Order 638034: review of literature portion for paper

Preventive Medicine 95 (2017) 26–37

Contents lists available at ScienceDirect

Preventive Medicine

journal homepage: www.elsevier.com/locate/ypmed

Review Article

Factors associated with parents’ attitudes to the HPV vaccination of their adolescent sons : A systematic review

Gorjana Radisic ⁎, Janine Chapman, Ingrid Flight, Carlene Wilson Flinders Centre for Innovation in Cancer, Flinders University, School of Medicine, GPO Box 2100, Adelaide SA 5001, Australia Cancer Council SA, 202 Greenhill Road, Eastwood, SA 5063, Australia

⁎ Corresponding author at: Flinders Centre for Innovati E-mail addresses: gorjana.radisic@flinders.edu.au (G. R

(C. Wilson).

http://dx.doi.org/10.1016/j.ypmed.2016.11.019 0091-7435/© 2016 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history: Received 15 March 2016 Received in revised form 15 November 2016 Accepted 24 November 2016 Available online 6 December 2016

Objective. The objective of the study was to identify factors associated with human papilloma virus (HPV) vac- cine acceptability in parents of adolescent boys. This information is critical to the development of approaches to optimise HPV vaccine uptake among this population group.

Methods. We performed a systematic search of the literature in addressing factors influencing parental atti- tudes to and beliefs about HPV vaccine and its acceptability for use. The findings were organised within the framework of the Health Belief Model (HBM) and summarised using a semi quantitative method.

Results. Eighteen studies met the inclusion criteria. Parental decisions were predominantly shaped by the per- ceived benefits of the vaccine; perceived risk of sons contracting the HPV infection, and having recommendations from health care providers. Fear of side effects and uncertainty about vaccine effectiveness, as well as cost and lack of healthcare, were barriers to HPV vaccination. Other factors such as knowledge, family characteristics, par- ent–child dialogue and egalitarian values appeared to be important when deciding whether to vaccinate boys.

Conclusions. HPV vaccine uptake among male adolescents is suboptimal. Future programs need to address the predictors of uptake by educating parents about the boys’ high susceptibility to infection, the benefits of vaccina- tion, and reduce concerns regarding perceived barriers. Additionally, uptake may be facilitated by encouraging health care provider endorsement, particularly in countries without government-funded immunisation programs.

© 2016 Elsevier Inc. All rights reserved.

Keywords: HPV vaccination Health behaviour Adolescent boys Parental attitudes and beliefs

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2. Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2.1. Inclusion/exclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.2. Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.3. Quality appraisal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.4. Data extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2.5. Data synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2.5.1. Categorisation of variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.5.2. Coding association of HPV vaccination-related factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 3.1. HBM-related factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3.1.1. Perceived risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 3.1.2. Perceived benefit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.1.3. Perceived barriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.1.4. Cues to action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

3.2. Other factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.2.1. Knowledge and awareness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

on in Cancer, Flinders University, School of Medicine, GPO Box 2100, Adelaide SA 5001, Australia. adisic), Janine.chapman@flinders.edu.au (J. Chapman), Ingrid.flight@flinders.edu.au (I. Flight), Carlene.wilson@flinders.edu.au

27G. Radisic et al. / Preventive Medicine 95 (2017) 26–37

3.2.2. Family characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.2.3. Parent–child dialogue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.2.4. Attitudes towards vaccinations in general . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.2.5. Miscellaneous beliefs around egalitarianism, child sexual activity, religion and culture . . . . . . . . . . . . . . . . . . . . . . 35 3.2.6. Other potential variables of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 4.1. Strengths and limitations of the study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Transparency document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

1. Introduction

Human papillomavirus (HPV) infection is the most common sexual- ly transmitted infection worldwide (Bosch et al., 2002; Chaturvedi, 2010). Its role in the aetiology of cervical, anal, penile and a subset of head and neck cancers, as well as genital warts is well established (Bosch et al., 2002; deMartel et al., 2012; Gillison et al., 2008; Giuliano et al., 2014). Three currently available HPV vaccines, Cervarix® (biva- lent vaccine against HPV16 and −18), Gardasil® (quadrivalent vaccine against HPV16, −18, −6, and −11) and Gardasil®-9 (nonavalent vac- cine against HPV16, −18, −31, −33, −45, −52, −58, −6, and −11) are effective against strains of the virus that are the major cause of pre- cancerous lesions (Frazer, 2010; Siddiqui et al., 2006; Zhai and Tumban, 2016).

Following the recommendation for vaccination of adolescent boys (Giuliano et al., 2011), a small but growing number of countries (includ- ing Australia, Austria, some provinces in Canada and Italy, to date) are providing government-funded HPV vaccination for eligible boys as well as girls. However, recent data from the National HPV Program Reg- ister in Australia indicate that boys are less likely than girls to participate in the free national school-based vaccination program, with fewer boys than girls finishing the recommended three dose course of vaccination (National HPV Program Register, 2015). Global data are consistent (Berenson and Rahman, 2012; Gilkey et al., 2012).

Considering that the HPV vaccine is most effective if given prior to sexual debut (11–13 years), uptake is critically dependent upon paren- tal acceptance of the vaccine for their sons. Identifying factors that are associated with parental acceptability regarding the vaccine in boys is important because parental decisions about the HPV vaccine will have a critical impact on the vaccine’s success in preventing HPV-associated diseases.

Social cognition models are considered an important tool for under- standing and predicting factors associated with health-related decision making. The Health Belief Model (HBM; Becker, 1974) is an established model that – along with others including the Theory of Planned Behav- iour, Protection Motivation Theory, and Precaution Adoption Process Model – has been previously utilised as a framework to predict and or- ganise factors associated with HPV vaccination (see Brewer and Fazekas, 2007; Dempsey et al., 2011; Gerend and Shepherd, 2012; Gainforth et al., 2012; Perez et al., 2015).

The HBM provides a basis for understanding decisions associated with vaccination acceptance or non-acceptance in terms of perceived risk (the extent to which a parent perceives their child to be at risk of se- rious health problems associated with HPV); beliefs concerning the ben- efits or efficacy of the vaccine; beliefs concerning the costs of, or barriers to, getting the vaccination (e.g. logistical barriers, fear of side effects and vaccine safety, financial hurdles), and finally, exposure to cues to action that might prompt the decision to be vaccinated. These cues may include a range of triggers, for example, recommendation from a health profes- sional or social influence (Conner and Norman, 2005).

Two systematic reviews by Brewer and Fazekas (2007) and Cunningham et al. (2014) employed constructs from the HBM as a framework for analysing decision-making about daughters' HPV

vaccine acceptability in the United States and African countries, respec- tively. Findings in relation to the HMB constructs in each review were similar across countries, indicating that perceived risk, in terms of sus- ceptibility to HPV infection; beliefs around the benefits of HPV vaccine; barriers including concerns about promoting sexual behaviour, associ- ated costs and accessibility, and cues to action such as receiving a recommendation from a health professional were underlying deter- minants of vaccine acceptability. As such, the authors conclude that the HBM is a useful framework for organising and understanding rel- evant factors associated with parental HPV vaccine decision-making in girls.

While a number of recent systematic reviews have assessed parental knowledge, attitudes and beliefs regarding HPV vaccination (e.g. Allen et al., 2010; Holman et al., 2014, Kessels et al., 2012; Trim et al., 2012), the current review considers parental acceptability of the HPV vaccina- tion in adolescent boys only. The use of the HBM as a guiding framework to structure the findings allows comparison to previous reviews (Brewer and Fazekas, 2007; Cunningham et al., 2014). However, it should be noted that, like most predictive social cognition models, the HBM is limited in its ability to cover all possible factors associated with a health outcome and has been found to have more utility in the prediction of behavioural intention (e.g. intention to vaccinate) than be- haviour per se (e.g. vaccine initiation) (Armitage and Conner, 2000). Therefore, the current review includes other modifying factors exam- ined in studies that are not formally part of the HBM, including vac- cine-related knowledge, which is likely a pre-requisite for decision- making (Waller et al., 2010). This review also separates findings relating to reported intention or willingness to vaccinate, and actual vaccine ini- tiation, for comparison purposes.

