Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies.
Level 3 - non-randomized controlled study
Individual-participant data meta-analysis of prospective observational cohort studies
PubMed 29676281 · doi:10.1016/S0140-6736(18)30134-X
What was done
Researchers conducted an individual-participant data meta-analysis pooling 83 prospective cohort studies from three large-scale data sources across 19 high-income countries: the Emerging Risk Factors Collaboration, EPIC-CVD, and the UK Biobank. The analysis included 599,912 current drinkers without baseline cardiovascular disease and with at least one year of follow-up. Investigators characterised dose-response associations and calculated hazard ratios per 100 g per week of alcohol intake, adjusted for study or centre, age, sex, smoking, and diabetes. Hazard ratios were corrected for long-term consumption variability using 152,640 serial alcohol assessments from 71,011 participants across 37 studies.
What was found
During 5.4 million person-years of follow-up, 40,310 deaths and 39,018 incident cardiovascular disease events were recorded. The threshold for lowest risk of all-cause mortality was at or below 100 g per week. Per 100 g per week higher alcohol intake, hazard ratios were 1.14 (95% CI, 1.10-1.17) for stroke, 1.06 (95% CI, 1.00-1.11) for coronary disease excluding myocardial infarction, 1.09 (95% CI, 1.03-1.15) for heart failure, 1.24 (95% CI, 1.15-1.33) for fatal hypertensive disease, and 1.15 (95% CI, 1.03-1.28) for fatal aortic aneurysm. Conversely, higher intake was associated with lower risk of myocardial infarction (HR 0.94, 95% CI, 0.91-0.97). Compared with drinking up to 100 g per week, consuming >100-200 g, >200-350 g, or >350 g per week was associated with approximately 6 months, 1-2 years, or 4-5 years shorter life expectancy at age 40, respectively.
Why it matters
The findings challenge guidelines that recommend higher upper drinking limits by demonstrating that consumption above 100 g per week is associated with increased all-cause mortality and higher risk for most cardiovascular subtypes.
Limits
The included studies were observational, which precludes causal conclusions. Alcohol consumption was measured by self-report and may be subject to reporting bias and misclassification. The analysis was restricted to current drinkers in high-income countries without baseline cardiovascular disease, which may limit generalizability to other populations. Unmeasured confounding from lifestyle or socioeconomic variables remains possible.
Cited by
- supports Consuming more than one alcoholic drink per day is detrimental to health.
- supports Any cardiovascular benefits associated with low alcohol intake are smaller than the associated increase in cancer risk, resulting in no net reduction in mortality.