Supervised consumption sites and infectious disease risk among people who inject drugs

Introduction: The rise of the synthetic opioid era has resulted in a new risk environment for people who inject drugs (PWID). In response, many governments have opened, or are considering opening, supervised consumption sites (SCS) to prevent fatal overdoses. In addition to prevention fatal overdoses, SCS use has been associated with decreased syringe and injection works sharing among those who access SCS. However, most of this research was done prior to the current era of synthetic opioids. The aims of this work were to: a) measure the association between SCS use and syringe sharing cross-sectionally and assess the potential impacts of multiple forms of bias, b) determine how SCS use patterns over time impact syringe sharing patterns, and d) estimate the association between SCS use and syringe sharing in Philadelphia, a context in which SCS do not yet exist. Methods: Using data from the Ontario Integrated Supervised Injection Services cohort study in Toronto, Ontario, Canada (OiSIS-Toronto), we measured the associations between self-reported SCS use and injection equipment sharing, conducted sensitivity analyses varying SCS use categories and restricting equipment sharing to syringes, and performed multiple bias analyses to estimate the impact of misclassification, selection bias, and unmeasured confounding on the observed association. We combined baseline data with follow-up data to calculate SCS use and syringe sharing trajectories using group-based trajectory models. Finally, using a transportability framework and data from the National HIV Behavioral Surveillance program in Philadelphia, we estimated the association between SCS use frequency and syringe sharing if SCS were implemented in Philadelphia. Results: Frequent SCS use was not associated with sharing injection equipment (aPR: 0.98; 95% CI: 0.77-1.24) and sensitivity analyses recategorizing exposure and outcome were of similar magnitude and significance. In multiple-bias analysis, the median bias-adjusted prevalence ratio was 1.04 (range: 0.83-1.42), suggesting no association between regular SCS use and syringe sharing. In group-based trajectory modeling, we identified four distinct SCS use trajectories, and three distinct syringe sharing trajectories. The majority of participants were in the non-syringe-sharing trajectory, and being in this trajectory was associated with being in a dynamic SCS trajectory group (i.e., increasing or decreasing use). When the cross-sectional association between SCS use frequency and syringe sharing was transported from Toronto to Philadelphia, the transported association was quite similar to the association in Toronto. Conclusions: Standardizing definitions of SCS use and injection equipment sharing may be helpful in reproducing results across populations. Sources of bias had little impact on the observed association between SCS use and syringe sharing among Toronto PWID. Group-based trajectory modeling can help harm reduction services assess the needs of client subpopulations. While the association between SCS use and syringe sharing was not significant after transporting to Philadelphia, similar methods could be applied to study other potential health impacts of SCS in new contexts.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.