2. Method

The review was developed following the preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidelines (Moher et al., 2009). A comprehensive search strategy was developed in consul- tation with a senior librarian. An extensive search across the electronic databases Ovid MEDLINE, PubMed, PsychInfo, Informit, Web of Science, Cochrane and Scopus was preformed to locate relevant articles. To iden- tify applicable words and MeSH (medical subject heading) terms, a pre- liminary manual search of the literature was performed. The combination of the following text words and MeSH terms was used: (human papilloma virus* or HPV* or papillomavirus*) AND (immunis* or immuniz* or vaccin*) AND (accept* or aware* or attitud* or knowledg* or predict* or percept* or decision* or choice* or motivat* or social norm* or value* or uptak* or educat* or barrier* or uncertain* or refus* or participat* or behavio?r* or attitud* or knowledg* or literate or literacy or cognitive* or intent*). The search was performed in July 2015 (ending date 10th of July 2015).

2.1. Inclusion/exclusion criteria

Studies were included if they addressed factors influencing parental attitudes to vaccination, intention to vaccinate or actual vaccination of

28 G. Radisic et al. / Preventive Medicine 95 (2017) 26–37

adolescent boys (9–18 years old) for HPV. Studies of parents with ado- lescent sons were included; studies with findings for sons and daugh- ters were only included if the data for sons were reported separately. The primary outcome measured in this study was parental acceptability of HPV vaccine. HPV acceptability was conceptualised as holding a pos- itive intention or willingness towards vaccinating sons in the future (vaccine intention), or by having consented to their sons being vaccinat- ed in the past (vaccine initiation). HPV non-acceptability was conceptualised as holding a negative intention or being unwilling to vaccinate sons in the future. The search was limited to English language and studies were included if they collected data following the licensure of HPV vaccine for males in October 2009. Both qualitative and quanti- tative studies were included. Only journal articles based on original re- search were included. Reviews, editorials, conference abstracts and dissertations were excluded.

2.2. Study selection

A total of 13,065 articles were identified. Following the removal of duplicates (n = 5656), 7409 titles were screened for inclusion. 5868 were excluded and 1541 were further reviewed. A further 1152 articles were excluded following abstract review and 389 records were fully reviewed based on the eligibility criteria. The reference sections of in- cluded articles and relevant reviews were also inspected for potentially eligible studies not retrieved by electronic search. A total of 18 studies met the criteria for the inclusion in this review: 16 studies were selected through screening of the records and 2 additional studies were sourced

Records identified through database searc (n=13065)

Records after duplicates removed (n=7409)

Title review (n=7409)

Abstract review (n=1541)

Full text review (n=389)

Studies meeting eligibility criteria (n=16)

Studies included in this review (n=18)

Fig. 1. Study selec

from reference lists. The results of the study selection process are shown in Fig. 1.

2.3. Quality appraisal

The Mixed Methods Assessment Tool (MMAT) developed by Pluye et al. (2009) was utilised by two independent reviewers (GR and CW) to evaluate the quality of the selected studies and to identify factors which might have introduced bias or limited the generalizability of the results. The MMAT was designed for use in systematic reviews with mixed studies and it allows researchers to appraise studies with di- verse common designs (qualitative, quantitative and mixed methods) (Pace et al., 2012; Plye and Hong, 2014). The MMAT awards scores based on four criteria: the methods used for sample selection; the ade- quacy of statistical tests; completeness of the dataset and the risk of misclassification bias. The final MMAT quality score can be 0, 25, 50, 75 or 100%, depending how many of the four criteria are met (0% = no criteria met and 100% = all criteria are met.)

2.4. Data extraction

Data extraction was performed by the lead author (GR). Two addi- tional authors (JC and CW) independently reviewed and summarised the studies. Any discrepancies or disagreements were resolved through discussion. Following the methodology used in previous reviews (Brewer and Fazekas, 2007; Cunningham et al., 2014), findings were organised using the guiding framework of the HBM, plus other relevant

h Medline: 2834

Pubmed: 496

Psych Info: 2804

Scopus: 3705

Informit: 76

Web of Science: 2988

Cochrane 162

Excluded titles: 5868

Excluded abstracts: 1152

Records excluded after

full text review (n=373)

Additional titles identified by

reference list search (n=2)

tion process.

Table 1 Characteristics of studies.

Author/country Study aim Study design/analysis type/theoretical model

Study setting Sample size Knowledge assessed (Y/N)/method

MMAT

Alexander et al., 2014

USA

Identify factors associated with the initiation of HPV vaccination of sons

Qualitative

– Interviews – Inductive content analysis – No theoretical model

Primary care clinic N = 21 dyad (parent and son, aged 13–17 years)

Parents asked to recall information about HPV vaccine given by HCP before the interview. Association between knowledge and vaccine acceptability not assessed

75%

Berenson and Rahman, 2012

USA

Identify factors associated with intention and initiation of HPV vaccination of sons, but did not distinguish between intention and initiation when reporting findings

Quantitative

– Cross-sectional survey (self-administered)

– Descriptive findings only – No theoretical model

Five publicly funded health clinics

N = 322 parents with ≥1 child aged 9–26 years

Parents asked about awareness and availability of HPV vaccine. Association between knowledge and vaccine acceptability not assessed

50%

Bianco et al., 2014

Italy

Identify factors associated with intention to vaccinate sons

Quantitative

– Cross-sectional survey (self-administered)

– Descriptive findings + partial statistical testing (logistic regression)

– No theoretical model

Seven public secondary schools

N = 565 parents with sons aged 10– 14 years

Yes

– True/false questionnaire about HPV infection and vaccination

50%

Cates et al., 2012 USA

Identify factors associated with intention to vaccinate sons following exposure to motivational material

Qualitative

– Focus groups and interviews – Coding categories using con-

stant comparative method – Questions and coding proto-

col derived from Health Be- lief Model

Community centres, Paediatric and adolescent health clinic

N = 29 (5 focus groups) parents with sons aged 11– 12 + N = 100 (interviews) parents with sons aged 9–13 years

Reported low parental awareness about HPV disease and HPV vaccine but method not reported. Association between knowledge and vaccine acceptability not assessed

100%

Donahue et al., 2014

USA

Identify factors associated with intention to vaccinate sons and explore reasons for non-vaccination

Quantitative

– Cross-sectional survey (web-based)

– Descriptive findings only – No theoretical model

National sample of parents who were members of Survey Sampling International (SSI) recruited via e mail

N = 779 parents with sons aged 11– 17 years

Yes

– Lack of knowledge was offered as a reason for vaccine non-acceptability in the survey

75%

Gainforth et al., 2012

Canada

Identify factors associated with intention to vaccinate sons following information provision

Quantitative

– Cross-sectional survey (self-administered)

– Statistical testing (logistic regression)

– Theory of planned behav- iour and Protection Motiva- tion Theory

Child centred community and sporting events

N = 137 parents with sons in grade 5.6 or 7

No 50%

Gilkey et al., 2012

USA

Identify factors associated with initiation of HPV vaccination of sons

Quantitative

– Cross-sectional survey (telephone-administered)

– Descriptive findings only – No theoretical model

The State of North Carolina, phone recruitment

N = 359 parents of sons aged 11– 17 years

Yes

– Lack of knowledge was of- fered as a reasons for vac- cine non-acceptability in the survey

100%

Griebeler et al., 2012

USA

Identify factors associated with intention and initiation of HPV vaccination of sons, but did not distinguish between intention and initiation when reporting findings

Quantitative

– Cross-sectional survey (self-administered)

– Descriptive findings only – No theoretical model

Paediatric clinic N = 102 parents of sons aged 9– 20 years (82% of boys ≤16)

Yes

– Likert scale about male health concerns and possible health outcomes

0–25%

Hansen et al., 2015

USA

Identify factors associated with intention to vaccinate sons

Qualitative

– Interviews – Thematic analysis – No theoretical model

Primary care clinic N = 45 parents (54% parents of sons aged 10– 18 years)

No 75%

Mortensen, 2010 Denmark

Identify factors associated with intention to vaccinate sons following brief information provision

Quantitative

– Cross-sectional survey (telephone-administered)

– Descriptive findings only – No theoretical model

National sample of Danish parents recruited through random digit dialling

N = 450 parents of sons aged 12– 15 years

Yes

– Survey offered knowledge-related an- swers to question about

75%

(continued on next page)

