Highly active antiretroviral therapy (HAART) among HIV-infected drug users: a prospective cohort study of sexual risk and injecting behaviour

Aims: To study sexual risk and injecting behaviour among HIV-infected drug users (DU) receiving highly active antiretroviral therapy (HAART). Design and setting: As part of an ongoing prospective cohort study, HIV-infected DU who commenced HAART (n = 67) were matched with those not starting HAART (n = 130) on CD4 cell counts, duration of cohort participation, age and calendar year of visit. Immunological and virological responses of the HAART-treated DU were compared with the HAART-treated homosexual men from the same cohort (n = 212). Measurements: Trends in behaviour and therapeutic response were tested with a logistic regression model adjusted for repeated measurements and a piecewise random effects model, respectively. Findings: Non-HAART users reported more episodes of injecting than HAART users. In both groups injecting declined over time with no effect of HAART initiation. Before HAART initiation an increase in sexual risk behaviour was observed among those who had been assigned to receive HAART; their sexual risk behaviour declined thereafter. No change in sexual risk behaviour was found among non-HAART users. Relative to homosexual men, DU had a similar initial therapeutic response, but DU started HAART at lower CD4 cell counts and higher viral load levels. Conclusion: DU who are treated with HAART are not increasing their risk behaviour, and their early response to HAART is similar to homosexual men. However, before the treated DU received HAART they were seen to inject less often than those not treated with HAART. This suggests that selection of potential HAART starters is based on limited drug use. Although the DU who commence HAART are a selected group, our results show that HIV-infected DU can be treated effectively. (Review' s abstract)