Rheumatologist's expertise in estimating risk of developing rheumatoid arthritis in patients with clinically suspect arthralgia: what is the value?

INTRODUCTION

Clinical expertise is paramount in medical decision-making. In the setting of arthralgia, this expertise is highly relevant in differentiating clinically suspect arthralgia (CSA) from other musculoskeletal symptoms. However, it remains unclear whether rheumatological expertise is also reliable in estimating the risk of progression to rheumatoid arthritis (RA) in CSA patients. Using clinical expertise is more time-efficient than applying criteria, such as the European Alliance of Associations for Rheumatology/American College of Rheumatology (EULAR/ACR) risk stratification criteria for RA development. This study assessed the accuracy of RA-risk estimation based on rheumatologists' clinical expertise and compared this to these criteria.

METHODS

501 CSA patients from the Leiden CSA cohort and placebo arm of the TREAT EARLIER trial were studied. At baseline, rheumatologists estimated RA risk (0-10 Numeric Rating Scale) informed by history, physical examination and laboratory results. The EULAR/ACR risk stratification criteria (without imaging) were calculated using baseline data. The outcome was RA development (2010 criteria) within 1 year, and the discrimination was compared.

RESULTS

Based on their expertise, rheumatologists estimated the risk of RA as a mean of 5 (SD 1.6) on a 0-10 scale. With an area under the curve (AUC) of 0.64 (95% CI 0.55 to 0.73), patients who did or did not develop RA were moderately differentiated. The RA risk was mainly overestimated compared with observed progression rates. In comparison, the EULAR/ACR risk stratification criteria in the same patients yielded an AUC of 0.91 (95% CI 0.87 to 0.95).

CONCLUSIONS

Rheumatologists' clinical expertise is inaccurate in assessing the risk of developing RA in patients with CSA, due to risk overestimation. This may support the use of established criteria when risk information has clinical or therapeutic implications.

TRIAL REGISTRATION NUMBER: NTR4853-trial-NL4599.