Abstract
BACKGROUND: The food environment is a driver of the double burden of malnutrition, influencing dietary intake by increasing or restricting access to foods. OBJECTIVES: The objective of this study was to assess the association between geospatial food environment indicators and the diet quality among postpartum women in rural Bangladesh. METHODS: Participants were women of infants enrolled in a cluster-randomized controlled trial from 2018 to 2020. Food vendor availability was defined as the number of food vendors within a specific household radius, and proximity was defined as the distance to the nearest vendor. Dietary intake was measured using a 7-d food frequency questionnaire collected at 3 mo, 6 mo, and 12 mo postpartum. Our primary outcome was nonstarchy staple food variety scores (FVS). Secondary outcomes included dietary diversity scores and individual food group consumption. To assess the association between food environment indicators and diet quality indicators, we fit linear regression models for the FVS outcome, Poisson regression models for the dietary diversity outcome, and logistic regression models for individual food group outcomes. RESULTS: A total of 5064 women were included in this analysis. Women reported consuming an average of 9.7 (standard deviation: 3.7) nonstarchy staple foods in the previous week. Women who lived in households with the highest market availability (≥7 markets within 1600 m) had an average of 0.84-unit (95% confidence interval: 0.53, 1.16) higher FVS compared with those in households with the lowest market availability (≤ 2 markets) (P < 0.001). Geospatial food environment indicators were not significantly associated with the odds of consuming less healthy food options. CONCLUSIONS: We found a positive relationship between market availability and diet quality for postpartum women in rural Bangladesh. However, more research is needed to understand which components of the food environment are associated with increased consumption of less healthy foods.