SAT0494\u2005HOME MONITORING OF INACTIVE DISEASE IN CHILDREN WITH JUVENILE IDIOPATHIC ARTHRITIS: PREDICTIVE VALUE OF EQ-5D-5L-Y

Abstract

Background: In recent years, juvenile idiopathic arthritis (JIA) research has shifted towards treat-to-target therapy based on clinical assessments and patient-reported outcomes (1). A well-known measurement of quality of life is the EQ-5D-5L (2). Herewith, we report preliminary results of a retrospective study using the child-friendly ‘EQ-5D-5L-Y’ with an E-health application (Reuma2GO) to monitor children with JIA in an outpatient setting. Objectives: To assess the relationship between dimensions of the health-related quality of life ‘EQ-5D-5L-Y’ questionnaire and conventional assessments for children with JIA, including the Juvenile Arthritis Multidimensional Assessment Report (JAMAR) and active joint count (AJC), and to investigate the potential of the EQ-5D-5L-Y as instrument for outpatient management. Methods: The study was designed as monocentric retrospective cohort study. Data from October 2017 to January 2019 were available for 70 patients with JIA. The relationships between individual dimensions of the EQ-5D-5L-Y, JAMAR and several clinical assessments were investigated. Furthermore, dimensions of the EQ-5D-5L-Y were investigated as possible predictors for binary disease activity using AJC > 0 as reference standard for active disease. Results: Seventy patients with JIA completed 115 EQ-5D-5L-Y and JAMAR questionnaires within two weeks before a clinical visit. Moderate to high correlations were found between the EQ-5D-5L-Y and JAMAR. Moreover, the best possible EQ-5D-5L-Y score, with and without health-related visual analogue scale (EQ-VAS), demonstrated high sensitivity (81.1%) and negative predictive value (84.8%) for active disease (Table 1). The few patients who were incorrectly classified as having inactive disease (false-negatives) did not have their medication changed at the clinical visit and experienced little to no impact of disease activity on their quality of life, as indicated by the JAMAR questionnaire.Table 1 Predictive value of EQ-5D-5L-Y dimensions for active disease using AJC > 0 as reference standard. Cut-off for ‘Active Disease’ Accuracy (%, CI) Sensitivity Specificity PPV NPV EQ-5D Mobility > 1 74.8% (65.8-82.4) 59.5% 82.1% 61.1% 81.0% EQ-5D Self-Care > 1 76.5% (67.7-83.9) 35.1% 96.2% 81.2% 75.8% EQ-5D Usual Activities > 1 67.8% (58.5-76.2) 73.0% 65.4% 50.0% 83.6% EQ-5D Pain/Discomfort > 1 62.6% (53.1-71.5) 62.2% 62.8% 44.2% 77.8% EQ-5D Anxiety/Depression > 1 73.9% (64.9-81.7) 51.4% 84.6% 61.3% 78.6% EQ-5D EQ-VAS <85 60.0% (50.4-69.0) 64.9% 57.7% 42.1% 77.6% EQ-5D Total Score > 5 60.0% (50.4-69.0) 81.1% 50.0% 43.5% 84.8% EQ-5D Score > 5 & EQ-VAS <85 52.2% (42.7-61.6) 83.8% 37.2% 38.8% 82.9% PPV: positive predictive value; NPV: negative predictive value. Conclusion: These results demonstrate the discriminatory value of the EQ-5D-5L-Y between active and inactive disease in our cohort of patients with JIA. High negative predictive value was found for the total EQ-5D-5L-Y score, with and without EQ-VAS. In conclusion, the EQ-5D-5L-Y could be a valuable instrument for monitoring children with JIA in an outpatient setting which could aid physicians with deciding whether a clinical visit is necessary or not. References [1] Filocamo G, et al. A new approach to clinical care of juvenile idiopathic arthritis: The juvenile arthritis multidimensional assessment report. J Rheumatol2011;38:938–953. [2] Janssen MF, et al. Measurement properties of the EQ-5D-5L compared to the EQ-5D-3L across eight patient groups: a multi-country study. Qual Life Res2013;22:1717–1727. Disclosure of Interests: Martijn J.H. Doeleman: None declared, Sytze De Roock: None declared, Nathan Buijsse: None declared, Mark Klein: None declared, Gouke J. Bonsel: None declared, V. Seyfert Shareholder of: VS is CEO and founder at MyOwnMed, Inc., Employee of: VS is CEO and founder at MyOwnMed, Inc., Nico Wulffraat: None declared, Joost F. Swart: None declared