for 24 hours each time. Once they had a mutually supportive conversation, and once they discussed a subject on which they disagreed. After the argument, the blister wounds the experiment produced healed more slowly. For the couples who showed more hostility on both occasions, the wounds healed at only 60% the rate of the couples with low hostility. On the morning of the day after the argument, their blood markers of inflammation (IL-6, tumour necrosis factor) had also risen more +- Benefit: Pooling 126 studies and more than 72,000 people, which is what a meta-analysis does. The better the quality of the marriage, the better the health, with a correlation coefficient r between .07 and .21. A correlation coefficient of 0 is no relationship at all and 1 is changing in perfect step. The correlation with a lower risk of death is r = .11. A smaller cardiovascular reaction during conflict is r = -.13. The authors say that all these effects are small, on a par with the association between health habits such as diet and health. In another crossover experiment, 42 healthy couples each stayed in hospital twice, for 24 hours each time. Once they had a mutually supportive conversation, and once they discussed a subject on which they disagreed. After the argument, the blister wounds the experiment produced healed more slowly. For the couples who showed more hostility on both occasions, the wounds healed at only 60% the rate of the couples with low hostility. On the morning of the day after the argument, their blood markers of inflammation (IL-6, tumour necrosis factor) had also risen more. - Evidence grade: B - Sources: Robles TF, Slatcher RB, Trombello JM, McGinn MM (2014). Marital quality and health: a meta-analytic review. Psychological Bulletin. ; Kiecolt-Glaser JK et al. (2005). Hostile marital interactions, proinflammatory cytokine production, and wound healing. Archives of General Psychiatry. -- Notes: The reason for grade B is that most of the pooled studies are only records that were followed, with no control group. So it cannot be told whether a bad relationship makes you ill or poor health makes the relationship worse. In some studies the association is still there after age and income are taken out. The wound experiment has only 42 couples in it. 「Argue about the thing and not the person, do not insult and do not sneer」 is a practice the author worked out from the 「hostile behaviour」 in the experiment, and no study compares it directly. Where the arguments are fierce and cannot be changed, the two of you can go together to marriage and family counselling, and that too has no separate literature. Where it has come to hitting, or to long-term verbal abuse and threats, that is no longer a matter of communication but of domestic violence: see item 43 in section 8 (if you are being abused at home). For deciding whether to go through with it, see the cost of getting out in item 16 in this section. +- Notes: The reason for grade B is that most of the pooled studies are only records that were followed, with no control group. So it cannot be told whether a bad relationship makes you ill or poor health makes the relationship worse. In some studies the association is still there after age and income are taken out. Robles also states in the original paper that some of the results are subject to publication bias. The studies that compared men and women explicitly found basically no difference between the sexes. This section found no pooled analysis directly comparing the mortality of a 「marriage with a bad relationship」 with not marrying at all, so 「a bad marriage does more harm than not marrying」 can only be an inference. The wound experiment has only 42 couples in it. 「Argue about the thing and not the person, do not insult and do not sneer」 is a practice the author worked out from the 「hostile behaviour」 in the experiment, and no study compares it directly. Where the arguments are fierce and cannot be changed, the two of you can go together to couple therapy, see item 20 in this section (couple therapy). Where it has come to hitting, or to long-term verbal abuse and threats, that is no longer a matter of communication but of domestic violence: see item 43 in section 8 (if you are being abused at home). For deciding whether to go through with it, see item 15 in this section (the cost of getting out). -### 19. Same-sex partners: while you are both clear-headed, get the power of attorney, the voluntary guardianship and the will done. In law you are not close relatives, and without these documents you have no right to sign and no right to inherit + + +### 18. Same-sex partners: while you are both clear-headed, get the power of attorney, the voluntary guardianship and the will done. In law you are not close relatives, and without these documents you have no right to sign and no right to inherit - Cost: A little money. Voluntary guardianship and a will are done at a notary office, and the notary fees differ from place to place, starting at a few hundred yuan. A will you write by hand costs nothing. It takes one or two half-days from start to finish. The hard part is the two of you having to sit down and talk about illness and death. - In plain language: Close relatives in law are a spouse, parents, children and siblings, and a partner is not among them. Without the documents in place, when one of you falls seriously ill the other cannot decide for them. When one of you dies, the other inherits not one yuan. The three documents each cover a stretch: the power of attorney for while you are alive, voluntary guardianship once the capacity to make decisions is lost, and the will for after your death. @@ -174,12 +155,20 @@ This section breaks 「whether marriage is worth it」 into separate accounts fo - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第二十三、三十三、一百六十五、一百七十三、一千零四十五、一千一百二十四、一千一百二十七、一千一百三十三至一千一百四十、一千一百四十二、一千二百一十九条）. (National Database of Laws and Regulations) ; Yichang Municipal Bureau of Justice (宜昌市司法局) (2021). 意定监护公证：通向彩虹的公证之路. 湖北省司法厅网站「实践荟萃」栏目. - Notes: The status a partner has in an inheritance is 「legatee under a will」, not heir. Once you know of the bequest, you have to express your acceptance within 60 days, and not expressing it counts as declining. The partner themselves, and anybody who stands to benefit from the will, cannot be a witness to it. Where several wills have been made, the last one governs: see item 2 in section 17 (a later will overturns an earlier one). Whether a voluntary guardian may sign the consent form for an operation differs from hospital to hospital, so ask the hospital you normally use in advance. If the family does not accept these documents, it comes to a court case: six months or more in the ordinary procedure at first instance, and legal fees are yours to pay. So get the documents done at a notary office where you can. For a flat bought together and money saved together, write a separate agreement on the basis of what each of you actually put in. For how voluntary guardianship works, see item 1 in section 17 (naming a future guardian in writing). The people who benefit are you and your partner. - -### 20. Do not marry while concealing your sexual orientation, and do not treat a marriage in name only as the easy way out: in law it is a real marriage, and the only way out is a divorce +### 19. Do not marry while concealing your sexual orientation, and do not treat a marriage in name only as the easy way out: in law it is a real marriage, and the only way out is a divorce - Cost: No money. The hard part is standing up to the pressure from the family to marry, which may take years to bear. - In plain language: Marrying while concealing your sexual orientation is not void in law and cannot be annulled, so the only way out is a divorce. A marriage in name only is a real marriage as well, and the money earned and the debts incurred after it count as the couple's. The side that was kept in the dark can, on divorce, ask for a larger share of the property and for compensation for mental harm, but whether the court agrees is not settled. -- Benefit: Article one thousand and fifty-one of the Civil Code says a marriage is void in only three situations: bigamy, a relationship of close relation that is forbidden for marriage, and not having reached the legal marriageable age. Articles one thousand and fifty-two and one thousand and fifty-three say only two situations can be annulled: marrying under coercion, and concealing a serious illness before the wedding. Article seventeen of Interpretation (One) on the Marriage and Family Book of the Civil Code provides that where a request is made to confirm a marriage void on any ground other than those three, the court rules to reject it. So the only way out is a divorce. A marriage in name only is a lawful marriage once registered. Article one thousand and sixty-two provides that wages, bonuses and remuneration for labour after the wedding are joint marital property. Article one thousand and sixty-four provides that a debt signed for jointly by the two, and a debt one of them took on for the everyday needs of the household, are joint marital debts. For damages on divorce, see Article one thousand and ninety-one, which is limited to bigamy, cohabiting with another person, domestic violence, abuse or abandonment of a family member, and 「having some other serious fault」. Article two of Interpretation (One) limits 「cohabiting with another person」 to living with an 「opposite-sex partner outside marriage」. So a same-sex relationship can only be argued as 「some other serious fault」. The compensation includes payment for mental harm, which the side kept in the dark has to claim at the same time as suing for the divorce; if the divorce is refused, it is not supported. Since the Civil Code came into force, no judgment of this kind published by a court has been found. Under the old law, an article posted on the website of the Jiangsu High People's Court in 2013 discussed a case in which a wife sued for divorce because her husband had concealed that he was homosexual. The article sets out two views. One holds that homosexuality was not among the faults recognised by the law at the time, so neither a larger share of the property nor mental compensation should be supported. The author prefers the other: treat the side kept in the dark as the 「party without fault」, take that into account when dividing the property, and also pay compensation for mental distress (nationwide) +- Benefit: Article one thousand and fifty-one of the Civil Code says a marriage is void in only three situations: bigamy, a relationship of close relation that is forbidden for marriage, and not having reached the legal marriageable age. Articles one thousand and fifty-two and one thousand and fifty-three say only two situations can be annulled: marrying under coercion, and concealing a serious illness before the wedding. Article seventeen of Interpretation (One) on the Marriage and Family Book of the Civil Code provides that where a request is made to confirm a marriage void on any ground other than those three, the court rules to reject it. So the only way out is a divorce. A marriage in name only is a lawful marriage once registered. Article one thousand and sixty-two provides that wages, bonuses and remuneration for labour after the wedding are joint marital property. Article one thousand and sixty-four provides that a debt signed for jointly by the two, and a debt one of them took on for the everyday needs of the household, are joint marital debts. For damages on divorce, see Article one thousand and ninety-one, which is limited to bigamy, cohabiting with another person, domestic violence, abuse or abandonment of a family member, and 「having some other serious fault」. Article two of Interpretation (One) limits 「cohabiting with another person」 to living with an 「opposite-sex partner outside marriage」. So a same-sex relationship can only be argued as 「some other serious fault」. The compensation includes payment for mental harm, which the side kept in the dark has to claim at the same time as suing for the divorce; if the divorce is refused, it is not supported. Since the Civil Code came into force, no judgment of this kind published by a court has been found. Under the old law, an article posted on the website of the Jiangsu High People's Court in 2013 discussed a case in which a wife sued for divorce because her husband had concealed that he was homosexual. The article set out two views. One held that homosexuality was not among the faults defined by the law at the time, so neither a larger share of the property nor mental compensation should be supported. The author preferred the other: treat the side kept in the dark as the 「party without fault」, give her extra consideration when the property is divided, and pay her mental consolation money as well (nationwide) - Evidence grade: A -- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零五十一至一千零五十四、第一千零六十二、第一千零六十四、第一千零七十九、第一千零八十七、第一千零九十一条）. (National Database of Laws and Regulations) ; Supreme People's Court (2020). Interpretation (One) on the Application of the Marriage and Family Book of the Civil Code, 关于适用《中华人民共和国民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第二、十七、八十六至九十条）. ; Zhu Laikuan, Wang Hongjuan (2013). 因同性恋导致离婚能否主张损害赔偿？江苏法院网. -- Notes: The larger share of the property and the mental compensation rest on a single analytical article that does not say how the case finally came out, so whether you can get it depends on the particular court. The Civil Code has one more ground than the old law, 「some other serious fault」, which gives the side kept in the dark something to argue with. The text of the law itself is grade A. Where both parties in a marriage in name only have same-sex partners, the partners are not protected by the marriage law between them, and the property has to be set out separately: see item 19 in this section (same-sex partners: voluntary guardianship and a will). A divorce from a marriage in name only also has to go through the 30-day cooling-off period or a court case: see item 16 in this section (a divorce by agreement has a 30-day cooling-off period). Where the agreement says that compensation is given up, a later suit will not be supported. The person who benefits first is the spouse who was kept in the dark, and second is yourself: once the concealment is found out, the divorce, the division of property and the custody of any children all arrive together. +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零五十一至一千零五十四、一千零六十二、一千零六十四、一千零七十九、一千零八十七、一千零九十一条）. (National Database of Laws and Regulations) ; Supreme People's Court (2020). Interpretation (One) on the Application of the Marriage and Family Book of the Civil Code, 关于适用《中华人民共和国民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第二、十七、八十六至九十条）. ; Zhu Laikuan, Wang Hongjuan (2013). 因同性恋导致离婚能否主张损害赔偿？江苏法院网. +- Notes: The larger share of the property and the mental compensation rest on a single analytical article that does not say how the case finally came out, so whether you can get it depends on the particular court. The Civil Code has one more ground than the old law, 「some other serious fault」, which gives the side kept in the dark something to argue with. The text of the law itself is grade A. Where both parties in a marriage in name only have same-sex partners, the partners are not protected by the marriage law between them, and the property has to be set out separately: see item 18 in this section (same-sex partners: voluntary guardianship and a will). A divorce from a marriage in name only also has to go through the 30-day cooling-off period or a court case: see item 15 in this section (a divorce by agreement has a 30-day cooling-off period). Where the agreement says that compensation is given up, a later suit will not be supported. The person who benefits first is the spouse who was kept in the dark, and second is yourself: once the concealment is found out, the divorce, the division of property and the custody of any children all arrive together. + +### 20. When a relationship is stuck in repeated arguments and cold shoulders, the two of you go to couple therapy together: waiting on your own will not get better on its own + +- Cost: It costs money. Each session is charged separately, the prices differ from one hospital to another, and this item gives no figures. In the studies it is generally once a week, and one trial ran for about 8 months from start to finish. The hard part is that both of you have to go, and that you have to keep going to the end. +- In plain language: Couples stuck in a bad patch who finish couple therapy together are markedly more satisfied with the relationship, and about seven in ten improve. Couples who only sit on the waiting list change very little. But the effect fades: followed to the fifth year, about half are still clearly better and about a quarter have already separated or divorced. +- Benefit: A meta-analysis in 2020 pooled 58 studies and 2092 couples. After couple therapy, the change in relationship satisfaction was g = 1.12 (credible range 0.92–1.31), which the authors call a large effect. g is an effect size, see the glossary in the README. For the couples allocated to the waiting list, over the same period g = 0.12 (−0.04 to 0.29), with no clear improvement. The worse the relationship was at the start, the more it improved. A review in 2012 puts it at 70% of the couples who took the treatment improving. Another randomised trial took 134 couples with long-standing and serious discord and gave them about 8 months of one of two behavioural couple therapies. At the end the change in satisfaction was d = 0.90 and 0.71. By year 5, 50.0% and 45.9% of the couples respectively were still clearly better. What the authors mean by clearly better, they call 「an improvement of clinical significance」. At the same time 25.7% and 27.9% had already separated or divorced +- Evidence grade: A (disputed) +- Sources: Roddy MK, Walsh LM, Rothman K, Hatch SG, Doss BD (2020). Meta-analysis of couple therapy: effects across outcomes, designs, timeframes, and other moderators. Journal of Consulting and Clinical Psychology, 88(7), 583-596. ; Lebow JL, Chambers AL, Christensen A, Johnson SM (2012). Research on the treatment of couple distress. Journal of Marital and Family Therapy, 38(1), 145-168. ; Christensen A, Atkins DC, Baucom B, Yi J (2010). Marital status and satisfaction five years following a randomized clinical trial comparing traditional versus integrative behavioral couple therapy. Journal of Consulting and Clinical Psychology, 78(2), 225-235. ; Rathgeber M, Bürkner PC, Schiller EM, Holling H (2019). The efficacy of emotionally focused couples therapy and behavioral couples therapy: a meta-analysis. Journal of Marital and Family Therapy, 45(3), 447-463. (the disputed side) ; Shadish WR, Baldwin SA (2003). Meta-analysis of MFT interventions. Journal of Marital and Family Therapy, 29(4), 547-570. (the disputed side) ; Karakurt G, Whiting K, van Esch C, Bolen SD, Calabrese JR (2016). Couples therapy for intimate partner violence: a systematic review and meta-analysis. Journal of Marital and Family Therapy, 42(4), 567-583. ; Standing Committee of the National People's Congress (全国人大常委会) (2018 revision). Mental Health Law, 精神卫生法（第二十三、五十一条）. (National Database of Laws and Regulations) +- Notes: Disputed. The dispute is over how long the effect lasts. In another meta-analysis that included only randomised trials, at the end g = 0.60 and after six months 0.44. At 12 months only the behavioural couple therapy still has data, g = 0.06, and almost nothing has been kept. Another review pooling 20 meta-analyses says an effect of clinical significance appears in 40% to 50% of people. 「Seven in ten improve」 and 「four to five in ten clearly improve」 use different criteria, so do not compare them. The five-year trial had no untreated control group, so it cannot be said whether divorces went up or down. The 2020 synthesis covered only opposite-sex couples, and this item found no randomised trial in China either, so discount it before you apply it to yourself. Where to go in China: the Mental Health Law requires psychotherapy to be done inside a medical institution, and a psychological counsellor may not carry out psychotherapy. You can ask the psychiatry department or the psychology department of a hospital whether they do couples therapy. When you look for an organisation outside a hospital, first ask what qualifications the people have and what training they have had. If there is domestic violence, do not go together first. The 2016 meta-analysis states that its conclusion applies only to mild to moderate violence between the two of you arising out of conflict. Violence characterised by controlling the other person or by serious harm is not covered. People in the field also worry about revenge after the sessions. If you have been hit, see item 43 in section 8 (if you are being abused at home) first. For how to talk during an argument see item 17 in this section (argue about the thing and not the person), and for deciding whether to go through with it see item 15 in this section (the cost of getting out). The people who benefit are you and your spouse. diff --git "a/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" "b/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" index fee69314..32a7a9d2 100644 --- "a/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" +++ "b/en/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" @@ -12,7 +12,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: C - Sources: National People's Congress (全国人大) (2018). Criminal Procedure Law, 刑事诉讼法（2018 年修正，第三十四、三十九、六十七、七十九、九十一、一百一十、一百一十二条）. - Notes: Those 37 days are counted from detention to the procuratorate's decision on whether to approve the arrest; they are not a limit on how long you may be held. After the arrest is approved there is still the custody period during the investigation, and the counting continues from there. Those 37 days are also when a lawyer who knows the field is most useful: getting to see you, handing the procuratorate a written view that the arrest should not be approved, and applying for release on bail. Looking for a lawyer after the prosecution has begun is too late. Every other item in this section is about the same thing: whether the other side has grounds to report you. - ### 2. Do not make or sell game cheats and cheats, including AI assistance that only reads the screen and never touches memory - Cost: No money. The price is giving up this kind of contract work and the income from 「small tools」. The hard part is that the money comes in quickly and it is not obvious who would come after you. @@ -21,7 +20,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第三款、第二百一十七条第六项）. ; Supreme People's Court and Supreme People's Procuratorate (2011). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases Endangering the Security of Computer Information Systems, 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第三条）. (reprinted by the Shenzhen Municipal Public Security Bureau) ; Supreme People's Procuratorate (2026). 上海静安：严惩制售游戏外挂黑灰产共护清朗网络空间. ; Supreme People's Procuratorate (2021). 2020 年度检察机关保护知识产权典型案例（案例五）. - Notes: The cheat cases that have been published are all the kind that read and write game memory and get round the anti-cheat. No official case has yet been seen for assistance that 「only takes screenshots and only simulates key presses」. But the ticket-grabbing case in item 3 shows that not touching memory and only sending requests can still amount to the crime of illegally obtaining data. Whether it is a crime in the end, and under which charge, is settled by the case-handling organ on the basis of the material handed in when the vendor reports it, and not by the author. Selling to 20 users is enough for 「serious circumstances」, and that is a small number. - ### 3. Do not write or sell ticket-grabbing, flash-sale, order-brushing or coupon-collecting scripts, not even one that only 「clicks the button automatically」 - Cost: No money, and no time either. @@ -30,7 +28,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ; Supreme People's Court and Supreme People's Procuratorate (2011). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases Endangering the Security of Computer Information Systems, 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ; Supreme People's Procuratorate (2026). 上海静安：查办利用外挂软件批量代抢火车票牟利案. ; Supreme People's Court (2025). 2024 年人民法院知识产权典型案例（案例七，抢票软件不正当竞争案）. - Notes: The script in the ticket-grabbing case does not change memory and does not crack anything; it is only 「faster than a person」, and it is still convicted under article 285 of the Criminal Law. Both taking money to grab tickets for others and selling the script count as making a profit. For the case where you use it once yourself, no official case was found. - ### 4. Let a crawler crawl only public pages that need no login, do not get round the anti-crawler measures, do not touch personal information, and do not sell what you crawl - Cost: No money. A crawler is a program that turns web pages automatically and pulls the data off them. Before you crawl, read the robots file and the interface authorisation note; robots is the file a site puts in its own root directory saying which pages may not be crawled. Do not crawl data that needs a login, needs a token, which is a temporary access credential, or that can only be got by cracking a signature. The hard part is that technically this data is there for the taking once you start, and holding out depends on nothing but your own restraint. @@ -39,7 +36,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百五十三条之一）. ; Supreme People's Court and Supreme People's Procuratorate (2017). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases of Infringing Citizens' Personal Information, 关于办理侵犯公民个人信息刑事案件适用法律若干问题的解释（第三、四、五条）. ; Supreme People's Procuratorate (2022). Typical Cases of Procuratorates Punishing the Crime of Infringing Citizens' Personal Information According to Law (Case Two), 检察机关依法惩治侵犯公民个人信息犯罪典型案例（案例二）. ; Supreme People's Procuratorate (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例一）. - Notes: The 2017 judicial interpretation provides that 「handing information that was lawfully collected on to somebody without that person's consent」 also counts as 「providing」. Exporting a table of users to an outside partner at work falls under that same sentence. 5000 records is only the size of a small spreadsheet. - ### 5. Do not take development contracts for gambling, fraud or pornography apps, and do not set up payment channels or unblock accounts for a 「client」 - Cost: No money. The price is turning down that kind of outsourcing work, where the quote is high and the requirement is vague. The hard part is that what the other side pays is often well above the going rate. @@ -57,7 +53,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百八十六条）. ; Supreme People's Procuratorate (2024). 检察机关「依法惩治网络犯罪 助力网络空间综合治理」典型案例（案例六）. ; Supreme People's Procuratorate (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例七）. - Notes: The account and password were given to you by the company, which does not mean you can still use them after you leave. Logging in with the original account is already 「illegal control」. The loss is worked out as the cost of restoring the data, and a few tens of thousands of yuan is enough to open a case. - ### 7. Do not take source code, customer lists or technical documents away when you leave, do not upload them to your own cloud drive, and do not reuse them at your next employer - Cost: No money. Before you leave, delete the company material from your own computer and cloud drive. The hard part is that some of those files you wrote line by line yourself and you are reluctant to let them go. @@ -66,7 +61,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百一十九条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Anti-Unfair Competition Law, 反不正当竞争法（2025 年修订，第十条）. ; Supreme People's Procuratorate (2025). 检察机关知识产权保护典型案例（案例七）. ; Supreme People's Procuratorate (2020). 2019 年度检察机关保护知识产权典型案例（案例十六）. ; Supreme People's Procuratorate (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例五）. - Notes: 「Code you wrote yourself」 is owned by the company as far as copyright goes; see item 13 in this section (code written with company resources belongs to the company). Taking it is taking the company's property. A customer list is business information and is likewise a trade secret. On top of any prison sentence, the company can bring a separate civil claim. - ### 8. Do not run your own programs on somebody else's computer, server or camera, and do not mine cryptocurrency on company machines - Cost: No money, and no time either. @@ -75,7 +69,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ; Supreme People's Court and Supreme People's Procuratorate (2011). Interpretation on Several Questions Concerning the Application of the Law in Handling Criminal Cases Endangering the Security of Computer Information Systems, 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十九、六十六条；2016 年文本为第二十七、六十三条，终身禁业为修正后的第六十六条第三款）. ; Supreme People's Procuratorate (2022). 山东滨州开发区：办理一起非法控制计算机信息系统案. ; Supreme People's Procuratorate (2021). 充分发挥检察职能 推进网络空间治理典型案例（案例六）. - Notes: No case of mining cryptocurrency on a company server published by the Supreme People's Procuratorate or the Supreme People's Court was found, so this item is written from the provision on 「controlling somebody else's system without permission」. 「Being barred for life from key posts」 is an extra consequence that the Cybersecurity Law attaches on top of a fine or a sentence, and for a technical worker it is heavier than a fine. - ### 9. Do not test somebody else's system without written authorisation, and 「out of good intentions」 and 「reporting it afterwards」 are not reasons for acquittal - Cost: No money. Before joining a vendor's SRC or a public bug bounty, get a document that sets out the scope of the authorisation, the list of targets and the period you may test for. An SRC is the security emergency response centre the vendor runs itself, which takes reports of vulnerabilities from outside. Once you have confirmed that the vulnerability exists, stop. The hard part is giving up having 「a quick try while you are at it」. @@ -93,7 +86,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: A - Sources: Ministry of Industry and Information Technology, Cyberspace Administration of China and Ministry of Public Security (工业和信息化部、国家互联网信息办公室、公安部) (2021). Provisions on the Administration of Security Vulnerabilities in Network Products, 网络产品安全漏洞管理规定（工信部联网安〔2021〕66 号，第二、四、九、十、十四条，2021 年 9 月 1 日施行）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十八、六十五条；2016 年文本为第二十六、六十二条，规定第十四条援引的是 2016 年文本的条号）. - Notes: The provisions put individuals squarely among the parties they cover, and 「I am only an amateur」 is not one of the exceptions. Submitting a vulnerability in a domestic system to an overseas bug bounty platform falls squarely under 「an unpublished vulnerability may not be given to anybody overseas」. Whether you are entitled to test at all is one question, and see item 9 in this section (do not test somebody else's system without written authorisation); what to do with a vulnerability once you have it is a separate question. A vulnerability found lawfully can still be punished for the way it is published, and a vulnerability found without authorisation is a problem both in the testing and in the publishing. Of the reporting channels, the Ministry of Industry and Information Technology platform and CNCERT are the most used. The provisions do not make a separate category for 「the vendor has been notified and still does not fix it」; in practice you keep a record of the negotiation and the reporting, and report to an official platform as well. - ### 11. Do not sell circumvention tools or VPN accounts, and do not set up this kind of node for other people - Cost: No money, and no time either. @@ -102,7 +94,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: B - Sources: State Council (国务院) (1997). Interim Provisions on the Administration of International Networking of Computer Information Networks, 计算机信息网络国际联网管理暂行规定（第六、十四条）. (reprinted by the Chongqing Municipal Rural and Agricultural Committee) ; National People's Congress (全国人大) (2020). Criminal Law, as amended by Criminal Law Amendment Eleven, 刑法（根据刑法修正案（十一）修正，第二百二十五条第四项）. - Notes: While this section was being written, no published case involving circumvention tools was found on the websites of the Supreme People's Court or the Supreme People's Procuratorate, and the way places handle it varies a great deal, so the grade is B. For your own use, the heaviest penalty is the warning plus a fine written in Article fourteen of the interim provisions. - ### 12. If you have signed a non-compete, send a written demand when the company does not pay the monthly compensation after you leave, and you can terminate it once 3 months have gone by; and for a post that never touched trade secrets you can ask the court to rule the clause has no effect - Cost: No money. When you leave, keep the original of the non-compete and the evidence that the company has not paid you month by month. @@ -111,7 +102,6 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2012). Labour Contract Law, 劳动合同法（2012 年修正，第二十三、二十四条）. (reprinted by the Fushun Municipal Emergency Management Bureau) ; Supreme People's Court (2020). Interpretation (One) on Several Questions Concerning the Application of the Law in Trying Labour Dispute Cases, 关于审理劳动争议案件适用法律问题的解释（一）（第三十六至四十条）. ; Supreme People's Court (2025). Interpretation (Two) on Several Questions Concerning the Application of the Law in Trying Labour Dispute Cases, 关于审理劳动争议案件适用法律问题的解释（二）（法释〔2025〕12 号，第十三至十五条）. ; Supreme People's Court (2025). 劳动争议典型案例（案例四）. - Notes: The other way round, where the agreement is valid, the company has paid month by month, and you still go to work for a competitor, you have to pay the liquidated damages as agreed. Having paid, you may also be required to serve out the rest of the non-compete period. A non-compete agreed during employment counts as well, even where the company pays nothing extra for it. - ### 13. Code written in working hours or with company resources belongs to the company; make your own open-source project with your own time and your own equipment, and do not mix in company code - Cost: No money. Read the intellectual property clause in the employment contract when you join. The hard part is that personal work and the company's work often share one computer and have to be kept apart. @@ -120,15 +110,14 @@ The measures in this section are personal freedom and money, and what you get ba - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Copyright Law, 著作权法（2020 年修正，第十八条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Patent Law, 专利法（2020 年修正，第六条）. - Notes: Software written straight into the category of 「copyright belongs to the unit」 almost never falls, for a programmer, into the category of 「copyright belongs to the author, and the unit has priority use for two years」. 「Written after work」 still counts as being made mainly with the equipment and materials of the unit as soon as a company computer, intranet or company business data is used. To keep a personal project clean, use your own machine, do not touch the company repository, and do not overlap with the company's business. - -### 14. If you build a product on open-source code such as GPL, open your source as the licence requires or change the licence; do not assume 「nobody is in charge」 +### 14. When you build a product on GPL and other open-source code, open the source as the licence requires or change the licence; do not treat it as 「nobody polices this」 -- Cost: No money. Before you use somebody else's open-source code, look at its licence, which is the set of conditions the author has written down. The GPL family requires a product that uses it to publish its source code too, which conflicts with selling closed source, and where it conflicts you change to something else. The hard part is giving up a library you have got used to. -- In plain language: Infringing copyright is made good by paying the other side's loss or the money you earned, and where the infringement is deliberate and the circumstances are serious it can reach 1 to 5 times that. One company hired a rival's former employees, produced substantially similar software, and argued that the other side's software was under GPLv2 and should have been open source. The Supreme People's Court held that whether the other side had open-sourced it made no difference to its copyright, and ordered it to stop the infringement and pay 500,000 yuan. -- Benefit: Infringing copyright is made good by paying the rights holder's actual loss, or by paying the money the infringer earned. Where neither can be worked out, it is set by reference to what it would cost to buy that licence properly. Deliberate infringement in serious circumstances can carry 1 to 5 times the compensation. Where the public interest is harmed, the illegal gains may also be confiscated and a fine imposed. The official case: (2021) Supreme People's Court Intellectual Property Chamber Civil Final No. 51. The defendant hired a former employee of the plaintiff and developed gateway software that was substantially similar to the plaintiff's. The defendant argued that the plaintiff's software was based on GPLv2 and should have been open source. The Supreme People's Court held that「开发者自身是否违反 GPLv2 协议和是否享有软件著作权是两个独立问题」, and ordered it to stop the infringement and pay 500,000 yuan (nationwide) +- Cost: No money. Before you use somebody else's open-source code, read its licence once (the conditions the author has written down). The GPL family requires a product that uses it to release its own source as well, which clashes with selling closed-source software; where it clashes, change to another one. The hard part is giving up a library you have got used to. +- In plain language: Infringing copyright is made good by the other party's loss or by the money you earned, and where the infringement is deliberate and the circumstances serious the damages can reach 1 to 5 times that. One company hired a competitor's former employee and produced substantially similar software, arguing that the other party's software was under GPLv2 and should have been open source. The Supreme People's Court held that whether the other party had opened its source made no difference to its copyright, and still ordered the infringement to stop and awarded 500,000 yuan in damages. +- Benefit: For infringement of copyright, the damages are the right holder's actual loss, or the money the infringer earned. Where neither can be worked out, the figure is set by reference to what a normal purchase of that licence would cost. Where the infringement is deliberate and the circumstances serious, damages of 1 to 5 times may be awarded. Where the public interest is harmed, the illegal gains may also be confiscated and a fine imposed. Official case: (2021) Supreme People's Court Intellectual Property Chamber Final No. 51. The defendant hired the plaintiff's former employee and developed gateway software substantially similar to the plaintiff's software. The defendant argued that the plaintiff's software was based on GPLv2 and should have been open source in the first place. The Supreme People's Court held that 「开发者自身是否违反 GPLv2 协议和是否享有软件著作权是两个独立问题」, that is, whether the developer has himself breached the GPLv2 agreement and whether he holds the software copyright are two separate questions, and ordered the infringement to stop with 500,000 yuan in damages (nationwide) - Evidence grade: B -- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Copyright Law, 著作权法（2020 年修正，第五十三、五十四条）. ; Intellectual Property Chamber of the Supreme People's Court (最高人民法院知识产权法庭) (2022). Copyright Infringement Case Involving 「Open-Source Software」（（2021）最高法知民终 51 号）. -- Notes: This case shows that「the other side used GPL code」cannot be used as your shield for copying it. In the other direction, a GPL author suing a commercial user who did not open-source it already has a judgment in force in China; the original text was not found on the website of the Supreme People's Court, so it is not cited here. The grade is B because whether you follow an open-source licence can only be handled through contracts and the general provisions of the copyright law, and there is no article written for it. +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Copyright Law, 著作权法（2020 年修正，第五十三、五十四条）. ; Supreme People's Court Intellectual Property Tribunal (最高人民法院知识产权法庭) (2022). Copyright infringement case involving open-source software（（2021）最高法知民终 51 号）. +- Notes: What the case shows is that 「the other side used GPL code」 is not a reason for you to copy it. The other way round, there is already a binding judgment in China from a GPL author suing a commercial user that had not opened the source, but this section did not find the text on the Supreme People's Court website, so nothing is cited. The grade is B because keeping to an open-source licence can only be governed through contracts and the general provisions of the Copyright Law; there is no dedicated article. ### 15. Collect only the personal information your app genuinely needs, get a guardian's consent for anyone under 14, and write the privacy policy clearly @@ -139,14 +128,14 @@ The measures in this section are personal freedom and money, and what you get ba - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Personal Information Protection Law, 个人信息保护法（第六、十三、三十一、六十六条）. - Notes: In practice,「the person directly responsible」very often lands on the people who write the code and build the product. The Ministry of Industry and Information Technology regularly reports apps taken down for breaches; this section did not verify which particular batch of reports it was, so nothing is cited. -### 16. Do the ICP filing before a website or app goes live, and keep logs for at least 6 months as the graded protection system requires +### 16. Before a website or app goes live, do the ICP filing first, and keep the logs for 6 months or more as the graded protection rules require -- Cost: A little money. The filing is free, and going through a cloud vendor's process takes a few weeks. The graded protection assessment is charged according to the level. Graded protection, known as 等保, is a set of security management measures by which the state grades systems by how important they are and requires websites and systems to meet them; a small site generally does not need the assessment. -- In plain language: The filing costs nothing, and going through a cloud vendor's process takes a few weeks. A domestic server without a filing cannot resolve the domain, and refusing to rectify leads to the site being closed. Not showing the filing number on the home page carries a fine of 5,000 to 50,000 yuan. Saying the filing is free of charge while actually charging carries a fine of 100,000 to 1,000,000 yuan. Logs have to be kept for a full 6 months, and refusing to rectify or causing consequences carries a fine of 10,000 to 100,000 yuan. -- Benefit: Commercial internet information services are subject to a licence, and non-commercial ones to a filing. Without a licence, or without a filing, you may not carry on. Where there is no filing, the telecommunications administration authority orders rectification, and where it is not rectified orders the website closed. Operating without a licence means the illegal gains are confiscated and a fine of 3 to 5 times the illegal gains imposed, or a fine of 100,000 to 1,000,000 yuan. Not showing the filing number on the home page carries a fine of 5,000 to 50,000 yuan. A network operator must, under the graded protection system, formulate security management rules, guard against attacks, keep network logs for not less than 6 months, and classify, back up and encrypt data. Failure to comply leads to an order to rectify and a warning. Refusing to rectify, or causing consequences, carries a fine of 10,000 to 100,000 yuan, and the person in charge 5,000 to 50,000 yuan (nationwide) +- Cost: The filing is free, and going through your cloud provider's process takes about a few weeks. The graded protection assessment is charged according to the level. Graded protection (等保) is a set of security measures the state grades by how important the system is and then requires a website or system to put in place; a small site generally does not need the assessment. +- In plain language: The filing costs nothing and the process through your cloud provider takes about a few weeks; a server inside China with no filing gets the site shut down. A filing that says it charges nothing but in fact charges a fee is fined 100,000 to 1,000,000 yuan. The logs have to be kept for a full 6 months, and a refusal to put it right, or consequences following from it, is fined 10,000 to 100,000 yuan. +- Benefit: Commercial internet information services are run under a licence, and non-commercial ones under a filing. Without a licence, or without a filing, you may not carry on at all. Where there is no filing, the telecommunications authority orders it to be put right, and where it is not put right the website is ordered closed. Operating without a licence brings confiscation of the illegal gains and a fine of 3 to 5 times the illegal gains, or a fine of 100,000 to 1,000,000 yuan. A website whose home page does not show the filing number is fined 5,000 to 50,000 yuan. A network operator has to work out security management rules under the graded protection system (等保), guard against attacks, keep network logs for not less than 6 months, and classify, back up and encrypt its data. Failure to comply brings an order to put it right and a warning. A refusal to put it right, or consequences following from it, is fined 10,000 to 100,000 yuan, and the person in charge 5,000 to 50,000 yuan (nationwide) - Evidence grade: A - Sources: State Council (国务院) (2011). Measures for the Administration of Internet Information Services, 互联网信息服务管理办法（2011 年修订，第四、十九、二十二条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十三、六十一条；2016 年文本为第二十一、五十九条）. -- Notes: A domestic server without a filing cannot resolve the domain at all. There are two common traps: filing as an individual non-commercial site while actually charging; and changing the domain or the entity without going back to change the filing. Keeping logs for 6 months is both a duty and a way of protecting yourself, because when something goes wrong you can show it was not you. +- Notes: Without a filing, a server inside China cannot resolve the domain at all. How to file, how much you are fined for not showing the filing number, and how to check whether a provider is qualified, see item 4 in section 26 (a server inside China needs a filing). Two traps are common: filing as a non-commercial personal site and in fact charging money; changing the domain or the entity and not going to change the filing. Keeping the logs for 6 months is both a duty and self-protection: if something happens, it proves it was not you. ### 17. Before offering a generative AI service to the public, do the security assessment and the algorithm filing as the rules require, and label the generated content diff --git "a/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" "b/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" index edcb4a76..59d3ed84 100644 --- "a/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" +++ "b/en/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" @@ -12,7 +12,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: B - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第五十六、一千零六十四条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2023). Company Law, 公司法（2023 年修订，第四十七、五十四条）. - Notes: 「亏得起」 means this: if the money is lost entirely, you still do not have to sell your home or borrow from relatives to plug the hole. For how many businesses survive, and for how many years on average, no verifiable official statistics could be found on the National Bureau of Statistics (国家统计局) or State Administration for Market Regulation (国家市场监督管理总局) websites this time. So the text gives no figures and only the legal consequences. That is why the grade is B. - ### 2. Do not sign a personal guarantee for company loans, and never let your spouse sign one either - Cost: No money. When a bank, a supplier or a landlord asks you to sign a 「股东个人连带担保」 or asks for 「配偶签字」, stop for a moment. Signing that page means you personally give up the protection that 「the company's debts do not reach the individual」 (giving up limited liability). Do not sign if you can avoid it. If you must sign, sign only a 「一般保证」 and write down the largest amount you are guaranteeing. The hard part is that the other side often says 「no signature, no loan」. You have to hold out against that. @@ -21,7 +20,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第六百八十一、六百八十七、六百八十八、一千零六十四条）. - Notes: When a bank lends to a small or micro business, it almost always asks a shareholder to sign a joint and several guarantee. You can still get the loan. What matters is understanding what that page gives up before you sign, and keeping the guaranteed amount inside the 「can afford to lose」 figure from item 1 in this section (start a business only with money you can afford to lose). For the general rules on promissory notes and guarantee documents, see item 18 in section 8. - ### 3. Choose the right entity before you open: a sole proprietorship and an ordinary partner pay to the end, only a limited liability company is 「limited」 - Cost: Registration is done on the local government services website and takes a few hours. After a limited liability company is set up you have to keep books and file tax returns every month, and that is a running cost: doing it yourself takes time, paying someone to keep the books costs money, and prices differ by place. @@ -30,7 +28,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第五十六条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2006). Partnership Enterprise Law, 合伙企业法（2006 年修订，第二条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2023). Company Law, 公司法（2023 年修订，第四、二十三、四十七、五十三、五十四条）. - Notes: The condition for the 「limited」 part is that company money and private money stay apart: do not mix the company account and your personal account, do not pay household expenses out of the company's money, and keep complete records. Do not put a large figure in the registered capital just because it looks better. Whatever you subscribe for has to be paid in real money within 5 years. Put in 1,000,000 and you have created a potential debt of 1,000,000. - ### 4. Do not act as a nominal shareholder, and do not hold shares on someone else's behalf - Cost: No money. If someone says 「lend me your ID to register as a shareholder」 or 「the shares go under your name for now」, refuse every time. @@ -39,7 +36,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2023). Company Law, 公司法（2023 年修订，第四十七、四十九、五十、五十四条）. ; Supreme People's Court (最高人民法院) (2020 revision). Provisions on Several Issues Concerning the Application of the Company Law of the People's Republic of China (III), 关于适用《中华人民共和国公司法》若干问题的规定（三）（法释〔2011〕3 号，2014 年法释〔2014〕2 号、2020 年法释〔2020〕18 号两次修正，第二十四、二十五条）:「如无法律规定的无效情形，人民法院应当认定该合同有效」「实际出资人未经公司其他股东半数以上同意，请求公司变更股东……人民法院不予支持」「名义股东处分股权造成实际出资人损失，实际出资人请求名义股东承担赔偿责任的，人民法院应予支持」. In plain terms: unless the law gives a ground of invalidity, the contract is valid; an actual investor who asks the company to change the shareholder registration without the consent of more than half the other shareholders is not supported by the court; where a nominal shareholder disposes of the equity and the actual investor suffers a loss, the court supports the actual investor's claim for compensation against the nominal shareholder. (National Laws and Regulations Database, 国家法律法规数据库) ; Supreme People's Court (最高人民法院) (2025). Interpretation on Several Issues Concerning the Application of the Company Law of the People's Republic of China (draft for comment), 关于适用《中华人民共和国公司法》若干问题的解释（征求意见稿）第九十条. - Notes: Between a nominal shareholder and the person who really put in the money, a nominee agreement holds unless the law makes it invalid. But if the person who really put in the money wants the registration changed to their own name, more than half of the other shareholders have to agree. If the nominal shareholder sells or pledges the shares, the court handles it under the good-faith acquisition rules of the Civil Code. Where the buyer was unaware, paid a reasonable price and completed the registration, the person who put in the money cannot get the shares back and can only claim compensation from the nominal shareholder. A draft for comment on a replacement for this judicial interpretation came out in September 2025, and the whole document is repealed once the final version takes effect. Use the 2020 revision, which was still in force as of September 2026. For the risk of being a nominal legal representative, see item 28 in section 8. - ### 5. Before you take a franchise, check the Ministry of Commerce filing, get the written disclosure, and put the 「cooling-off period」 into the contract - Cost: No money. Go to the Ministry of Commerce's 「商业特许经营信息管理系统」 website and check whether the brand has filed. Then require the other side to give you, before signing, written material covering everything it has to disclose, together with the contract text. A few hours is enough. @@ -57,7 +53,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: State Council (国务院) (2021). Regulations on the Registration of Market Entities, 市场主体登记管理条例（国务院令第 746 号，第十、十三、十四、十六、十九、二十一条）. ; State Administration for Market Regulation (国家市场监督管理总局) (2022). Detailed Rules for Implementing the Regulations on the Registration of Market Entities, 市场主体登记管理条例实施细则（市场监管总局令第 52 号，第十二、二十五、二十六条）. ; Ministry of Finance and National Development and Reform Commission (财政部、国家发展改革委) (2014). Notice on Cancelling, Suspending and Exempting a Batch of Administrative and Institutional Charges, 关于取消、停征和免征一批行政事业性收费的通知（财税〔2014〕101 号，附件 1）. - Notes: Registration itself is not charged for, and whatever you were charged is most likely an agent's errand fee. The price of cutting the company seal is set by the market. The General Office of the State Council has expressly ordered 「严禁指定公章制作单位制作公章」, so when someone says 「you must go to the designated place to cut the seal」 you do not have to accept that. How much you put in the registered capital does not affect the shopfront. It only sets the ceiling on your liability, and it has to be paid in full within 5 years; see item 3 in this section (choose the right entity before you open). If the business scope contains items that need a licence, you cannot open until the licence comes through; see item 7 in this section (trades that need a licence) - ### 7. In a trade that needs a licence, do not open before the licence comes through - Cost: No money. Before you open, find out whether the business you are starting needs a licence. Food service and selling food need a food business licence. Selling only whole packages made by someone else needs no licence, only a filing. Open for business once the licence is in place. @@ -66,7 +61,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: State Council (国务院) (2017). Measures for Dealing with Operations Without a Licence or a Business Licence, 无证无照经营查处办法（国务院令第 684 号，第五、六、十三条）. ; State Administration for Market Regulation (国家市场监督管理总局) (2023). Measures for the Administration of Food Business Licences and Filings, 食品经营许可和备案管理办法（市场监管总局令第 78 号，第四条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百二十五条）. - Notes: The licensing conditions for finance, medical care and education and training each have their own dedicated laws; this item verified only the general enforcement rules and the Criminal Law article. A bubble tea shop or a snack stall counts as food service and needs a food business licence. Renovating first and getting the licence afterwards, then trading before it comes through, is a common trap. For how the food business is divided into tiers and whether selling fresh meat and vegetables needs a certificate, see item 8 (work out which tier of food business you fall into). - ### 8. Work out which tier of food business you fall into: production and food service need a licence, selling only prepackaged food needs a filing, selling fresh meat and vegetables needs neither - Cost: The licence itself is not charged for. But the premises and the equipment have to meet the standard, and the on-site inspection plus the paperwork takes several weeks. If you sell meat you also have to keep the quarantine certificate and the certificate of conformity for meat quality inspection for every batch. @@ -75,7 +69,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025, third revision). Food Safety Law, 食品安全法（第三十五、三十六、一百二十二条）. ; State Council (国务院) (2021). Regulations on the Administration of Pig Slaughtering, 生猪屠宰管理条例（国务院令第 742 号，第二、二十三、三十一条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2021 revision). Animal Epidemic Prevention Law, 动物防疫法（第五十一、一百条）. ; Supreme People's Court (最高人民法院) (2024). Interpretation on Several Issues Concerning the Application of the Law in the Trial of Food and Drug Punitive Damages Cases, 关于审理食品药品惩罚性赔偿纠纷案件适用法律若干问题的解释（法释〔2024〕9 号，第四条）. - Notes: The thresholds for small workshops, small food service and food stalls are set by each province, so this item gives no specific figures. Go to the market regulation bureau website of the province you are in, search for 「食品小作坊小餐饮小摊贩管理条例」, and read the local version. Selling your own food online does not remove a single licence because it happens 「online」: whatever needs a licence still needs one. Beyond the licences, the general rules of this section apply. For the entity and the licence see item 6 (settle the name, premises and business scope before you register), and for whether other trades need a certificate see item 7 (trades that need a licence). How much the buyer can claim see item 31 in section 5 (ten times damages for food). - ### 9. Bagging it and selling makes it prepackaged food: the production date, the shelf life and the ingredients list on the label may not be missing - Cost: Printing a label costs from a few cents to a few tenths of a yuan a sheet. Every batch has to have its date and its ingredients filled in correctly, and that takes some of your time. @@ -93,7 +86,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025, third revision). Food Safety Law, 食品安全法（第七十一、七十三、一百二十五、一百四十条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2021 revision). Advertising Law, 广告法（第十七、五十八条）. - Notes: 「Health functions」 cannot be claimed by ordinary food either. The health functions a health food may claim can only come from the catalogue the state sets, and the product has to go through registration or filing, a step ordinary food does not have. For how buyers recognise this kind of wording, see item 10 in section 6 (do not spend a lot of money on health products to look after your health). Beyond the packaging and the advertisement, the description pages, the livestream scripts and the social media posts count as advertising content too, and once they are sent out the record exists. - ### 11. The food business has a criminal line: selling diseased or dead meat or goods over the limit is already enough, and adding toxic or harmful substances carries 5 years whatever the amount - Cost: No money. Do what item 8 (work out which tier of food business you fall into) says: keep the quarantine certificate and the purchase vouchers, and put only ingredients in the list whose approval number you can look up. @@ -102,7 +94,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020 revision). Criminal Law, 刑法（第一百四十一、一百四十三、一百四十四条）. (National Laws and Regulations Database, 国家法律法规数据库) ; Supreme People's Court and Supreme People's Procuratorate (最高人民法院、最高人民检察院) (2021). Interpretation on Several Issues Concerning the Application of the Law in Handling Criminal Cases Endangering Food Safety (法释〔2021〕24 号，第一、二、五、七、九、十七、十八条，2022 年 1 月 1 日起施行). (the full text reprinted by the Hebei Provincial Administration for Market Regulation) and (the full text reprinted by the Liaoning Provincial Department of Justice; the two copies were checked word for word against each other) - Notes: The dividing line with the three items above is whether food safety itself has been breached. No licence, a missing label item, an exaggerated health claim: those are punished with money. Selling diseased or dead meat, goods over the limit, adding additives freely, or putting non-food ingredients into food puts you into criminal procedure. The cheapest defence is still the set of actions in item 8 (work out which tier of food business you fall into). Take meat only from designated slaughterhouses. Keep the quarantine certificate and the certificate of conformity for meat quality inspection for every batch. Use only ingredients whose approval number you can look up. The 「meat cleared cheaply as a bargain」 and the 「offcuts from who knows where」 are what cause the most trouble. The Supreme People's Court gazette page that is the original source returned 502 this round, so two provincial official reprints are cited instead, and they were checked word for word against each other. - ### 12. Once you hold the licence you have a duty to file returns, and with no income you still file on time (file a sheet of zeros) - Cost: No money. File on the electronic tax bureau by month or by quarter; each time takes a few minutes. Paying someone to keep the books is a separate cost in the market. If you use staff, you also have to complete the social insurance registration within 30 days counting from the day the entity was established. The hard part is remembering to file on time in months with no business. @@ -111,7 +102,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: State Council (国务院) (2002). Detailed Rules for Implementing the Law on the Administration of Tax Collection, 税收征收管理法实施细则（国务院令第 362 号，第三十二、七十五条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2015). Law on the Administration of Tax Collection, 税收征收管理法（2015 年修正，第二十五、三十二、六十二、六十四条）. ; State Taxation Administration (国家税务总局) (2019). Announcement on Certain Matters Concerning Tax Collection and Administration, 关于税收征管若干事项的公告（国家税务总局公告 2019 年第 48 号，第三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Individual Income Tax Law, 个人所得税法（2018 年修正，第十二条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Social Insurance Law, 社会保险法（2018 年修正，第五十七条）. - Notes: After the 「多证合一」 reform you no longer register with the tax authority separately. Log in to the electronic tax bureau with the unified social credit code on the business licence, confirm the registration details, and declare which taxes you owe. Sole proprietorships and sole proprietorship enterprises file individual income tax on business income. First pay once 「在月度或者季度终了后十五日内」, that is, within fifteen days after the end of the month or of the quarter, then 「在取得所得的次年三月三十一日前办理汇算清缴」, the annual settlement return due before 31 March in the year after the income is obtained (totalling up the year, with a refund or a top-up). You file even if you never opened, and a nil return is still a return. Leave it alone for three months and you become an abnormal business household, invoices stop working, and you cannot use the simplified cancellation procedure either; see item 23 in this section (if it fails, exit through the proper procedure) - ### 13. Issue invoices only for real transactions, and small-scale taxpayers should use the whole exemption threshold - Cost: No money. Do not buy invoices, do not sell invoices, and do not issue invoices for someone else's 「走账」. Issue invoices and file tax on the basis of the business that actually happened. @@ -129,7 +119,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: State Taxation Administration (国家税务总局) (2025). 税务部门提醒广大纳税人——增强防范意识 警惕互联网涉税诈骗 (The tax department reminds all taxpayers to strengthen their awareness of prevention and beware of tax fraud on the internet). ; Yangzhou Tax Service, State Taxation Administration (国家税务总局扬州市税务局) (2023). 扬州防范涉税诈骗宣传篇 (Yangzhou guide to avoiding tax fraud). ; State Taxation Administration (国家税务总局) (2025). Measures for the Administration of Professional Tax Services (for Trial), 涉税专业服务管理办法（试行）（国家税务总局令第 58 号，第三十二条第七项、第三十三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Law of the People's Republic of China on Penalties for Administration of Public Security, 治安管理处罚法（2025 年修订，2026 年 1 月 1 日施行，第六十二条）. ; Supreme People's Court, Supreme People's Procuratorate and Ministry of Public Security (最高人民法院、最高人民检察院、公安部) (2016). Opinions on Several Issues Concerning the Application of the Law in Handling Criminal Cases such as Telecom and Online Fraud, 关于办理电信网络诈骗等刑事案件适用法律若干问题的意见（法发〔2016〕32 号）. - Notes: These scams pick out newly registered sole proprietorships and small companies. The script is 「the system shows you have not filed your taxes, if you do not deal with it today you will be fined」, and then they steer you to their own public account or mini-programme and charge a few hundred yuan at a time. The filing itself is free and takes you a few minutes. Using an agency in the market to keep your books is not illegal, but that agency is not a government department and it will not chase you in the name of the tax bureau. If you have already paid or transferred the money, follow item 2 in section 8 (stopping the payment after being scammed) and call 110 or 96110 at once to ask for the stop. - ### 15. Write 「定金」 when you take money, not 「订金」, put a figure on the liquidated damages, and vet a credit period as if it were a loan - Cost: No money. In the contract write the two characters 「定金」, with an amount not more than 20% of the total contract value. Take the money first and deliver afterwards. Before you agree to let the other side take the goods and pay later (granting a credit period), look into the other side first. @@ -138,7 +127,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第五百八十五、五百八十六、五百八十七、五百八十八条）. - Notes: 「订金」 has none of the rules that a deposit has under the Civil Code, and it is generally treated as an advance on payment. That is the standard view in practice, and this section has not verified any case law. A credit period is lending your own money to the other side interest-free; if they fail, it is a bad debt. Before granting one, go to the National Enterprise Credit Information Publicity System and look up the other side's record, to see whether it has been sued. - ### 16. Sign a written contract within the first month of employment and complete the social insurance registration within 30 days - Cost: The employer's share of the social insurance contributions has to go on being paid, and the rate differs from place to place. Signing the contract itself costs nothing; just use the model contract from the local human resources and social security department. @@ -147,7 +135,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第十、八十二条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2010). Social Insurance Law, 社会保险法（2018 年修正，第五十八、六十、八十四、八十六条）. - Notes: Double wages run from the second month right up to the full year, so the longer you delay the more it costs. For an employee with no signed contract and no social insurance, once there is a work injury, or once they go to labour arbitration after leaving, the back payment plus the fine plus the damages often come to far more than the money you saved. - ### 17. Pay wages according to the contract and the law, and put raises, loans and bonuses in writing - Cost: No money. Write the wage clearly into the contract. Raises, bonuses, money you advance and loans all go on a signed piece of paper. No verbal promises. Do not pay wages out of your own pocket on the company's behalf either. The hard part is following the same rules when a friend asks. @@ -156,7 +143,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: C - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第十七、三十条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第四百六十九、六百六十八条）. - Notes: This is not about refusing to help people; it is about helping them through the system too. Once favours outside the system become a habit, stopping them causes disputes. When a private loan is mixed into the company accounts and the company fails, the bad debt lands on you personally. The statutory parts have been verified, and the advice itself comes from experience, hence grade C. - ### 18. Sell before you build: use pre-orders and small orders to prove someone will buy, then put money into production - Cost: From tens to hundreds of yuan, to make a sample or a page. Then go and find some dozens of real customers and get them to pay in advance or place small orders. That is several weeks slower than starting production straight away. The hard part is holding back from making the goods first. @@ -165,7 +151,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: B - Sources: Camuffo A, Cordova A, Gambardella A, Spina C (2020). A Scientific Approach to Entrepreneurial Decision Making: Evidence from a Randomized Control Trial. Management Science 66(2):564-586. - Notes: There is only one such study, the sample is small, and it was done in Italy, so the grade is B. The title, the source and the abstract were checked against Crossref and Semantic Scholar, and the effect figures in the full text have not been verified for this section. In practice: if nobody will pay in advance and nobody will place a small order, do not believe that 「it will sell once it exists」. - ### 19. Once the sample is made, run it through the production checklist before you talk about starting - Cost: A few hundred yuan for one small trial batch. Then spend a few hours working out these items: the yield (how many out of 100 pieces can actually be sold), the supplier's minimum order quantity and payment terms, how much money the inventory ties up, packaging and freight, certification and licences, and the trademark. The hard part is holding back from skipping this step and starting production directly. @@ -183,7 +168,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第十七、二十四项，即刑法第二百一十四、二百二十条）. ; Supreme People's Court and Supreme People's Procuratorate (最高人民法院、最高人民检察院) (2025). Interpretation on Several Issues Concerning the Application of the Law in Handling Criminal Cases of Infringement of Intellectual Property Rights, 关于办理侵犯知识产权刑事案件适用法律若干问题的解释（法释〔2025〕5 号，2025 年 4 月 26 日起施行）. ; Standing Committee of the National People's Congress (全国人大常委会) (2019). Trademark Law, 商标法（2019 年修正，第五十七、六十四条；2026 年 6 月 26 日修订后为第七十二、七十八条，自 2027 年 1 月 1 日起施行）. and - Notes: Cosmetics, alcoholic drinks, mother and baby goods, digital products and clothing are the categories where these cases cluster. The common point is that the same product has a very wide gap in price. Labels like 「channel goods」, 「end-of-line stock」 and 「overseas duty-free」 do not replace the receipts. Beyond criminal punishment there is also an administrative penalty: where infringement is established, the infringing goods and the tools are confiscated, and where the turnover is 50,000 yuan or more a fine of up to 5 times may be imposed on top. If you find something wrong after the stock has arrived, stop selling and seal it at once, then contact the brand owner or the market regulation department. Do not go on selling, and do not destroy the goods or the records. Destroying evidence lands you squarely in the fifth situation above, which 「presumes you knew the goods were counterfeit」. For using someone else's artwork see item 21 (make your own artwork or buy a licence). There is one routine aimed squarely at this: someone turns up uninvited and offers to give you a batch of pens or other small goods free, to put in the shop, sell and split the takings. A few days later someone comes to buy, takes evidence on the spot, then claims in damages for infringement. Free goods fail all three conditions for escaping liability: there is no purchase contract and no payment record, the real supplier cannot be named, and the price is plainly abnormal. Not one element of 「legitimately obtained with the supplier identified」 is available. So goods that turn up unsolicited, come with no receipts and leave you only a WeChat ID should never go on the shelf. If you really want to sell on consignment, sign a written consignment contract, keep a copy of the other side's identity card and business licence, and write in who owns the goods and who is liable if there is infringement. - ### 21. The artwork on the goods, the packaging, the hang tags and the promotional images is either your own or licensed; changing a colour or adding an icon does not count as 「altered」 - Cost: Licences for a genuine stock library start at tens of yuan per image. Having a designer produce something original costs a few hundred to a few thousand yuan. For every image used on the goods, keep the licence certificate and the purchase record. The hard part is not using images you found at random by searching online. @@ -192,7 +176,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Copyright Law, 著作权法（2020 年修正，第五十二条第六项、第五十三条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Criminal Law Amendment (Eleven), 刑法修正案（十一）（第二十、二十一项，即刑法第二百一十七、二百一十八条）. ; Supreme People's Court and Supreme People's Procuratorate (最高人民法院、最高人民检察院) (2025). Interpretation on Several Issues Concerning the Application of the Law in Handling Criminal Cases of Infringement of Intellectual Property Rights, 关于办理侵犯知识产权刑事案件适用法律若干问题的解释（法释〔2025〕5 号）. ; Standing Committee of the National People's Congress (全国人大常委会) (2019). Trademark Law, 商标法（2019 年修正，第五十七条第二项；2026 修订后为第七十二条第二项）. - Notes: Three traps come up most often. First, printing cartoon characters and celebrity photos found by searching online onto clothes, cups and phone cases. Second, typefaces: most commercial typefaces need a licence of their own, and a free preview is not a free commercial licence. Third, unlicensed assets slipped into a design file made for a customer; the liability still sits with the person selling the goods. The low-effort route is to use an image library that carries a commercial licence and to save the licence page. Or have the designer state in the contract where the assets came from and how far the licence extends, with the designer responsible if something goes wrong. Beyond criminal punishment there is civil damages and an administrative fine: where the public interest is harmed, the illegal gains may be confiscated and a fine of 1 to 5 times the amount of illegal business operations imposed. For checking a trademark before production, see item 19 (once the sample is made, run it through the production checklist). In the other direction, if someone else uses your own images, you enforce your rights under these same rules. - ### 22. Do your job by the rules, and before a favour outside the rules, work out whose responsibility it is - Cost: No money. At work, follow the system, the medical orders and the process. Do not make decisions for other people beyond your own authority. Before you step in to help, ask yourself one question: if something goes wrong, who is responsible. The hard part is resisting a request to 「make an exception」 on the spot. @@ -201,7 +184,6 @@ This section counts money and legal liability, and is not converted into lifespa - Evidence grade: B - Sources: State Council (国务院) (2008). Regulations on Nurses, 护士条例（国务院令第 517 号，第十七条）. ; State Council (国务院) (2020). Regulations on Nurses, 护士条例（2020 年修订，第十七条）. (National Laws and Regulations Database, 国家法律法规数据库) ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一百八十四条）. - Notes: Within your own role, act by the rules, and rescuing someone in an emergency is protected by law. For everything else, first work out whose authority it is and whose responsibility it is, and do not let a private good intention stand in for a decision the system calls for. The Regulations on Nurses were revised in 2020 by State Council Order No. 726, which changed only the provisions on registration and left Article 17 untouched. This round the official text of the provisions in the Physicians Law about practising according to the diagnostic and treatment standards could not be opened, so they are not cited. - ### 23. If it fails, exit through the proper procedure: cancel through the simplified route if you can, go through bankruptcy if the assets do not cover the debts, and do not simply leave it alone - Cost: You can do the simplified cancellation yourself: post it on the National Enterprise Credit Information Publicity System for 20 days. Where debts are still outstanding, you first have to sort the accounts, collect what is owed and pay what is due; that step is called liquidation. Going through bankruptcy means applying to the court, which takes months. The hard part is not wanting to face it after a loss and putting it off indefinitely. diff --git "a/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" "b/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" index 9ffd35a4..e84e9853 100644 --- "a/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" +++ "b/en/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" @@ -6,14 +6,14 @@ This section says what to do beforehand, what to do next, and what not to do. Th This section only covers actions that take minutes. Recognising a scam, or being blackmailed with money, is handled on a scale of days and does not belong here. See item 3 in section 8 (hard anti-fraud rules) for the hard rules and the seven most common scams. See item 32 in section 8 (being blackmailed with private material) for demands for money backed by private photos or nude videos. If the money has already gone out, call 110 or 96110 at once and ask for the payment to be stopped. See item 2 in section 8 (stopping a payment). -### 1. Someone collapses and is not breathing: press hard on the centre of the chest, send someone else to call 120 and fetch an AED +### 1. Someone has fallen and is not breathing: press hard on the centre of the chest at once, and send someone else to call 120 and fetch an AED -- Cost: No money. A cardiopulmonary resuscitation course from the Red Cross or a hospital runs half a day to a day, and one course is enough for years. You can do it without any training: stack your hands at the midpoint of the line between the nipples, press down hard about 2 times a second, and do not stop. -- In plain language: The person you are most likely to press on is your own family. 79.2% of cardiac arrests happen at home. Where someone presses, the share discharged alive rises from 3.9% to 16.1%. If they do not answer and are not breathing, press straight away. Put your hands at the midpoint between the nipples, stack them, press hard, 2 times a second, and do not stop. Let the second person call 120 and fetch the AED. Your hands do not stop. -- Benefit: Pooling 79 studies and 142,740 people together (meta-analysis): among people whose heart stopped outside hospital (out-of-hospital cardiac arrest), only 7.6% in total were discharged alive. Where a bystander performed cardiopulmonary resuscitation, the share discharged alive rose from 3.9% to 16.1%. 53% had someone present who saw the collapse happen, and only 32% actually received compressions from a bystander. China's BASIC-OHCA registry recorded 38,227 emergency cases taken on by ambulance crews. In 30,282 of them (79.2%) the cardiac arrest happened at home. Among this group the bystander compression rate was 20.3%, and 1.2% were discharged alive. The order is: compress first, then let the second person call and fetch the AED. Once the machine is switched on, do what its voice tells you +- Cost: No money. A cardiopulmonary resuscitation course at the Red Cross or at a hospital takes half a day to a day, and one course is enough for years. You can do it without training: stack your hands at the midpoint of the line between the nipples, press down hard at about 2 compressions a second, and do not stop. +- In plain language: The person you are most likely to press on is someone in your own home: 79.2% of cardiac arrests happen at home. Where a bystander presses, the share who leave hospital alive rises from 3.9% to 16.1%. If they do not answer when you call and are not breathing, start pressing straight away. Put your hands stacked at the midpoint of the line between the nipples, press hard at 2 compressions a second, and do not stop. Send a second person to call 120 and fetch the AED, and do not let your hands stop. +- Benefit: Pooled across 79 studies and 142,740 people (meta-analysis): of people whose heart stopped outside hospital (out-of-hospital cardiac arrest), only 7.6% in all leave hospital alive. Where a bystander did cardiopulmonary resuscitation, the share leaving hospital alive rises from 3.9% to 16.1%. Someone was standing close enough to witness the collapse in 53% of cases, and only 32% actually got compressions from a bystander. China's BASIC-OHCA registry recorded 38,227 emergency cases taken on by ambulance crews. In 30,282 of them (79.2%) the cardiac arrest happened at home. Among that group, the bystander compression rate was 20.3% and the share leaving hospital alive 1.2%. The order is compressions first, and then a second person calls and fetches the AED. Once the machine is switched on, follow its voice instructions - Evidence grade: A - Sources: Sasson C et al (2010). Predictors of survival from out-of-hospital cardiac arrest: a systematic review and meta-analysis. Circulation: Cardiovascular Quality and Outcomes. ; Zheng J et al (2023). Incidence, process of care, and outcomes of out-of-hospital cardiac arrest in China: a prospective study of the BASIC-OHCA registry. The Lancet Public Health. ; National People's Congress (全国人大) (2020). Civil Code, 民法典第一百八十四条. -- Notes: Do not pinch the philtrum first, do not look for the family first, and do film it first. Do not hold back from pressing hard because you are afraid of breaking ribs. Article 184 of the Civil Code states that a person who gives emergency rescue voluntarily and injures the person helped bears no civil liability. +- Notes: For a child under 1 year the technique is different and breaths are needed: see item 44 in this section (infant cardiopulmonary resuscitation). Do not pinch the philtrum first, do not go looking for the family first, and do not film it first. Do not hold back from pressing hard for fear of breaking ribs. Article 184 of the Civil Code says this: where a person gives emergency rescue voluntarily and injures the person rescued, the rescuer bears no civil liability. ### 2. An elderly person falls or someone collapses: kneel down and speak to them, call 120, and do not lift them straight away. Walking away from a stranger is legal too, and if you do stop, do not move them by hand @@ -31,7 +31,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: A - Sources: Emberson J, Lees KR, Lyden P, et al. (2014). Effect of treatment delay, age, and stroke severity on the effects of intravenous thrombolysis with alteplase for acute ischaemic stroke: a meta-analysis of individual patient data from randomised trials. Lancet, 384(9958), 1929-1935. - Notes: The earlier the thrombolysis, the better it works. Past 4.5 hours it is close to useless. Call 120 rather than driving yourself, because the ambulance takes you straight to a hospital with a stroke centre and warns them in advance so they can prepare. Write down the exact time the symptoms started, because the doctor needs it to decide whether thrombolysis is possible. Do not give water or medicine by mouth, and do not wait to 「休息一下看看」 — rest a bit and see how it goes - ### 4. A sudden spinning feeling, unsteady standing, seeing double or one eye going black, or being unable to touch your own nose: treat it as a stroke and call 120 - Cost: No money. On top of the three actions of a crooked face, arms raised and speech (item 3 in this section, a crooked mouth and unclear speech), add two more. One is to have the person stand upright with the eyes closed and see whether they sway, then ask whether they see double or whether part of the field of view is missing. The other is to point at your own nose with a finger when you suspect a stroke, then close your eyes and point again. @@ -40,6 +39,7 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: Aroor S, Singh R, Goldstein LB (2017). BE-FAST (Balance, Eyes, Face, Arm, Speech, Time): Reducing the Proportion of Strokes Missed Using the FAST Mnemonic. Stroke, 48(2), 479-481. - Notes: A blocked vessel in the area of the cerebellum and brainstem (posterior circulation infarction) often causes neither a crooked mouth nor a weak arm. It only causes spinning, walking to one side, diplopia (seeing double), and choking on water or food, and it is most often mistaken for a neck problem, low blood sugar or an inner ear disorder. Missing the nose with your finger and not being able to stand with your eyes closed are signs that the cerebellum is involved. But no action rules a stroke out in reverse: if it comes on suddenly and does not settle, call 120 and write down the exact time of onset. The study looked at the records of one hospital only, and the authors say themselves that it needs further verification + ### 5. One eye suddenly goes black as if a blind had been pulled down, even if it clears within minutes: go to the emergency department as a stroke the same day - Cost: No money. Emergency assessment and the blood vessel checks take up half a day to a day. @@ -48,7 +48,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: Mac Grory B, Schrag M, Biousse V, et al. (2021). Management of Central Retinal Artery Occlusion: A Scientific Statement From the American Heart Association. Stroke, 52(6), e282-e294. ; Rothwell PM, Giles MF, Chandratheva A, et al. (2007). Effect of urgent treatment of transient ischaemic attack and minor stroke on early recurrent stroke (EXPRESS study): a prospective population-based sequential comparison. Lancet, 370(9596), 1432-1442. - Notes: The B is because EXPRESS was not a randomised trial. It compares two periods, before and after the change of practice, in the same population, and other conditions may also have changed between the two periods, which cannot be separated out. A black-out of vision lasting a few minutes that then clears by itself is most often put down to being tired or to low blood sugar. It is a form of transient ischaemic attack, and the next few days are exactly the window when strokes cluster. Losing vision in one eye with no pain, no redness and no watering of the eye deserves particular attention. Do not wait until tomorrow to book an eye clinic. Go straight to the emergency department of a hospital with a stroke centre. - ### 6. One eye swollen and painful and red, with a ring of rainbow around lights, plus headache, nausea and vomiting: go to the eye emergency department the same day - Cost: No money. Measuring the pressure inside the eye and examining the anterior chamber at the eye emergency department takes half a day. @@ -57,7 +56,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: Zhou L, Wu S, Wang Y, Bao X, Peng T, Luo W, Ortega-Usobiaga J (2022). Clinical presentation of acute primary angle closure during the COVID-19 epidemic lockdown. Frontiers in Medicine, 9, 1078237. ; Sung MS, Kim HJ, Park SW (2023). Predictors of long-term visual field outcome after an episode of acute primary angle closure. Clinical & Experimental Ophthalmology, 51(4), 291-299. ; Wang J, Wang J, Ng TK, Huang C (2025). Asymmetric intraocular pressure changes in dominant and contralateral eyes: the dark room prone provocative test. Seminars in Ophthalmology, 40(4), 325-331. - Notes: There are two reasons for the B. First, the two Wuhan groups were compared by going back through old records, and the lockdown itself brought other differences that cannot be separated from delay. Second, the Korean group is only 50 eyes, and it is retrospective as well. The trigger is not screen time. It is two things together: the pupil dilates in the dark, and looking down or lying face down for a long time pushes the lens forward. People who already have a shallow anterior chamber are easily triggered this way, so the real high-risk group is people over 50, people who are far-sighted, and people with a family history of angle-closure glaucoma. Young people playing games through the night hardly ever have an attack like this. The most common misjudgement during an attack is to treat it as gastroenteritis, migraine or high blood pressure, because the headache and vomiting are more obvious than the eye symptoms. Losing vision in one eye with no pain and no redness is a different matter, see item 5 in this section (one eye suddenly going black). For whether blue-light glasses are worth anything, see item 16 in section 6 (blue-light glasses). - ### 7. Chest pressure, tightness or squeezing that does not ease after more than 15 minutes: call 120, do not put up with it and do not drive yourself - Cost: No money. @@ -66,7 +64,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: Karkabi B, Meir G, Zafrir B, et al. (2021). Door-to-balloon time and mortality in patients with ST-elevation myocardial infarction. European Heart Journal - Quality of Care and Clinical Outcomes. - Notes: What the study compares is the delay after arrival at hospital, not the stretch from onset to arrival. The reasoning is the same either way: every extra stretch of delay kills off another patch of heart muscle. The pain can sit in the upper abdomen, the back or the chin, and the left arm can go numb. Diabetics and older people may have no pain at all, only breathlessness. Do not drive yourself, because losing consciousness on the road is a second accident - ### 8. Sudden tearing pain that travels from the chest through the back and on towards the waist: call 120 and say clearly that the pain is moving - Cost: No money. @@ -75,7 +72,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: Hagan PG, Nienaber CA, Isselbacher EM, et al. (2000). The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. JAMA, 283(7), 897-903. - Notes: The handling of a dissection is exactly the opposite of that of a heart attack. A heart attack needs anticoagulants and thrombolytics, drugs that stop the blood clotting, and using them on a dissection is like letting the tear bleed. So at the hospital say 「撕裂样，疼在往下走」 — tearing, the pain is moving downwards, so that the doctor thinks of a contrast CT. A large difference in blood pressure between the two arms, and no pulse in one of them, are the clues. People whose blood pressure is poorly controlled, people with Marfan syndrome, and people with an aortic aneurysm or dissection in the family are at higher risk. Again, do not drive yourself, and do not put up with it until tomorrow - ### 9. A sudden "the worst headache of my life" that reaches its worst within an hour: go to the emergency department for a head CT - Cost: The emergency visit itself costs very little. One head CT in the emergency department costs several hundred yuan. @@ -84,6 +80,7 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: A - Sources: Perry JJ, Stiell IG, Sivilotti MLA, et al. (2013). Clinical decision rules to rule out subarachnoid hemorrhage for acute headache. JAMA, 310(12), 1248-1255. - Notes: What matters is not how bad the pain is but how fast it reaches the worst. A headache that explodes within a few seconds to a minute is the danger signal; a migraine that builds up slowly is not. With vomiting, a stiff neck, photophobia or a brief loss of consciousness, take it more seriously still. A specificity of only 15.3% means that most of the people who go for a scan on the strength of this rule come out clear. But missing one and having a second bleed is death, and the scan is worth having. Do not take a painkiller to push it down and go back to sleep + ### 10. Two or three weeks to a few months after an elderly person hits their head, if they become unsteady, slow, sleepy or weak on one side: get a head CT - Cost: One head CT costs two or three hundred yuan. @@ -92,7 +89,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: Zhang J et al (2021). Expert consensus on drug treatment of chronic subdural hematoma. Chinese Neurosurgical Journal, 7(1), 47. ; Dziho A et al (2025). Global prevalence and incidence of chronic subdural hematoma: A systematic review. Brain and Spine, 5, 105893. ; Mathon B, Shotar E (2026). Diagnostic and therapeutic management of chronic subdural hematoma in elderly patients. Gériatrie et Psychologie Neuropsychiatrie du Vieillissement, 24(1), 30-41. - Notes: At the time of the knock there can be nothing wrong at all, and the elderly person may not even remember hitting their head. So when you notice at home that someone has recently become confused, walks unsteadily, keeps dozing off or has weakness in one hand, first go back and ask whether they have fallen or hit their head in the last two or three months, then get the CT. Do not put it straight down to senile dementia or to the after-effects of a previous stroke, because getting that wrong costs time. Elderly people taking aspirin, clopidogrel, warfarin or one of the new oral anticoagulants are at higher risk, and so are those who drink heavily over the long term. For fall prevention, see item 13 in section 1 (training balance and leg strength). - ### 11. One leg suddenly swollen, tight and painful to press: get seen as soon as you can. Add sudden breathlessness or chest pain and call 120 at once - Cost: An ultrasound of the blood vessels in the leg (a lower limb venous ultrasound) costs one or two hundred yuan. @@ -101,7 +97,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: Goldhaber SZ, Visani L, De Rosa M (1999). Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet, 353(9162), 1386-1389. ; WHO (2007). Study results released on travel and blood clots. - Notes: Both legs swollen together is usually down to the heart, the liver, the kidneys or how well the veins are working. One leg suddenly swollen and a girth larger than the other is the typical picture of a deep vein thrombosis. The situations where this happens are long journeys by car or plane, being bedridden after an operation or a fracture, a plaster cast, pregnancy and the weeks around childbirth, cancer, and taking the oral contraceptive pill. A clot that breaks loose and blocks the pulmonary artery is a pulmonary embolism: sudden breathlessness, more chest pain when you breathe in deeply, a fast heartbeat, dizziness, and coughing up blood (haemoptysis). At that point call 120 at once. 17.4% is the death rate among diagnosed cases of pulmonary embolism, not the risk you carry because your leg is swollen. Do not frighten yourself with that number. - ### 12. Heavy bleeding: press hard on the wound with your hand first, put a tourniquet on a limb if pressure will not hold, and call 120 at the same time - Cost: A windlass tourniquet costs from a few tens of yuan to a little over a hundred. Keep one in the car and one in the first aid kit. Learn the sequence 「扎在伤口上方、拧到血停、记时间」 — tie it above the wound, wind it until the bleeding stops, note the time. It takes a few minutes. @@ -110,7 +105,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: A - Sources: Teixeira PGR et al (2018). Civilian Prehospital Tourniquet Use Is Associated with Improved Survival in Patients with Peripheral Vascular Injury. Journal of the American College of Surgeons. ; Kragh JF et al (2009). Survival with emergency tourniquet use to stop bleeding in major limb trauma. Annals of Surgery. ; Chinese Medical Doctor Association Branch of Emergency Physicians (中国医师协会急诊医师分会) et al (2020). Expert consensus on emergency use of tourniquets. 中华急诊医学杂志 29(6):773-779: 「院前急救时，止血带一旦使用则不建议松开，除非得到可以替代的彻底止血的救治」 — in prehospital first aid, once a tourniquet has been used it is not recommended to loosen it, unless a treatment that stops the bleeding completely becomes available in its place, and 「最长使用时间不应超过2 h。但如果客观情况无法到医院救治或者无替代止血办法，则在得到正规救援前不解除止血带」 — the longest time of use should not exceed 2 h, but if the situation objectively makes it impossible to get to hospital for treatment and there is no alternative way to stop the bleeding, then do not release the tourniquet before proper rescue arrives. , full text - Notes: Do not use a rope, a shoelace or any other thin thing as a tourniquet. It will not grip an artery, and it cuts the skin. Once a tourniquet is on, do not loosen it now and then to let some blood through. Bleeding on the neck and the torso can only be pressed, never tied. In that battlefield dataset the tourniquet itself cost nobody an amputation, and there were 4 cases of short-lived nerve paralysis. The expert consensus of Chinese emergency physicians agrees: once it has been used before hospital, loosening it is not recommended unless another way of stopping the bleeding completely is available. For bleeding in the upper arm, put it on the upper third of the upper arm; for bleeding in the thigh, put it at the top of the thigh. Do not put it on the forearm, the lower leg, the elbow or the knee. The consensus recommends keeping the total time, up to arrival at hospital and after it, under 2 hours if at all possible, but where the hospital cannot be reached and there is no other way to stop the bleeding, do not release it until proper rescue arrives. A tourniquet improvised from a belt, a tie or a shirt sleeve will struggle to stop the bleeding on its own; you need a stick to wind it tight. - ### 13. Bitten or scratched by a dog or a cat through the skin: rinse alternately with soapy water and running water for 15 minutes, and get the vaccine the same day - Cost: The rinsing is free. The full course of rabies vaccine costs several hundred yuan. Grade III exposure also needs rabies immunoglobulin (ready-made antibodies given by injection), which costs from several hundred yuan to over a thousand. @@ -119,7 +113,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: A - Sources: General Office of the National Disease Control and Prevention Administration (国家疾控局综合司) and General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Notice on issuing the working standard for the prevention and handling of rabies exposure (2023 version), 关于印发狂犬病暴露预防处置工作规范（2023 年版）的通知. ; WHO. Rabies fact sheet, 狂犬病 实况报道. - Notes: Do not skip the handling because the animal is a domestic one that has been vaccinated. The standard looks at the grade the wound has reached (the exposure grade), not at where the animal came from. Do not dress the wound yourself. The ten-day observation method can only decide whether to stop the injections, not whether to start them. The same handling applies to bites from wild animals and from bats. - ### 14. After a scald: rinse immediately with cool running water for 20 minutes, and put on no toothpaste and no soy sauce - Cost: No money. It takes 20 minutes. @@ -128,6 +121,7 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: A - Sources: Griffin BR, Frear CC, Babl F, Oakley E, Kimble RM (2020). Cool Running Water First Aid Decreases Skin Grafting Requirements in Pediatric Burns: A Cohort Study of Two Thousand Four Hundred Ninety-five Children. Annals of Emergency Medicine, 75(1), 75-85. ；Griffin B, Cabilan CJ, Ayoub B, et al. (2022). The effect of 20 minutes of cool running water first aid within three hours of thermal burn injury on patient outcomes: A systematic review and meta-analysis. Australasian Emergency Care, 25(4), 367-376. - Notes: It is cool water, not ice water, and ice makes the damage worse. If clothing is stuck to the wound, do not tear it off; after rinsing, cover it with a clean cloth and go to hospital. Do not put on toothpaste, soy sauce, sesame oil or flour, because it changes the doctor's judgement of the depth and makes infection more likely. Rinsing again within three hours still helps + ### 15. A rash over the whole body, breathlessness or dizziness: handle it as anaphylactic shock, call 120 at once and say what you think it is - Cost: No money. A person already known to have a severe allergy pays for their own epinephrine pen, and it is a prescription medicine. @@ -136,7 +130,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B (strong recommendation in the guideline, but the underlying evidence grade is low) - Sources: Chinese Medical Education Association (中国医药教育协会) et al (2022). Clinical Practice Guidelines for Emergency Treatment of Anaphylactic Shock (2020). Frontiers in Pharmacology: 「Epinephrine is the first-line medicine in GRADE II to IV anaphylaxis」「Intramuscular epinephrine should be injected in the mid-anterolateral thigh」, upper limit for intramuscular injection in adults 0.5 mg and 0.3 mg for those under 14 years, repeated every 5–15 minutes if there is no response. ; Muraro A, et al. (2022). EAACI guidelines: Anaphylaxis (2021 update). Allergy. - Notes: The allergy guideline (EAACI) says of itself that in this field 「证据大多停留在很低的等级」 — the evidence is mostly at a very low grade. It lists a strong recommendation because the benefit is far larger than the risk, not because large trials exist. In China epinephrine is a prescription medicine, and most people do not have it to hand. What you can do is call 120, say 「怀疑过敏性休克」 — I suspect anaphylactic shock — and let the crew bring the drug. Anyone who has had a severe reaction should ask a doctor for an auto-injector and learn to inject into the outer mid-thigh (outer side, slightly to the front), which goes in through clothing. - ### 16. Someone has a fit and goes down: clear the area, lay them on their side, time it, and put nothing in their mouth - Cost: No money. @@ -145,7 +138,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: US CDC. First aid for seizures: 「Don't hold the person down or stop their movements」「Don't put anything in their mouth. This can hurt their teeth or jaw」「Don't offer water or food until they are fully alert」「Time the seizure to track if it lasts more than 5 minutes. If it does, seek immediate medical attention or call 911」. - Notes: The same page lists several other situations that need an ambulance at once: another fit soon after the first one ends, not being woken afterwards or having trouble breathing, being hurt during a fit, a fit in water, and a first-ever fit. When the fit stops, lay them on their side with the face pointing down, and give water only once they are fully awake. In a household with an epileptic patient, these are the steps that get used most. Doing the same for a stranger is less likely to pay you back, but it does not make you liable either (Article 184 of the Civil Code, see item 2 in this section, do not lift them straight away). - ### 17. A person with diabetes starts shaking, sweating cold and speaking unclearly: give 15 grams of sugar first, then re-test after 15 minutes - Cost: No money. Carry a few sugar cubes or a small box of juice with you. @@ -154,7 +146,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: A - Sources: American Diabetes Association (美国糖尿病学会) (2025). Standards of Care in Diabetes 2025, Chapter 6 Glycemic Goals and Hypoglycemia. Diabetes Care: grading 「Level 1: Glucose <70 mg/dL (<3.9 mmol/L) and ≥54 mg/dL (≥3.0 mmol/L)」「Level 2: Glucose <54 mg/dL (<3.0 mmol/L)」「Level 3: A severe event characterized by altered mental and/or physical status requiring assistance for treatment of hypoglycemia, irrespective of glucose level」; handling 「For most individuals, 15 g carbohydrates should be ingested」「Fifteen minutes after initial treatment, repeat the treatment if hypoglycemia persists」. - Notes: Give an unconscious person nothing to eat or drink, because it goes into the airway. Call 120 at once. The same standard also advises that anyone treated with insulin, or at high risk of hypoglycaemia, should be given a supply of glucagon. The wording is 「All individuals treated with insulin or who are at high risk of hypoglycemia...should be prescribed glucagon」. The nasal spray and the pre-filled injector are more usable than one that has to be mixed at the time. All of this assumes you are looking after yourself or a person with diabetes in your household, and the sugar you carry is carried for them. - ### 18. Someone is electrocuted: cut the power first, then move the source aside with a dry wooden stick or the like, never pull them with your hands - Cost: No money. @@ -163,7 +154,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: US CDC. What to Do to Protect Yourself From Electrical Hazards (First aid section). ; Spies C, Trohman RG (2006). Narrative review: Electrocution and life-threatening electrical injuries. Annals of Internal Medicine 145(7):531-537. - Notes: "Cut the power before touching the person" is purely a rule in your own interest: if you pull with your bare hands, the second person down is you. Do not deal yourself with a fallen high-voltage line or anything near a substation. Back away and wait for the electricity authority to cut the power. Cardiac arrest caused by electrocution is handled with the usual compressions and the usual defibrillation. Do not hold back because the label says electrical injury. - ### 19. A carbon monoxide alarm goes off, or everyone in the house gets a headache and nausea at once: get outside first, then call - Cost: No money. @@ -173,23 +163,23 @@ This section only covers actions that take minutes. Recognising a scam, or being - Sources: You J, Liu J, Zhou M (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly 2(2):32: 「In 2018, there were 11,523 deaths caused by carbon monoxide poisoning reported in China」. ; US CDC. Carbon monoxide poisoning: 「More than 400 Americans die from unintentional CO poisoning not linked to fires」「More than 100,000 visit an emergency department」「More than 14,000 are hospitalized」, common symptoms 「headache, dizziness, weakness, upset stomach, vomiting, chest pain, and confusion」. ; US Consumer Product Safety Commission. Carbon Monoxide Information Center: 「If you suspect CO poisoning, get outside to fresh air immediately, and then call 911」. - Notes: The symptoms look like a cold, and they also look like being drunk. Everyone in the house feeling ill at once and getting better outside is the strongest clue. Do not stay inside looking for the cause, and do not go back to turn off a valve. Get the people out and then call. For how to install the alarm, see item 3 in section 1. -### 20. Swallowed cleaning agent, pesticide or medicine: do not induce vomiting, take the bottle with you to hospital at once. Splashed in the eye or on the skin: rinse with plenty of clean water for 15 minutes +### 20. Accidental swallowing of a cleaning agent, a pesticide or medicine: do not induce vomiting, take the bottle with you to hospital at once; if it lands on the eyes or skin, flush with plenty of clean water for at least 15 minutes -- Cost: No money. Put cleaning agents, disinfectant, pesticides and medicines on a high shelf out of a child's reach, or in a locked cupboard. Never decant them into a drinks bottle. -- In plain language: Do not induce vomiting after someone swallows a cleaning agent, pesticide or medicine, because bringing a corrosive liquid back up burns the oesophagus a second time. First get them to drink a moderate amount of milk or egg white, then take the bottle and the instructions to hospital at once. If it is splashed on the skin or in the eye, rinse with plenty of clean water for a full 15 minutes, holding the eyelids open in the eye and not rubbing it. Mixing disinfectant with toilet cleaner gives off chlorine gas. -- Benefit: The China CDC Institute for Occupational Health and Poison Control (中国疾病预防控制中心职业卫生与中毒控制所) gives four kinds of handling. Swallowed (ingested into the digestive tract): get them to drink a moderate amount of milk or egg white at once, 「切记，不要催吐」 — remember, do not induce vomiting. Get to hospital as soon as possible, taking the packaging or the instructions with you, 「以便于医生快速判断成分并制定救治方案」 — so that the doctor can judge the ingredients quickly and work out a treatment plan. Inhaled (into the airway): move the person at once to a well-ventilated area, unbutton the collar and keep the airway clear; if there is severe coughing, breathlessness or difficulty breathing, get medical help at once. Skin contact: rinse the contact area at once with plenty of clean water for at least 15 minutes, taking care not to let the rinse splash into the eyes or onto other unbroken skin. Eye contact: rinse the eye at once with plenty of clean water or saline for at least 15 minutes. Hold the eyelids open while rinsing so the water reaches the eyeball and the inside of the lids, 「切勿揉搓眼睛」 — never rub the eyes. Go to the eye department at once afterwards. High-risk products are identified by their pH. Those with pH<3 are toilet cleaners, descalers, floor cleaners and stainless steel polishes. Some powerful drain unblockers also contain dilute sulphuric acid. Those with pH>11 are drain unblockers, heavy-duty kitchen degreasers, dishwasher powder, drain salts and cleaning powders. There is also a combination that causes harm on the spot: using disinfectant together with a toilet cleaner containing hydrochloric acid gives off chlorine gas, which irritates the airway and causes severe coughing and breathlessness, and can be fatal +- Cost: No money. Put cleaning agents, disinfectants, pesticides and medicines on a high shelf out of a child's reach, or in a cupboard with a lock. Never decant them into a drinks bottle. +- In plain language: Do not induce vomiting after someone swallows a cleaning agent, a pesticide or medicine. Bringing a corrosive liquid back up burns the oesophagus a second time. First give a moderate amount of milk or egg white, then take the bottle and the instructions straight to hospital. If it lands on the skin or in the eyes, flush with plenty of clean water for at least 15 minutes, hold the eyelids open to flush and do not rub. Mixing a disinfectant with a toilet cleaner gives off chlorine gas. +- Benefit: The China CDC Institute for Occupational Health and Poison Control sets out four routes of exposure. Swallowed (gastrointestinal intake): immediately drink a moderate amount of milk or egg white, 「切记，不要催吐」 — remember, do not induce vomiting. Get to hospital as soon as possible and take the packaging or the instructions with you, 「以便于医生快速判断成分并制定救治方案」 — so that the doctor can judge the ingredients quickly and work out a treatment plan. Inhaled (respiratory intake): move the person at once to a well-ventilated area, open the collar and keep the airway clear; with severe coughing, breathlessness or difficulty breathing, get medical help at once. On the skin: flush the area of contact with plenty of clean water for at least 15 minutes, taking care that the runoff does not splash into the eyes or onto other intact skin. In the eyes: flush the eyes with plenty of clean water or saline for at least 15 minutes. Hold the eyelids open so the water reaches the eyeball and the inside of the lids, 「切勿揉搓眼睛」 — never rub the eyes. Go to the eye department at once afterwards. Identify high-risk products by their pH. A pH<3 product is a toilet cleaner, a descaler, a floor cleaner or a stainless steel polish. Some powerful drain unblockers also contain dilute sulphuric acid. A pH>11 product is a drain unblocker, a kitchen degreaser, washing powder, blasting salt or scouring powder. There is also a combination that causes harm on the spot: a disinfectant used together with a toilet cleaner containing hydrochloric acid gives off chlorine gas, which irritates the airway, causes severe coughing and breathlessness, and can kill - Evidence grade: B -- Sources: China CDC Institute for Occupational Health and Poison Control (中国疾病预防控制中心职业卫生与中毒控制所) (2025). Holiday safety reminder: beware of strongly acid and strongly alkaline cleaning agents. -- Notes: The B is because the official explainer says only how to handle it and how to recognise it, and gives no figure for how many fewer incidents following it would prevent. The reason not to induce vomiting is that a corrosive liquid coming back up burns the oesophagus, the tube from throat to stomach, a second time. If you cannot tell what was swallowed, take the bottle, the instructions and any vomit to hospital together. Prevention costs far less than first aid: store things by category, keep them away from food and drink, screw the cap back on and throw the empty bottle away, and never pour anything into a mineral water bottle. Wear rubber gloves and goggles when using these products, open the window, and never mix types. For carbon monoxide see item 19, and for burns and scalds see item 14. +- Sources: China CDC Institute for Occupational Health and Poison Control (中国疾病预防控制中心职业卫生与中毒控制所) (2025). Summer holiday safety reminder: beware of strongly acidic and strongly alkaline cleaning agents. +- Notes: The B is because the official explainer only says how to handle it and how to recognise it, and gives no figure for how much less goes wrong if you follow it. The reason not to induce vomiting is that a corrosive liquid coming back up burns the oesophagus (the tube from the throat to the stomach) a second time. If you are not sure what was swallowed, take the bottle, the instructions and the vomit to hospital. Prevention is far cheaper than first aid: store by category, do not mix with food or drink, screw the cap shut when you have finished and throw the empty bottle away, and never pour the product into a mineral water bottle. Wear rubber gloves and goggles when you use it, open the window, and do not mix types. For carbon monoxide see item 19 in this section, and for burns and scalds see item 14 in this section. For chemicals such as acids and alkalis landing on the body, how to take the clothing off and how long to flush, see item 21 in this section (chemicals splashed on the body). -### 21. Acids, alkalis and other chemicals splashed on you: take off the contaminated clothing at once, rinse with plenty of running water, hold the eyelids open in the eye, and rinse for the full time before you go +### 21. Acids, alkalis and other chemicals splashed on the body: take off the contaminated clothing at once, flush with plenty of running clean water, hold the eyelids open to flush the eyes, and do not leave until the full time is up -- Cost: No money. On your first day at work, find where the eyewash station and the emergency shower are. Do not wait until something happens and then go looking. -- In plain language: Three steps. First, take off the contaminated clothes, shoes and socks at once, because the parts held against the body are the worst hurt. Second, rinse with plenty of running water, at least 15 minutes for both skin and eyes, and the Hunan Provincial Department of Emergency Management gives 20 to 30 minutes, so take the longer figure. Third, hold the eyelids open while rinsing the eye, and go straight to the eye department when you are done. Use water at room temperature, not hot water. -- Benefit: Three steps on site. First, immediately take off the contaminated clothes, shoes and socks: the chemical keeps being absorbed through the skin, and the parts held against the body are the worst hurt. Second, rinse with plenty of running water. The floor given by the China CDC Institute for Occupational Health and Poison Control is at least 15 minutes each for the skin and the eyes. The figure given by the Hunan Provincial Department of Emergency Management is 20 to 30 minutes for rinsing the wound, 「强烈的化学品要更长」 — strongly caustic chemicals need longer. Take the longer of the two figures. Third, in the eye you must 「撑开眼睑，确保水流充分接触眼球及眼睑内侧」 — hold the eyelids open and make sure the water reaches the eyeball and the inside of the lids — and 「切勿揉搓眼睛」 — never rub the eyes — then go at once to the eye department. Use room temperature or slightly warm water; hot water is forbidden. Put no ointment, no red potion and nothing else on a fresh wound, and do not wrap it in a dirty cloth, because whatever you put on it changes the doctor's judgement of the depth. In law, 「化学性皮肤灼伤」 and 「化学性眼部灼伤」 — chemical burns of the skin and of the eye — are both on the statutory list of occupational diseases, and cases arising at work are handled as an occupational disease and a work injury. When an accident involving acute occupational hazards happens, the employer 「应当及时组织救治、进行健康检查和医学观察，所需费用由用人单位承担」 — must organise rescue and treatment in good time, carry out health checks and medical observation, and bear the cost — (nationally) +- Cost: No money. On your first day at work, find out where the eyewash and the emergency shower are. Do not wait until there is an accident to go looking. +- In plain language: Three steps. First, take off at once the clothing and footwear the chemical has soaked into: the parts the clothing held against the body are the worst injured. Second, flush with plenty of running clean water, at least 15 minutes for the skin and for the eyes alike. The Hunan Provincial Emergency Management Department gives 20 to 30 minutes. Follow the longer one. Third, hold the eyelids open and flush the eyes, then go straight to the eye department. Use water at room temperature, not hot water. +- Benefit: Three steps at the scene. First, remove the contaminated clothing and footwear at once: the poison keeps being absorbed through the skin, and the parts the clothing held against the body are the worst injured. Second, flush with plenty of running clean water. The floor the China CDC Institute for Occupational Health and Poison Control gives is "at least 15 minutes" for the skin and for the eyes each. The figure the Hunan Provincial Emergency Management Department gives for flushing the wound surface is 20 to 30 minutes, "stronger chemicals need longer". Take the longer of the two. Third, "hold the eyelids open and make sure the water reaches the eyeball and the inside of the lids", "never rub the eyes", and go to the eye department at once afterwards. Use water at room temperature or slightly warm; hot water is forbidden. Put no ointment, no red potion and nothing else on a fresh wound, and do not wrap it in a dirty cloth: whatever goes on it stops the doctor judging how deep it goes. Legally, "chemical skin burn" and "chemical eye burn" are both on the statutory list of occupational diseases, and cases arising at work are handled as occupational disease and work injury. Where an accident involving acute occupational hazards happens, the employer "shall organise treatment in good time, carry out health checks and medical observation, and bear the costs required" (nationwide) - Evidence grade: B -- Sources: China CDC Institute for Occupational Health and Poison Control (中国疾病预防控制中心职业卫生与中毒控制所) (2025). Holiday safety reminder: beware of strongly acid and strongly alkaline cleaning agents. ; Hunan Provincial Department of Emergency Management (湖南省应急管理厅) (2015). Emergency handling and treatment of common chemical poisoning accidents. ; National Health Commission (国家卫生健康委) and three other departments (2024). Classification and catalogue of occupational diseases (in force from 2025-8-1). ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Law on the Prevention and Control of Occupational Diseases, 第三十七条. (reprinted by the Beijing Municipal People's Government) -- Notes: The B is because on the question of how long to rinse, one official document says 15 minutes and another says 20 to 30 minutes, and there is no single national standard figure. Take the chemical's name, its packaging or that safety data sheet (SDS) to hospital, because the doctor needs it to decide on a plan. Do not grab a different acid or alkali to 「中和」 — neutralise — the exposure: mixing the two gives off heat, and there is no way to control how much goes in. The 「中和液」 — neutralising solution — mentioned in the Hunan document means a special rinsing fluid prepared in advance by the workplace, not something you mix up on the spot from whatever is to hand. Brush dry powders such as lime powder and cement powder off first, then rinse with water (some emergency pages say concentrated sulphuric acid should be wiped dry first for the same reason; the guidance differs from place to place, so this book does not say it). For household cleaning agents and swallowed products see item 20. For protection before you start work and for medical checks, see items 10 and 11 in section 19 (the hazard notice before you start, and the protective equipment the employer must supply). +- Sources: China CDC Institute for Occupational Health and Poison Control (中国疾病预防控制中心职业卫生与中毒控制所) (2025). Summer holiday safety reminder: beware of strongly acidic and strongly alkaline cleaning agents. ；Hunan Provincial Emergency Management Department (湖南省应急管理厅) (2015). Emergency handling and treatment of common chemical poisoning accidents. ；National Health Commission and three other departments (国家卫生健康委等四部门) (2024). Classification and catalogue of occupational diseases (in force from 2025-8-1). ；Standing Committee of the National People's Congress (全国人大常委会) (2018). Law on the Prevention and Control of Occupational Diseases, 职业病防治法第三十七条. (reprinted by the Beijing Municipal People's Government) +- Notes: The B is because on how long to flush, one official document says 15 minutes and another says 20 to 30 minutes, and there is no single national standard figure. Take the name of the chemical, the packaging or the safety data sheet (SDS) to hospital: the doctor needs it to decide on a plan. Do not grab some other acid or alkali to "neutralise" it: the two give off heat when mixed, and there is no way to control how much you pour. The "neutralising solution" in the Hunan document means a purpose-made rinsing fluid the workplace prepared in advance, not something you hold against it at the scene. Brush dry powders such as lime powder and cement powder off first and flush afterwards (some local emergency pages say that even concentrated sulphuric acid should be wiped dry first; the wording differs from place to place, so this book does not say it). For cleaning agents at home and accidental swallowing, see item 20 in this section. For protection before you start work and the medical check-up, see items 9 and 10 in section 19 (hazard information before you start work, and protective equipment the unit must supply). ### 22. Dizzy, nauseated, not sweating or not fully conscious in high heat: move to shade at once, strip and pour water over the person to cool them. Give an unconscious person nothing to drink, and call 120 @@ -199,7 +189,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: US National Weather Service. Heat Related Illnesses. ; Yunnan Provincial CDC (云南省疾控局) (2026). With these symptoms you may have heatstroke, 出现这些症状，你可能是中暑了. ; China CDC. In hot weather, China CDC reminds you to guard against heatstroke: 「如有条件可用冷水浸泡的方法对无基础疾病的青壮年快速降温。老人和有基础疾病的患者可向皮肤喷水通过蒸发降温。降温应遵循"先降温，后转运" 的原则」 — where it is possible, healthy young and middle-aged adults with no underlying conditions can be cooled quickly by cold water immersion. Elderly people and patients with underlying conditions can be sprayed with water and cooled through evaporation. Cooling should follow the principle of "cool first, transport second". ; National Health Commission (国家卫生健康委) (2010). Key points for preventing heatstroke in high temperatures. ; Eifling KP et al (2024). Wilderness Medical Society Clinical Practice Guidelines for the Prevention and Treatment of Heat Illness: 2024 Update. Wilderness & Environmental Medicine 35(1_suppl):112S-127S. (the text site refuses automated fetching, so the specific recommendations were checked against the update summary on the Wilderness Medical Society's own site: 「cold water immersion … is the most important method of treating severe heat illness」「If CWI is unavailable, evaporative cooling is the recommended second-line treatment. … generally half as fast as CWI」) ; Lipman GS et al (2019). Same guidelines, 2019 version. - Notes: Do not wait to see if it 「缓一缓」 — eases off a little. Heat stroke can go from dizziness to coma in a few dozen minutes. Do not pour water into an unconscious person, because it goes into the airway. Where to put the ice packs, the two sources differ. China CDC says the neck, the armpits and the groin. The Wilderness Medical Society says the research does not show any cooling effect from those sites. Its advice is that where ice packs are all you have, cover the whole body with them, and where there are not enough, put them on the cheeks, the palms of the hands and the soles of the feet. Both say to use cold water immersion where it is possible. - ### 23. Anyone with heatstroke who is conscious: small sips of cool dilute salt water or a sports drink, often and little. Do not gulp, and do not drink alcohol or high-sugar drinks - Cost: No money. A sports drink costs a few yuan a bottle. @@ -217,24 +206,23 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: US FEMA Ready.gov. Home Fires. ; Ministry of Emergency Management (应急管理部) (2019). How to get yourself out of a fire in a public place: 「逃生时不要乘坐电梯，当火势不大时，要尽量往楼层下面跑」「可用湿毛巾或口罩蒙鼻，匍匐或弯腰撤离」 — when escaping, do not take the lift; when the fire is not severe, run towards the lower floors as far as you can; you can cover the nose with a wet towel or a mask, and crawl or bend low as you leave. ; Guizhou Provincial Fire and Rescue Corps (贵州省消防救援总队) (2020). Everyday fire knowledge: three points for escaping, 消防常识丨逃生三要点: 「湿毛巾对折3次（叠成8层），除烟率比较高，但烟和毒气浓度过高时，过滤效果就无法保证了。另外，毛巾过湿会导致呼吸困难」 — a wet towel folded 3 times (into 8 layers) filters smoke well, but when smoke and toxic gas are too thick to allow any filtering, and a towel that is too wet makes breathing difficult. - Notes: Do not take the lift, because a power cut traps you inside. Do not go back for property. Do not run upwards unless you have confirmed the roof door is open. Do not jump. Fold a wet towel 3 times into 8 layers and do not put too much water in it, because a towel that is too wet makes breathing harder. When the smoke and toxic gas are too thick, a towel will not save you. - -### 25. You see someone drowning: call for help and dial 110/120, reach in with a pole, throw something that floats, and do not go into the water yourself +### 25. You see someone drowning: shout for help and call 110/120, hand over a pole and throw something that floats, and do not go into the water yourself -- Cost: No money. The hard part is holding back the urge to jump in. -- In plain language: A drowning person pushes the rescuer under by instinct, so whoever goes in may drown too. The order is: shout for help → call 110/120 → pass over a long pole or a rope → throw a life ring, an empty bucket or a plastic bottle → when the person is out and not breathing, start compressions at once. Do not form a chain of hands and go into the water. About 300,000 people drown worldwide every year, and it is the fourth leading cause of death in children aged 1 to 4. -- Benefit: WHO: about 300,000 people drown worldwide every year, 92% of them in low- and middle-income countries. Drowning is the fourth leading cause of death in children aged 1 to 4 and the third in those aged 5 to 14. The intervention WHO recommends is 「安全救援和复苏培训」 — safe rescue and resuscitation training — which means learning first the way to help without going into the water. The order is: shout for help → make the call → pass over a long pole or a rope and throw a life ring, an empty bucket or a plastic bottle → when the person is out and not breathing, start compressions at once +- Cost: No money. The hard part is holding back the jump. +- In plain language: A drowning person instinctively drags the rescuer under, and anyone who goes into the water may drown too. The order is: shout for help → call 110/120 → hand over a long pole or a rope → throw a life ring, an empty bucket or a plastic bottle → if the person is ashore and not breathing, start compressions at once. Never form a chain of people holding hands and go in. About 300,000 people drown every year worldwide, and it is the fourth leading cause of death among children aged 1 to 4. +- Benefit: WHO: about 300,000 people drown every year worldwide, 92% of them in low- and middle-income countries. Drowning is the fourth leading cause of death among children aged 1 to 4 and the third among those aged 5 to 14. The intervention the WHO recommends is "safe rescue and resuscitation training", which means learning the ways to help without going into the water. The order is: shout for help → make the call → hand over a long pole or a rope, and throw a life ring, an empty bucket or a plastic bottle → if the person is ashore and not breathing, start compressions at once - Evidence grade: C -- Sources: WHO. Drowning fact sheet, 溺水 实况报道. -- Notes: The C is because no official text could be found for the rule "do not go into the water". There is only WHO's death figures and WHO's direction of "safe rescue". Do not link arms and form a chain into the water. If you have not been trained, do not try to drag someone out, because a drowning person pushes the rescuer under. "Do not go into the water" is also a rule in your own interest: the person who goes in drowns too, whoever it happens to be. The person you are most likely to save is your own child, and item 12 in section 1 (keeping a child in sight near water) covers how to prevent it. +- Sources: WHO. Drowning fact sheet. +- Notes: The C is because no official text could be found for the rule "do not go into the water yourself"; there is only the WHO death figures and the direction of "safe rescue". Never form a chain of people holding hands and go in. If you are not trained, do not go in after someone: a drowning person drags the rescuer under. "Do not go into the water yourself" is a rule in your own interest too: the person who goes in may drown, and it makes no difference who was in the water. If you do have to rescue someone, it will most often be your own child; for prevention see item 12 in section 1 (keep children in sight whenever they are near water). Where someone has fallen through the ice, do not run out onto the ice to pull them out either; for how to handle ice see item 41 in section 1 (bore a hole and measure the thickness before you step onto outdoor ice). -### 26. Someone is choking and cannot speak: stand behind them and give 5 back blows followed by 5 abdominal thrusts, and if they collapse start cardiopulmonary resuscitation +### 26. Someone is choking and cannot speak: stand behind them, give 5 back blows and 5 abdominal thrusts, and if they go down start cardiopulmonary resuscitation -- Cost: No money. Look at the movement once: make a fist above the navel, put your other hand over it, and thrust sharply inwards and upwards. -- In plain language: First ask 「你噎住了吗，能说话吗」 — are you choking, can you speak. If they can still cough hard and make a sound, do nothing and let them cough. If they cannot speak, stand behind them and give 5 back blows followed by 5 abdominal thrusts, repeating until the object comes out or the person collapses. If they collapse, start cardiopulmonary resuscitation and send someone to call the emergency number. For a pregnant woman or a very large person, thrust on the chest instead. Do not reach into the mouth for something you cannot see. -- Benefit: The US National Library of Medicine text says: first ask 「你噎住了吗，能说话吗」 — are you choking, can you speak. 「能用力咳嗽、能说话的不要动手」 — if they can cough hard and can speak, do not intervene, because coughing itself usually brings the object up. If they cannot speak, 「反复做 5 次拍背接 5 次腹部冲击，直到异物出来或人失去意识」 — repeat 5 back blows followed by 5 abdominal thrusts until the object comes out or the person loses consciousness. 「失去意识就开始心肺复苏」 — once they lose consciousness, start cardiopulmonary resuscitation — and send someone to call the emergency number. For a pregnant woman or someone with obesity, do chest thrusts instead +- Cost: No money. Watch the movement once: make a fist above the navel, cover it with the other hand, and thrust sharply inwards and upwards. +- In plain language: First ask 「你噎住了吗，能说话吗」 — are you choking, can you speak. If they can still cough hard and make a sound, do nothing and let them cough. If they cannot speak, stand behind them, give 5 back blows followed by 5 abdominal thrusts, and repeat until the object comes out or the person goes down. If they go down, start cardiopulmonary resuscitation and get someone to call the emergency number. For a pregnant woman or a very large person, thrust on the chest instead. Do not reach into the mouth for something you cannot see. +- Benefit: The US National Library of Medicine text: first ask 「你噎住了吗，能说话吗」 — are you choking, can you speak. "Anyone who can cough hard and can speak should not be touched", because the cough itself often brings the object out. If the person cannot speak, "repeat 5 back blows followed by 5 abdominal thrusts until the object comes out or the person loses consciousness". "If consciousness is lost, start cardiopulmonary resuscitation", and get someone to call the emergency number at the same time. For pregnant women and obese people, do chest thrusts instead - Evidence grade: B -- Sources: US National Library of Medicine, MedlinePlus. Choking – adult or child over 1 year. ; Fuzhou Municipal Health Commission (福州市卫生健康委员会) (2026). The Heimlich manoeuvre has been updated: 「1岁以上幼儿、儿童及成人：从旧版单一腹部冲击改为"5次背部拍击+5次腹部冲击"的循环」「中国红十字系统急救培训一直都是按"5次背部拍击+5次腹部冲击"循环进行的方式进行教学」 — for toddlers, children and adults over 1 year old: the old single abdominal thrust has been replaced by a cycle of 5 back blows plus 5 abdominal thrusts; first aid training in the Chinese Red Cross system has always taught it as a cycle of 5 back blows plus 5 abdominal thrusts. -- Notes: Do not reach into the mouth for something you cannot see, because you push it deeper. Do not give back blows to someone who can still cough. Choking happens most often at your own dining table, so the people who benefit are mostly your family. The teaching in China is the same set of moves: the Fujian Red Cross emergency first aid training centre says that for children over 1 year and for adults, the Chinese Red Cross system has always taught a cycle of 5 back blows plus 5 abdominal thrusts. Many older tutorials online teach only abdominal thrusts, and that is the old version. +- Sources: US National Library of Medicine, MedlinePlus. Choking – adult or child over 1 year. ；Fuzhou Municipal Health Commission (福州市卫生健康委员会) (2026). The Heimlich manoeuvre has been updated: 「1岁以上幼儿、儿童及成人：从旧版单一腹部冲击改为"5次背部拍击+5次腹部冲击"的循环」「中国红十字系统急救培训一直都是按"5次背部拍击+5次腹部冲击"循环进行的方式进行教学」 — for children over 1 year and for adults, instead of the single abdominal thrust of the old version the cycle is now "5 back blows plus 5 abdominal thrusts"; the Red Cross first aid training system in China has always taught the cycle of "5 back blows plus 5 abdominal thrusts". +- Notes: Do not reach into the mouth for something you cannot see: you push it further in. Do not strike a person who can still cough. Choking most often happens at your own dining table, and the people who benefit are mostly your family. Teaching in China is the same set: the Fujian Red Cross Emergency Rescue Training Centre says that for children over 1 year and for adults the Red Cross system in China has always taught the cycle of 5 back blows plus 5 abdominal thrusts. Plenty of older tutorials online teach only the abdominal thrust, and that is the old version. For a child under 1 year do not strike the abdomen, and strike the chest instead: see item 43 in this section (infant choking). ### 27. Lost in the Gobi, in a desert or in an uninhabited area, or the vehicle has broken down: stay where you are or by the vehicle, report your location first, then get shade and warmth, ration the water and do not wander @@ -244,7 +232,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: C - Sources: US National Park Service. Mojave National Preserve Safety. ; US National Park Service. Death Valley Safety. ; US National Park Service. Desert Driving Safety. ; Chongqing Municipal Emergency Management Bureau (重庆市应急管理局) (2024). Emergency science: the distress signals you need to know in the wild: 「燃放三堆火焰是国际通行的求救信号，将火堆摆成三角形」「三声短三声长，再三声短；间隔1分钟之后再重复」 — lighting three fires is the internationally recognised distress signal, and lay the fires out in a triangle; three short, three long, then three short; repeat after an interval of 1 minute. ; Ministry of Culture and Tourism (文化和旅游部) (2018). The Ministry of Culture and Tourism reminds visitors to choose outdoor adventure travel carefully: 「谨慎前往无人区和人烟稀少区」「尽量多车同行，提前准备对讲机，必要时租用卫星电话」 — go into uninhabited and thinly populated areas with care; travel in as many vehicles as possible, prepare walkie-talkies in advance, and rent a satellite phone if necessary. - Notes: The C is because all of this is official guidance rather than research data. The water quantity, the notification, the signal coverage, staying by the vehicle and the distress signals are all official wording. No official text could be found for the "groups of three" calls that circulate, so they have been rewritten as the official three fires and three short, three long, three short. Do not hike at midday looking for the route. Do not drink urine. Do not cut your drinking to save water. - ### 28. Shaking, slurred speech and unsteady walking mean hypothermia: change the wet clothing, wrap them up, warm the trunk first, give a hot sweet drink and no alcohol - Cost: An aluminium foil survival blanket costs a few yuan, plus one spare dry garment. Learning to read the signs costs nothing. @@ -262,7 +249,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: WHO. Animal bites fact sheet, 动物咬伤 实况报道. ; WHO (2023). Snakebite envenoming fact sheet, 蛇咬伤中毒 实况报道. ; Lai R et al (2024). The Chinese guideline for management of snakebites. World Journal of Emergency Medicine 15(5):333-355: 「An estimated 250,000–280,000 cases of venomous snakebites occur in China each year, affecting mainly patients aged ≥50 years」. ; Yan S et al (2026). Clinical epidemiology of snakebite envenoming in hospitals 11 provinces of Yangtze River Basin and southern regions of China. PLoS Neglected Tropical Diseases: 「withdrawal (4346, 10.65%), and death (23, 0.06%)」. - Notes: Take a photo of the snake if you can, but do not go and catch it. Do not drink alcohol, do not run about, do not suck with your mouth, do not apply ice, and do not tie a rope until it turns purple. China has no official national count of snakebite deaths. The in-hospital death rate in those 11 provinces looks low, but more than a tenth of the patients left before treatment was finished, and what became of them is not in the figures. - ### 30. In an earthquake, first work out what building you are in: in a building built to seismic code, drop where you are. In an old building that will come down, the door is a few steps away, so go out - Cost: No money. Look in advance at how many steps it is out of your own front door, which way you go, and which table is solid. @@ -271,7 +257,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: China Earthquake Administration (produced by the Anhui Provincial Earthquake Administration). Earthquake: hide or run? (explainer video, subtitle transcript), 地震来了，是躲还是跑？. ; Armenian HK, Noji EK, Oganesian AP (1992). A case-control study of injuries arising from the earthquake in Armenia, 1988. Bulletin of the World Health Organization, 70(2), 251-257. ; US FEMA Ready.gov. Earthquakes. ; Mahue-Giangreco M, Mack W, Seligson H, Bourque LB (2001). Risk factors associated with moderate and serious injuries attributable to the 1994 Northridge Earthquake. Annals of Epidemiology, 11(5), 347-357. - Notes:Disputed, and the two conclusions are about two different kinds of building. The China Earthquake Administration itself makes "take cover" conditional on 「符合抗震设防要求的建筑内」 — being inside a building that meets seismic design requirements. It also calls for 「因地制宜」「因人而异」 — adapting to the place and to the person — so you weigh up whether the building resists earthquakes, where you are standing, and whether you can run at all. The Armenian buildings came down whole, so there was nowhere to take cover and going out was the better option. What US FEMA gives for housing in the United States is "drop, cover your head and neck, hold on, and do not run outside", and the injuries there come mainly from falling objects and from falls while moving during the shaking. The Northridge study also warns that reaching out to catch something, or to hold up something that "looks steady", makes the injury worse. So asking only "do I run" is no use. You have to judge whether the building will come down and whether you can reach open ground within seconds. In a building put up to code after 2001, where you have to go down the stairs or through the entrance hall, drop where you are. In a single-storey house, an old brick house, or a self-built house with no seismic design, the door is a few steps away, so go out. The Armenian study compared only survivors who were hurt with survivors who were not, did not count the dead, and does not say whether the effect of storey height was adjusted for. People who can run may well have lived on the ground floor, so read the figures down. In no circumstances jump from a window, take the lift, or go down the stairs while the building is shaking. Once outside, keep clear of tall buildings, overpasses, electricity poles, glass curtain walls and advertising hoardings. - ### 31. Bears, wild boar and a pack of wild dogs: do not run, do not scream, do not stare them down, and back away slowly side-on - Cost: No money. Take a walking pole into the hills, walk in company, and make some noise so the animals know you are coming. The hard part is standing still instead of running. @@ -280,7 +265,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: US National Park Service. Staying Safe Around Bears. ; General Office of the National Disease Control and Prevention Administration (国家疾控局综合司) and General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Working standard for the prevention and handling of rabies exposure (2023 version), 狂犬病暴露预防处置工作规范（2023 年版）. - Notes: The B covers bears only. The rules for wild boar and wild dogs, namely do not run, back away slowly, and get to high ground or put something between you and the animal, are experience, and count as grade C. The official guidance from the National Forestry and Grassland Administration and from local forestry bodies could not be found this time. Do not use food to lure the animal, do not walk up close for a photograph, and do not take a dog into hills where there is wild boar. Bears in China are mostly black bears and brown bears, so read the play-dead and do-not-play-dead advice separately for the two. - ### 32. Caught in a thunderstorm in the wild: get into a vehicle or building with a hard roof, and keep away from ridges, lone trees, open water and metal - Cost: No money. The price is that the trip is interrupted for half an hour or more. The hard part is turning back while the thunder is still far off. @@ -298,7 +282,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: A - Sources: US CDC. Yellow Book, high-altitude travel and altitude illness: 「Any unacclimatized traveler proceeding to a sleeping altitude of ≥2,450 m (≥8,000 ft)—and sometimes lower—is at risk for altitude illness」; rate of ascent, 「500 m (1,650 ft) per night in sleeping altitude once above 3,000 m (9,800 ft)」, with one extra acclimatisation day for every 1000 m of climb; acetazolamide for prevention, 「125 mg twice a day; 250 mg twice a day if >100 kg body weight」; 「If symptoms become worse despite rest or treatment at the same elevation」 descent is required. ; Wilderness Medical Society (2024). Clinical Practice Guidelines for the Prevention, Diagnosis and Treatment of Acute Altitude Illness: 2024 Update. Wilderness & Environmental Medicine. - Notes: What counts is 「睡在多高」 — how high you sleep — not how high you got to during the day. A headache with nausea, unsteady walking, a change in consciousness, or breathlessness even at rest are all signs to come down. Descent takes priority over oxygen and over medicine, and carrying on upwards while breathing oxygen is wrong. - ### 34. A tick has attached itself: pull it straight out with fine-tipped tweezers held against the skin, and put on no oil and use no flame - Cost: A pair of fine-tipped tweezers. @@ -307,7 +290,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: US CDC. After a tick bite: 「Grasp the tick as close to the skin's surface as possible using clean fine-tipped tweezers」「Pull tick away from the skin with steady, even pressure」「Do not use petroleum jelly, heat, nail polish, or other substances to try and make the tick detach from the skin. This may agitate the tick and force infected fluid from the tick into the skin」. ; Ren Q et al (2026). Epidemiological characteristics of severe fever with thrombocytopenia syndrome (SFTS) in China, 2010–2024: a fifteen-year national surveillance analysis. BMC Infectious Diseases 26:1497: 「The number of reported cases (6,602 cases) and deaths (306 cases) in 2024 were both the highest levels since the discovery of this disease」「The average CFR was 4.79%, and the CFR increased with age, reaching 9.15% among those aged 85 years and above」. - Notes: Write down the date when you have pulled it out. If you get a fever, a rash or weakness in the weeks that follow, tell the doctor you were bitten by a tick. If the doctor does not ask and you do not say, it is easily treated as a cold. - ### 35. Drink all water in the wild boiled: a rolling boil for one minute is enough, and there is no need to boil it longer because of the altitude - Cost: Carrying an extra set of cooking gear and the weight of the fuel. @@ -316,7 +298,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: US CDC. Yellow Book, water disinfection for travellers: 「All organisms that may cause illness from drinking water are killed within seconds at boiling temperature」「CDC recommends that travelers boil water for a full minute to account for user variability in identifying boiling points and to add a margin of safety」「Although the boiling point for water decreases with increasing elevation, at common travel elevations the temperature needed to achieve boiling is still well above the temperature required to inactivate enteric pathogens」. - Notes: The same page also says you do not strictly have to bring it to a rolling boil: holding it at 60 ℃ for 30 minutes works just as well, and boiling is simply the end point you can see without a thermometer. If you cannot boil, use a filter or a chemical disinfection tablet; each of the two misses some pathogens, so boil the water if you can. Stream water and snowmelt that look clear can carry parasites just the same, and appearance is not a test. - ### 36. Demanded of money by a stranger in an uninhabited area: give him the money, do not use your hands, note his features, and call the police once you are clear - Cost: The cash you are carrying. The hard part is keeping your temper down. @@ -325,7 +306,6 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: C - Sources: National People's Congress (全国人大) (revised 2020). Criminal Law, 刑法（第二十条、第二百六十三条）. (consolidated text reprinted by the Beijing Municipal Public Security Bureau Traffic Management Bureau) - Notes: The C is because there is no direct literature on the claim that paying and not fighting is safer. The statute only sets out his risk and your boundary. Do not follow him to a secluded place to 「取钱」 — get the money. Do not hand over your phone password. Do not get into his vehicle. Paying him first applies only where you are stopped face to face and your personal safety is threatened, and you should still call the police once you are clear. Being demanded money online with private photos or nude videos is the opposite case, and you must not pay a single yuan. See item 32 in section 8 (being blackmailed with private material). - ### 37. You come across a group of people fighting: back off and walk away. Do not wade in, do not stand and watch, do not pick up the weapons on the ground. If you are going to call the police, move to a safe distance and dial 110 - Cost: No money. The hard part is holding back both impulses: going in to pull them apart, and standing there to watch the show. After you call the police you may be called back to give a statement, which costs at most half a day. @@ -343,33 +323,32 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: Beijing Municipal People's Government. Cannot tell before from after? Six questions and six answers on HIV pre-exposure and post-exposure prophylaxis, 事前、事后分不清？六问六答全面了解艾滋病病毒暴露前/暴露后预防. ; China CDC Center for STD and AIDS Prevention and Control (中国疾病预防控制中心性病艾滋病预防控制中心) (2020). Technical guidelines for HIV post-exposure prophylaxis (trial), 艾滋病病毒暴露后预防技术指南（试用）. - Notes: Where to get it: the designated HIV clinic named by the local CDC, usually under infectious diseases (the sexually transmitted infection and AIDS clinic). Many cities now have 24-hour clinics, and Chongqing's published figures are 29 clinics across 26 districts and counties, of which 14 are open around the clock. Before you go, phone the local CDC to ask which one has the medicine, and do not waste time in the emergency department. What counts as an "exposure" includes sex without protection, a condom that broke, a needlestick, and blood onto a mucous membrane. Whether the medicine is needed is the doctor's call. You have to complete the full 28 days, and missing a few doses makes it clearly less effective. Afterwards, follow the doctor's instructions and come back for checks at 4 to 6 weeks and at 3 months. The blocking medicine is only a remedy. For everyday prevention and testing, see items 30 and 31 in section 1 (use a condom throughout, and get tested if you have had higher-risk sex). - -### 39. You were hurt rescuing someone and put money in: claim against the person who caused it and against medical insurance first, then apply to have the rescue recognised +### 39. You are hurt or out of pocket after a rescue: go after the person who caused it and after medical insurance first, then apply for official recognition as someone who acted for justice -- Cost: No money. Applying means handing in documents and waiting for a decision, and making several trips, over a few weeks to a few months. -- In plain language: If rescuing someone has cost you, you do not have to write it off as bad luck. Someone injured while doing a righteous act is treated first and billed afterwards. The medical bill is paid by the person who caused the harm, and where that person has run off or cannot pay, it goes through basic medical insurance. Where there is no person to blame, the person you rescued is legally obliged to give you appropriate compensation. Apply at the same time to have the rescue recognised as a righteous act, and if you were injured, have the work injury insurance benefits applied. -- Benefit: Article one hundred and eighty-three of the Civil Code: 「因保护他人民事权益使自己受到损害的，由侵权人承担民事责任，受益人可以给予适当补偿。没有侵权人、侵权人逃逸或者无力承担民事责任，受害人请求补偿的，受益人应当给予适当补偿」 — where a person suffers damage in order to protect the civil rights and interests of others, the tortfeasor bears the civil liability and the beneficiary may give appropriate compensation. Where there is no tortfeasor, or the tortfeasor has absconded or is unable to bear the civil liability, and the victim requests compensation, the beneficiary shall give appropriate compensation. Guobanfa [2012] No. 39 sets out what happens to someone injured while doing a righteous act. Medical institutions have to set up a fast track and 「坚持『先救治、后收费』的原则」 — follow the principle of treating first and charging afterwards. The wording on the medical bill is 「因紧急救治发生的医疗费用，有加害人或责任人的，由加害人或责任人承担；无加害人或责任人以及加害人或责任人逃逸或者无力承担的，按规定通过基本医疗保障制度解决」 — medical costs arising from emergency treatment are borne by the person who caused the harm or is responsible where there is one; where there is no such person, or that person has absconded or is unable to pay, they are resolved through the basic medical insurance system as prescribed. In a death, 「符合烈士评定条件的，依法评定为烈士」 — where the conditions for being honoured as a martyr are met, the person is honoured as a martyr in accordance with the law; 「属于视同工伤情形的，享受一次性工亡补助金以及相当于本人 40 个月工资的遗属特别补助金」 — where the case is treated as a work injury, the family receives the one-off work-related death benefit plus a special benefit to the family equivalent to 40 months of the person's wages. In a disability, 「符合享受工伤保险待遇条件的，依据《工伤保险条例》落实相应待遇；不符合的，由民政部门评定伤残等级并落实相应待遇」 — where the conditions for work injury insurance benefits are met, the corresponding benefits are applied under the Regulations on Work-Related Insurance; where they are not, the civil affairs department assesses the level of disability and applies the corresponding benefits. On work, 「优先纳入就业援助，予以重点支持」 — priority is given in employment assistance, with focused support — and public welfare posts 「优先安排符合条件的见义勇为人员」 — give priority to people who qualify for recognition as having acted righteously. On housing, 「优先纳入住房保障体系，优先配租、配售保障性住房或发放住房租赁补贴」 — priority is given in the housing security system, with priority for rented or sold affordable housing, or a housing rental subsidy +- Cost: No money. Applying means handing in documents, waiting for a decision and making several trips: weeks to months. +- In plain language: Putting yourself into a rescue does not have to mean writing the loss off. Where the person acting for justice is injured, treatment comes first and payment after. The medical bill is paid by the person who caused the harm; if they have run away or cannot pay, it goes through basic medical insurance. Where there is no person who caused the harm, the person you rescued is required by law to give you appropriate compensation. Apply for official recognition at the same time: where you were injured, the work injury insurance benefits apply. +- Benefit: Article 183 of the Civil Code: "where a person suffers damage while protecting the civil rights and interests of another, the tortfeasor bears the civil liability, and the beneficiary may give appropriate compensation. Where there is no tortfeasor, or the tortfeasor has absconded or is unable to bear the civil liability, and the victim requests compensation, the beneficiary shall give appropriate compensation". 国办发〔2012〕39 号 sets out what happens to people injured while acting for justice. Medical institutions must set up a green channel and "adhere to the principle of treating first and charging afterwards". The text on medical costs reads: "medical expenses arising from emergency treatment are borne by the tortfeasor or the person responsible where there is a tortfeasor or a responsible person; where there is no tortfeasor or responsible person, or where the tortfeasor or responsible person has absconded or is unable to bear it, they are settled through the basic medical insurance system as prescribed". Where the person dies, "those who meet the conditions for being assessed as martyrs shall be assessed as martyrs by law"; where the case counts as a work injury, the dependants receive "a one-off work-death benefit and a special dependant benefit equal to 40 months of the person's wages". Where the person is disabled, "those who meet the conditions for work injury insurance benefits have the corresponding benefits applied under the Regulations on Work-Related Injury Insurance; where they do not, the civil affairs department assesses the disability grade and the corresponding benefits are applied". On employment, they are "given priority for inclusion in employment assistance and given key support", and posts in the public interest "give priority to people who acted for justice and qualify". On housing, they are "given priority for inclusion in the housing support system, given priority in the allocation or sale of affordable housing, or given a housing rental subsidy" - Evidence grade: B -- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一百八十三条）. ; General Office of the State Council forwarding the Ministry of Civil Affairs and other departments (国务院办公厅转发民政部等部门) (2012). Opinions on strengthening the protection of the rights and interests of people who act righteously (国办发〔2012〕39 号). -- Notes: The B is because what you can get is fixed by the national documents, but how much money there is and how the recognition is decided are set by each province's own regulation on righteous acts, the figures differ from place to place, and this round did not check them province by province. You normally apply to the public security organ of the place where the incident happened, or to the body that handles righteous acts (a foundation). Recognition turns on evidence: the receipt from the police report, the CCTV, the names and telephone numbers of witnesses, and the hospital records, and you keep all of it there and then. If you have to sue, the civil rights of someone who acted righteously already fall within the scope of free legal aid. For the national standard of the one-off work-related death benefit and for how a level of disability converts into money, see items 15 and 16 in section 19. For how to apply for legal aid, see item 3 in section 7. +- Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典第一百八十三条. ；General Office of the State Council, forwarding the Ministry of Civil Affairs and other departments (国务院办公厅转发民政部等部门) (2012). Opinions on Strengthening the Protection of the Rights and Interests of People Who Act for Justice (《关于加强见义勇为人员权益保护的意见》, 国办发〔2012〕39 号). +- Notes: The B is because what you can get is fixed by the national documents, but how much money is paid and how recognition works are settled by each province's own regulations on acting for justice, and the figures differ from place to place; they were not checked province by province this time. You apply to the public security organ where the incident happened, or to the body that handles cases of people who acted for justice (a foundation). Whether you are recognised turns on evidence: the police report receipt, the CCTV at the scene, the names and telephone numbers of witnesses, and the hospital records. Keep them at the time. If you intend to sue, the civil rights of a person who acted for justice already fall within the scope of free legal aid. For the national standard of the one-off work-death benefit and for how a disability grade converts into money, see items 14 and 15 in section 19. For how to apply for legal aid, see item 3 in section 7. -### 40. A knife, a length of steel bar or a shard of glass has gone into the body: press around the object to control the bleeding, and do not pull it out +### 40. A knife, a steel bar or glass has pierced the body: press around the object to stop the bleeding, and do not pull it out -- Cost: No money. The hard part is holding back the instinct to pull it out first and staying with the person until the ambulance comes. -- In plain language: Pulling it out can cause heavy bleeding, and it makes it harder for the doctor to judge how deep the wound goes. Here is what to do: call 120, press around the object with your hand or with a dressing to control the bleeding, and pad the object so it does not move. If organs are showing through a wound in the belly, do not push them back, cover them with a clean moist dressing. If a wound in the chest is bubbling, do not seal it with plastic sheet, because sealing it can collapse the lung. -- Benefit: The International Federation of Red Cross and Red Crescent Societies' International First Aid Guidelines 2025 say: 「急救的人应当压住扎进身体的异物（比如刀）周围，把异物稳住，不要把它拔出来」 — the person giving first aid should press around the object embedded in the body, such as a knife, and hold the object steady without pulling it out. 「不要从腹部或胸部的伤口里拔出任何扎入的异物，拔出来可能造成大出血，也会让医生更难判断伤得有多重」 — do not pull any embedded object out of a wound in the abdomen or chest, because pulling it out can cause heavy bleeding and makes it harder for the doctor to judge how serious the injury is. 「扎在身体里的异物应当稳住；异物随脉搏跳动的，就让它跳，松松地稳住即可」 — an object embedded in the body should be held steady; if it moves with the pulse, let it move, and steady it loosely. 「急救的人不应把露出来的内脏塞回体内」 — the person giving first aid should not push exposed organs back inside. An open chest-wall wound that is not bleeding outwards 「可以敞着、不加敷料，因为把伤口封死可能让空气积在胸腔里，使肺部分或完全塌陷（气胸）」 — can be left open with no dressing, because sealing the wound can let air collect in the chest cavity and partly or completely collapse the lung, which is a pneumothorax. Every penetrating wound of the chest or abdomen needs emergency help called at once +- Cost: No money. The hard part is holding back the instinct to pull it out first, and staying with the person until the ambulance comes. +- In plain language: Pulling it out can cause severe bleeding, and it makes the wound harder for the doctor to judge. What to do: call 120, press around the object with your hand or a dressing to stop the bleeding, and pad it so it does not shift. Where organs are showing through a wound in the abdomen, do not push them back; cover them with a clean damp dressing. A wound in the chest that is bubbling must not be sealed over with plastic sheet, because sealing it can collapse the lung. +- Benefit: The 2025 international first aid guidelines of the International Federation of Red Cross and Red Crescent Societies: "the first aider should press around an object embedded in the body, such as a knife, steady the object and not pull it out". "Do not remove any embedded object from a wound in the abdomen or chest, because removal can cause severe bleeding and makes it harder for the doctor to judge how serious the injury is". "An object embedded in the body should be stabilised; if it moves with the pulse, let it move and simply hold it steady". "The first aider should not push exposed organs back into the body". An open chest wall wound that is not bleeding out "may be left open and undressed, because sealing the wound can let air collect in the chest cavity and partially or completely collapse the lung (pneumothorax)". All penetrating chest and abdominal wounds need emergency help called at once - Evidence grade: B - Sources: International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, pages 256–257 and 267–268. -- Notes: The B is because this is common practice in the guidelines, with no randomised trial behind it, and you cannot divide the injured into a group you pull from and a group you do not. The only exception is when the object is blocking the airway and the person cannot breathe. At the hospital, go straight to the desk by the emergency department door that sorts cases by severity (the triage desk) and say 「穿透伤」 — penetrating injury. That is the level that goes straight into the resuscitation room, so do not queue at the registration window (see item 8 in section 24). If something has really gone through, call 120, and the ambulance will warn the hospital to prepare. For pressing a heavy bleed and for putting on a tourniquet, see item 12. +- Notes: The B is because this is common practice in the guidelines, with no randomised trial: nobody could split the injured into a "pull" group and a "do not pull" group to test it. The one exception is an object blocking the airway, where the person cannot get enough air. At the hospital go straight to the desk by the emergency department entrance that sorts cases by severity, the triage desk, and say "penetrating injury". That is the category that goes straight into the resuscitation room; do not queue at the registration window (see item 7 in section 24). Where something really has gone through, call 120, and the ambulance will warn the hospital in advance so it can prepare. For how to press down on severe bleeding and how to put on a tourniquet, see item 12 in this section. -### 41. If you suspect a fracture, do not move that part again: support it and keep it still, and apply cold for no more than 20 minutes. Do not straighten it yourself +### 41. If you suspect a fracture, do not move that part again: support it and keep it still, apply cold for no more than 20 minutes, and do not straighten it yourself -- Cost: No money. Find a board, a rolled-up magazine or clothing to use as support. Put a layer of cloth between the cold thing and the skin. -- In plain language: Treat any injury to a limb as a possible fracture to begin with. Being able to stand and being able to move do not rule it out, and only an X-ray settles it. Keep the injured part still, holding it with your hand, or support it where it already is with a triangular bandage or a board. Apply cold as soon as you can, for no more than 20 minutes at a time. Do not straighten it yourself. Where broken bone has come through the skin, control the bleeding and cover it with a clean dressing, and do not push the bone back in. -- Benefit: The International Federation of Red Cross and Red Crescent Societies published the International First Aid Guidelines 2025. It says that 「手脚上的任何一处伤，都应当先按可能的骨折来处理」 — any injury anywhere on a limb should first be treated as a possible fracture. It also says 「确诊骨折唯一准确的办法是拍 X 光。拿不准就当骨折办」 — the only accurate way to confirm a fracture is an X-ray, and where you are not sure, treat it as a fracture. The purpose of immobilising it is 「尽量减少活动和疼痛」 — to reduce movement and pain as far as possible. Applying cold, 「一次最多连续 20 分钟，这样可以减轻疼痛；冷得比这更久可能损伤皮肤」 — for no more than 20 minutes at a time, which eases the pain; staying cold for longer than that can damage the skin. 「伤到股骨这类长骨时，光是骨头本身出的血就可能致命」 — when a long bone such as the femur is injured, the bleeding from the bone alone can be fatal. Emergency help should be called where the pain or swelling is marked, where there are signs of shock, where there is a fracture of the femur (thigh bone), or where the limb is clearly in the wrong position. For a fracture that is bent out of line, both of two conditions have to hold before you consider straightening it on the spot. The first is that the place is remote or the facilities are limited, and the limb below the bend is cold and pale, or the person giving first aid cannot immobilise it well enough to move it. The second is that the rescuer is trained. For an open fracture, where bone has come through the skin, the priorities are controlling the bleeding and preventing infection +- Cost: No money. Find a hard board, a rolled-up magazine or a bundle of clothing to use as a splint. Put a layer of cloth between the skin and anything cold. +- In plain language: Treat every injury to an arm or a leg as a fracture until proven otherwise. Being able to stand and to move does not rule it out; only an X-ray can settle it. Keep the injured part still, supporting it with your hand, or hold it in the position it was in with a triangular bandage or a hard board. Apply cold early, no more than 20 minutes at a time. Do not straighten it yourself. Where the broken bone has broken through the skin, stop the bleeding and cover it with a clean dressing, and do not push the bone back in. +- Benefit: The International Federation of Red Cross and Red Crescent Societies published its 2025 international first aid guidelines. They say that "any injury to a limb should be treated as a possible fracture". They also say that "the only accurate way to confirm a fracture is an X-ray; where you are not sure, treat it as a fracture". The purpose of splinting is to "reduce movement and pain as far as possible". Cold is for "no more than 20 minutes at a time, which reduces the pain; staying cold for longer than this can damage the skin". "Where a long bone such as the femur is injured, the bleeding from the bone itself can be fatal". Call for emergency help where the pain or swelling is marked, where there are signs of shock, where the femur (thigh bone) is fractured, or where the limb is clearly in the wrong position. Where a fracture is bent out of line, straightening it on the spot (reduction) is only considered when two conditions are both met. First, the place is remote (or the conditions are limited) and the limb below the bend is cold and pale, or the first aider cannot splint it well enough to move the person. Second, the rescuer is trained. The priority in an open fracture (where the broken bone has broken through the skin) is stopping the bleeding and preventing infection - Evidence grade: B - Sources: International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, pages 301–302. -- Notes: The B is because all of this is common practice. Traction splinting of a femur (thigh bone) fracture before hospital rests on a single study that went back through old records, and the reliability of that evidence is very low. The result was less blood transfused within 24 hours and a shorter stay, and the difference in deaths and complications is too small to make out. As for whether to reduce the fracture first, putting the bone back where it belongs, or to immobilise it as it is, no evidence was found at all. Three common mistakes need fixing: being able to stand and walk does not mean there is no fracture; use only cold and not heat, and do not alternate between them; and a sprain does not get better by "working it out", so rest first. For the rule on not moving an elderly person when a fracture is suspected after a fall, see item 2. For a swollen leg after a plaster cast or a period in bed, watch for a clot and see item 11. For whether to have a disability assessment after treatment, and whether to apply for a disability certificate, see items 10 and 11 in section 24. +- Notes: The B is because all of this is common practice. Putting traction and a splint on a femur (thigh bone) fracture before the hospital rests on one study of old medical records, and the reliability of that evidence is very low. The result was less blood transfused within 24 hours and a shorter stay in hospital, with the difference in deaths and complications too small to make out. Whether to reduce first (put the bone back where it was) or to fix it as it is, no evidence was found at all. Three common misunderstandings have to go: being able to stand and walk does not mean there is no fracture; use cold and not heat, and do not alternate between them; a sprain does not "work itself out with movement", so rest first. For the rules on moving an elderly person who falls and may have a fracture, see item 2 in this section. For warning of a clot where one leg swells after a cast or a period in bed, see item 11 in this section. For whether to have a disability assessment and whether to get a disability certificate after treatment, see items 9 and 10 in section 24. ### 42. After a sexual assault: get somewhere safe and call 110. Before the examination do not wash, do not wash your clothes and do not tidy the room, and get to hospital within 72 hours @@ -379,3 +358,21 @@ This section only covers actions that take minutes. Recognising a scam, or being - Evidence grade: B - Sources: Beijing Municipal People's Procuratorate (2024). Through the law and through drama: watching the series 九部的检察官, The Ninth Inspector, how should minors facing unlawful infringement protect themselves? (reprinted from China National Radio's China Voice), 法眼看剧丨看《九部的检察官》 未成年人面对不法侵害该如何保护自己？（转载中央广播电视总台中国之声）. ; Guizhou Provincial Department of Civil Affairs (2024). A handbook on preventing sexual assault of minors, for parents and children to read (reprinted from the People's Daily), 未成年人防性侵手册，请家长孩子查收！（转载人民日报）. ; Beijing Xicheng District People's Procuratorate (2018). Why they no longer stay silent, 她们为什么不再沉默……. ; World Health Organization (2021). Emergency contraception (fact sheet), 紧急避孕（实况报道）. ; Wuhan Municipal Health Commission (2020). Safe and sensible use of emergency contraceptives, 安全合理使用紧急避孕药. ; Beijing Municipal People's Government. Cannot tell before from after? Six questions and six answers on HIV pre-exposure and post-exposure prophylaxis, 事前、事后分不清？六问六答全面了解艾滋病病毒暴露前/暴露后预防. - Notes: The B is because these are official practice and prescribing advice, and there is no figure for how much higher the conviction rate becomes after following it. The people who benefit are mostly you and your family. The 72 hours for emergency contraception is the figure on the Chinese package insert; the World Health Organization says it works within 5 days, but the later you take it the worse it does, so do not delay. If you are already pregnant, emergency contraception does nothing. Tell the truth when you report it, hiding nothing and exaggerating nothing. If the person tries to settle it privately afterwards, offers you money or threatens you, keep a record of it and do not delete it. You can also think it over before reporting it, but the evidence disappears with time. Whether or not you report it, get to hospital inside the time limit. For recording, see item 41 in section 8 (recording). Where the person afterwards uses photographs, or the police report, to pressure you, that is a crime as well. See item 32 in section 8 (being blackmailed with a police report or photographs). If this happens to a friend, go with them to the police and to hospital, and do not start by telling them to take a shower and lie down. + +### 43. An infant under 1 year is choking and cannot cry out: lay the child face down and give 5 back blows, turn the child over and give 5 chest thrusts, alternate the two, and do not thrust on the abdomen + +- Cost: No money. Watch the movement once, or sit a first aid course for infants: minutes to half a day. The hard part is getting the child turned over and face down in the middle of the panic. +- In plain language: If the child can still cough and cry out, let the child cough and watch. If the child cannot cry out and the face is turning blue, get someone to call 120, lay the child face down along your forearm, and give 5 back blows with the heel of your hand between the two shoulder blades. If nothing comes out, turn the child over and give 5 quick thrusts in the middle of the chest. Alternate the two, and do not thrust on the abdomen. If the child stops responding, start cardiopulmonary resuscitation instead. +- Benefit: The American Heart Association and the American Academy of Pediatrics published their 2025 pediatric basic life support guidelines. The guidelines say that suffocation, such as an object blocking the airway, "is one of the leading causes of death in infants and children". What infants most often choke on is liquid. A child who is still coughing and still making a sound has a mild obstruction: let the child cough, and watch whether it gets worse. A weak cough or no cough, no cry, and a bluish face means a severe obstruction. With a severe obstruction, call for emergency help first, then "repeat 5 back blows followed by 5 chest thrusts until the object comes out or the infant stops responding". The guidelines do not recommend abdominal thrusts on infants, because this can injure the organs inside the abdomen. Chest thrusts now use the heel of one hand, because it presses deeper than the old two-finger method. When an infant stops responding, start cardiopulmonary resuscitation, with compressions first and no pulse check. Remove any object you can see before giving breaths. The guidelines state plainly: "do not reach blindly with a finger". The 2025 international first aid guidelines of the International Federation of Red Cross and Red Crescent Societies say the same thing. First give up to 5 firm back blows between the two shoulder blades; if that does not work, turn the infant over and give up to 5 chest thrusts; if that does not work either, call for emergency help at once. It presses on the chest by encircling the chest with both hands and using the two thumbs. The Fujian Red Cross Emergency Rescue Training Centre says that for infants under 1 year the method is "5 back blows + 5 chest thrusts". For the back blows, lay the infant face down on your forearm with the head lower than the body +- Evidence grade: B +- Sources: Joyner BL et al (2025). Part 6: Pediatric Basic Life Support: 2025 American Heart Association and American Academy of Pediatrics Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 152(suppl 2):S424–S447. , full text at ；International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, pages 238–240. ；Fuzhou Municipal Health Commission (福州市卫生健康委员会) (2026). The Heimlich manoeuvre has been updated. +- Notes: The B is because what supports this set of techniques is mainly case reports, and there is no figure for "how many fewer deaths if you follow it". The benefit is graded large on judgement: an airway that is completely blocked stops the child's heart within minutes. Almost all the beneficiaries are your own children. For children over 1 year and for adults, thrust on the abdomen instead: see item 26 in this section (choking, back blows plus abdominal thrusts). The technique for thrusting the chest differs from source to source. The US guidelines and the new Chinese Red Cross version use the heel of one hand, the International Red Cross uses two thumbs, and older tutorials use two fingers. The 2025 US guidelines no longer recommend the two-finger method because it does not press deep enough. In an emergency, any one of them beats doing nothing. Even after the object comes out, go to hospital: a fragment may be left in the lung. Keep coins, small toys and things like that out of a child's reach. What to do once the child stops responding, see item 44 in this section (infant cardiopulmonary resuscitation). + +### 44. An infant under 1 year is unresponsive and not breathing normally: put 120 on speakerphone and do infant cardiopulmonary resuscitation, 30 compressions then 2 breaths + +- Cost: No money. Infant cardiopulmonary resuscitation has to be practised once on a course: half a day to a day. The hard parts are breathing into the child's mouth and nose, and being willing to press hard. +- In plain language: If tapping the soles of the feet gets no response and there is no normal breathing, put 120 on speakerphone and do not let your hands stop. Encircle the chest with both hands and press with the two thumbs; where you cannot get both arms round, use the heel of one hand. Press down about 4 cm, at about 2 compressions a second. After every 30 compressions, seal your mouth over the child's mouth and nose and give 2 breaths. An infant's heart usually stops because breathing has failed, and cardiopulmonary resuscitation with breaths saves more of them. +- Benefit: The American Heart Association and the American Academy of Pediatrics published their 2025 pediatric basic life support guidelines. The guidelines say that in infants and children the leading cause of cardiac arrest is still a failure of breathing. Among infants who go into cardiac arrest outside hospital, only 6.6% leave hospital alive. Where an ordinary person comes on an infant or a child who is unresponsive, not breathing normally and showing no signs of life, they "should start cardiopulmonary resuscitation at once", and should not check for a pulse. Compression depth is at least one third of the chest's front-to-back diameter, which for an infant is "about 1.5 inches (4 cm)". The rate is 100 to 120 a minute. Working alone, give 30 compressions to 2 breaths. With two rescuers, give 15 compressions to 2 breaths. On an infant, press with the heel of one hand, or with both hands encircling the chest and the two thumbs. Where you cannot encircle the chest, use the heel of the hand. The two-finger method has been withdrawn because it does not press deep enough. The guidelines also say that for infants and children in cardiac arrest outside hospital, adding breaths to the compressions improves survival. One US registry counted 3900 children under 18 in cardiac arrest outside hospital, of whom 59.4% were infants. Where a bystander did cardiopulmonary resuscitation, the adjusted share leaving hospital alive was 13.2%, against 9.5% where nobody did. Adjusted OR 1.57 (about 57% higher chance of surviving), 95% CI 1.25–1.96 (credible range). Among infants, cardiopulmonary resuscitation with breaths is associated with higher survival, and the children who survive also recover their brains better. Compressions without breaths give a result close to doing nothing. The 2025 international first aid guidelines of the International Federation of Red Cross and Red Crescent Societies say this: to check whether an infant responds, pick the child up or tap the soles of the feet. If you are alone, put the emergency number on speakerphone and call yourself. If there is no phone within reach that you can use while pressing, do 1 minute of cardiopulmonary resuscitation and then go and call. The Capital Institute of Pediatrics has a popular explainer that the Beijing Municipal Health Commission reprinted. It says: 「如果只有1个人在场，先进行2分钟的心肺复苏后呼叫120救护车」 — if only one person is present, first do 2 minutes of cardiopulmonary resuscitation and then call 120 for an ambulance. The compression depth is 「婴儿4cm」 — infant 4cm. For under 1 year, the breaths go in 「口对口鼻」 — mouth to mouth and nose +- Evidence grade: B +- Sources: Joyner BL et al (2025). Part 6: Pediatric Basic Life Support: 2025 American Heart Association and American Academy of Pediatrics Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 152(suppl 2):S424–S447. , full text at ；Naim MY et al (2017). Association of Bystander Cardiopulmonary Resuscitation With Overall and Neurologically Favorable Survival After Pediatric Out-of-Hospital Cardiac Arrest in the United States: A Report From the Cardiac Arrest Registry to Enhance Survival Surveillance Registry. JAMA Pediatrics 171(2):133–141. ；International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, pages 186–188. ；Capital Institute of Pediatrics (首都儿科研究所) (2022). 【儿研所儿医说】Cardiopulmonary resuscitation: first aid knowledge every parent needs, 《心肺复苏：爸妈必备的急救知识》 (reprinted by the Beijing Municipal Health Commission). ；European Resuscitation Council (2025). ERC Guidelines 2025 for the lay public (the chapter on Paediatric Basic Life Support). +- Notes: The B is because the survival figures come from a registry study that only followed records, and because they cover all children together: the abstract does not give the figures for infants alone. The benefit is graded large on the figures for all children: leaving hospital alive rises from 9.5% to 13.2%, about 40% more relatively. Almost all the beneficiaries are your own children. There are three differences from item 1 in this section (pressing hard on the centre of the chest). First, infants need breaths, and compressions without breaths work poorly on infants. Second, you use two thumbs or the heel of one hand and press about 4 cm, rather than stacking both hands. Third, where you are alone and cannot press and call at the same time, press first and call afterwards. How long to press first differs from source to source: the Capital Institute of Pediatrics and the US guidelines use 2 minutes for a single rescuer, and the International Red Cross uses 1 minute. A phone on speakerphone removes the problem. The European Resuscitation Council teaches giving 5 breaths first and then compressing; the US guidelines teach compressions first and breaths after. Both beat doing nothing. Old Chinese tutorials teach two-finger compressions, and the 2025 US guidelines have dropped that technique. Where choking caused it, look in the mouth before giving breaths and remove only an object you can see: see item 43 in this section (infant choking). diff --git "a/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" "b/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" index 1c9fbac5..aa2e56a3 100644 --- "a/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" +++ "b/en/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" @@ -12,7 +12,6 @@ This section measures money and personal information. The moment somebody signs - Evidence grade: A - Sources: Doerfler P, Thomas K, Marincenko M, et al. (2019). Evaluating Login Challenges as a Defense Against Account Takeover. The World Wide Web Conference (WWW '19). - Notes: The same study also found that this kind of challenge sometimes locks the real owner out. 52% of genuine users did not get in on the first attempt, but 97% of them did get in eventually. Start with your email, because most other accounts can be recovered through email - ### 2. Give your email a password of its own, and do not reuse it on any other site - Cost: No money. A password manager stores it so that you do not have to remember it. The hard part is dropping the habit of using one password everywhere. @@ -21,7 +20,6 @@ This section measures money and personal information. The moment somebody signs - Evidence grade: C - Sources: US CISA. Use Strong Passwords. - Notes: If you cannot remember them, use the password manager built into your browser. It keeps the password for each site for you, which is far better than one password everywhere. Do not store passwords in your WeChat favourites or in a notes app - ### 3. Set a lock-screen password on the phone and a PIN on the SIM card - Cost: No money. One setting for the lock-screen password and one for the PIN. @@ -30,7 +28,6 @@ This section measures money and personal information. The moment somebody signs - Evidence grade: C - Sources: the author's own experience, no direct literature - Notes: You set the PIN under 「SIM card lock」 in the phone's settings. The factory code is usually 1234 or 0000. Three wrong attempts in a row and you need the carrier's PUK code to open it. So write that code on paper as soon as you have set the PIN - ### 4. If you lose your phone, work in this order: report the SIM card, lock it remotely, change your passwords, call the police, freeze the bank card - Cost: No money. The whole sequence takes a bit over ten minutes. @@ -39,7 +36,6 @@ This section measures money and personal information. The moment somebody signs - Evidence grade: C - Sources: US FCC. Protect Your Smart Device. ; the order of the steps is the author's own experience ; for replacing an identity card see section 7, and for a loan taken out in your name see the entry on credit records in section 8 - Notes: Write down the customer service numbers of the three carriers in advance: China Mobile 10086, China Unicom 10010, China Telecom 10000. Also note which city your own number was registered in, because the operator will ask. You can call the operator from somebody else's phone and report the card there too - ### 5. If your card is used by someone else, report and freeze it first, then call the police, then claim from the bank: proving 「that you made the payment」 is the bank's duty - Cost: No money. The moment you see anything odd on the card, report it or freeze it at once. Keep the police report, the record of the card report, and every transaction notice the bank sends you. If the card is still with you, make a small enquiry or a small deposit nearby, so you hold a record showing the real card was in your hands at the time. The hard part is holding back the urge to argue with customer service first. Report and freeze first. @@ -48,7 +44,6 @@ This section measures money and personal information. The moment somebody signs - Evidence grade: A - Sources: Supreme People's Court (最高人民法院) (2021). Provisions on Several Issues Concerning the Trial of Civil Disputes over Bank Cards, 关于审理银行卡民事纠纷案件若干问题的规定（第四、五、七、十四、十五条）. - Notes: Two situations leave you carrying the liability. The first is that you did not look after the card, the password and the verification codes properly and were at fault yourself. The original wording is 「未尽妥善保管义务具有过错」, which means you failed to keep them properly and were at fault, and you carry it in proportion to your fault. So do not tell other people your password, and do not forward your verification codes to them (see item 1 in this section, which prefers a prompt on your phone for two-factor verification). The second is failing to report the card in time and letting the loss carry on growing, in which case the extra part is yours. So the first step is always to report and freeze, not to argue with customer service first. The same rules apply to non-bank payment institutions. If their promotional material promises 「先行赔付」, payment before recovery, in specific and definite terms, you can hold them to it and ask them to pay first. If you were tricked into transferring the money yourself, that is a different route: see item 2 in section 8 (if you spot the fraud, call 110 or 96110 at once and ask for the payment to be stopped). - ### 6. Now and then look at the signed-in devices and authorised apps on your accounts, and clear out the ones you no longer use - Cost: No money. Each time takes a few minutes. The hard part is that nobody reminds you, so you have to remember to look. @@ -57,7 +52,6 @@ This section measures money and personal information. The moment somebody signs - Evidence grade: C - Sources: the author's own experience, no direct literature - Notes: WeChat, Alipay, your email, your Apple account and your Android account all have this option. If you find a device you do not recognise, choose sign out of all devices and then change your password - ### 7. Do not tap 「agree to all」 to get an app working: if the information is not necessary, refusing it does not let them refuse you the service - Cost: No money. The hard part is holding back from tapping 「agree to all」. @@ -66,7 +60,6 @@ This section measures money and personal information. The moment somebody signs - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Personal Information Protection Law, 个人信息保护法. China NPC Net. : Article six 「收集个人信息，应当限于实现处理目的的最小范围，不得过度收集个人信息」 ; Article sixteen 「个人信息处理者不得以个人不同意处理其个人信息或者撤回同意为由，拒绝提供产品或者服务；处理个人信息属于提供产品或者服务所必需的除外」 ; Article fifteen 「基于个人同意处理个人信息的，个人有权撤回其同意。个人信息处理者应当提供便捷的撤回同意的方式」 - Notes: The test is whether the information is necessary to provide that service. A map needs your location; a torch does not need your contacts. After installing an app, go first to the app permissions page in your phone's settings and switch off the permissions you do not need. When you do need one, choose allow just this once. - ### 8. You have the right to see, copy, correct and delete your own personal information, and you can sue if you are refused - Cost: No money. You only need to complain or sue if the other side drags its feet. A real court case runs from several months, and you pay the legal fees yourself, so complaining first is better value. The hard part is chasing them repeatedly while they stall. diff --git "a/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" "b/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" index 39e5c6d9..589b1494 100644 --- "a/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" +++ "b/en/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" @@ -12,7 +12,6 @@ This section measures money. The places where renting loses you the most are the - Evidence grade: A - Sources: State Council (国务院) (2025). Regulations on Housing Leasing, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十条）. - Notes: On the day you move out, take photographs and video together with the landlord. Record the readings on the water, electricity and gas meters, and photograph the walls and the floor as well. If the deposit is reduced without reason, complain first to the local housing leasing administration. If the amount is small, go through small-claims procedure: see section 8 - ### 2. If the water or power is cut, the locks are changed, or you are threatened at the door, call the police and keep the record: the rules ban these ways of making you move out - Cost: No money. One telephone call to the police, which you can make on the spot @@ -21,7 +20,6 @@ This section measures money. The places where renting loses you the most are the - Evidence grade: A - Sources: State Council (国务院) (2025). Regulations on Housing Leasing, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十二条）. - Notes: You call the police so that there is an attendance record, and it serves as evidence when you later claim for the loss. Notify the local housing leasing administration at the same time - ### 3. An agent may not collect and pay the rent and deposit for you: the money goes straight to the landlord - Cost: No money. Check the payee before you pay, which takes a few minutes. The hard part is standing firm when the agent says you should transfer the money to it @@ -30,7 +28,6 @@ This section measures money. The places where renting loses you the most are the - Evidence grade: A - Sources: State Council (国务院) (2025). Regulations on Housing Leasing, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第二十五条）. - Notes: Before you pay, check that the payee is the name on the title certificate. If it is not the owner, ask questions and get a written authority. An agent may only charge the agency fee for bringing the two sides together - ### 4. Before you rent a long-term serviced apartment, check its escrow account, and do not pay a year up front to save money - Cost: No money. One check takes a bit over ten minutes @@ -39,7 +36,6 @@ This section measures money. The places where renting loses you the most are the - Evidence grade: A - Sources: State Council (国务院) (2025). Regulations on Housing Leasing, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十九条）. - Notes: Check at signing whether a loan is attached. If an instalment app pays the apartment a year's money in one go and you repay it monthly, you keep paying after the operator runs off. Do not sign that kind of contract - ### 5. If the flat is sold during the lease, the lease stays valid and you do not have to move - Cost: No money @@ -48,7 +44,6 @@ This section measures money. The places where renting loses you the most are the - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第七百二十五条）. - Notes: The condition is that you have taken possession lawfully, which is what the article calls lawful possession and use. Keep the contract, the transfer records and the evidence of when you moved in - ### 6. Before you sign, check the title certificate and the mortgages, and pay every sum by transfer with a note of the purpose - Cost: No money. One check takes half an hour @@ -57,7 +52,6 @@ This section measures money. The places where renting loses you the most are the - Evidence grade: C - Sources: the author's own experience, no direct literature ; the same reasoning about keeping evidence and noting the purpose of a transfer is in the two items in section 8 on bride price and promissory notes - Notes: You can look up the ownership information at the local real estate registration centre. At signing, have the landlord attend in person; if the landlord cannot come, get a written power of attorney. If a second landlord sublets to you, you need the original landlord's written consent - ### 7. If an agent collects the purchase money for a second-hand flat, it must go through the agent's dedicated client settlement account at the bank - Cost: No money. The hard part is insisting on the dedicated account, which may mean arguing with the agent for a while @@ -66,7 +60,6 @@ This section measures money. The places where renting loses you the most are the - Evidence grade: A - Sources: Ministry of Housing and Urban-Rural Development (住房城乡建设部), National Development and Reform Commission and Ministry of Human Resources and Social Security (人力资源社会保障部) (2011). Measures for the Administration of Real Estate Brokerage, 房地产经纪管理办法（令第 8 号，2016 年第 29 号令修改）第二十四条：「房地产交易当事人约定由房地产经纪机构代收代付交易资金的，应当通过房地产经纪机构在银行开设的客户交易结算资金专用存款账户划转交易资金。」 第十八条「房地产经纪服务实行明码标价制度……在经营场所醒目位置标明房地产经纪服务项目、服务内容、收费标准」；第十九条「两家或者两家以上房地产经纪机构合作开展同一宗房地产经纪业务的，只能按照一宗业务收取佣金，不得向委托人增加收费」；第十七条 on arranging loans, registering property and other services「应当向委托人说明服务内容、收费标准等情况，经委托人同意后，另行签订合同」. - Notes: The purchase money runs from several hundred thousand to several million yuan, and it is the single largest sum in this book, so do not transfer it by WeChat to a broker. There are two more common ways of being overcharged. The first is two agencies working the same deal together, which may carry only one commission. The second is arranging a loan or handling the transfer of title, which is a different service and needs its own contract with the charges told to you in advance. See item 6 in this section for paying every sum by transfer with a note of the purpose (check the title certificate and the mortgages before you sign). - ### 8. Do not rent a partitioned flat: the smallest unit for rent is a room as originally designed, and kitchens, toilets and balconies may not be lived in - Cost: No money. What it costs you is that on the same budget you either live further out or live in a smaller place diff --git "a/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" index 21f1415c..db414f9c 100644 --- "a/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" +++ "b/en/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -12,7 +12,6 @@ This section measures all-cause mortality and money. After a chronic illness is - Evidence grade: A - Sources: Simpson SH, Eurich DT, Majumdar SR, et al. (2006). A meta-analysis of the association between adherence to drug therapy and mortality. BMJ, 333(7557), 15. - Notes:Disputed. This is also the figure in the book that most needs reading with a discount. In the same pooled analysis, people who took a placebo (a dummy drug) for the full course also had a lower mortality (0.56, 0.43 to 0.74). That means a large part of the benefit comes from 「the people who keep taking their tablets are simply more careful about their health」. The authors call this confounder the healthy adherer effect. The harm of stopping the drug itself is nonetheless clear, and especially so for blood pressure medicine, anticoagulants and anti-rejection drugs - ### 2. Get the outpatient chronic and special disease status first and the cross-region filing second, and hypertension, diabetes, chemotherapy and radiotherapy, dialysis and anti-rejection treatment settle directly across regions - Cost: No money. Before you travel, do the cross-region medical filing once in the medical insurance app, which you can do on your phone. @@ -21,7 +20,6 @@ This section measures all-cause mortality and money. After a chronic illness is - Evidence grade: A - Sources: Office of the National Healthcare Security Administration and General Office of the Ministry of Finance (国家医保局办公室、财政部办公厅) (2024). Notice on Safely and in Order Expanding the Range of Outpatient Chronic and Special Diseases Settled Directly across Provinces, 关于稳妥有序扩大跨省直接结算门诊慢特病病种范围的通知（医保办发〔2024〕19 号）. - Notes: The condition is that you do both: get the outpatient chronic and special disease status in your home area first, then make the cross-region filing. How far the 5 added conditions have been opened up varies from place to place, so follow what your local medical insurance department publishes - ### 3. Do your follow-up tests at the interval the doctor gives, and write the figures from each one in the same notebook - Cost: Each test costs from tens to several hundred yuan. A trip to the hospital plus waiting for the results takes some of your time. The hard part is having to go again after a while, and it is easy to keep putting it off. @@ -30,7 +28,6 @@ This section measures all-cause mortality and money. After a chronic illness is - Evidence grade: C - Sources: the author's own experience, no direct literature - Notes: Record at least four things each time: the date, the value, the medicine you were taking, and the dose. Photographing the lab report on your phone is fine, but keep all of it in one place rather than scattered - ### 4. Do not stop proper treatment in order to try a folk remedy or a health supplement - Cost: No money, and you save the money you would have spent on the supplement. It takes no time. The hard part is standing firm when other people keep telling you to try it. @@ -39,15 +36,14 @@ This section measures all-cause mortality and money. After a chronic illness is - Evidence grade: C - Sources: the author's own experience, no direct literature ; for the related evidence see section 6 - Notes: If you want to try a supplement, tell your treating doctor first. Some supplements interact with prescription drugs, for example by changing the level of anticoagulants and anti-rejection drugs in your blood - -### 5. For a stable chronic illness, a community health centre can prescribe 12 weeks of medicine at a time +### 5. A stable chronic illness: a community health centre can prescribe up to 12 weeks of medicine at once -- Cost: No money. Go to the community health centre and ask one question, whether they can write a long-term prescription. That is all it takes. -- In plain language: If your condition is stable and the medicine has not changed, the community health centre can prescribe enough for 12 weeks in one go. That saves the half day every month of travelling to a big hospital and queueing to register. Within the same insurance area, the reimbursement rate at a grassroots hospital is also usually about ten percentage points higher than at the hospital above it. -- Benefit: No monthly trip to a big hospital to queue for a prescription. Within the same insurance area, also called the pooling region, the reimbursement rate at a grassroots hospital is in principle about 10 percentage points higher than at the hospital above it +- Cost: No money. Ask at the community health centre whether they can issue a long-term prescription. That is one question. +- In plain language: If your illness is stable and you have not changed the medicine you take, the community health centre can give you 12 weeks of it in one go. That saves the half-day of travelling to a big hospital every month to queue and register. In the same insurance region, the reimbursement rate for treatment at a primary-level hospital is usually about ten percentage points higher than at a hospital one level up. +- Benefit: You no longer travel to a big hospital every month to queue for a prescription. In the same insurance region (also called the pooled region), the reimbursement rate for treatment at a primary-level hospital is in principle about 10 percentage points higher than at a hospital one level up - Evidence grade: A -- Sources: State Council General Office (国务院办公厅) (2026). Several Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System, 关于加快建设分级诊疗体系的若干措施. : 「对于符合条件的慢性病患者，基层医疗卫生机构单次可开具不超过12周用药的长期处方。」 The same document says 「原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距」, in principle the inpatient reimbursement rates inside the pooling region are spread by about 10 percentage points at each level, and it requires higher-level hospitals to run outpatient clinics for common chronic conditions such as hypertension, diabetes and chronic obstructive pulmonary disease at the grassroots level -- Notes: Whether you 「meet the conditions」 is for the doctor to decide, and it generally means the diagnosis is settled and the treatment plan is stable. It is the same medicine as before; what you save is the half day of going there each month and the registration fee. If the grassroots level does not stock the medicine you need, ask whether they keep a shortage register and can deliver. +- Sources: General Office of the State Council (国务院办公厅) (2026). Several Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System, 关于加快建设分级诊疗体系的若干措施. : 「对于符合条件的慢性病患者，基层医疗卫生机构单次可开具不超过12周用药的长期处方。」; the same document: 「原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距」, and it requires higher-level hospitals to run outpatient clinics for common chronic conditions such as hypertension, diabetes and chronic obstructive pulmonary disease at the primary level; the same document: 「三级医院要聚焦急危重症和疑难复杂疾病，加强转诊会诊和住院服务，逐步酌减常见病复诊和诊断明确、病情稳定的慢性病等普通门诊」, 「并将专家团队普通门诊向基层医疗卫生机构延伸」 +- Notes: Whether you 「meet the conditions」 is decided by the doctor; in general it means the diagnosis is already clear and the treatment plan is stable. The medicine is the same medicine. What you save is the half-day of the monthly trip and the registration fee. If the primary level does not stock the kind you need, ask whether they keep a register of missing drugs and can deliver them. The same document also requires tertiary hospitals to cut back step by step on ordinary outpatient clinics for chronic illnesses of this kind, and to move the ordinary clinics of the same specialists into the community. If you keep treating the big hospital as your place to get medicine, getting an appointment there will only get harder. ### 6. Before you sign up with a family doctor in the community, ask which items are covered by medical insurance and which you pay for yourself @@ -57,7 +53,6 @@ This section measures all-cause mortality and money. After a chronic illness is - Evidence grade: A - Sources: State Council General Office (国务院办公厅) (2026). Several Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System, 关于加快建设分级诊疗体系的若干措施. : 「加强基层门诊付费与签约服务政策联动，基本服务包按规定纳入医保支付；个性化服务包由签约基层医疗卫生机构按程序向县级卫生健康部门备案，费用由个人支付。」 - Notes: The value of signing up is that somebody follows your illness over the long term, not how many items are bundled into the package. When somebody sells you an 「upgrade package」, ask three questions first: is this the basic package or a personalised one, has it been filed, and what does it cost for a year. - ### 7. Once diabetes is diagnosed, have the back of your eyes examined once, then at the interval the doctor gives; have your feet checked once a year - Cost: One dilated examination of the back of the eye costs from tens to several hundred yuan. The foot check is normally part of the outpatient visit and costs nothing extra. The hard part is remembering to go when your eyes and feet give you no trouble at all. @@ -66,7 +61,6 @@ This section measures all-cause mortality and money. After a chronic illness is - Evidence grade: A - Sources: American Diabetes Association (美国糖尿病学会) (2026). Standards of Care in Diabetes 2026, Chapter 12, Retinopathy, Neuropathy and Foot Care, 糖尿病诊疗标准 2026·第 12 章 视网膜病变、神经病变与足部护理. Diabetes Care: 「People with type 2 diabetes should have an initial dilated and comprehensive eye examination by an ophthalmologist or optometrist at the time of the diabetes diagnosis.」「Adults with type 1 diabetes should have an initial dilated and comprehensive eye examination by an ophthalmologist or optometrist 5 years after the onset of diabetes.」「If there is no evidence of retinopathy from one or more annual eye exams and glycemic indicators are within the goal range, then screening every 1–2 years may be considered.」「Perform a comprehensive foot evaluation at least annually to identify risk factors for ulcers and amputations.」 - Notes: For type 2 diabetes you have the examination at the time of diagnosis, and for type 1 you start five years after the onset. The difference is that type 2 has often been quietly high for years. If one or two examinations have found nothing and your blood sugar is on target, you can stretch it to once every one to two years. If the feeling in your foot has become dull, look at the sole once a day yourself, using a mirror if that is easier. - ### 8. If you have had a kidney stone, drink 2.5 to 3 litres of water a day and keep salt under 6 grams - Cost: No money. Turn drinking water throughout the day into a habit. On days when you sweat more you have to drink more still, and the hard part is keeping count every day. @@ -76,11 +70,11 @@ This section measures all-cause mortality and money. After a chronic illness is - Sources: Borghi L, Meschi T, Amato F, Briganti A, Novarini A, Giannini A (1996). Urinary volume, water and recurrences in idiopathic calcium nephrolithiasis: a 5-year randomized prospective study. The Journal of Urology, 155(3), 839-843. ; National Institute for Health and Care Excellence (2019). Renal and ureteric stones: assessment and management. NICE guideline NG118，第 1.8.1 条. - Notes: Recurrence is down by about 55% and it costs nothing, so by the ordering rules of this book it is the best value for money in this section. What you watch is the 24-hour urine volume, with a target of 2 to 2.5 litres or more, not how much you drink in. People who sweat more and people who work in high temperatures have to drink more. That trial covered patients having a first attack of idiopathic calcium stone. Stone composition is not one thing: uric acid stones, infection stones and cystine stones each have their own answer. Anyone who has had a recurrence is worth having a stone composition analysis and a 24-hour urine test, and only then deciding whether to add a drug. Do not cut calcium out in order to prevent stones: see item 21 in section 6. If the pain in your back and belly is severe enough to be colicky, go to hospital; drinking more water does not flush it out. Blood in the urine without pain is a separate thing to investigate: see item 27 in section 1 -### 9. Once gout is diagnosed, take a urate-lowering drug for the long term and keep the blood urate below 360 µmol/L +### 9. Once gout is diagnosed, take urate-lowering medicine for the long term, push the blood urate below 360 µmol/L and keep it there -- Cost: Allopurinol costs from a few yuan to tens of yuan a month. At the start you test the blood urate every few weeks to adjust the dose, and once it is stable every few months. For the first 3 to 6 months you also take a drug that prevents attacks. The hard part is keeping to it when nothing hurts. -- In plain language: Gout needs long-term urate-lowering treatment that holds the urate below 360 µmol/L, and painkillers taken only when it hurts are no substitute. Among patients whose dose was adjusted by a nurse against the lab results, 95% were on target after two years, and only 30% of those seen in the ordinary way were. You still cannot stop once you are on target. Diet does not bring the level down to that line, and giving up alcohol brings it down by only 1.6 mg/dL. -- Benefit: One controlled trial randomised into two groups covered 517 adults who had had an attack in the previous 12 months. One group was nurse-led: the nurse explained what the disease was, then adjusted the urate against the targets until it was on target. The other group saw a general practitioner as usual. At 2 years the share with blood urate down to <360 µmol/L (6 mg/dL) was 95% against 30%. The odds of being on target were 3.18 times those of the usual-care group (RR 3.18, 95% CI 2.42 to 4.18, and that interval is the credible range of the figure), P<0.0001. The secondary measures in year 2, attack frequency, tophi and quality of life, were also better than in the usual-care group. Each extra year of life gained, counted in quality-adjusted life years, cost £5,066. The same trial also ran a follow-up study: questionnaires went to 438 people and 82% came back. Both groups reported their own number of attacks over the previous 12 months. The median in the on-target group was 0 attacks, with the middle half lying between 0 and 0. In the usual-care group it was 1 attack (0 to 3), P<0.001. The range in brackets is called the interquartile range. The share still taking a urate-lowering drug was 1.19 times that of the usual-care group (adjusted RR 1.19, 1.09 to 1.30). The 2020 American College of Rheumatology gout guideline strongly recommends starting urate-lowering treatment in three kinds of people: those with tophi under the skin, those in whom imaging shows gout has eaten defects into the bone, and those with ≥2 attacks a year. The method is to keep monitoring the blood urate and adjust the dose against the results until it is on target, with a target of <6 mg/dL. Allopurinol is the first choice, and it is also the first choice in people with stage 3 or worse chronic kidney disease, with a starting dose of ≤100 mg/day. Give an anti-inflammatory preventive drug at the same time as you start, for at least 3 to 6 months. The same guideline also cites a series of case observations. Those patients stopped the drug after long-term good control, and their blood urate stayed below <7 mg/dL after stopping. Among them, only 13% had no attack at all over the 5-year follow-up (27/211). On diet, people who limit or give up alcohol have blood urate 1.6 mg/dL lower than people who do not, and one serving of beer raises blood urate by 0.16 mg/dL. A healthy diet, the Mediterranean diet and the DASH diet have a smaller effect. There is disagreement over which urate-lowering drug to use first. The American College of Rheumatology guideline strongly recommends allopurinol as first choice, on the grounds that it works when the dose is high enough, is well tolerated, is safe and is cheap. It also makes a conditional recommendation: people on febuxostat who have a history of cardiovascular disease, or who have just had a cardiovascular event, should switch if another drug is available. That recommendation matches the black-box warning the FDA added to febuxostat. One of the grounds for it is the CARES trial, in which 6190 gout patients with coexisting cardiovascular disease were randomly assigned febuxostat or allopurinol. In the febuxostat group the risk of all-cause death was 1.22 times that of the allopurinol group (HR 1.22, 95% CI 1.01 to 1.47, about 22% higher). Cardiovascular death was 1.34 times (1.03 to 1.73, about 34% higher). The European FAST trial reached a different conclusion. It enrolled 6128 patients over 60 with cardiovascular risk factors, and febuxostat produced no more major cardiovascular events than allopurinol (HR 0.85, 0.70 to 1.03). The death rate was 7.2% on febuxostat and 8.6% on allopurinol. For people taking a drug that pushes urate out through the kidneys, the American College of Rheumatology conditionally advises against alkalinising the urine, because there is no evidence that it works and the certainty of that evidence is very low. It also requires that these people be told to drink more water, and people known to have kidney stones or stage 3 or worse chronic kidney disease should not use this class of drug +- Cost: Allopurinol costs from a few yuan to a few tens of yuan a month. At the start you test the blood urate every few weeks to adjust the dose, and once it is stable every few months. For the first 3 to 6 months you also take a medicine that prevents attacks. The hard part is that you have to keep taking it when nothing hurts. +- In plain language: Gout needs long-term urate-lowering medicine that brings the urate below 360 µmol/L; you cannot just take painkillers when it hurts. Among patients whose dose a nurse helps adjust against the lab results, 95% reached the target after two years, against only 30% of those treated as usual. You cannot stop the medicine once you have reached the target either. Diet does not bring the level down to that line, and giving up alcohol lowers it by only 1.6 mg/dL. +- Benefit: A controlled trial with 517 adults who had an attack in the past 12 months, randomised into two groups. One group was led by a nurse, who explained what the illness was and then adjusted the urate against the target figures. The other group saw a general practitioner as usual. At 2 years the proportion with blood urate down to <360 µmol/L (6 mg/dL) was 95% against 30%. The chance of reaching the target was 3.18 times that of the usual-care group (RR 3.18, 95% CI 2.42–4.18, that interval is the credible range for the figure), P<0.0001. The secondary outcomes in year 2 — attack frequency, gouty tophi and quality of life — were also better than in the usual-care group. Each extra year of life gained, converted for quality of life (quality-adjusted life years), cost £5,066. The same trial also ran a follow-up study: 438 people were sent a questionnaire and 82% returned it. Both groups reported their own number of attacks over the past 12 months. The median in the group that reached the target was 0 attacks (the middle half fell between 0–0). In the usual-care group it was 1 attack (0–3), P<0.001. The range in brackets is called the interquartile range. The proportion still taking urate-lowering medicine was 1.19 times that of the usual-care group in the group that reached the target (adjusted RR 1.19, 1.09–1.30). The 2020 American College of Rheumatology gout guideline strongly recommends starting urate-lowering treatment in three kinds of people: those with gouty tophi under the skin, those whose imaging shows gout eating defects into the bone, and those with ≥2 attacks a year. The practice is to keep monitoring the blood urate and adjust the dose against the results until the target is met, which is <6 mg/dL. Allopurinol is the first choice, and it is also the first choice for people with stage 3 or higher chronic kidney disease, with a starting dose of ≤100 mg/day. Preventive anti-inflammatory medicine is given at the same time when treatment starts, for at least 3 to 6 months. The same guideline also cites a set of case observations. These patients stopped the medicine after good long-term control, and their blood urate still stayed below <7 mg/dL after stopping. Among them, only 13% (27/211) had no attack at all during the 5-year follow-up. On diet: people who limit or give up alcohol have blood urate 1.6 mg/dL lower than people who do not, and one serving of beer raises blood urate by 0.16 mg/dL. Healthy eating, the Mediterranean diet and the DASH diet make a smaller difference. There is disagreement over which urate-lowering drug to use first. The American College of Rheumatology guideline strongly recommends allopurinol as the first choice, on the grounds that it works when the dose is right, is well tolerated, safe and cheap. It also recommends conditionally: people on febuxostat who have a history of cardiovascular disease, or who have just had a cardiovascular event, should switch to something else where there is something else to switch to. That recommendation matches the black-box warning the FDA put on febuxostat. One of the bases is the CARES trial, in which 6190 gout patients with coexisting cardiovascular disease were put on febuxostat or allopurinol at random. In the febuxostat group the risk of all-cause death was 1.22 times that of the allopurinol group (HR 1.22, 95% CI 1.01–1.47, about 22% higher). Cardiovascular death was 1.34 times (1.03–1.73, about 34% higher). The European FAST trial reached a different conclusion. It enrolled 6128 patients aged over 60 with cardiovascular risk factors, and febuxostat did not produce more major cardiovascular events than allopurinol (HR 0.85, 0.70–1.03). Deaths were 7.2% in the febuxostat group and 8.6% in the allopurinol group. For people taking a drug that makes the kidneys excrete urate, the American College of Rheumatology advises against alkalinising the urine as a conditional recommendation, because there is no evidence that it works and the certainty of the evidence is very low. It also requires telling these people to drink more water. People with known kidney stones, or stage 3 or higher chronic kidney disease, should not use this kind of drug - Evidence grade: A -- Sources: Doherty M, Jenkins W, Richardson H, et al. (2018). Efficacy and cost-effectiveness of nurse-led care involving education and engagement of patients and a treat-to-target urate-lowering strategy versus usual care for gout: a randomised controlled trial. Lancet, 392(10156), 1403-1412. ; Abhishek A, Jenkins W, La-Crette J, Fernandes G, Doherty M (2020). Nurse-led care is preferred over GP-led care of gout and improves gout outcomes: results of Nottingham Gout Treatment Trial follow-up study. Rheumatology, 59(3), 575-579. ; FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760. ; White WB, Saag KG, Becker MA, et al.; CARES Investigators (2018). Cardiovascular Safety of Febuxostat or Allopurinol in Patients with Gout. New England Journal of Medicine, 378(13), 1200-1210. ; Mackenzie IS, Ford I, Nuki G, et al.; FAST Study Group (2020). Long-term cardiovascular safety of febuxostat compared with allopurinol in patients with gout (FAST): a multicentre, prospective, randomised, open-label, non-inferiority trial. Lancet, 396(10264), 1745-1757. ; Chinese Society of Endocrinology (中华医学会内分泌学分会) (2020). Guidelines for the Diagnosis and Treatment of Hyperuricaemia and Gout (2019), 中国高尿酸血症与痛风诊疗指南(2019). 中华内分泌代谢杂志, 36(1), 1-13. -- Notes:Disputed: guidelines in different countries disagree about which drug to use. The US guideline makes allopurinol first choice and says people with cardiovascular disease should switch off febuxostat where they can, but the European FAST trial did not find febuxostat more dangerous. In China you can also buy benzbromarone, which pushes urate out through the kidneys. The guidelines of the Chinese Medical Association (the 2019 version and the 2024 update) give separate recommendations on the choice of drug and on alkalinising the urine. Whether to add sodium bicarbonate to alkalinise the urine when taking benzbromarone: the US guidelines say there is no evidence, so follow your own doctor's advice, and do not leave out the water. **It is worth testing HLA-B*5801 before a Han Chinese person takes allopurinol.** It is an inherited gene type that governs the immune response, and one test result stays the same for life. Among people with this gene the figure is 7.4% among the Han, Korean and Thai populations and only 0.7% among white and Hispanic people. The risk of allopurinol hypersensitivity syndrome in Asian and African people after taking allopurinol is 3 times that of white people, and in its severe form the skin peels off over large areas, which can kill. The 2020 American College of Rheumatology gout guideline suggests, for these two groups, considering testing before starting the drug. **Starting urate-lowering treatment is itself more likely to trigger an attack**, so the guideline clearly requires an anti-inflammatory preventive drug during the first 3 to 6 months. 「It got more painful after I took the urate-lowering drug, so I stopped」 is the commonest way this goes wrong. When it hurts, what you do is add an anti-inflammatory drug as the doctor ordered, not stop the urate-lowering drug. The target value is written differently by different sources: <6 mg/dL is about 360 µmol/L, and Chinese lab sheets mostly use µmol/L. A high urate on a health check with no attack ever is a different question: see item 19 in section 6. For sugary drinks and alcohol, see items 7 and 20 in section 2 (do not drink sugary drinks, and drink little or no alcohol) +- Sources: Doherty M, Jenkins W, Richardson H, et al. (2018). Efficacy and cost-effectiveness of nurse-led care involving education and engagement of patients and a treat-to-target urate-lowering strategy versus usual care for gout: a randomised controlled trial. Lancet, 392(10156), 1403-1412. ; Abhishek A, Jenkins W, La-Crette J, Fernandes G, Doherty M (2020). Nurse-led care is preferred over GP-led care of gout and improves gout outcomes: results of Nottingham Gout Treatment Trial follow-up study. Rheumatology, 59(3), 575-579. ; FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760. ; White WB, Saag KG, Becker MA, et al.; CARES Investigators (2018). Cardiovascular Safety of Febuxostat or Allopurinol in Patients with Gout. New England Journal of Medicine, 378(13), 1200-1210. ; Mackenzie IS, Ford I, Nuki G, et al.; FAST Study Group (2020). Long-term cardiovascular safety of febuxostat compared with allopurinol in patients with gout (FAST): a multicentre, prospective, randomised, open-label, non-inferiority trial. Lancet, 396(10264), 1745-1757. ; Chinese Medical Association, Branch of Endocrinology (中华医学会内分泌学分会) (2020). 中国高尿酸血症与痛风诊疗指南(2019). 中华内分泌代谢杂志, 36(1), 1-13. +- Notes:Disputed. Countries' guidelines disagree about which drug to lower urate with. The US guideline makes allopurinol the first choice and says people with cardiovascular disease should get off febuxostat where they can, but the European FAST trial did not find febuxostat more dangerous. Benzbromarone is also sold in China; it works by making the kidneys excrete urate. The Chinese Medical Association guidelines (the 2019 edition and the 2024 update) give separate recommendations on choosing the drug and on alkalinising the urine. On whether to add sodium bicarbonate to alkalinise the urine when taking benzbromarone, the US guideline says there is no evidence; follow your doctor's advice, and do not skip the extra water. **Before a Han Chinese person takes allopurinol, it is worth testing for HLA-B\*5801**. It is an inherited gene type that governs the immune response, and one test gives a result that never changes. Among Han Chinese, Korean and Thai people, 7.4% carry this gene; among white and Hispanic people it is only 0.7%. Asian and Black people who take allopurinol run three times the risk of white people of developing allopurinol hypersensitivity syndrome, and in severe cases that disease strips the skin off over large areas and can kill. The 2020 American College of Rheumatology gout guideline suggests considering a test first for these two groups. **Starting urate-lowering treatment actually makes an attack more likely at first**, so the guideline explicitly requires a preventive anti-inflammatory medicine at the same time for the first 3 to 6 months. 「It got more painful once I started the urate-lowering drug, so I stopped」 is the commonest way this goes wrong. When the pain comes, what you do is add an anti-inflammatory drug as the doctor ordered, not stop the urate-lowering drug. Guides write the target value differently: <6 mg/dL is about 360 µmol/L, and Chinese lab reports mostly use µmol/L. A high urate found on a check-up with no attack ever is a different matter: see item 19 in section 6. For sugary drinks and alcohol, see items 7 and 19 in section 2 (do not drink sugary drinks, drink little or no alcohol) diff --git "a/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" "b/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" index e5d1f91c..b40370bf 100644 --- "a/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" +++ "b/en/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" @@ -2,7 +2,7 @@ # 17. When there are elderly at home -This section measures money and personal freedom. With elderly people at home, most of the trouble is not a shortage of money; it is not having said the words clearly and signed the paper in advance. The last item covers pressure ulcers in someone bedridden long-term, and it is measured by mortality rather than compared with the other items here. +This section measures money and personal freedom. With elderly people at home, most of the trouble is not a shortage of money; it is not having said the words clearly and signed the paper in advance. The last two items cover pressure ulcers in someone bedridden long-term and cataract surgery. Both are measured by mortality rather than compared with the other items here. ### 1. While the older person is still clear-headed, name in writing who will be their guardian @@ -12,7 +12,6 @@ This section measures money and personal freedom. With elderly people at home, m - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第三十三条）. - Notes: This kind of document is called voluntary guardianship, 意定监护, which means a guardian you name yourself in advance. That person comes before the default order the law sets out. An older person living alone, children living in another city, or a complicated family situation all make it more necessary to do this in advance. Have it notarised at a notary office - ### 2. Make a will, and remember that a later one overturns an earlier one, and that a notarised will no longer comes first - Cost: Writing a will by hand costs nothing. Having it notarised costs from several hundred to over a thousand yuan. The hard part is finding the words: you have to talk to the family about what comes after you die @@ -21,7 +20,6 @@ This section measures money and personal freedom. With elderly people at home, m - Evidence grade: A - Sources: National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千一百三十四条至第一千一百四十二条）. - Notes: A will handwritten by the testator has to be written out by them from beginning to end, signed, and dated. A printed will needs more than two witnesses. The testator and the witnesses have to sign every page and give the date. Heirs, people named in the will to receive something, and their close relatives may not act as witnesses. Get the formalities wrong and the whole will is void, which is the commonest way this goes wrong - ### 3. Keep the older person's money in a separate account, and set a rule that two people confirm any large payment - Cost: No money. One conversation with the family and you have agreed a figure. The hard part is how to raise it, so that the older person does not think you are watching them @@ -30,7 +28,6 @@ This section measures money and personal freedom. With elderly people at home, m - Evidence grade: C - Sources: the author's own experience, no direct literature ; for the general rules on fraud see section 8 and section 13 - Notes: Do not put it as 「I am afraid you will be cheated」. Put it as 「the family's money goes into one set of accounts」 and the older person takes it far more easily. Do not hand all of the older person's money to the children to keep either, because that produces a different kind of dispute - ### 4. Give the older person one sentence they can say at any time to get out of the situation - Cost: No money. One conversation with the older person, a few minutes is enough @@ -39,7 +36,6 @@ This section measures money and personal freedom. With elderly people at home, m - Evidence grade: C - Sources: the author's own experience, no direct literature - Notes: Use it together with item 3 in this section (keep the older person's money in a separate account). Do the fall-proofing changes round the house while you are at it: see section 1. The sentence gives the older person a way out so that he does not have to stand his ground on the spot; it is not a way of watching him. When he is surrounded by sellers, the hardest part is saying no in front of a roomful of people, and that is harder than telling whether they are genuine - ### 5. Stay away from every 「invest in your retirement」 scheme that asks the older person to pay first: cards, bed reservations, retirement flats, living-out schemes and products for older people are one kind of illegal fundraising - Cost: No money. Read the list of names below to the older person once, and it takes a few minutes. After that, refuse them straight away: do not go to the lectures, do not go to the visits, do not leave your phone number. The hard part is that once the older person has been a few times and knows the people, it is not easy to talk them round @@ -48,7 +44,6 @@ This section measures money and personal freedom. With elderly people at home, m - Evidence grade: A - Sources: Ministry of Civil Affairs (民政部) (2023). Risk Notice on Illegal Fundraising in the Field of Elderly Care Services, 关于养老服务领域非法集资的风险提示. (published on the Chinese government website) ; Ministry of Civil Affairs (民政部) and China Banking and Insurance Regulatory Commission (中国银保监会) (2021). Risk Notice on Illegal Fundraising in the Field of Elderly Care, 关于养老领域非法集资的风险提示. ; State Council (国务院) (2021). Regulations on the Prevention and Control of Illegal Fundraising (State Council Order No. 737, articles two, twenty-five and thirty), 防范和处置非法集资条例（国务院令第 737 号，第二、二十五、三十条）. ; Supreme People's Procuratorate (最高人民检察院) (2022). Typical Cases of Procuratorates Cracking Down on and Rectifying Fraud against the Elderly, 检察机关打击整治养老诈骗犯罪典型案例. ; Supreme People's Court (最高人民法院) (2025). Typical Cases of Punishing Fraud Against People's Livelihoods According to Law (case five, the Ni Mourong fraud case), 依法惩治涉民生领域诈骗犯罪典型案例（案例五 倪某荣诈骗案）. - Notes: You do not judge by 「this one looks official」, you judge by 「does it ask you to pay first, and does it promise a return」. A retirement home is in the business of selling services, so once the talk turns to investment, rebates or getting your money back, do not believe it. Being taken to lectures, on trips or to free clinics is not itself illegal and is very hard to stop. What you can stop is the moment of paying, and the two-person confirmation rule in item 3 in this section works with it. If the money has already been paid, call 110 or 96110 at once to ask for the payment to be stopped, following item 2 in section 8, and report it at the same time to the local lead department in charge of dealing with illegal fundraising. Raising money against your home is a different route: see item 6 in this section (reverse-mortgage insurance for the elderly). - ### 6. Apart from the insurance companies' reverse-mortgage insurance for the elderly, stay away from every other 「raise money against your home」 scheme, and never mortgage the flat to buy investment products - Cost: No money. Do not hand the property certificate at home to anybody who comes to the door in the name of 「raising money against your home」. For mortgages, guarantees and loans, show the agreement to your children before you sign. The hard part is that the people who come are often warm and attentive and have made many trips, and the older person finds it hard to say no @@ -66,7 +61,6 @@ This section measures money and personal freedom. With elderly people at home, m - Evidence grade: A - Sources: National Healthcare Security Administration and seven other departments (国家医保局等八部门) (2026). Notice on Issuing the Implementation Plan for Accelerating the Establishment of the Long-Term Care Insurance System, 关于印发《加快建立长期护理保险制度实施方案》的通知. : the people covered are 「按规定参保缴费且失能状态长期持续（一般为6个月以上），经申请通过评估认定的失能人员」, those who have paid contributions as required and whose disability has continued for a long time, generally more than 6 months, and who are recognised as disabled after applying and passing an assessment, and 「制度起步阶段保障重度失能人员」, the scheme covers severely disabled people at the stage where it starts ; 「待遇享受不设起付标准。符合规定的长期护理服务费用，按未就业城乡居民参保政策参保的，基金支付比例为50%左右；按单位职工参保政策参保的，基金支付比例为70%左右，退休人员享受单位职工参保待遇」, benefits carry no deductible threshold; for qualifying long-term care service costs the fund pays around 50% for people insured under the scheme for residents who are not in employment, and around 70% for people insured under the employed-worker scheme, and retired people get the employed-worker treatment ; 「参保人员基金年度最高支付限额不超过统筹地区上年度城乡居民人均可支配收入的50%」, the annual maximum the fund pays for one person is not more than 50% of the previous year's per-capita disposable income of urban and rural residents in the pooling region ; 「待遇享受固定等待期原则上按照6个月设置」, the fixed waiting period for benefits is in principle set at 6 months ; also see the Opinions of the General Offices of the CPC Central Committee and the State Council on Accelerating the Establishment of the Long-Term Care Insurance System, 中共中央办公厅、国务院办公厅《关于加快建立长期护理保险制度的意见》. - Notes: The disability has to 「长期持续（一般为6个月以上）」, to have continued for a long time, generally more than 6 months; a few weeks in hospital does not reach the condition for applying. There is also a 6-month waiting period before you can draw the benefit. Costs that are not care, such as bed fees and meal costs, are not paid by long-term care insurance. Medical costs that medical insurance should pay are not paid by it either. If you are already receiving work-injury living and care payments, you do not receive this on top. This money and the care allowance inside the two disability allowances are two sets of formalities and two separate sums, and they do not clash: see item 8 in section 7 (the two allowances). Care at home and in the community is reimbursed at a slightly better rate, so you do not have to go into an institution. The scheme is being put in place region by region, and the assessment standards differ. Ring 12393 or the medical insurance agency in your home area first, and ask whether it has started locally and how to book an assessment. If the assessment is not passed, do not give up: you can apply again once the disability has become worse. - ### 8. If someone at home is bedridden long-term, treat pressure ulcers as the main enemy: get an electric pressure-relief mattress, turn them on a schedule, and look at the bony parts every day - Cost: A home electric alternating-pressure air mattress starts at a few hundred yuan. Each turn takes a few minutes. The hard part is that you have to do it every two or three hours, through the night as well, and one person cannot manage it alone @@ -75,3 +69,12 @@ This section measures money and personal freedom. With elderly people at home, m - Evidence grade: B - Sources: Porter SB, Pla R, Chow JH, et al. (2022). Preoperative Pressure Ulcers, Mortality, and Complications in Older Hip Fracture Surgery Patients. JAAOS Global Research & Reviews, 6(11). ; Ottaviani S, Rondanina E, Longo E, et al. (2026). Impact of pressure ulcers and frailty on long-term mortality: a prospective cohort study of hospitalized older adults. European Geriatric Medicine, 17(2), 665-672. ; Shi C, Dumville JC, Cullum N (2018). Support surfaces for pressure ulcer prevention: A network meta-analysis. PLoS ONE, 13(2), e0192707. ; Latimer SL, Chaboyer WP, Probst S, et al. (2026). Repositioning for pressure injury prevention in adults. Cochrane Database of Systematic Reviews, 6, CD009958. - Notes: There are three reasons for the B. First, the two mortality studies only show that 「people with pressure ulcers did worse afterwards」, which is not the same as 「preventing pressure ulcers brings the death rate down」. Second, the randomised trials on the turning interval rest on very shaky conclusions. Third, the evidence on the mattress is harder, but what the studies compare are hospital mattresses, and applying it to a home mattress is an extrapolation. Check the base of the spine, the heels, the hips and the shoulder blades every day for a red mark that does not fade under pressure; that is routine nursing practice and has no separate literature. For the care services that long-term care insurance pays for, see item 7 in this section. If one leg suddenly swells, treat it as a deep vein thrombosis: see item 11 in section 13. For the period bedridden after a fall or a serious injury, see item 34 in section 1 + +### 9. Where an older person's cataract is already affecting how they see the road, go to the eye department for an assessment and an operation; where both eyes have it, do not leave the second one too long + +- Cost: A fair amount of money. First take the older person to the eye department for one examination and let the doctor decide whether it should be done. One eye is one operation, and many hospitals do it as day surgery, so the person can go home the same day. Where both eyes have a cataract, it has to be done in two operations. How much you pay yourself depends on which intraocular lens, that is, the artificial lens, is put in and how the local medical insurance reimburses it, so ask before the operation. +- In plain language: Women over 70 who have the first eye's cataract done as soon as possible fall about 34% less often over a year, and fewer of them have a fracture. But the number of people who fall is not lower. Whether the second eye is done makes no clear difference to falls. Older people who have the operation are about 16% lower in hip fracture within a year. +- Benefit: A randomised trial took 306 women over 70 with a cataract. Half had the first eye operated on within about 4 weeks and the other half waited the usual 12 months. Over a year, the people operated on as soon as possible fell about 34% less often. The original figure is a fall-rate ratio of 0.66, credible range 95% CI 0.45 to 0.96. The people who fell at least once were 49% in the surgery group and 45% in the waiting group, which is no lower. The people who fell more than twice were 18% in the surgery group and 25% in the waiting group. In the surgery group 4 people had a fracture (3%), in the waiting group 12 people (8%). The Cochrane review included this trial and concluded that surgery on the first eye in women reduced the number of falls, and surgery on the second eye did not. The randomised trial of the second eye took 239 women over 70. The fall rate in the surgery group was 32% lower, but the difference may be chance (fall-rate ratio 0.68, 95% CI 0.39 to 1.19). In a 5% sample of the US Medicare system there were 1113640 cataract patients over 65 (2002 to 2009). This study only followed records and had no control group. The people who had the operation were about 16% lower in the chance of a hip fracture within a year (OR 0.84, 95% CI 0.81 to 0.87). The absolute difference is 0.20 percentage points, about 2 fewer per 1000 people +- Evidence grade: B +- Sources: Harwood RH, Foss AJ, Osborn F, Gregson RM, Zaman A, Masud T (2005). Falls and health status in elderly women following first eye cataract surgery: a randomised controlled trial. British Journal of Ophthalmology, 89(1), 53-59. ; Foss AJ, Harwood RH, Osborn F, Gregson RM, Zaman A, Masud T (2006). Falls and health status in elderly women following second eye cataract surgery: a randomised controlled trial. Age and Ageing, 35(1), 66-71. ; Gillespie LD, Robertson MC, Gillespie WJ, et al. (2012). Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews, 9, CD007146. ; Tseng VL, Yu F, Lum F, Coleman AL (2012). Risk of fractures following cataract surgery in Medicare beneficiaries. JAMA, 308(5), 493-501. ; Desapriya E, Subzwari S, Scime-Beltrano G, et al. (2010). Vision improvement and reduction in falls after expedited cataract surgery: systematic review and metaanalysis. Journal of Cataract and Refractive Surgery, 36(1), 13-19. ; Meuleners LB, Lee AH, Ng JQ, Morlet N, Fraser ML (2012). First eye cataract surgery and hospitalization from injuries due to a fall: a population-based study. Journal of the American Geriatrics Society, 60(9), 1730-1733. ; Meuleners LB, Fraser ML, Ng J, Morlet N (2014). The impact of first- and second-eye cataract surgery on injurious falls that require hospitalisation: a whole-population study. Age and Ageing, 43(3), 341-346. ; Keay L, Ho KC, Rogers K, et al. (2022). The incidence of falls after first and second eye cataract surgery: a longitudinal cohort study. Medical Journal of Australia, 217(2), 94-99. ; National Health Commission (国家卫生健康委) (2022). The Fourteenth Five-Year National Eye Health Plan, 「十四五」全国眼健康规划（2021-2025年）. +- Notes: Disputed. Pooling the two randomised trials, the number of people who fell is not clearly lower (OR 0.81, 95% CI 0.55 to 1.17). In the whole-population hospital records for Western Australia, in the year after the first eye was done, hospital admissions for injuries from falls went up by 27%. In the stretch between the two operations, admissions of this kind were 2.14 times what they had been before the operation. A comparison like this before and after cannot separate the effect of the operation from the effect of getting older. In another Australian follow-up study, the number of falls fell most after the second eye was done. So where both eyes have a cataract, do not leave the second one too long. In the stretch between the two operations, be especially careful when walking and on stairs. Cochrane also included a trial that treated vision problems in older people, and the number of falls went up by about 57% instead. The operation is about whether you can see clearly; to fall less you also need item 13 in section 1 (balance and leg strength). Whether the bones can take a fall, see item 39 in section 1 (a bone density check) and item 40 in section 1 (treating osteoporosis once it is found). The national eye health plan states that cataract surgery to restore sight is already carried out in every county, so you do not have to go to a hospital in a big city. The people who benefit are mainly your parents and grandparents. diff --git "a/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" index c860be89..7aa6c77d 100644 --- "a/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" +++ "b/en/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -12,7 +12,6 @@ This section works out two accounts, money and time. As in section 10, where the - Evidence grade: A - Sources: General Offices of the CPC Central Committee and the State Council (中共中央办公厅、国务院办公厅) (2025). Implementation Plan for the Child-Rearing Subsidy Scheme, 育儿补贴制度实施方案. - Notes: This is the nationwide minimum. Some provinces and cities have their own local subsidy on top, and you can receive both. Three years come to 10800 yuan, which is a very small part of the cost of raising a child, so do not use it as the reason for deciding whether to have one - ### 2. Maternity leave is 98 days, and the maternity allowance is paid by the maternity insurance fund at the employer's average monthly wage for the previous year - Cost: No money. The employer makes the declaration, so you do not have to go anywhere @@ -21,7 +20,6 @@ This section works out two accounts, money and time. As in section 10, where the - Evidence grade: A - Sources: State Council (国务院) (2012). Special Provisions on Labour Protection for Female Employees, 女职工劳动保护特别规定（国令第 619 号，第七条、第八条）. - Notes: 98 days is the minimum set by the state. Most provinces add a reward leave of more than 60 days for childbirth in their population and family planning regulations, so 158 days in total is the usual starting point. Which party pays the wages during that reward leave is set differently in each province - ### 3. Know this rule: your pay may not be cut and you may not be dismissed because of pregnancy, childbirth or breastfeeding - Cost: No money @@ -30,7 +28,6 @@ This section works out two accounts, money and time. As in section 10, where the - Evidence grade: A - Sources: State Council (国务院) (2012). Special Provisions on Labour Protection for Female Employees, 女职工劳动保护特别规定（国令第 619 号，第五条、第九条、第十三条）. - Notes: If you are really dismissed in breach of the law, the route to a claim is in section 19. The evidence to collect is the written notice of a transfer to another post or a pay cut, together with the chat records, and you should keep both - ### 4. Count the time account by 「whose time is taken up」, not by 「how hard it is」 - Cost: No money; what it costs is time. Before the age of 3 the child takes up nearly the whole day. The hard part is that it is every single day, and there is no grinding it out for a few days and being finished @@ -39,7 +36,6 @@ This section works out two accounts, money and time. As in section 10, where the - Evidence grade: C - Sources: the author's own experience, no direct literature ; for the measure used see section 10 - Notes: The figures do not have to be exact. What is useful is that the two of you sit down with the same table before the birth and agree who does what, so that there is no argument afterwards - ### 5. Split the money account into three stages: 0 to 3 years old, compulsory education, and after compulsory education - Cost: There is no fixed figure for what it costs; different cities and different ways of looking after the child are far apart. The hard part is sitting down quietly and working out the three stages separately, instead of guessing a number @@ -48,7 +44,6 @@ This section works out two accounts, money and time. As in section 10, where the - Evidence grade: C - Sources: the author's own experience, no direct literature - Notes: First count the money you are certain to receive, such as the child-rearing subsidy, the maternity allowance and medical insurance reimbursements, and only then see what is still missing. Do not base a decision on the 「how many tens of thousands of yuan it takes to raise a child」 figures that circulate online; the measure behind them is not stated, so you do not know what they counted or what they left out - ### 6. Having a child for the sake of the older generation, for the sake of the marriage, and for the sake of your own old age: keep a separate account for each - Cost: No money; what it costs is one serious conversation. The hard part is saying it straight rather than going round it diff --git "a/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" "b/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" index f6e187c1..c520454c 100644 --- "a/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" +++ "b/en/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" @@ -1,8 +1,8 @@ -[← Back to the contents](../README.md) +[← 回总目录](../README.md) # 19. Working, leaving, and work injuries -This section is about money: how it is calculated, how the documents are signed, and how the time limits run. The first three items cover the money you should be getting while you are still employed, the middle six cover leaving the job, and the last seven cover occupational disease and work injuries. The money for a work injury is an order of magnitude larger than for a dismissal, and the time limits are firmer. Damage caused by an occupational disease cannot be undone. +This section is about money: how the amounts are worked out, how the documents are signed, and how the deadlines run. The first three items cover the money you should be getting while you are still in the job. The next six cover leaving, and the seven after that cover occupational disease and work injuries. The money on a work injury is an order of magnitude larger than on a dismissal, and the deadlines are harder. Damage caused by an occupational disease cannot be put back. Three more items sit at the end of the section: what to do if you are treated badly at work, how to get a certificate of leaving, and how the income tax on severance works. ### 1. Overtime pay runs in three bands at 1.5, 2 and 3 times, and if it is not paid you complain to the labour inspectorate, with a further 50% to 100% added if the deadline passes @@ -12,7 +12,6 @@ This section is about money: how it is calculated, how the documents are signed, - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). Labour Law, 劳动法（2018 年第二次修正，第三十六、四十一、四十三、四十四条）. (published by the State Administration for Market Regulation) ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第八十五条第三项）. - Notes: Rest days and statutory holidays are counted separately. The statutory holidays are only the few fixed days in a year, such as New Year's Day, Spring Festival, Labour Day and National Day. A long weekend created by shifting a day off is counted as a rest day, and is paid at 200%. 「The company requires approval for overtime」 and 「I stayed of my own accord」 do not mean the company can pay nothing, but they make your evidence harder to produce, so keep the assignment messages and the approval records. If the company wants to use the two special working-time systems, the comprehensive working-hours system and the irregular working-hours system, it needs the labour administration's approval first; it is not the company's own decision. For the time limit on arbitration of a dispute, see item 19 in section 8 - ### 2. Annual leave is 5, 10 or 15 days according to total length of service, and days you cannot take are paid at 300% of your daily wage - Cost: No money. Add up the years you have worked for all employers, and your social insurance contribution record is the proof. The hard part is refusing to sign the form 「voluntarily giving up annual leave」 when the company hands it to you. @@ -21,7 +20,6 @@ This section is about money: how it is calculated, how the documents are signed, - Evidence grade: A - Sources: State Council (国务院) (2007). Regulations on Paid Annual Leave for Employees, 职工带薪年休假条例（国务院令第 514 号，第二、三、四、五、七条）. ; Ministry of Human Resources and Social Security (人力资源和社会保障部) (2008). Measures for the Implementation of Paid Annual Leave for Enterprise Employees, 企业职工带薪年休假实施办法（人社部令第 1 号，第三、四、五、十、十一、十二、十五条）. - Notes: Cumulative working time has to be added up across employers, so if you have changed jobs do not use only the length of service with your current employer. **Only where the employee 「因本人原因且书面提出不休」, does not take the leave for their own reasons and asks in writing, may the employer pay only the ordinary wage**, so do not sign the form 「voluntarily giving up annual leave」. If you take personal leave totalling more than 20 days and the employer does not dock your pay, you get no annual leave that year. A sick leave longer than the corresponding number of months also disqualifies you: less than 10 years of service means 2 months, 10 to 20 years means 3 months, and more than 20 years means 4 months. Home-visit leave, marriage and bereavement leave, maternity leave and the work-injury period of paid recovery do not count towards annual leave. - ### 3. The probation period has a legal ceiling, the wage may not be below 80%, and it can only be agreed once - Cost: No money. Before you sign, spend a few minutes matching the term written in the contract against the length of the probation period. @@ -30,24 +28,23 @@ This section is about money: how it is calculated, how the documents are signed, - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第十九、二十、二十一、八十三条）. - Notes: The three commonest unlawful practices are these. First, a contract of only 1 year with a 3-month probation period agreed. Second, moving the person to another post after they have passed and running a second probation period. Third, 「no social insurance during probation」, because the obligation to enrol starts on the first day of work and has nothing to do with probation; for the employer's side of the obligation see item 16 in section 12. Being dismissed during probation does not mean getting nothing at all; it depends on whether the reason the employer gives is one of those the law allows. Where the termination is unlawful, the 2N in item 6 applies. - ### 4. Work out N before a dismissal: one month's wage for each full year, and half a month for under six months -- Cost: No money. One calculation on a calculator, ten minutes. -- In plain language: The economic compensation due on dismissal is worked out like this: a full year of service gives one month's wage, six months to a year counts as a full year, and under six months gives half a month. The monthly wage here is the average over the 12 months before you leave, and bonuses, allowances and subsidies all count. If the wage is above three times the local previous-year average monthly wage, it is capped at three times, and the length of service counts for at most 12 years. -- Benefit: The economic compensation set by the Labour Contract Law is the money the employer owes on a dismissal, and it is commonly called N. The method is 「每满一年支付一个月工资……六个月以上不满一年的，按一年计算；不满六个月的，向劳动者支付半个月工资」, one month's wage for each full year of service, six months to a year counted as a full year, and half a month's wage where it is under six months. If the monthly wage is above three times the local previous-year average monthly employee wage, the calculation is capped at three times, and the length of service counts for at most twelve years +- Cost: No money. One calculation with a calculator, ten minutes. +- In plain language: The economic compensation due when you are dismissed is worked out like this: a full year of service gives one month's wage, six months or more but under a year counts as a full year, and under six months gives half a month. The monthly wage here is the average over the 12 months before you leave, and bonuses, allowances and subsidies all count in. Where the wage is above 3 times the local average monthly wage for the previous year, it is capped at 3 times, and the years of service count for at most 12 years. +- Benefit: The economic compensation set by the Labour Contract Law is the money the employer owes you when it dismisses you, and it is commonly called N. The method is 「每满一年支付一个月工资……六个月以上不满一年的，按一年计算；不满六个月的，向劳动者支付半个月工资」, that is, one month's wage for each full year of service, a full year counted where the service is six months or more but under a year, and half a month's wage where it is under six months. Where the monthly wage is above three times the local previous-year average monthly employee wage, the calculation is capped at three times, and the length of service counts for at most twelve years - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law of the People's Republic of China (in force from 2008-1-1), 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第四十七条）. -- Notes: Monthly wage means the average wage over the twelve months before the employment contract is terminated. Bonuses, allowances and subsidies all count, and it is not only the basic wage. This is the point companies most easily cut down. +- Notes: 「月工资」 means the average wage over the twelve months before the employment contract is terminated. Bonuses, allowances and subsidies all count in, and it is not only the basic wage. This is the part a company most easily pushes down. Whether this money has to pay income tax, see item 18 in this section (the tax on severance) -### 5. If the company dismisses you without 30 days' notice, it has to pay one more month's wage +### 5. If the company dismisses you without giving 30 days' notice, it has to pay one month's wage on top - Cost: No money. -- In plain language: If the company lets you go without 30 days' written notice, it has to give you one month's wage on top of N, which is what is usually called N+1. When you negotiate, list that extra month separately, and do not let the company fold it into N. -- Benefit: The Labour Contract Law gives an employer terminating a contract under that article two options. One is 「提前三十日以书面形式通知劳动者本人」, to notify the worker in person in writing thirty days in advance. The other is 「额外支付劳动者一个月工资后，可以解除劳动合同」, to pay the worker one month's wage and then terminate the contract. If they want you gone at once, they have to find the extra month's wage. That extra month is usually called payment in lieu of notice, and it is counted separately from N +- In plain language: Where the company dismisses you because you can no longer work after an illness, because you cannot do the work, or because the circumstances have changed substantially, and does not give you 30 days' written notice, it has to pay one month's wage on top of N. That is what is usually called N+1. When you negotiate, put that one month down as a separate line, and do not let the company fold it into N. +- Benefit: The Labour Contract Law gives an employer terminating an employment contract under this article two ways of doing it. One is 「提前三十日以书面形式通知劳动者本人」, to notify the worker in person in writing thirty days in advance. The other is 「额外支付劳动者一个月工资后，可以解除劳动合同」, to pay the worker one month's wage and then terminate the contract. If they want you gone at once, they have to find that extra month's wage. That extra month is usually called severance in lieu of notice, and it is a separate amount counted outside N. The one month is worked out on your wage in the last month, while N is worked out on the average wage over the 12 months before you leave, so the two use different bases - Evidence grade: A -- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law of the People's Republic of China (in force from 2008-1-1), 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第四十条）. -- Notes: So the usual outcome of a lawful redundancy is N+1. In the negotiation, work out N and the 1 separately, and do not let the company count the extra month inside N +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law of the People's Republic of China (in force from 2008-1-1), 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第三十六、三十九、四十、四十一、四十六、四十七、八十七条）. ; State Council (国务院) (2008). Regulations on the Implementation of the Labour Contract Law, 劳动合同法实施条例（国务院令第 535 号）第二十条：「用人单位依照劳动合同法第四十条的规定，选择额外支付劳动者一个月工资解除劳动合同的，其额外支付的工资应当按照该劳动者上一个月的工资标准确定」. +- Notes: That extra month only appears in the three situations in Article forty of the Labour Contract Law: after an illness or an injury not caused by work, where the medical period has ended and you can do neither your old job nor the one arranged for you; where you cannot do the work and still cannot after training or a transfer to another post; and where the objective circumstances at the time the contract was signed have changed substantially and negotiation fails. No other situation carries that month. An economic layoff (Article forty-one) only requires the company to explain the situation 30 days ahead to the trade union or to all employees. Where the company proposes termination and the two sides agree, it is N. Where you are dismissed for a fault of your own such as serious disciplinary breach, there is no compensation. Where the company terminates the contract unlawfully it is 2N. Where one of the three situations in Article forty applies, work out N and the 1 separately when you negotiate, and do not let the company count the extra month inside N ### 6. Where the company terminates the contract unlawfully, the compensation is twice the economic compensation standard @@ -57,7 +54,6 @@ This section is about money: how it is calculated, how the documents are signed, - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law of the People's Republic of China (in force from 2008-1-1), 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第八十七条）. - Notes: This is what is usually called 2N. Whether a termination is unlawful depends on the particular circumstances. Consult a legal aid service or a lawyer first, then decide whether to negotiate or go to arbitration; the route is in section 7 - ### 7. Do not sign 「resignation for personal reasons」, because the moment you sign there is no N - Cost: No money. The hard part is standing up to the pressure to sign on the spot. @@ -66,53 +62,44 @@ This section is about money: how it is calculated, how the documents are signed, - Evidence grade: B - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law of the People's Republic of China (in force from 2008-1-1), 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第三十八条、第四十六条）. - Notes: The line companies use is 「sign first so the process can move on, the compensation comes separately」. If you want it, have it written into the agreement on the spot; if it is not written down, do not sign. Not signing on the day has no legal consequence, so sleep on it and decide in the morning - ### 8. Before you leave, save your payslips, attendance records, employment contract, social insurance record and chat logs - Cost: No money. Half an hour to save the material. The hard part is remembering to do it before you go. -- In plain language: Once the computer and the accounts are handed back, the payslips, attendance records, contract, social insurance record and chat logs are gone for good. At arbitration, what you earned and whether you worked overtime still rest on the material in your own hands. Save your own employment records; do not take away the company's source code or customer list. +- In plain language: Once the computer and the accounts are handed back, the payslips, attendance records, contract, social insurance record and chat logs are gone for good. At arbitration, what you earned and whether you worked overtime still rest on the material in your own hands. What you are saving is your own employment records; do not take away the company's source code or customer list. - Benefit: Once the computer and the accounts are handed back, this material cannot be obtained any more. In arbitration the burden of proof leans towards the employer, but what you earned and whether you worked overtime still rest on the evidence in your own hands - Evidence grade: C - Sources: the author's own experience, no direct literature ; for the route to enforce your rights see section 7 -- Notes: What you save is your own employment records. Do not take the company's source code, customer list or technical documents while you are at it; for the risk of taking them see section 11 +- Notes: What you save is your own employment records. Do not take the company's source code, customer list or technical documents while you are at it; for the risk of taking them see section 11. When you leave, ask the employer for a certificate of leaving as well, see item 17 in this section (certificate of leaving) -### 9. Do two things as soon as you leave: register as unemployed and claim the unemployment insurance benefit, and read the non-compete agreement - -- Cost: No money. -- In plain language: Do two things as soon as you have left. First, register as unemployed and claim the unemployment insurance benefit. Second, read the non-compete agreement. The benefit runs from the day you register, so a month late means a month less. Only people who lost their job rather than choosing to leave can receive it, which is another reason for item 7 in this section (do not sign 「resignation for personal reasons」). -- Benefit: There are four conditions for receiving the unemployment insurance benefit: a full year of unemployment insurance contributions, losing the job rather than choosing to leave, which is 非因本人意愿中断就业, an interruption of employment not caused by the person's own will, having registered as unemployed, and still looking for work. The years you have paid decide whether the maximum is 12, 18 or 24 months. For how to claim online, see item 1 in section 7 (claim the unemployment insurance benefit online once you are unemployed). A non-compete restriction lasts at most 2 years. Through those 2 years the company has to pay you compensation every month. Where the contract does not say how much, it is 30% of your average wage over the 12 months before you left, and it may not be below the local minimum wage. Only where the company has not paid for 3 consecutive months for its own reasons may you ask for this agreement to be terminated, see item 12 in section 11 (what to do if you signed a non-compete agreement and the company does not pay the compensation). The non-compete restriction has a time limit, but the parts you still have to keep to stay in force until then -- Evidence grade: C -- Sources: the author's own experience, no direct literature ; the figures cited come from State Council (国务院) (1999). Regulations on Unemployment Insurance, 失业保险条例（国务院令第 258 号，第十四、十六、十七条：「失业保险金自办理失业登记之日起计算」）. and Standing Committee of the National People's Congress (全国人大常委会) (2012). Labour Contract Law, 劳动合同法（2012 年修正，第二十三、二十四条）. (reprinted by the Fushun Municipal Emergency Management Bureau) -- Notes: It is graded C because the action of 「registering on the day you leave」 has no direct literature; the figures in the sources field are the words of the regulations themselves. The unemployment benefit does not expire, but it runs from the day you register, so an earlier registration means one more day of money. For the claim materials and the online entry point, see section 7. For the compensation standard and the conditions for ending a non-compete restriction, see section 11. Signing 「resignation for personal reasons」 does not only cost you N, it loses the unemployment insurance benefit as well; see item 7 in this section (do not sign 「resignation for personal reasons」) -### 10. Before you take a post with dust, noise or chemicals, check whether the contract states the hazards; the three occupational health examinations are arranged and paid for by the employer +### 9. Before you take a post with dust, noise or chemicals, check whether the contract states the hazards; the three occupational health examinations are arranged and paid for by the employer - Cost: No money. When you sign, read through the part of the contract on 「notification of occupational disease hazards」 clause by clause. Keep your own copy of every examination report. The hard part is asking in person and getting the documents into your hands in person. - In plain language: When you sign, the employer must state truthfully in the contract what occupational disease hazards the post carries, how to guard against them, and what benefits attach. If it hides them from you, you can refuse to do the work, and the employer may not dismiss you for that. The three occupational health examinations, before you start, while you are in the post, and when you leave, are arranged and paid for by the employer. Without the examination on leaving, the employer may not terminate the contract. -- Benefit: The Law on the Prevention and Control of Occupational Diseases provides that when the employment contract is signed the employer has to make four things clear: what occupational disease hazards this work may produce, what consequences follow, how to guard against them, and what benefits attach. The words of the article are 「如实告知劳动者，并在劳动合同中写明，不得隐瞒或者欺骗」, tell the worker truthfully, write it into the employment contract, and do not conceal or deceive. If after the contract is signed you are moved to harmful work the contract did not describe, the employer has to tell you truthfully again and agree the contract terms with you. Where the employer fails to do this, 「劳动者有权拒绝从事存在职业病危害的作业，用人单位不得因此解除与劳动者所订立的劳动合同」, the worker has the right to refuse work that carries occupational disease hazards, and the employer may not terminate the contract on that ground. There are three examinations in all: before starting, while in the post, and on leaving. 「职业健康检查费用由用人单位承担」, the cost of the occupational health examination is borne by the employer, so the employer pays and the results are given to you in writing. A person who has not had the examination before starting may not be put on work involving exposure to occupational disease hazards. A person whose physical condition is unsuited to this work, and who therefore has a contraindication, may not be put on the contraindicated work either. Where health harm related to the job is found, the person has to be moved off the post and properly placed elsewhere. **「对未进行离岗前职业健康检查的劳动者不得解除或者终止与其订立的劳动合同」, an employment contract may not be terminated or ended with a worker who has not had the occupational health examination before leaving.** On leaving you 「有权索取本人职业健康监护档案复印件，用人单位应当如实、无偿提供，并在所提供的复印件上签章」, have the right to ask for a copy of your own occupational health monitoring file, and the employer must supply it truthfully and free of charge and put its seal on the copy. The employer has to make a true photocopy of that occupational health monitoring file, give it to you free, and stamp it (nationwide) +- Benefit: The Law on the Prevention and Control of Occupational Diseases provides that when the employment contract is signed the employer has to make four things clear: what occupational disease hazards this work may produce, what consequences follow, how to guard against them, and what benefits attach. The words of the article are 「如实告知劳动者，并在劳动合同中写明，不得隐瞒或者欺骗」, tell the worker truthfully, write it into the employment contract, and do not conceal or deceive. If after the contract is signed you are moved to harmful work the contract did not describe, the employer has to tell you truthfully again and agree the contract terms with you. Where the employer fails to do this, 「劳动者有权拒绝从事存在职业病危害的作业，用人单位不得因此解除与劳动者所订立的劳动合同」, the worker has the right to refuse work that carries occupational disease hazards, and the employer may not terminate the contract on that ground. There are three examinations in all: before starting, while in the post, and on leaving. 「职业健康检查费用由用人单位承担」, the cost of the occupational health examination is borne by the employer, so the employer pays and the results are given to you in writing. A person who has not had the examination before starting may not be put on work involving exposure to occupational disease hazards. A person whose physical condition is unsuited to this work, and who therefore has a contraindication, may not be put on the contraindicated work either. Where health harm related to the job is found, the person has to be moved off the post and properly placed elsewhere. **「对未进行离岗前职业健康检查的劳动者不得解除或者终止与其订立的劳动合同」, an employment contract may not be terminated or ended with a worker who has not had the occupational health examination before leaving.** On leaving you 「有权索取本人职业健康监护档案复印件，用人单位应当如实、无偿提供，并在所提供的复印件上签章」, have the right to ask for a copy of your own occupational health monitoring file, and the employer must supply it truthfully and free of charge and put its seal on the copy. The employer has to make a true photocopy of that occupational health monitoring file, give it to you free of charge, and seal the copy (nationwide) - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). Law on the Prevention and Control of Occupational Diseases, 职业病防治法（2018 年修正，第三十三、三十五、三十六条）. (reprinted by the Beijing Municipal People's Government) -- Notes: Three questions at the interview tell you roughly whether the company is legitimate: does this post carry occupational disease hazards, does the contract say so, and who pays for the examination before you start. If they cannot answer, or say 「just start and we will see」, weigh the risk yourself. The examination on leaving is the one most often forgotten: without it, the employer may not terminate or end the employment contract. That examination is also the key document for applying for an occupational disease diagnosis later, and the same goes for the copy of the file, so get it into your hands before you go. An occupational disease is handled as a work injury, and the benefits are covered from item 12 onwards. +- Notes: Three questions at the interview tell you roughly whether the company is legitimate: does this post carry occupational disease hazards, does the contract say so, and who pays for the examination before you start. If they cannot answer, or say 「just start and we will see」, weigh the risk yourself. The examination on leaving is the one most often forgotten: without it, the employer may not terminate or end the employment contract. That examination is also the key document for applying for an occupational disease diagnosis later, and the same goes for the copy of the file, so get it into your hands before you go. An occupational disease is handled as a work injury, and the benefits are covered from item 11 onwards. -### 11. Damage from dust, noise and toxic chemicals is irreversible: the employer must supply protective equipment, and you may refuse work with no protection +### 10. Damage from dust, noise and toxic chemicals is irreversible: the employer must supply protective equipment, and you may refuse work with no protection - Cost: No money. The hard part is wearing a particle mask, earplugs and goggles all the time, and refusing unprotected work, and the price of that is being called 「fussy」 by colleagues. - In plain language: Pneumoconiosis, noise-induced deafness and chemical eye burns are all occupational diseases in law, and once the lungs or the hearing are damaged there is no way back. You have the right to know what hazards the post carries and the right to demand protective facilities and protective equipment from the employer. You do not have to do unprotected work. If the employer cuts your pay or dismisses you for that, it does not count. - Benefit: From 2025-8-1 the occupational diseases in law are 12 major classes and 135 kinds, including 4 open clauses. The first class is occupational pneumoconiosis. It includes silicosis, coal workers' pneumoconiosis, graphite pneumoconiosis, carbon black pneumoconiosis, asbestosis and talc pneumoconiosis. It also includes cement pneumoconiosis, mica pneumoconiosis, potter's pneumoconiosis, aluminium pneumoconiosis, welder's pneumoconiosis and foundry worker's pneumoconiosis, plus one open clause. Occupational diseases of the ear, nose, throat and mouth include noise-induced deafness and blast-induced deafness. The occupational eye diseases include chemical eye burns and photochemical keratitis. All of this damage is irreversible, and no treatment can repair the lungs or the hearing. The law gives you four rights you can use on the spot. First, to know what occupational disease hazards are present in this workplace, what consequences they cause, and what protective measures should be taken. Second, to require the employer to provide qualified protective facilities and personal protective equipment and to improve the working conditions. Third, to criticise, report and accuse practices that break the laws on the prevention and control of occupational diseases, and practices that endanger life and health. Fourth, to 「拒绝违章指挥和强令进行没有职业病防护措施的作业」, refuse unlawful instructions and orders to carry out work with no occupational disease protection; you do not have to take an instruction that breaks the rules or take unprotected work. And 「因劳动者依法行使正当权利而降低其工资、福利等待遇或者解除、终止与其订立的劳动合同的，其行为无效」, where an employer reduces a worker's wages or benefits, or terminates or ends the employment contract, because the worker lawfully exercised a legitimate right, that act is void. If it cuts your pay, cuts your treatment or dismisses you because you used these rights, it does not count (nationwide) - Evidence grade: A - Sources: National Health Commission (国家卫生健康委), Ministry of Human Resources and Social Security (人力资源社会保障部), National Disease Control and Prevention Administration (国家疾控局) and All-China Federation of Trade Unions (中华全国总工会) (2024). Classification and Catalogue of Occupational Diseases, 职业病分类和目录（2025 年 8 月 1 日起实施）. ; for the count of 12 major classes and 135 kinds in the new catalogue see National Health Commission (国家卫生健康委) (2024). The new catalogue adjusts occupational diseases to 12 major classes and 135 kinds, 新版目录将职业病调整为 12 大类 135 种. ; Standing Committee of the National People's Congress (全国人大常委会) (2018). Law on the Prevention and Control of Occupational Diseases, 职业病防治法（第三十九条）. (reprinted by the Beijing Municipal People's Government) -- Notes: You have to recognise the right mask: only a particle respirator stops dust, and an ordinary gauze mask or a disposable surgical mask does not. If the employer does not supply earplugs for a noisy post, buy your own; a pair costs a few yuan, and lost hearing does not come back. 「Work two years, save up enough and leave」 does not hold with pneumoconiosis. Silicosis can appear many years after you stop working with dust, and it keeps getting worse. That is why the examination on leaving matters so much (see item 10 in this section, the three occupational health examinations arranged and paid for by the employer). The new catalogue also adds two classes. One is occupational musculoskeletal disorders, where the entry is carpal tunnel syndrome, limited to manufacturing workers who use the wrist repeatedly for long hours or do heavy work. The other is occupational mental and behavioural disorders, where the entry is post-traumatic stress disorder, limited to police, health workers, fire and rescue and other emergency response workers who take part in handling sudden incidents. For what to do when a chemical splashes on the skin, see item 21 in section 13. +- Notes: You have to recognise the right mask: only a particle respirator stops dust, and an ordinary gauze mask or a disposable surgical mask does not. If the employer does not supply earplugs for a noisy post, buy your own; a pair costs a few yuan, and lost hearing does not come back. 「Work two years, save up enough and leave」 does not hold with pneumoconiosis. Silicosis can appear many years after you stop working with dust, and it keeps getting worse. That is why the examination on leaving matters so much (see item 9 in this section, the three occupational health examinations arranged and paid for by the employer). The new catalogue also adds two classes. One is occupational musculoskeletal disorders, where the entry is carpal tunnel syndrome, limited to manufacturing workers who use the wrist repeatedly for long hours or do heavy work. The other is occupational mental and behavioural disorders, where the entry is post-traumatic stress disorder, limited to police, health workers, fire and rescue and other emergency response workers who take part in handling sudden incidents. For what to do when a chemical splashes on the skin, see item 21 in section 13. -### 12. If you are hurt at work or hit on the way to or from work, the first step is to get it recognised as a work injury, and if the employer does not report it you do +### 11. If you are hurt at work or hit on the way to or from work, the first step is to get it recognised as a work injury, and if the employer does not report it you do - Cost: No money, and one trip to the social insurance administration. You have to bring three documents: the application for recognition as a work injury, proof of the employment relationship, and a medical diagnosis certificate. - In plain language: An injury at work counts as a work injury. Being hit on the way to or from work, where the main responsibility is not yours, also counts. But without recognition as a work injury, the medical costs, the wages paid during recovery and the disability allowance all fall away. The employer has to report within 30 days. If the employer does not, you, or your family, or the trade union can report it yourself within one year to the social insurance administration. - Benefit: Without recognition as a work injury, the medical costs, the wages paid during recovery, which are called 停工留薪期工资, and the disability allowance all fall away. Where the employer does not report within 30 days, it pays the work-injury benefits due for that period out of its own pocket - Evidence grade: A -- Sources: State Council (国务院) (2010 revision). Regulations on Work-Related Injury Insurance (State Council Order No. 586), 工伤保险条例（国务院令第 586 号）. Article fourteen lists seven cases that 「应当认定为工伤」, must be recognised as a work injury, including 「（六）在上下班途中，受到非本人主要责任的交通事故或者城市轨道交通、客运轮渡、火车事故伤害的」, on the way to or from work, suffering harm in a traffic accident, urban rail accident, passenger ferry accident or train accident where the main responsibility is not the person's own. Article seventeen: 「所在单位应当自事故伤害发生之日或者被诊断、鉴定为职业病之日起30日内，向统筹地区社会保险行政部门提出工伤认定申请」, the employer where the person works shall, within 30 days of the day the injury happens or of the day the occupational disease is diagnosed or determined, apply to the social insurance administration of the pooling region for recognition as a work injury ; 「用人单位未按前款规定提出工伤认定申请的，工伤职工或者其近亲属、工会组织在事故伤害发生之日或者被诊断、鉴定为职业病之日起1年内，可以直接向用人单位所在地统筹地区社会保险行政部门提出工伤认定申请」, where the employer does not apply as the previous paragraph requires, the injured worker, or their close relatives or trade union organisation may, within 1 year of the day the injury happens or of the day the occupational disease is diagnosed or determined, apply direct to the social insurance administration of the pooling region where the employer is located ; 「用人单位未在本条第一款规定的时限内提交工伤认定申请，在此期间发生符合本条例规定的工伤待遇等有关费用由该用人单位负担」, where the employer does not submit the application within the time limit in the first paragraph, the employer bears the work-injury benefit costs incurred in the meantime that fall under these Regulations. Article nineteen: 「职工或者其近亲属认为是工伤，用人单位不认为是工伤的，由用人单位承担举证责任」, where the worker or their close relatives consider it a work injury and the employer does not, the burden of proof is on the employer. Article twenty: 「自受理工伤认定申请之日起60日内作出工伤认定的决定」, the decision recognising a work injury is made within 60 days of accepting the application. -- Notes: 1 year is a hard deadline, and after it there is no application to be made. If the employer says it is not a work injury, the burden is on the employer to prove it. You still have to keep material: photographs from the day, colleagues' contact details, the medical records and the attendance record. Article sixteen lists three cases that are not recognised: intentional crime, intoxication or drug use, and self-injury or suicide. Article fifteen adds 「视同工伤」, treated as a work injury: sudden illness causing death during working hours at the workplace, or death despite rescue within 48 hours. +- Sources: State Council (国务院) (2010 revision). Regulations on Work-Related Injury Insurance (State Council Order No. 586), 工伤保险条例（国务院令第 586 号）. Article fourteen lists seven cases that 「应当认定为工伤」, must be recognised as a work injury, including 「（六）在上下班途中，受到非本人主要责任的交通事故或者城市轨道交通、客运轮渡、火车事故伤害的」, on the way to or from work, suffering harm in a traffic accident, urban rail accident, passenger ferry accident or train accident where the main responsibility is not the person's own. Article seventeen: 「所在单位应当自事故伤害发生之日或者被诊断、鉴定为职业病之日起30日内，向统筹地区社会保险行政部门提出工伤认定申请」, the employer where the person works shall, within 30 days of the day the injury happens or of the day the occupational disease is diagnosed or determined, apply to the social insurance administration of the pooling region for recognition as a work injury ; 「用人单位未按前款规定提出工伤认定申请的，工伤职工或者其近亲属、工会组织在事故伤害发生之日或者被诊断、鉴定为职业病之日起1年内，可以直接向用人单位所在地统筹地区社会保险行政部门提出工伤认定申请」, where the employer does not apply as the previous paragraph requires, the injured worker, or their close relatives or trade union organisation may, within 1 year of the day the injury happens or of the day the occupational disease is diagnosed or determined, apply direct to the social insurance administration of the pooling region where the employer is located ; 「用人单位未在本条第一款规定的时限内提交工伤认定申请，在此期间发生符合本条例规定的工伤待遇等有关费用由该用人单位负担」, where the employer does not submit the application within the time limit in the first paragraph, the employer bears the work-injury benefit costs incurred in the meantime that fall under these Regulations. Article nineteen: 「职工或者其近亲属认为是工伤，用人单位不认为是工伤的，由用人单位承担举证责任」, where the worker or their close relatives consider it a work injury and the employer does not, the burden of proof is on the employer. Article twenty: 「自受理工伤认定申请之日起60日内作出工伤认定的决定」, the decision recognising a work injury is made within 60 days of accepting the application. Article fifty-five, for 「工伤认定申请不予受理的决定不服的」, dissatisfaction with a decision not to accept an application for recognition as a work injury, and 「对工伤认定结论不服的」, dissatisfaction with the conclusion of the recognition, 「可以依法申请行政复议，也可以依法向人民法院提起行政诉讼」, an administrative reconsideration may be applied for according to law, and an administrative suit may also be brought before the people's court according to law. ; Supreme People's Court (2014). Provisions on Several Questions Concerning the Trial of Administrative Cases Involving Work Injury Insurance, 关于审理工伤保险行政案件若干问题的规定（法释〔2014〕9 号）第七条：「由于不属于职工或者其近亲属自身原因超过工伤认定申请期限的，被耽误的时间不计算在工伤认定申请期限内」, where the application period is passed for a reason that is not the employee's own or that of their close relatives, the time lost is not counted inside the application period, and the cases listed are 「（一）不可抗力；（二）人身自由受到限制；（三）属于用人单位原因；（四）社会保险行政部门登记制度不完善；（五）当事人对是否存在劳动关系申请仲裁、提起民事诉讼」, force majeure, personal freedom being restricted, the cause lying with the employer, an imperfect registration system at the social insurance administrative department, and a party applying for arbitration or bringing a civil action over whether an employment relationship exists. +- Notes: The 1 year has to be taken seriously, so do not let it run out. But time lost for a reason that is not your own does not count inside that 1 year: force majeure, personal freedom being restricted, the cause lying with the employer, an imperfect registration system at the social insurance department, and applying for arbitration or bringing a court case to establish whether an employment relationship exists. So where the employer sits on the report until the period has passed, apply anyway and bring the evidence of the employer's delay. If the social insurance department does not accept the application, or you disagree with the conclusion, you can apply for an administrative reconsideration and you can also sue directly. Where the employer says it is not a work injury, the employer has to prove it. But you also have to keep material: photographs from the day, colleagues' contact details, the medical records and the attendance record. Article sixteen lists three cases that are not recognised: intentional crime, intoxication or drug use, and self-injury or suicide. Article fifteen adds 「视同工伤」, treated as a work injury: sudden illness causing death during working hours at the workplace, or death despite rescue within 48 hours. -### 13. Do not believe 「getting to my desk counts as a work injury」: if you feel suddenly unwell, call 120 first, and do not hurry to clock in +### 12. Do not believe 「getting to my desk counts as a work injury」: if you feel suddenly unwell, call 120 first, and do not hurry to clock in - Cost: No money. - In plain language: Hurrying to the company when you feel suddenly unwell will not get you a single extra yuan. Being treated as a work injury requires the illness to begin during working hours and at the workplace, so falling ill on the way without arriving is not covered in the first place. Nor is it covered if rescue goes beyond 48 hours. The few dozen minutes you lose trying to meet the conditions are paid for out of your own chance of survival. @@ -120,8 +107,7 @@ This section is about money: how it is calculated, how the documents are signed, - Evidence grade: A - Sources: Regulations on Work-Related Injury Insurance, article fifteen, paragraph one, item (one): 「在工作时间和工作岗位，突发疾病死亡或者在48小时之内经抢救无效死亡的」 is treated as a work injury ; article fourteen, item (six) only lists as a work injury the case of suffering harm in a traffic accident, urban rail accident, passenger ferry accident or train accident on the way to or from work where the main responsibility is not the person's own, and sudden illness on the way is not in it. - Notes: Both conditions have to be met together: during working hours, and at the workplace. Falling ill on the way without arriving does not apply. If someone falls ill at the post and is taken home or to hospital and dies despite rescue within 48 hours, it can still apply. What counts is where the person was when the illness began, not where they finally died. Short videos put this the wrong way round when they say 「get to the company and there is money」. What changes the outcome is whether those few dozen minutes after the illness began were spent in hospital. And 48 hours is a hard line: rescue going beyond 48 hours does not count instead. - -### 14. If the employer has not enrolled you in work-injury insurance, the benefits still exist, and the employer pays the same amounts in full +### 13. If the employer has not enrolled you in work-injury insurance, the benefits still exist, and the employer pays the same amounts in full - Cost: No money. If the employer does not accept it, you have to go to arbitration or even to court, which runs from several months. The hard part is that you have to keep it moving for those months. - In plain language: If the employer has not enrolled you in work-injury insurance, the benefits still exist. They are simply paid in full by the employer on the same items and the same standards, with nothing missing. Not enrolling adds a penalty on the employer: an order to make up the contributions, a late payment charge of five hundredths of one per cent per day, and if it is still not paid, a fine of 1 to 3 times the amount outstanding. 「The company bought no work-injury insurance, so it cannot pay」 is untrue. @@ -129,8 +115,7 @@ This section is about money: how it is calculated, how the documents are signed, - Evidence grade: A - Sources: Regulations on Work-Related Injury Insurance, article sixty-two, paragraph two: 「依照本条例规定应当参加工伤保险而未参加工伤保险的用人单位职工发生工伤的，由该用人单位按照本条例规定的工伤保险待遇项目和标准支付费用。」 The first paragraph of the same article orders the employer to enrol and make up the contributions within a deadline, and says 「自欠缴之日起，按日加收万分之五的滞纳金；逾期仍不缴纳的，处欠缴数额1倍以上3倍以下的罚款」. - Notes: 「The company bought no work-injury insurance for you, so it cannot pay」 is the commonest untrue sentence in this kind of case. Not enrolling only adds a penalty to the employer; your benefits do not shrink. - -### 15. Once the injury is stable, get a labour capacity assessment; the disability grade converts straight into money +### 14. Once the injury is stable, get a labour capacity assessment; the disability grade converts straight into money - Cost: No money. Going through the labour capacity assessment takes a stretch of waiting before and after. - In plain language: Once the injury is stable, have the labour capacity assessed, and the grade converts straight into money. The one-off disability payment is worked out on your own wage: grade five 18 months, grade six 16 months, grade seven 13 months, grade eight 11 months, grade nine 9 months, grade ten 7 months. Before the assessment result comes out, do not sign a 「lump-sum settlement」 agreement. @@ -139,7 +124,7 @@ This section is about money: how it is calculated, how the documents are signed, - Sources: Regulations on Work-Related Injury Insurance, article twenty-one: 「经治疗伤情相对稳定后存在残疾、影响劳动能力的，应当进行劳动能力鉴定」, where after treatment the injury is relatively stable and there is a disability affecting the ability to work, a labour capacity assessment shall be carried out ; article twenty-two: 「劳动功能障碍分为十个伤残等级，最重的为一级，最轻的为十级」, loss of labour function is divided into ten disability grades, the most serious being grade one and the least grade ten ; article thirty-six, the benefits for grades five and six, being a one-off disability payment of 18/16 months of the person's own wage and a monthly disability allowance of 70%/60% of the person's own wage ; article thirty-seven, the benefits for grades seven to ten, being 13/11/9/7 months of the person's own wage. - Notes: At grades five and six the employment relationship is kept, the person is still an employee of that employer, and the disability allowance is paid monthly. For grades seven to ten, when the contract ends or the person proposes to terminate it, there are two other payments: the one-off work-injury medical subsidy, paid by the fund, and the one-off employment subsidy, paid by the employer, with the amount set by the province. Do not sign a 「lump-sum settlement」 agreement before the assessment result comes out. -### 16. There are three payments on a death at work, and you have to tell them apart: the funeral subsidy, the pension for dependants, and the one-off death benefit +### 15. There are three payments on a death at work, and you have to tell them apart: the funeral subsidy, the pension for dependants, and the one-off death benefit - Cost: No money, and one set of formalities. - In plain language: A death at work brings three payments, and do not negotiate only one of them. The funeral subsidy is 6 months of the local previous-year average monthly employee wage. The pension for dependants is paid monthly on the deceased's own wage, at 40% a month for a spouse and 30% a month for each other dependant. The one-off death benefit is 20 times the previous year's national per-capita disposable income of urban residents, and that single figure applies nationwide. @@ -147,8 +132,7 @@ This section is about money: how it is calculated, how the documents are signed, - Evidence grade: A - Sources: Regulations on Work-Related Injury Insurance, article thirty-nine: 「（一）丧葬补助金为6个月的统筹地区上年度职工月平均工资；（二）供养亲属抚恤金按照职工本人工资的一定比例发给由因工死亡职工生前提供主要生活来源、无劳动能力的亲属。标准为：配偶每月40%，其他亲属每人每月30%，孤寡老人或者孤儿每人每月在上述标准的基础上增加10%……（三）一次性工亡补助金标准为上一年度全国城镇居民人均可支配收入的20倍。」. ; National Bureau of Statistics (国家统计局) (2026). Resident Income and Consumption Expenditure for 2025, 2025 年居民收入和消费支出情况. - Notes: The one-off death benefit is the same nationwide, and the amount is updated each year with the figures the National Bureau of Statistics publishes. Do not accept the argument 「the local standard here is lower」 when you negotiate the compensation. - -### 17. If you are bullied, insulted or obstructed at work for a long time, do not just endure it: write it down and keep the evidence, then take a complaint, a police report or arbitration according to what it was +### 16. If you are bullied, insulted or obstructed at work for a long time, do not just endure it: write it down and keep the evidence, then take a complaint, a police report or arbitration according to what it was - Cost: No money. It takes time to write it down: the date, who was present, the other person's exact words, and keep the originals of the chat screenshots and emails. Arbitration and court take several months. Changing job is the highest price of all, and whether to do it is yours to work out. - In plain language: People who are bullied at work have about 60% more risk of a heart attack or stroke later, and the more they are bullied the higher the risk. People who are bullied are about twice as likely as others to have thoughts of suicide a few years on. Being sworn at and put upon is not a small thing that passes once you have endured it. @@ -156,3 +140,20 @@ This section is about money: how it is calculated, how the documents are signed, - Evidence grade: B - Sources: Xu T et al. (2019). Workplace bullying and workplace violence as risk factors for cardiovascular disease: a multi-cohort study. European Heart Journal. ; Nielsen MB, Nielsen GH, Notelaers G, Einarsen S (2015). Workplace Bullying and Suicidal Ideation: A 3-Wave Longitudinal Norwegian Study. American Journal of Public Health. ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Labour Contract Law, 劳动合同法（第三十八条第二款、第四十六条、第八十八条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千零一十条）, full text reprinted by the Supreme People's Procuratorate ; Standing Committee of the National People's Congress (全国人大常委会) (2022 revision). Law on the Protection of Women's Rights and Interests, 妇女权益保障法（第二十五条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Law on the Mediation and Arbitration of Labour Disputes, 劳动争议调解仲裁法（第四十三条）, full text reprinted by the Ankang Municipal Human Resources and Social Security Bureau - Notes: The reason for the B is that the health figures come from follow-up records, not from a grouped experiment, and 「being bullied」 is reported by the person themselves. There are four steps you can take first. First, start writing it down today, and for recording see item 41 in section 8 (recording). Second, put the complaint in writing through the employer's internal channels and keep a record that you did. Third, if you are beaten, locked up or publicly humiliated, call the police. Fourth, if you are forced out over unpaid wages or failure to enrol in social insurance, work through item 7 in this section (do not sign a voluntary resignation) and item 8 in this section (save the evidence before you leave). For how the economic compensation is worked out, see item 4 in this section (work out N before a dismissal). Plain neglect, exclusion and obstruction mostly do not match any particular provision of law, and the routes open to you are mainly an internal complaint, a transfer to another post, and changing job. Arbitration runs for at most 60 days from acceptance, and a first-instance court case runs from 6 months, with the legal fees your own. If you can no longer hold on, ring 12356 first: see item 25 in section 1 (12356). This book does not argue that you should resign the moment you are upset; whether changing job is worth it is yours to work out. +### 17. When you leave, ask the employer for a certificate of leaving that states the term of the contract, the date you leave, your post and your years of service + +- Cost: No money. Ask for it in person before you go, or send one written message asking for it, and keep the record that you asked. The hard part is having to open your mouth about it when the relationship has already broken down. Where the employer puts it off and does not give it to you, you complain or apply for arbitration, which takes from a few tens of days to several months. +- In plain language: Whether you were dismissed, whether the contract simply ran out, or whether you resigned, the employer has to give you a certificate of leaving when you leave, and has to move your file and your social insurance over within 15 days. The certificate has to state the term of the contract, the date you leave, your post and how many years you worked there. You need it to register as unemployed. Where the employer does not give it to you, the labour department orders it corrected, and if you suffer loss because of it the employer has to pay it. +- Benefit: The Labour Contract Law provides that the employer must issue a certificate when the employment contract is terminated or ended. The article does not distinguish who proposed it: a contract that runs out and is not renewed, a resignation of your own accord, and a dismissal are all covered. The employer also has to complete the transfer formalities for your file and your social insurance relationship within 15 days. The implementing regulations set out four things the certificate must state: the term of the employment contract, the date of termination or ending, the post, and the years of service with that employer. The Regulations on Unemployment Insurance provide that an employee who becomes unemployed is to take the certificate issued by the employer and register as unemployed with the social insurance agency without delay. The unemployment insurance benefit runs from the day the registration is made. Where the employer does not issue a written certificate, the labour administrative department orders it corrected, and where it causes you loss the employer has to pay it. The law separately provides that where a new employer takes on somebody who has not yet terminated the contract with the old employer and causes the old employer loss, it bears joint and several liability. The certificate of leaving is your written evidence that you have already terminated the contract with the former employer (nationwide) +- Evidence grade: A +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2012 revision). Labour Contract Law, 劳动合同法（第五十、八十九、九十一条）. ; State Council (国务院) (2008). Regulations on the Implementation of the Labour Contract Law, 劳动合同法实施条例（国务院令第 535 号，第二十四条）. ; State Council (国务院) (1999). Regulations on Unemployment Insurance, 失业保险条例（国务院令第 258 号，第十六条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Law on the Mediation and Arbitration of Labour Disputes, 劳动争议调解仲裁法（第二、二十七、四十三、五十三条）. ; State Council (国务院) (2004). Regulations on Labour Security Inspection, 劳动保障监察条例（国务院令第 423 号，第十七条）. ; State Council (国务院) (2006). Measures for the Payment of Litigation Costs, 诉讼费用交纳办法（国务院令第 481 号，第六、十三、二十九条）. +- Notes: The article does not tie issuing the certificate to handing over the work; what is paid only after the handover is the economic compensation. Where the employer puts it off, send one written message asking again. Where it still does not give it to you, ring 12333 or complain to the labour inspection office of the human resources and social security bureau. A complaint is free, and once the case is opened it is finished within 60 working days, with a further 30 working days allowed for a complicated case. To make the employer pay you compensation for the loss, go to labour arbitration, which is also free and has to be applied for within 1 year. After accepting the case, arbitration closes within 45 days, with at most a further 15 days. You have to prove the loss yourself with evidence, such as a registration as unemployed that was held up for want of the certificate, or a refusal by the new employer. If you are not satisfied with the award, go to the court, where an ordinary procedure at first instance runs from 6 months. The court's acceptance fee for a labour dispute is 10 yuan per case, and legal fees are not part of the litigation costs, so you pay them yourself. The reason you left is not one of the four things the regulations require to be written. Where the certificate says 「个人原因主动辞职」, resigned of your own accord for personal reasons, and the fact is not so, raise it on the spot; for the reason, see item 7 in this section (do not sign a voluntary resignation). For how to claim the unemployment insurance benefit, see item 1 in section 7 (the unemployment insurance benefit). + +### 18. When you get severance on leaving, look at the income tax first: the part within three times the local previous-year average employee wage is exempt, the excess is worked out separately, and it is not added to that year's wages + +- Cost: No money. Look up the local previous-year average employee wage, then spend a few minutes checking how the employer has deducted the tax. The individual income tax app shows the filed detail of every item of income. +- In plain language: The one-off payment you get for terminating the employment contract is not taxed on the part within three times the local previous-year average employee wage. The excess is taxed on its own, and is not added to that year's wages. Where the employer has folded the whole amount into that month's wages and taxed it, ask the employer first, then ring 12366 and ask the tax authority. +- Benefit: A document of the Ministry of Finance and the State Taxation Administration provides that one-off compensation income received where an individual terminates the employment relationship with an employer, which includes the economic compensation, living allowances and other allowances. The part within three times the local previous-year average employee wage is exempt from individual income tax. The part over three times is not added to that year's comprehensive income and is taxed separately on the comprehensive income tax rate table. On that table the part of the year's income up to 36000 yuan is taxed at 3%, the part from 36000 yuan to 144000 yuan at 10%, and the bands above that rise to 45%. As an example, if the local previous-year average employee wage is 100,000 yuan, three times that is 300,000 yuan. If you get 360,000 yuan of compensation, only the 60,000 yuan above that figure is taxed, which comes to 3480 yuan. The same 360,000 yuan worked out entirely on this table would come to 58080 yuan. This has applied since 2019-1-1, and no end date is written (nationwide) +- Evidence grade: A +- Sources: Ministry of Finance and State Taxation Administration (财政部、税务总局) (2018). Notice on the Question of How the Preferential Policies Are to Be Applied after the Individual Income Tax Law Was Amended, 关于个人所得税法修改后有关优惠政策衔接问题的通知（财税〔2018〕164 号，第五条第（一）项、第八条、第九条）. (law and policy database of the State Administration of Taxation) ; Ministry of Finance and State Taxation Administration (财政部、税务总局) (2023). Announcement on Continuing to Apply the Individual Income Tax Policy for the Single Annual Bonus, 关于延续实施全年一次性奖金个人所得税政策的公告（2023 年第 30 号）. (law and policy database of the State Administration of Taxation) ; Standing Committee of the National People's Congress (全国人大常委会) (2018 revision). Individual Income Tax Law, 个人所得税法（第三、九条，个人所得税税率表一）. +- Notes: In the same document, the item on the single annual bonus carries an end date, which was later extended to 2027-12-31. The item on compensation for terminating the employment relationship carries no end date, and does not follow the bonus item. The document names only the economic compensation, living allowances and other allowances; it does not name severance in lieu of notice or the compensation for an unlawful termination. Whether those two count is not clear, so ring 12366 and ask the local tax authority. Which figure to use for the 「local previous-year average employee wage」 is also settled by the local tax authority. For how the compensation itself is worked out, see item 4 in this section (work out N before a dismissal), item 5 in this section (no 30 days' notice given) and item 6 in this section (the compensation for an unlawful termination). The one-off payment you get on early retirement or on being kept inside the enterprise under an arrangement has its own method and this item does not apply to it. diff --git "a/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" "b/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" index dc112540..30732d87 100644 --- "a/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" +++ "b/en/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" @@ -12,7 +12,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: A - Sources: Moon RY, Carlin RF, Hand I, Task Force on Sudden Infant Death Syndrome and the Committee on Fetus and Newborn (2022). Sleep-Related Infant Deaths: Updated 2022 Recommendations for Reducing Infant Deaths in the Sleep Environment. Pediatrics, 150(1), e2022057990. - Notes: 「床上不放任何软东西」means no pillows, no duvets, no cot bumpers, no soft toys and no shaping pillows. No breastfeeding pillow either, however safe it looks. A reclined cradle or a baby seat is not a place to sleep. Within this section, this is the item where the difference is largest and where you most need to hold your line against older relatives. - ### 2. Give the first hepatitis B shot within 24 hours of birth - Cost: No money. The hepatitis B vaccine is on the national immunisation programme, so it is already free. The hospital gives it on the day the baby is born, and you do not have to go anywhere else for it. @@ -21,7 +20,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: A - Sources: World Health Organization. Hepatitis B fact sheet. - Notes: If the mother's hepatitis B surface antigen is positive, the baby also needs a shot of hepatitis B immune globulin after birth. Settle this at the antenatal checks rather than after the birth. For the screening the mother herself needs, see section 1 - ### 3. Complete every vaccine on the national immunisation programme, free throughout; if you miss one, only the unfinished doses are made up - Cost: No money. The law says 「接种单位接种免疫规划疫苗不得收取任何费用」, that is, a vaccination provider may not charge any fee for vaccines on the national immunisation programme. You have to make several trips to a vaccination clinic according to the month age. Keep the vaccination certificate safe and do not lose it. @@ -30,7 +28,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2019). Vaccine Administration Law, 疫苗管理法（第四十八、四十九条）. (reprinted by the Yunnan Provincial Health Commission, 云南省卫生健康委员会); National Disease Control and Prevention Administration and National Health Commission (国家疾病预防控制局、国家卫生健康委) (2026). Questions and Answers on the Children's Immunisation Schedule for the National Immunisation Programme and the Explanations, 2026 edition, 《国家免疫规划疫苗儿童免疫程序及说明（2026 年版）》解读问答. - Notes: The most common mistake is 「孩子体质不好，先别打」, that is, the child has a weak constitution, so hold off for now. The official position is this. An "allergic constitution", a family history of allergy and a previous allergic reaction to food or medicine are none of them reasons to withhold a vaccine. Atopic dermatitis (eczema), urticaria, allergic rhinitis and bronchial asthma can all be vaccinated as long as the condition is stable. Antihistamines and inhaled steroids do not stop vaccination either. Physiologic and breast-milk jaundice in newborns are also fine. Only two things stop a child: a previous acute severe allergic reaction to the same vaccine, or a severe allergy to a known component of the vaccine. Children with immune deficiency, and children on chemotherapy or radiotherapy or immunosuppressants, need an individual assessment, so explain the situation at the vaccination clinic and decide there. For the first hepatitis B dose, see item 2 in this section. Non-programme vaccines you pay for yourself carry a vaccine fee and a vaccination service fee, which have nothing to do with the free schedule described here. - ### 4. Breastfeed only for the first 6 months, not even water; from 6 months add solid food and keep breastfeeding - Cost: No money, and you save money on formula. What it costs is time, because feeding takes a good part of every day. The hard part is keeping it up, since you cannot stop in the first 6 months. @@ -39,7 +36,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: A - Sources: World Health Organization. Breastfeeding. ; General Office of the National Health Commission (国家卫生健康委办公厅) (2024). Service Guide for Assessing Nutrition and Feeding of Infants and Young Children (for Trial), 婴幼儿营养喂养评估服务指南（试行）（国卫办妇幼函〔2024〕452 号）. - Notes: If breast milk is not enough or your body does not allow it, use formula. The gap between breast milk and formula is far smaller than the one for safe sleep. How often to add solid food: 1 to 2 times a day at 6 to 8 months, and 2 to 3 times a day at 9 to 12 months. What the child eats in a day should cover at least four of the seven categories. When a child has severe eczema or an egg allergy, whether to avoid peanuts, see item 12 in this section (do not avoid peanuts). - ### 5. Make up formula with water above 70 ℃, let it cool before feeding, and throw away what is left - Cost: No money. Each time you wait a few extra minutes for the feed to cool. The hard part is waiting for the water to come down before mixing, when the baby is crying with hunger. @@ -48,7 +44,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: A - Sources: Beary MA, Daly SE, Baker J, Snyder AB (2025). Assessing Hot Water Reconstitution Instructions and Labeling of Powdered Infant Formula to Ensure Cronobacter spp. Reduction. Journal of Food Protection, 88(9), 100571. ; Sima WG, Legesse T, Girma S, et al. (2025). Emerging microbial risks: Cronobacter sakazakii in powdered infant formula for infants under six months of age in Ethiopia. BMC Microbiology, 26, 307. ; World Health Organization & FAO (2007). Safe preparation, storage and handling of powdered infant formula: guidelines. - Notes: The 70 ℃ threshold comes from a public health guideline. Beary and colleagues put it this way: 「针对高危婴儿照护者的公共卫生指南建议用加热到至少 70 ℃（158 ℉）的水冲调配方奶粉以灭活微生物」, that is, public health guidelines for the carers of high-risk infants recommend preparing formula with water heated to at least 70 ℃ (158 ℉) in order to inactivate microbes. Preterm babies, babies with a low birth weight and babies under 2 months carry the highest risk. Do not use water that has sat in a thermos all day. Do not leave a made-up feed standing at room temperature until the next feed. - ### 6. No honey before 1 year old - Cost: No money @@ -57,7 +52,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: A - Sources: US CDC. Preventing Botulism. - Notes: What counts is honey water, complementary food mixed with honey, and cakes containing honey. The same logic applies to home-made pickles and fermented foods, which babies should not eat either. - ### 7. The vitamin K shot at birth is not optional - Cost: 0 yuan to a few tens of yuan. This is normally already covered by the hospital bill for the birth, and there is nothing extra to pay. @@ -75,7 +69,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: A - Sources: Pantell RH, Roberts KB, Adams WG, et al. (2021). Evaluation and Management of Well-Appearing Febrile Infants 8 to 60 Days Old. Pediatrics, 148(2), e2021052228. - Notes: At this age a serious bacterial infection can also show up as nothing but a fever. So looking 「精神还行」, bright and alert, is not a reason to watch the baby at home. Do not give a fever medicine on your own first and then wait and see. - ### 9. Never shake a baby, however tired and angry you are - Cost: No money. The hard part is stopping your hands when you are too tired or too angry. @@ -84,7 +77,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: B - Sources: Spies EL, Klevens J (2016). Fatal Abusive Head Trauma Among Children Aged <5 Years — United States, 1999–2014. MMWR 65(20):505-509. ; Choudhary AK et al. (2018). Consensus statement on abusive head trauma in infants and young children. Pediatric Radiology 48(8):1048-1065. - Notes: When you really are about to lose control, put the baby back in the cot and walk away for a few minutes. Letting the baby cry for a while is far safer than holding the baby and shaking. If you are caring for the baby alone and are chronically short of sleep, remember this one above all. For what to do with your own temper, see section 3 - ### 10. Judge nappies on three things rather than the brand: fit, how often you change them, and whether they have been named in a spot check - Cost: A few hundred to a few thousand yuan a month, depending on the grade you buy @@ -93,7 +85,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: C - Sources: State Administration for Market Regulation (市场监管总局) (2026-06-22). The State Administration for Market Regulation led the formation of a joint investigation team to check on 「婴幼儿纸尿裤甲酰胺问题」：「针对媒体反映的"婴幼儿纸尿裤甲酰胺问题"，市场监管总局、工业和信息化部、国家卫生健康委、国家疾控局高度重视，成立联合调查组，核查婴幼儿纸尿裤甲酰胺有关问题，并依法依规处理。有关情况将及时公布。」. ; reprinted by the Chinese Government Website (中国政府网). - Notes: When the joint investigation team was set up it only said 「有关情况将及时公布」, that the situation would be published in due course. TODO (to be verified: the conclusion of the investigation. As of 30 September 2026, no conclusion on nappies could be found on the SAMR website or in recent national spot-check notices). How to buy in practice: expensive does not mean safe, and imported does not mean compliant. Buy a small pack and try it for a week, look for nappy rash and marks on the skin, and only then buy in bulk. Nappy rash comes mainly from changing too infrequently and from poor breathability, and has less to do with the brand than people imagine. For how to check the spot-check notices, see section 5 - ### 11. For big items, work through borrow, then second-hand, then new, and do not buy the whole set at once - Cost: No extra money, and what you save usually adds up to thousands of yuan. The hard part is resisting the urge to buy everything at once. @@ -102,7 +93,6 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: C - Sources: the author's own experience, no direct literature; for the evidence on car seats, see section 1, and for impulse spending, see section 5 - Notes: This book neither recommends nor rules out confinement centres, early education classes or baby swimming pools. They are all large sums you can choose to spend or not, and the 24-hour cooling-off period described in section 5 applies to them. - ### 12. Severe eczema or an egg allergy is not a reason to avoid peanuts: add them early under a doctor's guidance, but never a whole nut - Cost: A jar of peanut butter is a few tens of yuan. You feed it a few times a week, which fits in with what you already do. Before you start, take the child to see a doctor once for an assessment. @@ -111,3 +101,12 @@ This section measures infant mortality and money. It takes in only the things th - Evidence grade: A - Sources: Du Toit G, Roberts G, Sayre PH, et al. (2015). Randomized trial of peanut consumption in infants at risk for peanut allergy. New England Journal of Medicine, 372(9), 803-813. ; General Office of the National Health Commission (国家卫生健康委办公厅) (2020). Core Messages on Health Education for Infant and Young Child Feeding, 婴幼儿喂养健康教育核心信息. ; Perkin MR et al. (2016). Randomized Trial of Introduction of Allergenic Foods in Breast-Fed Infants. New England Journal of Medicine, 374(18), 1733-1743. (the EAT trial of ordinary infants, referred to in the Notes) - Notes: **The form must be peanut butter thinned out or peanut powder mixed into complementary food. Never give a whole peanut.** The feeding guidance from the National Health Commission is plain about it: 「整粒花生、坚果、果冻等食物易吸入气管，引起窒息，婴幼儿应当避免食用」, that is, whole peanuts, nuts, jellies and similar foods are easily drawn into the airway and cause suffocation, and infants and young children should avoid eating them. For what to do when someone is choking and cannot speak, see item 26 in section 13. This applies only to high-risk children: those with severe eczema, or those already allergic to eggs. **See a doctor for an assessment before you start; do not try it at home.** Before enrolment the LEAP trial gave every child a skin test, and children whose skin test reaction was greater than 4 mm were excluded and never given peanuts. The evidence in ordinary children is much weaker. In a separate trial of 1162 ordinary breast-fed infants, the early-introduction group had 5.6% food allergy against 7.1% in the usual group on an intention-to-treat basis, which is not a significant difference (P=0.32). It is significant only among the children who actually followed the plan (2.4% against 7.3%). That way of counting tends to overstate the effect, and the authors themselves say the intention-to-treat analysis failed to show efficacy, while also showing that early introduction is safe. China's official feeding documents currently say nothing about whether to introduce allergenic foods early or to avoid them, so this item rests on the international trial evidence. Note the timing too: LEAP started at 4 months, while China starts complementary food at 6 months, see item 4 in this section (breastfeed only for the first 6 months). At what age a high-risk child starts and how to start, ask the doctor. + +### 13. Watch for these signs of jaundice in a newborn: yellow within 24 hours of birth, getting yellower, hard to wake and not feeding, pale stools — go to hospital the same day + +- Cost: Registration and a bilirubin test, a few hundred yuan at most. If phototherapy is needed, it is usually about two days in hospital. The hard part is making the trip even when the baby looks perfectly all right. +- In plain language: More than four in five newborns go a little yellow, and most of them clear it on their own. Go to hospital the same day if the yellow appears less than 24 hours after birth, or if it deepens, or if the baby is sleepier than usual and cannot be woken, will not feed, or the body goes floppy or stiff. Pale stools may mean biliary atresia, and that needs checking as soon as possible. Too much bilirubin damages the brain and leaves a disability for life. +- Benefit: The 2022 guideline of the American Academy of Pediatrics says that 「超过 80% 的新生儿会有某种程度的黄疸」, that is, more than 80% of newborns have some degree of jaundice. Too much bilirubin causes acute bilirubin encephalopathy and kernicterus. The guideline calls kernicterus 「永久的致残性神经系统疾病」, that is, a permanent disabling neurological disease. Its signs include cerebral palsy with involuntary twisting movements of the hands and feet, eyes that cannot look upwards, poorly developed enamel on the first teeth, and hearing damage. The guideline lists jaundice appearing within 24 hours of birth as a risk factor for developing severe jaundice. The National Health Service (NHS) of the United Kingdom gives parents two bands. The first band means calling the emergency number or going straight to the emergency department. It is a baby with jaundice who is also sleepier than usual or cannot be woken, will not feed, whose body is floppy or stiff, or who has jerking movements. A baby with jaundice less than 24 hours after birth counts in this band too. The second band means booking a doctor as soon as possible. It is a baby more than 24 hours old whose jaundice is not improving or is getting worse, whose urine is dark yellow or brown, or whose stool is pale cream in colour. Global estimates put the number of newborns in 2010 with Rh haemolytic disease (caused by incompatibility between the mother's and the baby's blood group) or with extreme hyperbilirubinaemia at about 480,000. 24% of those children died and 13% developed kernicterus. Of the children who survived kernicterus, more than 83% were left with one kind or more of impairment. Pale stools may mean biliary atresia, and the Kasai operation (葛西手术) has to be done within 60 days for the outlook to be clearly better. Taiwan has printed a stool colour card into the child health handbook of every newborn since 2004. The proportion having the operation within 60 days rose from 60% in 2004 to 74.3% in 2005. In a separate nationwide study of Taiwan, after the colour card was introduced the mortality of children with biliary atresia fell from 26.2% to 15.9%. This is a before-and-after comparison, not a grouped trial +- Evidence grade: B +- Sources: Kemper AR, Newman TB, Slaughter JL, et al. (2022). Clinical Practice Guideline Revision: Management of Hyperbilirubinemia in the Newborn Infant 35 or More Weeks of Gestation. Pediatrics, 150(3), e2022058859. ; NHS. Jaundice in babies (reviewed 2026-03-24). ; NICE (2010, revised 2023). Jaundice in newborn babies under 28 days (CG98), 1.2.6、1.7.1. ; Bhutani VK, Zipursky A, Blencowe H, et al. (2013). Neonatal hyperbilirubinemia and Rhesus disease of the newborn: incidence and impairment estimates for 2010 at regional and global levels. Pediatric Research, 74(Suppl 1), 86-100. ; Hsiao CH, Chang MH, Chen HL, et al. (2008). Universal screening for biliary atresia using an infant stool color card in Taiwan. Hepatology, 47(4), 1233-1240. ; Lee M, Chen SC, Yang HY, et al. (2016). Infant Stool Color Card Screening Helps Reduce the Hospitalization Rate and Mortality of Biliary Atresia: A 14-Year Nationwide Cohort Study in Taiwan. Medicine, 95(12), e3166. ; National Health and Family Planning Commission (国家卫生计生委) (2017). 国家基本公共卫生服务规范（第三版），0～6 岁儿童健康管理服务规范. +- Notes: Phototherapy (照蓝光, light therapy) shines a special kind of light on the baby's skin to turn the bilirubin into a form the body can get rid of easily. Do not substitute sitting in the sun at home: the American Academy of Pediatrics says it is hard to do in a way that is safe and does not burn the skin, so it does not recommend it as treatment. Do not give water or glucose water to bring the jaundice down either; the guideline rules that out too. You cannot judge how high the bilirubin is by looking at how yellow the skin is. The NICE guideline in the United Kingdom says not to rely on the eye alone: if you are not sure, go and have it measured. Babies born before 38 weeks, those whose brothers or sisters had phototherapy, and those who have not had enough breast milk are more likely to go deeply yellow. The follow-up follows the national basic public health service specifications: within 1 week of discharge a health worker visits the home. At 28 to 30 days after birth you go to the community health centre or the township health centre for a follow-up visit. Both checks look at the jaundice. If the doctor booked a bilirubin re-test when you were discharged, go at the time agreed. Where a full-term baby has been yellow for more than 14 days, or a preterm baby for more than 21 days, the NICE guideline requires one test for conjugated bilirubin and a look at whether the stool is pale. Jaundice itself does not delay the vaccinations, see item 3 in this section (vaccinations). For newborn screening, see item 13 in section 27 (heel-prick screening). \ No newline at end of file diff --git "a/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" "b/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" index 00811c4b..1c4da971 100644 --- "a/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" +++ "b/en/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" @@ -12,7 +12,6 @@ This section measures money and personal freedom, and several of the items also - Evidence grade: A - Sources: State Council (国务院) (2026). Provisions on the Administration of Exit and Entry, 关于出境入境管理的规定（国令第 841 号）. (Article two); State Council (国务院) (2023). Regulations on Consular Protection and Assistance, 领事保护与协助条例（国令第 763 号）. (Articles nineteen and twenty-one); Consular Affairs Department of the Ministry of Foreign Affairs. Safety notices, 安全提醒. - Notes: The notices change with the situation, so this book does not copy the list. Check that page yourself before you travel. As of 7 September 2026 the column still carried these 「暂勿前往」 notices: Eswatini (2026-08-25), Palestine (2023-10-10), Sudan (2023-04-17). Four more were Afghanistan (2023-02-17), Syria (2023-01-04), Peru (2022-12-29) and Somalia (2022-11-24). There are also separate notices for northern Myanmar and for some provinces of the Democratic Republic of the Congo. Travel agencies have a duty to inform you as well. Before joining a group tour you can ask for a written statement of the risk at the destination. - ### 2. Save 12308 and your local consular protection number in your phone, and copy them onto paper in your wallet - Cost: No money. Save two numbers and write one copy down. A few minutes. @@ -21,7 +20,6 @@ This section measures money and personal freedom, and several of the items also - Evidence grade: A - Sources: Ministry of Foreign Affairs and Chinese Consulate-General in Myanmar (外交部、中国驻缅甸使领馆) (2023). Consular reminder on the current security situation in northern Myanmar, 关于当前缅北地区安全局势的领事提醒. ; State Council Information Office (国务院新闻办) (2023). The State Council's regular policy briefing on consular protection and assistance. - Notes: The local embassy or consulate has its own consular protection number, and it is usually faster than the switchboard. Before you leave, look up the country on the Chinese Consular Affairs website and save that number alongside the switchboard number. If your phone is lost or runs flat, numbers saved only on it are no use. So write a copy on paper and keep it in your wallet, or give a copy to someone at home. - ### 3. Know what consular protection can and cannot do: it can visit you, it cannot get you out, and you pay the bill - Cost: No money. It takes a few minutes to read this boundary clearly. @@ -30,7 +28,6 @@ This section measures money and personal freedom, and several of the items also - Evidence grade: A - Sources: State Council (国务院) (2023). Regulations on Consular Protection and Assistance, 领事保护与协助条例（国令第 763 号）. （第九、十、十四、二十六条） - Notes: A consulate does not pay your fines, does not pay your medical bills and does not buy your ticket, and it cannot change the conclusion of a court in the country you are in. What it can give you is information on local lawyers, translators, medical care and funeral services, and help reaching relatives and finding relief. So put the money and the insurance in place before you travel, and do not build a plan on the idea that the state will cover you if something happens. - ### 4. Buy a policy that covers overseas medical treatment and medical evacuation, not only flight-delay cover - Cost: A short-term overseas travel policy costs a few tens to a few hundred yuan. @@ -39,7 +36,6 @@ This section measures money and personal freedom, and several of the items also - Evidence grade: C - Sources: State Council (国务院) (2023). Regulations on Consular Protection and Assistance, 领事保护与协助条例（国令第 763 号）. （第二十六条）; US Centers for Disease Control and Prevention (美国疾病控制与预防中心). Travelers' Health: Travel Insurance：「Medical evacuation insurance covers emergency transportation from a remote area to a high-quality hospital, which could otherwise cost more than $100,000.」. ; the rest is the author's own experience, no direct literature - Notes: The figure of $100,000 is a floor, not a range of costs. No official or industry published range of costs was found, so this book gives no specific range. Confirm three things before you buy. First, whether the country you are going to is covered. Second, whether high-risk activities such as diving, skiing and motorcycling are excluded. Third, whether the insurer settles with the hospital directly or pays you back after you pay up front. Buying a policy is just as necessary for visa-free countries. - ### 5. Treat every 「境外高薪招聘」 offer as fraud, and know that being tricked into telecom fraud can cost you the right to leave the country - Cost: No money. The hard part is turning down with your own hand a job that looks very good. @@ -48,7 +44,6 @@ This section measures money and personal freedom, and several of the items also - Evidence grade: A - Sources: State Council (国务院) (2026). Provisions on the Administration of Exit and Entry, 关于出境入境管理的规定（国令第 841 号）. （第三、四条）; Ministry of Justice, Ministry of Public Security and National Immigration Administration (司法部、公安部、国家移民局) (2026). Questions and Answers on the Provisions of the State Council on the Administration of Exit and Entry. - Notes: The warning signs are more or less fixed: a monthly wage far above the same job, no requirement for education or experience, no proper contract, being told to gather in a border city first, visa handling done for you and your passport kept by them. There is already a dedicated consular reminder for northern Myanmar. If you do go, your personal freedom and your passport are in someone else's hands and you cannot leave when you want to. - ### 6. Cash withdrawals abroad are capped at 100,000 yuan a year, counted across every card in your own name - Cost: No money. Estimate before you leave how much cash this trip needs, and use cards for the rest. The hard part is changing the habit: if a card will do, do not take out cash. @@ -58,14 +53,14 @@ This section measures money and personal freedom, and several of the items also - Sources: State Administration of Foreign Exchange (国家外汇管理局) (2017). Notice on Regulating Large Cash Withdrawal Transactions with Bank Cards Abroad, 关于规范银行卡境外大额提取现金交易的通知（汇发〔2017〕29 号, Articles one, two, three, four, six and seven, in force from 1 January 2018）. ; State Administration of Foreign Exchange (国家外汇管理局) (2017). Answers by the responsible officials of the foreign exchange authority to reporters on the questions of regulating large cash withdrawal transactions with bank cards abroad. ; China UnionPay International (中国银联国际). ATM cash withdrawals abroad, 境外ATM取款. - Notes: The cap is counted by calendar year and starts again on 1 January. Only withdrawals abroad stop; card payments are not affected. If you really need a lot of cash abroad, buy the foreign exchange first under the Administrative Measures on Foreign Exchange of Individuals, then take the foreign currency notes out of the country. Withdrawal itself has two traps, and the UnionPay International page 「境外ATM取款」 sets them out. First, a dual-logo card cannot use the UnionPay channel at some ATMs in some countries. Second, some overseas ATMs show a surcharge on the screen, and that money goes to the local acquirer and is 「与银联无关」, nothing to do with UnionPay. The same page also warns: 「由于信用卡取款需支付透支利息，建议持卡人境外取款使用银联借记卡」, that is, because a credit card withdrawal incurs overdraft interest, holders are advised to use a UnionPay debit card for cash withdrawals abroad. If the card is lost, swallowed or used fraudulently, see item 5 in section 14 (report the card lost and freeze it before calling the police). -### 7. Photograph your passport, visa and identity card and keep a copy in the cloud, and if they are lost, report first and then get a travel document +### 7. Photograph your passport, visas and ID card and keep a copy in the cloud; if they are lost, report it first and then get a travel document - Cost: No money. Take a few photographs before you leave. -- In plain language: If your passport is lost, a consulate can issue you a travel document for the journey home. You have to be able to prove who you are first. With photographs of the data page, of the visa pages and of your entry and exit records to hand, the replacement goes through much faster. You also have to report the loss locally and get a receipt, which the exit check and the insurance claim will both ask for. Phones are often lost together with the passport, so do not keep the photographs only on the phone. -- Benefit: If your passport is lost, the embassy or consulate abroad can issue you a travel document for the journey home, but you have to prove your identity first. With clear photographs of the data page, of the visa pages and of your entry and exit records to hand, the difference in how long the replacement takes is large. At the same time you have to report the loss locally and get a receipt. Many countries' exit checks and insurance claims ask for this document. The way into the consulate is the two consular protection numbers you saved in item 2 in this section +- In plain language: If you lose your passport, the embassy or consulate can issue you a travel document to get home. The condition is that you can prove who you are. With photographs of the biodata page, of the visa pages and of your entry and exit record to hand, the replacement is much faster. You also have to report the loss to the local police first and get a report receipt; immigration checks and insurance claims both want that piece of paper. The phone is often lost together with the passport, so do not keep the photographs only on the phone. +- Benefit: If you lose your passport, the embassy or consulate abroad can issue you a travel document to return home, but you have to prove your identity first. With clear photographs of the passport biodata page, of the visa pages and of your entry and exit record to hand, the time it takes to replace it differs a great deal. At the same time you have to report the loss to the local police and get a report receipt. Many countries require that document both at immigration checks and for insurance claims. The way in to the embassy or consulate is the two consular protection numbers saved in item 2 - Evidence grade: C - Sources: State Council (国务院) (2023). Regulations on Consular Protection and Assistance, 领事保护与协助条例（国令第 763 号）. （第十六条）; the author's own experience, no direct literature -- Notes: How to store them: take the photographs and put them in two different places, for example one copy in cloud storage and one sent to a relative. Print a paper copy as well and keep it away from the original. Photograph your bank cards and the prescriptions for the medicines you use regularly too. A phone and a passport are usually lost together, so storing them only on the phone is the same as not storing them. +- Notes: The way to store them is to put the photographs in two different places once you have taken them: one copy in a cloud drive and one sent to a family member. Also print a paper copy and keep it away from the original. Photograph your bank cards and the prescriptions of the medicines you use often as well. The phone and the passport are usually lost together, so storing them only on the phone is the same as not storing them. Some cloud drives ask for a verification code by text message, and if you lose the phone you may not be able to get in. Before you leave, try signing in on another device and check that it opens without this phone. Identity document photographs are sensitive information, so a storage method with encryption is safer. But nobody can open an encrypted method whose password has been forgotten, so the copy sent to your family cannot be left out. ### 8. Before driving abroad, check whether the destination recognises a Chinese licence; most 「国际驾照」 sold online are waste paper @@ -75,7 +70,6 @@ This section measures money and personal freedom, and several of the items also - Evidence grade: A - Sources: United Nations Treaty Collection. Convention on Road Traffic, Geneva, 19 September 1949 — status of treaties. - Notes: In most countries driving without a licence is a criminal offence, or it carries a heavy penalty. And if there is an accident the insurance will very likely refuse to pay, which is the real cost. There are two ways to check: look up the destination country's 「中国公民须知」 on the Chinese Consular Affairs website, or ask the car hire company for a written confirmation. Hong Kong, Macao, Taiwan and some countries have separate arrangements. - ### 9. Before you use an agency for a visa, study abroad or immigration, ask whether it is registered with the immigration authorities - Cost: No money. Ask once whether it is registered, and keep the written material. @@ -84,7 +78,6 @@ This section measures money and personal freedom, and several of the items also - Evidence grade: A - Sources: State Council (国务院) (2026). Provisions on the Administration of Exit and Entry, 关于出境入境管理的规定（国令第 841 号）. （第七、八、十、十二、十三条） - Notes: An agency that keeps saying 「包过」, guaranteed approval, 「保签」, guaranteed visa, and 「材料我们帮你做」, we prepare the documents for you, is already doing what is expressly forbidden. If it is caught, the mark is left on your visa record. A problem with an application you filled in yourself costs you at most a refusal this time round. Using false documents and being found to have submitted false information costs a good deal more. Policy advice and information lookup that are not done for money do not count as agency service. - ### 10. Do not write an invitation letter for a stranger, and do not let anyone handle entry and exit formalities in your name - Cost: No money. The hard part is turning down a favour to someone's face. @@ -93,7 +86,6 @@ This section measures money and personal freedom, and several of the items also - Evidence grade: A - Sources: State Council (国务院) (2026). Provisions on the Administration of Exit and Entry, 关于出境入境管理的规定（国令第 841 号）. （第三、十一条） - Notes: This is the same kind of risk as 「不帮陌生人带东西」 in section 8: once documents are issued in your name, the consequences land on you. Where a public servant or a member of the armed forces uses an agency to get foreign nationality or overseas residence status in breach of the rules, the agency may not only refuse; it also has to report the matter to the supervisory authority. - ### 11. Before you go abroad, send your itinerary, address and travelling companions to one family member and agree how often you will make contact - Cost: No money. Send one message before you leave. diff --git "a/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" "b/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" index 4e966744..669a950e 100644 --- "a/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" +++ "b/en/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" @@ -12,7 +12,6 @@ The first half of this section covers which money in an entertainment venue is w - Evidence grade: A - Sources: State Council (国务院) (2006). Regulations on the Administration of Entertainment Venues, 娱乐场所管理条例（国令第 458 号, Articles twenty and twenty-one）. ; State Council (国务院) (2002). Regulations on the Administration of Business Premises for Internet Access Services, 互联网上网服务营业场所管理条例（国令第 363 号, Article twenty-one）. - Notes: If any one of these is true, change venue: the private room is underground, there is only one exit, the corridor is piled with junk, or the emergency exit sign is unlit. A fire in a place like this means heavy smoke, narrow corridors and many people, and most of them have been drinking. The time you have to get out is measured in seconds. The regulation says the number to report to must be on the warning signs inside the venue. - ### 2. Ask for the price list before you order, which venues have to display, and refuse 「play first, pay later」 - Cost: No money. Ask for the list once. The hard part is asking the price in front of other people, which is embarrassing. @@ -21,7 +20,6 @@ The first half of this section covers which money in an entertainment venue is w - Evidence grade: A - Sources: State Council (国务院) (2006). Regulations on the Administration of Entertainment Venues, 娱乐场所管理条例（国令第 458 号, Articles twenty-nine and fifty-three）. - Notes: How to do it: ask for a price list when you enter the private room, paper or electronic, and photograph it as soon as you have it. After ordering, ask for a confirmed bill. If a waiter recommends a 「特调」, a special mix, or a 「果盘」, a fruit platter, always ask the price before ordering. Call 12315 there and then if there is a dispute about the bill. Do not pay first and defend your rights afterwards, because once you have paid, producing the evidence is much harder. - ### 3. If someone offers you 「something」 inside the venue, walk out at once; both letting someone use it and supplying it are not doing a friend a favour - Cost: No money. Stand up and walk out. The hard part is leaving the table in front of everybody. @@ -30,7 +28,6 @@ The first half of this section covers which money in an entertainment venue is w - Evidence grade: A - Sources: State Council (国务院) (2006). Regulations on the Administration of Entertainment Venues, 娱乐场所管理条例（国令第 458 号, Articles fourteen, thirty and forty-two）. - Notes: If somebody takes out an unidentified powder, a tablet or an electronic cigarette cartridge and asks you to 「试试」, try it, leave at once and do not stay to watch. The risk you carry personally is heavier than the venue's. Using drugs yourself is in itself subject to administrative punishment. Letting a friend 「use it for a moment」 in a private room you have booked, or in your own home, is harbouring another person for drug use, and that is a crime. - ### 4. Do not eat sweets and snacks handed to you by strangers, do not drink a drink that has been out of your sight, and do not take a cigarette cartridge passed to you - Cost: No money. Turn down everything handed to you. If your own glass has been out of your sight, get a new one. @@ -39,17 +36,7 @@ The first half of this section covers which money in an entertainment venue is w - Evidence grade: A - Sources: Supreme People's Procuratorate (最高人民检察院) (2021). Wujiang District, Suzhou, Jiangsu: prosecutors reveal new drugs dressed in trendy clothing. ; National Medical Products Administration, Ministry of Public Security and National Health Commission (国家药监局、公安部、国家卫生健康委) (2023). Announcement on Adjusting the Catalogue of Narcotic Drugs and Psychotropic Drugs (2023 No. 120). (reprinted by the Shanghai Municipal Drug Administration, which records that the original page on nmpa.gov.cn returned 412 on this machine); Standing Committee of the National People's Congress (全国人大常委会) (2025). Public Security Administration Punishments Law (2025 revision, Articles eighty-four and eighty-five). ; National People's Congress (全国人大) (1997). Criminal Law (Article three hundred and fifty-three). ; Standing Committee of the National People's Congress (全国人大常委会) (2007). Narcotics Control Law (Articles thirty-three, thirty-eight and forty-seven). (reprinted by the Guangdong Provincial People's Government). - Notes: This is not only about entertainment venues. School gates, gatherings with friends and long-distance coaches are the same. Whether what is handed to you is a sweet, a drink or a cigarette cartridge makes no difference. Being drugged is often followed by robbery, sexual assault or blackmail, and there are no published official figures for that, so this is a judgement from experience. Replacing a drink the moment it has been out of your sight saves far more trouble than arguing afterwards. For what to do when somebody in the venue produces an unidentified powder, tablet or cigarette cartridge, see item 3 in this section. For carrying things for other people, see item 29 in section 8. - -### 5. Use your own identity card at an internet café, and neither lend yours nor use someone else's - -- Cost: No money. -- In plain language: By law, an internet café must check and register the identity card of each person who gets online, and must record the details of the session, keeping them for at least 60 days. The public security and culture departments can ask for them at any time. While that period runs the café may not alter or delete the records. Whatever you use to get online, what was done on that machine in that period is recorded under that name. -- Benefit: The regulation reads: 「互联网上网服务营业场所经营单位应当对上网消费者的身份证等有效证件进行核对、登记，并记录有关上网信息。登记内容和记录备份保存时间不得少于 60 日」, that is, an operator of a business premises providing internet access services shall check and register the valid documents of the consumers using it, such as their identity cards, and shall record the relevant details of their use. The registration and the backup of the records must be kept for not less than 60 days. When the cultural administration department or the public security authority queries them lawfully, the café must supply them. During the retention period they may not be altered or deleted. So whatever you use to get online, what was done on that machine in that period is recorded under that name. -- Evidence grade: A -- Sources: State Council (国务院) (2002). Regulations on the Administration of Business Premises for Internet Access Services, 互联网上网服务营业场所管理条例（国令第 363 号, Article twenty-three）. -- Notes: This is the same thing as 「身份证不借人」 in section 9. Internet cafés may not admit minors, and a sign barring minors has to hang at a visible place by the entrance. Entertainment venues likewise do not admit minors. If you help a minor 「刷卡进网吧」, get into an internet café with your own card, the café is fined, but the record stays on your document. - -### 6. Do not book murder mystery games or escape rooms set inside a residential building or below the second basement level +### 5. Do not book murder mystery games or escape rooms set inside a residential building or below the second basement level - Cost: No money. Ask which floor the address is on before you book. - In plain language: A 2022 notice from five departments states that a script entertainment venue may not be set up inside a residential building, nor below the first underground level. These places are dark, full of partitions and full of flammable props, and the door often only opens once you have solved a puzzle. Once there is a fire, getting out is much harder than from an ordinary shop. Look at the address and the floor on the review page before you book, and change venue if it does not comply. @@ -57,8 +44,7 @@ The first half of this section covers which money in an entertainment venue is w - Evidence grade: A - Sources: Ministry of Culture and Tourism, Ministry of Public Security, Ministry of Housing and Urban-Rural Development, Ministry of Emergency Management and State Administration for Market Regulation (文化和旅游部、公安部、住房和城乡建设部、应急管理部、市场监管总局) (2022). Notice on Strengthening the Management of Script Entertainment Venues, 关于加强剧本娱乐经营场所管理的通知. - Notes: These places are dark, full of partitions and full of flammable props, and the door often needs a puzzle to be solved before it opens. Once there is a fire, getting out is much harder than from an ordinary shop. You can see the floor, and whether it is a residential building, on the review page before you book. If it does not comply, change venue. - -### 7. When your mood is bad, walk or run: the effect size of exercise against depression rises with the intensity +### 6. When your mood is bad, walk or run: the effect size of exercise against depression rises with the intensity - Cost: No money. Half an hour each time. The hard part is that when your mood is worst is exactly when you least want to move, so you have to push yourself. - In plain language: When your mood is bad, brisk walking or slow running shows the clearest improvement. Next come yoga, strength training and tai chi or qigong, and the differences between them are small. The higher the intensity, the better the effect, and strength training and yoga are the easiest to keep up. The evidence here is of low quality. What can be said is that moving beats not moving, so pick the kind you can keep up. @@ -67,7 +53,7 @@ The first half of this section covers which money in an entertainment venue is w - Sources: Noetel M, Sanders T, Gallardo-Gómez D, et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ, 384, e075847. - Notes:Disputed. The authors say so themselves: of the 218 studies, only one was clean enough not to be swayed by its design. On the standard rating scale, brisk walking and slow running count only as "low" certainty, and the rest as "very low". Participants cannot be blinded to exercise, and knowing they are exercising makes them feel better; that expectation cannot be removed. The direction of the conclusion still holds: moving beats not moving, keep the intensity up, and pick the kind you can keep up. This is the expansion of 「情绪低落时先动起来」 in section 3, and it does not replace proper treatment after a diagnosis. -### 8. When anxiety is affecting your life, a mindfulness-based stress reduction course is no worse than the usual anti-anxiety drug +### 7. When anxiety is affecting your life, a mindfulness-based stress reduction course is no worse than the usual anti-anxiety drug - Cost: The course costs a little money. It runs 8 weeks, one session a week, and you practise at home every day. The hard part is practising every day for 8 weeks running. - In plain language: In people diagnosed with an anxiety disorder, 8 weeks of mindfulness-based stress reduction improves symptoms almost as much as the anti-anxiety drug escitalopram. Nearly eighty per cent of the people on the drug had an adverse reaction, and some dropped out because of it. About fifteen per cent of the people on the course had one, and nobody dropped out. The course has to be a standardised 8-week course; an ordinary meditation app does not count. @@ -75,8 +61,7 @@ The first half of this section covers which money in an entertainment venue is w - Evidence grade: A - Sources: Hoge EA, Bui E, Mete M, Dutton MA, Baker AW, Simon NM (2023). Mindfulness-based stress reduction vs escitalopram for the treatment of adults with anxiety disorders: a randomized clinical trial. JAMA Psychiatry, 80(1), 13-21. - Notes: The trial used the standardised 8-week mindfulness-based stress reduction course (MBSR), and an ordinary meditation app is not the same thing. It applies to people who already meet the diagnostic criteria for an anxiety disorder, and if you need a doctor you still need one. If you have been putting off treatment out of worry about drug side effects, this is an option with comparable strength of evidence. - -### 9. To settle down in the moment, do 5 minutes of 「cyclic sighing」: a two-stage inhale and a lengthened exhale +### 8. To settle down in the moment, do 5 minutes of 「cyclic sighing」: a two-stage inhale and a lengthened exhale - Cost: No money. 5 minutes a day. - In plain language: 5 minutes of breathing practice a day, kept up for a month, improves mood and slows breathing more than mindfulness meditation of the same length. Of the breathing methods, 「循环叹息」, cyclic sighing, which lengthens the exhale, is the best one. To do it, breathe in through the nose, then take a short top-up breath to fill the lungs, then breathe out slowly through the mouth. This rests on a single trial, where the rating comes from the participant himself, and it costs nothing. @@ -84,8 +69,7 @@ The first half of this section covers which money in an entertainment venue is w - Evidence grade: B - Sources: Balban MY, Neri E, Kogon MM, et al. (2023). Brief structured respiration practices enhance mood and reduce physiological arousal. Cell Reports Medicine, 4(1), 100895. - Notes: To do it, breathe in through the nose, take a short top-up breath to fill the lungs, then breathe out slowly and fully through the mouth, and repeat for 5 minutes. There is only this one trial. It is done at home, whether it works is reported by the person filling in the questionnaire, and the number of people is small. The effect is only a little better than meditation and it does not solve problems. But it costs nothing, takes 5 minutes and can be done anywhere, so it is good value for money. - -### 10. Treat 「meeting people regularly」 as a health expense, and do not only reach out when your mood is bad +### 9. Treat 「meeting people regularly」 as a health expense, and do not only reach out when your mood is bad - Cost: A little money, plus the time spent meeting people. The hard part is arranging to see people when your mood is good. - In plain language: People who rarely see other people have a risk of death over the same period about thirty per cent higher. People who feel lonely are about a quarter higher, and people who live alone about thirty per cent higher, and the pattern is more marked among people under 65. All of these come from following people over time, and it may be that people who are already in poor health are the ones more likely to live alone. @@ -93,8 +77,7 @@ The first half of this section covers which money in an entertainment venue is w - Evidence grade: A - Sources: Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D (2015). Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspectives on Psychological Science, 10(2), 227-237. - Notes:Disputed. All of this comes from associations observed in groups of people rather than from a controlled trial, so cause and effect cannot be separated. People who are already in poor health are more likely to live alone and to have fewer social contacts. The authors treat social relationships as a health factor of the same order as smoking and obesity, but no randomised trial has shown that increasing social contact lowers mortality. This does not conflict with 「减少让你消耗的人际关系」 in section 3: that one cuts the relationships that drain you, and this one protects the few that restore your energy. - -### 11. If you live near green space, make more use of it: walking, a stroll and sitting down all count +### 10. If you live near green space, make more use of it: walking, a stroll and sitting down all count - Cost: No money. You do it on the way past. - In plain language: The better the greenery within 500 metres of your home, the lower the risk of death over the same period. For each 0.1 rise in the vegetation index it is about 4% lower. What is measured is the greenery around the home, not how many times you go to a park. People who live in greener neighbourhoods are often richer, and the air is better as well. So do not read this as going to a park extending your life; it is enough to take the green space around your home into account when you choose where to live. diff --git "a/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" "b/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" index cf9aa9ba..3b482a8b 100644 --- "a/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" +++ "b/en/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" @@ -12,15 +12,14 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). Compulsory Education Law (second revision of 2018, Articles eleven, fourteen and fifty-eight). (published by the Ministry of Education); State Council (国务院) (2002). Provisions on the Prohibition of Child Labour (State Council Order No. 364, Articles two, six, seven, ten and thirteen). - Notes: The person who benefits is yourself. Working after you turn 16 is lawful. But the jobs that are keen to take on a 16-year-old with no qualification are exactly the ones described in item 10 in this section (choosing skills that resist automation): fixed processes, the easiest for a machine to take over. The overtime pay, annual leave and work injury benefits described in section 19 all depend on there being an employment relationship and on evidence. A black job gives you neither, and those are the things you cannot get. - -### 2. Put "is studying worth it" into the mortality account too: each extra year of education lowers the adult risk of death by about 1.9% - -- Cost: No money. What it costs is the time and energy of those extra years of study; for the tuition, see item 4 in this section (secondary vocational tuition and grants). The hard part is carrying on without giving up in the middle. -- In plain language: Each extra year of study lowers the risk of death in adulthood by about 1.9% on average. The effect is clearest between 18 and 49, where each year is worth about 2.9%, and above 70 it is only 0.8%. People who can keep studying are often better off to begin with, and how much of this is the effect of the study itself cannot be separated. -- Benefit: All the earlier studies were gathered up and pooled together, a systematic review and meta-analysis. The search ran from month 1 of 1980 to month 5 of 2023, and finally included 603 studies, 70 locations in 59 countries and 10,355 observations. Education and adult mortality, which is deaths from all causes added together, show a dose-response relationship: the more study, the larger the fall. Each extra year of education lowers the risk of death by 1.9% on average (95% uncertainty interval 1.8 to 2.0, which means the true value falls roughly within this range). For people aged 18 to 49 each extra year lowers it by 2.9% (2.8 to 3.0). For people over 70 the fall is 0.8% (0.6 to 1.0). No difference was seen between men and women. Split by national level of development, using the Social Population Index, no difference was seen either. +### 2. Put 「whether studying does any good」 into the mortality account too: every extra year of education lowers the adult risk of death by about 1.9% + +- Cost: No money. What you spend is the time and the energy of those extra years of study; for the tuition see item 4 in this section (secondary vocational tuition and grants). The hard part is keeping it up without giving up in the middle. +- In plain language: Every extra year of study goes with a risk of death in adulthood that is on average about 1.9% lower. The effect is clearest between 18 and 49, where every extra year is about 2.9% lower; above 70 it is only 0.8% lower. People who can keep studying are often better off to begin with, and studies of this kind cannot tell how much of the effect is the study itself. +- Benefit: All the earlier studies were gathered together and calculated in one go (a systematic review and meta-analysis). The search ran from 1980-1 to 2023-5 and finally included 603 studies, 70 sites in 59 countries and 10355 observations. Education and adult mortality (death from all causes added together) follow a dose-response relationship (the more you study, the larger the drop). Every extra year of education lowers the risk of death on average by 1.9% (95% uncertainty interval 1.8–2.0, which is roughly where the true value lies). For people aged 18 to 49 every extra year gives 2.9% (2.8–3.0). For people above 70 it gives 0.8% (0.6–1.0). No difference was found between men and women. None was found when the figures were split by level of national development (the social population index) either - Evidence grade: A (disputed) - Sources: IHME-CHAIN Collaborators (2024). Effects of education on adult mortality: a global systematic review and meta-analysis. The Lancet Public Health, 9(3), e155-e165. -- Notes:Disputed. The authors report publication bias themselves (p<0.0001), which means the studies with the prettier results are likelier to get published. The results also differ a long way across the studies. Everything included here is observational follow-up, not a trial that assigned people to groups. Years of schooling are tangled up with family income, region and occupation, and this kind of study cannot separate how much of the 1.9% comes from the schooling itself and how much comes from the fact that people who can keep studying were better off to begin with. So it shows strongly that more schooling and a longer life point the same way, but it does not mean that if you go and study you will live 1.9% longer. The person who benefits is yourself. +- Notes:Disputed. The authors themselves report publication bias (p<0.0001), which means a study with a good-looking result is more likely to get published. The results also differ a good deal from one study to another. Everything included is follow-up observation, not a trial in which people are split into groups. Years of schooling are tangled up with family circumstances, region and occupation, and a study of this kind cannot separate how much of the 1.9% comes from the study itself from how much comes from 「people who can keep studying were better off to begin with」. The opposite direction cannot be separated either: people who were ill or had mental problems as children are more likely to drop out and more likely to die young, and that makes the effect of studying look larger than it really is. So it does show firmly that more study and living longer point the same way, but that is not the same as 「if you study you will live 1.9% longer」. The benefit is graded small: what a reader is usually looking at is three or four more years of study, and at 1.9% a year that comes to less than a tenth in all, which counts as small against the line used throughout this book. The person who benefits is yourself ### 3. Before deciding whether a qualification has lost its value, look at the national education structure: in every 100,000 people only 15,467 have a university-level education @@ -30,7 +29,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: National Bureau of Statistics (国家统计局) and the Office of the Leading Group for the Seventh National Population Census of the State Council (国务院第七次全国人口普查领导小组办公室) (2021). Communiqué of the Seventh National Population Census, 第七次全国人口普查公报（第六号）: education level of the population, 人口受教育情况. - Notes: These figures only say how scarce a qualification is among the population. Whether studying is worth it for a particular person has to be worked out with the method in item 6 in this section (how to calculate study or work). One more thing to note: the census counts the whole national population, old people and small children included. The share of young people who go on to university is much higher than this figure, so comparing it with your own age group overstates your advantage. Used the other way round, it does answer the claim that studying is useless, because the denominator is the whole labour force competing with you for the same jobs. The person who benefits is yourself. - ### 4. When you cannot afford it, work through the policy first: most secondary vocational tuition is already free, grants are 2300 yuan, and student loans go up to 20,000 yuan a year - Cost: No money. Before term starts, ask the school's student financial aid office once and hand in the papers once. @@ -39,7 +37,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Ministry of Finance, National Development and Reform Commission, Ministry of Education and Ministry of Human Resources and Social Security (财政部、发展改革委、教育部、人力资源社会保障部) (2012). Opinions on Expanding the Scope of the Tuition-Free Policy for Secondary Vocational Education and Further Improving the National Grant System (Cai Jiao [2012] No. 376). ; the relevant department heads of the Ministry of Finance, the Ministry of Education and the Ministry of Human Resources and Social Security (财政部、教育部、人力资源社会保障部有关司局负责人) (2024). Answers to reporters on the questions of adjusting the student financial aid policy. ; Ministry of Finance, Ministry of Education, People's Bank of China and National Financial Regulatory Administration (财政部、教育部、人民银行、金融监管总局) (2024). Notice on Adjusting and Improving the Relevant Policies on National Student Loans (Cai Jiao [2024] No. 188). - Notes: The people who benefit are yourself and your children. Which students the tuition waiver covers and how many bands the grant has are both set by each province. Private schools are subsidised up to the standard of the equivalent public course locally, and you still pay the difference, so check the local documents before you enrol. A student loan is a loan, not a grant, and it has to be repaid after you graduate. But it separates "I cannot pay the tuition now" from "are these years worth it", so do not treat running out of money as the conclusion. The training subsidies and temporary assistance available during unemployment are covered in section 7; if the household really has lost its income, apply for those first. - ### 5. Not getting into a general senior high school does not mean the road ends: secondary vocational schools have linked admissions and separate examinations, and skilled jobs may lower the education requirement - Cost: No money. Before you fill in the application choices, check the vocational education admissions platform in your province once. @@ -57,7 +54,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: B - Sources: Zhang J, Zhao Y, Park A, Song X (2005). Economic returns to schooling in urban China, 1988 to 2001. Journal of Comparative Economics, 33(4), 730-752. - Notes: Before you do the sum, take off the tuition waivers and grants in item 4 in this section (secondary vocational tuition and grants), and set against them the account in item 2 in this section (an extra year of education and mortality). The data runs up to 2001, covers urban areas only and cannot be used as today's figures. The global average of 9% in item 7 of this section is the same. If you really work it through this way, the answer differs from person to person. High tuition, a major that does not match the job and a long course can come out genuinely not worth it, but that shows only that this particular programme is not worth it, and says nothing about study being useless. Several things also sit outside the formula: the education thresholds set by law, see item 5 in this section (linked admissions and separate examinations in secondary vocational schools); resistance to being replaced, see item 10 in this section (choosing skills that resist automation); and whether you can change industry. The person who benefits is yourself. - ### 7. Keep the baseline in mind: the global average private return to an extra year of education, the part that lands on your own income, is about 9% a year - Cost: No money. Keep this figure in mind and use it as a ruler for weighing up any "learn something" plan. @@ -66,7 +62,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Psacharopoulos G, Patrinos HA (2018). Returns to investment in education: a decennial review of the global literature. Education Economics, 26(5), 445-458. ; World Bank Policy Research Working Paper 8402. - Notes: This is the global average for an extra year of study. It is not the return on signing up for a course, and it is not a Chinese figure; it is an order of magnitude for comparison. Its use is to give every training plan one ruler: if a course costs you 30,000 yuan and half a year, you should ask whether it can bring an income change equivalent to one year of formal education. Note too that a global average hides large differences between fields and between countries. - ### 8. Before paying for a certificate, check whether it is in the national catalogue of vocational qualifications or on the Ministry of Human Resources list of registered assessment bodies - Cost: No money. It takes a check of ten-odd minutes. @@ -75,7 +70,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Ministry of Human Resources and Social Security (人力资源社会保障部) (2022). Notice on the Special Governance of "Boutique Certificates" in Technical and Vocational Skills. - Notes: Check in this order. First, is this occupation in the National Catalogue of Vocational Qualifications (《国家职业资格目录》)? If it is not in the catalogue, it is not a national vocational qualification. Second, is the issuing body an assessment body registered with the human resources department? Third, go to the official platform for looking up skill talent assessment certificates and see whether the certificate number can be found. If it cannot, do not enrol at all. Anything carrying 「国际」, international, 「高级」, senior, or 「注册」, registered, and promising 「包过」, a guaranteed pass, or 「挂靠有收入」, income from affiliation, carries the highest risk. - ### 9. For training, go through the government subsidy channel first, and do not pay for a commercial course straight away - Cost: No money, or a large reduction. The training itself takes time. The hard part is resisting the urge to pay for a course on impulse and looking up the subsidy catalogue first. @@ -84,7 +78,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: State Council Employment Promotion and Labour Protection Leading Group (国务院就业促进和劳动保护工作领导小组) (2025). Guiding Opinions on Carrying Out Large-Scale Vocational Skill Improvement Training. - Notes: How to claim it and how much you can get are set by each province. The way in is the vocational training subsidy catalogue on the website of the local human resources and social security bureau, and the public training bases. The vocational training subsidies available during unemployment, where the same person may benefit at most 3 times, and the social insurance subsidies, are covered in section 7. People who are employed should ask their employer about the employee education funds first. Looking through the subsidy catalogue before choosing what to learn saves far more than choosing first and then hunting for a subsidy. - ### 10. When choosing a skill, look first at whether it needs hands and on-the-spot judgement, because those are hardest for automation to replace - Cost: No money. Before you choose, run the jobs you are considering through this dimension. @@ -93,7 +86,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: B - Sources: Frey CB, Osborne MA (2017). The future of employment: how susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, 254-280. - Notes:Disputed. The study is an unpublished working paper from 2013, and what it rates is whether a whole occupation can be handed to a machine. Later work broke the question down into individual tasks and got a much lower high-risk share. It also does not say in which year any of this happens. When you use it, do not fix on the figure of 47%; look at the three places where it says machines are stuck. On-site installation, maintenance and nursing are more resistant to replacement than pure paperwork, and that is exactly why. So "learning a skill" should first and foremost mean one that puts you on site, facing a non-standard environment. - ### 11. Before you start, check whether this skill has a graded assessment, and whether the grade can be turned into money - Cost: No money. One check is enough. @@ -102,7 +94,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: B - Sources: State Council Employment Promotion and Labour Protection Leading Group (国务院就业促进和劳动保护工作领导小组) (2025). Guiding Opinions on Carrying Out Large-Scale Vocational Skill Improvement Training. ; Ministry of Human Resources and Social Security (人力资源社会保障部) (2021). Implementation Plan for the "Skilled China Action". - Notes: This is the direction of policy, not a wage increase already being paid out, and how far it has been implemented varies a great deal between localities and employers. The part you can act on is this: when choosing a skill, check whether it has a corresponding national occupational standard and graded pathway, because only a skill with a pathway can later rise in pay by grade. A "skill" with neither a grade nor an industry entry requirement leaves you to find work on your own after you finish. - ### 12. Do not take "the pay is high online" as evidence of a shortage; look up the local list of occupations in short supply and the subsidy catalogue - Cost: No money. Look at two local documents. @@ -120,7 +111,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: C - Sources: State Council Employment Promotion and Labour Protection Leading Group (国务院就业促进和劳动保护工作领导小组) (2025). Guiding Opinions on Carrying Out Large-Scale Vocational Skill Improvement Training. - Notes: There is one criterion: after the course, is there a specific post you can send a CV to? A programme lasting a year and a half, after which you are told to "gain more experience first", costs far more than its price. The stopping condition described in section 4 applies here too: write it down before you start, "I stop if I have not found work by the time I reach this level". - ### 14. After studying, close the book and test yourself, and do not go back and reread - Cost: No money, and no increase in the total time. Take the time you would have spent rereading and use it to write from memory with the book shut, or to do questions. The hard part is that self-testing exposes what you did not remember on the spot, which feels worse than rereading. @@ -129,7 +119,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Yang C, Luo L, Vadillo MA, Yu R, Shanks DR (2021). Testing (quizzing) boosts classroom learning: A systematic and meta-analytic review. Psychological Bulletin, 147(4), 399–435. ; Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 14(1), 4–58. - Notes: Self-testing is there to force you to pull things out of memory, not to score you. How to do it: after finishing a section, shut the book and write the key points from memory, then turn back and check. Corrective feedback works better, so when you check the answers look at where you went wrong. The benefit is graded "large" on the basis of the effect size and the grading. Both of these sources measure grades rather than hours, so they have not been tested against a time-saving threshold. Self-testing and item 15 in this section (spread the same time across several days) are the two highest-rated methods in the same review, and they work well together. - ### 15. Spread the same time across several days rather than finishing it in one go - Cost: No money, and the total time does not change either. Only the schedule changes: the same 3 hours, split into three days of 1 hour each. The hard part is starting a few days ahead instead of putting it off to the last night. @@ -138,7 +127,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Cepeda NJ, Pashler H, Vul E, Wixted JT, Rohrer D (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354–380. ; Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques. Psychological Science in the Public Interest, 14(1), 4–58. - Notes: Best used on top of item 14 in this section (test yourself with the book shut): spread the self-testing itself across several days. Most of the material in the Cepeda meta-analysis is word-list memory and the like, and no general number of days is given for the best gap for complicated content. "The further away the test, the longer the gap" is the rule the text gives, with no specific schedule. Cramming the night before does work at the time; the problem is how fast it fades. - ### 16. Do not make underlining, rereading and summarising your main study method - Cost: No money, and it saves time. The hard part is that these are the easiest to do and give the strongest feeling of "I am studying". @@ -147,7 +135,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 14(1), 4–58. - Notes: Highlighting is fine for marking the places you are going to test yourself on. The problem is treating the highlighting itself as the learning. The review also makes clear that the bottom band does not mean zero: the conditions under which these work are narrow, and they need more research. For what to do instead, see item 14 in this section (test yourself with the book shut) and item 15 in this section (spread the same time across several days). - ### 17. Mix question types when you practise, and do not do twenty of one kind in a row - Cost: No money. Change from working through the workbook in order by question type to doing it in a mixed order. The hard part is that accuracy is lower on the spot when you mix them, which feels like going backwards. @@ -156,7 +143,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Brunmair M, Richter T (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators. Psychological Bulletin, 145(11), 1029–1052. - Notes: The review quoted in item 16 in this section (do not make underlining your main study method) rates interleaved practice in the middle band, on the grounds that there was not enough research at the time. Put the two together and the use is this: mix practice for content where you have to "recognise which kind this is and which method applies", such as question types in mathematics, case files and which rule of law applies; practise in blocks material you have to memorise, such as vocabulary words and definitions of terms. - ### 18. Do not pick a study method on the basis of "I am visual and he is auditory" - Cost: No money. You simply stop picking material and courses by this classification, and stop paying for a style assessment. @@ -174,7 +160,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Pan SC, Rickard TC (2018). Transfer of test-enhanced learning: meta-analytic review and synthesis. Psychological Bulletin, 144(7), 710-756. - Notes: The word "recite" normally runs two things together: going over the material once more, which is rereading and underlining, and actively recalling with the book shut. The first is rated lowest in the review that ranks ten study methods; to see the ratings themselves, see item 16 in this section (do not make underlining, rereading and summarising your main study method). What you want is the second: active recall with the book shut, described in item 14 in this section (test yourself with the book shut), and done across several days, see item 15 in this section (spread the same time across several days). How to do it is to work out first how you will use this later: whether you have to pick the right answer on a paper, or judge it on the spot, and then set questions for yourself in that form. It is also worth doing the reverse, because transfer in the reversed direction was the weakest in this meta-analysis, so both directions need practice. - ### 20. Work out which series and which grade you are in before you apply for a professional title, then find the application channel from the nature of your employer - Cost: No money. Look up your province's application notice for that year once. It takes about half an hour. @@ -183,7 +168,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Ministry of Human Resources and Social Security Order No. 40 (人力资源和社会保障部令第 40 号) (2019). Interim Provisions on the Administration of Professional Title Assessments. : Articles two, thirteen, fourteen, fifteen, sixteen, seventeen and eighteen; General Office of the Ministry of Human Resources and Social Security (人力资源社会保障部办公厅) (2020). Notice on Further Improving Professional Titles for Private Enterprises. : 「在专业技术人才密集的创业孵化基地、高新技术开发区、科技园区等地设立职称申报受理服务点，或通过人才中介服务机构、工商联、行业协会商会、学会等社会组织受理民营企业专业技术人才职称申报」「民营企业专业技术人才被派驻外地连续工作一年以上的，可按有关规定在派驻地申报职称评审」; Xinhua News Agency (新华社) (2024). Professional title assessments reach their peak: how to check the information, and what has changed? : 「我国共设有经济、工程、卫生、出版等27个职称系列」「符合条件的自由职业者，可根据属地原则申报参加当地人社部门组织的职称评审」; General Office of the CPC Central Committee and General Office of the State Council (中共中央办公厅、国务院办公厅) (2016). Opinions on Deepening the Reform of the Professional Title System. : 「不将论文作为评价应用型人才的限制性条件」「对职称外语和计算机应用能力考试不作统一要求」「符合条件的高技能人才，可参加工程系列专业技术人才职称评审」「公务员不得参加专业技术人才职称评审」 - Notes: There is no national list of what material to prepare. Each series and each province issues an application notice every year, and the list of material, the application dates and the achievement requirements are all written into it. So the first step is to find that year's notice on the website of the provincial human resources and social security department. A paper is no longer required in every case: the 2016 reform opinions say not to make a paper a hard threshold for applied professionals, and no unified requirement is set for the foreign language and computer exams in professional titles. Skilled workers have a route too: highly skilled people who meet the conditions may take part in professional title assessments in the engineering series. Civil servants may not take part in professional title assessments. - ### 21. Junior and intermediate titles such as accounting are won by a national unified exam, so check your education and years of work before you register - Cost: The registration fee is set by each province. Preparing for the exam takes several months of your spare time. @@ -192,7 +176,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: General Office of the CPC Central Committee and General Office of the State Council (中共中央办公厅、国务院办公厅) (2016). Opinions on Deepening the Reform of the Professional Title System. : 「专业技术人才取得职业资格即可认定其具备相应系列和层级的职称，并可作为申报高一级职称的条件。初级、中级职称实行全国统一考试的专业不再进行相应的职称评审或认定。」; Ministry of Human Resources and Social Security and Ministry of Finance (人力资源社会保障部、财政部) (2019). Guiding Opinions on Deepening the Reform of the Professional Title System for Accounting Personnel. : 「助理会计师、会计师实行全国统一的会计专业技术资格考试」「高级会计师采取考试与评审相结合方式，正高级会计师一般采取评审方式」; Ministry of Finance and others (财政部等) (2024). Notice on the Linkage Between Professional Degrees in Accounting and Professional Technical Qualifications in Accounting. - Notes: Whether your series is examined, assessed, or examined and assessed together is set by the "guiding opinions on deepening the reform of the professional title system" for that series. The Ministry of Human Resources and Social Security and the competent department of each industry have issued these series by series, and they can be searched for in the policy document library on the Chinese Government website. The registration dates and the papers for the exam are those given in that year's examination announcement. Civil servants may sit the accounting exam, but may not take part in accounting professional title assessments. - ### 22. Do not pay an agency to assess your title for you, do not buy a ghostwritten paper, and do not falsify your material: if it is checked out, the title is withdrawn and a record is kept for 3 years - Cost: No money. The hard part is resisting the advertisements promising a pass for money. @@ -201,7 +184,6 @@ This section measures money and time, and item 2 measures mortality as well. The - Evidence grade: A - Sources: Ministry of Human Resources and Social Security Order No. 40 (人力资源和社会保障部令第 40 号) (2019). Interim Provisions on the Administration of Professional Title Assessments. : Article thirty-nine; Ministry of Human Resources and Social Security (人力资源社会保障部) (2024). Interim Measures for the Supervision of Professional Title Assessments. : Articles five, fifteen, seventeen and twenty-seven; General Office of the Ministry of Human Resources and Social Security (人力资源社会保障部办公厅) (2022). Notice on Further Improving Professional Title Assessments. : 「未经备案的职称评审委员会，评审结果不纳入全国职称评审信息查询验证系统」; Xinhua News Agency (新华社) (2024). Professional title assessments reach their peak: how to check the information, and what has changed? : 「大家可通过www.12333.gov.cn网页、手机12333客户端、人力资源社会保障部门户网站、“人力资源和社会保障部”微信公众号等多种渠道进行登录」 - Notes: There is a way to tell real from false: a title from a proper assessment can be found in the national professional title assessment information query platform. The way in is the 12333 website, the 12333 mobile app and the website of the Ministry of Human Resources and Social Security. A title awarded by an assessment committee that was not registered cannot be found in that system. So anything promising 「包过」, a guaranteed pass, 「不用本人参加」, no need for you to attend, or 「证书查不到也能用」, usable even if the certificate cannot be found, can be judged as suspect. The person who mainly benefits is yourself, and the benefit is graded on the basis of avoiding a withdrawn title and a 3-year record of dishonesty. - ### 23. Getting a professional title does not mean a pay rise: first find out whether your employer appoints by post quota, or whether being awarded does not mean being appointed - Cost: No money. Ask the human resources department of your employer once. diff --git "a/en/book/24-\347\234\213\347\227\205.md" "b/en/book/24-\347\234\213\347\227\205.md" index ebfae8cd..a1e4b0f2 100644 --- "a/en/book/24-\347\234\213\347\227\205.md" +++ "b/en/book/24-\347\234\213\347\227\205.md" @@ -6,23 +6,23 @@ This section measures money and time. It covers how medical care itself works: w This section rests mainly on the Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》), issued in April 2026 by the General Office of the State Council (国务院办公厅). Tiered diagnosis and treatment means deciding which level of hospital to go to first according to how serious the illness is. That document ties going to the grassroots level directly to saving money. Local implementation is at different stages, and the text gives the national rules. For your own pooling region, follow what the local medical insurance and health authorities say. The pooling region is where your medical insurance is enrolled, usually one city. -### 1. See common illnesses at the community health centre first, get referred upwards level by level, and the inpatient deductible carries on counting +### 1. For an ordinary illness, start at the community clinic and go up by referral step by step from the grassroots level, and the inpatient deductible carries on -- Cost: No money. You have to make one trip to the community first. The hard part is resisting the urge to book a specialist straight away. -- In plain language: Go to the community first, and let the community refer you upwards to a large hospital for an admission. The part you pay yourself is then counted once. Go straight to a tertiary hospital for an admission and be referred back to the community for rehabilitation afterwards, and you may pay that part twice. Referred from a large hospital back into the community to carry on treatment, the same illness is not counted again during that period. -- Benefit: For insured patients referred upwards level by level through a primary healthcare provider, the inpatient deductible at the higher hospital can be counted continuously. The deductible is the part of each admission you pay yourself until medical insurance starts reimbursing. Continuous counting means the two hospitals together count it once, so you do not pay it again. Inpatients referred back down to the grassroots level by a higher hospital are not given a separate deductible within the same disease cycle +- Cost: No money. You have to make one trip to the community clinic first. The hard part is holding back the urge to go straight for a specialist's appointment. +- In plain language: Go to the community clinic first, and let the community clinic refer you upwards to the big hospital for the admission. The part of the bill you pay yourself counts once only. If you go straight to a tier-three hospital for the admission and are then referred back to the community clinic for rehabilitation, you may pay that part twice. Where you are referred back from the big hospital to the community clinic and stay there, it is not counted a second time for the same illness within that period. +- Benefit: For insured patients referred up step by step through primary care, the inpatient deductible at the higher hospital can be counted continuously. The deductible is the part of each admission that you pay yourself, up to the point at which medical insurance starts paying. Counting it continuously means the two hospitals together count it once, and you do not have to pay it again from scratch. An inpatient referred back down from the higher hospital to the grassroots level is not given a separate deductible for the same period of illness - Evidence grade: A -- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「统筹地区内经基层医疗卫生机构逐级转诊的参保患者，在上级医院的住院起付线可连续计算；由上级医院下转至基层医疗卫生机构的住院患者，同一疾病周期内不再另设住院起付线。」, that is, for insured patients in the pooling region referred upwards level by level through a primary healthcare provider, the inpatient deductible at the higher hospital may be counted continuously; inpatient patients referred down to a primary healthcare provider by a higher hospital are not given a separate inpatient deductible within the same disease cycle. -- Notes: If you go straight to a tertiary hospital for an admission and are then referred back to the grassroots level for rehabilitation, you may pay that part twice; going through a referral counts it once. Referral rules are set separately by each province, each city and each medical consortium, so ask the community before you go. +- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「统筹地区内经基层医疗卫生机构逐级转诊的参保患者，在上级医院的住院起付线可连续计算；由上级医院下转至基层医疗卫生机构的住院患者，同一疾病周期内不再另设住院起付线。」 — for insured patients in the pooling region referred up step by step through primary care, the inpatient deductible at the higher hospital may be counted continuously; an inpatient referred back down to the primary care institution is not given a separate inpatient deductible within the same period of illness. +- Notes: If you go straight to a tier-three hospital for the admission and then go back to the grassroots level for rehabilitation, you may pay that part twice, whereas going by referral counts it once. Referral rules are set separately by each province, each city and each medical consortium, so ask the community clinic before you go. Do not go to the big hospital any more for prescriptions in the stable phase of a chronic illness: this kind of ordinary outpatient service is being cut back at the big hospitals, and the community clinic can prescribe 12 weeks of medicine at a time; see item 5 in section 16 (the community clinic prescribes 12 weeks of medicine at a time). -### 2. Within one pooling region (where your medical insurance is enrolled), the lower the hospital level the higher the reimbursement rate, about 10 percentage points apart +### 2. In the same pooling region (where your medical insurance is enrolled), the lower the level of the hospital the higher the share reimbursed, with a gap of about 10 percentage points -- Cost: No money. The hard part is resisting the urge to run to a large hospital. -- In plain language: For the same inpatient bill, the lower the hospital level, the more you get reimbursed. Localities have to spread the levels apart, by about ten percentage points for each step up. When something that could be finished at the community is done at a large hospital, you pay about a tenth more of the same bill yourself. New medical insurance funds also tilt towards the grassroots level. -- Benefit: The document requires localities to spread out the inpatient reimbursement level across healthcare providers of different grades, in principle by about 10 percentage points per level. On that gap, every step up the hospital ladder on the same inpatient bill costs you about a tenth more of your own money. The medical insurance funds added each year also tilt towards the grassroots level +- Cost: No money. The hard part is holding back the urge to run to the big hospital. +- In plain language: On the same bill for an admission, the lower the level of the hospital, the more is reimbursed. Each place has to spread the levels apart, with about ten percentage points between each step up. Take a hospital bill of 10,000 yuan: if the reimbursed share falls from 80% to 70%, what you pay yourself rises from 2000 yuan to 3000 yuan. Newly added medical insurance funds also tilt towards primary care. +- Benefit: The document requires each place to spread apart the inpatient reimbursement level of medical institutions of different grades, in principle by about 10 percentage points per step. On that gap, going up one hospital level for the same bill adds about 10% of the reimbursable amount to what you pay yourself. Take a reimbursable bill of 10,000 yuan: the reimbursed share falls from 80% to 70%, so what you pay yourself rises from 2000 yuan to 3000 yuan. The extra 1000 yuan is a tenth of the total, but half more than what you paid before. The actual ratios differ from place to place, and the deductible and the ceiling also change the difference; the figure here is only an illustration based on the 10 percentage points in the document. Newly added annual medical insurance funds also tilt towards primary care - Evidence grade: A -- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「因地制宜适当拉开参保人员在不同等级医疗卫生机构的住院报销水平，原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距。」, that is, according to local conditions, suitably widen the inpatient reimbursement level for insured patients across healthcare providers of different grades, in principle leaving a gap of about 10 percentage points between each level within the pooling region. 「具备条件的地方，可按分级诊疗导向酌情拉开参保人员在不同等级医疗卫生机构的门诊报销水平。」, that is, localities in a position to do so may, in line with the tiered diagnosis and treatment policy, widen the outpatient reimbursement level across different grades as they see fit. -- Notes: For the same inpatient bill, the difference between what you pay at the grassroots level and at a tertiary hospital can run to several thousand yuan. Do not go up for something the community can finish. Not every illness belongs at the community health centre. +- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「因地制宜适当拉开参保人员在不同等级医疗卫生机构的住院报销水平，原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距。」「具备条件的地方，可按分级诊疗导向酌情拉开参保人员在不同等级医疗卫生机构的门诊报销水平。」 — in line with local conditions, appropriately widen the gap in inpatient reimbursement between medical institutions of different grades, in principle about 10 percentage points per step within the pooling region; where conditions allow, widen the gap in outpatient reimbursement between grades at the direction of tiered diagnosis and treatment. +- Notes: On the same bill for an admission, the difference between paying at the grassroots level and paying at a tier-three hospital can be several thousand yuan. Where the illness can be dealt with at the grassroots level, do not run upwards. Not every illness has to be seen at the community clinic. ### 3. If you need a large hospital, go through a grassroots referral or the hospital's referral centre, and do not buy appointments from scalpers @@ -32,7 +32,6 @@ This section rests mainly on the Measures on Accelerating the Building of a Tier - Evidence grade: A - Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「牵头医院要为基层医疗卫生机构预留一定比例的号源和床位，并及时接诊经基层转诊患者。」, that is, the lead hospital must reserve a set share of appointment slots and beds for primary healthcare providers and see patients referred up from them promptly. 「医疗机构要强化转诊服务统一管理，设立转诊中心或指定固定部门承担患者转诊服务工作，到2027年实现全覆盖。」, that is, healthcare institutions must strengthen unified management of referral services, set up a referral centre or name a fixed department to take on patient referral work, reaching full coverage by 2027. - Notes: The reserved appointments are for referred patients only, so if you cannot get one, go through a referral. A scalper is not only expensive; you may also end up in a department that does not match. If you cannot find the referral centre, ring the hospital's service line and ask. - ### 4. Before going out of province for treatment, ask at home first; in principle the need is assessed by a deputy chief physician or above - Cost: No money. One more question before you set off. @@ -41,17 +40,7 @@ This section rests mainly on the Measures on Accelerating the Building of a Tier - Evidence grade: A - Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「原则上由二、三级医院副主任医师及以上职称人员评估患者跨统筹地区、跨省异地就医的必要性。」, that is, in principle the need for a patient to receive treatment across pooling regions and across provinces is assessed by staff at deputy chief physician level or above at secondary and tertiary hospitals. 「进一步强化异地就医直接结算管理服务，跨省临时外出就医人员报销水平与参保地同级别医疗机构报销水平保持合理差异。」, that is, further strengthen direct settlement of treatment away from the enrolled area, keeping the reimbursement level for people going out of province temporarily for treatment at a reasonable difference from the level at providers of the same grade inside the enrolled area. - Notes: Temporary treatment away from home and long-term residence away from home are two different registrations, and they reimburse at different levels. For how to register and which illnesses can be settled directly, see section 16. Do not travel first and think about reimbursement afterwards. - -### 5. Stop using a tertiary hospital as a prescription shop; send stable chronic illness back to the grassroots level - -- Cost: No money. The hard part is dropping the habit of trusting only big hospitals. -- In plain language: From now on tertiary hospitals mainly deal with critical illnesses and difficult cases. Routine follow-ups for common illnesses, and ordinary outpatient clinics for stable chronic illness, will be cut back little by little. Ordinary clinics run by the same specialists move to the community. If you keep using a big hospital as a prescription shop, getting an appointment there will get harder and harder. -- Benefit: Tertiary hospitals must focus on critical illnesses and difficult and complex diseases, strengthen referral, joint consultation and inpatient services, and gradually cut back, as they see fit, ordinary outpatient clinics such as routine follow-ups for common illnesses and diagnosed, stable chronic illness. Appointments for that kind of ordinary clinic will be withdrawn little by little. Higher hospitals will extend the ordinary outpatient clinics of their specialist teams down to the grassroots level, so the same specialists hold their ordinary clinics in the community -- Evidence grade: A -- Sources: General Office of the State Council (国务院办公厅) (2026). Measures on Accelerating the Building of a Tiered Diagnosis and Treatment System (《关于加快建设分级诊疗体系的若干措施》). : 「三级医院要聚焦急危重症和疑难复杂疾病，加强转诊会诊和住院服务，逐步酌减常见病复诊和诊断明确、病情稳定的慢性病等普通门诊。」, that is, tertiary hospitals must focus on critical illnesses and difficult and complex diseases, strengthen referral, joint consultation and inpatient services, and gradually reduce, as they see fit, ordinary outpatient clinics such as routine follow-ups for common illnesses and for diagnosed, stable chronic illness. 「紧密型医联体内上级医院要在基层医疗卫生机构开设高血压、糖尿病、慢性阻塞性肺疾病等常见病、慢性病门诊，提供诊疗和健康管理服务，并将专家团队普通门诊向基层医疗卫生机构延伸。」, that is, inside a tightly integrated medical consortium, higher hospitals must open clinics for common illnesses and chronic illness such as hypertension, diabetes and chronic obstructive pulmonary disease at primary healthcare providers, supply diagnosis, treatment and health management there, and extend the ordinary outpatient clinics of their specialist teams down to the grassroots level. -- Notes: This is the direction of policy, not a suggestion. Ordinary clinics at big hospitals will get harder to book over time. Stable chronic illness can get 12 weeks of medicine at once at the grassroots level; see section 16. - -### 6. After every visit, keep your own copy of the records, the test reports and the images +### 5. After every visit, keep your own copy of the records, the test reports and the images - Cost: No money. A hospital may charge a copying fee under the rules, and the rate has to be published. The hard part is remembering to go and copy them after every visit. - In plain language: Reading and copying your own medical records is a right the law gives you. Outpatient records, inpatient notes, lab sheets, images, operation and anaesthesia records, pathology material and itemised bills can all be copied. The hospital runs a dedicated window for the requests and may charge only a copying fee. Changing hospital, seeing another doctor, claiming on private insurance or fighting a dispute all rest on this pile of paper. @@ -59,16 +48,16 @@ This section rests mainly on the Measures on Accelerating the Building of a Tier - Evidence grade: A - Sources: State Council (国务院) (2018). Regulations on the Prevention and Handling of Medical Disputes (《医疗纠纷预防和处理条例》) (State Council Order No. 701), Article sixteen: 「患者有权查阅、复制其门诊病历、住院志、体温单、医嘱单、化验单（检验报告）、医学影像检查资料、特殊检查同意书、手术同意书、手术及麻醉记录、病理资料、护理记录、医疗费用以及国务院卫生主管部门规定的其他属于病历的全部资料。」, that is, patients have the right to read and copy their outpatient records, inpatient notes, temperature charts, medical orders, lab sheets (test reports), medical imaging materials, consent forms for special examinations, consent forms for surgery, operation and anaesthesia records, pathology material, nursing records, medical costs and all other material belonging to the records as required by the health authority of the State Council. 「患者要求复制病历资料的，医疗机构应当提供复制服务，并在复制的病历资料上加盖证明印记……可以收取工本费，收费标准应当公开。」, that is, where a patient asks for the records to be copied, the healthcare institution must provide the copying service and stamp a certifying mark on the copy; it may charge a copying fee, and the rate must be published. 「患者死亡的，其近亲属可以依照本条例的规定，查阅、复制病历资料。」, that is, where the patient has died, close relatives may read and copy the records under these Regulations. Article fifteen: 「任何单位和个人不得篡改、伪造、隐匿、毁灭或者抢夺病历资料。」, that is, no unit and no individual may alter, forge, conceal, destroy or seize medical records. ; National Health and Family Planning Commission and National Administration of Traditional Chinese Medicine (国家卫生计生委、国家中医药管理局) (2013). Administrative Provisions on Medical Records in Healthcare Institutions (2013 edition) (《医疗机构病历管理规定（2013 年版）》): healthcare institutions 「应当指定部门或者专（兼）职人员负责受理复制病历资料的申请」, that is, must designate a department, or a member of staff full-time or part-time, to take the applications to copy records, and the copy is made with the applicant present and stamped with a certifying mark. - Notes: There is a dedicated window for copying, so you do not have to plead with a doctor. The list of what you can ask for is in Article sixteen of these Regulations, item by item. First the outpatient records, inpatient notes, temperature charts, medical orders, lab sheets and imaging. Then the operation and anaesthesia records, pathology material, nursing records and the itemised bill. Put what you get into one folder in date order. The most important items inside are the discharge summary, the operation record, the pathology report and the key images. Keep the readings for a chronic illness in a single notebook; see section 16. - -### 7. If you doubt the diagnosis or treatment, ask on the spot for the records to be sealed, with both sides present, a list drawn up and one copy each +### 6. If you doubt the diagnosis or treatment, ask on the spot for sealing the records, with both sides present, a list drawn up and one copy each - Cost: No money. Ask the hospital to seal the records and wait for both sides to turn up and complete the formalities. The hard part is saying that sentence in front of a doctor. - In plain language: If you have doubts about the treatment, you can ask for the records to be sealed. Sealing requires both the patient side and the doctor side to be present, and a list to be drawn up, signed and stamped, with one copy each. Where the records are not finished, seal the part already completed. If you suspect an infusion, a blood transfusion, an injection or a drug caused the harm, the bag of fluid, the vial and the giving set have to be sealed as well. -- Benefit: The Regulations provide that 「发生医疗纠纷需要封存、启封病历资料的，应当在医患双方在场的情况下进行」, that is, where records must be sealed or unsealed because a medical dispute has arisen, this must be done with both the patient side and the doctor side present. What is sealed may be the original or a copy, and the healthcare institution keeps it. Where the records are not complete and sealing is required, the completed part is sealed first and the rest sealed once it is finished. The healthcare institution must draw up a list of the sealed records, signed or stamped by both sides, with one copy each. Where an infusion, a blood transfusion, an injection or a drug is suspected of having caused harm, both sides must jointly seal and unseal the physical items at the scene. Where testing is needed, the two sides must jointly appoint an institution with testing qualifications. Where they cannot agree on a joint appointment, the health authority of the county-level people's government where the institution is located designates one. Where a transfusion is suspected of having caused harm and the blood has to be sealed, the institution must notify the blood station to send someone to attend. There is also a hard limit: 「任何单位和个人不得篡改、伪造、隐匿、毁灭或者抢夺病历资料。」, that is, no unit and no individual may alter, forge, conceal, destroy or seize medical records. Once the dispute is resolved after sealing, or 3 years pass with no further request to resolve it, the institution may unseal the records itself (nationwide, in force from 2018-10-1) +- Benefit: The regulations provide that "where records must be sealed or unsealed because a medical dispute has arisen, this must be done with both the patient side and the doctor side present". What is sealed may be the original or a copy, and the healthcare institution keeps it. Where the records are not complete and sealing is required, the completed part is sealed first, and the rest is sealed once it is finished. The healthcare institution must draw up a list of the sealed records, signed or stamped by both sides, with one copy each. Where an infusion, a blood transfusion, an injection or a drug is suspected of having caused harm, both sides must jointly seal and unseal the physical items at the scene. Where testing is needed, the two sides must jointly appoint an institution with testing qualifications. Where they cannot agree on a joint appointment, the health authority of the county-level people's government where the institution is located designates one. Where a transfusion is suspected of having caused harm and the blood has to be sealed, the institution must notify the blood station to send someone to attend. There is also a hard limit: "no unit and no individual may alter, forge, conceal, destroy or seize medical records". Once the dispute is resolved after sealing, or 3 years pass with no further request to resolve it, the institution may unseal the records itself (nationwide, in force from 2018-10-1) - Evidence grade: A - Sources: State Council (国务院) (2018). Regulations on the Prevention and Handling of Medical Disputes (《医疗纠纷预防和处理条例》) (State Council Order No. 701, Articles fifteen, twenty-four and twenty-five). -- Notes: Sealing and copying are two different things. Copying is what you should do as a matter of course; see item 6 in this section (keeping your own copy of the records and images). Sealing fixes the evidence when a dispute arises, and it guards against entries written or changed afterwards. If you suspect an infusion, a blood transfusion or a drug caused the harm, seal the bag of fluid, the vial and the giving set along with the records, and do not let anyone carry them off and bin them. Healthcare institutions must publish in a prominent place the routes, the procedures and the contact details for resolving medical disputes, and must set up a single complaints department. When something happens, go to that department rather than arguing with the doctor on duty. -### 8. For serious injuries go straight to the emergency triage desk, not to the registration queue +- Notes: Sealing and copying are two different things. Copying is what you should do as a matter of course; see item 5 in this section (keeping your own copy of the records and images). Sealing fixes the evidence when a dispute arises, and it guards against entries written or changed afterwards. If you suspect an infusion, a blood transfusion or a drug caused the harm, seal the bag of fluid, the vial and the giving set along with the records, and do not let anyone carry them off and bin them. Healthcare institutions must publish in a prominent place the routes, the procedures and the contact details for resolving medical disputes, and must set up a single complaints department. When something happens, go to that department rather than arguing with the doctor on duty. + +### 7. For serious injuries go straight to the emergency triage desk, not to the registration queue - Cost: No money. A few more steps inside the door to find the triage desk. Say the most dangerous symptom in one sentence. - In plain language: Emergency care is not ordered by who arrived first. Patients are sorted into four levels by how serious they are, and the sorting happens at the triage desk at the emergency entrance, not at the registration window. Skip the triage and you have no level in the system, however badly you are hurt, so you move along the queue at the speed of the queue. On arrival, report the most dangerous thing first: heavy bleeding, breathlessness, loss of consciousness, something embedded in the body, crushing pain in the chest. @@ -77,7 +66,7 @@ This section rests mainly on the Measures on Accelerating the Building of a Tier - Sources: General Office of the National Health Commission (国家卫生健康委办公厅) (2024). Notice on Issuing the Medical Quality Control Indicators for Six Specialties including Emergency Medicine (2024 edition) (《关于印发急诊医学等6个专业医疗质量控制指标（2024年版）的通知》(国卫办医政函〔2024〕150号)), Annex 1, indicator four, 「急诊分级分诊执行率」, that is, the rate at which emergency graded triage is carried out. ; Beijing Municipal Health Commission (北京市卫生健康委员会) (2019). Beijing Plan for Strengthening Emergency Pre-triage and Grading, and its Annex, the Beijing Standard for Emergency Pre-triage and Grading in Hospitals (Trial) (《北京市加强急诊预检分诊分级工作方案》and 《北京市医院急诊预检分诊分级标准（试行）》): 「坚持急危重症优先就诊原则。按照患者病情的严重程度，遵循从重到轻、从病情迅速变化到相对稳定的原则，合理安排患者就诊顺序。」, that is, keep to the principle that critical cases are seen first; according to how serious the patient's condition is, follow the order from severe to mild and from rapidly changing to relatively stable, and arrange the order of treatment accordingly. 「在急诊科入口的显著位置，设置急诊预检分诊区域」, that is, set up an emergency triage area in a prominent position at the entrance to the emergency department. - Notes: The four levels are the same nationwide. How long each level waits before being seen is set by each province and city, and the Beijing text is a version you can check word for word. The level turns on readings such as pulse, blood pressure and blood oxygen, and on the symptoms, not on how much pain you feel. Level 1 has hard criteria, and these are some of them. A heart rate above 180 per minute, or below 40 per minute. A systolic blood pressure below 70 mmHg, which is the upper number in a blood pressure reading. Blood oxygen below 80%. Shock, a confirmed heart attack, or an acute impairment of consciousness. If you think the triage has put you too low, add the most dangerous symptom and say it again. If your condition changes while you wait, go straight back to the triage desk to be assessed again. When things are genuinely critical, ringing 120 is faster than going yourself: the paramedic assesses you against the same standard first and tells the hospital to prepare (see section 13). -### 9. No money, no papers, and nobody who can say who you are: emergency care still has to come first +### 8. No money, no papers, and nobody who can say who you are: emergency care still has to come first - Cost: No money. - In plain language: The state has an emergency medical assistance fund that pays emergency treatment charges for two kinds of people: those whose identity cannot be established, and those who cannot pay. A hospital may not refuse or delay resuscitation, and an ambulance service may not decline to send an ambulance over money. So do not sit at home and endure it because you have no money, and do not be afraid to call an ambulance and owe the bill. This fund covers emergency treatment only. An admission afterwards still goes through medical insurance and medical assistance. @@ -85,29 +74,29 @@ This section rests mainly on the Measures on Accelerating the Building of a Tier - Evidence grade: A - Sources: General Office of the State Council (国务院办公厅) (2013). Guiding Opinions on Establishing the Emergency Medical Assistance Scheme (《关于建立疾病应急救助制度的指导意见》(国办发〔2013〕15号)). ; National Health and Family Planning Commission (国家卫生和计划生育委员会) (2014). Measures for the Administration of Pre-hospital Medical Emergency Care (《院前医疗急救管理办法》(委令第3号), Articles thirteen, twenty-two, twenty-three and twenty-five). - Notes: The fund pays only the emergency treatment part, and it is the hospital that applies to the fund afterwards, so there is nothing for you to collect at a window. If you have medical insurance, use it first as usual. The application process and the ceiling differ by province. In the transfer principles, the nearest place and the most urgent place come before the patient's wishes, so the ambulance does not necessarily take you to the hospital you named. When things are critical, the nearest place that can treat you is what counts. Where what you pay yourself afterwards is still too heavy, use medical assistance; see item 10 in section 7 (medical assistance). - -### 10. Get the disability assessment after treatment has finished, because doing it early gets you a lower grade +### 9. Get the disability assessment after treatment has finished, because doing it early gets you a lower grade -- Cost: The assessment fee runs from a few hundred yuan to a few thousand yuan, and the party that applies usually pays it up front. You have to wait until treatment has finished or the clinical outcome has settled, usually three to six months after the injury and sometimes longer. The hard part is resisting the urge to settle it while the injury is still healing. +- Cost: The assessment fee runs from a few hundred yuan to two thousand yuan, and the party that applies usually pays it up front. You have to wait until treatment has finished or the clinical outcome has settled, usually three to six months after the injury and sometimes longer. The hard part is resisting the urge to settle it while the injury is still healing. - In plain language: How much you get paid depends mainly on the grade, and the grade depends on how much functional impairment is left once treatment has finished. The standard says the assessment must wait until treatment has finished or the clinical outcome has settled. Go for an assessment while the bone has not healed or the plate is still in place and you are assessed as you are that day, which means lowering your own grade. An ordinary fracture that has healed without affecting function does not reach any grade. -- Benefit: The Standard on the Grading of Disability from Human Body Injury states that the timing of the assessment 「应在原发性损伤及其与之确有关联的并发症治疗终结或者临床治疗效果稳定后进行鉴定」, that is, the assessment should be carried out after treatment of the primary injury and of complications clearly related to it has finished, or after the clinical treatment outcome has settled. The standard 「将人体损伤致残程度划分为 10 个等级，从一级（人体致残率 100%）到十级（人体致残率 10%），每级致残率相差 10%」, that is, divides the degree of disability from human body injury into 10 grades, from grade one (a disability rate of 100%) to grade ten (a disability rate of 10%), with 10% between each grade. It turns on 「人体组织器官结构破坏、功能障碍及其对医疗、护理的依赖程度，适当考虑由于残疾引起的社会交往和心理因素影响」, that is, structural damage to tissues and organs, functional impairment, and the degree of dependence on medical and nursing care, with due consideration of the effect of the disability on social interaction and psychological factors. The fracture-related clauses among the ten grades include these. First: 「一椎体压缩性骨折（压缩程度达 1/3）或者粉碎性骨折」, that is, compression fracture of one vertebral body with compression of 1/3 or more, or a comminuted fracture. Second: 「骨盆两处以上骨折或者粉碎性骨折，畸形愈合」, that is, more than two fractures of the pelvis or a comminuted fracture, with malunion. Third: 「四肢任一大关节（踝关节除外）功能丧失 25% 以上」, that is, loss of more than 25% of the function of any major joint of the limbs, excluding the ankle. Disability at the same site and of the same kind may not be counted twice. Where there is more than one disability, the grades must be stated separately (nationwide, in force from 2017-1-1) +- Benefit: The Standard on the Grading of Disability from Human Body Injury states that the timing of the assessment "should be after treatment of the primary injury and of complications clearly related to it has finished, or after the clinical treatment outcome has settled". The standard "divides the degree of disability from human body injury into 10 grades, from grade one (a disability rate of 100%) to grade ten (a disability rate of 10%), with 10% between each grade". It turns on "structural damage to tissues and organs, functional impairment, and the degree of dependence on medical and nursing care, with due consideration of the effect of the disability on social interaction and psychological factors". The fracture-related clauses among the ten grades include these. First: "compression fracture of one vertebral body (compression of 1/3 or more) or a comminuted fracture". Second: "more than two fractures of the pelvis or a comminuted fracture, with malunion". Third: "loss of more than 25% of the function of any major joint of the limbs, excluding the ankle". Disability at the same site and of the same kind may not be counted twice. Where there is more than one disability, the grades must be stated separately (nationwide, in force from 2017-1-1) - Evidence grade: A -- Sources: Supreme People's Court, Supreme People's Procuratorate, Ministry of Public Security, Ministry of State Security and Ministry of Justice (最高人民法院、最高人民检察院、公安部、国家安全部、司法部) (2016). Announcement on the Issue of the Standard on the Grading of Disability from Human Body Injury (2016 4 18 published, 2017 1 1 in force), clauses 4.2, 4.4, 4.5, 5.10.6 and 6.2 of the standard. -- Notes: There are three different disability assessments, and they do not substitute for one another. Assault and other personal injury, and traffic accidents, go through the judicial Standard on the Grading of Disability from Human Body Injury, which is used to work out the disability compensation (see section 9). A work injury goes through the assessment of capacity for work, which uses a different set of ten grades and governs the benefits for a work injury (see section 19). To claim the disability benefits available to disabled people you need a disability certificate as well (see item 11, disability certificate). Have the records, the operation record and the follow-up images ready before the assessment (see item 6, keeping your own copy of the records and images); missing images are the commonest reason a grade comes out low. Where this injury is mixed up with a long-standing condition, the assessment opinion also has to say how much of the resulting disability this injury accounts for. That line has a direct effect on how much you are paid. +- Sources: Supreme People's Court, Supreme People's Procuratorate, Ministry of Public Security, Ministry of State Security and Ministry of Justice (最高人民法院、最高人民检察院、公安部、国家安全部、司法部) (2016). Announcement on the Issue of the Standard on the Grading of Disability from Human Body Injury (《关于发布〈人体损伤致残程度分级〉的公告》, published 2016-4-18, in force 2017-1-1), clauses 4.2, 4.4, 4.5, 5.10.6 and 6.2 of the standard. +- Notes: There are three different disability assessments, and they do not substitute for one another. Assault and other personal injury, and traffic accidents, go through the judicial Standard on the Grading of Disability from Human Body Injury, which is used to work out the disability compensation (see section 9). A work injury goes through the assessment of capacity for work, which uses a different set of ten grades and governs the benefits for a work injury (see section 19). To claim the disability benefits available to disabled people you need a disability certificate as well (see item 10, disability certificate). Have the records, the operation record and the follow-up images ready before the assessment (see item 5, keeping your own copy of the records and images); missing images are the commonest reason a grade comes out low. Where this injury is mixed up with a long-standing condition, the assessment opinion also has to say how much of the resulting disability this injury accounts for. That line has a direct effect on how much you are paid. -### 11. If treatment leaves real functional impairment, apply to the county-level disabled persons' federation at your registered place of residence for a disability certificate +### 10. If functional impairment really is left after treatment, apply for a disability certificate at the county-level disabled persons' federation where your household registration is -- Cost: The certificate itself carries no administrative charge. The assessment fee and the photo fee are in principle yours to pay, and you can apply for a reduction if you have difficulty. Take your identity card, your household register and 3 passport-size photos on a white background to the county-level disabled persons' federation at your registered place of residence. Once the assessment passes, the result is posted in the village or community for 5 working days, and the federation finishes its review within 10 working days. The hard part is making these trips and waiting out that time. -- In plain language: A disability certificate is the threshold for claiming disabled persons' benefits, not a certificate of injury. There are seven categories: vision, hearing, speech, limb, intellectual, psychiatric and multiple disabilities, each graded one to four. The process is to apply, have the application accepted, be assessed at a designated institution, be posted, be approved after review, and be issued, and it is valid for ten years and replaced free of charge on expiry. If you do not pass you do not pass, and this is not something that can be arranged through someone you know. -- Benefit: The disability certificate 「是认定残疾人及其残疾类别、残疾等级的合法凭证，是残疾人依法享有国家和地方政府优惠政策的重要依据」, that is, is the lawful evidence for establishing that a person is disabled and for the category and grade of the disability, and the main basis on which disabled people lawfully enjoy the preferential policies of the state and local governments. The standard for the assessment is the national standard Classification and Grading of Disabilities (《残疾人残疾分类和分级》, GB/T 26341-2010). The certificate rests on voluntary application and local administration. People with vision, hearing, speech, limb, intellectual, psychiatric or multiple disabilities may all apply, and the grade runs from one to four. The county-level federation takes the application and issues the certificate, on the basis of the assessment conclusion reached by a hospital or specialist institution designated by the province. Where the conclusion meets the standard, it is posted in the village or community for five working days, and the county-level federation 「在十个工作日内审核完毕」, that is, completes its review within ten working days. 「办理残疾人证不收取工本费」, that is, no administrative charge is collected for a disability certificate. The assessment and photo costs are 「原则上由申请人个人自理；有条件的地方可由当地财政予以补贴，对特殊困难的申请人应协调有关部门予以减免」, that is, in principle paid by the applicant; where local conditions allow they may be subsidised by the local finance, and for applicants in special difficulty the relevant departments must be coordinated to reduce or waive them. The certificate is valid for ten years and is replaced free of charge on expiry. Where the category or grade changes, you may apply to be assessed again. Where you object to the conclusion, you may apply to the prefecture-level federation within ten working days for a fresh assessment. Where the objection stands, the provincial expert committee on disability assessment gives the final conclusion (nationwide, in force from 2018-1-1) +- Cost: The certificate itself carries no handling fee. The assessment fee and the photo fee are in principle yours to pay, and you can apply for a reduction if you are in difficulty. Take your identity card, your household register and 3 photos two inches in size on a white background to the county-level disabled persons' federation where your household registration is. Once the assessment passes, the result is posted in the village or community for 5 working days, and the federation finishes its review within 10 working days. The hard part is making those trips and waiting out that period. +- In plain language: A disability certificate is the threshold for claiming disabled persons' benefits. It is not a certificate of injury. There are seven categories: vision, hearing, speech, limbs, intellectual, mental and multiple disabilities, each graded from one to four. The process is application, acceptance of the application, assessment at a designated institution, posting of the result, review and approval, and issue of the certificate. It is valid for ten years and is replaced free of charge when it expires. If you do not qualify, you do not qualify. This is not something an acquaintance can fix for you. +- Benefit: The disability certificate "is the legal proof for determining that a person is disabled and of their category and grade of disability, and is the important basis on which disabled people lawfully enjoy the preferential policies of the state and of local governments". The assessment standard is the national standard 《残疾人残疾分类和分级》 (GB/T 26341-2010). The certificate rests on two rules: applying is voluntary, and it is administered where the applicant lives. The seven categories are vision, hearing, speech, limbs, intellectual, mental and multiple disabilities, and the disability grade runs from one to four. The county-level disabled persons' federation takes the application and issues the certificate, on the basis of the assessment conclusion made by a hospital or professional body designated by the province. Where the conclusion meets the standard, it is posted in the village or community for five working days, and the county-level federation "finishes its review within ten working days". "The issue of a disability certificate carries no handling fee". The assessment and photo fees "are in principle to be borne by the applicant personally; where local conditions allow they may be subsidised by the local finance, and for applicants in special difficulty the relevant departments shall be coordinated to reduce or waive them". The certificate is valid for ten years and is replaced free of charge when it expires. Where the category or the grade changes, you may apply for a fresh assessment. Where you dispute the conclusion, you may apply to the prefecture-level or city-level disabled persons' federation for a fresh assessment within ten working days. If you are still dissatisfied, the provincial expert committee on disability assessment makes the final decision (nationwide, in force from 2018-1-1) - Evidence grade: A -- Sources: China Disabled Persons' Federation and National Health and Family Planning Commission (中国残疾人联合会、国家卫生和计划生育委员会) (2017). Measures for the Administration of the Disability Certificate of the People's Republic of China (2017 5 27 published, 2018 1 1 in force; Articles two, three, five, seven, nine, seventeen, nineteen, twenty-two and twenty-five). -- Notes: Work out whether it is worth it before you apply. The certificate does not hand you money by itself. What it connects you to is the two subsidies for disabled people: disabled people in means-tested income support households get the living subsidy, and those at grades one and two who need long-term care get the care subsidy; see item 8 in section 7 (the two subsidies). The certificate also connects you to local concessions on rehabilitation, assistive devices, tax, employment and transport. These standards differ a great deal by province, so check the local federation's website first and then decide whether the trip is worth it. It is one of three separate things and does not replace the other two: the judicial injury grade and the assessment of capacity for work (see item 10, getting the disability assessment after treatment has finished). The certificate is for the holder only and may not be lent to anyone else. A certificate whose condition no longer meets the standard is cancelled, and you may not apply again within one year of cancellation. Multiple disabilities are graded by the most severe category among them. -### 12. Thank a doctor who saved your life with a letter of thanks, a banner or a satisfaction rating, not with cash: the code bans the property, not the gratitude +- Sources: China Disabled Persons' Federation and National Health and Family Planning Commission (中国残疾人联合会、国家卫生和计划生育委员会) (2017). Measures for the Administration of the Disability Certificate of the People's Republic of China (《中华人民共和国残疾人证管理办法》, published 2017-5-27, in force 2018-1-1; Articles two, three, five, seven, nine, seventeen, nineteen, twenty-two and twenty-five). +- Notes: Work out the money before you decide whether to apply. The certificate itself pays nothing out. What it connects you to is the two allowances for disabled people: a living allowance for disabled people in means-tested income support households, and a care allowance for those at grade one or two who need long-term care; see item 8 in section 7 (the two allowances). The certificate also connects you to the local concessions on rehabilitation, assistive devices, taxes, employment and transport. These standards differ a great deal between provinces, so look them up on the local federation's website before you decide whether the trip is worth making. It is one of three different things, along with the disability grade from a judicial assessment and the assessment of capacity for work after a work injury, and none of them substitutes for another (see item 9, get the disability assessment after treatment has finished). The certificate is for the holder only and may not be lent to anyone else. A disability that no longer meets the standard has its certificate cancelled, and no fresh application is allowed within one year of the cancellation. Where someone has multiple disabilities, the grade is set by the most severe category among them. + +### 11. Thanking the doctor who saved your life: go by thank-you letter, banner and satisfaction rating, and not by red envelope, because the code bans property, not thanks -- Cost: No money. Write one letter of thanks, or fill in one rating form on the satisfaction platform, ten minutes or so. -- In plain language: Money handed over in advance buys no priority. Appointment slots, beds, scarce drugs and surgery slots are all barred from being traded for a benefit. Afterwards you are free to thank the doctor who saved your life, but do not send cigarettes, alcohol or a red envelope, because taking it is a breach of the rules. There are three ways of saying thanks that really reach the person: fill in a satisfaction rating properly, write a letter of thanks naming the doctor or give a banner, and keep the follow-up appointments. -- Benefit: Article seven of the Nine Standards for Clean Practice by Staff of Healthcare Institutions (《医疗机构工作人员廉洁从业九项准则》(国卫医发〔2021〕41号)): 「维护诊疗秩序，不破坏就医公平。坚持平等原则，共建公平就医环境。严禁利用号源、床源、紧缺药品耗材等医疗资源或者检查、手术等诊疗安排收受好处、损公肥私」, that is, maintain the order of treatment and do not damage fairness in access to care; uphold the principle of equality and build a fair environment for access to care; it is strictly forbidden to take any benefit by using medical resources such as appointment slots, bed slots, scarce drugs and consumables, or arrangements for examinations and surgery, and to seek private gain at public expense. Article eight: 「共建和谐关系，不收受患方『红包』。恪守医德、严格自律」, that is, build harmonious relations and do not accept 「red envelopes」 from patients; keep to professional ethics and strict self-discipline. The same article adds: 「严禁索取或者收受患者及其亲友的礼品、礼金、消费卡和有价证券、股权、其他金融产品等财物」, that is, it is strictly forbidden to ask for or accept from patients and their relatives gifts, cash, stored-value cards, securities, equity, other financial products or other property. Meals and entertainment paid for by the patient side are covered too: 「严禁参加其安排、组织或者支付费用的宴请或者旅游、健身、娱乐等活动安排」, that is, it is strictly forbidden to take part in banquets, travel, fitness or entertainment arranged, organised or paid for by the patient side. The document also requires healthcare institutions to carry their own responsibility, carry out inspections and 「坚决查处违规行为」, that is, investigate and punish violations firmly. Satisfaction ratings were written into the performance assessment of tertiary public hospitals. The basis is the Opinions of the General Office of the State Council on Strengthening the Performance Assessment of Tertiary Public Hospitals (《国务院办公厅关于加强三级公立医院绩效考核工作的意见》(国办发〔2019〕4号)). It states: 「三级公立医院绩效考核指标体系由医疗质量、运营效率、持续发展、满意度评价等4个方面的指标构成」, that is, the index system for assessing tertiary public hospitals consists of indices in four areas: quality of care, operational efficiency, sustainable development and satisfaction ratings. It also states: 「医院满意度由患者满意度和医务人员满意度两部分组成……」, that is, hospital satisfaction consists of two parts, patient satisfaction and staff satisfaction. It goes on to say that one should 「通过门诊患者、住院患者和医务人员满意度评价，衡量患者获得感及医务人员积极性」, that is, measure through the satisfaction ratings of outpatients, inpatients and staff how much patients feel they have gained and how motivated the staff are. There is also a single national platform for these ratings: 「国家建立公立医院满意度管理制度……」, that is, the state establishes a satisfaction rating system for public hospitals. The timetable is: 「2019年3月底前，全国三级公立医院全部纳入国家卫生健康委满意度调查平台。各地要应用国家卫生健康委满意度调查平台，将调查结果纳入三级公立医院绩效考核」, that is, by the end of March 2019, all tertiary public hospitals nationwide are brought into the satisfaction survey platform of the National Health Commission; localities are to use that platform and bring the results into the performance assessment of tertiary public hospitals. Hospitals must also handle complaints. The basis is Article two of the Measures for the Administration of Complaints by Healthcare Institutions (《医疗机构投诉管理办法》(国家卫生健康委令第3号), in force 2019). Handling complaints means that a patient, with regard to medical services, administration, quality and safety of care and the like, 「提出意见、建议或者投诉请求」, that is, puts forward an opinion, a suggestion or a complaint. The hospital is responsible for investigating, dealing with and replying. 「二级以上医疗机构应当设置医患关系办公室或者指定部门（以下统称投诉管理部门）统一承担投诉管理工作」, that is, healthcare institutions at secondary level and above must set up a doctor-patient relations office or designate a department, collectively called the complaints department, to take on the handling of complaints in one place. Its duties include 「统一受理投诉，调查、核实投诉事项，提出处理意见，及时答复患者」, that is, receiving complaints in one place, investigating and verifying the matters complained of, putting forward a view on how to handle them, and replying to the patient in good time. Healthcare institutions 「应当鼓励工作人员主动收集患者对医疗服务、医疗质量安全等方面的意见和建议」, that is, must encourage staff to collect patients' opinions and suggestions on medical services, quality and safety of care and the like (nationwide) +- Cost: No money. Writing a thank-you letter, or filling in one rating on the satisfaction platform, takes a little over ten minutes. +- In plain language: Paying money in advance buys no priority. Appointment slots, beds, drugs in short supply and arrangements for surgery are all forbidden as something to exchange for a favour. Afterwards, if you want to thank the doctor who saved your life, you may do so, but do not send tobacco, alcohol or a red envelope: if he accepts, that is a breach of the rules. There are three forms of thanks that really land on him: fill in the satisfaction rating carefully, write a thank-you letter by name or send a banner, and go back for the follow-up the doctor asked for. +- Benefit: The Nine Codes for Clean Practice by Staff of Medical Institutions (《医疗机构工作人员廉洁从业九项准则》, 国卫医发〔2021〕41 号), seventh item: "maintain the order of diagnosis and treatment, and do not undermine fairness in access to care. Uphold the principle of equality and build an environment of fair access together. It is strictly forbidden to accept benefits by using appointment slots, bed sources, drugs and consumables in short supply and other medical resources, or examinations, surgery and other arrangements for diagnosis and treatment, and to seek private gain at public expense". The eighth item: "build harmonious relations, and do not accept the patient's red envelope. Keep to medical ethics and strict self-discipline". The same item also states that it is "strictly forbidden to ask for or accept from patients and their relatives gifts, cash, gift cards, negotiable securities, equity, other financial products and other property". A meal or entertainment paid for by the patient side is the same: it is "strictly forbidden to take part in banquets, travel, fitness, entertainment or other activities arranged, organised or paid for by them". The document also requires medical institutions to carry out their main responsibility, run inspections and "firmly investigate and punish violations". Satisfaction ratings were written into the performance assessment of tertiary public hospitals. The basis is the Opinions of the General Office of the State Council on Strengthening the Performance Assessment of Tertiary Public Hospitals (《国务院办公厅关于加强三级公立医院绩效考核工作的意见》, 国办发〔2019〕4 号). The document says: "the performance assessment indicator system of tertiary public hospitals consists of indicators in 4 areas: quality of care, operational efficiency, sustainable development and satisfaction rating". It also states: "hospital satisfaction consists of two parts, patient satisfaction and staff satisfaction……" and goes on to say that the purpose is to "measure patients' sense of gain and the motivation of staff through satisfaction ratings of outpatients, inpatients and staff". This rating also has a national platform: "the state establishes a satisfaction rating system for public hospitals……". The timetable is: "by the end of March 2019, all tertiary public hospitals nationwide are brought into the satisfaction survey platform of the National Health Commission. Localities are to apply the National Health Commission satisfaction survey platform and bring the survey results into the performance assessment of tertiary public hospitals". Hospitals also have to handle complaints. The basis is the Measures for the Administration of Complaints in Medical Institutions (《医疗机构投诉管理办法》, National Health Commission Order No. 3, in force 2019), Article 2. Complaint management means that patients, over the medical service itself, the medical management and the quality and safety of care, "put forward opinions, suggestions or a complaint request". The hospital is responsible for investigating, handling and giving feedback. "Medical institutions at grade two and above shall set up a doctor-patient relations office or a designated department (collectively the complaints department) to take on complaint management in a unified way". The duties of the complaints department include "receiving complaints in a unified way, investigating and verifying the matters complained of, putting forward handling opinions, and replying to the patient in good time". Medical institutions "should encourage staff to collect patients' opinions and suggestions on medical services and on the quality and safety of care actively" (nationwide) - Evidence grade: A -- Sources: National Health Commission, National Healthcare Security Administration and National Administration of Traditional Chinese Medicine (国家卫生健康委、国家医保局、国家中医药局) (2021). Notice on Issuing the Nine Standards for Clean Practice by Staff of Healthcare Institutions (《关于印发医疗机构工作人员廉洁从业九项准则的通知》(国卫医发〔2021〕41号)). State Council Gazette ; General Office of the State Council (国务院办公厅) (2019). Opinions on Strengthening the Performance Assessment of Tertiary Public Hospitals (《关于加强三级公立医院绩效考核工作的意见》(国办发〔2019〕4号)). ; National Health Commission (国家卫生健康委) (2019). Measures for the Administration of Complaints by Healthcare Institutions (《医疗机构投诉管理办法》(委令第3号)). -- Notes: The person who benefits is yourself, and so is the person who saved you. Wanting to say thank you after a night of resuscitation is ordinary, and you may say it; only the form has to change. A red envelope or a case of cigarettes shifts the risk onto the other person, while a letter of thanks, a banner and a satisfaction rating are the forms the rules recognise. If you have already handed something over and the other person pushed it back, do not read that as thinking the amount was too small or as putting on airs, because they are required to refuse it. Keep two other things apart as well. Handing a doctor a red envelope is a matter of professional discipline and of the hospital. Handing property to the people handling or enforcing a case is the crime of bribery under the Criminal Law, and the two are not on the same scale; see item 40 in section 8 for that one. Hospitals widely accept a letter of thanks or a banner, but the claim that it 「会进个人和科室记录」, that is, goes into the individual's and the department's record, is based on experience and is not in any document. What has a document behind it is the rule bringing satisfaction survey results into the performance assessment of tertiary public hospitals. If you have doubts about the diagnosis or the treatment itself, ask on the spot for the records to be sealed (see item 7 in this section, sealing the records), and keep your own copy of the records and images (item 6, keeping your own copy of the records and images). \ No newline at end of file +- Sources: National Health Commission, National Healthcare Security Administration and National Administration of Traditional Chinese Medicine (国家卫生健康委、国家医保局、国家中医药局) (2021). Notice on Issuing the Nine Codes for Clean Practice by Staff of Medical Institutions (《关于印发医疗机构工作人员廉洁从业九项准则的通知》, 国卫医发〔2021〕41 号). Gazette of the State Council ；General Office of the State Council (国务院办公厅) (2019). Opinions on Strengthening the Performance Assessment of Tertiary Public Hospitals (《关于加强三级公立医院绩效考核工作的意见》, 国办发〔2019〕4 号). ；National Health Commission (国家卫生健康委) (2019). Measures for the Administration of Complaints in Medical Institutions (《医疗机构投诉管理办法》, Order No. 3). +- Notes: The people who benefit are you and the person who saved you. Wanting to say thank you after an all-night resuscitation is human. You can thank them; only the form has to change. A red envelope, tobacco and alcohol push the risk onto the other side, while a thank-you letter, a banner and a satisfaction rating are the forms the system recognises. Where you have already handed something over and they pushed it back, do not read it as them being short of money or playing hard to get: they have to do that under the rules. And keep two things apart. A red envelope to a doctor is a matter of professional discipline and of the hospital, while handing property to the officials handling or enforcing a case is bribery under the Criminal Law. The two are not on the same scale, and the second is item 40 in section 8. Hospitals widely accept thank-you letters and banners, but the claim that they "go into the personal and department record" rests on experience and no document requires it. What is set down in a document is the rule bringing the satisfaction survey results into the performance assessment of tertiary public hospitals. Where you have doubts about the diagnosis or treatment itself, ask on the spot for sealing the records (see item 6 in this section, sealing the records), and keep your own copy of the records and images (item 5, keeping your own copy of the records and images) diff --git "a/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" "b/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" index b3f137f9..d80b1645 100644 --- "a/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" +++ "b/en/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" @@ -14,7 +14,6 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (1958). Regulations on the Registration of Households, 中华人民共和国户口登记条例（第八条）. (posted on the Guangdong Provincial Government portal); State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，自 2026 年 3 月 30 日起施行，第二十三条）. - Notes: Someone bedridden for a long time who dies at home counts as an expected death. You can ask the institution that treated him while he was alive to issue the certificate, or the community health centre or the township health centre for the area you live in. Whoever treated him issues the death certificate, see item 2 in this section (the death certificate). If you cannot tell which kind it is, treat it as unexpected and call 110 first; once the police have been there and finished their inspection, they will still issue the certificate. If you tidy the scene up before calling, nobody can help you afterwards. Do not move the body yourself, do not wash and dress the body, and do not take away the medicine bottles or the notes. - ### 2. The death certificate is the key to everything that follows: whoever treated the person issues it, for an expected death at home you go to the community health service institution, and it is issued within one day - Cost: No money. @@ -23,7 +22,6 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Evidence grade: A - Sources: Office of the National Health and Family Planning Commission (国家卫生计生委办公厅) (2014). Interim Standards for the Registration of Death Information, 人口死亡信息登记管理规范（试行）（国卫办规划发〔2014〕68 号，第九条）. (forwarded by the Shenzhen Municipal Civil Affairs Bureau); National Health Commission (国家卫生健康委), Ministry of Public Security (公安部), Ministry of Civil Affairs (民政部), National Administration of Traditional Chinese Medicine (国家中医药局), National Administration of Disease Control and Prevention (国家疾控局) (2026). Notice on Strengthening the Registration of Medical Death Certificates of Residents and the Management of Electronic Licences, 关于加强居民死亡医学证明信息登记和电子证照管理工作的通知（国卫规划发〔2026〕5 号）. (forwarded by the Fujian Provincial Health Commission) - Notes: The cancellation of the household registration, the cremation, the withdrawal of the housing provident fund, the payments under social insurance, the insurance claim and the inheritance all need it. Get it and make several copies at once, and ask for the electronic licence as well where one can be issued. You only get one chance of a replacement, so do not lose the original. If the name or the identity number is wrong, have it corrected on the spot, because finding it out when the inheritance is being processed means going round every counter again. - ### 3. Collecting, storing and cremating the body can only be done by a funeral home, a hospital mortuary does not hold funerals, and storage is normally no more than 3 days - Cost: No money. @@ -32,7 +30,6 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Evidence grade: A - Sources: State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，自 2026 年 3 月 30 日起施行，第八、九、二十四、二十五、二十七、二十八、二十九条）. - Notes: A 「one-stop service」 that turns up after someone dies in hospital is usually not from the funeral home. The collection itself is a basic item and it has a published price, see item 6 in this section (the list of basic items). Keep the cremation certificate safe; you need it for the inheritance, the insurance and moving the ashes. Whether the area you live in applies cremation or burial is fixed by the provincial government. The funeral customs of ethnic minorities are protected by Article 9 of the regulations. Where a body that should be cremated is buried instead, or is buried and a grave built outside a cemetery or an eco-burial area, the civil affairs department orders it corrected within a set period (that is, put it back the way the rule says). - ### 4. If you dispute the cause of death, ask for the autopsy within 48 hours and have the medical records sealed at the same time - Cost: No money. The hard part is remembering to ask at that moment. @@ -41,7 +38,6 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Evidence grade: A - Sources: State Council (国务院) (2018). Regulations on the Prevention and Handling of Medical Disputes, 医疗纠纷预防和处理条例（国务院令第 701 号，第二十四、二十六条）. - Notes: Saying afterwards that something felt wrong from the start puts the responsibility on the party that delayed. Once the body is cremated there is no autopsy to be had, so this has to be settled before you get in touch with the funeral home. How the records are sealed is in section 24. - ### 5. Cancelling the household registration has a statutory deadline: in a city before the burial, in the countryside within one month - Cost: No money. Take the death certificate and the household register and go to the police station once. @@ -50,7 +46,6 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (1958). Regulations on the Registration of Households, 中华人民共和国户口登记条例（第八条）. (posted on the Guangdong Provincial Government portal) - Notes: While the registration is still open, the transfer of inherited property, the payments you are entitled to and the repayment of the pension all stick, because every one of those counters looks at the registration status. If the pension keeps being paid, that is someone receiving it wrongly and it will have to be handed back eventually. While you are at the police station, ask as well: how the social insurance and the medical insurance payments are stopped, and how the funeral subsidy is claimed. Those payments are in item 9 in this section (the provident fund balance and the payments under social insurance). - ### 6. Funeral services are split into basic items and non-basic items, there is a list of the basic items, and the charges are set by law - Cost: No money. Spend a few minutes looking at the list before you set off. @@ -59,7 +54,6 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Evidence grade: A - Sources: State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，2025 年 11 月 14 日国务院第 72 次常务会议修订通过，「自2026年3月30日起施行」）. 第五条：「殡葬服务分为基础项目和非基础项目，收费政策由国务院发展改革部门、财政部门会同国务院民政等部门制定。」「国家制定殡葬服务基础项目清单，将遗体接运、遗体存放、遗体告别、遗体火化、骨灰寄存、生态安葬以及政府举办的殡葬服务机构提供的骨灰格位安葬等纳入清单范围」；第五十条「殡葬服务实行清单化管理并动态调整，禁止在清单之外设立项目、收取费用。」 - Notes: A province, an autonomous region or a municipality may add items on top of the national list. Ask first which items count as basic items and what the local charge is, and only then talk about the rest. The price of an item on the list is fixed by the rules; the items off the list are yours to choose. Tell the two kinds apart and only then do you have anything to say about the part you have been overcharged. There is nothing wrong with giving someone a decent send-off, and no one is telling you to make do cheaply; where the money goes is your own decision. - ### 7. If you are sold a bundle you did not ask for, charged above the posted price, or charged several times for one item split into several, write it down and complain - Cost: No money. The hard part is still standing your ground at that moment. @@ -68,6 +62,7 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Evidence grade: A - Sources: State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，2025 年 11 月 14 日国务院第 72 次常务会议修订通过，「自2026年3月30日起施行」）. 第四十二条：「殡葬服务机构不得违反有关规定擅自设立收费项目，或者采取分解项目、扩大范围收费等形式变相提高收费标准。」第四十三条列举的价格违法行为含「（一）不按规定明码标价，或者在标价之外加价或者收取未标明的费用」「（三）采取捆绑、附加不合理条件或者不合理限制等方式，强制或者变相强制销售丧葬用品、提供殡葬相关服务并收取费用」「（四）强制或者变相强制丧属接受第三方的有偿服务，或者利用相关服务网点重复收取费用」；第五十一条要求民政部门与政务服务便民热线建立信息共享监督机制，「对丧属反映的殡葬服务问题，应当会同有关部门及时予以处理并将处理结果告知丧属」 - Notes: Price violations are dealt with by the market regulation authorities, and the funeral service itself by the civil affairs department. Ask for a price list on the spot, keep the receipts and the messages, and complaining afterwards is still in time. + ### 8. Funeral agents, agents who buy the goods, planners and hosts of the ceremony: intermediaries like these must file with the county civil affairs department - Cost: No money. One question is enough: has it filed? @@ -76,7 +71,6 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Evidence grade: A - Sources: State Council (国务院) (2026). Regulations on Funeral and Interment Administration, 殡葬管理条例（国务院令第 824 号，2025 年 11 月 14 日国务院第 72 次常务会议修订通过，「自2026年3月30日起施行」）. 第四十一条：「除殡葬服务机构外，从事殡葬服务代理、用品代购、策划主持、信息咨询等殡葬相关服务活动的组织和个人，应当向县级人民政府民政部门备案。」「对存在以欺骗等手段违规收费、诱导大操大办、强迫接受服务、倒卖逝者信息等损害丧属权益行为的依法处置，并在殡葬服务信息系统中进行标注，向社会公开。」 - Notes: Selling on the information of the dead is the commonest trap in this trade. If a pile of sales calls arrives the moment the paperwork is done, the information has almost certainly been sold on. Online memorial platforms are covered by the same rule. Article 45 of the regulations forbids them from pushing consumption, and forbids them from handling the information of the dead in breach of the law. - ### 9. The money sitting in different places has to be collected one at a time: the provident fund balance, the payments under social insurance, the payments for a work injury - Cost: No money, but you have to visit several counters. Before you go, prepare the death certificate and the proof of the relationship. @@ -85,7 +79,6 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Evidence grade: A - Sources: State Council (国务院) (2026). Regulations on the Housing Provident Fund, 住房公积金管理条例（国务院令第 844 号修改，2026 年 9 月 20 日起施行）第二十四条末款. :「职工死亡或者被宣告死亡的，职工的继承人、受遗赠人可以提取职工住房公积金账户内的存储余额；无继承人也无受遗赠人的，职工住房公积金账户内的存储余额纳入住房公积金的增值收益。」；the three payments under the Regulations on Work Injury Insurance, see section 19. ；Standing Committee of the National People's Congress (全国人大常委会) (2018 修正). Social Insurance Law, 社会保险法（第十四、十七条）：「个人死亡的，个人账户余额可以继承」「参加基本养老保险的个人，因病或者非因工死亡的，其遗属可以领取丧葬补助金和抚恤金」. (National Database of Laws and Regulations)；Ministry of Human Resources and Social Security (人力资源社会保障部) and Ministry of Finance (财政部) (2021). Interim Measures on the Survivor Benefits of Enterprise Employees under the Basic Old-Age Insurance for Employees, 企业职工基本养老保险遗属待遇暂行办法（人社部发〔2021〕18 号，2021 年 9 月 1 日起施行，第四、五条），the original is not posted online, and the articles were checked word for word against the quotation in the administrative reconsideration decision of the People's Government of Huadu District, Guangzhou (花都府行复〔2022〕226 号):「丧葬补助金的标准，按照参保人员死亡时本省……上一年度城镇居民月人均可支配收入的2倍计算」「缴费年限不满5年的，发放月数为3个月……缴费年限30年以上的，按照30年计算，发放月数最高为24个月」. ；Fujian Provincial Department of Human Resources and Social Security (福建省人力资源和社会保障厅). Frequently asked questions:「每领取1年基本养老金减少1个月，发放月数最低为9个月」「累计缴费年限不足5年的，其遗属待遇（含丧葬补助金和抚恤金）标准不得超过其个人缴费之和」. - Notes: The provident fund is the one most often forgotten, because it is not in a bank account and it does not flow automatically into the distribution of the estate. What is left in the personal account under the old-age insurance and the survivor benefits are worked out and paid separately, so you go to the social insurance counter of the place where the person was insured and deal with both at once. What each place asks for is different. Where the contributions come to under 5 years, the survivor benefits together come to no more than what the person paid in. This formula comes from the rules for the old-age insurance of enterprise employees; for the survivor benefits under the insurance for public institutions and for urban and rural residents, ask at the counter. The formal requirements for a will and for an inheritance are in section 17. - ### 10. The deceased's accounts and personal information: close relatives may ask to look at them, take a copy, have them corrected, or have them deleted - Cost: No money. You only need to report to the authorities or sue if the platform refuses to cooperate. @@ -93,4 +86,4 @@ The funeral side follows the new Regulations on Funeral and Interment Administra - Benefit: The law says plainly: where a natural person dies, close relatives, for their own lawful and proper interests, may exercise the rights set out in that chapter to look at, take a copy of, correct and delete the deceased's relevant personal information. Close relatives may ask the platform to let them see the information, take a copy, put right the errors and delete it - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Personal Information Protection Law, 个人信息保护法. 中国人大网. 第四十九条：「自然人死亡的，其近亲属为了自身的合法、正当利益，可以对死者的相关个人信息行使本章规定的查阅、复制、更正、删除等权利；死者生前另有安排的除外。」；第五十条「个人信息处理者应当建立便捷的个人行使权利的申请受理和处理机制。拒绝个人行使权利的请求的，应当说明理由」「个人可以依法向人民法院提起诉讼」 -- Notes: The article has one clause, 「死者生前另有安排的除外」 — where the person made arrangements during his lifetime, either by naming an account successor or by leaving a clear statement of wishes, you follow what he left behind. So while the person is still lucid it is worth writing down where the accounts and the passwords are to go, see section 17. Cancelling an account and deleting personal information are two different things, see section 14. \ No newline at end of file +- Notes: The article has one clause, 「死者生前另有安排的除外」 — where the person made arrangements during his lifetime, either by naming an account successor or by leaving a clear statement of wishes, you follow what he left behind. So while the person is still lucid it is worth writing down where the accounts and the passwords are to go, see section 17. Cancelling an account and deleting personal information are two different things, see section 14. diff --git "a/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" "b/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" index ec9ff15a..228502a2 100644 --- "a/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" +++ "b/en/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" @@ -12,7 +12,6 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: State Council (国务院) (2023). Regulations on the Supervision and Administration of Non-Bank Payment Institutions, 非银行支付机构监督管理条例（国务院令第 768 号，第二、六、八、二十七、四十七条）. ; People's Bank of China (中国人民银行) (2024). Detailed Rules for the Implementation of the Regulations on the Supervision and Administration of Non-Bank Payment Institutions, 非银行支付机构监督管理条例实施细则（中国人民银行令〔2024〕第 4 号，第六十八条）. ; National People's Congress (全国人大) (1997). Criminal Law, 刑法（第二百二十五条第三项）. ; Supreme People's Court (最高人民法院) and Supreme People's Procuratorate (最高人民检察院) (2019). Interpretation on Several Issues Concerning the Application of Law in Handling Criminal Cases of Illegal Fund Payment and Settlement Services and Illegal Foreign Exchange Trading, 关于办理非法从事资金支付结算业务、非法买卖外汇刑事案件适用法律若干问题的解释（第一、三、四条）. - Notes: Designing the money so that it never comes into your account is the only safe way. The trade calls this kind of collection-and-payment-on-behalf 「secondary clearing」, but the word is not in any official document. The detailed rules use the wording 「变相从事支付业务」 — taking on payment business in a disguised way, that is, moving money for others on the user's instruction without the approval. The interpretation of the Supreme People's Court (最高人民法院) and the Supreme People's Procuratorate (最高人民检察院) lists only three situations: cashing out through fabricated transactions, cashing out through a company account, and cashing out through cheques. It does not name a platform collecting and paying on behalf of others, so one cannot say that a platform doing that 「certainly amounts to a crime」 - ### 2. A website that charges money needs a licence, matching transactions is a different licence, and a free site only needs a filing - Cost: The filing costs nothing. The Value-Added Telecommunications Business Licence (增值电信业务经营许可证) costs money, and it decides 「向用户收钱的网站能不能开」 — whether a website that charges its users is allowed to open. To get that licence you need a company first. A business operating only inside one province needs registered capital of at least 1 million yuan. A business operating across provinces needs at least 10 million yuan. The review takes 60 days, and the licence is valid for 5 years @@ -21,7 +20,6 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: State Council (国务院) (2000). Measures for the Administration of Internet Information Services, 互联网信息服务管理办法（国务院令第 292 号，2024 年 12 月 6 日第二次修订，第三、四、七、十九条）. ; State Council (国务院) (2024). Decision on Amending and Repealing Certain Administrative Regulations, 关于修改和废止部分行政法规的决定（国务院令第 797 号，第十八项）. ; State Council (国务院) (2000). Regulations on Telecommunications, 电信条例（2016 年第二次修订，第七、九、十三、六十九条）. ; Ministry of Industry and Information Technology (工业和信息化部) (2017). Measures for the Administration of Licences for the Conduct of Telecommunications Business, 电信业务经营许可管理办法（工信部令第 42 号，第六、九、十一、十四条）. - Notes: Whether what you are doing is in the EDI class is for you to work out by matching the definition in the Catalogue of Telecommunications Business (《电信业务分类目录》). No official document says in so many words that an e-commerce platform with matching and fund handling must hold an EDI licence. The service guide of the Ministry of Industry and Information Technology (工业和信息化部) only says 「按照业务界定申请相应的电信业务经营许可」 — apply for the telecommunications business licence that matches how the business is defined. So whether you need one has to be judged from the definition yourself. One more thing: an individual cannot get this licence; the applicant has to be a company established in accordance with the law - ### 3. Streaming needs the Internet Culture Business Licence, and the licence for audio-visual programmes is basically out of reach for a private company - Cost: The Internet Culture Business Licence (网络文化经营许可证), called the web culture licence in short, decides 「能不能靠直播、游戏解说这类网络表演赚钱」 — whether you may earn money from online performances such as livestreaming or game commentary. It is approved by the provincial cultural administrative department, which must give a result within 20 days of accepting the application, and it is valid for 3 years, with the renewal filed 30 days before it expires. The Audio-Visual Programme Licence (视听节目许可) decides 「能不能在网上放节目」 — whether you may put programmes online, and it requires the applicant to be a wholly state-owned or a state-controlled entity @@ -30,7 +28,6 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: Ministry of Culture (文化部) (2016). Measures for the Administration of the Business of Online Performance, 网络表演经营活动管理办法（文市发〔2016〕33 号，第二、四、六、十三、二十条）. ; Ministry of Culture (文化部) (2011). Interim Provisions on the Administration of Internet Culture, 互联网文化管理暂行规定（文化部令第 51 号，2017 年经文化部令第 57 号修订，第七、八、九、十六、二十一、二十八条）. ; National Radio and Television Administration (广电总局) and Ministry of Information Industry (信息产业部) (2007). Provisions on the Administration of Internet Audio-Visual Programme Services, 互联网视听节目服务管理规定（第 56 号令，2015 年修订，第二、七、八、二十四条）. ; Cyberspace Administration of China (国家网信办) and six other departments (2021). Guiding Opinions on Strengthening the Standardised Management of Online Livestreaming, 关于加强网络直播规范管理工作的指导意见（国信办发文〔2021〕3 号，第 9 项）. - Notes: Which business goes with which licence is set out most clearly in the document of the seven departments. A platform that earns money from livestreaming needs the web culture licence plus the ICP filing, and the ICP filing is the registration the website does with the Ministry of Industry and Information Technology (工业和信息化部). Putting programmes online needs either the audio-visual programme licence or the completed registration. Publishing news needs the Internet News Information Service Licence. One more thing: the Interim Provisions on the Administration of Internet Culture (互联网文化管理暂行规定) reprinted on the government website in 2025 are the unrevised 2011 text, so quote the version currently in force from the Ministry of Culture and Tourism (文化和旅游部) - ### 4. A server inside China needs a filing, and the access provider must itself hold the value-added telecommunications licence - Cost: The filing costs nothing. It is submitted together with the server at the provider, and it usually takes a few days to a few weeks. Ten minutes is enough to check once whether a provider is qualified @@ -39,6 +36,7 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: Ministry of Information Industry (信息产业部) (2005). Measures for the Filing of Non-Commercial Internet Information Services, 非经营性互联网信息服务备案管理办法（信产部令第 33 号，2024 年经工信部令第 68 号修订，第五、十三、十八、二十二条）. ; Ministry of Industry and Information Technology (工业和信息化部) (2017). Measures for the Administration of Licences for the Conduct of Telecommunications Business, 电信业务经营许可管理办法（工信部令第 42 号，第二十、二十四条）. ; Ministry of Industry and Information Technology (工业和信息化部) (2015). Catalogue of Telecommunications Business (2015 edition), 电信业务分类目录（2015 年版，B11、B14、B21、B25）. - Notes: Check whether the provider holds a licence before you talk about the price. To check a licence, use the MIIT integrated telecommunications business market management information system, tsm.miit.gov.cn, and search by the full name of the company or by the licence number. To check a website filing, use beian.miit.gov.cn. A 「data centre」 without the matching licence leaves you not knowing even who to appeal to on the day it fails. Putting the machine outside China does get round the filing. But where the users are in China and the money is in China, not one of the obligations written in items 5 to 10 in this section goes away, and you add the cost of sending data out of the country. + ### 5. Let users sell on the platform and you have to verify, register, report the information and keep it for three years - Cost: No money. You have to verify the sellers' qualifications in the back office and build a registration file. Report the tax information once a quarter and keep the transaction data for three years. The hard part is that this goes on all year; it is not something you do once @@ -47,7 +45,6 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). E-Commerce Law, 电子商务法（第十、二十七、二十八、三十一、八十条）. ; State Administration for Market Regulation (市场监管总局) (2021). Measures for the Supervision of Online Trading, 网络交易监督管理办法（市场监管总局令第 37 号，第八、二十四、二十五、三十一条）. ; State Council (国务院) (2025). Provisions on the Reporting of Tax-Related Information by Internet Platform Enterprises, 互联网平台企业涉税信息报送规定（国务院令第 810 号，第二、三、四、十条）. ; State Taxation Administration (国家税务总局) (2025). Announcement on Matters Concerning the Reporting of Tax-Related Information by Internet Platform Enterprises, 关于互联网平台企业报送涉税信息有关事项的公告（税务总局公告 2025 年第 15 号）. - Notes: The seller has to register as a market entity as well. The law lists a few cases where you do not: 「个人销售自产农副产品、家庭手工业产品，个人利用自己的技能从事依法无须取得许可的便民劳务活动和零星小额交易活动」 — a person selling their own agricultural and sideline products or home handicrafts, and a person using their own skills in convenience service activities that need no licence under the law and in occasional small-value transactions. 「零星小额」 — occasional and small in value — has a fixed figure: Article 8 of the Measures for the Supervision of Online Trading sets it at an annual turnover of 100,000 yuan. On the platform side there are two more things to do: mark the registered sellers and the unregistered ones separately, and do not allow the ratings left by consumers to be deleted - ### 6. You are answerable for what users post: a review system, a reporting channel, and stop the transmission and report at once when you find something unlawful - Cost: Money is needed to pay people to do the reviewing, and it is the platform's largest ongoing expense. The reporting channel and a record of every action taken both have to be kept. The hard part is that somebody has to watch it every day; one day off and something can go wrong @@ -56,7 +53,6 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: Cyberspace Administration of China (国家互联网信息办公室) (2019). Provisions on the Governance of the Ecology of Network Information Content, 网络信息内容生态治理规定（网信办令第 5 号，第八、九、十、十六、三十七条）. ; Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第四十九、六十九条）. ; State Council (国务院) (2000). Measures for the Administration of Internet Information Services, 互联网信息服务管理办法（第十五、十六、二十三条）. - Notes: You do not have to think of the review rules yourself; copy them. The nine kinds of content that may not be posted are set out clearly in Article 15 of the Measures for the Administration of Internet Information Services. If you livestream there are three more things: you have to 「建立直播内容审核平台」 — set up a platform for reviewing the content of the stream, you have to be able to cut off a livestream that is running at any moment, which the article calls 「即时阻断」 — instant blocking, and you have to keep the content of the stream and the logs for 60 days. The video material of an online performance is likewise kept for no less than 60 days - ### 7. If you provide information posting or instant messaging, you must require real identity information from users - Cost: Money is needed to connect a real-name verification channel, and these channels charge per check. Design the product on the principle of a real name in the back office and a free choice in the front: store the real name in the back office, show a nickname on the page @@ -65,7 +61,6 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2025). Cybersecurity Law, 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十六、六十四条；2016 年文本为第二十四、六十一条）. ; Cyberspace Administration of China (国家互联网信息办公室) (2016). Provisions on the Administration of Internet Livestreaming Services, 互联网直播服务管理规定（第十二、十六条）. ; Cyberspace Administration of China (国家网信办) and six other departments (2021). Interim Measures for the Administration of Online Livestreaming Marketing, 网络直播营销管理办法（试行）（国信办发文〔2021〕5 号，第八条）. - Notes: The article numbers have moved. The Cybersecurity Law was amended in 2025 and comes into force on 2026-1-1. The clause on real names moved from Article 24 to Article 26, and the penalties moved from Article 61 to Article 64. The wording did not change. If the material in your hands still cites the old numbers, they have to be corrected - ### 8. Do not open livestreaming accounts for people under 16, and handle tipping money in bands by age - Cost: No money. Add one step at each of the two points, registration and payment: first check the age, then confirm that the guardian agreed @@ -74,6 +69,7 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2020). Law on the Protection of Minors, 未成年人保护法（第七十四、七十六条）. ; State Council (国务院) (2023). Regulations on the Protection of Minors Online, 未成年人网络保护条例（国务院令第 766 号，第三十一、四十四条）. ; General Office of the Cyberspace Administration of China (中央网信办秘书局) (2026). Notice on Strengthening the Standardised Management of Online Livestreaming Tipping, 关于加强网络直播打赏规范管理的通知（第 7 项）. ; Supreme People's Court (最高人民法院) (2020). Guiding Opinions on the Proper Handling of Civil Cases Related to the Novel Coronavirus Epidemic under the Law, 关于依法妥善审理涉新冠肺炎疫情民事案件若干问题的指导意见（二）（法发〔2020〕17 号，第九条）. - Notes: If a refund really is disputed, handle it in the direction that is best for the minor. The wording on this has changed. The opinions of the four departments in 2022 and the notice of the National Radio and Television Administration (广电总局) in 2020 both said that minors may not tip at all. The new notice of April 2026 changed it to bands at 8 and 16, but did not say that the earlier two documents are repealed, and the authorities have not said how the two fit together. So the sentence 「no minor may ever tip」 is no longer accurate. + ### 9. Handle an infringement notice in time, and if nothing follows within 15 days of passing on the counter-notice, put the content back - Cost: No money. You need a fixed process: receive the notice, pass it to the seller, take the seller's counter-notice, and publish the result. Keep a record at every step @@ -82,7 +78,6 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2018). E-Commerce Law, 电子商务法（第四十二、四十三、四十四、四十五、八十四条）. ; National People's Congress (全国人大) (2020). Civil Code, 民法典（第一千一百九十五、一千一百九十六、一千一百九十七条）. - Notes: If you are not an e-commerce platform but an ordinary website, the notice-and-takedown rules of the Civil Code apply. The difference is how long you have to wait after the other side counters: the E-Commerce Law gives a clear 15 days, while the Civil Code only says 「合理期限」 — a reasonable period. Publishing the notice you received, the seller's counter-notice and the result of your handling is an express requirement of Article 44 of the E-Commerce Law - ### 10. Do not put users' information outside the country in passing; transfers out of China have statutory conditions and a headcount threshold - Cost: No money. The least trouble is to keep the user data inside China. Where it really has to go out, use the standard contract or get certified @@ -91,7 +86,6 @@ You want to build a website or an app. This section answers four questions: what - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2021). Personal Information Protection Law, 个人信息保护法（第三十八、三十九、四十条）. ; Cyberspace Administration of China (国家互联网信息办公室) (2024). Provisions on Promoting and Standardising the Cross-Border Flow of Data, 促进和规范数据跨境流动规定（网信办令第 16 号，第三、四、五、七、八条）. ; Cyberspace Administration of China (国家互联网信息办公室) (2022). Measures for the Security Assessment of Data Transfers Out of China, 数据出境安全评估办法（网信办令第 11 号，第四条）. - Notes: Having the server outside China is not the same as 「no data has gone out」. A user registers inside China and you send their information to a machine outside China, and that is a transfer out. There is one class that does not have to take the three routes of the security assessment, the certification and the standard contract: cross-border shopping, cross-border delivery and booking a flight or a hotel, because these are 「为订立、履行个人作为一方当事人的合同」 — transfers genuinely necessary to conclude or perform a contract to which the individual is a party. One more thing to watch is how the number of people is counted: it is counted 「自当年 1 月 1 日起累计」 — from 1 January of that year and accumulated, not over the twelve months counting back from today - ### 11. Choosing a server: first ask whether you can put up with the downtime, then whether anybody maintains it, and compare prices last - Cost: Cloud servers from the main cloud providers run from several hundred to several thousand yuan a month. A VPS from a small provider starts at a few tens of yuan a month. A VPS is a small virtual server. Colocation for a dedicated server starts at several thousand yuan a year, and the hardware you have to buy yourself @@ -103,4 +97,4 @@ You want to build a website or an app. This section answers four questions: what ## Licence -The text is published under [CC BY 4.0](../LICENSE). If you reprint or adapt it, say where it came from and link the original. The code is under [MIT](../LICENSE-CODE). \ No newline at end of file +The text is published under [CC BY 4.0](../LICENSE). If you reprint or adapt it, say where it came from and link the original. The code is under [MIT](../LICENSE-CODE). diff --git "a/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" "b/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" index 7e5dff49..734132f0 100644 --- "a/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" +++ "b/en/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" @@ -14,7 +14,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: De-Regil LM, Peña-Rosas JP, Fernández-Gaxiola AC, Rayco-Solon P (2015). Effects and safety of periconceptional oral folate supplementation for preventing birth defects. Cochrane Database of Systematic Reviews, (12), CD007950. - Notes: What matters is starting before conception. The fetal neural tube closes within 28 days of fertilisation. By the time a pregnancy test turns positive, most of that window has gone. So if you plan to have a child, start 3 months beforehand. The same pooled review found no clear effect of folic acid on cleft lip and palate, congenital heart disease or miscarriage. Do not expect it to treat anything else - ### 2. Before 13 weeks of pregnancy open a Mother and Child Health Handbook at the community health centre, and use up your free prenatal checks - Cost: No money. It is part of the national basic public health service programme. You have to make several trips according to how far along you are @@ -23,7 +22,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: National Health and Family Planning Commission (国家卫生计生委) (2017). National Standard for Basic Public Health Services, Third Edition（国家基本公共卫生服务规范（第三版））. ; Jingjiang Municipal People's Government (靖江市人民政府). National Basic Public Health Service Programme — Maternal and Child Health Management Services (local published service list)（国家基本公共卫生服务项目——孕产妇健康管理服务（地方服务清单公开））. - Notes: Those 5 checks are the free floor, not the whole of prenatal care. The hospital will add ultrasound, blood glucose and fetal heart monitoring where it sees fit, and you pay for those. The earlier you open the handbook, the earlier the pregnancy risk grading is worked out. Pregnancy risk grading puts pregnant women into tiers by risk level. Unless a high-risk tier is found, you cannot be referred in time to a hospital that can take the case. If you live in a district other than your registered one, open the handbook where you live now. There is no need to go back to your registered address - ### 3. Test for HIV, syphilis and hepatitis B at the first prenatal check, and free blocking treatment is available if any is positive - Cost: No money. The programme for preventing mother-to-child transmission provides the tests free @@ -32,7 +30,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: General Office of the National Health Commission (国家卫生健康委办公厅) (2020). Working Specification on the Prevention of Mother-to-Child Transmission of HIV, Syphilis and Hepatitis B (2020 Edition)《预防艾滋病、梅毒和乙肝母婴传播工作规范（2020 年版）》. (reprinted by the Xinjiang Uygur Autonomous Region Health Commission (新疆维吾尔自治区卫生健康委员会)) ; State Council (国务院) (2006). Regulations on the Prevention and Control of AIDS (Articles forty-three and forty-four)《艾滋病防治条例》. (reprinted by the Beijing Municipal Health Commission (北京市卫生健康委员会)) - Notes: The earlier these three are tested, the better, because blocking and treatment both take time. Article three of the same regulations also states that 「任何单位和个人不得歧视艾滋病病毒感染者、艾滋病病人及其家属」. The lawful rights to marry, to find work, to see a doctor and to go to school are protected by law. For everyday prevention and testing see section 1. For post-exposure blocking after high-risk behaviour see section 13 - ### 4. Not one cigarette and not one drink during pregnancy, and nobody in the household smokes indoors - Cost: No money, and you also save the money you would spend on tobacco and alcohol. The hard part is getting everyone you live with to stop as well @@ -41,7 +38,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: Centers for Disease Control and Prevention. Smoking During Pregnancy. - Notes: Smoke from someone else indoors counts for the fetus, because it is the pregnant woman who breathes it in. So family members should not smoke indoors either. Alcohol is the same: no amount of drinking is known to be safe, and the simplest rule during pregnancy is to drink nothing at all. For what tobacco and alcohol cost an adult in their own right see section 2 - ### 5. If you have risk factors for pre-eclampsia, start one low-dose aspirin tablet a day after 12 weeks of pregnancy - Cost: Tens of yuan. One tablet a day, taken right through to the birth. The hard part is never forgetting @@ -50,7 +46,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: US Preventive Services Task Force (2021). Aspirin Use to Prevent Preeclampsia and Related Morbidity and Mortality: Preventive Medication. - Notes: The high-risk categories are: pre-eclampsia in a previous pregnancy, twins or more this time, and chronic high blood pressure. Also type 1 or type 2 diabetes before pregnancy, kidney disease, and autoimmune conditions such as systemic lupus erythematosus or antiphospholipid syndrome. If you are in none of these categories, do not take it on your own. A doctor has to prescribe this, so give your full medical history at the prenatal check and let the doctor judge. What is sold in China is mostly the 100 mg enteric-coated tablet. Take it as the doctor instructs - ### 6. Have one screening for gestational diabetes at 24 weeks of pregnancy or later, and do not begrudge the sugar drink - Cost: Tens to a little over a hundred yuan. You go fasting and give several blood samples. It takes half a morning @@ -59,7 +54,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: US Preventive Services Task Force (2021). Gestational Diabetes: Screening. - Notes: Among the women who test positive, most can keep their blood sugar under control by changing diet and exercise, and only a minority need insulin. Do not skip this one test because you are afraid of what a positive result would mean. After the birth you have to test your blood sugar again: women who have had gestational diabetes are clearly at higher risk of type 2 diabetes later - ### 7. Memorise this list of 「立刻去医院」 signs; it counts during pregnancy and for a year after the birth - Cost: No money. Five minutes to memorise it, and it is best if the people you live with know it too @@ -68,6 +62,7 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: Centers for Disease Control and Prevention. Hear Her — Urgent Maternal Warning Signs. ; World Health Organization. Maternal mortality fact sheet. - Notes: What this list is for is telling 「再忍忍」 apart from 「必须马上走」. A headache together with blurred vision and pain in the upper abdomen points to pre-eclampsia (the high blood pressure disorder of pregnancy). One lower leg swollen and painful points to deep vein thrombosis (see section 13). By the World Health Organization's account, severe bleeding after birth can take the life of a healthy woman within hours. Go straight to hospital as soon as any one item appears, and do not spend a night at home first + ### 8. If your waters break, lie down flat where you are, raise your hips and call 120; do not walk about and do not bathe - Cost: No money @@ -76,7 +71,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: B - Sources: Beijing Municipal Health Commission (北京市卫生健康委员会). Weekly first aid topic: what to do when a pregnant woman's waters break early（每周急救话题：准妈妈羊水早破怎样办？）