HPV Vaccine Delivery Practices by Primary Care Physicians

From July 2018 to September 2018, we administered surveys to national networks of physicians who had agreed to participate in surveys about vaccine policy issues. The Colorado Multi-institutional Review Board approved this study as exempt research.

As part of the Vaccine Policy Collaborative Initiative, a rapid survey mechanism used to assess physician attitudes about vaccine issues, we surveyed members of networks of pediatricians and FPs recruited from the memberships of the American Academy of Pediatrics (AAP) and the American Academy of Family Physicians (AAFP). Physicians were eligible if they spent ≥50% of their time providing primary care. We performed quota sampling²⁴  to ensure networks were similar to the AAP and AAFP memberships with respect to region, urban versus rural location, and practice setting. In previous work, we demonstrated that survey responses from network physicians were similar to those of physicians randomly sampled from American Medical Association physician databases with respect to reported demographic characteristics, practice attributes, and attitudes about vaccination issues.²⁴ 

We used 4-point Likert scales to assess physicians’ strength of recommendation of the HPV vaccine for patients in different age groups, physicians’ frequency of using different vaccine discussion styles, physicians’ perceived barriers to vaccination and extent of agreement with statements about the change from a 3-dose to a 2-dose schedule in younger teenagers, and physicians’ experiences with HPV vaccine delivery. Physicians were classified as using a “presumptive”⁴¹  (or “announcement”)³⁰  style if they reported almost always or always introducing the HPV vaccine by saying, “We’ve got 3 vaccines today: Tdap, HPV and meningococcal vaccines.” Physicians were classified as using a “conversational”^30,41  (or “participatory”)⁴¹  style if they reported almost always or always saying, “Are you interested in getting HPV vaccine for your child today?” National advisory panels of 6 AAP members and 6 AAFP members representing different states pretested the survey, and it was then pilot tested among 13 pediatricians and 10 FPs.

The survey was administered July 2018 to September 2018, through the Internet or US mail, depending on physician preference. We sent the Internet group an initial e-mail with up to 8 e-mail reminders, and we sent the mail group an initial mailing and up to 2 additional mail reminders. Nonrespondents in the Internet group were also sent up to 2 mail surveys in case of problems with e-mail. We patterned the mail protocol on Dillman’s Tailored Design Method.⁴² 

We pooled Internet and mail surveys for analyses because studies have revealed that physician attitudes are similar when obtained by different methods.^42–44  Physicians who did not deliver the HPV vaccine in their practice were excluded from analysis. We compared respondents with nonrespondents using t tests for continuous variables and Pearson’s χ² tests for categorical variables. A multivariable analysis was conducted, with the outcome of reporting a ≥50% refusal or deferral of HPV vaccination rate among 11- to 12-year-old patients within each sex. We used a log-Poisson model with robust error estimation to calculate relative risks. Independent variables included physician and practice characteristics (specialty, number of providers in practice, proportion of adolescent patients, proportion of privately insured patients) and physician HPV vaccine recommendation style and delivery experiences. We used a cutoff of P < .25 for inclusion of variables into the model. We used a stepwise backward elimination procedure, in which the least significant predictor in the model was eliminated sequentially. At each step, estimates were checked to make sure other variables were not affected by dropping the least significant variable. This resulted in the retention of only those factors that were significant at P < .05 in the final model, with the exception of specialty, which we forced into the final model. All analyses were performed by using SAS version 9.4 (SAS Institute, Inc, Cary, NC).

The overall response rate was 65% (588 of 908); 70% (317 of 456) among pediatricians and 60% (271 of 452) among FPs. Of respondents, 280 (48%) responded via the Internet. Fifteen pediatricians and 43 FPs did not administer the HPV vaccine and were excluded. Reasons for not administering the HPV vaccine were not explored in this survey. In Table 1, we compare respondents and nonrespondents and show additional characteristics available only for respondents. Among FPs, respondents were significantly younger than nonrespondents and had more providers in their practices.

