OBJECTIVE: To compare the incremental drug-related expenditures and the incremental number of years of life gained in a hypothetical heroin users population treated with Subutex? or with methadone. METHODS: A cost-effectiveness analysis is conducted, wherein the expected life-time drug costs and benefits of the two drug therapies are calculated using a Markov cohort similuation with four different health states: 1) heroin use, 2) maintenance treatment, 3) relapse to heroin, and 4) death. This analysis takes the viewpoint of the provincial drug plan, BC Pharmacare. Information on mortality rates and duration of treatment were obtained from a literature review and from administrative data from France. The model starts with a hypothetical cohort of 1,000 heroin users with potential access to methadone or Subutex? maintenance treatments on their twenty-fifth birthday. The results are provided in 2000 Canadian dollars. RESULTS: All costs and benefits are discounted by 3% in the base case analysis. By providing 1,000 heroin users with potential access to Subutex? and methadone maintenance treatments, the present value of the incremental benefit of treatment are 4,430 and 3,137 years of life gained, respectively. The Subutex? maintenance treatment is more expensive, with an annual drug treatment cost varying from $5,749 to $3,387 depending on the number of “take-home” privileges, compared with $3,158 for methadone maintenance treatment. The incremental cost of providing Subutex? maintenance instead of methadone is $8,927 for each additional year of life gained. In almost all cases, small changes are observed in the marginal cost-effectiveness ratio when values are varied within ranges of uncertainty. CONCLUSION: Introducing Subutex? maintenance treatment in British Colombia will attract a wider spectrum of opiate addicts into substitution treatments, will reduce the use of heroin, will decrease the spread of HIV and HCV, will reduce criminal activity and will decrease mortality rates.