UCLA Study Finds AI Ambient Scribes Reduce Documentation Time and Improve Physician Well-Being
A randomized clinical trial at UCLA compared two ambient AI scribe systems and found meaningful reductions in documentation time per note and improvements in physician burnout measures, though AI-generated notes occasionally contained clinically significant inaccuracies.
UCLA Health has published results from a randomized clinical trial comparing two ambient AI scribe platforms — Nabla and Microsoft DAX — against standard documentation workflows. Physicians using Nabla saw average note-writing time decrease by 41 seconds per note, from 4 minutes 30 seconds to 3 minutes 49 seconds, representing a clinically meaningful reduction in daily documentation burden.
Both AI tools showed modest but statistically significant improvements in validated measures of physician burnout, cognitive workload, and work exhaustion. Stanford Health Care contributed a parallel analysis that revealed large reductions in task load and burnout scores with statistical significance.
A broader multi-system quality improvement study confirmed these findings, showing ambient AI scribes were associated with significant reduction in burnout and cognitive task load. Physicians also perceived the tools could improve patient access to care and increase attention on patient concerns during encounters.
However, the research surfaced an important safety concern: physicians reported that AI-generated notes occasionally contained clinically significant inaccuracies, most commonly omissions of discussed information or pronoun errors. This underscores the necessity of verification workflows before notes are finalized in the medical record.