Remote Care in People With Rheumatoid Arthritis (NOR-Flare)

Can Machine Learning and Patient-reported Outcomes be Used in Remote Care in People With Rheumatic Diseases

This study is a 24-months, non-inferiority randomized, controlled trial with two parallel arms to determine if a new follow-up strategy for patients with RA is non-inferior in maintaining comprehensive disease control measured as simultaneous maintenance of structural, functional and clinical treatment target at 2-year follow-up compared to the conventional follow-up regimen with regular hospital visits.

The study will include Norwegian adult males and females with rheumatoid arthritis. Eligible patients that consent to participation will be randomized to two groups: Control group: conventional follow-up strategy with blood tests, patient-reported outcomes (PROs), and pre-scheduled visits at the hospital every 6th month. Remote monitoring: monthly remote monitoring of PROs and triage of patients using an algorithm will guide healthcare providers in scheduling patients for a video consultation or face-to-face hospital visits. Participants will be followed for 24 months. Primary outcome is proportionn maintaining comprehensive disease control measured as simultaneous maintenance of structural, functional and clinical treatment target at 2-year follow-up1. Structural: Assessed with radiographs of hands and feet according to the van der Heijde modified Sharpe score (subscores for erosions (0-280) and joint space narrowing (0-168)), with a total range of 0-448. Maintenance of structural treatment target is defined as change in total score <1 unit/year (<2 units from inclusion to 2-year follow-up). Functional: Measured by Modified Health Assessment Questionnaire (MHAQ) measured on a scale from 0.00 to 3.00, where a change of 0.25 is considered clinical important3. Maintenance of functional treatment target is defined as a worsening <0.25 from inclusion to 2-year follow-up. Clinical: Measured by DAS28, categorized into remission (<2.6), low disease activity (2.6 to ≤3.2), moderate disease activity (3.2 to ≤5.1) and high disease activity (>5.1). Maintenance of clinical treatment target is defined disease activity category at 2-year follow-up ≤ baseline category. We will use a 15% non-inferiority margin. The study will comprise an internal pilot study the first 6 months for all participants in the intervention group. The study will also include qualitative research including semi-structured interviews and observations of patients in the intervention group and health professionals involved in the study. The interviews will explore experiences with remote monitoring and video consultations.

Conventional follow-up strategy with blood tests and face-to-face visits at the hospital every 6 months

Monthly remote monitoring of patient-reported outcomes and triage of patients using an algorithm will guide healthcare providers in scheduling patients for a video consultation or face-to-face hospital visits.

Patients in the conventional/usual care arm will be treated according to current conventional follow-up regimen with PROs, blood tests, and face-to-face visits with an experienced nurse or a rheumatologist every 6 months. They can contact the hospital if they are experience worsening of their disease.

The patients in the remote monitoring arm will use a web-app (Youwell) for self-reporting patient reported outcomes (PROs) and CRP/ESR, displaying results for PROs over time, and for synchronous (video) or asynchronous (chat) communication with healthcare providers. The patients will receive a SMS reminder for "tasks" (e.g., self-reporting PROs or registering results from blood tests) every month. A study coordinator/nurse will monitor the PROs and blood test (CRP/ESR) results, and respond to chat messages. Based on the algorithm, a triaging functionality in the Youwell platform will aid healthcare providers in highlighting which patients needs attention.

Comprehensive disease control measured as simultaneous maintenance of structural, functional and clinical treatment target at 2-year follow-up. Structural: Radiographs of hands and feet according to the van der Heijde modified Sharpe score, with a total range of 0-448. Maintenance of structural treatment target is defined as change in total score <1 unit/year (<2 units from inclusion to 2-year follow-up). Functional: Modified Health Assessment Questionnaire (MHAQ) measured on a scale from 0.00 to 3.00, where a change of 0.25 is considered clinical important. Maintenance of functional treatment target is defined as a worsening <0.25 from inclusion to 2-year follow-up. Clinical: DAS28, categorized into remission (<2.6), low disease activity (2.6 to ≤3.2), moderate disease activity (3.2 to ≤5.1) and high disease activity (>5.1). Maintenance of clinical treatment target is defined disease activity category at 2-year follow-up ≤ baseline category.

Joint damage progression from inclusion to 2-year follow-up assessed with radiographs of hands and feet according to the van der Heijde modified Sharpe score (subscores for erosions (0-280) and joint space narrowing (0-168)), with a total range of 0-448. Measured as a continuous variable.

Measured with patient global assessment of disease activity (0-10 scale, 0=no disease activity), estimated as change across all timepoints

Measured with patient global assessment of disease activity (0-10 scale, 0=no disease activity), estimated as change across all timepoints

Measured with EQ5D-5L, 5 questions used to calculate an utility score (0-1, 1= best health), assessed as a total score across all timepoints

Measured with CDAI (patient global assessment of disease activity, number of tender and swollen joints, physician assessment of disease activity). Remission/low disease activity defined as CDAI <10.

