Stockwell · Journal of studies on alcohol and drugs 2024 · systematic review and meta-analysis of longitudinal cohort studies · n=107 studies (4,838,825 participants)

Why Do Only Some Cohort Studies Find Health Benefits From Low-Volume Alcohol Use? A Systematic Review and Meta-Analysis of Study Characteristics That May Bias Mortality Risk Estimates.

Cited 29 times in the scientific literature.

Level 3 - non-randomized controlled study

Systematic review and meta-analysis of observational cohort studies

PubMed 38289182 · doi:10.15288/jsad.23-00283 · record verified 2026-08-29

What was done

The authors conducted a systematic review and meta-analysis of 107 longitudinal cohort studies (724 estimates, 4,838,825 participants, 425,564 deaths) evaluating low-volume alcohol use (1 drink/week [>1.30 g ethanol/day] to 2 drinks/day [<25 g ethanol/day]) and all-cause mortality. Higher-quality studies were defined as having a mean cohort age of 55 years or younger, follow-up past age 55, and excluding former and occasional drinkers from abstainer reference groups. Mixed linear regression modeled relative risks (RRs) across study subgroups and exploratory adjustments for smoking and socioeconomic status.

What was found

Higher-quality studies controlling for reference group contamination and cohort age found no significant mortality reduction for low-volume drinkers compared to abstainers (RR = 0.98, 95% CI [0.87, 1.11]). Studies not meeting these quality criteria reported a statistically significant lower risk of mortality for low-volume drinkers (RR = 0.84, 95% CI [0.79, 0.89]). Studies adjusting for smoking and/or socioeconomic status reported lower risk estimates, but analyses restricted to nonsmoking cohorts showed an RR of 1.16 (95% CI [0.91, 1.41]) for low-volume drinkers.

Why it matters

Health benefits frequently attributed to moderate drinking appear to be artifacts of lifetime selection bias, specifically classifying unhealthy former drinkers as abstainers. When these biases are eliminated, low-volume alcohol consumption provides no detectable survival advantage.

Limits

The underlying evidence is limited to observational cohort designs with self-reported alcohol intake, which remains susceptible to measurement error and recall bias. The abstract does not provide cause-specific mortality breakdowns (e.g., cardiovascular disease versus cancer) or detailed assessments of heavy episodic drinking patterns.

Cited by