証拠の質 4.5/5
8次元のレビュー評価。基準は 品質ルーブリック 。各次元は1〜5で評価。
- D1 出典への根拠
- 5/5
- D2 出典の権威性
- 5/5
- D3 算術
- 4/5
- D4 不確実性
- 4/5
- D5 範囲
- 5/5
- D6 文章
- 4/5
- D7 認識の誠実性
- 4/5
- D8 注意事項の完全性
- 5/5
● あなたの要因 — このリスクをクリック ▾ して表示
- あなたの要因
≈ 同じくらいの確率
認知リスク
アルコールは、文化的なフレーミングと数字が反対方向に進む稀な恐怖です。裕福な国のほとんどの成人は、日常的な飲酒を「理想的ではないが、それほど危険でもない」ものとして、死亡率のレバーというより食事上の悪習に近いものとして位置づけています。集計された計算はこれに反します。アルコールは世界で年間約2.6 millionの死亡、全死亡の約4.7%の原因であり、米国では年間約178,000件の死亡——米国の死亡の約20分の1——の原因です。1人あたりの死亡寄与は喫煙と同じオーダーですが、それに付随する社会的な恐怖はそうではありません。この項目は、飲酒者と非飲酒者を混ぜた人口平均ではなく、実際に定期的に飲酒する人の生涯帰属死亡率についてのものです。
概算: ほとんどの成人は大量飲酒が悪いと知っているが、生涯のアルコール死亡率を10分の1をはるかに下回ると推測する
出典: 編集部の直感(調査に基づかない)
実際のリスク
世界で年間約2.6 millionの死亡(全死亡の約4.7%)
世界、全年齢、アルコールに起因する疾患
計算を表示
参照サブグループ:米国の食事ガイドラインの閾値を超えて——男性は週14標準ドリンク超、女性は週7標準ドリンク超——長期の禁酒期間なく成人期の30年以上にわたって定期的に飲酒する成人。約15%という見出しは、このサブグループの生涯アルコール帰属死亡率の丸められた中間点であり、文献における方法論的な不一致を反映して約10%から約20%の間に括られています。基準:(1) WHOの2024年アルコールと健康に関する世界現状報告書は、世界で年間2.6 millionの死亡をアルコール消費に帰属させており(2019年データ)、全死亡の4.7%——男性死亡の6.7%、女性死亡の2.4%——に相当します。(2) CDC MMWR 2024年は、2020-2021年の過度のアルコール使用による米国の死亡を年間平均178,307件、米国の死亡の約20分の1と報告しており、2016-2017年の137,927件から29%増加しています。(3) 約380 millionの米国成人にわたり、大量飲酒者が飲酒人口の少数派でありながらアルコール帰属死亡の大多数を占めることを踏まえると、生涯にわたる大量飲酒者サブグループの年齢加重年間ハザードを40-50の成人期年にわたって累積すると、生涯帰属死亡率は10-20%の範囲になります。(4) 見出しの数字は喫煙の数字の約半分であり——これは、アルコール帰属死亡が喫煙と同様の集計規模を持つが曝露あたりのハザード比はより小さい(大量飲酒は全死因死亡を約1.5-2.5倍上昇させるのに対し、現在の喫煙は約3倍)という大まかな直感と一致します。スコープはsubgroup_lifetimeと宣言されています。これは一般人口の生涯リスクではなく生涯にわたる大量飲酒者1人あたりの確率であり、他のLikelierページの人口スコープの生涯数字と直接比較できないためです。regional_breakdownの中程度および軽度の飲酒者の行は見出しより低く、J字カーブに関する残りの方法論的論争を反映しています(ソース4および5を参照)。
注意事項: この項目は具体的に、成人期の大半にわたって大量飲酒の閾値で定期的に飲酒する人の生涯帰属死亡率であり、一般人口の平均ではありません。飲酒者と非飲酒者を平均した他の…
この項目は具体的に、成人期の大半にわたって大量飲酒の閾値で定期的に飲酒する人の生涯帰属死亡率であり、一般人口の平均ではありません。飲酒者と非飲酒者を平均した他のLikelierページの人口スコープの生涯数値(がん、心疾患、脳卒中)とは直接比較できません。サブグループの定義が重要です。世間の会話における「定期的な飲酒者」は、しばしばここで用いる週14ドリンク超(男性)/週7ドリンク超(女性)の閾値よりはるかに軽いものを意味します。軽度および中程度の飲酒者は、regional_breakdownの行に反映されているとおり、分布のはるかに低い部分に位置します。また、軽度の飲酒が保護的か、中立か、軽度に有害かについて、文献には活発な方法論的論争があります。GBD 2016年(Lancet 2018年)は、歴史的なJ字カーブは禁酒者の異質性と病的離脱者バイアスの産物であり、週ゼロドリンクが全死因の健康損失を最小化すると論じました。