abstract
- In Mexico, the comparative effectiveness of budget allocation versus technical water treatment performance in reducing childhood diarrheal mortality remains unestablished. We conducted a nationwide predictive analysis of under-five diarrheal mortality across all 32 Mexican states from 2012 to 2022, integrating data from Mexican National Water Commission (CONAGUA) through their National Water Information System (SINA) on government budget allocation and chlorination efficiency. We employed Generalized Additive Models (GAM) with rolling-window cross-validation to quantify the relative importance of technical versus financial water management interventions. The analysis achieved robust predictive performance (mean R2 = 0.74, MAE = 2.11 deaths per 100,000 children). Chlorination efficiency demonstrated superior predictive power (55.7% relative importance) compared to budget allocation (44.3% relative importance) in reducing child mortality. Southern states consistently showed both the highest mortality rates (Chiapas: 45.0 per 100,000 in 2022) and lowest chlorination efficiency (<80%), while northern states maintained low mortality with high technical performance (>95% efficiency). Technical water quality indicators outperform budget-based metrics in predicting child mortality outcomes. Our methodology provides a replicable framework for evidence-based water policy optimization in similar middle-income contexts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.