More Evidence Raising Questions About Whether “Moderate” Drinking Has Any Benefit

Low amounts of alcohol are purported to have health benefits, but even if advantages exist there are many questions about dose, frequency, to whom they might pertain, and with what risks. To assess the association between alcohol consumption and mortality, investigators linked national mortality data to English population health survey data from over 30,000 people aged ≥ 50 who were followed for 7–10 years.

• Although analyses that compared those who drank with those who did not and analyses only adjusted for age showed a number of significant associations, there were no significant associations between weekly alcohol consumption and mortality for men ≥ 65 years, or women 50–64 years old, when compared with never drinking.
• Younger men drinking 9–11 US standard drinks* in a week had reduced mortality (hazard ratio [HR], 0.5).
• Older women drinking < 1–2 occasions in a month or < 9 drinks in a week had reduced mortality (HRs, 0.7–0.8).
• The above adjusted** analyses compared those who drank with never drinkers, and excluded former drinkers from the noncurrent drinking category. In analyses in which those who drank were compared with occasional drinkers (< 1-2 times in a month), there were no significant associations between alcohol consumption and mortality.

Comments:

These analyses detected associations between drinking and mortality for very specific bands of age and consumption that disappeared in more sophisticated analyses. Of note, most associations were in the beneficial direction even though they were not statistically significant, and there was likely limited power. Nonetheless, the differences in results seen when the reference group is non-drinking versus never drinking, and when the analyses are adjusted for some potential confounders, are dramatic, suggesting that the worse health among non-drinkers is not because they don’t drink but rather because of other negative exposures. This study contributes to the mounting evidence that the associations between drinking and health benefits seen in prior research are not causal.

Richard Saitz, MD, MPH

Reference:

Knott CS, Coombs N, Stamatakis E, Biddulph JP. All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts. BMJ. 2015;350:h384.