Roles of Artificial Intelligence--Based Synthetic Data in Health Economics and Outcomes Research

This is the authors' abstract. We don't add key points for this paper.

OBJECTIVES: We aim to raise awareness of potential applications of synthetic data within the health economics and outcomes research (HEOR) community. METHODS: We provide a concise overview of synthetic data, including data generation and types. We then discuss 3 major data-associated challenges and how synthetic data may be used to address them. Finally, we discuss data utility, privacy protection, potential concerns of its applicability, and future research direction. RESULTS: The use of synthetic data is an alternative privacy protection technique to enhance data availability, strengthen the robustness of findings for underrepresented populations, and alleviate data insufficiency issues in rare disease research. More studies are needed to explore synthetic data use and address data challenges in HEOR studies. Furthermore, the development of an evaluation framework is encouraged to better support the integration of synthetic data into the HEOR field. CONCLUSIONS: Synthetic data provide a unique opportunity to overcome data-related challenges in HEOR.