Adherence to feeding guidelines among HIV-infected and HIV uninfected mothers in a rural district in Uganda.

Objective: To describe the infant feeding behaviour of HIV-infected and HIV-uninfected mothers, and identify factors influencing adherence to infant feeding guidelines.

Methods: Analytical cross-sectional study. Methods: Bushenyi, rural district in South-western Uganda Methods: One hundred and ninety four mothers who had a child less than 12 months of age. About half, 94 (48.5%), of these were HIV-infected. Methods: Proportion of mothers who exclusively breastfed, complementary fed, replacement fed, and adhered to feeding guidelines.

Results: Most (84.5%, 164/194) of the mothers had ever breastfed their infants, the rest had exclusively replacement fed since birth. Among children less than six months who were breastfeeding, 31.5% (34/108) were exclusively breastfeeding and the rest were mixed feeding. HIV-infected mothers were more likely than HIV-uninfected mothers to exclusively breastfeed (Crude Odds Ratio [COR], 3.61, 95% Confidence Interval [CI] 1.42-9.21). For infants older than six months, complementary feeding was more common among HIV-uninfected (100%) than HIV-infected mothers (41.7%; P < 0.001). Among infants of all ages, none of the HIV-uninfected and 45% of HIV-infected mothers were replacement feeding (p < 0.001). More than a half (59.8%) of the mothers adhered to infant feeding guidelines. The only independent predictor of adherence after multivariate analysis was mother ever attending infant feeding counselling (AOR 9.03; 95% CI 4.03-20.25). Only 35% of mothers reported ever attending infant feeding counselling.

Conclusions: Infant feeding counselling was associated with improved adherence to feeding guidelines. Since infant feeding counselling is low in this population there is need for scale-up of this essential service.