James Lachaud and colleagues’ 2025 study in JAMA Network Open examines mortality among adults experiencing homelessness and mental illness who participated in Canada’s At Home/Chez Soi trial. The original trial randomized 2,255 people across five cities to Housing First with supportive services or usual services; this secondary analysis linked 2,108 participants to administrative records. Although published in July 2025, the study used mortality follow-up extending through March 2019. The researchers found no statistically significant mortality difference between the groups. The randomized design and extensive record linkage strengthen the analysis, while unavailable causes of death and differing observation periods across cities limit interpretation. These findings concern mortality in this population and do not establish that providing housing lacks value.
For community-safety practice, the study raises an important question about what institutions remain responsible for after a person receives housing. Placement should begin a period of sustained attention to health, material security, and social connection, rather than become the point at which a service system declares its work complete. A mortality outcome also cannot stand in for every purpose of housing policy: having a secure place to live is itself a substantive improvement that deserves evaluation on its own terms. Communities should ask whether their programs can document continuity of care and respond to needs that remain after placement. The study’s contribution is to encourage a more demanding account of support, while avoiding the mistaken conclusion that an uncertain mortality effect justifies withholding housing.



Leave a Reply