Background The OMERACT group proposed minimal disease activity (MDA) as a treatment target, given the current treatment possibilities and limitations. Whereas a Disease Activity Score 28 (DAS28)-based definition of MDA has been proposed1, definitions based on the Simplified Disease Activity Index (SDAI), Clinical Disease Activity Index (CDAI), and Routine Assessment of Patient Index Data 3 (RAPID3) have not, despite the increasing use of these indices.
Objectives To define SDAI, CDAI, and RAPID3-based definition of MDA for use in routine care.
Methods We analyzed 15,101 patients registered in the Japanese National Database (NinJa 2015). As the OMERACT group proposed, patients with tender joint count (TJC) of 0, swollen joint count (SJC) of 0, and erythrocyte sedimentation rate (ESR) ≤10 mm/hour or patients with five of the following seven criteria, namely, pain ≤2, SJC ≤1, TJC ≤1, HAQ ≤0.5, physician's global ≤1.5, patient's global ≤2, and ESR ≤20, were considered to be in MDA.1 The ROC curve was used to obtain the best cut-off points for the SDAI, CDAI, and RAPID3-based definitions of MDA, which emerged as good predictors of MDA as defined by the core dataset. To compare the usefulness of the indices, the interclass correlation of MDA in DAS28, SDAI/CDAI, and RAPID3 was compared to that of low disease activity (LDA).
Results 57.6% of patients (5,629/9,767) were categorized as having MDA, and 29% of patients (4,003/13,781) were categorized as Boolean remissions. In the ROC analysis, the area under the curve for DAS28, SDAI, CDAI, and RAPID3 was 0.911, 0.955, 0.953, and 0.930, respectively. Based on the Youden index, SDAI ≤5.3, CDAI ≤4.8, and RAPID3 ≤5 were defined as SDAI, CDAI, and RAPID3-based MDA, respectively. The sensitivity and specificity of the DAS28, SDAI, CDAI, and RAPID3-based definitions were higher than those of the DAS28-based definition, with a sensitivity of 81.5%, 89.2%, 88.8%, and 90.0%, respectively, and a specificity of 83.5%, 89.4%, 89.9%, and 88.4%, respectively. Each index-based definition of MDA showed better interclass correlation than that of LDA; DAS28 vs CDAI/SDAI: MDA of 0.643/0.662 and LDA of 0.540/0.540; DAS28 vs RAPID3: MDA of 0.541 and LDA of 0.482; CDAI/SDAI vs RAPID3: MDA of 0.677/0.671 and LDA of 0.433/0.425.
Conclusions SDAI ≤5.3, CDAI ≤4.8, and RAPID3 ≤5, values two points higher than each remission criterion, may provide a more stringent therapeutic goal than LDA in clinical practice.
Wells GA, et al. Minimal disease activity for rheumatoid arthritis: a preliminary definition. J Rheumatol 2005;32:2016–24.
Acknowledgements Supported in part by a Health and Labor Sciences Research Grant from the Ministry of Health, Labor, and Welfare of Japan.
Disclosure of Interest None declared