Public Health Nursing

0737-1209/© 2017 Wiley Periodicals, Inc.

doi: 10.1111/phn.12318

POPULATIONS AT RISK ACROSS THE LIFESPAN: POPULATION STUDIES

Using Twitter to Understand Public

Perceptions Regarding the #HPV
Vaccine: Opportunities for Public
Health Nurses to Engage in Social
Marketing
Jessica Keim-Malpass, PhD, RN,1,2 Emma M. Mitchell, PhD, RN,1 Emily Sun, BSN Student,1 and
Christine Kennedy, PhD, RN, FAAN,1,2
1
University of Virginia School of Nursing, Charlottesville, Virginia; and 2Department of Pediatrics, University of Virginia School of Medicine,
Charlottesville, Virginia

Correspondence to:
Jessica Keim-Malpass, University of Virginia School of Medicine, P.O. Box 800782, Charlottesville, VA 22908. E-mail: Jlk2t@virginia.edu

ABSTRACT Objectives: Given the degree of public mistrust and provider hesitation regarding
the human papillomavirus (HPV) vaccine, it is important to explore how information regarding
the vaccine is shared online via social media outlets. The purpose of this study was to evaluate
the content of messaging regarding the HPV vaccine on the social media and microblogging site
Twitter, and describe the sentiment of those messages. Design and Sample: This study utilized a
cross-sectional descriptive approach. Over a 2-week period, Twitter content was searched hourly
using key terms #HPV and #Gardasil, which yielded 1,794 Twitter posts for analysis. Each post
was then analyzed individually using an a priori coding strategy and directed content analysis.
Results: The majority of Twitter posts were written by lay consumers and were sharing commentary
about a media source. However, when actual URLs were shared, the most common form of share
was linking back to a blog post written by lay users. The vast majority of content was presented as
polarizing (either as a positive or negative tweet), with 51% of the Tweets representing a positive
viewpoint. Conclusions: Using Twitter to understand public sentiment offers a novel perspective to
explore the context of health communication surrounding certain controversial issues.

Key words: cancer, health communication, HPV, public health nursing practice.

Background added the HPV vaccination to the immunization

The advent of the human papillomavirus (HPV)
vaccines has been described as a profound public
health achievement, and is used for the prevention
of cervical, vulvar, vaginal, anal, penile, and
oropharyngeal cancer. The first quadrivalent ver-
sion of the HPV vaccine was approved by the US
Food and Drug Administration (FDA) in 2006 and
the newest nonavalent vaccine was approved for
use in 2015 (Holman et al., 2014). The Advisory
Committee on Immunization Practices (ACIP)
schedule for adolescent girls aged 11–12 in 2006,
and recommendations extended to males in 2011
(Markowitz et al., 2014). Despite the tremendous
public health implications for the prevention of
cancer (National Cancer Institute, 2016), uptake
and completion of the three-series HPV vaccine
among adolescents and young adults residing in the
United States has been low (CDC, 2013). It is esti-
mated that roughly 50% of young women and 30%
of young men initiate the vaccine and far fewer

