Alcohol — What the Global Evidence Actually Shows
The largest study of its kind, and what it changed
Alcohol is the substance where the gap between popular perception and the underlying evidence is widest — worth covering carefully rather than repeating either the old "a glass of red wine is good for your heart" narrative or a vague "drink in moderation" line.
The Global Burden of Disease Findings
A systematic analysis covering 195 countries and territories from 1990 to 2016, drawing on 694 data sources on consumption and 592 studies on risk, found that alcohol use caused close to 3 million deaths globally in 2016 alone, including roughly 12% of deaths among men aged 15-49[2]. Critically, the modelled risk curve showed health risk rising from the first drink, with no protective effect at low levels of consumption once all outcomes (not just cardiovascular ones) were accounted for — directly contradicting the popular idea that light-to-moderate drinking carries a net health benefit.
Mechanism
Alcohol functions as a neurotoxin. It's officially classified as a Group 1 carcinogen, meaning there's strong, sufficient evidence that it causes cancer in humans — that classification is about the strength of the evidence, not a ranking of how dangerous it is compared to other substances in the same tier, which also includes tobacco smoke and asbestos. In practice, alcohol is linked to breast cancer, colorectal cancer, reduced grey matter volume, and liver damage even at moderate intake levels. It also measurably disrupts sleep: deep and REM sleep are fragmented even at low doses, and it takes 3-5 days after a single drinking session for sleep architecture to fully recover — meaning a few drinks on a Friday can still be costing you sleep quality by Monday, even if you feel fine.
Alcohol also suppresses testosterone and raises oestrogen, by increasing activity of an enzyme called aromatase that converts one into the other. That has downstream effects on muscle protein synthesis — relevant if you're drinking regularly while also trying to build or maintain muscle, since the hormonal shift works against that goal.
Avoid drinking within 3 hours of sleep where possible, given how reliably alcohol fragments deep and REM sleep even at low doses.
Signs Worth Paying Attention To
Population risk curves describe averages; noticing a pattern in your own drinking is a separate, more personal question. A few signs worth taking seriously, drawn from the indicators used in standard clinical screening tools:
Needing more to feel the same effect than you used to — a classic tolerance signal, and often the earliest one to show up.
Regularly drinking more, or for longer, than you intended to once you started.
Morning anxiety, shakiness, or a racing heart the day after drinking, even from a moderate amount — a sign your nervous system is rebounding from alcohol's effects overnight.
Wanting to cut back but finding it hard to actually do so — the gap between intention and behaviour is itself a meaningful signal, separate from how much you're drinking in absolute terms.
None of these alone is a diagnosis, but two or more together are worth raising with a doctor.
Section takeaway
The largest alcohol evidence review ever conducted found no level of drinking that improves health once the full range of outcomes is considered — a materially different conclusion from the older, cardiovascular-only "moderate drinking is fine" narrative. Alongside the population data, needing more to feel the same effect, drinking more than intended, morning after-effects, and difficulty cutting back are the personal warning signs worth noticing.