Answers are grounded in this paper and cite their sources.
Introduction ChatGPT is an artificial intelligence (AI)-driven chat technology that has gained popularity as a patient resource for a variety of clinical questions, including common orthopedic conditions. The purpose of this study was to assess ChatGPT 3.5’s effectiveness in providing accurate and high-quality answers to common patient questions about Dupuytren’s contracture (DC). Methods Ten questions were obtained from the “Frequently Asked Questions” from ten well-known healthcare institution websites. These questions were input into ChatGPT 3.5’s free online chatbot. The responses were recorded and analyzed by two senior authors. Using the DISCERN score, Flesch-Kincaid Grade Level, and Journal of the American Medical Association (JAMA) Benchmark criteria, the responses were graded for accuracy and readability. Results Out of the ten “Frequently Asked Questions,” there were eight poor responses and two fair responses. The average Flesch-Kincaid Grade Level was 14.3 (range: 12.3-16.6), indicating a college graduate reading level. The JAMA Benchmark criteria score was zero for all responses due to the absence of citations. Overall, the chatbot’s answers were inconsistent but remained unbiased and appropriately referred patients to healthcare professionals for further guidance. Conclusion ChatGPT can provide supplemental information for patients seeking educational material about Dupuytren’s contracture. However, it is questionable if ChatGPT is suitable for patient use given the advanced reading level, absence of citations, and lack of a comprehensive discussion it provides in its response. It is crucial to recognize the individuality of each patient and the importance of personal discussions about their diagnosis, treatment, and concerns with a licensed healthcare provider.
No takes yet. Share an insight, caveat, or question.
Kane et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: