We compared the measurement properties of 2 proxy versions (proxy-proxy perspective, P1 and proxy-patient perspective, P2) of EQ-5D-Y-3L (Y-3L) and EQ-5D-Y-5L (Y-5L) for assessing the health-related quality of life (HRQoL) of pediatric patients with asthma or eczema.
We recruited pediatric patients with asthma or eczema and their parents or legal guardians from 2 tertiary hospitals in Singapore. Patients completed Y-3L, Y-5L, and a disease-specific HRQoL questionnaire. Proxies completed either P1 or P2 version of Y-3L and Y-5L. Asthma control and eczema severity were assessed by clinicians. A subset of patient-proxy dyads was reassessed (follow-up duration: 101.3 ± 63.6 days). The known-groups validity, responsiveness to improvement, and patient-proxy agreement of both proxy versions were assessed using the Y-3L index score (3L-Index), Y-5L level sum score (5L-LSS), and EQ visual analog scale scores.
We surveyed 328 child-proxy dyads (P1: 154 dyads; P2: 174 dyads). P2 (47.1%-56.3%) exhibited lower ceiling effects than P1 (53.9%-60.4%). All P1-derived scores differed as expected between Global Initiative for Asthma- or Validated Investigator Global Assessment for Atopic Dermatitis-defined known groups but not all P2-derived scores differed as expected. P1/P2 standardized effect size based on improvement in P1/P2-generated scores was 0.48/0.21 for 3L-Index and 0.28/0.42 for 5L-LSS. Patient-proxy agreement was low for all proxy versions (Intraclass correlation coefficient, ICC: 0.22-0.42). Proxies reported better HRQoL than patients regardless of the version used.
Neither proxy version of the 2 EQ-5D-Y questionnaires displayed a clear superiority, suggesting that assessment perspective may not affect proxy-reported HRQoL of children with asthma or eczema. Researchers and clinicians should select the proxy perspective based on the purpose of evaluation.