Infant feeding practices in a high HIV prevalence rural district of KwaZulu-Natal, South Africa

Aim: To describe infant feeding practices at birth and at 14 weeks post-partum in the Ugu-North Health District, KwaZulu-Natal, South Africa. Methods: A prospective, cohort study design was used. Mothers who delivered over a one-month period were interviewed at birth and 14 weeks later. Results: Initially, 168 mothers were interviewed within 24 hours of delivery, of whom 117 (70%) were contactable at 14 weeks post-partum. The vast majority (96%) initiated breast-feeding at birth. At birth, less than one-third (55/168 [32.7%]) of mothers declared an intention to both breast and formula (mix) feed in the next 14 weeks, but by the 14th week post-partum over three-quarters (89/117 [76.1%]) actually practised mixed feeding. At 14 weeks, the prevalence of exclusive breast-feeding was 18%: 52% of infants were offered water and 73% solids. The majority (20/23 [87%]) of HIV infected mothers chose to breastfeed their infants at birth. Nevertheless, they were significantly more likely to formula feed their infants compared to HIV negative mothers (3/23 [13.0%] vs 2/145 [1.4%], OR 10.73, 95% CI 1.34 - 99.16, p = 0.02). By 14 weeks, only 11% of HIV positive mothers were still exclusively breast-feeding, while almost two-thirds (12/19 [63%]) practised mixed feeding. This change was mainly ascribed to their need to return to school (40%) or to work (20%). Conclusions: Most infants were fed inappropriately by 14 weeks of age. The failure to maintain exclusive breast-feeding, despite high initiation rates, is of greatest concern. Routine prevention of mother-to-child transmission of HIV services was ineffective in influencing mothers to follow any feeding regimen exclusively.