A total of 551 participants were randomized to treatment in three heroin-dependence treatment trials participating in the Australian National Evaluation of Pharmacotherapies for Opioid Dependence (NEPOD) project. A total of 272 patients (49%) received methadone maintenance, 238 (43%) received buprenorphine maintenance and 41 (7%) participants received levo-α-acetyl-methadol (LAAM). A total of 63% of participants in the methadone maintenance group were in treatment in the third month, with an average treatment episode lasting 69 days. This compares with 51% of participants in the buprenorphine maintenance group with an average treatment episode of 60 days and 71% of participants in the LAAM group with an average treatment episode of 75 days. The results of the cost-effectiveness analysis suggested that, for the primary outcome measure of imputed change in heroin-free days, compared with methadone maintenance, LAAM was the most cost-effective treatment, followed by buprenorphine maintenance. No statistically significant differences were found in the cost-effectiveness of methadone maintenance, buprenorphine maintenance and LAAM. Given the limited information available regarding the relative cost-effectiveness of pharmacotherapies for opioid maintenance treatment, the data reported herein provide valuable information to policy makers and treatment providers.