With over 1 billion registered users and 7 billion videos, YouTube is a hugely popular platform for information sharing—including medical advice. A recent study found that the platform houses over 600,000 videos on the topic of prostate cancer alone—a 1,000-fold increase in the last 10 years—and over 75 percent of those videos included misinformative content (Eur Urol 2018; doi.org/10.1016/j.eururo.2018.10.056).
The study examined the top 150 videos on prostate screening and treatment, with a combined total of over 1.3 million views, finding that 115 of those contained biased information, ranging from alternative medicine recommendations that are not evidence-based to outdated professional recommendations. With a greater understanding of patients' need for information related to their conditions, clinicians can provide patients and their families with verified sources of evidence-based information.
Prevalence of Misinformation
For study lead Stacy Loeb, MD, the lack of evidence-based content is a huge concern for patient safety. In a recent focus study, Loeb and colleagues found that prostate cancer patients did not have a lack of information, but a difficulty navigating it to discern what is accurate and what pertains to their individual situations.
“In our society as a whole, the fact that there is a lot of circulating misinformation on social networks has become a major issue,” Loeb said. “With this study, we were interested in looking at whether this issue occurs in prostate cancer-related content, what the quality of that information is, and if there are similar issues in health information online as there is in other segments of society.”
Findings from the study found that not only was there a high prevalence of misinformation in online content, but that the information was reaching a huge number of patients with a total reach of over 6 million viewers between YouTube videos and related comments.
When searching for prostate cancer videos on the platform, the first page of results for the terms “prostate cancer screening” and “prostate cancer treatment” have over 100,000 views each—demonstrating the popularity of YouTube for information regarding the disease. That popularity, Loeb said, threatens the ability for patients to gather unbiased information that reflects current recommendation guidelines.
“Though guidelines are publicly available,” Loeb noted, “many patients turn to platforms such as YouTube to seek more information about a recent diagnosis, screening recommendations, or treatment options.”
That tendency can lead patients to outdated or biased information, especially with regards to recent changes in recommendations from the U.S. Preventive Services Task Force finalized in 2017.
“In 2011, there was a recommendation against prostate cancer screening, but in 2017 the group issued a new draft that instead recommended that men should be told of the pros and cons of screening and be a part of the decision-making process,” she said.
In addition, current recommendations prescribe active surveillance as a treatment option for low-rate prostate cancer and there are new drugs for advanced prostate cancer, Loeb continued.
“Medicine is a very rapid and dynamic area, but social networks with user-generated content remain online for perpetuity,” Loeb stated. In some cases, continued, information in online videos may have been accurate at one time but are no longer true, and in other cases, content was biased from the start. “Patients watching outdated or biased videos would be getting advice that is the opposite of what is currently recommended.”
For example, one video with more than 300,000 views recommended injecting herbal supplements into the prostate for the treatment of prostate cancer—something that is not part of recommended treatment options and is not evidence-based, Loeb said. Many videos had less dramatic uses of misinformation, such as outdated recommendations.
Validating Video Quality
To determine the quality of information, the researchers used the validated DISCERN quality criteria for consumer health information, which incorporates key elements of prostate cancer guidelines, including whether the video or its comments describes benefits and risks or supports shared decision-making. The study also included rankings on intended audience, favoring new technology, recommending complementary/alternative medicine, commercial bias, and the extent of misinformation.
A subset of 50 videos was also graded on the Flesch-Kincaid reading level of written transcripts and the Agency for Healthcare Research and Quality Patient Education Materials Assessment Tool, a systematic method for evaluating understandability and actionability that allows viewers to identify what they can do based on the information presented. Results on those metrics found that more videos (75%) had a moderate to good description of benefits than harms (53%) and only half of the videos provided moderate to good support for shared decision-making. A quarter of videos were biased toward the use of new technology and 19 percent recommended complementary or alternative medicine. The Flesch-Kincaid reading level for videos was 12th grade.
Of the 150 videos studied, only five were authored by patients. The largest number of videos were authored by health care providers, such as individual doctors, clinics, or universities, followed by news media outlets such as television segments, with the third largest group being societies and organizations such as foundations and academic journals. A smaller number of videos were produced by health or wellness channels and commercial media.
“We found that health and wellness channels had a low quality of information, and many videos from patients had misinformation,” Loeb said. “The highest quality of information was from societies or organizations, including governmental groups and foundations.”
Most videos had clear objectives, a logical flow of information, and good audio quality, but few discussed multiple options, provided sources, or summarized information. Interestingly, the study found that commercial/health-wellness and patient-generated videos had the highest engagement by views per month and thumbs up per view, but had worse quality than videos from health care providers and professional/government groups.
Improving Patient Access to Information
“Unfortunately, a lot of videos use jargon and don't necessarily make it easy to understand or explain clearly what action steps to take next,” Loeb noted. “This can be useful information for health care professionals and stakeholders involved in creating educational content—be mindful to make good patient education materials that break down information using everyday language and actions in a way that consumers will know what to do.”
Loeb added that in her own experience, patients come to appointments having seen things online that are tough to explain and may not be evidence-based. “Sometimes we end up spending a lot of time as myth busters during the medical visit because of patients' exposure to misinformative content online.
“Ultimately, most people being treated for prostate cancer are under the care of a doctor and will therefore receive recommended treatments. However, if half of the visit is spent discussing content the patient saw online that isn't a recommended option, we have even less time to discuss the treatments that do fall under recommendation guidelines.”
Situations like those can leave patients with uncertainty about how to proceed with their treatment. “As health care professionals, we have to recognize that patients have a need for information about their conditions that can be provided one-on-one during the patient-clinician encounter,” Loeb noted. “It's important to incorporate a short amount of time during that encounter to highlight good sources of information for patients to use.”
Kelly Wolfgang is a contributing writer.