Parental willingness to have children vaccinated against COVID-19 in Geneva, Switzerland: a cross-sectional population-based study

DOI: https://doi.org/10.57187/smw.2023.40049

Introduction

Vaccination against COVID-19 has proven efficacy to reduce transmission, limit disease severity, and prevent hospitalisations and post-COVID syndrome, including in children as young as five years [1–4]. Mass vaccination in Switzerland began on 23 December 2020, with priority given to individuals aged 65 years and older, those with chronic diseases deemed ‘particularly vulnerable to COVID-19’ and health workers. Subsequently, vaccination eligibility progressively extended to include all persons aged ≥16 years (May 2021), children aged ≥12 years (June 2021) and children aged ≥5 years (December 2021) [5]. Since vaccines became available, public and scientific debate has been ongoing on the need for and relevance of vaccinating children, whereas the vaccination of adolescents is strongly recommended and generally accepted [6–8]. Although the disease burden (infection severity and short-term effects) in children and adolescents has been relatively low, when infected with SARS-CoV-2, they can transmit the virus to others [9]. Children and adolescents with COVID-19 have a low risk of multisystemic inflammatory syndrome [10] and a low to moderate risk of long COVID [11–13], but they may face educational disruptions, social isolation and poor wellbeing [14–16]. Available vaccination data highlight the safety of the COVID-19 vaccination among children and adolescents [3, 17, 18]. In Switzerland, COVID-19 vaccination is recommended for children and adolescents to protect them from severe forms of COVID-19 and post-COVID syndrome [2, 19]. It is also highly recommended for children and adolescents with a chronic illness and those who are in close contact with others at high risk, such as people with a weakened immune system [20]. In addition, from a public health perspective, vaccination among children and adolescents will de facto increase general vaccination coverage and potentially reduce viral circulation in the population.

Worldwide, parental willingness to vaccinate their children for common childhood diseases varies greatly across populations, often depending on psychosocial, sociocultural, socioeconomic and political factors [21–23]. Understanding the factors associated with parental willingness to vaccinate children is important for tailoring vaccination campaigns with the objective of an overall vaccination increase. Among adults, COVID-19 vaccine acceptance has evolved since the beginning of the pandemic as a function of its evolution and the establishment of containment measures [24–26].

As the pandemic continues, with new variants driving new waves of transmission, and the immune landscape in the population shows decreasing immune protection against emerging variants, vaccination will continue to play a key role in minimising infections, protecting health care workers and vulnerable patients and preventing overburdening health care systems. Monitoring the level of vaccination acceptance in the population, including among children and adolescents, is essential for this. This study aimed to examine parental willingness to have children vaccinated against COVID-19 and explored potentially related sociodemographic factors.

Methods

Study design, setting and sample

This study was embedded in the longitudinal digital cohort study Specchio-COVID19. The cohort was launched in December 2020 to follow the participants of COVID-19 serosurveys conducted in the canton of Geneva, Switzerland. Its main objective is to monitor COVID-19-related symptoms, SARS-CoV-2 seroconversion, and the overall impact of the pandemic on several dimensions of health and socioeconomic factors for at least 2 years [27]. The participants were randomly selected from the general population at three time points: for the first serosurvey, April–June 2020, participants aged 20–75 years were enrolled from a previous general health survey (Bus Santé) representative of the population of Geneva [25]; for the second and third serosurveys (November–December 2020 and June–July 2021), participants were recruited from random samples provided by the Federal Statistics Office of individuals living in Geneva aged 0–64 years [29, 30].

Following participation in one of these serosurveys, participants were invited to join the Specchio-COVID19 cohort study, consisting of periodic online questionnaires and potential serological follow-up [31]. Upon registration, an initial questionnaire collected sociodemographic and lifestyle characteristics and general health-related information. Self-reported SARS-CoV-2 infections and vaccination status were updated through monthly questionnaires. The vaccination questionnaire designed for this study was sent out to participants on 15 February 2022, five weeks after vaccination became available for children aged ≥5 years. Responses up to 13 March 2022 were included in the analysis. Data on the age, sex and education distribution in the overall population of the canton of Geneva were obtained from the Cantonal Office of Statistics of Geneva [32].

