Adolescents’ receipt of the human papillomavirus (HPV) vaccine is primarily a decision of their parents. Health care providers’ recommendation increases parents’ likelihood of initiating the HPV vaccine for their child.1 Rates of HPV vaccine recommendations and the proportion of up-to-date adolescents have improved over time2 ; however, in 2017–2018, 7.3 million vaccine-eligible US adolescents were unvaccinated.1 Notably, despite a provider recommendation, parents of 60.6% of unvaccinated adolescents had no intention to initiate the HPV vaccine series.1 It is unclear whether the sentiment of HPV vaccine hesitancy is growing (because of the rise of the antivaccination movement), has dissipated (with a growing body of evidence pertaining to postlicensure efficacy and the safety of the HPV vaccine), or has remained unchanged over time. To understand the trajectory of acute HPV vaccine hesitancy, we evaluated trends in the proportion of parents lacking the intent to initiate HPV vaccination for their unvaccinated adolescent despite having received a provider recommendation.
Methods
We analyzed the 2012–2018 National Immunization Survey (NIS)–Teen, a national survey representative of the US adolescent population. Each year, the Centers for Disease Control and Prevention conducts a telephonic survey of US households with adolescents aged 13 to 17 years to examine coverage and vaccination behaviors. Adult caregivers most knowledgeable about their adolescent’s vaccination status answer questions regarding 7 adolescent vaccines (measles, hepatitis B, hepatitis A, varicella, tetanus, meningitis, and HPV). Each participant in the NIS is assigned a weight that allows estimates from the surveyed adolescents to be combined to obtain national estimates that reflect the true relative proportions of these groups in the nation as a whole. The survey includes questions on receipt of vaccine recommendation from a health care provider, number of vaccine doses received, and parental intent to vaccinate. A detailed description of the survey methodology and questionnaire are available on the NIS Web site.3 For this study, HPV-unvaccinated adolescents (ie, those who received 0 doses of the HPV vaccine) who reported receiving a vaccine recommendation from a health care provider were identified. Parental lack of intent in this subset was determined by using the question “How likely is it that your teen will receive HPV shots in the next 12 months?”; parents’ responses of “not likely at all” or “not too likely” were identified as lack of intent.
We used survey-weighted frequency procedures to derive nationally representative estimates of the proportion of adolescents who did not initiate the HPV vaccine series despite providers’ recommendations. Subsequently, the proportions of adolescents with parental lack of intent to vaccinate were estimated. Analyses were stratified by adolescents’ sex. Linear regression was used to test for trends. Multivariable logistic regression models compared the probability of parental lack of intent in 2018 vs 2012, adjusting for age, sex, race and ethnicity, poverty status, and the region of residence of the adolescent. Statistical significance was tested at P < .05. Analyses were performed per the NIS-Teen analytical guideline and adjusted for strata and weights by using survey procedures in SAS 9.4 (SAS Institute, Inc, Cary, NC).
Results
The proportion of unvaccinated adolescents who were recommended the HPV vaccine by their provider increased from 27.0% in 2012 to 49.3% in 2018 (Ptrend < .0001). Among these adolescents, parental lack of intent to initiate the HPV vaccine series increased from 50.4% in 2012 to 64.0% in 2018 (Ptrend = .003) (Fig 1A). The lack of intent increased among parents of boys from 44.4% to 59.2% (Ptrend = .0097) (Fig 1B) and girls from 54.1% to 68.1% (Ptrend = .0061) (Fig 1C). Comparing 2018 to 2012, the difference in predicted probability (dPP) for parental lack of intent to vaccinate despite a provider recommendation was 13.6% (confidence interval [CI]: 13.2%–14.0%) (Table 1). Findings were consistent for boys (dPP = 13.9% [CI: 13.3%–14.5%]) and girls (dPP = 14.5% [CI: 14.0%–14.9%]).
