One Drink a Day Is Not Only Not Good for You, It’s Bad for You

A drink a day has been thought to benefit cardiovascular risk and even overall survival, but recent systematic reviews focused on the most rigorous studies have raised serious doubts. Researchers performed individual-participant data analysis (n=599,912 people who currently drink, 71,011 with serial measurements) of 83 prospective cohort studies (5.4 million person-years of follow-up). Analyses were adjusted for age, sex, diabetes, and smoking.

• All-cause mortality increased linearly at >7 drinks/week.
• The risk for non-fatal myocardial infarction was 7% lower for each additional 7 drinks/week.
• >1 drink/week was associated with a linear increase in the risk for stroke.
• >1 drink/day was linearly associated with heart failure and death from other cardiovascular disease.
• The risk for other cardiovascular disease was 6% higher above 7 drinks/week. Approximately 6 drinks/week was associated with lower risk, but 5 was not associated with any protection or risk compared with lower amounts.
• Further adjustment, for cholesterol, fibrinogen, smoking amount, education, occupation, physical activity, general health, red meat consumption, and anti-hypertensive use did not change the results.
• Reducing drinking from 14 to <7 drinks/week would increase life expectancy by 1-2 years for a 40-year-old man.

Comments: This study highlights a challenge in using observational data to study disease prevention with a pharmacological substance—the margins between doses (averages of self-reports) having different effects are slim. Randomized trials of ethanol for cardiovascular prevention are unlikely to be conclusive since a number of precise dose levels would need to be administered for years. These results suggest that aside from the known cancer risks of low amounts of drinking, just above a drink a day increases risk for cardiovascular disease and death. Public health recommendations for risky drinking limits (currently 14 drinks/week for men in the US) should likely be adjusted accordingly.

Richard Saitz, MD, MPH

Reference: Wood AM, Kaptoge S, Butterworth AS et al. Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. 2018;391:1513–1523.