An Evaluation of Educational Neurological Eye Movement Disorder Videos Posted on Internet Video Sharing Sites

Optic disc drusen (ODD) are hyaline deposits located within the optic nerve head. Peripapillary retinal nerve fiber layer (RNFL) thinning is associated with the high prevalence of visual field defects seen in ODD patients. The goal of this study was to investigate the characteristics of patients with ODD and to compare the peripapillary RNFL thickness to the extent of visual field defects and anatomic location (superficial or buried) of ODD.; ; Retrospective, cross sectional study.; ; A total of 149 eyes of 84 ODD patients were evaluated. Sixty-five percent were female and 76% had bilateral ODD. Of 149 eyes, 109 had superficial ODD and 40 had buried ODD. Peripapillary RNFL thinning was seen in 83.6% of eyes, where optical coherence tomography was performed (n = 61). Eyes with superficial ODD had greater mean peripapillary RNFL thinning (P ≤ 0.0001) and visual field defects (P = 0.002) than eyes with buried ODD. There was a correlation between mean peripapillary RNFL thinning and visual field defects as measured by perimetric mean deviation (R-0.66; P = 0.0001). The most frequent visual field abnormalities were arcuate and partial arcuate defects.; ; Peripapillary RNFL thickness correlates with anatomic location (superficial or buried) of ODD. Frequency and extent of visual field defects corresponded with anatomic location of ODD and peripapillary RNFL thickness, suggesting increased axonal damage in patients with superficial ODD.

Original Contribution An Evaluation of Educational Neurological Eye Movement Disorder Videos Posted on Internet Video Sharing Sites Simon J. Hickman, PhD, FRCP Background: Internet video sharing sites allow the free dissemination of educational material. This study investigated the quality and educational content of videos of eye movement disorders posted on such sites. Methods: Educational neurological eye movement videos were identified by entering the titles of the eye movement abnormality into the search boxes of the video sharing sites. Also, suggested links were followed from each video. The number of views, likes, and dislikes for each video were recorded. The videos were then rated for their picture and sound quality. Their educational value was assessed according to whether the video included a description of the eye movement abnormality, the anatomical location of the lesion (if appropriate), and the underlying diagnosis. Results: Three hundred fifty-four of these videos were found on YouTube and Vimeo. There was a mean of 6,443 views per video (range, 1-195,957). One hundred nineteen (33.6%) had no form of commentary about the eye movement disorder shown apart from the title. Forty-seven (13.3%) contained errors in the title or in the text. Eighty (22.6%) had excellent educational value by describing the eye movement abnormality, the anatomical location of the lesion, and the underlying diagnosis. Of these, 30 also had good picture and sound quality. The videos with excellent educational value had a mean of 9.84 "likes" per video compared with 2.37 for those videos without a commentary (P , 0.001). The videos that combined excellent educational value with good picture and sound quality had a mean of 10.23 "likes" per video (P = 0.004 vs videos with no commentary). There was no significant difference in the mean number of "dislikes" between those videos that had no commentary or which contained errors and those with excellent educational value. Department of Neurology, Royal Hallamshire Hospital, Sheffield, United Kingdom. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the full text and PDF versions of this article on the journal's Web site (www. jneuro-ophthalmology.com). S. J. Hickman is on the editorial board of the NOVEL education Web site. The author has no relevant financial disclosures. Address correspondence to Simon J. Hickman, PhD, FRCP, Department of Neurology, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, United Kingdom; E-mail: [e-mail] Hickman: J Neuro-Ophthalmol 2016; 36: 33-36 Conclusions: There are a large number of eye movement videos freely available on these sites; however, due to the lack of peer review, a significant number have poor educational value due to having no commentary or containing errors. The number of "likes" can help to identify videos with excellent educational value but the number of "dislikes" does not help in discerning which videos have poor educational value. Journal of Neuro-Ophthalmology 2016;36:33-36 doi: 10.1097/WNO.0000000000000292 © 2015 by North American Neuro-Ophthalmology Society T here is now a wealth of medical educational material on Internet video sharing sites, such as YouTube and Vimeo (1). Videos allow the demonstration of difficult concepts by using simulation, diagrams, dynamic illustrations, and real patients. However, because there is no peer review before videos are uploaded onto these sites, it can be difficult