Spatially Varying Regression Coefficient Model For Predicting Stunting Hotspots In Indonesia

Abstract

Stunting is a significant issue, particularly in the context of Indonesia. Identifying crucial risk factors is crucial for mitigating and developing effective strategies to control stunting. A Bayesian approach was employed to develop a regression model that incorporates spatial variation, allowing risk factors to vary across different districts and cities. The aim was to obtain the most optimal regression model. The analysis revealed that the impact of immunization varies across districts and cities in Indonesia when it comes to explaining the differences in stunting prevalence. The hotspot prediction results indicate that most urban districts in Indonesia remain hotspot areas, with a stunting risk exceeding 20%. The government must ensure the effective implementation of the immunization program in order to mitigate the prevalence of stunting in Indonesia. The novelty of this research lies in the use of Bayesian approaches to spatial analysis in identifying and understanding stunting risk factors as well as the prediction of stunting hotspots in Indonesia. This approach provides in-depth insight into local variations in the prevalence of stunting and the effectiveness of health interventions, which supports more effective and targeted policy development.