HPV vaccine acceptance, utilization and expected impacts in the US

Abstract

Human papillomavirus (HPV) vaccines represent a remarkable opportunity for the primary prevention of cervical cancer and other HPV-related diseases. With almost four years of vaccine availability now accrued in the United States (US.), data are beginning to accumulate about vaccine utilization patterns and how these may be affected by public opinions about the vaccines. This article describes the burden of HPV infection and related disease in the US, and reviews what is currently known about HPV vaccine utilization among adolescent and young adult females in this country. In addition, we report on emerging data on the personal and attitudinal factors that appear to influence HPV vaccine utilization and discuss how these data may be useful for designing future interventions to improve uptake of these vaccines. Finally, we re-examine cost-effectiveness studies of HPV vaccines, taking into account updated information on utilization of, and public attitudes about, these vaccines.

Burden of HPV-Related Diseases in the U.S.

HPV is the most common sexually transmitted disease in the United States. Less commonly, HPV can be acquired through mother-to-child transmission, autoinoculation and sexual abuse. More than half of all sexually active individuals will acquire a genital HPV infection in their lifetime, and there are over six million new cases each year in the US.1 Adolescents and young adults are at higher risk of acquiring HPV infection than other age groups.2 Nationally representative epidemiological data from the National Health and Nutrition Examination Survey indicate that, among sexually active females, the prevalence of HPV infection was highest in those aged 20–24 (44.8%).3 Though infection rates are high nationally, most infections resolve spontaneously, without evidence of clinically apparent disease. For example, approximately 90% of infections never progress beyond cervical intraepithelial neoplasia (CIN) grade 2.4 However, persistent infection with high-risk or oncogenic types of HPV has been established as the cause of virtually all cervical cancers.5 An estimated 11,270 new cervical cancer cases and 4,070 associated deaths were expected among American women in 2009.6 HPV infection is also implicated in a substantial proportion of anal (85%),7 vaginal (70%),7 vulvar (40%)8 and penile (40%)8 cancers, though these cancers are much less common than cervical cancer. Further, there is accumulating molecular and epidemiological evidence for an etiologic role of HPV infection in oral cancers in men and women; about 25% of mouth and 35% of throat cancers are thought to be HPV-associated.9 Infections with low-risk HPV types 6 and 11 are associated with genital warts. In the US, it is estimated that 1% of the population has HPV infection causing clinically visible genital warts.10 Though not malignant, genital warts can cause significant psychosocial morbidity11 and typically require multiple physician visits for diagnosis, treatment and follow-up.12

HPV Vaccines

Two HPV vaccines are currently available on the US public market. Gardasil^® is a quadrivalent vaccine produced by Merck and Co., and provides protection against HPV types 6, 11, 16 and 18. This vaccine has been shown in clinical trials to have nearly 100% efficacy in females against persistent infection with vaccine-targeted types of HPV (see Bonanni et al.13 for a comprehensive review of HPV vaccines) and also to be nearly 100% effective in preventing cervical dysplasia (CIN2⁺), vaginal/vulvar dysplasia and genital warts associated with the HPV types included in the vaccine. Cervarix^® is a bivalent vaccine produced by GlaxoSmithKline against HPV types 16 and 18. Like Gardasil^®, this vaccine demonstrates similarly high efficacy among females in the prevention of vaccine-type persistent infection and cervical dysplasia (CIN2⁺). Longitudinal studies of Gardasil^® have demonstrated sustained clinical efficacy up to five years, and also a robust anamnestic response following a booster dose, suggesting that the initial 3-dose series established immune memory.14,15 Longitudinal studies of Cervarix demonstrate efficacy out to 6.4 years.16 A randomized prospective trial comparing the two available HPV vaccines found higher serum neutralizing antibody titers and a greater incidence of adverse reactions, mostly injection site reactions, after vaccination with Cervarix.17 Whether the higher immune response offers any clinical benefits remains to be determined. For both vaccines, studies are ongoing to determine the efficacy over longer time periods.

