Factors associated with housing stability and criminal convictions among people experiencing homelessness and serious mental illness: Results from a Housing First study

Background: Housing First (HF) facilitates immediate access to independent housing with community-based supports for people experiencing homelessness and serious mental illness (PEHSMI). Despite positive outcomes associated with HF, studies have infrequently investigated factors that are associated with adverse outcomes once in HF. This thesis investigates factors which hinder housing stability following randomization to HF and factors associated with criminal convictions prior to and following randomization to HF. Methods: Three investigations were conducted using data from the Vancouver At Home study, which contains two randomized controlled trials each involving randomization to HF vs. treatment as usual (TAU) among PEHSMI. Using self-reported data, the first investigation examined the effect of experiencing homelessness in childhood or youth on housing stability (≥90% of days stably housed) after receiving HF (TAU excluded). The second investigation retrospectively examined factors associated with criminal convictions during the five-year period preceding baseline. The third investigation examined factors associated with criminal convictions after receiving HF (TAU excluded). Provincial administrative data were combined with self-reported baseline data for the second and third investigations. Results: 1) Among participants randomized to HF (n=297), those who had experienced homelessness in childhood or youth had significantly lower odds of housing stability. 2) Prior to study baseline, seven variables were significantly associated with criminal convictions among participants (n=425), such as drug dependence, psychiatric hospitalization, and irregular frequency of social assistance payments (vs. regular). 3) Following receipt of HF (n=255), five variables were significantly associated with criminal convictions, including daily drug use, daily alcohol use, and having received addictions counselling among others. Conclusions: Results underscore social marginalization as contributing to poorer housing stability in HF and criminal convictions while in HF and prior to enrollment among PEHSMI. Further supports are needed to facilitate improvements for a greater proportion of HF clients. HF providers may be able to identify clients with additional support needs related to housing stability and criminal convictions by asking about the factors found to be significant in analyses