1
2 Public Health Nursing
complete the three-series doses (CDC, 2013).
Healthy People 2020 objectives include increasing
HPV vaccination series completion for adolescents
ages 13–15 to 80% by the year 2020, and it remains
clear that increasing vaccination uptake will con-
tinue to be a public health priority (US Department
of Health and Human Services Immunizations and
Infectious Disease, 2016).
Research to date has assessed barriers to
uptake and has identified several key areas of possi-
ble intervention that require more immersive and
purposeful research methods specific to the local
environment, primary care infrastructure, and cul-
ture of care (Holman et al., 2014; Niccolai & Han-
sen, 2015). Identified barriers from the available
literature include: (1) health care provider (HCP)
barriers encompassing knowledge gaps, variations
in practice, inability to make a concrete clinical rec-
ommendations to parents/caregivers, and inade-
quate reimbursement infrastructure; (2)
parental/caregiver barriers including the
expressed need for more information about the vac-
cination, variations in health beliefs (inclusive of
the belief that their child does not need the vaccine
yet because they are not sexually active and the
belief that they are too young to be receiving the
vaccination), concerns about the vaccine safety pro-
file and potential for adverse events and (3) access
to care barriers including lack of insurance cover-
age, concerns over direct cost and co-pays associ-
ated with medical visits, lack of a patient-centered
medical home, sporadic contact with the health
care system, and lack of primary care infrastructure
to adequately monitor for series completion (Hol-
man et al., 2014; Niccolai & Hansen, 2015).
It is imperative to understand the overall con-
text of health communication and public health
messaging. Individuals rely on media, and social
media as important sources of health information
and public health campaigns must take into
account cultural norms, developmental considera-
tions, and how opinions and sentiments may be
shared through social media (Krieger et al., 2013;
Zhang, Gotsis, & Jordan-Marsh, 2013). Early after
the HPV vaccine was introduced, discussions sur-
rounding the vaccine were polarized with much of
the early language focused on sexually transmitted
infections among young girls, and fear centered on
vaccine initiation leading to sexual debut (Wailoo,
Livingston, Epstein, & Aronowitz, 2010). Arguably,
the early public health messaging was not as
focused on cancer prevention, and instead had
strong undertones of sexual health (Wailoo et al.,
2010).
Parental/caregiver and provider attitudes
regarding the HPV vaccine have been the focus of
many recent research and public health efforts. Even
so, current evidence suggests that tested educational
interventions do not routinely result in increases in
uptake of the vaccine (Fu, Bonhomme, Cooper,
Joseph, & Zimet, 2014). Some researchers have sug-
gested the need to have innovative public health
paradigms to sustain increases in HPV vaccine uti-
lization, including innovations in health communica-
tion (Niccolai & Hansen, 2015). One such modality is
understanding the context of HPV vaccine communi-
cation and promotion as a social marketing entity
and the importance of delivering reliable health
information to the public. Social marketing is the
“activity, set of institutions, and processes for creat-
ing, communicating, delivering, and exchanging
offerings that have value for customers, clients, part-
ners, and society at large.” (Nowak, Gellin, MacDon-
ald, Butler, & SAGE Working Group on Vaccine
Hesitancy 2015). Social marketing can be used as a
public health framework used to develop interven-
tions for the greater public health good, and involves
techniques from commercial marketing concerning
the four Ps: product, price, place, and promotion.
(Roncancio et al., 2016). To date, social marketing
has been applied to interventions focused on healthy
diets, increasing physical activity, and breastfeeding
(Roncancio et al., 2016) and only recently has been
conceptualized as a public health framework to
decrease vaccine hesitancy (Cates, Shafer, Diehl, &
Deal, 2011; Hull et al., 2014; Nowak et al., 2015;
Roncancio et al., 2016).
With the advent of multiple social media plat-
forms and user-generated content, understanding
how information regarding the HPV vaccine is gen-
erated and communicated to the general public
(i.e., promotion) is central to this goal and a neces-
sary first step to intervention development using
the social marketing framework. Twitter is an ideal
place for initial analyses because it comprises over
230 million users, provides a means of data mining
to assess sentiment of content, and is most popular
among young adults (Clark et al., 2015). Further-
more, understanding the sentiment (overall nega-
tive or positive messaging as a proxy for public
 Keim-Malpass et al.: Using Twitter to Understand Public Perceptions Regarding the #HPV Vaccine 3

perception) is critically useful for researchers and TABLE 1. Codebook/Variable List

