Chinese registry of rheumatoid arthritis: IV. Correlation and consistency of rheumatoid arthritis disease activity indices in China

Background: Disease activity indices (DAIs) including disease activity score 28 (DAS28), simplified disease activity index (SDAI), and clinical disease activity index (CDAI) have been widely used in clinical practice and research studies of rheumatoid arthritis (RA). The objective of our study was to evaluate the correlation and concordance among different DAIs in Chinese patients with RA. Methods: A cross-sectional study, including patients enrolled in the Chinese registry of rheumatoid arthritis from November 2016 to August 2018, was conducted. The correlations were evaluated using Spearman correlation coefficient and concordance with Bland-Altman plots, quadratic weighted kappa, and discordance rates in the crosstab. For other indices, the optimal cutoff points corresponding to SDAI remission were explored through receiver operating characteristic curve analysis. Results: A total of 30,501 patients were included, of whom 80.46% were women. Most individuals were with moderate disease activity or high disease activity. High correlations among DAS28-erythrocyte sedimentation rate (ESR) and DAS28-C-reactive protein (CRP), SDAI and CDAI were observed. Similarly, the weighted kappa value among the indices was high. In Bland-Altman plots, a positive difference between DAS28-ESR and DAS28-CRP was observed, with an absolute difference of >1.2 in 3079 (10.09%) patients. In crosstab, approximately 30% of the patients were classified into different groups. Concordance values between SDAI remission and the optimal cutoff points of DAS28-ESR, DAS28-CRP, and CDAI were 3.06, 2.37, and 3.20, respectively. Conclusions: Although DAIs had high correlations and weighted kappa values, the discordance between DAIs was significant in Chinese patients with RA. The four DAIs are not interchangeable.

Automated summary

The study on Chinese patients with RA showed high correlations and concordance among different DAIs, but significant discordance was also observed. The four DAIs are not interchangeable in clinical practice for assessing disease activity in RA patients.