Peer-reviewed research
Comparative evaluation of ambient digital scribe systems in clinical documentation
Weiss E et al., Journal of the American Medical Informatics Association, 2026 · DOI 10.1093/jamia/ocag157
Read the originalRead on October 16, 2026
- What it is
- Clinicians at Rochester Regional Health recorded two simulated visits and gave identical recordings to three commercial ambient scribe systems. Fifty-six physicians rated the notes, and reviewers checked each note against its recording and counted hallucinations, omissions and inaccuracies.
- How much weight it carries
- Peer-reviewed, published online ahead of print in September 2026. The authors call it exploratory: two simulated visits and three vendors cannot rank products reliably. They say larger samples are needed before the framework drives purchasing decisions.
Facts taken from this source
- In a Rochester Regional Health audit of three commercial scribes given identical simulated recordings, the vendor with the fewest errors had 3, against 8 and 9 for the other two.
“The audit identified fewest errors in Vendor B's notes (n = 3), followed by Vendor A (n = 8) and Vendor C (n = 9).” (Results)