ChatGPT Provided Excellent or Satisfactory Answers to Questions about de Quervain’s Tenosynovitis, but still Requires Verification by Health Care Professionals

Objective. ChatGPT, an AI chatbot by OpenAI, is increasingly utilized in patient and resident education due to its ability to provide information across various medical topics. However, the accuracy and reliability of its responses, particularly for specialty-specific conditions, are still being evaluated. This study aims to assess ChatGPT’s accuracy in answering standardized questions on de Quervain’s tenosynovitis to determine its effectiveness in supporting medical education. Materials and methods. Frequently asked questions regarding de Quervain’s tenosynovitis were identified from medical websites, and ten questions were selected by consensus among the authors. On July 4, 2024, these questions were posed to ChatGPT (version 4), and the responses were evaluated for accuracy and quality using the ChatGPT Response Rating System, with consensus reached among authors in cases of differing opinions. Results. Based on the ChatGPT Response Rating System, 30% (n = 3) of the responses scored a 1 (excellent, requiring no clarification), 50% (n = 5) scored a 2 (satisfactory, needing minimal clarification), and 20% (n = 2) scored a 3 (satisfactory, requiring moderate clarification). Extraneous details were omitted to condense the responses into summary sentences focused on the main answer. Conclusions. ChatGPT provided excellent or satisfactory answers to questions about de Quervain’s tenosynovitis, indicating its potential as a supplementary tool in medical education for this condition. However, the accuracy of its information can vary, necessitating verification by medical experts, and while not fully ready for integration into medical education, ChatGPT is expected to improve with further refinement of its data inputs.