Availability of HPV Vaccines in the US

Availability of, and recommendations for, HPV vaccines vary significantly from country to country. In the US Gardasil^® was the sole HPV vaccine available from 2006 to late 2009. Gardasil^® was initially licensed by the US Food and Drug Administration specifically for females ages 9–26 years.18 Subsequently, the US Advisory Committee on Immunization Practices (ACIP) recommended Gardasil^® for all 11–12 year girls, with “catch up” vaccination suggested for females ages 13–26 who had not previously received the vaccine, and vaccination allowable for girls as young as age 9 based on an individual's circumstances.19 The vaccine is covered under the Vaccines for Children (VFC) program, a national vaccine subsidy program that provides vaccines free of charge to children ≤18 years who are under- or un-insured for vaccines or are from certain native ethnic groups.20 Most private insurers also cover Gardasil^®, though some limit the age range for which it is available.

In September 2009 licensure for Gardasil^® was expanded in the US to include males ages 9–26 years for the prevention of genital warts.21 Unlike the strong and universal recommendation for female HPV vaccination, the ACIP recommended a “permissive” HPV vaccination strategy for males.22 Under a permissive strategy, provider discretion dictates whether the vaccine is recommended or even discussed. However, a permissive recommendation “allows” providers to administer the vaccine to male patients when it is requested. As with females, male vaccination with Gardasil^® is covered under the VFC program for those ≤18 years old who are under or uninsured for the vaccine. Historically, vaccines covered under the VFC program quickly become a covered benefit among those with private insurance.23 However, a “permissive” vaccine strategy for only one gender has not been undertaken previously in the US. Thus it is unclear how such a recommendation will impact private insurance coverage for male HPV vaccination.

In addition to male HPV vaccination, 2009 also marked the licensure of a second HPV vaccine, Cervarix, in the US.24 Given that this vaccine does not provide protection against genital warts (because it does not include HPV-6 and HPV-11 as antigens), the vaccine was approved for females only, specifically for those between the ages of 10–25 years. It is unknown how the availability of two competing HPV vaccine products will affect overall HPV vaccine utilization or acceptability. Because the two vaccines are similar in price (Cervarix,$385; Gardasil^®, $390 for the three-dose series),25 consumer and/or insurance company costs are not likely to be a main driver of any differential utilization of these vaccines. Future studies will be needed to assess why consumers and insurance payers select one vaccine over another.

Current Levels of HPV Vaccine Utilization in the US

In the US, the VFC program provides vaccines to nearly 50% of American children under the age of 18,26 and is therefore a major influential factor affecting vaccine uptake for children in this age range. Unfortunately, a similar vaccine subsidy program does not exist uniformly in all states for under or un-insured individuals 19 years or older. Because of this, it is helpful to consider HPV vaccine utilization levels separately for those who are ≤18 years old (i.e., “adolescents”) and those 19 years or older (i.e., “adults”).

Adolescents.

For adolescent females, both national and regional studies in the US suggest that HPV vaccine utilization is suboptimal. The US Centers for Disease Control and Prevention evaluated HPV vaccine utilization among 13–17 year old girls in 2008 using a “teen” module of their annual National Immunization Survey (NIS).27 This survey uses parental report and medical record confirmation to determine vaccine utilization levels nationally. At the end of 2008, which corresponds to approximately 2½ years of Gardasil^® availability in the US, only 37.2% of 13–17 year old girls had initiated the 3-dose series (i.e., gotten ≥1 dose of the vaccine) and far fewer, only 17.9%, had completed it. There was significant state-to-state variability in vaccine utilization, with levels ranging from as low as 18.5% in Georgia to as high as 57.4% in Rhode Island (for ≥1 dose of the vaccine). In general, states that adopted earlier and/or more comprehensive policies regarding insurance coverage for HPV vaccines appeared to have higher levels of vaccine utilization than states with later or less comprehensive coverage policies.28 An additional finding was disparities in vaccine utilization by race and socio-economic status. Specifically, more Hispanics (44.4%) and Native Americans (52.8%) initiated the vaccine series than Non-Hispanic Whites (35.0%) or Blacks (35.7%). Similarly, those living below the poverty level were more likely to initiate the series (46.4%) than those living above (35.8%). However, whites and those living above the poverty level were more likely to complete the vaccine series than their comparative groups, raising the notion that these groups have different barriers to series initiation versus series completion.

