EUVIMEDPoznaj EUVIMED
Open access

Patient Attitudes Toward Ambient Voice Technology: Preimplementation Patient Survey in an Academic Medical Center

Gary Leiserowitz iD, Jeff Mansfield iD, Scott MacDonald iD, Melissa Jost iD

JMIR Medical Informatics · 2025-11-27 · 13 · e77901

DOI: 10.2196/77901 ↗

At a glance

Journal article in JMIR Medical Informatics, published online on 2025-11-27. Written by Gary Leiserowitz and 3 co-authors. Institution: University of California Davis Medical Center, Sacramento, CA, United States. Major MEDLINE topics: Electronic Health Records, Artificial Intelligence, Patients. Cited 25 times so far (Crossref); 23 references. The full text is freely available in PubMed Central.

AbstraktSubjectsDane bibliograficzneCiteRelated studies

Abstract (original language)

Background

Many institutions are in various stages of deploying an artificial intelligence (AI) scribe system for clinic electronic health record (EHR) documentation. In anticipation of the University of California, Davis Health’s deployment of an AI scribe program, we surveyed current patients about their perceptions of this technology to inform a patient-centered implementation.

Objective

We assessed patient perceptions about current clinician EHR documentation practices before implementation of the AI scribe program, and preconceptions regarding the AI scribe’s introduction.

Methods

We conducted a descriptive preimplementation survey as a quality improvement study. A convenience sample of 9171 patients (aged ≥18 years) who had a clinic visit within the previous year, was recruited via an email postvisit survey. Patient-identified demographics (age, gender, and race and ethnicity) were collected. The survey included rating scales on questions related to the patient perception of the AI scribe program, plus open-ended comments. Data were collated to analyze patient perceptions of including AI Scribe technology in a clinician visit.

Results

In total, 1893 patients completed the survey (20% response rate), with partial responses from another 549. Sixty-three percent (n=1205) of the respondents were female, and most were 51 years and older (87%, n=1649). Most patients identified themselves as White (69%, n=1312), multirace (8%, n=154), Latinx (7%, n=130), and Black (2%, n=42). The respondents were not representative of the overall clinic populations and skewed more toward being female, ages 50 years and older, and White in comparison. Patients reacted to the current EHR documentation system, with 71% (n=1349) feeling heard or sometimes heard, but 23% (n=416) expressed frustrations that their physician focused too much on typing into the computer. When asked about their anticipated response to the use of an AI scribe, 48% (n=904) were favorable, 33% (n=630) were neutral, and 19% (n=359) were unfavorable. Younger patients (ages 18-30 years) expressed more skepticism than those aged 51 years and older. Further, 42% (655/1567) of positive comments received indicated this technology could improve human interaction during their visits. Comments supported that the use of an AI scribe would enhance patient experience by allowing the clinician to focus on the patient. However, when asked about concerns regarding the AI scribe, 39% (515/1330) and 15% (203/1330) of comments expressed concerns about documentation accuracy and privacy, respectively. Providing previsit patient education and obtaining permission were viewed as very important.

Conclusions

This patient survey showed that respondents are generally open to the use of an AI scribe program for EHR documentation to allow the clinician to focus on the patient during the actual encounter rather than the computer. Providing patient education and obtaining consent before using AI are important components to gain patient trust. Caution about the results is appropriate, given the low response rate and nonrepresentative profile.

Subjects

MeSH terms (MEDLINE)

Electronic Health RecordsArtificial IntelligencePatientsHumansFemaleMaleMiddle AgedAdultAcademic Medical CentersSurveys and QuestionnairesAgedAdolescentYoung Adult

Author keywords

AIAI scribeEHRambient voice technologyartificial intelligenceelectronic health recordshealth information privacypatient perceptions and attitudespreimplementation patient surveyscribe program

Authors and institutions

  • University of California Davis Medical Center, Sacramento, CA, United States.

ORCID profiles: Gary Leiserowitz, Jeff Mansfield, Scott MacDonald, Melissa Jost

Cite

APA

Leiserowitz, G., Mansfield, J., MacDonald, S., & Jost, M. (2025). Patient Attitudes Toward Ambient Voice Technology: Preimplementation Patient Survey in an Academic Medical Center. JMIR Medical Informatics, 13, e77901. https://doi.org/10.2196/77901

Vancouver

Leiserowitz G, Mansfield J, MacDonald S, Jost M. Patient Attitudes Toward Ambient Voice Technology: Preimplementation Patient Survey in an Academic Medical Center. JMIR Medical Informatics. 2025;13:e77901. doi:10.2196/77901

BibTeX
@article{Leiserowitz2025,
  author = {Leiserowitz, Gary and Mansfield, Jeff and MacDonald, Scott and Jost, Melissa},
  title = {Patient Attitudes Toward Ambient Voice Technology: Preimplementation Patient Survey in an Academic Medical Center},
  journal = {JMIR Medical Informatics},
  year = {2025},
  volume = {13},
  pages = {e77901},
  doi = {10.2196/77901}
}
Start for free →