Abstract
PURPOSE Data confirm that high rates of human papillomavirus (HPV) vaccination have not been achieved despite strong clinician endorsement of the vaccine. We conducted a study of primary care clinicians to assess the broad range of health care delivery, health policy, and attitudinal factors influencing vaccination uptake and opportunities for informed decision making.
METHODS We implemented a mixed methods study in RIOS Net, a primary care practice–based research network in New Mexico. We first conducted qualitative, in-depth interviews with primary care clinicians, health policy makers, and immunization experts, and followed up with a confirmatory survey distributed to RIOS Net clinician members.
RESULTS Health service delivery challenges emerged as the greatest barrier to HPV vaccination, specifically the lack of capacity to track and distribute reminders to eligible patients. Clinicians also reported variations in counseling approaches attributable to both age and emphasis on the cancer prevention benefits of the vaccine. There was no evidence of sociocultural influences on vaccine decision making, nor did concerns about perceived overprotection emerge.
CONCLUSIONS Our findings, based on a long-term program of research, suggest that both patients’ attributes and health system delivery are most influential in HPV vaccination coverage challenges. Interventions targeting innovative communication techniques, as well as health system changes that build on efforts toward coordinated care and utilization of other venues to promote vaccination, will be necessary to address these challenges.
- health care delivery
- cervical cancer
- mixed methods research
- health services research
- vaccination
- adolescent health services
INTRODUCTION
The Centers for Disease Control and Prevention (CDC) reports that the human papillomavirus (HPV) vaccine rates trail those of other vaccines for teenagers: tetanus, diphtheria, and pertussis (Tdap), and menningitis.1 Nationally, 57.3% reported getting the first dose of HPV in 2013, a modest increase from 2012 rates (53.8%). Full coverage (not considering the time frame in which the doses were given) is also less than ideal: among girls aged 13 to 17 years, 37.6% received all 3 doses.2 National HPV vaccination data for younger eligible recipients (eg, aged 9 to 12 years) are not provided by the CDC. Given new challenges in the health service delivery environment,3 innovative approaches for increasing HPV vaccination coverage are needed.
In the period before the vaccine’s release, research was aimed at understanding factors that might predict vaccination uptake. Although identification of such factors varied, there was strong consensus that primary care clinicians would play an important role in HPV vaccine dissemination.4–6 During the prelicensure period, we conducted a qualitative study in primary care settings to better understand and anticipate the challenges and opportunities for HPV vaccination. In this journal, we reported the 4 major domains that emerged from our research, which included (1) the complexity of HPV counseling in the clinical encounter, (2) the recommended age, (3) system/compliance issues, and (4) external factors.7
Given the lower-than-anticipated uptake of the HPV vaccine and associated challenges with promoting shared decision making, we followed up on our original study that examined the conditions into which the HPV vaccine would be introduced. The purpose of this research was to assess the broad range of health policy and health service delivery factors that influenced HPV vaccine in the postlicensure period and to identify opportunities for strategic interventions to enhance both uptake and shared decision making. In this article, we report findings from the current mixed methods study that provides a unique basis from which to understand a range of attitudinal, counseling, and practice-system perspectives about HPV vaccination over an 8-year period.
METHODS
Study Design and Setting
We conducted a sequential mixed methods study, including exploratory key informant interviews and a subsequent survey designed to confirm and expand upon these qualitative findings. This study (similar to our earlier research) was conducted in RIOS Net, a practice-based network in New Mexico. The University of New Mexico Human Research Protections Office approved the research protocol.
Key Informant Interviews
Sample and Recruitment
In addition to conducting semistructured interviews with RIOS Net clinicians based in academic, community, and school-based health settings, we also purposefully sampled and interviewed key informant stakeholders from state government, professional coalitions, and advocacy organizations focused on immunizations. We utilized a snowball sampling approach for recruitment, first contacting key informants known to the research team for their clinical and/or health policy interest in the HPV vaccine, who then identified other individuals with similar interests and/or involvement in HPV vaccination efforts.8
Data Collection
We conducted interviews from May 2009 to March 2010, mostly in person, with the exception of 2 completed over the telephone. Our semistructured interview guide covered a range of topics, including clinicians’ views and experiences implementing the HPV vaccine, perceived attitudes and decision-making influences of parents and caregivers, and health policy implications (ie, mandates for school entry). We ended data collection when we reached consensus that no new information was being provided. We digitally recorded and transcribed all interviews verbatim and offered participants a $50 gift card in appreciation of their time.
