A Comparison of the Psychometric Properties of the EQ-5D-Y-3L and EQ-5D-Y-5L Using Paediatric Multi-Instrument Comparison (P-MIC) Study Data

Mina Bahrampour; Nancy Devlin; Renee Jones; Kim Dalziel; Brendan Mulhern; On behalf of the QUOKKA (Quality of Life in Kids: Key Evidence for Decision Makers in Australia) Team

doi:10.1007/s40273-024-01379-7

A Comparison of the Psychometric Properties of the EQ-5D-Y-3L and EQ-5D-Y-5L Using Paediatric Multi-Instrument Comparison (P-MIC) Study Data

Pharmacoeconomics. 2024 May 15. doi: 10.1007/s40273-024-01379-7. Online ahead of print.

Authors

Mina Bahrampour¹, Nancy Devlin², Renee Jones², Kim Dalziel², Brendan Mulhern³; On behalf of the QUOKKA (Quality of Life in Kids: Key Evidence for Decision Makers in Australia) Team

Affiliations

¹ Centre for Health Economics Research and Evaluation (CHERE), University of Technology Sydney, Sydney, NSW, Australia. mina.bahrampour@uts.edu.au.
² Health Economics Unit, School of Population and Global Health, The University of Melbourne, Melbourne, VIC, Australia.
³ Centre for Health Economics Research and Evaluation (CHERE), University of Technology Sydney, Sydney, NSW, Australia.

PMID: 38748193
DOI: 10.1007/s40273-024-01379-7

Abstract

Background: The EQ-5D-Y is a generic health-related quality of life (HRQoL) measure suitable for children and adolescents. There are two versions of the instrument, which differ in response levels: the three- (Y-3L) and five-level (Y-5L) versions. The Y-5L has been developed to improve the psychometric performance of the Y-3L.

Objective: This study aims to assess the psychometric properties of Y-5L and to compare the performance of Y-5L with Y-3L in separate self- and proxy-reported samples.

Methods: Both versions of the instrument were administered to children/adolescents (self-report) or caregivers (proxy-report) of Australian children aged 5-18 years. The data were gathered as part of the Australian Paediatric Multi-Instrument Comparison (P-MIC) study. The Y-5L and Y-3L were evaluated in terms of ceiling effects, criterion validity, inconsistency, informativity, and test-retest reliability in both proxy and self-complete populations.

Results: Overall, 5945 respondents completed both the Y-3L and Y-5L, with 2083 proxy and 3862 self-completed responses. Ceiling effects were lower in the Y-5L than the Y-3L. Items from the same domains show the strongest correlation for both samples. Regarding informativity, the Y-5L demonstrated more discriminatory power, indicated by having a higher Shannon diversity index in all domains of the Y-5L compared with the Y-3L. Test-retest showed fair agreement between responses at baseline and follow-up.

Conclusion: The Y-5L has demonstrated moderate validity and fair reliability for use in measuring HRQoL in children and adolescents, both when self-reported by children or proxy reported. Overall, Y-5L is a dependable and valid extension from the Y-3L.

Abstract

Grants and funding