P-0021 ● Active

“Moderate drinking has no net health benefit.”

Community stance
No votes yet
§
Meaning & context

For decades, cohort studies showed a "J-curve": light and moderate drinkers lived longer and had less heart disease than abstainers. That finding underpinned the idea that a glass of wine a day is good for you.

The J-curve is now widely attributed to a statistical artefact. Many "abstainers" in those studies were former drinkers who quit because of illness, and moderate drinkers tend to be healthier and wealthier in other ways. Studies that correct for this, and genetic studies that sidestep it, mostly find no protective effect, while alcohol's link to several cancers is clear even at low doses. The remaining dispute is whether any benefit survives for specific groups, such as older adults and heart disease.

Moderate drinkingContested
▶

Usually up to about one standard drink a day for women and two for men (around 10–20 g of alcohol per day), though definitions vary by country.

Net health benefit
▶

Lower overall mortality or disease burden than not drinking at all, after weighing any cardiovascular benefit against cancer and other risks.

Filter by type:
↕
Arguments
6 total
↑ Supporting3
Empirical

When studies are corrected for the "sick quitter" problem and other biases, the apparent longevity advantage of moderate drinkers disappears.

Zhao and colleagues' meta-analysis of 107 cohort studies (4.8 million participants) found no significant reduction in all-cause mortality for low-volume drinkers (under 25 g/day) compared with lifetime non-drinkers once study design biases were adjusted for; risk increased at higher intakes, sooner for women.

Zhao, Stockwell, Naimi, Churchill, Clay & Sherk (2023). Association Between Daily Alcohol Intake and Risk of All-Cause Mortality. JAMA Network Open, 6(3), e236185.

Limitation

Several authors have links to alcohol-control research and advocacy. Adjusting for bias involves modelling choices that can themselves shift results.

Causal

Genetic studies, which use inherited variants that make people drink less as a natural experiment, find no protective effect of alcohol on stroke or heart disease, only harm rising with intake.

In the China Kadoorie Biobank (around 500,000 adults), Millwood and colleagues used genotypes affecting alcohol metabolism and found that higher alcohol intake increased stroke risk roughly linearly, with no protective effect at moderate levels. Holmes and colleagues found carriers of an alcohol-reducing variant had lower coronary heart disease risk.

Millwood et al. (2019). Conventional and genetic evidence on alcohol and vascular disease aetiology. The Lancet, 393(10183), 1831–1842; Holmes et al. (2014). BMJ, 349, g4164.

Limitation

Mendelian randomisation assumes the genetic variants affect health only through alcohol, which may not hold. Most of the Chinese sample's drinking was spirits.

Empirical

Alcohol causes at least seven types of cancer, and risk rises from the first drink. Any cardiovascular benefit would have to outweigh this.

In January 2025 the US Surgeon General issued an advisory that alcohol is the third leading preventable cause of cancer in the US, causally linked to cancers of the breast, colorectum, oesophagus, liver, mouth, throat and larynx. The WHO stated in 2023 that no safe level of alcohol consumption has been identified for cancer risk.

U.S. Surgeon General (2025). Alcohol and Cancer Risk advisory; WHO Regional Office for Europe (2023). No level of alcohol consumption is safe for our health. The Lancet Public Health, 8(1), e6–e7.

Limitation

Absolute cancer risk from low intake is small; for women, one drink a day raises lifetime breast cancer risk by a few percentage points relative to baseline.

No arguments of this type on this side.
↓ Opposing3
Empirical

Moderate alcohol raises HDL cholesterol and reduces blood clotting, plausible mechanisms for protection against heart disease. Randomised feeding trials confirm these biomarker effects.

A meta-analysis of experimental studies found moderate alcohol intake raised HDL cholesterol and adiponectin and lowered fibrinogen, consistent with reduced coronary risk.

Brien, Ronksley, Turner, Mukamal & Ghali (2011). Effect of alcohol consumption on biological markers associated with risk of coronary heart disease. BMJ, 342, d636.

Limitation

Raising HDL through other means (e.g. drugs) has not reduced heart attacks in trials, so HDL is a weak surrogate.

Empirical

For older adults, whose main risk is heart disease rather than cancer, global analyses still find small net benefits of low drinking. The answer depends on age and where you live.

The Global Burden of Disease 2020 alcohol analysis found that for people aged 40 and over without other health risks, small amounts of alcohol (around one standard drink or less per day) may bring some health benefits, mainly lower cardiovascular disease, while young people gained nothing.

GBD 2020 Alcohol Collaborators (2022). Population-level risks of alcohol consumption by amount, geography, age, sex, and year. The Lancet, 400(10347), 185–235.

Limitation

The GBD's earlier 2018 analysis concluded that the safest level is zero; differences between the two show how model-dependent the estimates are.

Empirical

The cardiovascular association is large and consistent across many cohorts and countries, for heart disease specifically, not just overall mortality. That consistency is hard to explain by bias alone.

Ronksley and colleagues' meta-analysis of 84 prospective studies found light-to-moderate drinkers had about 25% lower coronary heart disease mortality than non-drinkers.

Ronksley, Brien, Turner, Mukamal & Ghali (2011). Association of alcohol consumption with selected cardiovascular disease outcomes. BMJ, 342, d671.

Limitation

Residual confounding by wealth, social engagement and health behaviours is hard to rule out in any observational study, and genetic studies point the other way.

No arguments of this type on this side.
≈
Areas of agreement
3 points
Empirical

Heavy drinking increases mortality and the risk of many diseases.

Empirical

Alcohol is a cause of several cancers, with risk rising with intake.

Empirical

Early studies overstated benefits by including former drinkers who quit due to illness among abstainers.

P-0021 · Last edited Oct 26, 2026 · 1 revision