Understanding the determinants of vaccine hesitancy and vaccine confidence among adolescents: a systematic review

ABSTRACT

Vaccine hesitancy (VH) in the age of adolescence is a major public health issue, though it has not been widely examined in the scientific literature. This systematic review aims to address the determinants of VH among adolescents aged 10–19. PubMed, Scopus, and Web of Science were searched from the inception until 11 December 2020. Articles in English, assessing adolescents’ attitudes toward vaccination in terms of hesitancy and/or confidence were considered eligible. Out of 14,704 articles, 20 studies were included in the qualitative analysis. Quality assessment was performed through the Appraisal tool for Cross-Sectional Studies (AXIS). A better knowledge of vaccine-preventable diseases, a higher confidence in vaccines, as well as an active involvement in the decision-making process showed a positive relationship with adolescents’ vaccine uptake. These aspects should be considered to plan tailored interventions for the promotion of vaccination among adolescents and to reduce VH. Major limitations of this review are represented by the high heterogeneity of the tools used in the primary studies and the lack of standardization in outcomes definitions. Future research is needed to disentangle the interrelationship among the different determinants of VH in this age group.

Background

Vaccine hesitancy (VH) encompasses different phenomena, from delay in vaccine uptake to complete refusal. Actually, vaccine-hesitant individuals may refuse or delay some or all vaccines or even accept them albeit being unsure in doing so.^1,2

In fact, the World Health Organization (WHO) defines VH as ‘a delay in acceptance or refusal of vaccines despite the availability of vaccine services.’³ VH is a complex and context-specific issue that varies according to time, place, and vaccine types.⁴ The findings of a systematic review conducted by the dedicated working group of the Strategic Advisory Group of Experts (SAGE) on Immunization on routinely recommended childhood vaccines concluded that many factors are associated with VH and that there is not a unique group of determinants behind VH in all settings. For instance, a higher education showed a direct relationship with VH in China, USA, and Israel and an inverse relationship in Greece and Pakistan, while a lower education was associated with VH in Nigeria and Kyrgyzstan and was inversely related to VH in USA.⁵

According to the “3Cs” model, VH is linked to complacency, convenience, and confidence. Complacency is defined as the perceived risk of contracting the disease; when it is low vaccination can be deemed an unnecessary preventive action. Convenience is defined as the perceived level of access to vaccinations; it depends on physical availability, affordability, geographical accessibility, ability to understand information (language and health literacy), and appeal of immunization services (the quality of the service). Confidence is defined as the trust in: (1) the effectiveness and safety of vaccines; (2) the system that delivers them, including the reliability and competence of health services and health professionals; and (3) the motivations of the policy-makers who decide on the recommended vaccines.

The more complex VH Matrix includes determinants that belong to three categories: contextual factors (influences due to historic, socio-cultural, environmental, health system/institutional, economic, or political factors), individual and group influences (due to personal perception of the vaccine or to social environment/peer), and vaccine/vaccination-specific issues (directly related to vaccine or vaccination).⁴

VH has been paid attention by supranational organizations since the second decade of the 21st century. In 2015, a report was published by the European Center for Disease Prevention and Control (ECDC)^6,7 and important recommendations were issued by the SAGE on Immunization of the WHO.⁸ Although models explaining VH and tools to address it have been published, VH still represents a current issue worldwide.⁹

A study performed on 65,819 individuals from 67 countries showed different vaccine sentiments across world regions. In particular, the European region reported the highest mean-averaged negative responses for vaccine importance, safety, and effectiveness. Negative vaccine-safety perceptions were particularly alarming in the European region with seven out of the ten most negatively reporting countries to vaccine safety located in Europe.¹⁰

Negative sentiments challenge vaccination programs and might prevent reaching the target of vaccination coverage. This justify why VH has been declared by the WHO as one of the top ten global health threats in 2019.¹¹

In order to make it possible to implement effective actions counteracting VH, the understanding of its determinants is of utmost importance. The determinants of VH have been addressed in several population subgroups, mostly parents,¹² and healthcare professionals.¹³ Parents play a relevant role because they are also responsible for decisions on childhood vaccinations. Similarly, healthcare professionals are important for their role in influencing vaccination-related decisions of the general population.

Interestingly, other population subgroups are progressively catching the attention, namely pregnant women¹⁴ or elderly and specific groups of patients, in particular in respect to influenza vaccination.¹⁵ To the best of our knowledge, adolescents represent the less studied target. Actually, there has been an increase in papers on adolescents’vaccinations acceptance⁵ but they mainly focused on vaccination against Human Papilloma Virus (HPV).^16,17

Nevertheless, depending on country-specific immunization programmes, vaccinations intended for adolescents also include those against hepatitis B, tetanus, diphtheria, pertussis, measles, mumps, rubella, varicella, meningococcal diseases and, for risk groups, influenza, and pneumococcal diseases.¹⁸ Several strategies can be implemented to improve vaccination uptake among adolescents, namely, health education programmes, school-based vaccination, financial incentives, or mandatory vaccination. They can target adolescents, parents, and healthcare providers but the body of evidence on their efficacy is of lo- to-moderate certainty and a deeper understanding of factors that influence uptake would be deserved.¹⁹

Adolescent vaccination has become a major health priority. In the light of a life-course immunization, adolescents should be paid attention as target of both primary immunization and boosting shots and catch up programmes.^20,21 Furthermore, 25% of the global population is represented by adolescents, but their vaccine uptake remains low.²²

The objective of the present review is to identify main reasons behind VH and vaccine confidence in the adolescent population, including individual, contextual, and vaccine-/vaccination-related issues. The synthesis of these information will be useful to address VH and, consequently, set up intervention to increase adolescents’ vaccination coverage.

Materials and methods

A systematic review was conducted and reported according to the Preferred Reporting Items for Systematic Reviews (PRISMA).²³

After structuring a search string, three electronic databases were searched to retrieve studies exploring reasons of VH and vaccine confidence in the adolescent population according to pre-defined inclusion and exclusion criteria. Then, data were extracted and synthesized qualitatively and a methodological quality assessment of included articles was performed.

