Ambient Artificial Intelligence Versus Human Scribes in the Emergency Department - 20/04/26
, Derick Jones, MD, MBA, MHI, Laura Walker, MD, MBA, Rachel Lindor, MD, JD, John Schupbach, MD, MBA, Aidan Mullan, MA, Heather Heaton, MD, MSAbstract |
Study objective |
To compare the use of ambient artificial intelligence (AI) versus human scribes in the emergency department in terms of note quality and time spent in the electronic health record.
Methods |
A quality improvement pilot was performed with 5 early adopters from December 2024 to January 2025. Physicians were assigned to a human or AI scribe. Two physicians, blinded to the chart’s origin, scored notes using the Physician Documentation Quality Instrument (PDQI-9). We accessed our electronic health record for time metrics and note contributions and compared PDQI-9 scores, time metrics, and note contribution between groups.
Results |
There were 710 visits, 284 with human scribes (123 adult and 161 pediatric) and 426 with AI-assisted charting (271 adult, 155 pediatric). PDQI-9 scores were similar for adults, but AI scribes scored lower for pediatric patients (41.36 versus 42.25, adjusted risk ratio [aRR] = − 1.89 [95% confidence interval (CI) − 3.58 to − 0.20]). More time in the electronic health record notes section per patient was spent when using AI scribes (adult: 4.3 versus 1.8 minutes, aRR = 2.38 [95% CI 1.85 to 3.05]; pediatric: 3.5 versus 1.6 minutes, aRR = 2.21 [95% CI 1.94 to 2.51]). Note length was similar but physicians contributed significantly more characters per note when using AI (adult: 60.1% versus 30.8%, adjusted mean differences = 32.9 [95% CI 20.8 to 45.0]; pediatric: 62.3% versus 27.1%, adjusted mean differences = 35.2 [95% CI 29.7 to 40.7]).
Conclusion |
In comparison to human scribes, AI scribes were associated with more time spent in the electronic health record notes section, more physician note contribution, and similar to lower quality notes.
Le texte complet de cet article est disponible en PDF.Keywords : Ambient scribe, Artificial intelligence, Emergency medicine
Plan
| Please see page 562 for the Editor’s Capsule Summary of this article. |
|
| Supervising editor: David L. Schriger, MD, MPH. Specific detailed information about possible conflict of interest for individual editors is available at editors . |
|
| Author contributions : JM, DJ, LW, and HH formulated the study concept and design. JM, RL, JS, and HH contributed to data acquisition. AM performed statistical analysis. JM, DJ, and HH were primarily responsible for data interpretation and drafting of the manuscript. All authors provided manuscript edits and approved of the manuscript. JM takes responsibility for the paper as a whole. |
|
| All authors attest to meeting the four ICMJE.org authorship criteria: (1) Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND (2) Drafting the work or revising it critically for important intellectual content; AND (3) Final approval of the version to be published; AND (4) Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. |
|
| Data sharing statement: Data is available upon request to Dr. Morey at morey.jacob@mayo.edu . |
|
| Funding and support: By Annals ' policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org/ ). |
|
| Presentation information: This work was presented at American College of Emergency Physicians (ACEP) Research Forum 2025. |
|
| A podcast for this article is available at www.annemergmed.com . |
Vol 87 - N° 5
P. 561-568 - mai 2026 Retour au numéroBienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.
Déjà abonné à cette revue ?
