New Study: AI Voice Documentation Cuts Charting Time by Up to 93% in Nurse-Led Pilot
A new peer-reviewed study published in AJN (August 2026) offers one of the clearest looks yet at what happens when nurses lead the design and rollout of an AI-powered voice documentation tool — and the results are striking.
The setup. Researchers at a large academic medical center, Cedars-Sinai, ran a six-month pilot on a 48-bed medical-surgical unit, involving 90 RNs and 33 CNAs. Staff used a mobile app (Aiva Nurse Assistant) to dictate observations by voice; an AI model converted the dictation into structured documentation, which nurses then reviewed and approved before it synced to the EHR.
The Results
- Documentation timeliness improved across every measured task. Chlorhexidine bathing documentation dropped from 60 minutes to 4. Patient repositioning fell from 65 to 7 minutes. Urine output logging went from 42 to 6 minutes. Meal documentation shrank from 167 minutes to just 15.
- Patient experience rose across all six Press Ganey nursing domains, with gains of 7 to 41 percentile points. "Nurses kept you informed" jumped from the 57th to the 98th percentile.
- Staff satisfaction was strong: of 63 survey respondents, 41 said the tool reduced documentation time, and 42 said they enjoyed using it.
- Incidental overtime nearly halved, dropping from an average of 97 hours per month pre-pilot to 48.5 hours during the pilot.
- Adoption was near-universal. Usage grew from 14,231 entries in the first month to over 35,000 by month six, with all 123 staff members using the app by the pilot's end.
The authors credit much of the program's success to a structured change-management approach (Kotter's eight-step model) and, critically, to keeping nurses at the center of the tool's design from day one — not just as end users, but as co-creators.
Want the Full Study? Download the complete peer-reviewed article, including methodology, the Press Ganey breakdown, and lessons learned, below.
