Objective: To examine the test-retest reliability of the rheumatoid arthritis (RA) core disease activity measures and derived composite indices.
Methods: Twenty-eight stable RA patients had two complete assessments within one week, which included the seven RA core disease activity measures and derived disease activity indices (DAS28, SDAI, CDAI, RAPID3, RADAI). The intraclass correlations (ICC), the smallest detectable difference (SDD), and minimal detectable change as percentage of the maximum score (MDC%) were estimated as measures of test-retest reliability.
Results: Correlations for the disease activity indices were high. SDDs (MDC%) to detect a true improvement or deterioration with 95% confidence were for DAS28 1.32 (14.4%), SDAI 8.26 (9.6%), CDAI 8.05 (10.6%), RAPID3 1.48 (14.8%), and for RADAI 1.49 (14.9%).Thus, SDDs were rather high, and the MDC% were of similar magnitude of 10-15% as for all 7 core data set measures.
Conclusions: SDDs of the DAS28, SDAI and CDAI were close to limits to detect important improvement. Clinicians should be aware of measurement error. Nonetheless, RA core data set measures and indices obtained from a health professional, laboratory, and patient self-report had similar reliability.