Maternal Diet, Breastmilk Vitamin B12 Concentration, and Infant Feeding in the Context of Food Insecurity and Hunger in East Africa

Poor adherence to recommended infant and young child feeding practices, low dietary diversity, and persistent food insecurity impair maternal and child health and development in sub-Saharan Africa. Behavior change interventions can improve child feeding, but there is a paucity of evidence that they can improve maternal dietary diversity. Low maternal consumption of animal source foods can increase the risk of vitamin B12 inadequacy in breastmilk. Vertical transmission of HIV via breastmilk has led to infants being breastfed for shorter amounts of time, increasing the importance of nutrient rich complementary foods. Animal source foods are particularly nutrient dense, but can be difficult for poor families to attain because they are expensive. These challenges combine to create a difficult environment for optimal breastfeeding, complementary feeding, and maternal dietary diversity during lactation.;Research to measure the efficacy of a potentially scalable behavior change intervention to improve child feeding and maternal dietary diversity in a resource-constrained setting was conducted within a cluster-randomized trial in western Kenya. The nutrition behavior change intervention increased the likelihood of early initiation of breastfeeding by 1.5 times, decreased the risk of early cessation of exclusive breastfeeding 62%, and modestly increased maternal dietary diversity compared to the control. Strong evidence of reporting bias was detected for the outcome early initiation of breastfeeding. Seasonality and household hunger were strongly associated with maternal dietary diversity.;Correlates of breastmilk vitamin B12 concentration were measured within the same trial. Median maternal breastmilk B12 concentration was 152.9 pmol/L [(79.0, 242.7) (Q1, Q3)] and 90% of women had concentrations <362 pmol/L, the recommended adequate level. Seventy-three percent of women consumed B12 below the estimated average requirement, and the majority of animal source food intake in the last week was tea with milk (median six days) and a small fish dish, omena (median two days). Younger mothers (adjusted beta (95% CI) -.01 (-.03, -.01), p=0.006) had higher concentrations of breastmilk vitamin B12 when controlling for confounding.;Research to describe correlates of optimal infant feeding for HIV-exposed infants was nested within a pilot project in coastal Tanzania. Participants were surveyed about socio-demographic characteristics and infant feeding data were collected over 19 days. Exclusive breastfeeding prevalence was 72% and half of infants older than six months no longer breastfeeding received animal source foods on a daily basis. Enrollment facility and HIV disclosure patterns had unexpected associations with infant feeding practices.;In conclusion, diets adequate in animal source foods appear difficult to attain for women in rural western Kenya and infants in coastal Tanzania. The reported improvements in feeding infants under six-months old suggest that behavioral interventions can be effective at message transmission. Evidence of behavior change is lacking, and there is a clear need for biomarker data to validate self-reported breastfeeding practices. The high prevalence of food insecurity and household hunger, with clear associations between household hunger and maternal dietary diversity, indicate the need for targeted programs to improve maternal nutrition.