A Large Language Model–Powered Multiagent Framework Emulating Standardized Patients in Clinical Communication Skills Training: Development and Evaluation Study

Category	Detail
Condition
Key Mechanisms	Multiagent framework with specialized subagents for simulating standardized patients (SPs) in clinical communication training.
Target Population
Care Setting

Proficient clinical communication skills are essential for accurate diagnosis and effective physician-patient relationships.
Standardized patients (SPs) are effective for training but face challenges like resource demands and SP fatigue.
Recent advances in large language models (LLMs) enable the development of virtual patients (VPs) as scalable alternatives to SPs.
The study evaluates a multiagent framework designed to improve simulation fidelity and interaction performance.
Ethical approval was obtained, and the study adhered to ethical standards outlined in the Declaration of Helsinki.

Implement a multiagent framework with specialized subagents to enhance training.

Conduct comprehensive evaluations to assess performance compared to single-LLM approaches.

Current solutions may lack flexibility and robust evaluation metrics for role-playing fidelity.

Medical students and trainees in clinical settings.

Training using VPs can enhance clinical communication skills through realistic simulations.

Incorporate diverse training methods, including role-play and digital strategies.
Focus on enhancing interaction performance and scalability in training simulations.

Clinical Scorecard: A Multiagent Framework Utilizing Large Language Models to Simulate Standardized Patients for Training in Clinical Communication Skills: Development and Assessment Study