Associations Between Historically Redlined Districts and Racial Disparities in Current Obstetric Outcomes

IMPORTANCE: Despite much higher health care expenditure than comparable countries, striking racial and ethnic disparities exist in obstetric outcomes in the United States. A multifaceted exploration of the factors influencing these disparities, including the legacy of structural racism, is important to improve health outcomes for all. OBJECTIVE: To characterize the association of the historic racially discriminatory home loan practice of redlining with disparities in modern obstetric outcomes. DESIGN, SETTING, AND PARTICIPANTS: In this retrospective cohort study of a 9-county birth certificate database in the Finger Lakes region of New York state from 2005 to 2018, modern obstetric outcomes were matched with regions classified by the federal government for mortgage loan servicing based on racially discriminatory criteria from the 1940 Home Owners’ Loan Corporation map (HOLC; also known as the redline map). Patients with a live birth recorded in the data system with a recorded home zip code within the historic HOLC categories were included. Data were analyzed from July to December 2019. EXPOSURE: Regions previously categorized by historic, racially discriminatory criteria. MAIN OUTCOMES AND MEASURES: Each HOLC area was analyzed for the primary outcome of preterm birth and secondary outcomes of obstetric and medical complications, with logistic regression to address regional and patient-level covariates. RESULTS: From 2005 until 2018, there were 64 804 live births within the 15 zip codes overlaying historic HOLC regions. Prevalence of preterm birth increased with decreasing HOLC categories, from the lowest overall preterm birth rate of 217 of 2873 births (7.55%) in the zip code historically defined as “Best” or “Still Desirable” and the highest overall preterm birth rate of 427 of 3449 births (12.38%) in the zip code historically defined as “Hazardous.” These associations with preterm birth remained significant in logistic regression controlling for poverty levels and educational attainment (adjusted odds ratio, 1.46; 95% CI, 1.08-1.97) and parental race (adjusted odds ratio, 1.38; 95% CI, 1.25-1.53). CONCLUSIONS AND RELEVANCE: In this cohort study, the linkage of historic and modern community data sets with an obstetric data set offered the opportunity to characterize modern obstetric disparities associated with a system of historic inequity. The persistence of these findings after correcting for contemporary community socioeconomic characteristics suggest potential influences of a system of profound structural inequity that ripple forward in time, with impacts that extend beyond measurable socioeconomic inequity.