We studied the effects of a fixed dose of fluoxetine (20 mg) or cognitive psychotherapy in a 16 week trial of patients with dysthymic disorder. More patients assigned to fluoxetine dropped out of the 16 week treatment (33%) than those assigned to cognitive therapy (9%), but this difference did not attain statistical significance. Both treatments showed improvement over baseline conditions at 8 weeks and further improvement at 16 weeks. There were no statistically significant group differences in treatment response. No follow-up data were collected so the enduring effects of the treatments are unknown. An optimal treatment for dysthymic disorder may be combined psychotherapy and pharmacotherapy for a longer period of time.