G Model JVAC 12467 16

ARTICLE IN PRESS
Vaccine xxx (2011) xxxxxx

Contents lists available at SciVerse ScienceDirect

Vaccine
journal homepage: www.elsevier.com/locate/vaccine

1 2

Sorting through search results: A content analysis of HPV vaccine information online
Q1

3 4 5

Kelly Madden , Xiaoli Nan, Rowena Briones, Leah Waks

Q2 Department of Communication, University of Maryland, College Park, MD, United States

6 7 8 9 10 11

a r t i c l e

i n f o

a b s t r a c t
Introduction: Surveys have shown that many people now turn to the Internet for health information when making health-related decisions. This study systematically analyzed the HPV vaccine information returned by online search engines. HPV is the most common sexually transmitted disease and is the leading cause of cervical cancers. Methods: We conducted a content analysis of 89 top search results from Google, Yahoo, Bing, and Ask.com. The websites were analyzed with respect to source, tone, information related to specic content analyzed through the lens of the Health Belief Model, and in terms of two content themes (i.e., conspiracy theories and civil liberties). The relations among these aspects of the websites were also explored. Results: Most websites were published by nonprot or academic sources (34.8%) and governmental agencies (27.4%) and were neutral in tone (57.3%), neither promoting nor opposing the HPV vaccine. Overall, the websites presented suboptimal or inaccurate information related to the ve behavioral predictors stipulated in the Health Belief Model. Questions related to civil liberties were present on some websites. Conclusion: Health professionals designing online communication with the intent of increasing HPV vaccine uptake should take care to include information about the risks of HPV, including susceptibility and severity. Additionally, websites should include information about the benets of the vaccine (i.e., effective against HPV), low side effects as a barrier that can be overcome, and ways in which to receive the vaccine to raise individual self-efcacy. 2011 Published by Elsevier Ltd.

Article history: Received 1 June 2011 Received in revised form 17 August 2011 Accepted 10 October 2011 Available online xxx Keywords: HPV vaccine Internet Websites Health belief model

12 13 14 15 16 17

18

1. Introduction Genital human papillomavirus (HPV) is the most common sexually transmitted infection [1]. HPV is the leading cause of cervical cancers, a major health concern for women. Approximately 12,000 women are diagnosed with cervical cancer each year in the United States alone, with over a million aficted with the disease worldwide [1]. Two vaccines have been approved to prevent the spread of HPV. Gardasil, licensed in 2006, and Cervarix, approved in 2009, are available for girls as young as 9, recommended for females ages 11 and 12, and encouraged for women ages 1326 [2,3]. Gardasil is also available for males ages 926. Although HPV vaccines have been shown to be highly effective [4], vaccine uptake rate among young females in the United States is still low. According to the 2009 National Immunization Survey for Teens, approximately 44% of adolescent females received the rst dose of the HPV vaccine, with only 27% receiving all three doses [5].

19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

Corresponding author at: Department of Communication, 2130 Skinner Building, University of Maryland, College Park, MD 20742-7635, United States. Tel.: +1 570 660 6743; fax: +1 301 314 9471. E-mail addresses: kmadden@umd.edu, kkm kelly@hotmail.com (K. Madden). 0264-410X/$ see front matter 2011 Published by Elsevier Ltd. doi:10.1016/j.vaccine.2011.10.025

To understand why people get or do not get the vaccine, it is important to ascertain what information is available to them in decision-making. Approximately 80% of American adults are searching for health information online, making it the third most popular online activity [6]. People start with search engines when looking for health information online rather than navigate directly to medical portals or sites of medical societies [7]. Additionally, the Internet was named as one of the most preferred places to get information about HPV vaccines [8]. This study then seeks to understand the nature of online messages people are exposed to when they use search engines to nd information about HPV vaccines. Research on vaccine information online in general has uncovered a great deal of opposition to vaccines. The materials on anti-vaccine websites focused on the danger and ineffectiveness of vaccines and also included information related to civil liberties (e.g., vaccination mandate violates parental rights) and conspiracy theories (e.g., vaccines are a hoax) [9]. Other misinformation that exists on anti-vaccine websites includes incorrect interpretations of scientic reports [9,10]. Additionally, research has found that using different keywords in online searches, such as vaccination versus immunization, can result in dramatically different information about vaccines [11].

