Patients with rheumatoid arthritis acquire sustainable skills for home monitoring: a prospective dual-country cohort study (ELECTOR clinical trial I).

Objective In an eHealth setting, to investigate intra- and inter-rater reliability and agreement of joint assessments and disease activity score (DAS28CRP) in patients with rheumatoid arthritis (RA) and test the effect of repeated joint assessment training. Methods Patients with DAS28CRP ≤5.1 were included in a prospective cohort study (clinicaltrials.gov: NCT02317939). Intra-rater reliability and agreement of patient-performed joint counts were assessed through completion of five joint assessments over a two-month period. All patients received training on joint assessment at baseline, only half of the patients received repeated training. A subset of patients was included in an appraisal of inter-rater reliability and agreement comparing joint assessments completed by patients, healthcare professionals (HCPs), and ultrasonography. Cohen’s Kappa coefficients and intraclass correlation coefficients (ICC) were used for quantifying of reliability of joint assessments and DAS28CRP. Agreement was assessed using Bland-Altman plots. Results Intra-rater reliability was excellent with ICC of 0.87 (95%CI: 0.83 to 0.90) and minimal detectable change of 1.13. ICCs for inter-rater reliability ranged between 0.69 and 0.90 (good to excellent). Patients tended to rate DAS28CRP slightly higher than HCPs. In patient receiving repeated training, a mean difference in DAS28CRP of -0.08 was observed (limits of agreements of -1.06 and 0.90). After 2 months, reliability between patients and HCPs was similar between groups receiving single or repeated training. Conclusion Patient-performed assessments of joints and DAS28CRP in an eHealth home-monitoring solution were reliable and comparable with HCPs. Patients can acquire the necessary skills to conduct a correct joint assessment after initial and thorough training.