The reliability, functional quality, understandability, and actionability of fall prevention content in YouTube: an observational study

Background Falls are common but dangerous in the elderly. More and more seniors are searching for healthcare information online. YouTube has become the world's most popular video streaming platform. Albeit thousands of fall prevention videos are available on YouTube, their reliability, functional quality, understandability, and actionability have not been verified. Methods The top 300 watched videos on YouTube related to fall prevention were retrieved. After exclusion, all qualified sample videos were evaluated by three validated assessment instruments (the PEMAT scale, the HONCode scale, and the DISCERN instrument) regarding their reliability, functional quality, understandability, and actionability. Each video's length, number of views/likes/comments, forms of expression, and the uploader's profile were collected as well. The Wilcoxon rank sum test was performed for further analysis from the perspective of expression forms and uploaders' identities. Results One hundred thirty-seven videos (45.67%) were qualified as sample videos, and individuals/organizations with medical backgrounds posted 54.01% of them. Most of the excluded videos (n = 163) were irrelevant (n = 91, 55.83%), and commercial (n = 52, 31.90%). The median video length for sample videos was 470 seconds. The DISCERN instrument indicated that 115 videos (83.94%) were of moderate to high overall quality. Medical practitioners and organizations gained the highest scores in functional quality and reliability (P < 0.05), while they also tended to use technical terms more often (mean = 3.15). The HONCode scale suggested a lack of traceability was common. The most popular and actionable form of expression was workout (n = 58, median score = 86.90, P < 0.05), while monolog and keynote presentations scored the highest in understandability (no significant difference between them). The PEMAT scale suggested videos uploaded by medical teams were the easiest to be understood (P = 0.011 and P < 0.001, respectively), whereas they were less actionable than those made by fitness trainers (P = 0.039 and P < 0.001, respectively). Conclusions Cooperation between the medical team and fitness trainers is expected for better health promotion. Plain language is advised, and sources should be provided. As for expression form, monolog or keynote presentations, plus workout clips, might be the most effective.