Prevalence of child undernutrition measures and their spatio-demographic inequalities in Bangladesh: an application of multilevel Bayesian modelling

Abstract

Micro-level statistics on child undernutrition are highly prioritized by stakeholders for measuring and monitoring progress on the sustainable development goals. In this regard district-representative data were collected in the Bangladesh Multiple Indicator Cluster Survey 2019 for identifying localised disparities. However, district-level estimates of undernutrition indicators - stunting, wasting and underweight - remain largely unexplored. This study aims to estimate district-level prevalence of these indicators as well as to explore their disparities at sub-national (division) and district level spatio-demographic domains cross-classified by children sex, age-groups, and place of residence. Bayesian multilevel models are developed at the sex-age-residence-district level, accounting for cross-sectional, spatial and spatio-demographic variations. The detailed domain-level predictions are aggregated to higher aggregation levels, which results in numerically consistent and reasonable estimates when compared to the design-based direct estimates. Spatio-demographic distributions of undernutrition indicators indicate south-western districts have lower vulnerability to undernutrition than north-eastern districts, and indicate significant inequalities within and between administrative hierarchies, attributable to child age and place of residence. These disparities in undernutrition at both aggregated and disaggregated spatio-demographic domains can aid policymakers in the social inclusion of the most vulnerable to meet the sustainable development goals by 2030.

Keywords: Rural-urban disparities; Small area estimation; Spatial and cross-sectional correlations; Stunting; Underweight; Wasting.

Fig. 1

Standard errors (SE) and generalized variance function (GVF) model-based SEs are plotted against the corresponding estimates under original and square-root scales for stunting, wasting and underweight. Empty diamonds and filled circles refer to direct and GVF model-based SEs. Orange and sky-blue colours refer to rural and urban domains, respectively

Fig. 2

Prevalence of stunting, wasting and underweight for the cross-classified domains of children age-group, sex and place of residence

Fig. 3

Prevalence of stunting, wasting and underweight for the cross-classified domains of children age-group, sex, place of residence and division

Fig. 4

Distribution of estimated stunting, wasting and underweight (in %) with their coefficient of variation (CV in %) for the disaggregation level of district (D), district-sex (DS), district-age (DA), district-residence (DR), district-residence-sex (DRS), district-residence-age (DRA), and district-residence-age-sex (DRAS)

Fig. 5

Spatial distribution of district level stunting through univariate map (left), bivariate map (middle) and bivariate key for the prevalence estimates and their standard errors (right)

Fig. 6

Spatial distribution of district level wasting through univariate map (left), bivariate map (middle) and bivariate key for the prevalence estimates and their standard errors (right)

Fig. 7

Spatial distribution of district level underweight through univariate map (left), bivariate map (middle) and bivariate key for the prevalence estimates and their standard errors (right)

Fig. 8

District level map for the prevalence of stunting, wasting and underweight by the place of residence

Fig. 9

District level prevalence of stunting, wasting and underweight by the place of residence for Dhaka, Khulna and Sunamganj districts estimated by design-based (black circle dot) and model-based (coloured triangle dots) estimators (with 95% CI)

Fig. 10

District level prevalence of stunting, wasting and underweight by the children age-groups for Dhaka, Khulna and Sunamganj districts estimated by design-based (black circle dot) and model-based (coloured triangle dots) estimators (with 95% CI)

Fig. 11

Prevalence of stunting, wasting and underweight for cross-classified domains of children age-groups and place of residence for Dhaka, Khulna and Sunamganj districts estimated by design-based (black circle dot) and model-based (coloured triangle dots) estimators (with 95% CI)

Fig. 12

Prevalence of stunting for the detailed level domains of 8 districts as representative of 8 divisions estimated by design-based (black circle dot) and model-based (triangle coloured dots) estimators (with 95% CI)