Artificial Intelligence

Reshaping HEOR—responsibly

Artificial intelligence (AI) has quickly risen to the top of the trends list as its use has come to permeate virtually every aspect of life in recent years, including healthcare research and decision making. In health economics and outcomes research (HEOR), AI is accelerating how research and evidence is generated, analyzed, and communicated while reinforcing the importance of responsible use and human oversight.

AI tools can process massive volumes of data far more quickly than traditional methods, helping researchers and decision makers extract insights that support better healthcare decisions. Two main types of AI are being used across all industries: large language models (LLMs), which take large quantities of text and qualitative data and synthesize it into meaningful nuggets; and machine learning (ML), which essentially trains a computer to do prediction modeling.

How does AI support healthcare decision making?

In HEOR, AI and machine learning are being used to:

• Analyze complex data at scale — Organize and interpret large, high-dimensional datasets, including real-world evidence, to uncover patterns related to safety, effectiveness, and value
• Interpret data from devices — Make sense of information that may otherwise overwhelm the system, which is beneficial as massive amounts of new data are becoming available through continued innovation and use of digital health technologies and wearables
• Accelerate evidence generation — Perform tasks such as literature reviews, data structuring, and modeling more efficiently, allowing researchers to focus on interpretation and analysis
• Advance predictive modeling — Forecast the long-term outcomes and costs associated with different treatments, to help assess the value and cost-effectiveness of healthcare interventions
• Support personalized medicine — Analyze patient data to determine which treatments are likely to be most effective for individual patients
• Enhance research efficiency — Automate time-intensive processes such as data cleaning, integration, and model development

Why does responsible use matter?

As AI use continues to expand, so does the need for safeguards, as, for example, recent news has uncovered how AI-generated reports have cited studies that do not actually exist. Human oversight remains essential to ensure accuracy, transparency, and authentic patient-centered decision making. Responsible AI frameworks help ensure that AI augments, rather than replaces, human judgment.