Research reveals AI's potential to transform hospital reporting processes

The study results, published in the October 21, 2024 online edition of the New England Journal of Medicine (NEJM) AI, found an AI system using large language models (LLMs) can accurately process hospital quality measures, achieving 90% agreement with manual reporting, which could lead to more efficient and reliable approaches to health care reporting.

Researchers of the study, in partnership with the Joan and Irwin Jacobs Center for Health Innovation at UC San Diego Health (JCHI), found that LLMs can perform accurate abstractions for complex quality measures, particularly in the challenging context of the Centers for Medicare & Medicaid Services (CMS) SEP-1 measure for severe sepsis and septic shock.

Aaron Boussina, postdoctoral scholar and lead author of the study at UC San Diego School of MedicineThe integration of LLMs into hospital workflows holds the promise of transforming health care delivery by making the process more real-time, which can enhance personalized care and improve patient access to quality data. As we advance this research, we envision a future where quality reporting is not just efficient but also improves the overall patient experience."

Traditionally, the abstraction process for SEP-1 involves a meticulous 63-step evaluation of extensive patient charts, requiring weeks of effort from multiple reviewers. This study found that LLMs can dramatically reduce the time and resources needed for this process by accurately scanning patient charts and generating crucial contextual insights in seconds.

By addressing the complex demands of quality measurement, the researchers believe the findings pave the way for a more efficient and responsive health care system.

Other key findings of the study found that LLMs can improve efficiency by correcting errors and speeding up processing time; lowering administrative costs by automating tasks; enabling near-real-time quality assessments; and are scalable across various health care settings.

Co-authors of this study include Shamim Nemati, Rishivardhan Krishnamoorthy, Kimberly Quintero, Shreyansh Joshi, Gabriel Wardi, Hayden Pour, Nicholas Hilbert, Atul Malhotra, Michael Hogarth, Amy Sitapati, Karandeep Singh, and Christopher Longhurst, all with UC San Diego.

This study was funded, in part, by the National Institute of Allergy and Infectious Diseases (1R42AI177108-1), the National Library of Medicine (2T15LM011271-11 and R01LM013998) and the National Institute of General Medical Sciences (R35GM143121 and K23GM146092) and JCHI.

Source:

University of California - San Diego

Journal reference: