Debunking Palliative Care Myths: Assessing the Performance of Artificial Intelligence Chatbots (ChatGPT vs. Google Gemini)
Prakash Gondode, Vaishali Mahor, Deepa Rani, Rupavath Ramkumar, Pooja Yadav
Abstract
Palliative care plays a crucial role in comprehensive healthcare, yet misconceptions among patients and caregivers hinder access to services. Artificial Intelligence (AI) chatbots offer potential solutions for debunking myths and providing accurate information. This study aims to evaluate the effectiveness of AI chatbots, ChatGPT and Google Gemini, in debunking palliative care myths. Thirty statements reflecting common palliative care misconceptions were compiled. ChatGPT and Google Gemini generated responses to each statement, which were evaluated by a palliative care expert for accuracy. Sensitivity, positive predictive value, accuracy, and precision were calculated to assess chatbot performance. ChatGPT accurately classified 28 out of 30 statements, achieving a true-positive rate of 93.3% and a true-negative rate of 3.3%. Google Gemini achieved perfect accuracy, correctly classifying all 30 statements. Statistical tests showed no significant difference between chatbots' classifications. Both ChatGPT and Google Gemini demonstrated high accuracy in debunking palliative care myths. These findings suggest that AI chatbots have the potential to effectively dispel misconceptions and improve patient education and awareness in palliative care.