Data Analysis
The research team utilized an iterative analytic process by organizing the transcripts into categories reflecting the roles of stakeholders who participated in the interviews (see above). Each research team member reviewed 2 to 3 transcripts to identify key themes, after which time the group met to compare emergent themes and develop initial coding structures. Subsequently, 4 members of the team (A.L.S., C.M.G., A.S., A.B.) coded the remaining transcripts from each participant category independently and finalized the coding structure by reaching agreement about the application of codes. Transcripts were imported into NVivo 8 (QRS International). We first analyzed the data without reference to the findings from our earlier research. Once this initial period of analysis was completed, we then integrated the 4 domains from our previous work as a starting point from which to compare and contrast emergent themes and identify new areas as relevant.
Network Survey
Sample and Survey Development
After we analyzed the qualitative data, we developed a survey questionnaire to address the general themes that emerged from the qualitative findings. Included were questions about clinicians’ attitudes about the vaccine, their experience with implementing the Advisory Committee on Immunization Practices (ACIP) recommendations, and their views about the influence of health policy (eg, mandatory vaccination) on their practices. We used visual analog scales to indicate strength of agreement (or disagreement), with 0 = strongly disagree and 100 = strongly agree. It was administered in both electronic and paper formats. The electronic version utilized the Web-based application Research Electronic Data Capture (REDCap). The entire questionnaire can be found in Supplemental Appendix 1, at http://www.annfammed.org/content/13/4/354/suppl/DC1.
We took 2 steps to ensure that the questionnaire captured the perspectives of clinicians who administer the HPV vaccine for adolescent patients aged 9 to 18 years. First, we did not send the questionnaire to internal medicine clinician members of the network who were unlikely to order the HPV vaccine for adolescent patients. Second, we included a screening question that invited participants to complete the questionnaire only after confirmation that they currently ordered the vaccine.
Data Collection
We conducted the survey from May to August 2011. We administered the questionnaire through 4 weekly e-mail solicitations with an embedded survey link. We then sent out 2 more rounds of the questionnaire in paper format to clinicians’ at their practice locations. Two randomly selected respondents were drawn from each solicitation round and received $50 gift certificates.
Data Analysis
The Web-based questionnaires were automatically captured in REDCap; we hand-entered the paper-based questionnaires, double entering 25% of them to ensure data entry reliability. We exported data from REDCap into SAS 9.2 software (SAS Institute) for analysis. We computed descriptive statistics for all items including means and standard deviations of continuous variables and frequencies of categorical items.
As a final analytic step, we integrated the survey findings with our qualitative thematic template to assess the degree to which these data supported and/or modified the 4 major domains.
RESULTS
We conducted key informant interviews with 25 individuals and a total of 98 of 158 eligible RIOS Net clinicians responded to the survey (62%). Demographic characteristics are displayed in Table 1. We provide a comparison of the findings described below with our prelicensure findings in Table 2 and a full summary of survey results in Tables 3 and 4.
Survey Respondent Demographics (n=98)
Pre- and Postlicensure Comparison Table
Summary of Questionnaire Responses
Clinician Questionnaire Summary
HPV Counseling in the Clinical Encounter
Clinicians reflected on their challenges about discussing HPV infections with their adolescent patients, similar to those they had anticipated in the prelicensure period. They reported having difficulty offering the vaccine during acute episodic care visits, which constitute most of their encounters with adolescents. Counseling about the vaccine has become more complex with the introduction of Cervarix (a vaccine designed to prevent HPV types 16 and 18 infections), and conversations to distinguish between vaccine choices raises challenges. For example, in response to the statement, “I am prepared to provide counseling to patients about the difference between Gardasil [a vaccine designed to prevent HPV types 6, 11, 16, and 18 infections, appropriate for boys also] and Cervarix,” the mean response on the questionnaire was 42.1, based on a scale from 0 (strongly disagree) to 100 (strongly agree).
Clinicians reported emphasizing the cancer prevention benefits of the vaccine to avoid what they perceive as a potentially uncomfortable discussion about adolescent sexual behavior, particularly with parents in the examination room. As one clinician stated, “I talk about it as preventing cervical cancer. If they want to discuss the sexual aspects, they’re going to have to bring it up, because I won’t.” Survey respondents strongly endorsed the statement, “In presenting the HPV vaccine, I emphasize its cancer prevention benefits” (mean = 87.2). An adolescent immunization advocate strongly promoted not explicitly linking the vaccine with sexual activity, saying, “We don’t want this to be the sex vaccine because it’s not. It’s a cervical cancer–preventing vaccine or a HPV vaccine or a warts vaccine….”
