Abstract Body

HCV treatment models based on an individual’s drug use network have the potential to accelerate HCV elimination through increased rates of treatment uptake and reduced rates of reinfection among injection partners. Data to support the feasibility of this approach is limited.

Persons who reported recent (within < 1 year of enrollment) injection drug use were iteratively recruited from an urban infectious diseases clinic. We conducted detailed egocentric network inventories in which participants enumerated all network members including injection partners. These Egos (initial person recruited) received a brief intervention which included provision of information about HCV and its treatment and were instructed to recruit members of their injection network for HCV/HIV testing and, if positive, linkage to care. Egos received $10 for each listed member who presented for evaluation. Multivariable logistic regression analysis was conducted using generalized estimating equations (GEE) to assess for factors associated with the successful recruitment of ≥ one drug using network member.

Between January and August 2018, 67 PWID with active injection drug use and HCV (with or without prior treatment) completed egocentric network surveys with the following characteristics: Median age, 54 years (interquartile range (IQR) 45-58); male, 72%; Black, 81%; homeless, 50%; unemployed, 87%; mean income, $735/month; prior incarceration (median time incarcerated, 4 years), 97%. In this group 26 (38%) had been previously HCV treated of which 12(18%) reported previous HCV cure. Egos reported injecting heroin (40%) and cocaine + heroin (37%), and 42% ≥ daily injection in the last 30 days. PWID reported a median of 7 (IQR 5-10) network members of which a median of 3 (IQR 1-5) were injection partners. Mean network density (proportion of ego’s network members that are connected controlling for network size) was 0.6. Of the 67 Egos, 27 recruited ≥ 1 drug using network member (range 1-5). In multivariate analysis, Egos were more likely to successfully recruit if they had been treated for HCV (Odds ratio (OR) 4.1, 95% Confidence Interval (CI) 1.1-16.1), were injecting at least daily (0R 3.4, 95% CI 0.9-11.7) and reported a dense network (OR 9.0, 95% CI 1.0-74.2).

HCV treated PWID may be particularly effective at recruiting their drug using network members for HCV testing and linkage to care. Further work is needed to systematically assess network recruitment methods for HCV treatment