Many programs and policies in Canada and around the world have preventing or reducing homelessness as a goal. Do these programs really work? What kinds of programs have been evaluated? What are the different approaches used by researchers to evaluate strategies to end homelessness?
To answer these questions, we have recently developed a literature review that summarizes research on interventions that aim to reduce or prevent homelessness. Here’s a brief summary of that work. We started with the question: What works and for whom? Our specific goals were to understand the different populations for whom interventions have been tested, and the types of interventions evaluated, as well as to create an inventory of the indicators used in program evaluations. We’re hoping this will help people who are planning program evaluations, as well as highlight gaps in the evaluation of initiatives to end homelessness.
Here are some of the highlights from the review:
Different people, different needs: More research is needed on what works for different groups. The main focus of most published evaluations in the peer-reviewed literature was people with mental illness, with little attention given to differences in gender, age, ethnicity, or substance use. This is especially the case for evaluations of permanent independent housing (such as housing first). This is concerning given the increasing number of homeless women, youth, and families, including people from diverse ethnic backgrounds and those with minority sexual orientations. The gap in evaluation of housing interventions for First Nations, Inuit and Metis peoples is particularly concerning given the over-representation of Aboriginal people among homeless populations in Canada There is also a lack of research on independent housing programs for people whose main concern is substance use or addiction. Future research on homelessness interventions would benefit from a wider focus on what eases transitions out of homelessness for different groups of people with distinct needs (e.g. women, people with substance use problems, youth).
What about health? Housing and health are closely linked. A lack of housing has dire consequences for health, and people who are homeless suffer disproportionately from many health conditions. We didn’t find very many evaluations that looked at participants’ health outcomes beyond psychological functioning, even though health would likely improve with better housing. There is a need for future evaluation that focuses on changes in health (both physical and mental) related to housing.
What about the social and economic conditions that lead to homelessness? Mostly, the evaluations measured success by looking at whether program participants became or remained housed after leaving the program. They also often reported on participants’ thoughts about how effective the programs were. These are important things to look at, but the evaluations tended to focus on individual circumstances without referring to the broader social conditions that create homelessness, like poverty, racism, discrimination, housing policy, welfare policy, and cost of living. Evaluations that include discussion of these factors would give a fuller picture of the context, which can tell us something about why the program failed or succeeded and under what conditions similar programs could have a positive impact.
Social and economic conditions matter, so why don’t we change them? There are policy-level initiatives to end homelessness designed to address the social conditions that make it so hard for some people to find housing. While we did find some evaluations of policy level initiatives to end homelessness in this review, more are needed. Thus, we suggest an increased focus on evaluating broader systemic responses to end homelessness. The causes of homelessness are complex and require a multi-pronged approach; we know that systemic changes are needed if programs are going to achieve the goals of preventing, ending or reducing homelessness.
In undertaking program evaluation, we need to ask what are the broader social and economic conditions that impact the success of any program and move to questions regarding "what works for whom and in what context?"