Number of unvaccinated US adolescents who were recommended the HPV vaccine by a health care provider and the proportion with parental lack of intent to initiate the HPV vaccine. A, The overall number of US unvaccinated adolescents from 2012 to 2018 who were recommended the HPV vaccine by a health care provider and the overall proportion of these adolescents with no parental intent to initiate the HPV vaccine. B, The number of unvaccinated boys who were recommended the HPV vaccine by a health care provider and the proportion of boys with no parental intent to initiate the HPV vaccine. C, The number of unvaccinated girls who were recommended the HPV vaccine by a health care provider and the proportion of girls with no parental intent to initiate the HPV vaccine. The final analytical sample consisted of adolescents with nonmissing data on provider recommendation and parental intent (n = 42 859; weighted n = 12 991 729); adolescents with missing data were excluded.
Number of unvaccinated US adolescents who were recommended the HPV vaccine by a health care provider and the proportion with parental lack of intent to initiate the HPV vaccine. A, The overall number of US unvaccinated adolescents from 2012 to 2018 who were recommended the HPV vaccine by a health care provider and the overall proportion of these adolescents with no parental intent to initiate the HPV vaccine. B, The number of unvaccinated boys who were recommended the HPV vaccine by a health care provider and the proportion of boys with no parental intent to initiate the HPV vaccine. C, The number of unvaccinated girls who were recommended the HPV vaccine by a health care provider and the proportion of girls with no parental intent to initiate the HPV vaccine. The final analytical sample consisted of adolescents with nonmissing data on provider recommendation and parental intent (n = 42 859; weighted n = 12 991 729); adolescents with missing data were excluded.
Predicted Probability for Parental Lack of Intent to Initiate the HPV Vaccine Among Unvaccinated Adolescents Despite a Provider Recommendation, NIS-Teen 2012–2018
| Predicted Probability, % (95% CI)a | dPP, % (95% CI)a | |
|---|---|---|
| Overall | ||
| 2012 | 50.3 (50.0–50.6) | Reference |
| 2018 | 64.0 (63.7–64.2) | 13.6 (13.2–14.0) |
| Boys | ||
| 2012 | 50.1 (49.6–50.5) | Reference |
| 2018 | 63.9 (63.6–64.3) | 13.9 (13.3–14.5) |
| Girls | ||
| 2012 | 49.2 (48.9–49.5) | Reference |
| 2018 | 63.7 (63.4–64.0) | 14.5 (14.0–14.9) |
| Predicted Probability, % (95% CI)a | dPP, % (95% CI)a | |
|---|---|---|
| Overall | ||
| 2012 | 50.3 (50.0–50.6) | Reference |
| 2018 | 64.0 (63.7–64.2) | 13.6 (13.2–14.0) |
| Boys | ||
| 2012 | 50.1 (49.6–50.5) | Reference |
| 2018 | 63.9 (63.6–64.3) | 13.9 (13.3–14.5) |
| Girls | ||
| 2012 | 49.2 (48.9–49.5) | Reference |
| 2018 | 63.7 (63.4–64.0) | 14.5 (14.0–14.9) |
Predicted probabilities were adjusted for age, sex, race and ethnicity, poverty status, and region of residence of the adolescents.
Discussion
The proportion of unvaccinated adolescents who reported receiving an HPV vaccine recommendation increased from 2012 to 2018; however, parental HPV vaccine hesitancy also increased simultaneously for these adolescents. The surge in HPV vaccine misinformation on social media (for instance, deaths [“Girl dies shortly after receiving HPV vaccine”], injury [“Thousands of UK Girls Injured by HPV”], and government conspiracy [“Tyrannical NY legislators looking to force toxic HPV vaccine on sixth graders”]), may have contributed to this upward trajectory in hesitancy among parents.4 With parents of only 50% of unvaccinated adolescents reportedly receiving HPV vaccine recommendation in 2018, there is an opportunity to further improve the recommendation rates. The data also suggest a need to improve the quality and effectiveness of vaccine recommendations to increase parental HPV vaccine acceptance.5 Information on the recommendation quality was unavailable in NIS-Teen; therefore, results of this study should be interpreted within the context of this caveat.
