The Role of Large Language Models in Medical Education: Applications and Implications

Abstract

Large language models (LLMs) such as ChatGPT have sparked extensive discourse within the medical education community, spurring both excitement and apprehension. Written from the perspective of medical students, this editorial offers insights gleaned through immersive interactions with ChatGPT, contextualized by ongoing research into the imminent role of LLMs in health care. Three distinct positive use cases for ChatGPT were identified: facilitating differential diagnosis brainstorming, providing interactive practice cases, and aiding in multiple-choice question review. These use cases can effectively help students learn foundational medical knowledge during the preclinical curriculum while reinforcing the learning of core Entrustable Professional Activities. Simultaneously, we highlight key limitations of LLMs in medical education, including their insufficient ability to teach the integration of contextual and external information, comprehend sensory and nonverbal cues, cultivate rapport and interpersonal interaction, and align with overarching medical education and patient care goals. Through interacting with LLMs to augment learning during medical school, students can gain an understanding of their strengths and weaknesses. This understanding will be pivotal as we navigate a health care landscape increasingly intertwined with LLMs and artificial intelligence.

Keywords: AI; ChatGPT; LLM; artificial intelligence in health care; autoethnography; large language models; medical education.

Figure 1

Example of using ChatGPT to help brainstorm differential diagnoses (left). Follow-up questions could include, for example, which physical exam maneuvers (right), laboratory studies, or diagnostic tests could be used to narrow the selection of each differential diagnosis.

Figure 2

Example of using ChatGPT to generate an interactive medical practice case.

Figure 3

Example of applying ChatGPT to past practice exams. In this case, the student is using a multiple-choice question from a previous midterm that they answered incorrectly. The answer key provided for the exam was insufficient at explaining the physiologic reasoning behind the correct answer.

Figure 4

Demonstration of a negative use case. This example dialogue illustrates a scenario where a user requests the single most probable diagnosis in an ambiguous clinical scenario, and ChatGPT responds with an assertive and convincing, yet likely incorrect, response.

Figure 5

Examples of how ChatGPT can be integrated into medical education: practicing differential diagnoses, streamlining the wide array of study resources to assist with devising a study plan, serving as a simulated patient or medical professor for interactive clinical cases, helping students review multiple-choice questions or generating new questions for additional practice, digesting lecture outlines and generating materials for flash cards, and organizing information into tables to help build scaffolding for students to connect new information to previous knowledge.

Figure 6

A few examples of how ChatGPT may be integrated into health care, derived from current news sources and research projects within the clinical informatics community.