Variability of Hepatitis C Treatment Cascade Outcomes among People Who Inject Drugs across Geographically Diverse Clinics in the US: The HERO Study

Heterogeneity of outcomes across different clinical trial study sites is often inevitable. Understanding how outcomes differ by site is important for planning future programs and studies. We examined the extent of heterogeneity of hepatitis C virus (HCV) treatment cascade outcomes among persons who inject drugs (PWIDs) across sixteen clinical sites utilized in the HERO Study-a pragmatic randomized trial of HCV treatment support. Treatment cascade outcomes included averages of overall treatment adherence and proportions of treatment initiation, treatment completion, sustained virologic response (SVR) test completion, and SVR achievement. The HERO study utilized 16 clinical sites across the United States (US): eight opioid treatment programs (OTPs) and eight community health centers (CHCs). Variability of the outcomes across the 16 clinical sites was assessed using ranges and intraclass correlation coefficients (ICC) estimated from mixed-effects linear or logistic regression models. Treatment initiation was analyzed in the intention-to-treat (ITT) sample (N = 755); treatment completion, adherence, and SVR test completion in the modified ITT (mITT) sample, which is the sample that initiated treatment (N = 623); and SVR achievement in the mITT and per-protocol (PP, N = 501) samples. Across the 16 clinical sites, the range observed in the averages of overall treatment adherence was from 68% to 81% [ICC = 0.026 (0.005, 0.054)], and the ranges of proportions observed were from 68% to 96% for treatment initiation [ICC (95% CI) = 0.086 (0.051, 0.155)], 60% to 100% for treatment completion [ICC = 0.049 (0.008, 0.215)], 54% to 95% for SVR test completion [ICC = 0.096 (0.006, 0.177)], 46% to 90% for SVR achievement in the mITT sample [ICC = 0.070 (0.014, 0.122)], and 76% to 100% for SVR achievement in the PP sample [ICC = 0.143 (0.021, 0.422)]. The variability of the outcomes across 16 US sites treating HCV among PWIDs appears to be substantial in view of the ranges and ICC values of the outcomes. It is imperative to develop tailored interventions to target the sources of variability and reduce barriers at the patient, provider, clinic, and state policy levels to facilitate more equitable access to HCV treatment and reduce heterogeneity in treatment outcomes.