Patterns and Predictors of Housing Relocations Experienced By Housing First Tenants

Background and Purpose: Existing evidence demonstrates that housing first (HF), which provides immediate access to independent, scatter-site apartments and ongoing supportive services, is an effective intervention for addressing chronic homelessness.  Prior studies have shown two-year retention rates in HF programs of around 85% and high levels of satisfaction among housing first tenants.  However, there has been little research to date on relocations between apartments experienced by HF tenants after they have entered a HF program.  Assessing the extent to which HF tenants experience an initial—and possibly subsequent relocations—as well as the dynamics and nature of these relocations, is important for arriving at a more nuanced understanding of the residential stability of HF tenants.  Therefore, the present study aims to: 1) assess the extent to which HF tenants experience relocations due either to housing loss or housing fit; 2) assess the relationship between HF tenants’ socio-demographic characteristics and type of housing relocation.

Methods: The study uses data from an HF program in a large urban area to examine housing relocations among a cohort of 266 HF tenants who entered the program between 2008 and 2013.  Program data include dates of initial move-in to an apartment, as well as dates and reasons for all subsequent moves to different apartments.  We use these data to classify relocations experienced by HF tenants as either being due to housing loss (which includes negative events such as eviction or incarceration) or housing fit (which includes factors beyond a tenant’s control such as code violations or clinical concerns).  We use event history analysis, to examine the extent, timing and types of relocations experienced by HF tenants, and to assess the relationship between relocations and tenant socio-demographic and clinical characteristics.

Results:  Among HF tenants in the study cohort, more than 50% experienced at least one relocation over the course of the study period, and roughly one third experienced two or more relocations. Nearly 80% of initial relocations were due to reasons having to do with housing fit rather than due to housing loss, and subsequent moves were also more commonly due to housing fit than housing loss.  The event history models identified salient socio-demographic and clinical predictors of relocations.

Conclusions: Study findings show that relocations of tenants in HF first programs are quite common, but that these relocations are more frequently due to housing fit, rather than housing loss. In this sense, relocation of tenants in HF programs would seem to occur for largely for reasons that such relocations occur for members of the general population.  We discuss implications of these findings for HF programs.