public health nurses to understand HPV vaccine
Variable Codebook examples
messaging within the lay and social media. The aim
date and time of tweet
of this study was to evaluate the content of messag-
search type #HPV or #Gardasil
ing regarding the HPV vaccine on the social media self-described user type lay consumer, advocate/provider,
and microblogging site Twitter, and describe the political/religious affiliation,
sentiment of those messages. A secondary aim of news media organization,
this study was to explore sentiment by type of user. unofficial lay media
To the authors’ knowledge, this study provides the (i.e., blogger), and undetermined
category of tweet awareness, share research
first to describe user-generated Twitter content findings, share media source,
regarding the HPV vaccine. This study is a descrip- offer opinion, other
tive exploratory study with the intention of being sentiment positive, negative, neutral
hypothesis-generating for future social marketing URL shared
interventions pertaining to the HPV vaccine and (if applicable)
URL type from organization (i.e. American
there were no a priori hypotheses for this pilot
Academy of Pediatrics), from
project. lay media (i.e. Glamour
magazine), link to blog
(unofficial), link to scientific
Method peer-reviewed study,
journalism (digital or print
Design and sample media)
Institutional review was exempted due to the fact native tweet versus
that the data were publically archived data. A retweet
hashtags used
descriptive, cross-sectional, qualitative approach
Topsy sentiment Range from 1 to 100
was used to address the aim of this study. Topsy score at
(http://topsy.com), an analytic tool that maintains time of inquiry
all Tweets shared since 2006, was used as a search
engine for the Twitter content. Topsy also updates
the archive of Tweets in real-time and has a search primary analysis. We also assessed the Twitter pro-
function for ease of abstraction. file associated with the Tweet to determine how the
Over a 2-week period in June 2015, Topsy was originator of the message described themselves in
searched hourly for current Tweets using search their profile (if applicable).
terms #HPV and #Gardasil (the most widely used This study utilized deductive content analysis
HPV vaccine). All Tweets with those search terms (Hsieh & Shannon, 2005) to apply the a priori cod-
were analyzed for study inclusion. Criteria for study ing strategy. Abstracted Tweets were assessed indi-
inclusion were twofold: (1) content was in the Eng- vidually by the primary abstractor (E.S.) and
lish language and (2) the tweet had to focus on the assigned classifications based on the a priori cate-
vaccine, not the virus itself. A priori categories were gories following the definitions provided in the
determined based on previous research that used codebook. The primary abstractor also: (1) deter-
Twitter to determine sentiment of controversial mined if the Tweet was an overall positive, nega-
public health topics (Prabhu et al., 2014; Ram- tive, neutral message based on the content and
agopalan, Wasiak, & Cox, 2014). A codebook was what was being shared (i.e., manually assigned sen-
developed that included relevant definitions of cate- timent), (2) maintained a field notebook for daily
gories and examples of each variable collected and observations, reflexive practices, and (3) recorded
is found in Table 1. exemplar tweets. The primary abstractor also main-
Each Tweet included in the study was tained documentation on any major media outlet
abstracted (maintained in a Word document) and releasing HPV-related content, and documented
was considered a data element. If a URL linking to any practice and policy-related content to ensure
a web site was included in the primary Tweet, the that there were no changes in practice or vaccine
URL was also assessed as a component of the recommendations during this study period.
4 Public Health Nursing

