Maasstad Hopsital & Erasmus University Rotterdam, Netherlands
Disclosure: Disclosure information not submitted.
Marijke van den Dikkenberg1, Martijn Kuijper2, Marc Kok1, Deirisa Lopes Barreto2 and Angelique Weel1, 1Department of Rheumatology and Clinical immunology, Maasstad Hospital, Rotterdam, Netherlands, 2Department of Rheumatology and Clinical immunology, Maasstad Hospital, Rotterdam
Background/Purpose: In modern Rheumatoid Arthritis (RA) care remote consults (via phone or video call) are increasing. Whereas the Disease Activity Scores (DAS) requires a patient visit, Patient Reported Outcome Measures (PROM) like the Rheumatoid Arthritis Impact of Disease questionnaire (RAID), might serve as a remote triage tool for type of consult. However, for this, the discriminative ability of RAID in relation to DAS needs to be evaluated.
Methods: In this cross-sectional analysis, data of chronic RA patients (diagnosed by a rheumatologist, >6 months) of a large trainee hospital in Rotterdam, the Netherlands were used. Patients completed the RAID within a time-window of up to 7 days before the DAS28CRP in the period from December 2017 to December 2021. Patients were divided based on their DAS28CRP scores (lowDAS < 3.2≥highDAS), subsequently an ROC-curve and AUC were obtained and cut-off scores for the RAID were determined in two ways; maximum Youden-index and absence of false negatives (FN).
Results: Divided over lowDAS (76.8%) and highDAS (23.2%), n=328 patients had respectively a median (p25-p75) RAID score of 2.88 (1.16-5.00) and 5.93 (4.90-7.04). With an AUC=0.815 (95%CI=0.767-0.864), the maximum Youden-index (0.508) reflected RAID=4.79 and FN=4.9%. FN=0% was obtained at a maximum score of RAID=0.97. The cut-off score of RAID=2 (from literature) resulted in FN=27.7%.
Conclusion: In this cross-sectional study the RAID performed well in relation to the DAS. Based on the assumption that lowDAS patients are qualified for remote consults and taking into account the %FN, we advise to offer remote consults to patients scoring RAID < 4.79. Receiver operating characteristics (ROC) curve Disclosures: M. van den Dikkenberg, None; M. Kuijper, None; M. Kok, None; D. Lopes Barreto, None; A. Weel, None.
Receiver operating characteristics (ROC) curve