Child Mortality Prediction in Bangladesh: A Decade-Long Validation Study

TL;DR

This study validates and compares machine learning models for child mortality prediction in Bangladesh over a decade, revealing socioeconomic biases and identifying the most effective neural architecture for targeted health interventions.

Contribution

It introduces a novel neural network architecture optimized via genetic algorithms and demonstrates its superiority over XGBoost in predicting child mortality, accounting for socioeconomic disparities.

Findings

Neural network outperformed XGBoost with AUROC 0.76 vs. 0.73.

Model performance correlated with regional poverty levels.

Approximately 1300 additional at-risk children identified annually.

Abstract

The predictive machine learning models for child mortality tend to be inaccurate when applied to future populations, since they suffer from look-ahead bias due to the randomization used in cross-validation. The Demographic and Health Surveys (DHS) data from Bangladesh for 2011-2022, with n = 33,962, are used in this paper. We trained the model on (2011-2014) data, validated it on 2017 data, and tested it on 2022 data. Eight years after the initial test of the model, a genetic algorithm-based Neural Architecture Search found a single-layer neural architecture (with 64 units) to be superior to XGBoost (AUROC = 0.76 vs. 0.73; p < 0.01). Additionally, through a detailed fairness audit, we identified an overall "Socioeconomic Predictive Gradient," with a positive correlation between regional poverty level (r = -0.62) and the algorithm's AUC. In addition, we found that the model performed at its highest levels in the least affluent divisions (AUC 0.74) and decreased dramatically in the wealthiest divisions (AUC 0.66). These findings suggest that the model is identifying areas with the greatest need for intervention. Our model would identify approximately 1300 additional at-risk children annually than a Gradient Boosting model when screened at the 10% level and validated using SHAP values and Platt Calibration, and therefore provide a robust, production-ready computational phenotype for targeted maternal and child health interventions.