34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

Please cite this article in press as: Madden K, et al. Sorting through search results: A content analysis of HPV vaccine information online. Vaccine (2011), doi:10.1016/j.vaccine.2011.10.025

G Model JVAC 12467 16 2

57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102

ARTICLE IN PRESS
K. Madden et al. / Vaccine xxx (2011) xxxxxx

Studies have also examined the nature of information specically related to the HPV vaccine presented in print news sources [1214], Internet news articles [15], social networking sites or blogs [16], and YouTube videos [17]. All these studies were conducted either before approval of the rst HPV vaccine or within two years after the rst HPV vaccine was licensed, which limits the generalizability of their results to the more recent online discourse surrounding the vaccines. More importantly, very few of the previous studies specically investigated information about the HPV vaccine contained in top ranked websites returned by online searches through popular search engines, with the exception of Habel et al. [15], who explored HPV vaccine information from top online news search engines, and Tozzi et al. [18], who investigated credibility, accessibility, design, and content categories related to information accuracy and currency. Neither of these studies provided a theoretical framework for the basis of their analysis. We seek to improve on previous research by systematically analyzing websites in terms of the presence or absence of information pertaining to important predictors of vaccine acceptance within the framework of the Health Belief Model (HBM) [19]. The HBM specically identies ve factors susceptibility, severity, benets, barriers, and self-efcacy that predict whether people will perform a particular health behavior. Perceived susceptibility addresses the extent to which the person perceives they are vulnerable to the particular health problem. If people believe they are not vulnerable to a health problem, they are less likely to act to prevent this problem [20]. Perceived severity refers to an individuals belief that not acting to prevent the health problem will lead to severe consequences. Lower perceived severity reduces intentions to enact a health behavior. In terms of perceived benets, if individuals believe that the health behavior will reduce the threat (thus providing benets to the person), they are more likely to adopt the behavior. Perceived barriers to performing the behavior also play into the decision, as those who believe the costs of adopting the behavior are too high will not perform the behavior. Finally, perceived self-efcacy refers to a persons belief in his or her ability to perform the behavior, which will positively affect motivation to perform the behavior. Using the HBM as a framework, we analyze information available on the websites returned by an online search in terms of the ve predictors of vaccine acceptance. Additionally, we examine the sources of the websites, the tones of the websites, and possible associations between the type of source and the tone of the website. We also investigate the presence of anti-vaccine themes related to conspiracy theories and civil liberties and possible associations between the type of source and mention of either anti-vaccine theme.

