Characterizing the clinical and economic burden of COVID-19 among individuals with immunocompromising conditions in Ontario, Canada - a matched, population-based observational study.

Abstract

Aims: COVID-19 continues to be associated with substantial burden among immunocompromised patients (IC). This study aimed to describe and compare outcomes during and following COVID-19 hospitalizations among IC and non-IC patients.

Methods: Patients hospitalized with COVID-19 (January 2020-March 2023) were identified in Ontario health administrative claims databases. All eligible IC (≥1 of solid organ or stem cell transplant; hematological malignancy; rheumatoid arthritis; multiple sclerosis; or primary immunodeficiency) were matched (1:4) to eligible non-IC. Clinical, resource, and costburden were assessed during and post-hospitalization. Multivariate regression models were used to estimate relative risks (RRi), rates (RRa), and corresponding 95% confidence intervals (CIs), adjusting for neighborhood deprivation, long-term care residency, baseline comorbidities, and COVID-19 vaccination status.

Results: 9,283 IC hospitalized (mean age 68.7 years; 52.1% female) were matched to 37,127 non-IC. During index hospitalization, IC had greater risks of intensive care unit admission (RRi = 1.06 [1.01-1.12]), ventilation (RRi = 1.27 [1.19-1.36]), and all-cause mortality (RRi = 1.34 [1.27-1.41]) compared to non-IC. Within 30-days post-discharge, IC had greater rates of all-cause readmission (RRa = 1.33 [1.26-1.40]), emergency departments admission (RRa = 1.13 [1.08-1.18]), home oxygen use (RRi = 1.35 [1.15-1.58]), and COVID-19-related rehabilitation (RRa = 1.52 [1.22-1.89]), resulting in 21% (16%-25%) and 51% (45%-58%) greater costs in hospital and post-discharge, respectively. All-cause mortality remained approximately 5% higher for IC vs. non-IC at 30- and 60-days post-discharge (p < .001). Resource use remained elevated among IC with 57% (50%-64%) greater costs within 180 days post-discharge.

Limitations: Unmeasured confounding remains; hospital prescription data were not available such that treatments for COVID-19 were not captured. Attribution of post-discharge resource use and costs to COVID-19 was subject to greater uncertainty further from the index hospitalization.

Conclusion: IC experienced more severe COVID-19 hospitalization outcomes compared to non-IC. COVID-mitigating policies and prophylactic treatments are needed to protect immunocompromised populations.