Search strategy

Web of Science, PubMed, and Scopus were queried to retrieve potential eligible articles published from the inception until 11 December 2020.

A search string was created on the basis of the PICO model (P, population/patient; I, intervention/indicator; C, comparator/control; and O, outcome).

The string ((“adolescent”[MeSH Terms] OR adolescent* OR teenager* OR “young adult”[MeSH Terms] OR “students”[MeSH Terms]) AND ((vaccine AND confidence) OR (vaccine AND hesitancy) OR (vaccine attitudes)) was launched on PubMed and then adapted to the other two databases.

The reference lists of included articles were hand searched to look for additional eligible studies.

Inclusion/exclusion criteria

Articles on determinants of adolescents’ attitudes or behaviors in respect to vaccination with any study design (quantitative or qualitative) and published in English were included in the systematic review. For the purpose of the study, we used the WHO 2018 ‘adolescents’ definition, according to which “Adolescence is the phase of life between childhood and adulthood, from ages 10 to 19.”²⁴ Therefore, we considered studies eligible if the mean age of the study population fell between 10 and 19 years.

We excluded systematic reviews, non-empirical studies, conference, abstracts, editorials, commentaries, book reviews, and abstracts not accompanied by a full text. Furthermore, animal and modeling studies were also excluded. Studies that took into consideration parents’ attitudes alongside those of their adolescent children were excluded as well, unless data from the two age groups could be distinguished. In this case, only data from adolescents were extracted and included in our analysis.

Moreover, articles mentioning ‘college/university students’ but not specifying the age of the study population and those considering interventions aimed to increase vaccination confidence among teenagers were also not included.

Study selection

After removing duplicate records, four researchers (C.C., T.E.L., R.M., M.S.) independently screened articles first by title and abstract and then on the basis of the full texts. Disagreements were resolved through discussion among the review team members. The study selection was performed from December 2020 to January 2021.

Data extraction

The four researchers above-mentioned conducted data extraction from the end of January to February 2021. A dedicated data extraction form developed on Excel was used to gather the following information for each eligible study:

1. Study identification (first author, title, journal, and publication year)

2. Study characteristics (period, country, and design)

3. Sample characteristics (sample size, age, gender, and socio-cultural-economic context)

4. Vaccines/vaccinations investigated

5. Study outcome(s)

6. Research tools used (face-to-face/online/self-administered questionnaire, interview, and focus groups)

7. Preparatory materials provided

Data synthesis

According to data extraction, articles explored different types of vaccinations, adopted different tools for investigation (e.g., questionnaires, surveys, interviews etc.) and were performed in various contexts. Because of all these reasons and the heterogeneity of the information collected, data synthesis was conducted only qualitatively.

Quality assessment

The four researchers independently conducted the methodological quality assessment of included articles. Disagreements were resolved by discussion with a fifth researcher (Ch.C.). As the selected studies were all cross-sectional, the Appraisal tool for Cross-Sectional Studies (AXIS) was used to assess the methodological quality.²⁵ This tool, developed by a consensus, offers the possibility to appraise the methodological quality of articles, based on several specific criteria. It consists of 20 items and allows to record ‘yes,’ ‘no,’ or ‘don’t know’ for each criterion and to add short comments.²⁵

To summarize the results of the quality assessment, the articles were grouped into three categories: good (studies satisfying at least 75% of the quality criteria), moderate (studies satisfying from 55% to 74% of the quality criteria), and poor (studies satisfying less than 55% of the quality criteria) quality.

Results

Results of the search strategy

A total of 14,704 articles were retrieved from the three databases. After removing duplicates, the remaining 11,103 records were screened by title and abstract. The full texts of 45 papers were retrieved for the assessment of final eligibility.

Of these, 20 articles^26–45 met eligibility criteria and were included in the systematic review. They were all cross-sectional studies. The selection process is reported in Figure 1.

Figure 1.

Flowchart of the selection process. SLR: systematic literature review.

Results of the quality assessment

As illustrated in Table 1, quality assessment was performed for all the 20 articles evaluated in the analysis.

Table 1.

Quality assessment of the included studies using AXIS quality assessment tool

Study	Question 1 (aims/objectives of the study)	Question 2 (study design)	Question 3 (sample size)	Question 4 (target/reference population)	Question 5 (sample frame)	Question 6 (selection process)	Question 7 (non responders)	Question 8 (risk factor and outcome variables)	Question 9 (instruments/ measurements)	Question 10 (statistical significance)	Question 11 (methods)	Question 12 (basic data)	Question 13 (response rate concerns) *	Question 14 (information about non-responders)	Question 15 (results consistency)	Question 16 (results presentation)	Question 17 (discussions and conclusions)	Question 18 (limitations of the study)	Question 19 (funding sources/conflicts of interest) *	Question 20 (ethical approval)
Kreidl 2020	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	Yes
Marshall 2019	Yes	Yes	No	Yes	Yes	No	No	Yes	Yes	No	No	Yes	No	No	Yes	Yes	Yes	Yes	No	Yes
Hilton 2013	Yes	Yes	No	Yes	Yes	No	No	Yes	No	No	No	Yes	No	No	Yes	Yes	Yes	No	No	Yes
Pennella 2020	Yes	Yes	No	Yes	Yes	No	No	Yes	No	No	Yes	Yes	Don’t Know	No	Yes	Yes	Yes	Yes	No	Yes
Lavelle 2019	Yes	Yes	No	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	Yes
Choi 2013	Yes	Don’t Know	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	Yes
Wang 2016	Yes	Don’t Know	Don’t Know	Yes	Yes	No	No	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	Yes
Griffin 2018	Yes	Yes	No	Yes	No	No	No	Yes	Yes	No	No	Yes	No	No	Yes	Yes	Yes	Yes	No	No
Read 2010	Yes	Yes	No	Yes	Yes	No	No	Yes	Yes	Yes	Yes	Yes	Don’t Know	No	Yes	Yes	Yes	Yes	Don’t Know	Yes
Herman 2019	Yes	Yes	No	Yes	Yes	No	No	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	Yes
Pelullo 2018	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	Don’t Know	Yes	Yes	Yes	Yes	Yes	No	Yes
Khurana 2015	Yes	Yes	No	Yes	No	No	No	Yes	Yes	No	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	No
Caskey 2007	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	No
Costantino 2020	Yes	Yes	Yes	Yes	Don’t know	No	No	Yes	Yes	Yes	Yes	No	No	No	Yes	Yes	Yes	Yes	No	Yes
Pelucchi 2010	Yes	Yes	No	Yes	No	No	No	Don’t Know	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	Yes
Stocker 2013	Yes	Yes	No	Yes	No	No	No	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	No	Yes
Huon 2020	Yes	Yes	No	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	No	No	Yes	Yes	Yes	Yes	No	Yes
Forsner 2015	Yes	Yes	No	Yes	No	No	No	Yes	No	No	No	No	Yes	No	No	Yes	Yes	Yes	No	Yes
Nabirye 2020	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes	No	Yes
Maier 2015	Yes	Yes	No	Yes	Yes	No	No	Yes	No	Yes	Yes	Yes	Don’t Know	No	Yes	Yes	Yes	Yes	No	Yes

