A Feasibility Study to Assess Virtual Patients Authenticity on Building Confidence in Students’ Clinical Decision-Making

Abstract

There is limited access to real patients during healthcare training, presenting a critical need for realistic, scalable simulations to develop students’ clinical reasoning skills. This study investigates how LLMs-supported virtual patient interactions impact healthcare students’ confidence in clinical reasoning and decision-making. We developed a web-based framework featuring a photorealistic patient named Laura, guided by a large language model dialogue. After the interaction, students completed surveys assessing their confidence, perceived authenticity, and cognitive engagement. 

Our research project has two primary aims: (1) How confident do students feel in making clinical decisions after interacting with an LLM-supported virtual patient?, and (2) How do user engagement (dialogue turns) and perceptions of the simulation relate to confidence gain in real patient interactions? Preliminary results reveal that the authenticity of the interaction significantly predicted students’ reported confidence gain. These findings suggest that virtual patient simulations can positively impact the users’ confidence– offering insight into the future integration of virtual educational tools in healthcare education.

Accessibility Summary:

In accordance with Title II regulations this content meets all points of exemption as Archived web content and/or Preexisting conventional electronic documents.