What Hundreds of Millions of Patients With Dry Eye Will Find on YouTube: A Quality and Reliability Research of the YouTube Videos.

Abstract

The aim of this study was to evaluate the demographic features, quality, and reliability of YouTube videos addressing dry eye disease (DED). The term "dry eye disease" has been searched on YouTube. The first 500 videos that emerged using the defined search term were evaluated. Duplicated-split videos, videos shorter than 60 seconds, videos with a language other than English or videos with an unintelligible English accent, and videos unrelated to DED were excluded. Video uploaders, types, origins, durations, and viewer interactions of the videos were noted. DISCERN, the Global Quality Score, and the Video Quality Score (created by the authors) were used to evaluate the video quality. Of the 500 videos, 262 videos were excluded, and the remaining 238 videos were evaluated. Videos were of moderate quality on all 3 scoring systems. The medical institute and academic society videos had the highest quality ( P < 0.05). The quality of videos uploaded by physicians was significantly lower than medical institute and academic society videos and was higher than the others ( P < 0.05). User interactions and video duration were weakly positively correlated with the video quality ( P < 0.05). Among YouTube videos on DED, only a minority are of good or excellent quality. Videos uploaded by medical institutes or academic societies scored higher in quality than those uploaded by physicians, which, in turn, scored higher than those uploaded by all others. User interactions were weakly correlated with quality values, suggesting user interactions are not good indicators of the quality of YouTube videos on DED.