Evaluating the Quality and Safety of Clinical Notes Generated by Ambient Digital Scribe Platforms using Simulated Ambulatory Encounters

Presentation Time: 04:06 PM - 04:18 PM

Abstract Keywords: Documentation Burden, Large Language Models (LLMs), Evaluation, Healthcare Quality, Patient Safety, Standards, Artificial Intelligence
Primary Track: Applications
Programmatic Theme: Clinical Informatics

Ambient digital scribe (ADS) platforms have become increasingly utilized for clinical documentation in the ambulatory setting despite little data regarding their performance. We assessed the accuracy, quality, and safety of output generated by five ADS platforms using simulated ambulatory encounters. Significant performance variability was observed across platforms and cases, indicating a need for standardized evaluation to facilitate consumer decision-making.

Speaker(s):
Taylor Anderson, MD
Stanford Healthcare

Author(s):
Vishnu Mohan, MD, MBI, FACP, FAMIA - Oregon Health & Science University; David Dorr, MD, MS, FACMI, FAMIA, FIAHSI - Oregon Health & Science University; Jeffrey Gold, MD - Oregon Heatlh & Science University;