AI Chatbot Outperforms Human Physicians in Quality and Empathy of Patient Responses

The use of artificial intelligence in healthcare has been a hot topic in recent years, with many hoping that AI technology could help alleviate the burden on healthcare professionals and improve patient outcomes. Now, a new study has found that an AI chatbot assistant called ChatGPT may be able to provide high-quality and empathetic responses to patient questions, outperforming physicians in some cases.

The study, published in JAMA Internal Medicine, evaluated the ability of ChatGPT to provide quality and empathetic responses to patient questions in comparison to those written by physicians. The team of licensed healthcare professionals compared the responses to 195 randomly drawn patient questions from a social media forum, where a verified physician had responded to a public question. Chatbot responses were generated by entering the original question into a fresh session on December 22 and 23, 2022, without prior questions having been asked in the session. The original question, along with anonymized and randomly ordered physician and chatbot responses, were then evaluated in triplicate by the team of evaluators.

The study found that in 78.6% of evaluations, the evaluators preferred ChatGPT's responses to those of the physicians. Chatbot responses were also rated significantly higher for both quality and empathy than physician responses. In fact, the proportion of responses rated as good or very good quality and empathetic or very empathetic were 3.6 times and 9.8 times higher, respectively, for the chatbot than for physicians.

The rapid expansion of virtual healthcare has led to a surge in patient messages, leading to more work and burnout among healthcare professionals. With the results of this study, the use of AI assistants like ChatGPT may be able to help alleviate some of the burden on healthcare professionals by aiding in drafting responses to patient questions. This technology could be especially helpful in clinical settings, where chatbots could draft responses that physicians could then edit.

While the results of this study are promising, further exploration of this technology is warranted in clinical settings. Randomized trials could assess whether using AI assistants like ChatGPT might improve responses, lower clinician burnout, and ultimately improve patient outcomes.

In conclusion, the use of artificial intelligence in healthcare is an exciting prospect, and the results of this study suggest that AI assistants like ChatGPT may be able to aid in drafting high-quality and empathetic responses to patient questions. As this technology continues to develop, we may see a future where chatbots become an integral part of patient care, helping to improve outcomes for both patients and healthcare professionals.