Documentation load is a major driver of burnout in emergency medicine, and the ED’s pace only intensifies the pressure. Ambient AI scribes offer a hands‑free way to capture encounters and ease charting—but how often are these tools actually used, and do they meaningfully change documentation time? Join Dr. Kopel as she unpacks what early use reveals about the promise and limitations of AI‑generated documentation in emergency care.

Carl Preiksaitis, Al’ai Alvarez, Maia Winkel, Mia Karamatsu, Ian Brown, Neetha Sama, Luke Morris, Jae-yeon Lee, Allie Gubbels, Eileen Wahl, Anna Frye, Christian Rose. Ambient Artificial Intelligence Scribe Adoption and Documentation Time in the Emergency Department, Annals of Emergency Medicine, 2026, ISSN 0196-0644, https://doi.org/10.1016/j.annemergmed.2025.12.017.

Background:

Burnout among emergency physicians is at an all‑time high, and documentation demands are consistently identified as a major driver.

In the ED, the pressure is amplified by rapid patient turnover, constant interruptions, and the need to make time‑sensitive decisions—leaving little room to complete notes during clinical care.
Emergency physicians spend a significant portion of each shift entering data, writing notes, and managing EHR tasks, often extending this work beyond scheduled hours.
Ambient AI scribes—systems that listen to patient‑clinician interactions and generate draft documentation—have emerged as a potential solution to reduce this burden.
Early studies in outpatient and primary care settings suggest these tools may reduce after‑hours charting and improve perceived efficiency, but their performance and adoption in the ED remain largely unexplored.
Understanding how emergency physicians choose to use ambient AI—and how its use relates to objective metrics like documentation time and note characteristics—is essential for evaluating its real‑world impact.

Methods:

Study Design:

This retrospective observational study was conducted at a tertiary academic medical center between February to October 2025 (soon after offering the tool).

Electronic health record audit logs were analyzed from encounters where attending physicians could optionally use the ambient AI tool to generate clinical notes from patient-clinician conversations

Population:

Adult emergency department encounters managed by a single attending physician were eligible
Excluded visits with human scribes, resuscitation areas, and pediatric patient encounters

Exposure / Comparison:

Ambient AI scribe encounters vs standard documentation encounters without AI (such as typing or dictating notes).

Outcomes:

On-shit documentation time
Total EHR time
Note length

**Additional descriptive outcomes included adoption patterns by physician and AI use by ED care zone, patient acuity, and interpreter requirement.

Analysis:

Median time comparisons between ambient versus standard encounters
Descriptive analysis of utilization patterns

Results:

Sample size: of 8,740 eligible ED encounters, 976 (11.2%) used ambient AI.

Adoption patterns:

35 of 92 attendings (38%) used the tool at least once
Use was highly skewed however, with 9 physicians accounting for 70.5% of all ambient encounters
AI use was more common in
- Lower acuity zones: telemedicine, vertical care / chair-based ambulatory areas
- Encounters without interpreters

Documentation time:

On-shift documentation: 2:45 min with AI vs 3:50 min Standard
- Difference -1:05 min (28% reduction)
Total EHR time: 8:39 min AI vs 10:21 Standard
- Difference -16%

AI generated notes were also noted to be shorter overall compared with standard notes (median 9,233 characters vs 10,142 characters)

Conclusions:

Early implementation of ambient AI scribes in the ED showed overall low adoption but highly skewed usage among attending physicians.
When used, the technology was associated with reduced documentation time and shorter notes.
Physicians preferentially used the tool for lower acuity encounters without language barriers.

Limitations:

Limited generalizability: a single-center study that preferentially involved lower acuity encounters
Selection bias: physicians chose when to use AI
Early adoption phase: utilization patterns may change with increased familiarity and practice with the software
Lack of assessment of documentation quality or patient outcomes
Potential confounding variables from physician workflow differences. For example, a physician who was a higher AI-utilizer may have been a more efficient documenter prior to integration of the tool, which could skew results in favor of less time spent in the EHR when the tool is used.

Final Recommendations:

While this study is not practice-changing, it sheds light on a potential way to reduce documentation and EHR time for ED providers (and therefore potentially help lessen this known factor of burnout in providers?). Larger studies need to be conducted to determine whether this translates to clinically meaningful workflow implementations or reduced physician burnout, and should also evaluate what the implementation of AI scribe software means for documentation accuracy and quality as well as patient outcomes.

AUTHORSHIP

Written by: Nicole Kopel MD, PGY-3 University of Cincinnati Department of Emergency Medicine

Editing, Posting, and Audio Editing by Anita Goel, MD; Associate Professor, APD of UC EM Residency Program, and Co-editor of Tamingthesru.com

Cite as: Kopel, N., Goel, A. Real-Time Echo Intelligence: Ambient AI Scribe in the ED: Game-Changer or Just Hype? TamingtheSRU.com. www.tamingthesru.com/blog/journal-club/ambient-ai-scribe-for-ed-charting. 4/15/2026.

Blog