Simulated Patient and AI-based Roleplay for History-taking

NCT06766383 | Completed DMC

• 2 other identifiers: GPT for history-taking Trainin2023jyxm1150
• Study Type: interventional
• Target: 56
• Locations: 1 country
• Sites: 1

Brief Summary

The goal of this clinical trial is to evaluate whether AI-based simulated patient training can improve clinical reasoning and history-taking skills in medical students. The main questions it aims to answer is: Does GPT-based simulated patient training improve medical students' history-taking skills compared to traditional role-playing methods? Participants will: Participants in the intervention group perform medical history-taking conversations with an AI-simulated patient. Receive AI-generated structured feedback on their performance. The control group participated in role-playing exercises with instructors who acted as patients, receiving feedback after each session. Complete standardized assessments to evaluate clinical reasoning and decision-making skills.

Conditions

History-taking Training; GPT; Medical Education

Keywords

History-taking Training; Artificial Intelligence; Medical Education; GPT Simulation

Outcome Measures

Primary Outcomes (1)

• structured clinical examination to measure students' abilities in history taking

  Pre and Post Training: This primary endpoint measures improvements in students' clinical skills using a structured clinical examination conducted both before and after the training. The total score is out of 100 points, aggregated from four components: History Collection (30 points): Assesses the level of detail in the chief complaint and symptoms (10 points), the ability to understand and identify patient information (10 points), and the appropriateness and logic of follow-up questions (10 points). Clinical Reasoning (30 points): Evaluates the thoroughness of diagnostic thinking (15 points) and efficiency in processing information and forming clinical judgments (15 points). Communication Skills (20 points): Includes interaction with patients (10 points) and clarity of information delivery (10 points). Professional Behavior (20 points): Measures adherence to clinical procedural norms (10 points) and professional attitude towards patients (1

  From enrollment to the end of treatment at 4 weeks

Secondary Outcomes (2)

• Training Feedback Assessment

  From enrollment to the end of treatment at 4 weeks

• Student Satisfaction Survey

  From enrollment to the end of treatment at 4 weeks

Study Arms (2)

intervention group

EXPERIMENTAL

using GPT-simulated patients

Behavioral: GPT-simulated patients

control group

ACTIVE COMPARATOR

using traditional role-playing

Behavioral: traditional role-playing

Interventions

GPT-simulated patients BEHAVIORAL

GPT-simulated patients

intervention group

traditional role-playing BEHAVIORAL

traditional role-playing

control group

Eligibility Criteria

• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)

You may qualify if:

• Students enrolled in a medical or health-related professional program.
• Clinical coursework completion:
• Students who have completed basic clinical coursework, including foundational topics in patient communication and clinical reasoning.
• Age group:
• Participants aged 18-30 years to reflect typical medical student demographics.
• Language proficiency:
• Adequate proficiency in the language used for the study (e.g., English) to effectively interact with the AI-simulated patient and understand feedback.
• Willingness and availability:
• Students who provide informed consent and are willing to participate in all required study activities, including training sessions and assessments.
• Access to technology:
• Ability to access and use the necessary technology, such as a computer and internet connection, to complete the simulated patient interactions.

You may not qualify if:

• Students not enrolled in medical or related health professions programs. Students in early years of study who have not completed basic clinical coursework.
• Technical limitations:
• Inability to access or use the required technology (e.g., computer or online platforms).
• Time constraints:
• Inability to complete all required training and assessments within the study timeline.
• Language barriers:
• Insufficient proficiency in the language used for the study (e.g., English) to effectively communicate or understand the simulated patient interactions.
• Health-related factors:
• Conditions that may impair participation, such as significant hearing or vision impairment, or severe psychological conditions.
• Prior participation in similar studies:
• Students who have already participated in studies involving similar AI-based simulated patient training or feedback.

Sponsors & Collaborators

• zhen wang: lead

Study Sites (1)

The Second Affiliated Hospital of Anhui Medical University

Hefei, Anhui, 230601, China

Location

Related Publications (1)

• Wang Z, Fan TT, Li ML, Zhu NJ, Wang XC. Feasibility study of using GPT for history-taking training in medical education: a randomized clinical trial. BMC Med Educ. 2025 Jul 10;25(1):1030. doi: 10.1186/s12909-025-07614-9.

  PMID: 40640776 DERIVED

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: OUTCOMES ASSESSOR
• Purpose: OTHER
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR INVESTIGATOR
• PI Title: physician

Study Record Dates

• First Submitted: December 27, 2024
• First Posted: January 9, 2025
• Study Start: October 7, 2024
• Primary Completion: December 15, 2024
• Study Completion: December 15, 2024
• Last Updated: January 14, 2025

Record last verified: 2024-10

Locations

CN (1)