Health-Related Quality of Life Multi-Instrument Comparison for the Use in Children Aged 2-4 Years Old

OBJECTIVES: To test and compare the psychometric performance of the adapted EQ-5D-Y-3L/5L for use in children aged 2-4 years against CHU9D and HUI2/3 in children aged 2-4 years.

METHODS: Survey data of children aged 2-4 years were obtained from the Australian Pediatric Multi-Instrument Comparison (P-MIC) study. Psychometric performance of the instruments was assessed by; feasibility, convergent validity (Spearman’s correlations), ceiling effects, known-group validity (Cohen’s D effect sizes for pre-specified groups), test re-test reliability (intra-class correlation coefficients), and responsiveness (standardized response mean effect sizes for changes in health) by health status using level summed scores.

RESULTS: A total of 842 parents of children aged 2-4 years completed the survey with 33.1% of children having a special healthcare need. All instruments were easy to complete. High ceiling effects were observed for the adapted EQ-5D-Y-3L/5L and HUI2/3. The adapted EQ-5D-Y-3L had the largest Cohen’s D effect sizes for the majority of known groups analyses (54%) followed by CHU9D (38%). The adapted EQ-5D-Y-3L/5L had the largest interclass correlation coefficients for test re-test reliability (ICC:0.81,0.83). For responsiveness of improved health, the adapted EQ-5D-Y-3L, CHU9D, and HUI2/3 had the largest effect sizes (SRM:0.44-0.46). For responsiveness of worsened health, the adapted EQ-5D-Y-3L and HUI2/3 had the largest effect sizes (SRM:0.23).

CONCLUSIONS: The adapted EQ-5D-Y-3L/5L and CHU9D showed overall good psychometric performance. The HUI2/3 performed poorly for almost all known-groups analyses compared to the other instruments. The adapted EQ-5D-Y-3L appears to have the best psychometric performance, based on tests performed, for the use in children aged 2-4 years.