Skip to main content


Presenting at the American College of Rheumatology Convergence 2023 in San Diego

7 November 2023

Presenting at the American College of Rheumatology Convergence 2023, San Diego

I am presenting a poster titled ‘Patient Reported Outcome Measures for Rheumatoid Arthritis Disease Activity: using Rasch measurement theory and cognitive interviewing to achieve more meaningful measurement’ at the American College of Rheumatology Convergence 2023 in San Diego on Monday 13th November in a session named ‘RA – Diagnosis, Manifestations, and Outcomes Poster II’



This poster summarises the quantitative and qualitative research I have done on Patient Reported Outcome Measures (PROMs) for Rheumatoid Arthritis (RA) Disease Activity (DA).

There is a nice similarity in these two aspects, as this was first use of both qualitative research for content validity and RMT analyses in the field of RA DA PROMs. Completing work on content validity also brings this work full circle from the outcome of the systematic review that was the first part of my PhD, as this indicated a lack of content validity in this field.

The next step will be to develop a computer adaptive test or electronic PROM based on these results. A first step will be to develop new items within the relevant domains with patients so that instructions and response options are relevant to those who will be answering them. The hope is to then assess the psychometric properties of a new PROM that uses these items and then build this PROM into an intervention with weekly monitoring of RA DA fed back to healthcare professionals involved in the patients’ care to make informed decisions on treatment and management.


The first table here gives the comprehensive list of PROMs and groups of items that were included in the questionnaire and were therefore available for analysis. It is important to be transparent so that researchers know which PROMs I have analysed, especially as I have made use of items from these PROMs to find those that can be used to measure the construct of RA DA.

In the first figure, I show which items from which domains were taken forward on the basis of multiple exploratory factor analyses. This shows that 30 items from six domains – tenderness and swelling, pain, patient global, fatigue physical functioning and stiffness   were carried forward. Items from the pain (area-specific), physical functioning (specific), discomfort/fear, mood and swelling domains were not carried forward.

After showing that patient global must be split into two domains of disease activity and general health, the second figure provides justification for stating that general health and fatigue measure a separate construct to RA DA. Here we see the item characteristic curves for the testlets created by adding up the items in the domains. The observed dots are far from the line predicted by RMT and thus have large fit residuals, indicating a departure from construct of RA DA.

The final table provides exemplar quotes relating to the four themes that were evident in the think aloud section of the cognitive interviews. There is variability in symptoms like pain, tenderness and swelling and stiffness that isn’t always easy to summarise into a single answer. Some item instructions and response options were not straightforward, with one participant unable to answer a fatigue item that stated ‘due to your Rheumatoid Arthritis’. This linked onto the final category of teasing out RA symptoms. This participant was unable to distinguish between fatigue caused by her RA and that caused by her breast cancer.

Aspects of the poster


Following on from a systematic review of these PROMs following COSMIN guidelines, I analysed data from these PROMs using Rasch measurement theory (RMT), firstly to assess their psychometric properties and then to see if any combination of items could be used to measure the construct of RA DA. I discovered that items in the tenderness and swelling, pain, disease activity, physical functioning and stiffness domains could be used to measure this construct, whilst items in the general health and fatigue domains measured some other construct.


I then conducted cognitive interviews with 20 participants to better understand the comprehensiveness, comprehensibility and relevance of the items that could be used to measure the construct of RA DA. Participants provided opinions that confirmed the results of the quantitative data analyses, this providing initial evidence of content validity, and also made suggestions of further questions they would be happy to answer alongside them.


I have been very lucky to have presented at multiple conferences over the course of my PhD, including the British Society of Rheumatology (BSR) conference, the International Society of Quality of Life Research (ISOQOL) conference, the National PROMs Research Conference and the Health and Care Research Wales conference. Across these conferences, I’ve presented the plan for my PhD or a single component of it, whether it was the systematic review, the results of the quantitative analyses or the results of the qualitative analyses. For this final presentation of my PhD, and at the largest conference I’ve ever attended, I wanted to bring together both the quantitative and qualitative analyses and summarise how the results complement each other.

Links to other blogs






Abstract handout for ACR Convergence

-Blog by Tim Pickles

The Centre for Trials Research is funded through Welsh government by Health and Care Research Wales.