Key points
- Emergency physicians in a community hospital ED spent 43% of their time on data entry and 28% in direct contact with patients.1
- Emergency physicians were interrupted 9.7 times an hour, against 3.9 for office-based primary care physicians, so the scribe has to hold each patient's draft through every interruption.2
- In 10,344 ED encounters, ambient scribe use was linked to 72.6 seconds less on-shift documentation per encounter, about 24 minutes across 20 encounters.4
- A structured handover programme cut medical errors by 23% across 10,740 admissions, so handover belongs in the template.6
- wai drafts the note during the encounter and flags any visit that ends without a note, for the clinician to review and sign.78
Summary
An AI scribe for emergency medicine drafts the note while the clinician is at the bedside and builds handover into it, through constant interruptions. In the largest single-centre ED analysis so far, ambient scribe use was linked to about 24 minutes less documentation across 20 encounters.4
An emergency or urgent care shift ends with every chart signed and the clinician walking out on time. Evidence from emergency departments shows a scribe moving documentation from after the shift back into it, while the clinician spends the encounter with the patient.
See it in your clinicThirty minutes on your own patients and doctors.
Request a demoSee an urgent care encounter drafted at the bedside and the chart ready for your signature before the patient leaves.
| Note element | What the scribe can draft from the encounter | What the clinician confirms before signing |
|---|---|---|
| Presenting complaint and triage | Complaint in the patient's words, onset and timing | Triage category and time seen |
| History | History of the complaint, medicines, allergies, relevant negatives | Red-flag questions asked and answered |
| Examination | Findings spoken aloud during the examination | Findings examined silently, and anything not examined |
| Investigations | Tests discussed and results read out | Every result reviewed, with the time |
| Reassessment | Change in symptoms and response to treatment | Time of reassessment and observations |
| Decision and disposition | Discharge, admission or referral as discussed | The clinical reasoning in the clinician's words |
| Discharge advice | Advice given and when to return | Safety-net advice the patient understood, and who they will see next |
| Handover block | Summary, outstanding actions and contingencies | Accepted by the receiving clinician, with name and time |
| Letter to the patient's own doctor | Drafted from the same encounter | Reviewed and signed |
Where the emergency physician's hour goes
In a US community hospital ED, physicians spent a mean 43% of their time on data entry and 28% in direct contact with patients, with mouse clicks approaching 4,000 in a busy 10-hour shift.1
A 2026 time and motion study of 20 emergency physicians in a high-volume urban US ED found a median 34.1% of shift minutes on the computer and 26.9% with patients.3 Record logs showed a further median 1.3 hours on the computer after the scheduled shift.3
Interruptions are the design constraint
Emergency physicians observed in five community hospitals were interrupted an average of 9.7 times an hour, against 3.9 for primary care physicians in their offices.2 They spent 37.5 minutes of each hour managing three or more patients at once.2
A scribe for that floor keeps each patient's draft with that patient and closes no encounter without a note. wai holds the audio, shows a plain warning and runs a watchdog check, so every visit ends with a note.8
What the ED evidence shows so far
A retrospective cohort of 10,344 encounters managed by 100 attending physicians at a US tertiary academic ED, from February 2025 to March 2026, linked ambient scribe use to 72.6 seconds less on-shift documentation per encounter.4 The authors put that at about 24 minutes per 8-hour shift across 20 encounters, with notes 690 characters shorter.4
A 2026 scoping review found six ED studies, with 6 to 24 emergency physicians each, reporting better cognitive workload, shorter time to first note and higher throughput.5 It listed fragmented workflow, environmental noise and note accuracy as the ED's particular challenges, and one study found charting slower with an AI scribe than with a human scribe.5 These are category findings, so test a scribe on your own floor.
Handover belongs in the note
In a study of a structured handover programme across nine hospitals and 10,740 admissions, the medical error rate fell by 23%, from 24.5 to 18.8 per 100 admissions, and preventable adverse events fell by 30%.6 Oral handover time did not change, at about 2.5 minutes per patient.6
The programme standardised written and oral handover with a mnemonic.6 An ED template carries the same discipline in a short handover block the scribe fills from the encounter and the clinician confirms before the shift changes.
