abstract details

The summaries are free for public use. ARTHROS will continue to add and archive summaries of articles deemed relevant to ARTHROS by our Faculty.

Variation and implications of treatment decisions in early rheumatoid arthritis: results from a nationwide cohort study


Yates M1, MacGregor AJ2, Ledingham J3, Norton S4, Bechman K1, Dennison EM5, Galloway JB1. Rheumatology (Oxford). 2019 Dec 5. pii: kez550. doi: 10.1093/rheumatology/kez550. [Epub ahead of print]

Author Information

1 The Centre for Rheumatic Diseases, School of Immunology, Infection & Inflammatory Disease, King's College London, London.

2 Norwich Medical School, University of East Anglia, Norwich.

3 Department of Rheumatology, Queen Alexandra Hospital, Portsmouth.

4 Institute of Psychiatry, Kings College London, London.

5 MRC Epidemiology Unit, University of Southampton, Southampton, UK.



Trial data have provided an evidence base to guide early treatment in RA. Few studies have investigated rheumatologists' adherence to guidelines, and subsequent impact on outcomes. The objectives of this study are to characterize baseline prescribing for patients with RA across the National Health Service, identifying treatment decisions that associate with patient outcomes.


A nationwide audit of RA collected information on treatment choices, DAS and sociodemographic factors at baseline. Treatment response was assessed at 3 months. Multilevel regression models were used to characterize departmental variations in prescribing. Heat maps were used to visualize geographical variation. Mixed effects regression models were constructed to assess the relationship between treatment decisions and disease outcomes, adjusting for patient and department level covariates.


A total of 7154 patients with a diagnosis of RA were recruited from 136 departments. There was broad variation in prescribing choices, even between departments close to one another, with evidence of substantial deviation from guidelines. Over 75% of patients received glucocorticoids, fewer than half received combination conventional DMARDs. Early glucocorticoid therapy associated with achieving a good treatment response [odds ratio 1.93 (95% CI 1.31, 2.84), P-value = 0.001]. The association was maintained following propensity modelling and imputation.


Guideline adherence varies between departments and cannot be explained by case-mix alone. Departments that prescribe early adjunctive steroid achieve better short-term outcomes. Further research should work to ensure that the early arthritis evidence base translates into better outcomes for patients.