After 30 days with an AI scribe, fewer clinicians with burnout.
51.9%38.8%
After 30 days38.8%
Before51.9%
The cognitive load related to notes and after-hours documentation time also decreased.
View the study
Evidence
What the research says
Recent research results on AI in medicine. These studies concern other tools: their results do not measure AtomicMD.
Try it free
A question? info@atomicmd.ca
Physician workload
- 46%
of Canadian physicians report a high level of burnout.
CMA and Ipsos, 2026 - 49.8%
of American emergency physicians have at least one symptom of burnout: the most affected specialty.
AMA, 2026 - 34.1%
of an emergency physician’s shift is spent at the computer, compared with 26.9% with patients.
JMIR Medical Informatics, 2026
Clinical reasoning
What language models bring to physicians’ reasoning.
5 of 5experiments
On difficult clinical cases, a language model outperformed the physician baselines in each of the five experiments.
View the study+6.5points
On the management decision score, with the help of a language model, on cases drawn from real encounters.
View the study51.7%with a model’s help
With a model’s help51.7%
With a search engine44.4%
Unaided36.1%
The correct diagnosis among the top ten hypotheses.
View the study85%the model as first opinion
The model as first opinion85%
The model as second opinion82%
Conventional resources75%
Diagnostic accuracy when the clinician works with a language model.
View the study71.4%with a language model
With a language model71.4%
Conventional resources42.6%
Diagnostic reasoning score, without lengthening the time spent per case.
View the study79.4%for a specialized model
A specialized model79.4%
Emergency physicians65.4%
Correct diagnoses, under simulated conditions.
View the studyWhat AtomicMD does with it
A differential suggested from your dictation: what you named, the possibilities we think of less, and the ones you can’t afford to miss. Each comes with what supports it, what makes it less likely and what remains to check.
Suggested orders come with their reason. You choose what to keep, set aside or add.
Documentation
What AI scribes change in the time spent writing, and in burnout.
16minutes
Less documentation time per eight hours of scheduled appointments, and half a visit more per week.
View the study−28%in the emergency department
With an AI scribe2 min 45 s
Without3 min 50 s
On-shift documentation time, median.
View the study30.7%after 84 days
After 84 days30.7%
Before52.6%
Clinicians with burnout, before and after 84 days with an AI scribe.
View the study0.36hours
Less time on notes per day, less work exhaustion, and notes of the same quality.
View the study15,791hours
Of documentation saved in one year, in a large health system.
View the studyIn Canada, 21% of physicians already use AI tools, and 59% of them say these tools have already lightened their administrative load.
What AtomicMD does with it
You dictate; AtomicMD prepares the note, the orders, the forms and the billing, and you review before signing.
The physician decides
The World Health Organization, the Collège des médecins du Québec and the American Medical Association, which speaks of “augmented intelligence”, hold the same principle: AI assists, the physician decides.
What AtomicMD does with it
Nothing goes into the record without your action. Every suggestion is reviewed, corrected and signed, and each one carries its reason, so you can judge it.
All references
18
- Canadian Medical Association and Ipsos. “CMA 2025 National Physician Health Survey.” February 2026. ipsos.com
- American Medical Association. “AMA: Physician burnout rates are falling, specialty gaps remain.” April 16, 2026. ama-assn.org
- Henreid AJ et al. “Quantifying and Visualizing Emergency Physician Workflow: Observational Time-Motion Study.” JMIR Medical Informatics, 2026;14:e85983. doi:10.2196/85983
- Brodeur PG et al. “Performance of a large language model on the reasoning tasks of a physician.” Science, 2026;392(6797):524-527. doi:10.1126/science.adz4433
- McDuff D et al. “Towards accurate differential diagnosis with large language models.” Nature, 2025;642:451-457. doi:10.1038/s41586-025-08869-4
- Everett SS et al. “From tool to teammate in a randomized controlled trial of clinician-AI collaborative workflows for diagnosis.” npj Digital Medicine, 2026;9(1):409. doi:10.1038/s41746-026-02545-1
- Goh E et al. “GPT-4 assistance for improvement of physician performance on patient care tasks: a randomized controlled trial.” Nature Medicine, 2025;31(4):1233-1238. doi:10.1038/s41591-024-03456-y
- Qazi IA et al. “Large language model diagnostic assistance for physicians in a lower-middle-income country: a randomized controlled trial.” Nature Health, 2026;1(2):198-205. doi:10.1038/s44360-025-00007-8
- Guo Y et al. “Development and prospective shadow evaluation of a domain-specific large language model for emergency neurological diagnosis.” npj Digital Medicine, 2026;9(1):470. doi:10.1038/s41746-026-02644-z
- Olson KD et al. “Use of Ambient AI Scribes to Reduce Administrative Burden and Professional Burnout.” JAMA Network Open, 2025;8(10):e2534976. doi:10.1001/jamanetworkopen.2025.34976
- Rotenstein LS et al. “Changes in Clinician Time Expenditure and Visit Quantity With Adoption of Artificial Intelligence-Powered Scribes: A Multisite Study.” JAMA, 2026;335(16):1408-1417. doi:10.1001/jama.2026.2253
- Preiksaitis C et al. “Ambient Artificial Intelligence Scribe Adoption and Documentation Time in the Emergency Department.” Annals of Emergency Medicine, 2026;87(5):569-574. doi:10.1016/j.annemergmed.2025.12.017
- You JG et al. “Ambient Documentation Technology in Clinician Experience of Documentation Burden and Burnout.” JAMA Network Open, 2025;8(8):e2528056. doi:10.1001/jamanetworkopen.2025.28056
- Afshar M et al. “A Pragmatic Randomized Controlled Trial of Ambient Artificial Intelligence to Improve Health Practitioner Well-Being.” NEJM AI, 2025;2(12). doi:10.1056/aioa2500945
- Tierney AA et al. “Ambient Artificial Intelligence Scribes: Learnings after 1 Year and over 2.5 Million Uses.” NEJM Catalyst, 2025;6(5). doi:10.1056/CAT.25.0040
- World Health Organization. “Ethics and governance of artificial intelligence for health: Guidance on large multi-modal models.” 2024. who.int
- Collège des médecins du Québec. “Scribe et intelligence artificielle.” cmq.org
- American Medical Association. “Augmented intelligence in medicine.” ama-assn.org
Also worth reading: Trust, your data and your patients.
Try it free
A question? info@atomicmd.ca