Publications
Article Results BAnner
Back to browse results
Spatial heterogeneity and correlates of child malnutrition in districts of India
Authors: Khan J, and Mohanty SK
Source: BMC Public Health, 18(1):1027; DOI: 10.1186/s12889-018-5873-z
Topic(s): Child health
Nutrition
Spatial analysis
Country: Asia
  India
Published: AUG 2018
Abstract: BACKGROUND: Despite sustained economic growth and reduction in money metric poverty in last two decades, prevalence of malnutrition remained high in India. During 1992-2016, the prevalence of underweight among children had declined from 53% to 36%, stunting had declined from 52% to 38% while that of wasting had increased from 17% to 21% in India. The national average in the level of malnutrition conceals large variation across districts of India. Using data from the recent round of National Family Health Survey (NFHS), 2015-16 this paper examined the spatial heterogeneity and meso-scale correlates of child malnutrition across 640 districts of India. METHODS: Moran's I statistics and bivariate LISA maps were used to understand spatial dependence and clustering of child malnutrition. Multiple regression, spatial lag and error models were used to examine the correlates of malnutrition. Poverty, body mass index (BMI) of mother, breastfeeding practices, full immunization, institutional births, improved sanitation and electrification in the household were used as meso scale correlates of malnutrition. RESULTS: The univariate Moran's I statistics was 0.65, 0.51 and 0.74 for stunting, wasting and underweight respectively suggesting spatial heterogeneity of malnutrition in India. Bivariate Moran's I statistics of stunting with BMI of mother was 0.52, 0.46 with poverty and?-?0.52 with sanitation. The pattern was similar with respect to wasting and underweight suggesting spatial clustering of malnutrition against the meso scale correlates in the geographical hotspots of India. Results of spatial error model suggested that the coefficient of BMI of mother and poverty of household were strong and significant predictors of stunting, wasting and underweight. The coefficient of BMI in spatial error model was largest found for underweight (ß?=?0.38, 95% CI: 0.29-0.48) followed by stunting (ß?=?0.23, 95% CI: 0.14-0.33) and wasting (ß?=?0.11, 95% CI: 0.01-0.22). Women's educational attainment and breastfeeding practices were also found significant for stunting and underweight. CONCLUSION: Malnutrition across the districts of India is spatially clustered. Reduction of poverty, improving women's education and health, sanitation and child feeding knowledge can reduce the prevalence of malnutrition across India. Multisectoral and targeted intervention in the geographical hotspots of malnutrition can reduce malnutrition in India. KEYWORDS: India; Malnutrition; Spatial heterogeneity; Stunting; Underweight; Wasting
Web: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6098604/