A multiple baseline across subjects design was used to test the efficacy of cognitive therapy for obsessive-compulsive disorder (OCD). Six people with OCD received 10-18 sessions of weekly, outpatient cognitive therapy. Assessment included both structured interviews and diary data. For three to four of the six patients, stable baseline periods were followed by reductions of symptoms during intervention. Two clients met stringent criteria for Jacobson and Truax's (J. Consulting Clin. Psychol. 59 (1991) 12) recovered status at posttest according to the Yale-Brown Obsessive-Compulsive Scale (Arch. Gen. Psychiatry 46 (1989) 1006). For the group, large pretest-posttest effect sizes were found.