OBJECTIVES. The aims of this study were to examine mothers' intention to vaccinate their daughters and themselves against human papillomavirus and to determine which demographic, behavioral, and attitudinal factors were associated with intention to vaccinate daughters.
METHODS. We surveyed 10 521 US mothers, all nurses, between June 2006 and February 2007. Multivariable logistic regression models were used to determine which of the following factors were associated with a mother's intention to vaccinate a 9- to 12-year-old daughter: demographic factors, gynecologic history, belief that one's daughter should have regular Papanicolaou testing, beliefs about Papanicolaou testing outcomes (3-item scale), and beliefs about human papillomavirus vaccines (7-item scale measuring beliefs about human papillomavirus vaccine efficacy, impact of vaccination on sexual and Papanicolaou screening behaviors, severity of and susceptibility to human papillomavirus, and anticipated clinician recommendations).
RESULTS. Of the 8832 mothers who completed a survey (84% response rate), 7207 had a daughter. Among mothers with a daughter, 48% intended to vaccinate a daughter if she were 9 to 12 years of age, 68% if she were 13 to 15 years of age, and 86% if she were 16 to 18 years of age. Forty-eight percent intended to receive the vaccine themselves if recommended. In multivariable regression models, variables significantly associated with intention to vaccinate a 9- to 12-year-old daughter included belief that one's daughter should have regular Papanicolaou testing and beliefs about human papillomavirus vaccines.
CONCLUSIONS. In this first national study of mothers' attitudes about human papillomavirus vaccines, mothers' intention to vaccinate a daughter <13 years of age was lower than intention to vaccinate an older daughter, contrasting with national recommendations to target 11- to 12-year-old girls for vaccination. Educational interventions designed to affect mothers' willingness to vaccinate daughters should focus on human papillomavirus vaccine efficacy, behavioral impact of vaccination, perceived risk of human papillomavirus, and clinician support for vaccination.
Human papillomavirus (HPV) infection is a highly prevalent sexually transmitted infection (STI) that may cause genital warts, cervical cancer, and other malignancies. These health conditions are responsible for substantial morbidity, mortality, and expense globally. A quadrivalent HPV vaccine was licensed by the US Food and Drug Administration in June 2006, and a bivalent HPV vaccine is expected to be licensed soon. Both vaccines are highly effective in preventing persistent infection with HPV-16 and -18, high-risk types that are causally associated with cervical and other anogenital cancers, and in preventing cervical cancer precursor lesions caused by those HPV types.1,2 The quadrivalent vaccine is licensed for girls and women 9 to 26 years of age, and the US Advisory Committee on Immunization Practices (ACIP) has recommended that clinicians target girls 11 to 12 years of age for HPV vaccination.3 Clinical trials are ongoing that explore the efficacy of HPV vaccination in older women, and an HPV vaccine may be licensed for women >26 years of age in the future.
Parental attitudes about HPV vaccines will be key determinants of adolescent vaccination. Previous studies have demonstrated that parents generally support vaccination of their children, but some parents have specific concerns about their child receiving an STI vaccine, and some parents are reluctant to immunize daughters in the age range targeted for vaccination.4–8 Most published studies enrolled parents from a limited geographic area and were conducted before vaccine licensing. Thus, we conducted a national, postlicensing study involving mothers of adolescent and young adult women. The objectives were to (1) characterize mothers' attitudes about HPV vaccines and intention to make sure that daughters in different age groups were vaccinated, (2) explore mothers' intention to be vaccinated themselves, if the vaccine were recommended for women their age, and (3) determine which demographic, behavioral, and attitudinal factors were associated with intention to vaccinate a daughter in different age groups.
PARTICIPANTS AND METHODS
The study sample comprised mothers of children participating in the Growing Up Today Study (GUTS), a national longitudinal study of US adolescents.9 The mothers were women participating in the Nurses' Health Study II (NHS II), a national longitudinal cohort study of 116 686 female nurses, established in 1989. In 1996, participants in NHS II who had a child between 9 and 14 years of age were asked if their child could participate in a new longitudinal study (GUTS). The baseline questionnaire was returned by 16 882 boys and girls, representing 12 717 mothers. This study was conducted by using data from a mailed survey of mothers conducted twice between June 2006 and February 2007 (ie, within the first year after approval of the quadrivalent vaccine by the US Food and Drug Administration). Of the 12 717 NHS II participants who were originally involved (in 1996) as mothers of GUTS participants, 10 521 (83%) received this survey. The remaining mothers did not receive a survey, because during the previous 10 years their child had declined to participate in GUTS, the mother had declined to participate in the NHS II, the child or the mother had died, or the mother had not responded to the last NHS II survey. Mothers completed the survey items for this study only once, even if they had more than 1 child enrolled in GUTS, and mothers included in these analyses were those who reported having at least 1 daughter. The study was approved by the Brigham and Women's Hospital institutional review board.
