FRI0657 Construct validity and responsiveness of three stiffness items in rheumatoid arthritis: an application of the omeract filter 2.1. (12th June 2018)

• Record Type: Journal Article
• Authors: Craig, E.T.; Bingham, C.O.; Bartlett, S.J.; Haque, U.; Leong, A.; Orbai, A.-M.
• Is Part Of: Annals of the rheumatic diseases. Volume 77(2018)Supplement 2
• Page Start: 849
• Page End: 849
• Publication Date: 2018-06-12
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2018-eular.2142
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Stiffness is frequent and important to people with rheumatoid arthritis (RA), but measuring it is an unmet need. Qualitative research with people with RA has shown that stiffness duration, severity and interference have face and content validity. These items are being evaluated to develop a patient reported outcome measure (PROM) for stiffness. Objectives: To evaluate individual stiffness item measurement properties using the OMERACT Filter 2.1. Methods: Consecutive patients in a longitudinal RA cohort were enrolled. Participants completed 3 items assessing stiffness severity, interference, and duration. PROMs collected included patient global assessment (PGA), pain VAS, self-reported flare (Y/N), PROMIS measures, and patient global impression of change (PGIC). Physician assessments included 28 tender/swollen joint counts, and global (EGA). Disease activity indices were CDAI and DAS28-CRP. Stiffness construct validity was assessed using Spearman correlation coefficients. To assess discrimination in groups expected to change vs stable groups, we calculated standardized response means (SRMs) using multiple anchors. Results: 196 patients with RA were included. Construct validity: all stiffness items correlated highly with pain; moderately with DAS-28 CRP and CDAI; and low with joint counts and EGA. Stiffness severity and interference correlated highly with PGA, pain interference and fatigue. Stiffness interference correlated highly with physical function. Discrimination : all stiffness items had moderate effect size for resolution of flare, and for 'much improved' by PGIC. Interference had moderate effect for improved CDAI category and for 'little improved' by PGIC. Stiffness duration had large effect for new RA flare, and 'much worse' by PGIC; severity and interference had moderate effect for RA flare; severity had moderate effect for worsened CDAI and for 'little worse' and 'much worse' by PGIC (Tbl). Conclusions: In this cohort of RA patients there is emerging evidence for good construct validity for stiffness duration, severity and interference; moderate discrimination for improvement in RA disease state, and moderate to high discrimination for worsening RA. Additional items are being tested to assemble an instrument for measuring stiffness in RA. Disclosure of Interest: None declared