Spatial distribution and socio-demographic determinants of adolescent malnutrition in India using comprehensive national nutrition survey

The Present study has a two-fold objective; firstly, to examine the prevalence of adolescent malnutrition including underweight, stunting, and overweight/obesity and its socio-demographic determinants across Indian states; and secondly, to assess scenario-based future risk prediction of malnutrition among adolescents with particular emphasis on regional disparities across four distinct zones of India classified to understand spatial distribution of malnutrition. We used Comprehensive National Nutrition Survey (CNNS, 2016–18) individual-level data; study analyzed a sample of 35,830 adolescents aged 10–19 years. The burden of adolescent malnutrition was estimated for the entire country by dividing all Indian States in four zones based strictly on the official sampling classifications established by the CNNS framework. Multivariable logistic regression was employed to examine the socio-demographic factors associated with malnutrition including underweight, stunting, and overweight/obesity. Scenario-based zone-wise risk predictions were generated by calculating marginal predicted probabilities from the fitted regression models to estimate how hypothetical changes in key socio-demographic determinants would impact adolescent malnutrition levels across the four distinct zones. The overall prevalence of underweight was highest in zone 4 (30.25%), while stunting was highest in zone 3 (33.29%), spatial disparities that are primarily driven by lower educational level and wealth Overweight/obesity emerged as a growing concern in zone 4 (36.90%), particularly among richer households. The risk prediction map revealed that zone 1 has the highest predicted risk (> 0.4) due to high levels of poverty, limited access to healthcare, and persistent rural–urban disparities, followed by zone 3 (0.3–0.4), while zone 4 exhibited lowest predicted risk (< 0.2). Zone-specific risks highlight the need for targeted, region-sensitive interventions-prioritizing underweight in north and central part of the country, stunting in eastern part of the nation, and overweight/obesity in the southern part of the country. The integration of predictive analytics in public health planning can significantly improve policy responsiveness and resource allocation.