were unavailable at the time of coding. The remaining 89 websites constituted the nal working sample for the current study. 2.2. Coding scheme The content analysis consisted of rst coding for the source of the website (i.e., the individual or organization that published the website). Eight categories of source were included: governmental agencies, nonprot/academic organizations, pharmaceutical/forprot companies, consumer generated content, news sources, advocacy groups or professional associations, medical centers/hospitals, encyclopedic medical sites, and other. The tone of the website was categorized into neutral (i.e., does not approve or disapprove the HPV vaccine), negative (i.e., disapproves the vaccine), positive (i.e., approves the vaccine), and ambiguous (i.e., contains both disapproving and approving information). Additionally, whether the website contained reference to conspiracy theories (e.g., government/companies/doctors are only out to make money on the vaccine; vaccine is a hoax) or civil liberties (e.g., government wants to mandate the vaccine and threatens citizens freedom) was coded. Finally, we coded for HBM factors including (a) benets, specically effectiveness of the vaccine (high effectiveness, low effectiveness, or no information); (b) barriers, including physical health risks (high risk, low risk, or no information) and psychological risks, dened as the suffering of possible mental or emotional trauma or stress related to the vaccine, for example, the stigma associated with STIs or increased sexual activity/promiscuity (present or absent); (c) susceptibility, or likelihood of contracting the HPV virus (high susceptibility, low susceptibility, or no information); (d) severity, or whether HPV is linked to cervical cancer (yes or no); and (e) self-efcacy, dened as steps one could take to obtain the vaccine (present or absent). 2.3. Coding procedure The unit of analysis was the individual website. If little information was presented on the initial page, coders were instructed to go one level down, or click only on links contained within the initial page and internal to that page [22]. Coders were instructed not to click on additional links on the subsequent page (i.e. go another level down). This procedure was adopted because some websites contained very limited information on the opening page. To assess what information people would nd, coders needed to explore initial links. Two coders, who were blind to the research questions, coded all of the websites about the HPV vaccines. The coders were instructed on the denitions of the categories and their operationalizations. A third coder (one of the authors) independently coded 15% of the websites in the sample (n = 14). Intercoder reliability was assessed with Krippendorffs alpha, which was 0.95 overall. In several categories, intercoder reliability was 1.00, including effectiveness, severity, psychological risks, self-efcacy, civil liberties, and conspiracy theories. Additional intercoder reliability scores were 0.90 for source, 0.86 for susceptibility, 0.83 for tone, and 0.71 for health risks, all acceptable results considering the conservative nature of Krippendorffs alpha [23]. 2.4. Data analysis Data were analyzed using SPSS for Windows (Version 16.0; SPSS, Inc., Chicago, IL). Descriptive statistics were run to assess the frequencies of source, tone, and the factors of the HBM. Additionally, Chi-square analyses were performed to analyze relations between tone of the website and source of the website and between source of the website and HBM factors.

119 120

121

122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149

150

151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170

103

2. Methods 2.1. Website sample The most frequently used online search engines are Google, Yahoo, Bing, and Ask.com [21]. We used these four search engines on October 30, 2010 to locate websites related to the HPV vaccine. The following search terms were used: HPV vaccine, HPV vaccination, HPV immunization, human papillomavirus vaccine, human papillomavirus immunization, Gardasil, and Cervarix. Prior content analyses have used similar terms to locate information about vaccination [9] and indicated the importance of using both vaccination and immunization as keywords [11]. Previous research found that people choose to view one of the rst 10 websites returned by a search 97% of the time [7]. Therefore we retained the top 10 results from each search performed in our study. Duplicate websites were eliminated and sponsored links were not selected. As a result, a total of 96 websites were retained. Of those 96 websites, seven

104

105 106 107 108 109 110 111 112 113 114 115 116 117 118

171

172 173 174 175 176 177

Please cite this article in press as: Madden K, et al. Sorting through search results: A content analysis of HPV vaccine information online. Vaccine (2011), doi:10.1016/j.vaccine.2011.10.025

G Model JVAC 12467 16

ARTICLE IN PRESS
K. Madden et al. / Vaccine xxx (2011) xxxxxx 3

Fig. 1. Source and tone of websites.

178

3. Results Of the 89 websites coded, 31 (34.8%) websites were from nonprot or academic organizations. The second largest source category was governmental agencies, with 22 websites (24.7%). Additionally, 14 websites (15.7%) were developed by pharmaceutical companies, 11 (12.4%) website pages were published by news organizations, ve (5.6%) were consumer generated, three (3.4%) fell under the category of encyclopedic medical websites, 2 (2.2%) were published by professional organizations, and one website was Q3 developed by a medical center or hospital (see Fig. 1). The websites tone was also recorded. The majority of the websites were neutral in tone, as 51 (57.3%) of the websites were neutral toward the HPV vaccines. Additionally, 25 (28.1%) of the websites were positive in tone, or supportive of the HPV vaccines, 7 (7.9%) were negative in tone, or disapproving of the HPV vaccines, and 6 (6.7%) websites were ambiguous in tone, indicating that the site provided information both positive and negative in tone (see Fig. 2). An example of positive tone is demonstrated on a webpage from the pharmaceutical manufacturer GlaxoSmithKline: Vaccination with CERVARIX can help stop cervical cancer before it starts. Really! CERVARIX is your shot at stopping cervical cancer before it starts.

