CAN A SINGLE QUESTION ON FUNCTIONAL IMPAIRMENTS FACILITATE THE IDENTIFICATION OF EARLY INFLAMMATORY ARTHRITIS? A LARGE CROSS-SECTIONAL DERIVATION AND VALIDATION STUDY

Background: Early treatment initiation in inflammatory arthritis (IA), and specifically in RA, is associated with improved outcomes but requires early identification of synovitis. However, healthcare professionals other than rheumatologists, e.g. general practitioners (GPs), experience difficulties in detecting early IA by joint examination. In this light, two Early Arthritis Recognition Clinics (EARCs) were initiated in the Netherlands in 2010. EARCs are easy-access outpatient rheumatology clinics, intermediary between primary and secondary care, to which GPs can refer patients when in doubt about the presence of IA (instead of ‘watchful waiting’). Although this approach markedly reduced referral delay,[1] it may not be easily implemented in other places due to shortage of rheumatologists. Therefore, other new tools to support early identification of IA are needed. Objectives: Aiming for simple and time-efficient use in daily practice, we evaluated if just a single question on functional impairments could aid the identification of early IA. Methods: Data from two EARCs in the Netherlands were studied. Between 2010 and 2014, 997 patients with suspected IA visited the Leiden EARC (derivation cohort). Patients visiting the Groningen EARC (2010–2014, n = 506) and Leiden EARC (2015–2018, n = 557) served as validation cohorts. Physical functioning was assessed with the Health Assessment Questionnaire Disability Index (HAQ); IA by joint examination by rheumatologists. Discriminative abilities for IA of the 20 HAQ-questions were compared based on the area under the curve (AUC). For the best discriminating question, test characteristics and odds ratios (ORs) were calculated. ORs were adjusted for clinical characteristics that were previously reported to be predictive of the presence of IA (gender, age ≥60 years, symptom duration, acuteness of symptom onset, morning stiffness \u003e60 minutes, number of painful joints, presence of patient-reported swollen joint(s) and difficulty with making a fist)[2] using multivariable logistic regression. Results: IA was identified in 43% (derivation cohort), 53% and 35% (validation cohorts). In the derivation cohort, presence of IA associated with higher mean HAQ-scores (0.84 versus 0.73, p=0.003). The HAQ-question on difficulties with dressing yielded the highest AUC (0.58), which equalled discriminative ability of the total HAQ-score (AUC 0.55). Responses to this question were dichotomised into a simple binary score since loss of discriminative ability was minimal (AUC 0.57). Presence of any difficulties with dressing yielded ORs for IA of 1.80 (95%CI 1.39–2.33) in the derivation cohort; 2.00 (1.39–2.87) and 2.14 (1.48–3.10) in the validation cohorts. After adjustments for clinical characteristics the presence of any difficulties with dressing remained associated with the presence of IA; OR 1.71 (1.27–2.32) in derivation cohort and 1.64 (1.08–2.50) and 1.87 (1.20–2.92) in the validation cohorts. The prevalence of IA in case of presence of difficulties with dressing (positive predictive value) ranged 42–60% (see Figure). Conclusion: A yes/no answer on a simple question (“Are you able to dress yourself, including shoelaces and buttons?”) was helpful in discriminating patients with and without IA. Findings were validated in independent 1.5-line settings and need to be validated further in primary care. This is a step forward to arrive at practical tools that are helpful for GPs in identifying early IA. References: [1] van Nies JA et al. Improved early identification of arthritis: evaluating the efficacy of Early Arthritis Recognition Clinics. Ann Rheum Dis. 2013;72(8):1295–301. [2] ten Brinck RM et al. Development and validation of a clinical rule for recognition of early inflammatory arthritis. BMJ Open. 2019 Feb 22;8(11):e023552. Disclosure of Interests: Bastiaan van Dijk: None declared, Hanna W. van Steenbergen: None declared, Ellis Niemantsverdriet: None declared, Elisabeth Brouwer Consultant of: Roche (consultancy fee 2017 and 2018 paid to the UMCG), Speakers bureau: Roche (2017 and 2018 paid to the UMCG), Annette van der Helm - van Mil: None declared