Background No single “gold standard” measure is available to assess patients with rheumatoid arthritis (RA) in clinical trials and routine care, as in hypertension, diabetes, and other diseases. Therefore, an index of several measures, such as a DAS28 (Disease Activity Score-28) and CDAI (Clinical Disease Activity Index), based on 7 RA core data set measures; is needed. However, the only quantitative data in many (most) patients in routine rheumatology care are laboratory test results. RAPID3 (routine assessment of patient index data), which includes only patient self-report scores, is considerably more feasible than DAS28 or CDAI for routine care, distinguishes active from control treatments in RA clinical trials similarly and is correlated significantly with these indices. A minimal clinically important improvement (MCII) to interpret changes in clinical trials and clinical care has not been established for RAPID3
Objectives To estimate MCII of RAPID3, and compare results to MCIIs of DAS28 and CDAI.
Methods Post hoc analyses were performed of a reported longitudinal study of 250 patients with active RA (1). All 7 RA core data set measures were collected at baseline and after treatment escalation with prednisone 1 month later or with disease modifying medications or biologic agents 4 months later (1). Patient judgment of improvement in arthritis status was obtained as “improved”, “the same” or “worsened”, and analyzed in relation to changes in RAPID3, DAS28 and CDAI. RAPID3 is the sum of 3 0–10 measures: physical function on a HAQ recalculated from 0–3 to 0–10, pain and patient global estimate on 0–10 VAS (visual analog scales), total=0–30. DAS28-ESR (erythrocyte sedimentation rate) and CDAI were computed as described in the literature. Changes in all indices, standardized response means (SRM), MCIIs as changes that had a specificity of 0.80 for improvement based on receiver-operating characteristic curves, and MCII as a proportion of the maximum score were computed.
Results Among 250 patients, 167 (66.8%) reported improvement. RA activity and SRMs improved similarly per the 3 indices (Table). ROC curve areas were ≥0.77 (Table). MCIIs with specificity for improvement of 0.80 were -3.5 for RAPID3, -1.17 for DAS28-ESR, and -12.5 for CDAI. MCIIs were in a similar range of 11.6% to 16.8% of maximum score (Table).
Conclusions MCIIs for RAPID3, DAS28, and CDAI were in a similar range. Knowledge concerning MCII thresholds can improve interpretation of data from clinical trials and routine clinical care.
Ward, M et al, Ann Rheum Dis 2015, 74:1691–1696.
Disclosure of Interest I. Castrejon: None declared, M. Ward: None declared, M. Bergman: None declared, L. Guthrie: None declared, M. Alba: None declared, T. Pincus Shareholder of: Health Report Services, Inc