The Journal of Health Communication: Hesitant or Not? The Association of Age, Gender, and Education with Potential Acceptance of a COVID-19 Vaccine: A Country-level Analysis

This article was published in The Journal of Health Communication on March 13, 2021. It was co-authored by Jeffrey V. Lazarus, Katarzyna Wyka, Lauren Rauh, Kenneth Rabin, Scott Ratzan, Lawrence O. Gostin, Heidi J. Larson & Ayman El-Mohandes.

As the SARS-CoV2 pandemic reaches a year of global spread and a new surge of infections ravages many countries, manufacturers and governments are rushing to conclude studies and gain approval for COVID-19 vaccines that many people worldwide may not yet be willing to accept (Dror et al., 2020; Guarascio & Mason, 2020; Y. H. Khan et al., 2020; Jeffrey V. Lazarus et al., 2020; Palamenghi, Barello, Boccia, & Graffigna, 2020). The inconsistency to date of global governmental policy responses to the pandemic (Hale et al., 2020; Lazarus et al., 2020; Scarpetti, Webb, & Hernandez-Quevedo, 2020) suggests that implementation practices in regard to policy, education, communication, distribution, and availability of a new COVID-19 vaccine may also be inconsistent. Human factors—beliefs, perceptions, and behaviors—that influence rates of vaccine acceptance may additionally play a key role in achieving acceptable levels of community immunity required to slow the spread of the pandemic (Schoch-Spana, Brunson, Long, Ruth, & Trotochaud, 2020).

In June 2020 we surveyed 13,426 people in 19 countries and determined 71.5% of participants would be very or somewhat likely to take a COVID-19 vaccine (Lazarus et al., 2020). More recent opinion polling suggests a very wide range of likely vaccine acceptance, from as low as 40% in the United States (Talev, 2020) to as high as 87% in India (Foley, 2020). Indeed, the projected vaccine acceptance rate in our data also shows high variability when segmented at the national level and across demographic groups.

An individual’s intention to receive a vaccine hinges on complex systems of trust, in which individual perceptions of government, the medical industry, pharmaceutical companies, and employers all play a part (Figueiredo de, Simas, Karafillakis, Paterson, & Larson, 2020; Larson et al., 2018; Lazarus et al., 2020). From a review of the substantial and growing body of general research on vaccine hesitancy, confidence, and uptake, we know that motivators to vaccinate, including perceptions of risk, can vary by age, gender, and education-level (Deml et al., 2019; Klein & Pekosz, 2014; Larson et al., 2018; MacDonald, 2015; Yaqub, Castle-Clarke, Sevdalis, & Chataway, 2014). In addition, the highly charged social, political, and media climate surrounding the COVID-19 vaccine, in particular, appears to have bred new forms of hesitancy and new microclimates of fear and distrust (Burki, 2020; Chou & Budenz, 2020; Dror et al., 2020; Y. H. Khan et al., 2020). The specific challenges of this vaccine and the large-scale coverage needed for community protection will require sophisticated understanding of the drivers of the willingness to vaccinate, and accurate identification of potential pockets of heightened distrust.

A number of current scenarios may further amplify distrust, including the vaccine’s effectiveness (especially if it is below the efficacy reported at the end of the development phase), the speed of its development, the number of doses required, and the novelty of its underlying technology. All of these factors suggest an urgent need to prepare diverse and targeted messaging to preempt vaccine hesitancy and defuse an increasingly vocal opposition (Altay & Mercier, 2020; Schoch-Spana et al., 2020). This study aims to contribute additional insights into the demographics that may predict who is likely to accept or decline immunization in selected countries across the globe. We also include findings on the possible impact of trust in one’s employer to assess conceivable opportunities for businesses to motivate potentially vaccine-hesitant employees.

Materials and Methods

Study participants (n = 13,426) were recruited into the COVID-SCORE study in June 2020. Participants were from 19 countries (Brazil, Canada, China, Ecuador, France, Germany, India, Italy, Mexico, Nigeria, Poland, Russia, Singapore, South Africa, South Korea, Spain, Sweden, United Kingdom, United States). Sample sizes per country ranged from 619 to 773. The online questionnaire was administered by Emerson College (USA) after obtaining their written, informed consent about the survey and this project. No personally identifiable information was collected or stored. Sampling and further methodological details are described elsewhere (Jeffrey V. Lazarus et al., 2020; Jeffrey V., 2020). This study was approved by Emerson College, USA (IRB protocol number 20–023-F-E-6/12), with an expiration date of 11 June 2021.

We analyzed two survey questions pertaining to COVID-19 vaccine acceptance 1) If a COVID-19 vaccine is proven safe and effective and is available to me, I will take it; and 2) I would follow my employer’s recommendation to get a COVID-19 vaccine once the government has approved it as safe and effective. Response options were recorded on a 5-point Likert scale, ranging from completely agree to completely disagree.

