Responsiveness of the 3-level and 5-level EQ-5D Youth Versions in Patients with Idiopathic Scoliosis

Abstract

Introduction: A new version of EQ-5D-Y increasing the number of response levels from 3 (3LY) to 5 (5LY) has been recently introduced. We developed the Chinese language version of 5LY for Hong Kong and compared the validity and reliability of 5LY and 3LY in adolescent idiopathic scoliosis patients. Since the ability of both 3LY and 5LY to capture change in health among children and adolescents is unknown, this study aims to investigate the responsiveness of the 5LY in idiopathic scoliosis patents, in comparison with 3LY. Methods: A total of 129 children or adolescents attending the paediatric spine clinics of a tertiary hospital in Hong Kong, were consecutively recruited between October 2017 to January 2018. After informed consent was obtained, eligible patients were asked to fill in the two youth versions of EQ-5D questionnaire (3LY and 5LY) and interviewed with questions assessing their socio-demographics and clinical characteristics. At 3-month follow-up, 111 (86.0%) patients completed the single-item Global Rating on Change Scale (GRS, for determining worsened, unchanged, or improved global health), 3LY and 5LY, with the remaining being unable to be contacted. Among those indicating unchanged in global health from baseline to follow-up, agreement in responses to each 3LY and 5LY item was calculated. As EQ-5D-Y value set was not available, EQ-5D-3L and EQ-5D-5L value sets for Chinese adult population were adopted for calculating EQ-5D scores. Mean changes in EQ-5D-Y scores over the past 3 months in patients with ‘worsen’, ‘unchanged’ and ‘improved’ health were calculated. A change in the utility scores of 0.04 points was considered as clinically important difference. Receiver operating characteristic (ROC) curve analysis was performed to examine the instruments’ responsiveness to change in health measured by GRS (worsen, unchanged and improved health). Area under the curve (AUC) was used as the measure of responsiveness, with AUC>0.6 being considered as adequate responsiveness. Results: Most patients (82.9%) reported no change in global health, while about 12.6% and 4.5% of them felt better and worse, respectively, compared to baseline. Among those reporting unchanged health, the ‘Looking after myself’ item exhibited the largest proportion of agreement in responses (5LY: 96.36%; 3LY: 95.50%), followed by ‘Mobility’ (5LY 90.91%; 3LY 90.99%), ‘Usual activities’ (5LY 83.64%; 3LY 87.39%), ‘Pain / discomfort’(5LY 68.18%; 3LY 76.58%), and ‘Feeling worried/sad/unhappy’ (5LY 66.36%; 3LY 72.07%). In the improved or worsen group, the 3-month follow-up 5LY and 3LY scores were higher or lower compared with baseline, respectively. Comparing patients with unchanged and worsened health, stable patients had higher 5LY and 3LY scores, with the mean difference being 0.05 and 0.10, respectively (AUC 0.70 and 0.69, respectively). Patients reporting improved/unchanged health got higher scores than patients in the worsened group, with a positive mean difference of 0.06 and 0.10 for 5LY and 3LY, respectively (AUC 0.69 and 0.69). Conclusions: The 5LY, the latest youth version of EQ-5D, is as responsive as the 3LY in the idiopathic scoliosis patients. Both instruments had adequate ability to predict change in overall health.