Both Patients and Plastic Surgeons Prefer Artificial Intelligence-Generated Microsurgical Information.

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Microsurgery Education

Background:

• The increasing use of AI in patient education necessitates evaluating AI-generated health information.
• This study compares AI-generated content (ChatGPT) with professional organization content (ASRM) for microsurgery.

Purpose of the Study:

• To assess the accuracy, comprehensiveness, clarity, and readability of AI-generated microsurgery information versus professional society content.
• To determine preferences for AI-generated versus professionally curated microsurgery patient information among surgeons and lay individuals.

Main Methods:

• Six surgeons blindly evaluated microsurgery responses from ASRM and ChatGPT for accuracy, comprehensiveness, and clarity.
• Thirty lay individuals compared AI-generated and ASRM materials to determine preference.
• Readability was assessed using six standard formulas; statistical analysis included paired t-tests.

Main Results:

• ChatGPT responses were significantly more comprehensive and clearer than ASRM content.
• Surgeons (70.7%) and lay individuals (55.9%) preferred AI-generated materials.
• No inaccuracies were found in either ChatGPT or ASRM content.
• Both sources exceeded recommended readability levels, with AI content being more complex.

Conclusions:

• AI-generated patient-facing microsurgery materials are preferred by both surgeons and laypersons for their comprehensiveness and clarity.
• AI-generated content demonstrated no inaccuracies.
• Both AI and professional society materials present readability challenges, exceeding recommended levels for patient understanding.