Within each specialty, a larger percentage of physicians made a strong recommendation for HPV vaccination for older adolescents compared with adolescents 11 to 12 years old (Fig 1). For each age group, a larger percentage of pediatricians than FPs made a strong recommendation. Sixty-five percent of pediatricians and 42% of FPs reported almost always or always using a presumptive style, whereas 16% of pediatricians and 24% of FPs almost always or always used a conversational style of introduction. Overall, 40% of physicians used standing orders for HPV vaccination, 66% had a computer-based system that could report adolescents who needed an HPV vaccine, and 42% had an electronic alert in the medical record if a patient needed an HPV vaccine (no significant differences by specialty).

A majority in both specialties strongly agreed or somewhat agreed that it is important that adolescents be vaccinated before they engage in physical intimacy (including kissing), that they encounter more resistance to HPV vaccination compared to other adolescent vaccines because it is not required by schools in their state, and that they encounter less resistance from patients and parents to beginning the HPV vaccine series at age 13 years than at age 11 years (Fig 2). Approximately one-third of physicians in both specialties reported anticipating an uncomfortable conversation if they recommended vaccination for patients at age 11 to 12 years.

Physicians reported high rates of refusal or deferral among 11- to 12-year-old patients, with lower levels among older adolescents (Fig 3); FPs reported higher rates than pediatricians at all patient ages, and physicians in both specialties were significantly more likely to report refusal or deferral for their male patients than for their female patients.

Factors significantly associated with physicians reporting a ≥50% refusal or deferral rate among 11- to 12-year-old patients included the following: not strongly recommending to 11- to 12-year-old patients, not almost always or always using a presumptive recommendation style, strongly agreeing that they encounter less resistance to HPV vaccination from patients age 13 years versus patients age 11 years, and anticipating an uncomfortable discussion when recommending to 11- to 12-year-old patients (Table 2). The multivariable results for both sexes were similar.

Perceived barriers to HPV vaccination that ≥20% of physicians in either specialty reported as major were the following: misinformation parents receive from the Internet or social media, parental concerns about the safety of the HPV vaccine, parents not thinking the HPV vaccine was necessary for their daughters or sons, and opposition to vaccination for moral or religious reasons (Fig 4).

Physicians, especially FPs, frequently were incorrect or reported not knowing about the number of doses recommended in different scenarios or whether additional vaccination with 9 valent HPV should be offered to adolescents who were fully immunized previously with 2 valent HPV or 4 valent HPV (Fig 5). Regarding dosing intervals, among pediatricians, 74% reported routinely recommending the second HPV dose 6 months after the first, and 25% reported routinely recommending it at 12 months; among FPs, corresponding percentages were 88% at 6 months and 12% at a 12 months (P < .0001).

The vast majority of physicians strongly or somewhat agreed that the 2-dose schedule is as efficacious as a 3-dose schedule for adolescents <15 years and that the new schedule facilitated completion and initiation of the series among adolescents in this age group, with pediatricians being more likely to strongly agree with all of these statements than FPs (Fig 6). However, more than half of FPs and close to one-third of pediatricians strongly or somewhat agreed that having different recommendations for adolescents ≥15 years old and younger adolescents created confusion among patients and parents or practice staff.

Despite the existence of highly effective and safe HPV vaccines, national vaccination coverage remains suboptimal. This is the first national survey of physician attitudes regarding the new ACIP recommendations for a 2-dose schedule in younger adolescents, and it offers a glimpse into physician practices with respect to current HPV vaccine recommendations and delivery practices. Our data reveal that a high proportion of physicians are recommending the HPV vaccine, although more are recommending it for older adolescents than for those 11 to 12 years old. A lower proportion of FPs compared with pediatricians strongly recommend the HPV vaccine to patients of all ages evaluated. Slightly more than half of pediatricians and less than half of FPs reported using a presumptive recommendation style, and less than half of those in both specialties used standing orders or electronic alerts in HPV vaccine delivery. Physicians reported high rates of refusal, especially by 11- to 12-year-old patients, who are the target population for vaccination according to national guidelines. Physicians were more likely to report high refusal rates by 11- to 12-year-old patients if physicians did not use a presumptive style, if physicians did not strongly recommend vaccination to young adolescents, or if physicians anticipated less resistance to vaccination from patients older than 11 or 12 years. Although physicians demonstrated some knowledge gaps about the 2-dose recommendations, the majority believed that the revised schedule was facilitating both initiation and completion of the HPV vaccine series among adolescents <15 years of age.

