Seasonal variation in the association between household food insecurity and child undernutrition in Bangladesh: Mediating role of child dietary diversity

Abstract

Household food insecurity (HFI) and child dietary diversity (CDD) are variable across seasons. We examined seasonal variation in HFI and child undernutrition association and tested how CDD mediates this association. We analyzed data for 26,353 children aged 6-59 months drawn from nationally representative cross-sectional Food Security and Nutrition Surveillance Project data collected during 2012-2014 in Bangladesh across three seasons annually: Post-Aman harvest (January-April); Monsoon (May-August); and Post-Aus harvest (September-December). Multivariable logistic regression analysis adjusted for individual, maternal, household and geographical characteristics reveals that children of food-insecure households were more likely than food-secure households to be stunted (adjusted odds ratio, AOR: 1.12; 95% confidence interval, CI: 1.02-1.23; p < 0.05), wasted (AOR: 1.21; 95% CI: 1.05-1.39; p < 0.01) and underweight (AOR: 1.16; 95% CI: 1.04-1.3; p < 0.01). CDD mediated 6.1% of the total effect of HFI on underweight. These findings varied across seasons. HFI was associated with greater odds of underweight during Monsoon (AOR: 1.32; 95% CI: 1.08-1.62; p < 0.01) and Post-Aus (AOR: 1.21; 95% CI: 1.06-1.37; p < 0.01) while wasting during Post-Aus (AOR: 1.65; 95% CI: 1.35-2.01; p < 0.001). CDD largely mediated the total effect of HFI on underweight during the Post-Aman in 2012-2014 (23.2%). CDD largely mediated the total effect of HFI on wasting (39.7%) during Post-Aman season in 2014 and on underweight (13.7%) during the same season in 2012. These findings demonstrate that HFI is seasonally associated with child undernutrition and mediated by CDD as well in Bangladesh and seasonality and diversity should be considered while designing appropriate population-level food-based interventions to resolve child undernutrition.

Keywords: Bangladesh; child dietary diversity; food insecurity; nutrition; seasonality.

Figure 1

Relationship between household food insecurity, child dietary diversity and child undernutrition

Figure 2

Prevalence of undernutrition and consumption of minimum diversified diet among children 6–59 months of age by household food insecurity in Bangladesh during 2012–2014. ***p < 0.001, **p < 0.01 and *p < 0.05.

Figure 3

Association between household food insecurity and child undernutrition in Bangladesh during 2012–2014. ***p < 0.001, **p < 0.01 and *p < 0.05. All the models were separately fitted at an overall level and for all three seasons. All the models were adjusted for the age of child in months, child sex, child morbidity (acute respiratory infections and diarrhea), mother's education, maternal age at birth, maternal stature, household size, wealth quintiles of households, place of residence (rural and urban) and division. AOR, adjusted odds ratio; CI, confidence interval.