Survival analysis and predictors of mortality among adult HIV/AIDS patients initiated antiretroviral therapy from 2010 to 2015 in Dubti General Hospital, Afar, Ethiopia: A retrospective cohort study

Abstract

BackgroundAlthough antiretroviral therapy (ART) is well accepted to increase survival of patients with HIV/AIDS, AIDS related deaths continue to be a major problem in sub-Saharan Africa like Ethiopia. Studies have showed variable findings in the survival status of patients with HIV/AIDS initiating ART, and there was no such study in the study area. Therefore, purpose of this study was to determine the survival and predictors of mortality among HIV/AIDS patients starting taking ART in Dubti General Hospital, Afar, Ethiopia. MethodsA 5 year retrospective cohort study was performed among 702 HIV/AIDS patients aged ≥15 years that started ART between December 31, 2010, and December 31, 2015 in Dubti General Hospital, Afar, Ethiopia. A simple random sampling technique was used to select the study subjects from each WHO stage based stratum. Socio-demographic, clinical and survival status data were extracted by reviewing patients’ records. Data were analyzed by using SPSS Version 21. Kaplan-Meier and Cox-regression models were used to estimate survival, and explore predictors of mortality. Variables with a p value of <0.05 in multivariate Cox regression analysis were considered statistically significant. ResultsAmong 702 study participants, 82 (11.7%) died during follow up, and the overall incidence rate of mortality was 5.81 per 100 person-years. Identified predictors of mortality were being not married (AHR = 3.71, 95% CI: 1.97–6.99), had no formal education (AHR = 2.33, 95% CI: 1.33–4.38), bedridden functional status (AHR = 5.91, 95% CI: 2.71–12.88), advanced WHO stage III and IV (AHR = 4.36, 95% CI: 2.20–8.64), BMI 16–18.4 kg/m2 (AHR = 3.03, 95% CI: 1.50–6.13), and BMI<16.0 kg/m2 (AHR = 5.47; 95% CI: 2.85–10.50), CD4 count ≤50 cells/mm3 (AHR = 6.62, 95% CI: 4.73–8.52), hemoglobin <8 g/dl (AHR = 5.21; 95% CI: 2.64–10.26), not used cotrimoxazole prophylaxis therapy (AHR = 2.78, 95% CI: 1.61–4.73), stavudine based regimen (AHR = 2.34, 95% CI: 1.32–4.13), and zidovudine based regimen (AHR = 2.49, 95% CI: 1.41–4.39). ConclusionHigh mortality was observed in this cohort, and participants with stage III and IV, low CD4 count, low hemoglobin level, bed ridden functional status, low BMI should be closely monitored even with the scarce resources. In addition, the use of cotrimoxazole prophylaxis therapy should be more encouraged to increase survival.