Objectives: Large language models such as ChatGPT have been used to generate text in a conversational manner, and may be of use in providing patient information in a urological setting. This study evaluated the accuracy, presence of omissions, and preferability of traditional patient information to the large language models ChatGPT and Bing Chat. Methods: Eight common questions regarding urolithiasis and prostate cancer were selected from traditional patient information and posed to ChatGPT and Bing Chat. Responses from all sources were then evaluated by seven urologists in a blinded fashion for accuracy, omissions, and preferability. Results: We found that 96.43% of ratings of traditional patient information sources were rated accurate, compared to 94.6% for ChatGPT and Bing Chat; 7.1% of ratings of traditional patient information were rated as containing harmful omissions, compared to 10.71% for ChatGPT and 21.4% for Bing Chat; and 55.4% of rater first preferences were given to ChatGPT, compared to 35.7% for traditional patient information and 8.9% for Bing Chat. Conclusions: ChatGPT provided responses of a similar accuracy and preferability to traditional sources, highlighting its potential as a supplementary tool for urological patient information. However, concerns remain regarding omissions and complexity in model-generated responses.

Objectives Many people are turning to alternatives to the conventional doctor-patient relationship, such as webbased search engines and video forums for their health care information. We undertook this study to investigate the quality of videos and information on renal cancer available on the streaming platform YouTube. Methods We completed a search of YouTube (www.YouTube.com) in September 2021 with the term “kidney cancer.” The first 120 videos found which met the inclusion criteria (English speaking, duration greater than one minute, greater than 500 views, renal cancer addressed) were selected. We recorded information including duration, view count, likes, dislikes, comments, publisher, and author. The modified DISCERN tool and Global Quality Score (GQS) questionnaire were used to assess the quality of the included videos. The level of misinformation was assessed using a Likert 5-point scale. Descriptive statistics were used to analyse the collected data. A 2-sample t test was used to further analyse the quality assessment tool results before, during, and after 2016. Results Most videos were published during or after 2016 (63.3%), were predominantly created in North America (77.5%), and were presented by health care professionals (60%). The median length of the videos was 4.23 (1.01 to 65.55) minutes, and the median number of views was 3087 (514 to 228 152). The median number of likes and dislikes was 24 and 5, respectively. The median modified DISCERN score was 3, the median GQS score was 3, and the grading for overall level of misinformation was moderate. Conclusion The quality of information accessed from YouTube on kidney cancer is of a low to moderate overall standard with significant levels of misinformation. YouTube should not be used alone for educational purposes on renal cancer by patients or the public. It is best used in conjunction with information and advice from a medical practitioner and the health care system.