512220

research-article2014
JHI21310.1177/1460458213512220Health Informatics JournalMadathil et al.

Article

Health Informatics Journal

2015, Vol. 21(3) 173­–194
Healthcare information © The Author(s) 2014
Reprints and permissions:
on YouTube: A systematic review sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1460458213512220
jhi.sagepub.com

Kapil Chalil Madathil, A Joy Rivera-Rodriguez,

Joel S Greenstein and Anand K Gramopadhye
Clemson University, USA

Abstract
This article reviews the peer-reviewed literature addressing the healthcare information available on YouTube.
Inclusion and exclusion criteria were determined, and the online databases PubMed and Web of Knowledge
were searched using the search phrases: (1) YouTube* AND Health* and (2) YouTube* AND Healthcare*. In
all, 18 articles were reviewed, with the results suggesting that (1) YouTube is increasingly being used as a
platform for disseminating health information; (2) content and frame analysis were the primary techniques
employed by researchers to analyze the characteristics of this information; (3) YouTube contains misleading
information, primarily anecdotal, that contradicts the reference standards and the probability of a lay user
finding such content is relatively high; (4) the retrieval of relevant videos is dependent on the search term
used; and (5) videos from government organizations and professional associations contained trustworthy
and high-quality information. YouTube is used as a medium for promoting unscientific therapies and drugs
that are yet to be approved by the appropriate agencies and has the potential to change the beliefs of
patients concerning controversial topics such as vaccinations. This review recognizes the need to design
interventions to enable consumers to critically assimilate the information posted on YouTube with more
authoritative information sources to make effective healthcare decisions.

Keywords
consumer health information, e-health, health information on the Web, healthcare information on
YouTube, video-sharing sites

Introduction
Video-sharing sites are popular sources of information. YouTube, the most well known of these
sites, exceeds 2 billion views per day,1 with a new video being uploaded on average every minute
and an average user spending at least 15 min a day on the site.2 A recent Health Information

Corresponding author:
A Joy Rivera-Rodriguez, Department of Industrial Engineering, Clemson University, 130-C Freeman Hall, Clemson, SC
29634, USA.
Email: [email protected]
174 Health Informatics Journal 21(3)

National Trends Survey (HINTS) reports a substantial increase in the use of the Internet for retriev-
ing health information. Recent surveys have found that 8 of 10 Internet users accessed health
information online.3,4
Patients with chronic illnesses in particular are increasingly relying on Internet-based resources
to manage their conditions.5 According to surveys conducted by the Pew Research Center, deci-
sions made by 75 percent of such patients on how to treat their condition were influenced by the
knowledge acquired through online health information searches.5–8 These results suggest that a
platform like YouTube has the potential to serve as an important vehicle for sharing and dissemi-
nating timely health-related information, both in its function as a repository of videos and as a
social networking interface where users can interact and socialize. In fact, this interaction may
increase its influence on healthcare decisions, extending its benefits beyond being a diagnostic aid
or an educational tool for healthcare conditions and their management to being a source for infor-
mation sharing among patients coping with illness.9,43
However, healthcare providers and government agencies alike have expressed concern about
the veracity and quality of the information available on this platform,10–13 primarily for two rea-
sons: increased use of YouTube to post anecdotal information and more importantly, its minimal
guidelines and interventions regulating the content of the material uploaded on the site. These
issues raise questions about the trustworthiness of this information source and the risk of dissemi-
nating misleading information. As a result, government agencies and researchers are currently
examining the use of social media as a platform for communicating health information. This
article reviews the peer-reviewed literature investigating the content and veracity of healthcare
information available on YouTube. It discusses the implications of these studies and provides sug-
gestions for future research to support the dissemination of reliable information to healthcare
consumers.

Method
Inclusion and exclusion criteria
The inclusion criteria for the studies included in this review were as follows: (1) the domain of the
article was related to health care; (2) the primary focus of the article was to understand the type
of healthcare information available on YouTube; (3) the article was published in a peer-reviewed
journal; (4) the article detailed the methodology of the study reported; and (5) the article was
available in English. Articles were excluded if they contained only conceptual or theoretical dis-
cussions on the potential use of YouTube and similar social media for healthcare information
dissemination.

Search strategy
The online databases Web of Knowledge, an academic citation and indexing service, and PubMed,
the MEDLINE database of references and abstracts on life sciences and biomedical topics, were
searched using the following search phrases: (1) YouTube* AND Health* and (2) YouTube* AND
Healthcare*. In this review, the term health care was used to include studies that focus on proce-
dures and methods that could be used by a healthcare consumer to maintain and improve their
physical and mental health. The initial search was conducted on the second week of October
2012, and a subsequent search was conducted on the third week of November 2012. These
searches yielded 94 articles from the Web of Knowledge database and 77 articles from PubMed.
The initial screening was done by the first author (K.C.M.). The second author (A.J.R-R.)
Chalil Madathil et al. 175

Figure 1. Selection process for study inclusion.

separately reviewed the screened articles to ensure that the articles met the inclusion criteria. The
discrepancies were discussed by the two researchers until a consensus was met. In all, 31 articles
met the inclusion criteria, and a search of their references yielded 2 additional articles appropriate
for this review. Among the 33 articles subsequently screened for content, 10 did not meet the
inclusion criteria. Of the 23 articles remaining, 2 analyzed the linguistic features and social sup-
port mechanisms of the videos and user comments, and another focused on identifying the per-
sonal health information present on YouTube. These articles were excluded due to their narrow
scope. Two studies investigating smoking and smoking cessation on YouTube which focused on
understanding the influence of camouflaged advertisements supporting the use of tobacco were
also excluded. Thus, in total, 18 articles were included in this review. The selection process for
study inclusion is shown in Figure 1.
176 Health Informatics Journal 21(3)

Results
Of the 18 articles reviewed, 11 were from the United States, 5 from Europe, 1 from Canada, and 1
from India. A total of 11 studies focused on the informational content available for various health
conditions, 5 focused on the portrayal of controversial medical issues such as vaccination and
organ donation, 1 focused on the quality of information, and 1 focused on the portrayal of dentistry.
Table 1 summarizes the key findings of the 18 articles reviewed.

