An Electronic Health Record-Integrated, Large Language Model-Powered Tool to Triage Surgical Patients.
Large language models-based tool identified surgical patients needing hospitalist comanagement with high sensitivity.
An Electronic Health Record-Integrated, Large Language Model-Powered Tool to Triage Surgical Patients.
IMPORTANCE: Surgical comanagement (surgical comanagement) is an evidence-based care model in which hospitalists jointly manage medically complex perioperative patients alongside surgical teams.
To assess whether surgical comanagement eligibility triage can be automated.
DESIGN, SETTING, AND PARTICIPANTS: This prospective, unblinded quality improvement study was conducted at Stanford Health Care from September 2025 to February 2026.
correctly flagged most patients who truly needed hospitalist comanagement
Overall, 14 of 16 surgical comanagement attending physicians participated in the study.
Among 6193 triaged surgical cases (median [IQR] age, 60.2 [41.6-71.3] years; 3036 [49.0% female]), 1582 (25.5%) were recommended for hospitalist consultation.
Post hoc medical record review suggested that most discrepancies reflected modifiable gaps in clinical criteria, institutional workflow, or physician practice variability rather than large language models misclassification, which accounted for 2 of 19 false-negative cases (11%).
In this prospective quality improvement study of an large language models-powered, electronic health record-integrated, human-in-the-loop AI system, AI-enabled screening tools accurately augmented surgical patient triage.