The proxy problem: Child report versus parent report in health-related quality of life research