Measurement and valuation of health among children and adolescents using the EQ-5D-Y : methodological considerations and implications

Abstract

Introduction: Around the world, life expectancy is increasing but not all years are lived in full health. Hence it is important to look beyond survival rates and also measure healthrelated quality of life. Measurement of health-related quality of life can be used to evaluate population health and treatment effects in clinical practice and to assess the cost-effectiveness of treatments or interventions. There has been increased interest in measuring health-related quality of life among children and adolescents and several instruments have been developed for this purpose. The uniqueness of health in younger ages needs to be considered, as childhood and adolescence usually encompass vulnerable periods. The ability to self-report health is related to the cognitive and emotional development of a person, irrespective of age. Health-related quality of life instruments can be divided into condition-specific and generic instruments. The EQ-5D-Y is a generic health-related quality of life instrument, developed to be suitable for self-completion by children from the age of eight years. The instrument covers five dimensions of health (mobility; looking after myself; doing usual activities; having pain or discomfort; feeling worried, sad or unhappy), with either three or five severity levels.

There is a lack of population reference data for the EQ-5D-Y in Sweden and the use of the instrument in psychological or psychiatric conditions is relatively limited. Valuation of health is needed to combine responses related to the different dimensions and severity levels into a single index value, as different dimensions of health might affect the overall health status differently. Several methodological studies have been undertaken to derive values for the EQ- 5D-Y instrument and value sets have been developed either using experience-based values derived from children themselves or by asking an adult general population to value described health states for a child. There is no value set developed for the EQ-5D-Y in Sweden, and there is a lack of knowledge regarding people’s perceptions of valuing health states for children.

Aim: The overall aim of this thesis was to increase knowledge on measuring health-related quality of life among children and adolescents in various contexts and to explore methods for valuing health states using the EQ-5D-Y instrument.

Methods: The four papers of this thesis focused on population health, applications of the EQ-5D-Y instrument in psychiatric inpatient care of children and adolescents, exploration of experience-based values and people’s perceptions of valuing health states. In Papers I and III, data from a cross-sectional survey among a general population of adolescents were used to develop population reference data for the EQ-5D-Y-3L instrument and to explore experience-based values. Potential participants were asked to complete the instrument during school hours. In Paper I, regression analysis was used to investigate associations between problems reported in the EQ-5D-Y dimensions and mean Visual Analogue Scale (VAS) score and sex, age, self-reported comorbidity and parents’ socio-economic status. In Paper III, regression analysis was used to investigate if and how problems reported in the EQ-5DY- 3L predicted the VAS value. Paper II included patients in child and adolescent psychiatric inpatient care, with the aim to measure health-related quality of life and to assess the feasibility and construct validity of the EQ-5D-Y-5L by investigating correlations between the EQ-5D-Y-5L dimensions EQ VAS and the Strengths and Difficulties Questionnaire domains. In Paper IV, adolescents’ and adults’ perceptions of valuing health states for children were explored in a qualitative study. Adolescents and adults from the general population participated in individual Zoom interviews and were asked to complete several valuation tasks (Time Trade-Off (TTO) and Discrete Choice Experiment (DCE)) in accordance with a standardised valuation protocol. This was followed by a semi-structured interview. Qualitative content analysis was performed.

Findings: In Paper I, girls in general reported more problems than boys in the EQ-5D-Y-3L dimensions and lower mean EQ VAS scores. Respondents with one or both parents unemployed reported more problems than those with both parents employed. The instrument was able to distinguish in health status between those with and without self-reported comorbidity. In Paper II, participants reported problems at all severity levels in most of the EQ-5D-Y-5L dimensions. As expected, most problems were reported in the mood dimension. The mean EQ VAS score was found to be very low. Feasibility was supported and construct validity indicated as some of the expected correlations between the EQ-5D-Y-5L and the Strength and Difficulties Questionnaire were found. In Paper III, the largest decrements in VAS values were observed for the mood dimension. All models performed similarly in terms of monotonicity and goodness of fit. In Paper IV, the qualitative content analysis resulted in two main categories ‘Thoughts and feelings when valuing children’s health states’ and ‘Strategies when valuing children’s health states’. Participants felt terrible and doubtful when asked to choose between life years and health for a 10-year-old child and making decisions for a child was perceived as difficult and emotional. It was questioned who has the right to value health states for children.

Conclusions: The papers in this thesis have contributed to increased knowledge on the measurement and valuation of health among children and adolescents. Population reference data for Sweden are now available for the EQ-5D-Y-3L and initial support for the use of the EQ-5D-Y-5L instruments within the context of child and adolescent psychiatric inpatient care is provided. Adolescents were able to value health states with the VAS, TTO and DCE methods, which indicates that it is possible to capture aspects that are important to young people, a group for whom decisions regarding treatments and interventions will be made. Mental health was expressed as the most important dimension of health when valuing health states.