Maternal health service coverage and health financing in relation to infant mortality in Africa

Infant mortality remains an important public health challenge across Africa despite improvements in maternal and child health. Maternal healthcare coverage, health financing, socioeconomic conditions, and environmental factors may jointly shape country-level mortality patterns, but their associations may differ across modelling approaches. This study examined these relationships across African countries using complementary econometric, nonlinear, and predictive methods. The study used a balanced analytical panel constructed through interpolation and endpoint carrying of missing observations for 30 African countries over 2000–2024 ( N = 750). Infant mortality was the outcome, while skilled birth attendance, antenatal care, current health expenditure per capita, GDP per capita, female secondary enrolment, urbanisation, and basic sanitation were examined as explanatory variables. Two-way fixed-effects estimation with Driscoll–Kraay standard errors provided the principal specification. A parsimonious pooled cross-sectionally augmented error-correction model, a bootstrap PPP-adjusted panel threshold analysis, sensitivity tests, and temporally ordered XGBoost with Shapley Additive Explanations (SHAP) provided complementary evidence. In the baseline model, skilled birth attendance was associated with lower infant mortality (β = −0.0028, p = 0.001), while higher GDP per capita and sanitation coverage were also negatively associated with mortality; urbanisation showed a positive association. Current health expenditure and antenatal care were not statistically significant at the 5% level. The dynamic model showed a significant negative short-run association for health expenditure but no statistically significant individual long-run coefficients. The PPP-adjusted threshold analysis identified an exploratory candidate of Int$109.74 per capita, but the bootstrap test did not support a statistically significant threshold ( p = 0.563). XGBoost achieved an out-of-sample R² of 0.773 for the 2019–2023 temporal holdout, declining to 0.332 under leave-one-country-out spatial-temporal validation. SHAP analysis identified sanitation as the strongest predictor of mortality. Sensitivity analyses showed that several coefficient directions were stable to country exclusion and removal of 2024, but significance weakened under country-specific trends and first-difference specifications. Infant mortality across African countries is associated with an interconnected set of maternal-health, socioeconomic, financing, and environmental conditions, but the strength of these relationships is model-dependent. The evidence does not support a universal health-expenditure threshold. Policies should therefore emphasise context-sensitive improvements in maternal healthcare access, effective health financing, sanitation, and broader socioeconomic conditions rather than a single financing benchmark.

Authors

Institutions

Publication Details

Journal
Discover Public Health
Published
2026-11-18
DOI
https://doi.org/10.1186/s12982-026-03343-6
Primary Topic
Global Maternal and Child Health
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Maternal health service coverage and health financing in relation to infant mortality in Africa

Okechukwu Alexander Okeke
1 links
Discover Public Health
Global Maternal and Child Health
article

Maternal health service coverage and health financing in relation to infant mortality in Africa

Okechukwu Alexander Okeke
article en

Abstract

Infant mortality remains an important public health challenge across Africa despite improvements in maternal and child health. Maternal healthcare coverage, health financing, socioeconomic conditions, and environmental factors may jointly shape country-level mortality patterns, but their associations may differ across modelling approaches. This study examined these relationships across African countries using complementary econometric, nonlinear, and predictive methods. The study used a balanced analytical panel constructed through interpolation and endpoint carrying of missing observations for 30 African countries over 2000–2024 ( N = 750). Infant mortality was the outcome, while skilled birth attendance, antenatal care, current health expenditure per capita, GDP per capita, female secondary enrolment, urbanisation, and basic sanitation were examined as explanatory variables. Two-way fixed-effects estimation with Driscoll–Kraay standard errors provided the principal specification. A parsimonious pooled cross-sectionally augmented error-correction model, a bootstrap PPP-adjusted panel threshold analysis, sensitivity tests, and temporally ordered XGBoost with Shapley Additive Explanations (SHAP) provided complementary evidence. In the baseline model, skilled birth attendance was associated with lower infant mortality (β = −0.0028, p = 0.001), while higher GDP per capita and sanitation coverage were also negatively associated with mortality; urbanisation showed a positive association. Current health expenditure and antenatal care were not statistically significant at the 5% level. The dynamic model showed a significant negative short-run association for health expenditure but no statistically significant individual long-run coefficients. The PPP-adjusted threshold analysis identified an exploratory candidate of Int$109.74 per capita, but the bootstrap test did not support a statistically significant threshold ( p = 0.563). XGBoost achieved an out-of-sample R² of 0.773 for the 2019–2023 temporal holdout, declining to 0.332 under leave-one-country-out spatial-temporal validation. SHAP analysis identified sanitation as the strongest predictor of mortality. Sensitivity analyses showed that several coefficient directions were stable to country exclusion and removal of 2024, but significance weakened under country-specific trends and first-difference specifications. Infant mortality across African countries is associated with an interconnected set of maternal-health, socioeconomic, financing, and environmental conditions, but the strength of these relationships is model-dependent. The evidence does not support a universal health-expenditure threshold. Policies should therefore emphasise context-sensitive improvements in maternal healthcare access, effective health financing, sanitation, and broader socioeconomic conditions rather than a single financing benchmark.

Discover Public Health
University of Port Harcourt (NG)
Good health and well-being, Reduced inequalities
Global Maternal and Child Health
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon