Predictors of survival in HIV-infected patients under antiretroviral treatment in selected hospitals of Ethiopia

Introduction: In spite of the availability of large number of researches that address issues about HIV/AIDS in Ethiopia, evidence on survival and mortality indicators are insignificant.Objective: To assess survival and indicators for mortality of HIV-positive patients receiving ART in Ethiopia.Method: A retrospective facility-based register review was done in 31 purposively selected hospitals across the country. A 15 years data from medical records were included in the analysis. Cox proportional hazards regression was done to identify factors independently associated with survival. Relative risks (hazard ratio) with 95% CI and two-sided test of significance was used to measure the association between dependent and independent variables.Result: From the total 83,876 (56.12%) patient records reviewed, 65,333 (43.71%) were females and the rest were males. Among the total patient records, 10.91% were record of dead people, 47.1% were alive, 22.84% were transferred out, and 13.3% were lost to follow up. About three fourth of patients (77.6 %) were actively working, while 11.65% and 6.71% were ambulatory and bed ridden patients respectively. Death rate among patients on ART was 11.1 per 1,000 Person-year while it is 12.2 per 1,000 Person-year among patients on Pre ART. According to the Kaplan Meier Life Table, the survival probability of patients on ART in the first four years was greater than 95%, and gradually decreased to 78% survival by the end of the 15 years follow-up. Cox proportional hazard model shows that Region, sex, CD4 cell count at ART initiation, functional status, CD4 count at outcome, weight, and age were independent factors associated with death.Conclusion: The overall survival rate in this study was high. Low rate of survival was predicted by sex, region, CD4 count at ART initiation, functional status, and CD4 count at outcome, weight, and age.