Modelling Acute Malnutrition in Ethiopia: A Bayesian Distributional Regression Approach

Author(s): Fekade Getabil Habtewold, Butte Gotu Arero.

Child wasting, a severe form of acute malnutrition remains a pressing public health concern in Ethiopia. This study aimed to identify key determinants influencing child wasting, measured through Weight-for-Height Z-scores (WHZ), and to explore spatial variations using a Bayesian Distributional Regression (BDR) framework. By analyzing data from the 2019 Ethiopian Demographic and Health Survey (EDHS), the study examined the complex relationships between socio-economic, demographic, and geographical factors affecting WHZ. Unlike traditional regression approaches that focus solely on mean effects, BDR enabled a comprehensive assessment of the full distribution of WHZ, offering deeper insights into nutritional disparities. The results indicated notable regional differences in WHZ, with the Somali region showing the lowest values. BDR analysis highlighted maternal education and household wealth as crucial factors positively influencing mean WHZ. Conversely, child age and malaria incidence were negatively associated with WHZ, indicating their detrimental effects on child nutrition. Environmental factors also played a role, as higher mean temperatures were linked to lower WHZ and increased variability. Further, the sigma model indicated that maternal education and moderate household wealth contributed to reducing WHZ variability, suggesting their stabilizing effects on child nutrition outcomes. Additionally, non-linear relationships were observed in the effects of child age, mean temperature, precipitation, and population density, emphasizing the complexity of these interactions. Model fit diagnostics confirmed the reliability of the analysis. The study underscores the multifaceted nature of child wasting in Ethiopia, driven by socio-economic, health, and environmental determinants. Addressing these factors through targeted interventions is essential to enhance child nutritional outcomes. Future research should focus on understanding causal pathways and evaluating the effectiveness of proposed interventions to ensure sustainable solutions for combating child malnutrition.