29G. Radisic et al. / Preventive Medicine 95 (2017) 26–37

Table 1 (continued)

Author/country Study aim Study design/analysis type/theoretical model

Study setting Sample size Knowledge assessed (Y/N)/method

MMAT

the reasons for HPV vac- cine refusal

Mortensen et al., 2015

Denmark

Identify factors associated with intention to vaccinate sons in four European countries: UK, Germany, France and Italy

Quantitative

– Cross-sectional survey (self-administered)

– Descriptive findings + partial statistical tests (chi-square)

– No theoretical model

International sample (4 countries, UK, Germany, France and Italy); In France-randomly extracted from Yellow Pages; in the UK-50% omnibus and 50% yellow pages; in Germany and Italy-door to door recruitment

N = 1837 parents of sons aged 11– 17 years

Yes

– Lack of knowledge was offered as a response to barriers of vaccination in the survey

50%

Perez et al., 2015 Canada

Identify factors associated with intention and initiation of HPV vaccination of sons (reported intention and initiation findings separately)

Mixed-method

– Cross-sectional survey (self-administered)

– Open-ended questions – Descriptive findings only – Precaution Adoption Pro-

cess Model

National sample of parents who were registered as a part of the national panel of a polling and market research firm

N = 2874 parents of sons aged 9– 16 years

Yes

– Lack of knowledge was offered as a reason for vaccine non-acceptability the survey

75%

Perkins et al., 2013

USA

Identify factors associated with intention and initiation of HPV vaccination of sons following a short educational paragraph (reported intention and initiation findings separately)

Quantitative

– Cross-sectional survey (self-administered)

– Descriptive findings + partial statistical testing (logistic regression)

– No theoretical model

Paediatric and adolescent practices (academic and urban health centres)

N = 120 parents of sons (mean age 14 years)

Yes

– True/false questionnaire about HPV, but associa- tion between knowledge and vaccine acceptability not assessed

50%

Reiter et al., 2010

USA

Identify factors associated with mother’s intention to vaccinate sons (if free or if cost involved) following informative statement

Quantitative

– Cross-sectional survey (web-based)

– Statistical testing (t test; linear regression)

– No theoretical model

National sample of parents that were existing members of the national panel of U.S. households

N = 406 mothers of sons aged 9– 18 years

Yes

– Likert scale about aware- ness of HPV vaccine availability for boys

50%

Reiter et al., 2013

USA

Identify factors associated with the initiation of HPV vaccination of sons

Quantitative

– Before and after survey (baseline and one year at follow up after brief infor- mation provision)

– Descriptive findings + partial statistical tests (t-tests)

– Health Belief Model

National sample Baseline: N = 547 parents and Follow up: N = 327 parents of sons aged 11– 16 years

Yes

– Lack of knowledge was offered as a reason for vaccine non-acceptability the survey

100%

Schuler and Coyne-Beasley, 2015

USA

Identify factors associated with intention to vaccinate sons

Quantitative

– Cross-sectional survey (self-administered)

– Descriptive findings only – No theoretical model

Paediatric clinic N = 267 parents of sons aged 9– 21 years

Yes

– True/false questionnaire about HPV, but associa- tion between knowledge and acceptability not assessed

75%

Taylor et al., 2014

USA

Identify factors associated with the initiation of HPV vaccination of sons

Cross sectional (web based survey)

– Statistical test: Logistic re- gression

– No theoretical model applied

National sample of parents that were part of the SSI national sample, recruited through e mail

N = 758 parents of sons aged 11– 17 years

No 75%

Tisi et al., 2013 Italy

Identify factors associated with intention and initiation re: HPV vaccination of sons following information provision, but did not distinguish between intention and initiation when reporting findings

Cross sectional (survey)

– Descriptive findings only – No theoretical model

applied

Parents selected from the registry list of the Province of Brescia

N = 161 parents of sons aged 11– 15 years

Yes

– Lack of knowledge was offered as a reason for vaccine non-acceptability the survey

50%

Note: Findings are reported only when they represent independent results for sons while data on sons and daughters combined are excluded if they could not be separately interpreted. MMAT = Mixed Methods Assessment Tool.

30 G. Radisic et al. / Preventive Medicine 95 (2017) 26–37

31G. Radisic et al. / Preventive Medicine 95 (2017) 26–37

factors (described below). A considerable heterogeneity was demon- strated in the quantitative studies with a lack of uniformity of the evi- dence they offered, therefore, consistent with the earlier reviews, the findings are presented as a narrative synthesis rather than meta-analy- sis. The majority of studies reported descriptive findings in relation to the variables of interest, while others reported results of statistical tests for some variables but not others (details in Table 1). Due to this, the findings in the review are consolidated using a summary percent- age, calculated by dividing the number of respondents who answered affirmatively to the factor (associated with their HPV acceptance or non-acceptance) by the total number of participants who were asked the question. p values are reported where statistical significance of a predictor variable was available.

2.5. Data synthesis

The findings were then categorised to summarise the state of the lit- erature for different variables. In order to give an indication of the strength of each variable associated with parental HPV acceptability, we drew upon methodology used in previous reviews of health behav- iour (Sallis et al., 2000), adapted to accommodate the studies in the cur- rent review which cannot be classified on the basis of statistical significance. The following coding rules were applied.

2.5.1. Categorisation of variables Conceptually-similar variables were categorised under the broad

HBM headings: perceived risk; perceived benefits; perceived barriers; cues to action, and other relevant variables including knowledge and awareness; family characteristics including parental and child age and ethnicity; the nature of parent–child dialogue; attitudes to vaccinations in general, and holding miscellaneous beliefs around egalitarianism, child sexual activity, or religion and culture. Variables are not shown in the summary tables unless comparison findings from two or more studies were available. Although three or more comparisons have been used previously (Sallis et al., 2000), two was chosen in the current review because of the comparatively small number of studies available. For clarity and ease of reference, the findings associated with ‘HPV ac- ceptability’ and ‘non-acceptability’ are listed in separate tables (Tables 2 and 3).

2.5.2. Coding association of HPV vaccination-related factors In each variable category, the percentages of participants answering

affirmatively to the variable are reported for each study. All studies reporting between 1 and 29% of total participant agreement are listed under the column ‘low association’; studies reporting between 30 and 59% of total participant agreement are listed ‘medium association’, and those reporting 60–100% of total participant agreement are listed under ‘high association.’ Where a study reported a finding with a statis- tically significant result rather than a percentage, this was listed under ‘high association’. If more than one variable under the same category was reported in a single paper, the percentages were averaged and one score used. The exception to this is Mortensen et al. (2015) who present a cross-cultural comparison of four studies from four countries. Results from this paper were treated as separate studies and labelled by the country of administration.

The ‘Summary code’ column in each table contains a code to sum- marise the state of the literature for that variable. Each variable was coded low, med, or high association depending on which column had the greatest number of studies listed in it. ‘+’ was added when there were four or more studies and N50% fell into this category. The ‘?’ code indicates a variable that has been studied twice or more with an indeterminate or inconsistent result regarding the strength of the findings.

3. Results

Study characteristics are summarised in Table 1 in terms of the study location, setting, sample and aims (to investigate vaccine intention, ini- tiation or both); the research design; type of analysis used and whether the study uses a theoretical model; assessment of knowledge, and the MMAT quality assessment scores. Of the final 18 studies, 12 were con- ducted in the USA; with a further two studies in Canada, two in Italy and two in Denmark. Fourteen were quantitative studies; one mixed- method and 3 were qualitative, comprising focus groups and inter- views. The majority of quantitative studies (n = 13) were cross-section- al surveys. Four of the selected studies reported using social cognition models to investigate the determinants of vaccination. Nine studies re- ported findings relating to the intention to vaccinate; two studies re- ported separate findings for both intention and initiation; four studies reported findings related to vaccine initiation, and three studies did not distinguish between intention and initiation when discussing HPV vaccine acceptability. The majority of studies (n = 15) gave an indica- tion of the level of knowledge of HPV or the HPV vaccine in their sample using a variety of methods, but only 10 directly reported findings of the association between knowledge and the vaccine acceptability outcome (Table 1).

