Effect of drinking on all-cause mortality in women compared with men: a meta-analysis.

BACKGROUND: Alcoholic beverages are consumed by humans for a variety of dietary, recreational, and other reasons. It is uncertain whether the drinking effect on risk of all-cause mortality is different between women and men. We conducted a meta-analysis to evaluate the effect of drinking on the risk of all-cause mortality in women compared with men.

METHODS: We selected cohort studies with measures of relative risk (RR) and 95% confidence interval (CI) for all-cause mortality for drinkers versus nondrinkers by sex. Sex-specific RR and 95% CI were used to estimate the female-to-male ratio of RR (RRR) and 95% CI. Pooled estimates of RRR across studies were obtained by the fixed-effects model or the random-effects model (if heterogeneity was detected). Second-order fractional polynomials and random effects meta-regression models were used for modeling the dose-risk relationship.

RESULTS: Twenty-four studies were considered eligible. A total of 2,424,964 participants (male: 1,473,899; female: 951,065) were enrolled and 123,878 deaths (male: 76,362; female: 47,516) were observed. Compared with nondrinkers, the pooled female-to-male RRR for drinkers was 1.07 (95% CI: 1.02, 1.12). Subgroup analyses showed that the increased risk among female drinkers appeared to be consistent. J-shaped dose-response relationship was confirmed between alcohol and all-cause mortality in men and women, respectively. Moreover, the female-to-male RRR of all-cause mortality were 1.52 (95% CI: 1.01, 2.29), 1.95 (95% CI: 1.08, 3.49), and 2.36 (95% CI: 1.15, 4.88), respectively, for those who consumed 75, 90, and 100 g/day of alcohol.

CONCLUSIONS: Females had an increased risk for all-cause mortality conferred by drinking compared with males, especially in heavy drinkers. The present study suggested that female drinkers, particularly heavy drinkers, should moderate or completely reduce their level of consumption to have a health benefit.