. ; Beijing Shunyi District Maternal and Child Health Hospital (北京市顺义区妇幼保健院). What to do when the waters break at home in late pregnancy（孕晚期在家破水怎么办？）. - Notes: The three signs that labour is close are regular contractions, the bloody show, and the waters breaking. You can pack calmly for the first two. Breaking of the waters is the only one that means lying down on the spot and calling for a car. Bathing, squatting on the toilet and driving yourself after the waters break are all classic mistakes. Write down the time the waters broke and the colour of the fluid, and tell the doctor. Yellow or green fluid, or fluid with flakes of meconium in it, points to fetal distress (the fetus is short of oxygen inside the uterus). If regular contractions, a clear drop in the baby's movement, and fever with shivering are happening at the same time, stress to the doctor that this is an emergency - ### 9. If you want pain relief in labour, say so directly. It does not raise the risk of caesarean section - Cost: Several hundred to over a thousand yuan, and the price differs from place to place. You have to ask the hospital and the anaesthesia department in advance whether they can do it. The hard part is raising it yourself @@ -85,7 +79,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: Anim-Somuah M, Smyth RMD, Cyna AM, Cuthbert A (2018). Epidural versus non-epidural or no analgesia for pain management in labour. Cochrane Database of Systematic Reviews, (5), CD000331. ; National Healthcare Security Administration (国家医保局) (2024). Rebuilding the obstetric service price items to help build a childbirth-friendly society《重构产科服务价格项目 助力构建生育友好型社会》. - Notes: The two claims that spread widest are that pain relief in labour means a back problem for life, and that it makes the birth hard so an operation is needed. Neither is founded. Cochrane pooled epidural analgesia with the other methods and concluded that it has no effect on the risk of caesarean section and none on long-term back pain. The real side effects are these: low blood pressure, fever, weakness in the legs, difficulty passing urine, and a longer first and second stage of labour with more likelihood of needing oxytocin. Whether to have it is yours to decide, but do not give it up on the strength of a rumour. Whether it is available also depends on whether the hospital has an anaesthetist on site 24 hours a day. Ask when you go for the prenatal check - ### 10. Without a medical indication do not ask for a caesarean section, and do not have an operation to pick an auspicious date - Cost: No money. The hard part is standing your ground against what the family says @@ -94,7 +87,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: World Health Organization (2015). WHO Statement on Caesarean Section Rates（WHO/RHR/15.02）. - Notes: When there is a real medical need, a caesarean section is an operation that saves lives. Do it, and do not tough it out. What to avoid is an operation that is not medically needed, whether out of fear of pain, to pick an auspicious date, or to fit in with the age at which the child starts school. A caesarean section is a large abdominal operation, recovery is slower, and the next pregnancy carries higher risks of placenta previa, placenta accreta and uterine rupture. That bears directly on whether you still want a second child - ### 11. Check your maternity insurance status before conception; a spouse who is not working can also claim the maternity medical costs - Cost: No money. One call to the social insurance or medical insurance enquiry line @@ -103,7 +95,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (2010). Social Insurance Law of the People's Republic of China (Articles fifty-four and fifty-six)《中华人民共和国社会保险法（第五十四、五十六条）》. (reprinted by the Guangdong Provincial Tax Service, State Taxation Administration (国家税务总局广东省税务局)) - Notes: Two things have to be asked in advance, because neither can be put right afterwards. The first is whether the number of consecutive months of contributions is enough: every place sets a minimum, and below it you get nothing. The second is whether a birth outside the place of enrolment has to be filed for in advance: without the filing you may have to pay the money yourself first and claim it by hand afterwards, and sometimes you cannot claim it at all. For the number of days of maternity leave and how the maternity allowance is worked out see item 2 in section 18 - ### 12. Get the Medical Certificate of Birth done before you leave the hospital; decide the name in advance and do not get a character wrong - Cost: No money. It is done as part of the discharge formalities @@ -112,7 +103,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: Standing Committee of the National People's Congress (全国人大常委会) (1994, amended 2017). Law of the People's Republic of China on Maternal and Infant Health (Article twenty-three)《中华人民共和国母婴保健法（第二十三条）》. - Notes: Before filling it in, check the parents' names, their identity card numbers and the child's name character by character. The characters used in the name have to be ones the household registration system can produce, and rare characters make every later errand harder. Changing the name or reissuing the certificate means going through a supplementary or a replacement issue, which is far more trouble than checking it on the spot. Keep the certificate on its own once you have it, as carefully as the household register and the identity card - ### 13. Do not refuse either the newborn heel-prick screening or the hearing screening - Cost: Free in most regions, or tens of yuan. Some extended items are yours to pay for. The baby's heel is pricked once after birth @@ -121,7 +111,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: Ministry of Health (卫生部) (2009). Measures for the Administration of Screening for Neonatal Diseases (Ministry of Health Order No. 64, Articles three, ten and eleven)《新生儿疾病筛查管理办法（卫生部令第 64 号，第三、十、十一条）》. - Notes: Article eleven requires that before the screening the items, the conditions, the method, the sensitivity (how much it can find) and the cost are all explained truthfully to the guardian, who then signs consent. So you will be given an informed consent form. Read it before you sign: which conditions are the ones the state requires, and which are extra items you pay for. A positive first screen is not a diagnosis. Most are ruled out at the repeat test, so do not fall apart at this stage. If the first hearing screen is not passed, go for the repeat test when you are told to. Do not decide on your own that the child can hear well enough - ### 14. Enrol the child in resident medical insurance as soon as you have the Medical Certificate of Birth; do not wait for household registration - Cost: Resident medical insurance is paid once a year, a few hundred yuan @@ -130,7 +119,6 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: State Council official website (中国政府网) (2024). Regular policy briefing on a sound long-term mechanism for basic medical insurance enrolment《健全基本医疗保险参保长效机制国务院政策例行吹风会》. - Notes: Preterm birth, jaundice needing phototherapy and newborn pneumonia are the common reasons for a newborn to be admitted, and they run into thousands of yuan. Being enrolled or not makes a large difference. The rules on how many days after birth you must enrol in order to get cover from the day of birth itself are not the same everywhere. For how it is actually applied, follow what the local medical insurance authority says. One call to 12393 before you leave the hospital settles it - ### 15. Go to the police station to register the birth within one month of the child being born - Cost: No money. One trip to the household registration window at the police station @@ -139,6 +127,7 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Evidence grade: A - Sources: State Council (国务院) (1958). Regulations of the People's Republic of China on Household Registration (Article seven)《中华人民共和国户口登记条例（第七条）》. (reprinted by the Guangdong Provincial People's Government (广东省人民政府)) - Notes: Bring all of these: the Medical Certificate of Birth, both parents' identity cards and household registers, and the marriage certificate. Whether to register with the father or with the mother is yours to choose, but the rules of the two places differ and both the school catchment and the place of insurance are affected. Decide before you go; changing it later is a nuisance. A child born outside marriage can be registered in the same way + ### 16. Do not skip the 42-day postpartum check; it is also the screening for postpartum depression - Cost: No money, it is part of the basic public health service. It takes half a day. The hard part is that you have just left the maternity ward and cannot be bothered @@ -146,4 +135,4 @@ The numbers the World Health Organization gives: in 2023 about 260,000 women wor - Benefit: The National Standard for Basic Public Health Services (Third Edition)《国家基本公共卫生服务规范（第三版）》sets out maternal and child management, with 1 postpartum visit and 1 health check at 42 days postpartum. Locally published lists of services show that this check includes lab tests (a full blood count and a routine vaginal swab) and 1 screening for postpartum depression. Serious complications can occur at any time within a year after the birth. In the list 「立刻去医院」 in item 7, 「有伤害自己或伤害孩子的念头」, 「产后阴道出血或异常分泌物」 and 「一侧肢体严重肿痛」 are all postnatal - Evidence grade: A - Sources: National Health and Family Planning Commission (国家卫生计生委) (2017). National Standard for Basic Public Health Services, Third Edition（国家基本公共卫生服务规范（第三版））. ; Jingjiang Municipal People's Government (靖江市人民政府). National Basic Public Health Service Programme — Maternal and Child Health Management Services (local published service list)（国家基本公共卫生服务项目——孕产妇健康管理服务（地方服务清单公开））. -- Notes: Postpartum depression is not fussiness, and a word of encouragement from the family does not fix it. It has its own screening questionnaire, it can be treated, and it can be recovered from. For what to do if you have thoughts of suicide see section 1. For the lawful way out when a baby really cannot be brought up see item 20 in section 9. This check also looks at how the wound has healed, how the uterus has recovered and how the pelvic floor is working. Contraception is discussed at this visit too: no periods after the birth does not mean you cannot get pregnant \ No newline at end of file +- Notes: Postpartum depression is not fussiness, and a word of encouragement from the family does not fix it. It has its own screening questionnaire, it can be treated, and it can be recovered from. For what to do if you have thoughts of suicide see section 1. For the lawful way out when a baby really cannot be brought up see item 20 in section 9. This check also looks at how the wound has healed, how the uterus has recovered and how the pelvic floor is working. Contraception is discussed at this visit too: no periods after the birth does not mean you cannot get pregnant diff --git "a/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" "b/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" index 86af9faa..13d015f3 100644 --- "a/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" +++ "b/en/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" @@ -12,8 +12,8 @@ This section does not judge what anyone wants to look like; that is each person' - In plain language: People with anorexia nervosa die at about 5.9 times the rate of others their age. Follow 1000 people for one year and 5.1 of them die. One in five of those deaths is suicide. The way in is dieting: students at 44 secondary schools were followed for three years, and girls who dieted severely were 18 times as likely to develop an eating disorder as girls who did not. If you want to control your weight, move more, and do not eat so little that you are hungry. - Benefit: 36 studies pooled. The anorexia nervosa group was followed for 166,642 person-years in total (the years each person was followed, added up). Their standardised mortality ratio is 5.86, that is, their probability of dying is about 5.9 times that of others their age. Deaths are 5.1 per 1000 person-years, that is, follow 1000 people for one year and 5.1 of them die. About 1/5 of the people who die, die by suicide. The standardised mortality ratio for bulimia is 1.93, and for other eating disorders 1.92, each about 1.9 times that of others their age. Another study was done in 44 secondary schools in Victoria, Australia: students aged 14–15 followed for 3 years, and the same group surveyed in 6 waves over the three years. New eating disorders among girls were 21.8 per 1000 person-years, that is, follow 1000 girls for one year and 21.8 new cases appear. Girls who dieted severely had 18 times the risk of those who did not diet, and moderate dieters 5 times. The analysis then adjusted for previous dieting and psychiatric symptoms, taking out the effect of "had dieted before" and "already had psychiatric symptoms". After that adjustment, BMI (the measure of fatness worked out from weight divided by height squared), amount of exercise and sex no longer predicted new eating disorders. The authors conclude that "using exercise rather than food restriction to control weight in adolescents appears to carry a lower risk of eating disorders" - Evidence grade: A -- Sources: Arcelus J et al. (2011). Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry. ；Patton GC et al. (1999). Onset of adolescent eating disorders: population based cohort study over 3 years. BMJ. -- Notes: "Severe dieting" here means keeping food far below what the body needs over a long period, not skipping one meal or cutting out late-night snacks. If you have already started vomiting, feel guilty after bingeing, or cannot stop weighing yourself, go to a psychiatry department or the clinical psychology department of a general hospital. This is an illness with standard treatment; it is not a question of willpower. For BMI and mortality see item 33 in section 2; that item is about the long-term weight range, not about ways of losing weight. +- Sources: Arcelus J et al. (2011). Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry. ; Patton GC et al. (1999). Onset of adolescent eating disorders: population based cohort study over 3 years. BMJ. +- Notes: "Severe dieting" here means keeping food far below what the body needs over a long period, not skipping one meal or cutting out late-night snacks. If you have already started vomiting, feel guilty after bingeing, or cannot stop weighing yourself, go to a psychiatry department or the clinical psychology department of a general hospital. This is an illness with standard treatment; it is not a question of willpower. For BMI and mortality see item 32 in section 2; that item is about the long-term weight range, not about ways of losing weight. ### 2. Before injections, thread embedding or surgery, check two things: whether the clinic's licence lists "medical cosmetology", and whether the person operating is the designated attending physician @@ -23,7 +23,6 @@ This section does not judge what anyone wants to look like; that is each person' - Evidence grade: A - Sources: Ministry of Health (卫生部) (2002). Measures for the Administration of Medical Cosmetology Services（医疗美容服务管理办法）(Ministry of Health Order No. 19, articles two, eight, eleven, sixteen, twenty and twenty-four). ；National People's Congress (全国人大) (1997). Criminal Law（刑法）(article three hundred and thirty-six). - Notes: Article sixteen of the same Measures also provides that a cosmetic procedure may only be done in a medical cosmetology institution, or in a hospital or clinic that has a medical cosmetology department. So injections given at someone's home, or work done privately in a studio, already fail on the location alone. Article twenty states that a minor or anyone else who cannot decide for themselves may have cosmetic medicine only with the consent of a guardian, signed by a parent. Facials, massage and nail care are everyday beauty services and are not covered by these rules. But once a needle or a scalpel is involved, it counts as medical cosmetology. - ### 3. For fillers, avoid the bridge of the nose, the space between the eyebrows and the forehead, and never use an unlicensed injector to save money - Cost: No money. You simply inject somewhere else, or not at all. @@ -32,7 +31,6 @@ This section does not judge what anyone wants to look like; that is each person' - Evidence grade: B - Sources: Beleznay K et al. (2019). Update on Avoiding and Treating Blindness From Fillers: A Recent Review of the World Literature. Aesthet Surg J. - Notes: These numbers were counted case by case from cases reported in journals. They contain only the number of people who were hurt, not the total number of injections given. So you cannot work out "how likely one injection is to cause blindness"; you can only see which sites go wrong most often and how little can be done afterwards. That is also why the evidence is graded B. What a proper institution does to lower risk is routine anyway: first map the vessels under the place to be injected, use a blunt needle rather than a sharp one, inject small amounts while withdrawing, and have hyaluronidase ready beforehand. Every one of them requires the person doing it to be a trained doctor, which is the reason not to use an unofficial channel. Before you go, check it against item 2 in this section (check the clinic's licence and the attending physician). - ### 4. Do not buy diet pills, diet coffee, slimming sweets or enzyme plums that promise "fast weight loss" - Cost: No money. It also saves the money you would spend on these products. The hard part is accepting that there is no such thing as dropping several jin in a week. @@ -41,7 +39,6 @@ This section does not judge what anyone wants to look like; that is each person' - Evidence grade: A - Sources: James WPT et al. (2010). Effect of sibutramine on cardiovascular outcomes in overweight and obese subjects. NEJM. ；Tianjin Municipal Commission for Market Regulation (天津市市场监督管理委员会). Beware of "sibutramine hydrochloride" and "phenolphthalein" illegally added to some foods（警惕部分食品中违法添加「盐酸西布曲明」和「酚酞」）. - Notes: SCOUT enrolled people who already had heart disease or diabetes, a high-risk group. The trial did not show that healthy young people would have the same events. But because nobody can tell who is high-risk, the regulator simply took it off the market. Another substance often smuggled in is phenolphthalein; it was originally a laxative, and taking too much of it over a long period upsets the body's electrolytes and can bring on an arrhythmia. How to tell: a proper diet drug is a prescription drug, and its approval number can be looked up at the pharmacy and the hospital. Anything sold under the flag of "coffee", "sweets", "plums" or "meal replacement" that promises to drop several jin in a week should by definition have no drug effect; if it has one, something has been added. The original text of 国食药监办〔2010〕432 号 is on the National Medical Products Administration website; opening it on this machine has always returned error 412, so a provincial regulator's page repeating it is cited instead. The two sentences "the risks outweigh the benefits" and "production, sale and use stopped in October 2010" can be checked word for word there. - ### 5. Do not use anabolic steroids ("muscle-building injections", "oral steroids") to build muscle - Cost: No money, and it saves the money for the drugs. The hard part is accepting that muscle grows slowly. @@ -59,7 +56,6 @@ This section does not judge what anyone wants to look like; that is each person' - Evidence grade: B - Sources: State Administration for Market Regulation (国家市场监督管理总局) (2023). Measures for the Administration of the Quality Supervision of Drug Distribution and Use（药品经营和使用质量监督管理办法）(SAMR Order No. 84). ；State Administration for Market Regulation (国家市场监督管理总局) (2022). Measures for the Supervision and Administration of Online Drug Sales（药品网络销售监督管理办法）(SAMR Order No. 58, articles eight to ten). - Notes: The law can be checked word for word, but nobody has run a quantitative study on how much fewer goes wrong when you take a medicine as prescribed than when you buy it and take it yourself, so the grade is B for "hard to quantify". The steroids and the sex hormones covered in items 5 and 7 in this section follow the same rules. When it really is time to rely on medicine to lose weight, the doctor first looks at your body mass index, whether you have other illnesses, and whether there is any reason you cannot take the drug, then decides whether to prescribe and which one. That judgement cannot be bought. - ### 7. Use sex-hormone drugs only when a doctor prescribes them and you have regular check-ups; do not buy online, do not raise the dose yourself - Cost: Regular registrations, plus blood tests for the check-ups. A few hundred yuan a year, taking up several half-days. @@ -68,7 +64,6 @@ This section does not judge what anyone wants to look like; that is each person' - Evidence grade: A - Sources: Getahun D et al. (2018). Cross-sex Hormones and Acute Cardiovascular Events in Transgender Persons: A Cohort Study. Ann Intern Med. ；State Administration for Market Regulation (国家市场监督管理总局) (2022). Measures for the Supervision and Administration of Online Drug Sales（药品网络销售监督管理办法）(SAMR Order No. 58, article nine). - Notes: The researchers set out the weaknesses of the study themselves: how many people bought their drugs outside the hospital system they cannot see, so the line of "was the drug taken or not" was never clean. That is all the more reason to put the use of the drug inside regular follow-up. This study only looks at whether the vessels had an accident; it does not answer whether hormone therapy should be used or what it buys you, and that is another body of evidence. Every place where sex hormones are used is the same, including the pill, hormone replacement around menopause, and testosterone replacement. Anyone who buys online, doses from an internet post, and never has clotting, liver and kidney markers checked, runs the risk in the stretch where nobody is watching, and that has nothing to do with why the drug was taken. The proper route in China is the endocrinology department, or the relevant specialty, of a proper medical institution: assessment, prescription, follow-up. - ### 8. If "thinking you are ugly" has got to the point of checking the mirror repeatedly and wanting to change repeatedly, get a body-image assessment before talking about surgery - Cost: One registration at a psychiatry department or a clinical psychology department, a few dozen yuan. The hard part is making that trip before you book the operation. diff --git "a/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" "b/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" index 46d5a6a2..86a24d83 100644 --- "a/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" +++ "b/en/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" @@ -14,7 +14,6 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: A - Sources: Mostofsky E, et al. (2012). Risk of acute myocardial infarction after the death of a significant person in one's life: the Determinants of Myocardial Infarction Onset Study. Circulation. ；Carey IM, et al. (2014). Increased risk of acute cardiovascular events after partner bereavement: a matched cohort study. JAMA Internal Medicine. - Notes: Hand the pill box to somebody else to keep, and do not rely on your own memory. Do not drive long distances in those days. Do not try to carry it with alcohol; the account for alcohol is in section 2. How to recognise a heart attack or a stroke and how to call an ambulance, see section 13. Both studies followed groups of people and looked back, so they cannot separate how many of them were going to have an event anyway. But both point the same way and cover the same period. If there is nobody to hand the pill box to and you are certain to be alone in the house for those days, see item 6 in this section. - ### 2. In the week you are told you have a serious illness, do not go for the results on your own, and put off every decision other than treatment - Cost: No money. Have somebody go with you to pick up the results. Push back by a few days any big decision other than treatment. @@ -23,7 +22,6 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: A - Sources: Fang F, et al. (2012). Suicide and cardiovascular death after a cancer diagnosis. New England Journal of Medicine. - Notes: These are data on cancer patients. There is no source of the same weight for other serious diagnoses. The number of people involved is actually small: a suicide rate of 2.50/1000 person-years in the first week, that is 2.5 cases per 1000 people per year. This is written down not to frighten you but to show that somebody has to be there in the days just after you find out. Long-term management after diagnosis, reimbursement and follow-up are in section 16 and section 24. If you have nobody to go with, see item 6 in this section (put the neighbours and the community list in place of "that person"). - ### 3. After losing your job, fix your daily routine, your medical insurance and the rhythm of your job hunt first; do not spend the whole day at home - Cost: No money. Go and register as unemployed and claim unemployment insurance benefits first, then put job hunting into a fixed schedule. The hard part is getting up on time when nobody is making you. @@ -32,7 +30,6 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: A - Sources: Roelfs DJ, et al. (2011). Losing life and livelihood: a systematic review and meta-analysis of unemployment and all-cause mortality. Social Science & Medicine. - Notes: This is a study that followed groups of people and looked back, so it cannot tell cause from effect. People who were already in poor health were also more likely to lose their job. But "after taking out the effect of lifestyle habits the effect is 24% lower" gives you somewhere to start. What you can claim after losing your job, where to claim it and how to keep social insurance going, see section 7. What compensation you are owed on dismissal, and not to sign a voluntary resignation, see section 19. - ### 4. If a relative died by suicide, accident or murder, do not expect to tough it out; go and look for professional help - Cost: Calling 12356 costs nothing. To see a doctor, registering for a psychological outpatient clinic costs what an ordinary outpatient appointment costs. The hard part is being the one who asks. @@ -41,7 +38,6 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: A - Sources: Erlangsen A, et al. (2017). Association Between Spousal Suicide and Mental, Physical, and Social Health Outcomes: A Longitudinal and Nationwide Register-Based Study. JAMA Psychiatry. ；Pitman A, et al. (2014). Effects of suicide bereavement on mental health and suicide risk. The Lancet Psychiatry. ；Kristensen P 等 (2012). Bereavement and mental health after sudden and violent losses: a review. Psychiatry. （the review referred to in the notes） - Notes: A sudden, violent death means an accident, a murder, a disaster. After a death like this, the proportions of post-traumatic stress disorder, depression and grief that has got stuck are all higher than after an illness. Recovery is slower as well. What to do when you have thoughts of suicide yourself, see item 25 in section 1. - ### 5. In the first six months after a spouse dies, hand meals, sleep, the medication for a chronic illness and the follow-up appointments to one fixed person to keep an eye on them - Cost: No money. Find a specific person, a child, a brother or sister, or a friend, and have them come round or telephone on a set schedule. The hard part is asking someone to take this on. @@ -50,14 +46,15 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: A - Sources: Moon JR, et al. (2011). Widowhood and mortality: a meta-analysis. PLoS ONE. ；Shor E, et al. (2012). Widowhood and mortality: a meta-analysis and meta-regression. Demography. - Notes: This is a study that followed groups of people and looked back; other factors cannot be taken out of it. A married couple's lifestyle and illnesses are similar to begin with. But the pattern that "the first six months are far higher than the ones after" is the same in two independent meta-analyses, and it is worth treating as a window of care. Men's risk is higher, and they also ask for help less often. If there is a widowed older person at home, sort out the accounts, guardianship and wills together as in section 17. If you cannot find such a person (you live alone, you have no children left, or your children are gone too), hand this to the community and to your phone, see item 6 in this section. -### 6. If you have neither relatives nor friends, replace "the person keeping an eye on you" with three things: a neighbour who can get into your home, the community's visiting list, and an emergency contact in your phone + +### 6. If you have neither relatives nor friends, replace 「the person who keeps an eye on you」 with three things: a neighbour who can get into your home, the community visiting list, and an emergency contact in your phone -- Cost: No money. The hard part is speaking up, telling a neighbour and the residents' committee "I live on my own, something has just happened at home". -- In plain language: With neither relatives nor friends, there are two things to look after: that somebody finds out if something goes wrong, and that you take your medication and eat on time. The medication and the meals rest on you: write a list, set an alarm, do not rely on memory. Somebody finding out rests on three things: leave a spare key with a neighbour, ask the residents' committee whether you can be put on the visiting and care list, and set up an emergency contact in your phone and not switch it off. -- Benefit: 44,573 outpatients, all of whom had or were at risk of atherothrombosis (the kind of condition where plaques grow inside the arteries and clots form easily). Doctors in several countries registered them into the international REACH project, entry from 2003-12 to 2004-12, and they were followed for 4 years. 8,594 of them (19%) lived alone. Among those living alone, all-cause death over the 4 years was 14.1%, against 11.1% for those living with others. Cardiovascular death 8.6% against 6.8%. Both log-rank P<0.01, so the difference is unlikely to be chance. The gap differs by age group (interaction P=0.03). At 45 to 65, all-cause death living alone against living with others was 7.7% against 5.7%. After taking out other conditions, HR 1.24 (95% CI 1.01–1.51), that is about twenty-four per cent higher. At 66 to 80 it was 13.2% against 12.3%, HR 1.12 (1.01–1.26), about ten per cent higher. Above 80 it was 24.6% against 28.4%, HR 0.92 (0.79–1.06), no rise seen. A 2022 opinion from the Ministry of Civil Affairs (民政部) and ten other departments provides for visiting and care services. The subjects are elderly people who live alone, have empty-nest households, have children left behind, are disabled or severely disabled, and families in special family-planning circumstances. The forms of service are 「定期上门入户、电话视频、远程监测」, that is regular visits to the home, telephone and video, and remote monitoring. Townships (subdistricts) do the headcount and the screening, village (residents') committees help; they record the basic information and 「接受探访关爱服务的意愿」, the willingness to receive visiting and care services, and bring it into grassroots grid management, which means each area is assigned to a named person to keep an eye on it. By the end of 2025 「确保特殊困难老年人月探访率达到 100%」, that is make sure the monthly visiting rate for elderly people in special difficulty reaches 100%, and 「对失能的特殊困难老年人，每月探访关爱不少于一次」, at least one visit a month for disabled elderly people in special difficulty. The document also advocates that households install smart call systems, smart water meters, health monitoring products and elderly monitoring devices, so that 「一旦监测异常，能够及时预警，并同步向紧急联系人发送提示信息」, that is as soon as monitoring shows an abnormality an alert can be raised in time and a message sent at the same time to the emergency contact. Where service staff find an emergency they 「应当第一时间协助拨打紧急求助电话」, that is they shall assist in dialling the emergency number at the first opportunity +- Cost: No money. The hard part is speaking up: saying to a neighbour or to the residents' committee 「I live alone and something has just happened at home」. +- In plain language: When you have neither relatives nor friends, there are two things to look after: that somebody finds out if something happens, and that you take your medicine and eat on time. The medicine and the meals depend on you, so write them as a list and set an alarm instead of relying on memory. Somebody finding out depends on three things: leave a spare key with a neighbour, ask the residents' committee whether you can go on the visiting-care list, and set up an emergency contact on your phone and not switch it off. +- Benefit: 44,573 outpatients, all of them at risk of atherothrombosis or with a history of it (the kind of trouble in which plaques grow inside the vessels and clots block them easily). Doctors in several countries registered them into the international project REACH and followed them from 12 December 2003 to 12 December 2004, over 4 years. 8,594 of them (19%) lived alone. Among those living alone, all-cause death over the 4 years was 14.1%, against 11.1% among those living with somebody. Cardiovascular death was 8.6% against 6.8%. Both log-rank tests P<0.01, so the difference is unlikely to be chance. The gap differs between age groups (interaction P=0.03). At 45 to 65, death among those living alone against those living with somebody was 7.7% against 5.7%. After adjusting for the other conditions, HR 1.24 (95% CI 1.01–1.51), that is about a quarter higher. At 66 to 80 it was 13.2% against 12.3%, HR 1.12 (1.01–1.26), about a tenth higher. Above 80 it was 24.6% against 28.4%, HR 0.92 (0.79–1.06), and there was no rise at all. The 2022 opinion of the Ministry of Civil Affairs and ten other departments requires a visiting-care service to be set up. It covers older people who live alone, live apart from their children, were left behind, are disabled, are severely disabled, or come from families with special circumstances under the family planning policy. The service takes the form of 「regular home visits, telephone and video, remote monitoring」. Townships (subdistricts) do the first sweep to find the cases and village (residents') committees help, recording basic information and 「the willingness to receive visiting-care services」 and putting people into the grassroots grid management system, which means each area is assigned to a named person who keeps an eye on it. By the end of 2025 「the monthly visiting rate for elderly people in special difficulty is to reach 100%」, and 「for disabled elderly people in special difficulty, visiting care at least once a month」. The document also advocates that households install smart call systems, smart water meters, health monitoring products and elderly monitoring devices, 「so that when the monitoring shows an abnormality it can give a warning in time and send a message to the emergency contact at the same time」. Where staff find an urgent problem, they 「shall help dial the emergency number at the first opportunity」 - Evidence grade: B -- Sources: Udell JA, et al. (2012). Living alone and cardiovascular risk in outpatients at risk of or with atherothrombosis. Archives of Internal Medicine. ；Ministry of Civil Affairs (民政部) and ten other departments (2022). Guiding Opinions on Carrying Out Visiting and Care Services for Elderly People in Special Difficulty（关于开展特殊困难老年人探访关爱服务的指导意见）(民发〔2022〕73 号). -- Notes: How to do the three things. First, give a spare key or the temporary code of your door lock to a neighbour or to the building management, and say "I am on my own for the next few days". You do not need to be close to them; you only need somebody to know you are in the house. Second, tell the community residents' committee that you live alone and ask whether you can be put on the visiting and care list. Among the elderly, those who live alone, have empty-nest households or have no children left are named official subjects for the service; while you are there, ask whether your area has free smart call devices or smart water meters to install. Third, set up an emergency contact and a medical emergency card in your phone, and do not silence it or switch it off for those days. There are two reasons for the B. First, the link between living alone and mortality comes from a study that followed a group and looked back, so the reverse explanation, that people who were already in poor health and worse off are more likely to live alone, cannot be taken out; the original paper itself says the result remains to be confirmed. Second, the visiting and care document covers elderly people only. There is no matching scheme for anyone under 60. What you can rely on is one line in that document, which requires grid workers and social workers to 「及时发现家庭变故、失业、失学等心理危机风险」, that is identify in good time psychological crisis risks such as family upheaval, job loss and dropping out of school (see item 11 in this section, call 12356 and have somebody to talk to). Why "somebody in the room" is worth so much, see item 1 in section 13 (pressing on the chest). Of cardiac arrests outside hospital, 79.2% happen at home. Where somebody is there doing compressions, 16.1% are discharged alive, against 3.9% where nobody is. Nobody in the room puts you in the lower band. Two more things you can do in passing. A funeral itself means going to the funeral home, the community health service, the police and the housing provident fund, several counters in all (see section 25). Put that into a fixed schedule of one trip out a day, and the arrangements for going out during those days are already there. The hotline can be called as often as you like; it is not a single permitted call (see item 11 in this section, call 12356 and have somebody to talk to). "Living alone" is itself a long-term risk factor. The mortality ratio for living alone is 1.32, about thirty per cent higher, see item 10 in section 22 (treating regular contact with other people as a health expense). But this item only covers the next few days; the long-term account is a separate one. +- Sources: Udell JA, et al. (2012). Living alone and cardiovascular risk in outpatients at risk of or with atherothrombosis. Archives of Internal Medicine. ; Ministry of Civil Affairs (民政部) and ten other departments (2022). Guiding Opinions on Carrying Out Visiting-Care Services for Elderly People in Special Difficulty, 关于开展特殊困难老年人探访关爱服务的指导意见（民发〔2022〕73 号）. +- Notes: How to do the three things in practice. First, give a spare key, or the temporary code of your door lock, to a neighbour or to the building management, and tell them 「I am on my own for a few days」. You do not need to be close friends; you only need somebody to know that you are in the flat. Second, tell the community residents' committee that you live alone and ask whether you can be put on the visiting-care list. Among older people, those who live alone, live apart from their children or have lost their children are named outright as the service's target groups; while you are there, ask whether your area has free smart call devices or smart water meters of that kind. Third, set up an emergency contact and a medical emergency card on your phone, and for the next few days do not put it on silent and do not switch it off. There are two reasons for the grade of B. First, the association between living alone and mortality comes from a study that looked back over existing population data, and you cannot remove the reverse explanation that 「people who were already in poor health or in poor conditions are more likely to live alone」; the paper itself says the result awaits confirmation. Second, the visiting-care document covers older people only. There is no matching special scheme for anybody under 60. What you can rely on is one line in the document. It requires grid workers and social workers to 「identify in good time psychological crisis risks such as family upheaval, job loss and dropping out of school」 (see item 11 in this section, call 12356 and have somebody to talk to). Why 「somebody in the flat」 is worth so much: see item 1 in section 13 (pressing hard on the chest). Of the cardiac arrests that happen outside hospital, 79.2% happen at home. Where somebody next to them does the pressing, 16.1% leave hospital alive, against only 3.9% where nobody does. Nobody in the flat means you land in the lower band. Two other things come with the same trip. The funeral arrangements themselves mean going to several counters: the funeral home, the community health service, public security and the housing provident fund (see section 25). If you make that a fixed part of going out once a day, you already have the reason to be out of the house these days. The hotline can be called again and again; it is not a one-off call (see item 11 in this section, call 12356 and have somebody to talk to). 「Living alone」 is itself a long-term risk factor. The mortality ratio for living alone is 1.32, about a third higher: see item 9 in section 22 (treat 「meeting people regularly」 as a health expense). But this entry is only about the next few days; the long-term account is counted separately. ### 7. After a parent dies, tell the child the truth, let them attend the farewell, and do not send them somewhere else to "change the scenery" @@ -67,7 +64,6 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: B - Sources: Li J, et al. (2014). Mortality after parental death in childhood: a nationwide cohort study from three Nordic countries. PLoS Medicine. - Notes: Do not make up "has gone away on a long trip" when you talk to the child; the arrangements for guardianship and support have to be put in writing. The grade is not B because the numbers are weak. This is a follow-up study of the entire populations of three countries, and the numbers themselves would support an A. It is B because between that number and the specific practice of "how to talk to the child and whether to let him attend the farewell" there is no direct trial evidence. The authors say so themselves: after bereavement there is a shortage of data on the parent-child relationship and on how children live. The increase contains both an inborn-constitution part and a long-term effect left behind by the bereavement. Naming a guardian and arranging property, see section 17. Child support and schooling, see section 18. - ### 8. If grief has gone six months and you are still stuck and cannot get through your days, register with psychiatry or clinical psychology - Cost: The registration fee plus the treatment. This kind of treatment runs in courses; in the trials it averaged 16 sessions over 19 weeks. The hard part is being the one who goes and registers. @@ -76,7 +72,6 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: A - Sources: Lundorff M, et al. (2017). Prevalence of prolonged grief disorder in adult bereavement: A systematic review and meta-analysis. Journal of Affective Disorders. ；Prigerson HG, et al. (2009). Prolonged grief disorder: Psychometric validation of criteria proposed for DSM-V and ICD-11. PLoS Medicine. ；Shear MK, et al. (2005). Treatment of complicated grief: a randomized controlled trial. JAMA. - Notes: Do not count the months yourself. The real test is whether it is affecting eating, sleep, work and looking after the children; if it is, go and register. The departments are called psychiatry, psychology or clinical psychology, and every public psychiatric specialty hospital has them. This diagnosis goes under different names in different sources: complicated grief, and prolonged grief disorder. The standards also differ on how long it has to last: some say 6 months, some say 12. The 9.8% figure comes from 14 studies that differ quite a lot in method, and the authors warn that it may not carry over to other populations. - ### 9. Do not go straight to paid grief counselling; first check whether your grief really has got stuck (the symptoms in item 8) - Cost: No money, and you save the counselling fee instead. @@ -85,6 +80,7 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: A - Sources: Currier JM, et al. (2008). The effectiveness of psychotherapeutic interventions for bereaved persons: a comprehensive quantitative review. Psychological Bulletin. ；Wittouck C, et al. (2011). The prevention and treatment of complicated grief: a meta-analysis. Clinical Psychology Review. ；Bonanno GA, et al. (2002). Resilience to loss and chronic grief: a prospective study from preloss to 18-months postloss. Journal of Personality and Social Psychology. ；Stroebe M 等 (2007). Health outcomes of bereavement. Lancet. （the review referred to in the notes） - Notes: The point is "not everybody has to do it", not "nobody should do it". People at high risk of the kind covered by item 4 (a relative died by suicide, accident or murder) and people who have already got stuck, of the kind covered by item 8 (grief has got stuck, go to psychiatry), are exactly the ones for whom there is evidence that it should be done. The conclusion of a Lancet review says the same: grief is not an illness, and most people do not need professional intervention. Spend the effort on the people at high risk. And on those who already have complicated grief, depression after bereavement, or a stress disorder. + ### 10. After you separate or divorce, do not let medical insurance lapse and do not lean on alcohol - Cost: No money. Keep the medical insurance going and stop the alcohol. The hard part is that this is a period when leaning on alcohol is easy. @@ -93,7 +89,6 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: A - Sources: Sbarra DA, et al. (2011). Divorce and Death: A Meta-Analysis and Research Agenda for Clinical, Social, and Health Psychology. Perspectives on Psychological Science. - Notes: The original only says "clearly higher" and compares the groups; it gives no multiple that can be cited, so no figure is given here. The benefit is graded medium on the basis of "you know the direction, and there is no reduction you can apply a threshold to". Property, debts and bride price in a divorce, see section 8 and section 10. Arrangements for the children, see section 18. - ### 11. To have somebody to talk to, call 12356; minors and teenagers call 12355; to see a doctor, register for a psychological outpatient clinic - Cost: No money. Both hotlines are free. A psychological outpatient clinic is registered for like an ordinary outpatient appointment. @@ -102,7 +97,6 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: B - Sources: National Health Commission (国家卫生健康委) and 25 other departments (2026). Notice on Issuing the Implementation Plan for a Sound Social Psychological Service System and Crisis Intervention Mechanism（关于印发健全社会心理服务体系和危机干预机制实施方案的通知）(国卫医政发〔2026〕8 号). - Notes: The grade is B because the counselling rooms in villages and communities and the psychological outpatient clinics at county level are targets for 2030, not things every place has now. 12356 and 12355, on the other hand, you can call today. When 12356 opened, and how long it is answered each day, see item 25 in section 1. Psychological counselling and psychological therapy are not the same thing. Only a psychiatrist can make a diagnosis or write a prescription, and the treatment has to be done at a qualified medical institution. The "healing" and "spiritual counselling" sold by the hour on the market are not part of this system. - ### 12. In the first three months after a shock, put every irreversible big decision off - Cost: No money. You are only putting the decision off, and you may miss some opportunities because of it. The hard part is that at the time it feels as though the opportunity will not wait. @@ -111,12 +105,11 @@ The procedures at the time a relative dies, the body, and the money you can clai - Evidence grade: C - Sources: The author's own experience, no direct literature; the wording "family upheaval, job loss and dropping out of school" as psychological crisis risks is from National Health Commission (国家卫生健康委) and 25 other departments (2026). Implementation Plan for a Sound Social Psychological Service System and Crisis Intervention Mechanism（健全社会心理服务体系和危机干预机制实施方案）(国卫医政发〔2026〕8 号). - Notes: If there is nobody around you who has no stake in the matter, call 12356 and say it out loud there (see item 11 in this section, call 12356 and have somebody to talk to), or write it down yourself and read it back three days later. The "three months" is a line the author drew himself, with no literature behind that specific length, so it is graded C. The common scams that target the bereaved and older people are in section 17 (investing in retirement, reverse mortgages on a home) and section 6 (fortune telling). The traps in guaranties and IOUs are in section 8. There are a few payments you have to go and claim yourself, and they are not on the "put it off" list: the housing provident fund balance, social insurance benefits, and the three payments on a work-related death, see section 25 and section 19. - -### 13. Do not use death as a way to pay off debts: life insurance does not pay for a suicide in the first two years, a work injury is not recognised, and the debts still come out of the estate first +### 13. Do not use death as a way of paying off debts: life insurance within two years does not pay out, a work injury is not recognised, and the debts are still taken out of the estate first -- Cost: No money. You are only getting this one account straight. -- In plain language: A life insurance contract does not pay out if you commit suicide within two years; it only returns the part of the policy value that can be returned. A work injury is not recognised either, and you get none of the three payments on a work-related death. The debts are taken out of the estate you leave first, and your family gets only what is left over. They also have to bear the health cost described in item 4 in this section (bereavement after suicide or a violent death). -- Benefit: Article forty-four of the Insurance Law of the People's Republic of China（中华人民共和国保险法）covers contracts in which the payment of insurance money is conditional on the death of the insured person. Counting from the date the contract was made or the date the contract's effect was restored, if the insured person commits suicide within two years, the insurer does not bear the responsibility for paying the insurance money. Where the person committing suicide is a person without capacity for civil conduct, that is, someone who in law cannot take responsibility for their own actions, this does not apply. In that case the insurance company shall return the cash value of the policy as agreed in the contract, that is, it only returns the part of the policy value that can be returned and does not pay the sum assured. The sum assured on a life insurance policy is usually in the tens of thousands of yuan, and the benefit is therefore graded large. Item three of article sixteen of the Regulations on Work-Related Injury Insurance（工伤保险条例）provides: 「自残或者自杀的，不得认定为工伤或者视同工伤」, that is, where a person injures themselves or commits suicide, it shall not be recognised as a work injury or treated as one. The three payments on a work-related death under article thirty-nine all fall away as a result. First the funeral subsidy, paid at 6 months of the local average monthly wage of employees in the previous year. Second the pension for dependents. Third the one-off work-related death benefit, at 20 times the national per-capita disposable income of urban and rural residents in the previous year. Article one thousand one hundred and fifty-nine of the Civil Code（民法典）provides: 「分割遗产应当清偿被继承人依法应当缴纳的税款和债务」, that is, when an estate is divided it shall pay off the taxes and debts that the deceased was lawfully liable for, so the estate pays the taxes and debts first and only what is left is divided. Article one thousand one hundred and sixty-one provides: 「继承人以所得遗产实际价值为限清偿」, that is, an heir repays debts within the actual value of the estate received, so only to the extent of what was inherited and with nothing more to be found from their own pocket, and one who renounces the inheritance bears no repayment responsibility +- Cost: No money. You are only working out what this account really comes to. +- In plain language: A life insurance policy does not pay out if you kill yourself within two years; you only get back what can be withdrawn from the policy. A work injury is not recognised either, and none of the three payments on a work-related death are made. Debts owed are taken out of the estate you leave first, and the family only gets what is left. They also have to bear the health cost described in item 4 in this section (bereavement after a death by suicide or violence). +- Benefit: Article 44 of the Insurance Law covers contracts under which the payment is conditional on the death of the insured person. Counted from the date the contract is formed or from the date its effect is restored, where the insured person commits suicide within two years the insurer is not liable to make the payment. It does not apply where the person who commits suicide is a person without civil capacity, that is, somebody who in law cannot answer for their own actions. In that case the insurer shall return the cash value of the policy as agreed in the contract, which means only the part of the policy that can be withdrawn, and it does not pay the sum insured. The sum insured on a life policy usually runs to tens of thousands of yuan, so the benefit is graded large. Article 16, paragraph 3 of the Regulations on Work-Related Injury Insurance says: where a person injures themselves or commits suicide, it shall not be recognised as a work injury or treated as one. The three payments on a work-related death in Article 39 therefore all fall through. First the funeral subsidy, paid as 6 months of the average monthly wage of employees locally in the previous year. Second the pension for dependants. Third the one-off payment on a work-related death, 20 times the per-capita disposable income of urban residents nationwide in the previous year. Article 1159 of the Civil Code says: when the estate is divided, the taxes and debts the deceased was legally liable for shall be paid off first, which means the estate pays the tax and the debts first and only what is left is divided. Article 1161 says that the heir pays within the limit of the actual value of the estate received, which means only the part that was inherited is used and you do not put your own money in; a person who renounces the inheritance need not bear the liability - Evidence grade: A -- Sources: Standing Committee of the National People's Congress (2015 revision). Article forty-four of the Insurance Law of the People's Republic of China（中华人民共和国保险法）. ；State Council (2010 revision). Regulations on Work-Related Injury Insurance（工伤保险条例）（国务院令第 586 号）articles sixteen and thirty-nine. ；National People's Congress (2020). Articles one thousand one hundred and fifty-nine and one thousand one hundred and sixty-one of the Civil Code of the People's Republic of China（中华人民共和国民法典）. -- Notes: Only the money account is counted here. The other items in this section and in section 1 are about not getting to this point. Article forty-four of the Insurance Law covers only the first two years; it is written out so that the provision is complete, not to give you a timetable. The health cost to the family, see item 4 in this section (a spouse whose partner died by suicide has a 70% to 80% higher risk of a newly diagnosed mental disorder within 5 years). What to do when the thought comes up, see items 25 and 32 in section 1, and the after-effects once someone is saved, see item 33 in section 1. The standards for the three payments on a work-related death, see item 16 in section 19. The specific procedures for the estate and the debts, see item 9 in section 25 (collect the housing provident fund and the social insurance benefits one by one). \ No newline at end of file +- Sources: Standing Committee of the National People's Congress (全国人大常委会) (2015, as amended). Article 44 of the Insurance Law of the People's Republic of China. ; State Council (国务院) (2010, revised). Regulations on Work-Related Injury Insurance, 工伤保险条例（国务院令第 586 号）, articles 16 and 39. ; National People's Congress (全国人大) (2020). Articles 1159 and 1161 of the Civil Code of the People's Republic of China. +- Notes: This entry only counts the money. The other items in this section and in section 1 are about not getting to this point. Article 44 of the Insurance Law covers only the first two years; it is written here to give the whole of the article, not to set out a timetable. The health cost to the family is in item 4 in this section (a spouse whose partner died by suicide runs a seventy to eighty per cent higher risk of a new mental disorder within 5 years). What to do when the thought first appears: see items 25 and 32 in section 1, and the after-effects once somebody has been brought back: see item 33 in section 1. The standards for the three payments on a work-related death are in item 15 in section 19. The actual procedures for the estate and the debts are in item 9 in section 25 (going to claim the housing provident fund and social insurance benefits one by one) diff --git "a/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" "b/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" index 7533f38c..dbdbf5f3 100644 --- "a/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" +++ "b/en/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" @@ -2,9 +2,9 @@ # 30. School-age children -Section 20 covers the first year of a child's life. Section 18 works out what it costs in money and time to raise a child. The dozen or so years between 3 and 18 are written here. The item on managing screens by age starts at 0 years old, because the official bands are 0–3, 3–6 and school age, and splitting them up makes them less useful. What this section answers is: when "do not let it interfere with the schoolwork" and "the body" collide, which things cannot wait. These things have time windows, and once a window has passed it cannot be made up. Whether schoolwork should be that competitive is not discussed here; this section gives the windows and the numbers. +Section 20 covers the first year after a child is born. Section 18 works out what raising a child costs in money and time. The dozen-odd years in between, from 3 to 18, are in this section. The item on managing screen time by age starts at 0, because the official bands are three: 0–3, 3–6 and school age, and writing them out separately only makes it harder to use. What this section answers is: when "do not let it delay schoolwork" and "the body" collide, which things cannot wait. Those things have a time window, and once it has passed they cannot be made up. Whether schoolwork should be this competitive is not discussed here; this section gives only the windows and the numbers. -The first eight items and the last two measure death and the physical harm that really lands. Item 9 (do not buy products claiming to "cure myopia") measures money. Item 10 (the written rules on sleep, homework, physical education and ranking) and item 11 (taking a leave of absence while keeping the student place) measure time. Drowning, traffic, helmets and the HPV vaccine are in section 1. Babies and toddlers are in section 20. Children being scammed and top-ups in games are in section 5. The lines minors themselves should not cross are in section 9. How to recognise an emergency and how to call an ambulance, see section 13. Children after a parent dies are in section 29. None of this is repeated here. +The first eight items, item 12 (dilated refraction), item 13 (fissure sealants) and items 15 to 17, count death and the real physical and psychological harm that is left behind. Items 16 and 17 among them are about how a family manages a child: no hitting and no shouting, and going to a parenting course that teaches specific techniques. Item 9 (not buying products that claim to "cure" myopia) counts money. Item 10 (the written rules on sleep, homework and physical education rankings), item 11 (taking a leave of absence while keeping the student record) and item 14 (a child who loves computer games) count time. Drowning, traffic, helmets and the HPV vaccine: see section 1. Infants: see section 20. Children being deceived and top-ups for computer games: see section 5. The lines a minor should not cross: see section 9. How to recognise an emergency and how to call a vehicle: see section 13. Children after a parent dies: see section 29. None of that is repeated in this section. ### 1. If a child says the pain came on suddenly and is getting worse, go to hospital at once; do not let him sit it out until school is over @@ -14,7 +14,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: A - Sources: Mellick LB, et al. (2019). A Systematic Review of Testicle Survival Time After a Torsion Event. Pediatric Emergency Care. - Notes: Say something to the boy in advance: if down there it suddenly hurts badly, say so at once, there is no need to be embarrassed about it. Testicular torsion is commonest in adolescent boys, and a good many sit on it because it is hard to bring up. Other conditions counted in hours the same way are acute appendicitis (which perforates if left), diabetic ketoacidosis and meningitis. This section has not checked item by item how much damage a delay of a few hours does in those. The signal they share is this: it starts suddenly, it keeps getting worse, and it is unlike anything that has happened before. How to recognise an emergency and how to call an ambulance, see section 13. - ### 2. Do not put off treatment that has to be done for the sake of "after the exam"; some windows follow bone age, not the examination timetable - Cost: No money. The only decision is whether to do it early or late. Leave and follow-up appointments take time; the cost of the treatment itself is separate. The hard part is that the brace has to be worn for at least 18 hours a day, and the child has to put up with it the whole time. @@ -23,7 +22,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: A - Sources: Weinstein SL, et al. (2013). Effects of bracing in adolescents with idiopathic scoliosis. New England Journal of Medicine. - Notes: Scoliosis is used as an example here because it has a randomised trial, because it has a clear time window, and because it happens to be one of the focus items of the student medical examination (see item 7 in this section). Others of the same kind are strabismus, amblyopia, dental caries, and operations for congenital heart disease where you can pick the time. What they have in common is that the later you do it the worse the result, or the higher the cost. Ask the specialist about the specific window. If you are afraid of missing lessons, a child can take a leave of absence during the compulsory education stage, for up to 1 year, and the student place is kept, see item 11 in this section. - ### 3. If a child is bullied, report it to the school the same day and ask for a written response; where it involves hitting, taking money or spreading rumours, call the police - Cost: No money. Go to the head teacher, and also to the school's body for handling bullying. To call the police, dial 110. The hard part is standing up to advice like "don't make a fuss". @@ -32,7 +30,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: A - Sources: van Geel M, et al. (2014). Relationship between peer victimization, cyberbullying, and suicide in children and adolescents: a meta-analysis. JAMA Pediatrics. ；Ministry of Education (教育部) (2021). Provisions on School Protection of Minors（未成年人学校保护规定）（教育部令第 50 号，articles twenty-one, twenty-two and twenty-three）. - Notes: If a child has an injury whose origin cannot be explained, suddenly refuses to go to school, and things keep going missing, ask about it first using the signals in article twenty-two. This analysis pools several studies together; it only follows records and has no control group, so it cannot show that bullying alone causes suicide, but the direction and the size of the risk are clear. The Plan on Strengthening the Comprehensive Governance of Bullying among Primary and Secondary School Students《加强中小学生欺凌综合治理方案》additionally requires a strict distinction between bullying and the rough-and-tumble play of ordinary students. So the school is obliged to make a finding, and cannot settle the matter with one line about children playing. - ### 4. Two hours outdoors for the child every day; this is the only way of preventing myopia that randomised trials support - Cost: No money. But you have to set aside time every day, and it will mostly have to come out of the time for homework and for online classes. The hard part is doing it every day, not a few times. @@ -41,6 +38,7 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: A - Sources: He M, et al. (2015). Effect of Time Spent Outdoors at School on the Development of Myopia Among Children in China: A Randomized Clinical Trial. JAMA. ；General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents（防控儿童青少年近视核心知识十条）（国卫办妇幼函〔2023〕278 号）. - Notes: This trial only ran for 3 years, and the children who took part all started at 6 years old. A 9.1 percentage point difference in the proportion with myopia can stop some people from becoming myopic; it cannot guarantee that somebody will not become myopic. The same Ten Key Points gives several other practices. For reading and writing, "one foot, one fist, one inch". After 20 minutes of looking at something close, look into the distance for at least 20 seconds. A primary or secondary school student using an electronic screen for anything other than study, no more than 15 minutes at a time and no more than 1 hour a day in total. "Being outdoors" and "having a physical education lesson" are not the same thing; standing outside at break counts too. + ### 5. No screens from 0 to 3 years old, as few as possible from 3 to 6, and no more than 1 hour a day for primary and secondary school students using a screen for anything other than study - Cost: No money. The hard part is taking the screen out of the child's hands, and the parents cutting down their own use in front of the child. @@ -49,7 +47,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: B - Sources: General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents（防控儿童青少年近视核心知识十条）（国卫办妇幼函〔2023〕278 号）. ; Ministry of Education (教育部) and seven other departments (2018). Implementation Plan for the Comprehensive Prevention and Control of Myopia in Children and Adolescents（综合防控儿童青少年近视实施方案）（教体艺〔2018〕3 号）. ; Foreman J, Salim AT, Praveen A, et al. (2021). Association between digital smart device use and myopia: a systematic review and meta-analysis. The Lancet Digital Health, 3(12), e806-e818. ; Madigan S, McArthur BA, Anhorn C, Eirich R, Christakis DA (2020). Associations between screen use and child language skills: a systematic review and meta-analysis. JAMA Pediatrics, 174(7), 665-675. - Notes: The grade is B because what the official documents give are banded rules, not numbers measured in a trial. The lower bound of the odds ratio in the research sits exactly at 1.00, and the authors only dare say "possibly associated". So treat it as a good house rule to carry out, not as "no screens means no myopia". What is really supported by a randomised trial is item 4 in this section (2 hours outdoors). Preschool children need their eyes checked too: the same document states that at 1–3 years old, 4–6 years old and after the age of 7 all children should have regular refractive screening to see how much far-sightedness reserve is left; the method is in item 12 in this section (dilation and refraction testing). A meta-analysis has also been done on the relationship between screens and children's language skills: 42 studies, 18905 people, the longer the screen time the weaker the language skills (r=−0.14, 95% CI −0.18 to −0.10). But watching together, and watching programmes with educational content, are positively associated in the other direction (r=0.16 and 0.13). That is association, not causation; all it shows is that replacing talking with a child by putting a screen in front of them is not worth doing. - ### 6. Treat myopia as damage that stays: the deeper the degree, the more retinal detachment, macular disease and glaucoma later on - Cost: No money. You have nothing to do about it yourself; what has to be done is in items 4 and 5 (2 hours outdoors, managing screens by age). @@ -58,7 +55,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: A - Sources: Haarman AEG, Enthoven CA, Tideman JWL, Tedja MS, Verhoeven VJM, Klaver CCW (2020). The complications of myopia: a review and meta-analysis. Investigative Ophthalmology & Visual Science, 61(4), 49. ; General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents（防控儿童青少年近视核心知识十条）（国卫办妇幼函〔2023〕278 号）. - Notes: Everything pooled into this is observational research, and the odds ratios are unusually large. The credible range for high myopia on macular degeneration runs from 230 to 3104, which means the studies underneath it differ a great deal; the figure can only be read for its order of magnitude and must not be used as an exact value. Most of these complications only show up in middle and old age. Myopia itself is not a disability; the great majority of people have normal vision once they wear glasses to correct it. But it is not a matter of putting on a pair of glasses and being done with, and it is worth spending effort on delaying it and controlling it. How to follow up after diagnosis, see item 12 in this section (dilation and refraction testing). Do not buy products claiming they can cure it, see item 9 in this section (curing myopia). - ### 7. Read the report from the annual student health examination yourself, and take any abnormal item to a hospital the same year - Cost: No money. In the compulsory education stage the examination fee is paid out of the school's public budget, and parents do not pay separately. @@ -67,7 +63,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: B - Sources: National Health Commission (国家卫生健康委) and Ministry of Education (教育部) (2021). Measures for the Administration of Health Examinations of Primary and Secondary School Students (2021 Edition)《中小学生健康体检管理办法（2021 年版）》(国卫医发〔2021〕29 号). - Notes: The grade is B because this is a document telling schools how to do it, not a study of effects. What it guarantees is "examine every year, examine these items, follow up anything abnormal"; it does not guarantee that somebody will act on what is found. The examination is only a first screen, not a diagnosis. Poor vision needs dilation and refraction testing in the eye department (item 12 in this section). Abnormal curvature of the spine needs the orthopaedic or spinal surgery department (item 2 in this section, do not put off treatment to wait for an exam). Both of those steps have to be taken by a parent taking the child there. - ### 8. Screen children aged 12 to 18 for depression once; do not take the school's psychological questionnaire for a diagnosis - Cost: 0 to several tens of yuan, filling in a screening questionnaire at a community centre or an outpatient clinic. The 12355 and 12356 hotlines cost nothing to call. @@ -76,7 +71,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: B - Sources: US Preventive Services Task Force (2022). Screening for Depression and Suicide Risk in Children and Adolescents: US Preventive Services Task Force Recommendation Statement. JAMA. ；General Office of the Ministry of Education (教育部办公厅) (2025). Ten Measures to Further Strengthen the Mental Health Work for Primary and Secondary School Students《进一步加强中小学生心理健康工作十条措施》（教基厅〔2025〕2 号）. - Notes: The grade is B because what the original gives is a rating of the strength of the recommendation, not an effect that can be turned into a number. It is also a US recommendation; China has no matching official screening guidance. A positive screening result only means you have to do one more assessment; it is not a diagnosis. What to do when there are thoughts of suicide, see item 25 in section 1. How to handle it at home, see section 29. - ### 9. Do not buy products and services claiming they can "cure myopia" or "reduce the degree" - Cost: No money, and you save the money as well. @@ -85,6 +79,7 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: B - Sources: General Office of the State Administration for Market Regulation（市场监管总局办公厅）(2021). Notice on Carrying Out a Special Rectification Campaign Against Unlawful Commercial Marketing and Publicity for Myopia Prevention and Control Products for Children and Adolescents（关于开展儿童青少年近视防控产品违法违规商业营销宣传专项整治行动的通知）. - Notes: The grade is B because what it rests on is the standard the regulators use when they take action, not something measured by comparing products in a trial. But the statement that myopia cannot be cured is the same from the health department and the regulators: the Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents《防控儿童青少年近视核心知识十条》says「近视可防可控不可逆」, that is, myopia can be prevented and controlled but not reversed. The two things that really have evidence behind them are item 4 in this section (outdoors) and item 12 in this section (standard refraction testing, glasses and follow-up). For blue-light-blocking glasses, see section 6. + ### 10. Sleep, homework, physical education and ranking all have written rules, and where the school cannot meet them you can raise it - Cost: No money. Knowing the provisions is enough; if necessary raise it with the school or with the local education administration. The hard part is being the one who raises it. @@ -93,7 +88,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: B - Sources: General Office of the Ministry of Education (教育部办公厅) (2021). Notice on Further Strengthening the Management of Sleep for Primary and Secondary School Students《关于进一步加强中小学生睡眠管理工作的通知》（教基厅函〔2021〕11 号）. ；General Office of the Ministry of Education (教育部办公厅) (2025). Ten Measures to Further Strengthen the Mental Health Work for Primary and Secondary School Students《进一步加强中小学生心理健康工作十条措施》（教基厅〔2025〕2 号）. - Notes: Both of these are documents issued by the authorities in charge. How far each school carries them out varies a great deal, which is why this item is graded B. Their use is that you know which things the school was supposed to be doing anyway, so you have something to point to when you talk to them. The sleep account for adults, see section 2 and section 3. - ### 11. If a child cannot carry on any more they can take a leave of absence, and the school must keep their student place, for up to 1 year - Cost: No money. A parent or other legal guardian applies to the school. The hard part is having to accept that the child stops for a while. @@ -102,7 +96,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: B - Sources: Ministry of Education (教育部) (2025). Measures for the Administration of Student Status for Primary and Secondary School Students《中小学生学籍管理办法》（教基〔2025〕1 号，article sixteen）. ；General Office of the Ministry of Education (教育部办公厅) (2025). Ten Measures to Further Strengthen the Mental Health Work for Primary and Secondary School Students《进一步加强中小学生心理健康工作十条措施》（教基厅〔2025〕2 号）. - Notes: Before you go and do it, ask three things: what medical certificate is required, how the student status is recorded in the system during the leave, and whether coming back means returning to the original year or dropping a year. The grade is B because what conditions a leave requires, what materials have to be handed in and how the return is joined up are all set by the provincial education department; at the national level there is only a framework. The Provisions on School Protection of Minors《未成年人学校保护规定》also require compulsory education schools to register and file students who have dropped out, taken a leave of absence or been away for a long time, so a leave of absence is something the school has handled before, not a special exception for you alone. - ### 12. If poor vision is found, have dilation and refraction testing done at a hospital, and afterwards go back at the interval the doctor gives you - Cost: Tens to hundreds of yuan for the examination and for the glasses. After the dilation the child will be sensitive to light and unable to see near things clearly, and it takes half a day to a day or two to settle. @@ -111,7 +104,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: B - Sources: General Office of the National Health Commission (国家卫生健康委办公厅) (2023). Ten Key Points on the Prevention and Control of Myopia in Children and Adolescents（防控儿童青少年近视核心知识十条）（国卫办妇幼函〔2023〕278 号）. - Notes: The grade is B because this is the method given in an official popular-science document, not an original study with specific figures. High myopia brings retinal detachment and similar trouble, so the follow-up is not only about changing glasses. The "poor vision" found on the student health examination is only a screening result; it still has to be taken to hospital for one complete eye examination, see item 7 in this section. - ### 13. Get pit and fissure sealants once the first permanent molars are through - Cost: Tens to a hundred or two hundred yuan per tooth. No anaesthetic and no drilling, and it takes a few minutes. Many places have a free programme for children who have reached the age. @@ -120,7 +112,6 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: A - Sources: Ahovuo-Saloranta A, et al. (2017). Pit and fissure sealants for preventing dental decay in permanent teeth. Cochrane Database of Systematic Reviews. - Notes: The first permanent molar (the big tooth that will not be replaced) generally comes through at around 6 years old, the second at around 12. The sealant has to go on while the tooth has only just come through and before it has decayed; once there is a cavity all you can do is fill it. Many regions list sealants for children who have reached the age as a free service, so ask the local CDC or the community health centre. This section did not find a national document whose wording can be checked word for word, so it does not state the coverage or the age band. Dental caries is also a focus item of the student health examination, see item 7 in this section. - ### 14. If your child loves games, first check whether sleep, homework and time outdoors have been squeezed out; do not only watch how long the play goes on for - Cost: No money. Once a week, note what time your child goes to bed, whether the homework was handed in, and whether there was time outside. @@ -129,6 +120,7 @@ The first eight items and the last two measure death and the physical harm that - Evidence grade: B - Sources: Stevens MW, Dorstyn D, Delfabbro PH, King DL. (2021). Global prevalence of gaming disorder: A systematic review and meta-analysis. Australian and New Zealand Journal of Psychiatry, 55(6), 553–568. ；Vuorre M, Johannes N, Magnusson K, Przybylski AK. (2022). Time spent playing video games is unlikely to impact well-being. Royal Society Open Science, 9(7), 220411. ；National Press and Publication Administration (国家新闻出版署) (2021). Notice on Further Strict Management to Effectively Prevent Minors from Becoming Addicted to Online Games《关于进一步严格管理切实防止未成年人沉迷网络游戏的通知》（国新出发〔2021〕14 号）. - Notes: The grade is B because the prevalence changes too much with the questionnaire, and the well-being study only followed people for six weeks. The commonest cost of games is that they squeeze out sleep and time outdoors. The written requirement on sleep is in item 10 in this section (sleep, homework, physical education), outdoors is in item 4 in this section (2 hours outdoors a day), and screen time is in item 5 in this section (no more than 1 hour a day for anything other than study). Top-ups and refunds are in item 9 in section 5 (a child topping up or tipping on a phone). When school marks, sleep and the relationship with the family all fall apart together, and advice does not stop it, take the child to psychiatry or to child and adolescent psychology. The beneficiary here is the child. + ### 15. If your child says they like the same sex, do not scold them, do not turn them out of the house, and do not send them for "correction": the attitude at home is related to whether they will attempt suicide - Cost: No money. The hard part is that the parent has to work through their own feelings first, and not say anything harsh at the time. @@ -136,4 +128,22 @@ The first eight items and the last two measure death and the physical harm that - Benefit: A survey in China covered 506 secondary schools in 7 provinces, 150822 pupils from the first year of junior high to the third year of senior high. Among those who had attempted suicide in the past year, boys who liked only the opposite sex were 2.2%, boys who liked only the same sex 6.9%, and boys who liked both 12.2%. For girls the figures are 3.1%, 8.9% and 10.9%. That works out at about 3 to 5.5 times the figure for classmates who like only the opposite sex. The researchers took out age, academic pressure, family income and being bullied. After that, the OR for those two groups of boys is 3.13 (95% CI 2.28–4.28) and 3.83 (2.85–5.14), that is two to three times higher. A meta-analysis of 35 studies works out an OR of 3.50 (2.98–4.12) for attempted suicide among sexual minority adolescents. A US study interviewed 224 homosexual and bisexual young people aged 21 to 25 and asked them to recall the reaction at home during adolescence. Compared with peers whose families hardly rejected them, the OR for having attempted suicide where family rejection was high is 8.4. In a separate study, where parents themselves tried to change their child's sexual orientation, the OR for the child having attempted suicide is 3.08 (1.39–6.83). Where parents also took the child to see a therapist or a religious figure, it is 5.07 (2.38–10.79) (China / several countries / United States) - Evidence grade: B - Sources: Huang Y, Li P, Guo L, et al. (2018). Sexual minority status and suicidal behaviour among Chinese adolescents: a nationally representative cross-sectional study. BMJ Open, 8(8), e020969. ；di Giacomo E, Krausz M, Colmegna F, Aspesi F, Clerici M. (2018). Estimating the Risk of Attempted Suicide Among Sexual Minority Youths: A Systematic Review and Meta-analysis. JAMA Pediatrics, 172(12), 1145–1152. ；Ryan C, Huebner D, Diaz RM, Sanchez J. (2009). Family rejection as a predictor of negative health outcomes in white and Latino lesbian, gay, and bisexual young adults. Pediatrics, 123(1), 346–352. ；Ryan C, Toomey RB, Diaz RM, Russell ST. (2020). Parent-Initiated Sexual Orientation Change Efforts With LGBT Adolescents: Implications for Young Adult Mental Health and Adjustment. Journal of Homosexuality, 67(2), 159–173. ；Standing Committee of the National People's Congress (2024 revision). Law on the Protection of Minors（未成年人保护法）（article seventeen）. ；Standing Committee of the National People's Congress (2015). Anti-Domestic Violence Law（反家庭暴力法）（article two）. -- Notes: The grade is B because all of these are one-off surveys, and the degree of rejection at home is recalled in adulthood, so they can show an association and not who caused what. Parents do not have to accept anything straight away; first do not do anything that harms the child. In law, parents may not abuse or abandon a child, and may not commit domestic violence against a child. Regular verbal abuse and intimidation count as domestic violence as well. If the child's mood is low and they say they do not want to live, call 12356 first, see item 25 in section 1 (call 12356 when there are thoughts of suicide). You can also go to psychiatry following item 8 in this section (one depression screen between 12 and 18 years old). For why "correction" is not to be touched, see item 28 in section 6 (do not carry out sexual orientation conversion). The beneficiaries here are the child and you. \ No newline at end of file +- Notes: The grade is B because all of these are one-off surveys, and the degree of rejection at home is recalled in adulthood, so they can show an association and not who caused what. Parents do not have to accept anything straight away; first do not do anything that harms the child. In law, parents may not abuse or abandon a child, and may not commit domestic violence against a child. Regular verbal abuse and intimidation count as domestic violence as well. If the child's mood is low and they say they do not want to live, call 12356 first, see item 25 in section 1 (call 12356 when there are thoughts of suicide). You can also go to psychiatry following item 8 in this section (one depression screen between 12 and 18 years old). For why "correction" is not to be touched, see item 28 in section 6 (do not carry out sexual orientation conversion). The beneficiaries here are the child and you. +### 16. Do not hit your child, and do not shout at them or call them stupid: children who are often hit and shouted at have more behaviour and emotional problems + +- Cost: No money. The hard part is stopping your hand and your mouth at the moment the anger comes, and then switching to a way of managing the child that does not involve hitting or shouting. What to switch to is in item 17 in this section (a parenting course that teaches specific techniques). +- In plain language: Children who are hit often are more aggressive and have more psychological problems, are more distant from their parents, and are hit harder more often. Whether hitting makes a child obedient on the spot is not something the evidence settles. Children who at 13 are often shouted at or called stupid get into more trouble and are more depressed afterwards, and how warm the parents are the rest of the time does not cancel that out. +- Benefit: Start with hitting. One meta-analysis counts only one form of it, "an open hand on the buttocks or the limbs". It pools 75 studies, 111 effect sizes and 160927 children. Of the 17 outcomes, 13 are clearly connected with being hit, and the direction is bad in every one of them. d is the size of the effect: 0.2 counts as small and 0.5 as moderate. Aggression in childhood d = 0.37 (95% CI 0.13–0.61, credible range). Externalising problems 0.41, which means hitting people and breaking rules. Internalising problems 0.24, which means anxiety and depression. Mental health problems 0.53, a poor relationship with the parents 0.51, low self-esteem 0.15, poor cognitive ability 0.17. Physical abuse inflicted by the parents 0.64. Antisocial behaviour in adulthood 0.36, mental health problems 0.24. The 111 effect sizes put together give d = 0.33 (0.29–0.38). Whether the child obeys on the spot is not clearly connected with being hit (d = 0.14, −0.19–0.47), and that one rests on only 5 effect sizes. A second review looks only at studies that followed children forwards, 69 of them. Physical punishment consistently predicts more behaviour problems later, and it was not connected with any long-term good outcome. 7 of those studies looked at how often the hitting happened, and 5 found that the more the hitting, the worse the problems. Now shouting. One study in the United States followed 976 families with both parents at home. When the child was 13 and the parents often shouted, swore or called the child stupid or lazy, getting into trouble and depressive symptoms both increased between the ages of 13 and 14. How warm the parents were the rest of the time does not cancel that association. The reverse holds as well: children getting into more trouble at 13 were shouted at more afterwards. In law, parents may not commit domestic violence against a child. Article two of the Anti-Domestic Violence Law lists the forms domestic violence takes: beating, tying up, maiming, restricting personal freedom, and persistent verbal abuse and intimidation. Article twelve of that Law requires a guardian to carry out family education in a civil manner. Where parents infringe a child's rights, the residents' committee and the village committee must advise them and stop them, and where the circumstances are serious they must report it to the police. Where the police, the procuratorate or the court find out, they must admonish the parents and may order them to take family education guidance (nationwide) +- Evidence grade: A +- Sources: Gershoff ET, Grogan-Kaylor A. (2016). Spanking and child outcomes: Old controversies and new meta-analyses. Journal of Family Psychology, 30(4), 453–469. ; Heilmann A, Mehay A, Watt RG, et al. (2021). Physical punishment and child outcomes: a narrative review of prospective studies. Lancet, 398(10297), 355–364. ; Wang MT, Kenny S. (2014). Longitudinal links between fathers' and mothers' harsh verbal discipline and adolescents' conduct problems and depressive symptoms. Child Development, 85(3), 908–923. ; Ferguson CJ. (2013). Spanking, corporal punishment and negative long-term outcomes: a meta-analytic review of longitudinal studies. Clinical Psychology Review, 33(1), 196–208. ; Larzelere RE, Kuhn BR. (2005). Comparing child outcomes of physical punishment and alternative disciplinary tactics: a meta-analysis. Clinical Child and Family Psychology Review, 8(1), 1–37. ; Standing Committee of the National People's Congress (2015). Anti-Domestic Violence Law（反家庭暴力法）（articles two and twelve）. ; Standing Committee of the National People's Congress (2024 revision). Law on the Protection of Minors（未成年人保护法）（articles eleven, seventeen and one hundred and eighteen）. ; Standing Committee of the National People's Congress (2021). Law on Promoting Family Education（家庭教育促进法）（articles twenty-three and fifty-three）. +- Notes: Disputed. Nearly all of these studies only followed records and had no control group, so they cannot say who caused what. In the spanking meta-analysis, 72% of the studies only looked at whether two things appeared together, or relied on recall afterwards. Another meta-analysis included only studies that followed children forwards; after taking out the child's earlier behaviour, the link between spanking and externalising problems was only 0.07, and with internalising problems only 0.10. The authors of that one think the effect is small. A third meta-analysis found that "conditional spanking" does better than 10 of 13 other methods of discipline at reducing disobedience and antisocial behaviour. It also found that hitting too hard, or making hitting the main method, does worse than other methods. So what the other side can show is this: a little spanking, done conditionally, may not be worse than other methods. Heavy spanking and constant hitting and shouting, both sides agree, are worse. An occasional slap and a raised voice are not stated by the statute to be domestic violence; what it names is beating and persistent verbal abuse and intimidation. The main beneficiary is the child. If somebody else in the household is hitting the child, how to call the police and how to apply for a protection order: see item 43 in section 8 (personal safety protection order). Anyone who sees a child being hit over a long period in another family has the right to step in and stop it, or to report it to the police, the civil affairs department or the education department. For what is going on in a stranger's home, calling 110 or telling the residents' committee is enough; do not go to the house yourself and take it on. + + +### 17. If the child is hard to manage and you cannot stop yourself hitting or shouting, go to a parenting course that teaches specific techniques + +- Cost: The parenting schools run by schools, kindergartens and communities are public-benefit and charge nothing. The courses with evidence behind them run weekly or every other week, one to two hours each, from four weeks to twenty-four weeks. You have to work through the exercises at home. The hard part is finishing the course and practising every day. +- In plain language: In trials where assignment was by lot, a course that teaches parents specific techniques reduced the child's hitting and rule-breaking, and it improved the parents' mood. Parents also hit and shouted at their children less; the effect was largest just after the last session, was still there one or two years later, but was much smaller. Shanghai has run this kind of course too, and it worked. +- Benefit: Cochrane pooled 13 trials, 10 of them assigned by lot, with 1078 parents. The children were 3 to 12 years old and all of them had behaviour problems. Among the parents who took the course, the children's behaviour problems went down. Rated by the parents themselves the effect is SMD −0.53 (95% CI −0.72 to −0.34, credible range), and rated independently by others −0.44 (−0.77 to −0.11). SMD and d are the same ruler, and 0.5 counts as moderate. The parents' own mental health improved, −0.36 (−0.52 to −0.20). Negative or harsh discipline became less frequent: −0.77 (−0.96 to −0.59) rated by the parents, −0.42 (−0.67 to −0.16) rated by others. The authors describe these as short-term effects. A second meta-analysis screened out 346 randomised trials, 60 of which measured behaviour such as hitting and shouting at children, across 22 countries. Hitting and shouting fell when the course had just ended, d = −0.46 (−0.59 to −0.33). After 1 to 6 months it was −0.24 (−0.37 to −0.11). After 7 to 24 months it was −0.18 (−0.34 to −0.02), and the later the measurement, the smaller it is. Most of these figures are reported by the parents themselves. There are trials in China as well. In Shanghai, 81 parents were assigned by lot and some took the Triple P group course. After the course, the children's adjustment problems and the parents' discipline and mood were all better than in the parents still waiting for their course, and still were six months later. In the countryside, 191 adults looking after children at one kindergarten took part; some worked through an online course on their own and then joined two group discussions. Two weeks after it, the child abuse they reported was about 33% lower than among those who were only handed a book to read (IRR 0.67). Three months later, the children's behaviour problems had fallen as well (d = −0.40). The Law on Promoting Family Education requires parents to learn about family education on their own initiative. Primary and secondary schools and kindergartens may run parenting schools and organise public-benefit family education guidance at regular intervals. Residents' committees and village committees may run community parenting schools. Governments at provincial level and above have to run public-benefit online parenting schools and service hotlines. Institutions giving family education guidance may not organise profit-making training (nationwide, in force from 2022-1-1) +- Evidence grade: A +- Sources: Furlong M, McGilloway S, Bywater T, Hutchings J, Smith SM, Donnelly M. (2012). Behavioural and cognitive-behavioural group-based parenting programmes for early-onset conduct problems in children aged 3 to 12 years. Cochrane Database of Systematic Reviews, (2), CD008225. ; Backhaus S, Leijten P, Jochim J, Melendez-Torres GJ, Gardner F. (2023). Effects over time of parenting interventions to reduce physical and emotional violence against children: a systematic review and meta-analysis. eClinicalMedicine, 60, 102003. ; Guo M, Morawska A, Sanders MR. (2016). A Randomized Controlled Trial of Group Triple P With Chinese Parents in Mainland China. Behavior Modification, 40(6), 825–851. ; Wang W, Zhang H. (2026). A self-directed parenting program to reduce the risk of child maltreatment in rural China: A cluster randomized controlled trial. Child Abuse & Neglect, 173, 107937. ; Standing Committee of the National People's Congress (2021). Law on Promoting Family Education（家庭教育促进法）（articles eighteen, twenty-five, thirty-one, thirty-eight and forty）. +- Notes: The grade is A because there are pooled figures from Cochrane and from several randomised trials. The courses in the trials have a few things in common. The person leading the course teaches specific techniques, such as playing with the child, praising, rewarding and disciplining. The parents practise and rehearse in the class, and there is homework at home. Courses that only hand out material, or only discuss, were left out, so the evidence does not cover a single lecture. What the public-benefit parenting schools here teach, and how they teach it, differs from place to place and it may not be this kind of course. When you pick a course, look at whether it runs over several weeks and makes you practise specific techniques. Names you will often see are Triple P, Incredible Years and Parent-Child Interaction Therapy (PCIT). This book does not recommend particular institutions or paid products. The effect of less hitting and shouting gets smaller after a year or two, so you have to keep practising after the course ends. The beneficiaries are the child and you. If the child already has serious emotional problems, take them to psychiatry or to a child psychology clinic, see item 8 in this section (depression screening). diff --git "a/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" "b/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" index 2859191f..14b66ba2 100644 --- "a/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" +++ "b/en/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" @@ -16,7 +16,6 @@ Which path you take is your own business. This section offers two things only: w - Evidence grade: A - Sources: Standing Committee of the National People's Congress (2021). Military Service Law（兵役法）（article twenty）. （published by the Ministry of National Defense）；Central Organization Department of the CPC Central Committee (2019). Provisions on the Recruitment of Civil Servants《公务员录用规定》（articles eighteen and nineteen）. （published on the Communist Party member website）；Ministry of Human Resources and Social Security and Ministry of Emergency Management (2021). Measures for the Recruitment of Firefighters into the National Comprehensive Fire and Rescue Force《国家综合性消防救援队伍消防员招录办法》（人社部发〔2021〕58 号，articles five and six）. ；State Council and Central Military Commission (2022). Regulations on Civilian Personnel of the People's Liberation Army of China《中国人民解放军文职人员条例》（second revision, Order No. 757, articles nine and nineteen）. （National Database of Laws and Regulations）；State Council (2014). Regulations on the Administration of Personnel of Public Institutions《事业单位人事管理条例》（国务院令第 652 号，article eight）. ；State Council (2022). Regulations on the Promotion of the Development of Individual Businesses《促进个体工商户发展条例》（国务院令第 755 号，article two）. ；State Council (1988, revised 2014). Interim Regulations on Higher Education Self-Study Examinations《高等教育自学考试暂行条例》（articles three and twenty）. （National Database of Laws and Regulations）；Standing Committee of the National People's Congress (2018). Compulsory Education Law（义务教育法）（article fourteen）. （published by the Ministry of Education）；General Office of the State Council (2024). Notice on Forwarding the Implementation Measures for Publicly Funded Education of Students on the Bridging Programme for Undergraduates and Postgraduates at Normal Universities Directly under the Ministry of Education from the Ministry of Education and Other Departments《教育部直属师范大学本研衔接师范生公费教育实施办法》（国办发〔2024〕27 号，articles two and five）. ；National Health Commission and 7 other departments (2019). Notice on Employment Placement and Performance Management of Rural Order-Directed Free-Trained Medical Students《关于做好农村订单定向免费培养医学生就业安置和履约管理工作的通知》. ；State Council (2012). Regulations on the Administration of Overseas Labour Cooperation《对外劳务合作管理条例》（国务院令第 620 号，articles five, seven and eight）. - Notes: Having reached the threshold, you still have to get through the political criteria, the medical examination and the political assessment. These have no published scoring standard and cannot be found in the provisions. The entry points for secondary vocational schools, technical schools and integrated vocational education are in item 5 in section 23, and the financial help available when the family cannot afford it is in item 4 in section 23. An individual business does not look at your education, but opening a shop is not easy. The capital, the personal guarantee and the cost of closing down are in section 12. A personal guarantee means that if the company cannot repay, you have to repay it yourself. Whether the money to start with can be borrowed as a loan where the state pays part of the interest is in item 16 in this section. The two of publicly funded teacher training students and contract medical students are settled on the day you fill in your choices for the national entrance examination, and what they cost is written out in item 13 in this section. For working abroad the threshold is not on you, it is on whether the company is qualified; how to check that is in item 14 in this section. Which path is better value is not compared here. - ### 2. Register for military service before 31 October in the year you turn eighteen; a conscript serves two years of active duty - Cost: No money. Registration is filled in once online and takes a few minutes. If you really do serve, what it costs is two years of full-time work, and you cannot do anything else at the same time in those two years. @@ -25,7 +24,6 @@ Which path you take is your own business. This section offers two things only: w - Evidence grade: A - Sources: General Office of the State Council and General Office of the Central Military Commission (2025). Provisions on Military Service Registration《兵役登记工作规定》（国办发〔2025〕41 号）. ；Standing Committee of the National People's Congress (2021). Military Service Law（兵役法）（2021 revision, articles twenty, twenty-five, twenty-six, fifty-four and fifty-seven）. （published by the Ministry of National Defense）；State Council and Central Military Commission (2023). Regulations on Military Conscription《征兵工作条例》（国务院、中央军委令第 759 号，article five）. - Notes: Registration and being called up are two different things, and the cost differs by a long way. The set of penalties that stops you leaving the country within two years only applies to people who back out after being called up, see item 3 in this section (refusing military service). Women are not required to register initially, and are called up according to the needs of the armed forces. Soldiers moving on to a military academy, and good soldiers being promoted to officers, have their conditions written into the internal rules of the armed forces, which cannot be looked up from outside. The Military Service Law says only two things: active-duty officers are selected and recruited from among 「表现优秀的现役士兵」, serving soldiers with outstanding performance, and the military academies are not subject to the call-up age limit when they recruit students. This book does not give the specific score lines and the quotas. - ### 3. If you refuse to serve after being called up, you cannot leave the country or go back to school within two years, and you cannot join the civil service or a state-owned enterprise - Cost: No money. The cost is that before you sign up you have to settle in your own mind whether you are going. @@ -34,14 +32,15 @@ Which path you take is your own business. This section offers two things only: w - Evidence grade: A - Sources: Standing Committee of the National People's Congress (2021). Military Service Law（兵役法）（2021 revision, articles fifty-seven and fifty-eight）. （published by the Ministry of National Defense） - Notes: What is counted here is personal freedom and the right to be employed, not money. This list of sanctions applies to only two kinds of person. One is "a citizen called up who refuses or evades being called up for active service". The other is someone who, after enlisting, is struck off the rolls, has military rank stripped, or is held criminally responsible. Merely failing to register for military service is not in it; military service registration is item 2 in this section. The fine presupposes that the person "refuses to correct it", and it is the county-level people's government that determines it and imposes it. How much the fine is varies from place to place and this book does not give it. Whether you go is yours to decide, but decide it before you sign up. -### 4. The two solidest things a university student gets from serving: tuition reimbursed up to 20000 yuan a year, and the student place kept for two years with a separate plan for the postgraduate entrance examination + +### 4. The two most solid benefits for university students who enlist: tuition reimbursed up to 20000 yuan a year, and the place kept for two years with a special plan for the postgraduate entrance examination -- Cost: No money. After being discharged you hand in the documents the school asks for and apply, and it takes one or two trips. -- In plain language: Students on campus, graduating students and new students who go to serve have their tuition paid by the state for the period of service. For undergraduate and junior college students it is up to 20000 yuan each per year, and for postgraduates up to 25000 yuan each per year, so over two years that is up to forty or fifty thousand yuan. Your student place is kept, and within two years of being discharged you may enrol or go back. The postgraduate entrance examination has a separate plan that admits only discharged student soldiers. -- Benefit: A 2024 notice from the Ministry of Finance and three other departments says that students at higher education institutions going to serve military service have three kinds of support: reimbursement of tuition, repayment on behalf of the student of state student loans used for tuition, and a reduction of tuition. Repaying the loan on the student's behalf means the state pays that student loan off. The standards for all three are 「相应调整为本专科学生每人每年最高不超过 20000 元、研究生每人每年最高不超过 25000 元」, that is, adjusted so that undergraduate and junior college students get up to 20000 yuan each per year and postgraduates up to 25000 yuan each per year. The new standards apply from the autumn term of 2024. Exactly how much is reimbursed depends on this line: 「按学生实际缴纳的学费或获得的国家助学贷款（包括本金及其全部偿还之前产生的利息）两者金额较高者执行」, that is, it is paid at whichever is larger, the tuition the student actually paid or the state student loan obtained, including the principal and all the interest accrued before it is fully repaid. There is a further line: 「超出标准部分不予补偿、代偿或减免」, the part above the standard is not reimbursed, not repaid on the student's behalf and not reduced. For the student place, look at the Law on the Protection of Veterans《退役军人保障法》. It states: 「现役军人入伍前已被普通高等学校录取或者是正在普通高等学校就学的学生，服现役期间保留入学资格或者学籍，退役后两年内允许入学或者复学，可以按照国家有关规定转入本校其他专业学习」, that is, a student who has been admitted by an ordinary higher education institution before enlisting, or who is studying at one, keeps their admission place or student place during active service, and within two years of being discharged may enrol or go back, and may transfer to another programme in the same institution in accordance with the provisions of the state. The same law also states: 「高等学校根据国家统筹安排，可以通过单列计划、单独招生等方式招考退役军人」, that is, higher education institutions, in accordance with the overall arrangements of the state, may recruit veterans through separately allocated plans and separate enrolment. For the postgraduate entrance examination, look at the 2026 notice from the General Office of the Ministry of Education. The plan is called the "Plan for Discharged Student Soldiers"「退役大学生士兵计划」. The notice states: 「在全国硕士研究生招生计划总规模内单列下达、专项使用，不得调整为普通计划或其他专项计划。招生单位自主确定……初试成绩要求（一般不高于本单位相应学科专业普通计划考生初试成绩要求）……同等条件下服役时间长的考生优先录取」, that is, it is allocated separately within the overall national plan for postgraduate enrolment and used for that purpose only, and may not be converted into the ordinary plan or any other special plan. The enrolling institution decides for itself the requirements for the first-round examination score (generally no higher than the first-round score required of candidates in the ordinary plan for the same discipline). Under equal conditions, candidates who served longer are admitted first. Allocated separately and used for that purpose only means this part of the quota is set aside and can be used only to recruit discharged student soldiers. In 2026 this plan is carried by 532 ordinary higher education institutions (nationwide) +- Cost: No money. After being discharged you hand in the papers the school asks for and apply, which means one or two trips. +- In plain language: If students still at university, recent graduates and new entrants enlist, the state pays the tuition for those years of study, reimbursed against what you actually paid: up to 20000 yuan a year for undergraduate and junior-college students, and up to 25000 yuan a year for postgraduate students. Where the tuition was already low, you only get back what you actually paid. Your place is kept for two years. For the postgraduate entrance examination there is a separate special plan that only takes retired university-student soldiers. +- Benefit: A 2024 notice from the Ministry of Finance and three other departments says that college students going into military service get three things: a tuition reimbursement, repayment of the national student loan used for tuition, and a tuition reduction. Repayment of the loan means the state pays that loan off for you. The standards for all three are 「adjusted accordingly to a maximum of no more than 20000 yuan per undergraduate or junior-college student per year, and no more than 25000 yuan per postgraduate student per year」. The new standards apply from the autumn term of 2024. How much you actually get depends on this sentence: 「the higher of the tuition the student actually pays and the national student loan received (including the principal and all the interest accrued before it is repaid in full)」. The sentence after it is 「any part above the standard shall not be reimbursed, repaid or reduced」. So what you get each year is the smaller of 「the higher of the actual tuition and the student loan」 and the cap. If your tuition is 5000 yuan a year, you get 5000 yuan back. How many years you get is counted by the years of study, not by the years of service. The 2019 measures on the administration of student aid funds say: 「the years of study already completed before enlistment are the years for tuition reimbursement or repayment of the national student loan; the remaining required years of study completed after being discharged and returning are the years for tuition reduction」. Graduates who enlist get the tuition for the whole length of their course. On keeping your place, see the Law on the Protection of Veterans《退役军人保障法》. It says: 「students already admitted to an ordinary higher education institution before enlistment, or already studying in one, keep their admission qualification or their student status during their service, and are allowed to enrol or return within two years of discharge, and may transfer to another subject in the same institution according to the state's rules」. The same law also says: 「higher education institutions may, in line with the state's overall arrangements, recruit veterans by separate plans and separate admissions」. For the postgraduate entrance examination, see the 2026 notice from the General Office of the Ministry of Education. The plan is called the 「special plan for retired university-student soldiers」. The notice says: 「issued separately and used exclusively within the total scale of the national postgraduate admissions plan; it shall not be turned into the ordinary plan or any other special plan. The admitting institution decides for itself … its requirements for the initial examination results (generally no higher than those for candidates under the ordinary plan for the corresponding subject) … candidates with longer service are admitted first where the conditions are equal」. Issued separately and used exclusively means this slice of the places is set aside on its own and can only be used to recruit retired university-student soldiers. In 2026 the plan is run by 532 ordinary higher education institutions (nationwide) - Evidence grade: A -- Sources: Ministry of Finance, Ministry of Education, People's Bank of China and National Financial Regulatory Administration (2024). Notice on Adjusting and Improving Relevant Policies on State Student Loans（关于调整完善国家助学贷款有关政策的通知）（财教〔2024〕188 号）. ；Ministry of Finance and four other departments (2019). Measures for the Administration of Student Financial Aid Funds《学生资助资金管理办法》（财科教〔2019〕19 号，article five）. ；Standing Committee of the National People's Congress (2020). Law on the Protection of Veterans《退役军人保障法》（articles thirty-four and thirty-five）. （published by the Ministry of National Defense）；General Office of the Ministry of Education (2025). Notice on Doing Well the 2026 Enrolment Work under the Separate Postgraduate Plan for Discharged Student Soldiers《关于做好 2026 年退役大学生士兵专项硕士研究生招生计划招生工作的通知》（教学厅〔2025〕8 号）. ；Ministry of Veterans Affairs (2022). 23 points on education and training policy for discharged soldiers「退役士兵教育培训政策 23 条」: 「从2022年招生起，高职（专科）毕业生及在校生（含高校新生）应征入伍，退役后完成高职（专科）学业的，申请专升本，免于参加文化课考试」(based on 教学厅〔2021〕8 号). ；Ministry of Education (2026). 2027 National Postgraduate Enrolment Work Management Provisions（2027 年全国硕士研究生招生工作管理规定）（教学〔2026〕2 号，articles twenty-two, twenty-four and twenty-five），published on the China postgraduate admissions website. ；Ministry of Education (2024). 2025 National Postgraduate Enrolment Work Management Provisions（2025 年全国硕士研究生招生工作管理规定）（教学〔2024〕4 号，article twenty-two）. -- Notes: There are two more routes. First, exemption from the cultural examination for the upgrade to a bachelor's degree: a junior college student goes to serve, finishes the junior college course after being discharged, and can apply to be upgraded to a bachelor's degree without the examination, with the school testing practical vocational skill and admitting on merit. This has applied from the 2022 enrolment onwards, and the specific rules are in your province's notice for that year. Second, 10 points added to the first-round examination for the postgraduate entrance examination: a discharged student soldier who sits the postgraduate entrance examination within 3 years of being discharged gets 10 points added to the total first-round score. If you have applied under the separate plan you do not get the points; the two cannot both be enjoyed. Those who have been awarded the second-class merit or above may apply to be admitted to postgraduate study without the first-round examination. The list of institutions running the separate plan changes every year; in 2025 it was 523. The tuition standards have been adjusted twice: in 2019 they were 8000/12000, in the autumn of 2023 they became 16000/20000, and in the autumn of 2024 they became 20000/25000. The summary the Ministry of Veterans Affairs published in 2022 still has the earliest version in it; do not copy it. +- Sources: Ministry of Finance, Ministry of Education, People's Bank of China and National Financial Regulatory Administration (财政部、教育部、中国人民银行、金融监管总局) (2024). Notice on Adjusting and Improving the Relevant Policies on the National Student Loan, 关于调整完善国家助学贷款有关政策的通知（财教〔2024〕188 号）. ; Ministry of Finance and four other departments (财政部等五部门) (2019). Measures for the Administration of Student Aid Funds, 学生资助资金管理办法（财科教〔2019〕19 号，article 5; the years covered for enlistment aid: 「入伍前已完成规定的修业年限，即为学费补偿或国家助学贷款代偿的年限；退役复学后接续完成规定的剩余修业年限，即为学费减免的年限」; for graduates, 「学费补偿或国家助学贷款代偿的年限，分别按照国家规定的相应学制计算」）. ; Standing Committee of the National People's Congress (全国人大常委会) (2020). Law on the Protection of Veterans《退役军人保障法》（articles 34 and 35）. （published by the Ministry of National Defense）; General Office of the Ministry of Education (教育部办公厅) (2025). Notice on the Work of Enrolment under the 2026 Special Plan of Postgraduate Study for Retired University-Student Soldiers, 关于做好 2026 年退役大学生士兵专项硕士研究生招生计划招生工作的通知（教学厅〔2025〕8 号）. ; Ministry of Veterans Affairs (退役军人事务部) (2022). Policy item 23 on the education and training of retired soldiers: 「从2022年招生起，高职（专科）毕业生及在校生（含高校新生）应征入伍，退役后完成高职（专科）学业的，申请专升本，免于参加文化课考试」（under 教学厅〔2021〕8 号）. ; Ministry of Education (教育部) (2026). 2027 National Postgraduate Admissions Work Administration Provisions, 2027 年全国硕士研究生招生工作管理规定（教学〔2026〕2 号，articles 22, 24 and 25）, published on the postgraduate admissions website. ; Ministry of Education (教育部) (2024). 2025 National Postgraduate Admissions Work Administration Provisions, 2025 年全国硕士研究生招生工作管理规定（教学〔2024〕4 号，article 22）. +- Notes: There are two more routes. First, moving up to a bachelor's course without the culture exam: a junior-college student who enlists and finishes the junior-college course after discharge can apply to move up without sitting the exam, and the school takes the best of them after testing their vocational skills. This applies from the 2022 intake; the exact method is in your own province's notice for that year. Second, 10 extra points on the initial examination for the postgraduate entrance: a retired university-student soldier who takes the postgraduate entrance examination within 3 years of discharge gets 10 points added to the total for the initial examination. Once you have applied under the special plan there are no extra points; the two cannot both be taken. Those who have been awarded a second-class merit or above can apply to skip the initial examination and go straight to postgraduate study. The list of institutions running the special plan changes every year; in 2025 it was 523. The tuition standard has been adjusted twice: in 2019 it was 8000/12000, from autumn 2023 it became 16000/20000, and from autumn 2024 it became 20000/25000. The 2022 summary from the Ministry of Veterans Affairs still shows the earliest version, so do not copy it. ### 5. After being discharged the government arranges work in only six situations; those who take up self-employment must report within 30 days @@ -68,7 +67,6 @@ Which path you take is your own business. This section offers two things only: w - Evidence grade: A - Sources: Central Organization Department of the CPC Central Committee, Ministry of Human Resources and Social Security and eight other departments (2021). Notice on Implementing the Fourth Round of the "Three Supports and One Assistance" Plan for University Graduates「关于实施第四轮高校毕业生『三支一扶』计划的通知」（人社部发〔2021〕32 号，articles seven, eight, nine, eleven, twelve and thirteen）. ；General Office of the CPC Central Committee and General Office of the State Council (2017). Opinion on Further Guiding and Encouraging University Graduates to Work at the Grassroots Level《关于进一步引导和鼓励高校毕业生到基层工作的意见》（articles five, sixteen and eighteen）. ；National Office for the Administration of the Western Plan for University Student Volunteer Service (2026). 2026—2027 Enrolment Agreement for the Western Plan for University Student Volunteer Service「2026—2027 年度大学生志愿服务西部计划招募协议书」. - Notes: The fourth round of the "Three Supports and One Assistance" plan covers 2021 to 2025. The document for the fifth round and the notice for 2026 were not found this time, so check that year's recruitment announcement before you sign up. In 2025 the central finance supported the recruitment of 34430 people. The implementation plan for the Western Plan could not be found online, and the policy section of its own website would not open. For the benefits after it ends, go by the document from the two General Offices that covers all grassroots service programmes. What you get for those two years is a fiscal subsidy, not a market wage, and that cost has to be counted together with the targeted recruitment two years later. - ### 8. Special-post teachers who finish the three-year service period, pass the assessment and want to stay should be put on the establishment in good time - Cost: No money. The cost is three years teaching in a rural compulsory education school, and before you start you have to pass the teacher qualification examination. The hard part is lasting those three years. @@ -86,7 +84,6 @@ Which path you take is your own business. This section offers two things only: w - Evidence grade: A - Sources: Ministry of Human Resources and Social Security and Ministry of Emergency Management (2021). Measures for the Recruitment of Firefighters into the National Comprehensive Fire and Rescue Force《国家综合性消防救援队伍消防员招录办法》（人社部发〔2021〕58 号，articles five, six, seven and nine）. ；State Council and Central Military Commission (2022). Regulations on Civilian Personnel of the People's Liberation Army of China《中国人民解放军文职人员条例》（second revision, Order No. 757 of the State Council and the Central Military Commission, in force 2023-1-1; articles nine, eighteen, nineteen and twenty）. （National Database of Laws and Regulations） - Notes: The Regulations on Promoting the Employment and Entrepreneurship of Veterans《退役军人就业创业促进条例》, in force from 2026-8-1, set out two more things. First, a certain proportion of firefighter recruitment has to be set aside for discharged non-commissioned officers and conscripts. Second, when recruiting police auxiliary staff, veterans are to be given priority under equal conditions. The number of places, the posts and the specific conditions for both of these routes each year are set by that year's announcement; this book only writes down the thresholds that are fixed in the rules. Firefighting is a post in a disciplined service, so weigh up the occupational risk before you go in. - ### 10. A qualification is not a one-off purchase: the self-study examination has no limit on age or on the education you already have, and the adult entrance examination and the open university can both complete it - Cost: Several hundred yuan a year in registration and course materials. The cost is long-term self-study, or going to classes in your spare time; the shortest part-time route to a bachelor's degree, studied while working, takes two and a half years. The hard part is keeping it up over a long period. @@ -103,7 +100,6 @@ Which path you take is your own business. This section offers two things only: w - Evidence grade: A - Sources: Ministry of Human Resources and Social Security and eight other departments (2021). Guiding Opinion on Safeguarding the Labour Rights and Interests of Workers in New Forms of Employment《关于维护新就业形态劳动者劳动保障权益的指导意见》（人社部发〔2021〕56 号，articles two and eight）. ；General Office of the State Council (2020). Opinion on Supporting Flexible Employment through Multiple Channels《关于支持多渠道灵活就业的意见》（国办发〔2020〕27 号）. - Notes: The original says the household registration restriction is to be "lifted"（放开）, and it leaves an exception for the very largest cities; it does not say the restriction has been abolished nationwide. How far it has actually been lifted depends on the city where you enrol; ask there. The benefit is graded large because what a break in contributions loses is the pension and the medical treatment themselves, which is worth far more than ten thousand yuan. But the cost is real too, thousands of yuan a month paid in full by yourself. For making up a break in social insurance contributions and how the years are added together, see item 18 in section 7 (a break in social insurance contributions). And separately, selling agricultural produce or daily goods at a place and time designated by the government, or using your own skill to do convenience work that lawfully requires no licence, does not need a business licence. That is written in 国办发〔2020〕27 号. - ### 12. Delivering food, driving a ride-hailing car, carrying freight in the same city: the platform pays the work-injury cover premium job by job and you pay nothing - Cost: No money; the platform pays this. What you have to do is check whether the platform you take jobs from is inside the pilot. There are two ways: look for "work injury cover" on the protection or benefits page in the app, or telephone the human resources and social security bureau of your province and ask whether that platform has reported in. @@ -129,7 +125,6 @@ Which path you take is your own business. This section offers two things only: w - Evidence grade: A - Sources: State Council (2012). Regulations on the Administration of Overseas Labour Cooperation《对外劳务合作管理条例》（国务院令第 620 号，articles five, seven, eight, nine, ten, eleven, twelve, thirteen, twenty-one, twenty-three, twenty-five, twenty-six, twenty-nine, thirty-one, thirty-seven, thirty-nine, forty and forty-three）. - Notes: The compliant procedure is this: the company first signs a written labour cooperation contract with the foreign employer, and then signs a written service contract or labour contract with you. The contract has to state the content of the work, the period, the remuneration, the working hours, the leave, social insurance and the risks to personal safety. Within 10 working days of signing, the company has to file with the commerce department in charge. The company also has to arrange training for you in skill, in safety precautions, in a foreign language and in the local law and customs, and buy you accident and injury insurance for the period of working abroad. A counterparty that will not let you see the contract, that asks for a deposit, or that hurries you out first and promises to complete the formalities later, is the kind of company to avoid. The overseas high-pay recruitment scams and the telecom fraud compounds are in item 5 in section 21; the safety reminders to check before you leave, and 12308, are in section 21, and are not repeated here. The Regulations set no education or age threshold for the person going out to work; the threshold is on the company's qualification, and on whether the country of work will give permission. The company can only sign a contract with an enterprise or organisation approved in that country, and not with an individual abroad. The reserve fund is used by the department in charge when a company defaults, and you cannot withdraw it yourself. So if something really goes wrong, complain to the commerce department in charge first; article twenty gives you that right. Sending more than 100 people to the same country or region requires accompanying management staff to be arranged and filed with the embassy or consulate. Projects with few people have no such protection, so keep your own copies of the contract and your transfer records. - ### 15. Working remotely for a foreign company from home: under the tax law this money is domestic income, you file the personal income tax yourself, and the remittance goes through as "remuneration for employment" - Cost: No money. One settlement and payment a year. The window runs from 3-1 to 6-30 of the following year. Where the remittance exceeds the annual total, you also have to have an employment contract and proof of income ready. The hard part is remembering to file this tax yourself. @@ -138,6 +133,7 @@ Which path you take is your own business. This section offers two things only: w - Evidence grade: A - Sources: State Council (2018). Implementation Regulations of the Individual Income Tax Law of the People's Republic of China（中华人民共和国个人所得税法实施条例）（国务院令第 707 号，articles three and six）. ；Standing Committee of the National People's Congress (2018). Individual Income Tax Law of the People's Republic of China（中华人民共和国个人所得税法）(seventh revision of 2018, articles six, nine, ten and eleven). ；Policies and Regulations Database of the State Taxation Administration. Individual Income Tax Law of the People's Republic of China（中华人民共和国个人所得税法）（article six）. ；People's Bank of China (2006). Measures for the Administration of Individual Foreign Exchange（个人外汇管理办法）（中国人民银行令〔2006〕第 3 号，articles seven and nine）. ；State Administration of Foreign Exchange (2007). Detailed Rules for Implementation of the Measures for the Administration of Individual Foreign Exchange（个人外汇管理办法实施细则）（汇发〔2007〕1 号，articles two and ten）. - Notes: Which category the tax is worked out under depends on what your relationship with the other side is. If you sign an employment contract and the work is arranged by them, it is wages and salaries. If you take jobs project by project and hand over results, it is remuneration for personal services. The two come out with a different tax, so work out which one you are before you file. When the money comes in, up to the equivalent of 50,000 US dollars a year can be settled at the bank on your identity card alone; for the part over that, give the employment contract and the proof of income to the bank. Do not look for someone to split the receipts with so as to get round the quota. What is described here is the case where you are in China and do not leave the country. Actually going abroad to work for a foreign employer is a different matter, see item 14 in this section (if you want to work abroad, first check whether the company has the qualification for overseas labour cooperation). Four further questions this book has not yet written about. Each of them may have a real cost, but no basis can at present be checked word for word. First, whether this income is liable to value-added tax and whether an invoice has to be issued on your behalf. Second, whether doing this for a long time requires a business licence or registration as an individual business. Third, whether you can claim rights against the foreign employer under the labour law, for example annual leave, overtime pay and severance. Fourth, the export of data and confidentiality. + ### 16. Not being in an establishment and going into business yourself: look at the guaranteed entrepreneurship loan for the starting money, up to 300,000 yuan for an individual, with the state paying half the interest - Cost: No money to ask about it. Go once to the human resources and social security department for the eligibility check, and then go once each to the guarantee institution and to the bank that handles the loan (the bank that makes the loan). The hard part is that borrowed money still has to be repaid. @@ -145,4 +141,4 @@ Which path you take is your own business. This section offers two things only: w - Benefit: The conditions for an individual applicant are in article six of the Measures for the Administration of Special Funds for the Development of Inclusive Finance（财金〔2023〕75 号）《普惠金融发展专项资金管理办法》. The first condition is that you belong to a key group for employment. The document lists the registered unemployed in towns and cities, people with employment difficulties (including people with disabilities), veterans, and people released from prison. It also lists university graduates (including former village college students and students returning from study abroad), and workers and unemployed people in enterprises cutting excess capacity. And returning migrant workers starting a business, online shopkeepers, people who have escaped poverty, and farmers starting their own businesses in the countryside. The second condition is that there may be no other loan in the name: 「除助学贷款、脱贫人口小额信贷、住房贷款、购车贷款、5 万元以下小额消费贷款（含信用卡消费）以外，申请人提交创业担保贷款申请时，本人及其配偶无其他贷款」, that is, apart from student loans, small loans for people who have escaped poverty, housing loans, car loans and small consumer loans below 50,000 yuan (including credit card spending), when an applicant submits the application for an entrepreneurship guaranteed loan, neither the applicant nor their spouse may have any other loan. Article nine: 「对符合条件的创业担保贷款，财政部门给予贷款实际利率 50% 的财政贴息。对展期、逾期的创业担保贷款，财政部门原则上不予贴息」, that is, for a qualifying entrepreneurship guaranteed loan, the finance department gives a fiscal interest subsidy of 50% of the actual interest rate of the loan. For an entrepreneurship guaranteed loan whose term is extended or that is overdue, the finance department in principle gives no interest subsidy. An interest subsidy means the state pays half of the interest. Extending the term means pushing the repayment date further out; being overdue means the due date has passed and nothing has been paid. In neither case does the state pay any more. Article ten: in principle the requirement for a counter-guarantee is removed for an entrepreneurship guaranteed loan of 100,000 yuan or less applied for by an individual, so there is no need to find somebody else or to put up property as security for the guarantee institution. Article eleven: individual entrepreneurship guaranteed loans inside the scope of grant and subsidy support are 「额度不超过 30 万元，贷款期限不超过 3 年」, that is, the amount does not exceed 300,000 yuan and the term of the loan does not exceed 3 years. Where people go into business together, the figure may be raised appropriately, up to 110% of the sum of the individual ceilings. The interest rate ceiling for grant and subsidy support comes in three bands: LPR+50BPs in the eastern region, LPR+150BPs in the central and western regions, and LPR+250BPs in areas that have escaped poverty. LPR is the loan prime rate quoted by banks, and BPs are the unit in which an interest rate is measured; the more is added, the higher the interest. Those who 「还款积极、带动就业能力强、创业项目好」, that is, repay actively, have a strong capacity to bring employment and have a good business project, may apply again after the term ends, up to 3 times in total. Article eight: the applicant first applies to the local human resources and social security department for the eligibility check, and then submits the application to the guarantee institution and to the bank handling the loan. Article twelve: with the agreement of the provincial people's government, localities may widen the scope of support, raise the ceiling and raise the proportion of the interest subsidy (nationwide) - Evidence grade: A - Sources: Ministry of Finance (2023). Notice on Issuing the Measures for the Administration of Special Funds for the Development of Inclusive Finance《普惠金融发展专项资金管理办法》（财金〔2023〕75 号，articles six, eight, nine, ten, eleven and twelve）. -- Notes: The state paying half the interest does not change the fact that "the money borrowed has to be repaid". The position taken in item 1 in section 12 is unchanged: for a business, only take money you can afford to lose, do not touch the family savings, and do not sign a personal guarantee for a company loan, and your spouse should certainly not sign it either. If you have decided to borrow, this kind of policy loan is the lowest rate you will get; do not touch consumer loans or private lending. The ceiling and the scope may be raised by each province; for how much will actually be approved, go by what the local human resources and social security department says that year. The 2020 notice from the pandemic period, with the "up to 200,000 yuan for an individual", was a temporary document; read these Measures now. Whether you open a shop by registering as an individual business or incorporate a company, the licence, the tax and getting out are in section 12. The training subsidy, the social insurance subsidy and the work placement during unemployment are in item 13 in section 7 (vocational training subsidies). \ No newline at end of file +- Notes: The state paying half the interest does not change the fact that "the money borrowed has to be repaid". The position taken in item 1 in section 12 is unchanged: for a business, only take money you can afford to lose, do not touch the family savings, and do not sign a personal guarantee for a company loan, and your spouse should certainly not sign it either. If you have decided to borrow, this kind of policy loan is the lowest rate you will get; do not touch consumer loans or private lending. The ceiling and the scope may be raised by each province; for how much will actually be approved, go by what the local human resources and social security department says that year. The 2020 notice from the pandemic period, with the "up to 200,000 yuan for an individual", was a temporary document; read these Measures now. Whether you open a shop by registering as an individual business or incorporate a company, the licence, the tax and getting out are in section 12. The training subsidy, the social insurance subsidy and the work placement during unemployment are in item 13 in section 7 (vocational training subsidies). diff --git "a/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" "b/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" index 00f24979..eb8f8495 100644 --- "a/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" +++ "b/en/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" @@ -16,7 +16,6 @@ How to reach an embassy or consulate if something happens to you while you are a - Evidence grade: A - Sources: Chinese Service Center for Scholarly Exchange（中国（教育部）留学服务中心）. Accredited Institutions Search（认证院校查询）. ；Ministry of Education Network for the Supervision and Administration of Education Involving Foreign Parties（教育部教育涉外监管信息网）. - Notes: What an agency tells you in words, that it is "recognised by the Ministry of Education", does not count; go by what you find yourself when you look it up online. The list changes: a school can be on it when you enrol and off it when you graduate, so check once a year while you are studying. Sino-foreign cooperative education programmes have a separate system of certification and registration. For those who enrolled after 2008, you can look up your own registration number on that network using your name and your identity card number. - ### 2. Watch the court case over the new F-1 rule in the United States: "at most four years, and leave within 30 days of finishing" was to take effect on 2026-09-15 and was postponed by the court the day before, so for now it is still "until you finish" - Cost: No money. Copy the programme end date on your I-20 into your calendar and set a reminder 90 days ahead. The I-20 is the certificate of eligibility for study that the school gives you. Also look at the notices from the school's international student office once a term to see whether there is any new development in the case @@ -25,7 +24,6 @@ How to reach an embassy or consulate if something happens to you while you are a - Evidence grade: A - Sources: DHS (2026). Establishing a Fixed Time Period of Admission and an Extension of Stay Procedure for Nonimmigrant Academic Students, Exchange Visitors（published 2026-07-17, to take effect 2026-09-15）. ；Presidents' Alliance on Higher Education and Immigration v. U.S. Department of Homeland Security, No. 1:26-cv-13799-FDS (D. Mass. Sept. 14, 2026), court docket. ；Yale Office of International Students & Scholars (2026-09-14). Important Update: Court Action on the D/S Rule. ；AILA (2026). One Day Before Taking Effect, Federal Court Postpones the F, J, and I Fixed Admission Period Rule. ；8 CFR 214.2(f). - Notes: This item is as at 2026-09-25. The government can appeal to the Court of Appeals for the First Circuit, and the Department of Homeland Security can also amend the rule and publish it again. So if you are studying for several years, do not assume the old rule will last. The court has set 2026-10-02 for a further hearing on how both sides have got on, and after that date look at the notice from your school once more. If the new rule does take effect later, an extension has to be applied for separately before the period ends; it does not carry on by itself. Under the old rule, the departure deadline is different if you stop studying before finishing, so ask the international student office first. The J visa is in the same rule and the details are not the same as for F-1, so look at your own visa category. - ### 3. The hours you may work are a hard limit written into the rules of all four countries, and one hour over is a breach of your visa conditions - Cost: No money. Keep a record of your own hours week by week. The hard part is that employers say "nobody checks a few extra hours", and that is not true @@ -34,7 +32,6 @@ How to reach an embassy or consulate if something happens to you while you are a - Evidence grade: A - Sources: 8 CFR 214.2(f)(9). ；UK Home Office. Immigration Rules Appendix Student（ST26.1、ST26.5）. ；IRCC. Work off campus as an international student. ；Australian Government Department of Home Affairs. Student visa (subclass 500). - Notes: Off-campus work in the United States has to be authorised first. While the I-765 application for the work permit is still being processed, you may not start work. Canada's older permit says 20 hours a week on it; as long as you still meet the conditions, 24 hours is in fact possible, and the current rules govern. The United Kingdom's prohibition on self-employment is drawn widely: taking private jobs, running an online shop and working as a freelancer all count. - ### 4. The root of the status is "studying full time": during a leave of absence, a withdrawal or the gap between schools you may not work - Cost: No money. Before you stop classes, take a leave of absence or change schools, go to the school's international student office and ask what your status is @@ -43,7 +40,6 @@ How to reach an embassy or consulate if something happens to you while you are a - Evidence grade: A - Sources: IRCC. Work off campus as an international student. ；U.S. Department of Homeland Security, Study in the States. Working in the United States. ；UK Home Office. Immigration Rules Appendix Student（ST26.1）. - Notes: If the academic pressure is heavy and you are thinking of reducing your course load, ask first: how little still counts as full time. This one bites earlier than working does. If you are expelled by the school, or the programme is terminated, the status is normally broken the same day. Do not wait for an official letter from the school before you act. - ### 5. United States: you must report your new address within 10 days of moving - Cost: No money. Submit the change of address online once; it takes ten minutes @@ -52,7 +48,6 @@ How to reach an embassy or consulate if something happens to you while you are a - Evidence grade: A - Sources: 8 CFR 265.1. ；USCIS. AR-11, Alien's Change of Address Card. - Notes: Even after you have told the immigration authorities, the SEVIS record at the school (your file of international student status) has to be updated separately. The other three countries also require you to notify the immigration authorities or the school when you move. The deadline and the method differ from country to country; follow the page of the immigration department of the country you are in. - ### 6. Before you leave and while you are studying, take a look at the Ministry of Education's study-abroad warnings - Cost: No money. Look once before you leave, and once a term after that @@ -61,6 +56,7 @@ How to reach an embassy or consulate if something happens to you while you are a - Evidence grade: A - Sources: Ministry of Education Network for the Supervision and Administration of Education Involving Foreign Parties（教育部教育涉外监管信息网）. Warning information（预警信息）. ；Ministry of Education issues the fourth study-abroad warning of 2025（教育部发布 2025 年第 4 号留学预警）. ；Warning item one（第 1 号）. - Notes: A warning is not a ban, and it does not mean you cannot go; it gives you an official basis for weighing the risk there. The list of warnings changes with the situation, and this section does not keep it up to date over time; go by the page current at the time on the website. Personal safety abroad and consular protection, see section 21. + ### 7. Australia: OSHC has to cover the whole stay with no gap in the middle, and without proof of insurance you may be refused entry - Cost: Several thousand yuan up to more than ten thousand yuan. Buy it in one go for the length of the visa; if you are there several years, buy several years' cover @@ -69,7 +65,6 @@ How to reach an embassy or consulate if something happens to you while you are a - Evidence grade: A - Sources: Australian Government Department of Home Affairs. Student visa (subclass 500). - Notes: Where the school arranges it, ask for the name of the insurer, the start and end dates of the policy and the policy number; these go into the visa application. Buying too short a period causes trouble with the renewal, so it is better to buy a little past the date you graduate. - ### 8. United Kingdom: besides the visa fee there is a health surcharge charged by the year; a student pays 776 pounds a year, paid in full at the time of application - Cost: Visa fee 558 pounds. Health surcharge 776 pounds a year. Both are paid in full at the time of application @@ -78,7 +73,6 @@ How to reach an embassy or consulate if something happens to you while you are a - Evidence grade: A - Sources: UK Government. Student visa. ；UK Government. Pay for UK healthcare as part of your immigration application. - Notes: Both the amounts and the exchange rate change, so work from the current figures on the two pages in the sources. A degree course, and anyone aged 18 or over, is normally allowed at most 5 years; below degree level it is 2 years. Use that number of years to work out how much surcharge to allow for. - ### 9. Leave time for the certification before you come back: certification by the Service Center takes 10 to 20 working days - Cost: Apply and pay online. Preparing the documents plus the assessment and the review, counted in working days, comes to around a month in all @@ -87,7 +81,6 @@ How to reach an embassy or consulate if something happens to you while you are a - Evidence grade: A - Sources: Chinese Service Center for Scholarly Exchange（中国（教育部）留学服务中心）. Online Service Hall · Academic Qualification Certification（网上服务大厅·学历学位认证）. - Notes: The documents you have to hand in are these: the diploma, your passport or travel permit, your residence card or visa endorsement, a photograph of your identity document, and a statement of authorisation. Entry and exit records for the period you were abroad are pulled by the system itself, so do not throw away the visa pages you were carrying while you were abroad. The Service Center also issues announcements on its website saying which certificates have been invalidated and whose applications for certification have been suspended. Before you start, check whether your own school is one of them. - ### 10. More schools every year are put under "enhanced review" of certification, and the route of buying a diploma no longer works - Cost: No money. Before you sign up, search the name of the school once in the Service Center's announcements @@ -95,4 +88,4 @@ How to reach an embassy or consulate if something happens to you while you are a - Benefit: The Chinese Service Center for Scholarly Exchange has been issuing announcements applying enhanced certification review to the academic qualification certification of some foreign institutions (a separate extra check on the diplomas of these schools). The most recent is number nine (2025-10-28). The Service Center has also issued announcements suspending the acceptance of certification applications for individual overseas institutions, and reminders that people have used academic qualification certification from abroad and outside the territory to carry out fraud - Evidence grade: A - Sources: Chinese Service Center for Scholarly Exchange（中国（教育部）留学服务中心）. Announcement on Applying Enhanced Certification Review to the Academic Qualification Certification of Some Foreign Institutions (Nine)《关于对部分国外院校学历学位认证加强认证审查的公告（九）》(2025-10-28). ；Chinese Service Center for Scholarly Exchange（中国（教育部）留学服务中心）. -- Notes: "Enhanced review" does not mean the certification will definitely be refused, but it does mean you will be asked for more documents and that it will take longer. Any agent promising "certification guaranteed" or "expedited through inside channels" is a scam; the Service Center has issued specific warnings about fraud. Having somebody write your thesis or take your online courses for you across the border, if it is found out, not only voids the qualification but may also affect your visa record. \ No newline at end of file +- Notes: "Enhanced review" does not mean the certification will definitely be refused, but it does mean you will be asked for more documents and that it will take longer. Any agent promising "certification guaranteed" or "expedited through inside channels" is a scam; the Service Center has issued specific warnings about fraud. Having somebody write your thesis or take your online courses for you across the border, if it is found out, not only voids the qualification but may also affect your visa record. diff --git "a/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" index 80313bbd..963ac533 100644 --- "a/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" +++ "b/en/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -2,15 +2,15 @@ # 33. Living after a disability -This section is about the days after a disability has already happened. How to keep a disability from happening is in section 1 and section 13, and is not repeated here. +This section is about the days after a disability has happened. How to prevent one is in section 1 and section 13, and is not repeated here. -Four measures are counted separately and are not converted into one another. The emergency and complication items are counted by mortality. The subsidies, benefits and lawsuits items are counted by money. The schooling and paperwork items are counted by time. The psychiatric admission, capacity and guardianship items are counted by personal freedom. +Four measures are counted separately and are not converted into each other. The items on emergencies and complications are counted in mortality. The items on subsidies, benefits and lawsuits are counted in money. The items on schooling and paperwork are counted in time. The items on psychiatric admission, capacity for conduct and guardianship are counted in personal freedom. -The World Health Organization's briefing gives three numbers. Worldwide about 1.3 billion people have a significant functional disability, which is 16% of the world's population, roughly one in every six people. Some disabled people die up to 20 years earlier than people without a disability. Disabled people run twice the risk of depression, asthma, diabetes, stroke, obesity or poor oral health. A large part of those gaps is not caused by the disability itself, but by not being able to see a doctor, go to school, find work or get out of the house. So half the items in this section are about the body, and half about procedures and rights. +The World Health Organization's briefing gives three numbers. About 1.3 billion people worldwide have a significant functional impairment, 16% of the world's population, roughly one in every six people. Some disabled people die up to 20 years earlier than people who are not disabled. Disabled people have twice the risk of depression, asthma, diabetes, stroke, obesity or poor oral health. A large part of those gaps is not caused by the disability itself but by not being able to see a doctor, not being able to go to school, not being able to find work and not being able to get out of the house. So half the items in this section are about the body and half are about paperwork and rights. -Five things are already written elsewhere; here they are only pointed to, not repeated. How to get a disability certificate, and how the seven categories and grades one to four are assessed: see item 11 in section 24 (disability certificate). When to have the disability assessment done: see item 10 in section 24 (disability assessment). How to claim the two disability subsidies: see item 8 in section 7 (the two subsidies). For a work injury, the labour capacity assessment and how much a disability grade converts to in money: see item 15 in section 19 (labour capacity assessment). How to apply for long-term care insurance with a severe loss of function: see item 7 in section 17 (long-term care insurance). That item is written in the section on when there are elderly at home, but this benefit has no age limit, and young people can claim it too. +Five things already written up elsewhere are only pointed to here, not repeated. How to get a disability certificate, the seven categories and how levels one to four are assessed: see item 10 in section 24 (disability certificate). When to have a disability assessed: see item 9 in section 24 (disability assessment). How to claim the two disability subsidies: see item 8 in section 7 (the two subsidies). How a work injury's assessment of labour capacity and disability level convert into money: see item 14 in section 19 (assessment of labour capacity). How to apply for long-term care insurance for severe loss of ability: see item 7 in section 17 (long-term care insurance). That item is written in the section on having elderly at home, but this benefit has no age limit and young people can get it too. -Most items in this section benefit you and your family, which is the top two grades. Where they affect other people, that is written separately in the notes. +In most items in this section the beneficiary is you and your family, which puts them in the top two tiers. Where other people are involved, that is written separately in the notes. ### 1. After a spinal cord injury, a sudden severe headache and sweating: sit the person up straight first, loosen the clothing, and call 120 at the same time @@ -20,7 +20,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: B - Sources: Krassioukov A, Warburton DE, Teasell R, Eng JJ; Spinal Cord Injury Rehabilitation Evidence Research Team (2009). A systematic review of the management of autonomic dysreflexia after spinal cord injury. Archives of Physical Medicine and Rehabilitation, 90(4):682-695. - Notes: The grade is B because those three moves at the scene have no randomised controlled trial behind them, only consensus and physiological reasoning. The review itself says that most of these measures are supported only by uncontrolled studies. The drugs are the doctor's business; do not give a blood-pressure drug at home yourself. Laying the person flat will push the blood pressure higher, so sit them up straight. Family and care workers need to know these three steps too. Do not just ride it out as an ordinary headache. - ### 2. The first ten years after becoming disabled are the high-risk period for suicide; do not treat it as something to think your way out of - Cost: No money. Make one phone call, or register once at a hospital. @@ -29,7 +28,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Savic G, DeVivo MJ, Frankel HL, Jamous MA, Soni BM, Charlifue S (2018). Suicide and traumatic spinal cord injury - a cohort study. Spinal Cord, 56(1):2-6. - Notes: This study looked only at spinal cord injury, so applying it to blindness, amputation, burns and other causes of disability is an extrapolation. What it can tell you is the time window: the risk does not end at discharge, and you have to watch it for ten years. To have somebody to talk to, call 12356: see item 11 in section 29 (12356). Do not stock sleeping pills and pesticide at home: see item 25 in section 1 (thoughts of suicide). If grief and mood are stuck in the same place after half a year, register at a psychiatry or clinical psychology department: see item 8 in section 29 (grief past half a year). - ### 3. Psychiatric admission follows the principle of voluntariness; only two circumstances allow it without the person's consent - Cost: No money. Remember the two circumstances so that you can say them when you need them. @@ -38,7 +36,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Mental Health Law of the People's Republic of China（中华人民共和国精神卫生法）(2018 amendment, articles thirty, thirty-one, thirty-two, thirty-five and seventy-eight). - Notes: The other way round, not just anybody can be sent in either. Article twenty-eight provides that where a person suspected of having a mental disorder has already carried out an act of self-harm or of endangering another's safety, or is at risk of doing so, their close relatives, their work unit or the local public security organ must immediately take measures to stop it and send them to a medical institution for a diagnosis of mental disorder. A medical institution that receives such a person may not refuse to make a diagnosis. So "sent in for a diagnosis" and "kept in for admission" are two doors: the first is low, the second is high. Article seventy-five lists the medical institution's own liability; carrying out restraint or isolation against the rules, or diagnosing a person without a mental disorder as having one, leads to discipline up to withdrawal of the practising certificate. What to do when somebody near you has an urge to hurt someone: see section 8. - ### 4. The person who cares long-term for a disabled family member needs somebody watching their own health - Cost: No money. Hand a share of the caring work to somebody else, and get a check-up yourself. @@ -47,7 +44,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Schulz R, Beach SR (1999). Caregiving as a risk factor for mortality: the Caregiver Health Effects Study. JAMA, 282(23):2215-2219. - Notes:Disputed: The lower bound of the credible range in this study sits exactly on 1.00, so the possibility of "no difference" has not been ruled out. The population is older Americans living with a spouse, caring for an older spouse who is disabled; applying it to people caring for a younger disabled relative is an extrapolation. What it can tell you is the direction: the people who find it a strain run higher risk. Hand over as much of the caring as you can; ask the local federation of disabled persons and the civil affairs department about respite and day-care services. The carer should not stop their own chronic illness medication: see item 1 in section 16 (take the medicine as prescribed). The person who benefits from this item is yourself. - ### 5. If you sit in a wheelchair long-term, switch to a pressure-relief cushion and check your sitting bones and tailbone every day - Cost: A pressure-relief cushion runs from a few hundred yuan to over a thousand. One look a day, one minute. @@ -56,7 +52,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: B - Sources: Brienza D, Kelsey S, Karg P, et al. (2010). A randomized clinical trial on preventing pressure ulcers with wheelchair seat cushions. Journal of the American Geriatrics Society, 58(12):2308-2314. - Notes: There are two reasons for the grade B. First, the sample was only 232 people, the P value of 0.04 on the primary endpoint had just crossed the line, and the secondary endpoint showed no difference. Second, the population is very old nursing-home residents, so applying it to young wheelchair users is an extrapolation. In the trial everyone was first fitted with a well-fitting wheelchair, so "fits" and "cushion" work together; do not change only the cushion. For pressure ulcers during long bed rest, an air mattress and turning on a schedule: see item 8 in section 17 (pressure ulcers). How to apply for the subsidy for fitting basic assistive devices: see item 7 in this section (ask everything at the county federation). - ### 6. Do not buy therapies and devices that "can cure paralysis, blindness or deafness" - Cost: No money, it saves money. The hard part is that the family is holding on to hope and cannot be talked out of it. @@ -66,14 +61,14 @@ Most items in this section benefit you and your family, which is the top two gra - Sources: Advertising Law of the People's Republic of China（中华人民共和国广告法）(2021 amendment, articles four, sixteen and seventeen). - Notes: There is only one test: see whether it dares to promise everything. The rehabilitation department of a proper hospital will not promise you that you will stand up after a number of courses of treatment. When you meet this kind of advertising, report it to 12315 or to the local market regulator. Health supplements use the same script: see item 10 in section 6 (health supplements). If you have already bought and want your money back, follow the method for online shopping and pre-paid consumption: see item 29 in section 5 (online shopping follows the platform's rules). The people who benefit from this item are yourself and your family. -### 7. Once the disability certificate is issued, go to the county federation of disabled persons and ask about everything you can apply for in one visit +### 7. Once the disability certificate is issued, go to the county federation and ask what you can apply for, all in one visit -- Cost: No money. One trip to the county federation and a morning of asking. There is more than one department to visit afterwards. -- In plain language: The certificate itself pays nothing; it is a key. What hangs off it is several separate benefits, and each one you have to apply for yourself — none of them comes to you on its own. On the first visit to the federation, go through the list of items one by one; that saves trips back and forth later. -- Benefit: The certificate unlocks at least six benefits, each with its own basis. One: the living allowance for persons with disabilities in difficulty and the nursing allowance for persons with severe disabilities — see item 8 in section 7 (the two allowances). Two: rehabilitation assistance for children with disabilities — see item 8 in this section (rehabilitation assistance). Three: a subsidy for fitting basic assistive devices, based on Article twenty-six of the Regulations on Disability Prevention and Rehabilitation of Persons with Disabilities (《残疾预防和残疾人康复条例》). Four: a subsidy for home accessibility modifications — see item 9 in this section (ramps, handrails and bathroom modifications). Five: quota hiring and employment services — see item 10 in this section (declare the certificate). Six: a personal income tax reduction — see item 11 in this section (personal income tax reduction). Article twenty-six of the Regulations also states that the state provides basic rehabilitation services to poor and severely disabled people in urban and rural areas through key rehabilitation projects, and 「按照国家有关规定对基本型辅助器具配置给予补贴」, that is, subsidises the fitting of basic assistive devices in accordance with relevant state provisions. Article twenty provides that the relevant departments of people's governments at county level and above, and the federations of disabled persons, shall use community resources to set up rehabilitation premises or purchase services, and carry out community rehabilitation work such as rehabilitation guidance, daily-living-skills training, rehabilitation nursing, assistive-device fitting, information consultation and referrals (nationwide, in force from 2017-7-1) +- Cost: No money. One trip to the county federation, and one morning of asking. There is more than one department to go to afterwards. +- In plain language: The disability certificate pays nothing itself; it is a key. What hangs off it is several benefits, each independent of the others, and you have to apply for each one yourself; none of them comes to you. On your first visit to the federation, go through the items on the list one by one and ask about them. That saves you several trips back and forth. +- Benefit: At least six benefits hang off the certificate, and each has its own basis. First, the living allowance for disabled people in difficulty and the care allowance for severely disabled people: see item 8 in section 7 (the two subsidies). Second, rehabilitation assistance for disabled children: see item 8 in this section (rehabilitation assistance). Third, the subsidy for basic assistive devices, whose basis is Article twenty-six of the Regulations on Disability Prevention and Rehabilitation of Persons with Disabilities（残疾预防和残疾人康复条例）. Fourth, the subsidy for adapting the home to be barrier-free: see item 9 in this section (ramps, handrails and bathroom modifications). Fifth, quota employment and employment services: see item 10 in this section (say up front that you hold the certificate). Sixth, the reduction in individual income tax: see item 11 in this section (individual income tax reduction). Article twenty-six of those Regulations also states that the state provides basic rehabilitation services for disabled people in difficulty in urban and rural areas and for severely disabled people by carrying out key rehabilitation programmes, and 「按照国家有关规定对基本型辅助器具配置给予补贴」. Article twenty provides that the relevant departments of people's governments at county level and above and the federations shall make use of community resources, set up rehabilitation premises or purchase services, and carry out community rehabilitation work such as rehabilitation guidance, training in the activities of daily living, rehabilitation care, fitting of assistive devices, information and referral (nationwide, in force from 2017-7-1) - Evidence grade: A -- Sources: State Council (2017). Regulations on Disability Prevention and Rehabilitation of Persons with Disabilities (《残疾预防和残疾人康复条例》) (articles twenty and twenty-six). -- Notes: The process of getting the certificate itself: see item 11 in section 24 (disability certificate). Most subsidy standards and specific measures are set at the provincial level, and the same benefit can differ several times over between two provinces. Write down the name of each item when you ask, then go home and look up the original document again on the local federation's website and the government website, by name; do not rely on what the person at the counter tells you. Do not expect to ask everything in one visit — go back and ask again after a year, policies change. +- Sources: State Council (国务院) (2017). Regulations on Disability Prevention and Rehabilitation of Persons with Disabilities（残疾预防和残疾人康复条例）(Articles twenty, twenty-six). +- Notes: The procedure for getting the certificate itself is in item 10 in section 24 (disability certificate). The subsidy standards and the detailed measures are set mostly at provincial level, and the same item can differ severalfold between two provinces. As you ask, write down the name of each item, and at home look up the original document by that name on the local federation's website and on the government's website; do not rely only on what the counter says. Do not expect to get it all in one visit; go back and ask again a year later, because the policies change. ### 8. If the child is under 7 and has a disability or autism, apply to the county federation for rehabilitation assistance @@ -83,7 +78,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: State Council (2018). Opinions on Establishing a Rehabilitation Assistance System for Children with Disabilities (《国务院关于建立残疾儿童康复救助制度的意见》) (Guo Fa [2018] No. 20). ; State Council (2017). Regulations on Disability Prevention and Rehabilitation of Persons with Disabilities (《残疾预防和残疾人康复条例》) (article twenty-six). - Notes: The assistance standards are set by each province; first search the local federation's website for 「残疾儿童康复救助实施办法」. The designated-institution list is also in the federation's hands. The document requires the federation, together with the relevant departments, to tighten supervision of entry and exit for designated rehabilitation institutions, establish regular inspection and comprehensive evaluation mechanisms, and set up a blacklist. For an institution not on the list you pay every yuan yourself, and if something goes wrong it is hard to hold anyone to account. The beneficiary of this item is your child, which puts it in the second tier. Institutions that promise to "guarantee a cure" are handled under item 6 in this section (can cure paralysis). - ### 9. Ramps, handrails and bathroom modifications at home can be subsidised by a government at county level or above - Cost: Applying costs nothing. Most of the work itself is subsidised, but you may still have to cover part of it. You will have to visit the federation and the subdistrict office. @@ -92,7 +86,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: B - Sources: Barrier-free Environment Construction Law of the People's Republic of China（中华人民共和国无障碍环境建设法）(articles eighteen, nineteen, sixty-two and sixty-three). - Notes: The grade is B because the law only writes 「应当给予适当补贴」 — shall give appropriate subsidies; how much is paid and who counts as qualifying is handed entirely to the localities, with no nationally uniform number. First search the local federation's website for 「困难重度残疾人家庭无障碍改造」. Quotas are mostly allocated by year, so queue early. Adding a lift to a residential building and modifying stairwells is a different process, through the housing and urban-rural development department. Where the barrier-free facilities in a public place fall short, complain to the competent department under Article sixty-two — it has to answer you. - ### 10. When job hunting, say up front that you hold the certificate; hiring you offsets a payment for the firm - Cost: No money. One extra sentence in the interview. The hard part is bringing it up. @@ -101,7 +94,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: State Council (2007). Regulations on Employment of Persons with Disabilities (《残疾人就业条例》) (articles eight and nine). ; Ministry of Finance, State Taxation Administration and China Disabled Persons' Federation (2015). Measures for the Collection and Use of the Disability Employment Security Fund (《残疾人就业保障金征收使用管理办法》) (Cai Shui [2015] No. 72, articles six, eight and sixteen). ; Law of the People's Republic of China on the Protection of Persons with Disabilities（中华人民共和国残疾人保障法）(article thirty-six). - Notes: For a small or micro enterprise registered within the last three years with 20 staff or fewer, this reason does not hold — do not count on it. The fund's reduction and exemption policies have been adjusted year after year; what applies is whatever the Ministry of Finance announces that year. Also, quota hiring requires real employment. Taking a wage for a nominal post, or letting a unit use you to offset the fund, is fraud, and you would be dragged into it too. Recognition as a person having difficulty finding work, and the social insurance allowance: see item 12 in section 7 (persons having difficulty finding work). Vocational training allowance: see item 13 in section 7 (vocational training allowance). The person who benefits from this item is yourself. - ### 11. Personal income tax can be reduced for disabled people; phone the provincial tax bureau to find out by how much - Cost: No money. Tick one box when filing, or submit filing materials once. @@ -110,7 +102,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Individual Income Tax Law of the People's Republic of China（中华人民共和国个人所得税法）(2018 amendment, article five). ; Law of the People's Republic of China on the Protection of Persons with Disabilities（中华人民共和国残疾人保障法）(article thirty-six). - Notes: The benefit is graded "small" because the law itself gives no number; in most provinces the annual reduction comes to a few hundred to a few thousand yuan. It is not much money, but one filing keeps working for years, so it is worth the trip. How much your own province reduces is decided only by the local tax bureau's announcement; comparison tables passed around online are out of date year-round. If you run a small shop of your own, the administrative and institutional fees are waived under Article thirty-six of the Law on the Protection of Persons with Disabilities — raise it when you apply for the licence. - ### 12. Guide dogs may be taken into public places and public transport; blind people ride city buses free - Cost: No money. If you are stopped, you have to open your mouth and quote the statute, and there may be a standoff for a while. @@ -128,7 +119,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Ministry of Education (教育部) and China Disabled Persons' Federation (中国残联) (2017). Provisions on the Administration of the National Unified Examination for Admission to Regular Higher Education Institutions by Candidates with Disabilities (Teaching〔2017〕4, Articles five, seven, eight, nine and eighteen)《残疾人参加普通高等学校招生全国统一考试管理规定（教学〔2017〕4 号，第五条、第七条、第八条、第九条、第十八条）》. - Notes: The application dates are set by each province and written into that year's rules for signing up for the examination. Miss the date and there is no second one. So check the provincial education examination authority's website in the first term of the final year of high school. Article eighteen also covers the postgraduate entrance examination, the adult higher education entrance examination and other state education examinations, which may apply by reference. The rule that a school must not refuse admission during the school years is item 14 in this section (schools must not refuse admission). The person who benefits from this entry is you or your child - ### 14. A disabled child applying for a school place: the school must not refuse; when the child cannot get to school the education authority arranges teaching at home - Cost: No money. You have to hand in one application and there may be one assessment. When the school makes excuses, you go up to the education authority @@ -137,7 +127,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: State Council (国务院) (2017). Regulations on Education for Persons with Disabilities (Articles seven, twenty, twenty-three, twenty-nine and fifty-two)《残疾人教育条例（第七条、第二十条、第二十三条、第二十九条、第五十二条）》. - Notes: A school saying 「我们这儿没条件」 in conversation does not count. Make it give a written reply, and take that reply to the county education bureau. The assessment is done by an expert committee the education bureau commissions. The words in Article twenty-three, 「学习要求可以有适度弹性」, are useful: you can cite them when a school tries to push a child out on the grounds of results. For keeping evidence of bullying at school and which procedure the school must follow, see item 3 in section 30 (school bullying). For reasonable accommodation in the national college entrance examination see item 13 in this section (reasonable accommodation). The person who benefits from this entry is your child - ### 15. You can still get a driving licence without the right leg or without both legs; the vehicle category is called C5 - Cost: The tuition is about the same as for an ordinary licence, a few thousand yuan in most places. The car has to be converted to manual controls. A medical check every three years @@ -146,7 +135,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Ministry of Public Security Order No. 172 (公安部令第 172 号). Provisions on the Application for and Use of Motor Vehicle Driving Licences (Articles seventy-eight and eighty-five, Annex 1)《机动车驾驶证申领和使用规定（第七十八条、第八十五条、附件 1）》. - Notes: A much lower bar is the motorised wheelchair for disabled people. Article one hundred and nineteen of the Road Traffic Safety Law《道路交通安全法》puts it among non-motor vehicles, so no licence is needed. Article fifty-eight says that a motorised wheelchair for disabled people may not exceed fifteen kilometres per hour in a non-motor vehicle lane. A C5 car needs a hand-operated accelerator and brake. Have the conversion done by a qualified manufacturer, and then go to the vehicle administration office to change the registration. If you do not submit the proof of physical condition within one year after three scoring periods end, the licence is cancelled - ### 16. Do your rehabilitation in a proper institution that has a rehabilitation medicine department, and let the therapist set the intensity; more is not better - Cost: Inpatient rehabilitation costs several hundred yuan a day, and the part you pay yourself varies from place to place. The time works out in months. You have to practise every day, and it is very dull @@ -155,7 +143,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Langhorne P, Ramachandra S; Stroke Unit Trialists' Collaboration (2020). Organised inpatient (stroke unit) care for stroke: network meta-analysis. Cochrane Database of Systematic Reviews, 4:CD000197. ; AVERT Trial Collaboration group (2015). Efficacy and safety of very early mobilisation within 24 h of stroke onset (AVERT): a randomised controlled trial. Lancet, 386(9988):46-55. - Notes:Disputed: both pieces of evidence come from stroke, and applying them to spinal cord injury, amputation, burns and other causes of disability is an extrapolation. What the AVERT result shows is that 「越早越猛」 does not hold, not that early activity is useless. In the trial the great majority of people in both groups were out of bed within 24 hours; what differed was the frequency and the total amount. When you choose an institution, check whether it has a rehabilitation medicine department and full-time rehabilitation therapists and occupational therapists. Do not judge it by the two characters 「康复」 on the sign. For referral between the grassroots level and the top-tier hospital, and how the deductible is worked out, see item 1 in section 24 (referral step by step). For community-based rehabilitation and assistive devices see item 7 in this section (ask at the county federation of disabled persons) - ### 17. If your hearing is going, get a hearing aid fitted; the reason is to hear, not to prevent dementia - Cost: A hearing aid runs from several thousand to several tens of thousands of yuan, and most of it is yours to pay. It has to be fitted, and you go back several times for adjustments @@ -164,6 +151,7 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Lin FR, Pike JR, Albert MS, et al.; ACHIEVE Collaborative Research Group (2023). Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): a multicentre, randomised controlled trial. Lancet, 402(10404):786-797. - Notes:Disputed: the claim you often see online, that hearing aids cut cognitive decline by nearly half, comes from a subgroup inside this trial. It is not the main result, and the main result is that there was no difference between the two groups. 「配助听器」 and 「防痴呆」 have to be judged separately: the first is worth doing, the second should not be used as the reason. A candidate with a hearing disability sitting the national college entrance examination can be exempted from the foreign language listening test; see item 13 in this section (reasonable accommodation). A person with a hearing loss must wear a hearing device while driving; see item 15 in this section (C5) + ### 18. To have an adult declared 「管不了自己的事」 you apply to the court; the family's word does not count - Cost: An application to the court carries a court fee, usually tens of yuan up to a hundred yuan. The assessment fee is separate, usually several thousand yuan, and the applicant advances it. The whole thing takes several months. @@ -172,7 +160,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Civil Code of the People's Republic of China (Articles twenty-one, twenty-two, twenty-three and twenty-four)《中华人民共和国民法典（第二十一条、第二十二条、第二十三条、第二十四条）》. - Notes: The application goes through the court's special procedure, which is faster than an ordinary civil case but still takes several months. You have to submit a diagnosis from a hospital, and in most cases the court will also commission a judicial assessment. The assessment fee is usually several thousand yuan, advanced by the applicant. Signing for someone, or disposing of his property, without going through this procedure may later be found invalid, and a flat that has been sold may have to be returned. One more point to keep in mind: limited capacity for civil conduct does not mean nothing can be done. Things that match his intelligence and mental condition, and things that bring him pure benefit, he can do himself. Do not act for him across the board. For how the guardian is determined see item 19 in this section (guardian) - ### 19. A guardian for an adult is determined in the order set by law; if the person under guardianship injures someone else, the guardian pays - Cost: No money. Where there is a dispute you go to the residents' committee, the villagers' committee, the civil affairs department or the court @@ -181,7 +168,6 @@ Most items in this section benefit you and your family, which is the top two gra - Evidence grade: A - Sources: Civil Code of the People's Republic of China (Articles twenty-eight, thirty-one, thirty-three, one thousand one hundred and eighty-eight and one thousand one hundred and eighty-nine)《中华人民共和国民法典（第二十八条、第三十一条、第三十三条、第一千一百八十八条、第一千一百八十九条）》. - Notes: Article thirty-three is voluntary guardianship. Fixing the person in writing while he is still of sound mind saves most of the disputes that follow; for how to write it see item 1 in section 17 (voluntary guardianship). Be careful with the compensation part: handing the person over to a care home or a relative to look after him does not take the responsibility off the guardian, and the person entrusted answers for his own fault only. Guardian liability insurance for the person under guardianship is therefore worth having. Article seventy-eight of the Mental Health Law《精神卫生法》also says that a guardian who abandons a patient, or who fails in his duties and causes harm, bears compensation liability under the law - ### 20. Refused a job, kept off the permanent staff or underpaid because of a disability: you can sue, but work out the cost of the process first - Cost: Legal fees run from several thousand to several tens of thousands of yuan, and you pay them yourself. An ordinary procedure at first instance takes 6 months or more, and can be extended. You have to preserve the evidence yourself. @@ -189,4 +175,4 @@ Most items in this section benefit you and your family, which is the top two gra - Benefit: Article three of the Law on the Protection of Persons with Disabilities《残疾人保障法》says: 「禁止基于残疾的歧视。禁止侮辱、侵害残疾人。」 Article thirty-eight says: 「在职工的招用、转正、晋级、职称评定、劳动报酬、生活福利、休息休假、社会保险等方面，不得歧视残疾人。」 The same article also provides that the unit where a disabled employee works shall, according to the characteristics of the disabled employee, provide suitable working conditions and labour protection, and shall, as actually necessary, adapt the workplace, the equipment and the living facilities. Article sixty-four says: 「违反本法规定，在职工的招用等方面歧视残疾人的，由有关主管部门责令改正；残疾人劳动者可以依法向人民法院提起诉讼。」 Article four of the Regulations on the Employment of Persons with Disabilities《残疾人就业条例》says: 「禁止在就业中歧视残疾人。」 Article thirteen says that an employer shall not discriminate against disabled employees in promotion, advancement, grading of professional title, pay, social insurance, living benefits and the like. Article forty-two of the Legal Aid Law《法律援助法》says that an applicant who has documents proving that he belongs to a specific group such as 「无固定生活来源的未成年人、老年人、残疾人等特定群体」 is exempted from any check on whether he is financially in difficulty. Article forty-five says that where a legal aid institution provides legal aid services to elderly people or to persons with disabilities, it shall provide accessible facilities, equipment and services as the circumstances require (nationwide) - Evidence grade: A - Sources: Law of the People's Republic of China on the Protection of Persons with Disabilities (Articles three, thirty-eight and sixty-four)《中华人民共和国残疾人保障法（第三条、第三十八条、第六十四条）》. ; State Council (国务院) (2007). Regulations on the Employment of Persons with Disabilities (Articles four and thirteen)《残疾人就业条例（第四条、第十三条）》. ; Legal Aid Law of the People's Republic of China (Articles forty-two and forty-five)《中华人民共和国法律援助法（第四十二条、第四十五条）》. -- Notes: The cost of the process has to be worked out first. An ordinary procedure at first instance takes 6 months or more and can be extended; a simplified procedure takes 3 months. Legal fees are not among the litigation costs listed in Article six of the Regulations on the Payment of Litigation Costs《诉讼费用交纳办法》, and the 「败诉方负担」 in Article twenty-nine does not include legal fees. So take the two cheap routes first: complain to the human resources and social security department or to the disabled persons' federation, which must order the discrimination stopped; and apply for legal aid at the same time, because Article forty-two exempts disabled people from the check on financial hardship. Collecting evidence is the hardest part. When you are turned down at an interview the other side seldom leaves a written reason, so keep a recording and keep the messages. Once you are employed, the employment dispute goes to labour arbitration; for keeping evidence and the time limit see item 8 in section 19 (keep the evidence before you leave). The person who benefits from this entry is you yourself \ No newline at end of file +- Notes: The cost of the process has to be worked out first. An ordinary procedure at first instance takes 6 months or more and can be extended; a simplified procedure takes 3 months. Legal fees are not among the litigation costs listed in Article six of the Regulations on the Payment of Litigation Costs《诉讼费用交纳办法》, and the 「败诉方负担」 in Article twenty-nine does not include legal fees. So take the two cheap routes first: complain to the human resources and social security department or to the disabled persons' federation, which must order the discrimination stopped; and apply for legal aid at the same time, because Article forty-two exempts disabled people from the check on financial hardship. Collecting evidence is the hardest part. When you are turned down at an interview the other side seldom leaves a written reason, so keep a recording and keep the messages. Once you are employed, the employment dispute goes to labour arbitration; for keeping evidence and the time limit see item 8 in section 19 (keep the evidence before you leave). The person who benefits from this entry is you yourself diff --git "a/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" "b/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" index 74e9ea9b..db0ad05d 100644 --- "a/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" +++ "b/en/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" @@ -18,7 +18,6 @@ In this section the beneficiaries are you and your family, which is the top two - Evidence grade: B - Sources: Larson AM, Polson J, Fontana RJ, et al.; Acute Liver Failure Study Group (2005). Acetaminophen-induced acute liver failure: results of a United States multicenter, prospective study. Hepatology, 42(6):1364-1372. ；National Medical Products Administration (国家药监局) (2020). Announcement on Revising the Package Inserts for Immediate-Release and Extended-Release Acetaminophen Preparations (2020 No. 15), and Annex 2, Requirements for Revising the Package Insert of Immediate-Release Acetaminophen Over-the-Counter Medicines《对乙酰氨基酚常释制剂非处方药说明书修订要求》, reprinted by the Hunan Provincial Drug Administration (湖南省药品监督管理局). ；21 CFR 201.326(a)(1)(iii)(A) Liver warning. - Notes: Acetaminophen is also called paracetamol. Compound cold medicines whose generic name carries the two characters for acetaminophen (氨酚), such as those sold as acetaminophen with pseudoephedrine, or as acetaminophen, caffeine and chlorpheniramine, contain this ingredient. If you are not sure, take every box to the pharmacist in the shop and ask. The grade is B because the United States figures come from case registration with no control group, so nobody can work out how many incidents reading the ingredient list would prevent. The benefit is graded "large" by the severity of the outcome: close to three in ten acute liver failure patients die. - ### 2. Do not give children aspirin, nimesulide or metamizole for a fever, and do not give minors any cold medicine that contains metamizole - Cost: No money. When you buy a fever reducer, look at the generic name and the ingredient list. @@ -27,7 +26,6 @@ In this section the beneficiaries are you and your family, which is the top two - Evidence grade: B - Sources: Belay ED, Bresee JS, Holman RC, Khan AS, Shahriari A, Schonberger LB (1999). Reye's syndrome in the United States from 1981 through 1997. New England Journal of Medicine, 340(18):1377-1382. ；Notice on Strengthening the Administration of Oral Nimesulide Preparations (国食药监安〔2011〕209号). ；National Medical Products Administration (国家药监局) (2020). Announcement on Revising the Package Inserts of Metamizole-Related Products (2020 No. 34), with Annexes 1 to 3. - Notes: Metamizole is not the first choice for adults either. The insert says it may cause agranulocytosis, aplastic anaemia and anaphylactic shock, and it is used for a short time only when the illness is severe and no other effective medicine is available. Zhongganling sounds like an ordinary cold medicine, and it contains metamizole. For which fever reducer a child should take, and how much by body weight, ask a paediatrician or a pharmacist. A child under 3 months old with a fever goes straight to hospital, see item 8 in section 20 (go straight to hospital at 38 ℃). The grade is B because the Reye's syndrome figures come from surveillance reports, and the fall in cases and the issue of the warning stand in sequence rather than in a controlled trial. The benefit is graded "large" because after the warning cases fell from 555 a year to fewer than 36. - ### 3. If you are over 60, have bled from the stomach, or take anticoagulants or steroids, ask a doctor before taking a painkiller such as ibuprofen - Cost: No money. It takes asking a doctor or a pharmacist once. @@ -36,7 +34,6 @@ In this section the beneficiaries are you and your family, which is the top two - Evidence grade: A - Sources: Coxib and traditional NSAID Trialists' (CNT) Collaboration; Bhala N, Emberson J, Merhi A, et al. (2013). Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. Lancet, 382(9894):769-779. ；21 CFR 201.326(a)(2)(iii)(A) Stomach bleeding warning. - Notes: The paper's conclusion is about high-dose use. The paper gives no separate figure for occasional use over a day or two at the dose on the insert. Many compound cold medicines also contain this ingredient, so taking them together with a painkiller repeats it; read the ingredient list first. If a doctor has put you on low-dose aspirin for the long term to prevent blood clots, do not stop it yourself, and do not add a second painkiller yourself. - ### 4. Do not give a child under 2 compound cold medicines or cough medicines on your own - Cost: No money, and it also saves the money you would spend on medicines. @@ -45,7 +42,6 @@ In this section the beneficiaries are you and your family, which is the top two - Evidence grade: B - Sources: National Medical Products Administration (国家药监局) (2021). Announcement on Revising the Package Inserts of 14 Products Including the Paracetamol, Chlorpheniramine and Dextromethorphan Oral Solution (2021 No. 57), with annexes. ；Smith SM, Schroeder K, Fahey T (2014). Over-the-counter (OTC) medications for acute cough in children and adults in community settings. Cochrane Database of Systematic Reviews, (11):CD001831. - Notes: Many of these 14 products carry the two characters for acetaminophen (氨酚) in their names, and they contain acetaminophen, so taking them together with a fever reducer repeats it; see item 1 in this section (acetaminophen). One trial in that review found honey better than placebo, but honey must not be given below 1 year of age, see item 6 in section 20 (honey). The grade is B because the review holds few trials for each class of drug and the authors could not pool them into a figure. The benefit is graded "medium" as a judgement: no benefit has been shown for children, and the risks are overdose and repeated ingredients. - ### 5. After 20 weeks of pregnancy, do not take a painkiller such as ibuprofen on your own - Cost: No money. @@ -54,6 +50,7 @@ In this section the beneficiaries are you and your family, which is the top two - Evidence grade: B - Sources: U.S. Food and Drug Administration (2020-10-15). FDA recommends avoiding use of NSAIDs in pregnancy at 20 weeks or later because they can result in low amniotic fluid. Drug Safety Communication. - Notes: The FDA warns that many over-the-counter medicines for colds, flu and insomnia also contain these ingredients, so read the ingredient list. United States over-the-counter labelling used to warn only about the last 3 months of pregnancy; this time the line moved back to about 20 weeks. Low-dose aspirin for the prevention of pre-eclampsia, see item 5 in section 27 (low-dose aspirin). The grade is B because the figures come from adverse event reports, and there is no way to know how many people took the drug without harm. The benefit is graded "medium" as a judgement: the outcome is serious, but it is rare. + ### 6. Buy omeprazole yourself for acid reflux and heartburn, and take it for at most 7 days; if you have difficulty swallowing, vomit blood or pass black stool, go straight to hospital - Cost: No money. Take it for the number of days the insert gives. @@ -62,7 +59,6 @@ In this section the beneficiaries are you and your family, which is the top two - Evidence grade: B - Sources: National Medical Products Administration (国家药监局) (2022). Announcement on the Conversion of Omeprazole Enteric-Coated Tablets from a Prescription Medicine to an Over-the-Counter Medicine (2022 No. 68), and Annex 2, Over-the-Counter Package Insert Template《非处方药说明书范本》. - Notes: Clopidogrel is the anti-clotting drug that people who have had a stent put in or have had a heart attack often take. Omeprazole weakens it, so if you take it, ask your doctor first. The grade is B because these are the rules in the package insert, and no study has worked out how many incidents following them would prevent. The benefit is graded "medium" as a judgement: what it prevents is a stomach bleed or stomach cancer being caught late, but most acid reflux and heartburn is not serious. - ### 7. Do not ask a doctor for antibiotics for an ordinary cold - Cost: No money, and it also saves the money you would spend on medicines. @@ -71,7 +67,6 @@ In this section the beneficiaries are you and your family, which is the top two - Evidence grade: A - Sources: Kenealy T, Arroll B (2025). Antibiotics for the common cold and acute purulent rhinitis. Cochrane Database of Systematic Reviews, 11:CD000247. - Notes: This version was republished in 2025, and the abstract states that the literature search ran up to 2013. Antibiotics are prescription medicines in China, so a pharmacy has to sell them against a prescription; do not take leftover antibiotics from home either. The benefit is graded "medium" as a judgement: what it prevents is adverse reactions, not death. - ### 8. When you have diarrhoea, start with oral rehydration salts; do not give anti-diarrhoeal medicines to children under 5 - Cost: A box of oral rehydration salts costs a few yuan to a dozen-odd yuan. Keep one box at home. @@ -80,7 +75,6 @@ In this section the beneficiaries are you and your family, which is the top two - Evidence grade: A - Sources: World Health Organization (2005). The treatment of diarrhoea: a manual for physicians and other senior health workers, 4th revision. ；Hahn S, Kim S, Garner P (2002). Reduced osmolarity oral rehydration solution for treating dehydration caused by acute diarrhoea in children. Cochrane Database of Systematic Reviews, (1):CD002847. - Notes: When you buy it, check the osmolarity on the insert and choose the hypotonic formulation. The manual states that blood in a child's stool is most often bacterial dysentery, which needs a doctor and antibiotics; rehydrating at home alone is not enough. The benefit is graded "medium" as a judgement: the figures compare two rehydration solutions, not drinking against not drinking. - ### 9. Do not take painkillers for more than 15 days a month, or compound painkillers for more than 10 days, or the headache may be the medicine - Cost: No money. You have to keep count of the days you take medicine, and you have to hold back from taking it whenever you reach for it. @@ -88,4 +82,22 @@ In this section the beneficiaries are you and your family, which is the top two - Benefit: The third edition of the International Classification of Headache Disorders of the International Headache Society sets out the diagnostic criteria. A person who already has a primary headache disorder, who has headache on 15 or more days a month, and who has regularly overused headache medicines for more than 3 months. For non-opioid painkillers such as acetaminophen, aspirin and ibuprofen, the threshold for overuse is 15 or more days a month. For compound painkillers the threshold is 10 or more days a month. A compound painkiller is two or more kinds of pain-relieving drug put together, or one with an added helper ingredient such as caffeine, for example acetaminophen with codeine. If you take several non-opioid painkillers in turn, add the days up. The classification states that epidemiological evidence from several countries shows that of the people with headache on 15 or more days a month, more than half have this type of headache. Most patients improve after they stop the drug they have overused, and preventive treatment for headache works better too. - Evidence grade: B - Sources: Headache Classification Committee of the International Headache Society (2018). The International Classification of Headache Disorders, 3rd edition. Cephalalgia, 38(1):1-211. ；ICHD-3 online version 8.2 Medication-overuse headache (including 8.2.3 and 8.2.5). -- Notes: A painkiller whose name carries the character for coffee (咖) usually has caffeine in it, and the limit for a compound painkiller is 10 days. If the number of days you have a headache in a month keeps rising, see a neurologist instead of only adding medicines. The grade is B because the classification gives a diagnostic threshold, not a controlled trial. The benefit is graded "medium" as a judgement: what it buys back is days without headache, and it has nothing to do with death. \ No newline at end of file +- Notes: A painkiller whose name carries the character for coffee (咖) usually has caffeine in it, and the limit for a compound painkiller is 10 days. If the number of days you have a headache in a month keeps rising, see a neurologist instead of only adding medicines. The grade is B because the classification gives a diagnostic threshold, not a controlled trial. The benefit is graded "medium" as a judgement: what it buys back is days without headache, and it has nothing to do with death. +### 10. Clear out expired medicines and leftovers you will never use on a regular schedule, packaging and all, into the hazardous waste bin; do not leave them at home + +- Cost: No money. Go through the medicine cabinet once a year; it takes a dozen-odd minutes. +- In plain language: Nearly half the medicines young children swallow without their parents knowing are prescription tablets. About 30% of the prescription drugs they swallow are their grandparents'. Clearing out what nobody uses means one less thing a child can find. Put what you clear out into the hazardous waste bin with its packaging, or hand it in at a pharmacy that takes old medicines back. +- Benefit: US emergency department surveillance data estimate that between 2004 and 2013 children under 6 who swallowed a medicine without their parents knowing accounted for about 640,000 emergency department visits (estimated figure 640161). In 2013 alone there were about 59,000 (59092). Between 2010 and 2013 the commonest were oral prescription tablets, at 45.9%. Among prescription drugs the commonest were opioid painkillers (13.8%) and benzodiazepine sleeping pills and sedatives (12.7%). 5 US poison control centres surveyed the parents who called for help in 2017. Where a child had swallowed a prescription drug, 30.7% had swallowed a grandparent's. More than half (51.5%) of the prescription drugs had already been taken out of their original packaging by an adult before the child got hold of them. China's National Hazardous Waste Catalogue (2025 edition)《国家危险废物名录（2025 年版）》lists waste medicines and their packaging produced in everyday household life as hazardous waste in domestic waste. Shanghai's sorting guide puts waste medicines in the hazardous waste category and states that it includes expired medicines, medicine capsules, tablets and medicine packaging, with 「废弃药品宜连带包装一并投放」, that is, waste medicines should be put in with their packaging. Shanghai also provides that waste and expired medicines collected by pharmacies from households are put in temporary transfer points set up by each district's market regulation department. The US FDA says the best way to deal with expired and unwanted medicines is to hand them in at a collection point. +- Evidence grade: B +- Sources: Lovegrove MC, Weidle NJ, Budnitz DS (2015). Trends in Emergency Department Visits for Unsupervised Pediatric Medication Exposures, 2004-2013. Pediatrics, 136(4):e821-e829. ; Agarwal M, Lovegrove MC, Geller RJ, et al. (2020). Circumstances Involved in Unsupervised Solid Dose Medication Exposures among Young Children. Journal of Pediatrics, 219:188-195.e6. ; Ministry of Ecology and Environment (生态环境部) and others (2024). National Hazardous Waste Catalogue (2025 Edition)《国家危险废物名录（2025年版）》, annex, List of Hazardous Waste Exempt from Management《危险废物豁免管理清单》, item 1. ; Shanghai Municipal Administration of Greening and City Appearance (上海市绿化和市容管理局) (2024). Guide to the Sorting of Domestic Waste in Shanghai (2024 Edition)《上海市生活垃圾分类投放指引（2024版）》(沪绿容〔2024〕278号). ; Shanghai Municipal Administration of Greening and City Appearance, Shanghai Environmental Protection Bureau and Shanghai Food and Drug Administration (2018). Notice on Regulating the Whole-Procedure Management of Hazardous Waste in the Sorting of Domestic Waste. ; U.S. Food and Drug Administration (2025). Where and How to Dispose of Unused Medicines. +- Notes: The sorting lists differ from place to place; go by the list posted at the waste collection point in your housing estate. Shanghai puts the dregs of Chinese herbal medicine in wet waste, which is not hazardous waste. Keep the medicines you are keeping in their original boxes: do not decant them into a pill organiser or another bottle, and put them somewhere a child cannot reach, see item 20 in section 13 (accidental swallowing). Do not take leftover antibiotics from the medicine cabinet yourself, see item 7 in this section (antibiotics). Where there is no collection point in the United States, people are told to mix the medicines into coffee grounds or cat litter, seal the bag and put it in ordinary rubbish; a few very strong painkillers may be flushed down the toilet. That is the American practice; in China, follow the local sorting rules. It is graded B because the US figures are only registration, and no study has worked out how many accidental swallows clearing out the cabinet would prevent. The benefit is graded small: the figures you have are emergency department visits, not deaths, and the effect of the act of clearing out cannot be worked out either. + + +### 11. Do not drink alcohol while you are on cephalosporins, metronidazole or similar antibiotics, or in the 7 days after you stop; leave alone medicines that contain alcohol such as Huoxiang Zhengqi water and cough syrup + +- Cost: No money. Not drinking for a bit over a week, and looking at the ingredients to see whether a medicine contains ethanol before you take anything else. +- In plain language: Some cephalosporins, metronidazole and drugs like them, mixed with alcohol, make your face flush, your heart race, your chest tight and make you vomit; in severe cases your blood pressure drops. It is not only spirits and beer: Huoxiang Zhengqi water, cough syrup and other medicines containing alcohol can set it off too. Keep away from all of them while you are taking the antibiotic and for 7 days after you stop. +- Benefit: This is called a disulfiram-like reaction. A 2023 piece on safe medication use from the Fujian Provincial Drug Administration says that drinking alcohol before or after taking some cephalosporin medicines can trigger this reaction, and 「会严重损害健康甚至导致生命危险」, that it can seriously damage health and even endanger life. The symptoms include flushing of the face, sweating and palpitations, and in some cases difficulty breathing, falling blood pressure and anaphylactic shock. The drugs it lists, following the Clinical Drug Use Information of the Pharmacopoeia of the People's Republic of China《中华人民共和国药典临床用药须知》, include cefazolin, ceftriaxone, cefoperazone, metronidazole, tinidazole and furazolidone. It states that 「用药后 7 天内禁止饮酒和食用含酒精的药物、食物等」, that for 7 days after taking the medicine you must not drink alcohol or take medicines, foods and so on that contain alcohol, giving chocolate liqueurs and Huoxiang Zhengqi water as examples. The reason is that the enzyme the drug inhibits commonly takes 4 to 5 days to recover. A hospital in Guangzhou compiled 82 cases over 2 years. 57 were caused by cephalosporins (69.51%) and 10 by nitroimidazoles (metronidazole, tinidazole and the like) (12.20%). In 40 cases the alcohol came from an alcoholic drink (48.78%), and in 31 from a medicine containing ethanol (37.80%). The swallowed medicines included Huoxiang Zhengqi water, cough syrup and Leonurus liquid extract. Of the 82 cases 68 were mild, 9 moderate and 5 severe; all eased within 1 to 3 hours, and none died. The US metronidazole instructions require no alcohol during treatment and for at least 3 days after stopping, and also say not to use products containing propylene glycol. +- Evidence grade: B +- Sources: Fujian Provincial Drug Administration (福建省药品监督管理局, Fujian Provincial Drug Science Popularisation and Regulatory Data Centre) (2023). Original science popularisation piece for the 2023 National Safe Medication Month — why you cannot drink alcohol when taking cephalosporins. ; 房财富, 涂星, 刘韬, 黄红兵, 邱孟, 梁蔚婷 (2015). Analysis of 82 Cases of Disulfiram-Like Reaction in Our Hospital, 我院双硫仑样反应82例分析. 中国药房, 26(29):4089-4091. ; Pfizer (2026). FLAGYL 375 (metronidazole) capsules, prescribing information: Contraindications, Interaction with Alcohol. DailyMed. +- Notes: Not every cephalosporin does this, but an ordinary person cannot tell which is which, so following this rule whichever one you are on is the simplest course. When a doctor prescribes a cephalosporin or metronidazole, say of your own accord whether you have been drinking in the last few days. That popularisation piece advises not drinking for 3 days before the medicine either, while admitting there is no authoritative view on the length of that interval. The moment your face flushes and your heart races, stop anything containing alcohol. The popularisation piece says mild cases settle on their own and severe ones need a hospital. Do not drink while taking acetaminophen either, see item 1 in this section (acetaminophen). It is graded B because the evidence is a set of case records and package inserts, with nothing to compare against. The benefit is graded small: all 82 cases eased within a few hours, nobody died, and only 5 were severe. diff --git "a/en/docs/\345\210\232\347\241\256\350\257\212\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216.md" "b/en/docs/\345\210\232\347\241\256\350\257\212\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216.md" new file mode 100644 index 00000000..cc7625d5 --- /dev/null +++ "b/en/docs/\345\210\232\347\241\256\350\257\212\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216.md" @@ -0,0 +1,50 @@ +[← Back to the contents](../README.md) + +# After a chronic illness diagnosis: the first week, the first three months, and after that + +This checklist is for people who have just been diagnosed with a chronic illness such as high blood pressure, diabetes, high blood lipids or gout. The items in this book that apply are spread across several sections. They are put in order of time here. Each line gives the action to take and then points at where it is written. Why it is worth doing, the numbers and the sources are all in the item being pointed at. If a line does not match your illness, skip it. + +## The week of the diagnosis + +1. If the illness is serious, have someone go with you on the day you collect the report, and do not stay at home alone afterwards. Put things like resigning or selling your flat off until later. See item 2 in section 29 (in the week you are told you have a serious illness). +2. If the doctor has prescribed medicine, start taking it as prescribed today, and do not stop when you begin to feel better. See item 1 in section 16 (take the medicine for the full course as the doctor ordered). +3. Take the medicine for high blood pressure or high blood lipids regularly. Do not stop it on your own. See item 12 in section 2 (if you have high blood pressure or high blood lipids, take the medicine as prescribed). +4. If you have high blood pressure, buy an electronic blood pressure monitor and measure at home, until the readings reach target. See item 7 in section 1 (measure your blood pressure, and take drugs to bring it down to target). +5. When the prescription is being written, ask the doctor whether there is a generic version picked in centralised procurement. See item 12 in section 5 (generic drugs that passed the consistency evaluation and drugs picked in centralised procurement). +6. If you are diagnosed with type 2 diabetes, have the back of your eyes examined with dilation at the time. If you have type 1 diabetes, start at 5 years after the onset. See item 7 in section 16 (once diabetes is diagnosed, have the back of your eyes examined once). +7. Once gout is diagnosed, start urate-lowering medicine for the long term and push the blood urate below 360 µmol/L. See item 9 in section 16 (once gout is diagnosed, take urate-lowering medicine for the long term). +8. Han people taking allopurinol: ask the doctor first whether the HLA-B*5801 genotype should be tested. See item 9 in section 16 (once gout is diagnosed, take urate-lowering medicine for the long term). +9. If a check-up found high uric acid and you have never had an attack, that does not count as gout. Do not start urate-lowering drugs yet. See item 19 in section 6 (do not start urate-lowering drugs because a check-up found high uric acid and you have never had an attack). +10. If you smoke, set a quit date and stop in one go on that day. See item 4 in section 2 (set a quit date and stop in one go on that day). +11. Do not try to quit on willpower alone. Go to a smoking cessation clinic or a respiratory department to get the medicine. See item 3 in section 2 (do not try to quit on willpower alone, get the medicine first). + +## The first three months + +1. Get a notebook. At each follow-up test write down the date, the figures, the medicine you are taking and the dose. See item 3 in section 16 (write the figures from each one in the same notebook). +2. After every visit, put away your own copy of the records, the test reports and the images. See item 5 in section 24 (keep your own copy of the records, the test reports and the images). +3. In the place where you are insured, get the outpatient chronic and special disease status done. If you often go to hospital out of town, do the cross-region filing as well. See item 2 in section 16 (get the outpatient chronic and special disease status first and the cross-region filing second). +4. If someone at home has employee medical insurance, ask them to do the family binding once. The money in their personal account can then pay for your treatment and your medicine. See item 13 in section 5 (bind your family once in the medical insurance app). +5. Before you sign up with a family doctor in the community, ask which items medical insurance covers and which you pay for yourself. See item 6 in section 16 (before you sign up with a family doctor in the community, ask which items are covered). +6. Replace the salt in your home with low-sodium salt. If your kidney function is impaired or you are taking a potassium-sparing drug, ask the doctor first. See item 9 in section 2 (replace the salt in your home with low-sodium salt). +7. Start walking every day. The target is 7000–8000 steps, or 150–300 minutes of brisk walking a week. See item 11 in section 2 (walk 7000–8000 steps a day). +8. Drink less alcohol. If you can manage not to drink at all, do not drink. See item 19 in section 2 (drink less alcohol, or none at all). +9. If you drink every day and stopping makes you shake and panic, do not stop on your own. Go to hospital first. See item 20 in section 2 (if you drink every day and stopping makes you shake and panic, do not stop on your own). +10. If somebody tells you to stop the medicine and try a folk remedy, do not stop. If you want to take a supplement, tell the doctor who is treating you first. See item 4 in section 16 (do not stop proper treatment in order to try a folk remedy or a health supplement). +11. Do not spend a lot of money on health supplements or herbal pastes to "look after your health". See item 10 in section 6 (do not spend a lot of money on health supplements and other health products, herbal pastes or tonics). +12. In the first 3 to 6 months of urate-lowering treatment, gout attacks are more likely. When it hurts, add an anti-inflammatory medicine as prescribed. Do not stop the urate-lowering medicine. See item 9 in section 16 (once gout is diagnosed, take urate-lowering medicine for the long term). +13. Before you buy a cold medicine or a painkiller yourself, read the ingredient list. Acetaminophen can only count once. See item 1 in section 34 (read the ingredient list: acetaminophen can only count once). +14. If you are taking anticoagulants or steroids, ask a doctor before taking a painkiller such as ibuprofen. See item 3 in section 34 (ask a doctor before taking a painkiller such as ibuprofen). + +## Once the illness is stable + +1. If the illness is stable and the medicine has not changed, get 12 weeks of it in one go from the community clinic. See item 5 in section 16 (a community health centre can prescribe up to 12 weeks of medicine at once). +2. Go for follow-up tests at the interval the doctor gives, and write each set of figures in the same notebook. See item 3 in section 16 (do your follow-up tests at the interval the doctor gives). +3. After you switch to a drug from centralised procurement, keep an eye on the figures in the notebook. If they really do get worse, go to the doctor. Do not stop the medicine on your own. See item 12 in section 5 (prefer generic drugs that passed the consistency evaluation and drugs picked in centralised procurement). +4. If you have diabetes, have your feet checked once a year, and the back of your eyes at the interval the doctor gives. See item 7 in section 16 (once diabetes is diagnosed, have the back of your eyes examined once). +5. Once your urate has reached target, keep taking the medicine. Do not stop. See item 9 in section 16 (once gout is diagnosed, take urate-lowering medicine for the long term). +6. If you have had a kidney stone, drink 2.5–3 litres of water a day and keep salt under 6 grams. See item 8 in section 16 (if you have had a kidney stone, drink 2.5 to 3 litres of water a day and keep salt under 6 grams). +7. If you have cardiovascular disease, have a flu vaccine in the autumn every year. See item 20 in section 1 (people with cardiovascular disease and older adults should get a flu vaccine every year). +8. For an ordinary illness, start at the community clinic. If you have to be admitted to hospital, go up by referral step by step through the community. The inpatient deductible carries on counting. See item 1 in section 24 (start at the community clinic and go up by referral step by step from the grassroots level). +9. Do not move up a level for an illness that can be dealt with at the grassroots level. In the same pooling region, each level you go up lowers the share reimbursed by about 10 percentage points. See item 2 in section 24 (the lower the level of the hospital the higher the share reimbursed). +10. If you need a large hospital for a specialist, go through a community referral or the hospital's referral centre. Do not buy appointments from scalpers. See item 3 in section 24 (go through a grassroots referral or the hospital's referral centre). +11. If you are planning to go to another province for treatment, ask at home before you set off. What comes back may be a good deal less than at home. See item 4 in section 24 (before going out of province for treatment, ask at home first). \ No newline at end of file diff --git "a/en/docs/\345\255\251\345\255\220\345\207\272\347\224\237\345\211\215\345\220\216\350\246\201\345\212\236\347\232\204\344\272\213.md" "b/en/docs/\345\255\251\345\255\220\345\207\272\347\224\237\345\211\215\345\220\216\350\246\201\345\212\236\347\232\204\344\272\213.md" new file mode 100644 index 00000000..0c7bf0b0 --- /dev/null +++ "b/en/docs/\345\255\251\345\255\220\345\207\272\347\224\237\345\211\215\345\220\216\350\246\201\345\212\236\347\232\204\344\272\213.md" @@ -0,0 +1,68 @@ +[← Back to the contents](../README.md) + +# What to do before and after a child is born: a checklist in order of time + +The book is arranged by subject, so everything about pregnancy, childbirth and looking after a baby is spread across several sections. This checklist puts it in order of time, from preparing to conceive through to the child's first birthday. Each line only says what to do. Why it is worth doing and how strong the evidence is, look at the item in the brackets. Start from the heading for the stage you have reached. + +## One: preparing to conceive + +1. Take one 0.4 mg folic acid tablet every day, starting 3 months before you conceive. See item 1 in section 27 (start 0.4 mg of folic acid every day before conception). +2. Telephone the social insurance or medical insurance office once and ask whether your months of continuous maternity insurance contributions are enough. See item 11 in section 27 (check your maternity insurance status before conception). +3. Tell the people you live with that nobody is going to smoke indoors from now on. See item 2 in section 2 (do not smoke at home or in the car). +4. If someone at home lives on your income, insure the earner first with term life insurance. See item 40 in section 5 (insure the earner first with term life insurance). + +## Two: after conception, before the birth + +1. Before 13 weeks of pregnancy, open a Mother and Child Health Handbook (《母子健康手册》) at the community health centre. That is where you pick up the free prenatal checks. See item 2 in section 27 (before 13 weeks of pregnancy open a Mother and Child Health Handbook). +2. At the first prenatal check, test for HIV, syphilis and hepatitis B. You do not pay for the tests yourself. See item 3 in section 27 (test for HIV, syphilis and hepatitis B at the first prenatal check). +3. Not one cigarette and not one drink for the whole of pregnancy. See item 4 in section 27 (not one cigarette and not one drink during pregnancy). +4. Give your full medical history at the prenatal check. If you have risk factors for pre-eclampsia, the doctor decides whether to start low-dose aspirin after 12 weeks of pregnancy. See item 5 in section 27 (if you have risk factors for pre-eclampsia, start one low-dose aspirin tablet a day). +5. At 24 weeks of pregnancy or later, have one screening for gestational diabetes. See item 6 in section 27 (have one screening for gestational diabetes at 24 weeks of pregnancy or later). +6. After 20 weeks of pregnancy, do not take a painkiller such as ibuprofen on your own. See item 5 in section 34 (do not take a painkiller such as ibuprofen on your own). +7. Learn the list of 「立刻去医院」 signs, and make sure the people you live with learn it too. See item 7 in section 27 (memorise this list of 「立刻去医院」 signs). +8. At a prenatal check, ask whether the hospital can give pain relief in labour. See item 9 in section 27 (if you want pain relief in labour, say so directly). +9. Without a medical indication, do not ask for a caesarean section. See item 10 in section 27 (without a medical indication do not ask for a caesarean section). +10. If you are not giving birth in the place where you are insured, ask in advance whether you need an out-of-area maternity filing. See item 11 in section 27 (check your maternity insurance status before conception). +11. Maternity leave is at least 98 days, and 15 days of it can be taken before the birth. See item 2 in section 18 (maternity leave is 98 days). +12. If your employer cuts your pay or moves you to another post because of the pregnancy, keep the written notice and the chat records. See item 3 in section 18 (your pay may not be cut and you may not be dismissed because of pregnancy). +13. Borrow the cot and the pram first, and buy second-hand if you cannot borrow them. Do not buy a car seat or a mattress second-hand. See item 11 in section 20 (for big items, work through borrow, then second-hand, then new). +14. Buy nappies in a small pack first and try them for a week. Only then buy in bulk. See item 10 in section 20 (judge nappies on three things rather than the brand). +15. Decide the name in advance and check that the household registration system can produce every character in it. See item 12 in section 27 (decide the name in advance and do not get a character wrong). + +## Three: during the hospital birth + +1. If your waters break, lie down flat where you are, raise your hips and call 120. Do not walk about. See item 8 in section 27 (if your waters break, lie down flat where you are, raise your hips and call 120). +2. Start feeding the baby within the first hour after the birth. See item 4 in section 20 (breastfeed only for the first 6 months, not even water). +3. Give the first hepatitis B shot within 24 hours of the birth. If the mother's hepatitis B surface antigen is positive, the child also gets the hepatitis B immune globulin. See item 2 in section 20 (give the first hepatitis B shot within 24 hours of birth). +4. The vitamin K shot has to be given. If you gave birth at home or at a place that is not a proper institution, ask about it yourself. See item 7 in section 20 (the vitamin K shot at birth is not optional). +5. Do not refuse either the heel-prick screening or the hearing screening. Before you sign the consent form, check which items you have to pay for yourself. See item 13 in section 27 (do not refuse either the newborn heel-prick screening or the hearing screening). + +## Four: before you leave the hospital + +1. Get the Medical Certificate of Birth (《出生医学证明》) done, and check the parents' and the child's names and identity numbers character by character. See item 12 in section 27 (get the Medical Certificate of Birth done before you leave the hospital). +2. Use the Medical Certificate of Birth to enrol the child in resident medical insurance online. You do not have to wait for the household registration. If you are not sure, telephone 12393 and ask. See item 14 in section 27 (enrol the child in resident medical insurance as soon as you have the Medical Certificate of Birth). +3. Use the car seat from the moment you start the journey home. Under 1 year old it goes in facing backwards, and the child is not held in your arms. See item 10 in section 1 (use a child car seat for children under 4). + +## Five: the first month after the birth + +1. Within one month, go to the police station to register the birth. See item 15 in section 27 (go to the police station to register the birth within one month of the child being born). +2. After the household registration is done, apply for the child-rearing subsidy where the child's hukou is. It is 3600 yuan per child per year. See item 1 in section 18 (the national child-rearing subsidy is 3600 yuan per child per year). +3. Put the baby to sleep on their back, on a firm flat surface, in the same room but not the same bed, with nothing soft on the bed. See item 1 in section 20 (put the baby to sleep on their back, on a firm flat surface). +4. If you are formula-feeding, use water above 70 ℃. Throw away what is left. See item 5 in section 20 (make up formula with water above 70 ℃). +5. A baby under 3 months old at 38 ℃ goes straight to hospital. See item 8 in section 20 (a baby under 3 months old at 38 ℃ goes straight to hospital). +6. When you are about to lose your temper, put the baby back in the cot and walk away for a few minutes. See item 9 in section 20 (never shake a baby, however tired and angry you are). +7. The list of 「立刻去医院」 signs for the mother counts for the whole of the first year after the birth. See item 7 in section 27 (memorise this list of 「立刻去医院」 signs). + +## Six: from the end of the first month to one year old + +1. Go to the 42-day postpartum check. It includes a screening for postpartum depression. See item 16 in section 27 (do not skip the 42-day postpartum check). +2. Go to the vaccination clinic for the national immunisation programme vaccines according to the child's age in months. The DTaP starts at 2 months. See item 3 in section 20 (complete every vaccine on the national immunisation programme). +3. Breastfeed only for the first 6 months. From 6 months add solid food, and every day include an iron-rich animal food. See item 4 in section 20 (from 6 months add solid food and keep breastfeeding). +4. No honey before 1 year old. See item 6 in section 20 (no honey before 1 year old). +5. If the baby has severe eczema or an egg allergy, see the doctor first and then decide whether to add peanuts early. See item 12 in section 20 (severe eczema or an egg allergy is not a reason to avoid peanuts). +6. When a child under 2 has a cold, do not give compound cold medicines on your own. See item 4 in section 34 (do not give a child under 2 compound cold medicines or cough medicines). +7. Fit limiters on the windows and the balcony. Window screens do not count as protection. See item 11 in section 1 (if there are children at home, fit window and balcony limiters). +8. No phones, tablets or other screens from 0 to 3 years old. See item 5 in section 30 (no screens from 0 to 3 years old). +9. After you go back to work, if the child is not yet one year old, you get 1 hour a day for breastfeeding. See item 3 in section 18 (your pay may not be cut and you may not be dismissed because of pregnancy). +10. Bind the employee medical insurance personal account to the family, and the child's treatment and medicine can be paid from it. See item 13 in section 5 (bind your family once in the medical insurance app). +11. At the annual individual income tax settlement, fill in the deduction for care of infants and toddlers under 3. See item 2 in section 5 (claim every special additional deduction). \ No newline at end of file diff --git "a/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" "b/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" index 8bf3e720..749a0a1c 100644 --- "a/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" +++ "b/en/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" @@ -2,7 +2,7 @@ # How the body tells the time, and why night shifts hurt -This is the long-form version of item 40 in section 2 (the longer you work night shifts, the higher your cardiovascular risk). That item only works out the arithmetic. This essay sets out the reasoning behind it. +This is the long-form version of item 39 in section 2 (the longer you work night shifts, the higher your cardiovascular risk). That item only works out the arithmetic. This essay sets out the reasoning behind it. The question to answer is whether the body really does know what time it is. If it does, why is working night shifts not a matter of getting used to it after a few goes? @@ -140,9 +140,9 @@ The two measures above end at metabolism and cardiovascular measures on one side ## 9. What this essay does not cover -How much higher night shifts make cardiovascular risk, and how it is worked out by years, is in item 40 in section 2 (the longer you work night shifts, the higher your cardiovascular risk). +How much higher night shifts make cardiovascular risk, and how it is worked out by years, is in item 39 in section 2 (the longer you work night shifts, the higher your cardiovascular risk). -How to make up the sleep after a late night is in item 39 in section 2 (after a late night, make up the sleep the next evening, do not bank it up for the weekend). +How to make up the sleep after a late night is in item 38 in section 2 (after a late night, make up the sleep the next evening, do not bank it up for the weekend). How long you should sleep at night and how regular the schedule should be is in item 13 in section 2 (sleep about 7 hours a night, keep the schedule fixed). diff --git "a/en/docs/\350\242\253\350\243\201\344\272\206\344\271\213\345\220\216\345\205\210\345\201\232\344\273\200\344\271\210.md" "b/en/docs/\350\242\253\350\243\201\344\272\206\344\271\213\345\220\216\345\205\210\345\201\232\344\273\200\344\271\210.md" new file mode 100644 index 00000000..53bc88ed --- /dev/null +++ "b/en/docs/\350\242\253\350\243\201\344\272\206\344\271\213\345\220\216\345\205\210\345\201\232\344\273\200\344\271\210.md" @@ -0,0 +1,54 @@ +[← Back to the contents](../README.md) + +# What to do first after being laid off + +What has to be done after a layoff or a dismissal is spread across section 19, section 7 and several other sections. It is put in order of time below. Each step only says what to do. The details, the numbers and the reasoning are all in the item pointed at just before the brackets. If you have only been underpaid and are still working, go straight to item 2 in section 7 (if your wages are owed to you, complain to labour inspection first). + +## The same day, before you sign anything + +1. If the company hands you a form saying 「resignation for personal reasons」, do not sign it. Sign it and the economic compensation and the unemployment insurance benefit usually both go with it. See item 7 in section 19 (do not sign 「resignation for personal reasons」). +2. If the company says 「sign first so the paperwork moves along, the compensation comes separately」, ask for the compensation to be written into the agreement on the spot. If they will not write it down, do not sign. Not signing on the day carries no legal consequence. See item 7 in section 19 (do not sign 「resignation for personal reasons」). +3. Turn on the recorder during the conversation. For a conversation you are party to, you do not need the other person's consent first. See item 41 in section 8 (turn on the recorder for phone calls and meetings that could turn). +4. Before you hand back the computer and the accounts, save your payslips, attendance records, employment contract, social insurance record and chat logs. See item 8 in section 19 (before you leave, save your payslips, attendance records, employment contract). +5. Save only the documents relating to your own employment. Do not take the company's source code, customer lists or technical documents away. See item 7 in section 11 (do not take source code, customer lists or technical documents away when you leave). +6. Hand over your account access. Do not delete the data and do not change the password, even if the company owes you money. See item 6 in section 11 (hand over your account access when you leave). + +## The same day to the first few days: work out the money + +1. Work out N first: one month's wage for each full year, and half a month for under six months. See item 4 in section 19 (work out N before a dismissal). +2. The monthly wage is worked out from the average over the 12 months before you leave. Bonuses, allowances and subsidies all count in. This is the part a company is most likely to cut. See item 4 in section 19 (work out N before a dismissal). +3. If the company did not give you 30 days' written notice, it has to pay one month's wage on top of N. Put that month down separately when you negotiate. See item 5 in section 19 (if the company dismisses you without giving 30 days' notice). +4. If the company terminated the contract unlawfully, the compensation is 2N. Ask legal aid or a lawyer whether it was unlawful before you decide whether to negotiate or go to arbitration. See item 6 in section 19 (the compensation is twice the economic compensation standard). +5. Annual leave you did not get through is settled off when you leave, worked out by how much of the year you have already done. See item 2 in section 19 (annual leave is 5, 10 or 15 days according to total length of service). +6. Add the unpaid overtime pay to the same figure. See item 1 in section 19 (overtime pay runs in three bands at 1.5, 2 and 3 times). + +## The first week + +1. Go and register as unemployed. The unemployment insurance benefit is counted from the day you register, so a month late means a month less. See item 1 in section 7 (if you have lost your job, apply online first for unemployment insurance benefit). +2. The registration is done at a public employment service office such as an employment service centre. It is free. See item 5 in section 7 (to find work, start with free public employment services). +3. Apply for the unemployment insurance benefit on the 「掌上 12333」 app or the electronic social insurance card mini-programme. You can only get it if you have paid in for a year and you did not resign of your own accord. See item 1 in section 7 (if you have lost your job, apply online first for unemployment insurance benefit). +4. While you are on the unemployment insurance benefit, the unemployment insurance pays your employee medical insurance. You do not have to enrol separately. See item 1 in section 7 (if you have lost your job, apply online first for unemployment insurance benefit). +5. If you cannot afford to sue, telephone 12348 and apply for legal aid. You no longer have to go back to your home district for a certificate of financial difficulty. See item 3 in section 7 (if you cannot afford to sue, apply for legal aid). +6. If you signed a non-compete, keep the original. The company has to pay the compensation monthly, and if it does not pay for 3 months for reasons of its own you can ask for it to be terminated. See item 12 in section 11 (if you have signed a non-compete). +7. Fix your daily routine: get up at the same time, put the job search into the diary at fixed times, and do not spend the whole day at home. See item 3 in section 29 (after losing your job, fix your daily routine). + +## The first month + +1. If wages or overtime pay are being withheld, telephone 12333 and complain to labour inspection first, then apply for labour arbitration. Neither route costs anything. See item 2 in section 7 (if your wages are owed to you, complain to labour inspection first). +2. If an unlicensed workshop employed you, go to labour inspection and arbitration all the same. Money owed by a foreman is chased to the main contractor. Only work done privately for a household or an individual goes to court. See item 22 in section 7 (if you are owed work pay, first find out who employed you). +3. The labour arbitration has to be started within 1 year of leaving. After that the other side only has to say 「too late」. See item 19 in section 8 (1 year for labour arbitration). +4. To find work, start with the public employment service offices. Job placement and career guidance are both free. Do not go to an agency that charges. See item 5 in section 7 (to find work, start with free public employment services). +5. Do not only send out resumes when you job hunt. Learn the technique, set a definite target, and ask people you know to pass your name on. See item 14 in section 7 (when job hunting, do not only send out resumes; use a system). +6. If you hear 「pay first, then start work」, 「train first, then a job is arranged」 or 「recruit people」, refuse on the spot. See item 15 in section 7 (do not sign a "training loan"). +7. If you want to learn a skill, look up the local human resources and social security bureau's 「vocational training subsidy catalogue」 first. Do not pay for a course yourself. See item 13 in section 7 (while unemployed, go and claim the vocational training subsidy). +8. See whether you qualify for recognition as a person with difficulty finding work, for example on grounds of age or long-term unemployment. Once you have it, your own social insurance contributions carry a subsidy, and you can also queue for a public-interest post. See item 12 in section 7 (after registering as unemployed, seek recognition as a person with difficulty finding work). + +## The months after that + +1. Cover living costs out of the emergency fund first. Do not borrow, and do not sell investments at a loss. See item 27 in section 5 (save 3 to 6 months of living costs as an emergency fund). +2. In the first three months, put every irreversible big decision off. Selling your flat, investing the whole severance payment in one go, going into a partnership and standing guarantor for someone all count. See item 12 in section 29 (in the first three months after a shock, put every irreversible big decision off). +3. If the unemployment insurance benefit has run out and you still have no work, do not let the medical insurance lapse. You can pay for resident medical insurance yourself year by year, but after a lapse there is a 3-month period during which treatment is not reimbursed. See item 9 in section 7 (do not let the 400 yuan a year of resident medical insurance lapse). +4. A gap in the pension does not reset it to zero. It is counted by the years you have accumulated. Do not pay somebody to keep it going for you. See item 18 in section 7 (do not panic when social insurance contributions lapse). +5. If the household's income falls below the local means-tested income support line, apply for it. See item 7 in section 7 (if your income is below the local means-tested income support line, apply for means-tested income support). +6. If you cannot hold on and you want somebody to talk to, telephone 12356. It is free. See item 11 in section 29 (to have somebody to talk to, call 12356). +7. For money you have still not got back, watch the one-year limit counted from the day you left. See item 19 in section 8 (1 year for labour arbitration). \ No newline at end of file diff --git a/tools/i18n/README.md b/tools/i18n/README.md index d96f95c1..bd81f53a 100644 --- a/tools/i18n/README.md +++ b/tools/i18n/README.md @@ -65,9 +65,66 @@ node tools/site/build.mjs | `join-ready.sh` | 一轮把两种语言里所有「已经翻全」的节挨个 join 掉，省得一次次敲 | | `structure.mjs` | 一节（或一块）的结构骨架 + 逐条比对。`join` 和 `check` 共用同一份判据 | | `check.mjs` | 拿中文原文对全部译文，CI 的「译文检查」job 跑它 | +| `check-docs.mjs` | 长文（`docs/`）里的条目指路有没有跟着原文的条号走，CI 也跑 | +| `rebase.mjs` | 中文改了之后把已有译文搬到新条号上（删条并条之后用这个，见下） | +| `diff-scope.mjs` | 每节的删/改/增清单，用来估工作量 | +| `diff-delta.mjs` | 真改了的条目拆到句子级，按「只加了一句」「增删都有」「整段重写」分三档 | +| `put.mjs` | 把一条译文写进它该在的块，按条号找，找不到就退出码 1 | | `anchors.mjs` | 译本里逐字照抄的那几样（成本标签注释、字段顺序、不许动的数字、指路、链接） | | `glossary.{en,vi}.mjs` | 词表。同一件事全书一个说法，靠它 | +## 中文原文改了之后（rebase） + +作者改了中文（自己写的，或者 merge 上游），条目数变了、条号顺延了，已有的译文还停在旧的条号上。 +这时候**不能直接跑 `split.mjs` 或 `resplit.mjs`**：删条并条之后条号整体顺延，按条号配会把 +译文挂到别的条目上，而且不报错——结构检查只看条号在不在范围内，顺延后的号仍在范围内 +（上游 2026-10 那次删了 9 条、第 2 节两条并成一条）。 + +```bash +git archive <改动前的 commit> book | tar -x -C /tmp/oldbook +node tools/i18n/diff-scope.mjs /tmp/oldbook/book en # 看看要动多少 +node tools/i18n/rebase.mjs /tmp/oldbook/book en # 搬（加 --dry 只报不写） +node tools/i18n/rebase.mjs /tmp/oldbook/book vi +bash tools/i18n/join-ready.sh en vi # 搬完合一趟 +node tools/i18n/check.mjs && node tools/i18n/check-docs.mjs +``` + +`rebase.mjs` 做三件事： + +1. **按条目标题配对**（标题是作者写的动作句，改写之外一般不动；全书没有两条标题相同）。 + 配不上的、标题变了的、正文真变了的，那几条译文留在中文等着人翻——`join-ready.sh` + 会因为块里还有 `- 成本：` 而跳过那一节，这是设计，不是坏了。 +2. **改条号和指路条号**。删条之后中文原文的「第 20 条」变成「第 19 条」，译文的 + `item 20` 不同改就是指错条目，而 `check.mjs` 拿的是译文的指路跟中文原文逐个比。 +3. **把搬不动的报出来**，一节一节点名，不猜：上游删条并条后没跟着改的指路、跨删除点的 + 区间引用、跟着原文改了内容的条目、节标题或导读真改了的节。 + +**改指路要卡住「这是不是条目引用」。** 英文的 `xref.pattern` 里有一条光秃秃的 +`\bitems?\s+(\d+)`，来源栏里「劳部发〔1994〕309 号，第 59 条」译成 `item 59` 照样命中， +APA 的 `item 705` 也命中。照单全改就是把法条条款号当条目号，指到别的条目上去。卡口是拿 +**旧原文**那一条的指路当白名单（译文里的号是旧号，是不是条目引用也只有旧原文说了算）； +拿新原文的号去卡旧号，删条之后两边对不上，译文里的旧号一个都不改，比报错更坏—— +指路静默指错还不报。 + +**跑之前先确认没有别的活儿在动块。** `rebase.mjs` 会重写 `.i18n/<语言>/**` 全部块文件， +翻译任务进行中跑它会把已翻的条目一起盖回中文（2026-10-02 有子代理这么干过一轮， +en、vi 各有十几个条目被冲掉要重翻）。翻译任务并行的时候只让它改块，别同时跑 rebase。 + +## 长文（docs/）不在 check.mjs 的扫描范围里 + +`check.mjs` 逐条比的是 `book/` 里的条目。长文（`docs/*.md`）是另一批文件，全是条目指路， +`check.mjs` 一个字都不看——「孩子出生前后要办的事」那种按时间排的动作清单，引用指错条目， +读者照着做的那一步就错了，比正文里指错更难受。单独用 `check-docs.mjs` 查：拿译文引的 +(节, 条) 集合跟中文原文比，多了少了都报。 + +判据只比集合，不逐字比锚点：锚点是给人看的，长文里本来就允许截断（中文原文自己就在截， +「药按医嘱吃满」是「药按医嘱吃满，别感觉好了就停」的截断），译文截断的地方和长度都不一样， +逐字比只会报出一堆假警报。要拦的是条号顺延——译文那一份还在引旧号，读者点进去落在隔壁 +那条上——集合一比就露馅，而且不依赖译文的措辞。括号里的锚点对不上得人看。 + +2026-10-02 查出过一处真错：上游在第 2 节删条并条之后，正文这边 `rebase.mjs` 搬了，长文这边 +没人搬，`docs/生物钟和夜班.md` 的「第 2 节第 39 条」在英越两语还写着 item 40，而 2.40 +已经是「买预包装食用油」。已改，并把这个检查挂进 CI 的「译文检查」job。 ## 为什么译文不用人审 `check.mjs` 查的都是「翻着翻着丢了东西」：节数条数、六个字段的有无与顺序、成本标签注释、 diff --git a/tools/i18n/check-docs.mjs b/tools/i18n/check-docs.mjs new file mode 100644 index 00000000..793216d9 --- /dev/null +++ b/tools/i18n/check-docs.mjs @@ -0,0 +1,74 @@ +// 校长文（docs/*.md）里的条目指路：译文引的条号节号跟中文原文还对不对得上。 +// +// node tools/i18n/check-docs.mjs # 两种语言都查 +// node tools/i18n/check-docs.mjs en +// +// 为什么单独一个：check.mjs 逐条比 book/ 里的条目，不看 docs/。长文全是条目引用， +// 而且是按时间排的动作清单——引用指错条目，读者照着做的那一步就错了，比正文里 +// 指错更难受。