Abstract
Chat Generative Pre-trained Transformer (ChatGPT) has emerged as a new technology for physicians and patients to obtain medical information. Our aim was to assess the ability of ChatGPT 4.0 to deliver high-quality information in response to commonly asked questions and management recommendations for Barrett's esophagus (BE). Twenty-nine questions (14 clinical vignettes and 15 frequently asked questions (FAQ)) on BE were entered into ChatGPT 4.0. Using a 5-point Likert scale, three gastroenterologists with expertise in BE rated the 29 ChatGPT responses for accuracy, completeness, empathy, use of excessive medical jargon, and appropriateness to send to patients. Three separate gastroenterologists generated responses to the same 15 FAQs on BE. A group of blinded patients with BE evaluated both ChatGPT and gastroenterologist responses on quality, clarity, empathy and which of the two responses was preferred. Gastroenterologists rated ChatGPT responses as mostly accurate overall (4.01 out of 5) with 79.3% of responses completely accurate or mostly accurate with minor errors. When compared to gastroenterologist responses, the patient panel rated ChatGPT responses to be of significantly higher quality (4.42 vs. 3.07 out of 5) and empathy (4.33 vs. 2.55 out of 5) (p < 0.0001). In conclusion, ChatGPT 4.0 provides generally accurate and comprehensive information about BE. Patients expressed a clear preference for ChatGPT responses over those of gastroenterologists, finding responses from ChatGPT to be of higher quality and empathy. This study highlights the potential use of ChatGPT 4.0 as an adjunctive tool for physicians to provide real-time, high-quality information about BE to their patients.