GBD 2020年(Lancet 2022年)はこれを部分的に撤回し、40歳超の成人については理論上のリスク最小曝露レベルは小さいがゼロではなく、地域によって異なることを示しました。このページの見出しの数字は、結果として生じる不確実性の幅の中間点であり、個人の予測としてではなく、大量飲酒者の死亡率のオーダー単位の較正として読むべきです。個人の結果は、強度、期間、飲酒パターン(継続的か集中的か)、年齢、性別、併存するタバコ使用、慢性ウイルス性肝炎の状態、および遺伝的・環境的修飾因子の長いリストに依存します。WHOの2.6MおよびCDCの178Kの集計値は方法論を越えて頑健です。不確実性が存在するのはサブグループごとの変換においてです。
リスクはどう変わるか
見出しの数値は大きく異なる状況の平均です。シナリオや文脈による確率の違いは以下の通りです:
6.7分の1 · 15%
Headline subgroup. Mid-point of a 10-20% band; alcohol-attributable mortality concentrated in liver disease, cancers, cardiovascular disease, and injuries.
33分の1 · 3.0%
Substantially reduced hazard vs heavy drinking but non-zero attributable mortality per GBD 2016; GBD 2020 suggests some age-dependent attenuation of this figure.
125分の1
Small attributable fraction; the historical 'J-curve' protective effect is disputed by GBD 2016 as an abstainer-heterogeneity artefact.
—
Zero alcohol-attributable mortality — this entry measures excess attributable risk only.
バーの長さと濃淡は、これらのシナリオを他のリスクとではなく、互いに比較して順位付けしたものです。正確な確率は各項目の横に表示されています。
関連するリスク
似たテーマの他のリスク — 関連する不安を探るために。
このリスクが関わる決断
このリスク次第で変わる選択。そのトレードオフをどう見るか。
比較対象を選択
The headline number for a lifelong heavy drinker sits somewhere around 1 in 7, bracketed between 1 in 10 and 1 in 5 depending on how you draw the subgroup. At the aggregate level the picture is less ambiguous: the World Health Organization’s 2024 Global Status Report attributes about 2.6 million deaths a year worldwide to alcohol consumption — 4.7% of all deaths, 6.7% of male deaths, 2.4% of female deaths. In the United States the CDC’s 2024 MMWR analysis puts the average at 178,307 deaths a year in 2020-2021, up 29% from 137,927 in 2016-2017, a figure that works out to roughly 1 in 20 US deaths and shortens each affected life by an average of 24 years, for a total of about 4 million years of potential life lost per year. That 24-year average life-lost-per-death figure is unusually large compared with other chronic-disease mortality on this site, because alcohol-attributable deaths are concentrated at younger ages than tobacco-attributable ones.