Key: 1 = “Yes,” 0 = “No,” 0 = “Don’t know” *Item is reverse scored (i.e., 0 is a positive, counts as a point)

Ten^26–35, out of 20 studies (50%) were evaluated of ‘good quality,’ while nine^36–44 out of 20 (45%) were evaluated of ‘moderate quality.’ Only 1 study⁴⁵ out of 20 (5%) had a poor quality, for this reason it was excluded. As a consequence of this, a total of 19^26–44 studies of ‘moderate’ and ‘good’ quality were considered in the descriptive analysis.

The objective of the study (criterion n. 1), the reference population for the sample (criterion n. 4), results presentation (criterion n. 16), discussion, and conclusions justified by the results (criterion n. 17) were well descripted by all the 19 (100%) studies.

On the contrary, none of the included studies met the criterion n.7, which asks for measures undertaken to address and categorize non-responders.

Furthermore, 18 out of 19 articles^{26–36,38–44} (95%) clearly discussed the limitations of their studies (criterion n. n.18), and 16 articles^{26–35,38,39,41–44} (84%) met the quality standards about methods reproducibility (criterion n.11).

Characteristics of the included studies

The majority of the included studies (53%) were conducted in European countries: one³¹ in Austria, one³⁰ in France, one⁴⁴ in Germany, one⁴⁰ in Ireland, three^28,41,42 in Italy, one³⁹ in Romania, and one³⁷ in Scotland. Seven^{26,29,32,36,38,42,43,} out of 19 (37%) were performed in US countries or counties.

Only one²⁷ study was conducted in Asia, Hong Kong, one³⁵ in Australia, and another one³³ in Africa. Included articles were published from 2010 to 2020.

As far as study vaccines/vaccinations are concerned, studies were grouped into three categories:

1. Focusing on HPV vaccine only (12^{26–28,30,33,38–44} out of 19 articles – 63%);

2. Considering all vaccines in general (five^31,34–37 out of 19–25%);

3. Examining all vaccines in general with a specific focus on HPV vaccine (two^29,32 out of 19–10%).

Half of the included articles^{28,29,31,32,34–37,41,44} (53%) considered samples balanced between females and males. Five studies^{26,27,33,40,43} took into account only female, while two studies^30,38 only males. Only two articles^39,42 did not describe the gender distribution of the study population. The description of the characteristics of the included studies is reported in Table 2.

Table 2.

Description of the characteristics of the studies included in the systematic review