Finally, the sentiment score derived from Topsy TABLE 2. Sample Characteristics
(from 0 to 100) was recorded at the time of
Frequency/percent
abstraction for each data element. Along with oper- Variable or mean/SD
ating as a search engine for archived Tweets, Topsy
Search type
simultaneously uses natural language processing to #HPV 1,341 (74.7)
deliver its own sentiment score delivered as an #Gardasil 453 (25.3)
updated analytic. Natural language processing User type
involves using an algorithm and advanced com- Lay consumer 1,331 (74.2)
puter science to include the language used in the Advocate/medical or 171 (9.5)
research affiliation
context of the data element (i.e., Tweets including a Political/religious affiliation 73 (4.1)
specific search term) to give an overall rating of the News media 126 (7.0)
sentiment of a search at any given time. The senti- Unofficial lay media 79 (4.4)
ment score ranges from 0 (most negative) to 100 (i.e. blogger)
(most positive) and are updated with each refresh Undetermined 14 (0.8)
Category
of the search term. Inclusion of Topsy sentiment
Awareness 150 (8.4)
score at time of search was used as a secondary Research findings 30 (1.7)
and exploratory data element in this analysis. Topsy Share media source 1,389 (77.4)
was created primarily for the business and market- Opinion 146 (8.1)
ing sectors and has not been validated in health Other 79 (4.4)
research. Thus, we made the decision to include the Sentiment
Positive 912 (50.8)
manual data abstraction of sentiment as our pri- Negative 781 (43.5)
mary source for outcome ascertainment. Therefore, Neutral 102 (5.8)
the Topsy sentiment score was included for URL type
exploratory descriptive analysis and to determine if Organization 140 (7.8)
there was alignment between the manually derived Lay media 724 (40.5)
Peer review article 107 (6.0)
sentiment and natural language processing-derived
Digital/print media 470 (26.2)
sentiment (i.e., Topsy score). There was one pri- N/A no URL share 351 (19.6)
mary abstractor (E.S.) but a nested sample of 10% Tweet origination
of the tweets was analyzed by a second abstractor Native tweet 1,449 (80.8)
(J.K.M.) with a 100% uniform concordance in data Retweet 341 (19.0)
classification of the assigned categories. Descriptive Topsy sentiment score 35.25 (16.31) Mean/SD
Range (2.0–71.0)
statistics of the different categories were calculated
and included mean/standard deviation and fre-
quency/percent.
blog posts) in contrast to large market journalism,
Results public health, or scientific resources. The majority
of the content was self-initiated in the form of a
A total of 1,794 Tweets were included for analysis. native tweet (as opposed to a retweet). Only 51% of
There were no major changes in policy, practice, or the Tweets shared were classified as “positive” and
regulation of the HPV vaccine during the study 43.5% of the Tweets were considered “negative”.
time period. General characteristics of the Tweets Interestingly, only 5.75% of the Tweets were classi-
can be found in Table 2. The majority of self-identi- fied as “neutral” indicating that most of the mes-
fied users were lay consumers (which included par- saging on Twitter regarding the HPV vaccine were
ents) and the second most common group of users either in support of or against the vaccine adminis-
was advocates (including medical/research person- tration. The mean Topsy sentiment score during
nel and natural health practitioners). Less than 5% the period of data abstraction was 35.25 (represent-
of the users were classified as “advocates” self-iden- ing an overall negative sentiment).
tified as nurses. Tweets were commonly written to Table 3 identifies the category of tweet by user
share a media source, but of those shared tweets, type. The largest frequency/percentage of content
the most common URL type was of lay media (e.g., shared was identified as lay consumers sharing
 Keim-Malpass et al.: Using Twitter to Understand Public Perceptions Regarding the #HPV Vaccine 5

TABLE 3. User Type by Category of Tweet

User type (n and %) Awareness Research findings Shared media source Opinion Other Total
Lay consumer 80 (6.0) 7 (0.5) 1,057 (79.4) 113 (8.5) 74 (5.6) 1,331
Advocate 39 (22.8) 3 (1.8) 124 (72.5) 5 (2.9) 0 171
Political/religious 12 (16.4) 1 (1.4) 38 (52.1) 21 (28.8) 1 (1.4) 73
News media 15 (11.9) 17 (13.5) 90 (71.4) 3 (2.4) 1 (0.8) 126
Lay media 3 (3.8) 2 (6.3) 67 (84.8) 4 (5.1) 3 (3.8) 79
Undetermined 1 (7.1) 0 13 (92.9) 0 0 14

