This study compared the efficacy of buprenorphine to methadone for decreasing cocaine use in patients with combined opioid and cocaine use. Participants (n = 51) were enrolled in a 26-week treatment program and randomly assigned to either buprenorphine or methadone. Dosing was double-blind and double-dummy. Patients were stabilized on either 8 mg sublingual buprenorphine or 50 mg oral methadone, with dose increases given in response to continued illicit cocaine use or opioid use through week 16 of treatment. Maximum doses possible were 16 mg buprenorphine and 90 mg methadone. Average doses achieved were 11.2 mg buprenorphine and 66.6 mg methadone; 49% of the patients received the maximum doses possible. Urine samples were collected three times per week, and there was no significant difference in the rate of cocaine positive urines for the intent-to-treat sample (69% for buprenorphine versus 63% for methadone). For patients who remained in treatment through the flexible dosing period (n = 28), there were significant decreases in cocaine positive urines over time (P < 0.01), but no significant differences between groups or group x time effects. Buprenorphine and methadone were equally effective on measures of treatment retention, urine results for opioids, and compliance with attendance and counseling. These results demonstrate no selective efficacy of either buprenorphine or methadone in attenuating cocaine use in this population, but do provide further support for the equivalent efficacy of buprenorphine and methadone in the treatment of opioid dependence.