上游删条并条之后条号顺延，正文这边有 rebase.mjs 搬，长文这边没人搬： +// 2026-10-02 合成后 docs/生物钟和夜班.md 的「第 2 节第 39 条」在英越两语还写着 +// item 40，而 2.40 已经是「买预包装食用油」，夜班那条落到 2.39 上去了。 +// +// 判据是「两边引的 (节, 条) 集合相同」，不是逐字比锚点： +// · 锚点是给人看的，长文里本来就允许截断（中文原文自己就在截，「药按医嘱吃满」 +// 是「药按医嘱吃满，别感觉好了就停」的截断），译文截断的地方和截断的长度 +// 都不一样，逐字比只会报出一堆假警报。 +// · 真正要拦的是条号顺延：译文那一份还在引旧号，读者点进去落在隔壁那条上。 +// 集合一比就露馅，而且不依赖译文的措辞。 +// 所以这里只收「译文引了原文没引的」和「原文引了译文没引的」两种。 +// 括号里的锚点对不对得上，得人看，见 check-refs 对正文的同一套办法。 + +import { readFileSync, existsSync, readdirSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; +import { LOCALES, SOURCE, strings } from '../site/locales.mjs'; + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); + +const argv = process.argv.slice(2); +const locales = argv.length ? argv : LOCALES.map(l => l.code).filter(c => c !== SOURCE.code); + +/** 一门语言的长文里，引用到的 (节, 条) 集合。只认带节号的那一支：长文里裸的 + 「第 N 条」是法条，不是条目引用（CLAUDE.md 记的规矩）。 */ +function refsOfDoc(text, xref) { + const out = new Set(); + for (const m of String(text).matchAll(new RegExp(xref.pattern, 'g'))){ + const g = m.groups ?? {}; + if (g.sec === undefined || !g.nums) continue; + for (const part of String(g.nums).split(new RegExp(xref.listSep))){ + const n = part.trim(); + if (/^\d+$/.test(n)) out.add(g.sec + ':' + n); + } + } + return out; +} + +const zhDir = (SOURCE.contentDir ? SOURCE.contentDir + '/' : '') + 'docs/'; +let bad = 0, files = 0, pairs = 0; + +for (const locale of locales) { + const L = LOCALES.find(l => l.code === locale); + const S = strings(locale); + if (!S.xref) continue; + const docsDir = (L.contentDir ? L.contentDir + '/' : '') + 'docs/'; + if (!existsSync(resolve(ROOT, docsDir))) continue; + + for (const f of readdirSync(resolve(ROOT, docsDir)).filter(x => x.endsWith('.md')).sort()) { + if (!existsSync(resolve(ROOT, zhDir + f))) continue; // 原文还没有这篇，不比 + const want = refsOfDoc(read(zhDir + f), strings(SOURCE.code).xref); + if (!want.size) continue; // 原文没引条目 + const got = refsOfDoc(read(docsDir + f), S.xref); + files++; + pairs += want.size; + const extra = [...got].filter(x => !want.has(x)); + const missing = [...want].filter(x => !got.has(x)); + if (!extra.length && !missing.length) continue; + bad++; + console.log(`${locale} docs/${f}`); + if (extra.length) console.log(` 译文多引了（原文没有）：${extra.join(' ')}`); + if (missing.length) console.log(` 译文漏引了（原文有）：${missing.join(' ')}`); + } +} +console.log(`\n长文指路：${files} 篇共 ${pairs} 处引用，对不上 ${bad} 篇`); +process.exit(bad ? 1 : 0); \ No newline at end of file diff --git a/tools/i18n/diff-delta.mjs b/tools/i18n/diff-delta.mjs new file mode 100644 index 00000000..93fb8357 --- /dev/null +++ b/tools/i18n/diff-delta.mjs @@ -0,0 +1,164 @@ +// 把「正文真改了」的条目拆到句子级：报出每条多了一句、删了一句还是整段换了， +// 好按工作量分派翻译任务。 +// +// node tools/i18n/diff-delta.mjs <旧 book 目录> +// node tools/i18n/diff-delta.mjs <旧 book 目录> en +// +// 判据和 rebase.mjs 一致：按标题配对，按映射改过指路条号之后逐字比，还是不一样 +// 就是真改了。输出按「只加了一句」「只删了一句」「整段重写」三档归类。 + +import { readFileSync, existsSync, readdirSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; +import { LOCALES, SOURCE, strings } from '../site/locales.mjs'; +import { refsOf } from './structure.mjs'; + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); + +const argv = process.argv.slice(2); +const baseDir = argv[0]; +const only = argv[1] ?? null; // 'en' / 'vi' / null（全看） +if (!baseDir || !existsSync(resolve(ROOT, baseDir))) { + console.error('用法：node tools/i18n/diff-delta.mjs <旧 book 目录> [en|vi]'); + process.exit(2); +} + +function pieces(text) { + const lines = text.split('\n'); + const at = lines.map((l, i) => (/^### /.test(l) ? i : -1)).filter(i => i >= 0); + const byN = new Map(); + for (let k = 0; k < at.length; k++) { + const stop = k + 1 < at.length ? at[k + 1] : lines.length; + const seg = lines.slice(at[k], stop); + while (seg.length && !seg[seg.length - 1].trim()) seg.pop(); + byN.set(/^### (\d+)\./.exec(seg[0])[1], seg); + } + return byN; +} + +/** 一条拆成 { 字段名: 内容 }，注释行和标题行单放 */ +function fields(seg, S) { + const out = { __tag: '' }; + for (const l of seg) { + const tag = /^$/.exec(l); + if (tag) { out.__tag = tag[1]; continue; } + if (/^### /.test(l)) { out.__title = l; continue; } + for (const [k, name] of Object.entries(S.fields)) { + const m = new RegExp('^- ' + name.replace(/[：:]$/, '') + '[：:](.*)$').exec(l); + if (m) { out[k] = m[1]; break; } + } + } + return out; +} + +/** 中文字符串切句：句末标点加下一段引号/括注收尾。够用就行，判据是人读。 */ +const sentences = s => String(s ?? '') + .split(/(?<=[。；！？])/) + .map(x => x.trim()) + .filter(Boolean); + +/** 按句配对：旧句里出现在新句中的算没动，剩下的就是删的/加的 */ +function delta(oldS, newS) { + const setOld = new Set(sentences(oldS)); + const setNew = new Set(sentences(newS)); + return { + added: sentences(newS).filter(x => !setOld.has(x)), + removed: sentences(oldS).filter(x => !setNew.has(x)), + }; +} + +const zhS = strings(SOURCE.code); +const srcDir = (SOURCE.contentDir ? SOURCE.contentDir + '/' : '') + 'book/'; +const secs = readdirSync(resolve(ROOT, srcDir)).filter(x => x.endsWith('.md')).sort(); +const secNames = secs.map(f => f.replace(/\.md$/, '')); + +const buckets = { addOnly: [], mixed: [], rewrite: [] }; +let total = 0; + +for (const f of secs) { + const baseFile = `${baseDir}/${f}`; + if (!existsSync(resolve(ROOT, baseFile))) continue; + const oldE = pieces(read(baseFile)); + const newE = pieces(read(srcDir + f)); + const name = f.replace(/\.md$/, ''); + const map = new Map(); + // 先按标题配一遍，得出本节 old->new + const byTitle = new Map(); + for (const [n, seg] of oldE) { + const t = seg[0].replace(/^### \d+\.\s*/, '').trim(); + if (!byTitle.has(t)) byTitle.set(t, []); + byTitle.get(t).push({ n, seg }); + } + const claimed = new Set(); + const pairs = []; + for (const [n, seg] of newE) { + const t = seg[0].replace(/^### \d+\.\s*/, '').trim(); + const hit = (byTitle.get(t) || []).find(x => !claimed.has(x.seg)); + if (!hit) continue; + claimed.add(hit.seg); pairs.push([n, hit.n, seg, hit.seg]); + } + for (const [n, o, seg] of pairs) map.set(o, n); + + // 指路条号改写（中文原文自己那一份） + const renum = t => { + let out = '', last = 0; + for (const m of String(t).matchAll(new RegExp(zhS.xref.pattern, 'g'))) { + const g = m.groups ?? {}; + const key = g.nums !== undefined ? 'nums' : g.nums2 !== undefined ? 'nums2' : g.nums3 !== undefined ? 'nums3' : null; + if (!key) continue; + const allow = new Set(refsOf(t, zhS.xref).map(r => (r.startsWith('s') ? r.slice(1) : r.startsWith('*') ? r.slice(1) : null)).filter(Boolean)); + const span = g[key]; + const sub = span.replace(/\d+/g, d => (allow.has(d) && map.get(d) ? String(map.get(d)) : d)); + out += String(t).slice(last, m.index) + m[0].replace(span, sub); + last = m.index + m[0].length; + } + return out + String(t).slice(last); + }; + + for (const [n, o, newSeg, oldSeg] of pairs) { + const nf = fields(newSeg, zhS), of = fields(oldSeg, zhS); + // 逐栏比：只差指路条号的不算改动 + const diffFields = []; + for (const k of new Set([...Object.keys(nf), ...Object.keys(of)])) { + if (k === '__title') continue; + const a = renum(of[k] ?? '').replace(/\s+/g, ''); + const b = (nf[k] ?? '').replace(/\s+/g, ''); + if (a !== b) diffFields.push(k); + } + if (!diffFields.length) continue; + total++; + + let addOnly = true, big = false; + const detail = []; + for (const k of diffFields) { + const d = delta(renum(of[k] ?? ''), nf[k] ?? ''); + if (d.removed.length) addOnly = false; + if (d.added.some(x => x.length > 60)) big = true; + detail.push({ k, ...d }); + } + const label = `${name} 第 ${n} 条`; + const row = { label, fields: detail }; + if (addOnly && !big) buckets.addOnly.push(row); + else if (big || d2heavy(detail)) buckets.rewrite.push(row); + else buckets.mixed.push(row); + } +} + +/** 增删都有的，且删掉的句子不短——多半是整段重写，人读一遍比逐句补更省事 */ +function d2heavy(detail) { + return detail.some(d => d.removed.some(x => x.length > 40) && d.added.some(x => x.length > 40)); +} + +for (const [k, v] of Object.entries(buckets)) { + console.log(`\n=== ${k}（${v.length} 条）===`); + for (const r of v) { + console.log(`\n${r.label}`); + for (const d of r.fields) { + console.log(` [${d.k}]`); + for (const a of d.added) console.log(` + ${a.slice(0, 110)}`); + for (const a of d.removed) console.log(` - ${a.slice(0, 110)}`); + } + } +} +console.log(`\n正文真改了 ${total} 条：只加 ${buckets.addOnly.length}，增删都有 ${buckets.mixed.length}，整段重写 ${buckets.rewrite.length}`); \ No newline at end of file diff --git a/tools/i18n/diff-scope.mjs b/tools/i18n/diff-scope.mjs new file mode 100644 index 00000000..41dfbca3 --- /dev/null +++ b/tools/i18n/diff-scope.mjs @@ -0,0 +1,100 @@ +// 报告中文原文改了什么，好知道哪些译文要重翻：拿旧原文和现原文逐条配对。 +// +// node tools/i18n/diff-scope.mjs <旧 book 目录> # 两种语言 +// node tools/i18n/diff-scope.mjs <旧 book 目录> en +// +// 配对不按条号（上游删条并条，条号会整体顺延，按号配会把译文挂到别的条目上）， +// 按条目标题配：标题是作者写的动作句，删条并条之外的改动一般不动它。标题没命中的 +// 再按整段正文的前 60 字配。全书没有两条标题相同，所以标题是可靠的键。 +// +// 三类结果：删掉的（译文整条去掉，条号顺延）、标题变了的（同一件事被改写，要重翻）、 +// 新增的（要翻）。标题和正文都没变的，译文原样留着，只需改条号和交叉引用。 + +import { readFileSync, existsSync, readdirSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; +import { LOCALES, SOURCE } from '../site/locales.mjs'; + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); + +const argv = process.argv.slice(2); +const baseDir = argv[0]; +const locales = argv[1] ? [argv[1]] : LOCALES.map(l => l.code).filter(c => c !== SOURCE.code); +if (!baseDir || !existsSync(resolve(ROOT, baseDir))) { + console.error('用法：node tools/i18n/diff-scope.mjs <旧 book 目录> [en|vi]'); + process.exit(2); +} + +/** 一份正文拆成按条号索引的条目 */ +function pieces(text) { + const lines = text.split('\n'); + const at = lines.map((l, i) => (/^### /.test(l) ? i : -1)).filter(i => i >= 0); + const byN = new Map(); + for (let k = 0; k < at.length; k++) { + const stop = k + 1 < at.length ? at[k + 1] : lines.length; + const seg = lines.slice(at[k], stop); + while (seg.length && !seg[seg.length - 1].trim()) seg.pop(); + byN.set(/^### (\d+)\./.exec(seg[0])[1], { head: seg[0], body: seg.slice(1).join('\n') }); + } + return byN; +} + +const titleOf = h => h.replace(/^### \d+\.\s*/, '').trim(); +const keyOf = e => e.body.replace(/\s+/g, '').slice(0, 60); + +const totals = { same: 0, reword: 0, fresh: 0, gone: 0 }; +const perSection = []; + +for (const code of locales) { + const locale = LOCALES.find(l => l.code === code); + const tgtDir = `${locale.contentDir ? locale.contentDir + '/' : ''}book/`; + const srcDir = (SOURCE.contentDir ? SOURCE.contentDir + '/' : '') + 'book/'; + for (const f of readdirSync(resolve(ROOT, srcDir)).filter(x => x.endsWith('.md')).sort()) { + const baseFile = `${baseDir}/${f}`; + if (!existsSync(resolve(ROOT, baseFile))) continue; + if (!existsSync(resolve(ROOT, tgtDir + f))) continue; + const oldE = pieces(read(baseFile)); + const newE = pieces(read(srcDir + f)); + + // 旧条目按标题和按正文开头各建一张表。claims 记已配上的， + // 免得两条旧条目抢同一条新条目（并条之后标题会撞上一次）。 + const byTitle = new Map(), byKey = new Map(); + for (const [n, e] of oldE) { + const t = titleOf(e.head); + if (!byTitle.has(t)) byTitle.set(t, []); + byTitle.get(t).push({ n, e }); + if (!byKey.has(keyOf(e))) byKey.set(keyOf(e), []); + byKey.get(keyOf(e)).push({ n, e }); + } + const claimed = new Set(); + + const gone = [], reword = [], fresh = []; + for (const [n, e] of newE) { + const t = titleOf(e.head); + let hit = (byTitle.get(t) || []).find(x => !claimed.has(x.e)); + if (hit) { + claimed.add(hit.e); + if (keyOf(hit.e) !== keyOf(e)) { reword.push(`${n}（旧 ${hit.n}）`); totals.reword++; } + else totals.same++; + continue; + } + hit = (byKey.get(keyOf(e)) || []).find(x => !claimed.has(x.e)); + if (hit) { claimed.add(hit.e); totals.same++; continue; } + fresh.push(n); totals.fresh++; + } + for (const [n, e] of oldE) if (!claimed.has(e)) { gone.push(`${n}「${titleOf(e.head).slice(0, 26)}」`); totals.gone++; } + + if (gone.length || reword.length || fresh.length) { + perSection.push({ code, f, gone, reword, fresh }); + } + } +} + +for (const { code, f, gone, reword, fresh } of perSection) { + console.log(`\n${code} ${f}`); + if (gone.length) console.log(` 删掉 ${gone.length}：${gone.join(' ')}`); + if (reword.length) console.log(` 改写 ${reword.length}：${reword.join(' ')}`); + if (fresh.length) console.log(` 新增 ${fresh.length}：第 ${fresh.join('、')} 条`); +} +console.log(`\n合计：正文与标题都没变 ${totals.same}，标题变（要重翻）${totals.reword}，新增（要翻）${totals.fresh}，删掉（译文去掉、条号顺延）${totals.gone}`); diff --git a/tools/i18n/put.mjs b/tools/i18n/put.mjs new file mode 100644 index 00000000..fd62bf31 --- /dev/null +++ b/tools/i18n/put.mjs @@ -0,0 +1,51 @@ +// 把一条译文写进它该在的块。 +// +// node tools/i18n/put.mjs <语言> <节名> <条号> <译文文件> +// +// 译文文件是完整的条目，含 `### N. ` 那一行和下面六栏。 +// +// 为什么要有这个脚本：条目散在 .i18n/<语言>/<节名>/partNN.md 里，哪一条在第几块 +// 靠数行号很容易数错（2026-10-02 数错了一次，把第 19 条的译文写进了第 13 条那块， +// 脚本直接抛错才没写坏）。这个脚本按条号去找，找不到就退出码 1，一个字都不写。 +import { readFileSync, writeFileSync, existsSync, readdirSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8'); +const write = (p, t) => writeFileSync(resolve(ROOT, p), t); + +const [locale, sec, numStr, ...rest] = process.argv.slice(2); +const file = rest[0]; +if (!locale || !sec || !numStr || !file) { + console.error('用法：node tools/i18n/put.mjs <语言> <节名> <条号> <译文文件>'); + process.exit(2); +} +if (!existsSync(resolve(ROOT, file))) { console.error(`译文文件不存在：${file}`); process.exit(2); } + +const body = read(file).replace(/\n+$/, '\n'); +const dir = `.i18n/${locale}/${sec}`; +// 先确认译文那一行的条号跟参数一致，写错节号要当场拦住 +const head = /^### (\d+)\. /.exec(body); +if (!head) { console.error(`${file} 开头不是「### N. …」`); process.exit(2); } +if (head[1] !== numStr) { + console.error(`译文文件写的是第 ${head[1]} 条，参数写的是第 ${numStr} 条`); + process.exit(2); +} + +let hit = null; +for (const f of readdirSync(resolve(ROOT, dir)).filter(x => /^part\d+\.md$/.test(x)).sort()){ + const p = `${dir}/${f}`; + const t = read(p); + if (!new RegExp(`^### ${numStr}\\. `, 'm').test(t)) continue; + if (hit) { console.error(`第 ${numStr} 条同时出现在 ${hit} 和 ${p}，先去查块是怎么分的`); process.exit(1); } + hit = { p, t }; +} +if (!hit) { console.error(`${dir} 里没有第 ${numStr} 条`); process.exit(1); } + +const lines = hit.t.split('\n'); +const at = lines.findIndex(l => l.startsWith(`### ${numStr}. `)); +let stop = lines.length; +for (let i = at + 1; i < lines.length; i++) if (/^### \d+\. /.test(lines[i])) { stop = i; break; } +write(hit.p, lines.slice(0, at).concat(body.replace(/\n$/, ''), lines.slice(stop)).join('\n')); +console.log(`${locale} ${sec} 第 ${numStr} 条 → ${hit.p}`); diff --git a/tools/i18n/rebase.mjs b/tools/i18n/rebase.mjs new file mode 100644 index 00000000..32845085 --- /dev/null +++ b/tools/i18n/rebase.mjs @@ -0,0 +1,320 @@ +// 上游改了中文原文之后，把已有的译文搬到新的条号上。 +// +// node tools/i18n/rebase.mjs <旧 book 目录> # 两种语言 +// node tools/i18n/rebase.mjs <旧 book 目录> en +// node tools/i18n/rebase.mjs <旧 book 目录> --dry # 只报不写 +// +// 为什么不能直接跑 split.mjs：删条并条之后条号整体顺延，按条号配会把译文挂到 +// 别的条目上（上游这次删了 9 条、第 2 节两条并成一条）。resplit.mjs 也是按条号 +// 配的，并条之后会把两条并成一条的译文拆错。这个脚本按条目标题配——标题是作者 +// 写的动作句，改写之外一般不动。 +// +// 搬的时候要改两处，两处都会让译文对不上： +// ① 条号本身（### 12. → ### 11.）。 +// ② 译文里的指路条号。中文原文第 2 节删了一条，原文里「第 2 节第 30 条」跟着改成 +// 第 29 条，译文里的 "item 30 in section 2" 不同改就是指错条目，而 check.mjs +// 是拿译文的指路跟中文原文逐个比，条号对不上直接失败。节号不变（34 节没增减）， +// 所以只改条号。 +// +// 三种结果：标题和正文都没变的，译文搬过去（条号和指路改好）；标题变了的、正文 +// 变了的、新增的，译文那边留中文原文，等着人翻。搬不动的（并条之后指路指向的条号 +// 在新原文里没有了、跨删除点的区间引用）报出来，不猜。 +// +// 跑完 join.mjs 一校就知道有没有搬坏。 + +import { readFileSync, writeFileSync, existsSync, readdirSync, rmSync, mkdirSync } from 'node:fs'; +import { resolve, dirname } from 'node:path'; +import { fileURLToPath } from 'node:url'; +import { LOCALES, SOURCE, strings } from '../site/locales.mjs'; +import { splitSection } from './split.mjs'; + + +const ROOT = resolve(dirname(fileURLToPath(import.meta.url)), '../..'); +const read = p => readFileSync(resolve(ROOT, p), 'utf8').replace(/\r\n/g, '\n'); +const write = (p, t) => { + mkdirSync(resolve(ROOT, p).replace(/\/[^/]+$/, ''), { recursive: true }); + writeFileSync(resolve(ROOT, p), t); +}; + +const argv = process.argv.slice(2); +const dry = argv.includes('--dry'); +const baseDir = argv.find(a => !a.startsWith('--')); +const locales = argv.includes('--all') || !argv.some(a => LOCALES.some(l => l.code === a)) + ? LOCALES.map(l => l.code).filter(c => c !== SOURCE.code) + : argv.filter(a => LOCALES.some(l => l.code === a)); +if (!baseDir || !existsSync(resolve(ROOT, baseDir))) { + console.error('用法：node tools/i18n/rebase.mjs <旧 book 目录> [en|vi] [--dry]'); + process.exit(2); +} + +/** 一份正文拆成「头部 + 按条号索引的条目」，条目是行数组 */ +function pieces(text) { + const lines = text.split('\n'); + const at = lines.map((l, i) => (/^### /.test(l) ? i : -1)).filter(i => i >= 0); + let end = lines.length; + while (end > 0 && !lines[end - 1].trim()) end--; + if (end > 0 && /^\[.*\]\(/.test(lines[end - 1])) end--; + const head = at.length ? lines.slice(0, at[0]) : []; + const byN = new Map(); + for (let k = 0; k < at.length; k++){ + const stop = k + 1 < at.length ? at[k + 1] : end; + const seg = lines.slice(at[k], stop); + while (seg.length && !seg[seg.length - 1].trim()) seg.pop(); + byN.set(/^### (\d+)\./.exec(seg[0])[1], seg); + } + return { head, byN, tail: lines.slice(end) }; +} + +const titleOf = l => l.replace(/^### \d+\.\s*/, '').trim(); +const bodyOf = seg => seg.slice(1).join('\n'); +const keyOf = seg => bodyOf(seg).replace(/\s+/g, '').slice(0, 60); +const squash = s => s.replace(/\s+/g, ''); + +// ---------------------------------------------------------------- 1. 条号映射 +// 每节一张 oldNum -> newNum。搬得动的才进这张表：并条和真改写的那几条搬不过去， +// 指路指着它们的地方一并报出来（见下面的 dangling）。 +const srcDir = (SOURCE.contentDir ? SOURCE.contentDir + '/' : '') + 'book/'; +const secs = readdirSync(resolve(ROOT, srcDir)).filter(x => x.endsWith('.md')).sort(); +const secNames = secs.map(f => f.replace(/\.md$/, '')); +const targetSecName = n => secNames[Number(n) - 1]; // 「第 N 节」的第 N 个文件名 +const maps = {}; // { 节名: Map<旧条号, 新条号> } +const fresh = {}; // { 节名: [新条号] } 要人翻的 +const rewords = []; // [{ 节名, 条号 }] 正文真改了、译文要重翻的 +const pending = []; // 配好对、待验正文的 +const zhXref = strings(SOURCE.code).xref; +for (const f of secs){ + const name = f.replace(/\.md$/, ''); + const baseFile = `${baseDir}/${f}`; + if (!existsSync(resolve(ROOT, baseFile))) { console.log(`跳过 ${f}：旧原文里没有这一节`); continue; } + const oldE = pieces(read(baseFile)); + const newE = pieces(read(srcDir + f)); + const byTitle = new Map(), byKey = new Map(); + for (const [n, seg] of oldE.byN){ + const t = titleOf(seg[0]); + if (!byTitle.has(t)) byTitle.set(t, []); + byTitle.get(t).push({ n, seg }); + if (!byKey.has(keyOf(seg))) byKey.set(keyOf(seg), []); + byKey.get(keyOf(seg)).push({ n, seg }); + } + const claimed = new Set(); + const paired = []; // [新条号, 旧条号, 新正文, 旧正文] + const todo = []; + for (const [n, seg] of newE.byN){ + const t = titleOf(seg[0]); + const hit = (byTitle.get(t) || []).find(x => !claimed.has(x.seg) && keyOf(x.seg) === keyOf(seg)) + ?? (byTitle.get(t) || []).find(x => !claimed.has(x.seg)) + ?? (byKey.get(keyOf(seg)) || []).find(x => !claimed.has(x.seg)); + if (!hit){ todo.push(n); continue; } + claimed.add(hit.seg); + paired.push([n, hit.n, bodyOf(seg), bodyOf(hit.seg)]); + } + maps[name] = new Map(paired.map(([n, o]) => [o, n])); + fresh[name] = todo; + pending.push({ name, paired, map: maps[name] }); +} + +// 配好对之后逐条验：标题对得上但正文动了的，要分两种——只是指路条号跟着顺延 +// （译文照同样的规则改一下就行），还是真的改了内容（译文得重翻）。判据是把旧 +// 正文里的指路按映射改一遍，再跟新正文逐字比：一样就是顺延，不一样就是真改了。 +// 先把全部映射建好再验：一条的指路可能指着同一节里排在它后面的条目，那条的映射 +// 这时还没建，边建边验会把它当成「指路没动」。 +for (const { name, paired, map } of pending){ + const edited = []; + for (const [n, o, newBody, oldBody] of paired){ + if (squash(oldBody) === squash(newBody)) continue; + const allow = sourceRefs(newBody, zhXref, name); + if (squash(remapRefs(oldBody, zhXref, name, map, allow, [])) === squash(newBody)) continue; + map.delete(o); + fresh[name].push(n); + edited.push(n); + } + if (edited.length) rewords.push(`${name} 第 ${edited.join('、')} 条`); +} + +// ---------------------------------------------------------------- 2. 改指路条号 +// 各语言的指路写法从 locales/.json 的 xref 拿，和检索页把「第 X 条」变成 +// 可点引用、以及 structure.mjs 的 refsOf 是同一套正则，所以这里改的就是读者 +// 真能点到的那几处。 +// +// 数字一律逐个替换（\d+），分隔符原样留着——「items 8, 10 and 11」这种形状不该被 +// 脚本改写成别的样子。区间另算，见下面的 findRanges。 +function remapRefs(text, xref, secName, map, allow, dangling) { + if (!xref) return text; + let out = '', last = 0; + for (const m of text.matchAll(new RegExp(xref.pattern, 'g'))){ + const g = m.groups ?? {}; + // 节号那组（sections / phần）不动：34 节没有增减 + const key = g.nums !== undefined ? 'nums' : g.nums2 !== undefined ? 'nums2' : g.nums3 !== undefined ? 'nums3' : null; + if (!key) continue; + const target = g.sec ? g.sec : null; // 缺省就是本节 + const secOf = target ? targetSecName(target) : secName; + const mapOf = target ? maps[secOf] : map; + if (!mapOf){ continue; } // 指向的小节没了，原样留着 + const span = g[key]; + const sub = span.replace(/\d+/g, d => { + // allow 是「中文原文这一条里当作条目引用写的号」，只有在里面才改。 + // 不这么卡住就是乱改：英文的 xref 正则里有一条光秃秃的 \bitems?\s+(\d+)， + // 来源栏里「劳部发〔1994〕309 号，第 59 条」译成 "item 59" 也照样命中， + // 那是法条条款号不是条目号，改了就把引用指到别的条目上（实测这一类有 7 处）。 + if (!allow.has(secOf + ':' + d)) return d; + const nn = mapOf.get(d); + if (nn === undefined){ dangling.push(`${secName} 译文指路「第 ${target ?? '本节'}${d} 条」，新原文里没有对应条目`); return d; } + return String(nn); + }); + out += text.slice(last, m.index) + m[0].replace(span, sub); + last = m.index + m[0].length; + } + return out + text.slice(last); +} +/** 区间引用跨了删除点就不能逐个替换：「items 8 to 11」里的第 9 条要是被删了， + 新的应该写成「items 8, 10 and 11」，脚本猜不出该用哪种分隔，报出来让人看。 */ +function findRanges(text, xref, secName, allow) { + if (!xref) return []; + const out = []; + for (const m of text.matchAll(new RegExp(xref.pattern, 'g'))){ + const g = m.groups ?? {}; + const key = g.nums !== undefined ? 'nums' : g.nums2 !== undefined ? 'nums2' : g.nums3 !== undefined ? 'nums3' : null; + if (!key) continue; + const r = new RegExp(xref.rangeRe).exec(g[key]); + if (!r) continue; + const a = +r[1], b = +r[2]; + const secOf = g.sec ? targetSecName(g.sec) : secName; + // 和 remapRefs 同一个卡口：原文没把它当条目引用的（多半是法条条款号），不看 + if (![...Array(b - a + 1)].every((_, i) => allow.has(secOf + ':' + (a + i)))) continue; + const map = maps[secOf]; + if (!map) continue; + const moved = []; + for (let i = a; i <= b; i++) if (map.get(i) !== i) moved.push(`${i}->${map.get(i) ?? '没了'}`); + if (moved.length) out.push(`${secName}：区间 ${m[0]}（${moved.join('、')}）`); + } + return out; +} + +/** 一段正文里当条目引用写出来的号，格式「节名:条号」，本节的写本节名。 + xref 传哪门语言的写法就按那门语言解析——中文原文传中文的，译文传译文的 + （结构检查那边就是这么比的，所以这套判定要跟 refsOf 一致）。 + + 这里没有直接用 structure.mjs 的 refsOf：它把节号和条号拼成一个字符串 + （「第 6 节第 26 条」出来是 's626'，节号 6 条号 26 连在一起），本节与别节 + 的引用混在一处，比对时要拆开才分得清。拼起来的形状本身也有歧义——第 13 节 + 第 22 条和第 1 节第 322 条都出来是 's1322'——所以这里按 xref 的分组各读各的， + 区间和分隔符的处理照 refsOf 那份（listSep / rangeRe）。 */ +function sourceRefs(body, xref, secName) { + const out = new Set(); + if (!xref) return out; + const range = new RegExp(xref.rangeRe); + const sep = new RegExp(xref.listSep); + for (const m of String(body).matchAll(new RegExp(xref.pattern, 'g'))){ + const g = m.groups ?? {}; + if (g.secs !== undefined){ // 「第 8、24 节」：只有节号 + for (const s of String(g.secs).split(sep)) if (s.trim()) out.add(targetSecName(s.trim())); + continue; + } + const key = g.nums !== undefined ? 'nums' : g.nums2 !== undefined ? 'nums2' : g.nums3 !== undefined ? 'nums3' : null; + if (!key) continue; + const sec = g.sec ? targetSecName(g.sec) : secName; + if (!sec) continue; + const spec = String(g[key]).trim(); + const push = t => out.add(sec + ':' + t); + if (range.test(spec)){ // 「第 8 到 11 条」摊成一个一个 + const r = range.exec(spec); + const a = +r[1], b = +r[2]; + if (b >= a && b - a <= 40) for (let i = a; i <= b; i++) push(i); + continue; + } + for (const part of spec.split(sep)){ const t = part.trim(); if (t) push(t); } + } + return out; +} + +// ---------------------------------------------------------------- 3. 搬 +let carried = 0, untranslated = 0; +const allDangling = [], allRanges = [], heads = [], stale = []; +if (rewords.length){ + console.log(`\n正文真改了、译文要重翻 ${rewords.length} 处:`); + for (const r of rewords) console.log(' ' + r); +} +for (const locale of locales){ + const L = LOCALES.find(x => x.code === locale); + const S = strings(locale); + const tgtDir = (L.contentDir ? L.contentDir + '/' : '') + 'book/'; + for (const f of secs){ + const name = f.replace(/\.md$/, ''); + const bookFile = tgtDir + f; + if (!existsSync(resolve(ROOT, bookFile))) { console.log(`跳过 ${locale} ${f}：译文目录里没有`); continue; } + const oldBook = pieces(read(bookFile)); + const cut = splitSection(read(srcDir + f)); + const map = maps[name]; + if (!map) continue; + + const dangling = [], ranges = []; + // 译文的条号就是旧原文的条号（两边条目数 run 前已核对一致），按号取译文， + // 再用 map 把旧号改成新号。反查 map 得到旧号：新号到旧号。 + const newToOld = new Map([...map].map(([o, n]) => [n, o])); + const blocks = cut.blocks.map((entries, i) => entries.map(entryLines => { + const num = /^### (\d+)\./.exec(entryLines[0])[1]; + const oldNum = newToOld.get(num); + const seg = oldNum !== undefined ? oldBook.byN.get(oldNum) : null; + if (!seg){ untranslated++; return entryLines; } // 新增或改写的，留中文等着翻 + // 卡口用旧原文这一条的指路，不是新原文那一条的。译文里的号是旧号， + // 「这是不是条目引用」这件事也只有旧原文说了算（法条条款号混在同一段里）。 + // 用新原文的号去卡旧号，删条之后两边对不上，译文里的旧号就会一个都不改—— + // 比报错更坏，指路静默指错条目还不报。 + const allow = sourceRefs(bodyOf(seg), zhXref, name); + const rest = remapRefs(bodyOf(seg), S.xref, name, map, allow, dangling); + ranges.push(...findRanges(bodyOf(seg), S.xref, name, allow)); + // 搬过来的译文，指路必须跟新原文对得上。这是 rebase 自己的一份校验： + // 上游删条并条之后没跟着改的指路（指向被删掉或被并掉的那条），结构检查 + // 那边只会说「指路不见了」，看不出是哪一处指错了条目。这里点名报出来。 + // 译文那边要用它自己那门语言的指路写法解析，否则解析不出来，等于没比。 + const want = sourceRefs(bodyOf(entryLines), zhXref, name); + const got = sourceRefs(rest, S.xref, name); + for (const r of want) if (!got.has(r)) + stale.push(`${locale} ${name} 第 ${num} 条：译文的指路没有新原文的「${r}」，原文指向的那条被删或被并了`); + carried++; + // 标题行只改条号，标题文字是译文，原样留（节标题另说，见下面 head 的处理） + return [`### ${num}. ${titleOf(seg[0])}`, ...rest.split('\n')]; + })); + + // 节首（第 0 块里的回目录那行 + 「# N. 节名」+ 导读）单独判。判据比的是 + // 旧原文和现原文的节首，不是译文和现原文——译文本来就是另一种语言。 + // 三档：原文没动就搬译文节首；只是指路条号顺延（导读里「第 20 条」改成 + // 第 19 条），译文节首照同样的规则改一下就行；真加了句子就留中文待翻—— + // 节名或导读一动，侧栏（读 README）和正文两处都得改，只改一处 check.mjs + // 就会拿目录里的节名跟正文比。 + const oldZhHead = pieces(read(`${baseDir}/${f}`)).head.join('\n'); + const newZhHead = cut.head.join('\n'); + const norm = s => s.replace(/\s+/g, ''); + let head = newZhHead; + if (norm(oldZhHead) === norm(newZhHead)) head = oldBook.head.join('\n'); + else { + const empty = [], d = []; + const renum = remapRefs(oldZhHead, zhXref, name, map, sourceRefs(oldZhHead, zhXref, name), empty); + if (norm(renum) === norm(newZhHead)){ + head = remapRefs(oldBook.head.join('\n'), S.xref, name, map, sourceRefs(newZhHead, zhXref, name), d); + heads.push(`${locale} ${name}：节首只是指路条号顺延，已按映射改好`); + } else heads.push(`${locale} ${name}：节首（节名或导读）真改了，留在中文待翻`); + } + allDangling.push(...dangling.map(d => `${locale} ${d}`)); + allRanges.push(...ranges.map(r => `${locale} ${r}`)); + + const dir = `.i18n/${locale}/${name}`; + if (dry){ + console.log(`${locale} ${name}：${blocks.length} 块，待翻 ${fresh[name].length} 条`); + continue; + } + rmSync(resolve(ROOT, dir), { recursive: true, force: true }); + blocks.forEach((body, i) => { + write(`${dir}/part${String(i).padStart(2, '0')}.md`, + (i === 0 ? head + '\n' : '') + body.flat().join('\n') + '\n'); + }); + if (cut.tailLink) write(`${dir}/tail.txt`, cut.tailLink + '\n'); + console.log(`${locale} ${name}：${blocks.length} 块，搬过 ${blocks.flat().length - fresh[name].length} 条，待翻 ${fresh[name].length} 条（第 ${fresh[name].join('、')} 条）`); + } +} + +console.log(`\n搬过 ${carried} 条，留中文待翻 ${untranslated} 条`); +if (allDangling.length){ console.log(`\n指路搬不动的 ${allDangling.length} 处（要人工看）:`); for (const d of [...new Set(allDangling)]) console.log(' ' + d); } +if (stale.length){ console.log(`\n译文的指路跟新原文对不上 ${stale.length} 处（上游删条并条后没跟着改，要人工定）:`); for (const s of [...new Set(stale)]) console.log(' ' + s); } +if (allRanges.length){ console.log(`\n区间引用跨了删除点 ${allRanges.length} 处（要人工改写法）:`); for (const r of [...new Set(allRanges)]) console.log(' ' + r); } +if (heads.length){ console.log(`\n节首要人工看 ${heads.length} 处:`); for (const h of [...new Set(heads)]) console.log(' ' + h); } diff --git a/tools/i18n/structure.mjs b/tools/i18n/structure.mjs index 4c02da51..d73e209b 100644 --- a/tools/i18n/structure.mjs +++ b/tools/i18n/structure.mjs @@ -48,18 +48,24 @@ function cnNumber(s) { return total > 0 ? String(total) : null; } -/** 一栏里所有「第<中文数字>条 / 编 / 号 / 款」这类法条编号换算出来的数字。 +/** 一栏里所有法条编号里的中文数字换算成的阿拉伯数字。 * - * 一条法条可以并排写：「第一千零五十二、一千零五十三条规定」——「第」只出现在头一个， - * 后面几个是顿号连着的。所以先把「第…条」整串切出来，再按顿号/逗号逐个读。 */ + * 原文的编号是汉字（「第一千零四十五条」），译文按 BRIEF 要求写成「Điều 1045」—— + * 那个 1045 是原文就有的，不算译文自己加的数。 + * + * 一条法条可以并排写：「第一千零五十二、一千零五十三条」「第十六、十七条」—— + * 「第」只在头一个，后面几个是顿号连着的。所以先把「第…条」整串切出来（串里 + * 允许顿号继续接数字），再按顿号逐个读。只认「第 + 数字 + 条」的话，第二、三个 + * 编号全都读不到，译文里对应的数字就被报成「多了原文没有的」（越南文那边每条 + * 法条都踩一次）。 + */ function cnNumbersOf(field) { const out = new Set(); const D = '[零〇一二三四五六七八九十百千]'; - for (const m of String(field).matchAll(new RegExp(`第${D}{1,8}[条款项号章节][^。；;]{0,40}`, 'g'))){ + const run = new RegExp(`第${D}{1,8}(?:[、,，]${D}{1,8})*[条款项号章节]`, 'g'); + for (const m of String(field).matchAll(run)){ for (const p of m[0].slice(1).split(/[、,，]/)){ - const digits = new RegExp(`^${D}{1,8}`).exec(p.trim()); - if (!digits) continue; - const v = cnNumber(digits[0]); + const v = cnNumber(p.replace(/[条款项号章节]$/, '')); if (v) out.add(v); } } diff --git a/vi/README.md b/vi/README.md index bc457693..f3b52b99 100644 --- a/vi/README.md +++ b/vi/README.md @@ -32,7 +32,7 @@ Trợ lý trợ giọng nói cho AI chạy được với Claude Code và Codex. Bài dài -[Kết hôn có đáng không](docs/结婚划不划算.md) · [Danh mục trang bị cứu hộ tại nhà](docs/家庭应急装备清单.md) · [Có nên dừng lại khi người lạ gặp chuyện](docs/遇到陌生人出事该不该停.md) · [Một nền tảng cần những giấy phép nào](docs/做平台要办哪些证.md) · [Nhịp sinh học và ca đêm](docs/生物钟和夜班.md) +[Kết hôn có đáng không](docs/结婚划不划算.md) · [Danh mục trang bị cứu hộ tại nhà](docs/家庭应急装备清单.md) · [Có nên dừng lại khi người lạ gặp chuyện](docs/遇到陌生人出事该不该停.md) · [Một nền tảng cần những giấy phép nào](docs/做平台要办哪些证.md) · [Nhịp sinh học và ca đêm](docs/生物钟和夜班.md) · [Bị sa thải thì làm gì trước](docs/被裁了之后先做什么.md) · [Việc cần làm trước và sau khi có con](docs/孩子出生前后要办的事.md) · [Vừa được chẩn đoán bệnh mạn tính](docs/刚确诊慢性病之后.md) Ngôn ngữ khác @@ -257,10 +257,10 @@ Phần này giữ chỗ dễ hiểu khi có thể, nhưng vài thuật ngữ th 13. [Tình huống khẩn cấp: làm gì trước](book/13-紧急情况.md): ngừng tim, đột quỵ và đột quỵ tuần hoàn sau, đột quỵ mắt, nhồi máu cơ tim, bóc tách động mạch chủ, đau đầu xuất huyết ầm, tụ máu dưới màng não mãn tính, thuyết tắc phổi, chảy máu nặng, vết cắn, bỏng và phỏng, sốc phản vệ, động kinh, hạ đường huyết, điện giật, ô-xít cacbon, nuốt hóa chất và bỏng hóa chất, vật thể xâm nhập vào cơ thể, cố định gãy xương, lừa đảo, đe doạ quyền riêng tư, say nắng, hoả hoạn, chết đuối, đi lạc, hạ thân nhiệt, rắn cắn, động đất, thú hoang, sét, say vì độ cao, ve ký sinh, uống nước ngoài trời. Và chuyện người ta có cứu được hay không: làm sao đỡ người lớn tuổi bị ngã lên, làm gì khi bước vào một cuộc đánh nhau, và ai trả tiền khi việc giúp người làm bạn bị thương. Thước đo: sống sót và tiền, mấy mục cuối còn đụng tới tự do cá nhân. 14. [Tài khoản và an toàn thông tin](book/14-账号与信息安全.md): xác minh yếu tố thứ hai, mật khẩu, thẻ SIM, mất điện thoại, lừa đảo thẻ, thiết bị đang đăng nhập, quyền của ứng dụng, nhận diện khuôn mặt, quyền kiểm tra và quyền xoá. Thước đo: tiền / thông tin cá nhân. 15. [Thuê nhà và mua nhà](book/15-租房与买房.md): tiền đặt cọc, bị đuổi bằng vũ lực, môi giới giữ tiền, ký quỹ tiền, một lần bán không phá hợp đồng thuê, kiểm tra giấy tờ sở hữu, một tài khoản riêng cho khoản tiền, phòng chia, và đừng mua đất ở nông thôn khi mang hộ khẩu thành phố, chỉ thuê nhà ở nông thôn. Thước đo: tiền. -16. [Sống với bệnh mãn tính](book/16-得了慢性病之后怎么活.md): uống thuốc, quyết toán bệnh ngoại tỉnh cho bệnh mãn tính và bệnh nặng ngoại trú, giữ lại kết quả xét nghiệm, không bỏ thuốc để thử cách chữa dân gian, đơn thuốc dài, đăng ký với bác sĩ gia đình, sàng lọc biến chứng, phòng sỏi thận tái phát, và hạ gout về mức mục tiêu. Thước đo: tử vong toàn nguyên / tiền. +16. [Sống với bệnh mãn tính](book/16-得了慢性病之后怎么活.md): uống thuốc, quyết toán bệnh ngoại tỉnh cho bệnh mãn tính và bệnh nặng ngoại trú, giữ lại kết quả xét nghiệm, không bỏ thuốc để thử cách chữa dân gian, đơn thuốc dài, đăng ký với bác sĩ gia đình, sàng lọc biến chứng, phòng sỏi thận tái phát, và hạ gout về mức mục tiêu. Thước đo: tử vong toàn nguyên / tiền. Bài dài: [docs/刚确诊慢性病之后.md](docs/刚确诊慢性病之后.md). 17. [Khi nhà có người lớn tuổi](book/17-家里有老人.md): giám hộ tự nguyện, các hình thức di chúc, tài khoản và cách chào hàng, lừa đảo đầu tư và thế chấp ngược, bảo hiểm chăm sóc dài hạn, và phòng loét do nằm liệt lâu. Thước đo: tiền / tự do cá nhân, trong đó mục loét do nằm liệt nói về tử vong. 18. [Nuôi con có đáng không](book/18-养孩子划不划算.md): trợ cấp chăm trẻ, chế độ thai sản và thưởng thai sản, ba giai đoạn được bảo vệ, cách tính thời gian, cách tính tiền. Thước đo: tiền / thời gian. -19. [Đi làm, nghỉ việc và tai nạn lao động](book/19-在职离职和工伤.md): tiền làm thêm giờ, nghỉ phép năm, thử việc; thông báo nguy cơ nghề nghiệp và ba kỳ khám, chống bụi và tiếng ồn; N, bồi thường thay việc làm, 2N, đừng ký từ nguyện nghỉ việc, giữ lại chứng cứ; hạn công nhận tai nạn lao động, chủ lao động không bảo hiểm, đánh giá khả năng lao động, và trợ cấp khi tử vong. Thước đo: tiền. +19. [Đi làm, nghỉ việc và tai nạn lao động](book/19-在职离职和工伤.md): tiền làm thêm giờ, nghỉ phép năm, thử việc; thông báo nguy cơ nghề nghiệp và ba kỳ khám, chống bụi và tiếng ồn; N, bồi thường thay việc làm, 2N, đừng ký từ nguyện nghỉ việc, giữ lại chứng cứ; hạn công nhận tai nạn lao động, chủ lao động không bảo hiểm, đánh giá khả năng lao động, và trợ cấp khi tử vong. Thước đo: tiền. Bài dài: [docs/被裁了之后先做什么.md](docs/被裁了之后先做什么.md). 20. [Chăm trẻ mới sinh](book/20-刚出生的孩子怎么带.md): cho nằm an toàn, mũi tiêm viêm gan B đầu tiên, lịch tiêm chủng, sữa mẹ và thức ăn dặm, nhiệt độ khi pha sữa bằng công thức, mật ong, vitamin K, mức nhiệt độ nào thì phải đi khám, không bao giờ lắc trẻ, tã giấy và mua số lượng lớn, và cho trẻ dị ứng cao ăn sớm đậu phộng để phòng dị ứng. Thước đo: tử vong trẻ sơ sinh / tiền. 21. [Đi nước ngoài và an toàn ở nước ngoài](book/21-出国旅行与境外安全.md): các mức cảnh báo đi lại, 12308, giới hạn của sự bảo vệ của lãnh sự, bảo hiểm y tế ở nước ngoài, lời mời việc lương cao ở nước ngoài, hạn mức rút tiền mặt mỗi năm ở nước ngoài, mất giấy tờ, bằng lái xe ở nước ngoài, và đăng ký qua đại lý. Thước đo: tiền / tự do cá nhân. 22. [Thả lỏng: nơi chốn và cách giảm căng thẳng](book/22-怎么放松.md): cửa thoát hiểm, niêm yết giá, ranh giới quanh ma túy, đồ người khác đưa cho, đăng ký thật ở tiệm mạng, chọn nơi chơi phòng giải đố án mạng; vận động, chánh niệm, hơi thở, bạn bè, cây xanh. Thước đo: tiền / tự do cá nhân, cùng sức lực / tử vong toàn nguyên. @@ -268,7 +268,7 @@ Phần này giữ chỗ dễ hiểu khi có thể, nhưng vài thuật ngữ th 24. [Khám bệnh: tốn ít tiền hơn và đi lại ít hơn](book/24-看病.md): hệ thống phân tầng và chuyển tuyến, mức đồng chi trả cộng dồn liên tục, khác biệt về tỷ lệ hoàn trả, lịch khám dành riêng, đánh giá có cần điều trị nơi khác hay không, giữ và niêm phong hồ sơ bệnh, bốn mức phân loại cấp cứu, trợ giúp cấp cứu khi không đủ tiền, khi nào nên đánh giá khuyết tật, làm sao có giấy chứng nhận khuyết tật, và không đưa phong bì cho bác sĩ. Thước đo: tiền / thời gian. 25. [Sau khi có người qua đời](book/25-人走了以后要办什么.md): gọi công an và giấy chứng tử, nhận thi thể và hỏa thiêu, khiếu nại về nguyên nhân chết và giải phẫu, huỷ đăng ký hộ khẩu, danh mục tang vật cơ bản, xử phạt giá, đăng ký qua đại lý, số dư quỹ tích góp nhà ở và quyền lợi bảo hiểm xã hội, và quyền về thông tin cá nhân của người đã mất. Thước đo: tiền. 26. [Làm một trang web hoặc nền tảng: giấy phép, đăng ký, máy chủ](book/26-做一个网站或平台.md): vạch đỏ thanh toán và hạch toán, giấy phép và đăng ký ICP, điều kiện cho phát số và dịch vụ âm thanh hình, xác minh nền tảng và kê khai thuế, quản trị nội dung, đối chiếu danh tính thật, người chưa thành niên, thông báo và gỡ bỏ, xuất dữ liệu, và chọn máy chủ. Thước đo: tự do cá nhân / tiền. Bài dài: [docs/做平台要办哪些证.md](docs/做平台要办哪些证.md). -27. [Mang thai và sinh con: từ que thử đến giấy tờ](book/27-怀孕和生产.md): axit folic, sổ sức khoẻ bà mẹ và khám thai miễn phí, sàng lọc ba bệnh lây nhiễm và chặn lây truyền từ mẹ sang con, hút thuốc và uống rượu khi mang thai, aspirin và đái thường thai kỳ, dấu hiệu phải vào viện ngay, làm gì khi vỡ ối, gây mê tủng sống, chỉ định mổ đẻ, bảo hiểm thai sản, giấy khai sinh, sàng lọc trẻ sơ sinh, bảo hiểm và đăng ký hộ khẩu, và kiểm tra 42 ngày sau khi sinh. Thước đo: tử vong và tiền. +27. [Mang thai và sinh con: từ que thử đến giấy tờ](book/27-怀孕和生产.md): axit folic, sổ sức khoẻ bà mẹ và khám thai miễn phí, sàng lọc ba bệnh lây nhiễm và chặn lây truyền từ mẹ sang con, hút thuốc và uống rượu khi mang thai, aspirin và đái thường thai kỳ, dấu hiệu phải vào viện ngay, làm gì khi vỡ ối, gây mê tủng sống, chỉ định mổ đẻ, bảo hiểm thai sản, giấy khai sinh, sàng lọc trẻ sơ sinh, bảo hiểm và đăng ký hộ khẩu, và kiểm tra 42 ngày sau khi sinh. Thước đo: tử vong và tiền. Bài dài: [docs/孩子出生前后要办的事.md](docs/孩子出生前后要办的事.md). 28. [Đừng phá hỏng cơ thể vì ngoại hình](book/28-别为了外形把身体搞坏.md): ăn kiêng cực đại và rối loạn ăn uống, phòng khám thẩm mỹ và hai giấy chứng nhận mà bác sĩ thực hiện phải có, chất làm đầy mặt và những vùng trên mặt bị mù, sản phẩm giảm cân bị trộn sibutramin trái phép, steroid đồng hoạt, đơn thuốc và tái khám với thuốc giảm cân và hooc môn sinh dục, và đánh giá hình ảnh cơ thể. Thước đo: tử vong (một kết quả sức khoẻ), với hai mục về thẩm mỹ còn đụng tới tự do cá nhân. 29. [Sau một cú đánh nặng](book/29-遭遇重大打击之后.md): cửa sổ tim mạch trong tháng đầu sau khi mất người thân, tuần đầu sau khi được chẩn đoán bệnh nặng, mất việc, sáu tháng sau khi mất người bạn đời, tự tử dẫn tới tang, làm gì khi không còn gia đình lẫn bạn bè để thay người đó, trẻ mất cha mẹ, đăng ký khám ở đâu khi nỗi buồn bị kẹt, ly hôn, 12356 và 12355, không quyết định không thể đảo ngược trong giai đoạn căng thẳng trực tiếp, và vì sao dùng cái chết để dàn xếp nợ không có tác dụng. Thước đo: tử vong toàn nguyên, bốn mục về tiền và quyền lợi tính vào tiền. 30. [Trẻ đang đi học](book/30-上学以后的孩子.md): cấp cứu tính theo giờ chứ không vì thi, xử lý bạo hành, 2 giờ ngoài trời mỗi ngày, báo cáo sức khoẻ ở trường, sàng lọc trầm cảm ở thanh niên, sản phẩm quảng cáo chữa khỏi cận thị, quy tắc cứng về giấc ngủ và bài tập, nghỉ học mà vẫn giữ hồ sơ học tập, khúc xạ giãn đồng tử và tái khám, và chất bít khe răng. Thước đo: tử vong và kết quả sức khoẻ, cộng thêm một hai mục về tiền và một hai mục về thời gian. @@ -277,7 +277,7 @@ Phần này giữ chỗ dễ hiểu khi có thể, nhưng vài thuật ngữ th 33. [Sống sau khi bị khuyết tật](book/33-残疾之后怎么活.md): ba bước tại chỗ khi bị tăng huyết áp thần kinh tự động, cửa sổ tự sát trong mười năm sau chấn thương gây khuyết tật, nguyên tắc nhận viện chủ động để chữa bệnh tâm thần và hai ngoại lệ của nó, nguy cơ tử vong của người chăm sóc, đệm chống loét cho xe lăn, lừa đảo "chữa bệnh", sáu quyền lợi cần hỏi sau khi có giấy chứng nhận, bảo hiểm chăm sóc dài hạn không chỉ dành cho người lớn tuổi, trợ cấp phục hồi chức năng cho lứa tuổi 0 đến 6, trợ cấp làm nhà thuận tiện cho người khuyết tật, việc làm có chỉ tiêu và quỹ bảo đảm việc làm cho người khuyết tật, giảm thuế thu nhập cá nhân, chó dẫn đường và miễn vé, điều kiện hợp lý trong kỳ thi tốt nghiệp THPT, nghĩa vụ không được từ chối suất và học phí tại nhà, giấy phép lái hạng C5, cách chọn nơi phục hồi chức năng, máy trợ thính, đánh giá năng lực và giám hộ. Thước đo: tử vong / tiền / thời gian / tự do cá nhân. 34. [Đừng dùng thuốc trong nhà sai cách](book/34-家里的常备药别吃出事.md): đừng dùng paracetamol gấp đôi, không aspirin, nimesulide hay metamizole cho trẻ bị sốt, các nhóm dễ bị hại dạ dày khi dùng ibuprofen, không tự ý dùng thuốc cảm kết hợp cho trẻ dưới 2 tuổi, không tự ý dùng ibuprofen sau tuần 20 thai kỳ, không tự ý dùng omeprazole quá 7 ngày, không dùng kháng sinh khi bị cảm, bù nước trước khi chữa tiêu chảy và không cho trẻ dùng thuốc cầm tiêu chảy, và thêm thuốc giảm đau chỉ khiến đau đầu nặng hơn. Thước đo: tử vong. -Trong mỗi phần, các mục đi từ hiệu quả chi phí tốt nhất xuống kém nhất. Một tên phần như "Đừng chết sớm" hay "Đừng phí thời gian" gọi tên cái kết quả mà phần đó cố ngăn. Chuyện một mục nên làm hay không nên làm thì do tiêu đề mục quyết. Với các bài dài, xem [docs/家庭应急装备清单.md](docs/家庭应急装备清单.md), [docs/做平台要办哪些证.md](docs/做平台要办哪些证.md), [docs/结婚划不划算.md](docs/结婚划不划算.md), [docs/遇到陌生人出事该不该停.md](docs/遇到陌生人出事该不该停.md) và [docs/生物钟和夜班.md](docs/生物钟和夜班.md). Hồ sơ đối chiếu từng nguồn được ghi ở [docs/核实记录](docs/核实记录/). +Trong mỗi phần, các mục đi từ hiệu quả chi phí tốt nhất xuống kém nhất. Một tên phần như "Đừng chết sớm" hay "Đừng phí thời gian" gọi tên cái kết quả mà phần đó cố ngăn. Chuyện một mục nên làm hay không nên làm thì do tiêu đề mục quyết. Với các bài dài, xem [docs/家庭应急装备清单.md](docs/家庭应急装备清单.md), [docs/做平台要办哪些证.md](docs/做平台要办哪些证.md), [docs/结婚划不划算.md](docs/结婚划不划算.md), [docs/遇到陌生人出事该不该停.md](docs/遇到陌生人出事该不该停.md) và [docs/生物钟和夜班.md](docs/生物钟和夜班.md), [docs/被裁了之后先做什么.md](docs/被裁了之后先做什么.md), [docs/孩子出生前后要办的事.md](docs/孩子出生前后要办的事.md) và [docs/刚确诊慢性病之后.md](docs/刚确诊慢性病之后.md). Hồ sơ đối chiếu từng nguồn được ghi ở [docs/核实记录](docs/核实记录/). Tệp `index.html` ở thư mục gốc kho là trang tra cứu trên mạng. Nó lọc các mục theo từ khoá, theo phần, theo mức bằng chứng và theo các chiều chi phí (tiền bỏ ra, thời gian bỏ ra, ý chí phải có), và đọc dữ liệu thẳng từ tệp này. Nó xuất hiện sau khi bật GitHub Pages trong phần cài đặt kho (Deploy from a branch, branch main, directory /). `tools/epub/` giữ script dựng sách điện tử, và `cd tools/epub && npm ci && npm run build` tạo ra một tệp EPUB trong `dist/` ở máy bạn. GitHub Actions chạy đúng script đó sau mỗi lần nội dung thay đổi và cập nhật bản Release. diff --git "a/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" "b/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" index 030f1475..4cfb6250 100644 --- "a/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" +++ "b/vi/book/01-\344\270\215\350\246\201\346\227\251\346\255\273.md" @@ -28,14 +28,15 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: B - Nguồn: Marshall SW và cộng sự (1998). Fatal residential fires: who dies and who survives? JAMA. ; USFA (2022). Fatal Fires in Residential Buildings (2018-2020), Topical Fire Report Series 22(2). ; You J và cộng sự (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly. - Ghi chú: Lý do để ở mức B: phía báo động khói chỉ có một nghiên cứu, cách làm là so ngược hộ có sự cố với hộ không có, không phải thử nghiệm chia nhóm ngẫu nhiên. Bản thân báo động ô-xít cacbon không có nghiên cứu nào chứng minh giảm tỷ lệ tử vong. Lợi ích ở đây suy ra từ việc các ca chết vì ngộ độc dồn vào mùa đông và dồn trong nhà. -### 4. Đến hạn thì thay ống mềm gas và bếp, không tự sửa ống, từ chối luôn người đến nhà bán gas +### 4. Ống gas và bếp gas đến hạn thì thay, đừng tự sửa ống, công ty gas đến nhà bán hàng thì từ chối thẳng -- Chi phí: Một ống mềm gas đạt chuẩn vài chục tệ. Bếp thay mỗi vài năm một lần. Khi thay thì tiện tay xem đã quá hạn chưa. -- Nói thẳng: Một ống mềm đạt chuẩn vài chục tệ. Đến hạn phải thay, đó là nghĩa vụ của người dùng ghi trong Quy định quản lý khí thành phố. Tự sửa ống, tự tháo đồng hồ gas là những việc quy định cấm. Cùng quy định đó còn viết, công ty gas không được bắt bạn mua sản phẩm và dịch vụ do nó chỉ định. Người đến nhà nói「phải thay」thì từ chối thẳng. -- Lợi ích: Quy định xếp việc「及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管」là nghĩa vụ của người dùng: bếp và ống mềm đã hết hạn hay bị nhà nước loại bỏ thì phải thay. Quy định cũng liệt kê「擅自安装、改装、拆除户内燃气设施和燃气计量装置」là hành vi bị cấm, tự sửa ống hay tự tháo đồng hồ gas đều nằm trong đó. Quy định nói rõ, đơn vị kinh doanh gas không được yêu cầu người dùng mua sản phẩm do nó chỉ định, cũng không được yêu cầu người dùng chấp nhận dịch vụ do nó cung cấp -- Mức bằng chứng: A -- Nguồn: 国务院 (Hội đồng Nhà nước) (2010). Quy định quản lý khí thành phố, 城镇燃气管理条例（国务院令第 583 号）. 第二十七条: 「燃气用户应当遵守安全用气规则，使用合格的燃气燃烧器具和气瓶，及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管等」 — người dùng gas phải tuân thủ quy tắc dùng gas an toàn, dùng thiết bị đốt gas và bình gas đạt chuẩn, và kịp thời thay mọi thiết bị đốt gas hoặc ống nối bị nhà nước yêu cầu loại bỏ hay đã hết hạn sử dụng. Các hành vi bị cấm ở 第二十八条 gồm 「（一）擅自操作公用燃气阀门」「（二）将燃气管道作为负重支架或者接地引线」「（四）擅自安装、改装、拆除户内燃气设施和燃气计量装置」 — tự ý vặn van gas công cộng, dùng ống gas làm giá đỡ tải hay làm dây tiếp đất, và tự ý lắp, sửa, tháo thiết bị gas trong nhà cùng thiết bị đo gas. 第二十条 cấm đơn vị kinh doanh gas 「（六）要求燃气用户购买其指定的产品或者接受其提供的服务」 — yêu cầu người dùng gas mua sản phẩm do họ chỉ định hoặc chấp nhận dịch vụ họ cung cấp. 第二十九条 cho phép người dùng khiếu nại về phí và dịch vụ với cơ quan quản lý gas, và 「有关部门应当自收到投诉之日起15个工作日内予以处理」 — cơ quan có trách nhiệm xử lý trong 15 ngày làm việc kể từ khi nhận được khiếu nại. ；深圳市住房和建设局 (Sở Nhà ở và Xây dựng Thâm Quyến) (2024). Kiến thức dùng gas an toàn (khoản 2), 安全用气小知识（二）: 「立即关闭家里的燃气总阀门」— đóng ngay van gas tổng trong nhà, 「迅速开门、开窗进行通风」— mở cửa và cửa sổ để thông gió, 「一定不要开、关家里的电灯、抽油烟机等电器」— tuyệt đối không bật tắt đèn hay máy hút mùi trong nhà, 「人员应该迅速撤离，转移到安全的地方，并且及时拨打燃气公司的报警电话」— mọi người phải rút ra ngay, sang chỗ an toàn, rồi gọi số báo động của công ty gas. ；张家港市应急管理局 (Cục Quản lý Khẩn cấp Trường Gia Cang) (2024). Kiến thức dùng gas an toàn, 燃气安全使用知识: 「不要情急之下打开排气扇进行通风」— đừng vội bật quạt thông gió, 「应远离现场之后才能打电话求助」— phải rời khỏi hiện trường rồi mới gọi cầu cứu. -- Ghi chú: Có người mang danh「kiểm tra an toàn」đến nhà, kiểm tra xong là chào bán van tự đóng và báo động. Quy định không cho phép công ty gas bắt bạn mua sản phẩm do nó chỉ định, bạn từ chối được, khiếu nại cũng được. Ngửi thấy mùi gas thì tắt van gas tổng trước, gas bình thì tắt van trên bình. Rồi mở cửa mở cửa sổ thông gió, đừng bật quạt thông gió. Đèn, máy hút mùi các thiết bị đó đừng bật cũng đừng tắt, bật lửa cũng đừng đụng. Sau đó mọi người rút ra ngoài, đã rời hiện trường thì gọi số báo động của công ty gas. Ngộ độc ô-xít cacbon là chuyện khác, xem phần 13. +- Chi phí: Một ống gas đạt chuẩn vài chục tệ. Bếp gas thay mỗi vài năm. Khi thay thì tiện tay nhìn xem đã hết hạn chưa. +- Nói thẳng: Một ống gas đạt chuẩn vài chục tệ. Đến hạn thì thay, đây là nghĩa vụ của người dùng mà《城镇燃气管理条例》quy định rõ. Tự sửa ống, tháo đồng hồ khí là những việc quy định cấm rõ ràng. Cùng quy định đó còn viết, công ty gas không được bắt bạn phải mua sản phẩm và dịch vụ mà nó chỉ định. Có người đến nhà nói「phải thay」thì bạn từ chối thẳng được. +- Lợi ích: Quy định ghi「及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管」là nghĩa vụ của người dùng, tức khi bếp và ống gas hết hạn hoặc bị nhà nước loại bỏ thì bạn phải thay. Quy định còn liệt kê「擅自安装、改装、拆除户内燃气设施和燃气计量装置」vào những việc bị cấm, tự tay sửa ống hay tháo đồng hồ khí đều nằm trong đó. Quy định đồng thời nói rõ, doanh nghiệp khí không được bắt người dùng phải mua sản phẩm mà nó chỉ định, cũng không được bắt người dùng chấp nhận dịch vụ mà nó cung cấp +- Mức bằng chứng: C +- Nguồn: Hội đồng Nhà nước (国务院) (2010). 城镇燃气管理条例（国务院令第 583 号）第二十七条：「燃气用户应当遵守安全用气规则，使用合格的燃气燃烧器具和气瓶，及时更换国家明令淘汰或者使用年限已届满的燃气燃烧器具、连接管等」；第二十八条禁止行为含「（一）擅自操作公用燃气阀门」「（二）将燃气管道作为负重支架或者接地引线」「（四）擅自安装、改装、拆除户内燃气设施和燃气计量装置」；第二十条燃气经营者不得「（六）要求燃气用户购买其指定的产品或者接受其提供的服务」；第二十九条用户可就收费、服务向燃气管理等部门投诉，「有关部门应当自收到投诉之日起15个工作日内予以处理」. ；Sở Nhà ở và Xây dựng thành phố Thâm Quyến (深圳市住房和建设局) (2024). 安全用气小知识（二）：「立即关闭家里的燃气总阀门」「迅速开门、开窗进行通风」「一定不要开、关家里的电灯、抽油烟机等电器」「人员应该迅速撤离，转移到安全的地方，并且及时拨打燃气公司的报警电话」. ；Cục Quản lý Khẩn cấp thành phố Trương Gia Cảng (张家港市应急管理局) (2024). 燃气安全使用知识：「不要情急之下打开排气扇进行通风」「应远离现场之后才能打电话求助」. +- Ghi chú: Nguyên văn quy định đối chiếu được từng chữ, nhưng nó chỉ nói việc thay ống đến hạn là nghĩa vụ của bạn, không nói thay xong giảm được bao nhiêu tai nạn. Thay ống theo đúng hạn giảm được bao nhiêu vụ tai nạn khí, không tìm được con số thống kê chính thức, nên chỉ cho mức C. Có người mang danh「kiểm tra an toàn」đến nhà, kiểm tra xong bán van tự đóng và báo động. Điều 20, khoản (6) của quy định không cho phép công ty gas bắt bạn mua sản phẩm nó chỉ định, bạn từ chối được, khiếu nại cũng được. Ngửi thấy mùi khí, tắt van khí tổng trước; khí bình khí thì tắt van bình. Sau đó mở cửa, mở cửa sổ thông gió, đừng bật quạt thông gió. Đèn, máy hút mùi, những thiết bị điện này đừng bật cũng đừng tắt, bật lửa cũng đừng đụng vào. Rồi mọi người rút ra, đi khỏi hiện trường rồi gọi số báo động của công ty khí. Ngộ độc ô-xít cacbon là chuyện khác, xem phần 13. + ### 5. Không hái, không mua, không ăn nấm hoang, mọi cách nhận biết kiểu dân gian đều không có cơ sở - Chi phí: Không tốn tiền. Chỗ khó là từ bỏ một món ăn hoang theo mùa. @@ -61,14 +62,15 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: A - Nguồn: Ettehad D và cộng sự (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Lu J và cộng sự (2017). Prevalence, awareness, treatment, and control of hypertension in China (China PEACE Million Persons Project). Lancet. - Ghi chú: Đi đo huyết áp một lần trước đã. Ở Trung Quốc, trong 10 người cao huyết áp thì chưa đến 1 người kiểm soát được, nên lần đo này là một trong những việc hiệu quả chi phí cao nhất trong sách. Hạ tới bao nhiêu thì coi là đạt (130 hay 140) còn đang bàn, nhưng việc biết mình cao huyết áp và hạ nó xuống thì không ai tranh cãi. -### 8. Từ 35 tuổi, chỉ cần thừa cân là đi xét nghiệm glucose lúc nhịn một lần, bình thường thì cứ ba năm xét lại +### 8. Từ 35 tuổi, cứ thừa cân là đi xét đường huyết lúc nhịn một lần, xét ra bình thường thì ba năm xét lại -- Chi phí: Xét nghiệm glucose lúc nhịn mười mấy tệ. Xét nghiệm hemoglobin glycate vài chục tệ. Đến bệnh viện rút máu một lần là xong. -- Nói thẳng: Bỏ ra vài chục tệ để rút một ống máu là biết mình đang ở giai đoạn tiền đái thường, khi chưa có triệu chứng nào. Khuyến cáo sàng lọc chính thức của Mỹ là: người 35 đến 70 tuổi thừa cân hoặc béo thì đi xét nghiệm. Kết quả bình thường thì ba năm xét lại một lần. -- Lợi ích: Nhóm dịch vụ phòng ngừa Hoa Kỳ là cơ quan đưa ra khuyến cáo sàng lọc chính thức của Mỹ. Khuyến cáo hiện tại là: sàng lọc tiền đái thường và đái thường tuýp 2 ở người lớn 35 đến 70 tuổi thừa cân hoặc béo, kể cả khi chưa có triệu chứng nào. Thừa cân là chỉ số khối cơ thể ≥25, béo là ≥30. Khuyến cáo này mang mức khuyến nghị B. Tài liệu còn viết:「对血糖正常的成年人，每 3 年筛查一次是合理的做法」 — với người lớn có đường huyết bình thường, sàng lọc mỗi 3 năm một lần là hợp lý +- Chi phí: Xét đường huyết lúc nhịn hơn mười tệ. Xét đường huyết trung bình ba tháng vài chục tệ. Ra bệnh viện lấy một ống máu là xong. +- Nói thẳng: Bỏ ra vài chục tệ lấy một ống máu thì biết mình đã có tiền đái chưa, lúc chưa có triệu chứng nào. Khuyến cáo sàng lọc chính thức của Mỹ là: người 35 đến 70 tuổi, thừa cân hoặc béo, đi xét. Xét ra bình thường thì ba năm xét lại một lần. Mỹ tính thừa cân từ chỉ số khối cơ thể 25, với người châu Á nới lên 23; tiêu chuẩn Trung Quốc là từ 24 đã tính thừa cân. +- Lợi ích: Nhóm dịch vụ phòng ngừa Hoa Kỳ là cơ quan Mỹ đưa ra khuyến cáo sàng lọc. Khuyến cáo hiện tại là: sàng lọc tiền đái và đái thường loại 2 cho người lớn 35 đến 70 tuổi, thừa cân hoặc béo, dù chưa có triệu chứng nào. Thừa cân là chỉ số khối cơ thể ≥25, béo là ≥30. Với người Mỹ gốc Á, khuyến cáo hạ ranh giới xuống chỉ số khối cơ thể ≥23, vì ở cùng một chỉ số khối cơ thể thì người châu Á nguy cơ đái thường cao hơn. Tiêu chuẩn ngành của Trung Quốc là: chỉ số khối cơ thể từ 24.0 đến 28.0 là thừa cân, từ 28.0 trở lên là béo. Mức khuyến nghị của khuyến cáo này là B. Khuyến cáo còn viết rõ:「对血糖正常的成年人，每 3 年筛查一次是合理的做法」— với người lớn đường huyết bình thường thì mỗi 3 năm sàng lọc một lần là hợp lý. - Mức bằng chứng: A -- Nguồn: US Preventive Services Task Force (2021). Screening for Prediabetes and Type 2 Diabetes. -- Ghi chú: Phát hiện tiền đái thường không có nghĩa là phải uống thuốc. Khá nhiều người chỉ cần đổi lối sống là xoay ngược lại được. Ở người lớn Trung Quốc, số biết mình mắc đái thường luôn là thiểu số, nhiều người đợi đến lúc đã có biến chứng mới biết +- Nguồn: US Preventive Services Task Force (2021). Screening for Prediabetes and Type 2 Diabetes. ：「at a lower BMI (≥23) if the patient is Asian American」；Văn phòng Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委办公厅) (2024). 居民体重管理核心知识（2024 年版）释义：「根据《成年人体重判定》（WS/T 428-2013）行业标准……24.0≤BMI＜28.0 为超重，BMI≥28.0 为肥胖」. +- Ghi chú: Xét ra tiền đái không có nghĩa là phải uống thuốc. Khá nhiều người chỉ cần đổi lối sống là đã đưa nó trở lại bình thường. Ở người lớn Trung Quốc, số biết mình mắc đái thường vẫn là thiểu số, nhiều người chỉ biết khi đã có biến chứng. + ### 9. Không chạy quá tốc độ, không lái xe sau khi uống rượu - Chi phí: Không tốn tiền. Chạy chậm hơn một chút thì mỗi chuyến mất thêm vài phút. Chỗ khó là ở bàn ăn cơm phải uống ít đi. @@ -93,23 +95,25 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: B - Nguồn: Spiegel CN, Lindaman FC (1977). Children can't fly: a program to prevent childhood morbidity and mortality from window falls. American Journal of Public Health：「Significant reduction in falls resulted, particularly in the Bronx, where reported falls declined 50 percent from 1973 to 1975」— số vụ ngã giảm đáng kể, nổi bật ở Bronx, nơi số vụ ngã được báo cáo giảm một nửa từ 1973 đến 1975, và bản sửa đổi bộ luật vệ sinh thành phố New York năm 1976「to require that landlords provide window guards in apartments where children ten years old and younger reside」— buộc chủ nhà phải lắp thanh chắn cửa sổ ở căn hộ có trẻ em từ mười tuổi trở xuống. ；(2021). Unintentional Window Falls in Children and Adolescents. Academic Pediatrics：38.840 lượt khám cấp cứu ghi nhận trong hệ thống giám sát tổn thương điện tử toàn quốc từ tháng 1 năm 2007 đến tháng 8 năm 2017，「The majority of falls occurred in children under the age of 6 and were related to falls from a second story or below」— phần lớn vụ ngã xảy ra ở trẻ dưới 6 tuổi và liên quan đến rơi từ tầng hai trở xuống. - Ghi chú: Không đặt giường, sofa hay thùng ở mép cửa sổ và mép ban công, những thứ trẻ leo lên được. Lưới chống muỗi chỉ chặn muỗi, đẩy là rơi, đừng coi là thanh chắn. Cũng không phải chỉ ai ở tầng cao mới cần lắp: trong số liệu của Mỹ, phần lớn vụ ngã xảy ra ở tầng hai hoặc thấp hơn. -### 12. Trẻ nhỏ ở gần nước thì không được rời mắt, chèo thuyền hay bơi ngoài vùng nước an toàn thì mặc áo phao + +### 12. Trẻ em ở gần nước thì không được rời mắt, chèo thuyền hay bơi ngoài trời thì mặc áo phao -- Chi phí: Áo phao 50 đến 200 tệ. Ngoài ra còn phải bỏ ra phần tâm trí để trông trẻ. -- Nói thẳng: Khi rơi xuống nước mà có mặc áo phao, xác suất bị chết đuối chỉ còn khoảng một nửa. Đến nay đuối nước vẫn là nguyên nhân tử vong số một ở trẻ 1 đến 14 tuổi Trung Quốc. Tỷ lệ tử vong do đuối nước ở nông thôn gấp khoảng hai lần thành phố. -- Lợi ích: Một nghiên cứu dùng số liệu của Bộ Cảnh gia Hoa Kỳ, ghép từng cặp người có điều kiện gần giống nhau để so sánh. Người ngồi trên thuyền du lịch bị rơi xuống nước mà mặc áo phao có nguy cơ chết đuối thấp hơn khoảng một nửa. Giá trị gốc là RR 0.51 sau hiệu chỉnh, khoảng tin cậy 95% CI 0.35 đến 0.74. Ở Trung Quốc, tỷ lệ tử vong do đuối nước dưới 20 tuổi giảm từ 6.60 trên 100.000 người năm 2013 xuống 3.28 trên 100.000 người năm 2021. Nông thôn gấp khoảng 2 lần thành phố. Đuối nước vẫn là nguyên nhân tử vong số một ở trẻ 1 đến 14 tuổi. Năm 2021, trong số ca tử vong do tổn thương ở nhóm 0 đến 19 tuổi, đuối nước chiếm 31.1%, tai nạn giao thông đường bộ chiếm 27.9% +- Chi phí: Áo phao 50 đến 200 tệ. Ngoài ra còn phải bỏ ra phần sự chú ý dành cho việc trông con. +- Nói thẳng: Khi rơi xuống nước mà có mặc áo phao thì xác suất bị chết đuối chỉ khoảng một nửa so với không mặc. Đến nay đuối nước vẫn là nguyên nhân tử vong đứng đầu ở trẻ 1 đến 14 tuổi Trung Quốc. Tỷ lệ tử vong đuối nước ở nông thôn cao gấp khoảng hai lần thành phố. +- Lợi ích: Nghiên cứu làm trên số liệu của Bộ Cảnh vệ Bờ biển Hoa Kỳ ghép từng cặp người có điều kiện gần giống nhau rồi so sánh. Với người đi thuyền du lịch rồi rơi xuống nước, nguy cơ chết đuối khi mặc áo phao thấp hơn khoảng một nửa. Giá trị gốc là RR 0.51 sau khi hiệu chỉnh, khoảng tin cậy 95% CI 0.35 đến 0.74. Tỷ lệ tử vong đuối nước ở người dưới 20 tuổi Trung Quốc giảm từ 6.60 trên 100.000 người năm 2013 xuống 3.28 trên 100.000 người năm 2021. Nông thôn cao gấp khoảng 2 lần thành phố. Đuối nước vẫn là nguyên nhân tử vong đứng đầu ở trẻ 1 đến 14 tuổi. Năm 2021, trong số ca tử vong do tổn thương ở nhóm 0 đến 19 tuổi, đuối nước chiếm 31.1%, giao thông đường bộ chiếm 27.9% - Mức bằng chứng: A - Nguồn: Cummings P và cộng sự (2011). Association between wearing a personal floatation device and death by drowning among recreational boaters. Injury Prevention. ; Li Z và cộng sự (2023). Unintentional Drowning Mortality Among Individuals Under Age 20 — China, 2013–2021. China CDC Weekly. ; Zhou J và cộng sự (2024). Injury Mortality of Children and Adolescents Aged 0–19 Years — China, 2010–2021. China CDC Weekly. -- Ghi chú: Mức A chỉ áp dụng cho phần áo phao.「Không rời mắt」là đồng thuận chung, tính riêng chỉ ở mức C, không có thử nghiệm chia nhóm ngẫu nhiên, và sau này cũng sẽ không có. Ở Trung Quốc, các ca chết đuối tập trung ở nông thôn, mùa hè, và ở nam giới 15 đến 19 tuổi. +- Ghi chú: Mức A chỉ đúng phần áo phao.「Không rời mắt」là đồng thuận chung, tính riêng chỉ được mức C, không có thử nghiệm chia nhóm ngẫu nhiên, sau này cũng sẽ không có. Ở Trung Quốc, ca chết đuối nước tập trung ở nông thôn, mùa hè, và nam giới 15 đến 19 tuổi. Mùa đông lên băng chơi, xem mục 41 của phần này (đo độ dày băng ngoài trời trước). -### 13. Từ 60 tuổi, tập giữ thăng bằng và sức chân, cải tạo phòng tắm và cầu thang trong nhà +### 13. Trên 60 tuổi tập thăng bằng và sức chân, cải tạo phòng tắm và cầu thang trong nhà -- Chi phí: Mỗi tuần tập 2 đến 3 lần, mỗi lần 30 phút thái cực hoặc bài tập giữ thăng bằng. Thảm chống trượt, thanh vịn, đèn ngủ trong phòng tắm vài chục đến vài trăm tệ. Chỗ khó là phải tập bền nhiều năm. -- Nói thẳng: Tập giữ thăng bằng mỗi tuần hai ba lần, thái cực chẳng hạn, số lần ngã ở người cao tuổi giảm khoảng hai phần rưỡi, số người từng ngã giảm 15%. Làm chống trượt và lắp thanh vịn trong nhà thì giảm thêm khoảng hai phần mười. Ngã là nguyên nhân tử vong do tổn thương số một ở người Trung Quốc trên 65 tuổi, và hơn một nửa số vụ xảy ra ngay trong nhà họ. -- Lợi ích: Cochrane gộp 108 thử nghiệm, 23.407 người lại với nhau. Tập thể dục làm số lần ngã ở người cao tuổi giảm 23%. Giá trị gốc là tỷ lệ ngã RaR 0.77, khoảng tin cậy 95% CI 0.71 đến 0.83. Số người từng ngã giảm 15% (RR 0.85). Mời người đến nhà đánh giá an toàn và cải tạo, tỷ lệ ngã giảm khoảng hai phần mười (RR 0.81, 95% CI 0.68 đến 0.97). Tập thái cực, nguy cơ ngã thấp hơn khoảng 29% (RR 0.71). Ở Trung Quốc, ngã là nguyên nhân tử vong do tổn thương số một ở người trên 65 tuổi. Trong năm 2018, 55.97% số vụ ngã được theo dõi ở người cao tuổi xảy ra tại nhà +- Chi phí: Mỗi tuần tập 2 đến 3 lần, mỗi lần 30 phút tập thái cực quyền hoặc bài tập thăng bằng. Thảm chống trượt, tay vịn, đèn ngủ trong phòng tắm vài chục đến vài trăm tệ. Chỗ khó là phải tập lâu dài. +- Nói thẳng: Mỗi tuần tập thăng bằng hai ba lần, chẳng hạn thái cực quyền, số lần người cao tuổi bị ngã giảm khoảng hai phần mười, số người từng bị ngã giảm 15%. Làm chống trượt trong nhà, lắp tay vịn thì giảm thêm khoảng hai phần mười. Ngã là nguyên nhân tử vong do tổn thương đứng đầu ở người cao tuổi trên 65 tuổi Trung Quốc, và hơn một nửa rơi ngay trong nhà mình. +- Lợi ích: Cochrane gộp 108 thử nghiệm, 23.407 người lại với nhau. Tập thể dục làm số lần ngã ở người cao tuổi giảm 23%. Giá trị gốc là tỷ lệ nguy cơ ngã RaR 0.77, khoảng tin cậy 95% CI 0.71 đến 0.83. Số người từng bị ngã giảm 15%（RR 0.85）. Mời người tới nhà làm đánh giá an toàn và cải tạo thì tỷ lệ ngã giảm khoảng hai phần mười（RR 0.81, 95% CI 0.68 đến 0.97）. Tập thái cực quyền thì nguy cơ ngã thấp hơn khoảng 29%（RR 0.71）. Ở Trung Quốc, ngã là nguyên nhân tử vong do tổn thương đứng đầu ở người cao tuổi trên 65 tuổi. Trong số ca ngã của người cao tuổi theo giám sát năm 2018, có 55.97% xảy ra trong nhà - Mức bằng chứng: A - Nguồn: Sherrington C và cộng sự (2019). Exercise for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Gillespie LD và cộng sự (2012). Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. ; Lu Z và cộng sự (2021). Characteristics of Falls Among Older People — China, 2018. China CDC Weekly. -- Ghi chú: Mục này áp dụng cho người trên 60 tuổi sống ở nhà mình, không ở viện dưỡng lão. Các nghiên cứu trên đếm số lần ngã, không đếm số người chết. Đường đi từ một cú ngã đến gãy xương hông rồi đến tử vong là rất ngắn. +- Ghi chú: Mục này áp dụng cho người trên 60 tuổi sống trong nhà mình, không ở viện dưỡng lão. Nghiên cứu trên đếm số lần ngã, không phải số người chết. Đường đi từ cú ngã tới gãy xương hông rồi tới cái chết của người cao tuổi rất ngắn. Không nhìn rõ đường cũng ngã nhiều hơn, bệnh đục thủy tinh thể ảnh hưởng tầm nhìn thì xem mục 9 ở phần 17 (phẫu thuật đục thủy tinh thể). Xương có chịu được cú ngã không, xem mục 39 của phần này (xét mật độ xương). + ### 14. Xét nghiệm 5 chỉ số viêm gan B, không có kháng thể thì tiêm vaccine - Chi phí: Xét nghiệm một lần vài chục tệ. Vaccine viêm gan B cho người lớn gồm 3 mũi, tổng cộng 100 đến 300 tệ. Tiêm hết trong nửa năm. @@ -142,6 +146,7 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: A - Nguồn: US Preventive Services Task Force (2024). Breast Cancer: Screening. - Ghi chú: Bắt đầu sàng lọc từ tuổi nào thì mỗi nước khác nhau, ở Trung Quốc có hướng dẫn và chương trình địa phương bắt đầu từ 45 tuổi. Người có cha mẹ, con cái hay anh chị em ruột mắc ung thư vú, và người mang đột biến BRCA, thuộc nhóm nguy cơ cao, phải tự đi tìm bác sĩ để lập kế hoạch riêng, mục này không áp dụng + ### 18. Phụ nữ trên 30 tuổi làm xét nghiệm sàng lọc ung thư cổ tử cung, ưu tiên xét nghiệm HPV - Chi phí: Xét nghiệm HPV một lần 100 đến 300 tệ. Kết quả âm tính thì 5 năm xét một lần. Lấy mẫu chỉ mất vài phút. @@ -150,7 +155,6 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: A - Nguồn: Sankaranarayanan R và cộng sự (2009). HPV screening for cervical cancer in rural India. New England Journal of Medicine. - Ghi chú: Người đã tiêm vaccine HPV vẫn phải sàng lọc, vì vaccine không phủ hết các type virus. Chỗ đặc biệt của nghiên cứu này là: nó là thử nghiệm sàng lọc chia nhóm ngẫu nhiên, lấy chính số người tử vong làm kết quả, và chỉ làm một vòng sàng lọc. - ### 19. Từ 45 đến 50 tuổi làm xét nghiệm sàng lọc ung thư trực tràng và ruột già, bằng xét nghiệm miễn dịch hoá học tìm máu ẩn trong phân hoặc nội soi đại tràng - Chi phí: Xét nghiệm phân vài chục tệ, mỗi 1 đến 2 năm làm một lần. Nội soi đại tràng vài trăm đến hơn một nghìn tệ, kết quả âm tính thì 10 năm làm một lần. Trước khi nội soi phải chừa một ngày để làm sạch ruột. @@ -199,6 +203,7 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: A - Nguồn: National Lung Screening Trial Research Team (2011). Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. New England Journal of Medicine. - Ghi chú: Mục này chỉ áp dụng cho người hút thuốc nguy cơ cao. Trong cùng thử nghiệm đó, 24.2% người chụp CT liều thấp có kết quả khả nghi. Trong số này 96.4% chỉ là hoảng loạn thật, tức dương tính giả. Người không hút thuốc đi làm xét nghiệm này là tiền mua lo âu. + ### 25. Bị trầm cảm hoặc có ý nghĩ tự sát thì bấm 12356, trong nhà không tích trữ thuốc ngủ và thuốc trừ sâu - Chi phí: Không tốn tiền. Chỉ cần gọi một cuộc. Ngoài ra cất khoá thuốc, không để thuốc trừ sâu trong nhà. @@ -207,7 +212,6 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: B - Nguồn: Zalsman G và cộng sự (2016). Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry. ; Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委) (2024). 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）— Thông báo về việc dùng số điện thoại 12356 làm đường dây nóng hỗ trợ tâm lý thống nhất toàn quốc. - Ghi chú: Lý do cho mức B: phần lớn bằng chứng về hạn chế phương tiện đến từ nghiên cứu sinh thái, cách làm là so số liệu trước và sau khi chính sách ban hành. Bản thân đường dây nóng không có bằng chứng nào cho thấy giảm tỷ lệ tử vong. Ý nghĩ tự sát thường chỉ kéo dài mấy phút đến mấy giờ, dời xa thứ có thể gây chết một bước cũng đã là can thiệp. Dữ liệu ở quy mô thời gian này, và kết cục dài hạn sau lần không thành công, xem mục 32 của phần này (ý nghĩ tự sát vừa nảy thì kể ngay một người bên cạnh). Sau khi được cứu, còn lại những gì, xem mục 33 của phần này. - ### 26. Chuẩn bị bình chữa cháy, chăn chữa cháy, mặt nạ thở để chạy thoát và túi sơ cứu, kiểm tra mỗi năm một lần - Chi phí: Một bình chữa cháy, một chăn chữa cháy, mỗi người một mặt nạ thở, thêm một túi sơ cứu, cả bộ vài trăm tệ. Mua một lần là xong, sau đó mỗi năm dành mười phút kiểm tra. @@ -232,14 +236,15 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: A - Nguồn: Vlachopoulos CV, Terentes-Printzios DG, Ioakeimidis NK, Aznaouridis KA, Stefanadis CI (2013). Prediction of cardiovascular events and all-cause mortality with erectile dysfunction: a systematic review and meta-analysis of cohort studies. Circulation: Cardiovascular Quality and Outcomes, 6(1), 99-109. ；DailyMed (thư viện dữ liệu dược phẩm của Thư viện Y khoa Quốc gia Hoa Kỳ, 美国国立医学图书馆). VIAGRA (sildenafil citrate) tablet, CONTRAINDICATIONS. - Ghi chú: Cả 14 nghiên cứu đó đều theo dõi một nhóm người rồi xem kết quả, không phải thử nghiệm chia nhóm ngẫu nhiên. Vì thế「nhiều hơn bốn phần mười」là liên hệ, không phải quan hệ nhân quả. Thỉnh thoảng một hai lần thì không tính, phải kéo dài trên ba tháng mới xem là vấn đề. Những thứ cần khám là cùng một bó với mục 7 và 8 của phần này: đo huyết áp, xét đường huyết lúc nhịn, cộng mỡ máu, hút thuốc và vòng eo. Đừng mua「thuốc bổ sung」cường dương trên mạng. Thuốc đúng loại là thuốc kê đơn, tờ hướng dẫn liệt kê nhóm nitrat là chống chỉ định, người đang dùng nitroglycerin dạng tương tự thì không được dùng, dùng chung sẽ làm huyết áp tụt mạnh. Cải thiện bằng cách nào xem mục 29 của phần này (giảm cân, cai thuốc làm chức năng cương tốt lên). Lợi ích ghi「vừa」chứ không ghi「lớn」: 44% là chênh lệch nguy cơ giữa hai nhóm người, không phải「khám một lần giảm được bao nhiêu tử vong」. Cũng không ghi「nhỏ」: loại tín hiệu như mục 27 của phần này (máu trong nước tiểu nhìn thấy bằng mắt) chỉ có tỷ lệ phát hiện, còn ở đây những thứ phát hiện ra như huyết áp cao, tiểu đường, mỡ máu bất thường đều có sẵn cách điều trị. -### 29. Giảm cân, cai thuốc, giữ huyết áp và đường huyết ổn định, chức năng cương cũng tốt lên theo +### 29. Giảm cân, cai thuốc lá, kiểm soát huyết áp và đường huyết thì chức năng cương tốt lên theo -- Chi phí: Không tốn tiền, còn có thể tiết kiệm. Chỗ khó là duy trì lâu dài: kiềm chế ăn, vận động, cai thuốc. -- Nói thẳng: Kiểm soát được các yếu tố nguy hiểm tim mạch thì bản thân chức năng cương cũng tốt lên. Dùng một bảng chấm điểm, trung bình tăng được 2 đến 3 điểm. Tổng điểm của bảng này là 25 điểm, từ 22 điểm trở lên là bình thường. -- Lợi ích: Một phân tích tổng hợp gộp 6 thử nghiệm đối chứng ngẫu nhiên, 740 nam giới. Can thiệp là đổi lối sống, hoặc dùng thuốc kiểm soát yếu tố nguy hiểm tim mạch, theo dõi ít nhất 6 tuần. Kết quả: điểm bảng chấm chức năng cương (IIEF-5) tăng trung bình 2.66 điểm（95% CI 1.86 đến 3.47, đây là khoảng tin cậy）. Bỏ 2 thử nghiệm dùng statin (143 người), 4 thử nghiệm còn lại chỉ dựa vào đổi lối sống (597 người) vẫn tăng 2.40 điểm（1.19 đến 3.61）. IIEF-5 là bảng hỏi 5 câu, tổng điểm từ 5 đến 25. Nguyên văn chia thành năm bậc theo điểm, từ 22 đến 25 điểm là không rối loạn, từ 5 đến 7 điểm là nặng +- Chi phí: Không tốn tiền, có khi còn tiết kiệm. Chỗ khó là duy trì lâu dài: kiềm chế ăn, vận động, cai thuốc. +- Nói thẳng: Kiểm soát được các yếu tố nguy cơ tim mạch thì bản thân chức năng cương cũng tốt lên. Chấm điểm bằng một bảng hỏi thì điểm trung bình tăng được 2 đến 3 điểm. Bảng hỏi này tổng 25 điểm, từ 22 điểm trở lên là bình thường. +- Lợi ích: Một phân tích tổng hợp gộp 6 thử nghiệm đối chứng ngẫu nhiên, 740 nam giới lại với nhau. Can thiệp là đổi lối sống, hoặc dùng thuốc kiểm soát yếu tố nguy cơ tim mạch, theo dõi ít nhất 6 tuần. Kết quả: điểm bảng hỏi chức năng cương（IIEF-5）tăng trung bình 2.66 điểm（95% CI 1.86 đến 3.47, đây là khoảng tin cậy）. Bỏ 2 thử nghiệm dùng statin（143 người）ra thì 4 thử nghiệm còn lại, thuần đổi lối sống（597 người）, vẫn tăng 2.40 điểm（1.19 đến 3.61）. IIEF-5 là bảng hỏi 5 câu, tổng điểm từ 5 đến 25. Nguyên văn chia điểm thành năm bậc, 22 đến 25 điểm là không có rối loạn, 5 đến 7 điểm là nặng - Mức bằng chứng: A - Nguồn: Gupta BP, Murad MH, Clifton MM, Prokop L, Nehra A, Kopecky SL (2011). The effect of lifestyle modification and cardiovascular risk factor reduction on erectile dysfunction: a systematic review and meta-analysis. Archives of Internal Medicine, 171(20), 1797-1803. ；Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM (1999). Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. International Journal of Impotence Research, 11(6), 319-326. -- Ghi chú: Sáu thử nghiệm này can thiệp khác nhau, có thử nghiệm giảm cân cộng vận động, có thử nghiệm dùng thuốc kiểm soát huyết áp và mỡ máu. Tổng cộng chỉ có 740 người. Nên hướng thì rõ, còn cụ thể tốt đến đâu thì tuỳ từng người. Lợi ích ghi「nhỏ」là vì điểm cuối chỉ là điểm bảng hỏi, không có con số kiểu「sau này bớt được bao nhiêu bệnh」. Cùng mấy động tác ấy ở nơi khác lại có điểm cuối cứng: cai thuốc xem mục 1 ở phần 2 (cai thuốc, càng sớm càng tốt), giảm cân xem mục 33 ở phần 2 (giữ BMI ở mức 20–25), huyết áp xem mục 7 của phần này (đo huyết áp, cao thì uống thuốc). Bản thân việc uống thuốc không phải vùng cấm, điều kiện là thuốc kê đơn do bác sĩ kê.「Thực phẩm chức năng」bán trên mạng hay trộn thêm thành phần loại này, liều không rõ, cách phân biệt xem mục 4 ở phần 28 (không mua thuốc giảm cân hứa hẹn nhanh). +- Ghi chú: 6 thử nghiệm đó can thiệp khác nhau, có cái giảm cân kèm vận động, có cái dùng thuốc kiểm soát huyết áp và mỡ máu. Tổng cộng chỉ có 740 người. Nên hướng thì rõ, còn cụ thể tốt thêm được bao nhiêu thì tùy người. Lợi ích ghi「nhỏ」là vì điểm cuối chỉ có điểm bảng hỏi, không có con số「về sau mắc bao nhiêu bệnh hơn」. Mấy việc đó ở nơi khác có điểm cuối cứng: cai thuốc lá xem mục 1 ở phần 2 (cai thuốc lá, càng sớm càng tốt), giảm cân xem mục 32 ở phần 2 (giữ chỉ số khối cơ thể ở mức 20–25), huyết áp xem mục 7 của phần này (đo huyết áp, cao thì uống thuốc). Uống thuốc không phải chuyện cấm, với điều kiện đó là thuốc bác sĩ kê đơn.「Thực phẩm bảo vệ sức khoẻ」bán trên mạng hay trộn thêm các thành phần loại này vào, liều không rõ, cách phân biệt xem mục 4 ở phần 28 (đừng mua thuốc giảm cân hứa giảm nhanh). + ### 30. Dùng bao cao su suốt quá trình quan hệ tình dục, không dùng chung kim tiêm với người khác @@ -249,22 +254,24 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: A - Nguồn: Giannou FK, Tsiara CG, Nikolopoulos GK, et al. (2016). Condom effectiveness in reducing heterosexual HIV transmission: a systematic review and meta-analysis of studies on HIV serodiscordant couples. Expert Review of Pharmacoeconomics & Outcomes Research, 16(4), 489-499. ；Fonner VA và cộng sự (2016). Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS. （phòng trước phơi nhiễm PrEP） - Ghi chú: Chênh lệch giữa「dùng không đều」và「không dùng」rất nhỏ, phải dùng mỗi lần, suốt quá trình mới có tác dụng. Dùng chung kim tiêm là con đường lây hiệu quả khác, độc lập với quan hệ tình dục, tiêm ma tuý, xăm mình, xuyên lỗ tai đều tính vào. Nhóm nguy cơ cao còn dùng được thuốc phòng trước phơi nhiễm (PrEP), tức uống thuốc trước để chặn nhiễm. Gộp 18 nghiên cứu lại, mọi thử nghiệm có trên bảy phần mười người uống thuốc đúng giờ thì người dùng PrEP có nguy cơ nhiễm HIV thấp hơn nhóm giả dược khoảng bảy phần mười. Giá trị gốc là tỷ số nguy cơ 0.30, khoảng tin cậy 95% CI 0.21 đến 0.45. Ở những thử nghiệm có ít người uống thuốc đúng giờ thì không thấy tác dụng bảo vệ. PrEP là thuốc kê đơn, đi hỏi khoa bệnh truyền nhiễm hoặc trung tâm kiểm soát bệnh, đừng tự mua trên mạng. Sau khi đã phơi nhiễm thì xử lý thế nào, xem phần 13 -### 31. Từng có hành vi nguy cơ cao thì xét nghiệm AIDS một lần, miễn phí ở Trung tâm kiểm soát bệnh và giữ kín kết quả +### 31. Đã từng có hành vi nguy cơ cao thì đi xét HIV/AIDS một lần, trung tâm kiểm soát bệnh miễn phí, kết quả được giữ kín -- Chi phí: Không tốn tiền. Một chuyến tới Trung tâm kiểm soát bệnh hoặc cơ sở y tế được chỉ định, rút một ống máu. Chỗ khó là bước mở miệng đi xét nghiệm. -- Nói thẳng: Đi xét nghiệm AIDS ở Trung tâm kiểm soát bệnh không mất tiền. Theo《全国艾滋病检测工作管理办法》, nhân viên không được tiết lộ tên, địa chỉ và kết quả xét nghiệm của bạn. Kết quả dương tính không phải bản án tử hình. Người uống thuốc đều đặn đẩy lượng virus xuống dưới ngưỡng phát hiện sẽ không lây sang bạn tình. Nghiên cứu đó theo dõi hơn bảy trăm cặp đôi và hơn bảy mươi nghìn lần quan hệ không dùng bao cao, không có ca nào lây trong cặp. -- Lợi ích: 《艾滋病防治条例》quy định rõ: ở cấp huyện trở lên, cơ quan quản lý y tế địa phương sẽ chỉ định một số cơ sở y tế. Các cơ sở này「应当为自愿接受艾滋病咨询、检测的人员免费提供咨询和初筛检测」— phải tư vấn và xét nghiệm sàng lọc ban đầu miễn phí cho người tự nguyện muốn được tư vấn và xét nghiệm AIDS. Xét sớm thì có tác dụng thật. Thuốc kháng virus hiện nay đẩy được lượng virus xuống dưới ngưỡng phát hiện. Tổ chức Y tế Thế giới nói「接受抗病毒治疗且病毒载量检测不到的感染者不会把 HIV 传染给性伴侣」— người nhiễm HIV đang điều trị kháng virus và không còn phát hiện được lượng virus sẽ không lây HIV sang bạn tình. Nghiên cứu PARTNER theo dõi 782 cặp đôi cùng giới, mỗi cặp một người nhiễm một người không nhiễm, gọi tên khoa học là serum khác nhau. Tổng cộng 1593 cặp năm, tức cộng số năm theo dõi của từng cặp. Trong thời gian đó có 76088 lần quan hệ không dùng bao cao và 15 ca nhiễm mới. Sau khi so gen virus bằng phân tích phát sinh học, cả 15 ca đều không đến từ bạn tình. Lây trong cặp bằng không. 《全国艾滋病检测工作管理办法》còn quy định nhân viên phòng xét nghiệm AIDS không được tiết lộ tên, địa chỉ, kết quả xét nghiệm và các thông tin liên quan của người mắc AIDS hay người nhiễm HIV +- Chi phí: Không tốn tiền. Một chuyến tới trung tâm kiểm soát bệnh hoặc cơ sở y tế được chỉ định, rút một ống máu. Chỗ khó là bước mở miệng đi xét. +- Nói thẳng: Đi xét HIV/AIDS ở trung tâm kiểm soát bệnh không tốn tiền. Không có sự đồng ý của chính bạn thì không đơn vị, cá nhân nào được tiết lộ tên, địa chỉ, nơi làm việc và bệnh sử của bạn. Xét ra dương tính cũng không phải kết án tử hình: người uống thuốc đều đặn đủ để nén tải lượng vi-rút xuống mức không đo ra sẽ không lây sang bạn tình nữa. Nghiên cứu đó theo hơn bảy trăm cặp bạn tình và hơn bảy mươi nghìn lần quan hệ không dùng bao cao su, lây giữa hai người tình là không có ca nào. +- Lợi ích:《艾滋病防治条例》Điều hai mươi ba quy định rõ. Cơ quan chuyên ngành y tế của chính quyền địa phương từ cấp huyện trở lên sẽ chỉ định một số cơ sở y tế. Những cơ sở đó「应当为自愿接受艾滋病咨询、检测的人员免费提供咨询和初筛检测」— phải cho người tự nguyện nhận tư vấn, xét nghiệm HIV/AIDS tư vấn và xét sàng lọc ban đầu miễn phí. Xét sớm thì có tác dụng thật: điều trị kháng vi-rút hiện nay nén được tải lượng vi-rút xuống mức không đo ra. Công chức của Tổ chức Y tế Thế giới là「接受抗病毒治疗且病毒载量检测不到的感染者不会把 HIV 传染给性伴侣」— người nhiễm đang điều trị kháng vi-rút mà tải lượng vi-rút không đo ra thì không lây HIV sang bạn tình. Nghiên cứu PARTNER theo dõi 782 cặp cùng giới tính, mỗi cặp đều một bên nhiễm, một bên không, tên khoa học gọi là huyết thanh không đồng nhất. Tổng cộng 1593 cặp năm (số năm theo dõi của từng cặp cộng lại). Trong thời gian đó có 76088 lần quan hệ không dùng bao cao su, xuất hiện 15 ca nhiễm mới. So sánh gen vi-rút rồi làm phân tích phát sinh loài thì cả 15 ca đều không đến từ bạn tình, lây lan giữa hai người tình bằng không. Điều ba mươi chín của《艾滋病防治条例》quy định:「未经本人或者其监护人同意，任何单位或者个人不得公开艾滋病病毒感染者、艾滋病病人及其家属的姓名、住址、工作单位、肖像、病史资料以及其他可能推断出其具体身份的信息。」 - Mức bằng chứng: A -- Nguồn: Bộ Y tế (卫生部) (2006). Quy định quản lý việc xét nghiệm AIDS toàn quốc, 全国艾滋病检测工作管理办法. ；Ủy ban Nhà nước (国务院) (2006). Nghị định phòng ngừa và xử lý AIDS, 艾滋病防治条例（第三、二十三、三十八条）, gồm cấm phân biệt đối xử với người nhiễm, quyền được tư vấn và xét nghiệm miễn phí, và nghĩa vụ khai báo của người nhiễm. （Ủy ban Y tế và Sức khỏe thành phố Bắc Kinh, 北京市卫生健康委员会, đăng lại）；Rodger AJ, Cambiano V, Bruun T, et al. (2019). Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. The Lancet, 393(10189), 2428-2438. ；Trung tâm Kiểm soát và Phòng ngừa bệnh Quảng Đông (广东省疾病预防控制中心). Giáo trình phổ thông về thời gian cửa sổ xét nghiệm AIDS, 艾滋病检测窗口期科普. （con số về thời gian cửa sổ） -- Ghi chú: Thời gian cửa sổ quyết định xét nghiệm vào lúc nào mới có ý nghĩa. Cơ sở của Trung tâm kiểm soát bệnh Quảng Đông (广东省疾控中心) là: xét nghiệm axit nucleic khoảng 1 tuần là phát hiện được, xét nghiệm kháng nguyên kết hợp kháng thể thế hệ thứ tư khoảng 2 tuần, xét nghiệm kháng thể thế hệ thứ ba khoảng 3 tuần. Nên chờ ít nhất 2 tuần rồi hãy xét, cách 2 đến 4 tuần xét lại một lần. Sau khi lần hành vi nguy cơ cao nhất đã tròn 3 tháng, trên 99.99% có thể loại trừ nhiễm. Que thử tự cho dương tính không có nghĩa là chẩn đoán, phải tới Trung tâm kiểm soát bệnh hoặc cơ sở chính quy để kiểm chứng. Điều 3 của nghị định quy định không cá nhân và tổ chức nào được phân biệt đối xử với người nhiễm và gia đình họ. Quyền về hôn nhân, việc làm, khám chữa bệnh, nhập học được pháp luật bảo vệ. Điều 38 quy định thêm rằng người nhiễm có nghĩa vụ「将感染或者发病的事实及时告知与其有性关系者」— kịp thời báo cho người có quan hệ tình dục với mình biết sự thật mình đã nhiễm hoặc đã mắc bệnh. Sàng lọc ba bệnh trong thai kỳ và phòng lây truyền từ mẹ sang con xem ở phần 27 -### 32. Vừa nảy ý tưởng tự sát thì nói với một người bên cạnh, giao mấy chục phút ấy ra +- Nguồn: Hội đồng Nhà nước (国务院) (2006). 艾滋病防治条例（第三、二十三、三十八、三十九条）. （Ủy ban Y tế và Sức khoẻ thành phố Bắc Kinh đăng lại）；Rodger AJ, Cambiano V, Bruun T, et al. (2019). Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. The Lancet, 393(10189), 2428-2438. ；Trung tâm Kiểm soát và Phòng ngừa bệnh tỉnh Quảng Đông (广东省疾病预防控制中心). 艾滋病检测窗口期科普. （con số về cửa sổ thời gian; trang gốc đã bị gỡ, liên kết là bản lưu trang web ngày 2026-09-15） +- Ghi chú: Cửa sổ thời gian quyết định xét vào lúc nào thì mới có ý nghĩa. Công chức của Trung tâm Kiểm soát và Phòng ngừa bệnh tỉnh Quảng Đông nói: xét nghiệm nucleic khoảng 1 tuần, xét kháng nguyên kết hợp kháng thể thế hệ thứ tư khoảng 2 tuần, xét kháng thể thế hệ thứ ba khoảng 3 tuần thì phát hiện được. Nên đợi ít nhất 2 tuần rồi hãy xét, cách 2 đến 4 tuần xét lại một lần. Sau mốc cuối cùng hành vi nguy cơ cao đủ 3 tháng thì trên 99.99% có thể loại trừ nhiễm. Bộ thử tự chẩn đoán hiện dương tính không có nghĩa là chẩn đoán, phải tới trung tâm kiểm soát bệnh hoặc cơ sở chính quy để đối chiếu lại. Điều ba của quy định nói rõ, không đơn vị, cá nhân nào được phân biệt đối xử với người nhiễm và gia đình họ; quyền lợi về hôn nhân, việc làm, khám chữa bệnh, nhập học được pháp luật bảo vệ. Điều ba mươi tám đồng thời quy định, người nhiễm có nghĩa vụ「将感染或者发病的事实及时告知与其有性关系者」— báo kịp cho người có quan hệ tình dục với mình biết chuyện đã nhiễm hoặc đã phát bệnh. Sàng lọc ba bệnh trong thai kỳ và chặn lây từ mẹ sang con xem phần 27. + +### 32. Ý nghĩ tự sát vừa nảy lên thì kể ngay một người bên cạnh, giao mấy chút phút ấy ra -- Chi phí: Không tốn tiền. Chỗ khó là lúc mở miệng, và thừa nhận rằng mấy tiếng đó mình không ở một mình được. -- Nói thẳng: Trong số người đã tự sát không thành công, gần một nửa nói rằng từ lúc nảy ý tưởng đến lúc ra tay không quá 10 phút. Nên thứ phải trấn an thường chỉ là mấy chục phút đó. Ba phần tư trong số họ có tiếp xúc với người bên cạnh trong khoảng thời gian ấy, chỉ là không mở miệng. Người phải vào bệnh viện vì tự làm hại mình, sau 1 năm khoảng 2% chết vì tự sát, sau trên 9 năm khoảng 7%. Đa số về sau không chết vì tự sát. -- Lợi ích: 82 người tự sát không thành công được phỏng vấn trong vòng 3 ngày sau khi nhập viện. 47.6% (39 người) nói rằng từ lần đầu nảy ý tưởng tự sát cho tới khi ra tay không quá 10 phút. Người có khoảng thời gian này kéo dài hơn thì điểm trên thang đo ý định tự sát rõ ràng cao hơn. Điểm trên thang đo hành vi bốc đồng không liên quan đến độ dài của khoảng thời gian. 76.8% (63 người) nói rằng trong khoảng thời gian đó họ có tiếp xúc nào đó với bạn tình, gia đình hoặc bạn bè. Kết luận của tác giả: khoảng thời gian để can thiệp thường rất ngắn, chuyên viên ở giai đoạn này có rất ít chỗ để xen vào, nên phải phổ biến dấu hiệu nhận biết và cách can thiệp cho người thường. Một bài phân tích tổng hợp khác tổng hợp 90 nghiên cứu theo dõi. Ở người đến khám vì tự làm hại mình, tỷ lệ tự làm hại lại mà không chết sau 1 năm, lấy con số trung vị là 16%. Sau 1 năm, tỷ lệ chết vì tự sát là 2%. Ở nhóm theo dõi trên 9 năm, khoảng 7% chết vì tự sát +- Chi phí: Không tốn tiền. Chỗ khó là lúc đó mở miệng, và chịu nhận ra mấy giờ ấy mình không ở một mình được. +- Nói thẳng: Gần một nửa số người từng tự sát không thành công nói rằng, từ lúc ý nghĩ nảy lên tới lúc ra tay không quá 10 phút, nên phần phải trụ qua thường chỉ là mấy chút phút ấy. Ba phần tư số người trong khoảng thời gian đó có tiếp xúc với người bên cạnh, chỉ là không mở miệng. Người từng vào viện vì tự làm hại mình, sau 1 năm khoảng 2% chết vì tự sát, trên 9 năm khoảng 7%, tuyệt đối phần lớn sau đó không chết vì tự sát. +- Lợi ích: 82 người từng tự sát không thành công được phỏng vấn trong 3 ngày nhập viện: 47.6%（39 người）nói, từ lần đầu tiên trong đợt này ý nghĩ tự sát nảy lên tới lúc thật sự ra tay không quá 10 phút. Người mà quá trình này kéo dài hơn một chút thì điểm ý định tự sát rõ ràng cao hơn. Điểm trên thang đo tính bốc đột không có liên quan gì với độ dài của quá trình. 76.8%（63 người）nói, trong khoảng thời gian đó có tiếp xúc nào đó với bạn tình, người nhà hoặc bạn bè. Kết luận của tác giả là: cửa sổ can thiệp thường rất ngắn, chuyên gia trong giai đoạn này can thiệp được rất hạn chế, nên phải phổ biến cách nhận biết dấu hiệu và cách can thiệp cho người thường. Một bài tổng quan hệ thống khác gộp 90 nghiên cứu theo dõi. Người tới khám vì tự làm hại mình, tỷ lệ sau 1 năm lại tự làm hại mình mà không chết, lấy giá trị trung vị là 16%. Sau 1 năm chết vì tự sát là 2%. Theo dõi trên 9 năm, khoảng 7% chết vì tự sát - Mức bằng chứng: B - Nguồn: Deisenhammer EA, Ing CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry, 70(1), 19-24. ；Owens D, Horrocks J, House A (2002). Fatal and non-fatal repetition of self-harm: systematic review. British Journal of Psychiatry, 181, 193-199. -- Ghi chú: Lý do xếp B: nghiên cứu của Deisenhammer chỉ có 82 người, lấy ở một bệnh viện, dựa vào trí nhớ kể lại sau sự việc của chính người trong cuộc. Bài của Owens đưa ra con số giữa của 90 nghiên cứu làm rất khác nhau, chứ không gộp chúng thành một số. Chính bài đó còn viết rõ nguy cơ tự sát của người tự làm hại mình vẫn cao hơn người thường hàng trăm lần. Vì vậy「绝大多数人后来没有死于自杀」nói về cả nhóm này, không phải bảo đảm cho một cá nhân, bệnh vẫn phải khám. Đưa tư liệu gây chết ra xa và 12356 xem mục 25 của phần này. Khi tâm trạng tụt thì làm vài việc sau, xem mục 19 ở phần 3. Di chứng sau khi được cứu xem mục 33 của phần này. Khi người bên cạnh lộ ra ý tưởng ấy mà bạn làm được gì, xem mục 15 ở phần 8 +- Ghi chú: Lý do cho mức B: nghiên cứu của Deisenhammer chỉ có 82 người, ở một bệnh viện, dựa vào người bệnh nhớ lại sau sự việc. Bài của Owens đưa ra tỷ lệ lấy giá trị trung vị của 90 nghiên cứu làm khác nhau, không phải gộp chúng thành một con số. Chính bài đó còn viết rõ, nguy cơ tự sát của người tự làm hại mình vẫn cao hơn người thường hàng trăm lần. Vì thế câu「tuyệt đối phần lớn sau đó không chết vì tự sát」nói về cả nhóm người bị hại này, không phải bảo đảm cho một người nào cả, có bệnh thì vẫn phải khám. Dời xa phương tiện gây chết và gọi 12356 xem mục 25 của phần này. Mấy việc nên làm trước khi tâm trạng xuống thấp xem mục 18 ở phần 3. Di chứng còn lại sau khi được cứu xem mục 33 của phần này. Khi người bên cạnh để lộ ý nghĩ ấy thì bạn làm được gì, xem mục 15 ở phần 8. + ### 33. Đừng coi「cứu được」là chốt chặn cuối cùng: uống thuốc trừ sâu hay hít khí than, cấp cứu giữ được mạng nhưng không giữ được phổi và não @@ -315,3 +322,31 @@ Phần này nói về ba thứ: cái chết do tai nạn và bạo lực như ta - Mức bằng chứng: A - Nguồn: Grant RM, Lama JR, Anderson PL, et al. (2010). Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. New England Journal of Medicine, 363(27), 2587–2599. ；McCormack S, Dunn DT, Desai M, et al. (2016). Pre-exposure prophylaxis to prevent the acquisition of HIV-1 infection (PROUD): effectiveness results from the pilot phase of a pragmatic open-label randomised trial. The Lancet, 387(10013), 53–60. ；Molina JM, Capitant C, Spire B, et al. (2015). On-demand preexposure prophylaxis in men at high risk for HIV-1 infection. New England Journal of Medicine, 373(23), 2237–2246. ；Fonner VA, Dalglish SL, Kennedy CE, et al. (2016). Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS, 30(12), 1973–1983. ；Xu JJ, Huang XJ, Liu XC, et al. (2020). Consensus statement on human immunodeficiency virus pre-exposure prophylaxis in China. Chinese Medical Journal, 133(23), 2840–2846. （bản tiếng Trung, 《中国 HIV 暴露前预防用药专家共识》）；World Health Organization (2019). What's the 2+1+1? Event-driven oral pre-exposure prophylaxis to prevent HIV for men who have sex with men (WHO/CDS/HIV/19.8). ；Ủy ban Nhà nước (国务院办公厅) (2024). Kế hoạch kiểm soát và phòng chống AIDS Trung Quốc (2024–2030), 中国遏制与防治艾滋病规划（2024—2030 年）. - Ghi chú: Người trong các thử nghiệm chủ yếu là nam giới có quan hệ tình dục với nam, nên nhóm này cũng là nhóm nên đi hỏi nhất. Người khác có hành vi nguy cơ cao cũng có thể hỏi bác sĩ xem có hợp với mình không. Trước khi kê thuốc bắt buộc phải xác nhận là chưa nhiễm, người đã nhiễm mà uống thì virus kháng thuốc, kết quả que thử tự không làm căn cứ được. Người có viêm gan B mà ngừng thuốc có thể làm viêm gan tái phát, nên phải xét trước khi kê. Tổ chức Y tế Thế giới chỉ khuyến nghị dùng theo nhu cầu cho nam giới có quan hệ tình dục với nam, hợp với người trung bình dưới 2 lần mỗi tuần. Cách uống là trước quan hệ tình dục từ 2 đến 24 giờ uống 2 viên, sau đó ở 24 giờ và 48 giờ mỗi lần uống 1 viên. Phụ nữ và người có viêm gan B chỉ có thể uống mỗi ngày. Ở các thử nghiệm dùng theo nhu cầu, nhóm uống thuốc có tác dụng phụ ở dạ dày ruột và thận nhiều hơn nhóm giả dược, nên không được bỏ tái khám. Trung Quốc bắt đầu thí điểm trong nhóm nam giới có quan hệ tình dục với nam từ năm 2019, kế hoạch năm 2024 viết là「规范实施」. Chính sách miễn phí toàn quốc không tìm thấy. Đi kê ở khoa bệnh truyền nhiễm hoặc phòng khám da liễu và AIDS của bệnh viện AIDS được chỉ định tại địa phương, gọi trung tâm kiểm soát bệnh địa phương trước để hỏi bệnh viện nào kê được. Thuốc không phòng được giang mai, lậu và các bệnh lây qua đường tình dục khác, bao cao su vẫn phải dùng, xem mục 30 của phần này (dùng bao cao su suốt quá trình). Đã xảy ra hành vi nguy cơ cao thì xử lý bù, xem mục 38 ở phần 13 (lấy thuốc phòng sau phơi nhiễm trong vòng 72 giờ). Người hưởng lợi chủ yếu là bạn, và người bạn đời cũng được che chở. + +### 39. Phụ nữ từ 65 tuổi nên làm một lần xét mật độ xương bằng X-quang năng lượng kép, phụ nữ sau mãn kinh có yếu tố nguy cơ loãng xương thì không cần đợi tới 65 tuổi + +- Chi phí: Tới bệnh viện làm một lần xét mật độ xương bằng X-quang năng lượng kép (gọi tắt là DXA), đo thắt lưng và hông, nằm vài phút là xong. Tự trả thường một hai trăm tệ, giá ở mỗi nơi khác nhau. Siêu âm gót chân, kiểu giẫm một cái trong gói khám sức khoẻ, chỉ sàng lọc được chứ không chẩn đoán được. +- Nói thẳng: Phụ nữ từ 65 tuổi trở lên, và phụ nữ sau mãn kinh có yếu tố nguy cơ loãng xương, nên đi xét mật độ xương một lần. Xét rủi ro trước, rồi mới cho người rủi ro cao dùng thuốc, gãy xương hông giảm được khoảng 17%. Quy ra tuyệt đối, cứ 1000 người đi xét thì khoảng 5 người không gãy xương hông. Tỷ lệ tử vong toàn nguyên không thấy thay đổi. +- Lợi ích: Khuyến cáo Nhóm dịch vụ phòng ngừa Hoa Kỳ ban hành năm 2025 là: phụ nữ từ 65 tuổi trở lên làm sàng lọc loãng xương (khuyến nghị mức B). Phụ nữ sau mãn kinh, chưa tới 65 tuổi, nhưng đánh giá thì nguy cơ gãy xương cao, cũng được khuyến cáo sàng lọc (mức B). Với nam giới thì bằng chứng chưa đủ nên không có lập trường. Bài tổng hợp bằng chứng làm cho khuyến cáo này gộp 3 thử nghiệm ngẫu nhiên, 42009 phụ nữ cao niên châu Âu. Cách sàng lọc là đánh giá nguy cơ gãy xương trước, người rủi ro cao mới xét mật độ xương, xét ra rủi ro cao thì cho dùng thuốc. Nhóm sàng lọc gãy xương hông thấp hơn khoảng 17%. Giá trị gốc RR 0.83, khoảng tin cậy 95% CI 0.73 đến 0.93. Quy ra tuyệt đối, cứ 1000 người thì sàng lọc giúp giảm 5 ca gãy xương hông. Trong đó, thử nghiệm SCOOP của Anh chọn 12483 phụ nữ 70 đến 85 tuổi, nhóm sàng lọc gãy xương hông thấp hơn khoảng 28%（HR 0.72, 95% CI 0.59 đến 0.89）, tỷ lệ tử vong không thấy khác biệt. Hướng dẫn chẩn đoán và điều trị bản 2017 của Hiệp hội Y khoa Trung Quốc đề nghị: phụ nữ từ 65 tuổi, nam giới từ 70 tuổi thì làm xét mật độ xương. Chưa tới tuổi đó mà có một hay nhiều yếu tố nguy cơ loãng xương thì cũng nên xét. Khảo sát toàn quốc 2017 đến 2018, tỷ lệ mắc loãng xương ở phụ nữ trên 40 tuổi là 20.6%, ở nam giới là 5.0%（Trung Quốc） +- Mức bằng chứng: A +- Nguồn: US Preventive Services Task Force, Nicholson WK, Silverstein M, et al. (2025). Screening for Osteoporosis to Prevent Fractures: US Preventive Services Task Force Recommendation Statement. JAMA. ; Kahwati LC, Kistler CE, Booth G, et al. (2025). Screening for Osteoporosis to Prevent Fractures: A Systematic Evidence Review for the US Preventive Services Task Force. JAMA. ; Shepstone L, Lenaghan E, Cooper C, et al. (2018). Screening in the community to reduce fractures in older women (SCOOP): a randomised controlled trial. The Lancet. ; Hiệp hội Y khoa Trung Quốc, Chi nhánh Loãng xương và Bệnh khoáng chất xương (中华医学会骨质疏松和骨矿盐疾病分会) (2017). 原发性骨质疏松症诊疗指南(2017). 中华骨质疏松和骨矿盐疾病杂志, 10(5), 413-443（表 5 骨密度测量的临床指征）. ; Wang L, Yu W, Yin X, et al. (2021). Prevalence of Osteoporosis and Fracture in China: The China Osteoporosis Prevalence Study. JAMA Network Open. +- Ghi chú: Tranh cãi. Kết cục chính định sẵn của SCOOP là mọi kiểu gãy xương do loãng xương, nhóm sàng lọc không giảm rõ（HR 0.94, 95% CI 0.85 đến 1.03）, chỉ giảm ở gãy xương hông. Cả ba thử nghiệm đều là「đánh giá rủi ro trước, rồi xét mật độ xương」, không thử nghiệm nào so riêng chỉ xét mật độ xương. Bài tổng hợp bằng chứng còn ước tính, trong số người xét ra có 11.8% đến 24.1% là chẩn đoán quá mức, tức suốt đời vốn không gãy xương vì lý do đó. Nam giới có nên xét không, Mỹ nói bằng chứng chưa đủ, hướng dẫn Trung Quốc đặt ở trên 70 tuổi, cứ theo lời bác sĩ. Mục này quản lý chuyện xương có chịu được cú ngã không, giảm cú ngã thì nhờ mục 13 của phần này (tập thăng bằng và sức chân), làm cả hai cùng lúc. Xét ra loãng xương rồi thì làm gì, xem mục 40 của phần này (xét ra loãng xương thì dùng thuốc). Người hưởng lợi chủ yếu là bạn, cùng cha mẹ bạn. + +### 40. Xét ra loãng xương, hoặc đã từng gãy xương chỉ vì ngã nhẹ, thì xin bác sĩ kê thuốc chống loãng xương và dùng đều, đừng thay bằng bổ sung canxi + +- Chi phí: Cần đơn thuốc, dùng liên tục mấy năm. Loại thường dùng có thuốc uống mỗi tuần một lần, cũng có thuốc tiêm truyền mỗi năm một lần. Mới bắt đầu dùng thuốc thì mỗi năm xét lại mật độ xương, ổn định rồi thì 1 đến 2 năm xét một lần. Chỗ khó là không có triệu chứng vẫn phải dùng đều. +- Nói thẳng: Người đã xác định loãng xương, hoặc đã từng gãy xương chỉ vì ngã nhẹ, dùng thuốc theo phác đồ của bác sĩ thì gãy xương hông giảm được khoảng một nửa, gãy xương sống giảm hơn một nửa. Người cao tuổi sau khi mổ gãy xương hông mà bắt đầu tiêm thuốc mỗi năm một lần thì trong hai năm gãy xương lại giảm khoảng ba phần mười lăm. Chỉ bổ sung canxi không thay thế được mấy loại thuốc đó. +- Lợi ích: Gãy xương giòn là kiểu gãy xương xảy ra khi chỉ chịu chấn thương nhẹ hoặc trong sinh hoạt thường ngày. Hướng dẫn bản 2017 của Hiệp hội Y khoa Trung Quốc chia người phải dùng thuốc thành ba nhóm. Một là người từng gãy xương giòn ở cột sống hoặc hông. Hai là người có chỉ số T của mật độ xương từ -2.5 trở xuống, có gãy xương hay không cũng vậy. Ba là người khối lượng xương thấp mà từng gãy xương giòn ở cổ tay, xương cánh tay hoặc khung chậu, hoặc người tính ra nguy cơ gãy xương cao. Hướng dẫn còn viết「目前尚无充分证据表明单纯补钙可以替代其他抗骨质疏松药物治疗」— hiện chưa có đủ bằng chứng cho thấy chỉ bổ sung canxi có thể thay thế các thuốc chống loãng xương khác. Năm 2025 Cochrane cập nhật tổng quan về natri alendronat. Ở phụ nữ sau mãn kinh đã có loãng xương hoặc đã gãy xương, nhóm dùng thuốc gãy xương hông thấp hơn khoảng 51%. Giá trị gốc RR 0.49, khoảng tin cậy 95% CI 0.25 đến 0.96. Gãy xương sống có triệu chứng thấp hơn khoảng 55%（RR 0.45, 95% CI 0.28 đến 0.73）. Thử nghiệm HORIZON chọn 2127 người đã mổ gãy xương hông, tuổi trung bình 74.5. Bắt đầu truyền acid zoledronic mỗi năm một lần trong vòng 90 ngày sau mổ, thời gian theo dõi trung vị là 1.9 năm. Gãy xương mới giảm từ 13.9% xuống 8.6%, giảm khoảng 35%. Tử vong giảm từ 13.3% xuống 9.6%, giảm khoảng 28%. Hướng dẫn viết, khoảng 20% người chết vì các biến chứng khác nhau trong vòng một năm sau gãy xương hông. Trong khảo sát toàn quốc 2017 đến 2018, phụ nữ đã xác định loãng xương hoặc đã gãy xương chỉ có 1.4% dùng thuốc, nam giới chỉ có 0.3%（Trung Quốc） +- Mức bằng chứng: A +- Nguồn: Hiệp hội Y khoa Trung Quốc, Chi nhánh Loãng xương và Bệnh khoáng chất xương (中华医学会骨质疏松和骨矿盐疾病分会) (2017). 原发性骨质疏松症诊疗指南(2017). 中华骨质疏松和骨矿盐疾病杂志, 10(5), 413-443（表 10 抗骨质疏松症药物治疗适应证；钙剂一节；疗效监测一节）. ; Wells GA, Hsieh SC, Peterson J, et al. (2025). Alendronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database of Systematic Reviews. ; Lyles KW, Colón-Emeric CS, Magaziner JS, et al. (2007). Zoledronic acid and clinical fractures and mortality after hip fracture. New England Journal of Medicine. ; Cummings SR, Lui LY, Eastell R, Allen IE (2019). Association Between Drug Treatments for Patients With Osteoporosis and Overall Mortality Rates: A Meta-analysis. JAMA Internal Medicine. ; Kahwati LC, Kistler CE, Booth G, et al. (2025). Screening for Osteoporosis to Prevent Fractures: A Systematic Evidence Review for the US Preventive Services Task Force. JAMA. ; Wang L, Yu W, Yin X, et al. (2021). Prevalence of Osteoporosis and Fracture in China: The China Osteoporosis Prevalence Study. JAMA Network Open. +- Ghi chú: Tranh cãi. Một phân tích tổng hợp gộp 38 thử nghiệm không thấy loại thuốc này làm giảm tỷ lệ tử vong toàn nguyên（RR 0.98, 95% CI 0.91 đến 1.05）. Xét riêng acid zoledronic thì cũng không đáng kể（RR 0.88, 95% CI 0.68 đến 1.13）. Kết luận của tác giả là loại thuốc này chỉ nên dùng để giảm gãy xương. Độ tin cậy của mấy con số đó trong Cochrane cũng bị đánh giá là thấp đến trung bình. Ở phụ nữ chưa gãy xương, nguy cơ gãy xương thấp, natri alendronat không thấy tác dụng rõ với gãy xương hông（RR 0.76, 95% CI 0.43 đến 1.32）. Tác dụng phụ hiếm như hoại tử xương hàm, gãy xương đùi không điển hình, thử nghiệm vài năm không nhìn thấy hết, nên phải tái khám đúng hạn, không tự ý tăng liều hay ngừng thuốc. Người này hợp tiêm, người kia hợp uống, chọn cái nào do bác sĩ quyết. Làm sao để xét ra loãng xương, xem mục 39 của phần này (xét mật độ xương). Người hưởng lợi chủ yếu là bạn, cùng cha mẹ bạn. + + +### 41. Lên băng ngoài trời thì khoan lỗ đo độ dày trước, dưới 15 xentimét thì đừng lên, đừng tới chỗ băng trên sông và quanh hố đào giấu + +- Chi phí: Cách ít tốn tiền nhất là đi sân trượt chính thức có người quản lý, không phải tốn thêm tiền. Nếu cứ đi băng ngoài trời thì mang theo khoan băng hoặc cuốc băng để khoan lỗ đo độ dày. Mua thêm một đôi gai cứu hộ, tức hai cây đinh nhọn có tay cầm, treo trước ngực. Tất cả cộng lại vài chục đến một hai trăm tệ. Cứ đi được một đoạn thì đo một lần, mất thêm chút thời gian. +- Nói thẳng: Công chức chính thức trong nước nói, băng dày 15 xentimét thì người lớn mới lên trượt được. Mỹ và Canada đặt đường ranh giới ở 10 xentimét, và chỉ tính băng trong suốt mới đóng. Băng bạc màu chỉ chắc bằng một nửa, độ dày phải nhân đôi. Cùng một mảng băng, chỗ này dày một tấc, cách vài bước có khi chỉ một hai tấc, nên phải đi vừa đi vừa đo. +- Lợi ích: Tài liệu phổ biến của Cục Khí tượng Trung Quốc nói, lớp băng dày 15 xentimét thì người lớn lên trượt được. Băng ở sân trượt ngoài trời cũng phải dày 15 xentimét mới mở. Sở Thủy lợi thành phố Bắc Kinh nói, sân trượt tự nhiên chính thức cần băng dày 15 xentimét, sau đó một tuần nhiệt độ giữ trong khoảng -8℃ đến 1℃, không có gió lớn, thì mới mở. Trên băng ngoài trời thường có lỗ do người câu cá đục, đục ra rồi khó đóng lại chắc. Hướng dẫn của Sở Tài nguyên Thiên nhiên bang Minnesota chỉ nói về băng trong suốt mới đóng: dưới 4 inch (khoảng 10 xentimét) thì đừng lên, 4 inch thì đi bộ và câu cá trên băng được. Băng trắng do tuyết tan rồi đóng lại chỉ chắc khoảng một nửa băng trong suốt, độ dày phải tính nhân đôi. Cùng một mặt nước, chỗ này băng dày một feet, cách vài feet chỉ một hai inch. Băng trên dòng chảy thường nguy hiểm, nhất là gần suối, gần cầu và cống. Băng ven bờ cũng có thể mỏng hơn chỗ ở xa. Cơ quan Khí tượng Quốc gia Hoa Kỳ cũng đặt ngưỡng đi bộ, câu cá trên băng, trượt băng ở mức 4 inch. Trung tâm An toàn và Sức khoẻ Nghề nghiệp Canada dẫn lại khuyến nghị của Hội Chữ thập đỏ Canada: vài người trượt băng, đi bộ thì cần 15 xentimét, tụ họp trượt băng đông người thì cần 20 xentimét. Đây đều là hướng dẫn an toàn, không nghiên cứu nào tính được làm theo thì chết được bao nhiêu người (toàn quốc / Hoa Kỳ, Canada) +- Mức bằng chứng: C +- Nguồn: Cục Khí tượng Trung Quốc (2019). 想要冰上健步走，多厚的冰才安全？ ；Sở Thủy lợi thành phố Bắc Kinh (2024). 北京正规河湖冰场还未达开放标准，别冒险滑「野冰」. ；Minnesota Department of Natural Resources. General ice thickness guidelines. ；Minnesota Department of Natural Resources. Ice safety. ；Minnesota Department of Natural Resources. Steps for surviving immersion. ；National Weather Service. Winter Weather Safety: Ice and Frost. ；Canadian Centre for Occupational Health and Safety. Working on or near Ice Covered Water - Basic Information. +- Ghi chú: Hướng dẫn chính thức đều nói không có băng nào an toàn tuyệt đối, đo đủ độ dày chỉ là ngưỡng, không phải bảo đảm. Đầu đông mới đóng băng và đầu xuân bắt đầu tan băng là lúc nguy hiểm nhất. Lỡ rơi xuống, đừng cởi áo, không khí giữ trong áo dày giúp bạn nổi. Quay người về hướng đã đi, phía đó băng chắc nhất. Đặt cánh tay lên mặt băng, cắm gai xuống, đẩo chân ra sau, kéo người lên. Lên rồi đừng đứng dậy, nằm ngửa lăn ra khỏi lỗ, rồi sưởi ấm càng sớm càng tốt, người lạnh quá thì phải vào bệnh viện. Thấy người khác rơi vào hố băng, đừng chạy tới kéo, cách làm xem mục 25 ở phần 13 (thấy người sắp chết đuối thì đừng tự xuống nước): gọi người, báo 110, rồi đưa gậy, thả dây. Người hưởng lợi chủ yếu là bạn và đứa trẻ bạn đưa đi chơi trên băng. diff --git "a/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" "b/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" index e0ed8dcd..f697b414 100644 --- "a/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" +++ "b/vi/book/02-\344\270\215\350\246\201\346\205\242\346\205\242\346\255\273.md" @@ -86,14 +86,15 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: B - Nguồn: Wang K, Matsuyama Y, Kiuchi S và cộng sự (2026). Routine oral health practices and all-cause mortality. Journal of Dentistry. ; Ko MJ, Seo S, So JS và cộng sự (2026). Deteriorated oral health and function as risk factors for physical disability and mortality in community-dwelling older adults: a systematic review and meta-analysis. European Geriatric Medicine. - Ghi chú:Tranh cãi. Các nghiên cứu này chỉ ghi nhận và theo dõi. Người răng xấu thường sức khoẻ chung và tình hình tài chính cũng kém hơn. Răng xấu làm kéo cả cơ thể xuống, hay cơ thể yếu kéo răng theo, thì không rõ. Nhưng việc này tốn rất ít tiền, và bệnh nướu răng cùng tình trạng mất răng vốn đã ảnh hưởng đến ăn uống -### 11. Mỗi ngày đi bộ 7000–8000 bước +### 11. Đi 7000–8000 bước mỗi ngày, hoặc đi bộ nhanh tích luỹ 150–300 phút mỗi tuần -- Chi phí: Không tốn tiền. Mỗi ngày đi bộ 60–90 phút, chia vào đường đi làm và chuyến đi chợ cũng được. Chỗ khó là phải đi bộ mỗi ngày. -- Nói thẳng: Người mỗi ngày đi bộ 5800 bước có xác suất tử vong trong cùng thời gian thấp hơn khoảng bốn phần mười so với người chỉ đi bộ 3500 bước. Lên tới 7800 bước thì giảm khoảng bốn phần rưỡi. Từ đó lên nữa con số giảm thêm một chút rồi đi ngang. Người trên 60 tuổi thì 6000–8000 bước là đủ. -- Lợi ích: Gộp 15 nghiên cứu theo dõi một nhóm người, tất cả chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. 15 nghiên cứu đó gồm 47471 người và 3013 ca tử vong. Chia người thành bốn nhóm theo số bước mỗi ngày, trung vị của nhóm thấp nhất là 3553 bước/ngày. Trung vị của ba nhóm còn lại lần lượt là 5801, 7842, 10901 bước/ngày. Rủi ro tử vong của ba nhóm đó thấp hơn nhóm thấp nhất lần lượt khoảng 40%, 45%, 53% (HR 0.60, 0.55, 0.47). Ở người trên 60 tuổi, lợi ích đi ngang sau 6000–8000 bước mỗi ngày. Ở người dưới 60 tuổi thì đi ngang sau 8000–10000 bước. Một phân tích gộp khác gom nhiều nghiên cứu lại với nhau: tác dụng bắt đầu từ khoảng 3867 bước mỗi ngày. Từ đó, cứ thêm 1000 bước mỗi ngày thì rủi ro tử vong giảm khoảng 15%. +- Chi phí: Không tốn tiền. Tính theo bước: mỗi ngày đi 60–90 phút, chia vào đường đi làm và đi chợ. Tính theo thời gian: mỗi ngày 20–45 phút. Khó ở chỗ phải giữ được lâu dài. +- Nói thẳng: Người mỗi ngày đi khoảng 7800 bước có xác suất chết trong cùng kỳ thấp hơn khoảng bốn phần rưỡi so với người chỉ đi 3500 bước. Tính theo thời gian cũng vậy: đi bộ nhanh 150–300 phút mỗi tuần thấp hơn khoảng ba phần mươi so với không tập thể dục. Tăng đến một mức rồi thì không còn giảm nữa; trên 60 tuổi đi 6000–8000 bước là đủ. Hai cách tính, đạt một trong hai là được. +- Lợi ích: Gộp 15 nghiên cứu theo dõi một nhóm người, chỉ ghi nhận, không chia nhóm. 15 nghiên cứu này có 47471 người, 3013 ca tử vong. Chia người thành bốn nhóm theo số bước mỗi ngày, nhóm thấp nhất có trung vị 3553 bước/ngày. Ba nhóm còn lại có trung vị 5801, 7842, 10901 bước/ngày. Rủi ro tử vong của ba nhóm này thấp hơn nhóm thấp nhất khoảng 40%, 45%, 53% (HR 0.60, 0.55, 0.47). Ở người trên 60 tuổi, sau khi đi 6000–8000 bước thì thêm nữa lợi ích bằng đi. Ở người dưới 60 tuổi, lợi ích bằng sau 8000–10000 bước. Một phân tích gộp nhiều nghiên cứu: bắt đầu có hiệu quả từ khoảng 3867 bước/ngày. Từ đó, mỗi thêm 1000 bước/ngày, rủi ro tử vong giảm khoảng 15%. Tính theo thời gian: tổng hợp nhiều nghiên cứu theo dõi (chỉ ghi nhận, không chia nhóm), người đạt 1–2 lần mức tối thiểu trong hướng dẫn có rủi ro tử vong thấp hơn người không tập khoảng 31% (HR 0.69). Lượng này là 7.5–15 MET·h/tuần, tương đương đi bộ nhanh 150–300 phút/tuần. Người chưa đạt mức tối thiểu cũng thấp hơn khoảng 20% (HR 0.80). Người đạt 3–5 lần mức tối thiểu thấp hơn khoảng 39% (HR 0.61), đây là chỗ giảm nhiều nhất. Tăng thêm nữa thì không thấp hơn nữa, cũng không có hại (trên 10 lần là HR 0.69). Một nghiên cứu tổng hợp khác dùng gia tốc kế đo trực tiếp: tứ phần với hoạt động cường độ trung bình và cao nhất có rủi ro tử vong thấp hơn tứ phần ít hoạt động nhất khoảng 48% (HR 0.52). Gia tốc kế là thiết bị đeo trên người để đo hoạt động thực tế. - Mức bằng chứng: A -- Nguồn: Paluch AE và cộng sự (2022). Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. Lancet Public Health. ; Banach M và cộng sự (2023). The association between daily step count and all-cause and cardiovascular mortality: a meta-analysis. European Journal of Preventive Cardiology. -- Ghi chú: Đoạn từ 4000 bước lên 7000 bước là hiệu quả chi phí tốt nhất, không cần cố đạt mười nghìn bước. Cần nhớ các nghiên cứu này chỉ ghi nhận và theo dõi: nhóm đi bộ ít nhất có lẫn người vốn bị bệnh không đi nổi. Họ đi ít vì sắp chết, không phải đi ít nên chết sớm. Nên khoảng cách mà các nghiên cứu này báo ra thường đã phóng đại. Dù vậy xu hướng càng đi bộ càng tốt thì rõ ràng. Mục này và mục 14 (mỗi tuần 150–300 phút vận động cường độ vừa) nói cùng một việc, chỉ khác cách đếm số, đạt một trong hai là được. +- Nguồn: Paluch AE và cùng tác giả khác (2022). Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. Lancet Public Health. ；Banach M và cùng tác giả khác (2023). The association between daily step count and all-cause and cardiovascular mortality: a meta-analysis. European Journal of Preventive Cardiology. ；Arem H và cùng tác giả khác (2015). Leisure time physical activity and mortality: a detailed pooled analysis of the dose-response relationship. JAMA Internal Medicine. ；Ekelund U và cùng tác giả khác (2019). Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis. BMJ. +- Ghi chú: Số bước và thời gian tập nói cùng một chuyện, chỉ khác cách đếm, đạt một trong hai là được. Đoạn từ 4000 bước lên 7000 bước là đáng tiền nhất, không nhất thiết phải gom cho đủ mười nghìn bước. Cần lưu ý loại nghiên cứu này chỉ theo dõi ghi nhận: trong nhóm đi ít, vận động ít có lẫn người vốn đã bệnh nên không đi nổi. Họ sắp không kham nổi mới vận động ít, không phải vì vận động ít mà chết sớm. Vì vậy khoảng cách tính ra thường lớn hơn thực. Với mấy nghiên cứu dùng thiết bị đo hoạt động thực tế, thời gian theo dõi ngắn, người tham gia thiên về già, lợi ích thật không lớn bằng「rủi ro giảm còn 0.52 lần」. Nhưng xu hướng càng vận động càng tốt thì rõ ràng. + ### 12. Có cao huyết áp, mỡ máu cao thì uống thuốc đúng y lệnh, đừng tự ý ngừng - Chi phí: Thuốc hạ huyết áp và statin mua theo cơ chế tập trung dược phẩm mỗi tháng vài tệ đến vài chục tệ. Mỗi ngày uống một lần, không tốn nhiều thời gian. Chỗ khó là phải uống tiếp, không được nghĩ khỏe rồi là bỏ. @@ -102,24 +103,15 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Ettehad D và cộng sự (2016). Blood pressure lowering for prevention of cardiovascular disease and death: a systematic review and meta-analysis. Lancet. ; Cholesterol Treatment Trialists' (CTT) Collaboration (2010). Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170 000 participants in 26 randomised trials. Lancet. ; Chowdhury R và cộng sự (2013). Adherence to cardiovascular therapy: a meta-analysis of prevalence and clinical consequences. European Heart Journal. - Ghi chú: Chỉ làm theo khi bác sĩ đã đánh giá rằng bạn cần thuốc, người khoẻ mạnh thì không. Con số của nhóm "kiên trì uống thuốc" đến từ các nghiên cứu chỉ ghi nhận và theo dõi, người uống được đều đặn vốn đã kỷ luật hơn, nên lợi ích bị nói quá. Con số từ hai nhóm thử nghiệm chia nhóm ngẫu nhiên ở trên đáng tin hơn. Phần này không tách riêng mục nào về đường huyết: bằng chứng cho thấy hạ đường huyết thấp hơn giúp giảm tỷ lệ tử vong toàn nguyên không nhất quán bằng huyết áp và mỡ máu. -### 13. Mỗi đêm ngủ khoảng 7 tiếng, giữ giờ sinh hoạt ổn định +### 13. Ngủ khoảng 7 giờ mỗi đêm, giờ sinh hoạt giữ nguyên -- Chi phí: Không tốn tiền. Phải dành thời gian ra để ngủ, phần lớn người lấy từ lúc lướt điện thoại. Chỗ khó là cố định giờ lên giường và giờ thức dậy. -- Nói thẳng: Người lâu dài ngủ dưới 7 tiếng có xác suất tử vong trong cùng thời gian cao hơn khoảng một phần mười. Giờ sinh hoạt càng loạn thì rủi ro càng cao. Người ngủ đều đặn nhất có xác suất tử vong thấp hơn người loạn nhất từ hai phần mười đến gần một nửa. Nói cách khác, giờ sinh hoạt đều đặn hay không nói lên rủi ro tử vong rõ hơn là ngủ bao nhiêu tiếng. -- Lợi ích: Gộp 16 nghiên cứu theo dõi, tất cả chỉ ghi nhận và theo dõi, không chia nhóm đối chứng, gồm 1.380.000 người và 113.000 ca tử vong. Người ngủ ít có rủi ro tử vong cao hơn khoảng 12% (RR 1.12). Người ngủ nhiều cao hơn khoảng 30% (RR 1.30). Một bảng tổng hợp khác chia người theo thời lượng ngủ, nhóm 7 tiếng có rủi ro thấp nhất. Dưới 7 tiếng, mỗi khoảng thiếu 1 tiếng thì rủi ro cao hơn khoảng 6% (RR 1.06). Trên 7 tiếng, mỗi giờ thêm vào rủi ro cao hơn khoảng 13% (RR 1.13). Ngân hàng sinh học Anh có dữ liệu vòng đeo tay của khoảng 61.000 người, cũng chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. Chia người thành năm nhóm theo mức độ đều đặn của giờ sinh hoạt. Bốn nhóm đều đặn hơn có rủi ro tử vong thấp hơn nhóm loạn nhất 20%–48%. Mức độ đều đặn của giờ sinh hoạt dự báo tử vong tốt hơn thời lượng ngủ. Một nghiên cứu khác của Ngân hàng sinh học Anh đo giờ ngủ bằng vòng đeo tay ở 51.562 người, trong đó 3.853 người mắc bệnh tim mạch mới trong thời gian nghiên cứu. Chênh lệch giờ sinh học xã hội là khoảng cách giữa điểm giữa của giờ ngủ ngày thường và điểm giữa của giờ ngủ cuối tuần. Người có chênh lệch giờ sinh học xã hội từ 2 tiếng trở lên thì nguy cơ tim mạch cao hơn khoảng 30% (HR 1.30, 95% CI 1.11–1.54). Mối liên hệ này vẫn còn ở người có thời lượng ngủ bình thường. -- Mức bằng chứng: A -- Nguồn: Cappuccio FP và cộng sự (2010). Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. ; Yin J và cộng sự (2017). Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. JAHA. ; Windred DP và cộng sự (2024). Sleep regularity is a stronger predictor of mortality risk than sleep duration: A prospective cohort study. Sleep. ; Kumar N, Krishnamurthy S (2026). Social jet lag is associated with incident cardiovascular disease independent of sleep duration and cardiac genetic risk. Journal of Internal Medicine. -- Ghi chú: Cần sửa là ngủ quá ít và giờ sinh hoạt loạn, không nhất thiết phải rút ngắn giấc ngủ có chủ ý. Người ngủ nhiều có rủi ro cao hơn, nhiều khả năng là nguyên nhân ngược lại: trầm cảm, bệnh mãn tính và ngưng thở khi ngủ khiến người ta ngủ nhiều hơn. Còn về việc giữ giờ sinh hoạt đều đặn thì hiện có hai nghiên cứu theo dõi hỗ trợ, cả hai chỉ ghi nhận và theo dõi, không chia nhóm đối chứng, đứng riêng thì là bằng chứng mức B. Sau một đêm thức khuya thì bù lại thế nào, xem mục 39 của phần này (tối hôm sau ngủ bù luôn). -### 14. Mỗi tuần cộng dồn 150–300 phút vận động cường độ vừa, đi bộ nhanh là đủ - -- Chi phí: Không tốn tiền. Mỗi ngày 20–45 phút. Chỗ khó là duy trì lâu dài. -- Nói thẳng: Người mỗi tuần đi bộ nhanh 150–300 phút có xác suất tử vong trong cùng thời gian thấp hơn khoảng ba phần mười so với người không tập. Không đạt mục tiêu thì vẫn thấp hơn hai phần mười. Tập gấp ba đến năm lần mục tiêu thì con số đi xuống đáy, thấp hơn khoảng bốn phần mười. Tăng thêm thì không thấp hơn nữa, nhưng cũng không có hại. -- Lợi ích: Gộp nhiều nghiên cứu theo dõi lại với nhau (tất cả chỉ ghi nhận và theo dõi, không chia nhóm đối chứng): người có mức vận động bằng 1 đến 2 lần mức tối thiểu trong khuyến cáo có rủi ro tử vong thấp hơn người không tập khoảng 31% (HR 0.69). Mức đó là 7.5–15 MET·h mỗi tuần, tương đương khoảng 150–300 phút đi bộ nhanh mỗi tuần. Người không đạt mức tối thiểu vẫn thấp hơn khoảng 20% (HR 0.80). Người đạt từ 3 đến 5 lần mức tối thiểu thấp hơn khoảng 39% (HR 0.61), và đó là chỗ giảm nhiều nhất. Tăng thêm thì không thấp hơn nữa, nhưng cũng không có hại (hơn 10 lần là HR 0.69). Một nghiên cứu gộp khác dùng máy đo gia tốc: nhóm một phần tư hoạt động mạnh nhất theo hoạt động cường độ vừa đến cao có rủi ro tử vong thấp hơn nhóm một phần tư ít hoạt động nhất khoảng 48% (HR 0.52). Máy đo gia tốc là thiết bị đeo lên người để đo trực tiếp mức hoạt động. +- Chi phí: Không tốn tiền. Phải bỏ thời gian ra để ngủ, phần lớn người đổi thời gian lướt điện thoại lấy thành giấc ngủ. Khó ở chỗ phải cố định giờ lên giường và giờ thức dậy. +- Nói thẳng: Người ngủ dưới 7 giờ lâu ngày có xác suất chết trong cùng kỳ cao hơn khoảng một phần mười. Giờ sinh hoạt càng loạn, rủi ro càng cao. Người ngủ đều đặn nhất có tỷ lệ thấp hơn người loạn nhất từ hai phần mười đến gần năm phần mười. Nói cách khác, giờ sinh hoạt có quy định hay không cho thấy rủi ro tử vong rõ hơn cả thời gian ngủ bao lâu. +- Lợi ích: Gộp 16 nghiên cứu theo dõi, chỉ ghi nhận, không chia nhóm, 138 vạn người, 11.3 vạn ca tử vong. Người ngủ ít có rủi ro tử vong cao hơn khoảng 12% (RR 1.12). Người ngủ nhiều cao hơn khoảng 30% (RR 1.30). Một tổng hợp khác chia theo thời lượng ngủ: nhóm 7 giờ có rủi ro thấp nhất. Dưới 7 giờ, mỗi giờ ít đi (tức mỗi 1 giờ) thì cao hơn khoảng 6% (RR 1.06). Trên 7 giờ, mỗi giờ nhiều thêm (tức mỗi 1 giờ) thì cao hơn khoảng 13% (RR 1.13). Ngân hàng Sinh học Anh có khoảng 6.1 vạn người có dữ liệu đồng hồ đeo tay, chỉ ghi nhận, không chia nhóm. Chia người thành năm bậc theo mức độ đều đặn của giờ sinh hoạt. Bốn bậc đều đặn hơn có rủi ro tử vong thấp hơn bậc loạn nhất 20%–48%. Giờ sinh hoạt quy định hay không dự đoán tử vong tốt hơn cả thời gian ngủ. Một nghiên cứu khác của Ngân hàng Sinh học Anh đo giờ ngủ của 51.562 người bằng đồng hồ đeo tay, trong đó 3.853 người mới mắc bệnh tim mạch. Chênh lệch xã hội là độ lệch giữa điểm giữa của giấc ngủ ngày thường và cuối tuần. Người có chênh lệch xã hội từ 2 giờ trở lên có rủi ro bệnh tim mạch cao hơn khoảng 30% (HR 1.30, 95% CI 1.11–1.54). Mối liên hệ này vẫn còn ở những người có thời lượng ngủ bình thường. - Mức bằng chứng: A -- Nguồn: Arem H và cộng sự (2015). Leisure time physical activity and mortality: a detailed pooled analysis of the dose-response relationship. JAMA Internal Medicine. ; Ekelund U và cộng sự (2019). Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis. BMJ. -- Ghi chú: Mục này và mục 11 (mỗi ngày đi bộ 7000–8000 bước) chỉ cần chọn một. Cần nhớ các nghiên cứu này chỉ ghi nhận và theo dõi. Mấy nghiên cứu đo hoạt động bằng thiết bị thì thời gian theo dõi ngắn, người tham gia phần lớn lớn tuổi, nên trong con số có lẫn thành phần "bị bệnh rồi mới ngừng vận động". Lợi ích thật không lớn bằng tỷ số rủi ro 0.52. - -### 15. Mỗi tuần chơi môn bóng vợt ba lần, mỗi lần 45 phút +- Nguồn: Cappuccio FP và cùng tác giả khác (2010). Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. ；Yin J và cùng tác giả khác (2017). Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. JAHA. ；Windred DP và cùng tác giả khác (2024). Sleep regularity is a stronger predictor of mortality risk than sleep duration: A prospective cohort study. Sleep. ；Kumar N, Krishnamurthy S (2026). Social jet lag is associated with incident cardiovascular disease independent of sleep duration and cardiac genetic risk. Journal of Internal Medicine. +- Ghi chú: Cần sửa là ngủ quá ít và giờ sinh hoạt loạn, không nhất thiết phải cố rút ngắn thời gian ngủ. Người ngủ nhiều có rủi ro cao, nhiều khả năng là nguyên nhân quay ngược: trầm cảm, bệnh mạn tính, ngừng thở khi ngủ khiến người ta ngủ nhiều hơn. Ngoài ra, điểm giờ sinh hoạt đều đặn hiện có hai nghiên cứu theo dõi ủng hộ, đều chỉ ghi nhận không chia nhóm, đứng riêng thì thuộc bằng chứng B. Về chuyện thiếu ngủ rồi ngủ bù, xem mục 38 của phần này (sau khi thức khuya thì đêm hôm sau ngủ bù). +### 14. Mỗi tuần chơi môn bóng vợt ba lần, mỗi lần 45 phút - Chi phí: Phí thuê sân mỗi lần vài chục tệ. Khoảng 2 tiếng mỗi tuần. Chỗ khó là phải tìm đủ người và sân, rồi tuần nào cũng phải đi. - Nói thẳng: Người thường xuyên chơi quần vợt, cầu lông, bóng bàn có xác suất tử vong trong cùng thời gian bằng khoảng một nửa người không chơi, và tử vong tim mạch thấp hơn khoảng 56%. Người bơi thấp hơn khoảng 28%. Chạy bộ và bóng đá không thấy khác biệt trong bộ dữ liệu này. @@ -127,7 +119,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Oja P, Kelly P, Pedisic Z và cộng sự (2017). Associations of specific types of sports and exercise with all-cause and cardiovascular-disease mortality: a cohort study of 80 306 British adults. British Journal of Sports Medicine, 51(10), 812-817. - Ghi chú:Tranh cãi. Các nghiên cứu này chỉ ghi nhận và theo dõi. Người chơi thể thao vốn đã khỏe hơn, giao tiếp cũng nhiều hơn. Chỗ chạy bộ không thấy khác biệt cũng cho thấy yếu tố quyết định có khi là "người loại nào chọn môn nào". Đừng vì mục này mà nghĩ chạy bộ vô ích, phần này còn có mục khác tính theo tổng lượng vận động. -### 16. Cộng dồn các đoạn vận động rải rác như leo cầu thang, đi nhanh lên năm phút mỗi ngày +### 15. Cộng dồn các đoạn vận động rải rác như leo cầu thang, đi nhanh lên năm phút mỗi ngày - Chi phí: Không tốn tiền, cũng không cần dành thêm giờ tập. Chỗ khó là nhớ leo cầu thang, đi nhanh hơn thay vì chọn cách tiện. - Nói thẳng: Người hoàn toàn không tập thể thao, chỉ cần mỗi ngày có ba đoạn vận động mạnh một hoặc hai phút thì xác suất tử vong trong cùng thời gian vẫn thấp hơn khoảng bốn phần mười so với người không có đoạn nào. Tử vong tim mạch thấp hơn gần một nửa. Vận động mạnh ở đây là loại như leo cầu thang, đi nhanh để kịp giờ. @@ -135,7 +127,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Stamatakis E, Ahmadi MN, Gill JMR và cộng sự (2022). Association of wearable device-measured vigorous intermittent lifestyle physical activity with mortality. Nature Medicine, 28, 2521-2529. - Ghi chú:Tranh cãi. Nghiên cứu này đo hoạt động bằng thiết bị đeo, đáng tin hơn là bảo người ta điền bảng hỏi rồi nhớ lại. Nhưng nó vẫn chỉ ghi nhận và theo dõi, không làm thử nghiệm chia nhóm đối chứng, và chỉ theo dõi trong 6.9 năm. Đối tượng là người hoàn toàn không tập, nên mục này không áp dụng cho người đã tập đều -### 17. Mỗi tuần tập sức mạnh 30–60 phút +### 16. Mỗi tuần tập sức mạnh 30–60 phút - Chi phí: Squat không tạp, chống đẩy không tốn tiền. Tập mỗi tuần 1–2 lần, mỗi lần 20–30 phút. Chỗ khó là sắp xếp được vào lịch trong tuần. - Nói thẳng: Người tập sức mạnh 30–60 phút mỗi tuần có xác suất tử vong trong cùng thời gian thấp hơn một đến hai phần mười so với người không tập. Tập nhiều hơn không khá hơn. Tập sức mạnh cùng vận động aerobic thì hiệu quả nhất. @@ -143,7 +135,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Momma H và cộng sự (2022). Muscle-strengthening activities are associated with lower risk and mortality in major non-communicable diseases: a systematic review and meta-analysis of cohort studies. British Journal of Sports Medicine. - Ghi chú: Các nghiên cứu này chỉ ghi nhận và theo dõi, mức vận động do chính người tham gia tự báo. Bằng chứng cho thấy tập quá tay thì tệ hơn rất yếu, không có lý do gì phải dè chừng vì điều đó. Tập sức mạnh còn giúp người lớn tuổi tránh ngã và giữ cơ, những điều đó nằm ở phần 1. -### 18. Đừng ngồi liên tục quá lâu, thi thoảng đứng dậy vận động +### 17. Đừng ngồi liên tục quá lâu, thi thoảng đứng dậy vận động - Chi phí: Không tốn tiền, đứng dậy mấy lần cũng không tốn nhiều thời gian. Chỗ khó là ngồi xuống là quên mất thời gian, nên phải đặt nhắc. - Nói thẳng: Người ngồi lâu nhất có xác suất tử vong trong cùng thời gian bằng 2.6 lần người ngồi ít nhất. Ngồi liền một đoạn dài còn mang thêm một khoản rủi ro riêng. Nhưng chỉ cần mỗi ngày có 60–75 phút hoạt động cường độ vừa là phần rủi ro cộng thêm do ngồi gần như được bù lại. @@ -151,7 +143,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Diaz KM và cộng sự (2017). Patterns of Sedentary Behavior and Mortality in U.S. Middle-Aged and Older Adults: A National Cohort Study. Annals of Internal Medicine. ; Ekelund U và cộng sự (2016). Does physical activity attenuate, or even eliminate, the detrimental association of sitting time with mortality? A harmonised meta-analysis of data from more than 1 million men and women. Lancet. - Ghi chú: Kết luận đáng tin là lúc nào vận động nhiều quan trọng hơn lúc nào ngồi lâu, và khi lượng vận động đủ thì rủi ro do ngồi gần như biến mất. Cần nhớ các nghiên cứu này chỉ ghi nhận và theo dõi, trong nhóm ngồi nhiều nhất có rất nhiều người vốn đã bị bệnh, nên con số 2.63 lần rõ ràng có chứa thành phần "bị bệnh rồi mới ngồi nhiều". Trong cùng bảng gộp đó, rủi ro xem hơn 3 tiếng ti vi mỗi ngày tồn tại ở mọi mức vận động, nhóm hoạt động nhiều nhất chỉ đẩy mốc đó ra tới 5 tiếng. -### 19. Ăn ít thịt chế biến (thịt chua, thịt xông khói, xúc xích, thịt đóng hộp) +### 18. Ăn ít thịt chế biến (thịt chua, thịt xông khói, xúc xích, thịt đóng hộp) - Chi phí: Không tốn tiền, có khi còn tiết kiệm, cũng không mất thời gian. Chỗ khó là bớt chính miếng mình thích. - Nói thẳng: Người ăn nhiều thịt chua, thịt xông khói và xúc xích nhất có xác suất tử vong trong cùng thời gian cao hơn khoảng hai phần mười so với người ăn ít nhất. Mỗi phần thịt chế biến ăn thêm mỗi ngày cũng làm con số đó tăng thêm hai phần mười. @@ -160,33 +152,32 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Nguồn: Larsson SC, Orsini N (2014). Red meat and processed meat consumption and all-cause mortality: a meta-analysis. American Journal of Epidemiology. ; Schwingshackl L và cộng sự (2017). Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. American Journal of Clinical Nutrition. ; Johnston BC và cộng sự (2019). Unprocessed Red Meat and Processed Meat Consumption: Dietary Guideline Recommendations From the NutriRECS Consortium. Annals of Internal Medicine. (hướng dẫn NutriRECS, phe bên kia) - Ghi chú:Tranh cãi. Hướng dẫn NutriRECS dùng hệ thống chấm bằng chứng thông dụng (GRADE) và đánh giá các nghiên cứu này là "không đáng tin lắm". Nó chỉ đưa ra một khuyến nghị rất yếu: cứ ăn như hiện tại. Hai phe tranh nhau là bằng chứng có đủ chắc hay không, chứ không phải hướng. Không nghiên cứu nào nói thịt chế biến có lợi. Thịt đỏ chưa chế biến ảnh hưởng nhỏ và không thấy khác biệt có ý nghĩa thống kê. Cái cần kiềm là thịt chế biến. -### 20. Uống ít rượu hoặc không uống +### 19. Uống ít hoặc không uống rượu -- Chi phí: Không tốn tiền, còn tiết kiệm tiền, cũng không mất thời gian. Chỗ khó là lúc ở bàn tiệc phải nhận rượu. -- Nói thẳng: Quá 100 g rượu thuần mỗi tuần thì uống càng nhiều chết càng sớm. 100 g rượu thuần khoảng bằng 2.5 L bia. Người 40 tuổi mỗi tuần uống 100–200 g sống ít khoảng nửa năm; người uống 200–350 g sống ít một đến hai năm; người vượt 350 g sống ít bốn đến năm năm. Xét tổng thể sức khoẻ thì lượng đáng dùng nhất là không uống. -- Lợi ích: Gộp 83 nghiên cứu theo dõi lại với nhau, chỉ ghi nhận và theo dõi, không chia nhóm đối chứng, gồm 600.000 người có uống rượu. Rủi ro tử vong thấp nhất ở mức không quá 100 g rượu thuần mỗi tuần. Lượng đó khoảng 2.5 L bia 5% hoặc 300 mL rượu trắng 40 độ. Người 40 tuổi uống 100–200 g mỗi tuần có tuổi thọ ngắn hơn khoảng 6 tháng. Người uống 200–350 g mỗi tuần mất 1–2 năm. Người vượt 350 g mất 4–5 năm. Cả ba mốc đó bản gốc báo cáo theo năm sống. Nghiên cứu Giao tải bệnh toàn cầu GBD 2016 gộp mọi loại thiệt hại sức khoẻ và đặt lượng uống ít rủi ro nhất là 0 ly mỗi tuần. Một bảng gộp khác đã hiệu chỉnh sai lệch trong thiết kế nghiên cứu. Người uống 1.3–24 g mỗi ngày có rủi ro tử vong thấp hơn khoảng 7% so với người không bao giờ uống (RR 0.93), nhưng mục này không có ý nghĩa thống kê. Người uống 45–64 g mỗi ngày cao hơn khoảng 19% (RR 1.19). Người uống trên 65 g mỗi ngày cao hơn khoảng 35% (RR 1.35). +- Chi phí: Không tốn tiền, còn tiết kiệm tiền nữa, và không tốn thời gian. Khó ở chỗ lúng túng lúc phải từ chối chén rượu trong các dịp giao tiếp. +- Nói thẳng: Sau khi uống quá 100 g rượu nguyên chất mỗi tuần, uống càng nhiều thì chết càng sớm. 100 g rượu nguyên chất xấp xỉ bằng 2.5 L bia. Người 40 tuổi uống 100–200 g mỗi tuần sống ít đi khoảng nửa năm; uống 200–350 g sống ít đi một đến hai năm; vượt 350 g sống ít đi bốn đến năm năm. Tính cả sức khoẻ, lượng đáng giá nhất là không uống. +- Lợi ích: Gộp 83 nghiên cứu theo dõi về sau, chỉ ghi nhận, không chia nhóm, 600.000 người có uống rượu. Điểm thấp nhất của rủi ro tử vong nằm ở mức không quá 100 g rượu nguyên chất mỗi tuần. Lượng này xấp xỉ 2.5 L bia 5%, hoặc 300 mL rượu bạc 40 độ. Người 40 tuổi uống 100–200 g mỗi tuần mất khoảng 6 tháng tuổi thọ. Uống 200–350 g mỗi tuần mất 1–2 năm. Vượt 350 g mất 4–5 năm. Cả ba mức trên, nguyên văn báo bằng năm tuổi thọ. Nghiên cứu Giao tải bệnh toàn cầu GBD 2016 gộp mọi tổn hại sức khoẻ lại, lượng uống có rủi ro thấp nhất là 0 ly/tuần. Một tổng hợp khác đã hiệu chỉnh sai lệch về thiết kế nghiên cứu. Người uống 1.3–24 g mỗi ngày so với người cả đời không uống có rủi ro tử vong thấp hơn khoảng 7% (RR 0.93). Nhưng nghiên cứu này không thấy khác biệt có ý nghĩa thống kê. Người uống 45–64 g mỗi ngày cao hơn khoảng 19% (RR 1.19). Trên 65 g mỗi ngày cao hơn khoảng 35% (RR 1.35). - Mức bằng chứng: A -- Nguồn: Wood AM và cộng sự (2018). Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. ; GBD 2016 Alcohol Collaborators (2018). Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. ; Zhao J và cộng sự (2023). Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Network Open. ; Di Castelnuovo A và cộng sự (2006). Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Archives of Internal Medicine. (phe bên kia) -- Ghi chú:Tranh cãi. Phe cho rằng uống ít có lợi dựa vào một bảng gộp 34 nghiên cứu theo dõi dài hạn. Nó báo rằng người uống ít có tỷ lệ tử vong toàn nguyên thấp hơn tối đa 17%–18%. Lượng mà nó đưa ra là nam không quá 4 ly mỗi ngày, nữ không quá 2 ly. Trong mức đó, người uống nhiều hơn lại có tỷ lệ tử vong thấp hơn. Phe bên kia cho rằng nhóm "không uống" đã lẫn người bỏ rượu vì bệnh và người vốn sức khoẻ đã kém. Có một bảng gộp chỉ vào đúng những nhiễu đó, sau khi trừ đi thì tác dụng bảo vệ biến mất. Nghiên cứu Giao tải bệnh toàn cầu cũng đặt 0 là an toàn nhất. Cách đọc an toàn là: uống ít không rõ là có rút ngắn tuổi thọ hay không, uống nhiều thì chắc chắn có hại. "Bắt đầu uống rượu vì sức khoẻ" không có cơ sở. Muốn uống ít hơn thì xem mục 22 của phần này (đếm số ly trước rồi hỏi bác sĩ). Người uống rượu mỗi ngày không nên tự bỏ dứt khoát, xem mục 21 của phần này (đừng tự bỏ dứt khoát) -### 21. Người uống rượu mỗi ngày, ngừng lại là run tay và hồi hộp, đừng tự cai bằng sức mình +- Nguồn: Wood AM và cùng tác giả khác (2018). Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. ；GBD 2016 Alcohol Collaborators (2018). Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. ；Zhao J và cùng tác giả khác (2023). Association Between Daily Alcohol Intake and Risk of All-Cause Mortality: A Systematic Review and Meta-analyses. JAMA Network Open. ；Di Castelnuovo A và cùng tác giả khác (2006). Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Archives of Internal Medicine. （phía bên tranh luận） +- Ghi chú: Tranh cãi. Phe cho rằng「uống ít có lợi」dựa trên một tổng hợp của 34 nghiên cứu theo dõi dài hạn. Nó báo: người uống ít có tỷ lệ tử vong toàn nguyên nhân thấp nhất khoảng 17%–18%. Lượng nó đưa ra là nam không quá 4 ly mỗi ngày, nữ không quá 2 ly. Trong mức đó, người uống nhiều hơn lại có tỷ lệ tử vong thấp hơn. Phe bên kia cho rằng nhóm「không uống」lẫn người bỏ rượu vì bệnh và người vốn sức khoẻ đã kém. Có một tổng hợp chuyên trừ các nhiễu này ra, làm xong thì tác dụng bảo vệ biến mất. Nghiên cứu Giao tải bệnh toàn cầu cũng cho rằng 0 là an toàn nhất. Cách đọc chắc chắn là: uống ít không làm tổn hại tuổi thọ rõ rệt, uống nhiều thì chắc chắn có hại.「Bắt đầu uống rượu vì sức khoẻ」không có cơ sở. Muốn uống ít thì làm sao, xem mục 21 của phần này (đếm số ly trước rồi nói chuyện với bác sĩ). Người uống mỗi ngày không tự ngừng cứng, xem mục 20 của phần này (đừng tự cai cứng) +### 20. Người uống rượu mỗi ngày, ngừng là run tay hồi hộp, đừng tự cai cứng -- Chi phí: Khám khoa tâm thần hoặc khoa nghiện rượu một lần, mấy tệ đến mấy chục tệ. Trường hợp nặng thì phải nằm viện mấy ngày. -- Nói thẳng: Người uống rượu nhiều mỗi ngày mà ngừng đột ngột sẽ có phản ứng cai rượu. Nhẹ thì run tay, ra mồ hôi, hồi hộp, mất ngủ. Nặng thì co giật hoặc lên cơn sảng run, đó là cấp cứu có thể gây chết. Na Uy theo dõi hơn 30.000 người, trong nhóm từng bị sảng run thì mỗi năm về sau có 8% chết. Muốn cai thì đi bệnh viện trước, nói rõ mỗi ngày uống bao nhiêu. -- Lợi ích: Một tổng quan lâm sàng năm 2026 ghi rõ: người uống rượu nhiều lâu ngày, sau khi ngừng đột ngột hoặc giảm mạnh, sẽ có hội chứng cai rượu, với các biểu hiện gồm run, kích thích thần kinh thực vật, lo âu, mất ngủ, rối loạn cảm giác, cơn động kinh và sảng run, trong chẩn đoán nội khoa thông thường là「hay gặp và có thể đe dọa tính mạng」. Một nghiên cứu mẫu theo thời gian dựa trên sổ đăng ký toàn quốc của Na Uy lấy 36287 người được chẩn đoán nghiện rượu, trạng thái cai rượu hoặc sảng run trong giai đoạn 2009 đến 2015: nhóm sảng run có tỷ lệ tử vong hằng năm 8.0%, nhóm trạng thái cai rượu 5.0%, nhóm nghiện rượu 3.6%. Tỷ số tử vong chuẩn hoá của nhóm sảng run là 9.8 (95% CI 8.9–10.7, đây là khoảng tin cậy), nghĩa là rủi ro tử vong cao khoảng 9.8 lần người cùng tuổi nói chung. +- Chi phí: Một lượt khám ở khoa Tâm thần hoặc khoa Nghiện nghiệp rượu, mấy tệ đến mấy chục tệ. Nặng thì phải nằm viện vài ngày. +- Nói thẳng: Người uống rượu nhiều mỗi ngày ngừng đột ngột sẽ có phản ứng cai. Nhẹ thì run tay, ra mồ hôi, hồi hộp, mất ngủ; nặng thì co giật hoặc mê sảng run rút (loạn thần do cai rượu), đó là cấp cứu có thể chết người. Na Uy theo dõi hơn ba mươi nghìn người, trong nhóm từng bị mê sảng run rút, mỗi năm về sau có 8% chết. Muốn cai thì đi bệnh viện trước, nói rõ mỗi ngày uống bao nhiêu. +- Lợi ích: Một tổng quan lâm sàng năm 2026 ghi rõ: người uống rượu nhiều lâu ngày, sau khi ngừng đột ngột hoặc giảm mạnh sẽ có hội chứng cai, biểu hiện gồm run, kích thích thần kinh thực vật, lo âu, mất ngủ, rối loạn cảm giác, cơn co giật và mê sảng run rút, trong chẩn đoán nội khoa thường phổ biến「và có thể đe dọa tính mạng」. Cơ sở dữ liệu quốc gia của Na Uy gồm 36287 người được chẩn đoán rượu nghiện, trạng thái cai rượu hoặc mê sảng run rút trong giai đoạn 2009 đến 2015: tỷ lệ tử vong hằng năm của nhóm mê sảng run rút là 8.0%, nhóm trạng thái cai rượu là 5.0%, nhóm rượu nghiện là 3.6%. Tỷ số tử vong chuẩn hóa của nhóm mê sảng run rút là 9.8 (95% CI 8.9–10.7, đây là khoảng tin cậy), nghĩa là rủi ro tử vong khoảng 9.8 lần người bình thường cùng tuổi. - Mức bằng chứng: B - Nguồn: Caputo F, Lungaro L, Costanzini A, De Giorgio R, Addolorato G (2026). Alcohol withdrawal syndrome in hospitalized patients: a practical review. European Journal of Internal Medicine, 107103. ；Bramness JG, Heiberg IH, Høye A, Rossow I (2023). Mortality and alcohol-related morbidity in patients with delirium tremens, alcohol withdrawal state or alcohol dependence in Norway: A register-based prospective cohort study. Addiction, 118(12), 2352–2359. -- Ghi chú: Tự đánh giá thế nào: gần như ngày nào cũng uống, một ngày không uống thì run tay, ra mồ hôi, hồi hộp, mất ngủ, sáng dậy phải uống một chút mới dễ chịu, thì chỉ cần trúng một dấu hiệu trong số đó là đừng tự ngừng đột ngột. Bệnh viện thường dùng thuốc nhóm benzodiazepin để người bệnh chịu qua mấy ngày đó, đồng thời bổ sung thiamine (vitamin B1). Để ở mức B là vì nghiên cứu mẫu của Na Uy nói về「người từng bị sảng run sau đó thế nào」, không phải「tự cai cứng thì ra sao」, giữa hai câu hỏi đó còn là suy luận. Rượu vẫn phải cai, chỉ là cai ở bệnh viện. Mỗi tuần uống bao nhiêu thì tính là nhiều, xem mục 20 của phần này (uống ít hoặc không uống). Muốn uống ít hơn thì làm sao, xem mục 22 của phần này (đếm số ly trước rồi hỏi bác sĩ) - -### 22. Muốn uống ít rượu hơn, trước tiên tính xem một tuần uống bao nhiêu, rồi nói chuyện vài phút với bác sĩ +- Ghi chú: Tự đánh giá thế nào: uống gần như ngày nào, một ngày không uống là run tay, ra mồ hôi, hồi hộp, mất ngủ, buổi sáng phải uống chút mới thấy dễ chịu, chỉ một dấu hiệu thôi cũng đừng tự ngừng cứng. Cách làm thường quy của bệnh viện là dùng thuốc nhóm benzodiazepin để chịu qua mấy ngày đó, đồng thời bổ sung thiamin (vitamin B1). Xếp B là vì nghiên cứu mạch Na Uy nói「người từng bị mê sảng run rút sau đó thế nào」, không phải「tự cai cứng thì ra sao」, giữa hai câu hỏi đó còn là suy luận. Rượu vẫn nên cai, chỉ là đổi thành đi bệnh viện cai. Một tuần uống bao nhiêu là nhiều, xem mục 19 của phần này (uống ít hoặc không uống), muốn uống ít thì làm sao, xem mục 21 của phần này (đếm số ly trước rồi nói chuyện với bác sĩ). +### 21. Muốn uống ít rượu hơn, trước tiên tính xem một tuần uống bao nhiêu, rồi nói chuyện vài phút với bác sĩ -- Chi phí: Không tốn tiền. Mỗi ngày ghi số ly vào sổ là xong. Hỏi thêm một câu ở phòng khám cũng không tốn thêm tiền. -- Nói thẳng: Bác sĩ dành mấy phút để xem lượng rượu bạn uống, nói rõ rủi ro, giúp bạn đặt một mục tiêu. Sau một năm, trung bình mỗi tuần uống ít đi 20 g rượu thuần, tương đương khoảng nửa lít bia. Nói chuyện lâu hơn không có thêm lợi ích. Người đã không ngừng được thì ở nước ngoài có hai loại thuốc đã được chứng minh có tác dụng, cứ điều trị khoảng 12 người thì có thêm 1 người không quay lại kiểu uống một lần là nhiều. -- Lợi ích: Gộp 34 thử nghiệm chia nhóm ngẫu nhiên, 15197 người, lại với nhau: người được can thiệp ngắn sau một năm uống ít đi mỗi tuần 20 g rượu thuần so với người chỉ được can thiệp tối thiểu hoặc không được can thiệp (95% CI ít hơn 12 đến 28 g, đây là khoảng tin cậy; bằng chứng chất lượng trung bình). Lượng rượu nền của người tham gia trong các nghiên cứu này trung bình 244 g mỗi tuần. Can thiệp ngắn nghĩa là lời khuyên hoặc tư vấn lối sống không quá 5 lần, tổng thời gian dưới 60 phút; nói chuyện lâu hơn không cho thêm lợi ích nào. Về thuốc: gộp 122 thử nghiệm chia nhóm ngẫu nhiên, 22803 người, acamprosat mỗi 12 người được điều trị (95% CI 8 đến 26) thì có thêm 1 người không quay lại uống bất kỳ lượng nào; naltrexon uống mỗi ngày 50 mg, mỗi 12 người được điều trị (8 đến 26) thì có thêm 1 người không quay lại mức uống nặng. +- Chi phí: Không tốn tiền. Cứ ghi số ly vào một cuốn sổ mỗi ngày là xong. Hỏi thêm một câu ở phòng khám cũng không tốn thêm tiền. +- Nói thẳng: Bác sĩ dành mấy phút để xem lượng rượu bạn uống, nói rõ rủi ro, giúp bạn đặt một mục tiêu. Sau một năm, trung bình mỗi tuần uống ít đi 20 g rượu nguyên chất, khoảng nửa lít bia. Nói chuyện lâu hơn thì không có thêm lợi ích. Người đã không ngừng được thì ở nước ngoài có hai loại thuốc đã được chứng minh có tác dụng, cứ điều trị khoảng 12 người thì có thêm 1 người không quay lại kiểu uống một lần là nhiều. +- Lợi ích: Gộp 34 thử nghiệm chia nhóm ngẫu nhiên, 15197 người, lại với nhau: người được can thiệp ngắn sau một năm uống ít đi mỗi tuần 20 g rượu nguyên chất so với người chỉ được can thiệp tối thiểu hoặc không được can thiệp (95% CI ít hơn 12 đến 28 g, đây là khoảng tin cậy; bằng chứng chất lượng trung bình). Lượng rượu nền của người tham gia trong các nghiên cứu này trung bình 244 g mỗi tuần. Can thiệp ngắn nghĩa là lời khuyên hoặc tư vấn lối sống không quá 5 lần, tổng thời gian dưới 60 phút; nói chuyện lâu hơn không cho thêm lợi ích nào. Về thuốc: gộp 122 thử nghiệm chia nhóm ngẫu nhiên, 22803 người, acamprosat cứ điều trị 12 người (95% CI 8 đến 26) thì có thêm 1 người không quay lại uống bất kỳ lượng nào; naltrexon uống mỗi ngày 50 mg, cứ điều trị 12 người (8 đến 26) thì có thêm 1 người không quay lại mức uống nặng. - Mức bằng chứng: A - Nguồn: Kaner EF, Beyer FR, Muirhead C, et al. (2018). Effectiveness of brief alcohol interventions in primary care populations. Cochrane Database of Systematic Reviews, 2, CD004148. ；Jonas DE, Amick HR, Feltner C, et al. (2014). Pharmacotherapy for adults with alcohol use disorders in outpatient settings: a systematic review and meta-analysis. JAMA, 311(18), 1889–1900. -- Ghi chú: Mức lợi ích ghi「nhỏ」vì cả hai nguồn bằng chứng đo lượng uống, không đo tỷ lệ tử vong, và theo quy ước của cuốn sách này, chỉ có điểm thay thế thì ghi「nhỏ」. Phần thuốc đa số thử nghiệm làm ở nước ngoài, phần lớn người tham gia đã đạt tiêu chuẩn chẩn đoán nghiện rượu, và mọi thử nghiệm đều kèm can thiệp tâm lý xã hội, nên con số là「phần lợi ích cộng thêm ngoài cái nền đó」. Về chỉ định và khả năng kê đơn của hai loại thuốc này ở Trung Quốc, cứ theo bác sĩ và tờ hướng dẫn sử dụng thuốc, đừng tự mua qua mạng. Đã xuất hiện phản ứng cai rượu thì xem mục 21 của phần này (đừng tự cai cứng). +- Ghi chú: Mức lợi ích ghi「nhỏ」là vì cả hai nguồn bằng chứng đo lượng uống, không đo tỷ lệ tử vong, và theo quy ước của cuốn sách này, chỉ có điểm thay thế thì ghi「nhỏ」. Phần thuốc đa số thử nghiệm làm ở nước ngoài, phần lớn người tham gia đã đạt tiêu chuẩn chẩn đoán nghiện rượu, và mọi thử nghiệm đều kèm can thiệp tâm lý xã hội, nên con số là「phần lợi ích cộng thêm ngoài cái nền đó」. Về chỉ định và khả năng kê đơn của hai loại thuốc này ở Trung Quốc, cứ theo bác sĩ và tờ hướng dẫn sử dụng thuốc, đừng tự mua qua mạng. Đã xuất hiện phản ứng cai rượu thì xem mục 20 của phần này (đừng tự cai cứng). -### 23. Mỗi ngày ăn một nắm nhỏ hạt +### 22. Mỗi ngày ăn một nắm nhỏ hạt - Chi phí: Mỗi ngày ăn 28 g, một năm khoảng hai ba trăm tệ. Ăn tiện tay, không mất thời gian. - Nói thẳng: Người ăn hạt thường xuyên có xác suất tử vong trong cùng thời gian thấp hơn người không ăn khoảng hai phần mười. Ăn một lần mỗi tuần đã thấy khác biệt, thấp hơn khoảng một phần mười. Ăn đến mức mỗi ngày một nắm thì khác biệt rõ nhất. @@ -194,8 +185,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Bao Y, Han J, Hu FB, et al. (2013). Association of nut consumption with total and cause-specific mortality. New England Journal of Medicine, 369(21), 2001-2011. - Ghi chú: Mua loại nguyên vị, không thêm muối. Hạt có năng lượng không thấp, đừng ăn như đồ ăn vặt không kiểm soát. Cần nhớ các nghiên cứu loại này chỉ theo dõi dài hạn một nhóm người, người ăn hạt có lối sống nói chung tốt hơn, nên con số phải giảm bớt đi - -### 24. Đổi một phần thịt đỏ lấy cá và thịt gia cầm +### 23. Đổi một phần thịt đỏ lấy cá và thịt gia cầm - Chi phí: Không tốn tiền, cũng không mất thời gian. Đây là đổi loại thịt, không phải mua thêm. Chỗ khó là phải đổi một chút khẩu vị. - Nói thẳng: Người ăn thêm hai phần thịt chế biến hoặc thịt đỏ mỗi tuần có xác suất tử vong trong cùng thời gian cao hơn khoảng 3%. Đổi bằng cùng số phần cá hoặc thịt gia cầm thì không thấy tăng. Hiệu ứng của mục này nhỏ trong phần này, thuộc loại tiện tay là đổi, không phải loại phải gắng sức. @@ -204,8 +194,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Nguồn: Zhong VW, Van Horn L, Greenland P, et al. (2020). Associations of Processed Meat, Unprocessed Red Meat, Poultry, or Fish Intake With Incident Cardiovascular Disease and All-Cause Mortality. JAMA Internal Medicine, 180(4), 503-512. - Ghi chú: Hai phần mỗi tuần tạo ra khác biệt rất nhỏ, đừng trông chờ đổi thịt rồi sống thêm mấy năm. Cùng số tiền bỏ ra để ăn thịt thì có thể ưu tiên cá và thịt gia cầm - -### 25. Đổi một phần gạo trắng và bột mì tinh thành ngũ cốc nguyên hạt +### 24. Đổi một phần gạo trắng và bột mì tinh thành ngũ cốc nguyên hạt - Chi phí: Gạo lức, yến mạch, bánh mì nguyên cốc đắt hơn gạo trắng và bột mì tinh bột một chút, không mất thời gian. Chỗ khó là vị phải từ từ quen. - Nói thẳng: Người ăn thêm 90 g ngũ cốc nguyên mỗi ngày có xác suất tử vong trong cùng thời gian thấp hơn khoảng 17%. 90 g xấp xỉ ba phần. Ăn đến khoảng 200 g mỗi ngày thì con số vẫn đang giảm tiếp. @@ -213,8 +202,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Aune D và cộng sự (2016). Whole grain consumption and risk of cardiovascular disease, cancer, and all cause and cause specific mortality: systematic review and dose-response meta-analysis of prospective studies. BMJ. ；Schwingshackl L và cộng sự (2017). Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. American Journal of Clinical Nutrition. - Ghi chú: Không cần đổi hết sang ngũ cốc nguyên, đổi một nửa là đã nằm trong đoạn mà lợi ích thấy rõ nhất. Cần nhớ các nghiên cứu loại này chỉ ghi nhận và theo dõi, người ăn ngũ cốc nguyên vốn đã khoẻ hơn, nên con số có xu hướng lớn hơn thực tế. Kết quả giữa các nghiên cứu cũng chênh nhau khá xa (I² 83%, số này càng lớn thì kết luận của các nghiên cứu càng không nhất quán). - -### 26. Mỗi tuần uống trà từ ba lần trở lên +### 25. Mỗi tuần uống trà từ ba lần trở lên - Chi phí: Một năm mấy chục đến mấy trăm tệ. Pha một cốc mất mấy phút, không tốn nhiều thời gian. - Nói thẳng: Người có thói quen uống trà có xác suất tử vong trong cùng thời gian thấp hơn khoảng 15%. Tính từ tuổi 50, sống thêm trung bình 1.26 năm, và có thêm 1.41 năm không mắc bệnh tim mạch. @@ -222,8 +210,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Wang X, Liu F, Li J, et al. (2020). Tea consumption and the risk of atherosclerotic cardiovascular disease and all-cause mortality: The China-PAR project. European Journal of Preventive Cardiology, 27(18), 1956-1963. ；độc tố nấm men trong trà: Cui P và cộng sự (2020). Quantitative analysis and dietary risk assessment of aflatoxins in Chinese post-fermented dark tea. Food and Chemical Toxicology. ；Zhou H và cộng sự (2022). Mycotoxins in Tea (Camellia sinensis (L.) Kuntze): Contamination and Dietary Exposure Profiling in the Chinese Population. Toxins. ；chiết xuất trà xanh và tổn thương gan: Hu J và cộng sự (2018). The safety of green tea and green tea extract consumption in adults - Results of a systematic review. Regulatory Toxicology and Pharmacology. - Ghi chú:Tranh cãi. Các nghiên cứu này chỉ ghi nhận và theo dõi, và trong số nam giới Trung Quốc uống trà thì tỷ lệ hút thuốc cùng uống rượu cao. Tác giả đã trừ bớt một phần ảnh hưởng đó, nhưng trừ không sạch. Thêm một điểm nữa, đừng uống thứ nóng sôi, nhiệt độ của đồ uống nóng xem mục trong phần này về nhiệt độ đồ uống nóng. Có người lo trà chứa độc tố nấm men. Ở Trung Quốc đã kiểm tra 158 mẫu trà đen, có 2 mẫu phát hiện aflatoxin, tính theo lượng trà người ta uống thì lượng nạp vào thấp hơn giá trị có thể chấp nhận ở mức quốc tế. Một nghiên cứu khác kiểm tra 352 mẫu trà với 16 loại độc tố nấm men, chỉ có ochratoxin A trong trà đen là vượt giới hạn về trung bình. Tính theo lượng trà dân nước ta uống thì cả hai nghiên cứu đều không cho ra rủi ro khẩu phần. Giữ trà ở nơi khô, trà ẩm mốc thì đừng uống. Viên nang chiết xuất trà xanh là chuyện khác: nuốt một lượng lớn catechin nồng độ cao trong một lần thì có thể gây hại gan, trà pha không có vấn đề đó. Vậy mục này chỉ nói về trà pha, đừng lấy thực phẩm bổ sung dạng chiết xuất mà thay. - -### 27. Mỗi ngày uống ba đến bốn cốc cà phê, không đường không kem +### 26. Mỗi ngày uống ba đến bốn cốc cà phê, không đường không kem - Chi phí: Tự pha thì mỗi ngày mất một hai tệ. Pha một cốc mất mấy phút, không tốn nhiều thời gian. - Nói thẳng: Người mỗi ngày uống ba đến bốn cốc cà phê có xác suất tử vong trong cùng thời gian thấp hơn người không uống khoảng 17%. @@ -231,8 +218,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Poole R, Kennedy OJ, Roderick P, et al. (2017). Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ, 359, j5024. - Ghi chú:Tranh cãi. Tác giả ghi rõ bằng chứng gần như toàn bộ đến từ nghiên cứu theo dõi, và còn viết「cần thử nghiệm đối chứng ngẫu nhiên đáng tin cậy mới đánh giá được đó có phải quan hệ nhân quả hay không」. Nghĩa là phải chia ngẫu nhiên người tham gia thành hai nhóm để so sánh, chỉ nhìn nghiên cứu theo dõi thì không kết luận được nhân quả. Khi làm thì uống cà phê đen, thêm đường và kem sữa sẽ triệt tiêu lợi ích. Phụ nữ có thai, người bị loạn nhịp tim, lo âu, mất ngủ thì mục này không áp dụng - -### 28. Mỗi ngày ăn đủ 5 phần (khoảng 400 g) rau quả +### 27. Mỗi ngày ăn đủ 5 phần (khoảng 400 g) rau quả - Chi phí: Mỗi ngày mấy tệ đến mấy chục tệ. Mỗi ngày phải dành chút thời gian rửa và thái. Chỗ khó là phải mua và ăn đủ lượng mỗi ngày. - Nói thẳng: Người ăn thêm 200 g rau quả mỗi ngày có xác suất tử vong trong cùng thời gian thấp hơn khoảng một phần mười. Người ăn đủ 5 phần mỗi ngày thấp hơn người chỉ ăn 2 phần khoảng 13%. Ăn nhiều hơn nữa thì không thấp hơn nữa. @@ -240,17 +226,16 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Aune D và cộng sự (2017). Fruit and vegetable intake and the risk of cardiovascular disease, total cancer and all-cause mortality: a systematic review and dose-response meta-analysis of prospective studies. International Journal of Epidemiology. ；Wang DD và cộng sự (2021). Fruit and Vegetable Intake and Mortality: Results From 2 Prospective Cohort Studies of US Men and Women and a Meta-Analysis of 26 Cohort Studies. Circulation. - Ghi chú: Ăn đủ 5 phần là xong, không cần lên tới 10 phần, cả hai nghiên cứu trả lời giống nhau ở điểm này. Cần nhớ các nghiên cứu loại này chỉ ghi nhận và theo dõi, nhiễu rất rõ: người ăn nhiều rau khác người khác cả về thu nhập, giáo dục, mức vận động. Vậy nên「rủi ro thấp hơn khoảng 10%」đã là trần. - -### 29. Ăn ít thực phẩm siêu chế biến (khoai tây chiên, mì gói, bánh ngọt, đồ ăn nhanh) +### 28. Ăn ít thực phẩm siêu chế biến (khoai tây chiên, mì gói, bánh ngọt, đồ ăn nhanh) -- Chi phí: Không tốn tiền, chỉ là đổi chỗ mua. Tự nấu ăn, hoặc bỏ thời gian chọn nguyên liệu ít chế biến, ngày nào cũng phải tốn chút thời gian. Chỗ khó là phải chống lại thói quen「cứ tiện là được」trong thời gian dài. +- Chi phí: Không tốn tiền, chỉ là đổi sang mua thứ khác thôi. Tự nấu ăn, hoặc bỏ thời gian chọn nguyên liệu ít chế biến, ngày nào cũng phải tốn chút thời gian. Khó ở chỗ phải chống lại thói quen「cứ tiện là được」trong thời gian dài. - Nói thẳng: Người ăn nhiều thực phẩm siêu chế biến nhất có xác suất tử vong trong cùng thời gian cao hơn người ăn ít nhất khoảng hai phần mười, tử vong tim mạch cao hơn năm phần mười. Tuy nhiên loại bằng chứng này được đánh giá là không chắc chắn lắm, phải giảm bớt con số khi đọc. -- Lợi ích: Một bài tổng quan gộp nhiều nghiên cứu tổng hợp lại với nhau, và các nghiên cứu gốc bên dưới những bảng tổng hợp đó đều chỉ ghi nhận, không chia nhóm đối chứng. Kết quả: người ăn nhiều thực phẩm siêu chế biến có rủi ro tử vong cao hơn người ăn ít khoảng 21% (RR 1.21), tử vong do bệnh tim mạch cao hơn khoảng 50% (RR 1.50). Hai kết quả này được xếp mức bằng chứng lần lượt là「gợi ý mạnh」và「đáng tin cậy」. Nhưng theo cách chấm điểm bằng chứng thông dụng (GRADE), độ chắc chắn là thấp hoặc rất thấp. +- Lợi ích: Một bài tổng quan gộp nhiều nghiên cứu tổng hợp lại với nhau, và các nghiên cứu gốc bên dưới những bảng tổng hợp đó đều chỉ ghi nhận, không chia nhóm. Kết quả: người ăn nhiều thực phẩm siêu chế biến có rủi ro tử vong cao hơn người ăn ít khoảng 21% (RR 1.21), tử vong do bệnh tim mạch cao hơn khoảng 50% (RR 1.50). Hai kết quả này được xếp mức bằng chứng lần lượt là「gợi ý mạnh」và「đáng tin cậy」. Nhưng theo cách chấm điểm bằng chứng thông dụng (GRADE), độ chắc chắn là thấp hoặc rất thấp. - Mức bằng chứng: A - Nguồn: Lane MM và cộng sự (2024). Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. -- Ghi chú:Tranh cãi nằm ở ba điểm. Một là bộ phân loại này (NOVA) gom chung một nhóm các thực phẩm khác nhau khá xa về dinh dưỡng. Hai là nó trùng lặp rất nhiều với đồ uống có đường và thịt chế biến (mục 7 và 19), không phân biệt được chính xác yếu tố nào đang tác động. Ba là theo cách chấm điểm bằng chứng thông dụng (GRADE) thì mức độ cũng thấp. Vậy sau khi làm cho xong hai mục kia, mục này còn thêm được bao nhiêu lợi ích thì không chắc. Bằng chứng ở phe bên kia chính là mức đánh giá thấp mà cùng bài tổng quan đó đưa ra, hiện chưa có nghiên cứu gốc nào ra kết luận ngược lại. Vậy hãy làm cho xong mục 7 và 19 (không uống đồ uống có đường, ăn ít thịt chế biến) rồi hẵng xem mục này. +- Ghi chú: Tranh cãi nằm ở ba điểm. Một là bộ phân loại này (NOVA) gom chung một nhóm các thực phẩm khác nhau khá xa về dinh dưỡng. Hai là nó trùng lặp rất nhiều với đồ uống có đường và thịt chế biến (mục 7 và 18), không phân biệt được chính xác yếu tố nào đang tác động. Ba là theo cách chấm điểm bằng chứng thông dụng (GRADE) thì mức độ cũng thấp. Vậy sau khi làm cho xong hai mục kia, mục này còn thêm được bao nhiêu lợi ích thì không chắc. Bằng chứng ở phe bên kia chính là mức đánh giá thấp mà cùng bài tổng quan đó đưa ra, hiện chưa có nghiên cứu gốc nào ra kết luận ngược lại. Vậy hãy làm cho xong mục 7 và 18 (không uống đồ uống có đường, ăn ít thịt chế biến) rồi hẳng xem mục này. -### 30. Nấu ăn và sưởi ấm không đốt than củi, chuyển sang điện hoặc khí đốt +### 29. Nấu ăn và sưởi ấm không đốt than củi, chuyển sang điện hoặc khí đốt - Chi phí: Hộ gia đình nông thôn phải thay bếp, cộng thêm tiền nhiên liệu, một năm mấy trăm đến hơn một nghìn tệ. Thay một lần là xong, không chiếm thời gian dài hạn. Người ở thành phố vốn đã dùng điện hoặc khí đốt. - Nói thẳng: Người đốt than hoặc củi để nấu ăn có xác suất tử vong trong cùng thời gian cao hơn khoảng một phần mười; người đốt than củi để sưởi cao hơn khoảng 14%. Người đã đổi sang điện hoặc khí đốt có xác suất tử vong thấp hơn 13%–33% so với người vẫn đang đốt than củi. @@ -259,8 +244,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Nguồn: Yu K và cộng sự (2018). Association of Solid Fuel Use With Risk of Cardiovascular and All-Cause Mortality in Rural China. JAMA. ；Chen J, Hoek G (2020). Long-term exposure to PM and all-cause and cause-specific mortality: A systematic review and meta-analysis. Environment International. - Ghi chú: Loại nghiên cứu này chỉ ghi nhận và theo dõi. Hộ đổi nhiên liệu thường cũng giàu hơn, nên con số「rủi ro rơi còn 0.67 lần」có lẫn ảnh hưởng của hoàn cảnh gia đình. Còn với PM2.5 ngoài trời thì cá nhân làm được không nhiều, chỉ là chuyển nhà, đeo khẩu trang, dùng máy lọc không khí. Máy lọc không khí chưa có nghiên cứu nào lấy tỷ lệ tử vong làm kết quả cuối, nên phần này không đưa ra con số. - -### 31. Đồ nóng để nguội rồi hãy uống, không uống trà, canh hay cà phê sôi +### 30. Đồ nóng để nguội rồi hãy uống, không uống trà, canh hay cà phê sôi - Chi phí: Không tốn tiền, mỗi lần chỉ phải chờ thêm hai ba phút. Chỗ khó là nhịn không uống lúc thèm. - Nói thẳng: Uống trà rất nóng thì xác suất mắc ung thư thực quản gấp 8 lần so với uống trà ấm. Ngay cả khi chỉ「nóng」thì cũng gấp 2 lần. Rót ra mà dưới hai phút là uống thì rủi ro gấp 5 lần so với chờ trên bốn phút rồi uống. @@ -268,8 +252,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Islami F, Pourshams A, Nasrollahzadeh D, et al. (2009). Tea drinking habits and oesophageal cancer in a high risk area in northern Iran: population based case-control study. BMJ, 338, b929. ；Loomis D, Guyton KZ, Grosse Y, et al. (2016). Carcinogenicity of drinking coffee, mate, and very hot beverages. Lancet Oncology, 17(7), 877-878. - Ghi chú: Cơ quan Quốc tế Nghiên cứu Ung thư xếp đồ uống nóng trên 65 ℃ vào nhóm 2A, nghĩa là rất có thể gây ung thư. Cùng bản đánh giá đó kết luận bản thân cà phê không gây ung thư. Vùng Triều Châu – Sơn Đông và vùng Thái Hàng Sơn của Trung Quốc có nhiều ung thư thực quản, liên quan tới thói quen uống khi còn nóng - -### 32. Ban ngày ra ngoài hẵn nắng, đừng cả ngày không thấy ánh sáng +### 31. Ban ngày ra ngoài hẵn nắng, đừng cả ngày không thấy ánh sáng - Chi phí: Không tốn tiền. Mỗi ngàm mười mấy phút, đi làm về và nghỉ trưa là đủ. - Nói thẳng: Phụ nữ cố tránh nắng có tuổi thọ kỳ vọng thấp hơn người phơi nắng nhiều nhất 0.6 đến 2.1 năm. Người không hút thuốc nhưng tránh nắng thì sống lâu gần bằng người hút thuốc mà phơi nắng nhiều nhất. @@ -277,8 +260,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: B - Nguồn: Lindqvist PG, Epstein E, Nielsen K, et al. (2016). Avoidance of sun exposure as a risk factor for major causes of death: a competing risk analysis of the Melanoma in Southern Sweden cohort. Journal of Internal Medicine, 280(4), 375-387. - Ghi chú:Tranh cãi. Thụy Điển có vĩ độ cao, ít nắng, kết luận không chuyển thẳng sang Trung Quốc được. Người tránh nắng cũng có thể vốn đã yếu, không thích ra ngoài. Thêm nữa, đừng phơi đến mức cháy da, dưới ánh nắng gay gắt giữa trưa vẫn phải che, vì phơi nắng làm tăng nguy cơ ung thư da - -### 33. Giữ BMI trong khoảng 20–25, thừa cân thì giảm +### 32. Giữ BMI trong khoảng 20–25, thừa cân thì giảm - Chi phí: Không tốn tiền. Ăn uống cộng vận động thì ngày nào cũng phải tốn thời gian. Đây là mục tốn ý chí nhất của cả phần: phải kiềm chế ăn uống lâu dài, và giữ nguyên sau khi giảm còn khó hơn lúc giảm. - Nói thẳng: Xác suất tử vong thấp nhất ở đoạn BMI 20–25. BMI lên 27.5–30 thì cao hơn khoảng hai phần mười, 30–35 cao hơn khoảng bốn phần rưỡi, 35–40 gần gấp đôi. Người Đông Á nhạy với thừa cân hơn, mỗi 5 BMI tăng thì rủi ro cao hơn khoảng bốn phần mười. @@ -286,7 +268,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Global BMI Mortality Collaboration (2016). Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet. ；Flegal KM và cộng sự (2013). Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA. （phe bên kia） - Ghi chú:Tranh cãi. Có phe cho rằng「béo một chút lại sống lâu hơn」. Tổng hợp của phe đó cho ra: người thừa cân có rủi ro tử vong thấp hơn khoảng 6%, người béo phì nhẹ thấp hơn khoảng 5%. Hai phe chủ yếu khác nhau ở chỗ có nên loại bỏ ba nhóm dữ liệu: người hút thuốc, người đã mắc bệnh, và dữ liệu trong vài năm đầu sau khi bắt đầu nghiên cứu. Loại bỏ là vì người bệnh nặng sẽ sụt cân trước. Phân tích gộp 239 nghiên cứu nói trên, sau khi loại bỏ ba nhóm đó, rủi ro do thừa cân lại xuất hiện. Cần thêm một điểm nữa: không có thử nghiệm chia nhóm ngẫu nhiên nào chứng minh việc giảm cân bản thân nó làm giảm tỷ lệ tử vong toàn nguyên. Lợi ích ở trên là do so sánh ngang giữa những người béo khác nhau, không có nghĩa giảm đi sẽ giảm được tới mức đó. Người muốn giảm cân không cần bỏ công sắp xếp bữa ăn, bữa sáng và chế độ nhịn ăn 16:8 đều không có lợi ích thêm, xem mục 26 ở phần 6 (đừng trông chờ bữa sáng hoặc 16:8 giúp kiểm soát cân nặng) -### 34. Mỗi tuần ăn ớt từ bốn lần trở lên +### 33. Mỗi tuần ăn ớt từ bốn lần trở lên - Chi phí: Không tốn tiền, cũng không mất thời gian. - Nói thẳng: Người ăn ớt từ bốn lần mỗi tuần trở lên có xác suất tử vong trong cùng thời gian thấp hơn người ăn rất ít khoảng 23%, tử vong tim mạch thấp hơn khoảng một phần ba. @@ -294,8 +276,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: B - Nguồn: Bonaccio M, Di Castelnuovo A, Costanzo S, et al. (2019). Chili Pepper Consumption and Mortality in Italian Adults. Journal of the American College of Cardiology, 74(25), 3139-3149. - Ghi chú:Tranh cãi. Đây là nghiên cứu theo dõi ghi nhận, làm trong bối cảnh khẩu ăn Địa Trung Hải, người ăn cay có thể cả ăn uống nói chung sát hơn với cách ăn truyền thống địa phương. Thêm nữa, người bị trào ngược dạ dày lên thực quản, bị trĩ hay hội chứng ruột kích thích thì ăn cay sẽ nặng thêm triệu chứng, không cần cố - -### 35. Mỗi ngày uống một hai phần sữa hoặc sữa chua +### 34. Mỗi ngày uống một hai phần sữa hoặc sữa chua - Chi phí: Mỗi ngày mấy tệ, không mất thời gian. - Nói thẳng: Người uống từ hai phần sữa hoặc sữa chua mỗi ngày trở lên có xác suất tử vong trong cùng thời gian thấp hơn người không uống khoảng 17%. @@ -303,8 +284,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: B - Nguồn: Dehghan M, Mente A, Rangarajan S, et al. (2018). Association of dairy intake with cardiovascular disease and mortality in 21 countries from five continents (PURE): a prospective cohort study. Lancet, 392(10161), 2288-2297. - Ghi chú:Tranh cãi. PURE đưa vào rất nhiều nước thu nhập trung bình và thấp, ở đó mà uống sữa được mỗi ngày thì bản thân việc đó đã nói lên hoàn cảnh gia đình khá, tầng nhiễu này khó loại sạch. Ngoài ra người không dung nạp được đường sữa thì chuyển sang uống sữa chua hoặc sữa ít đường sữa, không cần cố uống sữa tươi - -### 36. Không cần kiêng trứng, nhưng đừng ăn ba bốn quả mỗi ngày +### 35. Không cần kiêng trứng, nhưng đừng ăn ba bốn quả mỗi ngày - Chi phí: Không tốn tiền, cũng không mất thời gian. - Nói thẳng: Người ăn thêm nửa quả trứng nguyên mỗi ngày có xác suất tử vong trong cùng thời gian cao hơn khoảng 7%. Đổi nửa quả đó ra lòng trắng trứng, cá, thịt gia cầm hoặc hạt thì rủi ro tử vong của từng nhóm lại thấp hơn. Không cần kiêng trứng, chỉ là đừng ăn ba bốn quả mỗi ngày. @@ -312,8 +292,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: B - Nguồn: Zhuang P, Wu F, Mao L, et al. (2021). Egg and cholesterol consumption and mortality from cardiovascular and different causes in the United States: A population-based cohort study. PLoS Medicine, 18(2), e1003508. - Ghi chú:Tranh cãi khá lớn: có những nghiên cứu theo dõi dài hạn quần số lớn, và cả những bảng gộp nhiều nghiên cứu, không thấy trứng có liên quan gì tới tỷ lệ tử vong. Hướng dẫn khẩu phần của Mỹ đã bỏ giới hạn cholesterol theo ngày. Thứ cần tránh chỉ là ăn ba bốn quả mỗi ngày, một ngày một quả thì không phải lo - -### 37. Có điều kiện thì ngâm mình, đừng chỉ tắm vòi sen +### 36. Có điều kiện thì ngâm mình, đừng chỉ tắm vòi sen - Chi phí: Tốn thêm chút tiền nước và điện. Mỗi lần nửa tiếng. - Nói thẳng: Người gần như ngày nào cũng ngâm mình có rủi ro sự kiện tim mạch thấp hơn người ngâm dưới hai lần mỗi tuần khoảng 28%, xuất huyết não thấp hơn khoảng 46%. @@ -322,8 +301,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Nguồn: Ukai T, Iso H, Yamagishi K, et al. (2020). Habitual tub bathing and risks of incident coronary heart disease and stroke. Heart, 106(10), 732-737. - Ghi chú:Tranh cãi. Đây là kết quả làm ra ở dân số Nhật với văn hoá tắm bồn, phần lớn hộ gia đình Trung Quốc không có điều kiện ngâm mình mỗi ngày. Ngoài ra đừng để nước quá nóng, cũng đừng ngâm quá lâu. Với người lớn tuổi và người bị bệnh tim mạch, nước quá nóng hoặc ngâm quá lâu lại thành nguy hiểm, mỗi năm ở Nhật có rất nhiều người lớn tuổi đột tử trong bồn tắm - -### 38. Giữ giấc ngủ trưa trong nửa tiếng, đừng quá một tiếng; nếu cứ phải ngủ một hai tiếng mới chịu thì đi tìm nguyên nhân +### 37. Giữ giấc ngủ trưa trong nửa tiếng, đừng quá một tiếng; nếu cứ phải ngủ một hai tiếng mới chịu thì đi tìm nguyên nhân - Chi phí: Không tốn tiền. Đặt một cái báo thức trước khi ngủ là xong, gần như không cần ý chí. - Nói thẳng: Người ngủ trưa không quá một tiếng thì rủi ro tử vong và bệnh mạn tính không tăng, đầu óc còn tỉnh táo hơn một chút. Người ngủ quá một tiếng có rủi ro bệnh tim mạch vành cao hơn khoảng ba phần mười, đái thường và béo phì cao hơn khoảng hai phần mười. Vậy hãy đặt báo thức, nửa tiếng là tự đánh thức mình. Nếu ngày nào cũng phải ngủ một hai tiếng mới chịu thì đó là tín hiệu cần đi khám. @@ -331,8 +309,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: B - Nguồn: Du P, Li J, Hua Z, v.v. (2026). Multiple Health Outcomes of Daytime Napping: A Comprehensive Umbrella Review. Public Health Reviews. ；Gao C, Cai R, Zheng X, v.v. (2026). Objectively Measured Daytime Napping Patterns and All-Cause Mortality in Older Adults. JAMA Network Open. ；Dashti HS và cộng sự (2021). Genetic determinants of daytime napping and effects on cardiometabolic health. Nature Communications. - Ghi chú:Tranh cãi. Hai nghiên cứu trên đều chỉ ghi nhận, không chia nhóm đối chứng. Người ngủ trưa lâu vốn dễ có ngưng thở khi ngủ, thiếu máu, suy giảm chức năng tuyến giáp, trầm cảm hoặc bệnh mạn tính khác. Trong câu「ngủ nhiều thì chết nhiều」 có bao nhiêu là do bản thân giấc ngủ trưa thì không rõ. Suy luận nhân quả bằng gen (Mendelian ngẫu nhiên, 453.000 người cộng 541.000 người) chỉ ủng hộ một điều: ngủ trưa càng thường xuyên thì huyết áp và vòng eo càng cao hơn một chút. Nó không tìm thấy ảnh hưởng nhân quả của giấc ngủ trưa lên bệnh tim mạch vành và đái thường. Vậy đừng coi「ngủ trưa ít đi」là cách hạ tỷ lệ tử vong. Nửa「đi tìm nguyên nhân」là khuyến nghị của tác giả, thuộc mức C: ban ngày mệt đến mức bắt buộc phải ngủ lâu thì trước hết xem giấc ngủ ban đêm, ngáy và ngạt thở trong lúc ngủ thì xem ngưng thở khi ngủ, rồi xem công thức máu và tuyến giáp. Giấc ngủ trưa ngắn làm sao để tỉnh táo, xem mục 11 ở phần 3 (buổi chiều mệt thì ngủ 10 phút). Ban đêm ngủ bao nhiêu lâu, xem mục 13 của phần này (mỗi đêm ngủ 7 tiếng). - -### 39. Thức khuya xong thì đêm hôm sau bù ngay, đừng dồn dập cuối tuần +### 38. Thức khuya xong thì đêm hôm sau bù ngay, đừng dồn dập cuối tuần - Chi phí: Không tốn tiền. Một hai tiếng bù vốn cũng là lúc bạn muốn ngủ. - Nói thẳng: Người ngủ thiếu mà đêm hôm sau không bù thì rủi ro tử vong cao hơn người có giờ sinh hoạt đều đặn khoảng 15%; người thiếu nhiều mà vẫn không bù thì cao hơn khoảng 42%. Người có bù thì không thấy rủi ro tăng. Vậy thức khuya xong thì đêm hôm sau lên giường sớm. @@ -340,18 +317,17 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: B - Nguồn: Li X, Zhang M, Li Z, v.v. (2026). Acute sleep rebound following sleep restriction is associated with reduced mortality risk. Nature Communications. - Ghi chú: Đây là nghiên cứu chỉ ghi nhận, không chia nhóm đối chứng. Người không bù được có thể vốn bận hơn, sức khoẻ kém hơn, tầng nhiễu đó không loại sạch được. Câu「hai nhóm đã bù không thấy rủi ro tăng」chỉ nói là không tìm ra khác biệt, ước lượng điểm của chúng vẫn lớn hơn 1, đã bù không có nghĩa là không sao. Vậy sau khi thức khuaya phải bù càng sớm càng tốt, cũng không vì thế mà thức khuaya tuỳ tiện. Trong nghiên cứu này, bù ngủ nghĩa là đêm đầu tiên sau khi thiếu ngủ ngủ thêm một lúc, không gồm việc dồn lại rồi ngủ một lần cuối tuần. Việc biến「thức khuya ngày làm việc, bù ngủ cuối tuần」thành một nhịp cố định hằng tuần gọi là chênh lệch giờ sinh học xã hội, bản thân nó liên quan tới bệnh tim mạch, xem mục 13 của phần này (mỗi đêm ngủ khoảng 7 tiếng, giờ sinh hoạt đều đặn). Mục về giờ dậy cố định cả cuối tuần nằm ở mục 2 ở phần 3 (giờ dậy cố định, cuối tuần cũng vậy), không mâu thuẫn với mục này: mục này bảo bạn lên giường sớm hôm đó, không phải bảo hôm sau ngủ tới trưa. - -### 40. Ca đêm làm càng lâu nguy cơ tim mạch càng cao, chuyển được ca thì chuyển sớm +### 39. Làm ca đêm càng lâu thì rủi ro tim mạch càng cao, đổi được chỗ thì nên đổi sớm -- Chi phí: Chuyển ca hoặc đổi việc có thể giảm thu nhập, tiền ca đêm cũng mất. Tìm việc tốn thời gian, quyết định cũng khó. Nếu tính khoản này vào lúc chọn việc thì chi phí bằng không. -- Nói thẳng: Người làm ca đêm mắc bệnh tim mạch nhiều hơn người không làm ca đêm khoảng 13%, tử vong tim mạch nhiều hơn khoảng 27%. Mà làm càng lâu thì càng tăng: mỗi 5 năm ca đêm thêm, bệnh tim mạch tăng thêm khoảng 7%. Vậy khoản này tính theo số năm, chuyển được ca thì chuyển sớm. Còn chuyện ca đêm gây ung thư thì bằng chứng mềm hơn nhiều so với lời đồn. -- Lợi ích: Gộp 23 nghiên cứu theo dõi lại với nhau, chỉ ghi nhận, không chia nhóm đối chứng. Người làm ca đêm có rủi ro sự kiện tim mạch cao hơn khoảng 13% (RR 1.13, 95% CI 1.10–1.16, đây là khoảng tin cậy). Tử vong tim mạch cao hơn khoảng 27% (RR 1.27, 1.18–1.36). Xét theo số năm: mỗi 5 năm ca đêm thêm, tỷ lệ mắc bệnh tim mạch tăng thêm khoảng 7% (RR 1.07, 1.04–1.09), tử vong tim mạch tăng thêm khoảng 5% (RR 1.05, 1.03–1.06). Tách theo từng bệnh: tỷ lệ mắc bệnh tim mạch vành cao hơn khoảng 22% (1.22, 1.16–1.28), tử vong do bệnh tim mạch vành cao hơn khoảng 22% (1.22, 1.10–1.36), tử vong do bệnh tim mạch thiếu máu cục bộ cao hơn khoảng 39% (1.39, 1.06–1.84), tử vong do đột quỵ cao hơn khoảng 49% (1.49, 1.04–2.12). Riêng tỷ lệ mắc đột quỵ thì không có ý nghĩa (1.06, 0.95–1.18). +- Chi phí: Đổi vị trí hoặc đổi việc có thể thu nhập ít hơn, tiền phụ ca đêm cũng mất. Tìm việc mất thời gian, quyết định cũng khó. Nếu tính khoản này vào lúc chọn việc thì chi phí bằng không. +- Nói thẳng: Người làm ca đêm mắc bệnh tim mạch nhiều hơn người không làm ca đêm khoảng 13%, tử vong tim mạch nhiều hơn khoảng 27%. Hơn nữa làm càng lâu thì càng nhiều: mỗi thêm 5 năm ca đêm thì bệnh tim mạch lại nhiều thêm khoảng 7%. Vậy tính theo số năm thì đổi được vị trí thì nên đổi sớm. Còn chuyện ca đêm gây ung thư thì bằng chứng mềm hơn nhiều so với lời đồn. +- Lợi ích: Gộp 23 nghiên cứu theo dõi lại với nhau, chỉ ghi nhận, không chia nhóm đối chứng. Người làm ca đêm có rủi ro sự kiện tim mạch cao hơn khoảng 13% (RR 1.13, 95% CI 1.10–1.16, đây là khoảng tin cậy). Tử vong tim mạch cao hơn khoảng 27% (RR 1.27, 1.18–1.36). Xét theo số năm: mỗi thêm 5 năm ca đêm thì tỷ lệ mắc bệnh tim mạch cao thêm khoảng 7% (RR 1.07, 1.04–1.09), tử vong tim mạch cao thêm khoảng 5% (RR 1.05, 1.03–1.06). Tách theo từng bệnh: bệnh mạch vành khởi phát cao hơn khoảng 22% (1.22, 1.16–1.28), tử vong do bệnh mạch vành cao hơn khoảng 22% (1.22, 1.10–1.36), tử vong do bệnh tim mạch thiếu máu cục bộ cao hơn khoảng 39% (1.39, 1.06–1.84), tử vong do đột quỵ cao hơn khoảng 49% (1.49, 1.04–2.12). Riêng phần tỷ lệ khởi phát đột quỵ thì không có ý nghĩa (1.06, 0.95–1.18). - Mức bằng chứng: A -- Nguồn: Xi J, Ma W, Tao Y, v.v. (2025). Association between night shift work and cardiovascular disease: a systematic review and dose-response meta-analysis. Frontiers in Public Health. ；Esposito G, Bravi F, Santucci C, v.v. (2025). Night shift work and breast cancer risk in healthcare workers: a systematic review and meta-analysis. Occupational Medicine. ；Shen QM, Li ZY, Tan YT, v.v. (2026). Night shift work and risk of total and site-specific cancer: results from a prospective cohort study among Chinese men. Scandinavian Journal of Work, Environment & Health. ；Czeisler CA, Johnson MP, Duffy JF, v.v. (1990). Exposure to bright light and darkness to treat physiologic maladaptation to night work. New England Journal of Medicine, 322(18), 1253-1259. -- Ghi chú: Tất cả đều là nghiên cứu chỉ ghi nhận, không chia nhóm đối chứng. Người làm ca đêm vốn khác người khác ở hút thuốc, cân nặng, thu nhập, hiệu chỉnh không sạch. Chuyện ca đêm gây ung thư thì bằng chứng mềm hơn nhiều so với lời đồn. Trong phân tích gộp 12 nghiên cứu, 12.132 ca ung thư vú, người từng làm ca đêm chỉ có rủi ro cao hơn khoảng 5%, khoảng tin cậy vượt qua 1 (RR 1.05, 0.96–1.14). Nhóm làm đủ 20 năm cao hơn khoảng 25% (1.25, 1.01–1.55), nhưng tác giả làm phân tích độ nhạy thì kết quả này không còn ý nghĩa, hiệu chỉnh thiên lệch xuất bản có thể thì gần 1, nguyên văn là mối liên hệ này「còn xa mới được xác lập」. Một nghiên cứu khác theo dõi nam giới Trung Quốc trong 16.1 năm với 8.202 ca ung thư không thấy ca đêm liên quan tới ung thư tổng thể và phần lớn các loại ung thư thường gặp, chỉ riêng ung thư tuyến tụy cao hơn khoảng 59% ở nhóm lũy kế từ 11 đến 20 năm (HR 1.59, 1.09–2.31). Vậy mục này tính sổ trên phía tim mạch, không tính theo ung thư. Không phải ai cũng nên rời khỏi ca đêm, khoản tính theo số năm phải tự bạn tính. Người đang làm thì trước hết sửa những gì sửa được: cai thuốc lá xem mục 1 của phần này (cai thuốc lá, càng sớm càng tốt), huyết áp và mỡ máu xem mục 12 của phần này (có ca huyết áp cao, mỡ máu cao thì uống thuốc đúng đơn đều đặn), sau ca đêm thì bù ngủ thế nào xem mục 39 của phần này (thức khuaya thì đêm hôm sau bù ngay). Có thêm hai cách đối phóng đáng thử: dồn ăn uống vào ban ngày, và dùng ánh sáng mạnh cộng với che tối nghiêm ngặt ban ngày để dịch đồng hồ sinh học. Cách sau là cách duy nhất được chứng minh thực sự dịch được đồng hồ sinh học. Dưới ánh sáng trong nhà bình thường, làm liên tiếp sáu ca đêm thì đồng hồ sinh học không nhúc nhích; đổi sang ánh sáng mạnh 7000 đến 12000 lux cộng với ban ngày gần như tối đen, bốn ngày là đảo ngược trọn vẹn. **Ánh sáng ban đêm phải đủ sáng, lúc ngủ ban ngày phải che tối nghiêm ngặt, và trên đường đi làm về phải đeo kính râm chặn ánh sáng buổi sớm, thiếu một trong ba thì vô nghĩa**, bỏ cái cuối thì coi như không làm. Cả hai cách đối phóng chỉ đi được tới mức chỉ số chuyển hoá và giấc ngủ, chưa thử nghiệm nào đo xem làm theo có hạ được rủi ro tim mạch hay không. Ngày nghỉ mà sống theo lịch ban ngày bình thường thì đồng hồ lại bị đẩy ngược lại. Về liều dùng, các con số gốc và phần nguồn còn lại, cùng với cách cơ thể nhận biết thời gian, đều ở [docs/生物钟和夜班.md](../docs/生物钟和夜班.md)。 +- Nguồn: Xi J, Ma W, Tao Y và cộng sự (2025). Association between night shift work and cardiovascular disease: a systematic review and dose-response meta-analysis. Frontiers in Public Health. ; Esposito G, Bravi F, Santucci C và cộng sự (2025). Night shift work and breast cancer risk in healthcare workers: a systematic review and meta-analysis. Occupational Medicine. ; Shen QM, Li ZY, Tan YT và cộng sự (2026). Night shift work and risk of total and site-specific cancer: results from a prospective cohort study among Chinese men. Scandinavian Journal of Work, Environment & Health. ; Czeisler CA, Johnson MP, Duffy JF và cộng sự (1990). Exposure to bright light and darkness to treat physiologic maladaptation to night work. New England Journal of Medicine, 322(18), 1253-1259. +- Ghi chú: Đây đều là nghiên cứu chỉ ghi nhận, không chia nhóm đối chứng. Người làm ca đêm vốn đã khác người khác ở hút thuốc, cân nặng và thu nhập, hiệu chỉnh không sạch. Chuyện ca đêm gây ung thư thì bằng chứng mềm hơn nhiều so với lời đồn. Trong bảng gộp 12 nghiên cứu với 12.132 ca ung thư vú, người từng làm ca đêm chỉ cao hơn khoảng 5%, khoảng tin cậy vắt qua 1 (RR 1.05, 0.96–1.14). Nhóm làm đủ 20 năm cao hơn khoảng 25% (1.25, 1.01–1.55), nhưng khi tác giả làm phân tích ảnh hưởng thì con số đó thành không có ý nghĩa, hiệu chỉnh thiên lệch xuất bản có thể có thì gần bằng 1, nguyên văn các ông viết liên hệ này「còn xa lắm mới được xác lập」. Một nghiên cứu khác theo dõi nam giới Trung Quốc trong 16.1 năm với 8.202 ca ung thư, ca đêm không gắn với ung thư nói chung cũng như phần lớn các loại ung thư thường gặp, chỉ có ung thư tuyến tụy ở nhóm cộng dồn 11 đến 20 năm là cao hơn khoảng 59% (HR 1.59, 1.09–2.31). Vậy mục này tính sổ theo bệnh tim mạch, không theo ung thư. Không phải ai cũng nên rời khỏi chỗ ca đêm, sổ tính theo số năm thì bạn phải tự tính. Người đang làm ca đêm thì sửa cái sửa được trước: bỏ thuốc lá xem mục 1 của phần này (bỏ thuốc lá, càng sớm càng tốt), huyết áp và mỡ máu xem mục 12 của phần này (có cao huyết áp, mỡ máu cao thì uống thuốc đúng y lệnh, đừng tự ý ngừng), sau ca đêm ngủ bù thế nào xem mục 38 của phần này (thức khuaya xong thì đêm hôm sau bù ngay). Còn hai cách đáng thử: gom ăn uống vào ban ngày, và dùng ánh sáng mạnh cộng với che tối kỹ ban ngày để dời đồng hồ sinh học. Cách sau là cách duy nhất được chứng minh là thật sự dời được đồng hồ sinh học. Dưới ánh sáng trong nhà bình thường, làm liền sáu ca đêm thì đồng hồ sinh học không nhúc nhích; đổi sang ánh sáng mạnh 7000 đến 12000 lux cộng với ban ngày tối gần như đen, bốn ngày là đảo ngược hẳn. **Ánh sáng ban đêm phải đủ sáng, lúc ngủ ban ngày phải che tối kỹ, và trên đường đi làm về phải đeo kính râm chặn ánh sáng buổi sớm, thiếu một trong ba là không được**, bỏ cái cuối cùng thì coi như làm bài rỗng. Cả hai cách chỉ dừng ở chỉ số chuyển hoá và giấc ngủ, chưa thử nghiệm nào đo được làm theo có hạ được rủi ro tim mạch không. Ngày nghỉ quay về nhịp sống ban ngày thì đồng hồ sinh học lại bị đẩy lùi. Liều dùng, số liệu gốc và các nguồn còn lại, cùng với cách cơ thể nhận biết thời gian, đều ở [docs/生物钟和夜班.md](../docs/生物钟和夜班.md). -### 41. Mua dầu ăn đóng gói có số SC, đừng mua dầu điều tự ép tẩy từ xưởng nhỏ bán rời +### 40. Mua dầu ăn đóng gói có số SC, đừng mua dầu điều tự ép tẩy từ xưởng nhỏ bán rời - Chi phí: Đổi chỗ mua thôi, tiền tương đương. Chỉ cần nhìn trên bao bì có số giấy phép sản xuất thực phẩm bắt đầu bằng SC hay không. Chỗ khó là thuyết phục người lớn tuổi quen mua dầu tự ép. - Nói thẳng: Đậu phộng bị ẩm mốc sinh aflatoxin, đây là chất gây ung thư đã rõ, và nó gây hại chủ yếu ở gan. Sau khi Quảng Châu quản được các xưởng nhỏ, tỷ lệ dân thường ăn dầu tự ép có xét nghiệm chức năng gan bất thường giảm khoảng ba phần rưỡi. Ở Quảng Tây, phụ nữ có thai ăn dầu điều tự ép có nguy cơ sinh con cân nặng thấp cao hơn khoảng chín phần mười, sinh non cao hơn khoảng tám phần mười. @@ -359,8 +335,7 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: B - Nguồn: Lei J, Li Y, Wang Y, v.v. (2024). The impact of small food workshops management regulations on aflatoxin B1 in home-made peanut oil and the liver function of high-consumption area residents: an interrupted time series study in Guangzhou, China. Frontiers in Public Health, 12. ；Zhong Y, Lu H, Jiang Y, v.v. (2024). Effect of homemade peanut oil consumption during pregnancy on low birth weight and preterm birth outcomes: a cohort study in Southwestern China. Global Health Action, 17. ；IARC (2012). Chemical Agents and Related Occupations. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol 100F. ；Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委), Cục Quản lý Dược phẩm và Thực phẩm quốc gia (国家食品药品监督管理总局) (2017). Tiêu chuẩn quốc gia về an toàn thực phẩm, Giới hạn độc tố nấm men trong thực phẩm (GB 2761-2017) - Ghi chú: Cả hai nghiên cứu chỉ ghi nhận và theo dõi, không có nhóm đối chứng. Kết quả cuối của nghiên cứu ở Quảng Châu là xét nghiệm chức năng gan, không phải ung thư gan và không phải tử vong. Trung vị độc tố trong dầu tự ép thực ra thấp hơn giới hạn 20 μg/kg mà tiêu chuẩn quốc gia đặt cho dầu điều, rủi ro đến từ việc ăn nhiều năm và từ những lô đơn lẻ cao bất thường. Giới hạn của tiêu chuẩn quốc gia áp dụng cho mọi loại dầu điều, nhưng dầu đóng gói phải qua giấy phép sản xuất và kiểm nghiệm trước khi xuất xưởng, còn xưởng nhỏ thì theo biện pháp quản lý của từng tỉnh và được lấy mẫu kiểm tra ít hơn nhiều. Người hưởng lợi từ mục này chủ yếu là bạn và người ăn cùng bạn. - -### 42. Nấu ăn bằng dầu thực vật thay cho mỡ lợn và bơ, không cần đổi dầu qua lại vì lý do sức khoẻ, cũng đừng trông chờ dầu hạt lanh bảo vệ tim +### 41. Nấu ăn bằng dầu thực vật thay cho mỡ lợn và bơ, không cần đổi dầu qua lại vì lý do sức khoẻ, cũng đừng trông chờ dầu hạt lanh bảo vệ tim - Chi phí: Không tốn thêm tiền. Dầu thực vật vốn đã là lựa chọn chính của đa số hộ gia đình. Chỗ khó là bỏ thói quen dùng mỡ lợn để xào và rưới cơm. - Nói thẳng: Ăn ít mỡ lợn, bơ và các loại mỡ bão hòa tương tự, đổi sang dầu thực vật thì các sự kiện như bệnh tim và đột quỵ giảm khoảng mười bảy phần trăm, nhưng rủi ro tử vong toàn nguyên không thấy thay đổi. Chuyển qua lại giữa dầu đậu nành, dầu ngô, dầu cải thì không có bằng chứng nào cho là khác nhau. Ăn nhiều dầu hạt lanh thì rủi ro bệnh tim cũng không thấy giảm. @@ -368,3 +343,11 @@ Phần này chỉ nhận hai loại rủi ro mạn tính: loại làm tỷ lệ - Mức bằng chứng: A - Nguồn: Hooper L, Martin N, Jimoh OF, v.v. (2020). Reduction in saturated fat for cardiovascular disease. Cochrane Database of Systematic Reviews, (5), CD011737. ；Hooper L, Al-Khudairy L, Abdelhamid AS, v.v. (2018). Omega-6 fats for the primary and secondary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews, (11), CD011094. ；Abdelhamid AS, Brown TJ, Brainard JS, v.v. (2020). Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease. Cochrane Database of Systematic Reviews, (3), CD003177. ；Hiệp hội Dinh dưỡng Trung Quốc (中国营养学会) (2022). Hướng dẫn khẩu phần cư dân Trung Quốc (2022). Nhà xuất bản Nhân dân Y tế - Ghi chú:Tranh cãi: có người cho rằng lượng omega-6 trong dầu đậu nành và dầu ngô quá nhiều nên gây viêm, do đó phải ăn ít và đổi dầu. Các thử nghiệm chia nhóm ngẫu nhiên trong bài omega-6 nói trên không thấy hại, tác giả tự chấm chất lượng bằng chứng là thấp. Mức lợi ích đặt là vừa dựa trên sự kiện tim mạch giảm 17%, còn tỷ lệ tử vong toàn nguyên thì không đổi. Dầu nào cũng phải kiểm soát lượng dùng, hướng dẫn khẩu phần khuyên mỗi người mỗi ngày dùng 25 đến 30 g dầu nấu ăn. +### 42. Khi xào và khi chiên rán thì bật máy hút mùi, từ lúc bật bếp cho tới khi nấu xong + +- Chi phí: Nhà đã có máy hút mùi thì chỉ thêm chút tiền điện. Nhà chưa có thì mua một máy, tính cả lắp đặt, mấy trăm đến hơn một nghìn tệ. Chỗ khó là mỗi lần bật bếp đều nhớ bật, đừng vì thấy ồn mà tắt. +- Nói thẳng: Phụ nữ Trung Quốc không hút thuốc, khi xào mà không bật hoặc rất ít bật máy hút mùi thì rủi ro mắc ung thư phổi cao khoảng hai lần người luôn bật. Người đã sống trong nhà có bếp thông gió kém thì rủi ro cũng cao hơn khoảng một nửa, nhà đốt than là rõ nhất. +- Lợi ích: Gộp 10 nghiên cứu hồi cứu bệnh nhân lại với nhau, đối tượng đều là phụ nữ Trung Quốc không hút thuốc. Nghiên cứu hồi cứu bệnh nhân là so bệnh nhân ung thư phổi với người không bệnh, rồi hỏi lại thói quen trong quá khứ. Người tiếp xúc với khói mỡ nấu ăn có cơ hội mắc ung thư phổi cao khoảng 1.74 lần (tỷ số khả năng, 95% CI 1.57–1.94, khoảng tin cậy). Trong đó có 4 nghiên cứu so người tiếp xúc khói mỡ mà không dùng máy hút mùi, cao khoảng 2.11 lần (1.54–2.89). Một nghiên cứu ở Đài Loan so 1302 nữ bệnh nhân ung thư phổi không hút thuốc với 1302 người không bệnh. Chỉ xem nhóm nấu ăn trên mười năm. Người có hơn hai phần ba số năm trong thời gian đó đều bật máy hút mùi, so với nhóm dưới một phần ba, có rủi ro thấp hơn khoảng một nửa (tỷ số khả năng 0.49, 0.32–0.76). Một nghiên cứu theo dõi về phía trước ở Thượng Hải gồm 7.1 vạn phụ nữ không hút thuốc, theo dõi từ 1996 đến 2009, phát hiện 429 ca ung thư phổi. Người tự nói đã sống trong nhà có bếp thông gió kém có rủi ro mắc ung thư phổi cao hơn khoảng 49% (HR 1.49, 1.15–1.95). Người đốt than mà bếp lại thông gió kém trên 20 năm thì khoảng 2.03 lần (HR 2.03, 1.35–3.05). Tính riêng yếu tố đã từng đốt than hay không thì không thấy khác biệt (HR 1.03, 0.84–1.26). Xào bằng loại dầu nào thì cũng không thấy khác biệt. +- Mức bằng chứng: B +- Nguồn: Xue Y, Jiang Y, Jin S, Li Y (2016). Association between cooking oil fume exposure and lung cancer among Chinese nonsmoking women: a meta-analysis. OncoTargets and Therapy, 9, 2987-2992. ; Chen TY, Fang YH, Chen HL và cộng sự (2020). Impact of cooking oil fume exposure and fume extractor use on lung cancer risk in non-smoking Han Chinese women. Scientific Reports, 10, 6774. ; Kim C, Gao YT, Xiang YB và cộng sự (2015). Home kitchen ventilation, cooking fuels, and lung cancer risk in a prospective cohort of never smoking women in Shanghai, China. International Journal of Cancer, 136(3), 632-638. +- Ghi chú: Tranh cãi: những con số ủng hộ việc bật máy hút mùi chủ yếu đến từ nghiên cứu hồi cứu bệnh nhân. Người bệnh nhớ lại quá khứ sau khi đã chẩn đoán, dễ nói ít số lần bật máy hút mùi, nên con số có thể lớn hơn thực. Trong nghiên cứu theo dõi về phía trước ở Thượng Hải, mức tăng rủi ro tập trung chủ yếu ở nhà đốt than mà thông gió kém. Riêng nhà chỉ đốt khí đốt thì bật hay không bật khác nhau bao nhiêu, nghiên cứu đó không đưa ra được con số riêng. Vậy mức bằng chứng chỉ để B, cỡ lợi ích đặt là lớn theo khoảng chênh lệch rủi ro gấp đôi. Nấu ăn mà không đốt than củi xem mục 29 của phần này (nấu ăn và sưởi ấm không đốt than củi). Người hưởng lợi chủ yếu là người thường nấu ăn trong nhà. diff --git "a/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" "b/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" index 6d41ad06..7c268728 100644 --- "a/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" +++ "b/vi/book/03-\344\270\215\350\246\201\346\265\252\350\264\271\347\262\276\345\212\233.md" @@ -2,7 +2,7 @@ # 3. Đừng phí sức -Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng chú ý bạn dùng được mỗi ngày, phản xạ nhanh bao nhiêu, sai sót bao nhiêu, và sau khi bị cắt ngang thì bao lâu mới quay lại được. Mọi con số đều đo trực tiếp, từ thí nghiệm, hoặc từ hồ sơ theo dõi một nhóm người (nghiên cứu mạch). Những số này không quy đổi ra tuổi thọ, cũng không quy đổi ra tỷ lệ tử vong. Phần lớn các mục trong phần này xếp B. Lĩnh vực này vốn vậy, không có nghĩa lời khuyên không đúng. Mục 20 nói về điều nên mong đợi khi làm việc với cảnh sát, bác sĩ, thủ quầy: nó dẫn quy định chứ không phải thí nghiệm, nên xếp C. Mục 15 (coi ý nghĩ「chuyện đó chắc chắn sẽ tệ hơn」là một triệu chứng) và mục 23 (coi ý nghĩ「người khác đòi mình phải hoàn hảo」là một triệu chứng) dẫn số về tỷ lệ tử vong và nguy cơ tự sát, khác đơn vị đo với các mục khác trong phần này, nên không so với nhau. +Phần này chỉ tính sức lực và thời gian. Có bốn thứ được xem: lượng chú ý bạn dùng được mỗi ngày, phản xạ nhanh bao nhiêu, sai sót ra bao nhiêu, và sau khi bị cắt ngang thì bao lâu mới quay lại được. Mọi con số đều đo trực tiếp, đến từ thử nghiệm, hoặc từ bản ghi theo dõi một nhóm người (nghiên cứu theo dõi). Những con số này không quy đổi ra tuổi thọ, cũng không quy đổi ra tỷ lệ tử vong. Phần lớn các mục trong phần này có mức bằng chứng là B. Lĩnh vực này vốn vậy, không có nghĩa lời khuyên không đúng. Mục 19 nói về thái độ nên có khi làm việc với cảnh sát, bác sĩ và nhân viên giao dịch, nó dẫn quy định pháp luật chứ không phải thử nghiệm, nên ghi mức C. Mục 15 (coi ý nghĩ bi quan là triệu chứng) và mục 22 (coi ý nghĩ「người khác đòi ta phải hoàn hảo」là triệu chứng) dẫn số về tỷ lệ tử vong và rủi ro tự sát, thước đo khác với các mục còn lại trong phần này, không so với nhau. ### 1. Tắt các thông báo không cần thiết, và để điện thoại ngoài tầm nhìn khi làm việc @@ -12,24 +12,23 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: B - Nguồn: Stothart, Mitchum & Yehnert (2015). The attentional cost of receiving a cell phone notification. Journal of Experimental Psychology: Human Perception and Performance. ; Ward, Duke, Gneezy & Bos (2017). Brain Drain: The Mere Presence of One's Own Smartphone Reduces Available Cognitive Capacity. Journal of the Association for Consumer Research. - Ghi chú: Đều là những tác vụ đơn lẻ trong phòng thí nghiệm. Cộng dồn mỗi ngày thì ra sao, chưa ai đo. Với công việc thật sự phải có mặt bất cứ lúc nào, chỉ giữ thông báo của vài đồng nghiệp là được. - -### 2. Giờ thức dậy cố định, cuối tuần cũng vậy +### 2. Giữ giờ thức dậy cố định, cuối tuần cũng vậy -- Chi phí: Không tốn tiền. Cuối tuần phải bỏ thói ngủ bù, cần một chút ý chí. -- Nói thẳng: Nhóm một phần năm có giờ sinh hoạt loạn nhất, đồng hồ sinh học lệch muộn hơn khoảng 2.5 giờ so với nhóm một phần năm đều đặn nhất. Nghĩa là sống lâu trong tình trạng lệch múi giờ. Học sinh nào giờ sinh hoạt càng đều thì kết quả học tập càng tốt. -- Lợi ích: 61 sinh viên đại học ghi nhận trong 30 ngày. Nhóm một phần năm có giờ sinh hoạt ít đều đặn nhất, đồng hồ sinh học (nhịp ngày đêm) lệch muộn hơn khoảng 2.5 giờ so với nhóm một phần năm đều đặn nhất. Cụ thể là giờ bắt đầu tiết melatonin là 00:08 so với 21:32. Khung giờ mệt nhất, muốn ngủ nhất cũng lệch muộn khoảng 1.8 giờ. Giờ sinh hoạt càng đều thì kết quả học tập càng tốt (hệ số tương quan r = 0.37). +- Chi phí: Không tốn tiền. Cuối tuần phải bỏ thói ngủ nướng, cần một chút ý chí. +- Nói thẳng: Nhóm một phần năm có giờ sinh hoạt lộn xộn nhất, đồng hồ sinh học muộn hơn nhóm một phần năm đều đặn nhất khoảng 2.5 giờ. Nghĩa là sống lâu dài trong chênh lệch múi giờ. Học sinh nào sinh hoạt càng đều thì thành tích càng tốt. +- Lợi ích: 61 sinh viên đại học được ghi nhận trong 30 ngày. Nhóm một phần năm có giờ sinh hoạt lộn xộn nhất, đồng hồ sinh học (nhịp ngày đêm) muộn hơn nhóm một phần năm đều đặn nhất khoảng 2.5 giờ. Cụ thể là giờ bắt đầu tiết melatonin là 00:08 so với 21:32. Khoảng thời gian mệt nhất, muốn ngủ nhất cũng muộn hơn khoảng 1.8 giờ. Sinh hoạt càng đều thì thành tích học tập càng tốt (hệ số tương quan r = 0.37). - Mức bằng chứng: B - Nguồn: Phillips et al. (2017). Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sleep/wake timing. Scientific Reports. -- Ghi chú: Chỉ có một nghiên cứu, số người ít, cũng không có nhóm đối chứng. Nhà nghiên cứu suy ra rằng khác biệt về đồng hồ sinh học chủ yếu đến từ lượng ánh sáng mỗi ngày khác nhau. Vì vậy giờ thức dậy cố định phải đi kèm việc tiếp xúc với ánh sáng buổi sáng mới có tác dụng. Thỉnh thoảng thức khuya rồi muốn ngủ bù, xem mục 39 ở phần 2 (ngủ bù vào tối hôm sau đêm thức khuya): cách làm là lên giường sớm hôm đó, không phải ngủ đến trưa hôm sau. +- Ghi chú: Chỉ có nghiên cứu này, số người ít, và cũng không có nhóm đối chứng chia nhau. Các nhà nghiên cứu suy ra rằng khác biệt về đồng hồ sinh học chủ yếu đến từ lượng ánh sáng mỗi ngày khác nhau. Vì vậy giữ giờ thức dậy cố định phải kèm việc tiếp xúc với ánh sáng buổi sáng thì mới có tác dụng. Thỉnh thoảng thức khuya xong bù lại thế nào, xem mục 38 ở phần 2 (thức khuya xong thì đêm hôm sau ngủ bù): cách làm là lên giường sớm vào tối hôm đó, không phải hôm sau ngủ tới trưa. -### 3. Ngủ đủ 7 đến 8 giờ mỗi đêm, đừng coi 6 giờ là đủ +### 3. Mỗi đêm ngủ đủ 7 đến 8 giờ, đừng coi 6 giờ là đủ -- Chi phí: Không tốn tiền. Mỗi ngày phải dành thêm 1 đến 2 giờ cho giấc ngủ. Lên giường sớm hơn thì cần một chút ý chí. -- Nói thẳng: Hai tuần liền mỗi đêm chỉ ngủ 6 giờ, hiệu năng não rơi xuống mức tương đương với hai ngày hai đêm không ngủ. Rắc rối là người ta không cảm nhận ra điều đó. Điểm buồn ngủ tự chấm tăng trong vài ngày đầu rồi đứng yên, người ngủ 6 giờ và người ngủ 4 giờ chấm điểm buồn ngủ bằng nhau. Vì vậy câu「tôi ngủ 6 giờ là đủ」bản thân nó là một triệu chứng thiếu ngủ. -- Lợi ích: 48 người lớn khỏe mạnh từ 21 đến 38 tuổi ngủ mỗi đêm 6 giờ trong 14 ngày liên tiếp. Hiệu năng nhận thức của họ giảm dần, rơi xuống mức tương đương với 2 đêm không ngủ hoàn toàn. Mức buồn ngủ tự chấm chỉ tăng trong vài ngày đầu, sau đó hầu như không tăng nữa. Nhóm ngủ 6 giờ và nhóm ngủ 4 giờ không khác nhau về buồn ngủ tự đánh giá. Đây là thử nghiệm đối chứng chia ngẫu nhiên người tham gia vào hai nhóm (RCT). Ngoài ra, thiếu ngủ trong thời gian ngắn (cắt giấc ngủ cấp tính, không quá 48 giờ) có cỡ tác động lên sự mất tập trung ở nhiệm vụ chú ý đơn giản là g = -0.776, và lên độ chính xác suy luận là g = -0.125. Phần này gộp 70 nghiên cứu lại. +- Chi phí: Không tốn tiền. Mỗi ngày phải nhường thêm 1 đến 2 giờ cho giấc ngủ. Lên giường sớm hơn thì cần một chút ý chí. +- Nói thẳng: Ngủ liên tục hai tuần chỉ 6 giờ mỗi đêm, khả năng của bộ não rơi xuống mức tương đương với hai ngày hai đêm không ngủ. Rắc rối là bản thân bạn không cảm nhận ra điều đó. Mức buồn ngủ bạn tự chấm chỉ tăng trong vài ngày đầu rồi đứng yên, người ngủ 6 giờ và người ngủ 4 giờ chấm điểm buồn ngủ như nhau. Vậy câu「6 giờ là đủ cho tôi」bản thân nó đã là triệu chứng thiếu ngủ. +- Lợi ích: 48 người lớn khoẻ mạnh từ 21 đến 38 tuổi, liên tục 14 ngày chỉ ngủ 6 giờ mỗi đêm. Khả năng nhận thức của họ giảm dần theo lũy kế, rơi xuống mức tương đương với 2 đêm không ngủ hoàn toàn. Còn mức buồn ngủ họ tự chấm chỉ tăng trong vài ngày đầu, sau đó hầu như không tăng nữa. Nhóm ngủ 6 giờ và nhóm ngủ 4 giờ không khác nhau về buồn ngủ tự đánh giá. Đây là thử nghiệm đối chứng chia người ngẫu nhiên thành hai nhóm (RCT). Ngoài ra, việc không ngủ trong thời gian ngắn (mất ngủ cấp tính, không quá 48 giờ) có cỡ hiệu ứng lên sự mất tập trung đơn giản là g = -0.776, và lên độ chính xác suy luận là g = -0.125. Phần này gộp 70 nghiên cứu lại. - Mức bằng chứng: A - Nguồn: Van Dongen, Maislin, Mullington & Dinges (2003). The cumulative cost of additional wakefulness. Sleep. ; Lim & Dinges (2010). A meta-analysis of the impact of short-term sleep deprivation on cognitive variables. Psychological Bulletin. -- Ghi chú: Cảm giác「không buồn ngủ」không phải bằng chứng cho thấy đã ngủ đủ. Người ngủ 6 giờ lâu dài chỉ quen về cảm giác, hiệu năng thì không theo kịp. Phần gộp nhiều nghiên cứu nói về việc không ngủ suốt đêm. 「Ngủ ít lâu dài」chỉ có một thử nghiệm đối chứng ngẫu nhiên hỗ trợ. +- Ghi chú: 「Cảm thấy không buồn ngủ」 không phải bằng chứng cho thấy đã ngủ đủ. Người ngủ 6 giờ lâu dài chỉ quen về cảm xúc, khả năng thì không theo kịp. Phần gộp nhiều nghiên cứu nói về việc không ngủ cả đêm. 「Ngủ ít lâu dài」 chỉ có một thử nghiệm đối chứng ngẫu nhiên đứng sau. Quan hệ giữa ngủ bao lâu và rủi ro tử vong tính riêng, xem mục 13 ở phần 2 (mỗi đêm ngủ khoảng 7 giờ): bên đó tính theo tỷ lệ tử vong, mốc 7 giờ có rủi ro thấp nhất. ### 4. Sau hai giờ chiều không dùng caffeine @@ -39,7 +38,6 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: A - Nguồn: Drake, Roehrs, Shambroom & Roth (2013). Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed. Journal of Clinical Sleep Medicine. ; Gardiner et al. (2023). The effect of caffeine on subsequent sleep: A systematic review and meta-analysis. Sleep Medicine Reviews. - Ghi chú: Nghiên cứu của Drake chỉ do một phòng thí nghiệm thực hiện, dùng 400 mg thì hơi lớn, khoảng hai ba ly cà phê. Người chuyển hóa caffeine chậm thì thời điểm phải dừng còn sớm hơn. Bớt caffeine đổi lại được thời lượng ngủ, và cộng thêm với mục 3 của phần này (ngủ đủ 7 đến 8 giờ mỗi đêm). - ### 5. Chuyển mail và tin nhắn sang xử lý theo lô, vài lần cố định mỗi ngày - Chi phí: Không tốn tiền. Mất chút thời gian nói rõ với đồng nghiệp rằng bạn trả lời bao lâu một lần. Kìm không mở mail suốt ngày thì cần một chút ý chí. @@ -48,7 +46,6 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: B - Nguồn: Kushlev & Dunn (2015). Checking email less frequently reduces stress. Computers in Human Behavior. - Ghi chú: Chỉ có một nghiên cứu, cỡ tác động vừa đến nhỏ, đo là áp lực chứ không phải lượng việc tạo ra. Tin nhắn tức thì chưa có nghiên cứu trực tiếp nào, ở đây chỉ suy ra theo. - ### 6. Khi làm việc cần suy nghĩ liên tục, chặn cả những lần gián đoạn chỉ vài giây - Chi phí: Không tốn tiền. Đóng cửa, đeo tai nghe, hoặc treo một biển「xin không làm phiền」. @@ -57,7 +54,6 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: B - Nguồn: Altmann, Trafton & Hambrick (2014). Momentary interruptions can derail the train of thought. Journal of Experimental Psychology: General. ; Mark, Gonzalez & Harris (2005). No task left behind? Examining the nature of fragmented work. CHI 2005. ; Mark, Gudith & Klocke (2008). The cost of interrupted work: More speed and stress. CHI 2008. - Ghi chú: Nghiên cứu tại hiện trường đó chia gián đoạn làm hai loại: một loại do người khác tìm đến bạn, một loại do chính bạn bấm vào. Đóng cửa chỉ chặn được loại trước. Loại sau phải dùng kèm mục 1 (tắt thông báo không cần thiết) và mục 5 (xử lý tin nhắn theo lô). - ### 7. Mỗi lúc làm một việc, đừng vừa họp vừa trả lời tin nhắn - Chi phí: Không tốn tiền. Kìm không trả lời tin nhắn trong cuộc họp thì cần một chút ý chí. @@ -75,7 +71,6 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: B - Nguồn: Chang, Aeschbach, Duffy & Czeisler (2015). Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. PNAS. - Ghi chú: Số người ít, thí nghiệm làm trong phòng thí nghiệm với điều kiện được kiểm soát. Với độ sáng và thời lượng sử dụng điện thoại như ngoài đời, ảnh hưởng có thể không lớn đến vậy. Chế độ ban đêm của điện thoại có ích bao nhiêu, nghiên cứu này không đo. - ### 9. Đến giờ là ngủ, không thức vì game, video ngắn hay nội dung khiêu dâm - Chi phí: Không tốn tiền. Khó ở chỗ chống lại quán tính「xem nốt tập nữa」suốt một thời gian dài. Để điện thoại ngoài phòng ngủ thì nhàn nhất. @@ -92,14 +87,15 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: B - Nguồn: Killgore, Balkin & Wesensten (2006). Impaired decision making following 49 h of sleep deprivation. Journal of Sleep Research. - Ghi chú: Nghiên cứu xét tình huống cực đoan là hai đêm liên tục không ngủ. Người bình thường thức đến hai giờ sáng sẽ ra sao, chưa ai đo trực tiếp. Nửa「không gửi tin nhắn」là kinh nghiệm của tác giả (mức C), gửi vào ngày hôm sau cũng chưa muộn. -### 11. Buồn ngủ buổi chiều thì ngủ 10 phút, đừng ngủ nửa tiếng +### 11. Buổi chiều buồn ngủ thì ngủ 10 phút, đừng ngủ nửa giờ -- Chi phí: Không tốn tiền. Mất 10 phút, thêm việc tìm chỗ nằm xuống hay úp mặt. -- Nói thẳng: Một giấc 10 phút tỉnh ngay tại chỗ, hiệu lực giữ được hai tiếng rưỡi. Ngủ 5 phút gần như vô ích. Ngủ 20 phút phải qua nửa tiếng mới có tác dụng. Ngủ 30 phút thì lúc tỉnh trước mắt có một đoạn mê man: giai đoạn đó còn làm kém hơn cả việc không ngủ. -- Lợi ích: Sau một đêm ngủ không đủ, giấc 10 phút cải thiện ngay buồn ngủ, mệt mỏi, sức lực và hiệu suất nhận thức, một phần lợi ích còn giữ qua 155 phút. Ngủ 5 phút hầu như không có lợi ích. Lợi ích của giấc 20 phút phải qua 35 phút mới xuất hiện. Ngủ 30 phút thì trước mắt có một đoạn lìa giấc: vừa tỉnh hiệu suất lại kém hơn, rồi mới khá lên. +- Chi phí: Không tốn tiền. Mất 10 phút, và tìm một chỗ nằm được hoặc úp xuống. +- Nói thẳng: Giấc ngủ ngắn 10 phút tỉnh ngay tại chỗ, tác dụng kéo dài hai tiếng rưỡi. Ngủ 5 phút gần như vô ích. Ngủ 20 phút phải chờ nửa giờ sau mới thấy hiệu quả. Ngủ 30 phút thì lúc vừa mở mắt còn một đoạn ngơ ngác, khoảng đó biểu hiện còn kém hơn là không ngủ. +- Lợi ích: Sau một đêm ngủ thiếu, giấc ngủ ngắn 10 phút cải thiện ngay buồn ngủ, mệt mỏi, sức lực và khả năng nhận thức; một phần lợi ích còn giữ tới sau 155 phút. Ngủ 5 phút gần như không có lợi ích. Lợi ích của giấc ngủ ngắn 20 phút chỉ xuất hiện sau 35 phút. Sau giấc ngủ ngắn 30 phút có trước một đoạn quánh ngủ: ngay sau khi tỉnh biểu hiện còn tệ hơn, rồi mới cải thiện. - Mức bằng chứng: B - Nguồn: Brooks & Lack (2006). A brief afternoon nap following nocturnal sleep restriction: Which nap duration is most recuperative? Sleep. -- Ghi chú: Chỉ có một nghiên cứu thí nghiệm trong phòng lab, người tham dự đều là người trẻ khỏe mạnh. Không ngủ được cũng đừng cố. Nằm nhắm mắt 10 phút có tác dụng hay không, không có nghiên cứu nào chống đỡ. Ngủ trưa quá một giờ lại đi cùng tỷ lệ tử vong và nguy cơ bệnh vành cao hơn, xem mục 38 ở phần 2 (giữ giấc trưa trong nửa tiếng). +- Ghi chú: Chỉ có một nghiên cứu trong phòng thí nghiệm, người tham gia đều là thanh niên khoẻ mạnh. Không ngủ được thì không cần cố. Nằm 10 phút với mắt nhắm có ích hay không, không có nghiên cứu nào đỡ. Ngủ trưa quá một giờ lại gắn với tỷ lệ tử vong và nguy cơ bệnh mạch vành cao hơn, xem mục 37 ở phần 2 (giữ giấc ngủ trưa trong nửa giờ). + ### 12. Làm việc ở văn phòng mở thì dùng nút bịt tai hoặc tìm phòng yên tĩnh cho việc cần ghi nhớ - Chi phí: Nút bịt tai vài chục tệ. Đặt trước một phòng họp yên tĩnh thì tốn chút thời gian. @@ -132,6 +128,7 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: B - Nguồn: Whitfield JB, Zhu G, Landers JG, Martin NG (2020). Pessimism is associated with greater all-cause and cardiovascular mortality, but optimism is not protective. Scientific Reports, 10, 12609. - Ghi chú:Tranh cãi. Đây chỉ là ghi nhận, không chia nhóm. Người ta sẽ nói「thật ra ốm trước nên mới bi quan」, tác giả phản hồi rằng nếu thật vậy thì tử vong ung thư cũng phải liên quan. Nhưng vẫn có thể còn yếu tố chưa đo. Lạc quan không có tác dụng bảo vệ, nên không cần ép mình「suy nghĩ tích cực」. Việc cần làm là đừng xem dự đoán bi quan là bản dự báo chính xác cho tương lai. + ### 16. Giảm bớt các mối quan hệ làm bạn hao mòn, học cách từ chối yêu cầu không muốn nhận - Chi phí: Không tốn tiền. Ngắn hạn sẽ có ma sát quan hệ, có thể mất đi một phần các mối quan hệ. Nói lời từ chối cần khá nhiều ý chí. @@ -140,16 +137,7 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: C - Nguồn: Rook (1984). The negative side of social interaction: Impact on psychological well-being. Journal of Personality and Social Psychology. - Ghi chú: Nhóm được hỏi khá đặc biệt, lại chỉ là bảng hỏi một lần, không thấy trước sau. Thứ đo là hạnh phúc, không phải sức lực. Đặt vào phần này là suy luận của tác giả. Mối quan hệ nào bị coi là「hao mòn」thì không có tiêu chuẩn khách quan, chỉ tự mình phán đoán. - -### 17. Chốt sẵn những quyết định giá trị thấp như mặc gì, ăn gì - -- Chi phí: Không tốn tiền. Lúc đầu tốn chút thời gian để định sẵn lựa chọn mặc định. -- Nói thẳng: Mục này không có bằng chứng chắc. Thí nghiệm gốc về「ra quyết định làm cạn ý chí」không tái hiện được trong hai lần làm lại quy mô lớn. Hai lần đó một lần hai nghìn người, một lần ba nghìn người, nhiều phòng thí nghiệm cùng làm, hiệu ứng cơ bản bằng không. Chốt sẵn mặc gì ăn gì vẫn đáng làm, vì nó tiết kiệm vài phút và vài lần do dự, chứ không phải「tiết kiệm ý chí」. -- Lợi ích: Không lượng hóa được. Thí nghiệm gốc cho rằng ra quyết định và tự kiểm soát cùng dùng chung một nguồn hạn chế (sự hao mòn bản ngã). Nhưng sau đó có hai lần làm lại quy mô lớn, cả hai đều đăng ký trước kế hoạch, nhiều phòng thí nghiệm làm lại theo đúng bản gốc, cả hai đều không ra cùng kết quả. Lần một là 23 phòng thí nghiệm, 2141 người, cỡ hiệu ứng d = 0.04, 95% CI [-0.07, 0.15] (khoảng tin cậy). Lần khác là 36 phòng thí nghiệm, 3531 người, d = 0.06. Chênh lệch nhỏ đến mức không tính được. -- Mức bằng chứng: C -- Nguồn: Baumeister, Bratslavsky, Muraven & Tice (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology. ; Hagger et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science. ; Vohs et al. (2021). A Multisite Preregistered Paradigmatic Test of the Ego-Depletion Effect. Psychological Science. -- Ghi chú:Tranh cãi.「Mệt vì quyết định」trong sách phổ thông được coi là kết luận chung, nhưng sau đó người khác làm lại theo đúng thí nghiệm gốc, phần lớn không ra cùng kết quả. Nó vẫn đáng làm vì gần như không tốn gì, còn thời gian tiết kiệm được là thật. Đừng mong nó cải thiện khả năng phán đoán của bạn vào buổi chiều. -### 18. Khi giận thì đi khỏi chỗ đó trước, coi đối phương như thời tiết chứ không như kẻ thù +### 17. Khi giận thì đi khỏi chỗ đó trước, coi đối phương như thời tiết chứ không như kẻ thù - Chi phí: Không tốn tiền. Lúc đó phải nhịn không cãi trả, quay lưng đi, rồi sau mới quyết định có xử lý hay không. Cần tập một chút. - Nói thẳng: Giải thích chuyện vừa xảy ra theo góc khác, hoặc đứng từ góc đối phương mà nghĩ, cả hai đều hạ cảm xúc thật được, cách sau hiệu quả hơn. Nén lại không lộ ra thì không có tác dụng. Người trừng phạt người gây sự thì sau đó nghĩ đi nghĩ lại về chuyện đó lâu hơn; người không trả thù thì quên mau hơn. Người ta ước tính trước cao cái sướng mà trả thù mang lại. @@ -157,32 +145,35 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: B - Nguồn: Webb TL, Miles E, Sheeran P (2012). Dealing with feeling: A meta-analysis of the effectiveness of strategies derived from the process model of emotion regulation. Psychological Bulletin. ; Carlsmith KM, Wilson TD, Gilbert DT (2008). The paradoxical consequences of revenge. Journal of Personality and Social Psychology. - Ghi chú: Phần gộp nhiều nghiên cứu đo là cường độ cảm xúc, không phải sức lực, hiệu ứng từ nhỏ đến trung bình. Phần trả thù chỉ có một nhóm thí nghiệm, nên chấm B.「Con thuyền trống」là chuyện ngụ ngôn trong chương「Sơn Mộc」của Trang Tử: thuyền trống đâm tới thì không giận, trên thuyền có người thì lại giận, nhưng va chạm là một. Đó là cách so sánh, không phải bằng chứng. Nếu thật có tổn thất, hãy đi lối pháp lý ở phần 8 và phần 9, rẻ hơn nhiều so với bùng phát ngay tại chỗ. -### 19. Khi tâm trạng xuống, làm trước mấy việc đáng giá nhất: vận động, phơi nắng, ngủ đúng giờ, nói với ai đó, gọi 12356 +### 18. Khi tâm trạng tụt, làm trước mấy việc hiệu quả chi phí nhất: vận động, phơi nắng, ngủ đúng giờ, nói với ai đó, gọi 12356 -- Chi phí: Không tốn tiền, hoặc tốn rất ít. Đi bộ nhanh, chạy bộ hay tập tạ vài lần mỗi tuần. Ra ngoài ban ngày để đón ánh sáng. Chốt giờ sinh hoạt. Khi không lội ra được thì gọi một cuộc điện thoại. -- Nói thẳng: Với trầm cảm, đi bộ nhanh và chạy bộ hiệu quả nhất, yoga và tập tạ kém hơn một chút, càng mạnh càng hiệu quả. Chiếu sáng hiệu quả rõ với trầm cảm theo mùa, và cũng hiệu quả với trầm cảm không theo mùa. Khóa tự giúp nhận thức hành vi trên mạng thì có tác dụng, lúc triệu chứng nặng hơn thì có người dẫn dắt sẽ tốt hơn. Khi không lội ra được, gọi 12356 trước. -- Lợi ích: Gộp 218 thử nghiệm đối chứng chia nhóm ngẫu nhiên, 14170 người (phân tích mạng). Nhóm đối chứng là những người cũng đang làm hoạt động khác. So với họ, đi bộ hoặc chạy bộ có cỡ hiệu ứng lên trầm cảm g = -0.62 (95% khoảng tin cậy -0.80 đến -0.45). Yoga là -0.55, tập tạ là -0.49. Càng mạnh càng hiệu quả. Gộp các nghiên cứu chiếu sáng: cỡ hiệu ứng 0.84 với trầm cảm theo mùa (0.60 đến 1.08), 0.53 với trầm cảm không theo mùa (0.18 đến 0.89). Còn một phân tích khác gộp dữ liệu từng bệnh nhân của các nghiên cứu. Nó xét liệu pháp nhận thức hành vi qua internet (khóa tự giúp trên mạng). Có người dẫn dắt hay không dẫn dắt đều tốt hơn nhóm đối chứng. Lúc trung bình đến nặng (PHQ-9 lớn hơn 9 điểm), có người dẫn dắt thì tốt hơn. 12356 là đường dây trợ giúp tâm lý toàn quốc, mở mỗi ngày không ít hơn 18 giờ. +- Chi phí: Không tốn tiền, hoặc tốn rất ít. Mỗi tuần vài lần đi bộ nhanh, chạy bộ hay tập tạ. Ban ngày ra ngoài đón nắng. Chốt giờ sinh hoạt. Khi không nghĩ ra được thì gọi một cuộc điện thoại. +- Nói thẳng: Với bệnh trầm cảm, đi bộ nhanh và chạy bộ hiệu quả nhất, càng mạnh càng hiệu quả, chi tiết xem mục 6 ở phần 22 (tâm trạng kém thì đi bộ hoặc chạy). Chiếu sáng hiệu quả rõ với trầm cảm theo mùa, và cũng hiệu quả với trầm cảm không theo mùa. Khoá tự giúp nhận thức - hành vi trên mạng thì có tác dụng, lúc triệu chứng nặng hơn thì có người hướng dẫn sẽ tốt hơn. Khi không nghĩ ra được, gọi 12356 trước. +- Lợi ích: Gộp 218 thử nghiệm đối chứng chia nhóm ngẫu nhiên, 14170 người (phân tích tổng hợp mạng lưới). Nhóm đối chứng là những người đang làm hoạt động khác tương tự. So với họ, đi bộ hoặc chạy bộ có cỡ hiệu ứng lên trầm cảm g = -0.62 (95% khoảng tin cậy -0.80 đến -0.45). Yoga là -0.55, tập tạ là -0.49. Càng mạnh càng hiệu quả. Gộp các nghiên cứu chiếu sáng lại: cỡ hiệu ứng 0.84 với trầm cảm theo mùa (0.60 đến 1.08), 0.53 với trầm cảm không theo mùa (0.18 đến 0.89). Còn một phân tích khác gộp dữ liệu của từng bệnh nhân trong các nghiên cứu. Nó xét liệu pháp nhận thức - hành vi qua internet (khoá tự giúp trên mạng). Có người hướng dẫn và không có người hướng dẫn đều tốt hơn nhóm đối chứng. Ở mức vừa đến nặng (PHQ-9 lớn hơn 9 điểm), có người hướng dẫn thì tốt hơn. 12356 là đường dây nóng hỗ trợ tâm lý toàn quốc, mở mỗi ngày không ít hơn 18 giờ. - Mức bằng chứng: A -- Nguồn: Noetel M et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ. ; Golden RN et al. (2005). The efficacy of light therapy in the treatment of mood disorders: a review and meta-analysis of the evidence. American Journal of Psychiatry. ; Karyotaki E et al. (2021). Internet-Based Cognitive Behavioral Therapy for Depression: A Systematic Review and Individual Patient Data Network Meta-analysis. JAMA Psychiatry. ; Ủy ban Y tế Quốc gia (国家卫生健康委) (2024). Thông báo về việc dùng số điện thoại đường dây trợ giúp tâm lý thống nhất toàn quốc「12356」（国卫医政函〔2024〕259 号）. -- Ghi chú: Tất cả những thứ này đều là biện pháp hỗ trợ cho trầm cảm nhẹ đến trung bình. Trầm cảm nặng đến trung bình, hoặc có ý nghĩ tự sát thì đi khám bác sĩ, và gọi 12356 trước (xem mục 25 ở phần 1). Các thử nghiệm về vận động phần lớn số người ít, và không thể che giấu ai đang vận động như khi thử thuốc. Nghiên cứu chiếu sáng chỉ đến năm 2003, số người cũng ít。「Ngủ đúng giờ」「nói với ai đó」là kinh nghiệm của chính tác giả, không có tài liệu riêng. Đừng coi「vận động」là yêu cầu đặt lên người trầm cảm, đi được mười phút là được. -### 20. Coi cảnh sát, bác sĩ, nhân viên quầy là người làm việc theo quy tắc, đừng coi là vai diễn: thứ đẩy việc đi được là giấy tờ và thời hạn, không phải cảm xúc +- Nguồn: Noetel M et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ. ; Golden RN et al. (2005). The efficacy of light therapy in the treatment of mood disorders: a review and meta-analysis of the evidence. American Journal of Psychiatry. ; Karyotaki E et al. (2021). Internet-Based Cognitive Behavioral Therapy for Depression: A Systematic Review and Individual Patient Data Network Meta-analysis. JAMA Psychiatry. ; Ủy ban Y tế và Sức khoẻ quốc gia (国家卫生健康委) (2024). Thông báo về việc dùng số điện thoại của đường dây nóng hỗ trợ tâm lý thống nhất toàn quốc「12356」, 关于应用"12356"全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. +- Ghi chú: Tất cả những thứ này đều là biện pháp hỗ trợ cho trầm cảm nhẹ đến vừa. Trầm cảm vừa đến nặng, hoặc có ý nghĩ tự sát, thì phải đi khám bác sĩ, và gọi 12356 trước (xem mục 25 ở phần 1). Các thử nghiệm về vận động phần lớn số người ít, và không thể giấu ai đang vận động như khi thử thuốc. Nghiên cứu chiếu sáng chỉ kéo dài tới năm 2003, số người cũng ít. 「Ngủ đúng giờ」「nói với ai đó」là kinh nghiệm của tác giả, không có tài liệu riêng. Đừng coi「vận động」là yêu cầu đặt lên người trầm cảm, đi được mười phút là được. + +### 19. Xem cảnh sát, bác sĩ và nhân viên giao dịch là người đi làm theo quy định, đừng xem là một vai diễn: thứ đẩy việc tới đích là giấy tờ và thời hạn, không phải cảm xúc -- Chi phí: Không tốn tiền. Cái đổi là kỳ vọng của bạn và thứ tự hành động. Lần đầu phải nén mong muốn「nói cho đến khi đối phương công nhận」. -- Nói thẳng: Cảnh sát, bác sĩ, nhân viên quầy đều là người làm việc theo quy tắc của vị trí, chịu sự quản lý của giấy tờ, thời hạn và thanh tra. Đối phương có thân thiện hay không gần như không liên quan đến việc việc của bạn có được làm hay không, hay mấy ngày xong. Vậy hãy xin biên nhận, ghi lại thời hạn, khiếu nại theo đúng kênh, đừng mong giảng道理 cho đối phương công nhận. -- Lợi ích: Đổi cách xử lý chuyện với cơ quan: từ「van xin, nổi giận, kể lể đi kể lể lại」sang「xin giấy tờ, ghi thời hạn, khiếu nại theo đúng kênh」. Thứ tiết kiệm được là thời gian cho việc nói đi nói lại và nghĩ đi nghĩ lại. Đây là điều chỉnh quan niệm một lần, không phải tiết kiệm thời gian mỗi ngày. Cái thật sự mang ra kiểm tra được là chính các thời hạn đó, vì chúng đều viết trong văn bản. Vụ việc hành chính an ninh tính từ ngày tiếp nhận, thời hạn giải quyết không quá 30 ngày; vụ việc trọng đại, phức tạp thì được cấp công an cao hơn một cấp phê duyệt để gia hạn 30 ngày. Vụ việc có người tố cáo mà quyết định không khởi kiện thì thông báo không khởi kiện phải được tống đạt trong 3 ngày. Người tố cáo trong 7 ngày có thể xin hòa giải lại, cơ quan công an ra quyết định trong 30 ngày. Không phục thì trong 7 ngày nữa xin xem xét lại ở cấp cao hơn. +- Chi phí: Không tốn tiền. Cái đổi là kỳ vọng và thứ tự hành động của bạn. Lần đầu phải kìm được thôi thúc「nói lý lẽ cho tới khi đối phương đồng ý」. +- Nói thẳng: Cảnh sát, bác sĩ và nhân viên giao dịch đều là người đi làm theo quy định chức vụ, bị giấy tờ, thời hạn và sự giám sát ràng buộc. Thái độ của đối phương tốt hay không gần như không liên quan tới việc của bạn có được giải quyết không, và mất bao nhiêu ngày. Vậy nên xin biên nhận, ghi thời hạn, khiếu nại đúng kênh, đừng nghĩ tới việc nói lý lẽ cho tới khi đối phương đồng ý. +- Lợi ích: Đổi cách làm việc với các cơ quan: từ「xin người, nổi giận, kể lại nỗi ấm phiền hết lần này tới lần khác」sang「xin giấy tờ, ghi nhớ thời hạn, khiếu nại đúng kênh」. Cái tiết kiệm là thời gian trao đổi lại nhiều lần và nhớ lại nhiều lần. Đây là một lần điều chỉnh tư tưởng, không phải mỗi ngày lại tiết kiệm thời gian. Thứ dùng để đối chiếu là chính các thời hạn này, chúng đều ghi rõ trong hồ sơ. Vụ án trật tự tính từ ngày tiếp nhận, thời hạn giải quyết không quá 30 ngày; vụ án phức tạp nghiêm trọng thì được kéo dài thêm 30 ngày nếu được cơ quan công an cấp trên chấp thuận. Vụ có người khiếu nại mà quyết định không khởi tố thì thông báo không khởi tố phải được gửi trong 3 ngày. Người khiếu nại có thể yêu cầu xét lại trong 7 ngày, cơ quan công an phải ra quyết định trong 30 ngày. Không chấp nhận thì trong 7 ngày nữa xin thẩm tra lên cấp trên. - Mức bằng chứng: C -- Nguồn: Bộ Công an (公安部) (2020). Quy định về thủ tục giải quyết vụ án hình sự của cơ quan công an (公安机关办理刑事案件程序规定)（公安部令第 159 号修正后，第一百七十一、一百七十八、一百七十九条）. ; Bộ Công an (公安部) (2013/2019 sửa đổi). Quy định về thủ tục giải quyết vụ việc hành chính của cơ quan công an (公安机关办理行政案件程序规定)（公安部令第 125 号、第 149 号）. -- Ghi chú: Lý do chấm C: chính câu「nhiều người coi người sau quầy là một vai diễn, chứ không phải người đi làm」không có tài liệu nào dẫn được. Chỉ có nửa quy tắc dẫn được: thời hạn, giấy tờ và kênh khiếu nại đều có văn bản rõ. Toàn bộ con số trong mục này lấy từ hai quy định của Bộ Công an mà mục 39 ở phần 8 dẫn (quy trình sau khi gọi công an). Đi theo quy trình không xử lý được「đối phương thái độ kém」, cũng không bảo đảm chắc chắn làm được, cái đổi chỉ là kỳ vọng của bạn và bước đầu tiên. Nhân viên quầy ngân hàng, quầy dịch vụ công cũng vậy: thao tác của họ đều có dấu, cần người thứ hai kiểm, sai thì bị truy trách. Vậy việc nên làm là giữ phiếu, xin biên lai, khiếu nại theo đúng kênh, đừng phán bừa theo「hắn có lừa mình không」. Thứ tự ưu tiên hợp pháp đều ghi rõ ra chỗ sáng, ví dụ cấp cứu phân loại theo tình trạng bệnh, không theo đến trước đến sau (mục 8 ở phần 24, phân loại khám cấp cứu). Muốn cảm ơn bác sĩ đã cứu mình thì gửi thư cảm ơn và bảng đánh giá hài lòng, xem mục 12 ở phần 24. Muốn「đưa chút gì để người ta để tâm hơn」thì với người đang xử lý vụ việc đó là tội hối lộ, mà luật còn quy định nặng hơn, xem mục 40 ở phần 8. Gặp đối xử phân biệt rõ ràng thì khiếu nại, đừng coi là luật ngầm mà làm theo. +- Nguồn: Bộ Công an (公安部) (2020). Quy định về thủ tục xử lý vụ án hình sự của cơ quan công an（公安机关办理刑事案件程序规定）（公安部令第 159 号修正后，第一百七十一、一百七十八、一百七十九条）. ；Bộ Công an (公安部) (2013/2019, sửa đổi). Quy định về thủ tục xử lý vụ việc hành chính của cơ quan công an（公安机关办理行政案件程序规定）（公安部令第 125 号、第 149 号）. +- Ghi chú: Lý do ghi mức C: chính nhận định「nhiều người coi người sau quầy giao dịch là một vai diễn chứ không phải một người đi làm」không có tài liệu nào dẫn được. Chỉ có nửa quy định là dẫn được: thời hạn, giấy tờ và kênh khiếu nại đều ghi rõ bằng văn bản. Toàn bộ số mà mục này dẫn đều đến từ hai quy chế của Bộ Công an ở mục 39 ở phần 8 (quy trình sau khi báo công an). Đi theo quy trình không quản được việc「thái độ kém của đối phương」, cũng không bảo đảm chắc là xong được, cái thay đổi chỉ là kỳ vọng và bước đầu tiên của bạn. Nhân viên giao dịch ngân hàng và quầy thủ tục của cơ quan nhà nước cũng vậy: thao tác của họ đều để lại dấu vết, phải kiểm tra lại, sai thì phải bị truy trách nhiệm. Vậy nên làm là giữ phiếu, xin lại bản sao, khiếu nại đúng kênh, đừng đoán bừa theo kiểu「anh ấy có lừa mình không」. Thứ tự ưu tiên hợp pháp đều nằm ở chỗ sáng, ví dụ cấp cứu xếp theo mức bệnh chứ không theo ai đến trước (mục 7 ở phần 24, phân loại bệnh nhân cấp cứu). Muốn cảm ơn bác sĩ đã cứu mình thì dùng thư cảm ơn và bảng đánh giá mức hài lòng, xem mục 11 ở phần 24. Muốn「tặng chút quà cho người ta chăm hơn」thì với người thi hành công vụ là tội hối lộ, và còn bị tăng mức phạt theo văn bản, xem mục 40 ở phần 8. Gặp đối xử rõ ràng khác biệt thì cứ khiếu nại, đừng học theo luật ngầm. -### 21. Đừng coi「người khác sống thế nào」bài đọc mỗi ngày: giới hạn hoặc tắt các app đua xem hoạt động của người cùng trang lứa + +### 20. Đừng coi việc「xem người khác sống thế nào」là phần bắt buộc phải đọc mỗi ngày: giới hạn hoặc tắt luôn các app lướt bảng tin của người cùng tuổi -- Chi phí: Không tốn tiền. Giới hạn app hay ngừng dùng tài khoản đều chỉ vài phút. Khó là sau đó không mở lại, phải chống quán tính lâu dài. -- Nói thẳng: Người quen so sánh trên mạng với người sống tốt hơn thường tâm trạng kém hơn, tự trọng thấp hơn. Người tắt Facebook bốn tuần thì hạnh phúc tăng nhẹ, nhưng hiểu biết thời sự cũng giảm rõ. Người giới hạn ba nền tảng mạng xã hội, mỗi nền tảng 10 phút mỗi ngày, thì ba tuần sau cô đơn và trầm cảm đều giảm. -- Lợi ích: Gộp nhiều nghiên cứu (phân tích tổng hợp): 54 mẫu độc lập, 94 cỡ hiệu ứng, N = 36.583, dùng mô hình biến thiên ngẫu nhiên ba tầng. Tương quan trung bình giữa việc so sánh xã hội đi lên trên mạng và sự thích nghi tâm lý kém là r = 0.330 (95% CI 0.289–0.370, khoảng tin cậy). Tách theo từng kết quả. Cảm xúc tiêu cực mang tính đánh giá xã hội như sợ bị coi thường là 0.438, lo âu 0.382. Trầm cảm 0.306, hạnh phúc thấp 0.268, tự trọng thấp 0.263. Tuổi, bối cảnh văn hóa và năm thu thập dữ liệu đều không làm đổi độ mạnh của liên hệ này. Nghiên cứu theo dõi lâu dài cũng không cho kết quả mạnh hơn khảo sát tại cùng một thời điểm. Còn một thử nghiệm đối chứng chia nhóm ngẫu nhiên, người tham dự là 2.743 người dùng Facebook Mỹ. Trong đó 61% sẵn sàng chấp nhận việc tắt với mức dưới 102 đô la. Họ được chia ngẫu nhiên vào nhóm tắt bốn tuần và nhóm đối chứng. Việc tắt giải phóng cho người tham dự trung bình 60 phút mỗi ngày. Chỉ số chủ quan về hạnh phúc tăng 0.09 độ lệch chuẩn. Tách ra: hài lòng với cuộc sống 0.12, lo âu 0.10, trầm cảm 0.09, vui 0.08. Sau khi trừ phần may mắn do so sánh nhiều lần, những con số này vẫn tính. Tác giả có đưa một mức tham chiếu, là các can thiệp tâm lý làm người ta tích cực hơn. Loại can thiệp đó nâng hạnh phúc chủ quan khoảng +0.34 độ lệch chuẩn và hạ trầm cảm khoảng -0.23 độ lệch chuẩn. Vậy hiệu ứng này vào khoảng 25% đến 40% so với một can thiệp tâm lý thường quy. Cùng kỳ cái giá là: chỉ số hiểu biết thời sự giảm 0.19 độ lệch chuẩn, mức độ đối lập trên lập trường chính trị giảm 0.16 độ lệch chuẩn. Sau khi thí nghiệm kết thúc, mức dùng Facebook của nhóm tắt vẫn thấp hơn nhóm đối chứng 12 phút mỗi ngày (0.31 độ lệch chuẩn, mức giảm 23%). Còn một thử nghiệm đối chứng chia nhóm ngẫu nhiên khác, người tham dự là 143 sinh viên năm nhất, chia ngẫu nhiên thành hai nhóm. Một nhóm giới hạn Facebook, Instagram, Snapchat, mỗi nền tảng 10 phút mỗi ngày, nhóm kia dùng như bình thường. Ba tuần sau, nhóm giới hạn thì cô đơn và trầm cảm giảm rõ. Cả hai nhóm thì lo âu và sợ bỏ lọt đều thấp hơn lúc đầu. +- Chi phí: Không tốn tiền. Đặt giới hạn cho ứng dụng, hoặc khoá tài khoản, đều chỉ mất vài phút. Chỗ khó là sau đó không mở lại nữa, phải chống lại thói quen lâu dài. +- Nói thẳng: Người quen so sánh mình với người khác trên mạng thường tâm trạng tệ hơn và lòng tự trọng thấp hơn. Người tắt Facebook bốn tuần thì mức hạnh phúc tăng nhẹ, nhưng hiểu biết về thời sự lại giảm. Người giới hạn ba nền tảng xã hội mỗi nền tảng còn 10 phút mỗi ngày thì sau ba tuần cả cô đơn lẫn trầm cảm đều giảm. +- Lợi ích: Gộp nhiều nghiên cứu lại (phân tích tổng hợp): 54 mẫu độc lập, 94 cỡ hiệu ứng, N = 36.583, dùng mô hình hiệu ứng ngẫu nhiên ba tầng. Tương quan trung bình giữa so sánh xã hội theo hướng lên trên trên mạng và thích nghi tâm lý kém là r = 0.330 (95% CI 0.289–0.370, khoảng tin cậy). Xét riêng từng kết quả cụ thể. Cảm xúc tiêu cực mang tính đánh giá xã hội như sợ bị coi thường là 0.438, lo âu là 0.382. Trầm cảm là 0.306, hạnh phúc thấp là 0.268, lòng tự trọng thấp là 0.263. Tuổi, bối cảnh văn hoá, năm thu thập dữ liệu đều không làm thay đổi độ mạnh của mối liên hệ này. Nghiên cứu theo dõi lâu dài cũng không cho kết quả mạnh hơn khảo sát tại một thời điểm. Còn một thử nghiệm đối chứng chia nhóm ngẫu nhiên, người tham gia là 2.743 người dùng Facebook tại Hoa Kỳ. Trong đó 61% người sẵn sàng chấp nhận tắt với mức dưới 102 đô la. Họ được chia ngẫu nhiên thành nhóm tắt bốn tuần và nhóm đối chứng. Với người tham gia bình thường, việc tắt giải phóng 60 phút mỗi ngày. Chỉ số hạnh phúc chủ quan tăng 0.09 điểm chuẩn. Các khoản riêng là hài lòng với cuộc sống 0.12, lo âu 0.10, trầm cảm 0.09, hạnh phúc 0.08. Sau khi trừ thành phần trùng hợp do so sánh nhiều lần, các con số này vẫn đứng vững. Tác giả đưa ra một mốc so sánh, đó là các can thiệp tâm lý làm người ta trở nên tích cực. Loại can thiệp đó làm chỉ số hạnh phúc chủ quan tăng khoảng +0.34 điểm chuẩn, và làm trầm cảm giảm khoảng -0.23 điểm chuẩn. Vậy hiệu ứng ở đây bằng khoảng 25% đến 40% can thiệp tâm lý thông thường. Cái giá trong cùng thời gian đó là: chỉ số kiến thức thời sự giảm 0.19 điểm chuẩn, mức độ đối đầu chính trị giảm 0.16 điểm chuẩn. Sau khi thử nghiệm kết thúc, nhóm tắt vẫn dùng Facebook ít hơn nhóm đối chứng 12 phút mỗi ngày (0.31 điểm chuẩn, giảm 23%). Còn một thử nghiệm đối chứng chia nhóm ngẫu nhiên nữa, người tham gia là 143 sinh viên đại học, được chia ngẫu nhiên thành hai nhóm. Một nhóm giới hạn Facebook, Instagram, Snapchat mỗi nền tảng 10 phút mỗi ngày, nhóm kia dùng bình thường. Sau ba tuần, nhóm có giới hạn giảm rõ rệt cả cô đơn lẫn trầm cảm. Ở cả hai nhóm, lo âu và nỗi sợ bỏ lỡ đều thấp hơn so với lúc đầu. - Mức bằng chứng: A -- Nguồn: Lei Y, Hu S, Sun Y, Zheng L (2026). "Looking up" linked to feeling down: a meta-analysis of online upward social comparison and psychological maladjustment. Frontiers in Psychology. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2020). The Welfare Effects of Social Media. American Economic Review, 110(3), 629–676. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2019). The Welfare Effects of Social Media. NBER Working Paper No. 25514 (các con số cụ thể mà mục này dẫn được đối chiếu từng chữ theo bản này, bản tóm tắt in không có con số) ; Hunt MG, Marx R, Lipson C, Young J (2018). No More FOMO: Limiting Social Media Decreases Loneliness and Depression. Journal of Social and Clinical Psychology, 37(10), 751–768. -- Ghi chú: Nửa「tương quan」chỉ cho thấy hai chuyện cùng xuất hiện, không đọc được thành chuyện nào gây ra chuyện nào, chính tác giả cũng nói cần thêm nghiên cứu theo dõi lâu dài. Nửa chia nhóm ngẫu nhiên nói được về quan hệ nhân quả, nhưng người tham dự là người dùng Facebook Mỹ trước bầu cử giữa kỳ 2018, đưa sang nền tảng Trung Quốc là lắp ghép gượng. Nghiên cứu Hunt chỉ có 143 sinh viên năm nhất, và chỉ có nghiên cứu đó. Cỡ hiệu ứng đều không lớn, đừng mong tắt app là giải quyết được vấn đề tâm trạng. Thứ cần chặn là loại dùng「xem người khác sống thế nào」; liên lạc với gia đình bạn bè, tra thông tin, tìm việc thì không tính. Nếu thật sự không giữ được thì bắt đầu từ tắt thông báo, tắt「người có thể quen」và tắt đề xuất ở trang đầu. Tính tổng thời gian màn hình xem mục 15 và 16 ở phần 4 (trần cứng cho video ngắn, không xem tin cuộn). Khi tâm trạng xuống làm gì trước xem mục 19 của phần này (vận động, phơi nắng, ngủ đúng giờ). Tìm lại cảm giác danh tính qua việc mua sắm xem mục 23 ở phần 6. Tiền bỏ thêm để「lên một bậc」xem mục 24 ở phần 6. -### 22. Khi thấy「ai cũng thấy mình mất mặt」thì chia đôi con số ước lượng đó +- Nguồn: Lei Y, Hu S, Sun Y, Zheng L (2026). "Looking up" linked to feeling down: a meta-analysis of online upward social comparison and psychological maladjustment. Frontiers in Psychology. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2020). The Welfare Effects of Social Media. American Economic Review, 110(3), 629–676. ; Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2019). The Welfare Effects of Social Media. NBER Working Paper No. 25514 (các con số cụ thể mà mục này dẫn được đối chiếu từng chữ với bản này, bản tóm tắt đã xuất bản không có số). ; Hunt MG, Marx R, Lipson C, Young J (2018). No More FOMO: Limiting Social Media Decreases Loneliness and Depression. Journal of Social and Clinical Psychology, 37(10), 751–768. +- Ghi chú: Nửa「tương quan」chỉ nói hai thứ xuất hiện cùng nhau, không đọc được thành ai gây ra ai, chính tác giả cũng nói còn cần thêm nghiên cứu theo dõi lâu dài. Nửa chia nhóm ngẫu nhiên thì nói được quan hệ nhân quả, nhưng người tham gia là người dùng Facebook trước bầu cử giữa kỳ năm 2018 tại Hoa Kỳ, mang sang các nền tảng của Trung Quốc là ép vừa. Nghiên cứu của Hunt chỉ có 143 sinh viên đại học, và chỉ một nghiên cứu đó. Cỡ hiệu ứng đều không lớn, đừng trông chờ tắt ứng dụng là giải quyết được chuyện tâm trạng. Cái cần chặn là loại sử dụng「xem người khác sống thế nào」, không tính liên lạc với người thân bạn bè, tra cứu thông tin, tìm việc. Thật sự không kìm được thì bắt đầu từ việc tắt thông báo đẩy, tắt「người có thể quen」và mục đề xuất ở trang chủ. Bài tính tổng thời gian nhìn màn hình xem ở mục 15 và 16 ở phần 4 (đặt trần cứng cho video ngắn, không đọc tin cuộn). Tâm trạng tụt thì làm gì trước, xem mục 18 của phần này (vận động, phơi nắng, ngủ đúng giờ). Mua sắm để tìm lại cảm giác bản thân xem mục 23 ở phần 6. Tốn thêm tiền chỉ để「nhích lên một nấc」xem mục 24 ở phần 6. + +### 21. Khi thấy「ai cũng thấy mình mất mặt」thì chia đôi con số ước lượng đó - Chi phí: Không tốn tiền. Chỉ là tự hỏi thêm một câu lúc căng thẳng: họ thật sự nhìn thấy sao. - Nói thẳng: Bước vào phòng có người mặc chiếc T-shirt đáng xấu hổ, người mặc nghĩ gần một nửa số người nhận ra ai được in trên áo, thực tế chưa đến một phần tư. Mặc chiếc T-shirt thấy tử tế thì con số ước lượng gấp sáu lần thực tế. Sau khi trò chuyện, người ta cũng đánh giá thấp mức đối phương thích mình. Vậy khi thấy「ai cũng nhớ mình mất mặt」, chia đôi vẫn còn cao. @@ -190,7 +181,7 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: B - Nguồn: Gilovich T, Medvec VH, Savitsky K (2000). The spotlight effect in social judgment: An egocentric bias in estimates of the salience of one's own actions and appearance. Journal of Personality and Social Psychology, 78(2), 211–222. ; Boothby EJ, Cooney G, Sandstrom GM, Clark MS (2018). The Liking Gap in Conversations: Do People Like Us More Than We Think? Psychological Science, 29(11), 1742–1756. - Ghi chú: Lý do chấm B: đều là nghiên cứu thí nghiệm trong phòng lab và nghiên cứu ở trường, bối cảnh là mặc quần áo và nói chuyện với người lạ. Chuyển sang「bị đồng nghiệp bàn tán」「nói sai trên họp」thì không có bằng chứng trực tiếp. Những con số này chỉ cho hướng ước cao và độ lớn đại khái, không dùng làm hệ số được. Ý kiến người khác cũng không phải thứ gì cũng quan trọng. Chỗ thật sự đang đánh giá bạn, như phỏng vấn, bảo vệ luận văn, đánh giá, báo cáo khám sức khỏe, thì đánh giá là thật, đã thế thì chuẩn bị. Cái cần giảm chỉ là cảm giác「mọi người đều đang nhìn mình」trong đời thường, chỗ đó thứ phải đối phó là chính thiên lệch ước lượng của bạn. -### 23. Xem câu「người khác đòi tôi hoàn hảo」là triệu chứng, không phải sự thật +### 22. Xem câu「người khác đòi tôi hoàn hảo」là triệu chứng, không phải sự thật - Chi phí: Không tốn tiền. Khó ở chỗ xem câu nghĩ đó là thứ cần xử lý, chứ không phải là phép đo chính xác yêu cầu thật của người khác. - Nói thẳng: Người luôn thấy「người quanh tôi đòi tôi hoàn hảo, không đạt thì sẽ bị phủ nhận」thì dễ có ý nghĩ tự tắt hơn. Mối liên hệ giữa「tôi tự đòi mình hoàn hảo」và ý nghĩ tự tắt yếu hơn nhiều, chỉ bằng khoảng một phần ba cách thứ nhất. Trừ những ý nghĩ đã có sẵn, cảm giác thứ nhất vẫn liên quan đến việc ý nghĩ tăng lên sau đó, chỉ là độ lớn không lớn. @@ -198,16 +189,17 @@ Phần này chỉ tính sức và thời gian. Nó nhìn bốn thứ: lượng c - Mức bằng chứng: B - Nguồn: Smith MM, Sherry SB, Chen S, Saklofske DH, Mushquash C, Flett GL, Hewitt PL (2018). The perniciousness of perfectionism: A meta-analytic review of the perfectionism-suicide relationship. Journal of Personality, 86(3), 522–542. - Ghi chú: Lý do chấm B: phần lớn nghiên cứu đưa vào là bảng hỏi một lần, không thấy trước sau. Độ mạnh tương quan ở mức trung bình (r khoảng .28), đại khái chỉ giải thích được vài phần trăm đến một phần mười chênh lệch. Thứ đo là ý nghĩ tự sát và hành động tự sát, không phải cái chết本身, nên quy mức lợi ích nhỏ với lý do「chỉ có điểm cuối thay thế」. Chính tác giả cũng nói nhóm lo lắng hoàn hảo và hành động tự sát này hình như có thiên lệch xuất bản. Thiên lệch xuất bản nghĩa là nghiên cứu kết quả không đẹp thì không được in. Bù phần nghiên cứu mất ước tính vào thì hướng kết luận không đổi. Tiêu chuẩn cao本身 không phải vấn đề, cũng không vì thế mà hạ yêu cầu. Khác nhau ở chỗ tiêu chuẩn đến từ đâu, và khi không đạt có thấy「mình thế là xong」hay không. Nó cũng như dự đoán bi quan ở mục 15 (xem câu「chắc chắn mọi chuyện sẽ tệ hơn」là triệu chứng), là thứ xử lý được như triệu chứng. Có ý nghĩ tự tắt thì gọi 12356 trước và chuyển phương tiện gây tử vong ra xa, xem mục 25 ở phần 1. Người quanh bạn bộc lộ ý nghĩ này thì bạn làm được gì, xem mục 15 ở phần 8. Sàng lọc trầm cảm cho trẻ xem mục 8 ở phần 30. -### 24. Khi giận đừng đập đồ, đấm bao cát, chạy một vòng để xả; hãy làm cơ thể chậm lại trước +### 23. Khi giận đừng trút bằng cách đập đồ, đấm bao cát, chạy một vòng; hãy làm cơ thể chậm lại trước -- Chi phí: Không tốn tiền. Lúc đang nóng, đổi sang vài phút làm cho tim chậm lại: thở chậm, ngồi xuống, yên một lát. Khó ở chỗ đúng lúc đó việc muốn nhất lại chính là xả ra. -- Nói thẳng: Đấm bao cát, đập đồ, đi chạy một vòng, những cách xả làm cơ thể kích thích hơn thì nhìn chung không hết giận. Hơi thở chậm, chánh niệm, thi định, những cách làm cơ thể yên lại thì ép được cơn giận và cái ham muốn xuống tay xuống chân. -- Lợi ích: Gộp 154 nghiên cứu (phân tích tổng hợp), tổng 184 mẫu, 10.189 người. Cách làm cơ thể bình tĩnh lại, như thở sâu, chánh niệm, thi định, làm giảm cơn giận và tính hung hăng. Cỡ hiệu ứng g = -0.63, 95% CI -0.82 đến -0.43 (khoảng tin cậy). Kết quả này ổn định ở người khác giới, chủng tộc, tuổi và văn hóa. Cách làm cơ thể kích thích hơn, như đấm bao cát, chạy bộ, đạp xe, nhìn chung không hiệu quả. Cỡ hiệu ứng của nó là g = -0.02, 95% CI -0.13 đến 0.09. Kết luận của tác giả là: xả cơn giận, đi chạy một vòng, đều không phải cách hết giận hiệu quả. +- Chi phí: Không tốn tiền. Lúc đang nóng, đổi sang làm vài phút việc làm nhịp tim chậm lại: thở chậm, ngồi xuống, bình tĩnh lại. Chỗ khó là lúc đó cái người ta muốn nhất lại đúng là trút giận. +- Nói thẳng: Đấm bao cát, ném đồ, đi chạy một vòng, những cách trút giận làm cơ thể hưng phấn hơn thì nhìn chung không giảm được cơn giận. Thở chậm, chánh niệm, thiền, những cách làm cơ thể tĩnh lại thì giảm rõ cơn giận và cơn muốn ra tay. +- Lợi ích: Gộp 154 nghiên cứu lại (phân tích tổng hợp), gồm 184 nhóm mẫu, 10.189 người. Những cách làm cơ thể bình tĩnh lại, như hít thở sâu, chánh niệm, thiền, có tác dụng giảm cơn giận và sự hung hăng. Cỡ hiệu ứng g = -0.63, 95% CI -0.82 đến -0.43 (khoảng tin cậy). Kết quả này ổn định ở người khác giới, khác chủng tộc, khác tuổi và khác văn hoá. Những cách làm cơ thể hưng phấn hơn, như đấm bao cát, chạy bộ, đạp xe, nhìn chung vô hiệu. Cỡ hiệu ứng của nó là g = -0.02, 95% CI -0.13 đến 0.09. Kết luận của tác giả là: trút giận, chạy một vòng, đều không phải cách trút giận hiệu quả. - Mức bằng chứng: A - Nguồn: Kjærvik SL, Bushman BJ (2024). A meta-analytic review of anger management activities that increase or decrease arousal: What fuels or douses rage? Clinical Psychology Review. -- Ghi chú: Thứ đo là cơn giận và tính hung hăng, không phải sức lực. Ghi口径 vào thời gian là cách xếp của tác giả: ít cãi một trận thì ít dọn một lần. Cách dùng được ngay lúc đó xem mục 9 ở phần 22 (thở vòng), còn mục 18 của phần này (khi giận thì đi khỏi chỗ đó trước). Chạy bộ không phải vô dụng. Nhìn dài hạn thì nó hiệu quả với tâm trạng xuống, xem mục 7 ở phần 22 (tâm trạng kém thì đi bộ hoặc chạy). Chỉ là không hợp để hết giận ngay tại chỗ. Lúc đang nóng đừng lái xe, đừng gửi tin nhắn, đừng quyết định. Nếu có người đang đánh bạn, rời đi trước rồi gọi công an. Lúc đó thứ cần xử lý không phải cảm xúc của bạn, xem mục 43 ở phần 8 (bị bạo lực gia đình). +- Ghi chú: Cái đo là cơn giận và sự hung hăng, không phải sức lực. Ghi vào thước đo thời gian là cách tác giả xếp loại: ít cãi nhau một lần là ít dọn dẹp hậu quả một lần. Cách dùng ngay tại chỗ xem mục 8 ở phần 22 (thở dài theo vòng tròn), và mục 17 của phần này (khi giận thì rời khỏi chỗ đó). Chạy bộ không phải là vô dụng. Nhìn dài hạn, nó có tác dụng với tâm trạng tụt, xem mục 6 ở phần 22 (tâm trạng tệ thì đi bộ hoặc chạy). Nó chỉ không hợp để trút giận tại chỗ. Lúc nóng máu đừng lái xe, đừng nhắn tin, đừng ra quyết định. Nếu có người đang đánh bạn, hãy rời đi và báo công an. Lúc đó cần xử lý không phải cảm xúc của bạn, xem mục 43 ở phần 8 (bạo hành gia đình). + -### 25. Có một chuyện đè nặng trong lòng, có thể viết suốt mấy ngày liền tư tưởng và cảm xúc của mình ra, nhưng đừng trông chờ nó chữa bệnh +### 24. Có một chuyện đè nặng trong lòng, có thể viết suốt mấy ngày liền tư tưởng và cảm xúc của mình ra, nhưng đừng trông chờ nó chữa bệnh - Chi phí: Không tốn tiền. Cách làm thường thấy trong nghiên cứu là viết 3 đến 4 lần, mỗi lần 15 đến 20 phút. Viết một chuyện khiến bạn khó chịu, viết ra những suy nghĩ và cảm xúc sâu nhất của bạn - Nói thẳng: Viết những chuyện đè nặng lòng ra thì trung bình có ít lợi ích, nhưng rất nhỏ. Kết luận của các tổng hợp cũng không nhất quán, có hai tổng hợp tính ra là không có hiệu quả. Nó gần như không tốn tiền, có thể thử. Vấn đề tình cảm nặng thì đừng dùng nó thay cho việc đi khám bác sĩ diff --git "a/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" "b/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" index 39302372..457b8e4d 100644 --- "a/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" +++ "b/vi/book/04-\344\270\215\350\246\201\346\265\252\350\264\271\346\227\266\351\227\264.md" @@ -12,7 +12,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: A - Nguồn: Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. - Ghi chú: Cách này chỉ hiệu quả với việc bạn thật sự muốn làm. Nếu bản thân mục tiêu bạn không muốn thì viết thêm bao nhiêu câu cũng vô ích. Thêm nữa, các nghiên cứu này lấy mẫu chủ yếu là học sinh và sinh viên, xét chủ yếu thói quen sức khoẻ, chuyển sang công việc chưa chắc còn hiệu quả như vậy. - ### 2. Trước khi bắt tay, viết điều kiện dừng - Chi phí: Không tốn tiền. Lúc bắt tay dành 5 phút, viết rõ「đến ngày nào nào đó mà chưa đạt chỉ số nào đó thì dừng」. Chỗ khó là đến ngày đó phải nhận, dừng tay trước mặt người ta thì hơi mất thể diện. @@ -21,7 +20,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: B - Nguồn: Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. - Ghi chú: Điều kiện dừng phải định lúc vừa bắt tay, khi tâm trạng còn bình tĩnh. Định sau sẽ bị chi phí đã bỏ vào dẫn lệch. Khi cân nhắc chỉ nhìn phần đầu tư tương lai và phần thu về tương lai, xem mục 3. Thêm nữa,「sẵn sàng rút ra」chỉ là một kết luận phụ trong bản tổng hợp nhiều nghiên cứu, chưa có ai tính riêng nó tiết kiệm được bao nhiêu thời gian. - ### 3. Khi quyết định có tiếp tục hay không, chỉ nhìn phần sẽ đầu tư và phần sẽ thu về, không nhìn phần đã bỏ - Chi phí: Không tốn tiền. Chỗ khó là phải chấp nhận một đằng: thời gian đã bỏ vào trước đó mất rồi, không lấy lại được. @@ -30,7 +28,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: B - Nguồn: Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35, 124–140. ; Roth, S., Robbert, T., & Straus, L. (2015). On the sunk-cost effect in economic decision-making: A meta-analytic review. Business Research, 8(1), 99–138. - Ghi chú: Thí nghiệm rạp hát của Arkes và Blumer làm như sau. Rạp hát đại học Ohio bán vé cả mùa, ngẫu nhiên chào bán ba mức giá cho người mua: giá gốc 15 đô la, rẻ hơn 2 đô la, rẻ hơn 7 đô la. Nửa đầu mùa, nhóm mua giá gốc xem trung bình 4.11 buổi, hai nhóm kia là 3.32 và 3.29 buổi. Nửa sau mùa thì ba nhóm không còn khác biệt rõ. Roth và cộng sự thấy hiệu ứng này nhạt dần theo thời gian, người lớn tuổi chịu ảnh hưởng ít hơn. - ### 4. Ước lượng thời hạn thì dựa vào thời gian thực tế của việc cùng loại trước đây, không suy ra từ kế hoạch - Chi phí: Không tốn tiền. Trước khi ước lượng, dành vài phút xem lại việc cùng loại lần trước thực tế mất bao lâu. @@ -39,7 +36,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: B - Nguồn: Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. ; Flyvbjerg, B. (2006). From Nobel Prize to project management: Getting risks right. Project Management Journal, 37(3), 5–15. ; Halkjelsvik, T., & Jørgensen, M. (2012). From origami to software development: A review of studies on judgment-based predictions of performance time. Psychological Bulletin, 138(2), 238–271. - Ghi chú: Nghiên cứu của Buehler chỉ lấy mẫu sinh viên đại học Canada, là một nghiên cứu đơn lẻ và số người ít, 37 người và khoảng 40 người mỗi nhóm. Ước lượng theo số liệu lịch sử của dự án cùng loại được gọi là dự báo theo lớp tham chiếu, cách làm cụ thể trong kỹ thuật xem ở Flyvbjerg. Halkjelsvik tổng hợp nhiều nghiên cứu rồi chỉ ra rằng ước lượng ngắn hơn thường gặp hơn ước lượng dài hơn, nhưng chênh lệch bao nhiêu thay đổi rất nhiều theo loại việc và theo phương pháp nghiên cứu. - ### 5. Gửi đề nghị trước cuộc họp, không có đề nghị thì không họp; họp được đứng thì cứ đứng - Chi phí: Không tốn tiền. Người tổ chức dành 5 đến 10 phút trước cuộc họp để viết đề nghị. Họp đứng thì cần người tham dự phối hợp. @@ -48,7 +44,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: B - Nguồn: Leach, D. J., Rogelberg, S. G., Warr, P. B., & Burnfield, J. L. (2009). Perceived meeting effectiveness: The role of design characteristics. Journal of Business and Psychology, 24(1), 65–76. ; Bluedorn, A. C., Turban, D. B., & Love, M. S. (1999). The effects of stand-up and sit-down meeting formats on meeting outcomes. Journal of Applied Psychology, 84(2). - Ghi chú: Thí nghiệm về họp đứng làm nhiệm vụ nhóm 5 người trong phòng thí nghiệm. Không thể chuyển nguyên kết luận này sang họp dài và hội nghị lớn. Leach đo là người tham dự tự cảm thấy hiệu quả ra sao, không phải khối lượng việc đã làm được. - ### 6. Bớt số cuộc họp, chuyện nào giải quyết bằng văn bản bất đồng bộ được thì khỏi họp - Chi phí: Không tốn tiền. Cần đồng nghiệp và cấp trên phối hợp. Chỗ khó là viết rõ một đoạn văn bản tốn sức hơn nói một câu. @@ -57,7 +52,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: B - Nguồn: Rogelberg, S. G., Leach, D. J., Warr, P. B., & Burnfield, J. L. (2006). "Not another meeting!" Are meeting time demands related to employee well-being? Journal of Applied Psychology, 91(1). ; Luong, A., & Rogelberg, S. G. (2005). Meetings and more meetings: The relationship between meeting load and the daily well-being of employees. Group Dynamics: Theory, Research, and Practice, 9(1). - Ghi chú:Tranh cãi. Rogelberg 2006 thấy quan hệ giữa số cuộc họp và hạnh phúc không nghiêng hẳn về một phía. Với vị trí mà công việc phụ thuộc cao vào người khác, họp nhiều lại không phải là điều xấu. Thêm nữa, chưa có nghiên cứu nào so trực tiếp「chuyển sang trao đổi bằng văn bản bất đồng bộ」với việc họp, mục này là suy luận của tác giả. - ### 7. Tách việc lớn thành các việc con rồi mới ước lượng thời gian, rồi mới bắt tay - Chi phí: Không tốn tiền. Bản thân việc tách mất 5 đến 15 phút. @@ -75,7 +69,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: B - Nguồn: Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. - Ghi chú: Nói ngày bạn tự đặt cho người khác biết thì nó gần với ngày do người khác đặt hơn. Nghiên cứu xác nhận là hạn chót do người khác đặt có tác dụng. Loại tự đặt cho mình, như canh giờ hay đồng hồ pomodoro, chưa ai đo trực tiếp, chỗ này là suy ra theo. - ### 9. Hiểu việc trì hoãn là đang né thứ khó chịu, đừng hiểu là chưa đủ cố gắng - Chi phí: Không tốn tiền. Đổi hướng phán đoán: hỏi trước chỗ nào của việc đó làm bạn khó chịu, rồi mới nghĩ cách. @@ -84,7 +77,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: A - Nguồn: Steel P (2007). The nature of procrastination: a meta-analytic and theoretical review of quintessential self-regulatory failure. Psychological Bulletin, 133(1), 65–94. - Ghi chú: Cỡ lợi ích ghi là「vừa」, vì các nghiên cứu này chỉ chỉ ra hướng, chưa đo được tiết kiệm bao nhiêu giờ. Hành động cụ thể nằm ở mục 1 của phần này (viết thành「mấy giờ, ở đâu, gặp gì thì làm gì」), mục 7 (tách việc lớn thành việc con), mục 8 (đặt một ngày cho việc không có hạn chót bên ngoài). Đây là bản tổng hợp của các nghiên cứu tương quan, không xác định được cái nào dẫn tới cái nào. - ### 10. Đặt sẵn thứ mình cần dùng, đẩy xa thứ mình không muốn đụng, đừng trông chờ nhịn tại chỗ - Chi phí: Không tốn tiền, mỗi lần vài phút. Đưa điện thoại sang phòng bên cạnh, bày sách và máy tính ra bàn. @@ -93,7 +85,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: B - Nguồn: Hollands GJ, Carter P, Anwer S, et al. (2019). Altering the availability or proximity of food, alcohol, and tobacco products to change their selection and consumption. Cochrane Database of Systematic Reviews, 9, CD012573. - Ghi chú: Lý do xếp mức B: cả 24 nghiên cứu mà tổng quan này đưa vào đều về thức ăn, không có mục nào về rượu hay thuốc lá; 14 nghiên cứu làm trong phòng thí nghiệm và đều đến từ nước có thu nhập cao. Nên chuyển「đẩy xa thì ít chạm vào」sang điện thoại và trò chơi là suy nối, không có thử nghiệm trực tiếp nào. Riêng việc đưa điện thoại ra khỏi tầm nhìn thì có người đã đo trực tiếp, xem mục 1 ở phần 3 (tắt thông báo không cần thiết, khi làm việc thì đưa điện thoại ra khỏi tầm nhìn). - ### 11. Muốn bắt bản thân làm được thì đặt tiền vào, nhưng trước hết hãy hỏi mình có chịu đặt không - Chi phí: Phương án dạng ký quỹ đòi trước một khoản tiền của chính mình, làm không được thì mất. Trong thí nghiệm, khoản ký quỹ là 150 đô la. @@ -102,7 +93,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: A - Nguồn: Halpern SD, French B, Small DS, et al. (2015). Randomized trial of four financial-incentive programs for smoking cessation. New England Journal of Medicine, 372(22), 2108–2117. - Ghi chú: Thử nghiệm này đo việc bỏ thuốc lá, chuyển sang viết luận văn hay tập thể dục thì là suy nối, nên cỡ lợi ích ghi là「vừa」. Khi chọn phương án, xem trước bản thân mình có thật sự dùng nó không. Cách phần đông số không chịu chấp nhận thì dù tác dụng mạnh đến đâu cũng vô ích với họ. Nên cứ chọn loại mình chịu dùng, đặt nhỏ hơn còn hơn không đặt. Cách bỏ thuốc lá thì xem mục 3 ở phần 2 (thuốc cai thuốc lá). - ### 12. Thói quen mới thì tính theo tháng, đừng tính theo「21 ngày」 - Chi phí: Không tốn tiền. Điều cần là lặp lại cùng một hành động mỗi ngày trong cùng một bối cảnh, kéo dài vài tháng. @@ -111,15 +101,14 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: B - Nguồn: Lally P, van Jaarsveld CHM, Potts HWW, Wardle J (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998–1009. - Ghi chú: Chỉ có một nghiên cứu này, 96 người, và các hành động đều đơn giản (chẳng hạn ăn một phần trái cây sau bữa ăn). Việc phức tạp cần bao lâu, như chạy ba lần mỗi tuần hay viết mỗi ngày, thì chưa ai đo. Nên sau hai tuần mà vẫn thấy khó thì không thể coi là bằng chứng thất bại. Muốn sự lặp lại thật sự xảy ra thì gắn hành động với một bối cảnh cố định, xem mục 1 của phần này (viết thành「mấy giờ, ở đâu, gặp gì thì làm gì」). - -### 13. Khi việc trì hoãn đã ảnh hưởng tới cuộc sống thì dùng tài liệu tự giúp có thử nghiệm ngẫu nhiên hỗ trợ, không bắt buộc phải tìm nhà trị liệu +### 13. Khi việc trì hoãn đã ảnh hưởng tới cuộc sống, dùng tài liệu tự học có thử nghiệm ngẫu nhiên hỗ trợ, không nhất thiết phải đi tìm nhà trị liệu -- Chi phí: Một cuốn sách tự giúp vài chục tệ, có loại miễn phí. Phải bỏ ra khoảng mười tuần, mỗi tuần đọc một phần và làm một bài tập. -- Nói thẳng: Việc trì hoãn nặng đến mức gây đau đớn thì chữa được. Luyện theo tài liệu tự giúp nhận thức-hành vi trong mười tuần, dù có người hướng dẫn hay tự làm hoàn toàn, đều rõ ràng hơn là không làm gì cả. Có người hướng dẫn và không có không thấy khác nhau, nên tự làm theo cũng có tác dụng. -- Lợi ích: 150 người được chia ngẫu nhiên thành ba nhóm: tự giúp có nhà trị liệu hướng dẫn, tự giúp không hướng dẫn, và nhóm đối chứng danh sách chờ. Sau mười tuần, cỡ hiệu ứng giữa các nhóm trên thang đo trì hoãn thuần túy là d = 0.70 ở nhóm có hướng dẫn (95% CI 0.29–1.10, đây là khoảng tin cậy), d = 0.50 ở nhóm không hướng dẫn (0.10–0.90); trên thang đo trì hoãn phi lý tính lý là 0.81 (0.40–1.22) và 0.69 (0.29–1.09) tương ứng. Tỷ lệ đạt cải thiện ở mức có ý nghĩa lâm sàng: nhóm có hướng dẫn 31.3% đến 40.0%, nhóm không hướng dẫn 24.0% đến 36.0%. Hai nhóm không khác nhau có ý nghĩa ở bất kỳ kết quả nào. +- Chi phí: Một cuốn sách tự học vài chục tệ, có cuốn miễn phí. Phải bỏ ra khoảng mười tuần, mỗi tuần đọc một phần và làm một bài tập. +- Nói thẳng: Trì hoãn nặng đến mức khiến người ta khó chịu thì chữa được. Làm theo tài liệu tự học theo hướng tiếp cận nhận thức – hành vi trong mười tuần thì cả khi có người hướng dẫn lẫn khi tự làm một mình, đều rõ ràng hơn nhiều so với không làm gì. Không thấy khác biệt giữa có người hướng dẫn và không có, nên tự làm theo cũng có tác dụng. +- Lợi ích: 150 người chia ngẫu nhiên thành ba nhóm: tự học có nhà trị liệu hướng dẫn, tự học không có hướng dẫn, và nhóm đối chứng chờ danh sách. Sau mười tuần, cỡ tác động giữa các nhóm trên thang đo trì hoãn thuần tuý là: nhóm có hướng dẫn d = 0.70 (95% CI 0.29–1.10, đây là khoảng tin cậy), nhóm không hướng dẫn d = 0.50 (0.10–0.90). Trên thang đo trì hoãn phi lý tính lý, hai nhóm tương ứng là 0.81 (0.40–1.22) và 0.69 (0.29–1.09). Tỷ lệ đạt mức cải thiện có ý nghĩa lâm sàng: nhóm có hướng dẫn 31.3% đến 40.0%, nhóm không hướng dẫn 24.0% đến 36.0%. Hai nhóm không khác biệt có ý nghĩa ở bất kỳ kết quả nào. - Mức bằng chứng: A - Nguồn: Rozental A, Forsell E, Svensson A, Andersson G, Carlbring P (2015). Internet-based cognitive-behavior therapy for procrastination: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 83(4), 808–824. -- Ghi chú: Người tham gia là người lớn Thụy Điển, tài liệu dùng bằng tiếng Thụy, tiếng Trung không có sẵn bộ cùng loại, nên cỡ lợi ích ghi là「vừa」. Các thành phần trong tài liệu đều là những thứ quen thuộc: kích hoạt hành vi, thí nghiệm hành vi, kiểm soát kích thích, phần giải thích về động lực và cách làm việc. Riêng kiểm soát kích thích nằm ở mục 10 của phần này (đẩy xa thứ mình không muốn đụng). Nếu việc trì hoãn đi kèm tâm trạng trầm chuyên rõ rệt hay lo âu thì trước hết làm theo mục 19 ở phần 3 (vận động, phơi nắng, ngủ đúng giờ, nói với ai, gọi 12356). +- Ghi chú: Người tham gia là người trưởng thành ở Thụy Điển, tài liệu dùng bằng tiếng Thụy Điển, bộ cùng như vậy chưa có sẵn trong tiếng Trung, nên cỡ lợi ích ghi là「vừa」. Thành phần trong tài liệu cũng chỉ là mấy thứ quen thuộc: kích hoạt hành vi, thí nghiệm hành vi, kiểm soát kích thích, phần giải thích về động lực và cách làm việc. Riêng kiểm soát kích thích nằm ở mục 10 của phần này (đẩy thứ mình không muốn đụng ra xa). Nếu trì hoãn đi kèm tâm trạng buồn rõ rệt hoặc lo âu, trước hết làm theo mục 18 ở phần 3 (vận động, ra nắng, ngủ đúng giờ, nói chuyện với ai đó, gọi 12356). ### 14. Dùng mức lương theo giờ của bạn để quyết định việc nhà nào thuê ngoài @@ -130,14 +119,14 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Nguồn: Whillans, A. V., Dunn, E. W., Smeets, P., Bekkers, R., & Norton, M. I. (2017). Buying time promotes happiness. Proceedings of the National Academy of Sciences. ; 国家统计局 (Cục Thống kê Quốc gia) (2024). 第三次全国时间利用调查公报（第二号）. - Ghi chú: Trước hết tính mức lương theo giờ, công thức là thu nhập sau thuế một tháng ÷ số giờ làm việc thực tế trong một tháng. Rồi xem việc nhà này thuê ngoài tốn bao nhiêu. Nếu đơn giá thuê ngoài thấp hơn mức lương theo giờ của bạn, và thời gian tiết kiệm được bạn thật sự dùng để kiếm tiền hoặc nghỉ ngơi, thì thuê ngoài. Mức lương theo giờ và giá thuê ngoài khác nhau với từng người, kết luận phải tự tính. Thêm nữa, Whillans đo là sau khi bỏ tiền mua thời gian thì người ta có vui không, không phải tiết kiệm được bao nhiêu giờ. -### 15. Đặt một giới hạn cứng cho video ngắn và lướt màn hình không mục đích +### 15. Đặt một trần cứng cho thời gian xem video ngắn và lướt màn hình không mục đích -- Chi phí: Không tốn tiền. Hành động chỉ mất vài phút: tắt thông báo đẩy, xoá ứng dụng, hoặc đặt cho ứng dụng một giới hạn thời gian mỗi ngày. Chỗ khó là kiềm tay, nhịn không bấm vào. -- Nói thẳng: Người Trung Quốc lên mạng trung bình 5 giờ 37 phút mỗi ngày. Trong số người thật sự lên mạng trong ngày, là 6 giờ 3 phút. Năm 2018, với cùng cách tính đó, con số này mới chỉ 2 giờ 42 phút. Đây là con số lớn nhất trong phần này. Mỗi ngày bớt một giờ thì một năm thêm 365 giờ. -- Lợi ích: Cư dân Trung Quốc lên mạng trung bình 5 giờ 37 phút mỗi ngày. Chỉ tính những người thật sự lên mạng trong ngày, trung bình 6 giờ 3 phút. Người dùng mạng trong ngày chiếm 92.9%. Đây là số liệu năm 2024. Năm 2018, với cùng cách tính, chỉ có 2 giờ 42 phút. Đây là con số lớn nhất trong phần này. Mỗi ngày bớt một giờ trong đó tức là mỗi năm thêm 365 giờ. Theo Trung tâm Thông tin Mạng lưới Trung Quốc (中国互联网络信息中心), tính đến tháng 6 năm 2025, người dùng mạng lên mạng trung bình 30.6 giờ mỗi tuần, nhiều hơn nửa năm trước 1.9 giờ. Người dùng video ngắn có 1.068.000.000 người, chiếm 95.1% người dùng mạng. +- Chi phí: Không tốn tiền. Thao tác chỉ mất vài phút: tắt thông báo đẩy, xoá ứng dụng, hoặc đặt cho ứng dụng một trần thời gian mỗi ngày. Chỗ khó là kiềm chế tay, nhịn không mở ra. +- Nói thẳng: Người Trung Quốc lên mạng trung bình 5 giờ 37 phút mỗi ngày. Trong số người thật sự lên mạng trong ngày, là 6 giờ 3 phút. Năm 2018 tính cùng cách thì chỉ có 2 giờ 42 phút. Đây là con số lớn nhất trong phần này. Bớt đi một giờ mỗi ngày thì một năm có thêm 365 giờ. +- Lợi ích: Cư dân Trung Quốc lên mạng trung bình 5 giờ 37 phút mỗi ngày. Chỉ tính những người thật sự lên mạng trong ngày, trung bình 6 giờ 3 phút. Người có dùng mạng trong ngày chiếm 92.9%. Đây là số liệu năm 2024. Năm 2018 tính cùng cách thì chỉ có 2 giờ 42 phút. Đây là con số lớn nhất trong phần này. Bớt đi một giờ trong số đó thì mỗi năm có thêm 365 giờ. Theo Trung tâm Thông tin Mạng lưới Internet Trung Quốc, tính đến tháng 6 năm 2025, người lên mạng trung bình 30.6 giờ mỗi tuần, hơn sáu tháng trước là 1.9 giờ. Người xem video ngắn có 1.068 triệu, chiếm 95.1% số người lên mạng. - Mức bằng chứng: C -- Nguồn: 国家统计局 (Cục Thống kê Quốc gia) (2024). 第三次全国时间利用调查公报（第二号）. ; 国家统计局 (Cục Thống kê Quốc gia) (2019). 2018年全国时间利用调查公报. ; 中国互联网络信息中心 (Trung tâm Thông tin Mạng lưới Trung Quốc) (2025). 第 56 次《中国互联网络发展状况统计报告》. -- Ghi chú: Thời gian lên mạng gồm cả công việc, học tập và chuyện xã giao, không phải lúc nào cũng là lãng phí. Xếp mức C vì không có nghiên cứu nào tính được「sau khi hạn chế thì tiết kiệm bao nhiêu và số thời gian đó đi đâu」, trong tay chỉ có số liệu tổng. Riêng nhóm「lướt xem người khác sống thế nào」thì có thử nghiệm chia nhóm ngẫu nhiên đo được tiết kiệm bao nhiêu và tâm trạng thay đổi ra sao, xem mục 21 ở phần 3. +- Nguồn: 国家统计局 (Cục Thống kê Quốc gia) (2024). 第三次全国时间利用调查公报（第二号）. ；国家统计局 (Cục Thống kê Quốc gia) (2019). 2018年全国时间利用调查公报. ；中国互联网络信息中心 (Trung tâm Thông tin Mạng lưới Internet Trung Quốc) (2025). 