Data collected in the COVID-19 vaccination questionnaire

For all participants, the questionnaire collected information regarding COVID-19 vaccination status, acceptance and perceived utility of vaccination in children. For those with children aged 5–17 years, supplementary questions collected information about their willingness to have their children vaccinated against COVID-19 and their reasons for or against this.

The demographic and socioeconomic variables were categorised as follows: education as primary (compulsory education or no formal education), secondary (apprenticeships, secondary school and specialised schools) and tertiary education (university degree, higher professional education and doctorates); income, according to household composition-specific categories from cantonal household income statistics [33], as low (below the first quartile of the general population), middle (between the first and third quartiles) and high (above the third quartile), considering both self-reported household income and household composition (living alone with or without children, in a relationship with or without children, or in a shared apartment with other adults) from the baseline questionnaire; birth country as Switzerland, other high-income country, and low- or middle-income country; employee status as employee, independent, unemployed, retired and other economically inactive; having a chronic disease as yes/no; household composition as alone, single with children, with partner and children, and living with others but not children; smoking status was current, former and never. We categorised having children into four age-based groups: only 5–11 years, only 12–15 years, only 16–17 years, and mixed ages.

As an outcome, vaccination status was defined as yes/no according to the question “Have you received at least one dose of the COVID-19 vaccine?” Intention to vaccinate children was defined as “intend to or already vaccinated”, “do not know” and “not intending to”.

Statistical analysis

In descriptive statistics, we reported the number and percentage for categorical variables and the median and interquartile range (IQR) for continuous variables. We used logistic regression to assess the association of demographic, socioeconomic and health-related factors with COVID-19 vaccination status. Logistic regression models were run for all the following variables, adjusting for sex and age: education, household income, employment status, household composition, having a chronic disease and smoking status. For each variable, we also ran multivariable logistic regressions adjusting for age, sex, education (n = 1 missing) and income (n = 3 missing; n = 790 unknown/unwilling to respond and excluded from analyses). A multinomial regression model was used to assess the relationship between the predictor variables (age, sex, education, income, child’s age group) and the willingness of parents to have their children vaccinated. All regression models were complete case analyses. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated through the exponentiation of estimated coefficients. Statistical significance was taken at the level of p<0.05 a priori. All analyses were conducted using R 4.1.1 (R Foundation for Statistical Computing, Vienna, Austria), distributed under the GNU General Public License. The code used in the regression analyses is available at https://github.com/UEP-HUG/2022-vacc-opinion-public.

Ethical considerations

All participants of the Specchio-COVID19 longitudinal study provided informed and written consent upon enrolment in the study. Ethical approval for the study was obtained from the Cantonal Research Ethics Commission of Geneva, Switzerland (project number 2020-00881). The protocol of the overarching study (Specchio-COVID19) has been published elsewhere [31].

Results

Sample characteristics

Of the 6,816 individuals invited, 4,275 chose to participate (62.7% overall response rate; see supplementary Table 1), of whom 1,383 participants (1/3 of the total sample) were parents of children aged 5–17 years and constituted our analytical sample. Among these, 69.3% were aged 35–49 years, and 56.8% were women. Most participants were professionally active (88.2%), had completed tertiary education (73.6%) and reported a middle household income (51.3%; table 1). Among participants with children aged 16–17 years, 87.0% reported that their children were already vaccinated. This decreased to 65.5% among those with children aged 12–15 years and 8.7% among participants with children aged 5–11 years. For a description of the overall sample and a comparison to the general population of Geneva, see supplementary tables 2–4. For an overview of sociodemographic factors associated with vaccination status in the overall sample, see supplementary tables 5–6.

Willingness of parents to have children vaccinated against COVID-19

Parental intention to vaccinate their children differed considerably according to their children’s age group (with a marked decrease in younger age groups) and their own vaccination status (figure 1, supplementary table 7). Overall, 13.4% of parents with adolescents aged 16–17 years indicated no intention to vaccinate them, 2.6% indicated being undecided and 84.0% indicated their children had already received the vaccine. Among parents of children aged 12–15 years, 31.0% indicated having no intention to vaccinate their children, 7.0% indicated being undecided and 60.9% indicated having already vaccinated their children. Among parents with children aged 5–11 years, 59.6% indicated having no intention to vaccinate them, 20.9% indicated not knowing yet and 21.2% indicated having already vaccinated their children or intending to (supplementary table 7). For all child age groups, unvaccinated parents indicated not intending to vaccinate their children more frequently than vaccinated parents did (figure 1).

