Attempting Randomized Housing First Research in a Community Context: Reflections on Failure.

The Housing First (HF) model of permanent supportive housing for individuals experiencing chronic homelessness has a strong evidence base that has largely been driven by researchers in the field of community psychology in partnership with community-based organizations. However, important gaps in the HF literature remain. Implementing rigorous research designs to further the evidence for HF requires immense resources to fund both the housing intervention and the research activities. In the absence of such resources, university-community partnerships may be established to integrate research within business-as-usual services and utilize existing housing units. This first person account presents a "post-mortem" exploration of an attempt to conduct a randomized trial of scattered-site and single-site approaches to HF within a community context from the perspectives of multiple stakeholders involved in the endeavor. Despite strengths of the research collaborative, the project did not come to completion due to a series of both insurmountable and avoidable barriers. Yet, the experience illuminated several potential challenges researchers and housing providers conducting work in this area may encounter, such as ever-changing homeless service system policies that may impact research and organizational procedures. Lessons learned and recommendations for preventing or overcoming systems-level barriers and potential challenges within the university-community partnership are described.