Study	Geographic area	Country	Sample size	Sample gender	Vaccine	Range Age	Recruitment (setting/modalities)	Type of questionnaire/interview	Materials	Social factors (ethnicity/religion/economic-status/education/parents education)
1.Nabirye et al 2020	Africa	Uganda	407	100% Female	HPV	10–15 y	Mbale district in Eastern Uganda (56 villages)	Face to face interview	No materials provided	Race/ethnicity/religion: 75% from rural areas. 71.3% of the Gishu tribe 41.5% Muslim faith.
2. Choi et al 2013	Asia	Hong Kong	2167	100% Female	HPV	11–18 y	13 secondary schools	Self-administered questionnaires	Short description of the HPV vaccine and its market price range	Education: 62.3% junior secondary (S1-S3); 20.6% senior secondary (S4-S5); 17% Sixth Form (S6-S7)
3. Wang et al 2016	Oceania	Australia	502	51% Female 49% Male	All	15–17 y	Through an online panel company, Pureprofile	National online survey	No materials provided	Primarily high socio-economic status; 59% from metropolitan regions.
4. Griffin et al 2018	America	USA	105	62% Female 38% Male	All	14–18 y	Public high school in Long Island, NY	Face to face questionnaire	No materials provided	Not reported
5. Read et al 2010		USA	175	100% Female	HPV	11–14 y	Urban adolescent medicine clinic	Face to face questionnaire	No materials provided	Race/ethnicity: African Caribbean 39.3%; African American 50.6%; Hispanic 10.1%. Education: girls currently in high school (23.9%) or had less than a high school education (76.1%)
6. Herman et al 2019		USA	336	54% Female 46% Male	All+ Focus on HPV	11–13 y	Middle school of a small city in Upstate New York	Face to face questionnaire	No materials provided	Education: 7th and 8th grades students. Race/ethnicity: 71.3% White non-Latin; 7.2% Black non-Latin; 6.6% Latin; 3.9% Asian; 0.8% American Indian; 8.4% Mixed race (including White); 1.8% Mixed race (not White)
7. Khuranaa et al 2015		USA	154	100% Male	HPV	11–21 y	Two private solo practitioner general pediatric practices in Maryland suburbs	Face to face questionnaire	No materials provided	Education: 22.5% 6th grade or lower; 30.5% 7th-8th grade; 27.8% 9th-11th grade; 19.2% 12th grade or higher. Race/ ethnicity: 71.7% Caucasian; 15.1% Asian; 13.2% Other.
8. Caskey et al 2017		USA	412	100% Female	HPV	13–17 y	Research panel of more than 60,000 U.S. households developed and maintained by the survey research firm Knowledge Networks (Menlo Park, CA).	Self-administered questionnaires	No materials provided	Education: 99% less than high school; 1% high school. Race/ethnicity: 61% White, non-Hispanic; 17% Hispanic; 13% Black, non-Hispanic; 9% Other, non-Hispanic.
9. Pennella et al 2020		USA	168	Not reported	HPV	11–12 y	The Shelby County School district in the Memphis, Tennessee-area	Face to face questionnaire. Written responses were transcribed into Microsoft Excel and then subsequently uploaded to MAXQDA, a standard software for coding and analyzing qualitative data	Lesson to introduce HPV (what it is, how it spreads, and how it potentially leads to cancer) and discussion of vaccines, calling attention to the HPV vaccine.	Not reported
10. Lavelle et al 2019		USA	316	54% Female 46% Male	All+ Focus on HPV	13–17 y	Knowledge Panel (GfK Custom Research, LLC, New York, NY), a probability-based online research panel	Face to face interview	No materials provided	Education: less than high school (100%). Race/Ethnicity: White, non-Hispanic 76.4%; Black, non-Hispanic 6.9%; Other, non-Hispanic 8.1%; Hispanic 2.1%
11. Kreidl et al 2020	Europe	Austria	367	41% Female 58% Male 1% Unknown	All	14 y	210 classes with schoolchildren in grades 8 and 9	Self-administered questionnaires	General information about the study	Race/ethnicity: 92.4% born in Austria; 30.8% migratory background
12. Costantino et al 2020		Italy	1702	52% Female 48% Male	HPV	11–14 y	18 first-grade secondary schools (accounting for 2469 students) located in Palermo.	Self-administered questionnaires	Informative note on the objectives and purposes of the study. Brief frontal lesson conducted by medical doctors and researchers; set of slides relating to STD prevention with a particular focus on HPV infection, related diseases, and vaccination.	Race/ethnicity: 98.1% Italian; 64.2% Palermo municipality.
13. Pelucchi et al 2010		Italy	863	75% Female 25% Male	HPV	14–20 y	2 middle schools in Milan (one private and one public), 5 high schools specializing in classical, linguistic or scientific studies (2 private and 3 state run) in the greater Milan area and Varese, and 1 state-run technical school in Milan.	Self-administered questionnaires	No materials provided	Mother’s education: 14.4% middle school diploma; 46% high school diploma; 38.8% degree; 0.8% missing values. Father’s education: 20.3% middle school diploma; 37% high school diploma; 4.,4% degreee; 0,3% missing values.
14. Stocker et al 2013		Germany	442	54% Female 46% Male	HPV	14–19 y	10th grade students from a sample of 14 schools in Berlin.	Self-administered questionnaires	Informative note of the study procedures.	Race/ethnicity:94% born in Germany; 20.4% first generation migrants; 27.2% second generation migrants
15. Huon et al 2020		France	145	100% Male	HPV	13–18 y	Middle and high school students attending schools in the Loire-Atlantique department	Online questionnaire	No materials provided	Education: 34.5% in the middle school; 65.5% in high school. Geographical distribution (parents): 59.8% town; 40.2 rural. Socio-professional category (parents): 0.8% agricultural worker/farmer; 3.9% craftsperson and head of a company; 4.,7% employee; 6.3% intermediate professions; 30.7% white-collar and gray-collar workers; 11% laborer; 0% retirees; 5.5% individuals not in the workforce
16. Maier et al 2015		Romania	524	Not reported	HPV	16–18 y	Two general high schools in Bucharest	Self-administered questionnaires	No materials provided	Education: intention to attend college 99.23%. Economic status: 80.91% families with medium monthly household income.
17. Marshall et al 2019		Ireland	50	100% Female	HPV	14 − 16 y	A list of second-level schools (n = 67) and education centers (n = 12) was compiled and classified according to the Irish Pobal HP Deprivation Indices. The recruitment poster, informed consent form and cover letter detailing the project overview were sent to each school principal.	Focus groups	Suggestions and detailed discussion points proposed by the facilitator.	Not reported
18. Hilton et al 2013		Scotland	59	49% Female 51% Male	All	13–18 y	Through posters, leaflets and advertisements placed in settings including schools, community facilities and sport facilities. Setting: local community facilities.	Focus groups, discussion.	No materials provided	Socially deprived and advantaged areas
19. Pelullo et al 2018		Italy	772	53% Female 47% Male	All	11–19 y	Random sample of 5 public schools in the geographic area of the Campania region, in the South of Italy	Self-administered questionnaire	Oral instructions about filling in the questionnaire.	Race/ethnicity: 95.5% Italian; 4.5% other. Education: 38.4% middle school; 61.6% high school. Father’s education: 1.7% no formal education; 3% elementary; 22.5% middle school; 38.5% high school; 34.3% college degree. Mother’s education: 1.9% no formal education; 3.3% elementary; 21.4% middle school; 37.4% high school; 36% college degree.

Outcome categories

All the included articles focused their attention on determinants that might influence adolescents’ vaccine confidence.

Although the majority of them – 15^{26,27,29–33,36–41,43,44} out of 19 (80%) – took also into consideration the reasons underlying VH.

