Model-Based Prioritisation of Local Government Areas for Maternal Mortality Reduction in Nigeria

Abstract Background Nigeria continues to bear one of the world’s highest maternal and neonatal mortality burdens, with the 2018 NDHS estimating a maternal mortality ratio (MMR) of 512 per 100,000 live births and United Nations inter-agency estimates suggesting it may have exceeded 1,047 per 100,000 by 2020. Efficient allocation of scarce resources requires transparent and reproducible approaches for identifying high-burden Local Government Areas (LGAs). This study developed and validated a nationwide LGA-level prioritisation framework to support targeted interventions for Maternal and Neonatal Mortality Reduction Initiatives in Nigeria. Methods Four complementary approaches were implemented using routine health information, household survey, and operational data. These included: (1) ranking LGAs using four-year (2020–2023) cumulative and median routine maternal mortality ratios (MMR); (2) equal-weight composite risk scoring across multiple scenarios combining service-coverage, readiness, and mortality indicators; (3) projection of survey-derived service coverage indicators from 2018 to 2023; and (4) a survey-anchored mortality signal derived from pregnancy-related deaths using the 2018 sibling-history data. Cross-scenario convergence was used to identify consistently high-risk LGAs. Results Across methods, a consistent cluster of high-risk LGAs emerged, predominantly located in northern Nigeria. Routine MMR approaches identified substantial overlap in priority LGAs, while composite and projection analyses highlighted persistent service-coverage and readiness deficits. The survey-based pregnancy-related death approach produced a shortlist of 172 LGAs for initial programme implementation, closely aligning with patterns observed in the other methods. Conclusions Triangulation of routine, survey, and operational data provides a robust and defensible framework for subnational prioritisation of maternal and neonatal mortality interventions. This approach enables targeted investment in high-burden LGAs and offers a replicable model for evidence-based programme targeting in data-constrained settings.