179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198

Fig. 2. Tone of websites overall.

Many leading health authorities, including the Centers for Disease Control and Prevention (CDC), support cervical cancer vaccination for girls and young women. An example of negative tone is this excerpt from a blog: The truths about Gardasil and cervical cancer are suppressed. I am not just talking about the controversial adverse reports of embolisms, strokes, motor neuron degeneration, numbness, muscle weakness, paralysis, heart disorders, skin disorders, extreme fatigue, debilitating headaches, recurring dizziness, seizures, and death. I am talking about the plain, unequivocal truth about cervical cancer and Gardasil. The blog goes on to provide information dismissing Gardasils effectiveness and the prevalence of cervical cancer. We examined whether there was a relation between tone of the website and source of the website. We excluded four source categories that had ve or fewer websites (consumer generated information, nongovernmental advocacy groups, medical centers/hospitals, and encyclopedic medical sites). The remaining four source categories were submitted to a series of Chi-square tests. A signicant relation was found between tone of the website and source of the website ( 2 (9, N = 78) = 27.970, p < 0.01). Although the majority of news sources were coded as neutral (45.5%), 27.3% of news sources were coded as negative, and 18.2% were positive in tone. In comparison, none of the governmental agencies websites or the pharmaceutical websites was negative in tone, and two (6.5%) websites posted by nonprot or academic organizations were negative in tone. Information related to tone of website by source is also included in Fig. 1. Content of the websites was assessed with respect to mentions of conspiracy theories and civil liberties and the ve factors of the Health Belief Model. Two websites (2%) made reference to possible conspiracy theories and eight sites (9%) mentioned civil liberties associated with HPV vaccination. Arguments for a conspiracy theory were evident on a webpage from Natural News: If true, this information reveals details of an enormous public health fraud being perpetrated on the American people, involving FDA ofcials, Big Pharma promoters, and even the governors of states like Texas. An example of a reference to civil liberties includes this excerpt from a webpage published by the National Conference of State Legislatures: Some people who support availability of the vaccine do not support a school mandate, citing concerns about the drugs cost, safety, and parents rights to refuse.

199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239

Please cite this article in press as: Madden K, et al. Sorting through search results: A content analysis of HPV vaccine information online. Vaccine (2011), doi:10.1016/j.vaccine.2011.10.025

G Model JVAC 12467 16 4

ARTICLE IN PRESS
K. Madden et al. / Vaccine xxx (2011) xxxxxx

or gave ways in which to obtain the vaccine and 44 (49.4%) did not provide such information (see Fig. 4). We also examined if there was any association between the type of source and specic website content through Chi-square tests. A signicant relation was found between source and self-efcacy, ( 2 (3, N = 78) = 12.913, p < 0.01). Governmental sources were the most likely to report steps to get the vaccine with 72.7% providing this information. News sources were the least likely to provide self-efcacy information, with only one of the 11 news websites providing information about steps to get the vaccine. 4. Discussion As young women and even young men and their caregivers consider whether or not to receive the HPV vaccine, it is important to understand the information available for them to make such a decision. An increasing number of people turn to the Internet for health information, and most visit top search engine sites to begin their research. The current study investigated the nature of HPV vaccine information on websites returned by searching Google, Yahoo, Bing, and Ask.com through a systematic content analysis of 89 websites, specically in the context of the Health Belief Model. Although past research has failed to provide signicant evidence of the impact of online information on health behavior [24,25], several recent studies have suggested or shown that exposure to online health information can affect behavioral intentions and actions specically in the context of vaccination [26,27], making it important to explore how websites may be used as a tool to facilitate decision-making by those considering HPV vaccination. Web-based interventions pose a series of advantages compared to ofine communication programs, including increased interactivity, more exibility, and reduced costs [28,29]. Search engine results may also provide individuals with information opposed to as well as similar to their previously held views [30], which increases the chance for attitude change. On the other hand, the web also makes it easier for negative vaccine information to circulate and undermine vaccination attitudes and behaviors, and perceptions of risks related to vaccines. Therefore, efforts need to be undertaken to counteract such negative information online. Overall, we found that the majority of websites were neutral or positive in tone, supporting the HPV vaccine, and few websites were negative in tone or in opposition to people receiving the vaccine. This nding may appear encouraging considering prior content analyses [9,10] have demonstrated that online searches for the term