Methodological characteristics and analytical techniques

YouTube offers four ways of sorting search results: (1) relevance, (2) upload date, (3) view count,
and (4) rating. Of the 18 studies reviewed here, 14 (77.8%) sorted their results based on relevance,
1 (5%) on view count, and 2 (11.1%) employed both relevance and view count; the remaining
study27 employed a snowball technique, viewing a video, followed by viewing those suggested by
YouTube as being relevant to emulate a typical strategy followed by a typical user. All studies
subsequently applied content analysis by reviewing the videos and coding them according to view
count, clip length, the source of the video, comments, and viewer ratings. Table 2 shows these ele-
ments extracted from YouTube by the individual studies.
As Table 2 indicates, view count and clip length were most frequently extracted for analysis,
with 87 percent of the studies using view count and 57 percent using clip length to compare the
videos.
Two of the 18 studies supplemented this content analysis with framing analysis,16,25 a method
that analyzes the frames in the media from an affective and a cognitive perspective. One of these
studies focused on the portrayal of organ donation on YouTube16 and the second on obesity.25 For
example, in the study examining obesity, framing analysis was used to examine the videos for
causal claims and solutions pertaining to obesity. This involved analyzing the frames of the video
for behavioral, biological, and systematic causal claims and treatment options. In 89 percent of the
18 studies reviewed, two or more reviewers analyzed the content and subsequently computed the
level of inter-rater agreement (IRA). The IRA ranged from 0.6 to 1.00. The specific analytical
techniques used in the studies are seen in Table 3 below.
Seven (38.8%) studies evaluated the video content by categorizing its tone as positive, negative,
or neutral/ambiguous based on the portrayal of the issue/condition.9,12,14,15,18,24,29 Generally, videos
were categorized as positive if their approach encouraged consumers to seek an intervention, and
negative if they discouraged viewers from doing so. The study on organ donation by Tian16 applied
this analysis to the text comments, categorizing the videos studied based on the poster’s behavioral
intention.
Nine (50%) studies analyzed the content based on the source.10,12,15–18,25,26,28 The primary source
types used were organizations, including governmental agencies, non-profit, and for-profit organi-
zations; independent users; and news agencies. Two (11.1%) studies15,25 analyzed the content
focusing on videos from the media; the categories included podcasts, TV news coverage, TV docu-
mentaries, TV commercials, and university lectures. Five (27.7%) studies20,21,24,26,28 categorized
the content based on the intended audience of the video, with patients, care givers, and healthcare
professionals being the primary types.
In all, 16 (88.89%) studies9,10,12,14,17–28 assessed the credibility and veracity of the information in
the video by assessing its quality in relation to that needed for informed healthcare decision-
making. These studies focused on understanding how this information compared to reference
standards. Finally, 5 (27.78%) studies14,16,21,23,25,27 used inductive coding, meaning they categorized
the information based on the general themes identified while conducting the analysis.
Table 1. Main results of reviewed studies.

Author Objective Search criteria used Number of videos Design method Key findings
by the authors analyzed
Keelan To understand Searched using the 153 Categorized the videos • 48% of the videos portrayed a
Chalil Madathil et al.

et al.14 how immunization keywords vaccination as negative, positive, positive message, 32% negative,
information is and immunization. or ambiguous based and 20% ambiguous.
portrayed on on the portrayal of • Negative videos had a higher
YouTube. the immunization number of views and higher mean
information. ratings.
The scientific claims • 50% of the videos posted did not
made by the videos were support immunization.
classified as substantiates • Information in negative videos
or contradicts standard often contradicted the reference
immunization guide. standard.
Ache and To understand how Searched using the 146 Videos were categorized • 75% of the video clips portrayed
Wallace15 HPV is portrayed on keywords Gardasil, as TV news coverage, HPV vaccination positively and
YouTube. cervical cancer TV commercials, TV 25.3% portrayed it negatively.
vaccination, and HPV documentary, university • 32% of the video clips generated at
vaccination. lecture/educational, least one posted comment.
or YouTube member/ • 61% of the user-generated content
subscriber-created portrayed HPV negatively.
content. • Video clips portraying HPV
Comments categorized vaccination negatively were
as positive, negative, or significantly longer than those
neutral regarding HPV carrying a positive message.
vaccination.
Analysis conducted
to understand the
relationship between
the number of views and
user-posted comments.
(Continued)
177
 178

Table 1. (Continued)

Author Objective Search criteria used Number of videos Design method Key findings
by the authors analyzed
Tian16 To examine Searched using 355 The content of videos • 96% of the organ donation videos
YouTube content the keyword organ was classified as positive, had a positive valence.
on organ donation donation. negative, or neutral based • 35% were produced by individuals,
and audience on the tone of the post. 34% by non-profit organizations,
response. The comments were and 7% were television programs
coded as positive, featuring interviews with
negative, or neutral representatives from non-profit
and further analyzed to organizations.
understand if the viewer • Kidney and liver donations were
intended to sign a donor the organs most frequently
card after viewing the discussed.
video or if the viewer had • The organ donation process was
signed one. the most discussed item.
• 90% of the comments had a
positive valence.

