Despite the high prevalence of HPV infection among the American population, numerous surveys have documented relatively low levels of knowledge about the virus or its consequences. A survey in 2004 found that 70% of women had never heard of HPV and almost 90% have never discussed HPV with a healthcare provider.⁵ A survey of college students, a particularly high-risk population, found that only 27% correctly identified HPV as the most common STI.⁶ More recent data show that awareness of HPV is increasing, although the majority of Americans remain unaware of HPV or its strong ties to cervical cancer.^7,8 A review by Zimet et al.⁹ confirmed poor levels of knowledge about HPV, Pap testing, and cervical cancer among U.S. women. However, it is expected that knowledge and awareness will increase with time as the media and industry marketing continue to bring attention to HPV and the vaccine.

Because medical interventions, such as vaccination, for children younger than 18 generally require parental consent, decisions to vaccinate adolescents will most likely be made by a parent.¹⁰ The review by Zimet et al.⁹ found that across 10 studies, parents expressed moderate to strong interest in vaccinating their adolescent daughter against HPV. Opposition to vaccination was associated with concerns about vaccine safety and efficacy, poor understanding of HPV, cost of the vaccine, and viewing one's child as being at low risk of an STI.^11–15

Research shows that female adolescents, like their parents, are also in favor of HPV vaccination. Numerous studies about adolescent attitudes toward vaccination, recently summarized by Zimet et al.,⁹ conclude that the majority of adolescent females are in favor of vaccination. Correlates of vaccination include parent and physician recommendation, a belief in vaccine efficacy, greater sexual risk taking, and greater knowledge about HPV.

The Advisory Committee on Immunization Practices (ACIP) recommends the vaccine for girls aged 11 and 12 but it is approved for use in females aged 9–26.¹⁶ This decision could be considered controversial, as many conservatives cited fears that vaccinating adolescents could send a message condoning sexual activity before marriage.^17,18 More recently, these groups have refined their positions to be in favor of vaccination, as long as it is not mandatory.¹⁹

The survey consisted of two parts. The first part assessed awareness of HPV, the vaccine, and exposure to news or health ads about HPV or the vaccine in the media. Next, participants were randomly assigned to read one of three framing paragraphs (FP), each of which emphasized a different aspect of the vaccine: the vaccine protects against cervical cancer; the vaccine protects against cervical cancer and an STI; the vaccine protects against cervical cancer and an STI and may or may not lead to increased sexual promiscuity among the vaccinated. All three FPs included general information on disease prevalence and symptoms, as well as vaccine dosage frequency, cost, and potential side effects. The average length of the three FPs was 160 words and the readability was at a 9th grade level. The Appendix presents the complete texts of the three FPs.

Females were asked about their intentions to vaccinate under two conditions: if they had to pay for the vaccine and if the vaccine were provided at little or no cost. Participants, both male and female, were asked if they had at least one daughter in particular age groups (0–8 years old, 9–14 years old, 15–17 years old, and 18–26 years old). If respondents indicated that they had one or more daughters in a particular age group, they were asked about their intentions to vaccinate or recommend vaccination to their daughter. Main effects of each age group were assessed by one-way ANOVA.

The results of HPV awareness and media exposure are presented in Table 1. Overall, the majority of participants (58.0%) had not heard about a new vaccine to prevent cervical cancer. Forty-five percent of females and 39% of males reported they had heard about the vaccine. Roughly one third of participants aged ≥45 had heard about the vaccine, whereas only 16% of those aged 18–29 and 21% of those 30–44 had heard about the vaccine. Non-Hispanic white participants were significantly more likely to have heard of the vaccine than any other racial or ethnic group (χ²=32.2, df=3, p<0.001), as were those with at least a college degree (χ²=36.6, df=3, p<0.001). Almost twice as many participants with a college degree or higher (42.9%) had heard of the vaccine as those with just a high school diploma (27.1%). Participants whose annual household income was >$75,000 were most likely to have heard of the vaccine, whereas those with incomes <$25,000 were least likely to have heard of the vaccine (χ²=14.5, df=3, p=0.002). Parents of children <age 18 were less likely to have heard about the vaccine than nonparents (χ²=6.7, df=, p=<0.010).

