Discrete choice experiments or best-worst scaling? A qualitative study to determine the suitability of preference elicitation tasks in research with children and young people.

ABSTRACT

BACKGROUND: Ordinal tasks are increasingly used to explore preferences for health states. This study aimed to determine the suitability of two ordinal preference elicitation tasks (discrete choice experiments (DCE) and best-worst scaling (BWS)) for use with children and young people to generate health state utility values. The study explored children's understanding, the relationship between their age and level of understanding, and how many tasks they felt they could complete. METHODS: Children aged 11-16 years were recruited from a secondary school in South Yorkshire, UK. Participants were asked to 'think aloud' as they completed a computer-based survey that contained both DCE and BWS tasks relating to dental caries (tooth decay) health states. Health states involved descriptions of the impact of tooth decay on children's daily lives. One-to-one semi-structured interviews were then held with participants, with use of a topic guide. Qualitative data were transcribed verbatim and analysed thematically. RESULTS: A total of 33 children (12 male, 21 female) participated, comprising 5-6 children from each school year group. Children expressed a preference for BWS and demonstrated a better understanding of these tasks than DCE. There was no clear relationship between children's level of understanding and age. Children felt they could manage between 8 and 10 BWS tasks comfortably. CONCLUSION: This study suggests that BWS tasks are the most appropriate type of preference elicitation task to value health states for children and young people aged 11-16 years to complete.