264 265 266 267 268 269 270 271 272 273

274

275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305

Fig. 3. HBM content on websites. Notes: effectiveness how did the website describe the effectiveness of the HPV vaccine; health risks how did the website describe the level of physical health risks associated with receiving the vaccine; susceptibility how did the website describe ones susceptibility to HPV infection.

240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263

In terms of the ve factors of the HBM, 52 (58.4%) websites indicated that the HPV vaccine was highly effective, six (6.7%) indicated low effectiveness of the vaccine, and 31 (34.8%) did not mention effectiveness. For health risks, 11 (12.4%) indicated high health risks associated with the HPV vaccine, 32 (36%) reported low risks, and 46 (51.7%) did not mention any risks. For susceptibility, 53 (59.6%) reported high susceptibility, six (6.7%) reported low susceptibility, and 30 (33.7%) did not report susceptibility (see Fig. 3). For psychological risks, six (6.7%) of the websites relayed information about psychological risks, while 83 (93.3%) did not describe any psychological risks. One example of a psychological risk was from a nursing center resource page: Will the HPV vaccine encourage sexual activity among teenagers? Concerned Women for America (CWFA) states, Giving the vaccine to young girls before they are sexually active provides them with a false sense of security, possibly leading to risky sexual behavior that would not have occurred had the threat of cervical cancer been present. A quote by 17-yearold Simmone Leslie in the New York Amsterdam News suggests that this concern may be valid: Hopefully, it will be available [Gardasil] so you wont need parental consent because not every teenager would want their parents to know whether or not they are sexually active. In terms of severity, 66 (74.2%) described the link between HPV and cervical cancer and 23 (25.8%) failed to describe the connection. Additionally, 45 (50.6%) encouraged self-efcacy

Fig. 4. Additional HBM content on websites. Notes: psychological risks whether the website discussed any psychological risk associated with receiving the vaccine; severity whether the website reported the link between HPV infection and cervical cancer; self-efcacy whether the website discussed steps to get the vaccine.

Please cite this article in press as: Madden K, et al. Sorting through search results: A content analysis of HPV vaccine information online. Vaccine (2011), doi:10.1016/j.vaccine.2011.10.025

G Model JVAC 12467 16

ARTICLE IN PRESS
K. Madden et al. / Vaccine xxx (2011) xxxxxx 5
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437

