Objective: To compare the initiation of highly active antiretroviral therapy (HAART) in HIV-infected patients according to sex, route of HIV acquisition and education, and to assess the impact of differences in utilization on the probability of progression to AIDS.
Design and setting: Swiss HIV Cohort Study, a national prospective multi-centre study.
Participants: A total of 3342 patients, including 1007 (30%) women. HIV was acquired through injection drug use in 1155 (35%) cases and through sex between men in 1172 (35%). Twenty-eight per cent (957) of participants had attained only the minimum level of schooling. At baseline, the median CD4 cell count was 269x10(6)/l cells, median HIV-1 RNA was 4.3 log10 copies/ml and 2917 (87%) were free of AIDS.
Methods: Kaplan-Meier life tables and Cox proportional hazards regression.
Results: During 7007 person-years of follow-up 2285 (69%) patients started HAART and 318 (10%) developed a new AIDS event. In multivariable analysis controlling for CD4 cell count, viral load and disease stage at baseline, the probability of starting HAART was lower in injection drug users compared with men who have sex with men, hazard ratio 0.63 (95% confidence intervals 0.56-0.70) and in patients with minimum schooling compared with those with vocational training, hazard ratio 0.82 (0.75-0.91). The risk of progression to AIDS was similar among men and women, patients with a history of injecting drug use, and patients with lower educational attainment in both univariable and multivariable analysis.
Conclusion: HIV-infected injecting drug users and those with lower levels of educational attainment start HAART later than other patient groups. The deferred initiation of therapy in these patients does not, however, appear to translate into an increased risk of clinical disease progression. This observation has important implications for treatment policy and the design of future clinical trials.