Methods and definitions

Drug-related infectious diseases such as HIV and hepatitis B and C are among the most serious health consequences of drug use. Even in countries where HIV prevalence in injecting drug users (IDUs) is low other infectious diseases, such as hepatitis B/C, sexually transmitted diseases, TB, tetanus, botulism, hepatitis A, HTLV and other infections may dis-proportionally affect drug users. IDUs are the target group for measuring prevalence of drug-related infections. They are defined as any person who has ever in their lifetime injected a drug for non-medical purposes. In practice almost all data on IDUs collected by the EMCDDA relates to ‘ever injectors’ among active drug users who are in contact with drug services.

The EMCDDA is systematically monitoring HIV and hepatitis B and C among injecting drug users (prevalence of antibodies, or other specific markers in the case of hepatitis B). This is as a complement to existing notification and case reporting systems that follow trends in counts of cases. National notification data are often unreliable due to under-diagnosis, under-reporting, misclassification of injecting risk and the fact that large proportions are asymptomatic cases (hepatitis B/C). In addition, HIV case reporting has not been implemented in some of the countries most affected by AIDS while trends in HIV case reports depend on testing coverage and are not necessarily consistent with trends in measured sero-prevalence. Other infections may in the future be added to the EMCDDA monitoring system (e.g. sexually transmitted infections, tuberculosis) while a rapid alert system is being maintained to report outbreaks of serious infections such as tetanus and wound botulism.

To improve HIV and hepatitis B/C monitoring in IDUs the EMCDDA follows two lines of work:

Collecting existing prevalence data (HIV and hepatitis B/C) and notification data (hepatitis B/C only; HIV case reports are obtained from EuroHIV) in aggregate format using a standard data reporting form (‘standard table 9’).
Stimulating new sero-behavioural studies in injecting drug users and stimulating increased screening of IDUs and data collection in routine settings such as drug treatment, by maintaining an expert network to discuss methods and work towards common protocols.

The EMCDDA has developed draft guidelines for the national focal points to collect the existing prevalence and notification data and it is working on a toolkit or ‘framework protocol’ for primary data collection through sero-prevalence studies. This is based on a draft consensus protocol prepared by an expert network of longitudinal (cohort) studies.

To further improve the comparability of prevalence data in IDUs, data are collected and reported on prevalence of HIV and hepatitis in young IDUs (under age 25) and new IDUs (who have injected less than 2 years). These indicators, and especially the data for new IDUs, are more sensitive to changes in incidence than is prevalence in all IDUs. In practice the target group differs slightly between settings: sero-prevalence data from needle exchanges by definition refer to current injectors (defined as having injected in the last 12 months) while data from hepatitis notifications or public health laboratories may be partly based on ex-injectors, so additional methodological data such as service setting are also collected.

Following discussions during the annual meetings of the EU expert network, a new sheet was included in the standard table in order to collect information on key behavioural characteristics of the IDUs in the studies’ samples. Main items include HIV testing and risk behaviour (e.g. needles or paraphernalia sharing) and other variables related to the risk of contracting a blood-borne infectious disease, e.g. homelessness or sex work.

The aggregate prevalence data collection through the standard reporting form has been successful. In few years time a general overview could be given of HIV and hepatitis B/C prevalence among IDUs in all EU Member States, going back to 1996 and in part even before. Many countries are able to provide up to date data with national coverage and in many cases there is regional breakdown or data from key regions or cities, often unpublished and recent. These data have proven useful to provide a general overview of the situation, showing regional variation in levels and trends. Although in general they show a stable prevalence of HIV and hepatitis among IDUs, they served to signal some increases in HIV or hepatitis among subgroups of IDUs in some countries.

However, the data are subject to important limitations: the use of varying source-types/settings (drug treatment, low-threshold, prisons etc.) that may result in different biases, in some cases non-adherence to the basic case definition of ‘ever-IDUs’ that by inclusion of non-IDUs may lead to potentially serious downward bias, small sample sizes and other problems. Improving data quality and comparability proves difficult, as this depends on influencing often well-established data producing systems. Also, to get quality information on trends over time from routine diagnostic data (as opposed to well-defined prevalence studies) it is necessary to understand selection procedures for being tested, and if possible to work towards more standardisation in the criteria for screening IDUs in contact with services.

New methodological improvements in the field of data collection include these projects:

Developing a more detailed DRID protocol including guidelines for primary data collection and recommended core and optional items;
A pilot study to assess the feasibility of collecting prevalence data on HCV in young people from public health laboratories;
Understanding potential protective factors in countries with low prevalence of HIV (see also EU action plan on drugs 2005 to 2008);
Joint analysis of data from cohort studies with community-wide recruitment.

For more information see http://www.emcdda.europa.eu/?nnodeid=1375.