Background: Remote management of rheumatoid arthritis (RA) using patient self-assessment of disease and patient-reported outcomes has potential to inform timely clinical decisions on disease management, reduce burden on busy rheumatology services and promote effective self-management. However, it is unclear how patients’ self-reporting relates to treatment decisions.

Objectives: To determine the agreement between remote treatment decisions based on patient self-assessment questionnaire assessed blindly by a health professional and treatment decisions based on routine outpatient monitoring appointments.

Methods: This was a prospective real-world study including patients (who gave informed consent) enrolled in an observational study of RA patients starting a new biologic therapy.

Enrolled patients continued with their usual care and usual clinic monitoring. In addition, they completed at home, self-assessment questionnaires at monthly intervals, including: two self-reported components which are collected as part of routine clinical practice (joint stiffness and flare), visual analogue scales for pain, global health and fatigue; and HAQ-DI and self-efficacy scales (Arthritis Self-Efficacy Pain and Other Symptoms subscales) at inclusion visits, 12 and 24 months.

Remote treatment decisions were made by an independent (blinded) health professional, based on the self-assessment questionnaires and information collected in the study: medical history, ongoing therapies for RA, clinical outcomes, adverse events and toxicity. In this analysis, the independent blinded clinician did not have the same information as the routine hospital visit clinician (blood results and joints assessment).

The remote decisions were matched with the hospital visit decisions (within 2 weeks) and the measure of agreement between the 2 raters (independent blinded health professional and clinician at outpatient appointment) were evaluated using kappa coefficient: <0.2, 0.21-0.40, 0.41-0.60, 0.61-0.80 and 0.81-0.99 representing poor, fair, moderate, substantial and almost perfect agreement respectively.

Results: A total of 72 RA patients were recruited into the sub-study: mean (SD) age 57.8 (11.6), disease duration 11.7 (10.3) and 52 (87%) were female. The pre-categorised remote decisions were: no change to biologic, stop biologic, add concomitant DMARD, reduction/removal of a concomitant DMARD and bring in for review. There were 57 matched decisions between the independent health professional and the outpatient clinicians. The outpatient clinician made 7 changes to biologic and 18 non-biologic therapy changes, while the remote health professional made 1 change to biologic and 17 changes to a non-biologic DMARD including bringing in for review. The self-assessment questionnaires reported 34 RA flares of which 21 had resolved. In the matched decisions, there was only one adverse event that needed stopping treatment, identified by both the remote and the outpatient treatment. The independent health professional and the outpatient clinician had a ‘fair’ agreement on changes to biologic therapy (Kappa = 0.226, p = 0.007) and overall changes to RA therapy (Kappa = 0.24, p = 0.07).

Conclusion: Remote RA monitoring using patient self-assessment and outcome measures was feasible with fair agreement on treatment decisions. Further work is required on understanding the importance of adding blood test monitoring to remote decision-making.

Disclosure of Interests: None declared