“Moderate” Drinking May Not Reduce Mortality Risk

Some believe that drinking small amounts of alcohol can decrease mortality. But the beneficial effects, if any, seem to peak at < 1 drink a day, and are based on studies with serious methodological challenges, not the least of which is the association between “moderate” drinking and numerous healthy characteristics not caused by alcohol (like past cancer screening or educational achievement). Even more important is the limitation that studies do not consider people who die (e.g., from drinking-related injuries) or stop drinking before a middle- or older-aged cohort is recruited. Scientists are beginning to question whether the observed associations between lower levels of alcohol consumption and decreased mortality are causal.
To address these serious limitations in prior studies, investigators performed a systematic review and meta-analysis of 87 studies (including data on 3,998,626 people) that reported associations between alcohol consumption and all-cause mortality.

The relative risk for all-cause mortality when compared with lifetime abstinence was:

• 0.86 for low-volume drinking (average 1.3 – < 25g/day) in unadjusted analyses.
• 0.89 for low-volume and 0.86 for occasional drinking (average < 1.3g/day, not biologically plausibly associated) in analyses adjusted for selected covariates.
• 0.97 (95% confidence interval, 0.88–1.07) for low-volume drinking in analyses adjusted for age, sex, race, smoking, drinking measure adequacy, study biases, exclusion of ill participants, and follow-up.

Comments:

This analysis does not address whether alcohol is associated with surrogate markers of disease (although an oft-cited review* of heart disease biomarkers found 9 of 13 studied were not). This analysis also did not review whether alcohol is associated with specific causes of death such as injury, cirrhosis, cancer, or heart disease. Nonetheless, many people care about their risk of death regardless of cause and this study addressed major methodological limitations of prior reports. When adjusting for study flaws there appears to be no protective effect of low-volume drinking on mortality. Although the usual response to observational studies of “moderate” drinking is to call for randomized trials, these results suggest a lack of evidence for an effect and could raise ethical questions about doing so.

Richard Saitz, MD, MPH

Reference:

Stockwell T, Zhao J, Panwar S, et al. Do “moderate” drinkers have reduced mortality risk? A systematic review and meta-analysis of alcohol consumption and all-cause mortality. J Stud Alcohol Drugs. 2016;77(2):185–198.