Pandey To understand the Searched using the 142 Two researchers • 61% of videos had useful
et al.17 use of YouTube keywords swine flu, categorized the videos information about H1N1.
as an information H1N1 influenza, and as useful, misleading, or • 23% of the videos were misleading.
dissemination influenza in June 2009 news updates based on • Total viewership share of useful
mechanism by and looked for videos their characteristics. videos was 71% and that of
organizations, posted over the past Videos were also misleading videos 18%.
independent users, 3 months. categorized based on • 12% of the useful videos were
and news agencies the source: CDC, the from the CDC, with a viewership
during the initial UN and WHO, the Red share of 47%.
phase of the H1N1 Cross, news agencies, and • No differences in viewership/day
outbreak. independent users. for useful and misleading videos.
• The misleading videos portrayed
messages against vaccination and
H1N1 as a manmade conspiracy.
Health Informatics Journal 21(3)
Table 1. (Continued)

Author Objective Search criteria used Number of videos Design method Key findings
by the authors analyzed
Steinberg To understand the Three sets of 51 videos in total Two urology residents • 73% of the videos had fair to poor
et al.18 quality of prostate searches. First set analyzed, including analyzed the videos information content.
Chalil Madathil et al.

cancer–related included the search 14 PSA videos, 5 for informational and • More than two-thirds of the
information on terms prostate-specific radiotherapy videos, scientific content, rating videos were in favor of PSA testing
YouTube and to antigen, prostate and 32 surgery them as excellent, fair, or treatment of prostate cancer.
investigate the cancer and PSA and videos. and poor. In addition • Surgical videos depicted
presence of bias prostate cancer and classified, the bias as the technical aspects of a
against screening diagnosis. The second for, against, or neutral/ prostatectomy.
and treatment. set used the search unclear. • Advertisements supported robotic
terms prostate cancer prostatectomy with minimal
and radiotherapy, information about the technology.
and prostate and • Videos on PSA testing were
radiotherapy. The heavily biased, with no videos
third set involved arguing against PSA screening for
prostatectomy, prostate prostate cancer.
cancer and surgery, • YouTube may not be a reliable
and robot and prostate source of information for
cancer. consumers seeking information to
better understand the screening
for and management of prostate
cancer.
Murugiah To understand the Searched using 52 Two researchers • 69% of the videos illustrated the
et al.19 content and quality keywords CPR, categorized the source correct compression–ventilation
of CPR information Cardiopulmonary of videos as a private ratio.
available on resuscitation, BLS and agency, guideline bodies • 64% gave information on the
YouTube. Basic life support. such as the Red Cross, location of chest compressions.
emergency medical • 35% provided information on the
technicians, rate of chest compressions.
(Continued)
179
Table 1. (Continued)
180

Author Objective Search criteria used Number of videos Design method Key findings
by the authors analyzed
CPR instructors and • 40% gave information on the depth
physicians, individuals of chest compressions.
with unspecified • Videos omitted several important
credentials and news steps. Several vital steps were
programs. depicted incorrectly.
The content of the • The best videos were not the
videos was ranked on most viewed ones.
a scale from 0 to 8,
based on the accuracy of
demonstration.
Knosel and To understand the Searched using the 60 Videos were analyzed • The quality of information
Jung20 content and bias of keyword orthodontist by an orthodontist, a provided in the videos was
orthodontic videos OR orthodontic OR non-dentist adult, and an low; the majority were pro-
on YouTube. In orthodontics OR adolescent with no history orthodontics.
addition, to evaluate braces. of orthodontic treatment. • A majority of the videos were
the changes in the The videos scored from user-generated content.
attitude of a non- 1 to 10 based on the • The probability of a video being
dentist adult and an information available on an advertisement was high when
adolescent toward orthodontics classified sorted by relevance.
orthodontics. based on the intention
of the video, its source,
and the presence of bias
(whether orthodontics
was portrayed positively
or negatively).
Fat et al.9 To understand Searched using the 28 Two researchers • 60%–64% of the videos showcased
the quality and keywords infantile analyzed the videos and infantile spasms.
veracity of videos spasm, spasm, epileptic rated their technical • 71% of the videos portrayed
on infantile spasms spasm, and West quality, diagnosis of good clinical examples of infantile
(childhood epileptic syndrome. infantile spasms, and the spasms.
encephalopathy) • The search term played a key role
Health Informatics Journal 21(3)

to investigate if in finding a good quality video.

YouTube could
Table 1. (Continued)

Author Objective Search criteria used Number of videos Design method Key findings
by the authors analyzed
be used as an potential of the video • Educating patients on search
educational platform to be used as a teaching strategies and selection practices
Chalil Madathil et al.

for both parents and resource. is vital to ensure the veracity and
medical staff. quality of information.
Gooding To understand Searched using the 32 The videos were analyzed • 37% of the videos had sessions
and the informational keywords music using a rubric designed to with older adults; 33% were
Gregory21 content of music therapy and music evaluate therapy specific videos for special populations, and
therapy-related therapy session information. In addition, 15% were medical music therapy
videos on YouTube. the top five videos sessions while 60% of the videos
found when sorted by were documentaries produced
number of views were by educational institutions,
further analyzed for organizations, and clinics.
their conformance to • 20% of the top videos fell short of
the professional level of the requirements mandated by the
practice competencies AMTA.
suggested by the AMTA. • None of the video clips had
information on accreditation.
Stamelou To understand Searched using 29 Videos of patients • 66% were psychogenic and 34%
et al.22 the content keywords Dystonia, purported to have a were organic.
on movement Parkinson’s, chorea, movement disorder • 53% recommended treatment for
disorders on myoclonus, tics, and were analyzed to a psychogenic movement disorder
YouTube. tremor. understand if the and 20% recommended treatment
disorder was psychogenic for an organic movement disorder.
(disorders not related • Treatment recommendations
to a medical disorder, ranged from the use of
usually associated with immunosuppressive agents to the
psychological factors) use of craniosacral massage and
or organic (disorder herbal treatments.
caused by a detectable • The key finding was that the
physiological change in an information was misleading,
organ). providing inaccurate depictions of
181

the disease and its treatment.

