Although the adverse effects associated with prenatal alcohol exposure (PAE) are well known, many women continue to drink heavily during pregnancy, putting their infants at risk for fetal alcohol spectrum disorders. Animal studies have shown that choline supplementation can mitigate effects of PAE on growth and development. Choline, an essential nutrient, serves as a methyl-group donor for DNA methylation and is a constituent of the neurotransmitter acetylcholine and a precursor to major components of cell membranes. In an R21 feasibility trial, 70 heavy drinkers were randomly assigned to receive a daily dose of 2g of choline or a placebo from initiation of antenatal care to delivery in Cape Town, South Africa, where the incidence of heavy drinking during pregnancy and fetal alcohol syndrome are among the highest in the world. When compared with infants in the placebo arm, infants in the choline-treated arm were more likely to meet criterion for eyeblink conditioning, demonstrated markedly better recognition memory on the Fagan Test of Infant Intelligence, which is known to have predictive validity for school-age IQ, and had better postnatal gains in weight and head circumference. Key features of this study included the higher choline dose (4.4 times adequate intake (AI), compared to 1.7-2.5 in previous human studies) and initiation of treatment early in pregnancy. We are now conducting a fully-powered, double-blind, randomized, placebo-controlled choline supplementation trial in heavy drinking pregnant women from a rural community in South Africa (1) to assess the effectiveness of maternal choline supplementation during pregnancy to mitigate effects of PAE on three primary outcomes: infant recognition memory and postnatal growth restriction (weight and head circumference); (2) to assess the efficacy of this supplementation for mitigating alcohol effects on the following secondary outcomes: infant eyeblink conditioning, postnatal length, and information processing speed; (3) to use innovative methods in causal inference analysis to examine protocol adherence as an important source of variation in treatment efficacy and to identify sociodemographic factors associated with non-compliance in order to facilitate implementation of the intervention protocol in clinical settings; and (4) in exploratory analyses, to examine whether maternal choline supplementation is particularly effective in women with lower dietary choline intake or poor nutritional status.