Patterns and effects of social integration on housing stability, mental health and substance use outcomes among participants in a randomized controlled Housing First trial

Introduction: Recent research has shown that social integration, involving community integration and social support, can be protective resources for the health of homeless individuals. However, it is not clear how social integration affects health in the transition from homelessness into housing, and subsequent housing retention. This paper examines, through mixed quantitative and qualitative methods, how social integration changes over time, and how these changes relate to housing stability, substance use and mental health outcomes among a sample of homeless individuals experiencing mental illness participating in the At Home/Chez Soi Housing First randomized controlled trial in Canada. Methods: Longitudinal quantitative data (baseline, 6, 12, 24 month) and qualitative data (18 month) from the five trial sites (Moncton, Montreal, Toronto, Winnipeg and Vancouver) were examined using mixed models, mediation analyses, and thematic analysis. Results: Social integration (i.e., social network size, social interest, psychological integration) increased over time among study participants, with social network size increasing significantly more among Housing First participants than Treatment as Usual participants. Social network size mediated the effect of the intervention on the percentage of days in stable housing, indicating that the Housing First intervention may have increased participants' social network size, which in turn increased the percentage of days stably housed. No significant mediation of social integration on the effects of the intervention on mental health and substance use outcomes was found. Findings from the qualitative interviews support and expand upon these quantitative findings.