Standard demographic questions included age, gender, and level of education. For this analysis the following age categorizations were considered 1) <50 or ≥50 (age group indicated in flu vaccination guidelines, 2) <40 or ≥40 (approx. mean age of the sample (median = 37)). Gender was categorized as male or female. Education was categorized as low (less than high school), medium (high school or some college), or high/very high (bachelor’s degree or postgraduate degree).

Analytic Approach

Analyses were stratified by country. First, each vaccine acceptance question was explored by demographic variables using descriptive statistics. Next, univariable and multivariable logistic regression was used, where the outcome was defined as 1 if a respondent answered, “completely agree” or “somewhat agree” and 0 for any other response. The independent demographic variables were age groups, gender, and education. Odds ratios and corresponding 95% confidence intervals (on a log scale) showing unadjusted and adjusted associations are reported.


Acceptability of COVID-19 Vaccine if Generally Available

Women in France, Germany, Russia, and Sweden indicated stronger willingness to accept COVID-19 vaccine than men. Older age (<50 vs. ≥50) was a significant factor in Canada, Poland, Sweden, and the UK. In China, an opposite trend was observed, with younger individuals stating they were more likely to accept a vaccine. Results were not significantly different when comparing respondents aged <40 vs. ≥40 (data not shown).

Educational differences also mattered substantially in some countries. Individuals with high/very high levels of education in Ecuador, France, Germany, India, and the US were more likely to say they would accept a vaccine, but higher education was linked to lower vaccine acceptance in Canada, Spain, and the UK (Table 1).

Table 1. Vaccine acceptance (general and when recommended by employer) by demographics by country


Adjusted associations indicated a consistent relationship between selected demographic factors and willingness to accept vaccine in France and Germany (older age, females, and high/very high/very high education), Sweden (older age, females), Poland, and the UK (older age), Russia (females), Ecuador, India, and US (high/very high education). Of note in Canada, Spain, and the UK high/very high education may be linked to lower vaccine acceptance (Figure 1).

Figure 1. Adjusted associations between vaccine acceptance (general) and demographics by country (odds ratios and 95%CI (log scale))

Reference categories: Female, Age<50, Low education.



Acceptability of COVID-19 Vaccine if Recommended by One’s Employer

Women were substantially more likely to follow an employer’s vaccination recommendation in Brazil and the US, while men were more likely to accept their employer’s recommendation in India and South Korea.

Age differences were significant in Brazil, Ecuador, Mexico, and South Africa, where older individuals (<50 vs ≥50) were more likely to accept their employer’s vaccination recommendation, while in France and the US, younger individuals were more likely to do so. Results were consistent when respondents age <40 were compared with those age ≥40; however, younger individuals in Italy, Poland, and Russia were more likely to indicate their willingness to accept an employer’s vaccination recommendation.

Individuals with high/very high education were significantly more likely to accept an employer’s recommended vaccination in Ecuador, France, India, Mexico, Sweden, and the US (Table 1).

Consistent adjusted associations between demographic factors and willingness to accept employer-recommended vaccine were evident in Brazil (females, older age), Ecuador and Mexico (older age and higher education), France (younger age and higher education), India (males and lower education), South Africa (older age), South Korea (males), and the US (females, younger age, higher education). (Figure 2).

Figure 2. Adjusted associations between vaccine acceptance (recommended by employer) and demographics by country (odds ratios and 95%CI (log scale))

Reference categories: Female, Age<50, Low education.



