Large-scale placebo controlled clinical trials assessing the efficacy of medications for the treatment of drug dependence have generally been limited to alcohol, cocaine and nicotine dependent populations. The purpose of the present study was to assess the early (1-2 week) clinical effectiveness of buprenorphine versus placebo in an opioid dependent population. The study used a parallel-group design with a behavioral choice component to compare buprenorphine (a mu-opioid partial agonist) to placebo for the treatment of opioid dependence. Opioid dependent volunteer patients participated in a 14-day study to assess the effectiveness and patient acceptance of this new pharmacotherapy for the treatment of opioid dependence. Patients were randomly assigned to placebo (n = 60) or 2 mg (n = 60) or 8 mg (n = 30) daily sublingual buprenorphine. All doses were administered double-blind. On days 6-13 all patients could request a dose change, knowing that their new dose would be randomly chosen from the remaining 2 alternatives. Compared to placebo, patients given buprenorphine (independent of dose) showed greater time on initial dose, requested fewer dose changes, used less illicit opioids (assessed by urinalysis), and rated dose adequacy higher. These results demonstrate that a placebo controlled study with a behavioral choice component is an effective means of assessing the potential efficacy and acceptability of new pharmacotherapies for opioid dependence.