Photo Credit: Thinkhubstudio
ChatGPT4’s answers to common questions from patients with systemic lupus erythematosus (SLE) were reliable and empathetic, according to a study published in Rheumatology. The researchers curated 95 frequently asked questions, 40 in English and 55 in Chinese, and ran them by ChatGPT4 and five rheumatologists. The responses were scored on a 0-10 scale, with higher scores reflecting a better response, for scientific validity, logical consistency, comprehensibility, completeness, satisfaction level, and empathy by both patients and other rheumatologists. Rheumatologists rated ChatGPT4’s responses higher for satisfaction (mean difference, 0.537) and empathy (0.460) compared with rheumatologist responses. Patients rated the ChatGPT4 and rheumatologist responses similarly across all 6 domains. The researchers also conducted a subgroup analysis by language and determined that ChatGPT4 responses transcended language barriers for patients. They concluded that their study demonstrates ChatGPT4’s potential for improving consults with patients who have SLE.