306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371

vaccination tend to result in a fair amount of anti-vaccination information. The difference in tone may be due to greater acceptance of the vaccine or the waning of the initial worry about the untested nature of the vaccine. Differences in results of tone between this study and others could be due to procedural difference as well. We attempted to include more search terms than were typically used in prior analyses and relied on multiple popular search engines rather than a single search engine to arrive at the most comprehensive evaluation. This qualication must then be applied to all comparisons hereafter. Some may suspect that those in vaccine critical communities have found alternative forums for criticism (new media channels such as MySpace or YouTube). Some research exploring these outlets found coverage to be more positive than negative [16,17]. More recent analyses, however, indicate a greater amount of negative discourse [31]. Additional research is needed to examine social network platforms, where the individual layperson is able to engage freely in the conversation surrounding the HPV vaccine. More than a quarter of news sources were negative in tone in contrast to websites in the other source categories, which were overwhelmingly positive or neutral in tone toward the HPV vaccine. Although not surprising that the news media may take a more critical stance on the vaccine than governmental agencies or pharmaceutical websites, this negative coverage may undermine governmental efforts aimed at increasing levels of vaccination. Very few websites mentioned conspiracy theories related to governmental agencies or pharmaceutical companies, but nearly one in ten websites did mention threats to civil liberties, indicating that parents concern about mandating the HPV vaccine is a critical issue alive in the HPV vaccine discussion. This has been a continuing debate, as Gostin and DeAngelis [32] found that legislation attempts to make HPV vaccinations mandatory in schools has undermined public condence and created a backlash of concerned parents. To deal with this particular issue, Charo [33] recommended optout programs for parents, in which they can choose to have their children vaccinated without interfering with their right to attend school. For those making the decision of whether or not to get themselves or their children vaccinated, complete and thorough information about the vaccine seems imperative. HPV vaccines have been reported to be effective at preventing the virus and subsequently cervical cancer. More than half of the websites in our sample reported high effectiveness of the HPV vaccines and a third of the websites did not mention effectiveness of the HPV vaccines. Only a few reported low effectiveness. From this analysis one might expect that people searching for information online about HPV vaccines would come to the conclusion that the vaccine is effective. However, people may interpret missing information about vaccine effectiveness to be as much of a cause for concern as information indicating low effectiveness. Habel et al. [15] lamented the fact that media coverage of the HPV vaccine failed to provide accurate information about the effectiveness of the vaccine and noted incorrect information about effectiveness might lead to delays in immunization and fewer people choosing to receive the vaccine. Additionally, in our content analysis, we found 74% of websites made the connection between HPV and cervical cancer, while 26% of websites failed to provide a link. The percentage of websites making this connection is a slight increase from the percentage of online news articles (70%) establishing the link between HPV and cervical cancer in 2006 [15], but a sharp decrease from the percentage of ofine news articles and broadcasts providing the connection (99%) in 2005 and 2006 [14]. Calloway et al. [12] argued that the failure to connect cervical cancer to HPV might inuence the publics acceptance of the vaccine. Despite scientic research linking HPV with cervical cancer, a quarter of the top websites returned by

Internet searches in our content analysis did not provide readers with information about the connection. The majority of the top websites returned by search engines indicated a high level of susceptibility to HPV, but more than a third of the websites did not include information about susceptibility, despite the fact that HPV is the most common sexually transmitted disease. A survey study [34] found that nearly half of young women were unaware of the prevalence of HPV and considered themselves not at risk to contract the virus. If a third of the websites returned by top search engines fail to provide information about susceptibility, young women may continue to be misinformed. Certain types of sources provided more information than other types of sources in terms of self-efcacy. Governmental agency websites, the second largest source category, were the most likely to provide steps to get the HPV vaccine. This aligns with Tristrams [35] recommendation for HPV vaccination programs to collaborate with governmental organizations. News sources, however, continue to provide little information related to self-efcacy. It may also be important to note that few websites (5.6%) were user generated. Among the ve we located, two of the webpages were blogs that negatively portrayed the HPV vaccine, two were users responding to other users questions in online forums, and the fth webpage was simply an informational blog about the HPV vaccine. The lack of user-generated sources in the top search results was countered by the abundance of non-prot and academic sources, which, according to Tozzi et al. [18], provide more accurate and clearly written content, as well as transparency for and accountability to users. The superior quality of these sites may have led them to be more frequently visited, more likely to appear in the top search results, and thus potentially more consequential in the decision to get vaccinated against HPV. There are a few limitations of our study that might be addressed or present opportunities for future research. First, although the use of a variety of search terms may be seen as a strength of the study, all search terms used by individuals to search for information related to HPV vaccination may not have been included in the list. Surveying individuals about which search terms they actually use is an important step for future research. Additionally, our data cannot provide an analysis of the way in which people may use websites from different sources and ultimately the way in which they are inuenced by different sources. An important direction for future research, therefore, is to understand the impact of information source on how HPV vaccine information is perceived and utilized in decision-making. Furthermore, risk perceptions are key in predicting health behavior, more so than an objective account of information. Our study could not provide answers to individual reception of the presented information. However, understanding the information available that may inuence perceptions may be seen as an important formative research step. Another limitation of this research is the lack of information on when the studied websites were last updated. This information would be crucial for determining whether the HPV vaccine information presented was reective of an ongoing trend or simply a residual from past debate. Because our study was conducted after the controversy over vaccine mandates had peaked, it is possible that some of the negative discourse we discovered was an artifact from the previous debate. This study was limited to search results returned from search engines within the United States. Other studies may seek to investigate online information available about the HPV vaccine in other countries, as Tozzi et al. have done [18]. In conclusion, our analysis showed that while most information available about the HPV vaccine online is positive or neutral, negative information exists as well. Health professionals need to consider the potential impact of negative information and devise

