Article Text

FRI0203 Is Patient-Physician Discordance in Global Assessment in Early Spondyloarthritis Varying Over Time: The Desir Cohort
  1. C. Desthieux1,
  2. A. Molto2,
  3. A. Saraux3,
  4. B. Fautrel1,
  5. L. Gossec1
  1. 1Rheumatology, Pitié-Salpêtrière Hospital - Paris 6 - Pierre et Marie Curie University - GRC 08 - IPLESP
  2. 2Rheumatology, Descartes University - Cochin Hospital, Paris
  3. 3Rheumatology, CHU Brest – La Cavale Blanche Hospital, Brest, France


Background In the management of chronic diseases, recommendations insist on the need to work in partnership with the patient. Discordance in the assessment of disease by different stakeholders is problematic in any chronic inflammatory disease, with regards to treatment decisions and shared decision-making. It is critical in spondyloarthritis (SpA) care where objective measures are often limited and subjective or patient-reported information becomes the only available for treatment decision making. Furthermore, there may be a reference shift in early disease leading to changes in discordance over time.

Objectives The aim of our study was to assess patient-physician discordance in global assessment of disease in early SpA and its evolution over time.

Methods DESIR (Devenir des Spondyloarthropathies Indifférenciées Récentes) is a French, multicenter, longitudinal cohort of patients with early inflammatory back pain suggestive of axial SpA. Patient- and Examiner-Global Assessment (PGA and EGA) were rated with a 0-10 numeric scale, every 6 months during 3 years. PGA and EGA were assessed respectively as “How do you rate the effect of your illness on your general condition during the previous week?” and “Overall assessment of the activity of the rheumatic disease”. Discordance was defined by the absolute difference PGA–EGA>2 (1). Mean PGA, EGA and discordance were analysed at each visit. Patients consistently discordant were calculated as an agreement statistic (kappa coefficient) between inclusion and 36 months. There was no imputation of missing data.

Results A total of 437 patients had complete data available over 36 months. The mean age was 34.5±8.7 years. 55.2% were female and mean symptoms duration was 18.0±10.7 months. Mean PGA values were always higher than mean EGA values (e.g. at 12 months, respectively 4.2±2.7 versus 3.2±2.3). PGA and EGA decreased over follow up (mean decrease, 1.4 (27.3%) for PGA, 1.6 (36.6%) for EGA). In all, 123 (28.2%) patients had discordance at inclusion and there was no change over time with also 123 patients with discordance at 36 months. Agreement statistics for discordance between inclusion and 36 months indicated it is not always the same patients who have discordance (kappa=0.03). Indeed, among patients with discordance at least once at inclusion or 36 months, 82.3% were discordant only once.

Conclusions PGA was around 4/10 and always higher than EGA in early SpA over 3 years. Around 28% of patients had discordance between PGA and EGA in early SpA, which is rather lower than discordance described in rheumatoid arthritis (around 36%). Over 3 years of follow up, although patients probably had a better understanding of their disease, discordance in global assessment did not decrease (no “reference shift”). Over 2 time points, 3 years apart, most patients with discordance were only discordant once, indicating that discordance is not a stable trait. This goes against the hypothesis that discordance could be a marker of personality or of comorbidity, e.g., widespread pain syndrome.


  1. Khan et al. Arthritis Care Res. févr 2012;64(2):206-14.

Disclosure of Interest None declared

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.