A recent study by Cedars-Sinai investigators explores the potential of ChatGPT, an artificial intelligence (AI) chatbot, to help patients with cirrhosis and liver cancer by offering easy-to-understand information on basic knowledge, lifestyle, and treatments. Published in the peer-reviewed journal Clinical and Molecular Hepatology, the study sheds light on the AI system’s possible role in clinical practice.
ChatGPT’s Role in Empowering Patients
Cirrhosis and liver cancer patients often have unmet needs and insufficient knowledge about managing and preventing complications. Their treatments can be complex and challenging to manage. According to Brennan Spiegel, MD, MSHS, director of Health Services Research at Cedars-Sinai and co-corresponding author of the study, ChatGPT has the potential to empower patients and improve health literacy for different populations, despite its limitations.
AI’s Potential in Patient Education
ChatGPT has already demonstrated its potential for medical professionals by writing basic medical reports and answering medical student examination questions. Alexander Kuo, MD, medical director of Liver Transplantation Medicine at Cedars-Sinai, notes that personalized education AI models could help increase patient knowledge and education.
Investigators tested ChatGPT’s accuracy by presenting it with 164 frequently asked questions in five categories. Two liver transplant specialists independently graded the answers provided by the AI model. The questions covered basic knowledge, diagnosis, treatment, lifestyle, and preventive medicine.
The study found that ChatGPT answered about 77% of the questions correctly, with high levels of accuracy in 91 questions across various categories. The specialists grading the responses indicated that 75% of the responses for basic knowledge, treatment, and lifestyle were comprehensive or correct but inadequate. The proportion of responses that were “mixed with correct and incorrect data” was:
- 22% for basic knowledge
- 33% for diagnosis
- 25% for treatment
- 18% for lifestyle
- 50% for preventive medicine
The AI model also offered practical advice for patients and caregivers regarding next steps after a new diagnosis. However, the study made it clear that advice from a physician remains superior.
Yee Hui Yeo, MD, first author of the study and a clinical fellow in the Karsh Division of Gastroenterology and Hepatology at Cedars-Sinai, said that ChatGPT struggled to provide tailored recommendations based on the inquirer’s region due to varied recommendations from different professional societies. Nonetheless, researchers remain hopeful that the AI model will become more accurate in addressing questions according to the inquirers’ locations.
While more research is needed to examine ChatGPT’s role in patient education, Brennan Spiegel believes it can be a valuable adjunctive tool for physicians. It is not meant to replace medical professionals, but to provide access to reliable and accurate health information that is easy to understand. The goal is to empower patients and improve health literacy for those facing challenging conditions such as cirrhosis and liver cancer.
The Cedars-Sinai team, including authors Jamil Samaan, Hirsh Trivedi, Aarshi Vipani, Walid Ayoub, Ju Dong Yang, and Omer Liran, is hopeful that their findings can contribute to the ongoing development and application of AI tools like ChatGPT in the medical field. The study serves as a starting point for further exploration into how AI can improve patient outcomes and complement the work of physicians in managing complex medical conditions.