media sources. Behind lay consumers, advocates
(public health, medical, or research affiliation) were
the next most common group and they most fre-
quently shared media sources and presented infor-
mation about HPV vaccine awareness. Table 4
describes the overall sentiment by user type and
highlights that the majority of lay consumers initi-
ated negative tweets regarding the HPV vaccine.
Not surprisingly, those who were identified as
health advocates initiated a high frequency of con-
tent that was described as “positive”. Those who
identified as having a political or religious affilia-
tion initiated a much higher frequency of negative
tweets (vs. positive content). Table 5 describes
exemplar tweets with descriptions of the user, the
written Tweet, and sentiment.
Discussion
Our study assessed real-time sentiment and user
characteristics of those tweeting about the #HPV
vaccination or #Gardasil over a 2-week period in
June 2015. The content expressed was overwhelm-
ingly polarizing and tweets represented either a
positive or negative view of the vaccine, and the
vast majority of the content was disseminated by
lay consumers. To our knowledge, this is the first
study exploring the sentiment of content focused
on the HPV vaccination.
Given the uncertainty some adolescents, young
adults and parents may feel about receiving the
HPV vaccination (Holman et al., 2014), it is vitally
important to understand the role social media sites
like Twitter have in cultivating and sharing public
health messages, as well as the potential for evi-
dence-based health communication. Because of the
140 character limit on Twitter, information is
shared in a condensed matter which makes rapid
dissemination of varying viewpoints possible. How-
ever, messages on microblogs like Twitter are often
TABLE 4. User Type by Sentiment
User type
(n and %) Positive Negative Neutral Total
Lay consumer 594 (44.7) 656 (49.3) 80 (6.0) 1,330
Advocate 163 (95.3) 6 (3.5) 2 (1.2) 171
Political/religious 8 (11.0) 63 (86.3) 2 (2.7) 73
News media 97 (77.0) 19 (15.1) 10 (7.9) 126
Lay media 44 (55.7) 31 (94.9) 4 (5.1) 79
Undetermined 6 (42.9) 6 (42.9) 2 (14.3) 14
subjective and it is easy to present erroneous infor-
mation with little ability to fact-check because of
rapid and transient dissemination as well as ease in
re-tweeting nonfact-checked information (Zhang
et al., 2013).
Most of the URL links shared on Twitter dur-
ing this study period linked back to lay media (e.g.,
blogs) and not peer-reviewed evidence-based
resources or content from trusted public health
resources (e.g., the Centers for Disease Control).
Previous research has described how information
regarding the HPV vaccine is searched for (P
ıas-
Peleteiro & Martino
n-Torres, 2013). P
ıas-Peleteiro


and Martinon-Torres (2013) conducted a study
assessing the information and recommendations
provided by Google Web SearchTM (i.e., Google) in
relation to the web searches conducted for HPV
vaccination, indications for females and males, and
possible adverse effects. They found that 72% of
websites offer information that is favorable about
the HPV vaccination, but with varying degrees of
content/detail. Furthermore, nearly 28% of the
websites contained content that was dissuasive or
against HPV vaccination, the most frequent being
lay media forums or blogs, which supports our own
exploratory findings (P
ıas-Peleteiro & Martino
n-
Torres, 2013).
While users considered “advocates” overwhelm-
ingly offered positive tweets, there is a defined need
6 Public Health Nursing

TABLE 5. Exemplar Tweets

User information
Category Tweet Sentiment
OB-GYN The Independent claims HPV vaccine unsafe. Positive
Advocate Science says The Independent is wrong.
http://wp.me/p13KVf-1iB
Mother Thousands of UK Girls Injured by HPV Negative
Lay consumer Vaccine – go.shr.lc/1I8p5ya via @EUTimes
A MUST SEE #WakeUpAmerica#tcot
#tyranny #democrats
Holistic psychiatrist What will it take? ‘Tens of thousands’ of Negative
Advocate teenage girls fall ill after routine HPV jab
http://buff.ly/1cs5lKr
Investigative journalist #SB277 Tyrannical NY legislators looking to Negative
News media force toxic HPV vaccine on 6th graders.
S509 http://j.mp/1HYLHUA
Mother Years ago I did not vaccinate my child against Positive
Lay consumer HPV (Gardasil). Today I made a difference
choice for Kid #2.#VACCINATEYOURKIDS
Christian Nurses and Midwives Concerns re HPV jab on front page Negative
Advocate @Independentind.pn/1RB6iQz I raised concern
about it 2011cmfblog.org.uk/2011/05/12/hpv. . .
@CMFethics
Physician If we get it right, we may eliminate genial Positive
Advocate warts/HPV-related cancers through herd
vaccination. Is that too much to ask? #bashh15
PhD student (cancer research) Re HPV vaccine: A girl I knew died from cervical Positive
Advocate cancer at age 20. Witnessing what she went
through, I didn’t hesitate to get vaccinated.
Mother FDA-approved Gardasil 9: Malfeasance or Negative
Lay consumer Stupidity? http://sanevax.org/fda-approved-
gardasil-9-malfeasance-or-stupidity/#.
VXSkDpSiB9w.twitter
Mother/pediatrician Latest HPV vaccine recommendations: Positive
Advocate http://aapnews.aappublications.org/content/
36/6/16.full
Actress Just for the hopper..I am not anti vaccines. . .. . . Negative
Lay consumer But that is up to a parent to decide which ones
and in which forms. . .the government. . .
. . . Should not have the right to mandate a child’s
medical path..talk about opening up a can of worms!
There ARE side effects of vaccines.
As an example: research the side effects of the vaccine
for HPV on young girls. This vaccine was pushed
heavily..find out the results
YOUR CHILD, YOUR CHOICE. . .but at least chose
after getting ALL the possible side effects..:) and
NEVER let a government mandate force you
Mother The Gardasil vaccine deregulates the immune system Negative
Lay consumer and that allows latent infections and viruses

