Project SHIELD: Police Education Partnership to Support Public Health in Kentucky

Abstract In this study, we will assess the impact and cost-effectiveness of the Safety and Health Integration in the Enforcement of Laws on Drugs (SHIELD) police education program to align public safety with public health in rural Appalachia. A robust body of literature indicates certain policing practices, such as arrest and syringe confiscation, can exacerbate drug-related harm among people who inject drugs (PWID). We have successfully implemented in SHIELD in diverse U.S. and international settings to evaluate knowledge, attitudes, and behaviors related to street-level encounters with PWID, however its impact on policing behaviors in a U.S. setting is unknown. We will leverage a unique research ecosystem of a cohort of over 700 people who do and not inject drugs in Eastern Kentucky, an Appalachian setting with high burdens of social determinants of health, overdose risk, justice involvement, and infectious disease transmission. By evaluating the effectiveness of this intervention in people who use drugs, we will have rigorous, unbiased estimates of the impact of SHIELD, rather than relying on outcomes self-reported by police. With support from state and local law enforcement organizations, we will assess the impact of SHIELD by accomplishing the following three aims: 1) Determine the effectiveness of SHIELD on proximal (arrest, referral) and distal substance use related outcomes (syringe sharing, opioid use) among PWID; 2) Examine mediating processes explaining the impact of police arrests on receptive syringe sharing through changes in harm reduction use, drug treatment, unemployment, and homelessness among PWID; 3) Estimate the observed and projected impact of SHIELD on fatal/non-fatal overdose and HCV transmission among people who do and do not inject drugs. We will deliver the SHIELD training to approximately 300 law enforcement officers in this region and follow them semi- annually for 12 months. Primary outcomes will be arrest and referral to harm reduction services or drug treatment as reported by participants in the parallel cohort of people who use drugs. We will apply novel quasi- experimental and causal mediation analyses and triangulate these effect estimates with self-reported data from police. These data will be used to inform a mathematical model and economic evaluation to determine its impact and cost-effectiveness on substance use related outcomes, including HCV incidence and overdose. Findings from this study will directly inform its effectiveness, public health impact, and value-for-money to inform evidence-based decision-making. Given the increasingly recognized role of justice system involvement as a social determinant of health and driver of overdose risk, the aims of this proposal are highly responsive to the objective of the HEAL Initiative to provide evidence-based solutions to end the opioid crisis.