Alcohol volume and pattern have been previously demonstrated to predict cardiovascular disease (CVD) in these data. The objective of this study is to assess whether personality is a mediator in this relationship. Interview data from 1154 men and women aged 18–64 in Winnipeg, Canada were linked to health care utilization and mortality records, with 8 years of follow-up. Cox regressions were performed for men and women on time to first event for physician visits, hospitalizations, and deaths due to coronary heart disease (CHD), hypertension, or other CVD. Models linking drinking to CHD outcomes were unaffected by personality covariates. Models of hypertension and other CVD outcomes were significantly affected by inclusion of personality covariates, but the only significant individual personality covariate was in the effect on other CVD in men: Ego strength had a protective effect [hazard ratio (HR)=0.95]. Personality covariates as measured by the Eysenck Personality Questionnaire (EPQ) and Barron's Ego Strength Scale make only a modest contribution to statistical models of the relation between drinking and cardiovascular health.