for advocacy within professional organizations,
researchers, and public health clinicians. In this
analysis, there was a paucity of users sharing HPV
vaccine-related content that self-identified as
nurses. Public health nurses have the potential to
play vital roles in becoming influencers and cancer
prevention advocates using sustained social market-
ing strategies and mobilizing support through social
media, including Twitter. Our findings suggest that
trusted, evidence-based resources are not the most
 Keim-Malpass et al.: Using Twitter to Understand Public Perceptions Regarding the #HPV Vaccine
commonly presented and shared content regarding
the HPV vaccine on Twitter, and there is a sus-
tained need for incorporating purposeful content
that disseminates evidence and engaging with
potential stakeholders in the context of the role of
the public health nurse.
There are several limitations in this study that
should be addressed. There may be content that
was missed because it was not attributed with a
hashtag. Additionally, given the cross-sectional
approach, the overall sentiment can be shifted with
one large news event or public health campaign.
We were able to examine this and there were no
major public health events, HPV-related stories, or
practice changes during the study period time. Fur-
thermore, this study did not account for the fre-
quency of the Tweet based on user; therefore, it
was possible for multiple Tweets from the same
user to be included in this study, which may skew
the analysis. Even though we did not account for
this formally in the study, in a naturalistic setting
(i.e., what users would find if they searched for
#HPV on Twitter in real-time settings), users would
be accessing content that may include multiple
posts from the same person, so our analysis
accounts for how users would be experiencing the
context in real-time. Finally, even though a nested
sample was also analyzed for concordance, there
was only one researcher attributing sentiment for
the majority of tweets, which may lead to a biased
analysis. Future intervention development utilizing
the social marketing framework and sentiment of
public health communication would warrant a more
formal intra- and inter-rater reliability assessment.
Even so, a strength of this study was the use of a
researcher to verify tweet sentiment without relying
solely on natural language processing (i.e., Topsy
analytic). In the context of natural language pro-
cessing, slang and abbreviations can change the
sentiment and may not be recognized solely by
computational informatics. Additionally, when a
URL is shared, it is very difficult to accurately
assign sentiment without reading the article itself
to understand the context of that article. We were
able to offer information on user type, and senti-
ment by user type, which offers a unique perspec-
tive to this study and transcends other large-scale
analyses.
The Internet stands as a prominent source of infor-
mation for the general public and has implications
7
for targeted social marketing strategies, particularly
as more users engage in dialog on social media

