Impact of Risk Heterogeneity on the Feasibility of Hepatitis C Elimination Among People Who Inject Drugs: A Modelling Study

Abstract

Although previous modelling work indicates treatment of < 10 people who inject drugs (PWID) per 100 person-years (PY) could achieve hepatitis C virus (HCV) elimination targets in many settings, these models frequently make simplifying assumptions of heterogeneity in infection risk. Here, we evaluated the impact of incorporating risk heterogeneity in transmission models on the predicted effects of interventions and the feasibility of HCV elimination in high-burden settings. We built an individual-based model of HCV transmission informed by detailed data from a cohort of PWID in Baltimore, MD, including an individual- and time-varying risk multiplier on the force of infection. We contrasted these risk-informed models to risk-agnostic models, ignoring this heterogeneity, and explored various levels of treatment and harm reduction scale-up from 2020 to 2030. Risk-agnostic models routinely estimated greater reductions in incidence (8%–19% higher for treatment rates of 10–90 per 100 PY) and greater numbers of infections averted per treatment course compared to otherwise equivalent populations modelled with risk heterogeneity. Elimination targets were only achieved in risk-informed models when treating 90 PWID per 100 PY. Expanding harm reduction services dramatically improved the impact of elimination programs, particularly in averting new infections soon after treatment scale-up initiation. Achieving HCV elimination targets among PWID in high-burden settings will require substantial improvements in treatment access and harm reduction services. Models that ignore the unequal distribution of HCV risk, including the correlation between reinfection risk and onward transmission, can result in inappropriately optimistic estimates of the feasibility of elimination.