This population-based cohort study assessed the comparative burden of COVID-19 in adults with systemic autoimmune rheumatic diseases (SARDs) versus immunocompetent adults with and without chronic comorbidities in Quebec, Canada, across major pandemic phases from March 2020 to April 2023. The primary focus was on SARS-CoV-2 infection incidence and severe outcomes defined as hospitalization and intensive care unit admission or death (ICU/death).
Researchers used administrative health datasets to assemble three adult cohorts (>18 years): 81,434 individuals with SARDs, 2,020,356 individuals with chronic comorbidities, and 3,144,700 individuals without comorbidities. Incidence estimates for SARS-CoV-2 infection and severe COVID-19 were derived as crude and age- and sex-standardized rates.
Comparative risk analyses employed Cox proportional hazards models to derive adjusted hazard ratios (aHRs) for SARS-CoV-2 infection, hospitalization, and ICU/death. Analyses were stratified by three pandemic periods: pre-vaccination (March–December 2020), vaccination pre-Omicron (December 2020–December 2021), and Omicron (December 2021–April 2023).
Data access was restricted by provincial privacy rules; the Quebec Ministry of Health and the Régie de l'assurance maladie du Québec (RAMQ) hold the data. The study was conducted under the legal mandate of the Quebec National Director of Public Health, with Research Ethics Board approval from CHU de Québec–Université Laval and a waiver of informed consent under that mandate. Study code is available from the authors upon reasonable request.
Cohort sizes were large: 81,434 adults with SARDs, 2,020,356 adults with comorbidities, and 3,144,700 adults without comorbidities. Across pandemic phases, adjusted SARS-CoV-2 infection rates were similar between the SARDs group and adults without comorbidities.
However, risks for severe outcomes were consistently higher among individuals with SARDs. During the pre-vaccination period, the SARDs cohort demonstrated markedly elevated risks relative to adults without comorbidities: hospitalization aHR 3.4 (95% CI 2.8–4.1) and ICU/death aHR 4.0 (95% CI 2.8–5.7). These higher risks persisted into the Omicron period: hospitalization aHR 4.5 (95% CI 4.0–4.9) and ICU/death aHR 4.4 (95% CI 3.6–5.4).
The excess burden among SARDs was evident despite broadly comparable infection rates, indicating that disease status and related factors influenced progression to severe disease rather than acquisition alone.
Among SARD diagnostic groups, rheumatoid arthritis, psoriatic arthritis, and systemic lupus erythematosus accounted for the greatest contribution to COVID-19 burden. The study reports that hospitalization and mortality among younger adults with rheumatoid arthritis were comparable to those observed in older adults without comorbidities, underscoring age alone did not fully account for risk differences.
Treatment-related risk stratification identified patients receiving highly immunosuppressive therapies as having the greatest risk. Those treated with rituximab or mycophenolate experienced the highest hospitalization and ICU/death hazards, particularly during the Omicron period: hospitalization aHR 11.9 (95% CI 9.9–14.3) and ICU/death aHR 17.6 (95% CI 13.0–23.9). These treatment-specific aHRs indicate substantially elevated risk relative to the reference group (adults without comorbidities).
The findings indicate that adults with SARDs faced a persistently elevated risk of severe COVID-19 outcomes throughout the pandemic phases analyzed, even when infection rates were similar to immunocompetent peers. The disproportionate severity was most pronounced among patients on highly immunosuppressive regimens such as rituximab and mycophenolate.
Clinically and from a public health perspective, these results support prioritizing individuals with SARDs—and especially those receiving potent immunosuppression—when planning protective measures, vaccination strategies, therapeutic access, and targeted interventions during future public health emergencies.
The study is a preprint and has not been peer-reviewed; authors explicitly note it should not be used alone to guide clinical practice.
Data are not publicly available because of privacy restrictions. The Quebec Ministry of Health and RAMQ retain the datasets and limit access to authorized personnel. Requests for data access should be directed to the Ministère de la Santé et des Services sociaux du Québec. The study protocol received Research Ethics Board approval from CHU de Québec–Université Laval, and the study was performed under the legal public health mandate that allowed a waiver of informed consent. The study code can be requested from the authors.
The study reports funding from the Canadian Immunization Research Network and support from the Ministère de la Santé et des Services sociaux du Québec. Specific authors disclosed additional funding sources for this work or unrelated work as reported in the competing interest statement. Other authors declared no conflicts of interest. The preprint is posted under a CC-BY-NC-ND 4.0 International license.