Participants completed a paper-and-pencil questionnaire assessing the following independent variables (Table 1): demographic factors, history of Papanicolaou screening in the mother and her daughter, history of HPV infection or a related condition (eg, genital warts, abnormal Papanicolaou tests, cervical dysplasia, or cervical cancer), history of smoking, frequency of communication with one's daughter about Papanicolaou testing, belief about the importance of regular Papanicolaou testing for one's daughter, personal beliefs about Papanicolaou testing (3-item scale), and beliefs about HPV vaccines (7-item scale). Personal beliefs about Papanicolaou testing and HPV vaccines were measured by using scales that were adapted from our previous work and informed by a conceptual model guided by theories of health behavior such as the theory of planned behavior, social cognitive theory, and the health belief model.10–12 Individual items for both scales are shown in Table 2.
The items assessing the outcome variables, intention to vaccinate one's daughter and oneself, were preceded by an introductory paragraph about HPV and HPV vaccines. The paragraph stated: (1) HPV is an extremely common infection that is sexually transmitted and may cause genital warts, abnormal Papanicolaou tests, cervical cancer, and other cancers; and (2) the quadrivalent HPV vaccine has the potential to prevent ∼90% of genital warts and 70% of cervical cancers, and is most effective if given before sexual initiation. The outcome variables were (1) the mother's reported likelihood of vaccinating a daughter in each of 3 age groups (9–12, 13–15, and 16–18 years), if she had a daughter in each age group and given that the vaccine was safe, affordable, effective, and recommended for each age group, and (2) the mother's reported likelihood of receiving the vaccine herself, if HPV vaccination were recommended for women her age. The 5 possible responses ranged from extremely likely to extremely unlikely. We defined those who intended to vaccinate a daughter as those who reported that they were extremely or somewhat likely to make sure a daughter in each age group was vaccinated. We defined those who did not intend to vaccinate a daughter as those who reported they were neither likely nor unlikely, somewhat unlikely, or extremely unlikely to make sure a daughter was vaccinated.
Because the ACIP recommends targeting 11- to 12-year-old girls for vaccination, we first focused on intention to vaccinate a daughter in this age range as an outcome variable. We conducted unadjusted and adjusted logistic regression modeling to determine if the following factors were associated with intention to vaccinate a 9- to 12-year-old daughter: demographic factors, history of Papanicolaou screening, history of HPV, history of smoking, communication with daughter about Papanicolaou testing, belief about the importance of Papanicolaou testing, personal beliefs about Papanicolaou testing, and beliefs about HPV vaccines. Those variables associated with the 2 outcome variables at P ≤ .10 in unadjusted regression models were entered into the multivariable regression model. Because previous studies have shown that parents are reluctant to immunize younger (eg, ≤12 years old) daughters compared with older (eg, ≥16 years old) daughters, we also examined whether there were differences in the factors associated with a mother's intention to vaccinate daughters in varying age groups. We categorized the outcome variable as follows: (1) intends to vaccinate older and younger daughters (ie, 16–18 and 9–12 years of age); (2) intends to vaccinate older but not younger daughters (ie, 16–18 but not 9–12 years of age); and (3) intends to vaccinate daughters in neither age group (ie, neither 16–18 nor 9–12 years of age). We did not create a group including mothers who intended to vaccinate only younger daughters, because this represented a very small fraction of participants (<1%). Using multinomial logistic regression, we determined which factors were associated with intention to vaccinate older and younger daughters, and intention to vaccinate older daughters only, compared with intention to vaccinate daughters in neither age group (the reference group).
Of the 10 521 eligible mothers who received a survey, 8832 returned a completed survey (response rate 84%), and 7207 mothers reported on the survey that they had at least 1 daughter. The age of the child enrolled in the GUTS ranged from 19 to 24 years of age, but information was not available as to the number or ages of daughters. Demographic characteristics of the sample are shown in Table 1. Cronbach's coefficient α for the scale measuring beliefs about Papanicolaou testing was .76 and for beliefs about HPV vaccines was .75, indicating good internal consistency reliability.
