Equal Protection Under the Law: The Measurement of Structural Racism and Health Disparities

This dissertation focuses on measuring the health effects of exposure to structural racism, or racism that forms through historic and contemporary policies that produce a disparate impact by race. After defining various forms of racism and describing how structural racism relates to health outcomes, I examine three distinct measures of exposure to structural racism arising from housing and criminal justice policies in relation to birth outcomes (Aim 1) and 30-year cardiovascular risk (Aims 2 and 3). In each of these studies I hypothesize that exposure to structural racism will be more harmful to Blacks than Whites. This hypothesis is grounded in research and theory suggesting that, despite color-blind policies that on the surface appear to affect all people in the same way, the racially disparate impacts of the policies are simultaneously privileging Whites while disadvantaging Blacks. The first measure of structural racism is in utero exposure to the Flint Water Crisis (FWC) emergency declaration, which is considered to be a racialized stressor due to Flint’s long history of racial segregation. The sample includes Michigan women (outside of Flint) who were pregnant before and during the declaration of a State of Emergency. Here I examine the relationship between the FWC declaration and birth outcomes between babies born to Black and White mothers using a before and after linear regression analytic methodology. I find that both babies born to Black and White mothers after the emergency declaration have lower birthweight, gestational age, and size-for gestational age compared to babies born before the declaration, but exposed babies born to Black mothers have marginally significantly greater decline in gestational age compared to exposed babies born to White mothers. The second measurement is direct exposure to the police through self-reported police encounters. Data are from a nationally representative sample of young adults interviewed during the War on Drugs, which included foundational policies that created incarceration disparities in the US. Using a standard linear regression with domain analysis, I examine the association between sex-specific levels of police encounters and the Framingham 30-year cardiovascular risk score, which predicts a major cardiovascular event occurring in the next 30 years. I find that the relationship between police encounters and cardiovascular risk differs by race, where a high level of police encounters is associated with higher risk for cardiovascular disease among Whites but lower risk for cardiovascular disease among Blacks. The final measure of structural racism is community- and school-level segregation, using standard measures of residential segregation and a novel measure of school segregation. Data are from a nationally representative sample of adolescents entering adulthood. Here I examine the association between residential and school segregation and the Framingham 30-year cardiovascular risk score using a general estimating equations analytic methodology. I find that a higher level of residential segregation measured with the Index of Concentration of Extremes is associated with an increased cardiovascular disease risk for both Blacks and Whites, but the association is stronger for Blacks. In addition, I find that a higher level of school segregation measured with the Index of Concentration of Extremes is associated with increased cardiovascular disease risk in Blacks but decreased risk in Whites. Overall, I find mixed support for the central study hypothesis. I conclude with suggestions for future research and policy considerations.PhDEpidemiological ScienceUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/169682/1/kallgood_1.pd