Clinical Scorecard: Bridging the Divide: A Cross-Sectional Study on the Role of Large Language Models in Emergency Medicine Publications
At a Glance
Category
Detail
Condition
Use of large language models (LLMs) in emergency medicine academic publications
Key Mechanisms
LLMs assist with literature summarization, language polishing, and drafting manuscript sections but pose risks including factual inaccuracies, plagiarism, bias, and lack of accountability
Target Population
Authors, editors, and reviewers involved in emergency medicine research publications
Care Setting
Academic and scientific publishing within emergency medicine journals
Key Highlights
LLMs offer efficiency gains in academic writing but raise ethical and integrity concerns.
There is significant variability in LLM guidance adoption across medical specialties and publishers.
Emergency medicine journals have not been comprehensively assessed for LLM-related editorial policies prior to this study.
Guideline-Based Recommendations
Diagnosis
Identify presence or absence of explicit LLM use guidance in journal editorial policies.
Use keywords such as large language model, AI, ChatGPT, and generative AI to detect relevant directives.
Management
Require authors to disclose any use of LLMs in manuscript preparation.
Mandate human accountability for accuracy and integrity of AI-generated content.
Include directives for authorship, content generation, and image creation related to LLM use.
Monitoring & Follow-up
Journals should regularly update and publicly display LLM-related policies.
Editorial teams and reviewers should be aware of and enforce LLM guidance.
Risks
Potential for factually inaccurate or hallucinated content generated by LLMs.
Risk of plagiarism and perpetuation of biases from training data.
Lack of accountability for AI-generated text threatens scholarly integrity.
Patient & Prescribing Data
Not applicable (focus on academic publication processes in emergency medicine).
Not applicable.
Clinical Best Practices
Ensure transparency by disclosing any use of LLMs in manuscript preparation.
Maintain human oversight and responsibility for all AI-generated content.
Adopt and adhere to specialty-specific editorial policies on LLM use.
Publish clear and accessible guidelines for authors, editors, and reviewers regarding LLMs.
Regularly review and update policies to keep pace with technological advances.
