The Future of Data-Driven HEOR Decision-Making Powered by Generative AI: How Soon is Now?


Xiaoyan Wang, PhD, Intelligent Medical Objects, Rosemont, IL, USA, Hua Xu, PhD, Yale University School of Medicine, New Haven, CT, USA, Abeed Sarker, PhD, Emory University School of Medicine, Atlanta, GA, USA and Ergin Soysal, MD PhD, Eli Lilly and Company, Indianapolis, IN, USA


This session will present the findings of the rapidly advancing generative AI, particularly Large Language Models (LLMs) across clinical development ranging from systematic literature review, in depth RWE, social media listening to RWD-informed clinical trials for decision making. Furthermore, it will highlight the unique impact of AI-enabled insight generation from different perspectives including health systems, pharmaceutical and technology industries.


Everyday billions of data points are collected in diverse health data platforms. Expanding health data and innovation in AI modeling are coinciding with health challenges of a scale we’ve never seen — creating urgency for true and rich insights for decision making. In this session, Dr. Xiaoyan Wang from IMO will delve into the advancements of generative AI and LLMs, focusing specifically on her recent research in AI-enabled evidence synthesis from literature and clinical RWD for oncology clinical trials (12’). Dr. Hua Xu from Yale University will provide an overview of the Observational Health Data Sciences and Informatics (OHDSI) consortium he leads on AI methods to scale the use of clinical notes in RWE studies of mental health (12’). Dr. Abeed Sarker from Emory University will present on how AI has powered ”social listening” on self-reported outcomes and the formation of auto-growing chronic pain cohorts from social media platforms for clinical development of treatment strategies (12’). Dr. Ergin Soysal from Eli Lilly will share his insights from a pharmaceutical industry perspective on integrating AI into clinical trials, and whether AI and LLMs can reduce costs and accelerate the scientific process in HEOR (12’)

We will review the opportunities and challenges of AI-powered insight generation, assessing its reliability for adequate, well-controlled investigations and confirmatory evidence for payers and regulatory decision-making (all panelists). Attendees are invited to engage with the panelists through a digital poll and Q&A during this session (12’).




Medical Technologies