JMIR Publications

ABSTRACT

Background:

The transition to clinical clerkships can be difficult for medical students as it requires synthesis and application of preclinical information into diagnostic and therapeutic decisions. ChatGPT–a Generative Language Model (GLM) with many medical applications due to its creativity, memory, and accuracy–can help students in this transition.

Objective:

This paper models ChatGPT 3.5’s ability to perform interactive clinical simulations, as well as show such a tool’s benefit to medical education.

Methods:

Simulation starting prompts were refined on ChatGPT 3.5 using Google Chrome. Starting prompts were selected based on assessment format, stepwise progression of simulation events and questions, free-response question type, responsiveness to user inputs, post-scenario feedback, medical accuracy of feedback.

Results:

Two starting prompts were chosen. Prompt 1 was developed through three test simulations and used successfully in two simulations. Prompt 2 was developed through 10 additional test simulations and used successfully in one simulation.

Conclusions:

These simulations let students practice novel parts of the clinical curriculum, such as forming independent diagnostic and therapeutic impressions over an entire patient encounter. Furthermore, the simulations adapt to user inputs in a way that replicates real life more accurately than premade question bank clinical vignettes. Finally, ChatGPT can create potentially unlimited free simulations, which increases access for lower socioeceonomic status (SES) medical students. However, no tool is perfect, and ChatGPT is no exception; there are concerns about simulation accuracy and replicability that need to be addressed to further optimize ChatGPT as an educational resource.