Comparison of composite disease activity indices for rheumatoid arthritis

• Published in: Modern rheumatology Vol. 21; no. 2; pp. 134 - 143
• Main Authors: Matsui, Toshihiro, Kuga, Yoshiaki, Nishino, Jinju, Kaneko, Atsushi, Eto, Yoshito, Tohma, Shigeto
• Format: Journal Article
• Language: English
• Published: Japan Springer Japan 01.04.2011; Informa Healthcare
• Subjects: Comparative studies; Disease management; Indexes; Medicine; Medicine & Public Health; Original Article; Orthopedics; Rheumatoid arthritis; Rheumatology; DAS28; CDAI; Rheumatoid arthritis; Disease activity index; SDAI
• DOI: 10.1007/s10165-010-0367-6
• ISSN: 1439-7595

To evaluate the composite disease activity indices for rheumatoid arthritis (RA), we compared disease activities and the changes therein calculated using the Disease Activity Score based on 28 joint counts using erythrocyte sedimentation rate (DAS28-ESR), DAS28-CRP (C-reactive protein), Simplified Disease Activity Index (SDAI), and Clinical Disease Activity Index (CDAI) in a cohort of 1,412 patients with RA. The median (1st; 3rd quartile) scores were 4.20 (3.31; 5.14) for DAS28-ESR, 3.44 (2.59; 4.36) for DAS28-CRP, 13.6 (7.49; 21.1) for SDAI, and 12.0 (6.9; 18.9) for CDAI. Absolute scores and their changes were significantly correlated ( p  < 0.0001) in all combinations among these four disease activity indices; however, their correlations were lower in males than in females. Correlations between disease activity indices and the clinical and acute phase reactant variables were different according to disease activity index, sex and age. A comparison of the number of patients in each disease activity category according to the disease activity indices using kappa-statistics revealed an almost perfect agreement between SDAI and CDAI (κ = 0.871), a moderate agreement between DAS28-ESR and SDAI (κ = 0.415) or CDAI (κ = 0.427), but only fair agreement between DAS28-ESR and DAS28-CRP (κ = 0.329). For the selection of a disease activity index for an evaluation of RA patients, both the convenience and the characteristics of the respective disease activity index should be considered.