ChatGPT and Other Large Language Models in Medical Education - Scoping Literature Review.

Area of Science:

• Medical Education
• Artificial Intelligence
• Natural Language Processing

Background:

• The rapid emergence of large language models (LLMs) presents new opportunities and challenges in medical education.
• Understanding the current landscape of research on LLMs in this field is crucial for guiding future development and implementation.

Purpose of the Study:

• To systematically review and summarize scientific publications on the use of LLMs in medical education within their first year of availability.
• To identify trends, common research areas, and limitations in the existing literature.

Main Methods:

• A scoping literature review was performed following PRISMA guidelines.
• Searches were conducted across five scientific literature databases using predefined terms.
• 145 studies were included from an initial pool of 1509 results.

Main Results:

• The majority of included studies focused on evaluating the capabilities of LLMs in passing medical examinations.
• Additional research explored the advantages, disadvantages, and potential applications of LLMs in medical training.
• A significant limitation identified was the scarcity of empirical research and the general lack of methodological rigor in published studies.

Conclusions:

• The current body of research on LLMs in medical education is nascent and predominantly descriptive.
• There is a critical need for more methodologically sound empirical studies to advance the field.
• A proposed research agenda aims to enhance the quality and impact of future investigations into LLMs in medical education.