Study objective-Although the total number ofselfpoisonings in England and Wales has dropped by 32%, the number involving methadone and/or heroin rose by 900% in 1974-92. Because ofconcern about the role of methadone in this increase, the part played by methadone and heroin in poisoning deaths in England and Wales in 1974-92 was investigated. Design-A proportional mortality design was used to study whether the ratio between deaths involving methadone or heroin and other substances had increased. Opiate addiction is a strong suicide risk factor,' and opiate addicts are also at increased risk of accidental death, principally through overdose.2 Substitution with methadone has become one of the mainstays of the treatment of opiate addiction and the proportion of opiate addicts using methadone has risen sharply over the past decades. ' Clinical research suggests that methadone maintenance reduces mortality (including overdose mortality) among clients.4 However, like heroin, methadone is dangerous in overdose.5 Marks6 and Newcombe7 have suggested, on the basis of analyses of cause of death returns and estimated numbers of addicts taking methadone and heroin respectively, that in the UK as a whole methadone in overdose kills proportionately more people than heroin. A limitation ofthis finding is that it depends crucially on unproven assumptions concerning the number of methadone and heroin users at risk in the population. Nevertheless, these authors suggest that "the current vogue for methadone in the management of addiction should be reviewed".6If the risk of death through overdose were indeed higher among addicts taking methadone compared with those taking heroin, and given the relatively rapid increase of methadone use in the addict population,' the proportion of overdose deaths involving methadone would rise at a faster pace than that involving heroin. We examine the contribution of methadone and heroin overdoses to total mortality from accidental and intentional self poisoning in England and Wales between 1974 and 1992. The proportional mortality design used eliminates the need to make unsafe assumptions concerning the numbers of addicts in the population taking methadone or heroin. (1982 not available).8 Hand-searching rather than relying on aggregate figures published by the OPCS8 or the Home Office9 was necessary in order to identify deaths which had involved heroin without methadone and deaths which had involved both substances simultaneously. Deaths were divided into three categories according to the substances involved as follows: * Methadone with or without heroin regardless of other substances, * Heroin without methadone regardless of other substances, * Deaths not involving methadone or heroin.The time period was divided into six successive three year spans. In the main tabulation (table 1) the three classes of death (accident, suicide, and undetermined) were combined and ratios were calculated, firstly, of deaths involving methadone with or without heroin versus all other deaths, and, secondly...