CoChat

Ambient artificial intelligence scribes: utilization and impact on documentation time

Open access

The DOI matches its Crossref record; 21 of the 22 references it cites were found in a registry or an index; 5 of the 13 sources its claims cite were checked for support.

Abstract

To quantify utilization and impact on documentation time of a large language model-powered ambient artificial intelligence (AI) scribe. This prospective quality improvement study was conducted at a large academic medical center with 45 physicians from 8 ambulatory disciplines over 3 months. Utilization and documentation times were derived from electronic health record (EHR) use measures. The ambient AI scribe was utilized in 9629 of 17 428 encounters (55.25%) with significant interuser heterogeneity. Compared to baseline, median time per note reduced significantly by 0.57 minutes. Median daily documentation, afterhours, and total EHR time also decreased significantly by 6.89, 5.17, and 19.95 minutes/day, respectively. An early pilot of an ambient AI scribe demonstrated robust utilization and reduced time spent on documentation and in the EHR. There was notable individual-level heterogeneity. Large language model-powered ambient AI scribes may reduce documentation burden. Further studies are needed to identify which users benefit most from current technology and how future iterations can support a broader audience.

Overview

CoChat Summary

TL;DR

This prospective quality improvement study evaluated a large language model-powered ambient AI scribe (DAX Copilot) among 45 physicians across 8 ambulatory specialties over 3 months. The scribe was used in 55.25% of encounters and was associated with significant reductions in median documentation time per note and daily EHR time, with notable interuser heterogeneity.

Key findings

  • DAX Copilot was utilized in 9629 of 17 428 encounters (55.25%); median physician-level utilization was 52.5% (IQR 17.86%–80.97%). HPI and A&P SmartSections had higher median utilization (39.7% and 14.8%) than Results and PE SmartSections (2.6% and 1.6%).
  • Median time per note decreased by 0.57 minutes (IQR −1.3 to −0.13; T-statistic −5.19, P<.001). Median user average time per note fell from 4.86 minutes prepilot (IQR 2.89–6.47) to 3.64 minutes during the pilot (IQR 2.45–5.46).
  • Median daily documentation time decreased by 6.89 minutes/day (IQR −22.37 to −0.65; T-statistic −4.48, P<.001). Median daily afterhours EHR time decreased by 5.17 minutes/day (IQR −21.32 to 3.82; T-statistic −2.65, P=.01), and median daily total EHR time decreased by 19.95 minutes/day (IQR −39.34 to −3.64; T-statistic −5.85, P<.001).
  • There was significant interuser heterogeneity: some physicians used DAX for almost all encounters, others had minimal utilization, and a small number spent more time per note compared to baseline.

Methods

  • Prospective quality improvement study at a single academic medical center (Stanford Health Care) with 45 physicians from 8 ambulatory disciplines. The pilot ran October 2023–January 2024, with a 2-month baseline prepilot period from May 28, 2023 to July 29, 2023.
  • DAX Copilot was integrated with Epic; the mobile app recorded physician-patient interactions, and an LLM generated drafts of HPI, PE, Results, and A&P SmartSections for insertion into existing note templates. Exclusions included existing scribe support, working exclusively with trainees, lacking an iPhone, non-English patient preferred language, and low-utility license relinquishment.
  • Utilization and documentation times were derived from Epic Signal EHR use measures. Average note time and daily documentation, afterhours, and total EHR time were weighted per physician and compared between prepilot and pilot periods using one-sample t-tests against a null of no change; analysis used Python 3.11.4 with statsmodels.

Limitations

  • The paper states: small sample size, volunteer and selection bias, potential secular trends, inability to analyze time outcomes at the level of individual patient encounters, lack of comparison to alternate strategies, English-speaking patients only, and predominance of primary care physicians.
  • The paper also states that EHR use measures have known limitations; afterhours and total EHR time were calculated per calendar day without normalization to encounter volume, potentially underestimating impact, and the 5-second inactivity timer likely underestimates total documentation time and the absolute magnitude of change.
  • Evident from the text: this was a single-center, non-randomized pilot with convenience and purposive sampling, a 50-license limit, rolling onboarding, and a short 3-month evaluation period, which limits causal inference and generalizability.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1093/jamia/ocae304

Cites

21 of 21 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

4 Valid · 0 Not valid · 1 Not checked · 32 Own results and interpretations

17 more claims

Create a free account to see all 37 claims and to verify the 2 references still open.

See all claims — free account
  • Probably not a substantial claim. 8
  • It may state two separate claims as one. 8

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

1 more claim about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 21 references

Cited by

See all 155

Metadata

Authors
Ma, S. P., Liang, A. S., Shah, S. J., Smith, M., Jeong, Y., Devon-Sand, A., Crowell, T., Delahaie, C., Hsia, C., Lin, S., Shanafelt, T., Pfeffer, M. A., Sharp, C., & Garcia, P.
Affiliations
  • Stanford Medicine (US)
  • Stanford Health Care (US)
  • Department of Medicine, Stanford University School of Medicine , Stanford, CA 94305,
Corresponding author
  • Ma, Stephen P
  • Liang, April S
Venue
Journal of the American Medical Informatics Association
Published
2024-12-17
Volume
32
Issue
2
Pages
381-385
ISSN
  • 1067-5027 (ISSN-L)
  • 1527-974X
OpenAlex
W4405499002
Cited by
155
References
22
Citations by year
2026: 99 · 2025: 56
FWCI
47.13
Type
journal article
Language
en
Open access
Locations
  • Journal of the American Medical Informatics Association (publishedVersion)
  • PubMed Central (publishedVersion)
  • Europe PMC (PubMed Central) (publishedVersion)
  • PubMed (publishedVersion)
Topics
  • Health Sciences › Health Professions › Health Information Management › Electronic Health Records Systems
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Health Sciences › Nursing › Research and Theory › Nursing Diagnosis and Documentation
Keywords
ambient AI scribes, large language models, electronic health records, documentation burden, physician time use, outpatient care
MeSH terms
Academic Medical Centers, Artificial Intelligence, Documentation, Humans, Prospective Studies, Time Factors, Electronic Health Records, Quality Improvement

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Ambient artificial intelligence scribes: utilization and impact on documentation time. https://app.cochat.ai/papers/10.1093/jamia/ocae304
The paper
Ma, S. P., Liang, A. S., Shah, S. J., Smith, M., Jeong, Y., Devon-Sand, A., Crowell, T., Delahaie, C., Hsia, C., Lin, S., Shanafelt, T., Pfeffer, M. A., Sharp, C., & Garcia, P. (2025). Ambient artificial intelligence scribes: utilization and impact on documentation time. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocae304

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.