Recommended Age
Clinicians differentiated between having discussions about the vaccine’s benefits and offering the vaccine to adolescent patients. Survey respondents reported that a discussion about the vaccine is useful to initiate sexual risk counseling for older adolescents, as exemplified by response to the question, “In presenting the HPV vaccine, I see it as an opportunity to open up a discussion about adolescent sexual activity (mean = 73.9).” In the qualitative interviews, clinicians expressed a preference to have these discussions with older adolescents (eg, those aged 15 to 18 years) because the conversation was more in line with their developmental status. As one clinician stated, “To me, 11 and 12 is a little bit young…because [for] some, their body is barely starting to change. And lots of times they’re even afraid to ask. And it’s like, ‘What do you mean; I’ve got to protect myself from cervical cancer? I’ve never even had a period.’”
The average age at which clinicians reported offering the HPV vaccine to adolescent patients was 10.7, however, even younger than the earliest ACIP-recommended age (11 to 12 years). When asked at what age they felt most comfortable first offering the vaccine, the vast majority stated that they preferred younger ages: 21.4% chose 9 to 10 years, whereas 65.3% chose the recommended age of 11 to 12 years; only 13.3% selected categories that included adolescents aged 13 years and older.
System and Compliance Issues
Clinicians strongly endorsed the role of system-level barriers to HPV vaccination, noting that opportunities to offer the vaccine to their adolescent patients are limited, especially in the context of visits addressing other acute health issues or behavioral or mental health needs. Exactly one-half of clinician respondents indicated that they do not schedule appointments to offer the second and third doses. Clinicians responding to the questionnaire estimated that only 9% of their patients received all 3 doses in the recommended 6-month time period, and that 28% of their adolescent patients aged 9 to 18 years had received the first dose. Few clinicians (22%) reported having reminder systems in their clinics. As one family physician noted, “The burden’s on the patients. To be honest, I don’t have the personnel to be calling the patients.”
External Factors: Sociocultural Beliefs, Media, and Policy
Despite well-publicized concerns that adolescents might feel overprotected from HPV (and even other sexually transmitted diseases) upon vaccination in the prelicensure period, clinicians reported that this fear did not play a major role in actual HPV vaccination decisions. Consistent with the literature, most clinicians reported that parents had not raised such concerns during discussions about the vaccine,9 and recent findings from a national sample of girls aged 12 to 18 years old confirm that incidence of sexually transmitted infections does not increase with HPV vaccination.10 When asked on the questionnaire to anticipate whether “the HPV vaccine will lead females aged 9 to 18 years to feel overprotected and more likely to engage in risky sexual behavior,” the mean response was 23.2. Further, despite concerns that diverse cultural considerations (eg, religious or social conservatism) would influence vaccination uptake, clinicians did not report variations in receptivity to the vaccine across the different populations that they serve (mostly Hispanic and American Indian).
DISCUSSION
In this follow-up to our original study, we triangulated data from a mixed methods approach that led us to examine the range of implementation challenges which affect current HPV vaccination practices and to compare and contrast these circumstances with those identified in our previous research. Based on this longitudinal program of research situated in a Southwestern practice-based research network, we offer recommendations about how this information can be used to develop contextually appropriate interventions.
Age-Based Counseling Strategies
Similar to the findings of Vadaparampil et al,11 we found that primary care clinicians continue to struggle with many, though not all, of the challenges anticipated from our earlier study. Perhaps the most prescient concern was that clinicians recognized the likelihood of framing the HPV vaccine differently based on patient age. Indeed, this prediction emerged as a central theme as clinicians in this current study expressed a preference for introducing the vaccine through a cancer prevention lens to younger girls (aged 9 to 12 years), while opportunistically integrating sexual risk behavior counseling with older adolescents (aged 15 to 18 years). This finding is consistent with other recently published qualitative research that found lower self-reported vaccination rates when clinicians considered the timing of sexual activity as compared with clinicians who emphasized a cancer prevention message.12,13 Further, a recent (2013) CDC report found that 3-dose completion rates are 25.8% at 13 years and 48.2% by 17 years,2 and results from 2 national surveys of physicians found that clinicians were more likely to offer HPV vaccines to older female adolescents10 and reported more vaccine refusals among parents of younger adolescents.14 This pattern is especially troubling, given that the antibody response is more robust in younger adolescents (aged 9 to 15 years).13
The data we report show there are subjective interpretations of what constitutes the most appropriate strategy, and interventions that provide guidance to health care professionals might include criteria for choosing one approach rather than another. It may be necessary to tailor counseling strategies in such a way that emphasize cancer prevention or sexual risk reduction based on consideration of patient and contextual attributes (eg, age, cultural group, and level of parental involvement).15 Furthermore, parents and their vaccine-eligible adolescent children may be attracted to interactive media (computer kiosks) or applications for electronic devices that would help them to independently review information and make decisions before they reach the examination room.16 Although clinicians did not detect differences in vaccine receptivity based on ethnicity and socioeconomic status, our published research17 and that of other researchers18 show that even within ethnic groups, parents and adolescents have different decision-making approaches.