n-Torres,
sites like Twitter (P
ıas-Peleteiro & Martino
2013). Because Twitter is such a mainstream social
media outlet with a myriad of data mining tech-
niques, it offers the opportunity to assess the
impact of purposeful social marketing activities,
health policy, and public health interventions on
sentiment over time. While the central purpose of
this study was to study sentiment and overall user
viewpoint, areas of needed future research include:
(1) further exploration to fact-check the content
presented on online forums, blog sites, and social
media for validity, (2) assessing where adolescents,
young adults, and parents/caregivers go to consume
health-related information regarding the HPV vac-
cination, (3) understanding the role of health liter-
acy and electronic health literacy in how
information is appraised, consumed, and acted
upon, and (4) the public health nursing role in
health communication specific to evidence-based
dissemination at a literacy level aimed at lay con-
sumers.
Public health nurses involved with HPV vaccine
administration or public health campaigns and
researchers involved in cancer prevention strategies
have a vested interest in understanding the public
discourse focused on HPV vaccination, particularly
as the nation struggles to maintain expected out-
comes in vaccination uptake and completion. Public
health nurses can advocate for the use of evidence-
based cancer prevention strategies by engaging with
public discourse surrounding the vaccine. Specifi-
cally, it is imperative for public health nurses to
identify themselves as nurses, and to provide access
to evidence-based practices (e.g., HPV vaccination)
and link to trusted, open-access resources that
stakeholders can engage with. Public health nurses
must be mindful of information appraisal and spe-
cial populations, such as those with limited health
literacy. Additionally, public health nurses can use
the social marketing framework to help specifically
engage with interventions to improve the use of
HPV vaccination.
Acknowledgments
We thank the UVA School of Nursing Office of Nursing Research
for their support in providing a research intern (ES).
Conflict of interest: None Declared/The Author(s) declare(s)
that there is no conflict of interest.
8 Public Health Nursing
Funding: J.K.M. and E.M.M. were financially supported
through a grant from NIH/NCI P30CA044579 (PI: Loughran; Pro-
ject PI: Keim-Malpass).
Ethical approval: The ethics committee of University of Vir-
ginia approved this as an exempt protocol IRB # 2014044300.
Contributorship: J.K.M. researched literature and conceived
the study, was involved in protocol development, gaining ethics
approval, and overall data analysis. E.S. was the lead data abstractor
and conducted the coding of all data. All authors contributed to
overall study development, reviewed and edited the manuscript,
and approved the final version of the manuscript.
References
Cates, J. R., Shafer, A., Diehl, S. J., & Deal, A. M. (2011). Evaluating
a county-sponsored social marketing campaign to increase
mothers’ initiation of HPV Vaccine for their pre-teen
daughters in a primarily rural area. Social Marketing Quar-
terly, 17(1), 4–26. Retrieved from http://doi.org/10.1080/
15245004.2010.546943
CDC (2013). National and state vaccination coverage among adoles-
cents aged 13 through 17 years—United States, 2010.
MMWR: Morbidity and Mortality Weekly Report, 62(34),
685–708.
Clark, E. M., Jones, C. a., Williams, J. R., Kurti, A. N., Nortotsky,
M. C., Danforth, C. M., et al. (2015). Vaporous marketing:
Uncovering pervasive electronic cigarette advertisements
on twitter, 1–9. http://doi.org/10.1371/journal.pone.
0157304
Fu, L. Y., Bonhomme, L.-A., Cooper, S. C., Joseph, J. G., & Zimet,
G. D. (2014). Educational interventions to increase HPV
vaccination acceptance: A systematic review. Vaccine, 32
(17), 1901–1920. Retrieved from http://doi.org/10.1016/
j.vaccine.2014.01.091
