Factors associated with parental reasons for “no-intent” to vaccinate female adolescents with human papillomavirus vaccine: National Immunization Survey - Teen 2008–2012

Background 1) To identify socio-demographic factors associated with parental “no-intent” for their 13–17 year old unvaccinated daughter to receive the human papillomavirus (HPV) vaccine series within the next twelve months, 2) to describe patterns in “no-intent” by socio-demographic factors, and 3) to identify socio-demographic factors associated with parental reasons for “no-intent”. Methods Data from 2008–2012 National Immunization Survey - Teen (NIS - Teen) were examined in this study. Parents with “no-intent” to vaccinate their daughters were asked to identify reasons for their decision. All responses were categorized into five domains identified as barriers to receive the HPV vaccine series: 1) Safety and Effectiveness Concerns; 2) Systemic Barriers; 3) Vaccine Misinformation; 4) Lack of Knowledge about the Vaccine; and 5) Socio-cultural Barriers. Multivariable logistic regression models were performed to address the study objectives. Results Number of people in the household, household income, mother’s age, education, health insurance, recommendation of a health care provider, and the survey year were significantly associated with parental “no-intent”. Race/ethnicity, mother’s education, marital status, recommendation of a health care provider, household income, age of the unvaccinated daughter, and the survey year, were significantly associated with one or more domains identified as barriers to receive the HPV vaccine. Conclusions This study identified sub-groups of parents across different socio-demographic factors with “no-intent” for their adolescent daughters to receive the HPV vaccine. Developing strategies that target educational tools towards the identified sub-groups of parents about the purpose, safety, and efficacy of the HPV vaccine, and HPV infection, may help increase HPV vaccine acceptance, initiation and completion rates.


Background
In the United States (U.S.), human papillomavirus (HPV) is the most prevalent sexually transmitted infection (STI) with up to 24.5% HPV infection among females aged 14-19 years [1][2][3]. HPV types 6 and 11, cause 90% of genital warts and HPV types 16 and 18, cause 70% of cervical cancers [4]. Oncogenic HPV also causes vulvar, vaginal, anal cancers [5,6]. In 2012, there were two highly efficacious HPV vaccines on the market licensed by The Food and Drug Administration (FDA); the bivalent vaccine (Ceravix) which provides protection against high-risk oncogenic HPV types 16 and 18 only, and the quadrivalent vaccine (Gardasil) which provides protection against types 16 and 18, as well as two low risk types 6 and 11 [7][8][9]. In 2014, the 9-valent (Gardasil 9) virus-like particle HPV vaccine was approved which has the potential to increase cervical cancer prevention from 70% to 90% [10,11]. The Healthy People 2020 goal for HPV vaccine coverage among female adolescents aged 13-15 years in the U.S. is 80% [12]. Among females aged 14-19 years, Markowtiz and colleagues reported a decline in the prevalence of vaccine preventable HPV infections from 11.5% in 2003-2006 (the pre-vaccine era) to 5.1% in 2007-2010 (the post-vaccine era) [13]. Despite a significant increase in HPV vaccination coverage during 2007-2011, the increase in the percentage of unvaccinated girls with at least one missed opportunity for HPV vaccination from 20.8% to 84.0% is a major public health concern [14]. These numbers highlight the potential of the HPV vaccine, the importance of developing educational strategies for parents about the vaccine, and the need for targeting vaccination campaigns towards adolescents to reduce HPV infections and their sequelae, most notably genital warts and cervical cancer.
A considerable number of studies have focused on the factors associated with initiation and completion of the HPV vaccine series [15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30]. For example, two studies reported older age of adolescent females was associated with initiation and completion of HPV vaccination [22,23]. Only recently, the research focus has shifted to understand the population of unvaccinated teens, factors associated with "intent", and parental reasons for "no-intent" for their daughters to receive the HPV vaccine [25][26][27][28][29][30][31][32][33][34][35]. However, such studies were limited in that they only reported the most common reasons cited by parents for "no-intent" to vaccinate their adolescents, while failing to identify the underlying socio-demographic characteristics of such parents. Parental attitudes and beliefs regarding vaccines in general and the HPV vaccine in particular, affect the likelihood that their children will receive vaccines [36]. It is particularly important that we identify these socio-demographic sub-groups and develop targeted strategies to improve acceptance and uptake of the HPV vaccine.
