Since December 2019, when Wuhan, China, became the centre of an outbreak of the disease COVID-19 (1), which subsequently spread around the world, the most interest has been focused on developing effective treatments of COVID-19 or developing safe and effective vaccines that are essential for the effective management of the pandemic (2).

In April 2021 (the time of our study), four vaccines had undergone phase 3 of a clinical trial; their efficacy had been proved and they were authorised for use in the European Union by the European Medicines Agency (EMA) (3). However, the mere presence of approved vaccines is not sufficient. So-called herd immunity is also needed to get the pandemic successfully under control. Based on the currently available epidemiological data, for a vaccine with a declared 90%–95% efficacy, the required herd immunity level would be 63%–76% (4, 5). To reach this limit, widespread vaccine acceptance and a willingness to be vaccinated are needed. However, studies on COVID-19 vaccination acceptance suggest that low vaccination rates have remained an issue of concern, while vaccine hesitancy and refusal are increasing (6, 7). In general, low willingness to get vaccinated is a growing problem, as also acknowledged by the WHO, which in 2019 (prior to the COVID-19 pandemic) labelled vaccine hesitancy among the top 10 threats to global health (8). Thus, there is a need to investigate what influences the willingness to be vaccinated. In the future, another situations similar to the current pandemic may arise and the willingness to be vaccinated may be essential in dealing with them.

Studies on COVID-19 vaccination hesitancy or refusal have shown that the common reason for COVID-19 hesitation or rejection is the fear of vaccine safety (9, 10) or its side effects, the belief that vaccines are not necessary (10), are effective (9–11), perceived unknown/short duration of immunity (11), or refusal of the vaccination generally (10). However, many factors can affect vaccination attitudes and can be complicated by the multifaceted nature of this phenomenon. Factors influencing the hesitation in accepting the vaccination could include cultural, sociodemographic, political, cognitive, psychological and spiritual factors (12–16). Thus, analysis of such factors in connection with COVID-19 vaccination is needed for guiding interventional measures for minimising the group of people refusing COVID-19 vaccines.

In connection with COVID-19 vaccination, research has shown that hesitation/refusing is associated, for example, with female gender, younger age, lower education (11, 17–20), living in a partnership (19), rural housing (11, 20), lower household income (11, 18–20), better health condition (17, 21), spirituality (22), religiosity (23, 24), belief in conspiracy theories (19, 25, 26) and poorer knowledge about COVID-19 (27). But to the best of our knowledge, this is one of the first studies that assesses the relationship between knowledge about COVID-19 vaccines with a tendency to refuse a COVID-19 vaccine, and in this relationship to compare the declared knowledge against the real knowledge.

Moreover, in April 2021 (the survey time), the Czech Republic (CZ) was the most affected country in Europe in terms of the number of positive cases per million inhabitants and the third-worst affected country in the world (28). CZ is also a country where vaccination is free for all, so there are no financial barriers affecting the willingness to be vaccinated. Therefore, it represents an interesting research area for examining the willingness to get vaccinated and the factors influencing this attitude. Thus, this study aims: 1) to examine the prevalence of people willing/refusing a COVID-19 vaccine in CZ and 2) to determine which of the assessed factors (sociodemographic, government trust, knowledge about COVID-19 vaccines, personal characteristics, anxiety, or depression) are related to the unwillingness to get vaccinated in such an “extreme” environment.

The sociodemographic characteristics of the sample are presented in Table 1. The final sample consisted of 1,401 participants (mean age = 50.46; SD = 15.73; 55.7% male). Most of the respondents (58.9%) reported COVID-19 vaccine acceptance. Further, 19.2% of respondent trusted the government, 16.3% were undecided and 20.9% of respondent trusted that government is managing the pandemic and 16.8% did not know.

Regarding the personal characteristics, the respondents achieved the following mean scores: openness 32.96 (SD = 5.79), conscientiousness 31.25 (SD = 4.91), extraversion 24.29 (SD = 5.09), agreeableness 31.46 (SD = 4.48), neuroticism 22.67 (SD = 5.87), OASIS 9.35 (SD = 4.18), ODSIS 8.69 (SD = 4.33). Regarding the knowledge the mean value of the declared knowledge was 3.36 (SD = 1.12), which corresponds roughly to intermediate knowledge. The mean value of the real knowledge was 1.02 (SD = 1.24), which means that, on average, respondents answered only one question correctly. So the real knowledge was much lower than the declared knowledge.

