Assessing ChatGPT's Reliability as an Educational Resource for Information on Breast Reduction Surgery

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Plastic Surgery

Background:

• Artificial intelligence (AI) and machine learning platforms like ChatGPT offer accessible health information.
• Concerns exist regarding the accuracy and reliability of AI-generated medical content.
• Breast reduction surgery (BRS) is a common procedure addressing physical and emotional burdens of macromastia.

Purpose of the Study:

• To evaluate the quality of ChatGPT's responses to common patient questions about breast reduction surgery.
• To assess the accuracy and thoroughness of AI-generated medical information for BRS.

Main Methods:

• A 15-question survey of typical patient inquiries regarding BRS was developed.
• ChatGPT's responses were analyzed by five board-certified plastic surgeons.
• Responses were categorized as appropriate, not thorough, or inaccurate.

Main Results:

• 92% of ChatGPT's responses regarding BRS were deemed accurate.
• 6% of responses lacked sufficient detail, and 2.7% were inaccurate.
• No statistical significance was found between accuracy, thoroughness, and inaccuracy categories.

Conclusions:

• ChatGPT can serve as a supplementary tool for basic patient understanding of BRS.
• AI-generated information should not substitute professional medical advice from a plastic surgeon.
• Patients can use AI to formulate more informed questions for surgical consultations.