Factors associated with body mass index in people living with HIV/AIDS attending Thika district hospital comprehensive care clinic, Kenya

Abstract

Background: loday despite advanced therapeutic success of highly active antiretroviral therapy (HAART) in the developed countries, HIV related weight loss still continue to be important co-morbidity in many patients. The situation is worse in developing countries due to limited resources and high HIV prevalence. Though H1V/AIDS is a widespread problem in Kenya, there is little information about the nutritional status of PLWHA in the country. Objective: The main objective of this study was to determine factors associated with Body Mass Index (BMI) as a measure of nutritional status in PLWHA attending Thika District Hospital Comprehensive Care Clinic (TDH CCC). Study design: The study design was a case control study in which cases were PLWHA with BMI less thanl8.5kg/m2 and controls PLWHA with BMI greater or equal tol8.5kg/m2. Study site/pop: The study centre was TDH CCC and study population was adults living with HIV/AIDS being followed up at TDH CCC. All HIVpositive patients above 18 years attending TDHCCC were eligible for selection as cases or control, cases (n=99 -those with BMI<18.5kg/m2) and controls (n=99-those with BMI >18.5kg/m2). Analysis: Univariate, bivariate and multivariate analyses were done. Statistical - chi-square and standard regression techniques were used. As a measure of relative risk of malnutrition, odds ratios (ORs) and their 95% confidence intervals (CIs) computed from a logistic regression model. For all the analyses, two-sided tests were used together with the 5% level of significance with p-valucs of <0.05 being considered to be significant. In the analysis of the association between BMI and socio- dcmographic/lifestylc characteristics only income was significantly associated with reduced risk of malnutrition as measured by BMI. In particular earning Kenya shillings 5000 9991 and >10,000 were associated with 69% and 85% risk reduction of low BMI (P=0.019 and 0.006), respectively. There was a tendency for the risk of low BMI to be less among the females compared to the males although this was not statistically significant (P=0.299). Higher education than primary level was associated with reduced risk of malnutrition as measured by BMI with 46% and 1% reduction among those with secondary and tertiary education respectively. Smoking was associated with 35% risk of low BMI compared to those not smoking but this was not statistically significant (p=0.77). Those taking alcohol were 1.4 times at risk of low BMI compared to those not taking alcohol. The major conclusions derived from this study wee that being a female, higher education level than primary and higher income were associated with reduced risk of malnutrition as measured by BMI. Smoking was associated with increased risk of low BMI by 35% and taking alcohol was associated with 1.4 times risk of low BMI but this was not statistically significant. Thus the hypothesis “socio-economic/demographic and lifestyle characteristics are not associated with BMI in PLWHA in TDH CCC” was rejected. Recommendations: Nutritional status amongst PLWHA requires to be monitored regularly and it is important to educate PLWHA as well as empower them economically to mitigate the adverse effects of low BMI in relation to rapid progression of HIV infection. There is need to do a surveillance to determine the prevalence rate of smoking/alcohol consumption among PLWHA so that intervention programs can be implemented to counter the adverse effects of these drugs amongst PLWHA. Key words: PLWHA, Kenya, socio-demographics, lifestyle, BMI, Case- control study