Insights from Preventability Assessments Across 42 State and City Maternal Mortality Reviews in the United States.

Abstract

Background: The rising trend in maternal mortality over the past three decades sets the United States (U.S.) apart from all other high-income countries. Multidisciplinary state and city Maternal Mortality Review Committees (MMRCs) conduct comprehensive reviews of maternal deaths, including assessments of preventability and contributing factors.

Objective(s): Assess preventability of and contributing factors to maternal mortality in the U.S.

Study design: This study is a secondary analysis of cross-sectional, population-based data from the most recent, publicly available MMRC data from 40 state and 2 cities in the U.S. Preventability was analyzed among all deaths during pregnancy or within one year postpartum from any cause (pregnancy-associated deaths, PAD) and deaths during pregnancy or within one year postpartum from causes related to pregnancy or its management, but not from accidental causes (pregnancy-related deaths, PRD). We also explored preventability by cause-of-death and contributing factors grouped as community, patient-family, provider, facility, and health system factors.

Results: Of deaths that occurred after 2010, between 53%-93.8% of PADs and 45%-100% of PRDs were deemed preventable across the 42 states and cities. Across the ten states reporting PRD preventability by cause-of-death, MMRCs deemed preventable >90% of deaths from preeclampsia-eclampsia and mental health conditions, >80% of deaths from hemorrhage and cardiovascular conditions, about 70% of deaths from infection and thrombotic embolism, and about 40% of deaths from amniotic fluid embolism and stroke. A total of 3,345 contributing factors were described in MMRC reports from 14 states in relation to 739 PRDs. While collectively patient-family and provider factors were most frequently noted as contributing to PRDs, the contribution of such factors varied between 6%-56% and 18%-42.3%, respectively, across the states. Based on data from 20 MMRCs with available information, racism or discrimination were noted in relation to 37.7% of PRDs.

Conclusions: A large proportion of PADs and PRDs in the U.S. are preventable. However, likely due to differences in MMRC membership, available data, and judgement employed to determine preventability, wide variation exists in the proportion of deaths deemed preventable and factors identified as contributing to such deaths across states. There is need to reevaluate the definitions, structure, and outputs for maternal death preventability assessments currently employed by a majority MMRCs to adequately inform state and national programming and policies.