Trends in acute drug-related deaths

Trends in drug-related deaths vary from country to country, and even between regions or cities within a country. With these limitations in mind, some general trends can be outlined for the EU countries. A marked increase in the number of drug-related deaths was observed during the 1980s and early 1990s (Figure 15a). During the period 1990–2000 the overall increasing trend continued, albeit at a lower rate. In 2000, 8 731 acute drug deaths were reported throughout the EU, compared with 6 394 in 1990 (an increase of 36.5 %) (Figure 15b) (31).

Figure 15

Trends in acute drug-related deaths in the EU, 1985–2000


A few countries did not provide data for some years (see Statistical Table 24: Number of ‘acute drug-related deaths’ recorded in EU countries (according to national definitions used to report cases to the EMCDDA), 1985–2001). To correct this situation, the computation method defined in Project CT99.RTX.04, EMCDDA (2001) was used.

In all, ten countries provided information for 2001 and six did not. The index for 2001 has not been computed. The trends for those countries that provided information can be seen in Figure 15 OL: Trends in acute drug-related deaths among EU countries, 1985–2001. Numbers of cases per country per year are presented in Statistical Table 24.

Sources: Reitox national reports 2002, taken from general mortality registries or special registries (forensic or police). Based on national definitions as presented in Box 9 OL: Definitions of ‘acute drug-related death’ in EU Member States, as used in the EMCDDA annual report and reported in national reports.


This overall trend may reflect the rapid expansion of heroin injection in many European countries during the 1980s and early 1990s. Recent trends might be related to the apparent stabilisation of the prevalence of problem drug use in some countries, a decrease in injection in others and, probably, the increase of treatment availability – including substitution programmes. On the other hand, the increasing age of problem drug users and polydrug use may contribute to an increase in the number of fatalities.

It should be emphasised that within the EU different, and sometimes divergent, national trends exist (Figure 15 OL). Many EU countries continue to report increases in drug deaths up to the present or very recently, while some report stabilisations or decreases. Changes in reporting procedures and reporting quality should always be taken into account when interpreting national or local trends.

Several countries reported a stabilisation or a downward trend in the number of acute drug deaths during the second half of the 1990s. France and Spain showed an apparently consistent decreasing trend. Although these findings may be subject to the limitations of under-reporting (in the case of France) or limited coverage (only five big cities in Spain), it is important to investigate further the reasons for these trends. Changes in the route of heroin administration and a substantial expansion of treatment programmes (especially substitution treatment) might be influencing these trends (32). In other countries (Germany, Italy and Austria), the number of drug deaths has fluctuated in recent years and, thus, although the number of drug deaths appears to have stabilised, albeit at a high level, there is no guarantee that a decrease for two or three years will not be followed by an increase.

Still other countries have reported substantial upward trends in the second half of the 1990s (e.g. Greece, Ireland, Portugal, Finland and Norway); in some cases these may be partly due to improved reporting. In Portugal, this trend was reversed in 2000, and the same appears to have occurred in Finland in 2001. These increases might be related to the later expansion of heroin use in these countries, although improved reporting systems may have played a role in some cases.

The United Kingdom exhibited a steady, although moderate, increasing trend in acute drug deaths until 2000 according to the traditional definition (which includes some medicine-related cases; Office for National Statistics), but if heroin–morphine cases are considered independently a much steeper upward trend is evident: a fivefold increase in 2000 compared with 1993 in England and Wales (Griffiths, 2003) (33).

Finally, in some countries trends are less clearly defined, or changes are difficult to interpret because of the switch between ICD classifications (from 9th to 10th edition), changes in national definitions or the small number of cases involved.

The fact that there are no indications of an overall decrease at EU level emphasises the fact that drug-related health damage is far from being overcome, and stresses the need for further investigation of risk factors for drug-related deaths as well as appropriate targeted interventions.

(31)  Note that numbers of and trends in drug-related deaths for the EU as a whole are available only up to 2000. Nine countries reported 2001 data, but for seven countries only 2000 data were available at the time of data collection (autumn 2002). Whenever possible, 2001 information was used for countries that reported it.

(32)  See EMCDDA annual report (2002a, p. 17), Figure 10, ‘Route of administration of heroin in Spain, first treatments 1991–2000’ and the section ‘Treatment’ and Table 3 of this report: Development of substitution treatment in the 15 EU Member States and Norway.

(33)  The recently developed ‘Drug strategy definition’ focuses on drugs of abuse (drugs controlled under the Misuse of Drugs Act of 1971 but not indirectly related deaths such as deaths from AIDS) and is relatively similar to the EMCDDA definition. Using the drug strategy definition, the number of reported cases almost doubled in England and Wales between 1993 (864) and 2001 (1 623). The application of the EMCDDA definition produces 1 606 cases in England and Wales in 2000 and 1 443 in 2001.