The Impact of GenAI on Health Economics: Observations and Applications

OBJECTIVES: Generative AI (GenAI) is revolutionizing the pharmaceutical and healthcare industries. Although only nascent in the field of health economic modelling, application of GenAI can significantly improve the quality and efficiency of model development. With quick advances of GenAI in recent years, this research aimed to identify its most recent use in the field of health economic modelling.

METHODS: A literature review was conducted in EMBASE to identify publications involving cost-effectiveness models developed entirely or with the support of GenAI. The search was restricted to widely used GenAI tools, including GPT, Copilot, Jasper, CodeWhisperer, Perplexity, Midjourney, Vertex and Gemini. No time restriction was applied to the search. Title and abstracts were screened by two independent reviewers before full text review was conducted.

RESULTS: The search identified 292 initial studies. The review revealed a limited number of publications integrating GenAI into cost-effectiveness modelling. Following eligibility criteria, only 2 studies were included as relevant. An additional 14 studies, which did not implement GenAI but explored its theoretical applications, were also summarized to elucidate the emerging trends in the scientific community. The two identified studies used GenAI for different purpose: one study attempted to replicate partition survival models using GenAI with high level of success. Support for operational tasks using Copilot tools was explored in the second study with very limited success.

CONCLUSIONS: Despite early promises, the integration of GenAI in cost-effectiveness modelling remains in its infancy. Challenges identified include the need for robust validation methodologies, transparency in model development, and adaptation to diverse healthcare settings. Ethical considerations regarding data privacy, algorithmic transparency, and stakeholder acceptance also pose substantial hurdles. Moving forward, concerted efforts are necessary to standardize methodologies, enhance reproducibility, and address regulatory concerns to realize the full potential of GenAI in improving the efficiency and accuracy of cost-effectiveness analyses in healthcare decision-making.