Long COVID Risk in Autoimmune Rheumatic Disease: USMLE Review

A retrospective cohort study published online Oct. 14 in Arthritis & Rheumatology found that patients with autoimmune rheumatic diseases (AIRD) had a higher risk of long COVID than matched patients without AIRD. In the matched cohort, long COVID occurred in 2.8% of patients with AIRD and 1.5% of those without (adjusted hazard ratio 1.56). The finding is an association from observational data, but it touches several topics that show up on USMLE: autoimmunity, immunosuppressive drugs, and post-infectious syndromes.

What happened

Namrata Singh, M.D., of the University of Washington in Seattle, and colleagues used data from the National Clinical Cohort Collaborative to estimate how common long COVID is in patients with AIRD and which factors are linked to it. They compared patients with AIRD against non-AIRD controls using 1:1 nearest-neighbor propensity score matching.

The matched cohort included 187,106 patients. Long COVID occurred in 2.8% of the AIRD group and 1.5% of the non-AIRD group. In a multivariable analysis, AIRD was associated with a higher risk of long COVID (adjusted hazard ratio, 1.56).

Within the AIRD group, these factors were associated with a higher risk of long COVID:

• Higher acute COVID-19 severity
• Pulmonary disease
• Obesity
• Exposure to rituximab or glucocorticoids

Sjögren syndrome was also associated with a higher risk. The authors called for targeted monitoring and further research into mechanisms and prevention strategies for long COVID in patients with autoimmune diseases.

Why it matters for long COVID and autoimmune disease

Long COVID, also called post-acute sequelae of SARS-CoV-2 infection (PASC), is a heterogeneous syndrome. Patients can have fatigue, dyspnea, cognitive complaints, and other symptoms that persist well after the acute infection. Researchers still debate its mechanisms. Proposed hypotheses include persistent viral antigen, immune dysregulation, autoantibody production, and endothelial or microvascular injury. These are hypotheses, and this study did not test them.

Patients with AIRD are an obvious group to examine. They already have dysregulated immunity, many take immunosuppressants, and they may have organ involvement, such as lung disease, that overlaps with post-COVID symptoms. This study suggests that, in a large US dataset, AIRD patients were more likely to be labeled with long COVID. It does not show why.

Note the absolute numbers too. The difference between 2.8% and 1.5% is 1.3 percentage points. A hazard ratio of 1.56 describes relative risk over time in the model. It does not mean 56% of patients are affected. Both the relative and absolute measures help you interpret the result.

What to know for the USMLE: long COVID, AIRD, and immunosuppression

• Autoimmune rheumatic diseases include rheumatoid arthritis (anti-CCP, RF), SLE (ANA, anti-dsDNA, anti-Smith), and Sjögren syndrome (anti-Ro/SSA, anti-La/SSB). The source reports an association with Sjögren syndrome specifically. Other individual diseases are not detailed in the available summary.
• Sjögren syndrome classically presents with keratoconjunctivitis sicca and xerostomia from lymphocytic infiltration of lacrimal and salivary glands. It carries an increased risk of B-cell lymphoma (often MALT type).
• Rituximab is a monoclonal antibody against CD20 on B cells. It causes B-cell depletion and can lead to hypogammaglobulinemia, blunted vaccine responses, hepatitis B reactivation, and, rarely, PML. Know that it is used in RA, ANCA-associated vasculitis, and B-cell lymphomas.
• Glucocorticoids inhibit NF-κB and reduce transcription of pro-inflammatory cytokines. They cause lymphopenia, demargination-related neutrophilia, and eosinopenia. Long-term effects include adrenal suppression, osteoporosis, hyperglycemia, and cataracts. Immunosuppression from steroids increases infection risk generally.
• Post-infectious syndromes are a recurring theme: reactive arthritis, post-streptococcal glomerulonephritis, Guillain-Barré syndrome after Campylobacter, and now post-viral syndromes like PASC. Be ready to distinguish an acute infection from a delayed immune-mediated or post-infectious process.
• Study design: This is a retrospective cohort study, which compares exposed and unexposed groups using existing data. It gives relative risk or hazard ratios and cannot establish causation.
• Propensity score matching reduces measured confounding by balancing covariates between groups. It cannot fix unmeasured confounding.
• Hazard ratio is the relative rate of an event over time, usually from a Cox proportional hazards model. An HR above 1 means a higher hazard in the exposed group.
• Confounding vs. effect modification: Obesity, pulmonary disease, and acute severity may act as confounders or mediators here. Exam questions often test whether you can tell these apart.

Limitations

Several caveats apply, and some are specific to what we can see in the source.

It is an association. This was a retrospective cohort study, so it cannot show that AIRD, or any drug, causes long COVID. Propensity score matching balances measured variables only.

Drug exposure and disease activity are tangled. Rituximab and glucocorticoid exposure were associated with higher long COVID risk among AIRD patients. But patients receiving these drugs often have more active or severe disease, and that may drive the association rather than the drugs themselves. The summary does not tell us how the authors handled this, so avoid reading the finding as "these drugs cause long COVID."

Details are limited. The available source does not specify how long COVID was defined or ascertained, the follow-up period, which AIRD diagnoses were included beyond Sjögren syndrome, or vaccination status. Any of these could affect interpretation. In an electronic health record-based dataset, how the condition is identified and coded matters.

Overlapping symptoms. Fatigue, dyspnea, and pain are common in rheumatic disease and in long COVID. In general, it can be hard to separate a flare or baseline disease symptoms from a post-COVID syndrome. Whether that applied here cannot be judged from the summary.

Absolute risk is low. Even in the AIRD group, 2.8% developed long COVID in this analysis. This is a population-level finding, not a prediction for any individual patient, and it is not a basis for personal medical decisions.

FAQ

Do autoimmune diseases cause long COVID?

This study cannot say. It found that AIRD was associated with a higher risk of long COVID, but it was a retrospective cohort study. Observational data can show a link, not a cause.

Which factors were linked to long COVID in patients with AIRD?

Higher acute COVID-19 severity, pulmonary disease, obesity, and exposure to rituximab or glucocorticoids were associated with a higher risk. Sjögren syndrome was also associated with a higher risk.

What does an adjusted hazard ratio of 1.56 mean on an exam?

It means that, after adjusting for covariates in the model, the hazard of long COVID was about 1.56 times higher in the AIRD group than in the non-AIRD group. It is a relative measure. The absolute rates in this study were 2.8% versus 1.5%.

Key takeaways

• In a matched US cohort of 187,106 patients, long COVID occurred in 2.8% of those with autoimmune rheumatic disease versus 1.5% of those without (adjusted HR 1.56). This is an observational association.
• Among AIRD patients, acute COVID-19 severity, pulmonary disease, obesity, Sjögren syndrome, and rituximab or glucocorticoid exposure were linked to higher risk. Confounding by disease severity and treatment indication cannot be excluded.
• For USMLE, connect this to rituximab (anti-CD20), glucocorticoid effects, Sjögren syndrome (anti-Ro/La), and study-design concepts: retrospective cohort, propensity matching, hazard ratio, and confounding.

Source: Medical Xpress report on Singh N, et al., Arthritis & Rheumatology (2026), DOI: 10.1002/art.70361. This article is for education and is not medical advice.