|| Checking for direct PDF access through Ovid
To assess the utility of the 2003 revised World Health Organization (WHO) criteria [initiating highly active antiretroviral therapy (HAART) in stage IV, in stage III plus CD4 cell count < 350 × 106 cells/l, or in stage I or II plus CD4 cell count < 200 × 106 cells/l] relative to other scenarios of HAART initiation.Progression to AIDS and death in 292 patients taking HAART and 974 not taking HAART in a South African institution in 1992–2001, stratifying patients by baseline CD4 cell count and WHO stage.HAART was associated with decreased AIDS [adjusted rate ratio [ARR], 0.16; 95% confidence interval (CI), 0.08–0.31) and death (ARR, 0.10; 95% CI, 0.06–0.18). Benefit of HAART was significant across all WHO stages plus CD4 cell counts. The greatest number of deaths averted was in stages IV [74.0/100 patient-years (PY); 95% CI, 50.2–84.5) and III (32.8/100 PY; 95% CI, 22.4–40.9). AIDS cases averted in stage III (22.0/100 PY; 95% CI, 6.1–26.9) were higher than in stage I and II with CD4 cell count < 200 × 106 cells/l (8.9/100 PY 95% CI, 5.6–13.3). Treatment initiation for symptomatic disease resulted in greater benefits than using any CD4 cell thresholds. Application of WHO criteria increased the treatment-eligible proportion from 44.5% to 56.7% (P < 0.05) but did not prevent more death (P > 0.05) than treating symptomatic disease.Implementation of the revised WHO guidelines in sub-Saharan Africa may result in a significantly increased number of individuals eligible for treatment but would not be as effective a strategy for preventing death as treating symptomatic disease.