Psychometric validation of EQ-5D-Y in children with asthma and their caregivers

Background: The child-friendly EQ-5D (EQ-5D-Y) has been available in both a 3-level (Y-3L) and a 5-level (Y-5L) version. The study aimed to assess the measurement properties of both versions in paediatric patients with asthma and their caregivers. Methods: Patients and their caregivers completed the self-complete and proxy Y-5L and Y-3L questionanires, respectively, twice in two consecutive visits. Test-retest reliability and patient-caregiver agreement of Y-3L index, Y-5L level sum score (LSS) and EQ Visual Analogue Scale (VAS) were assessed using intraclass correlation coefficient (ICC). Known-groups validity compared scores of patients with different asthma control statuses. Responsiveness to improved overall health and asthma control was assessed uisng standardized effect size (SES). Results: Seventy-six patients (age range: 8-16.7 years) and their caregivers participanted. The ICC for test-retest reliability of the Y-3L, Y-5L and EQ VAS scores ranged from 0.744 to 0.898 for the slef-complete version and 0.525 to 0.767 for the proxy version. As hypothesized, Y-3L, Y-5L and EQ VAS scores derived from both EQ-5D-Y versions indicated that well or partially controlled patients had better health than uncontrolled patients. SES values ranged from 0.20 to 1.40 for self-complete version and 0.39 to 1.26 for proxy version. ICC for parient-caregiver agreement ranged from 0.556 for Y-5L LSS to 0.833 for Y-3L index. Conclusion: The results support the relaibility, validity, and responsiveness of both the self-complete and proxy versions of the two EQ-5D-Y questionnaires for assessing health outcomes of asthmatic children. Moreover, the self-complete and proxy-reported versions appears to have good agreement.