Abstract
Dear Editor,
We appreciate the insightful and constructive feedback provided by the authors regarding our recent study. We welcome the opportunity to clarify the methodological and analytical aspects of our research.
Firstly, the selection of questions was based on Google’s “People Also Ask” (PAA) section, which reflects algorithm-driven user behavior. This approach has been adopted in previous studies analyzing artificial intelligence tools and online search patterns.1,2 Nevertheless, we acknowledge the limitation of not including patient-based qualitative data, and we fully support future efforts to incorporate survey-based or interview-derived patient perspectives.
Secondly, academic validity was assessed by two orthopedic specialists through reference to existing literature and clinical guidelines. The evaluation was supported by established readability and quality tools such as DISCERN, FKGL, SMOG, and FRES, which are widely recognized and used in similar orthopedic contexts.3,4 We agree, however, that expert-based assessment carries inherent subjectivity and could be strengthened in future research through interdisciplinary and patient-centered validation models.
Third, the statistical methods used—such as Fisher’s exact test and Mann–Whitney U—were explicitly described in the “Materials and Methods” section. The difference between ChatGPT’s academic consistency (83.3%) and Google’s (58.3%) was statistically significant. Still, we recognize that the absence of specific p-values in the main text might limit interpretability, and we aim to improve transparency in future reports.
Additionally, we observed that ChatGPT responses tended to be more academically detailed, while Google answers were shorter and easier to read. This finding is consistent with prior literature suggesting a trade-off between comprehensiveness and readability. 5 We believe simplified language, visual aids, and personalized content generation could help bridge this gap in future AI health communication tools.
In conclusion, while this study represents a preliminary analysis, we believe the thoughtful commentary provided lays the groundwork for more robust, multi-layered, and patient-centered future studies. We sincerely thank the authors again for their valuable contributions.
Sincerely, Aydilek A., Karadamar Ö.L.
