This study examined the relations between maternal feeding behaviors (responsive and nonresponsive) and children's nutritional status at 24 months of age in rural Bangladesh where child undernutrition is high. Data were collected on 4846 mother‐child dyads enrolled in a complementary food supplementation trial. Anthropometry (weight, length) and child diet (24‐h recall) were collected. A context‐specific 11‐item feeding behaviors module was developed; factor analysis revealed 2 theoretically derived constructs of feeding: responsive (5 items) and forceful (6 items). Each was dichotomized to reflect low and high feeding behaviors, with approximately 2/3 of the sample reporting optimal feeding behaviors for both constructs (high responsive [70%], low forceful [66%]). In addition, a 2×2 factorial variable was created using the dichotomized constructs to produce 4 categories: 1) high responsive/low forceful; 2) high responsive/high forceful; 3) low responsive/high forceful; and 4) low responsive/low forceful. Dietary diversity score (DDS; range 0–7) was derived using the diet recall data and was used to define minimum dietary diversity (MDD; score of 4 or more). Relations were examined using multivariate linear/logistic regression models, adjusting for confounders, intervention status, and study design. Responsive feeding was positively associated with DDS and meeting MDD (β: 0.09, 95% CI: 0.01, 0.17; OR: 1.2, 95% CI: 1.03, 1.3, respectively). No significant relations were seen with growth outcomes (p>0.05). Conversely, forceful feeding was negatively associated with WAZ, WLZ, and DDS (β: −0.16, 95% CI: −0.22, −0.11; β: −0.19, 95% CI: −0.25, −0.14; β: −0.12, 95% CI: −0.20, −0.03, respectively) and positively with wasting, underweight, and meeting MDD (OR: 1.38, 95% CI: 1.22, 1.56; OR: 1.55, 95% CI: 1.33, 1.81; OR: 0.82, 95% CI: 0.72, 0.94, respectively). No relationship was seen with LAZ or stunting (p>0.05). Using the 2×2 factorial variable, we found that as compared to children of mothers who reported optimal feeding (combination of high responsive and low forceful behaviors), children of mothers who reported using a combination of high force and low responsivity had lower WAZ (β: −0.18, 95% CI: −0.28, −0.09), WLZ (β: −0.24, 95% CI: −0.33, −0.15), and DDS (β: −0.19, 95% CI: −0.32, −0.06), and were more likely to be wasted (OR: 1.62, 95% CI: 1.28, 2.04), underweight (OR: 1.38, 95% CI: 1.14, 1.68) and less likely to achieve MDD (OR: 0.73, 95% CI: 0.71, 0.98). In the same way, children of mothers who reported using both high force and high responsivity had lower LAZ (β: −0.07, 95% CI: −0.14, −0.004), WAZ (β: −0.16, 95% CI: −0.25, −0.09), WLZ (β: −0.16, 95% CI: −0.23, −0.10), and DDS (β: −0.12, 95% CI: −0.22, −0.01) and were more likely to be wasted (OR: 1.43, 95% CI: 1.18, 1.73), underweight (OR: 1.36, 95% CI: 1.16, 1.59) and less likely to achieve MDD (OR: 0.82, 95% CI: 0.70, 0.96). Children of mothers who reported using a combination of low force and low responsivity had lower LAZ (β: −0.08, 95% CI: −0.16, −0.01) and were less likely to meet MDD (OR: 0.84, 95% CI: 0.71, 0.98). Non‐significant relations are not presented. Findings suggest that even in the context of high undernutrition, non‐responsive feeding behaviors (especially forceful) are associated with poor growth and dietary diversity among young children. Future longitudinal research is needed to better understand causality and test the efficacy of responsive feeding interventions on the promotion of child growth and diet.Support or Funding InformationFunded by USDA, NIFA, the Johns Hopkins Center for Global Health, the Bill and Melinda Gates Foundation, and the Johns Hopkins Sight and Life Global Nutrition Research Institute.
Read full abstract