Our findings imply that recommendations alone will not lead to substantial improvements in vaccine uptake. Providers should proactively use tools such as motivational interviewing and presumptive announcements when they encounter hesitancy.6 Strategic national plans, such as the recently launched Vaccinate with Confidence program, will be crucial for empowering providers to tackle vaccine communication issues and to build public trust in vaccinations.7 Implementation of evidence-based strategies at both the clinic and community levels to address vaccine hesitancy will be important for accelerating the nation’s progress toward HPV vaccination goals.
Drs Sonawane and Deshmukh conceptualized and designed the study, supervised data analysis, drafted the initial manuscript, and reviewed and revised the manuscript; Ms Zhu and Ms Yueh-Yun Lin collected data, conducted the initial analyses, and reviewed and revised the manuscript; Drs Damgacioglu, Montealegre, and Lin coordinated data analysis and critically reviewed the manuscript for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: Supported by the National Cancer Institute of the National Institutes of Health under award number R01CA232888. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Funded by the National Institutes of Health (NIH).
Comments
RE: Response to Brewer et al.
To simply explain our findings-let's assume in 2012, of 1 million US adolescents, 460,000 (46%) were unvaccinated (received 0 doses), and of those adolescents, parents of 138,000 (30%) had received a provider recommendation. Among the parents who received a provider recommendation, 69,000 parents (50%) indicated, "they will not initiate the HPV vaccine series in the next 12 months" (i.e., despite having received a provider recommendation). In 2018 (for the sake of simplicity, let's assume that the population size was constant ), given the improvements in uptake, 300,000 (30%) adolescents were unvaccinated (i.e., coverage improved). Of these, 150,000 (50%) had received a provider recommendation (% of parents of unvaccinated adolescents receiving recommendations also increased as we documented in our study). Now, among 150,000 parents who received a recommendation, if we assume that similar to 2012, parents of approximately 50% of adolescents were reluctant to initiate the series, the number of parents lacking intent will be 75,000. However, what we found was the lack of intent in this subset was higher ("64%" [our outcome of interest]) in 2018, i.e., 96,000 parents responded that they would not initiate the series despite having received a provider recommendation. Note that parents who lack the intent to initiate the HPV vaccine series have increased from 69,000 to 96,000, despite increased coverage. (Although we assumed a constant denominator for this example, newer vaccine-eligible adolescents will get added to the cohort each year given the increasing population size.)
Let's walk through the actual numbers in the NIS-Teen database. In the 2012 survey, parents of 31,792 teens participated. Of the adolescents, 17,832 were unvaccinated and 2,456 reported a lack of intent despite provider recommendation. In 2018, 38,706 participated, 12,798 were unvaccinated and 3,823 reported a lack of intent despite recommendation. The survey-weighted estimates of vaccine refusal despite having received a provider recommendation were 50.3% in 2012 and 63.9% in 2018. Although the NIS does not recommend calculation of the proportion using the overall sample (individuals outside of the domain), if we divide the absolute numbers above, the proportions based on the overall samples (total respondents) will be 7.7% (in 2012) and 9.9% (in 2018), which is higher in 2018. [Note: The NIS-Teen collects response on the parental intention to vaccination only among unvaccinated adolescents; therefore, the NIS analytical guidelines recommend that the analysis should be performed within this subpopulation, i.e., the domain (please see the NIS analytical guidelines for the 'domain analysis': https://www.cdc.gov/vaccines/imz-managers/nis/downloads/NIS-TEEN-PUF17-D...).]
There is no perfect linear inverse relationship between "receipt of recommendation" and "lack of intent". Providers should be applauded for the impressive improvement in recommendations. However, the rise in proportion of parents who are reluctant to initiate the HPV vaccine series despite having received a provider recommendation is a question worth exploring and we hope future studies will be able to provide answers.
