Drug-related infectious diseases (15)

Prevalence and trends in HIV and AIDS

Infectious diseases related to injecting drug use have a major impact on the economic and social costs of drug use, even in countries where HIV prevalence is low (Postma et al., 2001; Godfrey et al., 2002). They can be prevented by providing injecting drug users (IDUs) with information about disease transmission, clean needles (Hurley et al., 1997; Commonwealth of Australia, 2002) and vaccination against hepatitis A and B viruses. Effective treatment for HIV and HCV infections is now available and, as the prevalence of infections is often very high among IDUs, policies aimed at prevention and treatment of infections are highly cost-effective.

The EMCDDA is systematically monitoring the prevalence of HIV infection and hepatitis B and C among IDUs in the European Union. Information on aggregate prevalence (overall and within subgroups) is collected from different routine settings (e.g. drug-treatment or needle-exchange programmes, prisons) as well as from special studies (16). Although the data are difficult to compare because they come from a variety of sources, they provide an overall impression of differences between countries, regions and settings. More importantly, following trends over time, especially prevalence in young and in new injectors, provides crucial information on the spread of infections among IDUs and the success of preventative policy measures. The longer-term aim is to improve data quality and comparability from existing routine sources and to set up truly comparable local European seroprevalence studies among IDUs.

Figure 11

Prevalence of HIV infection among IDUs in the EU Member States and Norway

Figure 11


Figures in brackets are local data. Differences between countries should be interpreted with caution because of the different source types and use, in some cases, of local or self-reported data. The colour for each country indicates the midpoint of the range of prevalence estimates obtained from different data sources.

This summary map is meant to give a global overview of HIV prevalence among IDUs in the EU. In this map data are reported for the most recent year available. Data from samples with no information on IDU status were excluded. If this led to exclusion of sources that clearly improve generalisability (e.g. national data, out-of-treatment data) data from more than one year were combined. Data for Italy, Portugal and Norway are limited to HIV prevalence among IDUs in treatment and are not representative of HIV prevalence among IDUs who are not in treatment. Having health problems is one selection criterion for admission to drug treatment in some countries or cities (Greece, Portugal, Rome); because of long waiting lists or special programmes for infected IDUs, this may result in upward bias of prevalence. Prevalence in this map should not be compared with previous versions to follow changes over time, as inclusion of sources may vary according to data availability. For time trends, methodological detail and for sources see Statistical Table 12: Prevalence of HIV infection (percentage infected) among injecting drug users in the EU and Norway; and Box 6 OL: Data sources – prevalence.

Source: Reitox national focal points. For full details and primary sources see Statistical Table 10: Summary table of prevalence of HIV infection among injecting drug users in the EU; and Statistical Table 12.


The data that are available suggest that the prevalence of HIV infection among IDUs varies greatly between, as well as within, countries. The levels of infection reported by different sources vary from about 1 % in the United Kingdom (surveys and unlinked anonymous screening) to over 30 % in Spain (routine diagnostic tests in drug treatment), but they are, in general, stable (17). This overall picture has not changed in recent years (Figure 11).

In some countries and regions, HIV prevalence has remained extremely high among IDUs since 1995. Although in most cases this reflects old epidemics, special prevention efforts are very important (e.g. efforts to prevent transmission to new IDUs, to sexual partners of IDUs and from mother to child). Prevalence was over 25 % in some regions and cities (18).

Trends in HIV prevalence provide important information for making and evaluating policy. More action is needed if trends suggest that levels of infection are increasing but may be unnecessary if infection levels appear to be declining. However, even in areas where prevalence is stable or decreasing, new infections may still occur. In recent years, increases in HIV transmission in (subgroups of) IDUs may have occurred in regions or cities in Spain, Ireland, Italy, the Netherlands, Austria, Portugal, Finland and the United Kingdom, although in some of these countries large decreases were also recorded (19). The Italian data demonstrate that in countries with significant numbers of infections national averages are of limited value, and breakdowns by smaller regions or cities are important to evaluate the success of prevention (Figure 11 OL). However, few countries can yet provide national data broken down by region. To facilitate the detection of trends over time, prevalence data should, ideally, be supplemented by notifications of newly diagnosed cases. Although not yet available for the countries with the highest prevalences, and still highly dependent on testing patterns, in Finland, notification data have helped reveal new increases in transmission rates. Recently, Portugal has started to provide HIV notification data, revealing much higher rates per million population than the other reporting countries (20).

HIV prevalence among young IDUs may provide further feedback on the effectiveness of prevention measures, as infection in young people will typically be more recent than in the IDU population as a whole. Although sample sizes are small, these data suggest that infections have occurred in young IDUs in several regions in recent years (21).

For some countries, information is available on HIV prevalence among new injectors. This is a much better indicator of recent HIV infections and may reflect incidence of HIV infection, thus providing stronger evidence for the effectiveness of prevention measures. Assuming that injectors who have been injecting for less than two years have, on average, been injecting for one year, prevalence in that group may provide an estimate of incidence. The available data suggest that incidence per 100 person-years of exposure among new injectors (95 % confidence intervals) may vary from 0–3.7 in England and Wales (0/122, 1998), through 0.8–11.4 in the Belgian Flemish Community (3/77, 1998–99) and 4.4–14.5 in Coimbra, Portugal (12/127, 1999–2000), to 4.4–15.5 in France (11/111, self-reported serostatus in needle-exchange scheme attendees, 1998) (22). However, the data from Belgium, France and Portugal are from routine testing and may be affected by selection bias. (The data for Portugal may be an overestimation because of the selection criteria of detoxification units, which give priority to problematic and/or seropositive drug users.)

AIDS data provide little information or following trends in new HIV infections because of the long incubation time of HIV before onset of symptoms of AIDS and the major improvements in HIV treatment that delay the onset of AIDS. However, AIDS incidence does reflect trends in the burden of disease from HIV infection in the different countries. The incidence of AIDS has fallen in most countries since about 1996 as a result of improved treatment of HIV infection and possibly lower infection rates in the 1990s. In Portugal, latest data show that the incidence of AIDS among IDUs has been decreasing since 1999 (Figure 12 OL). This may indicate increased uptake of HIV treatment consistent with reports from the drug-treatment system.

(15) A more detailed insight into this issue is provided in the 2001 Annual report, Chapter 3, Selected issues – Drug-related infectious diseases.

(16) See the EMCDDA web site for more detail on methods and guidelines.

(17) For more detail on these data and for original sources, see Statistical Table 10: Summary table of prevalence of HIV infection among injecting drug users in the EU; and Statistical Table 12: Prevalence of HIV infection (percentage infected) among injecting drug users in the EU and Norway.

(18)  See Box 4 OL: Areas with high HIV prevalence, increases and decreases in HIV transmission among IDUs in some EU countries, HIV prevalence among young IDUs.

(19) See Box 4 OL.

(20) See Statistical Table 15: HIV infections newly diagnosed among injecting drug users in the EU and Norway.

(21) See Box 4 OL.

(22) See Statistical Table 15: HIV infections newly diagnosed among injecting drug users in the EU and Norway.