Defining Criteria for Disease Activity States in Systemic Juvenile Idiopathic Arthritis Based on the Systemic Juvenile Arthritis Disease Activity Score

Abstract

Objective: Our objective was to develop and validate cutoff values in the systemic Juvenile Arthritis Disease Activity Score 10 (sJADAS10) that distinguish the states of inactive disease (ID), minimal disease activity (MDA), moderate disease activity (MoDA), and high disease activity (HDA) in children with systemic juvenile idiopathic arthritis, based on subjective disease state assessment by the treating pediatric rheumatologist. Methods: The cutoff definition cohort was composed of 400 patients enrolled at 30 pediatric rheumatology centers in 11 countries. Using the subjective physician rating as an external criterion, six methods were applied to identify the cutoffs: mapping, calculation of percentiles of cumulative score distribution, the Youden index, 90% specificity, maximum agreement, and receiver operating characteristic curve analysis. Sixty percent of the patients were assigned to the definition cohort, and 40% were assigned to the validation cohort. Cutoff validation was conducted by assessing discriminative ability. Results: The sJADAS10 cutoffs that separated ID from MDA, MDA from MoDA, and MoDA from HDA were ≤2.9, ≤10, and >20.6, respectively. The cutoffs discriminated strongly among different levels of pain, between patients with and without morning stiffness, and among patients whose parents judged their disease status as remission or persistent activity or flare or were satisfied or not satisfied with current illness outcome. Conclusion: The sJADAS cutoffs revealed good metrologic properties in both definition and validation cohorts and are therefore suitable for use in clinical trials and routine practice.