Construct validation of the Health Utilities Index and the Child Health Questionnaire in children undergoing cancer chemotherapy.

Abstract

The objective of this study was to evaluate the construct validity of two questionnaire-based measures of health-related quality of life (HRQL) in children undergoing cancer chemotherapy: the Health Utilities Index (HUI) and the Child Health Questionnaire (CHQ). Subjects were children hospitalised for chemotherapy. To examine construct validity: (1) a priori expected relations between CHQ concepts and HUI attributes were examined; (2) HUI and CHQ summary scores were compared to visual analogue scale (VAS) scores. Ease of completion was rated using a 5-point categorical scale and completion time was recorded. A total of 36 subjects were included. The maximum score was seen in 15 (47%) of HUI3 assessments. As predicted, CHQ body pain was moderately correlated with HUI3 pain (r=0.51), CHQ physical functioning was moderately correlated with HUI2 mobility (r=0.58) and CHQ mental health was moderately correlated with HUI2 emotion (r=0.53). Only the CHQ psychosocial subscale (and not HUI) was correlated with VAS (r=0.44). The CHQ and the HUI were both easy to use. The HUI questionnaires required less time to complete (mean=3.1, s.d.=1 min) compared with CHQ (mean=13.1, s.d.=3.4 min, P<0.0001). In conclusion, HUI and CHQ demonstrated construct validity in children undergoing cancer chemotherapy. The Health Utilities Index is subject to a ceiling effect whereas CHQ requires more time to complete.