In a recent study among unvaccinated females aged 15-25 years, Liddon and colleagues described the correlates of the main reasons for foregoing vaccination in females [33]. However, to our knowledge, factors associated with various self-disclosed parental reasons for "no-intent" to vaccinate their daughters have not been studied in a nationally representative sample of unvaccinated female adolescents aged 13-17 years.
In this study, we used a sub-sample of unvaccinated females aged 13-17 years from the National Immunization Survey -Teen (NIS -Teen), 2008-2012, to address three specific objectives: 1) To identify socio-demographic factors associated with parental "no-intent" for their [13][14][15][16][17] year old unvaccinated daughter to receive the human papillomavirus (HPV) vaccine series in the next twelve months, 2) to describe patterns in "no-intent" by sociodemographic factors, and 3) to identify socio-demographic factors associated with parental reasons for "no-intent".

National Immunization Survey-Teen (NIS-Teen)
Data from 2008-2012 NIS-Teen surveys were used for the analyses [37]. The NIS-Teen survey is implemented annually by the National Center for Immunization and Respiratory Diseases and the National Center for Health Statistics of the Centers for Disease Control and Prevention. It represents a stratified national probability sample of households in the U.S. that include all 50 states, the District of Columbia, and the U.S. Virgin Islands. Details about methodology and weighting procedures, have been previously published [38]. All estimates were based on adolescents with adequate provider data. The observations from the U.S. Virgin Islands were excluded from all analyses. The final sample size was 23,722. This study was exempt from review by Kent State University's IRB because NIS -Teen data are publicly available and deidentified.
No-intent to receive the HPV vaccine series Parent (or guardian) of a female adolescent with no history of HPV vaccination at the time of the interview was asked the question "how likely is it the teen will receive HPV vaccinations in the next 12 months?". Of the five possible responses, parents who responded "very likely" or "somewhat likely" were considered as having "intent" and those who responded "not too likely", "not likely at all", or "not sure/do not know" were considered as having "no-intent" for their daughter to receive the HPV vaccine series within the next 12 months. The percentage of parents who responded "not sure/do not know" was very small (<3%). Parents with "no-intent" were then asked the question, "What is the main reason [your teen] will not receive HPV shots in the next 12 months? " Parents were allowed to report one or more reasons from a list which CDC created. For the present analyses, we created the following five domains for the reasons parents reported: i) Safety and Effectiveness Concerns, ii) Systemic Barriers, iii) Vaccine Misinformation, iv) Lack of Knowledge about the Vaccine, and v) Sociocultural Barriers. Table 1 shows the questions that went into each domain created based on qualitative groupings. Each domain is treated as a binary outcome in our analyses: "Yes" if a parent reported at least one of the reasons and "No" if no reason was reported in the corresponding domains. The unadjusted weighted prevalence estimates with 95% confidence intervals (CIs) were computed to describe the socio-demographic characteristics of unvaccinated female adolescents. Univariate analyses were performed to test for associations between sociodemographic factors and parental "intent" vs. "no-intent" for their daughters to receive the HPV vaccine series within the next 12 months. Multivariable logistic regression analysis was performed to determine the factors associated with "no-intent". To describe how the socio-demographic factors associated with parental "no-intent" changed over time, we computed the unadjusted weighted prevalence estimates with 95% CIs for each survey year. The sample size for these analyses was 21,467. To determine how the socio-demographic factors associated with parental reasons for "no-intent", we performed multivariable logistic regression analysis treating the five domains as binary outcomes (Yes/No) correspondingly. The sample size for these analyses was 12,274.
Sampling weights provided in the 2008-2012 NIS -Teen public use data were used to obtain population-based estimates. All analyses were performed using SAS® version 9.3 (Cary, NC, USA) survey procedures (PROC SURVEY-FREQ, PROC SURVEYMEANS, PROC SURVEYLOGIS-TIC) to account for the complex sampling design of the NIS-Teen data. All analyses were conducted in 2014.  Table 2 presents the socio-demographic characteristics of the unvaccinated adolescent females in this study. Sixty percent were White non-Hispanic, 71.6% of mothers of unvaccinated females were married, 33.6% were college graduates, and 46.6% of were between the ages 35-44. Sixty-three percent reported that a healthcare provider had not recommended that their daughter receive the HPV vaccine series.

