The Clinical Outcomes of a Canadian Hospital at Home Program: A Matched Cohort Analysis

Importance The safety of Hospital at Home (HaH) has been demonstrated by many high-quality studies; however, there is a paucity of data on Canadian HaH programs. Objective This study evaluates the clinical outcomes of the HaH program in Victoria, British Columbia, Canada. Design and Setting We conducted an observational quality outcomes analysis comparing HaH and brick-and-mortar (BAM) groups using data from the national Discharge Abstract Database (DAD). Risk adjustment was applied to control for clinical and demographic differences. Adjusted outcome rates were estimated using multivariable regression models, and group differences were assessed with two-proportions Z-tests on predicted probabilities. Primary statistical conclusions were based on regression-derived effect estimates and their corresponding confidence intervals. Participants Among 25,624 abstracts reviewed, 1,129 (4.2%) were HaH patients. Results HaH patients had a lower incidence of Hospital Harm (5.0% vs. 7.8%) including hospital-acquired delirium (0.8% vs. 2.0%), but a higher fall rate (20.4 vs. 8.2 per 1,000). HaH patients had a lower 30-day readmission rate (9.1% vs. 13.7%), and a lower 30-day mortality rate (1.0% vs. 2.8%). The Actual vs. Expected Length of Stay ratio was significantly higher for HaH patients (1.59 vs 1.19). Conclusions Victoria’s HaH program appears to offer a safe alternative to BAM hospitalization, with comparable or better clinical outcomes in several key areas. These findings support the expansion of HaH in Canada.