Medical videos on YouTube often provide incomplete information

Implantable cardioverter defibrillator (ICD) information on YouTube is of low and highly-variable quality, according to research presented at the American College of Cardiology (ACC) Quality Summit 2022. As more patients turn to the internet for medical education, the study identified an opportunity for medical institutions to help patients by understanding what information they are getting online and how institutions can improve online resources in the future.

"Patients are likely seeking out information about their medical conditions and procedures online primarily due to convenience and familiarity with the internet, particularly YouTube," said Thomas Lee, MD, resident physician at Rutgers New Jersey Medical School and the study's lead author.

"Most people only have limited time to talk to a clinician about an ICD placement, so it would make sense that they would take to the internet to learn more about it. While we can't stop our patients from getting information online, nor would we want to curtail their access to information about their conditions, it is important that the medical community work to ensure higher quality information that provides a full picture of a treatment or procedure is available."

According to the researchers, internet behavior data suggest that 80% of individuals search online for health information, with 56% searching for information on medical treatments or procedures. A recent Health Information National Trends Survey estimated 40% of Americans viewed YouTube videos regarding health and medical information in 2020.

In this study, researchers performed a YouTube search using the term "Implantable Cardioverter Defibrillator," before excluding lecture-styled videos, surgical and/or operating room videos, non-English language videos without subtitles, videos with irrelevant content, videos without audio or subtitles, and duplicates. Of the first 75 videos identified, 25 were excluded. The remaining 50 were categorized by uploader type: academic institution (29), medical device company (13) or independent uploader (8).

Video quality was assessed using modified DISCERN criteria (range: 0-25), which was developed to help patients judge information regarding treatment options and is used in patient-education literature to assess overall quality, bias and reliability of content. Using prior studies, the qualitative score cut-offs were: very poor (<9), poor (9-13), average (14-17), good (18-21) and excellent (22-25). Video quality was also assessed by an electrophysiologist-reviewed Novel Content Score (range: 0-10), which was designed to represent the information an electrophysiologist would present to a patient considering an ICD procedure. Similar qualitative score cut-offs were included: very poor (0-2.5), poor (2.6-40), average (4.1-6.0), good (6.1-8.0) and excellent (8.1-10). The researchers assessed video popularity through secondary outcomes, including views, likes, duration, days since upload and comments.

Overall, the videos assessed were found to be of poor quality and highly variable across both scoring systems. The average modified DISCERN score was 12.58 and the average Novel Content Score was 3.68. The researchers found videos uploaded by medical device companies had significantly lower modified DISCERN scores compared to academic institutions and independent uploaders. Earlier search result position had a significant positive correlation with the Novel Content Score, but not with the modified DISCERN score. There was also a significant positive correlation between video duration and both scoring criteria. None of the metrics of video popularity were found to correlate with video quality.

According to the researchers, the study demonstrates an opportunity for medical institutions to help patients by implementing new strategies to improve video content, quality and visibility as more patients seek online resources for patient education. For example, researchers found a majority of the videos failed to mention inappropriate shock as a possible outcome for ICDs.

"It's so important that patients have access to high quality information because of how challenging it can be to interact with medical professionals. Oftentimes a patient will have to wait weeks or even months before they can confidently have their questions answered," Lee said. "The issue is that on YouTube so much of the informative videos seem like they're high quality because of a hospital setting or doctor narrating, but in reality, the video fails to convey the complete picture of an ICD placement."