The interesting feature of this entry in the Likelier catalogue is the cultural asymmetry relative to smoking. US adult smoking rates have fallen by roughly 73% since 1965 (from 42.4% to 11.6% in 2022) under the weight of calibrated public fear: the warning labels, the tax regime, the indoor-smoking bans, and the near-universal adult intuition that smoking is dangerous. Alcohol use has not fallen in anything like the same way, despite an aggregate per-capita mortality contribution that sits in the same order of magnitude. The perceived/actual gap runs the opposite direction from smoking’s: people genuinely underestimate the per-capita mortality burden. This is why Likelier tags the entry underrated rather than calibrated. The hazard ratio per unit of exposure is smaller for alcohol than for tobacco — heavy drinking raises all-cause mortality by roughly 1.5 to 2.5x versus smoking’s roughly 3x — but the base population exposed is much larger, so the totals converge.
Where the headline doesn’t apply: almost everywhere outside the narrowly-defined heavy-drinker subgroup, and the definition of “regular” is doing a lot of work. The figure on this page is for someone consistently above the US dietary-guideline thresholds of 14 drinks per week for men and 7 per week for women, across most of adult life. Light and moderate drinkers sit on much lower parts of the distribution, and the shape of that lower tail is itself disputed. GBD 2016 (The Lancet, 2018) argued that the historical “J-curve” protective effect of light drinking was an artefact of abstainer heterogeneity — lifelong non-drinkers were being compared against a mixed reference group that included sick-quitters and former heavy drinkers, which made light drinking look protective relative to a denominator it wasn’t actually competing against. The paper’s headline conclusion was that the consumption level minimising all-cause health loss is zero standard drinks per week. GBD 2020 (The Lancet, 2022) partially walked that back by showing that for adults over 40 the theoretical minimum-risk exposure level is small but non-zero, and varies by region. Both findings coexist in the current literature; neither changes the 2.6M-per-year global aggregate, and neither changes the fact that the lifetime attributable mortality for someone who drinks heavily across decades is in the same order of magnitude as smoking.
関連する豆知識
根拠台帳
以下の各数値は各出典が報告した内容であり、引用した原文の抜粋と算出方法を記載しています。リンクをクリックして直接確認できます。
3/6 件の出典が引用元と一字一句一致することを独立して検証済み
-
[1] World Health Organization — Alcohol — fact sheet 検証済み
Alcohol — fact sheet- 統計値
Around 2.6 million deaths per year globally attributable to alcohol consumption (2019); 6.7% of all male deaths and 2.4% of all female deaths- 抜粋
“"Worldwide, around 2.6 million deaths were caused by alcohol consumption in 2019. [...] In 2019, alcohol use was responsible for 6.7% of all deaths among men and 2.4% of all deaths among women." ”
- 出典データ
- 2024-06-25
- アクセス日
- 2026-04-11 · アーカイブ版
- 検証
- グラウンディング監査の際、抜粋を引用元から独立して再取得し、原文と一字一句一致することを確認しました。
- 計算過程
- WHO’s 2.6 million figure is the canonical global headline. Across a global adult population of ~6 billion, that is ~0.43 per 1,000 adults per year averaged across the full population (drinkers plus abstainers). The per-regular-drinker rate is several times higher because abstainers and light drinkers account for a large share of the denominator but a small share of the attributable mortality. WHO notes the 2.6M figure is down from the ~3.0M figure reported in earlier releases, partly due to methodology changes and partly due to declining per-capita consumption in some regions. Used as the primary global anchor.
- 独立性
- WHO draws on the Global Information System on Alcohol and Health (GISAH) and the IHME Global Burden of Disease alcohol module. Partially overlapping with the GBD 2016 and GBD 2020 Lancet papers cited below.
