Accuracy and readability analysis of ChatGPT responses to frequently asked questions about olecranon fractures.

PurposeThis study aimed to evaluate the reliability, quality, and readability of ChatGPT-generated responses to common patient questions about olecranon fractures.MethodsIn this cross-sectional study, twenty frequently asked questions were identified using Google's "People also ask" feature and submitted to ChatGPT-4o in separate sessions between October 1 and October 10, 2025. Two orthopaedic surgeons independently assessed the responses using the DISCERN instrument and Global Quality Score (GQS). Readability was analysed using Flesch Reading Ease (FRE), Flesch-Kincaid Grade Level (FKGL), Simple Measure of Gobbledygook (SMOG), and Gunning Fog Index (GFI).ResultsMean DISCERN and GQS scores (51.25 ± 4.13 and 4.10 ± 0.58) indicated good quality and reliability. The mean FRE was 62.00 ± 5.12, and mean FKGL, SMOG, and GFI values (9.23, 8.42, 9.37) corresponded to a ninth-grade reading level. DISCERN and GQS were strongly correlated (r = 0.913, p < 0.001) and negatively associated with FKGL.ConclusionChatGPT-4o produced coherent responses of generally good quality and reliability regarding olecranon fractures; however, readability levels exceeded those recommended for patient comprehension. These findings suggest that while ChatGPT may serve as a supportive informational tool, clinician guidance remains essential to ensure appropriate patient understanding and prevent potential misinformation.