Tenants with additional needs: when housing first does not solve homelessness.

Background: At Home/Chez-Soi was a 24 month randomized controlled trial of Housing First (HF) conducted in five Canadian cities.

Aims: This article attempts to identify the characteristics of participants who experienced housing instability one year after entering HF.

Methods: Those defined as experiencing housing instability were housed <50% of the last 9 months of the first year, excluding time in institutions, unless they were housed 100% of the past 3 months.

Results: Only 13.5% of HF participants (n = 157/1162) met criteria for housing instability. Several variables were significant predictors of instability in between-group comparisons and multiple regression analyses: residence in Winnipeg, cumulative lifetime homelessness, percent of previous 3 months spent in jail, and community psychological integration; while residence in Moncton and a diagnosis of PTSD or panic disorder predicted stability. The predictive models were weak, identifying correctly only 3.8% of individuals that failed to achieve housing stability.

Conclusions: It is not possible to predict confidently at baseline who will experience early housing instability in HF. There are certain individual characteristics that might be considered risk factors. Providing HF to all individuals who qualify for a HF program remains the most valid way to administer admission to housing.