A study by Dempsey et al.29 investigated HPV vaccine utilization patterns in further depth using a regional sample from a University-based health system in Michigan. As with national studies, this study found vaccine utilization to be low among adolescents—only 15% of the 11–18 year old cohort had completed the HPV series in the 15 months after it became available. Interesting differences in vaccine series initiation and completion were again noted. Specifically, African-Americans and those with public health insurance were more likely to initiate the series than other races or private insurance, respectively. However, an opposite pattern was found when analyzing completion of the three dose series (as well as receipt of only two doses)—African-Americans and those with public health insurance had lower series completion rates than their comparative groups. Together with national data, these findings raise an important question of whether HPV vaccines will actually impact the rates of cervical cancer in the US at a population-level. For example, if those at highest risk of developing cervical cancer, that is minorities and those of lower SES, are the least likely to be vaccinated, then the overall population-level impact of HPV vaccination on cervical cancer may be only minimal. Studies are currently ongoing to assess the clinical impacts of receiving only two, rather than three, vaccine doses.30

Dempsey et al. also found disparities in vaccines series initiation, but not series completion, with regard to child age and provider specialty. Vaccine series initiation was significantly lower among 11–12 year olds (the preferred age for vaccination) than older adolescents and gynecologists initiated the vaccine series among their eligible patients less frequently than family practitioners or pediatricians. However, there were no differences by age or by provider specialty in the likelihood of receiving second or third doses once the series had been begun. These data suggest that improving the willingness to initiate the series may be an important target for future interventions.

There have been several additional studies on adolescent vaccine utilization among smaller populations. These studies report utilization levels ranging from 10–58%. Many of this variation can be attributed to the population studied, whether series initiation or completion was analyzed, and the duration from the time of vaccine licensure to when the data was collected.31–36 Future studies will need to examine how patterns of utilization change over time, and how factors associated with failure to vaccinate during adolescence may or may not be related to adult HPV vaccination behaviors. Vaccine utilization levels in US appear to be much lower than in many other developed countries, which have coverage levels >70% for series completion.37 Reasons for this discrepancy are likely multi-factorial, but may be related to different systems of insurance coverage for the vaccine (i.e., centralized in Australia, a “patchwork” of private and public plans in the US) and the fact that the US lacks a systematic mechanism for delivering vaccines to adolescents in the school setting, which has been the primary setting for HPV vaccination in countries like the United Kingdom and Australia. Future studies will also need to examine the impact of cost and insurance coverage on vaccine accessibility.

Adults.

HPV vaccines are recommended by the ACIP universally for adult females through age 26 years. Natural history studies demonstrate that older adolescents and young adults are at highest risk of contracting HPV infection when compared to other age groups.3 Despite this, HPV vaccine utilization among young adult women appears to be lagging considerably behind the already low utilization levels of adolescents. In 2008 Jain et al.38 reported results from an adult module of the NIS study demonstrating that as of mid 2007, only 10% of US women aged 18–26 years reported having initiated the vaccine series. No data was reported in this study regarding completion of the three dose series among adult females. The low levels of adult HPV vaccination highlight the need for improving adolescent vaccination levels, so as to ensure that the vaccine is provided prior to the onset of sexual activity.