Table 2 shows the associations of selected factors COVID-19 vaccination refusal. A significantly higher tendency toward vaccination refusal was observed among women, people living without a partner, self-employed people (adjusted model), unemployed people (adjusted model), respondents living in a town, believers outside the church and also respondents who did not trust the government or those who thought that government was not managing the pandemic.

In contrast, a lower tendency to vaccination refusal was significantly associated with age, with a 3% decrease in the odds ratio (p ˂ 0.001) for each year. Also, disabled/old pensioners (crude model) and respondents with a higher education (crude model) were less likely to refuse vaccination.

Table 3 shows the associations of COVID-19 vaccine refusal with COVID-19 knowledge, information source and information verification. Respondents’ declared knowledge (crude and adjusted models) and their real knowledge (crude models) about the COVID-19 vaccines were associated with a decreased tendency towards vaccination refusal, with associations ranging from OR = 0.79 (p˂0.001) for declared knowledge to OR = 0.87 (p˂0.05) for real knowledge (crude models). Regarding the source of information, in the crude model on real knowledge, respondents who obtained information about vaccination from experts (doctor or other experts) were 25% less likely to refuse a COVID-19 vaccine, and on declared knowledge, respondents who obtained information about vaccination from social sources (e.g., friends, social media or e-mail) were 1.45-times more likely to refuse a COVID-19 vaccine (p ˂ 0.05). Verification of information did not significantly affect the willingness to be vaccinated.

Table 4 shows associations of COVID-19 vaccine refusal with personal characteristics, anxiety and depression. Regarding personal characteristics, an increased extroversion score was significantly associated with a higher probability of COVID-19 vaccine refusal, with a 17% (crude) or 16% (adjusted) increase in the odds ratios (p ˂ 0.05) for each point of the extroversion score. Furthermore, a higher level of neuroticism was related to a lower chance of refusing vaccination, with a 17% decrease in the odds ratios (p ˂ 0.01, adjusted). Moreover, the table reveals that respondents’ higher level of depression was connected with a higher tendency to refuse the vaccination. Other associations were not significant.

This study aimed to examine the prevalence of people accepting/refusing a COVID-19 vaccine and to determine which factors affect the unwillingness to get vaccinated in the Czech Republic, one of the countries most affected by COVID-19 infection in the world (28). The results showed that only 59% of respondents were willing to get vaccinated. Respondents who were more likely to reject the vaccine were: female, younger, living without a partner, self-employed and unemployed, living in a town, believers outside a church, and who did not trust the government or did not trust that the government is managing the pandemic, who obtained information about the vaccine from social media, were extroverts and those who felt depressed. In contrast, respondents who were less likely to refuse the vaccine were: disabled or old-age pensioners, people with higher education, respondents with better knowledge of the COVID-19 vaccines, who obtained information about vaccines from experts and who had higher scores in neuroticism.

We found that 58.9% of respondents reported COVID-19 vaccine acceptance, 18.7% vaccine refusal and 22.4% hesitancy. The worldwide willingness to be vaccinated was estimated in a meta-analysis by Nehal et al. (36) at 66% and the European willingness was 67%. At the time of the study, CZ was the third-worst affected country in the world (28). Given that a higher risk of infection is associated with a higher willingness to be vaccinated (21), it would be expected that in this situation the willingness to be vaccinated would be higher in CZ. However, CZ is below European and global estimates. Although some sociodemographic characteristics influence the willingness to be vaccinated, as described below, there are also other factors that could reduce the willingness to be vaccinated in comparison to other countries. Some of these possible factors are presented in the second part of the Discussion.

Our second observation was that females and younger people were more likely to refuse a COVID-19 vaccine, which is in line with majority of other studies (11, 17–20, 23, 37–39), however, a minority of studies where men were more likely to refuse a COVID-19 vaccine (40, 41). Women and younger people do not belong to the risk group for severe and critically ill COVID-19 patients (42), so they may not have felt as threatened. In contrast, pensioners belong to the risk group (due to their age) and we observed, like Freeman et al. (19), a higher willingness to be vaccinated. Consequently, we also found that people with higher education were less likely to refuse a COVID-19 vaccine which is in line with other studies (11, 17–19, 23). Furthermore, similar to other studies that have found religious people to be less willing to be vaccinated (23, 24), our study also shows that religiosity had an influence on vaccine refusal but only in case of believers outside the church. It is possible that believers are more influenced by the official position of the Roman-Catholic Church (the most widespread church in the Czech Republic), which recommended vaccination (43). It is thus possible that not religiosity itself but other factors (e.g., spirituality, especially some of its forms highlighting one’s inner power and individuality) are a barrier to vaccine uptake (22).

