Conversational AI Study

Not Applicable

Not yet recruiting

• Conditions: Medical Education
• Interventions: Other: Conversational Artificial Intelligence Simulations

• Registration Number: NCT06459479
• Lead Sponsor: Stanford University

Brief Summary

This is a prospective, collaborative, mixed-methods study which includes anesthesiology trainees receiving one active intervention - a conversational artificial intelligence (CAI) simulation. The goal is to explore the capability of AI to provide high-fidelity simulations that can ultimately improve how healthcare professionals handle difficult conversations such as disclosing medical errors or mistakes.

Detailed Description

Not available

Study Timeline

• Study First Submitted: 2024-06-10
• Study First Submitted QC: 2024-06-10
• Study First Posted: 2024-06-14
• Status Verified: 2025-05
• Last Update Submitted: 2025-05-15
• Last Update Posted: 2025-05-16
• Study Start: 2025-08-01
• Primary Completion: 2026-07-31
• Study Completion: 2026-07-31

Recruitment & Eligibility

• Status: NOT_YET_RECRUITING
• Sex: All
• Target Recruitment: 20

Inclusion Criteria

• Hospital providers, including nurses, physician assistants, physicians, physician trainees, and other healthcare workers at Lucile Packard Children's Hospital (LPCH) and affiliated facilities who communicate with patients daily will be included

Exclusion Criteria

• Participants with reported severe motion sickness
• Nausea
• Seizure disorder
• Currently using chronotropic heart medications, such as β blockers

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: SINGLE_GROUP

Arm && Interventions

Group	Intervention	Description
AI Group	Conversational Artificial Intelligence Simulations	Participant will be using OpenAI's ChatGPT4o's application. For the purposes of this medical error scenario, the CAI will have a standardized prompt to simulate an upset parent whose child recently had surgery in which there was a medical error involving a tooth injury from an intubation. Following a scripted briefing, simulation will be initiated.

Primary Outcome Measures

Name	Time	Method
Evaluate acceptance of Conversational Artificial Intelligence Simulations using Guided Discussion Interview Question Guide	Before simulation	Guided Discussion Interview Question Guide contains 6 questions focusing 3 areas including Attitudes and Opinions, Perceptions and Summation question. All the questions are open-ended questions.

Secondary Outcome Measures

Name	Time	Method
Evaluation of Conversational Artificial Intelligence Simulations usability via the System Usability Scale (SUS)	Immediately after simulation	Measured with the System Usability Scale (SUS). The scale has 10 items. Scores ranges from 1-5 (1 = strongly disagree and 5 = strongly agree)
Evaluation of Conversational Artificial Intelligence Simulations usability via the User Experience Questionnaire (UEQ-S).	Immediately after simulation	UEQ-S, includes eight items (four from the pragmatic scales Efficiency, Perspicuity, Dependability, and four from the hedonic scales Stimulation and Novelty). Scores range from -3 (min) to 3 (max)
Evaluation of the reliability of AI quantitative assessment of non-technical performance compared to expert human assessors	immediately after simulation	The Anaesthetist Non-Technical Skills (ANTS) tool previously published by Fletcher et al. translated in French. The ANTS scoring system uses four categories assessing task management, teamworking, situation awareness and decision-making (1 to 4 points by categories). The minimum score is 4 and the maximum 16