Four hundred and twenty-three alcohol dependent subjects were enrolled into a 12-week randomized, double-blind, placebo-controlled study to determine the safety and efficacy of the 5-HT2 receptor antagonist, ritanserin (2.5 mg/day or 5 mg/day), in reducing alcohol intake and craving. All subjects received 1 week of single-blind placebo prior to randomization into the 11-week double-blind phase. Additionally, all subjects received weekly individual sessions of manual-guided cognitive-behavioral therapy. Comparing the single-blind period with endpoint, there was approximately a 23% reduction in drinks/day; 34% fall in the total number of drinking days/week; 22% decrease in drinks/drinking day; and a 37% diminution in alcohol craving for all treatment groups. All treatment groups experienced a beneficial clinical outcome as assessed by the Clinical Global Impression Scale. There was, however, no significant difference between treatment groups on any of these measures of alcohol drinking, craving, or clinical outcome. Subjects were of relatively high social functioning at baseline, and this did not change significantly during treatment. Treatment groups did not differ significantly on either medication compliance or reported adverse events. Ritanserin treatment was associated with a dose-related prolongation of subjects' QTc interval recording on the electrocardiogram. These results suggest that alcohol dependent subjects can show marked clinical improvement within a structured alcohol treatment program. These findings do not support an important role for ritanserin in the treatment of alcohol dependence.