In terms of the MMAT, the quality of the included studies varied widely (total scores from 0 to 100%), suggesting that the majority of the studies were at some risk of bias (Table 1). Methodological problems encompassed selection bias caused by sampling strategy, variation and inconsistencies in measurements of the reported out- comes, variability in outcome definition and the reported response rates. The use of convenience sampling in certain settings was noted among studies. Several studies did not report the response rate, while others used only partial statistical testing to assess the outcome variables.

Table 2 (factors associated with acceptability) and Table 3 (factors associated with non-acceptability) summarise the results and strength of association with variables from studies reporting quantitative find- ings (n = 15). Findings that relate specifically to vaccine initiation are marked with an asterisk. Table 4 summarises the results from qualita- tive studies (n = 3). The following section describes the findings summarised in the tables under the main variable headings. Qualitative findings are indicated when discussed.

3.1. HBM-related factors

3.1.1. Perceived risk Of the 18 studies, 10 studies (56%) asked about the perceived risk of

adolescent boys acquiring a disease and whether it would influence the decision to vaccinate them. Parents in six studies (Bianco et al., 2014; Gainforth et al., 2012; Mortensen, 2010; Mortensen et al., 2015 (Germa- ny), Mortensen et al., 2015 (Italy), Mortensen et al., 2015 (UK)) believed their sons to be at risk of contracting the HPV infection and this was pos- itively associated with intention to vaccinate, although the spread of findings across low, medium and high association categories left the overall strength of the association undetermined (Table 2). However, the association with vaccine acceptability was supported by a qualita- tive study showing that fear of disease in sexually-active sons was asso- ciated with vaccine intention (Cates et al., 2012: Table 4). As expected, the parental perception that sons were not at risk was also associated with non-acceptability of the vaccine: parents who believed their sons were not at risk of HPV; or that the HPV infection was not serious in males were less likely to initiate the vaccination, or believe the vaccine to be necessary or suitable (Berenson and Rahman, 2012; Bianco et al., 2014; Gilkey et al., 2012; Griebeler et al., 2012 Mortensen, 2010; Reiter et al., 2013). The strength of the association with perceived risk and non-acceptability was consistent but low (Table 3).

Table 2 Variables positively associated with parental acceptability of HPV vaccination in boys.

Low association (selected by 1–29% of sample)

Medium association (selected by 30–59% of sample)

High association (selected by 60–100% of sample, or statistically sig. finding) Summary codea

HBM-related variables Perceived risk

Belief that child is at risk/ fear of disease 4% (Mortensen, 2010); 19% (Mortensen et al., 2015: UK)

42% (Mortensen et al., 2015; Germany); 50% (Mortensen et al., 2015: Italy)

p = 0.06 (Bianco et al., 2014); p = 0.05 (Gainforth et al., 2012)

? (Even split)

Perceived benefit Protect males from cancers/other disease (labelled)

18% (Bianco et al., 2014: GWarts); 13% (Mortensen, 2010: GWarts)

36% (Mortensen, 2010: GWarts); 46% (Bianco et al., 2014)

67% (Mortensen, 2010); 76% (Griebeler et al., 2012); 95% (Perkins et al., 2013); 89% (Griebeler et al., 2012: GWarts); 94% (Perkins et al., 2013: GWarts); 73% (Griebeler et al., 2012: Lifelong infection); 59% (Mortensen et al., 2015: UK, STDS); 63% (Mortensen et al., 2015: France, STDS); p b 0.01 (Gainforth et al., 2012)

High+

Protect son's future partner/both sexes 24% (Mortensen et al., 2015: UK); 19% (Mortensen et al., 2015: France); 9% (Mortensen, 2010)

36% (Bianco et al., 2014); 45% (Perez et al., 2015); 63%b (Perez et al., 2015)

87% (Griebeler et al., 2012); 93% (Perkins et al., 2013); 71% (Bianco et al., 2014); p b 0.01 (Reiter et al., 2010: If vaccine is free OR $400)

High

Welcome any protection against cancer/disease 25% (Mortensen, 2010); 25% (Mortensen et al., 2015: France)

51% (Mortensen et al., 2015: Germany); 49% (Mortensen et al., 2015: Italy)

92% (Tisi et al., 2013) ? (≈Even split)

Cues to action Recommendation from HCP/GP/visit HCP in the past year

18% (Bianco et al., 2014); 13% (Mortensen et al., 2015: France); 10% (Perez et al., 2015); 13%b

(Perez et al., 2015); 4% (Mortensen, 2010)

48%b (Perkins et al., 2013) 76% (Perkins et al., 2013); p b 0.01b (Reiter et al., 2013); p b 0.05b (Taylor et al., 2014)

Low+

Vaccine required or included in NIP 15% (Bianco et al., 2014); 9% (Mortensen, 2010) Low Social norms/influence of others 12% (Schuler and Coyne-Beasley, 2015) p b 0.05 (Gainforth et al., 2012) ?

Other Knowledge and awareness

Knowledge: higher knowledge of HPV/HPV vaccination/additional information

p b 0.05 (Reiter et al., 2010: if vaccine is free OR $400); p = 0.02 (Bianco et al., 2014)

High

Family characteristics Age of parents: younger p b 0.05b (Taylor et al., 2014); p b 0.01 (Perkins et al., 2013:

parents b40) High

Experience of cancer or genital warts (labelled) in family

9% (Bianco et al., 2014); 8% (Mortensen, 2010); 1% (Mortensen, 2010: GWarts)

Low

Ethnic background: non-white parents more willing to vaccinate than white parents

p b 0.05b (Taylor et al., 2014: non-white = black or Hispanic); p b 0.05 (Reiter et al., 2010: non-white = African American, if vaccine is $400)

High

Parent–child dialogue Having more frequent parent–child discussions around topic

p b 0.05b (Reiter et al., 2013); p b 0.05b (Taylor et al., 2014) High

Attitudes to vaccination in general General pro-vaccination beliefs 19% (Bianco et al., 2014); 21% (Mortensen,

2010); 23% (Mortensen et al., 2015: UK) p = 0.01 (Bianco et al., 2014) Low+

Intention to vaccinate p b 0.01b (Perkins et al., 2013); p b 0.05b (Reiter et al., 2013) High Egalitarian values

Belief that it's a shared responsibility 13% (Mortensen, 2010) 44% (Mortensen et al., 2015: Germany); 53% (Mortensen et al., 2015: Italy)

Med

Belief that both genders should have equal rights to vaccinations

5% (Tisi et al., 2013) p b 0.01 (Mortensen et al., 2015: Germany); p b 0.01 (Mortensen et al., 2015: Italy)

High

a Total low, medium or high association summary, with + where there are four or more studies and N50% fall into this category, ? where there is an indeterminate distinction between association categories. b Findings associated with vaccine initiation. All other findings are related to intention to vaccinate, or do not distinguish between intention and initiation.

3 2

G .R

adisic et

al./ Preven

tive M edicin

e 95

(20 17)

2 6–

37

Table 3 Variables positively associated with parental non-acceptability of HPV vaccination in boys.

Low association (selected by 1–29% of sample)

Medium association (selected by 30–59% of sample)

High association (selected by 60–100% of sample, or statistically sig. finding) Summary code1

HBM-related variables Perceived risk

Belief that son is not at risk/unsuitable/not serious in males

27% (Berenson and Rahman, 2012); 3% (Mortensen, 2010: M of 2 items); 17% (Bianco et al., 2014); 11% (Gilkey et al., 2012); 13% (Griebeler et al., 2012); 1% (Reiter et al., 2013)

Low+

Perceived barriers Vaccine too new 8% (Reiter et al., 2013: Time 2) 33% (Berenson and Rahman, 2012) ? (Low/med) Associated financial costs 7% (Berenson and Rahman, 2012); 7% (Bianco et al.,

2014); 1% (Perez et al., 2015) p b 0.01 (Reiter et al., 2010) Low+

Fear of side effects/safety 17% (Donahue et al., 2014: M of 2 items); 29% (Mortensen, 2010); 27% (Perez et al., 2015); 7% (Gilkey et al., 2012); 10% (Reiter et al., 2013: Time 2); 23% (Mortensen et al., 2015: UK)

59% (Schuler and Coyne-Beasley, 2015); 53% (Perkins et al., 2013); 46% (Mortensen et al., 2015: France); 38% (Griebeler et al., 2012)