(Continued)
Table 1. (Continued)
182

Author Objective Search criteria used Number of videos Design method Key findings
by the authors analyzed
Sajadi and To understand Searched all sites 30 Videos were analyzed • 45% of the videos were
Goldman23 the availability, using the keyword for informative content informative.
quality, and incontinence. on incontinence and • 64% were from healthcare
variety of urinary The initial 30 results categorized based on providers or organizations.
incontinence– analyzed further. the source—healthcare • Of the 12 commercial videos,
related content on providers, organizations, 92% were advertisements and 8%
YouTube, Facebook, patients, and commercial were for surgical procedures for
and Twitter. videos. incontinence.
Knosel and To understand the Searched using 120 videos analyzed, Two researchers analyzed • The title of the video and the
Jung24 informational value, keyword dentist OR including the first set the videos using a number of views did not have any
source, and bias of dentists OR dentist’s of 30 videos when questionnaire and ranked relationship with the informational
dentistry-related OR dental OR sorted by relevance them on a scale from 1 value of the video.
videos on YouTube. dentistry. and the most to 10 based on how they • The primary aim was
viewed under the portrayed dentistry. The entertainment, followed by
system-generated intention, credibility, and advertisement and education.
filter categories All the presence of bias also • The videos under the category
and Education. evaluated. “Education” were found to be
more than the videos available
using the category “All.”
Yoo and To understand the Searched using the 417 videos analyzed. Videos were analyzed • 37% were user-generated content,
Kim25 portrayal of obesity keywords Obesity and Every fifth video by two researchers 21% were television news,
in obesity-related Obese from the search and classified as TV and 15% were public service
videos. results was selected news coverage, public announcements.
for analysis. service announcement, • 59% made causal claims about
excerpt from TV obesity, with 74% mentioning
documentary, excerpt behavioral causes followed by
from entertainment, systemic and biological causes.
user-generated content, • Consuming unhealthy food and a
podcast, and others. In sedentary lifestyle were portrayed
addition, each media as the major causes of obesity.
frame was analyzed and
Health Informatics Journal 21(3)

classified based
Table 1. (Continued)

Author Objective Search criteria used Number of videos Design method Key findings
by the authors analyzed
on its portrayal of the • Behavioral changes were suggested
causal factors of obesity as the most common treatment.
Chalil Madathil et al.

(behavioral, biological,
and systemic). The
treatment options
provided in the video
were also analyzed.
Briones To understand Searched using the 172 The researchers • 36% were from news sources,
et al.12 the portrayal of terms HPV vaccine, employed content 13% were consumer-generated,
HPV vaccines with HPV vaccination, HPV analysis and coded the 9% were from medical centers,
emphasis on their immunization, human characteristics of the 9% were from advocacy groups,
source and tone. papilloma virus vaccine, video, with a particular 8% were from non-profit
To investigate human papilloma virus emphasis on the tone of organizations, 4% were from
the relationship immunization, Gardasil, the message and on the pharmaceutical companies, 2%
between the source and Cervarix. The content. were from governmental agencies,
and the tone of the search results sorted and 1% from a professional
video. based on relevance. association.
• 52% had a negative tone and 33% a
positive tone.
• Videos that disapproved of the
HPV vaccine were more liked than
those that approved its use.
• No significant relationship was
found between the source and tone.
Pant et al.26 To understand Searched using the 104 Videos were • Videos from professional societies
the accuracy of terms acute myocardial classified as personal and lectures were longer.
information available infarction, heart experience, news • Only 6% featured the different
on AMI. attack, acute coronary reports, professional aspects of myocardial infarction;
syndrome, and ST- groups, pharmaceutical these videos were among the least
elevation myocardial companies, lectures from viewed.
infarction. medical institutions, and
183

other media.
(Continued)
 184

Table 1. (Continued)

Author Objective Search criteria used Number of videos Design method Key findings
by the authors analyzed
All videos were analyzed • Only 17% of the videos
for the following items: emphasized preventive aspects,
The depiction of the the majority being advertisements
pathophysiology of the from companies.
disease, including its signs • 23% provided information for
and symptoms patients and 31% for healthcare
The measures to be providers.
taken in case of positive • Information available on YouTube
AMI laboratory tests and on myocardial infarction is
preventive measures with misleading and often not regulated.
an emphasis on lifestyle
management
The treatments and the
complications associated
with the disease.

Singh To understand the Searched using the 102 Two researchers • 30% of the videos were misleading.
et al.10 quality and veracity term Rheumatoid analyzed the information • The average reliability score for
of information Arthritis. and classified it as useful the useful videos was 3.2.
on rheumatoid or misleading, rating the • 73.4% were advertisements.
arthritis. quality on a 1–5 scale, • Videos from university channels
with 5 being excellent. and professional organizations
Videos categorized as disseminated useful information.
useful were further • No significant differences in the
analyzed for the veracity viewership/day between useful and
and reliability of the misleading information.
content.
Health Informatics Journal 21(3)
Table 1. (Continued)

Author Objective Search criteria used Number of videos Design method Key findings
by the authors analyzed
Chalil Madathil et al.

Stephen To understand Searched using the 22 The researchers • The age of the audience ranged
and the type of term pelvic floor searched and analyzed from 35 to 64 years.
Cumming27 information available exercise. the results using the • Music was used as background in
on YouTube on snowball technique to 36% of the clips.
pelvic floor muscle simulate real-world use.
exercise. Each video underwent
content analysis, with the
characteristics of each
being recorded in three
stages over a 7-month
period.
Clerici et To understand the Searched using 149 The videos were classified • Only 1 video had complete and
al.28 information available the terms pediatric based on the source and adequate information.
on YouTube soft-tissue sarcoma, the type of the content. • 1 video had misleading
pertaining to rhabdomyosarcoma, They were categorized information.
rhabdomyosarcoma and soft-tissue sarcoma as useful, misleading, • Only 16% of the videos depicted
and soft-tissue in children. personal experience, and useful content.
sarcomas, common not pertinent. • 83% of the videos were personal
in children. experiences.
• Videos depicting a patient story
recorded a higher number of
views than those talking about the
medical and treatment aspects of
the disease.

HPV: human papillomavirus; CPR: cardiopulmonary resuscitation; AMI: acute myocardial infarction; CDC: Centers for Disease Control and Prevention; UN: United Nations;
WHO; World Health Organization; AMTA: American Music Therapy Association; PSA: prostate-specific antigen.
185
 186

Table 2. Methodological elements.

