Background Since the disease activity of patients with RA tends to fluctuate between visits and disease flares are easily missed during regular visits, patients might benefit from a more closely spaced determination of disease activity, which could be realised by iMonitor¹. iMonitor is a Software Medical Device developed and funded by Pfizer. This online system allows patients to complete three kinds of patient-reported outcome measures (PROMs): the Health Assessment Questionnaire (HAQ), Rheumatoid Arthritis Impact of Disease (RAID) and Rheumatoid Arthritis Disease Activity Index-5 (RADAI-5). iMonitor might contribute to identification of patients who need additional medical attention in between visits and reduction of visits for patients with stable disease activity.

Objectives To determine the degree to which the PROM-scores in iMonitor are associated with DAS28. Moreover, PROM preferences and completion rates were studied.

Methods Patients were recruited at Bernhoven (Uden, the Netherlands) by an announcement on the hospital website, leaflets and meetings. Instruction classes were organised in which researchers assisted patients in using iMonitor. Patients indicated which PROM(s) they want to complete in iMonitor and chose reminder email frequency (weekly, two-, four, six, or eight-weekly). Descriptive analyses were used to describe characteristics of the study population. Scatter plots with regression equations were performed with DAS28 as dependent and PROM as independent variable to determine the association between DAS28 and PROMs. Moreover, Pearson's correlations were calculated. PROM-scores within the fourteen day window before and after DAS28 assessment were included.

Results In total 33 patients with RA were included, seventeen of them were female (52%). Mean (±SD) age was 56±11 years. Seventeen patients (52%) attended the instruction classes. Majority of patients (n=10) chose all three PROMs to complete, nine patients chose RAID+RADAI-5, seven chose HAQ+RAID, three chose RAID, two chose RADAI-5 and two chose HAQ+RADAI-5. From March 2016 until December 2016, 435 RAID-values, 329 RADAI-5-values and 222 HAQ-values were gathered. When taking PROM-values within the fourteen day window before and after DAS28 assessment, 159 DAS28-values could be coupled to 320 PROM-values. Regression analysis showed the following proportions of explained variance (R²): 0.17 for HAQ, 0.32 for RAID and 0.29 for RADAI-5. Pearson's correlation coefficients were 0.41 for HAQ, 0.57 for RAID and 0.54 for RADAI-5. Most chosen reminder email frequency was four weeks (n=21). Completion rates (measured until December 31, 2016) were 65% for patients with one week PROM-frequency and for patients with two, four, six and eight week frequency completion rates were 39%, 24%, 30% and 0%, respectively.

Conclusions RAID and RADAI-5 were moderately associated with DAS28 and showed highest proportions of explained variance. The association between HAQ and DAS28 was weaker. Patients receiving a weekly reminder email showed highest completion rates. This pilot study is a first step towards personalised healthcare and patient involvement in online remote monitoring.

References

1. Medical and Educational Goods and Services (MEGS): iMonitor. 2014 Available from: http://www.pfizer.co.uk/content/medical-and-educational-goods-and-services-megs-imonitor.

References

Disclosure of Interest L. Renskers Grant/research support from: Pfizer, S. Rongen: None declared, S. Teerenstra: None declared, M. Hulscher: None declared, P. van Riel Grant/research support from: Pfizer