Attention: Restrictions on use of AUA, AUAER, and UCF content in third party applications, including artificial intelligence technologies, such as large language models and generative AI.
You are prohibited from using or uploading content you accessed through this website into external applications, bots, software, or websites, including those using artificial intelligence technologies and infrastructure, including deep learning, machine learning and large language models and generative AI.
The Quality of Medical Information on the Internet
By: Stacy Loeb, MD, MSc, PhD (Hon) | Posted on: 05 Oct 2021
The Internet is an important source of health information for our patients and their families. Among adult Internet users, 72% say they have searched online for information about a range of health issues, the most popular being specific diseases and treatments.1 The Internet can provide a valuable source of support and community, with 26% of adult Internet users reporting that they have read or watched someone else’s health experience and 16% saying they have gone online to find other people who share the same health condition within the past 12 months.
However, many online resources fall short in providing useful health information for the lay public. One major problem is the understandability and actionability of online information, which includes information from health care professionals and organizations. Unfortunately, we have found that many resources are presented above the recommended reading level for consumer health information and do not provide clear action steps for viewers.2
In addition, online resources may not present information that is culturally tailored or that reflects the diversity of the population. For example, we reported under-representation of Black men in YouTube videos about clinical trials for prostate cancer.3 We are currently studying the impact of online information on racial disparities in prostate cancer in a grant from the Department of Defense.
Another significant problem is the dissemination of biased and/or misinformative information about health through online networks. For example, we reported that, among the top 150 YouTube videos in searches for “prostate cancer” and “bladder cancer,” more than a fifth contained a moderate to large amount of misinformation.4,5 Another new study reported on the top 50 news articles shared on Facebook and other social networks about 4 different cancers.6 Among these 200 articles, 32.5% contained misinformation and 30.5% contained harmful information. Importantly, misinformative articles received more engagement from viewers.
The problem of online misinformation is not limited to oncology. Online content that is poor quality, false, misleading or commercially biased has been reported in female pelvic medicine, endourology, sexual medicine and infertility.7
Social media may be particularly problematic in this regard. Misinformation can be rapidly broadcast to large audiences in a “viral” way that was not previously possible. This phenomenon may be exacerbated by interactions between users and by algorithms that prioritize popular content.8
Unfortunately, it is logistically infeasible for medical experts to manually review the vast quantity of online information. Therefore, an automated method to identify potentially misinformative content is needed. In an interdisciplinary collaboration, our group created a machine learning algorithm incorporating multimodal features (eg meta-data, linguistic and acoustic features) to identify misinformation in YouTube videos about prostate cancer.9 The preliminary model had 74% accuracy, which we hope to improve through additional training. In the future, machine learning may facilitate a “smarter” search filter to help flag misinformation and prioritize information that is likely to be of higher quality.
In the meantime, it is important for health care providers to direct patients to vetted sources of additional information. As the saying goes, “The solution to pollution is dilution.” Active participation of experts in the creation of online content can help to raise the signal above the noise and provide evidence-based information about urological health to the public.
Dr. Loeb is supported by the Department of Defense and the Prostate Cancer Foundation.
- Pew Research Center: The Social Life of Health Information. Pew Research Center 2014. Available at https://www.pewresearch.org/fact-tank/2014/01/15/the-social-life-of-health-information/. Accessed August 8, 2021.
- Herbert AS, Nemirovsky A, Hess DS et al: An evaluation of the readability and content-quality of pelvic organ prolapse YouTube transcripts. Urology 2021; 154: 120.
- Borno HT, Zhang S, Bakke B et al: Racial disparities and online health information: YouTube and prostate cancer clinical trials. BJU Int 2020; 126: 11.
- Loeb S, Sengupta S, Butaney M et al: Dissemination of misinformative and biased information about prostate cancer on YouTube. Eur Urol 2019; 75: 564.
- Loeb S, Reines K, Abu-Salha Y et al: Quality of bladder cancer information on YouTube. Eur Urol 2021; 79: 56.
- Johnson SB, Parsons M, Dorff T et al: Cancer misinformation and harmful information on Facebook and other social media: a brief report. J Natl Cancer Inst 2021; https://doi.org/10.1093/jnci/djab141.
- Loeb S, Taylor J, Borin JF et al: Fake news: spread of misinformation about urological conditions on social media. Eur Urol Focus 2019; 6: 437.
- Lee JJ, Kang KA, Wang MP et al: Associations between COVID-19 misinformation exposure and belief with COVID-19 knowledge and preventive behaviors: cross-sectional online study. J Med Internet Res 2020; 22: e22205.
- Hou R, Perez-Rosas V, Loeb S et al: Towards Automatic Detection of Misinformation in Online Medical Videos. Available at https://arxiv.org/abs/1909.01543.