Socio-demographic factors associated with no-intent to receive the HPV vaccine series
During 2008-2012, 59.9% (95% CI: 58.7%-61.1%) of parents of unvaccinated adolescent females indicated that they did not intend to have their daughters vaccinated against HPV within the next 12 months. Table 3 describes the socio-demographic characteristics of unvaccinated females across "no-intent" vs. "intent" groups and presents the factors significantly associated with parents' "no-intent" to vaccinate their daughters: number of people in the household (Adjusted Odds Ratio: AOR: Patterns in no-intent to receive the HPV vaccine series, 2008-2012 Table 4 presents patterns across survey years in parental "no-intent" for their daughters to receive the HPV vaccine series within the next 12 months, by sociodemographic characteristics. Among parents who did Additionally, there was a significant increase through survey years 2009-2012 when compared to survey year 2008 among the following factors: mothers with a college education, mothers 45 years of age or older, and higher income were all associated with an increase in the proportion of parents who reported "no-intent".
Factors associated with reasons for "no-intent" to receive the HPV vaccine series Of the 12,274 parents who reported "no-intent" for their daughters to receive the HPV vaccine series within the next 12 months, 92.3% of parents cited one reason, 6.6% cited two, and 1.1% cited three or more reasons with the maximum being seven reasons. The proportion of parents who reported one or more reasons which fall in one of more of the five domains are as follows: 43.8% vaccine misinformation, 18.0% safety and effective concerns, 17.0% lack of knowledge about the vaccine, 16.0% systemic barriers, 5.5% socio-cultural reasons. Socio-demographic factors associated with parental reasons for "no-intent" for their daughters to receive the HPV vaccine series within the next 12 months is presented in Table 5. Lack of knowledge about the vaccine as a reason for "no-intent" were significantly greater among female adolescents 15

Discussion
This study is one of the first to identify sub-groups of parents across different socio-demographic factors who reported "no-intent" and reasons for "no-intent" for their daughters to receive the HPV vaccine series within the next 12 months in a nationally representative sample of unvaccinated female adolescents 13-17 years of age.
Though the proportion of HPV vaccine uptake among female adolescents has significantly increased over time, this study found that three out of five parents of unvaccinated females reported "no-intent" to vaccinate their daughters within the next 12 months (2008: 55.1% ; 2009: 60.6% ; 2010: 64.6% ; 2011: 59.8% ; 2012: 60.3%). In addition, this study describes the patterns over time in socio-demographic factors of parents who reported "no-intent". The study findings provide useful information to develop strategies for outreach and education geared toward health care providers whose lack of recommendation for the HPV vaccine significantly influenced parental "intent" for their daughters to receive the HPV vaccine series, to address parents' concerns about the HPV vaccine across different socio-demographic characteristics. This may help improve the vaccine uptake among females 13-17 years of age.
Consistent with previous reports [25,32,[39][40][41], provider recommendation was observed to significantly influence parental "intent" for their daughters to receive the HPV vaccine series; more than sixty percent of parents of unvaccinated females received no recommendation from their health care provider (2008: 70.1% ; 2009: 62.8% ; 2010: 63.4% ; 2011: 60.3% ; 2012: 55.9%). Among parents with "no-intent" who reported systemic barriers, 21.6% received a provider recommendation, compared to 3.9% of parents who did not receive a provider recommendation for their adolescent daughters. Among parents with "no-intent" who reported Safety and Effectiveness Concerns, 28.1% received a provider recommendation compared to 13.3% of parents who did not receive a provider recommendation for their adolescent daughters. Lastly, compared to parents in the survey year 2008, parents in the survey years 2009-2012 were significantly more likely to report "no-intent" and cite Safety and Effectiveness Concerns as their reason, yet less like to cite Lack of Knowledge about the Vaccine as a reason for "no-intent". This may suggest that over time knowledge about the vaccine has increased, but along with that the safety and effectiveness concerns have also increased. These findings not only highlight the importance of a provider recommendation, but also the missed      opportunities health care providers have to effectively communicate the health benefits of the HPV vaccine, specifically with regard to vaccine safety and effectiveness. Additionally, a provider recommendation would help to facilitate an improvement in vaccine acceptance and vaccination intentions [42][43][44][45].
