Machine-Learning Implications on Global Antenatal Care Improvement and Equity Disparities: The Wealth Effect, Regional Dynamics and Model-Based

Francis Ayiah-Mensah *

Department of Mathematics, Statistics and Actuarial Science, Takoradi Technical University, Sekondi-Takoradi, Ghana.

Emmanuel Mensah Baah

Department of Mathematics, Statistics and Actuarial Science, Takoradi Technical University, Sekondi-Takoradi, Ghana.

Emmanuel Harris

Department of Statistics and Actuarial Science, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana.

Samuel Kwame Okai

Department of Mathematics, Statistics and Actuarial Science, Takoradi Technical University, Sekondi-Takoradi, Ghana.

*Author to whom correspondence should be addressed.


Abstract

The study evaluated the progress of antenatal care worldwide using records from the UNICEF Maternal Health Dataset and determined the correlation between prosperity and improvements in antenatal care coverage. Mean coverage of antenatal care rose by 61.9 in 2013 to 71.6 in 2023, and the best-performing quartile improved by the greatest margin of +14.50 percentage points, with the best performance being in rural areas, which improved by +12.73 points. The relationship between wealth and baseline coverage change was found to be significantly affected by wealth, with main effects of wealth and the interaction. This was further confirmed by a nonparametric mediation statistic, which showed a positive correlation between wealth and baseline coverage, but the indirect effect of wealth was not significant; therefore, wealth does not mediate the change. Among the six machine learning models, the XGBoost model had the best predictive accuracy, followed by LightGBM and Multivariate Adaptive Regression Splines (MARS). The hot spot mapping showed that South Asia and East and Southern Africa had the highest rates of pro-poor progression. The findings offer a novel way to holistically assess changes in antenatal care, providing evidence-based interventions to hasten pro-equity developments towards achieving the third Sustainable Development Goal of Good Health. All the respective National Programs should employ machine-learning-focused early warning analytics to detect geographic regions with stagnation, and offer direct outreach services and preparedness of the Community and Services.

Keywords: Maternz health, global health equity, wealth gradient, moderation analysis, mediation analysis, health coverage change


How to Cite

Ayiah-Mensah, Francis, Emmanuel Mensah Baah, Emmanuel Harris, and Samuel Kwame Okai. 2025. “Machine-Learning Implications on Global Antenatal Care Improvement and Equity Disparities: The Wealth Effect, Regional Dynamics and Model-Based”. Asian Journal of Pregnancy and Childbirth 8 (1):527-42. https://doi.org/10.9734/ajpcb/2025/v8i1187.

Downloads

Download data is not yet available.