Figure 1 Parental intention to vaccinate children by age group and own vaccination status. 

Parental vaccination status is positively associated with their intention to have their children vaccinated, across all age groups, though this association was weakest for the 5–11 years age group. “Vacc.” means parents having received at least one vaccine dose (those with at least one child aged 5–11 n = 705; 12–15 n = 548; 16–17 n = 339), and “Non-vacc.” refers to non-vaccinated parents (5–11 n = 126; 12–15 n = 97; 16–17 n = 48). If a parent has children in more than one age group, they are counted twice in the figure.

Sociodemographic factors associated with parental intention to vaccinate children against COVID-19

In comparison with those willing to have their children vaccinated, regardless of age, parents with secondary or primary education, and those with middle or low household income levels, had greater odds of refusing COVID-19 vaccination for their children or deciding differently depending on the child’s age (table 2). No association existed with the parent’s sex, whereas being a younger parent was associated with willingness to have children vaccinated, whatever their age (table 2).

Compared with parents who had children only in the 16–17 years age group, parents with children only in the 12–15 years age group had three times greater odds of refusing to have their children vaccinated (3.08; 1.61–5.91); parents with children only in the 5–11 years age group had almost 20 times greater odds of refusing to have their children vaccinated (19.8; 10.3–38.1).

Reasons for parental intention for or against vaccinating children

Among parents in favour of vaccinating their children, the most frequently chosen reason was “wish my child to be able to lead a normal life (go to school, do activities, travel, etc.)”, reported by more than three quarters of parents, regardless of their child’s age group (figure 2). The other most reported reasons varied in frequency according to the child’s age group. For instance, the intention to vaccinate children to protect them against long COVID symptoms was reported by 51.5% of parents with children aged 16–17 years, 55.8% of parents with children aged 12–15 years and 74.7% of parents with children aged 5–11 years. Similarly, protection against SARS-CoV-2 infection was reported as a reason by 49.4%, 56.1% and 71.8% of parents with children aged 16–17 years, 12–15 years and 5–11 years, respectively. With a similar frequency according to the child’s age group, over half of the parents also reported “to get out of the public health crisis more quickly” and “wish to protect those close to my child (grandparents or other family members)” as reasons for vaccinating their children (figure 2).

Figure 2 Reasons for deciding or wanting to vaccinate children.

Parents of children in the given age groups were asked their reasons for deciding or intending to vaccinate their children. Multiple choices were possible. Parents of children aged 5–11 years were not asked about the need for a COVID certificate because this age group did not require one for any activities in Switzerland.

Among parents not intending to vaccinate their children, “my child has already had COVID-19” was reported by 74.2%, 55.3% and 66.0% of those with children aged 16–17 years, 12–15 years and 5–11 years, respectively (figure 3). A “worry about lack of long-term knowledge regarding COVID-19 vaccines” was reported by 59.7%, 60.2% and 44.3% of parents with children aged 16–17 years, 12–15 years and 5–11 years, respectively. A “worry about COVID-19 vaccines’ side effects” was reported by 54.8%, 44.7% and 36.0% of parents with children aged 16–17 years, 12–15 years and 5–11 years, respectively. “My child does not want to get vaccinated” was reported by 41.9%, 35.2% and 16.2% of parents with children aged 16–17 years, 12–15 years and 5–11 years, respectively. Very few participants reported being against vaccines in general (figure 3).

Figure 3 Reasons for deciding not to vaccinate children.

Parents of children in the given age groups were asked their reasons for not wanting to vaccinate or not being sure about vaccinating their children. Multiple choices were possible.

Discussion

To our knowledge, this is the first population-based study in Switzerland to assess parental willingness to have their children vaccinated against COVID-19 since vaccination was officially approved for children and adolescents. Our study began 5 to 8 weeks after vaccination became available to all children aged ≥5 years. We found that parental willingness to have their children vaccinated differed considerably according to their children’s age and their own vaccination status.