A physician recommendation has been shown to be 1 of the most important factors in parental vaccine acceptance,^27–29  and the lack of a strong recommendation has been identified as an important barrier to HPV vaccination.^22,45–48  Our data reveal substantial increases in the strength of recommendation when compared with a survey conducted ∼5 years ago by using similar methodology.⁴⁹  Percentages of pediatricians strongly recommending the HPV vaccine have increased from 60% in the 2013 survey to 85% in the 2018 survey for 11- to 12-year-old girls and from 52% to 83% for 11- to 12-year-old boys. Smaller increases have occurred among FPs during the same period (from 59% to 72% for 11- to 12-year-old girls and from 41% to 66% for 11- to 12-year-old boys), although the disparity in strength of recommendations between specialties persists.⁴⁹  Although the overall percentage of physicians who strongly recommended the HPV vaccine for patients age 11 to 12 years was high in the current study, it was 13 to 15 absolute percentage points lower than that for patients age 13 to 14 years old in both specialties.

Lower rates of strong recommendations for HPV vaccination among FPs have important implications given their prominent role in adolescent health care delivery. A recent national analysis revealed that the proportion of adolescents seen by FPs increases from 25% among 11-year-old male patients to 57% among 18-year-old male patients and from 31% among 11-year-old female patients to 41% among 18-year-old female patients.⁵⁰  The differences we observed by specialty are reflective of previous literature regarding childhood^51–56  and adolescent^49,57–62  vaccines and highlight the fact that FPs need to remain a major focus of vaccine education and practice improvement strategies to improve HPV vaccine delivery.

Reported rates of deferrals or refusals were high among our respondents, especially for patients ages 11 to 12 years and among FPs and appear relatively stable from 5 years ago.⁴⁹  Physicians, especially FPs, reported higher rates of deferral or refusal for their male patients compared with female patients in every age group. This may be related to the longer duration of recommendations for female patients or to a higher perceived benefit of the HPV vaccine for female patients. Physicians who perceive higher refusal or deferral rates among 11- to 12-year-old patients more frequently were those who (1) did not strongly recommend vaccination in this age group, (2) anticipated uncomfortable discussions with parents of 11- to 12-year-old patients, and (3) expected to encounter more resistance from parents of 11- to 12-year-old patients versus older adolescents. This finding is consistent with those of previous studies demonstrating that physicians are less likely to discuss the HPV vaccine with patients aged 11 to 12 years if they believe parents are likely to defer.^49,63,64  However, other literature has revealed that physicians may considerably overestimate the amount of resistance to vaccination that they are likely to encounter.⁶⁵  Our data also reveal that physicians who report high levels of refusal or deferral are also less likely to use a presumptive recommendation style. To our knowledge, this has not previously been reported. The circular nature of provider anticipation of refusal or deferral potentially leading to a weaker recommendation style and less persistence in responding to parental hesitancy⁶⁶  could be creating a self-perpetuating cycle within a subgroup of physicians.

Our data suggest that improvements are needed in how the HPV vaccine is being recommended. A presumptive⁴¹  style of initiating HPV vaccine discussions uses words that convey an assumption of vaccination and does not discuss the HPV vaccine in a different manner than other adolescent vaccines, whereas, a conversational³⁰  style engages parents in an open-ended discussion about the HPV vaccine without linguistic presupposition of vaccination. A presumptive approach has been shown to be associated with higher HPV acceptance compared with a conversational approach in multiple studies.^30–33  Updated implementation guidance from the AAP includes the use of a strong pediatrician recommendation with wording consistent with a presumptive style,³⁹  and the AAFP Web site refers FPs to Centers for Disease Control and Prevention talking points that included recommendations for “same way” as for other adolescent vaccines and “same day” recommendations. Our data indicate that large percentages of physicians, including a majority of pediatricians, have adopted the presumptive communication style in initiating HPV vaccine discussion, but there are still substantial proportions who have not. In addition to improving physician communication styles, HPV vaccine delivery could also be optimized by increased use of evidence-based methods, including standing orders and alert systems in the medical record to remind providers of the need for vaccination at the point of care, both of which are recommended by the Community Preventive Services Task Force.^67,68 

