Predictors of exclusive breast feeding among HIV-positive mothers in North Rift region of western Kenya

Abstract

Exclusive breastfeeding (EBF) for the first six months of infants’ lives is a cost effective intervention which can avert 13-15% of the 9 million deaths of children under 5 years old in resource poor settings. In areas where HIV prevalence is high, especially in sub-Saharan Africa, EBF has the added advantage of reducing the risk of mother-to-child transmission (MTCT) of HIV. However, EBF rates have been shown to be lower in resource poor settings, than the World Health Organization recommended 90% in spite of PMTCT counseling that is expected to increase prevalence of EBF. This low prevalence is a public health problem, especially in developing countries; hence, understanding predictors of EBF is essential for development of relevant intervention strategies. The North Rift is home to the most established comprehensive care clinics (CCCs) in Kenya, yet the overall prevalence of exclusive breastfeeding in the Rift Valley which encompasses the North Rift is 35%, with no specific data for the North Rift. Predictors of EBF vary widely between and within countries, necessitating the need to collect context-specific data. A cross-sectional study was conducted in the North Rift, Kenya to determine predictors of EBF among HIV-positive mothers attending five CCCs including: Kitale District hospital (DH), Turbo Health Centre (HC), Moi Teaching and Referral Hospital, Kabarnet DH and Mosoriot HC were included. Objectives were to determine prevalence of EBF; socio-demographic, cultural and economic predictors of EBF; and from these determine the main predictors of EBF. HIV-positive mothers (n=297) attending the CCCs, selected by random sampling, were included. Socio-demographic, cultural and economic information was collected from the mothers using open and closed ended questions. Qualitative data was collected in Focus group discussions (FGDs) conducted among the mothers until saturation, and key informant interviews conducted among purposively selected health workers, to corroborate data generated from the questionnaire. Socio-demographic, cultural and economic predictors of EBF were identified using bivariate and multivariate regression analysis. Main predictors of EBF were then identified using multiple logistic regression analysis. Prevalence of EBF was 63% confirming disparity with the recommended 90% prevalence. The main predictors of EBF were: OR (95% CI): education level 17.67(0.906, 2.512); knowledge 17.85(3.806, 8.372); stigma 0.19(0.092, 0.394); traditional beliefs 0.03(0.007, 0.154). Improving education of these women, enhancing their knowledge on breastfeeding, exploring interventions to address stigma and traditional beliefs can contribute to enhancing the prevalence of EBF in the North Rift as a step towards achieving the WHO target. Lessons on best practices from Mosoriot and Turbo should be explored as a starting point.