Use of a Large Language Model to Assess Clinical Acuity of Adults in the Emergency Department

Key Points Question Can a large language model (LLM) accurately assess clinical acuity in the emergency department (ED)? Findings This cross-sectional study of 251 401 adult ED visits investigated the potential for an LLM to classify acuity levels of patients in the ED based on the Emergency Severity Index across 10 000 patient pairs. The LLM demonstrated accuracy of 89% and was comparable with human physician classification in a 500-pair subsample. Meaning These findings suggest that LLMs could accurately identify higher-acuity patient presentation when given pairs of presenting histories extracted from patients’ first ED documentation.