A head-to-head comparison of five-level (EQ-5D-5L-Y) and three-level EQ-5D-Y questionnaires in paediatric patients

Purpose The aim of this study was to assess the psychometric properties of a youth version of the EQ-5D five-level questionnaire (5LY) and its three-level version (3LY) in a sample of Chinese paediatric patients. Methods A consecutive sample of idiopathic scoliosis patients were recruited from a referral outpatient scoliosis center at Hong Kong, China in October 2017 and completed the two versions of EQ-5D-Y. Redistribution properties in each dimension of EQ-5D-Y were analyzed between 5LY and 3LY by logistics regressions. Absolute reduction and relative reduction in ceiling effects from the 3LY to the 5LY were calculated. Test-retest reliability was assessed by examining the Gwet's agreement coefficient (Gwet's AC) for five individual dimension responses over the 2-week period. Results A total of 129 idiopathic scoliosis patients completed the two versions of EQ-5D-Y at baseline assessment, among which 70 patients completed the test-retest interview in 2-3 weeks after baseline assessment. For redistribution properties, the proportion of inconsistency was low in all the dimensions, ranging from 0.0% ("Usual activities") to 3.9% ("Pain/discomfort"). Ceiling effects were reduced in four dimensions. "Usual activities" dimension showed significant reduction (absolute and relative reductions: 3.9% and 4.3%; p = 0.025) and the "worried/sad/unhappy" dimension showed the largest significant reduction in ceiling effects (absolute and relative reductions: 7.8% and 9.8%; p = 0.012). The 3LY and 5LY showed very good agreement (>80%) of individual dimension responses between two assessments, except for the "worried/sad/unhappy" dimension in 3LY. Conclusion Through this head-to-head comparison, the 5LY had significant improvements in ceiling effects in two dimensions when compared to 3LY but other measurement properties of 3LY and 5LY performed similar in the idiopathic scoliosis patient group.