-
[2] US Centers for Disease Control and Prevention — Facts About U.S. Deaths from Excessive Alcohol Use
Facts About U.S. Deaths from Excessive Alcohol Use- 統計値
About 178,000 US deaths per year from excessive alcohol use (2020-2021); shortened lives by an average of 24 years; ~4 million years of potential life lost per year- 抜粋
“"About 178,000 people die from excessive drinking each year. [...] This was a 29% increase from just a few years earlier (2016-2017), when there were an estimated 138,000 deaths per year. [...] Shortened the lives of those who died by an average of 24 years. [...] This resulted in a total of about 4 million years of potential life lost." ”
- 出典データ
- 2024-02-29
- アクセス日
- 2026-04-11 · アーカイブ版
- 計算過程
- CDC’s ~178,000 US deaths per year is the domestic headline. Across ~3.3 million total US deaths per year (2021), that is ~5.4% — roughly 1 in 20 US deaths. The 24-year average life-expectancy loss per death is notable: alcohol-attributable deaths are concentrated at younger ages than tobacco-attributable deaths, which is why the aggregate years-of-life-lost figure (~4 million per year) is disproportionately large relative to the headcount. Used as the domestic anchor and as the basis for the "~1 in 20 US deaths" plain-English framing.
- 独立性
- CDC derives the 178,000 figure from the ARDI (Alcohol-Related Disease Impact) application, which ultimately draws on the same 58 alcohol- attributable causes tracked in the CDC MMWR source below. Treat CDC and MMWR as partially dependent — MMWR is the primary analysis, the facts-stats page is its plain-language republication.
-
[3] CDC Morbidity and Mortality Weekly Report (Esser, Sherk, Liu, Naimi) — Deaths from Excessive Alcohol Use — United States, 2016-2021
Deaths from Excessive Alcohol Use — United States, 2016-2021- 統計値
Average annual deaths from excessive alcohol use rose 29.3% from 137,927 (2016-2017) to 178,307 (2020-2021); age-standardized rates rose from 38.1 to 47.6 per 100,000- 抜粋
“"Average annual number of deaths from excessive alcohol use [...] increased 29.3%, from 137,927 during 2016-2017 to 178,307 during 2020-2021. [...] age-standardized death rates increased from 38.1 per 100,000 population [...] to 47.6 during 2020-2021. [...] deaths from excessive alcohol use among males increased approximately 27%, from 94,362 per year to 119,606 [...] among females increased approximately 35%, from 43,565 per year to 58,701." ”
- 出典データ
- 2024-02-29
- アクセス日
- 2026-04-11 · アーカイブ版
- 計算過程
- The MMWR is the primary analysis behind the CDC headline. The 29% increase from 2016-2017 to 2020-2021 is large enough to be meaningful even after accounting for the COVID-era drinking-pattern shock. The male/female split (119,606 vs 58,701) drives the sex-stratified numbers in the assumptions block and the fact that men carry ~67% of the US alcohol-attributable mortality burden. The age-standardized 47.6 per 100,000 figure, applied only to the subset of US adults who drink regularly at heavy-drinker levels, anchors the per-subgroup lifetime figure used in the normalized block.
- 独立性
- Same methodology as the CDC alcohol facts page above; the two sources are the same underlying analysis presented at different levels of detail. Treat as one combined line of evidence, not as independent verification.
-
[4] The Lancet (GBD 2016 Alcohol Collaborators) — Alcohol use and burden for 195 countries and territories, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016 検証済み
Alcohol use and burden for 195 countries and territories, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016- 統計値
Alcohol accounted for 2.2% of age-standardized female deaths and 6.8% of age-standardized male deaths globally in 2016; the level of alcohol consumption that minimized harm across health outcomes was zero standard drinks per week- 抜粋
“"Alcohol use was the seventh leading risk factor for both deaths and DALYs in 2016, accounting for 2.2% (95% uncertainty interval [UI] 1.5-3.0) of age-standardised female deaths and 6.8% (5.8-8.0) of age-standardised male deaths. [...] The level of alcohol consumption that minimised harm across health outcomes was zero (95% UI 0.0-0.8) standard drinks per week." ”
- 出典データ
- 2018-09-22
- アクセス日
- 2026-04-11 · アーカイブ版
- 検証
- グラウンディング監査の際、抜粋を引用元から独立して再取得し、原文と一字一句一致することを確認しました。
- 計算過程
- GBD 2016 is the most cited peer-reviewed analysis arguing that the historical "J-curve" protective effect of light drinking was an artefact of abstainer heterogeneity and sick-quitter bias, and that the consumption level minimising all-cause health loss is zero. This is the methodological basis for the regional_breakdown "light drinker" row being non-zero rather than protective. The sex-stratified attributable fractions (2.2% F, 6.8% M) line up closely with the WHO 2019 figures (2.4% F, 6.7% M), providing independent cross-check on the global attributable share.