Measured with DAS28 (patient global assessment of disease activity, CRP/ESR, number of tender and swollen joints). Remission/low disease activity defined as DAS28 <3.2.

Measured with DAS28 (patient global assessment of disease activity, CRP/ESR, number of tender and swollen joints). DAS28 score: <2.6= remission; 2.6-<3.2=low disease activity; 3.2 -5.1= moderate disease activity; >5.2= high disease activity

Measured with CDAI (patient global assessment of disease activity, number of tender and swollen joints, physician assessment of disease activity). CDAI score: <=2.8 = remission; >2.8 - <=10 = low disease activity; >10 - <=22 = moderate disease activity; >22 = high disease activity

Costs related to self-reported healthcare use in primary and secondary health care. Patients will be asked if they have been seeking healthcare (yes/no), if yes, they will be asked to specify type of healthcare use and time used. This information will also be collected from national register data.

Patient-reported experience of a significant worsening of symptoms (reflecting a flare in disease activity), response options: no, yes, uncertain. If yes or uncertain, they will be asked which date the flare occurred and the number of days it lasted.

Patient-reported experience of a significant worsening of symptoms (reflecting a flare in disease activity), response options: no, yes, uncertain. If yes or uncertain, they will be asked which date the flare occurred and the number of days it lasted.

Item no.6 from Work Productivity and Activity Impairment (WPAI): self-reported activity impairment on a NRS 0-10; higher value indicate worse outcome.

Item no.6 from Work Productivity and Activity Impairment (WPAI): self-reported activity impairment on a NRS 0-10; higher value indicate worse outcome.

Self-reported travel distance (km and time) and way of transport (walking, bicycle, privat car, public transportation, taxi, airplane, other) in conjunction with consultation at the hospital

1 item from Pittsburgh Sleep Quality Index, self-reported sleep impairment in last month due to pain with 4 response categories ranging "Not during the past month" to "Three or more times a week" ; higher value indicate worse outcome.

Physician count of swollen joints; MCP 1-5, PIP 1-5, wrist, elbows, shoulders, ankles, MTP 1-5, at regular visits, extra visits, withdrawls, early discontinuation

Physician count of tender joints; MCP 1-5, PIP 1-5, wrist, elbows, shoulders, ankles, MTP 1-5, at regular visits, extra visits, withdrawls, early discontinuation

one question assessing the patient acceptable symptom state last week, 5 response options from very good to very bad.

eHEALS questionnaire, 9 items measured on a 5 point Likert scale, with higher score indicating better health literacy

Patient-reported self-efficacy for using different digital devices, secure login and digital health services

Self-efficacy/confidence in using smartphone, tablet, computer, app's, secure login and digital health services. 6 items with Likert scale response categories: Never used, Very bad, Bad, Neither good nor bad, Good, Very good. Score range 1-5; higher scores indicate higher self-efficacy.

1 item with five point response options ranging from "Very satisfied" to "Very dissatisfied", higher value indicate better outcome

Telehealth usability questionnaire (TUQ) 21 items scored from 1-7, higher score indicate higher satisfaction with telehealth

Inclusion Criteria: Male or non-pregnant, non-nursing female ≥18 years of age at screening Patients with a diagnosis of RA who fulfil the 2010 ACR/EULAR diagnostic criteria24 (see Appendix 5, 10.4) Medical treatment with cs/ts/bDMARDs (incl. prednisolone) considered stable by the healthcare provider the last 6 months Low disease activity or remission (CDAI<10 / DAS28<3.2) at inclusion <2 swollen joints Not deemed inappropriate for remote monitoring by the healthcare provider Capable of understanding and signing an informed consent form Access to a smartphone or tablet Able to speak and understand Norwegian language Exclusion Criteria: Medical conditions: Major co-morbidities, such as severe malignancies, severe diabetes mellitus, severe infections, uncontrollable hypertension, severe cardiovascular disease (NYHA class III or IV), severe respiratory diseases, and/or cirrhosis. Indications of active tuberculosis (TB) Treated with intravenous DMARD (e.g., rituximab and infliximab) Diagnostic assessments: Abnormal renal function, defined as serum creatinine >142 µmol/L in female and >168 µmol/L in male, or glomerular filtration rate (GFR) <40 mL/min/1.73 m2 Abnormal liver function (defined as Aspartate Transaminate (AST)/Alanine Transaminase (ALT) >3 x upper normal limit), active or recent hepatitis Leukopenia and/or thrombocytopenia Other: Pregnancy and/or breastfeeding (current at screening or planned within the duration of the study) Severe psychiatric or mental disorders, alcohol abuse or other substance abuse, language barriers or other factors which makes adherence to the study protocol impossible. Deemed unsuitable for remote monitoring by medical doctor

Data sharing of anonymous data may be possible upon reasonable request sent to the principal investigator.