Trends in autoimmune rheumatic disease diagnoses before and after the COVID-19 pandemic in England: a population-based cohort study using OpenSAFELY

Lay summary

In this study, we used data in OpenSAFELY to describe how the COVID-19 pandemic impacted the number of people with newly recorded diagnoses of 10 rheumatological conditions in England. We showed that the number of newly recorded diagnoses fell for most rheumatological conditions during the first year of the pandemic. For many conditions, the number of diagnoses recovered by the second and third years of the pandemic. However, some diseases were more affected than others: persistent deficits in diagnoses were particularly evident for psoriatic arthritis and giant cell arteritis, whereas the number of newly recorded diagnosis has increased more than expected for axial spondyloarthritis. Importantly, this study demonstrates how routinely collected health data can be used to monitor inequity across diseases, and inform strategies to increase case capture and reduce diagnostic delay.

Abstract

Background: During the first year of the COVID-19 pandemic, diagnosis rates were reported to have declined for several autoimmune rheumatic diseases in numerous countries. It remains unclear whether diagnosis rates have since returned to pre-pandemic levels. This study aimed to evaluate diagnosis rates across autoimmune rheumatic diseases in the pre-pandemic and post-pandemic periods, with comparisons by disease, age group, sex, ethnicity, and socioeconomic status.

Methods: With NHS England approval, this population-level, observational cohort study used primary care and hospital admission data from all adults (aged ≥18 years) in England registered with general practices using TPP health record software, via the OpenSAFELY platform. Incident diagnosis rates for ten autoimmune rheumatic diseases were evaluated between April 1, 2016, and March 31, 2025. Expected diagnosis rates after pandemic onset (March, 2020) were modelled using Prophet time-series forecasting and compared with observed rates.

Findings: 23 353 040 adults had data available for analysis and were included in the reference population. Mean age was 46·2 years (SD 18·8). 11 776 490 (50·4%) of 23 353 040 individuals were female and 11 576 545 (49·6%) were male; 18 581 585 (79·6%) were White, 1 923 140 (8·2%) were Asian or Asian British, 700 025 (3·0%) were Black or Black British, 332 490 (1·4%) were mixed ethnicity, 691 475 (3·0%) were Chinese or other ethnicity, and 1 124 325 (4·8%) had unknown ethnicity. In the first year of the pandemic, diagnosis rates sharply declined across all autoimmune rheumatic diseases except for small vessel vasculitis. As of March, 2025, cumulative reductions in diagnoses remained for psoriatic arthritis (–12 320 diagnoses, 95% prediction interval [PI] –12 780 to –11 860; percentage difference –24·9%, 95% PI –25·6 to –24·2), rheumatoid arthritis (–11 690 diagnoses, –13 280 to –10 110; –9·8%, –11·0 to –8·6), and giant cell arteritis (–5660 diagnoses, –6130 to –5200; –15·2%, –16·3 to –14·1). By contrast, axial spondyloarthritis diagnoses increased above pre-pandemic levels (2040 diagnoses, 1670 to 2410; 8·9%, 7·1 to 10·7), with diagnosis rates in women exceeding those in men from 2023 onwards. Early diagnostic deficits in connective tissue diseases were followed by compensatory increases during the pandemic recovery period.

Interpretation: Persistent reductions in diagnoses remain evident for psoriatic arthritis, rheumatoid arthritis, and giant cell arteritis 5 years after the onset of the pandemic; contrasting a post-pandemic increase in axial spondyloarthritis diagnoses, particularly in women. Data-driven disease surveillance could be used to identify drivers of these changes and address the long-term impact of delayed diagnosis.