Rethinking Pediatric Asthma Education Through Large Language Model Generation and Simplification: Randomized Double-Blind Study - Summary - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

Reevaluating Pediatric Asthma Education Using Large Language Models for Content Generation and Simplification: A Randomized Double-Blind Trial

  • By

  • Tianyi Xu

  • Yong Yin

  • Xi Zhang

  • Wei Wei

  • Jiajun Yuan

  • Hansong Wang

  • Guodong Ding

  • Wenjie Xue

  • Ziwei Chen

  • Sixin Xie

  • Huiqin Niu

  • Jie Xi

  • Shuzhu Lin

  • Xiaoli Tang

  • Liebin Zhao

  • August 26, 2026

Share

Objective:

To compare the quality, credibility, usefulness, and adoption intention of pediatric asthma education materials generated by expert authors and those generated by GPT-4o.

Approach:
  • Study Design: A multicenter, randomized, parallel-group, double-blind study conducted in Shanghai, assessing health education materials from published expert-authored responses, GPT-4o-generated responses, and GPT-4o-simplified responses.
Key Findings:
  • The study assessed three types of educational materials: published expert-authored responses, GPT-4o-generated responses, and GPT-4o-simplified responses.
  • Participants evaluated the materials based on quality, credibility, usefulness, and adoption intention.
Interpretation:

The study aims to provide insights into the effectiveness of LLM-generated materials in pediatric asthma education, focusing on user perspectives.

Limitations:
  • The study did not evaluate clinical, therapeutic, preventive, or behavioral outcomes.
  • The sample may not fully represent all healthcare settings.
Conclusion:

The findings will contribute to understanding the role of LLMs in generating effective health education materials for pediatric asthma.

Sources:

Original Source(s)

Related Content