Smaller, regional studies of adult females demonstrate similarly low levels of vaccine utilization.35,39,40 A notable divergence however was a study by Chao et al.32 that showed remarkably high levels of vaccine series completion (47%) among adult female members of a large managed-care organization in Southern California. Also interesting was the finding that adult completion of the HPV series was higher than that of adolescents (42%). Reasons underlying this unexpected age-based difference in utilization are unclear but may relate to policies or health promotion practices unique to the managed-care environment and/or the fact that all women included in the study had insurance coverage for the vaccine. Importantly, this finding suggests that when the vaccine is covered by insurance and is accessible within a health system, a large proportion of adult women will opt to get vaccinated.

HPV Vaccine Acceptability and Its Impact on Utilization

Utilization of HPV vaccines is affected directly by patient, parent and provider attitudes about the vaccines. A significant number of studies examining HPV vaccine acceptability were performed prior to the vaccine's licensure in the US. Results of these studies are now being confirmed in the post-licensure era. These studies highlight specific attitudinal barriers to vaccination which may be important targets for future educational campaigns to increase HPV vaccine utilization.

Females.

In pre-licensure studies, characteristics such as child age, perceived access to the vaccine, societal norms, religious background and perceptions about disease severity and susceptibility were all hypothesized to drive uptake of these vaccines among females in the US (reviewed in ref. 42) Post-licensure studies of HPV vaccination have confirmed the influence of many of these factors.31,33,35,38,39,43,44 For example, in a study of 189 females aged 13–26 years, Conroy et al. demonstrated that insurance coverage for the vaccine, and a belief that clinicians, parents and/or partners approved of vaccination were associated with a significantly higher odds of initiating the vaccine series.35 Interestingly, those who had a history of abnormal Papanicolaou smears were less likely to get vaccinated in this study than those without this history—a finding that diverges from pre-licensure attitudinal studies and warrants further exploration.45 In another study, Dempsey et al. used qualitative methods to assess the reasons why mothers did or did not allow their adolescent daughter to be vaccinated. It was found that while both groups of mothers perceived both risks and benefits to vaccination, the balance of these two factors differed significantly between the two groups.43 Gottleib et al. found that among 889 parents of 10–18 year old girls in North Carolina, main reasons for not having their daughter vaccinated included needing more information about the vaccine, believing that their daughter was too young to be at risk for HPV infection and not having visited the doctor where the vaccine could be provided.33 Beliefs about vaccine safety and necessity have also been documented to drive adult women's HPV vaccination behaviors.38 In addition, several studies have documented the strong influence that a doctor's recommendation can have on parents' and patients' HPV vaccination decisions.31,39,43,44 Knowledge gained from these studies can help inform future educational interventions to improve HPV vaccine uptake. For example, taken together these studies' results suggest that accurately describing the risks versus benefits to HPV vaccination and risk of HPV infection during adolescence and young adulthood may be effective messages for educational interventions.

Attitudinal barriers to HPV vaccination among college students merits special consideration as these women often have unique vaccination constraints that are not present among other populations. For example, though most women who attend college are of the age where they can self consent for HPV vaccination, many are still covered under their parents' health care insurance policies.46 This raises issues of confidentiality since parents often receive itemized bills which would indicate when the HPV vaccine had been provided. College women who are not under their parents insurance frequently have coverage within that college's health system.47 However, this coverage can vary considerably with regard to HPV vaccines. The few studies that have examined HPV vaccine utilization in relation to college student status suggest that access to the vaccine may be easier for college-enrolled women than similarly aged individuals who are not enrolled in college.39,41 In addition, personal beliefs are thought to be an important influence of HPV vaccine acceptance among college-age individuals.48,49 In a study of 256 females ages 18–26 attending a university-based gynecologic clinic at a large, public Midwestern university, perceived parental approval regarding HPV vaccination, perceived vulnerability to genital HPV infection, and belief that HPV vaccine is important for maintaining health were significantly associated with increased intention (Patel DA et al., unpublished data).

Males.