In this study, mothers with a high school education or higher were more likely to report "no-intent" for their daughters to receive the HPV vaccine series within the next 12 months. Other studies have reported similar findings [28,35,46]. In addition, we found mothers with higher education were more likely to cite Safety and Effectiveness Concerns as a reason for "no-intent". This highlights the importance of appropriately addressing safety and effectiveness concerns among all mothers, particularly those with higher education. Also, we found older mothers were more likely to report "no-intent" compared to younger mothers, and this result is also consistent with other studies [47,48]. Older mothers, Black non-Hispanic, Hispanic, and parents of "Other" ethnicity were more likely to report "Lack of Knowledge about the Vaccine" as a reason for "no-intent". Consistent with our finding, recent studies in Latino parents' reported low HPV vaccine knowledge as one of the main barriers to vaccine receipt [41,49]. Furthermore, parents of older adolescents Black non-Hispanic and parents with annual household income more than $35,000, and were more likely to cite "Socio-cultural Barriers". These findings highlight the importance of disseminating health communication about HPV across various demographics.
Overall, this study found that the patterns in sociodemographic factors between proportions of parents reporting "no-intent" were similar across years 2008-2012. These patterns highlight that HPV vaccine acceptance beliefs among parents have not significantly changed over time, thereby impacting vaccine uptake among certain sub-groups of adolescents identified in this study. Considering that the HPV vaccine is intended to protect against HPV infection and pre-cancerous lesions caused by high-risk HPV types, delaying the initiation or refusing the vaccine altogether by these subgroups of parents for their daughters is a matter of public health concern.
We acknowledge that the analysis of this study has some limitations. In the years 2008-2010, during which cell-phone-only households were increasing, the NIS-Teen sampling frame included only households that had landline telephones. For the first time in 2011, NIS-Teen utilized a dual-frame (both landline and cell-phone sampling frames) but the cellular household response rate was low compared to the landline household response rate in both 2011 and 2012. As a result, non-response and non-coverage bias may exist, even after adjusting sampling weights. Geographic region in the US, which we did not consider in our study, may be helpful to better understand and address parents' reported reasons for "no-intent" to vaccinate their adolescent daughters [24]. Also, information about adolescent daughter being a first child or not would shed some insights into parents' awareness of HPV vaccine. Lastly, parental no-intent to vaccinate their 13-17 years old daughters within the next 12 months might not necessarily result in not vaccinating beyond the 12-month period, as intention and action often vary.

Conclusions
This study identified sub-groups of parents across different socio-demographic factors with "no-intent" for their daughters of 13-17 years of age to receive the HPV vaccine series within the next 12 months, and their reasons for such a decision. Developing strategies that target educational tools towards the identified sub-groups of parents about the purpose, safety, and efficacy of the HPV vaccine, and HPV infection, may help increase HPV vaccination acceptance, and initiation and completion rates among female adolescents. In addition, strategies to improve vaccine uptake should focus on developing communication tools for health care providers to better address parents' reasons for "no-intent" "no-intent" to initiate the HPV vaccine series in the next 12 months; b Sample size with complete data on all the covariates in the logistic regression models c Logistic regression models include all the variables listed; d Safety and effectiveness concerns category included the cited reasons of safety and side effects, new vaccine, need more information, and effectiveness concerns; e Systemic barriers includes: health care provider not recommended, cost, vaccine not available, no doctor/obstetrician/gynecologist/doctor visit scheduled, and not a school requirement; f Vaccine misinformation includes: not needed/not necessary, adolescent not sexually active, not appropriate age, and concern of increased sexual activity; g lack of knowledge about the vaccine included the response lack of knowledge; and h Sociocultural reasons included: family/parental decision, do not believe in immunizations, and against religion/orthodox i Includes non-Hispanic other racial/ethnic groups and biracial; j Never married/widowed/divorced/separated; k SCHIP = State Children's Health Insurance Program; IHS = Indian Health Service