When analyzed by gender, age, and education level, our results indicate a high level of heterogeneity at the national level. This is consistent with prior research findings regarding the relationship between vaccine hesitancy and either held beliefs or misinformation amongst various segments of the population (Butler & MacDonald, 2015; Dubé et al., 2018; MacDonald, Butler, & Dubé, 2018). Identifying the underlying factors associated with the choice to vaccinate oneself and one’s children will vary significantly by location and the effect of individual-level factors. Such variance can be attributed to complex socio-environmental, psychological, and cultural influences (Deml et al., 2019; Dubé et al., 2018; Dubé, Gagnon, Nickels, Jeram, & Schuster, 2014; Habersaat & Jackson, 2020; Jarrett, Wilson, O’Leary, Eckersberger, & Larson, 2015; Kumar et al., 2012; Larson, Jarrett, Eckersberger, Smith, & Paterson, 2014; Yaqub et al., 2014). For these reasons, and especially given the significant resurgence of the COVID-19 pandemic, careful planning for widespread acceptance of a COVID-19 vaccine is extremely relevant and time sensitive and will require an awareness of preexisting data as well new insights regarding media preference and governmental action or inaction obtained during the current crisis. Hesitancy associated specifically with a COVID-19 vaccine could be a unique reflection of mistrust in what may be perceived as politically motivated and nonscientific decisions and policies (Editors, 2020; Warren, Forrow, Hodge, & Truog, 2020). Recent research suggests that an individual’s political beliefs, as well as the attributes of the vaccine itself (e.g. where it was manufactured and which regulatory mechanisms were involved in its approval, as well as production and procurement), could impact perceptions of the vaccine, its efficacy and safety, and consequently the willingness to accept it (Kreps et al., 2020). Similarly, confidence in and demand for a vaccine may vary as a result of specific pandemic-related circumstances or experiences that influence public perceptions and patterns of belief (Determann et al., 2014; Scherr, Jensen, & Christy, 2017; Talarek et al., 2020). This suggests that an “environmental analysis” of the impact and severity of the pandemic and the adoption of associated mitigating measures (e.g. facemask wearing) may indicate which strategies will be necessary to promote the vaccine effectively in specific communities and populations. The appropriateness of the content and format of messaging about the vaccine is critical, both as it pertains to groups with a lower level of general or health and vaccine literacy and a heightened degree of cultural isolation and disenfranchisement and, as discovered in this analysis, some people with relatively high levels of educational attainment. Likewise, who the communication is from, e.g. which political leaders or parties, community or religious leaders, or other key stakeholders, must be carefully considered.

Different reasons for vaccine refusal or hesitancy—such as fear of side-effects, distrust in pharmaceutical companies, or a belief in a “chemical-free” lifestyle—are associated with different vaccination behaviors (Determann et al., 2014; Navin, Wasserman, Ahmad, & Bies, 2019). In other words, widespread sentiments of vaccine hesitancy, when explored at the individual-level, may turn out to have different root causes, leading to variations in vaccination choice, such as choosing to accept some vaccines over others, or agreeing to vaccinate only under particular circumstances (e.g. an altered schedule) (Navin et al., 2019). Such “hyper-local” heterogeneity of vaccine sentiments creates the risk of pockets of unvaccinated or under-vaccinated people within larger communities of vaccinated individuals. Even when these pockets are comparatively few and far between, their presence risks undermining the conditions necessary for sufficient levels of herd immunity. This behavioral pattern has been observed in a number of settings including in Ultra-Orthodox Jewish communities in Israel and the UK (Letley et al., 2018; Muhsen et al., 2012) and neighborhoods and villages in India and Pakistan (T. Khan & Qazi, 2013; Onnela et al., 2016). When there is an outbreak in an under-vaccinated pocket, those who are most vulnerable (e.g. infants and individuals with autoimmune disorders) are at a heightened risk. Greater understanding of where these potential pockets may lie helps prepare health authorities and communication experts to tailor messages aimed at the specific fears or concerns of these subgroups.

To date, much of the research aimed at understanding the association of gender and uptake has focused specifically on the HPV vaccine: pregnant woman or mothers who express hesitancy to vaccinate their children (Askelson, 2012; Cox, Cox, Sturm, & Zimet, 2010; Shafer, Cates, Diehl, & Hartmann, 2011; Shui, Kennedy, Wooten, Schwartz, & Gust, 2005; Topuzoğlu, Ay, Hidiroglu, & Gurbuz, 2007; Weiner, Fisher, Nowak, Basket, & Gellin, 2015). This research suggests that women are very often the gatekeepers to health decisions for their families, and in countries where results indicate a positive association between gender and vaccine acceptance, this finding could lead to more targeted messages. Similarly, in countries where women were more likely to accept their employer recommended vaccine—such as Brazil and the US—messages promoting COVID-19 vaccination that come from an employer or in the form of a company-provided vaccination campaign could have a positive impact beyond the employee base itself. The higher tendency of women in countries like France, Germany, Sweden, and Russia to accept a vaccine could reflect women’s higher level of empathy amongst women for the safety of families and communities, but also for the most vulnerable amongst us. Another explanation may lie in the disproportionate economic impact the pandemic has had on women and reflect their hope that widespread vaccine coverage could permit them to return to previously routine employment status and childcare arrangements (UN Women, 2020). Nonetheless, more attention to designing and disseminating messages suited to a more hesitant male audience may be an important focus for some countries in the immediate future.

