UC San Diego

ABSTRACT OF THE DISSERTATION

Behavioral and contextual factors shaping HIV risk environments among people who inject drugs in Tijuana, Mexico: implications for prevention and treatment

by

Jennifer P. Jain

Doctor of Philosophy in Interdisciplinary Research on Substance Use

University of California San Diego, 2019

San Diego State University, 2019

Professor Steffanie A. Strathdee, Chair

Professor Eileen V. Pitpitan, Co-Chair

Background: HIV prevalence among people who inject drugs (PWID) in Tijuana, Mexico is 22 times higher than that among the general population. Among PWID, HIV transmission is shaped by factors in the physical, social, economic and policy risk environments. This dissertation explored factors associated with incident HIV infections sex differences in HIV-related risk factors and their potential underlying mechanisms among PWID in Tijuana, Mexico.

Methods: Data were drawn from a prospective cohort study of PWID in Tijuana, (N=651) chapters 2 and 3, and (N=734) chapter 4. Chapter 2 assesses sex work as a mediator in the association between female sex and elevated HIV incidence using modern mediational analyses. Chapter 3 identifies factors associated with time to HIV seroconversion using Cox regression and assesses the robustness of these associations using novel sensitivity analyses. Chapter 4 explores sex differences in the environmental and psychological correlates of injection risk, using generalized linear regression.

Results: Chapter 2 demonstrates that HIV incidence is significantly higher among female PWID (1.75 per 100 person years [PY], 95% confidence interval [CI] 1.2- 2.7, vs. 0.95 per 100 PY, 95% CI=0.6-1.4), and that sex work mediates the association between female sex and HIV incidence (proportion mediated=86.0%, p=0.01). Chapter 3 highlights sex work (adjusted hazard ratio [aHR]=2.20, 95% CI=1.03-4.68); methamphetamine injection (aHR=2.13, 95% CI=1.04-4.35); and methamphetamine and heroin co-injection (aHR=2.09, 95% CI=1.13-3.85) as drivers of HIV seroconversion, and shows that these associations are relatively robust to unmeasured confounding. In chapter 4 among both sexes, using syringes from a safe source, and safe injection self-efficacy were associated with lower injection risk (b=0.87, 95% CI=0.82-0.94, and b=0.80, 95% CI=0.76-0.84, respectively). Among females, incarceration (b=1.22, 95% CI=1.09-1.36) and syringe confiscation (b=1.16, 95% CI=1.01-1.33) were associated with elevated injection risk, whereas, among males sex work (b=1.16, 95% CI=1.04-1.30), and polysubstance use (b=1.22, 95% CI=1.13-1.31) were associated with elevated injection risk.

Conclusions: Interventions should be sex-specific, incorporate sexual risk reduction, consider drug use behaviors, and address environmental influences of HIV risk among PWID in Tijuana.