第 56 次《中国互联网络发展状况统计报告》. +- Ghi chú: Thời gian lên mạng gồm cả làm việc, học tập và chuyện xã giao, không phải lúc nào cũng là phí phạm. Xếp mức C là vì chưa nghiên cứu nào tính được sau khi giới hạn thì tiết kiệm lại được bao nhiêu, số thời gian đó lại đi đâu, trong tay chỉ có số liệu tổng. Riêng nhóm「lướt xem người khác sống thế nào」thì có thử nghiệm ngẫu nhiên đo được tiết kiệm bao nhiêu, tâm trạng thay đổi bao nhiêu, xem mục 20 ở phần 3. ### 16. Không xem truyền hình và tin tức chạy liên tục, gom thông tin cần xem vào một khung giờ cố định @@ -147,7 +136,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: C - Nguồn: U.S. Bureau of Labor Statistics (2026). American Time Use Survey — 2025 Results. ; 国家统计局 (Cục Thống kê Quốc gia) (2019). 2018年全国时间利用调查公报. - Ghi chú: Trong tay chỉ có số liệu tổng. Chưa nghiên cứu nào tính được sau khi bớt xem truyền hình và tin tức thì thời gian tiết kiệm được đi đâu. Cách「gom lại xem một lúc」là kinh nghiệm của tác giả. - ### 17. Bỏ thời gian một lần để học phím tắt và tự động hoá của phần mềm hay dùng - Chi phí: Không tốn tiền. Phải bỏ một lần vài tiếng đến vài chục tiếng để học. Chỗ khó là thời gian đầu mới bắt đầu còn chậm hơn dùng chuột. @@ -156,7 +144,6 @@ Phần này chỉ tính thời gian. Nó nhìn mỗi ngày, mỗi tuần, mỗi - Mức bằng chứng: B - Nguồn: Lane, D. M., Napier, H. A., Peres, S. C., & Sándor, A. (2005). Hidden costs of graphical user interfaces: Failure to make the transition from menus and icon toolbars to keyboard shortcuts. International Journal of Human-Computer Interaction. - Ghi chú: Chỉ học những thao tác bạn dùng hơn mười lần mỗi ngày, học cả mọi thứ là một kiểu lãng phí khác. Thí nghiệm đo hiệu suất đó chỉ có 6 người tham gia, phần tóm tắt cũng không nêu số giây tiết kiệm được mỗi thao tác. - ### 18. Khi chọn nơi ở thì đặt thời lượng đi làm lên đầu tiêu chí, rút ngắn thời gian đi một chiều - Chi phí: Chi phí cao. Có thể phải trả tiền thuê cao hơn, hoặc chuyển sang một căn nhỏ hơn. Bản thân việc chuyển nhà cũng tốn thời gian. diff --git "a/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" "b/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" index 8cedda52..41f631fe 100644 --- "a/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" +++ "b/vi/book/05-\344\270\215\350\246\201\346\265\252\350\264\271\351\222\261.md" @@ -12,7 +12,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: C - Nguồn: Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2021, sửa đổi 2025). Biện pháp quản lý giao dịch qua mạng（总局令第 37 号，第十八条）. - Ghi chú: Với dịch vụ bạn thực sự dùng thường xuyên, trả cả năm thường rẻ hơn trả từng tháng, giữ nguyên tự động gia hạn lại là hợp lý. Cần tắt đi là những gói bạn đã quên từ lâu mà vẫn bị trừ tiền. - ### 2. Mỗi năm từ tháng Ba đến tháng Sáu làm một lần tổng kết thuế thu nhập cá nhân, khai đủ các khoản trừ bổ sung chuyên ngành - Chi phí: Không tốn tiền. Làm trên app「个人所得税」, mười mấy phút là xong. @@ -21,7 +20,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Tổng cục Thuế quốc gia (国家税务总局) (2025). Biện pháp quản lý việc tổng kết và nộp thuế thu nhập cá nhân tính trên thu nhập tổng hợp（第六条）kèm phần giải thích. ；Bộ Tài chính, Tổng cục Thuế (2023). Thông báo về việc tiếp tục thực hiện chính sách liên quan đến tổng kết và nộp thuế thu nhập cá nhân tính trên thu nhập tổng hợp. - Ghi chú: Người phải nộp thuế bù cũng phải làm, không làm là vi phạm pháp. Khoản trừ bổ sung chuyên ngành có bảy mục: giáo dục con cái, đào tạo tiếp, điều trị bệnh hiểm nặng, lãi vay mua nhà, tiền thuê nhà, nuôi dưỡng người cao tuổi, chăm sóc trẻ dưới 3 tuổi. Trong đó lãi vay mua nhà và tiền thuê nhà chỉ được chọn một, không thể khai cả hai. - ### 3. Quỹ tích góp nhà ở không chỉ dùng để mua nhà: thuê nhà, sửa sang, phí quản lý nhà ở đều rút được, từ ngày 20 tháng 9 năm 2026 áp dụng quy định mới - Chi phí: Không tốn tiền. Đi làm thủ tục rút một lần là xong. @@ -30,7 +28,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2026). Quyết định của Hội đồng Nhà nước về việc sửa đổi《Quy định quản lý quỹ tích góp nhà ở》（国务院令第 844 号，2026 年 7 月 31 日国务院第 93 次常务会议通过，「自2026年9月20日起施行」）: Điều 二十四 sau khi sửa đổi「职工有下列情形之一的，可以提取职工住房公积金账户内的存储余额：（一）支付房租的；（二）购买、建造、翻建、大修自住住房的；（三）偿还购房贷款本息的；（四）装修自住住房的；（五）支付自住住房物业费的；（六）离休、退休的；（七）完全丧失劳动能力，并与单位终止劳动关系的；（八）出境定居的；（九）国务院批准的其他住房消费情形。」— người lao động gặp một trong chín trường hợp sau thì được rút số dư tiền còn lại trong tài khoản quỹ tích góp nhà ở của mình：（一）đang trả tiền thuê nhà；（二）mua, xây, xây lại, sửa lớn nhà ở của mình；（三）trả gốc và lãi khoản vay mua nhà；（四）sửa sang nhà ở của mình；（五）trả phí quản lý nhà ở của chính nhà mình ở；（六）nghỉ hưu, về hưu；（七）mất hoàn toàn năng lực lao động và đã chấm dứt quan hệ lao động với đơn vị；（八）ra nước ngoài định cư；（九）các trường hợp tiêu dùng nhà ở khác được Hội đồng Nhà nước phê duyệt.；Điều 二十五「住房公积金管理中心应当自受理申请之日起3日内作出准予提取或者不准提取的决定」— trung tâm quản lý quỹ tích góp nhà ở phải quyết định cho rút hay không cho rút trong 3 ngày kể từ ngày tiếp nhận đơn；Điều 十八, tỷ lệ đóng「均不得低于职工上一年度月平均工资的5％」— không được thấp hơn 5％ mức lương trung bình hằng tháng của người lao động trong năm trước. - Ghi chú: Số tiền trong tài khoản là lương của chính bạn, cộng thêm phần đơn vị đóng thêm. Để yên số tiền đó thì nó chỉ sinh lãi ở mức lãi suất rất thấp. Quy định còn ghi: khi người lao động chết, hoặc bị tuyên bố đã chết, người thừa kế và người được tặng có thể rút số dư tài khoản. Hồ sơ phải nộp gồm những gì, cụ thể làm thế nào, mỗi địa phương do trung tâm quản lý quỹ tích góp nhà ở địa phương quy định. Trước khi đi, hãy tra hướng dẫn thủ tục của trung tâm nơi bạn. - ### 4. Mỗi năm tính lại gói điện thoại và băng thông một lần, không dùng tới thì hạ bậc, nhà mạng không chuyển được thì khiếu nại - Chi phí: Không tốn tiền. Kiểm tra một lần hoá đơn, mười mấy phút. Chỗ khó là giữ được nhân viên không hạ bậc cho bạn. @@ -39,7 +36,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Bộ Công nghiệp và Thông tin (工业和信息化部) (2019). Quy chế quản lý dịch vụ chuyển mạng giữ số thuê bao（工信部信管〔2019〕242 号）. - Ghi chú: Trước hết xem trong app của cửa hàng nhà mạng lượng dữ liệu và số phút gọi bạn đã dùng thật trong sáu tháng gần nhất, rồi chọn gói theo đúng lượng dùng đó. Nhân viên nói「降档得到营业厅办」「合约期内不能改」thì bạn yêu cầu cơ sở bằng văn bản. Không có thì gọi 12300 để khiếu nại. - ### 5. Không mua xổ số - Chi phí: Không tốn tiền. Không mua là tiết kiệm tiền. @@ -48,7 +44,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Bộ Tài chính (财政部) (2015). Thông báo về việc quy định và tăng cường quản lý tỷ lệ cấu thành quỹ xổ số（财综〔2015〕94 号）. ；Bộ Tài chính (2025). Thông báo số 8 năm 2025 của Bộ Tài chính (tình hình huy động và phân bổ quỹ công ích xổ số năm 2024). ；Quy tắc trò chơi xổ số kép của Xổ số Phúc lợi Trung Quốc（第十四条）. - Ghi chú: Dài hạn trung bình chắc chắn lỗ, điều này do hệ thống đặt sẵn, không liên quan tới may rủi. Mua cũng được, nhưng chỉ nên coi như khoản chi giải trí mỗi lần 2 tệ, không phải khoản đầu tư. - ### 6. Thấy「高收益」「保本」「稳赚」thì đi luôn - Chi phí: Không tốn tiền. Chỉ cần dừng lại một chút lúc nghe mấy chữ đó. @@ -57,7 +52,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Quách Thụ Thanh (郭树清) (2018). Phòng ngừa và giải quyết rủi ro tài chính, nỗ lực vượt qua những cổng then chốt (bài phát biểu tại Diễn đàn Lujiazui lần thứ 10, trên trang Cục Quản lý và Giám sát Tài chính quốc gia):「在打击非法集资过程中，努力通过多种方式让人民群众认识到，高收益意味着高风险，收益率超过6%的就要打问号，超过8%的就很危险，10%以上就要准备损失全部本金」. ；Cục Quản lý Tài chính địa phương thành phố Thiên Tinh đăng lại. ；Ngân hàng Nhân dân Trung Quốc và cơ quan khác (2018). Ý kiến chỉ đạo về việc quy định công tác quản lý tài sản của các tổ chức tài chính（银发〔2018〕106 号，第二条、第六条）. ；Hội đồng Nhà nước (2021). Quy chế phòng ngừa và xử lý huy động tiền phi pháp（国务院令第 737 号，第二十五条）. - Ghi chú: Ba mốc 6%, 8%, 10% là người giám sát vạch theo kinh nghiệm, không phải điều luật. Câu nói đó được đưa ra khi nói về việc đánh triệt huy động tiền phi pháp, và nhắm vào sản phẩm quản lý tài sản và công ty đầu tư hứa trả cao. Không phải lời khuyên đầu tư. - ### 7. Không dùng khoản trả tối thiểu trên thẻ tín dụng, không mở trả góp hay vay tiêu dùng để mua - Chi phí: Không tốn tiền. Bạn phải để dành đủ tiền mặt trả toàn bộ trước ngày đến hạn. Chỗ khó là nhịn không dùng trả góp để dàn nhẹ một khoản chi lớn. @@ -66,6 +60,7 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: C - Nguồn: Ngân hàng Nhân dân Trung Quốc (中国人民银行) (2020). Thông báo về việc đẩy mạnh cải cách thị trường hoá lãi suất thấu chi thẻ tín dụng（银发〔2020〕327 号）. - Ghi chú: Con số 18.25% là nhân thẳng「万分之五 × 365」, chưa tính lãi kép. Nếu cộng dồn theo tháng thì còn cao hơn. Có ngân hàng cho chiết khấu, hãy theo lãi suất quy đổi cả năm ghi trên hoá đơn của chính bạn. Khi thật sự không xoay được dòng tiền, hãy tìm kênh chính quy có lãi suất thấp hơn để vay trước, đừng để khoản trả tối thiểu cộng dồn lãi. + ### 8. Không tặng thưởng người livestream, không nạp tiền vào game, không đặt hàng bốc đầu - Chi phí: Không tốn tiền. Dành mấy phút làm ba việc: tắt thanh toán không cần mật mã, đặt hạn mức cho từng giao dịch và cho từng ngày, xoá các phương thức thanh toán đã liên kết trong ứng dụng. Chỗ khó là nhịn ngay tại chỗ không bấm vào. @@ -74,7 +69,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: B - Nguồn: Cục Quản lý Báo chí và Xuất bản quốc gia (国家新闻出版署) (2024). Doanh thu thị trường game Trung Quốc năm nay vượt 3257 tỷ tệ, lập kỷ mới（《Báo cáo ngành game Trung Quốc năm 2024》, Nhóm công tác game của Hiệp hội Âm thanh, số và truyền hình Trung Quốc）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Ý kiến chỉ đạo về việc xử lý đúng đắn theo pháp luật một số vụ án dân sự liên quan dịch bệnh viêm phổi do vi rút corona (二)（法发〔2020〕17 号，第 9 条）. ；Cục Quản lý Báo chí và Xuất bản quốc gia (2021). Thông báo siết chặt quản lý, thực sự ngăn người chưa thành niên nghiện game trực tuyến（国新出发〔2021〕14 号）. - Ghi chú: Xếp mức B vì chưa ai đo trực tiếp「không tặng thưởng thì tiết kiệm được bao nhiêu」, chỉ có số liệu tổng của ngành game. Hai cách tắt thanh toán không cần mật mã và đặt hạn mức là kinh nghiệm thực hành, nếu tính riêng thì là mức C. Nguyên văn Điều 十九 và Điều 一百四十五 của Bộ luật Dân sự đã được dẫn nguyên văn ở mục 9 (con nạp tiền, tặng thưởng trên điện thoại). Không phải lời khuyên đầu tư. - ### 9. Con nạp tiền, tặng thưởng trên điện thoại: khoản chi lớn của trẻ từ tám tuổi trở lên, cha mẹ không chấp thuận thì có thể đòi lại - Chi phí: Không tốn tiền. Phải bỏ thời gian thu thập chứng cứ và thương lượng với nền tảng, nền tảng thường từ chối trước, hoàn tiền sau. Không thống nhất được thì chỉ còn đi kiện tụng, mà việc đó mất nhiều tháng trở lên, phí luật sư phải tự bỏ. Nên chuẩn bị đủ hồ sơ rồi đi qua nền tảng và 12315 là hơn. Chỗ khó là con không chịu nói, bạn phải kiên nhẫn hỏi cho rõ. @@ -83,7 +77,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự. Bản toàn văn do Viện kiểm sát nhân dân tối cao đăng lại. 第十九条：「八周岁以上的未成年人为限制民事行为能力人，实施民事法律行为由其法定代理人代理或者经其法定代理人同意、追认；但是，可以独立实施纯获利益的民事法律行为或者与其年龄、智力相适应的民事法律行为。」— người chưa thành niên từ tám tuổi trở lên là người có năng lực hành vi hạn chế, hành vi pháp lý dân sự phải do người đại diện theo pháp luật thực hiện, hoặc phải được người đại diện theo pháp luật đồng ý hay chấp thuận sau; nhưng có thể tự thực hiện hành vi pháp lý dân sự thuần túy mang lại lợi ích, hoặc hành vi pháp lý dân sự tương xứng với tuổi và trình độ của mình. 第二十条：「不满八周岁的未成年人为无民事行为能力人，由其法定代理人代理实施民事法律行为。」— người chưa thành niên dưới tám tuổi là người không có năng lực hành vi, hành vi pháp lý dân sự do người đại diện theo pháp luật thực hiện thay. 第一百四十五条：「限制民事行为能力人实施的纯获利益的民事法律行为或者与其年龄、智力、精神健康状况相适应的民事法律行为有效；实施的其他民事法律行为经法定代理人同意或者追认后有效。」「法定代理人未作表示的，视为拒绝追认。」— hành vi pháp lý dân sự thuần túy mang lại lợi ích, hoặc tương xứng với tuổi, trình độ và tình trạng tâm thần của người có năng lực hành vi hạn chế thì có hiệu lực; các hành vi pháp lý dân sự khác phải được người đại diện theo pháp luật đồng ý hoặc chấp thuận sau mới có hiệu lực. Người đại diện theo pháp luật không tỏ thái độ thì coi như từ chối chấp thuận sau. - Ghi chú: Chỗ cốt lõi là chứng cứ: bạn phải đưa ra được bằng chứng chứng minh thao tác trên điện thoại là trẻ, không phải bạn. Tài liệu dùng được gồm: thời điểm nạp tiền tập trung vào giờ học hoặc giờ cha mẹ đi làm, lịch sử dùng thiết bị, lịch sử chat và lịch sử chơi game của trẻ, và bản giải thích của chính trẻ.「与年龄智力相适应」không có một mức tiền chung, một skin game vài chục tệ và một lần tặng thưởng mấy vạn tệ không phải một việc. Phòng trước còn dễ chịu hơn: đặt trần thanh toán không cần mật mã, bật chế độ thanh thiếu niên, đừng đưa mật mã thanh toán cho trẻ. Trường hợp bạn tự tặng thưởng, nạp tiền khi nóng đầu, xem mục「không tặng thưởng người livestream, không nạp tiền vào game, không đặt hàng bốc đầu」trong phần này. - ### 10. Dạy trẻ nhớ một câu: bất cứ ai bảo trẻ cầm điện thoại của cha mẹ ra thao tác, hoặc đọc mã xác minh, đều là kẻ lừa - Chi phí: Không tốn tiền. Kể cho trẻ nghe một lượt sáu loại lừa đảo bên dưới, rồi thỏa thuận với trẻ một câu「bị lừa rồi thì nói với con trước, con không mắng». Đồng thời đặt trần thanh toán không cần mật mã, bật chế độ thanh thiếu niên, không đưa mật mã thanh toán cho trẻ. Chỗ khó là sau khi bị lừa thật thì nhịn không nổi cơn giận. @@ -92,15 +85,15 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: B - Nguồn: Trung tâm chống lừa đảo thành phố Quanzhou (泉州市反诈骗中心) (2025). Bài học đầu năm học｜Chống lừa đảo là môn bắt buộc của thanh niên, 6 cái bẫy lớn này nhất định phải tránh!（Cục Công an tỉnh Phúc Kiên đăng lại） - Ghi chú: Xếp mức B vì nguồn chính thức đưa ra là phần tháo gở lời lẻ của kẻ lừa và các ca lẻ, không có tỷ lệ vụ án, cũng không có số liệu thống kê về hiệu quả phòng ngừa. Phải tách khỏi mục 9 (con nạp tiền, tặng thưởng trên điện thoại): tiền trẻ tự nạp, tự tặng thưởng thì có thể đòi hoàn theo Bộ luật Dân sự. Tiền bị kẻ lừa lấy thì điều đó không áp dụng, chỉ có thể báo công an để chặn chuyển tiền theo mục 2 ở phần 8. Có đòi lại được không thì tuỳ tiền còn nằm trong tài khoản hay không. Câu dọa cốt lõi của kẻ lừa với trẻ là「con vi phạm rồi, cha mẹ con sẽ vào tù」. Nói rõ điều này với trẻ từ trước, bản thân câu đó đã là dấu hiệu của lừa đảo, hữu ích hơn là hỏi lại sau. Câu「bị lừa thì nói trước, con không mắng」phải thực sự làm được. Trẻ sợ bị mắng nên giấu không nói, đó là nguyên nhân chung khiến thiệt hại trong các ca này càng lăn lóc càng lớn. Bản rộng hơn của kiểu giả danh người quen, và hoán đổi khuôn mặt bằng AI, xem mục 3 và 4 ở phần 8（quy tắc cứng chống lừa đảo, có chuyển tiền thì treo máy）. - -### 11. Từ chối bảo hành mở rộng cho thiết bị điện tử +### 11. Từ chối gói bảo hành mở rộng (延保) cho thiết bị điện tử -- Chi phí: Không tốn tiền. Cái giá phải trả là sau khi hết thời hạn bảo hành, nếu phải sửa thì bạn tự bỏ tiền, nhưng chuyện này xác suất không lớn. -- Nói thẳng: Bảo hành mở rộng thường bán giá bằng 10% đến 50% giá gốc hàng hoá. Người bán bảo hành mở rộng kiếm hơn hàng thường mười mấy lần. Best Buy của Mỹ năm 2003 đã kiếm được hơn một nửa lợi nhuận nhờ bảo hành mở rộng, trong khi bảo hành mở rộng chỉ chiếm 3% đến 4% doanh thu. Chênh lệch đó chính là số tiền bạn trả thừa. -- Lợi ích: Bảo hành mở rộng thường được định giá bằng 10% đến 50% giá gốc hàng hoá. Ước tính ngành được nghiên cứu dẫn lại cho biết biên lợi nhuận gộp trung bình của bảo hành mở rộng là 50% đến 60%: bán ra 100 tệ thì 50 đến 60 tệ là lãi. Xấp xỉ gấp 18 lần biên lợi nhuận gộp của hàng thường. Năm 2003, bảo hành mở rộng chỉ chiếm 3% đến 4% doanh thu của Best Buy nhưng đóng góp hơn 50% lợi nhuận (Mỹ, số liệu bán lẻ thập niên 2000) +- Chi phí: Không tốn tiền. Cái giá phải trả là sau khi hết thời hạn bảo hành, nếu phải sửa thì bạn tự bỏ tiền, nhưng chuyện đó ít xảy ra. +- Nói thẳng: Gói bảo hành mở rộng thường được bán bằng 10% đến 50% giá gốc. Người bán bảo hành mở rộng kiếm hơn mười lần bình thường so với bán hàng thường, khoản chênh lệch đó chính là số tiền bạn phải trả thêm. Những bộ phận chính như động cơ máy giặt, tủ lạnh, nhà nước vốn đã quy định bảo hành 3 năm. Gặp gói bảo hành mở rộng rẻ bất thường, lấy giá gói chia cho số tiền một lần sửa, kết quả là xác suất hỏng phải vượt bao nhiêu trong mấy năm đó thì mới hoà vốn. +- Lợi ích: Giá của gói bảo hành mở rộng thường bằng 10% đến 50% giá gốc của sản phẩm. Ước tính của ngành được nghiên cứu dẫn lại nói biên lợi nhuận gộp trung bình của gói bảo hành mở rộng là 50% đến 60%: bán ra 100 tệ thì có 50 đến 60 tệ là lãi. Khoảng này gấp khoảng 18 lần biên lợi nhuận gộp của hàng thường. Năm 2003, bảo hành mở rộng chỉ chiếm 3% đến 4% doanh thu của Best Buy, nhưng mang về hơn 50% lợi nhuận (Hoa Kỳ, số liệu bán lẻ thập niên 2000). Gặp gói bảo hành mở rộng có giá thấp hơn nhiều so với khoảng đó, bạn tự tính một phép: giá gói bảo hành mở rộng ÷ số tiền một lần sửa = tỷ lệ hỏng cần có để hoà vốn. Chỉ khi bạn ước tính xác suất hỏng ít nhất một lần trong mấy năm bảo hành lớn hơn con số đó thì mới đáng mua. Ví dụ gói bảo hành mở rộng giá 25 tệ, giả sử một lần sửa tốn 300 tệ, xác suất hỏng ít nhất một lần phải vượt khoảng 8% thì mới hoà vốn. Con số 300 tệ ở đây chỉ là ví dụ, không phải số liệu thống kê. Tính thì trừ trước phần nhà nước đã quy định bảo hành. Thời hạn ba bảo đảm theo quy định nhà nước tính từ ngày kê hoá đơn: máy giặt và tủ lạnh gia dụng 1 năm cho toàn máy, 3 năm cho bộ phận chính. Bộ phận chính của máy giặt là động cơ, bộ hẹn giờ, bộ điều khiển chương trình và tụ điện; của tủ lạnh là máy nén, bộ ổn định nhiệt và những bộ phận khác. Phần bảo hành mở rộng trùng với thời hạn ba bảo đảm là số tiền mua lại thứ vốn đã có (toàn quốc, từ 1995) - Mức bằng chứng: B -- Nguồn: Chen T, Kalra A, Sun B (2009). Why Do Consumers Buy Extended Service Contracts? Journal of Consumer Research 36(4):611-623. -- Ghi chú: Con số biên lợi nhuận gộp đó là bài báo dẫn lại ước tính của ngành (Business Week 2004), không phải bài báo tự đo. Cách tính ở trên áp dụng cho người bình thường tự bỏ được phí sửa một lần. Người sống bằng một thiết bị mà không có máy dự phòng thì phải tính riêng. Bảo hiểm vỡ màn hình điện thoại cũng phải tính riêng: trước hết ước lượng bao lâu bạn làm rơi một lần, rồi xem tiền phí bảo hiểm có đáng không. +- Nguồn: Chen T, Kalra A, Sun B (2009). Why Do Consumers Buy Extended Service Contracts? Journal of Consumer Research 36(4):611-623. ；Ủy ban Kinh tế và Thương mại quốc gia cùng các cơ quan khác (国家经贸委等) (1995). Quy định trách nhiệm sửa chữa, thay thế, hoàn trả đối với một phần hàng hóa (部分商品修理更换退货责任规定)（国经贸〔1995〕458 号, Điều 8: thời hạn ba bảo đảm tính từ ngày kê hoá đơn; phụ lục《Danh mục một phần hàng hóa thực hiện ba bảo đảm (lô thứ nhất)》máy giặt, tủ lạnh gia dụng 1 năm, bộ phận chính 3 năm). Văn bản quy phạm hành chính của Cục Quản lý thị trường quốc gia (国家市场监督管理总局). +- Ghi chú: Con số biên lợi nhuận gộp đó là bài báo dẫn lại ước tính của ngành (Business Week 2004), không phải bài báo tự đo. Công thức ở trên áp dụng cho người bình thường tự bỏ tiền được một lần sửa. Người kiếm ăn bằng một máy mà không có máy dự phòng thì phải tính riêng. Loại bảo hiểm kính vỡ điện thoại cũng phải tính riêng: trước hết ước tính bao lâu mình làm rơi một lần, rồi xem tiền phí có đáng không. Gói bảo hành mở rộng rất rẻ thì mua cũng chỉ lỗ chừng mấy chục tệ. Trước khi mua, nhìn rõ ba điều: tính từ ngày nào (tính từ ngày bạn mua, mấy năm đầu trùng với thời hạn ba bảo đảm), bên nào bảo hành là người bán, thương hiệu hay công ty bên thứ ba, phí ra tận nhà và phí thay linh kiện có tính riêng hay không. Cũng nên xem hướng dẫn sử dụng và thẻ bảo hành, thời hạn bảo hành mà chính thương hiệu cam kết có thể dài hơn thời hạn ba bảo đảm. + ### 12. Ưu tiên thuốc generic đã đạt đánh giá tính nhất quán và thuốc trong gói đấu thầu tập trung @@ -118,7 +111,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2021). Ý kiến chỉ đạo về việc xây dựng và hoàn thiện cơ chế bảo đảm thanh toán chung cho khám bệnh ngoại trú của bảo hiểm y tế cơ bản người lao động（国办发〔2021〕14 号）第（五）项：「个人账户主要用于支付参保人员在定点医疗机构或定点零售药店发生的政策范围内自付费用。可以用于支付参保人员本人及其配偶、父母、子女在定点医疗机构就医发生的由个人负担的医疗费用，以及在定点零售药店购买药品、医疗器械、医用耗材发生的由个人负担的费用。探索个人账户用于配偶、父母、子女参加城乡居民基本医疗保险等的个人缴费。个人账户不得用于公共卫生费用、体育健身或养生保健消费等不属于基本医疗保险保障范围的支出。」— tài khoản cá nhân chủ yếu dùng để trả phần chi phí trong phạm vi chính sách do người tham gia bảo hiểm tự bỏ ra tại cơ sở khám chữa bệnh tuyến chỉ định hoặc hiệu thuốc tuyến chỉ định. Có thể dùng để trả phần chi phí do người tham gia bảo hiểm và vợ chồng, cha mẹ, con cái tự bỏ ra khi khám chữa bệnh tại cơ sở tuyến chỉ định, cũng như phần tiền tự bỏ ra khi mua thuốc, thiết bị y tế, vật tư y tế tại hiệu thuốc tuyến chỉ định. Cần thử nghiệm dùng tài khoản cá nhân cho khoản đóng bảo hiểm y tế cơ bản cư dân đô thị và nông thôn của vợ chồng, cha mẹ, con cái. Tài khoản cá nhân không được dùng cho chi phí y tế cộng đồng, tập thể dục, hay các khoản chi tiêu dùng chăm sóc sức khoẻ không thuộc phạm vi bảo hiểm. - Ghi chú: Rất nhiều người có vài nghìn tệ nằm trong tài khoản cá nhân, trong khi người cao tuổi trong nhà tự bỏ tiền mua thuốc. Liên kết một lần thì có hiệu lực dài hạn. Hầu hết các nơi gọi là「家庭共济」, làm được ngay trong app bảo hiểm y tế địa phương. Cần chú ý một vạch đỏ ở chiều ngược lại: tài khoản cá nhân không dùng cho các khoản chi như tập thể dục, chăm sóc sức khoẻ. Mang nó tới hiệu thuốc mua thực phẩm chức năng để rút tiền mặt là gian lận bảo hiểm. - ### 14. Uống nước máy đã đun sôi, không mua nước đóng chai quanh năm - Chi phí: Mua một ẩm. Mỗi lần đun nước tốn một ít thời gian, cộng thêm chút tiền khí đốt hoặc tiền điện. @@ -128,14 +120,14 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Nguồn: Ủy ban Phát triển và Cải cách thành phố Bắc Kinh. Bảng giá nước sinh hoạt dân dụng. ；Cục Quản lý thị trường quốc gia (国家市场监督管理总局), Hệ thống công khai toàn văn tiêu chuẩn quốc gia. GB 5749-2022 Tiêu chuẩn vệ sinh nước uống sinh hoạt. ；Nhóm điều tra liên hợp thành phố Hàng Châu (2025). Thông báo về kết quả điều tra tình hình cấp nước bất thường ở một số khu chung cư quận Dư Hàng của thành phố Hàng Châu (ngày 23 tháng 7 năm 2025), Tân Hoa Xinh đăng lại ；Chính quyền quận Dư Hàng thành phố Hàng Châu. Dữ liệu giám sát chất lượng nước nguồn quận Dư Hàng tháng 4／5／6 năm 2025（Tập đoàn Quản lý Cấp nước Dư Hàng Hàng Châu công bố, đơn vị kiểm nghiệm là Công ty TNHH Giám sát Môi trường Gia Kiếm Hàng Châu, bảng PDF phụ lục 1, mục 24「粪大肠菌群」） 、、；Cục An ninh mạng và An ninh mạng của Bộ Công an (公安部网络安全保卫局) (2025). Bịa「thông báo tình huống công an」giả, dựng tin đồn「nước phân」, Thao Tư Hào bị bị giữ - Ghi chú:Tranh cãi. Cái gọi là「đạt chuẩn」là kiểm tra nước ra khỏi nhà máy và nước ở cuối đường ống. Khi nguồn nước hoặc mạng lưới có sự cố, chỉ tiêu này mất kiểm soát trong thời gian ngắn. Ngày 16 tháng 7 năm 2025, nước thô dẫn vào nhà máy nước Nhân Hợp quận Dư Hàng có mùi bất thường, nước máy ở một số khu chung cư Nhân Hợp và Lưng Chủ có mùi thối. Kết luận chính thức là tảo phân huỷ trong điều kiện thiếu ô-xi, sinh ra các chất sulfide. Sau khi đổi nguồn nước, không phát hiện tổng số vi khuẩn đường ruột và Escherichia coli trong nước ra khỏi nhà máy, mọi hạng mục đều đạt tiêu chuẩn nước uống sinh hoạt GB 5749-2022. Sau đó có 7 cán bộ bị kỷ luật, tiền nước tháng 7 miễn toàn bộ, còn được đổi miếng lọc miễn phí. Nhưng cùng nguồn nước đó, chỉ số vi khuẩn đường ruột chí ỉ đã đi lên liên tục mấy tháng trước sự cố. Cửa rút nước Dư Hành Vĩnh Thắng tháng 4 là 10 cái/L, tháng 5 là 7300 cái/L, tháng 6 là 16000 cái/L. Giới hạn của loại III trong tiêu chuẩn nước mặt GB 3838 là 10000 cái/L. Cùng tháng, hồ Tứ Lĩnh 28000 cái/L, hồ Mã Đầu Sơn 24000 cái/L cũng vượt. Đun sôi chỉ diệt được vi sinh vật, không bỏ được chất tạo mùi, cũng không bỏ được chất hoá chất hoà tan trong nước. Ngửi thấy mùi lạ, hoặc nước vàng hay đục, hãy dừng uống, báo cho doanh nghiệp cấp nước và 12345, tạm chuyển sang uống nước đóng chai, đừng trông đợi đun thêm mấy phút để áp chế. Nhưng đây là sự cố cục bộ, xảy ra thỉnh thoảng, địa phương cũng đã bồi thường theo từng đợt, cho nên vẫn không thành lý do mua nước đóng chai quanh năm. Tin trên mạng「粪水管接错」là thông báo tình huống công an bịa ra, người đăng đã bị bị giữ (xem phần 9). Giá 1 tệ 500 ml của nước đóng chai là giả định để tính sổ, hãy tính lại theo giá thật ở thành phố của bạn. Khu chung cư cũ có cơ sở cấp nước lần hai không đạt, có thể lắp thêm bộ lọc, hoặc chỉ mua nước uống trong trường hợp đó. Giá nướmỗi nơi một khác, số liệu Bắc Kinh chỉ dùng để nhìn mức độ. -### 15. Không giao dịch cổ phiếu thường xuyên +### 15. Đừng giao dịch cổ phiếu liên tục, thị trường lên rầm rộ rồi đổ sập thì càng phải ít ra tay hơn -- Chi phí: Không tốn tiền. Chỗ khó là thấy tin là muốn vào ngay, phải nhịn lại. -- Nói thẳng: Người mua bán cổ phiếu thường xuyên nhất, trung bình mỗi năm chỉ kiếm 11.4%, cùng thời gian đó toàn thị trường là 17.9%. Mấy phần trăm chênh lệch đó gần như bị phí môi giới trả lúc mua bán ăn mất. Nếu không tính phí, người giao dịch thường và người giao dịch ít kiếm được gần như nhau. -- Lợi ích: Nghiên cứu này xem 66.465 tài khoản gia đình tại nhà môi giới chiết khấu Mỹ, trong giai đoạn 1991 đến 1996. Nhóm giao dịch thường xuyên nhất có lợi nhuận quy đổi cả năm 11.4%, tức trung bình mỗi năm kiếm 11.4%. Cùng thời gian đó toàn thị trường là 17.9%, bình quân một hộ gia đình là 16.4%. Bình quân tỷ lệ luân chuyển cổ phiếu của một hộ trong một năm là 75%, tức trong một năm bán đi ba phần tư số cổ phiếu đang cầm. Nhóm giao dịch thường xuyên và nhóm ít giao dịch chênh lệch rất nhỏ về lợi nhuận trước phí, nên khoảng cách chủ yếu đến từ chi phí giao dịch phải trả lúc mua bán (Mỹ, 1991 đến 1996) +- Chi phí: Không tốn tiền. Chỗ khó là thấy tin là muốn ra tay ngay, thấy người ta đang kiếm tiền thì muốn tăng vốn theo, phải nhịn được. +- Nói thẳng: Những hộ gia đình Mỹ giao dịch gần như liên tục, mỗi năm chỉ kiếm được 11.4%, cùng thời gian đó toàn bộ thị trường là 17.9%, khoảng cách gần như toàn bộ là phí giao dịch. Trong vòng lên rầm rộ rồi đổ sập năm 2014 đến 2015 ở Trung Quốc, 85% hộ gia đình có tài khoản nhỏ nhất giao dịch quá thường xuyên. Họ kiếm được ít hơn 250 tỷ tệ so với nhóm cứ cầm không bán, khoảng ba phần mười số vốn. +- Lợi ích: Nghiên cứu này xem 66.465 tài khoản gia đình tại các công ty chứng khoán chiết khấu ở Mỹ, trong giai đoạn 1991 đến 1996. Nhóm giao dịch gần như liên tục có lợi nhuận hằng năm 11.4%, tức là trung bình mỗi năm kiếm được 11.4%. Cùng thời gian đó toàn bộ thị trường là 17.9%, một hộ gia đình bình quân là 16.4%. Tỷ lệ luân chuyển bình quân của một hộ gia đình trong một năm là 75%, tức trong một năm đã đổi mất ba phần tư số cổ phiếu đang cầm. Nhóm giao dịch liên tục và nhóm không giao dịch liên tục chỉ khác nhau rất ít về lợi nhuận gộp (tiền kiếm được trước khi trừ phí), nên khoảng cách chủ yếu đến từ chi phí giao dịch phải trả lúc mua bán (Hoa Kỳ, 1991 đến 1996). Dữ liệu trong nước: nghiên cứu dùng dữ liệu tài khoản theo từng ngày của Sở Giao dịch chứng khoán Thượng Hải, phủ gần như toàn bộ khoảng 40 triệu tài khoản. Thời gian lấy mẫu là từ tháng 7 năm 2014 đến tháng 12 năm 2015, tổng cộng 18 tháng. Trong 18 tháng đó chỉ số chung của sàn Thượng Hải tăng hơn 150% trước đã. Nó đạt đỉnh 5166.35 điểm ngày 12 tháng 6 năm 2015, rồi lại giảm 40% vào cuối tháng 12. Các tài khoản gia đình chia thành bốn nhóm theo giá trị cổ phiếu lúc đầu kỳ. Nhóm nhỏ nhất chiếm 85% tổng số tài khoản gia đình, nhóm lớn nhất chiếm 0.5%. Tổng giá trị cổ phiếu lúc đầu kỳ của hai nhóm gần bằng nhau, lần lượt là 880 tỷ tệ và 808 tỷ tệ. Kết quả là 85% nhóm nhỏ nhất mất 250 tỷ tệ do chủ động giao dịch, còn 0.5% nhóm lớn nhất kiếm được 254 tỷ tệ. Ở đây「tổn thất」là so với việc cứ nắm giữ suốt cùng giai đoạn thị trường đó, không làm gì cả. Mỗi con số khoảng bằng ba phần mười giá trị cổ phiếu lúc đầu kỳ của nhóm tương ứng, con số trước là 28%, sau là 31%. Để đối chiếu, hai năm rưỡi trước tháng 6 năm 2014 thị trường khá yên. Trong thời gian đó, mười tám tháng bất kỳ thì khoảng cách cùng loại chỉ bằng 1% đến 3% giá trị lúc đầu kỳ. Trong vòng thị trường này, tài khoản gia đình trung bình khoảng ba tuần đã đổi một lượt số cổ phiếu đang cầm, gần 18 lần một năm (Trung Quốc, Thượng Hải, 2014 đến 2015; thời kỳ đối chiếu 2012 đến 2014) - Mức bằng chứng: A -- Nguồn: Barber BM, Odean T (2000). Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors. The Journal of Finance 55(2):773-806. -- Ghi chú: Số liệu này là của Mỹ thập niên 1990. Cơ cấu hoa hồng và thuế đóng cửa của cổ phiếu A khác, nhưng「luân chuyển càng nhiều, lợi nhuận càng thấp」thì vẫn đúng. Không phải lời khuyên đầu tư. +- Nguồn: Barber BM, Odean T (2000). Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors. The Journal of Finance 55(2):773-806. ；An L, Lou D, Shi D (2022). Wealth redistribution in bubbles and crashes. Journal of Monetary Economics 126:134-153. ；bản thảo bài làm việc do cùng các tác giả công bố (tháng 10 năm 2021, các số liệu trong nước đối chiếu theo đúng bản đó). +- Ghi chú: Số liệu Mỹ là của thập niên 1990, cơ cấu hoa hồng và thuế đóng dấu của cổ phiếu A khác, nhưng nguyên lý「luân chuyển càng nhiều, lợi nhuận càng thấp」 là như nhau. Nghiên cứu trong nước không nói là đừng mua cổ phiếu. Nó tính rằng trong cùng một giai đoạn thị trường, phía hộ gia đình tài khoản nhỏ càng chủ động giao dịch thì tiền càng chảy sang bên đối tác. Số tiền đó không vào túi tổ chức mà vào túi cá nhân có tài khoản lớn nhất: trong thời gian này khoảng chín phần mười khối lượng giao dịch của sàn Thượng Hải đến từ tài khoản nhà đầu tư lẻ. Câu「chín phần mười nhà đầu tư lẻ lỗ」lưu truyền trên mạng không có số liệu chính thức nào để đối chiếu, mục này không dùng con số đó. Tỷ lệ luân chuyển có thể đối chiếu: trong vòng thị trường này tài khoản gia đình trong nước đổi gần 18 lần một năm, còn nhóm gia đình ở Mỹ chỉ đổi mất ba phần tư trong một năm. Ở những năm thị trường yên, khoảng cách này trong nước chỉ khoảng một phần trăm đến ba phần trăm, nên lúc thị trường nóng nhất mới là lúc cần nhịn nhất. Không phải lời khuyên đầu tư. ### 16. Không vay tiền để đầu tư, không dùng đòn bẩy, không mua thứ mình không hiểu @@ -145,7 +137,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: B - Nguồn: Ủy ban Chứng khoán Trung Quốc (中国证监会) (2015). Biện pháp quản lý nghiệp vụ tài chính và chứng khoán của công ty chứng khoán（证监会令第 117 号，第十二条）. ；Sở Giao dịch chứng khoán Thượng Hải. Giải thích《Quy chế thi hành giao dịch tài chính và chứng khoán của Sở Giao dịch chứng khoán Thượng Hải》(tài liệu giáo dục nhà đầu tư). - Ghi chú: Câu「không hiểu thì không mua」không có tài liệu nào chứng minh trực tiếp, thuộc đồng thuận, nếu tính riêng thì là mức C. Dùng vay tiêu dùng hoặc rút tiền từ thẻ tín dụng để đầu tư cũng là đòn bẩy hình thức khác, lãi suất của chúng xem mục 7 của phần này (không dùng khoản trả tối thiểu trên thẻ tín dụng). Không phải lời khuyên đầu tư. - ### 17. Thay quỹ quản lý chủ động bằng quỹ chỉ số phạm vi rộng cho phần giữ dài hạn - Chi phí: Không tốn tiền. Cái giá phải trả là bạn từ bỏ khả năng chọn trúng quản lý quỹ nổi tiếng. Chỗ khó là khi chỉ số giảm, bạn cũng phải chịu chung, không bỏ chạy giữa chừng. @@ -162,7 +153,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: C - Nguồn: Ủy ban Chứng khoán Trung Quốc (中国证监会) (2023). Lãnh đạo các bộ phận liên quan của Ủy ban Chứng khoán trả lời phóng viên về cải cách phí quỹ công. ；Cục Chứng khoán Xiamen (厦门证监局) (2024).「Quỹ lợi ích chủ động mới đăng ký thống nhất thực hiện trần "mức phí quản lý không quá 1.2%, mức phí lưu ký không quá 0.2%"」. ；Fama EF, French KR (2010). Xem mục 11. - Ghi chú: Công thức này chỉ làm theo công thức, với giả định hai quỹ kiếm được như nhau trước khi trừ phí. Thực tế còn phải tính thêm phí mua, phí bán lại, và phí dịch vụ bán hàng của lô cổ phần loại C. Không phải lời khuyên đầu tư. - ### 19. Đừng đặt hết tiền vào một cổ phiếu, một nền tảng, một căn nhà - Chi phí: Không tốn tiền. Chỗ khó là phải chấp nhận một điều: bạn sẽ không đúng lúc nào đặt trúng vào thứ đáng nhất. @@ -180,7 +170,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Bộ Tài chính, Tổng cục Thuế (财政部、税务总局) (2024). Thông báo thực hiện chính sách ưu đãi thuế thu nhập cá nhân cho lương hưu cá nhân trên toàn quốc（Thông báo số 21 năm 2024 của Bộ Tài chính và Tổng cục Thuế, ngày 12 tháng 12 năm 2024）. ；Bộ Nhân sự và Xã hội cùng năm bộ khác (2022). Biện pháp thực hiện lương hưu cá nhân（第七、八、九、十二、十三条）. - Ghi chú: Tài khoản này「vận hành khép kín」, tiền gửi vào thì không rút ra được. Chỉ có bốn trường hợp được nhận: đã tới tuổi nhận lương hưu cơ bản, mất hoàn toàn năng lực lao động, ra nước ngoài định cư, và các trường hợp khác theo quy định của nhà nước. Vì thế nó không phải khoản tích luỹ, mà là một sắp xếp thuế phải khoá nhiều năm. Hãy tính rõ mỗi phần thu nhập bạn kiếm thêm thì phải nộp bao nhiêu thuế, rồi quyết định gửi bao nhiêu. Đừng vì「móc lông vịt」mà khoá cả khoản dự phòng vào (khoản dự phòng xem mục 27). Tiền trong tài khoản còn phải do bạn tự chọn sản phẩm để đầu tư, và có thể lỗ. Cách chọn sản phẩm giống mục 17 và 18 của phần này (quỹ chỉ số phạm vi rộng, chọn quỹ có mức phí thấp). Chứng từ khấu trừ do Nền tảng Quản lý Thông tin Lương hưu cá nhân cấp. Bạn chọn được thời điểm khấu trừ: một là khấu trừ ngay lúc công ty trả lương và thu thuế, cách khác là khấu trừ khi tổng kết vào năm sau (xem mục 2). - ### 21. Phòng tập thì trả theo lượt hoặc theo chu kỳ ngắn, trừ khi bạn đã có hơn một năm đi đều đặn - Chi phí: Mua theo lượt thì giá mỗi lượt đắt hơn một chút so với chia ra thẻ cả năm. Bạn phải thành thật ước lượng mình mỗi tháng đi được mấy lần. @@ -189,7 +178,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: B - Nguồn: DellaVigna S, Malmendier U (2006). Paying Not to Go to the Gym. American Economic Review 96(3):694-719. - Ghi chú: Đây chỉ là một nghiên cứu, dùng số liệu Mỹ. Nhưng cái đằng sau thì rất phổ biến: người ta ước lượng cao hơn sức tự kiểm của mình trong tương lai. Nếu bạn đã có một năm liền đi từ hai buổi mỗi tuần trở lên, thì thẻ cả năm đúng là rẻ hơn. - ### 22. Trả tiền trước phải ký hợp đồng bằng văn bản; người bán có rủi ro phải ngừng nhận tiền, trước khi bỏ chạy bạn có quyền đòi lại số dư chưa tiêu - Chi phí: Không tốn tiền. Chỗ khó là kiên quyết đòi một bản hợp đồng bằng văn bản tại chỗ, không chấp nhận「nạp tiền trước đã」. @@ -198,7 +186,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2024). Quy chế thi hành Bộ luật Bảo vệ quyền lợi người tiêu dùng của Cộng hoà nhân dân Trung Quốc（国务院令第 778 号，2024 年 7 月 1 日施行）. 第二十二条：「经营者以收取预付款方式提供商品或者服务的，应当与消费者订立书面合同，约定商品或者服务的具体内容、价款或者费用、预付款退还方式、违约责任等事项。」「经营者未按照约定提供商品或者服务的，应当按照消费者的要求履行约定或者退还预付款。」「经营者出现重大经营风险，有可能影响经营者按照合同约定或者交易习惯正常提供商品或者服务的，应当停止收取预付款。经营者决定停业或者迁移服务场所的，应当提前告知消费者……消费者……有权要求经营者继续履行提供商品或者服务的义务，或者要求退还未消费的预付款余额。」— khi người bán cung cấp hàng hoá hoặc dịch vụ theo cách nhận tiền trả trước thì phải lập hợp đồng bằng văn bản với người tiêu dùng, thỏa thuận nội dung cụ thể của hàng hoá hay dịch vụ, giá hay phí, cách hoàn tiền trả trước, trách nhiệm khi vi phạm và các việc liên quan. Người bán không cung cấp hàng hoá hay dịch vụ đúng thỏa thuận thì phải theo yêu cầu của người tiêu dùng mà thực hiện, hoặc hoàn lại tiền trả trước. Người bán gặp rủi ro kinh doanh lớn, có thể ảnh hưởng tới việc cung cấp hàng hoá hay dịch vụ bình thường theo thỏa thuận hợp đồng hoặc thói quen giao dịch, thì phải ngừng nhận tiền trả trước. Người bán quyết định đóng cửa hoặc chuyển địa điểm phục vụ thì phải báo trước cho người tiêu dùng…… người tiêu dùng…… có quyền yêu cầu người bán tiếp tục nghĩa vụ cung cấp hàng hoá hay dịch vụ, hoặc yêu cầu hoàn lại số dư tiền trả trước chưa tiêu；Điều 五十 về chế tài phạt. - Ghi chú: Quy chế này cho bạn cơ sở để kiện tụng sau, không phòng ngừa được trước. Người bán mà bỏ chạy thật thì tiền nhiều khả năng đã không còn nằm trong tài khoản. Vì thế hãy làm theo mục「phòng tập trả theo lượt hoặc theo chu kỳ ngắn」trong phần này, nạp ít vào hơn, gặp chuyện thật thì dựa vào quy chế này mà đòi quyền. Cùng quy chế đó, Điều 二十七 còn quy định: người bán nhận tiền đặt cọc phải thỏa thuận trước về cách thức, thủ tục và thời hạn hoàn lại, không được đặt điều kiện hoàn tiền phi lý. - ### 23. Món mua lớn không thiết yếu thì đặt thời gian chờ, mua hàng trên mạng thì dùng luôn quyền trả hàng trong bảy ngày không cần lý do - Chi phí: Không tốn tiền. Thứ bạn muốn sẽ phải chờ thêm một ngày mới có. Phí vận chuyển khi trả hàng thường do bạn tự bỏ. Chỗ khó là nhịn không đặt hàng ngay trong ngày. @@ -207,7 +194,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: C - Nguồn: Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2020, sửa đổi). Biện pháp tạm thời về trả hàng trong bảy ngày không cần lý do đối với hàng mua trên mạng（总局令第 31 号）. - Ghi chú: Bản thân「thời gian chờ」là kinh nghiệm thực hành, không có tài liệu nào chứng minh trực tiếp. Khi thời gian chờ không giữ được, còn quyền trả hàng trong bảy ngày không cần lý do. Trước khi trả hàng, xác nhận hai việc: món hàng đó có nằm trong danh sách không được trả không, và bao bì còn nguyên vẹn không. - ### 24. Đừng tích trữ hàng vì giá gạch chéo và đợt khuyến mãi lớn - Chi phí: Không tốn tiền. Cái giá phải trả là có những mức giảm giá thật bạn cũng bỏ lỡ. Bạn cần nhớ giá thường ngày của mấy món mình hay mua. Chỗ khó là ngày khuyến mãi lớn nhịn không tích trữ. @@ -224,7 +210,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: C - Nguồn: Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (中国银保监会) (2022). Quy tắc công khai thông tin sản phẩm bảo hiểm nhân thọ trên một năm（银保监规〔2022〕24 号）. - Ghi chú:Tranh cãi. Với người hoàn toàn không tiết kiệm được tiền thì bảo hiểm loại hoàn tiền bắt họ đóng đúng hạn, tác dụng đó có giá trị thực tế. Lời khuyên ở trên nhắm vào người bình thường tự kiểm mình được.「Mua loại chỉ bảo vệ, lấy tiền phí tiết kiệm đi tự đầu tư」tốt hơn「mua loại hoàn tiền」bao nhiêu, phải xem từng sản phẩm, phần này không có số liệu hệ thống nào để dẫn. Không phải lời khuyên đầu tư. - ### 26. Mua đủ bảo hiểm trách nhiệm dân sự bên thứ ba: hạn mức bảo hiểm bắt buộc thống nhất toàn quốc và không cao, phần vượt ra do túi bạn trả - Chi phí: Bảo hiểm trách nhiệm dân sự bên thứ ba thương mại, mỗi năm vài trăm tới hơn nghìn tệ. Mức bảo hiểm mua càng cao thì tiền phí tính trên mỗi đơn vị bảo hiểm càng rẻ. @@ -242,7 +227,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: C - Nguồn: Hội đồng Nhà nước (国务院) (2015). Quy chế bảo hiểm tiền gửi（国务院令第 660 号，第五条）. - Ghi chú: Khoảng「3 đến 6 tháng」là kinh nghiệm thông dụng của mọi người, không tìm được nguồn gốc chính thức hay học thuật, nên xếp mức C. Số tiền tính theo chi phí cố định hằng tháng của bạn, không tính theo thu nhập. Không phải lời khuyên đầu tư. - ### 28. Trước khi trả nợ vay mua nhà sớm, làm một bài so sánh, đừng theo cảm tính - Chi phí: Không tốn tiền. Dành nửa tiếng để tính một bảng. Bạn phải thành thật ước lượng mức sinh lời ổn định sau thuế mà mình lấy được. @@ -251,7 +235,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: C - Nguồn: Trung tâm Liên ngân hàng liên ngành, được Ngân hàng Nhân dân Trung Quốc uỷ quyền (2026). Thông báo lãi suất báo giá thị trường vay (LPR) ngày 20 tháng 8 năm 2026. - Ghi chú: Ở đây chỉ đưa ra cách tính, không kết luận. Trường hợp lợi nhuận đầu tư ổn định vượt lãi suất vay mua nhà là ít. Nhưng dù chọn bên nào, bạn cũng phải để dành đủ tiền mặt dự phòng. Ngoài ra hãy xem trong hợp đồng có điều khoản phạt trả nợ sớm hay không. Không phải lời khuyên đầu tư. - ### 29. Mua hàng trên mạng thì nhìn quy tắc của nền tảng và luật, không nghe người livestream, không tin「đánh giá tốt」 - Chi phí: Không tốn tiền. Trước khi đặt hàng, xem thêm điều khoản hậu mãi và quy tắc về tiền cọc. Sau khi nhận hàng, giữ lại đơn hàng, ảnh chụp màn hình phát lại của buổi livestream và lịch sử chat. Chỗ khó là không bị không khí trong buổi livestream dẫn dắt. @@ -260,7 +243,6 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2013, sửa đổi). Bộ luật Bảo vệ quyền lợi người tiêu dùng（第四十四、五十五条）. （Cục Quản lý thị trường quốc gia đăng lại）；Cục Quản lý không gian mạng Trung Quốc cùng sáu bộ khác (国家网信办等七部门) (2021). Biện pháp quản lý tiếp thị trên mạng (thử nghiệm)（第十八、二十八条）. ；Ủy ban Thường vụ Quốc hội (2025, sửa đổi). Luật cạnh tranh không công bằng（第九、二十五条）. （Hội đồng Quỹ hưu trí xã hội quốc gia đăng lại）；Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự（第五百八十六、五百八十七条）. ；Ủy ban Thường vụ Quốc hội (2021). Luật bảo vệ thông tin cá nhân（第二十四条）. （Cục Thống kê quốc gia đăng lại）；Cục Quản lý thị trường tỉnh Sơn Đông (2025). Sáu vụ cạnh tranh không công bằng trên mạng điển hình. - Ghi chú: Việc「nền tảng nào hiệu quả chi phí tốt nhất, dễ gặp hàng kém nhất」không có số liệu so sánh chính thức nào để dẫn, nên cuốn sách này không xếp hạng nền tảng. Chỉ tra được kết quả kiểm tra mẫu toàn bộ (xem mục 32, tra thông báo kiểm tra mẫu của quốc gia). Sau khi Luật cạnh tranh không công bằng sửa đổi năm 2025, quảng cáo giả chuyển từ Điều 8 cũ sang Điều 9. Chế tài phạt chuyển từ Điều 20 cũ sang Điều 25. Luật mới có hiệu lực từ ngày 15 tháng 10 năm 2025. Khi viết tài liệu, đừng chép số điều cũ. Trong loạt vụ án của Sơn Đông, có một công ty tổ chức「người ve chai」làm 75 giao dịch giả với số vốn hơn 68 vạn tệ, bị phát hiện, còn con số phạt thì không ghi trong thông báo. Quyền trả hàng trong bảy ngày không cần lý do, xem mục 23 của phần này (dùng luôn quyền trả hàng trong bảy ngày không cần lý do). - ### 30. Hàng mua trong livestream có vấn đề, trước hết yêu cầu nền tảng cung cấp thông tin người bán và người livestream, nền tảng phải cung cấp - Chi phí: Không tốn tiền. Chỉ cần đưa ra yêu cầu với nền tảng. @@ -277,6 +259,7 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (2025, sửa đổi lần thứ ba). Luật An toàn thực phẩm (食品安全法, điều một trăm bốn mươi tám). (Cơ sở dữ liệu luật và quy định quốc gia, có hiệu lực từ 2025-12-1) ; Tòa án nhân dân tối cao (2024). Giải thích về một số vấn đề pháp lý liên quan đến việc áp dụng pháp luật trong các vụ án tranh chấp về tiền phạt trong lĩnh vực thực phẩm và dược phẩm (关于审理食品药品惩罚性赔偿纠纷案件适用法律若干问题的解释) (法释〔2024〕9, các điều một, sáu, bảy, tám, mười hai, mười ba, mười bốn, mười lăm và mười sáu, có hiệu lực từ 2024-8-22). (Công báo Tòa án nhân dân tối cao) - Ghi chú: Thứ tự của bằng chứng quan trọng hơn số tiền. Chụp ảnh và quay video trước, đưa vào khung bao bì, ngày sản xuất, vật lạ và bản ghi mua hàng, rồi hãy liên hệ người bán hoặc nền tảng. Đừng bỏ đi hay rửa món đồ trước khi trao đổi với ai. 12315 làm việc hoà giải, không phải phán quyết. Người bán cứ nhất quyết không nhận thì chỉ còn đường kiện tòa, cách tính vụ kiện nhỏ xem mục 22 ở phần 8 (mua qua mạng, giao dịch đồ cũ bị lừa). Một điều nữa: ăn ra tóc, giun, thứ lạ như vậy có tính là「không đạt tiêu chuẩn an toàn thực phẩm」hay không, thường phải do cơ quan quản lý thị trường nhận định hoặc qua giám định, không phải do bạn nói. Nên câu「ăn ra một sợi tóc là chắc chắn có một nghìn tệ」không đứng vững. Gian lận với hàng hoá thường đền gấp ba lần, dưới 500 tệ thì tính bằng 500 tệ, xem mục 29 của phần này (mua qua mạng tin vào quy tắc của nền tảng và điều luật). Nếu tự làm thực phẩm để bán thì cần giấy phép nào, ghi nhãn thế nào, vì sao không được nói thuốc chữa bệnh, đều có quy định riêng. Chi tiết xem mục 8 đến 11 ở phần 12 (giấy phép thực phẩm, nhãn, vạch đỏ về công dụng chữa bệnh và vạch đỏ hình sự). Ngược lại, biến việc đòi bồi thường thành một nghề thì có vạch đỏ rõ ràng. Một, cố tình tạo ra hình ảnh giả để người bán bị coi là vi phạm pháp sản xuất kinh doanh, rồi dùng khiếu nại hay kiện tòa để đòi tiền bồi thường, thì có dấu hiệu tống tiền, Tòa phải chuyển đầu mối và hồ sơ cho cơ quan công an. Hai, kiện tòa dựa trên hình ảnh giả đã tạo thì Tòa bác yêu cầu; nếu cấu thành vụ án giả dối thì phạt tiền hoặc bị giữ tuỳ mức độ, và người bán còn có thể kiện ngược lại bạn về tội xâm phạm danh dự. Mục này thuộc hàng người hưởng là bạn và gia đình. + ### 32. Mua món đồ lớn thì tra thông báo kiểm tra mẫu quốc gia, chứng nhận 3C và nhãn tiết kiệm năng lượng trước - Chi phí: Không tốn tiền. Vài phút, tra trên trang chính thức của Cục Quản lý thị trường quốc gia và của cơ quan chứng nhận. @@ -309,6 +292,7 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Cục Quản lý thị trường quốc gia (2023). Hướng dẫn (thử nghiệm) về hành vi kinh doanh hộp mù (盲盒经营行为规范指引（试行）) (国市监稽发〔2023〕39, các điều chín, mười một, mười bảy, hai mươi mốt và hai mươi ba). - Ghi chú: Văn bản này là hướng dẫn, quản lý người kinh doanh phải làm gì, bản thân nó không phải căn cứ xử phạt. Thực sự có chuyện thì xử lý theo Luật Chất lượng sản phẩm, Luật Cạnh tranh không lành mạnh, Luật Bảo vệ quyền lợi người tiêu dùng. Có ba trường hợp là lý do cụ thể để bạn khiếu nại: không công bố xác suất, xác suất đã công bố không khớp với thực tế, mua phải hộp rỗng. Trước khi khiếu nại, quay video màn hình lại quy tắc trên trang và quá trình mở hộp. Việc sau khi mở bao bì không áp dụng quyền đổi trong bảy ngày không cần lý do có một tiền đề: người bán đã thông báo đầy đủ, bạn cũng đã xác nhận khi mua từng lần, và không được chọn sẵn thay bạn. Bán cho người chưa thành niên có quy định khác: người bán không được bán cho trẻ dưới 8 tuổi, từ 8 tuổi trở lên phải xác nhận đã có sự đồng ý của người giám hộ. Trẻ nạp tiền điện thoại rồi tặng thưởng thì xin hoàn tiền thế nào, xem mục 9 của phần này (trẻ dùng điện thoại nạp tiền tặng thưởng). Thời hạn tỉnh táo 24 giờ cho khoản mua lớn không thiết yếu, xem mục 23 (đặt thời hạn tỉnh táo cho khoản mua lớn không thiết yếu). Vòng lặp「mua xong nhạt dần, rồi lại mua tiếp」, xem mục 23 ở phần 6 (đừng trông chờ mua đồ để cải thiện tâm trạng). + ### 36. Quan tâm thực phẩm có thêm gì thì mua loại đóng gói sẵn: quầy hàng bán lẻ không bắt buộc ghi bảng thành phần - Chi phí: Không tốn tiền. Cùng một món, loại đóng gói sẵn thường đắt hơn loại bán lẻ một chút. Liếc bảng thành phần, mấy giây. @@ -325,15 +309,7 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Odean T (1998). Are Investors Reluctant to Realize Their Losses? The Journal of Finance 53(5):1775-1798. - Ghi chú: Mục này quản lý「có bán hay không」, mục 15 (không giao dịch cổ phiếu thường xuyên) quản lý「đừng liên tục ra tay」, hai mục không mâu thuẫn. Định quy tắc trước rồi làm theo, số lần ra tay lại ít hơn. Mua thêm để hạ giá vốn thì xét về số học là không sai, mua thêm một chút thì giá vốn trung bình thấp hơn một chút. Chỗ sai là coi đó là cách lật lời, tác dụng thực tế là làm lớn thêm số tiền bạn đặt vào một món cổ phiếu đó, xem mục 19 (đừng đặt hết tiền vào một món cổ phiếu). Đoạn về thuế tính theo hệ thống thuế Mỹ, Trung Quốc không áp dụng, phần còn lại vẫn đúng. Đây là dữ liệu tài khoản ở Mỹ thập niên 1980 đến 1990, chi phí giao dịch và hệ thống thuế của cổ phiếu A đều khác. Không phải lời khuyên đầu tư. -### 38. Lúc thị trường tăng giảm dữ dội thì bớt ra tay, đừng tăng vốn mua bán khi nóng nhất - -- Chi phí: Không tốn tiền. Khó nhất là lúc người ta khác đang kiếm tiền thì bạn không tăng cược theo. -- Nói thẳng: Sàn chứng khoán Thượng Hải đã tính sổ cho vòng tăng giảm dữ dội 2014 đến 2015 trên 40 triệu tài khoản. 85% hộ gia đình có tài khoản nhỏ nhất vì giao dịch thường xuyên mà kiếm được ít hơn 250 tỷ tệ so với cứ nắm giữ không động. Con số đó bằng khoảng ba phần mười giá trị cổ phiếu ban đầu của họ. Trong hai năm rưỡi thị trường đi ngang, chênh lệch này chỉ một phần trăm đến ba phần trăm. -- Lợi ích: Nghiên cứu dùng dữ liệu tài khoản theo ngày của Sàn chứng khoán Thượng Hải, phủ toàn bộ khoảng 40 triệu tài khoản. Thời gian mẫu từ 2014-7 đến 2015-12, tổng cộng 18 tháng. Trong 18 tháng đó chỉ số chung Thượng Hải tăng hơn 150% trước. Nó đạt đỉnh 5166.35 điểm ngày 2015-6-12, rồi lại giảm 40% vào cuối tháng 12. Tài khoản gia đình chia thành bốn nhóm theo giá trị vốn đầu kỳ. Nhóm nhỏ nhất chiếm 85% toàn bộ tài khoản gia đình, nhóm lớn nhất chiếm 0.5%. Tổng giá trị cổ phiếu đầu kỳ của hai nhóm gần bằng nhau, lần lượt là 880 tỷ tệ và 808 tỷ tệ. Kết quả là 85% nhỏ nhất mất 250 tỷ tệ vì giao dịch chủ động, 0.5% lớn nhất kiếm được 254 tỷ tệ. Ở đây「mất」là so với việc cứ nắm giữ suốt cùng giai đoạn, không làm gì. Mỗi con số bằng khoảng ba phần mười giá trị cổ phiếu đầu kỳ của nhóm đó, một bên 28%, một bên 31%. Để đối chiếu, hai năm rưỡi trước tháng 6 năm 2014 thị trường khá đi ngang. Trong thời gian đó lấy bất kỳ 18 tháng nào, chênh lệch cùng loại chỉ bằng 1% đến 3% giá trị đầu kỳ. Trong vòng tăng giá này, tài khoản gia đình trung bình khoảng ba tuần đổi hết cổ phiếu đang cầm, gần 18 lần mỗi năm (Trung Quốc, Thượng Hải, 2014 đến 2015; thời kỳ đối chiếu 2012 đến 2014) -- Mức bằng chứng: A -- Nguồn: An L, Lou D, Shi D (2022). Wealth redistribution in bubbles and crashes. Journal of Monetary Economics 126:134-153. ; Bản thảo làm việc do cùng các tác giả công bố (2021-10, số liệu của mục này đối chiếu theo đúng bản đó). -- Ghi chú: Nghiên cứu không nói đừng mua cổ phiếu. Nó tính rằng trong cùng một giai đoạn, phía tài khoản nhỏ càng giao dịch chủ động thì tiền càng chảy sang bên đối thủ, và mốc so sánh là「cứ nắm giữ không động」. Người lấy đi số tiền đó cũng không phải tổ chức, mà là nhóm cá nhân có tài khoản lớn nhất: trong thời gian này khoảng chín phần mười khối lượng giao dịch của Sàn chứng khoán Thượng Hải đến từ tài khoản nhà đầu tư nhỏ. Câu「chín phần mười nhà đầu tư nhỏ lỗ」lưu truyền trên mạng không tìm được số liệu chính thức để đối chiếu, mục này không dùng con số đó. Có thể so với tỷ lệ luân chuyển. Trong vòng tăng giá này tài khoản gia đình hoán đổi gần 18 lần mỗi năm, còn hộ gia đình Mỹ trong mục 15 (không giao dịch cổ phiếu thường xuyên) mỗi năm chỉ hoán đổi ba phần tư. Không phải lời khuyên đầu tư. -### 39. Muốn mua tài sản nước ngoài như cổ phiếu Mỹ thì đi qua quỹ QDII hoặc kênh Hồng Kông, đừng mở tài khoản ở công ty chứng khoán nước ngoài, cũng đừng dùng hạn mức đổi tiền hằng năm để mua +### 38. Muốn mua tài sản nước ngoài như cổ phiếu Mỹ thì đi qua quỹ QDII hoặc kênh Hồng Kông, đừng mở tài khoản ở công ty chứng khoán nước ngoài, cũng đừng dùng hạn mức đổi tiền hằng năm để mua - Chi phí: Không tốn tiền. Mua quỹ QDII ngay trong ứng dụng ngân hàng hoặc chứng khoán của bạn, mất mấy phút. Cái giá phải trả là số loại sản phẩm mua được ít hơn ở công ty chứng khoán nước ngoài. - Nói thẳng: Với các công ty chứng khoán nước ngoài như Tiger, Futu, Longbridge, từ 2026-5 không được mở tài khoản cho người ở trong nước nữa. Tài khoản đã mở thì trong hai năm chỉ được bán ra và chuyển tiền ra, không được mua thêm. Hạn mức đổi tiền cũng không được dùng để mua cổ phiếu nước ngoài, người vi phạm thì năm đó và hai năm sau không được dùng hạn mức này. Muốn mua tài sản nước ngoài thì đi qua quỹ QDII hoặc kênh Hồng Kông. @@ -341,15 +317,16 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Ủy ban Chứng khoán và Phái đoàn Trung Quốc (2026-05-22). Tám cơ quan, trong đó có Ủy ban Chứng khoán và Phái đoàn, ban hành chung《Phương án thực hiện cải tổ toàn diện hoạt động kinh doanh chứng khoán, phái đoán, quỹ ngoài quốc gia trái phép》. ; Ủy ban Chứng khoán và Phái đoàn Trung Quốc (2026-05-22). Lãnh đạo các bộ phận liên quan của Ủy ban Chứng khoán và Phái đoàn trả lời báo chí về《Phương án thực hiện cải tổ toàn diện hoạt động kinh doanh chứng khoán, phái đoán, quỹ nước ngoài trái phép》. ; Ủy ban Chứng khoán và Phái đoàn Trung Quốc (2026-05-22). Ủy ban Chứng khoán và Phái đoàn xử lý nghiêm các vụ công ty như Tiger kinh doanh vượt biên giới trái phép. ; Cục Ngoại hối quốc gia (2021). Đơn xin mua ngoại tệ của cá nhân (Phụ lục 1). - Ghi chú: Mấy văn bản này quản lý cơ quan nước ngoài, không ghi chế tánh cho chính người đầu tư. Người đã có tài khoản ở công ty chứng khoán nước ngoài thì trong hai năm bán ra và chuyển tiền về là xong. Ba kênh hợp pháp đều có điều kiện mở khác nhau, hỏi rõ ở ngân hàng hoặc công ty chứng khoán của bạn. Cổ phiếu Mỹ đã tăng nhiều trong quá khứ, không có nghĩa sau này cũng vậy. Mua quỹ QDII cũng theo cách ở mục 17 (quỹ chỉ số rộng) và mục 19 (đừng đặt hết tiền vào một món cổ phiếu, một nền tảng, một căn nhà), chọn chỉ số rộng, đừng đặt vào một thị trường duy nhất. Rút tiền ở nước ngoài có hạn mức riêng, xem mục 6 ở phần 21 (rút tiền ở nước ngoài). Không phải lời khuyên đầu tư. -### 40. Chỉ mua bảo hiểm cho khoản thiệt hại mà bạn không gánh nổi, khoản thiệt hại gánh nổi thì để quỹ dự phòng đảm nhiệm +### 39. Chỉ mua bảo hiểm cho loại tổn thất mà bạn không gánh nổi, tổn thất gánh nổi thì để khoản dự phòng đỡ -- Chi phí: Không tốn tiền. Việc cần làm là liệt kê một lượt các khoản thiệt hại lớn nhà bạn có thể gặp, xem khoản nào tiền tiết kiệm không gánh nổi. Một lần, nửa tiếng. -- Nói thẳng: Tiền phí bảo hiểm bạn trả không được hoàn lại hết, một phần phải trả chi phí và lợi nhuận của công ty bảo hiểm. Nên đáng mua là khoản thiệt hại hiếm khi xảy ra, một khi xảy ra thì nhà bạn không lấy ra nổi, ví dụ bệnh nặng, lái xe gây thương tích cho người khác, người kiếm tiền trong nhà mất. Khoản thiệt hại nhỏ mà tiền tiết kiệm gánh nổi thì để quỹ dự phòng đảm nhiệm. -- Lợi ích: Trong tiền phí bảo hiểm công ty thu vào, chỉ một phần được trả ra làm tiền bồi hoàn. Phần còn lại phải trả chi phí bán hàng và vận hành, cộng lợi nhuận của công ty. Cơ quan quản lý yêu cầu công ty bán bảo hiểm sức khoẻ ngắn hạn cho cá nhân, cứ nửa năm công bố một lần「综合赔付率」trên trang chính thức. Đó là tỷ lệ tiền đã trả so với tiền phí, muốn biết một công ty trả ra bao nhiêu thì tra trang của công ty. Nên với một khoản thiệt hại, mua bảo hiểm lâu dài tính trung bình là lỗ một chút. Đổi lại bạn trải một khoản chi lớn có thể dìm chết cả nhà thành một khoản nhỏ mỗi năm. Thiệt hại càng nhỏ, càng dễ bạn trả nổi, phần đổi này càng không đáng. Theo tiêu chí đó, mấy món cuốn sách đã viết là đáng mua. Bảo hiểm y tế một năm xem mục 20 ở phần 7 (bảo hiểm y tế một năm). Người có xe xem mục 26 của phần này (bảo hiểm trách nhiệm bên thứ ba). Người đi nước ngoài xem mục 4 ở phần 21 (y tế ở nước ngoài). Nhà có người sống dựa vào thu nhập của bạn thì xem mục 41 của phần này (bảo hiểm nhân thọ có thời hạn) (toàn quốc, từ 2021) +- Chi phí: Không tốn tiền. Việc phải làm là liệt kê những khoản tổn thất lớn nhà có thể gặp, rồi xem khoản nào tiền tiết kiệm không gánh nổi. Mất khoảng một tiếng rưỡi cho một lần. +- Nói thẳng: Tiền phí bảo hiểm bạn đóng không được trả lại hết, một phần làm chi phí và lợi nhuận của công ty bảo hiểm. Vì thay đáng mua là loại tổn thất hiếm gặp, gặp rồi thì nhà không trả nổi, ví dụ bệnh nặng, lái xe gây thương tích cho người khác, người kiếm tiền trong nhà mất. Tổn thất nhỏ mà tiền tiết kiệm gánh nổi thì giao cho khoản dự phòng. +- Lợi ích: Trong số tiền phí bảo hiểm công ty thu vào, chỉ một phần được chi ra dưới dạng tiền bồi thường. Phần còn lại làm chi phí bán hàng, chi phí vận hành và lợi nhuận của công ty. Cơ quan quản lý yêu cầu công ty bán bảo hiểm sức khoẻ ngắn hạn cho cá nhân công bố「tỷ lệ bồi thường tổng hợp」trên trang chính thức của mình nửa năm một lần. Chỉ số đó xấp xỉ là số tiền đã bồi thường so với tổng tiền phí, muốn biết một công ty bồi thường bao nhiêu thì vào trang chính thức của công ty tra. Vì thay với một khoản tổn thất, mua bảo hiểm thì xét trung bình dài hạn là lỗ một chút tiền. Đổi lại là biến một khoản chi lớn có thể làm sập nhà thành một khoản nhỏ mỗi năm. Tổn thất càng nhỏ, càng dễ gánh, phép đổi đó càng không đáng. Theo tiêu chuẩn này, mấy loại cuốn sách đã viết là đáng mua. Bảo hiểm y tế một năm xem mục 20 ở phần 7 (bảo hiểm y tế một năm). Người có xe xem mục 26 của phần này (bảo hiểm trách nhiệm dân sự bên thứ ba). Người đi nước ngoài xem mục 4 ở phần 21 (bảo hiểm y tế ở nước ngoài). Nhà có người sống dựa vào thu nhập của bạn thì xem mục 40 của phần này (bảo hiểm nhân thọ có thời hạn) (toàn quốc, từ 2021) - Mức bằng chứng: C -- Nguồn: Văn phòng Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2021). Thông báo về việc chuẩn hoá hoạt động bảo hiểm sức khoẻ ngắn hạn (银保监办发〔2021〕7, tỷ lệ tổng hợp chi trả được công bố nửa năm một lần trên trang chính thức của công ty). -- Ghi chú: Đặt mức C vì không có số liệu chính thức nào để tính xem loại bảo hiểm nào mua là đáng, ở đây đưa ra cách cân đối. Bảo hiểm vỡ màn hình, bảo hiểm chậm chuyến bay, thiệt hại vài trăm tệ, bản thân bạn trả nổi, theo tiêu chí này không mua cũng được. Bảo hiểm bệnh hiểm nặng trả cho khoản thu nhập mấy năm ngừng làm vì bệnh, người có tiền tiết kiệm đủ gánh một hai năm ngừng việc thì có thể cân nhắc không mua. Bảo hiểm y tế cơ bản là bảo hiểm xã hội, không theo cách cân đối này, vẫn phải đóng, xem mục 9 ở phần 7 (bảo hiểm y tế cư dân). Khoản thiệt hại nhỏ dựa vào đâu, xem mục 27 của phần này (quỹ dự phòng). Mục này thuộc người hưởng là bạn và gia đình. -### 41. Nhà có người sống dựa vào thu nhập của bạn thì mua bảo hiểm nhân thọ có thời hạn cho người kiếm tiền trước, đừng mua cho con trước +- Nguồn: Văn phòng Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (中国银保监会办公厅) (2021). Thông báo quản lý việc chuẩn bị hoá hoạt động bảo hiểm sức khoẻ ngắn hạn (关于规范短期健康保险业务有关问题的通知)（银保监办发〔2021〕7 号, tỷ lệ bồi thường tổng hợp công bố nửa năm một lần trên trang chính thức của công ty). +- Ghi chú: Đặt mức C vì không có con số chính thức nào tính được loại bảo hiểm nào mua là đáng, ở đây đưa ra cách cân đối. Loại bảo hiểm kính vỡ, bảo hiểm chậm chuyến bay, tổn thất chỉ vài trăm tệ, bạn tự trả được, theo tiêu chuẩn này có thể không mua. Bảo hiểm bệnh nghiêm trọng bồi thường khoản thu nhập mấy năm bạn nghỉ việc vì bệnh, người có tiền tiết kiệm đủ sống sốt một hai năm nghỉ việc thì có thể cân nhắc không mua. Bảo hiểm y tế cơ bản là bảo hiểm xã hội, không cân theo cách này, vẫn phải đóng, xem mục 9 ở phần 7 (bảo hiểm y tế cho cư dân). Tổn thất nhỏ dựa vào đâu, xem mục 27 của phần này (khoản dự phòng). Mục này thuộc hàng người hưởng là chính bạn và gia đình. + +### 40. Nhà có người sống dựa vào thu nhập của bạn thì mua bảo hiểm nhân thọ có thời hạn cho người kiếm tiền trước, đừng mua cho con trước - Chi phí: Một năm vài trăm đến vài nghìn tệ, tuỳ tuổi, mức bảo hiểm và số năm mua (ước tính sơ bộ của tác giả, không có số liệu chính thức). Trước khi mua phải khai báo trung thực tình trạng sức khoẻ, một lần mất nửa tiếng. - Nói thẳng: Bảo hiểm nhân thọ có thời hạn chỉ bảo vệ trong số năm cố định, trong những năm đó nếu người đó mất thì nhà nhận được một khoản tiền. Nó dành cho người thân sống dựa vào thu nhập của bạn, ví dụ tiền vay mua nhà chưa trả hết, con còn nhỏ. Không ai sống dựa vào bạn thì không cần. Con không kiếm tiền, bảo hiểm tử vát mua cho con còn bị luật giới hạn mức bồi thường. @@ -357,7 +334,8 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: C - Nguồn: Ủy ban Giám sát và Quản lý bảo hiểm Trung Quốc (2011). Biện pháp quản lý điều khoản bảo hiểm và mức phí bảo hiểm của công ty bảo hiểm nhân thọ (人身保险公司保险条款和保险费率管理办法) (Quy định của Ủy ban Giám sát và Quản lý bảo hiểm số 3 năm 2011, các điều tám và mười hai). ; Ủy ban Thường vụ Quốc hội (2015, sửa đổi). Luật Bảo hiểm của Cộng hoà Nhân dân Trung Quốc, các điều ba mươi ba và ba mươi tư. - Ghi chú: Đặt mức C vì phía chính thức chỉ có định nghĩa và hạn mức, còn「bảo hiểm người lớn trước」suy ra từ「ai mất thu nhập thì nhà không sống nổi」. Con gặp chuyện thì nhà không mất đi một khoản thu nhập. Người kiếm tiền trong nhà mất thì tiền vay mua nhà và chi phí nuôi con vẫn còn. Khi tiền e hẹp, mua bảo hiểm nhân thọ có thời hạn và bảo hiểm y tế một năm cho người kiếm tiền trước, rồi mới đến bảo hiểm y tế cho con. Khai báo trung thực và giữ nguyên việc gia hạn xem thế nào, xem mục 20 ở phần 7 (bảo hiểm y tế một năm). Mục này thuộc hàng người hưởng thứ ②, là vợ hoặc chồng và con của bạn. -### 42. Ký bảo hiểm nhân thọ trên một năm rồi hối hận, hủy trong thời hạn cân nhắc, tiền phí trả lại gần hết + +### 41. Ký bảo hiểm nhân thọ trên một năm rồi hối hận, hủy trong thời hạn cân nhắc, tiền phí trả lại gần hết - Chi phí: Không tốn tiền. Gọi đường dây khách hàng của công ty bảo hiểm hoặc ra quầy, mất mười mấy phút. Khó nhất là sau khi ký, dành 15 ngày để bình tĩnh đọc hợp đồng một lượt. - Nói thẳng: Bảo hiểm nhân thọ trên một năm đều có thời hạn cân nhắc. Mua qua ngân hàng thì 15 ngày kể từ khi nhận được bản hợp đồng, trong thời gian đó hủy bảo hiểm bị trừ nhiều nhất 10 tệ phí giấy tờ. Qua thời hạn cân nhắc rồi mới hủy thì chỉ lấy lại được phần tiền trong hợp đồng quy định hoàn lại, thường ít hơn số tiền đã trả. Công ty bảo hiểm gọi điện hỏi thăm bạn có hiểu rõ không, chưa hiểu thì đừng nói là hiểu rõ. @@ -365,27 +343,29 @@ Phần này chỉ tính tiền: tiết kiệm được bao nhiêu, mỗi năm l - Mức bằng chứng: A - Nguồn: Văn phòng Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2019). Biện pháp quản lý hoạt động ngân hàng làm đại lý bảo hiểm (商业银行代理保险业务管理办法) (银保监办发〔2019〕179, các điều ba mươi sáu, sáu mươi bảy và sáu mươi tám). ; Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2019). Biện pháp quản lý bảo hiểm sức khoẻ (银保监会令 2019 年第 3 号) (điều mười lăm). ; Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2022). Quy tắc công bố thông tin sản phẩm bảo hiểm nhân thọ trên một năm (银保监规〔2022〕24, bản mô tả sản phẩm phải nêu thời điểm bắt đầu, số ngày và quyền của thời hạn cân nhắc). ; Ủy ban Giám sát và Quản lý bảo hiểm Trung Quốc (2010). Quy định về các dịch vụ cơ bản trong hoạt động bảo hiểm nhân thọ (人身保险业务基本服务规定) (Quy định số 4 năm 2010 của Ủy ban Giám sát và Quản lý bảo hiểm, điều mười lăm). ; Ủy ban Thường vụ Quốc hội (2015, sửa đổi). Luật Bảo hiểm của Cộng hoà Nhân dân Trung Quốc, các điều mười lăm và bốn mươi bảy. ; Cục Giám sát Tài chính địa phương Ninh Hạ (2026-06-16). Cảnh báo rủi ro về việc phòng chống đại lý hủy bảo hiểm trái phép. - Ghi chú: Thời hạn cân nhắc là quy định dành cho bảo hiểm nhân thọ trên một năm. Bảo hiểm y tế một năm, bảo hiểm tai nạn, quy định không yêu cầu, có hay không phải xem điều khoản. Mua không qua ngân hàng thì số ngày theo điều khoản trong hợp đồng, nhận bản hợp đồng bảo hiểm thì mở mục「犹豫期」trước. Bảo hiểm tích lũy, bảo hiểm chia lợi nhuận số tiền phí phải trả lớn, thời hạn cân nhắc đáng dùng nhất, xem mục 25 của phần này (bảo hiểm tiêu dùng). Lợi ích đặt là「vừa」vì thứ tiết kiệm được là phần chênh lệch giữa tiền phí đã trả và giá trị tiền mặt, thường từ vài trăm đến vài nghìn tệ, không có số liệu chính thức. Mục này thuộc người hưởng là chính bạn. -### 43. Trước khi ký ở quầy ngân hàng, xem kỹ trong tay mình là tiền gửi hay bảo hiểm +### 42. Trước khi ký ở quầy ngân hàng, nhìn rõ trong tay là tiền gửi hay bảo hiểm -- Chi phí: Không tốn tiền. Xem thêm hai phút bìa hợp đồng trước khi ký. -- Nói thẳng: Ngân hàng có thể bán hộ bảo hiểm, nhưng không được nói thành tiền gửi. Bìa hợp đồng bảo hiểm phải in chữ lớn「保险合同」và tên công ty bảo hiểm, không được có chữ「存款」「储蓄」. Trên bìa là công ty bảo hiểm thì số tiền đó không phải tiền gửi, hủy sau thời hạn cân nhắc sẽ lỗ. Người trên 65 tuổi mua ở ngân hàng, nguyên tắc chỉ mua loại có lợi nhuận xác định. -- Lợi ích: Quy định ghi rõ, ngân hàng không được bán lẫn bảo hiểm với tiền gửi tiết kiệm, quỹ, sản phẩm quản lý tài sản của ngân hàng. Không được đơn giản so lợi nhuận của bảo hiểm với tiền gửi, không được nói lợi nhuận không chắc chắn thành lợi nhuận được bảo đảm. Giấy tờ bảo hiểm và tài liệu quảng bá không được dùng tên và biểu trưng của ngân hàng, không được có chữ「存款」「储蓄」「与银行共同推出」. Bìa hợp đồng bảo hiểm phải in「保险合同」bằng chữ không nhỏ hơn cỡ 72 và in tên công ty bảo hiểm bằng chữ không nhỏ hơn cỡ số hai. Có ba nhóm người mua bảo hiểm, nguyên tắc chỉ được bán sản phẩm có quyền lợi bảo hiểm xác định: người mua bảo hiểm trên 65 tuổi; người mua sản phẩm trả góp hằng kỳ trên 60 tuổi; người ghi thu nhập năm thấp hơn thu nhập bình quân tiêu dùng của người dân địa phương. Hợp đồng của ba nhóm người này còn phải chuyển sang công ty bảo hiểm thẩm định thủ công, không được tự động xuất hồ sơ tại điểm giao dịch. Nhân viên công ty bảo hiểm không được bán bảo hiểm trong khu vực ngân hàng. Bảo hiểm trên một năm bán qua ngân hàng có thời hạn cân nhắc 15 ngày, hủy sau thời hạn cân nhắc thì có thiệt hại (toàn quốc, từ 2019-10) +- Chi phí: Không tốn tiền. Trước khi ký thì nhìn thêm hai phút trang bìa hợp đồng. +- Nói thẳng: Ngân hàng có thể làm đại lý bán bảo hiểm, nhưng không được nói đó là tiền gửi. Bìa hợp đồng bảo hiểm phải in chữ lớn「hợp đồng bảo hiểm」và tên công ty bảo hiểm, không được có chữ「tiền gửi」「tiết kiệm」. Bìa để tên công ty bảo hiểm thì số tiền đó không phải tiền gửi, hủy sau thời hạn cân nhắc sẽ lỗ. Người trên 65 tuổi mua ở ngân hàng, nguyên tắc chỉ được mua loại lợi nhuận chắc chắn. +- Lợi ích: Quy định ghi rõ, ngân hàng không được bán lẫn bảo hiểm với tiền gửi, quỹ, sản phẩm quản lý tiền của ngân hàng. Không được ví tỉ lệ lợi nhuận của bảo hiểm một cách đơn giản với tiền gửi, không được nói lợi nhuận không chắc chắn là lợi nhuận được bảo đảm. Giấy tờ bảo hiểm và tài liệu quảng bá không được dùng tên và nhãn hiệu của ngân hàng, không được có chữ「tiền gửi」「tiết kiệm」「phát hành cùng với ngân hàng」loại này. Bìa hợp đồng bảo hiểm phải in「hợp đồng bảo hiểm」bằng chữ cỡ không nhỏ hơn 72 và in tên công ty bảo hiểm bằng chữ cỡ không nhỏ hơn cỡ nhị. Có ba nhóm người mua bảo hiểm, nguyên tắc chỉ được bán sản phẩm có lợi ích hợp đồng chắc chắn: người mua bảo hiểm trên 65 tuổi; người mua sản phẩm đóng phí theo kỳ hạn mà trên 60 tuổi; người khai thu nhập hằng năm thấp hơn thu nhập bình quân của người dân địa phương. Hợp đồng của ba nhóm này còn phải chuyển cho công ty bảo hiểm thẩm định thủ công, không được cấp phiếu tự động tại điểm giao dịch. Nhân viên công ty bảo hiểm không được bán bảo hiểm trong điểm giao dịch của ngân hàng. Bảo hiểm trên một năm mà ngân hàng bán có thời hạn cân nhắc 15 ngày, qua thời hạn cân nhắc rồi hủy thì có tổn thất (toàn quốc, từ tháng 10 năm 2019) - Mức bằng chứng: B -- Nguồn: Văn phòng Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (2019). Biện pháp quản lý hoạt động ngân hàng làm đại lý bảo hiểm (商业银行代理保险业务管理办法) (银保监办发〔2019〕179, các điều ba mươi, ba mươi mốt, ba mươi hai, ba mươi sáu, bốn mươi hai và bốn mươi chín). -- Ghi chú: Đặt mức B vì quy định rất rõ ràng, nhưng không có số liệu chính thức về「người bị đánh lừa mất bao nhiêu tiền」. Người lớn tuổi trong nhà dễ gặp nhất khi đi gửi tiền ở ngân hàng, đi cùng thì giúp xem bìa. Đã ký rồi thì trong 15 ngày hủy theo mục 42 của phần này (thời hạn cân nhắc). Thấy mình bị đánh lừa thì khiếu nại thế nào, xem mục 44 của phần này (đại lý hủy bảo hiểm). Mục này thuộc người hưởng là bạn và người lớn tuổi trong nhà. -### 44. Muốn hủy bảo hiểm thì tự tìm công ty bảo hiểm, đừng tìm「đại lý hủy bảo hiểm」, thấy mình bị đánh lừa thì gọi 12378 khiếu nại +- Nguồn: Văn phòng Ủy ban Ngân hàng và Bảo hiểm Trung Quốc (中国银保监会办公厅) (2019). Biện pháp quản lý hoạt động ngân hàng làm đại lý bảo hiểm (商业银行代理保险业务管理办法)（银保监办发〔2019〕179 号, các điều ba mươi, ba mươi mốt, ba mươi tư, ba mươi sáu, bốn mươi tư và bốn mươi chín）. +- Ghi chú: Đặt mức B vì quy định viết rất rõ, nhưng không có số liệu chính thức「người bị hiểu sai lỗ bao nhiêu」. Người lớn tuổi trong nhà hay gặp chuyện này nhất khi đi gửi tiền ở ngân hàng, bạn đi cùng thì nhìn giúp trang bìa. Người đã ký rồi thì trong 15 ngày hủy theo mục 41 của phần này (thời hạn cân nhắc). Nghĩ là mình bị hiểu sai rồi thì khiếu nại thế nào, xem mục 43 của phần này (đại lý hủy bảo hiểm trái phép). Mục này thuộc hàng người hưởng là chính bạn và người lớn tuổi trong nhà. + +### 43. Muốn hủy bảo hiểm thì tự tìm công ty bảo hiểm làm, đừng tìm「đại lý hủy bảo hiểm」, nghĩ là mình bị hiểu sai rồi thì gọi 12378 khiếu nại -- Chi phí: Không tốn tiền. Tự gọi đường dây khách hàng của công ty bảo hiểm hoặc ra quầy, đi thêm một hai chút so với giao cho bên trung gian. -- Nói thẳng: Người quảng cáo trên mạng「hoàn tiền toàn bộ, không thành công không thu phí」phần lớn là đại lý hủy bảo hiểm trái phép. Họ muốn lấy từ ba phần mười đến năm phần mười số tiền hoàn lại, còn đòi giấy tờ tùy thân, thẻ ngân hàng và mã xác nhận của bạn. Đã hủy bảo hiểm thì sự bảo vệ cũng mất. Thấy mình bị đánh lừa thì trước tiên gọi khách hàng của công ty bảo hiểm, không hợp thì gọi 12378, miễn phí suốt quá trình. -- Lợi ích: Tháng 2026-2, Cục Quản lý Tài chính, Văn phòng Quản lý không gian mạng Trung ương, Bộ Công an, Ngân hàng Nhân dân Trung Quốc và Ủy ban Chứng khoán và Phái đoàn Trung Quốc cùng ban hành cảnh báo rủi ro. Cảnh báo nói những cách nói như「全额退保」là thông tin sai sự thật, không phù hợp với Luật Bảo hiểm và các luật quy định khác. Người làm「bảo vệ quyền lợi hộ」thu phí dịch vụ cao, bạn muốn dừng giữa chừng còn có thể bị kiện vì「违约」. Họ còn thu thập thẻ SIM, thẻ ngân hàng, hợp đồng bảo hiểm, địa chỉ nhà, trường học của con bạn. Số liệu đó bán lại thì bạn có thể gặp lừa đảo qua điện thoại và tiêu ý trái phép trên thẻ. Cảnh báo tháng 6-2026 của Cục Giám sát Tài chính địa phương Ninh Hạ viết chi tiết hơn. Loại trung gian này thỏa thuận là nếu hủy bảo hiểm thành công thì thu phí đại lý từ 30% đến 50% số tiền hủy bảo hiểm, nếu bỏ dở giữa chừng thì phải trả phạt nặng. Họ dạy người ta bịa lý do như bán hàng gây hiểu lầm, ký hộ để đi khiếu nại. Người làm theo có thể vi phạm Luật Xử phạt quản lý trật tự công cộng, nghiêm trọng hơn có thể cấu thành đồng phạm vu khống oan hoặc tống tiền. Hủy bảo hiểm thì trách nhiệm bảo hiểm kết thúc ngay. Sau này mua bảo hiểm cùng loại có thể tăng giá hoặc bị từ chối vì đã lớn tuổi hoặc đã phát hiện vấn đề sức khoẻ. Nếu thật sự bị đánh lừa thì trước tiên khiếu nại và thương lượng với khách hàng chính thức của công ty bảo hiểm, không hợp thì gọi 12378, đường dây khiếu nại và bảo vệ quyền lợi người tiêu dùng ngân hàng và bảo hiểm, miễn phí suốt quá trình. Nếu chỉ khó khăn trả tiền phí trong thời gian ngắn, trước tiên hỏi công ty bảo hiểm có làm được giảm mức đóng một lần, vay theo hợp đồng bảo hiểm hay không. Giảm mức đóng một lần nghĩa là sau này không phải đóng tiếp, mức bảo hiểm giảm theo tỷ lệ (toàn quốc, 2026) +- Chi phí: Không tốn tiền. Tự gọi đường dây khách hàng của công ty bảo hiểm hoặc ra quầy, chỉ phải đi thêm một hai chuyến so với giao cho môi giới. +- Nói thẳng: Mấy quảng cáo trên mạng rao「hủy bảo hiểm trọn vẹn, không thành công không thu phí」đa phần là đại lý hủy bảo hiểm trái phép. Họ lấy ba đến năm phần mười số tiền hủy bảo hiểm trả lại, còn đòi cả giấy tờ tùy thân, thẻ ngân hàng và mã xác minh của bạn. Hủy bảo hiểm xong thì sự bảo đảm cũng mất. Nghĩ là mình bị hiểu sai, trước hết gọi khách hàng của công ty bảo hiểm, không nói chuyện được thì gọi 12378, toàn bộ miễn phí. +- Lợi ích: Tháng 2 năm 2026, Cục Tài chính và Ngân hàng, Văn phòng Xây dựng mạng thông tin Trung Quốc, Bộ Công an, Ngân hàng Nhân dân Trung Quốc và Ủy ban Chứng khoán Trung Quốc cùng ban hành cảnh báo rủi ro. Cảnh báo nói các câu như「hủy bảo hiểm trọn vẹn」là thông tin sai, không khớp với Luật Bảo hiểm và các quy phạm khác. Người làm「đòi quyền qua đại lý」thu phí dịch vụ cao, bạn muốn dừng giữa chừng còn có thể phải ra tòa vì「vi phạm hợp đồng」. Họ còn thu thập thẻ SIM, thẻ ngân hàng, hợp đồng bảo hiểm, nơi ở, trường học của con bạn. Những thông tin đó bán lại, bạn có thể dính lừa đảo qua điện thoại và mất tiền do gian lận. Cảnh báo tháng 6 năm 2026 của Cục Giám sát Tài chính Ninh Hạ viết chi tiết hơn. Loại môi giới này thỏa thuận thu phí đại lý từ 30% đến 50% số tiền hủy bảo hiểm, làm nửa chừng thì phải nộp phạt vi phạm cao. Họ còn dạy người ta bịa lý do như bị tư vấn sai, bị ký hộ để khiếu nại. Người làm theo có thể vi phạm Luật xử phạt quản lý trật tự công cộng, trường hợp nặng còn có thể bị tình nghi cùng tham gia vu khống, ép bức tống tiền. Hủy bảo hiểm xong thì trách nhiệm bảo hiểm chấm dứt ngay. Sau này mua lại loại bảo hiểm tương tự có thể bị tăng giá hoặc từ chối bán vì lớn tuổi hơn hoặc khám ra vấn đề sức khoẻ. Bị hiểu sai thật thì trước hết khiếu nại trao đổi với khách hàng chính thức của công ty bảo hiểm, không nói chuyện được thì gọi 12378, đường dây khiếu nại và bảo vệ quyền lợi cho người tiêu dùng ngân hàng và bảo hiểm, toàn bộ miễn phí. Chỉ là tạm thời không đủ tiền đóng thì hỏi công ty bảo hiểm có làm được dịch vụ giảm mức bảo hiểm còn phí, vay bằng hợp đồng bảo hiểm hay không. Giảm mức bảo hiểm còn phí nghĩa là sau này không phải đóng tiền nữa, mức bảo hiểm giảm theo tỷ lệ (toàn quốc, năm 2026) - Mức bằng chứng: B -- Nguồn: Cục Quản lý Tài chính, Văn phòng Quản lý không gian mạng Trung ương, Bộ Công an, Ngân hàng Nhân dân Trung Quốc và Ủy ban Chứng khoán và Phái đoàn Trung Quốc (2026-02-06). Cảnh báo rủi ro về việc cảnh giác «video ngắn và livestream của các đối tượng "bảo vệ quyền lợi" trái pháp» (关于警惕不法「代理维权」短视频及直播陷阱的风险提示, đăng lại trên trang của Văn phòng Quản lý không gian mạng Trung ương). ; Cục Giám sát Tài chính địa phương Ninh Hạ (2026-06-16). Cảnh báo rủi ro về việc phòng chống đại lý hủy bảo hiểm trái phép. ; Cục Giám sát Tài chính tỉnh Hồ (2025-07-23). Cảnh báo rủi ro về việc phòng chống ngành công ngầm bảo hiểm「hủy bảo hiểm」trái phép (关于防范非法「代理退保」黑灰产的风险提示). -- Ghi chú: Cục Giám sát Tài chính tỉnh Hồ năm 2025 đã thông báo một vụ. Một người phụ nữ bị người「bạn」khuyên đi đại lý hủy bảo hiểm, đã nộp「phí tư vấn」, cuối cùng vẫn hủy theo giá trị tiền mặt. Ba tháng sau khi hủy, bà được chẩn đoán ung thư vú, hợp đồng bảo hiểm bệnh hiểm nặng đó đã mất. Có hủy hay không phải tự tính: bảo vệ còn dùng được mà không trả nổi thì trước tiên hỏi giảm mức đóng một lần. Tự hủy trong thời hạn cân nhắc, xem mục 42 của phần này (thời hạn cân nhắc). Mục này thuộc người hưởng là chính bạn. -### 45. Ghi người thụ hưởng là một người cụ thể trên hợp đồng bảo hiểm, có chuyện thì nhớ đi nhận tiền: bảo hiểm nhân thọ trong 5 năm, bảo hiểm khác trong 2 năm phải thực hiện yêu cầu +- Nguồn: Cục Tài chính và Ngân hàng, Văn phòng Xây dựng mạng thông tin Trung Quốc, Bộ Công an, Ngân hàng Nhân dân Trung Quốc, Ủy ban Chứng khoán Trung Quốc (2026-02-06). Cảnh báo rủi ro về việc cảnh giác「đòi quyền qua đại lý」trái pháp trong video ngắn và livestream (中央网信办官网转载). ；Cục Giám sát Tài chính Ninh Hạ (2026-06-16). Cảnh báo rủi ro về việc phòng chống đại lý hủy bảo hiểm trái phép. ；Cục Giám sát Tài chính Hồ Nam (2025-07-23). Cảnh báo rủi ro về việc phòng chống nghề đen đại lý hủy bảo hiểm trái phép. +- Ghi chú: Năm 2025 Cục Giám sát Tài chính Hồ Nam đã thông báo một ca. Một người phụ nữ bị「người quen」khuyên đi làm đại lý hủy bảo hiểm, đóng「phí tư vấn」, cuối cùng vẫn phải hủy bảo hiểm theo giá trị tiền mặt. Ba tháng sau khi hủy, bà được chẩn đoán ung thư vú, hợp đồng bảo hiểm bệnh nghiêm trọng đó đã mất. Có hủy hay không thì tự tính: bảo hiểm còn dùng được thì khi không đủ tiền đóng, hỏi giảm mức bảo hiểm còn phí trước. Tự hủy trong thời hạn cân nhắc, xem mục 41 của phần này (thời hạn cân nhắc). Mục này thuộc hàng người hưởng là chính bạn. + +### 44. Ghi người thụ hưởng là một người cụ thể trên hợp đồng bảo hiểm, có chuyện thì nhớ đi nhận tiền: bảo hiểm nhân thọ trong 5 năm, bảo hiểm khác trong 2 năm phải thực hiện yêu cầu - Chi phí: Không tốn tiền. Khi mua bảo hiểm thì điền một lần, sau này đổi người thụ hưởng phải báo công ty bảo hiểm bằng văn bản. - Nói thẳng: Không ghi người thụ hưởng, hoặc người thụ hưởng đã mất trước, thì tiền bảo hiểm tính là tài sản thừa kế, phải dùng để trả nợ của người đã mất trước. Ghi người thụ hưởng là「vợ, họ tên」, sau đó ly hôn, lúc có chuyện thì coi như không ghi. Người nhà phải biết có hợp đồng bảo hiểm này: bảo hiểm nhân thọ phải yêu cầu trong 5 năm kể từ khi biết có chuyện, bảo hiểm khác là 2 năm. - Lợi ích: Luật Bảo hiểm quy định người thụ hưởng do người được bảo hiểm hoặc người mua bảo hiểm chỉ định. Người mua bảo hiểm chỉ định thì phải được người được bảo hiểm đồng ý. Có ba trường hợp tiền bảo hiểm tính là tài sản thừa kế của người được bảo hiểm, chia theo quy tắc thừa kế. Một, không chỉ định người thụ hưởng, hoặc chỉ định không rõ ràng. Hai, người thụ hưởng chết trước người được bảo hiểm, và không có người thụ hưởng khác. Ba, người thụ hưởng bị mất hoặc từ bỏ quyền thụ hưởng theo pháp luật, và không có người thụ hưởng khác. Khi người thụ hưởng và người được bảo hiểm chết trong cùng một sự việc, không phân biệt được trước sau thì mặc định người thụ hưởng chết trước. Số tiền tính là tài sản thừa kế phải trả nợ thuế và nợ của người đã mất trước, phần còn lại mới chia. Giải thích của Tòa án nhân dân tối cao ghi rõ: ghi người thụ hưởng là「họ tên kèm quan hệ」, đến lúc có chuyện mà quan hệ đã thay đổi thì coi như không chỉ định người thụ hưởng. Đổi người thụ hưởng mà không báo công ty bảo hiểm thì không có hiệu lực với công ty bảo hiểm. Thời hiệu khởi kiện để nhận tiền: bảo hiểm nhân thọ là 5 năm, bảo hiểm khác là 2 năm, đều tính từ ngày bạn biết hoặc phải biết có chuyện. Mức bảo hiểm của bảo hiểm nhân thọ thường ở mức hàng trăm nghìn tệ, nên lợi ích đặt là lớn (toàn quốc, sửa đổi 2015, có hiệu lực từ 2021) - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (2015, sửa đổi). Luật Bảo hiểm của Cộng hoà Nhân dân Trung Quốc, các điều hai mươi sáu, ba mươi chín, bốn mươi mốt và bốn mươi hai. ; Tòa án nhân dân tối cao (2020, sửa đổi). Giải thích về một số vấn đề liên quan đến việc áp dụng Luật Bảo hiểm của Cộng hoà Nhân dân Trung Quốc, phần ba (关于适用《中华人民共和国保险法》若干问题的解释（三）), điều chín, khoản hai, điểm ba, và điều mười. ; Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc, Điều một nghìn một trăm năm mươi chín. -- Ghi chú: Hết thời hiệu khởi kiện thì công ty bảo hiểm có quyền từ chối trả tiền, ra tòa cũng khó lấy lại. Nên người nhà ít nhất phải biết bạn đã mua bảo hiểm ở những công ty nào. Sau khi ly hôn, kết hôn lại, sinh con, quay lại đổi người thụ hưởng một lần. Sau khi người đó mất, đi lấy tiền ở từng nơi, xem mục 9 ở phần 25 (đi lấy ở từng nơi). Dùng cái chết để trả nợ thì không đi được, xem mục 13 ở phần 29 (trả nợ). Mục này thuộc hàng người hưởng thứ ②, là vợ hoặc chồng và con của bạn. \ No newline at end of file +- Ghi chú: Hết thời hiệu khởi kiện thì công ty bảo hiểm có quyền từ chối trả tiền, ra tòa cũng khó lấy lại. Nên người nhà ít nhất phải biết bạn đã mua bảo hiểm ở những công ty nào. Sau khi ly hôn, kết hôn lại, sinh con, quay lại đổi người thụ hưởng một lần. Sau khi người đó mất, đi lấy tiền ở từng nơi, xem mục 9 ở phần 25 (đi lấy ở từng nơi). Dùng cái chết để trả nợ thì không đi được, xem mục 13 ở phần 29 (trả nợ). Mục này thuộc hàng người hưởng thứ ②, là vợ hoặc chồng và con của bạn. diff --git "a/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" "b/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" index a09e3cc1..947d74f2 100644 --- "a/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" +++ "b/vi/book/06-\345\217\215\351\235\242\346\270\205\345\215\225.md" @@ -12,7 +12,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: A - Nguồn: Sesso HD et al. (2012). Multivitamins in the prevention of cardiovascular disease in men: the Physicians' Health Study II randomized controlled trial. JAMA. ；US Preventive Services Task Force (2022). Vitamin, Mineral, and Multivitamin Supplementation to Prevent Cardiovascular Disease and Cancer. JAMA. ；Gaziano JM 等 (2012). Multivitamins in the Prevention of Cancer in Men. JAMA. （备注里那项癌症结果） - Ghi chú:Tranh cãi: Trong cùng nhóm bác sĩ nam đó, nghiên cứu còn xem tới ung thư. Nhóm uống có tỷ lệ mắc ung thư thấp hơn khoảng 8% (0,86–0,998, P=0,04, vừa đủ vượt ngưỡng 「không giống ngẫu nhiên」). Tỷ lệ chết vì ung thư thấp hơn khoảng 12% (0,77–1,01). Khoảng này vượt qua 1, nên có thể chỉ là ngẫu nhiên. Lợi ích nhỏ này rất nhỏ, và chỉ thấy ở bác sĩ nam. Người thực sự bị thiếu chất dinh dưỡng, phụ nữ có thai (bổ sung axit folic), và người ăn quá thiếu lệch, không tính vào đây, cứ bổ sung theo lời bác sĩ - ### 2. Đừng uống dầu cá thông thường để phòng bệnh tim mạch - Chi phí: Khoảng 1 đến 3 tệ mỗi ngày, 400 đến 1.000 tệ mỗi năm. Uống không tốn thời gian. @@ -21,7 +20,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: A - Nguồn: Manson JE et al. (2019). Marine n-3 Fatty Acids and Prevention of Cardiovascular Disease and Cancer. NEJM. ；ASCEND Study Collaborative Group (2018). Effects of n-3 Fatty Acid Supplements in Diabetes Mellitus. NEJM. ；Bhatt DL 等 (2019). Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia. New England Journal of Medicine. （备注里那项高纯度处方鱼油试验） - Ghi chú:Tranh cãi: Có một thử nghiệm thực sự có hiệu quả. Thử nghiệm REDUCE-IT có 8.179 người vốn đã mắc bệnh tim mạch hoặc đái tháo đường, đều đang dùng statin, mỡ máu triglyceride vẫn cao (135–499 mg/dL). Họ uống mỗi ngày 4 g thuốc EPA tinh khiết, tỷ lệ gặp sự cố là 17,2% so với 22,0%. Nguy cơ thấp hơn khoảng một phần tư (HR 0,75, 0,68–0,83). Nhưng đó là thuốc kê đơn liều cao kê cho bệnh nhân nguy cơ cao, không giống người thường uống dầu cá dạng thực phẩm chức năng. Ở đây cũng không tính việc ăn cá. - ### 3. Đừng bổ sung vitamin D cho người không thiếu vitamin D chỉ để kéo dài sống - Chi phí: Khoảng 0.2 đến 0.5 tệ mỗi ngày, 100 đến 200 tệ mỗi năm. Số tiền không nhiều, cái giá thật là tưởng rằng mình đã phòng ngừa xong. @@ -30,7 +28,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: A - Nguồn: Manson JE et al. (2019). Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease. NEJM. ；Neale RE et al. (2022). The D-Health Trial: a randomised controlled trial of the effect of vitamin D on mortality. Lancet Diabetes Endocrinol. - Ghi chú:Tranh cãi: Cả hai thử nghiệm đều không cố ý tuyển người thực sự thiếu vitamin D, nên kết luận chỉ áp cho 「người vốn không thiếu」. Người xét ra thực sự thiếu, người bị loãng xương, người lâu không hền nắng, trẻ nhỏ, vẫn bổ sung theo lời bác sĩ. Ở đây chỉ xem tử vong toàn nguyên, ung thư và bệnh tim mạch, những thứ khác không xem - ### 4. Đừng uống thực phẩm bổ sung chống oxy hóa để phòng ung thư (β-carotene, vitamin E, vitamin A) - Chi phí: Khoảng 0.5 đến 2 tệ mỗi ngày. Với người hút thuốc, cái giá thật là nguy cơ ung thư phổi tăng theo. @@ -39,7 +36,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: A - Nguồn: Bjelakovic G et al. (2012). Antioxidant supplements for prevention of mortality in healthy participants and patients with various diseases. Cochrane Database Syst Rev. ；The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group (1994). The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. NEJM. ；Omenn GS et al. (1996). Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. NEJM. - Ghi chú: Cơ quan chính thức của Hoa Kỳ chấm điểm các biện pháp phòng ngừa USPSTF, năm 2022 xếp β-carotene và vitamin E ở mức D, nghĩa là khuyến nghị không dùng. Người hút thuốc và người từng tiếp xúc với bụi amianhạt đã chịu hại vì thế. Chất chống oxy hóa có sẵn trong rau củ quả không tính vào đây, cứ ăn. - ### 5. Đừng trông chờ glucosamine/chondroitin chữa viêm khớp gối - Chi phí: Khoảng 2 đến 5 tệ mỗi ngày, 700 đến 1.800 tệ mỗi năm. @@ -57,7 +53,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: A - Nguồn: Hemilä H, Chalker E (2013). Vitamin C for preventing and treating the common cold. Cochrane Database Syst Rev. - Ghi chú:Tranh cãi: Trong chính tổng quan đó, uống hằng ngày quả thật rút ngắn thời gian khó chịu khi bị cảm. Người lớn ngắn hơn khoảng 8% (3%–12%), trẻ em ngắn hơn 14% (7%–21%). Tính ra mỗi lần cảm bệnh ít hơn khoảng nửa ngày. Vận độn viên marathon, người trượt tuyết, lính hành quân ở vùng khí lạnh là ngoại lệ. Trong 5 thử nghiệm, 598 người, số người bị cảm giảm khoảng một nửa (RR 0.48, 0.35–0.64). Đa số không thuộc hai nhóm này - ### 7. Đừng tự làm 「PET-CT toàn thân」 hay 「gói xét nghiệm chất chỉ dấu ung thư」 khi không có triệu chứng - Chi phí: Một lần PET-CT toàn thân khoảng 7000 đến 10000 tệ, cộng thêm một lần nhiễm xạ. Gói chất chỉ dấu ung thư vài trăm tệ. Cái giá thật nằm ở sau đó: nếu ra một kết quả 「bất thường」 thật ra chỉ là hốt, gọi là dương tính giả, phần tiếp theo là chụp lại, lấy mẫu mô xét nghiệm, mổ, tiền và sự khổ nằm hết ở đoạn đó. @@ -66,7 +61,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: A - Nguồn: US Preventive Services Task Force (2018). Screening for Ovarian Cancer: US Preventive Services Task Force Recommendation Statement. JAMA（cùng khuyến nghị đó còn có trang chính thức của USPSTF）. 、；Furtado CD et al. (2005). Whole-body CT screening: spectrum of findings and recommendations in 1192 patients. Radiology. - Ghi chú: Phần chất chỉ dấu ung thư có thử nghiệm đối chứng ngẫu nhiên đứng sau. Phần CT toàn thân và PET-CT chỉ có loại theo dõi 「xem người đã làm thì sau đó ra sao」, không có nhóm đối chứng, nên bằng chứng yếu hơn một bậc, tương đương mức B. Vài sàng lọc thật sự có bằng chứng, như ung thư ruột giàu, ung thư cổ tử cung, ung thư vú, và người nguy cơ cao làm CT ngực liều thấp, là chuyện khác, xem phần 1. Người đã có triệu chứng, hoặc người đã xác định ung thư mà làm PET-CT, đó là khám bệnh chứ không phải sàng lọc, không so được - ### 8. Đừng trông chờ vòng tay hay đồng hồ đo chuyển động giúp bạn giảm cân - Chi phí: Thiết bị 200 đến 2000 tệ. Ngoài ra mỗi ngày còn phải phân tâm nhìn mấy con số đó. @@ -75,7 +69,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: B - Nguồn: Jakicic JM et al. (2016). Effect of Wearable Technology Combined With a Lifestyle Intervention on Long-term Weight Loss: The IDEA Randomized Clinical Trial. JAMA. - Ghi chú: Ở đây chỉ có một thử nghiệm, lại dùng thiết bị đeo ở cánh tay kiểu đầu thập niên 2010, nên không thể nói 「mọi thiết bị đeo đều vô ích」. Bản thân việc vận động có hiệu quả, xem phần 2. Kết luận chỉ một câu: muốn giảm cân, thiết bị không phải khoản tiền bắt buộc phải chi - ### 9. Đừng trả tiền bản giá cao hơn để mua thực phẩm hữu cơ vì 「khỏe hơn」 - Chi phí: Thường là 1 đến 3 lần giá thực phẩm thường cùng loại. @@ -84,15 +77,14 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: B - Nguồn: Smith-Spangler C et al. (2012). Are organic foods safer or healthier than conventional alternatives?: a systematic review. Ann Intern Med. - Ghi chú: Nghiên cứu xem thẳng sức khỏe sau khi ăn rất ít. Vì vậy kết luận là 「không có bằng chứng nào cho thấy nó tốt hơn」, chứ không phải 「đã chứng minh nó không có ích」. Ảnh hưởng tới môi trường, phúc lợi động vật, khẩu vị, ở đây không bàn. Ăn nhiều rau quả vốn đã có ích, có phải hữu cơ hay không không ảnh hưởng điều đó - -### 10. Đừng bỏ tiền lớn mua thực phẩm chức năng, cao thuốc bắc, đồ bồi bổ để 「điều chỉnh cơ thể」 +### 10. Đừng bỏ tiền lớn vào thực phẩm bảo vệ sức khoẻ (保健食品), thang thuốc bồi bổ (膏方), đồ bồi bổ để「bồi bổ cơ thể」 -- Chi phí: Từ vài trăm tệ tới hàng chục nghìn tệ, tùy loại. Sau khi đóng thành hộp quà tặng, giá thường nhân lên mấy lần. -- Nói thẳng: Cơ quan quản lý thực phẩm chức năng quy định, trên bao bì phải in một câu: thực phẩm chức năng không phải là thuốc, không thay thế thuốc điều trị bệnh. Những thành phần phổ biến như đa vitamin, dầu cá, vitamin D, sáu mục đầu phần này đã lần lượt tra rồi, đều không thấy tác dụng. Khoản tiền này phần lớn là để mua cho cha mẹ, mang tới tận nhà. Tấm lòng đó không có gì đáng nói, chỉ là chọn nhầm món. -- Lợi ích: Cục Quản lý thị trường quốc gia quy định, mặt trước bao bì thực phẩm chức năng phải ghi bằng diện tích không nhỏ hơn 20% một câu: 「thực phẩm chức năng không phải là thuốc, không thay thế thuốc điều trị bệnh」. Cùng quy định đó còn khóa công dụng của thực phẩm chức năng trong phạm vi 「bổ sung chất dinh dưỡng trong khẩu phần, duy trì hoặc cải thiện trạng thái sức khỏe cơ thể, hoặc giảm yếu tố nguy cơ gây bệnh」. Chính cơ quan quản lý không công nhận nó chữa được bệnh. Vài thành phần phổ biến nhất trong đó, đa vitamin, dầu cá, vitamin D, chất chống oxy hóa, glucosamine, vitamin C, cho ra kết quả gì trong thử nghiệm đối chứng chia nhóm ngẫu nhiên, 6 mục đầu phần này đã liệt kê từng mục. Nó có vẻ đáng tiền vì câu 「không chữa bệnh, chỉ điều chỉnh」 đúng là nói gì cũng không sai, và không ai chứng minh được nó sai. Mà khi tặng quà mà bỏ số tiền này thì dựa vào tình cảm chứ không dựa vào bằng chứng +- Chi phí: Từ vài trăm tệ đến hơn mười nghìn tệ. Đóng thành hộp quà tặng thì giá thường nhân lên vài lần. +- Nói thẳng: Mấy thành phần thường gặp như vitamin đa, dầu cá, vitamin D đã được sáu mục trước của phần này tra từng thành phần một, không thấy thành phần nào có tác dụng. Câu trên bao bì「thực phẩm bảo vệ sức khoẻ không phải là thuốc, không thay thế thuốc để điều trị bệnh」là cơ quan quản lý thực phẩm bảo vệ sức khoẻ bắt buộc phải in. Số tiền này phần lớn là mua cho cha mẹ, mang đến nhà, cái tấm lòng thì không có gì đáng nói, chỉ là chọn nhầm đồ. +- Lợi ích: Đổi lại là khoản tiền đã tiết kiệm. Lý do nằm ở 6 mục trước của phần này: mấy thành phần phổ biến nhất trong thực phẩm bảo vệ sức khoẻ, vitamin đa, dầu cá, vitamin D, chất chống oxy hoá, glucosamine, vitamin C, đều không thấy lợi ích như mong đời trong các thử nghiệm đối chứng chia nhóm ngẫu nhiên, con số của từng thành phần nằm ở cột lợi ích của 6 mục đó. Quy định của cơ quan quản lý chỉ nói được phép tuyên bố những gì, không thể coi đó là bằng chứng cho thấy nó vô dụng. Cục Quản lý thị trường quốc gia quy định mặt trước bao bì thực phẩm bảo vệ sức khoẻ phải in một câu chiếm diện tích không nhỏ hơn 20%「thực phẩm bảo vệ sức khoẻ không phải là thuốc, không thay thế thuốc để điều trị bệnh」. Cùng quy định đó còn khoanh vùng công dụng của thực phẩm bảo vệ sức khoẻ trong phạm vi「bổ sung chất dinh dưỡng từ thức ăn, duy trì và cải thiện trạng thái sức khoẻ cơ thể hoặc làm giảm các yếu tố nguy cơ bệnh tật」. Nó có vẻ đáng tiền vì câu「không chữa bệnh, chỉ bồi bổ」nói thế nào cũng không sai, không ai chứng minh được nó là sai. Hơn nữa lúc tặng quà mà bỏ số tiền này thì dựa vào tình cảm, không phải dựa vào bằng chứng - Mức bằng chứng: C -- Nguồn: Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2019). 市场监管总局就《保健食品标注警示用语指南》和《保健食品原料目录与保健功能目录管理办法》有关情况举办专题新闻发布会. -- Ghi chú: Kết luận là 「không có bằng chứng ủng hộ」, chứ không phải 「đã chứng minh không có tác dụng」, nên chỉ cho mức C. Sản phẩm ghi rõ thành phần thì tra nghiên cứu tương ứng theo mấy mục trước. Người xét ra thực sự thiếu dinh dưỡng, người đang trị bệnh, cứ theo lời bác sĩ. Cái cần bỏ là sản phẩm, không phải tấm lòng. Người già tự mua và đang ăn, miễn là không dùng nó thay thuốc đang uống và không sinh vấn đề gì, thì không cần lên tận nhà cãi lại, việc sắp xếp vaccine và xét nghiệm sàng lọc cần làm cấp thiết hơn. Đổi số tiền đó sang những thứ sau, từng mục đều tra được trong sách. Đưa người già đi tiêm vaccine cúm hằng năm, xem mục 20 ở phần 1. Vaccine thủy đậu cho người trên 50 tuổi xem mục 21 ở phần 1. Vaccine phế cầu cho người trên 65 tuổi xem mục 22 ở phần 1. Mua một máy đo huyết áp, nhắc người đó uống đủ thuốc hạ huyết áp theo đơn và đưa huyết áp về mức mục tiêu, xem mục 7 ở phần 1. Sửa phòng tắm và cầu thang trong nhà, cùng tập thăng bằng và sức chân, xem mục 13 ở phần 1. Đến tuổi cần sàng lọc ung thư, đưa người đó đi làm một lần, xem mục 17 đến 19 ở phần 1. Người già đã nằm liệt giường lâu, phòng loét ép và bảo hiểm chăm sóc dài hạn xem mục 7 và 8 ở phần 17. Muốn mang quà tới nhà thì đổi sang trái cây, gạo, thứ người đó thích, thực tế hơn cao thuốc. Cái thật sự cần chặn là hai loại: một là đầu tư tiền vào chế độ dưỡng lão phải trả trước, hai là loại sẽ thay thế thuốc, xem mục 5 ở phần 17 +- Nguồn: Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2019). Cục Quản lý thị trường quốc gia tổ chức họp báo chuyên đề về tình hình liên quan đến《Hướng dẫn dùng cảnh báo trên nhãn thực phẩm bảo vệ sức khoẻ》và《Quy chế quản lý danh mục nguyên liệu và danh mục chức năng của thực phẩm bảo vệ sức khoẻ》. +- Ghi chú: Kết luận là「không có bằng chứng ủng hộ」, không phải「đã chứng minh là vô hiệu」, nên chỉ cho mức C. Sản phẩm ghi rõ thành phần thì tra nghiên cứu tương ứng theo mấy mục trước. Người tra ra thiếu chất dinh dưỡng, người đang điều trị bệnh thì làm theo lời bác sĩ. Cái cần bỏ là sản phẩm, không phải tấm lòng. Người lớn tuổi tự mua, đang uống rồi, chỉ cần không dùng nó thay thuốc đang uống và không sinh vấn đề gì thì không cần lên nhà cãi nhau chuyện này, sắp xếp đủ vắc xin và xét nghiệm sàng lọc cần thiết quan trọng hơn. Đổi số tiền rằng sang những thứ sau đây thì mục nào trong sách cũng tra được ra nguồn. Đưa người lớn tuổi đi tiêm vắc xin cúm hằng năm, xem mục 20 ở phần 1. Vắc xin viêm da dây thần kinh từ 50 tuổi, xem mục 21 ở phần 1. Vắc xin phế cầu khuẩn cho người trên 65 tuổi, xem mục 22 ở phần 1. Mua máy đo huyết áp, giám sát bà ấy uống thuốc hạ huyết áp đủ theo đơn và hạ huyết áp về mức chuẩn, xem mục 7 ở phần 1. Sửa phòng tắm và cầu thang trong nhà, cùng tập giữ thăng bằng và sức chân, xem mục 13 ở phần 1. Xét nghiệm sàng lọc ung thư đúng tuổi cần làm, đưa bà ấy đi làm một lần, xem mục 17 đến 19 ở phần 1. Người lớn tuổi đã nằm liệt giường lâu, phòng loét và bảo hiểm chăm sóc dài hạn xem mục 7 và 8 ở phần 17. Muốn mang đồ đến nhà thì đổi ra trái cây, gạo, thứ bà ấy thích, thực tế hơn thang thuốc bồi bổ. Cái thật sự cần chặn là hai loại: một loại là bỏ tiền trước vào nơi đầu tư nuôi dưỡng, một loại là thứ thay thế thuốc, xem mục 5 ở phần 17 ### 11. Đừng uống men vi sinh dài ngày khi ruột không có vấn đề @@ -111,7 +103,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: B - Nguồn: Buijze GA et al. (2016). The Effect of Cold Showering on Health and Work: A Randomized Controlled Trial. PLOS ONE. ；Cain T et al. (2025). Effects of cold-water immersion on health and wellbeing: A systematic review and meta-analysis. PLOS ONE. - Ghi chú:Tranh cãi: Số lần xin nghỉ việc quả thật giảm 29%, chênh lệch này không giống ngẫu nhiên. Nhưng tổng số ngày bị bệnh không giảm, nên giống 「có bệnh vẫn đi làm」 hơn là thật sự ít bệnh. Vận độn viên dùng nước lạnh để hồi phục sau buổi tập, là chuyện khác, ở đây không bàn. Tắm nước lạnh bản thân không gây hại cho người khỏe mạnh, chỉ đừng coi nó là một khoản đầu tư sức khỏe - ### 13. Đừng mua các sản phẩm giải độc, thụt ruột, enzyme, nước kiềm - Chi phí: Vài trăm tới vài nghìn tệ. Loại dùng nhịn đói để 「giải độc」 còn có thể làm đường huyết tụt và thiếu dinh dưỡng. @@ -120,7 +111,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: C - Nguồn: Klein AV, Kiat H (2015). Detox diets for toxin elimination and weight management: a critical review of the evidence. J Hum Nutr Diet. ；Fenton TR, Huang T (2016). Systematic review of the association between dietary acid load, alkaline water and cancer. BMJ Open. - Ghi chú: Chỉ có thể nói 「không có bằng chứng ủng hộ」, nên chỉ cho mức C. Chất thải chuyển hóa sinh ra trong cơ thể vốn do gan và thận xử lý. Chỉ cần hai cơ quan này không bệnh thì không cần làm thêm điều gì để 「giải độc」 - ### 14. Đừng ép mình uống nước chỉ để đủ 「8 ly nước mỗi ngày」 - Chi phí: Không tốn tiền. Tốn một chút thời gian và sức lực để nhớ uống, ban đêm cũng phải đi tiểu nhiều hơn. @@ -129,7 +119,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: C - Nguồn: Valtin H (2002). "Drink at least eight glasses of water a day." Really? Is there scientific evidence for "8 x 8"? Am J Physiol Regul Integr Comp Physiol. - Ghi chú: Cái cần bỏ là con số 「mỗi ngày phải đủ mấy ly」, không phải nói uống nhiều nước là có hại. Khát thì uống, nhìn màu nước tiểu, thế là đủ. Người làm việc trong nhiệt độ cao, người vận động mạnh, người từng bị sỏi thận, và người lớn tuổi không thấy khát thì cần chủ động uống, không so được - ### 15. Đừng bỏ tiền xem bói, xem tarot, xem chiêm tinh rồi ra quyết định - Chi phí: Không tốn tiền. Chỗ khó là từ bỏ một lối tắt để ra quyết định, sau đó phải tự nghĩ. @@ -138,15 +127,15 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: B - Nguồn: Forer BR (1949). The fallacy of personal validation: A classroom demonstration of gullibility. Journal of Abnormal and Social Psychology 44(1):118-123（表 1 评分分布：2 分 1 人、3 分 4 人、4 分 18 人、5 分 16 人，N=39）. ；Carlson S (1985). A double-blind test of astrology. Nature 318:419-425（表 2：首选 116 份，挑中 40 份，随机期望 38.5，占星师预计 58.5；「The data are consistent with chance, inconsistent with astrological hypothesis」）. ，第二作者主页全文 ；最高人民检察院 (2026). 河南息县：历经三年持续追踪打掉一批网络算命诈骗团伙. - Ghi chú: Cả hai bài đều chỉ là một nghiên cứu riêng lẻ. Bài của Forer không ghi thẳng điểm trung bình, 4.26 là tính từ phân bố trong bảng 1 của nguyên văn. Ở đây chỉ nói 「nó dự đoán không chính xác」, không đánh giá tín ngưỡng nào. Bỏ ra mấy chục tệ để giải trí thì không tính, dùng nó để quyết định nghỉ việc, kết hôn, đầu tư mới là chỗ lỗ hơn tiền. - -### 16. Đừng mua kính chống ánh sáng xanh để 「bảo vệ thị lực」, cũng đừng tin 「nhìn màn hình vài tháng là mắt hỏng」, nhưng mắt sưng đau đỏ thì coi là cấp cứu +### 16. Đừng mua kính chống ánh sáng xanh để「bảo vệ thị lực」, cũng đừng tin「nhìn màn hình mấy tháng là mắt mù」, nhưng mắt nhức sưng đỏ thì coi là cấp cứu -- Chi phí: Không tốn tiền, còn tiết kiệm được phần giá hơn vài trăm tới hơn nghìn tệ gắn thêm trên mặt kính. -- Nói thẳng: Kính lọc ánh sáng xanh không có tác dụng, không mua thì tiết kiệm được vài trăm tới hơn nghìn tệ tiền chênh lệch. So với kính thường, người đeo có mắt mỏi hay không, nhìn được bao xa cũng không khác, và không tìm được bằng chứng nào nó bảo vệ đáy mắt. Nhìn màn hình lâu thấy mắt rát và khô thì khỏi lại được, thứ thật sự gây hại mắt là ánh hàn hồ, tia cực tím, laser, và cơn glôcôm cấp nặng. -- Lợi ích: Tổng quan hệ thống Cochrane năm 2023 gom 17 thử nghiệm đối chứng chia nhóm ngẫu nhiên, mỗi thử nghiệm 5 đến 156 người, thời gian theo dõi từ chưa tới 1 ngày tới 5 tuần. Vì dữ liệu quá ít, quy mô mỗi thử nghiệm khác nhau, thời gian theo dõi cũng không thống nhất, tác giả không gộp chúng thành một tổng số. Mọi kết quả dưới đây so với kính không lọc ánh sáng xanh. Khi theo dõi chưa tới một tuần, mức mỏi mắt người bệnh tự báo 「có thể không khác biệt」, độ chắc của bằng chứng thấp. Trong một thử nghiệm 120 người, hai nhóm chênh 9.76 đơn vị. Khoảng tin cậy của chênh lệch này là 95% CI −33.95 đến 53.47. Khoảng đó vượt qua cả âm lẫn dương, nghĩa là không thấy khác biệt. Tần số thấp nhấp nháy ranh giới là một chỉ số khách quan đo mức mỏi mắt, hai nhóm 「có thể gần như không khác biệt」, độ chắc thấp. Thị lực tốt nhất sau khi đeo kính có kính 「chắc là không ảnh hưởng」. Hai nhóm chênh 0.00 logMAR (95% CI −0.02 đến 0.02). Mục này chỉ có 1 nghiên cứu, 156 người, độ chắc trung bình. Ảnh hưởng tới độ tỉnh táo ban ngày thì không nói rõ được (2 thử nghiệm, 42 người, độ chắc rất thấp). Bài tổng quan này cũng không tìm được bằng chứng ủng hộ việc nó bảo vệ vùng hoàng thể mắt +- Chi phí: Không tốn tiền, còn tiết kiệm được vài trăm đến hơn nghìn tệ tiền phí cộng thêm trên mặt kính. +- Nói thẳng: Kính chống ánh sáng xanh không có tác dụng, không mua thì tiết kiệm được vài trăm đến hơn nghìn tệ phí cộng thêm. So với kính thường, người đeo không khác gì về mắt có mỏi hay không, và nhìn rõ bao nhiêu, cũng không tìm thấy bằng chứng bảo vệ đáy mắt. Nhìn màn hình nhiều gây mắt mỏi và khô, nghỉ một lát và chớp mắt nhiều là khỏi. Thứ thật sự làm hỏng mắt là ánh hàn quang hàn điện, tia cực tím, laser, và cơn đột quỵ cấp của bệnh glôcôm. +- Lợi ích: Tổng quan hệ thống Cochrane năm 2023 gom 17 thử nghiệm đối chứng chia nhóm ngẫu nhiên, mỗi thử nghiệm từ 5 đến 156 người, thời gian theo dõi từ chưa đầy 1 ngày đến 5 tuần. Vì dữ liệu quá ít, các thử nghiệm tuyển người khác nhau, thời gian theo dõi cũng không nhất quán, các tác giả không gộp chúng thành một tổng số. Các kết quả dưới đây đều so với kính không lọc ánh sáng xanh. Ở thời gian theo dõi dưới một tuần, mức mỏi mắt tự báo cáo「có thể không khác biệt」, độ chắc của bằng chứng thấp. Một thử nghiệm 120 người, hai nhóm chênh nhau 9.76 đơn vị. Khoảng tin cậy của chênh lệch này là 95% CI −33.95 đến 53.47. Khoảng đó vắt qua cả dương lẫn âm, nghĩa là không thấy khác biệt. Tần số nhấp nháy phản biên là một chỉ số khách quan đo độ mỏi mắt, hai nhóm「có thể gần như không khác biệt」, độ chắc thấp. Thị lực tốt nhất sau khi đeo kính điều chỉnh「chắc không có ảnh hưởng」. Hai nhóm chênh nhau 0.00 logMAR (95% CI −0.02 đến 0.02). Chỉ số này chỉ có 1 nghiên cứu, 156 người, độ chắc trung bình. Ảnh hưởng đến độ tỉnh táo ban ngày thì không nói rõ (2 thử nghiệm, 42 người, độ chắc rất thấp). Bản tổng quan này cũng không tìm thấy bằng chứng ủng hộ tác dụng bảo vệ hoàng điểm vàng ở đáy mắt - Mức bằng chứng: A - Nguồn: Singh S, Downie LE, Anderson AJ, Keller PR, White SJ, Ang M, Wolffsohn JS (2023). Blue-light filtering spectacle lenses for visual performance, sleep, and macular health in adults. Cochrane Database of Systematic Reviews, 8, CD013244. -- Ghi chú: Nhìn màn hình gây mỏi mắt và khô mắt, biểu hiện là mắt rát, mắt khô, nhìn lâu thì mờ. Nghỉ một lát, chớm mắt nhiều hơn là khỏi. Nó có hồi phục được, không có nghĩa là 「mắt đã hỏng」. Câu này là đồng thuận của ngành mắt, không phải kết luận của bài tổng quan Cochrane trong cột nguồn, nên chỉ tính là mức C. Thứ thật sự gây tổn thương mắt không hồi phục được trong thời gian ngắn là nguồn sáng mạnh như ánh hàn hồ, tia cực tím, laser. Mắt bị ánh hàn hồ làm tổn thương gọi là 「viêm mắt do ánh sáng điện」, là bệnh nghề nghiệp theo luật. Cách chặn là kính bảo hộ và mặt nạ, không phải kính chống ánh sáng xanh, xem mục 11 ở phần 19 (phòng chống bụi, tiếng ồn, chất độc hoá học). Thức khuya dậy nhìn màn hình thì cái cần phòng không phải bản thân màn hình, mà có một ngoại lệ: mắt sưng đau đỏ, nhìn đèn có một vòng cầu vồng, kèm đau đầu, buồn nôn, nôn ói, đó là cơn glôcôm góc kín cấp tính, vài ngày là có thể ép hỏng dây thị giác, ngày đó phải đi cấp cứng mắt, xem mục 6 ở phần 13 (cấp cứu mắt). Ở chỗ tối đồng tử giãn, cúi đầu lâu đẩy thể thủy tinh về phía trước, hai việc đó làm nên thể bệnh. Người có nguy cơ cao là trên 50 tuổi, viễn thị, tiền đình nông. Khô mắt do màn hình kéo dài sẽ thành bệnh mạn tính phải đi khám, không phải ngủ một giấc là khỏi. Mắt cứ khó chịu, hoặc thị lực thật sự giảm, hãy đi khám mắt tìm nguyên nhân, đừng tự đổi mặt kính là xong. Trẻ em và thanh niên phòng cận thị thế nào, xem mục 4, 12 và 9 ở phần 30 (mỗi ngày 2 giờ ngoài trời, khúc xạ sau khi nhỏ mắt, không mua sản phẩm 「chữa khỏi cận thị」). +- Ghi chú: Nhìn màn hình gây mỏi thị giác và khô mắt, biểu hiện là mắt rát, mắt khô, nhìn lâu thì mờ. Nghỉ một lát, chớp mắt nhiều là khỏi. Nó có hồi phục được, không đồng nghĩa với「làm hỏng mắt」. Câu này là đồng thuận của ngành mắt, không phải kết luận của bản tổng quan Cochrane trong mục nguồn, nên chỉ tính là mức C. Thứ thật sự gây tổn thương mắt không thể hồi phục trong thời gian ngắn là nguồn sáng mạnh như ánh hàn quang hàn điện, tia cực tím, laser. Mắt bị ánh hàn quang hàn điện gây ra gọi là「viêm mắt do ánh sáng điện」, là bệnh nghề nghiệp theo luật. Chặn nó bằng kính bảo hộ và mặt nạ, không phải bằng kính chống ánh sáng xanh, xem mục 10 ở phần 19 (phòng chống bụi, tiếng ồn, chất độc hoá học). Thức khuya dán sát màn hình thì cái cần phòng không phải bản thân màn hình, mà là một ngoại lệ: mắt nhức sưng đỏ, nhìn đèn thấy một vòng cầu vồng, kèm đau đầu buồn nôn nôn ói, đó là cơn kịch phát glôcôm góc kín cấp, vài ngày là có thể làm hỏng dây thị giác, ngày hôm đó phải đi cấp cứng mắt, xem mục 6 ở phần 13 (cấp cứng mắt). Đồng tử giãn ra trong bóng tối, cúi đầu lâu rồi đẩy thể thuỷ tinh về phía trước, hai việc đó là nguyên nhân gây ra cơn kịch phát, người có nguy cơ cao là người trên 50 tuổi, viễn thị, tiền đình nông. Khô mắt do màn hình kéo dài cũng biến thành bệnh mạn phải đi khám bác sĩ, không phải ngủ một giấc là khỏi. Mắt khó chịu kéo dài, hoặc thị lực thật sự giảm, đi khám mắt để tìm ra nguyên nhân, đừng tự đổi mặt kính rồi cho là xong. Trẻ em và thanh niên phòng cận thế nào, xem mục 4, 12 và 9 ở phần 30 (2 giờ ngoài trời mỗi ngày, đo khúc xạ sau khi nhỏ mắt, không mua sản phẩm「chữa cận」). + ### 17. Đừng coi hiến máu là cách giữ sức khoẻ: 「giải độc, hạ mỡ máu, phòng nhồi máu cơ tim」 không có thử nghiệm chia nhóm ngẫu nhiên nào đỡ, hiến thường xuyên thì đúng là dễ thiếu sắt @@ -156,15 +145,15 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: A - Nguồn: Salonen JT, Tuomainen TP, Salonen R, Lakka TA, Nyyssönen K (1998). Donation of blood is associated with reduced risk of myocardial infarction. The Kuopio Ischaemic Heart Disease Risk Factor Study. American Journal of Epidemiology, 148(5), 445-451. ；Ullum H, Rostgaard K, Kamper-Jørgensen M, et al. (2015). Blood donation and blood donor mortality after adjustment for a healthy donor effect. Transfusion, 55(10), 2479-2485. ；Zacharski LR, Chow BK, Howes PS, et al. (2007). Reduction of iron stores and cardiovascular outcomes in patients with peripheral arterial disease: a randomized controlled trial. JAMA, 297(6), 603-610. ；Di Angelantonio E, Thompson SG, Kaptoge S, et al. (2017). Efficiency and safety of varying the frequency of whole blood donation (INTERVAL): a randomised trial of 45 000 donors. Lancet, 390(10110), 2360-2371. ；Beard JL, Borel MJ, Derr J (1990). Impaired thermoregulation and thyroid function in iron-deficiency anemia. American Journal of Clinical Nutrition, 52(5), 813-819. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (1997). Luật Hiến máu Cộng hòa Nhân dân Trung Hoa（中华人民共和国献血法）（第二、九、十四条）. ；Ủy ban Quốc gia Y tế và Sức khoẻ cùng các cơ quan khác (国家卫生健康委等) (2024-01-08). 关于进一步做好无偿献血者激励奖励工作的通知. - Ghi chú: Cái cần bỏ là 「đi hiến máu để tốt cho cơ thể mình」, không phải bản thân việc hiến máu. Máu dùng cho người khác, theo thang người hưởng của cuốn sách thì rơi vào bậc thứ ④ (người xa lạ). Phần quay lại được cho chính bạn là phần nhà nước dành riêng. Điều 14 của Luật Hiến máu (献血法) quy định người hiến máu tự nguyện dùng máu trong bệnh viện thì được miễn các khoản phí lấy máu, bảo quản, tách chiết và kiểm nghiệm, còn vợ chồng và người thân trực hệ được miễn hoặc giảm theo quy định của chính quyền cấp tỉnh. Thông báo tháng 1 năm 2024 của Ủy ban Quốc gia Y tế và Sức khoẻ ghi rõ 28 tỉnh đã áp dụng việc miễn trực tiếp phí dùng máu cho người hiến tự nguyện và người thân khi xuất viện tại bệnh viện trong tỉnh. Người nhận được Giải thưởng công đóng hiến máu tự nguyện còn được hưởng chính sách 「ba miễn」 tại địa phương: miễn phí đi phương tiện công cộng, miễn phí vào các công viên do nhà nước đầu tư, miễn phí khám chữa bệnh ngoại trú thường ở bệnh viện công lập. Muốn hiến thì theo đúng khoảng cách luật định. Điều 9 của Luật Hiến máu (献血法) quy định mỗi lần thường 200 ml, tối đa không quá 400 ml. Hai lần cách nhau không ít hơn 6 tháng. Tuổi khuyến nghị là từ 18 đến 55 tuổi. Đừng rút ngắn khoảng cách tới mức tối thiểu chỉ để lấy danh dự hay chỉ tiêu tập thể. Phụ nữ, người cân nhẹ, người kinh nguyệt nhiều vốn đã thiếu sắt, người hiến nhiều lần có thể tự trả tiền xét ferritin một lần. Sau khi hiến mà cứ mệt, sợ lạnh, hoặc buổi tối nằm xuống khó chịu trong chân đến nỗi phải cử động (gọi là hội chứng chân không yên), đừng tự đoán là 「tổn khí」, hãy xét công thức máu và ferritin. Nếu thiếu sắt thì bổ sung theo lời bác sĩ, đồng thời kéo dài khoảng cách hiến máu. Phần bằng chứng người về việc sợ lạnh chỉ có 10 người, dùng để giải thích vì sao lại vậy, không phải nói có bao nhiêu người bị như thế - -### 18. Đừng dựa vào gói khám sức khoẻ tổng quát để phòng bệnh, hãy làm đúng những xét nghiệm sàng lọc có bằng chứng theo tuổi và giới tính +### 18. Đừng dựa vào gói khám sức khoẻ tổng quát để phòng bệnh, hãy làm đúng mấy xét nghiệm sàng lọc có bằng chứng theo tuổi và giới -- Chi phí: Không tốn tiền, còn tiết kiệm được vài trăm đến vài nghìn tệ gói khám mỗi năm. Đổi lại bạn phải tự nhớ xét những mục nào, mỗi mục cách nhau bao lâu thì làm lại. -- Nói thẳng: Người khám sức khoẻ tổng quát định kỳ và người không khám, tỷ lệ tử vong như nhau, số người chết vì bệnh tim mạch và vì ung thư cũng bằng nhau. Làm đúng những xét nghiệm sàng lọc có bằng chứng theo tuổi và giới tính là đủ. Từ 35 tuổi trở lên, mỗi năm được đo huyết áp miễn phí một lần. Từ 65 tuổi trở lên, mỗi năm có một lần khám sức khoẻ miễn phí. -- Lợi ích: Bản cập nhật năm 2019 của tổng quan hệ thống Cochrane gồm 17 thử nghiệm đối chứng chia nhóm ngẫu nhiên. Khi tính tỷ lệ tử vong toàn nguyên, gộp 11 thử nghiệm trong đó, 233.298 người tham gia, 21.535 ca tử vong. Nhóm khám sức khoẻ tổng quát y hệt nhóm đối chứng, RR 1.00. RR là tỷ số nguy cơ, 1 nghĩa là hai bên như nhau. 95% CI 0.97–1.03 là khoảng tin cậy. Khi tính tử vong do bệnh tim mạch, gộp 9 thử nghiệm, 170.227 người, 6.237 ca tử vong, RR 1.05 (95% CI 0.94–1.16). Khi tính tử vong do ung thư, gộp 8 thử nghiệm, 139.290 người, 3.663 ca tử vong, RR 1.01 (95% CI 0.92–1.12). Cả hai cũng không thấy khác biệt. Kết luận của tác giả là 「khám sức khoẻ tổng quát khó có ích」. Chương trình dịch vụ công cộng y tế cơ bản quốc gia cho cư dân từ 35 tuổi trở lên đo huyết áp miễn phí một lần mỗi năm, và cho người cao tuổi từ 65 tuổi trở lên khám sức khoẻ miễn phí một lần mỗi năm. Lần khám miễn phí này gồm thăm khám thể chất, công thức máu, xét nghiệm nước tiểu, đường huyết lúc nhịn đói và điện tâm đồ. Chức năng gan xét alanin aminotransferase, aspartat aminotransferase, bilirubin toàn phần. Chức năng thận xét creatinin huyết thanh, urê máu. Mỡ máu xét tổng cholesterol, triglyceride, cholesterol mật độ thấp, cholesterol mật độ cao. Cuối cùng còn có siêu âm bụng gan mật tụy lách (toàn quốc) +- Chi phí: Không tốn tiền, còn tiết kiệm mỗi năm vài trăm đến vài nghìn tệ tiền gói khám. Cái đổi lại là bạn phải tự nhớ xét nghiệm nào, bao lâu xét lại một lần. +- Nói thẳng: Người đi khám sức khoẻ tổng quát định kỳ có tỷ lệ tử vong y như người không khám, chết vì bệnh tim mạch, ung thư cũng bằng nhau. Làm đúng mấy xét nghiệm sàng lọc có bằng chứng theo tuổi và giới là đủ. Người từ 35 tuổi trở lên, mỗi năm đo huyết áp miễn phí một lần; người trên 65 tuổi, mỗi năm có một lần khám sức khoẻ miễn phí. +- Lợi ích: Bản cập nhật năm 2019 của tổng quan hệ thống Cochrane gom 17 thử nghiệm đối chứng chia nhóm ngẫu nhiên. Khi tính tỷ lệ tử vong toàn nguyên, gộp 11 thử nghiệm, 233.298 người tham gia, 21.535 ca tử vong. Nhóm khám sức khoẻ tổng quát giống nhóm đối chứng, RR 1.00. RR là tỷ số nguy cơ, 1 nghĩa là hai bên như nhau. 95% CI 0.97–1.03 là khoảng tin cậy. Khi tính tử vong do bệnh tim mạch, gộp 9 thử nghiệm, 170.227 người, 6.237 ca tử vong, RR 1.05 (95% CI 0.94–1.16). Khi tính tử vong do ung thư, gộp 8 thử nghiệm, 139.290 người, 3.663 ca tử vong, RR 1.01 (95% CI 0.92–1.12). Hai chỗ này cũng không thấy khác biệt. Kết luận của các tác giả là「khám sức khoẻ tổng quát khó có ích」. Chương trình dịch vụ công cộng cơ bản về y tế của nhà nước cấp phép đo huyết áp miễn phí mỗi năm một lần cho cư dân từ 35 tuổi trở lên, và cấp phép khám sức khoẻ miễn phí mỗi năm một lần cho người cao tuổi từ 65 tuổi trở lên. Lần khám miễn phí này gồm khám thể chất, công thức máu toàn phần, công thức máu niệu, đường huyết lúc nhịn và điện tâm đồ. Chức năng gan xét men chuyển hoá AST, men chuyển hoá ALT, bilirubin toàn phần. Chức năng thận xét creatinine huyết thanh, urê máu. Mỡ máu xét cholesterol toàn phần, triglyceride, cholesterol mật độ thấp, cholesterol mật độ cao. Cuối cùng còn siêu âm bụng gan mật tụy lách (toàn quốc) - Mức bằng chứng: A -- Nguồn: Krogsbøll LT, Jørgensen KJ, Gøtzsche PC (2019). General health checks in adults for reducing morbidity and mortality from disease. Cochrane Database of Systematic Reviews, 1, CD009009. ；Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委) (2017). 国家基本公共卫生服务规范（第三版）（老年人健康管理服务、高血压患者健康管理服务）. -- Ghi chú: Cái cần bỏ là 「lấy gói khám hằng năm làm biện pháp phòng bệnh」, không phải nói đừng làm xét nghiệm nào cả. Thêm nữa, từ 35 tuổi trở lên mỗi năm đo huyết áp miễn phí một lần, từ 65 tuổi trở lên mỗi năm khám sức khoẻ miễn phí một lần, vốn đã do chương trình dịch vụ công cộng y tế cơ bản quốc gia cấp, không tốn tiền túi mình. Mục nào có lý do đáng xét thì cứ xét. Huyết áp, đường huyết, viêm gan B, xem mục 7, 8 và 14 ở phần 1. Vú, cổ tử cung, trực tràng và đại tràng, xem mục 17, 18 và 19 ở phần 1. Vi khuẩn Helicobacter pylori và CT liều thấp, xem mục 23 và 24 ở phần 1. Đã từng có hành vi nguy cơ cao thì đi xét, xem mục 31 ở phần 1. Từ bao nhiêu tuổi bắt đầu xét, mỗi bao lâu xét lại, đều ghi ở đó. Khám do cơ quan tổ chức, người khác trả tiền thì cứ đi. Cái cần chặn là tự bỏ tiền mua gói và tự thêm mục. Người đã mắc bệnh mạn tính, người có triệu chứng, hoặc xét nghiệm bác sĩ chỉ định theo bệnh thì không tính vào đây, đó là khám bệnh chứ không phải sàng lọc. Mục không có bằng chứng thường gặp nhất trong gói là chất chỉ dấu ung thư và chụp hình toàn thân, xem mục 7 của phần này (gói PET-CT toàn thân và chất chỉ dấu ung thư). Xét ra acid uric máu cao mà chưa từng đau, hay sỏi mật mà chưa từng đau, xem mục 19 của phần này (acid uric cao không triệu chứng) và mục 20 của phần này (sỏi mật không triệu chứng). Ngoài ra, phần lớn thử nghiệm Cochrane thu thập đều làm ở các nước thu nhập cao, nơi khám bệnh vốn đã thuận tiện, nên khu vực thiếu nguồn lực y tế không nên sao chép thẳng kết luận này. +- Nguồn: Krogsbøll LT, Jørgensen KJ, Gøtzsche PC (2019). General health checks in adults for reducing morbidity and mortality from disease. Cochrane Database of Systematic Reviews, 1, CD009009. ；Ủy ban Y tế và Kế hoạch gia đình quốc gia (国家卫生计生委) (2017). Quy định về dịch vụ công cộng cơ bản về y tế quốc gia (bản thứ ba) (dịch vụ quản lý sức khoẻ người cao tuổi, dịch vụ quản lý sức khoẻ người bị cao huyết áp). +- Ghi chú: Cái cần bỏ là「coi gói khám hằng năm là biện pháp phòng bệnh」, không phải bảo mọi xét nghiệm đều không cần làm. Hơn nữa, người từ 35 tuổi được đo huyết áp miễn phí mỗi năm một lần, người trên 65 tuổi có lần khám miễn phí mỗi năm, vốn do chương trình dịch vụ công cộng cơ bản về y tế của nhà nước cấp, không phải bỏ tiền túi. Có lý do phải xét thì cứ xét. Huyết áp, đường huyết, viêm gan B xem mục 7, 8, 14 ở phần 1. Vú, cổ tử cung, trực tràng xem mục 17, 18, 19 ở phần 1. Vi khuẩn H. pylori và CT liều thấp xem mục 23, 24 ở phần 1. Xét mật độ xương cho phụ nữ trên 65 tuổi, xem mục 39 ở phần 1 (xét mật độ xương). Có hành vi nguy cơ cao thì đi xét, xem mục 31 ở phần 1. Từ mấy tuổi bắt đầu xét, bao lâu xét lại một lần đều có ghi ở đó. Khám do đơn vị tổ chức, người khác trả tiền thì cứ đi, cái cần chặn là tự bỏ tiền mua gói và tự thêm xét nghiệm. Người đã có bệnh mạn, người có triệu chứng, hoặc xét nghiệm bác sĩ chỉ định theo bệnh thì không tính, đó là khám bệnh chứ không phải sàng lọc. Xét nghiệm trong gói không có bằng chứng thường gặp nhất là chất chỉ dấu ung thư và chụp toàn thân, xem mục 7 của phần này (gói PET-CT toàn thân và chất chỉ dấu ung thư). Kết quả ra axit uric cao mà chưa từng đau, sỏi mật mà chưa từng đau thì làm gì, xem mục 19 của phần này (axit uric cao không triệu chứng) và mục 20 (sỏi mật không triệu chứng). Ngoài ra, phần lớn thử nghiệm Cochrane gom vào làm ở nước có thu nhập cao, nơi khám bệnh vốn đã thuận tiện, nên khu vực thiếu nguồn lực y tế không nên cứ thế chuyển kết luận này. + ### 19. Đừng bắt đầu uống thuốc hạ acid uric chỉ vì xét sức khoẻ ra acid uric cao mà chưa từng đau lần nào @@ -175,14 +164,14 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Nguồn: FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760（表 1 及其脚注）. ；Badve SV, Pascoe EM, Tiku A, et al. (2020). Effects of Allopurinol on the Progression of Chronic Kidney Disease. New England Journal of Medicine, 382(26), 2504-2513. ；Doria A, Galecki AT, Spino C, et al. (2020). Serum Urate Lowering with Allopurinol and Kidney Function in Type 1 Diabetes. New England Journal of Medicine, 382(26), 2493-2503. - Ghi chú: Cái cần bỏ là 「chưa từng lên cơn mà đã uống thuốc hạ acid uric cả đời」, không phải nói acid uric cao thì không sao. Cùng hướng dẫn đó có liệt kê ngoại lệ. Dưới đây là vài trường hợp có thể cân nhắc bắt đầu uống. Một là lần lên cơn đầu tiên đi kèm bệnh thận mạn tính giai đoạn 3 trở lên. Hai là acid uric máu >9 mg/dL (khoảng 535 µmol/L). Ba là từng bị sỏi đường tiết niệu. Người đã từng lên cơn, người có u tophi, người chụp chiếu thấy xương đã bị xói mòn thì nên uống thuốc. Cụ thể xem mục 9 ở phần 16 (chẩn đoán gout thì uống thuốc hạ acid uric dài hạn). Trong dân tộc Hán, khoảng 7.4% người mang một kiểu gen gọi là HLA-B\*5801, còn người da trắng chỉ 0.7%. Người mang kiểu gen này uống allopurinol dễ bị phát ban thuốc nặng đe dọa tính mạng, nguy cơ hội chứng quá mẫn ở người châu Á gấp 3 lần người da trắng. Đây vốn là loại thuốc có thể chưa cần uống, không thiếu gì một rủi ro lần này. Nếu thật sự muốn bắt đầu uống, hãy đi xét kiểu gen trước, xem mục 9 ở phần 16 (chẩn đoán gout thì uống thuốc hạ acid uric dài hạn). Hai thử nghiệm về thận đó làm trên người bệnh thận mạn tính và người bị đái thái tuổi 1, không thể mở rộng thành 「acid uric cao hoàn toàn vô hại với thận」. Chúng chỉ bác bỏ một điều, là 「uống thuốc hạ acid uric để bảo vệ thận」 -### 20. Đừng cắt túi mật phòng bệnh chỉ vì xét sức khoẻ ra sỏi mật mà chưa từng đau lần nào +### 20. Đừng cắt túi mật phòng bệnh chỉ vì khám ra sỏi mật mà chưa từng đau -- Chi phí: Không tốn tiền, còn tiết kiệm được một ca mổ dưới gây mê toàn thân và một lần nằm viện. Đổi lại bạn phải nhớ vài dấu hiệu nào vừa xuất hiện là phải đi bệnh viện ngay. -- Nói thẳng: Phần lớn người có sỏi trong túi mật nhưng chưa từng đau sẽ không bao giờ đau. Italy theo dõi 580 người như vậy gần chín năm, 78% không hề có triệu chứng. Một nhóm 135 người ở Hoa Kỳ được theo dõi khoảng bốn năm, chỉ 7% phải mổ. Hướng dẫn chẩn đoán và điều trị của Anh viết thẳng: chỉ phát hiện tình cờ trong túi mật có sỏi, không có triệu chứng thì không cần điều trị. -- Lợi ích: Khảo sát dân số MICOL của Italy siêu âm cho 11.229 người từ 29 đến 69 tuổi. Có 856 người được phát hiện sỏi mật, trong đó 580 người không triệu chứng, theo dõi trung bình 8.7 năm. Trong 580 người đó, 453 người (78.1%) không hề có triệu chứng, 61 người (10.5%) có triệu chứng nhẹ, 66 người (11.4%) có triệu chứng nặng. Trong suốt thời gian theo dõi có tổng cộng 189 ca cắt túi mật, trong đó 41.3% làm trên người không triệu chứng. Kết luận của tác giả là 「điều trị chờ xem vẫn là cách xử lý hiệu quả với phần lớn bệnh nhân」, điều trị chờ xem nghĩa là quan sát trước, không mổ. Một HMO ở Hoa Kỳ có số liệu theo dõi, HMO là một loại cơ sở y tế quản lý theo hợp đồng, 691 người được chẩn đoán sỏi mật, theo dõi trung bình 78±61.6 tháng. Trong đó 135 người không triệu chứng được theo dõi 58±50.2 tháng (trung vị 46.3 tháng), chỉ 10% xuất hiện triệu chứng sỏi mật, 7% phải mổ. Trong tổng 50 ca tử vong của cả nhóm chỉ có 2 ca liên quan đến đường mật, và cả hai đều ở nhóm có triệu chứng. Kết luận của tác giả là 「bệnh nhân có sỏi im lặng không cần mổ trước khi xuất hiện triệu chứng」, sỏi im lặng là sỏi không kêu, không lên cơn. Hướng dẫn CG188 của NICE tại điều 1.2.1 viết như sau. 「对胆囊和胆道均正常、偶然发现无症状胆囊结石的人，应告知其在出现症状前不需要治疗」 — với người có túi mật và đường mật bình thường, phát hiện tình cờ sỏi túi mật không triệu chứng, cần thông báo rằng họ không cần điều trị trước khi xuất hiện triệu chứng. Cắt bỏ phòng bệnh trông có lợn vì trực giác 「đã sớm muộn cũng đau, thành cắt một lần lúc còn trẻ」 quá thuận lợi +- Chi phí: Không tốn tiền, còn tiết kiệm một ca mổ gây mê toàn thân và một lần nằm viện. Cái đổi lại là phải nhớ những dấu hiệu nào vừa xuất hiện là phải đi bệnh viện ngay. +- Nói thẳng: Người có sỏi trong túi mật mà chưa từng đau, trong gần chín năm nay khoảng tám phần trăm không có triệu chứng. Không đụng vào nó, mỗi năm tỷ lệ gặp viêm túi mật, viêm tụy cấp loại này khoảng một phần nghìn đến ba phần nghìn. Cắt túi mật cũng không có rủi ro bằng không, khoảng 1.5% người bị tổn thương đường mật. Nên cả hướng dẫn của châu Âu lẫn của nước Anh đều nói không có triệu chứng thì chưa cắt. +- Lợi ích: Khảo sát dân số MICOL của Ý siêu âm cho 11.229 người từ 29 đến 69 tuổi. Ra sỏi ở 856 người, trong đó 580 người không có triệu chứng, theo dõi trung bình 8.7 năm. Trong 580 người đó, 453 người (78.1%) không có triệu chứng, 61 người (10.5%) xuất hiện triệu chứng nhẹ, 66 người (11.4%) xuất hiện triệu chứng nặng. Trong thời gian theo dõi có tổng cộng 189 ca cắt túi mật, trong đó 41.3% làm trên người không có triệu chứng. Kết luận của tác giả là「theo dõi chờ vẫn là cách xử lý có hiệu quả với phần lớn bệnh nhân」, theo dõi chờ nghĩa là quan sát trước, không mổ. Một tổ chức y tế có trả phí ở Mỹ làm thống kê theo dõi, đây là một kiểu cơ sở khám chữa bệnh có quản lý, 691 người được xác nhận có sỏi mật, theo dõi trung bình 78±61.6 tháng. Trong đó 135 người không có triệu chứng được theo dõi 58±50.2 tháng (trung vị 46.3 tháng), chỉ 10% xuất hiện triệu chứng sỏi mật, 7% phải mổ. Trong 50 ca tử vong của toàn nhóm chỉ 2 ca liên quan đến đường mật, và cả hai đều ở nhóm có triệu chứng. Kết luận của tác giả là「bệnh nhân có sỏi không triệu chứng không cần mổ trước khi xuất hiện triệu chứng」, sỏi không triệu chứng là sỏi không kêu, không phát. Hướng dẫn CG188 của NICE tại khoản 1.2.1 viết rằng「Với người có túi mật và đường mật bình thường, phát hiện tình cờ có sỏi túi mật không triệu chứng, cần thông báo là không cần điều trị trước khi xuất hiện triệu chứng」. Hướng dẫn về sỏi mật năm 2016 của Hiệp hội Nghiên cứu bệnh gan châu Âu viết rõ rủi ro của cả hai bên cạnh nhau. Người không có triệu chứng mỗi năm khoảng 0.7% đến 2.5% xuất hiện triệu chứng. Mỗi năm gặp viêm túi mật cấp, viêm tụy cấp, vàng da do tắc mật (đường mật bị nghẽn, người ửng vàng) hoặc viêm đường mật là 0.1% đến 0.3%. Sau cơn đau quặn đầu tiên, biến chứng mới lên tới 1% đến 3% mỗi năm, nên phần lớn người sẽ đau một lần trước khi xảy ra chuyện nặng. Phía cắt trung, trong 51041 ca cắt túi mật ghi trong sổ đăng ký toàn quốc của Thụy Điển có 1.5% bị tổn thương đường mật. Trong đó khoảng một phần năm (0.3%) là đường mật đứt một phần hoặc hoàn toàn. Người bị tổn thương đường mật có tỷ lệ tử vong trong một năm là 3.9%, người không bị là 1.1%. Hướng dẫn còn viết, sau khi cắt có 10% đến 40% bệnh nhân vẫn khó chịu, hoặc khỏi một thời gian rồi lại tái phát. Kết luận của hướng dẫn là: người không có triệu chứng mà mổ, dù mổ hở hay mổ nội soi, đều không kéo dài tuổi thọ, vì rủi ro tử vong và biến chứng của bản thân ca mổ vượt quá rủi ro khi không cắt. Chờ có triệu chứng rồi mới xử lý thì tổng chi phí cũng thấp hơn. Một tính toán năm 1983 đã đưa ra con số cụ thể: ở nam giới 30 tuổi, cắt phòng bệnh so với theo dõi chờ thì sống ít đi trung bình 4 ngày; ở tuổi 50 thì ít đi 18 ngày. Đến nay chưa có thử nghiệm đối chứng chia nhóm ngẫu nhiên nào so cắt với không cắt. Cắt phòng bệnh trông như đáng vì trực giác「đã sớm muộn cũng đau, thà cắt một lần lúc còn trẻ」 - Mức bằng chứng: A -- Nguồn: Festi D, Reggiani ML, Attili AF, et al. (2010). Natural history of gallstone disease: Expectant management or active treatment? Results from a population-based cohort study. Journal of Gastroenterology and Hepatology, 25(4), 719-724. ；McSherry CK, Ferstenberg H, Calhoun WF, Lahman E, Virshup M (1985). The natural history of diagnosed gallstone disease in symptomatic and asymptomatic patients. Annals of Surgery, 202(1), 59-63. ；Viện Quốc gia về Chất lượng và Chăm sóc Sức khoẻ (National Institute for Health and Care Excellence) (2014). Gallstone disease: diagnosis and management. NICE guideline CG188，第 1.2.1、1.3.1 条. -- Ghi chú: Ở đây chỉ nói một tình huống: sỏi nằm trong túi mật và chưa từng đau. **Sỏi ở đường mật chính là chuyện khác.** Điều 1.3.1 của cùng hướng dẫn đó ghi rõ, sỏi ở đường mật chính dù có triệu chứng hay không thì vẫn phải làm sạch sỏi trong đường mật và cắt túi mật. Gặp các tình huống sau thì phải đi bệnh viện ngay trong ngày, đừng cứ dựa vào kết luận này mà chịu đựng: đau quặn liên tục vùng thượng bụng phải hay vùng bụng trên, sốt kèm rét run, lòng trắng mắt hoặc da vàng, nước tiểu đậm màu. Khi bác sĩ xem xét kết quả chụp rồi khuyên mổ, hãy hỏi rõ lý do nào, ví dụ đường mật bị ảnh hưởng, có polyp kèm theo, thành túi mật bất thường, đừng dùng kết luận này để bác bỏ bác sĩ. Nó chỉ chặn một việc là 「không lý do nào, chỉ vì phát hiện sỏi thì cắt」. Hai nghiên cứu nêu trên đều chỉ theo dõi một nhóm người rồi nhìn về sau, không chia nhóm đối chứng. Nhưng chúng muốn trả lời là 「khể đụng vào nó thì sẽ ra sao」 chứ không phải 「cách điều trị nào tốt hơn」, dùng ở đây là hợp lý. Ngoài ra hai nghiên cứu này theo dõi từ 4 đến 9 năm, theo dõi lâu hơn thì tỷ lệ xuất hiện triệu chứng sẽ tăng lên +- Nguồn: Festi D, Reggiani ML, Attili AF, et al. (2010). Natural history of gallstone disease: Expectant management or active treatment? Results from a population-based cohort study. Journal of Gastroenterology and Hepatology, 25(4), 719-724. ；McSherry CK, Ferstenberg H, Calhoun WF, Lahman E, Virshup M (1985). The natural history of diagnosed gallstone disease in symptomatic and asymptomatic patients. Annals of Surgery, 202(1), 59-63. ；National Institute for Health and Care Excellence (2014). Gallstone disease: diagnosis and management. NICE guideline CG188，第 1.2.1、1.3.1 条. ；European Association for the Study of the Liver (2016). EASL Clinical Practice Guidelines on the prevention, diagnosis and treatment of gallstones. Journal of Hepatology, 65(1), 146-181. ；Törnqvist B, Strömberg C, Persson G, Nilsson M (2012). Effect of intended intraoperative cholangiography and early detection of bile duct injury on survival after cholecystectomy: population based cohort study. BMJ, 345, e6457. ；Ransohoff DF, Gracie WA, Wolfenson LB, Neuhauser D (1983). Prophylactic cholecystectomy or expectant management for silent gallstones. A decision analysis to assess survival. Annals of Internal Medicine, 99(2), 199-204. ；Gurusamy KS, Samraj K (2007). Cholecystectomy versus no cholecystectomy in patients with silent gallstones. Cochrane Database of Systematic Reviews, CD006230. ；反方：Venneman NG, Buskens E, Besselink MG, et al. (2005). Small gallstones are associated with increased risk of acute pancreatitis: potential benefits of prophylactic cholecystectomy? American Journal of Gastroenterology, 100(11), 2540-2550. ；中华医学会外科学分会胆道外科学组, 中国医师协会外科医师分会胆道外科医师委员会 (2022). 胆囊良性疾病外科治疗的专家共识（2021版）. 中华外科杂志, 60(1), 4-9. （全文需订阅，未取得）；同上 (2022). 《胆囊良性疾病外科治疗的专家共识（2021版）》解读. 中华外科杂志, 60(4), 337-342. +- Ghi chú:Tranh cãi. Chuyên đồng thuận của các chuyên gia bản 2021 trong phẫu thuật ngoại khoa đường mật tại Trung Quốc xếp một phần sỏi túi mật không triệu chứng vào chỉ định phẫu thuật. Bài giải thích trong cùng tạp chí nói rõ một trong các điểm còn tranh cãi nằm ở đây. Bản toàn văn chuyên đồng thuận phải trả phí để đọc, cụ thể là mấy trường hợp nào thì chưa đối chiếu được với nguyên văn (TODO: cần đối chiếu). Ý kiến phản biện ở một chỗ khác là sỏi nhỏ: sỏi kích thước dưới 5 milimét liên quan nhiều hơn với viêm tụy. Một tính toán ở Hà Lan nói loại người này cắt phòng bệnh có thể sống thêm hoặc sống ít hơn, tuỳ viêm tụy phổ biến đến đâu và chết người đến đâu. Hướng dẫn của châu Âu cũng liệt kê ngoại lệ: vỏ túi mật vôi hoá (túi mật sứ), có đồng thời polyp túi mật trên 1 centimet, hai loại này nên cắt. Khi cần mổ bụng lớn vì lý do khác, có thể cân nhắc cắt luôn. Mục này nói về đúng một tình huống: sỏi nằm trong túi mật và chưa từng đau. **Sỏi ở ống mật chủ là chuyện khác**. Khoản 1.3.1 của cùng hướng dẫn đó viết rõ, sỏi ở ống mật chủ dù có triệu chứng hay không đều phải làm sạch sỏi trong đường mật và cắt túi mật. Gặp các tình huống sau thì ngày hôm đó phải đi bệnh viện, đừng cứ dựa vào kết luận này mà chịu: đau quặn kéo dài ở hạ sườn phải hoặc vùng thượng bụng, sốt điên rối, lòng trắng mắt hoặc da vàng, nước tiểu sậm màu. Khi bác sĩ xem ảnh rồi khuyên mổ thì hỏi rõ ông ta căn cứ vào lý do nào, chẳng hạn đường mật bị ảnh hưởng, có polyp đi kèm, vỏ túi mật bất thường, đừng dùng kết luận này để phản bác ông ta. Nó chỉ chặn việc「không có lý do gì, chỉ vì thấy sỏi là cắt」. Hai nghiên cứu nêu trên đều chỉ theo dõi một nhóm người đi tiếp, không có nhóm đối chứng. Nhưng chúng muốn trả lời là「không đụng vào thì sẽ ra sao」, chứ không phải「cách điều trị nào tốt hơn」, dùng ở đây là hợp lý. Ngoài ra hai nghiên cứu này theo dõi từ 4 đến 9 năm, nếu lâu hơn nữa thì tỷ lệ xuất hiện triệu chứng sẽ tăng lên ### 21. Đừng bỏ canxi hoàn toàn để phòng sỏi thận @@ -201,7 +190,6 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: B - Nguồn: Forer BR (1949). The fallacy of personal validation: A classroom demonstration of gullibility. Journal of Abnormal and Social Psychology. ；Cục Quản lý thị trường quốc gia (国家市场监督管理总局) (2026). 市场监管总局公布一批传统工艺市场「打假清源」典型案例. - Ghi chú: Kết luận chỉ có một: công năng của nó không thể kiểm chứng, và không đo được khả năng dự báo nào. Không đánh giá tín ngưỡng hay phong tục nào, cũng không nhắm vào các mục đích tặng quà, lưu niệm, truyền lại. Cần tách hai việc: công năng mà người bán tuyên bố có quy định văn bản làm căn cứ, có thể đi khiếu nại; còn các lời như 「đổi vận」 vừa không có nghiên cứu, vừa không chứng minh được là giả, chỉ có thể tự kiềm chế bằng ngân sách. Trước khi mua, hãy coi nó là món trang sức thường, tự hỏi một câu 「nếu nó không thay đổi gì cả, giá này tôi có mua không」. Cách kiểm tra chất liệu và giấy chứng nhận, vì sao coi nó là đầu tư là không đáng, xem mục 33 và 34 ở phần 5 (chuỗi tay và đồng hồ đắt tiền tính theo số tiền đã tiêu, đá quý thì nhận báo cáo CMA). Bỏ tiền xem bói thì xem mục 15 của phần này (xem bói, tarot, chiêm tinh). - ### 23. Đừng trông chờ mua sắm để cải thiện tâm trạng hay cảm giác về vị thế của mình - Chi phí: Không tốn tiền. Chỗ khó là lúc tâm trạng tệ mà nhịn không đặt hàng. @@ -210,33 +198,33 @@ Phần này gom những thứ「trông rẻ mà lại có tác dụng」. Mở n - Mức bằng chứng: B - Nguồn: Dittmar H, Bond R, Hurst M, Kasser T (2014). The relationship between materialism and personal well-being: A meta-analysis. Journal of Personality and Social Psychology, 107(5), 879-924. ；Van Boven L, Gilovich T (2003). To do or to have? That is the question. Journal of Personality and Social Psychology, 85(6), 1193-1202. - Ghi chú:Tranh cãi: Phần lớn nghiên cứu được thu thập hỏi một lớp người tại cùng một thời điểm, xem hai thứ có khớp nhau không, không phân biệt được cái nào trước cái nào. Chính tác giả cũng viết trong kết luận rằng muốn thấy rõ chiều hướng thì cần thêm thử nghiệm và nghiên cứu theo dõi dài hạn. Vậy không thể đọc thành 「mua đồ làm người ta không vui」, chỉ có thể đọc thành 「trông chờ mua đồ để thành vui thì không mấy khả thi」. Hơn nữa liên hệ rất yếu: r = -0.19 chỉ giải thích được vài phần trăm khác biệt, giữa người với người rất khác nhau. Cả hai nghiên cứu đều dùng mẫu ở nước ngoài, không có dữ liệu cùng loại cho người Trung Quốc để viện dẫn. Ở đây không phản đối việc bỏ tiền cho thứ mình thích, cái bị phản đối là coi mua sắm là cách điều chỉnh tâm trạng, vì làm thế thì lần sau tâm trạng tệ bạn vẫn đặt hàng. Cụ thể tiết kiệm tiền thì xem phần 5. Làm gì khi tâm trạng tệ mà thật sự có bằng chứng, xem nửa sau của phần 22 và phần 29. Khoản ngân sách cộng thêm để 「mạnh hơn người ta một bậc」, xem mục 24 của phần này (dời lên một bậc). - -### 24. Đừng bỏ thêm tiền đổi nhà, đổi xe, đổi giới xung quanh chỉ để 「dời lên một bậc trong nhóm người xung quanh」 +### 24. Đừng bỏ thêm tiền đổi nhà, đổi xe, đổi giới xung quanh chỉ để「được lên một bậc」 -- Chi phí: Không tốn tiền. Việc phải làm chỉ là trước khi cộng thêm ngân sách, hỏi thêm một câu: khoản chênh lệch này mua chính cái đồ, hay mua vị trí của tôi trong nhóm người đó. Chỗ khó là khi mọi người xung quanh đều đổi, mà bạn không theo. -- Nói thẳng: So với người có thu nhập ngang nhau, hàng xóm càng kiếm được nhiều thì người đó càng báo cáo hạnh phúc thấp hơn. Điều này chỉ xảy ra với hàng xóm có cùng trình độ học vật và hay gặp gỡ. Nên khoản tiền bỏ thêm để dọn vào nơi mà ai cũng dư dả hơn mình, cái mua được là đẩy mình xuống một bậc. Nhưng khi thu nhập của cả hai cùng tăng, người ta vẫn thấy dễ sống hơn, mức tuyệt đối vẫn có tác dụng. -- Lợi ích: Khảo sát Gia đình và Hộ gia đình toàn quốc Hoa Kỳ (NSFH) phỏng vấn lại cùng một lứ người trong hai vòng. Tác giả đối chiếu bộ dữ liệu này với thu nhập trung bình địa phương. Thu nhập trung bình địa phương (PUMA) được suy ra từ mẫu 5% của tổng điều tra dân số năm 1990. Sau khi trừ thu nhập của chính người đó và một loạt đặc điểm cá nhân, kết quả là: thu nhập của hàng xóm càng cao thì hạnh phúc tự báo của bản thân càng thấp. Hiệu ứng lớn, đổi cách tính vẫn giữ nguyên, và rất có ý nghĩa. Cách diễn đạt của tác giả là mức hạnh phúc giảm do 「thu nhập hàng xóm tăng」 và do 「thu nhập của bản thân giảm cùng mức đó」 nằm ở cùng một bậc lớn. Khi đổi sang so một người với chính họ trước và sau (gọi là thêm hiệu ứng cố định cá nhân), hệ số giữ nguyên dấu và gần nguyên giá trị, chỉ có mức ý nghĩa giảm xuống còn 10%. Khi giới hạn lại để chỉ so trong cùng một tiểu bang và cùng một vòng khảo sát (gọi là hiệu ứng cố định tiểu bang và vòng khảo sát), kết quả hầu như không đổi. Điều đó cho thấy hiệu ứng không phải do 「người hay chuyển tới nơi nghèo vốn đã vui hơn」, cũng không phải do một yếu tố chưa đo được ở cấp tiểu bang. Hiệu ứng mạnh hơn ở người hay gặp hàng xóm, không có ý nghĩa ở người chỉ gặp bạn bè ngoài khu phố. Người thuê nhà và người có nhà không khác nhau, điều này không ủng hộ cách giải thích 「chỗ đắt thì giá nhà cao, nên số tiền thực chi tiêu được còn ít」. Hạnh phúc chỉ giảm theo thu nhập của hàng xóm cùng trình độ học vật, gần như không phản ứng với thu nhập của hàng xóm khác trình độ. Hơn nữa nó chủ yếu truyền qua 「mức hài lòng với tình cảnh vật chất của bản thân」, không phải qua mức hài lòng với các mặt khác như đời sống gia đình. Cuối cùng, nếu thu nhập của bản thân và hàng xóm cùng tăng theo cùng tỷ lệ, hạnh phúc vẫn tăng. +- Chi phí: Không tốn tiền. Việc phải làm chỉ là trước khi cộng thêm vào ngân sách thì hỏi thêm một câu: khoản tiền cộng thêm này mua chính cái đồ, hay mua vị trí của tôi trong nhóm người đó. Chỗ khó là lúc tất cả mọi người xung quanh đều đổi mà bạn không theo. +- Nói thẳng: So với những người có thu nhập bằng nhau, hàng xóm kiếm càng nhiều thì người đó báo cáo hạnh phúc càng thấp. Hiện tượng này chỉ xảy ra giữa những người hàng xóm có cùng trình độ và hay qua lại. Nên số tiền phải bỏ thêm để dọn vào một nơi mà ai cũng dư dả hơn mình thì mua được là đẩy chính mình xuống một bậc. Nhưng khi thu nhập cùng tăng, người ta vẫn thấy dễ chịu hơn, mức tuyệt đối vẫn có tác dụng. +- Lợi ích: Khảo sát về hộ gia đình và dân cư của toàn nước Mỹ (NSFH) đã hai lần quay lại phỏng vấn cùng một nhóm người. Tác giả đối chiếu dữ liệu này với thu nhập trung bình của địa phương. Thu nhập trung bình của địa phương (PUMA) được suy ra từ mẫu 5% của tổng điều tra dân số năm 1990. Sau khi trừ thu nhập bản thân và một loạt đặc điểm cá nhân, kết quả là: thu nhập của hàng xóm càng cao thì hạnh phúc người tự báo cáo càng thấp. Hiệu ứng lớn, đổi cách tính vẫn ổn định, và rất có ý nghĩa. Cách diễn đạt của tác giả là hiệu ứng「thu nhập của hàng xóm tăng」và「thu nhập của chính mình giảm cùng mức đó」làm hạnh phúc giảm ở cùng một cấp độ. Khi đổi sang so chính người đó với chính mình trước sau (gọi là thêm hiệu ứng cố định cá nhân), hệ số không đổi chiều, độ lớn cũng gần, chỉ có mức ý nghĩa giảm xuống còn 10%. Khi giới hạn lại trong cùng một tiểu bang và giữa các vòng khảo sát (gọi là hiệu ứng cố định tiểu bang và vòng khảo sát), kết quả gần như không đổi. Điều này cho thấy nó không phải do「người hay chuyển đến nơi nghèo vốn vui hơn」, cũng không phải do một yếu tố chưa đo được ở cấp tiểu bang. Hiệu ứng mạnh hơn ở người hay qua lại với hàng xóm, không đáng kể ở người chỉ chơi với bạn bè ngoài khu phố. Người thuê nhà và người có nhà không khác nhau, điều này không ủng hộ cách giải thích「chỗ đắt thì giá nhà cao, nên số tiền thật sự còn lại ít」. Hạnh phúc chỉ giảm theo thu nhập của người hàng xóm cùng trình độ, gần như không phản ứng với thu nhập của người hàng xóm khác trình độ. Hơn nữa nó chủ yếu truyền qua「mức hài lòng với hoàn cảnh vật chất của bản thân」, không phải qua mức hài lòng với các mặt khác như đời sống gia đình. Cuối cùng, nếu thu nhập của chính mình và của hàng xóm cùng tăng theo cùng tỷ lệ thì hạnh phúc vẫn tăng - Mức bằng chứng: B -- Nguồn: Luttmer EFP (2005). Neighbors as Negatives: Relative Earnings and Well-Being. The Quarterly Journal of Economics, 120(3), 963–1002. ；Luttmer EFP (2004). Neighbors as Negatives: Relative Earnings and Well-Being. NBER Working Paper No. 10667（phần diễn giải và cỡ lớn được viện dẫn ở mục này được đối chiếu từng chữ theo bản này）. -- Ghi chú: Có vài lý do đặt mức B: dữ liệu đến từ bảng hỏi có phỏng vấn lại nhiều lần cùng một lứ người, chỉ có đúng nghiên cứu này, mẫu lại ở nước ngoài; kết quả đo là hạnh phúc do người trả lời tự báo, không phải tiền. Tác giả đúng là đã dùng thống kê để loại vài loại nhiễu, ví dụ so một người trước và sau, so trong cùng tiểu bang và cùng vòng, nhưng xét cho cùng không phải thử nghiệm ngẫu nhiên. Người Trung Quốc cũng không có dữ liệu cùng loại để viện dẫn. Cỡ lợi ích đặt 「vừa」 là theo đúng cách tính của mục 23 của phần này (đừng trông chờ mua sắm cải thiện tâm trạng). Số tiền tiết kiệm được đúng là có thể lên tới mức vạn tệ. Nhưng 「tiết kiệm được khoản này đổi lại bao nhiêu hạnh phúc」 thì không có con số nào đỡ, nên không tính là 「lớn」 ở mức vạn tệ. Ở đây không phản đối việc bỏ tiền cho căn nhà rộng hơn, khu chung cư yên tĩnh hơn, quãng đường đi làm ngắn hơn, những thứ đó mua chính cái đồ. Cái cần chặn là lý do 「đồng nghiệp đều đổi, mình không được kém họ」. Xếp ưu tiên khi chọn chỗ ở thì xem mục 18 ở phần 4 (khi chọn chỗ ở, đặt thời gian đi làm lên đầu). Mua sắm để điều chỉnh tâm trạng thì xem mục 23 của phần này (đừng trông chờ mua sắm cải thiện tâm trạng). Trên mạng cứ so lên xem mình ở đâu thì xem mục 21 ở phần 3 (đừng coi 「người ta sống thế nào」 là bài đọc bắt buộc mỗi ngày) +- Nguồn: Luttmer EFP (2005). Neighbors as Negatives: Relative Earnings and Well-Being. The Quarterly Journal of Economics, 120(3), 963–1002. ；Luttmer EFP (2004). Neighbors as Negatives: Relative Earnings and Well-Being. NBER Working Paper No. 10667（bản thảo làm việc, câu diễn đạt và cấp độ được dẫn trong mục này đối chiếu theo đúng bản đó）. +- Ghi chú: Lý do đặt mức B có vài điểm: dữ liệu đến từ bảng hỏi có quay lại cùng một nhóm người nhiều lần, chỉ có một nghiên cứu, mẫu lại ở nước ngoài; kết quả đo là hạnh phúc do người trả lời tự báo, không phải tiền. Tác giả quả thật đã dùng thống kê để loại vài loại nhiễu, chẳng hạn so trước sau của cùng một người, so giữa các vòng khảo sát trong cùng một tiểu bang, nhưng cuối cùng vẫn không phải thử nghiệm ngẫu nhiên. Người Trung Quốc cũng không có dữ liệu cùng loại để viện dẫn. Mức lợi ích đặt là「vừa」là theo đúng cách tính của mục 23 của phần này (đừng trông chờ mua sắm để cải thiện tâm trạng). Số tiền tiết kiệm được đúng là có thể lên tới mức vạn tệ. Nhưng「tiết kiệm được số tiền đó thì đổi lại được bao nhiêu hạnh phúc」không có số liệu nào đỡ, nên không tính là「lớn」ở mức vạn tệ. Ở đây không phản đối việc bỏ tiền cho căn nhà lớn hơn, khu chung cư yên tĩnh hơn, đi làm ngắn hơn, những thứ đó mua chính cái đồ. Cái cần chặn là một lý do duy nhất:「đồng nghiệp đều đổi, tôi không thể kém họ」. Nên xếp hạng nơi ở theo tiêu chí gì, xem mục 18 ở phần 4 (khi chọn nơi ở thì đặt thời gian đi làm lên trước). Mua sắm để điều chỉnh tâm trạng, xem mục 23 của phần này (đừng trông chờ mua sắm để cải thiện tâm trạng). Luôn so với người khác trên mạng, xem mục 20 ở phần 3 (đừng coi「người ta sống thế nào」 là mục phải đọc mỗi ngày) -### 25. Đừng tin 「ý chí giống cơ bắp, dùng một lúc là hết」 +### 25. Đừng tin「ý chí giống cơ, dùng một lát là hết」 -- Chi phí: Không tốn tiền. Cái cần sửa chỉ là một cách nói. -- Nói thẳng: Cách nói phổ biến là làm xong một việc tốn sức tự kiểm soát thì việc kế tiếp dễ phá quy tắc hơn, nên phải dùng ý chí tiết kiệm. 23 phòng thí nghiệm, 2141 người làm lại thí nghiệm này theo cùng một bộ phương án, kết quả gần như bằng không. Đừng dùng 「hôm nay ý chí đã hết」 để giải thích cho bản thân nữa, cũng đừng bỏ tiền mua khoá học dạy cách tích trữ ý chí. -- Lợi ích: Một nghiên cứu lặp lại có đăng ký trước, nhiều phòng thí nghiệm, 23 phòng thí nghiệm, tổng cộng 2141 người tham gia, làm lại mô hình nhiệm vụ tuần tự theo bộ phương án chuẩn hoá của Sripada và cộng sự: làm trước một nhiệm vụ tiêu tốn sức tự kiểm soát, rồi làm nhiệm vụ thứ hai cũng cần sức tự kiểm soát. Kết quả gộp cho thấy hiệu ứng hao mòn bản thân rất nhỏ, và 95% khoảng tin cậy có chứa 0 (d = 0.04, 95% CI −0.07 đến 0.15). Tác giả dựa vào đó bàn về việc mô hình 「tự kiểm soát là một nguồn lực hữu hạn」 còn đứng vững hay không. +- Chi phí: Không tốn tiền. Cái cần đổi chỉ là một cách nói. +- Nói thẳng: Cách nói phổ biến là: làm xong một việc tốn ý chí thì việc sau dễ vỡ hơn, nên phải tiết kiệm ý chí mà dùng. 23 phòng thí nghiệm, 2141 người làm lại thí nghiệm này theo cùng một bộ phương pháp, kết quả gần như bằng không. Đừng dùng「hôm nay ý chí đã hết」 để giải thích cho bản thân nữa, cũng đừng bỏ tiền mua lớp dạy cách tích trữ ý chí vì chuyện đó. +- Lợi ích: Một nghiên cứu lặp lại đa phòng thí nghiệm có đăng ký trước, 23 phòng thí nghiệm, tổng cộng 2141 người tham gia, làm lại mô hình nhiệm vụ tuần tự theo bộ phương pháp chuẩn hoá của Sripada và cộng sự: làm trước một nhiệm vụ tiêu tốn tự kiểm soát, rồi làm nhiệm vụ thứ hai cần tự kiểm soát. Kết quả gộp cho thấy hiệu ứng tự hao mòn rất nhỏ, và khoảng tin cậy 95% chứa 0 (d = 0.04, 95% CI −0.07 đến 0.15). Tác giả dựa vào đó bàn xem mô hình「tự kiểm soát là nguồn lực hữu hạn」còn đứng vững không. Một lần làm lại khác cũng đăng ký trước phương pháp, 36 phòng thí nghiệm, 3531 người, cỡ hiệu ứng d = 0.06, nhỏ đến mức không tính được. - Mức bằng chứng: A -- Nguồn: Hagger MS, Chatzisarantis NLD, Alberts H, et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science, 11(4), 546–573. -- Ghi chú: Cái cần bỏ là mô hình 「tự kiểm soát là một nguồn lực bị hao mòn hết」, không phải phủ nhận 「người ta sẽ mệt」. Ngủ thiếu và làm việc liên tục quá lâu khiến hiệu suất giảm, đó là chuyện khác, xử lý theo mấy mục về giấc ngủ và giờ làm ở phần 3. Cách làm thật sự có thử nghiệm chia nhóm ngẫu nhiên đỡ là đổi môi trường và đổi cách viết, xem mục 10 ở phần 4 (đưa món không muốn chạm tới ra xa) và mục 1 ở phần 4 (viết thành 「mấy giờ, ở đâu, gặp gì thì làm gì」). Những mục khác của phần này tính bằng tiền, riêng mục này tính bằng thời gian và sức lực, hai bên không quy đổi qua lại. +- Nguồn: Hagger MS, Chatzisarantis NLD, Alberts H, et al. (2016). A Multilab Preregistered Replication of the Ego-Depletion Effect. Perspectives on Psychological Science, 11(4), 546–573. ；Vohs et al. (2021). A Multisite Preregistered Paradigmatic Test of the Ego-Depletion Effect. Psychological Science. +- Ghi chú: Cái cần bỏ là mô hình「tự kiểm soát là nguồn lực sẽ bị hao mòn hết」, không phải bỏ ý niệm「người ta sẽ mệt」. Ngủ không đủ giấc, làm việc liên tục quá lâu khiến hiệu quả giảm, đó là chuyện khác, xử lý theo mấy mục về giấc ngủ và giờ làm ở phần 3. Cách làm thật sự có thử nghiệm ngẫu nhiên hỗ trợ là đổi môi trường và đổi cách viết, xem mục 10 ở phần 4 (đưa thứ mình không muốn đụng vào ra xa hơn) và mục 1 ở phần 4 (viết thành「mấy giờ, ở đâu, gặp gì thì làm việc đó」). Cố định sẵn mặc gì, ăn gì vẫn đáng làm, vì gần như không tốn chi phí. Nó tiết kiệm được mấy phút và vài lần do dự, không phải tiết kiệm ý chí, đừng trông chờ nó giúp bạn suy nghĩ tốt hơn vào buổi chiều. Các mục khác của phần này tính ra tiền, ở đây tính ra thời gian và sức lực, hai bên không quy đổi qua lại. + -### 26. Đừng trông chờ ăn sáng hay chế độ nhịn ăn 16:8 giúp giữ cân nặng, hãy chọn khung giờ ăn nào bạn giữ được lâu dài +### 26. Đừng trông chờ ăn sáng hay chế độ ăn 16:8 giúp bạn kiểm soát cân nặng, hãy chọn khung giờ ăn mà bạn theo được lâu dài -- Chi phí: Không tốn tiền. Phần tiết kiệm được là chút sức lực mà mỗi ngày bạn phải tranh luận với giờ ăn. -- Nói thẳng: Cách nói phổ biến là bỏ bữa sáng sẽ khiến người ta ăn bùng ở bữa trưa và bữa tối, tổng cả ngày lại nhiều hơn. Trong thử nghiệm thì không phải: người không ăn sáng mỗi ngày ăn ít hơn hơn hai trăm kcal, cân nặng còn nhẹ hơn một chút. So với chỉ ăn ít đi, chế độ 16:8 sau một năm cũng không khác biệt về cân nặng. Người lâu ngày không ăn sáng quả thật mắc bệnh tim mạch nhiều hơn khoảng một phần mươi bảy, còn tỷ lệ tử vong toàn nguyên thì không thấy khác biệt. -- Lợi ích: Xem trước lời nói 「bỏ bữa sáng sẽ khiến người ta ăn nhiều hơn về sau」. Một bài phân tích tổng hợp gộp 13 thử nghiệm ngẫu nhiên. Người được xếp vào nhóm ăn sáng mỗi ngày ăn vào nhiều hơn nhóm không ăn sáng 259.79 kcal (95% CI 78.87–440.71, đây là khoảng tin cậy). Cân nặng cao hơn 0.44 kg (0.07–0.82). Người không ăn sáng không bù lại ở bữa trưa và bữa tối. Các thử nghiệm này theo dõi rất ngắn, theo cân nặng trung bình 7 tuần, theo lượng ăn trung bình 2 tuần. Xem tiếp chế độ 16:8, tức chỉ ăn trong một khoảng giờ cố định. Một thử nghiệm cho 116 người lớn thừa cân làm trong 12 tuần. Nhóm chỉ ăn từ 12 giờ trưa tới 8 giờ tối và nhóm ăn đều ba bữa mỗi ngày chênh nhau −0.26 kg (−1.30 đến 0.78), không có ý nghĩa. Một thử nghiệm khác làm ở Quảng Châu, 139 người bệnh béo phì, theo dõi 12 tháng. Ăn trong khung giờ kèm kiểm soát calo giảm nhiều hơn chỉ kiểm soát calo 1.8 kg (−4.0 đến 0.4), cũng không có ý nghĩa. Cuối cùng xem bỏ bữa sáng lâu dài có gây hại không. Trong tổng hợp 2.380.000 người, người không ăn sáng có nguy cơ bệnh tim mạch cao hơn khoảng 17% (OR 1.17, 1.09–1.26), tử vong do tim mạch cao hơn khoảng 49% (OR 1.49, 1.20–1.84). Nhưng đây là những nghiên cứu chỉ ghi nhận, không chia nhóm. Trong nghiên cứu gốc theo dõi từ 17 đến 23 năm, tỷ số nguy cơ tử vong toàn nguyên là 1.19, khoảng tin cậy 0.99 đến 1.42, vượt qua 1, không tính là khác biệt. +- Chi phí: Không tốn tiền. Phần tiết kiệm được là chút sức lực mỗi ngày phải tranh cãi với khung giờ ăn. +- Nói thẳng: Cách nói phổ biến là bỏ ăn sáng sẽ khiến trưa và tối ăn hỉn, tổng cả ngày lại nhiều hơn. Thử nghiệm thì không phải: người không ăn sáng mỗi ngày ăn ít hơn hơn hai trăm calo, cân nặng còn nhẹ hơn một chút. So với chỉ giảm ăn, chế độ 16:8 sau một năm cân nặng cũng không khác. Người lâu ngày không ăn sáng thì mắc bệnh tim mạch đúng là nhiều hơn khoảng một phần bảy, còn tỷ lệ tử vong toàn nguyên thì không thấy khác biệt. +- Lợi ích: Xem cách nói「bỏ ăn sáng sẽ khiến người đó ăn nhiều hơn về sau」đã. Một phân tích tổng hợp gom 13 thử nghiệm ngẫu nhiên lại. Người được xếp vào nhóm ăn sáng, lượng nạp mỗi ngày cao hơn nhóm không ăn sáng 259.79 kcal (95% CI 78.87–440.71, đây là khoảng tin cậy). Cân nặng cao hơn 0.44 kg (0.07–0.82). Người không ăn sáng không phải là gộp bù lại vào bữa trưa và bữa tối. Thời gian theo dõi của các thử nghiệm này đều ngắn, xét cân nặng trung bình 7 tuần, xét lượng nạp trung bình 2 tuần. Xem tiếp chế độ 16:8, tức chỉ ăn trong một khoảng giờ cố định. Một thử nghiệm cho 116 người trẻ thành thừa cân làm 12 tuần. Nhóm chỉ ăn từ 12 giờ trưa đến 8 giờ tối so với nhóm ăn đúng ba bữa mỗi ngày, chênh lệch cân nặng là −0.26 kg (−1.30 đến 0.78), không có ý nghĩa. Một thử nghiệm khác làm ở Quảng Châu, 139 bệnh nhân béo phì, theo dõi 12 tháng. Ăn trong khung giờ kèm kiểm soát calo, so với chỉ kiểm soát calo thì giảm nhiều hơn 1.8 kg (−4.0 đến 0.4), cũng không có ý nghĩa. Cuối cùng xem việc lâu ngày không ăn sáng có hại không. Trong tổng hợp 2.380.000 người, người không ăn sáng có nguy cơ bệnh tim mạch cao hơn khoảng 17% (OR 1.17, 1.09–1.26), tử vong do tim mạch cao hơn khoảng 49% (OR 1.49, 1.20–1.84). Nhưng đây là những nghiên cứu chỉ ghi nhận, không chia nhóm. Trong nghiên cứu gốc theo dõi 17 đến 23 năm, tỷ số nguy cơ tử vong toàn nguyên là 1.19, khoảng tin cậy 0.99 đến 1.42, có vượt qua 1, coi như không khác biệt. - Mức bằng chứng: A - Nguồn: Sievert K, Hussain SM, Page MJ, 等 (2019). Effect of breakfast on weight and energy intake: systematic review and meta-analysis of randomised controlled trials. BMJ. ；Lowe DA, Wu N, Rohdin-Bibby L, 等 (2020). Effects of Time-Restricted Eating on Weight Loss and Other Metabolic Parameters in Women and Men With Overweight and Obesity. JAMA Internal Medicine. ；Liu D, Huang Y, Huang C, 等 (2022). Calorie Restriction with or without Time-Restricted Eating in Weight Loss. New England Journal of Medicine. ；Zhang H, Zhang S, Liu Y, 等 (2025). The association between skipping breakfast and cardiovascular disease: a meta analysis. Frontiers in Cardiovascular Medicine. ；Rong S, Snetselaar LG, Xu G, 等 (2019). Association of Skipping Breakfast With Cardiovascular and All-Cause Mortality. Journal of the American College of Cardiology. ；Fernandes-Alves D, Teixeira GP, Guimarães KC, Crispim CA (2026). Systematic Review and Meta-analysis of Randomized Clinical Trials Comparing Time-Restricted Eating With and Without Caloric Restriction for Weight Loss. Nutrition Reviews. ；Sun T, Zhang L, Lu Y, 等 (2024). Non-linear relationship between the first meal time of the day and gallstone incidence in American adults. Frontiers in Nutrition. ；Liu T, Wang Y, Wang X, 等 (2023). Habitually Skipping Breakfast Is Associated with the Risk of Gastrointestinal Cancers: Evidence from the Kailuan Cohort Study. Journal of General Internal Medicine. -- Ghi chú: Tranh cãi. Con số 17% về bệnh tim mạch đến từ những nghiên cứu chỉ ghi nhận, không chia nhóm. Người không ăn sáng vốn đã dễ hút thuốc, uống rượu, ngồi nhiều, thu nhập thấp hơn, mà không hiệu chỉnh sạch được. Trong 17% đó có bao nhiêu là do bữa sáng, nói không rõ. Chỉ số tử vong toàn nguyên vốn đã vượt qua 1. Chế độ 16:8 cũng còn tranh cãi: gộp 30 thử nghiệm, 1341 người, với điều kiện hai nhóm ăn bằng nhau, nhóm ăn trong khung giờ giảm nhiều hơn 1.46 kg (−2.65 đến −0.26), đổi lại mất thêm 0.41 kg khối lượng cơ. Nhưng phần lớn các thử nghiệm này rất ngắn, thử nghiệm dài và nghiêm nhất là thử nghiệm mười hai tháng thì nói không có khác biệt. Nhịn ăn lâu còn liên quan tới túi mật. Trong một khảo sát cắt ngang 6.547 người, bữa đầu tiên trong ngày muộn hơn mỗi giờ một chút thì tỷ lệ phát hiện sỏi mật cao hơn khoảng 5% (OR 1.05, 1.02–1.08); người ăn bữa đầu muộn từ 9 giờ tới 14 giờ cao hơn khoảng 49% so với người ăn sớm hơn (OR 1.49, 1.24–1.77). Cơ chế là mật đọng trong túi mật càng lâu càng dễ kết tinh thành sỏi. Nhưng đây là khảo sát tại một thời điểm, cũng có thể là do đã có vấn đề túi mật rồi mới đổi thói quen ăn. Nghiên cứu mẫu Kailuan ở Trung Quốc theo dõi 5.6 năm, có 369 ca ung thư đường tiêu hoá: người không ăn sáng có nguy cơ ung thư đại tràng và trực tràng gấp 2.32 lần (1.34–4.01), ung thư túi mật và đường mật ngoài gan gấp 5.43 lần (1.34–21.93). Khoảng tin cậy của con số sau rộng tới 21.93, nghĩa là số ca quá ít, chỉ coi là manh mối. Gộp lại, xét riêng chuyện giữ cân nặng thì giờ ăn ảnh hưởng rất nhỏ, đừng phí tâm lực vì nó. Còn chuyện bỏ hẳn bữa sáng lâu dài có tốt không, bằng chứng hiện có toàn là tương quan, không chứng minh được nhân quả, nhưng hướng thì nhất quán nghiêng về bất lợi; muốn an toàn thì ăn bình thường, chỉ đừng trông nó giúp giảm cân. Nguyên nhân chính của viêm dạ dày và loét dạ dày không phải là bụng rỗng, mà là Helicobacter pylori và uống thuốc giảm đau lâu dài; mục nào cần xét thì xem mục 23 ở phần 1 (xét Helicobacter pylori, dương tính thì diệt). Xét sức khoẻ ra sỏi mật mà chưa từng đau thì làm gì, xem mục 20 của phần này (xét sức khoẻ ra sỏi mật mà chưa từng đau, đừng cắt túi mật phòng bệnh). Muốn giữ cân nặng, thứ có bằng chứng là ăn gì và ăn bao nhiêu, xem mục 28 ở phần 2 (mỗi ngày đủ 5 phần rau quả), mục 29 ở phần 2 (ăn ít thực phẩm chế biến công nghiệp) và mục 33 ở phần 2 (giữ BMI ở 20–25). Người bị đái thái, người đang uống thuốc hạ đường huyết, hay phụ nữ mang thai thì không áp dụng, bỏ một bữa có thể hạ đường huyết, hãy làm theo chỉ định của bác sĩ. Nhịn ăn cực đoan và nôn kích thực là chuyện khác, xem mục 1 ở phần 28 (đừng dùng ăn kiêu cực đoan, nhịn ăn hoặc nôn kích thực để giữ cân nặng). +- Ghi chú:Tranh cãi. Con số 17% về bệnh tim mạch đến từ nghiên cứu chỉ ghi nhận, không chia nhóm. Người không ăn sáng vốn dễ hút thuốc, uống rượu, ngồi nhiều, thu nhập thấp hơn, các yếu tố đó hiệu chỉnh không sạch. Nên trong 17% đó có bao nhiêu do bữa sáng bản thân gây ra thì không nói rõ. Chỉ số tử vong toàn nguyên vốn đã vượt qua 1. Chế độ 16:8 cũng còn tranh cãi: gom 30 thử nghiệm, 1341 người, với điều kiện hai nhóm ăn lượng bằng nhau thì nhóm ăn trong khung giờ giảm nhiều hơn 1.46 kg (−2.65 đến −0.26), cái giá là mất thêm 0.41 kg khối lượng cơ nạc. Nhưng phần lớn các thử nghiệm này ngắn, thử nghiệm 12 tháng dài nhất và nghiêm ngặt nhất lại nói không có khác biệt. Nhịn lâu còn liên quan đến túi mật. Trong một khảo sát cắt ngang 6.547 người, bữa đầu tiên trong ngày muộn mỗi giờ thì tỷ lệ phát hiện sỏi mật cao hơn khoảng 5% (OR 1.05, 1.02–1.08); người đẩy bữa đầu tiên ra 9 giờ đến 14 giờ cao hơn khoảng 49% so với người ăn sớm hơn (OR 1.49, 1.24–1.77). Cơ chế là mật nằm trong túi mật càng lâu càng dễ kết tinh thành sỏi. Nhưng đây là khảo sát tại một thời điểm, cũng có thể là do đã có vấn đề túi mật rồi mới đổi thói quen ăn. Ở Trung Quốc, cohort Kailuan theo dõi 5.6 năm, 369 ca ung thư tiêu hoá: người không ăn sáng có nguy cơ ung thư trực tràng và ruột già gấp 2.32 lần (1.34–4.01), ung thư túi mật và đường mật ngoài gan gấp 5.43 lần (1.34–21.93). Khoảng tin cậy của con số sau rộng tới 21.93, nghĩa là số ca quá ít, chỉ coi là manh mối. Xem tổng thể thì xét kiểm soát cân nặng, khung giờ ăn có ảnh hưởng rất nhỏ, đừng bỏ sức lực vì nó. Còn việc lâu ngày hoàn toàn không ăn sáng có tốt không thì bằng chứng hiện có toàn là tương quan, không chứng minh được quan hệ nhân quả, nhưng chiều hướng nhất quán nghiêng về bất lợi, muốn chắc chắn thì ăn bình thường, chỉ đừng trông chờ nó giúp bạn giảm cân. Nguyên nhân chính của viêm dạ dày và loét dạ dày không phải bụng đói mà là vi khuẩn H. pylori và dùng thuốc giảm đau lâu dài, mục cần xét xem mục 23 ở phần 1 (xét vi khuẩn H. pylori, dương tính thì diệt trùng). Khám ra sỏi mật mà chưa từng đau thì làm gì, xem mục 20 của phần này (khám ra sỏi mật mà chưa từng đau thì đừng cắt túi mật phòng bệnh). Muốn kiểm soát cân nặng, thứ thật sự có bằng chứng là ăn gì và ăn bao nhiêu, xem mục 27 ở phần 2 (mỗi ngày đủ 5 phần rau quả), mục 28 ở phần 2 (ăn ít thực phẩm chế biến sẵn) và mục 32 ở phần 2 (giữ BMI ở 20–25). Người bị đái tháo đường, người đang uống thuốc hạ đường huyết, người đang mang thai không áp dụng được, bỏ một bữa có thể hạ đường huyết, làm theo lời bác sĩ. Nhịn ăn cực đoan và nôn sau ăn là chuyện khác, xem mục 1 ở phần 28 (đừng dùng ăn uống cực đoan, nhịn ăn hay nôn sau ăn để kiểm soát cân nặng). ### 27. Đừng tin 「thủ dâm tổn hại sức khoẻ」「nhịn thái dâm mới được cải thiện」, điều cần xem thật sự là có làm lỡ giấc ngủ, công việc và quan hệ hay không diff --git "a/vi/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" "b/vi/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" index 4a8be53d..e59568d3 100644 --- "a/vi/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" +++ "b/vi/book/07-\346\262\241\351\222\261\347\232\204\346\227\266\345\200\231\346\200\216\344\271\210\346\264\273.md" @@ -4,24 +4,22 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được miễn bao nhiêu, được dịch vụ gì, phải đáp ứng điều kiện gì, tất cả đều ghi ở đây, không quy đổi thành tuổi thọ. Chính sách đổi mỗi năm, mỗi nơi mỗi khác, nên số tiền và ngưỡng đều theo những gì địa phương của bạn công bố trong năm đó. Các mục được sắp như sau: mục không tốn tiền, thủ tục ít, có tác dụng nhanh thì đứng trước. Mục phải tích lũy điều kiện lâu dài, mỗi nơi khác nhau nhiều thì đứng sau. Mục chỉ dựa vào kinh nghiệm, không có văn bản chính thức thì để cuối cùng. -### 1. Mất việc thì xin trợ cấp bảo hiểm thất nghiệp trực tuyến trước +### 1. Mất việc thì xin trợ cấp thất nghiệp trực tuyến trước -- Chi phí: Không tốn tiền. Nộp đơn bằng điện thoại được, cả ba cửa đều được: ứng dụng「掌上 12333」, tiểu trình thẻ bảo hiểm xã hội điện tử trong WeChat hoặc Alipay, và nền tảng dịch vụ bảo hiểm xã hội công cộng quốc gia. Có thẻ bảo hiểm xã hội hoặc căn cước trong tay là làm được, phần lớn các nơi không phải chạy tận nơi. Điều kiện trước là đã làm xong đăng ký thất nghiệp. -- Nói thẳng: Đơn vị và bản thân bạn đóng đủ 1 năm, việc mất việc không phải do bạn chủ động xin thôi, làm xong đăng ký thất nghiệp thì nhận được tiền. Đóng đủ 1 năm đến dưới 5 năm, nhận nhiều nhất 12 tháng. Đủ 5 năm đến dưới 10 năm, nhiều nhất 18 tháng. Từ 10 năm trở lên, nhiều nhất 24 tháng. Trong thời gian nhận tiền, bảo hiểm thất nghiệp đóng thay phần bảo hiểm y tế người lao động cho bạn. -- Lợi ích: Phải đáp ứng đồng thời ba điều kiện. Một là đơn vị và bản thân bạn đã đóng bảo hiểm thất nghiệp tổng cộng đủ 1 năm. Hai là việc mất việc không phải do bạn chủ động xin thôi (pháp luật gọi là「非因本人意愿中断就业」). Ba là đã làm đăng ký thất nghiệp và có ý định tìm việc. Thời gian được nhận tính theo số năm đã đóng. Tổng cộng đủ 1 năm đến dưới 5 năm, dài nhất 12 tháng. Đủ 5 năm đến dưới 10 năm, dài nhất 18 tháng. Từ 10 năm trở lên, dài nhất 24 tháng. Số tiền mỗi tháng do chính quyền tỉnh quy định, mỗi tỉnh khác nhau. Số tiền thấp hơn lương tối thiểu địa phương, cao hơn trợ cấp người có thu nhập thấp ở thành phố. Mục tiêu của nhà nước là dần đưa lên tới 90% mức lương tối thiểu. Trong thời gian nhận tiền trợ cấp thất nghiệp, bạn vẫn tham gia bảo hiểm y tế người lao động và được hưởng quyền lợi bảo hiểm y tế cơ bản. Khoản phí bảo hiểm y tế cơ bản phải đóng được lấy từ quỹ bảo hiểm thất nghiệp, bạn không phải đóng (toàn quốc, quy định áp dụng từ 1999) +- Chi phí: Không tốn tiền. Có điện thoại thì nộp đơn được ngay, có ba cửa: ứng dụng「掌上 12333」, tiện ích thẻ bảo hiểm xã hội điện tử trong WeChat hoặc Alipay, Cổng dịch vụ công bảo hiểm xã hội quốc gia. Có thẻ bảo hiểm xã hội hoặc giấy tờ tùy thân là làm được, phần lớn địa phương không phải đến nơi. Điều kiện là đã đăng ký thất nghiệp. +- Nói thẳng: Đơn vị và bạn đóng đủ 1 năm, lại không phải do bạn chủ động xin nghỉ, làm xong đăng ký thất nghiệp thì nhận được tiền. Đóng đủ 1 năm đến dưới 5 năm, nhận tối đa 12 tháng. Từ 5 năm đến dưới 10 năm, tối đa 18 tháng. Từ 10 năm trở lên, tối đa 24 tháng. Trong lúc nhận tiền, bảo hiểm y tế người lao động do bảo hiểm thất nghiệp đóng thay. +- Lợi ích: Phải thoả đồng thời ba điều kiện. Một là đơn vị và bản thân bạn đóng bảo hiểm thất nghiệp cộng dồn đủ 1 năm. Hai là không phải do bạn chủ động xin nghỉ (pháp luật gọi là「không phải do ý chí của chính người mà mất việc」). Ba là đã làm đăng ký thất nghiệp và có ý định tìm việc. Thời gian nhận tính theo số năm đã đóng. Cộng dồn đủ 1 năm đến dưới 5 năm, dài nhất 12 tháng. Từ 5 năm đến dưới 10 năm, dài nhất 18 tháng. Từ 10 năm trở lên, dài nhất 24 tháng. Mỗi tháng trả bao nhiêu do chính quyền cấp tỉnh quyết, mỗi tỉnh một khác. Số tiền thấp hơn mức lương tối thiểu địa phương, cao hơn trợ cấp người có thu nhập thấp ở thành phố. Mục tiêu của nhà nước là dần nâng lên 90% mức lương tối thiểu. Trong thời gian nhận trợ cấp thất nghiệp, bạn vẫn tham gia bảo hiểm y tế người lao động và hưởng quyền lợi bảo hiểm y tế cơ bản. Phần bảo hiểm y tế cơ bản phải đóng do quỹ bảo hiểm thất nghiệp chi trả, bạn không phải đóng (toàn quốc, quy định có hiệu lực từ 1999) - Mức bằng chứng: A -- Nguồn: Quốc vụ viện (国务院) (1999). Quy định về bảo hiểm thất nghiệp（失业保险条例）（国务院令第 258 号，第十四、十七、十八条）. ；Bộ Nhân sự và Xã hội (人力资源社会保障部). Cẩm nang thủ tục xin trợ cấp bảo hiểm thất nghiệp trên Nền tảng dịch vụ chính phủ nhân sự và bảo hiểm xã hội toàn quốc（全国人社政务服务平台·失业保险待遇申领办事指南）. ；Cục Việc làm thuộc Ủy ban Cải cách và Phát triển quốc gia (国家发展改革委就业司) (2022). Mức trợ cấp thất nghiệp sẽ dần nâng lên bằng 90% lương tối thiểu（失业保险金标准，将逐步提高至最低工资的 90%）. ；Sở Nhân sự và Xã hội tỉnh Liêu Ninh (辽宁省人力资源和社会保障厅) (2019). Thông báo về việc điều chỉnh tiếp mức trợ cấp thất nghiệp（关于进一步调整失业保险金标准的通知）：「为落实《人力资源社会保障部财政部关于调整失业保险金标准的指导意见》(人社部发〔2017〕71号〕"逐步将失业保险金标准提高到最低工资标准的90%"等要求」. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2010). Luật Bảo hiểm xã hội（社会保险法）（第四十八条）. （Kho luật và chính sách Cục Tính thuế quốc gia） -- Ghi chú: Ai tự chủ động nghỉ việc thì không tính là「非因本人意愿」, không nhận được. Trong thời gian nhận trợ cấp, bảo hiểm y tế người lao động do bảo hiểm thất nghiệp đóng thay. Số tiền mỗi tỉnh chênh lệch rất nhiều, theo những gì nơi bạn tham gia bảo hiểm công bố. Cụm「最低工资 90%」lấy từ ý kiến chỉ đạo năm 2017 của Bộ Nhân sự và Xã hội cùng Bộ Tài chính（人社部发〔2017〕71 号）, nguyên văn là「逐步将失业保险金标准提高到最低工资标准的90%」. Đó là mục tiêu, không phải hiện nay đã trả theo 90%. Trang gốc trên cổng Bộ Nhân sự và Xã hội không mở được trên máy này, nên câu đó được đối chiếu theo trích dẫn trong văn bản của Sở Nhân sự và Xã hội tỉnh Liêu Ninh. - -### 2. Bị nợ lương thì khiếu nại với thanh tra lao động trước, rồi mới xin hoà giải lao động, cả hai đường đều không tốn tiền, phần lớn vụ có kết quả trong vài tháng +- Nguồn: Hội đồng Nhà nước (1999). Quy định về bảo hiểm thất nghiệp, 失业保险条例（国务院令第 258 号，第十四、十六、十七、十八条；第十六条：「失业保险金自办理失业登记之日起计算」）. ；Bộ Nhân sự và Xã hội. Cổng dịch vụ công nhân sự và xã hội toàn quốc·Hướng dẫn thủ tục xin trợ cấp thất nghiệp. ；Cục Việc làm, Ủy ban Phát triển và Cải cách Quốc gia (2022). 失业保险金标准，将逐步提高至最低工资的 90% — mức trợ cấp thất nghiệp sẽ dần nâng lên 90% mức lương tối thiểu. ；Sở Nhân sự và Xã hội tỉnh Liao Ning (2019). Thông báo điều chỉnh thêm mức trợ cấp thất nghiệp:「为落实《人力资源社会保障部财政部关于调整失业保险金标准的指导意见》(人社部发〔2017〕71号〕"逐步将失业保险金标准提高到最低工资标准的90%"等要求」. ；Ủy ban Thường vụ Quốc hội (2010). 社会保险法（第四十八条）. （Kho chính sách và pháp luật Cục Thuế nhà nước） +- Ghi chú: Bạn chủ động xin nghỉ thì không tính là「không phải do ý chí của chính người」, không nhận được. Nên khi nghỉ việc đừng ký vào chỗ「nguyên nhân cá nhân chủ động nghỉ」, xem mục 7 ở phần 19. Trợ cấp thất nghiệp tính từ ngày làm đăng ký thất nghiệp, đăng ký muộn một tháng thì mất một tháng tiền. Sau khi rời công ty thì việc đầu tiên là đi đăng ký. Đăng ký cần có giấy xác nhận nghỉ việc do công ty cấp, xem mục 17 ở phần 19 (giấy xác nhận nghỉ việc). Có phải giữ thỏa thuận cạnh tranh, công ty không trả bồi thường thì xử lý thế nào, xem mục 12 ở phần 11 (đã ký thỏa thuận cạnh tranh). Trong thời gian nhận trợ cấp, bảo hiểm y tế người lao động do bảo hiểm thất nghiệp đóng thay. Mức tiền ở mỗi tỉnh khác nhau nhiều, theo nơi bạn tham gia bảo hiểm mà nơi đó công bố. 「Lương tối thiểu 90%」 lấy từ chỉ thị năm 2017 của Bộ Nhân sự và Xã hội cùng Bộ Tài chính (số 71 năm 2017), nguyên văn là「逐步将失业保险金标准提高到最低工资标准的90%」. Đó là mục tiêu, không phải hiện nay đã trả 90%. Trang gốc trên cổng Bộ Nhân sự và Xã hội không mở được trên máy này, câu này đối chiếu theo trích dẫn trong văn bản của Sở Nhân sự và Xã hội tỉnh Liao Ning. +### 2. Bị nợ lương thì khiếu nại giám sát lao động trước, không được thì xin hoà giải lao động, cả hai đường đều miễn phí -- Chi phí: Không tốn tiền, khiếu nại và hoà giải đều không thu phí. Cần chuẩn bị hợp đồng lao động, bảng lương, sổ chấm công, lịch sử chuyển khoản và tin nhắn, chạy tới cục nhân sự xã hội hoặc ủy ban hoà giải 1 đến 3 chuyến. Xin hoà giải có thời hạn 1 năm (pháp luật gọi là thời hiệu khởi kiện). Người còn làm ở đơn vị đó mà bị nợ lương thì không bị giới hạn 1 năm này. Người đã nghỉ việc thì phải nộp trong vòng 1 năm kể từ ngày nghỉ. Khó ở chỗ phải chạy đi chạy lại, phải tìm từng mẩu chứng cứ, dễ nản giữa chừng. -- Nói thẳng: Cả hai đường đều không mất tiền. Gọi 12333 hoặc khiếu nại với bộ phận thanh tra lao động ở cục nhân sự xã hội, sau khi khởi kiện phải xác minh xong trong 60 ngày làm việc, có thể buộc công ty trả tiền trong thời hạn, quá hạn không trả thì còn phải bồi thêm 50% đến 100%. Không được thì xin hoà giải lao động, kết thúc trong 45 ngày. Nhưng「kết thúc」không có nghĩa là「tiền về tài khoản」: công ty không còn tài sản, chủ doanh nghiệp bỏ trốn, thắng kiện cũng có thể không được tiền. -- Lợi ích: Bước đầu là khiếu nại. Gọi 12333, hoặc khiếu nại với bộ phận thanh tra lao động ở cục nhân sự xã hội địa phương. Pháp luật quy định xác minh xong trong 60 ngày làm việc kể từ ngày khởi kiện, vụ phức tạp thì được kéo dài thêm 30 ngày làm việc. Xác minh đúng thì có thể buộc công ty trả tiền trong thời hạn. Quá hạn vẫn không trả thì phải trả thêm một khoản bồi thường bằng 50% đến 100% số tiền phải trả. Năm 2024 các cơ quan thanh tra trên toàn quốc xử lý 13.7 vạn vụ, thu hồi 86.5 tỷ tệ tiền lương cho 65.4 vạn người. Trong tay có giấy nợ hoặc bảng lương, sự việc nợ lương rõ ràng, còn có thể xin tòa cấp cấp xét cấp lệnh chi trả. Lệnh chi trả là chỉ thị bắt công ty trả tiền do tòa ban, tòa phải ban trong 15 ngày kể từ ngày nhận đơn. Bước thứ hai là xin hoà giải. Pháp luật quy định kết thúc trong 45 ngày kể từ ngày nhận đơn, có thể kéo dài thêm 15 ngày. Người đòi lương có thể xin thi hành trước: không chờ xét xong toàn bộ, thi hành trước một phần. Số tiền nợ không vượt quá 12 lần lương tối thiểu theo tháng của địa phương thì thuộc loại một trọng một quyết: phán quyết của trọng tài là kết quả cuối, công ty không được khởi kiện. Năm 2024 toàn quốc tiếp nhận 425.7 vạn vụ, giải quyết xong 415.6 vạn vụ. Tổng tiền giải quyết 934.7 tỷ tệ. Tỷ lệ thành công hòa giải 79.6%, tỷ lệ kết thúc ở trọng tài 98.2%, tỷ lệ phán quyết không đi tiếp 73.7%. Người ta nói việc này phải kéo dài「nửa năm một năm」, nói đến phần vụ sau khi qua trọng tài lại kiện ra tòa: xét sơ thẩm 6 tháng (loại đơn giản là 3 tháng), cộng thêm xét phúc âm 3 tháng. Năm 2024 tòa xét xong 8.8 vạn vụ đòi tiền công lao động, thu hồi 285.4 tỷ tệ, 998 người bị kết án về tội không chịu trả tiền công lao động (toàn quốc) +- Chi phí: Không tốn tiền, khiếu nại và hoà giải đều không thu phí. Cần chuẩn bị hợp đồng lao động, bảng lương, sổ chấm công, lịch sử chuyển khoản và tin nhắn, phải chạy tới Sở Nhân sự và Xã hội hoặc hội đồng hoà giải 1 đến 3 chuyến. Xin hoà giải có thời hạn 1 năm (pháp luật gọi là thời hiệu hoà giải). Người còn làm ở đơn vị đó mà bị nợ lương thì không chịu giới hạn 1 năm này. Người đã nghỉ việc thì phải nộp trong vòng 1 năm kể từ ngày nghỉ. Khó ở chỗ phải chạy nhiều chuyến, sưu tầm từng chứng từ, dễ nản trong chừng. +- Nói thẳng: Cả hai đường đều không mất tiền. Gọi 12333 hoặc khiếu nại tại bộ phận giám sát lao động của Sở Nhân sự và Xã hội, sau khi khởi kiện có 60 ngày làm việc để kết luận, có thể yêu cầu công ty trả tiền trong thời hạn, quá hạn không trả thì còn phải bồi thường thêm 50% đến 100%. Không được thì xin hoà giải lao động, 45 ngày để kết thúc. Nhưng「kết thúc」không phải là「tiền về tay」: công ty không còn tài sản, ông chủ bỏ đi, thắng kiện cũng có thể không được tiền. +- Lợi ích: Bước một là khiếu nại. Gọi 12333, hoặc khiếu nại tại bộ phận giám sát lao động của Sở Nhân sự và Xã hội địa phương. Pháp luật quy định sau khi khởi kiện có 60 ngày làm việc để kết luận, vụ phức tạp có thể kéo dài thêm 30 ngày làm việc. Xác nhận đúng thì có thể yêu cầu công ty trả đúng hạn. Quá hạn không trả thì phải trả thêm một khoản bồi thường bằng 50% đến 100% số tiền phải trả. Năm 2024, cơ quan giám sát toàn quốc xử lý 137.000 vụ, thu hồi 8.650.000.000 tệ tiền lương cho 654.000 người. Trong tay có giấy nợ hoặc bảng lương, sự việc nợ lương rõ ràng, còn có thể trực tiếp xin tòa cấp cơ sở ra lệnh thanh toán. Lệnh thanh toán là chỉ thị bắt công ty trả tiền do tòa ban, tòa phải ra trong 15 ngày kể từ khi nhận. Bước hai là xin hoà giải. Pháp luật quy định sau khi tiếp nhận có 45 ngày để kết thúc, có thể kéo dài thêm 15 ngày. Người đòi lương có thể xin thi hành trước: không chờ xét xử hết vụ, thi hành một phần trước. Số tiền nợ không quá 12 lần lương tối thiểu tháng địa phương thì thuộc loại một lần phân xử là chung cực: phán quyết của hội đồng hoà giải là kết quả cuối, công ty không được khởi kiện. Năm 2024, toàn quốc tiếp nhận 4.257.000 vụ, giải quyết xong 4.156.000 vụ. Tổng giá trị các vụ kết thúc là 93.470.000.000 tệ. Tỷ lệ hoà giải thành công 79.6%, tỷ lệ kết thúc bằng phán quyết 98.2%, tỷ lệ kết thúc chung cực 73.7%. Có người nói việc này kéo dài「nửa năm một năm」, là nói phần vụ sau hoà giải lại kiện ra tòa: xét sơ thẩm 6 tháng (3 tháng nếu theo thủ tục đơn giản), cộng phúc án 3 tháng. Năm 2024, tòa xét xong 88.000 vụ đòi tiền lương, thu hồi 28.540.000.000 tệ, 998 người bị kết án tội không trả tiền lương (toàn quốc) - Mức bằng chứng: A -- Nguồn: Bộ Nhân sự và Xã hội (人力资源社会保障部) (2025). Báo cáo thống kê công trình nhân sự và bảo hiểm xã hội năm 2024. ；Tòa án nhân dân tối cao (最高人民法院) (2025). Báo cáo công tác của Tòa án nhân dân tối cao. ；Quốc vụ viện (国务院) (2004). Quy định về thanh tra bảo đảm lao động（劳动保障监察条例）（国务院令第 423 号，第九、十七、二十六条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007). Luật Hoà giải và trọng tài lao động（劳动争议调解仲裁法）（主席令第八十号，第二十七、四十三、四十四、四十七、五十、五十三条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2012, sửa đổi). Luật Hợp đồng lao động（劳动合同法）（第三十、八十五条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2023, sửa đổi). Luật Tố tụng dân sự（民事诉讼法）（第一百零九、一百五十二、一百六十四、一百八十三、二百二十五、二百二十七条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2011). Bản sửa đổi (tám) Bộ luật Hình sự（刑法修正案（八））第四十一条, tức là 刑法第二百七十六条之一 của Bộ luật Hình sự. ；Tòa án nhân dân tối cao (最高人民法院) (2013). Giải thích về một số vấn đề pháp lý khi xét các vụ án hình sự về việc không chịu trả tiền công lao động（法释〔2013〕3 号，第三条）. ；Quốc vụ viện (国务院) (2019). Quy định bảo đảm việc trả lương cho công nhân nông nghiệp（保障农民工工资支付条例）（国务院令第 724 号，第十、二十六、三十、三十一、三十二、四十一条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2021). Luật Trợ giúp pháp lý（法律援助法）（第三十一、四十二条）. -- Ghi chú: Khiếu nại và hoà giải đều có ích, nhưng phải phân biệt「kết thúc」với「tiền về tài khoản」. Tỷ lệ kết thúc 98.2% chỉ nói thủ tục đã đi hết. Công ty không có tài sản, chủ doanh nghiệp bỏ trốn thì thắng kiện vẫn không được tiền. Tòa án nhân dân tối cao chỉ công bố tổng số tiền đã thi hành được (2.3 nghìn tỷ tệ năm 2024) và nói「执行完毕率保持高位」, còn cuối cùng bao nhiêu tiền nợ lương thật sự vào túi người bị thiếu thì không có số liệu chính thức. Cụm「仲裁终结率」trong báo cáo không được định nghĩa, mục này hiểu là「sau khi có phán quyết thì không còn đi kiện ra tòa」. Nợ lương đến mức nào thì đủ đi khởi kiện hình sự. Một là nợ một người từ 3 tháng trở lên, số tiền từ 5000 đến 2 vạn tệ trở lên. Trường hợp khác là nợ từ 10 người trở lên, tổng cộng từ 3 vạn đến 10 vạn tệ trở lên. Với hai trường hợp này, đã bị buộc trả mà vẫn không trả thì đủ đi khởi kiện. Tòa án cấp cao và viện kiểm sát của từng tỉnh ấn định và công bố con số cụ thể trong khoảng đó, nên trước khi khiếu nại nên tra con số của tỉnh mình, và nêu ra tận mặt khi khiếu nại. Công nhân nông nghiệp tại công trình còn có quy định riêng về tài khoản lương riêng và việc nhà thầu chính phải trả trước, nên tìm nhà thầu chính còn hơn tìm đội trưởng. Công nhân vào thành phố làm việc xin trợ giúp pháp lý để đòi lương thì không bị xét khó khăn tài chính. Nguyên văn Luật Hợp đồng lao động và Luật Tố tụng dân sự được trích từ trang sao chép của Cục Tính thuế và Cục Quản lý thị trường tỉnh Quý Châu (đều là gov.cn), vì trang của Đại biểu Quốc hội Trung Quốc không mở được. Bản sửa đổi (tám) Bộ luật Hình sự không tải được trên trang Đại biểu Quốc hội Trung Quốc, nên đã dùng curl mở ra để đối chiếu. - +- Nguồn: Bộ Nhân sự và Xã hội (2025). 2024 年度人力资源和社会保障事业发展统计公报 — Báo cáo thống kê phát triển nhân sự và xã hội năm 2024. ；Tòa án nhân dân tối cao (2025). 最高人民法院工作报告 — Báo cáo công tác của Tòa án nhân dân tối cao. ；Hội đồng Nhà nước (2004). 劳动保障监察条例（国务院令第 423 号，第九、十七、二十六条）. ；Ủy ban Thường vụ Quốc hội (2007). 劳动争议调解仲裁法（主席令第八十号，第二十七、四十三、四十四、四十七、五十、五十三条）. ；Ủy ban Thường vụ Quốc hội (2012 修正). 劳动合同法（第三十、八十五条）. ；Ủy ban Thường vụ Quốc hội (2023 修正). 民事诉讼法（第一百零九、一百五十二、一百六十四、一百八十三、二百二十五、二百二十七条）. ；Ủy ban Thường vụ Quốc hội (2011). 刑法修正案（八）第四十一条，即刑法第二百七十六条之一. ；Tòa án nhân dân tối cao (2013). 关于审理拒不支付劳动报酬刑事案件适用法律若干问题的解释（法释〔2013〕3 号，第三条）. ；Hội đồng Nhà nước (2019). 保障农民工工资支付条例（国务院令第 724 号，第十、二十六、三十、三十一、三十二、四十一条）. ；Ủy ban Thường vụ Quốc hội (2021). 法律援助法（第三十一、四十二条）. +- Ghi chú: Khiếu nại và hoà giải có ích, nhưng phải phân biệt「kết thúc」và「tiền về tay」. Tỷ lệ kết thúc 98.2% chỉ nói thủ tục đã đi hết. Công ty không còn tài sản, ông chủ bỏ đi, thắng kiện vẫn không được tiền. Tòa án nhân dân tối cao chỉ công bố tổng số tiền thi hành được (năm 2024 là 2,3 nghìn tỷ tệ) và nói「tỷ lệ thi hành xong ở mức cao」, không có số liệu chính thức về bao nhiêu phần tiền lương thực sự vào túi người bị nợ. Cụm「tỷ lệ kết thúc chung cực」báo cáo không định nghĩa, mục này hiểu là「sau khi có phán quyết thì không còn kiện ra tòa」. Nợ lương đến mức nào đó là đủ để khởi kiện hình sự. Một là nợ một người trên 3 tháng, số tiền từ 5.000 đến trên 2 vạn tệ. Hai là nợ trên 10 người, cộng dồn từ 3 vạn đến trên 10 vạn tệ. Trong hai trường hợp đó, sau khi bị yêu cầu trả mà vẫn không trả thì đủ để khởi kiện. Tòa án nhân dân tỉnh và viện kiểm sát tỉnh ấn định con số cụ thể trong khoảng đó rồi công bố, nên tra trước khi khiếu nại và nêu ra khi đến nơi. Công nhân lao động trên công trường còn có quy định tài khoản lương riêng và nhà thầu chính phải trả trước, tìm nhà thầu chính có tác dụng hơn tìm đối tác thầu. Hai đường này đều cần có quan hệ lao động. Xưởng và cửa hàng chưa đăng ký kinh doanh vẫn tính là quan hệ lao động, vẫn khiếu nại và hoà giải được. Làm việc riêng cho gia đình hoặc cá nhân, ví dụ làm giúp việc nhà, thì không phải quan hệ lao động, phải ra tòa, xem mục 22 của phần này (xem trước người thuê bạn là ai). Người lao động vào thành phố xin trợ giúp pháp lý để đòi lương thì không bị xem xét tình trạng khó khăn tài chính. Nguyên văn Luật hợp đồng lao động và Luật tố tụng dân sự lấy từ trang Cục Thuế nhà nước và Cục Quản lý thị trường tỉnh Quý Châu đăng lại (đều trên gov.cn), vì trang của Quốc hội nhân dân Trung Quốc không mở được. Bản sửa đổi (tám) của Bộ luật Hình sự không lấy được từ trang Quốc hội, đã mở bằng curl để đối chiếu. ### 3. Không đủ tiền kiện tụng thì xin trợ giúp pháp lý, các vụ đòi lương, cấp dưỡng, tai nạn lao động vốn đã nằm trong phạm vi - Chi phí: Không tốn tiền. Trợ giúp pháp lý không thu một đồng nào của người được trợ giúp. @@ -30,7 +28,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2003). Quy định về trợ giúp pháp lý（法律援助条例）（国令第 385 号，第十条、第十一条、第十二条）. ；Lãnh đạo Bộ Tư pháp (司法部有关负责人) (2021). Trả lời phóng viên về việc thực thi Luật Trợ giúp pháp lý. - Ghi chú: Luật Trợ giúp pháp lý có hiệu lực từ ngày 1 tháng 1 năm 2022, có một thay đổi. Trước đây phải「提交经济困难证明」. Nay đổi thành bản thân người khai đúng tình hình tài chính của mình là được, không phải về quê mở giấy nữa. Cửa vào là trung tâm trợ giúp pháp lý và trung tâm dịch vụ pháp lý công cộng ở địa phương, cũng có thể gọi 12348. - ### 4. Không còn đường lui thì đến trạm cứu hộ, ở ăn ở và có vé về quê - Chi phí: Không tốn tiền. Tự đến trạm quản lý cứu hộ địa phương xin giúp, cũng có thể nói với công an hoặc bộ phận dân vụ để họ đưa bạn đi. Khi vào phải khai trung thực họ tên và đồ đạc mang theo. Khó ở chỗ phải nói với người khác rằng mình không có tiền, không có chỗ đi. @@ -39,6 +36,7 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2003). Biện pháp cứu hộ và quản lý người lang thang xin ăn không nơi nương tựa ở thành phố（城市生活无着的流浪乞讨人员救助管理办法）（国务院令第 381 号，第五、六、七条）. ；Bộ Dân vụ (民政部) (2003). Quy tắc chi tiết về thi hành Biện pháp cứu hộ và quản lý người lang thang xin ăn không nơi nương tựa ở thành phố（城市生活无着的流浪乞讨人员救助管理办法实施细则）（民政部令第 24 号，第十一、十二条）. - Ghi chú: Xin giúp và rời đi đều tự nguyện, không phải trại tạm giam. Trạm cho ăn ở cơ bản, không phát tiền mặt. Chứng từ đi xe cấp theo nơi đăng ký hộ khẩu hoặc nơi cư trú, trước khi xin giúp hãy nghĩ kỹ mình đi đâu. + ### 5. Tìm việc thì dùng dịch vụ việc làm công cộng miễn phí và thị trường việc lẻ trước, đừng qua môi giới thu phí - Chi phí: Không tốn tiền. Đi đăng ký ở cơ quan dịch vụ việc làm công cộng cấp huyện trở lên, ví dụ trung tâm dịch vụ việc làm, thị trường lao động. Thị trường việc lẻ địa phương cũng đăng ký được. Không biết địa chỉ thì gọi 12333 hỏi. @@ -47,7 +45,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007, sửa đổi 2015). Luật thúc đẩy việc làm（第三十五条）. ；Bộ Nhân sự và Xã hội và bốn cơ quan khác (人力资源社会保障部等五部门) (2022). Ý kiến về việc tăng cường xây dựng thị trường việc lẻ, hoàn thiện dịch vụ tìm việc và tuyển dụng（人社部发〔2022〕38 号）. - Ghi chú: Đăng ký thất nghiệp là thủ tục trước khi lĩnh bảo hiểm thất nghiệp và xác nhận người khó khăn việc làm, làm luôn cho xong. Làm việc lẻ trả theo ngày thì đối chiếu tiền công ngay tại chỗ, giữ lại cách liên hệ của đối phương và bản ghi công việc. - ### 6. Gặp biến cố đột ngột thì xin trợ cấp tạm thời trước - Chi phí: Không tốn tiền. Nộp đơn lên chính quyền xã, thị trấn hoặc ủy ban quản lý khu phố, nói rõ ở nhà đã xảy ra chuyện gì. Ví dụ hỏa hoạn, tai nạn giao thông, người nhà đột ngột mắc bệnh nặng, hoặc chi tiêu bắt buộc hằng ngày đột ngột tăng lên. Sau khi thẩm định và niêm yết công khai, phòng dân sự cấp huyện phê duyệt. Chuẩn bị giấy tờ theo danh sách địa phương đưa, thường phải đi 1 đến 2 lần. @@ -56,15 +53,14 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: B - Nguồn: Quốc vụ viện (国务院) (2014, sửa đổi 2019). Biện pháp tạm thời về cứu trợ xã hội（国务院令第 649 号，第四十七、四十八、四十九条）. - Ghi chú: Không yêu cầu bạn đã là hộ nghèo. Thường phát một lần, số tiền các nơi chênh lệch nhiều, lấy số địa phương công bố trong năm đó làm chuẩn. - -### 7. Thu nhập dưới mức trợ cấp người có thu nhập thấp địa phương thì đi xin trợ cấp +### 7. Thu nhập thấp hơn mức trợ cấp người có thu nhập thấp ở địa phương thì xin trợ cấp người có thu nhập thấp -- Chi phí: Không tốn tiền. Các thành viên gia đình sống chung nộp đơn bằng văn bản lên chính quyền xã, thị trấn hoặc ủy ban khu phố nơi đăng ký hộ khẩu. Tự làm bất tiện thì có thể ủy quyền cho ủy ban dân cư hoặc ủy ban khu phố làm thay. Phải chịu kiểm tra thu nhập và tài sản của gia đình. Thường đi 2 đến 3 lần, mang căn cước, sổ hộ khẩu và chứng minh thu nhập, tài sản. Khó ở chỗ phải bày hết của cải ra cho người kiểm tra, còn phải chờ niêm yết công khai. -- Nói thẳng: Tổng thu nhập cả nhà chia cho số người, dưới mức trợ cấp địa phương, tài sản trong nhà cũng đúng quy định, thì xin được. Cuối 2024, cả nước mức trợ cấp thành phố bình quân 798.1 tệ mỗi người mỗi tháng, nông thôn 593.9 tệ. Phải về nơi đăng ký hộ khẩu để xin. Có nhà, có xe, có tiền tiết kiệm, hoặc con cái có khả năng nuôi bạn, có thể không được duyệt. -- Lợi ích: Tổng thu nhập cả nhà chia cho số người, dưới mức trợ cấp địa phương, tài sản trong nhà cũng nằm trong quy định, thì nhận được trợ cấp. Cuối 2024, cả nước mức trợ cấp thành phố bình quân 798.1 tệ một người một tháng, nông thôn 593.9 tệ một người một tháng. Thành phố trợ cấp 625.0 vạn người, nông thôn trợ cấp 3361.5 vạn người. Mức trợ cấp tính như sau: lấy chi tiêu bình quân đầu người của cư dân địa phương trong năm trước, nhân với một tỷ lệ định lượng, sau một khoảng thời gian điều chỉnh lại một lần. Nói trắng ra là lấy số tiền bình quân mỗi người ở địa phương tiêu trong năm trước, trừ đi một phần rồi ra (bình quân cả nước, mỗi nơi một khác) +- Chi phí: Không tốn tiền. Một thành viên gia đình sống chung nộp một đơn viết ra nơi chính quyền xã, thị trấn hoặc văn phòng đại diện khu phố. Tự làm không tiện thì có thể nhờ ủy ban thôn hoặc ủy ban tổ dân cư làm hộ. Phải chấp nhận bị kiểm tra thu nhập và tài sản của cả nhà. Thường phải chạy 2 đến 3 chuyến, mang theo giấy tờ tùy thân, sổ hộ khẩu và giấy chứng minh thu nhập, tài sản. Khó ở chỗ phải bày ra hết tài sản của nhà cho người ta xem, rồi còn phải chờ công khai. +- Nói thẳng: Tổng thu nhập của cả nhà chia cho số người, thấp hơn mức trợ cấp người có thu nhập thấp ở địa phương, mà tài sản trong nhà cũng hợp quy định thì có thể xin. Cuối năm 2024, bình quân toàn quốc mức trợ cấp người có thu nhập thấp ở thành phố là 798.1 tệ mỗi người mỗi tháng, ở nông thôn là 593.9 tệ. Được duyệt thì trả phần chênh lệch giữa mức trợ cấp và thu nhập mỗi người mỗi tháng của nhà. Có nhà, có xe, có tiền để dành, hoặc con cái có khả năng phụng dưỡng bạn thì có thể không được duyệt. +- Lợi ích: Tổng thu nhập của cả nhà chia cho số người, thấp hơn mức trợ cấp người có thu nhập thấp ở địa phương, mà tài sản trong nhà cũng nằm trong phạm vi quy định thì được nhận trợ cấp. Cuối năm 2024, bình quân toàn quốc mức trợ cấp người có thu nhập thấp ở thành phố là 798.1 tệ/người·tháng, ở nông thôn là 593.9 tệ/người·tháng. Thành phố bảo đảm cho 6.250.000 người, nông thôn bảo đảm cho 33.615.000 người. Tiền trợ cấp trả hàng tháng theo phần chênh lệch: mức trợ cấp ở địa phương trừ đi thu nhập bình quân đầu người của nhà. Ví dụ mức ở địa phương là 798.1 tệ, nhà mỗi người mỗi tháng có 500 tệ, thì mỗi người mỗi tháng nhận 298.1 tệ, không phải 798.1 tệ. Mức trợ cấp ấy được đặt như sau: lấy chi tiêu bình quân đầu người cư dân của năm trước ở địa phương, nhân với một tỷ lệ định lượng, rồi định kỳ lại điều chỉnh một lần. Nói thẳng là lấy mức bình quân một người ở địa phương tiêu bao nhiêu tiền trong năm trước, rồi quy ra một hệ số (bình quân toàn quốc, mỗi nơi khác nhau) - Mức bằng chứng: A -- Nguồn: Quốc vụ viện (国务院) (2014, sửa đổi 2019). Biện pháp tạm thời về cứu trợ xã hội（国务院令第 649 号，第九、十、十一条）. ；Cục Thống kê Quốc gia (国家统计局) (2025). Các ngành xã hội phát triển tốt, phúc lợi người dân không ngừng tăng — báo cáo thống kê xã hội dân sinh kể từ「十四五」. ；Bộ Dân vụ và ba cơ quan khác (民政部等四部门) (2024). Ý kiến hướng dẫn về việc xác định và điều chỉnh chuẩn trợ cấp người có thu nhập thấp（民发〔2024〕16 号）. -- Ghi chú: Trợ cấp là tuyến giữ cuối cùng, phải về nơi đăng ký hộ khẩu để xin. Nhà, xe, tiền tiết kiệm trong nhà vượt chuẩn địa phương, hoặc con cái có khả năng nuôi bạn, đều có thể không được duyệt. Duyệt rồi còn có lợi khác: đóng bảo hiểm y tế cư dân được hỗ trợ (bảo hiểm y tế cư dân, xem mục 9), khám bệnh đi được trợ cấp y tế (trợ cấp y tế, xem mục 10), xin trợ cấp pháp lý không kiểm tra khó khăn tài chính (trợ cấp pháp lý, xem mục 3). +- Nguồn: Hội đồng Nhà nước (国务院) (2014, 2019 修订). Quy định lâm thời về trợ giúp xã hội (《社会救助暂行办法》)(国务院令第 649 号，第九、十、十一条；第十二条：「县级人民政府民政部门按照共同生活的家庭成员人均收入低于当地最低生活保障标准的差额，按月发给最低生活保障金」). ；Cục Thống kê quốc gia (国家统计局) (2025). 社会事业向好发展 民生福祉不断增进——「十四五」以来社会民生统计报告. ；Bộ Dân vụ và ba bộ khác (民政部等四部门) (2024). 关于进一步做好最低生活保障标准确定调整工作的指导意见（民发〔2024〕16 号）. +- Ghi chú: Trợ cấp người có thu nhập thấp là lớp bảo đảm cuối cùng, phải về nơi đăng ký hộ khẩu mà xin. Nhà, xe, tiền để dành trong gia đình vượt quá chuẩn địa phương, hoặc con cái có khả năng phụng dưỡng bạn, thì đều có thể không được duyệt. Được duyệt rồi còn có lợi ích khác: mua bảo hiểm y tế cư dân được trợ giúp (bảo hiểm y tế cư dân xem mục 9), khám bệnh thì đi theo trợ cấp y tế (trợ cấp y tế xem mục 10), xin trợ giúp pháp lý thì không bị xét điều kiện khó khăn tài chính (trợ giúp pháp lý xem mục 3). ### 8. Người có giấy chứng nhận người khuyết tật thì đi xin hai khoản trợ cấp cho người khuyết tật @@ -74,7 +70,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2015). Ý kiến về việc thiết lập toàn diện chế độ trợ cấp sinh hoạt cho người khuyết tật khó khăn và trợ cấp chăm sóc cho người khuyết tật nặng（国发〔2015〕52 号）. - Ghi chú: Chuẩn do chính quyền tỉnh đặt, mỗi nơi chênh lệch lớn, mỗi tháng từ vài chục tệ đến vài trăm tệ. Khi xin mang theo giấy chứng nhận người khuyết tật thế hệ hai. Người nào bất tiện thì ủy quyền cho người nhà làm thay - ### 9. Đừng để bảo hiểm y tế cư dân bị đứt quãng mỗi năm 400 tệ, hộ khó khăn được miễn giảm - Chi phí: Phần bạn tự đóng, mỗi người mỗi năm không dưới 400 tệ. Đây là tiêu chuẩn quốc gia năm 2024、2025, mỗi nơi có thể cao hơn mức này. Phải đóng trong đúng khoảng thời gian thu phí tập trung cố định mỗi năm. Người đặc biệt khó khăn được hỗ trợ toàn bộ, khoản này do nhà nước trả hết. Đối tượng trợ cấp người có thu nhập thấp được hỗ trợ định mức, nhà nước trả một phần cố định. @@ -84,14 +79,14 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Nguồn: Cục Quản lý bảo hiểm y tế quốc gia (国家医保局) (2024). Giải thích chính sách về Thông báo liên quan đến công tác bảo hiểm y tế cơ bản cho cư dân đô thị và nông thôn năm 2024（医保发〔2024〕19 号）. ；Cục Bảo hiểm y tế tỉnh Hà Bắc và các cơ quan khác (河北省医疗保障局等) (2025). Thông báo về việc làm tốt công tác bảo hiểm y tế cơ bản cho cư dân đô thị và nông thôn năm 2025（冀医保发〔2025〕6 号，地方文件）. ；Cục Bảo hiểm y tế tỉnh An Hui và các cơ quan khác (安徽省医疗保障局等) (2025). Thông báo về việc làm tốt công tác bảo hiểm y tế cơ bản cho cư dân đô thị và nông thôn năm 2025（淮北市医保局转载）：「按照国家医保局、财政部、国家税务总局《关于做好2025年城乡居民基本医疗保障有关工作的通知》（医保发〔2025〕22号）要求」「居民医保人均财政补助标准较上年提高30元，达到每人每年不低于700元」. ；Văn phòng Quốc vụ viện (国务院办公厅) (2021). Ý kiến về việc hoàn thiện chế độ bảo hiểm y tế và trợ cấp cho bệnh hiểm nghèo nặng（国办发〔2021〕42 号）. ；Văn phòng Quốc vụ viện (国务院办公厅) (2024). Ý kiến hướng dẫn về việc hoàn thiện cơ chế lâu dài tham gia bảo hiểm y tế cơ bản（国办发〔2024〕38 号）. ；Cục Quản lý bảo hiểm y tế quốc gia (国家医保局) (2026). Báo cáo chính phủ năm 2026 nhắc đến các công tác bảo hiểm y tế này. - Ghi chú: Thông báo năm 2025 của Cục Quản lý bảo hiểm y tế quốc gia（医保发〔2025〕22 号）không được đăng công khai trên trang của Cục và trang Chính phủ Trung Quốc. Hai con số 700 tệ và 400 tệ trùng nhau trong văn bản của Hà Bắc và An Hui, bản An Hui ghi rõ là theo thông báo này mà đặt. Trong thời gian lĩnh bảo hiểm thất nghiệp, bảo hiểm y tế do quỹ bảo hiểm thất nghiệp đóng thay, không cần tham gia trùng. -### 10. Mắt bệnh nặng thì đi theo bảo hiểm y tế, bảo hiểm bệnh hiểm nghèo, trợ cấp y tế và đăng ký khám chữa bệnh ngoài địa phương, không đụng tới vay tiền trực tuyến +### 10. Mắc bệnh nặng thì đi theo bảo hiểm y tế, bảo hiểm bệnh nan y, trợ cấp y tế và đăng ký khám ngoài tỉnh trước, đừng đụng vào vay tiền trên mạng -- Chi phí: Không tốn tiền. Thứ tự là cố định, làm từng bước một. Bước một, trước khi nằm viện ở tỉnh khác, đăng ký qua ứng dụng nhỏ「国家异地就医备案」trong tài khoản WeChat chính thức của Cục Quản lý bảo hiểm y tế quốc gia（国家医保局）. Bước hai, khi ra viện thì quyết toán thẳng. Bước ba, sau khi bảo hiểm chi trả mà phần bạn tự bỏ vẫn quá nặng, xin trợ cấp y tế ở chính quyền xã hoặc văn phòng đường phố. Bước bốn, người còn đi làm thì xin nghỉ phép theo thời gian y tế; thời gian y tế là khoảng nghỉ việc vì bệnh mà không bị sa thải; thuộc tai nạn lao động thì đi theo bảo hiểm tai nạn lao động. Bước năm, không trả nổi thẻ tín dụng thì chủ động tìm ngân hàng thương lượng trả góp. Bước sáu, thật sự không đủ thì lên nền tảng quyên góp từng cá nhân do Bộ Dân vụ chỉ định để gây quỹ. Giấy tờ chủ yếu là giấy chẩn đoán, bảng chi phí, phiếu quyết toán, phải chạy 2 đến 4 lượt. Khó ở chỗ đang nằm bệnh mà vẫn phải đi từng lượt làm thủ tục. -- Nói thẳng: Thứ tự là cố định. Trước khi nằm viện ở tỉnh khác thì đăng ký trong tài khoản WeChat của Cục Quản lý bảo hiểm y tế quốc gia（国家医保局）, ra viện thì quyết toán thẳng, không phải tạm ứng toàn bộ trước. Sau khi bảo hiểm chi trả mà bạn vẫn không trả nổi, xin trợ cấp y tế ở chính quyền xã hoặc đường phố. Người còn đi làm có thời gian y tế từ 3 đến 24 tháng tính theo thâm niên, trong khoảng đó công ty không được sa thải bạn. Đừng đụng tới vay tiền trực tuyến và vay ngoài lãi cao. -- Lợi ích: Nhà nước gộp bảo hiểm y tế cơ bản, bảo hiểm bệnh hiểm nghèo và trợ cấp y tế gọi là「三重制度」, từng tầng đỡ cho tầng dưới. Người thuộc diện trợ cấp người có thu nhập thấp và người khó khăn đặc biệt, chi phí y tế đúng quy định được trợ cấp không dưới 70%. Người bệnh nặng vốn không nghèo, chữa bệnh rồi mới nghèo, có thể tự nêu đơn xin trợ cấp. Đã đăng ký trước khi nằm viện ở tỉnh khác thì ra viện quyết toán thẳng, không phải tạm ứng toàn bộ rồi quay về nơi tham gia bảo hiểm để hoàn. Nghỉ việc vì bệnh có thời gian y tế từ 3 đến 24 tháng, cụ thể bao lâu tùy thâm niên. Trong khoảng đó công ty không được chấm dứt hợp đồng lao động, tiền lương nghỉ bệnh không được thấp hơn 80% mức lương tối thiểu địa phương. Công nhân bị tai nạn lao động có thời gian giữ lương điều trị, tức khoảng chữa thương không đi làm, lương và phúc lợi cũ vẫn được trả, công ty trả theo tháng, thường không quá 12 tháng. Nợ thẻ tín dụng vượt khả năng trả nợ mà bạn vẫn muốn trả, có thể thương lượng với ngân hàng một khoản trả góp theo điều chỉnh riêng, xếp lại kế hoạch trả nợ, dài nhất 5 năm, đã ký thỏa thuận thì ngân hàng phải ngừng đòi. Nền tảng quyên góp từng cá nhân hỗ trợ người góp tiền trên mạng phải do cơ quan dân vụ của Quốc vụ viện chỉ định. Nền tảng phải xác minh thông tin nhờ cứu có thật hay không. Tiền quyên góp chỉ được chuyển cho chính người nhờ cứu hoặc tài khoản do người đó cung cấp (toàn quốc) +- Chi phí: Không tốn tiền. Thứ tự là cố định, làm từng bước một. Bước một, trước khi nằm viện ở tỉnh khác, đăng ký qua tiện ích nhỏ「国家异地就医备案」trong tài khoản WeChat chính thức của Cục Quản lý bảo hiểm y tế quốc gia. Bước hai, quyết toán thẳng khi ra viện. Bước ba, sau khi bảo hiểm hoàn trả mà phần bạn tự bỏ ra vẫn quá nặng, thì xin trợ cấp y tế ở chính quyền xã, thị trấn hoặc văn phòng đại diện khu phố. Bước bốn, đang làm việc thì xin nghỉ bệnh theo thời gian nghỉ chữa bệnh, tức khoảng thời gian nghỉ làm vì bệnh mà không bị sa thải; thuộc tai nạn lao động thì đi theo bảo hiểm tai nạn lao động. Bước năm, không trả nổi thẻ tín dụng thì chủ động tìm ngân hàng thương lượng trả góp. Bước sáu, thật sự không đủ thì mới lên nền tảng cầu viện tư nhân do Bộ Dân vụ chỉ định để quyên góp. Hồ sơ chủ yếu là giấy chẩn đoán, bảng chi phí và phiếu quyết toán, phải chạy 2 đến 4 chuyến. Khó ở chỗ đang nằm bệnh mà vẫn phải chạy từng chuyến làm thủ tục. +- Nói thẳng: Thứ tự là cố định. Trước khi nằm viện ở tỉnh khác thì đăng ký trong tài khoản WeChat của Cục Quản lý bảo hiểm y tế quốc gia, ra viện là quyết toán thẳng, không phải ứng trước toàn bộ. Hoàn trả xong mà bạn vẫn không đủ thì xin trợ cấp y tế ở chính quyền xã, thị trấn hoặc văn phòng đại diện khu phố. Đang làm việc thì tuổi công quyết định thời gian nghỉ chữa bệnh, từ 3 đến 24 tháng, trong thời gian đó công ty không được sa thải bạn. Đừng vay tiền trên mạng và đừng vay tiền lãi cao ngoài ngân hàng. +- Lợi ích: Nhà nước gộp bảo hiểm y tế cơ bản, bảo hiểm bệnh nan y và trợ cấp y tế lại gọi là「ba lớp chế độ」, đỡ từng lớp xuống dưới. Người thuộc diện trợ cấp người có thu nhập thấp và người khó khăn đặc biệt, phần chi phí khám chữa hợp quy định được trợ cấp ở mức không thấp hơn 70%. Người mắc bệnh nặng vốn không nghèo mà chữa bệnh thì nghèo, có thể tự mình đề nghị trợ cấp thêm. Đăng ký trước khi nằm viện ở tỉnh khác thì ra viện quyết toán thẳng, không phải ứng trước toàn bộ rồi về nơi tham gia bảo hiểm mà hoàn trả. Nghỉ làm vì bệnh có thời gian nghỉ chữa bệnh từ 3 đến 24 tháng, cụ thể bao lâu tùy tuổi công. Trong thời gian đó công ty không được chấm dứt hợp đồng lao động, tiền lương nghỉ bệnh không được thấp hơn 80% mức lương tối thiểu địa phương. Công nhân bị tai nạn lao động có thời gian nghỉ lương để chữa thương, tức khoảng thời gian chữa thương không đi làm, lương và phúc lợi cũ vẫn được trả như cũ, do đơn vị trả hàng tháng, thường không quá 12 tháng. Dư nợ thẻ tín dụng vượt khả năng trả mà bản thân vẫn muốn trả thì có thể thương lượng với ngân hàng một khoản trả góp theo nhu cầu, sắp lại kế hoạch trả nợ, dài nhất 5 năm, đã ký thỏa thuận thì ngân hàng phải ngừng đòi. Nền tảng cầu viện tư nhân quyên góp thay cho chính người đó phải được Bộ Dân vụ của Hội đồng Nhà nước chỉ định. Nền tảng phải xác minh thông tin cầu viện có thật không. Tiền quyên góp chỉ được chuyển cho chính người cầu viện, hoặc tài khoản do người đó cung cấp (toàn quốc) - Mức bằng chứng: A -- Nguồn: Văn phòng Quốc vụ viện (国务院办公厅) (2021). Ý kiến về việc hoàn thiện chế độ bảo hiểm và trợ cấp y tế cho bệnh hiểm nghèo nặng（关于健全重特大疾病医疗保险和救助制度的意见）（国办发〔2021〕42 号）. ；Cục Quản lý bảo hiểm y tế quốc gia (国家医保局) (2025). Khám chữa bệnh ngoài địa phương ở tỉnh khác thì hoàn thế nào? ；Bộ Lao động (劳动部) (1994). Quy định thời gian y tế cho công nhân doanh nghiệp mắc bệnh hoặc bị thương không do lao động（企业职工患病或非因工负伤医疗期规定）（劳部发〔1994〕479 号，第二、三条）. ；Bộ Lao động (劳动部) (1995). Ý kiến về việc thực hiện Luật Lao động（关于贯彻执行《劳动法》若干问题的意见）（劳部发〔1995〕309 号，mục 59）. （Sở Nhân sự và Xã hội tỉnh Cát Lâm đăng lại）；Quốc vụ viện (国务院) (2010, sửa đổi). Quy định bảo hiểm tai nạn lao động（工伤保险条例）（国务院令第 586 号，第三十、三十三条）. ；Ủy ban Giám sát ngân hàng (银监会) (2011). Biện pháp quản lý hoạt động thẻ tín dụng của ngân hàng thương mại（商业银行信用卡业务监督管理办法）（银监会令 2011 年第 2 号，第七十条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2023). Quyết định sửa đổi Luật Từ thiện（关于修改慈善法的决定）（第一百二十四条）. ；Bộ Dân vụ và bốn cơ quan khác (民政部等五部门) (2024). Biện pháp quản lý nền tảng dịch vụ nhờ cứu cá nhân qua mạng（个人求助网络服务平台管理办法）（令第 75 号，第三、十一、十三、十七条）. -- Ghi chú: Tỷ lệ trợ cấp y tế bao nhiêu, mỗi năm tối đa bao nhiêu, do nơi bạn tham gia bảo hiểm tự quy định. Thời gian y tế chia sáu bậc, tuỳ bạn đã làm việc tổng cộng bao nhiêu năm và ở đơn vị này bao lâu. Hết thời gian y tế thì công ty có thể chấm dứt hợp đồng một cách hợp pháp, nên xem trước mình đang ở bậc nào. Người cần cấp cứu cũng nằm trong danh sách phải đăng ký. Trang của Cục Quản lý bảo hiểm y tế không nói cấp cứu có được tính là đăng ký tự động không, có đăng ký bổ sung sau được không, nên gọi điện hỏi nơi tham gia bảo hiểm. Đừng đụng tới vay tiền trực tuyến và vay ngoài lãi cao. Tòa án chỉ bảo vệ tới mức bằng 4 lần LPR kỳ hạn một năm (không vay lãi cao, xem mục 15). Phần vượt quá mức đó không được bảo vệ, khoản gốc vẫn phải trả. Nằm trên giường bệnh cũng không có sức để đánh kiện kiểu đó. Bệnh nặng cần cấp cứu là chuyện khác: không có tiền vẫn phải gọi 120, bệnh viện không được từ chối, đẩy hoặc chậm trễ cứu chữa, phần cấp cứu đó do quỹ cứu hộ khẩn cấp bệnh tật chi trả, xem mục 9 ở phần 24 (cấp cứu cũng phải cứu trước). Trang gốc của văn bản 劳部发〔1995〕309 号 không mở được trên cổng Bộ Nhân sự và Xã hội, nội dung lấy từ trang sao chép của Sở Nhân sự và Xã hội tỉnh. +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2021). Ý kiến về việc hoàn thiện chế độ bảo hiểm y tế và trợ cấp cho bệnh nan y, bệnh hiểm nghèo (《关于健全重特大疾病医疗保险和救助制度的意见》)(国办发〔2021〕42 号). ；Cục Quản lý bảo hiểm y tế quốc gia (国家医保局) (2025). 跨省异地就医如何直接报销？. ；Bộ Lao động (劳动部) (1994). Quy định về thời gian nghỉ chữa bệnh cho công nhân viên doanh nghiệp mắc bệnh hoặc bị thương không do lao động (《企业职工患病或非因工负伤医疗期规定》)(劳部发〔1994〕479 号，第二、三条). ；Bộ Lao động (劳动部) (1995). Ý kiến về một số vấn đề khi thi hành Luật Lao động (《关于贯彻执行〈劳动法〉若干问题的意见》)(劳部发〔1995〕309 号, mục 59). （bản đăng lại của Sở Nhân sự và Xã hội tỉnh Cát Lâm）；Hội đồng Nhà nước (国务院) (2010 修订). Quy định về bảo hiểm tai nạn lao động (《工伤保险条例》)(国务院令第 586 号，第三十、三十三条). ；Ủy ban Tài chính (银监会) (2011). Biện pháp quản lý nghiệp vụ thẻ tín dụng của ngân hàng thương mại (《商业银行信用卡业务监督管理办法》)(银监会令 2011 年第 2 号，第七十条). ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Quyết định sửa đổi Luật Từ thiện (《慈善法》)(第一百二十四条). ；Bộ Dân vụ và bốn bộ khác (民政部等五部门) (2024). Biện pháp quản lý nền tảng dịch vụ cầu viện tư nhân (《个人求助网络服务平台管理办法》)(令第 75 号，第三、十一、十三、十七条). +- Ghi chú: Trợ cấp y tế hoàn trả bao nhiêu phần trăm, một năm nhiều nhất bao nhiêu, thì nơi bạn tham gia bảo hiểm tự quy định. Thời gian nghỉ chữa bệnh chia sáu bậc, tuỳ bạn đã làm việc tổng cộng bao nhiêu năm và ở đơn vị này bao nhiêu năm. Thời gian nghỉ chữa bệnh hết thì công ty có thể hợp pháp chấm dứt hợp đồng, nên xem trước mình đang ở bậc nào. Người được cấp cứu cũng nằm trong danh sách phải đăng ký. Trang của Cục Quản lý bảo hiểm y tế quốc gia không nói cấp cứu có tính là đăng ký tự động không, có bổ sung sau được không, hãy gọi hỏi nơi bạn tham gia bảo hiểm. Đừng vay tiền trên mạng và đừng vay tiền lãi cao ngoài ngân hàng. Tòa chỉ bảo vệ tới mức bằng 4 lần LPR kỳ hạn một năm (đừng vay tiền lãi cao, xem mục 15). Phần vượt mức đó không được bảo vệ, vốn vẫn phải trả đầy đủ. Nằm giường bệnh cũng không có sức để đánh kiện loại đó. Bệnh nặng do tai nạn lại là chuyện khác: không có tiền vẫn phải gọi 120, bệnh viện không được từ chối, đẩy trách nhiệm hoặc chậm trễ cứu chữa, phần cấp cứu ấy do quỹ trợ cấp khẩn cấp khi bệnh chi trả, xem mục 8 ở phần 24 (cấp cứu cũng phải cứu trước). Trang nguyên văn công văn 309 năm 1995 không mở được trên website Bộ Nhân sự và Xã hội, nội dung lấy từ bản đăng lại của sở nhân sự và xã hội các tỉnh. ### 11. Mất thẻ căn cước thì đi làm lại ngay, gấp thì xin thẻ tạm trước @@ -101,7 +96,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2003). Luật Thẻ căn cước cư dân（居民身份证法）（主席令第四号，第十二、二十条）. ；Bộ Công an (公安部) (2005). Biện pháp quản lý thẻ căn cước tạm của cư dân（临时居民身份证管理办法）（公安部令第 78 号，第二、七、九、十二、十七条）. - Ghi chú: Không có thẻ căn cước thì tìm việc, thuê chỗ ở, mở tài khoản, nhận trợ cấp đều không làm được, đây là điều kiện tiên quyết của mọi thủ tục. Luật Thẻ căn cước cư dân có sửa đổi năm 2011, mục này dẫn bản công bố năm 2003, điều khoản phải theo bản đang hiệu hành. Có làm ở nơi khác được không thì theo thông báo của công an địa phương. - ### 12. Làm xong đăng ký thất nghiệp thì hãy xin công nhận là người khó khăn việc làm, để nhận trợ cấp bảo hiểm xã hội hoặc làm công việc công ích - Chi phí: Không tốn tiền. Nộp đơn xin công nhận ở cơ quan dịch vụ việc làm công cộng cấp đường phố hoặc cộng đồng. Điều kiện công nhận mỗi nơi khác nhau, thường gặp là tuổi đã lớn, thất nghiệp lâu ngày, thuộc hộ có thu nhập thấp, người khuyết tật, hộ không có ai có việc làm. Phải làm xong đăng ký thất nghiệp thì mới xin được. Khó ở chỗ phải tự đối chiếu từng điều kiện một, và tiền bảo hiểm xã hội còn phải tạm ứng trước. @@ -110,7 +104,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: B - Nguồn: Bộ Tài chính và Bộ Nhân sự và Xã hội (财政部、人力资源社会保障部) (2023, sửa đổi). Biện pháp quản lý quỹ trợ cấp việc làm（就业补助资金管理办法）（财社〔2017〕164 号，2023 年 12 月修订印发）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007, sửa đổi 2015). Luật thúc đẩy việc làm（就业促进法）（第五十二、五十三条）. - Ghi chú: Lương công việc công ích không cao, nhưng có bảo hiểm xã hội, hợp để làm cầu nối. Trợ cấp bảo hiểm xã hội phần lớn đóng trước hoàn sau, phải tạm ứng tiền trước. Điều kiện cụ thể hỏi 12333 hoặc đường phố. - ### 13. Trong thời gian thất nghiệp thì đi nhận trợ cấp đào tạo nghề, trợ cấp thực tập việc làm và trợ cấp bảo hiểm xã hội, đừng tự bỏ tiền học khóa - Chi phí: Không tốn tiền. Bản thân việc đào tạo đã chiếm không ít thời gian của bạn. Khó ở chỗ phải học liên tục hết khóa, đỗ thi lấy chứng chỉ thì mới được trợ cấp. @@ -119,7 +112,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Bộ Tài chính và Bộ Nhân sự và Xã hội (财政部、人力资源社会保障部) (2023). Biện pháp quản lý quỹ trợ cấp việc làm（就业补助资金管理办法）（第五条、第七条、第十条）. - Ghi chú: Danh mục và mức trợ cấp cụ thể do từng tỉnh quy định, trước hết tra「职业培训补贴目录」trên cổng Bộ Nhân sự và Xã hội địa phương. Sinh viên tốt nghiệp thuộc hộ có thu nhập thấp, hộ không có ai có việc làm, còn sinh viên đang nợ vay vốn học bổng hoặc sinh viên khuyết tật, có trợ cấp tìm việc, khởi nghiệp một lần. - ### 14. Tìm việc đừng chỉ nộp hồ sơ, hãy dùng cách có hệ thống: học kỹ thuật, đặt mục tiêu, nhờ người khác giúp - Chi phí: Không tốn tiền. Mỗi ngày phải dành 1 đến 2 giờ, làm liên tục mấy tuần. Khóa học có phí thì không cần đăng ký, thay bằng tư vấn nghề miễn phí của cơ quan dịch vụ việc làm công cộng. Khó ở chỗ phải duy trì hằng ngày khi không ai giám sát, lại còn phải chủ động mở lời nhờ người khác giúp. @@ -128,6 +120,7 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Liu S, Huang JL, Wang M (2014). Effectiveness of job search interventions: a meta-analytic review. Psychological Bulletin 140(4): 1009–1041. - Ghi chú: Phần lớn các nghiên cứu này làm ở châu Âu và châu Mỹ, dạng là các khóa nhóm hoàn chỉnh. Chuyển sang Trung Quốc là mang kết quả từ nơi khác áp vào, không phải đo trong nước. Tìm hỗ trợ xã hội một cách cụ thể là nhờ người quen giới thiệu và nội bộ, đừng ngượng ngùng không dám mở lời. + ### 15. Không nộp tiền cọc, không giữ giấy tờ, không ký「培训贷」, không vào đa cấp, không vay lãi cao - Chi phí: Không tốn tiền. Gặp ba kiểu lời mời「nộp tiền trước rồi lên ca」「đào tạo trước rồi xếp việc」「kéo người khác vào」 thì từ chối ngay tại chỗ. Giữ lại lịch sử chat và chứng từ chuyển khoản. Khó ở chỗ càng thiếu tiền càng muốn đánh cược, nói「không」 ngay tại chỗ không phải dễ. @@ -136,7 +129,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007). Luật Hợp đồng lao động（劳动合同法）（主席令第六十五号，第九、八十四条）. ；Bộ Lao động và Bảo hiểm xã hội (劳动保障部) (2007). Quy định về dịch vụ việc làm và tuyển dụng việc làm（就业服务与就业管理规定）（劳动保障部令第 28 号，第十四、五十五、五十八条）. ；Bộ Nhân sự và Xã hội cùng bốn cơ quan khác (人力资源社会保障部等五部门) (2026). Cảnh báo nguy cơ「招转培」「培训贷」và các hình thức tương tự. ；Quốc vụ viện (国务院) (2005). Quy định cấm đa cấp（禁止传销条例）（国务院令第 444 号，第二、七、二十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Quyết định sửa đổi Quy định về một số vấn đề pháp lý khi xét vụ án vay ngoài（关于修改《关于审理民间借贷案件适用法律若干问题的规定》的决定）（法释〔2020〕6 号，第二十六条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Điều chỉnh trần lãi vay ngoài theo giới hạn bảo vệ của tòa thành「4 lần LPR kỳ hạn một năm」. - Ghi chú: Mức 4 lần LPR là trần lãi mà tòa án sẵn sàng bảo vệ, không phải「vay rồi không phải trả」, khoản gốc vẫn phải trả. Bản giải thích tư pháp này tháng 12 năm 2020 lại sửa một lần, thứ tự điều khoản có thay đổi, phải theo bản đang hiệu hành. Càng gấp dùng tiền thì càng dễ bị những chiêu này nhắm vào. Trước hết hãy đi những đường chính thống ở đầu phần này: trợ cấp bảo hiểm thất nghiệp, khiếu nại thanh tra lao động, trợ giúp pháp lý, trạm cứu hộ, trợ cấp tạm thời. - ### 16. Khó ở nhà thì xếp hàng chờ nhà ở công lập trước - Chi phí: Nộp đơn không thu phí. Nộp hồ sơ ở cơ quan bảo đảm nhà ở cấp thành phố, huyện, đồng thời bằng văn bản đồng ý cho họ kiểm tra những gì bạn khai. Thiếu giấy tờ thì họ nói một lượt cần bổ sung mấy thứ. Thời hạn xếp hàng chờ nhà gọi là thời gian chờ, thường không quá 5 năm. @@ -145,7 +137,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: B - Nguồn: Bộ Xây dựng và Nhà ở (住房和城乡建设部) (2012). Biện pháp quản lý nhà ở công lập cho thuê（公共租赁住房管理办法）（住房和城乡建设部令第 11 号，第七、八、十条）. - Ghi chú: Đây là nước xa, không cứu được ngọn lửa gần, nhưng xếp hàng không tốn tiền, đăng ký sớm thì được đến sớm. Các nơi còn có loại khác như nhà ở cho thuê có bảo đảm, quy tắc khác nhau, vào trang của Sở Xây dựng và Nhà ở địa phương tra. - ### 17. Ép chi phí ở và ăn xuống mức thấp nhất: thuê chung theo tháng thay vì thuê theo ngày, tự nấu ăn và bếp ăn trợ giúp thay cho đồ ăn giao - Chi phí: Tìm nhà phải mất 1 đến 3 ngày. Tiền cọc thường bằng 1 tháng tiền thuê (ước tính sơ bộ của tác giả). Tự nấu ăn thì mỗi ngày chiếm 30 đến 60 phút. Khó ở chỗ phải tự làm ngày này qua ngày khác, kiềm chế không gọi đồ ăn. @@ -154,7 +145,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: B - Nguồn: Cục Thống kê Quốc gia (国家统计局) (2026). Tình hình thu nhập và chi tiêu của dân cư năm 2025. ；Bộ Dân vụ và mười cơ quan khác (民政部等十一部门) (2023). Phương án hành động tích cực phát triển dịch vụ bếp ăn trợ giúp cho người cao niên（积极发展老年助餐服务行动方案）（民发〔2023〕58 号）. ；Sở Nhân sự và Xã hội tỉnh Tứ Xuyên (四川省人力资源和社会保障厅) (2026). Mức lương tối thiểu ở các tỉnh, vùng tự trị, thành phố trực thuộc trung ương trên toàn quốc（截至 2026 年 1 月 1 日，转载人社部官网表）. - Ghi chú: Nhà thuê theo ngày và nhà thuê theo giờ quy ra tháng thì đắt hơn thuê chung rất nhiều, đây là đánh giá theo kinh nghiệm, không có số liệu chính thức. Điểm cấp cơm cho người cao niên có mở cho người không cao niên không, giá bao nhiêu, mỗi nơi khác nhau. Khi tìm việc phải hỏi rõ có bao ăn ở không, việc bao ăn ở thì xoá thẳng hai khoản lớn kia. Trang bảng gốc của Bộ Nhân sự và Xã hội không mở được, số liệu lấy từ trang sao chép của Sở Nhân sự và Xã hội tỉnh Tứ Xuyên. - ### 18. Bảo hiểm xã hội đứt quãng thì đừng hoảng: lương hưu tính theo số năm tích luỹ, bảo hiểm y tế nộp bù theo quy định - Chi phí: Không tốn tiền. Bảo hiểm hưu trí cư dân và bảo hiểm y tế cư dân có thể tự đóng theo năm. Bảo hiểm xã hội của người lao động bị đứt thì không cần bỏ tiền tìm người đứng tên hộ cho「đừng đứt kỳ」. @@ -163,7 +153,6 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2010, sửa đổi 2018). Luật Bảo hiểm xã hội（社会保险法）（第十六、十九、二十七条）. ；Văn phòng Quốc vụ viện (国务院办公厅) (2024). Ý kiến hướng dẫn về việc hoàn thiện cơ chế lâu dài tham gia bảo hiểm y tế cơ bản（关于健全基本医疗保险参保长效机制的指导意见）（国办发〔2024〕38 号）. - Ghi chú: Số năm bảo hiểm y tế người lao động không phải đóng nữa sau khi về hưu, luật viết là「quy định của nhà nước」, số mỗi nơi thực hiện khác nhau, nam nữ cũng thường khác nhau, gọi 12393 hoặc hỏi Cục Quản lý bảo hiểm y tế nơi tham gia. Đứt đóng chủ yếu ảnh hưởng các điều kiện ở địa phương đòi hỏi đóng liên tục đủ số tháng, như mua nhà, lập hộ tịch, điểm cư dân, các quy tắc này do từng nơi tự quy định, tra ở địa phương. Bỏ tiền tìm người đứng tên hộ đóng hộ, tiền có thể mất trắng, cũng có rủi ro pháp lý, không nên làm. - ### 19. Từng vào tù, từng phá sản, từng bị đưa vào danh sách không thi hành án, pháp luật vẫn còn đường bắt đầu lại, trước hết làm hết thủ tục - Chi phí: Không thu phí. Người vừa hết án tù thì đến văn phòng tư pháp hoặc đường phố nơi đăng ký hộ khẩu hay nơi đang ở báo cáo, xin điều trợ cấp tổ chức lại cuộc sống (việc chính quyền sắp xếp và giúp đỡ). Người đã bị đưa vào danh sách người bị thi hành án không thi hành thì trả xong phần phải trả, rồi xin tòa án thi hành án xoá tên. Cá nhân tham gia bảo hiểm đủ 3 năm ở Thâm Quyến thì có thể xin phá sản cá nhân. Mỗi bước đều phải chuẩn bị giấy tờ, trình bày đối diện, rồi còn có thời gian quan sát 1 đến 3 năm. Khó ở chỗ phải kể lại chuyện đã qua nhiều lần, và thời gian quan sát vẫn phải chịu. @@ -172,15 +161,15 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2026). Luật Cải tạo nhà tù（监狱法）（2026 年修订，第五十九、六十、六十二条）. ；Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Quốc vụ viện (中共中央办公厅、国务院办公厅) (2010). Thông báo chuyển tải Ý kiến của Ủy ban Trung ương xây dựng và quản lý an ninh xã hội về việc tăng cường công tác tổ chức lại cuộc sống và giúp đỡ cho người vừa hết án tù và được giải tội（转发《中央社会治安综合治理委员会关于进一步加强刑满释放解除劳教人员安置帮教工作的意见》的通知）（中办发〔2010〕5 号）. （Chính quyền quận Tư Dương, thành phố Dương Châu đăng lại）；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2018). Luật Tố tụng hình sự（刑事诉讼法）（2018 年修正，第二百八十六条）. ；Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao, Bộ Công an và Bộ Tư pháp (2022). Biện pháp thi hành niêm phong lý lịch phạm tội của người chưa thành niên（关于未成年人犯罪记录封存的实施办法）（第九、十五条）. （Sở Công an tỉnh Hồ Nam đăng lại）；Tòa án nhân dân tối cao (最高人民法院) (2017, sửa đổi). Một số quy định về công bố thông tin người bị thi hành án không thi hành（关于公布失信被执行人名单信息的若干规定）（第二、十条）. ；Ủy ban Thường vụ Đại biểu Nhân dân thành phố Thâm Quyến (深圳市人大常委会) (2020). Quy định về phá sản cá nhân của Khu kinh tế đặc biệt Thâm Quyến（深圳经济特区个人破产条例）（第二、二十三、九十五、九十七、一百零三条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2006). Luật Phá sản doanh nghiệp（企业破产法）（主席令第五十四号，第二、七条）. - Ghi chú: Tiền án của người thành niên không niêm phong. Điều 100 của Bộ luật Hình sự đòi hỏi phải khai trung thực việc mình đã bị xử phạt hình sự khi nhập quân hay tìm việc. Các nghề công chức, thẩm phán, luật sư, giáo viên và việc thẩm tra lý lịch chính trị có hạn chế riêng theo pháp luật. Bản đang hiệu hành của Luật Cải tạo nhà tù trước ngày 1 tháng 11 năm 2026 tương ứng với điều 37, 38, số điều lấy theo bản có hiệu lực. Văn bản 中办发〔2010〕5 号 không tìm thấy trang gốc trên cổng Chính phủ trung ương, lấy từ trang sao chép của chính quyền quận. Trong thời gian quan sát của phá sản cá nhân thì không được đi hạng nhất máy bay, không được tiêu vào cao, không được mua nhà; các khoản nợ như bồi thường do cố ý xâm phạm thân thể người khác thuộc 8 loại không được xoá; phần miễn trừ giành được bằng gian lận còn có thể bị thu hồi. Được xoá khỏi danh sách không thi hành án không có nghĩa lịch sử tín dụng cũng sạch theo, đó là hai hệ thống khác nhau. -### 20. Trước khi mắc bệnh nặng, mua thêm một năm bảo hiểm chi phí y tế hoặc bảo hiểm bệnh hiểm nghèo ngoài bảo hiểm y tế cơ bản, xem kỹ bốn chữ「保证续保」 + +### 20. Trước khi mắc bệnh nặng, ngoài bảo hiểm y tế cơ bản thì mua một bảo hiểm y tế hoặc bảo hiểm bệnh hiểm nặng hạn một năm, và đọc kỹ bốn chữ「保证续保」 -- Chi phí: Người trẻ mua bảo hiểm chi phí y tế bồi hoàn theo chi phí, hiệu lực một năm, mỗi năm vài trăm tệ. Bồi hoàn theo chi phí nghĩa là chi trả theo đúng số tiền bạn đã bỏ ra. Bảo hiểm bệnh hiểm nghèo mỗi năm từ vài nghìn tệ trở lên (tác giả ước lượng, không có số chính thức). Trước khi mua bạn phải khai trung thực tình trạng sức khỏe, mất chừng nửa tiếng một lần. -- Nói thẳng: Bảo hiểm này bù cho phần bạn phải tự trả sau khi bảo hiểm y tế đã thanh toán, và cả khoản thu nhập mất khi bạn nghỉ làm. Trước khi mua, tìm cho ra bốn chữ「保证续保」. Không có bốn chữ đó thì sang năm sản phẩm có thể ngừng bán hoặc tăng giá. Tình trạng sức khỏe phải khai trung thực, giấu thì khi xảy ra sự cố sẽ bị từ chối bồi hoàn. Trước tiên đóng bảo hiểm y tế cơ bản đã, rồi hãy cân nhắc mục này. -- Lợi ích: Cơ quan quản lý chia sản phẩm này thành hai loại. Bảo hiểm chi phí y tế chi trả theo đúng số tiền y tế bạn đã bỏ ra. Bảo hiểm bệnh hiểm nghèo, khi được chẩn đoán đúng một trong những bệnh ghi trong hợp đồng, sẽ trả thẳng một khoản theo mức đã thỏa thuận. Cả hai loại đều bù cho phần bạn phải tự trả sau khi bảo hiểm y tế đã thanh toán, và cho khoản thu nhập mất khi nghỉ làm. Thời hạn bảo hiểm từ một năm trở xuống mà điều khoản không ghi「保证续保」thì thuộc bảo hiểm sức khỏe ngắn hạn. Quá một năm, hoặc có ghi「保证续保」, mới là bảo hiểm sức khỏe dài hạn. Nghĩa của「保证续保」là: đến kỳ gia hạn, công ty phải tiếp tục bảo hiểm theo đúng điều khoản cũ và mức phí đã thỏa thuận. Công ty bảo hiểm không được gợi ý cùng một người mua lặp lại bảo hiểm chi phí y tế có chức năng bảo vệ giống hoặc tương tự. Mua hai hợp đồng, cùng một khoản tiền cũng chỉ được chi trả một lần. Định nghĩa bệnh của bảo hiểm bệnh hiểm nghèo có quy chuẩn thống nhất trong ngành, gồm 28 bệnh nặng và 3 bệnh nhẹ. Ung thư ác tính, nhồi máu cơ tim cấp, di chứng sau đột quỵ là những bệnh cốt lõi (toàn quốc, áp dụng từ 2019、2020) +- Chi phí: Người trẻ mua bảo hiểm y tế bồi thường hạn một năm, mỗi năm vài trăm tệ. Bảo hiểm y tế bồi thường là loại hoàn theo số tiền bạn thực chi. Bảo hiểm bệnh hiểm nặng mỗi năm từ mấy nghìn tệ trở lên (tác giả ước chừng, không có số liệu chính thức). Trước khi mua phải khai trung thực tình trạng sức khoẻ, tốn nửa tiếng cho một lần. +- Nói thẳng: Loại bảo hiểm này bù đúng phần bạn phải tự bỏ sau khi bảo hiểm y tế hoàn, cộng thêm thiệt hại do ngừng việc. Trước khi mua, nhận ra đúng bốn chữ「保证续保」: không có bốn chữ này thì năm sau có thể ngừng bán hoặc tăng giá. Tình trạng sức khoẻ phải khai trung thực, giấu thì bị từ chối bồi thường. Hãy đóng bảo hiểm y tế cơ bản trước, rồi mới tính tới mục này. +- Lợi ích: Cơ quan nhà nước chia loại sản phẩm này thành hai. Bảo hiểm y tế hoàn theo chi phí y tế bạn thực chi. Bảo hiểm bệnh thì khi chẩn đoán đúng một trong các bệnh đã thoả thuận trong hợp đồng thì trả thẳng một khoản tiền theo mức đã thoả thuận. Cả hai đều bù đúng phần bạn phải tự bỏ sau khi bảo hiểm y tế hoàn, cùng thiệt hại do ngừng việc. Thời hạn bảo hiểm từ một năm trở xuống mà điều khoản không ghi bảo đảm gia hạn, thì là bảo hiểm sức khoẻ ngắn hạn. Trên một năm, hoặc đã ghi bảo đảm gia hạn, mới là bảo hiểm sức khoẻ dài hạn. 「保证续保」 nghĩa là: khi đến hạn gia hạn, công ty phải tiếp tục bảo hiểm theo điều khoản cũ và mức phí đã thoả thuận. Công ty bảo hiểm không được xúi giục cùng một người mua trùng loại bảo hiểm y tế bồi thường có chức năng bảo hiểm giống nhau hoặc tương tự. Mua hai bản thì một khoản tiền cũng chỉ hoàn một lần. Định nghĩa bệnh của bảo hiểm bệnh hiểm nặng có quy chuẩn thống nhất của ngành, gồm 28 loại bệnh nặng và 3 loại bệnh nhẹ. Ung thư ác tính, nhồi máu cơ tim cấp, diệt vằng sau đột quỵ là các bệnh chính (toàn quốc, từ năm 2019, 2020) - Mức bằng chứng: B -- Nguồn: Cục Giám soát bảo hiểm và ngân hàng (银保监会) (2019). Biện pháp quản lý kinh doanh bảo hiểm sức khỏe（银保监会令 2019 年第 3 号，第二、四、四十一条）. ；Hiệp hội Bảo hiểm Trung Quốc và Hiệp hội Y khoa Trung Quốc (中国保险行业协会、中国医师协会) (2020). Hỏi đáp về quy chuẩn sử dụng định nghĩa bệnh của bảo hiểm bệnh hiểm nghèo（2020 年修订版）. ；Cục Giám soát bảo hiểm và ngân hàng (银保监会) (2020). Thông báo về việc sử dụng《Bảng tỷ lệ mắc bệnh hiểm nghèo của ngành bảo hiểm nhân thọ Trung Quốc（2020）》（银保监发〔2020〕51 号）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2015 sửa đổi). Luật Bảo hiểm nước Cộng hòa Nhân dân Trung Hoa第十六条（如实告知与合同成立二年后不得解除）. -- Ghi chú: Chọn mức B là vì phía chính phủ chỉ đưa ra định nghĩa và quy tắc, không có con số「mua rồi tiết kiệm được bao nhiêu」.「百万医疗险」chỉ là tên bán hàng, về pháp luật nó là bảo hiểm sức khỏe ngắn hạn. Điều khoản không ghi「保证续保」thì sang năm có thể ngừng bán hoặc tăng giá, trước khi mua hãy tìm bốn chữ đó trong điều khoản. Hai hợp đồng bảo hiểm chi phí y tế chi trả theo chi phí thực tế, cùng một khoản tiền không được chi trả hai lần, đừng mua chồng lên nhau. Đã có bệnh mà vẫn mua và không khai trung thực, khi xảy ra sự cố sẽ bị từ chối bồi hoàn. Khai trung thực có thời hạn hai năm: hợp đồng tròn hai năm, công ty bảo hiểm không thể lấy lý do bạn không khai trung thực để hủy hợp đồng, xảy ra việc là phải bồi hoàn. Trong hai năm mà bị phát hiện cố ý không khai, việc xảy ra trước khi hợp đồng bị hủy sẽ không được bồi hoàn, tiền phí cũng không hoàn. Những khoản thiệt hại nào xứng đáng để bảo hiểm gánh, xem mục 40 ở phần 5 (khoản thiệt hại không gánh nổi). Trước tiên đóng bảo hiểm y tế cư dân (mục 9) đã, rồi hãy cân nhắc mục này. Quy chuẩn định nghĩa bệnh do hiệp hội ngành đưa ra dưới sự hướng dẫn của Cục Giám soát bảo hiểm và ngân hàng, không phải văn bản của chính phủ. - +- Nguồn: Ủy ban Giám sát Ngân hàng và Bảo hiểm (2019). 健康保险管理办法（银保监会令 2019 年第 3 号，第二、四、四十一条）. ；Hiệp hội Bảo hiểm Trung Quốc, Hiệp hội Bác sĩ Trung Quốc (2020). 重大疾病保险的疾病定义使用规范（2020 年修订版）答记者问. ；Ủy ban Giám sát Ngân hàng và Bảo hiểm (2020). 关于使用《中国人身保险业重大疾病经验发生率表（2020）》有关事项的通知（银保监发〔2020〕51 号）. ；Ủy ban Thường vụ Quốc hội (2015 修正). 中华人民共和国保险法第十六条（如实告知与合同成立二年后不得解除）. +- Ghi chú: Để ở mức B vì cơ quan nhà nước chỉ đưa ra định nghĩa và quy tắc, không có con số「mua vào tiết kiệm được bao nhiêu」. 「百万医疗险」 là cách gọi lúc bán hàng, theo quy định nó là bảo hiểm sức khoẻ ngắn hạn. Điều khoản không ghi「保证续保」thì năm sau có thể ngừng bán hoặc tăng giá, trước khi mua hãy tìm bốn chữ này trong điều khoản. Hai bảo hiểm y tế hoàn theo chi phí thực chi thì một khoản tiền không thể hoàn hai lần, đừng mua chồng. Đã mắc bệnh mà vẫn mua bảo hiểm, lại không khai trung thực, thì khi xảy ra sự cố sẽ bị từ chối bồi thường. Khai trung thực có thời hạn hai năm: hợp đồng thành lập đủ hai năm thì công ty bảo hiểm không còn được chấm dứt hợp đồng với lý do bạn đã không khai trung thực, xảy ra sự cố phải bồi thường. Trong hai năm bị phát hiện cố tình không khai, thì sự cố xảy ra trước khi chấm dứt hợp đồng không được bồi thường, tiền phí cũng không hoàn. Những thiệt hại nào đáng dùng bảo hiểm để đỡ, xem mục 39 ở phần 5 (thiệt hại không chịu nổi). Hãy đóng bảo hiểm y tế cư dân (mục 9 của phần này) trước, rồi mới tính tới mục này. Quy chuẩn định nghĩa bệnh do hiệp hội ngành ban hành dưới hướng dẫn của Ủy ban Giám sát Ngân hàng và Bảo hiểm, không phải văn bản của chính phủ. ### 21. Thật sự không có chỗ ngủ, nơi mở cửa 24 giờ chỉ cứu vãn được một đêm, ưu tiên đến trạm cứu hộ - Chi phí: Khu chờ sân bay, nhà ga thường không mất tiền. Quán ăn nhanh, quán internet thì phải tiêu ít tiền. Ngủ không ngon, đồ đạc có nguy cơ bị lấy cắp. Khó nhất là cả đêm không dám ngủ thật. @@ -189,3 +178,11 @@ Phần này tính tiền và bảo hiểm. Nhận được bao nhiêu, được - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có văn bản chính thức - Ghi chú: Rủi ro an toàn là có thật. Giữ đồ sát người, không nhận đồ uống và đồ ăn người lạ đưa, tránh góc khuất. Ban đêm đừng ngủ ở dưới cầu, ven sông, công trường. Ngủ ngoài trời trong thời tiết cực đoan có thể mất mạng. Trạm cứu hộ ở mục 4 nuôi ăn nuôi ở miễn phí, đến đó trước. Một số nơi nửa đêm sẽ dọn người, đừng xung đột với bảo vệ. +### 22. Bị nợ lương công nhân, trước hết phân biệt ai là người thuê bạn + +- Chi phí: Không tốn tiền. Trước hết làm rõ ba việc: đối phương có đang mở cửa hàng, đặt nhà máy, làm ăn không; công việc có được giao thầu lại từ công ty hoặc công trường không; bạn có làm việc trong nhà người ta không. Đối phương có giấy phép hay không, tra trên Hệ thống công khai thông tin tín dụng doanh nghiệp quốc gia. +- Nói thẳng: Xưởng, cửa hàng chưa đăng ký kinh doanh mà nợ lương thì vẫn khiếu nại giám sát lao động, xin hoà giải lao động được, chủ xưởng bản thân cũng có thể bị truy đuổi. Công trường mà đối tác thầu nhận thầu lại, nợ tiền thì tổng thầu hoặc chủ đầu tư trả. Chỉ làm việc riêng cho gia đình hoặc cá nhân, ví dụ làm giúp việc nhà, thì không phải tranh chấp lao động, phải ra tòa khởi kiện. +- Lợi ích: Điều 2 của Luật hợp đồng lao động giới hạn đơn vị sử dụng lao động ở doanh nghiệp, tổ chức kinh tế cá thể, đơn vị phi lợi nhuệ dân sự tư nhân và các tổ chức khác, người tự nhiên không thuộc diện. Giải thích tư pháp của Tòa án nhân dân tối cao về tranh chấp lao động liệt kê vài loại tranh chấp không phải tranh chấp lao động. Một là tranh chấp giữa gia đình hoặc cá nhân với người làm giúp việc nhà. Hai là tranh chấp giữa thợ cá thể với người làm thuê, người học nghề. Ba là tranh chấp giữa hộ kinh doanh nông thôn với người lao động. Mấy loại này phải ra tòa kiện như tranh chấp dân sự thông thường, thời hiệu khởi kiện 3 năm. Đơn vị sử dụng lao động không có giấy phép kinh doanh không nằm trong ngoại lệ đó. Điều 93 của Luật hợp đồng lao động quy định, đơn vị sử dụng lao động không đủ tư cách kinh doanh hợp pháp mà người lao động đã làm việc, đơn vị đó hoặc nhà đầu tư của họ phải trả tiền lương, bồi thường thường hưu và bồi thường tổn thất. Điều 29 của bản giải thích đó quy định, khi người lao động có tranh chấp với đơn vị sử dụng lao động chưa đăng ký kinh doanh, giấy phép bị thu hồi, hoặc hết hạn hoạt động mà vẫn tiếp tục kinh doanh, thì phải đưa đơn vị sử dụng lao động hoặc nhà đầu tư của họ vào vị thế bị đơn. Điều 33 của Quy định giám sát bảo hiểm lao động quy định, với đơn vị không có giấy phép kinh doanh hoặc đã bị thu hồi giấy phép theo quy định mà vẫn có hành vi sử dụng lao động, thì cơ quan giám sát bảo hiểm lao động tiến hành giám sát. Người cá thể nhận thầu mà tuyển dụng lao động trái pháp luật, gây thiệt hại cho người lao động, thì tổ chức giao thầu và người nhận thầu cá thể chịu trách nhiệm liên đới về bồi thường (Điều 94 của Luật hợp đồng lao động). Chủ đầu tư hoặc tổng thầu gói thầu, chia thầu cho cá nhân hoặc đơn vị không đủ tư cách kinh doanh hợp pháp, dẫn tới nợ lương công nhân lao động di cư, thì chủ đầu tư hoặc tổng thầu trả (Điều 36 của Quy định bảo đảm thanh toán tiền lương công nhân lao động di cư) (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (2012 修正). 劳动合同法（第二、九十三、九十四条）. （Kho chính sách và pháp luật Cục Thuế nhà nước đăng lại）；Tòa án nhân dân tối cao (2020). 关于审理劳动争议案件适用法律问题的解释（一）（法释〔2020〕26 号，第二、二十九条）. ；Hội đồng Nhà nước (2004). 劳动保障监察条例（国务院令第 423 号，第三十三条）. ；Hội đồng Nhà nước (2019). 保障农民工工资支付条例（国务院令第 724 号，第三十六条）. ；Quốc hội nhân dân (2020). 民法典（第一百八十八条）. ；Hội đồng Nhà nước (2006). 诉讼费用交纳办法（国务院令第 481 号，第六、二十九条）. ；Ủy ban Thường vụ Quốc hội (2023 修正). 民事诉讼法（第一百五十二、二百二十五条）. （Cục Quản lý thị trường tỉnh Quý Châu đăng lại） +- Ghi chú: Hãy phân biệt theo thứ tự này. Xem trước công việc có được giao lại từ công ty hay công trường không, có thì tìm công ty giao thầu hoặc tổng thầu. Quy định tài khoản lương riêng trên công trường và tổng thầu phải trả trước, xem mục 2 của phần này (bị nợ lương thì khiếu nại giám sát lao động trước). Xem tiếp đối phương có đang làm ăn không: mở cửa hàng, đặt nhà máy, mở xưởng, có giấy phép hay không thì đều đi theo giám sát và hoà giải, khi hoà giải thì đưa cả chủ xưởng vào danh sách bị đơn. Phần còn lại là làm việc riêng cho gia đình hoặc cá nhân thì mới ra tòa. Ra tòa phải tự khởi kiện, tự đưa chứng cứ, xét sơ thẩm theo thủ tục thông thường mất từ 6 tháng trở lên. Phí luật sư không nằm trong khoản phí tố tụng mà tòa buộc bên thua trả, thuê luật sư phải tự bỏ tiền, không đủ thì xin trợ giúp pháp lý, xem mục 3 của phần này (trợ giúp pháp lý). Trong tay có giấy nợ thì có thể xin lệnh thanh toán trước, số tiền nhỏ thì đi theo thủ tục tố tụng giá trị thấp, xem mục 22 ở phần 8 (bị lừa trong giao dịch cũ, tố tụng giá trị thấp). Thời hiệu khởi kiện 3 năm, xem mục 19 ở phần 8 (bảo vệ quyền lợi có thời hạn). Đi đường nào cũng không bảo đảm tiền về tay, đối phương không có tài sản thì thắng kiện cũng không được tiền. diff --git "a/vi/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" "b/vi/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" index 4c2d537e..a950fbff 100644 --- "a/vi/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" +++ "b/vi/book/08-\345\210\253\346\212\212\350\207\252\345\267\261\346\220\255\350\277\233\345\216\273.md" @@ -12,7 +12,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). Pháp luật An toàn giao thông đường bộ, 道路交通安全法（2021 年修订，第七十、一百零一条）. （bản đăng lại của Cục Cảnh sát giao thông Công an Bắc Kinh）；Hội đồng Nhà nước (国务院) (2017). Pháp luật thực thi về An toàn giao thông đường bộ, 道路交通安全法实施条例（国务院令第 405 号，2017 年修订，第八十六、九十二条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第一百三十三条）. ；Hiệp hội Bảo hiểm Trung Quốc (中国保险行业协会) (2020). Điều khoản mẫu bảo hiểm xe thương mại bản 2020, 机动车商业保险示范条款（2020 版）（第九、二十二条）. - Ghi chú: Có người bị thương thì cứu người trước, rồi mới báo công an. Nếu vì cứu người mà buộc phải di chuyển hiện trường thì đánh dấu lại chỗ cũ của xe và của người.「Va chạm một cái không có cảm giác nên lái đi」về sau cũng có thể bị tính là bỏ trốn, không chắc thì báo công an. Mục bảo hiểm trích điều khoản mẫu, đó là mẫu thống nhất của ngành bảo hiểm, bạn thực sự bồi thường được bao nhiêu thì lấy hợp đồng bảo hiểm trong tay mình làm chuẩn. - ### 2. Phát hiện bị lừa thì gọi ngay 110 hoặc 96110 yêu cầu dừng chi trả, đừng tự mình đi tra xét trước - Chi phí: Không tốn tiền. Gọi một cuộc điện thoại, nói rõ thời điểm chuyển khoản, số tiền và tài khoản của đối phương. Giữ lại ảnh chụp màn hình chuyển khoản và lịch sử trò chuyện. @@ -21,7 +20,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). Pháp luật Chống lừa đảo viễn thông và mạng, 反电信网络诈骗法（第二十、三十四条）. ；Công an tỉnh Phúc Kiến (福建省公安厅) (2023). 96110 来电，请务必接听. （trang của công an cấp tỉnh） - Ghi chú: Bất kỳ ai tự xưng là công an, viện kiểm sát, tòa án mà bảo bạn chuyển tiền vào 「tài khoản an toàn」 thì đều là lừa đảo. Pháp luật chỉ viết 「建立止付冻结制度」, tức là bảo ngân hàng giữ tiền lại. Đuổi được bao nhiêu còn tùy lúc bạn báo công an thì tiền còn trong tài khoản hay không. Mô tả về Ứng dụng 「国家反诈中心」 và 96110 trích từ trang của Công an tỉnh Phúc Kiến, trang gốc trên website Bộ Công an không mở được. - ### 3. Hãy nhớ quy tắc chống lừa đảo: đừng tin vội cuộc gọi, đừng lộ thông tin, đừng bấm link, chuyển khoản thì kiểm lại nhiều lần, bảy loại lừa đảo phổ biến nhất đều có hình dáng này - Chi phí: Không tốn tiền. Hãy nhớ 「ba không một nhiều」, đọc qua bảy loại dưới đây, cài thêm Ứng dụng 「国家反诈中心」, số 96110 gọi đến thì nhất định phải nghe. Bất cứ ai bảo bạn chuyển tiền, chia sẻ màn hình, tải App lạ, bấm link, báo mã xác thực thì cúp máy trước, rồi gọi số điện thoại chính thức để kiểm tra. Khó ở chỗ đối phương thường hối rất gấp, bạn phải nhịn, chịu cúp máy trước. @@ -30,7 +28,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: B - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). Pháp luật Chống lừa đảo viễn thông và mạng, 反电信网络诈骗法（第八、二十、三十四条）. ；Bộ Công an (公安部) (2026). Họp báo Bộ Công an: thông báo tình hình, đặc điểm mới của tội phạm lừa đảo viễn thông và mạng hiện nay và những điểm chính trong phòng chống（ngày 12 tháng 6 năm 2026）：「以上这10类案件就占了全部电诈案件的85%。其中，刷单返利诈骗发案最高，占所有电诈案件的25%；虚假网络投资理财诈骗损失最高，约占全部电诈案件损失的40%」. ；Công an tỉnh Phúc Kiến (福建省公安厅) (2023). 96110 来电，请务必接听. （trang của công an cấp tỉnh, có khẩu quyết 「来电不轻信，信息不透露，链接不点击，转账多核实」 và lời cảnh báo 「警惕 192 开头的诈骗电话」）；Công an tỉnh Phúc Kiến (福建省公安厅) (2023 到 2024). Chuyên mục kỹ năng nhận biết lừa đảo：刷单 ；投资理财 ；贷款 ；冒充客服 ；冒充公检法 ；杀猪盘 ；色诱刷单 - Ghi chú: Tự xưng công an, nhân viên chăm sóc khách hàng, lãnh đạo, quản trị rủi ro của nền tảng mà bảo bạn 「phối hợp điều tra」「hủy thành viên」「chạy đơn hoàn tiền」「thông tin nội bộ đầu tư」 thì đều là một khuôn như nhau. Vay thật không cần nộp tiền đặt cọc, công an viện kiểm sát tòa án không có 「tài khoản an toàn」, nhớ hai câu này là chặn được hơn một nửa. Xếp mức B vì khẩu quyết này chỉ là cách tuyên truyền của phía chính thức, không có con số 「nhớ rồi thì bị lừa ít đi bao nhiêu」. Khẩu quyết và cách phân tích bảy loại lấy từ trang của công an cấp tỉnh. 10 loại lừa đảo xảyra nhiều mà Bộ Công an báo cáo năm 2026 còn bao gồm giả mạo nhân viên chăm sóc khách hàng của sàn thương mại điện tử và giao hàng, tín dụng vay tiền, giao dịch ảo trong game online, kết bạn, hẹn hò online, giả mạo lãnh đạo, người quen và đổi, hoàn vé máy bay, khuôn đều như nhau, cuối cùng đều bắt bạn chuyển tiền. - ### 4. Thấy mặt trong video, nghe giọng trong điện thoại đều không tính là đã kiểm chứng; liên quan đến chuyển khoản thì cúp máy trước, dùng số cũ trong danh bạ của mình gọi lại - Chi phí: Không tốn tiền. Gọi thêm một cuộc điện thoại, đợi thêm vài phút. Hàng ngày少 đăng video thấy rõ mặt và đoạn thu âm dài lên tài khoản công khai. Khó ở chỗ đối phương thường hối rất gấp, bạn phải nhịn không chuyển khoản ngay tại chỗ. @@ -48,7 +45,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). Bộ luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第三十四、三十五、三十九、五十二、一百一十九、一百二十条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百四十三条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). Luật Bồi thường nhà nước, 国家赔偿法（2012 年修正，第十七、三十三条）. （bản đăng lại của Cục Thống kê Quốc gia） - Ghi chú: Đừng lén đi gặp đối phương, cũng đừng đi tìm người nhà của họ. Bất kỳ cuộc 「thương lượng riêng」nào cũng có thể bị hiểu là bạn đã nhận. Lịch sử trò chuyện, lịch sử gọi điện, lịch sử hành trình là vật liệu để bạn tự chứng minh mình trong sạch, xóa đi là tự xóa cả chứng cứ của mình. Nhớ nhầm, việc báo việc sau khi điều tra không thành thì không tính là tội tố cáo vu khống. Mục này áp dụng cho người bị buộc tội, không phân nam nữ. Ba mục triển khai xem mục 33 của phần này（trách nhiệm của bên bịa đặt sự thật）、mục 34 của phần này（không đủ chứng cứ thì phải tuyên vô tội và khiếu nại, xử lại）、mục 35 của phần này（bồi thường nhà nước sau khi bị giam tính theo ngày） - ### 6. Sau khi xảy ra chuyện hãy chủ động nói rõ: đầu thú nhiều nhất giảm được bốn phần mười hình phạt, con đường 「né thời hiệu truy tố」 đã bị chặn - Chi phí: Không tốn tiền. Tự mình đến cơ quan công an đầu thú, hoặc ở nguyên chỗ gọi điện báo công an, rồi kể lại chuyện một cách trung thực. Khó là chống đỡ được ý nghĩ 「chờ thêm xem, biết đâu không ai điều tra」. @@ -57,7 +53,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大). Bộ luật Hình sự nước Cộng hòa Nhân dân Trung Hoa, bản hợp nhất sau khi sửa đổi lần thứ mười một, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第六十七、八十七、八十八条. ；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). Hướng dẫn định lượng hình sự cho các tội phổ biến（thử điểm）, 关于常见犯罪的量刑指导意见（试行）（法发〔2021〕21 号，2021 年 7 月 1 日施行）三（六）（七）（十四）. Toàn văn được Viện kiểm sát nhân dân thành phố Trương Dịch tỉnh Cam Túc đăng lại. ；Tòa án nhân dân tối cao (最高人民法院) (2010). Một số vấn đề cụ thể về xử lý đầu thú và lập công, 关于处理自首和立功若干具体问题的意见（法发〔2010〕60 号）一. Công báo Tòa án nhân dân tối cao. - Ghi chú: Người được lợi của mục này chính là bạn. Đầu thú chỉ là tình tiết giảm nhẹ, tội vẫn bị định, giảm bao nhiêu do tòa án quyết định, hướng dẫn định lượng hình sự yêu cầu siết chặt mức giảm nhẹ với tội bạo lực nghiêm trọng và tội ma túy. Tai nạn giao thông có cách tính riêng: ở lại hiện trường, cứu người, báo với cơ quan công an thì tính là tự đầu thú, nhưng vì những việc này vốn là nghĩa vụ pháp lý nên mức giảm nhẹ phải siết chặt; chạy rồi sau đó đầu thú vẫn tính là đầu thú, chỉ là phải lấy khung hình phạt pháp luật nặng hơn làm cơ sở, rồi mới quyết định có giảm không, giảm bao nhiêu, cách làm xem mục 1 của phần này（tai nạn giao thông thì dừng xe, cứu người, báo công an trước）. Mời luật sư trước và khai trung thực không mâu thuẫn, xem mục 5 của phần này（bị buộc tội thì mời luật sư trước）: luật sư giúp bạn đối chiếu xem bản ghi lời khai viết đúng không, không phải giúp bạn bịa ra cách nói. - ### 7. Đã uống rượu thì đừng cầm vô lăng, kể cả xe điện và 「chỉ dời xe một chút」 - Chi phí: Người đưa đón thay thế hoặc đi taxi vài chục tệ. Trước khi tụ họp thì thống nhất trước hôm nay ai không uống. Khó là khi bị ép uống rượu ở chỗ đông người thì phản lại. @@ -66,7 +61,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). Pháp luật An toàn giao thông đường bộ, 道路交通安全法（2021 年修订，第九十一条）. （bản đăng lại của Cục Cảnh sát giao thông Công an Bắc Kinh）；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). Sửa đổi Bộ luật Hình sự（九）, 刑法修正案（九）（第八条，即刑法第一百三十三条之一）. ；Hiệp hội Bảo hiểm Trung Quốc (中国保险行业协会) (2020). Điều khoản mẫu bảo hiểm xe thương mại bản 2020, 机动车商业保险示范条款（2020 版）（第九、二十二条）. - Ghi chú: Say rượu lái xe là tội hình sự, sẽ để lại án tích, ảnh hưởng việc kiểm tra chính trị của bản thân và một số tư cách nghề nghiệp. 「Uống rượu」và 「say rượu」được kẻ đường theo nồng độ cồn trong máu, ranh giới cụ thể do tiêu chuẩn quốc gia quy định. Phần này chưa đối chiếu con số đó, đừng ước lượng cảm tính xem mình đã tới mức nào. - ### 8. Đừng cho ai mượn thẻ ngân hàng, sim điện thoại hay tài khoản thanh toán, 「cho thuê tài khoản nhận tiền」 không phải việc làm thêm - Chi phí: Không tốn tiền. Có người nói「mượn thẻ đi qua tài khoản một chuyến, cho bạn vài trăm tệ」, từ chối là được. @@ -75,7 +69,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). Pháp luật Chống lừa đảo viễn thông và mạng, 反电信网络诈骗法（第三十一、四十四条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). Sửa đổi Bộ luật Hình sự（九）, 刑法修正案（九）（第二十九条，即刑法第二百八十七条之二）. - Ghi chú: Nói「tôi không biết nó mang thẻ đi làm gì」không chặn được. Khi phá án người ta sẽ suy ngược từ hoàn cảnh bên ngoài xem bạn có biết không: từng nhận tiền cảm ơn, dòng tiền rõ ràng bất thường, đều có thể bị tính là bạn biết. Thẻ bị người khác dùng để nhận tiền赃 thì cho dù cuối cùng bạn không cấu thành tội phạm, tài khoản vẫn bị đóng băng, bản thân vẫn bị xử kỷ luật. - ### 9. Mỗi năm miễn phí kiểm tra báo cáo tín dụng của mình hai lần, xem có khoản vay và thẻ nào không phải mình mở không - Chi phí: Không tốn tiền. Kiểm tra trên mạng, mười phút là xong. @@ -84,7 +77,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2013). Điều lệ quản lý ngành tín dụng, 征信业管理条例（国令第 631 号，第十六条、第十七条、第二十五条）. - Ghi chú: Thẻ căn cước bị người khác chiếm dụng để vay online, mở thẻ, bản thân thường phải đến lúc mua nhà vay vốn bị từ chối mới phát hiện, lúc đó đã kéo dài mấy năm rồi. Cổng kiểm tra là website chính thức của Trung tâm tín dụng Ngân hàng Nhân dân Trung Hoa, hoặc ứng dụng ngân hàng trên điện thoại. Đừng dùng ứng dụng 「kiểm tra tín dụng」của bên thứ ba, loại ứng dụng đó bản thân nó đang thu thập thông tin của bạn - ### 10. Xung đột thì báo công an trước, đừng ra tay, người ra tay trước gần như chắc chắn thiệt - Chi phí: Không tốn tiền. Lùi đến chỗ đông người rồi báo công an, để hành động của đối phương nằm trong camera giám sát. Khó là nhịn không cãi lại, không đẩy. @@ -102,16 +94,14 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二十条）. ；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院)、Bộ Công an (公安部) (2020). Hướng dẫn về việc áp dụng đúng pháp luật chế độ phòng vệ chính đáng, 关于依法适用正当防卫制度的指导意见（法发〔2020〕31 号，mục 5, 6, 7, 10, 11, 13, 14 và 19）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2018). 第十二批指导性案例（检例第 45 号陈某正当防卫案、检例第 47 号于海明正当防卫案）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). Bộ luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第九十一条）. Toàn văn do Viện kiểm sát Khu tự trị dân tộc Hồi Ninh Hạ đăng lại - Ghi chú: Việc mặc định vẫn là mục 10 nói về lùi ra báo công an（xung đột thì báo công an trước, đừng ra tay）. Hai người đánh nhau không tính là phòng vệ, đối phương ngã rồi hoặc chạy rồi mà bạn còn đuổi đánh cũng không tính. Chỉ đến lúc không còn đường lui mới xét chuyện đánh lại. Sau đó có ba việc phải làm. Thứ nhất, sau khi báo công an thì đừng rời hiện trường. Thứ hai, giữ ngay camera và nhân chứng, hướng dẫn yêu cầu công an đi lấy录像录音 của hiện trường ngay trong giờ đầu tiên, hỏi nhân chứng ngay trong giờ đầu tiên. Thứ ba, trước khi bị cảnh sát hỏi chính thức thì mời luật sư trước, xem mục 5（bị buộc tội thì mời luật sư trước）. Cuối cùng vụ việc bị hủy, không truy tố hay tuyên vô chất thì những ngày bị giam có thể xin bồi thường nhà nước theo từng ngày, xem mục 35（bồi thường nhà nước sau khi bị giam）. Ra tay phòng vệ cho người khác cũng được pháp luật cho phép. Nhưng tay không đi can vào vụ đánh nhau giữa người lạ thì rủi ro phải tính riêng, xem mục 37 ở phần 13（chứng kiến一群人đánh nhau）. Người được lợi của mục này chủ yếu là chính bạn và người thân. - -### 12. Mắc mứa với ai — bị nợ lương, bị sa thải, bị lừa tiền — thì đi khiếu nại, hoà giải lao động, khởi kiện, đừng đi tính sổ với người ta +### 12. Bị kẻ khác ghét, bị nợ lương, bị sa thải, bị lấy tiền, thì đi theo khiếu nại, hoà giải, khởi kiện, đừng đi tìm đánh nhau -- Chi phí: Không tốn tiền, khiếu nại và hoà giải lao động đều không thu phí. Phải chạy một hai chuyến, chờ vài tháng. Khó là trong vài tháng đó nhịn không ra tay. -- Nói thẳng: Đi con đường theo pháp luật, kết quả xấu nhất là tiền đòi không về. Đi trả thù thì kết quả là tiền vẫn đòi không về, còn phải tự lấy mạng đổi. Vụ ở无锡: 徐某金 không lấy được bằng tốt nghiệp, lại không vừa ý với tiền công thực tập, cầm dao gây án khiến 8 người chết 17 người bị thương,被判死刑 và đã thi hành. Nợ lương thì đi khiếu nại và hoà giải lao động, năm 2024 cả nước nhờ con đường này追回 lương cho 654.000 người. -- Lợi ích: Con đường theo pháp luật có số chính thức. Năm 2024 cả nước thanh tra lao động xử lý 137.000 vụ,追回 lương cho 654.000 người với 8.650.000.000 tệ. Còn có 998 người bị xử tù vì tội không trả thù lao lao động. Nợ lương thì khiếu nại với thanh tra lao động thế nào, xin hoà giải lao động thế nào, xem mục 2 ở phần 7. Bảng giá của con đường trả thù viết ở Điều hai trăm ba mươi hai Bộ luật Hình sự:「故意杀人的，处死刑、无期徒刑或者十年以上有期徒刑；情节较轻的，处三年以上十年以下有期徒刑」. Ví dụ chính thức: 徐某金 thi không đạt, không lấy được bằng tốt nghiệp, lại không vừa ý với tiền công thực tập. Ngày 16 tháng 11 năm 2024, anh ta về Viện Cao đẳng Công nghệ Ngôn tình无锡 cầm dao gây án, khiến 8 người chết, 17 người bị thương. Tòa sơ thẩm xử tội cố ý giết người với hình phạt tử hình, t剥夺政治权利终身, ngày 20 tháng 1 năm 2025 được Tòa án nhân dân tối cao chuẩn准 thi hành. Người anh ta không vừa ý là trường học và khoản tiền công đó, nhưng người bị hại lại là bạn học, khoản tiền công đó anh ta cũng không nhận được một xu (toàn quốc) +- Chi phí: Không tốn tiền, khiếu nại và hoà giải đều không thu phí. Phải chạy một hai chuyến, chờ mấy tháng. Khó ở chỗ phải nhịn được trong mấy tháng đó. +- Nói thẳng: Đi theo đường pháp luật thì kết quả tệ nhất vẫn là không đòi được tiền. Đi báo thù thì kết quả vẫn là không đòi được tiền, còn phải tự đi đền mạng. Vụ việc ở Vô Dung, anh Hứ Kim bút không lấy được bằng tốt nghiệp, lại bất mãn khoản thù lao thực tập, cầm dao gây án làm 8 người chết, 17 người bị thương, bị kết án tử hình và đã thi hành. Đòi lương bị nợ mà đi khiếu nại và hoà giải thì năm 2024 trên toàn quốc đường này đã giúp 654.000 người lấy lại 8.650.000.000 tệ. +- Lợi ích: Đường pháp luật có số liệu chính thức. Năm 2024, giám sát lao động toàn quốc xử lý 137.000 vụ, thu hồi 8.650.000.000 tệ tiền lương cho 654.000 người. Còn có 998 người bị kết án tội không trả tiền lương. Bị nợ lương thì khiếu nại giám sát lao động ra sao, xin hoà giải lao động ra sao, xem mục 2 ở phần 7. Bảng giá của đường báo thù nằm trong Điều 232 của Bộ luật Hình sự:「故意杀人的，处死刑、无期徒刑或者十年以上有期徒刑；情节较轻的，处三年以上十年以下有期徒刑」. Ca thực tế chính thức: anh Hứ Kim thi trượt, không lấy được bằng tốt nghiệp, rồi bất mãn khoản thù lao thực tập. Ngày 16 tháng 11 năm 2024, anh quay lại trường Công nghệ Vô Dung cầm dao gây án, làm 8 người chết, 17 người bị thương. Phúc thẩm tuyên tử hình theo tội giết người, tước quyền chính trị vĩnh viễn, ngày 20 tháng 1 năm 2025 được Tòa án nhân dân tối cao phê chuẩn thi hành. Thứ hắn bất mãn là trường và khoản thù lao đó, kẻ bị hại lại là bạn cùng khóa, khoản thù lao ấy hắn cũng không nhận được đồng nào (toàn quốc) - Mức bằng chứng: A -- Nguồn: Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百三十二条）. ；Tân Hoa xã (新华社) (2024). 江苏宜兴一校园发生持刀伤人事件致 8 死 17 伤. （chuyển bản thông báo của cảnh sát Nghi Hưng, chứa số người thương vong và nguyên văn động cơ）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 无锡校园持刀伤人案罪犯徐加金被执行死刑. ；Bộ Nhân lực和社会保障 (人力资源社会保障部) (2025). 2024 年度人力资源和社会保障事业发展统计公报. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 最高人民法院工作报告（「对驾车冲撞行人的樊维秋、校园持刀行凶的徐加金等判处死刑」）. -- Ghi chú: Thực tập sinh còn đi học thì thường không tính là đã thiết lập quan hệ lao động với đơn vị. Tiền công thực tập đòi không được thì chỉ còn cách nhờ trường đứng ra xử lý, hoặc ra tòa khởi kiện, không đi được con đường hoà giải lao động. Bị nợ lương thì khiếu nại cụ thể thế nào, hoà giải thế nào, xem mục 2 ở phần 7. Đường đỏ khi đi đòi nợ xem mục 15 ở phần 9（đòi nợ thì không bắt giữ người, không giam người）. Đi con đường theo pháp luật cũng có thể đòi không về tiền, vì vụ việc kết thúc không đồng nghĩa tiền đã về tài khoản. Đi trả thù thì tiền vẫn đòi không về, còn phải tự lấy mạng đổi. - +- Nguồn: Quốc hội nhân dân (1997). 刑法（第二百三十二条）. ；Tân Hoa Xã (2024). 江苏宜兴一校园发生持刀伤人事件致 8 死 17 伤. （Chuyển thông cáo của công an Di Hưng, có số người chết bị thương và nguyên văn động cơ）；Viện kiểm sát nhân dân tối cao (2025). 无锡校园持刀伤人案罪犯徐加金被执行死刑. ；Bộ Nhân sự và Xã hội (2025). 2024 年度人力资源和社会保障事业发展统计公报. ；Tòa án nhân dân tối cao (2025). 最高人民法院工作报告（「对驾车冲撞行人的樊维秋、校园持刀行凶的徐加金等判处死刑」）. +- Ghi chú: Người thực tập còn đang học thì đại đa số không được coi là đã thiết lập quan hệ lao động với đơn vị. Đòi được khoản thù lao thực tập thì chỉ có thể nhờ nhà trường phối hợp, hoặc ra tòa kiện, không đi được con đường hoà giải lao động. Làm việc riêng cho gia đình hoặc cá nhân mà bị nợ tiền thì cũng phải ra tòa, xem mục 22 ở phần 7 (phân biệt trước ai là người thuê bạn). Bị nợ lương thì cụ thể khiếu nại ra sao, hoà giải ra sao, xem mục 2 ở phần 7. Ranh giới khi đòi nợ xem mục 15 ở phần 9 (đòi nợ không giam người, không bắt người). Đi đường pháp luật cũng có thể không đòi được tiền, vì kết thúc vụ án không có nghĩa tiền về tay. Đi báo thù thì tiền vẫn không đòi được, còn phải tự đi đền mạng. ### 13. Dù căm đến đâu cũng đừng ra tay với người không liên quan: lái xe đâm vào đám đông, gây án ở nơi công cộng thì xử tội gây nguy hiểm cho an toàn công cộng bằng phương pháp nguy hiểm, khởi tố từ ba năm, có người chết thì tử hình - Chi phí: Không tốn tiền. Khó là trong lúc nóng giận nén được ý nghĩ này lại. @@ -120,15 +110,15 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第一百一十四、一百一十五条）. ；Tân Hoa xã (新华社) (2024). 广东珠海驾车撞人事件造成 35 人死亡 43 人受伤. （chuyển bản thông báo của Công an thành phố Chu Hải）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2024). 樊维秋以危险方法危害公共安全案一审宣判，判处死刑，剥夺政治权利终身. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 珠海驾车撞人案罪犯樊维秋被执行死刑. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 法律：治安管理处罚法（2025 年修订，第二十九、五十条）. - Ghi chú: 「Suy cho cùng tôi cũng không muốn sống nữa」trong định lượng hình sự không phải là mức giảm. Báo cáo công tác năm 2025 của Tòa án nhân dân tối cao nói về loại vụ này là「đối với những kẻ tội ác tày trời thách thức đường lối pháp luật và đạo đức, kiên quyết xử lý nghiêm, nặng và nhanh theo pháp luật」. Chỉ mới nói suông cũng đã vi phạm pháp rồi. Nói suông về việc phóng hỏa, nổ bom, đầu độc这类危害公共安全的事，把公共秩序搅乱了的，拘留 5 đến 10 ngày, còn phạt được dưới 1000 tệ. Viết thư đe dọa, hoặc dùng cách khác đe dọa tính mạng thân thể người khác, cũng vi phạm pháp. Loại này phải bị giữ dưới 5 ngày, hoặc phạt dưới 1000 tệ（Pháp luật Xử phạt hành chính về trật tự công cộng có hiệu lực từ ngày 1 tháng 1 năm 2026）. Đã nghĩ đến bước này rồi thì xem mục 14（xuất hiện ý nghĩ 「kéo một người chết theo」thì xử như cấp cứu）. - -### 14. Nổi lên ý nghĩ 「kéo một người chết theo」「cùng chết」thì xử như cấp cứu: rời hiện trường, đưa chìa khóa xe và dao cho người khác, gọi 12356 +### 14. Khi nảy ra ý「kéo một người chết cùng」「cùng chết cùng sống thì cùng chết」, hãy xử lý như cấp cứu: rời hiện trường, đưa chìa khóa xe và dao cho người khác, gọi 12356 -- Chi phí: Không tốn tiền. Khó là thừa nhận mình lúc này không ổn, cũng khó là đưa đồ vào tay người khác. -- Nói thẳng: Loại ý nghĩ này từ lúc nảy ra đến lúc ra tay thì cửa sổ rất ngắn. Trong 82 người tự tử không thành, 47.6% nói từ lần đầu nảy ý đến khi ra tay không quá 10 phút. Nên việc cần làm không phải 「nghĩ cho thông」, mà là trước hết tách vô lăng, dao, van ga, dây thừng và bản thân ra một bước,熬过 mười phút này. Rồi gọi 12356. -- Lợi ích: Cửa sổ từ lúc nảy ý đến lúc ra tay của loại ý nghĩ này rất ngắn. Trong phỏng vấn 82 người tự tử không thành trong vòng ba ngày nhập viện, 47.6% nói từ lúc ý nghĩ lần đầu xuất hiện đến khi làm không quá 10 phút. Một cuộc tổng quan hệ thống 10 năm（gộp các nghiên cứu trong mười năm lại soát一遍）cho rằng bằng chứng về việc hạn chế phương tiện gây tử vong đang ngày càng mạnh. Sau khi kiểm soát thuốc giảm đau, số ca tự tử liên quan giảm 43%. Sau khi thêm hàng rào ở các điểm nhảy lầu, tự tử giảm 86%（79% đến 91%）. Tách vô lăng, dao, van ga, dây thừng ra một bước, chờ的就是 mười phút này qua đi. 12356 là đường dây trợ giúp tâm lý thống nhất toàn quốc, mỗi ngày nghe máy không dưới 18 tiếng (toàn quốc, từ ngày 1 tháng 5 năm 2025) +- Chi phí: Không tốn tiền. Khó ở chỗ phải thừa nhận rằng lúc này mình không ổn, cũng khó ở chỗ phải đưa đồ vật cho người khác. +- Nói thẳng: Loại ý nghĩ này, từ lúc nảy sinh đến lúc ra tay, có cửa sổ rất ngắn. Trong 82 người từng thử tự sát, 47.6% nói rằng từ lần đầu nảy ý đến khi ra tay không quá 10 phút. Vậy việc cần làm không phải là「nghĩ cho ra」, mà là trước hết tách vô-lăng, dao, van gas, dây thừng và bản thân ra một bước, để qua mười phút đó. Rồi gọi 12356. +- Lợi ích: Loại ý nghĩ này tới lúc ra tay có cửa sổ rất ngắn. Trong các buổi phỏng vấn 82 người từng thử tự sát trong vòng ba ngày sau khi nhập viện, 47.6% nói rằng từ lần nảy ý đầu tiên đến lúc thực hiện không quá 10 phút. Một tổng quan hệ thống trong 10 năm (gộp lại rà soát các nghiên cứu trong mười năm) cho rằng bằng chứng về hiệu quả của việc hạn chế phương tiện gây tử vong đang ngày càng mạnh. Sau khi kiểm soát thuốc giảm đau, các vụ tự sát liên quan giảm 43%. Sau khi bảo vệ các điểm nhảy từ trên cao, tự sát giảm 86% (từ 79% đến 91%). Tách vô-lăng, dao, van gas, dây thừng ra một bước, thứ chờ đợi chính là mười phút đó trôi qua. 12356 là đường dây trợ giúp tâm lý thống nhất toàn quốc, mỗi ngày nghe không dưới 18 giờ (toàn quốc, từ 1 tháng 5 năm 2025) - Mức bằng chứng: B -- Nguồn: Deisenhammer EA 等 (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry. ；Zalsman G 等 (2016). Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry. ；Ủy ban Y tế và Sức khỏe Quốc gia (国家卫生健康委) (2024). 关于应用「12356」全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. -- Ghi chú: Lý do xếp mức B là cả hai nghiên cứu này đều lấy từ tài liệu研究自杀，lấy áp vào「suy nghĩ muốn làm hại người khác」thì không có nghiên cứu trực tiếp. Nghiên cứu của Deisenhammer mới chỉ có 82 người, mới làm ở một bệnh viện. Trạng thái không ổn thì đừng lái xe, cũng đừng đi ca trực nào mà phải chịu trách nhiệm an toàn cho người khác, bàn giao ca trước đã. Muốn làm hại chính mình thì xem mục 25 ở phần 1（có ý nghĩ tự tử thì gọi 12356）. Tâm trạng xuống thấp thì làm gì trước, xem mục 19 ở phần 3. +- Nguồn: Deisenhammer EA 等 (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry. ；Zalsman G 等 (2016). Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry. ；Ủy ban Y tế và Sức khoẻ quốc gia (2024). 关于应用「12356」全国统一心理援助热线电话号码的通知（国卫医政函〔2024〕259 号）. +- Ghi chú: Để ở mức B vì cả hai bằng chứng đó đều đến từ tài liệu nghiên cứu về tự sát, chuyển sang「xung hăng muốn làm hại người khác」thì không có nghiên cứu trực tiếp. Nghiên cứu của Deisenhammer còn chỉ có 82 người, và chỉ làm ở một bệnh viện. Lúc trạng thái không ổn thì đừng lái xe, cũng đừng đi trực cao loại phải gánh trách nhiệm an toàn cho người khác, giao cao rồi nói sau. Muốn làm hại chính mình thì xem mục 25 ở phần 1 (có ý nghĩ tự sát thì gọi 12356). Lúc tâm trạng trầm thì làm gì trước, xem mục 18 ở phần 3. + ### 15. Người thân cận nói 「không ai muốn sống yên」, 「tao đưa con đi cùng」 thì đừng coi là lời giận: người thân gần có thể đưa họ đi khám, công an đã nhận báo cáo cũng phải xử lý @@ -138,7 +128,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). Pháp luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订，第二十九、五十条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012, sửa đổi 2018). Luật Sức khoẻ Tâm thần, 精神卫生法（第二十八、二十九、三十、三十五条）. ；Cục Điều tra Tai nạn Hàng không (BEA), Pháp (2016). Final Report BEA2015-0125.en: Accident on 24 March 2015 at Prads-Haute-Bléone to the Airbus A320-211 registered D-AIPX operated by Germanwings. - Ghi chú: Xét theo người được lợi, mục này tính tới vợ chồng, cha mẹ, con cái sống cùng bạn. 