A recent meta-analysis of 44 studies worldwide, most of which were conducted before vaccination was approved for children and adolescents, found that parental intention to vaccinate their children was 60.1% overall, although it ranged from 25.6% in Saudi Arabia to 92.2% in Zambia [8]. This meta-analysis also found that, overall, 22.9% of participants intended not to vaccinate their children, and 25.8% were unsure [8].

In our previous cross-sectional vaccination survey, conducted before vaccination was available for children, we showed that less than half of parents (45.6%) intended to have their children vaccinated in the canton of Geneva. The gradient in intention increased with children’s age from 38.6% for children aged 6–10 years to 55.9% for children aged 16–18 years [24]. In a sample of Swiss German-speaking parents, 58.7% reported their intention to have their children vaccinated. However, this result might be overestimated because parents were recruited at a vaccination centre [34].

These new findings reflect those of international studies conducted since vaccination was approved for children aged 12 years and older. A Canadian study found that around 90% of parents had already vaccinated or were very likely to vaccinate their children aged 12–17 years, whereas more than half were willing to vaccinate their younger children (0–11 years) [35]. In the US, a study found that around two thirds of parents had already vaccinated or intended to vaccinate their children aged 12–17 years, and 60% intended to vaccinate their children aged 5–11 years [36]. In both studies, similarly to our findings, parents were significantly more likely to vaccinate their children if they were themselves vaccinated [35, 36].

Our findings are also in line with those of previous studies showing that parental willingness to vaccinate their children decreased progressively with the age of the child, from 84.0% for parents of children aged 16–17 years to 19.5% for parents of children aged 5–11 years [8, 34–36]. However, in the studies from the US and Canada, these decreases were not as pronounced as in our findings [35, 36].

The COVID certificate was established for adolescents aged 16 years or above in Switzerland, and for those aged 12 years or above in neighbouring France, which likely explains why around three quarters of parents reported choosing to vaccinate their children so they could “lead a normal life”. Similar to reports regarding willingness to vaccinate children against other diseases, we found that perceived vaccine safety was an important determinant of acceptance [37, 38]. In addition, in Switzerland, vaccination is not recommended for children who have already had COVID, which likely explains this reason being the most frequently reported among parents not intending to vaccinate their children.

As reported in other countries, we found that higher parental education and household income levels were associated with a higher likelihood of vaccinating children. Although we found no association with the parent’s sex, previous studies reported that fathers were more likely to intend to vaccinate their children, with mothers more concerned about the potential undesirable side effects [8, 34–36]. This lack of association may be because women were overrepresented in our sample.

Parental hesitancy to vaccinate their children against COVID-19 appears to be higher compared with other recommended vaccines, although our study did not consider vaccine hesitancy for children aged <5 years. For instance, a recent nationwide survey in Switzerland found that 26% of parents were vaccine-hesitant regarding the official recommendations for polio–pertussis–tetanus and measles vaccines during the first year of life [39, 40]. Notably, our results showed that even among parents not intending to vaccinate their children against COVID-19, only a small proportion (<5%) reported being against vaccines in general.

Implications for public health policies

As the pandemic enters new phases, with newly emerging virus variants driving new waves and exerting variable pressure on health systems, maintaining and reinforcing prevention strategies remains important, particularly including the vaccination of children and adolescents. Information campaigns must still emphasise that even though vaccines’ effectiveness in preventing infections can decrease over time and with the emergence of new variants of concern, they still confer a high level of protection against severe COVID-19, hospitalisation and death [2, 4]. Vaccination of children and adolescents remains an essential component of the prevention of severe COVID-19 and hospitalisation, as well as preventing long COVID and psychosocial and educational difficulties. As such, given the decreased willingness to vaccinate younger children, vaccination programs must frame their outreach message, particularly for socioeconomically disadvantaged parents, who are less likely to vaccinate their children. Identifying specific groups with the lowest level of acceptance of further COVID-19 vaccination is essential for effective science and public health communication, building general trust in vaccines and tackling socioeconomic inequalities. This will, in turn, strengthen other vaccination campaigns and prepare for future pandemics.

