OBJECTIVE: To estimate the survival of AIDS patients reported in HFSE from 1986 to 2006. METHODS: We used the extended proportional hazards model of Cox considering the diagnosis date as a time-dependent variable, which divided the analysis in two periods: <1996 and ≥1996 (after the introduction of antiretrovirals in HFSE). RESULTS: Among the 1.300 cases analyzed, 62% were males. The following variables showed significant hazard ratios (HR): criteria for notification based on the quantification of serum CD4<350 (HR=0.187); group diagnosed after 1996 (HR=0.355); age >50 years (HR=1.386); dysfunction of the central nervous system (HR=1.570), cachexia (HR=1.526); Kaposi's sarcoma (HR=1.376); candidiasis (HR=1.295). The overall survival rate at 5 years was 35.7% and at 10 years, 6.3%. For cases reported through the CD4 criteria, survival at 5 and 10 years was 89.6%. CONCLUSION: This study adds evidence to the increased survival of AIDS patients and points the CD4 notification criteria and group diagnosed after 1996 as the variables with the greatest association with this increase.
acquired immunodeficiency syndrome; survival analysis; epidemiologic surveillance services; epidemiological surveillance