「Không ai muốn sống yên」 thường đã bao gồm cả bạn, và pháp luật trao quyền đưa đi khám đúng cho người thân gần, người ngoài muốn quản cũng không được, chỉ có thể báo công an. Đưa đi khám chỉ dùng cho tín hiệu nguy hiểm có thật, lời giận lúc cãi nhau không tính, cũng không dùng nó để nhốt người nhà không nghe lời vào bệnh viện. Cùng bộ luật đó còn viết: sau khi khám lại hoặc giám định mà xác định người đó không bị rối loạn tâm thần nặng thì bệnh viện không được giữ họ nhập viện（Điều ba mươi lăm）. Không phân biệt được thì gọi 12356 hỏi trước, hoặc báo 110 cho công an tới nơi. Kết luận điều tra chuyến bay 9525 của Germanwings là: không có cách nào để chính quyền và chủ lao động ngăn được phi hành đoàn phó bay vào ngày hôm đó. Nguyên nhân là 「无论是他本人，还是任何其他人——医生、同事或者家属——都没有告知他们」. Cả 150 người trên máy đều tử vong, gồm 2 phi hành đoàn, 4 tiếp viên và 144 hành khách. - ### 16. Trên mạng thì đừng chửi người, đừng bịa tin, đừng chia sẻ tin chưa kiểm chứng; bị tấn công trên mạng thì lưu bằng chứng rồi báo công an - Chi phí: Không tốn tiền. Trước khi đăng thì nghĩ thêm mười giây. Bị tấn công thì chụp màn hình, quay video, lưu lại đường dẫn và ID tài khoản, trước hết khiếu nại với nền tảng, rồi báo công an. Khó là nhịn không chửi lại lúc bị chửi. @@ -147,7 +136,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). Pháp luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订，第二十九、五十条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百四十六条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). Sửa đổi Bộ luật Hình sự（九）, 刑法修正案（九）（第十六条，增加刑法第二百四十六条第三款）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). 刑法修正案（九）（第二百四十六条新增第三款）. Toàn văn đăng lại bởi Vụ Xử lý thư khiếu nại và thư án dân chủng quốc gia (国家信访局) - Ghi chú: Với hai tội xúc phạm và vu khống, nguyên tắc là 「có người báo thì mới xử lý」, nghĩa là phải tự bạn ra tòa kiện. Công an thường không tự khởi kiện, chứng cứ chỉ tự bạn tích góp. Nhưng có hai ngoại lệ. Một là trừ 「trường hợp gây tổn hại nghiêm trọng đến trật tự xã hội và lợi ích quốc gia」, trường hợp này công an có thể trực tiếp làm việc đó (khoản hai Điều hai trăm bốn mươi sáu của Bộ luật Hình sự). Hai là bị vu khống trên mạng mà bản thân thu thập chứng cứ thật sự khó, tòa có thể yêu cầu công an phụ giúp điều tra (khoản ba cùng điều đó, thêm vào bằng Sửa đổi thứ chín Bộ luật Hình sự). Đã chia sẻ lại thì vẫn phải chịu trách nhiệm, biết là giả mà vẫn phát ra ngoài thì vẫn tính là 「bịa sự thật」. Bản thân bị tấn công trên mạng rồi xử lý thế nào, xem mục 37 của phần này（bị tấn công trên mạng: chặn bảo vệ và lưu giữ bằng chứng trước, rồi chọn cách）. Số điều của tội xúc phạm, vu khống đổi từ Điều bốn mươi hai của luật cũ sang Điều năm mươi của luật mới. - ### 17. Đọc hết tờ giấy trước khi ký, không ký hộ người khác, không ký trên giấy trắng - Chi phí: Không tốn tiền. Dành thêm vài phút, đọc từ đầu đến cuối hợp đồng, giấy mượn, giấy bảo lãnh, hồ sơ nhận việc. Chỗ không hiểu thì hỏi tại chỗ, rồi chụp ảnh lưu lại. Khó là đọc chậm rãi trước mặt đối phương. @@ -156,7 +144,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第四百六十五、五百零九条）. - Ghi chú: 「Không ký hộ người khác, không ký giấy trắng, ghi rõ mục đích sử dụng trên bản sao thẻ căn cước」 là cách làm theo kinh nghiệm, không có văn bản chính thức riêng nào. Chúng cùng một mục đích: đừng để chữ ký của bạn bị ghép vào một hợp đồng có hiệu lực pháp lý mà bạn lại không tránh được phải thừa nhận. Chữ ký điện tử và xác nhận bằng nhận diện khuôn mặt cũng tính là ký. - ### 18. Cho vay thì viết giấy mượn cho đủ, trước khi bảo lãnh cho người khác hãy nghĩ xem mình có sẵn lòng trả thay không - Chi phí: Không tốn tiền. Dùng một tờ giấy ghi rõ người cho vay, người vay, số tiền, lãi suất, thời hạn và cách trả, hai bên cùng ký. Tiền chuyển qua tài khoản để lại dấu vết, không đưa tiền mặt. @@ -165,7 +152,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第六百六十八、六百八十一、六百八十六、六百八十七条）. - Ghi chú: Phần lãi vượt quá 4 lần LPR thì tòa không bảo vệ, xem phần 7. Ghi số thẻ căn cước và ghi 「vay」 trên giấy mượn là cách làm theo kinh nghiệm. Trên giấy bảo lãnh có chữ 「liên」 hay không, quyết định bạn đứng sau người vay hay đứng ngang hàng với họ để bị đòi, phải đọc kỹ trước khi ký. - ### 19. Bảo vệ quyền lợi có thời hạn: thời hiệu khởi kiện dân sự là 3 năm, hoà giải lao động là 1 năm, đối phương chỉ cần một câu 「quá thời hiệu」 là đủ - Chi phí: Không tốn tiền. Đòi nợ thì đừng chỉ nghĩ trong đầu, dùng WeChat, tin nhắn, email hoặc thư bảo đảm để lại dấu vết. @@ -174,6 +160,7 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一百八十八、一百八十九、一百九十二、一百九十三、一百九十五、一百九十六、一百九十七条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). Luật Hoà giải và Trọng tài tranh chấp lao động, 劳动争议调解仲裁法（第二十七条）. （bản đăng lại của Sở Nhân sự và Xã hội thành phố An Khang） - Ghi chú: Đây có lẽ là việc rẻ nhất cả sách: một tin nhắn đòi nợ có dấu vết là 3 năm tính lại từ đầu; không làm gì thì có thể mất trắng quyền thắng kiện. Thỏa thuận trước 「tôi không viện dẫn thời hiệu」 thì không có giá trị, hai bên thỏa thuận kéo dài hay rút ngắn 3 năm cũng không có giá trị. Ngược lại, quá thời hiệu mà đối phương vẫn đồng ý trả thì không được viện dẫn thời hiệu để từ chối trả. Đã tự nguyện trả rồi thì cũng không được đòi lại. Giấy mượn và bảo lãnh xem mục 18 của phần này（giấy mượn và bảo lãnh）. Giai đoạn bị thi hành án xem mục 20 của phần này（bị khởi kiện, bị thi hành án thì báo cáo tài sản thật）. Tiền làm thêm giờ và tiền nghỉ phép năm chưa dùng thì đi theo bộ hoà giải lao động, xem mục 1 và 2 ở phần 19. + ### 20. Bị khởi kiện, bị thi hành án thì báo cáo tài sản thật, trả được bao nhiêu thì trả bấy nhiêu, đừng chuyển nhà và tiền cho người thân hoặc công ty - Chi phí: Không tốn tiền. Báo cáo tài sản theo yêu cầu của tòa: hiện có gì, và trong vòng một năm trước ngày nhận thông báo thi hành án có gì. Trả được thì trả trước. Không trả nổi thì thương lượng hoà giải, thương lượng trả góp với người xin thi hành án, người đó chính là bên đòi nợ. @@ -182,7 +169,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2002). Giải thích về Điều ba trăm mười ba của Bộ luật Trị phạt và Cải cách Dân chủ nước Cộng hòa Nhân dân Trung Hoa, 关于《中华人民共和国刑法》第三百一十三条的解释. ；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2024). Giải thích một số vấn đề về áp dụng pháp luật khi xét các vụ án hình sự không thi hành án, quyết định（法释〔2024〕13 号，第三、四、五、六、八、十一、十二条）. Trang công bố của Tòa án nhân dân tối cao, đã đối chiếu toàn văn với bản đăng lại của Nhà tù thành phố Quảng Châu, tỉnh Quảng Đông. 、；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). 刑法修正案（九）（第三十九条，修改刑法第三百一十三条）. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第五百三十八、五百三十九、五百四十条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 修正). Pháp luật Tố tụng Dân sự, 民事诉讼法（第一百一十四、一百一十八、二百五十二条）. 贵州省市场监管局转载： - Ghi chú: Câu 「ghi tên nhà vào công ty làm nhà ở cho nhân viên, chuyển hết tiền cho người thân」 đang lưu truyền trên mạng không phải lỗ hổng, mà đúng là trường hợp cho không và chuyển nhượng giả mà giải thích của tòa nêu đích danh. Người thân và công ty giúp bạn thao tác bị tính là đồng bọn, tài sản phải truy thu thì vẫn bị truy thu. Ngược lại, thật sự không có tiền thì không cấu thành tội này. Để biết mình có phải 「có tiền mà không trả」 không, trước hết phải trừ khoản sinh hoạt cần thiết cho bạn và gia đình bạn phải nuôi. Trước khi viện kiệm sát khởi tố, bạn trả lại một phần hoặc toàn bộ mà tình tiết lại nhẹ thì có thể không khởi tố. Trả trước khi có bản án sơ thẩm thì có thể được giảm nhẹ, thậm chí miễn phạt. Bị đưa vào danh sách, bị hạn chế tiêu dùng rồi làm sao, xem mục 21 của phần này（bị hạn chế tiêu dùng hoặc bị đưa vào danh sách không thi hành án）. Thi hành xong thì khôi phục thế nào, xem mục 19 ở phần 7（sau khi lỡ bị đưa vào danh sách thì làm lại thế nào）. - ### 21. Bị hạn chế tiêu dùng hoặc bị đưa vào danh sách không thi hành án, trước hết xem mình bị đưa vào theo điều khoản nào, có thể sửa thì xin sửa - Chi phí: Không tốn tiền. Đến tòa thi hành án hỏi rõ mình bị đưa vào theo khoản nào. Chuẩn bị chứng tứ đã trả tiền, đã làm theo bản án（chứng tứ thi hành）, hoặc chuẩn bị tài liệu bảo lãnh. Món chi cần cho sinh hoạt và kinh doanh thì phải xin từng lần một. Khó là phải tự đi nhiều chuyến, xin nhiều lần. @@ -191,7 +177,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Tòa án nhân dân tối cao (最高人民法院) (2017 修正). Một số quy định về công bố thông tin về người bị thi hành án không thi hành án, 关于公布失信被执行人名单信息的若干规定（第一、二、三、四、十条）. ；Tòa án nhân dân tối cao (最高人民法院) (2015 修正). Một số quy định về hạn chế tiêu dùng cao của người bị thi hành án và các tiêu dùng liên quan, 关于限制被执行人高消费及有关消费的若干规定（法释〔2015〕17 号，第三、八、九条）：「（九）乘坐G字头动车组列车全部座位、其他动车组列车一等以上座位等其他非生活和工作必需的消费行为」. - Ghi chú: Hạn chi tiêu là hạn tiêu dùng cao. Quy định này ra đời năm 2010, năm 2015 được sửa bằng 法释〔2015〕17 号, tên đổi thành 《……及有关消费的若干规定》, mục này viết theo văn bản hiện hành sau sửa đổi. Lần sửa đổi năm 2015 có ba điểm đáng lưu ý. Một là thêm loại thứ chín, mọi chỗ ngồi trên tàu cao tốc đều tính. Hai là bỏ lớp giới hạn 「bằng tài sản của mình」, trước đây viết là 「không được có các hành vi sau trả phí bằng tài sản của mình」. Ba là đơn vị bị hạn chế tiêu dùng thì người thực sự kiểm soát cũng bị hạn chế, nhưng tiêu dùng tư, dùng tiền cá nhân thì có thể xin tòa thi hành án. Danh sách không thi hành án và hạn chế tiêu dùng là hai việc khác nhau, có thể bạn chỉ dính vào một cái. Cả hai cũng không có nghĩa là lịch sử tín dụng bị bẩn, bị xoá khỏi danh sách không sửa lịch sử tín dụng, xem mục 19 ở phần 7（sau khi lỡ bị đưa vào danh sách thì làm lại thế nào）. Vợ chồng, con cái dùng tiền mình tự kiếm được thì có bị hạn chế không, văn bản hiện hành không nói rõ, gặp thì hỏi tòa thi hành án. - ### 22. Mua sắm trên mạng, giao dịch đồ cũ bị lừa thì khiếu nại nền tảng trước, báo công an sau, rồi mới tính có đáng kiện hay không - Chi phí: Khiếu nại nền tảng và báo công an đều không tốn tiền. Khi khởi kiện, số tiền tranh chấp dưới 1 vạn tệ thì phí tiếp nhận 50 tệ; đi theo thủ tục giản dị, hoà giải hoặc rút đơn thì giảm thêm một nửa. Khởi kiện, xét xử, thi hành án mỗi bước ít nhất nửa ngày đến một ngày, khởi kiện trên mạng thì bớt được chuyến đi chạy về. Phải lưu đơn hàng, lịch sử trò chuyện, lịch sử chuyển tiền và phiếu vận chuyển. Khó là đừng bỏ cuộc giữa đường khi phải đi lại nhiều lần. @@ -200,7 +185,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). Giải thích một số vấn đề về áp dụng cụ thể pháp luật khi xét các vụ án lừa đảo, 关于办理诈骗刑事案件具体应用法律若干问题的解释（法释〔2011〕7 号，第一条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百六十六条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). Pháp luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订，第五十八条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 修正). Pháp luật Tố tụng Dân sự, 民事诉讼法（第十六、一百六十四、一百六十五、一百六十八条）. （bản đăng lại của Ủy ban Phát triển và Cải cách Thượng Hải）；Tòa án nhân dân tối cao (最高人民法院) (2022 修正). Giải thích về việc áp dụng Pháp luật Tố tụng Dân sự, 关于适用《民事诉讼法》的解释（第二十条）. ；Hội đồng Nhà nước (国务院) (2006). Quy định về thu phí tố tụng, 诉讼费用交纳办法（国务院令第 481 号，第十三、十五、十六条）. - Ghi chú: Đáng kiện hay không thì tính như sau: lấy mức lương theo giờ sau thuế của bạn nhân với số giờ phải bỏ ra, cộng thêm phí tiếp nhận của tòa. Khởi kiện, xét xử, có thể còn phải xin thi hành án, mỗi việc tính nửa ngày đến một ngày. Tính ra rồi so với số tiền bạn muốn đòi lại, rồi trừ bớt, ước lượng xem thắng kiện thì có thi hành được tiền thật không. Có kiện hay không, tự bạn quyết theo bảng tính đó. Sau khi pháp luật xử phạt hành chính về trật tự công cộng sửa đổi năm 2025 thì số điều đổi hết, trộm cắp và lừa đảo từ Điều bốn mươi chín của luật cũ sang Điều năm mươi tám. Sau khi pháp luật tố tụng dân sự sửa đổi năm 2023 thì thủ tục tụng thọp nằm ở Điều trăm sáu mươi lăm. Hai chỗ này đừng chép tài liệu cũ. 「Khiếu nại nền tảng trước」 là kinh nghiệm: nền tảng có ký quỹ và quy tắc xử phạt, xử lý nhanh nhất. 法释〔2011〕7 号 trên trang của viện kiệm sát không ghi số hiệu văn bản, số hiệu còn phải làm rõ với trang bản gốc của Tòa án nhân dân tối cao. - ### 23. Tặng tiền lớn khi yêu đương và trong hôn nhân, nghĩ kỹ trước khi chuyển, đã chuyển thì nguyên tắc không đòi lại được - Chi phí: Không tốn tiền. Trước khi chuyển khoản lớn thì nghĩ một câu: đó là vay hay tặng. Là vay thì viết giấy mượn, ghi chú 「vay」 trong phần chuyển tiền. Khó là mở lời nói câu đó. @@ -209,6 +193,7 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第六百五十八、六百六十三条）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). Một số quy định về áp dụng pháp luật khi xét các vụ án tranh chấp liên quan đến tiền cưới, 关于审理涉彩礼纠纷案件适用法律若干问题的规定（法释〔2024〕1 号，第三条）. - Ghi chú: 「Những số tiền đặc biệt như 520, 1314 thường được xác định là tặng」 là cách nói thường gặp trong thực tiễn, phần này không tìm được bản gốc chính thức của Tòa án nhân dân tối cao, nên không coi là kết luận. Món quà đã công chứng tại văn phòng công chứng thì không thể nói muốn rút lại là rút lại. Một khoản chuyển tiền là vay hay là tặng thì cuối cùng vẫn phải xem chứng cứ, một câu 「khoản này tôi cho bạn mượn trước」 trong lịch sử trò chuyện hữu ích hơn nhiều so với tranh cãi sau đó. + ### 24. Tiền cưới thì chuyển khoản và ghi chú, giữ lại lịch sử trò chuyện; cưới để đòi tiền thì đòi lại được - Chi phí: Không tốn tiền. Tiền cưới, ba món vàng, tiền đổi lời đều nên chuyển khoản và ghi rõ mục đích trong phần ghi chú. Chuyện chỉ nói miệng thì để lại dấu vết trong lịch sử trò chuyện, hoặc nhờ người nhà có mặt xác nhận. @@ -217,15 +202,14 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零四十二条）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). Một số quy định về áp dụng pháp luật khi xét các vụ án tranh chấp liên quan đến tiền cưới, 关于审理涉彩礼纠纷案件适用法律若干问题的规定（法释〔2024〕1 号，第二、五、六条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Giải thích (một) về việc áp dụng phần Gia đạo và Hôn nhân của Bộ luật Dân sự, 关于适用《民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第五条）. - Ghi chú: Tiền cưới có tính là 「số tiền quá cao」 không thì phải xem thu nhập khả dụng bình quân đầu người ở nơi bên đưa tiền và hoàn cảnh gia đình họ, cả nước không có một con số thống nhất. Phải trả lại bao nhiêu thì thẩm phán quyết từng vụ. Bên đưa tiền và bên nhận tiền áp dụng cùng một quy tắc, bên đưa tiền thì giữ bằng chứng. - -### 25. Tài sản trước khi cưới không phải sợ, việc thêm tên vợ chồng và tiền của cha mẹ thì viết rõ trước +### 25. Tài sản trước khi kết hôn thì không phải sợ, nhà cộng tên và tiền cha mẹ đóng thì viết rõ trước -- Chi phí: Tự viết thỏa thuận tài sản bằng văn bản thì không tốn tiền, ra công chứng thì phí riêng. Khi cha mẹ đóng tiền thì nhờ họ viết một dòng giải trình bằng văn bản. Cả việc này tốn mấy tiếng một lần. Khó là mở lời bàn chuyện tiền với người nhà. -- Nói thẳng: Tài sản trước khi cưới là tài sản cá nhân của bạn, không biến thành tài sản chung chỉ vì cưới lâu năm. Vợ chồng có thể bằng văn bản thỏa thuận tài sản trước và sau khi cưới thuộc về mỗi người. Cần để ý một điểm: nói sẽ thêm tên người kia vào nhà thì trước khi đăng ký sang tên vẫn rút lại được, đăng ký xong là của người kia. Khi ly hôn, bên giấu tiền, chuyển tiền, tiêu tán hoặc bịa nợ thì có thể được chia ít hoặc không được chia. -- Lợi ích: Tài sản trước khi cưới là tài sản cá nhân, không biến thành tài sản chung chỉ vì cưới lâu năm. Vợ chồng có thể bằng văn bản thỏa thuận: tài sản trước khi cưới và thu nhập sau khi cưới thuộc về từng người, thuộc về chung, hoặc một phần thuộc chung, thỏa thuận này ràng buộc cả hai bên. Cha mẹ đóng tiền mua nhà trước khi cưới thì coi là tặng cho con mình, trừ khi nói rõ là tặng cho cả hai người. Cha mẹ đóng tiền sau khi cưới thì xử lý theo thỏa thuận, không có thỏa thuận thì xử lý theo nguyên tắc tài sản chung. Nếu trước khi cưới hoặc trong hôn nhân có thỏa thuận tặng bất động sản của một bên cho bên kia, hoặc để thành của hai người cùng sở hữu thì bên tặng có thể rút lại trước khi đăng ký sang tên, đăng ký xong là của người kia. Khi ly hôn, bên giấu, chuyển, bán, tiêu tán tài sản chung, hoặc bịa nợ thì có thể được chia ít hoặc không được chia (toàn quốc) +- Chi phí: Tự viết một thỏa thuận tài sản bằng văn bản thì không tốn tiền, ra cơ quan công chứng thì phải trả phí riêng. Khi cha mẹ đóng tiền, nhờ họ viết một dòng giải trình bằng văn bản. Cả việc này tốn một lần vài giờ. Khó ở chỗ mở lời bàn chuyện tiền với người nhà. +- Nói thẳng: Tài sản trước khi kết hôn là của riêng bạn, không phải cưới nhau lâu năm thế nào thì nó thành tài sản chung. Vợ chồng có thể bằng văn bản thỏa thuận tài sản trước và sau khi kết hôn thuộc riêng mỗi bên. Có một điểm phải để ý: nói sẽ cộng tên đối phương vào nhà thì trước khi làm thủ tục đổi tên vẫn đổi ý được, đã đăng ký xong thì của đối phương. Khi ly hôn, bên giấu tiền, chuyển tài sản, tiêu tán hoặc bịa ra khoản nợ thì được chia ít hơn hoặc không được chia. +- Lợi ích: Tài sản có trước khi kết hôn là tài sản cá nhân, không phải cưới nhau lâu năm thế nào thì nó thành tài sản chung. Vợ chồng có thể bằng văn bản thỏa thuận: tài sản trước khi kết hôn và thu nhập sau khi kết hôn thuộc riêng mỗi bên, thuộc chung, hay một phần thuộc chung, thỏa thuận này ràng buộc cả hai bên. Trước khi kết hôn mà cha mẹ đóng tiền mua nhà thì coi là quà tặng cho con mình, trừ khi nói rõ là cho cả hai người. Sau khi kết hôn mà cha mẹ đóng tiền thì xử lý theo thỏa thuận, không có thỏa thuận thì theo nguyên tắc của tài sản chung. Nếu trước khi kết hôn hoặc trong hôn nhân thỏa thuận tặng bất động sản của một bên cho bên kia, hoặc để hai người cùng sở hữu, thì trước khi làm thủ tục đổi tên, bên tặng có quyền rút lại, đã đăng ký xong thì của bên kia. Khi ly hôn, bên giấu, chuyển, bán, tiêu tán tài sản chung hoặc bịa ra khoản nợ thì được chia ít hơn hoặc không được chia (toàn quốc) - Mức bằng chứng: A -- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零六十三、一千零六十五、一千零九十二条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). Giải thích (một) về việc áp dụng phần Gia đạo và Hôn nhân của Bộ luật Dân sự, 关于适用《民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第二十九、三十一、三十二条）. -- Ghi chú: Thỏa thuận tài sản có hiệu lực với cả nam lẫn nữ. Muốn dùng nó để chống lại nợ của chồng với người ngoài thì điều kiện là chủ nợ phải biết hai vợ chồng có thỏa thuận đó. Người đó không biết thì không ràng buộc được họ. Thỏa thuận không công chứng vẫn có hiệu lực, công chứng chỉ để sau này khỏi tranh cãi về chữ ký và ngày tháng. +- Nguồn: Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự (《民法典》)(第一千零六十三、一千零六十五、一千零九十二条). ；Tòa án nhân dân tối cao (最高人民法院) (2020). Giải thích về việc áp dụng Bộ luật Dân sự, phần Hôn nhân và Gia đình (một) (法释〔2020〕22 号，第二十九、三十一、三十二条). +- Ghi chú: Thỏa thuận tài sản có hiệu lực như nhau với cả hai bên. Muốn dùng nó để chống lại khoản nợ của người phối ngẫu với người ngoài thì điều kiện là chủ nợ biết bạn có thỏa thuận này. Họ không biết thì không ràng buộc được họ. Thỏa thuận không công chứng vẫn có hiệu lực, công chứng chỉ để sau này khỏi tranh cãi về chữ ký và ngày tháng. Cha mẹ đóng tiền mua nhà thì viết thế nào là cho vay hay cho tặng, xem mục 11 ở phần 10 (cha mẹ đóng tiền mua nhà). ### 26. Trước khi cưới xem tình trạng hôn nhân của đối phương, lấy cưới làm vỏ để đòi tiền thì truy cứu tội lừa đảo @@ -235,7 +219,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零五十一条）. ；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百五十八、二百六十六条）. - Ghi chú: 「Lừa cưới」 bản thân không phải một tội danh. Có theo tội lừa đảo hay không thì phải xem ngay từ đầu người kia đã có ý định chiếm đoạt tiền, đã bịa lời để lấy tiền, và còn phải để công an khởi kiện điều tra. Có đủ chuẩn 「số tiền lớn」 không thì số tiền do giải thích tư pháp quy định, các nơi có thể điều chỉnh, phần này chưa làm rõ. Người bị lừa cưới không phân nam nữ, có cả vụ bị lừa tiền cưới lẫn vụ bị lừa đồ theo vợ. - ### 27. Sửa chữa tại nhà, kiểm định xe, sửa sang nhà thì hỏi giá trước khi để ai đó làm, và đứng cạnh xem suốt quá trình - Chi phí: Không tốn tiền. Trước khi làm thì hỏi rõ cách tính tiền và sẽ thay những bộ phận nào, chụp ảnh báo giá lại, hoặc bảo bên kia gửi bằng văn bản. Sửa xong phải có chứng tứ sửa chữa và hoá đơn, giữ lại bộ phận cũ. Còn phải đứng cạnh xem một hai tiếng. Khó là giữ được sự chú ý suốt quá trình, không rời đi. @@ -244,7 +227,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: B - Nguồn: Bộ Thương mại (商务部) (2012). Quy định quản lý ngành dịch vụ sửa chữa thiết bị điện tử gia dụng, 家电维修服务业管理办法（商务部令 2012 年第 7 号，第五、七、九、十四条）. ；Tổng cục Quản lý thị trường quốc gia (市场监管总局) (2022). Quy định về niêm yết giá công khai và cấm gian lận giá, 明码标价和禁止价格欺诈规定（市场监管总局令第 56 号，第五、八、十九条）. ；Hội đồng Nhà nước (国务院) (2024). Quy định thực thi Luật Bảo vệ quyền lợi người tiêu dùng, 消费者权益保护法实施条例（国务院令第 778 号，第九、十条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2013 修正). Luật Bảo vệ quyền lợi người tiêu dùng, 消费者权益保护法（第八、十条）：「消费者有权根据商品或者服务的不同情况，要求经营者提供……服务的内容、规格、费用等有关情况」「有权拒绝经营者的强制交易行为」. （Cơ sở dữ liệu luật pháp quốc gia） - Ghi chú: Bị thu thêm thì gọi 12315 trước, việc xử phạt do cơ quan thương mại và quản lý thị trường thực hiện, bạn không phải tự mình ra tòa. Để xếp mức B là vì quy định dù là chính thức nhưng 「đứng cạnh xem tiết kiệm được bao nhiêu」 không có con số, chủ yếu dựa vào kinh nghiệm. Hỏi giá, giữ chứng tứ, giữ bộ phận cũ thì ai cũng nên làm như nhau, không cần nhìn mặt người. Quy định về thiết bị gia dụng chỉ áp dụng cho sửa chữa thiết bị gia dụng; kiểm định xe và sửa sang nhà thì áp dụng quy định niêm yết giá công khai và quy định thực thi. Luật Bảo vệ quyền lợi người tiêu dùng ở cấp trên cũng ghi hai quyền này. Điều tám nói bạn có quyền yêu cầu người bán cho biết nội dung, quy cách, chi phí của dịch vụ. Điều mười nói bạn có quyền từ chối giao dịch cưỡng ép. - ### 28. Đừng làm 「người đứng tên」, đừng đưa thẻ căn cước cho người khác để đăng ký công ty - Chi phí: Không tốn tiền. Có người nhờ bạn 「đứng tên thôi, không phải quản gì」 thì từ chối là được. @@ -253,7 +235,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: B - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Luật Công ty, 公司法（2023 年修订，第十、十一、一百八十、一百九十一条）. - Ghi chú: Nguyên văn Luật Công ty chỉ nói trách nhiệm thuộc về ai. Hậu quả thường gặp hơn của việc đứng tên là: khi công ty bị thi hành án, người đại diện theo pháp luật cũng bị hạn chế tiêu dùng, thuế và xử phạt hành chính cũng có thể kéo bạn vào. Những hậu quả đó đến từ pháp luật và quy chế thi hành án khác, phần này chưa đối chiếu nguyên văn nên xếp mức B. Đã đứng tên rồi thì Luật Công ty cho phép bạn từ chối, công ty phải chỉ định người đại diện theo pháp luật mới trong 30 ngày. - ### 29. Ra nước ngoài thì đừng mang đồ giúp người lạ, đừng nhận hộ bưu kiện không rõ nguồn gốc - Chi phí: Không tốn tiền. Từ chối là được. @@ -262,7 +243,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: B - Nguồn: Lực lượng lãnh sự Bộ Ngoại giao (外交部领事司) (2023). Những lời khuyên dành cho công dân Trung Quốc chuẩn bị ra nước ngoài, 对准备出国的中国公民有哪些建议. - Ghi chú: Trang nhắc nhở chính thức của Tổng cục Hải quan và Bộ Công an về việc này không mở được, mục này chỉ dẫn trang của Lực lượng lãnh sự Bộ Ngoại giao. Và nguyên văn chỉ nêu nguyên tắc, không nêu điều luật, nên xếp mức B. Nhận hộ bưu kiện trong nước cũng vậy, nhất là loại bảo bạn ứng tiền trước rồi hoàn lại sau. - ### 30. Nuôi chó thì phải dắt dây: không dắt dây mà xảy ra chuyện thì chịu trách nhiệm không lỗi (dù bạn có lỗi hay không cũng phải bồi thường), bồi thường đến cùng - Chi phí: Một sợi dây vài chục tệ. Khó là mỗi lần ra ngoài đều nhớ dắt dây. @@ -271,6 +251,7 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典. Toàn văn đăng lại bởi Viện kiểm sát nhân dân tối cao. 第一千二百四十五条：「饲养的动物造成他人损害的，动物饲养人或者管理人应当承担侵权责任；但是，能够证明损害是因被侵权人故意或者重大过失造成的，可以不承担或者减轻责任。」第一千二百四十六条：「违反管理规定，未对动物采取安全措施造成他人损害的，动物饲养人或者管理人应当承担侵权责任；但是，能够证明损害是因被侵权人故意造成的，可以减轻责任。」第一千二百四十七条：「禁止饲养的烈性犬等危险动物造成他人损害的，动物饲养人或者管理人应当承担侵权责任。」第一千二百四十九条：遗弃、逃逸的动物在此期间造成损害，由原饲养人或者管理人承担责任 - Ghi chú: Tiền bồi thường vết cắn của chó gồm tiền khám chữa, tiền công mất việc, vắc-xin dại và globulin miễn dịch, một lần trên vạn tệ là chuyện thường thấy. Vết thương nặng, hoặc trúng phải trẻ em thì số tiền còn cao hơn. Bỏ mặc cũng không thoát trách nhiệm: bị cắn trong thời gian bị bỏ rơi, đào thoi thì vẫn tính vào người nuôi trước đó. Sau khi bị cắn xử lý thế nào, xem phần 13. + ### 31. Đối phương uống tới mất ý thức, đã ngủ, hoặc có thể dưới 14 tuổi thì không quan hệ tình dục - Chi phí: Không tốn tiền. Cái giá phải trả là dừng lại ngay lúc đó. Khó là dừng được tại chỗ. @@ -279,7 +260,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: B - Nguồn: Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百三十六条）. ；Quốc hội (全国人大) (2020 年修正). Bộ luật Hình sự, 刑法（第二百三十七条）. （bản đăng lại và hợp nhất của Cục Quản lý giao thông đường bộ Công an Bắc Kinh）；Viện kiểm sát nhân dân quận Tây Thành, Bắc Kinh (北京市西城区人民检察院) (2018). 她们为什么不再沉默……. - Ghi chú: Xếp mức B là vì điều luật chỉ có bốn chữ 「phương tiện khác」. 「Có ép rượu rồi quan hệ tình dục vẫn là cưỡng ép」 lấy từ bài phổ biến pháp luật của viện kiểm sát, không phải giải thích tư pháp. Điều về tội cưỡng ép chỉ nói về phụ nữ. Nam giới bị xâm hại kiểu này thì truy cứu theo tội cưỡng ép sách tế, nên đối phương là nam hay nữ thì cũng không được đụng vào. Nếu bạn là bên bị hại thì làm gì trước, xem mục 42 ở phần 13（sau khi bị xâm hại tính dục）. Khi bạn tự uống tới mất trí nhớ thì hai rủi ro bị buộc tội và bị tống tiền cùng tồn tại, xem mục 32 của phần này（đối phương nắm nhược điểm đòi tiền thì báo công an）. - ### 32. Sau khi quan hệ tình dục, quay video trần, đối phương dọa gọi công an, gửi ảnh, báo với cơ quan của bạn để đòi tiền thì không đưa một đồng nào, không xoá một bản ghi nào, báo công an ngay - Chi phí: Không tốn tiền. Chụp màn hình tài khoản, lịch sử trò chuyện và mã nhận tiền của đối phương, mang theo điện thoại tới trạm công an. Khó là nhịn được ý nghĩ 「cho tiền để xong chuyện」. @@ -288,7 +268,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020 年修正). Bộ luật Hình sự, 刑法（第二百七十四条）. （bản đăng lại và hợp nhất của Cục Quản lý giao thông đường bộ Công an Bắc Kinh）；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2013). Giải thích một số vấn đề về áp dụng pháp luật khi xét các vụ án tống tiền, 关于办理敲诈勒索刑事案件适用法律若干问题的解释（法释〔2013〕10 号）. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零三十二、一千零三十三条）. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千一百九十五条）：「权利人有权通知网络服务提供者采取删除、屏蔽、断开链接等必要措施。通知应当包括构成侵权的初步证据及权利人的真实身份信息」「未及时采取必要措施的，对损害的扩大部分与该网络用户承担连带责任」. （Cơ sở dữ liệu luật pháp quốc gia）；Cục Công an tỉnh Phúc Kiến (福建省公安厅) (2024). Vì một tin nhắn 「约炮」 mà tôi gánh trên 23 vạn tệ nợ. ；Báo Kiểm sát (检察日报) (2018). Hai lần bị kết án không hối hận, đăng ảnh riêng tư để đe dọa người trên mạng（bản đăng lại trên trang Viện kiểm sát nhân dân quận Cố Lâu, thành phố Hà Châu）. ；Báo Kiểm sát (检察日报) (2022). Đàn ông dọa công khai ảnh trần để ép 10 phụ nữ quan hệ tình dục, sàng lọc dữ liệu lớn 「báo công an」！（trên trang Viện kiểm sát nhân dân tối cao）. - Ghi chú: Ba mức tiền do mỗi tỉnh ấn định trong khung mà giải thích tư pháp đưa ra. 2000 đến trên 5000 tệ tính là 「số tiền lớn」. 3 vạn đến trên 10 vạn tệ tính là 「số tiền rất lớn」. 30 vạn đến trên 50 vạn tệ tính là 「số tiền đặc biệt lớn」. Tỉnh của bạn rơi vào con số nào thì theo tiêu chuẩn địa phương. Tống tiền ba lần trở lên trong hai năm thì mỗi lần số tiền không lớn cũng vẫn tính là 「nhiều lần」. Đừng tự xoá lịch sử trò chuyện, xoá ảnh, huỷ tài khoản, chặn là xong, đó là tự xoá bằng chứng của chính mình, xem mục 5 của phần này（bị triệu tập thì mời luật sư trước, không xoá bản ghi）. Bị khởi kiện điều tra thật thì những bản ghi đó đồng thời là tài liệu để bạn tự chứng minh. Cũng đừng thử 「thương lượng chia nhiều khoản」. Tống tiền bằng quay video trần phần lớn do nhóm ở nước ngoài làm, họ vừa quay màn hình vừa dùng mã độc để lấy danh bạ của bạn. 「Gửi cho cả gia đình」 là câu đàm kỳ; sau khi báo công an vẫn khởi kiện được. Dọa bằng 「tôi sẽ nói với người nhà bạn là bạn thích giống tính」 để đòi tiền cũng là tống tiền. Ở thành phố Diêm Trung, tỉnh Giang Tô có người hẹn gặp qua ứng dụng hẹn hò người đồng giới rồi đòi tiền như vậy, hơn 14.500 tệ; người bị hại báo công an thì họ bị khởi kiện, trước đó họ đã bị phạt tù hai lần vì cùng thủ đoạn. Dùng ảnh trần ép người ta quan hệ tình dục thì truy cứu theo tội cưỡng ép. Ở huyện Tượng Sơn, tỉnh Chiêu có người lừa được ảnh trần của nhiều phụ nữ trên nền tảng xã hội, rồi dọa tung ảnh, xác minh có 10 người bị hại. Một người nhận được tin đe dọa đã báo công an, hai người chưa từng gặp mặt, viện kiểm sát vẫn xử lý là cưỡng ép chưa thành công, cuối cùng khởi tố theo tội cưỡng ép và tội cưỡng ép sách tế. Vậy nên đừng gửi ảnh trần cho người quen qua mạng. Ngược lại, nếu bạn là bên bị hại mà đi đòi bồi thường, số tiền lớn cũng không có nghĩa là tống tiền, xem mục 36 của phần này（bản thân là bên bị hại mà đi đòi bồi thường）. Ảnh thật sự bị đăng lên mạng thì có thể yêu cầu nền tảng xoá. Điều một nghìn một trăm chín mươi lăm của Bộ luật Dân sự quy định bạn có quyền yêu cầu nền tảng xoá, chặn, ngắt liên kết, trong đơn phải kèm chứng cứ sơ bộ về hành vi xâm phạm và thông tin nhận dạng thật của bạn. Nền tảng nhận thông báo mà xử lý không kịp thì phần thiệt hại mở rộng do nó và người đăng cùng chịu trách nhiệm liên đới. - ### 33. Bị người khác bịa sự thật báo cáo thì có thể yêu cầu truy cứu: đủ mức xử phạt hành chính thì bị giữ từ 5 ngày, đủ cấu thành tội thì phạt dưới 3 năm - Chi phí: Không tốn tiền. Đến cơ quan công an báo việc, nói rõ đối phương đã bịa sự thật. Khó là đưa ra được thứ chứng minh họ cố tình bịa. @@ -297,7 +276,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). Pháp luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订，第五十条）. （bản đăng lại toàn văn của Cục Công an Bắc Kinh）；Quốc hội (全国人大) (1997). Bộ luật Hình sự, 刑法（第二百四十三条）. - Ghi chú: Muốn truy cứu đối phương thì chỉ chứng minh 「tôi không làm」 chưa đủ, còn phải chứng minh họ 「bịa sự thật」 và 「cố ý vu khống」. Phần sau khó chứng minh hơn nhiều, không chắc truy cứu được. Cả nước không có số liệu công khai về việc khởi kiện và xử lý vu khống, cuốn sách này tra không ra, nên không viết tỷ lệ, cũng không bàn phân bố theo giới tính. Sau khi bị buộc tội thì phải đưa chứng cứ và tìm cách cứu vãn thế nào, xem mục 34 và 35 của phần này（không đủ chứng cứ thì phải tuyên vô tội, bồi thường nhà nước）. - ### 34. Không đủ chứng cứ thì vốn đã phải tuyên vô tội, lời khai bị ép phải bị loại; đã tuyên thì còn khiếu nại và xử lại - Chi phí: Đi bộ thủ tục thì không tốn tiền, nhưng khiếu nại và xét lại đều rất tốn thời gian. Thường phải thuê luật sư, phí luật sư vài nghìn tệ trở lên. Người khó khăn về kinh tế có thể xin trợ giúp pháp lý miễn phí, xem mục 5 của phần này（bị buộc tội thì mời luật sư trước）. Khó là chờ qua từng vòng mà không bỏ cuộc. @@ -306,6 +284,7 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). Pháp luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第五十五、五十六、二百、二百五十二、二百五十三条）. - Ghi chú: Đây là những đường ranh giới mà pháp luật viết ra, còn kết quả thì không được bảo đảm. Cuốn sách này không có số liệu chính thức đối chiếu được về tỷ lệ tuyên vô tội trong các vụ án hình sự ở Trung Quốc hay tỷ lệ khiếu nại được xét lại, nên không viết tỷ lệ, cũng không bình luận vụ cụ thể. Có ba việc bạn tự làm được. Một, sau lần hỏi cung đầu tiên thì ủy quyền cho luật sư, xem mục 5 của phần này（bị buộc tội thì mời luật sư trước）. Hai, đọc hết bản ghi từng chữ rồi hãy ký. Ba, đừng nhận tội việc mình chưa làm để 「ra tù sớm」, nhận tội nhận phạt thì chỉ được giảm nhẹ, không tránh được trách nhiệm. Đủ điều kiện luật định thì tòa phải xét xử lại. + ### 35. Bị giữ rồi vụ được rút, không khởi tố hoặc tuyên vô tội thì đi xin bồi thường nhà nước, tiền tính theo ngày - Chi phí: Không tốn tiền. Gửi đơn yêu cầu bồi thường đến cơ quan đã ra quyết định bắt giữ, bắt, khởi tố hoặc tuyên án. @@ -314,16 +293,14 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012 年修正). Luật Bồi thường nhà nước, 国家赔偿法（第十七、十九、三十三、三十五条）. （bản đăng lại của Cục Thống kê Quốc gia）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). Tiêu chuẩn mới nhất về tiền bồi thường xâm phạm tự do cá nhân của công dân. ；Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院) (2015). Giải thích một số vấn đề về áp dụng pháp luật khi xét các vụ án bồi thường hình sự, 关于办理刑事赔偿案件适用法律若干问题的解释（法释〔2015〕24 号，第七、八条）. ；Quốc hội (全国人大) (2018). Pháp luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第十六、一百七十七、一百八十一、二百八十四、二百九十条）. . Điều mười chín của Luật Bồi thường nhà nước và giải thích đó dẫn số điều của Pháp luật Tố tụng Hình sự năm 2012（Điều mười lăm, khoản hai Điều trăm bảy mươi ba, khoản hai Điều hai trăm bảy mươi ba, Điều hai trăm bảy mươi chín）, sau khi sửa đổi năm 2018 tương ứng là Điều mười sáu, khoản hai Điều trăm bảy mươi bảy, khoản hai Điều hai trăm tám mươi tư, Điều hai trăm chín mươi - Ghi chú: Tiêu chuẩn tính theo ngày này được cập nhật mỗi tháng 5 theo mức lương bình quân năm của đơn vị thành phố không tự trả do Cục Thống kê Quốc gia công bố. Khi dùng thì tra con số của năm đó, đừng áp thẳng con số trong mục này. Nhận được quyết định không khởi tố, trước hết xem căn cứ ghi là gì. Ghi 「không có sự thật phạm tội」 hoặc 「không đủ chứng cứ」 thì có thể xin bồi thường. Ghi 「tội nhẹ」, hoặc đã quá thời hiệu truy tố thì nguyên tắc những ngày bị giam không được bồi thường. Không đồng ý với việc không khởi tố tương đối thì sau khi nhận quyết định có thể khiếu nại với viện kiểm sát trong 7 ngày. Tiền bùi đắp tổn thương tâm do cơ quan xử án ấn định theo hậu quả, không có công thức thống nhất. Việc bồi thường thuộc về cơ quan đã ra quyết định giam bạn（công an, viện kiểm sát hoặc tòa án）, đi tìm cơ quan đó trước. Không đồng ý về cách xử lý của nó thì theo thủ tục của Luật Bồi thường nhà nước đi lên cấp trên. Cách tính thật thì xem quyết định trong vụ Quốc Lợi ở mục 36 của phần này（bản thân là bên bị hại mà đi đòi bồi thường）. Quốc Lợi bị giam 1826 ngày. Tiền bồi thường tính theo tiêu chuẩn của năm ra quyết định, không theo năm bị giam, dùng mức lương bình quân ngày của công nhân viên cả nước năm 2016 là 258.89 tệ. Phần hạn chế tự do cá nhân được bồi thường khoảng 473.000 tệ（472733.14 tệ）, cộng thêm tiền bùi đắp tổn thương tâm khoảng 165.000 tệ（165450 tệ）. Mười vạn tệ tiền bùi đắp tổn thương tâm mà bản thân xin được 8888 vạn tệ, phần vượt không được chấp nhận. Cũng nên biết những khoản nào không được bồi thường: quyết định đó xác định tiền công mất việc, bồi thường về sức khoẻ, tổn thất tài sản, phí luật sư, tiền sinh hoạt trong trại tạm giam đều không thuộc phạm vi bồi thường của nhà nước. - -### 36. Bản thân là bên bị hại mà đi đòi bồi thường: gọi 12315, khởi kiện hoặc thuê luật sư; đừng một mình đi gặp đối phương, đừng gộp 「cho tiền」 với 「tôi không đăng」 thành một câu +### 36. Nếu chính mình là người bị hại đi yêu cầu bồi thường, dùng 12315, khởi kiện hoặc thuê luật sư, đừng một mình đến hẹn của đối phương, đừng gộp「trả tiền」với「tôi không công kích」thành một câu -- Chi phí: Không tốn tiền. Gọi 12315, ra tòa khởi kiện, hoặc ủy thác luật sư đưa ra yêu cầu. Khi gặp nói chuyện thì ghi âm toàn bộ. Khó là nhịn được, không tự đi nói chuyện riêng với đối phương. -- Nói thẳng: Bản thân là bên bị hại mà đòi bồi thường bên xâm phạm thì có căn cứ pháp lý, không phải vì 「đòi nhiều」 là tự động thành tống tiền. Nhưng cái giá có thể rất lớn: Quốc Lợi đầu tiên bị tuyên năm năm tù vì tội tống tiền, sau đó mới được xử lại và tuyên vô tội. Vậy nên đi con đường có để lại dấu vết, đừng một mình tới nơi đối phương chỉ định để nói chuyện, đừng gộp 「cho tiền」 và 「tôi sẽ không đăng」 thành một điều kiện để nói ra. -- Lợi ích: Vụ án chính thức: sau khi năm 2008 cơ quan có liên quan công bố một số lô sữa bột 「施恩」 có chứa melamine, kết quả kiểm tra của con gái Quốc Lợi là 「双肾中央集合系统内可见数个点状强回声」. Quốc Lợi nhiều lần yêu cầu bồi thường từ nhà phân phối và công ty Thận Nhiệt, rồi đưa tin ra báo chí. Tháng 6 năm 2009 hai bên đạt thoả thuận, công ty Thận Nhiệt bồi thường 40 vạn tệ, Quốc Lợi viết văn bản xác nhận không còn truy cứu và từ bỏ yêu cầu bồi thường. Sau đó công ty báo công an, Quốc Lợi bị Tòa án nhân dân huyện Triều An tuyên năm năm tù vì tội tống tiền. Viện kiểm sát nhân dân tỉnh Quảng Đông đưa ra đề nghị xét lại. Đề nghị cho rằng 「chứng cứ hiện có không đủ để chứng minh Quốc Lợi có chủ ý chiếm đoạt」. Là người giám hộ hợp pháp, Quốc Lợi 「có quyền yêu cầu nhà sản xuất sữa bồi thường, hành vi đòi bồi thường của họ có căn cứ pháp lý」. Đề nghị còn viết, 「người tiêu dùng có quyền phê phán, chỉ trích qua các phương tiện truyền thông đại chúng hành vi làm tổn hại quyền lợi hợp pháp của họ」. Ngày 7 tháng 4 năm 2017, Tòa án nhân dân tỉnh Quảng Đông xét lại tuyên Quốc Lợi vô tội. Lý do là 「sự thật bản án xác định Quốc Lợi nhằm chiếm đoạt vô đoạt, dùng phương pháp đe dọa, uy hiếp để cưỡng ép chiếm đoạt tài sản thì không rõ ràng, chứng cứ không đủ, bản chất hành vi của Quốc Lợi không vượt quá phạm vi tranh chấp dân sự」. Quốc Lợi bị giam 1826 ngày, từ ngày 23 tháng 7 năm 2009 đến ngày 22 tháng 7 năm 2014 (Quảng Đông, xét lại năm 2017, tuyên vô tội) +- Chi phí: Không tốn tiền. Thông qua 12315, ra tòa khởi kiện, hoặc nhờ luật sư đưa ra. Khi gặp nói chuyện thì ghi âm suốt. Khó ở chỗ phải nhịn được, không tự đi nói chuyện riêng với đối phương. +- Nói thẳng: Chính mình là người bị hại mà yêu cầu bồi thường thì có cơ sở pháp lý, không tự động biến thành tống tiền chỉ vì「đòi giá cao」. Nhưng cái giá có thể rất lớn: Quả Lý trước bị kết án tội tống tiền năm năm, sau đó mới được xét xử lại tuyên vô tội. Vậy nên đi đường có để lại dấu vết, đừng một mình tới nơi đối phương chỉ định để nói, đừng gộp「trả tiền」với「tôi sẽ không công kích」thành một điều kiện nói ra. +- Lợi ích: Ca thực tế chính thức: năm 2008, sau khi cơ quan liên quan công bố một số lô sữa bột「施恩」có chứa melamine, kết quả xét nghiệm của con gái Quả Lý là「双肾中央集合系统内可见数个点状强回声」. Quả Lý nhiều lần yêu cầu bồi thường nhà phân phối và công ty Thắc Ân, đồng thời phơi bày trên báo chí. Tháng 6 năm 2009, hai bên đạt thoả thuận, công ty Thắc Ân bồi thường 400.000 tệ, Quả Lý lập văn bản nói không truy tố và từ bỏ yêu cầu bồi thường. Sau đó công ty báo cáo, Quả Lý bị Tòa án nhân dân huyện Triêu An kết án tội tống tiền, phạt tù năm năm. Viện kiểm sát nhân dân tỉnh Quảng Đông đưa ra kiến nghị xét xử lại. Kiến nghị cho rằng「现有证据不足以证实郭利主观上具有非法占有的目的」. Với tư cách người giám hộ hợp pháp, Quả Lý「有权利向奶粉的生产厂家索赔，其索赔行为有法律依据」. Kiến nghị còn viết「消费者对损害其合法权益的行为，有权通过大众传播媒介予以揭露、批评」. Ngày 7 tháng 4 năm 2017, Tòa án nhân dân tỉnh cao cấp Quả Đông xét xử lại tuyên Quả Lý vô tội. Lý do là「原审认定郭利以非法占有为目的，使用威胁、要挟的方法，强行索取财物行为的事实不清，证据不足，郭利的行为性质未超出民事纠纷的范畴」. Quả Lý bị giam từ ngày 23 tháng 7 năm 2009 đến ngày 22 tháng 7 năm 2014, tổng cộng 1826 ngày (Quảng Đông, năm 2017 xét xử lại thay đổi bản án) - Mức bằng chứng: B -- Nguồn: Viện kiểm sát nhân dân tỉnh Quảng Đông (广东省人民检察院) (2017). Đề nghị của Viện kiểm sát nhân dân tỉnh Quảng Đông đề nghị sửa án, tuyên Quốc Lợi vô tội. （bản đăng lại của Viện kiểm sát nhân dân thành phố Hà Nguyên）；Ủy ban Bồi thường của Tòa án nhân dân tỉnh Quảng Đông (广东省高级人民法院赔偿委员会) (2018). Vụ Quốc Lợi xin bồi thường nhà nước với lý do được xét lại tuyên vô tội, Tòa án nhân dân trung cấp thành phố Triều Châu, tỉnh Quảng Đông bồi thường（（2018）粤委赔 11 号）. -- Ghi chú: Xếp mức B là vì đây chỉ là một vụ, không phải tỷ lệ có thể khái quát. Nó đồng thời nói rõ hai điều: bản thân việc đòi bồi thường là hợp pháp, và bạn vẫn có thể bị công an lập vụ để điều tra. Cách đi rủi ro thấp nhất là giao việc đòi bồi thường cho luật sư, hoặc ra tòa khởi kiện trực tiếp, bản thân không tiếp xúc riêng với đối phương. Đặc biệt cần cẩn trọng khi đối phương hẹn bạn tới nơi của họ để 「nói chuyện để giải quyết」, việc báo công an và khởi kiện thường xảy ra tại nơi của đối phương. Đã ký thoả thuận hoà giải thì làm theo thoả thuận, ký xong lại mở một vụ đòi bồi thường lớn khác thì dễ bị hiểu là 「đe dọa」. Sau khi bị khởi kiện thì xử lý thế nào, xem mục 5, 34 và 35 của phần này（mời luật sư trước, không đủ chứng cứ thì phải tuyên vô tội, bồi thường nhà nước）. Trường hợp đối phương nắm nhược điểm đòi tiền, xem mục 32 của phần này（đối phương dọa gọi công an, gửi ảnh để đòi tiền）. - +- Nguồn: Viện kiểm sát nhân dân tỉnh Quảng Đông (2017). 广东省检察院再审建议对郭利案改判无罪. （Trên cổng Viện kiểm sát nhân dân tỉnh Quảng Đông）；Ủy ban Bồi thường Tòa án nhân dân tỉnh cao Quảng Đông (2018). 郭利以再审无罪为由申请、广东省潮州市中级人民法院赔偿的国家赔偿案件（（2018）粤委赔 11 号）. +- Ghi chú: Để ở mức B vì đây chỉ là một ca, không phải tỷ lệ có thể khái quát. Nó đồng thời nói rõ hai việc: yêu cầu bồi thường bản thân là hợp pháp, và bạn vẫn có thể bị công an điều tra như một vụ án. Cách làm rủi ro thấp nhất là giao việc yêu cầu bồi thường cho luật sư, hoặc trực tiếp ra tòa khởi kiện, bản thân không tiếp xúc riêng với đối phương. Cần hết sức cảnh giác khi đối phương hẹn bạn tới nơi của họ để「giao lỏng giải quyết」, vì báo cáo và khởi kiện thường xảy ra ở nơi đối phương. Đã ký thoả thuận hoà giải thì làm theo thoả thuận, ký xong lại mở một vụ đòi tiền lớn khác thì dễ bị hiểu là「đe dọa」. Sau khi bị khởi kiện xử lý thế nào, xem mục 5, 34, 35 của phần này (thuê luật sư trước, thiếu chứng cứ thì phải tuyên vô tội, bồi thường của nhà nước). Trường hợp đối phương cầm nhược điểm để tống tiền bạn, xem mục 32 của phần này (đối phương dọa báo cảnh sát, dọa gửi ảnh để đòi tiền). ### 37. Bị tấn công trên mạng: bật chống bảo vệ và lưu giữ bằng chứng trước, rồi chọn trong ba con đường nền tảng, lệnh cấm, báo công an - Chi phí: Không tốn tiền. Tại chỗ dành mười mấy phút để chống bảo vệ và lấy chứng cứ. Nếu đi tới bước xin lệnh cấm hoặc tự khởi kiện thì phải tới tòa, có thể còn phải thuê luật sư. Khó là khi bị chửi thì đi lưu giữ bằng chứng trước, không đi cãi lại. @@ -332,6 +309,7 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Tòa án nhân dân tối cao (最高人民法院)、Viện kiểm sát nhân dân tối cao (最高人民检察院)、Bộ Công an (公安部) (2023). Hướng dẫn về việc trị phạt theo pháp luật các hành vi vi phạm pháp phạm tội bạo lực trên mạng, 关于依法惩治网络暴力违法犯罪的指导意见（法发〔2023〕14 号, mục 4, 11, 12 và 15）. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第九百九十七条）. Toàn văn đăng lại bởi Viện kiểm sát nhân dân tối cao ；Văn phòng Quản lý thông tin Internet quốc gia (国家互联网信息办公室)、Bộ Công an (公安部)、Bộ Văn hóa và Du lịch (文化和旅游部)、Tổng cục Phát thanh Truyền hình Quốc gia (国家广播电视总局) (2024). Quy định quản lý thông tin bạo lực trên mạng, 网络暴力信息治理规定（第十五、二十三、二十五、二十六条, có hiệu lực từ ngày 1 tháng 8 năm 2024）. Công báo Hội đồng Nhà nước - Ghi chú: Xếp thứ tự theo 「nhanh và rẻ」: nền tảng nhanh nhất, rẻ nhất, lưu giữ bằng chứng và báo cáo trước. Tòa chậm nhất và đắt nhất, cuối cùng mới quyết định có kiện hay không. Cả ba con đường đều phải tự gánh chi phí trong quá trình. Lệnh cấm phải tự xin tòa. Tự khởi kiện là tự ra tòa đăng ký, tự lấy chứng cứ. Muốn viện kiểm sát tiếp tục, đổi sang tự khởi tố công thì phải trùng một trong năm tình huống nêu ở mục 12 của 法发〔2023〕14 号, không đủ thì chỉ có thể tự kiện. Có hai việc đừng làm. Một là không cãi lại, cãi lại sẽ biến bạn thành người bị xử phạt trong mục 16 của phần này（không chửi người, không bịa tin trên mạng）. Hai là không xoá nội dung trong tài khoản của bạn, đó là bằng chứng. Không chịu nổi thì gọi 12356, xem mục 25 ở phần 1. Yêu cầu nền tảng xoá thông tin cá nhân của bạn, xem mục 8 ở phần 14. Vì giúp người lạ mà bị quay, bị đăng lên mạng, xem [docs/遇到陌生人出事该不该停.md](../docs/遇到陌生人出事该不该停.md). + ### 38. Con đường 「mua bảo hiểm cho người ta trước rồi mới ra tay」 bị pháp luật chặn kín từ đầu: không lấy được phân xu nào, tội cộng lại tội cố ý giết người và lừa bảo hiểm - Chi phí: Không tốn tiền. @@ -340,7 +318,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015 修正). Luật Bảo hiểm của nước Cộng hòa Nhân dân Trung Hoa, 中华人民共和国保险法第二十七、三十四、三十九、四十三条. ；Quốc hội (全国人大) (2020). Bộ luật Dân sự của nước Cộng hòa Nhân dân Trung Hoa, 中华人民共和国民法典第一千一百二十五条. ；Ủy ban Thường vụ Quốc hội (全国人大常委会). Bộ luật Hình sự của nước Cộng hòa Nhân dân Trung Hoa, bản hợp nhất sau khi sửa đổi lần thứ mười một, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第一百九十八条、第二百三十二条. ；Báo Pháp luật (《法制日报》) (2020-01-13), đăng lại bởi Viện kiểm sát nhân dân tỉnh An Huy (安徽省人民检察院转载). Giết vợ lừa bảo hiểm lộ ra vụ án trong vụ án（vụ án điển hình về giám sát vụ án dân sự giả của Viện kiểm sát nhân dân tỉnh An Huy）. - Ghi chú: Người được lợi của mục này chính là bạn, mục đích là tránh tử hình hoặc chung thân. Số người chết và bản án trong vụ án lấy từ nguyên văn vụ án điển hình do cơ quan kiểm sát công bố, phần này không viết chi tiết nào ngoài các bước điều tra. Bịa chuyện tai nạn, thổi phóng thành tổn thất, các kiểu lừa bảo hiểm không liên quan đến mạng người vẫn là tội phạm, kể cả người làm chứng giúp thì cũng bị tính, xem mục 21 ở phần 9（bịa chuyện tai nạn, thổi phóng thành tổn thất để lừa bồi thường）. Mua bảo hiểm cho người nhà vốn là chuyện bình thường, cái cần ngăn là ý nghĩ 「mua bảo hiểm xong rồi mới ra tay」. Xử lý cơn thôi thúc muốn làm hại người nhà như một trường hợp cấp cứu, xem mục 14 và 15 của phần này（ý nghĩ tự nhiên xuất hiện trong mình, người bên cạnh nói lời đe dọa）. - ### 39. Báo công an thì lấy biên nhận tiếp nhận ngay tại chỗ, không khởi kiện thì phải có văn bản: trong 7 ngày xin xét lại, thêm 7 ngày xin xét lại lần hai, viện kiểm sát có thể yêu cầu công an khởi kiện - Chi phí: Không tốn tiền. Tại chỗ hỏi thêm một câu, xin được biên nhận tiếp nhận vụ án. Sau đó theo các mốc ngày nêu trên mà đi hỏi tiến độ. Khó là phải nhiều lần mở miệng hỏi. @@ -349,15 +326,16 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2018). Pháp luật Tố tụng Hình sự, 刑事诉讼法（2018 年修正，第一百一十二、一百一十三条）. ；Bộ Công an (公安部) (2020). Quyết định sửa đổi Quy định về trình tự xử lý vụ án hình sự của cơ quan công an（公安部令第 159 号）và toàn văn sau khi sửa. ；Bộ Công an (公安部) (2013/2019 修正). Quy định về trình tự xử lý vụ án hành chính của cơ quan công an（公安部令第 125 号, quyết định sửa đổi số 149 号）. 、；Hội đồng Nhà nước (国务院) (2011). Quy định về thanh tra cơ quan công an（国务院令第 603 号，第九条）. - Ghi chú: Người được lợi của mục này chính là bạn. Báo công an không nhất thiết sẽ được xử lý, nhưng có xử lý hay không, xử lý trong bao lâu thì đều có tiêu chuẩn và thời hạn bằng văn bản. Ghi nhớ ba thao tác theo thứ tự. Một, lấy biên nhận tiếp nhận ngay tại chỗ, ghi lại số vụ án và đơn vị tiếp nhận. Hai, hết thời hạn mà chưa có kết quả thì theo đúng thời hạn nêu trên mà đi hỏi tiến độ, hỏi 「vụ án đang ở bước nào」, đừng hỏi 「sao không ai xử lý tôi」. Ba, nhận được văn bản từ chối khởi kiện thì xem lý do trước, trong 7 ngày xin xét lại, thêm 7 ngày xin xét lại lần hai, đồng thời có thể đề nghị Viện kiểm sát nhân dân giám sát việc khởi kiện, ba con đường không mâu thuẫn nhau. Vụ án hành chính là kiểu đánh nhau, trộm nhỏ. Thời hạn giải quyết tính từ ngày tiếp nhận, không quá 30 ngày. Vụ án lớn, phức tạp thì được cơ quan công an cấp trên phê duyệt để kéo dài thêm 30 ngày. Thời gian giám định không tính vào đó. Thái độ phục vụ tệ thì khiếu nại qua thanh tra, vụ án vẫn được đuổi theo đúng trình tự khởi kiện. Nhặt được đồ rơi, tranh chấp hàng xóm, kiểu không phải vụ án hình sự, thì cơ quan công an xử theo vụ án hành chính hoặc hoà giải, thậm chí chỉ đăng ký ghi nhận, đừng hiểu là 「báo công an mà không ai xử lý」. Sau khi bị lừa đảo thì quy trình dừng chi trả thế nào, xem mục 2 của phần này（gọi 110 ngay và yêu cầu dừng chi trả）. Bị tấn công trên mạng xem mục 37 của phần này（bị tấn công trên mạng: lưu giữ bằng chứng trước rồi chọn cách）. - -### 40. Đừng tặng tiền, tặng thẻ cho người xử án, người thi hành: hối người thì bản thân cũng bị phạt, hối cán bộ giám sát, cán bộ thi hành hành chính, nhân viên tư pháp thì còn bị phạt nặng hơn +### 40. Đừng tặng tiền, tặng thẻ cho người xử lý vụ việc, người thi hành: hối hộ chính mình cũng bị xử, hối hộ cán bộ giám sát, thi hành, tư pháp còn bị phạt nặng hơn - Chi phí: Không tốn tiền. Chỉ là không làm. -- Nói thẳng: 「Một ít tiền để xong chuyện」 trong Bộ luật Hình sự là một tội độc lập, người bị bắt không chỉ là kẻ nhận tiền. Hối người thì phạt dưới 3 năm. Tình tiết nghiêm trọng thì phạt 3 đến 10 năm, đặc biệt nghiêm trọng thì phạt trên 10 năm hoặc chung thân. Hối cán bộ giám sát, cán bộ thi hành hành chính, nhân viên tư pháp vẫn là nhóm bị phạt nặng hơn theo văn bản. Bị ép mà đưa tiền, lại không nhận được lợi ích phi pháp thì không tính là hối người. -- Lợi ích: Điều ba trăm tám mươi chín của Bộ luật Hình sự định nghĩa hối người: 「为谋取不正当利益，给予国家工作人员以财物的，是行贿罪」, tức là để mong lợi ích phi pháp mà đưa của cải cho cán bộ, công chức nhà nước thì là tội hối người. Cùng điều còn viết: 「在经济往来中，违反国家规定，给予国家工作人员以财物，数额较大的，或者违反国家规定，给予国家工作人员以各种名义的回扣、手续费的，以行贿论处」, tức là trong quan hệ kinh tế mà vi phạm quy định của nhà nước, đưa cho cán bộ, công chức nhà nước tài sản có số tiền lớn, hoặc đưa hoa hồng, phí thủ dụch dưới danh nghĩa khác thì xử theo tội hối người. Điều đó cũng để một ngoại lệ: 「因被勒索给予国家工作人员以财物，没有获得不正当利益的，不是行贿」, tức là vì bị ép mà đưa tài sản cho cán bộ, công chức nhà nước mà không nhận được lợi ích phi pháp thì không phải hối người. Mức phạt ghi ở Điều ba trăm chín mươi của Bộ luật Hình sự（sửa bằng Sửa đổi thứ mười hai Bộ luật Hình sự, có hiệu lực từ ngày 1 tháng 3 năm 2024）. Nguyên văn là: 「对犯行贿罪的，处三年以下有期徒刑或者拘役，并处罚金；因行贿谋取不正当利益，情节严重的，或者使国家利益遭受重大损失的，处三年以上十年以下有期徒刑，并处罚金；情节特别严重的，或者使国家利益遭受特别重大损失的，处十年以上有期徒刑或者无期徒刑，并处罚金或者没收财产」. Có bảy tình huống phải phạt nặng hơn: 「（一）多次行贿或者向多人行贿的；（二）国家工作人员行贿的；（三）在国家重点工程、重大项目中行贿的；（四）为谋取职务、职级晋升、调整行贿的；（五）对监察、行政执法、司法工作人员行贿的；（六）在生态环境、财政金融、安全生产、食品药品、防灾救灾、社会保障、教育、医疗等领域行贿，实施违法犯罪活动的；（七）将违法所得用于行贿的」. Khoản cuối để lối giảm nhẹ khi chủ động khai báo: 「行贿人在被追诉前主动交待行贿行为的，可以从轻或者减轻处罚。其中，犯罪较轻的，对调查突破、侦破重大案件起关键作用的，或者有重大立功表现的，可以减轻或者免除处罚」(toàn quốc) +- Nói thẳng: 「Bỏ ít tiền cho xong việc」trong Bộ luật Hình sự là một tội độc lập, người bị bắt không chỉ có kẻ nhận tiền. Hối hộ bị xử dưới 3 năm. Tình tiết nghiêm trọng thì bị xử 3 đến 10 năm, đặc biệt nghiêm trọng thì bị xử trên 10 năm hoặc chung thân. Hối hộ cán bộ giám sát, thi hành hành chính, tư pháp thì vẫn là bậc phạt nặng hơn ghi rõ trong luật. Bị ép mà đưa tiền, lại không nhận được lợi ích phi pháp, thì không tính là hối hộ. +- Lợi ích: Điều 389 của Bộ luật Hình sự định nghĩa hối hộ:「为谋取不正当利益，给予国家工作人员以财物的，是行贿罪」. Cùng điều khoản đó còn viết:「在经济往来中，违反国家规定，给予国家工作人员以财物，数额较大的，或者违反国家规定，给予国家工作人员以各种名义的回扣、手续费的，以行贿论处」. Cũng để lại một ngoại lệ:「因被勒索给予国家工作人员以财物，没有获得不正当利益的，不是行贿」. Mức phạt ghi ở Điều 390 của Bộ luật Hình sự (sửa đổi theo Bản sửa đổi mười hai, có hiệu lực từ 1 tháng 3 năm 2024). Nguyên văn là:「对犯行贿罪的，处三年以下有期徒刑或者拘役，并处罚金；因行贿谋取不正当利益，情节严重的，或者使国家利益遭受重大损失的，处三年以上十年以下有期徒刑，并处罚金；情节特别严重的，或者使国家利益遭受特别重大损失的，处十年以上有期徒刑或者无期徒刑，并处罚金或者没收财产」. Có bảy tình tiết phải phạt nặng hơn:「（一）多次行贿或者向多人行贿的；（二）国家工作人员行贿的；（三）在国家重点工程、重大项目中行贿的；（四）为谋取职务、职级晋升、调整行贿的；（五）对监察、行政执法、司法工作人员行贿的；（六）在生态环境、财政金融、安全生产、食品药品、防灾救灾、社会保障、教育、医疗等领域行贿，实施违法犯罪活动的；（七）将违法所得用于行贿的」. Khoản cuối để lại lối miễn giảm khi chủ động khai:「行贿人在被追诉前主动交待行贿行为的，可以从轻或者减轻处罚。其中，犯罪较轻的，对调查突破、侦破重大案件起关键作用的，或者有重大立功表现的，可以减轻或者免除处罚」(toàn quốc) - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会). Bộ luật Hình sự của nước Cộng hòa Nhân dân Trung Hoa, bản hợp nhất sau khi sửa đổi lần thứ mười một, 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第三百八十九条. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Sửa đổi Bộ luật Hình sự của nước Cộng hòa Nhân dân Trung Hoa（十二）（điểm năm, sửa Điều ba trăm chín mươi của Bộ luật Hình sự; có hiệu lực từ ngày 1 tháng 3 năm 2024）. -- Ghi chú: Người được lợi của mục này chính là bạn. Đừng trông chờ hối một lần là né được phạt. Bên kia có đồng nghiệp, có sổ sách, có camera giám sát và thanh tra, thứ bạn đưa ra chính là bằng chứng khi họ bị điều tra, cũng là bằng chứng tội của bạn, hơn nữa loại hối người này là nhóm bị phạt nặng hơn theo văn bản. Thật sự gặp người đòi lợi ích thì đi theo con đường ở mục 39 của phần này（báo công an thì lấy biên nhận tiếp nhận ngay tại chỗ）: giữ lại biên nhận tiếp nhận và chứng từ, báo lên cơ quan thanh tra hoặc cơ quan kỷ luật đảng. Khoản ba Điều ba trăm tám mươi chín của Bộ luật Hình sự đặt điều kiện miễn trách là 「bị ép và không nhận được lợi ích phi pháp」, một khi bạn nhân cơ hội đó để lấy thứ vốn không lấy được thì điều kiện không còn. Phong bao đỏ trong bệnh viện thuộc bộ quy tắc khác, xem mục 12 ở phần 24. +- Nguồn: Ủy ban Thường vụ Quốc hội. 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第三百八十九条. ；Ủy ban Thường vụ Quốc hội (2023). 中华人民共和国刑法修正案（十二）（第五项，修改刑法第三百九十条；自 2024 年 3 月 1 日起施行）. +- Ghi chú: Mục này có lợi cho chính bạn. Đừng trông chờ hối hộ một lần là miễn phạt. Đối phương có đồng nghiệp, có sổ sách, có hình ảnh và giám sát, thứ bạn đưa ra là bằng chứng khi họ bị điều tra, đồng thời cũng là bằng chứng buộc tội chính bạn, mà loại hối hộ này vốn đã nằm trong bậc phạt nặng hơn. Thật sự gặp đối phương đòi lợi ích thì đi theo con đường ở mục 39 (gọi cảnh sát, nhận văn bản xác nhận tiếp nhận ngay tại chỗ): giữ lại văn bản xác nhận tiếp nhận và giấy tờ, phản ánh tới cơ quan giám sát hoặc cơ quan kỷ luật đảng. Điều 389 khoản 3 của Bộ luật Hình sự đưa ra điều kiện miễn là「bị ép mà không nhận được lợi ích phi pháp」, một khi bạn lợi dụng cơ hội này để lấy thứ vốn không lẽ ra phải có thì điều kiện đó không còn đúng. Hồi hào đưa trong bệnh viện thuộc bộ quy tắc khác, xem mục 11 ở phần 24. + + ### 41. Cuộc gọi và cuộc gặp có thể đổi thái độ thì bật ghi âm ngay: cuộc nói chuyện bạn tham gia thì không cần xin phép đối phương trước - Chi phí: Không tốn tiền. Bật ghi âm cuộc gọi trên điện thoại hoặc dùng ứng dụng ghi âm là được. Ghi từ đầu đến cuối, xong thì giữ nguyên tệp gốc, không cắt ghép, không ghi đè. @@ -366,7 +344,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 年修正, có hiệu lực từ ngày 1 tháng 1 năm 2024). Pháp luật Tố tụng Dân sự, 民事诉讼法（第六十六条）. Ủy ban Phát triển và Cải cách thành phố Thượng Hải đăng lại toàn văn ；Tòa án nhân dân tối cao (最高人民法院) (2022 年第二次修正). Giải thích về việc áp dụng Pháp luật Tố tụng Dân sự của nước Cộng hòa Nhân dân Trung Hoa, 关于适用《中华人民共和国民事诉讼法》的解释（第一百零六条）. ；Tòa án nhân dân tối cao (最高人民法院) (2019 年修正). Một số quy định về chứng cứ trong tố tụng dân sự, 关于民事诉讼证据的若干规定（第十四、十五、九十条）. ；Lưu Đan Ni (2023). Pháp luật trong phim 「Cô Ấy Đã Biến Mất」 (báo Tòa án nhân dân, đăng lại trên trang Tòa án nhân dân tối cao). - Ghi chú: Người được lợi là bạn và gia đình bạn. Trong tệp ghi âm nên để người, việc, số tiền và thời gian khớp nhau. Đừng chỉ hỏi 「khi nào bạn trả tôi」, có thể hỏi tự nhiên 「khoản 5 vạn tệ bạn cho tôi mượn năm ngoái, bạn định trả khi nào」. Tệp ghi âm đưa cho tòa, hội đồng trọng tài hoặc công an dùng thì được, đừng đăng lên mạng, công khai lời của người khác có thể dính vào tranh chấp quyền riêng tư và danh dự, xem mục 16 của phần này（không chửi người, không bịa tin trên mạng）. Có tệp ghi âm không có nghĩa là chắc thắng, tòa còn cân nhắc chứng cứ khác xem sự thật và mức độ liên quan. Tệp ghi âm chỉ giải quyết vấn đề có dùng được để chứng minh hay không, có thật sự kiện tụng, tốn bao lâu thì phải tính riêng. Tài liệu cần lưu trước khi nghỉ việc xem mục 8 ở phần 19（trước khi nghỉ việc thì lưu phiếu lương, bản chấm công）. Căn cứ của mục này là quy tắc tố tụng dân sự, cách dùng chứng cứ trong vụ án hình sự và hành chính có quy định khác. - ### 42. Tại hiện trường có chuyện thì chụp toàn cảnh trước, chụp quan hệ vị trí sau, cuối cùng chụp hư hại và vết thương, ảnh gốc và video gốc đừng xoá - Chi phí: Không tốn tiền. Dùng điện thoại quay liên tục vài chục giây tới vài phút. Trước hết bảo đảm an toàn cho bạn, không làm chậm việc cứu người và xử lý. @@ -375,7 +352,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: C - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 年修正, có hiệu lực từ ngày 1 tháng 1 năm 2024). Pháp luật Tố tụng Dân sự, 民事诉讼法（第六十六条）. Ủy ban Phát triển và Cải cách thành phố Thượng Hải đăng lại toàn văn ；Tòa án nhân dân tối cao (最高人民法院) (2019 年修正). Một số quy định về chứng cứ trong tố tụng dân sự, 关于民事诉讼证据的若干规定（第十四、十五条）. - Ghi chú: Người được lợi là bạn và gia đình bạn. Để xếp mức C là vì thứ tự chụp chỉ là kinh nghiệm, điều luật chỉ quy định ảnh, video có thể làm chứng cứ và phải nộp bản gốc. Có người bị thương, hoả hoạn, rò rỉ khí hoặc còn nguy hiểm khác thì cứu người, tránh đi, đừng để chuyện chụp ảnh làm chậm. Việc phải làm tại hiện trường tai nạn giao thông xem mục 1 của phần này（tai nạn giao thông thì dừng xe, cứu người, báo công an trước）. - ### 43. Bị bạo hành gia đình: trước hết báo công an để có biên bản ra hiện trường, rồi ra tòa xin lệnh bảo vệ an toàn cá nhân, không cần ly hôn trước, cũng không mất phí - Chi phí: Không tốn tiền. Xin lệnh bảo vệ an toàn cá nhân không phải nộp phí tố tụng, cũng không phải đặt bảo lãnh. Cái phải bỏ ra là thời gian báo công an, khám thương, viết đơn. Khó nhất là mở lời với người nhà. @@ -384,7 +360,6 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). Luật Chống bạo hành gia đình, 反家庭暴力法（第二、十三、十五、十六、十九、二十三、二十八、二十九、三十、三十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2022). Một số quy định về áp dụng pháp luật khi xét các vụ án lệnh bảo vệ an toàn cá nhân, 关于办理人身安全保护令案件适用法律若干问题的规定（法释〔2022〕17 号，第一、三、六、十、十二条）. ；Tòa án nhân dân tối cao (最高人民法院) (2016). Phúc trả về các vấn đề thủ tục về vụ án lệnh bảo vệ an toàn cá nhân（法释〔2016〕15 号，第一、二项）. Toàn văn đăng lại bởi Tòa án nhân dân huyện Bình Nguyên ；Stöckl H và cộng sự (2013). Mức độ phổ biến toàn cầu của giết người bạn đời thân mật: một tổng quan hệ thống kê. Lancet. ；Glass N và cộng sự (2008). Bóp cổ không gây tử vong là một yếu tố nguy cơ quan trọng với giết người phụ nữ. Journal of Emergency Medicine. - Ghi chú: Người được lợi chủ yếu là bạn và gia đình bạn. Tỷ lệ các vụ giết người là số liệu nước ngoài, nước ta không công bố thống kê cùng cách đo. Cần tích góp chứng cứ: biên bản ra hiện trường, văn bản cảnh cáo, kết quả khám thương, hồ sơ khám bệnh, bản cam kết do đối phương viết, lịch sử trò chuyện và tệp ghi âm đều dùng được. Tệp ghi âm xem mục 41 của phần này（ghi âm）. Xin lệnh bảo vệ không cần thuê luật sư, thường vài ngày là có kết quả. Muốn ly hôn hoặc muốn đòi bồi thường là một vụ kiện khác, bản án sơ thẩm theo thủ tục thông thường tốn ít nhất 6 tháng, phí luật sư tự bỏ ra. Người bị bạo hành gia đình có thể xin trợ giúp pháp lý, tòa cũng có thể thu phí tố tụng chậm, giảm hoặc miễn. Lệnh bảo vệ không có nghĩa là an toàn cá nhân đã được bảo đảm, trước mắt có nguy hiểm thì gọi 110 trước. Người bị ép hoặc hoảng sợ, không thể tự xin, thì người thân gần, cơ quan công an, hội liên hiệp phụ nữ, ủy ban nhân dân khu phố, ủy ban thôn có thể xin thay. Nếu người bị hại là bạn bè hoặc đồng nghiệp thì bạn có thể đưa họ đi báo công an, đi bệnh viện. Thấy trong nhà người lạ đang có người bị đánh thì gọi 110 là đủ. Đừng tự xông vào can ngăn, lý do xem mục 10 của phần này（xung đột thì báo công an trước）. - ### 44. Người nhà cờ bạc nợ nần, đừng vội trả thay: nợ cờ bạc pháp luật không bảo vệ, tiền vay để cờ bạc cũng không phải nợ chung của vợ chồng - Chi phí: Không tốn tiền. Khó là giữ được sự kiên nhẫn khi chủ nợ tìm đến cửa, người nhà khóc lóc cầu xin mà vẫn không rút tiền ra. Chủ nợ thật sự kiện thì bạn phải ra tòa. @@ -392,4 +367,4 @@ Phần này tính bằng tiền và tự do cá nhân, không quy đổi ra tu - Lợi ích: Thông báo năm 2017 của Tòa án nhân dân tối cao viết bốn điểm. Nợ phát sinh trong các hoạt động vi phạm pháp luật như cờ bạc, nghiện ma túy thì không được pháp luật bảo vệ. Chủ nợ biết hoặc phải biết người kia vay tiền để cờ bạc, nghiện ma túy mà vẫn cho vay thì không được pháp luật bảo vệ. Một bên vay tiền dưới tên mình rồi dùng vào việc vi phạm pháp luật, yêu cầu xử như nợ chung của vợ chồng thì tòa không chấp nhận. Chưa qua thủ tục xét xử thì không thể yêu cầu bên không vay chịu trách nhiệm dân sự (toàn quốc, từ năm 2017) - Mức bằng chứng: B - Nguồn: Tòa án nhân dân tối cao (最高人民法院) (2017). Thông báo về việc xử lý đúng quy củ các vụ án liên quan đến nợ vợ chồng, 关于依法妥善审理涉及夫妻债务案件有关问题的通知（法〔2017〕48 号，第二、四条）. -- Ghi chú: Để xếp mức B là vì đây là văn bản nội bộ của tòa, không có con số, chủ nợ biết hay không biết cũng phải xem chứng cứ. Thông báo này ban hành theo Luật Hôn nhân và gia đình thời đó, sau khi Bộ luật Dân sự có hiệu lực thì nhận diện nợ chung của vợ chồng thế nào, xem mục 12 ở phần 10（vợ chồng vay một khoản lớn）. Khoản vay mà chủ nợ không biết là để cờ bạc thì vẫn phải trả, chỉ là người vay tự trả. Muốn chứng minh số tiền đã đi vào cờ bạc thì cần có sao kê chuyển tiền, lịch sử trên nền tảng cờ bạc, hoặc hồ sơ xử lý của công an, nên giữ lại từ trước. Chủ nợ thật sự kiện thì bản án sơ thẩm theo thủ tục thông thường tốn ít nhất 6 tháng, thủ tục giản dị là 3 tháng. Phí luật sư mỗi người trả phần mình, bên thua không gánh. Chủ nợ tìm người vây khốn, giữ người thì xem mục 15 ở phần 9（đòi nợ không giữ người）, báo công an ngay. Nếu chính người đó đã cờ bạc tới mức muốn chết thì xem mục 37 ở phần 1（cờ bạc tới mức vay tiền rồi vẫn cờ thì gọi 12356 trước）. Người được lợi chủ yếu là vợ chồng và người thân trực hệ. \ No newline at end of file +- Ghi chú: Để xếp mức B là vì đây là văn bản nội bộ của tòa, không có con số, chủ nợ biết hay không biết cũng phải xem chứng cứ. Thông báo này ban hành theo Luật Hôn nhân và gia đình thời đó, sau khi Bộ luật Dân sự có hiệu lực thì nhận diện nợ chung của vợ chồng thế nào, xem mục 12 ở phần 10（vợ chồng vay một khoản lớn）. Khoản vay mà chủ nợ không biết là để cờ bạc thì vẫn phải trả, chỉ là người vay tự trả. Muốn chứng minh số tiền đã đi vào cờ bạc thì cần có sao kê chuyển tiền, lịch sử trên nền tảng cờ bạc, hoặc hồ sơ xử lý của công an, nên giữ lại từ trước. Chủ nợ thật sự kiện thì bản án sơ thẩm theo thủ tục thông thường tốn ít nhất 6 tháng, thủ tục giản dị là 3 tháng. Phí luật sư mỗi người trả phần mình, bên thua không gánh. Chủ nợ tìm người vây khốn, giữ người thì xem mục 15 ở phần 9（đòi nợ không giữ người）, báo công an ngay. Nếu chính người đó đã cờ bạc tới mức muốn chết thì xem mục 37 ở phần 1（cờ bạc tới mức vay tiền rồi vẫn cờ thì gọi 12356 trước）. Người được lợi chủ yếu là vợ chồng và người thân trực hệ. diff --git "a/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" "b/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" index 6ac65fa5..83616484 100644 --- "a/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" +++ "b/vi/book/09-\346\231\256\351\200\232\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\346\263\225\345\276\213\347\272\242\347\272\277.md" @@ -12,7 +12,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第二十九条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). 刑法修正案（九）（第三十二条，即刑法第二百九十一条之一第二款）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百九十一条之一）. （Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại） - Ghi chú: Biết là giả mà vẫn chuyển tiếp, với tự bịa ra thì công an dùng chung một điều luật. Câu「tôi chỉ chuyển lại thôi」không được miễn trừ. Ảnh chỉnh sửa và ảnh hiện trường giả do AI tạo ra cũng tính là bạn tự bịa. Bộ Công an (公安部) năm 2025 đã công bố nhiều đợt vụ án điển hình về tin đồn mạng liên quan tới thiên tai và nguy cơ. Lúc viết phần này trang của Bộ Công an không mở được, nên không dẫn vụ án cụ thể. - ### 2. Không đăng, không chuyển tiếp ảnh, biểu tượng cảm xúc và bình luận xúc phạm liệt sĩ, không đùa cợt với cờ đỏ và quốc ca - Chi phí: Không tốn tiền. Với nội dung liên quan tới liệt sĩ, cờ đỏ, quốc huy, quốc ca thì không bình luận, không sửa ảnh, không chuyển tiếp. Chỗ khó là khi người ta trong nhóm đùa tiếp, nhịn không nói theo một câu. @@ -21,7 +20,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第三十五条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 英雄烈士保护法（第二十二、二十六条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第三十五条，即刑法第二百九十九条之一）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百九十九条、第二百九十九条之一）. ；Tòa án nhân dân tối cao (最高人民法院) (2022). 涉英烈权益保护十大典型案例（案例二、案例三）. - Ghi chú: Điều này của Luật Xử phạt quản lý trật tự công cộng chỉ được thêm vào khi sửa đổi năm 2025. Trước đó, việc này chỉ có thể xử lý theo Bộ luật Hình sự, hoặc do viện kiểm sát khởi xướng vụ án dân sự vì lợi ích công cộng. Nhóm WeChat vài trăm người, trong mắt tòa án vẫn là nơi công cộng. - ### 3. Xem xong trên trang nước ngoài thì đóng lại, không chụp màn hình, không chuyển tiếp, không chuyển vào nhóm, cũng không nói theo ở phần bình luận - Chi phí: Không tốn tiền. Đóng trang là xong việc. Không lưu trữ, không chuyển tiếp, không chuyển sang nền tảng trong nước. Không dùng tài khoản trùng với danh tính của bạn để đăng nội dung chính trị lên nền tảng nước ngoài lâu dài. Cũng không nhận tiền bài và trợ cấp do tổ chức, đoàn thể, cá nhân nước ngoài cho. Chỗ khó là nhìn thấy nội dung gây khó chịu mà nhịn không chụp, không chia sẻ. @@ -30,7 +28,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (1997). 刑法（第四十五、五十五、五十六、一百零三、一百零五、一百零六、一百零七、一百一十一、一百一十三条，这几条未经修正案修改）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第十三条第二款、第七十一条第一项）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015). 刑法修正案（九）（新增第一百二十条之六）. 国家信访局转载全文 - Ghi chú: Các điều luật dẫn ở mục này phạt hành vi kêu gọi người khác, truyền ra ngoài, cung cấp cho người khác, chỉ xem thì không nằm trong đó. Tài liệu chủ nghĩa khủng bố, cực đoan là ngoại lệ mà điều luật ghi rõ, chỉ riêng việc giữ trong tay đã đủ để định tội (điều một trăm hai mươi, khoản sáu), nên loại này tuyệt đối không được lưu trữ. Nhìn thấy thì dừng lại tại đó: không chụp màn hình lưu, không chuyển tiếp, không bình luận, không rủ người khác cùng chuyển. Chuyển tiếp và nói theo ở phần bình luận, khi xử lý vụ án đều bị tính là phát tán, câu「tôi chỉ chuyển lại thôi」không được miễn trừ. Nội dung lấy từ nền tảng nước ngoài còn có thể bị phạt nặng hơn. Không chắc thật giả thì càng đừng chuyển theo, xem mục 1 của phần này (không chuyển tiếp tin không rõ thật giả). Công cụ vượt tường bị phạt thế nào, xem mục 11 ở phần 11. Chuyển nội dung loại này ra ngoài không mang lại lợi ích thực chất nào, cái giá lại là mức tù, tước quyền chính trị và tịch thu tài sản. Vụ án cùng loại công khai tra cứu được chỉ thấy ở báo chí nước ngoài và văn bản do gia đình công bố, ví dụ một blogger chuyên đăng bình luận chính trị trên nền tảng nước ngoài bị phán năm 2023 là 7 năm. Cuốn sách này không xin được bản án có thể đối chiếu qua kênh chính thức, nên chỉ viết theo điều luật, không kèm vụ án. - ### 4. Video khiêu dâm thì tự xem ở nhà, đừng gửi vào nhóm, đừng bán「tư liệu」, đừng lập nhóm - Chi phí: Không tốn tiền. Không chuyển tiếp, không lưu vào tệp trong nhóm, cũng đừng đi làm chủ nhóm hay quản trị viên của loại nhóm đó. @@ -39,7 +36,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第八十条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第三百六十三、三百六十四条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2010). 关于办理利用互联网、移动通讯终端、声讯台制作、复制、出版、贩卖、传播淫秽电子信息刑事案件具体应用法律若干问题的解释（二）（第三条）. - Ghi chú: Cả hai bộ luật này đều phạt hành vi「phát tán」, tự xem riêng không nằm trong điều khoản. Nhưng gửi vào nhóm một lần đã là phát tán cho cả nhóm.「情节严重」cần đến bao nhiêu vật thì quy định trong giải thích của tòa án, phần này chưa đối chiếu con số cụ thể. - ### 5. Việc「làm thêm」bảo bạn dùng thẻ của mình để nhận tiền, rút tiền, chuyển khoản thì không làm, dù được trả bao nhiêu hoa hồng - Chi phí: Không tốn tiền. Bỏ loại việc「trả tiền trong ngày, mang thẻ đi ngân hàng」. @@ -48,6 +44,7 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第三百一十二条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 检察机关依法惩治电信网络诈骗及其关联犯罪典型案例（案例九）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 依法惩治掩饰、隐瞒犯罪所得、犯罪所得收益犯罪典型案例（案例五）. - Ghi chú: Cho vay thẻ ngân hàng, giúp người khác「chạy tiền」gì hậu quả thế nào, xem mục 8 ở phần 8 (cho mượn thẻ ngân hàng). Mục này nói về việc tự rút tiền, chuyển khoản. Câu「tôi không biết tiền đó từ đâu ra」khi xét xử không nhất thiết được chấp nhận. Mức công vượt mức bất thường, bắt bạn quét mặt để phối hợp, giao dịch nhìn thế nào cũng thấy không bình thường, chừng đó là đủ để kết luận bạn thực ra biết. Kết quả nhẹ nhất cũng là bị giữ 10 ngày, cộng thêm một hồ sơ tội phạm. + ### 6. Có người rủ bạn làm「bọc vật liệu」để vay tiền, chia hoa hồng theo khoản vay, thì không làm việc nào cả - Chi phí: Không tốn tiền. Từ chối thẳng. Không cho mượn giấy tờ tùy thân và lịch sử tín dụng của chính mình. Không ký hợp đồng vay tiền chưa đọc, cũng không đi thay người khác ra ngân hàng ký trực tiếp. @@ -56,7 +53,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第十一条，修改刑法第一百七十五条之一）. ；Quốc hội Nhân dân (全国人大) (1997). 刑法（第八十八、一百九十三、二百八十条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部) (2022). 关于公安机关管辖的刑事案件立案追诉标准的规定（二）（第二十二、四十五条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). 骗贷流水线背后藏洗钱黑手（检察日报）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). 出境入境管理法（第十二条）. （Cục Quản lý Nhập cảnh Xuất cảnh quốc gia）；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 修正). 民事诉讼法（第二百六十六条）. （Cục Quản lý thị trường tỉnh Quý Châu đăng lại） - Ghi chú: Câu「lúc đó tôi định trả」không được chấp nhận trong vụ án đó. Cơ quan kiểm sát làm rõ những người này không có việc làm, không có tài sản cố định, vốn không thể trả nổi, tiền chia được lại tiêu hết, nên kết luận vốn dĩ không định trả. Người rút tiền hộ bọn còn có thể bị cộng thêm tội rửa tiền. Tiền vay nằm dưới tên ai, hợp đồng ai ký, người đó là người vay. Cho dù không cấu thành tội phạm thì khoản nợ này và dấu vết trên lịch sử tín dụng vẫn là của bạn. Cầm tiền rồi chạy ra nước ngoài không giải quyết được: sau khi khởi xướng mà trốn, dù bao nhiêu năm vẫn truy được. Người đã kết án mà chưa thi hành xong, đang bị coi là bị can hoặc bị cáo, đều không được xuất cảnh. Vụ án dân sự chưa kết thúc thì tòa án có thể không cho bạn xuất cảnh; người bị thi hành án không thi hành thì tòa án cũng hạn chế xuất cảnh. Rút tiền, chuyển khoản hộ người khác xem mục 5 của phần này (việc「làm thêm」bảo bạn dùng thẻ của mình nhận tiền); bị người khác mượn tên vay tiền thì phát hiện bằng cách nào, xem mục 9 ở phần 8. - ### 7. Nhặt được thẻ ngân hàng thì không rút tiền; nhặt được điện thoại, ví thì liên lạc chủ nhân hoặc giao cho công an, không giữ lại - Chi phí: Không tốn tiền, mất mấy phút. Gọi điện tìm chủ nhân, hoặc mang đồ đến đồn công an. @@ -65,7 +61,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 民法典（第三百一十四条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第一百九十六、二百七十条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2008). 关于拾得他人信用卡并在自动柜员机（ATM 机）上使用的行为如何定性问题的批复. - Ghi chú: Tội chiếm đoạt là「chỉ xử lý khi bị kiện」, phải do chủ nhân tự ra tòa khởi kiện, công an không chủ động xử lý. Nhưng dùng thẻ nhặt được để rút tiền thì khác, không cần chủ nhân kiện, công an khởi xướng trực tiếp.「Số tiền lớn」cụ thể là bao nhiêu do giải thích của tòa án quy định, phần này chưa đối chiếu. Sau khi giao nộp đồ cho cơ quan có liên quan, khi chủ nhân đến nhận thì phải trả phí giữ giữ (điều ba trăm mười bảy của Bộ luật Dân sự). - ### 8. Ưu đãi thì lấy bằng chính danh tính thật của bạn; không mua số ảo để đăng ký hàng loạt tài khoản nhỏ, không dựa lỗ hổng để lấy lặp - Chi phí: Không tốn tiền. Bỏ loại việc trong「nhóm nhiệm vụ háo mùa」. Thấy nền tảng có lỗ hổng thì đừng đụng vào. Chỗ khó là nhịn được cái thói coi ưu đãi là khoản thu nhập có thể nhân bản hàng loạt. @@ -74,6 +69,7 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百五十三条之一、第二百六十六条、第二百九十五条）. （Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại）；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理诈骗刑事案件具体应用法律若干问题的解释（法释〔2011〕7 号，第一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). 反电信网络诈骗法（第十四、三十一、三十八、四十二、四十四条）. ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第九百八十五条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 剑指电信网络诈骗｜不法分子盯上了电商经济的漏洞（检察日报）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 如此「薅羊毛」涉嫌诈骗（检察日报）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2024). 辽宁朝阳龙城区：办理系列电信网络诈骗案（法治日报）. - Ghi chú: Ranh giới không nằm ở chỗ「háo được bao nhiêu」mà ở chỗ có bịa đặt hay không. Dùng chính danh tính thật của bạn tham gia hoạt động, một tài khoản lấy một lần, mã ưu đãi dùng cho bản thân mình, đó là ưu đãi nền tảng cho bạn. Để lấy nhiều hơn mà đăng ký người dùng giả, bịa giao dịch, giấu việc mình đã hoàn tiền, là khiến đối phương tưởng là thật rồi mới giao đồ, bị xác định là lừa đảo, số tiền chỉ quyết định mức tù. Số tiền không đủ mức kết án vẫn có hậu quả: nền tảng có thể khoá tài khoản, thu hồi ưu đãi, còn có thể yêu cầu bạn hoàn lại theo nguyên tắc「tiền không được hưởng thì phải trả lại」. Tham gia lừa đảo viễn thông mà chưa tới mức phạm tội thì vẫn còn mức bị giữ và phạt tiền.「Trong nhóm có người dẫn, trả tiền theo đơn」không làm thay đổi bản chất của phần bạn tham gia. Dạy người khác cách làm, bán lại tài khoản và số điện thoại là những tội khác, nhiều tội sẽ cộng lại. Người hưởng lợi chính là bạn. Tự viết script, bán script xem mục 3 ở phần 11; bán thẻ, bán tài khoản xem mục 8 ở phần 8 (cho mượn thẻ ngân hàng) và mục 16 của phần này (không cho mượn giấy tờ tùy thân). + ### 9. Không ném bất cứ thứ gì ra ngoài từ ban công, cửa sổ; đầu thuốc lá, túi rác, chai rượu đều tính - Chi phí: Không tốn tiền, cũng không mất thời gian. Đầu thuốc lá, túi rác, chai rượu đều cho vào thùng rác trong nhà, không thả ra từ cửa sổ và ban công. @@ -82,7 +78,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第四十三条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第三十三条，即刑法第二百九十一条之二）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百九十一条之二）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). 依法惩治危害公共安全犯罪典型案例（案例一）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 专访最高法副院长李勇：让高空抛物行为人都能被处罚. - Ghi chú: Không trúng người vẫn có thể bị giữ, điều luật chỉ đòi hỏi「có nguy cơ」. Bộ luật Dân sự còn một quy định bồi thường dân sự riêng về việc ném đồ từ trên cao (điều một nghìn hai trăm năm mươi tư). Lúc viết phần này không mở được nguyên văn nên không dẫn. - ### 10. Không mua súng nhựa, súng bi, súng hơi trên mạng; không mua「súng đồ chơi」không rõ nguồn gốc - Chi phí: Không tốn tiền. Muốn bắn súng thì đến phòng bắn súng uy tín. @@ -91,7 +86,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第一百二十八条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2018). 关于涉以压缩气体为动力的枪支、气枪铅弹刑事案件定罪量刑问题的批复（法释〔2018〕8 号）. ；Tòa án nhân dân tối cao (最高人民法院) (2018). 《批复》的理解与适用. - Ghi chú: Chỗ mấu chốt khiến vụ án đó không khởi tố là xác định họ「không biết」. Người trưởng thành cố ý lên mạng mua súng bi, súng hơi, sau đó nói không biết đó là súng thì rất khó thành lập. Nách và dao quản lý thế nào, phần này chưa đối chiếu nguyên văn nên không viết. - ### 11. Mua máy bay không cân thì đăng ký tên thật trước; không bay ở sân bay, khu quân sự và không gian bị kiểm soát của thành phố; không phá giới hạn chiều cao - Chi phí: Đăng ký tên thật không tốn tiền, mất mấy phút là xong. Trước mỗi lần cất cánh, dùng kênh chính thức tra xem khu vực bạn định bay có phải không gian bị kiểm soát hay không. @@ -100,7 +94,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Hội đồng Bộ trưởng (国务院), Ủy ban Trung ương Quân đội (中央军委) (2023). 无人驾驶航空器飞行管理暂行条例（国令第 761 号，第十、十九、四十七、五十一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第四十六条）. ；Sở Công an tỉnh Hồ Bắc (湖北省公安厅) (2026). 公安部公布依法打击无人机「黑飞」违法犯罪八起典型案例（转载人民公安报）. - Ghi chú: Máy bay không cân và nhẹ bay ở không gian được phép bay thì không cần phê duyệt. Không cân là loại máy bay trọng lượng cất cánh dưới 0.25 ki lô gam. Không gian được phép bay là nơi ngoài không gian bị kiểm soát; còn khu vực quanh sân bay, cơ sở quân sự, cơ quan chính phủ thì phần lớn thuộc không gian bị kiểm soát. Trang nguyên văn của Bộ Công an không mở được lúc viết phần này, nên phần này dẫn trang Sở Công an tỉnh Hồ Bắc đăng lại. - ### 12. Không chụp lén người khác, không lắp camera trong phòng thuê, nhà nghỉ, phòng khách sạn - Chi phí: Không tốn tiền, cũng không mất thời gian. Không chụp lén người khác, cũng không lắp camera trong phòng thuê, nhà nghỉ, phòng khách sạn. @@ -109,7 +102,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第五十条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). 依法惩治非法生产、销售、使用窃听、窃照设备犯罪典型案例（案例二）. - Ghi chú: Hình chụp lén lấy ra bán hoặc phát tán thì xử lý theo tội khác, ví dụ sản xuất, buôn bán vật phẩm khiêu dâm để phi pháp, trong cùng loại vụ án có người bị phạt 10 năm. Lắp camera trong nhà mình không sao, lắp trong phòng người khác dùng thì sao. - ### 13. Chơi mậu thẩu, bài poker thì được; không ăn phần trăm, không làm cái, không tổ chức bàn thu tiền, không chơi cờ bạc trên mạng - Chi phí: Không tốn tiền. Người thân chơi giải trí số tiền nhỏ thì vẫn chơi. Không ăn phần trăm từ bàn, không làm cái, không tổ chức bàn thu tiền. Chỗ khó là khi có người rủ rê thì nhịn không đi nhận. @@ -118,6 +110,7 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: B - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第八十二条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第三百零三条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第三十六条，开设赌场刑期提高到五年以下、五年以上十年以下）. - Ghi chú: Bao nhiêu là「số tiền cược lớn」không có con số thống nhất toàn quốc, từng nơi tự quy định, nên mục này chỉ ở mức B. Ăn phần trăm từ bàn, thu tiền hộ nền tảng cờ bạc trực tuyến, khi xét xử thường bị xác định là mở sòng bạc. Nguyên văn giải thích của tòa án liên quan phần này chưa đối chiếu. Cũng không tìm được vụ án chính thức nào. + ### 14. Không ăn thịt hoang dã, không mua bán, không nuôi động vật được bảo vệ làm thú cưng - Chi phí: Không tốn tiền. Thú cưng chỉ nên mua giống có nguồn gốc hợp pháp, nhân giống nhân tạo. @@ -126,7 +119,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). 野生动物保护法（2022 年修订，第三十一、五十一、五十三条）. （Bộ Môi trường và Tự nhiên đăng lại）；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第三百四十一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第四十一条，即刑法第三百四十一条第三款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 关于办理破坏野生动物资源刑事案件适用法律若干问题的解释（法释〔2022〕12 号，第十三条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 鹦鹉小案 民生大案. - Ghi chú: Câu「thường không xử lý là tội phạm」chỉ nói tới nhóm nhân giống nhân tạo, nuôi làm thú cưng. Động vật bắt ngoài hoang dã, không rõ nguồn gốc thì không áp dụng. Loài nào được tính là động vật bảo vệ cấp quốc gia thì theo danh mục. Mua trước thì tra danh mục đã, đỡ hơn là sau khi có chuyện mới giải thích. - ### 15. Đòi nợ thì không giữ người, không giam người, không đuổi theo tới nhà ở - Chi phí: Không tốn tiền. Đối phương nợ tiền không trả thì ra tòa khởi kiện. Giấy nhận nợ viết thế nào, xem mục 18 ở phần 8 (giấy nhận nợ). Chỗ khó là khi đối phương chống trả thì nhịn không ra tay, không đi tới nhà họ. @@ -135,15 +127,14 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第四十七条）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百三十八条）. - Ghi chú: Câu「anh ta đúng là nợ tôi tiền」không có tác dụng trong điều này. Điều luật ghi riêng, vì đòi nợ mà giữ người, giam người cũng xử lý như giam giữ trái phép. Trên mạng tòa án của Trung Quốc có nhiều vụ án đều là giữ người để đòi một khoản nợ hợp pháp, bị phạt tù từ 7 tháng đến 1 năm. Lúc viết phần này trang đó không mở được nên không dẫn. - -### 16. Không cho mượn giấy tờ tùy thân, không dùng giấy tờ của người khác, cũng không dùng giấy tờ người khác để đăng ký, mở thẻ, mua vé +### 16. Không cho mượn thẻ căn cước, không dùng thẻ căn cước của người khác, cũng không lấy giấy tờ của người khác để đăng ký, mở thẻ, mua vé -- Chi phí: Không tốn tiền. Có người mượn thì từ chối là xong. Nếu phải đưa bản sao thì ghi rõ bản sao đó chỉ dùng cho việc nào lên nó. -- Nói thẳng: Cho mượn, cho thuê, chuyển nhượng giấy tờ tùy thân của mình cho người khác thì bị cảnh cáo và phạt tiền dưới 200 tệ. Mạo danh giấy tờ tùy thân của người khác, hoặc dùng giấy tờ lấy được bằng lừa đảo thì phạt tiền 200 đến 1000 tệ, hoặc bị giữ trong 10 ngày. Ở những nơi bắt buộc có danh tính thật như mua vé tàu, ở khách sạn mà dùng giấy tờ giả hoặc giấy tờ của người khác, nghiêm trọng thì bị phạt giam giữ hoặc kỉ luật. -- Lợi ích: Cho thuê, cho mượn, chuyển nhượng thẻ căn cưởc dân của mình cho người khác thì bị cảnh cáo và phạt tiền dưới 200 tệ. Mạo danh giấy tờ tùy thân của người khác, hoặc dùng giấy tờ tùy thân lấy được bằng lừa đảo thì phạt tiền 200 đến 1000 tệ, hoặc bị giữ dưới 10 ngày. Mua, bán, sử dụng giấy tờ tùy thân làm giả hoặc sửa đổi cũng thuộc mức này. Có những nơi theo quy định nhà nước bắt buộc phải xuất trình chứng minh danh tính. Ở những nơi đó mà dùng giấy tờ làm giả, sửa đổi hoặc lấy của người khác thì nghiêm trọng là phải kết án. Giấy tờ gồm thẻ căn cưởc, hộ chiếu, thẻ bảo hiểm xã hội, giấy phép lái xe. Hình phạt là giam giữ (nhốt ngắn hạn) hoặc kỉ luật (không giam nhưng bị hạn chế đi lại), còn phạt kèm hoặc chỉ phạt tiền, tức là vừa kết án vừa phạt tiền, cũng có thể chỉ phạt tiền (toàn quốc) +- Chi phí: Không tốn tiền. Có người xin mượn thì từ chối là được. Nếu phải đưa bản sao chụp thì ghi rõ trên bản sao là chỉ dùng cho việc gì. +- Nói thẳng: Cho thẻ căn cước của mình mượn, thuê hoặc chuyển nhượng cho người khác thì bị cảnh cáo và phạt tiền dưới 200 tệ. Mượn danh thẻ căn cước của người khác, hoặc dùng thẻ lấy được bằng cách lừa, thì phạt tiền 200 đến 1000 tệ, hoặc bị giữ 10 ngày. Mua vé tàu, ở khách sạn, những chỗ bắt buộc đúng tên như vậy mà dùng giấy tờ giả hoặc giấy tờ của người khác, nghiêm trọng thì bị kết án phạt tù giam hoặc kỉ luật. +- Lợi ích: Cho thuê, cho mượn hoặc chuyển nhượng thẻ căn cước dân của mình cho người khác thì bị cảnh cáo và phạt tiền dưới 200 tệ. Mượn danh thẻ căn cước của người khác, hoặc dùng thẻ căn cước lấy được bằng cách lừa, thì phạt tiền 200 đến 1000 tệ, hoặc bị giữ 10 ngày. Mua, bán, dùng thẻ căn cước giả hoặc đã sửa đổi thì cũng ở mức này. Có những chỗ mà quy định của nhà nước bắt buộc phải xuất trình giấy tờ chứng minh. Dùng giấy tờ giả, sửa đổi hoặc lấy trộm giấy tờ của người khác ở những chỗ đó, nghiêm trọng thì phải chịu hình phạt. Giấy tờ gồm thẻ căn cước, hộ chiếu, thẻ bảo hiểm xã hội, giấy phép lái xe. Hình phạt là phạt tù giam (giam ngắn hạn) hoặc kỉ luật (không giam nhưng bị hạn chế đi lại), kèm phạt tiền hoặc chỉ phạt tiền, kết án xong vẫn phạt tiền, có thể chỉ phạt tiền (toàn quốc) - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2011). 居民身份证法（2011 年修正，第十六、十七条）. （Cục Quản lý Giao thông và An toàn giao thông thành phố Bắc Kinh đăng lại）；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十条之一）. -- Ghi chú: Cho giấy tờ tùy thân ra, người ta có thể dùng nó để đăng ký công ty, mở thẻ ngân hàng. Hậu quả của hai việc đó xem mục 28 và 8 ở phần 8 (đừng làm người đại diện danh nghĩa, đừng cho mượn thẻ ngân hàng). Dùng giấy tờ của người khác để mua vé tàu, ở khách sạn, thuộc loại「hoạt động phải cung cấp chứng minh danh tính」mà điều luật nói đến. Mục này không tìm được vụ án chính thức nào. +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2011). 居民身份证法（2011 年修正，第十六、十七条）. （Cục Quản lý giao thông công an thành phố Bắc Kinh đăng lại）；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十条之一）. ；Hội đồng Nhà nước (国务院) (2002). 互联网上网服务营业场所管理条例（国令第 363 号，第二十三条）. +- Ghi chú: Thẻ căn cước cho mượn đi có thể bị dùng để đăng ký công ty, mở thẻ ngân hàng. Hậu quả của hai việc đó xem mục 28 và 8 ở phần 8 (đừng làm người đại diện pháp lý hộ danh, đừng cho ai mượn thẻ ngân hàng). Dùng thẻ căn cước của người khác để mua vé tàu, ở khách sạn thì thuộc「hoạt động có nghĩa vụ cung cấp giấy tờ chứng minh」mà điều luật nói. Mục này không tìm được vụ án chính thức nào. Vào quán internet dùng chung cũng phải dùng giấy tờ của mình. Quán internet phải kiểm tra thẻ căn cước khi đăng ký, nội dung đăng ký và lịch sử lên mạng lưu ít nhất 60 ngày, công an và cơ quan văn hoá lúc nào cũng yêu cầu được. Dùng thẻ của ai lên mạng thì trong khoảng thời gian đó, việc làm trên máy tính được ghi vào tên người đó. ### 17. Khi bị công an chặn hoặc đưa đi thì hợp tác, không đẩy, không đánh, không giật máy quay của người thi hành; đã uống rượu thì càng phải nhịn @@ -153,7 +144,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第六十一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 刑法修正案（十一）（第三十一条，即刑法第二百七十七条第五款）. ；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百七十七条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 关于办理袭警刑事案件适用法律若干问题的解释（高检发释字〔2025〕1 号，第一、八、十二条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 袭警罪入刑一月 检察机关批捕 405 人. - Ghi chú: Năm 2021 Viện kiểm sát nhân dân tối cao thông báo, trong tháng đầu tiên sau khi tội tấn công công an được định tội riêng đã phê chuẩn bắt 405 người. Trong 42 vụ ở tỉnh Giang Tô, 47.6% do gây rối sau khi uống rượu. Trong 16 vụ ở tỉnh Hồ Bắc có 12 vụ chỉ ở mức đấm đá, cắn xé. Ra tay với công an hỗ trỗ thì xử lý theo tội gây cản trở công vụ. Mục này không tìm được vụ án chính thức nào ghi rõ hoàn cảnh cá nhân. - ### 18. Người kia chưa đủ 14 tuổi thì không được quan hệ tình dục; câu「cô ấy đồng ý」không phải là lý do - Chi phí: Không tốn tiền. Không chắc tuổi đối phương thì không quan hệ tình dục. @@ -162,6 +152,7 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百三十六条）. ；Tòa án nhân dân tối cao (最高人民法院) (2015). 关于「与十四周岁以下幼女发生性关系是否要明知才构成犯罪」的答复. - Ghi chú: Nguyên văn ý kiến của Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao cùng Bộ Công an, Bộ Giáo dục về xử lý vụ án xâm hại tình dục người chưa thành niên, cùng trang trả lời năm 2003 của Tòa án nhân dân tối cao, đều không mở được lúc viết phần này. Vì vậy tiêu chuẩn xác định「biết」dẫn từ câu trả lời trên trang của Tòa án nhân dân tối cao. Cách nói trong điều luật là「chưa thành niên nữ」. Mục này không phân biệt người ra tay là nam hay nữ. + ### 19. Trước khi ra tay, tính rõ cái giá: bị giữ, phải bồi, kết án - Chi phí: Không tốn tiền, cũng không mất thời gian. Khó chỉ ở mấy giây trước lúc ra tay. @@ -170,7 +161,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，第九、五十一条）. ；Quốc hội Nhân dân (全国人大) (1997). 刑法（第二百三十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2022 修正). 关于审理人身损害赔偿案件适用法律若干问题的解释（法释〔2022〕14 号，第六至十二、十五条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部) (2023). 关于依法妥善办理轻伤害案件的指导意见. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 检察机关依法妥善办理轻伤害案件典型案例. - Ghi chú: Cách tránh xung đột, vì sao ra tay trước là thiệt, ranh giới tự vệ ở đâu, xem mục 10 ở phần 8 (xung đột và tự vệ). Cái giá của việc ra tay với người khác chỉ để trút giận, xem mục 11 đến 14 ở phần 8. Đánh nhau vì bất hòa dân sự, mức nhẹ hơn thì công an có thể hoà giải. Đánh thành thương tích nhẹ mà thương lượng xong với đối phương và đã trả thật thì có thể nhẹ hơn, thậm chí không khởi tố, nhưng đồng tiền bồi thường không được thiếu một đồng nào. Bị thương tới mức nào thì gọi là thương tích nhẹ do tiêu chuẩn định giá quy định, phần này chưa đối chiếu. Lương mất công, chi phí chăm sóc tính theo thu nhập thực tế và số ngày bị chậm, không có con số thống nhất. - ### 20. Sinh con ra không nuôi nổi thì lối thoát hợp pháp chỉ có một: đăng ký nhận nuôi ở cơ quan dân vụ; nhận tiền rồi đưa con cho người khác có thể bị kết án về tội bán trẻ em, bỏ trẻ không chăm thì là tội bỏ bê - Chi phí: Không tốn tiền. Đăng ký nhận nuôi ở cơ quan dân vụ cấp tỉnh trở lên thì phải nộp hồ sơ, còn phải chờ xét duyệt và đánh giá nuôi dưỡng, trước sau cũng tốn thời gian. @@ -179,7 +169,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (1997). 刑法（第二百四十、二百六十一条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部), Bộ Tư pháp (司法部) (2010). 关于依法惩治拐卖妇女儿童犯罪的意见（法发〔2010〕7 号，第 17 条）. Trang chính thức của Tòa án nhân dân tối cao ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第一千零九十四、一千一百零五条）. Chính quyền thành phố Diêm City đăng lại toàn văn phần nhận nuôi - Ghi chú: Chuyện có nên sinh con hay không không bàn ở đây. Sau khi con sinh ra, đi theo thứ tự này là tiện nhất. Bước một, sinh con ở bệnh viện, khám thai và nằm viện đẻ xem phần 27. Giấy chứng nhận sinh con là tiền đề của mọi thủ tục về sau. Bước hai, muốn cho con nuôi thì ra thẳng cơ quan dân vụ cấp tỉnh trở lên, không qua trung gian. Tiền đã đi qua tay là bản chất có thể đổi. Trong vòng một năm sau sinh, nếu xuất hiện「ý nghĩ tự làm hại mình hoặc hại đứa trẻ」thì xử lý như cấp cứu. Danh mục những lúc phải tới bệnh viện ngay trong thai kỳ và sau sinh xem mục 7 ở phần 27. Tái khám sau 42 ngày cũng chính là sàng lọc trầm cảm sau sinh, xem mục 16 ở phần 27. Có ý nghĩ tự sát thì gọi 12356, xem mục 25 ở phần 1. Mục này không kèm vụ án: các vụ án điển hình chính thức về bán con ruột, lượt này chưa đối chiếu từng vụ, nên chỉ dẫn nguyên văn điều luật và ý kiến. - ### 21. Không bịa tai nạn, không phóng đại thiệt hại để lừa bồi thường: đó là tội lừa đảo bảo hiểm, và người làm chứng, sửa xe, giám định giúp bạn cùng bị tính - Chi phí: Không tốn tiền, cũng không mất thời gian, không làm là xong. @@ -188,6 +177,7 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会). 中华人民共和国刑法（经刑法修正案十一修正的整合文本）第一百九十八条. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015 修正). 中华人民共和国保险法第二十七条. ；Cục Quản lý Tài chính ngân hàng quốc gia (国家金融监督管理总局) (2024). 关于印发《反保险欺诈工作办法》的通知. ；Viện kiểm sát nhân dân tỉnh An Huy đăng lại (安徽省人民检察院转载) 《法制日报》(2020-01-13). 杀妻骗保牵出案中案. - Ghi chú: Mấy việc dưới đây tưởng vô hại đều nằm trong năm trường hợp đó. Xảy ra tai nạn nhỏ thì tìm tiệm sửa xe「báo hơn một chút」, tính luôn vết thương vết hư cũ vào lần này, đâm nhẹ vào nhau rồi mới báo, không có chuyện gì mà bịa ra một chuyện. Bảo hiểm xe mua thế nào, xem mục 26 ở phần 5 (mua đủ bảo hiểm trách nhiệm bên thứ ba). Sau khi mua bảo hiểm cho người nhà mà cố ý gây ra cái chết cho họ, điều luật ghi rõ phải gộp nhiều tội, xem mục 38 ở phần 8 (mua bảo hiểm cho người nhà). Bảo hiểm y tế là bộ quy tắc khác, quẹt thẻ bảo hiểm y tế trống hoặc rút tiền từ tài khoản cá nhân của bảo hiểm y tế thì xử lý như lừa đảo, xem ghi chú của mục 13 ở phần 5 + ### 22. Không bán cơ quan của mình, cũng đừng đi tìm người hiến cho người khác: thận bán được hơn 2 vạn tệ, cùng một quả bán cho bệnh nhân tới 20 vạn tệ, tiền bị tịch thu lại còn phạt theo giao dịch 10 đến 20 lần - Chi phí: Không tốn tiền, cũng không tốn thời gian, chỉ cần đừng làm là được. @@ -196,7 +186,6 @@ Phần này tính hai thứ: tự do cá nhân của bạn, và tiền của b - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2023). Biện pháp hiến tặng và ghép bộ phận cơ thể người（国务院令第 767 号，第五、十、十一、三十七条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会). Bộ luật Hình sự nước Cộng hòa Nhân dân Trung Hoa（经刑法修正案十一修正的整合文本）第二百三十四条之一. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2012-04-09). Vụ「thiếu niên bán thận mua iPhone」đưa ra tòa. ；Trang web Tòa nhân dân trung cấp thành phố Lang Phong, Hà Bắc chuyển lại (2014-09-02). Phúc thẩm vụ mua bán thận lớn nhất cả nước, người tổ chức lấy 51 quả thận, lợi nhuận lên tới hàng chục triệu tệ. - Ghi chú: Người được lợi là chính bạn. Giao dịch bề ngoài là「một lần nhận hai vạn」, thực ra là đổi một cơ quan lấy một chiếc điện thoại. Kết luận giám định trong vụ thiếu niên bán thận là thương tích nặng, thương tật mức 3, nghỉ vài tháng không khỏi. Trong vụ đó, viện kiểm sát xác định 5 bị cáo vi phạm 刑法第二百三十四条第二款, truy tố về tội cố ý gây thương tích. Một trong các lý do là nhân viên y tế không kiểm tra「người hiến」có phải vị thành niên hay không trước khi cắt thận phải. Hôm nay chuyện đó tái diễn, người đi lôi kéo bị xử về tội tổ chức mua bán cơ quan người, còn việc lấy cơ quan của người chưa đủ mười tám tuổi vẫn bị xử về tội cố ý gây thương tích theo 第二百三十四条. Phẫu thuật của các môi giới đen làm trong nhà thuê hoặc nơi mượn, ngoài việc khớp bệnh thì không có xét nghiệm nào, cũng không có tái khám sau mồ, xảy ra biến chứng thì không tìm ai chịu trách nhiệm. Thực sự muốn hiến cho người thân, đi theo con đường của 第十一条: làm ở cơ sở y tế có tư cách ghép, qua hội đồng đạo đức duyệt, không thu tiền suốt quá trình. Hiến cơ quan sau khi chết là无偿, không nằm trong mục này. Cái giá cơ thể phải trả sau khi mất một quả thận, xem mục 35 ở phần 1 (quả thận còn lại làm việc thay hai quả). Đến mức nghèo không còn tiền mà phải tính chuyện bán cơ quan, trước tiên xem phần 7 (các kênh cứu trợ: bảo hiểm thất nghiệp, trợ cấp tạm thời, trợ cấp người có thu nhập thấp). Bẫy vay nóng mạng và cho vay「包装材料」, xem mục 6 của phần này. - ### 23. Đừng đi mại dâm: mức mặc định là giam 10 đến 15 ngày, không phải phạt chút tiền là xong - Chi phí: Không tốn tiền, cũng không tốn thời gian, chỉ cần đừng làm là được. diff --git "a/vi/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/vi/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" index 3d79d203..04fae0ed 100644 --- "a/vi/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" +++ "b/vi/book/10-\346\201\213\347\210\261\345\222\214\347\273\223\345\251\232\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -12,7 +12,6 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Joel S, Eastwick PW, Finkel EJ (2017). Is romantic desire predictable? Machine learning applied to initial romantic attraction. Psychological Science, 28(10), 1478-1489. ; Eastwick PW, Finkel EJ (2008). Sex differences in mate preferences revisited: do people know what they initially desire in a romantic partner? Journal of Personality and Social Psychology, 94(2), 245-264. - Ghi chú: Hai nghiên cứu chỉ nói「trước khi gặp thì không tính ra được」, không nói「cố thế nào cũng vô ích」. Có hợp nhau hay không thì gặp mới biết, cho nên bỏ thời gian quen thêm vài người, mỗi người gặp thật sự một lần, đáng hơn việc đặt cược hết vào một người. Đối tượng nghiên cứu đều là sinh viên đại học Âu Mỹ. Môi trường yêu đương ở Trung Quốc khác, áp lên mình thì phải giảm kỳ vọng. - ### 2. Dừng lại sau khi đối phương từ chối rõ ràng: cứ bám theo tiếp là vi phạm trật tự an ninh, không phải「chân thành」 - Chi phí: Không tốn tiền, cũng không tốn thời gian. Khó nhất là nhịn: không nhắn tin nữa, không đi đợi nữa. @@ -21,7 +20,6 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025 sửa đổi). Luật Xử phạt hành chính về trật tự an ninh công, 治安管理处罚法（第五十条第一款第五项、第二款）. （Viện kiểm sát nhân dân tối cao đăng lại） - Ghi chú:「Bám theo đến cùng rồi đối phương sẽ cảm động」không có bất kỳ nghiên cứu nào ủng hộ, nhưng lệnh giữ người và lệnh cấm tiếp cận thì ghi trắng đen trong văn bản. Hồ sơ để lại còn ảnh hưởng đến thi công chức, thi biên chế, kiểm tra chính trị, và cả một số chứng chỉ hành nghề của vài ngành. Đường ranh giữa theo đuổi một người và quấy rối một người, không nhìn bạn nghĩ gì, mà nhìn đối phương đã từng nói rõ từ chối hay chưa. - ### 3. Muốn biết đối phương có hứng thú hay không thì xem hành vi, đừng xem「tín hiệu」: ai chủ động, ai sắp xếp, ai theo sau - Chi phí: Không tốn tiền, cũng không tốn bao nhiêu thời gian. Khó nhất là đổi thói quen: đừng nghĩ đi nghĩ lại nét mặt và giọng điệu của đối phương, thay vào đó ghi lại sự thật. @@ -30,16 +28,14 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: B - Nguồn: Hall JA, Xing C, Brooks S (2015). Accurately detecting flirting: error management theory, the traditional sexual script, and flirting base rate. Communication Research, 42(7), 939-958. ; Perilloux C, Kurzban R (2015). Do men overperceive women's sexual interest? Psychological Science, 26(1), 70-77. （quan điểm bị tranh cãi） - Ghi chú:Tranh cãi. Có nghiên cứu khác cho rằng không phải đàn ông đánh giá quá cao ý định tình dục của phụ nữ, mà là phụ nữ khi điền问卷 báo ý định của mình thấp đi. Rốt cuộc ai phán đúng hơn thì vẫn chưa có kết luận. Nhưng hai phe đều đồng ý một chuyện: suy ngược ra suy nghĩ của đối phương qua những tín hiệu nhỏ như nét mặt, giọng điệu thì độ chính xác rất thấp. Hành vi nhìn thấy được thì tin cậy hơn nhiều: đối phương có chủ động mở lời không, có sắp xếp lần gặp sau không, khi bạn không đẩy tiếp thì đối phương có theo sau không. - -### 4. Quan hệ tốt hay không chủ yếu do cảm nhận của chính bạn trong quan hệ đó, không phải do điều kiện của đối phương +### 4. Mối quan hệ có tốt hay không, chủ yếu xem cảm nhận của chính bạn, không phải xem điều kiện của đối phương -- Chi phí: Không tốn tiền, cũng không tốn bao nhiêu thời gian. Khó nhất là đổi góc nhìn: cất danh sách chọn bạn đời đi, thay vào đó xem cách hai người sống cùng nhau. -- Nói thẳng: Điều nói lên quan hệ tốt hay không rõ nhất là cảm nhận của chính bạn, không phải điều kiện của đối phương. Ví dụ bạn thấy đối phương nghiêm túc hay không, có coi trọng bạn hay không, hai người cãi nhau nhiều hay không. Cảm nhận của chính bạn hữu dụng hơn lời của bạn đời từ hai đến bốn lần. Nhưng quan hệ này về sau khá lên hay xấu đi thì hầu như bộ câu hỏi nào cũng không tính ra được. -- Lợi ích: Phân tích máy học trên 29 phòng thí nghiệm và 43 bộ dữ liệu cặp đôi theo dõi dài hạn, tức là ghi lại cùng một nhóm cặp đôi suốt quá trình. Những mục dự báo chất lượng quan hệ mạnh nhất đều đến từ cảm nhận của chính người trong cuộc đối với quan hệ này: cảm nhận được sự cam kết của đối phương, sự coi trọng, mức hài lòng tình dục, cảm nhận đối phương hài lòng, xung đột. Trong các mục liên quan đến cá nhân, mạnh nhất là mức hài lòng cuộc sống, cảm xúc tiêu cực, trầm cảm, gắn kết tránh né, gắn kết lo âu. Các mục liên quan đến quan hệ này lúc mới bắt đầu khảo sát giải thích được nhiều nhất 45% phương sai. Phương sai là độ khác biệt về chất lượng quan hệ giữa các cặp với nhau, 45% là phần nói rõ được bằng các mục đó. Đến lúc nghiên cứu kết thúc thì giải thích được nhiều nhất 18%. Và sức dự báo của「các mục tự báo」gấp từ hai đến bốn lần so với「các mục do bạn đời báo」. Ngoài cảm nhận quan hệ do mình báo ra, nếu thêm khác biệt cá nhân và báo cáo của bạn đời vào thì không thêm được sức dự báo nào +- Chi phí: Không tốn tiền, cũng không tốn nhiều thời gian. Khó ở chỗ phải đổi góc nhìn: cất danh sách tiêu chí chọn bạn đời đi, thay bằng việc quan sát quá trình sống chung. +- Nói thẳng: Thứ nói rõ mối quan hệ có tốt hay không nhất là cảm nhận của chính bạn, không phải điều kiện của đối phương. Ví dụ bạn thấy đối phương có nghiêm túc không, có trân trọng bạn không, hai bạn cãi nhau nhiều không. Cảm nhận của bạn có tác dụng mạnh hơn lời của người yêu từ hai đến bốn lần. Nhưng mối quan hệ này sau này tốt lên hay xấu đi thì không câu hỏi nào tính được. +- Lợi ích: Phân tích bằng học máy trên 29 phòng thí nghiệm, 43 tập dữ liệu ghép cặp theo chiều dọc, tức là ghi chép theo dõi cùng một nhóm bạn đời suốt quãng thời gian. Những yếu tố dự đoán tốt nhất chất lượng mối quan hệ đều đến từ cảm nhận của chính các bên về mối quan hệ đó: cảm giác đối phương có cam kết, cảm giác được trân trọng, mức hài lònh tình dục, cảm giác đối phương hài lòng, xung đột. Trong các yếu tố liên quan tới cá nhân thì mạnh nhất là sự hài lòng về cuộc sống, tâm trạng tiêu cực, trầm cảm, kiểu gắn bó né tránh, kiểu gắn bó lo âu. Các yếu tố về mối quan hệ này, vào lúc khảo sát bắt đầu, giải thích được tối đa 45% phương sai. Phương sai là chênh lệch chất lượng mối quan hệ giữa bạn đời này với bạn đời khác, 45% là phần trong đó giải thích được bằng các yếu tố đó. Tới khi nghiên cứu kết thúc thì tối đa chỉ giải thích được 18%. Hơn nữa, sức dự đoán của「những yếu tố do chính mình báo cáo」gấp từ hai đến bốn lần sức dự đoán của「những yếu tố do bạn đời báo cáo」. Ngoài cảm nhận do bản thân báo cáo về mối quan hệ, có thêm vào khác biệt cá nhân và báo cáo của bạn đời thì không thêm được sức dự đoán nào. - Mức bằng chứng: A - Nguồn: Joel S, Eastwick PW, Allison CJ, et al. (2020). Machine learning uncovers the most robust self-report predictors of relationship quality across 43 longitudinal couples studies. Proceedings of the National Academy of Sciences, 117(32), 19061-19071. -- Ghi chú: Chính nghiên cứu đó cũng thấy rằng quan hệ này trong thời gian nghiên cứu khá lên hay xấu đi thì hầu như không tính ra được bằng bất kỳ nhóm chỉ số tự đánh giá nào trong bộ câu hỏi. Cho nên để phán quan hệ hiện tại tốt hay không, hỏi cảm nhận của chính mình chính xác hơn việc đối chiếu danh sách điều kiện của đối phương, nhưng đừng mong dự báo được nó sẽ ra sao sau này. Mục「tình cảm đừng chỉ hỏi có hay không」của phần này nói đúng chuyện ấy, ở đây liệt kê cụ thể là những mục nào. - +- Ghi chú: Cùng nghiên cứu đó còn phát hiện, trong thời gian nghiên cứu mối quan hệ tốt lên hay xấu đi thì không nhóm chỉ số tự đánh giá nào trong bảng hỏi tính được. Nên muốn đánh giá mối quan hệ hiện tại có tốt không thì hỏi cảm nhận của chính mình chính xác hơn là đối chiếu danh sách điều kiện của đối phương, nhưng đừng trông chờ dự đoán được nó sẽ thế nào về sau. Phần tính mối quan hệ tốt xấu gắn với sức khoẻ thân thể, xem mục 17 của phần này (coi chất lượng mối quan hệ là một khoản sức khoẻ). ### 5. Muốn kéo gần khoảng cách thì thay nhau trả lời một loạt câu hỏi từ nông đến sâu, đừng học thoại - Chi phí: Không tốn tiền. Tốn 45 phút. @@ -48,7 +44,6 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: B - Nguồn: Aron A, Melinat E, Aron EN, Vallone RD, Bator RJ (1997). The experimental generation of interpersonal closeness: a procedure and some preliminary findings. Personality and Social Psychology Bulletin, 23(4), 363-377. - Ghi chú: Mới chỉ làm thí nghiệm một lần, đối tượng là sinh viên đại học, chỉ đo mức thân thiết trong thời gian ngắn, không đo xem có phát triển thành quan hệ dài hạn hay không. Mức thân thiết trong thí nghiệm đến từ việc hai người thay nhau, nói dần vào chỗ sâu về tình hình thật của mình, không dựa vào thoại, chiêu thức hay biểu hiện đơn phía. Cho nên lúc trò chuyện có thể「tôi nói một chút thật, rồi hỏi lại một câu sâu bằng nhau」, hai bên thay phiên. - ### 6. Yêu xa không kém yêu gần, chỗ khó thật sự là ba tháng đầu sau khi dọn về sống cùng - Chi phí: Tốn tiền vé đường đi, còn tốn thời gian đường đi đường về. Số lần gặp ít hơn yêu gần. Khó nhất là chịu đựng lâu dài việc không gặp được mặt. @@ -66,7 +61,6 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Bộ Dân vụ (民政部) (2025). Báo cáo thống kê công trình dân vụ năm 2024（五（二）1 婚姻登记服务、注释 5）, bản trên trang và bản PDF. 、 - Ghi chú: Câu「tỷ lệ ly hôn trên 50%」hay gặp trên mạng là chia số cặp ly hôn trong năm cho số cặp kết hôn trong năm, 351.3 / 610.6 ≈ 57.5%. Cái này gọi là「tỷ lệ ly trên kết」. Hai số chia cho nhau không phải cùng một nhóm người, báo cáo thống kê cũng không có chỉ số này, nên không thể đọc thành「một nửa số hôn nhân sẽ ly hôn」. Muốn tính「xác suất cuối cùng một cuộc hôn nhân là ly hôn」 thì phải theo dõi cùng một nhóm người kết hôn từ đầu đến cuối, loại số liệu chính thức như vậy phần này không tìm thấy. - ### 8. Tính cả lợi ích sức khoẻ vào, nhưng phải hạ giá theo dữ liệu chỉ quan sát - Chi phí: Không tốn tiền, cũng không tốn thời gian. Chỉ cần nhớ một điều: các con số này nói là người đã kết hôn có tỷ lệ tử vong thấp hơn, đó là hai việc cùng xảy ra, không phải kết hôn làm bạn sống lâu hơn. @@ -75,7 +69,6 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A (Tranh cãi) - Nguồn: Manzoli L, et al. (2007). Marital status and mortality in the elderly: a systematic review and meta-analysis. Soc Sci Med 64:77–94. ；Roelfs DJ, et al. (2011). The rising relative risk of mortality for singles: meta-analysis and meta-regression. Am J Epidemiol 174(4):379–389. ；Wang Y, et al. (2020). Sex differences in the association between marital status and the risk of cardiovascular, cancer, and all-cause mortality: a systematic review and meta-analysis of 7,881,040 individuals. Glob Health Res Policy 5:4. - Ghi chú:Tranh cãi ở chỗ người kết hôn vốn đã được tuyển chọn trước: người khoẻ, thu nhập cao, giao tiếp nhiều thì vừa dễ kết hôn hơn, vừa dễ sống lâu hơn. Cả ba phân tích tổng hợp đều dùng dữ liệu quan sát trên nhóm dân cư có sẵn, chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. Chính tác giả cũng nói có thiên lệch công bố hoặc hạn chế về phương pháp. Roelfs còn thấy nghiên cứu càng làm nghiêm ngặt thì chênh lệch tính ra càng nhỏ. Các con số này chủ yếu lấy từ dân cư châu Âu và châu Mỹ, áp vào trong nước chỉ nên coi là đơn vị cỡ. - ### 9. Bảng tính thời gian thì tính theo「việc không được trả công」, nói rõ phân công rồi hãy đi đăng ký - Chi phí: Không tốn tiền. Trước khi cưới dành một hai tiếng, nói rõ việc nhà, chăm sóc người lớn tuổi trẻ nhỏ, ai phải nhường việc, tức các việc không được trả công, thành phân công rõ ràng, sau đó làm theo phân công. Khó ở chỗ phải nói thẳng ra, đừng kéo dài không nói. @@ -84,7 +77,6 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Cục Thống kê Quốc gia (国家统计局) (2024). Hai báo cáo công bố của khảo sát toàn quốc về sử dụng thời gian lần thứ ba（第三次全国时间利用调查公报（第二号、第三号））. 、；Cục Thống kê Quốc gia (国家统计局) (2019). Báo cáo khảo sát toàn quốc về sử dụng thời gian năm 2018. - Ghi chú: Cả hai khảo sát chỉ chia nhóm theo giới tính, thành thị hay nông thôn, độ tuổi, không chia theo đã kết hôn hay chưa. Nên「sau khi kết hôn việc nhà tăng thêm bao nhiêu」không có số liệu chính thức, chỉ có thể dùng chênh lệch giới tính làm mức trần. Số liệu là bình quân cả nước, phân công của hai người bạn hoàn toàn có thể thỏa thuận thành một kiểu khác. - ### 10. Bảng tính tiền thì xem quy tắc mặc định của pháp luật trước, rồi mới quyết định có cần thỏa thuận bằng văn bản không - Chi phí: Không tốn tiền. Đọc mấy điều về tài sản trong phần hôn nhân và gia đình của Bộ luật Dân sự. Nếu cần thì hai người ký một thỏa thuận tài sản bằng văn bản, pháp luật chỉ yêu cầu viết ra giấy, không cần công chứng. Khó ở chỗ phải nói trước chuyện tiền. @@ -93,7 +85,6 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人民代表大会) (2020). Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（中华人民共和国民法典）（2021 年 1 月 1 日起施行，第一千零六十二、一千零六十三、一千零六十五、一千零八十八条）. Cơ sở dữ liệu văn pháp quốc gia ；toàn văn điều khoản xem ở bản Công báo của Tòa án nhân dân tối cao đăng lại - Ghi chú: Tiền cước, đám cưới, nhà cưới không có thống kê chính thức, mục này không viết số. Khoản bồi thường ở điều 1088 cho bao nhiêu thì trước hết hai người tự bàn, thỏa thuận không xong thì tòa án quyết, không có công thức cố định. Thỏa thuận tài sản bằng văn bản dùng được cho cả hai bên: bên thu nhập cao dùng được, bên làm nhiều việc nhà và chăm sóc không được trả công cũng dùng được. - ### 11. Bố mẹ góp tiền mua nhà, lúc chuyển khoản thì viết rõ là cho hay là vay - Chi phí: Không tốn tiền. Ngày góp tiền thì viết một trang giải trình, ghi rõ khoản tiền đó tặng cho ai hay cho vay ai, bố mẹ góp tiền và con cái đều ký tên. Nếu ghi là khoản vay thì để cả hai vợ chồng ký, phần ghi chú trong chuyển khoản cũng viết cùng câu đó. @@ -102,6 +93,7 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Tòa án nhân dân tối cao (最高人民法院) (2020). Giải thích (một) về việc áp dụng phần hôn nhân và gia đình của Bộ luật Dân sự（关于适用《民法典》婚姻家庭编的解释（一））（法释〔2020〕22 号，第二十九条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). Giải thích (hai) về việc áp dụng phần hôn nhân và gia đình của Bộ luật Dân sự（关于适用《民法典》婚姻家庭编的解释（二））（法释〔2025〕1 号，第八条）. ；Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第一千零六十二条第一款第四项）. - Ghi chú: Viết rõ thì cả hai bên đều dùng được. Bố mẹ bỏ tiền không cần trông chờ tới lúc ly hôn mới bù thêm một tờ giấy vay nữa để giữ phần vốn góp. Sau khi giải thích (hai) có hiệu lực, việc tiền do ai bỏ ra đã là căn cứ để chia nhà, tên đứng trên giấy chứng nhận không còn quyết định thuộc về ai. Bên nhận tiền góp cũng không phải chờ tới lúc ly hôn mới biết khoản tiền đó tính là gì. Muốn làm theo kiểu khoản vay thì làm thật như khoản vay: hai bên nói rõ là vay, thỏa thuận cách trả, chuyển khoản ghi rõ, cần thì để lại chứng từ trả nợ. Thiếu một mắt xích nào thì lên tòa phải giải thích. + ### 12. Khoản vay lớn của một bên vợ chồng, nếu bạn không ký và không xác nhận sau, không tự động thành nợ của bạn - Chi phí: Không tốn tiền. Đừng ký「người vay chung」「người bảo lãnh」trên giấy vay hay giấy bảo đảnh của vợ chồng bạn. Cũng đừng viết trong tin nhắn kiểu「tiền này chúng tôi sẽ trả」, câu đó có thể bị hiểu là bạn đã xác nhận khoản nợ đó sau. Khó ở chỗ phải nhịn không đáp lời khi bị đòi tiền. @@ -110,7 +102,6 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第一千零六十四条）. ；Tòa án nhân dân tối cao (最高人民法院) (2020, sửa đổi lần thứ hai). Quy định về một số vấn đề pháp lý khi xét các vụ án vay ngoài（关于审理民间借贷案件适用法律若干问题的规定）（第十五、十六条）. （Công báo của Tòa án nhân dân tối cao） - Ghi chú: Lúc ly hôn chia tài sản, bố mẹ bên kia bất ngờ đưa ra một hợp đồng vay chỉ có chữ ký của vợ chồng bạn thì phải xem thành hai bước. Bước một xem khoản vay đó có thật là đã xảy ra hay không. Bạn nói tiền chưa từng được đưa, mà lý lẽ lại nói nối, thì tòa án sẽ xét kết hợp nhiều yếu tố. Xem trước số tiền vay, tiền được đưa bằng cách nào, năng lực tài chính của hai bên; sau đó xem hai bên vẫn giao dịch ra sao, tài sản có biến động tương ứng không, lời khai của người làm chứng thế nào. Bên kia chỉ đưa ra được chứng từ chuyển khoản thì vẫn phải tự chứng minh khoản vay đó có thật. Bước hai, khoản vay đã thành lập thì mới xem có phải nợ chung không. Quy tắc chung về giấy vay và giấy bảo đảnh xem mục 18 ở phần 8. - ### 13. Đi đăng ký chỉ cần thẻ căn cước cộng một bản cam kết có chữ ký, không cần sổ hộ khẩu, không thu tiền, cấp giấy ngay tại chỗ - Chi phí: Không tốn tiền. Cả hai người đều phải có mặt. Thường xong trong vài chục phút. @@ -119,7 +110,6 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2025). Quy định đăng ký hôn nhân（婚姻登记条例）（国令第 804 号，第二、六、八、十条）. - Ghi chú: Có bốn trường hợp cơ quan đăng ký không làm, và chỉ bốn thôi: chưa đến tuổi kết hôn theo pháp luật, không phải cả hai đều hoàn toàn tự nguyện, một bên hoặc cả hai bên đã có người vợ chồng, hai người là quan hệ huyết thống trực tiếp hoặc huyết thống bên thứ ba trở lên tính từ ông bà. Những khoản có thể thu tiền chỉ là dịch vụ phụ như chụp ảnh, sao chép. Bản thân việc đăng ký không thu đồng nào, gặp khoản gọi là「phí cấp giấy」「phí làm gấp」 thì từ chối thẳng và khiếu nại. Thời hạn 30 ngày suy nghĩ khi đăng ký ly hôn xem ở mục tính chi phí thoát ra trong phần này. - ### 14. Trước khi cưới nên làm một lần khám sức khoẻ tiền hôn nhân, mắc bệnh hiểm nghèo nặng thì phải nói thật với bên kia trước khi đăng ký - Chi phí: Phần lớn các nơi miễn phí hoặc giảm phí cho người đúng tuổi. Tự trả cũng chỉ vài trăm tệ. Mất nửa ngày. @@ -128,17 +118,7 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: B - Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第一千零五十三条）. ；Quốc vụ viện (国务院) (2025). Quy định đăng ký hôn nhân（婚姻登记条例）（国令第 804 号，第十二条）. ；Văn phòng Ủy ban Quốc gia Y tế và Sức khoẻ dân dân (国家卫生健康委办公厅) (2023). Kế hoạch nâng cao năng lực phòng chống dị tật bẩm sinh（出生缺陷防治能力提升计划（2023—2027 年））（国卫办妇幼发〔2023〕9 号）. ；Ủy ban Quốc gia Y tế và Sức khoẻ dân dân cùng bốn cơ quan khác (国家卫生健康委等五部门) (2020). Thông báo về việc tăng cường chăm sóc sức khoẻ tiền hôn nhân（关于加强婚前保健工作的通知）（国卫妇幼函〔2020〕205 号）：「积极倡导各地结合实际推行免费婚前检查」. - Ghi chú: Khám sức khoẻ tiền hôn nhân từ 2003 không còn là bước phải làm trước khi cấp giấy, có đi không là tuỳ chọn. Nhưng quy định「mắc bệnh hiểm nghèo nặng mà không nói thật」thật sự trao cho bên kia quyền ra tòa án hủy cuộc hôn nhân đó, phải đưa ra trong vòng một năm kể từ ngày biết. Cả nước không có quy định thống nhất rằng khám là miễn phí. Thông báo năm 2020 của Ủy ban Quốc gia Y tế và Sức khoẻ dân dân cùng bốn cơ quan khác chỉ viết「积极倡导各地结合实际推行免费婚前检查」. Vì vậy có miễn phí không, khám những hạng mục nào thì do từng nơi tự quy định, hỏi tại quầy một cửa của cơ quan bảo vệ bà mẹ và trẻ em địa phương hoặc của văn phòng đăng ký hôn nhân. - -### 15. Đừng chỉ hỏi giá trị tình cảm「có hay không」, hãy xem chất lượng quan hệ - -- Chi phí: Không tốn tiền. Trước khi cưới, dùng mấy lần cãi vã, mấy tháng sống chung để xem chất lượng quan hệ, đừng chỉ đối chiếu danh sách điều kiện. Khó ở chỗ phải nhịn không kết luận chỉ từ danh sách. -- Nói thẳng: Chất lượng hôn nhân càng cao thì người ta càng khoẻ, nhưng mối liên hệ này rất yếu, gần bằng mức ảnh hưởng của thói quen như ăn uống tới sức khoẻ. Muốn xem quan hệ có tốt không thì phải nhìn qua mấy lần cãi vã, mấy tháng sống chung, đừng chỉ đối chiếu danh sách điều kiện. Cuộc hôn nhân tệ có gây hại cho sức khoẻ nhiều hơn việc không kết hôn không, chưa có dữ liệu nào so trực tiếp. -- Lợi ích: Phân tích tổng hợp 126 nghiên cứu, hơn 72.000 người cho thấy chất lượng hôn nhân càng cao thì sức khoẻ càng tốt. Cỡ tác động r nằm trong khoảng 0.07 đến 0.21. r là mức độ hai việc biến đổi cùng nhau, 0 là không liên quan gì, 1 là đồng biến hoàn toàn. Trong đó r = 0.11 với nguy cơ tử vong, r = −0.13 với phản ứng tim mạch khi xảy ra xung đột. Cỡ này tương đương với ảnh hưởng của các hành vi sức khoẻ như ăn uống tới sức khoẻ. Trong các nghiên cứu có kiểm tra rõ khác biệt giới tính, hầu như không thấy nam nữ khác nhau. Một tổng quan trong lĩnh vực tim mạch chỉ ra rằng mức độ không hài lòng với hôn nhân và chất lượng hôn nhân có ảnh hưởng rõ rệt tới nguy cơ tim mạch -- Mức bằng chứng: B (Tranh cãi) -- Nguồn: Robles TF, et al. (2014). Marital quality and health: a meta-analytic review. Psychol Bull 140(1):140–187. ；Dhindsa DS, et al. (2020). Marital status and outcomes in patients with cardiovascular disease. Trends Cardiovasc Med 30(4):215–220. -- Ghi chú:Tranh cãi ở chỗ mối liên hệ này vốn đã rất yếu. Một phần kết quả có thiên lệch công bố. Hơn nữa phần lớn nghiên cứu chỉ khảo sát tại một thời điểm, không thấy rõ ai dẫn tới ai, chính Robles trong bài gốc cũng liệt kê các hạn chế đó. Phần này không tìm thấy phân tích tổng hợp nào so trực tiếp tỷ lệ tử vong giữa「hôn nhân chất lượng thấp」với「chưa kết hôn」, nên「hôn nhân tệ tệ hơn không kết hôn」hiện chỉ là suy luận, không phải kết luận bằng số. - -### 16. Tính chi phí thoát ra: ly hôn thuận lợi có 30 ngày suy nghĩ, ly hôn kiện tụng có điều kiện theo pháp luật +### 15. Tính chi phí thoát ra: ly hôn thuận lợi có 30 ngày suy nghĩ, ly hôn kiện tụng có điều kiện theo pháp luật - Chi phí: Không tốn tiền. Trước khi cưới dành chút thời gian đọc: nhỡ phải thoát ra thì mất bao lâu, đi theo thủ tục nào. Khó ở chỗ phải nghĩ chuyện chia tay trước khi cưới. - Nói thẳng: Ly hôn thuận lợi không phải ký xong là xong. Sau khi nộp đơn trong 30 ngày, bất kỳ bên nào cũng rút lại được. Hết hạn thì vẫn phải trong 30 ngày, hai người tự đến nhận giấy chứng nhận ly hôn, không đến thì coi như rút đơn. Bên kia không đồng ý thì chỉ còn đi kiện. Tòa án hoà giải trước, phải có bạo lực gia đình, cưới hai vợ, sống ly tân đủ hai năm và các trường hợp tương tự, hoà giải không xong mới xử cho ly hôn. @@ -146,25 +126,26 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人民代表大会) (2020). Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（中华人民共和国民法典）（第一千零七十六、一千零七十七、一千零七十九条）. ；Bộ Dân vụ (民政部) (2020). Thông báo về việc triển khai các quy định liên quan đến đăng ký hôn nhân trong Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（关于贯彻落实《中华人民共和国民法典》中有关婚姻登记规定的通知）（民发〔2020〕116 号）. ；Bộ Dân vụ (民政部) (2025). Báo cáo thống kê công trình dân vụ năm 2024. Nguồn tham khảo giống mục 7 của phần này（结婚率、离婚率各是什么口径） - Ghi chú: Thời hạn suy nghĩ chỉ áp dụng cho ly hôn thuận lợi. Ly hôn kiện tụng không có thời hạn suy nghĩ, nhưng kéo dài hơn. Phí tố tụng, phí luật sư mỗi nơi khác nhau, mục này không viết số. Chi phí thoát ra cao không có nghĩa là không nên cưới, nhưng trước khi cưới thì nên biết những điều này. - -### 17. Riêng phần「kết hôn vì bố mẹ」thì ghi một sổ riêng, đừng trộn vào sổ của bạn +### 16. Ghi riêng một sổ cho việc「cưới vì bậc trưởng bối」, đừng trộn vào sổ của chính mình -- Chi phí: Không tốn tiền. Trên giấy chia làm hai cột. Một cột là trong cuộc hôn nhân đó những lợi ích nào là của bạn, mấy mục ở đầu phần này tính được. Cột kia là những lợi ích nào là của bố mẹ: yên tâm, thể diện, không còn bị thúc cưới. Khó ở chỗ phải để cột của bố mẹ riêng ra, không gộp vào cột của mình. -- Nói thẳng: Phần lợi ích của bố mẹ, tức là yên tâm, thể diện, không còn bị thúc cưới, là có thật. Nhưng nó không tự động thành lợi ích sức khoẻ hay chất lượng quan hệ của bạn. Còn ba sổ là thời gian, tiền và chi phí thoát ra thì do bạn tự gánh hết. Chia hai cột trên giấy, viết rõ lợi ích nào của mình, lợi ích nào của bố mẹ, sổ này có đáng hay không thì bạn nhìn là biết. -- Lợi ích: Lợi ích của bố mẹ là có thật. Nhưng nó không tự động chuyển thành lợi ích sức khoẻ của bạn (mục 8 của phần này) hay chất lượng quan hệ (mục 15 của phần này). Còn sổ thời gian (mục 9 của phần này), sổ tiền (mục 10 đến 12 của phần này) và chi phí thoát ra (mục 16 của phần này) thì do hai vợ chồng tự gánh. Tách sổ ra rồi,「hiệu quả chi phí」của việc kết hôn như làm nhiệm vụ là cao hay thấp thì bạn tự thấy được +- Chi phí: Không tốn tiền. Trên giấy chia làm hai cột. Một cột là những lợi ích trong cuộc hôn nhân này thuộc về chính bạn, những mục trước trong phần này tính được. Cột còn lại là lợi ích của bậc trưởng bối: yên tâm, thể diện, không còn bị thúc ép cưới. Khó ở chỗ giữ cột của bậc trưởng bối riêng, không gộp vào cột của bạn. +- Nói thẳng: Lợi ích cho bậc trưởng bối, tức là yên tâm, thể diện, không còn bị thúc, là có thật. Nhưng nó không tự động biến thành lợi ích sức khoẻ hay chất lượng mối quan hệ của bạn. Còn sổ thời gian, sổ tiền và chi phí thoát ra thì ba sổ đó bạn tự gánh hết. Chia hai cột trên giấy, viết rõ lợi ích nào của mình, lợi ích nào của bậc trưởng bối, nhìn một lần là biết khoản đó đáng hay không. +- Lợi ích: Lợi ích cho bậc trưởng bối là có thật. Nhưng nó không tự động chuyển thành lợi ích sức khoẻ của bạn (mục 8) hay chất lượng mối quan hệ (mục 4, mối quan hệ tốt hay không xem cảm nhận của chính mình). Còn sổ thời gian (mục 9), sổ tiền (mục 10 đến 12) và chi phí thoát ra (mục 15) thì người trong cuộc gánh hết. Sau khi tách sổ,「cưới như hoàn thành nhiệm vụ」có đáng hay không thì bạn tự thấy ngay - Mức bằng chứng: C -- Nguồn: Kinh nghiệm của tác giả, thuộc loại gợi ý về phương pháp; phần dữ liệu xem các mục trước trong phần này -- Ghi chú: Cột của bố mẹ cũng có thể tính là một khoản rất nặng, đánh đổi thế nào thì bạn tự quyết. Danh sách điền chỗ trống xem tại [docs/结婚划不划算.md](../docs/结婚划不划算.md). -### 18. Coi quan hệ tốt xấu là một sổ sức khoẻ: cãi nhau thì nói vào việc, không nói vào người, đừng chửi, đừng chế giễu +- Nguồn: Kinh nghiệm của tác giả, dạng gợi ý về phương pháp; phần dữ liệu xem các mục trước trong phần này +- Ghi chú: Cột của bậc trưởng bối cũng có thể tính là một khoản rất nặng, cân nhắc thế nào thì bạn tự quyết. Danh sách điền chỗ trống xem [docs/结婚划不划算.md](../docs/结婚划不划算.md) +### 17. Hãy coi chất lượng mối quan hệ là một khoản sức khoẻ: cãi nhau thì nói về việc chứ không công kích con người, đừng chửi, đừng đào lại -- Chi phí: Không tốn tiền. Khi cãi nhau chỉ nói chuyện đang xảy ra trước mắt, không chửi, không chế giễu, không đào lại chuyện cũ. Khó ở chỗ phải kìm miệng lúc nóng máu. -- Nói thẳng: Vợ chồng có quan hệ tốt thì thân thể khoẻ hơn một chút, nguy cơ tử vong cũng thấp hơn một chút. Mối liên hệ này không lớn, gần bằng mức ảnh hưởng của ăn uống tới sức khoẻ. Trong một thí nghiệm, vợ chồng mà lúc cãi nhau luôn mang thái độ thù địch thì vết thương nhỏ do thí nghiệm tạo ra chỉ lành nhanh bằng sáu phần mười tốc độ của vợ chồng bình tĩnh. -- Lợi ích: Gộp 126 nghiên cứu, hơn 72.000 người lại tính (phân tích tổng hợp). Chất lượng hôn nhân càng cao thì thân thể càng tốt, hệ số tương quan r nằm trong khoảng .07 đến .21. Hệ số tương quan 0 là không liên quan gì, 1 là đồng biến hoàn toàn. Trong đó tương quan với việc nguy cơ tử vong thấp hơn là r = .11. Cãi nhau thì phản ứng tim mạch nhỏ hơn, r = -.13. Tác giả nói các tác động này đều nhỏ, gần bằng mức liên hệ giữa thói quen sức khoẻ như ăn uống với sức khoẻ. Một thí nghiệm giao cắt khác dùng 42 cặp vợ chồng khoẻ, mỗi cặp nằm viện hai lần, mỗi lần 24 giờ. Một lần để nói chuyện hỗ trợ lẫn nhau, một lần để bàn một việc có bất đồng. Sau khi cãi nhau xong, vết thương bọng nước do thí nghiệm tạo ra lành chậm hơn. Vợ chồng thể hiện thái độ thù địch nhiều hơn ở cả hai lần thì tốc độ lành vết thương chỉ bằng 60% tốc độ của vợ chồng ít thù địch hơn. Sáng hôm sau khi cãi nhau, các chỉ số viêm trong máu của họ (IL-6, yếu tố hoại tử u) cũng tăng nhiều hơn +- Chi phí: Không tốn tiền. Khi cãi nhau chỉ nói đúng sự việc trước mắt, không chửi, không đào lại, không nhắc chuyện cũ. Khó ở chỗ giữ miệng khi nóng. +- Nói thẳng: Vợ chồng có quan hệ tốt thì sức khoẻ một chút, nguy cơ chết cũng thấp hơn một chút. Mối liên hệ này không lớn, cỡ bằng mối liên hệ giữa khẩu phần và sức khoẻ. Trong một thí nghiệm, vợ chồng cãi nhau mà lúc nào cũng mang thái độ thù địch thì vết thương nhỏ do thí nghiệm tạo ra lành chậm hơn vợ chồng bình thản tới mức chỉ còn sáu phần mười tốc độ. +- Lợi ích: Gộp 126 nghiên cứu, hơn 72 nghìn người lại (phân tích tổng hợp). Chất lượng hôn nhân càng cao thì sức khoẻ càng tốt, hệ số tương quan r trong khoảng .07 đến .21. Hệ số tương quan 0 là không liên quan gì, 1 là thay đổi hoàn toàn đồng pha. Trong đó liên hệ với nguy cơ chết thấp hơn là r = .11. Cãi nhau mà phản ứng tim mạch nhỏ hơn là r = -.13. Tác giả nói các hiệu ứng này đều nhỏ, cỡ bằng mối liên hệ giữa các thói quen sức khoẻ như ăn uống và sức khoẻ. Một thí nghiệm chéo khác có 42 cặp vợ chồng khỏe mạnh, mỗi cặp vào viện hai lần, mỗi lần 24 giờ. Một lần nói chuyện hỗ trợ lẫn nhau, một lần thảo luận một việc có bất đồng. Sau khi cãi nhau, vết thương bọng nước do thí nghiệm tạo ra lành chậm hơn. Cặp nào ở cả hai lần đều thể hiện nhiều thái độ thù địch hơn thì tốc độ lành vết thương chỉ còn 60% so với cặp ít thù địch. Sáng hôm sau khi cãi nhau, các chỉ số viêm trong máu của họ (IL-6, yếu tố hoại tử u) cũng tăng nhiều hơn. - Mức bằng chứng: B - Nguồn: Robles TF, Slatcher RB, Trombello JM, McGinn MM (2014). Marital quality and health: a meta-analytic review. Psychological Bulletin. ；Kiecolt-Glaser JK 等 (2005). Hostile marital interactions, proinflammatory cytokine production, and wound healing. Archives of General Psychiatry. -- Ghi chú: Lý do để mức B là: phần lớn nghiên cứu được gộp chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. Nên không phân biệt được quan hệ xấu khiến người ta bị bệnh, hay thân thể xấu khiến quan hệ xấu đi. Có nghiên cứu sau khi trừ tuổi và thu nhập thì mối liên hệ vẫn còn. Thí nghiệm vết thương chỉ có 42 cặp vợ chồng. Cách làm「nói vào việc không nói vào người, không chửi không chế giễu」là tác giả suy ra từ「hành vi thù địch」trong thí nghiệm, không có nghiên cứu nào so trực tiếp. Cãi nhau dữ dội mà không sửa được thì hai người cùng đi làm tư vấn hôn nhân và gia đình, điểm này cũng không có tài liệu riêng. Cãi nhau mà đánh nhau, hoặc lạm ngạng, đe dọa lâu dài, thì không còn là chuyện giao tiếp nữa, đó là bạo lực gia đình, xem mục 43 ở phần 8 (bị bạo lực gia đình). Muốn cân nhắc có nên đi không thì xem chi phí thoát ra ở mục 16 của phần này (chi phí thoát ra). +- Ghi chú: Lý do để ở mức B: phần lớn các nghiên cứu gộp lại chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. Nên không phân biệt được là quan hệ kém khiến người ta ốm hay người ốm khiến quan hệ kém đi. Có nghiên cứu trừ đi tuổi và thu nhập thì mối liên hệ vẫn còn. Nguyên văn của Robles còn ghi rằng một phần kết quả có thiên lệch công bố. Nghiên cứu nào so sánh rõ nam với nữ thì gần như không tìm ra khác biệt giữa hai giới. Phần này không tìm thấy phân tích tổng hợp nào so trực tiếp tỷ lệ tử vong giữa「hôn nhân kém」với「không kết hôn」, nên「hôn nhân xấu còn gây hại hơn cả không cưới」chỉ là suy luận. Thí nghiệm vết thương chỉ có 42 cặp vợ chồng. 「Nói về việc chứ không công kích con người, không chửi không đào lại」là cách làm tác giả rút ra từ「hành vi thù địch」trong thí nghiệm, không có nghiên cứu nào so sánh trực tiếp. Cãi nhau dữ dội mà không sửa được, hai người cùng đi làm trị liệu đôi, xem mục 20 của phần này (trị liệu đôi). Cãi nhau mà đánh nhau, hay chửi bạo lực và đe dọa lâu dài, thì không còn là chuyện giao tiếp nữa mà là bạo lực gia đình, xem mục 43 ở phần 8 (bị bạo lực gia đình). Cân nhắc kỹ có nên đi hay không, chi phí thoát ra xem mục 15 của phần này (chi phí thoát ra). + -### 19. Đôi bạn cùng giới nên tranh thủ lúc cả hai còn minh mẫn, làm xong giấy uỷ quyền, giám hộ tự nguyện bằng văn bản và di chúc: về mặt pháp luật hai người không phải người thân, không làm thì không có quyền ký và không có quyền thừa kế + +### 18. Đôi bạn cùng giới nên tranh thủ lúc cả hai còn minh mẫn, làm xong giấy uỷ quyền, giám hộ tự nguyện bằng văn bản và di chúc: về mặt pháp luật hai người không phải người thân, không làm thì không có quyền ký và không có quyền thừa kế - Chi phí: Giám hộ tự nguyện bằng văn bản và di chúc thì làm ở văn phòng công chứng, phí công chứng mỗi nơi khác nhau, vài trăm tệ trở lên. Di chúc tự tay viết thì không tốn tiền. Trước sau phải dành một hai buổi nửa ngày. Khó ở chỗ hai người phải ngồi lại nói chuyện bệnh tật và cái chết. - Nói thẳng: Người thân theo pháp luật là vợ chồng, cha mẹ, con cái, anh chị em, bạn đời không phải. Không làm giấy tờ trước thì khi một bên nặng bệnh, bên kia không quyết được thay. Khi một bên qua đời, bên kia không thừa kế được phần nào. Ba bản giấy tờ mỗi bản phụ trách một đoạn: khi còn sống thì dựa vào giấy uỷ quyền, khi mất năng lực tâm thần thì dựa vào giám hộ tự nguyện bằng văn bản, sau khi mất thì dựa vào di chúc. @@ -172,12 +153,19 @@ Phần này tách câu hỏi「kết hôn có đáng không」thành mấy bản - Mức bằng chứng: A - Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第二十三、三十三、一百六十五、一百七十三、一千零四十五、一千一百二十四、一千一百二十七、一千一百三十三至一千一百四十、一千一百四十二、一千二百一十九条）. （Cơ sở dữ liệu văn pháp quốc gia）；Sở Tư pháp thành phố Ích Châu (宜昌市司法局) (2021). Công chứng giám hộ tự nguyện bằng văn bản: con đường công chứng tới cầu vồng. Mục「实践荟萃」trên trang của Sở Tư pháp tỉnh Hồ Bắc. - Ghi chú: Vị thế của người thừa kế của bạn đời là「người được tặng theo di chúc」, không phải người thừa kế. Biết có di chúc như vậy thì trong 60 ngày phải nói rõ là nhận, không nói là coi như từ bỏ. Bản thân bạn đời và những người có lợi ích liên quan với họ không được làm người chứng kiến di chúc. Đã lập nhiều di chúc thì lấy bản cuối cùng, xem mục 2 ở phần 17 (di chúc sau thì đảo ngược di chúc trước). Người giám hộ tự nguyện có ký trực tiếp được giấy đồng ý phẫu thuật không thì mỗi bệnh viện làm khác nhau, hỏi trước bệnh viện bạn hay đi. Người thân không công nhận các giấy tờ này thì phải ra tòa: xét sơ thẩm theo thủ tục thông thường từ 6 tháng trở lên, phí luật sư tự bỏ ra. Vì vậy giấy tờ càng nên làm ở văn phòng công chứng. Nhà mua chung, tiền gửi chung, thì viết thêm một thỏa thuận bằng văn bản theo số tiền góp thật. Làm giám hộ tự nguyện bằng văn bản thì xem mục 1 ở phần 17 (chỉ định bằng văn bản người giám hộ trong tương lai). Người được lợi là chính bạn và bạn đời của bạn. - -### 20. Đừng kết hôn mà giấu xu hướng tính dục, cũng đừng coi hôn nhân hình thức là cách cho nhanh: về mặt pháp luật đó là một cuộc hôn nhân thật, muốn thoát ra chỉ có thể ly hôn +### 19. Đừng giấu xu hướng tính dục mà kết hôn, cũng đừng coi hôn nhân hình thức là cách cho tiện -- Chi phí: Không tốn tiền. Khó ở chỗ phải chịu được áp lực bố mẹ thúc cưới, có thể phải gánh nhiều năm. -- Nói thẳng: Kết hôn mà giấu xu hướng tính dục thì về mặt pháp luật không phải là hôn nhân vô hiệu, cũng không hủy được, muốn thoát ra chỉ có ly hôn. Hôn nhân hình thức cũng là hôn nhân thật, tiền kiếm được sau khi cưới, nợ mắc sau khi cưới đều tính cho hai vợ chồng. Bên bị giấu khi ly hôn có thể yêu cầu chia nhiều tài sản hơn, bồi thường tổn thương tinh thần, nhưng tòa án có chấp nhận hay không thì chưa có kết luận. -- Lợi ích: Bộ luật Dân sự quy định hôn nhân vô hiệu chỉ có ba trường hợp: cưới hai vợ, có quan hệ huyết thống cấm kết hôn, chưa đến tuổi kết hôn theo pháp luật（第一千零五十一条）. Hôn nhân có thể hủy được chỉ có hai trường hợp: bị ép cưới, giấu bệnh hiểm nghèo nặng trước khi cưới（第一千零五十二、一千零五十三条）. Yêu cầu xác nhận hôn nhân vô hiệu với lý do ngoài ba trường hợp đó thì tòa án xử bác（第十七条 của Giải thích (một) về việc áp dụng phần hôn nhân và gia đình của Bộ luật Dân sự）. Vì vậy muốn thoát ra chỉ có thể ly hôn. Hôn nhân hình thức đã đăng ký vẫn là hôn nhân hợp pháp. Lương, thưởng, tiền công lao động sau khi cưới là tài sản chung của vợ chồng（第一千零六十二条）. Khoản nợ vay có cả hai vợ chồng cùng ký, hoặc một bên vay để chi phí sinh hoạt hằng ngày của gia đình, là nợ chung của vợ chồng（第一千零六十四条）. Bồi thường tổn thương khi ly hôn chỉ áp dụng cho cưới hai vợ, sống chung với người khác, bạo lực gia đình, ngược đãi bỏ rơi thành viên gia đình, và「lỗi lớn khác」（第一千零九十一条）. Giải thích (một) giới hạn「sống chung với người khác」là sống chung với「người khác giới tính ngoài hôn nhân」（第二条 của Giải thích (một)）. Vì vậy quan hệ đồng giới chỉ có thể được tranh đòi theo「lỗi lớn khác」. Phần bồi thường gồm cả bồi thường tổn thương tinh thần, bên bị giấu phải nêu ra cùng đơn kiện ly hôn, nếu không xử cho ly hôn thì không hỗ trợ. Sau khi Bộ luật Dân sự có hiệu lực, không tìm thấy phán quyết nào của tòa án được công bố cho loại vụ này. Thời kỳ luật cũ, một bài viết đăng năm 2013 trên trang của Tòa án nhân dân tỉnh Giang Tô bàn về vụ vợ kiện chồng vì chồng giấu chuyện đồng tính luyến. Bài viết nêu hai cách hiểu. Một cách cho rằng đồng tính luyến không thuộc lỗi theo pháp luật thời đó, không nên hỗ trợ chia nhiều tài sản cùng bồi thường tổn thương tinh thần. Tác giả đồng tình với cách còn lại: coi bên bị giấu là「bên không có lỗi」, khi chia tài sản nên quan tâm đến cô, và còn phải bồi thường tiền an ủi tinh thần (toàn quốc) +- Chi phí: Không tốn tiền. Khó ở chỗ phải chịu áp lực thúc ép cưới từ gia đình, có thể phải gánh nhiều năm. +- Nói thẳng: Kết hôn mà giấu xu hướng tính dục thì về mặt pháp lý không phải là hôn nhân vô hiệu, cũng không huỷ được, muốn thoát ra chỉ có thể ly hôn. Hôn nhân hình thức cũng là hôn nhân thật, thu nhập sau khi cưới và nợ phải trả đều tính theo vợ chồng. Bên bị giấu khi ly hôn có thể yêu cầu chia nhiều tài sản hơn, yêu cầu bồi thường tổn thương tinh thần, nhưng tòa có chấp nhận hay không thì chưa có kết luận. +- Lợi ích: Điều 1051 của Bộ luật Dân sự quy định hôn nhân vô hiệu chỉ có ba trường hợp: đã có vợ chồng mà kết hôn, có quan hệ huyết thống bị cấm kết hôn, chưa tới tuổi kết hôn theo quy định. Điều 1052, 1053 quy định huỷ được chỉ có hai trường hợp: bị ép cưới, giấu bệnh nặng trước khi cưới. Điều 17 của Giải thích (một) về phần hôn nhân và gia đình trong Bộ luật Dân sự quy định, yêu cầu xác nhận hôn nhân vô hiệu vì lý do ngoài ba trường hợp đó thì tòa án bác yêu cầu. Vậy muốn thoát ra chỉ có thể ly hôn. Hôn nhân hình thức đã đăng ký vẫn là hôn nhân hợp pháp. Điều 1062 quy định lương, thưởng, thu nhập lao động và các khoản khác sau khi cưới là tài sản chung của vợ chồng. Điều 1064 quy định, khoản nợ vay có cả hai vợ chồng cùng ký, hoặc một bên vay cho chi phí sinh hoạt gia đình thường ngày, là nợ chung của vợ chồng. Bồi thường tổn thương khi ly hôn xem Điều 1091, chỉ giới hạn ở đã có vợ chồng mà kết hôn, sống chung với người khác, bạo lực gia đình, ngược đãi hay bỏ rơi thành viên gia đình, và「có lỗi nghiêm trọng khác」. Điều 2 của Giải thích (một) giới hạn「sống chung với người khác」là sống chung với「người khác giới tính không cùng giới tính」. Nên quan hệ đồng tính chỉ có thể tranh cãi theo「lỗi nghiêm trọng khác」. Bồi thường gồm cả bồi thường tổn thương tinh thần, bên bị giấu phải nêu ra cùng lúc kiện ly hôn, tòa không cho ly hôn thì không hỗ trợ. Sau khi Bộ luật Dân sự có hiệu lực, không tìm thấy bản án loại này mà tòa án công khai. Thời luật cũ, năm 2013 trên trang mạng của Tòa án nhân dân tỉnh Giang Tô có đăng một bài bàn về vụ vợ kiện ly hôn vì chồng giấu xu hướng đồng tính. Bài liệt kê hai cách nhìn. Một là cho rằng đồng tính không thuộc loại lỗi theo luật thời đó, không nên hỗ trợ chia nhiều tài sản cũng không nên bồi thường tinh thần. Tác giả đồng tình với cách còn lại: coi bên bị giấu là「bên không có lỗi」, khi chia tài sản nên để ý đến cô ấy, và còn phải bồi thường tiền an ủi tinh thần (toàn quốc) - Mức bằng chứng: A -- Nguồn: Đại hội Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự（民法典）（第一千零五十一至一千零五十四、一千零六十二、一千零六十四、一千零七十九、一千零八十七、一千零九十一条）. （Cơ sở dữ liệu văn pháp quốc gia）；Tòa án nhân dân tối cao (最高人民法院) (2020). Giải thích (một) về việc áp dụng phần hôn nhân và gia đình của Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（关于适用《中华人民共和国民法典》婚姻家庭编的解释（一））（法释〔2020〕22 号，第二、十七、八十六至九十条）. ；Zhu Laikuan, Wang Hongjuan (2013). Chồng đồng tính luyến dẫn tới ly hôn thì có thể yêu cầu bồi thường tổn thương hay không? Trang mạng Tòa án Giang Tô. -- Ghi chú: Phần chia nhiều tài sản và bồi thường tổn thương tinh thần chỉ có một bài phân tích, bài đó không viết vụ cuối cùng xử ra sao, nên có nhận được hay không còn tuỳ từng tòa. Bộ luật Dân sự thêm vào luật cũ một mục「lỗi lớn khác」, cho bên bị giấu một cửa để tranh đòi. Bản thân điều khoản là mức A. Hai bên của hôn nhân hình thức đều có bạn đời cùng giới thì giữa hai bạn đời không được luật hôn nhân bảo vệ, tài sản phải viết riêng cho rõ, xem mục 19 của phần này (đôi bạn cùng giới làm giám hộ tự nguyện bằng văn bản và di chúc). Ly hôn hình thức cũng phải qua 30 ngày suy nghĩ hoặc ra tòa, xem mục 16 của phần này (ly hôn thuận lợi có 30 ngày suy nghĩ). Ly hôn thuận lợi mà viết là từ bỏ quyền bồi thường thì sau đó kiện sẽ không được hỗ trợ. Người được lợi trước hết là người vợ chồng bị giấu, sau đó là chính bạn: một khi chuyện giấu bị phát hiện thì ly hôn, chia tài sản, nuôi con sẽ dính vào nhau cùng lúc. \ No newline at end of file +- Nguồn: Quốc hội nhân dân (2020). 民法典（第一千零五十一至一千零五十四、一千零六十二、一千零六十四、一千零七十九、一千零八十七、一千零九十一条）. （Cơ sở dữ liệu pháp luật và pháp quốc quốc gia）；Tòa án nhân dân tối cao (2020). 关于适用《中华人民共和国民法典》婚姻家庭编的解释（一）（法释〔2020〕22 号，第二、十七、八十六至九十条）. ；Chu Lai Kuan, Vương Hồng Uyên (2013). 因同性恋导致离婚能否主张损害赔偿？江苏法院网. +- Ghi chú: Phần chia nhiều tài sản và bồi thường tinh thần chỉ có một bài phân tích, không nói vụ cuối cùng xử thế nào, nên có lấy được hay không phụ thuộc từng tòa án. Bộ luật Dân sự thêm một mục「lỗi nghiêm trọng khác」so với luật cũ, cho bên bị giấu một cửa để tranh cãi. Chính văn bản luật ở mức A. Hai bên trong hôn nhân hình thức đều có bạn đời cùng giới thì các bạn đời đó không được luật hôn nhân bảo vệ, tài sản phải viết riêng cho rõ, xem mục 18 của phần này (bạn đời cùng giới làm giám hộ tự nguyện bằng văn bản và di chúc). Ly hôn trong hôn nhân hình thức cũng phải qua 30 ngày cân nhắc hoặc ra tòa, xem mục 15 của phần này (ly hôn thuận có 30 ngày cân nhắc). Ly hôn thuận mà đã viết từ bỏ bồi thường thì sau đó kiện sẽ không được hỗ trợ. Người hưởng lợi trước tiên là người phối ngẫu bị giấu, kế đến là bạn: một khi chuyện giấu bị phát hiện thì ly hôn, chia tài sản, nuôi con cùng ập tới một lúc. +### 20. Mối quan hệ kẹt ở chỗ cãi nhau lặp đi lặp lại và im lặng, thì cả hai cùng đi làm trị liệu đôi: ngồi chờ thì gần như không tự khỏi + +- Chi phí: Phải tốn tiền. Tính theo buổi, mỗi bệnh viện một giá, mục này không viết con số. Trong nghiên cứu thường đi mỗi tuần một lần, một thử nghiệm kéo dài khoảng 8 tháng trước và sau. Khó ở chỗ cả hai người đều phải đi, mà còn phải kiên trì làm hết. +- Nói thẳng: Cặp đôi kẹt ở mối quan hệ mà cùng làm trị liệu đôi thì hài lòng với mối quan hệ rõ ràng hơn, khoảng bảy phần mười có cải thiện. Cặp chỉ xếp hàng chờ thì gần như không thay đổi gì. Nhưng hiệu quả có mờ dần: đến năm thứ 5, khoảng một nửa vẫn còn tốt rõ rệt, khoảng một phần tư đã ly trú hoặc ly hôn. +- Lợi ích: Một phân tích tổng hợp năm 2020 gộp 58 nghiên cứu, 2092 cặp đôi. Sau khi làm trị liệu đôi, mức thay đổi của sự hài lòng mối quan hệ g = 1.12 (khoảng tin cậy 0.92–1.31), tác giả nói đây là hiệu ứng lớn. g là cỡ hiệu ứng, xem bảng thuật ngữ trong README. Cặp được xếp vào nhóm chờ trong cùng thời gian đó thì g = 0.12 (−0.04 đến 0.29), không cải thiện rõ rệt. Lúc bắt đầu quan hệ càng tệ thì càng cải thiện nhiều. Một tổng quan năm 2012 viết rằng trị liệu đôi giúp 70% các cặp được điều trị có cải thiện. Một thử nghiệm chia ngẫu nhiên khác tìm 134 cặp vợ chồng lâu năm, bất hòa nặng, thực hiện hai dạng trị liệu đôi theo hành vi trong khoảng 8 tháng. Lúc kết thúc mức thay đổi sự hài lòng là d = 0.90 và 0.71. Đến năm thứ 5, lần lượt 50.0% và 45.9% cặp vợ chồng vẫn còn cải thiện rõ rệt. Cải thiện rõ rệt ở đây tác giả gọi là「có cải thiện mang ý nghĩa lâm sàng」. Đồng thời 25.7% và 27.9% đã ly trú hoặc ly hôn +- Mức bằng chứng: A（Tranh cãi） +- Nguồn: Roddy MK, Walsh LM, Rothman K, Hatch SG, Doss BD (2020). Meta-analysis of couple therapy: effects across outcomes, designs, timeframes, and other moderators. Journal of Consulting and Clinical Psychology, 88(7), 583-596. ；Lebow JL, Chambers AL, Christensen A, Johnson SM (2012). Research on the treatment of couple distress. Journal of Marital and Family Therapy, 38(1), 145-168. ；Christensen A, Atkins DC, Baucom B, Yi J (2010). Marital status and satisfaction five years following a randomized clinical trial comparing traditional versus integrative behavioral couple therapy. Journal of Consulting and Clinical Psychology, 78(2), 225-235. ；Rathgeber M, Bürkner PC, Schiller EM, Holling H (2019). The efficacy of emotionally focused couples therapy and behavioral couples therapy: a meta-analysis. Journal of Marital and Family Therapy, 45(3), 447-463. （Phe phủ đối lập）；Shadish WR, Baldwin SA (2003). Meta-analysis of MFT interventions. Journal of Marital and Family Therapy, 29(4), 547-570. （Phe phủ đối lập）；Karakurt G, Whiting K, van Esch C, Bolen SD, Calabrese JR (2016). Couples therapy for intimate partner violence: a systematic review and meta-analysis. Journal of Marital and Family Therapy, 42(4), 567-583. ；Ủy ban Thường vụ Quốc hội (2018 修正). 精神卫生法（第二十三、五十一条）. （Cơ sở dữ liệu pháp luật và pháp quốc quốc gia） +- Ghi chú: Tranh cãi ở chỗ hiệu quả giữ được bao lâu. Trong một phân tích tổng hợp khác chỉ thu thập thử nghiệm chia ngẫu nhiên, lúc kết thúc g = 0.60, nửa năm sau là 0.44. Đến 12 tháng chỉ còn trị liệu đôi theo hành vi có số liệu, g = 0.06, gần như không giữ được. Một tổng quan khác gộp 20 phân tích tổng hợp nói hiệu ứng mang ý nghĩa lâm sàng xuất hiện ở 40% đến 50% số người. 「Bảy phần mười có cải thiện」và「bốn đến năm phần mười cải thiện rõ rệt」dùng tiêu chuẩn khác nhau, đừng so sánh lẫn nhau. Thử nghiệm năm năm đó không có nhóm đối chứng không điều trị, nên không nói được ly hôn là tăng hay giảm. Nghiên cứu gộp năm 2020 toàn bộ là cặp đôi khác giới tính, mục này cũng không tìm thấy thử nghiệm chia ngẫu nhiên trong nước, áp vào trường hợp của bạn thì phải trừ đi chút. Ở trong nước làm ở đâu: Luật vệ sinh tâm thần quy định trị liệu tâm lý phải làm trong cơ sở y tế, nhân viên tư vấn tâm lý không được làm trị liệu tâm lý. Có thể hỏi khoa tâm thần hoặc khoa tâm lý ở bệnh viện xem có trị liệu đôi không. Tìm cơ sở ngoài bệnh viện thì hỏi trước điều kiện hành nghề và kiến thức đào tạo của người đó. Có bạo lực gia đình thì đừng đi cùng nhau. Phân tích tổng hợp năm 2016 ghi rõ, kết luận chỉ áp dụng cho bạo lực mức nhẹ đến trung bình, do hai bên đụng đấm nhau vì xung đột. Bạo lực mang đặc trưng kiểm soát đối phương, gây tổn thương nặng thì không nằm trong đó. Người trong nghề cũng lo rằng nói chuyện xong về nhà lại bị trả đũa. Bị đánh thì xem mục 43 ở phần 8 (bị bạo lực gia đình). Khi cãi nhau nói cách gì, xem mục 17 của phần này (cãi nhau thì nói về việc chứ không công kích con người), cân nhắc kỹ có nên đi không, xem mục 15 của phần này (chi phí thoát ra). Người hưởng lợi là bạn và người phối ngẫu. diff --git "a/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" "b/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" index e7c6f456..bd8e92f5 100644 --- "a/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" +++ "b/vi/book/11-\347\250\213\345\272\217\345\221\230\345\222\214\346\212\200\346\234\257\344\272\272\345\256\271\346\230\223\350\270\251\347\232\204\347\272\242\347\272\277.md" @@ -12,7 +12,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: C - Nguồn: Quốc hội Nhân dân (全国人大) (2018). 刑事诉讼法（2018 年修正，第三十四、三十九、六十七、七十九、九十一、一百一十、一百一十二条）. - Ghi chú: 37 ngày đó tính từ lúc bị giữ đến lúc viện kiểm sát quyết định phê chuẩn bắt hay không, không phải trần của thời gian giam. Sau khi phê chuẩn bắt còn có thời hạn giam trong giai đoạn điều tra, rồi tính tiếp từ đó. Luật sư rành việc cũng hữu ích nhất đúng trong 37 ngày ấy: gặp người bị bắt, nộp cho viện kiểm sát một ý kiến「không nên phê chuẩn bắt」, xin bảo lãnh tại ngoại. Đợi đến lúc khởi tố rồi mới tìm luật sư thì đã muộn. Các mục còn lại của phần này nói về cùng một điều: đối phương có lý do để báo việc hay không. - ### 2. Không làm và không bán hack game, kể cả công cụ hỗ trợ bằng AI「chỉ đọc màn hình, không đụng bộ nhớ」 - Chi phí: Không tốn tiền. Cái giá phải trả là bỏ các đơn thuộc loại đó cùng thu nhập từ「tiện ích nhỏ」. Chỗ khó là số tiền đó đến nhanh, lại không thấy ai sẽ truy cứu @@ -21,7 +20,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第三款、第二百一十七条第六项）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第三条）. （Cục Công an thành phố Thâm Quyến đăng lại）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). 上海静安：严惩制售游戏外挂黑灰产共护清朗网络空间. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 2020 年度检察机关保护知识产权典型案例（案例五）. - Ghi chú: Các vụ án hack đã công bố đều là loại đọc ghi bộ nhớ trò chơi, vượt cơ chế chống gian lận. Chưa thấy vụ án chính thức nào về công cụ「chỉ chụp màn hình, chỉ mô phỏng phím」. Nhưng vụ giành vé ở mục 3 cho thấy không đụng bộ nhớ, chỉ gửi yêu cầu, vẫn đủ để kết tội lấy dữ liệu trái phép. Cuối cùng có thành tội phạm hay không, tính theo tội nào, do cơ quan điều tra quyết định theo hồ sơ nhà sản xuất nộp lúc báo việc, không phải do tác giả quyết. Bán đủ 20 lượt người dùng đã là「情节严重」, mức đó rất nhỏ. - ### 3. Không viết và không bán script giành vé, đua đặt hàng, tạo đơn ảo, mò hàng miễn phí, kể cả loại chỉ「tự động bấm nút」 - Chi phí: Không tốn tiền, cũng không mất thời gian @@ -30,7 +28,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2026). 上海静安：查办利用外挂软件批量代抢火车票牟利案. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 2024 年人民法院知识产权典型案例（案例七，抢票软件不正当竞争案）. - Ghi chú: Script trong vụ giành vé không sửa bộ nhớ, không bẻ khóa, chỉ「nhanh hơn người」, vẫn bị kết tội theo Điều 285 của Bộ luật Hình sự. Nhận tiền để giành vé thay và bán script đều là hưởng lợi phi pháp. Trường hợp chỉ tự dùng một lần thì không tìm thấy vụ án chính thức nào. - ### 4. Crawler chỉ lấy trang công khai không cần đăng nhập, không vòng cơ chế chống thu thập, không đụng thông tin cá nhân, dữ liệu đã lấy thì không bán - Chi phí: Không tốn tiền. Crawler là chương trình tự động mở trang web rồi gom dữ liệu trên đó. Trước khi lấy, xem qua tệp robots và hướng dẫn cấp quyền của giao diện; robots là tệp website đặt ở thư mục gốc, ghi rõ những trang nào không được lấy. Dữ liệu phải đăng nhập, phải có token (một chuỗi giấy thông hành tạm), phải bẻ chữ ký mới có được thì đừng lấy. Chỗ khó là loại dữ liệu đó về kỹ thuật cứ thế là có, nhịn lại hoàn toàn dựa vào tự kiểm @@ -39,7 +36,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百五十三条之一）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2017). 关于办理侵犯公民个人信息刑事案件适用法律若干问题的解释（第三、四、五条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 检察机关依法惩治侵犯公民个人信息犯罪典型案例（案例二）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例一）. - Ghi chú: Giải thích tư pháp năm 2017 quy định「未经本人同意，把合法收集来的信息给别人」cũng tính là「提供」. Trong công ty mà xuất bảng người dùng cho đối tác bên ngoài thì cũng rơi vào câu đó. 5000 mục chỉ bằng một bảng nhỏ. - ### 5. Không nhận đơn phát triển ứng dụng cờ bạc, lừa đảo, khiêu dâm, không dựng cổng thanh toán cho「khách hàng», không mở khoá tài khoản bị khoá - Chi phí: Không tốn tiền. Cái giá là bỏ mấy đơn thuê ngoài giá cao mà yêu cầu lại nói không rõ. Chỗ khó là bên kia thường trả nhiều hơn hẳn một đơn bình thường @@ -48,6 +44,7 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十七条之二、第三百零三条第二款）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2023). 建赌博网站并出售，涉案近 70 亿元. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 上海闵行区：办理一起新型开设网络赌场案. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 依法惩治帮助信息网络犯罪活动及相关犯罪典型案例（案例一）. - Ghi chú: 「明知」không cần bên kia nói ra mục đích sử dụng. Báo giá cao hơn hẳn giá thị trường, chức năng là nạp, rút và tỷ lệ thắng, khách không chịu nói dùng để làm gì, thì khi điều tra đều được dùng để suy ra rằng bạn biết rõ. Kỹ thuật viên phát triển ứng dụng cờ bạc bị xử lý theo tội đồng phạm của tội lập casino, không phải mức trần 3 năm của tội giúp tin. Tội giúp tin là tên gọi rút gọn của「帮助信息网络犯罪活动罪」. + ### 6. Rời công ty thì bàn giao xong quyền truy cập tài khoản, không xoá cơ sở dữ liệu, không để lại backdoor, không đổi mật khẩu để khoá hệ thống, dù công ty còn nợ bạn - Chi phí: Không tốn tiền. Công ty nợ bạn lương thì đi hoà giải lao động để đòi lại, đừng tự làm (hoà giải lao động xem phần 7). Chỗ khó là lúc nóng giận phải nhịn không đụng vào hệ thống @@ -56,7 +53,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款、第二百八十六条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2024). 检察机关「依法惩治网络犯罪 助力网络空间综合治理」典型案例（案例六）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例七）. - Ghi chú: Mật khẩu tài khoản do công ty cấp không có nghĩa sau khi nghỉ việc vẫn dùng được. Dùng tài khoản cũ để đăng nhập là「非法控制」. Thiệt hại tính theo chi phí khôi phục dữ liệu, mấy vạn tệ là đủ để khởi kiện. - ### 7. Rời công ty không mang mã nguồn, danh sách khách hàng, tài liệu kỹ thuật đi, không đăng lên ổ đĩa mạng cá nhân, không dùng lại ở công ty sau - Chi phí: Không tốn tiền. Trước khi nghỉ việc, xoá sạch tài liệu công ty trong máy tính cá nhân và trên ổ đĩa mạng. Chỗ khó là có tệp do mình tự tay viết từng dòng, không nỡ @@ -65,7 +61,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百一十九条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 反不正当竞争法（2025 年修订，第十条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关知识产权保护典型案例（案例七）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2020). 2019 年度检察机关保护知识产权典型案例（案例十六）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2025). 检察机关依法惩治侵害企业数据安全犯罪 服务保障企业创新发展（案例五）. - Ghi chú: 「自己写的代码」cũng thuộc quyền tác giả của công ty, điểm này xem mục 13 (mã nguồn viết bằng nguồn lực công ty thì thuộc công ty). Mang đi là lấy đồ của công ty. Danh sách khách hàng là thông tin kinh doanh, cũng là bí mật thương mại. Ngoài án tù, công ty còn có thể kiện bồi thường dân sự riêng. - ### 8. Không chạy chương trình của bạn trên máy tính, máy chủ, camera của người khác, không dùng máy của công ty để đào mỏ - Chi phí: Không tốn tiền, cũng không mất thời gian @@ -74,7 +69,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第二款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十九、六十六条；2016 年文本为第二十七、六十三条，终身禁业为修正后的第六十六条第三款）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). 山东滨州开发区：办理一起非法控制计算机信息系统案. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 充分发挥检察职能 推进网络空间治理典型案例（案例六）. - Ghi chú: Dùng máy chủ công ty để đào mỏ thì không tìm thấy vụ án nào được Viện kiểm sát nhân dân tối cao hay Tòa án nhân dân tối cao công bố, nên mục này viết theo điều luật về「kiểm soát hệ thống của người khác khi không được phép」. 「终身不得从事关键岗位」là hậu quả kèm theo mà Luật An ninh mạng đặt ra ngoài khoản phạt tiền và án tù, với người làm kỹ thuật thì nặng hơn khoản phạt tiền. - ### 9. Không có văn bản cho phép thì không thử hệ thống của người khác,「vì thiện ý」và「báo cáo sau đó」đều không phải lý do để thôi tội - Chi phí: Không tốn tiền. Trước khi tham gia SRC của nhà sản xuất hoặc chương trình thử nghiệm công khai, phải có văn bản ghi rõ phạm vi được phép, danh sách mục tiêu và khung thời gian được thử. SRC là trung tâm ứng cứu an ninh do chính nhà sản xuất lập, chuyên nhận lỗ hổng báo từ bên ngoài. Xác minh tới mức lỗ hổng đúng là tồn tại thì dừng. Chỗ khó là từ bỏ thói「thử luôn một cái cho vui」 @@ -83,6 +77,7 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百八十五条第一款、第二款）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2011). 关于办理危害计算机信息系统安全刑事案件应用法律若干问题的解释（第一条）. （Cục Công an thành phố Thâm Quyến đăng lại）；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2017). 第九批指导性案例（检例第 36 号，卫梦龙、龚旭、薛东东非法获取计算机信息系统数据案）. - Ghi chú: Bản án hướng dẫn số 36 vốn là vụ bán dữ liệu để hưởng lợi. Dẫn nó chỉ để nói câu「đăng nhập vượt quá phạm vi được phép cũng là xâm nhập」, không phải bảo thử nghiệm thiện ý sẽ bị kết án nặng như nhau. Ý bạn định làm gì, sau đó có báo cáo hay không, không xoá được tội danh, chỉ có thể ảnh hưởng tới mức phạt và việc có khởi tố hay không. Nhà sản xuất cảm ơn trước rồi báo việc sau, hai việc đó không mâu thuẫn, cảm ơn không phải là cho phép. Khi viết phần này, trên trang của Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao không tìm thấy vụ án thử nghiệm thiện ý nào đối chiếu từng chữ, nên chỉ viết theo điều luật và tiêu chuẩn xử phạt. Muốn làm thì đi theo SRC công khai của nhà sản xuất, hoặc theo hợp đồng đo thử bằng văn bản, ghi rõ phạm vi, mục tiêu và khung thời gian vào hồ sơ. Chứng minh lỗ hổng bằng lượng bằng chứng tối thiểu, không lấy dữ liệu hàng loạt. Mức không đủ thành tội phạm vẫn chịu xử phạt hành chính và cấm hành nghề, mức phạt tiền và thời hạn thì giống mục 8 (không chạy chương trình của bạn trên máy của người khác). Sau khi phát hiện lỗ hổng thì xử lý thế nào, xem mục 10 (báo lỗ hổng theo quy định). + ### 10. Báo lỗ hổng theo quy định, trước khi vá không công bố chi tiết, không đưa công cụ khai thác, không đưa ra nước ngoài - Chi phí: Không tốn tiền. Báo lỗ hổng cho nhà cung cấp sản phẩm, hoặc cho nền tảng chính thức. Trước khi nhà sản xuất vá xong, không đưa chi tiết, cũng không đưa mã dùng để kiểm chứng @@ -91,7 +86,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Bộ Công nghiệp và Thông tin (工业和信息化部), Văn phòng Quản lý mạng quốc gia (国家互联网信息办公室), Bộ Công an (公安部) (2021). 网络产品安全漏洞管理规定（工信部联网安〔2021〕66 号，第二、四、九、十、十四条，2021 年 9 月 1 日施行）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十八、六十五条；2016 年文本为第二十六、六十二条，规定第十四条援引的是 2016 年文本的条号）. - Ghi chú: Quy định ghi rõ cá nhân vào đối tượng bị quản lý,「tôi chỉ là người nghiệp dư」không nằm trong ngoại lệ. Gửi lỗ hổng của hệ thống trong nước lên nền tảng thưởng lỗ hổng ở nước ngoài thì rơi đúng vào câu「lỗ hổng chưa công bố không được cung cấp ra nước ngoài」. Có đủ tư cách để thử hay không, xem mục 9 (không có văn bản cho phép thì không thử hệ thống của người khác), đó là chuyện khác với sau khi có lỗ hổng thì xử lý thế nào. Lỗ hổng tìm ra hợp pháp vẫn có thể bị phạt vì cách công bố; lỗ hổng tìm ra khi chưa được phép thì cả khâu thử lẫn khâu công bố đều có vấn đề. Trong các kênh báo cáo, nền tảng của Bộ Công nghiệp và Thông tin và Trung tâm ứng cứu Internet quốc gia (CNCERT) được dùng nhiều nhất. Quy định không tách riêng trường hợp「đã báo nhà sản xuất mà nhà sản xuất không vá」, nên cách làm thực tế là để lại hồ sơ quá trình thương lượng và báo cáo, đồng thời báo cho nền tảng chính thức. - ### 11. Không bán công cụ vượt tường, không bán tài khoản VPN, không dựng loại nút mạng đó cho người khác - Chi phí: Không tốn tiền, cũng không mất thời gian @@ -100,7 +94,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: B - Nguồn: Quốc vụ viện (国务院) (1997). 计算机信息网络国际联网管理暂行规定（第六、十四条）. （Ủy ban Nông nghiệp và Nông thôn thành phố Trùng Khánh đăng lại）；Quốc hội Nhân dân (全国人大) (2020). 刑法（根据刑法修正案（十一）修正，第二百二十五条第四项）. - Ghi chú: Khi viết phần này, trên trang của Tòa án nhân dân tối cao và Viện kiểm sát nhân dân tối cao không tìm thấy vụ án vượt tường nào được công bố, cách xử lý ở các địa phương cũng khác nhau khá nhiều, nên để mức B. Cá nhân tự dùng thì mức phạt nặng nhất là cảnh cáo cộng phạt tiền quy định ở khoản 14 của Quy định tạm thời. - ### 12. Đã ký thỏa thuận không cạnh tranh, sau khi nghỉ việc công ty không trả bồi thường hằng tháng thì gửi văn bản yêu cầu, đủ 3 tháng không trả thì có thể chấm dứt; vị trí chưa từng tiếp cận bí mật thương mại thì có thể yêu cầu xác nhận điều khoản không có hiệu lực - Chi phí: Không tốn tiền. Khi nghỉ việc, giữ lại bản gốc thỏa thuận không cạnh tranh, cùng bằng chứng công ty không trả tiền hằng tháng @@ -109,7 +102,6 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). 劳动合同法（2012 年修正，第二十三、二十四条）. （Cục Quản lý ứng cứu thành phố Phụ Thuận đăng lại）；Tòa án nhân dân tối cao (最高人民法院) (2020). 关于审理劳动争议案件适用法律问题的解释（一）（第三十六至四十条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 关于审理劳动争议案件适用法律问题的解释（二）（法释〔2025〕12 号，第十三至十五条）. ；Tòa án nhân dân tối cao (最高人民法院) (2025). 劳动争议典型案例（案例四）. - Ghi chú: Ngược lại, thỏa thuận có hiệu lực, công ty cũng đã trả tiền hằng tháng, mà bạn vẫn đi làm cho công ty cạnh tranh thì phải nộp phạt vi phạm theo điều khoản đã thoả thuận. Nộp xong còn có thể bị yêu cầu giữ hết phần thời gian không cạnh tranh còn lại. Thỏa thuận ký trong thời gian còn đi làm mà công ty không trả thêm tiền thì vẫn có hiệu lực. - ### 13. Mã nguồn viết trong giờ làm việc hoặc bằng nguồn lực công ty thì thuộc công ty, dự án mã nguồn mở của bạn thì tự làm bằng thời gian và thiết bị của bạn, không trộn mã công ty - Chi phí: Không tốn tiền. Khi nhận việc, đọc kỹ điều khoản sở hữu trí tuệ trong hợp đồng lao động. Chỗ khó là dự án cá nhân và việc của công ty thường nằm chung một máy, phải tách ra @@ -118,16 +110,14 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 著作权法（2020 年修正，第十八条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 专利法（2020 年修正，第六条）. - Ghi chú: Phần mềm bị xếp thẳng vào loại「quyền tác giả thuộc đơn vị」, lập trình viên hầu như không rơi vào loại「quyền tác giả thuộc tác giả, đơn vị ưu tiên sử dụng trong hai năm」. 「Viết sau giờ làm」mà có dùng máy tính công ty, mạng nội bộ hoặc dữ liệu nghiệp vụ công ty thì vẫn có thể bị tính là chủ yếu làm ra bằng thiết bị và tài liệu của đơn vị. Muốn làm sạch dự án cá nhân thì dùng máy của mình, không đụng kho mã công ty, không trùng với công việc công ty. - -### 14. Làm sản phẩm bằng mã nguồn mở như GPL, thì mở mã theo yêu cầu giấy phép hoặc đổi giấy phép, đừng coi là「không ai quản」 +### 14. Làm sản phẩm bằng mã nguồn mở như GPL, thì mở theo yêu cầu của giấy phép hoặc đổi giấy phép, đừng coi là「không ai quản」 -- Chi phí: Không tốn tiền. Trước khi dùng mã nguồn mở của người khác, xem giấy phép của nó (điều kiện sử dụng mà tác giả viết ra). Họ GPL yêu cầu sản phẩm có dùng nó cũng phải mở mã nguồn ra, xung đột với việc đóng mã rồi bán tiền, xung đột thì đổi. Chỗ khó là từ bỏ một thư viện đã dùng quen -- Nói thẳng: Xâm phạm quyền tác giả thì bồi thường theo thiệt hại của bên kia hoặc theo số tiền bạn kiếm được, cố ý và đặc biệt nghiêm trọng thì có thể bị phạt tới mức 1 đến 5 lần. Có công ty thuê nhân viên cũ của đối thủ, làm ra phần mềm về cơ bản giống nhau, rồi biện rằng phần mềm của bên kia dùng GPLv2, lẽ ra phải mở mã. Tòa án nhân dân tối cao cho rằng việc bên kia có mở mã hay không không ảnh hưởng tới quyền tác giả của nó, vẫn kết án buộc ngừng xâm phạm, bồi thường 50 vạn tệ. -- Lợi ích: Xâm phạm quyền tác giả thì bồi thường theo thiệt hại thực tế của chủ sở hữu quyền, hoặc theo số tiền người xâm phạm kiếm được. Cả hai đều không tính được thì căn cứ số tiền phải trả nếu mua bình thường giấy phép này. Cố ý xâm phạm và đặc biệt nghiêm trọng thì có thể bị phạt bồi thường 1 đến 5 lần. Làm tổn hại lợi ích công cộng thì còn có thể tịch thu thu nhập bất hợp pháp và phạt tiền. Vụ án chính thức: (2021) 最高法知民终 51 号. Bên cáo tuyển nhân viên cũ của nguyên đơn, phần mềm cổng phát triển ra về cơ bản giống phần mềm của nguyên đơn. Bên cáo biện bạo rằng phần mềm của nguyên đơn dựa trên GPLv2, vốn đã phải mở mã. Tòa án nhân dân tối cao xác định「việc bản thân nhà phát triển có vi phạm thỏa thuận GPLv2 hay không và việc có sở hữu quyền tác giả phần mềm là hai vấn đề độc lập」, kết án buộc ngừng xâm phạm, bồi thường 50 vạn tệ (toàn quốc) +- Chi phí: Không tốn tiền. Trước khi dùng mã nguồn mở của người khác, nhìn giấy phép của nó (điều kiện sử dụng tác giả ghi ra). Dòng GPL yêu cầu sản phẩm dùng nó cũng phải mở mã nguồn, xung đột với việc đóng mã nguồn rồi bán, xung đột thì đổi. Khó ở chỗ phải bỏ một thư viện đã quen dùng. +- Nói thẳng: Xâm phạm quyền tác giả thì bồi thường theo thiệt hại của bên bị xâm phạm hoặc theo số tiền bạn thu được, cố ý và tình tiết nghiêm trọng thì có thể bị phán đến gấp 1 đến 5 lần. Có công ty tuyển cựu nhân viên của đối thủ, làm ra phần mềm tương đương về nội dung, biện bạn rằng phần mềm của đối phương dùng GPLv2, vốn đã phải mở. Tòa án nhân dân tối cao cho rằng đối phương có mở mã nguồn hay không không ảnh hưởng tới quyền tác giả của nó, vẫn phán phải dừng vi phạm, bồi thường 500.000 tệ. +- Lợi ích: Xâm phạm quyền tác giả thì bồi thường theo thiệt hại thực tế của chủ quyền, hoặc theo số tiền người vi phạm thu được. Cả hai đều không tính được thì căn cứ số tiền phải trả để mua giấy phép này bình thường. Cố ý vi phạm mà tình tiết nghiêm trọng thì có thể bị phán bồi thường gấp 1 đến 5 lần. Còn gây hại lợi ích công cộng thì có thể bị tịch thu thu nhập bất hợp pháp và phạt tiền. Ca thực tế chính thức: (2021) số 51 của Tòa án nhân dân tối cao về quyền sở hữu trí tuệ. Bị cáo tuyển cựu nhân viên của nguyên đơn, phần mềm cổng giao hàng bị cáo phát triển tương đương về nội dung với phần mềm của nguyên đơn. Bị cáo biện bảo rằng phần mềm của nguyên đơn dựa trên GPLv2, vốn đã phải mở. Tòa án nhân dân tối cao xác định「开发者自身是否违反 GPLv2 协议和是否享有软件著作权是两个独立问题」, phán phải dừng vi phạm, bồi thường 500.000 tệ (toàn quốc) - Mức bằng chứng: B -- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 著作权法（2020 年修正，第五十三、五十四条）. ；Tòa chuyên môn về sở hữu trí tuệ của Tòa án nhân dân tối cao (最高人民法院知识产权法庭) (2022). 涉「开源软件」著作权侵权案（（2021）最高法知民终 51 号）. -- Ghi chú: Vụ án này cho thấy「bên kia dùng mã GPL」không thể dùng làm lá chắn khi bạn sao chép. Ngược lại, tác giả GPL đã kiện bên thương mại đóng mã ở trong nước và đã có bản án có hiệu lực, phần này không tìm thấy văn bản trên trang của Tòa án nhân dân tối cao nên không dẫn. Để mức B là vì có tuân thủ giấy phép mã nguồn mở hay không chỉ dựa được vào hợp đồng và các quy định chung trong luật bản quyền, không có điều luật riêng nào. - +- Nguồn: Ủy ban Thường vụ Quốc hội (2020). 著作权法（2020 年修正，第五十三、五十四条）. ；Tòa xử lý quyền sở hữu trí tuệ Tòa án nhân dân tối cao (2022). 涉「开源软件」著作权侵权案（（2021）最高法知民终 51 号）. +- Ghi chú: Vụ án này cho thấy「đối phương dùng mã GPL」không phải là lý do để bạn sao chép nó. Ngược lại, tác giả GPL kiện bên thương mại dùng mã nguồn đóng đã có phán quyết có hiệu lực trong nước, phần này không tìm thấy nguyên văn trên cổng Tòa án nhân dân tối cao nên không dẫn. Để ở mức B vì có tuân theo giấy phép mã nguồn mở hay không chỉ dựa vào hợp đồng và những quy định chung trong luật bản quyền, không có điều khoản chuyên biệt nào. ### 15. Ứng dụng chỉ thu thập thông tin cá nhân cần thiết cho nghiệp vụ, dưới 14 tuổi phải có sự đồng ý của người giám hộ, chính sách bảo mật phải viết rõ - Chi phí: Không tốn tiền. Trước khi đưa lên, đối chiếu danh sách quyền truy cập mà ứng dụng xin với chính sách bảo mật, quyền không cần thì đừng xin. Chỗ khó là bên sản phẩm luôn muốn giữ lại thêm vài dữ liệu「có thể sau này cần dùng」 @@ -136,15 +126,15 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Mức bằng chứng: B - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). 个人信息保护法（第六、十三、三十一、六十六条）. - Ghi chú: 「直接责任人员」trong thực tế thường rơi vào người viết mã và làm sản phẩm. Bộ Công nghiệp và Thông tin định kỳ công bố danh sách gỡ ứng dụng vi phạm, phần này không đối chiếu được đợt công bố cụ thể nào nên không dẫn. -### 16. Website, ứng dụng trước khi đưa lên phải đăng ký ICP, giữ nhật ký tối thiểu 6 tháng theo yêu cầu bảo vệ theo cấp độ + +### 16. Trước khi website, ứng dụng lên mạng thì làm đăng ký ICP, giữ nhật ký trên 6 tháng theo yêu cầu phân cấp bảo vệ -- Chi phí: Đăng ký miễn phí, đi theo quy trình của nhà cung cấp điện toán đám mây mất khoảng mấy tuần. Đánh giá bảo vệ theo cấp độ thì thu phí theo cấp. Bảo vệ theo cấp độ gọi tắt là đánh giá cấp độ, là một bộ biện pháp quản lý an toàn mà Nhà nước phân loại theo mức quan trọng của hệ thống và yêu cầu website, hệ thống phải thực hiện; website nhỏ thường không cần đánh giá -- Nói thẳng: Đăng ký không mất tiền, đi quy trình của nhà cung cấp đám mây mất khoảng mấy tuần. Máy chủ trong nước mà không đăng ký thì tên miền không phân giải được, không sửa thì bị đóng website. Không ghi số đăng ký trên trang chủ thì phạt tiền 5000 tệ đến 5 vạn tệ. Đăng ký nói không thu phí mà thực tế lại thu tiền thì phạt tiền 10 vạn đến 100 vạn tệ. Nhật ký phải giữ đủ 6 tháng, không sửa hoặc đã để lại hậu quả thì phạt tiền 1 vạn đến 10 vạn tệ. -- Lợi ích: Dịch vụ thông tin Internet có tính thương mại thì áp dụng giấy phép, không thương mại thì áp dụng đăng ký. Chưa có giấy phép hoặc chưa đăng ký thì không được làm. Chưa đăng ký thì cơ quan quản lý viễn thông yêu cầu sửa, không sửa thì yêu cầu đóng website. Kinh doanh mà chưa có giấy phép thì tịch thu thu nhập bất hợp pháp, phạt tiền bằng 3 đến 5 lần thu nhập bất hợp pháp, hoặc phạt tiền 10 vạn đến 100 vạn tệ. Trang chủ website không ghi số đăng ký thì phạt tiền 5000 tệ đến 5 vạn tệ. Đơn vị vận hành mạng phải theo hệ thống bảo vệ theo cấp độ (đánh giá cấp độ) để xây dựng quy chế quản lý an toàn, phòng chống tấn công, giữ nhật ký mạng không dưới 6 tháng, sao lưu và mã hoá dữ liệu theo phân loại. Không thực hiện thì yêu cầu sửa, đưa ra cảnh cáo. Không sửa hoặc đã gây hậu quả thì phạt tiền 1 vạn đến 10 vạn tệ, người phụ trách phạt tiền 5000 tệ đến 5 vạn tệ (toàn quốc) +- Chi phí: Đăng ký miễn phí, đi theo quy trình của nhà cung cấp đám mây thì khoảng mấy tuần. Đo kiểm phân cấp bảo vệ thì tính phí theo cấp. Phân cấp bảo vệ gọi tắt là等保, là bộ biện pháp quản lý an toàn do nhà nước phân cấp theo mức độ quan trọng của hệ thống, yêu cầu website và hệ thống thực hiện, website nhỏ thường không phải đo kiểm. +- Nói thẳng: Đăng ký không mất tiền, đi theo quy trình của nhà cung cấp đám mây thì khoảng mấy tuần, máy chủ trong nước không đăng ký thì bị đóng trang web. Đăng ký nói không thu phí mà thật ra có thu tiền thì phạt từ 100.000 đến 1.000.000 tệ. Nhật ký phải giữ đủ 6 tháng, không chịu sửa hoặc để xảy ra hậu quả thì phạt từ 10.000 đến 100.000 tệ. +- Lợi ích: Dịch vụ thông tin internet có tính thương mại thì phải xin phép, không tính thương mại thì phải đăng ký. Không có phép, không đăng ký thì không được làm. Không đăng ký thì cơ quan quản lý viễn thông buộc phải sửa, không sửa thì buộc đóng trang web. Kinh doanh không được phép thì bị tịch thu thu nhập bất hợp pháp, và phạt bằng 3 đến 5 lần thu nhập bất hợp pháp, hoặc phạt từ 100.000 đến 1.000.000 tệ. Trang chủ website không ghi số đăng ký thì phạt từ 5.000 đến 50.000 tệ. Đơn vị vận hành mạng phải theo hệ thống phân cấp bảo vệ (等保) để ra quy chế quản lý an toàn, phòng chống tấn công, lưu nhật ký mạng không dưới 6 tháng, sao lưu và mã hoá dữ liệu theo phân loại. Không thực hiện thì bị buộc phải sửa, bị cảnh cáo. Không sửa hoặc để xảy ra hậu quả thì phạt từ 10.000 đến 100.000 tệ, người phụ trách phạt từ 5.000 đến 50.000 tệ (toàn quốc) - Mức bằng chứng: A -- Nguồn: Quốc vụ viện (国务院) (2011). 互联网信息服务管理办法（2011 年修订，第四、十九、二十二条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十三、六十一条；2016 年文本为第二十一、五十九条）. -- Ghi chú: Máy chủ trong nước mà không đăng ký thì tên miền không phân giải được ngay. Có hai cái hố thường gặp: đăng ký là cá nhân phi thương mại nhưng thực tế lại thu tiền; đổi tên miền hoặc đổi chủ thể mà không đi làm thủ tục thay đổi đăng ký. Giữ nhật ký 6 tháng vừa là nghĩa vụ vừa là cách tự bảo vệ, khi có chuyện thì chứng minh được không phải mình làm. - +- Nguồn: Hội đồng Nhà nước (2011). 互联网信息服务管理办法（2011 年修订，第四、十九、二十二条）. ；Ủy ban Thường vụ Quốc hội (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十三、六十一条；2016 年文本为第二十一、五十九条）. +- Ghi chú: Máy chủ trong nước không đăng ký thì tên miền cơ bản không phân giải được. Đăng ký thì làm thế nào, không gắn số đăng ký bị phạt bao nhiêu, làm sao biết nhà cung cấp dịch vụ có chứng chỉ hay không, xem mục 4 ở phần 26 (đặt máy chủ trong nước thì phải đăng ký). Hai cái bẫy hay gặp: đăng ký là cá nhân không kinh doanh nhưng thực tế lại có thu tiền; đổi tên miền hoặc đổi chủ thể mà không đi làm thay đổi đăng ký. Giữ nhật ký 6 tháng vừa là nghĩa vụ vừa là tự bảo vệ, khi có sự cố thì chứng minh được việc đó không phải do mình. ### 17. Trước khi cung cấp dịch vụ AI tạo sinh cho công chúng, phải đánh giá an toàn và đăng ký thuật toán theo quy định, nội dung tạo ra phải gắn nhãn - Chi phí: Không tốn tiền, nhưng phải đi quy trình của cơ quan quản lý mạng, xin xong mất một khoảng thời gian. Công cụ nhỏ do cá nhân làm, chỉ cần không mở cho công chúng thì không kích hoạt bộ quy định này @@ -152,4 +142,4 @@ Phần này tính hai cuốn sổ: tự do cá nhân và tiền, đổi lại l - Lợi ích: Cung cấp dịch vụ trí tuệ nhân tạo tạo sinh có tính chất dư luận hoặc có khả năng vận động xã hội thì phải tiến hành đánh giá an toàn theo quy định của Nhà nước, đăng ký thuật toán theo quy định quản lý thuật toán đề xuất, cùng các thủ tục thay đổi, huỷ bỏ về sau. Tính chất dư luận hoặc khả năng vận động xã hội là khả năng ảnh hưởng tới nhận thức của công chúng, hoặc khả năng tổ chức vận động người ta. Bên cung cấp phải gắn nhãn cho nội dung do mô hình tạo ra như ảnh, video. Biện pháp này có hiệu lực từ ngày 15 tháng 8 năm 2023 (toàn quốc) - Mức bằng chứng: B - Nguồn: Văn phòng Quản lý mạng quốc gia (国家互联网信息办公室) và sáu cơ quan khác (2023). 生成式人工智能服务管理暂行办法（第十二、十七条）. -- Ghi chú: Biện pháp này không tự ghi mức phạt tiền là bao nhiêu. Vi phạm thì phải quay lại Luật An ninh mạng, Luật An toàn dữ liệu, Luật Bảo vệ Thông tin cá nhân để tìm mức phạt, nên để mức B. Vụ án dân sự về nội dung do AI tạo ra xâm phạm quyền tác giả, quyền hình ảnh của người khác, phần này không đối chiếu được văn bản trên trang của Tòa án nhân dân tối cao nên không dẫn. \ No newline at end of file +- Ghi chú: Biện pháp này không tự ghi mức phạt tiền là bao nhiêu. Vi phạm thì phải quay lại Luật An ninh mạng, Luật An toàn dữ liệu, Luật Bảo vệ Thông tin cá nhân để tìm mức phạt, nên để mức B. Vụ án dân sự về nội dung do AI tạo ra xâm phạm quyền tác giả, quyền hình ảnh của người khác, phần này không đối chiếu được văn bản trên trang của Tòa án nhân dân tối cao nên không dẫn. diff --git "a/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" "b/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" index ac321ff8..a4158acc 100644 --- "a/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" +++ "b/vi/book/12-\345\210\233\344\270\232\344\270\216\345\201\232\347\224\237\346\204\217.md" @@ -12,7 +12,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: B - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第五十六、一千零六十四条）. ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Luật Công ty, 公司法（2023 年修订，第四十七、五十四条）. - Ghi chú: 「亏得起」 nghĩa là: mất trắng số tiền đó, bạn cũng không phải bán nhà, không phải vay tiền người thân để lấp lỗ hổng. Có bao nhiêu doanh nghiệp sống sót, trung bình sống được mấy năm, loại thống kê chính thức đó lần này không tìm được bản gốc để đối chiếu trên trang Cục Thống kê Quốc gia và Cục Quản lý thị trường quốc gia. Vì vậy phần chính văn không ghi con số, chỉ ghi hậu quả pháp lý, để ở mức B. - ### 2. Đừng ký bảo đảm cá nhân cho khoản vay của công ty, người phối ngẫu càng đừng ký theo - Chi phí: Không tốn tiền. Khi ngân hàng, nhà cung cấp hay chủ nhà bảo bạn ký「bảo đảm liên đới cá nhân của cổ phân」hoặc đòi「chữ ký của người phối ngẫu」, hãy dừng lại. Ký trang đó có nghĩa là bạn tự tay trao đi lớp bảo vệ「nợ của công ty không tìm đến cá nhân」(từ bỏ trách nhiệm hữu hạn). Không ký được thì đừng ký. Bắt buộc phải ký thì chỉ ký「bảo đảm chung」, và ghi rõ số tiền nhiều nhất bạn bảo đảm. Khó nhất là bên kia hay nói「không ký thì không cho vay」, bạn phải chịu được. @@ -21,7 +20,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第六百八十一、六百八十七、六百八十八、一千零六十四条）. - Ghi chú: Ngân hàng cho vay doanh nghiệp nhỏ và siêu nhỏ gần như lúc nào cũng bắt cổ phân công ty ký bảo đảm liên đới. Vẫn vay được như thường, chỉ là trước khi ký phải hiểu rõ trang đó trao đi cái gì, rồi giữ số tiền bảo đảm trong mức「lỡ mất trắng」của mục 1 (chỉ dùng số tiền mà lỡ mất trắng vẫn chịu được để khởi nghiệp). Quy tắc chung về phiếu vay và giấy bảo đảm, xem mục 18 ở phần 8. - ### 3. Chọn đúng loại hình kinh doanh trước khi mở hàng: hộ cá thể và cổ phần công chung phải trả đến cùng, chỉ công ty hữu hạn mới「hữu hạn」 - Chi phí: Đăng ký làm online trên cổng dịch vụ công của địa phương, mỗi lần mất mấy tiếng. Sau khi lập công ty hữu hạn thì hằng tháng phải làm sổ sách và kê khai thuế, đây là khoản chi phí lâu dài: tự làm thì tốn thời gian, thuê người làm thay thì tốn tiền, giá ở mỗi nơi khác nhau. @@ -30,7 +28,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第五十六条）. ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2006). Luật Doanh nghiệp hợp tác, 合伙企业法（2006 年修订，第二条）. ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Luật Công ty, 公司法（2023 年修订，第四、二十三、四十七、五十三、五十四条）. - Ghi chú: Điều kiện của chữ「hữu hạn」là công và tư phải phân minh: tài khoản công ty và tài khoản cá nhân không dùng chung, không dùng tiền công ty trả chi phí trong nhà, sổ sách giữ đầy đủ. Vốn đăng ký đừng điền số lớn cho đẹp mắt. Cam kết góp bao nhiêu thì trong 5 năm phải thật sự đưa ra bấy nhiêu. Điền 1.000.000 tệ là tạo ra một khoản nợ tiềm tàng 1.000.000 tệ. - ### 4. Đừng làm cổ phân danh nghĩa, đừng giữ cổ phần thay người khác - Chi phí: Không tốn tiền. Có người nói「cho mượn giấy tờ tùy thân để làm cổ phân」hay「cổ phần tạm đứng tên bạn」, từ chối hết. @@ -39,7 +36,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023). Luật Công ty, 公司法（2023 年修订，第四十七、四十九、五十、五十四条）. ; Tòa án nhân dân tối cao (最高人民法院) (2020 修正). 关于适用《中华人民共和国公司法》若干问题的规定（三）（法释〔2011〕3 号，2014 年法释〔2014〕2 号、2020 年法释〔2020〕18 号两次修正，第二十四、二十五条）：「如无法律规定的无效情形，人民法院应当认定该合同有效」「实际出资人未经公司其他股东半数以上同意，请求公司变更股东……人民法院不予支持」「名义股东处分股权造成实际出资人损失，实际出资人请求名义股东承担赔偿责任的，人民法院应予支持」— Nói thẳng: hợp đồng có hiệu lực trừ khi luật có lý do vô hiệu; yêu cầu đổi đăng ký cổ phân mà không được hơn một nửa cổ phân khác đồng ý thì tòa không hỗ trợ; cổ phân danh nghĩa chuyển nhượng cổ phần gây thiệt hại cho nhà đầu tư thật thì tòa hỗ trợ yêu cầu bồi thường của nhà đầu tư thật với cổ phân danh nghĩa. （Cơ sở dữ liệu luật và quy định quốc gia, 国家法律法规数据库）; Tòa án nhân dân tối cao (最高人民法院) (2025). 关于适用《中华人民共和国公司法》若干问题的解释（征求意见稿）第九十条. - Ghi chú: Giữa cổ phân danh nghĩa và người thật sự bỏ tiền, thỏa thuận giữ hộ có hiệu lực nếu không rơi vào lý do vô hiệu. Nhưng người thật sự bỏ tiền muốn đổi tên đăng ký sang mình thì phải được hơn một nửa cổ phân khác đồng ý. Người đứng tên danh nghĩa bán hoặc thế chấp cổ phần thì tòa xử lý theo quy định thiện chí của Bộ luật Dân sự. Bên mua không biết, giá hợp lý, đã làm thủ tục đăng ký thì người bỏ tiền không lấy lại được cổ phần, chỉ có thể đòi người đứng tên danh nghĩa bồi thường. Tháng 9 năm 2025 đã có bản dự thảo giải thích mới thay cho bản giải thích tư pháp này, bản chính thức có hiệu lực thì toàn văn bị bãi bỏ, lấy bản sửa đổi năm 2020 còn hiệu lực tính đến tháng 9 năm 2026 làm chuẩn. Rủi ro làm đại diện pháp lý danh nghĩa, xem mục 28 ở phần 8. - ### 5. Trước khi nhượng quyền, kiểm tra hồ sơ đăng ký ở Bộ Thương mại, lấy bản công bố bằng văn bản, ghi「thời gian cân nhắc」vào hợp đồng - Chi phí: Không tốn tiền. Vào website「商业特许经营信息管理系统」của Bộ Thương mại, kiểm tra bên thương hiệu đã đăng ký hay chưa. Yêu cầu bên kia viết ra giấy tờ tất cả những điều phải nói trước khi ký, đưa bạn cùng bản hợp đồng. Mất mấy tiếng là xong. @@ -57,7 +53,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2021). Quy định về đăng ký chủ thể kinh doanh（市场主体登记管理条例）（国务院令第 746 号，第十、十三、十四、十六、十九、二十一条）. ; Cục Quản lý thị trường quốc gia (市场监管总局) (2022). Quy tắc chi tiết về thi hành Quy định đăng ký chủ thể kinh doanh（市场监管总局令第 52 号，第十二、二十五、二十六条）. ; Bộ Tài chính, Ủy ban Cải cách và Phát triển quốc gia (财政部、国家发展改革委) (2014). Thông báo về việc bãi bỏ, dừng thu và miễn thu một loạt khoản thu dịch vụ hành chính công（财税〔2014〕101 号，附件 1）. - Ghi chú: Bản thân việc đăng ký không thu phí, khoản bị thu phần lớn là phí chạy việc của bên làm thay. Giá khắc con dấu do thị trường quyết định. Văn phòng Quốc vụ viện ra lệnh「严禁指定公章制作单位制作公章」, nên có người nói「bắt buộc phải đến chỗ được chỉ định để khắc con dấu」thì bạn không cần phải nghe. Vốn đăng ký điền bao nhiêu không ảnh hưởng tới mặt tiền, chỉ quyết định giới hạn trách nhiệm của bạn, và phải nộp đủ trong 5 năm, xem mục 3 của phần này (chọn đúng loại chủ thể trước khi khai trương). Phạm vi kinh doanh có mục cần giấy phép thì chưa có giấy thì không được khai trương, xem mục 7 của phần này (ngành cần giấy phép). - ### 7. Ngành cần giấy phép thì chưa có giấy thì không khai trương - Chi phí: Không tốn tiền. Trước khi khai trương, tra rõ ngành bạn định làm có phải xin giấy phép không. Làm ăn uống, bán thực phẩm thì phải xin giấy phép kinh doanh thực phẩm. Chỉ bán thực phẩm đóng gói sẵn do người khác làm thì không cần giấy phép, chuyển sang kê khai. Có giấy rồi hãy mở cửa kinh doanh. @@ -66,7 +61,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2017). Biện pháp xử lý kinh doanh không phép không giấy phép（无证无照经营查处办法）（国务院令第 684 号，第五、六、十三条）. ；Cục Quản lý thị trường quốc gia (市场监管总局) (2023). Biện pháp quản lý giấy phép và kê khai kinh doanh thực phẩm（市场监管总局令第 78 号，第四条）. ；Đại biểu Quốc hội (全国人大) (1997). Bộ luật Hình sự（第二百二十五条）. - Ghi chú: Điều kiện cấp giấy phép trong các lĩnh vực tài chính, y tế, giáo dục đào tạo có luật riêng, mục này chỉ đối chiếu quy tắc xử lý chung và điều khoản của Bộ luật Hình sự. Cửa hàng trà sữa, quán ăn nhỏ thuộc dịch vụ ăn uống, phải xin giấy phép kinh doanh thực phẩm. Trang trí xong mới xin giấy, trước khi giấy về đã mở bán thử, là chỗ vấp thường gặp. Ngành thực phẩm chia làm mấy bậc, bán thịt rau tươi có phải xin giấy không, xem mục 8 (làm thực phẩm thì xem trước mình ở bậc nào). - ### 8. Làm thực phẩm thì xem trước mình ở bậc nào: sản xuất và ăn uống phải có giấy phép, chỉ bán đóng gói sẵn thì kê khai, bán thịt rau tươi không cần giấy - Chi phí: Bản thân giấy phép không thu tiền. Nhưng cơ sở và thiết bị phải đạt, kiểm tra thực tế cộng với thủ tục xin giấy mất mấy tuần. Bán thịt thì còn phải giữ giấy kiểm dịch và giấy xác nhận chất lượng thịt cho từng lô. @@ -75,7 +69,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2025, sửa đổi lần thứ ba). Luật An toàn thực phẩm（食品安全法）（第三十五、三十六、一百二十二条）. ；Quốc vụ viện (国务院) (2021). Quy định về quản lý giết lợn（生猪屠宰管理条例）（国务院令第 742 号，第二、二十三、三十一条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2021, sửa đổi). Luật Phòng chống dịch bệnh động vật（动物防疫法）（第五十一、一百条）. ；Tòa án nhân dân tối cao (最高人民法院) (2024). Giải thích về một số vấn đề pháp lý khi xét các vụ án tranh chấp về bồi thường mang tính răn đe trong lĩnh vực dược phẩm và thực phẩm（法释〔2024〕9 号，第四条）. - Ghi chú: Ngưỡng của xưởng nhỏ, nhà hàng nhỏ, rong quán thực phẩm do từng tỉnh đặt, mục này không ghi số cụ thể. Vào trang Cục Quản lý thị trường tỉnh bạn, tìm「食品小作坊小餐饮小摊贩管理条例」, xem bản của địa phương. Bán thực phẩm tự làm trên mạng không phải vì「trên mạng」mà thiếu một giấy phép, chỗ nào cần giấy phép thì vẫn phải có. Ngoài giấy phép ra còn có quy tắc chung của phần này: chủ thể và giấy phép xem ở mục 6 của phần này (chốt tên, địa điểm, phạm vi kinh doanh trước khi đăng ký), ngành khác có cần giấy phép không xem ở mục 7 của phần này (ngành cần giấy phép). Người mua có thể đòi bồi thường bao nhiêu xem ở mục 31 ở phần 5 (bồi thường gấp mười cho thực phẩm). - ### 9. Đóng túi để bán thì là thực phẩm đóng gói sẵn: trên nhãn, ngày sản xuất, hạn sử dụng, bảng thành phần thiếu mục nào cũng không được - Chi phí: In nhãn mỗi nhãn vài xu đến vài chục xu. Mỗi lô đều phải điền đúng ngày và thành phần, tốn một chút thời gian. @@ -84,6 +77,7 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2025, sửa đổi lần thứ ba). Luật An toàn thực phẩm（食品安全法）（第六十七、六十八、一百二十五条）. ；Ủy ban Quốc gia Y tế, Cục Quản lý thị trường quốc gia (国家卫生健康委员会、国家市场监督管理总局) (2025). Thông báo công bố《Quy tắc chung nhãn thực phẩm đóng gói sẵn》(GB 7718-2025) cùng 50 tiêu chuẩn an toàn thực phẩm và 9 văn bản sửa đổi（2025 年第 2 号）. （Ủy ban Y tế thành phố Lạc Sa trích lại toàn văn thông báo）；Tòa án nhân dân tối cao (最高人民法院) (2024). Giải thích về một số vấn đề pháp lý khi xét các vụ án tranh chấp về bồi thường mang tính răn đe trong lĩnh vực dược phẩm và thực phẩm（法释〔2024〕9 号，第六、八条）. （cách phân bậc giữa khiếm khuyết nhãn và thiếu mục bắt buộc）；Cục Quản lý thị trường thành phố Bắc Kinh (北京市市场监督管理局) (2026). Không tìm thấy ngày sản xuất? Nhãn thực phẩm sắp「thay mặt」! Những điểm chính trong quy định mới này. （ngày áp dụng, thời gian chuyển tiếp và yêu cầu mới về trình bày ngày） - Ghi chú: Lỗi nhãn chia làm hai loại, phạt bao nhiêu, bồi thường bao nhiêu đều theo cách chia này. Một là lỗi nhỏ, như chính tả, cỡ chữ không quy chuẩn, không ảnh hưởng an toàn thực phẩm cũng không khiến người ta hiểu nhầm. Loại này cơ quan quản lý thị trường chỉ yêu cầu sửa, không sửa mới phạt dưới 2000 tệ, người mua cũng không đòi được bồi thường gấp mười. Loại kia là thiếu mục bắt buộc, như bỏ ngày sản xuất, hạn sử dụng. Giải thích tư pháp nói rõ đây không tính là lỗi nhỏ: phạt tiền tính theo giá trị lô hàng, dưới 1 vạn tệ thì phạt 5000 tệ đến 5 vạn tệ, từ 1 vạn tệ trở lên thì phạt gấp 5 đến 10 lần giá trị lô hàng. Người mua còn có thể đòi bạn bồi thường gấp mười giá thực phẩm, dưới 1000 tệ thì tính bằng 1000 tệ (xem mục 31 ở phần 5, bồi thường gấp mười cho thực phẩm). Yêu cầu chi tiết từng điều của GB 7718-2025, mục này mới chỉ đối chiếu được ngày và cách trình bày trong thông báo và bài giải thích chính thức, thể thức cụ thể theo nguyên văn tiêu chuẩn. + ### 10. Thực phẩm thường không được nói là chữa bệnh: nhãn, hướng dẫn, quảng cáo và lời livestream đều tính - Chi phí: Không tốn tiền. Rà lại câu quảng cáo, trang giới thiệu sản phẩm và lời livestream. Khó ở chỗ nhịn không nói những câu như「hạ đường huyết, bao tiêu」「chữa táo bón」. @@ -92,7 +86,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2025, sửa đổi lần thứ ba). Luật An toàn thực phẩm（食品安全法）（第七十一、七十三、一百二十五、一百四十条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2021, sửa đổi). Luật Quảng cáo（广告法）（第十七、五十八条）. - Ghi chú:「Chức năng bảo vệ sức khoẻ」cũng không phải thứ thực phẩm thường được nói. Chức năng bảo vệ sức kho�è mà thực phẩm chức năng được tuyên bố chỉ được lấy từ danh mục nhà nước quy định, sản phẩm còn phải qua đăng ký hoặc kê khai, thực phẩm thường không có bước này. Người mua nhận ra loại lời này thế nào, xem mục 10 ở phần 6 (đừng bỏ nhiều tiền mua thực phẩm bảo vệ sức khoẻ để điều chỉnh cơ thể). Ngoài bao bì và quảng cáo, trang giới thiệu sản phẩm, lời livestream, văn bản đăng trên mạng xã hội cũng là nội dung quảng cáo, đăng ra là để lại dấu vết. - ### 11. Ngành thực phẩm có đường ranh giới hình sự: bán thịt chết bệnh, hàng vượt tiêu chuẩn là đủ thành tội, trộn thêm nguyên liệu độc hại thì không tính tiền, khởi tội năm năm - Chi phí: Không tốn tiền. Làm theo bộ mục 8 của phần này (làm thực phẩm thì xem trước mình ở bậc nào): giữ giấy kiểm dịch và chứng từ nhập hàng, bảng thành phần chỉ đưa loại có tra được số hiệu phê duyệt. @@ -101,7 +94,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020, sửa đổi). Bộ luật Hình sự（第一百四十一、一百四十三、一百四十四条）. （Cơ sở dữ liệu luật và quy phạm quốc gia）；Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao (最高人民法院、最高人民检察院) (2021). Giải thích về một số vấn đề pháp lý khi xử lý các vụ án hình sự gây hại an toàn thực phẩm（法释〔2021〕24 号，第一、二、五、七、九、十七、十八条，2022 年 1 月 1 日起施行）. （Cục Quản lý thị trường tỉnh Hà Bắ trích lại toàn văn）、（Sở Tư pháp tỉnh Liao Ninh trích lại toàn văn, hai bản đối chiếu từng chữ khớp nhau） - Ghi chú: Ranh giới với ba mục trước là có đụng tới an toàn thực phẩm hay không. Không xin giấy, nhãn thiếu mục, thổi tác dụng chữa bệnh thì bị phạt tiền. Bán thịt chết bệnh, hàng vượt tiêu chuẩn, thêm phụ gia bừa bãi, trộn nguyên liệu không phải thực phẩm thì đi vào thủ tục hình sự. Phòng thủ tiện nhất vẫn là bộ động tác ở mục 8 của phần này (làm thực phẩm thì xem trước mình ở bậc nào). Thịt chỉ nhập từ nhà máy giết cố định. Mỗi lô giữ giấy kiểm dịch và giấy xác nhận chất lượng thịt. Thành phần chỉ dùng loại tra được số hiệu phê duyệt. Thứ được gọi là「thịt xử lý giá rẻ」「phế tấm không biết từ đâu」thì hay xảy ra sự cố nhất. Lần này trang Công báo Tòa án nhân dân tối cao trả về lỗi 502, nên đã dẫn hai bản trích lại chính thức của tỉnh và đối chiếu từng chữ với nhau. - ### 12. Có giấy phép là bắt đầu có nghĩa vụ kê khai, không có doanh thu vẫn phải kê khai đúng hạn (nộp một tờ toàn số không) - Chi phí: Không tốn tiền. Tự kê khai theo tháng hoặc theo quý trên cổng thuế điện tử, mỗi lần vài phút. Thuê người làm kế toán thay là khoản chi khác ngoài thị trường. Đơn vị đã thuê người làm còn phải đăng ký bảo hiểm xã hội trong 30 ngày kể từ ngày thành lập. Khó ở chỗ những tháng không có việc cũng phải nhớ kê khai đúng hạn. @@ -110,7 +102,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2002). Quy tắc chi tiết về thi hành Luật Quản lý thuế thu（税收征收管理法实施细则）（国务院令第 362 号，第三十二、七十五条）. ; Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2015). Luật Quản lý thuế thu（税收征收管理法）（2015 年修正，第二十五、三十二、六十二、六十四条）. ; Cục Tính thuế quốc gia (国家税务总局) (2019). Thông báo về một số việc trong quản lý thuế（国家税务总局公告 2019 年第 48 号，第三条）. ; Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2018). Luật Thuế thu nhập cá nhân（个人所得税法）（2018 年修正，第十二条）. ; Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2018). Luật Bảo hiểm xã hội（社会保险法）（2018 年修正，第五十七条）. - Ghi chú: Sau「đăng ký một mã nhiều giấy」không phải đăng ký thuế riêng nữa. Lấy mã tín dụng xã hội thống nhất trên giấy phép để đăng nhập cổng thuế điện tử, xác nhận thông tin đăng ký, rồi kê khai những loại thuế mình phải nộp. Hộ kinh doanh cá thể và doanh nghiệp tư nhân nộp thuế thu nhập cá nhân theo thu nhập kinh doanh. Trước hết nộp một khoản「trong vòng mười lăm ngày sau khi kết thúc tháng hoặc quý」, rồi「làm quyết toán trước ngày ba mươi mốt tháng ba năm sau năm thu được tiền」(tính cả năm, trả nhiều hoàn ít). Không khai trương vẫn phải kê khai, kê khai toàn số không cũng là kê khai. Để yên ba tháng là thành hộ không bình thường, hoá đơn bị dừng, muốn đăng ký huỷ cũng không đi được thủ tục rút gọn, xem mục 23 của phần này (lỗ thì rút lui theo thủ tục). - ### 13. Hoá đơn chỉ xuất theo giao dịch thật, người nộp thuế quy mô nhỏ hãy dùng hết ngưỡng miễn thuế - Chi phí: Không tốn tiền. Không mua hoá đơn, không bán hoá đơn, cũng không giúp người khác「lập sổ qua hoá đơn」. Xuất hoá đơn và kê khai thuế theo việc kinh doanh thật sự xảy ra. @@ -119,6 +110,7 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2024). Luật Thuế giá trị gia tăng（增值税法）（2026 年 1 月 1 日施行，第二十三条）. ；Bộ Tài chính, Cục Tính thuế (财政部、税务总局) (2026). Thông báo về việc chuyển tiếp các chính sách ưu đãi thuế giá trị gia tăng sau khi Luật Thuế giá trị gia tăng có hiệu lực（财政部 税务总局公告 2026 年第 10 号，第一、二、六条）. ；Đại biểu Quốc hội (全国人大) (1997). Bộ luật Hình sự（第二百零五条）. - Ghi chú: Luật Thuế giá trị gia tăng có hiệu lực từ ngày 1 tháng 1 năm 2026. Hai thông báo giảm thuế cho người nộp thuế quy mô nhỏ trước đây, số 2023 年第 19 号 và số 2023 年第 1 号, theo đó ngừng thi hành hoặc bãi bỏ. Ngưỡng số tiền không đổi, nhưng căn cứ đã thay, mục này đã cập nhật theo văn bản mới. Trang nguyên văn Bộ luật Hình sự là bản năm 1997, Điều 205 sau này đã được sửa đổi về các bậc phạt tù, mục này chỉ ghi những bậc đã đối chiếu và không bị sửa. Miễn là thuế giá trị gia tăng, thuế thu nhập tính riêng. 「Tìm người mua hoá đơn để bù chi phí」là xuất hoá đơn khống, không phải tiết kiệm thuế. + ### 14. Ai tự xưng cơ quan thuế đòi bạn nộp tiền khai thuế đều là lừa đảo, làm thủ tục thuế chỉ đi qua cổng thuế điện tử và 12366 - Chi phí: Không tốn tiền. Nhận cuộc gọi, tin nhắn như「税务抽查」「账户异常」「不报税就罚款」, hoặc có người kết bạn thì cắt máy, không trả lời. Rồi gọi 12366, hoặc gọi số cố định của cơ quan thuế quản lý để kiểm chứng. @@ -127,7 +119,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Cục Thuế nhà nước (国家税务总局) (2025). 税务部门提醒广大纳税人——增强防范意识 警惕互联网涉税诈骗. ; Sở Thuế thành phố Dương Châu (国家税务总局扬州市税务局) (2023). 扬州防范涉税诈骗宣传篇. ; Cục Thuế nhà nước (国家税务总局) (2025). 涉税专业服务管理办法（试行）（国家税务总局令第 58 号，第三十二条第七项、第三十三条）. ; Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，2026 年 1 月 1 日施行，第六十二条）. ; Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部) (2016). 关于办理电信网络诈骗等刑事案件适用法律若干问题的意见（法发〔2016〕32 号）. - Ghi chú: Kiểu lừa này nhắm vào hộ kinh doanh cá thể và công ty nhỏ vừa đăng ký. Lời dẫn thường là「系统显示你没报税，今天不处理就罚款」, rồi dẫn bạn sang tài khoản chính thức hay tiểu ứng của chúng, thu một lần vài trăm tệ. Trong khi bảu kê đó vốn miễn phí, tự làm chỉ mất vài phút. Thuê tổ chức ngoài thị trường làm sổ sách kế toán thay không vi phạm pháp luật, nhưng đó không phải cơ quan nhà nước, cũng không dùng danh nghĩa cơ quan thuế để đòi bạn. Đã trả tiền hoặc chuyển khoản rồi thì làm theo mục 2 ở phần 8 (dừng chi trả khi bị lừa), gọi ngay 110 hoặc 96110 để yêu cầu dừng chi trả. - ### 15. Nhận tiền thì viết「定金」chứ đừng viết「订金」, tiền phạt vi phạm phải ghi rõ con số, hạn thanh toán thì soi như cho vay - Chi phí: Không tốn tiền. Trong hợp đồng viết đúng hai chữ「定金」, số tiền không quá 20% tổng giá trị hợp đồng. Nhận tiền rồi mới giao hàng. Trước khi đồng ý cho bên kia lấy hàng trước, trả tiền sau (cho bán hàng có hạn nợ), hãy tra đối tác một lượt. @@ -136,7 +127,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Quốc hội Nhân dân (全国人大) (2020). 民法典（第五百八十五、五百八十六、五百八十七、五百八十八条）. - Ghi chú:「订金」trong Bộ luật Dân sự không có bộ quy tắc như tiền đặt, thường chỉ được coi là khoản trả trước. Đây là cách nói lưu hành trong thực tiễn, phần này chưa kiểm chứng án lệ. Hạn thanh toán là cho vay tiền của mình không lãi cho bên kia, bên kia phá sản thì thành nợ xấu. Trước khi cho bán hàng có hạn nợ, vào Hệ thống công bố thông tin tín dụng doanh nghiệp quốc gia tra hồ sơ của bên kia, xem đã từng bị khởi kiện hay chưa. - ### 16. Người lao động tháng đầu phải ký hợp đồng bằng văn bản, 30 ngày đăng ký bảo hiểm xã hội - Chi phí: Phần do công ty đóng trong bảo hiểm xã hội thì phải nộp đều, tỷ lệ khác nhau theo địa phương. Bản thân việc ký hợp đồng không tốn tiền, dùng thẳng mẫu hợp đồng của cơ quan nhân sự và xã hội địa phương là được. @@ -145,7 +135,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007). 劳动合同法（第十、八十二条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2010). 社会保险法（2018 年修正，第五十八、六十、八十四、八十六条）. - Ghi chú: Lương gấp đôi tính từ tháng thứ hai cho tới đủ một năm, kéo dài bao lâu thì tốn bấy nhiêu. Nhân viên chưa ký hợp đồng, chưa đóng bảo hiểm xã hội, một khi xảy ra tai nạn lao động, hoặc sau khi nghị việc đi qua hòa giải lao động, thì khoản nộp bù cộng tiền phạt cộng tiền bồi thường, thường nhiều hơn hẳn số tiền đã tiết kiệm được. - ### 17. Lương trả theo hợp đồng và theo pháp luật, tăng lương, cho vay, thưởng đều phải qua văn bản - Chi phí: Không tốn tiền. Viết rõ mức lương trong hợp đồng. Tăng lương, thưởng, ứng trước, cho vay, đều ký giấy để lại. Không hứa miệng. Cũng không tự bỏ tiền túi ra trả lương thay công ty. Khó là người quen mở lời mà vẫn theo đúng quy tắc này. @@ -154,7 +143,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: C - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2007). 劳动合同法（第十七、三十条）. ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第四百六十九、六百六十八条）. - Ghi chú: Không phải là không giúp, mà là giúp cũng phải theo thể chế. Ăn huệ ngoài thể chế một khi thành thói quen, muốn dừng lại sẽ sinh tranh cãi. Vay riêng lẫn vào sổ công ty, công ty lỗ thì khoản nợ xấu rơi vào túi bạn. Phần điều luật đã đối chiếu, còn lời khuyên là kinh nghiệm, nên để mức C. - ### 18. Bán trước rồi mới sản xuất: dùng bán trước và đơn nhỏ để kiểm chứng có người mua, rồi mới bỏ tiền vào sản xuất - Chi phí: Vài chục đến vài trăm tệ, làm một mẫu hoặc một trang. Rồi đi tìm vài chục khách thật, bảo họ trả trước hoặc đặt đơn nhỏ. Cách này chậm hơn vài tuần so với khoảng sản xuất ngay. Khó là nhịt không làm hàng trước. @@ -163,7 +151,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: B - Nguồn: Camuffo A, Cordova A, Gambardella A, Spina C (2020). A Scientific Approach to Entrepreneurial Decision Making: Evidence from a Randomized Control Trial. Management Science 66(2):564-586. - Ghi chú: Chỉ có một nghiên cứu như vậy, cỡ mẫu nhỏ, lại làm ở Italia, nên để mức B. Tên bài, nguồn xuất bản và phần tóm tắt đã đối chiếu qua Crossref và Semantic Scholar, còn các con số hiệu quả trong toàn văn thì phần này chưa kiểm chứng. Xét tới thao tác cụ thể: nếu không ai chịu trả trước, không ai đặt đơn nhỏ, đừng tin「làm ra có người mua」. - ### 19. Làm xong mẫu thì so qua danh sách sản xuất hàng loạt, rồi hãy nói chuyện khoảng sản xuất - Chi phí: Vài trăm tệ cho một lần thử sản xuất số lượng nhỏ. Rồi dành vài tiếng để tính rõ mấy khoản: tỷ lệ thành phẩm (trong 100 sản phẩm thì bán được mấy cái), số lượng đặt hàng tối thiểu và hạn thanh toán của nhà cung cấp, tồn kho chiếm bao nhiêu tiền, bao bì và vận chuyển, chứng nhận và giấy phép, nhãn hiệu. Khó là nhịt không bỏ qua bước này để khoảng sản xuất ngay. @@ -181,7 +168,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). Bản sửa đổi (mười một) Bộ luật Hình sự（刑法修正案（十一））（第十七、二十四项，即刑法第二百一十四、二百二十条）. ；Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao (最高人民法院、最高人民检察院) (2025). Giải thích về một số vấn đề pháp lý khi xử lý các vụ án hình sự xâm phạm quyền sở hữu trí tuệ（法释〔2025〕5 号，2025 年 4 月 26 日起施行）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2019). Luật Nhãn hiệu（商标法）（2019 年修正，第五十七、六十四条；2026 年 6 月 26 日修订后为第七十二、七十八条，自 2027 年 1 月 1 日起施行）. 、 - Ghi chú: Mỹ phẩm, đồ uống, hàng mẹ và bé, hàng điện tử, quần áo là những nhóm hàng hay xảy ra vụ việc loại này. Điểm chung là cùng một mã hàng có khoảng chênh giá rất lớn. Các nói「hàng kênh」「hàng tồn cuối」「hàng miễn thuế nước ngoài」đều không thay thế được hoá đơn. Ngoài phạt tù còn có một mức xử phạt hành chính: xác định được vi phạm thì tịch thu hàng vi phạm và công cụ, doanh thu vi phạm từ 5 vạn tệ trở lên còn có thể bị phạt kèm không quá 5 lần. Đã nhập hàng về mới phát hiện sai thì dừng bán ngay, niêm phong lại, rồi liên hệ hãng hoặc cơ quan quản lý thị trường. Đừng bán tiếp, đừng huỷ hàng và sổ sách. Huỷ chứng cứ đúng là rơi vào tình huống thứ năm「bị coi là bạn biết hàng giả」nêu trên. Dùng hình ảnh của người khác xem ở mục 21 của phần này (hình ảnh thì tự làm hoặc mua giấy uỷ quyền). Có một kiểu lừa đảo nhắm vào đúng điều này: có người chủ động mang đến, nói mẻ bút hay món nhỏ khác cho bạn miễn phí, đặt trong cửa hàng bán rồi chia tiền. Vài ngày sau có người đến mua, lấy bằng chứng tại chỗ, rồi đòi bồi thường vì vi phạm. Hàng cho không thì thiếu đúng cả ba điều kiện để được miễn bồi thường: không có hợp đồng mua và chứng từ chuyển khoản, không nói được nhà cung cấp thật, giá rõ ràng bất thường. Không mục「có mua hợp pháp và nêu tên nhà cung cấp」nào dùng được. Vì vậy ai chủ động giao hàng tới, không đưa hoá đơn, chỉ để lại một tài khoản WeChat thì đều không lên kệ. Thật sự muốn làm đại lý, hãy ký hợp đồng đại lý bằng văn bản, giữ lại bản sao căn cước và giấy phép kinh doanh của bên kia, ghi rõ hàng thuộc về ai, vi phạm thì tính vào ai. - ### 21. Hình ảnh trên hàng, bao bì, treo thẻ và ảnh quảng cáo thì tự làm hoặc mua giấy uỷ quyền, đổi màu, thêm biểu tượng không tính là「đã sửa」 - Chi phí: Giấy uỷ quyền trong thư viện ảnh chính hãng bắt đầu vài chục tệ một ảnh. Thuê người thiết kế riêng thì vài trăm đến vài nghìn tệ. Mỗi ảnh dùng trên hàng đều giữ lại một chứng từ uỷ quyền và biên la mua. Khó ở chỗ nhịn không dùng ảnh tìm được trên mạng. @@ -190,7 +176,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). Luật Bảo hộ quyền tác giả（著作权法）（2020 年修正，第五十二条第六项、第五十三条）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2020). Bản sửa đổi (mười một) Bộ luật Hình sự（刑法修正案（十一））（第二十、二十一项，即刑法第二百一十七、二百一十八条）. ；Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao (最高人民法院、最高人民检察院) (2025). Giải thích về một số vấn đề pháp lý khi xử lý các vụ án hình sự xâm phạm quyền sở hữu trí tuệ（法释〔2025〕5 号）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2019). Luật Nhãn hiệu（商标法）（2019 年修正，第五十七条第二项；2026 年修订后为第七十二条第二项）. - Ghi chú: Có ba chỗ hay vấp. Một là in hình nhân vật hoạt hình, hình người nổi tiếng tìm được trên mạng lên quần áo, cốc, vỏ điện thoại. Hai là phông chữ: phông dùng thương mại phần lớn phải có giấy uỷ quyền riêng, xem trước miễn phí không có nghĩa dùng thương mại miễn phí. Ba là bản thiết kế làm cho khách có lẫn tư liệu chưa được uỷ quyền, trách nhiệm vẫn thuộc về người bán hàng. Muốn làm gọn thì dùng thư viện ảnh có giấy uỷ quyền thương mại, và lưu lại trang giấy uỷ quyền. Hoặc bắt bên thiết kế ghi rõ trong hợp đồng tư liệu lấy từ đâu, uỷ quyền trong phạm vi nào, có sự cố thì bên thiết kế chịu trách nhiệm. Ngoài phạt tù còn có bồi thường dân sự và phạt hành chính: gây tổn hại lợi ích công cộng thì có thể tịch thu thu nhập bất hợp pháp, và phạt gấp từ 1 đến 5 lần doanh thu vi phạm. Trước khi sản xuất, tra nhãn hiệu xem ở mục 19 của phần này (làm xong mẫu thì so qua danh sách sản xuất hàng loạt). Ngược lại, hình ảnh của bạn bị người khác dùng thì cũng theo bộ quy tắc này mà bảo vệ quyền lợi. - ### 22. Ở đúng vị trí thì làm theo quy trình, còn cái thiện ý ngoài quy trình thì nghĩ kỹ trách nhiệm thuộc về ai - Chi phí: Không tốn tiền. Khi làm việc thì theo quy định, theo chỉ định của bác sĩ, theo quy trình. Không thay người khác quyết định việc vượt ngoài thẩm quyền của mình. Trước khi giúp, tự hỏi một câu: có chuyện thì ai chịu trách nhiệm. Khó ở chỗ phải từ chối tận thời lời「linh động một chút». @@ -199,7 +184,6 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Mức bằng chứng: B - Nguồn: Quốc vụ viện (国务院) (2008). Quy định về y tá（护士条例）（国务院令第 517 号，第十七条）. ；Quốc vụ viện (国务院) (2020). Quy định về y tá（护士条例）（2020 年修订，第十七条）. （Cơ sở dữ liệu luật và quy phạm quốc gia）；Đại biểu Quốc hội (全国人大) (2020). Bộ luật Dân sự（第一百八十四条）. - Ghi chú: Trong phạm vi công vụ thì làm theo quy trình, cứu người trong tình huống khẩn cấp được pháp luật bảo vệ. Còn những việc khác, trước hết phải phân định đó là quyền của ai, trách nhiệm của ai, đừng dùng thiện ý riêng của mình thay cho quyết định mang tính quy trình. Quy định về y tá năm 2020 được sửa đổi bằng Quy định của Quốc vụ viện số 726, chỉ sửa mấy điều về đăng ký, Điều 17 không đổi một chữ. Các điều về việc hành nghề theo quy chuẩn khám chữa bệnh trong Luật Y khoa, lần này không mở được nguyên văn chính thức nên không dẫn. - ### 23. Lỗ thì rút lui theo thủ tục: huỷ được kiểu rút gọn thì huỷ, vỡ nợ thì đi phá sản, đừng để yên - Chi phí: Huỷ kiểu rút gọn bạn tự làm được: niêm yết trên Hệ thống công bố thông tin tín dụng doanh nghiệp quốc gia trong 20 ngày. Còn nợ chưa trả hết thì phải làm rõ sổ sách trước, cái phải thu thì thu, cái phải trả thì trả, bước này gọi là thanh lễ. Đi phá sản thì nộp đơn xin tòa, mất mấy tháng. Khó ở chỗ lỗ rồi không muốn đối mặt, cứ kéo dài không làm. @@ -207,4 +191,4 @@ Phần này chỉ tính tiền và trách nhiệm pháp lý, không quy đổi r - Lợi ích: Không nợ người khác và người khác không nợ mình, hoặc đã trả hết. Không nợ lương, không nợ bảo hiểm xã hội, không nợ thuế. Các nhà đầu tư đã viết cam kết bằng văn bản. Đủ mấy điều đó thì niêm yết 20 ngày không ai phản đối là đi được thủ tục huỷ kiểu rút gọn. Hộ cá thể không phải niêm yết, cơ quan có liên quan trong 10 ngày không phản đối thì huỷ luôn. Công ty không trả nổi khoản nợ đến hạn mà số nợ nhiều hơn số tài sản có (vỡ nợ), có thể xin tòa tái cơ cấu, thỏa thuận hoặc thanh lễ phá sản. Đã giải thể mà chưa thanh lễ, nợ nhiều hơn tài sản có, thì người phụ trách thanh lễ nên nộp đơn xin thanh lễ phá sản. Hậu quả khi để yên thì thế này: không niêm yết báo cáo thường niên đúng hạn thì bị đưa vào danh mục bất thường kinh doanh. Liên tục 2 năm không nộp báo cáo thường niên, đã vào danh mục bất thường kinh doanh mà không sửa, lại không liên lạc được ở địa chỉ đã đăng ký, thì bị thu hồi giấy phép kinh doanh. Người làm người đại diện theo pháp luật của doanh nghiệp bị thu hồi giấy phép và chịu trách nhiệm cá nhân, kể từ ngày bị thu hồi giấy phép trong 3 năm không được làm người đại diện theo pháp luật, giám đốc, giám sát, cán bộ quản lý của công ty. Đã vào danh mục bất thường kinh doanh thì kể cả huỷ kiểu rút gọn cũng không đi được (toàn quốc) - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2021). Quy định về đăng ký chủ thể kinh doanh（市场主体登记管理条例）（国务院令第 746 号，第三十一、三十二、三十三条）. ；Cục Quản lý thị trường quốc gia và năm cơ quan khác (市场监管总局等六部门) (2025). Cẩm nang huỷ doanh nghiệp (bản sửa đổi năm 2025)（2025 年第 52 号公告）. ；Ủy ban Thường vụ Đại biểu Nhân dân toàn quốc (全国人大常委会) (2006). Luật Phá sản doanh nghiệp（企业破产法）（第二、七条）. ；Quốc vụ viện (国务院) (2024, sửa đổi). Quy chế tạm thời về công khai thông tin doanh nghiệp（企业信息公示暂行条例）（国务院令第 777 号修订，第十八条）：「企业因连续2年未按规定报送年度报告被列入经营异常名录未改正，且通过登记的住所或者经营场所无法取得联系的，由县级以上市场监督管理部门吊销营业执照」. （Cơ sở dữ liệu luật và quy phạm quốc gia）；Quốc vụ viện (国务院) (2024). Quyết định sửa đổi và bãi bỏ một số quy chế hành chính（国务院令第 777 号）. ；Quy định về đăng ký chủ thể kinh doanh（第十二条第四项）与 Luật Công ty (2023, sửa đổi, Điều một trăm bảy mươi tám khoản một điểm bốn): người làm người đại diện theo pháp luật của doanh nghiệp bị thu hồi giấy phép kinh doanh vì vi phạm pháp luật mà chịu trách nhiệm cá nhân, kể từ ngày bị thu hồi giấy phép kinh doanh chưa quá 3 năm thì không được làm người đại diện theo pháp luật, giám đốc, giám sát, cán bộ quản lý cấp cao. ； -- Ghi chú: Quy chế tạm thời về công khai thông tin doanh nghiệp năm 2024 được sửa đổi bằng Quy định của Quốc vụ viện số 777, có hiệu lực từ ngày 1 tháng 5, sau đó không sửa thêm. Điều 17 của bản cũ「经营异常满 3 年列入严重违法企业名单」đã bị bỏ. Danh sách doanh nghiệp vi phạm nghiêm trọng, mất uy tín hiện nay quản lý doanh nghiệp bịa báo cáo thường niên, tình tiết nặng. Nhiều cách nói trên mạng vẫn là bản cũ, đừng sao chép. Chế độ phá sản cá nhân (kể cả hộ cá thể) hiện mới có ở vài địa phương thí điểm, phần này không đối chiếu nguyên văn nên không ghi. \ No newline at end of file +- Ghi chú: Quy chế tạm thời về công khai thông tin doanh nghiệp năm 2024 được sửa đổi bằng Quy định của Quốc vụ viện số 777, có hiệu lực từ ngày 1 tháng 5, sau đó không sửa thêm. Điều 17 của bản cũ「经营异常满 3 年列入严重违法企业名单」đã bị bỏ. Danh sách doanh nghiệp vi phạm nghiêm trọng, mất uy tín hiện nay quản lý doanh nghiệp bịa báo cáo thường niên, tình tiết nặng. Nhiều cách nói trên mạng vẫn là bản cũ, đừng sao chép. Chế độ phá sản cá nhân (kể cả hộ cá thể) hiện mới có ở vài địa phương thí điểm, phần này không đối chiếu nguyên văn nên không ghi. diff --git "a/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" "b/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" index 895abf9f..e63bd5dd 100644 --- "a/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" +++ "b/vi/book/13-\347\264\247\346\200\245\346\203\205\345\206\265.md" @@ -6,14 +6,14 @@ Phần này nói trước khi xảy ra chuyện thì làm gì, sau đó làm gì Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận ra lừa đảo, hoặc bị người ta cầm những thứ bắt để tống tiền, thì xử lý tính theo ngày, không thuộc phần này. Quy tắc cứng chống lừa đảo và bảy kiểu lừa đảo hay gặp, xem mục 3 ở phần 8 (quy tắc cứng chống lừa đảo). Bị tống tiền bằng ảnh riêng tư, clip nhạy cảm, xem mục 32 ở phần 8 (đối phương cầm nhúng để tống tiền). Tiền đã chuyển đi rồi thì gọi ngay 110 hoặc 96110 yêu cầu phong tỏa thanh toán, xem mục 2 ở phần 8 (phong tỏa thanh toán). -### 1. Người bị ngã và không thở: ấn mạnh vào giữa ngực, bảo người bên cạnh gọi 120 và đi lấy AED +### 1. Người ngã xuống đất không thở: ấn thật mạnh vào giữa ngực ngay, bảo người bên cạnh gọi 120 và tìm máy AED -- Chi phí: Không tốn tiền. Khóa hồi sinh tim phổi của Hội Chữ thập đỏ hoặc của bệnh viện kéo dài nửa ngày đến một ngày, học một lần là đủ dùng nhiều năm. Không học vẫn làm được: chồng hai tay đặt vào giữa đường nối hai núm vú, ấn mạnh xuống khoảng 2 lần mỗi giây, đừng dừng. -- Nói thẳng: Người bạn có khả năng cao nhất phải ấn chính là người nhà bạn: 79.2% các ca ngừng tim xảy ra ở trong nhà. Có người ấn, tỷ lệ sống sót đến lúc xuất viện tăng từ 3.9% lên 16.1%. Gọi không đáp, không thở thì ấn luôn. Vị trí là giữa đường nối hai núm vú, chồng hai tay ấn mạnh, 2 lần mỗi giây, không dừng. Bảo người thứ hai gọi 120 và đi lấy AED, tay bạn không dừng. -- Lợi ích: Gộp 79 nghiên cứu, 142.740 người lại với nhau (phân tích tổng hợp): trong số người tim ngừng ngoài bệnh viện (ngừng tim ngoài bệnh viện), tổng cộng chỉ 7.6% sống sót đến lúc xuất viện. Khi có người chứng kiến làm hồi sinh tim phổi, tỷ lệ sống sót đến lúc xuất viện tăng từ 3.9% lên 16.1%. 53% có người bên cạnh nhìn thấy lúc bị lên cơn, thật sự được người chứng kiến ấn ngực chỉ 32%. Sổ đăng ký BASIC-OHCA của Trung Quốc ghi 38.227 ca cấp cứu được tiếp nhận. Trong đó 30.282 ca (79.2%) ngừng tim xảy ra tại nhà. Trong nhóm này, tỷ lệ được người chứng kiến ấn ngực là 20.3%, sống sót đến lúc xuất viện là 1.2%. Thứ tự là ấn ngực trước, rồi bảo người thứ hai gọi điện và đi lấy AED. Máy bật lên thì làm theo giọng nói của nó +- Chi phí: Không tốn tiền. Lớp hồi sinh tim phổi của Hội Chữ thập đỏ hoặc của bệnh viện, nửa ngày đến một ngày, học một lần là đủ dùng nhiều năm. Chưa học thì vẫn làm được: chồng hai tay lại, đặt vào giữa đường nối hai núm vú, ấn xuống thật mạnh khoảng 2 lần mỗi giây, đừng dừng. +- Nói thẳng: Người bạn phải ấn nhiều nhất lại là người trong nhà bạn: 79.2% ca ngừng tim xảy ra ở trong nhà. Có người ấn thì tỷ lệ sống tới lúc xuất viện tăng từ 3.9% lên 16.1%. Gọi không đáp, không thở thì ấn luôn. Vị trí là giữa đường nối hai núm vú, chồng hai tay ấn thật mạnh, 2 lần mỗi giây, không dừng. Bảo người thứ hai gọi 120 và lấy máy AED, tay bạn đừng dừng. +- Lợi ích: Gộp 79 nghiên cứu, 142.740 người lại (phân tích tổng hợp): với người tim ngừng ngoài bệnh viện (ngừng tim ngoài bệnh viện), tổng cộng chỉ 7.6% sống tới lúc xuất viện. Người được người bên cạnh làm hồi sinh tim phổi thì tỷ lệ sống tới lúc xuất viện tăng từ 3.9% lên 16.1%. Có người chứng kiến cơn ngừng tim chiếm 53%, thực sự được người bên cạnh ấn chỉ 32%. Sổ đăng ký BASIC-OHCA của Trung Quốc ghi 38.227 ca cấp cứu tiếp nhận. Trong đó 30.282 ca (79.2%) ngừng tim xảy ra tại nhà. Với nhóm này, tỷ lệ được người bên cạnh ấn là 20.3%, sống tới lúc xuất viện là 1.2%. Thứ tự là ấn trước, rồi mới bảo người thứ hai gọi điện thoại và lấy máy AED. Máy bật lên thì làm theo giọng nó hướng dẫn - Mức bằng chứng: A -- Nguồn: Sasson C và cộng sự (2010). Predictors of survival from out-of-hospital cardiac arrest: a systematic review and meta-analysis. Circulation: Cardiovascular Quality and Outcomes. ; Zheng J và cộng sự (2023). Incidence, process of care, and outcomes of out-of-hospital cardiac arrest in China: a prospective study of the BASIC-OHCA registry. The Lancet Public Health. ; Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一百八十四条）. -- Ghi chú: Đừng bấm huyệt nhân trung trước, đừng đi tìm người thân trước, đừng quay video trước. Cũng đừng vì sợ bẻ gãy xương sườn mà không dám ấn mạnh. Điều 184 của Bộ luật Dân sự ghi rõ: người tự nguyện cứu hộ khẩn cấp mà làm người được cứu bị tổn thương thì người cứu hộ không chịu trách nhiệm dân sự. +- Nguồn: Sasson C và cộng sự (2010). Predictors of survival from out-of-hospital cardiac arrest: a systematic review and meta-analysis. Circulation: Cardiovascular Quality and Outcomes. ; Zheng J và cộng sự (2023). Incidence, process of care, and outcomes of out-of-hospital cardiac arrest in China: a prospective study of the BASIC-OHCA registry. The Lancet Public Health. ; Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（Điều 184）. +- Ghi chú: Trẻ dưới 1 tuổi dùng cách khác, phải có thổ khí, xem mục 44 của phần này (hồi sinh tim phổi trẻ sơ sinh). Đừng bấm huyệt nhân trung trước, đừng đi tìm người nhà trước, đừng quay video trước. Cũng đừng vì sợ ấn gãy xương sườn mà không dám ấn mạnh. Điều 184 của Bộ luật Dân sự ghi rõ: người tự nguyện cứu hộ khẩn cấp mà làm người được cứu bị thương thì người cứu hộ không phải chịu trách nhiệm dân sự. ### 2. Người cao tuổi ngã, người bị ngã: quỳ xuống gọi họ, gọi 120, đừng vội đỡ dậy; với người xa lạ thì bước đi cũng hợp pháp, mà đã dừng lại thì đừng bê người bằng tay @@ -23,7 +23,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Bộ Y tế (卫生部) (2011). 老年人跌倒干预技术指南（「老年人跌倒后的处理」）. Bản PDF đầy đủ ở Trung tâm kiểm soát và phòng ngừa bệnh Quảng Châu ，phần xử lý tại hiện trường có đầy đủ ở bản đăng lại của Chính phủ quận Quý Nam ; Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（第一百八十三、一百八十四条）. ; Tòa án nhân dân tối cao (最高人民法院) (2023). 人民法院抓实公正与效率践行社会主义核心价值观典型案例（案例十三，齐某某诉孙某某健康权纠纷案，辽宁省康平县人民法院）. ; Tòa án nhân dân tối cao (最高人民法院) (2022 年第二次修正，2022 年 4 月 10 日施行). 关于适用《中华人民共和国民事诉讼法》的解释（第九十条）. ; Điều một nghìn một trăm chín mươi tám của Bộ luật Dân sự như trên, bản đầy đủ do Viện kiểm sát nhân dân tối cao đăng lại ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2023 年修正，2024 年 1 月 1 日施行). Bộ luật Tố tụng dân sự, 民事诉讼法（第一百五十二、一百六十四条）. Bản đầy đủ đăng lại ở Ủy ban Phát triển và cải cách Thượng Hải ; Hội đồng Nhà nước (国务院) (2006). 诉讼费用交纳办法（国务院令第 481 号，第六、二十九条）. ; Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 治安管理处罚法（2025 年修订，2026 年 1 月 1 日起施行，第二十六条第一项、第五十条第五项）. - Ghi chú: Để ở mức B là vì hướng dẫn chỉ nói cách xử lý, không có con số「làm theo thì giảm được bao nhiêu ca tử vong」. Điều 184 miễn trách cho tổn thương phát sinh trong lúc cứu người, điều kiện là bạn tự nguyện và tình huống khẩn cấp. Những việc như「tự ý nhét người vào xe riêng chở đi bệnh viện」nó không lo, chờ 120 thì chắc hơn. Gọi 120 xong đừng tút máy, làm theo lời người trực tổng đài. Bộ phán đoán này trước hết dành cho người cao tuổi trong nhà và người quen. Đổi sang người không quen gặp ngoài đường thì bước đi không vi phạm luật, và cũng tiết kiệm nhất: pháp luật không bắt bạn phải cứu. Cái giá phải trả nếu dừng lại thì phải tính từng bước: thời gian, để lại số điện thoại, bị người nhà của họ tìm đến, bị kiện ra tòa, ồn đến cơ quan nơi làm việc, bị quay video và đăng lên mạng. Xem chi tiết [docs/遇到陌生人出事该不该停.md](../docs/遇到陌生人出事该不该停.md). Nếu quyết định dừng lại thì việc duy nhất bắt buộc phải làm là**đừng bê người bằng tay**, lý do nằm ở y học: dựng cứng có thể làm nặng thêm xuất huyết não hoặc tổn thương cột sống. Có một ngoại lệ: nếu nơi đó do bạn kinh doanh, bạn quản lý, hoặc hoạt động do bạn tổ chức thì bước đi có thể phải chịu trách nhiệm. Nghĩa vụ bảo đảm an toàn theo Điều 1198 của Bộ luật Dân sự đòi bạn gánh một phần trách nhiệm an toàn cho những người ở trong khu vực đó. Bản thân chuyện phòng ngã xem mục 13 ở phần 1 (tập thăng bằng và sức chân). Triệu chứng đến muộn sau khi người cao tuổi bị va đầu, xem mục 10 của phần này. Chuyện tiền sau khi cứu người mà bị thương, xem mục 39 của phần này. Bị công kích trên mạng thì lấy bằng chứng và báo công thế nào, xem mục 16 ở phần 8. - ### 3. Miệng lệch, một bên tay không có lực, nói không rõ: gọi 120 ngay, đừng chờ, đừng tự lái xe đi - Chi phí: Không tốn tiền. Ba động tác, xem một lần là biết: bảo người đó cười, giơ hai tay ngang thân, nói một câu trọn vẹn. @@ -32,7 +31,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: A - Nguồn: Emberson J, Lees KR, Lyden P, et al. (2014). Effect of treatment delay, age, and stroke severity on the effects of intravenous thrombolysis with alteplase for acute ischaemic stroke: a meta-analysis of individual patient data from randomised trials. Lancet, 384(9958), 1929-1935. - Ghi chú: Can thiệp mạch càng sớm càng hiệu quả, quá 4.5 giờ thì gần như không dùng được. Gọi 120 thay vì tự lái xe, vì xe cứu thương đưa thẳng tới bệnh viện có trung tâm đột quỵ và còn báo trước để bên đó chuẩn bị. Ghi lại thời điểm chính xác bắt đầu có triệu chứng, bác sĩ dùng nó để quyết định có can thiệp mạch được không. Không cho uống nước, không cho uống thuốc, đừng nghĩ「nghỉ một lát xem sao」. - ### 4. Chóng mặt xoay tròn không đứng vững, nhìn thấy hai hình, một mắt tối đi, hoặc chỉ tay không trúng mũi mình: vẫn coi như đột quỵ và gọi 120 - Chi phí: Không tốn tiền. Ngoài ba động tác「miệng lệch, giơ tay, nói」(mục 3, miệng lệch, nói không rõ) thì thêm hai động tác nữa. Một là bảo người đó đứng thẳng nhắm mắt, xem có chệnh không, rồi hỏi người đó nhìn vật có thành hai hình không, có mất một mảng không. Động tác còn lại là khi bạn nghi ngờ thì giơ ngón tay chỉ vào mũi, nhắm mắt chỉ lại một lần. @@ -50,7 +48,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Mac Grory B, Schrag M, Biousse V, et al. (2021). Management of Central Retinal Artery Occlusion: A Scientific Statement From the American Heart Association. Stroke, 52(6), e282-e294. ; Rothwell PM, Giles MF, Chandratheva A, et al. (2007). Effect of urgent treatment of transient ischaemic attack and minor stroke on early recurrent stroke (EXPRESS study): a prospective population-based sequential comparison. Lancet, 370(9596), 1432-1442. - Ghi chú: Để ở mức B là vì EXPRESS không phải thử nghiệm chia nhóm ngẫu nhiên. Nó so hai giai đoạn trước và sau khi đổi cách làm trên cùng một dân số, giữa hai giai đoạn đó các điều kiện khác cũng có thể đã thay đổi, không tách được. Mắt tối đi vài phút rồi tự hết là dạng dễ nhầm nhất với「mệt quá」「tụt đường huyết」rồi bỏ qua. Đó là một dạng cơn thiếu máu não thoáng qua, mấy ngày sau đó đúng là lúc đột quỵ xuất hiện nhiều nhất. Một mắt mất thị lực mà không đau, không đỏ, không chảy nước mắt thì càng phải coi trọng. Đừng chờ ngày hôm sau xếp hàng khám mắt ngoại trú, đi thẳng vào cấp cứu ở bệnh viện có trung tâm đột quỵ. - ### 6. Một mắt sưng đau, đỏ, nhìn đèn thấy một vòng cầu vồng, kèm đau đầu buồn nôn, nôn ói: đi cấp cứu mắt trong ngày - Chi phí: Không tốn tiền. Cấp cứu mắt đo nhãn áp và soi tiền đồ chiếm nửa ngày. @@ -59,7 +56,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Zhou L, Wu S, Wang Y, Bao X, Peng T, Luo W, Ortega-Usobiaga J (2022). Clinical presentation of acute primary angle closure during the COVID-19 epidemic lockdown. Frontiers in Medicine, 9, 1078237. ; Sung MS, Kim HJ, Park SW (2023). Predictors of long-term visual field outcome after an episode of acute primary angle closure. Clinical & Experimental Ophthalmology, 51(4), 291-299. ; Wang J, Wang J, Ng TK, Huang C (2025). Asymmetric intraocular pressure changes in dominant and contralateral eyes: the dark room prone provocative test. Seminars in Ophthalmology, 40(4), 325-331. - Ghi chú: Có hai lý do để ở mức B. Một là hai nhóm ở Vũ Hán được so bằng cách đọc lại hồ sơ bệnh cũ, bản thân việc phong tỏa còn tạo ra những khác biệt khác, không tách được khỏi yếu tố「kéo dài」. Hai là nhóm ở Hàn Quốc chỉ có 50 con mắt, cũng là nghiên cứu hồi cứu. Yếu tố kích thích không phải là nhìn màn hình, mà là hai việc chồng lên nhau: ở chỗ tối đồng tử giãn ra, còn cúi đầu hoặc nằm sấp lâu lại đẩy thủy tinh thể ra phía trước. Người vốn đã có tiền đồ nông thì dễ bị kích thích như vậy, nên nhóm thật sự có nguy cơ cao là người trên 50 tuổi, người viễn thị, và người có người thân trong nhà từng bị glô-cô-mơ góc đóng. Người trẻ thức cả đêm chơi game hầu như không lên cơn như vậy. Sai lệch dễ gặp nhất lúc lên cơn là nhầm sang viêm dạ dày ruột, đau nửa đầu, hoặc huyết áp cao, vì đau đầu và nôn biểu hiện rõ hơn triệu chứng ở mắt. Một mắt mất thị lực mà không đau không đỏ là chuyện khác, xem mục 5 (một mắt tối đi đột ngột). Kính chống ánh sáng xanh có ích không, xem mục 16 ở phần 6 (kính chống ánh sáng xanh). - ### 7. Đau tức ngực kiểu bóp nghẽn, khó thở, cứng ngực quá 15 phút không dịu: gọi 120, đừng chịu đựng và đừng tự lái xe - Chi phí: Không tốn tiền. @@ -68,7 +64,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Karkabi B, Meir G, Zafrir B, et al. (2021). Door-to-balloon time and mortality in patients with ST-elevation myocardial infarction. European Heart Journal - Quality of Care and Clinical Outcomes. - Ghi chú: Nghiên cứu đó so đoạn bị chậm sau khi đã vào bệnh viện, không phải đoạn từ lúc khởi phát đến lúc vào bệnh viện. Nhưng lý do thì vẫn vậy: chậm thêm một chút nghĩa là thêm một mảng cơ tim chết hoại. Cơn đau có thể nằm ở bụng trên, ở lưng, ở cằm, cũng có thể chỉ là tê một bên cánh tay. Người bị tiểu đường và người cao tuổi có thể không đau, chỉ thấy khó thở. Đừng tự lái xe, mất ý thức trên đường là một tai nạn thứ hai. - ### 8. Đau dữ dội như bị xé toạc, lan từ ngực ra sau lưng và xuống thắt lưng: gọi 120 và nói rõ「cơn đau đang di chuyển」 - Chi phí: Không tốn tiền. @@ -77,7 +72,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Hagan PG, Nienaber CA, Isselbacher EM, et al. (2000). The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. JAMA, 283(7), 897-903. - Ghi chú: Cách xử lý bóc tách động mạch ngược hẳn với nhồi máu cơ tim. Nhồi máu cơ tim cần thuốc chống đông, thuốc làm tan cục máu, tức là thuốc làm cho máu không đông lại; dùng thuốc đó lên bóc tách thì hậu quả là máu chảy qua chỗ rách. Vì vậy đến bệnh viện hãy chủ động nói「xé như thế, cơn đau đang đi xuống」để bác sĩ nghĩ đến việc chụp CT cải tiến. Huyết áp hai tay chênh nhau nhiều, một bên không sờ thấy mạch là dấu hiệu. Người kiểm soát huyết áp không tốt, người mắc hội chứng Marfan, người trong nhà từng bị phình động mạch hoặc bóc tách động mạch thì nguy cơ cao. Cũng đừng tự lái xe, cũng đừng chịu đựng đến ngày hôm sau. - ### 9. Đột nhiên một cơn「đau đầu nặng nhất đời», trong vòng một giờ đã tới lúc đau nhất: đi cấp cứu ngay để chụp CT sọ - Chi phí: Đi cấp cứu bản thân không tốn bao nhiêu tiền. Một lần chụp CT sọ ở khoa cấp cứu vài trăm tệ. @@ -95,7 +89,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Zhang J và cộng sự (2021). Expert consensus on drug treatment of chronic subdural hematoma. Chinese Neurosurgical Journal, 7(1), 47. ; Dziho A và cộng sự (2025). Global prevalence and incidence of chronic subdural hematoma: A systematic review. Brain and Spine, 5, 105893. ; Mathon B, Shotar E (2026). Diagnostic and therapeutic management of chronic subdural hematoma in elderly patients. Gériatrie et Psychologie Neuropsychiatrie du Vieillissement, 24(1), 30-41. - Ghi chú: Lúc va thì có thể hoàn toàn không sao, đến nỗi người cao tuổi còn không nhớ là mình đã va đầu. Nên trong nhà thấy「dạo này bỗng lú lẫn, đi loạng bước, lúc nào cũng muốn ngủ, một tay không có lực」thì trước hết quay lại hỏi trong hai ba tháng qua có ngã, va đầu không, rồi mới đi chụp CT. Đừng quy thẳng là「sao trí nhớ」hay「di chứng của lần đột quỵ trước」, quy sai là chậm. Người cao tuổi uống aspirin, clopidogrel, warfarin hoặc thuốc chống đông máu đường uống thế hệ mới có nguy cơ cao hơn, uống rượu nhiều lâu năm cũng vậy. Phòng ngã xem mục 13 ở phần 1 (tập thăng bằng và sức chân). - ### 11. Một chân sưng, căng, ấn vào thì đau: đi khám càng sớm càng tốt; thêm khó thở đột ngột hoặc đau ngực thì gọi 120 ngay - Chi phí: Một lần siêu âm mạch máu chân (siêu âm tĩnh mạch chi dưới) một hai trăm tệ. @@ -104,7 +97,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Goldhaber SZ, Visani L, De Rosa M (1999). Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet, 353(9162), 1386-1389. ; WHO (2007). Study results released on travel and blood clots. - Ghi chú: Hai chân cùng sưng thì phần lớn là vấn đề của tim, gan, thận hay chức năng tĩnh mạch. Một chân sưng đột ngột, to hơn chân kia một vòng, mới là dáng vẻ điển hình của huyết khối tĩnh mạch sâu. Dễ xảy ra với người đi xe dài, đi máy bay, nằm liệt sau mổ hoặc sau gãy xương, bó bột, mang thai và trước sau khi sinh, bị ung thư, dùng thuốc tránh thai. Cục máu bong ra rồi bịt động mạch phổi chính là thuyết phối khối phổi: khó thở đột ngột, hít một hơi sâu thì đau ngực hơn, nhịp tim nhanh, choáng, ho ra máu (ho máu). Đến mức đó thì gọi 120 ngay. 17.4% là tỷ lệ tử vong của các ca thuyết phối khối phổi đã được xác định, không phải「chân sưng thì nguy cơ lớn cỡ này」, đừng dùng con số đó để tự dọn mình. - ### 12. Máu chảy nhiều: dùng tay ấn thật chặt vào vết thương, tứ chi ấn không được thì dùng gar, đồng thời gọi 120 - Chi phí: Một gar kiểu xoay vặn vài chục đến hơn một trăm tệ, để trong xe và trong túi cấp cứu. Học một lần cách「buộc trên vết thương, vặn đến khi ngừng máu, ghi giờ」, mất mấy phút. @@ -113,7 +105,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: A - Nguồn: Teixeira PGR và cộng sự (2018). Civilian Prehospital Tourniquet Use Is Associated with Improved Survival in Patients with Peripheral Vascular Injury. Journal of the American College of Surgeons. ; Kragh JF và cộng sự (2009). Survival with emergency tourniquet use to stop bleeding in major limb trauma. Annals of Surgery. ; Chi hội Bác sĩ Cấp cứu của Hiệp hội Bác sĩ Trung Quốc (中国医师协会急诊医师分会) và cộng sự (2020). 止血带的急诊应用专家共识. 中华急诊医学杂志 29(6):773-779：「院前急救时，止血带一旦使用则不建议松开，除非得到可以替代的彻底止血的救治」「最长使用时间不应超过2 h。但如果客观情况无法到医院救治或者无替代止血办法，则在得到正规救援前不解除止血带」. ，bản đầy đủ - Ghi chú: Đừng lấy dây, dây giày, những thứ mảnh làm gar, vì không siết được động mạch và còn cứa rách da. Đã buộc gar thì đừng cứ vài phút lại nới ra「xả máu». Ở cổ và thân mình chỉ được ấn, không được buộc. Trong số liệu chiến trường đó, bản thân cái gar không làm ai bị cắt chi, có 4 ca xuất hiện tê liệt thần kinh tạm thời. Ý kiến chung của các chuyên gia cấp cứu Trung Quốc nói cùng ý: đã dùng gar trước bệnh viện thì không nên nới, trừ khi có cách khác thay thế để cầm máu triệt để. Chảy máu ở tay trên buộc vào một phần ba trên của tay trên, chảy máu ở đùi buộc vào gần hông. Không buộc ở cẳng tay, bắp chân, khuỷu tay, khuỷu chân. Ý kiến chung khuyến nghị tổng thời gian từ lúc trước bệnh viện tới sau bệnh viện cố không quá 2 giờ, nhưng nếu không tới được bệnh viện và không có cách nào khác để cầm máu thì không mở ra cho tới khi có cứu hộ chính quy. Gar tạm làm từ dây lưng, cà vạt, tay áo thì đứng một mình khó cầm được máu, phải kèm một que để vặn siết. - ### 13. Bị chó, mèo cắn hoặc cào rách da: xịt xen kẽ nước xà phòng và nước chảy trong 15 phút, trong ngày đi tiêm vắc-xin - Chi phí: Rửa không tốn tiền. Vắc-xin dại cả liệu trình vài trăm tệ. Mức phơi nhiễm độ ba còn phải tiêm thêm huyết thanh miễn dịch bệnh dại (tiêm thẳng kháng thể đã có sẵn), vài trăm đến hơn một nghìn tệ. @@ -122,7 +113,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: A - Nguồn: Văn phòng Tổng hợp Cục Kiểm soát bệnh quốc gia và Văn phòng Ủy ban Quốc gia Y tế (国家疾控局综合司、国家卫生健康委办公厅) (2023). 关于印发狂犬病暴露预防处置工作规范（2023 年版）的通知. ; WHO. 狂犬病 实况报道. - Ghi chú: Đừng vì「con nuôi trong nhà, đã tiêm rồi」mà bỏ qua. Quy chuẩn xem vết thương ở mức nào (mức độ phơi nhiễm), không xem con chó từ đâu ra. Đừng tự băng vết thương. Phương pháp quan sát mười ngày chỉ quyết định có dừng tiêm hay không, không quyết định có bắt đầu tiêm hay không. Bị động vật hoang dã, bị dơi cắn thì xử lý y hệt. - ### 14. Bị bỏng: xả ngay bằng nước mát chảy trong 20 phút, đừng bôi kem đánh răng hay nước tương - Chi phí: Không tốn tiền. Tốn mất 20 phút. @@ -140,7 +130,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B (khuyến cáo mạnh trong hướng dẫn, nhưng mức bằng chứng nền thấp) - Nguồn: Hiệp hội Giáo dục Y khoa Trung Quốc (中国医药教育协会) và cộng sự (2022). Hướng dẫn thực hành lâm sàng cấp cứu sốc phản vệ (2020). Frontiers in Pharmacology: 「Epinephrine is the first-line medicine in GRADE II to IV anaphylaxis」「Intramuscular epinephrine should be injected in the mid-anterolateral thigh」, liều tiêm bắp tối đa ở người lớn 0.5 mg, dưới 14 tuổi là 0.3 mg, không có phản ứng thì lặp lại mỗi 5–15 phút. ；Muraro A và cộng sự (2022). EAACI guidelines: Anaphylaxis (2021 update). Allergy. - Ghi chú: Bản hướng dẫn dị ứng này (EAACI) tự viết rằng trong lĩnh vực đó「证据大多停留在很低的等级」 — bằng chứng phần lớn dừng ở mức rất thấp. Nó xếp khuyến cáo mạnh vì lợi ích lớn hơn rủi ro nhiều, không phải vì có thử nghiệm lớn. Adrenaline ở Trung Quốc là thuốc kê đơn, người thường không có trong tay. Việc làm được là gọi 120, nói「怀疑过敏性休克」 — nghi ngờ sốc phản vệ — và bảo xe cứu thương mang thuốc đến. Người đã từng dị ứng nặng thì xin bác sĩ kê bút tiêm tự động, và học cách tiêm vào phần giữa phía ngoài đùi (phía ngoại, hơi chếch trước), xuyên qua cả quần vẫn tiêm được. - ### 16. Người ta co giật rồi ngã xuống: dọn chỗ xung quanh, cho nằm nghiêng, nhìn đồng hồ đếm thời gian, không đưa gì vào miệng - Chi phí: Không tốn tiền. @@ -149,7 +138,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Cấp cứu khi người bị động kinh co giật: 「Don't hold the person down or stop their movements」「Don't put anything in their mouth. This can hurt their teeth or jaw」「Don't offer water or food until they are fully alert」「Time the seizure to track if it lasts more than 5 minutes. If it does, seek immediate medical attention or call 911」 — đừng ấn giữ người đó xuống hay chặn cử động của họ; đừng đưa bất cứ thứ gì vào miệng, làm vậy có thể làm tổn thương răng hoặc hàm; chỉ cho ăn uống khi họ đã hoàn toàn tỉnh táo; đếm thời gian cơn co giật để biết có kéo dài quá 5 phút không, nếu có thì tìm cấp cứu y tế ngay hoặc gọi 911. - Ghi chú: Trang đó còn liệt thêm vài trường hợp phải gọi xe cứu thương ngay: lên cơn rất sớm sau khi một cơn vừa kết thúc, sau cơn không gọi tỉnh được hoặc khó thở, bị thương trong lúc co giật, xảy ra dưới nước, và cơn đầu tiên. Cơn dừng thì cho nằm nghiêng với miệng hướng xuống, chờ người đó tỉnh hẳn rồi hãy cho nước. Trong nhà có người bị động kinh thì đây là những thao tác dùng đến nhiều nhất. Làm vậy cho người xa lạ thì khả năng đổi lại cho bạn nhỏ hơn, nhưng bạn cũng không phải chịu trách nhiệm (Điều 184 của Bộ luật Dân sự, xem mục 2 của phần này, đừng vội vàng dựng người đó dậy). - ### 17. Người bị tiểu đường đột nhiên run, đổ mồ hôi lạnh, nói không rõ: cho 15 gram đường trước, 15 phút sau đo lại - Chi phí: Không tốn tiền. Mang theo vài viên đường vuông hoặc một hộp nhỏ nước trái cây. @@ -158,7 +146,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: A - Nguồn: Hiệp hội Đái thái đường Hoa Kỳ (美国糖尿病学会) (2025). Tiêu chuẩn chẩn đoán và điều trị đái thường 2025, chương 6 Mục tiêu đường huyết và hạ đường huyết. Diabetes Care: phân cấp「Level 1: Glucose <70 mg/dL (<3.9 mmol/L) and ≥54 mg/dL (≥3.0 mmol/L)」「Level 2: Glucose <54 mg/dL (<3.0 mmol/L)」「Level 3: A severe event characterized by altered mental and/or physical status requiring assistance for treatment of hypoglycemia, irrespective of glucose level」; xử lý「For most individuals, 15 g carbohydrates should be ingested」「Fifteen minutes after initial treatment, repeat the treatment if hypoglycemia persists」 — với đa số người, nên nạp 15 g carbohydrate; 15 phút sau khi xử lý lần đầu, lặp lại nếu hạ đường huyết còn. - Ghi chú: Không cho người không tỉnh ăn uống vào, sẽ hút vào khí quản, gọi 120 thẳng. Cùng bộ tiêu chuẩn đó còn khuyên người dùng insulin và người hay bị hạ đường huyết nên có sẵn glucagon. Nguyên văn là「All individuals treated with insulin or who are at high risk of hypoglycemia...should be prescribed glucagon」 — mọi người được điều trị bằng insulin hoặc có nguy cơ cao bị hạ đường huyết… nên được kê glucagon. Dạng xịt mũi và dạng tiêm đã đóng sẵn thuốc dùng tiện hơn dạng phải pha lúc đó. Những cách này mặc định là chuẩn bị cho chính bạn hoặc người bị tiểu đường trong nhà bạn, và đường bạn mang theo cũng là mang cho họ. - ### 18. Người ta bị điện giật: cắt điện trước, rồi dùng gậy khô hay thứ gì tương tự gạp nguồn điện ra, tuyệt đối không kéo bằng tay - Chi phí: Không tốn tiền. @@ -167,7 +154,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: US CDC. What to Do to Protect Yourself From Electrical Hazards (mục First aid). ；Spies C, Trohman RG (2006). Narrative review: Electrocution and life-threatening electrical injuries. Annals of Internal Medicine 145(7):531-537. - Ghi chú: 「先断电再碰人」 — cắt điện trước rồi mới chạm vào người — là quy tắc thuần vì lợi ích cho bản thân: kéo bằng tay thì người thứ hai ngã xuống chính là bạn. Đường cao thế rớt xuống đất hay khu vực gần thiết bị biến áp thì không tự xử lý, lùi ra xa chờ ngành điện cắt nguồn. Ngừng tim do điện giật thì ấn theo cách thường lệ, khử rung theo cách thường lệ, đừng vì đó là「电伤」mà không dám làm. - ### 19. Báo động carbon monoxide kêu, hoặc cả nhà cùng đau đầu buồn nôn: ra ngoài trước rồi mới gọi - Chi phí: Không tốn tiền. @@ -176,23 +162,24 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: You J, Liu J, Zhou M (2020). Number of Deaths due to Carbon Monoxide Poisoning by Month and by Place of Death — China, 2018. China CDC Weekly 2(2):32: 「In 2018, there were 11,523 deaths caused by carbon monoxide poisoning reported in China」 — năm 2018, Trung Quốc báo cáo 11.523 ca tử vong vì ngộ độc carbon monoxide. ；Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Ngộ độc carbon monoxide: 「More than 400 Americans die from unintentional CO poisoning not linked to fires」「More than 100,000 visit an emergency department」「More than 14,000 are hospitalized」 — hơn 400 người Mỹ chết vì ngộ độc CO ngoài ý muốn không liên quan đến hỏa hoạn; hơn 100.000 người đến phòng cấp cứu; hơn 14.000 người nhập viện, các triệu chứng thường gặp「headache, dizziness, weakness, upset stomach, vomiting, chest pain, and confusion」 — đau đầu, chóng mặt, yếu, rối loạn tiêu hoá, nôn, đau ngực và lú lẫn. ；Ủy ban An toàn Sản phẩm Tiêu dùng Hoa Kỳ. Trung tâm Thông tin Carbon Monoxide: 「If you suspect CO poisoning, get outside to fresh air immediately, and then call 911」 — nếu nghi ngờ ngộ độc CO, hãy ra ngoài hít không khí tươi ngay, rồi mới gọi 911. - Ghi chú: Triệu chứng giống cảm lạnh, lại giống say rượu. Cả nhà cùng khó chịu, ra ngoài là khỏe, đó là dấu hiệu rõ nhất. Đừng ở lại trong nhà đi tìm nguyên nhân, đừng quay lại đóng van, đưa người ta ra ngoài rồi mới gọi. Lắp báo động thế nào, xem mục 3 ở phần 1. -### 20. Uống nhầm chất tẩy rửa, thuốc trừ sâu hoặc thuốc: đừng gây nôn, mang theo vỏ chai đi viện ngay; dính vào mắt hoặc da thì xải bằng nhiều nước sạch trong 15 phút + +### 20. Uống nhầm chất tẩy rửa, thuốc trừ sâu, thuốc: đừng gây nôn, cầm theo chai vào viện ngay; dính vào mắt hoặc da thì xịt nhiều nước sạch ít nhất 15 phút -- Chi phí: Không tốn tiền. Để chất tẩy rửa, thuốc sát trùng, thuốc trừ sâu và thuốc ở chỗ cao trẻ không với tới, hoặc trong tủ có khoá. Tuyệt đối không rót sang chai đồ uống. -- Nói thẳng: Uống nhầm chất tẩy rửa, thuốc trừ sâu hoặc thuốc thì đừng gây nôn, nôn ra thì chất lỏng ăn mòn sẽ cháy lại thực quản một lần nữa. Trước hết cho uống một lượng sữa hoặc lòng trắng trứng vừa phải, mang theo vỏ chai và tờ hướng dẫn đi bệnh viện ngay. Dính vào da hoặc mắt thì xải bằng nhiều nước sạnh đủ 15 phút, mắt thì phải kéo mí ra mà xải, đừng xoa. Dùng thuốc sát trùng chung với thuốc tẩy bồn cầu sẽ sinh khí clo. -- Lợi ích: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (中国疾病预防控制中心职业卫生与中毒控制所) đưa ra bốn cách xử lý. Nuốt vào miệng (hấp thu qua đường tiêu hoá): cho uống ngay một lượng sữa hoặc lòng trắng trứng vừa phải,「切记，不要催吐」 — nhớ kỹ, đừng gây nôn. Đi viện càng nhanh càng tốt, mang theo vỏ bao bì hoặc tờ hướng dẫn,「以便于医生快速判断成分并制定救治方案」 — để bác sĩ nhanh chóng xác định thành phần và lập phương án điều trị. Hít vào (hấp thu qua đường hô hấp): chuyển ngay ra nơi thông thoáng, nới lỏng cổ áo để đường hô hấp thông; nếu ho dữ dội, thở gấp hoặc khó thở thì đi viện ngay. Tiếp xúc qua da: xải ngay vùng bị tiếp xúc bằng nhiều nước sạch trong ít nhất 15 phút, chú ý không để nước xải bắn vào mắt hoặc phần da khác còn nguyên vẹn. Tiếp xúc vào mắt: xải ngay mắt bằng nhiều nước sạch hoặc nước muối sinh lý trong ít nhất 15 phút. Khi xải phải kéo mí mắt ra, để nước tiếp xúc đủ với nhãn cầu và mặt trong mi mắt,「切勿揉搓眼睛」 — tuyệt đối không xoa mắt. Xải xong đi khoa mắt ngay. Sản phẩm nguy hiểm thì nhận theo độ pH. pH<3 là thuốc tẩy bồn cầu, thuốc tẩy cặn, thuốc lau sàn, thuốc đánh bóng thép không gỉ. Một số thuốc thông ống nước siêu mạnh còn chứa axit sunfuric loãng. pH>11 là thuốc thông ống nước, thuốc tẩy bếp siêu mạnh, bột rửa bát, muối thông ống, bột tẩy quần áo. Còn một cặp phối hợp gây họa ngay tại chỗ: dùng thuốc sát trùng chung với thuốc tẩy bồn cầu chứa axit clohydric sẽ sinh khí clo, kích ứng đường hô hấp, gây ho dữ dội và thở gấp, thậm chí đe doạ tính mạng +- Chi phí: Không tốn tiền. Cất chất tẩy rửa, nước khử trùng, thuốc trừ sâu và thuốc ở chỗ cao trên tầm tay trẻ con, hoặc trong tủ có khoá. Tuyệt đối không bỏ sang chai đồ uống. +- Nói thẳng: Uống nhầm chất tẩy rửa, thuốc trừ sâu, thuốc thì đừng gây nôn, chất lỏng ăn mòn nôn ra sẽ đốt cháy thực quản thêm một lần nữa. Trước hết uống một lượng sữa hoặc lòng trắng trứng vừa đủ, cầm chai và tờ hướng dẫn vào viện ngay. Dính vào da hoặc mắt thì xịt nhiều nước sạch ít nhất 15 phút, mắt thì phải kéo mí mắt ra xịt, đừng xoa. Nước khử trùng trộn với thuốc tẩy bồn cầu sẽ sinh khí clo. +- Lợi ích: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc chia thành bốn trường hợp. Nuốt vào miệng (qua đường tiêu hoá): uống ngay một lượng sữa hoặc lòng trắng trứng vừa đủ,「切记，不要催吐」— nhớ kỹ, đừng gây nôn. Đi viện càng nhanh càng tốt, mang theo vỏ bao bì hoặc tờ hướng dẫn,「以便于医生快速判断成分并制定救治方案」— để bác sĩ nhanh chóng xác định thành phần và lập phương án cứu chữa. Hít vào (qua đường hô hấp): chuyển ngay ra nơi thoáng mát, nới lỏng cổ áo cho thở thông; nếu có ho khan nặng, thở gấp hoặc khó thở thì đi viện ngay. Tiếp xúc qua da: xịt nhiều nước sạch vào chỗ tiếp xúc ít nhất 15 phút, chú ý đừng để nước rửa bắn vào mắt hoặc da chỗ khác còn nguyên. Tiếp xúc qua mắt: xịt nhiều nước sạch hoặc nước muối sinh lý vào mắt ít nhất 15 phút. Khi rửa thì kéo mí mắt ra, để nước tiếp xúc đầy đủ với nhãn cầu và mặt trong mi mắt,「切勿揉搓眼睛」— tuyệt đối đừng xoa mắt. Rửa xong đi khoa mắt ngay. Sản phẩm nguy hiểm thì nhận theo độ chua kiềm (pH). pH dưới 3 là thuốc tẩy bồn cầu, chất tẩy cặn, thuốc lau sàn, chất làm bóng thép không gỉ. Một số thuốc thông đường ống nước loại mạnh còn chứa axit sunfuric loãng. pH trên 11 là thuốc thông đường ống nước, thuốc tẩy rửa bếp loại mạnh, bột rửa bát, muối nổ, bột tẩy rửa. Còn một tổ hợp gây tai nạn ngay tại chỗ: nước khử trùng dùng cùng thuốc tẩy bồn cầu chứa axit clohidric sẽ sinh khí clo, kích thích đường hô hấp, gây ho khan nặng và thở gấp, thậm chí nguy hiểm tới tính mạng - Mức bằng chứng: B -- Nguồn: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (2025). Nhắc nhở an toàn kỳ nghỉ hè: cảnh giác chất tẩy rửa axit mạnh và bazơ mạnh. -- Ghi chú: Để ở mức B là vì tài liệu phổ biến chính thức chỉ nói cách xử lý và cách nhận ra, không có con số「làm theo thì giảm được bao nhiêu vụ」. Lý do không gây nôn là chất lỏng ăn mòn nôn ra sẽ cháy lại thực quản, cái ống từ cổ xuống dạ dày, một lần nữa. Không chắc mình uống nhầm thứ gì thì mang vỏ chai, tờ hướng dẫn cùng chất nôn đi bệnh viện. Phòng hơn cấp cứu nhiều: cất theo nhóm, không để lẫn với thức ăn đồ uống, dùng xong vặn kín nắp rồi vứt vỏ chai, tuyệt đối không đổ vào chai nước khoáng. Khi dùng thì đeo găng cao su và kính bảo hộ, mở cửa sổ thông gió, không trộn các loại với nhau. Về carbon monoxide xem mục 19 của phần này, về bỏng xem mục 14 của phần này. +- Nguồn: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (2025). Nhắc nhở an toàn kỳ nghỉ hè: cảnh giác với chất tẩy rửa axit mạnh và bazơ mạnh. +- Ghi chú: Để ở mức B là vì tài liệu phổ biến của cơ quan chính thức chỉ nói cách xử lý và cách nhận ra, không có con số「làm theo thì giảm được bao nhiêu tai nạn」. Lý do không gây nôn: chất lỏng ăn mòn nôn ra sẽ đốt cháy thực quản (cái ống nối cổ họng với dạ dày) thêm một lần nữa. Không chắc đã nuốt nhầm thứ gì thì cầm chai, tờ hướng dẫn và cả dịch nôn mang vào viện. Phòng ngừa rẻ hơn cấp cứu nhiều: cất theo loại, không để chung với thức ăn đồ uống, dùng xong vặn kín nắp rồi bỏ vỏ, tuyệt đối không đổi vào chai nước khoáng. Khi dùng đeo găng cao su và kính bảo hộ, mở cửa sổ thông gió, không trộn lẫn các loại khác nhau. Ngộ độc carbon ô-xít xem mục 19 ở phần này, bỏng và phỏng xem mục 14 ở phần này. Hóa chất axít bazơ bắn vào người thì cởi áo quần thế nào, rửa bao lâu, xem mục 21 ở phần này (hóa chất bắn vào người). -### 21. Axit, bazơ hay hóa chất khác văng lên người: cởi ngay quần áo bị nhiễm, xải bằng nhiều nước sạch chảy, mắt thì kéo mí ra mà xải, xải đủ thời gian rồi hãy đi +### 21. Hóa chất axít, bazơ bắn vào người: cởi ngay quần áo bị nhiễm, xịt nhiều nước sạch chảy, mắt thì kéo mí mắt ra xịt, xịt đủ thời gian rồi hãy đi -- Chi phí: Không tốn tiền. Ngày đầu đi làm hãy xem vị trí thiết bị rửa mắt và vòi xải khẩn cấp, đừng đợi xảy ra chuyện rồi mới đi tìm. -- Nói thẳng: Ba bước. Bước một, cởi ngay quần áo, giày, tất dính chất; chỗ quần áo úp sát vào người thì tổn thương nặng nhất. Bước hai, xải bằng nhiều nước sạch chảy, da và mắt đều ít nhất 15 phút, Sở Quản lý Khẩn cấp tỉnh Hunan đưa ra 20 đến 30 phút, lấy con số dài hơn. Bước ba, mắt thì kéo mí ra mà xải, xải xong đi thẳng khoa mắt. Dùng nước ở nhiệt độ phòng, không dùng nước nóng. -- Lợi ích: Ba bước tại hiện trường. Bước một, cởi ngay quần áo, giày, tất bị nhiễm: chất độc vẫn tiếp tục hấp thu qua da, chỗ bị quần áo úp sát thì tổn thương nặng nhất. Bước hai, xải bằng nhiều nước sạch chảy. Mức sàn mà Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc đưa ra là ít nhất 15 phút cho cả da và mắt. Sở Quản lý Khẩn cấp tỉnh Hunan đưa ra 20 đến 30 phút để xải vết thương,「强烈的化学品要更长」 — hoá chất ăn mòn mạnh cần lâu hơn. Lấy con số dài hơn trong hai mức đó. Bước ba, với mắt thì phải「撑开眼睑，确保水流充分接触眼球及眼睑内侧」 — kéo mí mắt ra, đảm bảo nước tiếp xúc đủ với nhãn cầu và mặt trong mi mắt — và「切勿揉搓眼睛」 — tuyệt đối không xoa mắt — rồi đi khoa mắt ngay. Dùng nước ở nhiệt độ phòng hoặc ấm nhẹ, cấm nước nóng. Vết thương mới không bôi dầu mỡ, thuốc đỏ hay bất cứ thứ gì, cũng không bọc bằng vải bẩn, vì thứ bôi lên sẽ làm bác sĩ khó đánh giá độ sâu. Về mặt pháp lý,「化学性皮肤灼伤」 và「化学性眼部灼伤」 — bỏng hoá chất ở da và ở mắt — đều có trong danh mục bệnh nghề nghiệp theo luật, trường hợp xảy ra trong công việc thì áp dụng bệnh nghề nghiệp và tai nạn lao động. Khi xảy ra tai nạn cấp tính do yếu tố nguy hại nghề nghiệp, người sử dụng lao động「应当及时组织救治、进行健康检查和医学观察，所需费用由用人单位承担」 — có nghĩa tổ chức cứu chữa kịp thời, làm kiểm tra sức khoẻ và quan sát y tế, các khoản chi phí do người sử dụng lao động chịu (toàn quốc) +- Chi phí: Không tốn tiền. Ngày đầu tiên lên ca thì nhớ xem vị trí thiết bị rửa mắt và vòi xịt toàn thân, đừng đợi xảy ra tai nạn rồi mới đi tìm. +- Nói thẳng: Ba bước. Bước một, cởi ngay quần áo, giày, tất dính chất. Chỗ quần áo úp giữ lại nhiều thì bị tổn thương nặng nhất. Bước hai, xịt nhiều nước sạch chảy, da và mắt đều ít nhất 15 phút. Sở Quản lý Khẩn cấp tỉnh Hồ Nam chốt 20 đến 30 phút, cứ theo mốc dài hơn. Bước ba, mắt thì kéo mí mắt ra xịt, xịt xong đi thẳng khoa mắt. Dùng nước ở nhiệt độ thường, không dùng nước nóng. +- Lợi ích: Ba bước tại hiện trường. Bước một, cởi ngay quần áo, giày, tất bị nhiễm: độc chất vẫn tiếp tục hấp thu qua da, chỗ quần áo úp giữ bị tổn thương nặng nhất. Bước hai, xịt nhiều nước sạch chảy. Mức tối thiểu do Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc đưa ra là da và mắt mỗi thứ「至少 15 分钟」— ít nhất 15 phút. Mốc của Sở Quản lý Khẩn cấp tỉnh Hồ Nam là rửa vết thương 20 đến 30 phút,「强烈的化学品要更长」— hóa chất mạnh thì phải lâu hơn. Trong hai mốc đó lấy mốc dài hơn. Bước ba, mắt phải「撑开眼睑，确保水流充分接触眼球及眼睑内侧」— kéo mí mắt ra, để nước tiếp xúc đầy đủ với nhãn cầu và mặt trong mi mắt,「切勿揉搓眼睛」— tuyệt đối đừng xoa mắt, rửa xong đi khoa mắt ngay. Dùng nước ở nhiệt độ thường hoặc hơi ấm, cấm dùng nước nóng. Trên vết thương tươi không bôi thuốc mỡ, nước đỏ hay bất cứ thứ gì, cũng không bọc vải bẩn, bôi vào sẽ ảnh hưởng tới việc bác sĩ đánh giá độ sâu. Về mặt pháp luật,「化学性皮肤灼伤」và「化学性眼部灼伤」đều nằm trong danh mục bệnh nghề nghiệp, xảy ra trong công việc thì áp dụng bệnh nghề nghiệp và tai nạn lao động. Khi xảy ra tai nạn bệnh nghề nghiệp cấp tính, chỗ sử dụng lao động「应当及时组织救治、进行健康检查和医学观察，所需费用由用人单位承担」— phải kịp thời tổ chức cứu chữa, làm kiểm tra sức khoẻ và theo dõi y tế, chi phí do chủ sử dụng lao động chịu (toàn quốc) - Mức bằng chứng: B -- Nguồn: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (2025). Nhắc nhở an toàn kỳ nghỉ hè: cảnh giác chất tẩy rửa axit mạnh và bazơ mạnh. ；Sở Quản lý Khẩn cấp tỉnh Hunan (湖南省应急管理厅) (2015). Xử lý và điều trị khẩn cấp các tai nạn ngộ độc hoá chất thường gặp. ；Ủy ban Sức khoẻ Quốc gia cùng ba bộ khác (2024). Phân loại và danh mục bệnh nghề nghiệp (có hiệu lực từ ngày 1 tháng 8 năm 2025). ；Ủy ban Thường vụ Quốc hội (2018). Luật Phòng chống và kiểm soát bệnh nghề nghiệp (Điều ba mươi bảy). (Chính phủ thành phố Bắc Kinh đăng lại) -- Ghi chú: Để ở mức B là vì chuyện xải bao lâu thì hai tài liệu chính thức khác nhau: một viết 15 phút, một viết 20 đến 30 phút, không có một con số chuẩn quốc gia thống nhất. Đi bệnh viện thì mang theo tên hoá chất, vỏ bao bì, hoặc bảng dữ liệu an toàn (SDS), bác sĩ cần nó để lập phương án. Đừng ném đại một loại axit hay bazơ khác vào để「中和」 — trung hoà: hai thứ trộn vào sinh nhiệt, lượng cho vào không kiểm soát được. «Dung dịch trung hoà» trong tài liệu của Hunan nói đến loại nước xải chuyên dụng mà đơn vị chuẩn bị sẵn, không phải lấy thứ có trong tay ở hiện trường ra dội vào nhau. Bột khô như vôi bột, xi măng bột thì quét sạch trước rồi mới xải nước (có nơi trang ứng cứu cứu nói cả axit sunfuric đậm đặc cũng phải lau khô trước, mỗi nơi nói một kiểu, cuốn sách này không viết). Về chất tẩy rửa trong nhà và uống nhầm xem mục 20 của phần này. Về phòng vệ trước khi đi làm và khám sức khoẻ xem mục 10 và 11 ở phần 19 (tờ thông báo nguy hại trước khi đi làm, và thiết bị bảo hộ đơn vị phải cấp). +- Nguồn: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (2025). Nhắc nhở an toàn kỳ nghỉ hè: cảnh giác với chất tẩy rửa axit mạnh và bazơ mạnh. ；Sở Quản lý Khẩn cấp tỉnh Hồ Nam (湖南省应急管理厅) (2015). Xử lý và điều trị khẩn cấp tai nạn ngộ độc hoá chất thường gặp. ；Bộ Y tế và Sức khoẻ Quốc gia và ba cơ quan khác (2024). Phân loại và danh mục bệnh nghề nghiệp（áp dụng từ ngày 1 tháng 8 năm 2025）. ；Ủy ban Thường vụ Quốc hội (2018). 职业病防治法（Điều 37）. （đăng lại bởi Chính phủ thành phố Bắc Kinh） +- Ghi chú: Để ở mức B là vì chuyện rửa bao lâu, hai tài liệu chính thức một bản ghi 15 phút, một bản ghi 20 đến 30 phút, không có con số quốc gia thống nhất. Đi viện thì mang theo tên hoá chất, vỏ bao bì, hoặc bảng dữ liệu an toàn (SDS), bác sĩ cần nó để lập phương án. Đừng tùy tiện nắm một loại axít hay bazơ khác để「中和」— trộn hai loại sẽ sinh nhiệt, lấy bao nhiêu cũng không kiểm soát được.「中和液」trong tài liệu của Hồ Nam nói tới dung dịch rửa chuyên dụng mà đơn vị chuẩn bị sẵn, không phải lấy đồ tại hiện trường để trung hoà. Bột vôi, bột xi măng loại bột khô thì quét sạch trước rồi mới xịt nước (một số trang cấp cứu nói rằng cả axít sunfuric đậm đặc cũng phải lau khô trước, chỗ này các nơi nói khác nhau, cuốn sách không viết). Chất tẩy rửa trong nhà và uống nhầm thì xem mục 20 ở phần này. Phòng vệ và khám sức khoẻ trước khi lên ca xem mục 9 và 10 ở phần 19 (trước khi lên ca phải xem thông báo nguy cơ, đơn vị phải cấp đồ dùng bảo hộ). ### 22. Chóng mặt, buồn nôn, không ra mồ hôi hoặc không tỉnh khi trời nóng: chuyển ngay ra chỗ mát, cởi áo quần xả nước để hạ nhiệt; người không tỉnh thì không cho uống nước, gọi 120 @@ -202,7 +189,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: US National Weather Service. Heat Related Illnesses. ; Cục Kiểm soát bệnh tỉnh Vân Nam (云南省疾控局) (2026). Có mấy triệu chứng sau thì bạn có thể đang say nắng, 出现这些症状，你可能是中暑了. ; Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc. Thời tiết nóng, Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc nhắc phòng say nắng: 「如有条件可用冷水浸泡的方法对无基础疾病的青壮年快速降温。老人和有基础疾病的患者可向皮肤喷水通过蒸发降温。降温应遵循"先降温，后转运" 的原则」 — nếu có điều kiện, dùng cách ngâm nước lạnh để hạ nhiệt nhanh cho người trẻ và trung niên khoẻ mạnh không có bệnh nền; người lớn tuổi và người có bệnh nền thì xịt nước lên da để hạ nhiệt bằng bay hơi; hạ nhiệt phải theo nguyên tắc "hạ nhiệt trước, vận chuyển sau". ; Ủy ban Sức khoẻ Quốc gia (2010). Những điểm chính về phòng say nắng khi nóng. ; Eifling KP và cộng sự (2024). Wilderness Medical Society Clinical Practice Guidelines for the Prevention and Treatment of Heat Illness: 2024 Update. Wilderness & Environmental Medicine 35(1_suppl):112S-127S. (trang nội dung từ chối cho công cụ tự động tải về, nên khuyến cáo cụ thể được đối chiếu theo bản tóm tắt cập nhật trên trang của Hiệp hội Y khoa Hoang dã: 「cold water immersion … is the most important method of treating severe heat illness」「If CWI is unavailable, evaporative cooling is the recommended second-line treatment. … generally half as fast as CWI」 — ngâm nước lạnh là cách quan trọng nhất để điều trị kiệt sức vì nóng nặng; nếu không ngâm được, hạ nhiệt bằng bay hơi là cách tuyến hai được khuyến cáo, chậm hơn khoảng một nửa)；Lipman GS và cộng sự (2019). Cùng bộ hướng dẫn đó, bản 2019. - Ghi chú: Đừng đợi xem「缓一缓」 — dịu đi một chút — rồi mới làm: kiệt sức vì nóng nặng có thể đi từ chóng mặt tới hôn mê trong mấy chục phút. Không đổ nước vào miệng người không tỉnh, sẽ bị sặc. Chỗ đặt túi đá thì hai nguồn nói khác nhau. Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc nói đặt ở cổ, nách và bẹn; Hiệp hội Y khoa Hoang dã nói các nghiên cứu không cho thấy đặt ở mấy chỗ đó có tác dụng hạ nhiệt. Khuyến cáo của họ là: nếu chỉ có túi đá thì phủ kín cả người, túi đá không đủ thì đặt ở má, lòng bàn tay và lòng bàn chân. Cả hai đều nói nếu có điều kiện thì dùng ngâm nước lạnh. - ### 23. Người bị say nắng mà còn tỉnh: uống nhiều lần mỗi lần ít, nước muối loãng mát hoặc nước thể thao; đừng uống hùng hục, đừng uống rượu và đồ uống nhiều đường - Chi phí: Không tốn tiền. Một chai nước thể thao vài tệ. @@ -211,6 +197,7 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Viện Y tế lao động và Kiểm soát Ngộ độc thuộc Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc (2014). Hướng dẫn phòng ngừa và xử lý khẩn cấp say nắng cho công chúng (bản 2014). ; US CDC/NIOSH. Heat-Related Illnesses. - Ghi chú: Sau khi mồ hôi đầm mà chỉ uống nước trắng thì muối trong máu bị loãng (hạ natri máu), nên nước phải có chút muối. Uống hùng hục một lúc dễ nôn, thứ nôn hút vào khí quản chết nhanh hơn cả say nắng. 「别喝冰镇的」 — đừng uống đá lạnh — là khuyến cáo trong bản hướng dẫn năm 2014 của Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc, lý do là sợ co thắt dạ dày, không có bằng chứng định lượng. Đừng hiểu nhầm thành「không được dùng nước lạnh」. xem mục 22 của phần này (khi chóng mặt buồn nôn dưới trời nóng thì cởi áo quần xả nước hạ nhiệt) nói về mấy cách hạ nhiệt. Đó là lau bằng nước mát, ngâm nước lạnh, và đặt túi đá ở cổ, nách, bẹn. Chính mấy cách này là hàng đầu trong những cách hạ nhiệt mà nguồn chính thức khuyến cáo, và chúng đặt lên người chứ không rót vào miệng. Hạ nhiệt luôn đứng trước bù nước. Người không tỉnh, nói không rõ, không cho một ngụm nào. + ### 24. Khi hoảng hoạt: bò sát đất, sờ cửa rồi mới mở, cửa nóng thì đừng mở, đi cầu thang chứ đừng đi thang máy, ra ngoài rồi đừng quay lại - Chi phí: Lắp trong nhà 1 đến 2 báo động khói độc lập, vài chục tệ. Xem trước hai lối ra của hành lang. @@ -219,24 +206,23 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: US FEMA Ready.gov. Home Fires. ；Bộ Quản lý Khẩn cấp (应急管理部) (2019). Cách tự cứu khi hoảng hoạt ở nơi công cộng: 「逃生时不要乘坐电梯，当火势不大时，要尽量往楼层下面跑」「可用湿毛巾或口罩蒙鼻，匍匐或弯腰撤离」 — khi thoát hiểm không được đi thang máy, lửa chưa lớn thì cố chạy xuống dưới các tầng; có thể dùng khăn ướt hoặc khẩu trang che mũi, bò sấp hoặc cúi người rút ra. ；Sư đoàn Cứu hoả và Cứu hộ tỉnh Quý Châu (2020). Kiến thức cứu hoả, ba điểm chính khi thoát hiểm: 「湿毛巾对折3次（叠成8层），除烟率比较高，但烟和毒气浓度过高时，过滤效果就无法保证了。另外，毛巾过湿会导致呼吸困难」 — khăn ướt gấp đôi 3 lần (thành 8 lớp) thì lọc khói khá tốt, nhưng khi nồng độ khói và khí độc quá cao thì không còn bảo đảm tác dụng lọc; thêm nữa, khăn quá ướt sẽ gây khó thở. - Ghi chú: Đừng đi thang máy, mất điện là mắc kẹt trong đó. Đừng quay lại lấy tài sản. Đừng chạy lên trên, trừ khi đã xác nhận cửa trên nóc mở. Đừng nhảy lầu. Khăn ướt phải gấp đôi 3 lần thành 8 lớp, đừng tẩm quá nhiều nước, quá ướt thì ngược lại khó thở. Khi khói và khí độc quá đặc thì khăn cũng chắn không được. - -### 25. Thấy người đang đuối nước: gọi người và báo 110/120, đưa sào hoặc thả vật nổi, tuyệt đối không xuống nước +### 25. Thấy có người đuối nước: gọi người và gọi 110/120, đưa gậy, thả vật nổi, bản thân đừng xuống nước -- Chi phí: Không tốn tiền. Khó là nhịn cơn thèm nhảy xuống. -- Nói thẳng: Người đang đuối nước sẽ theo bản năng ấn người cứu xuống dưới nước, và người xuống nước cũng có thể chết đuối. Thứ tự là: hô to gọi người → báo 110/120 → đưa sào dài, dây → thả phao cứu sinh, xô rỗng, chai nước khoáng → người lên bờ mà không thở thì ấn ngực ngay. Không nắm tay thành dây chuyền rồi xuống nước. Mỗi năm trên thế giới có khoảng 30 vạn người chết đuối, đây là nguyên nhân tử vong thứ tư ở trẻ từ 1 đến 4 tuổi. -- Lợi ích: WHO: mỗi năm trên thế giới có khoảng 30 vạn người chết đuối, 92% ở các nước thu nhập trung bình và thấp. Đuối nước là nguyên nhân tử vong thứ tư ở trẻ từ 1 đến 4 tuổi, thứ ba ở trẻ từ 5 đến 14 tuổi. Can thiệp WHO khuyến cáo là「安全救援和复苏培训」 — đào tạo cứu hộ an toàn và hồi sinh, chẳng hạn là học trước cách cứu mà không xuống nước. Thứ tự là: hô to gọi người → gọi điện thoại → đưa sào dài, dây, thả phao cứu sinh, xô rỗng, chai nước khoáng → người lên bờ sau mà không thở thì ấn ngực ngay +- Chi phí: Không tốn tiền. Khó nhất là nhịn cái thôi lao xuống nước. +- Nói thẳng: Người đuối nước theo bản năng sẽ ghì người cứu xuống dưới, người xuống nước cũng có thể chết đuối. Thứ tự là: hét gọi người → gọi 110/120 → đưa gậy dài, dây thừng → ném phao cứu sinh, xô rỗng, chai nước khoáng → người lên bờ mà không thở thì ấn ngực ngay. Đừng xuống nước theo kiểu xích nắm tay. Cả năm trên toàn cầu có khoảng 30 vạn người chết đuối, đây là nguyên nhân tử vong thứ tư ở trẻ 1 đến 4 tuổi. +- Lợi ích: WHO: cả năm trên toàn cầu có khoảng 30 vạn người chết đuối, 92% ở các nước thu nhập trung bình và thấp. Đuối nước là nguyên nhân tử vong thứ tư ở trẻ 1 đến 4 tuổi, thứ ba ở trẻ 5 đến 14 tuổi. Biện pháp mà WHO khuyến nghị là「安全救援和复苏培训」— huấn luyện cứu hộ an toàn và hồi sinh, ví dụ học cách cứu mà không phải xuống nước. Thứ tự là: hét gọi người → gọi điện thoại → đưa gậy dài, dây thừng, ném phao cứu sinh, xô rỗng, chai nước khoáng → người lên bờ mà không thở thì ấn ngực ngay - Mức bằng chứng: C -- Nguồn: WHO. Đuối nước, thông tin thực trạng. -- Ghi chú: Lý do để ở mức C là quy tắc「không xuống nước」 không tìm được nguyên văn chính thức, chỉ có số tử vong của WHO và hướng「cứu hộ an toàn」. Đừng nắm tay thành dây chuyền rồi xuống nước. Chưa qua huấn luyện thì đừng đi kéo người, người đuối nước sẽ ấn người cứu xuống dưới. 