Holman, D. M., Benard, V., Roland, K. B., Watson, M., Liddon, N.,
& Stokley, S. (2014). Barriers to human papillomavirus vac-
cination among US adolescents: A systematic review of the
literature. JAMA Pediatrics, 168(1), 76–82. Retrieved from
http://doi.org/10.1001/jamapediatrics.2013.2752
Hsieh, H.-F., & Shannon, S. E. (2005). Three approaches to qualita-
tive content analysis. Qualitative Health Research, 15(9),
1277–1288. Retrieved from http://doi.org/10.1177/
1049732305276687
Hull, P. C., Williams, E. A., Khabele, D., Dean, C., Bond, B., & San-
derson, M. (2014). HPV vaccine use among African Ameri-
can girls: Qualitative formative research using a
participatory social marketing approach. Gynecologic
Oncology, 132(Suppl), S13–S20. Retrieved from http://doi.
org/10.1016/j.ygyno.2014.01.046
Krieger, J. L., Katz, M. L., Eisenberg, D., Heaner, S., Sarge, M., &
Jain, P. (2013). Media coverage of cervical cancer and the
HPV vaccine: Implications for geographic health inequities.
Health Expectations: An International Journal of Public
Participation in Health Care and Health Policy, 16(3), e1–
e12. Retrieved from http://doi.org/10.1111/j.1369-7625.
2011.00721.x
Markowitz, L., Dunne, E., Saraiya, M., Chesson, H., Curtis, C., Gee,
J., et al. (2014). Human Papillomavirus Vaccination Rec-
ommendations of the Advisory Committee on Immuniza-
tion Practices (ACIP). MMWR Recommendations and
Reports, 63(5), 1–30.
National Cancer Institute. (2016). NCI-Designated Cancer Centers
Urge HPV Vaccination for the Prevention of Cancer.
Retrieved from https://www.mdanderson.org/content/da
m/mdanderson/documents/prevention-and-screening/
NCI_HPV_Consensus_Statement_012716.pdf
Niccolai, L. M., & Hansen, C. E. (2015). Practice- and community-
based interventions to increase human papillomavirus vac-
cine coverage: A Systematic Review. JAMA Pediatrics,
6520(7), 1–7. Retrieved from http://doi.org/10.1001/ja
mapediatrics.2015.0310
Nowak, G. J., Gellin, B. G., MacDonald, N. E., Butler, R., & SAGE
Working Group on Vaccine Hesitancy (2015). Addressing
vaccine hesitancy: The potential value of commercial and
social marketing principles and practices. Vaccine, 33(34),
4204–4211. Retrieved from http://doi.org/10.1016/j.vacc
ine.2015.04.039
P
ıas-Peleteiro, L., & Martin
on-Torres, F. (2013). What about the
human papillomavirus vaccine? Dr Google Human Vacci-
nes and Immunotherapeutics, 9(8), 1712–1719.
Prabhu, V., Lee, T., Loeb, S., Holmes, J. H., Gold, H. T., Lepor, H.,
et al. (2015). Twitter response to the United States Preven-
tive Services Task Force Recommendations against screen-
ing with prostate specific antigen. BJU International, 116,
65–71. Retrieved from http://doi.org/10.1111/bju.12748
Ramagopalan, S., Wasiak, R., & Cox, A. P. (2014). Using Twitter to
investigate opinions about multiple sclerosis treatments: A
descriptive, exploratory study. F1000Research, 216, 1–9.
Retrieved from http://doi.org/10.12688/f1000research.
5263.1
Roncancio, A. M., Ward, K. K., Carmack, C. C., Mu~ noz, B. T., Cano,
M. A., & Cribbs, F. (2016). Using social marketing theory
as a framework for understanding and increasing HPV Vac-
cine series completion among Hispanic adolescents: A qual-
itative study. Journal of Community Health. Retrieved
from http://doi.org/10.1007/s10900-016-0244-0
US Department of Health and Human Services Immunizations and
Infectious Disease. (2016). Healthy People 2020 Topics &
Objectives. Retrieved from http://www.healthypeople.gov/
2020/topicsobjectives2020/objectiveslist.aspx?topicid=23
Wailoo, K., Livingston, J., Epstein, S., & Aronowitz, R. (Eds.)
(2010). Three shots at prevention: The HPV vaccine and
the politics of medicine’s simple solutions. Baltimore, MD:
The Johns Hopkins University Press.
Zhang, C., Gotsis, M., & Jordan-Marsh, M. (2013). Social media
microblogs as an HPV vaccination forum. Human Vaccines
and Immunotherapeutics, 9(November), 2483–2489.
Retrieved from http://doi.org/10.4161/hv.25599