Mothers reported a higher likelihood of vaccinating an older, compared with a younger, daughter: 48% were extremely or somewhat likely to vaccinate a daughter if she were 9 to 12 years of age, 68% if she were 13 to 15 years of age, and 86% if she were 16 to 18 years of age. Forty-eight percent of mothers were extremely or somewhat likely to receive the vaccine themselves if recommended for women their age. Likelihood of vaccinating older versus younger daughters is shown in Figure 1. Just under half of mothers (48%) intended to vaccinate a younger and older daughter, 38% intended to vaccinate only an older daughter, 13% did not intend to vaccinate a daughter in either age group, and 0.2% intended to vaccinate only a younger daughter.
Variables significantly associated with intention (defined as extremely or somewhat likely) to vaccinate a younger (9–12 years old) daughter in unadjusted logistic regression models included: demographic factors, history of HPV and Papanicolaou testing, intention to get a Papanicolaou test, cigarette smoking, communication with daughter about Papanicolaou testing, belief that daughter should get a Papanicolaou test regularly, personal beliefs about Papanicolaou testing, and beliefs about HPV vaccines (Table 2, model 1). The odds of intention to vaccinate a daughter for participants who strongly agreed, versus strongly disagreed, with each of the items comprising the scale measuring beliefs about HPV vaccines ranged from 3.4 to 17.4. The highest odds ratios (>10.0) were those associated with beliefs that the HPV vaccine protects against cancer, and that vaccines are a good way to protect a daughter's health. In adjusted logistic regression models, variables significantly associated with intention to ensure a younger (9 to 12 years old) daughter received the vaccine included belief that one's daughter should get a Papanicolaou test regularly and beliefs about HPV vaccines (Table 3, model 1). A 1-unit increase in the mean HPV vaccine belief scale score was associated with almost 5 times the odds of intending to vaccinate one's 9- to 12-year-old daughter against HPV.
When the outcome variable was categorized in 3 ways to explore factors associated with intention to vaccinate daughters in different age groups, variables significantly associated with intention in unadjusted logistic regression models were largely consistent with the previously described analyses in which the outcome variable was intention to vaccinate a younger daughter (Table 2, model 2). However, the magnitude of the odds ratios associated with beliefs about HPV vaccines was higher in the latter set of analyses. The multinomial logistic regression models demonstrated that 4 variables were independently associated with intention to vaccinate older and younger daughters, and older but not younger daughters, compared with daughters in neither age group: history of HPV or HPV-related disease in the mother, mother's report of daughter having had a Papanicolaou test in the past year, mother's belief that daughter should get a Papanicolaou test regularly, and beliefs about HPV vaccines (Table 3, model 2). The odds of intention to vaccinate an older daughter increased by almost 15 times, and the odds of intention to vaccinate an older as well as a younger daughter increased by 50 times (compared with intention to vaccinate a daughter in neither age group), for each increase of 1 point in the beliefs about HPV vaccine scale score.
In this study of US mothers, 48% of participants reported that they would be likely to vaccinate a daughter <13 years of age against HPV. Our findings differ from those of previous studies, which have shown higher levels of intention. In recent surveys of parents conducted in the United States, Canada, and the United Kingdom, 74% to 81% of parents reported that they would agree to vaccinate daughters <13 years of age.4,5,7 It is possible that the lower rates of intention reported by the mothers in our study may be explained by differences in study sample, recruitment, and data collection. Although one might expect nurses to be even more accepting of vaccination than other parents, participants in this study had a relatively high mean income level, and higher socioeconomic status may be associated with lower acceptance of HPV and other vaccines.7 Two of the previous studies involved telephone interviews. Parents may provide more positive responses about intention to vaccinate in an interview than they would in a confidential written survey, if they consider positive responses to be desirable to the interviewer. In addition, each study presented HPV vaccination in a different context. For example, in the study of Canadian parents, investigators asked about vaccination in the context of a publicly funded, school-based vaccination program, which may be more feasible than privately funded programs in the United States. Finally, our response rate of 84% was higher than the response rates of previous studies, which ranged from 22% to 57%. It is possible that those who chose to participate in the previous studies were more interested in a vaccine-related study and more accepting of vaccines.
