ANALYZING PROPOSED TREATMENTS: The large body of literature on substitution treatments for drug abusers describes a variety of social settings and an extremely heterogeneous set of protocols. Establishing correlations between protocols and practical applications is thus a difficult task. METHADONE: The most widely studied substance is methadone. With methadone treatment, there is a decline in the amount of heroin used and in the number of injections. But there is no response with other drugs, leading to a real risk of increasing cocaine abuse. Methadone is used as a tool to decrease the risk of HIV although its impact is difficult to quantify. The most significant effect of methadone treatment is the social effect with a decrease in delinquency and in the number of drug-abuse related incarcerations. The consequences in terms of employment are less clear and vary depending on the social setting. The therapeutic window is very narrow with methadone and results are highly dependent on practical applications. A multidisciplinary and individualized approach is required. BUPHRENORPHINE: There has been less work on buphrenorphine. Like methadone, there is a dose/efficacy relationship. The results of comparative studies between the two agents are quite similar with a possible advantage for buphrenorphine because of the wider prescription spectrum and the possibility of longer intervals between administrations. EXTRAPOLATING MODELS: The implications of applying models developed in other countries to the social context in France still have to be analyzed on a scientific basis. New trial methodologies are needed to evaluate the social and psychological impact of individualized management.