To examine the differential effects of citalopram on alcohol consumption in nondepressed women and men with mild to moderate alcohol dependence. Prospective, placebo-controlled study. Sixty-one subjects (34 men and 27 women). After a 2-week baseline, subjects were randomly assigned to 12 weeks of citalopram (40 mg per day) (n = 15 women, 16 men) or placebo (n = 12 women, 18 men). All received brief standard psychosocial interventions. Alcohol Dependence Scale, Montgomery-Asberg Depression Scale, Michigan Alcohol Screening Test, State-Trait Anxiety Inventory and daily alcohol intake. Pretreatment sex differences were evident in alcohol consumption, alcohol dependence, alcohol-related problems and on anxiety and depression measures. After treatment, analyses of covariance with depression and anxiety scores as covariates revealed a differential benefit of citalopram for men. Men receiving citalopram reduced average drinks per day by 44%, whereas women exhibited a 27% decrease (p < 0.05). Men may benefit more than women from citalopram in the treatment of alcohol dependence. These findings highlight the importance of examining sex as a significant variable in evaluating response to pharmacotherapy.