Please cite this article in press as: Madden K, et al. Sorting through search results: A content analysis of HPV vaccine information online. Vaccine (2011), doi:10.1016/j.vaccine.2011.10.025

G Model JVAC 12467 16 6

438 439 440 441 442 443 444 445 446 447 448 449

ARTICLE IN PRESS
K. Madden et al. / Vaccine xxx (2011) xxxxxx [14] Kelly BJ, Leader AE, Mittermaier DJ, Hornik RC, Cappella JN. The HPV vaccine and the media: how has the topic been covered and what are the effects knowledge about the virus and cervical cancer? Patient Educ Couns 2009;77:30813. [15] Habel MA, Liddon N, Stryker JE. The HPV vaccine: a content analysis of online news stories. J Womens Health 2009;18:4017. [16] Keelan J, Pavri V, Balakrishnan R, Wilson K. An analysis of the human papilloma virus vaccine debate on MySpace blogs. Vaccine 2010;28:153540. [17] Ache KA, Wallace LS. Human papillomavirus vaccination coverage on YouTube. Am J Prevent Med 2008;35:38992. [18] Tozzi AE, Buonuomo PS, degli Atti MLC, Carloni E, Meloni M, Gamba F. Comparison of quality of internet pages on human papillomavirus immunization in Italian and in English. J Adolesc Health 2010;46:839. [19] Rosenstock L. Historical origins of the health belief model. Health Educ Monogr 1974;2:32835. [20] Carpenter CJ. A meta-analysis of the effectiveness of health belief model variables in predicting behavior. Health Commun 2010;25:6619. [21] Experian Hitwise. Search engine analysis. http://www.hitwise.com/ us/datacenter/main/dashboard-23984.html [accessed 14.04.11], in press. [22] Weare C, Lin WY. Content analysis of the World Wide Web: opportunities and challenges. Soc Sci Comput Rev 2000;18:27292. [23] Hayes AF, Krippendorff K. Answering the call for a standard reliability measure for coding data. Commun Methods Measures 2007;1:7789. [24] Whitten P, Buis L, Love B. An internet-based health information intervention for individuals with varying degrees of health literacy. In: International Communication Association Annual Conference. 2007. [25] Stoddard JL, Delucchi KL, Munoz RF, Collins NM, Prez Stable EJ, Augustson E, et al. Smoking cessation research via the internet: a feasibility study. J Health Commun 2005;10:2741. [26] Farbey P, Gagneur A, Pasquier JC. Determinants of A (H1N1) vaccination: cross-sectional study in a population of pregnant women in Quebec. Vaccine 2011;29:18249. [27] Betsch C, Renkewitz F, Betsch T, Ulshfer C. The inuence of vaccine-critical websites on perceiving vaccination risks. J Health Psychol 2010;15:44655. [28] Lustria M, Brown L, Davis R. 10 years of consumer health informatics: what have we learned about how to design successful web-based interventions? National Communication Association Annual Conference. 2007. [29] Cassell MM, Jackson C, Cheuvront B. Health communication on the internet: an effective channel for health behavior change? J Health Commun 1998;3:719. [30] Lev-On A. The democratizing effects of search engine use: on chance exposures and organizational hubs. In: Spink A, Zimmer M, editors. Information science and knowledge management. Heidelberg, Berlin: Springer; 2008. p. 13549. [31] Author. When vaccines go viral: an analysis of HPV vaccine coverage on YouTube. Health Commun, in press. [32] Gostin LO, DeAngelis CD. Mandatory HPV vaccination. J Am Med Assoc 2007;297:19213. [33] Charo A. Politics, parents, and prophylaxis mandating HPV vaccination in the United States. N Engl J Med 2007;356:19058. [34] Wong LP, Sam IC. Ethnically diverse female university students knowledge and attitudes toward human papillomavirus (HPV). HPV vaccination and cervical cancer. Eur J Obstetr Gynecol 2010;148:905. [35] Tristram A. HPV information needs. Best Pract Res Clin Obstetr Gynecol 2006;20:26777.
483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534

