Spatial Pattern Analysis and Determinants of Stunting Prevalence in Central Sulawesi, Indonesia: Using Linear Regression, Local Moran’s I, and Random Forest Approaches

Background: Stunting remains a significant public health issue in Indonesia, particularly in Central Sulawesi, where socio-economic and environmental factors contribute to its prevalence. Understanding these determinants is crucial for effective intervention strategies. Objective: This study aims to analyze the spatial distribution and predictors of stunting prevalence in Central Sulawesi, focusing on socio-economic and environmental factors. Methods: An observational design was employed, utilizing secondary data from the Central Sulawesi Provincial Health Department. Spatial analysis, including Moran’s I and Local Moran’s I, assessed spatial autocorrelation and identified outliers. Regression analysis and Random Forest modeling examined predictors of stunting prevalence. Results: The study found significant spatial clustering in stunting prevalence. Key socio-economic factors identified were maternal education and household income, with poverty being the most influential predictor. Random Forest analysis highlighted sanitation and access to health facilities as important, although access to clean water did not show a significant effect. Conclusion: The findings provide valuable insights into the socio-economic determinants of stunting and emphasize the need for targeted, comprehensive intervention strategies focusing on improving maternal education and addressing poverty, along with enhancing healthcare access in Central Sulawesi