This article presents a taxonomy of generative artificial intelligence (AI) for health economics and outcomes research (HEOR), explores emerging applications, outlines methods to improve the accuracy and reliability of AI-generated outputs, and describes current limitations.
Foundational generative AI concepts are defined, and current HEOR applications are highlighted, including for systematic literature reviews, health economic modeling, real-world evidence generation, and dossier development. Techniques such as prompt engineering (eg, zero-shot, few-shot, chain-of-thought, and persona pattern prompting), retrieval-augmented generation, model fine-tuning, domain-specific models, and the use of agents are introduced to enhance AI performance. Limitations associated with the use of generative AI foundation models are described.
Generative AI demonstrates significant potential in HEOR, offering enhanced efficiency, productivity, and innovative solutions to complex challenges. Although foundation models show promise in automating complex tasks, challenges persist in scientific accuracy and reproducibility, bias and fairness, and operational deployment. Strategies to address these issues and improve AI accuracy are discussed.
Generative AI has the potential to transform HEOR by improving efficiency and accuracy across diverse applications. However, realizing this potential requires building HEOR expertise and addressing the limitations of current AI technologies. Ongoing research and innovation will be key to shaping AI’s future role in our field.
What is it about? Generative artificial intelligence (AI), a branch of AI that creates new content, is increasingly influencing health economics and outcomes research. This technology is important because it can significantly improve efficiency and accuracy in various research tasks. However, the challenge lies in ensuring the scientific reliability of AI-generated outputs. The manuscript identifies gaps in the current understanding of AI's application in this field and suggests methods to improve its accuracy and dependability. The article contributes to a better understanding of how generative AI can transform health economics research by highlighting opportunities and challenges.
How was the research conducted? The article is based on a comprehensive review and synthesis of generative AI applications in health economics. Researchers identified uses of GenAI to explore systematic literature reviews, health economic modeling, real-world evidence generation, and dossier development. They reviewed these applications to understand the potential and limitations of AI in each area.
What were the results? The article found that generative AI holds significant potential in health economics and outcomes research, particularly in enhancing efficiency and productivity. Additional findings include AI's ability to streamline complex processes, such as systematic reviews and economic modeling. However, the article notes the persistent challenge of ensuring the scientific accuracy of AI-generated outputs, which remains a barrier to its full integration into research practices.
Why are the results important? These results are important for health technology assessment agencies as the findings highlight the need for guidelines and standards when using AI tools. In practical terms, these findings suggest that AI has the potential to make research processes faster and more efficient. Healthcare professionals, researchers, and decision makers can all benefit from these advancements by improving the quality and speed of research. Long-term, the results could lead to more innovative healthcare solutions and improved patient outcomes.
What are the strengths and weaknesses of this study? The article's strength lies in its comprehensive overview of AI applications across multiple domains within health economics. However, a limitation is that the field is advancing rapidly and the article may not include the latest applications.
By providing a clear understanding of generative AI's potential and challenges in health economics, the review offers valuable insights for patients, healthcare decision makers, and researchers. It emphasizes the need for ongoing research and collaboration to address current limitations and harness AI's full potential in improving healthcare outcomes.
Note: This content was created with assistance from artificial intelligence (AI) and has been reviewed and edited by ISPOR staff. For more information or for inquiries on ISPOR’s AI policy, click here or contact us at info@ispor.org.