Background/Purpose: A new approach to the measurement of disease activity in juvenile idiopathic arthritis (JIA) is based on the Juvenile Arthritis Disease Activity Score (JADAS). The feasibility of the JADAS for use in standard clinical practice may be enhanced by implementing a 3-item version, which does not include the acute phase reactant. Inflammatory markers frequently are not obtained or available at a visit. The lack of this information would hinder the potential to make immediate therapeutic decisions based on the JADAS and, thus, the potential benefit of intensifying therapy. As for the original tool, the use of the cJADAS as a guide to pursuing tight disease control requires cutoff values to identify the target disease activity states. Aim of the study was to determine cutoff values in the cJADAS that correspond to the states of inactive disease (ID), minimal disease activity (MDA) and high disease activity (HDA), and to the parent/child acceptable symptom state (PASS/CASS).

Methods: For the selection of cutoff values, data from a clinical database including 609 patients were used. Optimal cutoffs were determined against external criteria by calculating the 75th percentile of cumulative score distribution for ID, MDA, PASS and CASS and the 25th percentile cumulative score distribution for HDA, and through receiver operating characteristic curve analysis. External criteria were based on formal definitions for ID and MDA, on the subjective rating of satisfaction with illness outcome by parents and patients for PASS and CASS, and on the therapeutic decision made by the attending physician at the time of the visit for HDA. MDA and HDA cutoffs were developed separately for oligoarthritis and polyarthritis. The choice of optimal cutoffs was made on clinical and statistical grounds. Cross-validation was performed using 5 JIA patient samples that included a total of 1,421 patients, and was based on assessment of construct, discriminant, and predictive validity.

Results: The table shows the selected cutoff values for the cJADAS10 and the sensitivity (se) and specificity (sp) of each cutoff. In cross-validation analyses, the cutoffs revealed strong ability to discriminate between different levels of ACR Pedi response in 2 clinical trials (one on methotrexate and one on abatacept) and revealed good concordance with the subjective assessment of the disease state (remission, continued activity or flare) by the physicians and the parents. Furthermore, they proved able to predict functional and radiographic outcomes.

	cJADAS10 cutoff (se/sp)
Inactive disease	1 (84.1/87.0)
MDA oligoarthritis	1.5 (75.4/91.7)
MDA polyarthritis	3.5 (96.8/94.7)
PASS	4 (75.6/91.4)
CASS	3.5 (77.5/86.5)
HDA oligoarthritis	7 (74.9/93.5)
HDA polyarthritis	11 (76.9/93.2)

Conclusion: The cutoff values in the cJADAS that correspond to the states of ID, MDA, PASS, CASS, and HDA in JIA were developed. In cross-validation analyses, they proved to have strong construct and discriminant validity and ability to predict disease outcome.

---

« Back to ACR/ARHP Annual Meeting 2013

ACR Meeting Abstracts - https://acrabstracts.org/abstract/development-and-validation-of-the-cutoff-values-for-disease-activity-states-of-the-three-item-version-of-the-juvenile-arthritis-disease-activity-score/