Regional Variations in Maternal Mortality and Health Workforce Availability in the United States

Ann Intern Med. 2020 Dec 1;173(11 Suppl):S45-S54. doi: 10.7326/M19-3254.

Abstract

Background: Skilled, high-quality health providers and birth attendants are important for reducing maternal mortality.

Objective: To assess whether U.S. regional variations in maternal mortality rates relate to health workforce availability.

Design: Comparison of regional variations in maternal mortality rates and women's health provider rates per population and identification of a relationship between these measures.

Setting: U.S. health system.

Participants: Women of child-bearing age and women's health providers, as captured in federal data sources from the Centers for Disease Control and Prevention, Census Bureau, and Health Resources and Services Administration.

Measurements: Regional-to-national rate ratios for maternal mortality and women's health provider availability, calculated per population for women of reproductive age. Provider availability was examined across occupations (obstetrician-gynecologists, internal medicine physicians, family medicine physicians, certified nurse-midwives), in service-based categories (birth-attending and primary care providers), and across the entire women's health workforce (all studied occupations).

Results: Maternal deaths per population increased nationally from 2009 to 2017 and, in 2017, were significantly higher in the South and lower in the Northeast (P < 0.001) than nationally. The occupational composition and per-population availability patterns of the women's health workforce varied regionally in 2017. The South had the lowest availability in the nation for nearly every health occupation and category studied, and the Northeast had the highest. This exploratory analysis suggests that subnational levels of provider availability across a region may be associated with higher maternal mortality rates.

Limitations: No causal relationship was established. Nationally representative maternal mortality and health workforce data sources have well-known limitations. Low numbers of observations limit statistical analyses.

Conclusion: Regional variations in maternal mortality rates may relate to the availability of birth-attending and primary care providers.

Primary funding source: None.

MeSH terms

  • Adolescent
  • Adult
  • Female
  • Health Personnel / statistics & numerical data*
  • Humans
  • Maternal Mortality*
  • Middle Aged
  • Obstetrics / statistics & numerical data
  • Pregnancy
  • United States / epidemiology
  • Women's Health Services / statistics & numerical data*
  • Young Adult