Background The Bath Ankylosing Spondylitis (AS) Disease Activity Index (BASDAI) –an index of only patient-self-report measures - has been the most widely used measure in axial spondyloarthritis (SpA). More recently the AS Disease Activity Score (ASDAS) has been developed as an algorithm that combines BASDAI elements and patient global assessment with laboratory measures. Both indices are specifically designed for AS patients. In busy clinical settings, it is not feasible to ask different patients to complete different questionnaires, but simple for all patients to complete the same questionnaire, such as a multidimensional health assessment questionnaire (MDHAQ), with a Routine Assessment of Patient Index Data (RAPID3), which has proved useful in most rheumatic diseases.
Objectives To evaluate different aspects of validity of a modified version for RAPID3, a generic index including only patient reported outcomes, compared to AS specific measures.
Methods A prospective cohort of patients with inflammatory back pain suggestive of AS, Devenir des Spondylarthropathies Indifférenciées Récentes (DESIR) was studied. 461 patients who fulfilled Assessment of SpondyloArthritis international Society classification (ASAS) criteria for axial AS were included in the analysis. A modified version for RAPID3 including HAS for AS, the mean of back pain and peripheral pain and patient global was calculated. Different aspect of validity of this modified version for RAPID3 were tested including feasibility (ad hoc ranging), internal consistency (Cronbach's alpha), construct validity (Pearson's r coefficients between disease activity measures), discrimination between low/high disease activity according to different definitions (by standardised mean difference and agreement by weighted kappas), and responsiveness (ROC curves).
Results Construct validity is seen for the 3 indices, with significant correlations (Pearson's r range 0.73-0.84) (Table). Criterion validity versus patient global estimates ranged from 0.70-0.90 and from 0.61-0.62 versus physician global estimate, similar for the three indices. The percentage of patients with low to active disease ranged from 9 to 31% and with high activity from 18% to 56%, according to various measures. The strength of agreement was good for the modified RAPID3 and moderate for BASDAI and ASDAS-CRP versus patient global estimate; and moderate for ASDAS-CRP and fair for BASDAI and modified RAPID3 versus physician global estimate as reference. The strength of agreement of the modified RAPID3 with BASDAI and ASDAS-CRP was similar. Responsiveness was similar for all 3 indices. the modified RAPID3 appears the most feasible index in terms of time to complete, clarity, simplicity and acceptance (fewer missing data).
Conclusions This modified version for RAPID3 provide similar information to BASDAI and ASDAS-CRP. A generic measure for all rheumatic diseases may provide a feasible approach to quantitative assessment of AS patients in busy clinical settings.
Disclosure of Interest None declared