「Không xuống nước」 cũng là quy tắc thuần vì lợi ích cho bản thân: người xuống nước tự chết đuối, còn người ở trong nước thì ai cũng thế. Người bạn thật sự cần cứu phần lớn là con bạn, cách phòng nằm ở xem mục 12 ở phần 1 (trẻ ở gần nước không được rời mắt). +- Nguồn: WHO. 溺水 实况报道. +- Ghi chú: Để ở mức C là vì quy tắc「đừng xuống nước」không tìm được nguyên văn chính thức, chỉ có con số tử vong của WHO và hướng「cứu hộ an toàn」. Đừng xuống nước nắm tay thành xích. Chưa qua huấn luyện thì đừng đi kéo người, người đuối nước sẽ ghì người cứu xuống dưới.「Đừng xuống nước」cũng là quy tắc tự bảo vệ: người xuống nước tự chết đuối, người ở dưới nước thì ai cũng như nhau. Người phải cứu thường là con mình, cách phòng xem mục 12 ở phần 1 (trẻ em ở gần nước không rời mắt). Có người rơi qua lỗ băng, cũng đừng chạy lên mặt băng để kéo, cách phòng trên băng xem mục 41 ở phần 1 (ngoài hoang tính độ dày mặt băng trước). -### 26. Người bị nghẹn không nói nổi: đứng sau lưng, vỗ lưng 5 lần rồi đẩy bụng 5 lần, ngã xuống thì làm hồi sinh tim phổi +### 26. Có người bị nghẽn nói không ra tiếng: đứng sau lưng vỗ lưng 5 lần rồi đấm bụng 5 lần, ngã xuống thì làm hồi sinh tim phổi -- Chi phí: Không tốn tiền. Xem lại động tác một lần: một tay nắm lại đặt trên rốn, tay kia bọc lấy, đẩy nhanh vào trong và lên trên. -- Nói thẳng: Trước hết hỏi「bạn bị nghẹn phải không, nói được không」. Người còn ho mạnh, còn ra tiếng thì đừng can thiệp, cứ để họ ho. Người không nói nổi thì đứng sau lưng, vỗ lưng 5 lần rồi đẩy bụng 5 lần, lặp lại cho tới khi vật ra ngoài hoặc người đó ngã xuống. Ngã xuống thì làm hồi sinh tim phổi, đồng thời gọi người báo số cấp cứu. Phụ nữ có thai và người quá béo thì đẩy vào ngực. Đừng thò tay vào miệng móc thứ không nhìn thấy. -- Lợi ích: Nguyên văn của Thư viện Y khoa Quốc gia Hoa Kỳ: trước hết hỏi「bạn bị nghẹn phải không, nói được không」.「能用力咳嗽、能说话的不要动手」 — người còn ho mạnh, còn nói được thì đừng can thiệp, bản thân cơn ho đã phần lớn tự đẩy vật ra. Người không nói nổi thì「反复做 5 次拍背接 5 次腹部冲击，直到异物出来或人失去意识」 — lặp lại vỗ lưng 5 lần rồi đẩy bụng 5 lần cho tới khi vật ra ngoài hoặc người mất ý thức.「失去意识就开始心肺复苏」 — mất ý thức thì bắt đầu hồi sinh tim phổi — và gọi người báo số cấp cứu cùng lúc. Phụ nữ có thai và người béo thì làm đẩy vào ngực +- Chi phí: Không tốn tiền. Xem lại thao tác một lần: một tay nắm đấm đặt ngay trên rốn, tay kia úp lên, đấm nhanh vào trong và lên trên. +- Nói thẳng: Trước hết hỏi「Bạn bị nghẽn à, nói được không」. Người còn ho mạnh, còn ra tiếng thì đừng can thiệp, cứ để họ ho. Người không ra tiếng thì đứng sau lưng, vỗ lưng 5 lần rồi đấm bụng 5 lần, lặp lại cho tới khi vật ra ngoài hoặc người ngã xuống. Ngã xuống thì làm hồi sinh tim phổi, đồng thời bảo người đó gọi số điện thoại cấp cứu. Phụ nữ mang thai và người quá béo thì đấm vào ngực. Đừng thò tay vào miệng mò vật không nhìn thấy. +- Lợi ích: Nguyên văn của Thư viện Quốc gia Y tế Hoa Kỳ: trước hết hỏi「你噎住了吗，能说话吗」.「Người còn ho mạnh, còn nói được thì đừng can thiệp」, bản thân cơn ho đã phần lớn đẩy được vật lạ ra. Người không ra tiếng thì「反复做 5 次拍背接 5 次腹部冲击，直到异物出来或人失去意识」— lặp lại vỗ lưng 5 lần rồi đấm bụng 5 lần cho tới khi vật lạ ra ngoài hoặc người mất ý thức.「Mất ý thức thì bắt đầu hồi sinh tim phổi」, đồng thời bảo người đó gọi số điện thoại cấp cứu. Phụ nữ mang thai, người béo thì đấm vào ngực - Mức bằng chứng: B -- Nguồn: US National Library of Medicine, MedlinePlus. Choking – adult or child over 1 year. ；Ủy ban Sức khoẻ thành phố Phúc Châu (福州市卫生健康委员会) (2026). Phương pháp Heimlich đã cập nhật: 「1岁以上幼儿、儿童及成人：从旧版单一腹部冲击改为"5次背部拍击+5次腹部冲击"的循环」「中国红十字系统急救培训一直都是按"5次背部拍击+5次腹部冲击"循环进行的方式进行教学」 — trẻ nhỏ từ 1 tuổi trở lên, trẻ em và người lớn: chuyển từ cách chỉ đẩy bụng của bản cũ sang vòng lặp "5 lần vỗ lưng + 5 lần đẩy bụng"; đào tạo cấp cứu của hệ thống Hội Chữ thập đỏ Trung Quốc vẫn luôn dạy theo vòng lặp "5 lần vỗ lưng + 5 lần đẩy bụng". -- Ghi chú: Đừng thò tay vào miệng móc thứ không nhìn thấy, sẽ đẩy sâu hơn. Đừng vỗ vào người còn ho được. Nghẹn thường xảy ra nhất ở bàn ăn nhà bạn, nên người được lợi chủ yếu là người trong nhà. Cách dạy ở Trung Quốc cũng là bộ này: Trung tâm Đào tạo Cứu hộ Khẩn cấp của Hội Chữ thập đỏ tỉnh Phúc Kiến nói rằng với trẻ trên 1 tuổi và người lớn, hệ thống Hội Chữ thập đỏ Trung Quốc vẫn dạy theo vòng lặp vỗ lưng 5 lần rồi đẩy bụng 5 lần. Nhiều hướng dẫn cũ trên mạng chỉ dạy đẩy bụng, đó là bản cũ. +- Nguồn: US National Library of Medicine, MedlinePlus. Choking – adult or child over 1 year. ；Sở Y tế và Sức khoẻ thành phố Phúc Châu (福州市卫生健康委员会) (2026). 海姆立克急救法更新了：「1岁以上幼儿、儿童及成人：从旧版单一腹部冲击改为"5次背部拍击+5次腹部冲击"的循环」「中国红十字系统急救培训一直都是按"5次背部拍击+5次腹部冲击"循环进行的方式进行教学」. +- Ghi chú: Đừng thò tay vào miệng mò vật không nhìn thấy, sẽ đẩy nó sâu hơn. Đừng vỗ lưng người còn ho được. Nghẽn thường xảy ra nhất ở bàn cơm nhà mình, người hưởng lợi chủ yếu là người thân. Cách dạy ở trong nước cũng là bộ này: Trung tâm Đào tạo Cứu hộ Khẩn cấp Hội Chữ thập đỏ tỉnh Phúc Kiến nói, với trẻ trên 1 tuổi và người lớn, hệ thống Hội Chữ thập đỏ Trung Quốc vẫn dạy theo vòng lặp vỗ lưng 5 lần rồi đấm bụng 5 lần. Nhiều hướng dẫn cũ trên mạng chỉ dạy đấm bụng, đó là bản cũ. Trẻ dưới 1 tuổi không đấm bụng, chuyển sang đấm ngực, xem mục 43 ở phần này (trẻ sơ sinh bị nghẽn). ### 27. Lạc đường hoặc xe hỏng ở sa mạc Gobi, sa mạc, vùng không người ở: ở lại tại chỗ hoặc cạnh xe, báo vị trí trước, rồi tìm bóng mát và giữ ấm, tiết kiệm nước, đừng đi lang thang @@ -246,7 +232,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: C - Nguồn: US National Park Service. Mojave National Preserve Safety. ; US National Park Service. Death Valley Safety. ; US National Park Service. Desert Driving Safety. ; Cục Quản lý Khẩn cấp thành phố Trùng Khánh (重庆市应急管理局) (2024). Khoa học khẩn cấp, những tín hiệu cầu cứu bạn cần biết khi ở ngoài trời (bản tranh dài): 「燃放三堆火焰是国际通行的求救信号，将火堆摆成三角形」「三声短三声长，再三声短；间隔1分钟之后再重复」 — đốt ba đống lửa là tín hiệu cầu cứu quốc tế, xếp ba đống lửa thành hình tam giác; ba tiếng ngắn ba tiếng dài rồi ba tiếng ngắn, cách 1 phút thì lặp lại. ; Bộ Văn hoá và Du lịch (文化和旅游部) (2018). Bộ Văn hoá và Du lịch nhắc du khách cân nhắc khi chọn du lịch khám phá ngoài trời: 「谨慎前往无人区和人烟稀少区」「尽量多车同行，提前准备对讲机，必要时租用卫星电话」 — cẩn thận khi tới vùng không người ở và vùng ít dân cư; nên đi nhiều xe cùng nhau, chuẩn bị bộ đàm trước, cần thì thuê điện thoại vệ tinh. - Ghi chú: Để ở mức C là vì tất cả những thứ này là nhắc nhở chính thức, không phải số liệu nghiên cứu. Lượng nước uống, việc báo trước, vùng phủ tín hiệu, ở lại cạnh xe và tín hiệu cầu cứu đều là nguyên văn chính thức. Câu「ba tiếng một nhóm」 được truyền rộng không tìm được nguyên văn chính thức, nên đã viết lại thành ba đống lửa và ba tiếng ngắn, ba tiếng dài, ba tiếng ngắn của phần chính thức. Đừng đi bộ tìm đường vào buổi trưa. Đừng uống nước tiểu. Đừng vì tiết kiệm nước mà không uống. - ### 28. Run, nói không rõ, đi lảo đảo là thân nhiệt tụt: thay quần áo ướt, quấn lại, làm ấm thân trước, uống đồ uống nóng có đường, không uống rượu - Chi phí: Một tấm chăn cách nhiệt nhôm vài tệ, cộng thêm một bộ quần áo khô để dự phòng. Biết nhận ra dấu hiệu thì không tốn tiền. @@ -255,6 +240,7 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: US National Weather Service. During Extreme Cold. ; Dow J và cộng sự (2019). Wilderness Medical Society Clinical Practice Guidelines for the Out-of-Hospital Evaluation and Treatment of Accidental Hypothermia: 2019 Update. Wilderness & Environmental Medicine. - Ghi chú: Đừng xoa tay chân. Đừng ngâm bồn nước nóng. Đừng cứng kéo hất người đó đi, người thân nhiệt tụt nặng thì tim dễ ngừng. Đừng vì「hết run」mà tưởng đã khỏe, không run (hết cơn run vì lạnh) là nặng thêm. Toàn văn hướng dẫn của Hiệp hội Y khoa Hoang dã phải đăng ký mới đọc được, lần này chỉ đối chiếu phần tóm tắt. + ### 29. Bị rắn cắn: ngồi yên, tháo nhẫn và đồng hồ, đưa đi bệnh viện tiêm huyết thanh kháng nọc càng nhanh càng tốt; không rạch, không hút, không buộc chặt - Chi phí: Không tốn tiền. Biết trước bệnh viện nào gần đó có huyết thanh kháng nọc. Khó là nhịn cơnghè tự xử lý vết thương. @@ -263,7 +249,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: WHO. Cắn động vật, thông tin thực trạng. ; WHO (2023). Ngộ độc do rắn cắn, thông tin thực trạng. ; Lai R và cộng sự (2024). The Chinese guideline for management of snakebites. World Journal of Emergency Medicine 15(5):333-355: 「An estimated 250,000–280,000 cases of venomous snakebites occur in China each year, affecting mainly patients aged ≥50 years」 — ước tính mỗi năm ở Trung Quốc có 250.000 đến 280.000 ca rắn độc cắn, chủ yếu ở nhóm bệnh nhân từ 50 tuổi trở lên. ; Yan S và cộng sự (2026). Clinical epidemiology of snakebite envenoming in hospitals 11 provinces of Yangtze River Basin and southern regions of China. PLoS Neglected Tropical Diseases: 「withdrawal (4346, 10.65%), and death (23, 0.06%)」 — tự xuất viện (4346, 10.65%) và tử vong (23, 0.06%). - Ghi chú: Chụp được ảnh con rắn thì chụp, đừng đi bắt. Đừng uống rượu, đừng chạy, đừng hút bằng miệng, đừng chườm đá, đừng dùng dây buộc tới bầm tím. Trung Quốc không có thống kê chính thức số ca chết vì rắn cắn trên toàn quốc. Tỷ lệ tử vong nội trú ở 11 tỉnh đó trông thấp, nhưng có hơn một phần mười bệnh nhân tự xuất viện khi chưa điều trị xong, họ ra sao sau đó thì không nằm trong thống kê. - ### 30. Khi động đất, trước hết xem mình đang ở trong loại nhà nào: nhà xây đúng tiêu chuẩn chống chấn thì nằm tại chỗ, nhà cũ có thể sập và cửa cách vài bước thì ra ngoài - Chi phí: Không tốn tiền. Xem trước nhà mình ra cửa mất mấy bước, đi đường nào, cái bàn nào thì chắc. @@ -272,7 +257,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Cục Địa chấn Trung Quốc (do Sở Địa chấn tỉnh An Huy sản xuất). 地震来了，是躲还是跑？（video phổ biến, bản chép lời thoại phụ đề）. ；Armenian HK, Noji EK, Oganesian AP (1992). A case-control study of injuries arising from the earthquake in Armenia, 1988. Bulletin of the World Health Organization, 70(2), 251-257. ；US FEMA Ready.gov. Earthquakes. ；Mahue-Giangreco M, Mack W, Seligson H, Bourque LB (2001). Risk factors associated with moderate and serious injuries attributable to the 1994 Northridge Earthquake. Annals of Epidemiology, 11(5), 347-357. - Ghi chú:Tranh cãi. Hai kết luận nói về hai loại nhà khác nhau. Chính Cục Địa chấn Trung Quốc cũng viết điều kiện của việc「tránh tại chỗ」 là「符合抗震设防要求的建筑内」. Nó còn yêu cầu「因地制宜」「因人而异」 — tuỳ tình huống, tuỳ người — nên phải xem cả ba việc: nhà có chịu được động đất không, bạn đang đứng ở vị trí nào, bản thân có chạy nổi không. Loạt nhà ở Armenia sập cả khối, không có chỗ để tránh, ra ngoài mới là lựa chọn hơn. FEMA Hoa Kỳ khuyên cho nhà ở trong nước họ là「就地趴下、护住头颈、抓牢，不要往外跑」, thương tích ở đó chủ yếu đến từ vật rơi và do ngã khi di chuyển trong lúc rung. Nghiên cứu Northridge còn chỉ ra rằng đưa tay ra đỡ đồ, hay vịn vào thứ「trông chắc」, lại làm nặng thêm thương tích. Vì vậy chỉ hỏi「chạy hay không」 thì vô ích, phải đánh giá「tòa nhà này có sập không, bạn có kịp ra khoảng trống trong mấy giây không」. Nhà xây đúng quy chuẩn từ sau năm 2001, cần đi xuống cầu thang hoặc đi qua sảnh thì nằm tại chỗ. Nhà một tầng, nhà gạch cũ, nhà tự xây không có thiết kế chống chấn thì cửa ở ngay vài bước, ra ngoài. Nghiên cứu ở Armenia chỉ so người sống sót bị thương với không bị thương, không tính người tử vong, cũng không nói có trừ ảnh hưởng của độ cao tầng hay không. Người chạy ra ngoài được vốn đã có thể sống ở tầng thấp, số đó phải tính vào. Mọi trường hợp đều không nhảy lầu, không đi thang máy, không đi xuống cầu thang khi rung. Ra ngoài rồi thì tránh xa nhà cao lớn, cầu vượt, cột điện, tường kính và biển quảng cáo. - ### 31. Gặp gấu, lợn rừng, bầy chó hoang: đừng chạy, đừng hét, đừng nhìn thẳng vào mắt, nghiêng người lùi ra chậm - Chi phí: Không tốn tiền. Vào núi thì mang theo một cây gậy leo núi, đi theo nhóm, đi có tiếng động cho con vật biết bạn tới trước. Khó là đứng yên mà không chạy. @@ -281,7 +265,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: US National Park Service. Staying Safe Around Bears. ；Văn phòng Tổng hợp Cục Kiểm soát bệnh quốc gia, Văn phòng Ủy ban Sức khoẻ Quốc gia (国家疾控局综合司、国家卫生健康委办公厅) (2023). Quy định công tác phòng ngừa và xử lý sau phơi nhiễm bệnh dại (bản 2023). - Ghi chú: Mức B chỉ áp cho gấu. Riêng「đừng chạy, lùi chậm, tìm chỗ cao hoặc vật cản」 với lợn rừng và chó hoang là kinh nghiệm, tính là mức C, lần này không tìm được nhắc nhở chính thức của Cục Quản lý Lâm nghiệp và Phát triển Quốc gia cùng các cục lâm nghiệp địa phương. Đừng dùng thức ăn dỗ, đừng áp sát chụp ảnh, đừng dắt chó vào núi có lợn rừng. Ở Trung Quốc gấu chủ yếu là gấu đen và gấu nâu, giả chết hay không giả chết thì tách ra xem theo hai loại đó. - ### 32. Gặp giông ngoài hoang dã: chui vào xe có nóc cứng hoặc vào công trình, tránh đường sống lưng, cây đứng một mình, mặt nước và kim loại - Chi phí: Không tốn tiền. Cái giá phải trả là lịch trình bị gián đoạn hơn nửa giờ. Khó là khi tiếng sấm còn xa thì đã chịu quay về. @@ -290,6 +273,7 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: US National Weather Service. An toàn khi có sét ngoài trời: 「There is no safe place outside when thunderstorms are in the area」「Avoid open fields, the top of a hill or a ridge top. Stay away from tall, isolated trees or other tall objects」「Stay away from water, wet items, such as ropes, and metal objects, such as fences and poles」 — khi có giông dông trong khu vực thì không có chỗ nào ngoài trời là an toàn; tránh bãi trống, đỉnh đồi hoặc đường sống lưng, tránh xa cây cao đứng lẻ hoặc vật cao khác; tránh xa nước, vật ướt như dây thừng, và vật kim loại như hàng rào, cột; khi thật sự không đi được thì「If you are in a group, spread out to avoid the current traveling between group members」 — nếu đi thành nhóm, hãy tản ra để tránh dòng điện lan trong nhóm; muốn hoạt động lại thì「Wait 30 minutes until after the last rumble of thunder」 — chờ 30 phút sau tiếng sấm cuối cùng. - Ghi chú: Lều trại, chòi, dưới gốc cây, dưới phần đá nhô ra đều không tính là chỗ che. Xe có nóc cứng thì tính, xe mui trần và xe máy thì không. Nghe tiếng sấm là bắt đầu rút, đừng đợi mưa rơi xuống mới đi. + ### 33. Lên cao nguyên thì tăng dần theo độ cao ngủ (đêm đó ngủ ở cao bao nhiêu); triệu chứng nặng thì hạ xuống, đừng gắng chịu - Chi phí: Tốn thêm một đến hai ngày để thích nghi. Thuốc phòng ngừa là thuốc kê đơn, phải hỏi bác sĩ trước khi xuất phát. Khó là lịch trình chật mà vẫn chừa ra mấy ngày đó. @@ -298,7 +282,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: A - Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Sách vàng, du lịch độ cao lớn và bệnh độ cao: 「Any unacclimatized traveler proceeding to a sleeping altitude of ≥2,450 m (≥8,000 ft)—and sometimes lower—is at risk for altitude illness」 — bất kỳ người đi du lịch chưa thích nghi nào tới độ cao ngủ ≥2.450 m (≥8.000 ft), đôi khi còn thấp hơn, đều có nguy cơ bị bệnh độ cao; tốc độ leo「500 m (1,650 ft) per night in sleeping altitude once above 3,000 m (9,800 ft)」 — 500 m (1.650 ft) mỗi đêm tính theo độ cao ngủ, một khi đã vượt 3.000 m (9.800 ft), mỗi lần lên thêm 1000 mét thì cộng thêm một ngày thích nghi; thuốc phòng ngừa acetazolamide「125 mg twice a day; 250 mg twice a day if >100 kg body weight」; nếu「If symptoms become worse despite rest or treatment at the same elevation」 thì phải hạ xuống. ；Hiệp hội Y khoa Hoang dã (2024). Hướng dẫn thực hành lâm sàng về phòng ngừa, chẩn đoán và điều trị bệnh độ cao cấp tính (cập nhật 2024). Wilderness & Environmental Medicine. - Ghi chú: Cái cần tính là「ngủ ở cao bao nhiêu」, không phải ban ngày đã lên tới cao bao nhiêu. Đau đầu kèm buồn nôn, đi lảo đảo, ý thức thay đổi, hoặc không hẳn mình cũng thở không nổi, đều là tín hiệu phải đi xuống. Hạ xuống được ưu tiên trước hít oxy và uống thuốc, vừa hít oxy vừa leo tiếp thì làm sai. - ### 34. Bị ve cắn: dùng nhíp mỏm nhọn áp sát da rồi kéo ra thật đều, đừng bôi dầu, đừng đốt bằng lửa - Chi phí: Một cây nhíp mỏm nhọn. @@ -307,7 +290,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Sau khi bị ve cắn: 「Grasp the tick as close to the skin's surface as possible using clean fine-tipped tweezers」「Pull tick away from the skin with steady, even pressure」「Do not use petroleum jelly, heat, nail polish, or other substances to try and make the tick detach from the skin. This may agitate the tick and force infected fluid from the tick into the skin」 — dùng nhíp sạch mỏm nhọn kẹp ve sát bề mặt da nhất có thể; kéo ve ra khỏi da bằng lực đều đặn; không dùng dầu mỏ, nhiệt, sơn móng tay hay vật khác để làm ve rời da, vì làm vậy có thể kích thích ve và đẩy dịch nhiễm từ ve vào da. ；Ren Q và cộng sự (2026). Epidemiological characteristics of severe fever with thrombocytopenia syndrome (SFTS) in China, 2010–2024: a fifteen-year national surveillance analysis. BMC Infectious Diseases 26:1497: 「The number of reported cases (6,602 cases) and deaths (306 cases) in 2024 were both the highest levels since the discovery of this disease」「The average CFR was 4.79%, and the CFR increased with age, reaching 9.15% among those aged 85 years and above」 — số ca báo cáo (6.602 ca) và số ca tử vong (306 ca) trong năm 2024 đều ở mức cao nhất kể từ khi phát hiện bệnh này; tỷ lệ tử vong trung bình là 4.79% và tăng theo tuổi, đạt 9.15% ở nhóm từ 85 tuổi trở lên. - Ghi chú: Rút xong thì ghi lại ngày tháng. Trong vài tuần sau đó mà sốt, nổi ban, mệt rã rời thì khi đi khám hãy nói trước là đã bị ve cắn. Bác sĩ không hỏi mà bạn không nói thì dễ bị điều trị như bị cảm. - ### 35. Nước ngoài hoang dã thì cứ đun sôi rồi uống, sôi thêm một phút là đủ, không cần nấu lâu hơn vì ở cao - Chi phí: Phải mang thêm bộ bếp và nhiên liệu. @@ -316,7 +298,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Trung tâm Kiểm soát và Phòng ngừa bệnh Hoa Kỳ. Sách vàng, khử trùng nước cho người đi du lịch: 「All organisms that may cause illness from drinking water are killed within seconds at boiling temperature」「CDC recommends that travelers boil water for a full minute to account for user variability in identifying boiling points and to add a margin of safety」「Although the boiling point for water decreases with increasing elevation, at common travel elevations the temperature needed to achieve boiling is still well above the temperature required to inactivate enteric pathogens」 — mọi sinh vật có thể gây bệnh khi uống nước đều bị giết trong vài giây ở nhiệt độ sôi; CDC khuyên người đi du lịch đun nước sôi trọn một phút, để tính tới việc người dùng nhận ra điểm sôi không giống nhau và để chừa biên an toàn; tuy điểm sôi của nước giảm khi độ cao tăng, ở những độ cao du lịch thông thường thì nhiệt độ cần thiết để đun sôi vẫn cao hơn nhiều nhiệt độ cần thiết để vô hiệu hoá mầm bệnh đường ruột. - Ghi chú: Cùng trang đó còn viết rằng không nhất thiết phải đun thật sôi: giữ ở 60 ℃ trong 30 phút cũng hiệu quả như nhau, còn đun sôi chỉ là điểm kết thúc nhìn thấy được mà không cần nhiệt kế. Không đun được thì dùng máy lọc hoặc viên khử trùng hoá chất, mỗi thứ có những mầm bệnh mà thứ kia không chặn được, đun được thì cứ đun. Dòng suối nhìn trong, nước tuyết cũng có thể có ký sinh trùng, đừng đánh giá bằng mắt. - ### 36. Bị người lạ đòi tiền ngoài hoang đương: đưa tiền cho họ, không ra tay, ghi nhớ đặc điểm, thoát ra thì báo công an - Chi phí: Tiền mặt trong người. Khó là kìm được cơn tức. @@ -325,7 +306,6 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: C - Nguồn: Quốc hội (2020). Bộ luật Hình sự, 刑法（第二十条、第二百六十三条）. (bản tổng hợp do Cục Quản lý giao thông thuộc Công an thành phố Bắc Kinh đăng lại) - Ghi chú: Lý do để ở mức C là mệnh đề「đưa tiền mà không chống trả thì an toàn hơn」 không có tài liệu trực tiếp, điều luật chỉ nói rủi ro của bên kia và giới hạn của bạn. Đừng đi theo ra chỗ vắng để「rút tiền」. Đừng đưa mật khẩu điện thoại. Đừng lên xe của họ. Trả tiền trước chỉ áp dụng cho tình huống bị chặn trực diện, tính mạng bị đe doạ, thoát ra rồi vẫn báo công an. Bị người ta dùng ảnh riêng tư hoặc video bạn trong lúc trò chuyện tình cảm đòi tiền thì ngược lại, không đưa một đồng nào, xem mục 32 ở phần 8 (đối phương cầm điểm yếu đòi tiền). - ### 37. Gặp một nhóm đánh nhau: lùi ra rồi bỏ đi, đừng vào rách, đừng đứng xem, đừng nhặt đồ nằm trên mặt đất; nếu báo công an thì lùi tới khoảng cách an toàn rồi gọi 110 - Chi phí: Không tốn tiền. Khó là nhịn cả hai cơn thèm: bước vào rách, và đứng lại xem náo nhiệt. Sau khi báo công an thì có thể bị gọi lại làm một lần biên bản, tốn nhiều nhất là nửa ngày. @@ -334,6 +314,7 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Quốc hội (1997). Bộ luật Hình sự, 刑法（第二百三十二、二百三十四、二百九十二、三百零五、三百一十条）. ；Ủy ban Thường vụ Quốc hội (2025). Luật Xử phạt quản lý trật tự công cộng, 治安管理处罚法（2025 年修订，第三十、五十一条）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院), Bộ Công an (公安部) (2020). Ý kiến chỉ đạo về việc áp dụng đúng đắn hệ thống tự vệ, 关于依法适用正当防卫制度的指导意见（法发〔2020〕31 号, mục 5 và 9）. ；Quốc hội (2018). Bộ luật Tố tụng hình sự, 刑事诉讼法（2018 年修订，第六十二、一百一十、一百一十一、一百九十三条）. Toàn văn đăng lại trên Viện kiểm sát nhân dân dân tộc trực thuộc - Ghi chú: Để ở mức B là vì phần hậu quả có văn bản luật cứng, nhưng bước「lùi ra, báo công an, không vào rách」 thì không có nguyên văn chính thức, phải suy ra từ điều luật và từ cách xác định đánh nhau hai bên. 「Tôi đi khuyên can thôi」 không phải lá chắn, khi xác định người ta xem bạn có ra tay không, đánh vào ai. Bên kia đã nằm xuống hoặc đã bỏ chạy, bạn vẫn đuổi lên đánh thì không còn tính là tự vệ. Thấy một người bị vây đánh một chiều thì về mặt pháp lý bạn thật sự có quyền ngăn lại. Tự vệ không chỉ bảo vệ bản thân mình mà cả người khác. Nhưng vào bằng tay trần thì nhiều khi chỉ thêm một người bị thương. Gọi người, quay video từ xa, báo công an thì hữu ích hơn. 「Không vào, không đứng xem」 chỉ có lợi cho bạn. Lợi ích lớn nhất của việc gọi 110 rơi vào người đang nằm dưới đất, đó là bậc thấp nhất trong nhóm người được lợi mà cuốn sách này nói tới; có bỏ thời gian hay không thì bạn tự cân. Trước khi cân, cần biết mấy điều. Nghĩa vụ làm chứng vốn đã có: Điều sáu mươi hai của Bộ luật Tố tụng hình sự viết rõ rằng ai biết tình hình vụ án đều có nghĩa vụ làm chứng, không gọi cái điện thoại đó thì nghĩa vụ cũng không biến mất. Có thương tích nặng hoặc có người chết thì cảnh sát lấy hình giám sát vẫn tìm ra người có mặt, đứng xem thì ngược lại dễ bị ghi sổ nhất. Người báo việc không muốn công khai tên thì cơ quan xử lý「应当为他保守秘密」 — phải giữ kín danh tính cho họ (Điều một trăm mười một). Nói rõ địa điểm, số người, có cầm đồ vũ khí hay không là có thể treo máy, không cần ở lại tại chỗ. Bước tới mức bị nhắc nhở hoặc bị giữ dưới mười ngày là khi tòa triệu tập bạn ra tòa, bạn không đưa ra được lý do chính đáng mà vẫn không chịu làm chứng (Điều một trăm chín mươi ba), người đi đường hiếm khi tới bước này. Không báo công an không vi phạm luật: câu「有权利也有义务」 ở Điều một trăm mười của Bộ luật Tố tụng hình sự không có điều khoản phạt, và Bộ luật Hình sự cũng không có tội phạt cho người bình thường biết mà không báo, nên bỏ đi là lựa chọn hợp pháp. Sau khi bỏ đi chỉ còn hai vạch đỏ. Một là giúp giấu người hoặc làm chứng giả để bao bọc sau đó (Điều ba trăm mười của Bộ luật Hình sự, dưới 3 năm, trường hợp nghiêm trọng từ 3 đến 10 năm). Hai là cố tình khai chứng giả khi được yêu cầu làm chứng (Điều ba trăm linh năm, dưới 3 năm, trường hợp nghiêm trọng từ 3 đến 7 năm). Im lặng và nói đúng sự thật thì không sao, bịa chuyện thì mới có chuyện. Bản thân bạn bị kéo vào xung đột, xem mục 10 ở phần 8; bị người lạ đòi tiền, xem mục 36 của phần này; chuyện tiền sau khi cứu người mà bị thương, xem mục 39 của phần này. + ### 38. Có thể đã bị phơi nhiễm HIV: trong 72 giờ đi lấy thuốc ngăn chặn, càng sớm càng tốt - Chi phí: Một liệu trình 28 ngày thuốc giá hơn nghìn tệ, ở phần lớn các nơi phải tự trả. Tối đó phải chạy tới bệnh viện một chuyến. Khó là thừa nhận ngay tại chỗ rằng mình có thể đã bị lây, rồi ra khỏi nhà ngay. @@ -342,33 +323,32 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Mức bằng chứng: B - Nguồn: Chính phủ thành phố Bắc Kinh. Trước và sau khó phân biệt? Sáu câu hỏi sáu câu trả lời để hiểu toàn diện phòng ngừa trước và sau phơi nhiễm HIV. ；Trung tâm Kiểm soát và Phòng ngừa bệnh Trung Quốc, Trung tâm Kiểm soát bệnh lây qua đường tình dục và HIV (2020). Hướng dẫn kỹ thuật phòng ngừa sau phơi nhiễm HIV (bản thử nghiệm). - Ghi chú: Lấy thuốc ở đâu: cơ sở y tế chuyên HIV do Trung tâm Kiểm soát bệnh địa phương chỉ định, thường nằm ở khoa Bệnh truyền nhiễm (phòng khám bệnh lây qua đường tình dục và HIV). Nhiều thành phố đã có phòng khám mở 24 giờ, số liệu công khai của Trùng Khánh là 29 phòng khám ở 26 quận huyện, trong đó 14 phòng mở suốt ngày. Trước khi đi, gọi điện thoại Trung tâm Kiểm soát bệnh địa phương hỏi chỗ nào có thuốc, đừng phí thời gian ở khoa cấp cứu. Các tình huống được tính là「phơi nhiễm」 gồm quan hệ tình dục không dùng biện pháp bảo vệ, bao cao su rách, bị kim đâm, máu dính lên niêm mạc; có dùng thuốc hay không thì bác sĩ quyết định. Thuốc phải uống đủ 28 ngày, bỏ mấy lần thì hiệu quả giảm rõ rệt. Sau đó theo lệnh bác sĩ, xét lại ở tuần thứ 4 đến 6 và ở tháng thứ 3. Thuốc ngăn chặn chỉ là biện pháp khắc phục; phòng ngừa thường ngày và xét nghiệm xem mục 30 và 31 ở phần 1 (dùng bao cao su suốt quá trình, có quan hệ nguy cơ cao thì đi xét). - -### 39. Cứu người mà bị thương, mất tiền: trước hết tìm người gây hại và bảo hiểm y tế, rồi hãy đi khai để được xác nhận hành động nghĩa +### 39. Cứu người mà bị thương, mất tiền: tìm người gây hại và bảo hiểm y tế trước, rồi đi khai báo hành động dũng cảm để được xác nhận -- Chi phí: Không tốn tiền. Khai báo thì phải nộp hồ sơ, chờ kết quả xác nhận, chạy vài chuyến, mất vài tuần đến vài tháng. -- Nói thẳng: Cứu người mà thiệt hại cho bản thân thì không phải tự nhận xui xẻo. Người bị thương khi làm việc nghĩa thì cứu chữa trước, tính tiền sau. Phí y tế do người gây hại trả, người đó bỏ trốn hoặc không đủ khả năng thì đi theo bảo hiểm y tế cơ bản. Không có người gây hại thì người bạn cứu, theo luật, phải cho bạn bồi thường tương xứng. Đồng thời đi khai xác nhận hành động nghĩa, nếu bị thương thì thực hiện chế độ bảo hiểm tai nạn lao động. -- Lợi ích: Điều một trăm tám mươi ba của Bộ luật Dân sự:「因保护他人民事权益使自己受到损害的，由侵权人承担民事责任，受益人可以给予适当补偿。没有侵权人、侵权人逃逸或者无力承担民事责任，受害人请求补偿的，受益人应当给予适当补偿」 — khi vì bảo vệ quyền lợi dân sự của người khác mà bản thân bị tổn hại thì người gây hại chịu trách nhiệm dân sự, người hưởng lợi có thể cho bồi thường tương xứng; không có người gây hại, người gây hại trốn đi hoặc không đủ khả năng chịu trách nhiệm dân sự, mà người bị hại yêu cầu bồi thường thì người hưởng lợi phải cho bồi thường tương xứng. Công văn số 39 năm 2012 của Văn phòng Nội các quốc gia nói xử lý người bị thương khi làm việc nghĩa thế nào. Cơ sở y tế phải lập kênh thông xanh,「坚持『先救治、后收费』的原则」. Nguyên văn về khoản phí y tế là「因紧急救治发生的医疗费用，有加害人或责任人的，由加害人或责任人承担；无加害人或责任人以及加害人或责任人逃逸或者无力承担的，按规定通过基本医疗保障制度解决」 — phí y tế phát sinh do cứu chữa khẩn cấp, có người gây hại hoặc người chịu trách nhiệm thì do người đó chịu; không có người gây hại hoặc người chịu trách nhiệm, hoặc người đó trốn đi hay không đủ khả năng chịu, thì giải quyết theo quy định qua hệ thống bảo hiểm y tế cơ bản. Người chết thì「符合烈士评定条件的，依法评定为烈士」 — nếu đủ điều kiện được phong hàm anh thì được xếp là liệt sĩ theo luật; 「属于视同工伤情形的，享受一次性工亡补助金以及相当于本人 40 个月工资的遗属特别补助金」 — nếu thuộc trường hợp coi như tai nạn lao động thì được hưởng trợ cấp tử vong một lần cùng khoản trợ cấp đặc biệt cho người thân, bằng tiền lương 40 tháng của người đó. Người bị tàn tật thì「符合享受工伤保险待遇条件的，依据《工伤保险条例》落实相应待遇；不符合的，由民政部门评定伤残等级并落实相应待遇」 — nếu đủ điều kiện hưởng chế độ bảo hiểm tai nạn lao động thì thực hiện chế độ tương ứng theo Quy định về bảo hiểm tai nạn lao động; không đủ thì cơ quan dân sự xếp cấp độ tàn tật và thực hiện chế độ tương ứng. Về việc làm thì「优先纳入就业援助，予以重点支持」 — ưu tiên đưa vào hỗ trợ việc làm, có hỗ trợ trọng điểm, chức danh công ích thì「优先安排符合条件的见义勇为人员」 — ưu tiên xếp người làm việc nghĩa đủ điều kiện. Về nhà ở thì「优先纳入住房保障体系，优先配租、配售保障性住房或发放住房租赁补贴」 — ưu tiên đưa vào hệ thống bảo đảm nhà ở, ưu tiên cho thuê, bán nhà ở bảo đảm hoặc cấp trợ cấp thuê nhà +- Chi phí: Không tốn tiền. Khai báo phải nộp hồ sơ, chờ kết luận, chạy mấy chuyến, mất mấy tuần đến mấy tháng. +- Nói thẳng: Cứu người mà để bản thân vào thế không cần tự nhận là xui xẻo. Người hành động dũng cảm mà bị thương thì cứu chữa trước, tính tiền sau. Chi phí y tế do người gây hại trả, người đó bỏ trốn hoặc không đủ khả năng trả thì đi theo bảo hiểm y tế cơ bản. Không có người gây hại thì người bạn cứu phải bồi thường bù thích hợp theo luật. Đồng thời đi khai báo hành động dũng cảm để được xác nhận, người bị thương thì áp dụng chế độ của bảo hiểm tai nạn lao động. +- Lợi ích: Điều 183 của Bộ luật Dân sự:「因保护他人民事权益使自己受到损害的，由侵权人承担民事责任，受益人可以给予适当补偿。没有侵权人、侵权人逃逸或者无力承担民事责任，受害人请求补偿的，受益人应当给予适当补偿」— người vì bảo vệ quyền lợi dân sự của người khác mà bị tổn hương thì người gây thiệt hại chịu trách nhiệm dân sự, người hưởng lợi có thể cho bù thích hợp. Không có người gây thiệt hại, người gây thiệt hại bỏ trốn hoặc không đủ khả năng chịu trách nhiệm dân sự, mà người bị hại yêu cầu bù thì người hưởng lợi phải cho bù thích hợp. Văn bản số 国办发〔2012〕39 号 của Văn phòng Hội đồng Nhà nước nói cách xử lý người hành động dũng cảm bị thương. Cơ sở y tế phải mở lối đi ưu tiên,「坚持『先救治、后收费』的原则」— giữ nguyên tắc cứu chữa trước, tính tiền sau. Nguyên văn về phần chi phí y tế là「因紧急救治发生的医疗费用，有加害人或责任人的，由加害人或责任人承担；无加害人或责任人以及加害人或责任人逃逸或者无力承担的，按规定通过基本医疗保障制度解决」— chi phí y tế do cứu chữa khẩn cấp sinh ra, có người gây hại hoặc người chịu trách nhiệm thì do người đó chịu; không có người gây hại hoặc người chịu trách nhiệm, hoặc người đó bỏ trốn hoặc không đủ khả năng chịu, thì theo quy định giải quyết qua chế độ bảo hiểm y tế cơ bản. Người tử vong thì「符合烈士评定条件的，依法评定为烈士」— đủ điều kiện được xếp liệt sĩ thì xếp liệt sĩ theo luật;「属于视同工伤情形的，享受一次性工亡补助金以及相当于本人 40 个月工资的遗属特别补助金」— thuộc trường hợp coi như tai nạn lao động thì được hưởng tiền trợ cấp một lần khi chết vì tai nạn lao động cùng khoản trợ cấp đặc biệt cho người thân bằng tương đương 40 tháng lương của người đó. Người bị tàn tật thì「符合享受工伤保险待遇条件的，依据《工伤保险条例》落实相应待遇；不符合的，由民政部门评定伤残等级并落实相应待遇」— đủ điều kiện hưởng chế độ bảo hiểm tai nạn lao động thì áp dụng chế độ tương ứng theo《工伤保险条例》; không đủ thì cơ quan dân vụ đánh giá hạng khuyết tật và thực hiện chế độ tương ứng. Về việc làm thì「优先纳入就业援助，予以重点支持」— ưu tiên đưa vào hỗ trợ việc làm, có trọng điểm; việc làm vì lợi ích cộng đồng thì「优先安排符合条件的见义勇为人员」— ưu tiên bố trí người hành động dũng cảm đủ điều kiện. Về nhà ở thì「优先纳入住房保障体系，优先配租、配售保障性住房或发放住房租赁补贴」— ưu tiên đưa vào hệ thống bảo đảm nhà ở, ưu tiên cho thuê, bán nhà ở bảo đảm hoặc cấp trợ cấp thuê nhà - Mức bằng chứng: B -- Nguồn: Quốc hội (2020). Bộ luật Dân sự, 民法典（第一百八十三条）. ；Văn phòng Nội các quốc gia chuyển đạt Bộ Dân vụ và các bộ khác (国务院办公厅转发民政部等部门) (2012). Ý kiến về tăng cường bảo vệ quyền lợi của người làm việc nghĩa, 关于加强见义勇为人员权益保护的意见（国办发〔2012〕39 号）. -- Ghi chú: Để ở mức B là vì các chế độ được hưởng bị văn bản của Trung ương ấn định, nhưng thưởng bao nhiêu, đi xác nhận thế nào thì do điều lệ về việc nghĩa của từng tỉnh quy định, số ở mỗi nơi mỗi khác, lần này chưa kiểm tra từng tỉnh. Thường thì khai với cơ quan công an nơi xảy ra chuyện hoặc với đơn vị phụ trách việc nghĩa (quỹ). Có được xác nhận hay không phụ thuộc bằng chứng: giấy biên nhận báo việc, hình giám sát tại hiện trường, tên và số điện thoại của người chứng kiến, hồ sơ bệnh án, tất cả đều lấy ngay tại chỗ. Nếu phải ra tòa thì quyền lợi dân sự của người làm việc nghĩa vốn nằm trong phạm vi trợ giúp pháp lý miễn phí. Mức trợ cấp tử vong một lần theo tiêu chuẩn toàn quốc và cách quy đổi mức tàn tật thành tiền, xem mục 15 và 16 ở phần 19. Cách xin trợ giúp pháp lý, xem mục 3 ở phần 7. +- Nguồn: Quốc hội (全国人大) (2020). Bộ luật Dân sự, 民法典（Điều 183）. ；Văn phòng Hội đồng Nhà nước chuyển tiếp Bộ Dân vụ cùng các cơ quan khác (2012). 关于加强见义勇为人员权益保护的意见（国办发〔2012〕39 号）. +- Ghi chú: Để ở mức B là vì những chế độ được hưởng do văn bản toàn quốc quy định cứng, còn thưởng bao nhiêu và thủ tục công nhận thế nào thì do điều lệ hành động dũng cảm của từng tỉnh quyết, con số khác nhau ở mỗi nơi, lượt này chưa kiểm tra từng tỉnh. Thường thì khai báo với cơ quan công an nơi xảy ra tai nạn, hoặc với cơ quan chuyên trách về hành động dũng cảm (quỹ). Có được công nhận hay không phụ thuộc bằng chứng: biên nhận báo cảnh sát, hình cảnh tại hiện trường, họ tên và số điện thoại của người chứng kiến, hồ sơ bệnh án, lấy ngay tại chỗ. Nếu phải kiện tụng, quyền lợi dân sự của người hành động dũng cảm vốn nằm trong phạm vi trợ giúp pháp lý miễn phí. Mức toàn quốc của tiền trợ cấp một lần khi chết vì tai nạn lao động và cách quy đổi hạng khuyết tật thành tiền, xem mục 14 và 15 ở phần 19. Cách xin trợ giúp pháp lý xem mục 3 ở phần 7. -### 40. Dao, thép, thuỷ tinh cắm vào người: ấn quanh vật thể để cầm máu, đừng rút nó ra +### 40. Dao, thép, kính đâm vào người: ấn cầm máu quanh vật lạ, đừng rút nó ra -- Chi phí: Không tốn tiền. Khó là nhịn cơn bản năng「rút ra trước」, và ngồi giữ người đó chờ xe cứu thương. -- Nói thẳng: Rút ra có thể gây chảy máu ồ ạt, và cũng khiến bác sĩ khó đánh giá tổn thương sâu bao nhiêu. Cách làm: gọi 120, dùng tay hoặc miếng bông để ấn quanh vật thể mà cầm máu, đệm cho vật thể nằm yên không rung. Vết thương ở bụng mà lộ ruột ra thì đừng nhét lại, phủ một miếng gạc sạch ướt lên. Vết thương ở ngực mà đang sủi bọt thì đừng bịt kín bằng nilon, bịt kín có thể làm xẹp phổi. -- Lợi ích: Hướng dẫn cấp cứu quốc tế năm 2025 của Hiệp hội Liên hợp Chữ thập Đỏ và Trăng Lưỡi liền quốc tế:「急救的人应当压住扎进身体的异物（比如刀）周围，把异物稳住，不要把它拔出来」 — người cứu nạn nên ấn quanh vật thể cắm vào cơ thể (ví dụ con dao), giữ vật thể ổn định, không rút ra. 「不要从腹部或胸部的伤口里拔出任何扎入的异物，拔出来可能造成大出血，也会让医生更难判断伤得有多重」 — không rút bất cứ vật thể nào đang cắm ở vết thương bụng hoặc ngực ra, vì rút ra có thể gây chảy máu ồ ạt và khiến bác sĩ khó đánh giá tổn thương nặng bao nhiêu. 「扎在身体里的异物应当稳住；异物随脉搏跳动的，就让它跳，松松地稳住即可」 — vật thể cắm trong cơ thể phải được giữ ổn định; nếu vật thể nhảy theo mạch thì cứ để nó nhảy, giữ lỏng là được. 「急救的人不应把露出来的内脏塞回体内」 — người cứu nạn không nên nhét ruột lộ ra trở lại vào trong. Vết thương thành ngực hở không chảy máu ra ngoài thì「可以敞着、不加敷料，因为把伤口封死可能让空气积在胸腔里，使肺部分或完全塌陷（气胸）」 — có thể để hở, không băng bó, vì bịt kín vết thương có thể làm khí tích trong lồng ngực, khiến một phần hoặc toàn bộ phổi xẹp (tràn khí). Mọi vết thương xuyên thấm vùng ngực và bụng đều phải gọi cấp cứu ngay +- Chi phí: Không tốn tiền. Khó nhất là nhịn cái phản xạ「rút ra trước」, giữ người đó và chờ xe cứu thương. +- Nói thẳng: Rút ra có thể gây chảy máu nhiều, cũng khiến bác sĩ khó đánh giá tổn thương sâu bao nhiêu. Cách làm: gọi 120, dùng tay hoặc miếng bông để ấn quanh vật lạ cầm máu, chèn cho vật lạ nằm yên không rung. Vết thương ở bụng mà lộ ruột ra thì đừng nhét lại, đặt lên một miếng gạc ẩm sạch. Vết thương ở ngực mà đang sủi bọt thì đừng bịt kín bằng nilông, bịt kín có thể làm phổi xẹp. +- Lợi ích: Hướng dẫn cấp cứu quốc tế năm 2025 của Hiệp đoàn Liên hợp Chữ thập đỏ và Trăng lưỡi liền Quốc tế:「急救的人应当压住扎进身体的异物（比如刀）周围，把异物稳住，不要把它拔出来」— người cứu hộ phải ấn quanh vật lạ đâm vào cơ thể (ví dụ con dao), giữ vật lạ cho vững, đừng rút ra.「不要从腹部或胸部的伤口里拔出任何扎入的异物，拔出来可能造成大出血，也会让医生更难判断伤得有多重」— đừng rút bất kỳ vật lạ nào đâm vào qua vết thương ở bụng hoặc ngực, rút ra có thể gây chảy máu nhiều, cũng khiến bác sĩ khó đánh giá tổn thương nặng bao nhiêu.「扎在身体里的异物应当稳住；异物随脉搏跳动的，就让它跳，松松地稳住即可」— vật lạ đâm trong cơ thể phải được giữ cho vững; vật lạ nhảy theo mạch thì cứ để nó nhảy, giữ lỏng là được.「急救的人不应把露出来的内脏塞回体内」— người cứu hộ không nên nhét ruột lộ ra trở lại vào trong. Vết thương thành ngực hở không máu chảy ra ngoài「可以敞着、不加敷料，因为把伤口封死可能让空气积在胸腔里，使肺部分或完全塌陷（气胸）」— có thể để hở, không băng bó, vì bịt kín vết thương có thể làm không khí tích trong lồng ngực khiến phổi xẹp một phần hoặc toàn bộ (tràn khí). Mọi vết thương xuyên thành ngực và bụng đều phải gọi cấp cứu ngay - Mức bằng chứng: B - Nguồn: International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, trang 256–257, 267–268. -- Ghi chú: Để ở mức B là vì đây là cách làm được hướng dẫn công nhận, không có thử nghiệm chia nhóm ngẫu nhiên, tất nhiên không thể chia người bị thương thành hai nhóm「rút」 và「không rút」 để thử. Ngoại lệ duy nhất là khi vật thể chặn khí quản, người đó không thở nổi. Tới bệnh viện thì đi thẳng tới cái bàn phân loại cấp thiết trước cửa khoa cấp cứu (bàn tiếp đón và phân loại), báo「vết thương xuyên thấm」. Đây là mức phải vào phòng cấp cứu ngay, đừng xếp hàng ở quầy đăng ký khám (xem mục 8 ở phần 24). Thật sự bị xuyên thấm thì gọi 120, xe cứu thương sẽ báo trước bệnh viện để chuẩn bị tiếp nhận. Chảy máu ồ ạt thì ấn thế nào, đặt gar thế nào, xem mục 12 của phần này. +- Ghi chú: Để ở mức B là vì đây là cách làm được mọi người công nhận trong hướng dẫn, không có thử nghiệm ngẫu nhiên, tất nhiên không thể chia người bị thương thành hai nhóm「rút」và「không rút」để thử. Ngoại lệ duy nhất là vật lạ bịt đường hô hấp khiến người đó không thở được. Tới bệnh viện thì đi thẳng tới bàn phân loại mức độ ở cửa cấp cứu (bàn tiếp đón sàng lọc), báo là「vết thương xuyên」. Đây là mức phải vào phòng cấp cứu ngay, đừng xếp hàng ở quầy đăng ký khám (xem mục 7 ở phần 24). Đã bị đâm xuyên thì gọi 120, xe cứu thương sẽ báo trước cho bệnh viện chuẩn bị tiếp nhận. Chảy máu nhiều thì ấn thế nào và đặt gar thế nào xem mục 12 ở phần này. -### 41. Nghi ngờ gãy xương thì đừng cử động chỗ đó nữa, đỡ cho cố định, chườm lạnh trong vòng 20 phút, đừng tự nắn lại +### 41. Nghi gãy xương thì đừng cử động chỗ đó nữa, đỡ cho yên, chườm lạnh không quá 20 phút, đừng tự bẻ lại -- Chi phí: Không tốn tiền. Tìm một tấm ván cứng, tạp chí cuộn lại hoặc quần áo làm chỗ tựa. Chỗ chườm lạnh phải có một lớp vải ngăn ở giữa. -- Nói thẳng: Chân tay bị thương thì cứ coi là gãy xương trước, đứng được, cử động được cũng không loại trừ được, chỉ X quang mới xác định. Để vùng bị thương giữ nguyên vị trí, dùng tay đỡ, hoặc dùng khăn tam giác, ván cứng chống theo đúng vị trí cũ. Chườm lạnh càng sớm càng tốt, mỗi lần không quá 20 phút. Đừng tự nắn lại. Xương gãy xuyên ra da thì cầm máu trước, phủ gạc sạch, đừng ấn xương trở lại. -- Lợi ích: Hiệp hội Liên hợp Chữ thập Đỏ và Trăng Lưỡi liền quốc tế đã ban hành hướng dẫn cấp cứu quốc tế năm 2025. Trong đó viết「手脚上的任何一处伤，都应当先按可能的骨折来处理」 — bất kỳ chỗ nào trên chân tay bị thương, đều nên xử lý trước như là gãy xương có thể. Còn viết「确诊骨折唯一准确的办法是拍 X 光。拿不准就当骨折办」 — cách duy nhất chính xác để xác nhận gãy xương là chụp X quang, không chắc thì cứ coi như gãy xương. Mục đích của việc cố định là「尽量减少活动和疼痛」 — hạn chế cử động và đau nhất có thể. Chườm lạnh thì「一次最多连续 20 分钟，这样可以减轻疼痛；冷得比这更久可能损伤皮肤」 — mỗi lần tối đa 20 phút liên tục, như vậy giảm đau, chườm lâu hơn thế có thể làm tổn thương da. 「伤到股骨这类长骨时，光是骨头本身出的血就可能致命」 — khi bị thương ở xương dài như xương đùi, chỉ riêng máu chảy ra từ xương đã có thể gây tử vong. Khi đau hoặc sưng rõ, có dấu hiệu sốc, gãy xương đùi hoặc chi thể bị sai vị trí rõ ràng thì nên gọi cấp cứu. Gãy xương cong thành góc thì phải có đồng thời hai điều kiện mới nghĩ tới việc nắn về nguyên vị trí ngay tại chỗ (复位). Một là「地处偏远（或者条件有限），而且成角处下方的肢体发凉、发白，或者急救的人没法固定好以便搬动」 — ở nơi xa xôi (hoặc điều kiện hạn chế), và phần chi thể dưới chỗ gãy góc bị lạnh, tái, hoặc người cứu nạn không cố định được để cử động. Hai là người cứu nạn đã qua huấn luyện. Với gãy xương hở (xương gãy xuyên ra da) thì việc trước tiên là cầm máu và phòng nhiễm trùng +- Chi phí: Không tốn tiền. Tìm một tấm ván cứng, tạp chí cuộn lại hoặc quần áo để làm chỗ tựa. Chườm vật lạnh thì có một lớp vải ở giữa. +- Nói thẳng: Chân tay bị thương thì cứ coi như gãy xương trước, đứng được, cử động được cũng không loại trừ được, chỉ chụp X-quang mới xác định. Giữ chỗ bị thương không cử động, dùng tay đỡ, hoặc dùng khăn tam giác, tấm ván cứng chống lại ở đúng vị trí cũ. Chườm lạnh càng sớm càng tốt, mỗi lần không quá 20 phút. Đừng tự bẻ lại. Xương gãy xuyên da thì cầm máu trước, phủ gạc sạch, đừng ấn xương vào trong. +- Lợi ích: Hiệp đoàn Liên hợp Chữ thập đỏ và Trăng lưỡi liền Quốc tế đã ban hành hướng dẫn cấp cứu quốc tế năm 2025. Trong đó viết「手脚上的任何一处伤，都应当先按可能的骨折来处理」— bất cứ vết thương nào ở chân tay đều phải xử lý trước như là gãy xương. Còn viết「确诊骨折唯一准确的办法是拍 X 光。拿不准就当骨折办」— cách chính xác duy nhất để xác nhận gãy xương là chụp X-quang, không chắc thì cứ coi là gãy xương. Mục đích của việc cố định là「尽量减少活动和疼痛」— giảm bớt cử động và đau. Chườm lạnh thì「一次最多连续 20 分钟，这样可以减轻疼痛；冷得比这更久可能损伤皮肤」— mỗi lần tối đa 20 phút liên tục, như vậy giảm đau, lạnh lâu hơn thế có thể làm tổn thương da.「伤到股骨这类长骨时，光是骨头本身出的血就可能致命」— khi tổn thương xương dài như xương đùi, riêng máu từ xương đã có thể gây chết. Khi đau hoặc sưng rõ, có dấu hiệu sốc, gãy xương đùi (xương đùi) hoặc chi đứng sai hẳn vị trí thì phải gọi cấp cứu. Gãy xương chúc thì phải có đồng thời hai điều kiện mới nghĩ tới bẻ về nguyên vị trí tại hiện trường (đặt lại). Một là「地处偏远（或者条件有限），而且成角处下方的肢体发凉、发白，或者急救的人没法固定好以便搬动」— ở nơi xa xôi (hoặc điều kiện hạn chế), và phần chi nằm dưới chỗ chúc thì lạnh, tái, hoặc người cứu hộ không cố định được để vận chuyển. Hai là người cứu hộ đã qua huấn luyện. Với gãy xương hở (xương gãy xuyên da) thì việc trước tiên là cầm máu và chống nhiễm trùng - Mức bằng chứng: B - Nguồn: International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, trang 301–302. -- Ghi chú: Để ở mức B là vì đây đều là những cách làm được công nhận. Kéo giãn cố định cho vết gãy xương đùi trước khi đưa đi bệnh viện thì chỉ có một nghiên cứu hồi sát bệnh án cũ làm căn cứ, độ tin cậy của bằng chứng rất thấp. Kết quả là lượng máu truyền trong 24 giờ ít hơn, thời gian nằm viện ngắn hơn, còn khác biệt về tử vong và biến chứng nhỏ đến mức không nói rõ được. Còn về việc nên nắn về nguyên vị trí trước hay cố định nguyên trạng thì hẳn là không tìm thấy bằng chứng. Ba hiểu lầm thường gặp cần sửa: đứng được đi được không có nghĩa là không gãy xương; chỉ dùng lạnh không dùng nóng, cũng đừng luân phiên nóng lạnh; chấn thương trẹo không phải「vận động một chút là khỏi」, cứ nghỉ trước. Người lớn tuổi ngã mà nghi ngờ gãy xương thì có những điều cấm khi di chuyển, xem mục 2 của phần này. Sau khi bó bột hoặc nằm liệt giường mà một chân sưng lên thì cảnh giác huyết khối, xem mục 11 của phần này. Sau khi chữa xong có cần làm giám định mức tàn tật, có cần làm giấy chứng nhận người khuyết tật không, xem mục 10 và 11 ở phần 24. +- Ghi chú: Để ở mức B là vì tất cả những thứ này đều là cách làm được mọi người công nhận. Việc kéo nắn cố định gãy xương đùi (xương đùi) trước khi đưa đi viện chỉ có một nghiên cứu cũ rà lại bệnh án làm chỗ dựa, độ tin cậy rất thấp. Kết quả là lượng máu truyền trong 24 giờ ít hơn, thời gian nằm viện ngắn hơn, còn khác biệt về tử vong và biến chứng nhỏ tới mức không nói được. Còn chuyện nên đặt lại xương trước (đưa xương về nguyên vị trí) hay cố định nguyên trạng thì hoàn toàn không tìm được bằng chứng. Ba hiểu lầm thường gặp phải sửa: đứng được đi được không có nghĩa là không gãy xương; chỉ dùng lạnh không dùng nóng, cũng đừng luân phiên nóng lạnh; trật khớp không phải「vận động cho ra là khỏi」, nghỉ trước đã. Sau khi người cao tuổi ngã mà nghi gãy xương thì các điều cấm cử di chuyển xem mục 2 ở phần này. Sau khi bó bột hay nằm liệt giường mà một chân sưng lên thì cảnh giác huyết khối, xem mục 11 ở phần này. Sau khi chữa xong có cần làm giám định tàn tật, có cần xin giấy chứng nhận người khuyết tật hay không, xem mục 9 và 10 ở phần 24. ### 42. Sau khi bị cưỡng ép tình dục: tới nơi an toàn rồi gọi 110; trước khi khám lạm dụng thì đừng tắm, đừng giặt quần áo, đừng dọn phòng, trong 72 giờ hãy tới bệnh viện @@ -377,4 +357,22 @@ Phần này chỉ gom những việc phải làm trong vòng mấy phút. Nhận - Lợi ích: Kiểm sát viên của Viện kiểm sát nhân dân thành phố Bắc Kinh nói:「无论是未成年人还是成年人，遇到侵害，都应当第一时间拨打110报警」 — dù là người chưa thành niên hay người lớn, khi gặp xâm hại thì phải gọi 110 báo ngay lập tức. Báo công an và khám chữa kịp thời thì「便于侦查机关及时勘查现场、提取物证痕迹、固定证据」 — thuận tiện cho cơ quan điều tra khám hiện trường kịp thời, thu dấu vật chứng và vết tích, cố định bằng chứng. Sổ tay phòng chống xâm hại tình dục trên Nhân Dân Nhật Báo mà Sở Dân vụ tỉnh Quý Châu đăng lại viết:「一旦不幸遭遇性侵害，最重要的事情就是留存证据」 — một khi không may gặp xâm hại tình dục thì việc quan trọng nhất là lưu lại bằng chứng. Cần giữ「体液、毛发、皮屑、通信记录等证据」 — dịch vật cơ thể, lông tóc, vảy da, lịch sử liên lạc và các bằng chứng khác — và「被侵害后，不要擅自洗澡、洗衣物、收拾房间、扔东西」 — sau khi bị xâm hại thì không tự ý tắm, không giặt quần áo, không dọn phòng, không vứt bỏ đồ. Với người quen gây ra chuyện thì giữ lại lịch sử cuộc gọi và liên lạc. Bên kia sau đó đe doạ hay mua chuộc thì cũng phải để lại bằng chứng. Viện kiểm sát nhân dân quận Tây thành phố Bắc Kinh viết:「在接受身体检查之前尽量不要洗澡」 — trước khi thăm khám thể chất nên hạn chế tắm. Tổ chức Y tế Thế giới nói thuốc tránh thai khẩn cấp dùng được khi「女性在未采取有效的避孕保护措施时遭受性侵犯」 — phụ nữ bị xâm hại tình dục khi chưa có biện pháp tránh thai bảo vệ có hiệu quả. Khuyến cáo dùng trong 5 ngày, càng sớm càng hiệu quả. Thuốc tránh thai khẩn cấp chứa levonorgestrel (một loại nội tiết tố thai khoa), tỷ lệ mang thai ở phụ nữ đã dùng là 1.2% đến 2.1%. Viên levonorgestrel bán ở hiệu thuốc trong nước là thuốc không kê đơn. Ý kiến của Ủy ban Sức khoẻ thành phố Vũ Hán: sau quan hệ tình dục không có biện pháp bảo vệ thì「72小时以内，服药越早，预防妊娠效果越好」 — trong 72 giờ, uống thuốc càng sớm thì hiệu quả phòng mang thai càng tốt. Thuốc ngăn chặn HIV phải uống trong 2 giờ nếu có thể, muộn nhất không quá 72 giờ (toàn quốc) - Mức bằng chứng: B - Nguồn: Viện kiểm sát nhân dân thành phố Bắc Kinh (2024). Nhìn phim qua lăng kính pháp lý丨xem《九部的检察官》, người chưa thành niên gặp xâm hại trái pháp nên tự bảo vệ thế nào? (đăng lại từ Đài Phát thanh Trung Quốc của Tổng cục Phát thanh và Truyền hình quốc gia). ；Sở Dân vụ tỉnh Quý Châu (2024). Sổ tay phòng chống xâm hại tình dục cho người chưa thành niên, phát cho phụ huynh và các cháu! (đăng lại từ Nhân Dân Nhật Báo). ；Viện kiểm sát nhân dân quận Tây thành phố Bắc Kinh (2018). Vì sao họ không còn im lặng nữa…… ；Tổ chức Y tế Thế giới (2021). Tránh thai khẩn cấp (thông tin thực trạng). ；Ủy ban Sức khoẻ thành phố Vũ Hán (2020). Dùng thuốc tránh thai khẩn cấp an toàn và hợp lý. ；Chính phủ thành phố Bắc Kinh. Trước và sau khó phân biệt? Sáu câu hỏi sáu câu trả lời để hiểu toàn diện phòng ngừa trước và sau phơi nhiễm HIV. -- Ghi chú: Để ở mức B là vì đây là kinh nghiệm điều tra và khuyến cáo dùng thuốc của cơ quan chính thức, không có con số「làm theo thì tỷ lệ kết án cao hơn bao nhiêu」. Người được lợi chủ yếu là bạn và gia đình. Con số 72 giờ của thuốc tránh thai khẩn cấp là theo hướng dẫn trong nước, Tổ chức Y tế Thế giới nói dùng được trong 5 ngày, nhưng càng muộn càng tệ, đừng kéo dài. Đã có thai rồi thì thuốc tránh thai khẩn cấp không có tác dụng. Khi báo công an thì nói đúng sự thật, không giấu, không thổi phồng. Bên kia sau đó tìm bạn thương lượng, đưa tiền hay đe doạ thì đều để lại hồ sơ, đừng xoá. Nghĩ kỹ rồi báo cũng được, nhưng bằng chứng sẽ mất dần theo thời gian. Báo hay không báo thì cả hai trường hợp đều phải tới bệnh viện trong hạn. Chuyện ghi âm, xem mục 41 ở phần 8 (ghi âm). Bên kia sau đó dùng ảnh hoặc chuyện báo công an để dọa bạn thì cũng là phạm pháp, xem mục 32 ở phần 8 (đối phương dọa bằng việc báo công an, gửi ảnh để đòi tiền). Người bạn gặp chuyện này thì hãy đi cùng họ để báo công an và tới bệnh viện, đừng khuyên họ tắm hay nghỉ ngơi trước. \ No newline at end of file +- Ghi chú: Để ở mức B là vì đây là kinh nghiệm điều tra và khuyến cáo dùng thuốc của cơ quan chính thức, không có con số「làm theo thì tỷ lệ kết án cao hơn bao nhiêu」. Người được lợi chủ yếu là bạn và gia đình. Con số 72 giờ của thuốc tránh thai khẩn cấp là theo hướng dẫn trong nước, Tổ chức Y tế Thế giới nói dùng được trong 5 ngày, nhưng càng muộn càng tệ, đừng kéo dài. Đã có thai rồi thì thuốc tránh thai khẩn cấp không có tác dụng. Khi báo công an thì nói đúng sự thật, không giấu, không thổi phồng. Bên kia sau đó tìm bạn thương lượng, đưa tiền hay đe doạ thì đều để lại hồ sơ, đừng xoá. Nghĩ kỹ rồi báo cũng được, nhưng bằng chứng sẽ mất dần theo thời gian. Báo hay không báo thì cả hai trường hợp đều phải tới bệnh viện trong hạn. Chuyện ghi âm, xem mục 41 ở phần 8 (ghi âm). Bên kia sau đó dùng ảnh hoặc chuyện báo công an để dọa bạn thì cũng là phạm pháp, xem mục 32 ở phần 8 (đối phương dọa bằng việc báo công an, gửi ảnh để đòi tiền). Người bạn gặp chuyện này thì hãy đi cùng họ để báo công an và tới bệnh viện, đừng khuyên họ tắm hay nghỉ ngơi trước. + +### 43. Trẻ sơ sinh dưới 1 tuổi bị nghẽn khóc không ra tiếng: úp sấp xuống vỗ lưng 5 cái, lật lại ấn ngực 5 cái, làm xen kẽ, đừng ấn bụng + +- Chi phí: Không tốn tiền. Xem lại thao tác một lần, hoặc học lại lớp cấp cứu trẻ sơ sinh, mất mấy phút đến nửa ngày. Khó là lúc hối hả phải lật trẻ úp sấp xuống cho đúng. +- Nói thẳng: Trẻ còn ho, còn khóc ra tiếng thì cứ để trẻ tự ho, bạn đứng bên cạnh canh. Khóc không ra tiếng, mặt tím tái thì bảo người đó gọi 120, úp trẻ mặt xuống nằm úp trên cẳng tay, dùng gốc bàn tay vỗ 5 cái vào giữa hai bản vai. Không ra thì lật trẻ lại, ấn nhanh 5 cái vào chính giữa ngực, làm xen kẽ hai cách, đừng ấn bụng. Trẻ không phản ứng thì chuyển sang làm hồi sinh tim phổi. +- Lợi ích: Hiệp hội Tim Hoa Kỳ và Hiệp hội Nhi khoa Hoa Kỳ đã ban hành hướng dẫn hồi sinh tim phổi nền tảng cho trẻ em năm 2025. Hướng dẫn viết: nghẹt thở (ví dụ vật lạ bịt đường hô hấp)「是婴儿和儿童的主要死因之一」— là một trong các nguyên nhân tử vong chính ở trẻ sơ sinh và trẻ em. Thứ trẻ sơ sinh hay hít vào nhất là chất lỏng. Còn ho, còn ra tiếng là thể nhẹ, cứ để trẻ tự ho đồng thời theo dõi xem có nặng thêm không. Ho yếu, không ho được, khóc không ra, mặt tím tái là thể nặng. Thể nặng thì gọi cấp cứu trước, rồi「反复做 5 次拍背、接 5 次胸部冲击，直到异物出来或者婴儿失去反应」— lặp lại vỗ lưng 5 lần rồi đấm ngực 5 lần cho tới khi vật lạ ra ngoài hoặc trẻ sơ sinh không còn phản ứng. Hướng dẫn không khuyên đấm bụng cho trẻ sơ sinh, vì có thể làm tổn thương cơ quan trong bụng. Đấm ngực hiện nay khuyên dùng gốc bàn tay một tay, vì ấn sâu hơn cách dùng hai ngón tay trước đây. Trẻ sơ sinh không còn phản ứng thì bắt đầu hồi sinh tim phổi, ấn trước, không bắt mạch. Trước khi thổi khí, vật lạ nhìn thấy được thì lấy ra. Hướng dẫn nói rõ「不要用手指盲目去掏」— đừng mù quáng mò bằng ngón tay. Hướng dẫn cấp cứu quốc tế năm 2025 của Hiệp đoàn Liên hợp Chữ thập đỏ và Trăng lưỡi liền Quốc tế cũng là bộ này. Vỗ thật mạnh tối đa 5 cái vào giữa hai bản vai, không được thì lật trẻ lại ấn ngực tối đa 5 cái, vẫn không được thì gọi cấp cứu ngay. Cách ấn ngực của họ là ôm ngực bằng hai tay, dùng hai ngón cái ấn. Trung tâm Đào tạo Cứu hộ Khẩn cấp Hội Chữ thập đỏ tỉnh Phúc Kiến nói, trẻ sơ sinh dưới 1 tuổi phải làm「5 次背部拍击 + 5 次胸部冲击」— vỗ lưng 5 lần cộng đấm ngực 5 lần. Khi vỗ lưng thì để trẻ sơ sinh nằm úp trên cẳng tay, đầu thấp hơn thân +- Mức bằng chứng: B +- Nguồn: Joyner BL và cộng sự (2025). Part 6: Pediatric Basic Life Support: 2025 American Heart Association and American Academy of Pediatrics Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 152(suppl 2):S424–S447. , toàn văn xem ；International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, trang 238–240. ；Sở Y tế và Sức khoẻ thành phố Phúc Châu (福州市卫生健康委员会) (2026). 海姆立克急救法更新了. +- Ghi chú: Để ở mức B là vì cái chính đỡ cho bộ thao tác này chủ yếu là các báo cáo ca lâm sàng, không có con số「làm theo thì giảm được bao nhiêu người chết」. Mức lợi ích để「lớn」là đánh giá: một đường hô hấp bị bịt kín hoàn toàn sẽ làm tim trẻ ngừng trong vòng mấy phút. Người hưởng gần như luôn là con mình. Trẻ trên 1 tuổi và người lớn thì đấm bụng, xem mục 26 ở phần này (nghẹn thì vỗ lưng và đấm bụng). Dùng thao tác nào khi đấm ngực thì mỗi nơi viết khác nhau. Hướng dẫn Hoa Kỳ và bản mới của Hội Chữ thập đỏ trong nước dùng gốc bàn tay một tay, Hiệp đoàn Chữ thập đỏ Quốc tế dùng hai ngón cái, hướng dẫn cũ dùng hai ngón tay. Hướng dẫn Hoa Kỳ năm 2025 đã không còn khuyên dùng hai ngón tay, vì ấn không đủ sâu. Lúc cấp cứu mà dùng cách nào cũng hơn là không ấn. Vật lạ đã ho ra vẫn nên đi viện, có thể còn sót mảnh vụn trong phổi. Đồng tiền, đồ chơi nhỏ loại đó thì để ở chỗ trẻ với tới không được. Sau khi trẻ không phản ứng thì làm tiếp thế nào, xem mục 44 ở phần này (hồi sinh tim phổi trẻ sơ sinh). + +### 44. Trẻ sơ sinh dưới 1 tuổi không phản ứng, thở không bình thường: bật loa ngoài gọi 120, làm hồi sinh tim phổi trẻ sơ sinh: ấn 30 cái thổi 2 hơi + +- Chi phí: Không tốn tiền. Hồi sinh tim phổi trẻ sơ sinh phải học một lần, nửa ngày đến một ngày. Khó là thổi khí vào miệng mũi trẻ, và không dám ấn mạnh. +- Nói thẳng: Vỗ bàn chân mà không phản ứng, thở không bình thường thì bật loa ngoài gọi 120, tay không dừng. Ôm ngực bằng hai tay, dùng hai ngón cái ấn, không ôm nổi thì dùng gốc bàn tay một tay, ấn sâu khoảng 4 centimet, khoảng 2 cái mỗi giây. Cứ ấn 30 cái thì dùng miệng úp lên miệng mũi trẻ thổi 2 hơi. Tim trẻ sơ sinh ngừng phần lớn là do không thở được, làm hồi sinh tim phổi có thổi khí thì sống sót nhiều hơn. +- Lợi ích: Hiệp hội Tim Hoa Kỳ và Hiệp hội Nhi khoa Hoa Kỳ đã ban hành hướng dẫn hồi sinh tim phổi nền tảng cho trẻ em năm 2025. Hướng dẫn viết: tim trẻ sơ sinh và trẻ em ngừng, nguyên nhân chính vẫn là hô hấp có vấn đề. Trẻ sơ sinh bị ngừng tim ngoài bệnh viện chỉ có 6.6% sống tới lúc xuất viện. Người bình thường gặp trẻ sơ sinh hoặc trẻ em không phản ứng, thở không bình thường, không có dấu hiệu sống「应当立即开始心肺复苏」— phải bắt đầu hồi sinh tim phổi ngay, đừng bắt mạch. Độ sâu ấn tối thiểu là một phần ba đường kính trước sau của lồng ngực, với trẻ sơ sinh là「约 1.5 英寸（4 厘米）」— khoảng 1.5 inch (4 centimet). Tần số từ 100 đến 120 lần mỗi phút. Khi một người làm thì ấn 30 cái thổi 2 hơi. Khi hai người làm thì ấn 15 cái thổi 2 hơi. Trẻ sơ sinh dùng gốc bàn tay một tay, hoặc ôm ngực bằng hai tay rồi dùng hai ngón cái ấn. Không ôm nổi thì dùng gốc bàn tay. Cách ấn bằng hai ngón tay đã bị bỏ, vì ấn không đủ sâu. Hướng dẫn còn viết, với trẻ sơ sinh và trẻ em bị ngừng tim ngoài bệnh viện, thêm thổi khí vào sau khi ấn sẽ nâng được tỷ lệ sống. Một sổ đăng ký ở Hoa Kỳ thống kê 3900 trẻ dưới 18 tuổi bị ngừng tim ngoài bệnh viện, trong đó 59.4% là trẻ sơ sinh. Người có người bên cạnh làm hồi sinh tim phổi thì tỷ lệ sống tới lúc xuất viện sau điều chỉnh là 13.2%, không làm thì 9.5%. OR sau điều chỉnh 1.57 (xác suất sống sót cao hơn khoảng 57%), 95% CI 1.25–1.96 (khoảng tin cậy). Với riêng trẻ sơ sinh thì hồi sinh tim phổi có thổi khí gắn với tỷ lệ sống cao hơn, trẻ sống sót cũng hồi phục não tốt hơn. Chỉ ấn mà không thổi thì kết quả gần như không khác gì không ai làm. Hướng dẫn cấp cứu quốc tế năm 2025 của Hiệp đoàn Liên hợp Chữ thập đỏ và Trăng lưỡi liền Quốc tế viết: xem trẻ sơ sinh có phản ứng không thì bế trẻ lên hoặc vỗ bàn chân. Nếu chỉ có một mình bạn thì bật loa ngoài tự gọi số cấp cứu. Nếu bên cạnh không có chiếc điện thoại vừa ấn vừa gọi thì làm 1 phút hồi sinh tim phổi rồi hãy gọi. Viện Nghiên cứu Nhi khoa Bắc Kinh có một bài phổ biến, Sở Y tế và Sức khoẻ thành phố Bắc Kinh đăng lại. Bài đó viết:「如果只有1个人在场，先进行2分钟的心肺复苏后呼叫120救护车」— nếu chỉ có một người ở hiện trường thì làm hai phút hồi sinh tim phổi trước rồi gọi xe cứu thương 120. Biên độ ấn là「婴儿4cm」— trẻ sơ sinh 4 cm. Thổi khí cho trẻ dưới 1 tuổi phải「口对口鼻」— miệng úp miệng mũi +- Mức bằng chứng: B +- Nguồn: Joyner BL và cộng sự (2025). Part 6: Pediatric Basic Life Support: 2025 American Heart Association and American Academy of Pediatrics Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 152(suppl 2):S424–S447. , toàn văn xem ；Naim MY và cộng sự (2017). Association of Bystander Cardiopulmonary Resuscitation With Overall and Neurologically Favorable Survival After Pediatric Out-of-Hospital Cardiac Arrest in the United States: A Report From the Cardiac Arrest Registry to Enhance Survival Surveillance Registry. JAMA Pediatrics 171(2):133–141. ；International Federation of Red Cross and Red Crescent Societies (2025). International First Aid, Resuscitation and Education Guidelines 2025, trang 186–188. ；Viện Nghiên cứu Nhi khoa Bắc Kinh (2022). 【儿研所儿医说】心肺复苏：爸妈必备的急救知识（đăng lại bởi Sở Y tế và Sức khoẻ thành phố Bắc Kinh）. ；European Resuscitation Council (2025). ERC Guidelines 2025 for the lay public（một chương về Paediatric Basic Life Support）. +- Ghi chú: Để ở mức B là vì con số sống sót đến từ nghiên cứu sổ đăng ký chỉ ghi nhận và theo dõi, lại tính trên tất cả trẻ em chứ không riêng trẻ sơ sinh, phần tóm tắt không đưa con số riêng cho trẻ sơ sinh. Mức lợi ích để「lớn」là tính theo số của toàn bộ trẻ em: sống tới lúc xuất viện từ 9.5% lên 13.2%, tương đối thêm khoảng bốn phần mười. Người hưởng gần như luôn là con mình. Có ba điểm khác với mục 1 ở phần này (ấn mạnh vào giữa ngực). Một là trẻ sơ sinh phải có thổ khí, chỉ ấn không thổi thì với trẻ sơ sinh kém hiệu quả. Hai là dùng hai ngón cái hoặc gốc bàn tay một tay, ấn khoảng 4 centimet, không chồng hai tay ấn xuống. Ba là khi một người mà không vừa ấn vừa gọi điện thoại được thì ấn trước rồi hãy gọi. Ấn bao lâu thì mỗi nơi khác nhau: quy trình một người của Viện Nghiên cứu Nhi khoa Bắc Kinh và hướng dẫn Hoa Kỳ là 2 phút, Hiệp đoàn Chữ thập đỏ Quốc tế là 1 phút. Bật loa ngoài điện thoại thì không còn vấn đề này. Ủy ban Hồi sinh Tim phổi châu Âu dạy thổi 5 hơi trước rồi mới ấn, hướng dẫn Hoa Kỳ là ấn trước thổi sau, cả hai đều hơn là không làm. Hướng dẫn cũ ở trong nước dạy ấn bằng hai ngón tay, hướng dẫn Hoa Kỳ năm 2025 đã bỏ thao tác đó. Nếu vì nghẹn mà gây ra thì trước khi thổi khí, nhìn vào miệng một cái, chỉ lấy vật lạ mà nhìn thấy, xem mục 43 ở phần này (trẻ sơ sinh bị nghẽn). diff --git "a/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" "b/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" index 2477b6a2..c3a949af 100644 --- "a/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" +++ "b/vi/book/14-\350\264\246\345\217\267\344\270\216\344\277\241\346\201\257\345\256\211\345\205\250.md" @@ -12,7 +12,6 @@ Thước đo: tiền và thông tin cá nhân. Khi tài khoản của bạn bị - Mức bằng chứng: A - Nguồn: Doerfler P, Thomas K, Marincenko M, et al. (2019). Evaluating Login Challenges as a Defense Against Account Takeover. The World Wide Web Conference (WWW '19). - Ghi chú: Cùng nghiên cứu đó cũng thấy loại xác thực này đôi khi chặn nhầm chính người dùng thật. 52% người dùng thật lần đầu không đăng nhập được, nhưng cuối cùng 97% vẫn vào được. Hãy bật tính năng này cho email trước, vì phần lớn tài khoản khác đều dùng email để tìm lại mật khẩu. - ### 2. Đặt mật khẩu riêng cho email, không dùng lại ở bất kỳ trang nào khác - Chi phí: Không tốn tiền. Lưu vào trình quản lý mật khẩu thì không phải tự nhớ. Khó là bỏ được thói quen dùng một mật khẩu ở khắp mọi nơi. @@ -21,7 +20,6 @@ Thước đo: tiền và thông tin cá nhân. Khi tài khoản của bạn bị - Mức bằng chứng: C - Nguồn: US CISA. Use Strong Passwords. - Ghi chú: Không nhớ được thì dùng trình quản lý mật khẩu có sẵn trong trình duyệt. Nó giữ mật khẩu từng trang giúp bạn, tốt hơn nhiều so với dùng chung một mật khẩu khắp mọi nơi. Đừng lưu mật khẩu trong mục yêu thích của WeChat hay trong ứng dụng ghi chú. - ### 3. Đặt mật khẩu khóa màn hình điện thoại, đặt mã PIN cho thẻ SIM - Chi phí: Không tốn tiền. Mật khẩu khóa màn hình và mã PIN mỗi thứ đặt một lần là xong. @@ -30,7 +28,6 @@ Thước đo: tiền và thông tin cá nhân. Khi tài khoản của bạn bị - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp - Ghi chú: Mã PIN nằm ở mục「khóa thẻ SIM」trong phần cài đặt của điện thoại. Mã ban đầu khi mới xuất xưởng thường là 1234 hoặc 0000. Nhập sai ba lần liên tiếp thì phải dùng mã PUK do nhà mạng cung cấp mới mở được. Nên sau khi đặt xong, ghi mã đó ra giấy trước. - ### 4. Điện thoại bị mất thì làm theo đúng thứ tự này: báo mất thẻ SIM, khóa từ xa, đổi mật khẩu, báo công an, phong tỏa thẻ ngân hàng - Chi phí: Không tốn tiền. Làm trọn bộ mất mười mấy phút. @@ -39,7 +36,6 @@ Thước đo: tiền và thông tin cá nhân. Khi tài khoản của bạn bị - Mức bằng chứng: C - Nguồn: US FCC. Protect Your Smart Device. ；thứ tự các bước là kinh nghiệm của tác giả；làm lại giấy tờ tùy thân xem phần 7, vay với danh tính người khác xem mục về tín dụng ở phần 8 - Ghi chú: Nhớ trước số chăm sóc khách hàng của ba nhà mạng: Mobile 10086, Unicom 10010, Telecom 10000. Số điện thoại của bạn đăng ký ở thành phố nào thì ghi lại luôn, nhân viên tổng đài sẽ hỏi. Gọi chăm sóc khách hàng bằng điện thoại của người khác vẫn báo mất được. - ### 5. Thẻ bị dùng ăn cắp: báo mất và phong tỏa trước, báo công an sau, rồi yêu cầu ngân hàng bồi thường; chứng minh「là chính bạn đã dùng thẻ」là trách nhiệm của ngân hàng - Chi phí: Không tốn tiền. Thấy trên thẻ có động tĩnh bất thường thì báo mất hoặc phong tỏa ngay. Giữ lại biên bản báo công an, biên nhận báo mất, và các thông báo giao dịch ngân hàng gửi tới. Nếu thẻ vẫn còn trong tay bạn thì tới gần nơi đó làm một giao dịch nhỏ như tra cứu số dư hay rút tiền, để lại chứng từ chứng minh lúc xảy ra việc thẻ thật nằm trong tay bạn. Khó là nhịn đừng cãi lại với nhân viên tổng đài trước, cứ báo mất trước đã. @@ -48,7 +44,6 @@ Thước đo: tiền và thông tin cá nhân. Khi tài khoản của bạn bị - Mức bằng chứng: A - Nguồn: Tòa án nhân dân tối cao (最高人民法院) (2021). 关于审理银行卡民事纠纷案件若干问题的规定（第四、五、七、十四、十五条）. - Ghi chú: Có hai trường hợp bạn phải tự gánh chịu. Một là thẻ ngân hàng, mật khẩu, mã xác minh bạn giữ không cẩn thận, bản thân bạn có lỗi (nguyên văn là「未尽妥善保管义务具有过错」), lỗi bao nhiêu thì gánh tới bấy nhiêu. Vì vậy đừng nói mật khẩu cho người khác, đừng chuyển mã xác minh cho người khác (xem mục 1, xác thực hai lớp nên dùng xác nhận bằng cửa sổ trên điện thoại trước). Hai là không báo mất kịp thời, để tổn thất tiếp tục nới rộng, phần phát sinh thêm phải tự gánh. Vì vậy bước đầu tiên luôn là báo mất và phong tỏa, đừng gọi tổng đài mà cãi trước. Các tổ chức thanh toán bên thứ ba cũng áp dụng bộ quy tắc này. Trong tài liệu quảng bá của họ có ghi「先行赔付」, nếu cam kết cụ thể và rõ ràng thì có thể yêu cầu họ bồi thường trước theo đúng điều đó. Tiền bị người ta lừa bạn tự chuyển đi thì phải đi theo cách khác, xem mục 2 ở phần 8 (phát hiện bị lừa thì gọi ngay 110 hoặc 96110 yêu cầu phong tỏa thanh toán). - ### 6. Cứ một thời gian lại xem thiết bị đã đăng nhập và các ứng dụng đã cấp quyền của tài khoản, cái nào không dùng thì dọn đi - Chi phí: Không tốn tiền. Mỗi lần xem mất mấy phút. Khó là không có ai nhắc, phải tự nhớ để xem. @@ -57,7 +52,6 @@ Thước đo: tiền và thông tin cá nhân. Khi tài khoản của bạn bị - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp - Ghi chú: WeChat, Alipay, email, tài khoản Apple và tài khoản Android đều có mục này. Thấy thiết bị không nhận ra thì bấm đăng xuất khỏi toàn bộ, rồi đổi mật khẩu. - ### 7. Đừng bấm「đồng ý tất cả」chỉ để dùng ứng dụng: với thông tin không bắt buộc, bạn không đồng ý thì họ không được phép từ chối dịch vụ - Chi phí: Không tốn tiền. Khó là nhịn không bấm「đồng ý tất cả」. @@ -66,7 +60,6 @@ Thước đo: tiền và thông tin cá nhân. Khi tài khoản của bạn bị - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). 个人信息保护法. 中国人大网. ：第六条「收集个人信息，应当限于实现处理目的的最小范围，不得过度收集个人信息」；第十六条「个人信息处理者不得以个人不同意处理其个人信息或者撤回同意为由，拒绝提供产品或者服务；处理个人信息属于提供产品或者服务所必需的除外」；第十五条「基于个人同意处理个人信息的，个人有权撤回其同意。个人信息处理者应当提供便捷的撤回同意的方式」 - Ghi chú: Tiêu chí để đánh giá là thông tin đó có bắt buộc với dịch vụ này hay không. Bản đồ cần vị trí là bắt buộc, đèn pin xin danh bạ thì không. Cài ứng dụng xong, vào trang quyền ứng dụng trong phần cài đặt điện thoại và tắt những quyền không cần. Đến lúc thật sự dùng thì chọn chỉ cho phép một lần này. - ### 8. Bạn có quyền xem, sao chép, sửa và xóa thông tin cá nhân của mình, bị từ chối thì có thể kiện - Chi phí: Không tốn tiền. Chỉ khi bên kéo dài không làm thì mới cần khiếu nại hoặc kiện. Đi thật sự ra tòa thì tính từ mấy tháng trở lên, phí luật sư còn phải tự bỏ ra, nên khiếu nại trước hơn là hờn. Khó là khi bên kéo dài, phải đi đòi nhiều lần. diff --git "a/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" "b/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" index ff79fa77..cb8a5ff3 100644 --- "a/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" +++ "b/vi/book/15-\347\247\237\346\210\277\344\270\216\344\271\260\346\210\277.md" @@ -12,7 +12,6 @@ Cột đo của phần này là tiền. Chỗ dễ lỗ nhất khi thuê nhà l - Mức bằng chứng: A - Nguồn: Ủy ban Nhà nước (国务院) (2025). Quy định quản lý thuê nhà ở, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十条）. - Ghi chú: Ngày trả nhà, chụp ảnh và quay video cùng chủ nhà. Ghi lại chỉ số đồng hồ nước, đồng hồ điện, đồng hồ gas, tường và sàn cũng chụp lại. Tiền cọc bị giữ mà không có lý do thì trước hết khiếu nại với cơ quan quản lý nhà ở. Số tiền không lớn thì kiện theo thủ tục vụ án nhỏ, xem phần 8 - ### 2. Có người cắt nước cắt điện, đổi ổ khóa, đến tận cửa đe dọa để đuổi bạn đi: báo công an và giữ chứng cứ, quy định cấm đúng những cách này - Chi phí: Không tốn tiền. Một cuộc gọi báo công an, làm được ngay tại chỗ @@ -21,7 +20,6 @@ Cột đo của phần này là tiền. Chỗ dễ lỗ nhất khi thuê nhà l - Mức bằng chứng: A - Nguồn: Ủy ban Nhà nước (国务院) (2025). Quy định quản lý thuê nhà ở, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十二条）. - Ghi chú: Báo công an là để có biên bản xử lý, sau này yêu cầu bồi thường thì đó là chứng cứ. Đồng thời báo cơ quan quản lý nhà ở địa phương - ### 3. Môi giới không được thu hộ và chi hộ tiền thuê cùng tiền cọc, tiền phải vào thẳng tay chủ nhà - Chi phí: Không tốn tiền. Kiểm tra người nhận tiền trước khi trả, mất vài phút. Chỗ khó là giữ vững khi môi giới bảo bạn chuyển tiền vào cho nó @@ -30,7 +28,6 @@ Cột đo của phần này là tiền. Chỗ dễ lỗ nhất khi thuê nhà l - Mức bằng chứng: A - Nguồn: Ủy ban Nhà nước (国务院) (2025). Quy định quản lý thuê nhà ở, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第二十五条）. - Ghi chú: Trước khi trả tiền, hãy kiểm tra người nhận có đúng là tên trên giấy chứng nhận quyền sở hữu không. Nếu không phải chính chủ thì hỏi rõ và xin một văn bản ủy quyền bằng văn bản. Môi giới chỉ được thu phí môi giới cho việc se tiếp hai bên - ### 4. Thuê dài hạn căn hộ có dịch vụ thì xem tài khoản giám sát tiền trước, đừng trả cả năm một lúc để mua rẻ - Chi phí: Không tốn tiền. Xem một lần mất hơn mười phút @@ -39,7 +36,6 @@ Cột đo của phần này là tiền. Chỗ dễ lỗ nhất khi thuê nhà l - Mức bằng chứng: A - Nguồn: Ủy ban Nhà nước (国务院) (2025). Quy định quản lý thuê nhà ở, 住房租赁条例（国令第 812 号，2025 年 9 月 15 日施行）（第十九条）. - Ghi chú: Lúc ký, xem có bị gắn khoản vay hay không. Nếu một ứng dụng trả góp đưa cả năm tiền một lượt cho căn hộ và bạn trả lại theo từng tháng, thì căn hộ bỏ trốn bạn vẫn phải trả tiếp. Loại hợp đồng đó đừng ký - ### 5. Nhà bị bán trong thời gian thuê, hợp đồng thuê vẫn có hiệu lực, không phải dọn - Chi phí: Không tốn tiền @@ -48,7 +44,6 @@ Cột đo của phần này là tiền. Chỗ dễ lỗ nhất khi thuê nhà l - Mức bằng chứng: A - Nguồn: Ủy ban Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự, 民法典（第七百二十五条）. - Ghi chú: Điều kiện là bạn đã dọn vào ở hợp pháp, cách nói trong điều khoản là chiếm hưởng hợp pháp. Hãy giữ hợp đồng, lịch sử chuyển khoản và bằng chứng về thời điểm dọn vào - ### 6. Trước khi ký, kiểm tra giấy chứng nhận quyền sở hữu và tình trạng thế chấp, mọi khoản tiền đều chuyển khoản kèm ghi chú mục đích - Chi phí: Không tốn tiền. Xem một lần mất nửa tiếng @@ -57,7 +52,6 @@ Cột đo của phần này là tiền. Chỗ dễ lỗ nhất khi thuê nhà l - Mức bằng chứng: C - Nguồn: kinh nghiệm của tác giả, không có tài liệu trực tiếp; cùng lập luận về giữ chứng cứ và ghi chú khi chuyển khoản, xem hai mục về tiền cước và giấy nợ trong phần 8 - Ghi chú: Thông tin quyền sở hữu có thể tra tại trung tâm đăng ký bất động sản địa phương. Lúc ký, bảo chủ nhà trực tiếp đến dự, nếu không đến được thì xin một văn bản ủy quyền. Nếu chủ nhà thứ hai cho thuê lại cho bạn thì cần có văn bản đồng ý bằng văn bản của chủ nhà đầu tiên - ### 7. Nhà cũ mà môi giới thu hộ tiền mua nhà thì bắt buộc phải đi qua tài khoản tiền giao dịch riêng do môi giới mở ở ngân hàng - Chi phí: Không tốn tiền. Chỗ khó là kiên trì đòi dùng tài khoản riêng, có khi phải tranh cãi với môi giới mấy câu @@ -66,7 +60,6 @@ Cột đo của phần này là tiền. Chỗ dễ lỗ nhất khi thuê nhà l - Mức bằng chứng: A - Nguồn: Bộ Xây dựng và Đô thị (住房城乡建设部), Ủy ban Cải cách và Phát triển Quốc gia và Bộ Nhân sự và Xã hội (人力资源社会保障部) (2011). Biện pháp quản lý môi giới bất động sản, 房地产经纪管理办法（令第 8 号，2016 年第 29 号令修改）第二十四条：「房地产交易当事人约定由房地产经纪机构代收代付交易资金的，应当通过房地产经纪机构在银行开设的客户交易结算资金专用存款账户划转交易资金。」 第十八条「房地产经纪服务实行明码标价制度……在经营场所醒目位置标明房地产经纪服务项目、服务内容、收费标准」；第十九条「两家或者两家以上房地产经纪机构合作开展同一宗房地产经纪业务的，只能按照一宗业务收取佣金，不得向委托人增加收费」；第十七条 về thủ tục vay, đăng ký bất động sản và các dịch vụ khác「应当向委托人说明服务内容、收费标准等情况，经委托人同意后，另行签订合同」. - Ghi chú: Tiền mua nhà là vài trăm nghìn tệ đến vài triệu tệ, là khoản đơn lẻ lớn nhất trong cuốn sách này, đừng chuyển bằng WeChat cho môi giới viên. Còn hai kiểu thu thêm tiền thường gặp. Một là hai môi giới cùng làm một hợp đồng, kiểu này chỉ được thu một khoản hoa hồng. Hai là thủ tục vay và thủ tục sang tên, đó là dịch vụ khác, phải ký hợp đồng riêng và nói trước chi phí. Mọi khoản tiền đều chuyển khoản kèm ghi chú mục đích, xem mục 6 của phần này (trước khi ký, kiểm tra giấy chứng nhận quyền sở hữu và tình trạng thế chấp). - ### 8. Đừng thuê nhà bị chia ngăn: đơn vị cho thuê nhỏ nhất là căn phòng thiết kế ban đầu, bếp, nhà vệ sinh, ban công không được ở - Chi phí: Không tốn tiền. Cái giá phải trả là với cùng ngân sách thì hoặc ở xa hơn một chút, hoặc ở chật hơn một chút diff --git "a/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" index 6ca98bbf..ea7da846 100644 --- "a/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" +++ "b/vi/book/16-\345\276\227\344\272\206\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -12,7 +12,6 @@ Cột đo của phần này là tỷ lệ tử vong chung và tiền. Sau khi ph - Mức bằng chứng: A - Nguồn: Simpson SH, Eurich DT, Majumdar SR, et al. (2006). A meta-analysis of the association between adherence to drug therapy and mortality. BMJ, 333(7557), 15. - Ghi chú:Tranh cãi. Đây cũng là con số cần tính giảm nhất trong cuốn sách. Trong cùng phân tích tổng hợp đó, cả người uống đủ cả thuốc giả (placebo) cũng có tỷ lệ tử vong thấp (0.56, 0.43–0.74). Điều đó cho thấy phần lớn lợi ích đến từ 「người kiên trì uống thuốc vốn đã chú ý sức khoẻ hơn」. Tác giả gọi yếu tố bị trộn vào đó là hiệu ứng người tuân thủ khỏe mạnh. Nhưng cái hại của bản thân việc ngừng thuốc thì rõ ràng, đặc biệt với thuốc hạ huyết áp, thuốc chống đông máu và thuốc chống thải ghép - ### 2. Làm giấy công nhận bệnh mạn tính ngoại trú trước, rồi đăng ký khám chữa tại nơi khác, để huyết áp cao, tiểu đường, xạ trị hoá chất, lọc máu và chống thải ghép được quyết toán ngay tại nơi khác - Chi phí: Không tốn tiền. Trước khi đi, làm một lần đăng ký khám chữa bệnh tại nơi khác trên ứng dụng bảo hiểm y tế, làm xong ngay trên điện thoại. @@ -21,7 +20,6 @@ Cột đo của phần này là tỷ lệ tử vong chung và tiền. Sau khi ph - Mức bằng chứng: A - Nguồn: Văn phòng Cục Quản lý bảo hiểm y tế quốc gia và Văn phòng Bộ Tài chính (国家医保局办公室、财政部办公厅) (2024). Thông báo về việc mở rộng có trật tự và an toàn danh mục bệnh mạn tính ngoại trú được quyết toán trực tiếp liên tỉnh, 关于稳妥有序扩大跨省直接结算门诊慢特病病种范围的通知（医保办发〔2024〕19 号）. - Ghi chú: Điều kiện là làm đủ cả hai việc: làm giấy công nhận bệnh mạn tính ngoại trú ở nơi tham gia bảo hiểm trước, rồi đăng ký khám chữa tại nơi khác. Năm loại bệnh bổ sung đã mở đến đâu thì mỗi nơi tiến độ khác nhau, cứ theo nơi bảo hiểm y tế địa phương công bố - ### 3. Tái khám theo khoảng cách bác sĩ dặn, ghi chỉ số mỗi lần vào cùng một quyển sổ - Chi phí: Mỗi lần xét nghiệm vài chục đến vài trăm tệ. Đi bệnh viện một chuyến cộng thời gian chờ kết quả thì mất một ít thời gian. Chỗ khó là cứ sau một khoảng lại phải đi một lần, dễ hoãn lần này sang lần khác. @@ -30,7 +28,6 @@ Cột đo của phần này là tỷ lệ tử vong chung và tiền. Sau khi ph - Mức bằng chứng: C - Nguồn: kinh nghiệm của tác giả, không có tài liệu trực tiếp - Ghi chú: Mỗi lần ghi tối thiểu bốn thứ: ngày, giá trị chỉ số, thuốc đang uống, liều thuốc. Chụp ảnh phiếu xét nghiệm bằng điện thoại cũng được, nhưng phải để cùng một chỗ, đừng rải mỗi thứ một nơi - ### 4. Đừng bỏ điều trị chính thức chỉ để thử phương thuốc dân gian hay thực phẩm bảo vệ sức khoẻ - Chi phí: Không tốn tiền, còn tiết kiệm tiền mua thực phẩm bảo vệ sức khoẻ. Không mất thời gian. Chỗ khó là giữ vững khi người khác liên tục khuyên bạn thử. @@ -39,15 +36,14 @@ Cột đo của phần này là tỷ lệ tử vong chung và tiền. Sau khi ph - Mức bằng chứng: C - Nguồn: kinh nghiệm của tác giả, không có tài liệu trực tiếp; bằng chứng liên quan xem phần 6 - Ghi chú: Muốn thử thực phẩm bổ sung thì nói với bác sĩ điều trị trước. Có thực phẩm bổ sung tương tác với thuốc kê đơn, chẳng hạn làm thay đổi nồng độ thuốc chống đông máu và thuốc chống thải ghép trong máu - -### 5. Bệnh mạn tính đã ổn định thì ở bệnh viện cộng đồng một lần lấy được thuốc dùng 12 tuần +### 5. Bệnh mạn tính đã ổn định, ở bệnh viện cộng đồng xin được thuốc một lần trong 12 tuần -- Chi phí: Không tốn tiền. Đến trạm y tế cộng đồng hỏi một câu có kê được đơn dài hạn không, chỉ mất một câu. -- Nói thẳng: Người bệnh đã ổn định, thuốc cũng không đổi, ở trạm y tế cộng đồng một lần lấy đủ thuốc dùng 12 tuần. Bỏ được nửa ngày mỗi tháng đi bệnh viện lớn xếp hàng khám. Trong cùng nơi tham gia bảo hiểm, tỷ lệ hoàn tiền khi khám ở bệnh viện tuyến cơ sở thường cao hơn bệnh viện tuyến trên khoảng mười điểm phần trăm. -- Lợi ích: Không phải mỗi tháng đi bệnh viện lớn xếp hàng lấy thuốc. Trong cùng nơi tham gia bảo hiểm (còn gọi là khu vực điều hòa), tỷ lệ hoàn tiền khi nằm viện ở bệnh viện tuyến cơ sở về nguyên tắc cao hơn tuyến trên khoảng 10 điểm phần trăm +- Chi phí: Không tốn tiền. Tới trạm y tế cộng đồng hỏi một câu xem có kê được đơn thuốc dài hạn không, chỉ mất một câu. +- Nói thẳng: Người bệnh ổn định, thuốc đang uống cũng không đổi, thì ở trạm y tế cộng đồng xin đủ thuốc cho 12 tuần trong một lần. Bỏ được nửa ngày mỗi tháng phải chạy tới bệnh viện lớn xếp hàng lấy số khám. Ở cùng một khu vực tham gia bảo hiểm, tỷ lệ bồi thường của người nằm viện ở bệnh viện tuyến cơ sở thường còn cao hơn bệnh viện tuyến trên khoảng mười điểm phần trăm. +- Lợi ích: Không phải mỗi tháng chạy tới bệnh viện lớn xếp hàng lấy thuốc. Trong cùng một khu vực tham gia bảo hiểm (gọi là khu vực quy hoạch chung), tỷ lệ bồi thường của người nằm viện ở bệnh viện tuyến cơ sở về nguyên tắc cao hơn bệnh viện tuyến trên khoảng 10 điểm phần trăm - Mức bằng chứng: A -- Nguồn: Văn phòng Ủy ban Nhà nước (国务院办公厅) (2026). Một số biện pháp thúc đẩy xây dựng hệ thống chẩn đoán điều trị phân tầng. ：「对于符合条件的慢性病患者，基层医疗卫生机构单次可开具不超过12周用药的长期处方。」 — với người bị bệnh mạn tính đủ điều kiện, cơ sở y tế tuyến cơ sở có thể kê một lần đơn dài hạn tối đa 12 tuần thuốc.；cùng văn bản 「原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距」, và yêu cầu bệnh viện tuyến trên mở phòng khám bệnh mạn tính thường gặp như huyết áp cao, tiểu đường, bệnh phổi tắc nghẽn mạn tính tại tuyến cơ sở -- Ghi chú: 「符合条件」 do bác sĩ đánh giá, thường là người đã chẩn đoán rõ và phác đồ dùng thuốc cũng ổn định. Thuốc vẫn là thuốc đó, cái tiết kiệm là nửa ngày chạy bệnh viện mỗi tháng và phí khám. Ở tuyến cơ sở không có loại thuốc bạn cần, có thể hỏi xem có đăng ký báo thiếu thuốc và vận chuyển không. +- Nguồn: Văn phòng Ủy ban Nhà nước (国务院办公厅) (2026). Một số biện pháp thúc đẩy xây dựng hệ thống chẩn đoán điều trị phân tầng, 关于加快建设分级诊疗体系的若干措施. ：「对于符合条件的慢性病患者，基层医疗卫生机构单次可开具不超过12周用药的长期处方。」— với bệnh nhân mạn tính đủ điều kiện, cơ sở y tế tuyến cơ sở được kê một lần đơn thuốc dài hạn tối đa 12 tuần dùng thuốc；同文「原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距」— cùng văn bản quy định trong khu vực quy hoạch chung, tỷ lệ bồi thường tiền nằm viện của các cơ quan y tế kéo giãn theo từng bậc khoảng 10 điểm phần trăm, đồng thời yêu cầu bệnh viện tuyến trên mở khám ngoại trú cho các bệnh mạn tính thường gặp như huyết áp cao, tiểu đường, bệnh phổi tắc nghẽn mạn tính tại tuyến cơ sở；同文「三级医院要聚焦急危重症和疑难复杂疾病，加强转诊会诊和住院服务，逐步酌减常见病复诊和诊断明确、病情稳定的慢性病等普通门诊」— cùng văn bản quy định bệnh viện hạng ba tập trung vào cấp cứu nguy kịch và bệnh nặng, bệnh khó chẩn đoán phức tạp, tăng cường hội chẩn chuyển tuyến và dịch vụ nằm viện, dần giảm bớt khám ngoại trú thường cho bệnh thường gặp tái khám và cho bệnh mạn tính đã chẩn đoán rõ, bệnh đã ổn định；「并将专家团队普通门诊向基层医疗卫生机构延伸」— và đưa khám ngoại trú thường của nhóm chuyên gia xuống mở tại các cơ sở y tế tuyến cơ sở +- Ghi chú: Ai đủ điều kiện thì do bác sĩ quyết, thường là người đã chẩn đoán rõ và phương thuốc đã ổn định. Thuốc vẫn là thuốc cũ, cái tiết kiệm là nửa ngày mỗi tháng chạy đi lấy và tiền lệ khám. Ở tuyến cơ sở không có loại thuốc bạn cần thì hỏi xem có đăng ký thiếu thuốc và cấp phát không. Cùng văn bản đó còn yêu cầu bệnh viện hạng ba dần bỏ khám ngoại trú thường cho loại bệnh mạn tính này, khám ngoại trú thường của cùng nhóm chuyên gia chuyển xuống mở ở cộng đồng. Còn coi bệnh viện lớn là chỗ lấy thuốc thì sau này sẽ càng khó lấy số khám. ### 6. Trước khi đăng ký bác sĩ gia đình ở cộng đồng, hỏi rõ cái nào vào bảo hiểm y tế, cái nào tự trả @@ -57,7 +53,6 @@ Cột đo của phần này là tỷ lệ tử vong chung và tiền. Sau khi ph - Mức bằng chứng: A - Nguồn: Văn phòng Ủy ban Nhà nước (国务院办公厅) (2026). Một số biện pháp thúc đẩy xây dựng hệ thống chẩn đoán điều trị phân tầng. ：「加强基层门诊付费与签约服务政策联动，基本服务包按规定纳入医保支付；个性化服务包由签约基层医疗卫生机构按程序向县级卫生健康部门备案，费用由个人支付。」 — tăng cường liên kết giữa thanh toán khám ngoại trú tuyến cơ sở và chính sách dịch vụ theo hợp đồng đăng ký, gói dịch vụ cơ bản được đưa vào bảo hiểm y tế trả theo quy định; gói dịch vụ cá nhân hoá do cơ sở y tế tuyến cơ sở đã ký đăng ký với cơ quan y tế và sức khoẻ cấp huyện theo thủ tục, chi phí do cá nhân trả. - Ghi chú: Giá trị của việc đăng ký nằm ở chỗ có người theo dõi bệnh của bạn lâu dài, không nằm ở chỗ gói có bao nhiêu mục. Khi có người chào bán 「gói nâng cấp」, hãy hỏi ba câu: đây là gói cơ bản hay gói cá nhân hoá, đã đăng ký chưa, một năm bao nhiêu tiền. - ### 7. Xác định tiểu đường thì đi khám đáy mắt một lần, sau đó khám theo khoảng cách bác sĩ dặn; mỗi năm khám chân một lần - Chi phí: Một lần khám đáy mắt có giãn đồng tử vài chục đến vài trăm tệ. Khám chân thường đã nằm trong tiền khám, không phải trả thêm. Chỗ khó là lúc mắt và chân chưa khó chịu gì vẫn phải nhớ đi khám. @@ -66,7 +61,6 @@ Cột đo của phần này là tỷ lệ tử vong chung và tiền. Sau khi ph - Mức bằng chứng: A - Nguồn: Hiệp hội Đái thái Hoa Kỳ (美国糖尿病学会) (2026). Tiêu chuẩn chẩn đoán và điều trị đái thái 2026·chương 12 về bệnh lõ mạch võng mạc, bệnh thần kinh và chăm sóc bàn chân. Diabetes Care：「People with type 2 diabetes should have an initial dilated and comprehensive eye examination by an ophthalmologist or optometrist at the time of the diabetes diagnosis.」「Adults with type 1 diabetes should have an initial dilated and comprehensive eye examination by an ophthalmologist or optometrist 5 years after the onset of diabetes.」「If there is no evidence of retinopathy from one or more annual eye exams and glycemic indicators are within the goal range, then screening every 1–2 years may be considered.」「Perform a comprehensive foot evaluation at least annually to identify risk factors for ulcers and amputations.」 - Ghi chú: Người typ 2 phải khám ngay lúc xác định chẩn đoán, người typ 1 bắt đầu khám 5 năm sau khi bệnh khởi phát. Khác biệt là typ 2 thường đã tăng âm thầm mấy năm. Người đã khám một hai lần không thấy vấn đề và đường huyết cũng đạt chuẩn thì có thể giãn ra 1 đến 2 năm một lần. Người cảm giác ở chân bị tê dần thì mỗi ngày tự nhìn lòng bàn chân, soi bằng gương cũng được. - ### 8. Đã từng bị sỏi thận thì uống nước mỗi ngày 2,5–3 lít, hạ muối xuống dưới 6 g - Chi phí: Không tốn tiền. Biến việc uống nước bất cứ lúc nào thành thói quen. Ngày ra nhiều mồ hôi thì còn phải uống nhiều hơn, chỗ khó là ngày nào cũng nhớ. @@ -76,11 +70,11 @@ Cột đo của phần này là tỷ lệ tử vong chung và tiền. Sau khi ph - Nguồn: Borghi L, Meschi T, Amato F, Briganti A, Novarini A, Giannini A (1996). Urinary volume, water and recurrences in idiopathic calcium nephrolithiasis: a 5-year randomized prospective study. The Journal of Urology, 155(3), 839-843. ；Viện Quốc gia về Chất lượng và Chăm sóc Sức khoẻ (National Institute for Health and Care Excellence) (2019). Renal and ureteric stones: assessment and management. NICE guideline NG118, điều 1.8.1. - Ghi chú: Tái phát giảm khoảng 55%, lại không tốn tiền, theo cách xếp hạng của cuốn sách thì đây là mục hiệu quả chi phí cao nhất của phần này. Cái cần theo dõi là lượng nước tiết trong 24 giờ, mục tiêu là trên 2 đến 2,5 lít, không phải nhìn vào lượng đã uống vào. Người ra nhiều mồ hôi, người làm việc trong môi trường nóng thì phải uống nhiều hơn. Thử nghiệm đó làm trên người bị sỏi canxi nguyên ốc lần phát đầu tiên. Thành phần sỏi không chỉ một loại, sỏi axit uric, sỏi nhiễm trùng và sỏi xystin mỗi loại có cách xử lý riêng. Người đã từng tái phát thì nên làm một lần phân tích thành phần sỏi và xét nghiệm nước tiểu 24 giờ, rồi mới quyết định có thêm thuốc hay không. Đừng kiêng hẳn canxi chỉ để phòng sỏi, xem mục 21 ở phần 6. Khi đau quặn lưng bụng dữ dội thì phải đến bệnh viện, uống nhiều nước không đẩy được. Máu tiểu không đau thì còn có thứ khác cần xét, xem mục 27 ở phần 1 -### 9. Xác định bệnh gout thì uống thuốc hạ acid uric dài hạn, hạ acid uric máu xuống dưới 360 µmol/L rồi giữ nguyên ở đó +### 9. Chẩn đoán gout thì uống thuốc hạ acid uric lâu dài, ép acid uric máu xuống dưới 360 µmol/L và giữ nguyên thế -- Chi phí: Allopurinol vài tệ đến vài chục tệ mỗi tháng. Lúc mới đầu mỗi vài tuần xét nghiệm acid uric máu để chỉnh liều, ổn định rồi thì mỗi vài tháng xét nghiệm một lần. 3 đến 6 tháng đầu còn phải uống kèm một thuốc phòng ngừa cơn. Chỗ khó là lúc không đau vẫn phải uống liên tục. -- Nói thẳng: Gout cần uống thuốc hạ acid uric dài hạn, hạ acid uric xuống dưới 360 µmol/L, không thể chỉ uống thuốc giảm đau lúc đau. Người bệnh được y tá chỉnh liều theo kết quả xét nghiệm thì sau hai năm có 95% đạt chuẩn, còn người khám theo kiểu cũ chỉ 30%. Đạt chuẩn rồi vẫn không được ngừng thuốc. Nhịn ăn không hạ được xuống mức đó, bỏ rượu cũng chỉ giảm 1.6 mg/dL. -- Lợi ích: Một thử nghiệm đối chứng chia ngẫu nhiên thành hai nhóm, gồm 517 người lớn từng lên cơn trong 12 tháng trở lại. Một nhóm do y tá phụ trách, giải thích rõ cho bệnh nhân bệnh là gì, rồi chỉnh acid uric theo chỉ số cho tới khi đạt chuẩn. Nhóm kia khám như cũ với bác sĩ đa khoa. Sau 2 năm, tỷ lệ acid uric máu hạ xuống dưới <360 µmol/L (6 mg/dL) là 95% so với 30%. Khả năng đạt chuẩn gấp 3.18 lần nhóm thường (RR 3.18, 95% CI 2.42–4.18, đây là khoảng tin cậy của con số), P<0.0001. Ở năm thứ 2, tần suất lên cơn, u tophi và chất lượng sống, vài chỉ số phụ này cũng tốt hơn nhóm thường. Mỗi năm tuổi thọ tăng thêm, tính cả phần quy đổi theo chất lượng sống (số năm sống điều chỉnh theo chất lượng sống), tốn £5,066. Cùng thử nghiệm đó còn làm nghiên cứu theo dõi, phát bảng hỏi cho 438 người, thu về 82%. Cả hai nhóm đều tự báo cáo số lần lên cơn trong 12 tháng qua. Nhóm đạt chuẩn có trung vị 0 lần (nửa số người ở giữa rơi vào 0–0 lần). Nhóm thường là 1 lần (0–3), P<0.001. Khoảng trong ngoặc này gọi là khoảng tứ phân vị. Tỷ lệ vẫn còn uống thuốc hạ acid uric, nhóm đạt chuẩn gấp 1.19 lần nhóm thường (RR hiệu chỉnh 1.19, 1.09–1.30). Hướng dẫn gout 2020 của Hiệp hội Thấp khớp Hoa Kỳ mạnh mẽ khuyến nghị bắt đầu điều trị hạ acid uric cho ba nhóm người: người có u tophi dưới da, người chụp chiếu thấy gout làm xói mòn xương, người lên cơn ≥2 lần mỗi năm. Cách làm là theo dõi liên tục acid uric máu, chỉnh liều theo kết quả cho tới khi đạt chuẩn, mục tiêu <6 mg/dL. Thuốc lựa chọn đầu tiên là allopurinol, người bị bệnh thận mạn tính từ giai đoạn 3 trở lên cũng dùng nó trước, liều khởi đầu ≤100 mg/ngày. Lúc bắt đầu cho kèm thuốc chống viêm để phòng ngừa, uống ít nhất 3 đến 6 tháng. Cùng hướng dẫn đó còn dẫn một nhóm bệnh nhân được theo dõi. Sau khi kiểm soát tốt trong thời gian dài, những người này ngừng thuốc, và sau khi ngừng acid uric máu vẫn giữ <7 mg/dL. Trong số họ, người không lên cơn lần nào trong suốt 5 năm theo dõi chỉ chiếm 13% (27/211). Về ăn uống, người hạn chế rượu hoặc bỏ rượu có acid uric máu thấp hơn người không hạn chế 1.6 mg/dL, một phần bia làm acid uric máu tăng 0.16 mg/dL. Ăn uống lành mạnh, ăn theo kiểu Địa Trung Hải và ăn theo DASH thì hiệu quả càng nhỏ hơn. Có khác biệt về việc chọn thuốc hạ acid uric nào là thuốc đầu tiên. Hướng dẫn của Hiệp hội Thấp khớp Hoa Kỳ mạnh mẽ khuyến nghị allopurinol là thuốc đầu tiên, lý do là khi dùng đủ liều thì có tác dụng, dung nạp tốt, an toàn và rẻ. Hướng dẫn còn khuyến nghị có điều kiện: người đang uống febuxostat mà có tiền sử tim mạch, hoặc vừa có biến cố tim mạch, thì nếu có thuốc khác để thay thế nên thay. Khuyến nghị này khớp với cảnh báo hộp đen mà FDA gắn cho febuxostat. Một căn cứ là thử nghiệm CARES, 6190 người bị gout kèm bệnh tim mạch được chia ngẫu nhiên uống febuxostat hoặc allopurinol. Rủi ro tử vong toàn nguyên ở nhóm febuxostat gấp 1.22 lần nhóm allopurinol (HR 1.22, 95% CI 1.01–1.47, cao hơn khoảng 22%). Tử vong do tim mạch gấp 1.34 lần (1.03–1.73, cao hơn khoảng 34%). Thử nghiệm FAST ở châu Âu cho kết luận khác. Thử nghiệm này gồm 6128 người bệnh trên 60 tuổi có yếu tố nguy cơ tim mạch, sự kiện tim mạch chính ở febuxostat không nhiều hơn allopurinol (HR 0.85, 0.70–1.03). Tỷ lệ tử vong là 7.2% ở nhóm febuxostat và 8.6% ở nhóm allopurinol. Người uống thuốc làm tăng đào thải acid uric qua thận, Hiệp hội Thấp khớp Hoa Kỳ không khuyến nghị kiềm hóa nước tiểu, vì không có bằng chứng cho là có hiệu quả và độ chắc chắn của bằng chứng rất thấp. Hướng dẫn đồng thời yêu cầu dặn những người này uống nhiều nước, và người đã biết có sỏi thận hoặc bệnh thận mạn tính từ giai đoạn 3 trở lên thì không được dùng loại thuốc này +- Chi phí: Allopurinol mỗi tháng vài tệ đến mấy chục tệ. Lúc đầu cứ vài tuần xét nghiệm acid uric máu một lần để chỉnh liều, ổn định rồi thì vài tháng một lần. Trong 3 đến 6 tháng đầu còn phải uống kèm một thuốc phòng cơn lên. Chỗ khó là không đau vẫn phải uống liên tục. +- Nói thẳng: Gout đòi hỏi uống thuốc hạ acid uric lâu dài, ép acid uric xuống dưới 360 µmol/L, không thể chỉ uống thuốc giảm đau lúc đang đau. Bệnh nhân có y tá giúp chỉnh liều theo kết quả xét nghiệm thì sau hai năm 95% đạt chuẩn, còn người khám như bình thường chỉ 30%. Đạt chuẩn rồi vẫn không ngừng thuốc được. Dựa vào kiêng khem ăn không hạ được xuống mức đó, bỏ rượu cũng chỉ hạ 1.6 mg/dL. +- Lợi ích: Một thử nghiệm đối chứng chia ngẫu nhiên thành hai nhóm, 517 người lớn từng lên cơn gout trong 12 tháng trở lại. Một nhóm do y tá phụ trách, người bệnh được giải thích rõ bệnh này là gì, rồi chỉnh acid uric theo chỉ số cho tới khi đạt chuẩn. Nhóm kia khám bác sĩ đa khoa như bình thường. Ở thời điểm 2 năm, tỷ lệ acid uric máu xuống dưới <360 µmol/L (6 mg/dL) là 95% so với 30%. Khả năng đạt chuẩn gấp 3.18 lần nhóm thông thường (RR 3.18, 95% CI 2.42–4.18, khoảng này là độ tin cậy của con số), P<0.0001. Ở năm thứ hai, tần suất lên cơn, u tophi và chất lượng sống, các chỉ số phụ này cũng tốt hơn nhóm thông thường. Mỗi năm sống thêm đổi lại, tính cả phần quy về chất lượng sống (năm sống điều chỉnh chất lượng), tốn £5.066. Cùng thử nghiệm đó còn làm nghiên cứu theo dõi, 438 người nhận bảng hỏi, thu hồi được 82%. Cả hai nhóm tự báo số lần lên cơn trong 12 th qua. Nhóm đạt chuẩn có trung vị 0 lần (nửa giữa rơi vào 0–0 lần). Nhóm thông thường là 1 lần (0–3), P<0.001. Khoảng trong ngoặc gọi là khoảng tứ phân vị. Tỷ lệ vẫn đang uống thuốc hạ acid uric ở nhóm đạt chuẩn gấp 1.19 lần nhóm thông thường (RR hiệu chỉnh 1.19, 1.09–1.30). Hướng dẫn gout năm 2020 của Hiệp hội Thấp khớp Hoa Kỳ khuyến nghị mạnh việc bắt đầu hạ acid uric cho ba nhóm người: người có u tophi dưới da, người mà hình ảnh cho thấy gout đã ăn mòn xương thành khuyết, và người lên cơn từ 2 lần trở lên mỗi năm. Cách làm là theo dõi liên tục acid uric máu, chỉnh liều theo kết quả cho tới khi đạt chuẩn, mục tiêu <6 mg/dL. Thuốc chọn đầu tiên là allopurinol; người bệnh thận mạn tính giai đoạn 3 trở lên cũng chọn nó, liều khởi đầu ≤100 mg/ngày. Lúc bắt đầu thì cho kèm thuốc chống viêm để phòng, uống ít nhất 3 đến 6 tháng. Cùng hướng dẫn đó còn dẫn một nhóm quan sát ca lâm sàng. Những bệnh nhân này ngừng thuốc sau khi đã kiểm soát tốt dài ngày, thì acid uric máu vẫn giữ dưới <7 mg/dL. Trong số họ, suốt 5 năm theo dõi mà không lên cơn lần nào chỉ có 13% (27/211). Về ăn uống, người hạn chế rượu hoặc bỏ rượu có acid uric máu thấp hơn người không hạn chế 1.6 mg/dL, một phần bia làm acid uric máu tăng 0.16 mg/dL. Chế độ ăn khoẻ, chế độ ăn Địa Trung Hải và chế độ ăn DASH tác dụng còn nhỏ hơn nữa. Có bất đồng về việc nên chọn thuốc hạ acid uric nào làm lựa chọn đầu tiên. Hướng dẫn của Hiệp hội Thấp khớp Hoa Kỳ khuyến nghị mạnh allopurinol làm thuốc đầu tiên, lý do là khi dùng đủ liều thì có tác dụng, dung nạp tốt, an toàn và rẻ. Hướng dẫn còn khuyến nghị có điều kiện: người đang uống febuxostat mà có tiền sử tim mạch, hoặc vừa có biến cố tim mạch, thì nếu có thuốc khác để đổi thì nên đổi. Khuyến nghị này khớp với cảnh báo khung đen mà FDA gắn cho febuxostat. Một căn cứ là thử nghiệm CARES, 6190 bệnh nhân gout có bệnh tim mạch đi kèm được chia ngẫu nhiên uống febuxostat hoặc allopurinol. Nhóm febuxostat có nguy cơ tử vong do mọi nguyên nhân bằng 1.22 lần nhóm allopurinol (HR 1.22, 95% CI 1.01–1.47, cao khoảng 22%). Tử vong tim mạch là 1.34 lần (1.03–1.73, cao khoảng 34%). Thử nghiệm FAST của châu Âu lại cho kết luận khác. Nó đưa vào 6128 bệnh nhân trên 60 tuổi có yếu tố nguy cơ tim mạch, biến cố tim mạch chính ở nhóm febuxostat không nhiều hơn allopurinol (HR 0.85, 0.70–1.03). Tỷ lệ tử vong là 7.2% ở nhóm febuxostat và 8.6% ở nhóm allopurinol. Người uống thuốc tăng thải acid uric qua thận, Hiệp hội Thấp khớp Hoa Kỳ không khuyến nghị có điều kiện việc kiềm hoá nước tiểu, vì không có bằng chứng cho thấy có tác dụng và độ chắc của bằng chứng rất thấp. Hướng dẫn đồng thời yêu cầu dặn những người đó uống nhiều nước, còn người đã có sỏi thận hoặc bệnh thận mạn tính giai đoạn 3 trở lên thì không nên dùng loại thuốc này. - Mức bằng chứng: A -- Nguồn: Doherty M, Jenkins W, Richardson H, et al. (2018). Efficacy and cost-effectiveness of nurse-led care involving education and engagement of patients and a treat-to-target urate-lowering strategy versus usual care for gout: a randomised controlled trial. Lancet, 392(10156), 1403-1412. ；Abhishek A, Jenkins W, La-Crette J, Fernandes G, Doherty M (2020). Nurse-led care is preferred over GP-led care of gout and improves gout outcomes: results of Nottingham Gout Treatment Trial follow-up study. Rheumatology, 59(3), 575-579. ；FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760. ；White WB, Saag KG, Becker MA, et al.; CARES Investigators (2018). Cardiovascular Safety of Febuxostat or Allopurinol in Patients with Gout. New England Journal of Medicine, 378(13), 1200-1210. ；Mackenzie IS, Ford I, Nuki G, et al.; FAST Study Group (2020). Long-term cardiovascular safety of febuxostat compared with allopurinol in patients with gout (FAST): a multicentre, prospective, randomised, open-label, non-inferiority trial. Lancet, 396(10264), 1745-1757. ；Chi nhánh Nội tiết của Hiệp hội Y khoa Trung Quốc (中华医学会内分泌学分会) (2020). Hướng dẫn chẩn đoán và điều trị tăng axit uric máu và gout tại Trung Quốc (2019). 中华内分泌代谢杂志, 36(1), 1-13. -- Ghi chú:Tranh cãi: nên dùng thuốc hạ acid uric nào thì hướng dẫn của các nước khác nhau. Hướng dẫn Hoa Kỳ chọn allopurinol làm thuốc đầu tiên, người có bệnh tim mạch thì đổi được thì đừng dùng febuxostat, nhưng thử nghiệm FAST ở châu Âu không thấy febuxostat nguy hiểm hơn. Ở trong nước vẫn mua được benzbromarone, đó là thuốc làm tăng đào thải acid uric. Hướng dẫn của Hiệp hội Y khoa Trung Quốc (bản 2019 và bản cập nhật 2024) có khuyến nghị riêng về chọn thuốc và về kiềm hóa nước tiểu. Khi uống benzbromarone có cần thêm natri hydrocarbonat để kiềm hóa nước tiểu không thì hướng dẫn Hoa Kỳ cho là không có bằng chứng, cứ theo ý bác sĩ của bạn, uống nhiều nước thì không được bỏ. **Người Hán trước khi dùng allopurinol nên xét HLA-B\*5801**. Đây là một mã gen bẩm sinh phụ trách phản ứng miễn dịch, xét một lần thì kết quả không đổi suốt đời. Người mang gen này chiếm 7.4% trong người Hán, người Hàn Quốc và người Thái, còn người da trắng và người gốc Tây Ban Nha chỉ 0.7%. Người châu Á và người không thuộc chủng tộc Phi sau khi dùng allopurinol có nguy cơ bị hội chứng quá mẫn với allopurinol gấp 3 lần người da trắng, bệnh này nặng thì da bong tróc diện rộng, có thể đe dọa tính mạng. Hướng dẫn gout 2020 của Hiệp hội Thấp khớp Hoa Kỳ khuyến nghị với hai nhóm này là trước khi dùng thuốc có thể cân nhắc xét trước. **Mới bắt đầu hạ acid uric lại dễ kích thích cơn hơn**, nên hướng dẫn yêu cầu rõ ràng phải uống kèm một thuốc chống viêm phòng ngừa trong 3 đến 6 tháng đầu. 「Uống thuốc hạ acid uric xong đau hơn nên bỏ thuốc」 là cách thất bại phổ biến nhất của việc này. Đau lên thì phải làm là tăng thuốc chống viêm theo đơn bác sĩ, không phải ngừng thuốc hạ acid uric. Ngưỡng đạt chuẩn mỗi nơi viết khác nhau: <6 mg/dL quy đổi khoảng 360 µmol/L, phiếu xét nghiệm ở Trung Quốc thường dùng µmol/L. Thăm khám thấy acid uric cao mà chưa từng lên cơn là chuyện khác, xem mục 19 ở phần 6. Về đồ uống có đường và rượu, xem mục 7 và 20 ở phần 2 (không uống đồ uống có đường, uống ít rượu hoặc không uống) +- Nguồn: Doherty M, Jenkins W, Richardson H, et al. (2018). Efficacy and cost-effectiveness of nurse-led care involving education and engagement of patients and a treat-to-target urate-lowering strategy versus usual care for gout: a randomised controlled trial. Lancet, 392(10156), 1403-1412. ；Abhishek A, Jenkins W, La-Crette J, Fernandes G, Doherty M (2020). Nurse-led care is preferred over GP-led care of gout and improves gout outcomes: results of Nottingham Gout Treatment Trial follow-up study. Rheumatology, 59(3), 575-579. ；FitzGerald JD, Dalbeth N, Mikuls T, et al. (2020). 2020 American College of Rheumatology Guideline for the Management of Gout. Arthritis Care & Research, 72(6), 744-760. ；White WB, Saag KG, Becker MA, et al.; CARES Investigators (2018). Cardiovascular Safety of Febuxostat or Allopurinol in Patients with Gout. New England Journal of Medicine, 378(13), 1200-1210. ；Mackenzie IS, Ford I, Nuki G, et al.; FAST Study Group (2020). Long-term cardiovascular safety of febuxostat compared with allopurinol in patients with gout (FAST): a multicentre, prospective, randomised, open-label, non-inferiority trial. Lancet, 396(10264), 1745-1757. ；Chi nhánh Nội tiết của Hiệp hội Y khoa Trung Quốc (中华医学会内分泌学分会) (2020). Hướng dẫn chẩn đoán và điều trị tăng acid uric máu và gout của Trung Quốc (2019), 中国高尿酸血症与痛风诊疗指南(2019). 中华内分泌代谢杂志, 36(1), 1-13. +- Ghi chú: Tranh cãi. Dùng thuốc nào để hạ acid uric thì hướng dẫn các nước không giống nhau. Hướng dẫn Hoa Kỳ chọn allopurinol làm thuốc đầu tiên, người có bệnh tim mạch mà đổi được thì đừng uống febuxostat, nhưng thử nghiệm FAST của châu Âu không thấy febuxostat nguy hiểm hơn. Ở trong nước vẫn mua được benzbromarone, đó là thuốc tăng thải acid uric. Hướng dẫn của Hiệp hội Y khoa Trung Quốc (bản 2019 và bản cập nhật 2024) có khuyến nghị riêng cho việc chọn thuốc và việc kiềm hoá nước tiểu. Uống benzbromarone có cần thêm natri bicarbonate để kiềm hoá nước tiểu không, hướng dẫn Hoa Kỳ cho rằng không có bằng chứng, cứ theo ý bác sĩ của bạn, phần uống nhiều nước thì không được bỏ. **Người Hoa kỳ nên xét HLA-B\*5801 trước khi dùng allopurinol**. Đó là một kiểu gen vốn có sẵn, phụ trách phản ứng miễn dịch, xét một lần thì kết quả không đổi suốt đời. Người mang gen này chiếm 7.4% trong số người Hoa, người Hàn Quốc và người Thái, còn người da trắng và người gốc Tây Ban Nha chỉ 0.7%. Người gốc châu Á và người không thuộc sắc tộc da đen dùng allopurinol thì nguy cơ bị hội chứng quá mẫn với allopurinol gấp 3 lần người da trắng, bệnh này nặng thì da bong tróc diện rộng, có thể đe dọa mạng sống. Hướng dẫn gout năm 2020 của Hiệp hội Thấp khớp Hoa Kỳ khuyến nghị với hai nhóm người đó là trước khi dùng thuốc có thể cân nhắc xét trước. **Mới bắt đầu hạ acid uric lại dễ kích thích cơn hơn**, nên hướng dẫn yêu cầu rõ trong 3 đến 6 tháng đầu phải uống kèm một thuốc chống viêm để phòng. 「Uống thuốc hạ acid uric xong đau hơn nên bỏ thuốc」 là cách hỏng phổ biến nhất của việc này. Khi đau thì việc phải làm là theo đơn bác sĩ thêm thuốc chống viêm, chứ không phải ngừng thuốc hạ acid uric. Giá trị đạt chuẩn mỗi nơi viết khác nhau: <6 mg/dL quy đổi ra xấp xỉ 360 µmol/L, giấy xét nghiệm ở Trung Quốc hay dùng µmol/L. Xét nghiệm thấy acid uric cao mà chưa bao giờ lên cơn là chuyện khác, xem mục 19 ở phần 6. Đồ uống có đường và rượu xem mục 7 và 19 ở phần 2 (không uống đồ có đường, uống ít rượu hoặc không uống rượu) diff --git "a/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" "b/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" index 48de50e0..461d7db3 100644 --- "a/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" +++ "b/vi/book/17-\345\256\266\351\207\214\346\234\211\350\200\201\344\272\272.md" @@ -2,7 +2,7 @@ # 17. Khi nhà có người lớn tuổi -Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn rắc rối không phải vì thiếu tiền, mà là chưa nói cho rõ và chưa ký từ trước. Mục cuối nói về loét áp lực cho người nằm lâu, tính theo tỷ lệ tử vong, không so với các mục khác trong phần này. +Thước đo của mục này là tiền và tự do cá nhân. Khi nhà có người lớn tuổi, phần lớn rắc rối không phải vì thiếu tiền, mà vì chưa nói rõ và ký giấy từ trước. Hai mục cuối nói về loét do nằm liệt lâu và phẫu thuật đụng thủy tinh thể; hai mục đó tính theo tỉ lệ tử vong, không so với các mục khác trong mục này. ### 1. Trong khi người già còn minh mẫn, hãy viết giấy chỉ định trước người giám hộ tương lai @@ -12,7 +12,6 @@ Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn - Mức bằng chứng: A - Nguồn: Ủy ban Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự, 民法典（第三十三条）. - Ghi chú: Bản văn bản này gọi là giám hộ tự nguyện bằng văn bản, tức người do mình chỉ định từ trước. Người này đứng trước thứ tự giám hộ mặc định mà pháp luật đặt ra. Người già sống một mình, con cái ở nơi khác, nhà có quan hệ phức tạp thì càng cần làm trước. Nên đi công chứng ở văn phòng công chứng - ### 2. Hãy lập di chúc, nhớ bản lập sau gạt bản lập trước, di chúc công chứng không còn được ưu tiên - Chi phí: Tự tay viết một bản di chúc, không tốn tiền. Đi công chứng, vài trăm đến hơn nghìn tệ. Khó ở chỗ phải nói chuyện với người nhà về chuyện sau khi mình chết @@ -21,7 +20,6 @@ Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn - Mức bằng chứng: A - Nguồn: Ủy ban Đại biểu nhân dân toàn quốc (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千一百三十四条至第一千一百四十二条）. - Ghi chú: Di chúc tự tay viết phải do chính người viết viết tay từ đầu đến cuối, ký tên, ghi rõ năm tháng ngày. Di chúc in bằng máy cần hơn hai người chứng kiến. Người lập di chúc và người chứng kiến phải ký vào từng trang và ghi rõ năm tháng ngày. Người thừa kế, người được ghi vào di chúc để chia đồ, và thân thích của họ đều không được làm người chứng kiến. Viết sai hình thức là cả bản vô hiệu, đó là chỗ hỏng phổ biến nhất - ### 3. Tiền của người già để riêng một tài khoản, chi tiêu lớn thì đặt quy tắc hai người xác nhận - Chi phí: Không tốn tiền. Nói chuyện với người nhà một lần, đặt ra một con số là được. Khó ở chỗ mở lời thế nào, đừng để người già thấy bạn đang đề phòng ông ấy @@ -30,7 +28,6 @@ Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; về nguyên tắc chống lừa đảo chung xem phần 8 và phần 13 - Ghi chú: Đừng nói chuyện này là 「sợ ông bị lừa」. Hãy nói là 「tiền trong nhà ghi sổ chung」, người già dễ chấp nhận hơn. Cũng đừng giao hết tiền của người già cho con cái giữ, như thế sẽ nảy sinh một loại tranh chấp khác - ### 4. Cho người già một câu để nói ra bất cứ lúc nào mà thoát khỏi việc - Chi phí: Không tốn tiền. Nói chuyện với người già một lần, vài phút là đủ @@ -39,7 +36,6 @@ Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp - Ghi chú: Dùng cùng quy tắc xác nhận của hai người ở mục 3 của phần này (tiền của người già để riêng một tài khoản). Việc chống ngã trong nhà làm luôn tiện tay, xem phần 1. Câu này để người già có chỗ xuống, để ông ấy khỏi phải cãi lại ngay tại chỗ, không phải để đề phòng ông ấy. Bị đám bán hàng vây, khó nhất là mở lời từ chối trước mặt cả một căn phòng, còn khó hơn là phân biệt thật giả - ### 5. Mấy thứ「đầu tư dưỡng già」yêu cầu người già nộp tiền trước thì đừng đụng: mua thẻ, mua giường, mua căn hộ dưỡng già, dưỡng già du lịch, mua sản phẩm cho người già đều là một kiểu huy động vốn trái phép - Chi phí: Không tốn tiền. Đọc mấy cái tên dưới đây cho người già nghe một lượt, vài phút. Sau này gặp thì từ chối thẳng: không đi nghe giảng, không đi tham quan, không để lại số điện thoại. Khó ở chỗ nếu người già đã đi mấy lần, đã quen người thì khó khuyên lại @@ -48,7 +44,6 @@ Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn - Mức bằng chứng: A - Nguồn: Bộ Dân vụ (民政部) (2023). Cảnh báo rủi ro về huy động vốn trái phép trong lĩnh vực dưỡng già, 关于养老服务领域非法集资的风险提示. （đăng trên Trang web chính phủ Trung Quốc）；Bộ Dân vụ (民政部) và Ủy ban Giám sát Ngân hàng và Bảo hiểm Trung Quốc (中国银保监会) (2021). Cảnh báo rủi ro về huy động vốn trái phép trong lĩnh vực dưỡng già, 关于养老领域非法集资的风险提示. ；Quốc vụ viện (国务院) (2021). Bộ luật Phòng ngừa và xử lý huy động vốn trái phép（số 737 của Quốc vụ viện, Điều hai, hai mươi lăm và ba mươi）, 防范和处置非法集资条例（国务院令第 737 号，第二、二十五、三十条）. ；Viện kiểm sát nhân dân tối cao (最高人民检察院) (2022). Các vụ án điển hình của cơ quan kiểm sát đập phá và chấn chỉnh tội phạm lừa đảo nhằm vào người già, 检察机关打击整治养老诈骗犯罪典型案例. ；Tòa án nhân dân tối cao (最高人民法院) (2025). Các vụ án điển hình trừng trị lừa đảo liên quan sinh kế theo pháp luật（vụ năm, vụ lừa đảo của Nhi Mỗ Mung）, 依法惩治涉民生领域诈骗犯罪典型案例（案例五 倪某荣诈骗案）. - Ghi chú: Phán đoán không dựa vào 「cái này trông có vẻ chính quy」, mà dựa vào 「có bắt nộp trước không, có hứa trả lời không」. Cơ sở dưỡng già vốn là bán dịch vụ, trong lời mà nói đến đầu tư, hoàn lợi, thu hồi vốn thì đừng tin. Bị lôi đi nghe giảng, đi du lịch, đi khám miễn phí thì bản thân không vi phạm pháp, cũng rất khó chặn. Chặn được là bước móc ví, dùng cùng quy tắc xác nhận của hai người ở mục 3. Nếu đã nộp tiền rồi, thì gọi ngay 110 hoặc 96110 yêu cầu ngừng chi trả theo mục 2 ở phần 8, đồng thời báo việc cho bộ phận đầu mối ở địa phương phụ trách xử lý huy động vốn trái phép. Dưỡng già dựa vào nhà là lối đi khác, xem mục 6 （bảo hiểm ngược chiều thế chấp nhà ở cho người già） - ### 6. Trừ bảo hiểm ngược chiều thế chấp nhà ở cho người già của công ty bảo hiểm, mấy thứ「dưỡng già dựa vào nhà」còn lại đừng đụng, tuyệt đối không thế chấp nhà đi mua sản phẩm tài chính - Chi phí: Không tốn tiền. Sổ đỏ trong nhà đừng giao cho bất kỳ ai đến tận cửa dưới danh 「dưỡng già dựa vào nhà」. Các thỏa thuận thế chấp, bảo đảm, vay nợ thì trước khi ký đưa con cái xem một lượt. Khó ở chỗ người đến tận cửa thường rất nhiệt tình, đi lại nhiều chuyến, người già ngại từ chối @@ -66,7 +61,6 @@ Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn - Mức bằng chứng: A - Nguồn: Cục Quản lý bảo hiểm y tế quốc gia và bảy cơ quan khác (2026). 关于印发《加快建立长期护理保险制度实施方案》的通知. : đối tượng bảo hiểm là 「按规定参保缴费且失能状态长期持续（一般为6个月以上），经申请通过评估认定的失能人员」, 「制度起步阶段保障重度失能人员」; 「待遇享受不设起付标准。符合规定的长期护理服务费用，按未就业城乡居民参保政策参保的，基金支付比例为50%左右；按单位职工参保政策参保的，基金支付比例为70%左右，退休人员享受单位职工参保待遇」; 「参保人员基金年度最高支付限额不超过统筹地区上年度城乡居民人均可支配收入的50%」; 「待遇享受固定等待期原则上按照6个月设置」; xem thêm Văn phòng Trung ương Đảng Cộng sản Trung Quốc, Văn phòng Hội đồng Nhà nước 《关于加快建立长期护理保险制度的意见》. - Ghi chú: Mất khả năng phải 「长期持续（一般为6个月以上）」, nằm viện mấy tuần thì chưa đủ điều kiện xin. Trước khi được hưởng còn phải qua 6 tháng chờ. Các khoản không tính là chăm sóc như tiền giường, tiền ăn thì bảo hiểm chăm sóc dài hạn không trả. Phần chi phí y tế vốn phải do bảo hiểm y tế trả thì bảo hiểm chăm sóc dài hạn cũng không trả. Người đang nhận tiền chăm sóc dưỡng về tai nạn lao động thì không được hưởng trùng. Khoản tiền này và khoản trợ cấp chăm sóc trong hai khoản trợ cấp dành cho người khuyết tật là hai thủ tục, hai khoản tiền, không xung đột, xem mục 8 ở phần 7 (hai khoản trợ cấp). Chăm sóc tại nhà và tại cộng đồng được ưu ái hơn trong tỷ lệ hoàn, không nhất thiết phải vào cơ sở. Chính sách này từng địa phương triển khai theo lô, tiêu chuẩn đánh giá mức mất khả năng cũng khác nhau. Gọi 12393 hoặc số của cơ quan thực hiện bảo hiểm y tế nơi tham gia bảo hiểm, hỏi địa phương đã bắt đầu chưa và đặt lịch đánh giá thế nào. Đánh giá không đạt thì cũng đừng bỏ cuộc, khi tình trạng mất khả năng nặng hơn thì có thể xin lại. - ### 8. Khi trong nhà có người nằm liệt giường lâu ngày, hãy coi loét áp lực là kẻ thù số một: dùng nệm hơi điện, xoay người đúng giờ, mỗi ngày soi một lần chỗ xương nhô ra - Chi phí: Nệm hơi điện dùng trong nhà loại luân phiên áp lực, vài trăm tệ trở lên. Mỗi lần xoay người chỉ mất vài phút. Chỗ khó là cứ hai ba tiếng phải xoay một lần, ban đêm cũng không dừng, một người không gánh nổi @@ -75,3 +69,12 @@ Thước đo: tiền và tự do cá nhân. Nhà có người già, phần lớn - Mức bằng chứng: B - Nguồn: Porter SB, Pla R, Chow JH, et al. (2022). Preoperative Pressure Ulcers, Mortality, and Complications in Older Hip Fracture Surgery Patients. JAAOS Global Research & Reviews, 6(11). ；Ottaviani S, Rondanina E, Longo E, 等 (2026). Impact of pressure ulcers and frailty on long-term mortality: a prospective cohort study of hospitalized older adults. European Geriatric Medicine, 17(2), 665-672. ；Shi C, Dumville JC, Cullum N (2018). Support surfaces for pressure ulcer prevention: A network meta-analysis. PLoS ONE, 13(2), e0192707. ；Latimer SL, Chaboyer WP, Probst S, 等 (2026). Repositioning for pressure injury prevention in adults. Cochrane Database of Systematic Reviews, 6, CD009958. - Ghi chú: Chấm B vì ba lý do. Một, hai nghiên cứu tỷ lệ tử vong chỉ thấy 「có loét áp lực thì sau đó tình hình tệ hơn」, điều đó không có nghĩa là 「phòng được loét áp lực thì giảm được tỷ lệ tử vong」. Hai, kết luận của mấy thử nghiệm ngẫu nhiên về khoảng xoay người rất không chắc. Ba, bằng chứng về nệm hơi điện chắc hơn một chút, nhưng nghiên cứu so đệm giường bệnh viện, áp sang nệm dùng ở nhà là suy rộng ra. Soi mỗi ngày xương cùng, gót chân, hông, vai lưng, xem có 「vết đỏ ấn xuống không mất màu」 không, đây là việc chăm sóc thường lệ, không có tài liệu riêng. Dịch vụ chăm sóc mà bảo hiểm chăm sóc dài hạn hoàn trả, xem mục 7 của phần này. Một chân sưng đột ngột thì xử lý như huyết khối tĩnh mạch sâu, xem mục 11 ở phần 13. Khoảng thời gian nằm liệt giường lâu sau khi ngã hoặc chấn thương nặng, xem mục 34 ở phần 1. + +### 9. Người lớn tuổi bị đụng thủy tinh thể đã ảnh hưởng tới việc nhìn đường thì đi đánh giá ở khoa mắt để phẫu thuật; ai đụng cả hai mắt thì đừng kéo dài mắt thứ hai + +- Chi phí: Đưa người lớn tuổi đi khám mắt một lần trước, để bác sĩ quyết định có nên phẫu thuật không. Một mắt thì một lần phẫu thuật, nhiều bệnh viện làm theo kiểu ngoại trú, ngày đó về nhà được. Ai đụng thủy tinh thể cả hai mắt thì phải chia làm hai lần. Bạn tự bỏ ra bao nhiêu, tuỳ loại thủy tinh thể nhân tạo (kính mắt giả) được thay vào mắt và bảo hiểm y tế địa phương hoàn trả thế nào, hỏi rõ trước khi phẫu thuật. +- Nói thẳng: Phụ nữ trên 70 tuổi phẫu thuật đụng thủy tinh thể mắt thứ nhất càng sớm càng tốt, trong một năm số lần ngã giảm khoảng ba phần mười tư, số người gãy xương cũng giảm. Nhưng số người từng bị ngã thì không giảm. Có phẫu thuật mắt thứ hai hay không, xét riêng chuyện ngã thì không thấy khác biệt rõ rệt. Người lớn tuổi đã phẫu thuật có tỷ lệ gãy xương hông trong một năm thấp hơn khoảng một phần mười sáu. +- Lợi ích: Một thử nghiệm ngẫu nhiên gồm 306 phụ nữ trên 70 tuổi bị đụng thủy tinh thể. Một nửa phẫu thuật mắt thứ nhất trong khoảng 4 tuần, nửa còn lại chờ theo lệ thường là 12 tháng. Trong một năm, nhóm phẫu thuật sớm ngã ít đi khoảng 34% số lần ngã. Giá trị gốc là tỷ lệ ngã 0.66, khoảng tin cậy 95% CI 0.45 đến 0.96. Người ngã ít nhất một lần: nhóm phẫu thuật 49%, nhóm chờ 45%, không ít hơn. Người ngã từ hai lần trở lên: nhóm phẫu thuật 18%, nhóm chờ 25%. Nhóm phẫu thuật có 4 người gãy xương (3%), nhóm chờ có 12 người (8%). Tổng quan Cochrane đã đưa thử nghiệm này vào, kết luận là phẫu thuật mắt thứ nhất ở phụ nữ làm giảm số lần ngã, còn phẫu thuật mắt thứ hai thì không. Thử nghiệm ngẫu nhiên cho mắt thứ hai gồm 239 phụ nữ trên 70 tuổi. Nhóm phẫu thuật có tỷ lệ ngã thấp hơn 32%, nhưng khác biệt có thể chỉ là trùng hợp (tỷ lệ ngã 0.68, 95% CI 0.39 đến 1.19). Trong mẫu 5% của bảo hiểm y tế Hoa Kỳ có 1113640 bệnh nhân đụng thủy tinh thể từ 65 tuổi trở lên (từ 2002 đến 2009). Nghiên cứu này chỉ ghi nhận và theo dõi, không chia nhóm đối chứng. Người được phẫu thuật có xác suất gãy xương hông trong một năm thấp hơn khoảng 16% (OR 0.84, 95% CI 0.81 đến 0.87). Chênh lệch tuyệt đối là 0.20 điểm phần trăm, tức khoảng 2 ca ít hơn trên mỗi 1000 người +- Mức bằng chứng: B +- Nguồn: Harwood RH, Foss AJ, Osborn F, Gregson RM, Zaman A, Masud T (2005). Falls and health status in elderly women following first eye cataract surgery: a randomised controlled trial. British Journal of Ophthalmology, 89(1), 53-59. ; Foss AJ, Harwood RH, Osborn F, Gregson RM, Zaman A, Masud T (2006). Falls and health status in elderly women following second eye cataract surgery: a randomised controlled trial. Age and Ageing, 35(1), 66-71. ; Gillespie LD, Robertson MC, Gillespie WJ, et al. (2012). Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews, 9, CD007146. ; Tseng VL, Yu F, Lum F, Coleman AL (2012). Risk of fractures following cataract surgery in Medicare beneficiaries. JAMA, 308(5), 493-501. ; Desapriya E, Subzwari S, Scime-Beltrano G, et al. (2010). Vision improvement and reduction in falls after expedited cataract surgery: systematic review and metaanalysis. Journal of Cataract and Refractive Surgery, 36(1), 13-19. ; Meuleners LB, Lee AH, Ng JQ, Morlet N, Fraser ML (2012). First eye cataract surgery and hospitalization from injuries due to a fall: a population-based study. Journal of the American Geriatrics Society, 60(9), 1730-1733. ; Meuleners LB, Fraser ML, Ng J, Morlet N (2014). The impact of first- and second-eye cataract surgery on injurious falls that require hospitalisation: a whole-population study. Age and Ageing, 43(3), 341-346. ; Keay L, Ho KC, Rogers K, et al. (2022). The incidence of falls after first and second eye cataract surgery: a longitudinal cohort study. Medical Journal of Australia, 217(2), 94-99. ; Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委) (2022). 「十四五」全国眼健康规划（2021-2025年）. +- Ghi chú: Tranh cãi. Gộp cả hai thử nghiệm ngẫu nhiên lại, số người từng bị ngã không giảm rõ rệt (OR 0.81, 95% CI 0.55 đến 1.17). Trong hồ sơ nhập viện của toàn bộ dân số ở Tây Úc, trong năm sau khi làm xong mắt thứ nhất, số người phải nằm viện vì chấn thương do ngã lại tăng 27%. Trong khoảng giữa hai lần phẫu thuật, loại nhập viện đó gấp 2.14 lần so với trước phẫu thuật. Kiểu so sánh trước – sau này không tách được ảnh hưởng của phẫu thuật với ảnh hưởng của việc người ta già đi. Trong một nghiên cứu theo dõi khác ở Úc, sau khi làm xong mắt thứ hai thì số lần ngã giảm nhiều nhất. Nên với ai đụng thủy tinh thể cả hai mắt, đừng kéo dài mắt thứ hai. Trong khoảng giữa hai lần phẫu thuật, đi bộ và leo cầu thang phải càng cẩn thận càng tốt. Cochrane còn đưa vào một thử nghiệm xử lý vấn đề thị lực cho người lớn tuổi, số lần ngã lại tăng khoảng năm phần mười bảy. Phẫu thuật giải quyết chuyện nhìn có rõ không, còn muốn ít ngã thì còn phải dựa vào mục 13 ở phần 1 (giữ thăng bằng và sức chân). Xương có chịu nổi cú ngã không, xem mục 39 ở phần 1 (đo mật độ xương) và mục 40 ở phần 1 (phát hiện loãng xương thì dùng thuốc). Kế hoạch quốc gia về sức khoẻ mắt ghi rõ phẫu thuật đụng thủy tinh thể để phục hồi thị lực đã được triển khai rộng rãi ở cấp huyện, không nhất thiết phải tới bệnh viện ở thành phố lớn. Người hưởng lợi chủ yếu là cha mẹ và ông bà của bạn. diff --git "a/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" "b/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" index 0d2eb596..3778e80e 100644 --- "a/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" +++ "b/vi/book/18-\345\205\273\345\255\251\345\255\220\345\210\222\344\270\215\345\210\222\347\256\227.md" @@ -12,7 +12,6 @@ Phần này tính hai thứ: tiền và thời gian. Giống phần 10 tính chu - Mức bằng chứng: A - Nguồn: Văn phòng Trung ương Đảng Cộng sản Trung Quốc (中共中央办公厅), Văn phòng Hội đồng Nhà nước (国务院办公厅) (2025). 育儿补贴制度实施方案. - Ghi chú: Đây là mức sàn thống nhất toàn quốc. Có tỉnh, thành có thêm trợ cấp địa phương riêng, có thể nhận cộng với khoản này. Ba năm cộng lại 10800 tệ, trong chi phí nuôi con chỉ chiếm một phần rất nhỏ, đừng lấy nó làm căn cứ để quyết định có sinh hay không - ### 2. Nghỉ thai sản 98 ngày, trợ cấp thai sản do quỹ bảo hiểm thai sản chi trả theo lương bình quân hằng tháng của công nhân viên năm trước của đơn vị - Chi phí: Không tốn tiền. Đơn vị sử dụng lao động đi khai báo, bạn không phải chạy đi đâu @@ -21,7 +20,6 @@ Phần này tính hai thứ: tiền và thời gian. Giống phần 10 tính chu - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2012). 女职工劳动保护特别规定（国令第 619 号，第七条、第八条）. - Ghi chú: 98 ngày là mức tối thiểu do Nhà nước đặt ra. Điều lệ dân số và kế hoạch hoá của các tỉnh thường còn cộng thêm hơn 60 ngày nghỉ thưởng cho việc sinh con, tính cả thường là từ 158 ngày. Lương trong thời gian nghỉ thưởng do bên nào chi trả thì quy định ở mỗi tỉnh khác nhau - ### 3. Biết điều này: không được giảm lương hoặc sa thải vì mang thai, vì sinh con, vì cho con bú - Chi phí: Không tốn tiền @@ -30,7 +28,6 @@ Phần này tính hai thứ: tiền và thời gian. Giống phần 10 tính chu - Mức bằng chứng: A - Nguồn: Quốc vụ viện (国务院) (2012). 女职工劳动保护特别规定（国令第 619 号，第五条、第九条、第十三条）. - Ghi chú: Nếu thật sự bị sa thải vi phạm pháp thì đường đòi bồi thường xem phần 19. Trọng tâm khi lấy chứng cứ là thông báo bằng văn bản về việc chuyển vị trí, cắt giảm lương và lịch sử trò chuyện, phải giữ kỹ hai thứ đó - ### 4. Cuốn sổ thời gian tính theo「thời gian của ai bị chiếm mất」, không tính theo「có vất vả không」 - Chi phí: Không tốn tiền, cái mất là thời gian. Trước khi trẻ đủ 3 tuổi thì về cơ bản thời gian bị chiếm hết cả ngày. Chỗ khó là chuyện này lặp lại ngày này sang ngày khác, không thể cắn răng chịu mấy ngày rồi qua @@ -39,7 +36,6 @@ Phần này tính hai thứ: tiền và thời gian. Giống phần 10 tính chu - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; thước đo xem phần 10 - Ghi chú: Không cần tính số chính xác lắm. Cái hữu ích là hai người cùng nhìn một bảng để bàn chia việc trước khi sinh, khỏi sinh xong rồi cãi nhau - ### 5. Cuốn sổ tiền chia làm ba đoạn để tính: 0 đến 3 tuổi, giáo dục bắt buộc, sau giáo dục bắt buộc - Chi phí: Tốn bao nhiêu tiền thì không có con số cố định, khác nhau theo thành phố và theo cách nuôi, chênh nhau rất xa. Chỗ khó là phải bình tĩnh tách ba đoạn sổ ra tính riêng, không thể đoán bừa một con số bằng cảm tính @@ -48,7 +44,6 @@ Phần này tính hai thứ: tiền và thời gian. Giống phần 10 tính chu - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp - Ghi chú: Trước hết cộng vào các khoản chắc chắn nhận được như trợ cấp nuôi con, trợ cấp thai sản, bảo hiểm y tế hoàn trả, rồi xem còn thiếu bao nhiêu. Đừng dựa vào những con số「nuôi một đứa con tốn bao nhiêu triệu」lan truyền trên mạng, loại số đó không rõ thước đo, bạn không biết nó tính những gì và bỏ sót những gì - ### 6. Sinh vì người lớn, sinh vì hôn nhân, sinh vì lúc về già, mỗi kiểu ghi một cuốn sổ riêng - Chi phí: Không tốn tiền, cái mất là một cuộc nói chuyện nghiêm túc. Chỗ khó là phải nói thẳng ra, không vòng vo @@ -56,4 +51,4 @@ Phần này tính hai thứ: tiền và thời gian. Giống phần 10 tính chu - Lợi ích: Cách nghĩ giống phần 10. Kỳ vọng của người khác có thể tính là một yếu tố cân nhắc của bạn. Nhưng chi phí thì ghi vào sổ của chính bạn, không ghi vào sổ của họ. 「Nuôi con khi về già」càng phải tính riêng một cuốn, vì nó giả định năng lực tài chính và ý chí của một người khác hai mươi năm sau - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; xem phân tích cùng loại ở phần 10 -- Ghi chú: Hãy viết cả「hậu quả của việc không sinh」rồi đặt cạnh「hậu quả của việc sinh」để đối chiếu \ No newline at end of file +- Ghi chú: Hãy viết cả「hậu quả của việc không sinh」rồi đặt cạnh「hậu quả của việc sinh」để đối chiếu diff --git "a/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" "b/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" index 68e07118..3813f26b 100644 --- "a/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" +++ "b/vi/book/19-\345\234\250\350\201\214\347\246\273\350\201\214\345\222\214\345\267\245\344\274\244.md" @@ -2,7 +2,7 @@ # 19. Đi làm, nghỉ việc và tai nạn lao động -Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký thế nào, thời hạn tính thế nào. Ba mục đầu nói về số tiền bạn phải có trong tay khi còn đang làm việc. Sáu mục giữa nói về việc nghỉ việc. Bảy mục cuối nói về bệnh nghề nghiệp và tai nạn lao động. Tiền của tai nạn lao động lớn hơn tiền của sa thải một bậc, thời hạn cũng cứng hơn. Tổn thương do bệnh nghề nghiệp thì không hoàn lại được. +Phần này tính ra đều là tiền: tiền tính thế nào, chữ ký thế nào, hạn thời gian tính thế nào. Ba mục đầu nói về số tiền bạn phải nhận được khi còn đang làm việc. Sáu mục kế tiếp nói về nghỉ việc. Bảy mục sau đó nói về bệnh nghề nghiệp và tai nạn lao động. Tiền của tai nạn lao động lớn hơn tiền bị sa thải cả một bậc, và hạn thời gian cũng cứng hơn. Tổn thương do bệnh nghề nghiệp gây ra thì không hoàn được. Cuối phần còn ba mục: bị đơn vị bắt nạt thì làm sao, xin giấy xác nhận nghỉ việc thế nào, thuế thu nhập cá nhân trên khoản bồi thường khi nghỉ việc tính ra sao. ### 1. Tiền làm thêm giờ tính theo ba mức 1.5 lần, 2 lần, 3 lần; không trả thì báo thanh tra lao động, quá hạn không trả còn phải cộng thêm 50% đến 100% @@ -12,7 +12,6 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 劳动法（2018 年第二次修正，第三十六、四十一、四十三、四十四条）. （Cục Quản lý thị trường quốc gia đăng lại）；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 劳动合同法（第八十五条第三项）. - Ghi chú: Ngày nghỉ và ngày lễ tết theo luật phải tính riêng. Ngày lễ tết theo luật cả năm chỉ có Tết Nguyên Đán, Tết Nguyên Tiêu, Quốc tế Lao động, Quốc khánh và mấy ngày lễ hợp pháp khác. Ngày nghỉ liên tục do điều chỉnh lịch nghỉ để bù thì tính là ngày nghỉ, trả 200%. Hai câu「công ty quy định làm thêm giờ phải được duyệt」và「tôi tự nguyện ở lại」không có nghĩa công ty được phép không trả tiền, nhưng chúng làm bạn khó lấy bằng chứng hơn, nên phải giữ lại lệnh phân công và hồ sơ duyệt. Công ty muốn áp dụng hai chế độ giờ làm đặc biệt là tính giờ tổng hợp và giờ không xác định thì phải được cơ quan quản lý lao động chấp thuận trước, không tự quyết được. Thời hiệu khởi kiện khi còn tranh cãi xem mục 19 ở phần 8. - ### 2. Nghỉ phép năm tính theo thâm niên cộng dồn: 5, 10, 15 ngày; nghỉ không được thì đổi thành tiền bằng 300% lương ngày - Chi phí: Không tốn tiền. Cộng lại số năm bạn đã làm ở tất cả các đơn vị. Sổ đóng bảo hiểm xã hội chính là bằng chứng. Chỗ khó là phải ký được tờ「tự nguyện không nghỉ phép năm」khi công ty đưa ra. @@ -21,7 +20,6 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2007). 职工带薪年休假条例（国务院令第 514 号，第二、三、四、五、七条）. ；Bộ Nhân sự và Xã hội (2008). 企业职工带薪年休假实施办法（人社部令第 1 号，第三、四、五、十、十一、十二、十五条）. - Ghi chú: Thời gian làm việc cộng dồn phải cộng cả các đơn vị, đã từng đổi việc thì đừng tính theo thâm niên ở đơn vị hiện tại. **Chỉ khi người lao động「tự nguyện không nghỉ vì lý do của bản thân và có văn bản」thì đơn vị mới chỉ phải trả lương bình thường**, nên đừng ký tờ「tự nguyện không nghỉ phép năm」. Nghỉ phép việc tự sự tích luỹ từ 20 ngày trở lên mà đơn vị không trừ lương thì năm đó không được nghỉ phép năm. Nghỉ ốm vượt số tháng tương ứng thì cũng không được: thâm niên chưa đủ 10 năm là 2 tháng, từ 10 đến 20 năm là 3 tháng, trên 20 năm là 4 tháng. Phép thăm viếng, phép cưới tang, phép thai sản và thời gian lương được trả khi ngừng việc vì tai nạn lao động không tính vào nghỉ phép năm. - ### 3. Thời gian thử việc có trần theo luật, lương không được dưới 80%, và chỉ được thỏa thuận một lần - Chi phí: Không tốn tiền. Trước khi ký, dành vài phút để đối chiếu thời hạn ghi trong hợp đồng với độ dài thời gian thử việc. @@ -30,25 +28,22 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 劳动合同法（第十九、二十、二十一、八十三条）. - Ghi chú: Ba cách làm sai thường gặp nhất. Một là hợp đồng chỉ ký 1 năm nhưng lại thỏa thuận thời gian thử việc 3 tháng. Hai là sau khi hết thử việc thì chuyển vị trí, rồi lại có một thời gian thử việc nữa. Ba là「thời gian thử việc không đóng bảo hiểm xã hội」, nghĩa vụ đóng bảo hiểm xã hội tính từ ngày đầu tiên đi làm, không liên quan gì đến thử việc, nghĩa vụ của phía sử dụng lao động xem mục 16 ở phần 12. Bị sa thải trong thời gian thử việc không có nghĩa là mất trắng, phải xem lý do công ty đưa ra có đúng là các trường hợp luật cho phép sa thải hay không. Bị chấm dứt sai luật thì tính 2N theo mục 6 của phần này. - -### 4. Bị sa thải thì tính N trước: cứ đủ mỗi năm một tháng lương, chưa đủ sáu tháng thì tính nửa tháng +### 4. Bị sa thải thì tính rõ N trước: cứ mỗi năm làm đủ một năm thì một tháng lương, chưa đủ sáu tháng thì tính nửa tháng -- Chi phí: Không tốn tiền. Bấm máy tính một lần, mất mười phút. -- Nói thẳng: Khoản bồi thường kinh tế công ty phải trả khi bạn bị sa thải tính thế này: đủ một năm thì một tháng lương, đủ nửa năm đến chưa đủ một năm thì tính như một năm, chưa đủ nửa năm thì nửa tháng. Mức lương tháng ở đây là trung bình 12 tháng trước khi rời đi, có tính cả thưởng, trợ cấp, phụ cấp. Lương vượt 3 lần mức lương tháng bình quân của năm trước tại địa phương thì tính trần ở 3 lần, số năm công chỉ tính tối đa 12 năm. -- Lợi ích: Khoản bồi thường kinh tế theo hợp đồng lao động chính là khoản tiền đơn vị phải trả khi bạn bị sa thải, mọi người thường gọi là N. Cách tính là：「每满一年支付一个月工资……六个月以上不满一年的，按一年计算；不满六个月的，向劳动者支付半个月工资」, tức là đủ một năm thì một tháng lương, từ sáu tháng đến chưa đủ một năm thì tính như một năm, chưa đủ sáu tháng thì trả nửa tháng lương. Lương tháng vượt ba lần lương tháng bình quân của người lao động năm trước tại địa phương thì tính trần ở ba lần, và thâm niên chỉ tính tối đa đến mười hai năm +- Chi phí: Không tốn tiền. Cầm máy tính tính một lần, mười phút. +- Nói thẳng: Tiền bồi thường kinh tế mà công ty phải trả cho bạn khi bị sa thải tính thế này: làm đủ một năm thì một tháng lương, đủ nửa năm mà chưa đủ một năm thì tính như một năm, chưa đủ nửa năm thì nửa tháng lương. Lương tháng ở đây là mức bình quân của 12 tháng trước khi rời đi, tiền thưởng, trợ cấp và phụ cấp đều phải tính vào. Lương cao hơn 3 lần mức lương bình quân theo tháng của năm trước ở địa phương thì lấy 3 lần làm trần. Số năm công sức tối đa chỉ tính đến 12 năm. +- Lợi ích: Bồi thường kinh tế mà Bộ luật Hợp đồng lao động quy định chính là khoản tiền đơn vị phải trả khi bạn bị sa thải, ai cũng gọi là N. Cách tính là:「mỗi năm làm đủ một năm thì trả một tháng lương……từ sáu tháng trở lên mà chưa đủ một năm thì tính là một năm; chưa đủ sáu tháng thì trả cho người lao động nửa tháng lương」. Lương tháng cao hơn ba lần lương bình quân theo tháng của người lao động năm trước ở địa phương thì lấy ba lần làm trần, và thâm niên công sức chỉ tính đến mười hai năm - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第四十七条）. -- Ghi chú: Lương tháng ở đây là lương trung bình mười hai tháng trước khi hợp đồng lao động chấm dứt. Thưởng, trợ cấp và phụ cấp đều phải tính vào, không chỉ tính lương cơ bản. Đây là điểm công ty dễ làm thấp nhất. - -### 5. Công ty không báo trước 30 ngày mà cho bạn nghỉ thì còn phải trả thêm một tháng lương +- Ghi chú: Lương tháng ở đây là mức bình quân của mười hai tháng trước khi hợp đồng lao động chấm dứt. Tiền thưởng, trợ cấp và phụ cấp đều phải tính vào, không phải chỉ tính lương cơ bản. Chỗ này hay bị công ty tính thấp nhất. Khoản tiền đó có phải đóng thuế thu nhập cá nhân không, xem mục 18 của phần này (thuế thu nhập cá nhân trên khoản bồi thường) +### 5. Công ty sa thải bạn mà không báo trước 30 ngày thì còn phải trả thêm một tháng lương - Chi phí: Không tốn tiền. -- Nói thẳng: Công ty không báo trước 30 ngày bằng văn bản mà bắt bạn đi thì phải trả thêm một tháng lương ngoài N, đó là cái người ta gọi là N+1. Khi thương lượng, hãy tách tháng lương đó thành một khoản riêng, đừng để công ty gộp vào trong N. -- Lợi ích: Hợp đồng lao động quy định khi đơn vị sử dụng lao động chấm dứt hợp đồng theo điều này thì có hai cách. Cách một là「提前三十日以书面形式通知劳动者本人」. Cách hai là「额外支付劳动者一个月工资后，可以解除劳动合同」. Muốn bạn đi ngay thì phải bỏ thêm một tháng lương. Tháng lương thêm đó dân gọi là tiền báo trước, tính ngoài N +- Nói thẳng: Công ty sa thải bạn vì sau khi ốm không làm nổi, không làm nổi công việc, hoặc vì hoàn cảnh thay đổi lớn, mà không báo trước 30 ngày bằng văn bản, thì ngoài N còn phải trả thêm một tháng lương, đó là cái gọi là N+1. Khi thương lượng thì tách tháng đó thành một mục riêng, đừng để công ty gộp vào trong N. +- Lợi ích: Bộ luật Hợp đồng lao động quy định khi đơn vị sử dụng lao động chấm dứt hợp đồng theo điều này thì có hai cách. Một là「thông báo trước ba mươi ngày bằng văn bản cho chính người lao động」. Cách kia là「sau khi trả thêm cho người lao động một tháng lương thì có thể chấm dứt hợp đồng lao động」. Muốn bạn đi ngay thì phải bỏ thêm một tháng lương. Tháng lương dư ra đó dân gian gọi là tiền bù giáp ngạp, tính ngoài N thành một khoản riêng. Tháng đó tính theo lương tháng trước cùng của bạn, còn N tính theo lương bình quân của 12 tháng trước khi nghỉ việc, hai khoản khác nhau ngay cả cái tính gốc - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第四十条）. -- Ghi chú: Vì vậy kết quả thường gặp của một cuộc sa thải hợp pháp là N+1. Khi thương lượng, tính riêng N và 1, đừng để công ty gộp tháng lương thêm vào trong N. - +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第三十六、三十九、四十、四十一、四十六、四十七、八十七条）. ；Hội đồng Nhà nước (国务院) (2008). Quy chế thực thi Bộ luật Hợp đồng lao động, 劳动合同法实施条例（国务院令第 535 号）第二十条:「用人单位依照劳动合同法第四十条的规定，选择额外支付劳动者一个月工资解除劳动合同的，其额外支付的工资应当按照该劳动者上一个月的工资标准确定」. +- Ghi chú: Tháng lương thêm đó chỉ có trong ba trường hợp ở điều bốn mươi của Bộ luật Hợp đồng lao động: ốm hoặc bị thương không phải do công việc, hết thời gian điều trị mà không làm nổi công việc cũ cũng không làm nổi công việc được xếp lại; không làm nổi công việc, sau khi đào tạo hoặc điều chuyển vị trí vẫn không làm nổi; hoàn cảnh khách quan lúc ký hợp đồng thay đổi lớn, không thỏa thuận được. Các trường hợp khác không có tháng đó: sa thải vì lý do kinh tế (điều bốn mươi mốt) chỉ yêu cầu công ty báo trước 30 ngày cho công đoàn hoặc toàn thể người lao động; công ty đề xuất, hai bên thỏa thuận chấm dứt thì là N; bạn có lỗi như kỷ luật nghiêm trọng mà bị sa thải thì không có bồi thường; công ty chấm dứt sai luật thì là 2N. Thuộc ba trường hợp của điều bốn mươi thì khi thương lượng hãy tính N và 1 tách ra, đừng để công ty gộp tháng lương thêm vào trong N ### 6. Công ty chấm dứt hợp đồng sai luật thì tiền bồi thường bằng hai lần mức bồi thường kinh tế - Chi phí: Không tốn tiền. Đi qua hoà giải lao động phải chờ mấy tháng. Chỗ khó là mấy tháng đó bạn phải theo suốt, không được bỏ giữa chừng. @@ -57,7 +52,6 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第八十七条）. - Ghi chú: Đây là cái dân gọi là 2N. Cái gì tính là chấm dứt sai luật thì phải xem từng trường hợp. Hãy hỏi trợ giúp pháp lý hoặc luật sư trước, rồi mới quyết định thương lượng hay đi khởi kiện, các bước đi xem phần 7. - ### 7. Đừng ký「tự nguyện nghỉ việc vì lý do cá nhân」, ký là mất trắng cả khoản N - Chi phí: Không tốn tiền. Chỗ khó là phải chịu được áp lực bị bảo ký ngay tại chỗ. @@ -66,52 +60,40 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Mức bằng chứng: B - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). 中华人民共和国劳动合同法（2008 年 1 月 1 日施行）（第三十八条、第四十六条）. - Ghi chú: Câu công ty hay dùng là「ký trước để làm thủ tục, phần bồi thường tính riêng」. Muốn nhận thì ghi thẳng vào văn kiện ngay tại chỗ, không ghi thì không ký. Không ký trong ngày không có bất kỳ hậu quả pháp lý nào, để nguôi một đêm rồi hãy quyết. - -### 8. Trước khi nghỉ việc, lưu lại phiếu lương, bản chấm công, hợp đồng lao động, sổ bảo hiểm xã hội và lịch sử chat +### 8. Trước khi nghỉ việc thì lưu trước phiếu lương, bản chấm công, hợp đồng lao động, hồ sơ bảo hiểm xã hội và lịch sử trò chuyện -- Chi phí: Không tốn tiền. Dành nửa tiếng để lưu hồ sơ. Chỗ khó là phải nhớ lưu trước khi đi. -- Nói thẳng: Trả lại máy tính và tài khoản là phiếu lương, bản chấm công, hợp đồng, sổ bảo hiểm xã hội và lịch sử chat không lấy lại được nữa. Khi khởi kiện, lương bao nhiêu, có làm thêm giờ hay không, vẫn phải dựa vào hồ sơ trong tay bạn. Hãy lưu hồ sơ quan hệ lao động của chính bạn, đừng mang mã nguồn và danh sách khách hàng của công ty đi. -- Lợi ích: Một khi trả lại máy tính và tài khoản thì những hồ sơ đó không lấy lại được nữa. Khi khởi kiện, việc ai phải đưa bằng chứng tuy nghiêng về phía đơn vị hơn một chút, nhưng lương bao nhiêu, có làm thêm giờ hay không thì vẫn dựa vào bằng chứng trong tay bạn -- Mức bằng chứng: C -- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; các bước bảo vệ quyền lợi xem phần 7 -- Ghi chú: Hãy lưu hồ sơ quan hệ lao động của chính bạn. Đừng tiện tay mang mã nguồn, danh sách khách hàng và tài liệu kỹ thuật của công ty đi, rủi ro của việc mang đi xem phần 11. -### 9. Ngay sau khi nghỉ việc, làm hai việc: đăng ký thất nghiệp để lĩnh tiền bảo hiểm thất nghiệp, và đọc kỹ hợp đồng cạnh tranh - -- Chi phí: Không tốn tiền. -- Nói thẳng: Ngay sau khi nghỉ việc, làm hai việc. Một là đi đăng ký thất nghiệp và lĩnh tiền bảo hiểm thất nghiệp. Hai là đọc lại hợp đồng cạnh tranh. Tiền bảo hiểm thất nghiệp tính từ ngày làm thủ tục đăng ký, đăng ký muộn một tháng là mất một tháng tiền. Chỉ người mất việc vì lý do không phải do mình mới được lĩnh, đó cũng là một lý do nữa cho việc không nên ký theo mục 7 của phần này (đừng ký「tự nguyện nghỉ việc vì lý do cá nhân」). -- Lợi ích: Để lĩnh tiền bảo hiểm thất nghiệp có bốn điều kiện: đã đóng bảo hiểm thất nghiệp đủ 1 năm, mất việc vì lý do không phải do mình muốn đi (không thất nghiệp do chủ ý của bản thân), đã làm thủ tục đăng ký thất nghiệp, và vẫn đang đi tìm việc. Số năm đã đóng quyết định lâu nhất là được lĩnh 12, 18 hay 24 tháng. Cách đăng ký trên mạng xem mục 1 ở phần 7 (mất việc thì đăng ký tiền bảo hiểm thất nghiệp trên mạng trước). Hạn chế cạnh tranh dài nhất là 2 năm. Trong 2 năm đó công ty phải trả bạn bồi thường hằng tháng. Hợp đồng không ghi là bao nhiêu thì tính bằng 30% lương tháng trung bình 12 tháng trước khi nghỉ việc, và không được thấp hơn lương tối thiểu tại địa phương. Chỉ khi công ty liên tục 3 tháng không tráng vì lý do từ phía mình thì bạn mới được yêu cầu chấm dứt thoả thuận đó, xem mục 12 ở phần 11 (đã ký hợp đồng cạnh tranh mà công ty không trả bồi thường thì làm gì). Hạn chế cạnh tranh có thời hạn, nhưng trước ngày hết hạn thì vẫn phải giữ đúng thoả thuận +- Chi phí: Không tốn tiền. Dành nửa tiếng để lưu lại hồ sơ. Chỗ khó là phải nhớ lưu trước khi rời đi. +- Nói thẳng: Trả lại máy tính và tài khoản là phiếu lương, bản chấm công, hợp đồng, hồ sơ bảo hiểm xã hội và lịch sử trò chuyện không lấy lại được nữa. Lúc hoà giải lao động, lương bao nhiêu, có làm thêm giờ không, vẫn phải dựa vào hồ sơ bạn tự giữ. Hồ sơ cần lưu là hồ sơ quan hệ lao động của chính bạn, mã nguồn và danh sách khách hàng của công ty thì đừng mang theo. +- Lợi ích: Trả máy tính và tài khoản về là những hồ sơ đó không lấy lại được nữa. Lúc hoà giải lao động, ai phải đưa ra chứng cứ thì có phần nghiêng về phía đơn vị, nhưng lương bao nhiêu, có làm thêm giờ không, những điều đó vẫn phải dựa vào chứng cứ bạn tự giữ - Mức bằng chứng: C -- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; những con số dẫn ra từ Ủy ban Nhà nước (国务院) (1999). 失业保险条例（国务院令第 258 号，第十四、十六、十七条：「失业保险金自办理失业登记之日起计算」）. và Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012). 劳动合同法（2012 年修正，第二十三、二十四条）. （Sở Quản lý Khẩn cấp thành phố Phụ Thuận đăng lại） -- Ghi chú: Để ở mức C là vì hành động「ngày nghỉ việc đi làm ngay」không có tài liệu trực tiếp, còn những con số trong mục Nguồn đều là nguyên văn quy định. Tiền bảo hiểm thất nghiệp không hết hạn mất tác dụng, nhưng nó tính từ ngày làm thủ tục đăng ký, nên làm sớm một ngày là có thêm một ngày tiền. Hồ sơ và cửa đăng ký trên mạng xem phần 7. Mức bồi thường và điều kiện chấm dứt hạn chế cạnh tranh xem phần 11. Đã ký「tự nguyện nghỉ việc vì lý do cá nhân」thì không những không có N, tiền bảo hiểm thất nghiệp cũng mất luôn, xem mục 7 của phần này (đừng ký「tự nguyện nghỉ việc vì lý do cá nhân」). +- Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; đường đi bảo vệ quyền lợi xem phần 7 +- Ghi chú: Hồ sơ cần lưu là hồ sơ quan hệ lao động của chính bạn. Đừng tiện tay mang theo mã nguồn, danh sách khách hàng và tài liệu kỹ thuật của công ty, rủi ro khi mang theo xem phần 11. Lúc nghỉ việc còn phải xin đơn vị giấy xác nhận nghỉ việc, xem mục 17 của phần này (giấy xác nhận nghỉ việc) -### 10. Trước khi vào chỗ làm việc có bụi, tiếng ồn, hóa chất, xem hợp đồng có ghi nguy cơ nghề nghiệp không; ba lần khám sức khoẻ nghề nghiệp do đơn vị sắp xếp và trả tiền +### 9. Trước khi vào chỗ có bụi, tiếng ồn, hoá chất thì xem trong hợp đồng đã ghi nguy cơ chưa; ba kỳ khám sức khoẻ nghề nghiệp do đơn vị sắp xếp và bỏ tiền -- Chi phí: Không tốn tiền. Khi ký hợp đồng, đọc hết mục「职业病危害告知」từng điều một. Báo cáo mỗi lần khám sức khoẻ thì giữ lại một bản. Chỗ khó là phải hỏi thẳng mặt và xin được hồ sơ ngay tại chỗ. -- Nói thẳng: Khi ký hợp đồng, đơn vị phải ghi thật vào hợp đồng vị trí này có nguy cơ nghề nghiệp gì, bảo vệ bằng cách nào, được hưởng ưu đãi gì. Giấu bạn thì bạn có quyền từ chối làm việc đó, công ty không được vì thế mà sa thải bạn. Ba lần khám sức khoẻ nghề nghiệp gồm trước khi nhận việc, trong thời gian làm việc, và khi rời khỏi công việc, do đơn vị sắp xếp và trả tiền. Không làm khám sức khoẻ khi rời việc thì đơn vị không được chấm dứt hợp đồng với bạn. -- Lợi ích: Luật phòng và chữa bệnh nghề nghiệp quy định, lúc ký hợp đồng lao động, đơn vị phải nói rõ bốn điều: làm việc này có thể sinh ra nguy cơ nghề nghiệp nào, để lại hậu quả gì, phòng ngừa bằng cách nào, được hưởng ưu đãi gì. Nguyên văn điều luật là「如实告知劳动者，并在劳动合同中写明，不得隐瞒或者欺骗」, tức là phải nói thật với người lao động và ghi vào hợp đồng lao động, được giấu hay lừa dối. Sau khi ký hợp đồng mà lại điều chuyển bạn sang việc có hại chưa nói trong hợp đồng thì phải nói thật lại, và bàn với bạn để sửa điều khoản hợp đồng. Đơn vị không làm thì「người lao động có quyền từ chối làm công việc có nguy cơ nghề nghiệp, đơn vị sử dụng lao động không được vì thế mà chấm dứt hợp đồng lao động đã ký với người lao động」. Khám sức khoẻ nghề nghiệp có ba lần: trước khi nhận việc, trong thời gian làm việc, và khi rời việc. 「Chi phí khám sức khoẻ nghề nghiệp do đơn vị sử dụng lao động chịu」, tiền do đơn vị đó trả, kết quả phải báo bằng văn bản cho chính người đó. Người chưa làm khám trước khi nhận việc thì đơn vị không được sắp xếp làm công việc có tiếp xúc với nguy cơ nghề nghiệp. Người vốn không hợp với loại việc đó (có chống chỉ định nghề nghiệp) cũng không được sắp xếp làm các công việc bị chống chỉ định đó. Khám ra tổn thương sức khoẻ có liên quan đến công việc này thì phải điều người đó ra khỏi vị trí cũ và sắp xếp lại cho tử tế. 「Đối với người lao động chưa làm khám sức khoẻ nghề nghiệp trước khi rời việc thì không được chấm dứt hoặc kết thúc hợp đồng lao động đã ký với người đó」. Khi nghỉ việc thì bạn「có quyền yêu cầu bản sao hồ sơ theo dõi sức khoẻ nghề nghiệp của mình, đơn vị sử dụng lao động phải cung cấp thật và miễn phí, và đóng dấu lên bản sao đó」. Hồ sơ theo dõi sức khoẻ nghề nghiệp đó, đơn vị phải sao chép thật một bản miễn phí cho bạn, rồi đóng dấu lên bản sao (toàn quốc) +- Chi phí: Không tốn tiền. Lúc ký hợp đồng thì đọc hết từng dòng phần「thông báo nguy cơ bệnh nghề nghiệp」. Mỗi kết quả khám đều phải giữ một bản lại. Chỗ khó là phải hỏi trực tiếp và đòi được hồ sơ ngay tại chỗ. +- Nói thẳng: Lúc ký hợp đồng, đơn vị phải ghi thật vào hợp đồng vị trí đó có nguy cơ bệnh nghề nghiệp gì, phòng ngừa ra sao, được hưởng gì. Nếu giấu bạn thì bạn có quyền không nhận việc đó, đơn vị không được vì thế mà sa thải bạn. Ba lần khám sức khoẻ nghề nghiệp: trước khi vào việc, trong thời gian làm việc và lúc rời việc, đều do đơn vị sắp xếp và bỏ tiền. Không làm khám lúc rời việc thì đơn vị không được chấm dứt hợp đồng với bạn. +- Lợi ích: Luật phòng bệnh nghề nghiệp quy định, lúc ký hợp đồng lao động, đơn vị phải nói rõ bốn việc: làm việc đó có thể gây ra bệnh nghề nghiệp nào, để lại hậu quả gì, phòng ngừa ra sao, được hưởng gì. Nguyên văn điều luật là「如实告知劳动者，并在劳动合同中写明，不得隐瞒或者欺骗」. Sau khi đã ký hợp đồng mà lại điều chuyển bạn sang việc nguy hại chưa có trong hợp đồng thì phải thông báo thật lại, và bàn với bạn để sửa điều khoản hợp đồng. Đơn vị không làm được thì「劳动者有权拒绝从事存在职业病危害的作业，用人单位不得因此解除与劳动者所订立的劳动合同」. Tổng cộng ba lần khám: trước khi vào việc, trong thời gian làm việc, lúc rời việc.「职业健康检查费用由用人单位承担」, tiền do đơn vị chi, kết quả phải báo bằng văn bản cho chính người đó. Người chưa khám trước khi vào việc thì đơn vị không được bố trí làm việc có tiếp xúc với nguy cơ bệnh nghề nghiệp. Người vốn không phù hợp thể chất với kiểu việc đó (có chống chỉ định nghề nghiệp) thì cũng không được bố trí làm đúng những việc bị chống chỉ định ấy. Phát hiện tổn hại sức khoẻ liên quan tới công việc này thì phải điều chuyển người đó khỏi vị trí cũ và sắp xếp lại cho tử tế.**「对未进行离岗前职业健康检查的劳动者不得解除或者终止与其订立的劳动合同」**. Lúc nghỉ việc bạn「有权索取本人职业健康监护档案复印件，用人单位应当如实、无偿提供，并在所提供的复印件上签章」. Hồ sơ theo dõi sức khoẻ nghề nghiệp đó, đơn vị phải sao chép nguyên văn một bản cho bạn miễn phí, rồi đóng dấu lên bản sao (toàn quốc) - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 职业病防治法（2018 年修正，第三十三、三十五、三十六条）. （Chính phủ thành phố Bắc Kinh đăng lại） -- Ghi chú: Khi phỏng vấn, hỏi ba câu là đo được đại thể chỗ này có nghiêm không: vị trí này có nguy cơ nghề nghiệp không, hợp đồng có ghi không, tiền khám trước khi nhận việc do ai trả. Không trả lời được, hoặc trả lời「làm thử đã」, thì tự rủi ro. Khám sức khoẻ khi rời việc là cái dễ bị bỏ qua nhất: không làm thì đơn vị không được chấm dứt hoặc kết thúc hợp đồng lao động với bạn. Chính cuộc khám này sau này còn là hồ sơ then chốt để xin chẩn đoán bệnh nghề nghiệp, bản sao hồ sơ cũng vậy, phải xin được trước khi đi. Bản thân bệnh nghề nghiệp thì áp dụng như tai nạn lao động, các ưu đãi xem từ mục 12 trở đi. - -### 11. Tổn thương do bụi, tiếng ồn, độc chất hoá học thì không đảo ngược: đơn vị phải cấp phương tiện bảo vệ, công việc không có bảo vệ thì được từ chối +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). Luật phòng bệnh nghề nghiệp, 职业病防治法（2018 年修正，第三十三、三十五、三十六条）. （Chính quyền thành phố Bắc Kinh (北京市人民政府) đăng lại） +- Ghi chú: Lúc phỏng vấn mà hỏi ba câu thì đoán được đại thể chỗ này có nghiêm túc không: vị trí này có nguy cơ bệnh nghề nghiệp không, hợp đồng có ghi không, tiền khám trước khi vào việc do ai trả. Trả lời không nổi hoặc nói「làm trước đã」thì rủi ro tự tính lấy. Khám lúc rời việc là cái dễ bị bỏ sót nhất: không khám lúc rời việc thì đơn vị không được chấm dứt hoặc kết thúc hợp đồng lao động với bạn. Kết quả khám này sau này còn là hồ sơ then chốt để xin chẩn đoán bệnh nghề nghiệp, bản sao hồ sơ theo dõi cũng vậy, phải lấy được trước khi rời đi. Bản thân bệnh nghề nghiệp thì đi theo lối tai nạn lao động, quyền lợi xem mục 11 của phần này trở đi. +### 10. Tổn thương do bụi, tiếng ồn, chất độc hoá học thì không hồi phục được: đơn vị phải cấp đồ dùng bảo vệ, công việc không có biện pháp bảo vệ thì được từ chối -- Chi phí: Không tốn tiền. Chỗ khó là phải đeo khẩu trang chống bụi hạt, nút bịt tai, kính bảo hộ suốt thời gian, còn phải từ chối công việc không có bảo vệ, cái giá là đồng nghiệp có thể nói bạn「nhiều chuyện」. -- Nói thẳng: Bệnh bụi phổi, điếc do tiếng ồn, bỏng mắt do hoá chất đều là bệnh nghề nghiệp theo luật, phổi và thính lực hỏng thì không chữa lại được. Bạn có quyền biết vị trí này có nguy cơ gì, và có quyền đòi đơn vị cấp thiết bị và phương tiện bảo vệ. Công việc không có bảo vệ thì bạn không làm cũng được. Đơn vị vì thế mà giảm lương hoặc sa thải bạn thì đều không có hiệu lực. -- Lợi ích: Từ ngày 1 tháng 8 năm 2025, bệnh nghề nghiệp theo luật gồm 12 nhóm lớn 135 loại, trong đó có 4 điều khoản mở. Nhóm đứng đầu là bệnh phổi bụi nghề nghiệp. Trong đó có bệnh phổi do silic, do bụi than, do bụi than chì, do bụi than đen, do bụi asbest, do bụi đá phiến. Rồi đến bệnh phổi do bụi xi măng, do bụi mica, do bụi gốm, do bụi nhôm, do hàn điện, do bụi luyện kim, cộng thêm một điều khoản mở. Bệnh tai mũi họng và miệng nghề nghiệp có điếc do tiếng ồn và điếc do sốc. Bệnh mắt nghề nghiệp có bỏng mắt do hoá chất và viêm mắt do ánh điện. Những tổn thương này không đảo ngược, không có cách chữa nào lấy lại phổi và thính lực. Luật đưa ra bốn quyền có thể dùng ngay tại chỗ. Một là biết nơi làm việc có những nguy cơ nghề nghiệp nào, gây hậu quả gì, phải áp dụng biện pháp bảo vệ nào. Hai là yêu cầu đơn vị cấp thiết bị bảo vệ và phương tiện bảo vệ cá nhân đạt chuẩn, và cải thiện điều kiện làm việc. Ba là phê bình, kiện cáo và tố cáo những cách làm vi phạm quy định phòng chữa bệnh nghề nghiệp, cũng như cách làm nguy hại tới tính mạng và sức khoẻ. Bốn là「từ chối chỉ đạo sai quy định và từ chối bị ép làm công việc không có biện pháp phòng chữa bệnh nghề nghiệp」, chỉ đạo sai quy định và công việc không có bảo vệ thì bạn không làm cũng được. Và「nếu vì người lao động thực hiện hợp pháp quyền lợi của mình mà giảm lương, giảm ưu đãi hoặc chấm dứt, kết thúc hợp đồng lao động đã ký với họ thì hành vi đó vô hiệu」. Dùng bốn quyền này mà bị giảm lương, giảm ưu đãi hoặc sa thải thì những cách đó không có hiệu lực (toàn quốc) +- Chi phí: Không tốn tiền. Chỗ khó là phải đeo khẩu trang chống hạt, nút bịt tai và kính bảo hộ suốt thời gian, còn phải từ chối việc không có bảo vệ, cái giá là có thể bị đồng nghiệp bảo「nhiều chuyện quá」. +- Nói thẳng: Bệnh phổi bụi, điếc do tiếng ồn và bỏng mắt do hoá chất đều là bệnh nghề nghiệp theo luật, phổ và thính giác hỏng thì không chữa lại được. Bạn có quyền biết vị trí đó có nguy cơ gì, có quyền yêu cầu đơn vị cấp biện pháp bảo vệ và đồ dùng bảo vệ. Việc không có bảo vệ thì bạn không nhận. Đơn vị vì thế mà giảm lương hoặc sa thải bạn thì đều không có hiệu lực. +- Lợi ích: Từ ngày 1 tháng 8 năm 2025, bệnh nghề nghiệp theo luật gồm 12 nhóm lớn, 135 loại, trong đó có 4 điều khoản để mở. Nhóm đứng đầu là bệnh phổi bụi do nghề nghiệp. Trong đó có bệnh phổi bụi do silicat, bệnh phổi bụi của thợ mỏ than, bệnh phổi bụi do bột than chì, bệnh phổi bụi do bột than đen, bệnh phổi bụi do bụi asbestos, bệnh phổi bụi do bột phấn. Rồi đến bệnh phổi bụi do xi măng, bệnh phổi bụi do mica, bệnh phổi bụi của thợ gốm, bệnh phổi bụi do bột nhôm, bệnh phổi bụi của thợ hàn, bệnh phổi bụi của thợ đúc, cộng thêm một điều khoản để mở. Nhóm bệnh tai mũi họng và răng do nghề nghiệp có điếc do tiếng ồn và điếc do âm thanh nổ. Nhóm bệnh mắt do nghề nghiệp có bỏng mắt do hoá chất và viêm mắt do ánh điện. Những tổn thương này không hồi phục được, không có cách điều trị nào lấy lại phổi và thính giác. Luật đưa ra bốn quyền dùng được ngay tại chỗ. Một là biết trong nơi làm việc này có những nguy cơ bệnh nghề nghiệp gì, gây hậu quả ra sao, phải áp dụng biện pháp bảo vệ nào. Hai là yêu cầu đơn vị cấp biện pháp bảo vệ đạt chuẩn và phương tiện bảo vệ cá nhân, cải thiện điều kiện làm việc. Ba là phê bình, báo việc và tố cáo những làm việc vi phạm quy luật pháp lý phòng bệnh nghề nghiệp, cũng như những làm việc nguy hại tới sức khoẻ và tính mạng. Bốn là「拒绝违章指挥和强令进行没有职业病防护措施的作业」, chỉ đạo sai luật và việc không có bảo vệ thì bạn không nhận. Và「因劳动者依法行使正当权利而降低其工资、福利等待遇或者解除、终止与其订立的劳动合同的，其行为无效」. Dùng mấy quyền đó mà công ty giảm lương, giảm quyền lợi hoặc sa thải bạn thì những việc đó không có hiệu lực (toàn quốc) - Mức bằng chứng: A -- Nguồn: Ủy ban Quốc gia Y tế, Bộ Nhân sự và Xã hội, Cục Quốc gia Kiểm soát và Phòng ngừa bệnh, Tổng Liên đoàn Lao động (2024). 职业病分类和目录（2025 年 8 月 1 日起实施）. ；cách tính 12 nhóm lớn 135 loại trong danh mục mới xem Ủy ban Quốc gia Y tế (2024). 新版目录将职业病调整为 12 大类 135 种. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 职业病防治法（第三十九条）. （Chính phủ thành phố Bắc Kinh đăng lại） -- Ghi chú: Phải chọn đúng khẩu trang, khẩu trang chống bụi hạt mới chặn được bụi, khẩu trang vải thường và khẩu trang y tế dùng một lần không chặn được. Chỗ làm việc có tiếng ồn mà đơn vị không phát nút bịt tai thì tự mua, vài tệ một cặp, mất thính lực thì không bù lại được. 「Làm hai năm kiếm đủ tiền rồi đi」không đúng với bệnh bụi phổi. Bệnh phổi do silic có thể khởi phát nhiều năm sau khi ngừng làm việc với bụi, và vẫn tiếp tục nặng thêm. Nên khám sức khoẻ khi rời việc lại càng quan trọng (xem mục 10 của phần này, ba lần khám sức khoẻ nghề nghiệp do đơn vị sắp xếp và trả tiền). Danh mục mới còn thêm hai nhóm. Một là bệnh cơ xương khớp nghề nghiệp, trong đó có hội chứng ống cổ tay, chỉ áp dụng cho công nhân sản xuất dùng cổ tay lặp lại lâu hoặc dùng sức. Hai là rối loạn tâm thần và hành vi nghề nghiệp, trong đó có rối loạn căng thẳng sau chấn thương, chỉ áp dụng cho công an nhân dân, nhân viên y tế, lực lượng cứu hộ và các lực lượng ứng cứu khẩn cấp có tham gia xử lý sự cố. Xử lý tại chỗ khi hoá chất bắn vào người xem mục 21 ở phần 13. - -### 12. Bị thương lúc làm việc, hay bị xe đâm trên đường đi làm: việc đầu tiên là xin xác định tai nạn lao động, đơn vị không báo thì bạn tự báo +- Nguồn: Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委), Bộ Nhân sự và Xã hội (人力资源社会保障部), Văn phòng Tổng hợp Cục Kiểm soát bệnh quốc gia (国家疾控局), Tổng Liên đoàn Công nhân toàn quốc (全国总工会) (2024). Phân loại và danh mục bệnh nghề nghiệp, 职业病分类和目录（2025 年 8 月 1 日起实施）. ；Cách tính 12 nhóm lớn 135 loại trong danh mục mới xem Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委) (2024). Danh mục mới điều chỉnh bệnh nghề nghiệp thành 12 nhóm lớn 135 loại. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). Luật phòng bệnh nghề nghiệp, 职业病防治法（第三十九条）. （Chính quyền thành phố Bắc Kinh (北京市人民政府) đăng lại） +- Ghi chú: Khẩu trang phải chọn đúng, chỉ khẩu trang chống hạt mới chắn bụi, khẩu trang vải thường và khẩu trang y tế dùng một lần không chắn được. Ở chỗ có tiếng ồn mà đơn vị không phát nút bịt tai thì tự mua, vài tệ một cặp, mất thính rồi không bù lại được. 「Làm hai năm tích đủ tiền rồi nghỉ」không có hiệu lực với bệnh phổi bụi. Bệnh phổi bụi do silicat có thể mấy năm sau khi rời việc có bụi mới bị, và còn nặng dần. Cho nên khám lúc rời việc càng quan trọng (xem mục 9, ba lần khám sức khoẻ nghề nghiệp do đơn vị sắp xếp và bỏ tiền). Danh mục mới còn thêm hai nhóm. Một là bệnh cơ xương khớp do nghề nghiệp, trong đó là hội chứng ống cổ tay, chỉ dành cho công nhân sản xuất phải dùng cổ tay lặp lại lâu hoặc dùng lực nặng. Nhóm kia là rối loạn tâm thần và hành vi do nghề nghiệp, trong đó là rối loạn stress sau chấn thương, chỉ dành cho công an nhân dân, nhân viên y tế, lực lượng cứu hỏa và cứu nạn có khoản tham gia xử lý tình huống khẩn cấp. Xử lý tại chỗ khi hoá chất bắn vào người xem mục 21 ở phần 13. +### 11. Bị thương lúc làm việc, hay bị xe đâm trên đường đi làm về: việc đầu tiên là xin xác nhận tai nạn lao động, đơn vị không báo thì bạn tự báo -- Chi phí: Không tốn tiền, chỉ mất một chuyến tới cơ quan bảo hiểm xã hội. Mang theo ba loại giấy tờ: đơn xin xác định tai nạn lao động, giấy chứng minh quan hệ lao động, giấy chẩn đoán y tế. -- Nói thẳng: Bị thương lúc làm việc là tai nạn lao động. Bị xe đâm trên đường đi làm mà phần lỗi chính không ở bạn thì cũng là tai nạn lao động. Nhưng không xin xác định tai nạn lao động thì tiền khám chữa, lương được trả trong thời gian điều trị, và tiền trợ cấp thương tật đều mất trắng. Đơn vị phải báo trong 30 ngày. Đơn vị không báo thì bạn, hoặc người nhà và công đoàn, có thể tự báo với cơ quan bảo hiểm xã hội trong vòng một năm. -- Lợi ích: Không xin xác định tai nạn lao động thì tiền khám chữa, lương được trả trong thời gian điều trị (lương thời gian ngừng việc được trả), và tiền trợ cấp thương tật đều mất trắng. Đơn vị quá 30 ngày không báo thì trong thời gian đó, các khoản chi phí về ưu đãi tai nạn lao động phải trả thì chính đơn vị đó chịu +- Chi phí: Không tốn tiền, chạy một lượt tới cơ quan quản lý bảo hiểm xã hội. Cần mang ba loại hồ sơ: đơn xin xác nhận tai nạn lao động, chứng cứ quan hệ lao động, giấy chứng nhận chẩn đoán y tế. +- Nói thẳng: Bị thương lúc đi làm thì tính là tai nạn lao động. Bị xe đâm trên đường đi làm về mà lỗi chính không thuộc về bạn thì cũng tính là tai nạn lao động. Nhưng không xin xác nhận tai nạn lao động thì tiền khám chữa, lương vẫn được trả trong thời gian nghỉ dưỡng, tiền trợ cấp tàn tật đều mất trắng. Đơn vị phải báo trong vòng 30 ngày. Đơn vị không báo thì bạn, người nhà hoặc công đoàn có thể tự báo tới cơ quan quản lý bảo hiểm xã hội trong vòng một năm. +- Lợi ích: Không xin xác nhận tai nạn lao động thì tiền khám chữa, lương vẫn được trả trong thời gian nghỉ dưỡng (lương thời gian nghỉ giữ lương), tiền trợ cấp tàn tật đều mất trắng. Đơn vị để quá 30 ngày mà không báo thì trong khoảng thời gian đó, các khoản chi phí quyền lợi tai nạn lao động phải chi trả thì đơn vị tự bỏ ra - Mức bằng chứng: A -- Nguồn: Ủy ban Nhà nước (国务院) (2010 修订). 工伤保险条例（国务院令第 586 号）第十四条列了七种「应当认定为工伤」，含「（六）在上下班途中，受到非本人主要责任的交通事故或者城市轨道交通、客运轮渡、火车事故伤害的」；第十七条「所在单位应当自事故伤害发生之日或者被诊断、鉴定为职业病之日起30日内，向统筹地区社会保险行政部门提出工伤认定申请」「用人单位未按前款规定提出工伤认定申请的，工伤职工或者其近亲属、工会组织在事故伤害发生之日或者被诊断、鉴定为职业病之日起1年内，可以直接向用人单位所在地统筹地区社会保险行政部门提出工伤认定申请」「用人单位未在本条第一款规定的时限内提交工伤认定申请，在此期间发生符合本条例规定的工伤待遇等有关费用由该用人单位负担」；第十九条「职工或者其近亲属认为是工伤，用人单位不认为是工伤的，由用人单位承担举证责任」；第二十条「自受理工伤认定申请之日起60日内作出工伤认定的决定」. -- Ghi chú: Một năm là thời hạn cứng, quá hạn là không còn đường điều. Đơn vị nói không phải tai nạn lao động thì đơn vị phải đưa bằng chứng. Nhưng bạn cũng phải giữ hồ sơ: ảnh chụp ngày xảy ra, thông tin liên lạc của đồng nghiệp, hồ sơ khám chữa, bản chấm công. Điều mười sáu của nghị định liệt kê ba trường hợp không công nhận: cố ý phạm tội, say rượu hoặc nghiện ma túy, tự làm hại bản thân hoặc tự sát. Điều mười lăm lại có「视同工伤」, tức là coi như tai nạn lao động: chết vì bệnh phát đột ngột trong giờ làm và tại chỗ làm, hoặc chết sau cấp cứu không cứu được trong 48 giờ. - -### 13. Đừng tin「leo tới bàn giấy là được coi là tai nạn lao động」: thấy khó chịu đột ngột thì gọi 120 trước, đừng chạy tới chấm công +- Nguồn: Hội đồng Nhà nước (国务院) (2010, sửa đổi). Quy định về bảo hiểm tai nạn lao động, 工伤保险条例（国务院令第 586 号）điều mười bốn liệt kê bảy trường hợp「应当认定为工伤」, trong đó có「（六）在上下班途中，受到非本人主要责任的交通事故或者城市轨道交通、客运轮渡、火车事故伤害的」；điều mười bảy「所在单位应当自事故伤害发生之日或者被诊断、鉴定为职业病之日起30日内，向统筹地区社会保险行政部门提出工伤认定申请」「用人单位未按前款规定提出工伤认定申请的，工伤职工或者其近亲属、工会组织在事故伤害发生之日或者被诊断、鉴定为职业病之日起1年内，可以直接向用人单位所在地统筹地区社会保险行政部门提出工伤认定申请」「用人单位未在本条第一款规定的时限内提交工伤认定申请，在此期间发生符合本条例规定的工伤待遇等有关费用由该用人单位负担」；điều mười chín「职工或者其近亲属认为是工伤，用人单位不认为是工伤的，由用人单位承担举证责任」；điều hai mươi「自受理工伤认定申请之日起60日内作出工伤认定的决定」；điều năm mươi lăm, với「工伤认定申请不予受理的决定不服的」「对工伤认定结论不服的」thì「可以依法申请行政复议，也可以依法向人民法院提起行政诉讼」. ；Tòa án nhân dân tối cao (最高人民法院) (2014). Quy định về xét lý một số vấn đề trong vụ án hành chính về bảo hiểm tai nạn lao động, 关于审理工伤保险行政案件若干问题的规定（法释〔2014〕9 号）điều bảy:「由于不属于职工或者其近亲属自身原因超过工伤认定申请期限的，被耽误的时间不计算在工伤认定申请期限内」, các trường hợp liệt kê là「（一）不可抗力；（二）人身自由受到限制；（三）属于用人单位原因；（四）社会保险行政部门登记制度不完善；（五）当事人对是否存在劳动关系申请仲裁、提起民事诉讼」. +- Ghi chú: Một năm là thời hạn thật, đừng để trôi. Nhưng thời gian bị trễ không phải do bạn thì không tính vào một năm đó: bất khả kháng, bị hạn chế tự do cá nhân, lý do của đơn vị, hệ thống đăng ký của cơ quan bảo hiểm xã hội chưa hoàn chỉnh, đi hoà giải lao động hoặc kiện tụng dân sự để xác nhận có quan hệ lao động hay không. Nên đơn vị kéo dài không báo, kéo đến quá hạn thì bạn vẫn nộp đơn, mang theo chứng cứ đơn vị đã trễ. Không được tiếp nhận, hoặc không chấp nhận kết luận xác nhận, thì bạn có thể xin xét lại hành chính, cũng có thể kiện thẳng ra tòa. Đơn vị nói không phải tai nạn lao động thì đơn vị phải đưa ra chứng cứ. Nhưng bạn cũng phải giữ lại hồ sơ: ảnh chụp ngày xảy ra việc, thông tin liên lạc của đồng nghiệp, hồ sơ khám chữa, bản chấm công. Điều mười sáu của quy định liệt kê ba trường hợp không công nhận: cố ý phạm tội, say rượu hoặc nghiện ma tuý, tự hại hoặc tự sát. Điều mười lăm còn có「视同工伤」: chết vì bệnh phát đột ngột trong giờ làm và tại chỗ làm việc, hoặc chết sau cấp cứu không cứu được trong vòng 48 giờ. +### 12. Đừng tin「leo tới bàn giấy là được coi là tai nạn lao động」: thấy khó chịu đột ngột thì gọi 120 trước, đừng chạy tới chấm công - Chi phí: Không tốn tiền. - Nói thẳng: Khó chịu đột ngột mà chạy tới công ty không giúp bạn nhận thêm đồng nào. Coi như tai nạn lao động đòi hỏi người đó phát bệnh trong giờ làm việc và tại chỗ làm việc, phát bệnh trên đường chưa tới nơi thì vốn không tính. Cấp cứu quá 48 giờ thì ngược lại cũng không tính. Mấy chục phút chờ để khớp điều kiện là đổi lấy tỷ lệ sống sót của chính bạn. @@ -119,8 +101,7 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Mức bằng chứng: A - Nguồn: Điều mười lăm, khoản một, mục (一) của Quy định về bảo hiểm tai nạn lao động:「在工作时间和工作岗位，突发疾病死亡或者在48小时之内经抢救无效死亡的」视同工伤；Điều mười bốn, mục (六) chỉ xếp「受到非本人主要责任的交通事故或者城市轨道交通、客运轮渡、火车事故伤害的」khi đi trên đường là tai nạn lao động, phát bệnh đột ngột giữa đường thì không nằm trong đó. - Ghi chú: Hai điều kiện phải có cùng lúc: đang trong giờ làm, và đang ở chỗ làm việc. Phát bệnh trên đường chưa tới nơi thì không áp dụng. Phát bệnh lúc đã đến chỗ rồi được đưa về nhà hoặc đưa đi bệnh viện, chết sau cấp cứu không cứu được trong 48 giờ thì vẫn có thể áp dụng. Người ta xem lúc phát bệnh người đó ở đâu, không xem cuối cùng chết ở đâu. Video ngắn giải thích điều này thành「leo tới công ty là có tiền」, hướng ngược hẳn. Cái quyết định kết cục là mấy chục phút sau khi phát bệnh có đi bệnh viện không. Ngoài ra 48 giờ là ranh giới cứng, cấp cứu quá 48 giờ thì ngược lại không tính. - -### 14. Đơn vị không đóng bảo hiểm tai nạn lao động cho bạn thì ưu đãi tai nạn lao động vẫn có, và do đơn vị trả đủ theo đúng tiêu chuẩn +### 13. Đơn vị không đóng bảo hiểm tai nạn lao động cho bạn thì ưu đãi tai nạn lao động vẫn có, và do đơn vị trả đủ theo đúng tiêu chuẩn - Chi phí: Không tốn tiền. Đơn vị không công nhận thì phải đi qua hoà giải lao động, thậm chí ra tòa, ít nhất mấy tháng. Chỗ khó là mấy tháng đó phải theo suốt. - Nói thẳng: Đơn vị không đóng bảo hiểm tai nạn lao động cho bạn thì ưu đãi tai nạn lao động vẫn có. Chỉ khác ở chỗ đổi thành đơn vị tự trả đủ các khoản và theo đúng tiêu chuẩn, không bớt một phần nào. Không tham gia bảo hiểm là đơn vị bị phạt thêm: bị buộc đóng bù, bị tính phí trễ nộp theo ngày là năm phần mười nghìn, quá hạn còn bị phạt bằng một đến ba lần số tiền chậm nộp. 「Công ty không mua bảo hiểm tai nạn lao động nên không đền được」là lời nói sai. @@ -128,7 +109,7 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Mức bằng chứng: A - Nguồn: Điều sáu mươi hai, khoản hai của Quy định về bảo hiểm tai nạn lao động:「依照本条例规定应当参加工伤保险而未参加工伤保险的用人单位职工发生工伤的，由该用人单位按照本条例规定的工伤保险待遇项目和标准支付费用。」Khoản một cùng điều: bị buộc phải tham gia và đóng bù trong thời hạn,「自欠缴之日起，按日加收万分之五的滞纳金；逾期仍不缴纳的，处欠缴数额1倍以上3倍以下的罚款」. - Ghi chú: 「Công ty không mua bảo hiểm tai nạn lao động cho bạn nên không đền được」là câu nói sai phổ biến nhất trong loại việc này. Không tham gia bảo hiểm chỉ khiến đơn vị bị phạt thêm, ưu đãi của bạn không giảm. -### 15. Khi tình trạng thương tích đã ổn định thì đi đánh giá năng lực lao động, hạng khuyết tật quy đổi thẳng thành tiền +### 14. Khi tình trạng thương tích đã ổn định thì đi đánh giá năng lực lao động, hạng khuyết tật quy đổi thẳng thành tiền - Chi phí: Không tốn tiền. Đi một lượt quy trình đánh giá năng lực lao động, trước sau phải chờ một thời gian - Nói thẳng: Sau khi tình trạng thương tích ổn định thì đi đánh giá năng lực lao động, hạng đánh giá quy đổi thẳng thành tiền. Tiền trợ cấp khuyết tật một lần tính theo lương của chính bạn: hạng năm 18 tháng, hạng sáu 16 tháng, hạng bảy 13 tháng, hạng tám 11 tháng, hạng chín 9 tháng, hạng mười 7 tháng. Trước khi có kết quả đánh giá, đừng ký thoả thuận 「giải quyết một lần」 @@ -137,7 +118,7 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Nguồn: Quy định về bảo hiểm tai nạn lao động, 工伤保险条例 điều hai mươi mốt: 「经治疗伤情相对稳定后存在残疾、影响劳动能力的，应当进行劳动能力鉴定」; điều hai mươi hai: 「劳动功能障碍分为十个伤残等级，最重的为一级，最轻的为十级」; điều ba mươi sáu, chế độ cho hạng năm và hạng sáu, gồm tiền trợ cấp khuyết tật một lần 18/16 tháng lương của chính bạn và trợ cấp khuyết tật hằng tháng bằng 70%/60% lương của chính bạn; điều ba mươi bảy, chế độ cho hạng bảy đến hạng mười, gồm 13/11/9/7 tháng lương của chính bạn. - Ghi chú: Hạng năm, hạng sáu giữ quan hệ lao động, người đó vẫn là công nhân viên của đơn vị, hằng tháng nhận trợ cấp khuyết tật. Người ở hạng bảy đến hạng mười, khi hợp đồng hết hạn hoặc khi chính họ đề nghị chấm dứt thì có thêm hai khoản tiền: tiền trợ cấp y tế tai nạn lao động một lần do quỹ chi trả, tiền trợ cấp việc làm cho người khuyết tật một lần do đơn vị chi trả, mức bao nhiêu do tỉnh quyết định. Trước khi có kết quả đánh giá thì đừng ký thoả thuận 「giải quyết một lần」 -### 16. Phải phân biệt ba khoản tiền khi chết vì tai nạn lao động: tiền trợ cấp mai táng, tiền trợ cấp hưu trí người được nuôi dưỡng, tiền trợ cấp một lần khi chết vì tai nạn lao động +### 15. Phải phân biệt ba khoản tiền khi chết vì tai nạn lao động: tiền trợ cấp mai táng, tiền trợ cấp hưu trí người được nuôi dưỡng, tiền trợ cấp một lần khi chết vì tai nạn lao động - Chi phí: Không tốn tiền, làm một bộ thủ tục giấy tờ - Nói thẳng: Chết vì tai nạn lao động là ba khoản tiền, đừng chỉ thương lượng một khoản. Tiền trợ cấp mai táng bằng 6 tháng lương bình quân theo tháng của công nhân viên năm trước tại địa phương. Tiền trợ cấp hưu trí người được nuôi dưỡng trả theo tháng theo lương của người đã mất, vợ hoặc chồng mỗi tháng 40%, người thân khác mỗi người mỗi tháng 30%. Tiền trợ cấp một lần khi chết vì tai nạn lao động bằng 20 lần thu nhập bình quân của dân cư đô thị cả nước năm trước, cả nước chỉ có một con số này @@ -145,8 +126,7 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Mức bằng chứng: A - Nguồn: Quy định về bảo hiểm tai nạn lao động, 工伤保险条例第三十九条: 「（一）丧葬补助金为6个月的统筹地区上年度职工月平均工资；（二）供养亲属抚恤金按照职工本人工资的一定比例发给由因工死亡职工生前提供主要生活来源、无劳动能力的亲属。标准为：配偶每月40%，其他亲属每人每月30%，孤寡老人或者孤儿每人每月在上述标准的基础上增加10%……（三）一次性工亡补助金标准为上一年度全国城镇居民人均可支配收入的20倍。」. ；Cục Thống kê Quốc gia (国家统计局) (2026). Tình hình thu nhập và chi tiêu của dân cư năm 2025, 2025 年居民收入和消费支出情况. - Ghi chú: Tiền trợ cấp một lần khi chết vì tai nạn lao động là mức thống nhất cả nước, số tiền cập nhật hằng năm theo số liệu Cục Thống kê Quốc gia công bố. Khi thương lượng bồi thường, không chấp nhận lý do 「tiêu chuẩn địa phương của chỗ chúng tôi thấp」. - -### 17. Ở công ty lâu ngày bị bắt nạt, mắng, gây khó, đừng gồng chịu: ghi lại để có chứng cứ, rồi tuỳ tính chất mà khiếu nại, báo công an hoặc đi hoà giải +### 16. Ở công ty lâu ngày bị bắt nạt, mắng, gây khó, đừng gồng chịu: ghi lại để có chứng cứ, rồi tuỳ tính chất mà khiếu nại, báo công an hoặc đi hoà giải - Chi phí: Không tốn tiền. Phải tốn thời gian ghi chép: ngày tháng, người có mặt, nguyên văn lời của đối phương, ảnh chụp màn hình trò chuyện và giữ bản gốc email. Đi hoà giải lao động, ra tòa phải mất mấy tháng. Đổi việc thì cái giá cao nhất, đi hay không thì bạn tự tính - Nói thẳng: Người bị bắt nạt ở công ty thì về sau nguy cơ mắc bệnh tim, đột quỵ cao hơn khoảng sáu phần mười, bị bắt nạt càng nhiều thì nguy cơ càng cao. Người bị bắt nạt thì xác suất nảy sinh ý nghĩ tự sát vài năm sau cao gấp khoảng hai lần người khác. Chịu mắng, chịu khí ở nơi làm việc không phải chuyện nhỏ mà nhịn một chút là qua @@ -154,3 +134,19 @@ Phần này tính ra tiền: tiền tính thế nào, chữ trong giấy ký th - Mức bằng chứng: B - Nguồn: Xu T 等 (2019). Workplace bullying and workplace violence as risk factors for cardiovascular disease: a multi-cohort study. European Heart Journal. ；Nielsen MB, Nielsen GH, Notelaers G, Einarsen S (2015). Workplace Bullying and Suicidal Ideation: A 3-Wave Longitudinal Norwegian Study. American Journal of Public Health. ；Ủy ban Thường vụ Quốc hội nhân dân (全国人大常委会) (2007). Bộ luật Hợp đồng lao động, 劳动合同法（第三十八条第二款、第四十六条、第八十八条）. ；Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự, 民法典（第一千零一十条）, bản toàn văn do Viện kiểm sát nhân dân tối cao đăng lại ；Ủy ban Thường vụ Quốc hội nhân dân (全国人大常委会) (2022, sửa đổi). Luật bảo vệ quyền lợi phụ nữ, 妇女权益保障法（第二十五条）. ；Ủy ban Thường vụ Quốc hội nhân dân (全国人大常委会) (2007). Luật hoà giải và trọng tài tranh chấp lao động, 劳动争议调解仲裁法（第四十三条）, bản toàn văn do Sở Nhân sự và Xã hội thành phố An Khang đăng lại - Ghi chú: Lý do chấm B: các con số sức khoẻ đến từ theo dõi ghi chép, không phải thử nghiệm chia nhóm, và 「bị bắt nạt」 là do bản thân người bị hại tự khai. Có bốn bước làm được ngay. Bước một, từ hôm nay ghi chép lại, xin phép ghi âm xem mục 41 ở phần 8 (ghi âm). Bước hai, phản ánh bằng văn bản qua kênh chính thức của đơn vị, giữ lại bằng chứng đã phản ánh. Bước ba, bị đánh, bị nhốt, bị xúc phạm trước đông người thì báo công an. Bước bốn, bị ép nghỉ vì nợ lương, không đóng bảo hiểm xã hội thì làm theo mục 7 của phần này (đừng ký đơn tự nguyện nghỉ việc) và mục 8 của phần này (lưu chứng cứ trước khi nghỉ việc). Cách tính bồi thường kinh tế xem mục 4 của phần này (bị sa thải thì tính rõ N trước). Bị coi thường, bị loại trừ, bị gây khó đơn thuần thì phần lớn không khớp với điều luật nào, đường đi chủ yếu là khiếu nại nội bộ, xin đổi vị trí và đổi việc. Hoà giải lao động tính từ ngày tiếp nhận tối đa 60 ngày, ra tòa phân thủ cấp một tối thiểu 6 tháng, phí luật sư tự bỏ ra. Khi đã không chịu nổi nữa, gọi 12356 trước, xem mục 25 ở phần 1 (12356). Không phản đối việc chỉ vì khó chịu là nghỉ việc, cái giá của việc đổi việc thì bạn tự tính. +### 17. Lúc nghỉ việc thì xin đơn vị một giấy xác nhận nghỉ việc, ghi rõ thời hạn hợp đồng, ngày nghỉ việc, vị trí và số năm đã làm + +- Chi phí: Không tốn tiền. Trước khi đi thì xin trực tiếp, hoặc gửi một tin nhắn bằng văn bản để xin, giữ lại chuyện đã xin. Chỗ khó là quan hệ đã rạn thì vẫn phải mở miệng. Đơn vị kéo dài không đưa thì phải khiếu nại hoặc xin hoà giải lao động, trước sau mấy chục ngày tới mấy tháng. +- Nói thẳng: Dù là bị sa thải, hết hạn hợp đồng hay tự nghỉ việc, đơn vị cũng phải cấp cho bạn một giấy xác nhận nghỉ việc khi bạn rời đi, và chuyển hồ sơ lưu trữ cùng bảo hiểm xã hội trong 15 ngày. Trên giấy phải ghi thời hạn hợp đồng, ngày nghỉ việc, vị trí và số năm bạn làm ở chỗ này. Đăng ký thất nghiệp cần dùng nó. Đơn vị không đưa thì cơ quan lao động sẽ buộc sửa, và phải bồi thường nếu bạn bị thiệt hại. +- Lợi ích: Bộ luật Hợp đồng lao động quy định, đơn vị phải cấp giấy chứng nhận khi chấm dứt hoặc kết thúc hợp đồng lao động. Điều luật không phân biệt ai là người đề xuất, hết hạn hợp đồng không gia hạn, bạn tự nghỉ việc, bị sa thải đều nằm trong đó. Đơn vị còn phải làm thủ tục chuyển hồ sơ lưu trữ và quan hệ bảo hiểm xã hội cho bạn trong 15 ngày. Quy chế thực thi quy định giấy phải ghi bốn thứ: thời hạn hợp đồng lao động, ngày chấm dứt hoặc kết thúc, vị trí làm việc, số năm làm việc tại đơn vị. Quy định về bảo hiểm thất nghiệp quy định, sau khi thất nghiệp thì người lao động phải cầm giấy chứng nhận do đơn vị cấp, đi đăng ký thất nghiệp kịp thời ở cơ quan bảo hiểm xã hội. Tiền trợ cấp thất nghiệp tính từ ngày đăng ký thất nghiệp. Đơn vị không cấp giấy chứng nhận bằng văn bản thì cơ quan quản lý lao động buộc phải sửa. Gây thiệt hại cho bạn thì đơn vị phải bồi thường. Luật còn quy định riêng, đơn vị mới tuyển người chưa chấm dứt hợp đồng với đơn vị cũ mà gây thiệt hại cho đơn vị cũ thì phải chịu trách nhiệm bồi thường liên đới. Giấy xác nhận nghỉ việc chính là bằng chứng bằng văn bản rằng bạn đã chấm dứt hợp đồng với đơn vị cũ (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2012, sửa đổi). Bộ luật Hợp đồng lao động, 劳动合同法（第五十、八十九、九十一条）. ；Hội đồng Nhà nước (国务院) (2008). Quy chế thực thi Bộ luật Hợp đồng lao động, 劳动合同法实施条例（国务院令第 535 号，第二十四条）. ；Hội đồng Nhà nước (国务院) (1999). Quy định về bảo hiểm thất nghiệp, 失业保险条例（国务院令第 258 号，第十六条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2007). Luật hoà giải và trọng tài tranh chấp lao động, 劳动争议调解仲裁法（第二、二十七、四十三、五十三条）. ；Hội đồng Nhà nước (国务院) (2004). Quy định về giám sát bảo hiểm lao động, 劳动保障监察条例（国务院令第 423 号，第十七条）. ；Hội đồng Nhà nước (国务院) (2006). Biện pháp về việc nộp phí tụng, 诉讼费用交纳办法（国务院令第 481 号，第六、十三、二十九条）. +- Ghi chú: Điều luật không buộc việc cấp giấy chứng nhận với việc bàn giao công việc, bàn giao xong mới trả thì đó là bồi thường kinh tế. Đơn vị kéo dài không đưa thì trước tiên gửi tin nhắn bằng văn bản để xin lại một lần nữa. Vẫn không đưa thì gọi 12333 hoặc khiếu nại tới thanh tra lao động ở Sở Nhân sự và Xã hội. Khiếu nại không mất phí, sau khi tiếp nhận thì 60 ngày làm việc có kết quả, vụ phức tạp thêm tối đa 30 ngày làm việc. Muốn đòi đơn vị bồi thường thiệt hại thì đi hoà giải lao động, cũng không mất phí, phải nộp đơn trong 1 năm. Sau khi tiếp nhận thì 45 ngày kết thúc, nhiều nhất kéo thêm 15 ngày. Thiệt hại thì bạn phải tự đưa ra chứng cứ, chẳng hạn vì không có giấy nên đăng ký thất nghiệp bị chậm, hoặc bị đơn vị mới từ chối nhận. Không chịu phán quyết thì ra tòa, bản án sơ thẩm theo thủ tục thường kéo dài từ 6 tháng. Lệnh phí tiếp nhận vụ án lao động mà tòa thu là 10 tệ mỗi vụ, phí luật sư không nằm trong phí tụng, phải tự bỏ ra. Trong bốn thứ quy định bắt viết không có lý do nghỉ việc. Trên giấy mà ghi「个人原因主动辞职」mà sự thật không phải vậy thì nêu ra ngay tại chỗ, lý do xem mục 7 của phần này (đừng ký tự nguyện nghỉ việc). Cách nhận tiền trợ cấp thất nghiệp xem mục 1 ở phần 7 (tiền trợ cấp thất nghiệp). +### 18. Nhận tiền bồi thường khi nghỉ việc thì xem thuế thu nhập cá nhân trước: phần trong 3 mức lương bình quân người lao động của năm trước ở địa phương thì miễn thuế, phần vượt ra tính riêng, không gộp vào thu nhập của năm đó + +- Chi phí: Không tốn tiền. Tra mức lương bình quân người lao động của năm trước ở địa phương, rồi dành mấy phút đối chiếu xem đơn vị đã khấu trừ thuế thế nào. Trong ứng dụng Thuế thu nhập cá nhân có thể xem chi tiết kê khai của từng khoản thu nhập. +- Nói thẳng: Khoản tiền một lần bạn nhận được vì chấm dứt hợp đồng lao động thì phần trong 3 lần mức lương bình quân người lao động của năm trước ở địa phương không phải đóng thuế thu nhập cá nhân. Phần vượt ra thì tính thuế riêng, không cộng vào thu nhập của năm đó. Đơn vị mà gộp cả khoản này vào lương tháng rồi khấu trừ thuế thì trước tiên hỏi đơn vị cho rõ, rồi gọi 12366 hỏi cơ quan thuế. +- Lợi ích: Văn bản của Bộ Tài chính và Cục Thuế quy định, thu nhập một lần từ việc cá nhân chấm dứt quan hệ lao động với đơn vị sử dụng lao động, gồm bồi thường kinh tế, tiền trợ cấp sinh hoạt và các khoản trợ cấp khác. Phần trong 3 lần mức lương bình quân người lao động của năm trước ở địa phương thì miễn thuế thu nhập cá nhân. Phần vượt 3 lần không gộp vào thu nhập tổng hợp của năm đó, mà tính thuế riêng theo biểu thuế suất thu nhập tổng hợp. Biểu này, phần trong 36000 tệ cả năm thuế suất 3%, phần 36000 tệ đến 144000 tệ thuế suất 10%, cao hơn thì chia bậc tăng tới 45%. Ví dụ, lương bình quân người lao động của năm trước ở địa phương là 10 vạn tệ thì 3 lần là 30 vạn tệ. Nhận 36 vạn tệ bồi thường thì chỉ tính thuế 6 vạn tệ vượt ra, thành 3480 tệ. Cùng 36 vạn tệ đó mà tính cả khoản theo biểu này thì là 58080 tệ. Quy định này chạy từ ngày 1 tháng 1 năm 2019, không ghi ngày hết hiệu lực (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Bộ Tài chính và Cục Thuế (财政部、税务总局) (2018). Thông báo về việc nối tiếp các chính sách ưu đãi sau khi sửa đổi Luật Thuế thu nhập cá nhân, 关于个人所得税法修改后有关优惠政策衔接问题的通知（财税〔2018〕164 号，第五条第（一）项、第八条、第九条）. （Kho chính sách và pháp luật Cục Thuế nhà nước）；Bộ Tài chính và Cục Thuế (财政部、税务总局) (2023). Thông báo về việc tiếp tục áp dụng chính sách thuế thu nhập cá nhân đối với tiền thưởng một lần trong năm, 关于延续实施全年一次性奖金个人所得税政策的公告（2023 年第 30 号）. （Kho chính sách và pháp luật Cục Thuế nhà nước）；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018, sửa đổi). Luật Thuế thu nhập cá nhân, 个人所得税法（第三、九条，个人所得税税率表一）. +- Ghi chú: Trong cùng văn bản đó, khoản tiền thưởng một lần trong năm có ghi ngày hết hiệu lực, sau đó kéo dài tới ngày 31 tháng 12 năm 2027. Khoản bồi thường khi chấm dứt quan hệ lao động không ghi ngày hết hiệu lực, không đi theo khoản tiền thưởng. Văn bản chỉ nêu tên bồi thường kinh tế, tiền trợ cấp sinh hoạt và các khoản trợ cấp khác, không nêu riêng tiền bù giáp ngạp và tiền bồi thường khi chấm dứt sai luật. Hai khoản đó có được tính vào không, nếu không chắc thì gọi 12366 hỏi cơ quan thuế địa phương. 「Mức lương bình quân người lao động của năm trước ở địa phương」lấy con số nào thì cũng theo cơ quan thuế địa phương. Bản thân khoản bồi thường tính thế nào xem mục 4 của phần này (bị sa thải thì tính rõ N trước), mục 5 của phần này (không báo trước 30 ngày) và mục 6 của phần này (tiền bồi thường khi chấm dứt sai luật). Khoản tiền một lần khi nghỉ việc sớm, nghỉ việc hưu trí nội bộ thì có cách tính khác, không áp dụng quy định này. diff --git "a/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" "b/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" index 9ea837b6..92d69776 100644 --- "a/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" +++ "b/vi/book/20-\345\210\232\345\207\272\347\224\237\347\232\204\345\255\251\345\255\220\346\200\216\344\271\210\345\270\246.md" @@ -60,6 +60,7 @@ Cột đo của phần này là tỷ lệ tử vong trẻ sơ sinh và tiền. P - Mức bằng chứng: B - Nguồn: Fiesack S, Smits A, Rayyan M, et al. (2021). Nutrients, 13(11), 4109. ；Tursunov D, Yoshida Y, Yrysov K, et al. (2018). Nagoya Journal of Medical Science, 80(1), 11. - Ghi chú: Sinh con ở bệnh viện chính quy tại Trung Quốc, bệnh viện sẽ tiêm mũi này thường quy, không phải tự lo. Sinh tại nhà, hoặc tại cơ sở không chính quy, phải tự đi xác nhận. Dạng uống kém hiệu quả hơn dạng tiêm, ở đây tiêm là tiêm vào bắp + ### 8. Trẻ dưới 3 tháng tuổi, nhiệt độ chạm 38 ℃ thì đi bệnh viện luôn, không quan sát ở nhà - Chi phí: Khám bệnh cộng xét nghiệm vài trăm tệ @@ -99,4 +100,13 @@ Cột đo của phần này là tỷ lệ tử vong trẻ sơ sinh và tiền. P - Lợi ích: Thử nghiệm chia nhóm ngẫu nhiên LEAP của Anh. 640 trẻ sơ sinh được đưa vào nghiên cứu, điều kiện là bị eczema nặng, dị ứng trứng, hoặc cả hai. Khi nhập mẫu trẻ từ 4 đến 11 tháng tuổi. Chia ngẫu nhiên thành hai nhóm, một nhóm ăn đậu phộng đều đặn, một nhóm né hoàn toàn, đến 60 tháng tuổi thì xem tỷ lệ dị ứng đậu phộng. Trong 530 trẻ có xét nghiệm da âm tính với đậu phộng lúc nhập mẫu, nhóm né 13.7% dị ứng, nhóm ăn 1.9%（P<0.001）. Trong 98 trẻ đã dương tính yếu lúc nhập mẫu, nhóm né 35.3%, nhóm ăn 10.6%（P=0.004）. Biến cố bất lợi nghiêm trọng không khác nhau giữa hai nhóm. - Mức bằng chứng: A - Nguồn: Du Toit G, Roberts G, Sayre PH, 等 (2015). Randomized trial of peanut consumption in infants at risk for peanut allergy. New England Journal of Medicine, 372(9), 803-813. ；Văn phòng Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委办公厅) (2020). 婴幼儿喂养健康教育核心信息. ；Perkin MR 等 (2016). Randomized Trial of Introduction of Allergenic Foods in Breast-Fed Infants. New England Journal of Medicine, 374(18), 1733-1743. （thử nghiệm EAT với trẻ sơ sinh thường nêu trong ghi chú） -- Ghi chú: **Dạng thức chỉ có thể là sốt đậu phộng pha loãng, hoặc bột đậu phộng trộn vào thức ăn dặm, tuyệt đối không cho hạt đậu phộng nguyên.** Thông tin cốt lõi về nuôi dưỡng của Ủy ban Quốc gia Y tế và Sức khoẻ nói rõ: 「整粒花生、坚果、果冻等食物易吸入气管，引起窒息，婴幼儿应当避免食用」, tức hạt đậu phộng nguyên, hạt cây, thạch và các thức ăn tương tự rất dễ bị hút vào khí quản, gây ngạt thở, trẻ sơ sinh và trẻ nhỏ nên tránh ăn. Bị nghẹn thì làm gì, xem mục 26 ở phần 13 (có người bị nghẹn không nói nổi). Chỉ nhóm trẻ nguy cơ cao mới áp dụng: trẻ bị eczema nặng, hoặc đã dị ứng trứng. **Trước khi thêm phải đi khám bác sĩ để đánh giá, đừng tự thử ở nhà**: trước khi nhập mẫu, thử nghiệm LEAP làm xét nghiệm da cho từng trẻ, trẻ có phản ứng da lớn hơn 4 milimét bị loại khỏi thử nghiệm, không cho ăn. Với trẻ thường, bằng chứng yếu hơn nhiều. Trong một thử nghiệm khác gồm 1162 trẻ sơ sinh bú mẹ bình thường, tính theo nhóm thì nhóm cho sớm tỷ lệ dị ứng thức ăn là 5.6%, nhóm thường là 7.1%, chênh lệch không có ý nghĩa (P=0.32); chỉ trong nhóm trẻ thực sự làm theo mới thấy khác biệt (2.4% so với 7.3%). Cách tính này dễ nói quá hiệu quả, chính tác giả nói tính theo nhóm không chứng minh được hiệu lực, nhưng cũng cho thấy cho sớm là an toàn. Văn bản nuôi dưỡng chính thức của Trung Quốc hiện chưa có lập trường về 「nên cho sớm hay nên tránh thức ăn dễ dị ứng」, nên mục này viết theo bằng chứng thử nghiệm quốc tế. Về thời điểm cũng phải để ý: LEAP bắt đầu từ 4 tháng tuổi, còn Trung Quốc bắt đầu thêm thức ăn dặm khi đủ 6 tháng tuổi, xem mục 4 của phần này (sáu tháng đầu chỉ cho bú mẹ). Trẻ nguy cơ cao bắt đầu từ mấy tháng tuổi và bắt đầu thế nào, nghe bác sĩ. \ No newline at end of file +- Ghi chú: **Dạng thức chỉ có thể là sốt đậu phộng pha loãng, hoặc bột đậu phộng trộn vào thức ăn dặm, tuyệt đối không cho hạt đậu phộng nguyên.** Thông tin cốt lõi về nuôi dưỡng của Ủy ban Quốc gia Y tế và Sức khoẻ nói rõ: 「整粒花生、坚果、果冻等食物易吸入气管，引起窒息，婴幼儿应当避免食用」, tức hạt đậu phộng nguyên, hạt cây, thạch và các thức ăn tương tự rất dễ bị hút vào khí quản, gây ngạt thở, trẻ sơ sinh và trẻ nhỏ nên tránh ăn. Bị nghẹn thì làm gì, xem mục 26 ở phần 13 (có người bị nghẹn không nói nổi). Chỉ nhóm trẻ nguy cơ cao mới áp dụng: trẻ bị eczema nặng, hoặc đã dị ứng trứng. **Trước khi thêm phải đi khám bác sĩ để đánh giá, đừng tự thử ở nhà**: trước khi nhập mẫu, thử nghiệm LEAP làm xét nghiệm da cho từng trẻ, trẻ có phản ứng da lớn hơn 4 milimét bị loại khỏi thử nghiệm, không cho ăn. Với trẻ thường, bằng chứng yếu hơn nhiều. Trong một thử nghiệm khác gồm 1162 trẻ sơ sinh bú mẹ bình thường, tính theo nhóm thì nhóm cho sớm tỷ lệ dị ứng thức ăn là 5.6%, nhóm thường là 7.1%, chênh lệch không có ý nghĩa (P=0.32); chỉ trong nhóm trẻ thực sự làm theo mới thấy khác biệt (2.4% so với 7.3%). Cách tính này dễ nói quá hiệu quả, chính tác giả nói tính theo nhóm không chứng minh được hiệu lực, nhưng cũng cho thấy cho sớm là an toàn. Văn bản nuôi dưỡng chính thức của Trung Quốc hiện chưa có lập trường về 「nên cho sớm hay nên tránh thức ăn dễ dị ứng」, nên mục này viết theo bằng chứng thử nghiệm quốc tế. Về thời điểm cũng phải để ý: LEAP bắt đầu từ 4 tháng tuổi, còn Trung Quốc bắt đầu thêm thức ăn dặm khi đủ 6 tháng tuổi, xem mục 4 của phần này (sáu tháng đầu chỉ cho bú mẹ). Trẻ nguy cơ cao bắt đầu từ mấy tháng tuổi và bắt đầu thế nào, nghe bác sĩ. + +### 13. Canh mấy dấu hiệu vàng da ở trẻ sơ sinh: vàng trong 24 giờ đầu sau khi sinh, vàng đậm dần, gọi không dậy không bú, phân màu trắng thì trong ngày đó đi bệnh viện + +- Chi phí: Phí khám và đo bilirubin, dưới vài trăm tệ. Trẻ phải chiếu ánh sáng xanh thì thường phải ở bệnh viện khoảng hai ngày. Chỗ khó là lúc thấy con vẫn bình thường mà vẫn phải đi chuyến đó +- Nói thẳng: Hơn tám phần mười trẻ sơ sinh hơi vàng, đa số tự hết. Vàng da trước 24 giờ sau khi sinh, hoặc vàng đậm dần, ngủ li bì hơn bình thường gọi không dậy, không bú, thân mềm hoặc cứng, thì trong ngày đó phải đi bệnh viện. Phân màu trắng thì có thể là tắc mật tạo bàng, cũng phải đi xét sớm. Bilirubin cao quá thì tổn thương não, để lại tàn tật cả đời. +- Lợi ích: Hướng dẫn năm 2022 của Hiệp hội Nhi khoa Hoa Kỳ viết「超过 80% 的新生儿会有某种程度的黄疸」— hơn 80% trẻ sơ sinh bị vàng da ở mức nào đó. Bilirubin quá cao gây bệnh não bilirubin cấp và vàng da não. Hướng dẫn viết vàng da não là「永久的致残性神经系统疾病」— bệnh thần kinh gây tàn tật vĩnh viễn. Biểu hiện gồm bại liệt não với tay chân co giật không tự chủ, mắt không nhìn thẳng lên trên, men răng sữa kém phát triển và tổn thương thính giác. Hướng dẫn xếp「出生后 24 小时内出现黄疸」là yếu tố nguy cơ dẫn tới vàng da nặng. Cách chia hai bậc mà Hệ thống Dịch vụ Y tế Quốc gia Anh (NHS) nói với cha mẹ thế này. Bậc một thì gọi cấp cứu hoặc đi thẳng khoa cấp cứu. Bậc này là trẻ có vàng da, đồng thời ngủ li bì hơn bình thường hoặc gọi không dậy, không bú, thân mềm hoặc cứng, có giật. Trẻ vừa sinh chưa đủ 24 giờ mà đã vàng da thì cũng ở bậc này. Bậc hai thì hẹn bác sĩ càng sớm càng tốt. Bậc này là sau 24 giờ mà vàng da không khỏi hoặc nặng thêm, nước tiểu vàng đậm hoặc nâu, hoặc phân màu kem nhạt. Ước tính toàn cầu, năm 2010 có khoảng 480.000 trẻ sơ sinh mắc bệnh huyết tanh Rh (do mẹ và con khác nhóm máu) hoặc bilirubin cao cực kỳ nặng. Trong số trẻ đó 24% chết, 13% tiến thành vàng da não. Trong số trẻ sống sót với vàng da não, hơn 83% để lại một hay nhiều di chứng. Phân màu trắng thì có thể là tắc mật tạo bàng, bệnh này phải mổ Kasai trong 60 ngày thì tiên lượng mới rõ ràng hơn nhiều. Từ năm 2004, Đài Loan in thẻ so màu phân vào sổ sức khoẻ trẻ em của mọi trẻ sơ sinh. Tỷ lệ trẻ được mổ trong 60 ngày tăng từ 60% năm 2004 lên 74.3% năm 2005. Một nghiên cứu khác toàn dân Đài Loan cho thấy sau khi dùng thẻ so màu, tỷ lệ tử vong ở trẻ tắc mật tạo bàng giảm từ 26.2% xuống 15.9%. Đây là so sánh trước sau, không phải thử nghiệm chia nhóm +- Mức bằng chứng: B +- Nguồn: Kemper AR, Newman TB, Slaughter JL, et al. (2022). Clinical Practice Guideline Revision: Management of Hyperbilirubinemia in the Newborn Infant 35 or More Weeks of Gestation. Pediatrics, 150(3), e2022058859. ；NHS. Jaundice in babies（xem lại ngày 2026-03-24）. ；NICE (2010, sửa đổi 2023). Jaundice in newborn babies under 28 days (CG98), 1.2.6、1.7.1. ；Bhutani VK, Zipursky A, Blencowe H, et al. (2013). Neonatal hyperbilirubinemia and Rhesus disease of the newborn: incidence and impairment estimates for 2010 at regional and global levels. Pediatric Research, 74(Suppl 1), 86-100. ；Hsiao CH, Chang MH, Chen HL, et al. (2008). Universal screening for biliary atresia using an infant stool color card in Taiwan. Hepatology, 47(4), 1233-1240. ；Lee M, Chen SC, Yang HY, et al. (2016). Infant Stool Color Card Screening Helps Reduce the Hospitalization Rate and Mortality of Biliary Atresia: A 14-Year Nationwide Cohort Study in Taiwan. Medicine, 95(12), e3166. ；国家卫生计生委 (2017). 国家基本公共卫生服务规范（第三版），0～6 岁儿童健康管理服务规范. +- Ghi chú: Chiếu ánh sáng xanh (liệu pháp ánh sáng) là dùng loại ánh sáng đặc biệt chiếu lên da trẻ, làm bilirubin chuyển thành dạng dễ đào ra. Đừng phơi nắng thay ở nhà: Hiệp hội Nhi khoa Hoa Kỳ nói phơi nắng khó vừa an toàn vừa không cháy, nên không khuyến nghị dùng làm cách điều trị. Cũng đừng cho uống nước hay nước đường để hạ vàng da, hướng dẫn này cũng nói rõ không khuyến nghị. Nhìn bằng mắt độ sâu màu vàng không ước tính được bilirubin cao tới đâu. Hướng dẫn NICE của Anh viết rõ đừng chỉ dựa vào mắt, không chắc thì đi đo một lần. Trẻ sinh trước 38 tuần, có anh chị em đã từng chiếu ánh sáng xanh, hoặc bú mẹ chưa đủ thì dễ vàng nặng hơn. Tái khám theo Quy chuẩn dịch vụ công cộng cơ bản quốc gia: trong 1 tuần sau khi xuất viện, nhân viên y tế đến thăm tại nhà. Từ 28 đến 30 ngày sau khi sinh, tới trạm y tế cộng đồng hoặc trạm y tế xã theo dõi. Cả hai lần đều phải xem vàng da. Khi bác sĩ hẹn thêm lần đo bilirubin lúc xuất viện thì đi đúng hẹn. Trẻ đủ tháng mà vàng da quá 14 ngày, trẻ non tháng quá 21 ngày, hướng dẫn NICE yêu cầu đo một lần bilirubin liên kết, và xem phân có trắng không. Bản thân vàng da không cản tiêm vaccine, xem mục 3 của phần này (vaccine). Sàng lọc trẻ sơ sinh xem mục 13 ở phần 27 (xét gót chân). diff --git "a/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" "b/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" index c9e67ca6..4f51b93c 100644 --- "a/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" +++ "b/vi/book/21-\345\207\272\345\233\275\346\227\205\350\241\214\344\270\216\345\242\203\345\244\226\345\256\211\345\205\250.md" @@ -52,14 +52,16 @@ Cột đo của phần này là tiền và tự do. Các mục xếp theo hiệu - Mức bằng chứng: A - Nguồn: Cục Quản lý Ngoại hối nhà nước (国家外汇管理局) (2017). 关于规范银行卡境外大额提取现金交易的通知（汇发〔2017〕29 号, Điều một, hai, ba, bốn, sáu, bảy, có hiệu lực từ ngày 1 tháng 1 năm 2018）. ；Cục Quản lý Ngoại hối nhà nước (国家外汇管理局) (2017). 外汇局有关负责人就规范银行卡境外大额提取现金交易有关问题答记者问. ；Tập đoàn UnionPay Quốc tế Trung Quốc (中国银联国际). Rút tiền tại ATM ở nước ngoài, 境外ATM取款. - Ghi chú: Hạn mức tính theo năm dương lịch, ngày 1 tháng 1 bắt đầu lại từ đầu. Chỉ có việc rút tiền mặt ở nước ngoài bị dừng, chi tiêu bằng thẻ không bị ảnh hưởng. Nếu thật sự cần dùng nhiều tiền mặt ở nước ngoài thì theo Biện pháp quản lý ngoại hối của cá nhân (个人外汇管理办法), làm thủ tục mua ngoại tệ trước, rồi mang tiền ngoại tệ mặt ra khỏi nước. Bản thân việc rút tiền còn có hai cái bẫy, trang 「境外ATM取款」của UnionPay Quốc tế nói rõ. Một, thẻ có hai nhãn ở một số ATM tại một số nước không chọn được kênh UnionPay để rút. Hai, một số ATM ở nước ngoài hiện trên màn hình khoản phí thêm (surcharge), khoản tiền đó do tổ chức thanh toán tại chỗ thu, 「与银联无关」, không liên quan gì đến UnionPay. Cùng trang đó còn nhắc 「由于信用卡取款需支付透支利息，建议持卡人境外取款使用银联借记卡」, tức vì rút tiền bằng thẻ tín dụng phải trả lãi vượt hạn mức, nên khuyến nghị chủ thẻ rút tiền ở nước ngoài bằng thẻ ghi nợ UnionPay. Thẻ bị mất, bị nuốt hoặc bị dùng gian lận thì làm gì, xem mục 5 ở phần 14 (thẻ bị dùng gian lận thì báo mất và đóng băng trước, rồi báo công an) -### 7. Chụp ảnh hộ chiếu, visa, giấy tùy thân lưu một bản trên đám mây, mất rồi báo công an trước rồi xin giấy tờ du lịch + +### 7. Chụp ảnh hộ chiếu, visa, thẻ căn cước lưu một bản trên đám mây, mất rồi thì báo công an trước rồi làm giấy tờ du lịch -- Chi phí: Không tốn tiền. Trước khi đi, chụp vài bức ảnh. -- Nói thẳng: Mất hộ chiếu, đại sứ đoàn có thể cấp cho bạn một giấy tờ du lịch dùng để về nước. Điều kiện là bạn chứng minh được mình là ai. Có ảnh trang thông tin, ảnh trang visa và lịch sử nhập cảnh trong tay thì việc cấp lại nhanh hơn nhiều. Bạn còn phải báo công an tại chỗ trước để lấy biên nhận, kiểm tra xuất cảnh và bồi thường bảo hiểm đều cần tờ này. Điện thoại thường mất cùng hộ chiếu, đừng chỉ lưu ảnh trong điện thoại. -- Lợi ích: Mất hộ chiếu, đại sứ toàn tháp ở nước ngoài có thể cấp cho bạn giấy tờ du lịch dùng để về nước, nhưng cần bạn chứng minh danh tính trước. Có ảnh rõ trang thông tin hộ chiếu, ảnh trang visa và lịch sử nhập cảnh trong tay thì tốc độ cấp lại chênh lệch rất lớn. Đồng thời bạn phải báo cô an tại chỗ trước để lấy biên nhận báo án. Nhiều nước yêu cầu tài liệu này khi kiểm tra xuất cảnh và khi bồi thường bảo hiểm. Cửa vào để tìm đại sứ đoàn chính là hai số bảo vệ lãnh sự đã lưu ở mục 2 của phần này +- Chi phí: Không tốn tiền. Trước khi đi chụp mấy tấm ảnh. +- Nói thẳng: Mất hộ chiếu thì sứ quán dịch có thể cấp cho bạn một giấy tờ du lịch dùng để về nước. Điều kiện là bạn chứng minh được mình là ai. Trong tay có ảnh trang thông tin, ảnh trang visa và lịch sử nhập cảnh thì việc làm lại nhanh hơn nhiều. Bạn còn phải báo công an tại nơi đó trước để lấy biên bản, kiểm tra xuất cảnh và bồi hoàn bảo hiểm đều công nhận tờ giấy đó. Điện thoại thường mất cùng hộ chiếu, đừng chỉ lưu ảnh trong điện thoại. +- Lợi ích: Mất hộ chiếu thì cơ quan đại diện ngoại giao và lãnh sự ở nước ngoài có thể cấp cho bạn giấy tờ du lịch dùng để về nước, nhưng bạn phải chứng minh được thân phận trước. Trong tay có ảnh trang thông tin hộ chiếu rõ, ảnh trang visa và lịch sử nhập cảnh thì tốc độ làm lại chênh nhau rất lớn. Đồng thời bạn phải báo công an tại nơi đó trước để lấy biên bản báo cáo. Nhiều nước kiểm tra xuất cảnh và bồi hoàn bảo hiểm đều cần bộ hồ sơ này. Cửa vào tìm sứ quán dịch chính là hai số điện thoại bảo vệ lãnh sự đã lưu ở mục 2 của phần này - Mức bằng chứng: C - Nguồn: Hội đồng Nhà nước (国务院) (2023). Quy chếnh bảo vệ và hỗ trợ lãnh sự, 领事保护与协助条例（国令第 763 号）. （Điều mười sáu）；kinh nghiệm của tác giả, không có tài liệu trực tiếp -- Ghi chú: Cách lưu là chụp xong lưu vào hai nơi khác nhau, ví dụ một bản trên ổ đám mây, một bản gửi cho một người nhà. Ngoài ra in một bản giấy, cất chỗ khác với bản gốc. Chụp luôn thẻ ngân hàng và đơn thuốc thường dùng. Điện thoại và hộ chiếu thường mất cùng nhau, chỉ lưu trong điện thoại thì coi như không lưu +- Ghi chú: Cách lưu là chụp xong rồi đặt vào hai nơi khác nhau, ví dụ lưu một bản trên đám mây, gửi cho một người nhà một bản. Ngoài ra in một bản sao chụp giấy, cất chỗ khác với bản gốc. Thẻ ngân hàng, đơn thuốc dùng thường cũng chụp luôn. Điện thoại và hộ chiếu thường mất cùng nhau, chỉ lưu trong điện thoại là như không lưu. Đám mây nào mà đăng nhập cần mã xác minh gửi về điện thoại thì mất điện thoại là có thể không vào được. Trước khi đi, thử đăng nhập một lần trên thiết bị khác, xác nhận không nhờ chiếc điện thoại đó vẫn mở được. Ảnh giấy tờ tùy thân là thông tin nhạy cảm, chọn cách lưu có mã hoá thì an toàn hơn. Nhưng cách lưu có mã hoá mà quên mật khẩu thì ai cũng không mở được, nên bản gửi cho người nhà không được bỏ. + ### 8. Tự lái xe ở nước ngoài, trước tiên xác nhận nơi đó có công nhận giấy phép lái xe Trung Quốc không, 「giấy phép lái xe quốc tế」bán trên mạng đa phần là giấy vô dụng - Chi phí: Không tốn tiền. Trước khi đi, tra quy định của điểm đến một lần. @@ -91,4 +93,4 @@ Cột đo của phần này là tiền và tự do. Các mục xếp theo hiệu - Lợi ích: Khi công dân Trung Quốc mất tích ở nước ngoài, người thân có thể nhờ cơ quan ngoại giao ở nước ngoài giúp. Nguyên văn quy chếnh là 「驻外外交机构应当提供当地报警方式及其他获取救助的信息」, tức cơ quan ngoại giao ở nước ngoài phải cung cấp cách báo công an tại chỗ cùng các thông tin khác về cách tiếp cận cứu hộ. Sau khi công an địa phương lập hồ sơ, đại sứ đoàn sẽ thúc họ xử lý kịp thời. Các trợ giúp này có khởi động được hay không phụ thuộc vào việc ở trong nước có ai biết bạn đáng lẽ ở đâu, đáng lẽ liên lạc vào lúc nào - Mức bằng chứng: C - Nguồn: Hội đồng Nhà nước (国务院) (2023). Quy chếnh bảo vệ và hỗ trợ lãnh sự, 领事保护与协助条例（国令第 763 号）. （Điều mười ba）；kinh nghiệm của tác giả, không có tài liệu trực tiếp -- Ghi chú: Chỉ cần viết rõ bốn thứ: số hiệu chuyến bay, địa chỉ nơi ở, tên và số điện thoại người đi cùng, tần suất liên lạc đã hẹn. Nếu đi nơi sóng yếu hoặc tình hình an ninh kém, hãy viết thêm câu 「bao lâu không có tin thì báo công an」. Tất cả các thủ tục cứu hộ nói trên đều phải có người nhà biết lịch trình của bạn thì mới bắt đầu được \ No newline at end of file +- Ghi chú: Chỉ cần viết rõ bốn thứ: số hiệu chuyến bay, địa chỉ nơi ở, tên và số điện thoại người đi cùng, tần suất liên lạc đã hẹn. Nếu đi nơi sóng yếu hoặc tình hình an ninh kém, hãy viết thêm câu 「bao lâu không có tin thì báo công an」. Tất cả các thủ tục cứu hộ nói trên đều phải có người nhà biết lịch trình của bạn thì mới bắt đầu được diff --git "a/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" "b/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" index 9e64f88a..6817c47c 100644 --- "a/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" +++ "b/vi/book/22-\346\200\216\344\271\210\346\224\276\346\235\276.md" @@ -36,15 +36,7 @@ Cột đo của phần này là sức khoẻ và thời gian. Các mục xếp t - Mức bằng chứng: A - Nguồn: Viện kiểm sát nhân dân tối cao (最高人民检察院) (2021). 江苏苏州吴江区：办案检察官揭秘披着潮流外衣的新型毒品. ；Cục Quản lý Dược phẩm nhà nước, Bộ Công an, Ủy ban Quốc gia Y tế và Sức khoẻ (国家药监局、公安部、国家卫生健康委) (2023). 关于调整麻醉药品和精神药品目录的公告（2023 年第 120 号）. （Cục Quản lý Dược phẩm Thượng Hải đăng lại, trang gốc nmpa.gov.cn trả về 412 trên máy này）；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2025). Luật Xử phạt hành chính về trật tự công cộng, 治安管理处罚法（2025 年修订, Điều tám mươi tư, tám mươi lăm）. ；Đại hội Đại biểu Quốc dân (全国人大) (1997). Bộ luật Hình sự, 刑法（Điều ba trăm năm mươi ba）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2007). Luật chống ma túy, 禁毒法（Điều ba mươi ba, ba mươi tám, bốn mươi bảy）. （Chính phủ tỉnh Quảng Đông đăng lại） - Ghi chú: Không chỉ nơi giải trí, cửa trường học, tụ họp bạn bè, xe đường dài cũng vậy. Cho đồ ngọt, cho thức uống hay cho điếu thuốc điện tử thì đều thế. Sau khi bị cho thuốc thường đi kèm cướp đoạt, xâm hại tình dục và tống tiền, phần này không có thống kê chính thức công khai, thuộc phán đoán theo kinh nghiệm. Thức uống chỉ cần rời khỏi tầm mắt là đổi ly mới, dễ hơn nhiều so với truy đòi sau. Trong chỗ vui có người lấy ra bột không rõ, viên nén, điếu thuốc điện tử thì làm gì, xem mục 3 của phần này. Giúp người ta mang đồ xem mục 29 ở phần 8. -### 5. Vào quán Internet dùng giấy tùy thân của bạn, đừng cho mượn, cũng đừng dùng giấy của người khác - -- Chi phí: Không tốn tiền. -- Nói thẳng: Theo quy định, quán Internet phải kiểm tra và ghi đăng ký giấy tùy thân của người lên mạng, còn phải ghi lại thông tin lên mạng, lưu giữ không dưới 60 ngày. Cơ quan công an và cơ quan văn hoá có thể tra bất cứ lúc nào. Trong khoảng thời gian đó quán Internet không được sửa, cũng không được xoá. Dùng giấy của ai lên mạng thì trong khoảng thời gian đó, việc làm trên máy đó ghi vào tên người đó. -- Lợi ích: Nguyên văn quy chếnh: 「互联网上网服务营业场所经营单位应当对上网消费者的身份证等有效证件进行核对、登记，并记录有关上网信息。登记内容和记录备份保存时间不得少于 60 日」, tức đơn vị kinh doanh nơi cung cấp dịch vụ truy cập Internet phải kiểm tra, đăng ký giấy tùy thân và các giấy tờ hợp lệ khác của người tiêu dùng lên mạng, và ghi lại thông tin liên quan tới việc lên mạng. Nội dung đăng ký và bản sao lưu hồ sơ phải lưu giữ không dưới 60 ngày. Khi cơ quan quản lý văn hoá và cơ quan công an tra cứu theo đúng pháp luật, quán Internet phải cung cấp. Trong thời hạn lưu giữ không được sửa hay xoá. Vì vậy dùng giấy của ai lên mạng thì trong khoảng thời gian đó, việc làm trên máy đó ghi vào tên người đó -- Mức bằng chứng: A -- Nguồn: Hội đồng Nhà nước (国务院) (2002). Quy chếnh quản lý nơi kinh doanh dịch vụ truy cập Internet, 互联网上网服务营业场所管理条例（国令第 363 号, Điều hai mươi ba）. -- Ghi chú: Đây cùng một việc với mục 「không cho mượn giấy tùy thân」ở phần 9. Quán Internet không được phục vụ người chưa thành niên, ở cửa ra vào chỗ dễ thấy phải treo biển cấm người chưa thành niên. Nơi giải trí có ca hát cũng không phục vụ người chưa thành niên. Giúp người chưa thành niên 「quẹt giấy vào quán Internet」thì bị phạt là quán Internet, nhưng dữ liệu thì nằm lại trên giấy tờ của bạn -### 6. Đừng đến nơi chơi luận án hay phòng giải đố mở trong tòa nhà dân cư hoặc dưới tầng hầm một +### 5. Đừng đến nơi chơi luận án hay phòng giải đố mở trong tòa nhà dân cư hoặc dưới tầng hầm một - Chi phí: Không tốn tiền. Trước khi đặt, hỏi một câu địa chỉ ở tầng mấy. - Nói thẳng: Thông tư của năm 2022 của năm bộ ngành ghi rõ, địa điểm giải trí theo kịch bản không được mở trong tòa nhà dân cư, cũng không được mở dưới tầng hầm một. Loại chỗ này tối, vách ngăn nhiều, đạo cụ dễ cháy, cửa thường phải giải câu đố mới mở được. Một khi xảy ra cháy, chạy ra khó hơn cửa hàng thường rất nhiều. Trước khi đặt, xem địa chỉ và tầng trên trang đánh giá, không hợp lệ thì đổi chỗ. @@ -52,7 +44,7 @@ Cột đo của phần này là sức khoẻ và thời gian. Các mục xếp t - Mức bằng chứng: A - Nguồn: Bộ Văn hoá và Du lịch, Bộ Công an, Bộ Xây dựng và Nhà ở, Bộ Quản lý Khẩn cấp, Cục Quản lý thị trường quốc gia (文化和旅游部、公安部、住房和城乡建设部、应急管理部、市场监管总局) (2022). 关于加强剧本娱乐经营场所管理的通知. - Ghi chú: Đặc điểm của loại chỗ này là tối, vách ngăn nhiều, đạo cụ dễ cháy, cửa thường phải 「giải câu đố」mới mở được. Một khi xảy ra cháy, khó chạy ra hơn cửa hàng thường nhiều. Trước khi đặt, ngay trên trang đánh giá đã thấy được tầng, cũng thấy được có phải tòa nhà dân cư không. Không hợp lệ thì đổi chỗ luôn -### 7. Tâm trạng tệ thì đi bộ hoặc chạy, cỡ tác động chống trầm cảm (độ lớn hiệu quả) tỷ lệ thuận với cường độ +### 6. Tâm trạng tệ thì đi bộ hoặc chạy, cỡ tác động chống trầm cảm (độ lớn hiệu quả) tỷ lệ thuận với cường độ - Chi phí: Không tốn tiền. Mỗi lần khoảng nửa tiếng. Chỗ khó là lúc tâm trạng tệ lại càng không muốn động, phải tự đẩy mình một cái. - Nói thẳng: Khi tâm trạng tệ, đi bộ nhanh hoặc chạy chậm cho cải thiện rõ nhất. Kế tiếp là yoga, tập tạp, thái quyền và công năng, mấy cách này chênh lệch không nhiều. Cường độ càng cao thì hiệu quả càng tốt, tập tạp và yoga dễ duy trì nhất. Chất lượng bằng chứng ở lĩnh vực này hơi thấp, điều chắc chắn là vận động hơn không vận động, hãy chọn cái mà bạn duy trì nổi. @@ -60,7 +52,8 @@ Cột đo của phần này là sức khoẻ và thời gian. Các mục xếp t - Mức bằng chứng: A - Nguồn: Noetel M, Sanders T, Gallardo-Gómez D, et al. (2024). Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ, 384, e075847. - Ghi chú:Tranh cãi. Chính tác giả ghi rõ, trong 218 nghiên cứu chỉ có 1 nghiên cứu đủ cẩn thận, khó bị thiết kế làm sai lệch. Theo cách chấm điểm thông dụng, đi bộ nhanh và chạy chậm chỉ ở mức 「thấp」đáng tin, còn lại là 「rất thấp」. Vận động cũng không che được với người tham gia, người ta biết mình đang tập thì dễ cảm thấy khá hơn, không bỏ được phần kỳ vọng đó. Hướng kết luận vẫn dùng được: vận động hơn không vận động, cường độ đừng quá thấp, chọn cái mình duy trì nổi. Đây là phần mở rộ của mục 「khi tâm trạng tệ thì vận động trước đã」ở phần 3, không thay thế việc điều trị đúng quy trình sau khi đã chẩn đoán -### 8. Khi lo âu nhiều tới mức ảnh hưởng cuộc sống, lớp giảm căng thông chánh niệm hiệu quả không kém thuốc chống lo âu thường dùng + +### 7. Khi lo âu nhiều tới mức ảnh hưởng cuộc sống, lớp giảm căng thông chánh niệm hiệu quả không kém thuốc chống lo âu thường dùng - Chi phí: Đăng ký lớp học tốn một ít tiền. Khóa học 8 tuần, mỗi tuần học một buổi, còn phải luyện ở nhà mỗi ngày. Chỗ khó là luyện liên tục 8 tuần, ngày nào cũng phải luyện. - Nói thẳng: Ở người được chẩn đoán rối loạn lo âu, học lớp giảm căng thông chánh niệm 8 tuần thì mức cải thiện triệu chứng gần như giống uống thuốc chống lo âu escitalopram. Người uống thuốc có gần tám phần mười gặp tác dụng phụ, có người phải bỏ dở. Người học lớp khoảng một phần rưỡi gặp tác dụng phụ, không ai phải bỏ. Khóa học phải là khóa chuẩn hoá 8 tuần, một ứng dụng thiền nào cũng tính là không. @@ -68,7 +61,7 @@ Cột đo của phần này là sức khoẻ và thời gian. Các mục xếp t - Mức bằng chứng: A - Nguồn: Hoge EA, Bui E, Mete M, Dutton MA, Baker AW, Simon NM (2023). Mindfulness-based stress reduction vs escitalopram for the treatment of adults with anxiety disorders: a randomized clinical trial. JAMA Psychiatry, 80(1), 13-21. - Ghi chú: Thử nghiệm dùng khóa giảm căng thông chánh niệm chuẩn hoá 8 tuần (MBSR), một ứng dụng thiền nào cũng không thể ngang với nó. Nó dành cho người đã đạt tiêu chuẩn chẩn đoán rối loạn lo âu, nếu cần đi khám bác sĩ thì vẫn phải đi. Nếu vì lo tác dụng phụ của thuốc mà bạn cứ trì hoãn không điều trị, đây là một lựa chọn có sức bằng chứng tương đương -### 9. Muốn bình tĩnh lại ngay tại chỗ thì dùng 5 phút 「thở dài vòng lặp」: hít vào hai đoạn, kéo dài hơi thở ra +### 8. Muốn bình tĩnh lại ngay tại chỗ thì dùng 5 phút 「thở dài vòng lặp」: hít vào hai đoạn, kéo dài hơi thở ra - Chi phí: Không tốn tiền. Mỗi ngày 5 phút. - Nói thẳng: Làm bài tập thở 5 phút mỗi ngày, kiên trì một tháng, mức cải thiện tâm trạng và độ chậm của hơi thở đều tốt hơn thiền chánh niệm dài tương ứng. Trong các phương pháp thở, 「thở dài vòng lặp」có hơi thở ra kéo dài là tốt nhất. Cách làm là hít vào bằng mũi một hơi, rồi hít thêm một hơi nhỏ cho phổi đầy, sau đó thở ra chậm bằng miệng cho hết. Đây chỉ là một thử nghiệm, tốt hay không do người đó tự báo, nhưng không tốn tiền. @@ -76,7 +69,7 @@ Cột đo của phần này là sức khoẻ và thời gian. Các mục xếp t - Mức bằng chứng: B - Nguồn: Balban MY, Neri E, Kogon MM, et al. (2023). Brief structured respiration practices enhance mood and reduce physiological arousal. Cell Reports Medicine, 4(1), 100895. - Ghi chú: Cách làm là hít vào bằng mũi một hơi, hít thêm một hơi nhỏ trong thời gian ngắn cho phổi đầy, rồi thở ra chậm và dài bằng miệng cho hết, lặp lại trong 5 phút. Chỉ có một thử nghiệm này, làm ở nhà, tốt hay không do người đó tự điền bảng hỏi báo, số người cũng không nhiều. Hiệu quả chỉ hơn thiền một chút, không giải quyết được vấn đề. Nhưng nó không tốn tiền, chỉ 5 phút, làm được ở bất cứ lúc nào, hiệu quả chi phí cao -### 10. Coi 「gặp gỡ với người khác đều đặn」là một khoản chi cho sức khoẻ, đừng chỉ tìm người khi tâm trạng tệ +### 9. Coi 「gặp gỡ với người khác đều đặn」là một khoản chi cho sức khoẻ, đừng chỉ tìm người khi tâm trạng tệ - Chi phí: Tốn một ít tiền, còn phải bỏ thời gian đi gặp. Chỗ khó là lúc tâm trạng tốt vẫn phải chủ động rủ người. - Nói thẳng: Người ít giao lưu với mọi người có xác suất chết cùng kỳ cao hơn khoảng ba phần mười. Người cảm thấy cô đơn cao hơn khoảng một phần tư, người sống một mình cao hơn khoảng ba phần mười, người dưới 65 tuổi rõ hơn. Đây đều là quan sát theo dõi, cũng có thể vì người sức khoẻ vốn đã yếu thì dễ sống một mình hơn. @@ -84,11 +77,11 @@ Cột đo của phần này là sức khoẻ và thời gian. Các mục xếp t - Mức bằng chứng: A - Nguồn: Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D (2015). Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspectives on Psychological Science, 10(2), 227-237. - Ghi chú:Tranh cãi. Đây đều là mối liên hệ quan sát theo dõi một nhóm người, không phải thử nghiệm đối chứng có nhóm, nên không tách được nguyên nhân với hậu quả. Người sức khoẻ vốn đã yếu thì dễ sống một mình hơn, ít giao lưu hơn. Tác giả coi các quan hệ xã hội là một yếu tố sức khoẻ cùng cỡ với hút thuốc và béo phì, nhưng không có thử nghiệm ngẫu nhiên nào chứng minh 「tăng giao lưu xã hội có giảm tỷ lệ tử vong」. Mục này không mâu thuẫn với mục 「giảm bớt những quan hệ làm bạn hao hụt sức lực」ở phần 3: mục đó chặt bỏ những quan hệ làm bạn hao sức, mục này giữ lại vài mối quan hệ giúp bạn hồi phục sức lực -### 11. Ở gần cây xanh thì cứ dùng, đi bộ, đi dạo, ngồi nghỉ đều tính +### 10. Ở gần cây xanh thì cứ dùng, đi bộ, đi dạo, ngồi nghỉ đều tính - Chi phí: Không tốn tiền. Tiện đường thì làm luôn. - Nói thẳng: Cây xanh quanh chỗ ở càng tốt trong bán kính 500 mét thì xác suất chết cùng kỳ càng thấp, tính mỗi 0.1 điểm chỉ số thực vật thì thấp hơn khoảng 4%. Đo là cây xanh quanh chỗ ở, không phải số lần bạn đi công viên. Ở khu cây xanh tốt, người ở thường giàu hơn, không khí cũng tốt hơn. Vì vậy đừng hiểu là đi công viên là sống thêm, khi chọn chỗ ở thì tính cả cây xanh quanh đó vào. - Lợi ích: Đây là kết quả gộp 9 nghiên cứu theo dõi nhóm người nhiều năm (phân tích tổng hợp các nghiên cứu mẫu theo thời gian), liên quan tới 7 quốc gia, 8.320.000 người. Đo mức cây xanh quanh chỗ ở trong bán kính 500 mét, bằng chỉ số thực vật chuẩn hoá (NDVI, con số quy đổi độ phủ xanh mặt đất). Mỗi 0.1 điểm chỉ số cây xanh tăng thêm, tỉ số nguy cơ gộp của tỷ lệ tử vong toàn nguyên là 0.96 (95% CI 0.94–0.97), xác suất chết cùng kỳ thấp hơn khoảng 4%, giá trị thật nằm khoảng thấp hơn 3% tới thấp hơn 6%. Trong 9 nghiên cứu thì có 7 nghiên cứu thấy cây xanh tốt hơn thì tỷ lệ tử vong thấp hơn, và chênh lệch cũng khó do tình cờ. 2 nghiên cứu còn lại không thấy liên hệ - Mức bằng chứng: A - Nguồn: Rojas-Rueda D, Nieuwenhuijsen MJ, Gascon M, Perez-Leon D, Mudu P (2019). Green spaces and mortality: a systematic review and meta-analysis of cohort studies. The Lancet Planetary Health, 3(11), e469-e477. -- Ghi chú:Tranh cãi. Kết quả các nghiên cứu rất khác nhau (I²=95%, số này càng lớn thì các nghiên cứu càng không khớp nhau). Hơn nữa nó đo mức cây xanh quanh nơi ở, không liên quan gì tới số lần bạn đi công viên. Người ở nơi cây xanh tốt thường cũng giàu hơn, không khí tốt hơn, hay vận động hơn, mấy thứ đó không tách được với cây xanh thì cái nào là nguyên nhân. Đừng đọc thành 「đi công viên là sống thêm」, phần dùng được là tính cả cây xanh quanh nhà vào lúc chọn chỗ ở, chi phí bằng không. Con số NDVI này, đất trống và mặt bê tông gần 0, thảm thực vật dày gần 1. Nên 「0.1」xấp xỉ khác biệt giữa dãy phố trọc và dãy phố có hàng cây bàng đầy đủ \ No newline at end of file +- Ghi chú:Tranh cãi. Kết quả các nghiên cứu rất khác nhau (I²=95%, số này càng lớn thì các nghiên cứu càng không khớp nhau). Hơn nữa nó đo mức cây xanh quanh nơi ở, không liên quan gì tới số lần bạn đi công viên. Người ở nơi cây xanh tốt thường cũng giàu hơn, không khí tốt hơn, hay vận động hơn, mấy thứ đó không tách được với cây xanh thì cái nào là nguyên nhân. Đừng đọc thành 「đi công viên là sống thêm」, phần dùng được là tính cả cây xanh quanh nhà vào lúc chọn chỗ ở, chi phí bằng không. Con số NDVI này, đất trống và mặt bê tông gần 0, thảm thực vật dày gần 1. Nên 「0.1」xấp xỉ khác biệt giữa dãy phố trọc và dãy phố có hàng cây bàng đầy đủ diff --git "a/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" "b/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" index def46825..fba8f94e 100644 --- "a/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" +++ "b/vi/book/23-\345\255\246\344\273\200\344\271\210\346\212\200\350\203\275\345\210\222\347\256\227.md" @@ -12,15 +12,14 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 义务教育法（2018 年第二次修正，第十一、十四、五十八条）. （Bộ Giáo dục đăng lại）；Ủy ban Nhà nước (国务院) (2002). 禁止使用童工规定（国务院令第 364 号，第二、六、七、十、十三条）. - Ghi chú: Người hưởng lợi là bạn. Sau 16 tuổi thì đi làm công là hợp pháp. Nhưng những chỗ nhận một người 16 tuổi không học vấn đúng là loại mục 10 của phần này nói tới (chọn kỹ năng chống bị máy thay), quy trình cố định, dễ bị máy đẩy ra nhất. Tiền làm thêm giờ, nghỉ phép năm, ưu đãi tai nạn lao động mà phần 19 nói tới, đều đặt điều kiện là có quan hệ lao động và có bằng chứng. Việc chui không có cả hai, nên số tiền đó cũng không lấy được. - -### 2. Câu hỏi đi học có ích không cũng nên tính vào sổ tử vong: mỗi năm học thêm, nguy cơ tử vong ở tuổi trưởng thành giảm khoảng 1.9% - -- Chi phí: Không tốn tiền. Cái giá phải trả là thời gian và sức lực của mấy năm học thêm, phần học phí xem mục 4 của phần này (học phí và học bổng trường trung cấp chuyên nghiệp). Chỗ khó là phải học liên tục, không bỏ giữa chừng. -- Nói thẳng: Mỗi năm học thêm, nguy cơ tử vong sau khi thành người lớn trung bình thấp hơn khoảng 1.9%. Rõ nhất là người từ 18 đến 49 tuổi, mỗi năm thêm thì thấp hơn khoảng 2.9%; người trên 70 tuổi chỉ thấp hơn 0.8%. Người đọc được suốt thường vốn đã có điều kiện tốt, trong đó bao nhiêu là do chính việc học thì không tách được. -- Lợi ích: Gộp lại toàn bộ các nghiên cứu trước đây rồi tính chung (tổng quan hệ thống và phân tích tổng hợp). Khoảng tra cứu từ tháng 1 năm 1980 đến tháng 5 năm 2023, cuối cùng chọn 603 nghiên cứu, 59 nước, 70 địa điểm, 10355 quan sát. Giáo dục và tỷ lệ tử vong ở tuổi trưởng thành (tử vong tính chung mọi nguyên nhân) có quan hệ đáp lượng, đọc càng nhiều thì giảm càng nhiều. Mỗi năm học thêm, nguy cơ tử vong giảm trung bình 1.9% (95% khoảng không chắc chắn 1.8–2.0, giá trị thật nằm gần trong khoảng đó). Người từ 18 đến 49 tuổi mỗi năm thêm giảm 2.9% (2.8–3.0). Người trên 70 tuổi giảm 0.8% (0.6–1.0). Không thấy khác biệt giữa nam và nữ. Xét theo mức phát triển của từng nước (chỉ số dân số xã hội) cũng không thấy khác biệt +### 2. Cũng đưa「học có ích không」vào bảng tính tỷ lệ tử vong: mỗi thêm một năm học, nguy cơ tử vong ở người lớn giảm khoảng 1.9% + +- Chi phí: Không tốn tiền. Cái phải bỏ là thời gian và sức lực của mấy năm đi học thêm, phần học phí thì xem mục 4 của phần này (học phí và học bổng trường trung cấp chuyên nghiệp). Chỗ khó là phải học liên tục, không bỏ giữa chừng. +- Nói thẳng: Mỗi năm học thêm, nguy cơ tử vong sau khi trưởng thành trung bình thấp hơn khoảng 1.9%. Người 18 đến 49 tuổi rõ nhất, mỗi năm thêm thấp hơn khoảng 2.9%, người trên 70 tuổi chỉ thấp hơn 0.8%. Người học được liên tục thường vốn đã có điều kiện tốt, trong đó bao nhiêu là do chính việc học thì không phân biệt được. +- Lợi ích: Gom đủ các nghiên cứu trước đó rồi tính chung lại (tổng quan hệ thống và phân tích tổng hợp). Khoảng tra cứu từ tháng 1 năm 1980 đến tháng 5 năm 2023, cuối cùng đưa vào 603 nghiên cứu, 59 nước 70 địa điểm, 10355 quan sát. Giáo dục và tỷ lệ tử vong ở người lớn (cộng lại mọi nguyên nhân) có quan hệ liều – đáp ứng (học càng nhiều thì tử vong càng giảm). Mỗi năm học thêm, nguy cơ tử vong trung bình giảm 1.9% (95% khoảng không chắc chắn 1.8–2.0, giá trị thật nằm đại khái trong đoạn này). Người 18 đến 49 tuổi, mỗi năm thêm giảm 2.9% (2.8–3.0). Người trên 70 tuổi giảm 0.8% (0.6–1.0). Giữa nam và nữ không thấy khác biệt. Tách theo mức độ phát triển của quốc gia (chỉ số dân số và xã hội) cũng không thấy khác biệt - Mức bằng chứng: A (Tranh cãi) - Nguồn: IHME-CHAIN Collaborators (2024). Effects of education on adult mortality: a global systematic review and meta-analysis. The Lancet Public Health, 9(3), e155-e165. -- Ghi chú: Tranh cãi. Chính các tác giả cũng báo cáo thiên lệch công bố (p<0.0001), nghĩa là nghiên cứu cho kết quả đẹp dễ được đăng hơn. Kết quả giữa các nghiên cứu cũng chênh nhau khá xa. Các nghiên cứu được chọn đều là theo dõi quan sát, không phải thử nghiệm chia nhóm. Số năm đi học bị trộn lẫn với điều kiện gia đình, địa phương, nghề nghiệp, nên không tách được trong 1.9% đó bao nhiêu do chính việc học và bao nhiêu do「người đọc được suốt vốn đã có điều kiện tốt」. Vì vậy nó nói khá chắc rằng học nhiều hơn và sống lâu hơn là hai việc cùng hướng, chứ không có nghĩa bạn đi học là sẽ kéo dài 1.9%. Người hưởng lợi là bạn. +- Ghi chú: Tranh cãi. Chính tác giả cũng báo cáo thiên lệch công bố (p<0.0001), nghĩa là nghiên cứu có kết quả đẹp thì dễ được đăng hơn. Kết quả giữa các nghiên cứu cũng chênh nhau khá xa. Tất cả nghiên cứu đưa vào đều là theo dõi quan sát, không phải thử nghiệm chia người thành nhóm. Số năm học bị trộn lẫn với điều kiện gia đình, địa vị, nghề nghiệp, nên trong 1.9% đó có bao nhiêu do học, có bao nhiêu do「người học được liên tục vốn đã có điều kiện tốt」, loại nghiên cứu này không tách được. Chiều ngược lại cũng không tách được: người từ nhỏ đã bị bệnh, có vấn đề tâm lý thì dễ bỏ học và cũng dễ chết sớm, điều đó làm vai trò của việc học trông lớn hơn thực tế. Vì vậy nó chứng minh được khá vững rằng「học nhiều hơn và sống lâu hơn cùng chiều」, nhưng không đồng nghĩa với「bạn đi học là sống thêm được 1.9%」. Cỡ lợi ích để là nhỏ: người đọc thường chỉ đối mặt với chuyện học thêm ba bốn năm, tính 1.9% mỗi năm thì cộng lại chưa tới một phần mười, theo đường phân định của cả cuốn sách thì chỉ tính là nhỏ. Người hưởng lợi là bạn ### 3. Trước khi kết luận học vấn đã mất giá, xem cơ cấu học vấn cả nước: cứ 100.000 người thì chỉ 15467 người học trình độ đại học @@ -30,7 +29,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Cục Thống kê Quốc gia, Văn phòng Ban Chỉ đạo Tổng điều tra dân số lần thứ bảy của Hội đồng Nhà nước (2021). 第七次全国人口普查公报（第六号）——人口受教育情况. - Ghi chú: Người hưởng lợi là bạn. Các con số này chỉ nói học vấn có hiếm hay dày trong dân chúng. Với một người cụ thể thì đi học đáng không, phải tự tính theo cách làm ở mục 6 của phần này (học hay đi làm thì tính thế nào). Còn phải để ý: tổng điều tra đếm cả toàn bộ dân số, kể cả người già và trẻ em. Tỷ lệ người trẻ học tới đại học cao hơn con số này nhiều, so với người cùng tuổi sẽ thổi phóng lợi thế của mình. Ngược lại, dùng nó để bác câu「đi học vô ích」thì hợp lý, vì mẫu số chính là cả lực lượng lao động đang giành chỗ với bạn. Người hưởng lợi là bạn. - ### 4. Trường hợp「không đủ tiền để đi học」thì tính theo chính sách trước đã: học phí trường trung cấp chuyên nghiệp đã miễn phần lớn, học bổng 2300 tệ, học vay tối đa 2 vạn tệ một năm - Chi phí: Không tốn tiền. Trước khi khai giảng, hỏi một lần bộ phận tài trợ học sinh của trường, nộp một lần hồ sơ. @@ -39,7 +37,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Bộ Tài chính, Ủy ban Phát triển và Cải cách, Bộ Giáo dục, Bộ Nhân sự và Xã hội (2012). 关于扩大中等职业教育免学费政策范围 进一步完善国家助学金制度的意见（财教〔2012〕376 号）. ；Các cục chủ trì của Bộ Tài chính, Bộ Giáo dục, Bộ Nhân sự và Xã hội (2024). 就调整学生资助政策有关问题答记者问. ；Bộ Tài chính, Bộ Giáo dục, Ngân hàng Nhà nước, Cục Quản lý Tài chính (2024). 关于调整完善国家助学贷款有关政策的通知（财教〔2024〕188 号）. - Ghi chú: Người hưởng lợi là bạn và con cái. Miễn học phí cụ thể gồm những ai, học bổng chia mấy bậc, đều do tỉnh quy định. Trường tư thục được hỗ trợ theo mức trường công lập cùng chuyên ngành tại địa phương, phần chênh lệch vẫn phải tự trả, nên trước khi nhập học phải đối chiếu văn bản địa phương một lần. Học vay là khoản vay chứ không phải trợ cấp, ra trường phải trả lại. Nhưng nó tách「lúc này không đủ tiền học phí」và「mấy năm này có đáng không」thành hai việc riêng, đừng vội lấy thiếu tiền làm kết luận. Ở phần 7 đã viết về trợ cấp đào tạo và trợ cấp tạm thời trong thời gian thất nghiệp, nhà thật sự mất nguồn thu nhập thì hãy đi xin những khoản đó trước. - ### 5. Không đỗ được trường phổ thông không phải là hết đường: trường trung cấp chuyên nghiệp có tuyển sinh liên thông và thi riêng, tuyển dụng lao động kỹ thuật còn được hạ yêu cầu học vấn - Chi phí: Không tốn tiền. Trước khi điền nguyện vọng vào trường phổ thông, tra cổng tuyển sinh giáo dục nghề nghiệp của tỉnh một lần. @@ -48,6 +45,7 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2022). 职业教育法（2022 年修订，第三、三十七、五十三条）. （Bộ Giáo dục đăng lại） - Ghi chú: Người hưởng lợi là bạn và con cái. Đây là những lối đi theo luật, không phải bảo đảm trúng tuyển. Đào tạo liên thông gồm những ngành nào, tuyển sinh riêng chừa bao nhiêu tỷ lệ, kỳ thi nghề nghiệp học mấy môn, đều do tỉnh quy định. Trước khi điền nguyện vọng, nhất định phải tra cổng tuyển sinh giáo dục nghề nghiệp và chỉ tiêu tuyển sinh của tỉnh mình năm đó, đừng suy từ kinh nghiệm của tỉnh khác. Khi chọn ngành ở trường trung cấp chuyên nghiệp thì lượt qua một lần theo chiều của mục 10 của phần này (chọn kỹ năng chống bị máy thay): ngành phải tới hiện trường, môi trường làm việc không theo khuôn thì ít bị thay thế hơn. Chọn trường thì xem trường đó có nằm trong chỉ tiêu tuyển sinh giáo dục nghề nghiệp của tỉnh mình năm đó không. Giấy chứng chỉ trường cấp thì kiểm tra theo ba bước ở mục 8 của phần này (kiểm tra chứng chỉ có trong danh mục không). Ai quảng cáo「có chỗ làm chắc chắn」,「đảm bảo đỗ」thì đừng tin. + ### 6. Biến câu hỏi「học hay đi làm」thành một bài toán: ba năm lương kiếm được sớm, đổi lấy chênh lệch thu nhập mỗi năm của mấy chục năm sau - Chi phí: Không tốn tiền. Ngồi tính một lần, khoảng nửa tiếng. @@ -56,7 +54,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: B - Nguồn: Zhang J, Zhao Y, Park A, Song X (2005). Economic returns to schooling in urban China, 1988 to 2001. Journal of Comparative Economics, 33(4), 730-752. - Ghi chú: Trước khi tính, hãy trừ các khoản miễn học phí và học bổng nói ở mục 4 của phần này (học phí và học bổng trường trung cấp chuyên nghiệp), và tính thêm cái khoản ở mục 2 của phần này (một năm học thêm và tỷ lệ tử vong). Số liệu chỉ tới năm 2001, chỉ phủ vùng thành thị, không dùng thẳng cho hiện nay được. Mức trung bình toàn cầu 9% ở mục 7 của phần này cũng vậy. Tính thật theo cách này thì kết luận thay đổi theo từng người. Học phí cao, chuyên ngành không khớp với vị trí, số năm học lại dài, thì tính ra có thể thật sự không đáng, nhưng điều đó chỉ nói riêng chương trình đó không đáng, không nói học vô ích. Còn vài điều không đưa vào công thức được: ngưỡng học vấn theo luật, xem mục 5 của phần này (tuyển sinh liên thông và thi riêng của trường trung cấp chuyên nghiệp); khả năng chống bị thay thế, xem mục 10 của phần này (chọn kỹ năng chống bị máy thay); và khả năng đổi ngành. Người hưởng lợi là bạn. - ### 7. Ghi trước đường cơ sở: một năm học thêm, mức hoàn lại tư nhân trung bình toàn cầu (phần rơi vào thu nhập của chính mình) khoảng 9% mỗi năm - Chi phí: Không tốn tiền. Ghi nhớ con số này, coi nó là thước đo cho mọi phương án「học một chút」. @@ -65,7 +62,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Psacharopoulos G, Patrinos HA (2018). Returns to investment in education: a decennial review of the global literature. Education Economics, 26(5), 445-458. ；Bài làm việc nghiên cứu chính sách của Ngân hàng Thế giới 8402. - Ghi chú: Đây là mức trung bình toàn cầu của「đọc thêm một năm」, không phải mức lợi ích của「đăng một lớp」, cũng không phải con số cụ thể của Trung Quốc, chỉ dùng làm đại lượng tham chiếu. Công dụng của nó là cho mọi phương án đào tạo một thước đo: một khoá học bắt bạn bỏ ra 3 vạn tệ và nửa năm thì phải hỏi nó có tạo ra mức thay đổi thu nhập tương đương một năm giáo dục chính quy không. Còn phải để ý, mức trung bình toàn cầu che khuất chênh lệch rất lớn giữa ngành và giữa các nước. - ### 8. Trước khi trả tiền thi chứng chỉ, kiểm tra chứng chỉ đó có trong Danh mục chứng chỉ nghề nghiệp quốc gia hoặc trong danh sách cơ quan đánh giá đã đăng ký với Bộ Nhân sự và Xã hội không - Chi phí: Không tốn tiền. Kiểm tra một lần, mất mười mấy phút. @@ -74,7 +70,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Bộ Nhân sự và Xã hội (2022). 关于开展技术技能类「山寨证书」专项治理工作的通知. - Ghi chú: Kiểm tra theo đúng thứ tự này. Trước hết xem nghề đó có trong《国家职业资格目录》hay không, không có trong danh mục thì không phải chứng chỉ nghề nghiệp quốc gia. Rồi xem cơ quan cấp chứng chỉ có phải cơ quan đánh giá đã đăng ký với cơ quan nhân sự không. Cuối cùng vào nền tảng tra chứng chỉ đánh giá nhân tài kỹ năng chính thức, tra mã số chứng chỉ. Tra không ra thì đừng đăng ký. Loại nào có chữ「quốc tế」,「cao cấp」,「đăng ký」trong tên, hoặc hứa「đảm bảo đỗ」,「treo giấy phép có thu nhập」, thì rủi ro cao nhất. - ### 9. Đào tạo nên đi qua kênh trợ cấp của nhà nước trước, đừng vừa học vừa tự bỏ tiền đăng lớp thương mại - Chi phí: Không tốn tiền, hoặc giảm được rất nhiều. Bản thân khoá đào tạo thì vẫn tốn thời gian. Chỗ khó là nhịn được cái thói hứa điện thoại là bỏ tiền đăng lớp, thay vào đó tra danh mục trợ cấp một lần đã. @@ -83,7 +78,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Nhóm dẫn đầu xúc tiến việc làm và bảo vệ lao động của Hội đồng Nhà nước (2025). 关于开展大规模职业技能提升培训行动的指导意见. - Ghi chú: Cách nhận và nhận được bao nhiêu thì do từng tỉnh quy định. Cửa vào là mục「职业培训补贴目录」trên trang cơ quan nhân sự địa phương và cơ sở thực hành công cộng. Ở phần 7 đã viết về trợ cấp đào tạo nghề trong thời gian thất nghiệp (mỗi người tối đa được hưởng 3 lần) và trợ cấp bảo hiểm xã hội. Người đang làm việc thì hỏi kinh phí đào tạo người lao động của đơn vị trước. Tra danh mục trợ cấp rồi mới chọn học gì, tiết kiệm hơn nhiều so với chọn xong rồi đi tìm trợ cấp. - ### 10. Khi chọn kỹ năng, xem trước cái「có phải làm bằng tay, có phải phán đoán tại chỗ không」, loại này khó bị tự động hoá thay nhất - Chi phí: Không tốn tiền. Trước khi chọn, lượt việc bạn định học qua chiều này một lượt. @@ -92,7 +86,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: B - Nguồn: Frey CB, Osborne MA (2017). The future of employment: how susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, 254-280. - Ghi chú: Tranh cãi. Nghiên cứu này là một bản thảo bài báo năm 2013 chưa công bố chính thức, và điểm nó chấm là「cả nghề có bị máy đảm nhiệm được không」. Sau đó có nghiên cứu tách ra từng nhiệm vụ cụ thể để ước tính, kết quả tỷ lệ nguy cơ cao thấp hơn nhiều. Nghiên cứu cũng không nói chuyện này sẽ xảy ra vào năm nào. Dùng nó thì đừng nhìn con số 47%, hãy nhìn ba chỗ nó chỉ ra là máy vướng. Việc lắp đặt, sửa chữa, chăm sóc tại hiện trường chống bị thay hơn công việc giấy tờ thuần tuý, lý do nằm ở đó. Vì vậy khi「học một kỹ năng」thì nên chọn loại cần tới hiện trường, đối diện môi trường không theo khuôn. - ### 11. Trước khi học, xác nhận kỹ năng này có đánh giá theo cấp bậc không, cấp bậc có đổi ra tiền được không - Chi phí: Không tốn tiền. Chỉ cần tra một lần. @@ -101,7 +94,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: B - Nguồn: Nhóm dẫn đầu xúc tiến việc làm và bảo vệ lao động của Hội đồng Nhà nước (2025). 关于开展大规模职业技能提升培训行动的指导意见. ；Bộ Nhân sự và Xã hội (2021). 「技能中国行动」实施方案. - Ghi chú: Đây là hướng chính sách, không phải con số lương đã thực hiện được, mức độ thực thi khác nhau rất nhiều giữa các địa phương và doanh nghiệp. Phần làm theo được là: khi chọn kỹ năng thì xem nó có tiêu chuẩn nghề nghiệp quốc gia và lối đi theo bậc tương ứng không, có lối đi theo bậc thì sau này mới nói chuyện tăng lương theo bậc. Kỹ năng không có bậc, cũng không có ngưỡng vào nghề thì học xong chỉ có thể tự đi tìm việc. - ### 12. Đừng coi「trên mạng nói lương cao」là thiếu người, hãy tra danh mục nghề khan hiếm và danh mục trợ cấp tại địa phương - Chi phí: Không tốn tiền. Tra hai văn bản của địa phương. @@ -110,6 +102,7 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: C - Nguồn: Nhóm dẫn đầu xúc tiến việc làm và bảo vệ lao động của Hội đồng Nhà nước (2025). 关于开展大规模职业技能提升培训行动的指导意见. ；Cục Thống kê Quốc gia (2026). 2025年城镇单位就业人员年平均工资情况：「规模以上企业就业人员年平均工资为106080元，其中，中层及以上管理人员210016元，专业技术人员155491元，办事人员和有关人员94936元，社会生产服务和生活服务人员79857元，生产制造及有关人员80739元」. ；Trung tâm Hướng dẫn kỹ thuật đào tạo việc làm của Bộ Nhân sự và Xã hội (2023). 2022年第四季度全国招聘大于求职「最缺工」的100个职业排行（中国就业网）. - Ghi chú: Lương chính thức chia theo nghề chỉ chia tới năm nhóm vị trí lớn, không có nghề cụ thể. Số liệu năm 2025 của Cục Thống kê Quốc gia là doanh nghiệp quy mô trở lên có lương năm trung bình khoảng 10.6 vạn tệ (con số chính xác 106080 tệ). Quản lý cấp trung trở lên khoảng 21 vạn tệ (con số chính xác 210016 tệ), nhân viên kỹ thuật chuyên môn khoảng 15.5 vạn tệ (con số chính xác 155491 tệ), nhân viên văn phòng 94936 tệ, người sản xuất 80739 tệ, người dịch vụ sản xuất và sinh hoạt 79857 tệ. Bảng「100 nghề thiếu người nhất」của Bộ Nhân sự và Xã hội tra được trên Cổng việc làm Trung Quốc, kỳ mới nhất dừng ở quý bốn năm 2022, lập từ số liệu tuyển dụng và xin việc của các cơ quan dịch vụ việc làm công cộng ở 102 thành phố cả nước. Câu「nghề này kiếm được bao nhiêu một tháng」lan truyền trên mạng gần như đến từ giá niêm yết của nền tảng tuyển dụng hoặc lời tự kể của cá nhân. Phần bạn nhìn thấy phần lớn là nhóm đã kiếm được tiền, và không ai tính giờ làm, đường đi, bệnh nghề nghiệp, mùa cao mùa thấp cả. Muốn xem nhu cầu thật thì vào trang cơ quan nhân sự địa phương tra danh mục trợ cấp, rồi tra bảng xếp hạng nhu cầu vị trí của cơ quan dịch vụ việc làm công cộng. Cuối cùng hãy ra công trường, nhà máy, cửa hàng hỏi người đang làm việc. + ### 13. Với cùng số tiền và thời gian, ưu tiên chương trình ngắn hạn học xong là vào việc được ngay - Chi phí: Phải đóng học phí khoá ngắn hạn, số tiền không nhiều. Về thời gian thì phải bỏ ra một khoảng. Chỗ khó là khi học được nửa chừng chưa thấy kết quả thì dừng lại đúng điều kiện đã hẹn trước. @@ -118,7 +111,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: C - Nguồn: Nhóm dẫn đầu xúc tiến việc làm và bảo vệ lao động của Hội đồng Nhà nước (2025). 关于开展大规模职业技能提升培训行动的指导意见. - Ghi chú: Tiêu chí đánh giá chỉ một: học xong có một vị trí cụ thể để nộp hồ sơ không. Học cả năm, tốt nghiệp rồi còn phải「tích lũy thêm kinh nghiệm」thì chi phí thực tế cao hơn giá niêm yết nhiều. Điều kiện dừng mà phần 4 nói tới cũng áp dụng: trước khi bắt đầu, viết rõ「học tới mức nào mà vẫn chưa tìm được việc thì dừng」. - ### 14. Học xong thì đóng sách lại tự kiểm một lượt, đừng đọc lại từ đầu - Chi phí: Không tốn tiền, tổng thời gian cũng không phải tăng. Đổi khoảng thời gian trước đây dùng để đọc lại thành đóng sách viết ra hoặc làm bài. Chỗ khó là bài tự kiểm lộ ra ngay chỗ chưa nhớ, thấy nặng hơn đọc lại. @@ -127,7 +119,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Yang C, Luo L, Vadillo MA, Yu R, Shanks DR (2021). Testing (quizzing) boosts classroom learning: A systematic and meta-analytic review. Psychological Bulletin, 147(4), 399–435. ；Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 14(1), 4–58. - Ghi chú: Tự kiểm là để buộc bản thân bày ra thứ đang nằm trong đầu, không phải để chấm điểm. Cách làm: đọc xong một mục, đóng sách viết lại các ý chính, rồi lật trở lại đối chiếu. Có phản hồi sửa sai thì hiệu quả hơn, nên sau khi đối chiếu phải nhìn rõ sai ở đâu. Mức lợi ích ghi「lớn」là căn cứ cỡ hiệu quả và thang đánh giá. Hai tài liệu bằng chứng này đo điểm số, không đo số giờ, nên không áp ngưỡng tiết kiệm thời gian. Tự kiểm và mục 15 của phần này (chia cùng thời gian ra mấy ngày) là hai cách được xếp hạng cao nhất trong cùng một tổng quan đó, nên dùng kèm nhau. - ### 15. Chia cùng thời gian đó ra mấy ngày, đừng học một lạch xong - Chi phí: Không tốn tiền, tổng thời gian cũng không đổi. Chỉ có cách xếp lịch thay đổi: cùng 3 giờ đó, chia thành ba ngày mỗi ngày 1 giờ. Chỗ khó là phải bắt đầu trước vài ngày, không dồn đến đêm cuối cùng. @@ -136,7 +127,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Cepeda NJ, Pashler H, Vul E, Wixted JT, Rohrer D (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354–380. ；Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques. Psychological Science in the Public Interest, 14(1), 4–58. - Ghi chú: Dùng chung với mục 14 của phần này (đóng sách tự kiểm) thì tốt nhất: bản thân bài tự kiểm cũng chia ra mấy ngày làm. Phân tích tổng hợp của Cepeda chủ yếu dùng tư liệu nhớ danh sách từ, nên không đưa ra số ngày chung cho nội dung phức tạp. 「Càng thi xa thì khoảng cách càng dài」là quy tắc tác giả đưa ra, không kèm lịch cụ thể. Ôn dồn vào đêm trước khi thi thì hữu ích ngay tại chỗ, vấn đề là quên nhanh. - ### 16. Đừng coi gạch chân, đọc lại, viết tóm tắt là cách học chính - Chi phí: Không tốn tiền, ngược lại còn tiết kiệm thời gian. Chỗ khó là mấy cách đó dễ làm nhất, và cũng tạo cảm giác「tôi đang học」mạnh nhất. @@ -145,7 +135,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT (2013). Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychological Science in the Public Interest, 14(1), 4–58. - Ghi chú: Kẻ dòng vẫn dùng được để「đánh dấu chỗ lát nữa sẽ tự kiểm」, cái sai là coi việc kẻ dòng là học. Tổng quan cũng nói bậc thấp nhất không có nghĩa là bằng không, chỉ là điều kiện để có tác dụng hẹp và còn cần nhiều nghiên cứu hơn. Cách thay thế xem mục 14 của phần này (đóng sách tự kiểm) và mục 15 của phần này (chia cùng thời gian ra mấy ngày). - ### 17. Luyện xen kẽ nhiều dạng bài, đừng làm liên hai mươi bài cùng một dạng - Chi phí: Không tốn tiền. Thay vì làm bài tập theo thứ tự dạng, hãy làm theo thứ tự đảo. Chỗ khó là luyện xen kẽ lúc đó tỷ lệ đúng thấp hơn, có cảm giác như lùi. @@ -154,7 +143,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Brunmair M, Richter T (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators. Psychological Bulletin, 145(11), 1029–1052. - Ghi chú: Tổng quan dẫn ở mục 16 của phần này (đừng coi gạch chân là cách học chính) xếp luyện xen kẽ vào bậc trung bình, vì lúc đó số nghiên cứu chưa nhiều. Xem cả hai thì cách dùng là: nội dung cần「nhận ra đây là dạng nào, phải dùng cách nào」 thì luyện xen kẽ, chẳng hạn dạng bài toán, ca bệnh, áp dụng điều luật; nội dung cần học thuộc thì luyện theo khối, chẳng hạn từ vựng, định nghĩa. - ### 18. Đừng chọn cách học theo kiểu「tôi là người học bằng mắt, người kia là người học bằng tai」 - Chi phí: Không tốn tiền. Chỉ là không chọn tư liệu, không chọn khoá học theo cách phân loại đó nữa, và không bỏ tiền làm bài kiểm tra kiểu học nữa. @@ -163,6 +151,7 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Pashler H, McDaniel M, Rohrer D, Bjork R (2008). Learning Styles: Concepts and Evidence. Psychological Science in the Public Interest, 9(3), 105–119. - Ghi chú: Tổng quan đồng thời nói, điều đó không có nghĩa mọi phiên bản của lý thuyết phong cách học đều đã bị kiểm tra và bị bác bỏ, nhiều phiên bản chưa hề được kiểm tra. Cái cần tránh là bỏ tiền làm bài kiểm tra phong cách học và chọn lớp đào tạo theo phong cách. Tiêu chí chọn đào tạo xem mục 9 của phần này (đào tạo nên đi qua kênh trợ cấp của nhà nước) và mục 13 của phần này (ưu tiên chương trình ngắn hạn học xong là vào việc được ngay). + ### 19. Muốn học thuộc, hãy ra đề theo cách dùng về sau rồi tự kiểm tra, đừng chỉ đọc thuộc một lượt là xong - Chi phí: Không tốn tiền, cũng không tốn thêm thời gian. Chỉ có cách luyện là thay đổi: đổi「đọc thuộc một lượt」thành「ra đề theo cách dùng về sau rồi tự kiểm tra」. Khó ở chỗ phải nghĩ trước sau này dùng những thứ này vào việc gì. @@ -171,7 +160,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Pan SC, Rickard TC (2018). Transfer of test-enhanced learning: meta-analytic review and synthesis. Psychological Bulletin, 144(7), 710-756. - Ghi chú: Từ「học thuộc」thường gộp lẫn hai việc: đọc lại tư liệu (đọc đi đọc lại, gạch chân), và chủ động gọi lại khi đã đóng sách. Loại trước được xếp hạng thấp nhất trong tổng quan xếp hạng mười cách học thường dùng, muốn xem bản thân phần xếp hạng thì xem mục 16 (đừng coi gạch chân, đọc lại, viết tóm tắt là cách học chính). Cái cần làm là loại sau: chủ động gọi lại khi đã đóng sách, cụ thể xem mục 14 (đóng sách tự kiểm tra), và phải chia ra mấy ngày, xem mục 15 (chia ra mấy ngày). Cách làm là nghĩ trước sau này dùng vào việc gì: là phải chọn ra trên bài thi, hay phải phán đoán tại chỗ, rồi ra đề đúng kiểu đó để tự kiểm tra. Ngược lại, kiểm tra theo chiều đảo cũng đáng làm, vì trong phân tích tổng hợp này chiều đảo có chuyển đổi yếu nhất, cả hai chiều đều phải luyện tới. - ### 20. Xét chức danh nghề nghiệp: trước hết xem mình thuộc chuyên môn nào, bậc nào, rồi tìm cửa đăng ký theo tính chất đơn vị - Chi phí: Không tốn tiền. Tra thông báo đăng ký của năm đó ở tỉnh mình, khoảng nửa tiếng. @@ -180,7 +168,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Lệnh số 40 của Bộ Nhân sự và Xã hội (2019). 职称评审管理暂行规定. ：第二条、第十三条、第十四条、第十五条、第十六条、第十七条、第十八条；Văn phòng Bộ Nhân sự và Xã hội (2020). 关于进一步做好民营企业职称工作的通知. ：「在专业技术人才密集的创业孵化基地、高新技术开发区、科技园区等地设立职称申报受理服务点，或通过人才中介服务机构、工商联、行业协会商会、学会等社会组织受理民营企业专业技术人才职称申报」「民营企业专业技术人才被派驻外地连续工作一年以上的，可按有关规定在派驻地申报职称评审」；Tân Hoa Xã (2024). 职称评审进入高峰，信息如何查？有何新变化？ ：「我国共设有经济、工程、卫生、出版等27个职称系列」「符合条件的自由职业者，可根据属地原则申报参加当地人社部门组织的职称评审」；Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「不将论文作为评价应用型人才的限制性条件」「对职称外语和计算机应用能力考试不作统一要求」「符合条件的高技能人才，可参加工程系列专业技术人才职称评审」「公务员不得参加专业技术人才职称评审」 - Ghi chú: Cần chuẩn bị những tài liệu nào thì cả nước không có danh mục thống nhất. Mỗi chuyên môn, mỗi tỉnh hằng năm lại ra thông báo đăng ký, danh mục tài liệu, thời gian đăng ký và điều kiện thành tích đều ghi trong đó. Nên bước đầu tiên là tìm thông báo của năm đó trên trang Bộ Nhân sự và Xã hội của tỉnh. Luận án không còn bắt buộc phải có: ý kiến cải cách năm 2016 yêu cầu không lấy luận án làm ngưỡng cứng với nhân tài ứng dụng, kỳ thi ngoại ngữ chức danh và tin học cũng không có yêu cầu thống nhất. Người lao động kỹ thuật cũng có đường: lao động kỹ năng cao đủ điều kiện có thể dự đánh giá chức danh nghề nghiệp thuộc chuyên môn kỹ thuật. Công chức không được dự đánh giá chức danh nghề nghiệp. - ### 21. Chức danh sơ cấp, trung cấp ở kế toán thì lấy bằng kỳ thi thống nhất toàn quốc, đăng ký theo trình độ học vấn và số năm làm việc - Chi phí: Phí đăng ký trả theo tiêu chuẩn của từng tỉnh. Ôn thi thì chiếm mấy tháng thời gian ngoài giờ làm việc. @@ -189,7 +176,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「专业技术人才取得职业资格即可认定其具备相应系列和层级的职称，并可作为申报高一级职称的条件。初级、中级职称实行全国统一考试的专业不再进行相应的职称评审或认定。」；Bộ Nhân sự và Xã hội, Bộ Tài chính (2019). 关于深化会计人员职称制度改革的指导意见. ：「助理会计师、会计师实行全国统一的会计专业技术资格考试」「高级会计师采取考试与评审相结合方式，正高级会计师一般采取评审方式」；Bộ Tài chính và các cơ quan khác (2024). 关于做好会计专业学位与会计专业技术资格衔接有关工作的通知. - Ghi chú: Chuyên môn của bạn là thi, là xét đánh giá, hay kết hợp cả hai thì xem bản「Hướng dẫn để cải cách hệ thống chức danh nghề nghiệp ××」của chuyên môn đó. Bộ Nhân sự và Xã hội cùng các cơ quan chủ quản ngành đã ban hành theo từng chuyên môn, tra được trong kho chính sách của Cổng Chính phủ Trung Quốc. Thời gian đăng ký và các môn thi theo thông báo thi của năm đó. Công chức được dự kỳ thi kế toán, nhưng không được dự đánh giá chức danh nghề nghiệp kế toán. - ### 22. Đừng nhờ trung gian xét thay, đừng mua dịch vụ viết thay luận án, đừng làm giả hồ sơ: bị phát hiện thì huỷ chức danh và lưu hồ sơ lưu tín 3 năm - Chi phí: Không tốn tiền. Khó ở chỗ nhịn được quảng cáo「trả tiền là bảo đảm đỗ」. @@ -198,7 +184,6 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Mức bằng chứng: A - Nguồn: Lệnh số 40 của Bộ Nhân sự và Xã hội (2019). 职称评审管理暂行规定. ：第三十九条；Bộ Nhân sự và Xã hội (2024). 职称评审监管暂行办法. ：第五条、第十五条、第十七条、第二十七条；Văn phòng Bộ Nhân sự và Xã hội (2022). 关于进一步做好职称评审工作的通知. ：「未经备案的职称评审委员会，评审结果不纳入全国职称评审信息查询验证系统」；Tân Hoa Xã (2024). 职称评审进入高峰，信息如何查？有何新变化？ ：「大家可通过www.12333.gov.cn网页、手机12333客户端、人力资源社会保障部门户网站、“人力资源和社会保障部”微信公众号等多种渠道进行登录」 - Ghi chú: Có một cách phân biệt thật giả: chức danh nghề nghiệp do đánh giá đúng quy trình thì tra được trên nền tảng tra chức danh nghề nghiệp toàn quốc. Cửa vào là trang 12333, ứng dụng 12333 trên điện thoại, và trang Bộ Nhân sự và Xã hội. Chức danh do hội đồng xét đánh giá chưa đăng ký chấp nhận thì hệ thống này tra không ra. Nên hứa「bảo đảm đỗ」,「không cần bản thân tham dự」,「tra không thấy chứng chỉ vẫn dùng được」 thì về cơ bản có thể kết luận là có vấn đề. Người hưởng lợi chủ yếu là bạn, cỡ lợi ích được định theo「tránh bị huỷ chức danh và bị ghi mất uy tín 3 năm」. - ### 23. Có chức danh nghề nghiệp không có nghĩa lương tăng: trước hết hỏi rõ đơn vị xét theo tỷ lệ vị trí, hay xét rồi chưa chắc dùng - Chi phí: Không tốn tiền. Chỉ cần hỏi một lần bộ phận nhân sự của đơn vị. @@ -206,4 +191,4 @@ Phần này tính tiền và thời gian, riêng mục 2 tính thêm tỷ lệ t - Lợi ích: Điều 2 của Lệnh số 40 của Bộ Nhân sự và Xã hội: kết quả xét đánh giá chức danh nghề nghiệp là căn cứ quan trọng để sử dụng, đánh giá, thăng tiến nhân tài chuyên môn. Ý kiến cải cách năm 2016 quy định, cơ quan, đơn vị thực hiện đầy đủ quản lý theo vị trí thì thường xét chức danh nghề nghiệp trong tỷ lệ cơ cấu vị trí. Đơn vị không thực hiện quản lý theo vị trí, cùng chuyên môn phổ biến rộng và nghề mới, thì có thể tách việc xét và việc sử dụng. Khi nhân tài chuyên môn chuyển vùng, chuyển đơn vị thì chức danh nghề nghiệp được xét lại hoặc xác nhận theo thẩm quyền quản lý, trừ trường hợp nhà nước có quy định khác (toàn quốc) - Mức bằng chứng: B - Nguồn: Lệnh số 40 của Bộ Nhân sự và Xã hội (2019). 职称评审管理暂行规定. ：第二条、第二十九条；Văn phòng Trung ương Đảng Cộng sản Trung Quốc và Văn phòng Hội đồng Nhà nước (2016). 关于深化职称制度改革的意见. ：「对于全面实行岗位管理、专业技术人才学术技术水平与岗位职责密切相关的事业单位，一般应在岗位结构比例内开展职称评审。对于不实行岗位管理的单位，以及通用性强、广泛分布在各社会组织的职称系列和新兴职业，可采用评聘分开方式。」 -- Ghi chú: Mức bằng chứng đặt là B vì văn bản chỉ nói cách nối việc xét với việc sử dụng, không có con số thống nhất toàn quốc cho「được xét thì tăng bao nhiêu」. Phần làm theo được là: ở cơ quan, đơn vị và doanh nghiệp nhà nước, trước hết hỏi bộ phận nhân sự về số vị trí trống và quy tắc sử dụng của đơn vị mình. Ở doanh nghiệp tư nhân, trước hết hỏi người chủ xem họ có công nhận chứng chỉ này không, và có tăng lương không. \ No newline at end of file +- Ghi chú: Mức bằng chứng đặt là B vì văn bản chỉ nói cách nối việc xét với việc sử dụng, không có con số thống nhất toàn quốc cho「được xét thì tăng bao nhiêu」. Phần làm theo được là: ở cơ quan, đơn vị và doanh nghiệp nhà nước, trước hết hỏi bộ phận nhân sự về số vị trí trống và quy tắc sử dụng của đơn vị mình. Ở doanh nghiệp tư nhân, trước hết hỏi người chủ xem họ có công nhận chứng chỉ này không, và có tăng lương không. diff --git "a/vi/book/24-\347\234\213\347\227\205.md" "b/vi/book/24-\347\234\213\347\227\205.md" index 8fdef1d3..8a3dfc1a 100644 --- "a/vi/book/24-\347\234\213\347\227\205.md" +++ "b/vi/book/24-\347\234\213\347\227\205.md" @@ -6,23 +6,23 @@ Thước đo: tiền và thời gian. Phần này nói cách đi khám: vào b Phần này chủ yếu dựa vào《Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng》(关于加快建设分级诊疗体系的若干措施), do Văn phòng Hội đồng Nhà nước (国务院办公厅) ban hành tháng 4 năm 2026. Chẩn đoán và điều trị phân tầng nghĩa là căn cứ bệnh nặng hay nhẹ để quyết định vào bệnh viện cấp nào trước. Văn bản này gắn thẳng「đi tuyến cơ sở」với「tiết kiệm tiền」. Nhưng tiến độ thực hiện ở mỗi nơi khác nhau, phần chữ trong sách đưa ra quy định ở tầm quốc gia. Riêng với vùng gộp quỹ của bạn, cứ theo cách nói của cơ quan bảo hiểm y tế và cơ quan y tế địa phương. Vùng gộp quỹ là nơi bạn tham gia bảo hiểm y tế, thường là một thành phố. -### 1. Bệnh thường ngày xem ở tuyến cơ sở trước, chuyển tuyến lên từng bậc, mức đồng chi trả nội trú vẫn tính tiếp +### 1. Bệnh thông thường thì khám ở cộng đồng trước, đi theo chuyển tuyến từ tuyến cơ sở lên trên, mức đồng chi trả nằm viện tính nối tiếp -- Chi phí: Không tốn tiền. Phải đến tuyến cơ sở một lần trước. Chỗ khó là kiềm chế không đi lấy số khám chuyên gia ngay. -- Nói thẳng: Đi tuyến cơ sở trước, rồi từ đó chuyển lên bệnh viện lớn để nằm viện, phần tiền bạn tự trả chỉ phải ra một lần. Tự đi thẳng bệnh viện cấp ba nằm viện rồi sau đó chuyển về tuyến cơ sở tập luyện phục hồi, phần đó có thể phải trả hai lần. Được bệnh viện lớn chuyển về tuyến cơ sở nằm tiếp thì cùng một bệnh trong suốt khoảng thời gian ấy cũng không tính lại lần nữa. -- Lợi ích: Bệnh nhân có bảo hiểm y tế được chuyển tuyến từng bậc qua tuyến cơ sở thì mức đồng chi trả nội trú ở bệnh viện tuyến trên được tính liên tục. Mức đồng chi trả là phần mỗi lần nẩm viện bạn tự trả trước cho đến khi bảo hiểm y tế mới bắt đầu hoàn. Tính liên tục nghĩa là hai bệnh viện trước sau gộp lại chỉ tính một lần, không phải trả lại từ đầu. Bệnh nhân nằm viện được bệnh viện tuyến trên chuyển ngược xuống tuyến cơ sở thì trong cùng một chu kỳ bệnh không còn đặt mức đồng chi trả riêng nữa +- Chi phí: Không tốn tiền. Phải đi một chuyến tới trạm y tế cộng đồng. Khó là nhịn không đi thẳng để lấy số khám chuyên gia. +- Nói thẳng: Đi tới trạm y tế cộng đồng trước, rồi từ đó chuyển lên bệnh viện lớn để nằm viện, đoạn tiền bạn tự trả chỉ tính một lần. Tự chạy thẳng tới bệnh viện hạng A để nằm viện, sau đó chuyển về trạm y tế cộng đồng để phục hồi chức năng thì đoạn tiền đó có thể phải trả hai lần. Người từ bệnh viện lớn chuyển về trạm y tế cộng đồng nằm tiếp thì trong thời gian đó, cùng một bệnh đó không tính lại lần nữa. +- Lợi ích: Bệnh nhân có tham gia bảo hiểm y tế đi theo chuyển tuyến từ tuyến cơ sở thì mức đồng chi trả khi nằm viện ở bệnh viện tuyến trên được tính liên tục. Mức đồng chi trả là đoạn tiền mỗi lần nằm viện bạn phải tự trả trước, trả đủ rồi bảo hiểm y tế mới bắt đầu hoàn. Tính liên tục nghĩa là hai bệnh viện trước sau cộng lại chỉ tính một lần, không phải tự trả lại từ đầu. Bệnh nhân nằm viện được chuyển từ bệnh viện tuyến trên xuống lại tuyến cơ sở thì trong cùng một chu kỳ bệnh không còn đặt mức đồng chi trả riêng nữa - Mức bằng chứng: A -- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「统筹地区内经基层医疗卫生机构逐级转诊的参保患者，在上级医院的住院起付线可连续计算；由上级医院下转至基层医疗卫生机构的住院患者，同一疾病周期内不再另设住院起付线。」 -- Ghi chú: Tự đi thẳng bệnh viện cấp ba nằm viện rồi chuyển về tuyến cơ sở tập luyện phục hồi, phần tiền đó có thể phải trả hai lần, đi theo chuyển tuyến thì chỉ tính một lần. Quy định chuyển tuyến do từng tỉnh, từng thành phố và từng liên kết y tế tự đặt ra, trước khi đi nên hỏi tuyến cơ sở trước. +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「统筹地区内经基层医疗卫生机构逐级转诊的参保患者，在上级医院的住院起付线可连续计算；由上级医院下转至基层医疗卫生机构的住院患者，同一疾病周期内不再另设住院起付线。」— bệnh nhân có tham gia bảo hiểm y tế đi theo chuyển tuyến từ cơ sở y tế tuyến cơ sở thì mức đồng chi trả nằm viện ở bệnh viện tuyến trên được tính liên tục; bệnh nhân nằm viện được chuyển từ bệnh viện tuyến trên xuống cơ sở y tế tuyến cơ sở thì trong cùng một chu kỳ bệnh không còn đặt riêng mức đồng chi trả nằm viện nữa. +- Ghi chú: Tự chạy thẳng tới bệnh viện hạng A để nằm viện, rồi chuyển về tuyến cơ sở để phục hồi chức năng, đoạn tiền đó có thể phải trả hai lần, đi theo chuyển tuyến thì chỉ tính một lần. Quy tắc chuyển tuyến do từng tỉnh, từng thành phố và từng liên kết y tế tự đặt ra, trước khi đi hỏi trạm y tế cộng đồng một câu. Bệnh mạn tính ở thời kỳ ổn định thì đừng đi bệnh viện lớn xin thuốc nữa: loại khám ngoại trị thường của bệnh viện lớn sẽ dần bị bỏ, trạm y tế cộng đồng một lần có thể kê 12 tuần thuốc, xem mục 5 ở phần 16 (trạm y tế cộng đồng một lần kê thuốc tới 12 tuần). -### 2. Trong cùng một vùng gộp quỹ (nơi tham gia bảo hiểm y tế), bệnh viện cấp càng thấp thì tỷ lệ hoàn càng cao, chênh nhau khoảng 10 điểm phần trăm +### 2. Trong cùng một vùng gộp quỹ (nơi tham gia bảo hiểm y tế), cấp bệnh viện càng thấp thì tỷ lệ hoàn càng cao, chênh nhau khoảng 10 điểm phần trăm -- Chi phí: Không tốn tiền. Chỗ khó là kiềm chế không chạy lên bệnh viện lớn. -- Nói thẳng: Cùng một khoản viện phí, bệnh viện cấp càng thấp thì được hoàn càng nhiều. Các nơi phải kéo giãn khoảng cách này ra, mỗi bậc lên cao hơn chênh nhau khoảng mười điểm phần trăm. Việc có thể xong ở tuyến cơ sở mà chạy lên bệnh viện lớn thì cùng khoản chi phí đó bạn phải tự trả thêm khoảng một phần mười. Phần quỹ bảo hiểm y tế tăng thêm trong năm cũng nghiêng về tuyến cơ sở. -- Lợi ích: Văn bản yêu cầu các nơi kéo giãn mức hoàn phí nội trú giữa các cơ sở y tế khác cấp, về nguyên tắc mỗi bậc chênh nhau khoảng 10 điểm phần trăm. Với khoảng cách đó, cùng một khoản viện phí mỗi bậc bệnh viện lên cao hơn, bạn phải tự trả thêm khoảng một phần mười. Quỹ bảo hiểm y tế tăng thêm trong năm cũng nghiêng về tuyến cơ sở +- Chi phí: Không tốn tiền. Khó là nhịn không chạy thẳng tới bệnh viện lớn. +- Nói thẳng: Với một khoản chi phí nằm viện, bệnh viện cấp thấp hơn thì được hoàn nhiều hơn. Các nơi phải kéo dài khoảng cách, mỗi bậc lên trên chênh nhau khoảng mười điểm phần trăm. Ví dụ chi phí nằm viện một vạn tệ, tỷ lệ hoàn giảm từ tám phần mười xuống bảy phần mười thì số tiền tự trả đi từ 2000 tệ lên 3000 tệ. Phần quỹ bảo hiểm y tế tăng thêm cũng nghiêng về tuyến cơ sở. +- Lợi ích: Văn bản yêu cầu các nơi kéo rộng mức hoàn chi phí nằm viện giữa các cơ sở y tế khác cấp, về nguyên tắc mỗi bậc chênh nhau khoảng 10 điểm phần trăm. Với khoảng chênh đó, cùng một khoản chi phí nằm viện mỗi khi lên một bậc bệnh viện, phần bạn phải tự trả thêm ra khoảng 10% số tiền được hoàn của khoản đó. Ví dụ chi phí nằm viện được hoàn là 1 vạn tệ, tỷ lệ hoàn giảm từ 80% xuống 70%, số tiền tự trả đi từ 2000 tệ lên 3000 tệ. 1000 tệ phải trả thêm bằng một phần mười tổng chi phí, nhưng nhiều hơn một nửa số tiền bạn tự trả trước đó. Tỷ lệ thực tế ở mỗi nơi khác nhau, mức đồng chi trả và mức trần cũng làm khoảng chênh thay đổi, ở đây chỉ là ví dụ tính theo 10 điểm phần trăm của văn bản. Quỹ bảo hiểm y tế tăng thêm trong năm cũng hướng về tuyến cơ sở - Mức bằng chứng: A -- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「因地制宜适当拉开参保人员在不同等级医疗卫生机构的住院报销水平，原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距。」「具备条件的地方，可按分级诊疗导向酌情拉开参保人员在不同等级医疗卫生机构的门诊报销水平。」 -- Ghi chú: Cùng một khoản viện phí, ở tuyến cơ sở và ở bệnh viện cấp ba, phần bạn tự trả có thể chênh nhau vài nghìn tệ. Bệnh nào xong được ở tuyến cơ sở thì đừng chạy lên, và không phải bệnh nào cũng cần xem ở tuyến cơ sở. +- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「因地制宜适当拉开参保人员在不同等级医疗卫生机构的住院报销水平，原则上统筹地区内医疗卫生机构住院报销比例逐级拉开10个百分点左右的差距。」「具备条件的地方，可按分级诊疗导向酌情拉开参保人员在不同等级医疗卫生机构的门诊报销水平。」— tuỳ nơi nào mà khoảng cách mức hoàn chi phí nằm viện cho người tham gia bảo hiểm ở các cơ sở y tế khác cấp, về nguyên tắc trong vùng gộp quỹ thì tỷ lệ hoàn chi phí nằm viện kéo dài khoảng 10 điểm phần trăm mỗi bậc.「Nơi có điều kiện thì có thể theo định hướng chẩn đoán phân tầng mà khoảng cách mức hoàn khám ngoại trú cho người tham gia bảo hiểm ở các cơ sở y tế khác cấp được điều chỉnh cho phù hợp.」 +- Ghi chú: Cùng một khoản chi phí nằm viện, ở tuyến cơ sở và ở bệnh viện hạng A, số tiền tự trả chênh nhau có thể mấy nghìn tệ. Bệnh nào giải quyết xong được ở tuyến cơ sở thì đừng chạy lên trên, không phải bệnh nào cũng cần khám ở cộng đồng. ### 3. Muốn đi bệnh viện lớn, đi theo chuyển tuyến từ tuyến cơ sở hoặc trung tâm chuyển tuyến của bệnh viện, đừng tìm kẻ bán số khám @@ -32,7 +32,6 @@ Phần này chủ yếu dựa vào《Một số biện pháp nhằm đẩy nhanh - Mức bằng chứng: A - Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「牵头医院要为基层医疗卫生机构预留一定比例的号源和床位，并及时接诊经基层转诊患者。」「医疗机构要强化转诊服务统一管理，设立转诊中心或指定固定部门承担患者转诊服务工作，到2027年实现全覆盖。」 - Ghi chú: Số khám dành riêng là dành cho bệnh nhân chuyển tuyến, không giành được số khám thì đi theo chuyển tuyến. Tìm kẻ bán số khám ngoài chuyện đắt, còn có thể khiến bạn rơi vào khoa không đúng chuyên. Không tìm thấy trung tâm chuyển tuyến thì gọi đường dây dịch vụ của bệnh viện hỏi. - ### 4. Trước khi đi khám chữa bệnh ở tỉnh khác, hỏi một câu ở nơi bạn đang sống, tính cần thiết về nguyên tắc phải có bác sĩ phó trưởng khoa đánh giá - Chi phí: Không tốn tiền. Hỏi thêm một câu trước khi lên đường. @@ -41,17 +40,7 @@ Phần này chủ yếu dựa vào《Một số biện pháp nhằm đẩy nhanh - Mức bằng chứng: A - Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「原则上由二、三级医院副主任医师及以上职称人员评估患者跨统筹地区、跨省异地就医的必要性。」「进一步强化异地就医直接结算管理服务，跨省临时外出就医人员报销水平与参保地同级别医疗机构报销水平保持合理差异。」 - Ghi chú: 「临时外出就医」 và「长期异地居住」 là hai kiểu khai báo khác nhau, mức hoàn cũng khác nhau. Cách khai báo thế nào, những loại bệnh nào được thanh toán trực tiếp, xem phần 16. Đừng chạy tới nơi khác rồi mới nghĩ tới chuyện hoàn. - -### 5. Đừng còn lấy bệnh viện cấp ba làm chỗ lấy thuốc, bệnh mạn tính ở thời kỳ ổn định thì quay về tuyến cơ sở - -- Chi phí: Không tốn tiền. Chỗ khó là bỏ được thói quen「chỉ tin bệnh viện lớn」. -- Nói thẳng: Bệnh viện cấp ba về sau chủ yếu lo bệnh cấp nguy kịch và bệnh khó chẩn đoán. Khám lại bệnh thường ngày, và khám ngoại trú thường cho bệnh mạn tính đã ổn định, sẽ bị giảm dần. Khám ngoại trú thường của cùng một đội chuyên gia sẽ chuyển xuống mở ở tuyến cộng đồng. Còn lấy bệnh viện lớn làm chỗ lấy thuốc thì sau này càng khó lấy số khám. -- Lợi ích: Bệnh viện cấp ba phải tập trung vào bệnh cấp nguy kịch và bệnh khó, phức tạp, giảm dần khám ngoại trú thường cho bệnh thường ngày tái khám và bệnh mạn tính đã chẩn đoán rõ, ổn định. Số khám của loại khám ngoại trú thường ấy sẽ bị thu dần tuỳ tình hình. Bệnh viện tuyến trên sẽ kéo dài khám ngoại trú thường của đội chuyên gia xuống tuyến cơ sở, khám ngoại trú thường của cùng một đội chuyên gia sẽ mở ở tuyến cộng đồng -- Mức bằng chứng: A -- Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2026). Một số biện pháp nhằm đẩy nhanh xây dựng hệ thống chẩn đoán và điều trị phân tầng, 《关于加快建设分级诊疗体系的若干措施》. : 「三级医院要聚焦急危重症和疑难复杂疾病，加强转诊会诊和住院服务，逐步酌减常见病复诊和诊断明确、病情稳定的慢性病等普通门诊。」「紧密型医联体内上级医院要在基层医疗卫生机构开设高血压、糖尿病、慢性阻塞性肺疾病等常见病、慢性病门诊，提供诊疗和健康管理服务，并将专家团队普通门诊向基层医疗卫生机构延伸。」 -- Ghi chú: Đây là hướng chính sách, không phải lời khuyên. Sau này khám ngoại trú thường ở bệnh viện lớn sẽ càng khó lấy số. Bệnh mạn tính ở thời kỳ ổn định, ở tuyến cơ sở có thể kê thuốc một lần đủ 12 tuần, xem phần 16. - -### 6. Mỗi lần khám xong, tự giữ lại một bộ hồ sơ bệnh án, kết quả xét nghiệm và ảnh +### 5. Mỗi lần khám xong, tự giữ lại một bộ hồ sơ bệnh án, kết quả xét nghiệm và ảnh - Chi phí: Không tốn tiền. Bệnh viện được thu phí sao chép theo quy định, mức phí phải công khai. Chỗ khó là nhớ đi sao chép sau mỗi lần khám. - Nói thẳng: Quyền tra cứu và sao chép hồ sơ bệnh án của chính bạn là quyền mà luật trao cho bạn. Hồ sơ khám ngoại trú, sổ nhập viện, phiếu xét nghiệm, ảnh, hồ sơ phẫu thuật và gây mê, tư liệu bệnh lý, bảng chi phí đều sao chép được. Bệnh viện có một cửa sổ riêng tiếp nhận, chỉ được thu phí sao chép. Đổi bệnh viện, tìm bác sĩ khác xem lại, xin bảo hiểm thương mại, khi thật sự có tranh chấp, đều dựa vào bộ hồ sơ này. @@ -59,16 +48,16 @@ Phần này chủ yếu dựa vào《Một số biện pháp nhằm đẩy nhanh - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2018). Quy chế phòng ngừa và xử lý tranh chấp y tế（国务院令第 701 号）Điều mười sáu: 「患者有权查阅、复制其门诊病历、住院志、体温单、医嘱单、化验单（检验报告）、医学影像检查资料、特殊检查同意书、手术同意书、手术及麻醉记录、病理资料、护理记录、医疗费用以及国务院卫生主管部门规定的其他属于病历的全部资料。」「患者要求复制病历资料的，医疗机构应当提供复制服务，并在复制的病历资料上加盖证明印记……可以收取工本费，收费标准应当公开。」「患者死亡的，其近亲属可以依照本条例的规定，查阅、复制病历资料。」Điều mười lăm「任何单位和个人不得篡改、伪造、隐匿、毁灭或者抢夺病历资料。」；Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委), Cục Quản lý Y học cổ truyền quốc gia (国家中医药管理局) (2013). Quy định quản lý hồ sơ bệnh án của cơ sở y tế (bản 2013): cơ sở y tế「应当指定部门或者专（兼）职人员负责受理复制病历资料的申请」, việc sao chép thực hiện khi người xin có mặt và có đóng dấu chứng thực. - Ghi chú: Sao chép có cửa sổ riêng, không cần đi xin bác sĩ. Danh sách cần sao chép thì cứ theo Điều mười sáu của《Quy chế phòng ngừa và xử lý tranh chấp y tế》đọc. Trước là hồ sơ khám ngoại trú, sổ nhập viện, bảng nhiệt độ, phiếu y lệnh, phiếu xét nghiệm, tư liệu ảnh. Sau là hồ sơ phẫu thuật và gây mê, tư liệu bệnh lý, hồ sơ chăm sóc, bảng chi phí khám chữa bệnh. Lấy xong xếp vào một thư mục theo thứ tự thời gian. Trong đó quan trọng nhất là tóm tắt lúc ra viện, biên bản phẫu thuật, kết quả giải phẫu bệnh và ảnh then chốt. Chỉ số qua các lần của bệnh mạn tính thì ghi vào cùng một quyển sổ, xem phần 16. - -### 7. Có nghi ngờ về chẩn đoán và điều trị thì yêu cầu niêm phong hồ sơ ngay tại chỗ, hai bên cùng có mặt, lập biên bản, mỗi bên giữ một bản +### 6. Có nghi ngờ về chẩn đoán và điều trị thì yêu cầu niêm phong hồ sơ bệnh án ngay tại chỗ, hai bên cùng có mặt, lập danh mục, mỗi bên giữ một bản - Chi phí: Không tốn tiền. Nêu yêu cầu niêm phong với bệnh viện, rồi chờ cả hai bên đến làm thủ tục. Chỗ khó là nói được câu đó trước mặt bác sĩ. -- Nói thẳng: Có nghi ngờ về cách điều trị thì có thể yêu cầu niêm phong hồ sơ bệnh án. Việc niêm phong phải có cả bệnh nhân và cơ sở y tế cùng có mặt, lập biên bản, ký và đóng dấu, mỗi bên giữ một bản. Hồ sơ chưa viết xong thì niêm phong trước phần đã viết xong. Nghi ngờ do truyền dịch, truyền máu, tiêm, dùng thuốc thì túi dịch, ống thuốc và bộ truyền dịch cùng phải được niêm phong. -- Lợi ích: Quy chế quy định「发生医疗纠纷需要封存、启封病历资料的，应当在医患双方在场的情况下进行」. Phần được niêm phong có thể là bản gốc, cũng có thể là bản sao, do cơ sở y tế giữ. Hồ sơ chưa hoàn thành mà cần niêm phong thì niêm phong phần đã hoàn thành trước, phần làm sau thì niêm phong tiếp. Cơ sở y tế phải lập biên bản danh mục hồ sơ đã niêm phong, để hai bên ký hoặc đóng dấu, mỗi bên giữ một bản. Nghi ngờ truyền dịch, truyền máu, tiêm, dùng thuốc gây hậu quả xấu thì hai bên phải cùng niêm phong, mở niêm phong vật thể tại hiện trường. Cần xét nghiệm thì hai bên cùng giao cho một cơ sở có đủ điều kiện xét nghiệm. Không giao được thì cơ quan quản lý y tế của chính quyền cấp huyện nơi cơ sở y tế đang đặt trụ sở chỉ định. Nghi ngờ truyền máu gây hậu quả xấu mà cần niêm phong máu thì cơ sở y tế phải báo trạm máu cử người đến. Còn một ràng buộc cứng:「任何单位和个人不得篡改、伪造、隐匿、毁灭或者抢夺病历资料。」Sau khi niêm phong mà tranh chấp đã giải quyết, hoặc đã quá 3 năm không ai yêu cầu giải quyết nữa, cơ sở y tế có thể tự mở niêm phong (toàn quốc, áp dụng từ ngày 1 tháng 10 năm 2018) +- Nói thẳng: Có nghi ngờ về cách điều trị thì có thể yêu cầu niêm phong hồ sơ bệnh án. Khi niêm phong, bệnh nhân và cơ sở y tế phải cùng có mặt, lập danh mục, ký và đóng dấu, mỗi bên giữ một bản. Hồ sơ chưa viết xong thì niêm phong trước phần đã viết xong. Nghi ngờ do truyền dịch, truyền máu, tiêm, dùng thuốc thì túi dịch, ống thuốc và bộ truyền dịch cũng phải niêm phong cùng. +- Lợi ích: Quy chế quy định「发生医疗纠纷需要封存、启封病历资料的，应当在医患双方在场的情况下进行」— xảy ra tranh chấp y tế mà cần niêm phong, mở niêm phong hồ sơ bệnh án thì phải tiến hành khi cả bệnh nhân và cơ sở y tế đều có mặt. Phần được niêm phong có thể là bản gốc, cũng có thể là bản sao, do cơ sở y tế giữ. Hồ sơ chưa hoàn thành mà cần niêm phong thì niêm phong trước phần đã hoàn thành, phần làm sau thì niêm phong tiếp. Cơ sở y tế phải lập danh mục hồ sơ đã niêm phong, bệnh nhân và cơ sở y tế ký hoặc đóng dấu, mỗi bên giữ một bản. Nghi ngờ truyền dịch, truyền máu, tiêm, dùng thuốc gây hậu quả xấu thì hai bên phải cùng niêm phong, mở niêm phong vật thể tại hiện trường. Cần xét nghiệm thì hai bên phải cùng giao cho một cơ sở có đủ điều kiện xét nghiệm. Không giao được thì cơ quan quản lý y tế của chính quyền cấp huyện nơi cơ sở y tế đặt trụ sở chỉ định. Nghi ngờ truyền máu gây hậu quả xấu mà cần niêm phong máu thì cơ sở y tế phải báo trạm máu cử người đến. Còn một ràng buộc cứng:「任何单位和个人不得篡改、伪造、隐匿、毁灭或者抢夺病历资料。」— không đơn vị cá nhân nào được sửa, làm giả, giấu, huỷ diệt hay cướp hồ sơ bệnh án. Sau khi niêm phong mà tranh chấp đã giải quyết, hoặc đã đủ 3 năm mà không ai yêu cầu giải quyết nữa, cơ sở y tế có thể tự mở niêm phong (toàn quốc, áp dụng từ ngày 1 tháng 10 năm 2018) - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2018). Quy chế phòng ngừa và xử lý tranh chấp y tế（国务院令第 701 号，第十五、二十四、二十五条）. -- Ghi chú: Niêm phong và sao chép là hai việc khác nhau. Sao chép là việc nên làm khi bình thường, xem mục 6 (tự giữ hồ sơ và ảnh). Niêm phong là khi xảy ra tranh chấp thì ghim bằng chứng lại, để chống việc sau đó bổ sung viết hoặc sửa. Nghi ngờ do truyền dịch, truyền máu, dùng thuốc thì ngoài hồ sơ còn phải niêm phong cả túi dịch, ống thuốc và bộ truyền dịch, đừng để người ta dọn đi mà vứt bỏ. Cơ sở y tế phải công bố ở vị trí dễ thấy các đường giải quyết tranh chấp y tế, thủ tục và cách liên hệ, và lập một bộ phận tiếp nhận khiếu nại thống nhất. Có khiếu nại thì tìm bộ phận đó, đừng đi lý lẽ với bác sĩ trực. -### 8. Chấn thương nặng thì đi thẳng bàn tiền phân loại của khoa cấp cứu, đừng xếp hàng ở cửa sổ lấy số khám +- Ghi chú: Niêm phong và sao chép là hai việc khác nhau. Sao chép là việc nên làm khi bình thường, xem mục 5 ở phần này (tự giữ hồ sơ bệnh án và ảnh). Niêm phong là khi xảy ra tranh chấp thì ghim bằng chứng lại, chống việc sau đó bổ sung viết hoặc sửa. Nghi ngờ do truyền dịch, truyền máu, dùng thuốc thì ngoài hồ sơ bệnh án còn phải niêm phong cả túi dịch, ống thuốc và bộ truyền dịch, đừng để người ta dọn đi mà vứt bỏ. Cơ sở y tế phải công bố ở vị trí dễ thấy các đường giải quyết tranh chấp y tế, thủ tục và cách liên hệ, và lập một bộ phận tiếp nhận khiếu nại thống nhất. Có khiếu nại thì tìm bộ phận đó, đừng đi lý lẽ với bác sĩ trực. + +### 7. Chấn thương nặng thì đi thẳng bàn tiền phân loại của khoa cấp cứu, đừng xếp hàng ở cửa sổ lấy số khám - Chi phí: Không tốn tiền. Vào cửa đi thêm vài bước để tìm bàn tiền phân loại. Nói rõ bằng một câu triệu chứng nguy kịch nhất. - Nói thẳng: Cấp cứu không xếp theo thứ tự đến trước, mà xếp theo bệnh nặng nhẹ thành bốn bậc. Chỗ xếp bậc là bàn tiền phân loại ở cửa vào khoa cấp cứu, không phải cửa sổ lấy số khám. Không qua bàn phân loại thì trong hệ thống của bạn không có bậc nào, chấn thương nặng đến mấy cũng chỉ đi theo hàng chậm. Đến nơi hãy báo cáo trước triệu chứng nguy hiểm nhất: chảy máu nhiều, thở không ra, ý thức không tỉnh, trên người có vật đâm, đè nặng ở ngực. @@ -77,7 +66,7 @@ Phần này chủ yếu dựa vào《Một số biện pháp nhằm đẩy nhanh - Nguồn: Văn phòng Ủy ban Quốc gia Y tế (国家卫生健康委办公厅) (2024). Thông báo in chỉ tiêu chất lượng khám chữa bệnh cho 6 chuyên khoa trong đó có cấp cứu (bản 2024)（国卫办医政函〔2024〕150 号）, phụ lục 1, chỉ số bốn「急诊分级分诊执行率」. ；Ủy ban Y tế và Sức khoẻ thành phố Bắc Kinh (北京市卫生健康委员会) (2019). Phương án Bắc Kinh tăng cường tiền phân loại phân bậc ở khoa cấp cứu và phụ lục《Chuẩn phân loại phân bậc tiền kiểm ở khoa cấp cứu của các bệnh viện Bắc Kinh (thử nghiệm)》: 「坚持急危重症优先就诊原则。按照患者病情的严重程度，遵循从重到轻、从病情迅速变化到相对稳定的原则，合理安排患者就诊顺序。」「在急诊科入口的显著位置，设置急诊预检分诊区域」. - Ghi chú: Chia bốn bậc là thống nhất toàn quốc. Mỗi bậc phải chờ bao lâu thì địa phương tự quy định, bản của Bắc Kinh là bản có thể đối chiếu từng chữ. Xếp bậc dựa vào mạch, huyết áp, độ bão hòa oxy và triệu chứng, không dựa vào cảm giác đau của bạn. Chỉ tiêu cứng của bậc 1 có mấy điều. Một là nhịp tim lớn hơn 180 lần/phút, hoặc nhỏ hơn 40 lần/phút. Hai là huyết áp tâm thu thấp hơn 70 mmHg, tâm thu là chỉ số huyết áp cao. Ba là độ bão hòa oxy dưới 80%. Bốn là sốc, nhồi máu cơ tim xác định, rối loạn ý thức cấp. Thấy bàn phân loại xếp bạn thấp hơn thể nào thì bổ sung, nói rõ triệu chứng nguy hiểm nhất. Trong lúc chờ mà bệnh thay đổi thì lập tức quay lại bàn phân loại đánh giá lại. Khi thật sự nguy kịch thì gọi 120 còn nhanh hơn tự đến bệnh viện, bác sĩ ngoài bệnh viện sẽ xếp một bậc theo cùng bộ tiêu chí và báo trước bệnh viện để chuẩn bị tiếp đón (xem phần 13). -### 9. Không có tiền, không có giấy tờ, không nói rõ mình là ai, cấp cứu vẫn phải cứu trước +### 8. Không có tiền, không có giấy tờ, không nói rõ mình là ai, cấp cứu vẫn phải cứu trước - Chi phí: Không tốn tiền. - Nói thẳng: Nhà nước có một quỹ cứu hộ khẩn cấp khi bệnh, dùng riêng để trả chi phí cấp cứu cho hai loại người: người không xác định được danh tính, và người không đủ tiền trả. Bệnh viện không được từ chối hay trì hoãn cấp cứu, trạm cấp cứu cũng không được không xuất xe vì tiền. Vậy nên đừng ở nhà chịu đựng vì không có tiền, cũng đừng sợ nợ tiền mà không gọi xe cứu thương. Khoản tiền này chỉ lo phần cấp cứu, vào viện sau vẫn theo bảo hiểm y tế và trợ cấp y tế. @@ -85,29 +74,29 @@ Phần này chủ yếu dựa vào《Một số biện pháp nhằm đẩy nhanh - Mức bằng chứng: A - Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2013). Ý kiến chỉ đạo về việc xây dựng chế độ cứu hộ khẩn cấp khi bệnh（国办发〔2013〕15 号）. ；Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生和计划生育委员会) (2014). Biện pháp quản lý cấp cứu y tế ngoài bệnh viện（委令第 3 号，第十三条、第二十二条、第二十三条、第二十五条）. - Ghi chú: Quỹ chỉ trả phần「急救费用」, lại là cứu trước rồi sau đó cơ sở y tế mới xin quỹ, bạn không phải ra cửa sổ nhận tiền. Có bảo hiểm y tế thì vẫn dùng bảo hiểm y tế trước như bình thường. Thủ tục xin và hạn mức ở mỗi tỉnh khác nhau. Trong nguyên tắc chuyển bệnh,「就近、就急」đứng trước「患者意愿」, nên xe cứu thương không nhất thiết đưa bạn tới bệnh viện bạn chỉ định, lúc nguy kịch thì lấy bệnh viện gần nhất xử lý được là chính. Phần bạn tự trả về sau mà vẫn quá nặng thì làm trợ cấp y tế, xem mục 10 ở phần 7 (trợ cấp y tế). - -### 10. Giám định thương tật phải đợi điều trị kết thúc rồi hãy làm, làm sớm thì cấp độ bị chấp thấp +### 9. Giám định thương tật phải đợi điều trị kết thúc rồi hãy làm, làm sớm thì cấp độ bị chấp thấp - Chi phí: Phí giám định vài trăm đến hai nghìn tệ, thường do bên xin đơn trả trước. Phải đợi điều trị kết thúc, hoặc hiệu quả lâm sàng ổn định, thường là ba đến sáu tháng sau khi bị thương, có khi lâu hơn. Chỗ khó là lúc thương tích chưa lành đã muốn chốt sớm, phải kiềm chế. -- Nói thẳng: Đền bao nhiêu phần lớn tùy cấp độ được xếp, còn cấp độ thì tùy còn lại bao nhiêu tổn hương chức năng sau khi điều trị. Tiêu chuẩn ghi rõ phải làm sau khi điều trị kết thúc, hoặc khi hiệu quả điều trị lâm sàng ổn định. Xương chưa lành, đinh chưa lấy ra mà đi giám định thì người ta đánh giá đúng cái trạng thái lúc đó, tức là bạn tự ép cấp độ của mình xuống. Gãy xương thường đã lành hẳn mà không ảnh hưởng chức năng thì không được xếp cấp. -- Lợi ích: 《Phân cấp mức độ tàn tật do tổn thương cơ thể》quy định thời điểm giám định「应在原发性损伤及其与之确有关联的并发症治疗终结或者临床治疗效果稳定后进行鉴定」. Tiêu chuẩn「将人体损伤致残程度划分为 10 个等级，从一级（人体致残率 100%）到十级（人体致残率 10%），每级致残率相差 10%」. Căn cứ xếp là「人体组织器官结构破坏、功能障碍及其对医疗、护理的依赖程度，适当考虑由于残疾引起的社会交往和心理因素影响」. Trong mười cấp, các khoản liên quan tới gãy xương gồm mấy điều. Một là「一椎体压缩性骨折（压缩程度达 1/3）或者粉碎性骨折」. Hai là「骨盆两处以上骨折或者粉碎性骨折，畸形愈合」. Ba là「四肢任一大关节（踝关节除外）功能丧失 25% 以上」. Cùng một vị trí và cùng một tính chất tàn tật thì không tính trùng. Bị tàn tật từ hai chỗ trở lên thì phải ghi rõ từng cấp (toàn quốc, áp dụng từ ngày 1 tháng 1 năm 2017) +- Nói thẳng: Đền bao nhiêu phần lớn tùy cấp độ được xếp, còn cấp độ thì tùy còn lại bao nhiêu tổn thương chức năng sau khi điều trị. Tiêu chuẩn ghi rõ phải làm sau khi điều trị kết thúc, hoặc khi hiệu quả điều trị lâm sàng ổn định. Xương chưa lành, đinh chưa lấy ra mà đi giám định thì người ta đánh giá đúng cái trạng thái lúc đó, tức là bạn tự ép cấp độ của mình xuống. Gãy xương thường đã lành hẳn mà không ảnh hưởng chức năng thì không được xếp cấp. +- Lợi ích: 《Phân cấp mức độ tàn tật do tổn thương cơ thể》quy định thời điểm giám định là「应在原发性损伤及其与之确有关联的并发症治疗终结或者临床治疗效果稳定后进行鉴定」— phải giám định sau khi tổn thương gốc và biến chứng có liên quan chắc chắn với nó đã điều trị đến khi kết thúc, hoặc khi hiệu quả điều trị lâm sàng đã ổn định. Tiêu chuẩn「将人体损伤致残程度划分为 10 个等级，从一级（人体致残率 100%）到十级（人体致残率 10%），每级致残率相差 10%」— chia mức độ tàn tật do tổn thương cơ thể thành 10 cấp, từ cấp một (tỷ lệ tàn tật 100%) tới cấp mười (tỷ lệ tàn tật 10%), mỗi cấp chênh nhau 10%. Căn cứ đánh giá là「人体组织器官结构破坏、功能障碍及其对医疗、护理的依赖程度，适当考虑由于残疾引起的社会交往和心理因素影响」— mức phá huỷ cấu trúc mô cơ quan, rối loạn chức năng và mức phụ thuộc vào y tế chăm sóc, đồng thời có tính tới ảnh hưởng của việc tàn tật tới giao tiếp xã hội và yếu tố tâm lý. Trong mười cấp, các điều khoản liên quan tới gãy xương gồm mấy điều. Một là「一椎体压缩性骨折（压缩程度达 1/3）或者粉碎性骨折」— gãy xương đốt sống nén (mức nén tới 1/3) hoặc gãy vụn. Hai là「骨盆两处以上骨折或者粉碎性骨折，畸形愈合」— gãy xương chậu từ hai chỗ trở lên hoặc gãy vụn, lành biến dạng. Ba là「四肢任一大关节（踝关节除外）功能丧失 25% 以上」— mất trên 25% chức năng của bất kỳ khớp lớn nào ở tứ chi (trừ khớp cổ chân). Cùng một vị trí và cùng một tính chất tàn tật thì không tính trùng. Bị tàn tật từ hai chỗ trở lên thì phải ghi rõ từng cấp (toàn quốc, áp dụng từ ngày 1 tháng 1 năm 2017) - Mức bằng chứng: A -- Nguồn: Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao, Bộ Công an, Bộ An ninh quốc gia, Bộ Tư pháp (最高人民法院、最高人民检察院、公安部、国家安全部、司法部) (2016). Thông báo công bố《Phân cấp mức độ tàn tật do tổn thương cơ thể》(ban hành ngày 18 tháng 4 năm 2016, áp dụng từ ngày 1 tháng 1 năm 2017), điều 4.2, 4.4, 4.5, 5.10.6, 6.2 của tiêu chuẩn. -- Ghi chú: Ba bộ xếp tàn tật không dùng chung, đừng lẫn. Bị đánh, tai nạn giao thông và các tổn thương cá nhân khác thì dùng《Phân cấp mức độ tàn tật do tổn thương cơ thể》của giám định tư pháp, để tính tiền bồi thường tàn tật (xem phần 9). Tai nạn lao động thì làm khảo sát năng lực lao động, dùng bộ mười cấp tàn tật khác, và bộ đó quy định quyền lợi tai nạn lao động (xem phần 19). Muốn hưởng chính sách cho người khuyết tật thì phải xin riêng giấy chứng nhận người khuyết tật (xem mục 11, giấy chứng nhận người khuyết tật). Trước khi giám định, chuẩn bị đủ hồ sơ bệnh án, biên bản phẫu thuật, ảnh chụp lại (xem mục 6, tự giữ hồ sơ và ảnh), thiếu ảnh là nguyên nhân bị chấp cấp độ thấp nhất. Khi lần thương này lẫn với bệnh mạn tính sẵn có thì ý kiến giám định còn phải ghi rõ lần thương này chiếm bao nhiêu phần trong hậu quả tàn tật. Ô mục đó ảnh hưởng trực tiếp tới số tiền đền. +- Nguồn: Tòa án nhân dân tối cao, Viện kiểm sát nhân dân tối cao, Bộ Công an, Bộ An ninh quốc gia, Bộ Tư pháp (最高人民法院、最高人民检察院、公安部、国家安全部、司法部) (2016). Thông báo công bố《Phân cấp mức độ tàn tật do tổn thương cơ thể》（ban hành ngày 18 tháng 4 năm 2016, áp dụng từ ngày 1 tháng 1 năm 2017）, điều 4.2、4.4、4.5、5.10.6、6.2 của tiêu chuẩn. +- Ghi chú: Ba bộ xếp tàn tật không dùng chung, đừng lẫn. Bị đánh, tai nạn giao thông và các tổn thương cá nhân khác thì dùng《Phân cấp mức độ tàn tật do tổn thương cơ thể》của giám định tư pháp, để tính tiền bồi thường tàn tật (xem phần 9). Tai nạn lao động thì làm khảo sát năng lực lao động, dùng bộ mười cấp tàn tật khác, bộ đó quy định quyền lợi tai nạn lao động (xem phần 19). Muốn hưởng chính sách cho người khuyết tật thì phải xin riêng giấy chứng nhận người khuyết tật (xem mục 10 ở phần này, giấy chứng nhận người khuyết tật). Trước khi giám định, chuẩn bị đủ hồ sơ bệnh án, biên bản phẫu thuật, ảnh chụp lại (xem mục 5 ở phần này, tự giữ hồ sơ bệnh án và ảnh), thiếu ảnh là nguyên nhân bị chấp cấp độ thấp phổ biến nhất. Khi lần thương này lẫn với bệnh mạn tính sẵn có thì ý kiến giám định còn phải ghi rõ lần thương này chiếm bao nhiêu phần trong hậu quả tàn tật. Ô mục đó ảnh hưởng trực tiếp tới số tiền đền. -### 11. Sau khi điều trị xác nhận còn tổn hương chức năng, đến hiệp hội người khuyết tật cấp huyện nơi đăng ký hộ khẩu xin giấy chứng nhận người khuyết tật +### 10. Sau khi chữa xong mà thật sự còn tổn thương chức năng, đi tới hội liên hiệp người khuyết tật cấp huyện nơi có hộ khẩu để xin giấy chứng nhận người khuyết tật -- Chi phí: Bản giấy không thu phí sao chép. Phí đánh giá và phí ảnh về nguyên tắc do người xin tự trả, khó khăn thì xin giảm trừ. Mang theo giấy tờ tùy thân, sổ hộ khẩu và 3 tấm ảnh thẻ nền trắng cỡ hai inch đến hiệp hội người khuyết tật cấp huyện nơi đăng ký hộ khẩu. Kết quả đạt thì công bố tại thôn (cộng đồng) trong 5 ngày làm việc, hiệp hội xét xong trong 10 ngày làm việc. Chỗ khó là phải đi mấy chuyến và chờ khoảng thời gian ấy. -- Nói thẳng: Giấy chứng nhận người khuyết tật là ngưỡng để nhận quyền lợi của người khuyết tật, không phải giấy chứng minh mức độ thương tích. Chia thành bảy nhóm: thị lực, thính lực, ngôn ngữ, vận động, trí tuệ, tâm thần và đa dạng, mỗi nhóm từ cấp một đến cấp bốn. Quy trình là xin, tiếp nhận, đánh giá ở cơ sở được chỉ định, công bố, phê duyệt, cấp giấy, thời hạn mười năm, hết hạn đổi miễn phí. Không đạt thì là không đạt, đây không phải việc có thể nhờ người quen mà xin được. -- Lợi ích: Giấy chứng nhận người khuyết tật「是认定残疾人及其残疾类别、残疾等级的合法凭证，是残疾人依法享有国家和地方政府优惠政策的重要依据」. Tiêu chuẩn đánh giá là tiêu chuẩn quốc gia《Phân loại và phân cấp tàn tật》(GB/T 26341-2010). Làm giấy giữ nguyên tắc tự nguyện xin và quản lý theo địa phương. Cả bảy nhóm tàn tật về thị lực, thính lực, ngôn ngữ, vận động, trí tuệ, tâm thần và đa dạng đều có thể xin, cấp độ tàn tật từ một đến bốn. Hiệp hội người khuyết tật cấp huyện phụ trách tiếp nhận và cấp giấy, căn cứ kết luận đánh giá của bệnh viện hoặc cơ sở chuyên môn do tỉnh chỉ định. Kết luận đánh giá phù hợp tiêu chuẩn thì công bố tại thôn (cộng đồng) trong năm ngày làm việc, hiệp hội người khuyết tật cấp huyện「在十个工作日内审核完毕」.「办理残疾人证不收取工本费」. Phí đánh giá và phí ảnh「原则上由申请人个人自理；有条件的地方可由当地财政予以补贴，对特殊困难的申请人应协调有关部门予以减免」. Thời hạn của giấy là mười năm, hết hạn đổi miễn phí. Loại tàn tật hoặc cấp độ thay đổi thì có thể xin đánh giá lại. Có ý kiến khác về kết luận đánh giá thì trong mười ngày làm việc đến hiệp hội người khuyết tật cấp thành phố xin đánh giá lại. Vẫn còn ý kiến khác thì Ủy ban chuyên gia đánh giá tàn tật cấp tỉnh đưa ra kết luận cuối cùng (toàn quốc, áp dụng từ ngày 1 tháng 1 năm 2018) +- Chi phí: Giấy chứng nhận không thu phí cấp. Phí đánh giá và phí ảnh thì về nguyên tắc do người xin tự trả, hoàn cảnh khó thì có thể xin giảm. Mang theo giấy tờ tùy thân, sổ hộ khẩu và 3 tấm ảnh thẻ hai phân nền trắng tới hội liên hiệp người khuyết tật cấp huyện nơi có hộ khẩu. Kết quả đánh giá đạt thì công bố ở thôn (cộng đồng) trong 5 ngày làm việc, hội liên hiệp xem xét xong trong 10 ngày làm việc. Chỗ khó là phải chạy mấy chuyến và chờ khoảng thời gian đó. +- Nói thẳng: Giấy chứng nhận người khuyết tật là ngưỡng để nhận các chế độ dành cho người khuyết tật, không phải giấy chứng minh tình trạng thương tích. Chia thành bảy loại: thị giác, thính giác, ngôn ngữ, vận động, trí tuệ, tâm thần và đa khuyết tật, mỗi loại từ cấp một đến cấp bốn. Quy trình là xin, tiếp nhận, đánh giá ở cơ sở được chỉ định, công bố, xét duyệt, cấp giấy, giá trị mười năm, hết hạn đổi miễn phí. Không đạt thì là không đạt, đây không phải thứ tìm người quen mà xin được. +- Lợi ích: Giấy chứng nhận người khuyết tật「是认定残疾人及其残疾类别、残疾等级的合法凭证，是残疾人依法享有国家和地方政府优惠政策的重要依据」— là giấy tờ hợp pháp để xác định người khuyết tật cùng loại và cấp khuyết tật của họ, và là căn cứ quan trọng để người khuyết tật hưởng ưu đãi của Nhà nước và chính quyền địa phương theo luật. Tiêu chuẩn đánh giá là tiêu chuẩn quốc gia《Phân loại và xếp cấp khuyết tật》（GB/T 26341-2010）. Làm giấy giữ nguyên tắc tự nguyện xin và quản lý theo địa bàn. Người khuyết tật thuộc bảy loại thị giác, thính giác, ngôn ngữ, vận động, trí tuệ, tâm thần và đa khuyết tật đều có thể xin, cấp khuyết tật từ cấp một đến cấp bốn. Hội liên hiệp người khuyết tật cấp huyện phụ trách tiếp nhận và phát, căn cứ kết luận đánh giá của bệnh viện hoặc cơ sở chuyên môn do tỉnh chỉ định. Kết luận đánh giá đạt tiêu chuẩn thì công bố ở thôn (cộng đồng) trong năm ngày làm việc, hội liên hiệp người khuyết tật cấp huyện「在十个工作日内审核完毕」— xem xét xong trong mười ngày làm việc.「办理残疾人证不收取工本费」— làm giấy chứng nhận người khuyết tật không thu phí cấp. Phí đánh giá và phí ảnh「原则上由申请人个人自理；有条件的地方可由当地财政予以补贴，对特殊困难的申请人应协调有关部门予以减免」— về nguyên tắc do người xin tự bỏ; nơi có điều kiện thì ngân sách địa phương có thể trợ một phần, với người xin khó khăn đặc biệt phải phối hợp với các cơ quan liên quan để giảm. Giá trị giấy chứng nhận là mười năm, hết hạn thì đổi miễn phí. Loại hoặc cấp khuyết tật có thay đổi thì có thể xin đánh giá lại. Có ý kiến khác về kết luận đánh giá thì trong mười ngày làm việc có thể tới hội liên hiệp người khuyết tật cấp thành phố xin đánh giá lại. Vẫn còn ý kiến khác thì Ủy ban chuyên gia đánh giá khuyết tật cấp tỉnh đưa ra kết luận cuối cùng (toàn quốc, áp dụng từ ngày 1 tháng 1 năm 2018) - Mức bằng chứng: A -- Nguồn: Hiệp hội Người khuyết tật Trung Quốc (中国残疾人联合会), Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生和计划生育委员会) (2017). Biện pháp quản lý giấy chứng nhận người khuyết tật nước Cộng hoà nhân dân Trung Quốc（中华人民共和国残疾人证管理办法）(ban hành ngày 27 tháng 5 năm 2017, áp dụng từ ngày 1 tháng 1 năm 2018; Điều hai, ba, năm, bảy, chín, mười bảy, mười chín, hai mươi hai, hai mươi lăm). -- Ghi chú: Có làm hay không thì tính tiền trước đã. Bản giấy không trực tiếp phát tiền. Nó nối tới hai khoản trợ cấp cho người khuyết tật: người khuyết tật trong gia đình nhận trợ cấp thu nhập thấp nhận khoản trợ cấp sinh hoạt, người khuyết tật cấp một, cấp hai cần chăm sóc dài hạn nhận khoản trợ cấp chăm sóc, xem mục 8 ở phần 7 (hai khoản trợ cấp). Giấy còn nối tới các ưu đãi phục hồi chức năng, dụng cụ trợ giúp, thuế và phí, việc làm và giao thông ở các nơi. Tiêu chuẩn ở mỗi tỉnh khác nhau rất nhiều, hãy tra trên trang chính thức của hiệp hội người khuyết tật địa phương cho rõ rồi hãy quyết định có đi chuyến này không. Nó khác với cấp độ tàn tật của giám định tư pháp và với khảo sát năng lực lao động cho tai nạn lao động, đó là ba bộ riêng, không thay thế nhau (xem mục 10, giám định thương tật phải đợi điều trị kết thúc rồi hãy làm). Giấy chỉ dùng cho chính người đó, không cho mượn. Tình trạng tàn tật không còn phù hợp tiêu chuẩn thì bị thu hồi, sau khi thu hồi một năm không được xin lại. Người tàn tật đa dạng thì lấy loại nặng nhất trong các loại để xếp cấp. -### 12. Cảm ơn bác sĩ đã cứu bạn thì dùng thư cảm ơn, bằng khen và đánh giá mức hài lòng, đừng dùng bao lì xì: quy chuẩn cấm tài vật, không cấm lòng biết ơn +- Nguồn: Hội Liên hiệp Người khuyết tật Trung Quốc, Ủy ban Quốc gia Y tế và Kế hoạch gia đình (中国残疾人联合会、国家卫生和计划生育委员会) (2017). Quy chế quản lý giấy chứng nhận người khuyết tật của Cộng hoà Nhân dân Trung Quốc（ban hành ngày 27 tháng 5 năm 2017, áp dụng từ ngày 1 tháng 1 năm 2018；第二、三、五、七、九、十七、十九、二十二、二十五条）. +- Ghi chú: Chuyện làm hay không thì tính tiền trước đã. Giấy chứng nhận không phát tiền trực tiếp. Nó nối tới hai khoản trợ cấp dành cho người khuyết tật: người khuyết tật trong hộ gia đình được trợ cấp thu nhập thấp thì nhận trợ cấp sinh hoạt, người khuyết tật cấp một cấp hai cần chăm sóc lâu dài thì nhận trợ cấp chăm sóc, xem mục 8 ở phần 7 (hai khoản trợ cấp). Nó còn nối tới các ưu đãi về phục hồi chức năng, dụng cụ hỗ trợ, thuế, việc làm và giao thông ở các nơi. Các tiêu chuẩn này chênh nhau rất nhiều giữa các tỉnh, tra kỹ trên trang của hội liên hiệp người khuyết tật địa phương rồi hãy quyết định có chạy chuyến này không. Nó cùng cấp khuyết tật của giám định tư pháp, khảo sát năng lực lao động của tai nạn lao động là ba thứ khác nhau, không thay thế nhau (xem mục 9 ở phần này, giám định thương tật phải đợi điều trị kết thúc rồi hãy làm). Giấy chứng nhận chỉ dùng cho bản thân, không được đưa cho người khác. Tình trạng khuyết tật không còn đạt tiêu chuẩn thì bị thu hồi, trong một năm sau khi bị thu hồi thì không được xin lại. Khuyết tật đa thì lấy loại nặng nhất trong các loại để xếp cấp. + +### 11. Cảm ơn bác sĩ đã cứu bạn: dùng thư cảm ơn, băng đỏ tặng ơn và đánh giá mức độ hài lòng, đừng dùng phong bao; quy tắc cấm là tài sản, không phải lòng biết ơn -- Chi phí: Không tốn tiền. Viết một lá thư cảm ơn, hoặc điền một lần trên nền tảng đánh giá mức hài lòng, mười mấy phút. -- Nói thẳng: Chặn tiền trước không mua được quyền ưu tiên, số khám, giường bệnh, thuốc khan hiếm và lịch phẫu thuật đều bị cấm dùng để đổi lấy lợi ích. Sau khi chữa xong mà muốn cảm ơn bác sĩ cứu mạng thì cảm ơn được, nhưng đừng tặng thuốc lá, rượu hay bao lì xì, bác sĩ mà nhận là vi phạm. Có ba cách cảm ơn thật sự rơi vào tay bác sĩ: điền nghiêm túc một lần đánh giá mức hài lòng, viết thư cảm ơn có tên hoặc tặng một bằng khen, và tái khám đúng lời bác sĩ dặn. -- Lợi ích: 《Chín nguyên tắc về liêm chính khi làm việc của nhân viên cơ sở y tế》(国卫医发〔2021〕41 号) điều khoản thứ bảy: 「维护诊疗秩序，不破坏就医公平。坚持平等原则，共建公平就医环境。严禁利用号源、床源、紧缺药品耗材等医疗资源或者检查、手术等诊疗安排收受好处、损公肥私」. Điều khoản thứ tám:「共建和谐关系，不收受患方『红包』。恪守医德、严格自律」. Cùng điều khoản đó còn ghi「严禁索取或者收受患者及其亲友的礼品、礼金、消费卡和有价证券、股权、其他金融产品等财物」. Những bữa ăn, vui chơi do bên bệnh nhân trả tiền cũng vậy:「严禁参加其安排、组织或者支付费用的宴请或者旅游、健身、娱乐等活动安排」. Văn bản đồng thời yêu cầu cơ sở y tế thực hiện trách nhiệm chính, tiến hành giám sát và「坚决查处违规行为」. Đánh giá mức hài lòng đã được đưa vào công tác đánh giá hiệu quả của bệnh viện công lập cấp ba. Căn cứ là《Ý kiến của Văn phòng Hội đồng Nhà nước về tăng cường công tác đánh giá hiệu quả bệnh viện công lập cấp ba》(国办发〔2019〕4 号). Văn bản nói:「三级公立医院绩效考核指标体系由医疗质量、运营效率、持续发展、满意度评价等 4 个方面的指标构成」. Văn bản còn ghi:「医院满意度由患者满意度和医务人员满意度两部分组成……」. Sau đó nói tiếp, phải「通过门诊患者、住院患者和医务人员满意度评价，衡量患者获得感及医务人员积极性」. Bộ đánh giá này còn có nền tảng thống nhất toàn quốc:「国家建立公立医院满意度管理制度……」. Mốc thời gian là:「2019 年 3 月底前，全国三级公立医院全部纳入国家卫生健康委满意度调查平台。各地要应用国家卫生健康委满意度调查平台，将调查结果纳入三级公立医院绩效考核」. Bệnh viện còn phải quản lý khiếu nại. Căn cứ là《Biện pháp quản lý khiếu nại của cơ sở y tế》(国家卫生健康委令第 3 号, áp dụng từ năm 2019), Điều hai. Quản lý khiếu nại là việc bệnh nhân「提出意见、建议或者投诉请求」về hành vi khám chữa bệnh, quản lý y tế, an toàn chất lượng khám chữa bệnh. Bệnh viện phải chịu trách nhiệm điều tra, xử lý và phản hồi.「二级以上医疗机构应当设置医患关系办公室或者指定部门（以下统称投诉管理部门）统一承担投诉管理工作」. Nhiệm vụ của bộ phận tiếp nhận khiếu nại gồm「统一受理投诉，调查、核实投诉事项，提出处理意见，及时答复患者」. Cơ sở y tế「应当鼓励工作人员主动收集患者对医疗服务、医疗质量安全等方面的意见和建议」(toàn quốc) +- Chi phí: Không tốn tiền. Viết một thư cảm ơn, hoặc điền một lần trên nền tảng đánh giá mức độ hài lòng, mất mười mấy phút. +- Nói thẳng: Nhét tiền trước không mua được quyền ưu tiên: số khám, giường bệnh, thuốc khan hiếm và lịch mổ đều bị cấm dùng để đổi lấy lợi ích. Sau đó muốn cảm ơn bác sĩ đã cứu mạng bạn thì cảm ơn được, nhưng đừng tặng thuốc lá, rượu hay phong bao, bác sĩ mà nhận là vi phạm. Có ba cách cảm ơn thật sự rơi vào bác sĩ: điền nghiêm túc một lần đánh giá mức độ hài lòng, viết thư cảm ơn có ghi rõ tên hoặc tặng một tấm băng đỏ tặng ơn, tái khám theo đơn bác sĩ dặn. +- Lợi ích: 《Chín nguyên tắc làm việc liêm chính của nhân viên cơ sở y tế》（国卫医发〔2021〕41 号）hạng mục 7:「维护诊疗秩序，不破坏就医公平。坚持平等原则，共建公平就医环境。严禁利用号源、床源、紧缺药品耗材等医疗资源或者检查、手术等诊疗安排收受好处、损公肥私」— giữ trật tự chẩn đoán, không phá công bằng trong khám chữa bệnh, giữ nguyên tắc bình đẳng, cùng dựng môi trường khám chữa bệnh công bằng, tuyệt đối cấm nhận lợi ích bằng cách lợi dụng nguồn số khám, nguồn giường bệnh, thuốc vật tư khan hiếm hay tài nguyên y tế khác, hoặc lịch xét nghiệm phẫu thuật, để tư lợi. Hạng mục 8:「共建和谐关系，不收受患方『红包』。恪守医德、严格自律」— cùng xây dựng quan hệ hoà hợp, không nhận『phong bao』của bên bệnh nhân, giữ vững y đức nghề nghiệp, tự kiềm chế nghiêm. Cùng hạng mục đó còn ghi「严禁索取或者收受患者及其亲友的礼品、礼金、消费卡和有价证券、股权、其他金融产品等财物」— tuyệt đối cấm đòi hoặc nhận quà tặng, tiền mừng, thẻ tiêu và chứng khoán có giá trị, cổ phần, sản phẩm tài chính khác và các tài sản khác từ bệnh nhân cùng người thân của họ. Bữa ăn và vui chơi do bên bệnh nhân trả tiền cũng vậy:「严禁参加其安排、组织或者支付费用的宴请或者旅游、健身、娱乐等活动安排」— tuyệt đối cấm tham dự các cuộc tiệc, chuyến đi du lịch, tập thể dục, giải trí do họ sắp xếp, tổ chức hoặc trả tiền. Văn bản đồng thời yêu cầu cơ sở y tế thực hiện trách nhiệm chủ thể, tiến hành giám sát,「坚决查处违规行为」— kiên quyết xử lý hành vi vi phạm. Đánh giá mức độ hài lòng được đưa vào công tác đánh giá hiệu quả của bệnh viện công lập cấp ba. Căn cứ là《Ý kiến của Văn phòng Hội đồng Nhà nước về việc tăng cường công tác đánh giá hiệu quả bệnh viện công lập cấp ba》（国办发〔2019〕4 号）. Văn bản viết:「三级公立医院绩效考核指标体系由医疗质量、运营效率、持续发展、满意度评价等 4 个方面的指标构成」— hệ thống chỉ tiêu đánh giá hiệu quả của bệnh viện công lập cấp ba gồm chỉ tiêu của 4 phương diện: chất lượng khám chữa bệnh, hiệu quả vận hành, phát triển bền vững và đánh giá mức độ hài lòng. Văn bản còn ghi:「医院满意度由患者满意度和医务人员满意度两部分组成……」— mức độ hài lòng của bệnh viện gồm hai phần, hài lòng của bệnh nhân và hài lòng của nhân viên y tế. Rồi nói tiếp, phải「通过门诊患者、住院患者和医务人员满意度评价，衡量患者获得感及医务人员积极性」— thông qua đánh giá mức độ hài lòng của bệnh nhân ngoại trú, bệnh nhân nội trú và nhân viên y tế để đo cảm nhận của bệnh nhân và mức độ tích cực của nhân viên y tế. Bộ đánh giá này còn có nền tảng thống nhất trên cả nước:「国家建立公立医院满意度管理制度……」— Nhà nước xây dựng chế độ đánh giá mức độ hài lòng của bệnh viện công lập. Mốc thời gian là:「2019 年 3 月底前，全国三级公立医院全部纳入国家卫生健康委满意度调查平台。各地要应用国家卫生健康委满意度调查平台，将调查结果纳入三级公立医院绩效考核」— trước cuối tháng 3 năm 2019, toàn bộ bệnh viện công lập cấp ba trên cả nước phải vào nền tảng khảo sát mức độ hài lòng của Ủy ban Quốc gia Y tế và Sức khoẻ; các nơi phải dùng nền tảng khảo sát đó, đưa kết quả khảo sát vào công tác đánh giá hiệu quả của bệnh viện công lập cấp ba. Bệnh viện còn phải quản lý khiếu nại. Căn cứ là《Quy chế quản lý khiếu nại của cơ sở y tế》（国家卫生健康委令第 3 号，2019 年施行）điều 2. Quản lý khiếu nại nghĩa là bệnh nhân đưa ra「提出意见、建议或者投诉请求」về hành vi khám chữa bệnh, quản lý y tế, chất lượng an toàn khám chữa bệnh và các mặt khác. Bệnh viện phải chịu trách nhiệm điều tra, xử lý và phản hồi.「二级以上医疗机构应当设置医患关系办公室或者指定部门（以下统称投诉管理部门）统一承担投诉管理工作」— cơ sở y tế từ cấp hai trở lên phải lập văn phòng quan hệ bệnh nhân hoặc chỉ định một bộ phận (gọi chung là bộ phận tiếp nhận khiếu nại) để gánh việc quản lý khiếu nại. Nhiệm vụ của bộ phận này gồm「统一受理投诉，调查、核实投诉事项，提出处理意见，及时答复患者」— tiếp nhận khiếu nại, điều tra, xác minh nội dung khiếu nại, đưa ra ý kiến xử lý, trả lời bệnh nhân kịp thời. Cơ sở y tế「应当鼓励工作人员主动收集患者对医疗服务、医疗质量安全等方面的意见和建议」（全国）— nên khuyến khích nhân viên chủ động thu thập ý kiến và góp ý của bệnh nhân về khám chữa bệnh, chất lượng an toàn khám chữa bệnh và các mặt khác (toàn quốc) - Mức bằng chứng: A -- Nguồn: Ủy ban Quốc gia Y tế (国家卫生健康委), Cục Quản lý bảo hiểm y tế quốc gia (国家医保局), Cục Quản lý Y học cổ truyền quốc gia (国家中医药局) (2021). Thông báo in Chín nguyên tắc về liêm chính khi làm việc của nhân viên cơ sở y tế（国卫医发〔2021〕41 号）. Công báo của Hội đồng Nhà nước ；Văn phòng Hội đồng Nhà nước (国务院办公厅) (2019). Ý kiến về tăng cường công tác đánh giá hiệu quả bệnh viện công lập cấp ba（国办发〔2019〕4 号）. ；Ủy ban Quốc gia Y tế (国家卫生健康委) (2019). Biện pháp quản lý khiếu nại của cơ sở y tế（委令第 3 号）. -- Ghi chú: Người hưởng lợi là bạn, và cũng là người đã cứu bạn. Làm việc cứu người cả đêm rồi muốn cảm ơn là chuyện thường tình, cảm ơn được, chỉ cần đổi cách thức. Bao lì xì, thuốc lá và rượu đẩy rủi ro sang phía người nhận, còn thư cảm ơn, bằng khen và đánh giá mức hài lòng là những cách mà chế độ công nhận. Đã tặng mà bên nhận đẩy lại thì đừng hiểu là họ thấy ít hoặc làm bộ, đó là việc họ bắt buộc phải làm theo quy định. Ngoài ra cần phân biệt hai việc: tặng bao lì xì cho bác sĩ thuộc kỷ luật ngành và do bệnh viện quản lý, tặng tài vận cho công chức thụ lý và công chức thi hành thì là tội hối lộ theo Bộ luật Hình sự, hai việc không cùng cấp độ, cái sau xem mục 40 ở phần 8. Việc viết thư cảm ơn, tặng bằng khen thì bệnh viện thường nhận, nhưng câu「会进个人和科室记录」là kinh nghiệm thực tế, không có quy định nào viết thế. Có căn cứ trong văn bản là việc kết quả đánh giá mức hài lòng được đưa vào công tác đánh giá hiệu quả bệnh viện công lập cấp ba. Có nghi ngờ về bản thân việc khám chữa bệnh thì yêu cầu niêm phong hồ sơ ngay tại chỗ (xem mục 7 của phần này, niêm phong hồ sơ), đồng thời giữ lại hồ sơ và ảnh (mục 6, tự giữ hồ sơ và ảnh) \ No newline at end of file +- Nguồn: Ủy ban Quốc gia Y tế và Sức khoẻ, Cục Quản lý bảo hiểm y tế quốc gia, Cục Y học cổ truyền quốc gia (国家卫生健康委、国家医保局、国家中医药局) (2021). 关于印发医疗机构工作人员廉洁从业九项准则的通知（国卫医发〔2021〕41 号）. Công báo của Hội đồng Nhà nước ；Văn phòng Hội đồng Nhà nước (国务院办公厅) (2019). 关于加强三级公立医院绩效考核工作的意见（国办发〔2019〕4 号）. ；Ủy ban Quốc gia Y tế và Sức khoẻ (国家卫生健康委) (2019). 医疗机构投诉管理办法（委令第 3 号）. +- Ghi chú: Người hưởng lợi là chính bạn, và cả người đã cứu bạn. Thức trắng cứu hộ rồi muốn cảm ơn là chuyện con người, cảm ơn thì được, chỉ cần đổi hình thức thôi. Phong bao, thuốc lá và rượu là chuyển rủi ro sang cho phía bên kia; thư cảm ơn, băng đỏ tặng ơn và đánh giá mức độ hài lòng là những cách chế độ công nhận. Đã tặng mà bên kia từ chối rồi trả lại, đừng hiểu là họ chê ít hay dựng cao, đó là việc họ bắt buộc phải làm theo quy định. Ngoài ra cần phân biệt hai việc: tặng phong bao cho bác sĩ thuộc kỷ luật ngành và bệnh viện quản lý, tặng tài sản cho cán bộ xử án, chấp pháp thì là tội đưa hối theo Bộ luật Hình sự, hai việc không cùng cấp độ, cái sau xem mục 40 ở phần 8. Việc viết thư cảm ơn, tặng băng đỏ thì bệnh viện đón nhận phổ biến, nhưng câu「có ghi vào hồ sơ cá nhân và hồ sơ khoa」là kinh nghiệm nói, không có quy định văn bản nào. Có căn cứ văn bản là câu kết quả khảo sát mức độ hài lòng được đưa vào công tác đánh giá hiệu quả của bệnh viện công lập cấp ba. Có nghi ngờ về bản thân việc khám chữa bệnh thì yêu cầu niêm phong hồ sơ bệnh án ngay tại chỗ (xem mục 6 ở phần này, niêm phong hồ sơ bệnh án), và giữ lại hồ sơ bệnh án cùng ảnh (mục 5 ở phần này, tự giữ hồ sơ bệnh án và ảnh) diff --git "a/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" "b/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" index 9312ebfa..78002133 100644 --- "a/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" +++ "b/vi/book/25-\344\272\272\350\265\260\344\272\206\344\273\245\345\220\216\350\246\201\345\212\236\344\273\200\344\271\210.md" @@ -14,7 +14,6 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (1958). Quy chế đăng ký hộ khẩu nước Cộng hoà nhân dân Trung Quốc（中华人民共和国户口登记条例）(Điều tám). (đăng trên cổng thông tin điện tử của Chính phủ tỉnh Quảng Đông)；Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, áp dụng từ ngày 30 tháng 3 năm 2026, Điều hai mươi ba）. - Ghi chú: Nằm liệt giường lâu ngày, tử vong tại nhà thì tính là tử vong bình thường. Có thể tìm cơ sở đã khám cho người đó lúc còn sống để cấp giấy, cũng có thể tìm trạm y tế cộng đồng hoặc trạm y tế xã ở khu vực bạn ở. Giấy chứng minh tử vong thì ai cứu chữa người đó sẽ cấp, xem mục 2 (giấy chứng minh tử vong). Không phân biệt được thuộc loại nào thì cứ xử lý như tử vong bất thường, gọi 110 trước, cảnh sát tới nơi điều tra xong vẫn cấp được giấy. Nếu dọn sạch hiện trường rồi mới báo công an thì không ai giúp được bạn. Đừng tự di chuyển thi thể, đừng rửa và thay đồ cho người đó, cũng đừng cất đi lọ thuốc và mẩu giấy. - ### 2. Giấy chứng minh tử vong là chìa khoá cho mọi việc phía sau: ai cứu chữa thì người đó cấp, tử vong bình thường tại nhà thì tìm cơ sở y tế cộng đồng, ký trong một ngày - Chi phí: Không tốn tiền. @@ -23,7 +22,6 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Mức bằng chứng: A - Nguồn: Văn phòng Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委办公厅) (2014). Quy định quản lý thông tin tử vong dân số (thử nghiệm)（国卫办规划发〔2014〕68 号, Điều chín）. (Sở Dân vụ thành phố Thâm Quyến đăng lại)；Ủy ban Quốc gia Y tế, Bộ Công an, Bộ Dân vụ, Cục Quản lý Y học cổ truyền quốc gia, Cục Quản lý và Phòng ngừa bệnh quốc gia (2026). Thông báo về tăng cường đăng ký thông tin giấy chứng minh tử vong và quản lý giấy chứng nhận điện tử của dân chúng（国卫规划发〔2026〕5 号）. (Ủy ban Y tế và Sức khoẻ tỉnh Phúc Kiến đăng lại) - Ghi chú: Huỷ hộ khẩu, thiêu, rút quỹ tích góp nhà ở, nhận quyền lợi bảo hiểm xã hội, bồi hoàn bảo hiểm, làm thừa kế, việc nào cũng cần tờ giấy này. Lấy xong thì sao chép thêm vài bản, nơi nào cấp được giấy chứng nhận điện tử thì xin luôn. Chỉ có một lần để xin cấp lại, đừng đánh mất bản gốc. Tên hay số giấy tờ tùy thân bị sai thì sửa ngay tại chỗ, vì đến lúc làm thừa kế mới phát hiện thì mọi cửa sổ phải làm lại từ đầu. - ### 3. Vận chuyển, lưu giữ và thiêu thi thể chỉ làm được ở nhà tang lễ, kho lưu thi thể của bệnh viện không làm tang nghiệp, lưu giữ thường không quá 3 ngày - Chi phí: Không tốn tiền. @@ -32,7 +30,6 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, áp dụng từ ngày 30 tháng 3 năm 2026, Điều tám, chín, hai mươi tư, hai mươi lăm, hai mươi bảy, hai mươi tám, hai mươi chín）. - Ghi chú: Bọn「trọn gói」Ghé tới sau khi người mất ở bệnh viện thì phần lớn không phải người của nhà tang lễ. Bản thân việc vận chuyển thuộc mục cơ sở, có giá niêm yết, xem mục 6 (bảng giá mục cơ sở). Giấy chứng minh thiêu cất giữ kỹ, làm thừa kế, bảo hiểm, di dời tro cốt đều cần. Ở địa phương thiêu hay chôn đất do chính quyền cấp tỉnh quy định. Phong tục tang chế của dân tộc thiểu số được bảo vệ theo Điều chín của quy chế. Ai đáng hoả thiêu lại chôn đất, hoặc chôn, đắp mộ ngoài nghĩa trang và khu vực an táng sinh thái, thì cơ quan dân vụ sẽ yêu cầu sửa lại trong thời hạn (sửa lại cho đúng trong hạn). - ### 4. Bất đồng về nguyên nhân tử vong thì trong 48 giờ phải yêu cầu mổ xác, đồng thời niêm phong hồ sơ bệnh án - Chi phí: Không tốn tiền. Chỗ khó là lúc ấy còn nhớ ra mà yêu cầu. @@ -41,7 +38,6 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2018). Quy chế phòng ngừa và xử lý tranh chấp y tế（国务院令第 701 号, Điều hai mươi tư, hai mươi sáu）. - Ghi chú: Qua thời hạn rồi mới nói「lúc đầu tôi đã thấy không ổn」thì trách nhiệm lại rơi vào bên kéo dài. Thi thể đã thiêu thì không còn khả năng mổ xác, nên chuyện này phải chốt trước khi liên hệ nhà tang lễ. Cách niêm phong hồ sơ bệnh án xem phần 24. - ### 5. Huỷ đăng ký hộ khẩu có thời hạn theo luật: ở thành phố thì trước khi an táng, ở nông thôn thì trong một tháng - Chi phí: Không tốn tiền. Mang theo giấy chứng minh tử vong và sổ hộ khẩu, ra công an phường xí phong một chuyến. @@ -50,7 +46,6 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (1958). Quy chế đăng ký hộ khẩu nước Cộng hoà nhân dân Trung Quốc（中华人民共和国户口登记条例）(Điều tám). (đăng trên cổng thông tin điện tử của Chính phủ tỉnh Quảng Đông) - Ghi chú: Cứ để hộ khẩu còn nguyên thì chuyển nhượng thừa kế, nhận quyền lợi, rút bảo hiểm đều bị kẹt, các cửa sổ này đều xem tình trạng hộ khẩu. Lương hưu trí mà vẫn được phát thì đó là chiếm đoạt, sớm muộn cũng phải hoàn lại. Ra công an phường xí phóng tiện thể hỏi luôn: quyền lợi bảo hiểm xã hội và bảo hiểm y tế dừng thế nào, xin trợ cấp mai táng ra sao. Mấy khoản tiền đó xem mục 9 (số dư quỹ tích góp nhà ở, quyền lợi bảo hiểm xã hội). - ### 6. Dịch vụ tang chế chia thành mục cơ sở và mục ngoài cơ sở, mục cơ sở có bảng giá, mức thu do pháp luật quy định - Chi phí: Không tốn tiền. Trước khi lên đường, dành vài phút xem bảng giá. @@ -59,7 +54,6 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, được Hội đồng Nhà nước sửa đổi thông qua tại phiên họp thường vụ lần thứ 72 ngày 14 tháng 11 năm 2025,「自2026年3月30日起施行」）. Điều năm: 「殡葬服务分为基础项目和非基础项目，收费政策由国务院发展改革部门、财政部门会同国务院民政等部门制定。」「国家制定殡葬服务基础项目清单，将遗体接运、遗体存放、遗体告别、遗体火化、骨灰寄存、生态安葬以及政府举办的殡葬服务机构提供的骨灰格位安葬等纳入清单范围」；Điều năm mươi「殡葬服务实行清单化管理并动态调整，禁止在清单之外设立项目、收取费用。」 - Ghi chú: Tỉnh, khu tự trị, thành phố trực thuộc có thể thêm mục vào bảng của Nhà nước. Hãy hỏi rõ mục nào thuộc mục cơ sở, tiêu chuẩn thu ở địa phương là bao nhiêu, rồi hãy bàn phần còn lại. Giá mục trong bảng bị định theo quy định, ngoài bảng thì bạn chọn, phân biệt hai loại này thì phần bị thu thêm mới có căn cứ mà nói. Muốn làm tang nghiệm cho người thân cho ra dáng không có gì sai, cuốn sách này không khuyên bạn tiết kiệm, tiền tiêu ở đâu thì bạn tự quyết định. - ### 7. Gặp ép mua kèm, thêm tiền ngoài giá niêm yết, tách một dịch vụ thành mấy dịch vụ để thu tiền thì ghi lại và khiếu nại - Chi phí: Không tốn tiền. Chỗ khó là lúc ấy vẫn còn chịu làm rõ. @@ -68,6 +62,7 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, được Hội đồng Nhà nước sửa đổi thông qua tại phiên họp thường vụ lần thứ 72 ngày 14 tháng 11 năm 2025,「自2026年3月30日起施行」）. Điều bốn mươi hai: 「殡葬服务机构不得违反有关规定擅自设立收费项目，或者采取分解项目、扩大范围收费等形式变相提高收费标准。」Điều bốn mươi ba liệt kê hành vi vi phạm giá có「（一）不按规定明码标价，或者在标价之外加价或者收取未标明的费用」「（三）采取捆绑、附加不合理条件或者不合理限制等方式，强制或者变相强制销售丧葬用品、提供殡葬相关服务并收取费用」「（四）强制或者变相强制丧属接受第三方的有偿服务，或者利用相关服务网点重复收取费用」；Điều năm mươi một yêu cầu cơ quan dân vụ lập cơ chế chia sẻ thông tin giám sát với đường dây nóng phục vụ hành chính công,「对丧属反映的殡葬服务问题，应当会同有关部门及时予以处理并将处理结果告知丧属」 - Ghi chú: Xử lý hành vi vi phạm giá thuộc cơ quan quản lý thị trường, bản thân dịch vụ tang chế thuộc cơ quan dân vụ. Tại chỗ hãy đòi một tờ bảng giá, giữ chứng từ thu tiền và tin nhắn, sau đó ra khiếu nại vẫn kịp. + ### 8. Môi giới tang chế, môi giới mua tang phẩm, tổ chức và dẫn dắt tang sự phải đăng ký với cơ quan dân vụ cấp huyện - Chi phí: Không tốn tiền. Hỏi một câu xem đã đăng ký hay chưa. @@ -76,7 +71,6 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý tang chế（国务院令第 824 号, được Hội đồng Nhà nước sửa đổi thông qua tại phiên họp thường vụ lần thứ 72 ngày 14 tháng 11 năm 2025,「自2026年3月30日起施行」）. Điều bốn mươi mốt: 「除殡葬服务机构外，从事殡葬服务代理、用品代购、策划主持、信息咨询等殡葬相关服务活动的组织和个人，应当向县级人民政府民政部门备案。」「对存在以欺骗等手段违规收费、诱导大操大办、强迫接受服务、倒卖逝者信息等损害丧属权益行为的依法处置，并在殡葬服务信息系统中进行标注，向社会公开。」 - Ghi chú: 「倒卖逝者信息」là cái bẫy phổ biến nhất trong nghề này. Vừa làm xong thủ tục lại nhận ngay một loạt cuộc gọi quảng cáo thì phần lớn là thông tin đã bị chuyển nhượng. Nền tảng hành hương trực tuyến cũng chịu quản lý như vậy. Điều bốn mươi lăm của quy chế không cho phép nó xúi giục tiêu và cũng không cho phép xử lý thông tin người đã mất trái phép. - ### 9. Tiền nằm rải rác các nơi thì phải đi lấy từng chỗ: số dư quỹ tích góp nhà ở, quyền lợi bảo hiểm xã hội, quyền lợi tai nạn lao động - Chi phí: Không tốn tiền, nhưng phải đi mấy cửa sổ. Trước khi đi chuẩn bị giấy chứng minh tử vong và giấy chứng minh quan hệ thân nhân. @@ -85,7 +79,6 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2026). Quy chế quản lý quỹ tích góp nhà ở（国务院令第 844 号 sửa đổi, áp dụng từ ngày 20 tháng 9 năm 2026）, khoản cuối Điều hai mươi tư. : 「职工死亡或者被宣告死亡的，职工的继承人、受遗赠人可以提取职工住房公积金账户内的存储余额；无继承人也无受遗赠人的，职工住房公积金账户内的存储余额纳入住房公积金的增值收益。」；ba khoản trợ cấp ở Điều ba mươi chín của《Quy chế bảo hiểm tai nạn lao động》xem phần 19. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018 sửa đổi). Luật bảo hiểm xã hội（Điều mười bốn, mười bảy）: 「个人死亡的，个人账户余额可以继承」「参加基本养老保险的个人，因病或者非因工死亡的，其遗属可以领取丧葬补助金和抚恤金」. (Cơ sở dữ liệu văn bản quốc pháp)；Bộ Nhân sụ và Xã hội, Bộ Tài chính (人力资源社会保障部、财政部) (2021). Biện pháp tạm thời về quyền lợi cho thân nhân trong bảo hiểm hưu trí cơ sở của người lao động doanh nghiệp（人社部发〔2021〕18 号, áp dụng từ ngày 1 tháng 9 năm 2021, Điều bốn, năm）, bản gốc không đăng công khai, văn bản được đối chiếu theo trích dẫn nguyên văn trong quyết định giải quyết khiếu nại hành chính của Chính phủ quận Hoa Đô, thành phố Quảng Châu（花都府行复〔2022〕226 号）: 「丧葬补助金的标准，按照参保人员死亡时本省……上一年度城镇居民月人均可支配收入的2倍计算」「缴费年限不满5年的，发放月数为3个月……缴费年限30年以上的，按照30年计算，发放月数最高为24个月」. ；Sở Nhân sụ và Xã hội tỉnh Phúc Kiến (福建省人力资源和社会保障厅). Câu hỏi thường gặp: 「每领取1年基本养老金减少1个月，发放月数最低为9个月」「累计缴费年限不足5年的，其遗属待遇（含丧葬补助金和抚恤金）标准不得超过其个人缴费之和」. - Ghi chú: Khoản quỹ tích góp nhà ở hay bị bỏ sót nhất, vì nó không nằm trong tài khoản ngân hàng và cũng không tự động vào phần chia tài sản. Số tiền còn lại trong tài khoản bảo hiểm hưu trí cá nhân và quyền lợi cho thân nhân là hai khoản tính riêng và phát riêng, phải ra cửa sổ bảo hiểm xã hội nơi tham gia bảo hiểm làm cùng một lúc. Mỗi nơi đòi giấy tờ khác nhau. Thời gian đóng dưới 5 năm thì tổng quyền lợi cho thân nhân không vượt quá số tiền bản thân người đó đã đóng. Công thức này lấy từ biện pháp về bảo hiểm hưu trí người lao động doanh nghiệp, còn quyền lợi cho thân nhân trong bảo hiểm hưu trí của cơ quan, tổ chức, đơn vị công lập và bảo hiểm hưu trí dân cư thì hỏi tại cửa sổ. Yêu cầu về hình thức của di chúc và thừa kế xem phần 17. - ### 10. Tài khoản và thông tin cá nhân của người đã mất, người thân gần có quyền yêu cầu tra cứu, sao chép, sửa, xoá - Chi phí: Không tốn tiền. Chỉ khi nền tảng không hợp tác mới phải khiếu nại hoặc ra tòa. @@ -93,4 +86,4 @@ Phần tang chế dựa vào bản《Quy chế quản lý tang chế》mới, á - Lợi ích: Luật viết rõ: người tự nhiên mất, thì người thân gần vì lợi ích hợp pháp, chính đáng của mình có thể thực hiện các quyền như tra cứu, sao chép, sửa, xoá đối với thông tin cá nhân liên quan của người đã mất. Người thân gần có thể yêu cầu nền tảng đưa thông tin đó ra cho xem, sao chép một bản, sửa chỗ sai, xoá đi - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). Luật bảo vệ thông tin cá nhân. Trang Quốc hội nhân dân. Điều bốn mươi chín: 「自然人死亡的，其近亲属为了自身的合法、正当利益，可以对死者的相关个人信息行使本章规定的查阅、复制、更正、删除等权利；死者生前另有安排的除外。」；Điều năm mươi「个人信息处理者应当建立便捷的个人行使权利的申请受理和处理机制。拒绝个人行使权利的请求的，应当说明理由」「个人可以依法向人民法院提起诉讼」 -- Ghi chú: Trong điều luật có một câu「死者生前另有安排的除外」, người đã từng đặt người thừa kế tài khoản hoặc để lại ý muốn rõ ràng thì làm theo điều người đó đã để lại. Vậy nên khi người đó còn tỉnh, đáng viết ra tài khoản và mật khẩu nằm ở đâu, xem phần 17. Huỷ tài khoản và xoá thông tin cá nhân là hai việc khác nhau, xem phần 14. \ No newline at end of file +- Ghi chú: Trong điều luật có một câu「死者生前另有安排的除外」, người đã từng đặt người thừa kế tài khoản hoặc để lại ý muốn rõ ràng thì làm theo điều người đó đã để lại. Vậy nên khi người đó còn tỉnh, đáng viết ra tài khoản và mật khẩu nằm ở đâu, xem phần 17. Huỷ tài khoản và xoá thông tin cá nhân là hai việc khác nhau, xem phần 14. diff --git "a/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" "b/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" index b9a13c9d..af24d759 100644 --- "a/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" +++ "b/vi/book/26-\345\201\232\344\270\200\344\270\252\347\275\221\347\253\231\346\210\226\345\271\263\345\217\260.md" @@ -12,7 +12,6 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2023). 非银行支付机构监督管理条例（国务院令第 768 号，第二、六、八、二十七、四十七条）. ；Ngân hàng Nhân dân Trung Quốc (中国人民银行) (2024). 非银行支付机构监督管理条例实施细则（中国人民银行令〔2024〕第 4 号，第六十八条）. ；Quốc hội Nhân dân (全国人大) (1997). 刑法（第二百二十五条第三项）. ；Tòa án nhân dân tối cao (最高人民法院), Viện kiểm sát nhân dân tối cao (最高人民检察院) (2019). 关于办理非法从事资金支付结算业务、非法买卖外汇刑事案件适用法律若干问题的解释（第一、三、四条）. - Ghi chú: Thiết kế dòng tiền để nó không đi vào tài khoản của bạn là cách an toàn duy nhất. Trên thị trường gọi kiểu thu hộ rồi chi hộ như vậy là「二清」, nhưng từ này không có trong bất kỳ văn bản chính thức nào. Quy chế thi hành dùng cách nói「变相从事支付业务」— làm ngành thanh toán theo hình thức ngụy trang, tức là không được phê duyệt mà vẫn thay người dùng chuyển tiền theo mệnh lệnh của họ. Giải thích của Tòa án nhân dân tối cao (最高人民法院) và Viện kiểm sát nhân dân tối cao (最高人民检察院) chỉ liệt kê ba trường hợp: dựng giao dịch giả để rút tiền mặt, dùng tài khoản của đơn vị để rút tiền mặt, dùng séc để rút tiền mặt. Nó không nêu đích danh việc nền tảng thu hộ rồi chi hộ, nên không thể nói rằng nền tảng thu hộ chi hộ「一定构成犯罪」 - ### 2. Trang web thu tiền thì cần giấy phép, bên trung gian kết nối giao dịch là loại giấy phép khác, còn trang web miễn phí chỉ cần đăng ký khai báo - Chi phí: Đăng ký khai báo không tốn tiền. Giấy phép kinh doanh dịch vụ viễn thông giá trị gia tăng (增值电信业务经营许可证) thì có tốn tiền, và nó quyết định「向用户收钱的网站能不能开」— một trang web có thu tiền người dùng có được mở hay không. Muốn có giấy phép đó thì phải có công ty trước. Chỉ hoạt động trong một tỉnh thì vốn đăng ký không thấp hơn 100 vạn tệ. Hoạt động xuyên tỉnh thì không thấp hơn 1000 vạn tệ. Thời gian thẩm định 60 ngày, giấy phép có hiệu lực 5 năm @@ -21,7 +20,6 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2000). 互联网信息服务管理办法（国务院令第 292 号，2024 年 12 月 6 日第二次修订，第三、四、七、十九条）. ；Hội đồng Nhà nước (国务院) (2024). 关于修改和废止部分行政法规的决定（国务院令第 797 号，第十八项）. ；Hội đồng Nhà nước (国务院) (2000). 电信条例（2016 年第二次修订，第七、九、十三、六十九条）. ；Bộ Công nghiệp và Thông tin (工业和信息化部) (2017). 电信业务经营许可管理办法（工信部令第 42 号，第六、九、十一、十四条）. - Ghi chú: Việc bạn làm có thuộc loại EDI hay không thì tự đối chiếu với định nghĩa trong《电信业务分类目录》. Không có câu nào của cơ quan chính thức nói thẳng rằng「带撮合和资金处理的电商平台必须办 EDI 许可」. Hướng dẫn nghiệp vụ của Bộ Công nghiệp và Thông tin (工业和信息化部) chỉ nói「按照业务界定申请相应的电信业务经营许可」— xin giấy phép kinh doanh viễn thông tương ứng với cách định nghĩa ngành nghề. Nên có phải xin hay không thì chỉ tự suy ra từ định nghĩa. Còn một điểm: cá nhân không xin được giấy phép này, người xin phải là công ty được thành lập theo pháp luật - ### 3. Phát trực tiếp thì cần giấy phép kinh doanh văn hóa trên mạng, còn giấy phép chương trình nghe nhìn thì công ty tư nhân gần như không xin được - Chi phí: Giấy phép kinh doanh văn hóa trên mạng (网络文化经营许可证), gọi tắt là giấy phép văn hóa mạng, quyết định「能不能靠直播、游戏解说这类网络表演赚钱」— có được kiếm tiền từ biểu diễn trên mạng như phát trực tiếp hay thuyết minh trò chơi hay không. Cơ quan quản lý văn hóa cấp tỉnh phê duyệt, phải có kết quả trong 20 ngày kể từ khi nhận hồ sơ, giấy phép có hiệu lực 3 năm, phải đi gia hạn trước khi hết hạn 30 ngày. Giấy phép chương trình nghe nhìn (视听节目许可) quyết định「能不能在网上放节目」— có được phát chương trình trên mạng hay không, và nó đòi hỏi người xin là đơn vị sở hữu toàn phần của nhà nước hoặc đơn vị do nhà nước kiểm soát phần lớn @@ -30,7 +28,6 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Bộ Văn hóa (文化部) (2016). 网络表演经营活动管理办法（文市发〔2016〕33 号，第二、四、六、十三、二十条）. ；Bộ Văn hóa (文化部) (2011). 互联网文化管理暂行规定（文化部令第 51 号，2017 年经文化部令第 57 号修订，第七、八、九、十六、二十一、二十八条）. ；Cục Quản lý Phát thanh truyền hình (广电总局), Bộ Công nghiệp và Thông tin (信息产业部) (2007). 互联网视听节目服务管理规定（第 56 号令，2015 年修订，第二、七、八、二十四条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家网信办) và sáu cơ quan khác (2021). 关于加强网络直播规范管理工作的指导意见（国信办发文〔2021〕3 号，第 9 项）. - Ghi chú: Ngành nghề nào hợp với giấy phép nào thì văn bản của bảy cơ quan đó viết rõ nhất. Nền tảng kiếm tiền bằng phát trực tiếp cần giấy phép văn hóa mạng cộng thêm đăng ký ICP, đăng ký ICP là thủ tục đăng ký trang web với Bộ Công nghiệp và Thông tin (工业和信息化部). Phát chương trình trên mạng thì cần giấy phép chương trình nghe nhìn, hoặc hoàn tất đăng ký. Đăng tin tức thì cần giấy phép dịch vụ thông tin tin tức Internet. Ngoài ra,《互联网文化管理暂行规定》mà trang chính phủ Trung Quốc đăng lại năm 2025 là bản gốc năm 2011 chưa sửa đổi, khi trích dẫn phải theo văn bản đang có hiệu lực của Bộ Văn hóa và Du lịch (文化和旅游部) - ### 4. Máy chủ đặt trong nước thì phải đăng ký khai báo, và nhà cung cấp dịch vụ truy cập phải tự có giấy phép viễn thông giá trị gia tăng - Chi phí: Đăng ký khai báo không tốn tiền. Hồ sơ nộp cùng máy chủ tại nhà cung cấp, thường phải chờ mấy ngày đến mấy tuần. Kiểm tra một lần nhà cung cấp đã đủ điều kiện hay chưa thì mười phút là đủ @@ -39,6 +36,7 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Bộ Công nghiệp và Thông tin (信息产业部) (2005). 非经营性互联网信息服务备案管理办法（信产部令第 33 号，2024 年经工信部令第 68 号修订，第五、十三、十八、二十二条）. ；Bộ Công nghiệp và Thông tin (工业和信息化部) (2017). 电信业务经营许可管理办法（工信部令第 42 号，第二十、二十四条）. ；Bộ Công nghiệp và Thông tin (工业和信息化部) (2015). 电信业务分类目录（2015 年版，B11、B14、B21、B25）. - Ghi chú: Trước khi bàn giá, kiểm tra xem nhà cung cấp đã có giấy phép chưa. Tra giấy phép thì dùng Hệ thống thông tin quản lý tổng hợp thị trường kinh doanh viễn thông của Bộ Công nghiệp và Thông tin, tsm.miit.gov.cn, tìm theo tên đầy đủ của công ty hoặc theo số giấy phép. Tra đăng ký khai báo trang web thì dùng beian.miit.gov.cn. Một「机房」không có giấy phép tương ứng thì đến ngày xảy ra sự cố bạn cũng không biết kêu ai. Đặt máy ở nước ngoài thì đúng là tránh được đăng ký khai báo. Nhưng khi người dùng ở trong nước và tiền ở trong nước thì nghĩa vụ nào ghi ở mục 5 đến 10 của phần này cũng không bớt, lại cộng thêm một tầng chi phí chuyển dữ liệu ra nước ngoài + ### 5. Cho người dùng lên bán hàng thì nền tảng phải kiểm tra đăng ký, báo thông tin và lưu giữ ba năm - Chi phí: Không tốn tiền. Phải làm thủ tục kiểm tra tư cách người bán trong hệ thống nền trả, còn phải lập một hồ sơ đăng ký. Mỗi quý báo một lần thông tin liên quan thuế, dữ liệu giao dịch giữ ba năm. Chỗ khó là đây là việc lặp lại hằng năm, không phải làm một lần là xong. @@ -47,7 +45,6 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 电子商务法（第十、二十七、二十八、三十一、八十条）. ；Cục Quản lý Thị trường Quốc gia (市场监管总局) (2021). 网络交易监督管理办法（市场监管总局令第 37 号，第八、二十四、二十五、三十一条）. ；Hội đồng Nhà nước (国务院) (2025). 互联网平台企业涉税信息报送规定（国务院令第 810 号，第二、三、四、十条）. ；Cục Quản lý Thuế Nhà nước (国家税务总局) (2025). 关于互联网平台企业报送涉税信息有关事项的公告（税务总局公告 2025 年第 15 号）. - Ghi chú: Người bán cũng phải tự đi đăng ký chủ thể kinh doanh. Luật liệt kê vài trường hợp không phải đăng ký:「个人销售自产农副产品、家庭手工业产品，个人利用自己的技能从事依法无须取得许可的便民劳务活动和零星小额交易活动」.「零星小额」có một con số cụ thể, Điều 8 của Quy định quản lý và giám sát giao dịch trên mạng ấn định là giá trị giao dịch trong một năm là 10 vạn tệ. Phía nền tảng còn phải làm hai việc: đánh dấu riêng người bán đã đăng ký và chưa đăng ký, và không cho xóa đánh giá của người tiêu dùng - ### 6. Nội dung người dùng đăng lên là bạn phải quản: có cơ chế duyệt, có cửa sổ báo việc, thấy nội dung vi phạm thì dừng truyền ngay và báo cáo - Chi phí: Phải tốn tiền thuê người duyệt nội dung, đây là khoản chi lớn nhất và lâu dài nhất của nền tảng. Cửa sổ báo việc và bản ghi của mỗi lần xử lý đều phải để lại dấu vết. Chỗ khó là ngày nào cũng phải có người trực, ngừng một ngày là có thể xảy ra sự cố. @@ -56,7 +53,6 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2019). 网络信息内容生态治理规定（网信办令第 5 号，第八、九、十、十六、三十七条）. ；Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第四十九、六十九条）. ；Hội đồng Nhà nước (国务院) (2000). 互联网信息服务管理办法（第十五、十六、二十三条）. - Ghi chú: Quy tắc duyệt không cần tự nghĩ, cứ chép theo. Chín loại nội dung không được đăng, Điều 15 của Quy định quản lý dịch vụ thông tin Internet liệt kê rõ mồn một. Người phát trực tiếp còn phải làm thêm ba việc: phải「建立直播内容审核平台」— dựng nền tảng duyệt nội dung phát trực tiếp; phải cắt được một buổi đang phát bất cứ lúc nào, điều luật gọi việc đó là「即时阻断」; và phải lưu nội dung phát trực tiếp cùng nhật ký trong 60 ngày. Tư liệu video của biểu diễn trên mạng cũng không được ít hơn 60 ngày - ### 7. Cung cấp dịch vụ đăng thông tin và nhắn tin tức thì phải yêu cầu người dùng cung cấp thông tin danh tính thật - Chi phí: Phải kết nối một kênh xác minh tên thật, kênh loại này tính phí theo lượt. Sản phẩm nên thiết kế theo nguyên tắc「后台实名、前台自愿」— phía sau lưu tên thật, trên trang chỉ hiện biệt danh. @@ -65,7 +61,6 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2025). 网络安全法（2025 年修正，2026 年 1 月 1 日施行，第二十六、六十四条；2016 年文本为第二十四、六十一条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2016). 互联网直播服务管理规定（第十二、十六条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家网信办) và sáu cơ quan khác (2021). 网络直播营销管理办法（试行）（国信办发文〔2021〕5 号，第八条）. - Ghi chú: Số điều đã đổi chỗ. Luật An ninh mạng sửa đổi năm 2025, có hiệu lực từ ngày 1 tháng 1 năm 2026. Điều về tên thật chuyển từ Điều 24 sang Điều 26, điều về xử phạt chuyển từ Điều 61 sang Điều 64. Văn bản của điều không đổi. Tài liệu bạn đang cầm nếu vẫn dẫn số điều cũ thì phải sửa lại - ### 8. Không mở phát trực tiếp cho người chưa đủ 16 tuổi, tiền thưởng xử lý theo bậc tuổi - Chi phí: Không tốn tiền. Thêm một bước vào mỗi khâu đăng ký và thanh toán: trước hết xác định tuổi, sau đó xác nhận người giám hộ đã đồng ý. @@ -74,6 +69,7 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2020). 未成年人保护法（第七十四、七十六条）. ；Hội đồng Nhà nước (国务院) (2023). 未成年人网络保护条例（国务院令第 766 号，第三十一、四十四条）. ；Văn phòng Tổng thư ký Cục Quản lý Không gian mạng Trung Quốc (中央网信办秘书局) (2026). 关于加强网络直播打赏规范管理的通知（第 7 项）. ；Tòa án nhân dân tối cao (最高人民法院) (2020). 关于依法妥善审理涉新冠肺炎疫情民事案件若干问题的指导意见（二）（法发〔2020〕17 号，第九条）. - Ghi chú: Thực sự có tranh chấp hoàn tiền thì xử theo hướng có lợi nhất cho người chưa thành niên. Cách nói về việc này đã thay đổi. Ý kiến của bốn cơ quan năm 2022 và thông báo của Cục Quản lý Phát thanh truyền hình năm 2020 đều nói người chưa thành niên tuyệt đối không được thưởng tiền. Thông báo mới tháng 4 năm 2026 đổi thành chia bậc theo tám tuổi, mười sáu tuổi, nhưng không ghi rõ hai văn bản trước hết hiệu, cơ quan chính thức cũng không nói hai bên liên hệ với nhau thế nào. Nên câu「未成年人一律禁止打赏」bây giờ không còn chính xác + ### 9. Nhận được thông báo vi phạm thì phải xử lý kịp thời, chuyển bản giải trình mà sau 15 ngày không có động tĩnh thì khôi phục lại - Chi phí: Không tốn tiền. Phải dựng một quy trình cố định: nhận thông báo, chuyển cho người bán, nhận giải trình của người bán, công bố kết quả ra công khai. Mỗi bước đều lưu hồ sơ @@ -82,7 +78,6 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2018). 电子商务法（第四十二、四十三、四十四、四十五、八十四条）. ；Quốc hội Nhân dân (全国人大) (2020). 民法典（第一千一百九十五、第一千一百九十六、第一千一百九十七条）. - Ghi chú: Bạn không phải nền tảng thương mại điện tử mà chỉ là một trang web thường thì áp dụng quy tắc thông báo và gỡ bỏ của Bộ luật Dân sự. Khác biệt nằm ở chỗ sau khi bên kia giải trình thì bạn phải chờ bao lâu: Luật Thương mại điện tử ấn định rõ 15 ngày, còn Bộ luật Dân sự chỉ viết「合理期限」— một thời hạn hợp lý. Ngoài ra, việc công bố ra công khai thông báo đã nhận, giải trình của người bán và kết quả xử lý của bạn là yêu cầu rõ ràng tại Điều 44 của Luật Thương mại điện tử - ### 10. Đừng tùy tiện đặt thông tin người dùng ra nước ngoài, việc chuyển dữ liệu ra nước ngoài có điều kiện luật định và có ngưỡng số người - Chi phí: Không tốn tiền. Cách ít phiền nhất là giữ dữ liệu người dùng trong nước. Đúng là phải chuyển ra nước ngoài thì đi theo hợp đồng mẫu hoặc làm chứng nhận. @@ -91,7 +86,6 @@ Bạn muốn làm một trang web hoặc một ứng dụng. Phần này trả l - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2021). 个人信息保护法（第三十八、三十九、四十条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2024). 促进和规范数据跨境流动规定（网信办令第 16 号，第三、四、五、七、八条）. ；Cục Quản lý Không gian mạng Trung Quốc (国家互联网信息办公室) (2022). 数据出境安全评估办法（网信办令第 11 号，第四条）. - Ghi chú: Máy chủ đặt ở nước ngoài không đồng nghĩa với「数据没出境」. Người dùng đăng ký trong nước, bạn chuyển thông tin của họ sang máy ở nước ngoài, đó là chuyển ra nước ngoài. Có một loại không phải đi ba con đường đánh giá an toàn, chứng nhận và hợp đồng mẫu: mua sắm xuyên biên giới, gửi hàng xuyên biên giới, đặt vé máy bay và khách sạn, vì đó là「为订立、履行个人作为一方当事人的合同」— những chuyển dữ liệu thật sự cần thiết để lập hoặc thực hiện hợp đồng có cá nhân là một bên. Còn phải chú ý cách đếm số người: tính「自当年 1 月 1 日起累计」— từ ngày 1 tháng 1 năm đó trở đi, không phải đếm lùi mười hai tháng từ hôm nay - ### 11. Chọn máy chủ thì xem sập có chịu được không, xem có ai trực vận hành không, cuối cùng mới so giá - Chi phí: Máy chủ ảo của các nhà cung cấp dịch vụ đám mây lớn, mỗi tháng từ vài trăm đến vài nghìn tệ. VPS của nhà cung cấp nhỏ, mỗi tháng bắt đầu từ mấy chục tệ. VPS là một máy chủ ảo nhỏ. Thuê chỗ đặt máy chủ riêng thì mỗi năm bắt đầu từ vài nghìn tệ, và phần cứng phải tự mua. diff --git "a/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" "b/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" index 781604f4..63bd43fe 100644 --- "a/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" +++ "b/vi/book/27-\346\200\200\345\255\225\345\222\214\347\224\237\344\272\247.md" @@ -61,6 +61,7 @@ Con số của Tổ chức Y tế Thế giới: năm 2023, toàn cầu có kho - Mức bằng chứng: A - Nguồn: Centers for Disease Control and Prevention. Hear Her — Urgent Maternal Warning Signs. ；Tổ chức Y tế Thế giới. Maternal mortality fact sheet. - Ghi chú: Cái lợi của bảng này là tách 「nhịn thêm một chút」 khỏi 「phải đi ngay」. Đau đầu kèm nhìn thấy mờ, cộng thêm đau vùng bụng trên, là dấu hiệu hướng tới tiền sản giật (bệnh tăng huyết áp trong thai kỳ). Một bắp chân sưng và đau là dấu hiệu hướng tới huyết khối tĩnh mạch sâu (xem ở phần 13). Băng huyết sau sinh theo cách nói của Tổ chức Y tế Thế giới có thể cướp mạng một người phụ nữ khỏe mạnh trong vài giờ. Chỉ cần một dấu hiệu là đi viện thẳng, đừng ở nhà chờ qua đêm trước. + ### 8. Vỡ ối thì nằm bẹp tại chỗ, kê hông lên, gọi 120, không đi lại và không tắm - Chi phí: Không tốn tiền @@ -125,6 +126,7 @@ Con số của Tổ chức Y tế Thế giới: năm 2023, toàn cầu có kho - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (1958). Quy chế đăng ký hộ khẩu Trung Quốc, 中华人民共和国户口登记条例（第七条）. （Chính phủ tỉnh Quảng Đông, 广东省人民政府, đăng lại） - Ghi chú: Mang đủ những giấy này: Giấy chứng nhận sinh y tế, căn cước và sổ hộ khẩu của cả hai bố mẹ, giấy đăng ký kết hôn. Đăng ký theo bố hay theo mẹ thì tự chọn được, nhưng chính sách hai nơi khác nhau, cả tuyến trường lẫn nơi tham gia bảo hiểm đều bị ảnh hưởng. Suy nghĩ cho xong trước khi đi, sửa lại rất phiền. Trẻ sinh ngoài hôn nhân thì cũng có thể khai báo đăng ký sinh. + ### 16. Đừng bỏ lần tái khám 42 ngày sau khi sinh, lần đó đồng thời là sàng lọc trầm cảm sau sinh - Chi phí: Không tốn tiền, thuộc dịch vụ y tế công cộng cơ bản. Mất nửa ngày. Chỗ khó là vừa hết hậu sản thì lười nhúc nhích. @@ -132,4 +134,4 @@ Con số của Tổ chức Y tế Thế giới: năm 2023, toàn cầu có kho - Lợi ích: Trong phần quản lý sức khoẻ sản phụ của 《国家基本公共卫生服务规范（第三版）》 có, sau khi sinh, 1 lần thăm khám sau sinh và 1 lần khám sức khoẻ 42 ngày sau sinh. Danh mục dịch vụ công khai của địa phương quy định lần này gồm xét nghiệm (công thức máu, soi dịch âm đạo) và 1 lần sàng lọc trầm cảm sau sinh. Trong một năm sau khi sinh đều có thể xảy ra biến chứng nghiêm trọng. Trong bảng 「phải vào viện ngay」 ở mục 7 ở phần này, 「có ý nghĩ làm hại mình hoặc con」「chảy máu âm đạo hoặc dịch tiết bất thường sau khi sinh」「một bên chi thể sưng và đau nặng」 đều thuộc về giai đoạn sau sinh. - Mức bằng chứng: A - Nguồn: Ủy ban Quốc gia Y tế và Kế hoạch gia đình (国家卫生计生委) (2017). Quy chuẩn dịch vụ y tế công cộng cơ bản quốc gia, bản thứ ba. ；Chính quyền thành phố Tĩnh Giang (靖江市人民政府). Dự án dịch vụ y tế công cộng cơ bản quốc gia — quản lý sức khoẻ sản phụ (bảng danh mục dịch vụ công khai của địa phương). -- Ghi chú: Trầm cảm sau sinh không phải là ủng ta, một lời khuyên 「cứ thấy nhẹ tâm thôi」 của người nhà cũng không giải quyết được. Nó có bảng hỏi sàng lọc riêng, có cách điều trị, và có thể hồi phục. Khi có ý nghĩ tự sát thì xử lý ra sao xem ở phần 1. Khi đứa trẻ sinh ra mà thật sự không nuôi nổi thì lối thoát hợp pháp xem ở mục 20 ở phần 9. Lần tái khám này còn xem vết thương hồi phục thế nào, tử cung hồi phục thế nào, chức năng đáy chậu có tốt không. Phương pháp tránh thai cũng được bàn ở lần này: sau khi sinh không có kinh nguyệt không có nghĩa là không thụ thai. \ No newline at end of file +- Ghi chú: Trầm cảm sau sinh không phải là ủng ta, một lời khuyên 「cứ thấy nhẹ tâm thôi」 của người nhà cũng không giải quyết được. Nó có bảng hỏi sàng lọc riêng, có cách điều trị, và có thể hồi phục. Khi có ý nghĩ tự sát thì xử lý ra sao xem ở phần 1. Khi đứa trẻ sinh ra mà thật sự không nuôi nổi thì lối thoát hợp pháp xem ở mục 20 ở phần 9. Lần tái khám này còn xem vết thương hồi phục thế nào, tử cung hồi phục thế nào, chức năng đáy chậu có tốt không. Phương pháp tránh thai cũng được bàn ở lần này: sau khi sinh không có kinh nguyệt không có nghĩa là không thụ thai. diff --git "a/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" "b/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" index 6e667795..79229063 100644 --- "a/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" +++ "b/vi/book/28-\345\210\253\344\270\272\344\272\206\345\244\226\345\275\242\346\212\212\350\272\253\344\275\223\346\220\236\345\235\217.md" @@ -5,14 +5,15 @@ Phần này nói về mấy cách làm thay đổi ngoại hình bằng cách đụng vào cơ thể: nhịn ăn, uống thuốc, tiêm, mổ. Mấy việc này, phần lớn quyết định làm sau khi so với người khác, cái giá lại dồn hết vào bản thân mình, và một phần khá lớn không sửa lại được. Phần này không đánh giá ai muốn mình thành hình dạng nào; đó là việc của từng người. Cùng một mục tiêu, có thể tìm chỗ「có đơn thuốc, có giấy phép, có theo dõi lại」, cũng có thể「mua qua mạng, đến thẩm mỹ viện, tự tăng liều」; phần này chỉ so hai cách đó khác nhau về rủi ro bao nhiêu. Cách tính chủ yếu xem cơ thể thực sự bị gì, như mù mắt, nằm viện, tử vong. Hai mục về thẩm mỹ còn liên quan tới trách nhiệm hình sự của hành nghề y không đúng phép. -### 1. Đừng kiểm soát cân nặng bằng ăn kiêng cực đoan, nhịn ăn hay tự nôn; muốn giảm thì giảm bằng vận động +### 1. Đừng dùng ăn kiêng cực đoan, nhịn ăn hay nôn ra sau khi ăn để kiểm soát cân nặng; muốn giảm thì giảm từ phía vận động -- Chi phí: Không tốn tiền. Ngược lại còn tiết kiệm được tiền mua sản phẩm giảm cân. -- Nói thẳng: Người mắc chán ăn tâm thần có xác suất chết xấp xỉ 5.9 lần người cùng tuổi. Theo dõi 1000 người trong một năm thì chết 5.1 người. Trong số người chết, cứ năm người thì một người chết tự tử. Lối vào của nó là ăn kiêng: theo dõi học sinh 44 trường trung học trong ba năm, nữ sinh ăn kiêng nghiêm trọng có xác suất mắc rối loạn ăn uống gấp 18 lần nữ sinh không kiêng. Muốn kiểm soát cân nặng thì vận động nhiều hơn, đừng ăn ít đến mức đói. -- Lợi ích: Tổng hợp 36 nghiên cứu. Nhóm chán ăn tâm thần được theo dõi tổng cộng 166.642 người-năm (tổng số năm theo dõi của từng người). Tỷ lệ tử vong chuẩn hoá của họ là 5.86, tức xác suất chết xấp xỉ 5.9 lần người cùng tuổi. Cứ 1000 người theo dõi trong một năm thì có 5.1 ca tử vong, tức 1000 người theo dõi một năm chết 5.1 người. Trong số người chết, khoảng 1/5 chết tự tử. Tỷ lệ tử vong chuẩn hoá của chứng ăn vô độ (bulimia) là 1.93, của rối loạn ăn uống khác là 1.92, tức mỗi loại xấp xỉ 1.9 lần người cùng tuổi. Nghiên cứu khác làm ở 44 trường trung học tại Victoria, Úc, theo dõi học sinh 14–15 tuổi trong 3 năm, cùng một nhóm học sinh được khảo sát 6 đợt trong ba năm đó. Nữ sinh mắc rối loạn ăn uống mới là 21.8 ca trên 1000 người theo dõi một năm, tức 1000 nữ sinh theo dõi một năm xuất hiện 21.8 ca mới. Nữ sinh ăn kiêng nghiêm trọng có nguy cơ gấp 18 lần người không kiêng, người ăn kiêng mức vừa là 5 lần. Phân tích sau đó hiệu chỉnh cho việc đã từng ăn kiêng và triệu chứng tâm thần, loại bỏ ảnh hưởng của「trước đây đã từng ăn kiêng」và「vốn đã có triệu chứng tâm thần」. Sau khi loại bỏ, BMI (chỉ số béo gầy tính bằng trọng lượng chia chiều cao bình phương), lượng vận động và giới tính không còn dự báo rối loạn ăn uống mới. Tác giả kết luận「ở thanh thiếu niên, dùng vận động thay vì hạn chế ăn để kiểm soát cân nặng có vẻ làm giảm nguy cơ rối loạn ăn uống」 +- Chi phí: Không tốn tiền. Ngược lại còn tiết kiệm tiền mua các sản phẩm giảm cân. +- Nói thẳng: Người bị chứng ăn uống từ chối tâm thần có xác suất tử vong cao gấp khoảng 5.9 lần người cùng tuổi. Cứ 1000 người theo dõi một năm thì có 5.1 người chết. Trong số người chết, cứ năm người thì một người tự tử. Cửa vào của nó chính là ăn kiêng: theo dõi học sinh ở 44 trường trung học trong ba năm, nữ sinh ăn kiêng nặng có xác suất mắc rối loạn ăn uống cao gấp 18 lần nữ sinh không ăn kiêng. Muốn giữ cân thì tập thể dục nhiều hơn, đừng ăn thiếu đến mức đói. +- Lợi ích: Gộp 36 nghiên cứu lại tính chung. Nhóm người bị chứng ăn uống từ chối tâm thần cộng lại được theo dõi 166.642 người năm (số năm theo dõi của từng người cộng dồn). Tỷ số tử vong chuẩn hoá của họ là 5.86, tức xác suất tử vong cao gấp khoảng 5.9 lần người cùng tuổi. Mỗi 1000 người năm có 5.1 ca tử vong, tức 1000 người theo dõi một năm thì chết 5.1 người. Trong số người chết, khoảng 1/5 chết vì tự tử. Chứng ăn uống thừa có tỷ số tử vong chuẩn hoá 1.93, các rối loạn ăn uống khác là 1.92, tức mỗi loại cao gấp khoảng 1.9 lần người cùng tuổi. Một nghiên cứu khác làm ở 44 trường trung học tại bang Victoria, Úc, theo dõi học sinh 14–15 tuổi trong 3 năm, khảo sát cùng một nhóm học sinh 6 đợt trong ba năm đó. Nữ sinh mới mắc rối loạn ăn uống 21.8/1000 người năm, tức 1000 nữ sinh theo dõi một năm thì có 21.8 ca mới. Nữ sinh ăn kiêng nặng có nguy cơ mắc bệnh cao gấp 18 lần người không ăn kiêng, người ăn kiêng mức vừa là gấp 5 lần. Nghiên cứu còn hiệu chỉnh lại tiền sử ăn kiêng và triệu chứng tâm thần, trừ đi ảnh hưởng của「đã từng ăn kiêng」và「vốn đã có triệu chứng tâm thần」. Sau khi trừ đi, chỉ số BMI (chỉ báo béo gầy tính bằng cân nặng chia cho bình phương chiều cao), mức vận động và giới tính không còn dự đoán được ca rối loạn ăn uống mới. Kết luận của tác giả là「ở thanh thiếu niên, dùng vận động thay cho kiêng ăn để kiểm soát cân nặng thì nguy cơ rối loạn ăn uống có vẻ thấp hơn」 - Mức bằng chứng: A -- Nguồn: Arcelus J và cộng sự (2011). Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry. ；Patton GC và cộng sự (1999). Onset of adolescent eating disorders: population based cohort study over 3 years. BMJ. -- Ghi chú: Ở đây「ăn kiêng nghiêm trọng」nghĩa là duy trì lâu dài mức ăn thấp hơn nhiều so với nhu cầu cơ thể, không phải bỏ một bữa hay bỏ ăn khuya. Nếu đã tự nôn, ăn xong lại tự trách, lên cân rồi không dừng được thì đi khoa tâm thần hoặc khoa tâm lý lâm sàng của bệnh viện đa khoa. Đây là bệnh có cách điều trị đúng quy trình, không phải vấn đề ý chí. Quan hệ giữa BMI và tỷ lệ tử vong xem mục 33 ở phần 2; mục đó nói về khoảng cân nặng dài hạn, không nói về biện pháp giảm cân. +- Nguồn: Arcelus J et al. (2011). Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry. ；Patton GC et al. (1999). Onset of adolescent eating disorders: population based cohort study over 3 years. BMJ. +- Ghi chú: Ở đây「ăn kiêng nặng」là giảm ăn trong thời gian dài xuống thấp hơn nhiều so với nhu cầu của cơ thể, không phải là bỏ một bữa hay không ăn vặt khuya. Đã nôn ra sau khi ăn, ăn no rồi tự trách, cân đi cân lại không dừng thì đi khoa Tâm thần hoặc khoa Tâm lý lâm sàng ở bệnh viện đa khoa. Đây là bệnh có phương pháp điều trị theo quy định, không phải chuyện ý chí. Quan hệ giữa BMI và tỷ lệ tử vong xem mục 32 ở phần 2, mục đó nói về khoảng cân nặng dài hạn, không nói về cách giảm cân. + ### 2. Trước khi tiêm, cấy chỉ hay phẫu thuật, kiểm tra hai thứ: giấy phép cơ sở có ghi「y học thẩm mỹ」không, và người thao tác có phải bác sĩ chủ trị không - Chi phí: Không tốn tiền. Kiểm tra hai loại giấy tờ này chỉ mất vài phút. Cơ sở phải treo Giấy phép hành nghề cơ sở y tế ở đại sảnh. Văn bằng của bác sĩ tra được trên hệ thống tra cứu bác sĩ hành nghề của ngành y tế. diff --git "a/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" "b/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" index 7706ad08..82da0f77 100644 --- "a/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" +++ "b/vi/book/29-\351\201\255\351\201\207\351\207\215\345\244\247\346\211\223\345\207\273\344\271\213\345\220\216.md" @@ -14,7 +14,6 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: A - Nguồn: Mostofsky E, et al. (2012). Risk of acute myocardial infarction after the death of a significant person in one's life: the Determinants of Myocardial Infarction Onset Study. Circulation. ；Carey IM, et al. (2014). Increased risk of acute cardiovascular events after partner bereavement: a matched cohort study. JAMA Internal Medicine. - Ghi chú: Đưa hộp thuốc cho người khác giữ, đừng để mình tự nhớ. Mấy ngày đó đừng đi đường dài. Đừng dựa vào rượu để chịu đựng, xem phần 2. Nhồi máu cơ tim và đột quỵ nhận biết thế nào, gọi xe thế nào, xem phần 13. Cả hai đều là nghiên cứu nhìn lại dữ liệu nhóm người, không tách được trong đó có bao nhiêu người vốn đang sắp bệnh. Nhưng cả hai cùng chỉ một hướng và cùng một khoảng thời gian. Người không thể đưa hộp thuốc cho ai và chắc mấy ngày đó ở một mình trong nhà thì xem mục 6 của phần này - ### 2. Trong tuần bạn được báo mắc bệnh nặng, đừng đi lấy kết quả một mình, quyết định gì ngoài điều trị thì đẩy lùi - Chi phí: Không tốn tiền. Tìm một người đi cùng bạn lấy kết quả. Đẩy mấy quyết định lớn ngoài điều trị lùi vài ngày @@ -23,7 +22,6 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: A - Nguồn: Fang F, et al. (2012). Suicide and cardiovascular death after a cancer diagnosis. New England Journal of Medicine. - Ghi chú: Đây là số liệu của bệnh nhân ung thư. Đổi sang các bệnh nan y khác thì không có nguồn gốc tương đương. Người gặp chuyện thực ra rất ít: tỷ lệ tự tử trong tuần đầu là 2.50/1000 người-năm, tức 1.000 người theo dõi một năm thì có 2.5 ca. Viết ra không phải để dọa, mà để nói rõ mấy ngày đầu mới biết tin phải có người ở bên cạnh. Quản lý lâu dài sau khi chẩn đoán, bồi hoàn và tái khám xem phần 16 và phần 24. Không tìm được người đi cùng thì xem mục 6 của phần này (thay «người đó» bằng hàng xóm và danh sách của cộng đồng) - ### 3. Sau khi thất nghiệp, ổn định giờ sinh hoạt, bảo hiểm y tế và nhịp đi tìm việc trước, đừng ở nhà suốt ngày - Chi phí: Không tốn tiền. Đi đăng ký thất nghiệp, nhận tiền bảo hiểm thất nghiệp trước, rồi xếp việc tìm việc vào một lịch cố định. Khó ở chỗ không ai đuổi mà vẫn phải dậy đúng giờ @@ -32,7 +30,6 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: A - Nguồn: Roelfs DJ, et al. (2011). Losing life and livelihood: a systematic review and meta-analysis of unemployment and all-cause mortality. Social Science & Medicine. - Ghi chú: Đây là nghiên cứu nhìn lại dữ liệu nhóm người, không tách được chiều nhân quả. Người vốn sức khoẻ kém cũng dễ thất nghiệp hơn. Nhưng câu «sau khi trừ phần ảnh hưởng của thói quen sinh hoạt thì hiệu ứng thấp hơn 24%» cho một hướng để bắt tay. Thất nghiệp được nhận khoản gì, ở đâu nhận, bảo hiểm xã hội gia hạn thế nào thì xem phần 7. Bị sa thải thì được bồi thường bao nhiêu, đừng ký tự nguyện nghỉ việc thì xem phần 19 - ### 4. Người thân chết vì tự tử, tai nạn hoặc bị giết thì đừng trông chờ tự gượng, hãy chủ động đi tìm giúp đỡ chuyên môn - Chi phí: Gọi 12356 thì không tốn tiền. Muốn khám bác sĩ thì đăng ký khám tâm lý ngoại trú, giá như khám ngoại trú thường. Khó ở chỗ phải tự mở lời nhờ người @@ -41,7 +38,6 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: A - Nguồn: Erlangsen A, et al. (2017). Association Between Spousal Suicide and Mental, Physical, and Social Health Outcomes: A Longitudinal and Nationwide Register-Based Study. JAMA Psychiatry. ；Pitman A, et al. (2014). Effects of suicide bereavement on mental health and suicide risk. The Lancet Psychiatry. ；Kristensen P và cộng sự (2012). Bereavement and mental health after sudden and violent losses: a review. Psychiatry. （bản tổng quan trong phần Ghi chú） - Ghi chú: Cái chết đột ngột, bạo lực là tai nạn, bị giết, thảm hoạ. Sau loại chết này, tỷ lệ rối loạn sau chấn thương tâm lý, trầm cảm và nỗi buồn bị kẹt đều cao hơn sau khi chết bệnh. Hồi phục cũng chậm hơn. Khi bản thân có ý nghĩ tự tử thì xem mục 25 ở phần 1 - ### 5. Trong nửa năm đầu sau khi mất bạn đời, giao việc canh bữa cơm, việc ngủ, thuốc bệnh mạn tính và lịch tái khám cho một người cố định - Chi phí: Không tốn tiền. Tìm một người cụ thể, con cái, anh chị em hay bạn đều được, bảo người đó định kỳ đến nhà hoặc gọi điện. Khó ở chỗ phải mở lời giao việc này @@ -51,14 +47,14 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Nguồn: Moon JR, et al. (2011). Widowhood and mortality: a meta-analysis. PLoS ONE. ；Shor E, et al. (2012). Widowhood and mortality: a meta-analysis and meta-regression. Demography. - Ghi chú: Đây là nghiên cứu nhìn lại dữ liệu nhóm người, không gỡ được yếu tố khác lẫn vào. Vợ chồng vốn đã giống nhau về lối sống và bệnh tật. Nhưng quy luật «nửa năm đầu cao hơn nhiều so với sau đó» giống nhau trong cả hai phân tích tổng hợp độc lập, đáng coi là một giai đoạn cần có người trông. Nam có nguy cơ cao hơn và ít kêu cầu giúp đỡ hơn. Nhà có người lớn tuổi vừa mất bạn đời thì theo phần 17, làm cho rõ tài khoản, giám hộ và di chúc. Không tìm được người như vậy (sống một mình, không còn con, con cái cũng không còn) thì giao việc này cho cộng đồng và điện thoại, xem mục 6 của phần này -### 6. Không có thân nhân cũng không có bạn, hãy thay «người trông hộ bạn» bằng ba thứ: hàng xóm có vào được nhà bạn, danh sách thăm hỏi của cộng đồng, và người liên hệ khẩn cấp trong điện thoại +### 6. Không có người thân cũng không có bạn bè thì thay「người canh giữ bạn」bằng ba thứ: hàng xóm có vào được nhà bạn, danh sách thăm viếng của cộng đồng, và người liên hệ khẩn cấp trong điện thoại -- Chi phí: Không tốn tiền. Khó ở chỗ phải tự mở lời, nói với hàng xóm và ủy ban nhân dân một câu «tôi sống một mình, trong nhà vừa xảy ra chuyện» -- Nói thẳng: Khi không có thân nhân cũng không có bạn thì cần trông hai việc: có người phát hiện bạn khi xảy ra chuyện, và ăn uống dùng thuốc đúng giờ hằng ngày. Ăn uống dùng thuốc thì tự lo, viết thành danh sách, đặt báo thức, đừng trông vào trí nhớ. Có người phát hiện thì dựa vào ba thứ: để lại cho hàng xóm một chìa khoá dự phòng, hỏi ủy ban nhân dân xem có thể vào danh sách thăm hỏi chăm sóc không, và đặt sẵn người liên hệ khẩn cấp trong điện thoại, đừng tắt máy -- Lợi ích: 44.573 bệnh nhân ngoại trú, tất cả có nguy cơ hoặc tiền sử huyết khối mạch máu do xơ vữa (bệnh mạch máu có mảng bám trong lòng mạch, dễ bị cục máu đông nghẽn). Họ được bác sĩ nhiều nước đăng ký vào chương trình quốc tế REACH để theo dõi, nhập mẫu tháng 12 năm 2003 đến tháng 12 năm 2004, theo dõi 4 năm. Trong đó 8.594 người (19%) sống một mình. Người sống một mình, trong 4 năm, tử vong do mọi nguyên nhân là 14.1%, người ở cùng nhà là 11.1%. Tử vong tim mạch 8.6% so với 6.8%. Hai chỉ số log-rank P<0.01, khác biệt khó có thể là trùng hợp. Ở các nhóm tuổi khác nhau thì mức chênh lệch cũng khác nhau (P tương tác=0.03). Từ 45 đến 65 tuổi, tử vong người sống một mình so với người ở cùng nhà là 7.7% so với 5.7%. Sau khi cân bằng các yếu tố khác, HR 1.24 (95% CI 1.01–1.51), tức cao hơn khoảng hai phần tư. Từ 66 đến 80 tuổi là 13.2% so với 12.3%, HR 1.12 (1.01–1.26), cao hơn khoảng một phần mười. Trên 80 tuổi là 24.6% so với 28.4%, HR 0.92 (0.79–1.06), không thấy tăng. Ý kiến năm 2022 của Bộ Dân vụ cùng mười bộ khác quy định phải triển khai dịch vụ thăm hỏi chăm sóc. Đối tượng là người cao niên sống một mình, không có con, con ở lại quê, mất khả năng tự chăm sóc, khuyết tật nặng, gia đình có con đông theo kế hoạch sinh và các trường hợp đặc biệt. Hình thức dịch vụ là «thăm hỏi tại nhà theo định kỳ, điện thoại, hình ảnh, giám sát từ xa». Chính quyền xã, phường điều tra, ủy ban nhân dân xóm, khu phố hỗ trợ, nắm thông tin cơ bản và «ý muốn nhận dịch vụ thăm hỏi chăm sóc», rồi đưa vào quản lý lưới ở cấp cơ sở, tức chia theo khu vực và giao người phụ trách. Đến cuối năm 2025, «bảo đảm tỷ lệ thăm hỏi hằng tháng đạt 100% với người cao niên khó khăn đặc biệt», «với người cao niên khó khăn đặc biệt mất khả năng tự chăm sóc thì mỗi tháng thăm hỏi chăm sóc không ít hơn một lần». Văn bản còn khuyến nghị trang bị trong gia đình hệ thống gọi cầu thông minh, đồng hồ nước thông minh, sản phẩm theo dõi sức khoẻ, thiết bị giám sát người cao niên, «một khi phát hiện bất thường thì cảnh báo kịp thời và đồng thời gửi thông báo cho người liên hệ khẩn cấp». Nhân viên phụ vụ phát hiện vấn đề khẩn cấp thì «phải giúp gọi số cấp cứu ngay lập tức» +- Chi phí: Không tốn tiền. Chỗ khó là phải chủ động mở lời, nói với hàng xóm và ủy ban nhân dân phường một câu「tôi sống một mình, trong nhà vừa xảy ra chuyện」. +- Nói thẳng: Khi không có người thân cũng không có bạn bè, có hai việc phải lo: có người phát hiện ra bạn khi xảy ra chuyện, và ăn uống uống thuốc đúng giờ hằng ngày. Việc ăn uống uống thuốc tự lo, viết thành danh sách, đặt báo thức, đừng dựa vào trí nhớ. Việc có người phát hiện dựa vào ba thứ: để lại một chìa khoá dự phòng cho hàng xóm, tới ủy ban nhân dân phường hỏi có vào được danh sách thăm viếng chăm sóc không, và đặt sẵn người liên hệ khẩn cấp trong điện thoại, đừng để máy tắt nguồn. +- Lợi ích: 44.573 bệnh nhân ngoại trú, tất cả đều có nguy cơ hoặc tiền sử huyết khối do xơ vữa động mạch (loại bệnh mà trong mạch máu có mảng bám và dễ bị cục máu nghẽn). Họ được bác sĩ nhiều nước đăng ký vào chương trình quốc tế REACH để theo dõi, lấy nhóm từ tháng 12 năm 2003 đến tháng 12 năm 2004, theo dõi 4 năm. Trong đó 8.594 người (19%) sống một mình. Người sống một mình trong 4 năm có tỷ lệ tử vong do mọi nguyên nhân cộng lại là 14.1%, người ở cùng nhà là 11.1%. Tử vong tim mạch là 8.6% so với 6.8%. Hai chỉ số log-rank P<0.01, khác biệt khó có thể là ngẫu nhiên. Ở các nhóm tuổi khác nhau, mức chênh lệch không giống nhau (tương tác P=0.03). 45 đến 65 tuổi, tỷ lệ tử vong của người sống một mình so với người ở cùng nhà là 7.7% so với 5.7%. Sau khi đã cân bằng các yếu tố khác thì HR 1.24 (95% CI 1.01–1.51), tức cao khoảng hai mươi bốn phần trăm. 66 đến 80 tuổi là 13.2% so với 12.3%, HR 1.12 (1.01–1.26), cao khoảng mười phần trăm. Trên 80 tuổi là 24.6% so với 28.4%, HR 0.92 (0.79–1.06), ngược lại không thấy tăng. Ý kiến năm 2022 của Bộ Dân vụ cùng chín bộ khác quy định phải triển khai dịch vụ thăm viếng chăm sóc. Đối tượng là người cao tuổi sống một mình, vợ con đều đi xa, ở lại quê, mất khả năng tự phục vụ, tàn tật nặng, thuộc gia đình có kế hoạch sinh con đặc biệt. Cách phục vụ là「định kỳ tới tận nhà, gọi điện và gọi video, theo dõi từ xa」. Chính quyền xã phường rà soát, ủy ban nhân dân xã hỗ trợ, nắm thông tin cơ bản và「ý chị muốn được thăm viếng chăm sóc」rồi đưa vào quản lý lưới ở tuyến cơ sở, tức chia theo khu vực để giao người phụ trách theo dõi. Đến cuối năm 2025「bảo đảm tỷ lệ thăm viếng hằng tháng đối với người cao tuổi khó khăn đặc biệt đạt 100%」,「đối với người cao tuổi khó khăn đặc biệt mất khả năng tự phục vụ thì mỗi tháng thăm viếng chăm sóc không ít hơn một lần」. Văn bản còn khuyến nghị các hộ gia đình lắp hệ thống gọi cứu thông minh, đồng hồ nước thông minh, sản phẩm theo dõi sức khoẻ, thiết bị giám sát người cao tuổi,「khi phát hiện chỉ số bất thường thì cảnh báo kịp thời và đồng thời gửi thông báo cho người liên hệ khẩn cấp」. Nhân viên phục vụ phát hiện vấn đề khẩn cấp thì「phải giúp gọi đường dây cầu cứu khẩn ngay lập tức」 - Mức bằng chứng: B -- Nguồn: Udell JA, et al. (2012). Living alone and cardiovascular risk in outpatients at risk of or with atherothrombosis. Archives of Internal Medicine. ；Bộ Dân vụ và mười bộ khác (2022). Ý kiến hướng dẫn về việc triển khai dịch vụ thăm hỏi chăm sóc người cao niên khó khăn đặc biệt（民发〔2022〕73 号）. -- Ghi chú: Cụ thể cách dùng ba thứ đó. Một, đưa chìa khoá dự phòng hoặc mật khẩu tạm của ổ khóa cho một người hàng xóm hoặc bảo vệ, nói với người đó một câu «mấy ngày này tôi ở một mình». Không cần thân tình, chỉ cần có người biết bạn ở trong nhà. Hai, đến ủy ban nhân dân cộng đồng nói mình sống một mình, hỏi xem có thể vào danh sách thăm hỏi chăm sóc không. Trong số người cao niên thì người sống một mình, không có con, mất con là nhóm được chính thức xác định làm đối tượng dịch vụ, tiện thể hỏi xem địa phương có lắp miễn phí thiết bị gọi cầu thông minh, đồng hồ nước thông minh loại đó không. Ba, đặt sẵn người liên hệ khẩn cấp và thẻ cấp cứu y tế trong điện thoại, mấy ngày đó đừng để im hay tắt máy. Cho B vì hai lý do. Một là liên hệ giữa sống một mình và tỷ lệ tử vong đến từ nghiên cứu nhìn lại dữ liệu nhóm người, không gỡ được cách giải thích ngược rằng người vốn sức khoẻ kém, điều kiện kém thì dễ sống một mình hơn, chính bài gốc cũng viết kết quả này còn chờ xác nhận. Hai là văn bản về dịch vụ thăm hỏi chăm sóc chỉ phủ người cao niên. Người chưa đầy 60 tuổi không có chế độ riêng tương ứng. Có thể trông chờ là một câu trong văn bản đó, nó yêu cầu cán bộ lưới và nhân viên xã hội «kịp thời phát hiện rủi ro khủng hoảng tâm lý như biến cố gia đình, thất nghiệp, mất đi cơ hội học» (xem mục 11 của phần này, muốn có người để nói chuyện thì gọi 12356). «Có người ở trong nhà» đáng giá bao nhiêu thì xem mục 1 ở phần 13 (ép ngực). Ngừng tim xảy ra ngoài bệnh viện thì 79.2% xảy ra ở nhà. Có người bên cạnh làm ép ngực thì tỷ lệ xuất viện còn sống là 16.1%, không có người thì chỉ 3.9%. Trong nhà không có người nghĩa là rơi vào nhóm thấp. Thêm hai việc nữa làm tiện tay. Bản thân việc cưới tang cũng phải chạy nhiều cửa sổ ở nhà tang lễ, trạm y tế cộng đồng, công an, quỹ tích góp nhà ở (xem phần 25). Xếp việc đó thành lịch cố định mỗi ngày ra ngoài một lần thì các việc cần đi trong mấy ngày đó đã có sẵn. Đường dây nóng gọi nhiều lần được, không phải chỉ gọi một lần (xem mục 11 của phần này, muốn có người để nói chuyện thì gọi 12356). Bản thân việc «sống một mình» cũng là một yếu tố nguy cơ dài hạn. Tỷ số tử vong của người sống một mình là 1.32, cao hơn khoảng ba phần mười, xem mục 10 ở phần 22 (coi việc gặp gỡ định kỳ với người khác là một khoản chi cho sức khoẻ). Nhưng ở đây chỉ nói mấy ngày trước mắt, sổ sức khoẻ dài hạn thì tính riêng +- Nguồn: Udell JA, et al. (2012). Living alone and cardiovascular risk in outpatients at risk of or with atherothrombosis. Archives of Internal Medicine. ；Bộ Dân vụ cùng chín bộ khác (民政部等十部门) (2022). Ý kiến hướng dẫn về việc triển khai dịch vụ thăm viếng chăm sóc người cao tuổi khó khăn đặc biệt, 关于开展特殊困难老年人探访关爱服务的指导意见（民发〔2022〕73 号）. +- Ghi chú: Cụ thể ba thứ đó làm thế nào. Một, đưa chìa khoá dự phòng hoặc mật khẩu tạm của ổ khoá cho một hàng xóm hoặc bên quản lý nhà, và nói với họ một câu「mấy ngày này tôi ở nhà một mình」. Không cần quan hệ thân, chỉ cần có người biết là bạn đang ở trong nhà. Hai, tới ủy ban nhân dân cộng đồng nói rõ mình sống một mình, hỏi xem có vào được danh sách thăm viếng chăm sóc không. Trong người cao tuổi, người sống một mình, vợ con đều đi xa, mất người thân là nhóm mà chính quyền nêu rõ là đối tượng được phục vụ; tiện thể hỏi xem địa phương có lắp miễn phí thiết bị gọi cứu thông minh, đồng hồ nước thông minh các loại đó không. Ba, đặt sẵn người liên hệ khẩn cấp và thẻ cấp cứu y tế trong điện thoại, mấy ngày này đừng để chế độ im lặng, đừng tắt nguồn. Có hai lý do ghi mức B. Một là mối liên hệ giữa sống một mình và tỷ lệ tử vong đến từ nghiên cứu hồi cứu dữ liệu dân số, không bỏ được lớp giải thích ngược「người vốn đã yếu, hoàn cảnh kém thì dễ sống một mình hơn」, chính bài gốc cũng viết kết quả này còn cần xác nhận. Hai là văn bản về dịch vụ thăm viếng chăm sóc chỉ phủ cho người cao tuổi. Người chưa đủ 60 tuổi không có chế độ riêng nào ứng với. Cái có thể trông chờ là một điều trong văn bản, đòi lưới viên và nhân viên xã hội「kịp thời phát hiện các rủi ro khủng hoảng tâm lý như biến cố gia đình, thất nghiệp, mất đi cơ hội học」(xem mục 11 của phần này, muốn có người để nói chuyện thì gọi 12356). Vì sao「trong nhà có người」lại đáng giá, xem mục 1 ở phần 13 (ép ngực). Trong các ca ngừng tim xảy ra ngoài bệnh viện, 79.2% xảy ra ở nhà. Khi bên cạnh có người ép ngực, tỷ lệ sống sót xuất viện là 16.1%, không có ai thì chỉ 3.9%. Không có ai ở nhà tức là rơi vào nhóm thấp. Còn hai việc tiện tay nữa. Bản thân việc cầm tang vốn phải chạy nhiều cửa như nhà tang lễ, cơ quan y tế cộng đồng, công an, quỹ tích góp nhà ở (xem phần 25). Xếp nó thành lịch cố định mỗi ngày ra ngoài một lần thì chuyến đi trong mấy ngày đó có sẵn. Đường dây nóng gọi nhiều lần được, không phải chỉ gọi một lần (xem mục 11 của phần này, muốn có người để nói chuyện thì gọi 12356). Bản thân việc「sống một mình」vẫn là một yếu tố rủi ro dài hạn. Tỷ số tử vong của người sống một mình là 1.32, cao hơn khoảng ba mươi phần trăm, xem mục 9 ở phần 22 (coi việc gặp gỡ người quen đúng kỳ là một khoản chi cho sức khoẻ). Nhưng ở đây chỉ tính mấy ngày trước mắt, còn khoản dài hạn thì tính riêng. ### 7. Sau khi cha mẹ qua đời, nói thật với con, để con dự lễ tang, đừng đưa con đi nơi khác để «đổi không khí» @@ -68,7 +64,6 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: B - Nguồn: Li J, et al. (2014). Mortality after parental death in childhood: a nationwide cohort study from three Nordic countries. PLoS Medicine. - Ghi chú: Khi nói với con đừng bịa «đi công tác xa», việc giám hộ và nuôi dưỡng phải viết ra giấy. Cho B không phải vì con số yếu. Đây là nghiên cứu toàn bộ dân số của ba nước, bản thân số liệu đủ A. Cho B là vì từ con số này đến cách làm cụ thể «nói với con thế nào, có nên cho con dự lễ tang không» không có bằng chứng thử nghiệm trực tiếp. Tác giả cũng viết rõ là thiếu dữ liệu về quan hệ cha mẹ con và lối sống của trẻ sau khi mất người thân. Mức tăng vừa có phần do thể chất sẵn có, vừa có phần do ảnh hưởng dài hạn của việc mất người thân. Chỉ định giám hộ và sắp xếp tài sản xem phần 17. Cấp dưỡng trẻ và chuyện đi học xem phần 18 - ### 8. Nỗi buồn kéo dài quá nửa năm vẫn đứng yên, ngày nào cũng không nổi, thì đi đăng ký khoa tâm thần hoặc tâm lý lâm sàng - Chi phí: Phí đăng ký khám cộng phí điều trị. Loại điều trị này làm theo liệu trình, trong thử nghiệm thì trung bình 19 tuần làm 16 buổi. Khó ở chỗ phải tự bước ra đăng ký số khám đó @@ -77,7 +72,6 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: A - Nguồn: Lundorff M, et al. (2017). Prevalence of prolonged grief disorder in adult bereavement: A systematic review and meta-analysis. Journal of Affective Disorders. ；Prigerson HG, et al. (2009). Prolonged grief disorder: Psychometric validation of criteria proposed for DSM-V and ICD-11. PLoS Medicine. ；Shear MK, et al. (2005). Treatment of complicated grief: a randomized controlled trial. JAMA. - Ghi chú: Đừng tự đếm số tháng. Cách đánh giá đúng là «đã ảnh hưởng tới ăn, ngủ, đi làm, chăm con», khớp thì đi đăng ký khám. Khoa gọi là tâm thần, tâm lý hoặc tâm lý lâm sàng, bệnh viện chuyên khoa tâm thần công lập đều có. Chẩn đoán này trong tài liệu khác gọi là đau buồn phức tạp, cũng gọi là rối loạn đau buồn kéo dài. Các tiêu chuẩn khác nhau đòi kéo dài bao lâu cũng khác nhau, có bản viết 6 tháng, có bản viết 12 tháng. Con số 9.8% đến từ 14 nghiên cứu làm khác nhau khá xa, tác giả cũng nhắc nó có thể không chuyển được sang nhóm người khác - ### 9. Đừng vội trả tiền làm tư vấn đau buồn, xem trước nỗi buồn của mình có thật sự bị kẹt không (các triệu chứng ở mục 8) - Chi phí: Không tốn tiền, ngược lại còn tiết kiệm được một khoản phí tư vấn @@ -95,7 +89,6 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: A - Nguồn: Sbarra DA, et al. (2011). Divorce and Death: A Meta-Analysis and Research Agenda for Clinical, Social, and Health Psychology. Perspectives on Psychological Science. - Ghi chú: Bài gốc chỉ nói «tăng rõ rệt» và so sánh giữa các nhóm, không đưa ra hệ số có thể trích dẫn, nên ở đây không viết con số cụ thể. Cỡ lợi ích đặt là vừa, theo cách «biết hướng, không có mức giảm nào để áp». Tài sản, nợ và tiền cưới lúc ly hôn xem phần 8 và phần 10. Cách sắp xếp nuôi con xem phần 18 - ### 11. Muốn có người để nói chuyện thì gọi 12356, người chưa thành niên và thanh niên gọi 12355, muốn khám bác sĩ thì đăng ký khám tâm lý ngoại trú - Chi phí: Không tốn tiền. Cả hai đường dây nóng đều miễn phí. Khám tâm lý ngoại trú thì đăng ký như khám ngoại trú thường @@ -104,7 +97,6 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: B - Nguồn: Ủy ban Quốc gia Y tế và Sức khoẻ cùng 25 bộ (2026). Thông báo về ban hành phương án hoàn thiện hệ thống dịch vụ tâm lý xã hội và cơ chế can thiệp khủng hoảng（国卫医政发〔2026〕8 号）. - Ghi chú: Cho B vì phòng tư vấn ở xóm và cộng đồng, phòng khám tâm lý cấp huyện đều là mục tiêu đến năm 2030 mới phải xây, không phải hiện giờ nơi nào cũng có. Còn 12356 và 12355 thì gọi được ngay bây giờ. Thời điểm mở 12356 và mỗi ngày nghe trong bao lâu thì xem mục 25 ở phần 1. Tư vấn tâm lý và trị liệu tâm lý không phải một. Chẩn đoán và kê đơn thuốc phải do bác sĩ tâm thần thực hiện, điều trị phải làm ở cơ sở y tế có giấy phép. Loại «tư vấn chữa lành» và «tư vấn tâm linh» tính tiền theo giờ bán ngoài không nằm trong hệ thống này - ### 12. Trong ba tháng đầu sau biến cố, mọi quyết định lớn không thể đảo ngược đều đẩy lùi - Chi phí: Không tốn tiền, chỉ là đẩy quyết định lùi, có thể vì thế bỏ lỡ một số cơ hội. Khó ở chỗ lúc đó người ta thấy cơ hội không đợi mình @@ -113,12 +105,11 @@ Thủ tục lúc người thân qua đời, thi thể và khoản tiền phải - Mức bằng chứng: C - Nguồn: Kinh nghiệm của tác giả, không có tài liệu trực tiếp; cách diễn đạt «biến cố gia đình, thất nghiệp, mất đi cơ hội học» như rủi ro khủng hoảng tâm lý xem Ủy ban Quốc gia Y tế và Sức khoẻ cùng 25 bộ (2026). Phương án hoàn thiện hệ thống dịch vụ tâm lý xã hội và cơ chế can thiệp khủng hoảng（国卫医政发〔2026〕8 号）. - Ghi chú: Nếu bên cạnh không có người không có lợi ích gì với việc này thì gọi 12356 kể một lần (xem mục 11 của phần này, muốn có người để nói chuyện thì gọi 12356), hoặc tự viết ra rồi ba ngày sau đọc lại. «Ba tháng» là ranh giới do tác giả tự đặt, không có tài liệu nào chống đỡ thời hạn cụ thể này, nên ghi mức C. Những trò lừa đảo thường gặp nhắm vào người mất người thân và người già xem phần 17 (đầu tư chu cấp thu hưu, vay mua nhà bằng tiền hưu trí) và phần 6 (bói quẻ). Bẫy của việc đứng bảo lãnh và giấy nợ xem phần 8. Có vài khoản tiền phải chủ động đi làm, không nằm trong danh sách «đẩy lùi»: số dư quỹ tích góp nhà ở, quyền lợi bảo hiểm xã hội, ba khoản tiền khi chết vì tai nạn lao động, xem phần 25 và phần 19 - -### 13. Đừng dùng cái chết để trả nợ: bảo hiểm nhân thọ trong hai năm không bồi thường, tai nạn lao động không được công nhận, nợ vẫn bị trừ trước tiên từ di sản +### 13. Đừng dùng cái chết làm cách trả nợ: bảo hiểm nhân thọ trong hai năm đầu không bồi, tai nạn lao động không công nhận, và nợ vẫn bị lấy từ tài sản thừa kế trước -- Chi phí: Không tốn tiền, chỉ là làm cho số sách này rõ ràng -- Nói thẳng: Bảo hiểm nhân thọ không bồi thường khi tự tử trong hai năm đầu, chỉ hoàn lại số tiền trong hợp đồng mà được hoàn. Tai nạn lao động cũng không được công nhận, ba khoản tiền khi chết vì tai nạn lao động không có một xu nào. Nợ phải trả trước ra từ di sản bạn để lại, người nhà chỉ nhận phần còn lại. Họ còn phải gánh sức khoẻ mà mục 4 của phần này nói (mất người thân vì tự tử hoặc bạo lực) -- Lợi ích: Điều 44 của Bộ luật Bảo hiểm quy định về hợp đồng có điều kiện trả tiền bảo hiểm khi người được bảo hiểm chết. Tính từ ngày hợp đồng thành lập hoặc ngày hiệu lực được khôi phục, nếu người được bảo hiểm tự tử trong hai năm thì công ty bảo hiểm không phải nghĩa vụ trả tiền bảo hiểm. Không tính trường hợp người được bảo hiểm tự tử khi chưa có năng lực hành vi dân sự, tức người mà pháp luật không cho chịu trách nhiệm về hành vi của chính mình. Trong trường hợp đó công ty bảo hiểm phải hoàn trả giá trị tiền mặt của hợp đồng theo thỏa thuận, tức chỉ hoàn lại số tiền trong hợp đồng được hoàn, không trả số tiền bảo hiểm. Mức bảo hiểm nhân thọ thường ở cỡ vài trăm nghìn tệ, vì vậy cỡ lợi ích đặt là lớn. Điều 16, khoản thứ ba của Quy chế bảo hiểm tai nạn lao động quy định: tự tàn tật hoặc tự tử thì không được công nhận là tai nạn lao động, cũng không được coi là tai nạn lao động. Ba khoản quyền lợi khi chết vì tai nạn lao động ở Điều 39 vì thế mất trắng cả ba. Một là trợ cấp mai táng, trả bằng 6 tháng mức lương trung bình của người lao động năm trước tại địa phương. Hai là tiền tùy táng người nuôi dưỡng. Ba là trợ cấp một lần khi chết vì tai nạn lao động, bằng 20 lần thu nhập chi tiêu bình quân đầu người dân đô thị toàn quốc năm trước. Điều 1159 của Bộ luật Dân sự quy định: khi chia di sản thì phải trừ các khoản thuế và nợ mà người để lại có nghĩa vụ theo pháp luật, tức lấy di sản trả thuế và nợ trước, phần còn lại mới chia. Điều 1161 quy định: người thừa kế trả nợ trong giới hạn giá trị thực tế của di sản được nhận, tức chỉ dùng phần được thừa kế để trả, không phải bỏ thêm ra túi mình; người từ bỏ quyền thừa kế thì không phải chịu trách nhiệm trả nợ +- Chi phí: Không tốn tiền, chỉ là làm rõ cái sổ sách này. +- Nói thẳng: Hợp đồng bảo hiểm nhân thọ không bồi cho trường hợp tự sát trong hai năm đầu, chỉ hoàn lại phần tiền trong hợp đồng mà quy định được hoàn lại. Tai nạn lao động thì cũng không được công nhận, ba khoản tiền khi chết vì tai nạn lao động không có đồng nào. Nợ nợ phải trả lấy từ tài sản bạn để lại trước, người nhà chỉ nhận phần còn lại. Họ còn phải gánh cái giá sức khoẻ mà mục 4 của phần này (mất người thân vì tự sát hoặc bạo lực) nói tới. +- Lợi ích: Điều 44 của Luật Bảo hiểm quy định về những hợp đồng có điều kiện trả tiền bảo hiểm là khi người được bảo hiểm chết. Tính từ ngày hợp đồng được lập hoặc ngày hiệu lực của hợp đồng được khôi phục, nếu người được bảo hiểm tự sát trong vòng hai năm thì công ty bảo hiểm không phải trả tiền bảo hiểm. Trường hợp người được bảo hiểm tự sát khi chưa có năng lực hành vi dân sự, tức người mà pháp luật không công nhận chịu trách nhiệm cho hành vi của mình, thì không thuộc trường hợp trên. Trường hợp này công ty bảo hiểm phải hoàn lại giá trị tiền mặt của hợp đồng theo thỏa thuận, tức chỉ hoàn lại phần tiền mà hợp đồng cho hoàn lại, không trả mứn bảo hiểm. Mứn bảo hiểm của hợp đồng bảo hiểm nhân thọ thường ở mức hàng chục vạn tệ, vì vậy cỡ lợi ích được ghi là lớn. Khoản thứ ba của Điều 16 Quy định về bảo hiểm tai nạn lao động quy định: tự tàn tật hoặc tự sát thì không được công nhận là tai nạn lao động, cũng không được coi là tai nạn lao động. Ba khoản trợ cấp khi chết vì tai nạn lao động quy định ở Điều 39 vì thế mất trắng cả ba. Một là trợ cấp mai táng, trả bằng 6 tháng lương tháng bình quân của người lao động năm trước ở địa phương. Hai là trợ cấp hưu trí cho người được nuôi dưỡng. Ba là trợ cấp một lần khi chết vì tai nạn lao động, bằng 20 lần thu nhập khả dụng bình quân đầu tư dân của dân đô thị cả nước trong năm trước. Điều 1159 của Bộ luật Dân sự quy định: khi chia tài sản thì phải lấy tài sản để bù cho khoản thuế và nợ mà người đã mất có nghĩa vụ phải nộp theo pháp luật, tức lấy tài sản trước để trả thuế và trả nợ, phần còn lại mới chia. Điều 1161 quy định: người thừa kế trả nợ trong phạm vi giá trị thực tế của tài sản được thừa kế, tức chỉ lấy đúng phần được thừa kế ra trả, không phải bỏ thêm tiền của mình vào, và người từ bỏ thừa kế thì không phải chịu trách nhiệm trả nợ - Mức bằng chứng: A -- Nguồn: Ủy ban Thường vụ Quốc hội (2015 sửa đổi). Bộ luật Bảo hiểm nước Cộng hoà nhân dân Trung Quốc（中华人民共和国保险法第四十四条）. ；Hội đồng Nhà nước (2010 sửa đổi). Quy chế bảo hiểm tai nạn lao động（工伤保险条例，国务院令第 586 号，第十六条、第三十九条）. ；Quốc hội (2020). Bộ luật Dân sự nước Cộng hoà nhân dân Trung Quốc（中华人民共和国民法典第一千一百五十九条、第一千一百六十一条）. -- Ghi chú: Ở đây chỉ tính sổ sức khoẻ về tiền. Các mục khác của phần này và của phần 1 nói là đừng đi đến bước này. Điều 44 của Bộ luật Bảo hiểm chỉ nói về hai năm đầu, viết ra là để nêu đủ điều khoản, không phải để đặt ra một mốc thời gian. Sức khoẻ của người nhà xem mục 4 của phần này (vợ chồng có người chết vì tự tử thì trong 5 năm đầu nguy cơ mắc rối loạn tâm thần mới cao hơn bảy đến tám phần mười). Khi có ý nghĩ tự tử thì làm gì xem mục 25 và 32 ở phần 1, di chứng còn lại sau khi được cứu xem mục 33 ở phần 1. Mức ba khoản tiền khi chết vì tai nạn lao động xem mục 16 ở phần 19. Thủ tục cụ thể về di sản và nợ xem mục 9 ở phần 25 (lấy từng khoản quỹ tích góp nhà ở và quyền lợi bảo hiểm xã hội) +- Nguồn: Ủy ban Thường vụ Quốc hội (全国人大常委会) (2015, sửa đổi). Luật Bảo hiểm của Cộng hoà Nhân dân Trung Quốc（中华人民共和国保险法第四十四条）. ；Hội đồng Nhà nước (国务院) (2010, sửa đổi). Quy định về bảo hiểm tai nạn lao động（工伤保险条例）（国务院令第 586 号）第十六条、第三十九条. ；Quốc hội nhân dân (全国人大) (2020). Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc（中华人民共和国民法典）第一千一百五十九条、第一千一百六十一条. +- Ghi chú: Ở đây chỉ tính đúng khoản tiền. Mục khác của phần này và của phần 1 nói về việc đừng đi tới bước đó. Điều 44 của Luật Bảo hiểm chỉ nói tới hai năm đầu, viết ra là để nêu đủ điều khoản chứ không phải để đưa ra một cái thời hạn. Cái giá sức khoẻ cho người nhà xem mục 4 của phần này (người có vợ chồng chết vì tự sát thì trong 5 năm đầu nguy cơ mắc rối loạn tâm thần mới cao hơn bảy đến tám mươi phần trăm). Ý nghĩ đó vừa xuất hiện thì làm gì xem mục 25 và 32 ở phần 1, di chứng sau khi cứu lại xem mục 33 ở phần 1. Tiêu chuẩn ba khoản tiền khi chết vì tai nạn lao động xem mục 15 ở phần 19. Thủ tục cụ thể cho tài sản thừa kế và nợ nợ xem mục 9 ở phần 25 (đi từng nơi một để lấy quỹ tích góp nhà ở và các quyền lợi bảo hiểm xã hội) diff --git "a/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" "b/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" index 24352835..667e9085 100644 --- "a/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" +++ "b/vi/book/30-\344\270\212\345\255\246\344\273\245\345\220\216\347\232\204\345\255\251\345\255\220.md" @@ -1,9 +1,10 @@ -[← 回总目录](../README.md) +[← Về mục lục](../README.md) # 30. Trẻ đang đi học -Mục 20 nói về năm đầu tiên khi đứa trẻ vừa mới sinh. Mục 18 tính câu hỏi nuôi con có đáng không. -Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật sự rơi vào thân thể. Mục 9 (kỳ vọng khi làm việc với cảnh sát, bác sĩ, thủ quầy) dẫn quy định chứ không phải thí nghiệm, nên xếp C. Các mục còn lại xếp B, vì chủ yếu dựa vào một nghiên cứu hoặc khảo sát tự báo cáo. +Phần 20 nói về năm đầu tiên khi trẻ vừa sinh ra. Phần 18 tính xem nuôi trẻ tốn bao nhiêu tiền, bao nhiêu thời gian. Mười mấy năm giữa lứa tuổi 3 đến 18 thì nằm ở phần này. Mục quản lý màn hình theo tuổi viết bắt đầu từ 0 tuổi, vì cách chia bậc chính thức là 0～3 tuổi, 3～6 tuổi và nhà trường phổ thông, tách ra viết lại thì lại không tiện dùng. Phần này trả lời: khi「đừng làm lỡ việc học tập」va vào「sức khoẻ」thì có việc nào không thể đợi. Những việc đó có cửa sổ thời gian, qua rồi thì không bù lại được. Có nên học tập căng thế không, phần này không bàn, chỉ đưa ra cửa sổ và con số. + +Tám mục đầu, mục 12 (khám khúc xạ sau khi làm giãn đồng tử), mục 13 (hàn răng cắt khoé hốm răng) và mục 15 đến 17 tính vào tử vong và vào tổn hại thật sự về thể chất, tâm lý. Trong đó mục 16, 17 nói về việc quản trẻ ở nhà: không đánh không mắng, hãy đi học lớp dạy cha mẹ có cách làm cụ thể. Mục 9 (không mua sản phẩm「chữa khỏi cận thị」) tính vào tiền. Mục 10 (quy định rõ về giấc ngủ, bài tập, thể dục), mục 11 (nghỉ học mà vẫn giữ hồ sơ học tập) và mục 14 (trẻ thích chơi game) tính vào thời gian. Đuối nước, giao thông, mũ bảo hiểm, vắc-xin HPV xem ở phần 1. Trẻ sơ sinh và trẻ nhỏ xem ở phần 20. Trẻ bị lừa và nạp tiền trong game xem ở phần 5. Vạch đỏ mà người chưa thành niên tự bước vào xem ở phần 9. Nhận biết cấp cứu và gọi xe cứu thương xem ở phần 13. Trẻ sau khi bố mẹ qua đời xem ở phần 29. Những việc đó phần này không lặp lại. ### 1. Trẻ bảo đau dữ dội bất ngờ và đau hơn nữa thì đưa đi bệnh viện ngay, đừng để cháu nhịn đến tan học @@ -13,7 +14,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: A - Nguồn: Mellick LB, et al. (2019). A Systematic Review of Testicle Survival Time After a Torsion Event. Pediatric Emergency Care. - Ghi chú: Nói trước với bé trai một câu: nếu bỗng đau dữ dội bên dưới thì nói ngay, đừng ngượng. Xoắn tinh hoàn hay gặp ở các bé trai tuổi thanh niên, nhiều người nhịn vì khó mở lời. Những bệnh khác cũng tính theo giờ như vậy là viêm ruột thừa cấp (kéo dài sẽ thủng), nhiễm toan ceton do đái tháo đường, viêm màng não. Phần này chưa kiểm chứng từng bệnh một xem trễ mấy giờ thì hỏng đến đâu. Dấu hiệu chung của chúng là bắt đầu đột ngột, liên tục nặng thêm, khác hẳn mọi lần trước. Nhận biết cấp cứu và gọi xe cứu thương, xem phần 13. - ### 2. Không nên vì「chờ thi xong」mà trì hoãn việc chữa trị cần làm, có những cửa sổ thời gian đi theo tuổi xương chứ không theo lịch thi - Chi phí: Không tốn tiền, chỉ phải quyết định làm sớm hay làm muộn. Nghỉ và tái khám tốn thời gian, tiền chữa trị tính riêng. Chỗ khó là đai nẹo bắt đeo mỗi ngày ít nhất 18 giờ, trẻ phải chịu suốt thời gian đó. @@ -22,7 +22,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: A - Nguồn: Weinstein SL, et al. (2013). Effects of bracing in adolescents with idiopathic scoliosis. New England Journal of Medicine. - Ghi chú: Ở đây lấy vẹo cột sống làm ví dụ vì nó có thử nghiệm chia nhóm ngẫu nhiên, có cửa sổ thời gian rõ ràng, và vừa là một trong các mục chính của khám sức khoẻ học sinh (xem mục 7 của phần này). Cùng loại còn có lách, nhược thị, sâu răng, và những ca mổ bệnh tim bẩm sinh có thể chọn thời điểm. Điểm chung của chúng là làm càng muộn thì hiệu quả càng kém, hoặc cái giá càng lớn. Cửa sổ thời gian cụ thể thì hỏi bác sĩ chuyên khoa. Sợ ảnh hưởng giờ học, trong giai đoạn giáo dục bắt buộc có thể xin nghỉ học, tối đa 1 năm, học vị được giữ, xem mục 11 của phần này. - ### 3. Trẻ bị bắt nạt thì báo ngay trong ngày cho nhà trường và yêu cầu xử lý bằng văn bản, có đánh nhau, cướp tiền, tung tin thì báo công an - Chi phí: Không tốn tiền. Tìm giáo viên chủ nhiệm, tìm cả tổ chuyên trách xử lý bạo lực học đường của trường. Muốn báo công an thì gọi 110. Chỗ khó là phải chịu được lời khuyên kiểu「đừng làm ầm lên」. @@ -31,7 +30,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: A - Nguồn: van Geel M, et al. (2014). Relationship between peer victimization, cyberbullying, and suicide in children and adolescents: a meta-analysis. JAMA Pediatrics. ；Bộ Giáo dục (教育部) (2021). Quy định về bảo vệ học sinh vị thành niên, 未成年人学校保护规定（教育部令第 50 号，第二十一、二十二、二十三条）. - Ghi chú: Trên người trẻ có vết thương không giải thích được, bỗng nhiên không chịu đi học, đồ đạc bị mất đi đi lại, trước hết hãy hỏi theo các dấu hiệu nêu trong 第二十二条. Bản phân tích gộp nhiều nghiên cứu này chỉ ghi nhận và theo dõi, không có nhóm đối chứng, nên không chứng minh chỉ riêng bắt nạt đã gây tự sát, nhưng hướng và độ lớn của rủi ro thì rõ ràng. 《Phương án tăng cường quản lý toàn diện bạo lực học đường đối với học sinh tiểu học và trung học》《加强中小学生欺凌综合治理方案》còn yêu cầu phân biệt nghiêm ngặt bạo lực với chuyện đánh nhau vui đùa giữa học sinh. Vì vậy nhà trường có nghĩa vụ lập kết luận, không thể cho xong bằng một câu「trẻ con chơi thôi mà」. - ### 4. Cho trẻ ở ngoài trời đủ 2 giờ mỗi ngày, đây là cách phòng cận thị duy nhất hiện có thử nghiệm ngẫu nhiên đứng sau - Chi phí: Không tốn tiền. Nhưng mỗi ngày phải dành ra thời gian, phần lớn là phải bóc ra từ thời gian làm bài và học online. Chỗ khó là làm được mỗi ngày, không phải làm mấy lần là xong. @@ -49,7 +47,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: B - Nguồn: Văn phòng của Ủy ban Quốc gia Y tế (国家卫生健康委办公厅) (2023). Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên, 防控儿童青少年近视核心知识十条（国卫办妇幼函〔2023〕278 号）. ; Bộ Giáo dục và bảy bộ khác (教育部等八部门) (2018). Phương án triển khai phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên toàn diện, 综合防控儿童青少年近视实施方案（教体艺〔2018〕3 号）. ; Foreman J, Salim AT, Praveen A, et al. (2021). Association between digital smart device use and myopia: a systematic review and meta-analysis. The Lancet Digital Health, 3(12), e806-e818. ; Madigan S, McArthur BA, Anhorn C, Eirich R, Christakis DA (2020). Associations between screen use and child language skills: a systematic review and meta-analysis. JAMA Pediatrics, 174(7), 665-675. - Ghi chú: Cho mức B là vì cái cơ quan đưa ra là quy định chia theo nhóm tuổi, không phải con số đo từ thử nghiệm. Nghiên cứu cho tỷ số odds có cận dưới vừa đúng ở 1.00, nên tác giả chỉ dám viết「可能相关」. Vì vậy cứ coi đây là một quy tắc gia đình dễ thực hiện, đừng hiểu là「không xem màn hình thì không cận thị」. Cái có thử nghiệm ngẫu nhiên bảo chứng là mục 4 (2 giờ ở ngoài trời). Trẻ mẫu giáo cũng cần khám mắt: cùng tài liệu đó ghi rõ 1～3 tuổi, 4～6 tuổi và sau 7 tuổi đều nên làm khám khúc xạ định kỳ để xem còn lại bao nhiêu dự trữ thị lực, cách khám xem mục 12 của phần này (khám khúc xạ sau khi nhỏ mắt). Quan hệ giữa màn hình và khả năng ngôn ngữ của trẻ cũng đã có phân tích tổng hợp: 42 nghiên cứu, 18905 người, dùng màn hình lâu hơn thì khả năng ngôn ngữ yếu hơn (r=−0.14, 95% CI −0.18 đến −0.10). Nhưng xem chung, xem chương trình có nội dung giáo dục lại ra tương quan thuận (r=0.16 và 0.13). Đây là tương quan chứ không phải nhân quả, chỉ cho thấy dùng màn hình thay cho việc nói chuyện với trẻ là không đáng. - ### 6. Hãy coi cận thị là một tổn thương còn lại: độ càng cao thì sau này càng nhiều bong võng mạc, thoái hoa điểm vàng và glôcôm - Chi phí: Không tốn tiền. Bản thân cận thị không cần làm gì, việc phải làm nằm ở mục 4 và 5 của phần này (ngoài trời 2 giờ, quản lý màn hình theo tuổi). @@ -58,7 +55,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: A - Nguồn: Haarman AEG, Enthoven CA, Tideman JWL, Tedja MS, Verhoeven VJM, Klaver CCW (2020). The complications of myopia: a review and meta-analysis. Investigative Ophthalmology & Visual Science, 61(4), 49. ; Văn phòng của Ủy ban Quốc gia Y tế (国家卫生健康委办公厅) (2023). Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên, 防控儿童青少年近视核心知识十条（国卫办妇幼函〔2023〕278 号）. - Ghi chú: Những nghiên cứu được gom vào đây đều là chỉ ghi nhận và theo dõi, tỷ số odds lớn đến mức bất thường. Khoảng tin cậy của mục thoái hoa hoàng điểm ở cận thị cao trải từ 230 đến 3104, nghĩa là các nghiên cứu bên dưới rất khác nhau, con số này chỉ xem được độ lớn chứ không dùng làm giá trị chính xác. Phần lớn các biến chứng này chỉ lộ ra khi đã trung niên hoặc cao niên. Bản thân cận thị không phải là tàn tật, đa số người đeo kính là nhìn bình thường. Nhưng cận thị không phải là chuyện cứ mua một cặp kính là xong, đáng bỏ công sức để hoãn lại và kiềm soát nó. Sau khi xác định chẩn đoán thì tái khám theo bao lâu xem mục 12 của phần này (khám khúc xạ sau khi nhỏ mắt). Đừng mua sản phẩm quảng cáo là chữa khỏi, xem mục 9 của phần này (chữa khỏi cận thị). - ### 7. Báo cáo khám sức khoẻ học sinh hằng năm phải tự đọc một lượt, mục bất thường thì đưa đi khám trong năm đó - Chi phí: Không tốn tiền. Tiền khám sức khoẻ học sinh trong giai đoạn giáo dục bắt buộc do ngân sách công của trường chi trả, bố mẹ không phải bỏ thêm ra. @@ -67,7 +63,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: B - Nguồn: Ủy ban Quốc gia Y tế và Bộ Giáo dục (国家卫生健康委、教育部) (2021). Biện pháp quản lý khám sức khoẻ học sinh tiểu học và trung học (bản 2021), 中小学生健康体检管理办法（2021 年版）（国卫医发〔2021〕29 号）. - Ghi chú: Cho mức B vì đây là văn bản quy định trường phải làm thế nào, không phải nghiên cứu đo hiệu quả. Nó bảo đảm「mỗi năm khám, khám đúng những mục này, thấy bất thường thì phải theo dõi」, chứ không bảo đảm khám ra xong sẽ có người đứng ra xử lý. Khám sức khoẻ chỉ là sàng lọc một lượt, không phải chẩn đoán. Nhìn hở thì đi mắt làm khám khúc xạ sau khi nhỏ mắt (mục 12 của phần này). Bất thường cong cột sống thì đi khoa chỉnh hình hoặc khoa ngoại cột sống (mục 2 của phần này, đừng trì hoãn chữa trị vì chờ thi xong). Hai bước này đều phải bố mẹ tự đưa trẻ đi. - ### 8. Cho trẻ 12 đến 18 tuổi làm một lần sàng lọc trầm cảm, đừng lấy bài tâm lý học đường làm chẩn đoán - Chi phí: 0 đến mấy chục tệ, điền một bảng khảo sát sàng lọc ở trạm y tế cộng đồng hoặc phòng khám bệnh viện. Hai đường dây 12355 và 12356 gọi thì không mất tiền. @@ -76,7 +71,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: B - Nguồn: US Preventive Services Task Force (2022). Screening for Depression and Suicide Risk in Children and Adolescents: US Preventive Services Task Force Recommendation Statement. JAMA. ；Văn phòng của Bộ Giáo dục (教育部办公厅) (2025). Mười biện pháp tăng cường công tác sức khoẻ tâm lý cho học sinh tiểu học và trung học, 进一步加强中小学生心理健康工作十条措施（教基厅〔2025〕2 号）. - Ghi chú: Cho mức B vì nguyên bản đưa ra là xếp hạng mức khuyến nghị, không phải hiệu quả quy đổi được thành số. Đó còn là khuyến nghị của Mỹ, trong nước không có khuyến nghị sàng lọc chính thức tương ứng. Kết quả sàng lọc dương tính chỉ nói là cần đánh giá lại một lần, không có nghĩa là đã chẩn đoán. Khi trẻ có ý nghĩ tự sát thì xử lý xem mục 25 ở phần 1. Cách xử lý ở trong nhà xem phần 29. - ### 9. Đừng mua sản phẩm và dịch vụ quảng cáo là có thể「chữa khỏi cận thị」「giảm độ」 - Chi phí: Không tốn tiền, ngược lại còn tiết kiệm được khoản tiền đó. @@ -94,7 +88,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: B - Nguồn: Văn phòng Bộ Giáo dục (教育部办公厅) (2021). Thông báo về việc tăng cường quản lý giấc ngủ học sinh tiểu học và trung học（教基厅函〔2021〕11 号）. ；Văn phòng Bộ Giáo dục (教育部办公厅) (2025). Mười biện pháp tăng cường công tác sức khoẻ tâm lý cho học sinh tiểu học và trung học（教基厅〔2025〕2 号）. - Ghi chú: Cả hai đều là văn bản của cơ quan chủ quản. Mức độ các trường các nơi thực hiện chênh nhau rất nhiều, nên mục này để mức B. Chúng có ích là để bạn biết việc nào trường vốn đã phải làm, để khi đàm phán bạn có căn cứ trong tay. Về giấc ngủ của người lớn, xem phần 2 và 3. - ### 11. Con không chịu nổi thì có thể xin nghỉ học, trường bắt buộc giữ học vị cho cháu, tối đa 1 năm - Chi phí: Không tốn tiền. Cha mẹ hoặc người giám hộ hợp pháp khác nộp đơn lên nhà trường. Chỗ khó là phải chấp nhận cho con dừng học một thời gian. @@ -103,7 +96,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: B - Nguồn: Bộ Giáo dục (教育部) (2025). Quy định quản lý học vị học sinh tiểu học và trung học《中小学生学籍管理办法》（教基〔2025〕1 号，Điều mười sáu）. ；Văn phòng Bộ Giáo dục (教育部办公厅) (2025). Mười biện pháp tăng cường công tác sức khoẻ tâm lý cho học sinh tiểu học và trung học（教基厅〔2025〕2 号）. - Ghi chú: Trước khi đi làm thủ tục thì hỏi rõ ba việc: cần xin loại giấy tờ y tế nào, khoảng thời gian nghỉ học thì học vị được ghi thế nào trong hệ thống, khi trở lại học là về nguyên lớp hay xuống một lớp. Để mức B là vì điều kiện nghỉ học, hồ sơ phải nộp và cách nối lại khi trở lại học đều do cơ quan giáo dục cấp tỉnh tự quy định, phần cấp quốc gia chỉ dựng khung. Quy định về bảo vệ học sinh vị thành niên《未成年人学校保护规定》còn yêu cầu trường giáo dục bắt buộc lập hồ sơ báo cáo cho học sinh bỏ học, nghỉ học và nghỉ phép dài ngày, nên nghỉ học là việc trường đã làm qua, không phải mở ngoại lệ riêng cho bạn. - ### 12. Khám ra nhìn kém thì ra bệnh viện khám khúc xạ sau khi nhỏ mắt, sau đó tái khám theo khoảng cách bác sĩ chỉ định - Chi phí: Phí khám và phí kính từ mấy chục tới mấy trăm tệ. Sau khi nhỏ mắt, trẻ sợ ánh sáng và nhìn gần bị mờ, phải nửa ngày đến một hai ngày mới hết. @@ -112,7 +104,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: B - Nguồn: Văn phòng của Ủy ban Quốc gia Y tế (国家卫生健康委办公厅) (2023). Mười điểm cốt lõi về phòng và kiểm soát cận thị ở trẻ em và thanh thiếu niên, 防控儿童青少年近视核心知识十条（国卫办妇幼函〔2023〕278 号）. - Ghi chú: Để mức B là vì đây là cách làm do tài liệu phổ biến chính thức đưa ra, không phải nghiên cứu gốc có số liệu cụ thể. Cận thị nặng sẽ kéo theo những bệnh như bong võng mạc, nên tái khám không chỉ để đổi kính. Kết quả「nhìn kém」trong khám sức khoẻ học sinh chỉ là kết quả sàng lọc ban đầu, vẫn còn mang con đi bệnh viện làm một lần khám mắt đầy đủ, xem mục 7 của phần này. - ### 13. Sau khi răng cối chính mọc lên thì làm hàn khe răng - Chi phí: Mỗi răng mấy chục tới một hai trăm tệ. Không gây tê, cũng không phải mài răng, mỗi lần mấy phút. Nhiều nơi có chương trình miễn phí cho trẻ đủ tuổi. @@ -121,7 +112,6 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: A - Nguồn: Ahovuo-Saloranta A, et al. (2017). Pit and fissure sealants for preventing dental decay in permanent teeth. Cochrane Database of Systematic Reviews. - Ghi chú: Răng cối chính thứ nhất (răng lớn không thay nữa) thường mọc khoảng 6 tuổi, răng thứ hai khoảng 12 tuổi. Hàn phải làm lúc răng vừa mọc, chưa hỏng, đã sâu rồi thì chỉ còn vá. Nhiều khu vực đưa hàn khe răng cho trẻ đủ tuổi vào danh sách chương trình miễn phí, hỏi trung tâm kiểm soát bệnh địa phương hoặc trạm y tế cộng đồng. Trong phần này không tìm được văn bản phạm vi toàn quốc nào đối chiếu từng chữ, nên không viết phạm vi bảo hiểm và nhóm tuổi. Sâu răng cũng là một trong các mục trọng tâm hướng dẫn trong khám sức khoẻ học sinh, xem mục 7 của phần này. - ### 14. Con thích chơi game thì xem giấc ngủ, bài tập và ra ngoài vận động có bị chen mất không, đừng chỉ nhìn thời gian chơi - Chi phí: Không tốn tiền. Mỗi tuần nên biết một lần con đi ngủ lúc mấy giờ, bài tập có nộp không, có ra ngoài vận động không. @@ -130,6 +120,7 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: B - Nguồn: Stevens MW, Dorstyn D, Delfabbro PH, King DL. (2021). Global prevalence of gaming disorder: A systematic review and meta-analysis. Australian and New Zealand Journal of Psychiatry, 55(6), 553–568. ；Vuorre M, Johannes N, Magnusson K, Przybylski AK. (2022). Time spent playing video games is unlikely to impact well-being. Royal Society Open Science, 9(7), 220411. ；Cục Quản lý Báo chín và Xuất bản (国家新闻出版署) (2021). Thông báo về việc quản lý chặt hơn nữa, ngăn người chưa thành niên nghiện game trực tuyến（国新出发〔2021〕14 号）. - Ghi chú: Để mức B là vì tỷ lệ mắc thay đổi quá nhiều theo bảng hỏi, và nghiên cứu về hạnh phúc chỉ theo dõi sáu tuần. Cái giá phổ biến nhất của game là chen mất giấc ngủ và thời gian ở ngoài trời. Yêu cầu rõ ràng về giấc ngủ xem mục 10 của phần này（giấc ngủ, bài tập, thể dục）, thời gian ở ngoài trời xem mục 4 của phần này（2 giờ ở ngoài trời mỗi ngày）, thời lượng dùng màn hình xem mục 5 của phần này（không phải để học thì mỗi ngày không quá 1 giờ）. Nạp tiền và hoàn tiền xem mục 9 ở phần 5（trẻ nạp tiền và tặng quà trên điện thoại）. Kết quả học tập, giấc ngủ và quan hệ với người nhà cùng sụp xuống mà khuyên cũng không ngừng được, lúc đó đưa con đi khoa tâm thần hoặc khoa tâm lý trẻ em. Người hưởng lợi là con. + ### 15. Trẻ nói mình thích người cùng giới, đừng mắng, đừng đuổi khỏi nhà, đừng đưa đi「cải tận」: thái độ của gia đình có liên quan đến chuyện trẻ có tự sát hay không - Chi phí: Không tốn tiền. Chỗ khó là bố mẹ phải xử lý cảm xúc của mình trước, và lúc đó không nói lời độc ác. @@ -138,3 +129,21 @@ Tám mục đầu và hai mục cuối tính cái chết và tổn hại thật - Mức bằng chứng: B - Nguồn: Huang Y, Li P, Guo L, et al. (2018). Sexual minority status and suicidal behaviour among Chinese adolescents: a nationally representative cross-sectional study. BMJ Open, 8(8), e020969. ；di Giacomo E, Krausz M, Colmegna F, Aspesi F, Clerici M. (2018). Estimating the Risk of Attempted Suicide Among Sexual Minority Youths: A Systematic Review and Meta-analysis. JAMA Pediatrics, 172(12), 1145–1152. ；Ryan C, Huebner D, Diaz RM, Sanchez J. (2009). Family rejection as a predictor of negative health outcomes in white and Latino lesbian, gay, and bisexual young adults. Pediatrics, 123(1), 346–352. ；Ryan C, Toomey RB, Diaz RM, Russell ST. (2020). Parent-Initiated Sexual Orientation Change Efforts With LGBT Adolescents: Implications for Young Adult Mental Health and Adjustment. Journal of Homosexuality, 67(2), 159–173. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (sửa đổi năm 2024). Luật bảo vệ người chưa thành niên（未成年人保护法）（第十七条）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2015). Luật chống bạo hành gia đình（反家庭暴力法）（第二条）. - Ghi chú: Cho mức B vì đây đều là khảo sát một lần, mức độ phản đối của gia đình phải nhớ lại khi đã trưởng thành, nên chỉ cho thấy có liên quan chứ không cho biết ai gây ra ai. Bố mẹ không cần phải đồng ý ngay, chỉ cần đừng làm việc làm tổn hại con. Về mặt pháp luật, cha mẹ không được ngược đãi, bỏ rơi con, cũng không được áp dụng bạo hành gia đình với con. Chửi bới thường xuyên, đe dọa cũng tính là bạo hành gia đình. Trẻ tâm trạng u buồn, nói không muốn sống thì gọi 12356 trước, xem mục 25 ở phần 1 (khi có ý nghĩ tự sát thì gọi 12356). Cũng có thể đi khoa Tâm thần theo mục 8 của phần này (làm một lần sàng lọc trầm cảm cho trẻ 12 đến 18 tuổi). Vì sao đừng đụng vào「cải tận」, xem mục 28 ở phần 6 (không làm cải tận xu hướng tính dục). Người hưởng lợi là đứa trẻ, và cả bạn. +### 16. Quản trẻ thì đừng ra tay đánh, cũng đừng quát mắng hay chửi trẻ là ngu: trẻ hay bị đánh mắng thì vấn đề hành vi và vấn đề tâm lý đều nhiều hơn + +- Chi phí: Không tốn tiền. Khó ở chỗ phải dừng tay và dừng miệng đúng lúc cơn giận lên, rồi còn phải chuyển sang cách quản trẻ không đánh không mắng. Dùng cách thay thế nào thì xem mục 17 của phần này (lớp dạy cha mẹ dạy cách làm cụ thể). +- Nói thẳng: Trẻ hay bị đánh thì hành vi gây hại và vấn đề tâm lý nhiều hơn, xa bố mẹ hơn, và cũng hay bị đánh nặng hơn. Đánh có làm trẻ ngoan ngay lúc đó không, bằng chứng không rõ. Trẻ 13 tuổi hay bị quát, bị chửi là ngu thì sau đó gây rắc rối và trầm cảm nhiều hơn, bố mẹ bình thường có thân thiết đến đâu cũng không bù được. +- Lợi ích: Trước hết xem việc đánh. Một bài phân tích tổng hợp chỉ tính loại「vỗ mông bằng tay, hoặc đánh tay chân」. Nó gộp 75 nghiên cứu, 111 cỡ hiệu ứng, tổng cộng 160927 trẻ. Trong 17 kết quả thì 13 kết quả liên quan rõ với việc bị đánh, chiều đều là xấu. d là cỡ hiệu ứng, 0.2 tính là nhỏ, 0.5 tính là trung bình. Hành vi gây hại lúc nhỏ d = 0.37 (95% CI 0.13–0.61, khoảng tin cậy). Vấn đề ngoại tiết 0.41, tức đánh người, vi phạm quy tắc. Vấn đề nội tiết 0.24, tức lo âu, trầm cảm. Vấn đề sức khoẻ tâm thần 0.53, quan hệ với bố mẹ kém 0.51, lòng tự trọng thấp 0.15, năng lực nhận thức kém 0.17. Bị bố mẹ gây bạo hành thân xác 0.64. Hành vi phản xã hội khi lớn lên 0.36, vấn đề sức khoẻ tâm thần 0.24. Gộp 111 cỡ hiệu ứng lại thì d = 0.33 (0.29–0.38). Trẻ có vâng lời ngay lúc đó hay không, không thấy liên quan rõ với việc có bị đánh (d = 0.14, −0.19–0.47), mục này chỉ có 5 cỡ hiệu ứng. Một bài tổng quan khác chỉ xem các nghiên cứu có theo dõi trước sau, tổng cộng 69 nghiên cứu. Phạt tay thể ổn định dự báo trẻ sau đó có nhiều vấn đề hành vi hơn, không thấy nó gắn với kết quả tốt dài hạn nào. Trong đó có 7 nghiên cứu xem số lần bị đánh, 5 nghiên cứu thấy đánh càng nhiều thì vấn đề càng nặng. Tiếp theo xem quát mắng. Một nghiên cứu ở Mỹ theo dõi 976 gia đình có cả hai vợ chồng. Khi trẻ 13 tuổi, nếu bố mẹ hay quát, chửi thô tục, chửi trẻ là ngu hoặc lười, thì trong khoảng từ 13 đến 14 tuổi số vụ gây rắc rối và các triệu chứng trầm cảm đều tăng. Bố mẹ bình thường có thân thiết đến đâu thì cũng không bù được liên hệ này. Chiều ngược lại cũng đúng: trẻ 13 tuổi hay gây rắc rối thì sau đó càng hay bị mắng. Về mặt pháp luật, cha mẹ không được áp dụng bạo hành gia đình với con. Điều 2 của Luật Chống bạo hành gia đình (《反家庭暴力法》) liệt kê các hình thức bạo hành gia đình: đánh, trói, làm tổn thương, hạn chế tự do cá nhân, cùng với chửi bới và đe dọa thường xuyên. Điều 12 của luật đó yêu cầu người giám hộ phải giáo dục con bằng cách văn minh. Cha mẹ xâm phạm quyền lợi của trẻ thì ủy ban tổ dân cư, ủy ban thôn phải khuyên can, ngăn chặn, tình hình nghiêm trọng thì báo công an. Công an, viện kiểm sát, tòa án phát hiện thì phải nhắc nhở cha mẹ, và có thể yêu cầu họ tiếp nhận hướng dẫn giáo dục gia đình (toàn quốc) +- Mức bằng chứng: A +- Nguồn: Gershoff ET, Grogan-Kaylor A. (2016). Spanking and child outcomes: Old controversies and new meta-analyses. Journal of Family Psychology, 30(4), 453–469. ；Heilmann A, Mehay A, Watt RG, et al. (2021). Physical punishment and child outcomes: a narrative review of prospective studies. Lancet, 398(10297), 355–364. ；Wang MT, Kenny S. (2014). Longitudinal links between fathers' and mothers' harsh verbal discipline and adolescents' conduct problems and depressive symptoms. Child Development, 85(3), 908–923. ；Ferguson CJ. (2013). Spanking, corporal punishment and negative long-term outcomes: a meta-analytic review of longitudinal studies. Clinical Psychology Review, 33(1), 196–208. ；Larzelere RE, Kuhn BR. (2005). Comparing child outcomes of physical punishment and alternative disciplinary tactics: a meta-analysis. Clinical Child and Family Psychology Review, 8(1), 1–37. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2015). Luật Chống bạo hành gia đình (《反家庭暴力法》)(第二、十二条). ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2024 年修正). Luật bảo vệ người chưa thành niên (《未成年人保护法》)(第十一、十七、一百一十八条). ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2021). Luật thúc đẩy giáo dục gia đình (《家庭教育促进法》)(第二十三、五十三条). +- Ghi chú: Tranh cãi. Gần như các nghiên cứu này chỉ theo dõi ghi nhận, không có nhóm đối chứng, nên không nói được cái nào dẫn tới cái nào. Trong bài phân tích tổng hợp về vỗ mông, 72% nghiên cứu chỉ xem hai việc có xảy ra cùng nhau không, hoặc dựa vào trí nhớ sau sự kiện. Một bài phân tích tổng hợp khác chỉ thu những nghiên cứu có theo dõi trước sau, sau khi trừ hành vi vốn có của trẻ thì liên hệ giữa vỗ mông và vấn đề ngoại tiết chỉ còn 0.07, với vấn đề nội tiết chỉ còn 0.10. Tác giả của bài đó cho rằng ảnh hưởng rất nhỏ. Lại có một bài phân tích tổng hợp phát hiện「có điều kiện mà đánh」hiệu quả hơn 10 trong số 13 cách quản trẻ khác, khi nói đến giảm hành vi bất tuân và hành vi phản xã hội. Nó cũng phát hiện đánh quá tay hoặc lấy đánh làm chính thì kém hơn các cách khác. Nên phía đối lập chỉ chứng minh được điều này: đánh ít, có điều kiện thì chưa chắc kém hơn cách khác. Đánh nặng, đánh mắng thường xuyên thì cả hai phía đều cho là tệ hơn. Thỉnh thoảng đánh một cái, quát một câu, luật không nói thẳng là bạo hành gia đình, luật nêu tên là đánh và chửi bới, đe dọa thường xuyên. Người hưởng lợi chủ yếu là đứa trẻ. Người khác trong nhà đánh trẻ thì báo công an và xin lệnh bảo vệ, xem mục 43 ở phần 8 (lệnh bảo vệ an toàn cá nhân). Thấy nhà người ta đánh trẻ lâu dài thì bất kỳ ai cũng có quyền khuyên can, hoặc báo công an, bộ phận dân vụ, sở giáo dục. Với chuyện trong nhà người lạ thì gọi 110 hoặc báo ủy ban tổ dân cư là đủ, đừng tự đến nhà can thiệp bằng tay. + + +### 17. Trẻ khó quản, bản thân lúc nào cũng không nhịn được mà đánh mắng, thì đi học một khoá dạy cha mẹ có cách làm cụ thể + +- Chi phí: Lớp học cho cha mẹ do trường, trường mầm non và cộng đồng mở thì hoạt động phi lợi nhuận, không thu phí. Loại lớp có bằng chứng thì học mỗi tuần hoặc cách tuần, mỗi buổi một đến hai giờ, từ bốn tuần đến hai mươi bốn tuần. Về nhà phải luyện theo. Khó ở chỗ phải học cho đủ số buổi và luyện mỗi ngày. +- Nói thẳng: Lớp dạy cha mẹ cách quản trẻ cụ thể, trong thử nghiệm chia nhóm ngẫu nhiên, làm trẻ đánh người, không vâng lời giảm đi, tâm trạng cha mẹ cũng tốt hơn. Cha mẹ đánh mắng trẻ cũng giảm, rõ nhất ngay sau khi học xong, một hai năm sau vẫn còn nhưng nhỏ hơn nhiều. Thượng Hải cũng đã thử loại lớp này, có hiệu quả. +- Lợi ích: Cochrane gộp 13 thử nghiệm, trong đó 10 thử nghiệm chia nhóm ngẫu nhiên, tổng cộng 1078 cha mẹ. Trẻ từ 3 đến 12 tuổi, đều có vấn đề hành vi. Cha mẹ có học thì vấn đề hành vi của trẻ giảm. Cha mẹ tự đánh giá là SMD −0.53 (95% CI −0.72 đến −0.34, khoảng tin cậy), người ngoài đánh giá độc lập là −0.44 (−0.77 đến −0.11). SMD và d là cùng một thước đo, 0.5 tính là trung bình. Sức khoẻ tâm thần của cha mẹ cải thiện, −0.36 (−0.52 đến −0.20). Cách quản trẻ tiêu cực hoặc nghiêm khắc giảm, cha mẹ tự đánh giá −0.77 (−0.96 đến −0.59), người ngoài đánh giá −0.42 (−0.67 đến −0.16). Tác giả nói đây là hiệu quả ngắn hạn. Một bài phân tích tổng hợp khác lọc ra 346 thử nghiệm chia nhóm ngẫu nhiên, trong đó 60 thử nghiệm đo hành vi đánh mắng trẻ, rải ở 22 nước. Hành vi đánh mắng giảm ngay sau khi học xong, d = −0.46 (−0.59 đến −0.33). Sau 1 đến 6 tháng là −0.24 (−0.37 đến −0.11). Sau 7 đến 24 tháng là −0.18 (−0.34 đến −0.02), càng lâu càng nhỏ. Phần lớn số liệu này do chính cha mẹ tự báo. Ở trong nước cũng có thử nghiệm. Ở Thượng Hải có 81 cha mẹ chia nhóm ngẫu nhiên, một phần học lớp nhóm Triple P. Học xong thì vấn đề thích ứng của trẻ, cách quản trẻ và tâm trạng của cha mẹ đều tốt hơn nhóm chờ học, 6 tháng sau vẫn còn. Một trường mầm non ở nông thôn có 191 người lớn đưa trẻ đến học tham gia, một phần tự học khoá trực tuyến, rồi thêm hai buổi thảo luận nhóm. Sau hai tuần, hành vi gây bạo hành trẻ mà họ tự báo thấp hơn khoảng 33% so với nhóm chỉ được phát tài liệu để đọc (IRR 0.67). Sau ba tháng, vấn đề hành vi của trẻ cũng giảm (d = −0.40). Luật thúc đẩy giáo dục gia đình (《家庭教育促进法》) yêu cầu cha mẹ chủ động học kiến thức giáo dục gia đình. Trường phổ thông và trường mầm non được mở lớp học cho cha mẹ, định kỳ tổ chức hướng dẫn giáo dục gia đình phi lợi nhuận. Ủy ban tổ dân cư, ủy ban thôn được mở lớp học cha mẹ ở cộng đồng. Chính quyền cấp tỉnh trở lên phải mở lớp học cha mẹ trực tuyến phi lợi nhuận và đường dây nóng. Cơ quan hướng dẫn giáo dục gia đình không được tổ chức đào tạo vì lợi nhuận (toàn quốc, quy định có hiệu lực từ ngày 1 tháng 1 năm 2022) +- Mức bằng chứng: A +- Nguồn: Furlong M, McGilloway S, Bywater T, Hutchings J, Smith SM, Donnelly M. (2012). Behavioural and cognitive-behavioural group-based parenting programmes for early-onset conduct problems in children aged 3 to 12 years. Cochrane Database of Systematic Reviews, (2), CD008225. ；Backhaus S, Leijten P, Jochim J, Melendez-Torres GJ, Gardner F. (2023). Effects over time of parenting interventions to reduce physical and emotional violence against children: a systematic review and meta-analysis. eClinicalMedicine, 60, 102003. ；Guo M, Morawska A, Sanders MR. (2016). A Randomized Controlled Trial of Group Triple P With Chinese Parents in Mainland China. Behavior Modification, 40(6), 825–851. ；Wang W, Zhang H. (2026). A self-directed parenting program to reduce the risk of child maltreatment in rural China: A cluster randomized controlled trial. Child Abuse & Neglect, 173, 107937. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2021). Luật thúc đẩy giáo dục gia đình (《家庭教育促进法》)(第十八、二十五、三十一、三十八、四十条). +- Ghi chú: Đặt mức A vì có số liệu gộp từ Cochrane và nhiều thử nghiệm chia nhóm ngẫu nhiên. Các lớp trong thử nghiệm có vài điểm chung. Người dạy dạy cách làm cụ thể, ví dụ chơi cùng trẻ, khen, thưởng và kỷ luật. Cha mẹ tập, diễn lại trên lớp, về nhà còn có bài tập. Lớp chỉ phát tài liệu hoặc chỉ thảo luận thì bị loại ra, nên bằng chứng không phủ được hình thức chỉ nghe một buổi nói. Lớp học cha mẹ phi lợi nhuận trong nước dạy cái gì, dạy thế nào thì mỗi nơi một khác, chưa chắc là loại này. Khi chọn lớp thì xem có liên kết mấy tuần liền không, có buộc bạn tập cách làm cụ thể không. Những tên lớp thường gặp có Triple P, Incredible Years, 亲子互动治疗 (PCIT). Cẩm nang này không giới thiệu cơ quan hay sản phẩm có thu phí cụ thể. Hiệu quả giảm đánh mắng sau một hai năm nhỏ đi, học xong vẫn phải luyện tiếp. Người hưởng lợi là con bạn, và cả bạn. Trẻ đã có vấn đề tâm lý nặng thì đưa đi khoa Tâm thần hoặc khoa Tâm lý trẻ em, xem mục 8 của phần này (làm một lần sàng lọc trầm cảm cho trẻ 12 đến 18 tuổi). \ No newline at end of file diff --git "a/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" "b/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" index 2a56dec8..78c54beb 100644 --- "a/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" +++ "b/vi/book/31-\345\215\201\345\205\253\345\262\201\344\271\213\345\220\216\346\234\211\345\223\252\345\207\240\346\235\241\350\267\257.md" @@ -14,7 +14,6 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Mức bằng chứng: A - Nguồn: Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2021). Luật nghĩa vụ quân sự（兵役法）（第二十条）. （đăng trên Bộ Quốc phòng）；Ban Tổ chức Trung ương Đảng Cộng sản Trung Quốc (中共中央组织部) (2019). Quy định tuyển dụng công chức（公务员录用规定）（第十八、十九条）. （đăng trên Cổng Đảng viên）；Bộ Nhân sự và Xã hội, Bộ Môi trường và Tự nhiên (人力资源社会保障部、应急管理部) (2021). Biện pháp tuyển lính trong đội cứu hộ và cứu nạn tổng hợp quốc gia（国家综合性消防救援队伍消防员招录办法）（人社部发〔2021〕58 号，第五、六条）. ；Hội đồng Nhà nước, Ủy ban Quân sự Trung ương (国务院、中央军委) (2022). Quy chế nhân sự văn chứng Quân giải phóng nhân dân Trung Quốc（中国人民解放军文职人员条例）（sửa đổi lần hai, Lệnh số 757, 第九、十九条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Hội đồng Nhà nước (国务院) (2014). Quy chế quản lý nhân sự của cơ quan hành chính sự vệ（事业单位人事管理条例）（Hội đồng Nhà nước Lệnh số 652, 第八条）. ；Hội đồng Nhà nước (国务院) (2022). Quy chế thúc đẩy phát triển hộ kinh doanh cá thể（促进个体工商户发展条例）（Hội đồng Nhà nước Lệnh số 755, 第二条）. ；Hội đồng Nhà nước (国务院) (1988, sửa đổi 2014). Quy chế lâm thời về thi tự học bậc đại học（高等教育自学考试暂行条例）（第三、二十条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2018). Luật Giáo dục bắt buộc（义务教育法）（第十四条）. （đăng trên Bộ Giáo dục）；Văn phòng Hội đồng Nhà nước (国务院办公厅) (2024). Thông báo chuyển tiếp các biện pháp của Bộ Giáo dục và các bộ về chế độ giáo dục có học phí cho sinh viên sư phạm liên thông bậc đại học và sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục（国办发〔2024〕27 号，第二、五条）. ；Ủy ban Quốc gia Y tế và 7 bộ khác (国家卫生健康委等 7 部门) (2019). Thông báo về việc làm tốt công tác bố trí việc làm và quản lý thực hiện cam kết cho sinh viên y khoa được đào tạo miễn phí theo đơn đặt hàng nông thôn. ；Hội đồng Nhà nước (国务院) (2012). Quy chế quản lý hợp tác lao động ngoài biên giới（对外劳务合作管理条例）（Hội đồng Nhà nước Lệnh số 620, 第五、七、八条）. - Ghi chú: Đủ điều kiện rồi còn phải qua mấy chặng nữa là điều kiện chính trị, khám sức khoẻ và thẩm tra chính trị. Mấy chặng đó không có tiêu chuẩn chấm điểm công khai, tra văn bản cũng không có. Cửa vào tuyển sinh liên thông của trường trung cấp nghề, trường kỹ nghề và giáo dục nghề xem mục 5 ở phần 23; trợ cấp khi gia đình không gánh nổi học phí xem mục 4 ở phần 23. Hộ kinh doanh cá thể không xem trình độ, nhưng mở cửa hàng không dễ. Vốn, bảo lãnh cá nhân và chi phí rút lui khi đóng cửa xem phần 12. Bảo lãnh cá nhân nghĩa là công ty không trả được thì chính bạn phải trả. Tiền khởi đầu có vay được khoản vay mà Nhà nước góp một phần lãi suất không, xem mục 16 của phần này. Hai con đường sinh viên sư phạm có học phí và sinh viên y khoa định hướng đã được chốt vào ngày điền nguyện vọng kỳ thi gaokao, cái giá nằm ở mục 13 của phần này. Điều kiện của việc đi làm công ăn nước ngoài không nằm ở bạn mà ở chỗ công ty có đủ tư cách hay không, cách nhận biết xem mục 14 của phần này. Con đường nào đáng hơn thì phần này không so sánh. - ### 2. Năm thứ mười tám tuổi phải đăng ký nghĩa vụ quân sự trước ngày 31 tháng 10; quân nhân lính nghĩa vụ nhập ngũ đúng hai năm - Chi phí: Không tốn tiền. Đăng ký thì điền một lần trên mạng, vài phút là xong. Nếu thật sự đi nhập ngũ thì cái giá là hai năm toàn thời gian, trong hai năm đó không làm được việc khác song song. @@ -23,7 +22,6 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Mức bằng chứng: A - Nguồn: Văn phòng Hội đồng Nhà nước, Văn phòng Ủy ban Quân sự Trung ương (国务院办公厅、中央军委办公厅) (2025). Quy định về công tác đăng ký nghĩa vụ quân sự（兵役登记工作规定）（国办发〔2025〕41 号）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2021). Luật nghĩa vụ quân sự（兵役法）（sửa đổi năm 2021，第二十、二十五、二十六、五十四、五十七条）. （đăng trên Bộ Quốc phòng）；Hội đồng Nhà nước, Ủy ban Quân sự Trung ương (国务院、中央军委) (2023). Quy chế công tác truy xuất quân nhân（征兵工作条例）（Hội đồng Nhà nước, Ủy ban Quân sự Trung ương Lệnh số 759，第五条）. - Ghi chú: Đăng ký và nhập ngũ là hai chuyện, cái giá chênh nhau rất xa. Loạt hình phạt cấm xuất cảnh trong hai năm chỉ áp cho người sau khi truy xuất mà đổi ý, xem mục 3 của phần này (từ chối nghĩa vụ quân sự). Phụ nữ không bắt buộc đăng ký nghĩa vụ quân sự lần đầu, chỉ bị truy xuất tuỳ nhu cầu quân đội. Quân nhân thi đỗ trường quân, quân nhân có thành tích tốt được cấp quân sĩ (tiến lên quân sĩ), điều kiện đều nằm trong quy chế nội bộ của quân đội, ra ngoài tra không được. Luật nghĩa vụ quân sự chỉ viết hai câu: sĩ quan xuất phái chọn từ「表现优秀的现役士兵」, và việc tuyển học viên của trường quân không chịu hạn chế tuổi truy xuất. Cụ thể điểm chấm và số suất, sách này không nêu. - ### 3. Sau khi được truy xuất mà từ chối nghĩa vụ quân sự thì trong hai năm không được xuất cảnh, không được học tiếp, và không vào được cơ quan công chức hay doanh nghiệp nhà nước - Chi phí: Không tốn tiền. Cái giá phải trả trước khi đăng ký, vì phải nghĩ rõ có đi hay không. @@ -33,14 +31,14 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Nguồn: Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2021). Luật nghĩa vụ quân sự（兵役法）（sửa đổi năm 2021，第五十七、五十八条）. （đăng trên Bộ Quốc phòng） - Ghi chú: Ở đây tính là tự do cá nhân và điều kiện làm việc, không phải tiền. Loạt hình phạt này chỉ áp cho hai loại người. Một là「应征公民拒绝、逃避征集服现役的」. Hai là người nhập ngũ rồi bị xóa tên, bị khai trừ quân tịch, bị truy trách nhiệm hình sự. Người chỉ không làm đăng ký nghĩa vụ quân sự thì không nằm trong đó, đăng ký nghĩa vụ quân sự xem mục 2 của phần này. Điều kiện để phạt là「拒不改正」, do chính quyền huyện xác định và xử phạt. Phạt bao nhiêu thì các nơi tự quy định, sách này không viết. Có đi hay không thì bạn tự quyết, nhưng phải chốt trước khi đăng ký. -### 4. Hai ưu đãi rõ ràng nhất cho sinh viên đại học đi nhập ngũ: học phí được bù mỗi năm tối đa 20000 tệ, học vị giữ trong hai năm, và có chương trình riêng cho thi cao học +### 4. Hai điều thật nhất khi sinh viên đại học đi nghĩa vụ quân sự: học phí mỗi năm bù tối đa 20000 tệ, giữ học vị hai năm, và có một kế hoạch riêng cho tuyển sinh cao học -- Chi phí: Không tốn tiền. Sau khi giải ngũ thì nộp hồ sơ theo yêu cầu của trường, phải chạy một hai chuyến. -- Nói thẳng: Học sinh đang học, người sắp tốt nghiệp và tân sinh viên đi nhập ngũ thì trong thời gian phục vụ, học phí do Nhà nước chi trả. Sinh viên đại học hoặc cao đẳng mỗi người mỗi năm tối đa 20000 tệ, học sau đại học mỗi người mỗi năm tối đa 25000 tệ, hai năm thì nhiều nhất bốn vạn hoặc năm vạn. Học vị được giữ lại cho bạn, trong hai năm sau khi giải ngũ vẫn được nhập học hoặc học lại. Thi cao học lại có một chương trình riêng chỉ tuyển sinh viên đại học đã giải ngũ. -- Lợi ích: Thông báo năm 2024 của Bộ Tài chính và ba bộ khác nói sinh viên các trường đại học đi nghĩa vụ quân sự có ba ưu đãi: bù học phí, hoàn trả khoản vay học phí của Nhà nước, và giảm học phí. Hoàn trả nghĩa là Nhà nước trả thay khoản vay học phí đó. Mức của cả ba được「相应调整为本专科学生每人每年最高不超过 20000 元、研究生每人每年最高不超过 25000 元」, tức điều chỉnh thành mức sinh viên đại học hoặc cao đẳng mỗi người mỗi năm không quá 20000 tệ, học sau đại học mỗi người mỗi năm không quá 25000 tệ. Mức mới áp dụng từ học kỳ mùa thu năm 2024. Bù cụ thể bao nhiêu thì xem câu này:「按学生实际缴纳的学费或获得的国家助学贷款（包括本金及其全部偿还之前产生的利息）两者金额较高者执行」, tức tính theo số tiền lớn hơn trong hai khoản là học phí học sinh thật sự đã đóng hoặc khoản vay học phí của Nhà nước được cấp (kể cả gốc và toàn bộ lãi phát sinh trước khi trả hết). Có thêm câu「超出标准部分不予补偿、代偿或减免」, tức phần vượt quá mức không được bù, hoàn trả hay giảm. Phần học vị thì xem《Luật bảo đảm cho cựu chiến binh》. Nó viết:「现役军人入伍前已被普通高等学校录取或者是正在普通高等学校就学的学生，服现役期间保留入学资格或者学籍，退役后两年内允许入学或者复学，可以按照国家有关规定转入本校其他专业学习」, tức học sinh đã được trường đại học thường tuyển, hoặc đang học ở trường đại học thường, trước khi nhập ngũ thì trong thời gian phục vụ được giữ quyền nhập học hoặc học vị; sau khi giải ngũ, trong hai năm vẫn được nhập học hoặc học lại; và có thể chuyển sang học chuyên ngành khác của cùng trường theo quy định của Nhà nước. Luật đó còn viết:「高等学校根据国家统筹安排，可以通过单列计划、单独招生等方式招考退役军人」, tức trường đại học có thể dùng chỉ tiêu tách riêng, tuyển sinh riêng để tuyển cựu chiến binh theo kế hoạch Nhà nước sắp xếp. Phần thi cao học thì xem thông báo năm 2026 của Văn phòng Bộ Giáo dục. Chương trình tên là「退役大学生士兵计划」, tức chương trình dành cho sinh viên đại học đã giải ngũ. Thông báo viết:「在全国硕士研究生招生计划总规模内单列下达、专项使用，不得调整为普通计划或其他专项计划。招生单位自主确定……初试成绩要求（一般不高于本单位相应学科专业普通计划考生初试成绩要求）……同等条件下服役时间长的考生优先录取」, tức chỉ tiêu được tách riêng trong tổng quy mô tuyển sinh cao học toàn quốc và dùng riêng, không được chuyển thành chỉ tiêu bình thường hay chương trình riêng khác; cơ quan tuyển sinh tự quyết định yêu cầu về điểm kỳ thi đầu (về nguyên tắc không cao hơn yêu cầu điểm kỳ thi đầu của thí sinh chỉ tiêu bình thường cùng ngành chuyên ngành tương ứng); cùng điều kiện thì thí sinh phục vụ lâu hơn được tuyển trước. Tách riêng và dùng riêng nghĩa là phần chỉ tiêu đó được chia ra riêng, chỉ dùng để tuyển sinh viên đại học đã giải ngũ. Năm 2026 có 532 trường đại học thường đảm nhận chương trình này (toàn quốc) +- Chi phí: Không tốn tiền. Sau khi giải ngũ thì nộp hồ sơ theo yêu cầu của trường, chạy một hai chuyến là xong. +- Nói thẳng: Sinh viên đang học, sinh viên năm cuối và sinh viên mới nhập trường đi nghĩa vụ quân sự thì nhà nước trả hộ phần học phí cho những năm đó, bù theo số thực tế đã nộp, sinh viên bậc đại học và cao đẳng mỗi năm tối đa 20000 tệ, học viên sau đại học tối đa 25000 tệ. Học phí vốn đã thấp thì chỉ bù tới đúng số đã nộp. Học vị được giữ cho bạn hai năm. Phần tuyển sinh cao học lại có một kế hoạch riêng chỉ nhận cựu chiến binh là sinh viên đại học. +- Lợi ích: Thông báo năm 2024 của Bộ Tài chính cùng ba bộ khác nói rằng sinh viên trường đại học đi nghĩa vụ quân sự có ba loại đãi ngộ: bù hoàn học phí, bù trả khoản vay học tập quốc gia dùng để trả học phí, và giảm học phí. Bù trả khoản vay nghĩa là nhà nước trả thay khoản vay học tập đó. Mức của cả ba là「相应调整为本专科学生每人每年最高不超过 20000 元、研究生每人每年最高不超过 25000 元」, tức điều chỉnh thành học sinh đại học và cao đẳng mỗi người mỗi năm tối đa không quá 20000 tệ, học viên sau đại học mỗi người mỗi năm tối đa không quá 25000 tệ. Mức mới bắt đầu từ học kỳ mùa thu năm 2024. Cụ thể bù bao nhiêu thì xem câu này:「按学生实际缴纳的学费或获得的国家助学贷款（包括本金及其全部偿还之前产生的利息）两者金额较高者执行」, tức thực hiện theo số tiền lớn hơn trong hai khoản là học phí sinh viên thực sự đã nộp và khoản vay học tập quốc gia đã được cấp, gồm cả tiền gốc và toàn bộ lãi phát sinh trước khi trả hết. Câu sau nữa là「超出标准部分不予补偿、代偿或减免」, tức phần vượt mức thì không bù, không bù trả và không giảm. Vậy mỗi năm bạn nhận được là số nhỏ hơn giữa「khoản cao hơn trong học phí thực tế và khoản vay học tập」với mức trần. Học phí mỗi năm là 5000 tệ thì bù 5000 tệ. Bù mấy năm thì tính theo số năm học, không tính theo số năm phục vụ. Quy định năm 2019 về quản lý quỹ hỗ trợ học sinh viết:「入伍前已完成规定的修业年限，即为学费补偿或国家助学贷款代偿的年限；退役复学后接续完成规定的剩余修业年限，即为学费减免的年限」, tức số năm đã hoàn thành chương trình học theo quy định trước khi nhập ngũ là số năm được bù học phí hoặc bù trả khoản vay học tập quốc gia, còn số năm hoàn thành tiếp phần học còn lại theo quy định sau khi giải ngũ trở lại học là số năm được giảm học phí. Sinh viên năm cuối đi nghĩa vụ thì được bù học phí cho cả thời gian học. Về giữ học vị thì xem《Luật bảo đảm cho cựu chiến binh》. Luật viết:「现役军人入伍前已被普通高等学校录取或者是正在普通高等学校就学的学生，服现役期间保留入学资格或者学籍，退役后两年内允许入学或者复学，可以按照国家有关规定转入本校其他专业学习」, tức sinh viên đã được trường đại học thường môn tuyển, hoặc đang học ở trường đại học thường môn, trước khi nhập ngũ, trong thời gian phục vụ quân đội vẫn giữ quyền nhập học hoặc học vị, sau khi giải ngũ trong vòng hai năm được phép nhập học hoặc trở lại học, và có thể theo quy định của nhà nước chuyển sang học chuyên ngành khác của trường. Luật còn viết:「高等学校根据国家统筹安排，可以通过单列计划、单独招生等方式招考退役军人」, tức trường đại học có thể tuyển cựu chiến binh theo kế hoạch tổng hợp của nhà nước, bằng cách tách kế hoạch riêng, tuyển sinh riêng và các hình thức khác. Phần tuyển sinh cao học thì xem thông báo năm 2026 của Văn phòng Bộ Giáo dục. Kế hoạch này gọi là「退役大学生士兵计划」. Thông báo viết:「在全国硕士研究生招生计划总规模内单列下达、专项使用，不得调整为普通计划或其他专项计划。招生单位自主确定……初试成绩要求（一般不高于本单位相应学科专业普通计划考生初试成绩要求）……同等条件下服役时间长的考生优先录取」, tức được tách riêng và dùng chuyên biệt trong tổng quy mô kế hoạch tuyển sinh cao học toàn quốc, không được điều chỉnh thành kế hoạch thường hay kế hoạch chuyên biệt khác; cơ quan tuyển sinh tự quyết định yêu cầu điểm kỳ thi đầu, về nguyên tắc không cao hơn yêu cầu điểm kỳ thi đầu của thí sinh kế hoạch thường ứng với chuyên ngành của đơn vị đó; cùng điều kiện thì thí sinh đã phục vụ lâu hơn được ưu tiên nhận. Tách riêng và dùng chuyên biệt nghĩa là phần chỉ tiêu này được cắt riêng ra, chỉ dùng để nhận cựu chiến binh là sinh viên đại học. Năm 2026 kế hoạch này do 532 trường đại học thường môn thực hiện (toàn quốc) - Mức bằng chứng: A -- Nguồn: Bộ Tài chính, Bộ Giáo dục, Ngân hàng Nhân dân Trung Quốc, Cục Quản lý Tài chính (财政部、教育部、中国人民银行、金融监管总局) (2024). Thông báo về việc điều chỉnh hoàn thiện chính sách khoản vay học phí của Nhà nước（关于调整完善国家助学贷款有关政策的通知）（财教〔2024〕188 号）. ；Bộ Tài chính và bốn bộ khác (财政部等五部门) (2019). Quy chế quản lý quỹ hỗ trợ học sinh（学生资助资金管理办法）（财科教〔2019〕19 号，第五条）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2020). Luật bảo đảm cho cựu chiến binh（退役军人保障法）（第三十四、三十五条）. （đăng trên Bộ Quốc phòng）；Văn phòng Bộ Giáo dục (教育部办公厅) (2025). Thông báo về việc làm tốt công tác tuyển sinh theo chỉ tiêu chuyên biệt cho cao học năm 2026 dành cho sinh viên đại học đã giải ngũ（关于做好 2026 年退役大学生士兵专项硕士研究生招生计划招生工作的通知）（教学厅〔2025〕8 号）. ；Bộ Công tác cựu chiến binh (退役军人事务部) (2022). 23 điểm chính sách đào tạo cho quân nhân giải ngũ:「从2022年招生起，高职（专科）毕业生及在校生（含高校新生）应征入伍，退役后完成高职（专科）学业的，申请专升本，免于参加文化课考试」（theo 教学厅〔2021〕8 号）. ；Bộ Giáo dục (教育部) (2026). Quy định quản lý công tác tuyển sinh cao học toàn quốc năm 2027（2027 年全国硕士研究生招生工作管理规定）（教学〔2026〕2 号，第二十二、二十四、二十五条），đăng trên Cổng tuyển sinh cao học. ；Bộ Giáo dục (教育部) (2024). Quy định quản lý công tác tuyển sinh cao học toàn quốc năm 2025（2025 年全国硕士研究生招生工作管理规定）（教学〔2024〕4 号，第二十二条）. -- Ghi chú: Còn hai cửa nữa. Một là lên đại học cử nhân không phải thi môn văn hoá: sinh viên cao đẳng đi nhập ngũ, sau khi giải ngũ học xong cao đẳng thì có thể xin miễn thi môn văn hoá để lên đại học, trường kiểm tra kỹ năng nghề nghiệp rồi tuyển theo thứ tự. Cửa này bắt đầu từ tuyển sinh năm 2022, cách làm cụ thể xem thông báo năm đó của tỉnh. Hai là cộng 10 điểm ở kỳ thi đầu: sinh viên đại học đã giải ngũ mà trong vòng 3 năm sau khi giải ngũ đi thi cao học thì tổng điểm kỳ thi đầu cộng thêm 10 điểm. Đã đăng ký chương trình riêng thì không cộng điểm nữa, hai ưu đãi không cùng hưởng được. Người từng lập công chiến công từ bậc hai trở lên thì có thể xin miễn kỳ thi đầu để học thạc sĩ. Danh sách trường có chương trình riêng thay đổi mỗi năm, năm 2025 là 523 trường. Mức học phí đã điều chỉnh hai lần: năm 2019 là 8000/12000, mùa thu năm 2023 đổi thành 16000/20000, mùa thu năm 2024 đổi thành 20000/25000. Bản tóm tắt năm 2022 của Bộ Công tác cựu chiến binh vẫn ghi bản đầu tiên, đừng chép lại nguyên văn. +- Nguồn: Bộ Tài chính, Bộ Giáo dục, Ngân hàng Nhà nước Trung Quốc, Cục Quản lý Tài chính (财政部、教育部、中国人民银行、金融监管总局) (2024). Thông báo về điều chỉnh và hoàn thiện chính sách khoản vay học tập quốc gia（关于调整完善国家助学贷款有关政策的通知）（财教〔2024〕188 号）. ；Bộ Tài chính cùng bốn bộ khác (财政部等五部门) (2019). Quy định về quản lý quỹ hỗ trợ học sinh（学生资助资金管理办法）（财科教〔2019〕19 号，第五条；入伍资助年限：「入伍前已完成规定的修业年限，即为学费补偿或国家助学贷款代偿的年限；退役复学后接续完成规定的剩余修业年限，即为学费减免的年限」，毕业生「学费补偿或国家助学贷款代偿的年限，分别按照国家规定的相应学制计算」）. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2020). Luật bảo đảm cho cựu chiến binh（退役军人保障法）（第三十四、三十五条）. （đăng trên Bộ Quốc phòng）；Văn phòng Bộ Giáo dục (教育部办公厅) (2025). Thông báo về việc làm tốt công tác tuyển sinh kế hoạch chuyên biệt cao học dành cho cựu chiến binh là sinh viên đại học năm 2026（关于做好 2026 年退役大学生士兵专项硕士研究生招生计划招生工作的通知）（教学厅〔2025〕8 号）. ；Bộ Công tác cựu chiến binh (退役军人事务部) (2022). Chính sách đào tạo cho cựu chiến binh, 23 mục: 「从2022年招生起，高职（专科）毕业生及在校生（含高校新生）应征入伍，退役后完成高职（专科）学业的，申请专升本，免于参加文化课考试」（依据教学厅〔2021〕8 号）. ；Bộ Giáo dục (教育部) (2026). Quy định về công tác tuyển sinh cao học toàn quốc năm 2027（2027 年全国硕士研究生招生工作管理规定）（教学〔2026〕2 号，第二十二、二十四、二十五条）, đăng trên trang tuyển sinh cao học. ；Bộ Giáo dục (教育部) (2024). Quy định về công tác tuyển sinh cao học toàn quốc năm 2025（2025 年全国硕士研究生招生工作管理规定）（教学〔2024〕4 号，第二十二条）. +- Ghi chú: Còn hai đường nữa. Một là lên bập cao đẳng không phải thi văn hoá: sinh viên bập cao đẳng đi nghĩa vụ quân sự, sau khi giải ngũ học xong bập cao đẳng thì có thể xin được miễn thi đầu vào lên bập cao đẳng, nhà trường xem kỹ năng nghề rồi nhận những người đạt. Đường này áp dụng từ kỳ tuyển sinh năm 2022, cách làm cụ thể xem thông báo của tỉnh trong năm đó. Hai là cộng thêm 10 điểm vào kỳ thi đầu: cựu chiến binh là sinh viên đại học sau khi giải ngũ mà trong vòng 3 năm thi vào cao học thì được cộng thêm 10 điểm vào tổng điểm kỳ thi đầu. Đã đăng ký kế hoạch riêng thì không cộng điểm nữa, hai thứ không được cùng hưởng. Người từng lập công từ bậc hai trở lên thì có thể xin miễn kỳ thi đầu để học thạc sĩ. Danh sách trường mở kế hoạch riêng thay đổi mỗi năm, năm 2025 là 523 trường. Mức học phí đã điều chỉnh hai lần: năm 2019 là 8000/12000, mùa thu năm 2023 đổi thành 16000/20000, mùa thu năm 2024 đổi thành 20000/25000. Phần tóm tắt năm 2022 của Bộ Công tác cựu chiến binh vẫn còn bản đầu tiên, đừng chép theo. ### 5. Sau khi giải ngũ chỉ có sáu trường hợp được chính quyền bố trí việc làm; người tự tìm việc phải đến đăng ký trong vòng 30 ngày @@ -59,7 +57,6 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước, Ủy ban Quân sự Trung ương (国务院、中央军委) (2024). Quy chế sắp xếp cựu chiến binh（退役军人安置条例）（Hội đồng Nhà nước, Ủy ban Quân sự Trung ương Lệnh số 787，第二十一、二十二、七十四、八十一、八十五条）. ；Bộ Công tác cựu chiến binh (退役军人事务部) (2020). Mười câu hỏi về tiền trở cấp một lần (đăng lại từ《Cựu chiến binh Trung Quốc》số 10 năm 2020):「基本标准。士兵服现役每满一年，发给4500元」「退役金标准适时调整」. ；Sở Công tác cựu chiến binh tỉnh Quý Châu (贵州省退役军人事务厅) (2024). Tăng tiền trở cấp một lần, có thay đổi quan trọng!：「现在执行的标准是：士兵服现役每满一年，发给4500元」. ；Chính quyền quận Mộc Bình, thành phố Yên Đài (烟台市牟平区人民政府) (2025). Trích trong bản trả lời《Thông báo của Sở Công tác cựu chiến binh và Sở Tài chính thành phố Yên Đài về việc điều chỉnh mức trợ cấp kinh tế một lần cho quân nhân giải ngũ tự tìm việc》：「2024年退出现役的每服役一年补助6000元」. ；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2020). Luật bảo đảm cho cựu chiến binh（退役军人保障法）（第四十二、四十四、五十条）. （đăng trên Bộ Quốc phòng）；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2012). Luật bảo hiểm quân nhân（军人保险法）（第十六、二十三条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Bộ Tài chính, Cục Thuế, Bộ Công tác cựu chiến binh (财政部、税务总局、退役军人事务部) (2023). Thông báo về chính sách thuế hỗ trợ thêm cho quân nhân giải ngũ tự tìm việc khởi nghiệp và làm việc（关于进一步扶持自主就业退役士兵创业就业有关税收政策的公告）（2023 年第 14 号）. - Ghi chú: Quy chế chỉ viết rằng số tiền do cơ quan chủ quản quyết định riêng. Con số chính thức tra được là câu trong bài đăng năm 2020 trên trang Bộ Công tác cựu chiến binh「服现役每满一年，发给4500元」, cùng bài đó còn viết「适时调整」. Tỷ lệ tăng trong bài đó đã không khớp với quy chế đang có hiệu lực, phải theo quy chế. Bài đăng tháng 8 năm 2024 của Sở Công tác cựu chiến binh tỉnh Quý Châu, dưới quy chế mới, vẫn viết「现在执行的标准是：士兵服现役每满一年，发给4500元」. TODO（cần đối chiếu：không tìm được văn bản tiêu chuẩn chính thức, chỉ xác nhận được tới tháng 8 năm 2024 cơ quan cấp tỉnh vẫn ghi như vậy）. Ngoài ra đừng lẫn「trợ cấp kinh tế một lần」của địa phương với khoản này. Đó là một khoản tiền khác do các nơi tự quy định và tự chi trả, ví dụ Yên Đài năm 2024 trợ cấp mỗi năm phục vụ 6000 tệ. Số mới trong các hướng dẫn trên mạng phần lớn không có nguồn, có cái lại là gộp hai khoản đó vào nhau. Trong thời gian quân nhân lính nghĩa vụ phục vụ, gia đình còn có thể nhận một khoản ưu đãi gia đình riêng. Luật nghĩa vụ quân sự chỉ viết「标准由地方人民政府制定，中央财政给予定额补助」, cụ thể bao nhiêu thì hỏi chính quyền huyện nơi chấp thuận cho bạn nhập ngũ. Khoản thuế là mỗi năm giảm được tối đa bao nhiêu tệ thuế. Nếu việc kinh doanh không lên thì không dùng hết, nên cả mục này chỉ để mức lợi ích「vừa」. 《Quy chế thúc đẩy việc làm và khởi nghiệp cho cựu chiến binh》áp dụng từ ngày 1 tháng 8 năm 2026 bổ sung một điều: chứng chỉ nghề nghiệp của người lao động có tay nghề thi đỗ trong thời gian phục vụ thì sau khi giải ngũ có giá trị trên toàn quốc. - ### 7. Muốn vào hệ thống cơ quan công chức không chỉ có thi cao công chức: làm đủ hai năm ở chương trình phục vụ cơ sở thì dự được kế hoạch tuyển định hướng dành riêng cho 10% mỗi năm - Chi phí: Không tốn tiền. Cái giá là hai năm phục vụ ở cơ sở cấp xã. Hai năm đó bạn nhận trợ cấp ngân sách, không phải lương thị trường, chỗ khó là chịu được hai năm ấy. @@ -68,7 +65,6 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Mức bằng chứng: A - Nguồn: Ban Tổ chức Trung ương Đảng Cộng sản Trung Quốc, Bộ Nhân sự và Xã hội và mười bộ (中共中央组织部、人力资源社会保障部等十部门) (2021). Thông báo về việc thực hiện kế hoạch「三支一扶」đợt thứ tư cho sinh viên, học viên tốt nghiệp（关于实施第四轮高校毕业生「三支一扶」计划的通知）（人社部发〔2021〕32 号，第七、八、九、十一、十二、十三条）. ；Văn phòng Trung ương Đảng, Văn phòng Hội đồng Nhà nước (中共中央办公厅、国务院办公厅) (2017). Ý kiến về việc tiếp tục hướng dẫn và khuyến khích sinh viên, học viên tốt nghiệp đi làm việc ở cơ sở（关于进一步引导和鼓励高校毕业生到基层工作的意见）（第五、十六、十八条）. ；Văn phòng quản lý dự án Kế hoạch tình nguyện sinh viên đại học phía Tây (全国大学生志愿服务西部计划项目管理办公室) (2026). Thỏa thuận tuyển sinh Kế hoạch tình nguyện sinh viên đại học phía Tây năm học 2026—2027. - Ghi chú: Kế hoạch「三支一扶」đợt thứ tư áp dụng cho giai đoạn 2021 đến 2025. Lần này không tìm được văn bản của đợt thứ năm và thông báo năm 2026, trước khi đăng ký hãy tra thông báo tuyển sinh của năm đó. Năm 2025 ngân sách trung ương hỗ trợ tuyển 34430 người. Nguyên văn phương án thực hiện của Kế hoạch phía Tây không tra được trên mạng, trang chính thức mở mục chính sách không ra nội dung. Các ưu đãi sau khi đủ hạn thì theo văn bản của Văn phòng Trung ương Đảng và Văn phòng Hội đồng Nhà nước, vốn quản lý mọi chương trình phục vụ cơ sở. Hai năm đó chỉ nhận trợ cấp ngân sách chứ không phải lương thị trường, cái giá này phải tính cùng với kế hoạch tuyển định hướng sau hai năm. - ### 8. Giáo viên vị trí đặc biệt hết hạn ba năm, đạt kết quả đánh giá và còn muốn ở lại thì phải được chuyển thành biên chế đúng thời hạn - Chi phí: Không tốn tiền. Cái giá là ba năm dạy ở trường giáo dục bắt buộc nông thôn, trước khi vào làm còn phải thi lấy giấy chứng nhận giáo viên. Chỗ khó là ở được ba năm ấy. @@ -86,7 +82,6 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Mức bằng chứng: A - Nguồn: Bộ Nhân sự và Xã hội, Bộ Môi trường và Tự nhiên (人力资源社会保障部、应急管理部) (2021). Biện pháp tuyển lính trong đội cứu hộ và cứu nạn tổng hợp quốc gia（国家综合性消防救援队伍消防员招录办法）（人社部发〔2021〕58 号，第五、六、七、九条）. ；Hội đồng Nhà nước, Ủy ban Quân sự Trung ương (国务院、中央军委) (2022). Quy chế nhân sự văn chứng Quân giải phóng nhân dân Trung Quốc（中国人民解放军文职人员条例）（sửa đổi lần hai, Hội đồng Nhà nước, Ủy ban Quân sự Trung ương Lệnh số 757，áp dụng từ ngày 1 tháng 1 năm 2023；第九、十八、十九、二十条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia） - Ghi chú: 《Quy chế thúc đẩy việc làm và khởi nghiệp cho cựu chiến binh》áp dụng từ ngày 1 tháng 8 năm 2026 nêu thêm hai việc. Một là tuyển lính cứu hộ phải dành một tỷ lệ nhất định, chuyên tuyển quân nhân cấp bậc và quân nhân lính nghĩa vụ giải ngũ. Hai là khi tuyển lực lượng trợ giúp công an (cảnh sát phụ), cùng điều kiện thì ưu tiên tuyển cựu chiến binh. Số suất, vị trí và điều kiện cụ thể hằng năm của hai con đường này đều theo thông báo năm đó, sách này chỉ viết những ngưỡng cố định không đổi trong quy chế. Lính cứu hộ là vị trí trong đội kỷ luật, trước khi vào hãy nghĩ kỹ rủi ro nghề nghiệp. - ### 10. Bằng cấp không phải việc mua một lần là xong: thi tự học không giới hạn tuổi cũng không xem trình độ đã học, thi lớn và đại học mở đều bổ sung được - Chi phí: Phí đăng ký và giáo trình mỗi năm mấy trăm tệ. Cái giá là học tự thọn lâu dài, hoặc đi học buổi tối; bậc đại học học song song với việc làm (không rời việc) tối thiểu phải học hai năm rưỡi. Chỗ khó là phải kiên trì lâu dài. @@ -95,7 +90,6 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (1988, sửa đổi 2014). Quy chế lâm thời về thi tự học bậc đại học（高等教育自学考试暂行条例）（第三、二十、二十五、三十二条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Ủy ban Thường vụ Đại hội Đại biểu Quốc dân (全国人大常委会) (2018). Luật Giáo dục bậc cao（高等教育法）（第二十一条）. （Cơ sở dữ liệu văn bản pháp luật quốc gia）；Văn phòng Bộ Giáo dục (教育部办公厅) (2025). Thông báo về việc làm tốt công tác tuyển sinh các trường đại học dành cho người lớn toàn quốc năm 2025（关于做好 2025 年全国成人高校招生工作的通知）（教学厅〔2025〕6 号）kèm phụ lục《Cách thức tuyển sinh các trường đại học dành cho người lớn toàn quốc năm 2025》. ；Bộ Giáo dục (教育部) (2016). Ý kiến về việc làm tốt đại học mở（关于办好开放大学的意见）（教职成〔2016〕2 号）. ；Bộ Giáo dục (教育部) (2007). Thông báo về việc làm tốt hơn nữa việc quản lý các hình thức tuyển sinh của trường đại học（关于进一步做好高等学校各类招生管理工作的通知）（教发〔2007〕13 号）：「普通专升本招生计划按不超过当年应届普通高职（专科）毕业生5%的比例安排」. ；Cổng Chính phủ Trung Quốc (中国政府网) (2020). Năm nay mở rộng cao học 18.9 vạn, đại học cử nhân 32.2 vạn, hỗ trợ việc làm cho sinh viên tốt nghiệp, Phó Bộ trưởng Bộ Giáo dục Ô Thiết Hoa：「扩大普通高校专升本的规模，同比争取增加32.2万人」. ；Sở Khảo thử tỉnh Phúc Kiến (福建省教育考试院) (2026). Khi chuyển tiếp có dẫn「教育部高校学生司《关于做好2026年普通高等学校专升本考试招生工作的通知》（教学司函〔2025〕53号）」. - Ghi chú: Còn một hình thức đại học cử nhân dành cho sinh viên tốt nghiệp cao đẳng toàn thời gian. Văn bản công khai ở cấp quốc gia là thông báo năm 2007 của Bộ Giáo dục, trong đó ghi kế hoạch đại học cử nhân không quá 5% số sinh viên tốt nghiệp cao đẳng của năm đó. Từ năm 2020 Bộ Giáo dục mở rộng quy mô đại học cử nhân, năm đó tăng thêm 32.2 vạn người so với năm trước, mức 5% thực tế không còn được áp dụng. Công tác cụ thể hằng năm do Cục Quản lý Sinh viên trường đại học của Bộ Giáo dục phát công văn, năm 2026 là 教学司函〔2025〕53 号. Công văn đó không công bố ra ngoài, chỉ thấy được số văn bản trong các văn bản mà tỉnh chuyển tiếp. Đối tượng tuyển do từng tỉnh tự quy định, về cơ bản chỉ nhận tốt nghiệp cao đẳng toàn thời gian của chính tỉnh trong năm đó, nên phần nội dung không viết điều kiện của nó. Từ mùa thu năm 2025, giáo dục tiếp tục bậc đại học không còn gọi là「函授」「业余」mà thống nhất gọi là「非脱产」. Giáo dục tiếp tục bậc đại học chính là loại học bổ sung trình độ sau này, còn dùng tên cũ trong quảng cáo thì phải cảnh giác. Bổ sung trình độ và「học kỹ năng gì thì đáng tiền」là hai việc khác nhau, về sau xem phần 23. Chứng chỉ bán「lấy nhanh」và chứng chỉ chếch mạch xem mục 8 ở phần 23. - ### 11. Không vào cơ quan công chức thì làm việc linh hoạt: hưu trí và y tế phải tự tham gia bảo hiểm ở nơi làm việc, hạn chế về hộ khẩu đã được nới - Chi phí: Hưu trí và y tế bạn phải đóng toàn bộ, tính theo mức đóng của địa phương, mỗi tháng hơn nghìn tệ. Mức đóng là con số dùng làm căn cứ tính phí bảo hiểm. Đây là khoản chi cố định dễ bị tính sót nhất khi làm việc linh hoạt, chỗ khó là đều đặn đóng hằng tháng. @@ -104,7 +98,6 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Mức bằng chứng: A - Nguồn: Bộ Nhân sự và Xã hội và tám bộ (人力资源社会保障部等八部门) (2021). Ý kiến chỉ đạo về việc bảo vệ quyền lợi lao động của người lao động có hình thức việc làm mới（关于维护新就业形态劳动者劳动保障权益的指导意见）（人社部发〔2021〕56 号，第二、八条）. ；Văn phòng Hội đồng Nhà nước (国务院办公厅) (2020). Ý kiến về việc hỗ trợ việc làm linh hoạt theo nhiều kênh（关于支持多渠道灵活就业的意见）（国办发〔2020〕27 号）. - Ghi chú: Nguyên văn là「放开」hạn chế về hộ khẩu và vẫn để lỗ thông cho các thành phố rất lớn, không phải「toàn quốc đã bỏ hẳn」. Đến mức nào thì hỏi thành phố mà bạn sẽ tham gia bảo hiểm. Mức lợi ích để「lớn」vì đóng gián đoạn thì mất chính các chế độ hưu trí và y tế, giá trị vượt hẳn mấy vạn tệ. Nhưng chi phí cũng rõ ràng, tự đóng toàn bộ mỗi tháng hơn nghìn tệ. Sau khi đóng gián đoạn bảo hiểm xã hội thì khắc phục thế nào, số năm cộng dồn ra sao, xem mục 18 ở phần 7 (đóng gián đoạn bảo hiểm xã hội). Ngoài ra, bán nông sản, hàng tiêu dùng ở địa điểm và khung giờ do chính quyền quy định, hoặc dùng tay nghề của mình làm việc lợi ích không cần giấy phép, thì không phải đăng ký giấy phép kinh doanh. Điều này nằm trong 国办发〔2020〕27 号. - ### 12. Giao đồ ăn, chạy xe ô tô công nghệ, chở hàng trong thành phố: sàn trả phí bảo hiểm tai nạn lao động theo từng đơn, bạn không phải đóng - Chi phí: Không tốn tiền, khoản tiền đó sàn đóng. Việc bạn phải làm là xem sàn mình nhận việc có nằm trong phạm vi thí điểm không. Có hai cách: tìm「职业伤害保障」trong trang bảo vệ hoặc quyền lợi của ứng dụng, hoặc gọi điện cho cơ quan bảo hiểm xã hội của tỉnh hỏi xem sàn đó đã đăng ký chưa. @@ -122,7 +115,6 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Mức bằng chứng: A - Nguồn: Văn phòng Hội đồng Nhà nước (国务院办公厅) (2024). Thông báo chuyển tiếp các biện pháp của Bộ Giáo dục và các bộ về chế độ giáo dục có học phí cho sinh viên sư phạm liên thông bậc đại học và sau đại học của các trường đại học sư phạm trực thuộc Bộ Giáo dục（国办发〔2024〕27 号，第二、三、六、九、十三、十四、二十一条）. ；Ủy ban Quốc gia Y tế, Văn phòng Tổ chức cán bộ trung ương, Ủy ban Phát triển và Cải cách quốc gia, Bộ Giáo dục, Bộ Tài chính, Bộ Nhân sự và Xã hội, Cục Y học cổ truyền quốc gia (国家卫生健康委、中央编办、国家发展改革委、教育部、财政部、人力资源社会保障部、国家中医药局) (2019). Thông báo về việc làm tốt công tác bố trí việc làm và quản lý thực hiện cam kết cho sinh viên y khoa được đào tạo miễn phí theo đơn đặt hàng nông thôn. - Ghi chú: Bản năm 2024 đã đổi sinh viên sư phạm có học phí thành hình thức「liên thông», tức học xong bậc đại học thì học tiếp thạc sĩ giáo dục. Biện pháp cũ năm 2018 đã bãi bỏ, cái「tốt nghiệp đại học là dạy thẳng」còn truyền trên mạng là cách nói cũ. Người đã ký thỏa thuận trước khi bản này ban hành thì theo thỏa thuận cũ.「Có biên chế」không có nghĩa là「vào là xong」: biên chế phải bố trí được trong tổng số hiện có, vị trí cụ thể do thành phố tổ chức trường tuyển dụng riêng, bạn và trường cùng chọn nhau. Sinh viên y khoa định hướng hết hạn mà muốn đi thì được, văn bản ghi「服务期满，对自主择业的应予同意」. Người đã vi phạm cũng có cách xoá hồ sơ vi phạm, cái giá là theo thỏa thuận cũ, hoặc đổi sang cơ sở khác trong tỉnh và làm đủ ít nhất 6 năm nữa, phí và tiền phạt đã nộp không hoàn lại. Sinh viên y khoa còn phải tính thêm thời gian đào tạo quy chuẩn 3 năm, nên toàn chu kỳ dài hơn ngành sư phạm. Mỗi tỉnh còn có kế hoạch riêng cho sinh viên sư phạm có học phí và sinh viên y khoa định hướng, điều kiện do tỉnh quy định, trước khi điền nguyện vọng hãy xem tài liệu tuyển sinh năm đó của tỉnh mình. - ### 14. Muốn đi làm công ăn nước ngoài, trước hết tra công ty này có giấy phép hợp tác lao động ngoài biên giới không: công ty không được đòi bạn đặt tiền - Chi phí: Không tốn tiền. Trước khi xuất hành dành mười mấy phút để kiểm tra tư cách của công ty, rồi đọc kỹ cả hai bản hợp đồng. @@ -131,7 +123,6 @@ Chọn đường nào là việc của bạn. Phần này chỉ cung cấp hai t - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2012). Quy chế quản lý hợp tác lao động ngoài biên giới（对外劳务合作管理条例）（Hội đồng Nhà nước Lệnh số 620，第五、七、八、九、十、十一、十二、十三、二十一、二十三、二十五、二十六、二十九、三十一、三十七、三十九、四十、四十三条）. - Ghi chú: Trình tự làm đúng phải thế này: công ty trước hết ký hợp đồng hợp tác lao động bằng văn bản với nhà tuyển dụng nước ngoài, rồi ký hợp đồng dịch vụ bằng văn bản hoặc hợp đồng lao động với bạn. Hợp đồng phải ghi rõ nội dung công việc, thời hạn, mức thù lao, thời giờ làm việc, ngày nghỉ, bảo hiểm xã hội và rủi ro an toàn cá nhân. Sau khi ký, trong 10 ngày làm việc công ty phải báo cơ quan quản lý thương mại để đăng ký. Công ty còn bắt buộc phải sắp xếp cho bạn học kỹ năng, phòng chống xâm hại, ngoại ngữ và pháp luật tập quán địa phương, đồng thời mua cho bạn bảo hiểm tai nạn cá nhân trong thời gian làm việc ở nước ngoài. Bên kia không cho bạn xem hợp đồng, đòi tiền đặt, hay thúc bạn đi trước rồi bổ sung thủ tục sau thì chính là loại công ty phải tránh. Bảng lừa đảo tuyển dụng lương cao ở nước ngoài và khu công nghiệp lừa đảo xem mục 5 ở phần 21; nhắc nhở an toàn cần tra trước khi xuất hành và mã 12308 nằm ở phần 21, không lặp lại. Quy chế không đặt ngưỡng trình độ hay tuổi cho người đi làm công ăn, ngưỡng nằm ở tư cách của công ty và ở việc nước đi có cấp phép hay không. Công ty chỉ được ký hợp đồng với doanh nghiệp hoặc tổ chức được nước đó cho phép, không được ký với cá nhân ở nước ngoài. Khoản dự phòng do cơ quan chủ quản dùng vào khi công ty không trả nợ, bạn không tự rút được. Nên khi thật sự có sự cố vẫn phải khiếu nại với cơ quan quản lý thương mại trước, Điều hai mươi trao cho bạn quyền khiếu nại đó. Dự án cử trên 100 người sang cùng một nước hoặc vùng thì phải cử quản lý đi kèm và báo đại sứ quán, lãnh sự quán để đăng ký. Dự án ít người không có lớp bảo đảm đó, càng phải tự giữ hợp đồng và chứng tử chuyển khoản cho kỹ. - ### 15. Làm việc từ xa cho công ty nước ngoài ngay tại nhà, khoản tiền này theo luật thuế tính là thu nhập trong nước: thuế thu nhập cá nhân tự kê khai, tiền về nước đi theo khoản「thù lao lao động」 - Chi phí: Không tốn tiền. Mỗi năm quyết toán thuế một lần, thời gian là từ ngày 1 tháng 3 đến ngày 30 tháng 6 năm sau. Khi vượt tổng hạn mức cả năm thì còn phải chuẩn bị hợp đồng lao động và giấy chứng minh thu nhập. Chỗ khó là khoản thuế này bạn phải tự nhớ đi kê khai. diff --git "a/vi/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" "b/vi/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" index fd80b982..ae462b12 100644 --- "a/vi/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" +++ "b/vi/book/32-\345\207\272\345\233\275\347\225\231\345\255\246.md" @@ -16,7 +16,6 @@ Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra s - Mức bằng chứng: A - Nguồn: Trung tâm Dịch vụ Học bổng của Bộ Giáo dục Trung Quốc（中国（教育部）留学服务中心）. Tra cứu cơ sở được công nhận（认证院校查询）. ；Mạng lưới Giám sát và Quản lý Giáo dục có yếu tố nước ngoài của Bộ Giáo dục（教育部教育涉外监管信息网）. - Ghi chú: Cái mà đại gia nói miệng là 「Bộ Giáo dục công nhận」 không tính, hãy theo kết quả bạn tự tra trên mạng. Danh sách có thay đổi: lúc nhập học có mặt, lúc tốt nghiệp thì có khi không, nên trong thời gian học cứ tra lại mỗi năm. Chương trình giáo dục hợp tác Trung Quốc – nước ngoài có một hệ thống công nhận và đăng ký riêng. Người nhập học sau năm 2008 có thể dùng họ tên và số chứng minh nhân dân để tra số đăng ký của mình trên mạng giám sát giáo dục có yếu tố nước ngoài. - ### 2. Theo dõi vụ kiện về quy định F-1 mới ở Hoa Kỳ: 「tối đa bốn năm, học xong thì rời đi trong 30 ngày」 dự kiến có hiệu lực ngày 15 tháng 9 năm 2026, bị tòa án đình chỉ một ngày trước đó, hiện vẫn là 「cho tới khi học xong」 - Chi phí: Không tốn tiền. Chép ngày kết thúc chương trình trên I-20 vào lịch, đặt nhắc trước 90 ngày. I-20 là giấy chứng nhận đủ điều kiện du học do trường cấp cho bạn. Ngoài ra, mỗi học kỳ xem một lần thông báo của văn phòng sinh viên quốc tế của trường, để biết vụ kiện này có tiến triển gì mới không @@ -25,7 +24,6 @@ Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra s - Mức bằng chứng: A - Nguồn: DHS (2026). Establishing a Fixed Time Period of Admission and an Extension of Stay Procedure for Nonimmigrant Academic Students, Exchange Visitors（công bố 2026-07-17, dự kiến có hiệu lực 2026-09-15）. ；Presidents' Alliance on Higher Education and Immigration v. U.S. Department of Homeland Security, No. 1:26-cv-13799-FDS (D. Mass. Sept. 14, 2026), hồ sơ vụ án. ；Yale Office of International Students & Scholars (2026-09-14). Important Update: Court Action on the D/S Rule. ；AILA (2026). One Day Before Taking Effect, Federal Court Postpones the F, J, and I Fixed Admission Period Rule. ；8 CFR 214.2(f). - Ghi chú: Mục này có thời điểm chốt là ngày 25 tháng 9 năm 2026. Chính quyền có thể kháng nghị lên Tòa án Phúc án Tiểu bang thứ nhất, và Bộ An ninh nội địa cũng có thể sửa quy định rồi ban hành lại. Người phải học mấy năm thì đừng mặc định quy định cũ sẽ giữ nguyên mãi. Tòa đã ấn định ngày 2 tháng 10 năm 2026 mở phiên tiếp để nghe hai bên báo cáo tiến triển, sau ngày đó xem lại thông báo của trường. Nếu quy định mới sau này có hiệu lực, muốn gia hạn thì phải xin riêng trước khi thời hạn hết, không tự động nối tiếp. Dưới quy định cũ, người bỏ học sớm có thời hạn rời đi khác với người học đủ khóa, hãy hỏi văn phòng sinh viên quốc tế của trường trước. Thị thực J nằm trong cùng quy định đó, chi tiết khác với F-1, hãy tra theo loại thị thực của bạn. Cơ sở pháp lý cho việc hoãn nằm ở mục 705 của Đạo luật Thủ tục Hành chính Hoa Kỳ, cho phép tòa tạm dừng hiệu lực của một quy định. - ### 3. Số giờ làm thêm là vạch đỏ viết cứng ở cả bốn nước Mỹ, Canada, Anh và Úc, vượt một giờ là vi phạm điều kiện thị thực - Chi phí: Không tốn tiền. Tự ghi lại số giờ làm theo từng tuần. Chỗ khó là chủ lao động sẽ nói 「làm thêm vài giờ chẳng ai kiểm tra」, câu đó không đáng tin @@ -34,7 +32,6 @@ Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra s - Mức bằng chứng: A - Nguồn: 8 CFR 214.2(f)(9). ；UK Home Office. Immigration Rules Appendix Student（ST26.1、ST26.5）. ；IRCC. Work off campus as an international student. ；Australian Government Department of Home Affairs. Student visa (subclass 500). - Ghi chú: Việc làm ngoài trường ở Hoa Kỳ phải được duyệt trước. Khi hồ sơ xin giấy phép làm I-765 còn đang chờ xét, bạn không được bắt đầu làm. Giấy phép cũ ở Canada in số 20 giờ mỗi tuần, miễn là bạn vẫn đủ điều kiện thì thực tế làm được 24 giờ, theo quy định hiện hành. Quy định cấm tự kinh doanh ở Anh rất rộng: nhận việc riêng, mở cửa hàng trực tuyến, làm tự do đều tính vào. - ### 4. Gốc của tư cách là 「đang học toàn thời gian」: thời gian nghỉ học, bỏ học, khoảng trống khi chuyển trường đều không được làm thêm - Chi phí: Không tốn tiền. Trước khi ngừng lớp, xin nghỉ học hoặc đổi trường, hãy tìm văn phòng sinh viên quốc tế của trường và hỏi rõ tình trạng tư cách của bạn @@ -43,7 +40,6 @@ Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra s - Mức bằng chứng: A - Nguồn: IRCC. Work off campus as an international student. ；U.S. Department of Homeland Security, Study in the States. Working in the United States. ；UK Home Office. Immigration Rules Appendix Student（ST26.1）. - Ghi chú: Khi khối lượng học tập nặng và bạn muốn bớt môn, hãy hỏi cho rõ trước: bớt xuống bao nhiêu thì không còn tính là toàn thời gian. Chuyện này hỏng trước chuyện làm thêm. Bị trường đuổi học, hoặc chương trình bị chấm dứt, thì tư cách thường đứt ngay trong ngày. Đừng chờ đến khi trường gửi thư chính thức rồi mới hành động. - ### 5. Hoa Kỳ: sau khi chuyển nhà, báo địa chỉ mới trong vòng 10 ngày - Chi phí: Không tốn tiền. Nộp đơn đổi địa chỉ trực tuyến một lần, mất mười phút @@ -52,7 +48,6 @@ Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra s - Mức bằng chứng: A - Nguồn: 8 CFR 265.1. ；USCIS. AR-11, Alien's Change of Address Card. - Ghi chú: Báo với cơ quan nhập cảnh rồi, hồ sơ SEVIS ở trường (hồ sơ tư cách sinh viên quốc tế của bạn) vẫn còn phải cập nhật riêng. Ba nước còn lại cũng đều yêu cầu đổi địa chỉ thì báo cơ quan nhập cảnh hoặc trường. Hạn chót và cách làm khác nhau theo từng nước, làm theo trang web của cơ quan nhập cảnh nước bạn đang ở. - ### 6. Trước khi lên đường và trong thời gian học, xem qua cảnh báo du học của Bộ Giáo dục một lần - Chi phí: Không tốn tiền. Xem một lần trước khi lên đường, sau đó mỗi học kỳ xem một lần @@ -61,6 +56,7 @@ Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra s - Mức bằng chứng: A - Nguồn: Mạng lưới Giám sát và Quản lý Giáo dục có yếu tố nước ngoài của Bộ Giáo dục（教育部教育涉外监管信息网）. Thông tin cảnh báo（预警信息）. ；Bộ Giáo dục phát cảnh báo du học số 4 năm 2025. ；Số 1. - Ghi chú: Cảnh báo không phải lệnh cấm, cũng không phải bảo bạn không được đi, nó cho bạn một căn cứ chính thức để cân đối rủi ro nơi đó. Danh sách cảnh báo đổi theo tình hình, phần này không cập nhật dài hạn, cứ theo trang hiện hành trên website. An toàn cá nhân ở nước ngoài và bảo vệ lãnh sự, xem phần 21. + ### 7. Australia: OSHC phải phủ trọn suốt và không được đứt quãng, xuất cảnh mà không lấy ra được bằng chứng bảo hiểm thì có thể bị từ chối nhập cảnh - Chi phí: Mấy nghìn đến hơn mười nghìn tệ. Mua một lần theo thời hạn visa, học mấy năm thì mua mấy năm @@ -69,7 +65,6 @@ Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra s - Mức bằng chứng: A - Nguồn: Australian Government Department of Home Affairs. Student visa (subclass 500). - Ghi chú: Nếu trường làm thủ thì hãy hỏi rõ tên công ty bảo hiểm, ngày bắt đầu và kết thúc hợp đồng, số hợp đồng, những thứ này phải điền trong hồ sơ xin visa. Mua ngắn hơn thì lúc gia hạn sẽ vướng, nên mua dài đến sau khi tốt nghiệp thêm một chút. - ### 8. Anh: ngoài phí visa còn một khoản phụ thu y tế tính theo năm, sinh viên mỗi năm 776 bảng Anh, nộp một lần lúc xin - Chi phí: Phí visa 558 bảng Anh. Phụ thu y tế 776 bảng Anh mỗi năm. Cả hai khoản đều nộp một lần lúc xin @@ -78,7 +73,6 @@ Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra s - Mức bằng chứng: A - Nguồn: UK Government. Student visa. ；UK Government. Pay for UK healthcare as part of your immigration application. - Ghi chú: Số tiền và tỷ giá đều có thể đổi, cứ tính theo con số của kỳ hiện tại trên hai trang đó. Khóa học lấy bằng, từ 18 tuổi trở lên thì về nguyên tắc ở lâu nhất 5 năm, dưới bậc đại học là 2 năm. Ước theo số năm đó xem phải nộp phụ thu bao nhiêu. - ### 9. Trước khi về nước, chừa thời gian cho việc công nhận: công nhận của Trung tâm Dịch vụ Học bổng mất 10 đến 20 ngày làm việc - Chi phí: Nộp hồ sơ trên mạng, nộp tiền trên mạng. Chuẩn bị hồ sơ cộng thời gian xét duyệt, tính theo ngày làm việc, tổng cộng khoảng một tháng @@ -87,7 +81,6 @@ Khi đang ở nước ngoài mà có sự cố thì tìm đại sứ đoàn ra s - Mức bằng chứng: A - Nguồn: Trung tâm Dịch vụ Học bổng của Trung Quốc (Bộ Giáo dục) (中国（教育部）留学服务中心). Sảnh giao dịch trực tuyến, công nhận trình độ học vụ. - Ghi chú: Hồ sơ phải nộp gồm: bằng tốt nghiệp, hộ chiếu hoặc giấy thông hành, thẻ cư trú hoặc visa và dấu nhập cảnh, ảnh giấy tờ tùy thân, bản ủy quyền. Dữ liệu xuất nhập cảnh trong thời gian ở nước ngoài do hệ thống tự tra, nên đừng làm rách trang visa ở nước ngoài. Trên trang của Trung tâm Dịch vụ Học bổng còn có thông báo, nói giấy công nhận nào bị vô hiệu, trường nào tạm ngừng nhận hồ sơ công nhận. Trước khi làm thì xem đã có trường nào liên quan đến mình không. - ### 10. Các trường bị「siết lại việc công nhận」mỗi năm lại thêm, con đường mua bằng đã đi không còn - Chi phí: Không tốn tiền. Trước khi đăng ký, tra tên trường một lần trong thông báo của Trung tâm Dịch vụ Học bổng diff --git "a/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" "b/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" index 13fef8cc..7a17b4bb 100644 --- "a/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" +++ "b/vi/book/33-\346\256\213\347\226\276\344\271\213\345\220\216\346\200\216\344\271\210\346\264\273.md" @@ -2,15 +2,15 @@ # 33. Sống sau khi bị khuyết tật -Phần này viết về những ngày sau khi đã bị khuyết tật. Cách phòng ngừa gây khuyết tật nằm ở phần 1 và phần 13, ở đây không lặp lại. +Phần này nói về những ngày sau khi khuyết tật đã xảy ra. Phòng thế nào để không bị khuyết tật thì nằm ở phần 1 và phần 13, ở đây không lặp lại. -Bốn kiểu tính riêng nhau, không quy đổi cho nhau. Các mục về cấp chứng và biến chứng tính theo tỷ lệ tử vong. Các mục về trợ cấp, quyền lợi và kiện tụng tính theo tiền. Các mục về đi học và làm thủ tục tính theo thời gian. Các mục về nhập viện điều trị rối loạn tâm thần, năng lực hành vi và giám hộ tính theo tự do cá nhân. +Bốn thước đo tính riêng, không quy đổi qua lại. Mấy mục về cấp cứu và biến chứng thì tính theo tỷ lệ tử vong. Mấy mục về trợ cấp, quyền lợi và kiện tụng thì tính theo tiền. Mấy mục về đi học và làm thủ tục thì tính theo thời gian. Mấy mục về nằm viện vì rối loạn tâm thần, năng lực hành vi và giám hộ thì tính theo tự do cá nhân. -Tờ hướng dẫn của Tổ chức Y tế Thế giới đưa ra ba con số. Trên toàn cầu có khoảng 1,3 tỷ người bị suy giảm chức năng đáng kể, chiếm 16% dân số thế giới, trung bình cứ sáu người thì có một. Một số người khuyết tật chết sớm hơn người không khuyết tật tới 20 năm. Nguy cơ người khuyết tật mắc trầm cảm, hen, đái tháo đường, đột quỵ, béo phì hoặc sức khỏe miệng kém gấp hai lần. Trong những chênh lệch này, phần lớn không do bản thân khuyết tật gây ra, mà do không được khám bệnh, không được đến trường, không tìm được việc, không ra khỏi cửa. Nên các mục trong phần này một nửa là thân thể, một nửa là thủ tục và quyền lợi. +Tờ hướng dẫn của Tổ chức Y tế Thế giới đưa ra ba con số. Toàn cầu có khoảng 1,3 tỷ người có rối loạn chức năng đáng kể, chiếm 16% dân số thế giới, tức cứ sáu người thì có một. Có những người khuyết tật chết sớm hơn người không khuyết tật tối đa 20 năm. Người khuyết tật có nguy cơ bị trầm cảm, hen suỵn, đái tháo đường, đột quỵ, béo phì hoặc răng miệng kém gấp hai lần. Trong những khoảng cách này, một phần lớn không phải do bản thân khuyết tật gây ra, mà do không khám được bệnh, không học được, không xin được việc, không ra khỏi nhà. Nên các mục trong phần này, một nửa là về cơ thể, một nửa là về thủ tục và quyền. -Năm việc đã viết ở chỗ khác, ở đây chỉ dẫn đường, không lặp lại. Cách làm giấy chứng nhận người khuyết tật, bảy nhóm và cách chấm từ cấp 1 đến cấp 4, xem mục 11 ở phần 24 (giấy chứng nhận người khuyết tật). Thời điểm nào được làm giám định thương tật, xem mục 10 ở phần 24 (giám định thương tật). Cách nhận hai khoản trợ cấp cho người khuyết tật, xem mục 8 ở phần 7 (hai khoản trợ cấp). Khảo sát năng lực lao động của tai nạn lao động và thương tật đổi ra bao nhiêu tiền, xem mục 15 ở phần 19 (khảo sát năng lực lao động). Cách nộp đơn xin bảo hiểm chăm sóc dài hạn cho người mất năng lực nặng, xem mục 7 ở phần 17 (bảo hiểm chăm sóc dài hạn). Mục đó nằm ở phần 17 (Khi trong nhà có người già), nhưng khoản quyền lợi này không có giới hạn độ tuổi, người trẻ vẫn nhận được. +Năm việc đã viết ở nơi khác rồi, ở đây chỉ chỉ đường, không lặp lại. Giấy chứng nhận người khuyết tật làm thế nào, bảy loại và bốn bậc từ một đến bốn đánh giá ra sao, xem mục 10 ở phần 24 (giấy chứng nhận người khuyết tật). Giám định mức tàn tật phải đợi đến lúc nào mới làm, xem mục 9 ở phần 24 (giám định mức tàn tật). Hai khoản trợ cấp cho người khuyết tật nhận thế nào, xem mục 8 ở phần 7 (hai khoản trợ cấp). Giám định năng lực lao động cho tai nạn lao động và quy đổi mức tàn tật thành bao nhiêu tiền, xem mục 14 ở phần 19 (giám định năng lực lao động). Bảo hiểm chăm sóc dài hạn cho người mất khả năng tự chăm sóc nặng xin thế nào, xem mục 7 ở phần 17 (bảo hiểm chăm sóc dài hạn). Mục đó viết ở phần「có người cao tuổi trong nhà」, nhưng khoản trợ cấp này không có ngưỡng tuổi, người trẻ vẫn được nhận. -Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia đình bạn, thuộc hai mức cao nhất. Chỗ nào liên quan đến người khác thì ghi riêng trong Ghi chú. +Phần lớn các mục trong phần này, người hưởng lợi là bạn và gia đình, thuộc hai bậc cao nhất. Chỗ nào liên quan tới người khác thì viết riêng trong phần Ghi chú. ### 1. Sau chấn thương tủy sống mà đột nhiên đau đầu dữ dội và ra mồ hôi: đỡ ngồi ngay lên, nới lỏng quần áo, đồng thời gọi 120 @@ -20,7 +20,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: B - Nguồn: Krassioukov A, Warburton DE, Teasell R, Eng JJ; Spinal Cord Injury Rehabilitation Evidence Research Team (2009). A systematic review of the management of autonomic dysreflexia after spinal cord injury. Archives of Physical Medicine and Rehabilitation, 90(4):682-695. - Ghi chú: Chọn B là vì ba động tác tại chỗ đó không có thử nghiệm đối chứng ngẫu nhiên phía sau, chỉ có đồng thuận và suy luận sinh lý. Bản thân bài tổng kết cũng nói, phần lớn các cách này chỉ có nghiên cứu không đối chứng. Phần dùng thuốc là việc của bác sĩ, ở nhà không tự cho người ta thuốc hạ áp. Để người nằm thẳng xuống thì huyết áp còn cao hơn, nên phải đỡ ngồi dậy. Người nhà và người chăm cũng phải biết ba bước này. Đừng coi đó là đau đầu thường rồi bỏ qua. - ### 2. Mười năm đầu sau khi bị khuyết tật là giai đoạn nguy cơ tự tử cao, đừng tưởng chỉ cần nghĩ thoáng là xong - Chi phí: Không tốn tiền. Gọi một cuộc điện thoại, hoặc đi đăng ký một lần khám. @@ -29,7 +28,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Savic G, DeVivo MJ, Frankel HL, Jamous MA, Soni BM, Charlifue S (2018). Suicide and traumatic spinal cord injury - a cohort study. Spinal Cord, 56(1):2-6. - Ghi chú: Nghiên cứu này chỉ xem chấn thương tủy sống, áp sang mù lòa, cắt cụt, bỏng và các nguyên nhân khuyết tật khác thì là suy diễn lan. Nó cho biết được khung thời gian: nguy cơ không hết sau khi ra viện, phải theo dõi mười năm. Cần người bên cạnh nói chuyện thì gọi 12356, xem mục 11 ở phần 29 (12356). Đừng tích thuốc ngủ và thuốc trừ sâu trong nhà, xem mục 25 ở phần 1 (ý nghĩ tự tử). Đau buồn và cảm xúc mắc kẹt sau nửa năm vẫn thế thì đi đăng ký khoa tâm thần hoặc tâm lý lâm sàng, xem mục 8 ở phần 29 (sau nửa năm vẫn còn đau buồn). - ### 3. Nhập viện điều trị rối loạn tâm thần theo nguyên tắc tự nguyện, chỉ có hai trường hợp được làm mà không cần sự đồng ý của bản thân - Chi phí: Không tốn tiền. Ghi nhớ hai trường hợp, đến lúc cần là nói được ngay. @@ -38,7 +36,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Luật Sức khỏe Tâm thần của Cộng hoà Nhân dân Trung Quốc (中华人民共和国精神卫生法)（2018 年修正，第三十条、第三十一条、第三十二条、第三十五条、第七十八条）. - Ghi chú: Ngược lại, cũng không phải ai cũng bị cấm đưa đi khám. Điều 28 quy định, người nghi bị rối loạn tâm thần đã có hành vi tự gây hại cho bản thân hoặc gây nguy hại cho an toàn người khác, hoặc có nguy cơ đó, thì người thân thích, cơ quan công tác, công an địa phương phải ngay lập tức áp dụng biện pháp ngăn chặn, và đưa người đó đến cơ sở y tế để chẩn đoán rối loạn tâm thần. Cơ sở y tế khi tiếp nhận người đưa đến thì không được từ chối chẩn đoán. Nên「đưa đi chẩn đoán」và「giữ lại nhập viện」là hai cánh cửa, cánh trước thấp, cánh sau cao. Điều 75 liệt kê trách nhiệm của cơ sở y tế: thực hiện cách ly cưỡng chế trái quy định, chẩn đoán người không bị rối loạn tâm thần thành rối loạn tâm thần, thì bị xử lý cho đến khi bị thu hồi chứng chỉ hành nghề. Người bên cạnh có xung lực gây hại thì làm gì, xem phần 8. - ### 4. Người chăm lâu dài người khuyết tật trong nhà, sức khỏe của chính mình cũng phải có người nhìn - Chi phí: Không tốn tiền. Giao bớt phần việc chăm sóc đi một phần, tự đi khám sức khỏe một lần. @@ -47,7 +44,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Schulz R, Beach SR (1999). Caregiving as a risk factor for mortality: the Caregiver Health Effects Study. JAMA, 282(23):2215-2219. - Ghi chú:Tranh cãi: Giới dưới của khoảng tin cậy của nghiên cứu này đúng 1.00, khả năng「không có khác biệt» chưa bị loại hẳn. Đối tượng là người già Mỹ sống cùng phối ngẫu, người được chăm là phối ngẫu già mất năng lực, áp sang người chăm thân nhân khuyết tật trẻ tuổi thì là suy diễn lan. Nó cho biết được hướng: nhóm mệt thì nguy cơ cao hơn. Công việc chăm giao được thì giao bớt, dịch vụ nuôi dưỡng và chăm sóc ban ngày hỏi hội người khuyết tật địa phương và phòng dân vụ. Người chăm sóc đừng tự ngừng thuốc cho bệnh mạn của mình, xem mục 1 ở phần 16 (uống thuốc đủ theo chỉ định). Người được lợi của mục này là bạn. - ### 5. Người ngồi xe lăn lâu thì đổi đệm ngồi sang loại giảm áp, mỗi ngày xem một lượt xương ngồi và xương cùng - Chi phí: Một chiếc đệm giảm áp giá vài trăm tệ đến hơn một nghìn tệ. Mỗi ngày xem một lượt, một phút. @@ -56,7 +52,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: B - Nguồn: Brienza D, Kelsey S, Karg P, et al. (2010). A randomized clinical trial on preventing pressure ulcers with wheelchair seat cushions. Journal of the American Geriatrics Society, 58(12):2308-2314. - Ghi chú: Chọn B có hai lý do. Thứ nhất mẫu chỉ có 232 người, P của điểm chính 0.04 vừa vượt qua, điểm phụ không ra khác biệt. Thứ hai đối tượng là cư dân cao tuổi ở viện dưỡng lão, áp sang người dùng xe lăn trẻ tuổi thì là suy diễn lan. Trong thử nghiệm tất cả đều được lắp xe lăn vừa vặn trước, nên「vừa vặn」và「đệm」cùng có tác dụng, đừng chỉ đổi đệm. Phòng loét cho người nằm lâu: đệm khí và lật người đúng giờ, xem mục 8 ở phần 17 (loét). Cách nộp đơn xin trợ cấp lắp thiết bị hỗ trợ kiểu cơ bản, xem mục 7 của phần này (hỏi hết ở hội người khuyết tật cấp huyện). - ### 6. Đừng mua các liệu pháp và thiết bị「chữa khỏi bại liệt, mù lòa, điếc」 - Chi phí: Không tốn tiền, lại còn tiết kiệm được tiền. Khó là người nhà còn ôm hy vọng, khuyên không nổi. @@ -66,14 +61,14 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Nguồn: Luật Quảng cáo của Cộng hoà Nhân dân Trung Quốc (中华人民共和国广告法)（2021 年修正，第四条、第十六条、第十七条）. - Ghi chú: Cách phán đoán chỉ một: xem nó có dám nói hết hay không. Khoa phục hồi chức năng của bệnh viện chính quy không hứa với bạn mấy đợt điều trị là đứng dậy được. Gặp loại quảng cáo này thì báo 12315 hoặc cơ quan quản lý thị trường địa phương. Bộ thoại quảng cáo bảo phẩm cũng cùng một lối, xem mục 10 ở phần 6 (bảo phẩm). Đã mua rồi muốn đòi lại tiền, làm theo cách mua hàng qua mạng và tiêu dùng trả trước, xem mục 29 ở phần 5 (mua qua mạng theo quy tắc của nền tảng). Người được lợi của mục này là bạn và gia đình. -### 7. Sau khi giấy chứng nhận người khuyết tật đã được cấp, đến hội người khuyết tật cấp huyện hỏi hết những khoản có thể xin trong một lần +### 7. Có giấy chứng nhận người khuyết tật rồi thì đến hội người khuyết tật cấp huyện hỏi trọn một lần các khoản trợ cấp -- Chi phí: Không tốn tiền. Đi một chuyến tới hội người khuyết tật cấp huyện, hỏi cả buổi sáng. Sau đó còn phải đi nhiều cơ quan nữa, không chỉ một. -- Nói thẳng: Bản thân giấy chứng nhận người khuyết tật không phát tiền, nó là một chìa khóa. Phía sau nó là mấy khoản quyền lợi tách bạch với nhau, mỗi khoản đều phải tự nộp đơn, không khoản nào tự đến. Lần đầu đến hội người khuyết tật, hãy hỏi lần lượt từng mục trong danh sách, sẽ đỡ phải chạy đi chạy lại mấy chuyến. -- Lợi ích: Giấy chứng nhận mở ra ít nhất sáu khoản quyền lợi, mỗi khoản có căn cứ riêng. Một là trợ cấp sinh hoạt cho người khuyết tật khó khăn và trợ cấp chăm sóc cho người khuyết tật nặng, xem mục 8 ở phần 7 (hai khoản trợ cấp). Hai là trợ giúp phục hồi chức năng cho trẻ khuyết tật, xem mục 8 của phần này (trợ giúp phục hồi chức năng). Ba là trợ cấp lắp thiết bị hỗ trợ kiểu cơ bản, căn cứ là Điều hai mươi sáu của Quy định về Phòng ngừa khuyết tật và Phục hồi chức năng cho người khuyết tật (《残疾预防和残疾人康复条例》). Bốn là trợ cấp cải tạo tiện nghi không vật cản trong gia đình, xem mục 9 của phần này (làm lối đi, lan can và cải tạo phòng tắm). Năm là tuyển dụng theo tỷ lệ và dịch vụ việc làm, xem mục 10 của phần này (chủ động nói ra mình có giấy). Sáu là giảm thuế thu nhập cá nhân, xem mục 11 của phần này (giảm thuế cá nhân). Điều hai mươi sáu của Quy định này còn viết rằng nhà nước thông qua thực hiện các dự án phục hồi chức năng trọng điểm để cung cấp dịch vụ phục hồi chức năng cơ bản cho người khuyết tật nghèo ở đô thị và nông thôn, cho người khuyết tật nặng,「按照国家有关规定对基本型辅助器具配置给予补贴」— tức là trợ cấp cho việc lắp thiết bị hỗ trợ kiểu cơ bản theo quy định của nhà nước. Điều hai mươi quy định các cơ quan liên quan của chính quyền nhân dân cấp huyện trở lên và hội người khuyết tật phải tận dụng nguồn lực cộng đồng, lập địa điểm phục hồi chức năng hoặc mua dịch vụ, để làm công tác phục hồi chức năng tại cộng đồng như hướng dẫn phục hồi, huấn luyện khả năng sinh hoạt hằng ngày, chăm sóc phục hồi, lắp thiết bị hỗ trợ, tư vấn thông tin và chuyển giới thiệu (toàn quốc, có hiệu lực từ ngày 1 tháng 7 năm 2017) +- Chi phí: Không tốn tiền. Đến hội người khuyết tật cấp huyện một chuyến, hỏi một buổi sáng. Các cơ quan phải chạy sau đó không chỉ một. +- Nói thẳng: Bản thân giấy chứng nhận người khuyết tật không phát tiền, nó là một chìa khoá. Sau nó là mấy khoản trợ cấp độc lập với nhau, mỗi khoản đều phải tự mình xin, không tự tới. Lần đầu đến hội người khuyết tật, hãy hỏi lần lượt từng mục trong danh sách, đỡ hơn chạy đi chạy lại mấy chuyến. +- Lợi ích: Ít nhất sáu khoản trợ cấp nối với giấy chứng nhận, mỗi khoản có căn cứ riêng. Một là trợ cấp sinh hoạt cho người khuyết tật khó khăn và trợ cấp chăm sóc cho người khuyết tật nặng, xem mục 8 ở phần 7 (hai khoản trợ cấp). Hai là trợ giúp phục hồi chức năng cho trẻ khuyết tật, xem mục 8 của phần này (trợ giúp phục hồi chức năng). Ba là trợ cấp trang bị thiết bị hỗ trợ loại cơ bản, căn cứ là Điều 26 của Quy định về Phòng ngừa khuyết tật và Phục hồi chức năng cho người khuyết tật (《残疾预防和残疾人康复条例》). Bốn là trợ cấp cải tạo tiện nghi cho gia đình có người khuyết tật, xem mục 9 của phần này (lối đi, lan can, cải tạo phòng tắm). Năm là việc làm theo tỷ lệ quy định và dịch vụ việc làm, xem mục 10 của phần này (chủ động nói ra mình có giấy). Sáu là giảm thuế thu nhập cá nhân, xem mục 11 của phần này (giảm thuế thu nhập cá nhân). Điều 26 của Quy định này còn viết rằng nhà nước thông qua việc thực hiện các dự án phục hồi chức năng trọng điểm để cung cấp dịch vụ phục hồi chức năng cơ bản cho người khuyết tật nghèo ở thành thị và nông thôn, người khuyết tật nặng,「按照国家有关规定对基本型辅助器具配置给予补贴」. Điều 20 quy định, các cơ quan có liên quan ở cấp huyện trở lên của chính quyền nhân dân và hội người khuyết tật phải dùng nguồn lực cộng đồng để lập nơi phục hồi chức năng hoặc mua dịch vụ, làm công tác phục hồi chức năng ở cộng đồng: hướng dẫn phục hồi chức năng, luyện tập khả năng sinh hoạt hàng ngày, chăm sóc phục hồi chức năng, cấp thiết bị hỗ trợ, tư vấn thông tin và chuyển giới thiệu (toàn quốc, quy định có hiệu lực từ ngày 1 tháng 7 năm 2017) - Mức bằng chứng: A -- Nguồn: Hội đồng Nhà nước (国务院) (2017). Quy định về Phòng ngừa khuyết tật và Phục hồi chức năng cho người khuyết tật (《残疾预防和残疾人康复条例》)（第二十条、第二十六条）. -- Ghi chú: Quy trình làm giấy chứng nhận người khuyết tật, xem mục 11 ở phần 24 (giấy chứng nhận người khuyết tật). Tiêu chuẩn trợ cấp và cách làm cụ thể phần lớn do cấp tỉnh quy định, cùng một khoản ở hai tỉnh có thể chênh nhau mấy lần. Khi hỏi, ghi lại tên từng khoản một, về nhà tra lại văn bản gốc theo tên trên trang chính thức của hội người khuyết tật địa phương và trên cổng thông tin điện tử của chính quyền, đừng chỉ nghe lời nói miệng bên quầy. Đừng trông chờ hỏi trọn trong một lần, cứ sau một năm hỏi lại, chính sách có thay đổi. +- Nguồn: Hội đồng Nhà nước (国务院) (2017). Quy định về Phòng ngừa khuyết tật và Phục hồi chức năng cho người khuyết tật (《残疾预防和残疾人康复条例》)(第二十条、第二十六条). +- Ghi chú: Thủ tục làm giấy chứng nhận người khuyết tật xem mục 10 ở phần 24 (giấy chứng nhận người khuyết tật). Mức trợ cấp và cách làm cụ thể phần lớn do cấp tỉnh quy định, cùng một khoản ở hai tỉnh có thể chênh nhau vài lần. Khi hỏi thì ghi lại tên từng khoản, về nhà tra đúng tên đó trên trang chính thức của hội người khuyết tật địa phương và trên trang chính phủ, xem lại văn bản gốc, đừng chỉ nghe lời nói miệng tại quầy. Đừng trông chờ một lần hỏi là hết, cứ cách một năm đi hỏi lại, chính sách có thay đổi. ### 8. Trẻ dưới 7 tuổi có khuyết tật hoặc tự kỷ, đến hội người khuyết tật cấp huyện xin trợ giúp phục hồi chức năng @@ -83,7 +78,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2018). Ý kiến về việc lập chế độ trợ giúp phục hồi chức năng cho trẻ khuyết tật (《国务院关于建立残疾儿童康复救助制度的意见》)(国发〔2018〕20 号). ；Hội đồng Nhà nước (国务院) (2017). Quy định về Phòng ngừa khuyết tật và Phục hồi chức năng cho người khuyết tật (《残疾预防和残疾人康复条例》)（第二十六条）. - Ghi chú: Tiêu chuẩn trợ giúp do từng tỉnh quy định, trước hết tra trên trang chính thức của hội người khuyết tật địa phương「残疾儿童康复救助实施办法」. Danh sách cơ sở chỉ định cũng nằm trong tay hội. Văn bản yêu cầu hội phối hợp với các cơ quan liên quan siết chặt quản lý việc cơ sở phục hồi chức năng chỉ định được vào và được ra, lập cơ chế kiểm tra định kỳ và đánh giá tổng hợp, và lập danh sách cấm. Cơ sở không có trong danh sách thì bạn phải tự trả toàn bộ, có chuyện gì cũng khó truy trách nhiệm. Người được lợi của mục này là con bạn, thuộc mức thứ hai. Cơ sở nào hứa「chắc chắn khỏi」thì xử lý theo mục 6 của phần này (chữa khỏi bại liệt). - ### 9. Lối đi, lan can và cải tạo phòng tắm trong nhà, có thể xin trợ cấp từ chính quyền cấp huyện trở lên - Chi phí: Nộp đơn không tốn tiền. Phần cải tạo phần lớn có trợ cấp, bạn có thể phải bỏ thêm một phần. Phải đi hội người khuyết tật và ủy ban phường. @@ -92,7 +86,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: B - Nguồn: Luật Xây dựng môi trường không vật cản của Cộng hoà Nhân dân Trung Quốc (中华人民共和国无障碍环境建设法)（第十八条、第十九条、第六十二条、第六十三条）. - Ghi chú: Để ở mức B là vì luật chỉ viết「应当给予适当补贴」, trợ cấp bao nhiêu, ai thỏa điều kiện thì giao hết cho địa phương, không có con số thống nhất cả nước. Trước hết tra trên trang chính thức của hội người khuyết tật địa phương「困难重度残疾人家庭无障碍改造」. Hạn ngạch phần lớn chia theo năm, phải xếp hàng sớm. Lắp thang máy trong khu chung cư, cải tạo hành lang là một quy trình khác, đi theo cơ quan nhà ở và đô thị. Tiện nghi không vật cản ở nơi công cộng không đạt thì khiếu nại với cơ quan chủ trì theo Điều sáu mươi hai, cơ quan đó phải trả lời bạn. - ### 10. Khi tìm việc, chủ động nói ra mình có giấy, doanh nghiệp tuyển bạn thì được trừ một khoản tiền - Chi phí: Không tốn tiền. Thêm một câu trong buổi phỏng vấn. Khó là mở lời. @@ -101,7 +94,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2007). Quy định về Việc làm cho người khuyết tật (《残疾人就业条例》)(第八条、第九条). ；Bộ Tài chính, Cục Tổng cục Thuế nhà nước, Hiệp hội Người khuyết tật Trung Quốc (中国残联) (2015). Biện pháp thu, chi quỹ bảo đảm việc làm cho người khuyết tật (《残疾人就业保障金征收使用管理办法》)(财税〔2015〕72 号，第六条、第八条、第十六条). ；Luật Bảo vệ Người khuyết tật của Cộng hoà Nhân dân Trung Quốc (中华人民共和国残疾人保障法)（第三十六条）. - Ghi chú: Với doanh nghiệp vừa và nhỏ mới đăng ký dưới ba năm, dưới 20 người, lý do này không thành lập, đừng trông chờ. Chính sách giảm, miễn quỹ bảo đảm mấy năm nay cứ điều chỉnh, cụ thể theo thông báo của Bộ Tài chính trong năm. Ngoài ra, yêu cầu việc làm theo tỷ lệ là việc làm thật. Đứng tên hưởng lương, để đơn vị lấy bạn để khấu trừ quỹ bảo đảm, là hành vi lừa đảo, bạn cũng bị kéo vào. Xác định là người khó khăn trong việc làm và trợ cấp bảo hiểm xã hội, xem mục 12 ở phần 7 (người khó khăn trong việc làm). Trợ cấp đào tạo nghề, xem mục 13 ở phần 7 (trợ cấp đào tạo nghề). Người được lợi của mục này là bạn. - ### 11. Thuế thu nhập cá nhân của người khuyết tật có thể được giảm, giảm bao nhiêu thì gọi điện hỏi cơ quan thuế tỉnh - Chi phí: Không tốn tiền. Kê khai thì chọn một mục, hoặc nộp một bộ hồ sơ khai báo một lần. @@ -110,7 +102,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Luật Thuế thu nhập cá nhân của Cộng hoà Nhân dân Trung Quốc (中华人民共和国个人所得税法)（2018 年修正，第五条）. ；Luật Bảo vệ Người khuyết tật của Cộng hoà Nhân dân Trung Quốc (中华人民共和国残疾人保障法)（第三十六条）. - Ghi chú: Đặt mức lợi ích là「nhỏ」, vì bản thân luật không đưa ra con số, phần lớn các tỉnh miễn giảm mỗi năm ở mức vài trăm đến vài nghìn tệ. Số tiền không nhiều, nhưng làm một lần thì dùng được nhiều năm, đáng đi. Cụ thể tỉnh của bạn giảm bao nhiêu thì chỉ thông báo của cơ quan thuế địa phương mới có giá trị, bảng đối chiếu được chuyền tay trên mạng thường xuyên lỗi thời. Người tự mở cửa hàng nhỏ thì các loại phí hành chính, phí sự nghiệp công được miễn theo Điều ba mươi sáu của Luật Bảo vệ Người khuyết tật, lúc xin giấy phép nhớ nêu ra. - ### 12. Chó dẫn đường cho người mù được vào nơi công cộng và phương tiện giao thông công cộng, người mù đi xe buýt trong thành phố miễn phí - Chi phí: Không tốn tiền. Bị chặn thì phải mở miệng viện dẫn điều luật, có khi phải cãi lại một lúc. @@ -128,7 +119,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Bộ Giáo dục (教育部), Hội Liên hiệp Người khuyết tật Trung Quốc (中国残联) (2017). Quy định về việc tham gia kỳ thi tuyển sinh quốc gia vào trường đại học thường chính của người khuyết tật (教学〔2017〕4 号，第五条、第七条、第八条、第九条、第十八条). - Ghi chú: Thời hạn xin do từng tỉnh quy định, ghi trong quy chế đăng ký thi tuyển sinh của năm đó, trễ là mất. Vì vậy học kỳ một lớp mười hai phải vào trang của cơ quan kiểm tra giáo dục cấp tỉnh để xác nhận thời hạn. Điều mười tám còn quy định, các kỳ thi giáo dục quốc gia khác như thi sau đại học, thi tốt nghiệp trung học phổ thông người lớn cũng được áp dụng khi xin. Quy định ở giai đoạn đi học rằng nhà trường không được từ chối, xem mục 14 của phần này (nhà trường không được từ chối). Người hưởng mục này là bạn hoặc con bạn. - ### 14. Trẻ khuyết tật xin nhập học, nhà trường không được từ chối; trẻ không đến được trường thì phòng giáo dục sắp xếp giảng viên đến tận nhà - Chi phí: Không tốn tiền. Phải nộp một đơn, có thể phải qua một lần đánh giá. Khi nhà trường đẩy trách nhiệm thì phải lên tới phòng giáo dục. @@ -137,7 +127,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Hội đồng Nhà nước (国务院) (2017). Quy định về giáo dục người khuyết tật (残疾人教育条例，第七条、第二十条、第二十三条、第二十九条、第五十二条). - Ghi chú: Nhà trường nói miệng「chỗ chúng tôi không có điều kiện」thì không được. Yêu cầu nhà trường trả lời bằng văn bản, cầm bản trả lời đó đến phòng giáo dục cấp huyện. Việc đánh giá do phòng giáo dục ủy thác cho ủy ban chuyên gia. Câu「yêu cầu học tập có thể có mức độ linh hoạt」trong điều hai mươi ba rất có ích, gặp nhà trường dùng kết quả học tập làm lý do khuyên bỏ học thì dẫn câu này. Con bị bắt nạt ở trường thì lấy bằng chứng thế nào, nhà trường bắt buộc phải đi qua bước nào, xem mục 3 ở phần 30 (bắt nạt học đường). Biện pháp hỗ trợ hợp lý cho kỳ thi tuyển sinh đại học xem mục 13 của phần này (biện pháp hỗ trợ hợp lý). Người hưởng mục này là con bạn. - ### 15. Mất hoặc không cử động được hai chi dưới bên phải vẫn thi được giấy phép lái xe, loại xe được phép lái gọi là C5 - Chi phí: Học phí tương tự giấy phép lái xe thường, ở hầu hết các nơi vài nghìn tệ. Xe phải đội lại thành điều khiển bằng tay. Cứ ba năm kiểm tra sức khoẻ một lần. @@ -146,7 +135,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Lệnh số 172 của Bộ Công an (公安部令第 172 号). Quy định về việc cấp và sử dụng giấy phép lái xe cơ giới (机动车驾驶证申领和使用规定，第七十八条、第八十五条、附件 1). - Ghi chú: Loại xe dễ hơn nhiều là xe lăn có động cơ dành cho người khuyết tật. Điều 119 của Luật An toàn giao thông đường bộ xếp loại xe này vào nhóm phương tiện không có động cơ, nên không cần giấy phép lái xe; Điều 58 quy định, xe lăn có động cơ dành cho người khuyết tật khi chạy trong làn xe tự động không được vượt quá tốc độ tối đa mười lăm ki lô mét một giờ. Xe mã C5 phải lắp bước ga và phanh cơ, việc đội xe phải tìm nhà sản xuất có giấy phép, đội xong đi đăng ký đổi thông tin tại cơ quan quản lý phương tiện. Không nộp giấy chứng minh điều kiện thể chất trong vòng một năm sau khi hết ba chu kỳ điểm thì giấy phép lái xe sẽ bị thu hồi. - ### 16. Phục hồi chức năng làm ở cơ sở chính quy có khoa phục hồi chức năng, cường độ nghe người trị liệu, tập càng nặng càng không tốt - Chi phí: Nằm viện phục hồi chức năng mỗi ngày vài trăm tệ, phần tự trả thay đổi tuỳ địa phương. Thời gian tính theo tháng. Ngày nào cũng phải tập, rất chán. @@ -155,7 +143,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Langhorne P, Ramachandra S; Stroke Unit Trialists' Collaboration (2020). Organised inpatient (stroke unit) care for stroke: network meta-analysis. Cochrane Database of Systematic Reviews, 4:CD000197. ；AVERT Trial Collaboration group (2015). Efficacy and safety of very early mobilisation within 24 h of stroke onset (AVERT): a randomised controlled trial. Lancet, 386(9988):46-55. - Ghi chú:Tranh cãi: Cả hai bằng chứng này đều đến từ đột quỵ, áp sang chấn thương tủy sống, cắt chiết, bỏng và các nguyên nhân khuyết tật khác thì là suy diễn lan. Kết quả AVERT nói rằng「càng sớm càng mạnh」không đúng, không phải là「vận động sớm không có ích」. Trong thử nghiệm, đa số người ở cả hai nhóm đều xuống giường trong vòng 24 giờ, khác nhau ở tần suất và tổng lượng. Khi chọn cơ sở, xem nó có khoa phục hồi chức năng không, có nhân viên chuyên trách về phục hồi chức năng và trị liệu hoạt động tự do không, đừng chỉ nhìn hai chữ「phục hồi」trên biển hiệu. Chuyển viện giữa tuyến cơ sở và bệnh viện hạng A, mức đồng chi trả tính thế nào, xem mục 1 ở phần 24 (chuyển viện theo từng bậc). Phục hồi chức năng cộng đồng và lắp thiết bị hỗ trợ xem mục 7 của phần này (hỏi hết ở hội người khuyết tật cấp huyện). - ### 17. Nghe kém thì đi lắp máy trợ thính, lý do lắp là để nghe rõ, không phải để phòng chứng mất trí nhớ - Chi phí: Máy trợ thính vài nghìn tới vài chục nghìn tệ, phần lớn phải tự trả. Cần đo khớp, còn phải chỉnh vài lần. @@ -173,7 +160,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc (中华人民共和国民法典，第二十一条、第二十二条、第二十三条、第二十四条). - Ghi chú: Đơn này đi theo thủ tục đặc biệt của tòa án, nhanh hơn vụ án dân sự thường, nhưng vẫn mất mấy tháng. Phải nộp giấy chứng nhận chẩn đoán của bệnh viện, phần lớn tòa án còn ủy thác cho giám định tư pháp, phí giám định thường vài nghìn tệ, do người nộp đơn tạm trả trước. Không qua thủ tục này mà ký thay hoặc xử lý tài sản của người khác thì về sau có thể bị công nhận là vô hiệu, căn nhà đã bán cũng có thể phải hoàn trả. Còn một điểm phải nhớ: có năng lực hành vi dân sự bị giới hạn không có nghĩa là không làm được gì cả. Việc phù hợp với trình độ trí tuệ, tình trạng tâm thần của họ, và việc thuần lợi, thì họ tự làm được, đừng thay tất cả. Người giám hộ được xác định thế nào thì xem mục 19 của phần này (người giám hộ). - ### 19. Người giám hộ của người thành niên được xác định theo thứ tự luật định; người được giám hộ gây hại cho người khác thì người giám hộ bồi thường - Chi phí: Không tốn tiền. Khi có tranh chấp thì phải đến ủy ban nhân dân khu phố, ủy ban nhân dân làng, cơ quan dân vụ hoặc tòa án để giải quyết. @@ -182,7 +168,6 @@ Hầu hết các mục trong phần này mang lại lợi ích cho bạn và gia - Mức bằng chứng: A - Nguồn: Bộ luật Dân sự của Cộng hoà Nhân dân Trung Quốc (第二十八条、第三十一条、第三十三条、第一千一百八十八条、第一千一百八十九条). - Ghi chú: Điều ba mươi ba chính là giám hộ tự nguyện bằng văn bản. Khi người đó còn minh mẫn thì dùng văn bản chốt người được chọn, sau này khỏi phải tranh chấp phần lớn, viết cụ thể thế nào thì xem mục 1 ở phần 17 (giám hộ tự nguyện bằng văn bản). Đoạn bồi thường phải cẩn thận: giao người cho cơ sở nuôi dưỡng hay người thân chăm sóc thì trách nhiệm vẫn nằm ở người giám hộ, người được ủy thác chỉ chịu trách nhiệm tương ứng khi bản thân có lỗi. Vì vậy mua cho người được giám hộ một bảo hiểm trách nhiệm người giám hộ là có nghĩa. Điều bảy mươi tám của Luật Sức khỏe Tâm thần còn quy định, người giám hộ bỏ rơi người bệnh hoặc không làm tròn nghĩa vụ giám hộ mà gây hại thì phải chịu trách nhiệm bồi thường theo pháp luật. - ### 20. Bị từ chối tuyển, không được chuyển biên chế, bị trả lương thiếu vì khuyết tật thì có thể kiện, nhưng hãy tính chi phí trong quá trình trước - Chi phí: Phí luật sư vài nghìn tới vài chục nghìn tệ, do bạn tự trả. Bản án sơ thẩm theo thủ tục thông thường kéo dài từ 6 tháng, có thể kéo dài hơn. Bằng chứng bạn phải tự giữ. diff --git "a/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" index 200a57dc..9798aa4b 100644 --- "a/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" +++ "b/vi/book/34-\345\256\266\351\207\214\347\232\204\345\270\270\345\244\207\350\215\257\345\210\253\345\220\203\345\207\272\344\272\213.md" @@ -18,7 +18,6 @@ Mục trong phần này có người hưởng là bạn và gia đình bạn, th - Mức bằng chứng: B - Nguồn: Larson AM, Polson J, Fontana RJ, et al.; Acute Liver Failure Study Group (2005). Acetaminophen-induced acute liver failure: results of a United States multicenter, prospective study. Hepatology, 42(6):1364-1372. ；Thông báo của Cục Quản lý Dược phẩm Quốc gia về việc sửa đổi hướng dẫn sử dụng dược phẩm paracetamol phóng thích và phóng thích kéo dài (2020年第15号) kèm phụ lục 2《对乙酰氨基酚常释制剂非处方药说明书修订要求》, đăng lại bởi Sở Quản lý Dược phẩm tỉnh Hunan (湖南省药品监督管理局). ；21 CFR 201.326(a)(1)(iii)(A) Liver warning. - Ghi chú: Paracetamol còn gọi là 扑热息痛. Tên thông thường của thuốc cảm cúm phối hợp có hai chữ「氨酚」, như 氨酚伪麻, 氨咖黄敏, đều chứa thành phần này. Không chắc thì mang cả mấy hộp thuốc đến hiệu thuốc cho dược sĩ xem. Để ở mức B là vì nhóm số liệu Mỹ là đăng ký ca bệnh, không có nhóm đối chứng, không tính được xem bảng thành phần thì bớt được bao nhiêu sự cố. Cỡ lợi ích đặt là「lớn」là theo hậu quả: gần ba phần mườu ca suy gan cấp chết thẳng. - ### 2. Trẻ bị sốt đừng dùng aspirin, nimesulide và metamizole, thuốc cảm cúm chứa metamizole cũng không cho người chưa thành niên uống - Chi phí: Không tốn tiền. Khi mua thuốc hạ sốt thì nhìn tên thông thường và bảng thành phần. @@ -27,7 +26,6 @@ Mục trong phần này có người hưởng là bạn và gia đình bạn, th - Mức bằng chứng: B - Nguồn: Belay ED, Bresee JS, Holman RC, Khan AS, Shahriari A, Schonberger LB (1999). Reye's syndrome in the United States from 1981 through 1997. New England Journal of Medicine, 340(18):1377-1382. ；Thông tư về việc tăng cường quản lý việc sử dụng dạng uống của nimesulide (国食药监安〔2011〕209号). ；Thông báo của Cục Quản lý Dược phẩm Quốc gia về việc sửa đổi hướng dẫn sử dụng cho các chế phẩm liên quan đến metamizole (2020年第34号) kèm phụ lục 1 đến 3. - Ghi chú: Metamizole với người lớn cũng không phải lựa chọn hàng đầu. Hướng dẫn ghi rõ nó có thể gây mất bạch cầu hạt, thiếu máu bất tái sinh và sốc dị ứng, chỉ dùng ngắn hạn khi bệnh nặng cấp tính mà không có thuốc hiệu quả khác. 重感灵 nghe tên như thuốc cảm cúm thường, bên trong có metamizole. Trẻ nên dùng thuốc hạ sốt nào, liều bao nhiêu theo cân nặng thì hỏi bác sĩ nhi khoa hoặc dược sĩ. Trẻ dưới 3 tháng bị sốt thì đi thẳng bệnh viện, xem mục 8 ở phần 20 (38 ℃ thì đi thẳng bệnh viện). Để ở mức B là vì số liệu hội chứng Reye đến từ báo cáo giám sát, việc số ca giảm và việc phát cảnh báo chỉ là quan hệ trước sau, không phải thử nghiệm đối chứng. Cỡ lợi ích đặt là「lớn」, vì sau khi có cảnh báo số ca rơi từ 555 ca mỗi năm xuống dưới 36 ca. - ### 3. Người trên 60 tuổi, đã từng xuất huyết dạ dày, đang uống thuốc chống đông hoặc hormone, trước khi uống thuốc giảm đau như ibuprofen hãy hỏi bác sĩ - Chi phí: Không tốn tiền. Hỏi bác sĩ hoặc dược sĩ một lần. @@ -36,7 +34,6 @@ Mục trong phần này có người hưởng là bạn và gia đình bạn, th - Mức bằng chứng: A - Nguồn: Coxib and traditional NSAID Trialists' (CNT) Collaboration; Bhala N, Emberson J, Merhi A, et al. (2013). Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. Lancet, 382(9894):769-779. ；21 CFR 201.326(a)(2)(iii)(A) Stomach bleeding warning. - Ghi chú: Kết luận của bài báo nói về cách dùng liều cao. Rủi ro khi thỉnh thoảng uống một hai ngày, đúng liều trong hướng dẫn, bài này không đưa con số riêng. Nhiều thuốc cảm cúm phối hợp cũng chứa thành phần nhóm này, uống cùng thuốc giảm đau thì thành phần bị lặp, xem bảng thành phần trước. Nếu bác sĩ bảo bạn uống aspirin liều thấp lâu dài để phòng huyết khối thì đừng tự ngừng, cũng đừng tự thêm một loại thuốc giảm đau nữa. - ### 4. Trẻ dưới 2 tuổi bị cảm cúm, đừng tự cho uống thuốc cảm cúm phối hợp và thuốc ho - Chi phí: Không tốn tiền, lại còn tiết kiệm tiền mua thuốc. @@ -45,7 +42,6 @@ Mục trong phần này có người hưởng là bạn và gia đình bạn, th - Mức bằng chứng: B - Nguồn: Thông báo của Cục Quản lý Dược phẩm Quốc gia về việc sửa đổi hướng dẫn sử dụng cho 14 chế phẩm như 氨酚麻美口服溶液 (2021年第57号) kèm phụ lục. ；Smith SM, Schroeder K, Fahey T (2014). Over-the-counter (OTC) medications for acute cough in children and adults in community settings. Cochrane Database of Systematic Reviews, (11):CD001831. - Ghi chú: Trong 14 loại này, nhiều tên có chữ「氨酚」, cũng chứa paracetamol, uống cùng thuốc hạ sốt thì thành phần bị lặp, xem mục 1 của phần này (paracetamol). Trong bài tổng quan đó có một thử nghiệm thấy mật ong tốt hơn giả dược, nhưng trẻ chưa đầy 1 tuổi không được cho ăn mật ong, xem mục 6 ở phần 20 (mật ong). Để ở mức B là vì mỗi nhóm thuốc trong tổng quan đều có ít thử nghiệm, tác giả không gộp được ra con số. Cỡ lợi ích đặt là「vừa」là đánh giá: chưa có bằng chứng lợi ích nào ở trẻ, rủi ro chủ yếu là quá liều và thành phần bị lặp. - ### 5. Sau tuần thứ 20 của thai kỳ, đừng tự uống thuốc giảm đau như ibuprofen - Chi phí: Không tốn tiền. @@ -63,7 +59,6 @@ Mục trong phần này có người hưởng là bạn và gia đình bạn, th - Mức bằng chứng: B - Nguồn: Thông báo của Cục Quản lý Dược phẩm Quốc gia về việc chuyển viên nén bao phủ ruột omeprazole từ thuốc kê đơn sang thuốc không kê đơn (2022年第68号) kèm phụ lục 2《非处方药说明书范本》. - Ghi chú: Clopidogrel là thuốc chống huyết khối mà người đã đặt stent, đã từng bị nhồi máu cơ tim hay dùng. Omeprazole làm nó yếu đi, người đang uống thì hỏi bác sĩ trước. Để ở mức B là vì đây chỉ là quy định trong hướng dẫn, không có nghiên cứu nào tính được làm theo thì bớt được bao nhiêu sự cố. Cỡ lợi ích đặt là「vừa」là đánh giá: nó ngăn việc kéo dài xuất huyết dạ dày và ung thư dạ dày, nhưng phần lớn triệu chứng trào ngược và ợ nóng không nặng. - ### 7. Cảm cúm thường đừng xin bác sĩ kê kháng sinh - Chi phí: Không tốn tiền, lại còn tiết kiệm tiền thuốc. @@ -72,7 +67,6 @@ Mục trong phần này có người hưởng là bạn và gia đình bạn, th - Mức bằng chứng: A - Nguồn: Kenealy T, Arroll B (2025). Antibiotics for the common cold and acute purulent rhinitis. Cochrane Database of Systematic Reviews, 11:CD000247. - Ghi chú: Bản này phát hành lại năm 2025, phần tóm tắt ghi rõ đã tìm tài liệu đến năm 2013. Kháng sinh ở trong nước là thuốc kê đơn, hiệu thuốc phải bán theo đơn, kháng sinh còn dư trong nhà cũng đừng tự lấy ra uống. Cỡ lợi ích đặt là「vừa」là đánh giá: nó ngăn tác dụng phụ, không ngăn tử vong. - ### 8. Tiêu chảy thì uống dung dịch bù nước đường uống trước, trẻ dưới 5 tuổi không cho thuốc cầm tiêu chảy - Chi phí: Một hộp dung dịch bù nước đường uống vài tệ đến hơn mười tệ, nên để sẵn một hộp trong nhà. @@ -81,7 +75,6 @@ Mục trong phần này có người hưởng là bạn và gia đình bạn, th - Mức bằng chứng: A - Nguồn: World Health Organization (2005). The treatment of diarrhoea: a manual for physicians and other senior health workers, 4th revision. ；Hahn S, Kim S, Garner P (2002). Reduced osmolarity oral rehydration solution for treating dehydration caused by acute diarrhoea in children. Cochrane Database of Systematic Reviews, (1):CD002847. - Ghi chú: Khi mua, xem áp suất thẩm thấm ghi trên hướng dẫn, chọn công thức nhẩm. Sổ tay ghi rõ, trẻ đi ngoài có máu phần lớn là kiết lị, phải đi khám bác sĩ dùng kháng sinh, không thể chỉ bù nước ở nhà. Cỡ lợi ích đặt là「vừa」là đánh giá: con số trong tay so hai loại dung dịch bù nước với nhau, không phải so có uống với không uống. - ### 9. Thuốc giảm đau trong một tháng đừng uống quá 15 ngày, thuốc giảm đau phối hợp đừng quá 10 ngày, nếu không đau đầu có thể do thuốc - Chi phí: Không tốn tiền. Phải ghi nhớ số ngày uống thuốc, và phải kiềm chế không tự tiện uống. @@ -90,3 +83,21 @@ Mục trong phần này có người hưởng là bạn và gia đình bạn, th - Mức bằng chứng: B - Nguồn: Headache Classification Committee of the International Headache Society (2018). The International Classification of Headache Disorders, 3rd edition. Cephalalgia, 38(1):1-211. ；ICHD-3 bản trực tuyến 8.2 Medication-overuse headache (gồm 8.2.3, 8.2.5). - Ghi chú: Viên giảm đau mà tên có chữ「咖」thì hầu như đã thêm caffeine, tính như thuốc phối hợp, tức 10 ngày. Số ngày đau đầu trong tháng càng tăng dần thì đi khám khoa thần kinh, đừng chỉ dựa vào tăng liều. Để ở mức B là vì tiêu chuẩn phân loại đưa ra là ngưỡng chẩn đoán, không phải thử nghiệm đối chứng. Cỡ lợi ích đặt là「vừa」là đánh giá: nó đổi lại là những ngày ít đau đầu hơn, không liên quan tới tử vong. +### 10. Định kỳ dọn thuốc hết hạn và thuốc thừa không dùng tới, bỏ cả vỏ vào thùng rác chất độc, đừng để trong nhà + +- Chi phí: Không tốn tiền. Mỗi năm lục tủ thuốc một lần, mất mười mấy phút. +- Nói thẳng: Trong các thuốc trẻ tự uống nhầm mà không có người lớn nhắc, gần một nửa là viên thuốc kê đơn. Trong số thuốc kê đơn bị uống nhầm, khoảng ba phần mười là của ông bà. Dọn hết thuốc không dùng tới thì bớt một món trẻ có thể tìm thấy. Thuốc đã dọn thì bỏ cả vỏ vào thùng rác chất độc, hoặc gửi cho hiệu thuốc có thu gom thuốc cũ. +- Lợi ích: Dữ liệu giám sát phòng cấp cứu của Mỹ ước tính, từ năm 2004 đến 2013, trẻ dưới 6 tuổi tự uống thuốc không có người lớn nhắc đã khám cấp cứu khoảng 640.000 lượt (giá trị ước tính 640161 lượt). Riêng năm 2013 khoảng 59.000 lượt (59092 lượt). Từ 2010 đến 2013, loại phổ biến nhất là thuốc uống dạng viên kê đơn, chiếm 45.9%. Trong thuốc kê đơn thì phổ biến nhất là thuốc giảm đau nhóm opioid (13.8%) và thuốc an thần gây ngủ nhóm benzodiazepin (12.7%). Năm 2017, 5 trung tâm kiểm soát độc tại Mỹ hỏi các bậc phụ huynh gọi điện cầu cứu. Trong số người trẻ tự uống nhầm thuốc kê đơn, 30.7% uống thuốc của ông bà. Hơn một nửa (51.5%) số thuốc kê đơn đã bị người lớn lấy ra khỏi vỏ gốc trước khi đến tay trẻ. Danh mục chất thải nguy hại quốc gia của Trung Quốc (2025 年版) xếp thuốc thải loại phát sinh trong sinh hoạt gia đình và bao bì của thuốc vào chất thải nguy hại trong rác thải sinh hoạt. Hướng dẫn phân loại rác của Thượng Hải xếp thuốc thải vào rác chất độc, ghi rõ bao gồm thuốc hết hạn, thuốc viên nang, thuốc dạng viên, bao bì trong của thuốc,「nên bỏ thuốc thải kèm cả bao bì」. Thượng Hải còn quy định thuốc hết hạn thải của hộ gia đình do hiệu thuốc thu gom sẽ có điểm trung chuyển tạm do cơ quan quản lý thị trường của từng quận đặt. FDA của Mỹ nói, cách xử lý tốt nhất thuốc hết hạn và thuốc không dùng tới là giao cho điểm thu gom. +- Mức bằng chứng: B +- Nguồn: Lovegrove MC, Weidle NJ, Budnitz DS (2015). Trends in Emergency Department Visits for Unsupervised Pediatric Medication Exposures, 2004-2013. Pediatrics, 136(4):e821-e829. ；Agarwal M, Lovegrove MC, Geller RJ, et al. (2020). Circumstances Involved in Unsupervised Solid Dose Medication Exposures among Young Children. Journal of Pediatrics, 219:188-195.e6. ；生态环境部等 (2024). 国家危险废物名录（2025年版），附录《危险废物豁免管理清单》第 1 项. ；上海市绿化和市容管理局 (2024). 上海市生活垃圾分类投放指引（2024版）（沪绿容〔2024〕278号）. ；上海市绿化和市容管理局、上海市环境保护局、上海市食品药品监督管理局 (2018). 关于规范生活垃圾分类有害垃圾全程管理的通知. ；U.S. Food and Drug Administration (2025). Where and How to Dispose of Unused Medicines. +- Ghi chú: Danh mục phân loại rác ở các địa phương không giống hệt nhau, cứ theo bảng dán ở nhà rác của khu chung cư. Thượng Hải xếp bã cỏ của thuốc thảo dược vào rác ẩm, không phải rác chất độc. Thuốc còn để lại thì giữ trong hộp gốc, không đổ vào hộp chia liều hay bình khác, đặt ở chỗ trẻ không với tới được, xem mục 20 ở phần 13 (uống nhầm). Kháng sinh còn thừa trong nhà cũng đừng tự lấy ra uống, xem mục 7 của phần này (kháng sinh). Ở nước không có điểm thu gom thì Mỹ cho trộn thuốc vào bã cà phê, đất vệ sinh rồi bịt túi bỏ vào rác thường, một ít thuốc giảm đau tác dụng mạnh thì được phép xả xuống bồn cầu. Đó là cách làm ở Mỹ, trong nước thì theo quy định phân loại rác của địa phương. Đặt mức B vì mấy nhóm số liệu Mỹ đó chỉ là số đăng ký, chưa có nghiên cứu nào tính được dọn tủ thuốc sẽ giảm bao nhiêu ca uống nhầm. Mức lợi ích đặt là「nhỏ」: số liệu trong tay là số lượt cấp cứu, không phải số ca tử vong, cũng tính không ra tác dụng của việc dọn. + + +### 11. Khi uống kháng sinh nhóm cephalosporin, metronidazole, trong lúc uống và 7 ngày sau khi ngừng thì không uống rượu; thuốc có cồn như nước Hoa Hương Chính Khí, siro ho vẫn cũng không uống + +- Chi phí: Không tốn tiền. Một tuần hơn không uống rượu, và trước khi uống thuốc khác thì nhìn thành phần xem có ethanol không. +- Nói thẳng: Một phần cephalosporin, metronidazole gặp rượu sẽ làm người ta đỏ mặt, hồi hộp, tức ngực, nôn, nặng thì huyết áp tụt. Không chỉ rượu trắng và bia, thuốc có cồn như nước Hoa Hương Chính Khí, siro ho cũng gây ra. Suốt lúc uống thuốc đến 7 ngày sau khi ngừng thì đừng đụng vào. +- Lợi ích: Hiện tượng đó gọi là phản ứng kiểu disulfiram. Bài giáo dục sử dụng thuốc an toàn năm 2023 của Sở Dược quản lý tỉnh Phúc Kiên nói, uống rượu trước và sau khi dùng một số thuốc cephalosporin có thể gây phản ứng này,「có thể gây tổn hại nghiêm trọng cho sức khoẻ thậm chí đe doạ tính mạng」. Triệu chứng gồm đỏ mặt, ra mồ hôi, hồi hộp, thậm chí khó thở, huyết áp tụt, sốc phản vệ. Bài giáo dục liệt kê các thuốc theo《Hướng dẫn dùng thuốc lâm sàng của Dược điển Cộng hoà Nhân dân Trung Quốc》gồm cefazolin, ceftriaxone, cefoperazone, metronidazole, tinidazole, furazolidone và các thuốc khác. Bài viết rõ「trong 7 ngày sau khi dùng thuốc cấm uống rượu và ăn thuốc, thức ăn có cồn」, ví dụ có sô cô lao rượu và nước Hoa Hương Chính Khí. Lý do là loại enzyme bị thuốc ức chế phải mất 4 đến 5 ngày mới hồi phục. Một bệnh viện ở Quảng Châu tổng hợp 82 ca trong 2 năm. Trong đó 57 ca do nhóm cephalosporin (69.51%), 10 ca do nhóm nitroimidazole (metronidazole, tinidazole loại này) (12.20%). Rượu đến từ đồ uống có cồn là 40 ca (48.78%), đến từ thuốc có ethanol là 31 ca (37.80%). Các thuốc uống gồm nước Hoa Hương Chính Khí, siro ho, ca lỏng huyết thông và ích mẫu. Trong 82 ca có 68 ca nhẹ, 9 ca trung bình, 5 ca nặng, đều thuyên giảm trong 1 đến 3 giờ, không có ca tử vong. Tờ hướng dẫn dùng metronidazole của Mỹ yêu cầu trong lúc dùng thuốc và ít nhất 3 ngày sau khi ngừng không uống rượu, và không dùng sản phẩm có chứa propylen glycol. +- Mức bằng chứng: B +- Nguồn: 福建省药品监督管理局（福建省药品科普与监管数据中心）(2023). 2023年“全国安全用药月”原创科普小知识——为什么吃头孢不能喝酒. ；房财富, 涂星, 刘韬, 黄红兵, 邱孟, 梁蔚婷 (2015). 我院双硫仑样反应82例分析. 中国药房, 26(29):4089-4091. ；Pfizer (2026). FLAGYL 375 (metronidazole) capsules, prescribing information: Contraindications, Interaction with Alcohol. DailyMed. +- Ghi chú: Không phải loại cephalosporin nào cũng bị, nhưng người bình thường không phân biệt được, dù loại nào thì làm theo mục này là tiện nhất. Khi bác sĩ kê cephalosporin hoặc metronidazole, bạn chủ động nhắc rằng mấy ngày trước đó có uống rượu không. Bài giáo dục đó khuyên trước khi dùng thuốc 3 ngày cũng đừng uống, đồng thời thừa nhận khoảng cách này hiện chưa có nói chính thức. Vừa đỏ mặt hồi hộp thì ngay lập tức ngừng mọi thứ có cồn. Bài giáo dục nói ca nhẹ tự thuyên giảm được, ca nặng thì phải đưa đi bệnh viện. Trong lúc uống paracetamol cũng đừng uống rượu, xem mục 1 của phần này (paracetamol). Đặt mức B vì bằng chứng là tổng hợp ca bệnh và tờ hướng dẫn, không có nhóm đối chứng. Mức lợi ích đặt là「nhỏ」: cả 82 ca đó đều thuyên giảm trong vài giờ, không có ca tử vong, chỉ 5 ca nặng. diff --git "a/vi/docs/\345\210\232\347\241\256\350\257\212\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216.md" "b/vi/docs/\345\210\232\347\241\256\350\257\212\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216.md" new file mode 100644 index 00000000..35d0412a --- /dev/null +++ "b/vi/docs/\345\210\232\347\241\256\350\257\212\346\205\242\346\200\247\347\227\205\344\271\213\345\220\216.md" @@ -0,0 +1,50 @@ +[← Về mục lục](../README.md) + +# Vừa được chẩn đoán bệnh mạn tính: làm gì trong tuần đầu, ba tháng đầu và sau đó + +Danh sách này viết cho người vừa được chẩn đoán bệnh mạn tính kiểu cao huyết áp, tiểu đường, mỡ máu cao, gout. Các mục liên quan trong sách nằm rải rác ở nhiều phần, ở đây đã xếp theo thứ tự thời gian. Mỗi mục chỉ ghi hành động cần làm, phần chỉ dẫn đi kèm ngay sau. Vì sao làm, số liệu và nguồn ở đều nằm trong mục được trỏ tới. Mục nào không hợp với bệnh của bạn thì bỏ qua. + +## Tuần được chẩn đoán + +1. Bệnh nặng thì ngày lấy kết quả, nhờ một người đi cùng. Về nhà đừng ở một mình. Nghỉ việc, bán nhà các kiểu để sau. Xem mục 2 ở phần 29 (Trong tuần bạn được báo mắc bệnh nặng). +2. Bác sĩ bảo uống thuốc thì uống đúng đơn từ hôm nay. Thấy khá hơn cũng đừng ngừng. Xem mục 1 ở phần 16 (Uống thuốc đủ theo đơn của bác sĩ). +3. Thuốc cao huyết áp, mỡ máu cao thì uống đúng y lệnh. Đừng tự ý ngừng. Xem mục 12 ở phần 2 (Có cao huyết áp, mỡ máu cao thì uống thuốc đúng y lệnh). +4. Người cao huyết áp thì mua một máy đo huyết áp điện tử, tự đo ở nhà, hạ tới khi đạt chuẩn. Xem mục 7 ở phần 1 (Đo huyết áp, cao rồi thì uống thuốc hạ cho tới mức chuẩn). +5. Lúc kê đơn thì hỏi bác sĩ có thuốc tập trung đấu thầu hay thuốc generic không. Xem mục 12 ở phần 5 (Ưu tiên thuốc generic đã đạt đánh giá tính nhất quán và thuốc trong gói đấu thầu tập trung). +6. Xác định tiểu đường loại 2 thì ngay lúc đó đi khám đáy mắt có nhoè một lần. Tiểu đường loại 1 thì bắt đầu khám từ 5 năm sau khi bệnh khởi phát. Xem mục 7 ở phần 16 (Xác định tiểu đường thì đi khám đáy mắt một lần). +7. Chẩn đoán gout thì bắt đầu uống thuốc hạ acid uric lâu dài, ép acid uric máu xuống dưới 360 µmol/L. Xem mục 9 ở phần 16 (Chẩn đoán gout thì uống thuốc hạ acid uric lâu dài). +8. Người Han dùng allopurinol thì hỏi bác sĩ có cần xét HLA-B\*5801 trước không. Xem mục 9 ở phần 16 (Chẩn đoán gout thì uống thuốc hạ acid uric lâu dài). +9. Chỉ xét sức khoẻ ra acid uric cao mà chưa từng đau lần nào thì không phải gout. Chưa uống thuốc hạ acid uric. Xem mục 19 ở phần 6 (Đừng bắt đầu uống thuốc hạ acid uric chỉ vì xét sức khoẻ ra acid uric cao mà chưa từng đau lần nào). +10. Người hút thuốc thì chọn một ngày cai. Đến ngày đó bỏ một lần. Xem mục 4 ở phần 2 (Chọn một ngày cai, đến ngày đó bỏ một lần, đừng giảm dần trước). +11. Cai thuốc lá đừng chỉ dựa vào ý chí. Ra phòng khám cai thuốc lá hoặc khoa hô hấp để lấy thuốc. Xem mục 3 ở phần 2 (Đừng chỉ dựa vào ý chí để cai, hãy đi lấy thuốc trước). + +## Ba tháng đầu + +1. Chuẩn bị một quyển sổ. Mỗi lần tái khám ghi ngày, chỉ số, thuốc đang uống và liều. Xem mục 3 ở phần 16 (Tái khám theo khoảng cách bác sĩ dặn, ghi chỉ số mỗi lần vào cùng một quyển sổ). +2. Mỗi lần khám xong, tự giữ lại một bộ hồ sơ bệnh án, kết quả xét nghiệm và ảnh. Xem mục 5 ở phần 24 (Mỗi lần khám xong, tự giữ lại một bộ hồ sơ bệnh án, kết quả xét nghiệm và ảnh). +3. Ở nơi tham gia bảo hiểm y tế thì làm giấy công nhận bệnh mạn tính ngoại trú. Hay đi khám ở tỉnh khác thì đăng ký khám chữa tại nơi khác thêm nữa. Xem mục 2 ở phần 16 (Làm giấy công nhận bệnh mạn tính ngoại trú trước, rồi đăng ký khám chữa tại nơi khác). +4. Nhà có người tham gia bảo hiểm y tế người lao động thì nhờ người đó liên kết dùng chung tài khoản một lần. Tiền trong tài khoản cá nhân của họ dùng cho bạn khám và mua thuốc được. Xem mục 13 ở phần 5 (Làm một lần liên kết dùng chung tài khoản trên app bảo hiểm y tế). +5. Trước khi đăng ký bác sĩ gia đình ở cộng đồng, hỏi rõ cái nào vào bảo hiểm y tế, cái nào tự trả. Xem mục 6 ở phần 16 (Trước khi đăng ký bác sĩ gia đình ở cộng đồng, hỏi rõ cái nào vào bảo hiểm y tế, cái nào tự trả). +6. Muối ăn trong nhà đổi sang muối ít natri. Người suy thận không toàn phần hoặc đang uống thuốc giữ kali thì hỏi bác sĩ trước. Xem mục 9 ở phần 2 (Đổi muối ăn trong nhà sang muối ít natri (muối kali)). +7. Bắt đầu đi bộ mỗi ngày. Mục tiêu là 7000–8000 bước, hoặc đi bộ nhanh tích luỹ 150–300 phút mỗi tuần. Xem mục 11 ở phần 2 (Đi 7000–8000 bước mỗi ngày, hoặc đi bộ nhanh tích luỹ 150–300 phút mỗi tuần). +8. Uống ít rượu. Không uống được thì không uống. Xem mục 19 ở phần 2 (Uống ít hoặc không uống rượu). +9. Người uống rượu mỗi ngày, ngừng là run tay hồi hộp, đừng tự cai cứng. Ra bệnh viện trước. Xem mục 20 ở phần 2 (Người uống rượu mỗi ngày, ngừng là run tay hồi hộp, đừng tự cai cứng). +10. Có người khuyên ngừng thuốc để thử phương thuốc dân gian thì đừng ngừng. Muốn uống thực phẩm bổ sung thì nói với bác sĩ điều trị trước. Xem mục 4 ở phần 16 (Đừng bỏ điều trị chính thức chỉ để thử phương thuốc dân gian hay thực phẩm bảo vệ sức khoẻ). +11. Đừng bỏ tiền lớn vào thực phẩm bảo vệ sức khoẻ, thang thuốc bồi bổ, đồ bồi bổ để「bồi bổ cơ thể」. Xem mục 10 ở phần 6 (Đừng bỏ tiền lớn vào thực phẩm bảo vệ sức khoẻ (保健食品), thang thuốc bồi bổ (膏方), đồ bồi bổ để「bồi bổ cơ thể」). +12. Gout vừa bắt đầu hạ acid uric thì 3 đến 6 tháng đầu dễ lên cơn hơn. Đau thì uống thêm thuốc chống viêm theo y lệnh. Đừng ngừng thuốc hạ acid uric. Xem mục 9 ở phần 16 (Chẩn đoán gout thì uống thuốc hạ acid uric lâu dài). +13. Trước khi tự mua thuốc cảm cúm, thuốc giảm đau thì xem bảng thành phần trước. Paracetamol chỉ được chiếm một chỗ. Xem mục 1 ở phần 34 (Trước khi uống hai loại thuốc cảm cúm hoặc giảm đau cùng lúc, hãy xem bảng thành phần, paracetamol chỉ được chiếm một chỗ). +14. Người đang uống thuốc chống đông hoặc hormone thì hỏi bác sĩ trước khi uống thuốc giảm đau như ibuprofen. Xem mục 3 ở phần 34 (Người trên 60 tuổi, đã từng xuất huyết dạ dày, đang uống thuốc chống đông hoặc hormone, trước khi uống thuốc giảm đau như ibuprofen hãy hỏi bác sĩ). + +## Khi bệnh đã ổn định + +1. Bệnh đã ổn định, thuốc cũng không đổi, ra cộng đồng lấy thuốc một lần đủ dùng 12 tuần. Xem mục 5 ở phần 16 (Bệnh mạn tính đã ổn định, ở bệnh viện cộng đồng xin được thuốc một lần trong 12 tuần). +2. Tái khám theo khoảng cách bác sĩ dặn. Mỗi lần ghi vào cùng một quyển sổ. Xem mục 3 ở phần 16 (Tái khám theo khoảng cách bác sĩ dặn, ghi chỉ số mỗi lần vào cùng một quyển sổ). +3. Đổi sang thuốc tập trung đấu thầu thì theo dõi chỉ số trong sổ. Thấy thật sự xấu đi thì đi tìm bác sĩ. Đừng tự ý ngừng. Xem mục 12 ở phần 5 (Ưu tiên thuốc generic đã đạt đánh giá tính nhất quán và thuốc trong gói đấu thầu tập trung). +4. Người tiểu đường thì mỗi năm khám chân một lần. Khám đáy mắt theo khoảng cách bác sĩ dặn. Xem mục 7 ở phần 16 (Xác định tiểu đường thì đi khám đáy mắt một lần). +5. Người bị gout, acid uric đã về chuẩn vẫn uống tiếp. Không được ngừng. Xem mục 9 ở phần 16 (Chẩn đoán gout thì uống thuốc hạ acid uric lâu dài). +6. Người từng bị sỏi thận thì mỗi ngày uống 2,5–3 lít nước, hạ muối xuống dưới 6 g. Xem mục 8 ở phần 16 (Đã từng bị sỏi thận thì uống nước mỗi ngày 2,5–3 lít, hạ muối xuống dưới 6 g). +7. Người bệnh tim mạch thì mỗi năm mùa thu tiêm một mũi vaccine cúm. Xem mục 20 ở phần 1 (Người bệnh tim mạch và người cao tuổi tiêm vaccine cúm mỗi năm). +8. Bệnh thông thường thì khám ở cộng đồng trước. Muốn nằm viện thì đi theo chuyển tuyến từ tuyến cơ sở lên trên. Mức đồng chi trả nằm viện tính nối tiếp. Xem mục 1 ở phần 24 (Bệnh thông thường thì khám ở cộng đồng trước, đi theo chuyển tuyến từ tuyến cơ sở lên trên). +9. Bệnh khám ở tuyến cơ sở xong được thì đừng chạy lên trên. Trong cùng một vùng gộp quỹ, mỗi lần lên một cấp, tỷ lệ hoàn thấp hơn khoảng 10 điểm phần trăm. Xem mục 2 ở phần 24 (Trong cùng một vùng gộp quỹ (nơi tham gia bảo hiểm y tế), cấp bệnh viện càng thấp thì tỷ lệ hoàn càng cao, chênh nhau khoảng 10 điểm phần trăm). +10. Muốn đi bệnh viện lớn khám chuyên khoa thì đi theo chuyển tuyến từ cộng đồng hoặc trung tâm chuyển tuyến của bệnh viện. Đừng tìm kẻ bán số khám. Xem mục 3 ở phần 24 (Muốn đi bệnh viện lớn, đi theo chuyển tuyến từ tuyến cơ sở hoặc trung tâm chuyển tuyến của bệnh viện). +11. Định đi khám ở tỉnh khác thì trước khi đi hỏi một câu ở nơi đang sống. Số tiền hoàn về có thể thấp hơn khá nhiều so với khám ở nhà. Xem mục 4 ở phần 24 (Trước khi đi khám chữa bệnh ở tỉnh khác, hỏi một câu ở nơi bạn đang sống). diff --git "a/vi/docs/\345\255\251\345\255\220\345\207\272\347\224\237\345\211\215\345\220\216\350\246\201\345\212\236\347\232\204\344\272\213.md" "b/vi/docs/\345\255\251\345\255\220\345\207\272\347\224\237\345\211\215\345\220\216\350\246\201\345\212\236\347\232\204\344\272\213.md" new file mode 100644 index 00000000..ad0fbef6 --- /dev/null +++ "b/vi/docs/\345\255\251\345\255\220\345\207\272\347\224\237\345\211\215\345\220\216\350\246\201\345\212\236\347\232\204\344\272\213.md" @@ -0,0 +1,68 @@ +[← Về mục lục](../README.md) + +# Việc cần làm trước và sau khi có con: danh sách xếp theo thời gian + +Cả cuốn sách chia theo chủ đề, việc mang thai, sinh nở và chăm con nằm rải rác ở nhiều phần. Danh sách này xếp chúng theo thứ tự thời gian, từ lúc chuẩn bị mang thai đến khi con đủ một tuổi. Mỗi mục chỉ ghi cần làm gì. Vì sao làm và bằng chứng chắc tới đâu thì xem mục trong ngoặc ở cuối. Tới giai đoạn nào thì đọc từ tiêu đề con ở giai đoạn đó. + +## Một. Khi chuẩn bị mang thai + +1. Mỗi ngày uống một viên axit folic 0.4 miligam, bắt đầu uống trước 3 tháng. Xem mục 1 ở phần 27 (Chuẩn bị mang thai thì bắt đầu uống 0.4 miligam axit folic mỗi ngày, uống liên tới hết 3 tháng đầu thai kỳ). +2. Gọi điện một lần tới bảo hiểm xã hội hoặc bảo hiểm y tế, hỏi số tháng tham gia liên tục bảo hiểm sinh sản đã đủ chưa. Xem mục 11 ở phần 27 (Trước khi mang thai đã xác nhận tình trạng tham gia bảo hiểm sinh sản). +3. Nói với người ở cùng nhà, sau này ai cũng không hút thuốc trong nhà. Xem mục 2 ở phần 2 (Đừng hút thuốc trong nhà và trong xe, và cũng đừng để khách hút trong nhà). +4. Nhà có người sống dựa vào thu nhập của bạn thì mua bảo hiểm nhân thọ có thời hạn cho người kiếm tiền trước. Xem mục 40 ở phần 5 (Nhà có người sống dựa vào thu nhập của bạn thì mua bảo hiểm nhân thọ có thời hạn cho người kiếm tiền trước, đừng mua cho con trước). + +## Hai. Khi mang thai và trước ngày sinh + +1. Trước tuần thai kỳ thứ 13, ra trạm y tế cộng đồng lập Sổ sức khoẻ Mẹ và Con, khám thai miễn phí lấy hạn mức ở đây. Xem mục 2 ở phần 27 (Trước tuần thai kỳ thứ 13, ra trạm y tế cộng đồng lập Sổ sức khoẻ Mẹ và Con và dùng hết hạn mức khám thai miễn phí). +2. Lần khám thai đầu tiên thì loại luôn AIDS, giang mai, viêm gan B. Xét nghiệm không phải tự bỏ tiền. Xem mục 3 ở phần 27 (Lần khám thai đầu tiên thì loại luôn ba bệnh AIDS, giang mai, viêm gan B). +3. Suốt thai kỳ không hút một điếu thuốc, không uống một giọt rượu. Xem mục 4 ở phần 27 (Suốt thai kỳ không được điếu thuốc nào, không được uống rượu nào, người trong nhà cũng đừng hút trong nhà). +4. Lúc khám thai thì kể rõ bệnh sử. Có yếu tố nguy cơ cao tiền sản giật thì bác sĩ quyết định có uống aspirin liều thấp từ tuần thai kỳ thứ 12 không. Xem mục 5 ở phần 27 (Có yếu tố nguy cơ cao tiền sản giật thì từ tuần thai kỳ thứ 12 uống mỗi ngày một viên aspirin liều thấp). +5. Từ tuần thai kỳ thứ 24 trở đi, làm một lần sàng lọc đái thường thai kỳ. Xem mục 6 ở phần 27 (Từ tuần thai kỳ thứ 24 trở đi, làm một lần sàng lọc đái thường thai kỳ, đừng chê uống nước đường phiền). +6. Sau tuần thứ 20 của thai kỳ, đừng tự uống thuốc giảm đau như ibuprofen. Xem mục 5 ở phần 34 (Sau tuần thứ 20 của thai kỳ, đừng tự uống thuốc giảm đau như ibuprofen). +7. Học thuộc bảng「phải vào viện ngay」, và bảo người ở cùng nhà học luôn. Xem mục 7 ở phần 27 (Thuộc lòng bảng 「phải vào viện ngay」 này). +8. Lúc khám thai thì hỏi bệnh viện có làm đẻ không đau không. Xem mục 9 ở phần 27 (Muốn đẻ không đau thì cứ nói thẳng, nó không làm tăng nguy cơ mổ lấy thai). +9. Không có chỉ định y tế thì đừng chủ động xin mổ lấy thai. Xem mục 10 ở phần 27 (Không có chỉ định y tế thì đừng chủ động xin mổ lấy thai, cũng đừng mổ để chọn ngày). +10. Sinh ở nơi không phải nơi tham gia bảo hiểm y tế thì hỏi trước có cần đăng ký sinh con ở nơi khác không. Xem mục 11 ở phần 27 (Trước khi mang thai đã xác nhận tình trạng tham gia bảo hiểm sinh sản). +11. Nghỉ thai sản ít nhất 98 ngày. Trước ngày sinh có thể nghỉ trước 15 ngày. Xem mục 2 ở phần 18 (Nghỉ thai sản 98 ngày). +12. Cơ quan vì bạn mang thai mà giảm lương, đổi vị trí thì giữ lại thông báo bằng văn bản và lịch sử trò chuyện. Xem mục 3 ở phần 18 (Biết điều này: không được giảm lương hoặc sa thải vì mang thai, vì sinh con, vì cho con bú). +13. Giường cũi, xe đẩy trẻ thì mượn trước. Không mượn được thì mua đồ cũ. Ghế an toàn và nệm thì đừng mua đồ cũ. Xem mục 11 ở phần 20 (Với đồ lớn thì cứ theo thứ tự 「mượn, mua đồ cũ, mua mới」, đừng mua đủ một lần). +14. Bỉ thì mua bao nhỏ thử một tuần. Hợp rồi hãy mua nhiều. Xem mục 10 ở phần 20 (Bỉ thì không xem nhãn hiệu, xem ba điều). +15. Nghĩ trước tên con, và xác nhận hệ thống hộ khẩu nhập được mấy chữ đó. Xem mục 12 ở phần 27 (Trước khi xuất viện, làm Giấy chứng nhận sinh y tế, nghĩ trước tên con và đừng viết sai chữ). + +## Ba. Khi nằm viện sinh + +1. Vỡ ối thì nằm bẹp tại chỗ, kê hông lên, gọi 120. Không đi lại. Xem mục 8 ở phần 27 (Vỡ ối thì nằm bẹp tại chỗ, kê hông lên, gọi 120, không đi lại và không tắm). +2. Trong giờ đầu tiên sau khi sinh thì cho bú luôn. Xem mục 4 ở phần 20 (Sáu tháng đầu chỉ cho bú mẹ, không cần cho uống nước, từ 6 tháng thêm thức ăn dặm và tiếp tục cho bú mẹ). +3. Trong vòng 24 giờ sau khi sinh thì tiêm mũi đầu tiên vaccine viêm gan B. Mẹ dương tính kháng nguyên bề mặt viêm gan B thì tiêm thêm globulin miễn dịch viêm gan B cho con. Xem mục 2 ở phần 20 (Tiêm mũi đầu tiên vaccine viêm gan B trong vòng 24 giờ sau khi sinh). +4. Mũi tiêm vitamin K lúc sinh nhất định phải tiêm. Sinh ở nhà hoặc ở chỗ không đúng quy định thì hỏi một câu. Xem mục 7 ở phần 20 (Mũi tiêm vitamin K lúc sinh nhất định phải tiêm). +5. Sàng lọc máu gót chân và sàng lọc thính giác đừng từ chối. Trước khi ký giấy đồng ý bằng văn bản, xem rõ mục nào tự trà thêm. Xem mục 13 ở phần 27 (Cả sàng lọc máu gót chân trẻ sơ sinh lẫn sàng lọc thính giác, đừng từ chối). + +## Bốn. Trước khi xuất viện + +1. Làm Giấy chứng nhận sinh y tế. Đối chiếu từng chữ tên và số thẻ căn cước của bố mẹ và con. Xem mục 12 ở phần 27 (Trước khi xuất viện, làm Giấy chứng nhận sinh y tế, nghĩ trước tên con và đừng viết sai chữ). +2. Có Giấy chứng nhận sinh y tế thì làm bảo hiểm y tế cư dân cho con trên mạng, không cần đợi đăng ký hộ khẩu. Không chắc thì gọi 12393 hỏi. Xem mục 14 ở phần 27 (Có Giấy chứng nhận sinh y tế thì làm bảo hiểm y tế cho con ngay, đừng đợi đăng ký hộ khẩu xong). +3. Trên đường về nhà thì dùng ghế an toàn. Trẻ dưới 1 tuổi lắp ngược hướng, không bế trên tay. Xem mục 10 ở phần 1 (Trẻ dưới 4 tuổi thì dùng ghế an toàn, đừng bế trên tay). + +## Năm. Tháng đầu sau khi sinh + +1. Trong vòng một tháng thì ra công an đường phố khai báo đăng ký sinh. Xem mục 15 ở phần 27 (Trong vòng một tháng sau khi con sinh, ra công an đường phố khai báo đăng ký sinh). +2. Sau khi đăng ký hộ khẩu thì ra nơi có hộ khẩu của con xin trợ cấp nuôi con, mỗi đứa mỗi năm 3600 tệ. Xem mục 1 ở phần 18 (Cộng trước vào các khoản được nhận: trợ cấp nuôi con của Nhà nước mỗi đứa trẻ mỗi năm 3600 tệ, trả đến khi đủ 3 tuổi). +3. Cho con nằm ngửa, nằm trên mặt phẳng cứng, cùng phòng nhưng khác giường, trên giường không có gì mềm. Xem mục 1 ở phần 20 (Cho trẻ nằm ngửa, nằm trên mặt phẳng cứng, cùng phòng nhưng khác giường, trên giường không có gì mềm). +4. Pha sữa công thức bằng nước trên 70 ℃, phần còn lại thì đổi đi. Xem mục 5 ở phần 20 (Pha sữa công thức bằng nước trên 70 ℃, pha xong để nguội rồi cho uống, phần còn lại thì đổi đi). +5. Trẻ dưới 3 tháng tuổi, nhiệt độ chạm 38 ℃ thì đi bệnh viện luôn. Xem mục 8 ở phần 20 (Trẻ dưới 3 tháng tuổi, nhiệt độ chạm 38 ℃ thì đi bệnh viện luôn, không quan sát ở nhà). +6. Sắp không kiềm chế nổi thì đặt con lại vào giường cũi, người rời đi vài phút. Xem mục 9 ở phần 20 (Dù mệt mỏi, giận dữ đến đâu cũng đừng lắc đứa trẻ). +7. Bảng「phải vào viện ngay」 cho người mới sinh, trong vòng một năm sau sinh đều tính. Xem mục 7 ở phần 27 (Thuộc lòng bảng 「phải vào viện ngay」 này). + +## Sáu. Từ sau một tháng đến một tuổi + +1. Đi tái khám 42 ngày sau khi sinh, trong đó có một lần sàng lọc trầm cảm sau sinh. Xem mục 16 ở phần 27 (Đừng bỏ lần tái khám 42 ngày sau khi sinh, lần đó đồng thời là sàng lọc trầm cảm sau sinh). +2. Theo tuổi tháng thì ra phòng tiêm chuẩn, tiêm đủ các vaccine trong lịch tiêm chuẩn của quốc gia. Vaccine bạch cởa, ho cực và uốn lách bắt đầu từ 2 tháng tuổi. Xem mục 3 ở phần 20 (Tiêm đủ các loại vaccine trong lịch tiêm chuẩn của quốc gia, toàn bộ miễn phí, lỡ mất thì chỉ bù những liều chưa tiêm). +3. 6 tháng đầu chỉ cho bú mẹ. Đủ 6 tháng thì thêm thức ăn dặm, mỗi ngày phải có thực phẩm động vật nhiều sắt. Xem mục 4 ở phần 20 (Sáu tháng đầu chỉ cho bú mẹ, không cần cho uống nước, từ 6 tháng thêm thức ăn dặm và tiếp tục cho bú mẹ). +4. Không cho con dưới 1 tuổi ăn mật ong. Xem mục 6 ở phần 20 (Không cho con dưới 1 tuổi ăn mật ong). +5. Con bị eczema nặng hoặc dị ứng trứng thì khám bác sĩ trước, rồi mới quyết định có thêm đậu phộng sớm không. Xem mục 12 ở phần 20 (Con bị eczema nặng hoặc dị ứng trứng, đừng né đậu phộng). +6. Trẻ dưới 2 tuổi bị cảm cúm thì đừng tự cho uống thuốc cảm cúm phối hợp. Xem mục 4 ở phần 34 (Trẻ dưới 2 tuổi bị cảm cúm, đừng tự cho uống thuốc cảm cúm phối hợp và thuốc ho). +7. Lắp chặn cho cửa sổ và ban công. Lưới chống muỗi không tính là biện pháp bảo vệ. Xem mục 11 ở phần 1 (Nhà có trẻ nhỏ thì lắp chặn cửa sổ và ban công, lưới chống muỗi không tính là biện pháp bảo vệ). +8. Từ 0 đến 3 tuổi không cho dùng màn hình như điện thoại, máy tính bảng. Xem mục 5 ở phần 30 (Không cho màn hình ở tuổi 0 đến 3, hạn chế tối đa ở tuổi 3 đến 6). +9. Đi làm lại thì khi con chưa đủ một tuổi, mỗi ngày có 1 giờ cho con bú. Xem mục 3 ở phần 18 (Biết điều này: không được giảm lương hoặc sa thải vì mang thai, vì sinh con, vì cho con bú). +10. Liên kết tài khoản cá nhân của bảo hiểm y tế người lao động để dùng chung. Tiền đó dùng cho con khám và mua thuốc được. Xem mục 13 ở phần 5 (Làm một lần liên kết dùng chung tài khoản trên app bảo hiểm y tế). +11. Mỗi năm khi tổng kết thuế thu nhập cá nhân thì điền khoản trừ cho việc chăm trẻ dưới 3 tuổi. Xem mục 2 ở phần 5 (Mỗi năm từ tháng Ba đến tháng Sáu làm một lần tổng kết thuế thu nhập cá nhân, khai đủ các khoản trừ bổ sung chuyên ngành). diff --git "a/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" "b/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" index 9715084d..3d358a62 100644 --- "a/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" +++ "b/vi/docs/\347\224\237\347\211\251\351\222\237\345\222\214\345\244\234\347\217\255.md" @@ -2,7 +2,7 @@ # Cơ thể nhận ra thời gian ra sao, và ca đêm vì sao gây hại -Đây là bản viết dài của mục 40 ở phần 2 (làm ca đêm càng lâu thì nguy cơ tim mạch càng cao). Mục đó trong phần chính chỉ tính số, bài này nói lý do đằng sau. +Đây là bản viết dài của mục 39 ở phần 2 (làm ca đêm càng lâu thì nguy cơ tim mạch càng cao). Mục đó trong phần chính chỉ tính số, bài này nói lý do đằng sau. Câu hỏi cần trả lời là: cơ thể có thật sự biết bây giờ là mấy giờ không. Nếu biết, tại sao làm ca đêm không phải chuyện "thức thêm vài lần là quen". @@ -140,9 +140,9 @@ Hai cách đối phó ở trên, một cách dừng ở chỗ chuyển hóa và ## 9. Những phần bài này không nói -Ca đêm làm tăng nguy cơ tim mạch bao nhiêu, và tính theo số năm thì ra sao, ở mục 40 ở phần 2 (làm ca đêm càng lâu thì nguy cơ tim mạch càng cao). +Ca đêm làm tăng nguy cơ tim mạch bao nhiêu, và tính theo số năm thì ra sao, ở mục 39 ở phần 2 (làm ca đêm càng lâu thì nguy cơ tim mạch càng cao). -Sau một đêm thức khuya thì bù lại giấc ngủ thế nào, ở mục 39 ở phần 2 (thức khuya xong thì bù ngủ vào tối hôm sau, đừng dồn dập cuối tuần). +Sau một đêm thức khuya thì bù lại giấc ngủ thế nào, ở mục 38 ở phần 2 (thức khuya xong thì bù ngủ vào tối hôm sau, đừng dồn dập cuối tuần). Ban đêm nên ngủ bao lâu, và giờ giấc có đều hay không, ở mục 13 ở phần 2 (mỗi đêm ngủ khoảng 7 giờ, giữ giờ giấc cố định). diff --git "a/vi/docs/\350\242\253\350\243\201\344\272\206\344\271\213\345\220\216\345\205\210\345\201\232\344\273\200\344\271\210.md" "b/vi/docs/\350\242\253\350\243\201\344\272\206\344\271\213\345\220\216\345\205\210\345\201\232\344\273\200\344\271\210.md" new file mode 100644 index 00000000..892f0c42 --- /dev/null +++ "b/vi/docs/\350\242\253\350\243\201\344\272\206\344\271\213\345\220\216\345\205\210\345\201\232\344\273\200\344\271\210.md" @@ -0,0 +1,54 @@ +[← Về mục lục](../README.md) + +# Bị sa thải thì làm gì trước + +Việc phải làm sau khi bị cắt giảm, bị sa thải nằm rải rác ở phần 19, phần 7 và mấy phần khác. Danh sách dưới đây xếp theo thứ tự thời gian. Mỗi bước chỉ ghi làm gì, còn chi tiết, số liệu và căn cứ thì nằm ở mục được trỏ trước dấu ngoặc. Nếu chỉ bị nợ lương mà người lao động chưa đi khỏi thì xem ngay mục 2 ở phần 7 (bị nợ lương thì khiếu nại giám sát lao động trước). + +## Ngày đó: trước khi ký + +1. Công ty đưa tờ đơn「tự nguyện nghỉ việc vì lý do cá nhân」thì không ký. Ký là mất trắng cả khoản bồi thường kinh tế lẫn trợ cấp thất nghiệp. Xem mục 7 ở phần 19 (Đừng ký「tự nguyện nghỉ việc vì lý do cá nhân」, ký là mất trắng cả khoản N). +2. Công ty nói「ký trước để làm thủ tụng, phần bồi thường tính riêng」thì yêu cầu ghi khoản bồi thường vào văn kiện ngay tại chỗ. Không ghi thì không ký. Không ký trong ngày đó không có hậu quả pháp lý nào. Xem mục 7 ở phần 19 (Đừng ký「tự nguyện nghỉ việc vì lý do cá nhân」, ký là mất trắng cả khoản N). +3. Lúc nói chuyện thì bật ghi âm. Cuộc nói chuyện bạn tham gia thì không cần xin phép đối phương trước. Xem mục 41 ở phần 8 (Cuộc gọi và cuộc gặp có thể đổi thái độ thì bật ghi âm ngay). +4. Trước khi trả lại máy tính và tài khoản, lưu phiếu lương, bản chấm công, hợp đồng lao động, hồ sơ bảo hiểm xã hội và lịch sử trò chuyện. Xem mục 8 ở phần 19 (Trước khi nghỉ việc thì lưu trước phiếu lương, bản chấm công, hợp đồng lao động, hồ sơ bảo hiểm xã hội và lịch sử trò chuyện). +5. Chỉ lưu tài liệu về quan hệ lao động của bạn. Không mang mã nguồn, danh sách khách hàng, tài liệu kỹ thuật của công ty đi. Xem mục 7 ở phần 11 (Rời công ty không mang mã nguồn, danh sách khách hàng, tài liệu kỹ thuật đi). +6. Bàn giao xong quyền truy cập tài khoản, không xoá dữ liệu, không đổi mật khẩu, dù công ty còn nợ bạn. Xem mục 6 ở phần 11 (Rời công ty thì bàn giao xong quyền truy cập tài khoản). + +## Từ ngày đó tới mấy ngày sau: tính rõ tiền + +1. Tính N trước: cứ mỗi năm làm đủ một năm thì một tháng lương, chưa đủ sáu tháng thì tính nửa tháng. Xem mục 4 ở phần 19 (Bị sa thải thì tính rõ N trước: cứ mỗi năm làm đủ một năm thì một tháng lương, chưa đủ sáu tháng thì tính nửa tháng). +2. Lương tháng tính theo mức bình quân của 12 tháng trước khi rời đi. Thưởng, phụ cấp, trợ cấp đều phải tính vào. Chỗ này hay bị công ty làm thấp nhất. Xem mục 4 ở phần 19 (Bị sa thải thì tính rõ N trước: cứ mỗi năm làm đủ một năm thì một tháng lương, chưa đủ sáu tháng thì tính nửa tháng). +3. Công ty không báo trước bằng văn bản 30 ngày thì ngoài N còn phải trả thêm một tháng lương. Khi thương lượng thì tách tháng đó ra, nói riêng. Xem mục 5 ở phần 19 (Công ty sa thải bạn mà không báo trước 30 ngày thì còn phải trả thêm một tháng lương). +4. Nếu công ty chấm dứt sai luật thì tiền bồi thường bằng 2N. Có phải chấm dứt sai luật không, trước tiên hỏi trợ giúp pháp lý hoặc luật sư cho rõ, rồi hãy quyết định thương lượng hay đi hoà giải lao động. Xem mục 6 ở phần 19 (Công ty chấm dứt hợp đồng sai luật thì tiền bồi thường bằng hai lần mức bồi thường kinh tế). +5. Phần nghỉ phép năm chưa nghỉ hết thì khi rời việc tính theo số tháng đã làm trong năm mà trả một lượt. Xem mục 2 ở phần 19 (Nghỉ phép năm tính theo thâm niên cộng dồn). +6. Tiền làm thêm giờ còn nợ thì tính vào luôn. Xem mục 1 ở phần 19 (Tiền làm thêm giờ tính theo ba mức 1.5 lần, 2 lần, 3 lần). + +## Tuần đầu + +1. Đi làm đăng ký thất nghiệp. Trợ cấp thất nghiệp tính từ ngày làm đăng ký. Chậm một tháng thì mất một tháng trợ cấp. Xem mục 1 ở phần 7 (Mất việc thì xin trợ cấp thất nghiệp trực tuyến trước). +2. Đăng ký thất nghiệp làm ở cơ quan dịch vụ việc làm công cộng như trung tâm dịch vụ việc làm. Không thu tiền. Xem mục 5 ở phần 7 (Tìm việc thì dùng dịch vụ việc làm công cộng miễn phí và thị trường việc lẻ trước, đừng qua môi giới thu phí). +3. Xin trợ cấp thất nghiệp trên app「掌上 12333」hoặc tiểu chương trình nhỏ thẻ bảo hiểm xã hội điện tử. Phải đóng đủ 1 năm và không phải tự mình xin nghỉ việc thì mới được nhận. Xem mục 1 ở phần 7 (Mất việc thì xin trợ cấp thất nghiệp trực tuyến trước). +4. Trong thời gian nhận trợ cấp thất nghiệp, bảo hiểm y tế người lao động do quỹ bảo hiểm thất nghiệp đóng. Không phải tự đóng thêm. Xem mục 1 ở phần 7 (Mất việc thì xin trợ cấp thất nghiệp trực tuyến trước). +5. Không đủ tiền kiện tụng thì gọi 12348 xin trợ giúp pháp lý. Nay không cần về quê lấy giấy chứng minh khó khăn về tài chính. Xem mục 3 ở phần 7 (Không đủ tiền kiện tụng thì xin trợ giúp pháp lý). +6. Đã ký thỏa thuận không cạnh tranh thì giữ bản gốc. Công ty phải trả bồi thường hằng tháng. Vì lý do công ty mà 3 tháng không trả thì có thể xin chấm dứt. Xem mục 12 ở phần 11 (Đã ký thỏa thuận không cạnh tranh). +7. Ổn định giờ sinh hoạt: dậy đúng giờ, xếp việc tìm việc vào lịch cố định, đừng ở nhà suốt ngày. Xem mục 3 ở phần 29 (Sau khi thất nghiệp, ổn định giờ sinh hoạt, bảo hiểm y tế và nhịp đi tìm việc trước, đừng ở nhà suốt ngày). + +## Tháng đầu + +1. Lương, tiền làm thêm giờ bị kéo dài không trả thì trước hết gọi 12333 khiếu nại giám sát lao động, sau đó xin hoà giải lao động. Cả hai đường đều không thu tiền. Xem mục 2 ở phần 7 (Bị nợ lương thì khiếu nại giám sát lao động trước, không được thì xin hoà giải lao động, cả hai đường đều miễn phí). +2. Người thuê bạn là xưởng không có giấy phép thì vẫn đi theo giám sát lao động và hoà giải lao động. Tiền đối tác thi công nợ thì tìm chủ thầu tổng. Chỉ khi làm việc riêng cho gia đình hoặc cá nhân thì mới ra tòa. Xem mục 22 ở phần 7 (Bị nợ lương công nhân, trước hết phân biệt ai là người thuê bạn). +3. Hoà giải lao động phải được yêu cầu trong vòng 1 năm sau khi rời việc. Quá hạn thì đối phương chỉ cần một câu「quá thời hiệu」là xong. Xem mục 19 ở phần 8 (Bảo vệ quyền lợi có thời hạn: thời hiệu khởi kiện dân sự là 3 năm, hoà giải lao động là 1 năm). +4. Tìm việc thì tới cơ quan dịch vụ việc làm công cộng trước. Giới thiệu việc và hướng dẫn nghề đều miễn phí. Đừng tìm môi giới thu phí. Xem mục 5 ở phần 7 (Tìm việc thì dùng dịch vụ việc làm công cộng miễn phí và thị trường việc lẻ trước, đừng qua môi giới thu phí). +5. Tìm việc đừng chỉ nộp hồ sơ. Học kỹ thuật tìm việc, đặt mục tiêu rõ ràng, mở lời nhờ người quen giới thiệu. Xem mục 14 ở phần 7 (Tìm việc đừng chỉ nộp hồ sơ, hãy dùng cách có hệ thống: học kỹ thuật, đặt mục tiêu, nhờ người khác giúp). +6. Nghe nói「trả tiền trước rồi lên ca」,「đào tạo trước rồi sắp xếp việc」,「kéo người」thì từ chối ngay tại chỗ. Xem mục 15 ở phần 7 (Không nộp tiền cọc, không giữ giấy tờ, không ký「培训贷」, không vào đa cấp, không vay lãi cao). +7. Muốn học kỹ năng thì tra「Danh mục trợ cấp đào tạo nghề」ở cơ quan nhân sự và xã hội địa phương trước. Đừng tự bỏ tiền học khóa. Xem mục 13 ở phần 7 (Trong thời gian thất nghiệp thì đi nhận trợ cấp đào tạo nghề). +8. Xem mình có đủ điều kiện được công nhận là người khó khăn việc làm không, chẳng hạn lớn tuổi, thất nghiệp lâu dài. Công nhận được rồi thì tự đóng bảo hiểm xã hội có trợ cấp, còn xếp được việc công ích. Xem mục 12 ở phần 7 (Làm xong đăng ký thất nghiệp thì hãy xin công nhận là người khó khăn việc làm). + +## Mấy tháng sau đó + +1. Chi phí sinh hoạt lấy từ khoản dự phòng trước. Không vay tiền, cũng đừng bán thua lỗ những khoản đầu tư. Xem mục 27 ở phần 5 (Để dành trước khoản dự phòng bằng 3 đến 6 tháng chi phí, đặt ở chỗ rút ra được bất cứ lúc nào). +2. Trong ba tháng đầu, quyết định lớn không thể đảo ngược thì đẩy lùi. Bán nhà, dùng cả khoản bồi thường một lượt để đầu tư, góp vốn, bảo lãnh cho người khác đều tính vào. Xem mục 12 ở phần 29 (Trong ba tháng đầu sau biến cố, mọi quyết định lớn không thể đảo ngược đều đẩy lùi). +3. Đã hết trợ cấp thất nghiệp mà vẫn chưa tìm được việc thì đừng để bảo hiểm y tế bị trống. Bảo hiểm y tế cư dân tự đóng theo năm được. Đã đứt quãng rồi đóng lại thì có 3 tháng khám bệnh không được hoàn. Xem mục 9 ở phần 7 (Đừng để bảo hiểm y tế cư dân bị đứt quãng mỗi năm 400 tệ, hộ khó khăn được miễn giảm). +4. Bảo hiểm hưu trí đứt quãng không bị xoá số. Lương hưu tính theo số năm tích luỹ. Không cần bỏ tiền thuê người gắn tên đóng thay. Xem mục 18 ở phần 7 (Bảo hiểm xã hội đứt quãng thì đừng hoảng: lương hưu tính theo số năm tích luỹ, bảo hiểm y tế nộp bù theo quy định). +5. Thu nhập trong nhà rơi xuống dưới mức trợ cấp người có thu nhập thấp ở địa phương thì xin trợ cấp người có thu nhập thấp. Xem mục 7 ở phần 7 (Thu nhập thấp hơn mức trợ cấp người có thu nhập thấp ở địa phương thì xin trợ cấp người có thu nhập thấp). +6. Không chịu nổi mà muốn có người để nói chuyện thì gọi 12356, miễn phí. Xem mục 11 ở phần 29 (Muốn có người để nói chuyện thì gọi 12356). +7. Khoản tiền chưa đòi lại được thì canh mốc 1 năm sau khi rời việc. Xem mục 19 ở phần 8 (Bảo vệ quyền lợi có thời hạn: thời hiệu khởi kiện dân sự là 3 năm, hoà giải lao động là 1 năm).