DAS28, CDAI and SDAI cut-offs do not translate the same information: results from the Rheumatic Diseases Portuguese Register Reuma.pt

Objectives. The 28-joint DAS (DAS28), clinical disease activity index (CDAI) and simplified disease activity index (SDAI) are indices frequently used to assess disease activity in RA patients. Cut-off values were defined to classify the states of RA disease activity: remission, low, moderate and high. The aim of this work was to assess disease activity states classified by DAS28, CDAI and SDAI and to analyse their agreement in the Rheumatic Diseases Portuguese Register Reuma.pt. Methods. A total of 2795 patients and 14 440 visits were selected from Reuma.pt for analysis. Pearson's correlation coefficients (PCCs) were calculated for the three indices. McNemar's chi-squared tests, PCCs and kappa statistics were performed to analyse and compare the distribution of visits among all disease activity states and indices. Results. A strong correlation was found between the three indices throughout the 14 440 visits: r = 0.874 for DAS28/CDAI, r = 0.877 for DAS28/SDAI and r = 0.984 for CDAI/SDAI (all PCCs with P < 0.0001). However, when categorization in the different disease activity states was analysed, McNemar's chisquared tests and PCCs revealed significant disagreement between the cut-offs of the three indices. Conclusion. DAS28, CDAI and SDAI cut-offs do not translate into the same clinical information in Reuma.pt. Although this might be expected for the original DAS28 cut-offs, when compared with CDAI and SDAI significant disagreement was also found for the DAS28 modified cut-offs. For visits where patients are in CDAI or SDAI remission, we also find disagreement between these two indices, which may contradict previous conclusions that acute phase reactants add little to composite disease activity indices for RA.