Abstract / Summary
Background: COVID-19 increased maternal risk through infection and disruption of essential services, but maternal mortality was already changing before 2020, making simple pre-pandemic comparisons difficult. Objectives: To estimate country-level departures in maternal deaths and maternal mortality ratios (MMR) during 2020-2023 from projected pre-pandemic trajectories and examine whether maternal health-service disruptions were associated with MMR deviations. Methods: We fitted country-specific counterfactual models to WHO/UN MMEIG estimates from 2000-2019 and projected maternal deaths and MMR through 2023 under a no-pandemic scenario. Excess was defined as the positive difference between the MMEIG estimate and counterfactual value. In a self-selected subset of 39 countries (63 country-year observations), we modelled log(MMEIG-estimated MMR/counterfactual MMR) as a function of three service-disruption measures and four baseline or year covariates. Results: Across 195 countries and territories, cumulative excess maternal deaths were estimated at 68,489 (95% UI 34,706, 147,118); 113 had statistically detectable cumulative excess. The largest regional burdens were in South-East Asia, the Eastern Mediterranean, and Africa. In the 39-country subset, antenatal care, facility-based delivery, and postnatal care disruption were not clearly associated with signed MMR deviation in the full or separate models. No single country materially changed the estimated coefficients in leave-one-country-out analyses, and excluding Namibia, which had the two most unusual observations, did not alter the conclusions. Conclusions: Maternal mortality departed substantially but unevenly from pre-pandemic trajectories. The absence of associations with broad national disruption measures should not be interpreted as evidence that service disruption was unimportant; it instead highlights the need for more detailed measures of service access and quality, alongside protection of the maternal-care continuum during emergencies.