Objective: The Disease Activity Score including 28 joints (DAS28) and the Clinical Disease Activity Index (CDAI) were developed in order to provide a quantifiable measure of rheumatoid arthritis (RA) activity. The aim of this study was to evaluate the validity and internal consistency reliability for DAS28 and CDAI in patients with RA seen by rheumatologists in usual clinical care. We also compared proposed categories of high, moderate, and low activity and remission according to both scores.
Patients and methods: A sample of 2864 RA patients (2267 female, 597 male; mean age 58.5 yr, range 18-88 yr) were enrolled in this cross-sectional community-based study. Disease activity was assessed in each patient based on DAS28 and CDAI. Patients completed the Health Assessment Questionnaire (HAQ). Statistical evaluation was carried out by applying the Cronbach's values and principal component analysis (internal consistency reliability), the Pearson's coefficients, ANOVA and kappa statistic (convergent validity) and receiver operating characteristic (ROC) curve analysis (discriminant validity).
Results: Internal consistency testing of both scores indicated a reasonable difference, with Cronbach's alpha slightly higher for the DAS28. Interestingly, factor analysis revealed that the DAS28 constitutes a monocomponent measure in RA. Linear regression analysis showed a significant correlation between DAS28 and CDAI (p<0.0001). In addition, the DAS28 and CDAI were well correlated with HAQ (both at p level of <0.0001). The discriminatory power of both indices was good, without significant difference, but our results showed wide differences in both moderate/high disease activity and remission percentages (k=0.418).
Conclusions: DAS28 and CDAI are valid and simple acceptable ways to measure RA activity in the clinical practice, but disease activity categorized by these indices differ considerably. Further research is needed to resolve this issue.