Does housing improve healthcare utilization and costs? A longitudinal analysis of health administrative data linked to a cohort of individuals with a history of homelessness

Background — Individuals who are homeless have complex healthcare needs, which contribute to the frequent use of health services. In this study, we investigated the relationship between housing and healthcare utilization among adults with a history of homelessness in Ontario.

Methods — Survey data from a 4-year prospective cohort study were linked with administrative health records in Ontario. Annual rates of health encounters and mean costs were compared across housing categories (homeless, inconsistently housed, housed), which were based on the percentage of time an individual was housed. Generalized estimating equations were applied to estimate the average annual effect of housing status on healthcare utilization and costs.

Results — Over the study period, the proportion of individuals who were housed increased from 37% to 69%. The unadjusted rates of ambulatory care visits, prescription medications, and laboratory tests were highest during person-years spent housed or inconsistently housed and the rate of emergency department visits was lowest during person-years spent housed. Following adjustment, the rate of prescription claims remained higher during person-years spent housed or inconsistently housed compared with the homeless. Rate ratios for other healthcare encounters were not significant (P>0.05). An interaction between time and housing status was observed for total healthcare costs; as the percentage of days housed increased, the average costs increased in year 1 and decreased in years 2–4.

Conclusions — These findings highlight the effects of housing on healthcare encounters and costs over a 4-year study period. The rate of prescription medications was higher during person-years spent housed or inconsistently housed compared with the homeless. The cost analysis suggests that housing may reduce healthcare costs over time; however, future work is needed to confirm the reason for the reduction in total costs observed in later years.