Exploring the Policy Implications of High-Profile Police Violence : Freddie Gray Unrest

The goal was to determine whether cumulative exposure to violence in childhood and adolescence contributes to disparities in self-rated health among a nationally representative sample of US adolescents. The National Longitudinal Survey of Youth 1997 is an ongoing, 8-year (1997-2004), longitudinal, cohort study of youths who were 12 to 18 years of age at baseline (N = 8224). Generalized estimating equations were constructed to investigate the relationship between cumulative exposure to violence and risk for poor health. At baseline, 75% of subjects reported excellent or very good health, 21.5% reported good health, and 4.5% reported fair or poor health. Cumulative violence exposures (witnessed gun violence, threat of violence, repeated bullying, perceived safety, and criminal victimization) were associated with a graded increase in risk for poor health and reduced the strength of the relationship between household income and poor health. In comparison with subjects with no violence exposure, risk for poor self-rated health was 4.6 times greater among subjects who reported >or=5 forms of cumulative exposure to violence, controlling for demographic features and household income. Trend analysis revealed that, for each additional violence exposure, the risk of poor health increased by 38%. Adjustment for alcohol use, drug use, smoking, depressive symptoms, and family and neighborhood environment reduced the strength of the relationships between household income and cumulative exposure to violence scores and poor self-rated health, which suggests partial mediation of the effects of socioeconomic status and cumulative exposure to violence by these factors. In this nationally representative sample, social inequality in risk for poor self-rated health during the transition from adolescence to adulthood was partially attributable to disparities in cumulative exposure to violence. A strong graded association was noted between cumulative exposure to violence and poor self-rated health in adolescence and young adulthood.

The USA has very high rates of homicide by police compared to other high-income countries, with approximately 1000 civilians killed annually. The overwhelming majority of these police homicides are fatal shootings. Over the past 5 years, several comprehensive, real-time, data repositories, drawn largely from news reporting, have kept track of incidents in which civilians die during an encounter with the police and have become widely available. Data from these repositories, which are more complete than data available from federal data systems, have been used to explore fatal police shootings of civilians, often with a focus on racial disparities in police shootings of unarmed civilians, and have consistently found that police are more likely to shoot unarmed African American men than unarmed White men. Although numerous studies have examined how rates of police killings of civilians are related to several ecologic determinants of these events, no peer-reviewed study to date has examined the extent to which variation in police involved firearm homicides is explained by firearm prevalence while adjusting for violent crime rates (the most well-established ecologic factor associated with fatal police shootings). The current cross-sectional state-level analysis uses data on the number of civilians shot and killed by police in the line of duty, aggregated over 2015–2017. Data come from the Washington Post’s “Fatal Force Database”, which assembles the information from news reports and other sources. Data provided include information on whether the victim was armed, and, if so, with what weapon. Explanatory ecologic variables in our models include the violent crime rate, the percentage of the state population that is non-White, poverty rate, and urbanization, along with a validated proxy for firearm prevalence. We find that rates of police shooting deaths are significantly and positively correlated with levels of household gun ownership, even after accounting for the other explanatory variables. The association is stronger for the shooting of armed (with a gun) rather than unarmed victims.