Geographic Clustering of Social Injection Activity and Risk of Hepatitis C Virus Acquisition Among People Who Inject Drugs in Montreal

Author: Nanor Minoyan Stine Bordier Høj Marie-Pierre Sylvestre Andreea Adelina Artenie Valérie Martel-Laferrière Sarah Larney Didier Jutras-Aswad Julie Bruneau

Theme: Epidemiology & Public Health Research Year: 2022

High HCV incidence rates persist among PWID in Montreal, Canada, despite longstanding presence
of harm reduction in the city. Harm may be concentrated in specific neighbourhoods; targeting of
these areas may improve cost-effectiveness of HCV elimination efforts, including treatment-asprevention approaches. We estimated the association between residence in areas with high
clustering of social injection activity and HCV infection among PWID in Montreal.
Data were drawn from HEPCO, a prospective cohort study involving three-monthly HCV testing and
interviews with active PWID (2010-2017). At each visit, participants reporting past-month injection
with other PWID reported the location (postcode) of their latest social injection episode. We first
used these postcodes to identify areas exhibiting heightened clustering of social injection activity
(“hotspots”). HCV-negative cohort participants were then categorized as residing (or not) in a
hotspot, based on where they had slept most often in the past month. Incident HCV was defined as
an antibody+/RNA+ test. Associations were estimated using inverse-probability-weighted marginal
structural models to adjust for time-dependent confounding and non-differential dropout by age,
gender, housing/income stability, OAT, incarceration and prescription opioid injection. Weights were
estimated using pooled logistic regression; hazard ratios were obtained from a weighted Cox model.
Follow-up was censored at three years.
Participants were mostly white (89.8%), male (79.8%) with a median age of 40. Hotspots covered an
area of 5.78 km2 and were located in the downtown/Ville-Marie area. At baseline, participants
residing in hotspots were older and generally more vulnerable than those residing in outer areas. 99
infections were observed over 956.0 person-years (rate: 10.4/100py [95%CI: 8.5-1.3]). Weighted
models estimated a two-fold risk of infection in hotspots vs. areas outside (95%CI: 1.20-3.44).
Risk of HCV was elevated in social injection hotspots, supporting prioritization of HCV elimination
efforts to select geographic areas. Prevention may further require modifying structural determinants
of harm among PWID.
Disclosure of Interest Statement:
No author has financial or personal relationships with people or organizations that could
inappropriately influence their work. SBH receives part of her salary from a Canadian Institutes of
Health Research grant. VML holds research grants from Gilead Sciences and Merck and receives
advisor fees from Abbvie (unrelated to the current work). SL received untied educational grant funding
from Indivior. DJA receives study material from Insys Therapeutics for a trial unrelated to the current
work. JB receives research grants from Gilead Sciences and Abbvie and receives advisor fees from
Gilead Sciences and Abbvie (unrelated to the present work).

Download abstract