Recording acute poisoning deaths.

Recording deaths from acute poisoning/substance abuse is not straightforward. The International Classification of Diseases (ICD), used to code mortality statistics, is aimed towards recording the underlying cause of death such as suicide or drug dependence rather than gathering data on poisoning per se. Despite the inherent difficulties clear trends can be observed from the data available for England and Wales. There have been marked changes in the compounds featuring in suicidal poisoning in the last 35 years reflecting changes in the availability of poisons, notably carbon monoxide and prescription barbiturates. However, although the number of poisoning suicides has decreased in the recent years, suicides from other means have increased in males (suicides in 1999, 75% male), hence there has been little change in the annual total of suicides. There are also striking differences in drug abuse- and volatile substance abuse (VSA)-related deaths between males and females. Drug abuse-related fatal poisoning (83% male, 1979-1999, European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) definition which does not include suicide), largely attributable to heroin and methadone, increased markedly during the 1990s, with a sharp rise in deaths attributed to accidental poisoning, although deaths involving methadone are now declining. VSA-related deaths (90% male, 1971-1999, almost entirely accidental deaths), nowadays predominantly from abuse of fuel gases (liquefied petroleum gas, LPG) from, for example, cigarette lighter refills, have declined from a peak in the early 1990s and are now becoming manifest in an older age group. These two latter instances especially provide examples where ICD-derived fatal poisoning data are inadequate and a 'poisons oriented' approach to data collection and analysis is necessary.