Patient-reported Outcome Measures for Rheumatoid Arthritis Symptom Severity: development of a computer adaptive test from an item bank using Rasch measurement theory (SOCRATES)

End of project summary

Main Messages

Rheumatoid Arthritis (RA) is a common chronic inflammatory arthritis characterised by fluctuating symptom severity and recurrent flares, requiring ongoing monitoring and subsequent treatment adjustments over time. Disease activity (DA) monitoring is a standard of care in RA. The current DA assessments require laboratory tests and/or health care professional (HCP) input. Patient Reported Outcome Measures (PROMs), which are tools completed by patients to ascertain perceptions of their health, may therefore be preferable. However, there is no consensus on how to measure RA DA using a PROM.

The aim of this study was to assess the existing RA DA PROMs and other relevant PROMs, and begin developing a new RA DA PROM.

Multiple methods were used, which included:

1. a review of literature following up-to-date international guidelines;
2. analyses of data collected through questionnaires sent to people with RA (pwRA) across four South Wales University Health Boards;
3. analyses of discussions held with pwRA, and;
4. the development of an online tool that decides on question order.

The results showed that:

• following the up-to-date international guidelines, no existing RA DA PROMs can be recommended for future use;
• no existing RA DA PROMs, or other relevant PROMs, are fully valid, which means there is no evidence that they measure RA DA correctly;
• the Patient global domain questions fall into two distinct domains of Disease activity questions and General health questions, and these cannot be used interchangeably;
• initially, 12 questions from the domains of Pain, Disease activity, Tenderness and swelling, Physical functioning and Stiffness can be used to measure RA DA; • questions from the domains of General health and Fatigue cannot be used to measure RA DA. This is important because many RA DA outcomes include General health domain questions;
• through discussions with pwRA, there were no questions, or concepts, missing that should be covered;
• there were no concerns with the majority of questions. For the minority of questions that pwRA did have concern with, these could still be completed;
• these concerns were variability of response relating to the time of day and teasing out RA symptoms from those caused by other conditions;
• that an online tool that decides on question order does not provide a major advantage for the purpose of asking the 12 questions;
• finally, five questions, with one each the domains of Pain, Disease activity, Tenderness and swelling, Physical functioning and Stiffness, can be used to measure RA DA.

The next steps are to design these questions with pwRA and then test their ability to measure RA DA. After that, the plan is to build these questions into a weekly DA monitoring tool, where results would be provided to and facilitate discussion with HCPs to make better decisions about the best treatments for pwRA. This should allow pwRA to be as healthy as possible as soon as possible.