Background/Purpose: As a tool for self-monitoring and reporting RA disease activity, a smartphone app was developed to enable people with rheumatoid arthritis (RA) to record their disease activity, including self-reported 28 tender (28TJC) and swollen joint counts (28SJC).  Studies suggest self-joint counts by people with RA are relatively similar to those by rheumatologists (28TJC more similar than 28SJC) but little is known about the effects of training on self-joint count accuracy.  In collaboration with people with RA we refined techniques for joint self-examination and produced an instructional video.  The aims of this study were to determine how patient self-joint counts, and derived disease activity scores, compared to rheumatologist joint counts and whether this was affected by a self-training video.

Methods: Patients (n=100) meeting EULAR/ACR 2010 criteria for RA attending clinics in 3 centres self-reported RA disease activity measures using the app, including a self-examination joint count.  Patients were then randomised to either complete the self-report joint count a second time (Pt 2) with no further instruction (n=49) or to complete the self-report after watching the video (n=51).  Blinded to patient data, a rheumatologist completed a 28-joint count, which was taken as the “gold standard”. 

Results: Most participants were female (77%), had mean age 60.2 years (range 33-83 years), duration of RA 17.0 years (range 0.25-55 years) and 45% were using a biologic DMARD.  Video and no-video group participants did not differ on demographic or disease characteristics but the video group had lower smartphone ownership (68% versus 85%, p=0.042).  First and second patient-reported joint counts showed high correlation and did not change after watching the video, suggesting that watching the video had no training effect (Pearson R for 28TJC Rheum vs Pt 2 No video 0.568, Video 0.508; 28SJC Rheum vs Pt 2 No video 0.281, Video 0.386).  Compared to rheumatologist assessment, people with RA consistently reported higher 28TJC and 28 SJC (28TJC Pt 2 vs rheum mean diff 3.01 (SD 5.76), 28SJC Pt 2 vs rheum mean diff 1.49 (SD 5.49)) but these differences were not statistically significant. Patient-derived DAS28 was never more than one disease activity category higher than rheumatologist-derived DAS28.

Conclusion: People with RA are consistent in reporting of self-joint counts and a training video did not change reporting.  Although patients scored their joint counts higher than rheumatologists, the difference was small and not significant and would not invalidate the use of patient-reported scores for clinical purposes.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/self-joint-counts-by-people-with-rheumatoid-arthritis-does-a-video-increase-accuracy-and-does-it-matter/