System Issues
In our current study, health system delivery issues, including the lack of flexible tracking and reminder capabilities, were identified as the most important barriers to achieving broader HPV vaccination coverage. These barriers lead to missed opportunities, because adolescents come to the clinic most often for acute, episodic care.11 Most clinicians in this study reported having an electronic medical record system, though most respondents did not rely on this system to encourage vaccination visits. Limitations in tracking can lead to further problems, including potential over-vaccination and avoiding mixing the 2 available vaccines, as it is difficult to verify vaccination history.
Implications for Intervention
A major focus of our work is to address the tension between increasing vaccination uptake and ensuring informed decision making. Two strategies emerged from the research: improved communication and health systems change.
Improved Communication
Experience has shown that messages disseminated through various social media channels for teenagers, starting with the ACIP-recommended age and older, can be helpful. Social marketing and Internet-based interventions, however, can also be designed to support brief discussions between primary care clinicians with parents and teenagers in the clinical encounter, as well as dissemination of preventive health recommendations.19
Health System Changes
Health system delivery barriers, more than any other single factor, were identified by participants as the greatest limitation to achieving shared goals of increased vaccination uptake and informed decision making. Given the strong endorsement of the HPV vaccine among both clinicians and the general public, efforts to enhance vaccination may be best served by interventions aimed at improving delivery mechanisms rather than those strategies that focus on individual behavior change.
Optimizing newly required electronic medical systems to enhance tracking and generate timely reminders has been shown to increase vaccination rates.20 Other promising health system approaches include the creation of dedicated nurse appointments for the second and third dose of the HPV vaccine. There may be trade-offs to consider in such approaches, however, as these efforts to increase vaccine coverage may decrease opportunities for patient-clinician communication. Given that very few adolescents complete the 3-dose series in the recommended 6-month period, our previous research suggests that many adolescents and parents may need a basic refresher about the HPV vaccine, and it will be important for staff to be comfortable with these counseling needs.
Another approach to increasing compliance with the vaccine would be to incorporate the HPV vaccine as part of an adolescent immunization platform (Tdap, meningococcal vaccines).21–24 The infrequency of adolescent visits to primary care clinics during the ages of 13 to 17 years makes it all the more important to bundle immunizations as a package to ensure initiation and multidose completion.
Finally, further research is needed to reduce the burden on the primary health care system by identifying other venues for vaccination, such as school-based health centers. Students in these settings are easier to access, and the school year is long enough to complete vaccination in a multidose series.24,25
Limitations
Given that this research reflects the views and experiences of clinicians and immunization experts who work with underserved ethnic minority and rural populations, it is possible that these findings are relevant to these circumstances and not to conditions of vaccine delivery across the country. Our findings, however, are widely consistent with those reported in the broader literature while offering perspectives that may be uniquely relevant to primary care clinicians and researchers working in similar contexts.
In our earlier study, before the approval and release of the HPV vaccine, we sought to identify the range of issues relevant to implementing the vaccine in primary care settings. Now that the vaccine has been available for more than 8 years, persistent gaps in vaccination coverage show the need to overcome both implementation and informed decision-making barriers. Contextually sensitive interventions will be needed to address these challenges. Findings from this current study, based in a long-term program of research examining these evolving dynamics, provide guidance toward that goal.
Acknowledgments
The authors would like to thank Valerie Crespin-Trujillo for her important contributions to this research.
Footnotes
Conflicts of interest: authors report none.
Funding support: This project was supported [in part or in full] by the National Cancer Institute (1R21CA134259-01) and by the National Center for Research Resources and the National Center for Advancing Translational Sciences of the National Institutes of Health through Grant Number UL1 TR000041.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Previous presentations: Preliminary findings from this study have been presented at the 71st Annual Meeting of the Society for Applied Anthropology, Seattle, Washington, March 29–April 2, 2011; the North American Primary Care Research Group, Seattle, Washington, November 13–17, 2010; and the 69th Annual Meeting of the Society for Applied Anthropology, Santa Fe, New Mexico, March 19, 2009.
Supplementary materials: Available at http://www.AnnFamMed.org/content/13/4/354/suppl/DC1/
- Received for publication February 27, 2015.
- Revision received May 28, 2015.
- Accepted for publication June 9, 2015.
- © 2015 Annals of Family Medicine, Inc.