67% (Bianco et al., 2014); 64% (Mortensen et al., 2015: Germany); 76% (Mortensen et al., 2015: Italy)

Low

Doubts around vaccine efficacy/utility 6% (Donahue et al., 2014); 4% (Perez et al., 2015); 27% (Bianco et al., 2014: M of 2 items)

Low

Cues to action Lack of recommendation from HCP/GP 14% (Mortensen, 2010); 27% (Gilkey et al., 2012);

17% (Reiter et al., 2013: Time 2); 23% (Mortensen et al., 2015: France)

61% (Donahue et al., 2014) Low+

Lack of recent HCP appointment 9% (Donahue et al., 2014); 5% (Gilkey et al., 2012); 11% (Reiter et al., 2013: Time 2)

Low

Other Knowledge and awareness

Don't know enough/unaware about HPV or HPV vaccination in boys

15% (Mortensen et al., 2015: UK); 21% (Mortensen et al., 2015: France); 23% (Reiter et al., 2013: Time 2); 17% (Gilkey et al., 2012); 3% (Tisi et al., 2013); 9% (Perez et al., 2015)

52% (Mortensen et al., 2015: Germany, M of 2 items); 53% (Mortensen et al., 2015: Italy, M of 2 items); 32% (Donahue et al., 2014)

70% (Mortensen, 2010) Low+

Family characteristics Age of son: son too young 7% (Bianco et al., 2014); 7% (Mortensen, 2010); 9%

(Gilkey et al., 2012); 23% (Mortensen et al., 2015: UK); 2% (Reiter et al., 2013: Time 2)

33% (Berenson and Rahman, 2012); 38% (Griebeler et al., 2012)

Low+

Fathers less likely to intend/vaccinate than mothers

p = 0.05 (Perkins et al., 2013); p b 0.05 (Reiter et al., 2013)

High

Parent–child dialogue Belief in child's right to decide/wishes 8% (Bianco et al., 2014); 2% (Mortensen, 2010); 6%

(Donahue et al., 2014) 39% (Perkins et al., 2013); 54% (Griebeler et al., 2012)

Low

Attitudes to vaccination General anti-vaccination beliefs/actions 10% (Bianco et al., 2014); 11% (Mortensen, 2010); 8%

(Griebeler et al., 2012); 17% (Mortensen et al., 2015: France); 13% (Perez et al., 2015) 16% (Gainforth et al., 2012)

54% (Mortensen et al., 2015: Germany) 73% (Mortensen et al., 2015: Italy); p = 0.05 (Perkins et al., 2013)

Low+

Beliefs around child sexual activity Concerns about encouraging sexual activity/promiscuity

6% (Berenson and Rahman, 2012); 21% (Bianco et al., 2014); 9% (Donahue et al., 2014)

Low

Religious/cultural beliefs Goes against cultural or religious beliefs/places high importance on religious beliefs

39% (Mortensen et al., 2015: UK) p b 0.05 (Reiter et al., 2010: If vaccine is free)

? (Med-high)

1 Total low, medium or high association summary, with + where there are four or more studies and N50% fall into this category, ? where there is an indeterminate distinction between association categories.

3 3

G .R

adisic et

al./ Preven

tive M edicin

e 95

(20 17)

2 6–

37

Table 4 Variables positively associated with parental acceptability and non-acceptability of HPV vaccination in boys (qualitative studies).

Acceptability Non-acceptability

HBM related variables Perceived risk

Fear of disease (if child sexually active)

Cates et al. (2012)

Perceived benefit Protect males from future cancers/disease

Alexander et al. (2014)a, Hansen et al. (2015)

Protect son's future partner Cates et al. (2012) Perceived barriers

Associated financial costs Cates et al. (2012) Fear of side effects/safety Cates et al. (2012),

Alexander et al. (2014)

Doubts around vaccine efficacy

Cates et al. (2012)

Other Family characteristics

Age of son: son too young Hansen et al. (2015) Support recommended age of vaccination

Hansen et al. (2015)

Attitudes to vaccination Trust in vaccine safety Alexander et al.

(2014) Beliefs around child sexual activity Concerns about encouraging sexual activity/promiscuity

Hansen et al. (2015)

Note: Findings are presented as a content analysis. a Findings associated with vaccine initiation. All other findings are related to intention

to vaccinate.

34 G. Radisic et al. / Preventive Medicine 95 (2017) 26–37

3.1.2. Perceived benefit Twelve studies (67%) reported a link between the perceived benefits

of HPV vaccination and HPV vaccination acceptance (Bianco et al., 2014; Donahue et al., 2014; Gainforth et al., 2012; Griebeler et al., 2012; Mortensen, 2010; Mortensen et al., 2015; Perez et al., 2015; Reiter et al., 2010; Tisi et al., 2013: Table 2); (Alexander et al., 2014; Cates et al., 2012; Hansen et al., 2015: Table 4). The most consistently cited reason among parents endorsing the HPV vaccination was the desire to protect their sons from cancer and other diseases, including genital warts and general lifelong infection (Bianco et al., 2014; Gainforth et al., 2012; Griebeler et al., 2012; Mortensen, 2010; Mortensen et al., 2015 (France); Mortensen et al., 2015 (UK); Perkins et al., 2013). The parental desire to protect sons' health had the highest association (in terms of the highest % of reported participant agreement, see Table 2) with HPV acceptability than any other factor. A high association was also found between vac- cine intention and the desire to protect the son's future partner (Bianco et al., 2014; Griebeler et al., 2012; Mortensen, 2010; Mortensen et al., 2015 (France); Mortensen et al., 2015 (UK); Perez et al., 2015; Perkins et al., 2013; Reiter et al., 2010). It is also notable that future partner protection was associated with HPV initiation as well as intention (Perez et al., 2015) and was supported by qualitative findings in one study (Cates et al., 2012: Table 4). Additionally, the intention to vaccinate was associated with the welcoming of any protection against cancer, but again, studies were evenly distributed across categories therefore the strength of this association remains undetermined.

3.1.3. Perceived barriers Perceived barriers were associated with non-acceptability of the

vaccine. Of the 18 studies, 13 studies (11 quantitative and 2 qualitative) reported findings on barriers to vaccine acceptability. One of the most common barriers included fear of side effects and concern about vaccine safety (Bianco et al., 2014; Donahue et al., 2014; Gilkey et al., 2012; Griebeler et al., 2012; Mortensen, 2010; Mortensen et al., 2015 (France);

Mortensen et al., 2015 (Germany); Mortensen et al., 2015 (Italy); Mortensen et al., 2015 (UK); Perkins et al., 2013; Perez et al., 2015; Reiter et al., 2013; Schuler and Coyne-Beasley, 2015: Table 3). Parents believed there was insufficient evidence to prove beneficial effects of vaccine for boys, reporting that the vaccine is still too new (Berenson and Rahman, 2012; Reiter et al., 2013) or doubts about efficacy (Bianco et al., 2014; Donahue et al., 2014; Perez et al., 2015). Although concern about vaccine safety/efficacy and fear of side effects showed a relatively consistent association with vaccine non-acceptability, the strength of the association overall was low, with between 1 and 29% of total participants citing these barriers as reasons for non-acceptance across studies. These findings were, however, confirmed by parents in qualitative studies from Cates et al. (2012) and Alexander et al. (2014: Table 4). In addition to psychological constructs, cost-related barriers including financial hurdles and lack of adequate healthcare insurance showed an overall low association with HPV vaccine non-acceptance (Berenson and Rahman, 2012; Bianco et al., 2014; Perez et al., 2015; Reiter et al., 2010). Financial cost associated with the vaccine was also raised as a perceived barrier among interviewed parents (Cates et al., 2012: Table 4). In one study, significantly fewer parents were willing to vaccinate at the cost of US $400 (Reiter et al., 2010).