RE: Progress in HPV vaccine hesitancy
The paper by Sonawane and colleagues examines the important question of parental hesitancy regarding adolescent HPV vaccination. The authors reported an increase in hesitancy over a six year period, from 50% to 64%, among US parents asked about it. This apparent increase is due to progressively restricting the denominator in later years. Most importantly, the analytic choice masks a more general truth: HPV vaccine hesitancy actually fell among parents overall.
To make sense of this seeming paradox, let’s start by observing that HPV vaccine initiation increased during that time period. The National Immunization-Survey Teen (NIS-Teen) found that, in 2012, around 62.7% of US adolescents had not received HPV vaccination. By 2018, the percentage of unvaccinated children fell to 31.9%. Next, we can observe that NIS-Teen assessed hesitancy only among parents of unvaccinated children and thus had fewer parents to ask about hesitancy in later years. These two findings create the illusion of higher hesitancy when one looks at smaller and smaller subsets of parents over time.
An illustration helps clarify the paradox. Let’s walk through what this would look like for a cohort of 1,000 parents. In the 2012 survey, around 627 parents would have been asked the hesitancy question, and 316 said they were hesitant to vaccinate. While the authors described this is 50% hesitancy (316 of 627 parents), we assert that this is better understood as 32% hesitancy (316/1,000 parents). By the 2018 survey, only about 319 of the parents had unvaccinated adolescent children, and 204 said they were hesitant to vaccinate when asked about it. Again, the authors said hesitancy increased to 65% (204/319 parents), and we believe it decreased to 20% (204/1,000 parents). This illustration shows how important it is to clearly define the denominator when measuring hesitancy over time.
In contrast to Sonawane and colleagues, we conclude that HPV vaccine hesitancy decreased among US parents overall from 2012-2018, as rates of HPV vaccine recommendations and administration rose nationally. Clearly communicating this finding is of vital importance for encouraging pediatricians and other primary care providers to continue to build on the progress they have achieved in improving their communication and establishing HPV vaccination as a norm among US adolescents. Providers' hard work is paying off and should be applauded.
RE: Progress in HPV vaccine hesitancy
To make sense of this seeming paradox, let's start by observing that HPV vaccine initiation increased during that time period. The National Immunization-Survey Teen (NIS-Teen) found that, in 2012, around 62.7% of US adolescents had not received HPV vaccination. By 2018, the percentage of unvaccinated children fell to 31.9%. Next, we can observe that NIS-Teen assessed hesitancy only among parents of unvaccinated children and thus had fewer parents to ask about hesitancy in later years. These two findings create the illusion of higher hesitancy when one looks at smaller and smaller subsets of parents over time.
An illustration helps clarify the paradox. Let's walk through what this would look like for a cohort of 1,000 parents. In the 2012 survey, around 627 parents would have been asked the hesitancy question, and 316 said they were hesitant to vaccinate. While the authors described this is 50% hesitancy (316 of 627 parents), we assert that this is better understood as 32% hesitancy (316/1,000 parents). By the 2018 survey, only about 319 of the parents had unvaccinated adolescent children, and 204 said they were hesitant to vaccinate when asked about it. Again, the authors said hesitancy increased to 65% (204/319 parents), and we believe it decreased to 20% (204/1,000 parents). This illustration shows how important it is to clearly define the denominator when measuring hesitancy over time.
In contrast to Sonawane and colleagues, we conclude that HPV vaccine hesitancy decreased among US parents overall from 2012-2018, as rates of HPV vaccine recommendations and administration rose nationally. Clearly communicating this finding is of vital importance for encouraging pediatricians and other primary care providers to continue to build on the progress they have achieved in improving their communication and establishing HPV vaccination as a norm among US adolescents. Providers' hard work is paying off and should be applauded.