Study Clip View Source Comments Viewer Country Upload Title Description Web page Number of Original Age/gender
length count rating of origin date of video URL subscribers or repost of audience
Keelan et al.14 • • • •
Ache and Wallace15 • • • •
Tian16 • • • • • •
Pandey et al.17 • • • •
Steinberg et al.18 • • • • • •
Murugiah et al.19 • • •
Knosel and Jung20 • •
Fat et al.9 • •
Gooding and •
Gregory21
Stamelou et al.22 •
Sajadi and Goldman23
Knosel et al.24 • •
Yoo and Kim25 • • • • • •
Briones et al.12 • • • • • • • • • •
Pant et al.26 • • • •
Singh et al.10 • • • • •
Stephen and • •
Cumming27
Clerici et al.28 • • •
Health Informatics Journal 21(3)
Table 3. Analytical techniques and resulting categories.

Study Calculated Analysis techniques Content categorization

level of
inter-rater Content Framing Based on tone Based on Based on Based on Based on Based on
agreement analysis analysis of the videos/ source intention/ media information themes
comments intended presentation quality generated during
(bias) audience content analysis
Chalil Madathil et al.

Keelan et al.14 • • • • •
Ache and Wallace15 • • • •
Tian16 • • • • • •
Pandey et al.17 • • • •
Steinberg et al.18 • • • •
Murugiah et al.19 • • • •
Knosel and Jung20 • • • • •
Fat et al.9 • • •
Gooding and • • • • •
Gregory21
Stamelou et al.22 • • •
Sajadi and Goldman23 • • • •
Knosel et al.24 • • • • • •
Yoo and Kim25 • • • • • • •
Briones et al.12 • • • • •
Pant et al.26 • • • • •
Singh et al.10 • • • •
Stephen and • • •
Cumming27
187

Clerici et al.28 • • • • •
188 Health Informatics Journal 21(3)

Characteristics of information available on YouTube

The 18 articles reviewed fell into two primary categories: those focusing on specific healthcare
issues/conditions and those investigating health care–related videos related to diagnosis, treatment,
and disease management procedures.

Studies focusing a health issue. The five studies focusing on specific healthcare concerns analyzed
videos on vaccinations (n = 3), organ donation (n = 1), and obesity (n = 1). Keelan et al.14 found
that 32 percent of the videos they analyzed were not supportive of immunization and that these
videos received a higher mean star rating and more views than the supportive ones.14 More impor-
tantly, these videos contained information that contradicted the reference standards. Ache and Wal-
lace15 found that 25.3 percent of the videos they analyzed were not supportive of the vaccination
for human papillomavirus (HPV).15 The number of views for videos supporting and those not
supporting this vaccine were similar, suggesting that the chance of an information seeker finding
contradictory information is high, a problem compounded by the view comments, which contained
biased and inaccurate information. Briones et al.’s12 study focusing on the sources of HPV videos
suggested that the majority originated from news sources, followed by user- and organization-
generated content. However, the results of this study, conducted 3 years after that of Ache and
Wallace,15 found that 51.7 percent of the videos expressed negative attitude toward the HPV vac-
cine. This study did not find any significant differences between the source and the number of
views and the like/dislike rating. Videos with a negative tone and those that disapproved of the
vaccine were more liked than the ones that approved the vaccine.
In contrast, Tian16 found that the healthcare issue of organ donation was in general positively
portrayed: 95.8 percent of the videos he analyzed portrayed it positively and 2.5 percent portrayed
it negatively. In all, 92 percent of the comments analyzed were positive and 5.7 percent were nega-
tive. The negative video portrayals and viewer comments focused on the corruption involved with
this procedure, the high cost of the process, and the resulting unfair advantage for the wealthy. One
potential explanation for the high incidence of positive videos was that a large number of them
were posted by entities promoting organ donation. Specifically, 34.4 percent of the videos were
generated by organizations, 6.5 percent by TV interviews, and 35.2 percent by individuals and
students who support organ donation.
Yoo and Kim25 investigated the portrayal of obesity, finding that 36.5 percent (n = 417) of the
videos were user-generated, 20.6 percent TV news stories, and 14.9 percent public service
announcements. More than half, 59 percent, analyzed the causes of obesity, with the primary ones
being behavioral causes (74%) followed by systematic (36.6%) and biological/genetic (17.5%)
causes. The solution options found exhibited a similar trend, advocating behavioral changes, fol-
lowed by systematic solutions and medical treatment.

Studies investigating videos related to diagnosis, treatment, and disease management procedures. In all,
9 of the articles reviewed focused on assessing the veracity of the information in the videos
researched, which included a range of health problems and one exploring music therapy. For exam-
ple, Steinberg et al.18 examined the portrayal of prostate cancer and found that the informational
content was fair or poor in 73 percent (n = 51) of the videos with no significant differences related
to video content or the number of views. Though they found support for prostate-specific antigen
(PSA) testing, surgery, or radiotherapy in 69 percent of the videos, the informational content was
of moderate quality.
Murugiah et al.’s19 study, focusing on videos related to cardiopulmonary resuscitation (CPR),
found that 88.5 percent (n = 52) of the videos were meant for lay rescuers and the remaining for
Chalil Madathil et al. 189