3.1.4. Cues to action Nine studies examined factors that might prompt the decision to be

vaccinated. Having the vaccine recommended by health care providers (HCP) was a consistent factor in HPV vaccination acceptance in parents: five studies reported an overall small positive association between HCP recommendation and intention to vaccinate (Bianco et al., 2014; Mortensen, 2010; Mortensen et al., 2015 (France); Perez et al., 2015; Perkins et al., 2013), with findings from four studies also looking at vac- cine initiation (Perez et al., 2015; Perkins et al., 2013; Reiter et al., 2013; Taylor et al., 2014). In addition, having the vaccine endorsed by the Min- istry of Health or a publicly funded national immunisation program showed a small association with intention to vaccinate (Bianco et al., 2014; Mortensen, 2010) (Table 2). Conversely, a lack of recommenda- tion by HCPs and a lack of doctors’ appointments were associated with non-acceptance of the vaccination (Donahue et al., 2014; Gilkey et al., 2012; Mortensen, 2010; Mortensen et al., 2015 (France); Reiter et al., 2013); again showing a small but consistent association (Table 3). The final cue to action was social norms, or the influence of significant others, such as family members or friends, which was associated with positive vaccination intention (Gainforth et al., 2012; Schuler and Coyne-Beasley, 2015). As only a small number of studies reported this finding, the strength of the association remains undetermined (Table 2).

3.2. Other factors

3.2.1. Knowledge and awareness Generally, modest or limited knowledge about HPV infection and

vaccine availability for boys were observed in the majority (83%) of the reviewed studies. Parents often expressed a need for more informa- tion about HPV vaccine before making decision about their son's vacci- nation. Two of the 15 studies that provided a description of parental knowledge also reported how this directly related to vaccine acceptabil- ity, with greater awareness of HPV infection and vaccination being strongly associated with intention to vaccinate (Bianco et al., 2014; Reiter et al., 2010: Table 2). Ten studies reported a low yet consistent as- sociation between lack of awareness and vaccine non-acceptability (Donahue et al., 2014; Gilkey et al., 2012; Mortensen, 2010; Mortensen et al., 2015 (France); Mortensen et al., 2015 (Germany); Mortensen et al., 2015 (Italy); Mortensen et al., 2015 (UK); Perez et al., 2015; Reiter et al., 2013; Tisi et al., 2013: Table 3).

3.2.2. Family characteristics Socio-demographic characteristics such as age, ethnic background

and previous experience with cancer in the family were associated

35G. Radisic et al. / Preventive Medicine 95 (2017) 26–37

with outcomes. Parent’s younger age had a strong positive association with both vaccine intention and initiation in two studies (Perkins et al., 2013; Taylor et al., 2014), as did coming from a non-White ethnic background (Reiter et al., 2010; Taylor et al., 2014: Table 2). Fathers were also reported as less likely to accept the vaccine than mothers (Perkins et al., 2013; Reiter et al., 2013: Table 3). Previous experience with cancer or genital warts in the family had low association with in- tention to vaccine in a small group of parents (Bianco et al., 2014; Mortensen, 2010). In terms of the age of the son, the perception that the son was too young for the vaccine showed a small but consistent as- sociation with vaccine non-acceptability (Berenson and Rahman, 2012; Bianco et al., 2014; Gilkey et al., 2012; Griebeler et al., 2012; Mortensen, 2010; Mortensen et al., 2015 (UK); Reiter et al., 2013: Table 3).

3.2.3. Parent–child dialogue Having frequent and/or open discussions between parents and sons

about HPV infection and vaccination was strongly positively associated with vaccine initiation (Reiter et al., 2013; Taylor et al., 2014: Table 2). Parental respect of son's wishes or right to decide resulted in vaccine non-acceptability in five studies (Bianco et al., 2014; Donahue et al., 2014; Griebeler et al., 2012; Mortensen, 2010; Perkins et al., 2013: Table 3) if the boys rejected the vaccination.

3.2.4. Attitudes towards vaccinations in general General anti-vaccination beliefs showed a consistent, small associa-

tion with HPV vaccine non-acceptability (Bianco et al., 2014; Gainforth et al., 2012; Griebeler et al., 2012; Mortensen, 2010; Mortensen et al., 2015 (France); Mortensen et al., 2015 (Germany); Mortensen et al., 2015 (Italy); Perkins et al., 2013; Perez et al., 2015: Table 3). Interviewed parents who lacked trust in the safety of vaccines were also less likely to initiate the HPV vaccination (Alexander et al., 2014: Table 4). Likewise, intention to vaccinate was associated with general pro-vaccination beliefs, such as welcoming all vaccines for children (Bianco et al., 2014; Mortensen, 2010; Mortensen et al., 2015 (UK): Table 2). Parental intention to vaccinate was also strongly associated with HPV vaccine initiation in sons (Perkins et al., 2013; Reiter et al., 2013).

3.2.5. Miscellaneous beliefs around egalitarianism, child sexual activity, re- ligion and culture

Parents in several European countries (Italy, Denmark and Germa- ny) stated the importance of gender-neutral vaccination and recognised that both genders contribute to the transmission of HPV (Mortensen, 2010; Mortensen et al., 2015 (Germany); Mortensen et al., 2015 (Italy); Tisi et al., 2013). This issue of equal rights and shared responsi- bility for prevention was strongly associated with intention to vaccinate (Table 2). Parents also expressed concerns that the HPV vaccine might encourage or condone earlier onset of sexual activity, promiscuity and/or unprotected sex (Berenson and Rahman, 2012; Bianco et al., 2014; Donahue et al., 2014: Table 3; Hansen et al., 2015: Table 4). Vac- cine non-acceptance was also associated with the importance of reli- gious or cultural beliefs endorsed by some parents (Mortensen et al., 2015 (UK); Reiter et al., 2010: Table 3).

3.2.6. Other potential variables of interest Given the relatively few studies available, single variables with no

comparison studies may still be of interest to future research. Of partic- ular note is that having a daughter in the family as well as a son predict- ed mothers’ intention to vaccinate sons in one study, while mothers of sons only were less willing to vaccinate (Berenson and Rahman, 2012). Similarly, having an adolescent daughter who had already had the HPV vaccination (Reiter et al., 2010) and awareness of the benefits of the vaccination in girls (Bianco et al., 2014) and predicted vaccine ac- ceptability for sons in two other studies. An additional factor of interest is fear of needles, which was associated with parental non-acceptance in one study (Perkins et al., 2013).

4. Discussion

Parents in the selected studies were generally supportive of HPV vaccination for boys. Health behaviour constructs previously used to or- ganise and understand the factors associated with HPV vaccine in girls (cf. Brewer and Fazekas, 2007; Cunningham et al., 2014) were also asso- ciated with HPV vaccine acceptance here. The perceived benefit of HPV vaccination for preventing cancer and other diseases was the strongest, most consistently cited predictor of vaccine acceptability among par- ents in this review. This is in accordance with findings from the other two reviews where perceived benefit (effectiveness) was rated as the most important attribute of the HPV vaccine and was also positively as- sociated with vaccine acceptability (Brewer and Fazekas, 2007; Cunningham et al., 2014). The second most cited benefit strongly asso- ciated with vaccine acceptability – found in this review but not in the previous reviews – was parental perception about the importance of fu- ture partner protection. Parents who perceived the importance of future partner protection were more accepting of HPV vaccination for their sons. This suggests that HPV vaccination programs specifically targeting adolescent boys could emphasise the high prevalence of HPV infection and the benefits of HPV vaccination for them and their future partner’s health.

Other health behavioural constructs, although consistent, showed positive but small associations with vaccine acceptability. Parental ac- ceptance was higher among those who perceived their children to be at a greater risk of HPV infection and HPV related diseases. In contrast, parents who did not believe in the severity of HPV infection in males and the consequent necessity for vaccinating boys were less likely to en- dorse vaccination. This is in line with findings from Brewer and Fazekas (2007) where higher perceived susceptibility to infection was related to higher vaccine acceptance, while Cunningham et al. (2014) reported an unclear relationship. Future interventions should emphasise the high risk of acquisition of HPV among sexually active young males and the severity of diseases that affect them directly.

Perceived barriers to vaccination including doubts around efficacy of the vaccine, cost and fear of safety and side effects were the most cited reasons for vaccine non-acceptability, again in line with the findings from reviews of girls. Parents who did not understand the severity of vaccine-preventable diseases were more concerned about the side ef- fects of the vaccines than the diseases they prevent. This highlights the need for effective communication on risk in order to avoid parental misconceptions about vaccine safety and efficacy. In countries without government funded immunisation programs, the cost and financial hur- dles were a common reason for refusing HPV vaccination. Without uni- versal healthcare coverage, the financial cost of the HPV vaccine may remain an important barrier to provision and uptake in some countries. Also consistent with previous reviews, practitioner recommendation appeared to be an important cue to vaccination acceptability in boys. Parents seem to trust their healthcare provider to recommend vacci- nations; lack of prompts from health authorities are likely to result in vaccination refusal or hesitance. Increased recommendation is particularly important for the countries where HPV vaccination for males is not part of a universal, government funded school immunisation program.