Intention to vaccinate a daughter <13 years of age was substantially lower than intention to vaccinate older daughters, despite the fact that the ACIP recommends targeting vaccination to girls 11 to 12 years of age. This reported reluctance to vaccinate younger girls is similar to findings of previous studies of parents6,13,14 and clinicians.15–18 In a qualitative study that enrolled the mothers of 8- to 14-year-old girls, Waller et al13 found that although mothers were enthusiastic about vaccination in general, some mothers in the study were concerned that vaccination could lead to an increase in risky sexual behavior and other STIs, and would be uncomfortable discussing the HPV vaccine with younger girls. Although almost 90% of mothers in our study intended to vaccinate a daughter >16 years of age, the finding that fewer than 50% intended to vaccinate a daughter <13 years of age is concerning, because HPV vaccination is most effective if given before sexual initiation and, thus, exposure to genital HPV infection.19 Our results suggest that public health messages should focus on the importance of vaccinating younger adolescents and reinforce current ACIP guidelines.
Identification of the factors linked to intention to receive the HPV vaccine is important in that it can inform the development of evidence-based interventions designed to increase vaccination rates in young women. A large body of theoretical and empirical work has demonstrated that intention is strongly linked to health-related behaviors, such as the decision to be vaccinated.20,21 Thus, interventions that aim to increase intention by targeting the factors associated with intention may impact health-related behaviors. Mothers' beliefs about the importance of regular Papanicolaou testing and about HPV vaccines were associated with vaccinating a daughter <13 years of age. Mothers who had a history of HPV, who reported that their daughter had had Papanicolaou testing, who believed their daughter should have regular Papanicolaou testing, and who had more positive beliefs about HPV vaccination were more likely to report that they would vaccinate older daughters only, as well as both older and younger daughters, as compared with mothers who did not intend to vaccinate daughters in either age group. In summary, a group of factors including gynecologic history, beliefs about cervical cancer prevention, and beliefs about HPV vaccines seem to be key factors influencing mothers' intention to vaccinate their daughters against HPV. The very high odds ratios associated with beliefs about HPV vaccines suggest that these beliefs may be one of the most powerful modifiable factors in a mother's decision to have her daughter vaccinated.
The individual items comprising the scale measuring beliefs about HPV vaccination assessed perceived benefits of vaccination in general and HPV vaccination specifically (ie, protection against genital warts and cervical cancer), barriers to vaccination including adverse behavioral consequences of vaccination, severity of HPV, susceptibility to HPV and HPV-related disease, and belief that one's clinician would recommend vaccination. In univariate analyses, all were significantly associated with intention to vaccinate one's daughter, but the strongest predictors of intention were the belief that HPV vaccination would be the best protection against cervical cancer and that vaccination is a good way to protect a daughter's health. These findings are consistent with those of previous studies examining parental acceptance of HPV and other STI vaccines, in which parents consistently report a strong desire to protect their children from disease through vaccination.13,14,22–24 Previous studies have also shown that parental acceptance of HPV vaccination is associated with fewer perceived barriers (such as concern for riskier adolescent sexual behaviors postvaccination), perceived susceptibility to HPV, and physician endorsement.5–7 Collectively, the findings of our study and others demonstrate that a fairly consistent set of modifiable factors are associated with intention to vaccinate one's daughter against HPV. These factors may be readily targeted in office-based or public health interventions focusing on parents. For example, clinicians and health educators may be most effective in their efforts to encourage parents to consider vaccinating their children if they highlight the safety of HPV vaccines and their anticipated high efficacy in preventing cervical cancer, emphasize that HPV is common and that HPV-related diseases may be serious, and explicitly state their own support of vaccination.
One of the limitations of this study is that mothers were all nurses, most were white, and mean income was relatively high; thus, findings are not necessarily generalizable to all US mothers. Mothers who are nurses may be more likely than other mothers to believe in the utility and safety of vaccination25 and to value vaccination.26 Although all participants had at least 1 daughter, the age of each daughter was unknown, and it is likely that few had daughters 9 to 12 years of age at the time of this study, so the questions were in part hypothetical. The fact that all participants had older adolescent or young adult children may have influenced their responses regarding willingness to vaccinate older daughters. In this study, we assessed intention to receive the vaccine rather than actual vaccine acceptance. However, intention is one of the strongest predictors of actual behavior,21,27 and a recent study of parents in the United Kingdom demonstrated that actual uptake was similar to parental intention as reported in previous survey studies.8 Despite these limitations, this study has a number of unique strengths including the large number of participants, national study sample, assessment of attitudes postlicensing, and assessment of intention to vaccinate daughters of varying ages. In addition, nurses' beliefs may be especially important to assess; nurses may be important to target in interventions, because parents look to clinicians for guidance about vaccination.