Factors reported by physicians as important barriers to HPV vaccination (including perceived parental concerns about safety or effects of vaccination on sexual behavior and parents thinking that the HPV vaccine is not necessary or that there are too many vaccines at 1 visit) are consistent with previously identified barriers.²²  However, the barrier most frequently reported by physicians in both specialties (the effect of misinformation parents receive from the Internet or social media) is one that has not been frequently highlighted. Although social media have been recognized as a significant source of vaccine-critical content,^69–72  its contribution to low HPV vaccination acceptance has not been well studied. A recent survey of 1263 parents of US adolescents revealed that stories of HPV vaccine harms were more commonly found in social media than in traditional media or conversations and were associated with lack of initiation, delay, or refusal of HPV vaccination.⁷³  Along with these data, our findings highlight the importance of developing effective public health communication strategies to counter misinformation and highlight benefits of HPV vaccination via social media.

Given the relative recency of recommendations for the 2-dose schedule for adolescents initiating HPV vaccination before age 15 years, it is not surprising that there were knowledge gaps for proportions of physicians about when a third dose is recommended, the acceptability of using different HPV vaccines to complete the series, and the need for an additional 9 valent HPV vaccination in those already fully vaccinated with either 2 valent HPV or 4 valent HPV. Physician knowledge should increase over time, as will, presumably, the ability of the electronic medical record or immunization registry–based systems to forecast the need for HPV vaccination. Interestingly, the majority of physicians in both specialties reported recommending the second dose of the 2-dose schedule at a 6-month interval rather than 12 months later at the next annual well-child visit, although national guidelines allow for either. This may reflect an understanding of the importance of series completion as soon as possible. Importantly, primary care physicians strongly endorsed the belief that the 2-dose schedule was facilitating both initiation and completion of the HPV series.

There are important strengths and limitations to our data. We surveyed large, nationally representative samples of pediatricians and FPs and achieved high response rates. Responses of our sentinel physicians may not be fully generalizable, although previous work has revealed that the survey methods described yield similar responses to the most commonly employed method of sampling physicians nationally.²⁴  Nonrespondents might have different views than respondents, although our high response rates helps to mitigate against this source of bias. Most importantly, our data are based on self-report rather than direct observation and may not entirely reflect actual physician practice.

This snapshot of HPV vaccine delivery in primary care demonstrates room for improvement in the way physicians are communicating about the HPV vaccine and in their delivery practices. Most important in this regard is the continued finding that physicians who experience or expect to experience high rates of deferral or refusal may be anticipating and accommodating refusals by altering their recommendation strength and style. Such accommodations by physicians may perpetuate a lack of acceptance of the HPV vaccine among parents. Greater physician awareness about the potential of overestimating the degree of parental resistance to HPV vaccination and about the effectiveness of a strong recommendation for the HPV vaccine, delivered in the same way as for other adolescent vaccines and on same day as other adolescent vaccines, may be key to increasing acceptance among parents of 11- to 12-year-old patients. Increased use of available communication training materials and applications as well as further development of evidence-based messages for parents may be helpful in improving the way HPV vaccination is introduced.^74–78  Our data are encouraging in revealing substantial increases over the past 5 years in the percentage of physicians who report strongly recommending the HPV vaccine to 11- to 12-year-old patients. The findings also reveal that according to primary care physicians, the 2-dose schedule could result in meaningful increases in HPV vaccination initiation and completion among adolescents, leading to greater protection against HPV-associated cancers in the United States.

We thank Lynn Olson, PhD, and Karen O’Connor from the Department of Research, AAP; Bellinda Schoof, MHA, from the AAFP; and the leaders of the AAP and AAFP for collaborating in the establishment of the sentinel network in pediatrics and family medicine. We also thank all pediatricians and family physicians in the networks for participating and responding to this survey.

AAFP

American Academy of Family Physicians

AAP

American Academy of Pediatrics

ACIP

Advisory Committee on Immunization Practices

FP

family physician

HPV

human papillomavirus