Political Geographies of Health Equity and Criminalization in the U.S Overdose Crisis

Abstract

Since 1999, nearly one million U.S. adults have died from overdose. The ongoing Covid-19 pandemic has further accelerated the overdose crisis. Provisional data suggest that nearly 100,000 individuals died from overdose in the pandemic’s first year, and EMS related activations for overdose related cardiac arrests increased by more than 40% compared to 2018-2019. Initial empirical descriptions and public discourse of the overdose crisis focused narrowly on prescription opioids and among white adults in midlife, especially those in rural areas and in specific U.S regions such as Central Appalachia. However, overdose mortality is now increasing fastest among Black individuals and in large urban areas, with inequities increasing sharply during Covid-19.

Methodological limitations and lack of theoretical orientation of existing empirical analyses – especially the narrow focus on “opioids” and failure to incorporate structural racism and spatial processes – have resulted in an incomplete epidemiologic understanding of the current crisis, especially spatial variation within sociodemographic groups. Additionally, a growing body of work documents the role of the criminal legal system – including policing and mass incarceration – in perpetuating structural racism and increasing substance-use related harms. Despite consistent evidence demonstrating similar rates of illicit drug use across racial and ethnic groups, racist responses to the contemporary overdose crisis have emphasized public health approaches for white communities and criminal justice approaches, including community surveillance and mass incarceration, in communities of color. However, there is little understanding of how governmental budget allocations towards the racialized criminalization of drug use shape these risks.

In Chapter 1, we provide a cohesive, data-driven description of the spatial, temporal, and sociodemographic patterning of drug overdose epidemiology in the US, embedding the empirical description within historic context and alongside theories of structural racism and racial capitalism. Using Bayesian spatio-temporal models, we note several examples where the epidemiologic patterns are not consistent with prominent discourse of the overdose crisis. For example, we observed the highest rates of overdose mortality in large central metro counties, counties in Northern New Mexico and several western states. Importantly, these geographies and urbanicities were places where Black, Native American, and Latinx individuals experienced their highest mortality rates. Native Americans experienced the highest overdose mortality of any racial/ethnic group, and 1 in 5 Black individuals and 1 in 25 Latinx individuals, respectively, experienced higher mortality rates than white individuals in the same counties. In addition, we find that while the spatial distribution of overdose mortality among white individuals is largely consistent with prominent discourse, overdose mortality among Black, Latinx, and Native American individuals exhibits very different spatial patterns. Finally, overdose mortality rates during Wave 3 (2013-2018) increased fastest among sociodemographic groups and geographies often excluded from the “deaths of despair” narrative that emerged during the first epidemic waves, and while uncommon, decreases were most common among groups centered in prominent discourse.

In Chapter 2, we further explore heterogeneity in epidemic trajectories across space and sociodemographic groups. Using latent class trajectory methods, we identified 9 latent epidemic trajectory classes which broadly fell into four groups – [A] counties with a large Wave 1 surge followed by flat or decreasing overdose mortality rates; [B] counties with a large Wave 1 surge, leveling trends during Wave 2 followed by a resurgence in Wave 3; [C] counties with slower overdose mortality increases during Wave 1 and 2 followed by steepest increases during W3; and [D] counties with steady growth throughout the study period. We find that even within the same county, different sociodemographic groups experienced different distributions of epidemic trajectories, with differences most pronounced by race/ethnicity – 91% of Black individuals and 70% of Latinx individuals experienced accelerating overdose mortality during Wave 3 (i.e., groups [B] Level-Resurge and [C] W3 Surge) compared to less than half of Native American, white, and Asian/Pacific Islander individuals.

In Chapter 3, we systematically describe state and local governmental resource allocation towards racialized criminalization of drug use compared to resources allocated for public health and quantify the resultant impacts of this resource distribution on overdose mortality. Using 20 years of data from the U.S. Census of Government Finance, we find median state and local spending on police and corrections (i.e., carceral spending) was 2.1 (IQR: 1.5, 3.2) times higher than public health spending, and exceeded health spending in 90% of all counties. Using marginal structural models, we find that each additional percent increase in state and local police spending as a percentage of total expenditures was associated with a 6% increase in overdose mortality (RR: 1.06 [95% CI 1.01, 1.11]). Additionally, we find that the effect of police spending on overdose mortality was generally highest in groups whose drug use has been historically criminalized – namely, among Black individuals, males, and those aged 35-64 years. Results from our mediation analysis suggest that 35% of the total effect of police spending on overdose mortality is mediated through increased drug arrests. Finally, had all U.S. counties reduced their police spending to the 5th percentile (1.53% of total expenditures) between 1999-2018, we estimate that 22,083 (95% CI: 37,612, 10, 179) overdose deaths could have been prevented – 3% of all overdose deaths during this time period.

As demographics of the crisis shift toward sociodemographic groups and geographic areas where drug use has been historically subject to violent militarized policing and mass incarceration, addressing the current overdose crisis is increasingly a racial justice issue. Our results emphasize that there is no singular overdose crisis and underscores the importance of policy and practice in shaping the epidemic – namely, that systematic governmental investment in carceral systems, especially police spending, increases overdose mortality and further perpetuates the harms of structural racism and the ‘war on drugs.’ As such, resources must urgently be diverted from carceral systems and reinvested in community-led solutions that promote health, especially in Black and Indigenous communities most proximate to the pathogenic effects of structural racism and racial capitalism. Urgent and immediate action is needed to rectify the historic and ongoing harms of the racist ‘war on drugs’, strengthen the social safety net including universal health care, and address the fundamental causes of the crisis – including the inequitable distribution of material resources, exploitation of workers, and ongoing oppression under capitalism and white supremacy. Failure to do so will result in continued acceleration of the overdose crisis and increasing socioeconomic and racial inequities.