Integrated nutrition-sensitive interventions are associated with improvements in household food security and dietary diversity among vulnerable women in the host communities of Cox’s Bazar, Bangladesh: Findings from a pre-post comparison study

Introduction Nutrition-sensitive interventions aim to improve or at least safeguard the underlying causes of malnutrition, especially among the most vulnerable populations. Integrated nutrition-sensitive interventions are effective strategies for improving household food security and dietary diversity in low-income settings. In 2021, the European Union (EU) and UNICEF, in collaboration with the Government of Bangladesh and partners, implemented nutrition-sensitive interventions combining two income-generating activities (IGAs): backyard poultry rearing and tailoring with asset transfers, direct cash support, and nutrition education for vulnerable women in the host communities of Cox’s Bazar, Bangladesh. We evaluated the association of these interventions with household food security and dietary diversity among vulnerable women. Materials and methods We analyzed data from two rounds (2022 and 2024) of repeated cross-sectional surveys involving 1,255 vulnerable women (638 in 2022 and 617 in 2024) across four upazilas in Cox’s Bazar, Bangladesh. Covariates were initially imbalanced across groups because the interventions were not randomly assigned; therefore, we employed propensity score weighting to balance the covariates between the two groups. We used modified Poisson regression and binomial regression models to estimate the association between exposure and outcomes from the pseudo-population obtained by propensity score weighting. All the inferences in this manuscript were based on a 5% significance level. Results In 2024, vulnerable women had a 53% higher likelihood of household food security (RR = 1.53, 95% CI: 1.27–1.85, p < 0.001) than in 2022. Similarly, dietary diversity of vulnerable women increased markedly, showing over two and a half-fold higher likelihood of adequate dietary diversity in 2024 (RR = 2.78, 95% CI: 2.36–3.27, p < 0.001). The E-values corresponding to the associations of food security and dietary diversity with survey year were 2.44 and 5.00, respectively, on the risk ratio scale, which shows that the findings were less sensitive to unmeasured confounders. Conclusion Household food security and dietary diversity substantially increased over time among vulnerable women in the study areas. As the study was a two-repeated cross-sectional survey, residual confounding and other time-varying contextual factors cannot be excluded, so the observed differences should not be interpreted as definitive causal effects of the intervention. Future controlled or longitudinal evaluations are needed to establish the causal effectiveness and long-term sustainability of integrated nutrition-sensitive interventions.