As shown in Table 3, according to the findings, we chose to group the study outcomes into four different categories:

1. Vaccine-related issues (knowledge about the disease and the vaccine, vaccines’ effectiveness and safety);

2. Vaccination-related issues (source of information, vaccine costs, vaccination setting, injection fear and related pain);

3. Individual factors;

4. Socio-cultural-economic characteristics.

Table 3.

Description of the main findings assessed from the studies included in the systematic review

Study	Geographic area	Country	Vaccine	Outcome category
				1. Vaccine-related issues (knowledge about the disease and the vaccine, vaccines’ effectiveness and safety)	2. Vaccination-related issues (source of information, vaccine costs, vaccination setting, injection fear and related pain)	3.Individual factors	4. Socio-cultural-economic characteristics
Nabirye et al 2020	Africa	Uganda	HPV	- Lack of awareness was the main reason given by 45% of the adolescents for not receiving the vaccine - Factors that positively influenced uptake of the vaccine: receiving an explanation for possible HPV vaccine side effects	- Not aware of the number of doses that they must receive and others were not aware of the schedule or interval of the vaccines. - Reluctance to vaccinate, being afraid of vaccines, and myths about the vaccines. - Factors that positively influenced uptake of the vaccine: having many options from where to get the vaccine.	Not applicable	Not applicable
Choi et al 2013	Asia	Hong Kong	HPV	Not applicable	- Among schoolgirls who had not been vaccinated, 27.1% expressed willingness to be vaccinated at market price, 26.5% not willing to be vaccinated because the market price was too high, 46.4% not willing to be vaccinated for reason other than vaccine cost.	Not applicable	- The high price of HPV vaccines is a major barrier to vaccine acceptability
Wang et al 2016	Oceania	Australia	All	- Among the respondents: 64.1% strongly believed that vaccines were beneficial; 24.7% responded “moderately” beneficial; 6.4% “slightly” beneficial; 4.8% disagreeing or uncertain.	- Adolescents were less likely than adults to seek information about vaccination from media sources, and were more likely to seek information from social networks including families and schools	- Adolescents were more likely to prefer a joint decision with parents about vaccinations or to make a decision on their own.	Not applicable
Griffin et al 2018	America	USA	All	- A little over half 56% of participants reported that they felt sufficiently informed about vaccines and their safety. - The possible side effects of vaccines is a potential driver of vaccine hesitancy among a subset of adolescents.	- Almost half 47% of adolescents reported family physicians or other medical professionals as their primary source of vaccine information, followed by their parents/guardians 38%. - The most trusted source of information was family physician or other medical profession while the least trusted source for information about vaccines was social media.		Not applicable
Read et al 2010		USA	HPV	- Less than half 44.5% of respondents were interested in receiving the HPV vaccine, while the remainder were either uninterested 15.2% or unsure 40.2%. - Not knowing enough about the vaccine because the vaccine is too new was the main reason cited by the sample of adolescent girls 40% for not being interested in receiving the HPV vaccine. - Regarding attitudes, 41.3% of adolescents had a very favorable attitude by responding 9 or 10 toward the utility of vaccines in preventing diseases.	Not applicable	- Among respondents 47.2% believed that adolescents should make vaccination decisions for themselves. – Additionally, 31.4% of adolescent girls were not interested in receiving the HPV vaccine because they believed they were too young to think about or to have sex.	- Knowledge was significantly higher in those who had at least one graduated parent, those who had received information about the vaccines from physicians and those who needed additional information about the vaccines.
Herman et al 2019		USA	All+ Focus on HPV	- Boys were significantly more likely than girls to perceive vaccines to be very safe (48.4% vs 30.2%, p < .01) and very effective (49.7% vs 29.0%, p < .01).	Not applicable	- Approximately 1/3 of adolescents reported having a say in the decision to be vaccinated and 1/4 of students expressed a desire for specific information about vaccines.	- Adolescents in non-urban areas of Upstate New York were generally marginalized in the vaccine decision-making process
Khuranaa et al 2015		USA	HPV	- Vaccine acceptance was low among adolescent males (and their parents), and the majority of participants were uncertain if they wanted the vaccine - Boys who were unsure about wanting the vaccine indicated they would like to get the vaccine if it protected against genital warts (60%) or against cervical cancer in women (65%).	Not applicable	- History of sexual activity were associated with greater odds of acceptance for a vaccine against genital warts and only knowing someone who had received the vaccine was a significant predictor of conditional acceptance for a vaccine protecting females against cervical cancer	- Asian race was associated with lower odds of conditional acceptance compared with Caucasian race (p < .001) - Income >$75,000 was associated with higher odds of conditional acceptance of a vaccine against genital warts compared with income less than or equal to $75,000 (p = .011).
Caskey et al 2017		USA	HPV	Not applicable	- Older respondents were more likely to report cost as a barrier to vaccine receipt, compared to younger respondents. - More than 80% of respondents would be likely to receive the vaccine if recommended by a healthcare provider or parent, compared to 55%if close friends received the vaccine - Of those who reported a healthcare visit during the 6 months prior to the survey, just 1/3 reported that their healthcare provider had ever discussed HPV or the HPV vaccine.	- Among respondents 1/3 reported not being sexually active or concerns about vaccine safety as the most important reasons for foregoing vaccination. - Younger respondents were more likely to report lack of sexual activity as the reason for not receiving the vaccine, compared to older respondents	Not applicable