healthcare professionals. In all, 48 percent of total videos analyzed came from people with unknown
credentials. Though agencies such as the American Heart Association had YouTube channels, they
provided few videos showcasing the CPR technique. One of the important findings of this study
was that several videos skipped important steps in the CPR process and performed some steps
incorrectly. In addition, similar to the study by Shepherd29, no significant differences were found
in the number of views per day between the inaccurate and accurate CPR demonstration videos.
Stamelou et al.’s22 studies on the information quality of YouTube content on movement disor-
ders found a trend similar to the findings of Shepherd29 and Murugiah et al.19 suggesting that the
information was misleading, providing inaccurate depictions of movement disorder conditions and
their treatment.
This situation was also observed in videos on incontinence, with Sajadi and Goldman23 finding
under half (46%) of the videos contained useful information. Of this useful information, 64 percent
came from healthcare professionals or organizations. This study also found a proliferation of
advertisements from for-profit companies, with 92 percent of incontinence-related videos promot-
ing incontinence briefs or pads. In contrast, the content analysis of the videos on infantile spasms
conducted by Fat et al.9 suggested that 56 percent of the videos depicted high-quality videos, cor-
rectly diagnosing the condition in 60–64 percent of them. Fat et al.9 determined that the retrieval of
the videos was dependent on the search term used.
Pant et al.26 examined YouTube content for acute myocardial infarction (AMI). They found that
anecdotal videos from patients and videos from professional organizations primarily focused on
the symptoms, signs, and treatment of a heart attack, while videos from non-teaching hospitals and
graphic representations concentrated on the pathophysiology of the disease. The personal videos
featured individual experiences of the symptoms and signs of the disease (58%), followed by meth-
ods for prevention and treatment (45%) and perspectives on the news reports related to newer
diagnostic techniques (9%). Videos categorized as personal experiences recorded the maximum
number of dislikes. Out of 104 videos examined by Pant and colleagues, only 7 videos discussed
the need for immediate treatment when suspecting a myocardial infarction. Six of these came from
professional organizations.
In 2012, Clerici et al.28 analyzed videos on rhabdomyosarcoma. They found that 82.5 percent
(n = 149) of the videos involved personal experience, and only 16.7 percent of all the videos
included useful information. Only one, a video created by a physician, was completely accurate.
Approximately 7 percent of the videos provided information about rhabdomyosarcoma itself, 7.3
percent discussed treatment for it, and the rest depicted reactions to the loss of a person due to the
disease. This study concluded that there was a lack of quality videos on YouTube featuring infor-
mation on the various aspects of the disease including its treatment.
Singh et al.,10 studying videos related to rheumatoid arthritis, reported that 54.9 percent (n =
102) of the videos were found useful and 30.4 percent were misleading, including misinformation
about the causal mechanism of the condition and the promotion of unscientific therapies. In addi-
tion, 19 percent of these misleading videos portrayed the standard treatment methods as lethal and
pathogenic. About one-third (36.3%) of the videos were from independent users, followed by for-
profit organizations (22.5%), government entities/schools/professional organizations (21.5%), and
health information websites (19.6%). Notably, 73.9 percent of the videos from medical advertise-
ments and for-profit organizations were misleading. This study also found that there were no dif-
ferences in number of views between the useful and misleading videos, suggesting a high
probability for a lay user to encounter misleading information.
Finally, the study focusing on music therapy videos on YouTube conducted by Gooding and
Gregory21 found that although the therapy videos had adequate visual and audio fidelity, they
lacked effective narrative description and identification information. One of the top five music
190 Health Informatics Journal 21(3)

therapy videos exhibited poor quality according to the standards set by American Music Therapy
Association (AMTA).

Discussion
In an effort to make informed healthcare decisions, patients are increasingly turning to the Internet
to better understand their medical conditions and treatments. As seen in this review, YouTube hosts
videos providing information on the pathogenesis, diagnosis, treatment, and prevention of various
health conditions. The majority of the videos reviewed in the articles cited were narrative experi-
ences posted by individual users. This review of 18 articles provides six insights on the character-
istics of health-related information available on YouTube: (1) YouTube hosts health care–related
communication and health consumers are viewing this information; (2) public service announce-
ments from organizations, documentaries, and TV shows, and user-generated anecdotal content in
which users discuss their perspectives and their experiences were the most commonly found
sources; (3) misleading information is found on YouTube, and the probability of healthcare con-
sumers encountering such material during the information-seeking process is high; (4) reliable
postings from government/professional organizations are available; (5) overall, there was little
difference between the frequency of viewings between misleading and accurate posts; and (6) there
has been little research on developing interventions to increase the ease with which users can find
useful healthcare information on YouTube.
Three major safety concerns were identified when consumers use information obtained from
YouTube for health care decision-making: (1) YouTube is used as a medium for promoting unsci-
entific therapies that are yet to be approved by the appropriate agency,10 (2) YouTube contains
information contradicting reference standards/guidelines, and (3) YouTube has the potential to
change the beliefs of patients about controversial topics such as vaccinations.
Studies have suggested that pharmaceutical companies and for-profit institutions have a pres-
ence on YouTube and are increasingly using it to advertise their products.10,26 YouTube also con-
tains information on the use of drugs23 that are yet to be approved by the Food and Drug
Administration and that may be dangerous when used without medical supervision. Companies are
using YouTube portal to reach health consumers and circumvent government regulations.30–32
YouTube users can create and upload videos to express their opinions on healthcare topics.
Many of these videos contain information that negatively portrays public health interventions. The
results of content analyses suggested that the majority of video clips addressing vaccination por-
trayed it negatively, with the negative portrayal videos receiving a higher number of views than the
positive ones.14 The results of early studies conducted by Ache and Wallace15 in 2008 found that 32
percent of the videos on HPV vaccination were negative portrayals. A more recent study conducted
by Briones et al.12 found 51.7 percent of the videos on HPV vaccine portrayed it negatively. This
suggests an increased proliferation of content having a negative tone over the past few years. In
addition, with all three studies12,14,15 suggesting that the negative videos had a higher average num-
ber of likes than their positive counterparts, the probability of a lay user perceiving such videos as
the ones they should watch may significantly reduce the effectiveness of health campaigns.33
The educational value of showing videos providing instruction on healthcare procedures has
been demonstrated for both professionals and laypersons.45 People rely on YouTube to find such
demonstration videos and learn specific procedures such as CPR,19 pelvic floor muscle exercises,27
and music therapy.21 Video-based self-instruction, that is, learning from a video depicting a proce-
dure, has been found to be an effective way to understand a method.34,35,42 Hence, it is important
that video content depicting such procedures is accurate.44
Chalil Madathil et al. 191

