An admission cohort of 296 Australian methadone maintenance patients was followed over 15 years. The relative risks of death in and out of maintenance were calculated for two age groups, 20-29 and 30-39 years. Heroin addicts in both age groups were one-quarter as likely to die while receiving methadone maintenance as addicts not in treatment. This is because they were significantly less likely to die by heroin overdose or suicide while in maintenance. Methadone maintenance had no measurable effect on the risk of death through nonheroin overdose, violence or trauma, or natural causes. A meta-analysis showed the reduction in overall mortality was consistent with the results of cohort studies conducted in the United States, Sweden, and Germany. The combined results of the five studies again indicated that methadone maintenance reduced addicts' risk of death to a quarter, RR 0.25 (95% CI 0.19 to 0.33).