Evaluation of survival rates of patients on ART in Kibera Slums, Nairobi, Kenya

Abstract

Background: Recent improvements in access to Anti retroviral Therapy (ART) have radically reduced hospitalization rates and deaths associated with Human Immunodeficiency Virus (HIV) infection in both developed countries and sub-Saharan Africa. Majority of HIV infected populations in sub-Saharan Africa live in urban slums and do not have access to modem HIV care services due to informal nature of the settlements characterized by inadequate infrastructure to deliver effective health services. Not much is known about survival of patients on ART in the slum settings. This paper aims to identify factors associated with survival among adult patients on ART in resource poor, urban, sub-Saharan African setting.

Methodology: A total of 2011 patients from a prospective open cohort of HIV + patients on ART are used in this study. The patients were enrolled on treatment between March 2005 and September 2011. Kaplan-Meier methods and Cox's proportional hazards model were used to analyze the data. Key variables considered are socio-demographic characteristics, regimen, baseline CD4 count, functional status, World Health Organization (WHO) Staging and opportunistic infections.

Results: Females had better survival rates on ART than males (p=0.0016). Patients with baseline CD4 counts < 50 died faster than those with higher CD4 counts (p<O.OOOl). Major regimen (hazard ratio (HR) =1.8, 95% Confidence Interval (CI) [1.5-2.2]), WHO staging (HR=1.3, 95% CI [1.1-1.6]) had significant effect on survival of patients on ART. A test of the proportionality assumption through the computation of Schoenfeld residuals shows that the P-values for covariates CD4Baseline, WABART and WHOART are 0.941, 0.987 and 0.946 respectively at 95% CI implying that the application ,..of the Cox PH model is appropriate in this study.

Conclusion: CD4 cell count, patients changing regimen, major regirnen of treatment combination, functional status of the patient and Patient WHO staging at start of ART are major contributing factors to mortality among adult patients on ART. The program needs to focus keenly of male patients as they are more prone to die from AIDS related complications than the female participants. The findings are relevant for decisions on quality of service given to ART patients in slum settings. It is important for studies that apply survival models to test for ph assumptions to avoid biased results and use alternative models if need arises.