Estimating the impact of a police education program on hepatitis C virus transmission and disease burden among people who inject drugs in Tijuana, Mexico: A dynamic modeling analysis

Abstract

Background and aims: Criminalization of drug use and punitive policing are key structural drivers of hepatitis C virus (HCV) risk among people who inject drugs (PWID). A police education program (Proyecto Escudo) delivering training on occupational safety together with drug law content was implemented between 2015 and 2016 in Tijuana, Mexico, to underpin drug law reform implementation. We used data from a longitudinal cohort of PWID in Tijuana to inform epidemic modeling and assess the long-term impact of Escudo on HCV transmission and burden among PWID in Tijuana.

Methods: We developed a dynamic, compartmental model of HCV transmission and incarceration among PWID and tracked liver disease progression among current and former PWID. The model was calibrated to data from Tijuana, Mexico, with 90% HCV seroprevalence. We used segmented regression analysis to estimate impact of Escudo on recent incarceration among an observational cohort of PWID. By simulating the observed incarceration trends, we estimated the potential impact of the implemented (2-year reduction in incarceration) and an extended (10-year reduction in incarceration) police education program over a 50-year follow-up (2016-2066) on HCV outcomes (incidence, cirrhosis, HCV-related deaths and disability adjusted life-years averted) compared with no intervention.

Results: Over the 2-year follow-up, Proyecto Escudo reduced HCV incidence among PWID from 21.5 per 100 person years (/100py) (95% uncertainty interval [UI] = 15.3-29.7/100py) in 2016 to 21.1/100py (UI = 15.0-29.1/100py) in 2018. If continued for 10 years, Escudo could reduce HCV incidence to 20.0/100py (14.0-27.8/100py) by 2026 and avert 186 (32-389) new infections, 76 (UI = 12-160) cases of cirrhosis and 32 (5-73) deaths per 10 000 PWID compared with no intervention over a 50-year time horizon.

Conclusions: In Tijuana, Mexico, implementation of a police education program delivering training on occupational safety and drug law content appears to have reduced hepatitis C virus incidence among people who inject drugs.

Keywords: Epidemic modelling; HCV; PWID; Tijuana; incarceration; police education.

Figure 1.. Hepatitis C Virus transmission and disease progression model schematic among PWID and ex-PWID.

Boxes represent disease stages, and arrows represent transitions.

Figure 2.. Incarceration submodel schematic among PWID and relative risks (RR) of HCV transmission for each stage compared to a referent group of never incarcerated.

Figure 3.. Model fit to segmented regression analysis trend in relative proportion of PWID recently incarcerated in the 2 years after Escudo compared to pre-Escudo in 2016.

Solid line presents the mean projections with dashed lines representing the 95% confidence interval projections. Red dots represent the predicted trends after 1 and 2 years of Escudo implementation from the segmented regression analysis, with vertical lines representing the 95% confidence intervals of the data.

Figure 4.. Fifty-year trajectory for mean HCV incidence rate among PWID in Tijuana, Mexico for different Escudo implementation scenarios from 2010–2060.

Figure 5.. Relative reduction in HCV incidence rate among PWID in Tijuana, Mexico simulating a 2- and 10-year Escudo intervention effect.

Lines represent the median estimate, boxes 25–75% centiles, and whiskers 2.5–97.5% centile projections.

Figure 6.. Fifty-year impact in cumulative HCV deaths averted among current and former PWID for each Escudo implementation period, per 10,000 PWID