What question did this study set out to answer?

The aim is to evaluate the effectiveness of machine learning models compared to traditional logistic regression in predicting childhood diarrhea in Nigeria.

March 13, 2026Open Access

Comparative performance of machine learning models in predicting childhood diarrhea: implications for public health surveillance

Key Points

The aim is to evaluate the effectiveness of machine learning models compared to traditional logistic regression in predicting childhood diarrhea in Nigeria.
Analyzed data from the 2018 NDHS involving 33,924 children under five.
Used logistic regression to estimate adjusted odds ratios for diarrhea risk factors.
Implemented machine learning models including CatBoost, LightGBM, and XGBoost for comparison.
Assessed model performance based on feature importance ranking and predictive ability.
CatBoost showed the highest performance with an AUC of 0.706 and sensitivity of 71.3%.
Logistic regression also performed well with an AUC of 0.704.
Child's age and region were identified as key predictors across both model types.
Higher socioeconomic status and safe feces disposal were linked to lower odds of diarrhea.

Abstract

Diarrhea remains one of the leading causes of under-5 mortality in Nigeria. Traditional logistic regression has been used to assess risk factors; however, machine learning (ML) models complement the prediction capability by untying tangled interactions. This study presents an analysis of childhood diarrhea predictors in Nigeria using both approaches to inform strategic interventions. The 2018 NDHS (Nigeria Demographic and Health Survey) data were assessed, and 33,924 under-5 children’s data were analyzed. Predictor variables include socioeconomic, environmental, behavioral, and household factors. Adjusted odds ratios (aORs) were estimated using logistic regression, while ML models (CatBoost, LightGBM, XGBoost) were used to evaluate non-linear relationships. Performance of the models was compared based on feature importance rankings. Notably, there was a striking regional variation, with the highest prevalence of diarrhea recorded in the North-East (aOR = 2.04, p < 0.001). Region and child’s age were the most important predictors in both regression and ML models, with ML importance of 100 and 76.1, respectively. Higher socioeconomic status had a strongly protective association, as evidenced by a clear wealth gradient (Very rich: aOR = 0.57, p < 0.001). Similarly, safe disposal of children’s feces (aOR = 0.92, p = 0.031) was associated with reduced odds of diarrhea, as was higher maternal education (aOR = 0.76, p = 0.010). Among the ML models, CatBoost provided the best balance of performance (AUC = 0.706, Balanced Accuracy = 0.660, sensitivity = 71.3%), suggesting its utility in ruling out the condition, although all models, including LR (AUC = 0.704), showed a good discriminative ability. This study emphasizes the complementarity of logistic regression and ML to identifying diarrhea predictors. The principal findings indicate a strategic combination of geographically focused WASH (Water, Sanitation, and Hygiene) interventions, age-stratified prevention, and poverty alleviation strategies. These findings can inform Nigeria’s interventions to reduce diarrhea in children and achieve SDG 3.2.

Bookmark

View Full Paper

Cite This Study

Ashaolu et al. (Tue,) studied this question.

synapsesocial.com/papers/69b3ac3f02a1e69014ccdd01 https://doi.org/https://doi.org/10.1186/s42506-026-00212-7

Bookmark

View Full Paper