Examining the relationship between armed conflict and coverage of maternal and child health services in 35 countries in sub-Saharan Africa: a geospatial analysis.

Background: Armed conflict is on the rise in sub-Saharan Africa and affects public infrastructures, including health systems, although evidence on population health is sparse. We aimed to establish how these disruptions ultimately affect health service coverage.

Methods: We geospatially matched Demographic and Health Survey data with the Uppsala Conflict Data Program Georeferenced Events Dataset, covering 35 countries for the period from 1990 to 2020. We relied on linear probability models with fixed effects to capture the effect of nearby armed conflict (within 50 km of the survey cluster) on four service coverage indicators along the continuum of maternal and child health care. We also investigated effect heterogeneity by varying conflict intensity and duration, and sociodemographic status.

Findings: The estimated coefficients represent the decrease in the probability (in percentage points) of the child or their mother being covered by the respective health service following deadly conflicts within 50 km. Any nearby armed conflict was associated with reduced coverage for all examined health services, with the exception of early antenatal care: early antenatal care (-0·5 percentage points, 95% CI -1·1 to 0·1), facility-based delivery (-2·0, -2·5 to -1·4), timely childhood vaccination (-2·5, -3·1 to -1·9), and treatment of common childhood illnesses (-2·5, -3·5 to -1·4). For all four health services, the negative effects increased for high-intensity conflicts and were significant throughout. When examining conflict duration, we did not find negative effects on the treatment of common childhood illnesses in prolonged conflicts. The analysis on effect heterogeneity revealed that, except for timely childhood vaccination, the negative effects of armed conflict on health service coverage were more pronounced in urban settings.

Interpretation: Our findings suggest that health service coverage is significantly affected by contemporaneous conflict, but health systems can adapt to provide routine services, such as child curative services, in situations of prolonged conflict. Our analysis underlines the importance of studying health service coverage during conflict both at the finest possible scales and across different indicators, pointing at the need for differential policy interventions.

Funding: None.

Translations: For the French and Portuguese translations of the abstract see Supplementary Materials section.