Empirical evidence on structural racism as a driver of racial inequities in COVID-19 mortality

Abstract

ObjectiveThis study contributes to the literature by empirically testing the extent to which place-based structural racism is a driver of state-level racial inequalities in COVID-19 mortality using theoretically-informed, innovative approaches.MethodsCDC data are used to measure cumulative COVID-19 death rates between January 2020 and August 2022. The outcome measure is a state-level Black-White (B/W) ratio of age-adjusted death rates. We use state-level 2019 administrative data on previously validated indicators of structural racism spanning educational, economic, political, criminal-legal and housing to identify a novel, multi-sectoral latent measure of structural racism (CFI = 0.982, TLI = 0.968, and RMSEA = 0.044). We map B/W inequalities in COVID-19 mortality as well as the latent measure of structural racism in order to understand their geographic distribution across U.S. states. Finally, we use regression analyses to estimate the extent to which structural racism contributes to Black-White inequalities in COVID-19 mortality, net of potential confounders.ResultsResults reveal substantial state-level variation in the B/W ratio of COVID-19 death rates and structural racism. Notably, regression estimates indicate that the relationship between the structural racism and B/W inequality in COVID-19 mortality is positive and statistically significant (p < 0.001), both in the bivariate model (adjusted R2 = 0.37) and net of the covariates (adjusted R2 = 0.54). For example, whereas states with a structural racism value 2 standard deviation below the mean have a B/W ratio of approximately 1.12, states with a structural racism value 2 standard deviation above the mean have a ratio of just above 2.0.DiscussionFindings suggest that efficacious health equity solutions will require bold policies that dismantle structural racism across numerous societal domains.

The Future of Social Determinants of Health: Looking Upstream to Structural Drivers

“…To better align measurement strategies with core tenets of structural theories, a growing chorus of scholars recommend that quantitative research on the health consequences of structural oppression use latent variable approaches , which are well‐suited for minimizing measurement error and capturing unobserved, complex systems such as structural racism and structural sexism 9,20,55 . Indeed, several recent population health studies have demonstrated the utility of confirmatory factor analysis to estimate latent measures of state‐level and county‐level structural racism (using observed, reflective indicators in housing, educational, economic, political, and criminal–legal domains) 30,37,57 . Consistent with the idea that structural racism involves a multifaceted, interconnected system, these studies used latent measurement approaches to explicitly model relationships among the multisectoral reflective indicators of structural racism.…”

“…Findings from a study by Brown, Kamis, and Homan showed that a multisectoral latent measure of state-level structural racism provided a better model fit and accounted for more variation in Black-White inequities in COVID-19 mortality than individual indicators as well as a summative index of structural racism. 30 In addition, Chantarat and colleagues have contributed to the literature by using latent class analysis to identify distinct typologies of multisectoral structural racism and linking them to racialized birth outcomes. 58,59 Collectively, these studies point to promise of latent measurement approaches for advancing the scientific literature on the impact of structural oppression on health.…”

“…Consistent with the idea that structural racism involves a multifaceted, interconnected system, these studies used latent measurement approaches to explicitly model relationships among the multisectoral reflective indicators of structural racism. Findings from a study by Brown, Kamis, and Homan showed that a multisectoral latent measure of state‐level structural racism provided a better model fit and accounted for more variation in Black–White inequities in COVID‐19 mortality than individual indicators as well as a summative index of structural racism 30 . In addition, Chantarat and colleagues have contributed to the literature by using latent class analysis to identify distinct typologies of multisectoral structural racism and linking them to racialized birth outcomes 58,59 .…”

“…9,20,55 Indeed, several recent population health studies have demonstrated the utility of confirmatory factor analysis to estimate latent measures of state-level and county-level structural racism (using observed, reflective indicators in housing, educational, economic, political, and criminal-legal domains). 30,37,57 Consistent with the idea that structural racism involves a multifaceted, interconnected system, these studies used latent measurement approaches to explicitly model relationships among the multisectoral reflective indicators of structural racism. Findings from a study by Brown, Kamis, and Homan showed that a multisectoral latent measure of state-level structural racism provided a better model fit and accounted for more variation in Black-White inequities in COVID-19 mortality than individual indicators as well as a summative index of structural racism.…”

“…22,23 Structural racism and structural sexism refer to the systematic race and genderbased exclusion from resources, power, and opportunity across an array of societal domains. 15,24,22 New lines of research have emerged that are beginning to measure structural racism, structural sexism, structural lesbian, gay, bisexual, transgender plus (LGBT+) stigma, structural xenophobia, and other forms of structural oppression, and examine their links to health 3,19,20,22,[25][26][27][28][29][30] but this research is still in its infancy. Many additional forms of structural oppression such as cissexism, structural ageism, structural classism, and structural ableism have yet to be conceptualized, measured, or thoroughly examined as structural drivers of health.…”