Patient-caregiver agreement and test–retest reliability of the EQ-5D-Y-3L and EQ-5D-Y-5L in paediatric patients with haematological malignancies

Background In 2019, the EuroQol Group developed a ‘beta’ 5-level version of EQ-5D-Y (Y-5L) by increasing the number of descriptive levels to five for each health dimension, as compared to the standard 3-level EQ-5D-Y (Y-3L). Objective To assess patient-caregiver agreement and test–retest reliability of the Y-5L and Y-3L in paediatric patients with haematological malignancies. Methods Paediatric inpatients aged 8–17 years were interviewed with the Y-5L and Y-3L questionnaires twice, while their caregivers were interviewed at the same time using the proxy versions of the questionnaires. Patient-caregiver agreement and test–retest reliability were assessed using Gwet’s agreement coefficient (Gwet’s AC1) for EQ-5D dimensions and the intraclass correlation coefficient (ICC) for the EQ VAS. Results Ninety-six patient-caregiver dyads participated in the study. Patient-caregiver agreement on the EQ-5D-Y descriptive system was moderate to good for both the Y-3L and Y-5L, but poor on the EQ VAS. Test–retest reliability of the descriptive system was good to very good for the Y-3L and moderate to good for the Y-5L in children, and fair to good for both versions of EQ-5D-Y in proxies. The EQ VAS showed good test–retest reliability in both children and caregivers. In a subgroup analysis of results in younger patients aged 8–10 years, patient-caregiver agreement and test–retest reliability were also observed to range from moderate to very good. Conclusion Both the Y-3L and Y-5L descriptive systems showed acceptable patient-caregiver agreement and test–retest reliability when used to assess the HRQoL of children and adolescents with haematological malignancies, including in younger patients.