Discordance between non-zero physician's global scores and absence of active joints in juvenile idiopathic arthritis: multicenter vs. single-center cohorts

This study aims to compare the frequency of instances in which the physician's global assessment of disease activity (PhGA) was scored >0 despite the absence of active joints in children with juvenile idiopathic arthritis (JIA), using two multicenter patient datasets and one single-center dataset from a pediatric rheumatology center with expertise in clinimetric assessments.

Data were extracted from two multicenter datasets and one single-center dataset, comprising 9,081, 563, and 394 patients, respectively. Patients with an active joint count (AJC) of 0 were included. The PhGA and fulfillment of other criteria from the 2004 or 2011 Wallace definition of clinically inactive disease (CID) were assessed. UpSet plots were used to analyze the frequency and overlap of PhGA and CID items across the datasets.

Among patients with an AJC of 0, the percentage for whom a PhGA score >0 was the sole unmet CID criterion was 14.8% and 13.7% in the two multicenter datasets and 5.1% in the single-center dataset. The CID criteria that were most frequently not met when the PhGA was scored >0 were elevated acute-phase reactants (APRs) and morning stiffness lasting ≥15 min.

The discordance between the absence of active joints and a PhGA score >0 was less common in the single-center sample, suggesting that regular use and training may increase concordance between PhGA and AJC in patients without clinical signs of joint disease. APR elevation and parent-/patient-reported morning stiffness seemed to play a major role in prompting physicians to assign a non-zero global score.