Housing and support intervention and mortality among homeless adults with mental illnesses: a secondary analysis of a randomized clinical trial

Importance — Homelessness is an important risk factor for premature death, with individuals experiencing homelessness having substantially higher mortality rates than the general population.

Objective — To assess the association of housing and support interventions with mortality among individuals experiencing homelessness and mental illnesses.

Design, setting, and participants — This secondary analysis of a randomized clinical trial included 2255 homeless adults with mental illnesses. The study was conducted in 5 Canadian cities (Vancouver, Winnipeg, Toronto, Montreal, and Moncton). Recruitment took place from October 2009 to July 2011; mortality data were collected until March 30, 2019. Due to the complexity of accessing health administrative data, analyses were conducted and completed between February 2021 and December 2023.

Exposure — Participants were randomized to receive either the Housing First (HF) intervention, which provided immediate permanent, scattered-site housing and support through intensive case management or assertive community treatment to chronically homeless individuals, or treatment as usual (TAU).

Main outcomes and measures — Mortality rate ratios were ascertained at each site using health administrative databases. Adjusted hazard ratios were computed using Cox proportional hazard survival models. Random-effects meta-analysis was used to calculate pooled effect sizes across sites.

Results — Of the 2255 total participants, 2108 (93.5%) were successfully linked with health administrative data; among them, 1434 (68.0%) were male, with a mean (SD) age of 40.6 (11.5) years. Mortality rates were not different in the HF compared with TAU groups (pooled log mortality rate ratio, −0.07; 95% CI, −0.36 to 0.22). The pooled adjusted hazard ratio comparing mortality in the HF and TAU groups was 0.83 (95% CI, 0.43-1.22).

Conclusions and relevance — In this secondary analysis of a randomized clinical trial, the HF intervention was not directly associated with mortality risk. Research is needed to determine whether adjunctive interventions could reduce mortality among homeless individuals with mental illnesses.