The application of logistic regression in modeling of survival chances of HIV-positive patients under highly active antiretroviral therapy (HAART): A Case of Nyakach District, Kenya

Abstract

In Kenya the number of persons requiring antiretroviral Therapy (ART) in 2012 was over 90,000 while globally more than 1.4 million were under ART in the year 2012. Though ART treatment has decreased mortality of HIV associated deaths, there is still need for funding of ART services by donors and Ministry of health so as to prevent the loss caused by AIDS related deaths in the society. This study modeled using logistic regression the survival chances of the HIV positive patients under ART treatment in within Nyakach district. The study outlined the various socio-economic factors affecting the survival or death of the patients using logistic regression model. The sample of 320 patients was considered from patients from Katito health center from Nyakach district for a period of six months. The analysis of the data showed that risk factors consisting TB infection, kind of drug regimen, distance from the health facility, access to tap and portable water, the kind of marital status of patients and counselling session attendance by patients on ART affected their survival. Thus, the survival of patients under ART programs can be improved if we improve the access to socioeconomic factors like access to water, setting up many health facilities too provide health services at close distance to the affected and also we can bring behavioral change among HIV patients to attend counselling session and get pieces of advices on correct health measures and be-haviours.