Clinical Report: Reevaluating Pediatric Asthma Education Using LLMs
Overview
This study evaluates the effectiveness of large language models (LLMs) in generating pediatric asthma education materials. It compares the quality, credibility, usefulness, and adoption intention of materials generated by LLMs against expert-authored content, focusing on participant assessments.
Background
Asthma is the leading chronic respiratory condition in children, necessitating effective education for caregivers to manage symptoms and treatment. Traditional educational resources often face barriers such as accessibility and comprehensibility.
Data Highlights
No numerical data or trial results were provided in the source material, which may limit the interpretation of the findings.
Key Findings
LLM-generated responses were assessed for quality, credibility, usefulness, and adoption intention based on participant evaluations.
Three types of materials were compared: published expert-authored responses, GPT-4o-generated responses, and GPT-4o-simplified responses.
The study utilized a 19-item scale based on the information adoption model for evaluation.
Participants included medical professionals and families of pediatric patients.
Results indicated varying perceptions of the three material types among participants without editorial interpretation.
Clinical Implications
The findings suggest that LLMs can be a viable alternative for generating educational materials, potentially improving caregiver understanding of pediatric asthma. Further exploration of LLM applications in health education may enhance patient engagement and health literacy.
Conclusion
This study presents findings on the use of LLMs in pediatric asthma education, highlighting participant evaluations of the materials.