The findings of this study, in combination with results of the evolving literature on HPV vaccine acceptability, provide information that can be used to develop effective office-based, community, and public health interventions that may help to improve acceptance of HPV vaccination among mothers. The development of evidence-based key messages may increase vaccine acceptability among parents and help to maximize vaccine uptake.28
This study was supported by American Cancer Society grant RSGPB-04-009-01-CPPB (to Dr Frazier).
- Accepted September 24, 2008.
- Address correspondence to Jessica A. Kahn, MD, MPH, Cincinnati Children's Hospital Medical Center, Division of Adolescent Medicine, MLC 4000, 3333 Burnet Ave, Cincinnati, OH 45229. E-mail:
The results of this study were presented in part at the annual meeting of the Pediatric Academic Societies, AAP Presidential Plenary Session; May 4, 2008; Honolulu, HI.
Financial Disclosure: Dr Rosenthal received grant or research support and has been either a consultant or member of the advisory board or review panel for Merck, GlaxoSmithKline, and Sanofi-Pasteur; and Dr Zimet is a co-principal investigator for an investigator-initiated proposal funded by Merck, and he has served on the speaker's bureau for Merck and provided consultation to M2 Communications and SciMed. The other authors have no financial relationships relevant to this article to disclose.
What's Known on This Subject
Parental attitudes about HPV vaccines will be key determinants of adolescent vaccination, but previous studies have demonstrated that some parents have specific concerns about vaccinating their daughters. Additional information about HPV vaccine acceptability in US mothers is needed.
What This Study Adds
The findings of this study suggest that educational interventions designed to affect mothers' willingness to vaccinate their daughters against HPV should focus on HPV vaccine efficacy, behavioral impact of vaccination, perceived risk of HPV, and clinician support for vaccination.
- ↵Ault KA. Effect of prophylactic human papillomavirus L1 virus-like-particle vaccine on risk of cervical intraepithelial neoplasia grade 2, grade 3, and adenocarcinoma in situ: a combined analysis of four randomised clinical trials. Lancet.2007;369 (9576):1861– 1868
- ↵Paavonen J, Jenkins D, Bosch FX, et al. Efficacy of a prophylactic adjuvanted bivalent L1 virus-like-particle vaccine against infection with human papillomavirus types 16 and 18 in young women: an interim analysis of a phase III double-blind, randomised controlled trial. Lancet.2007;369 (9580):2161– 2170
- ↵Ogilvie GS, Remple VP, Marra F, et al. Parental intention to have daughters receive the human papillomavirus vaccine. CMAJ.2007;177 (12):1506– 1512
- ↵Brabin L, Roberts SA, Stretch R, et al. Uptake of first two doses of human papillomavirus vaccine by adolescent schoolgirls in Manchester: prospective cohort study. BMJ.2008;336 (7652):1056– 1058
- ↵Berkey CS, Rockett HR, Field AE, et al. Activity, dietary intake, and weight changes in a longitudinal study of preadolescent and adolescent boys and girls. Pediatrics.2000;105 (4). Available at: www.pediatrics.org/cgi/content/full/105/4/e56
- ↵Kahn JA, Rosenthal SL, Hamann T, Bernstein DI. Attitudes about human papillomavirus vaccine in young women. Int J STD AIDS.2003;14 (5):300– 306
- ↵Waller J, Marlow LA, Wardle J. Mothers' attitudes towards preventing cervical cancer through human papillomavirus vaccination: a qualitative study. Cancer Epidemiol Biomarkers Prev.2006;15 (7):1257– 1261
- ↵Dempsey AF, Zimet GD, Davis RL, Koutsky L. Factors that are associated with parental acceptance of human papillomavirus vaccines: a randomized intervention study of written information about HPV. Pediatrics.2006;117 (5):1486– 1493
- ↵Salmon DA, Moulton LH, Omer SB, et al. Knowledge, attitudes, and beliefs of school nurses and personnel and associations with nonmedical immunization exemptions. Pediatrics.2004;113 (6). Available at: www.pediatrics.org/cgi/content/full/113/6/e552
- Copyright © 2009 by the American Academy of Pediatrics