OBJECTIVES: EQ-5D-Y is a generic health related quality of life (HRQoL) measure suitable for children and adolescents. This measure has five dimensions; mobility, looking after self, doing usual activities, pain and feeling worried sad or unhappy. There are two versions of the instrument that differ in terms of response levels; a three (EQ-5D-Y-3L) (Y-3L) and a five level (EQ-5D-Y -5L) (Y-5L). The Y-5L has been developed to improve the psychometric performance of the Y-3L. This study aims to assess the psychometric properties of Y-5L and to compare the performance of Y-5L and Y-3L in proxy and self-report.

METHODS: Both versions of the instrument were administered to children/adolescents (self-report) or caregivers (proxy-report) from the Australian population. The data used were from the Australian Paediatric Multi-Instrument Comparison Study (P-MIC) [i]. The Y-5L and Y-3L were evaluated in terms of acceptability (ceiling effects), informativity (Shannon index), and validity (correlation) in both proxy and self-completion.

RESULTS: When proxy and self report data from Y-3L and Y-5L, there is some difference in the ceiling effect and Informativity between the two versions in both groups. The ceiling effect was lower for the Y-5L in both groups, and it was slightly higher in the proxy reports compare to self report. In both groups, items from same domains show a strong correlation. Regarding informativity, Y-5L version has more discriminatory power, with a higher Shannon diversity index in all dimensions of the Y-5L compared to the Y-3L.

CONCLUSIONS: Our results suggest Y-5L is comparable in terms of validity to the Y-3L. Proxies and children respond to the instruments differently. This has implications for the interpretation of HRQoL data from self and proxy reports.