The association of different CD4(+) cell counts with the same disease risk in treated and untreated populations reflects the effectiveness of highly active antiretroviral therapy (HAART) in persons with human immunodeficiency virus (HIV). Clinical progression of disease following initiation of HAART was determined for 679 HIV-infected men in the Multicenter AIDS Cohort Study by means of Kaplan-Meier survival analyses. Cox proportional hazards models were used to assess the effects of markers of HIV disease, antiretroviral history, and demographic factors. Men who had been followed since January 1993 (pre-HAART) were used to identify CD4(+) levels associated with the acquired immunodeficiency syndrome (AIDS)-free time equivalent to that of men starting HAART with CD4(+) cell counts of <200 cells/microl. Within 3.5 years following HAART initiation, 11.3% of the subjects developed AIDS and 8.5% died. Determinants of AIDS were a CD4(+) cell count of <200 cells/microl at initiation (relative hazard = 2.25, 95% confidence interval: 1.13, 4.49) and age >45 years at initiation (relative hazard = 1.92, 95% confidence interval: 0.98, 3.77). An increase in CD4(+) cell count of >50 cells/microl immediately after HAART initiation also improved prognosis (relative hazard = 0.34, 95% confidence interval: 0.16, 0.71). AIDS risk in men starting HAART with CD4(+) counts of <200 cells/microl (median = 132) was similar to that of non-HAART users with CD4(+) counts of 375-475 cells/microl (median = 432). The equivalence of disease progression to that of nonusers with approximately 300 more cells per microl demonstrates that HAART users have a broader reconstitution of the immune system beyond that of observed increases in CD4(+) cell count.