- 独立性
- GBD is the upstream source for the WHO fact sheet’s attributable-fraction figures; WHO republishes GBD alcohol module outputs. Treat as partially dependent on the WHO source above — they agree to one significant figure precisely because they share a modelling pipeline.
-
[5] The Lancet (GBD 2020 Alcohol Collaborators) — Population-level risks of alcohol consumption by amount, geography, age, sex, and year: a systematic analysis for the Global Burden of Disease Study 2020 検証済み
Population-level risks of alcohol consumption by amount, geography, age, sex, and year: a systematic analysis for the Global Burden of Disease Study 2020- 統計値
Among individuals aged 40 and older, the burden-weighted relative risk curve was J-shaped with a 2020 theoretical minimum-risk exposure level (TMREL) of 0.114-1.87 standard drinks per day; 59.1% of those consuming harmful amounts were aged 15-39- 抜粋
“"Among individuals aged 15-39 years in 2020, the TMREL varied between 0 (95% uncertainty interval 0-0) and 0.603 (0.400-1.00) standard drinks per day. [...] Among individuals aged 40 years and older, the burden-weighted relative risk curve was J-shaped for all regions, with a 2020 TMREL that ranged from 0.114 (0-0.403) to 1.87 (0.500-3.30) standard drinks per day. [...] Among individuals consuming harmful amounts of alcohol in 2020, 59.1% (54.3-65.4) were aged 15-39 years and 76.9% (73.0-81.3) were male." ”
- 出典データ
- 2022-07-14
- アクセス日
- 2026-04-11 · アーカイブ版
- 検証
- グラウンディング監査の際、抜粋を引用元から独立して再取得し、原文と一字一句一致することを確認しました。
- 計算過程
- GBD 2020 partially walks back the "zero is optimal" framing of GBD 2016 by showing an age-dependent TMREL — approximately zero for young adults, non-zero (and J-shaped) for older adults. This is the source for the methodological dispute flagged in the caveats and in the body text. Does not overturn the aggregate attributable-mortality numbers, only the claim that there is no safe level at any age. Used to motivate the uncertainty band on the moderate-drinker row of the regional_breakdown.
- 独立性
- Same GBD pipeline as GBD 2016 above and as the WHO attributable-fraction figures; treat as partially dependent. Included for the age-dependent TMREL finding, which is the most substantive methodological update in the literature since 2018.
-
[6] American Hospital Association (AHA) News, reporting CDC/NHIS data — CDC: Adult smoking hits 60-year low
CDC: Adult smoking hits 60-year low- 統計値
Cigarette smoking among US adults fell from 42.4% in 1965 to 11.6% in 2022, the lowest rate in 60 years- 抜粋
“"Cigarette smoking overall by U.S. adults decreased from 42.4% in 1965 to 11.6% in 2022." ”
- 出典データ
- 2025-03-06
- アクセス日
- 2026-07-03 · アーカイブ版
- 計算過程
- Grounds the body text's contrast between alcohol's flat public perception and smoking's decades-long decline. Not used in this entry's own probability calculation (which concerns alcohol- attributable mortality); cited only to support the smoking-rate comparison in the "cultural asymmetry" paragraph.
- 独立性
- Reports CDC/National Health Interview Survey smoking-prevalence data; independent of the WHO, CDC alcohol, and GBD sources used for this entry's headline alcohol figures.








