Patient Reported Outcome Measures for Rheumatoid Arthritis Disease Activity: Rasch measurement theory to identify items and domains

Objectives Disease activity (DA) monitoring is a standard of care in RA. There is demand for achieving this through patient-reported outcome measures (PROMs). The aim of this study was to determine which items could be used to measure the construct of RA DA, by analysing legacy PROMs, using Rasch measurement theory (RMT) analyses. Methods Questionnaires including 10 legacy PROMs were sent to people with RA to create original and validation datasets. Items were grouped according to OMERACT domains and analysed using principal component analysis. Based on separate domain RMT analyses of the original dataset, domain-level testlets were assessed to determine which items measure the construct of RA DA. The result was then replicated in confirmatory factor analyses bifactor models and RMT analyses of the validation dataset. Psychometric properties of legacy PROMs were also assessed in the original dataset. Results The total sample size was 691 (original: 398, validation: 293). The Patient global domain was split into General health and Disease activity domains under RMT. General health and Fatigue domain items measure a separate construct to the construct of RA DA. A set of 12 Pain, Disease activity, Tenderness and swelling, Physical functioning and Stiffness domain items can be used to measure the construct of RA DA. No legacy PROMs fully fit the Rasch measurement model. Conclusion General health and Disease activity domain items are not interchangeable. Twelve items form an item pool that can be used to measure the construct of RA DA. Legacy PROMs should not be recommended for use.