This study evaluated and compared the psychometric properties (ceiling effects, construct validity, and responsiveness) of EQ-5D-Y-3L (Y-3L), EQ-5D-Y-5L (Y-5L), and child health utility 9D (CHU9D) in pediatric patients with eczema.
Patients completed the Y-3L, Y-5L, CHU9D, and the Child Dermatology Life Quality Index and were followed up during their next visit. Clinicians assessed patients’ skin status using the Validated Investigator Global Assessment scale for Atopic Dermatitis. Ceiling effects of the 3 instruments were compared. Convergent validity was assessed using correlation with the Child Dermatology Life Quality Index; known-group validity was evaluated using F-statistics; and responsiveness was assessed using standardized effect sizes (SESs) among patients reporting improved health.
A total of 164 patients (mean age: 12.2 years; female: 47.6%) participated in the study. Of those, 67 patients (mean age:11.9; female: 53.8%) completed the follow-up survey. The ceiling effects of Y-3L, Y-5L, and CHU9D were 34.8%, 28.7%, and 9.8%, respectively. All known-group hypotheses, defined by the Validated Investigator Global Assessment scale for Atopic Dermatitis and scratching frequency, were met, with Y-5L showing the strongest performance in discriminating between patients based on symptom severity. Convergent validity was met in 70.0%, 70.0%, and 83.3% of hypotheses tested for Y-3L, Y-5L, and CHU9D, respectively. CHU9D (SES = 0.53) was relatively more responsive to change than Y-3L (SES= 0.18) or Y-5L (SES = 0.43).
Y-3L and Y-5L were more sensitive to clinical outcome differences, whereas CHU9D appeared to be more responsive to self-rated health improvement. Further research using clinical measures and more diverse eczema patient samples is needed to confirm these findings and support evidence-based instrument selection.