Pennella et al 2020		USA	HPV	- Among respondents 87% indicated that the HPV vaccination was a prevention measure; 52% spoke specifically of prevention of HPV transmission. - Students inquiries were mostly centered on HPV vaccine composition, administration, duration and how the vaccine interacts with the body.	- Among respondents 87% indicated that there was no concern around the HPV vaccine and that the HPV vaccine is good and helpful.	Not applicable	Not applicable
Lavelle et al 2019		USA	All+ Focus on HPV	- More likely to choose vaccines with 99% effectiveness than vaccines with 20%. - Vaccine effectiveness of 70% or 95% effectiveness did not impact adolescent choice.	- Adolescents are most sensitive to Primary Care Physicians recommendation, and out-of-pocket cost for vaccination in their decision to get vaccinated.	Not applicable	Not applicable
Kreidl et al 2020	Europe	Austria	All	- Lack of awareness: only 1/3 of participants knew about their right to decide to get vaccinated at age 14 and that they could chose to be vaccinated over parental objections. - Those who consider measles and HPV severe diseases had a significantly higher intention to be vaccinated - The level of knowledge about vaccine-preventable diseases is pretty scarce in the region	Not applicable	Not applicable	Not applicable
Costantino et al 2020		Italy	HPV	- Better attitudes and knowledge regarding HPV infection and vaccination are attributable to the formative role of families and schools. - Students that had previously received information about sexual transmitted diseases at home or at school showed statistically significantly higher knowledge scores on HPV than others.	- Parental awareness represented one of the main factors influencing the acceptance of the HPV vaccine among preadolescents, and was associated with the quantity and quality of information provided on HPV at school	Not applicable	Not applicable
Pelucchi et al 2010		Italy	HPV	- Both students and parents seem to underestimate the likelihood of HPV infection, and this is associated with a lower propensity for vaccination.	Not applicable	- Females more than males were aware that HPV infection could concern themselves and would undergo vaccination against HPV. - The students who were aware that HPV infection could affect themselves were more in favor of HPV vaccination, regardless of whether they were male or female.	Not applicable
Stocker et al 2013		Germany	HPV	- Reasons for being HPV-unvaccinated reported: concerns about side-effects 30.8% -concerns specifically about HPV vaccine safety -female students appeared to be more knowledgeable in terms of transmission of HPV than male students	-Among respondents 2/3 reported that the vaccination is too time consuming	- Reasons for being HPV-unvaccinated reported: Dissuasion from parents (40.2%), dissuasion from their physician (18.5%)	Not applicable
Huon et al 2020		France	HPV	- Lack of information among boys and their parents about HPV and its vaccination. - The probability of being amenable to being vaccinated was significantly higher when the children were worried about the consequences of being infected, and perceived that there was a risk of HPV infection, that the vaccine could be effective and protect against serious disease.	Children generally go along with their parent’s choice.	Not applicable	- The were not significant associations between the age of the children (p = .412), attending middle or high school (p = .176), being at a private or public facility, the socio-professional category of the parents (p = .105), living in an urban versus a rural area (p = .298), and the presence of a general practitioner in the town where they resided (p = .184), and the children’s favorable attitude toward HPV vaccination.
Maier et al 2015		Romania	HPV	- A very small proportion had heard of HPV infection, HPV vaccine and Papanicolau smear test, that is, 20.22%, 67.92% and 22.9%, respectively. - The most common reason for not receiving the HPV vaccine was the lack of information 80.6% followed by parents’ concerns regarding safety 11%, fear of pain 5.59% and not being sexually active 2.7%. - Among participants, 97.7% of the respondents declared interest in receiving more information about HPV.	Not applicable	Not applicable	- Monthly household income over 1000 euros and self-perceived good relationship with family members were statistically associated on a multivariate logistic regression analysis with a high HPV knowledge score and rate of vaccination.
Marshall et al 2019		Ireland	HPV	- Adolescents demonstrated poor knowledge and understanding of HPV, HPV-related diseases and/or the HPV vaccine. - Vaccination was viewed as an accepted norm, and participants recognized the importance and benefits of immunization.	- Adolescents discussed their fear of needles, fear of cancer, fear of vaccine side-effects and fear of the unknown; their fears did not discourage vaccination. - Expressed the view that the vaccine would not be offered free of charge if it was unsafe or ineffective.	- Participants relied on their parent/guardian to provide information and make healthcare decisions on their behalf.	Not applicable
Hilton et al 2013		Scotland	All	- Awareness of HPV infection (Girls were most aware of HPV infection but it was nonetheless subject to misunderstandings; boys commonly believed it to only affect girls, and both sexes demonstrated confusion about its relationship with cervical cancer) - Enthusiastic at the prospect of more vaccines becoming available for their age group	- Combined vaccines are preferred since they reduce the overall number of injections needed - Immunization in school were not always positive	Not applicable	Not applicable
Pelullo et al 2018		Italy	All	- Attitudes: 41.3% of adolescents had a very favorable attitude assessing the utility of vaccines in preventing diseases	Not applicable	- Regarding vaccination decisions, 47.2% believed that adolescents should make vaccination decisions for themselves	- Knowledge was significantly higher in those who had at least one graduated parent