The review also raised the need to increase awareness about the HPV vaccination for males since insufficient knowledge about male HPV-re- lated diseases and the availability of HPV vaccination for males were linked with vaccine non-acceptance. Requests by many parents for more information on HPV vaccination before deciding signifies the need for better educational campaigns. The delivery of relevant, clear in- formation about HPV infection, availability of the vaccine for boys and vaccine provision requirements could reduce misconceptions about the vaccine and assist the delivery, particularly to older parents or those with concerns regarding the age of the child and the potential for the promotion of sexual behaviour. Similarly, the review shows that fathers are less likely than mothers to initiate the vaccination in

36 G. Radisic et al. / Preventive Medicine 95 (2017) 26–37

sons, and therefore may benefit from targeted information provision. Culturally-sensitive promotional materials may also be useful, given the suggestion from a small number of studies that vaccine acceptance may be influenced by religious and cultural beliefs, and the ethnic back- ground of parents. Further research is warranted.

Additional finding that is unique to this review is that having a daughter, particularly if previously vaccinated, is associated with vac- cine acceptance in boys. Although we are unable to estimate the strength of this association, this suggests that parents of only sons may be in need of additional information. Interestingly, strong support was also found for gender-neutral vaccination among parents in European countries. This is consistent with the program goal of preventing the transmission of HPV-related diseases by highlighting the importance of shared responsibility, indicating a further potential target area for HPV awareness promotion in other countries.

4.1. Strengths and limitations of the study

The strengths of this review include comprehensive summary of ev- idence from diverse studies, large parental sample from several coun- tries and inclusion of studies following the licensure of HPV vaccine for males thus eliminating the bias of hypothetical acceptability in the pre recommendation studies. The separation of vaccine intention from vaccine initiation enabled the identification of factors associated with actual uptake. This is important because a positive intention is not al- ways translated into future behaviour. It was found that, to date, per- ceived benefit to future partner; having recommendations from a HCP, and having open and frequent discussions with boys about HPV and the vaccine were among salient factors specifically associated with up- take in boys. Future studies should aim to incorporate longitudinal fol- low-up designs where possible to provide more information regarding vaccine initiation rather than intention or willingness alone.

The current review also offered a comprehensive synthesis of find- ings. The heterogeneity of the included studies and inconsistencies in reporting challenged the quantification of findings: nonetheless it was possible to give a useful indication of the strength of the association be- tween vaccine acceptability and consistently reported variables. The quality of the studies included in the review was variable, with method- ological problems including a lack of reliable measurement. In the rap- idly growing field of HPV vaccination, this review highlights the existing heterogeneity and variation in measures used to asses HPV vac- cine acceptability. Researchers in the HPV vaccination field should be encouraged to utilise standardised, validated and reliable tools that are now available to measure HPV knowledge, attitudes and beliefs (e.g. Waller et al., 2013; Perez et al., 2016) to achieve better compari- sons across studies. In addition, several studies were limited to conve- nience samples in health care settings, restricting the generalizability to other populations. Studies were also characterised by reliance on cross-sectional design and parental self-report. As such, the potential for bias in the studies is high and findings should be interpreted with this in mind.

There are several limitations to this review. Our search returned studies from the developed countries only, hence limiting the generaliz- ability to the rest of the world, in particular to developing countries and more conservative societies. Furthermore, two thirds of the studies were from the USA, which may limit the generalizability of the results in other health care settings. However, with the exception of European parents that stressed the importance of gender equality, shared respon- sibility and equal rights in regards to HPV vaccination, no regional dif- ferences were observed in parental vaccination perceptions among different countries.

Finally, while the HBM was used as a guiding framework to organise the findings in the review, we recognise its limited ability to encompass all variables relevant to HPV vaccination. Thus, other prospective factors were added to augment the model. However, in line with previous re- views in girls (Brewer and Fazekas, 2007; Cunningham et al., 2014),

this review demonstrates the utility of the HBM for integrating qualita- tive and quantitative research findings to better understand the factors associated with HPV vaccination acceptability in the parents of boys.

5. Conclusion

Despite promulgation of recommendations to vaccinate boys for HPV, uptake remains low. Parents are critical to a successful program in adolescents. This review summarises the evidence for factors under- lying parental decision-making, identifying variables that are common to HPV vaccination in girls and unique to boys. It is essential to target those factors proven to promote vaccine acceptability in order to opti- mise vaccine uptake, particularly given the growing number of coun- tries now recommending or adding boys' participation to their HPV vaccination program. However, this number is still not optimal. Ad- dressing HPV infection in male adolescents is pertinent to remove the onus for preventive health behaviour away from one gender. As the in- formation about the role of oncogenic HPV in causing various cancers that affect males continues to strengthen, the prevention implications of the HPV vaccine becomes more obvious. To reduce further incidence of cancer and address underutilisation of HPV vaccine among adoles- cent boys, further research should aim to develop targeted, evidence- based strategies to increase uptake, and better understand factors that could influence decision-making at policy-level as well as the individual level.

Transparency document

The Transparency document associated with this article can be found, in online version.

References

Alexander, A.B., Stupianski, N.W., Herbenick, D., Reece, M., Zimet, G.D., 2014. What par- ents and their adolescent sons suggest for male HPV vaccine messaging. Health Psychol. 33 (5), 448–456.

Allen, J.D., Coronado, G.D., Williams, R.S., Glenn, B., Escoffery, C., Fernandez, M., Tuff, R.A., Wilson, K.M., Mullen, P.D., 2010. A systematic review of measure used in studies of humanpapillomavirus (HPV) vaccine acceptability. Vaccine 28 (24), 4027–4037.

Armitage, C.J., Conner, M., 2000. Social cognition models and health behaviour: a struc- tured review. Psychol. Health 15 (2), 173–189.

Becker, M.H., 1974. The health belief model and personal health behavior. Health Educ. Monogr. 2, 324–508.

Berenson, A.B., Rahman, M., 2012. Gender differences among low income women in their intent to vaccinate their sons and daughters against human papillomavirus infection. J. Pediatr. Adolesc. Gynecol. 25 (3), 218–220.

Bianco, A., Pillegi, C., Lozzo, F., Nobile, C.G.A., Pavia, M., 2014. Vaccination against human papilloma virus infection in male adolescents: knowledge, attitudes, and acceptabili- ty among parents in Italy. Hum. Vaccines Immunother. 10 (9), 2536–2542.

Bosch, F.X., Lorincz, A., Munoz, N., Meijer, C.J.L.M., Shah, K.V., 2002. The causal relation be- tween human papillomavirus and cervical cancer. J. Clin. Pathol. 55, 244–265.

Brewer, N.T., Fazekas, K.I., 2007. Predictors of vaccine acceptability: a theory-informed, systematic review. Prev. Med. 45, 107–114.

Cates, J.R., Ortiz, R., Shafer, A., Ramocki, L.S., Coyne-Beasley, T., 2012. Designing messages to motivate parents to get their preteenage sons vaccinated against human papillo- mavirus. Perspect. Sex. Reprod. Health 44 (1), 39–47.

Chaturvedi, A.K., 2010. Beyond cervical cancer: burden of other HPV-related cancers among men and women. J. Adolesc. Health S20–S26.

Conner, M., Norman, P., 2005. Predicting Health Behaviour. Bell & Bain Lts, Glasgow, UK. Cunningham, M.S., Davison, C., Aronson, K.J., 2014. HPV vaccine acceptability in Africa: a

systematic review. Prev. Med. 69, 274–279. deMartel, C., Ferlay, J., Franceschi, S., Vignat, J., Bray, F., Forman, D., Plummer, M., 2012.

Global burden of cancers attributable to infections in 2008: a review and synthetic analysis. Lancet Oncol. 13 (6), 607–615.

