System-wide AI scribe deployment and clinician documentation time at a 14-hospital health system
Across 1,637,227 outpatient encounters at a 14-hospital Boston health system, enrolment in the Heidi ambient AI scribe was associated with 1.21 fewer minutes of active note time per encounter, more among heavy users, and longer notes, with no change in same-day or after-hours note closure or in work relative value units.
- Model
- Heidi (Heidi Health)
- Field
- Medicine
- Date
- 2026-10-27
- Human collaborators
- Sara E. Booth, Jennifer P. Stevens
What was found
After system-wide availability from January 2025, 902 outpatient clinicians enrolled, of whom 605 used the tool and 297 did not. The study used an interrupted time series indexed on each clinician's enrolment month and adjusted for patient demographics and visit complexity, over notes from 1 July 2024 to 31 January 2026, with a secondary comparison of non-users, low users and high users. Note-active time fell by 1.21 minutes (95% CI 0.68 to 1.74) on enrolment and by 2.50 minutes among high users, and high users' notes grew by about 171 characters relative to non-users.
Novelty check
Europe PMC, 2026-10-29: TITLE_ABS:"scribe" AND TITLE_ABS:"artificial intelligence" AND TITLE_ABS:"documentation" returned 111 records, many of them 2026 evaluations of ambient scribes in single specialties, pilots or simulations. Ambient scribe studies are not new; this one stands out for unrestricted system-wide deployment, its scale and its use of measured note-active time rather than surveys. Within this registry 2025-11-26-ambient-ai-scribes-rct records a randomised trial of two other ambient scribes, DAX Copilot and Nabla; this is an observational deployment study of a different product.
Caveats and known objections
A retrospective quality-improvement study, not a trial; the authors flag possible unmeasured confounding and that AI-scribed notes could not be told apart from clinician-written ones after enrolment. One co-author is affiliated with Heidi Health, and the paper reports no primary funding source. Single health system in Massachusetts. The endpoints are surrogates, time and note length, with no measure of documentation accuracy, safety or patient outcome, and longer notes among high users cut both ways. Autonomy is ai-assisted: the scribe drafted notes for clinicians.
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
Flag this for triage
Signals order the review queue and nothing else. They are never published, and they never move a grade: that takes a citation.
Entry history (1 event)
- 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
whataifound.org. (2026). System-wide AI scribe deployment and clinician documentation time at a 14-hospital health system. whataifound.org: A Registry of AI Scientific and Mathematical Discoveries. https://whataifound.org/finding/2026-10-27-heidi-ai-scribe-health-systemBibTeX
@misc{whataifound-bethisraeldeaconessmedicalcenter-2026-system,
title = {System-wide AI scribe deployment and clinician documentation time at a 14-hospital health system},
author = {{whataifound.org}},
year = {2026},
howpublished = {whataifound.org: A Registry of AI Scientific and Mathematical Discoveries},
note = {Result by Beth Israel Deaconess Medical Center / Beth Israel Lahey Health. Verification: Peer reviewed. Autonomy: AI-assisted.},
url = {https://whataifound.org/finding/2026-10-27-heidi-ai-scribe-health-system}
}