Racism Lives Here: Leveraging Context and Place in the Conceptualization and Operationalization of Structural Racism in Health Research

Abstract

Structural racism is defined as “… the macrolevel systems, social forces, institutions, ideologies, and processes that interact with one another to generate and reinforce inequities among racial and ethnic groups.” Though structural racism is theorized as a contextual exposure, often due to data limitations, it is operationalized in health research using compositional measures such as segregation, or other measures reliant on the proportion of relevant groups in a given place. In this dissertation, in collaboration with my committee members, I explore underutilized data sources and novel theories to conceptualize and operationalize structural racism. First, I use a disaster-framework to understand the community-level health impact of the police killing of Mike Brown, Jr. in 2014 in Ferguson, Missouri, and the prolonged sociopolitical conflict which occurred in response to protest. Using a difference-in-difference design, I compare the rates of preterm birth in St. Louis County (where Ferguson is located), St. Louis City, and St. Charles County, for mothers who conceived during the Ferguson Uprising compared to mothers who conceived during the same time in the 3 years prior to the Ferguson Uprising. I compare the change in rates across time to Jackson County, Missouri, all Missouri Counties, and all U.S. counties. I found that mothers in the St. Louis region were significantly more likely to deliver preterm ( weeks gestation) or for the baby to be low birthweight ( g). Second, I use the Census of Governments dataset to examine the proportion of revenue generated through police and court-related fees and fines as a type of exploitation of residents, particularly Black residents, at the county level. I examined the relationship between fees & fines and odds of preterm birth and low birthweight deliveries. We found that controlling for maternal characteristics, living in areas most reliant on fees & fines for revenue was associated with an increased risk for adverse birth outcomes. Finally, we used data made available through the Home Mortgage Disclosure Act to determine if there is evidence of discrimination in home improvement loan denials. These data have primarily been used to assess discrimination in home purchase loans, however repairs, maintenance, and other home improvement tasks represent an important factor in homeownership. We found that non-Hispanic Black applicants were more likely to be denied home improvement loans. We also found that non-Hispanic white applicants were more likely to apply for loans in Black, and lower income neighborhoods, perhaps indicating areas likely to gentrify. Finally we assessed rates of home improvement loan denial and rates of poor mental health and short sleep duration. We found that sleep, but not mental health, was associated with higher rates of home improvement loan denials. Expanding how structural racism is measured in health research is particularly important as the chosen exposure influences how we interpret findings and proposed solutions and interventions.