Dempsey, A.F., Butchart, A., Singer, D., Clark, S., Davis, M., 2011. Factors associated with parental intentions for male human papillomavirus vaccination: results of a national survey. Sex. Transm. Dis. 38 (8), 769–776.

Donahue, K.L., Stupiansky, N.W., Alexander, A.B., Zimet, G.D., 2014. Acceptability of the human papillomavirus vaccine and reasons for non-vaccination among parents of ad- olescent sons. Vaccine 32 (31), 3883–3885.

Frazer, I.H., 2010. Measuring serum antibody to human papillomavirus infection or vacci- nation. Gynecol. Oncol. 118 (1 Suppl), S8–11.

Gainforth, H.L., Cao, W., Latimer-Cheung, A.E., 2012. Determinants of human papillomavi- rus (HPV) vaccination intent among three Canadian target groups. J. Cancer Educ. 27 (4), 717–724.

37G. Radisic et al. / Preventive Medicine 95 (2017) 26–37

Gerend, M.A., Shepherd, J.E., 2012. Predicting human papillomavirus vaccine uptake in young adult women: comparing the health belief model and theory of planned be- havior. Ann. Behav. Med. 44 (2), 171–180.

Gilkey, M.B., Moss, J.L., Moss, J.L., McRee, A.L., Brewer, N.T., 2012. Do correlates of HPV vac- cine initiation differ between adolescent boys and girls? Vaccine 30 (41), 5928–5934.

Gillison, M.L., Chaturvedi, A.K., Lowy, D.R., 2008. HPV prophylactic vaccines and the po- tential prevention of noncervical cancers in both men and women. Cancer 113 (S10), 3036–3046.

Giuliano, A.R., Palefsky, J.M., Goldstone, S., et al., 2011. Efficacy of quadrivalent HPV vac- cine against HPV infection and disease in males. N. Engl. J. Med. 364, 401–411.

Giuliano, A.R., Nyitray, A.G., Kreimer, A.R., Pierce Campbell, C.M., Goodman, M.T., Sudenga, S.L., Monsonego, J., Franceschi, S., 2014. EUROGIN 2014 roadmap: differences in human papillomavirus infection natural history, transmission and human papilloma- virus-related cancer incidenc by gender and anatomic site of infection. Int. J. Cancer 136, 2752–2760.

Griebeler, M., Feferman, H., Gupta, V., Patel, D., 2012. Parental beliefs and knowledge about male human papillomavirus vaccination in the US: a survey of a pediatric clinic population. Int. J. Adolesc. Med. Health 24 (4), 315–320.

Hansen, C.E., Credle, M., Shapiro, E.D., Niccolai, L.M., 2015. It all depends: a qualitative study of parents' views of human papillomavirus vaccine for their adolescents at ages 11–12 years. J. Cancer Educ. http://dx.doi.org/10.1007/s13187-014-0788-6.

Holman, D.M., Bernards, V., Roland, K.B., Watson, M., Liddon, N., Stokley, S., 2014. Barriers to human papillomavirus vaccination among US adolescents: a systematic review of the literature. JAMA Pediatr. 168 (10), 76–82.

Kessels, S.J., Marshall, H.S., Watson, M., Braunack-Mayer, A.J., Reuzel, R., Tooher, R.L., 2012. Factors associated with HPV vaccine uptake in teenage girls: a systematic review. Vaccine 30 (24), 3546–3556.

Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G., the PRISMA Group, 2009. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 339, 332–336.

Mortensen, G.L., 2010. Parental attitudes towards vaccinating sons with human papillo- mavirus vaccine. Dan. Med. Bull. 57 (12), A4230.

Mortensen, G.L., Adam, M., Idtaleb, L., 2015. Parental attitudes towards male human pap- illomavirus vaccination: a pan-European cross-sectional survey. BMC Public Health 15, 624.

National HPV vaccination Program Register, 2015. National (Australia) HPV 3 dose vacci- nation coverage for adolescents turning 15 years of age. available from. http://www. hpvregister.org.au/research/coverage-data.

Pace, R., Plye, P., Bartlett, G., Macaulay, A.C., Salsberg, J., et al., 2012. Testing the reliability and efficiency of the pilot mixed methods appraisal tool (MMAT) for systematic mixed studies review. Int. J. Nurs. Stud. 49, 47–53.

Perez, S., Shapiro, G.K., Brown, C.A., Dube, E., Ogilvie, G., Rosberger, Z., 2015. ‘I didn't even know boys could get vaccine’: parent's reasons for human papillomavirus (HPV) vac- cination decision making for their sons. Psycho-Oncology http://dx.doi.org/10.1002/ pon.3984 (Published online in Wiley Online Library (wileyonline library.com)).

Perez, S., Shapiro, G.K., Tatar, O., Joyal-Desmarais, K., Rosberger, Z., 2016. Development and validation of the human papillomavirus attitudes and beliefs scale in a national Canadian sample. Sex. Transm. Dis. 43 (10), 626–632.

Perkins, R., Apte, G., Marquez, C., Porter, C., Belizaire, M., Clark, J.A., Pierre-Joseph, N., 2013. Factors affecting human papillomavirus vaccine use among white, black and latino parents of sons. Vaccine Rep. 32 (1), e38–e44.

Pluye, P., Gagnon, M.P., Griffiths, F., Johnson-Lafleur, J., 2009. A scoring system for apprais- ing mixed methods research, and concomitantly appraising qualitative, quantitative and mixed methods primary studies in mixed studies reviews. Int. J. Nurs. Stud. 46, 529–546.

Plye, P., Hong, Q.N., 2014. Combining the power of stories and the power of numbers: mixed methods research and mixed studies reviews. Annu. Rev. Public Health 35, 29–45.

Reiter, P.L., McRee, A.L., Gotlieb, S.L., Brewer, N.T., 2010. HPV vaccine for adolescent males: acceptability to parents post-vaccine licensure. Vaccine 28 (38), 6292–6297.

Reiter, P.L., McRee, A.L., Kepper, J.K., Gilkey, M.B., Galbraith, K.V., Brewer, N.T., 2013. Lon- gitudinal predictors of human papillomavirus vaccination among a national sample of adolescent males. Am. J. Public Health 103 (8), 1419–1427.

Sallis, J.F., Prochaska, J.J., Taylor, W.C., 2000. A review of correlates of physical activity of children and adolescents. Med. Sci. Sports Exerc. 32 (5), 963–975.

Schuler, C.L., Coyne-Beasley, T., 2015. Has their son been vaccinated? Beliefs about other parents matter for human papillomavirus vaccine. Am. J. Mens Health 10 (4), 318–324.

Siddiqui, M., Asif, A., Perry, C.M., 2006. Human papillomavirus quadrivalent (types 6, 11, 16, 18) recombinant vaccine Gardasil R. Drugs 66 (9), 1263–1271.

Taylor, J.L., Zimet, G.D., Donahue, K.L., Alexander, A.B., Shew, M.L., Stupiansky, N.W., 2014. Vaccinating sons against HPV: results from a U.S. national survey of parents. PLoS One 9 (12), e115154.

Tisi, G., Salinaro, F., Apostoli, P., Bassani, R., Bellicini, A., Groppi, L., Donarini, P., Pecorelli, S., 2013. HPV vaccination acceptability in young boys. Ann. Ist. Super. Sanita 49 (3), 286–291.

Trim, K., Nagji, N., Elit, L., Roy, K., 2012. Parental knowledge, attitudes, and Behaviours to- wards human Pappilomavirus vaccination for their children: a systematic review from 2001–2011. Obstet. Gynecol. Int. 2012, 921236.

Waller, J., McCaffrey, K., Forrest, S., Szarewski, A., Cadman, L., Wardle, J., 2010. Awareness of human papillomavirus among women attending a well woman clinic. Sex. Transm. Infect. 79, 320–322.

Waller, J., Ostini, R., Marlow, L.A., McCaffery, K., Zimet, G., 2013. Validation of a measure of knowledge about human papillomavirus (HPV) using item response theory and clas- sical test theory. Prev. Med. 56 (1), 35–40.

Zhai, L., Tumban, E., 2016. Gradasil 9: a global survey of projected efficacy. Antivir. Res. http://dx.doi.org/10.1016/j.antiviral.2016.03.016 (Epub 2016 Apr 1).