BACKGROUND AND AIMS: Alcohol consumption has complex effects on myocardial infarction (MI) and ischemic stroke. We investigated the difference in associations according to drinking patterns (drinking frequency vs. amount per occasion) and sex.
METHODS: This population-based retrospective study included 11,595,191 subjects participating in national health examinations between 2009 and 2010. Using Cox regression analyses, we calculated MI and ischemic stroke risk according to weekly alcohol consumption, drinking frequency, and amount per occasion.
RESULTS: For MI, all weekly alcohol consumption amounts showed lower risk compared to non-drinkers: mild (adjusted hazard ratio [aHR], 0.78; 95% confidence intervals [CI], 0.77-0.79), moderate (aHR, 0.71; 95% CI, 0.70-0.73), and heavy (aHR, 0.74; 95% CI, 0.72-0.76). Drinking frequency and amount per occasion did not differ in MI risk. However, women showed increased risk with heavy drinking and >/=8 drinks per occasion. For ischemic stroke, a J-shaped association was observed for weekly alcohol consumption: mild (aHR, 0.91; 95% CI, 0.90-0.92), moderate (aHR, 0.94; 95% CI, 0.93-0.96), and heavy (aHR, 1.04; 95% CI, 1.02-1.06). Among women, ischemic stroke risk began to increase with moderate drinking. Given similar weekly alcohol consumption levels, ischemic stroke risk increased with higher frequency of drinking, not with amount per occasion.
CONCLUSIONS: Drinking frequency may be a more important risk factor for ischemic stroke than amount per occasion. Among women, the protective effect of alcohol against MI was not evident in heavy amounts, and the risk of ischemic stroke began to increase at lower levels compared to men.