for the viewer to identify educationally useful videos (2). The popularity of a video can be ascertained by the number of views it has had. In addition, the video sharing sites permit viewer feedback by the use of clicking that they "like" (on YouTube and Vimeo) or "dislike" (on just YouTube) a video (3,4). Viewers can also leave "comments" that can be a means of providing critical feedback (4,5). These have been termed the user engagement measurements (6). What is not clear is whether these metrics are helpful in identifying educationally useful content. Recently, a number of educational institutions and medical journals have started uploading content onto YouTube as a means of both disseminating knowledge but also to showcase the institution or journal (1). Since 2009, YouTube has organized educational content from college and university partners into YouTube EDU (7,8). One might suspect that videos from educational establishments would be more popular and more highly rated. However, in a study of health education videos on cardiovascular disease aimed at the public, there was no correlation 33 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution between a rating of the educational content and accuracy of the videos and any user engagement measurement (6). For the demonstration of eye movement disorders, videos are invaluable and DVDs are now included in some textbooks (9,10). A number of educational videos of eye movement disorders have been uploaded onto the video sharing sites. The aims of this study were to assess the number of these videos present, the range of disorders covered, the quality and educational value of these videos and whether any of the user engagement measurements are useful in differentiating the most educational useful videos from those which are less useful or which contain errors. METHODS The video sharing sites Break, Dailymotion, Metacafe, Vimeo, and YouTube were accessed. On these sites, searches were performed to identify videos containing neurological eye movement disorders. Search terms were obtained from the index of the book The Neurology of Eye Movements (9) (See Supplemental Digital Content, Table E1, http://links.lww.com/WNO/A167). Also, suggested TABLE 1. The metrics recorded for each video and the criteria by which the videos were judged The type of eye movement abnormality shown The number of views The length of the video in seconds A subjective rating of picture quality (ranging from 1 to 3) Depending on lighting, camera shake, focusing, and maintaining a stable view of the eyes A subjective rating of sound quality (ranging from 1 to 3) if a verbal commentary was present Depending on legibility and the presence of background noise Whether the subject was an adult or a child Whether the subject was identifiable Whether there was a statement confirming the subject's consent to upload the video Whether a commentary was present and the educational value of the commentary based on: Whether the commentary described the eye movement abnormality Whether the anatomical location of the lesion was discussed (if appropriate) Whether the underlying diagnosis/differential diagnosis was discussed Whether there were any errors in the title and/or commentary Whether the video was on a medical journal or educational establishment channel The number of "likes" recorded by previous viewers The number of "dislikes" (on YouTube only) recorded by previous viewers A review of any comments left 34 links were followed from each video and the channels of each uploader viewed to identify additional videos. The inclusion criteria were that the videos had to be neurological eye movement disorder videos of single human subjects uploaded for the purposes of education and, if the video had commentary, it had to be in English. Duplicate videos were not rerated. Videos uploaded up to December 31, 2014, were included. Table 1 lists the metrics recorded and the criteria by which each video was judged. The subjective rating of picture and sound quality ranged from 1 (poor) to 3 (good). If the age of the subject was unclear from the video, they were assumed to be an adult. Statistical analysis was performed in Numbers for Mac (Apple Inc, Cupertino, CA) and GraphPad Prism (GraphPad Software Inc, La Jolla, CA). Group comparisons were performed using the Mann-Whitney test. Comparison of categorical data was by Fisher exact test. RESULTS The number of thumbnail pictures assessed by entering the search terms were as follows: 1 on Break, 8,494 on Dailymotion, 29 on Metacafe, 18,758 on YouTube, and 520 on Vimeo. Only YouTube and Vimeo contained videos fulfilling the inclusion criteria (342 only on YouTube, 8 only on Vimeo, and 4 on both YouTube and Vimeo, making 354 in total). The types of eye movement abnormalities found on the sites are listed in Supplemental Digital Content, Table E2, http://links.lww.com/WNO/ A168. The total number of views for all of the videos was 2,280,997 with a mean of 6,443 per video (range, 1- 195,957). The