Social and policy characteristics associated with injurious shootings by police in US counties: A multilevel analysis, 2015-2020

Soc Sci Med. 2024 Oct 29:362:117460. doi: 10.1016/j.socscimed.2024.117460. Online ahead of print.

Abstract

From 2015 to 2020, shootings by police injured or killed 1769 people annually in the United States, disproportionately harming members of minoritized groups. Prior studies of the structural determinants of these inequities have largely examined state-level aggregations and fatal outcomes. This study aimed to: 1) describe state and county variation in fatal and nonfatal injurious shootings by police, and 2) analyze the relationship between state and county context and differences in county rates of injurious shootings by police. Injury data were developed from manual review of incidents listed in the Gun Violence Archive, then aggregated by county-year. Covariate selection was informed by theories of police use of force and the Social Basis of Disparities in Health conceptual framework. Fixed effects negative binomial regression models were estimated, nesting years within counties and states. Analyses controlled for county population, local reporting presence, and multiple measures of social conflict and community violence. From 2015 to 2020, 56% of counties experienced injurious shootings by police. Higher county rates of victimization were associated with greater state spending on police relative to health, county income inequality, prevalence of unmet substance use disorder needs, higher county firearm availability, and permitless concealed carry statutes. Firearm purchaser licensing polices were associated with lower incidence of injurious shootings by police. To prevent patterns of injurious shootings by police, policymakers should consider addressing undermanaged substance use disorder through crisis fund allocation and use, adopting stronger firearm licensing systems, and evaluating local strategies to combat inequities and strengthen non-policing responses to social needs.