Back to this edition · Cancer risk, exposure & prevention
Did light drinking mean no increased cancer risk?
What the researchers did
The team combined prospective cohort studies, meaning they followed groups of people over time and looked at how much alcohol those people already drank. Cancer diagnoses in drinkers were then compared with diagnoses in non-drinkers or occasional drinkers. After a broad search, 139 cohort studies were reviewed and 106 had enough quantitative information to be included in the pooled analysis. The review covered several cancer types, including esophageal, colorectal, breast and prostate cancer.
What counted as light drinking
Drinking was divided into four bands. Light drinking meant 0.01-12.4 g/day, which is roughly up to one standard drink because the analysis used 12 grams per glass. This is the group behind the question of whether light drinking was harmless.
No clear rise overall, but specific cancers did rise
When all cancer types were pooled, light drinkers had a relative risk of 1.02 (95% confidence interval 0.99 to 1.04), which was not a clear increase. That overall result hid differences by cancer site. Light drinking was associated with a higher risk of esophageal cancer, colorectal cancer, breast cancer in women, and prostate cancer in men. Higher drinking categories produced larger risk increases and added more cancer types, confirming a dose-response relationship.
What the finding does and does not mean
These are observational studies, so the results show an association rather than proof that a small drink directly causes cancer. The comparison group included occasional drinkers, and the studies varied in how alcohol intake was measured. Even so, because light drinking was linked to several common cancers, the authors concluded that no safe level of alcohol consumption was demonstrated for cancer risk.
Source and scope
Cancer risk based on alcohol consumption levels: a comprehensive systematic review and meta-analysis.
Epidemiology and health · 2023-10-16
Based on the abstract and selected full-text passages; the rest of the full text was not reviewed.
Original source · DOI 10.4178/epih.e2023092 · PMID 37905315
Prepared: 2026-09-14T18:41:54.961845+00:00 · Version a12fd67c090debe3