Strengths and limitations

The main strengths of this study are its population-based sample and the inclusion of data on sociodemographic and health-related characteristics, which allowed analysis of COVID-19 vaccination acceptance according to these factors. Importantly, participants were interviewed 6–10 weeks after vaccination was approved for children aged ≥5 years, which enabled their opinions to be based on real-life conditions instead of hypothetical vaccination of their children.

Several limitations of our study should be acknowledged. Although the participation rate in this study was relatively high, generalisation of the results presented here requires caution because our sample is not completely representative of the general adult population of the canton of Geneva. Our participants were older and had a higher level of education than the general adult population – a common characteristic of epidemiological population-based studies [41]. Participation required French literacy, internet access and digital literacy, potentially excluding part of the general population. Some other factors that may impact vaccination acceptance were not investigated, such as religious or political views, social network influence, and other indicators of socioeconomic conditions that are likely to influence individual perceptions and behaviour. Parents’ decisions regarding vaccination can also be influenced by their health care providers. In addition, especially for children and adolescents, we do not know about the presence of chronic diseases, which can impact the willingness of parents to have them vaccinated.

Conclusion

Parental willingness to vaccinate their children was high for adolescents aged 16–17 years, but it decreased considerably for younger adolescents and was markedly lower among unvaccinated parents. Underprivileged socioeconomic conditions were associated with being unvaccinated among adults and not intending to vaccinate their children. Given the proven efficacy of COVID-19 vaccines in preventing severe disease and hospitalisation, our results should be considered in designing future vaccination hesitancy research and in the communication and outreach strategies of vaccination programs targeting vaccine-hesitant groups.

Data availability statement

Coded study data that underlie the results reported in this article can be made available to the scientific community after deidentification and upon submission of a data request application to the investigators’ board via the corresponding author. The protocol of the overarching study (Specchio-COVID19) can be found at https://bmjopen.bmj.com/content/12/1/e055515.

Acknowledgements

We thank all the participants, without whom this study would not have been possible.

Authors’ contributions: HB, SS and IG designed the study. HB and SS designed the questionnaire for the survey. MEZ, CS and FP were involved in participant recruitment and implementation of the survey. NP conducted statistical analyses of the data. HB and CdM drafted the manuscript. EL and ALH reviewed the manuscript. All authors participated in the analysis interpretation and reviewed and approved the final manuscript.

Specchio-COVID19 study group: Specchio-COVID19 study group: Isabelle Arm-Vernez, Andrew S Azman, Fatim Ba, Oumar Ba, Delphine Bachmann, Jean-François Balavoine, Baysson H, Michael Balavoine, Hélène Baysson, Lison Beigbeder, Julie Berthelot, Patrick Bleich, Gaëlle Bryand Rumley, François Chappuis, Prune Collombet, Delphine Courvoisier, Alain Cudet, Carlos de Mestral Vargas, Paola D’Ippolito, Richard Dubos, Roxane Dumont, Isabella Eckerle, Nacira El Merjani, Antoine Flahault, Natalie Francioli, Marion Frangville, Idris Guessous, Séverine Harnal, Samia Hurst, Laurent Kaiser, Omar Kherad, Julien Lamour, Pierre Lescuyer, François L’Huissier, Fanny-Blanche Lombard, Andrea Jutta Loizeau, Elsa Lorthe, Chantal Martinez, Lucie Ménard, Lakshmi Menon, Ludovic Metral-Boffod, Benjamin Meyer, Alexandre Moulin, Mayssam Nehme, Natacha Noël, Francesco Pennacchio, Javier Perez-Saez, Giovanni Piumatti, Didier Pittet, Jane Portier, Klara M Posfay-Barbe, Géraldine Poulain, Caroline Pugin, Nick Pullen, Zo Francia Randrianandrasana, Aude Richard, Viviane Richard, Frederic Rinaldi, Jessica Rizzo, Khadija Samir, Claire Semaani, Silvia Stringhini, Stéphanie Testini, Didier Trono, Guillemette Violot, Nicolas Vuilleumier, Ania Wisniak, Sabine Yerly, María-Eugenia Zaballa.

References

Appendix: Supplementary tables and figures

The appendix is available in the pdf version of the article.