This paper is in the following e-collections/theme issues:
Estimation of Geographic Variation in Human Papillomavirus Vaccine Uptake in Men and Women: An Online Survey Using Facebook Recruitment
1School of Public Health, Division of Epidemiology and Community Health, University of Minnesota, Minneapolis, MN, United States
2School of Public Health, Division of Biostatistics, University of Minnesota, Minneapolis, MN, United States
Erik J Nelson, MPH, PhD
School of Public Health
Division of Epidemiology and Community Health
University of Minnesota
1300 South Second Street
Minneapolis, MN, 55454
Phone: 1 1 612 624 1818
Fax: 1 1 612 624 0315
1] and is the necessary cause of cervical cancer . HPV infections are also associated with other cancers (eg, anogenital and oropharyngeal) as well as genital warts [3,4]. In total, it is estimated that 5.2% of cancers in men and women worldwide are attributable to HPV .
Two vaccinations against HPV infection are currently licensed in the United States. The vaccinations were originally licensed for use in girls, but as of October 2011, the Advisory Committee on Immunization Practices extended their recommendation of the quadrivalent vaccine to include both boys and girls aged 11 or 12 years old [6,7]. However, vaccine uptake has been far lower than expected, with only about half of eligible young women receiving at least one dose of the vaccine . Initiation of the HPV vaccine series has been shown to be higher among minority adolescent girls; however, completion of the three-dose series is substantially lower among black and Hispanic adolescent girls compared to white adolescent girls . Although male vaccination data are very limited (due to a later date of approval of the HPV vaccine for boys), racial and income differences in terms of vaccine series initiation and completion have also been observed among adolescent boys .
Previous research on HPV vaccine coverage has used publicly available data from five national health surveys (National Survey of Family Growth, National Immunization Survey [NIS]-Teen, National Health and Nutrition Examination Survey, National Health Interview Survey [NHIS], and the Behavioral Risk Factor Surveillance System) [11-15]. These surveys are designed to gather information on a variety of health topics and ask only a few questions regarding HPV vaccination. However, none of these surveys address cervical cancer screening practices and potential barriers to screening or HPV vaccine receipt. In addition, due to the small number of responses in many geographic areas, local data from these surveys are routinely suppressed and aggregated to state boundaries in order to protect the confidentiality of survey respondents, which means that variations at a local level (ie, between counties or postal codes) cannot be adequately assessed. Further, these surveys have, to date, primarily surveyed adolescent girls; HPV vaccination practice data of adolescent boys are limited .
The Internet provides a unique point of contact to reach young adults for health research. Several studies have demonstrated that Internet-based research can be used to elicit high response rates at a fraction of the cost of traditional recruitment methods [16-18]. In addition, it has been shown that when compared to in-person interviews, Internet-based surveys have the potential to reach more respondents, include otherwise inaccessible populations, and reduce bias in responses as respondents may be willing to report more sensitive information online compared to in-person interviews [19-24]. A number of studies have also shown that recruitment via Facebook (the leading social media site with more than one billion active users worldwide) can be used to enroll representative samples of the general population [16,25-30]. This combination of reach, utility, and reduced cost indicates that social media networks can be a cost-effective medium for research.
The objective of this study was to estimate HPV vaccination practices among a local population of young adult men and women in the United States using an Internet-based recruitment strategy.
Men and women from Minnesota were surveyed about their HPV vaccination practices via the Internet from November 21, 2012, through January 31, 2013. Participants were English-speaking, aged 18-30 years, had a Facebook account, and resided in the greater Twin Cities Metropolitan Area (ie, within 25 miles of downtown Minneapolis, MN). This age range was used to target men and women who were eligible to receive the HPV vaccine, participate in cervical cancer screening (women), and able to provide informed consent. The Twin Cities Metropolitan Area was selected due to the variation of HPV-related cancer incidence rates exhibited in this area during the past 15 years, the high concentration of colleges and universities, and the large population of 18-30 year olds residing in this area . The University of Minnesota Institutional Review Board approved this study.
Facebook Recruitment Campaign
Participants were recruited online via Facebook advertisements (Figure 1). Tailored advertisements were used to target Facebook users who had profiles that matched the study inclusion criteria. The advertisement criteria were adjusted as needed to target specific postal codes with fewer responses in order to achieve a balanced sample of participants by postal code. Facebook uses an advertisement algorithm that automatically selects the best advertisement to display based on its performance and the advertiser’s bid ; 14 unique advertisements were created and approved by Facebook. For this study, multiple advertisements were submitted for auction simultaneously to create a continuous recruitment window in the event that a particular advertisement performed poorly. Bidding prices and advertisement availability (advertisements can be paused and released at the discretion of the advertiser) were monitored daily and adjusted as necessary until the intended number of completed questionnaires was obtained. The bidding price for advertisements ranged from US $0.75 to US $2.75, with a maximum daily budget of US $50. When a Facebook user clicked on the study advertisement, they were automatically redirected to the secure study website and invited to complete a questionnaire regarding HPV vaccination practices.
The Facebook Ads Manager was used to track the total number of impressions (each time an advertisement was displayed), the number of times an ad was clicked, the average cost-per-click, and the number of people reached (ie, the number of Facebook users that had an opportunity to view one of the study advertisements). Google Analytics software was used to tabulate the total number of visits, the unique visits, the average duration of visits, and the bounce rate (the percentage of visitors that visit a website and leave the site without further browsing) of the study website.