A growing number of studies in the US have investigated attitudes towards male HPV vaccination. Given that HPV vaccines have only recently been approved for males, all studies thus far have investigated the “theoretical” acceptance of the vaccine. Most studies have focused on acceptability of the vaccine for adult males,50–54 though a few have investigated parental acceptability of adolescent male HPV vaccination (ref. 55–57, Dempsey et al. unpublished). Findings are similar to that of pre-licensure studies of HPV vaccine acceptability for females—namely that the balance between risks and benefits, vaccine availability, vaccine cost, level of sexual experience and normative influences appear to be important factors.50,52 Interestingly, Dempsey et al. found that among a national sample of parents of boys, while most believed that male HPV vaccination was generally important (90%), far fewer intended to have their own sons vaccinated against HPV in the near future (49%) (Dempsey et al, unpublished). This result suggests that actual utilization of the HPV vaccine by males could be far less that what pre-licensure studies have predicted.

Re-Examining the Predicted Clinical and Economic Impacts of HPV Vaccination

Decision-analytic models have been increasingly used in the context of HPV vaccination to integrate complex epidemiological, clinical and economic data; to simulate population-level disease burden and potential long term outcomes; to examine the impact of uncertain model parameters; and to evaluate the benefits and cost-effectiveness of various vaccination strategies.58 Mathematical models of the clinical and economic impact of HPV vaccination prior to the clinical availability of these vaccines have been reviewed.58,59 These “theoretical” models have generally demonstrated clinical and cost benefits of HPV vaccination targeted at females, though one model developed by the National Cancer Institute suggests that HPV vaccination will have only minimal impacts on the rates of cervical cancer and abnormal Papanicolaou smears.60

The validity of these models must be explored with additional epidemiological data.59,61 For example, it will be important to re-examine the potential impact of HPV vaccination in light of the accumulating “real world” data on HPV vaccine utilization, how utilization is affected by attitudes about the vaccine, and other aspects affecting these models such as frequency of screening and HPV-related disease treatment. With regard to vaccine utilization, several pre-licensure mathematical models of HPV vaccines assumed vaccine coverage levels above 70%,62–67 predicting significant reductions in HPV-related cervical disease with estimated cost per QALY ranging from $3,000 to $24,300 gained by adding routine HPV vaccination of 12 year old girls to existing cervical cancer screening in the US.68 More recent analyses published after clinical availability of HPV vaccines have also been based on similarly high HPV vaccine coverage estimates (≥70%).68,69 However, because these studies assumed high levels of vaccine utilization, the clinical and economic impact of HPV vaccine may have been overestimated.70 Indeed, a more recent model that incorporated parental attitudes about the vaccine suggests that a level of 70% vaccine coverage will not be realized for many years, if attitudes about the vaccine remain the same.70 Future modeling work on the impact of HPV vaccination should incorporate a lower range of vaccine uptake into sensitivity analyses and should also try and parameterize the models using “real world” data on screening and treatment.

Conclusions

HPV infection is common in the US, particularly among adolescents and young adults. Data accrued from clinical and epidemiologic studies thus far provide reassurance regarding the safety and efficacy of currently available HPV vaccines. While vaccination against HPV has the potential to reduce both morbidity and mortality from HPV-associated disease, this will ultimately depend upon utilization of HPV vaccines among those at risk. HPV vaccine utilization in the US currently is low, and important disparities in vaccination related to age, race and other factors have begun to emerge. Understanding how these factors interact with attitudes about the vaccine to affect utilization is important. Incorporating lower levels of vaccination and population heterogeneities in vaccine utilization may be important to consider in future modeling studies in order to gain a clearer picture of the potential clinical and economic impact of HPV vaccination.

Conflicts of Interest

A.F.D. serves on an advisory board for Merck and Co., related to male HPV vaccination. D.A.P. has no conflicts of interest to declare.

Acknowledgements

A.F.D. is supported by the Bridging Interdisciplinary Research Careers in Women's Health (BIRCWH) program at the University of Michigan (K12 HD001438) D.A.P. is supported by a career development award through the National Cancer Institute (NIH/NCI CA120040).