ChatGPT and other large language models for childhood asthma

Area of Science:

• Pediatric Pulmonology
• Artificial Intelligence in Medicine
• Health Informatics

Background:

• Childhood asthma is a prevalent chronic condition with ongoing management challenges.
• Large language models (LLMs) like ChatGPT, Claude, and Gemini are increasingly accessible.
• The integration of LLMs into pediatric asthma care presents both opportunities and risks.

Purpose of the Study:

• To review the role of commercially available LLMs in childhood asthma management.
• To explore the fundamental principles, current evidence, and potential applications of LLMs in this context.
• To identify challenges and areas for future research in LLM utilization for pediatric asthma.

Main Methods:

• Narrative review of existing literature on LLMs and childhood asthma.
• Analysis of LLM capabilities in generating medically accurate and comprehensible responses.
• Examination of potential benefits for patients, families, and healthcare professionals.

Main Results:

• LLMs demonstrate capacity for generating medically accurate asthma-related information.
• Potential applications include patient education, information summarization for clinicians, and content tailoring.
• Identified risks include model 'hallucinations,' data bias, and privacy concerns.

Conclusions:

• LLMs show promise for enhancing childhood asthma care but require careful evaluation.
• Further research is crucial to assess safety, clinical utility, and real-world acceptance.
• Development of reliable, inclusive LLMs tailored for pediatric respiratory care is needed.