Peer-reviewed article
Unveiling predictive factors for household-level stunting in India: A machine learning approach using NFHS-5 and satellite-driven data.
Prashant Kumar Arya, Koyel Sur, Tanushree Kundu, Siddharth Dhote, Shailendra Kumar Singh ·
Summary
Written by AI, not yet AICS's view · claude-haiku-4-5-20251001This study used machine learning methods to identify factors that predict childhood stunting at the household level in India, where about 35% of children under 5 are stunted. Researchers combined data from the National Family Health Survey with satellite information about climate and vegetation. They found that poverty, household structure, parental education, and environmental factors like temperature and vegetation were the strongest predictors of stunting, with significant geographic clustering in central and eastern India.
Key points
- Approximately 35% of children under 5 in India are affected by stunting
- Random forest regression demonstrated the highest predictive accuracy among four machine learning models tested
- Proportion of households below the poverty line and dependency ratio consistently predicted stunting across all models
- Environmental variables including average temperature and leaf area index emerged as strong predictors of stunting
- Significant geographic clustering of high-stunting districts was found in central and eastern India
Limitations
This study identifies predictive associations between factors and stunting but does not test interventions or establish causal relationships. The analysis is observational and regional patterns may reflect unmeasured social or health system differences rather than direct environmental causation.
Abstract
OBJECTIVES: Childhood stunting remains a significant public health issue in India, affecting approximately 35% of children under 5. Despite extensive research, existing prediction models often fail to incorporate diverse data sources and address the complex interplay of socioeconomic, demographic, and environmental factors. This study bridges this gap by employing machine learning methods to predict stunting at the household level, using data from the National Family Health Survey combined with satellite-driven datasets. METHODS: We used four machine learning models-random forest regression, support vector machine regression, K-nearest neighbors regression, and regularized linear regression-to examine the impact of various factors on stunting. The random forest regression model demonstrated the highest predictive accuracy and robustness. RESULTS: The proportion of households below the poverty line and the dependency ratio consistently predicted stunting across all models, underscoring the importance of economic status and household structure. Moreover, the educational level of the household head and environmental variables such as average temperature and leaf area index were significant contributors. Spatial analysis revealed significant geographic clustering of high-stunting districts, notably in central and eastern India, further emphasizing the role of regional socioeconomic and environmental factors. Notably, environmental variables like average temperature and leaf area index emerged as strong predictors of stunting, highlighting how regional climate and vegetation conditions shape nutritional outcomes. CONCLUSIONS: These findings underline the importance of comprehensive interventions that not only address socioeconomic inequities but also consider environmental factors, such as climate and vegetation, to effectively combat childhood stunting in India.