Cox proportional hazard model to the survival status of HIV infected patients

Abstract

The paper is on the application of Cox proportional hazards model with a view to identify the factors that affect the survival and influence death  status of HIV infected patients. A secondary data from a retrospective cohort study based on test from HIV infected patients on antiretroviral treatment in the University of Port Harcourt Teaching Hospital was used. The result of the analysis showed that out of 330 registered HIV patients, 56 (34.2%) died during the study period and 274 (65.8%) were censored. Cox hazard regression analysis showed that the covariate HB and CD4 count were significant risk factors associated with survival in HIV infected patients.

Keywords: Cox regression model, Hazard ratio, Non-proportional hazards, Kaplan-Meier Estimator, Log-Rank Test, Survival Function.