Psychometric Properties of the Spanish Versions of EQ-5D-Y-3L and EQ-5D-Y-5L in Children with Cancer: A Comparative Study

Abstract

Background: The recent published version with five levels of response of EQ-5D-Y needs to be studied in children with chronic illness. For this, the aim of the present study was to assess and compare the psychometric properties of EQ-5D-Y-3L and EQ-5D-Y-5L in terms of feasibility, ceiling effect, redistribution properties, informativity and inconsistence responses in children with cancer. (2) Methods: A core set of self-report tools, including the Spanish version of EQ-5D-Y-3L and EQ-5D-Y-5L, were administered to children drawn from the population with cancer. EQ-5D-Y-3L and EQ-5D-Y-5L were evaluated in terms of feasibility, ceiling effects, redistribution properties and differences in absolute and relative informativity. (3) Results: A total of 73 children (9.7 ± 2.3 years old) from the population with cancer participated in the study. No missing data in the new EQ-5D-Y-5L were visualized, so the feasibility was acceptable. EQ-5D-Y-5L showed a low ceiling effect in all dimensions with relative changes from EQ-5D-Y-3L to EQ-5D-Y-5L of between 15.3% and 42.4% for the dimensions and 44.6% for the overall system. Compared to EQ-5D-Y-3L, EQ-5D-Y-5L provided a better distribution of the severity of the problem in the five levels of response. The absolute informativity (Shannon’s index) did not show statistically significant differences between EQ-5D-Y-3L and EQ-5D-Y-5L in all dimensions and the overall system. (4) Conclusions: EQ-5D-Y-5L is feasible, presenting a low ceiling effect and high discriminative power