Nutrition-sensitive programs in low- and middle-income countries often aim to improve child nutrition outcomes in part by empowering women. Although previous studies have found cross-sectional associations linking women's empowerment and child nutritional status, there is limited empirical evidence supporting the hypothesis that empowering women as part of an intervention will, in turn, improve child nutritional outcomes. We tested this hypothesis using two waves of data from a cluster-randomized controlled trial of a nutrition-sensitive agricultural program in Burkina Faso. With structural equation models, we examined whether four domains of women's empowerment—purchasing decisions, healthcare decisions, family planning decisions, and spousal communication—mediated the program's impact on reducing wasting and increasing hemoglobin among children who were three to 12 months old at the start of the two-year program. We found that improvements in women's empowerment in the domains of spousal communication, purchasing decisions, healthcare decisions, and family planning decisions contributed to the program's impact on reducing wasting with the largest share being attributable to spousal communication. Improvements in women's empowerment did not contribute to the increase in hemoglobin. These findings provide the first evidence from a randomized controlled trial that women's empowerment is a pathway by which a nutrition-sensitive program can improve child wasting. Programs that aim to improve child nutritional status should incorporate interventions designed to empower women.