total number of views per day for all of the videos was 1,633 with a mean of 4.61 per video (range: 0.003-72.09). The mean length of the videos was 63.1 seconds (range: 4-570 s). The mean picture quality was 1.86, and the mean sound quality was 2.52 in the 58 videos with audio commentary. There were 286 videos of adults and 68 of children. Only 3 videos explicitly stated that the subjects had consented for the video to be uploaded, although a further 32 were on channels uploaded by medical journals. The journals' consent forms contain provision for further dissemination of material (11-13). Therefore, in 319 videos (90.4%), it was not clear whether consent for uploading had been obtained, although the subjects were identifiable in 186 (52.5%) of all of the videos. Two hundred thirty-five videos (66.4%) contained some form of commentary. One hundred seventy-seven (50%) had a written commentary, 35 (9.9%) contained a verbal commentary, and 23 (6.5%) had both a written and verbal commentary, leaving 119 (33.6%) with no form of commentary apart from the title. Hickman: J Neuro-Ophthalmol 2016; 36: 33-36 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution Forty-seven (13.3%) contained errors in the title and/or in the text. Of these 47, it was not possible to identify clearly what the eye movement was in 6 videos. Appropriate corrections in the comments section were present in only 5 of these videos. Eighty (22.6%) had excellent educational value by describing the eye movement abnormality, the anatomical location of the lesion (if appropriate), and the underlying diagnosis. Of these, 30 also had good picture and sound quality (if the video contained a verbal commentary). These are listed in the Supplemental Digital Content, Table E3, http://links.lww.com/WNO/A169. A comparison of the user engagement measurements for the categories of videos described above is presented in Table 2. Thirty-two (9.0%) of the videos were found on channels supplied by the following journals: Journal of the American Medical Association, Neurology, and New England Journal of Medicine. An additional 9 (2.5%) were found on channels uploaded by University College Dublin and Kellogg Eye Center, University of Michigan Health System. The mean picture quality of these videos was 2.44 (P , 0.001 compared with videos not supplied by journals or educational institutions). Eleven had audio commentary with a mean sound quality of 2.91 (P = 0.02 compared with videos not supplied by journals or educational institutions). Eleven combined excellent educational value with good picture and sound quality (See Supplemental Digital Content, Table E3, http://links.lww.com/WNO/A169) (P = 0.0001 for the proportion of videos in this category when compared with videos not supplied by journals or educational institutions). Importantly, none of these 41 videos contained any errors. The mean number of views of these videos was 2,438 (Median 66) with a mean number of likes of 6.51 (Median 0) and dislikes of 0.32 (Median 0). The relatively low number of the user engagement measurements despite the apparent high quality of these videos probably reflects the shorter time these videos had been uploaded compared with videos not supplied by journals or educational establishments (mean 211 vs 1,295 days, P , 0.0001). DISCUSSION There are some excellent eye movement videos freely available on these Internet sites, covering a variety of neurological eye movement disorders. Due to the lack of peer review, a large number have poor educational value since they contain significant errors or have no commentary. This study demonstrated that user engagement measurements have only a limited role in signposting the most educationally useful videos. The number of "views" and "likes" may help to identify videos with good educational value, but the dislikes and comments sections seem to be less helpful in discerning which videos have poor educational content or which contain significant errors. Videos supplied by medical journals and educational institutions were rated as having better production values, were more likely to have greater educational value, and contained no errors. It is likely that the peer review process within the journals accounted for these findings. The major limitation of this study was that it was performed by a single observer who was not blinded as to the identity of the uploader of their host institution. The major subjective component of the study although was in assessing the picture and sound quality. The assessment of the educational value was more objective, based on the presence or absence of key educational metrics. A large number of videos were rated as having poor picture and sound quality. Many were taken in the outpatient clinic or on the wards with handheld devices. These videos often had poor focusing and camera shake, as well as background noise if a verbal commentary was provided at the time the video was taken. Picture quality could have been improved with better