We examined individual responses to cognitive-behavior therapy for insomnia in 51 persons with chronic pain to determine the rate of clinically significant change and to identify predictors of successful treatment response. Outcome measures consisted of the Pittsburgh Sleep Quality Index (PSQI) and diary measures of sleep latency and sleep continuity. Using reliable change indices, 57% of participants were statistically improved on the PSQI after 7 weeks of treatment, but only 18% were considered fully recovered from their sleep problems. No demographic variables predicted treatment response but persons who reliably improved on the PSQI had a lower sleep self-efficacy at baseline. Improvers showed a significant increase in sleep self-efficacy ratings and a decrease in self-reported levels of distress and pain-related disability. These results suggest that patients with insomnia secondary to chronic medical conditions can be helped with cognitive-behavior therapy, although most individuals continue to have mild or subthreshold sleep problems at posttreatment.