MEDICAL AI with ChatGPT

The field of medicine is constantly evolving, and the latest development involves artificial intelligence (AI) assistants like ChatGPT. A recent study published in JAMA Internal Medicine led by Dr. John W. Ayers from the Qualcomm Institute within the University of California, San Diego (UCSD) reveals the potential of AI assistants in the healthcare industry. The study compared written responses from physicians and those generated by ChatGPT to real-world health questions, and the results were eye-opening.

The study found that ChatGPT's responses were preferred by a panel of licensed healthcare professionals 79% of the time, and were rated as higher quality and more empathetic. The opportunities for improving healthcare with AI are immense, says Ayers, who is also the vice chief of innovation in the UCSD School of Medicine Division of Infectious Disease and Global Public Health. According to Ayers, AI-augmented care is the future of medicine.

However, the study also notes that directly answering patient questions accurately and empathetically is a different ballgame. While ChatGPT might be able to pass a medical licensing exam, it needs to be able to respond accurately to questions patients send to their doctors to be integrated into health systems and improve physician responses to patient queries.

To test ChatGPT's capabilities, the research team turned to social media to obtain a large and diverse sample of healthcare questions and physician answers that did not contain identifiable personal information. They randomly sampled 195 exchanges from Reddit's AskDocs where a verified physician responded to a public question, and provided the original question to ChatGPT to author a response. A panel of three licensed healthcare professionals assessed each question and the corresponding responses and compared responses based on information quality and empathy, noting which one they preferred.

The panel of healthcare professional evaluators preferred ChatGPT responses to physician responses 79% of the time. ChatGPT's responses were rated significantly higher in quality than physician responses and were more empathic. The study suggests that tools like ChatGPT can efficiently draft high quality, personalized medical advice for review by clinicians.

The study highlights that AI assistants like ChatGPT can potentially solve real-world healthcare delivery problems, and could impact patient health and physician performance. However, it is crucial to integrate AI assistants into healthcare messaging in the context of a randomized controlled trial to assess the impact of using AI assistants on outcomes for both physicians and patients.

Overall, the study suggests that a physician harnessing ChatGPT is the answer for better and empathetic care, and AI assistants can revolutionize the healthcare industry.