Estimating the size of the HIV epidemic among injecting drug users in Amsterdam.

Aim of this study was to assess the cumulative incidence of HIV-infection, AIDS and pre-AIDS death in the population of injecting drug users (IDU) in Amsterdam. By assuming equivalence, between a cohort of IDU and the IDU population, of the ratios of incidences of AIDS and pre-AIDS death to the number of HIV positive persons giving rise to these incidences, the numbers of HIV positive persons and pre-AIDS deaths in the population could be calculated, given that other parameters were known. Cohort study data on HIV prevalence and incidences of HIV infection, AIDS, and pre-AIDS death, were combined with national AIDS surveillance data. As of 1 October 1994, the estimated cumulative number of HIV positive IDU in Amsterdam was approximately 1280, far higher than a recent back-calculation estimate. Of the 1280, 204 HIV positive IDU had been diagnosed with AIDS, while about 270 had died pre-AIDS. The HIV prevalence of IDU residing in Amsterdam that were still alive and free of AIDS was hence estimated at around 800. Since the incidence of pre-AIDS death and AIDS exceeded the number of seroconversions during the past four years, the HIV epidemic among IDU in Amsterdam appears to be dwindling. A lower bound of the number of HIV positive IDU being alive, AIDS-free and living elsewhere in the Netherlands was roughly estimated at 600. Because of untimely deaths, only a limited number of HIV positive IDU can be expected to be diagnosed with AIDS in the future. Since these estimates are based upon some rather bold assumptions, they should be interpreted with caution and require further validation by independent sources.