EUVIMEDScopri EUVIMED
Randomisierte kontrollierte StudieOpen access

Effects of a Generative AI Virtual Patient on Serious Illness Communication Skills: Randomized Controlled Trial

Kurtis G Haut, Masum Hasan, Thomas Carroll, Ronald M Epstein, Taylan Sen, Ehsan Hoque

JMIR Medical Education · 2026-10-20 · 12 · e93034-e93034

DOI: 10.2196/93034 ↗

At a glance

Randomisierte kontrollierte Studie in JMIR Medical Education, published online on 2026-10-20. Written by Kurtis G Haut and 5 co-authors. Institution: Department of Computer Science, University of Rochester, 3013 Wegmans Hall, Rochester, NY, 14627, United States, +1 (585) 275-1351. Major MEDLINE topics: Communication, Patient Simulation. 47 references. The full text is freely available in PubMed Central.

AbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

Background

Serious illness communication (SIC) is a critical skill in medical education, especially in end-of-life care, yet clinicians often lack sufficient opportunities for deliberate practice. Standardized patients (SPs) remain the gold standard for SIC training, but they are limited by cost, availability, and consistency. To support wider access to SIC training, scalable alternatives are needed.

Objective

The aim of this study is to evaluate whether SOPHIE (Standardized Online Patient for Healthcare Interaction Education), an AI-powered virtual patient that provides real-time dialogue and automated feedback, improves SIC skills among clinicians and trainees.

Methods

We conducted a single-blind, parallel-group, randomized controlled trial from June to December 2024 via videoconferencing. Participants were randomized in a 1:1 ratio to the intervention (SOPHIE training) or control (reading module) group. Blinded SP actors and 4 independent raters evaluated outcomes preintervention and postintervention. The intervention group completed 3 interactive SOPHIE modules (Empathize, Be Explicit, and Empower) for 30 minutes, combining dialogue with an interactive virtual patient and automated feedback. Control participants reviewed reading materials covering the same framework for the same duration. The primary outcome was the change in communication skills (Empathize, Be Explicit, and Empower) assessed by SPs and blinded raters using a rubric based on an established clinical communication training framework. Secondary outcomes included self-reported confidence in communication skills.

Results

Fifty-one medical, nursing, and physician assistant students and practicing clinicians (mean age 30.6, SD 13.2 years; women: 40/51, 78.4%) were recruited from academic and clinical settings. Participants were stratified by training level and randomized; all completed the study. Compared with controls, SOPHIE participants demonstrated significantly greater improvement in Empower (β=.11, 95% CI .04-.17; P =.002), Be Explicit (β=.08, 95% CI .03-.13; P =.002), Empathize (β=.07, 95% CI .01-.13; P =.03), and overall communication score (β=.09, 95% CI .04-.14; P =.001). Here, β represents the additional pre-to-post improvement in the SOPHIE arm relative to the control arm, estimated from linear mixed effects models. SOPHIE participants also reported higher confidence in their communication skills than controls (mean score of 4.31 vs 3.83 on a 5-point Likert scale; P =.009).

Conclusions

Training with an AI-powered virtual patient significantly improved SIC skills and confidence compared with reading modules. Generative AI–based simulations may provide a scalable, accessible, and effective adjunct or alternative to traditional SP training in clinical education.

Subjects

MeSH terms (MEDLINE)

CommunicationPatient SimulationHumansFemaleAdultSingle-Blind MethodMaleGenerative Artificial IntelligencePhysician-Patient RelationsMiddle AgedClinical Competence

Author keywords

AIend-of-life caregenerative AIlarge language modelsmedical educationpalliative carerandomized controlled trialserious illness communicationsimulation-based trainingvirtual standardized patients

Authors and institutions

  • Department of Computer Science, University of Rochester, 3013 Wegmans Hall, Rochester, NY, 14627, United States, +1 (585) 275-1351.
  • Department of Medicine, Divisions of General Medicine and Palliative Care, University of Rochester Medical Center, Rochester, NY, United States.
  • Departments of Family Medicine and Medicine, University of Rochester Medical Center, Rochester, NY, United States.
  • Computer and Information Sciences, Niagara University, Niagara Falls, NY, United States.

Cite

APA

Haut, K. G., Hasan, M., Carroll, T., Epstein, R. M., Sen, T., & Hoque, E. (2026). Effects of a Generative AI Virtual Patient on Serious Illness Communication Skills: Randomized Controlled Trial. JMIR Medical Education, 12, e93034-e93034. https://doi.org/10.2196/93034

Vancouver

Haut KG, Hasan M, Carroll T, Epstein RM, Sen T, Hoque E. Effects of a Generative AI Virtual Patient on Serious Illness Communication Skills: Randomized Controlled Trial. JMIR Medical Education. 2026;12:e93034-e93034. doi:10.2196/93034

BibTeX
@article{Haut2026,
  author = {Haut, Kurtis G and Hasan, Masum and Carroll, Thomas and Epstein, Ronald M and Sen, Taylan and Hoque, Ehsan},
  title = {Effects of a Generative AI Virtual Patient on Serious Illness Communication Skills: Randomized Controlled Trial},
  journal = {JMIR Medical Education},
  year = {2026},
  volume = {12},
  pages = {e93034--e93034},
  doi = {10.2196/93034}
}
Start for free →