BACKGROUND: Evidence is often missing on social differentials in effects of nutrition interventions. We evaluated the adherence to and effect of prenatal food and micronutrient supplementations on mortality before the age of five years in different social groups as defined by maternal schooling. METHODS: Data came from the MINIMat study (Maternal and Infant Nutrition Interventions, Matlab), a randomized trial of prenatal food supplementation (invitation early, about 9 weeks [E], or at usual time, about 20 weeks [U] of pregnancy) and 30 mg or 60 mg iron with 400 μgm folic acid, or multiple micronutrients (Fe30F, Fe60F, MMS) resulting in six randomization groups, EFe30F, UFe30F, EFe60F, UFe60F, EMMS, and UMMS (n = 4436). Included in analysis after omissions (fetal loss and out-migration) were 3625 women and 3659 live births of which 3591 had information on maternal schooling. The study site was rural Matlab, Bangladesh. The main stratifying variable was maternal schooling dichotomized as <6 years and ≥6 years. We used Cox proportional hazard model for survival analyses. RESULTS: Overall, women having <6 years of schooling adhered more to food (81 vs. 69 packets, P=0.0001) but a little less to micronutrient (104 vs. 120 capsules, P = 0.0001) supplementation compared to women having more schooling, adjusted for maternal age (years), parity and body mass index (BMI, kg/m2) at week 8 pregnancy. Children of mothers with ≥6 years of schooling had lower under-five mortality, but the EMMS supplementation reduced the social difference in mortality risk (using standard program and schooling <6 years as reference; standard program and schooling ≥6 years HR 0.54, 95% CI 0.27-1.11; EMMS and schooling ≥6 years HR 0.28, 95% CI 0.12-0.70; EMMS and schooling <6 years HR 0.26, 95% CI 0.11-0.63), adjusted for maternal age (years), parity and body mass index (kg/m2) at week 8 pregnancy. CONCLUSIONS: The combination of an early invitation to prenatal food supplementation and multiple micronutrient supplementation lowered mortality in children before the age of five years and reduced the gap in child survival chances between social groups. The pattern of adherence to the supplementations was complex; women with less education adhered more to food supplementation while those with more education had higher adherence to micronutrients. TRIAL REGISTRATION: ISRCTN16581394.