Are You Still Not Integrating Gen-AI Frameworks in HEOR Workflows? It's No Longer Optional!
Author(s)
Tushar Srivastava, MSc;
ConnectHEOR, London, United Kingdom
ConnectHEOR, London, United Kingdom
Presentation Documents
OBJECTIVES: Health economics and outcomes research (HEOR) involves complex processes—from systematic literature reviews and economic model development to stakeholder communication and policy decision-making. Traditionally, these tasks are resource-intensive, prone to human error, and limited by manual workflows. This study examines how Gen-AI has evolved into an essential tool in HEOR, rather than a mere add-on, by automating and enhancing various stages of the HEOR pipeline.
METHODS: A targeted literature and industry review was conducted to identify use cases where Gen-AI significantly improves HEOR processes. Sources included interviews, recent publications, case studies, and conference proceedings (January 2022-December 2024). Examples highlighting efficiency gains, error reduction, and improved decision-making were catalogued under four key areas: health economic modeling, systematic reviews, RWE generation, and result dissemination.
RESULTS: Gen-AI automated conceptualization and code generation, speeding up model development by around 50%. AI-driven literature screening improved speed and accuracy by upto 98% in systematic reviews while summarization of text was performed using fine-tuned transformer models like BERT. Gen-AI transformed real world evidence analysis by converting unstructured clinical notes into analyzable data and identifying patient patterns for exploratory analyses. Mathematically fine-tuned LLMs can also be used to perform early cost-effective calculations for healthcare interventions. Gen-AI fostered compatibility among team members of varying domain-expertise by inter-converting information from one domain’s jargon to another. Multimodal LLMs converted outputs into accessible formats such as audio and visual formats.
CONCLUSIONS: Gen-AI is no longer an option but a necessity in HEOR. It offers systematic, scalable solutions that improve data handling, analysis, and communication across stakeholders. By embracing AI-driven methodologies, HEOR teams can produce more timely, accurate, and user-friendly evidence, strengthen the overall decision-making process in healthcare. As complexity of healthcare data continues to grow, integrating Gen-AI will be essential for organizations seeking to remain at the forefront of innovative and impactful HEOR practices.
METHODS: A targeted literature and industry review was conducted to identify use cases where Gen-AI significantly improves HEOR processes. Sources included interviews, recent publications, case studies, and conference proceedings (January 2022-December 2024). Examples highlighting efficiency gains, error reduction, and improved decision-making were catalogued under four key areas: health economic modeling, systematic reviews, RWE generation, and result dissemination.
RESULTS: Gen-AI automated conceptualization and code generation, speeding up model development by around 50%. AI-driven literature screening improved speed and accuracy by upto 98% in systematic reviews while summarization of text was performed using fine-tuned transformer models like BERT. Gen-AI transformed real world evidence analysis by converting unstructured clinical notes into analyzable data and identifying patient patterns for exploratory analyses. Mathematically fine-tuned LLMs can also be used to perform early cost-effective calculations for healthcare interventions. Gen-AI fostered compatibility among team members of varying domain-expertise by inter-converting information from one domain’s jargon to another. Multimodal LLMs converted outputs into accessible formats such as audio and visual formats.
CONCLUSIONS: Gen-AI is no longer an option but a necessity in HEOR. It offers systematic, scalable solutions that improve data handling, analysis, and communication across stakeholders. By embracing AI-driven methodologies, HEOR teams can produce more timely, accurate, and user-friendly evidence, strengthen the overall decision-making process in healthcare. As complexity of healthcare data continues to grow, integrating Gen-AI will be essential for organizations seeking to remain at the forefront of innovative and impactful HEOR practices.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
SA30
Topic
Study Approaches
Disease
No Additional Disease & Conditions/Specialized Treatment Areas