Fifty-six percent of participants reported that they had ever heard of HPV. Two thirds of females and 44% of males reported ever hearing of the virus. Middle-aged participants were significantly more likely than younger or older participants to have heard of HPV; 30% of those aged 30–45 and aged 46–60 had heard of HPV, compared with 20.8% of those 18–29 and 18.0% of those ≥60 (χ²=26.5, df=6, p<0.001), as were participants with higher levels of education (χ²=52.3, df=6, p<0.001) and household income (χ²=22.6, df=6, p<0.001). Parents were more likely than nonparents to have heard about HPV (χ²=8.7, df=1, p=0.013). Of those who had ever heard of HPV, 57.2% reported having seen or heard news or ads about HPV at least once per week or more, and 42.8% reported exposure to HPV news or ads as once per month or less. Significant differences in the frequency of media exposure were only seen by gender (66% of females, compared with 34% of males, reported exposure at least once per week or more (χ²=4.24, df=1, p=0.040)). Of females, only 11% indicated that they had talked to someone other than a healthcare provider about HPV.

Women who read that the vaccine protects against cervical cancer (FP 1) had significantly higher intentions to vaccinate themselves when the vaccine was available at little or no cost compared with women who read alternate versions containing information on STIs or the possibility of increased sexual promiscuity, F(2,325) = 5.74, p=0.004). Tukey HSD analysis showed that mean differences were significant between FP 1 and FP 2 and between FP 1 and FP 3. Full results of intentions to vaccinate self or female child are shown in Table 2. Sixty-three percent of women responded that they were very likely or somewhat likely to get the vaccine if it were available to women their age at little or no cost if they read FP 1, compared with 43.0% of women and 43.5% of women who read FP 2 and FP 3, respectively (χ²=16.7, df=8, p<0.05). When asked about their intentions to vaccinate if the vaccine were at a cost to them or their family, there were no discernible differences between conditions, F(2,325) = 1.64, nonsignificant). Intentions to vaccinate female children at different ages were not significantly different between experimental conditions; however, power calculations for each age group revealed that sample sizes may not have been large enough to detect small effects. For example, for the analysis of daughters aged 9–26, there were roughly 50 participants in each condition. Using α = 0.05, d=033, and σ (between group)=1.391, power (β) was 0.255.

Despite the lack of significant findings among intentions to vaccinate female children, important trends emerged that are worth noting. First, mean scores for intentions to vaccinate females >18 years and daughters aged 9–26 were highest among participants who read FP 1, that the vaccine protects against cervical cancer. When information about STIs was added (FP 2), intentions to vaccinate were reduced. Information about increased sexual promiscuity among the vaccinated had little effect on reducing intentions below the levels seen with FP 2. Additionally, across all three age categories, intentions to vaccinate were significantly higher when the vaccine was presented as being of little or no cost to themselves or their family. For example, 50.7% of females responded that they were very likely or somewhat likely to get the vaccine for themselves if eligible when the vaccine was offered at little or no cost, compared with 25.8% of females when the vaccine would be of cost to themselves or their family.

The recall item revealed only partial success in relaying information to participants and framing the vaccine under specific conditions. All three experimental groups had high percentages correctly responding that the paragraph they read described that the vaccine could reduce the number of women dying from cervical cancer. However, the majority of participants failed to accurately recall whether or not the paragraph they read contained information about the vaccine preventing an STI. More than half of the total sample (53.0%) failed to accurately recall whether or not their paragraph discussed the possibility of increased sexual promiscuity among the vaccinated. The distribution of responses to the recall items by condition are presented in Table 3.

A key finding from this study is that information framing had limited effects on intentions to vaccinate. Among adult women, framing had a significant effect on intentions to vaccinate themselves when the vaccine was presented as being of little or no cost to their family. When information describing HPV as an STI was not included, intentions to vaccinate were highest. When STI information was included, intentions dropped, although the decrease was not significant. When information about potential public controversy about sexual promiscuity was added to the STI information, intentions remained relatively constant. Our results are similar to those of a study conducted among Southern women that found that vaccination intentions among women were highest when the vaccine was framed as preventing cervical cancer rather than STI infections; however, the same study reported equal intentions to vaccinate daughters against the two disease outcomes.³⁵

