Peer reviewed AI-assisted

Longitudinal study finds DAX Copilot did not make primary care clinicians more efficient at Atrium Health

Comparing 112 primary care clinicians given Nuance's DAX Copilot ambient documentation with 103 similar clinicians without it over 180 days, Atrium Health found no significant difference in electronic health record time or financial measures, with small reductions in documentation time only in exploratory subgroups of heavy users, low-volume clinicians and family medicine.

Model
DAX Copilot (Nuance, Microsoft), since renamed Dragon Copilot
Field
Medicine
Date
2024-11-22
Human collaborators
Tsai-Ling Liu, Jeffrey A. Cleveland

What was found

Atrium Health was the first health system to use DAX Copilot. Clinicians using it were recruited between June and August 2023 and compared with clinicians from similar practices who did not, and electronic health record use and financial outcomes were modelled with linear mixed models over 180 days, controlling for length of use, age, gender, provider type, years in practice and baseline. The primary analyses were not significant. Exploratory analyses found small decreases in documentation hours for high users, who transferred at least 60% of DAX notes (means ratio 0.93, 95% CI 0.88-0.98), for low-volume clinicians (0.91, 0.83-0.99) and for family medicine (0.91, 0.85-0.98). The study was funded by Wake Forest University Health Sciences and registered as NCT06329427.

Novelty check

This registry already carries two later ambient-scribe studies, 2025-11-26-ambient-ai-scribes-rct, a randomised trial of DAX and Nabla against usual care, and 2026-10-27-heidi-ai-scribe-health-system; neither cites this one, which is a separate, non-randomised design. Europe PMC, 2026-11-16, ambient with DAX or Dragon and documentation or scribe returned 253 records, mostly later surveys and reviews. It reached this registry through Becker's Hospital Review coverage in November 2026, two years after publication, and is entered under its 2024 publication date.

Caveats and known objections

Not randomised: the comparison group came from similar practices, and clinicians who adopt a new tool may differ from those who do not in ways the covariates miss. The positive signals are exploratory subgroup results, and the high-user subgroup is defined by how much clinicians used the tool, so it partly selects for those it suited. The outcomes are time and money, not documentation quality, patient experience or burnout. The product has been updated and renamed since the 2023 study period. Autonomy is ai-assisted: the software drafts notes that clinicians review and transfer.

Trial registration

ClinicalTrials.govNCT06329427

That a study was registered, with the outcomes it declared and the date it declared them, so a published endpoint can be checked against the one planned. It does not certify the results, their analysis, or any clinical benefit; registry records are submitted by the sponsor and are not peer reviewed.

No independent check is recorded on this entry yet.

Reading the primary source closely enough to say whether it supports the claim counts as a check, and you are credited on the entry.

Or on GitHub: submit a check challenge the grade send a correction or send a pull request

Entry history (1 event)
  1. AddedEntered the registry graded Peer reviewed and AI-assisted.

Entries are never deleted. A grade that does not hold up is downgraded on the record, with the reason beside it.

Graded peer reviewed for verification and ai-assisted for autonomy. What these mean.

Cite this entry

Plain text
whataifound.org. (2024). Longitudinal study finds DAX Copilot did not make primary care clinicians more efficient at Atrium Health. whataifound.org: A Registry of AI Scientific and Mathematical Discoveries. https://whataifound.org/finding/2024-11-22-dax-copilot-atrium-efficiency
BibTeX
@misc{whataifound-atriumhealth-2024-efficiency,
  title        = {Longitudinal study finds DAX Copilot did not make primary care clinicians more efficient at Atrium Health},
  author       = {{whataifound.org}},
  year         = {2024},
  howpublished = {whataifound.org: A Registry of AI Scientific and Mathematical Discoveries},
  note         = {Result by Atrium Health. Verification: Peer reviewed. Autonomy: AI-assisted.},
  url          = {https://whataifound.org/finding/2024-11-22-dax-copilot-atrium-efficiency}
}

Related findings

← All medicine findings in the registry