The results were grouped into four different outcome categories: 1. Vaccine (knowledge of technology/disease, effectiveness, safety, side effects); 2. Vaccination (source of information, costs, setting, fears, pain); 3. Individual factors; 4. Socio-cultural-economic implications.

Vaccine-related issues (knowledge about the disease and the vaccine, vaccines’ effectiveness and safety)

Knowledge

Knowledge of vaccine-preventable diseases and associated risks, together with the awareness of available vaccines and the availability of correct information about them, are described as some of the major factors positively influencing adolescents’ vaccine acceptance.^33,36,42 On the contrary, lack of awareness and information about recommended vaccines was one of the main reasons given by adolescents for not receiving them.^30,31,33,37 Likewise, some of the included studies^37,38,42 confirmed that adolescents’ confidence on vaccines is higher if the disease is considered ‘severe,’ such as measles or HPV-related cervical cancer.

Concerning HPV, girls revealed to have a greater knowledge of HPV infection than boys and they also showed more favorable attitudes about HPV vaccination.³⁷

In accordance with one study,³⁴ girls compared to boys were 20% more likely of having a satisfactory knowledge about vaccine-preventable diseases (OR = 1.2; 95% CI 0.87–1.65). Boys were 20% (OR = 0.8; 95% CI 0.58–1.11) and 29% (OR = 0.71; 95% CI 0.48–1.05) less likely to show positive attitudes regarding the usefulness of vaccines to prevent diseases and to consider useful the information received about vaccinations with respect to girls. Some authors³⁹ reported that a significant higher percentage of girls than boys (45.0% vs 26.3%) was concerned about HPV infection and would have undergo HPV vaccination (68.3% vs 39.7%). According to another study,⁴⁴ a significant difference was observed between boys and either HPV-vaccinated or unvaccinated girls (13.6% vs 31.3% or 23.1%) in respect to the agreement with the statement ‘HPV infection occurs frequently.’ A significant difference was also observed between boys and HPV-vaccinated girls (43.9% vs 64.6%) in regard to the agreement on the statement reporting that HPV infection can cause premalignant lesions and carcinosis of cervix and penis.

Effectiveness/safety

Effectiveness and safety also represent two significant factors that positively affect the acceptability of vaccination among adolescents.^{29,30,32,36,44} In particular, one study³² reported that adolescents were more likely to choose vaccines with 99% effectiveness than those with 20%.

Boys were significantly more likely than girls to perceive vaccines to be effective (49.7% vs 29.0) and believe they are ‘very safe’ (48.4% vs 30.2%).²⁹ Nevertheless, another study did not find differences between male and females in respect to preventive potential of vaccinations toward infectious diseases.

Concerns about the potential side effects deriving from the vaccine and the lack of detailed information about them are described as two of the main reasons underlying VH in the adolescent population.^33,36,44

According to some authors,⁴⁴ a statistically significant difference was observed between HPV-unvaccinated and vaccinated girls when considering the agreement about the severity of vaccination side effects (23.1% vs 10.4%) and the risk of weakening of the immune system through vaccination as reasons avoiding vaccinations (9.2% vs 1.0%).

Vaccination-related issues (source of information, vaccine costs, vaccination setting, injection fear, and related pain)

Source of information

According to our results, the family and school environment (e.g., participating in school educative seminars on HPV, formative intervention scheduled during school hours, school-based vaccination educational programs)^28,35 seem to have a positive impact on adolescents’ attitude toward vaccinations. Indeed, some authors²⁸ described that a school-based educational intervention was strongly associated with an increased HPV vaccination confidence, knowledge, and uptake, with young girls significantly more likely than boys peers in the willingness to receive vaccination after the educational intervention. According to our findings, the main sources of information on vaccinations and the most trusted ones were family, schools, family doctors, and other medical professionals^{28,30,32,35,36.} On the contrary, the least reliable source of information were social media, even though they represent the most easily accessible.^{28,30,32,35,36}

Setting and costs

In some cases, having different options to access the vaccine (e.g., school, free clinics, etc.) is a factor that positively affect confidence in vaccination programmes and also leads to an increased uptake.³³

Some of the included studies,^26,27,32,40 conducted in different countries and contexts, reported vaccine cost as a potential barrier to vaccine uptake: a high cost might be a driver of VH among adolescents if compared to a low-cost vaccine or to a vaccine free of charge.

Injection/fears/pain

One study⁴⁰ of the included studies (5%) reported that the fear of the needle and the pain due to the injection are potential discouraging factors, while combined vaccines are significantly preferred since they reduce the overall number of injections.³⁷

Only one study³⁵ found a slightly higher number of females than males expressing fear of pain at the injection site.

Individual factors

Adolescents seem generally to prefer taking an active part in the decision-making process, since they feel able to deal with it by themselves^35. However, they also appreciate the support and the guidance of professionals, like medical doctors, or of their parents.³⁵

Data on HPV vaccination showed that vaccine confidence is greater in individuals with a previous history of sexual intercourses^26,38,43 and in those who know someone already vaccinated against HPV, such as school friends.^38.

Socio-cultural-economic characteristics

Adolescents’ attitudes toward vaccinations were significantly more favorable in those who had at least one graduated parent,^34,43 while there were not significant associations with the parents’ occupation.³⁰ Attending middle or high school or being at a private or public facility were not significantly associated with adolescents’ attitudes toward HPV vaccinations.³⁰ One study,²⁹ conducted in several schools of different areas in New York, showed that adolescents living in the suburbs are less involved in the vaccine decision-making process than those who live in urban areas. Monthly household income higher than 1,000 Euros³⁹ and annual income higher than 75,000³⁸ seemed to be associated with a better vaccination confidence and uptake.

Discussion

Key results of the systematic review

Our review addressed determinants of VH and vaccine confidence in the adolescent population being the first one, to the best of our knowledge, to provide a broad overview of the topic. In fact, our review focused on all types of vaccinations recommended to adolescents and looked at the total range of reasons behind VH in this age. Most of the articles included in the review actually dealt with HPV vaccination but 35% also addressed other vaccines. Interestingly, included papers were mostly published in the last three years, except for Wang et al. 2016³⁵ and Hilton et al. 2013,³⁷ and this showcases the increasing interest in the topic as a whole. In respect to determinants of VH, the findings showed that, as expected, confidence in vaccines effectiveness and safety plays a relevant role also among adolescents.^26,32,44 In addition, the lack of awareness and information on vaccine-preventable diseases and vaccines, often reported among adolescents,^30,36,39 negatively impact on vaccination attitudes.^{30,33,39,41,43} Other important determinants of vaccine acceptance, beside some individual and socio-economic factors, pertained to vaccination-related issues and included ease of access³³ and cost of vaccination.^27,32

