Associations Between Measures of Structural Racism and Acute Ischemic Stroke Incidence in the United States
Sai Polineni1, Praneet Polineni2, Daniel Santos3
1Icahn School of Medicine at Mount Sinai, New York, NY.
Background And Objectives:
Racial differences in socioeconomic characteristics are believed to be secondary to structural racism. While socioeconomic factors explain some of the racial disparity in stroke incidence at an individual level, little is known about the association between societal-level structural racism and incidence of acute ischemic stroke (AIS). We aimed to determine whether the geographic racial disparity in stroke incidence across the United States is associated with structural racism.
Methods:
We performed a national, population-based analysis of 71,078,619 adults (844,406 with incident AIS) aged 65 years and older who were enrolled in Medicare from January 1, 2016, to December 31, 2019. The primary exposure was a composite score calculated from 8 county-level measures of structural racism (segregation indices [delta, dissimilarity, isolation], Gini index, housing discrimination, educational attainment, employment, and income) that account for validated domains of structural racism based on an ecosocial model. The primary outcome was incident AIS. Marginal Cox models with data clustered at the county level were used to estimate the hazard ratio (HR) of AIS incidence, comparing Black individuals with White individuals. Separate marginal Cox models tested associations between each measure of structural racism and AIS incidence, with further testing to screen for interaction with the race variable.
Results:
The composite structural racism score identified significant geographic variation in structural racism across the United States (mean 0.818, SD 2.874, interquartile range 3.02). Black individuals had a 19% increased hazard of AIS compared with White individuals (HR 1.19, 95% CI 1.14-1.25, p < 0.0001). All constituent measures of structural racism, except for housing discrimination, were associated with AIS incidence independently of race. Each SD increase in the composite structural racism score was associated with an 18% increased incidence in AIS in the total population. This association interacted with race (p = 0.03), with a greater magnitude of association for White (HR 1.19, 95% CI 1.13-1.25, p < 0.0001) vs Black (HR 1.09, 95% CI 1.03-1.16, p = 0.0073) individuals.
Discussion:
There is significant county-level geographic variation in structural racism across the United States, and increasing levels of structural racism are associated with increased incidence of AIS, regardless of race.
