The aim of this study was to compare the efficacy, safety, and tolerability of sertraline and clomipramine in the treatment of obsessive-compulsive disorder (OCD). Outpatients with DSM-III-R defined OCD for 1 year or longer and scores of ≥ 20 on the YaleBrown Obsessive Compulsive Scale (Y-BOCS), ≥ 7 on the National Institute of Mental Health Global Obsessive-Compulsive Scale (NIMH-OC), ≥ 4 on the Clinical Global Impression Severity of Illness Scale (CGI-S) and ≤17 on the Hamilton Depression Scale (17 item HAMD) were randomized to sertraline (n = 86) or clomipramine (n = 82) once daily for 16 weeks. Initial daily doses of sertraline and clomipramine were 50 mg. After a minimum of 4 weeks, these doses could be increased by 50 mg increments every 2 weeks to a maximum of 200 mg daily if the response was thought inadequate. Efficacy was assessed at the end of 1, 2, 4, 6, 8, 12 and 16 weeks of therapy using the Y-BOCS, NIMH-OC, CGI-S, CGI Improvement Scale (CGI-I) and Clinical Anxiety Scale (CAS). One hundred sixty-eight patients were randomized and received at least one dose of double-blind medication; 86 received sertraline and 82 clomipramine. Mean final daily doses at final visit were clomipramine 90 mg (efficacy evaluable patients 101 mg, completers 110 mg), and sertraline 129 mg (efficacy evaluable patients 132 mg, completers 136 mg). Mean baseline Y-BOCS, NIMH-OC and CGI-S totals were 27.7, 10.1 and 5.5, respectively, for sertraline and 27.4, 9.9 and 5.5, respectively, for clomipramine. Sertraline demonstrated greater efficacy than clomipramine in the intent-to-treat patient group: mean baseline to final visit changes were 50.8% (Y-BOCS), 41.9% (NIMH-OC) and 37.7% (CGI-S) for sertraline and 42.9% (Y-BOCS), 33.8% (NIMH-OC) and 30.0% (CGI-S) for clomipramine (P < 0.05). The number of patients withdrawing because of adverse events was substantially greater for clomipramine (26%) than sertraline (11%) (P < 0.05). The most frequent adverse events for clomipramine were dry mouth (20%), anxiety (17%), constipation (16%), nausea (15%) and somnolence (11%), and for sertraline, diarrhea (12%) and nausea (12%). In this study, sertraline was more effective than clomipramine in the intent-to-treat analysis. The difference in efficacy between the treatments is almost wholly accounted for by a greater number of clomipramine withdrawals due to the poor patient acceptance of clomipramine. The superior tolerability of sertraline and the lower rate of premature treatment withdrawal relative to clomipramine may offer considerable quality of life and compliance benefits in the long-term management of a chronic disorder such as OCD.