Enhancing Post-Colonoscopy Patient Follow-Up Through Chatbots: A Quantitative Analysis

Aims In this study, we evaluate the effectiveness of chatbot technology, specifically ChatGPT 4.0 and Google GEMINI, in providing post-colonoscopy follow-up information to patients. Utilizing a structured questionnaire rated by independent reviewers on a 1-10 Likert scale, we explored three main metrics: accuracy, reliability, and user interface to compare the performance of each chatbot in handling patient inquiries about polyps and post-procedure care. This research aims to ascertain if these AI tools can reliably supplement traditional patient follow-up methods.

Methods The study evaluated if patients could trust ChatGPT and Google GEMINI for post-colonoscopy queries using a questionnaire rated by five reviewers on a 1-10 scale. We compared the chatbots on accuracy, reliability, and user interface based on patient questions about polyps and follow-up care. Data analysis involved descriptive statistics and t-tests to check significant differences between the chatbots, with p-values determining statistical significance.

Results Comparative analysis of ChatGPT 4.0 and Google Gemini across accuracy, reliability, and user interface showed ChatGPT 4.0 with higher scores: accuracy 7.99±0.71, reliability 7.76±0.63, user interface 7.80±0.67. Google Gemini scored lower: accuracy 3.67±1.42, reliability 3.90±1.17, user interface 3.51±0.69. T-tests confirmed ChatGPT 4.0 significantly outperforms Google Gemini, suggesting its better suitability for patient queries post-colonoscopy

Conclusions The study shows ChatGPT 4.0 outperforms Google GEMINI in accuracy, reliability, and usability for post-colonoscopy care. Notably better at addressing queries about polyps and follow-ups, ChatGPT 4.0 supports integrating AI chatbots to enhance traditional patient care methods, ensuring timely, accurate information. Further research is needed on the long-term effects of chatbot-assisted care on clinical outcomes and patient satisfaction