Book Description

Background: Interactions with police shape the HIV risk environment for people who inject drugs (PWID) by driving risky injection behaviors and harm reduction service avoidance. The SHIELD (Safety and Health Integration in the Enforcement of Laws on Drugs) police training in Tijuana, Mexico, is an intervention to improve PWID health by modifying police behavior. This dissertation 1) explores the global body of peer-reviewed literature on police practices and HIV risk among PWID and 2) examines police knowledge, attitudes, and behaviors relevant to PWID health in the context of the SHIELD training. Methods: Chapter 2 constitutes a systematic review of published research with quantitative associations between police practices and HIV and/or risky injection behaviors among PWID (n=8,201 abstracts, 175 manuscripts). Chapter 2 applies longitudinal logistic regression to examine the association between police knowledge of syringe possession law and extrajudicial arrests for syringe possession over 24 months following the SHIELD training in Tijuana (n=693). Chapter 3 uses log-binomial regression to identify police attitudes associated with support for officer-led referrals to drug treatment and syringe service programs (n=305). Results: Chapter 2 identified 27 studies with data on police practices and risk of HIV infection among PWID (n=5), risky injection behaviors (n=21) and harm reduction service avoidance (n=9) from diverse global settings. Chapter 3 establishes that training with the SHIELD model can police improve knowledge of syringe law and reduce self-reported extrajudicial arrests for syringe possession up to 24 months following the training (adjusted odds ratio [AOR]:0.87,95% confidence interval [CI]:0.85,0.90). Officers with correct knowledge of syringe possession law were 37% less likely to arrest PWID for syringe possession (AOR:0.63,CI:0.44,0.89), after controlling for sex and patrol assignment location. Chapter 4 showed that officer-held beliefs that MMT programs reduce criminal activity and SSPs increase the risk of NSI among police were significantly associated with support for officer-led referrals to drug treatment (Adjusted Prevalence Ratio [APR]=4.66,CI=2.05,9.18) and SSPs (APR=0.44,CI=0.27,0.71), respectively. Conclusions: Together, these findings highlight the deleterious role that drug law enforcement practices have on the HIV risk environment for PWID and sheds light on interventions to align police behavior with public health priorities.