Key points are not available for this paper at this time.
INTRODUCTION: The objective of this study was to identify clusters in the United States with the highest and lowest rates of maternal mortality, to pinpoint changes in maternal mortality rates, and to forecast maternal deaths by demographic variables. METHODS: A retrospective cohort study was used to report maternal deaths in the United States. The maternal mortality counts (as defined by ICD-10 codes) from 2010 to 2020 at the county level for the 48 contiguous states were obtained from the Centers for Disease Control and Prevention (CDC) via their Wonder application. Variables such as population density, environmental conditions, race, ethnicity, and physical and mental health conditions for each county were identified. A regression analysis and spatial cluster analysis of these variables generated a heat map to identify areas in the United States with high and low rates of maternal mortality. RESULTS: Spatial clusters in the 48 contiguous states averaged across all years were identified. There are discrete areas with higher maternal mortality, some broad, such as the Southeast region, and some specified to cities, such as New York. A regression model used multiple factors to predict maternal deaths from demographic variables. The four best predictors were high population density, air pollution, and Hispanic and African American ethnicities. Clusters were also identified with the lowest maternal mortality. CONCLUSION: The United States has the highest maternal mortality rate (MMR) among all high-income countries. Historically, MMR has steadily increased since 2016. This study found that the best predictors for maternal mortality were high population density, air pollution, and Hispanic and African American subpopulations. Such a study can be used by health agencies and beyond to delve into possible causes of elevated maternal mortality in certain geographic regions of the United States. More studies are indicated to determine causality and unearth strategies to improve maternal outcomes.
Solomon et al. (Wed,) studied this question.