Current debate on the topic

VH and its determinants in the adolescent population have been paid less attention as compared to other age groups. Published systematic reviews on adolescent population have mainly focused on HPV vaccination addressing specific groups, such as boys,⁴⁶ or determinants, such as knowledge and preferences.^47,48 On the contrary, the paper by Karafillakis et al.¹⁷ took a broader perspective addressing determinants of VH toward HPV vaccination in the whole European population, including parents and healthcare workers. They showed a higher percentage of people reporting doubts on safety and effectiveness of HPV vaccines among hesitant people. With this respect, the importance of confidence on vaccines was also emphasized by our findings, not only for HPV but also for other vaccinations. Undoubtedly, the lack of confidence is recognized as one of the most vital determinants of VH. This is also demonstrated by the Vaccine Confidence Project⁴⁹ effort in measuring the public confidence in immunization programmes in people aged 18 and older. Another important determinant is information issues. In fact, Karafillakis et al.¹⁷ identified insufficient knowledge or information as an underlying reason of VH in almost all articles they considered. Our review showed that information issues represent a challenge also in regard to vaccinations other than HPV. Luckily, our findings also attested that adolescents trust more family physicians or other medical professionals and parents than social media.^26,32,36,40 This aspect is closely linked to vaccination confidence as it is described as the “belief that vaccination – and by extension the providers and range of private sector and political entities behind it – serves the best health interests of the public.”⁵⁰ Adolescents’ trust in health professionals is a relevant piece of information, which adds up to similar results issued by the survey on Europeans’ attitudes toward vaccination of the European Commission.⁵¹ This information is of great potential interest in order to set up intervention to improve adolescents’ knowledge and awareness on vaccine-preventable diseases.⁵² In addition, the school context has a fundamental role in this process. In fact, previous research provides evidence that educating students through school-based educational programs, represents one of the best practices to promote vaccine awareness among adolescents.^28,53 In respect to vaccination-specific issues, our review pinpointed ease of access as a potential determinant, adding relevant information to the identification of access barriers in the adolescent population. The review authored by Thomson et al.⁵⁴ summarized the evidence on the reasons behind low vaccination uptake addressing five domains, namely access, affordability, awareness, acceptance, and activation. The authors concluded that access and affordability of vaccination can represent important barriers in many contexts, despite interventions aimed at increasing vaccine awareness and acceptance. This should be kept in mind to improve vaccine uptake, as the Authors showed that arranging the vaccination in universities and schools could be helpful. As already stated in the introduction section, other factors beyond vaccine-/vaccination-specific issues play a role in VH, namely, contextual and individual/group factors.⁴ Actually, also the findings of our review highlighted that both these factors can play a role. As far as contextual factors are concerned, the review showed a potential indirect relationship between VH and socio-economic conditions in terms of higher parents’ education and household income.^38,39 Nevertheless, these finding asks for confirmation, because results on the role of socio-economic status seem currently not conclusive.⁵ Eventually, individual/groups factors, such as knowing someone who got the vaccine and personal sexual behaviors,³⁸ as well as fear of the injection play a role.⁴⁹

Larson et al.⁵ pinpointed to the importance of parents’ awareness and dissuasion in vaccine uptake among adolescents. In this regard, a last relevant aspect needing further attention is that adolescents call for either an involvement in the decision to be vaccinated³⁵ or for a complete autonomy in decision.^34,35,43 These findings implicate some very sensitive and still debated ethical issues, including topics regarding minors’ medical decisions such as the right time for minors to make their own health decisions. Generally speaking, children are considered non-autonomous until they reach 18 years of age. Nevertheless, denying vaccination to adolescents because of parents’ choice could lead to important health threats at individual and society level. In light of this and considering that in some countries adolescents are already allowed to make own decision on sensitive or stigmatized healthcare services, some Authors call for laws enacting minors to consent to vaccination if at least 12 or 14 years old.^55,56 Another point is the involvement of adolescents in the vaccine-related decision that is generally ensured, albeit in different extent,⁵⁷ and whose final impact on vaccine uptake would deserve more investigations.

Limitations and strengths

This review has some limitations that should be considered when interpreting results:

1. Protocol of this systematic review was not registered;

2. Only articles published in English were included in the analysis, and this might have led to and underrepresentation of findings from certain countries. In fact, the great majority of studies included were from Europe and USA, whereas data available from other regions were dearth;

3. A potential bias in the selection of studies cannot be completely ruled out, even though selection was performed independently by four researchers working in couples supervised by a fifth senior researcher;

4. The heterogeneity of the tools used in the studies, namely, questionnaires/interviews/national surveys/focus-groups and the lack of homogeneity in outcomes definition prevented us making a meta-analysis and issuing more conclusive findings. Nevertheless, it should be said that the whole literature on VH and its determinants is still undermined by the lack of standardization of definitions (i.e., confidence, acceptance and uptake are generally used interchangeably), data collection, and analysis. Nonetheless, to the best of our knowledge, this is the first systematic review giving an overview of determinants of VH and vaccine confidence in the adolescent population. Furthermore, as a strength, most of the studies were judged of moderate-good quality.

Final remarks and suggestions for future research

In the context of VH, adolescents still represent a poorly investigated population. Nonetheless, they are an important target of both primary immunization with new vaccines and boosting shots and catch-up programmes. Adolescence is also a phase of life in which boys and girls begin to make significant choices in respect to their health and develop attitudes and behaviors that remain in adulthood. Improving adolescents’ vaccination uptake can provide substantial health benefits to individuals, society, and future generations. Indeed, further studies should better disentangle the role of VH determinants and their interrelationship and investigate how VH and its determinants influence the final vaccination uptake considering also other issues, such as access and affordability. Finally, new evidence would be worthwhile to address context-specific determinants of VH in the adolescent population.

Conclusion

This systematic review provides an overview of the total range of determinants of VH in the adolescent population advancing knowledge to what is already known for HPV. The knowledge of vaccine-preventable diseases and the confidence in vaccines, as well as an active involvement in the decision-making process were shown significant factors that positively influenced adolescents’ attitudes. These aspects should be taken into consideration to plan tailored interventions to reduce VH and then strengthen vaccine uptake among adolescents. Future research should be conducted to disentangle the interrelationship between different determinants of VH, also considering the role of contextual aspects and other challenges for vaccine uptake. This will be fundamental to develop context-specific interventions.

Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Declarations of interest: none

Disclosure of potential conflicts of interest

All the Authors declare that they have no conflict of interest.