In developing countries, several poor infant feeding practices are common, resulting in physical and intellectual developmental impairments. Good feeding practices are crucial, especially in the first year of life. In addition to the malnutrition currently suffered in developing countries, HIV/AIDS has worsened the clinical and nutritional status of both mothers and their children. HIV-infected women face difficulties in choosing among infant feeding options because they must balance the risks of postnatal HIV transmission and infant morbidity and mortality. Many mothers in low-income settings like Rwanda do choose to breast feed but this simply shifts the difficult decision to when to stop. However, few studies have explored feeding practices among HIV-exposed infants during the transition period when breastfeeding is stopped.

An Infant and Child Feeding Index (ICFI) was created in this study to (1) explore infant feeding practices, (2) assess their association with nutritional status among HIV-exposed infants, and (3) evaluate the magnitude of the association over time in Rwanda. Data analysis, using a cross-sectional time series model, suggests that ICFI is positively associated with Weight for Length Z-score (WLZ) and Weight for Age Z-score (WAZ). However, neither the ICFI nor any of its components were associated with the Length for Age Z-score. Other feeding variables related to behaviors, including initiation of breastfeeding within one hour after birth and ever breastfeeding, were not associated with the anthropometric indices, but these take place much earlier. There was no correlation between changes in the index over time and anthropometric indices. Mean LAZ, WLZ, and WAZ were all lower in male infants. In Rwanda, reorganizing nutritional units at the central level and integrating nutritional services at district and health facility levels are critical to delivering simple and concise nutritional messages that involve all types of orphans and vulnerable children.