In our study, another group refusing COVID-19 vaccination comprised people living without a partner. The opposite findings were reported by Freeman et al. (19) and Giuseppe et. Al (37). Further, other studies also came to different conclusions regarding vaccine refusal among the self-employed or unemployed (40, 41, 44) and people living in urban areas (11, 20). These discrepancies may be caused by various factors. First, when comparing the willingness/unwillingness to be vaccinated and its association with sociodemographic factors, the fact that different studies had various sample compositions (e.g., regarding age, gender, education) and used different designs should be taken into consideration. Hence, comparisons are only possible to a limited extent. Second, determining the sociodemographic factors of acceptability of vaccines is complex and context-specific (e.g., demographic or historical differences).

Furthermore, we found that, in general, only 19.2% of people trusted the government, and 20.9% believed the government was handling the COVID-19 pandemic. The proportion of people who trusted the government is significantly lower compared to other countries (e.g., Australia 80% and New Zealand 83%) (45). This is also confirmed by OECD data from 2020 (46), where CZ was among seven countries with the lowest level of trust in government. Moreover, one of the strongest relationships with vaccine refusal was distrust in the government or its managing of the pandemic. These observations align with the findings of other authors who mention that lower government trust and the measures they implemented during the COVID-19 pandemic were associated with delaying or refusing a COVID-19 vaccine (21, 47). Thus, low trust in the government could explain the low levels of vaccination uptake in CZ.

Further, we found that there were differences between declared knowledge about the COVID-19 vaccines and real knowledge (as assessed by control questions), and that real knowledge was lower than declared knowledge. This phenomenon is entirely consistent with the Dunning-Kruger effect, which describes that people tend to have an overly favourable opinion of their own abilities in many social and intellectual domains (48). However, though real knowledge was associated with a higher willingness to be vaccinated, the same held for declared knowledge, which corresponds to the findings of other authors (20, 49).

Although, as mentioned above, it does not seem to matter whether knowledge is real or claimed, our data suggest that the sources of information matter. Participants drawing information from experts were more likely to accept vaccines, which is consistent with other studies that have shown the important role of the healthcare workers as source of information with a positive impact on willingness to be vaccinated (50, 51). In contrast, participants drawing information from social media were more likely to refuse vaccines. Similar findings were observed by Reno et al. (52), who suggested that social media directly or indirectly increase vaccine hesitancy towards COVID-19 vaccination, while the opposite effect was observed for institutional websites. An explanation for this could be that social media have been identified as major vectors for the dissemination of misinformation and conspiracy theories (53–55), and at the same time, people who believe in COVID-19 conspiracy theories are less willing to get vaccinated (25, 26).

Further, our data also suggest that respondents with a higher score for extroversion had a higher probability of vaccine refusal. Conversely, people with a higher score for neuroticism had a lower likelihood of vaccine refusal. In contrast, Howard’s study (56) suggested that extroversion has positive effects on pro-vaccine outcomes and explained this by the fact that extroversion is associated with positive expectations (57), so they could have expected greater benefits from vaccination (56). However, this positivity can also be turned towards a positive expectation of a disease course (mild course) and an associated lower need to be vaccinated. In addition, extroverts are also likely to use social media (58). As we mentioned in the previous paragraph, social media could lead to less willingness to be vaccinated. Regarding neuroticism, the explanation could be that neuroticism is positively related to a perceived risk of infection (59), which may lead these respondents to be more willing to get vaccinated for their own safety.

Finally, we found that with depression, the chance of vaccine refusal also increased, while anxiety did not affect attitude toward vaccination. These findings are not consistent with other studies, which are, however, heterogenous themselves. Urrunaga-Pastor et al. (60) found that having depressive symptoms was associated with a higher probability of vaccine intention, but Bendau et al. (61) did not find any relationship. This inconsistency is probably due to different ways of measuring depression. Thus, it is not easy to interpret the findings, and this relationship needs to be verified in subsequent studies on a larger sample. With regards to anxiety, we found no significant association, which is consistent with study by Bendau et al. (61).