(OR10-02-26) Maternal Dietary Patterns in Pregnancy and Their Relationship to Birth Outcomes in Bangladesh: A Latent-Transition Analysis of the JiVitA-3 Trial

Objectives: Maternal diet can influence fetal growth; this analysis aims to examine the influence of dietary pattern changes during pregnancy on birth outcomes in rural Bangladesh.

Methods: Data were analyzed from JiVitA-3, a cluster-randomized trial of multiple micronutrient vs. iron–folic acid supplementation among rural Bangladeshi pregnant women. Dietary intake was assessed using seven-day food frequency questionnaires in early (~12 weeks gestational age) and late pregnancy (~32 weeks). Mean servings of eight of the ten Minimum Dietary Diversity – Women food groups (excluding Nuts and Grains, roots and tubers due to lack of variability) were used to construct latent profiles of dietary patterns at each time point. Latent transition analysis quantified pattern changes across pregnancy adjusting for socioeconomic status, season, and Ramadan fasting. Associations between late pregnancy dietary patterns and LBW were estimated using linear probability models adjusted for the transition from early to late pregnancy diet, maternal pre-pregnancy BMI, maternal age, infant sex, and infant gestational age at birth.

Results: Four dietary profiles were identified in both early and late pregnancy among n=18,672 women (selected from 24,652 singleton live births for complete covariates and anthropometry measured within 72 hours of birth). Patterns included: low diversity (71% early –73% late pregnancy), high dairy (20–24%), high Other Vitamin A (3–4%), and high pulses (2–3%). Most women (65%) maintained the same pattern across pregnancy with 35% transitioning; most transitions occurred between the high dairy and low diversity groups. Season and socioeconomic status were the main predictors of transitioning between dietary patterns. Late pregnancy dietary pattern but not early, showed an association with LBW (43%). High Dairy or High Other Vitamin A diets in late pregnancy were associated with a lower risk of LBW compared to the Low Diversity pattern (risk difference: -4.4 percentage points (95% CI: -6.2%, -2.7%) and -4.5% (-8.7%, -0.2%), respectively).

Conclusions: Socioeconomic and seasonal factors influenced dietary changes during pregnancy in rural Bangladesh, with implications for fetal growth.

Funding Sources: Gates Foundation, Edesia Nutrition