Few studies have evaluated efficacy of psychotherapies for pathological gambling. Pathological gamblers (N = 231) were randomly assigned to (a) referral to Gamblers Anonymous (GA), (b) GA referral plus a cognitive-behavioral (CB) workbook, or (c) GA referral plus 8 sessions of individual CB therapy. Gambling and related problems were assessed at baseline, 1 month later, posttreatment, and at 6-and 12-month follow-ups. CB treatment reduced gambling relative to GA referral alone during the treatment period and resulted in clinically significant improvements, with some effects maintained throughout follow-up ( ps = 0.05). Individual CB therapy improved some outcomes compared with the CB workbook. Attendance at GA and number of CB therapy sessions or workbook exercises completed were associated with gambling abstinence. These data suggest the efficacy of this CB therapy approach.