Feasibility, Validity and Differences in Adolescent and Adult EQ-5D-Y Health State Valuation in Australia and Spain: An Application of Best-Worst Scaling.
Dalziel K., Catchpool M., García-Lorenzo B., Gorostiza I., Norman R., Rivero-Arias O.
BACKGROUND: The measurement and valuation of health-related quality of life for and by young people are increasingly important, yet research on the impact of study perspective and validity of preferences obtained from young populations remains limited. OBJECTIVE: The objective of this study was to evaluate the feasibility and validity of collecting EQ-5D Youth version (EQ-5D-Y) preferences from adolescents, adults, and adults from a child perspective. METHODS: A profile case best-worst scaling (BWS) online survey was administered to representative Australian and Spanish adult (age ≥ 18 years) and child (age 11-17 years) samples. Adults were told to either answer from their own perspective or for a hypothetical 10-year-old child. Marginal best- and worst-choice frequencies, analysis of dominant choices, self-reported difficulty completing the tasks, and time to complete tasks were used to determine the validity of responses. RESULTS: In Australia, 2134 adults and 1010 adolescents completed the survey. In Spain, 2007 adults and 1000 adolescents completed it. Analysis of marginal choice frequencies and dominant choices indicated that the pattern of responses between adolescents and adults was similar. For Australian respondents, having no mobility problems was rated as best by adolescents, while adults rated having no pain and discomfort as 'best'. In Spain, both adults and adolescents rated no pain or discomfort as 'best'. Australian adolescents rated very worried, sad or unhappy as 'worst', while Spanish adolescents, Spanish adults and Australian adults rated a lot of pain and discomfort as 'worst'. CONCLUSIONS: Results suggest preferences from adolescents using direct BWS are valid. Our descriptive analysis also suggest that there are age-related and country-specific differences in elicitation values for the EQ-5D-Y.