The finding that describing HPV as an STI had negative effects on intentions to vaccinate suggests that a social stigma toward STIs may still exist or that most people believe a vaccine against an STI is personally unnecessary. A Centers for Disease Control and Prevention (CDC) study validated both of these arguments when focus group participants believed they were at low risk of contracting an STI and associated words such as “infidelity,” “promiscuity,” “shame,” and “divorce” with an STI.³⁶ Healthcare providers and communication experts must be cautious of the public's perception of HPV so as not to cause stigmatization of cervical cancer, which could be detrimental to the field of cancer prevention and control. Emphasizing the vaccine as a cancer prevention tool may alleviate some of the stigma associated with STIs.

Another key finding of this study is that, for some, cost of the vaccine may be a realistic barrier of intention to vaccinate. Intentions were higher among all age groups when the vaccine was presented as being of no cost to the participant or her family. The HPV vaccine is the most expensive of all currently recommended childhood vaccines.³⁷ Cost might not be a barrier to the youngest of eligible females, however, because many minors are eligible for free or low-cost vaccines through public or private programs. Those who might be most impacted by cost are young women aged 18–26, who are typically not covered by subsidy programs and are most likely to report being uninsured.³⁸ Even women with private insurance may face financial barriers to vaccination, as co-payments and deductibles can make vaccination prohibitively expensive. As attitudes against vaccination decrease when cost is not a factor, state legislation mandating affordable insurance coverage or expansion of vaccination programs for the uninsured could bring the vaccine within reach to many who are interested.

This study showed that information about the potential of the vaccine to lead to increased sexual promiscuity among those vaccinated had little or no effect on intentions. Despite a movement from conservative groups to link this message with the vaccine to thwart vaccination efforts and preserve abstinence efforts, it appears that the information has not affected vaccination intentions. Instead, some have suggested that the home environment and peers are most influential on adolescent sexual decision making.³⁹ Data from other vaccination studies support our finding as well. A study about Lyme disease vaccination and the risk compensation hypothesis found that whereas vaccinated respondents tended to reduce the frequency of protective behaviors after vaccination, their risk behaviors never exceeded those of the unvaccinated.⁴⁰

The findings from this study support previous research that an STI-associated stigma creates barriers to health behavior adoption.^41,42 This will undoubtedly have implications for communication strategies and vaccination adoption. A report issued by the CDC on communicating HPV information to the public warns public health professionals against focusing primarily on the sexually transmitted nature of HPV, which can distract from the more important health concern of cervical cancer.⁴³ The report is not suggesting that the medical community ignore the STI aspect, which could erode source credibility; rather, it suggests that those who are discussing HPV focus on the commonality of the virus and its serious link to cervical cancer. Our research supports these CDC guidelines and recommendations by providing limited evidence that the framing of health information may have an effect on health behavior intentions.

There are numerous strengths to this study, such as conducting a randomized controlled experiment with a nationally representative sample, but there are a few limitations that might affect the interpretation of results. Foremost, the partial success of the recall item suggests that not all participants were fully aware of the information they were receiving. This may be due to the complexity of the information that was being relayed or that participants' attitudes toward HPV were not influenced by the material. It has been suggested, however, that when the items involve objective recall of information rather than perceived effects, failure of the item to produce anticipated results may not directly jeopardize the findings of the study.⁴⁴ Failure to recall factual information contained in the FPs is not necessarily failure of a manipulation, as the goal of a framing manipulation is to invoke particular emotions or perceptions. Additionally, our measure of vaccine awareness did not include the acronym HPV and might not have fully captured awareness among those who only recognize the vaccine as the HPV vaccine.

As the vaccine was just being launched at the time of our survey and was not yet available to the public for use, we were unable to measure actual vaccination behavior. Whereas theories of health behavior change are premised on the idea that intention is the most immediate and important predictor of behavior⁴⁵ and meta-analyses of intention-behavior relations show significant and strong associations,^46,47 the findings of this study would be enhanced if actual vaccination behavior could have been measured. Only 10% of females who responded to the vaccination intention items were actually eligible for vaccination, based on current ACIP recommendations. Although it is possible that older women will be eligible for vaccination in the future, this may have compromised the generalizability of our findings.