Association between exclusive breastfeeding and maternal underweight during the first six months postpartum in Bangladesh: a cross-sectional propensity-score full-matching analysis

Abstract Background Maternal underweight remains a concern during the postpartum period in Bangladesh. This study examined the cross-sectional adjusted prevalence association between exclusive breastfeeding and maternal underweight among mothers of infants younger than six months. Methods We analysed the 2022 Bangladesh Demographic and Health Survey. The original complete-case sample included 423 nonpregnant, ever-married mothers whose most recent infant was aged 0–5 completed months, alive, living with the mother, and currently breastfeeding. Maternal underweight was defined as body mass index < 18.5 kg/m 2 . The primary analysis targeted the average treatment effect among eligible mothers within the common propensity-score support. Sampling-weighted propensity scores were used for full matching, with exact matching on completed infant age. A survey-weighted modified Poisson model adjusted for completed infant age and maternal age estimated the prevalence ratio; marginal standardization estimated adjusted prevalences and their difference. Results Exclusive breastfeeding was reported by 243 of 423 mothers, with a survey-weighted prevalence of 58.0% (95% CI 52.7%-63.1%). Maternal underweight occurred in 39 mothers, corresponding to 8.8% (95% CI 6.4%-12.1%). The common-support matched sample included 409 mothers and 36 underweight cases. Exclusive breastfeeding was associated with an adjusted prevalence ratio of 1.29 (95% CI 0.55–3.04; p = 0.552). Standardized underweight prevalence was 7.42% under non-exclusive breastfeeding and 9.60% under exclusive breastfeeding, a difference of 2.18 percentage points (95% CI -4.67 to 9.04; p = 0.532). Conclusion The estimate was compatible with both a lower and a higher prevalence of maternal underweight, and the confidence interval was wide. Because breastfeeding status and maternal body mass index were measured at the same survey visit, temporality and causality cannot be established. Further longitudinal assessment of maternal nutrition during lactation is warranted without weakening recommended breastfeeding support.