Vaccination was associated with shorter hospital stays and lower observed in-hospital mortality among adults receiving disease-modifying antirheumatic drugs or biologic therapies who were hospitalized with COVID-19, according to a retrospective cohort study.
The analysis covered 254 patients at seven UPMC hospitals in Central Pennsylvania. There were 144 vaccinated patients and 110 unvaccinated patients. The result is an association, not proof that vaccination caused the difference.
A wide gap in hospital outcomes
The median hospital stay was 6 days for vaccinated patients, compared with 10 days for unvaccinated patients. The reported p-value was less than 0.0001.
The Results section also reported a 37% reduction in length-of-stay odds among vaccinated patients. Because the study was observational, that figure describes an association in the dataset, not proof that vaccination itself shortened each admission.
The difference in observed in-hospital mortality was larger. Death occurred in 6.25% of vaccinated patients and 27.27% of unvaccinated patients, with a reported p-value below 0.0001.
After adjustment for key demographic and clinical covariates, the study reported an adjusted odds ratio of 0.157 for death. An odds ratio compares the odds of an outcome between groups; the paper described this result as an 84% reduction in the odds of death, with a 95% confidence interval of 0.068 to 0.361.
The booster analysis was narrower
The researchers also compared vaccinated patients who had received at least one booster with vaccinated patients who had not. Of the vaccinated patients, 128 had received at least one booster. Their reported hospital stay was 6 days, compared with 7.5 days for those without a booster; the p-value was 0.0414.
That comparison covered hospital length of stay only. The supplied results do not provide a booster-specific mortality estimate, so they cannot show whether boosters were linked to a lower risk of death in this cohort.
A focused hospital-based study
Vaccination status came from electronic health records and registry-linked coding data. The analysis used multivariable logistic and gamma regression models, with adjustment for key demographic and clinical covariates.
Participants were adults aged at least 18 who were hospitalized with COVID-19 between March 2020 and April 2025 while receiving immunomodulatory therapy. Hospital length of stay was the primary endpoint, and in-hospital mortality was the secondary endpoint.
The study also assessed vaccine type and SARS-CoV-2 variants, but the supplied analysis does not give separate outcome estimates for them. The booster result likewise concerns hospital stay, not mortality.
The association has clear limits
Because this was retrospective observational research, it cannot establish that vaccination caused the differences in length of stay or mortality. The study came from one health system and included only hospitalized adults receiving immunomodulatory therapy, which limits how broadly the findings can be applied.
The paper's summaries also contain a numerical inconsistency. Its Results section reports an 84% reduction in adjusted odds of death, while a conclusion key point reports 87%. That discrepancy needs clarification in the full report.
The authors interpret vaccination as independently associated with shorter hospital stays and lower mortality, and describe it as a critical protective strategy for immunosuppressed populations. The evidence presented supports that association within this hospitalized cohort, not a causal verdict.
The version of record was published on 21 August 2026. The authors reported no specific funding for the work. Study data are not publicly available under UPMC policies but may be obtained from the authors upon reasonable request and with UPMC permission.
Paper data and sources
Original title: COVID-19 outcomes among vaccinated and unvaccinated patients receiving disease-modifying antirheumatic drugs and biologic therapies: a retrospective multihospital cohort study.
Authors: Dat Le, Yijin Wert
Journal/Repository: Clinical rheumatology
Status: Peer-reviewed
First online: 2026-08-21
DOI: 10.1007/s10067-026-08371-w
Original paper · Full text