Objective: To determine the incidence of mortality of injecting drug users as a function of the duration of injecting drugs and HIV status, and to assess how these effects vary according to age at initiation and calendar period (before and after 1992).
Methods and design: Cohort of 376 intravenous heroin users admitted to detoxification between February 1987 and January 1990.
Setting: Patients referred from outpatient clinics of metropolitan Barcelona. Duration and characteristics of drug use were determined by interviews. Blood samples were collected during admission and analyzed for HIV, CD4+ cell count and different biologic parameters. Assessment of vital status and causes of death were obtained by hospital charts, death certificates, and autopsies.
Results: The study population consisted of 299 men and 77 women, whose mean age at entry was 26 years, mean duration of injecting drug use before admission 6.1 years; HIV seroprevalence at entry 70.2%. By the end of the follow-up (median 5.6 years), 21.8% of individuals had died (26.6% in HIV-positive, and 10.7% in HIV-negative injecting users). Based on Kaplan-Meier estimates, 10%, 20%, and 30% of HIV negative patients died by 8.7, 11.3 and 14.3 years, respectively, after initiating injecting drugs. The corresponding survival times for the seropositives were substantially lower: 6.6, 8.5, and 11.6 years, respectively. Overall, the survival time was significantly (p < .05) decreased by 22% in HIV-positive injecting drug users. Older age at initiation of injecting drug use was significantly (p < .05) associated with mortality in HIV-positive heroin users but it showed the opposite direction among HIV-negative people. Death rates in HIV-positive patients of the same duration of drug use were similar in periods before and after 1992 (relative hazard (RH) = 0.97; 95% confidence interval: 0.58-1.61). Although not statistically significant, the hazard of death in HIV-negative injecting drug users was substantially lower after 1992 (RH = 0.59).
Conclusions: Before introduction of potent antiretroviral therapies, HIV infection further increased rates of mortality that had already been heightened by injecting drug use. Furthermore, HIV infection modifies the effect of age at initiation and eliminates the seemingly downward trend of mortality in HIV-negative people.