The note holds up when it is read later
An emergency note is often read later by a complaints lead or a court. It should show the time of assessment, the reassessment, the decision and the advice given at discharge, in the clinician's own judgement and under their signature.
The clinician signs what the scribe drafts, and our guides on medico-legal risk and who signs an AI clinical note cover the detail.
How wai fits an urgent care shift
wai drafts the visit note during the consultation, structured the way the clinician writes, and prepares letters and suggested coding, all of which stay drafts until the clinician checks and signs them.7
For private urgent care, where the same clinician codes the visit and writes to the GP, those drafts arrive together from one encounter. The template guide shows how to shape an ED note the scribe writes into.
Where to start
Build one ED template with fixed fields for time of assessment, reassessment, disposition, discharge advice and a handover block. Run it for two weeks on a single stream, such as minor injuries or fast track, where encounters are short and frequent.
Measure the minutes from disposition to signed note and the share of charts finished before the shift ends, and compare against the two weeks before. Then widen to the rest of the department.
Bring a shift's worth of typical encounters. We show the drafts, the handover block and the signed chart on your own case mix.
Frequently asked questions
Does an AI scribe work in a noisy emergency department?
ED studies report noise as a challenge, along with fragmented workflow and note accuracy.5 Test the scribe in your own department before rolling it out, and review every draft before signing.
How much time does an AI scribe save in the ED?
In 10,344 encounters at one US academic ED, ambient scribe use was linked to 72.6 seconds less on-shift documentation per encounter, about 24 minutes across 20 encounters.4 Results vary by physician and case mix.
What happens when a clinician is interrupted mid-encounter?
wai holds the audio, shows a plain warning and runs a watchdog check, so the encounter still ends with a note to review.8
Can the scribe write the handover?
It can draft a handover block from the encounter for the clinician to confirm. Structured handover cut medical errors by 23% in a nine-hospital study.6
Is a scribe-drafted ED note medico-legally sound?
It is the clinician's note once reviewed and signed. wai notes, letters and suggested coding are drafts until a clinician checks and signs them.7
Is it suitable for private urgent care?
Yes. Short, frequent encounters are where documentation time adds up, and the note, the letter to the patient's own doctor and the suggested coding come from the same encounter.7
Sources
- 1American Journal of Emergency Medicine, Hill and colleagues, 20134000 clicks: a productivity analysis of electronic medical records in a community hospital EDRead 14 October 2026Read source
- 2Annals of Emergency Medicine, Chisholm and colleagues, 2001Work interrupted: a comparison of workplace interruptions in emergency departments and primary care officesRead 14 October 2026Read source
- 3JMIR Medical Informatics, Henreid and colleagues, 2026Quantifying and Visualizing Emergency Physician Workflow: Observational Time-Motion StudyRead 14 October 2026Read source
- 4JMIR AI, Preiksaitis and colleagues, 2026Ambient AI Scribes and Emergency Department Documentation Burden: Retrospective Cohort StudyRead 14 October 2026Read source
- 5Journal of Emergency Medicine, Gancz and colleagues, 2026Ambient Artificial Intelligence Scribes in the Emergency Department: A Scoping Review of Current EvidenceRead 14 October 2026Read source
- 6New England Journal of Medicine, Starmer and colleagues, 2014Changes in medical errors after implementation of a handoff programRead 14 October 2026Read source
- 7wai, 2026AI medical scribe for cliniciansRead 14 October 2026Read source
- 8wai, 2026wai changelogRead 14 October 2026Read source
Every chart signed on shift, and home when the shift ends.
Fifteen minutes on video with a clinician who uses wai daily, on encounters like your own. We reply within one working day.
Partner with us
See it on your own patients, then build it with us
Thirty minutes on your own patients, doctors and rules, in your market. We reply within one working day, and the clinics that partner with us early shape what we build next.
See what wai does between the visits