The typical information available from governing bodies included information on symptoms,
treatment and preventive methods, and risk factors. Some organizations interviewed patients who
narrated their experiences with a disease/condition. Singh et al.10 recommend that government,
professional organizations, and healthcare professionals actively participate on YouTube by devel-
oping and uploading such videos to YouTube, as studies have identified that the information pro-
vided by such sources is trustworthy. YouTube can become a powerful information dissemination
platform if healthcare professionals and organizations contribute to it.
Only minimal barriers can realistically be applied to video uploads due to the nature of such
video hosting portals as YouTube. However, there is a need to develop better algorithms and design
interfaces to provide results’ information that is accurate. In addition, integrating verified informa-
tion available from federal agencies such as MedlinePlus, the medical information glossary main-
tained by the National Library of Medicine, and public report cards generated by federal entities
such as Medicare (e.g. Medicare Hospital Compare) might increase the trustworthiness and verac-
ity of the information available. Incorporating features such as crowdsourcing, whereby current
YouTube users are encouraged to report inaccurate and misleading information, could be another
strategy for preventing the spread of misinformation. Interfaces that integrate YouTube videos with
the references from clinical research studies for the claims made could also enhance the dissemina-
tion of accurate information.
Anecdotal information from patients is becoming increasingly available to the public in the
form of YouTube videos and as discussion posts on peer-support groups.36 Studies have indicated
that when making treatment choices, such anecdotal information is heavily weighed by the health
consumers.37 Hence, there is also a need to integrate consumers’ narrative accounts into the
decision-making process by presenting anecdotal information in ways that complement the quan-
titative quality information provided by federal entities.
Seeking, filtering, and integrating useful, trustworthy, and valid sources of health information,
a complex cognitive activity in itself is becoming increasingly difficult, compounded by the
increasing amount of medical information available. Healthcare consumers need to develop skills
for accessing, comprehending, and effectively using the information available on the Internet. In
addition, as consumers use this information as a source for making decisions, issues such as trust
and credibility become important considerations. As this review suggests, there is a need to under-
stand how consumers make health care–related decisions and to identify effective ways for dis-
seminating trustworthy information on the Internet so that it becomes an effective part of their
healthcare decision-making process.

Conclusion
The Internet provides many opportunities for consumers to gain information on health care. These
resources can provide novel ways for consumers to gain information and share their experiences of
investigation, diagnosis, and treatment of disease.38–41 Studies on the use and implications of
YouTube for healthcare communication are still in the nascent stage. As seen from this review,
YouTube holds a vast amount of data pertaining to health care. Some of this information is mis-
leading or incorrect. With the results of recent surveys suggesting that YouTube is ranked as the
third most accessed website worldwide, the probability of disseminating such misleading informa-
tion to healthcare consumers is high and can have catastrophic implications. There is a significant
lack of research on developing interventions for the effective dissemination of YouTube videos for
healthcare communication.
YouTube could be used as a very effective information resource if guided search practices are
followed. Interventions need to be developed to assure that consumers are able to critically
192 Health Informatics Journal 21(3)

assimilate the information posted on YouTube with more authoritative information sources to make
effective healthcare decisions. To address this issue, we call for further cross-disciplinary research
focusing on human factors engineering and user-centered design to develop interventions that sup-
port effective filtering and integration of useful, trustworthy, and valid sources of health informa-
tion on social media, including YouTube.

Acknowledgements
All the authors have contributed to the following: (1) the conception and design of the study, or acquisition of
data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intel-
lectual content; and (3) have made final approval of the version to be submitted. All persons who qualify as
authors have been listed in this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publica-
tion of this article.

Funding
The authors received no financial support for the research, authorship, and/or publication of this article.

References
1. YouTube at five- 2 bn views a day. London, 2012, http://news.bbc.co.uk/2/hi/technology/8676380.stm
2. YouTube statistics. California, 2012, http://www.viralblog.com/research-cases/youtube-statistics/
3. Atkinson NL, Saperstein SL and Pleis J. Using the internet for health-related activities: findings from a
national probability sample. J Med Internet Res 2009; 11(1): e4.
4. Rutten LJ, Squiers L and Hesse B. Cancer-related information seeking: hints from the 2003 Health
Information National Trends Survey (HINTS). J Health Commun 2006; 11 (Suppl. 1): 147–156.
5. Fox S. Online health search 2006. Pew Internet & American Life Project, Washington, DC, October
2006.
6. Fox S and Purcell K. Chronic disease and the internet. Pew Internet & American Life Project, Washington,
DC, March 2010.
7. Fox S and Jones S. The social life of health information (Americans’ pursuit of health takes place within a
widening network of both online and offline sources). Pew Internet & American Life Project; California
HealthCare Foundation, Washington, DC, June 2009.
8. Fox S. The engaged e-patient population: people turn to the Internet for health information when the
stakes are high and the connection fast. 2008, http://www.pewinternet.org/Reports/2008/The-Engaged-
Epatient-Population.aspx
9. Fat MJ, Doja A, Barrowman N, et al. YouTube videos as a teaching tool and patient resource for infantile
spasms. J Child Neurol 2011; 26(7): 804–809.
10. Singh AG, Singh S and Singh PP. YouTube for information on rheumatoid arthritis—a wakeup call?
J Rheumatol 2012; 39(5): 899–903.
11. Lewis SP, Heath NL, Sornberger MJ, et al. Helpful or harmful? An examination of viewers’ responses
to nonsuicidal self-injury videos on YouTube. J Adolescent Health 2012; 51(4): 380–385.
12. Briones R, Nan X, Madden K, et al. When vaccines go viral: an analysis of HPV vaccine coverage on
YouTube. Health Commun 2012; 27(5): 478–485.
13. Allen A, Nguyen B, Nagalla R, et al. Social media and the IUD—a YouTube content analysis.
Contraception 2012; 86(3): 316.
14. Keelan J, Pavri-Garcia V, Tomlinson G, et al. YouTube as a source of information on immunization: a
content analysis. JAMA 2007; 298(21): 2482–2484.
15. Ache KA and Wallace LS. Human papillomavirus vaccination coverage on YouTube. Am J Prev Med
2008; 35(4): 389–392.
Chalil Madathil et al. 193

