Registry data map who fared worst when COVID-19 hit inflammatory arthritis and psoriasis
A large physician-reported analysis from the COVID-19 PsoProtect and Global Rheumatology Alliance registries has drawn a clearer line around which patients with psoriasis, psoriatic arthritis and axial spondyloarthritis were most likely to experience severe COVID-19. The dataset included 5,045 adult cases and used a three-level ordinal severity scale: no hospitalisation, hospitalisation without death, and death. Most patients stayed out of hospital, but the gradient in risk became obvious once age, sex, comorbidity burden, disease activity and treatment exposure entered the model.
The cohort itself was divided across three closely related immune-mediated diseases: 18.3% had psoriasis, 45.5% had psoriatic arthritis and 36.3% had axial spondyloarthritis. Across the full sample, 83.6% were not hospitalised, 14.6% were hospitalised and 1.8% died. Those proportions matter because they place the analysis squarely in the clinical territory that rheumatologists and dermatologists faced during repeated pandemic waves: deciding which patients might need closer surveillance, tighter disease control, or faster escalation of preventive care when respiratory viruses were circulating widely.
Older age was non-linearly associated with worse COVID-19 severity, a familiar pattern in infectious disease but still clinically important in this immune-mediated cohort. Male sex carried higher odds of severe disease, with an OR of 1.54 and a 95% CI of 1.30 to 1.83. Several comorbidity groups also tracked with worse outcomes, including cardiovascular, respiratory, renal, metabolic and cancer comorbidities, with ORs ranging from 1.25 to 2.89. Each of those conditions can narrow physiologic reserve in a different way: vascular disease and metabolic dysfunction amplify thromboinflammatory stress, chronic lung disease limits respiratory compensation, kidney disease reduces buffering capacity during acute illness, and cancer often signals both frailty and treatment-related immune compromise.
Disease control itself was not a passive background feature. Moderate or high disease activity, and glucocorticoid use, were both associated with increased odds of severe COVID-19 compared with remission or low disease activity and no glucocorticoids, with ORs from 1.39 to 2.23. In practical terms, uncontrolled systemic inflammation and steroid exposure appear to travel together as a risk cluster. Ongoing inflammation may worsen endothelial dysfunction and alter immune responses to viral infection, while glucocorticoids can blunt antiviral defense when they are already needed for inflammatory disease control. A patient entering a respiratory pandemic with active disease and chronic steroid exposure starts from a less forgiving baseline.
Male sex increased risk.
Treatment exposure and later pandemic periods were linked to lower odds of severe disease
Against that higher-risk background, several variables were associated with lower odds of severe COVID-19. Later pandemic time periods were linked to reduced severity, with ORs of 0.42 to 0.52 compared with the period up to 15 June 2020. This likely reflects a mix of factors rather than a single explanation: changing clinical practice, better recognition of COVID-19, improved inpatient management, vaccination-era effects in later periods, and differences in testing and case capture as the pandemic evolved. The registry design cannot disentangle those forces completely, but the directional trend is clinically intuitive.
Disease category also mattered. Compared with psoriatic arthritis, psoriasis was associated with lower odds of severe COVID-19, with an OR of 0.49 and a 95% CI of 0.37 to 0.65. The finding may reflect differences in systemic inflammatory burden, patient characteristics, or treatment patterns across the three disease groups. Psoriatic arthritis and axial spondyloarthritis often bring a heavier burden of systemic involvement and immunomodulatory treatment than skin-only disease, which may partly explain why the risk profile differed across diagnoses.
Baseline biologic exposure was associated with lower severity as well. Treatment with tumor necrosis factor inhibitors was linked to reduced odds of severe COVID-19, with an OR of 0.57 and a 95% CI of 0.44 to 0.73, compared with no disease-modifying antirheumatic drug. IL-17 inhibitors carried an OR of 0.62 with a 95% CI of 0.45 to 0.87, and IL-23 inhibitors or IL-12/23 inhibitors had an OR of 0.67 with a 95% CI of 0.45 to 0.98. These findings fit a biologically plausible pattern: targeted cytokine blockade may control the inflammatory disease without the broader immunosuppressive burden of uncontrolled systemic inflammation or chronic glucocorticoid exposure. In other words, the signal does not suggest that all immune modulation is harmful during viral infection; it suggests the opposite may be true when inflammation is better controlled and steroids are avoided.
Risk stratification now has a sharper bedside edge
This registry analysis gives clinicians a usable framework for anticipating who may do poorly during COVID-19 waves or future respiratory pandemics. The highest-risk profile clustered around older age, male sex, multiple comorbidities, active disease and glucocorticoid exposure. Patients with psoriasis alone, and those receiving TNF, IL-17 or IL-23 pathway inhibitors, showed lower odds of severe outcomes in this dataset. None of those associations should be treated as proof of causality, but they are strong enough to inform counseling, monitoring and shared decision-making.
For day-to-day practice, the implications are concrete. A patient with psoriatic arthritis, cardiometabolic disease and ongoing steroid use should not be viewed in the same way as a younger person with stable plaque psoriasis on a targeted biologic. The registry data support a more granular approach to risk assessment, especially when viral transmission rises and treatment interruptions, delayed care, or steroid escalation are being considered. The biologic therapies were not associated with worse COVID-19 severity here; the heavier risk burden came from the host factors and disease state surrounding them.
The study also leaves several questions open. The physician-reported registry format is well suited to rapid international capture, but it cannot remove all confounding, and the later-pandemic risk reduction may blend multiple eras of care and viral conditions. Even so, the central message is clinically durable: in people with psoriasis, psoriatic arthritis and axial spondyloarthritis, severe COVID-19 tracked more strongly with age, comorbidity, active disease and glucocorticoid exposure than with targeted biologic treatment.