lighting and also the use of a tripod or mount, which are inexpensive and available for both video cameras and smartphones. Editing and audio dubbing would also have helped a number of videos. Advice on how to use video cameras mounted on goggles also is available online (14). The other critically important issue when videoing patients is to obtain consent, both for the videoing itself but also for how the video is to be used and disseminated TABLE 2. A comparison of the user engagement measurements for different categories of videos Category No commentary Errors present Excellent educational value Excellent educational value and good picture/sound quality Total Number of Videos in This Category 119 47 80 30 Mean (Median) Number of Views Per Video 5,159 5,326 7,690 5,799 (710) (1,129) (2,197)* (2,215) Mean (Median) Mean (Median) Number of "Likes" Number of "Dislikes" Per Video Per Video 2.37 4.26 9.84 10.23 (0) (1) (2)† (2.5)‡ 0.26 0.42 0.39 0.52 (0) (0) (0) (0) *P = 0.02 vs no commentary. † P , 0.0001 vs no commentary. ‡ P = 0.004 vs no commentary. Hickman: J Neuro-Ophthalmol 2016; 36: 33-36 35 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution (15). It is important to comply with institution rules on the use of smart technology and handling patient data (16). It is worrisome that in over 90% of the videos viewed in this study, it was not clear whether there had been consent for uploading the video onto the video sharing sites. The policies of the hosting sites are that they expect that the uploader has obtained the necessary consent for the video to be uploaded (17,18). Due to the sheer volume of videos uploaded, it is not possible for the sites to check each video. A standard moderating service for educational videos on these sites has been proposed, which would go some way to improve the current system of "self-policing." However, this would prove difficult and costly to implement given the volume of new content added to the sites daily (19). REFERENCES 1. YouTube and Vimeo for Education. Available at: http://www. jiscdigitalmedia.ac.uk/guide/youtube-vimeo-education. Accessed October 30, 2014. 2. Azer SA, Al Eshaiwi SM, Al Grain HA, Al Khelaif RA. Nervous system examination on YouTube. BMC Med Educ. 2012;12:126. 3. Like or Dislike a Video. Available at: https://support.google. com/youtube/answer/6083270?hl=en. Accessed January 29, 2015. 4. Likes and Comments. Available at: http://vimeo.com/help/faq/ watching-videos/likes-and-comments. Accessed January 29, 2015. 5. Comment on Videos. Available at: https://support.google. com/youtube/topic/3014329?hl=en&ref_topic=6059684. Accessed January 29, 2015. 6. Desai T, Shariff A, Dhingra V, Minhas D, Eure M, Kats M. Is content really king? An objective analysis of the public's response to medical videos on YouTube. PLoS One. 2013;8: e82469. 36 7. Takushi S. YouTube Viewers Do Not Value Content or Credibility in Educational Videos. 2014. Available at: http://guardianlv. com/2014/03/youtube-viewers-do-not-value-content-orcredibility-in-educational-videos/. Accessed October 30, 2014. 8. #Education. Available at: https://www.youtube.com/channel/ UC3yA8nDwraeOfnYfBWun83g. Accessed February 21, 2015. 9. Leigh RJ, Zee DS. The Neurology of Eye Movements, 4th edition. Oxford, United Kingdom: Oxford University Press, 2006. 10. Wray SH. Eye Movement Disorders in Clinical Practice. Oxford, United Kingdom: Oxford University Press, 2014. 11. Consent for Publication of Identifying Material in the JAMA Network Journals. Available at: http://jama.jamanetwork.com/ DocumentLibrary/InstructionsForAuthors/PatientConsent.pdf. Accessed January 31, 2015. 12. Patient's Consent for the Publication of Material Relating to Him or Her in the New England Journal of Medicine. Available at: http://www.nejm.org/userimages/ContentEditor/ 1274822530041/patientid.pdf. Accessed January 31, 2015. 13. Consent for Publication of Photos, Videos, and Other Identifying Materials. Available at: http://www.neurology.org/ site/misc/auth2.xhtml#ELECTRONICFIGURE. Accessed January 31, 2015. 14. Hain TC. Digital Video Recorders and Eye Movements: How I Do it. 2014. Available at: http://www.dizziness-and-balance. com/practice/dvr.htm. Accessed February 2, 2015. 15. Hood CA, Hope T, Dove P. Videos, photographs, and patient consent. BMJ. 1998;316:1009-1011. 16. ECRI Institute. Judgment Call: Smartphone Use in Hospitals Requires Smart Policies. 2012. Available at: https://www.ecri. org/components/HDJournal/Pages/hd201210guid_ Smartphones.aspx?tab=2. Accessed February 1, 2015. 17. Terms of Service. 2014. Available at: http://vimeo.com/ terms. Accessed January 29, 2015. 18. Terms of Service. 2010. Available at: https://www.youtube. com/static?gl=GB&template=terms. Accessed January 26, 2015. 19. Sunderland N, Camm CF, Glover K, Watts A, Warwick G. A quality assessment of respiratory auscultation material on YouTube. Clin Med. 2014;14:391-395. Hickman: J Neuro-Ophthalmol 2016; 36: 33-36 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.