16. Tian Y. Organ donation on web 2.0: content and audience analysis of organ donation videos on YouTube.
Health Commun 2010; 25(3): 238–246.
17. Pandey A, Patni N, Singh M, et al. YouTube as a source of information on the H1N1 influenza pandemic.
Am J Prev Med 2010; 38(3): e1–e3.
18. Steinberg PL, Wason S, Stern JM, et al. YouTube as source of prostate cancer information. Urology
2010; 75(3): 619–622.
19. Murugiah K, Vallakati A, Rajput K, et al. YouTube as a source of information on cardiopulmonary
resuscitation. Resuscitation 2011; 82(3): 332–334.
20. Knosel M and Jung K. Informational value and bias of videos related to orthodontics screened on a
video-sharing web site. Angle Orthod 2011; 81(3): 532–539.
21. Gooding LF and Gregory D. Descriptive analysis of YouTube music therapy videos. J Music Ther 2011;
48(3): 357–369.
22. Stamelou M, Edwards MJ, Espay AJ, et al. Movement disorders on YouTube—caveat spectator. N Engl
J Med 2011; 365(12): 1160–1161.
23. Sajadi KP and Goldman HB. Social networks lack useful content for incontinence. Urology 2011; 78(4):
764–767.
24. Knosel M, Jung K and Bleckmann A. YouTube, dentistry, and dental education. J Dent Educ 2011;
75(12): 1558–1568.
25. Yoo JH and Kim J. Obesity in the new media: a content analysis of obesity videos on YouTube. Health
Commun 2012; 27(1): 86–97.
26. Pant S, Deshmukh A, Murugiah K, et al. Assessing the credibility of the “YouTube approach” to health
information on acute myocardial infarction. Clin Cardiol 2012; 35(5): 281–285.
27. Stephen K and Cumming GP. Searching for pelvic floor muscle exercises on YouTube: what individuals
may find and where this might fit with health service programmes to promote continence. Menopause
Int 2012; 18(3): 110–115.
28. Clerici CA, Veneroni L, Bisogno G, et al. Videos on rhabdomyosarcoma on YouTube: an example of the
availability of information on pediatric tumors on the web. J Pediatr Hematol Oncol 2012; 34(8): e329–e331.
29. Shepherd S. Marketing. The dawn of the fan. Health Serv J 2008; 21: 20–22.
30. Elkin L, Thomson G and Wilson N. Connecting world youth with tobacco brands: YouTube and the
internet policy vacuum on web 2.0. Tob Control 2010; 19(5): 361–366.
31. Freeman B and Chapman S. Is “YouTube” telling or selling you something? Tobacco content on the
YouTube video-sharing website. Tob Control 2007; 16(3): 207–210.
32. Kim K, Paek HJ and Lynn J. A content analysis of smoking fetish videos on YouTube: regulatory impli-
cations for tobacco control. Health Commun 2010; 25(2): 97–106.
33. Lau AY, Gabarron E, Fernandez-Luque L, et al. Social media in health—what are the safety concerns for
health consumers? HIM J 2012; 41(2): 30–35.
34. Lynch B, Einspruch EL, Nichol G, et al. Effectiveness of a 30-min CPR self-instruction program for lay
responders: a controlled randomized study. Resuscitation 2005; 67(1): 31–44.
35. Jones I, Handley AJ, Whitfield R, et al. A preliminary feasibility study of a short DVD-based distance-
learning package for basic life support. Resuscitation 2007; 75(2): 350–356.
36. Chalil Madathil K, Greenstein JS, Neyens DM, et al. An investigation of the informational needs of ovar-
ian cancer patients and their supporters. In: Proceedings of the human factors and ergonomics society’s
international annual meeting, San Diego, CA, 30 September–4 October 2013.
37. Fagerlin A, Wang C and Ubel PA. Reducing the influence of anecdotal reasoning on people’s health care
decisions: is a picture worth a thousand statistics? Med Decis Making 2005; 25(4): 398–405.
38. Chalil Madathil K, Koikkara R, Obeid J, et al. An investigation of the efficacy of electronic consenting
interfaces of research permissions management system in a hospital setting. Int J Med Inform 2013;
82(9): 854–863.
39. Sanderson IC, Obeid JS, Chalil Madathil K, et al. Managing clinical research permissions electronically:
a novel approach to enhancing recruitment and managing consents. Clin Trials 2013; 10(4): 604–611.
40. Obeid J, Gerken K, Chalil Madathil K, et al. Development of an electronic research permissions manage-
ment system to enhance informed consents and capture research authorizations data. In: Proceedings of
the AMIA clinical research informatics summit, San Francisco, CA, 18–22 March 2013.
194 Health Informatics Journal 21(3)

41. Chalil Madathil K, Koikkara R, Dorlette-Paul M, et al. An investigation of format modifications on the
comprehension of information in consent form when presented on mobile devices. In: Proceedings of the
human factors and ergonomics society 56th annual meeting, Boston, MA, 22–26 October 2012.
42. Lambert BG, Mushtaq I, Elliot M, et al. Surgical safety checklists. Watch how Great Ormond Street uses
WHO checklist on YouTube. BMJ 2009; 338: b518.
43. Hayanga AJ and Kaiser HE. Medical information on YouTube. JAMA 2008; 299(12): 1424–1425, author
reply 1425.
44. Kapp JM, LeMaster JW, Lyon MB, et al. Updating public health teaching methods in the era of social
media. Public Health Rep 2009; 124(6): 775–777.
45. Haines SL and Van Amburgh JA. A vidcasting project to promote the pharmacist’s role in public health.
Am J Pharm Educ 2010; 74(6): 97.