ChatGPT for Patient Counseling in Kidney Stone Clinic: Comment

Dear Editor,

We would like to share ideas on the publication “Evaluation of ChatGPT for Patient Counseling in Kidney Stone Clinic: A Prospective Study.” ¹ The study examined how two different levels of skill urologists responded to queries from patients with kidney stones on ChatGPT. ChatGPT outperformed urologists in accuracy, empathy, thoroughness, and practicality. The only category with discernible variance was significance. Surprisingly, urologists strongly agreed on the empathy dimension, implying that they were on par or even better than ChatGPT in expressing empathy to patients.

The small sample size of patients and questions in the study may have limitations that could skew the findings. Increasing the number of patients and research questions can lead to more reliable and robust outcomes. A larger sample size allows for a more representative sample of the population, which improves the findings' generalizability. Furthermore, increasing the number of questions allows for a more thorough understanding of the various queries that patients may have in real-life situations, ensuring that the study captures a fuller picture of patient experiences and concerns. These improvements would improve the validity and relevance of the study's findings.

Furthermore, the study omitted information regarding the specific training and background of the urologists who participated in the assessment, which may have affected the validity of their conclusions. The exclusion of information about the specific training and background of the urologists who took part in the assessment could have impacted the validity of the conclusions in a variety of ways. For example, urologists' knowledge and skill may have influenced the accuracy of their assessments and diagnoses.

Urologists with more training and experience may have been more likely to correctly recognize symptoms and make accurate decisions, resulting in different outcomes than less experienced urologists. In addition, the urologists' experience with specific diagnostic tools or approaches may have influenced their decision-making process, thus skewing the study's findings. As a result, incorporating information regarding the urologists' training and backgrounds could have offered valuable context for evaluating the study findings and verifying the validity of the conclusions.

To assure a more thorough analysis and to improve the study, researchers might think about increasing the sample size of questions and patient participants.

Further research may investigate the prospect of improving the accuracy and reliability of ChatGPT replies in the medical area, particularly in clinical settings such as kidney stones. Integrating competent human input may improve the quality of replies. Given that people are the major users of technology, following specific behavioral norms is critical. ²