How Should Artificial Intelligence (Not) Be Used in Health Preference Research?

ISSUE: Health preference studies have successfully informed various decisions across medical product development, such as endpoint selection, regulatory benefit-risk assessment, and reimbursement. The studies typically require the design of a new preference elicitation instrument that is resource-intensive to design, pre-test, field, and analyse. Artificial intelligence (AI) could provide efficiencies and allow novel approaches for health preference elicitation. OVERVIEW: The panel will discuss the value of AI in health preference research. Prof. Marshall will open the panel by introducing key stages of health preference studies where AI is expected to bring value. Dr. Heidenreich will then discuss the use of AI in designing study materials using real-life examples. Dr. Boeri will discuss how existing literature can be reviewed with AI to support the development of health preference studies. Dr. Tervonen will discuss the use of AI in the analysis of qualitative preference data and social media listening. The panel will close with a discussion between panellists and the audience on their experience with artificial intelligence in health preference research. HEOR scientists benefit from attending this intermediate panel.