ways to counteract false or negative statements, including information related to civil liberties. The negativity of news stories, in particular, is a cause for concern and may indicate the need for better training of journalists in relaying vaccine information. Additionally, the websites we studied presented suboptimal or inaccurate information related to the ve behavioral predictors stipulated in the HBM. Health professionals designing webpages with the aim of increasing HPV vaccination uptake should focus on providing sufcient and accurate information related to susceptibility, severity, benets, barriers, as well as self-efcacy, all of which have been shown to signicantly impact health behavior adoption. References
Q4
[1] Centers for Disease Control and Prevention. Genital HPV infection: fact sheet. http://www.cdc.gov/std/HPV/STDFact-HPV.htm [accessed 14.04.11], in press. [2] Centers for Disease Control and Prevention. HPV vaccines. http://www.cdc. gov/hpv/vaccine.html [accessed 14.04.11], in press. [3] Centers for Disease Control and Prevention FDA Licensure of Bivalent. Human papillomavirus vaccine (HPV2, Cervarix). http://www.cdc.gov/mmwr/ preview/mmwrhtml/mm5920a4.htm [accessed 14.04.11], in press. [4] Centers for Disease Control and Prevention Human papillomavirus. http://www.cdc.gov/vaccines/pubs/pinkbook/downloads/hpv.pdf [accessed 14.04.11], in press. [5] Centers for Disease Control and Prevention. Advisory committee on immunization practices. www.cdc.gov/vaccines/recs/acip/downloads/min-oct10. pdf [accessed 02.08.11], in press. [6] Fox S. (2011). Health information is a popular pursuit online. http://www. pewinternet.org [accessed 16.04.11], in press. [7] Eysenbach G, Kohler C. How do consumers search for and appraise health information on the World Wide Web? Qualitative study using focus groups, usability tests, and in-depth interviews. Br Med J 2002;324:5738. [8] Hughes J, Cates JR, Liddon N, Smith JS, Gottlieb SL, Brewer NT. Disparities in how parents are learning about the human papillomavirus vaccine. Cancer Epidemiol Biomarkers Prev 2009;18:36372. [9] Kata A. A postmodern pandoras box: anti-vaccination misinformation on the internet. Vaccine 2010;28:170916. [10] Wolfe RM, Sharp LK, Lipsky MS. Content and design attributes of antivaccination websites. J Am Med Assoc 2002;24:3245. [11] Wolfe RM, Sharp LK. Vaccination of immunization? The impact of search terms on the internet. J Health Commun 2005;10:53751. [12] Calloway C, Jorgensen CM, Saraiya M, Tsui J. A content analysis of news coverage of the HPV vaccine by U.S. newspapers, January 2002June 2005. Womens Health 2006;15:8039. [13] Hilton S, Hunt K, Langan M, Bedford H, Petticrew M. Newsprint media representations of the introduction of the HPV vaccination programme for cervical cancer prevention in the UK (20052008). Soc Sci Med 2010;70:94250.

450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482

Please cite this article in press as: Madden K, et al. Sorting through search results: A content analysis of HPV vaccine information online. Vaccine (2011), doi:10.1016/j.vaccine.2011.10.025