The COVID-19 pandemic is clearly vivid and urgent in peoples’ minds, and the morbidity and mortality associated with it has been felt directly in many communities. It is not theoretical. It is a harsh reality that could be drawn upon carefully to increase readiness to accept the vaccine. For example, this could be especially true amongst the highly vulnerable populations, such as Black Americans, who have, for historic reasons been wary of government-sponsored immunization initiatives (Guarascio & Mason, 2020; Warren et al., 2020). Even though older adults are at higher risk for developing acute illness when sick with COVID-19, our research indicated that the likelihood of older respondents to accept a potential vaccine was by no means consistent from country to country. Previous research on the behavior and beliefs of older adults regarding the influenza vaccine has also shown significant variation, which indicates that messages that focus solely on heightened personal risk cannot be relied on as the sole motivator to vaccinate (Abbas, Kang, Chen, Werre, & Marathe, 2018; Cameron et al., 2009; Gatwood et al., 2020; Santibanez, Mootrey, Euler, & Janssen, 2010). In countries where age indicated greater readiness to accept a vaccine—Canada, Poland, Sweden, and the UK— risk-based messages may ensure high coverage rates among older individuals and their families, but vaccination program managers should be prepared to confront resistance based on different independent variables such as socioeconomic status and education, and also personal beliefs and specific perceptions (Gatwood et al., 2020; Schmid, Rauber, Betsch, Lidolt, & Denker, 2017). Our results show that China may face barriers to achieving high COVID-19 vaccine coverage among older adults, and recent studies have pointed out increasing rates of vaccine hesitancy across the country (Ren et al., 2018; Yang, Penders, & Horstman, 2020). Further, in countries where younger individuals were less willing be vaccinated against COVID-19, awareness campaigns should target this age group.

Higher levels of health and vaccine literacy and consequent understanding of the need for widespread vaccination coverage have previously been associated with higher levels of education (L. R. Biasio et al., 2020; L. R. Biasio, 2017; Qin, Niu, Huang, & Xu, 2011; Vikram, Vanneman, & Desai, 2012). This may explain why education was substantially associated with acceptance of a vaccine in American, Ecuadorian, French, German, and Indian respondents in this study. On the other hand, however, higher levels of education seemed to correlate with a lower likelihood of vaccine acceptance among respondents from Canada, Spain, Sweden, and the UK. This paradox has been previously noted when gauging the education levels of mothers as a predictor of vaccine acceptance (Murfin, Irvine, Meechan‐Rogers, & Swift, 2019; Navin et al., 2019), which indicates that especially in high-income countries, education cannot invariably be relied upon as an indicator of vaccine acceptance. Moreover, higher education does not always indicate greater or more accurate knowledge of how vaccination works, and inaccurate perceptions of risk must also be considered when planning and designing risk communications and selecting media platforms.


As our findings suggest, the challenges of overcoming expected hesitancy to COVID-19 vaccination worldwide will not be met by any one communications strategy, message set, spokesperson selection, medium, or venue for dissemination. Our study represents an initial effort to delineate the diversity and extent of the challenges, but it also underscores at the very least that “one size will not fit all” when it comes to building public trust in a COVID-19 vaccine. More sophisticated and nationally- and community-relevant investigations must follow quickly.

We are also aware that vaccine communications too often take a “backseat” to the substantial and daunting challenges of developing vaccines, proving their safety and efficacy, scaling up their production consistent with good manufacturing practice, and assuring their widespread and equitable distribution and administration under safe conditions. We must further develop comprehensive, multistakeholder communication strategies and mitigate vaccine hesitancy before the logistical challenges of distribution and access to the recently approved COVID-19 vaccines become major obstacles. The dramatic setback in global economies associated with this pandemic may help vaccine advocates in justifying an immediate and aggressive investment by governments and other interested parties to help resolve some of the ambivalence associated with this new vaccine development and deployment. Such an investment will not only save human lives but also expedite the recovery of a global economy at the brink of collapse.

Achieving successful global immunization against COVID-19 will be among the greatest public health challenges that most of us will face in our lifetimes. And we believe that effective vaccine communications must play a central role in that success.


JVL acknowledges support to ISGlobal from the Spanish Ministry of Science, Innovation and Universities through the “Centro de Excelencia Severo Ochoa 2019–2023” Programme (CEX2018-000806-S) and from the Government of Catalonia through the CERCA Programme.

Declaration of interest statement

The authors declare no conflicts of interest.

Data availability statement

Data are available upon request from the corresponding author.

Ethics statement

This study was approved by Emerson College, USA (IRB protocol number 20-023-F-E6/12) with an expiration date of 11 June 2021. The online questionnaire was administered by Emerson College to gather information from respondents after obtaining their written, informed consent about the survey and this project. Equitable compensation per survey was applied ($2 per complete for Mturk data and increased up to US $3 in some countries) regardless of country being polled to comply with ethical compensation standards. No personally identifiable information was collected or stored.

Additional information


This work was supported by the City University of New York Graduate School of Public Health and Health Policy Foundation, Bocconi University, J. Fielding, the United States Council for International Business Foundation, and K. Rabin. Consensus Strategies was the funding recipient.