Date of Award

January 2013

Document Type

Open Access Thesis

Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

Kaveh Khoshnood



Exclusive breastfeeding (EBF) has been shown to result in lowered risk of postnatal vertical HIV transmission compared to mixed breastfeeding (Mixed BF). Early breastfeeding cessation, or not breastfeeding at all, is associated with an increase in morbidity and mortality amongst both HIV exposed and HIV unexposed infants. The South African tradition of giving nutritive and non-nutritive foods to breastfeeding infants in the first few months of life is unique within African nations and warrants closer examination to identify maternal determinants that lead to early mixed feeding practices, commonly used supplemental foods/liquids, the timing of supplemental food introduction, and maternal perceptions of what constitutes mixed feeding. Additionally, understanding the determinants of general early infant feeding intent and practices among HIV-positive and -negative mothers has become germane again in light of the most recent WHO guidelines and South Africa's updated national recommendations on prevention of mother to child transmission (PMTCT) and infant feeding practices in the context of HIV. Having this information will better direct PMTCT counseling, infant feeding interventions, and HIV-exposed uninfected (HEU) child focused interventions.


Participants and data were gathered in a larger prospective cohort study focused on the health and development of children born to HIV-positive mothers from 11 PMTCT clinics in the Tshwane region of South Africa. The larger study followed women who gave birth between January and July of 2011 and their children, Women were enrolled in the study at the time of birth and follow-up was conducted with all HIV positive women and a sample of one-third of HIV negative women at three months. This study used bivariate analyses to identify significant associations between maternal HIV status and self reported feeding choices at birth and feeding practices within the first 3 months of life.


Of the 511 mothers followed to 3 months, 260 were HIV-positive and 251 were HIV-negative. 240 (98.4%) HIV-negative women intended to breastfeed their baby at baseline and only 4 (1.6%) intended to formula feed as compared to 203 (81.2%) and 47 (18.8%) HIV-positive mothers, respectively. By 3 months, the majority, 206 (86.2%), of HIV-negative women had conducted Mixed BF with nutritive or non-nutritive supplemental foods, while 79 (30.9%) HIV-positive mothers maintained EBF, 78 (30.5%) had at some point applied Mixed BF, and 99 (38.7%) practiced formula feeding with no history of Mixed BF. Among HIV-positive women, those who had ever breastfed (n=198) by 3 months were more likely than those who had never breastfed (n=58) to: have a life partner (p=0.01), live in a home with fewer adults (p=0.02), report a higher degree of stigma (p=0.04), and posses a greater understanding of HIV transmission risk in breastfeeding (p<0.01). Additionally, compared to those who practiced Mixed BF (n=78), HIV-positive mothers who maintained EBF to 3 months (n=79) were more likely to report higher levels of stigma and negative support as well as possess a greater understanding of HIV transmission risks in breastfeeding. No significant differences were found between EBF HIV-positive mothers (n=79) and formula feeding HIV-positive mothers who had ever breastfed (n=41).


While early infant feeding practices were poor among HIV-positive and -negative women, HIV-positive women did maintain safer infant feeding practices to 3 months. Yet, only 30.9% of HIV-positive women were able to maintain EBF to 3 months and those women who practiced EBF felt more stigmatized and reported support that was more domineering. The fact that Mixed BF is the dominant form of early infant feeding in this geographic region, complicates the goals of PMTCT programs and obscures local, national and international infant feeding recommendations of EBF until 6 months for HIV-positive mothers. Generally it is assumed that mixed feeding most often involves giving formula, but in fact the substance that was most commonly given to babies in this cohort was muti, which may be based on parental beliefs that they are benefiting the health of their infants. A better understanding is needed of the motivations for introducing nutritive and non-nutritive foods/liquids against clinical recommendations in South Africa. Educational and behavioral interventions are needed to promote safe infant feeding practices for the general population, as well as HIV-positive mothers.


This is an Open Access Thesis.

Open Access

This Article is Open Access